Radiation Oncology: Rationale, Technique, Results, 9th Edition

1600 John F. Kennedy Blvd. Ste 1800 Philadelphia, PA 19103-2899

Radiation Oncology: Rationale, Technique, Results Copyright © 2010 by Mosby, Inc., an affiliate of Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions.

This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein).

Notice Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researches must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. With respect to any drug or pharmaceutical products identified, readers are advised to check the most current information provided (i) on procedures featured or (ii) by the manufacturer of each product to be administered, to verify the recommended dose or formula, the method and duration of administration, and contraindictions. It is the responsibility of practitioners, relying on their own experience and knowledge of their patients, to make diagnoses, to determine dosages and the best treatment for each individual patient, and to take all appropriate safety precautions. To the fullest extent of the law, neither the publisher nor the authors, contributors or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. The Publisher Previous editions copyrighted 2003, 1994, 1989, 1979, 1973, 1969, 1965, 1959

Library of Congress Cataloging-in-Publication Data Radiation oncology: rationale, technique, results. — 9th ed. / [edited by] James D. Cox, K. Kian Ang. P. ; cm. Includes bibliographical references and index. ISBN 978-0-323-04971-9 1. Cancer—Radiotherapy. I. Cox, James D. (James Daniel), 1938- II. Ang, K. K. (K. Kian) [DNLM: 1. Neoplasms—radiotherapy. QZ 269 R12803 2010] RC271.R3R3315 2010 616.99’40642-dc22 2009018138

Acquisitions Editor: Dolores Meloni Publishing Services Manager: Frank Polizzano Senior Project Manager: Pete Faber Design Direction: Ellen Zanolle

Printed in China Last digit is the print number:  9  8  7  6  5  4  3  2  1

ISBN: 978-0-323-04971-9

Contributors

Anesa Ahamad, M.D.

James D. Cox, M.D., F.A.C.R., F.A.S.T.R.O.

K. Kian Ang, M.D., Ph.D.

Christopher H. Crane, M.D.

Associate Lecturer, Faculty of Medical Sciences University of the West Indies at St. Augustine Trinidad and Tobago Professor and Deputy Chair, Department of Radiation ­Oncology Gilbert H. Fletcher Distinguished Memorial Chair The University of Texas M. D. Anderson Cancer Center Houston, Texas Matthew T. Ballo, M.D.

Associate Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Peter A. Balter, Ph.D., D.A.B.R.

Assistant Professor, Department of Radiation Physics The University of Texas M. D. Anderson Cancer Center Houston, Texas James D. Brierley, M.B.B.S., F.R.C.P.(UK), F.R.C.R., F.R.C.P.(C)

Professor, Department of Radiation Oncology University of Toronto Princess Margaret Hospital Toronto, Ontario Canada

Professor and Head, Division of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Bouthaina Dabaja, M.D.

Assistant Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Prajnan Das, M.D., M.S., M.P.H.

Assistant Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Marc E. Delclos, M.D.

Associate Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Lei Dong, Ph.D.

Associate Professor, Department of Radiation Physics The University of Texas M. D. Anderson Cancer Center Houston, Texas

Thomas A. Buchholz, M.D.

Patricia J. Eifel, M.D., F.A.C.R., F.A.S.T.R.O

Roger W. Byhardt, M.D.

Douglas B. Evans, M.D.

Min Rex Cheung, M.D., Ph.D.

Selim Y. Firat, M.D.

Professor and Chair, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Professor, Department of Radiation Oncology Medical College of Wisconsin Milwaukee, Wisconsin Assistant Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas

Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Professor and Chairman, Department of Surgery Medical College of Wisconsin Milwaukee, Wisconsin Assistant Professor, Department of Radiation Oncology Medical College of Wisconsin Milwaukee, Wisconsin Adam S. Garden, M.D.

Jay S. Cooper, M.D., F.A.C.R., F.A.C.R.O., F.A.S.T.R.O.

Director, Maimonides Cancer Center Chair, Department of Radiation Oncology Maimonides Medical Center Brooklyn, New York

Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas



vi

Contributors

Michael Gillin, Ph.D.

Sunil Krishnan, M.D.

Mary K. Gospodarowicz, M.D., F.R.C.P.C., F.R.C.R.(Hon)

Deborah A. Kuban, M.D.

Professor, Department of Radiation Physics The University of Texas M. D. Anderson Cancer Center Houston, Texas Professor and Chair, Department of Radiation Oncology University of Toronto Princess Margaret Hospital Toronto, Ontario Canada B. Ashleigh Guadagnolo, M.D.

Assistant Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Chul S. Ha, M.D.

Distinguished Chair in Radiation Oncology Professor and Chair, Department of Radiation Oncology The University of Texas Health Science Center at San Antonio Associate Director & Associate Physician in Chief Cancer Therapy and Research Center San Antonio, Texas

Associate Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Colleen A. Lawton, M.D., F.A.C.R.

Professor, Department of Radiation Oncology Medical College of Wisconsin Milwaukee, Wisconsin Andrew K. Lee, M.D., M.P.H.

Associate Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Valerae O. Lewis, M.D.

Associate Professor, Department of Orthopedic Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Zhongxing Liao, M.D.

Eric J. Hall D.Sc., F.A.C.R., F.R.C.R.

Higgins Professor Emeritus Columbia University New York, New York

Associate Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Michelle S. Ludwig, M.D., M.P.H.

David H. Hussey, M.D., F.A.C.R.

Professor Emeritus, Department of Radiation Oncology The University of Texas Health Science Center at San Antonio San Antonio, Texas Nora A. Janjan, M.D.

Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas

Resident, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Marvin L. Meistrich, Ph.D.

Ashbel Smith Professor, Department of Experimental Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Raymond E. Meyn, Jr., Ph.D.

Anuja Jhingran, M.D.

Associate Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas

Professor, Department of Experimental Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas

Stella K. Kim, M.D.

Luka Milas, M.D., Ph.D.

Ritsuko Komaki, M.D., F.A.C.R., F.A.S.T.R.O.

Michael F. Milosevic, M.D., F.R.C.P.C.

Associate Professor, Section of Ophthalmology Department of Head and Neck Surgery The University of Texas M. D. Anderson Cancer Center Houston, Texas Professor, Department of Radiation Oncology Gloria Lupton Tennison Endowed Professor in Lung Cancer Research The University of Texas M.D. Anderson Cancer Center Houston, Texas

Professor, Department of Experimental Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Professor, Department of Radiation Oncology University of Toronto Princess Margaret Hospital Toronto, Ontario Canada

Contributors

Radhe Mohan, Ph.D.

Professor and Chair, Department of Radiation Physics The University of Texas M. D. Anderson Cancer Center Houston, Texas

Elizabeth L. Travis, Ph.D.

Professor, Department of Experimental Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas

William H. Morrison, M.D.

Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Firas Mourtada, Ph.D.

Associate Professor, Department of Radiation Physics The University of Texas M. D. Anderson Cancer Center Houston, Texas Michael S. O’Reilly, M.D.

Associate Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas

Richard W. Tsang, M.D., F.R.C.P.(C)

Professor, Department of Radiation Oncology University of Toronto Princess Margaret Hospital Toronto, Ontario Canada Gauri R. Varadhachary, M.D., M.B.B.S.

Associate Professor, Department of Gastrointestinal Medical Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas B-Chen Wen, M.D.

Alan Pollack, M.D., Ph.D.

Chair and Sylvester Professor of Radiation Oncology The University of Miami Miller School of Medicine Miami, Florida

Professor, Department of Radiation Oncology Miller School of Medicine The University of Miami Miami, Florida

David L. Schwartz, M.D.

Robert A. Wolff, M.D.

Assistant Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Almon S. Shiu, Ph.D.

Professor, Department of Radiation Physics The University of Texas M. D. Anderson Cancer Center Houston, Texas George Starkschall, Ph.D.

Professor, Department of Radiation Physics The University of Texas M. D. Anderson Cancer Center Houston, Texas Eric A. Strom, M.D.

Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Gillian M. Thomas, B.Sc., M.D., F.R.C.P.C., F.R.C.R.(Hon)

Department of Radiation Oncology Edmond Odette Cancer Centre at Sunnybrook University of Toronto Toronoto, Ontario Canada

Professor of Medicine Department of Gastrointestinal Medical Oncology Division of Cancer Medicine The University of Texas M. D. Anderson Cancer Center Houston Texas Shiao Y. Woo, M.D.

Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas Gunar K. Zagars, M.D.

Professor, Department of Radiation Oncology The University of Texas M. D. Anderson Cancer Center Houston, Texas

vii

preface

Radiation oncology has undergone a major transformation in the past few years. Results of three-dimensional conformal radiation therapy and intensity-modulated radiation therapy have confirmed the promise that was evident from treatment-planning studies and early results of phase II and phase III trials. Advances in functional and metabolic imaging have augmented the tools for evaluating disease and planning treatments. Refinements of cytotoxic chemotherapy in conjunction with radiation therapy have reduced side effects and improved efficacy. The rapidly developing arena of molecular therapeutic agents and the emerging ability to select some agents based on molecular abnormalities in tumors promise to revolutionize combined modality therapy. Surgical procedures have emphasized preservation of structures, video assistance, and robotic tools such that the indications for adjuvant radiation therapy have been expanded and modified. This ninth edition has updated, replaced, and eliminated chapters from previous editions to seek the most contemporary discussions of principles and specific disease sites. New contributors have been enlisted and new subjects have been added. In recognition of the importance of color displays, most figures in the ninth edition are now in color. This greatly enhances the value of the figures, especially those that show dose distributions for specific diseases.

As in previous editions, we have sought consistency in the presentation of each chapter. Each begins with discussions of effects of treatments on normal tissues. Emphasis has been placed on synthesis of the available literature for clarity and brevity. The thrust of this approach is accessibility of radiation oncology to students and colleagues from other disciplines as well as physicians, physicist, and biologists from various countries. We are thankful to our colleagues who have worked arduously to meet the guidelines we have set for presenting their material in this manner. This edition has been prepared in the context of ongoing care of patients, research, and administration. We are grateful to those who have worked with us to permit its completion. We are indebted beyond words to Christine Wogan, who has edited this edition and the previous one. Her knowledge and advice has served to gather the thoughts and words of the many contributors into a more unified whole. Her patience and goodwill are without compare. James D. Cox, M.D. K. Kian Ang, M.D., Ph.D.

ix

Physical and Biologic Basis of Radiation Therapy

1

Eric J. Hall and James D. Cox HISTORICAL BACKGROUND PHYSICAL BASIS Types of Radiation Production of Radiation for Therapeutic Applications The Measurement and Meaning of Dose Physical Basis for Choice of Treatment Techniques CANCER BIOLOGY Mechanisms of Carcinogenesis The Multistep Nature of Cancer Functions of Oncogenes and Tumor Suppressor Genes Signal Transduction The Cell Cycle Cancer Genetics BIOLOGIC BASIS OF RADIATION THERAPY Radiation Chemistry DNA Breaks Chromosomal Aberrations

Cell Survival Curves Effects on Normal Tissues Effects on Tumors Tissue and Tumor Kinetics Systemic Effects Fractionation Effect of Tumor Volume on Tumor Control by Irradiation Predicting Tumor Sensitivity or Resistance to Radiation Modifiers of Radiation Effects Radiation Quality Use of Neutrons in Radiation Therapy Use of Protons in Radiation Therapy Use of Carbon Ions in Radiation Therapy Adjuncts to Irradiation Mutagenesis and Carcinogenesis

Historical Background Radiation has been an ever-present ingredient in the evolution of life on Earth. It is not something invented by human ingenuity in the technologic age; it has always been there. What is new and manmade is the extra radiation to which we are subjected, largely for medical purposes, but also from journeys in high-flying jet aircraft and from the nuclear reactors that are used to generate electrical power. X-rays were discovered in 1895 by the German physicist Wilhelm Conrad Roentgen. He found that “this new kind of ray” could blacken photographic film sealed in a container and stored in a drawer and could pass through materials opaque to light, including cardboard and wood. During a public demonstration of the production of ­  x-rays, Roentgen asked his colleague, Herr Kölliker, to put his hand in front of the x-ray machine, and with a sheet of photographic film, he made the first public radiograph displaying the bony structure of the hand. With this action, Roentgen became the father of diagnostic radiology and radiation physics. Some controversy exists about who was the first to use x-rays therapeutically. In 1897, Professor Freund demonstrated before the Vienna Medical Society the disappearance of a hairy mole by the use of x-rays, and by the turn of the century, x-rays had been used in Europe and America in primitive therapeutic applications. Parallel to the discovery of x-rays, Becquerel discovered radioactivity in 1898. Three years later, he performed what is arguably the first radiobiologic experiment when

he ­inadvertently left a container with 200 mg of radium in his vest pocket for 6 hours. He subsequently described the erythema of the skin that became evident in 2 weeks and the ulceration that developed and required several weeks to heal. During the early 1900s, radiobiologic experiments were conducted with simple biologic systems in parallel with the development of radiation therapy. One of the most wellknown results, which is still cited, was the so-called law of Bergonié and Tribondeau, which states that radiosensitivity is highest in tissues with the highest mitotic index and lowest in differentiated tissues. From 1912 to 1940, several ­investigators, first in Germany and later in the United Kingdom, demonstrated the dependence of radiation response on oxygen. The magnitude of the oxygen effect was determined by using seedlings of Vicia faba, and the possible implications for radiation therapy were discussed. In Paris in the 1920s and 1930s, famous experiments were performed in which the testes of rams were irradiated with x-rays. It proved to be impossible to sterilize the animals in a single dose without causing a severe reaction in the skin of the scrotum, but if the dose was fractionated over several weeks, sterilization could be achieved with little apparent skin damage. It was argued that the testes were a model for a rapidly growing tumor, whereas the skin represented a normal tissue response. On this basis, fractionation was introduced into clinical radiation therapy. Brachytherapy underwent a similar conceptual ­evolution after its first use in the early years of the 20th century. As





PART 1  Principles

fractionation was recognized as being advantageous in external irradiation, protraction in brachytherapy (i.e., using low-activity sources for longer periods) was thought to improve the therapeutic ratio, although radiobiologic studies of dose-rate effects were still decades away. This concept became the hallmark of the “Paris approach” to intrauterine and intravaginal radium therapy for cancer of the ­cervix. In Manchester, England, optimal arrangements of radium sources were sought to achieve a consistent dose rate and a more nearly homogeneous dose distribution through the tumor-bearing volume while sparing surrounding normal structures; this led to a more systematic application of the Paris concepts. Later, afterloading systems were developed in brachytherapy that used nonradioactive applicators so that the radioactive sources were introduced only after the desired relationships of sources were ensured; the afterloading approach reduces radiation exposure to the administering personnel. Large quantities of radium were gathered in some ­centers to produce a telecurietherapy unit (i.e., one that would permit the treatment of patients with gamma rays at a distance) in contrast to the placement of encapsulated radium in body cavities or directly into tumors (i.e., brachycurietherapy). Clinical experience with these units suggested advantages of high-energy radiation over the widely available 200 kilovolt-peak (kVp) x-ray generators. Between 1930 and 1950, technical advances permitted the development of much higher energy x-ray generators. In the 1950s, cobalt 60 (60Co) teletherapy units became widely available, and the first generation of medical linear accelerators was developed. In the wake of World War II and the use of atomic weapons on Hiroshima and Nagasaki, research in radiobiology developed rapidly. Significant milestones include the development of techniques to culture single mammalian cells in vitro in 1956 and to determine survival curves in vivo in 1959. These developments ushered in a greatly enhanced effort in radiation biology to understand conventional ­radiation therapy and suggested new horizons for improvements in treatment. In national laboratories on both sides of the Atlantic, radiation biology studies not related to ­radiation therapy were developing at the same time, involving basic studies of mutagenesis and carcinogenesis.

Physical Basis Types of Radiation Types of radiation of concern in this book are those with the capacity to produce ionizations and excitations during the absorption of energy in biologic material. The raising of an electron in an atom or molecule to a higher energy level, without the ejection of that electron from the atom or molecule, is called excitation. If the radiation has sufficient energy to eject one or more orbital electrons from the atom or molecule, the process is referred to as ionization, and the radiation is said to be ionizing radiation. The important characteristic of ionizing radiation is the localized release of large amounts of energy. The energy dissipated by an ionizing event is approximately 33 electron volts (eV), which is more than enough to break a strong chemical bond; for example, the energy associated with a carbon-carbon bond is 4.9 eV. Ionizing radiation produces substantial biologic effects for the relatively small total

amounts of energy involved because the energy is released locally in “packets” large enough to break chemical bonds and initiate the chain of events that lead ultimately to a biologic effect.

Electromagnetic Radiation Electromagnetic radiation (x-rays and gamma rays) is indirectly ionizing. These types of radiation do not themselves produce chemical and biologic damage, but when they are absorbed in the medium through which they pass, they give up their energy to produce fast-moving electrons by the Compton, photoelectric, or pair-production processes (Fig. 1-1). X-rays and gamma rays are forms of electromagnetic radiation that do not differ in nature or properties; the designation of x or gamma reflects the way in which they are produced. X-rays are produced extranuclearly, which means that they are generated in an electric device that accelerates electrons to high energy and then stops them abruptly in a target, made usually of tungsten or gold. Part of the kinetic energy, or energy of motion of the electrons, is converted into photons of x-rays. Gamma rays are produced intranuclearly (i.e., emitted by radioactive isotopes); they represent excess energy that is given off as the unstable nucleus breaks up and decays in its efforts to reach a stable form.

Particulate Radiation Other forms of radiation that are used experimentally and used or contemplated for radiation therapy include electrons, protons, alpha particles, neutrons, negative ­­ pi­mesons, and high-energy heavy ions. Electrons are light, negatively charged particles that can be accelerated to high energy and to a speed close to that of light by means of an electrical device such as a betatron or linear accelerator. Protons are positively charged particles with a mass nearly 2000 times that of an electron. Protons require more complex and expensive equipment to accelerate them to useful energies. Protons are increasingly used for radiation therapy because of the favorable dose distributions that can be obtained. For example, 160-megaelectron volt (MeV) protons have a range of about 12 cm in tissue. Alpha particles are nuclei of helium atoms, each consisting of two protons and two neutrons in close association. Because they have a net positive charge, they can be accelerated in large electrical devices similar to those used for protons. Alpha particles are also emitted during the decay of some radioactive isotopes. Neutrons are particles having a mass similar to that of protons, but they carry no electrical charge. Because they are electrically neutral, they cannot be accelerated in an electrical device but rather are produced when a charged particle, such as a deuteron or proton, is accelerated to high energy and then made to impinge on a suitable target material. Neutrons are indirectly ionizing because the first step in their absorption is for them to collide with nuclei of the atoms of the absorbing material and produce recoil protons, alpha particles, or heavier nuclear fragments. These charged particles are responsible for the biologic effects. Neutron beams have been used experimentally for radiation therapy at several centers because of their theoretical advantage over photons in low-oxygen (hypoxic) conditions, but interest in the use of neutron therapy has waned

1  Physical and Biologic Basis of Radiation Therapy COMPTON EFFECT

PHOTOELECTRIC EFFECT

e–

Vacancy in k-shell

e– e–

p +np + np + n p + np + np + n

e–

e–

e

id

c In

n

to

ho

e–

Incident photon

e–

e–

p nt

Fast electron

p +np + np + n p + np + np + n

e–

e–

Sc



Fast electron e–

e–

atte

red

pho

e–

ton

Characteristic x-rays PAIR PRODUCTION e–

Incident photon

e– p +np + np + n p + np + np + n

e–

e– e–

e+

Positron electron pair

e– e–

FIGURE 1-1.  The first step in the absorption of a photon of x-rays or gamma rays is the conversion of the energy of the photon into the kinetic energy of an electron, or electron-positron pair. At higher energies, when the energy of the incident photon greatly exceeds the binding energy of the planetary electrons in the atoms of the absorber, the Compton process dominates. The photon interacts with the electron in a classic “billiard-ball collision.” Part of the photon energy is given to the electron as kinetic energy, while the photon is deflected and has reduced energy. At lower energies, when the binding energy of the planetary electrons of the atoms of the absorber is not small compared with the photon energy, the photoelectric effect is most important. The photon disappears completely as it interacts with a bound electron. The electron is ejected with kinetic energy equal to the photon energy, less the energy required to overcome the electron bond. The vacancy caused by the removal of the electron must be filled by an electron dropping from an outer orbit, giving rise to a photon of characteristic radiation. At sufficiently high photon energies, the photon may interact with the powerful nuclear forces to produce an electron-positron pair. The first 1.02 MeV of photon energy is used to create the rest mass of the pair, and the remainder is distributed equally between them as kinetic energy.

because it has shown no advantages in terms of outcome over irradiation with other types of particles. Negative pi-mesons are negatively charged particles with a mass 273 times that of an electron. They are produced by a complex process that necessitates a huge linear accelerator or synchrocyclotron capable of accelerating protons to energies of 400 to 800 MeV. When pi-mesons are absorbed in biologic material, they behave like overweight electrons as long as they are relativistic (i.e., as long as their velocity is close to that of light). However, when they slow down, they spiral down the energy levels of an absorbing atom and are absorbed by the nucleus of that atom, which then explodes to produce fragments consisting of neutrons, alpha particles, and larger nuclear fragments. Radiation therapy with negative pi-mesons has been attempted experimentally in several countries but has been abandoned. Heavy ions are nuclei of elements such as nitrogen, carbon, neon, argon, or silicon that are positively charged because some or all of their planetary electrons have been stripped from them. To be useful, they must be accelerated to energies of thousands of millions of volts and can therefore be produced in only a very limited number of laboratories in the world. Most of the interest in heavy ions for biologic applications has focused on carbon ions, which combine the dose-distribution advantages of protons with an increase in biologic effectiveness toward the end of the

particle range. This topic is discussed further in the “Use of Carbon Ions in Radiation Therapy” section later in this chapter.

Production of Radiation for Therapeutic Applications When x-rays or gamma rays enter biologic material, energy is converted into chemical damage and heat. At the energy levels of most x-ray and gamma-ray sources in use, the primary events are the interactions of photons with electrons in the outer shells, resulting in scattering of the photons and electrons (i.e., Compton scattering). With higher-­energy photons, secondary electrons scatter more in the forward direction (i.e., in the direction of the primary beam). It takes some distance for the interactions to summate and reach a maximum, after which the energy of the beam dissipates by a constant fraction per unit depth. Figure 1-2 compares depth-dose characteristics of several radiation beams commonly used in radiation therapy. The insert in this figure demonstrates the physical basis for skin sparing; high-energy photons, unlike conventional x-rays, deposit the maximum dose below the skin surface. The most commonly used sources for external irradiation are listed in Table 1-1. Much overlap exists among the teletherapy sources. For example, there is relatively little difference in depth doses between the gamma rays from 60Co and 2- to 6-megavolt (MV) x-rays. However, the



PART 1  Principles 100

100 Percent depth dose

Protons with range shifter

90

Percent depth dose

80 70 Monoenergetic protons

60

80

2 mm Cu

6 MV

18 MV

60 40 20

50

0 0 2 4 6 8 10 12 14 16 18 20

40

Depth (mm)

18 MV

30 20

12 6 MeV MeV

10 0 0

5

10

15 Depth (cm)

6 MV

D-T neutrons

20 MeV

60Co

2 mm Cu 20

25

30

FIGURE 1-2.  Percentage depth-dose curves are shown for a variety of radiation types used in radiation therapy. These include xrays and γ-rays up to 18 MV, electrons of 6 to 20 MeV, and protons. The inset shows the pattern of absorption at shallow depths and provides a rationale for the skin-sparing effect. Cu, copper; MeV, megaelectron volt; MV, megavolt.

TABLE 1-1.    Teletherapy Sources

TABLE 1-2.    Therapeutic Isotopes

Unit

Mean Energy (MeV) of Photons

Isotope

150-440 kVp x-ray

0.06-0.14

137CS

0.66

60Co

teletherapy

teletherapy

1.25

4 MV linear accelerator

1.3

6 MV linear accelerator

1.8

20-24 MV betatron and linear ­accelerator

6.2-7.0

kVp, kilovolt peak; MeV, million electron volts; MV, megavolts.

edge of the beam produced by a linear accelerator is much sharper than that from cobalt units, which may be an important factor when radiation is delivered close to critical structures, such as the lens of the eye. The sources most widely used for intracavitary or interstitial therapy are listed in Table 1-2. The various brachytherapy sources have overlapping capabilities, but some have specific advantages. For example, iridium 192 (192Ir) is the only isotope listed that has been widely and satisfactorily used with an afterloading technique for interstitial therapy. Radium and cesium can be used in afterloading intracavitary applicators. Gold, iodine, and palladium can be used for permanent interstitial implants. When used appropriately and meticulously, teletherapy units and brachytherapy sources have had great success in the permanent control of a variety of malignant diseases. However, there is nothing inherent in any of these x-ray or gamma-ray generators that can ensure success independent of the expertise of the radiation oncologist.

The Measurement and Meaning of Dose Quantification of dose has been an important factor in the development of modern radiation therapy. The ­ability to prescribe and deliver specific amounts of radiation

Half-life

Energy (keV)

226Ra

1620 yr

830*

137Cs

30 yr

662*

198Au

2.7 days

412*

192Ir

73.8 days

370*

125I

60 days

28*

103Pd

16.97 days

21*

Photon

Beta 32P

14.3 days

1710†

90Sr/90Y

28.5 yr/2.7 days

550/2280†

188W/188Re

69.4 days/17 hr

350/2120†

186Re

3.8 days

1070†

62Zn/62Cu

9.3 hr/9.7 min

660/2930†

133Xe

5.2 days

360†

131I

8.0 days

600†

89Sr

50.5 days

1495†

166Ho

26.8 hr

1850†

*Average

energy. energy. keV, kiloelectron volt.

†Maximum

r­ epeatedly with precision and accuracy is essential if a ­given treatment is to be duplicated and if clinical data are to be accumulated and compared. Biologic effects in laboratory research and in clinical practice correlate well with absorbed dose, expressed as energy per unit mass of tissue. This correlation holds true for all types of radiation used in conventional radiation therapy, such as x-rays, gamma rays, fast neutrons, and electrons, but it is probably less satisfactory for very-high-energy heavy ions.

1  Physical and Biologic Basis of Radiation Therapy

For many years, the unit of dose commonly used was the rad, defined as an energy absorption corresponding to 100 erg/g. This unit has been replaced by the gray (Gy), defined as energy absorption of 1 J/kg. One gray is equivalent to 100 rad.

Measurement of Dose Dose can be measured directly, in terms of energy absorbed per unit mass, by measuring the temperature rise in the absorbing material. However, this can be done in only a few research centers because doses of several grays result in a temperature rise of only a few hundredths of a degree Celsius, which is exceedingly difficult to measure. For practical day-to-day use, what is measured is the ability of the x-rays or gamma rays to ionize air because such measurements are sensitive and relatively easy to make. Absorbed dose is then calculated on the basis of the ionization measurements. An ionization chamber consists of a cavity containing a gas that is surrounded by tissue-equivalent material. A central electrode within the cavity is maintained at a high ­­voltage relative to the cavity walls. Because the gas is an insulator, no charge passes across the gas until x-rays or gamma rays pass through the cavity and produce ionizations. The collection of the charges produced leads to the current between the central electrode and cavity walls, which can be detected with a sensitive instrument. In practice, most of the ionizations occur in the cavity wall and result in electrons that pass through the cavity. If the ionization chamber is connected across a capacitor, the dose meter is an integrating dose meter (i.e., it measures the ­total amount of incident radiation) (Fig. 1-3). If the chamber is connected across a



resistor and the current through that resistor is measured, the dose meter measures the instantaneous dose rate.

Patterns of Energy Deposition: Microdosimetry Dose is an average quantity (i.e., the average energy deposited per unit mass of tissue). However, at a microscopic level, energy from ionizing radiation is not deposited uniformly but tends to be localized along the tracks of charged particles. What is critical to the biologic effect produced by a given amount of radiation is the pattern of energy ­deposition. Figure 1-4 shows the tracks of secondary charged particles in cells irradiated with 1 centigray (cGy) of x-rays or 1 cGy of neutrons. In the case of x-rays, the charged particles set in motion are electrons, whereas in the case of neutrons, the secondary charged particles are protons.  A marked difference exists between the two. A dose of  1 cGy of x-rays results in essentially all cells being traversed by several particles. For neutrons, only a few cells are traversed by a single particle. Consequently, when the dose of x-rays is doubled, the average energy per cell is increased; however, in the case of neutrons, the number of cells traversed is increased, but the amount of energy per cell traversed does not change. This is an essential difference ­between the two types of radiation. Biologic effects correlate with dose for a given type of radiation (i.e., a bigger dose leads to a bigger biologic effect), but the same dose of a different type of radiation does not produce equal biologic effects because of the considerations referred to earlier. In other words, the different relative biologic effectiveness (RBE) of different types of radiation is largely a reflection of the different patterns of energy deposition.

Physical Basis for Choice of Treatment Techniques

DOSE-RATE METER Thimble chamber

R

INTEGRATING DOSE METER Thimble chamber

C

FIGURE 1-3.  An ionization chamber consists of a cavity fabricated from tissue-equivalent plastic and filled with an appropriate gas. A voltage is applied between the central electrode and the wall of the cavity. The gas is an insulator so that no current flows unless x-rays pass through the gas and ionize the gas into positive and negative ions that are collected by the voltage across the chamber. The charge generated is proportional to the x-ray intensity. If the charge is passed through a resistor (R), the voltage generated across it is a measure of the instantaneous dose rate. It is then a dose-rate meter. If the charge is passed onto a condenser (C), the change in voltage across the condenser is a measure of the total dose. It is then an integrating dose meter.

The determinants for choosing particular techniques in clinical radiation therapy are the limits of localization of the malignant tumor and especially its subclinical extensions and the availability of particular sources of radiation. Except for the few institutions that have accelerators ­capable of producing beams of heavy charged particles, treatment choices are based on the external beams listed in Table 1-1 or the intracavitary sources listed in Table 1-2. The particular techniques that are suitable for individual malignant tumors are found in chapters elsewhere in this book, but a few generalizations can be made here. The intent when administering ionizing radiation for the purpose of killing all cells in a malignant tumor is to deliver a dose that is sufficient to eradicate these cells but is limited with respect to the surrounding normal tissues. In light of the preceding discussion of dose and its meaning, it is worth considering the appropriate designations of dose in the clinical setting. The term tumor dose is inappropriate because dose distribution is rarely homogeneous throughout the tumor-containing volume. It is more appropriate to refer to minimum tumor dose because this dose determines control of the tumor. Similarly, it is most meaningful to refer to maximum doses in specific normal tissues surrounding the tumor because these doses determine the adverse effects of radiation therapy. In all cases, it is virtually meaningless to express a total dose without qualifying it by the fractionation regimen used (fractionation is discussed further later in this chapter).



PART 1  Principles 10-mGy (1 rad) x-rays

10-mGy (1 rad) neutrons

FIGURE 1-4.  When a population of cells is irradiated with 1 cGy of x-rays, many charged particle tracks traverse every cell. In the case of neutrons, only a small proportion of the cells are traversed by a charged particle track. Consequently, when the x-ray dose is increased, the average energy deposited per cell is increased. When the neutron dose is increased, the number of cells in which energy is deposited is increased. This is a fundamental difference between x-rays and neutrons.

A more detailed discussion of dose distributions and the related subject of selection of single fields of specific radiation beams or the use of combinations of fields is beyond the scope of this chapter, but these topics are addressed in standard texts. One principle of choosing treatment techniques can be discussed here. Given the contemporary availability of radiation beams, computers, and treatment-planning software, the temptation is to seek refinements of dose distribution that are conceptually satisfying but have little or no ­clinical importance. Such temptations are best resisted in the interest of daily reproducibility. To paraphrase a ­concept ­ espoused by the 14th-century philosopher, William of Occam, “complexity is not to be assumed without necessity.”

Cancer Biology Cancer is thought to be a clonal disorder because most types of cancer seem to arise from a single cell that has suffered a disruption in its regulatory mechanisms for proliferation and self-elimination. Malignant cells differ from their normal tissue counterparts in several ways. First, they are often immortal or at least have the ability to divide many more times than do cells from normal somatic tissues. Second, they often grow more rapidly than do cells of the same tissue of origin. Third, they have abnormal cellto-cell interactions, which result in their being able to invade, metastasize, and grow in foreign environments.

Mechanisms of Carcinogenesis Carcinogenesis seems to be a multistep process, with multiple genetic alterations occurring over extended periods. Sometimes, genetic alterations are carried in the germline and form cancer-predisposing syndromes (described later); however, heritable mutations are the exception. Most of

the alterations that lead to cancer are acquired in the form of somatic mutations. Cell proliferation is controlled through a series of positive and negative signals that affect cell division and differentiation. The acquisition of tumorigenicity results from changes that affect these signal-control points. The conversion of a cell to a malignant state can result from gain-of-function mutations that activate oncogenes, which are positive growth regulators; loss of function of a tumor suppressor gene product that normally acts as a negative growth regulator; or mutations that affect death-regulating genes that control apoptosis (i.e., programmed cell death). Oncogenes are found in a mutated or abnormally expressed form in many human cancers.1 Oncogenes can be activated by a point mutation, by a chromosomal rearrangement such as a translocation, or by gene amplification. Some human leukemias and lymphomas seem to be caused by specific chromosomal translocations that lead to oncogene activation in several different ways. Because oncogenes are gain-of-function mutations, only one of the two copies needs to be activated for their effects to be ­expressed; oncogenes act in a dominant (rather than a recessive) fashion. The concept of tumor suppressor genes arose from the observation that the in vitro fusion of a normal human ­fibroblast with a HeLa cervical carcinoma cell suppressed the expression of the malignant phenotype of the HeLa cell. Of the many tumor suppressor genes whose location and function are known, two of the most intensively studied are TP53 (formerly designated p53) and RB1 (formerly designated Rb).2,3 In contrast to oncogenes, tumor suppressor genes involve loss-of-function mutations, and both copies of the gene must be lost for their effects to be expressed (i.e., they act in a recessive fashion). In practice, the loss of both copies of a gene often occurs by somatic ­homozygosity, which is the process by which one chromosome of a pair is

1  Physical and Biologic Basis of Radiation Therapy

lost, a deletion occurs in the remaining chromosome, and the chromosome with the deletion replicates. A third class of genes associated with some forms of cancer affects apoptosis. The most relevant example is the BCL family of genes, which includes BCL2, BCL2L1 (formerly designated BclX), and BAX. BCL2 is deregulated by the chromosomal translocation t(14;18), which blocks apoptosis and contributes to the development of follicular lymphoma.

The Multistep Nature of Cancer The idea that carcinogenesis is a multistage process in which several distinct events occur, separated in time, is almost 70 years old. Skin cancer experiments in mice led to the development of the concepts of initiation, promotion, and progression as being stages in tumor development.  Genetic analysis of cells from solid tumors also suggests that multiple alterations, mutations, or deletions are ­needed in multiple signaling genes for cancer to form. In the case of colorectal cancer, a model has been proposed that correlates a series of 6 to 12 chromosomal and molecular events producing changes in the histopathology of normal epithelium during the formation and metastasis of colorectal cancer.4 These events are thought to involve a “mutator” phenotype in which multiple mutations are likely in cancer-associated genes (i.e., oncogenes and tumor suppressor genes). This leads to progression of the cancer and ultimately to its invasive and metastatic properties.

Functions of Oncogenes and Tumor Suppressor Genes The identification of genes associated with human ­cancer quickly led to studies of the function of these genes. Cells must repeatedly “decide” whether to divide, differentiate, or undergo apoptosis. Because all three outcomes affect cell number, it is not surprising that these decision pathways are the primary targets of oncogenes or tumor ­suppressor genes. Most tumor suppressor genes can be classified as gatekeepers or caretakers. Gatekeepers are genes that directly regulate the growth of tumors by inhibiting cell division or by promoting cell death. The function of these genes is rate limiting for tumor initiation and growth; both  alleles ­(maternal and paternal) must be lost or inactivated for a tumor to develop. Predisposed individuals inherit one damaged copy of such a gene and require only one additional mutation for tumor initiation. The identity of gatekeeper genes varies among tissues, such that inactivation of a given gene leads to predisposition to specific forms of cancer. For example, inherited mutations in the APC gene lead to ­colon cancer, and mutations in the VHL gene predispose an individual to develop kidney cancer. ­Because these gatekeeper genes are rate limiting for tumor ­initiation, they ­often tend to be mutated in sporadic cancer through somatic mutations and in the germlines of predisposed individuals. Inactivation of caretaker genes, in contrast, does not directly promote the growth of tumors but leads instead to genomic instability, thereby indirectly promoting growth by causing an increase in mutation rate. Such an increase in genetic instability can greatly accelerate the development of cancer, especially those types—such as colon cancer—that require numerous mutations for their full ­ development.



The targets of the accelerated mutation rate in cells with defective caretakers are the gatekeeper tumor suppressor genes or oncogenes, or both.

Signal Transduction An elaborate system of biochemical communication networks exists in eukaryotic cells to provide information to the nucleus regarding the environment outside the cell. Signal transduction is the term used to describe the flow of information from the cell’s outer membrane through the cytoplasm and into the nucleus. Multiple signal transduction cascades can operate simultaneously, with variable amounts of cross-talk occurring between them. Even a weak signal can elicit a powerful response because it is amplified by biologic processes in much the same way as a radio signal is amplified by an electronic device. Signals in the form of cytokines, low-molecular-weight hormones, growth factors, and other proteins that arrive at the outer plasma membrane can affect the cell by two mechanisms. In the first, the signaling factor binds to a specific cell surface receptor that spans the plasma membrane. On binding, these receptor molecules transduce a signal from the extracellular domain to the cytoplasmic domain on the inner surface of the plasma membrane; this transduction often leads to the activation of an enzyme such as a protein kinase. The second mechanism involves the migration of the signaling agent itself across the plasma membrane, followed by its binding to specific receptor molecules in the cytoplasm. In either case, a biochemical cascade ultimately ­allows the signal to reach its destination—the nucleus—and ­alter cellular behavior. These signal transduction processes involve signaling molecules that are bound together by ­protein-protein interactions and by the production of ­second-messenger molecules. One important signaling pathway involves GTP-binding proteins such as those produced by the RAS family; the RAS gene is often mutated in human cancer. Another important pathway is that associated with protein kinase C. Ionizing radiation can activate nuclear and membrane-cytoplasmic signal-transduction pathways; damage to deoxyribonucleic acid (DNA) modulates the transcriptional activity of TP53 through the activity of several kinases and by protein-protein interactions. These and other perturbations to signaling pathways can produce cell cycle arrest or apoptosis.

The Cell Cycle The ability of cells to produce exact, accurate copies of themselves is essential to the continuance of life; it is accomplished through highly organized processes that have been well conserved through evolution. Lack of fidelity in cellular reproduction, as manifested by alterations in DNA and chromosomes, is a hallmark of cancer. The division of the mammalian cell cycle into its constituent phases  (M, G1, S, and G2) was accomplished by radiation biologists in the 1950s. As detailed under “Tissue and Tumor Kinetics,” it is now known that progression through the cell cycle is governed by protein kinases, which are activated by cyclins. Each cyclin protein is synthesized at a discrete phase of the cycle—cyclins D and E in G1, cyclin A in S and G2, and cyclin B in G2 and M. Transitions in the cycle occur only when a given kinase activates the proteins required for progression.

10

PART 1  Principles

The arrest of cells at various positions in the cycle by the action of “checkpoint genes” is an important response to DNA damage. The two principal checkpoints are the G1/S and the G2/M boundaries, but a checkpoint also exists in S. G2/M is the best-known checkpoint after radiation damage; cells pause at G2/M to repair damage before initiating the complex process of mitosis. The hallmark of cancer is a lack of the ability to respond to signals that would normally cause the cell to stop progressing through the cycle and dividing. Of the tumor suppressor genes discovered so far in human cancers, only RB1 and TP53 seem to be directly implicated in cell cycle control; clearly, much more remains to be learned on this topic.

Cancer Genetics The link between cancer risk and genetic factors varies among tumor types. Up to 30% of rare childhood cancers occur in individuals who are predisposed to the condition, whereas only about 5% to 10% of common adult cancers are thought to be hereditary. Many instances of cancer predisposition are characterized by a distinctive pattern involving several different types of cancer in several different organs, sometimes accompanied by unusual physical features. In general, a family history of cancer, early age at onset, or the presence of multiple primary tumors strongly suggests the presence of a genetic disorder. Germline mutations can result in a predisposition to cancer in any of four ways. Most cases of hereditary cancer predisposition can be traced to germline mutations in tumor suppressor genes. (Tumor suppressor genes were first discovered through a study of familial predisposition to retinoblastoma.) Germline mutations that activate oncogenes are a second possible mechanism for hereditary predisposition to cancer, but the only disorders suspected of arising through this mechanism are multiple endocrine neoplasias. Medullary thyroid carcinoma and pheochromocytoma are a feature of these autosomal dominant syndromes, and physical abnormalities often accompany the malignancy. Third, any syndrome that includes DNA-repair defects is likely to increase the probability of mutations in genes that lead to cancer. Xeroderma pigmentosum, ataxia telangiectasia, Bloom syndrome, and Fanconi anemia are examples of relatively uncommon autosomal recessive syndromes that involve faulty DNA repair or replication.  A particularly interesting member of this class is hereditary nonpolyposis colon cancer, which is caused by a mutation in one of five mismatch repair genes; a mutation in one of these genes leads to a general instability known as the mutator phenotype. A fourth, less well understood category of hereditary disorders predisposes to cancer through unusual sensitivity to common carcinogens. For example, lung cancer is usually considered to have only a small genetic component, but genetic variations in the metabolism of the carcinogens in tobacco smoke may be important in determining who develops a malignancy. Genetic variations also may influence susceptibility to hepatic cellular carcinoma caused by aflatoxins. Several hereditary disorders are known to predispose to cancer, but understanding of those disorders at the molecular level is at various stages. Retinoblastoma displays the clearest pattern of a familial and sporadic form. Some researchers speculate that when all of the information is in, all common cancers will turn out to have familial and

sporadic components but that the genetics will be more complicated and the pattern far less obvious than those of retinoblastoma.

Biologic Basis of Radiation Therapy Radiation Chemistry Strong circumstantial evidence exists to suggest that DNA is the principal target for radiation damage that leads to the loss of reproductive integrity of a cell, although the discovery of the bystander effect has shown that damage can be passed from a “hit” cell to a “nonhit” neighboring cell through gap junction communications.5-8 When x-rays or gamma rays are absorbed in biologic material, the first step in the absorption process is the conversion of their energy to fast-moving recoil electrons. When neutrons are absorbed, the first step is the conversion of that energy into protons, alpha particles, or heavy nuclear fragments. Radiochemical lesions in the DNA may result from the direct action of these charged particles or from indirect action mediated by highly reactive free radicals (Fig. 1-5). Absorption of energy from direct action requires ionizations to occur within the DNA molecule itself. This is

INDIRECT ACTION n

to

o Ph

H OH• S A TS H P P G SP S C P S T S A P P S G CS

O e– p+

n

to

o Ph e– p+

2 nm 4 nm

DIRECT ACTION

FIGURE 1-5.  Direct and indirect actions of radiation are depicted. The structure of DNA is shown schematically. In direct action, a secondary electron resulting from absorption of an x-ray photon interacts with the DNA to produce an effect. In indirect action, the secondary electron interacts with, for example, a water molecule to produce a hydroxyl radical (OH·), which produces the damage to the DNA. The DNA helix has a diameter of about 20 Å (2 nm). It is estimated that free radicals produced in a cylinder with a diameter double that of the DNA helix can affect the DNA. Indirect action is dominant for ­sparsely ionizing radiation, such as x-rays. A, adenine; C, ­ cytosine; G, guanine; P, phosphorus; S, sugar; T, thymine. Modified from Hall EJ, Giaccia AJ. Physics and chemistry of radiation absorption. In: Radiobiology for the Radiologist. 6th ed. Philadelphia: Lippincott Williams & Wilkins, 2006, pp 5-15.

11

1  Physical and Biologic Basis of Radiation Therapy

the dominant process for radiation of high linear energy transfer (LET), such as neutrons or alpha particles. LET is discussed further under “Radiation Quality.” Alternatively, the radiation may interact with other atoms or molecules in the cell, particularly water, to produce free radicals that can diffuse far enough to reach and damage the critical targets. A free radical is an atom or molecule carrying an unpaired electron in its outer shell. The most important of these is probably the hydroxyl radical (OH•). Free radicals have a lifetime of about 10−5 seconds, which is long relative to the time for the initial interaction of the radiation with the atoms and molecules of the absorber (10−15 seconds). Experiments in which free-radical scavengers are added to biologic material during irradiation suggest that about two thirds of the damage produced by x-rays or gamma rays is mediated by free radicals. In the case of radiation of higher LET, the relative balance moves away from indirect action in favor of direct action. The component of the radiation damage mediated by free radicals is readily amenable to modification by the presence or absence of oxygen or other chemical compounds. In contrast, direct action cannot be modified by chemical means.

DNA Breaks DNA is a large molecule assembled as a double helix. Its two helical strands are held together by hydrogen bonds between the bases. The “backbone” of each strand consists of alternating sugar-phosphate groups. The sugar involved is deoxyribose. Attached to this backbone are four bases, the sequence of which specifies the genetic code. Two of the bases—thymine and cytosine—are single-ring groups (i.e., pyrimidines); the other bases—adenine and guanine— are double-ring groups (i.e., purines). The structure of a single strand of DNA is illustrated in Figure 1-6. Bases on opposite strands must be complementary; adenine pairs with thymine, and guanine pairs with cytosine (Fig. 1-7A). When cells are irradiated with x-rays, many breaks occur in single strands. These breaks can be observed and scored as a function of dose if the DNA is denatured and the supporting structure is stripped away. However, in intact DNA, single-strand breaks are of little biologic

consequence as far as cell death is concerned because  single-strand breaks are readily repaired using the opposite strand as a template (see Fig. 1-7B). If the repair is incorrect (i.e., misrepair), it may result in a mutation. If both strands of the DNA are broken but the breaks are well separated (see Fig. 1-7C), repair again occurs readily because the two breaks are handled separately. If the breaks in the two strands are opposite one another or are separated by only a few base pairs (see Fig. 1-7D), this may lead to a double-strand break (DSB) (i.e., the piece of chromatin snaps into two pieces). When this occurs as the result of the passage of a charged particle, considerable damage can be produced in the region, producing a “locally multiple damaged site.”6 A DSB is thought to be the most important lesion produced in chromosomes by radiation; as described in the next section, the interaction of two DSBs may result in cell death, mutation, or carcinogenesis.

Chromosomal Aberrations The most important of the chromosomal aberrations are exchange-type aberrations, which are formed when a DSB in each of two chromosomes, or chromosome arms, rejoins in an illegitimate way. A detailed discussion of cytogenetics is beyond the scope of this chapter; a variety of complex aberrations can occur, but in this chapter, attention focuses on four types of aberrations that are of special importance (Fig. 1-8). The consequences of DSBs in two separate prereplication chromosomes are shown in Figure 1-8A and B. In Figure  1-8A, the rejoining occurs in such a way as to produce a

G

A

G

B

P

O

CH2

O Sugar (deoxyribose)

O O

Phosphate

C

G

O O

C

P

Adenine

C

O

H O

Cytosine CH2

O O O

C

P

G

O

D

H O

Thymine CH2 O

O

O O

P O

H O

CH2

Guanine O

OH

H

FIGURE 1-6.  Structure of a single strand of DNA.

C

C

T

T

G

A

A

C

T

T

G

A

C

T

G

A

C G

A T

A T

T

A T

T A

T A

C G

C G

C G

A

A

G

T

T

C

A

A

G

T

T

C

T A

T

A T

A

C

T

T

G

A

G

A

A

T

A T

G C

T A

FIGURE 1-7.  Single- and double-strand breaks are caused by radiation. A, Two-dimensional representation of the normal DNA double helix. The based pairs carrying the genetic code are complementary; adenine pairs with thymine, and guanine pairs with cytosine. B, A break in one strand is of little significance because it is readily repaired, using the opposite strand as a template. C, Breaks in both strands, if well separated, are repaired as independent breaks. D, If breaks occur in both strands and are directly opposite each other or are separated by only a few base pairs, a double-strand break may occur in which the chromatid snaps into two pieces.

12

PART 1  Principles Translocation

Chromosome 9

Chromosome 22

A

Dicentric

B FIGURE 1-9.  A symmetric translocation between chromosomes 9 and 22 may bring together the BCR and ABL1 genes, a change associated with chronic myelogenous leukemia. Modified from Hall EJ, Giaccia AJ. Cancer Biology. In: Radiobiology for the Radiologist. 6th ed. Philadelphia: Lippincott Williams & Wilkins, 2006, pp 274-302.

C Ring

Dic

D Trans Deletion

FIGURE 1-8.  Four types of chromosomal aberrations can ­occur. A, If breaks occur in two prereplication chromosomes, a symmetric exchange may occur. This is consistent with cell viability, but the process can lead to activation of an oncogene. B, Breaks in two prereplication chromosomes may reconstitute to form a dicentric and an acentric fragment. This aberration is lethal to the cell at a subsequent division. C, Breaks in the two arms of a prereplication chromosome may rejoin to form a ring and an acentric fragment. This aberration is lethal to the cell. D, Two breaks in the same arm of a prereplication chromosome may result in a deletion. If the deletion is small, it may be consistent with cell viability. If the deleted piece of DNA includes a suppressor gene, it may result in carcinogenesis.

FIGURE 1-10.  Fluorescence in situ hybridization (FISH) uses probes to “paint” a given chromosome with a characteristic color. Multiplex-FISH paints each chromosome with a different color, enabling symmetric translocations to be then readily visualized. (Courtesy of Charles R. Geard, MD, Columbia University, New York, NY.)

symmetric translocation. This is compatible with cell viability but is occasionally associated with carcinogenesis. For example, if breaks occur in just the right places on chromosomes 9 and 22, the symmetric translocation that occurs may bring together the genes BCR and ABL (Fig. 1-9). This translocation is associated with chronic myelogenous leukemia. Symmetric translocations and other chromosomal abnormalities can be visualized in human cells by the technique of multiplex fluorescent in situ hybridization (FISH), in which the chromosomes are “painted” in different colors (Fig. 1-10). In Figure 1-8B, the two broken chromosomes rejoin in such a way as to produce a dicentric fragment (i.e., chromosome with two centromeres) and an acentric fragment. The acentric fragment will be lost at a subsequent mitosis 

because it has no centromere and therefore cannot be segregated ­ correctly to the daughter cells. The formation of a dicentric fragment is usually lethal to the cell. Another lethal aberration is the ring chromosome, the formation of which is illustrated in Figure 1-8C. In this case, a DSB is produced in each arm of the same prereplication chromosome. The rejoining of the “sticky” ends may result in an acentric fragment and a ring. This aberration is usually not consistent with cell viability. ­Figure 1-8D illustrates the formation of an interstitial deletion in which two DSBs occur close together in the same arm of the same chromosome. This may lead to the loss of a section of genetic material, which may not affect cell viability. However, an interstitial deletion may be a mechanism of carcinogenesis if the deleted genetic material happens to be a tumor ­  suppressor gene.

r+f

1  Physical and Biologic Basis of Radiation Therapy

Cell Survival Curves A cell survival curve describes the relationship between the absorbed dose of radiation and the proportion of cells that survive in the sense that they are able to grow into a colony, thereby demonstrating retention of their reproductive integrity.10 With modern tissue culture techniques, it is possible to establish cell lines from many different tissues and tumors. If these cells are seeded as single cells and exposed to radiation or to some other cytotoxic agent, it is then possible to count the proportion of cells that are able to form macroscopic colonies after graded doses of the cytotoxic agent. Cell survival curves for mammalian cells have a characteristic shape. If surviving fraction is plotted on a logarithmic scale against dose on a linear scale, the curve has an initial slope at low doses, at which the surviving fraction seems to be an exponential function of dose (Fig. 1-11). As the dose is increased, the curve bends and becomes progressively steeper. At very high doses, the curve tends to become straight again (i.e., the surviving fraction returns to being an exponential function of dose). Over the first few decades of survival, survival curves for mammalian cells closely approximate that of a linear quadratic form expressed as follows: S  eBD  CD

2

In the equation, S is the fraction of cells surviving a dose (D), and α and β are constants. Over a wider range of doses, a more complex relationship that contains at least three parameters usually proves to be a better fit. The parameters include an initial slope (D1), a final slope (D0), and some quantity that is a measure of the width of the shoulder. This quantity can be the extrapolation number (n) or the quasi-threshold dose (Dq). For densely ionizing radiation, the shoulder of the survival curve tends to disappear, and the surviving fraction then approximates an exponential function of dose over the entire dose range. For a wide range of cells derived from normal tissue or from tumors investigated in vitro, the D0

α Cell kill

Damage

β Cell kill

2

S = e−αD − βD

α /β Dose per fraction

FIGURE 1-11.  Dose-response curves for mammalian cells are adequately fitted by the linear-quadratic relationship, at least over the range of doses of concern in radiation therapy. The form of the equation is S = e−αD  −βD2, in which S is the fraction of cells surviving a dose D, and α and β are constants. Cell killing by the linear and quadratic terms is equal when αD = βD2. This occurs when the dose D = α:β.

13

tends to be within the range of 1 to 2 Gy and the extrapolation number, n, varies from little more than 1 to about 20.

Effects on Normal Tissues Clonogenic Survival End Points Although normal tissue cells can be difficult to maintain in culture for long periods, cell survival curves for normal tissues have occasionally been obtained through the use of ingenious techniques. The first example historically was for the survival of irradiated skin cells in the mouse. In this technique, devised by Withers,11-13 a superficial x-ray machine is used to irradiate a circular annulus to the massive dose of 30 Gy to produce a moat of sterilized cells. In the center of this moat is an isolated island of intact skin, protected during the first exposure by means of a small metal sphere. This small area of intact skin is then given a test dose, D, and subsequently observed for the regrowth of a nodule of skin. If one nodule regrows in one half of the areas exposed, it indicates that a single cell survived. By varying the area of skin irradiated and dose of the radiation, it is possible to calculate the number of cells per square centimeter surviving various doses of radiation and to construct a cell survival curve. When this is done, the D0 turns out to be about 1.35 Gy. Because the quantity scored is not the surviving fraction but rather the number of surviving cells per square centimeter of skin, no extrapolation number can be obtained. Instead, a second dose-response curve is constructed for split doses in which the radiation is administered in 2 equal fractions 24 hours apart. The horizontal separation between the dose-response curve for single and split doses is then Dq. This value is about 3.5 Gy for mouse skin and is very similar for human skin. Another normal tissue for which cell survival curves can be obtained is the “surviving crypts in the mouse jejunum system.” The intestinal epithelium represents a classic selfrenewing tissue. The crypts at the base of the villi contain dividing cells, which pass up the villi as they differentiate and become functional cells. A dose of radiation of several Gy sterilizes a proportion of the dividing cells in the crypts but does not affect the mature, differentiated, functioning cells in the villi. After a dose of radiation, the villi shrink because no replacement cells from the crypts exist to take the place of those sloughed away from the tips of the villi. If the test animal is killed 3 or 4 days after the irradiation and sections are made of the jejunum, the number of regenerating crypts can be counted, giving a measure of surviving cells per circumference. This value can then be plotted as a function of dose. It turns out with this system that the D0 is about 1.3 Gy, but the most interesting fact is that the Dq is very large, indicating the presence of a large amount of accumulation and repair of sublethal damage. A third normal tissue system with which cell survival curves can be obtained was developed by McCullough and Till.14 In this technique, a suspension of nucleated bone marrow cells is prepared from the bone marrow of a donor mouse and injected into a recipient mouse. Some of the circulating cells lodge in the spleen and produce nodules that can be counted after they have grown for 9 or 10 days; in effect, the investigator counts nodules in the spleen instead of colonies in a Petri dish. By comparing the number of spleen colonies resulting from cells taken from irradiated and unirradiated donor animals, the fraction of cells ­surviving a given dose can be calculated. The cell survival

14

PART 1  Principles

curve for these stem cells closely approximates an exponential function of dose (i.e., it has little or no shoulder, and the D0 is about 0.95 Gy). These cells are radiosensitive because they die an apoptotic rather than a mitotic death. Dose-response curves for these three normal tissue systems are shown in Figure 1-12.

Functional End Points Although dose-response curves for clonogenic cell survival can be obtained in a limited number of circumstances, in many other situations, it is more practical to obtain a doseresponse relationship for the loss of function of an organ or tissue. Organ function is related more to the proportion of functional cells remaining in an irradiated organ at a particular time than to the proportion of clonogenic, or stem, cells. The functional cell, not the stem cell, data determine tolerance doses in multifraction radiation therapy. ­Functional dose-response curves can be constructed from multifraction experimental data by making several simple assumptions: (1) the linear-quadratic model is an adequate fit of the dose-response data, at least up to doses of a few Gy; (2) the response of the tissue or organ is the same to each dose in a multifraction regimen; (3) the repair of ­sublethal damage is complete between multiple dose fractions; and (4) no cell division occurs during the ­fractionated regimen. The basis of the method is to expose groups of animals to various numbers of fractions, with the dose per fraction titrated to produce a given functional impairment. The reciprocal of the total dose is then plotted against the dose per fraction, and the values of α and β corresponding to the linear-quadratic form of the dose-response relationships can then be inferred from the intercept on the ­abscissa and from the slope of this plot. An example is shown in ­Figure 1-13.15 The extent to which the experimental data fit this

straight-line plot, which is a transcription of the ­ linear­quadratic curve, is an indication of how well the linear­quadratic relationship fits the basic dose-response data. Acute functional effects for which multifraction techniques have been used to obtain the parameters of the dose-  response relationship include the following: Acute desquamation in mouse, rat, and pig skin Human skin “tolerance” Callus formation after bone breakage in mice Mouse tail necrosis Mouse LD50/30 (i.e., dose that is lethal to 50% of the animals within 30 days) Rat tail necrosis For these acute effects, the ratio of α:β is usually close to  10 Gy, with the ratio being greater for the LD50 end points. Late functional effects for which multifraction experiments have been performed to obtain the parameters of the dose-response curve include the following: Cervical and lumbar spinal cord function in rats Kidney function in the rabbit, pig, or mouse Lung LD50 Lung (i.e., number of breaths/min) Bladder (i.e., urination frequency) Skin of the pig (i.e., late contraction) In these instances, the α:β ratio usually falls in the range of 1.5 to 4 Gy.

Calculations of Tumor Cell Kill Although the survival curve for cells exposed to graded single doses of radiation may be a linear-quadratic formulation, the effective survival curve for a fractionated regimen delivered as daily 2-Gy fractions is an exponential function of dose. The D0, the reciprocal of the slope of the curve

Reciprocal total dose (Gy−1)

Surviving fraction

1

0.04

1

Jejunal crypt cells

0.1

2

0.03

Skin cells

17 64

8

∴α / β = 10.1 Gy

16 32 Intercept =

0

0.01

Slope =

5 0.02

0.01 Colony forming bone marrow cells

β logeS = 0.00129 Gy −2

4

5

α = 0.013 Gy−1 logeS

10

15

20

Dose per fraction (Gy) 0

2

4

6

8

10

Dose (Gy)

FIGURE 1-12.  X-ray survival curves are reconstructed for three mammalian cell types using in vivo systems. Colony-forming units in the bone marrow are cultured according to the technique of J. E. Till and E. A. McCullough; skin colonies, by a ­technique devised by H.  R. Withers; and regenerating crypts in the intestinal epithelium, by a technique used by H. R. Withers and M. M. Elkind. The slopes of the survival curves are different, ­particularly in the width of the shoulder.

FIGURE 1-13.  Reciprocal of the total dose required to produce a given level of injury (i.e., acute skin reaction in mice) is a function of the dose per fraction in multiple equal doses. The overall time of these experiments was sufficiently short that proliferation could be neglected; the numbers of fractions are shown by each point. From the values of the intercept and slope of the best-fit line, the values of α and β and the ratio α:β for the dose-response curve for organ function can be determined. (Modified from Douglas BG, Fowler JF: The effect of multiple small doses of x-rays on skin reactions in the mouse and a basic interpretation. Radiat Res 1976;66:401-426.)

1  Physical and Biologic Basis of Radiation Therapy

and defined as the dose required to reduce the fraction of cells surviving to 37%, has a value of about 3 Gy for cells of human origin. This average value can differ significantly for different tumor types. For calculation purposes, it is often useful to use the D10, the dose required to kill 90% of the population: $  . × $ In the equation, 2.3 is the natural logarithm of 10. Two sample calculations follow. Example 1: A tumor consists of 109 clonogenic cells. The effective dose-response curve, given in daily dose fractions of 2 Gy, has no shoulder and has a D0 of 3 Gy. What total dose is required to give a 90% chance of tumor cure? Answer: To give a 90% probability of tumor control in a tumor containing 109 cells requires a cellular depopulation of 10−10. The dose to result in 1 decade of cell killing (D10) is given by the following: $  . × $  . ×   . 'Y Total dose for 10 decades of cell killing, therefore, is 10 × 6.9 = 69 Gy. Example 2: Suppose that in the previous example the clonogenic cells underwent three cell doublings during treatment. Approximately what total dose would then be required to achieve the same probability of tumor ­control? Answer: Three cell doublings would increase the cell number by 2 × 2 × 2 = 8. Consequently, about an extra ­decade of cell killing would be required, corresponding to an additional dose of 6.9 Gy. Total dose is 69 + 6.9 = 75.9 Gy.

Differential Effects on Normal Tissues In experimental animals and humans, normal tissues vary considerably in their sensitivity to ionizing radiation. Some of the least sensitive tissues (e.g., vagina, uterus, bile ducts) can tolerate a dose nearly 100 times that of very sensitive tissues (e.g., bone marrow, lens, gonads). The details of the effects that occur and the doses at which various effects can be seen are explored in the individual chapters of this textbook. However, some comparisons and generalizations of the biologic effects of ionizing radiation can be made here. Acute Effects. Certain tissues manifest evidence of radiation effects in a matter of hours to days; they can be classified as acutely responding tissues. In approximate ­order of clinical significance, these tissues are as follows: Bone marrow Ovary Testis Lymph node Salivary gland Small bowel Stomach Colon Oral mucosa Larynx Esophagus Arterioles Skin Bladder Capillaries Vagina

15

Tissues that do not reveal acute effects of clinical significance but show evidence of damage of clinical significance in a matter of weeks to a few months after irradiation are often included with acutely responding tissues. Tissues such as these may be better categorized as subacutely responding tissues. In order of decreasing sensitivity, these tissues include the following: Lung Liver Kidney Heart Spinal cord Brain Late Effects. If given sufficient doses of radiation, all tissues can manifest late effects. However, several tissues are characterized as showing little or no evidence of acute effects but having well-recognized late effects. These ­tissues include the following: Lymph vessels Thyroid Pituitary Breast Bone Cartilage Pancreas (endocrine) Uterus Bile ducts

Effects on Tumors Comparison of Normal and Malignant Cells Malignant cells differ little from normal cells in their responses to ionizing radiation, tending to be most sensitive in the same phases of the cell cycle. Most cells seem to undergo repair of sublethal damage if sufficient time elapses between two doses of radiation. Depending on the interval after irradiation, repopulation may result from the proliferation of clonogenic cells. Redistribution of cells as a function of their differing sensitivities in different phases of the cell cycle, with consequent radiation-induced synchrony, may occur in benign and malignant cells. However, only malignant cells are thought to undergo reoxygenation ­after the administration of ionizing radiation. The radioprotective effect of hypoxia in tumors is an important reason for the failure to control tumors with irradiation. The concepts of reoxygenation and radioprotection are discussed further under “Tissue and Tumor Kinetics.” Malignant tumors differ greatly in radiosensitivity. ­Although cell survival data suggest that different cell types have some inherent differences in radiosensitivity, the differences in sensitivities of tumors are the result of many factors, including inherent cellular sensitivity, hypoxic ­fraction, vascularity, and changes in vascularity between repeated applications of radiation. Normal tissues and malignant tumors have dose­response relationships that tend to be sigmoidal (Fig.  1-14). A certain level of dose must be reached before any response is seen, after which the rate of response increases rapidly, followed by a diminishing rate of response at the highest dosage levels. As expected from the logarithmic ­reduction in the surviving fraction of cells after irradiation, higher doses are required to control larger tumors. Unfortunately, the term radiosensitivity has often been used to describe the rapidity with which a tumor shrinks

Probability of local control or complications

16

PART 1  Principles 1.0 0.90

Local control

Complications

0.05 Dose

FIGURE 1-14.  The probability of local tumor control and the probability of complications are a sigmoid function of dose. If the two curves are well separated, it is possible to achieve a high rate of tumor control with a small complication rate. If the curves are closer together, as may be the case for large, resistant tumors, this favorable situation would not apply.

after irradiation. The rate of regression after irradiation is more appropriately called radioresponsiveness; it has no relationship to radiosensitivity at the cellular level. Shrinkage is more related to the rate of cell removal (i.e., to the cell loss factor, as discussed later under “Tumor Cell ­Kinetics”). Shrinkage, or response of a tumor, is an important element for determination of tumor control. Complete response is the only meaningful indicator of tumor control. Because of the heterogeneity of cells within a tumor, there is no certainty of permanent tumor control after a complete response. A partial response is a radiotherapeutic failure except in the infrequent cases of tumors that may reach a stable state after initial shrinkage because they have a large stromal component (e.g., desmoid tumors) or because they have residual mature elements after the malignant component is eradicated (e.g., ganglioneuroblastomas, teratocarcinomas).

Assays for Transplantable Tumors in Laboratory Animals Several assay systems have been developed for transplantable tumors in small laboratory rodents. The advantage of these systems is that they are reproducible and highly quantitative. The disadvantage is that the tumors involved are highly anaplastic, undifferentiated, and rapidly growing, but they remain encapsulated, so that they show little resemblance to spontaneous tumors in humans. The following paragraphs describe five basic assays that have been developed and used widely over the years. These assays vary in complexity and expense, and they have different advantages and disadvantages. Tumor-Control Dose Assay. The most direct experimental assay, which is parallel to the observation of patients treated with x-rays, is to determine the dose of radiation required to produce tumor control (i.e., tumorcontrol dose [TCD]) in irradiated animals.16 In this assay, tumors are transplanted into a large number of animals and allowed to grow to a certain size, and then groups of animals are treated with graded doses of radiation. The animals are then observed for some period, and the dose is calculated that produces tumor control in one half of the animals treated. This quantity is the TCD50. The disadvantage of this system is that it is costly, inasmuch as perhaps  100 animals are required to produce this single-figure ­estimate of the TCD50. The balancing advantage is that the tumor is treated in situ, with all of the complexities

of the tumor-host response, and the tumor remains undisturbed after treatment. Growth Delay Assay. In the growth delay assay, a large number of identical animals are used in which tumors of a given size are produced by transplantation. After irradiation, the tumors are measured daily so that the patterns of regression and regrowth can be established. The time in which a tumor grows back to its original size after irradiation is a measure of tumor response called the growth delay. For tumors in which shrinkage does not occur, an alternative end point is to express the time for the tumor to grow from size A to size B (perhaps from 2 to 8 mm) in irradiated animals compared with the comparable time in unirradiated controls. Dilution Assay. The dilution assay technique devised by Hewitt and Wilson17 was the method by which the first in vivo dose-response curve was obtained. In this technique, initially devised for lymphocytic leukemia and subsequently adapted for use with solid tumors, tumor-bearing animals are irradiated and then cells are withdrawn and diluted, and various numbers are inoculated into recipient animals, which are then observed to see if they develop a tumor. The TD50 (i.e., number of cells required to produce a tumor in one half of the animals inoculated) is determined for control animals and irradiated animals. The ratio of TD50 values for control and irradiated animals is the surviving fraction of cells for that particular radiation dose. The strength of this system is that the tumor is irradiated in situ. An additional advantage of this assay over the two methods referred to earlier is that the reproductive ­integrity of individual cells is determined rather than the response of the tumor as a whole. The disadvantage of the dilution assay is that after the solid tumors are irradiated, the tumor must be broken up and disaggregated. Although this practice does not seem to produce any significant artifacts in the case of radiation, it may produce spurious results in  evaluations of chemotherapy agents. This assay system is also time-consuming and expensive, inasmuch as many animals are required to produce a dose-response curve. For example, a control group of at least six animals and an irradiated group of six animals produces the same amount of information as that obtained from one control and one irradiated Petri dish in the in vitro assay. Lung Colony Assay. The lung colony assay system was developed and has been used largely in Canada.18 In this system, solid tumors are irradiated in situ and then removed, disaggregated into a single-cell suspension, and inoculated into recipient animals that are killed some weeks later. The number of metastases in the animals’ lungs is counted. This system, too, exploits the realism of in vivo irradiation and assessment of the reproductive integrity of survivors, but it suffers the disadvantages of requiring large numbers of animals for a given amount of data and requiring that the tumor be disaggregated after treatment, before its assay. In Vivo/In Vitro Assay. In the in vivo/in vitro assay, the tumor is irradiated in situ, removed, and disaggregated into a single-cell suspension. Various numbers of cells are then plated onto Petri dishes, and their reproductive integrity is assessed by their ability to produce colonies in vitro. This technique combines the advantages of the in vivo and in vitro assays; the tumors are irradiated in situ with all of the complexities of the in vivo milieu, and they are then assessed with the speed, accuracy, and economy of the in

17

1  Physical and Biologic Basis of Radiation Therapy

vitro cell culture technique. Only a limited number of tumors have been developed that can be transferred back and forth between the animal and the Petri dish; unfortunately, these are all highly undifferentiated tumors. This technique also has the same limitation as the dilution assay and the lung colony assay systems in that it involves the disaggregation of the tumor after treatment. These assay systems, developed initially for the assessment of response to x-rays, neutrons, and other types of radiation, have allowed a vast body of data to be accumulated for chemotherapy agents and hyperthermia. Use of these systems resulted in the development of the important concept that the presence of a proportion of hypoxic cells can limit the curability of a tumor with a single dose of x-rays, leading to the demonstration and explanation of the phenomenon of reoxygenation. Models that involve the use of transplantable tumors in small rodents have ­fallen into disfavor to some extent because the dose-response curves obtained do not give much more information than an in vitro survival curve that requires much less time and a fraction of the cost. Criticism levied against these ­ models is that the tumors are highly undifferentiated and not at all like spontaneous human tumors. Their response to hyperthermia and chemotherapy agents may give spurious results for various reasons that are described later.

Tissue and Tumor Kinetics The Cell Cycle Mammalian cells propagate and increase in number by mitosis. The division cycle for mammalian cells is illustrated in Figure 1-15. The principal phases of the cell cycle (M, G1, S, G2) were described by Howard and Pelc in 1953.19 During a complete cell cycle, the cell must accurately replicate the DNA during S phase and distribute an identical set of chromosomes equally to two progeny cells during M phase. Mitosis

Regulation of the cell cycle in eukaryotic cells ­ occurs by the periodic activation of different members of the ­cyclin-  dependent kinase (CDK) family. In its active form, each CDK is complexed with a particular cyclin. Different CDKcyclin complexes are required to phosphorylate several protein substrates, and that phosphorylation drives cell cycle events such as the initiation of DNA replication or the onset of mitosis. CDK-cyclin complexes are also vital in preventing the initiation of a cell cycle event at the wrong time. Extensive regulation of CDK-cyclin activity by several transcriptional and posttranscriptional mechanisms ensures the perfect timing and coordination of cell cycle events. The CDK catalytic subunit by itself is inactive, requiring association with a cyclin subunit and phosphorylation of a key threonine residue (usually T-160) to become fully active. The CDK-cyclin complex is reversibly inactivated by phosphorylation on a tyrosine residue (usually Y-15) located in the ATP-binding domain or by association with cyclin-­dependent kinase inhibitor (CKI, also known as celldirected inhibitor [CDI]) proteins. After the completion of the cell cycle transition, the complex is irreversibly inactivated by ubiquitin-mediated degradation of the cyclin subunit. The entry of cells into S phase is controlled by CDKs that are themselves sequentially regulated by cyclins D, E, and A. D-type cyclins act as growth-factor sensors, with their expression depending more on the extracellular cues than on the cell’s position in the cycle. Mitogenic stimulation governs their synthesis and the formation of complexes with CDK4 and CDK6, and catalytic activity of the assembled complexes persists through the cycle as long as mitogenic stimulation continues. Cyclin E expression in proliferating cells is normally periodic and maximal at the G1-S transition; throughout this interval, cyclin E enters into active complexes with its catalytic partner, CDK2. This view of the cell cycle and its regulation is illustrated in Figure 1-16.

G0 M

G2

EGF FGF PDGF

M

G1

Transcription factors

Cyclin B/A and CDK1

G2

Cyclin D1 and CDK4 Cyclin A and CDK2 S

Cyclin E and CDK2

G1

p53 Rb p21 inhibits CDKs

Late S

S

FIGURE 1-15.  With a conventional microscope, the only phase of the mammalian cell cycle that can be identified is mitosis, in preparation for which the chromosomes condense and become visible. With the use of autoradiography, the DNA synthetic phase (S) can be identified to occupy a discrete length of time in the cycle.

FIGURE 1-16.  Updated view of the phases of the cell cycle shows how the phases are regulated by the periodic activation of different members of the cyclin-dependent kinase (CDK) family. Various CDK-cyclin complexes are required to phosphorylate several protein substrates that drive key events, including the initiation of DNA replication and the onset of mitosis. EGF, epidermal growth factor; FGF, fibroblast growth factor; PDGF, platelet-derived growth factor.

PART 1  Principles

The only phase of the cycle that can be observed with a light microscope is the process of division itself, in preparation for which the chromosomes condense and become visible. To visualize the DNA-synthesis (S) phase, the technique of autoradiography must be used. In this technique, cells are fed one of the precursors needed to build DNA that is labeled with a radioactive isotope such as tritium. Only cells in the DNA-synthesis phase take up the tritium. If those cells are later fixed, stained, and covered with a layer of nuclear emulsion, the beta particles from the ­tritium pass through the nuclear emulsion and produce an image that is visible when the emulsion is fixed and stained. To avoid using radioactive material, cells are more often labeled by the incorporation of bromodeoxyuridine (BrdU), the presence of which can be detected by a ­fluorescent-­labeled antibody (Fig. 1-17). However, the principle remains the same.

A

B FIGURE 1-17.  A, Autoradiograph of cultured Chinese hamster cells flash-labeled with tritiated thymidine. Black grains indicate cells synthesizing DNA when they were labeled. The labeled mitotic cell (right) was in S phase when the culture was flash-­labeled but in M phase at staining and autoradiography. B, Photomicrograph of cells grown in the presence of bromodeoxyuridine (BrdU) and fixed and stained 20 hours later. First-generation mitotic cells show uniform staining of both chromatids of each chromosome. Inset, A second-generation mitotic cell (which incorporated BrdU during two S phases); the dark chromatids indicate BrdU incorporation by both strands of the DNA double helix, and the light chromatids by only one. (Courtesy of Charles R. Geard, MD, Columbia University, New York, NY.) Modified from Hall EJ, Giaccia AJ. Radiosensitivity and cell age in the mitotic cycle. In: Radiobiology for the Radiologist. 6th ed. Philadelphia: Lippincott Williams & Wilkins, 2006, pp 47-59.

The simplest experiment that can be done with a cell culture, a tissue, or a tumor is to flash-label it by exposing the cells or tissues for 15 to 20 minutes with the labeled DNA precursor and then to look at the samples after they have been autoradiographed. The labeling index (LI), the proportion of cells that are labeled, reflects the ratio of the length of the DNA-synthesis phase (Ts) to that of the total cell cycle (Tc): ,)  4S/4C

From the same sample, it is also possible to observe the mitotic index (MI), defined as the proportion of cells in mitosis. This index represents the ratio of the length of the mitotic time (Tm) to that of the Tc: -)  4M/4C

It is therefore possible in a simple experiment to observe the ratio of the lengths of mitosis and the DNA-­synthesis phase to that of the total cell cycle. However, these experiments involve the measurement of ratios and cannot be used to estimate the length of any phase of the cycle. To determine the length of the various phases of the cell cycle, a much more complex procedure is necessary—the percent-labeled mitoses technique. The cell culture, tissue, or tumor is flash-labeled with tritiated thymidine, after which samples are taken, fixed, stained, and autoradiographed hourly for some period that is longer than the anticipated cell cycle. A typical percent-labeled mitoses curve is shown in Figure 1-18.20 This technique follows the progress of the labeled cohort of cells through DNA synthesis, to the next mitosis, and through the cell cycle to another mitosis. The second wave of labeled mitoses is always smaller than the first because the cells in any population in vivo do not have a single cell cycle time but rather are characterized by a range of cell cycle times. A computer technique can be used to unfold the spectrum of cell cycle times from the decrease in height of the second peak and to obtain an estimate of the range of cell cycles involved. The width of the first wave of labeled mitoses at the 50% height gives an estimate of the Ts, and the time between the first and second peaks is an estimate of the average Tc. In many instances, particularly with normal tissues, the second peak is too indistinct to be recognized; then the cell cycle cannot be estimated from the distance between consecutive peaks. In such cases, the Ts must be obtained Percent labeled mitoses

18

100 80 60 40 20 0 0

2

4

6

8

10 12 14 16 18 20 22 24

Hours after injection of 3HTdR

FIGURE 1-18.  In the percent-labeled mitoses curve for a chemically induced carcinoma in the cheek pouch of a hamster, the squares and circles refer to two separate experiments. (From Brown JM. The effect of acute x-irradiation on the cell proliferation kinetics of induced carcinomas and their normal counterpart. Radiat Res 1970;43:627-653.)

1  Physical and Biologic Basis of Radiation Therapy

from the width of the first peak and the cell cycle determined by the observation of the LI: 4C  4S/,)

Population Kinetics of Tumors When animal tumors are studied with the percent-­labeled mitoses technique, it becomes evident that the cell cycle time is much shorter than the gross volume-doubling time. There are two reasons for this. First, not all cells in a ­growing tumor are actively cycling at any given time. By observing the uptake of tritium in mouse tumors, it is possible to can estimate that the growth fraction, defined as a proportion of cells that are actively cycling, ranges from 30% to 50%. From the cell cycle of individual cells and the growth fraction, it is possible to calculate the potential doubling time (Tpot). The potential doubling time is still much shorter than the actual gross volume-doubling time, and this difference is accounted for by the cell loss fraction (Φ), defined as the proportion of cells produced by mitosis that are lost from the tumor. These cells are lost largely into the necrotic “sinks” within the tumor, areas where cells die because they are remote from a source of oxygen and nutrients. The overall picture of tumor growth that emerges is that of a population of cells, some of which are dividing and others are quiescent, with the progeny of division largely being lost into necrotic areas of the tumor. For this reason, the gross volume-doubling time of the tumor, as a whole, is much longer than the cell cycle time. In humans, a limited number of population kinetic studies have been performed because it is not possible in most countries to give large doses of tritiated thymidine to patients. Generally, human tumors are observed to have gross volume-doubling times on the order of months, judging from repeated measurements of skeletal or pulmonary metastases. In contrast, 80% of human tumors have a cell cycle that lasts between 2 and 5 days. These numbers are consistent with the estimate of Gordon Steel21-23 that, on average, human tumors have a cell loss factor of 70% (i.e., almost three fourths of the cells produced by division are lost into necrotic areas within the tumor).

Clinical Applications Although results from population kinetic studies performed in the past are illuminating and instructive, they cannot be applied to an individual patient because of the long periods required to perform cell cycle analyses. For example, an autoradiograph takes about 6 weeks to be produced, which includes the time necessary for sufficient beta particles to be emitted by tritium and incorporated within the cells and to produce a visible image in the nuclear emulsion above the cells. The breakthrough in this area has been the application of flow cytometry.

Measurement of Potential Tumor Doubling Time The growth rate of a tumor is slower than the growth rate of individual dividing cells for two reasons. First, most ­tumors contain significant numbers of cells that are not cycling at any given time; the fraction of cells in cell cycle is the growth fraction. Second, many cells produced by division are lost, mainly in necrotic areas. The fraction of cells produced by mitosis that are lost from the tumor is known as the cell loss fraction (Φ).

19

Tpot accounts for the proportion of dividing cells but not for cell loss. It is a measure of the rate of increase of cells capable of continued proliferation, and it therefore determines the outcome of a treatment delivered in fractions over an extended period. Tumors with a short Tpot may repopulate if fractionation is extended over too long a period. Tpot can be calculated from the following equation: 4POT  Ȉ4S/,) In the equation, λ is a correction factor for the nonlinear distribution in time of the cells as they pass through the cycle (λ is between 1 and 0.67). The Tpot of human tumors can be estimated by flow cytometry analysis of cells from a biopsy specimen taken 6 hours after an injection of BrdU.24-26

Systemic Effects Systemic effects of ionizing radiation are determined by the proportion of the body irradiated, the specific organs included, and the dose received. At the extreme, the lethal effects of total-body irradiation are well recognized. In addition to results from experiments with animals, especially experiments in which investigators are seeking the dose that results in the death of one half of the animals (lethal dose in 50% [LD50]), limited data are available from accidental exposures to humans, such as that at Chernobyl. Three types of death from total-body exposures have been delineated. Doses on the order of 100 Gy or higher cause death in  a matter of hours from effects on the central nervous system and cardiovascular collapse. Intermediate doses of 10 to 20 Gy result in death in several days caused by elimination of the intestinal epithelium with intractable diarrhea and infection. Lower doses of 1 to 5 Gy result in depopulation of hematopoietic stem cells; death may occur several weeks after the exposure from infection or bleeding. The LD50/60 for healthy young adults without medical intervention is between 3 and 4 Gy. Individuals can survive doses perhaps twice as large with supportive treatment, including antibiotics and transfusions of blood products to help them over the nadir in peripheral blood counts. Single-dose total-body irradiation has been used alone or with systemic cytotoxic drugs as ablative therapy to eradicate malignant cells from the bone marrow (especially in cases of leukemia) or from other parts of the body, followed by autologous, syngeneic, or allogeneic reconstitution of the bone marrow. In clinical experiments in which single total-body doses of 5 to 10 Gy and fractionated doses of 12 to 16 Gy have preceded marrow transplants, it has become apparent that the limiting normal tissue is the lung; deaths from interstitial pneumonia have occurred weeks to months after irradiation. The effects of irradiation on immune responses are complex. When total-body irradiation is given in doses that are potentially lethal, the immune mechanisms throughout the body, both humoral and cell-mediated, are depressed. This is a direct result of the suppression of the progenitors of plasma cells and lymphocytes that produce immunoglobulins or participate in cell-mediated responses. The effects of local irradiation, particularly with regard to the volume irradiated and the amount of bone marrow within that volume, are only partially defined. Total lymphoid ­irradiation can have effects that are qualitatively similar to total-body

20

PART 1  Principles

irradiation but are quantitatively less ­similar. In the clinical setting, the malignant process itself may impair humoral and cell-mediated immunity in humans, and thereby, the contributions of the primary disease and the therapy are difficult to separate. Radiation intended as therapy for many common malignant conditions does not impair cellular immunity, as measured by the delayed hypersensitivity reaction. When very large volumes are irradiated, as takes place in the treatment of Hodgkin disease, cutaneous anergy may temporarily follow. Patients who have anergy, presumably related to the malignant process, often recover their delayed hypersensitivity response after radiotherapeutic elimination of the tumor. Depressions of the peripheral blood count from local irradiation are seen only when the volume irradiated is large and encompasses a significant portion of the active bone marrow. Irradiation of small areas does not significantly influence white blood cell count, platelet count, or hematocrit. The assumption that local irradiation is sufficient to cause profound pancytopenia can lead to serious delay in seeking the actual causes. Nausea and vomiting have been considered by patients and physicians to be a necessary accompaniment of any sort of radiation therapy. However, nausea and ­ vomiting may also accompany irradiation as a function of the suggestibility of the patient. Undoubtedly, the range of ­individual sensitivities to these occasional side effects of irradiation is considerable, but nausea is most likely to occur when the stomach or significant portions of the intestine are ­included within the treatment volume. The irradiation of a very large volume of tissue, even when the gastrointestinal tract is excluded (e.g., treatment with the mantle field), also can result in nausea. Nausea that is truly related to irradiation for one of the previous reasons occurs at the very start of a course of treatment. Serious delays in finding the causative factors can occur if it is assumed that the nausea and vomiting that appear well into or near the end of a course of treatment are related to the irradiation when they are not.

Fractionation The earliest attempts at therapeutic uses of the newly discovered x-rays, especially in Germany, involved single, large, caustic doses with profound effects on the tumor but severe injuries to normal tissues. More than 20 years passed before the advantages of repeated, brief applications of x-rays were appreciated. The most important conceptual development in the history of clinical radiation oncology came from studies of spermatogenesis in the ram, reported by Regaud in the early 1900s.27 Regaud demonstrated that a ram could not be sterilized by exposing its testes to a single dose of radiation without extensive damage to the skin of the scrotum, but if the radiation was given in a series of daily fractions, sterilization was possible without producing unacceptable skin damage. Because the testis could be considered a model for a rapidly growing tumor in which the skin of the scrotum represented a dose-limiting normal tissue, the strategy of multifraction radiation therapy was born. The efficacy of fractionation can be understood in terms of radiobiologic principles established with more relevant test systems. These principles have been described as the four Rs of radiobiology: Repair Reassortment of cells within the cell cycle

Repopulation Reoxygenation

Repair Sublethal damage repair and potentially lethal damage ­repair are defined in operational terms. The molecular ­basis of either is not understood, nor is it known whether the two forms of repair involve the same or different ­molecular processes. Sublethal damage repair is defined as the increase in survival observed when a dose of radiation is split into two (or more) fractions with a time interval between. Figure  1-19A shows a dose-response curve for cells given single doses up to the dose value D (solid line). If a dose D/2 is given and a time interval of several hours is allowed to elapse before the remaining portion of the dose is delivered, survival follows the dotted line as shown in Figure 1-19B (i.e., the shoulder of the curve must be re-expressed). ­ Figure  1-19B shows the result of an experiment in which two doses of D/2 were delivered, separated by various time intervals. Repair takes place with a half-life of about an hour and is essentially complete by about 3 hours. Sub­lethal damage repair seems to be a ubiquitous phenomenon; it has been demonstrated in cells in ­ culture, transplanted ­ tumor systems, and normal tissue assay ­systems. In the example of rapidly dividing cells of rodent origin grown in vitro, the half-time of repair is about an hour. However, the rapidity of repair varies considerably. For cells of human origin cultured in vitro, it may vary from a few minutes to several hours.28 Precise estimates for normal tissues in vivo are more difficult to make, but fractionation experiments suggest that the repair of sublethal damage may be very much slower in late-responding tissues.25 The extent of repair varies widely among different cells and tissues, tending to correlate with the size of the shoulder on the cell survival curve. One manifestation of sublethal damage repair is the effect of dose rate. Radiation delivered continuously at a low dose rate over some period approximates to an infinite number of infinitely small dose fractions. Sublethal damage repair occurs during the protracted exposure; as a consequence, the survival curve becomes shallower (i.e., D0 increases), and the shoulder has a tendency to disappear as the dose rate is progressively reduced (Fig. 1-20). Repair of potentially lethal damage is defined as the increase in cell survival that can be produced by manipulation of the postirradiation conditions. If, after irradiation, cells are kept in an environment that prevents cell division, more cells survive than if they are called on to divide soon after the exposure. This effect can be achieved with cells in culture by replacing the growth medium with saline or depleted medium, by lowering the temperature, or by maintaining the cells in plateau phase in culture conditions where they are confluent and have no room to proliferate. Potentially lethal damage repair occurs in cells of normal and malignant origin and has been demonstrated to occur in transplanted tumors in animals. If a tumor is left in situ for 6 to 24 hours after irradiation before being removed, subcultured, and transplanted, more cells survive than if the tumor is transplanted immediately. The naïve interpretation of these results is that some radiation damage that is potentially lethal can be repaired in a few hours, provided the cells are not called on to complete the complex task of mitosis during that time interval.

1  Physical and Biologic Basis of Radiation Therapy Dose

D/2

21

Time between split doses D

2

4

6

8

10

12

Cell cycle

1

0.1 D/2- + -D/2

0.01

A

Reassortment in fastcycling cells

Proliferation

Single dose

Repair SLD

Surviving fraction

Split dose

B

FIGURE 1-19.  A, Increase in cell survival is observed when a dose of radiation is delivered in 2 fractions separated by a time interval adequate for repair of sublethal damage (SLD). When the dose is split into 2 fractions, the shoulder must be expressed each time. B, The fraction of cells surviving a split dose increases as the time interval between the 2 dose fractions increases. As the time interval increases from 0 to 2 hours, the increase in survival results from the repair of sublethal damage. In cells with a long cell cycle or cells that are out of cycle, cell survival cannot be further increased by separating the dose by more than 2 or 3 hours.

Presumably, repair of potentially lethal damage also occurs in human tumors in vivo. It has been speculated that the ability to repair potentially lethal damage efficiently may be related to the resistance of some tumors, such as melanoma. Whether potentially lethal damage repair occurs in normal tissues in vivo is unknown because suitable assays are unavailable. However, it has been demonstrated in cells of normal tissue origin cultured in vitro.

Reassortment The radiosensitivity of mammalian cells varies according to their position in the cell cycle. In fast-growing cells in culture, such as Chinese hamster cells, which have a cell cycle of only 9 or 10 hours, cells in mitosis (M) or late in G2 usually are the most sensitive. They are characterized by a survival curve that is steep, with little or no initial shoulder. Cells late in the DNA-synthetic phase (late S) are most resistant and tend to have a survival curve with a broad shoulder. A similar pattern of radiosensitivity has been shown for the rapidly dividing crypt cells in the mouse intestinal epithelium, the only situation in vivo in which good synchronization can be achieved by the use of a drug such as hydroxyurea. Cells cultured in vitro that have a longer cell cycle (e.g., 24 to 30 hours) and therefore a long G1 phase seem to have a second period of resistance in G1. When an asynchronous population of cells is exposed to a single large dose of radiation, most of the survivors are in the resistant phases of the cell cycle (i.e., the surviving population is partially synchronized). A second dose of radiation is less effective than the first because the population is resistant. If a time interval is allowed, however, cells tend to reassort themselves and become asynchronous

again as cells move from resistant into sensitive phases of the cycle.

Repopulation Repopulation refers to regrowth of cells after irradiation. It was described earlier under “Tissue and Tumor Kinetics.”

Reoxygenation Reoxygenation is the process whereby cells in a tumor that are hypoxic after a dose of radiation become oxygenated again as the tumor shrinks or as the demand for oxygen is reduced. This process has been demonstrated in many transplanted animal tumors. The extent and rapidity of reoxygenation vary considerably in animal tumors, but little is known about the process in humans. Reoxygenation is detailed under “Modifiers of Radiation Effects.”

Fractionation and the Four Rs Dividing a dose into fractions spares normal tissues because of the repair of sublethal damage between dose fractions and because of the repopulation of cells, if the overall time is sufficiently long. At the same time, dividing a dose into fractions increases damage to the tumor as a result of ­reoxygenation. The effect of repopulation of normal tissue is rather complex. The extra dose that is required (because of ­fractionation) to produce a given level of damage does not increase until several weeks after the start of a multifraction regimen because proliferation does not occur until damage is expressed in the normal tissue. After damage to normal tissue is apparent, compensatory proliferation is rapid. However, all normal tissues are not the same. 

22

PART 1  Principles Dose (Gy) 0

2

4

6

8

10

12

14

100

Reaction Sites Low dose rate

Cell surviving fraction

TABLE 1-3.   Ratio of Linear to Quadratic Terms from Multifraction Experiments

10-1

Acute exposure

Dose-rate effect caused by repair of sublethal damage

10-2

α:β (Gy)

Early Reactions Skin

9-12

Jejunum

6-10

Colon

10-11

Testis

12-13

Callus

9-10

Late Reactions Spinal cord

1.7-4.9

Kidney

1.0-2.4

Lung

2.0-6.3

Bladder

3.1-7.0

From Fowler J. The second Klaas Breur memorial lecture. La Ronde—radiation sciences and medical radiology. Radiother Oncol 1983;1:1-22.

10-3

FIGURE 1-20.  The dose-rate effect is depicted. For sparsely ionizing types of radiation such, as x-rays and γ-rays, the biologic effectiveness of a given radiation dose decreases as the dose rate is lowered. This occurs when the exposure time becomes comparable to or longer than the half-time of repair of sublethal damage, so that sublethal damage is repaired during the protracted exposure. As the dose rate is progressively reduced, the slope of the survival curve becomes shallower (the final slope [D0] increases), and the shoulder of the survival curve disappears. The dose-response curve then approximates an exponential function of dose. Dose

In particular, a clear distinction exists between tissues that are rapidly responding, such as the skin, the intestinal epithelium, or the oral mucosa, and late-responding tissues, such as the spinal cord. Prolonging overall treatment time within the normal radiation therapy range has little ­sparing effect on late reactions, but it has a large sparing effect on early reactions. This axiom is of far-reaching importance in radiation therapy. Early reactions, including those in the skin and the mucosa, can be avoided by the simple expedient of prolonging the overall time. Although such a strategy avoids the problems of the early reactions, it has no effect whatsoever on the late reactions because compensatory proliferation in late-responding tissue does not occur during the time course of conventional ­radiation therapy.

Surviving fraction

Early- and Late-Responding Tissues

Late-responding tissue

Tumor and earlyresponding tissue

FIGURE 1-21.  The dose-response relationship for late­responding tissue is “curvier” than for early-responding tissue. The dose at which cell killing is equal by the linear and quadratic components is α:β. This is about 2 Gy for late-responding and 8 to 10 Gy for early-responding tissue. (Information courtesy of H. R. Withers, MD, UCLA Medical Center, Los Angeles, CA.)

Differences in the repair and efficiency of reassortment of cells within the cell cycle lead to quite different shapes of the dose-response curves for early-responding and late-responding tissues. Dose-response relationships for late-responding tissues tend to be more curved than for early-responding tissues.30 In terms of the linear-quadratic relationship between effect and dose, this translates into a larger ratio of α:β for early effects than for late effects. The difference in the shapes of the dose-response relationships is illustrated in Figure 1-21. Values of α:β for a range of tissues in experimental animals is shown in Table 1-3.31 These data came from multifraction experiments with rats and mice. In general, the value of α:β is on the order of 2 Gy for late-responding tissues but is closer to 8 or 10 Gy for early-responding tissues. Consideration of the shape of the dose-response relationship for early-responding and late-responding tissues leads to the important truism that fraction size is the dominant factor in determining late effects, whereas overall treatment time has little influence. In contrast, fraction size and overall treatment time determine the response of acutely responding tissues.

1  Physical and Biologic Basis of Radiation Therapy

The Strandqvist Plot and Nominal Standard Dose Early attempts to understand and account for fractionation in clinical radiation therapy gave rise to the well-known Strandqvist plot, in which the effective single dose was plotted as a function of the overall treatment time. Because all treatments were given as 3 or 5 fractions per week,  overall time in these plots carried with it the implication of a certain number of fractions. These plots commonly showed that the slope of the isoeffect curve for skin was approximately 0.33. The most important contribution in the area of fractionation, made by Ellis and colleagues32 with the ­introduction of the nominal standard dose system, was the recognition of the importance of separating overall treatment time from the number of fractions. This system was used to ­predict which radiation regimens would produce equivalent results with different numbers of fractions, but it has fallen out of favor.

Tumor Control and Dose-Time Considerations For many years, the basic determinant of total dose in the clinical setting was the tolerance of the surrounding normal tissues, especially the skin. With the advent of megavoltage treatment units and the attendant skin-sparing effect, greater attention has focused on the depth and distribution of dose correlated with permanent eradication of the disease. It became apparent that tumor-control probability for epithelial tumors is profoundly affected by small changes in the time-dose relationships, and the size of the tumor greatly influences the dose necessary to achieve a high probability of control. Although the time-dose relationships are of great importance for controlling epithelial tumors, the time factor seems to be of much less significance in the treatment of highly radiosensitive tumors (e.g., Hodgkin disease, malignant lymphomas, seminoma, Wilms’ tumor, neuroblastoma, retinoblastoma). The total dose is a more important determinant in the control of Hodgkin disease and nodular lymphomas, whereas the number of fractions and the time over which the dose is delivered are less important. ­Because deleterious effects on normal tissues are related to dose and to fractionation and time, radiation that is more fractionated and given over a longer period tends to have a greater margin of safety. Radiation in laboratory animals or humans is used, with rare exceptions, as a locoregional treatment. Although laboratory studies have consistently evaluated tumor control (TCD50) as the end point, clinical results are often ­expressed in terms of survival or cancer-free survival. The only direct measure of the effectiveness of irradiation is control of the tumor within the irradiated volume (i.e., ­ local control). Margin failure, the failure to appreciate the original extensions of the tumor, should be analyzed ­separately. The only measure of failure of local control (i.e., true recurrence) is regrowth of the tumor within the irradiated volume. Identifying failure of local control requires ­careful evaluation of the patient over months and years, and ­ attempts to predict failure of local control by other means have been unsuccessful. The rate at which the tumor ­ regresses and, consequently, the extent of residual tumor at the completion of irradiation have been evaluated in ­ animals and in humans and have not proved to

23

correlate with ultimate control of disease. Suggestions that ­ postirradiation biopsies might help were based on the expectation that tumor cell “viability” could be assessed histologically. This largely was proved incorrect in the ­ immediate postirradiation period. However, if a high growth fraction is present and most cells are actively cycling, eventual control may be predicted by repeat biopsies at the site of the original tumor. Arriagada and colleagues33 documented a high local failure rate of radiation to treat non–small cell carcinoma of the lung by using repeat bronchoscopy and biopsy 3 months after radiation therapy with and without induction chemotherapy. In that study, almost every patient with positive biopsy results developed recurrent disease and eventually died, and all long-term survivors had negative bronchial biopsy results. In contrast, biopsies of the prostate 2 years or more after external beam radiation therapy are much less predictive of clinical local failure,34 although they have been found to correlate with levels of prostatespecific antigen.35

Large-Dose Fractionation The strong influence of fraction size on late effects led most investigators to abandon the use of large fractions in the 1980s.36 Since that time, a resurgence of interest evolved from the ability to achieve highly conformal dose ­distributions by using advanced imaging for the planning and delivery of radiation treatment. Stereotactic radiation therapy for cerebral38 and pulmonary tumors37 uses between 1 and 5 fractions ranging from 12 Gy to 20 Gy each. The use of radiation fractions larger than traditional 1.8 Gy or 2.0 Gy is being studied for the treatment of disease at a variety of anatomic sites.

Multiple Fractions Per Day Most malignant tumors arising from the epithelium require total doses near or even above those tolerated by the surrounding normal tissues. Consideration of the different α:β ratios for early-responding tissues, including tumors, and late-responding tissues suggests an advantage to altered fractionation. Prolongation of treatment has the advantage of avoiding acute reactions and allowing adequate reoxygenation in tumors. However, excessive prolongation can decrease acute reactions without necessarily sparing late injury, and it may allow surviving tumor cells to proliferate during treatment. Two separate strategies use multiple fractions per day:  hyperfractionation and accelerated fractionation. The objectives of these two strategies are quite different. Both fractionation strategies involve multiple fractions per day, and in this context, it is important to ensure that the two fractions are separated by a sufficient time interval for the effects of the two doses to be independent, allowing the repair of sublethal damage from the first dose to be complete before the second dose is delivered. With cells cultured in vitro, the half-time of repair is usually about an hour, but repair may be much slower in late-responding tissues. Clinical trials indicate that for a given total dose delivered in a given number of fractions, the incidence of late effects was worse for average interfraction intervals of less than 4.5 hours.39 Current wisdom dictates an interfraction interval of 6 hours or more when multiple fractions per day are used.

24

PART 1  Principles

Hyperfractionation. The basic aim of hyperfractionation is to further separate early and late effects. “Pure” ­hyperfractionation may be defined as keeping the same ­total dose as in a conventional regimen in the same overall time but delivering it in twice as many fractions by the ­expedient of treating twice each day. However, this strategy would not be satisfactory because the total dose would have to be increased if the dose per fraction were ­decreased. In practice, “impure” hyperfractionation involves an increase in the total dose and sometimes a longer overall time, as well as many more fractions delivered twice daily. The intent of hyperfractionation is to reduce late effects while achieving the same or better tumor control and the same or slightly increased early effects. A large, controlled clinical trial of hyperfractionation for the treatment of head and neck cancer was conducted by the European Organisation for Research and Treatment of Cancer (EORTC).40 A hyperfractionated schedule of 80.5 Gy delivered in 70 fractions (1.15 Gy delivered twice daily) over a period of 7 weeks was compared with a conventional regimen of 70 Gy delivered in 35 fractions of 2 Gy over 7 weeks. Local tumor control at 5 years after the hyperfractionated protocol increased from 40% to 59%, and this increase was reflected in improved survival. No increase in late effects or complications was reported. These results led to the conclusion that hyperfractionation confers an unequivocal advantage in the treatment of oropharyngeal cancer. Two fractions per day may not be the limit of hyperfractionation. A further sparing of late effects by splitting the dose into more and more fractions of smaller and smaller size would occur as long as the dose fraction is still on the curved portion of the dose-response curve. Accelerated Treatment. The alternative strategy to hyperfractionation is accelerated treatment. “Pure” accelerated treatment may be defined as the same total dose delivered in one half of the overall time by the expedient of giving two or more fractions each day. In practice, this can never be achieved because the acute effects become limiting. A rest period must be imposed in the middle of the treatment, or the dose must be reduced slightly, with acute effects being the limiting factor. The intent of this accelerated treatment strategy is to reduce repopulation in rapidly proliferating tumors. Late effects should show little or no change because the number of fractions and the dose per fraction are unaltered. In addition to the hyperfractionation trial described previously, the EORTC conducted a large prospective randomized clinical trial of accelerated treatment for head and neck cancer (except that of the oropharynx).41 The accelerated treatment consisted of 72 Gy given in 45 fractions (1.6 Gy in 3 fractions per day) over a 5-week span, with a 2-week rest in the middle. The conventional control condition involved 35 fractions of 2 Gy for a total dose of 70 Gy in 7 weeks. The results of this trial showed a 15% increase in locoregional control for accelerated fractionation, but this increase did not translate into a survival advantage. As expected, acute effects were significantly increased, but the observed increase in late effects was decidedly not ­expected; moreover, some of those effects involved complications that proved to be lethal. This EORTC trial and several others testing accelerated treatment show that attempting to keep the total

dose as high as 66 to 72 Gy while shortening the overall time by as much as 2 to 3 weeks from a conventional 6 or 7 weeks leads to serious problems with late complications. There are probably two reasons for this. First, the late effects observed are consequential late damage (i.e., late damage developing out of the very severe acute  effects). Second, repair is probably incomplete between dose fractions, particularly for protocols involving 3 fractions per day, because any unrepaired damage in the first interval would accumulate in the second interval in each day. Moreover, the interval between fractions in the early years of the EORTC trial was only 4 hours. Continuous Hyperfractionated Accelerated Radiation Therapy Protocol: The CHART Trial. The strategy underlying the CHART trial was to use a low dose per fraction to minimize late effects and a very short overall time to minimize tumor proliferation.42 In this unique study, conducted at the Mount Vernon Hospital in association with the Gray Laboratory, patients were given a total dose of 50.4 to 54 Gy in 36 fractions over 12 consecutive days, with 3 fractions delivered daily and an interfraction interval of 6 hours. The dose per fraction was 1.4 to 1.5 Gy. By conventional standards, the total dose was very low, but it was delivered in a very short time. This protocol produced good local tumor control; despite the severe acute reactions experienced, patients reportedly favored the protocol because treatment was concluded quickly. The incidence of late effects in general was not increased, and by some ­measures, the incidence was decreased. The notable ­exception was damage to the spinal cord. Several myelopathies were recorded at total doses of 50 Gy, probably because an interfraction interval of 6 hours was not sufficient to allow full repair of sublethal damage in this tissue. CHART is the only fractionation schedule that resulted in a lower incidence of late complications. The severe but tolerable acute reactions associated with this protocol did not translate into late sequelae, probably because the total dose was relatively low. Effectiveness in tumor control was not lost, even at this low dose, because the overall treatment time was extremely short, minimizing tumor cell proliferation. Compliance was also high because the acute reactions did not peak and become uncomfortable until ­after the end of treatment. Radiation Therapy Oncology Group Trial 9003. In the largest fractionation trial conducted (1073 evaluable patients), trial 9003 by the Radiation Therapy Oncology Group (RTOG) compared a standard fractionation protocol (70 Gy delivered at 2.0 Gy per fraction) with three other fractionation protocols in patients with stage III to IV, M0 squamous carcinoma of the oral cavity, oropharynx, or supraglottic larynx or with stage II to IV carcinoma of the base of the tongue or the hypopharynx.43 Two of these protocols, one involving hyperfractionation of a total dose of 81.6 Gy given in 68 fractions twice daily over 7 weeks and the other involving accelerated fractionation by a concomitant boost with 30 fractions of 1.8 Gy in 6 weeks plus a second fraction of 1.5 Gy during the last 12 treatment days (total dose of 72.0 Gy given in 42 fractions over 6 weeks), produced significantly higher local control rates than did standard fractionation. A third accelerated fractionation scheme with a total dose of 67.2 Gy given in 42 fractions over 6 weeks with an interruption after 38.4 Gy produced a local control rate similar to that of

1  Physical and Biologic Basis of Radiation Therapy

25

the standard fractionation protocol. The incidence of acute effects was higher among those treated with the nonstandard fractionation regimens than for those treated with the standard regimen, but the late effects were similar. These results provided immediate opportunities for mathematical modeling of radiation effects, especially the interplay of time and total dose in treatment outcome.44,45 Accelerated Hyperfractionated Radiation ­ Therapy with Carbogen Breathing and Nicotinamide: The ARCON Protocol. Also worthy of mention is a strategy involving accelerated hyperfractionated radiation therapy while the patient breathes carbogen (95% O2 and 5% CO2) with nicotinamide. The goals of this regimen, known as the ARCON protocol, were to accelerate treatment to avoid tumor proliferation, to hyperfractionate the doses (small doses per fraction) to minimize late effects, to add carbogen breathing to overcome chronic hypoxia, and to add nicotinamide to overcome acute hypoxia. Clinical trials of this complex but imaginative protocol are being conducted.  Early results of a trial conducted in The Netherlands for patients with advanced laryngeal cancer were spectacular compared with historical controls.45a Results of a prospective, randomized trial have yet to be published.

of hypoxic cell radiosensitizers (discussed further under “Modifiers of Radiation Effects”). Nevertheless, in the case of relatively rapidly growing tumors such as head and neck cancer, ­overall treatment time can be a dominant factor in determining outcome. Overall Treatment Time and Local Control. One of the most important lessons from fractionation studies is that local control is lost when overall treatment time is prolonged. For head and neck cancer in particular, local control is reduced by 0.4% to 2.5% for each day that the overall treatment time is prolonged.44,45 In other words, after the first 4 weeks of a fractionated schedule, the first 0.61 Gy of each day’s dose fraction is required to overcome proliferation from the previous day. Similar estimates can be made for carcinoma of the cervix, in which a mean of 0.5% local control (range, 0.3% to 1.1%) is lost for each day that the overall time is prolonged. On the other hand, the overall treatment time is not so critical for cancer of the breast or prostate in which tumor cell proliferation is not so rapid. Although estimates of Tpot have not proved useful as predictors for individual patients, mean Tpot values for groups of patients are in accord with the importance—or lack thereof—of overall treatment time.

Lessons Learned from Fractionation Studies

Accelerated Repopulation

Trials involving changes in fractionation patterns have produced several important results. The first of these ­“lessons learned” is that hyperfractionation seems to ­ confer an ­unequivocal benefit in the treatment of head and neck ­cancer in terms of local control and survival, without ­producing any increase in late sequelae. However, accelerated treatment should be used cautiously given the findings from the EORTC trial of an unexpected increase in serious late complications. Particular caution is necessary if the ­spinal cord is in the treatment field for twice-daily treatments because repair of sublethal damage is slow in this tissue. A second lesson is that late effects depend ­primarily on total dose and dose per fraction; the overall treatment time (within the usual therapeutic range) has little influence. Another important lesson is that overall treatment time affects acute effects and tumor control. Gaps in treatment should be avoided because they lead to an increase in overall treatment time and a concomitant decrease in tumor control. The importance of overall treatment time was illustrated dramatically by Overgaard and colleagues46 in their retrospective analysis of three consecutive trials of the Danish Head and Neck Cancer (DAHANCA) study (Fig. 1-22).46 All three trials involved a total dose of 66 to 68 Gy. The first trial, a split-course regimen that extended over a total of 9.5 weeks, produced a 3-year local control rate of 32%. The second trial, involving 5 fractions per week given over 6.5 weeks, had a 3-year local control rate of 52%. The most recent trial, in which 6 fractions were given per week and the overall treatment time was shortened to 5.5 weeks, produced a 3-year local control rate of 62%. No change in late effects was observed among the three treatment groups, but as expected, the acute reactions increased as the overall treatment time decreased. These intriguing findings must be viewed with some caution, however, because the three treatment groups came from different trials conducted over a period of several years; moreover, the initial purpose of the trials was to investigate the usefulness

Radiation or any other cytotoxic agent can trigger surviving clonogens in a tumor to divide faster than they did before treatment. This concept of accelerated repopulation is illustrated for a rat rhabdomyosarcoma in Figure 1-23.47 Figure 1-23A shows the overall growth curve for the tumor and the shrinkage and regrowth that occurs after a single dose of 20 Gy. Figure 1-23B illustrates the repopulation of individual surviving cells, which after treatment are dividing with a cell cycle of only 12 hours—considerably faster than before irradiation. The rapid proliferation of surviving clonogens occurs while the tumor is overtly shrinking and regressing. This same phenomenon occurs in human tumors, ­although evidence to support this must be indirect. For ­example, from a survey of reports in the literature of ­radiation therapy of head and neck cancer, Withers and colleagues48-50 concluded that the dose required to achieve local control in 50% of the cases (TCD50) increased with overall treatment time beyond approximately 28 days. A dose increment of about 0.6 Gy/day is required to compensate for this accelerated repopulation of clonogens. Such a dose increment is consistent with a clonogen doubling time of 4 days, compared with a median value of about 60 days for unperturbed growth. The conclusion to be drawn from these findings is that radiation therapy, at least for head and neck cancer and probably in other instances, should be completed as soon as is practical after it has begun. It may be better to delay initiation of treatment than to introduce delays during treatment. If overall treatment time is too long, the effectiveness of later dose fractions will be adversely affected because the surviving clonogens in the tumor have been triggered into rapid repopulation. The experimental data referred to earlier all relate to radiation therapy. However, similar considerations may apply to chemotherapy or to a combination of radiation therapy and chemotherapy. As discussed in Chapter 4, some evidence exists to suggest that in some human malignancies,

26

PART 1  Principles

Split-course RT 66Gy/33 fractions/9.5 weeks (DAHANCA 2, 1979–1985)

Conventional RT 66Gy/33 fractions/6.5 weeks (DAHANCA 5 and 7, 1986–1996)

Accelerated RT 66Gy/33 fractions/5.5 weeks (DAHANCA 7, 1992–1996)

Tumor control (%)

80

9.5 weeks 6.5 weeks 5.5 weeks

60

77%

50%

65% 50%

55%

40 22% 20 0 Well-/moderately differentiated

Poorly differentiated

FIGURE 1-22.  Top, Overview of the fractionation schedules used in the three Danish Head and Neck Cancer (DAHANCA) trials. Bottom, Relationships of histopathologic grade, overall treatment time, and local tumor control in the three trials. Only the welldifferentiated or moderately differentiated tumors were significantly influenced by overall treatment time. (Modified from Overgaard J, Sand Hansen H, Overgaard M, et al. Importance of overall treatment time for the outcome of radiotherapy in head and neck carcinoma. Experience from the Danish Head and Neck Cancer Study, 6th International Meeting on Progress in Radio-Oncology, Salzburg, Austria, 13-17 May 1998. Bologna, Italy: Monduzzi Editore, 1998, pp 743-752.)

the results of radiation therapy are poorer if it is preceded by a course of chemotherapy. Accelerated repopulation that is triggered by the chemotherapy may be the ­explanation.

Effect of Tumor Volume on Tumor Control by Irradiation Presumably, larger tumors need larger irradiation doses if those tumors are to be controlled. However, this topic cannot be dealt with rigorously by using radiobiologic principles. Ultimately, the most pertinent information must be derived from clinical experience. Information from animal models cannot be extrapolated to the human situation, because no information on scaling factors is available. ­Another problem is the absence of experimental data on volume-­related variation in the tolerance of normal tissue to a given radiation dose because this tolerance ultimately limits the dose delivered and the strategy adopted. Clinical experience with common epithelial tumors sheds some light on the issue of tumor control as a function of tumor volume. Because local control of very radiosensitive tumors (e.g., Hodgkin disease, nodular lymphoma, seminoma) can be achieved readily regardless of tumor size, the most meaningful data come from irradiating less radiosensitive epithelial tumors such as squamous cell carcinomas and adenocarcinomas (Table 1-4). Although most larger tumors require higher doses for control, some exceptions exist, such as squamous cell carcinoma of the glottis or the skin. In the glottis, the tumor volume is so small that sufficiently high doses have been used since the ­earliest years of radiation therapy. In the skin, the wide range of fractionation regimens used and the difficulty of comparing those regimens make these tumors poor models

for other squamous cell carcinomas arising in the mucosal surfaces or viscera.

Predicting Tumor Sensitivity or Resistance to Radiation In day-to-day practice, radiation treatment schedules are designed for the average patient with a given type of malignancy at a given anatomic site. Considerable effort has been devoted to the identification and development of assays to predict the sensitivity of a particular tumor or normal tissue in an individual patient, with the goal of tailoring the treatment to each individual.51 No assay is suitable for routine use in the clinic, but the potential value of such an assay is enormous. Predictive assays ­focus on three factors: the intrinsic radiosensitivity of the ­tumor and normal tissues (which limits the dose that can be given), the tumor’s oxygen status, and proliferative ­potential of the tumor.

Intrinsic Radiosensitivity The radiosensitivity of normal tissues and tumors varies considerably among individuals; for any group of patients given the same treatment, some will experience more severe reactions than others, and a small proportion will experience unacceptable late sequelae. The means of prospectively identifying in advance which individuals are particularly sensitive to radiation would be of tremendous benefit for clinical treatment. Unfortunately, no assay has been identified that would be suitable for routine clinical use. A group at The University of Texas M. D. Anderson Cancer Center found a link between late effects and the radiosensitivity of fibroblasts grown from skin biopsy specimens.52 West and colleagues53-56 in the United Kingdom

1  Physical and Biologic Basis of Radiation Therapy 1

20 Gy of 300-kV x-rays

10

27

2

1

A

2

1

10−1

10−2

1

10−3

10−1

10−4

10−2

10−5

−20 −16 −12 −8 −4

10−3 0

4

8

Fraction of clonogenic cells

Relative tumor volume

2

12 16 20 24 28 32

B

Days after irradiation

FIGURE 1-23.  Growth curves of a rat rhabdomyosarcoma show shrinkage, growth delay, and subsequent recurrence after treatment with a single dose of 20 Gy of x-rays. A, Curve 1 is the growth curve of unirradiated control tumors; curve 2 is the growth curve of tumors irradiated at time t = 0, showing tumor shrinkage and recurrence. B, Variation in the fraction of clonogenic cells as a function of time after irradiation, obtained by removing cells from the tumor and assaying for colony formation in vitro. (From Hermens AF, Barendsen GW: Changes of cell proliferation characteristics in a rat rhabdomyosarcoma before and after irradiation. Eur J Cancer 1969;5:173-189.)

TABLE 1-4.   Relationships of Biologic Dose, Tumor Size, and Control by Irradiation Total Dose (Gy)*

Control (%)

Histology

Size

Squamous Adenocarcinoma

Subclinical (<106 cells)

95+

60

Squamous

<2 cm >4 cm

85 50

65

Squamous

2-4 cm

70

70

Squamous Adenocarcinoma

2-4 cm >4 cm

90 60

75+

Squamous

>4 cm

90

50

*Approximation

based on a minimum tumor dose of 2 Gy per f­raction and five fractions per week. Data from Shukovsky LJ, Fletcher GH. Time-dose and tumor ­volume relationships in the irradiation of squamous cell carcinoma of the tonsillar fossa. Radiology 1973;107:621-626; Fletcher GH, ­Shukovsky LJ. The interplay of radiocurability and tolerance in the irradiation of human cancers. J Radiol Electrol Med Nucl 1975;56:383-400; Fletcher GH, Shukovsky LJ. Isoeffect exponents for the ­production of dose-response curves in squamous cell carcinomas treated between 4 to 8 weeks. J Radiol Electrol Med Nucl 1976;57:825-827; Perez CA, Brady LW. Overview. In Perez CA, Brady LW (eds): Principles and Practice of Radiation Oncology, 2nd ed. Philadelphia: JB Lippincott, 1992, pp 1-63.

impractical for clinical use. Perhaps a more likely possibility for the future may be to screen patients who are to undergo radiation therapy for the presence of genes that may confer radiosensitivity or radioresistance.

Oxygen Status The oxygen status of a tumor can be assessed by the deposition of labeled nitroimidazoles in the tumor or with polarographic oxygen probes. Compounds labeled with the ­short-lived, gamma-emitting isotope iodine 123 (123I) also can be used in regions of low oxygen tension. The presence of hypoxia can be visualized by single photon emission computed tomography (SPECT).57 Eppendorf probes have a very fast response and can be moved quickly through a tumor under computer control to obtain an oxygen profile directly. A clinical trial in Germany of patients with advanced carcinoma of the cervix treated with radiation therapy has shown lower overall and recurrence-free survival rates among patients whose tumors exhibited median partial oxygen pressure (Po2) values less than or equal to 10 mm Hg as measured with the Eppendorf probe.58 Although this evidence was initially interpreted to show that radioresistance caused by hypoxia was important, later studies indicated that hypoxia leads to more aggressive and malignant tumors.59

Proliferative Potential found that radiation therapy was less likely to ­ produce local control of carcinoma of the cervix when the SF2   (i.e., fraction of tumor cells surviving 2 Gy) was greater than 0.55. However, no other studies have shown such promising results. Attempts to develop more rapid assays (e.g., assessments of chromosomal aberrations, DNA damage, cell growth rather than clonogenicity) have proved

The proliferative potential of a tumor can be expressed in terms of its Tpot, which takes into account cell cycle time and growth fraction but not cell loss. Tpot can be measured by labeling tumor biopsy specimens with BrdU and then using flow cytometry to estimate doubling time. In a trial of head and neck cancer conducted by the EORTC, local control of fast-growing tumors (Tpot < 4 days) was ­better

28

PART 1  Principles Dose (Gy) 0

2

4

6

8

100

10 12 14 16 18 20 22 24

1

Surviving fraction

Surviving fraction

OER = 2 at doses below 2 Gy

0.1

10−1

10−2

0.01

OER = 3 at large doses

FIGURE 1-24.  Cells irradiated in the presence of molecular oxygen are more sensitive to killing by x-rays than cells that are hypoxic (deficient in oxygen). The ratio of doses that produce the same level of biologic damage in the absence of oxygen and in the presence of oxygen is known as the oxygen enhancement ratio (OER). At high doses, the OER has a value of about 3; its value is close to 2 at doses below 2 Gy.

after accelerated fractionation than after conventional therapy, but no difference was found in local control of slow-growing tumors after either type of therapy.60 A later study found no correlation between Tpot and outcome after radiation therapy; although the LI and the length of the DNA-synthesis phase (Ts) correlated weakly with treatment outcome, that correlation was not strong enough for use as a predictor.61 Predictive assays are still in developmental stages. ­Although none of the markers explored has been reliable for predicting the response of a tumor to radiation therapy, some may show promise for identifying groups of patients who may benefit from altered treatment protocols such as accelerated treatment, hyperfractionation, bioreductive drugs, or neutron therapy.

Modifiers of Radiation Effects Oxygen and the Tumor Microenvironment The sensitivity of biologic material to sparsely ionizing radiation critically depends on the presence or absence of molecular oxygen. Survival curves for mammalian cells exposed to x-rays in the presence or absence of oxygen are shown in Figure 1-24. The ratio of doses needed under hypoxic and aerated conditions to achieve a given level of biologic effect is the oxygen enhancement ratio (OER). At high radiation doses, the OER has a value of between 2.5 and 3, and some evidence suggests that it may be reduced to about 2 for doses below 2 Gy. The magnitude of the oxygen effect seems to be the same for bacteria, plants, mammalian cells, and even some chemical systems, despite great differences in dose ranges. For oxygen to act as a sensitizer, it must be present during the radiation exposure, or at least during the lifetime of the free radicals that are involved in the indirect action of radiation, which is about 10−5 seconds. According to the oxygen fixation hypothesis, in the absence of oxygen, the chemical changes produced in the target molecule may

10−3 0

10

20

30

Dose (Gy)

FIGURE 1-25.  A biphasic survival curve is characteristically obtained for tumors from air-breathing mice. The steep initial portion of the survival curve reflects the killing of aerated cells; the shallower tail of the curve reflects the death of the more radioresistant hypoxic cells. (Courtesy of Sara Rockwell, MD, Yale University School of Medicine, New Haven, CT.)

be repaired; if oxygen is present, the damage is fixed (i.e., made permanent and irreparable). A question of obvious importance concerns the concentration of oxygen that is required to potentiate the effect of x-rays. The answer is that only a very small amount of ­oxygen is necessary to produce a sensitizing effect. A Po2 of about 3 mm Hg, corresponding to about 0.5%, is enough to take the radiation response halfway from fully hypoxic to fully oxygenated conditions. At concentrations characteristic of most normal tissues (about 30 mm Hg), the radiation response is essentially the same as that under fully aerated conditions or in the presence of 100% oxygen. The Importance of Oxygen in Tumor Response. Oxygen was suspected of influencing the radiosensitivity of tumors in vivo as early as the mid-1930s. However, in 1955, a paper written by Thomlinson and Gray62 led to ­overwhelming interest in the importance of oxygen in clinical radiation therapy. This paper, reporting results of a histologic study of bronchial carcinoma, concluded that tumor cords contained necrotic centers because of the limited distance that oxygen could diffuse in respiring tissue (100 to  180 μm). The authors proposed that close to the end of the diffusion range of oxygen, there would be a region, possibly two cell layers thick, in which the oxygen tension would be low enough to be intransigent to killing by x-rays but high enough to be viable. (This concept was later recognized as the mechanism underlying chronic hypoxia.) They postulated that these cells might constitute a focus for tumor regrowth and represent the limiting factor in the curability of some human tumors by x-rays. With the development of quantitative assay systems for transplanted tumors in experimental animals, by 1963, these tumors were shown to contain a proportion of viable hypoxic cells, and the

1  Physical and Biologic Basis of Radiation Therapy

Anoxic necrotic cell

Chronic hypoxia

Acute hypoxia

Normoxic

O2 70 µm

FIGURE 1-26.  Diagram illustrating the difference between chronic and acute hypoxia in a tumor. Chronic hypoxia results from the limited diffusion distance of oxygen in a respiring ­tissue that is actually metabolizing oxygen. Cells that become ­hypoxic for this reason remain hypoxic for long periods of time and eventually die and become necrotic. Acute hypoxia results from the temporary closing of tumor blood vessels. Some cells are intermittently hypoxic because normoxia is restored each time the blood vessel opens up again. At one moment one area of the tumor may temporarily be hypoxic, while minutes later a different area may be hypoxic. (Courtesy of J. Martin Brown, MD, Stanford University School of Medicine, Stanford, CA.) Modified from Hall EJ, Giaccia AJ. Oxygen effect and reoxygenation. In: Radiobiology for the Radiologist. 6th ed. Philadelphia: Lippincott Williams & Wilkins, 2006, pp 85-105.)

r­ esistance of these cells did constitute a limit for the sterilization of these tumors by single large doses of x-rays. Figure 1-25 shows a typical dose-response curve for the cells of an experimental animal tumor. Most of the cells have a dose response characteristic of fully aerated ­conditions, but some proportion of cells have a sensitivity characteristic of hypoxia. Many tumors of various histologic types have been evaluated in different animal species. The proportion of hypoxic cells in these tumors usually ranges from 10% to 15%. Although comparable data are not available for ­humans, experiments involving radiosensitizers indicate that the proportion of hypoxic cells in some skin nodules in humans may be similar. Chronic and Acute Hypoxia. Hypoxia in tumors can result from two quite different mechanisms (Fig. 1-26). As Thomlinson and Gray62 surmised, chronic hypoxia results from the limited distance that oxygen can diffuse in respiring tissues. Regions of acute hypoxia within a tumor result from the temporary blockage of a blood vessel. If the blockage were permanent, the cells downstream would eventually die and be of no further consequence. However, good evidence exists to suggest that tumor blood vessels open and close randomly and that different regions of the tumor become hypoxic intermittently. Acute hypoxia is the result of blood flow instability and is seldom a result of total ­stasis. At the moment when a radiation dose is delivered, some proportion of the tumor cells may be hypoxic; if the radiation were to be delayed, a different group of cells may be hypoxic. This concept of acute hypoxia was postulated

29

in the late 1970s by Brown63 and was later demonstrated unequivocally in rodent tumors. Areas of acute hypoxia and chronic hypoxia can be present in the same tumor,64 which may affect the choice of treatment for such tumors. One oxygen-sensing mechanism that has evolved to maintain cell and tissue homeostasis under chronic low-oxygen conditions involves the transcription factor hypoxia-inducible factor 1 (HIF-1) and its alpha subunit. The HIF-1α subunit protein is ­ rapidly degraded under aerobic conditions, but its stabilization under hypoxic conditions leads to the upregulation of several genes that promote tumor cell survival under those conditions. Preclinical studies have suggested that HIF-1α levels may be useful indicators of a tumor’s current ­oxygen status. Indicators of longer-lasting oxygenation conditions such as the chemical hypoxia marker pimonidazole have been explored in preclinical studies. The hope is that means of visualizing HIF1α levels, in combination with indicators of longer-lasting oxygenation conditions such as pimonidazole, can be used with clinical samples to indicate which areas of tumors may be experiencing transient changes in oxygen perfusion and to link the results with response to therapy.64 Reoxygenation. When an animal tumor that consists of a mixture of aerated and hypoxic cells is irradiated, the aerated cells are killed preferentially because of their greater sensitivity, and the survivors contain a preponderance of ­hypoxic cells. However, experimentation has shown that this situation is not static, but that over time after irradiation, the proportion of hypoxic cells tends to return to the preirradiation level. This shift of cells from the hypoxic to the aerated compartment is known as reoxygenation. Not surprisingly, the rapidity and extent of reoxygenation vary greatly among the different types of animal tumors that have been studied. In some tumors, such as mammary carcinomas, reoxygenation is rapid and complete. Predictably, these tumors can be readily eradicated, with acceptable normal tissue effects, by using a fractionated radiation schedule. Other animal ­ tumors, such as the transplanted osteosarcoma, reoxygenate inefficiently and slowly; they cannot be eradicated with any fractionated schedule of xrays without incurring substantial normal tissue damage. Reoxygenation cannot be studied in humans, and we are left to speculate on the patterns of reoxygenation in human tumors of different histologic types. The effect of reoxygenation on a fractionated course of radiation is illustrated in Figure 1-27. Each dose of ­radiation reduces the aerated compartment but has comparatively little effect on hypoxic cells. If the interval between radiation doses is long enough to allow reoxygenation to take place, hypoxic cells do not greatly affect the eventual outcome of the irradiation of the tumor. Few hypoxic cells are killed with each dose, but they are killed after they have become aerated as a result of the process of reoxygenation. In other words, tumors that reoxygenate quickly and efficiently can be sterilized with a fractionated course of x-rays. They cannot be sterilized with a single dose, because the resistance of the hypoxic cells would dominate. However, if the radiation is split into fractions delivered over time, with the interval between the fractions sufficient for reoxygenation to be complete, the presence of hypoxic cells has a minimal effect. Hypoxia and Tumor Progression. Evidence that lowoxygen conditions play an important role in malignant

30

PART 1  Principles REOXYGENATION

Ae

ra te d

ce

lls

X-rays

15% hypoxic cells

100% hypoxic cells

15% hypoxic cells

FIGURE 1-27.  Many animal tumors contain a mixture of aerated and hypoxic cells, with about 15% of the cells being deficient in oxygen. When a tumor is exposed to a dose of x-rays, a large proportion of the aerated cells is killed, and most of the surviving cells are hypoxic. If a time interval is allowed before a subsequent dose of radiation is administered, the initial pattern of aerated and hypoxic cells is restored. Consequently, if reoxygenation is rapid and complete between dose fractions, the presence of hypoxic cells does not affect the outcome of a multifraction regimen.

progression comes from studies of the correlation between tumor oxygenation and treatment outcome in patients and in laboratory studies of cells and animals. Findings from these studies are summarized briefly here. In a clinical study of radiation therapy for advanced cervical carcinoma performed in Germany, local control was found to be better for those tumors in which oxygen-probe measurements showed a Po2 of 10 mm Hg than for ­tumors in which the Po2 was less than 10 mm Hg. However, a similar improvement in outcome was observed for betteroxygenated tumors that were treated surgically rather than radiologically.54 This finding suggests that hypoxia may ­indicate tumor aggressiveness rather than conferring radioresistance to some cells. In a clinical study conducted in the United States, ­patients were given radiation therapy for soft tissue ­sarcoma; the extent of tumor oxygenation was found to be associated with the frequency of distant metastasis.66 Among patients with tumors having a Po2 of less than 10 mm Hg, 70% developed distant metastases, compared with 35% of those whose tumors had a Po2 of more than 10 mm Hg. These findings were particularly compelling because the primary tumor was eradicated in all patients regardless of the level of oxygenation; only the incidence of metastasis varied ­between groups. The results from this study argue strongly that tumor oxygenation level influences tumor aggressiveness. With regard to laboratory studies, at least three lines of evidence suggest a link between hypoxia and malignant progression.67 First, hypoxia seems to destabilize the genome by inducing gene amplification and increasing gene mutation. Although the mechanisms by which this takes place are far from clear, genomic instability results in the production of transformed cells with mutations that give them a survival advantage in adverse conditions. A second insight into how low-oxygen conditions may affect the aggressiveness of tumors is the observation that hypoxia ­initiates apoptosis in oncogenically transformed cells, functioning

to enhance the selection of cells with reduced apoptotic sensitivity. Cells with a defective apoptotic pathway from inactivation of the TP53 tumor suppressor gene or from overexpression of antiapoptotic genes such as BCL2 have a survival advantage in a low-oxygen environment. The third consequence of hypoxia is the induction of proangiogenic stimulatory factors. Of the factors described, vascular ­endothelial growth factor is highly specific in stimulating endothelial cell proliferation and migration.

Radiosensitizers of Hypoxic Cells Methods that have been proposed to eliminate foci of hypoxic cells within tumors, foci that are thought to lead to tumor regrowth after irradiation, include treatment in hyperbaric oxygen chambers, the use of chemical sensitizers, the use of hypoxic cytotoxins, and conversely, the use of radioprotectant substances to protect normal tissue. These methods are reviewed briefly in the remainder of this section; another approach, the use of high-LET radiation such as neutrons or heavy ions, is discussed under “Radiation Quality.” Hyperbaric Oxygen Treatment. After hypoxia was identified as being a possible source of tumor resistance, the use of hyperbaric oxygen was explored for cancer treatment, first by Churchill Davidson at St. Thomas Hospital in London and later by others. Clinical trials were performed with patients sealed in a chamber filled with pure oxygen raised to a pressure of 3 atmospheres. Results from these trials were difficult to interpret given the small numbers of subjects and the use of unconventional fractionation schemes involving a few large fractions delivered over short periods. The largest multicenter trials of hyperbaric oxygen, performed by the Medical Research Council in the United Kingdom, showed significant benefits in local control and survival for patients with carcinoma of the uterine cervix or advanced head and neck cancer but not for those with bladder cancer. Hyperbaric oxygen treatment has fallen into disuse, partly because of its cumbersome nature and partly because drugs were thought to be able to achieve the same end by simpler means. Chemical Radiosensitizers. Instead of trying to force oxygen into tissues by the use of high-pressure chambers, the emphasis in the chemical-radiosensitizing approach ­focused on the use of oxygen substitutes that diffused into poorly vascularized areas of tumors and achieved the ­desired effect by chemical means.68-70 The vital difference between these drugs and oxygen, on which their success depends, is that the sensitizers are not rapidly metabolized by the cells in the tumor through which they diffuse. They can penetrate further than oxygen and reach all of the ­hypoxic cells in the tumor, including those that are most remote from a blood supply. Three nitroimidazole-type radiosensitizers have been tested in clinical trials. Misonidazole produces appreciable sensitization of tumor cells in culture and in vivo. Of the 20 or so randomized, prospective, controlled clinical trials performed in the United States by the RTOG, none showed a statistically significant advantage for misonidazole, although several suggested a slight benefit. Moreover, the dose-limiting toxicity of misonidazole in clinical use was found to be peripheral neuropathy, which progressed to central nervous system toxicity if use of the drug was continued. This effect precluded misonidazole being used at adequate levels throughout multifraction regimens.

31

1  Physical and Biologic Basis of Radiation Therapy

Compounds Cytotoxic to Hypoxic Cells An alternative approach to designing drugs that preferentially radiosensitize hypoxic cells is to develop drugs that selectively kill hypoxic cells. It was pointed out at an early stage that the greater reductive environment of tumors might be exploited by developing drugs that are preferentially reduced to cytotoxic species in the hypoxic regions of tumors. The best-studied of these drugs is tirapazamine, which shows highly selective toxicity toward hypoxic cells in vitro and in vivo. Results from studies of this compound in laboratory and clinical trials are described in the ­following paragraphs. Laboratory Studies of Tirapazamine. Survival curves for Chinese hamster cells treated with various doses of tirapazamine are illustrated in Figure 1-28. The ratio of cytotoxicity to hypoxic cells versus aerated cells is about 100. In this system, tirapazamine is thought to be activated by the enzyme cytochrome P450. The smaller cytotoxicity ­ratio for cell lines of human origin (about 20) presumably reflects a different spectrum or different levels of enzymes.64. Results of an experiment in which a transplanted mouse carcinoma was treated with x-rays or tirapazamine, or both, are shown in Figure 1-29.72 The drug was injected  30 ­ minutes before each irradiation, which consisted of 2.5-Gy fractions given twice daily. The combination of ­tirapazamine and radiation was highly effective in ­reducing

101

O N

Surviving fraction

100

N

10−1

N NH2

O

Hypoxic cytotoxicity ratio

−2

10

10−3 10−4

Hypoxic cells Aerobic cells

10−5 100

101

102

103

104

Dose of SR-4233 (µM)

FIGURE 1-28.  Tirapazamine (SR-4233), synthesized by Stanford Research International, is cytotoxic to hypoxic cells. Its structure is shown in the inset. Dose-response curves are shown for Chinese hamster cells exposed for 4 hours to graded concentrations of the drug under aerated and hypoxic conditions. Cells deficient in oxygen are preferentially killed. (Courtesy of J. Martin Brown, MD, Stanford University School of Medicine, Stanford, CA.) 104 Median tumor volume (mm3)

The experience with misonidazole led to the evaluation of etanidazole, another 2-nitroimidazole with equivalent sensitization but much less toxicity than misonidazole. The lower neurotoxicity of etanidazole is a function of its ­shorter half-life in vivo plus its lower partition coefficient, so it penetrates poorly into nerve tissue and does not cross the blood-brain barrier. Controlled clinical trials by the RTOG in the United States and by a multicenter consortium in Europe showed no benefit from adding etanidazole to conventional radiation therapy. The third of the nitroimidazole-type radiosensitizers tested, nimorazole, is less effective as a radiosensitizer than misonidazole or etanidazole, but it is much less toxic, allowing very large doses to be given. In a Danish ­DAHANCA trial, this compound produced significant improvement in locoregional control rates and survival rates among patients with supraglottic or pharyngeal carcinoma relative to radiation therapy alone. Surprisingly, nimorazole has not been used elsewhere. Nicotinamide and Carbogen Breathing. Hypoxic cell radiosensitizers such as the nitroimidazoles were designed primarily to overcome chronic hypoxia, which is diffusionlimited hypoxia resulting from the inability of oxygen to diffuse further than about 100 μm through respiring ­tissue. However, hypoxia also arises through acute mechanisms (i.e., the intermittent blockage of blood ­vessels). ­Nicotinamide, a vitamin B3 analogue, has been shown in mouse tumors to prevent the transient fluctuations in tumor blood flow that lead to the development of acute hypoxia. The combination of nicotinamide to overcome acute ­hypoxia with carbogen breathing to overcome chronic ­hypoxia is the basis of the ARCON trials being ­conducted in several European centers. These trials also involve ­accelerated and hyperfractionated radiation therapy to avoid tumor proliferation and damage to late-responding normal tissues.

TPZ Saline

X-rays alone

103 X-ray + TPZ TPZ + x-ray

102

8 × 2.5 Gy 10

1

0

10

20

30

40

50

Days after first treatment

FIGURE 1-29.  Tumor volume is shown as a function of time after various treatments of a squamous cell carcinoma (SCCVII) transplantable mouse carcinoma. Tirapazamine (TPZ) or radiation alone had little effect. The combination of TPZ and x-rays caused significant growth delay, and giving radiation after the drug produced a slightly greater effect. (Modified from Brown JM, Lemmon MJ. Tumor hypoxia can be exploited to preferentially sensitize tumors to fractionated irradiation. Int J Radiat Oncol Biol Phys 1991;20:457-461.)

tumor ­ volume; giving the drug before the radiation was slightly more effective than giving the treatments in the reverse order. The effect of the combination was greater than additive. In parallel experiments using the same  x-ray and drug protocols, no evidence of radiosensitization or ­ additive cytotoxicity (in terms of skin reactions) was produced by adding tirapazamine to the radiation treatments. This finding substantiates the tumor selectivity of the radiation enhancement. Clinical Trials of Tirapazamine. Despite extensive laboratory findings in vitro and in vivo showing a ­greaterthan-additive effect for the combination of x-rays and

32

PART 1  Principles

t­ irapazamine, clinical trials have shown only modest success. In a phase III trial comparing cisplatin or cisplatin combined with tirapazamine for advanced (stage IIIB or IV) non–small cell lung cancer,73 patients given the combination therapy had twice the response rate and significantly longer survival than patients given cisplatin alone. Although tirapazamine did not increase systemic cisplatin toxicity, some nausea and muscle cramping were reported. The only reported trial showing a benefit from the combination of tirapazamine with radiation comes from Australia. In that trial, Rischin and colleagues74,75 showed tirapazamine, cisplatin, and radiation to be superior to fluorouracil, cisplatin, and radiation for patients with locally advanced head and neck cancer. Meta-analysis of Clinical Trials. Of the dozens of clinical trials performed in attempts to overcome the perceived problem of hypoxic cells in tumors, most have shown inconclusive or borderline results. In 1996, Overgaard and Horsman76 published the results of a meta-analysis of 10,602 patients treated in 82 randomized clinical trials involving hyperbaric oxygen, chemical sensitizers, carbogen breathing, or blood transfusions; the tumor sites studied included bladder, uterine cervix, central nervous system, head and neck, and lung. Overall, antihypoxic cell treatments were found to improve local tumor control by 4.6% and survival rate by 2.8%. The complication rate increased by only 0.6%, which was not statistically significant. The ­largest number of trials involved head and neck tumors, which also showed the greatest benefit. Overgaard and Horsman also concluded from the meta-analysis that the problem of hypoxia may be marginal in most adenocarcinomas and most important in squamous cell carcinomas.

Radioprotective Compounds In 1948, Harvey Patt and colleagues77 discovered that cysteine afforded mice considerable protection from death by whole-body irradiation. The dose required to produce lethality was almost doubled in animals that had been given the drug compared with those that had not. The structure of cysteine is illustrated as follows: NH2 SH – CH2 – CH COOH Almost simultaneously in Europe, Bacq and colleagues78 found that cysteamine was also a protector. The structure of this compound is illustrated as follows: SH – CH2 – CH2 – NH2 In the years that followed, many similar compounds were found to be effective protectors, and they all tended to have the same general structure: a free sulfhydryl (SH) group (or a potential sulfhydryl group) at one end of the molecule and a strong basic function, such as an amine or guanidine, at the other end, separated by a straight chain of two or three carbon atoms. The problem with these simple compounds was their toxicity. In the years ­ immediately

after World War II, in the wake of the first use of atomic weapons, the possibility of a compound that would protect against whole-body irradiation was of great interest to the military. Over the years, the Walter Reed Army Hospital in Washington, DC, synthesized several thousand compounds that were similar in structure to cysteine and cysteamine, with the aim of finding something less toxic. The first breakthrough in this research was the discovery that covering the sulfhydryl group with a phosphate allowed the dose resulting in a given level of toxicity to be doubled. Amifostine and Radiation Therapy. Of the many compounds produced in the Walter Reed program, the one used most often in radiation therapy is amifostine (WR2721). Amifostine is a prodrug; the presence of a terminal phosphorothioic acid group makes it relatively nonreactive, and it does not readily permeate cell membranes. Dephosphorylation by alkaline phosphatase, which is present in high concentrations in normal tissues and capillaries, converts amifostine to its active form, WR-1065. This metabolite readily enters normal cells by facilitated diffusion, where it scavenges free radicals generated by ionizing radiation or by some alkylating chemotherapeutic agents. It is taken up much more slowly by tumor tissues, perhaps because of the superior vascularization of normal tissues or by differences in the membrane structure of the tumor cells that impedes entry of this relatively hydrophilic compound. Whatever the underlying mechanism, the compound quickly floods normal tissues but penetrates tumors more slowly. Experiments with animal tumors have confirmed that concentrations of amifostine are high in several normal tissues shortly after drug administration but rise much more slowly in tumors. The extent to which amifostine protects normal tissues from radiation effects varies considerably among ­ tissue types.79 The hematopoietic system, the gut lining, and the salivary glands are generally well protected. However, because the drug does not cross the blood-brain barrier, it provides no protection to the brain and a disappointingly low level of protection to the lung. Ideally, for the maximum radioprotective effect, radioprotectants such as amifostine should be administered at the maximum tolerable concentration immediately before the radiation dose is delivered, so that the drug can be present in and protect normal tissues—but not tumor cells—when the radiation is delivered. Although radioprotectants such as amifostine have several potential applications in combination with radiation therapy, their clinical use has been slow in coming. Phase I toxicity trials of amifostine conducted in humans in the United States have shown that the dose-limiting toxicity is hypotension; this and other adverse effects such as sneezing and somnolence have tended to limit the amount of drug given to less than the dose needed to achieve maximum protection according to animal experiments. In a randomized clinical trial conducted in mainland China,80 100 patients with inoperable, unresectable, or recurrent adenocarcinoma of the rectum were stratified and randomly assigned to receive amifostine plus radiation therapy or radiation therapy only. The amifostine was administered 15 minutes before the radiation, 4 days each week, for 5 weeks. The amifostine group showed protection of the skin, mucous membranes, bladder, and pelvic structures against late,

1  Physical and Biologic Basis of Radiation Therapy

moderate, and severe reactions; none of the 34 evaluable patients in this group had such reactions, compared with 5 of 37 evaluable patients in the radiation-only group, a statistically significant difference. Moreover, amifostine afforded no apparent protection to the tumors. One of the worrisome factors in the experimental use of radioprotectors in the clinic is that their use is not failsafe. To exploit a benefit, radiation doses must be increased, with the confidence that the normal tissues are protected and that the extra dose will improve tumor response. If radioprotection does not occur, unacceptable normal tissue morbidity results. A more modest but achievable goal is to use a radioprotectant to reduce the troublesome side effects of radiation therapy. The RTOG conducted a phase III randomized clinical trial in which amifostine was found to reduce xerostomia (dry mouth) without affecting early tumor control in patients with head and neck cancer given radiation therapy.81 The drug was administered daily, 30 minutes before each dose fraction in a multifraction regimen. Three months after treatment, the incidence of xerostomia was significantly reduced in patients treated with amifostine; these patients also reported having less difficulty in eating or speaking and in the need for fluids and oral comfort aids. No difference was found in locoregional tumor control rates between patients given amifostine and those who were not. Radioprotectors and Chemotherapy. Although sulfhydryl compounds were initially developed as protectors against ionizing radiation, they also protect against the cytotoxic effects of several chemotherapeutic agents. Amifostine, for example, offers significant protection against the nephrotoxicity, ototoxicity, and neuropathy associated with cisplatin and against the hematologic toxicity associated with cyclophosphamide. In a trial conducted at the M. D. Anderson Cancer Center, investigators tested whether amifostine would reduce the acute toxicity associated with concurrent chemoradiation therapy.82 Amifostine was given twice each week at 20 to 30 minutes before doses of intravenous cisplatin, oral etoposide, or irradiation. Amifostine did reduce the incidence and severity of esophageal, pulmonary, and hematologic toxic effects but did not ­ affect survival time, suggesting that it did not have a

3-MeV proton track

Heavy ion track

33

tumor-­protective effect. This and other studies indicated that ­amifostine has no obvious antitumor activity, suggesting that its uptake by normal tissues may be different from that by malignant tissues. Mechanism of Action. The mechanisms most often implicated in sulfhydryl-mediated cytoprotection include free-radical scavenging, which protects against oxygenbased free radicals generated by ionizing radiation or ­alkylating agents, and hydrogen-atom donation, which facilitates direct chemical repair at sites of DNA damage. The protective effect of sulfhydryl compounds tends to parallel the oxygen effect and is maximal for sparsely ionizing, low-LET radiation such as x-rays or gamma rays and minimal for densely ionizing radiation such as low-energy alpha particles, which tend to damage DNA directly rather than generating free radicals.

Radiation Quality The energy deposited in biologic materials by radiation is in the form of ionizations and excitations that are not distributed at random; they tend to be localized along the tracks of individually charged particles. The pattern of this energy deposition varies substantially with the type of ionizing radiation involved (Fig. 1-30). For example, photons give rise to fast electrons, particles carrying a unit-negative electric charge and having a very small mass. Neutrons give rise to recoil protons or alpha particles, which are particles carrying 1 or 2 units of positive electric charge and having a mass approximately 2000 and 8000 times, respectively, that of an electron. The spatial distribution of the ionizing events they produce differs markedly. The tracks of energetic electrons, produced as a result of the absorption of x-ray photons, result in primary events that are well separated in space, and for this reason, x-rays are described as sparsely ionizing. Alpha particles give rise to individual ionizing events that occur close together, giving rise to tracks that consist of a column of ionizations; they are therefore said to be densely ionizing. The computer simulation of ionizing events from several different types of radiation against the background of a mammalian cell shown in Figure 1-30 is an attempt to illustrate the wide differences in ionization density associated with different types of radiation in common use.

Delta ray

Electron from 250-kV x-rays

FIGURE 1-30.  Computer simulations demonstrate sections of charged-particle tracks produced by different types of radiation passing through a strand of chromatin. Each cross represents a single ionization of the chromatin or the surrounding medium. Right track, Low–linear energy transfer (LET), 100-keV electron is typical of those produced by 250-kVp x-rays. Center track, High LET, high-energy iron ion produces a dense column of ionization; notice the high-energy, secondary delta ray coming out of the track. Left track, Medium LET, 3-MeV proton. The scale bar represents 50 nm. (Electron micrograph of the 30-nm chromatin fiber courtesy of Barbara Hamkalo, MD, University of California, Los Angeles, CA; calculated particle tracks and diagram courtesy of David Brenner, MD, New York, NY.)

34

PART 1  Principles Dose (Gy) 0

2

4

6

8

10

12

14

6

2

4 RBE

2

X-rays

0.1

OER

Surviving fraction

OER

8

1.0 RBE

1.0

0.1

3

10

0 0.1

1

10

100

1 1000

LET ∞, keV/µ RBE = 4.1

Neutrons RBE = 2.7

0.01 0

2

4

6

0.01 8

10

12

14

FIGURE 1-31.  The survival curve for x-rays is characterized by a broad initial shoulder, whereas for neutrons, the survival curve has little or no shoulder. Consequently, the relative biologic effectiveness (RBE) gets larger as the dose gets smaller. When a dose is fractionated, the RBE is larger for a given level of cell killing than if the dose is given in a single exposure because the large shoulder of the x-ray dose-response curve is repeated each time.

FIGURE 1-32.  Variation of the oxygen enhancement ratio (OER) and the relative biologic effectiveness (RBE) is a function of the linear energy transfer (LET) of the radiation involved. The data were obtained by using T1 kidney cells of human origin, which were irradiated with various naturally occurring alpha particles or with deuterons accelerated in the Hammersmith cyclotron. The RBE increases rapidly, and OER falls rapidly at about the same LET (about 100 keV/μ). (Modified from Barendsen GW. Radiobiological dose-effect relations for radiation characterized by a wide spectrum of LET: implications for their application in radiotherapy. In Proceedings of the Conference of Particle ­Accelerators in Radiation Therapy, Los Alamos Science Laboratory, Los Alamos, NM, October 2-5, 1972. LA-5180-C, 120-125. Oak Ridge, TN: U.S. Atomic Energy Commission, Technical Information Center, 1972.)

Linear Energy Transfer The quality of the radiation is described quantitatively in terms of LET, which is defined as the average energy imparted locally to the absorbing medium per unit length of track. The unit in which LET is commonly expressed is keV/μm; the kiloelectron volt is a unit of energy, and the micron (micrometer) is a unit of length. It is evident from Figure 1-30 that LET is very much an average quantity. In some regions along the track, for a considerable distance on the microscopic scale, no energy at all is deposited. In other regions, a cluster of ionizations occurs, and a large amount of energy is deposited in a small distance. The LET of 60Co gamma rays is about 0.2 keV/μm; that of 250-keV x-rays is about 2 keV/μm; that of 14-MeV neutrons is about  75 keV/μm; and that of heavy charged particles is ­anywhere from 100 to 2000 keV/μm. In general, for a given type of radiation, the LET goes down as the energy goes up. For ­example, 250-keV x-rays have a higher LET (and are more effective biologically) than are cobalt gamma rays. Likewise, for a given type of charged particle (e.g., alpha particle), a high-energy particle has a lower LET and less biologic ­effect per unit dose than a low-energy particle.

Linear Energy Transfer and Shape of Dose-Response Curve Dose-response curves from mammalian cells exposed to low-energy (high-LET) neutrons and low-LET x-rays are shown in Figure 1-31. When the LET of the radiation increases, the slope of the survival curve gets progressively steeper, and the shoulder gets smaller. The RBE of a given type of radiation is expressed in terms of standard radiation, usually taken to be low-LET photons having a LET value between 0.2 and 3 keV/μm. The RBE of neutrons is defined as the ratio of doses of x-rays to neutrons required to produce a given biologic effect. What biologic effect? No particular biologic effect is specified. At a surviving fraction of 10−2, the ratio of doses to produce the same effects is 2.7, whereas at a surviving fraction of 0.5,

the RBE is 4.0. The important point to make is that the value of the RBE for a given type of radiation (in this case, neutrons) varies with the level of biologic damage involved or, what amounts to the same thing, the size of dose involved. The effect of fractionation on RBE is illustrated in ­Figure 1-31. When a dose of x-rays or neutrons is split into multiple fractions, separated by sufficient time to allow sublethal damage repair to be completed, the shoulder of the doseresponse curve must be re-expressed with each fraction. For a given level of biologic damage, it is obvious that the RBE is greater for a schedule that involves multiple fractions than for a single exposure. The reason for this is that the large shoulder of the dose-response curves for x-rays must be repeated with each exposure, whereas for neutrons, only a small shoulder is repeated each time. For the fractionated regimen, the value of the RBE involved is that characteristic of the individual-dose fraction.

Relative Biologic Effectiveness and Linear Energy Transfer The variation of RBE as a function of LET has a characteristic shape (Fig. 1-32).83 As LET increases from 1 to 10 keV/μm, RBE increases slowly but steadily and then increases more rapidly beyond 10 keV/μm to reach a peak at about 100 keV/μm, after which RBE tends to fall for higher values of LET. The decrease of RBE at very high values of LET usually is attributed to the phenomenon of overkill. At these very high LETs, the density of ionization is such that if a charged particle track passes through a cell, it deposits far more energy than is required to kill the cell. Consequently, much of this energy is wasted because the cell cannot be killed more than once. Because so much ­energy is deposited in each cell, the radiation is relatively inefficient, and the killing effect per unit dose (which is what RBE amounts to) decreases.

1  Physical and Biologic Basis of Radiation Therapy

Linear Energy Transfer and Oxygen ­ Enhancement Ratio As LET increases, a corresponding decrease occurs in the dependence of cell killing on the presence of molecular oxygen, which is expressed in terms of the OER. This variation with LET is illustrated in Figure 1-32.83 ­­  Low-LET forms of radiation, such as x-rays or gamma rays, are characterized by a large OER of between 2.5 and 3. Heavy charged particles with an LET of several hundred keV/μm have an OER of 1 (i.e., cell killing is independent of the presence or absence of oxygen). ­Between these two extremes, neutrons with an LET of about 70 keV/μm show an intermediate OER of about 1.6. The value of the OER reflects the relative importance of the direct-killing effect of the radiation versus the indirect-killing effect, which is mediated by free radicals; this latter component of radiation damage depends on the presence or absence of oxygen.

Radiation Quality and Biologic Effectiveness RBE varies with radiation quality. In general, the more densely ionizing the radiation, the more biologically effective it is, at least up to a certain point. The biologic effectiveness of a certain type of radiation depends on several factors, including radiation quality, dose per fraction, dose rate, the presence or absence of oxygen, and the biologic system or cell type involved. In this context, radiation quality is measured in terms of the LET. The RBE depends on the dose level and the ­number of dose fractions (or, alternatively, the dose per fraction), because in general, the shapes of the dose­response relationship are different for types of radiation that differ substantially in LET. The RBE can vary with dose rate, too, because the slope of the dose-response curve for sparsely ionizing radiation varies with changing dose rate. In contrast, the biologic response to densely ionizing radiation depends little on the rate at which the radiation is delivered. The biologic system, end point, or cell type chosen to determine RBE has a marked influence on the ­values obtained. RBE values usually are high for tissues that ­ accumulate and repair a large amount of sublethal damage and consequently have large shoulders to their xray dose-response relationships, and RBE values are low for those that do not.

Use of Neutrons in Radiation Therapy Neutrons were first used in radiation therapy in the 1930s at the University of California at Berkeley. Their introduction was not based on any physical or radiobiologic ­rationale, and they were abandoned because of severe late effects produced in many patients. After World War II, the use of neutrons was reintroduced at the Hammersmith Hospital in England because neutrons have a lower OER than x-rays and because of the premise that hypoxic cells limit the curability of tumors by x-rays.84-86 After ­apparent early success, neutron machines were used for therapy in the United States, continental Europe, and ­ Japan. Controlled clinical trials for a wide range of tumors indicated little or no advantage for the new modality, except perhaps for a few disease sites, specifically prostate cancer, salivary gland tumors, and possibly soft tissue ­sarcomas.87,88 The biologic properties of neutrons differ from those of x-rays in several respects, apart from their lesser dependence on molecular oxygen for cell killing. Neutrons tend

35

to be associated with less repair of sublethal and potentially lethal damage, and they are associated with smaller variations in radiosensitivity according to the phase of the cell cycle. Perhaps because of these differences, neutron RBE values tend to be higher for slowly growing tumors. The rationale for the use of neutrons has evolved considerably over the years, but for the most part, the use of neutrons for radiation therapy has been abandoned.

Use of Protons in Radiation Therapy The biologic properties of high-energy protons do not differ significantly from those of x-rays. Protons produce ionizations and excitations in atoms along a track in much the same way as electrons are set in motion when x-rays are absorbed. The clinical use of protons is based on the superior dose distributions that can be obtained. Protons have a place in the treatment of tumors for which dose localization is important, such as tumors that are close to sensitive normal structures. Protons have been widely used to treat choroidal melanoma of the eye, sarcoma of the base of the skull, and chordomas. The increasingly impressive results of conformal radiation therapy, including intensity-modulated radiation therapy, in decreasing effects on normal tissues and allowing much higher total doses than have been used previously, suggest wider applicability of proton therapy. As a consequence, several elaborate proton facilities have been built or are under construction in the United States, Japan, and  Europe.89,90 This issue is discussed further in Chapters  2 and 40.

Use of Carbon Ions in Radiation Therapy Early investigations of the use of heavy ions for radiation therapy were done in the 1950s at the Lawrence Berkeley National Laboratory in the United States. Although interest in this therapy has declined over time in the United States, interest has been rekindled in Europe and Japan, with attention focused largely on high-energy carbon ions.91 The characteristic depth-dose profile of heavy ions, which they share with protons (i.e., dose increases with the depth of penetration), make them an attractive modality for the irradiation of deep-seated tumors. Another advantage shared by high-energy carbon ions (typically 300 MeV/μm 12C  ions) and protons is attractive physical dose distributions. Carbon ions also produce a substantial increase in RBE toward the end of the particle range. In principle, the good news of this latter feature is that the end of the particle’s range, and therefore the greatest antitumor effectiveness, can be located within the tumor. The bad news is that the appropriate RBE level to be used is unknown, which introduces an element of uncertainty. Compared with protons, carbon ions show less lateral scattering, leading to sharper beam edges (which is good). However, carbon ions produce a tail of lighter fragments beyond the Bragg peak, and they do not share the advantage of protons that the dose stops sharply at the end of the range of the primary particle.

Adjuncts to Irradiation Hyperthermia The use of hyperthermia to treat cancer has a history much longer than that of radiation. The first references are in an Egyptian papyrus roll 5000 years old. Just before the discovery of x-rays, it was observed that tumors regressed in

36

PART 1  Principles

patients with high fevers resulting from various bacterial infections.92 The interest in hyperthermia in modern times results from experiments with cells in vitro and with animal ­tumors and normal tissues. Several biologic effects of heat seem favorable for cancer therapy: • Heat kills; survival curves for heat are similar in shape to those for radiation, except that time at the elevated temperature replaces absorbed dose. • The age-response function for heat complements that for x-rays. S-phase cells that are resistant to x-rays are also sensitive to heat. • Cells that are at low pH and nutritionally deprived (more likely in tumors) are more sensitive to heat, although cells can adapt to pH changes in 80 to 100 hours and can lose their sensitivity to heat. • Hypoxia does not protect cells from heat as it does from x-rays. • Heat preferentially damages tumor vasculature. After heating, blood flow goes down in tumors but increases in normal tissues. This may result in an enhanced temperature differential between tumors and normal tissues. Good evidence exists to support this contention in transplantable animal tumors, but it is less clear in spontaneous human tumors. • Mild hyperthermia can promote tumor reoxygenation; this has been shown in animal experiments and in clinical studies. Although the biologic properties of heat seem favorable for its use in cancer therapy, physical devices that can produce uniform heating at a particular depth are not well developed, and adequate thermometry is also difficult. The basic laws of physics make the desired end difficult or impossible to achieve. Microwaves provide good localization at shallow depths, but localization is poor at greater depths, where low frequencies are needed. With ultrasound, deep heating can be achieved with focused arrays, but bones or air cavities cause distortions. Because of these considerations, in practice, primary or recurrent breast tumors and melanomas can be heated adequately with microwaves, and deep-seated tumors below the diaphragm can be ­heated adequately with ultrasound. For tumors at any other site, however, the current methods of heating pose a major problem. In contrast, the combination of interstitial implants of radioactive sources with hyperthermia generated by radiofrequency power applied to the implanted sources seems to be particularly promising because a good radiation dose distribution is combined with a good heat distribution. Many thousands of patients have been treated with ­hyperthermia. The general consensus is that hyperthermia alone has no role in the curative treatment of cancer; its usefulness is limited to palliation or retreatment of recurrent tumors. However, several clinical studies have shown that the combination of hyperthermia plus radiation produces more complete and partial tumor responses than does radiation alone. Several phase III randomized trials have shown a clear and significant benefit from the addition of hyperthermia to standard radiation therapy.93,94 The tumor sites evaluated include those of superficial localized breast cancer, recurrent or metastatic melanoma, and nodal metastases from head and neck cancer. In contrast, trials involving advanced breast cancer and neck nodes with 

full-dose radiation therapy did not show any benefit from the addition of hyperthermia. A novel and easily achievable means of producing modest levels of hyperthermia (41° to 42°C) is to trigger the ­release of drugs encapsulated in liposomes.95 Liposomes are spherical constructs, usually membranes, that can be filled with a cytotoxic chemotherapy agent such as doxorubicin or cisplatin. Injected liposomes tend to become trapped in the chaotic disorganized blood vessels of tumors, and the external application of a modest level of hyperthermia can cause them to extravasate into the tumor ­interstitium, depositing the chemotherapeutic agent. This form of ­hyperthermia-triggered drug release can have substantial therapeutic effects.96 In summary, the clinical value of hyperthermia for the routine treatment of cancer is still not clear, despite its proven efficacy in a few specific instances. The situations in which hyperthermia seemed to be effective involved its combination with external beam radiation therapy for the treatment of superficial tumors, for which adequate heating is not so difficult, or its combination with brachytherapy, for which heating is induced by implanted electrodes, inevitably involving all the complications and limitations of accessibility.

Cytotoxic Drugs Drugs with anticancer activity interact in complex and only partially understood ways with ionizing radiation. ­Although the combination is often used in clinical settings, such use is based on an empiric rationale with little regard for cellular interactions. Possible interactions of drugs with radiation are illustrated graphically in Figure 1-33.97 The drugs under consideration in this section are those with known dose-response curves; the interactions are considered at least potentially subadditive. However, they may be additive with regard to their effects on tumors and normal tissues. The object in combining anticancer drugs with ionizing radiation is to improve the therapeutic ratio relative to the use of either agent alone. It is beyond the scope of this chapter to consider cooperative interactions in which the effect of one agent is local, the other is systemic, and no interaction on tumors or normal tissues is assumed. Little is known about the interactions of cytotoxic drugs and radiation at the subcellular or molecular levels. Although much is known about the mechanisms of action of the various classes of drugs, such as alkylating agents, antimetabolites, plant alkaloids, and antibiotics, little is known about specific drug-radiation interactions. Several mechanisms by which cell death can be ­increased in vitro by the combined use of drugs and ­radiation can be identified: S-phase cytotoxicity, cell cycle ­ redistributions that increase drug or radiation effects, and ­ prevention of repair of potentially lethal damage. ­ Cellular studies have suggested preferred sequencing and timing of drugs and radiation types, as well as interactions with other treatment modalities such as hyperthermia. However, no ­ consistent correlations have been found between interactions that are apparent at the cellular level and in vivo effects. For ­example, hydroxyurea, a cell cycle S-phase–specific agent, enhances radiation effects on cells in tissue culture. ­ However, with few exceptions, the effects of this combination in vivo and in clinical settings have been ­disappointing.

1  Physical and Biologic Basis of Radiation Therapy One agent inactive by itself Sensitization

Fixed doses

One D/E curve

Cooperation Enhancement

37

Two D/E curves Supraadditive

Positive interaction Noninteractive

Noninteractive

Negative interaction Protection

Antagonism

Inhibition

Subadditive

Protection

Conceivably additive response

FIGURE 1-33.  Suggested terminology for drug-radiation interactions is derived from four experimental situations: the drug is inactive by itself, fixed doses of drug and radiation are combined, the dose-response (D/E curve) for one agent is studied with and without a fixed dose of another agent, and dose-response curves for both agents are available. (From Steel GG. Terminology in the description of drug-radiation interactions. Int J Radiat Oncol Biol Phys 1979;5:1145-1150.)

Clinical data suggest that normal tissues are at least as likely as tumors, if not more so, to be affected by drug­radiation interactions. Notable examples of enhanced acute and late effects with radiation therapy include those seen with simultaneous administration of doxorubicin, dactinomycin, fluorouracil, methotrexate, bleomycin, nitrosoureas, and possibly etoposide. In 1980, Tubiana98 stated that “the use of drugs during radiation therapy has never improved control of ­local ­lesions or reduced frequency of metastases from any ­tumor.” Clinical data convincingly show that local control can be enhanced with simultaneous chemotherapy and irradiation. SchaakeKoning and colleagues99 reported statistically significant increases in locoregional control of non–small cell carcinoma of the lung when cisplatin was given, weekly or daily, simultaneously with thoracic irradiation. Concurrent fluorouracil and pelvic irradiation has been shown to decrease local failure rates when given after resection of adenocarcinoma of the rectum.100 Improved local control has resulted from the use of vincristine and cyclophosphamide (with or without dactinomycin and doxorubicin) for Ewing’s sarcoma of bone and for embryonal rhabdomyosarcoma of the upper aerodigestive tract. Small cell carcinoma of the lung has been controlled much more consistently with concurrent ­chemotherapy and radiation therapy than with radiation therapy or chemotherapy alone. A French cooperative group101 led by investigators from Tubiana’s institution showed a ­significant reduction in the frequency of distant metastasis and improved long-term survival with induction chemotherapy (i.e., vindesine, lomustine, cisplatin, and  cyclophosphamide) followed by thoracic radiation therapy for non–small cell carcinoma of the lung, despite local control not being improved. Herskovic and colleagues102   reported significant decreases in rates of local failure and distant metastasis and increases in survival when fluorouracil and cisplatin were given concurrently with moderate-dose radiation therapy (50 Gy given in 25 fractions) ­compared

with higher-dose radiation therapy alone (64.8 Gy in  36 fractions).

Targeted Therapy Achieving tumor response with traditional chemotherapy agents such as the alkylating agents cisplatin or fluorouracil inevitably involves substantial toxicity to self-renewing normal tissues such as the gastrointestinal tract and bloodforming organs that are characterized by large growth ­fractions. Newer targeted therapeutic agents show some promise when given in combination with radiation therapy in terms of enhancing tumor control with little increase in normal-tissue toxicity. Cetuximab (Erbitux) can be considered a model for this class of agent; other examples that target specific cellular signaling pathways include trastuzumab (Herceptin), imatinib mesylate (Gleevec), and rituximab (Rituxan). Cetuximab is a recombinant human/mouse chimeric monoclonal antibody originally produced in mammalian cell culture. Cetuximab specifically binds to the epidermal growth factor receptor, blocking the phosphorylation and activation of receptor-associated kinases and resulting in inhibition of cell growth, induction of apoptosis, and decreased production of matrix metalloproteinases and vascular endothelial growth factor. Studies of cetuximab for the treatment of xenografted tumors in mice indicate that cetuximab given alone produces only a modest tumor-growth delay but that it shows substantial interaction with radiation.103 These promising preclinical results led to a phase III trial that demonstrated that adding cetuximab to high-dose radiation therapy for locoregionally advanced squamous cell carcinoma of the head and neck produced statistically significant extensions in overall survival time with minimal enhancement of toxicity (mostly skin reactions).104 The substantial effects of cetuximab and other forms of targeted therapy in experimental systems have translated into relatively modest gains

38

PART 1  Principles

in the clinic; however, this approach continues to be the subject of intense research.

Biological Response Modifiers Biological agents, once limited to derivatives of CalmetteGuérin bacillus, are being discovered at a rate unimaginable only a few years ago. Such agents have been considered possibly restorative of cell-mediated immunity in patients with immune suppression, whether from cancer or the modalities with which cancer is treated. Very few laboratory studies and even fewer clinical investigations are available describing possible interactions between local irradiation and biological agents.

Gene Therapy The term gene therapy covers several different ap-  proa­­ches,105 but all require some means of introducing a gene into tumor cells. The most common of these means are viral vectors. Among the several types of vectors used, retroviruses are convenient to work with but can infect only ­ dividing cells; adenoviruses can infect dividing and quiescent cells, but they evoke an immune response that can make their repeated use difficult. Derivatives of the herpes simplex virus are attractive because of their large size, which allows more material to be packaged in them, but they can be pathogenic and difficult to control. In most cases, the virus is used as a simple vector to deliver a gene of interest to the cells. For safety’s sake, the virus in these cases is usually engineered so that it cannot replicate. In a few cases, the virus itself is designed to be cytotoxic, and some means is needed to selectively limit its cytotoxicity to tumor cells. Several different forms of gene therapy are described briefly in the following sections. Suicide Gene Therapy. The strategy in suicide gene therapy is to transduce cells with a gene that can convert an inert prodrug (administered separately) into a toxic agent. One system that has been extensively investigated is the combination of the herpes simplex virus (HSV) thymidine kinase gene (HSV-TK) with ganciclovir.100 Several steps are involved in suicide gene therapy. The virus containing the HSV-TK gene, which phosphorylates the ganciclovir, is injected into the tumor, and the ganciclovir is administered systemically. Within the transduced cells, phosphorylation of the ganciclovir by the thymidine kinase converts it into a cytotoxic agent, killing the cells. Significant numbers of nontransduced cells also are killed (i.e., bystander effect) through various other transport and immune mechanisms. An alternative suicide gene system involves cytosine deaminase, which converts 5-fluorocytosine (nontoxic) to fluorouracil (toxic) within the cells.107 A fusion gene that combines the HSV-TK and cytosine deaminase systems in a single viral vector has also shown promise. Suicide gene therapy has led to tumor growth delays and some outright cures in animal models, but it has been less successful in clinical trials. Because of the ­ limited ­efficiency of gene delivery, suicide gene therapy needs to be combined with conventional radiation therapy.106   Phase I/II clinical trials of double-suicide gene therapy in combination with radiation therapy for men with prostate cancer have produced encouraging results.108,109 Cytotoxic Viruses. A variant of gene therapy involves the construction of cytotoxic viruses that are engineered to replicate and kill only cells with mutations in TP53, which

influences a variety of critical cellular functions including transcription, cell cycle checkpoints, DNA repair, ­apoptosis, and angiogenesis. To the extent that mutant TP53 is a hallmark of cancer, this treatment distinguishes between ­normal and cancer cells on that basis. This approach has produced evidence of tumor growth arrest in animal models and in early clinical trials.110-112 Immunomodulatory Gene Therapy. The basis of the immunomodulatory gene therapy approach (e.g., cancer vaccines), also known as molecular immunotherapy, is to provoke or stimulate a cellular immune response to cancer cells by injecting a vaccine consisting of suspensions of irradiated tumor cells that have been genetically engineered to express immune stimulatory molecules or tumor-specific antigens. This approach has been somewhat successful in animal models but generally is effective against only small tumor burdens. Tumor Suppressor Gene Therapy. In its purest form, tumor suppressor gene therapy seeks to replace a gene in which a mutation initiates or otherwise enhances the malignant phenotype. The goal of treatment may be to modify cell growth, invasiveness, or metastatic potential as much as it is to kill the cell. The TP53 gene has received the most attention as a target for such therapy because it is so commonly mutated in human cancer.113-122 Phase I/II clinical trials of tumor suppressor gene therapy have shown some positive results in the treatment of non–small cell lung ­cancer. The chief limitations to tumor suppressor gene therapy are the lack of information on exactly which target genes are essential for inducing or maintaining the malignant phenotype and the apparent necessity of multiple genetic changes for cancer to develop. Radiation-Activated Gene Therapy. The strategy behind radiation-activated gene therapy, also called radiogenetic therapy, combines the physics of radiation-­targeting technology with molecular aspects of gene therapy.  A ­ radiation-inducible promoter sequence is linked to a cyto­toxic agent, which is then “turned on” only in carefully delineated radiation fields.123-125 In one animal model system, the radiation-inducible Egr1 promoter is linked with tumor necrosis factor α and then packaged in an adenovirus. The adenovirus is injected into the tumor, but the cytokine is activated only in the target volume delineated by the radiation field, thereby minimizing total-body toxicity. Released in the tumor, the tumor necrosis factor causes vascular destruction and apoptosis of tumor cells. Efforts are underway to improve this approach by identifying and using more radiation-specific or tumor-specific promoter genes and more effective toxic agents. The strategies described here have all had some measure of success in reducing tumor growth, if not eliminating tumors entirely, in animal model systems, and some have progressed to phase II trials in patients. Some forms of gene therapy claim to evoke bystander effects in which more cells are killed than are initially transduced. Nevertheless, all of the strategies share the difficulty of transducing therapeutic genes into sufficiently large proportions of tumor cells. Growth reduction is common, but cures are rare. For this reason, a logical strategy would be to combine one or more of the gene therapy approaches with radiation therapy or chemotherapy, with the hope of translating a small increase in cell killing from the gene therapy into

1  Physical and Biologic Basis of Radiation Therapy

a large increase in tumor control from the conventional therapies. The guiding principle must be to identify combinations that have additive or synergistic effect with regard to tumor cell kill but have tolerable, non-overlapping toxic effects on normal tissues.

Mutagenesis and Carcinogenesis When radiation is absorbed in biologic material, several consequences can result, including cell death, hereditary effects, and carcinogenesis. Cell death is the end point of principal concern in radiation therapy, in which the object of the treatment is to kill (i.e., remove the reproductive integrity of) as many malignant cells as possible. It is also the end point of concern for some radiation effects on the developing embryo and fetus, which are cell-depletion phenomena. Hereditary effects may be promoted. Radiation may affect the germ cells in a way that does not kill the cells but allows them to survive and carry with them some legacy of the radiation exposure. These effects will not be expressed until a later generation. Carcinogenesis may occur. The radiation may affect a somatic cell, leading to a point mutation or a chromosome rearrangement that results, for example, in the release or expression of an oncogene. This may result in leukemogenesis or carcinogenesis. Hereditary effects and carcinogenesis are potential hazards for people who are occupationally exposed to ionizing radiation and for patients who are long-term survivors of radiation therapy. These effects are discussed in greater detail in the remainder of this chapter.

Hereditary Effects As early as 1927, Müller reported that exposure to x-rays could cause observable hereditary mutations in Drosophila melanogaster, including changes of eye color, the appearance of vestigial wings, and recessive lethal mutations.126 With the advent of the nuclear age after World War II, ­hereditary effects were considered to be the principal hazard of ionizing radiation, in part because solid tumors had not yet appeared in atomic bomb survivors. Hereditary diseases are classified in three principal categories: mendelian, chromosomal, and multifactorial. Mendelian diseases are caused by mutations in a single gene; they can be autosomal dominant, autosomal recessive, or sex linked. Chromosomal diseases are caused by gross abnormalities in the structure or number of chromosomes. Multifactorial diseases have a genetic component, but transmission cannot be described as a simple mendelian pattern. By far the most common type of hereditary diseases, multifactorial diseases include everything from cleft palate to coronary heart disease. Estimating Risks of Radiation-Induced Hereditary Effects. In the 1960s, relative mutation rates were assessed in the megamouse project, in which about 7 million mice were used.127,128 A species was chosen that had seven easily identified specific locus mutations, six of which involved coat color changes and the seventh a shortened ear. Male and female animals were irradiated with graded doses, subsequently bred, and mutations scored in the offspring. Conclusions from this project are as follows: 1. The radiosensitivity of different mutations varies substantially, by a factor of about 20.

39

2. A dose-rate effect exists in the mouse, unlike Droso­ phila, such that spreading radiation over time greatly reduces the number of mutations observed. 3. The male mouse is much more sensitive than the female, and at low dose rates, essentially all of the genetic load is carried by the male. This may be an artifact of the species because murine oocytes are exquisitely sensitive to radiation, and if they are killed by the radiation, mutations produced in them cannot be expressed. 4. The number of mutations produced by a given dose of radiation drops if a time interval is allowed to elapse between irradiation and mating. This finding is already used as a basis for human genetic counseling. If an individual is exposed to a large dose of radiation accidentally or as a consequence of medical procedures, a planned conception should be delayed to minimize the genetic consequences of the radiation. 5. The megamouse project indicates that the doubling dose is about 1 Gy. Estimates of the risk of radiation-induced hereditary ­effects in humans can be made with knowledge of two ­factors—the baseline rate of spontaneous mutations, which is known, and the doubling dose, which is not known for humans but has been estimated from mouse experiments to be about 1 Gy.129 Estimates of the risk of hereditary effects resulting from radiation exposure, published by the International Commission on Radiological Protection (ICRP), are 0.2% per sievert (Sv) for the whole population (including children) and 0.1%/Sv for the working population (those 18 to 65 years old). However, these risks can be considered in light of detriment, a term introduced to account for years of life lost or years of life spent impaired. The detriment from radiation-induced hereditary effects from radiation is far smaller than the detriment from  radiation-induced carcinogenesis or from the effects of  radiation on an embryo or fetus. Some direct information on the hereditary effects of radiation on humans is available from the detailed studies of the survivors of the atomic bomb attacks on Hiroshima and Nagasaki and their children. Among the many factors documented in the children are congenital defects, physical development, the sex of child, ­cytogenetic abnormalities, and electrophoretic variations in blood proteins.130 No statistically significant excess in hereditary effects has been found in these children, but in statistical terms, the numbers are relatively small compared with the number of animals used in the megamouse project.

Carcinogenesis In sharp contradistinction to the situation for the hereditary effects of radiation, an abundance of information exists concerning radiation carcinogenesis directly from humans. The human experience can be summarized as follows: 1. Early workers, largely physicists and engineers working around accelerators, showed an increased incidence of skin cancer and other forms of tumors. 2. Miners in Saxony in the late 1800s and early 1900s and, more recently, uranium workers in the Central Colorado Plateau showed an increased incidence of lung cancer as a result of inhaling radon in poorly ventilated mines.

40

PART 1  Principles

3. Workers who painted luminous dials on clocks and watches and who ingested radium by using their tongues to lick the tip of their brushes into a point subsequently developed bone tumors. 4. Patients in whom the contrast material Thorotrast (which contains radioactive thorium) was used subsequently developed liver tumors. 5. The survivors of Hiroshima and Nagasaki have shown a whole spectrum of malignancies, from leukemia to solid tumors, particularly cancer of the gastrointestinal tract, breast, thyroid, and lung. 6. Children given therapeutic doses of radiation for infected tonsillar or nasopharyngeal lymphoid tissue or what was perceived to be an enlarged thymus subsequently developed benign and malignant thyroid tumors and leukemias. 7. Children epilated by x-rays as part of the treatment for tinea capitis subsequently developed thyroid and skin tumors. 8. Patients with ankylosing spondylitis who received x-rays for the relief of pain subsequently had an increased incidence of leukemia. 9. Women who underwent fluoroscopy many times during the management of tuberculosis had an increased incidence of breast cancer. The best quantitative data come from the Japanese atomic bomb survivors. For solid tumors, risk seems to be a linear function of dose between about 0.2 and 2.5 Gy. For leukemia, a linear-quadratic relationship fits the data more closely. For protection purposes, it is assumed that these risks observed at high doses can be extrapolated linearly to low doses and that no dose threshold exists below which there is no risk. The Latent Period. In leukemia, the latency period (i.e., time between the radiation exposure and the appearance of the disease) is relatively short, with a peak excess incidence occurring by 7 to 10 years after irradiation and with the radiation-induced incidence disappearing by about 15 years after irradiation. In the case of solid tumors, the latent period may be considerably longer; for example, the excess incidence of solid tumors is still evident in the Japanese survivors more than 60 years after the atomic bomb attacks. Part of the latency may have resulted from a form of induction period before the initially transformed cell or cells start to divide to form a tumor or, alternatively, to a period before the tumor assumes the malignant characteristics of growth and spread. For example, at short intervals after irradiation, the thyroid gland contains tumors that are of a benign histologic character, and malignant tumors become detectable only at later stages. It is widely believed from animal studies that tumorigenesis consists of three separate stages: initiation, promotion, and progression. Some chemicals can only initiate, whereas others can only promote. Radiation seems to be capable of both. The perception of the latency period has changed. ­Radiation-induced tumors appear not at a fixed time ­after exposure, but rather at the age at which spontaneous  tumors of the same type are most prevalent. For example, ­radiation-induced breast cancers in women appear in middle age, regardless of whether the radiation exposure occurred when the women were teenagers or in their 30s. This fact suggests that radiation may initiate the process at a young age but that completion requires additional steps, some of which may depend on hormones.

Absolute versus Relative Risk. The preceding information leads to the discussion of relative versus absolute risk models in radiation carcinogenesis. Although they can seem unduly complex, use of one model over another can produce substantial differences in risk estimates of carcinogenesis, and some basic understanding of their principles is warranted. The absolute risk model assumes that radiation produces a discrete crop of malignancies that are unrelated to the natural or spontaneous incidence. This seems to be the case for leukemia; radiation-induced leukemia in Japanese survivors of atomic bomb blasts was observed for some period after the exposure, but the incidence subsequently returned to that characteristic of the control population. The relative risk model assumes that radiation increases the spontaneous incidence by some fixed factor. According to this model, the radiation-induced cancer incidence will increase with the age of the population as the spontaneous incidence increases. This seems to be the case for the Japanese atomic bomb survivors as the population ages and the data mature; a relative risk model would predict a large number of tumors occurring late in the life of the irradiated population. The currently favored model is a relative risk model with allowance for age at exposure and time since exposure. Another point of interest is that risk for three of the four common types of leukemia seems to be increased by radiation, but the risk of chronic lymphocytic leukemia has never been reported to increase from exposure to radiation. The ICRP, working with data published by the Committee on the Biological Effects of Ionizing Radiation (BEIR V)131 and the United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR 88),132 has generated the following risk estimates for ­ excess cancer ­ mortality from radiation exposure. For a working population (men and women between 18 and 65 years old), the lifetime risk of death from cancer is 0.08/Sv for high-dose or high-dose-rate radiation and 0.04/Sv for low-dose or low-dose-rate radiation. Comparable values for the general population are slightly higher because of the higher susceptibility of the young; those values are 0.1/Sv for high-dose, high-dose-rate radiation and 0.5/Sv  for low-dose, low-dose-rate radiation. Acknowledging the assumptions inherent in the relative-risk projection model, the ICRP further estimates that, on average, 13 to 15 years of life are lost for each radiation-induced cancer, with death occurring, on average, at 68 to 70 years of age.133 As the data from Japan have matured and more ­detailed information has become available, it has ­ become evident that the risk of radiation-induced cancer varies considerably with age at the time of exposure. In most cases, ­individuals exposed at an early age are much more ­ susceptible than are those exposed at older ages. This ­ difference seems to be the most dramatic for female breast cancer; children exposed before 15 years of age are most susceptible, whereas women 50 years old or older at ­ exposure show little or no excess risk. There are ­ exceptions to this general rule. Susceptibility to radiation-induced leukemia is relatively constant throughout life, and susceptibility to respiratory cancers increases in middle age. ­However, the overall risk drops drastically with age; children and young adults are much more susceptible to radiation-induced cancer than are middle-aged and older persons (Fig. 1-34).131,133

41

1  Physical and Biologic Basis of Radiation Therapy PERCENTAGE INCREASE IN RELATIVE RISK FOR RT vs. SURGERY %

Risk per unit dose (percent per Sv)

16 14 All years 12

All solid tumors

5+ years

10

10+ years

8 ICRP 60

0

BEIR V

6

10

20

30

40

50

60

SECOND CANCERS AFTER PROSTATE RT

4 % contribution to total number of radiationinduced second cancers

2 0 0

10

20

30

40

50

60

70

80

Age of acute exposure

FIGURE 1-34.  The attributable lifetime risk from a single, small dose of radiation at various ages at the time of exposure is described by two curves. Notice the dramatic decrease in radiosensitivity with age. The higher risk for the younger age groups is not expressed until late in life. These estimates are based on a relative risk model and on a dose and dose-rate effectiveness factor (DDREF) of 2. BEIR, Biological Effects of Ionizing Radiation study; ICRP, International Commission on Radiological Protection study. (Modified from ICRP. Recommendations. Annals of the ICRP publication 60. Oxford, England: Pergamon Press, 1990.)

Second Malignancies in Patients Given Radiation Therapy The risk of second malignancies after radiation therapy is a controversial topic for several reasons, among them that patients undergoing radiation therapy are often at high risk for a second cancer because of lifestyle factors, and these factors may be more dominant than the radiation risk. Although many single-institution studies involving radiation therapy for a variety of tumors have concluded that radiation is not associated with an increase in second malignancies, many of the studies were much too small to detect any increased incidence of second malignancies induced by the treatment. Larger studies have consistently indicated that radiation therapy is associated with a small but statistically significant enhancement in the risk of second malignancies, particularly among long-term survivors. To credibly answer the question of whether radiation therapy is associated with a risk of second malignancies, a study must have (1) a sufficiently large number of patients, (2) a suitable control group (i.e., patients with the same cancer treated by some means other than radiation), and (3) a sufficiently long follow-up period for radiationinduced solid tumors to manifest. Studies that satisfy these criteria describing patients treated for prostate cancer, cervical cancer, and Hodgkin disease are discussed briefly in the following sections. Second Malignancies after Treatment for Prostate Cancer. In the largest study of its kind, Brenner and colleagues134 used data from the National Cancer Institute’s Surveillance Epidemiology, and End Results (SEER) Program to evaluate outcomes among 51,584 men with ­prostate cancer treated with radiation and 70,539 treated

Sarcoma (in field) 6% Sarcoma (out of field) 2%

Rectum 12%

Bladder 37%

Lung 34%

Colon 9%

FIGURE 1-35.  Top, Percentage increase in the relative risk for all solid tumors (except prostate cancer) for individuals who underwent radiation therapy (RT) for prostate cancer relative to the risk for individuals who underwent surgery for prostate cancer. Bottom, Distribution of radiation-induced second cancer at 5+ years after the radiation therapy. (Illustration courtesy of David Brenner, MD, New York, NY; data from Brenner DJ, Curtis RE, Hall EJ, Ron E. Second cancers after radiotherapy for prostate cancer. Cancer 2000;88:398-406.) Modified from Hall EJ, Giaccia AJ. Radiation carcinogenesis. In: Radiobiology for the Radiologist. 6th ed. Philadelphia: Lippincott Williams & Wilkins, 2006, pp 135-155.

with surgery. No evidence was found of a difference in the risk of leukemia for radiation-therapy versus surgery patients, but the risk of a second solid tumor at any time after diagnosis was significantly greater after radiation therapy than after surgery (Fig. 1-35).134 This increased relative risk became greater with time, reaching 34% after 10 years or more. The greatest increases in relative risk were for tumors of the bladder (77%) and rectum (105%) at 10 years or more after diagnosis. The increased risk of developing sarcoma in the heavily irradiated tissues within the field amounted to 145% at 5 years or more, compared with surgical patients. The increase in relative risk for carcinoma of the lung that was exposed to relatively low doses (about 0.5 Gy) was close to that for carcinoma of the bladder, rectum, or colon, all of which were subject to much higher doses (typically more than 5 Gy). This pattern may reflect the fact that carcinoma originating in actively dividing cells or cells under hormonal control can be efficiently induced by relatively low doses of radiation, as evidenced by the atomic bomb survivors, but that the cancer risk at high doses decreases because of cell death. In contrast to this pattern for radiation-induced carcinomas, radiation-induced sarcomas usually appeared only in heavily irradiated sites (i.e., close to the treatment volume), where large radiation doses were needed to produce sufficient tissue damage to stimulate cellular renewal in otherwise dormant cells. Second Malignancies after Treatment for Cervical Cancer. In a landmark study, Boice and colleagues135 studied the risk of second malignancies as a consequence

42

PART 1  Principles

of ­radiation treatment for carcinoma of the uterine cervix. This huge international study, involving 42 investigators from 38 institutions, had a sample size of 150,000 patients and had an ideal control group for comparison, because cervical cancer is equally well treated by radiation or surgery. Some of the pertinent findings from this large study are summarized here. Very high doses (on the order of several hundred Gy) were found to increase the risk of developing cancer of the bladder, vagina, bone, uterine corpus, and cecum, as well as non-Hodgkin lymphoma. Risk ratios ranged from a high of 4.0 for bladder cancer to a low of 1.3 for bone cancer. A steep dose-response curve was observed for the risk of developing subsequent genital cancer, with a fivefold excess at doses of more than  150 Gy. Doses of several Gy were found to increase the risk of stomach cancer and leukemia. Somewhat surprisingly, radiation therapy for cervical cancer did not increase the overall risk of subsequent cancer of the small intestine, colon, ovary, vulva, connective tissue, or breast, nor did it increase the risk of Hodgkin disease, multiple myeloma, or chronic lymphocytic leukemia. The overall conclusions of this study were that radiation therapy, unlike surgery, was certainly associated with a risk of excess cancer and that the risks were highest among long-term survivors and concentrated among women who were irradiated at relatively young ages. Second Malignancies after Treatment for Hodgkin Lymphoma. In a 1996 report of the Late Effects Study Group, which followed 1380 children with Hodgkin ­disease, 17 of 483 girls in whom Hodgkin disease was diagnosed before the age of 16 years subsequently developed breast cancer, with radiation therapy implicated in most cases.136 The ratio of observed to expected cases in that study was 75.3. Another study involving 1641 patients treated for Hodgkin disease as children in five Nordic countries reported 16 cases of breast cancer, for a relative risk 17 times that of the general population.137 The largest study of this kind evaluated 3869 women in population-based registries participating in the SEER Program.138 All of these women had been given radiation therapy as initial treatment for Hodgkin disease. Breast cancer developed in a total of 55 patients, which represents a ratio of observed to expected cases of 2.24. However, the risk of subsequent breast cancer in women treated before the age of 16 years was 61%, with most tumors appearing 10 or more years after treatment. This finding agrees with those from previous studies showing the prepubertal and pubertal female breast to be very radiosensitive. The risk of breast cancer decreased as the age at therapy increased and was only slightly elevated in women who were 30 years old or older when they were treated. In one study,139 27 of 202 patients with Hodgkin disease treated with radiation developed a second cancer. Most of the second tumors appeared within or near the treatment field, most commonly in the lung and breast; however, the spectrum was large, and the risk was greatest among patients irradiated at young ages. In summary, studies of large numbers of patients clearly indicate that radiation therapy induces an excess of second cancers. In the young, the breast is especially sensitive to the carcinogenic effects of radiation. The latency period for the development of excess cancers can range from 10 years to decades after the radiation treatment.

REFERENCES   1. Bishop JM. Cellular oncogene retroviruses. Annu Rev Biochem 1983;52:301-354.   2. Kundson AG. Mutation and cancer: statistical study of ­retinoblastoma. Proc Natl Acad Sci U S A 1971;68:820-823.   3. Knudson A. Anti-oncogenes and human cancer. Proc Natl Acad Sci U S A 1993;90:10914-10921.   4. Fearon ER, Vogelstein B. A genetic model for colorectal ­tumorigenesis. Cell 1990;61:759-767.   5. Nagasawa H, Little JB. Induction of sister chromatid ­exchanges by extremely low doses of alpha-particles. Cancer Res 1992;52:6394-6396.   6. Zhou H, Randers-Pehrson G, Waldren CA, et al. Induction of a bystander mutagenic effect of alpha particles in mammalian cells. Proc Natl Acad Sci U S A 2000;97:2099-2104.   7. Zhou H, Suzuki M, Randers-Pehrson G, et al. Radiation risk to low fluences of alpha particles may be greater than we thought. Proc Natl Acad Sci U S A 2001;98:14410-14415.   8. Mothersill C, Seymour C. Radiation-induced bystander ­effects: past history and future directions. Radiat Res 2001;155:  759-767.   9. Ward JF. DNA lesions produced by ionizing radiation: locally multiplying damaged sites. In Wallace SS, Painter RB, (eds): I­onizing Radiation Damage to DNA: Molecular Aspects. New York: Wiley-Liss, 1990. 10. Puck TT, Markus PI. Action of x-rays on mammalian cells. J Exp Med 1956;103:656-666. 11. Withers HR. The dose-survival relationship for irradiation of epithelial cells of mouse skin. Br J Radiol 1967;40:187-194. 12. Withers HR. Recovery and repopulation in vivo by mouse skin epithelium cells during fractionated irradiation. Radiat Res 1967;32:227-239. 13. Withers HR, Elkind MM. Microcolony survival assay for cells of mouse intestinal mucosa exposed to radiation. Int J Radiat Biol 1970;17:261-267. 14. McCullough EA, Till JE. The sensitivity of cells from normal mouse bone marrow to gamma radiation in vitro and in vivo. Radiat Res 1962;16:822-832. 15. Douglas BG, Fowler JF: The effect of multiple small doses of x-rays on skin reactions in the mouse and a basic interpretation. Radiat Res 1976;66:401-426. 16. Suit H, Wette R. Radiation dose fractionation and tumor control probability. Radiat Res 1966;29:267-281. 17. Hewitt HB, Wilson CW. A survival curve for mammalian ­leukaemia cells irradiated in vivo. Br J Cancer 1959;13:69-75. 18. Hill RP, Bush RS. A lung colony assay to determine the radiosensitivity of the cells of a solid tumor. Int J Radiat Biol 1969;15:435-444. 19. Howard A, Pelc SR. The synthesis of deoxyribose nucleic acid and its relationship to the chromosome damage induced by  x-rays. Heredity 1953;6;(Suppl):261-273. 20. Brown JM. The effect of acute x-irradiation on the cell proliferation kinetics of induced carcinomas and their normal counterpart. Radiat Res 1970;43:627-653. 21. Steel GG. Cell loss as a factor in the growth rate of human ­tumors. Eur J Cancer 1967;3:381-387. 22. Steel GG. Cell loss from experimental tumors. Cell Tissue Kinet 1968;1:193-207. 23. Steel GG. The kinetics of cell proliferation in tumors. In Bond V (ed): Proceedings of the Carmel Conference on Time and Dose Relationships in Radiation Biology as Applied to Radiotherapy. BNL Report 50203 (C-57). Upton, NY: Brookhaven National Laboratory, 1969, pp 130–149. 24. Begg AC, McNally NJ, Shrieve D, et al. A method to measure the duration of DNA synthesis and the potential doubling time from a single sample. Cytometry 1985;6:620-625. 25. Begg AC, Moonen I, Hofland I, et al. Human tumor cell kinetics using a monoclonal antibody against iododeoxyuridine: intratumoral sampling variations. Radiother Oncol 1988;11:337-347. 26. Gray JW, Dolbeare F, Pallavicini MG, et al. Cell cycle analysis using flow cytometry. Int J Radiat Biol 1986;49:237-255.

1  Physical and Biologic Basis of Radiation Therapy 27. Regaud C, Nogier T. Sterilization roentgenienne totale et definitive, sans radiodermite, des testicules du Belier adulte: corditions de sa realisation. Compt Rend Soc Biol 1911;70:202-203. 28. Brenner DJ, Hall E. Conditions for the equivalence of continuous to pulsed low dose rate brachytherapy. Int J Radiat Oncol Biol Phys 1991;20:181-190. 29. Thames HD, Ang KK, Stewart FA, et al. Does incomplete repair explain the apparent failure of the basic LQ model to predict spinal cord and kidney responses to low doses per fraction. Int J Radiat Biol 1988;554:13-19. 30. Withers HR, Peters LJ, Kogelnik HC. The pathobiology of late effects of irradiation. In Meyn RE, Withers HR, (eds): Radiation Biology in Cancer Research. New York: Raven Press, 1980, pp 439-448. 31. Fowler JF. The second Klaas Breur memorial lecture: La Ronde—radiation sciences and medical radiology. Radiother Oncol 1983;1:1-22. 32. Ellis F. Dose time and fractionation: a clinical hypothesis. Clin Radiol 1969;20:1-7. 33. Arriagada R, LeChevalier T, Quoix E, et al. Effect of chemotherapy on locally advanced non–small cell lung carcinoma: a randomized study of 3,553 patients. Int J Radiat Oncol Biol Phys 1991;20:1183-1190. 34. Cox JD, Stoffel TJ. The significance of needle biopsy after irradiation for stage C adenocarcinoma of the prostate. Cancer 1977;40:156-160. 35. Dugan TC, Shipley WU, Young RH, et al. Biopsy after external beam radiation therapy for adenocarcinoma of the prostate: correlation with original histological grade and current prostate specific antigen levels. J Urol 1991;146:1313-1316. 36. Cox JD. Large dose fractionation (hypofractionation). Cancer 1985;55;(Suppl):2105-2111. 37. Flickinger JC, Kondziolka D, Lunsford LD, et al. A multi­institutional experience with stereotactic radiosurgery for  solitary brain metastasis. Int J Radiat Oncol Biol Phys 1994;28:797-802. 38. Onishi H, Arak, T, Shirato H, et al. Stereotactic hypofractionated high-dose irradiation for stage I nonsmall cell lung carcinoma: clinical outcomes in 245 subjects in a Japanese ­multiinstitutional study. Cancer 2004;101:1623-1631. 39. Cox JD, Pajak TF, Marcial VA, et al. Interfractional interval is a major determinant of late effects, with ­hyperfractionated radiation therapy of carcinomas of upper respiratory and digestive tracts: results from Radiation Therapy Oncology Group protocol 8313. Int J Radiat Oncol Biol Phys 1991;20:1191-1195. 40. Horiot J-C., Le Fur R, N’Guyen T, et al. Hyperfractionation versus conventional fractionation in oropharyngeal carcinoma: final analysis of a randomized trial of the EORTC cooperative group of radiotherapy. Radiother Oncol 1992;25:231-241. 41. Horiot J-C., Bontemps P, van den Bogaert W, et al. Accelerated fractionation (AF) compared to conventional fractionation (CV) improves loco-regional control in the radiotherapy of advanced head and neck cancers: results of the EORTC 22851 randomised trial. Radiother Oncol 1997;44:111-121. 42. Dische S, Saunders M, Barrett A, et al. A randomized multicentre trial of CHART versus conventional radiotherapy in head and neck cancer. Radiother Oncol 1997;44:123-136. 43. Fu KK, Pajak TF, Trotti A, et al. A Radiation Therapy ­Oncology Group (RTOG) phase III randomized study to compare ­hyperfractionation and two variants of accelerated ­fractionation to standard fractionation radiotherapy for head and neck ­squamous cell carcinomas: first report of RTOG 9003. Int J Radiat Oncol Biol Phys 2000;48:7-16. 44. Fowler JF, Harari P. Confirmation of improved local-regional control with altered fractionation in head and neck cancer. Int J Radiat Oncol Biol Phys 2000;48:3-6. 45. Withers HR, Peters LJ. Transmutability of dose and time. ­Commentary on the first report of RTOG 9003. Int J Radiat Oncol Biol Phys 2000;48:1-2. 45a. Kaanders JH, Pop LA, Maores HA, et al. ARCON: experience with 215 patients with advanced head-and-neck cancer. Int J Radiat Oncol Biol Phys 2002;52:769-778.

43

46. Overgaard J, Sand Hansen H, Overgaard M, et al. Importance of overall treatment time for the outcome of radiotherapy in head and neck carcinoma. Experience from the Danish Head and Neck Cancer Study, 6th International Meeting on Progress in Radio-Oncology, Salzburg, Austria, 13-17 May 1998. Bologna, Italy: Monduzzi Editore, 1998, pp 743-752. 47. Hermens AF, Barendsen GW. Changes of cell proliferation characteristics in a rat rhabdomyosarcoma before and after ­irradiation. Eur J Cancer 1969;5:173-189. 48. Withers HR, Peters LJ, Thames HD, et al. Hyperfractionation. Int J Radiat Oncol Biol Phys 1982;8:1807-1809. 49. Withers HR, Mason K, Reid BO, et al. Response of mouse ­intestine to neutrons and gamma rays in relation to dose fractionation and division cycle. Cancer 1974;34:3947. 50. Withers HR, Taylor MMG, Maciejewski B, et al. The hazard of accelerated tumor clonogen repopulation during radiotherapy. Acta Oncol 1988;27:131-146. 51. Chapman JD, Peters LJ, Withers HR. Prediction of Tumor ­Treatment Response. New York: Pergamon Press, 1989. 52. Geara FB, Peters LJ, Ang KK, et al. Prospective comparison of in vitro normal cell radiosensitivity and normal tissue reactions in radiotherapy patients. Int J Radiat Oncol Biol Phys 1993;27:1171-1179. 53. West CM. Intrinsic radiosensitivity as a predictor of patient ­response to radiotherapy. Br J Radiol 1995;68:827-837. 54. West CM, Davidson SE, Burt PA, et al. The intrinsic radiosensitivity of cervical carcinoma: correlations with clinical data. Int J Radiat Oncol Biol Phys 1995;31:841-846. 55. West CML, Davidson SE, Hendry JH, et al. Prediction of cervical carcinoma response to radiotherapy. Lancet 1991;338:818. 56. West CM, Davidson SE, Roberts SA, et al. Intrinsic radiosensitivity and prediction of patient response to radiotherapy for carcinoma of the cervix. Br J Cancer 1993;68:819-823. 57. Parliament MB, Chapman JD, Urtasun RC, et al. Noninvasive assessment of human tumour hypoxia with 123I-iodoazomycin arabinoside: preliminary report of a clinical study. Br J Cancer 1992;65:90-95. 58. Höckel M, Knoop C, Schlenger K, et al. Intratumoral pO2 ­predicts survival in advanced cancer of the uterine cervix. ­Radiother Oncol 1993;26:45-50. 59. Höckel M, Schlenger K, Aral B, et al. Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix. Cancer Res 1996;56:4509-4515. 60. Begg AC, Hofland I, Moonen L, et al. The predictive value of cell kinetic measurements in a European trial of accelerated ­fractionation in advanced head and neck tumors: an interim ­report. Int J Radiat Oncol Biol Phys 1990;19:1449-1453. 61. Begg AC, Haustermans K, Hart AAM, et al. The value of ­pretreatment cell kinetic parameters as predictors for radiotherapy outcome in head and neck cancer: a multicenter analysis. Radiother Oncol 1999;50:13-23. 62. Thomlinson RH, Gray LH. The histological structure of some ­ human lung cancers and the possible implications for ­radiotherapy. Br J Cancer 1955;9:539-549. 63. Brown JM. Evidence for acutely hypoxic cells in mouse ­tumours, and a possible mechanism of reoxygenation. Br J ­ Radiol 1979;52:650-656. 64. Sobhanifar S, Aquino-Parsons C, Stanbridge EJ, et al. Reduced expression of hypoxia-inducible factor-1alpha in perinecrotic regions of solid tumors. Cancer Res 2005;65:7259-7266. 65. Kwon SJ, Lee YJ. Effect of low glutamine/glucose on hypoxiainduced elevation of hypoxia-inducible factor-1alpha in ­human pancreatic cancer MiaPaCa-2 and human prostatic cancer  DU-145 cells. Clin Cancer Res 2005;11:4694-4700. 66. Brizel DM, Scully SP, Harrelson JM, et al. Tumor oxygenation predicts for the likelihood of distant metastases in human soft tissue sarcoma. Cancer Res 1996;56:941-943. 67. Giaccia AJ, Brown JM, Wouters B, et al. Cancer therapy and tumor physiology. Science 1998;279:12-13. 68. Adams GE. Chemical radiosensitization of hypoxic cells. Br Med Bull 1973;29:48-53.

44

PART 1  Principles

69. Adams GE. Hypoxic cell sensitizers for radiotherapy. In Becker FF (ed): Cancer, vol 6. New York: Plenum Press, 1977,  pp 181-223. 70. Adams GE, Bremner J, Stratford IJ, et al. Nitroheterocyclic compounds as radiosensitizers and bioreductive drugs. ­ Radiother Oncol 1991;20:85-91. 71. Brown JM, Siim BG. Hypoxia-specific cytotoxins in cancer ­therapy. Semin Radiat Oncol 1996;6:22-36. 72. Brown JM, Lemmon MJ. Tumor hypoxia can be exploited to preferentially sensitize tumors to fractionated irradiation. Int J Radiat Oncol Biol Phys 1991;20:457-461. 73. von Pawel J, von Roemeling R, Gatzemeier U, et al. Tirapazamine plus cisplatin in advanced non–small cell lung cancer: a report of the international CATAPULT I study group. Cisplatin and ­ Tirapazamine in Subjects with Advanced Previously Untreated Non–Small Cell Lung Tumors. J Clin Oncol 2000;18:  1351-1359. 74. Rischin D, Peters L, Hicks R, et al. Phase I trial of concurrent tirapazamine, cisplatin, and radiotherapy in patients with ­advanced head and neck cancer. J Clin Oncol 2001;19:535-542. 75. Rischin D, Peters L, Fisher R, et al. Tirapazamine, cisplatin, and radiation versus fluorouracil, cisplatin, and radiation in patients with locally advanced head and neck cancer: a randomized phase II trial of the Trans-Tasman Radiation Oncology Group (TROG 98.02). J Clin Oncol 2005;23:79-87. 76. Overgaard J, Horsman MR. Modification of hypoxia-induced radioresistance in tumours by the use of oxygen and sensitizers. Semin Radiat Oncol 1996;6:10-21. 77. Patt HM, Tyree EB, Straube RL, et al. Cysteine protection against x-irradiation. Science 1949;110:213-214. 78. Bacq ZM, Herve A, Lecompte J, et al. Protection contre le rayonnement x par la β-mercato-éthylamine. Arch Intern Physiol 1951;S9:442-447. 79. Yuhas JM, Spellman JM, Culo F. The role of WR2721 in ­radiotherapy and/or chemotherapy. In Brady L (ed): Radiation Sensitizers. New York: Masson Publishing, 1980. 80. Kligerman MM, Liu T, Liu Y, et al. Interim analysis of a ­randomized trial of radiation therapy of rectal cancer with/without WR-2721. Int J Radiat Oncol Biol Phys 1992;22:799-802. 81. Brizel D, Sander R, Wannenmacher M, et al. Randomized phase III trial of radiation ± amifostine in patients with head and neck cancer [abstract]. Proc Am Soc Clin Oncol 1998;17:1487. 82. Komaki R, Lee JS, Milas L, et al. Effects of amifostine on acute toxicity from concurrent chemotherapy and radiotherapy for inoperable non–small cell lung cancer: report of a randomized comparative trial. Int J Radiat Oncol Biol Phys 2004;58:  1369-1377. 83. Barendsen GW. Radiobiological dose-effect relations for radiation characterized by a wide spectrum of LET: implications for their application in radiotherapy. In Proceedings of the Conference of ­Particle Accelerators in Radiation Therapy, Los Alamos Science Laboratory, Los Alamos, NM, October 2-5, 1972. LA5180-C, 120-125. Oak Ridge, TN: U.S. Atomic Energy Commission, Technical Information Center, 1972. 84. Catterall M, Vonberg DD. Treatment of advanced tumors of head and neck with fast neutrons. Br Med J 1974;3:137-143. 85. Catterall M, Sutherland I, Bewley DK. First results of a ­clinical trial of fast neutrons compared with x or gamma rays in ­treatment of advanced tumors of the head and neck. Br Med J 1975;2:653-656. 86. Catterall M. Results of neutron therapy: differences, correlations, and improvements. Int J Radiat Oncol Biol Phys 1982;8:  2141-2144. 87. Batterman JJ. Clinical Application of Fast Neutrons: The ­Amsterdam Experience. Amsterdam: Rodipi, 1981. 88. Laramore GE, Krall JM, Thomas FJ, et al. Fast neutron radiotherapy for locally advanced prostate cancer: results of an RTOG randomized study. Int J Radiat Oncol Biol Phys 1985;11:1621-1627. 89. Weber DC, Trofimov AV, Delaney TF, et al. A treatment planning comparison of intensity modulated photon and proton therapy for paraspinal sarcomas. Int J Radiat Oncol Biol Phys 2004;58:1596-1606.

90. Sisterson J. Proposed new facilities for proton and ion beam therapy, July 2005. Particles Newslett 2005;36:1-11. Available at http://ptcog.web.psi.ch/Archive/ptles36.pdf (accessed ­January 2009). 91. Kraft G. Tumor therapy with heavy charged particles. Progress in Particle and Nuclear Physics 2000;45;(Suppl 2):S473-S544. 92. Coley WB. The treatment of malignant tumors by repeated ­inoculation of erysipelas: with a report of ten original cases. Am J Med Sci 1893;105:487-511. 93. Overgaard J, Gonzalez Gonzalez D, Hulschof MC, et al. Hyperthermia as an adjuvant to radiation therapy of recurrent or metastatic malignant melanoma. A multicentre randomized trial by the European Society of Hyperthermic Oncology. Int J Hyperthermia 1996;12:3-20. 94. Vernon CC, Hand JW, Field SB, et al. Radiotherapy with or without hyperthermia in the treatment of superficial localized breast cancer: results from five randomized controlled trials. International Collaborative Hyperthermia Group. Int J Radiat Oncol Biol Phys 1996;35:731-744. 95. Kong G, Anyarambhatla G, Petros WP, et al. Efficacy of liposomes and hyperthermia in a human tumor xenograft model: importance of triggered drug release. Cancer Res 2000;60:6950-6957. 96. Kong G, Braun RD, Dewhirst MW. Characterization of the effect of hyperthermia on nanoparticle extravasation from tumor vasculature. Cancer Res 2001;61:3027-3032. 97. Steel GG. Terminology in the description of drug-radiation interactions. Int J Radiat Oncol Biol Phys 1979;5:1145-1150. 98. Tubiana M. Les associations radiotherapie-chimeotherapie. J Eur Radiother 1980;1:107-114. 99. Schaake-Koning C, vanden Bogaert W, Dalesio O, et al. Effects of concomitant cisplatin and radiotherapy on inoperable nonsmall cell lung cancer. N Engl J Med 1992;326:524-530. 100. Krook JE, Moertel C, Gunderson L, et al. Effective surgical adjuvant therapy for high-risk rectal carcinoma. N Engl J Med 1991;324:709-715. 101. Le Chevalier T, Arriagada R, Tarayre M, et al. Significant effect of adjuvant chemotherapy on survival in locally advanced ­nonsmall cell lung carcinoma. J Natl Cancer Inst 1992;84:58. 102. Herskovic A, Martz K, Al-Sarraf M, et al. Combined ­chemotherapy and radiotherapy compared with radiotherapy alone in patients with cancer of the esophagus. N Engl J Med 1992;326:1593-1598. 103. Milas L, Mason K, Hunter N, et al. In vivo enhancement of ­tumor radioresponse by C225 antiepidermal growth factor ­receptor antibody. Clin Cancer Res 2000;6:701-708. 104. Bonner JA, Giralt J, Harari PM, et al. Cetuximab prolongs ­survival in patients with locoregionally advanced squamous cell carcinoma of head and neck: a phase III study of high dose radiation therapy with or without cetuximab [abstract]. J Clin Oncol 2004;22;(14S):5507. 105. Hall SJ. Chen S-H, Woo SLC. Gene therapy 97: the promise and reality of cancer gene therapy. Am J Hum Genet 1997;61:785-789. 106. Kim JH, Kim SH, Brown SL, et al. Selective enhancement by an antiviral agent of the radiation-induced cell killing of human glioma cells transduced with HSV-tk gene. Cancer Res 1994;54:6053-6056. 107. Khil MS, Kim JH, Mullen CA, et al. Radiosensitization by 5-fluorocytosine of human colorectal carcinoma cells in culture transduced with cytosine deaminase gene. Clin Cancer Res 1996;2:53-57. 108. Freytag SO, Stricker H, Pegg J, et al. Phase I study of ­replicationcompetent adenovirus-mediated double-suicide gene therapy in combination with conventional-dose three-dimensional conformal radiation therapy for the treatment of newly diagnosed, intermediate- to high-risk prostate cancer. Cancer Res 2003;63:7497-7506. 109. Freytag SO, Stricker H, Peabody J, et al. Five-year follow-up of trial of replication-competent adenovirus-mediated suicide gene therapy for treatment of prostate cancer. Mol Ther 2007;15:636-642. 110. Ganly I, Kirn D, Eckhardt G, et al. A phase I study of Onyx-015, an E1B attenuated adenovirus, administered intratumorally to patients with recurrent head and neck cancer. Clin Cancer Res 2000;6:798-806.

1  Physical and Biologic Basis of Radiation Therapy 111. Nemunaitis J, Khuri F, Ganly I, et al. Phase II trial of intratumoral administration of ONYX-015, a replication-selective ­adenovirus, in patients with refractory head and neck cancer.  J Clin Oncol 2001;19:289-298. 112. Vasey PA, Shulman LN, Campos S, et al. Phase I trial of intraperitoneal injection of the E1B-55-kd-gene-deleted adenovirus ONYX-015 (dl1520) given on days 1 through 5 every 3 weeks in patients with recurrent/refractory epithelial ovarian cancer.  J Clin Oncol 2002;20:1562-1569. 113. Gallardo D, Drazan KE, McBride WH. Adenovirus-based transfer of wild-type p53 gene increases ovarian tumor radiosensitivity. Cancer Res 1996;56:4891-4893. 114. Spitz FR, Nguyen D, Skibber JM, et al. Adenoviral-mediated wild-type p53 gene expression sensitizes colorectal cancer cells to ionizing radiation. Clin Cancer Res 1996;2:1665-1671. 115. Swisher SG, Roth JA, Nemunaitis J, et al. Adenovirus-mediated p53 gene transfer in advanced non–small cell lung cancer. J Natl Cancer Inst 1999;91:763-771. 116. Clayman GL, el-Naggar AK, Lippman SM, et al. Adenovirusmediated p53 gene transfer in patients with advanced recurrent head and neck squamous cell carcinoma. J Clin Oncol 1998;16:2221-2232. 117. Weill D, Mack M, Roth J, et al. Adenoviral-mediated p53 gene transfer to non–small cell lung cancer through endobronchial injection. Chest 2000;118:966-970. 118. Nemunaitis J, Swisher SG, Timmons T, et al. Adenovirus­mediated p53 gene transfer in sequence with cisplatin to ­tumors of patients with non–small cell lung cancer. J Clin Oncol 2000;18:609-622. 119. Modesitt SC, Ramirez P, Zu Z, et al. In vitro and in vivo ­adenovirus-mediated p53 and p16 tumor suppressor therapy in ­ovarian cancer. Clin Cancer Res 2001;7:1765-1772. 120. Lang FF, Bruner JM, Fuller GN, et al. Phase I trial of adenovirusmediated p53 gene therapy for recurrent glioma: biological and clinical results. J Clin Oncol 2003;21:2508-2518. 121. Pagliaro LC, Keyhani A, Williams D, et al. Repeated intravesical instillations of an adenoviral vector in patients with locally advanced bladder cancer: a phase I study of p53 gene therapy. J Clin Oncol 2003;21:2247-2253. 122. Clayman GL, Frank DK, Bruso PA, et al. Adenovirus-mediated wild-type p53 gene transfer as a surgical adjuvant in advanced head and neck cancers. Clin Cancer Res 1999;5:1715-1722. 123. Weichselbaum RR, Hallahan DE, Beckett MA, et al. Gene therapy targeted by radiation preferentially radiosensitizes tumor cells. Cancer Res 1994;54:4266-4269. 124. Hallahan DE, Mauceri HJ, Seung LP, et al. Spatial and ­temporal control of gene therapy using ionizing radiation. Nat Med 1995;1:786-791. 125. Staba MJ, Mauceri HJ, Kufe DW, et al. Adenoviral TNF-­alpha gene therapy and radiation damage tumor vasculature in a human malignant glioma xenograft. Gene Ther 1998;5:293-300. 126. Müller HJ. Artificial transmutation of the gene. Science 1927;66:84-87. 127. Russell WL. The effect of radiation dose rate and fractionation on mutation in mice. In Sobels FH (ed): Repair of ­Genetic Radiation Damage. New York: Pergamon Press, 1963, pp 205-217. 128. Russell WL. Effect of the interval between irradiation and ­conception on mutation frequency in female mice. Proc Natl Acad Sci U S A 1965;54:1552-1557. 129. United Nations Scientific Committee on the Effects of Atomic Radiation. Hereditary Effects of Radiation: The UNSCEAR 2001 Report to the General Assembly with Scientific Annex. New York: United Nations, 2001. 130. Neel JV, Schull WJ, Awa AA, et al. The children of parents ­exposed to atomic bombs: estimates of genetic doubling dose of radiation for humans. Am J Hum Genet 1990;46:1053-1072. 131. Committee on the Biological Effects of Ionizing Radiation. BEIR V: Health Effects of Exposure to Low Levels of Ionizing ­Radiation. Washington, DC: National Academy Press, 1990. 132. United Nations Scientific Committee on the Effects of Atomic Radiation. Sources and Effects of Ionizing Radiation. New York: United Nations, 1988.

45

133. International Commission on Radiological Protection. ICRP 60: 1990 Recommendations of the International Commission on ­ Radiological Protection. Oxford, England: Pergamon Press, 1990. 134. Brenner DJ, Curtis RE, Hall EJ, et al. Second cancers after ­radiotherapy for prostate cancer. Cancer 2000;15:398-406. 135. Boice JD Jr, Engholm G, Kleinman RA, et al. Radiation dose and second cancer risk in patients treated for cancer of the ­cervix. Radiat Res 1988;116:3-55. 136. Bhatia S, Robison LL, Oberlin O, et al. Breast cancer and other second neoplasms after childhood Hodgkin’s disease. N Engl J Med 1996;334:745-751. 137. Sankila R, Garwicz S, Olsen JH, et al. Risk of subsequent malig­nant neoplasms among 1,641 Hodgkin’s disease patients ­diagnosed in childhood and adolescence: a population-based  cohort study in the five Nordic countries. J Clin Oncol 1996;14:  1442-1446. 138. Travis LB, Curtis RE, Boice JD Jr, et al. Late effects of treatment for childhood Hodgkin’s disease. N Engl J Med 1996;335:  352-353. 139. Nyandoto P, Muhonen T, Joensuu H. Second cancers among long-term survivors from Hodgkin’s disease. Int J Radiat Oncol Biol Phys 1998;42:373-378.

ADDITIONAL READING Historical Background del Regato JA.Radiological Physicists. New York: American Institute of Physics, 1985. Hall EJ. Radiation and Life. New York: Pergamon Press, 1984. Hall EJ, Giaccia AJ. Radiobiology for the Radiologist. 6th ed. ­Philadelphia: Lippincott Williams & Wilkins, 2005.

Types of Radiation Goodwin PN, Quimby EH, Morgan RH. Physical Foundations of Radiology. New York: Harper & Row, 1970. Johns H, Cunningham J (eds): The Physics of Radiology, 4th ed. Springfield, IL: Charles C Thomas, 1983. Smith VP (ed): Radiation Particle Therapy. Philadelphia: American College of Radiology, 1976.

Production of Radiation for Therapeutic Applications  Hendee W (ed) Radiation Therapy Physics. St Louis: Mosby; 1981. Johns H, Cunningham J (eds). The Physics of Radiology, 4th ed. Springfield, IL: Charles C Thomas, 1983.

Dose: Its Measurement and Meaning Rossi HH. Neutron and heavy particle dosimetry. In Reed GW (ed): Radiation Dosimetry. Proceedings of the International School of Physics. New York: Academic Press, 1964.

Physical Basis for Choice of Treatment Techniques Khan F (ed): The Physics of Radiation Therapy. Baltimore: ­Williams & Wilkins, 1984. Levitt SH, Khan FM, Potish RA. Levitt and Tapley’s Technological Basis of Radiation Therapy: Practical Clinical Applications, 2nd ed. Philadelphia: Lea & Febiger, 1992.

Signal Transduction Amundson SA, Myers TG, Fornace AJ. Roles of p53 in growth arrest and apoptosis: putting on the brakes after genotoxic stress. Oncogene 1998;17:3287-3299.

The Cell Cycle Hartwell LH, Weinert TA. Checkpoints: controls that ensure the order of cell cycle events. Science 1989;246:629-634.

46

PART 1  Principles

Kastan MB, Zhan Q, El-Deiry WS, et al. A mammalian cell cycle checkpoint pathway utilizing p53 and GADD45 is defective in ataxia-telangiectasia. Cell 1992;71:587-597. McKenna W, Iliakis G, Weiss MC, et al. Increased G2 delay in ­radiation-resistant cells obtained by transformation of primary rat embryo cells with the oncogenes H-ras and v-myc. Radiat Res 1991;125:283-287.

Oncogenes and Tumor Suppressor Genes Bishop JM, Varmus HE. Functions and origins of retroviral transforming genes. In Weiss R, Teich N, Varmus H, Coffinm J (eds): RNA Tumor Viruses: Molecular Biology of Tumor Viruses, 2nd ed. New York: Cold Spring Harbor Laboratory, 1984, pp 990-1108. Cavanee WK. Tumor progression stage: specific losses of heterozygosity. Int Symp Princess Takamatsu Cancer Res Found 1989;20:33-42. Cole M. The myc oncogene: its role in transformation and differentiation. Annu Rev Genet 1986;20:361-384. Hartwell LH, Kastan MB. Cell cycle control and cancer. Science 1994;266:1821-1828. Rous P. A sarcoma of fowl transmissible by an agent separable from the tumor cells. J Exp Med 1911;13:397. Stanbridge EJ. Suppression of malignancy in human cells. Nature 1976;260:17-20.

Radiation Chemistry Boag JW. The action of ionizing radiation on dilute aqueous ­solutions. Phys Med Biol 1965;10:457-476. Ward JF. Biochemistry of DNA lesions. Radiat Res 1985;104:  S103-S111.

Cell Survival Curves Elkind MM, Sutton H. Radiation response of mammalian cells grown in culture. I. Repair of x-ray damage in surviving Chinese hamster cells. Radiat Res 1960;13:556-593. Puck TT, Markus PI. Action of x-rays on mammalian cells. J Exp Med 1956;103:653-666.

Effects on Normal Tissues Cox JD, Byhardt RW, Wilson JF, et al. Complications of radiation therapy and factors in their prevention. World J Surg 1986;10:  171-188. Denekamp J. Cell kinetics and radiation biology. Int J Radiat Biol 1986;49:357-380. Fowler JF, Kragt K, Ellis RE, et al. The effect of divided doses of 15 MeV electrons in the skin response of mice. Int J Radiat Biol 1965;9:241-252. Fowler JF, Morgan RL, Silvester JA, et al. Experiments with fractionated x-ray treatment of the skin of pigs. I. Fractionation up to 28 days. Br J Radiol 1963;36:188-196. Hendry JH, Thames HD. The tissue-rescuing unit. Br J Radiol 1986;59:628-630. Hopewell JW. Late radiation damage to the central nervous system: a radiobiological interpretation. Neuropathol Appl Neurobiol 1979;5:329-343. Hopewell JW. Mechanisms of the action of radiation on skin and underlying tissues. Br J Radiol 1986;19 (Suppl):39-51. Hopewell JW, Campling D, Calvo W, et al. Vascular irradiation damage: its cellular basis and likely consequences. Br J Cancer 1986;53:181-191. Hopewell JW, Morris AD, Dixon-Brown A. The influence of field size on the late tolerance of the rat spinal cord to single doses of  x-rays. Br J Radiol 1987;60:1099-1108. Kerr JRF, Searle J. Apoptosis: its nature and kinetic role. In Meyn RE, Withers HR (eds): Radiation Biology in Cancer Research. New York: Raven Press, 1980, pp 367-384. Michalowsli A. A critical appraisal of clonogenic survival assays in the evaluation of radiation damage to normal tissues. Radiother Oncol 1984;1:241-246. Phillips TL, Margolis L. Radiation pathology and the clinical response of lung and esophagus. Front Radiat Ther Oncol 1972;6:254-273.

Phillips TL, Ross G. A quantitative technique for measuring renal damage after irradiation. Radiology 1973;109:457-462. Phillips TL, Ross G. Time-dose relationships in the mouse esophagus. Radiology 1974;113:435-440. Rubin P. Late effects of chemotherapy and radiation therapy: a new hypothesis. Int J Radiat Oncol Biol Phys 1984;10:5-34. Rubin P, Casarett GW. Clinical Radiation Pathology, vol 1. ­Philadelphia: WB Saunders, 1968. Stewart FA. Mechanism of bladder damage and repair after treatment with radiation and cytostatic drugs. Br J Cancer 1986;7, (Suppl):280-291. Travis EL. Liao Z-X, Tucker SL. Spatial heterogeneity of the volume effect for radiation pneumonitis in mouse lung. Int J Radiat Oncol Biol Phys 1997;38:1045-1054. Travis EL, Tucker SL. Relationship between functional assays of radiation response in the lung and target cell depletion. Br J Cancer Suppl 1986;7:304-319. van der Kogel AJ. Mechanisms of late radiation injury in the spinal cord. In Meyn RE, Withers HR (eds): Radiation Biology in Cancer Research. New York: Raven Press, 1980, pp 461-470. van der Kogel AJ. Central nervous system radiation injury in small animal models. In Gutin PH, Leibel SA, Sheline GE (eds): Radiation Injury to the Nervous System. New York: Raven Press, 1991,  pp 91-112. van der Schueren E, Landuyt W, Ang KK, et al. From 2 Gy to 1 Gy per fraction: sparing effect in rat spinal cord? Int J Radiat Oncol Biol Phys 1988;14:297-300. Withers HR, Taylor JMG, Maciejewski B. Treatment volume and ­tissue tolerance. Int J Radiat Oncol Biol Phys 1988;14:751-759. Withers HR, Thames HD, Peters LJ, et al. Keynote address: Normal tissue radioresistance in clinical radiotherapy. In Fletcher GH, Nervi C, Withers HR (eds): Biological Bases and Clinical Implications of Tumor Radioresistance. New York: Masson, 1983,  pp 139-152.

Effects on Tumors Barendsen GW, Broerse JJ: Experimental radiotherapy of a rat ­rhabdomyosarcoma with 15 MeV neutrons and 300 kV x-rays. I. Effects of single exposures. Eur J Cancer 1969;5:373-391. Hewitt HB. Studies in the quantitative transplantation of mouse sarcoma. Br J Cancer 1953;7:367-383. Hewitt HB, Wilson CW. A survival curve for mammalian cells ­irradiated in vivo. Nature 1959;183:1060-1061. Hewitt HB, Chan DPS, Blake ER. Survival curves for clonogenic cells of a murine keratinizing squamous carcinoma irradiated in vivo or under hypoxic conditions. Int J Radiat Biol 1967;12:535-549. Hill EP, Bush RS. A lung colony assay to determine the radiosensitivity of the cells of a solid tumor. Int J Radiat Biol 1969;15:  435-444. Powers WE, Tolmach LJ. A multicomponent x-ray survival curve for mouse lymphosarcoma cells irradiated in vivo. Nature 1963;197:710-711.

Tissue and Tumor Kinetics Begg AC, McNally NJ, Schrieve DC, et al. A method to measure the duration of DNA synthesis and the potential doubling time from a single sample, Cytometry 1985;6:620-626. Denekamp J. The cellular proliferation kinetics of animal tumors. Cancer Res 1970;30:393-400. Dolbeare F, Beisker W, Pallavicini M, et al. Cytochemistry for BrdUrd/DNA analysis: stoichiometry and sensitivity. Cytometry 1985;6:521-530. Dolbeare F, Gratzner H, Pallavicini M, et al. Flow ­ cytometric ­measurement of total DNA content and incorporated ­bromodeoxyuridine. Proc Natl Acad Sci U S A 1983;80:55735577. Frindel E, Malaise EP, Tubiana M. Cell proliferation kinetics in five human solid tumors. Cancer 1968;22:661-662. Gray JW. Cell-cycle analysis of perturbed cell populations. ­Computer simulation of sequential DNA distributions. Cell Tissue Kinet 1976;9:499-516.

1  Physical and Biologic Basis of Radiation Therapy Mendelsohn ML. The growth fraction: a new concept applied to  tumors. Science 1960;132:1496. Mendelsohn ML. Principles, relative merits, and limitations of current cytokinetic methods. In Drewinko B, Humphrey RM (eds): Growth Kinetics and Biochemical Regulation of Normal and Malignant Cells.. Baltimore: Williams & Wilkins, 1977. Tubiana M, Malaise EP. Growth rate and cell kinetics in human ­tumors: some prognostic and therapeutic implications. In Symington T, Carter RL (eds): Scientific Foundations of Oncology.  St Louis: Mosby, 1976.

Systemic Effects Bond VP, Fliedner TM, Archambeau JO: Mammalian Radiation Lethality: A Disturbance in Cellular Kinetics. New York: Academic Press, 1965. Hemplemann LH, Lisco H, Hoffman JG. The acute radiation ­syndrome: a study of nine cases and a review of the problem. Ann Intern Med 1952;36:279-510. Lushbaugh CC. Reflections on some recent progress in human radiobiology. In Augenstein LG, Mason R, Zelle M (eds): Advances in Radiation Biology. New York: Academic Press, 1969. Shipman TL, Lushbaugh CC, Peterson D, et al. Acute radiation death resulting from an accidental nuclear critical excursion.  J ­Occup Med 1961;3:145-192. Vriesendorp HM, van Bekkum DW. Role of total body irradiation in conditioning for bone marrow transplantation. In Thierfelder S, Rodt H, Kolb HJ (eds): Immunobiology of Bone Marrow Transplantation. Berlin: Springer Verlag, 1980. Vriesendorp HM, van Bekkum DW. Total body irradiation. In Broerse JJ, MacVittie T (eds): Response to Total Body Irradiation in Different Species. Amsterdam: Martinus Nijhoff, 1984.

Fractionation Bedford JS, Hall EJ. Survival of HeLa cells cultured in vitro and ­exposed to protracted gamma irradiation. Int J Radiat Biol 1963;7:377-383. Bedford JS, Mitchell JB. Dose-rate effects in synchronous mammalian cells in culture. Radiat Res 1973;54:316-327. Begg AC, Hofland I, Van Glabekke M, et al. Predictive value of potential doubling time for radiotherapy of head and neck tumor patients: results from the EORTC Cooperative Trial 22851. Semin Radiat ­Oncol 1992;2:22-25. Belli JA, Shelton M. Potentially lethal radiation damage: repair by mammalian cells in culture. Science 1969;165:490-492. Ben-Hur E, Elkind MM, Bronx BV. Thermally enhanced radiosensitivity of cultured Chinese hamster cells: inhibition of repair of sublethal damage and enhancement of lethal damage. Radiat Res 1974;55:38-51. Dolbeare F, Beisker W, Pallavicini M, et al. Cytochemistry for BrdUrd/DNA analysis: stoichiometry and sensitivity. Cytometry 1985;6:521-530. Dolbeare F, Gratzner H, Pallavicini M, et al. Flow cytometric ­measurements of total DNA content and incorporated bromodeoxyuridine. Proc Natl Acad Sci U S A 1983;80:5573-5577. Elkind MM, Sutton H. Radiation response of mammalian cells grown in culture. I. Repair of x-ray damage in surviving Chinese hamster cells. Radiat Res 1960;13:556-593. Fowler JF. Dose response curves for organ function or cell survival. Br J Radiol 1983;56:497-500. Fowler JF. What next in fractionated radiotherapy? Br J Cancer 1984;49:285-300. Fowler JF. Potential for increasing the differential response between tumors and normal tissues: can proliferation rate be used? Int J Radiat Oncol Biol Phys 12:641-645 1986. Fowler JF, Tepper JE. Fractionation in radiation therapy. Semin Radiat Oncol 1992;2:1-72. Hall EJ. Radiation dose-rate: a factor of importance in radiobiology and radiotherapy. Br J Radiol 1972;45:81-97. Hall EJ. The biological basis of endocurietherapy. The Henschke Memorial  Lecture, 1984. Endocuriether Hypertherm Oncol 1985;1:141-151. Hall EJ, Bedford JS. Dose rate: its effect on the survival of HeLa cells irradiated with gamma rays. Radiat Res 1964;22:305-315.

47

Hansen O, Overgaard J, Hansen HS, et al. Importance of overall ­treatment time for the outcome of radiotherapy of advanced head and neck carcinoma: dependency on tumor differentiation. Radiother ­Oncol 1997;43:47-51. Hendry JH, Bentzen SM, Dale RG, et al. A modeled comparison of the effects of using different ways to compensate for missed treatment days in radiotherapy. Clin Oncol 1996;8:297-307. Little JB, Hahn GM, Frindel E, et al. Repair of potentially lethal radiation damage in vitro and in vivo. Radiology 1973;106:689-694. Mitchell JB, Bedford JS, Bailey SM. Dose-rate effects on the cell cycle and survival of S3 HeLa and V79 cells. Radiat Res 1979;79:520-536. Nakatsugawa S, Kumar A, Ono K, et al. Increased tumor curability by radiotherapy combined with PLDR inhibitors in murine ­cancers. In Prospective Methods of Radiation Therapy in Developing Countries, IAEA-TECDOC 266. Vienna: International Atomic Energy Agency, 1982, pp 77-86. Nakatsugawa S, Sugahara T, Kumar A. Purine nucleoside analogues inhibit the repair of radiation induced potentially lethal damage in mammalian cells in culture. Int J Radiat Biol 1982;41:343-346. Overgaard J, Hjelm-Hansen M, Johnsen LV, et al. Comparison of conventional and split-course radiotherapy as primary treatment in carcinoma of the larynx. Acta Oncol 1988;27:147-152. Overgaard J, Sand Hansen H, Sapru W, et al. Conventional radiotherapy as the primary treatment of squamous cell carcinoma of the head and neck: a randomized multicenter study of 5 versus 6 fractions per week—preliminary report from the DAHANCA 6 and 7 trial. ­Radiother Oncol 1996;40:S31. Peters LJ, Withers HR, Thames HD. Radiobiological bases for ­multiple daily fractionation. In Kaercher KH, Kogelnik HD, Reinartz G (eds): Progress in Radio-oncology, vol 2. New York: Raven Press, 1982. Phillips RA, Tolmach LJ. Repair of potentially lethal damage in  x-­irradiated HeLa cells. Radiat Res 1966;29:413-432. Saunders MI, Dische S, Barrett A, et al. Randomised multicentre trials of CHART vs. conventional radiotherapy in head and neck cancer and non–small cell lung cancer: an interim report. Br J Cancer 1996;73:1455-1462. Saunders MI, Dische S, Hong A, et al. Continuous hyperfractionated accelerated radiotherapy in locally advanced carcinoma of the head and neck region. Int J Radiat Oncol Biol Phys 1989;17:  1287-1293. Sinclair WK. Cyclic x-ray responses in mammalian cells in vitro. Radiat Res 1968;33:620-643. Sinclair WK. Radiation survival in synchronous and asynchronous Chinese hamster cells in vitro. In Biophysical Aspects of ­Radiation Quality. Proceedings of the Second IAEA Panel, Vienna, 1967. Vienna: IAEA, 1968. Sinclair WK. Dependence of radiosensitivity upon cell age. In Proceedings of the Carmel Conference on Time and Dose ­Relationships in Radiation Biology as Applied to Radiotherapy, Brookhaven National Laboratory Report 50203 (C-57). Upton, NY: Brookhaven National Laboratory, 1969, pp 97-107. Sinclair WK, Morton RA. X-ray sensitivity during the cell generation cycle of cultured Chinese hamster cells. Radiat Res 1966;29:  450-474. Suit H, Urano M. Repair of sublethal radiation injury in hypoxic cells of a C3H mouse mammary carcinoma. Radiat Res 1969;37:  423-434. Terasima R, Tolmach LJ. X-ray sensitivity and DNA synthesis in synchronous populations of HeLa cells. Science 1963;140:490-492. Tubiana N, Malaise E. Growth rate and cell kinetics in human tumors: some prognostic and therapeutic implications. In Symington T, Carter RL (eds): Scientific Foundations of Oncology. St Louis: Mosby, 1976. Weichselbaum RR, Little JB, Nove J. Response of human osteosarcoma in vitro to irradiation: evidence for unusual cellular repair activity. Int J Radiat Biol 1977;31:295-299. Weichselbaum RR, Schmitt A, Little JB. Cellular repair factors influencing radiocurability of human malignant tumors. Br J Cancer 1982;45:10-16. Withers HR. Isoeffect curves for various proliferative tissues in experimental animals. In Proceedings of Conference on Time-dose ­Relationships in Clinical Radiotherapy. Madison, WI: Madison ­Printing and Publishing, 1975.

48

PART 1  Principles

Withers HR. Response of tissues to multiple small dose fractions. Radiat Res 1977;71:24-33. Withers HR, Peters LJ, Taylor JMG, et al. Late normal tissue sequelae from radiation therapy for carcinoma of the tonsil: patterns of fractionation study of radiobiology. Int J Radiat Oncol Biol Phys 1995;33:563-568.

Modifiers of Radiation Effects Adams GE. Chemical radiosensitization of hypoxic cells. Br Med Bull 1973;29:48-53. Adams GE, Clarke ED, Gray P, et al. Structure-activity relationships in the development of hypoxic cell radiosensitizers. II. Cytotoxicity and therapeutic ratio. Int J Radiat Biol 1979;35:151-160. Bagshaw MA,Doggett RL,Smith KC.Intra-arterial 5-bromodeoxyuridine  and x-ray therapy. AJR Am J Roentgenol 1967;99:889-894. Barendsen GW. Impairment of the proliferative capacity of human cells in culture by alpha particles with differing linear energy transfer. Int J Radiat Biol 1964;8:453-466. Batterman JJ. Clinical Application of Fast Neutrons: the Amsterdam Experience. Amsterdam: Rodipi, 1981. Brown JM, Goffinet DR, Cleaver JE, et al. Preferential r­adiosensitization of mouse sarcoma relative to normal skin by chronic intra-­arterial infusion of halogenated pyrimidine analogs. J Natl Cancer Inst 1971;47:75-89. Brown JM, Lemmon MJ. Potentiation by the hypoxic cytotoxin SR4233 of cell killing produced by fractionated irradiation of mouse tumors. Cancer Res 1990;50:7745-7749. Catterall M. The treatment of advanced cancer by fast neutrons from the Medical Research Council’s cyclotron at Hammersmith Hospital, London. Eur J Cancer 1974;10:343-347. Catterall M. Results of neutron therapy: differences, correlations, and improvements. Int J Radiat Oncol Biol Phys 1982;8:2141-2144. Chapman JD, Whitmore GF. Chemical modifiers of cancer ­treatment. Int J Radiat Oncol Biol Phys 1984;10:1161-1813. Coleman CN. Hypoxic cell radiosensitizers: expectations and progress in drug development. Int J Radiat Oncol Biol Phys 1985;11:  323-329. Deschner EE, Gray LH. Influence of oxygen tension on x-ray-­induced chromosomal damage in Ehrlich ascites tumor cells ­irradiated in vitro and in vivo. Radiat Res 1959;11:115-146. Elkind MM, Swain RW, Alescio T, et al. Oxygen, nitrogen, recovery and radiation therapy. In Shalek R (ed): Cellular Radiation Biology. Baltimore: Williams & Wilkins, 1965. Field SB, Jones T, Thomlinson RH. The relative effect of fast neutrons and x-rays on tumor and normal tissue in the rat. II. Fractionation recovery and reoxygenation. Br J Radiol 1968;41:597-607. Fowler JF, Morgan RL. Pretherapeutic experiments with the fast neutron beam from the Medical Research Council cyclotron. VIII. General review. Br J Radiol 1963;36:115-121. Fowler JF, Morgan RL, Wood CAP. Pretherapeutic experiments with fast neutron beam from the Medical Research Council cyclotron. I. The biological and physical advantages and problems of neutron therapy. Br J Radiol 1963;36:77-80. Hall EJ. Radiobiology of heavy particle radiation therapy: cellular studies. Radiology 1973;108:119-129. Hall EJ, Astor M, Geard C, et al. Cytotoxicity of Ro-07-0582; enhancement by hyperthermia and protection by cysteamine. Br J Cancer 1977;35:809-815. Hall EJ, Graves RG, Phillips TL, et al. Particle accelerators in ­radiation therapy. Int J Radiat Oncol Biol Phys 1982;8:2041-2207. Hall EJ, Roizin-Towle L. Hypoxic sensitizers: radiobiological studies at the cellular level. Radiology 1975;117:453. Howes AE. An estimation of changes in the proportion and absolute numbers of hypoxic cells after irradiation of transplanted C3H mouse mammary tumors. Br J Radiol 1969;42:441-447. Kallman RF, Jardine LJ, Johnson CW. Effects of different schedules of dose fractionation on the oxygenation status of a transplantable mouse sarcoma. J Natl Cancer Inst 1970;44:369-377. Kinsella T, Mitchell J, Russo A, et al. Continuous intravenous ­infusions of bromodeoxyuridine as a clinical radiosensitizer. J Clin Oncol 1984;2:1144-1150.

Kinsella T, Mitchell J, Russo A, et al. The use of halogenated thymidine analogs as clinical radiosensitizers: rationale, current status, and future prospects: non-hypoxic cell sensitizers. Int J Radiat Oncol Biol Phys 1984;10:1399-1406. Mitchell J, Kinsella T, Russo A, et al. Radiosensitization of hematopoietic precursor cells (CFUc) in glioblastoma patients receiving ­intermittent intravenous infusions of bromodeoxyuridine (BUdR). Int J Radiat Oncol Biol Phys 1983;9:457-463. Mitchell J, Russo A, Kinsella T, et al. The use of non-hypoxic cell sensitizers in radiobiology and radiotherapy. Int J Radiat Oncol Biol Phys 1986;12:1513-1518. Moulder JE, Rockwell S. Hypoxic fractions of solid tumors: ­experimental techniques, methods of analysis and a survey of ­existing data. Int J Radiat Oncol Biol Phys 1984;10:695-712. Nordsmark M, Overgaard M, Overgaard J. Pretreatment oxygenation predicts radiation response in advanced squamous cell carcinoma of the head and neck. Radiother Oncol 1996;41:31-40. Overgaard J, Horsman MR. Modification of hypoxia induced ­radioresistance in tumours by the use of oxygen and sensitizers. Semin Radiat Oncol 1996;6:10-21. Overgaard J, Sand-Hansen H, Lindeløv B, et al. Nimorazole as a hypoxic radiosensitizer in the treatment of superglottic larynx and pharynx carcinoma. First report from the Danish Head and Neck Cancer Study (DAHANCA) protocol 5-85. Radiother Oncol 1991;20:143-150. Overgaard J, Sand-Hansen H, Overgaard M, et al. The Danish Head and Neck Cancer Study Group (DAHANCA) randomized trials with radiosensitizers in carcinoma of the larynx and pharynx. In Dewey WC, Eddington M, Fry RJM, et al (eds): Radiation Research.  A Twentieth Century Perspective, vol �������������������������������� II������������������������������ . New York: Academic Press, 1992, pp 573-577. Powers WE, Tolmach LJ. A multicomponent x-ray survival curve for mouse lymphosarcoma cells irradiated in vivo. Nature 1963;197:710-711. Report of the RBE Committee to the International ­ Commissions on Radiological Protection and on Radiological Units and ­Measurements. Health Phys 1963;9:357. Roizin-Towle LA, Hall EJ, Flynn M, et al. Enhanced cytotoxicity of melphalan by prolonged exposure to nitroimidazoles: the role of endogenous thiols. Int J Radiat Oncol Biol Phys 1982;8:757-760. Rose CM, Millar JL, Peacock JH, et al. Differential enhancement of melphalan cytotoxicity in tumor and normal tissue by ­misonidazole. In Brady LW, ed: Radiation Sensitizers. New York: Masson ­Publishing, 1980. Sutherland RM. Selective chemotherapy of non-cycling cells in an in vitro tumor model. Cancer Res 1974;34:3501. Sutherland RM (ed). Conference on chemical modification: radiation and cytotoxic drugs, Key Biscayne, Florida, 17-20 September 1981. Int J Radiat Oncol Biol Phys 1982;8:323-815. Thomlinson RH. Changes of oxygenation in tumors in relation to irradiation. Front Radiat Ther Oncol 1968;3:109-121. Thomlinson RH. Reoxygenation as a function of tumor size and histopathological type. In Proceedings of the Carmel Conference on Time and Dose Relationships in Radiation Biology as Applied to Radiotherapy. BNL Report 50203 (C-57). Upton, NY: Brookhaven National Laboratory, 1969, pp 242-254. Thomlinson RH, Dische S, Gray AJ, et al. Clinical testing of the radiosensitizers Ro-07-0582, III: response of tumors. Clin Radiol 1976;27:167-174. Utley JF, Marlowe C, Waddell WJ. Distribution of 35S-labeled WR2721 in normal and malignant tissues of the mouse. Radiat Res 1976;68:284-291. van Putten LM. Tumor reoxygenation during fractionated radiotherapy: studies with a transplantable osteosarcoma. Eur J Cancer 1968;4:173-182. Withers HR, Thames HD, Peters LJ. Biological bases for high RBE values for late effects of neutron irradiation. Int J Radiat Oncol Biol Phys 1982;8:2071-2076. Wong TW, Whitmore GF, Gulyas S. Studies on the toxicity and ­radiosensitizing ability of misonidazole under conditions of prolonged incubation. Radiat Res 1978;75:541-555.

1  Physical and Biologic Basis of Radiation Therapy Yuhas J. Active versus passive absorption kinetics as the basis for selective protection of normal tissues by S-2-(3-aminopropylamino)ethyl-phosphorothioic acid. Cancer Res 1980;40:1519-1524.

Adjuncts to Radiation Cox JD, Byhardt RW, Komaki R, et al. Interaction of thoracic irradiation and chemotherapy on local control and survival in small cell carcinoma of the lung. Cancer Treat Rep 1979;63:1251-1255. Dewey WC, Hopwood LE, Sapareto SA, et al. Cellular responses to combinations of hyperthermia and radiation. Radiology 1977;123:463-474. Eddy HA. Alterations in tumor microvasculature during ­hyperthermia. Radiology 1980;137:515-521. Elkind MM. Fundamental questions in the combined use of radiation and chemicals in the treatment of cancer. Int J Radiat Oncol Biol Phys 1979;5:1711-1720. Gerner EW, Boone R, Conner WG, et al. A transient thermotolerant survival response produced by single thermal doses in HeLa cells. Cancer Res 1976;36:1035-1040. Gerweck LE, Gillette EL, Dewey WC. Killing of Chinese hamster cells in vitro by heating under hypoxic or aerobic conditions. Eur J Cancer 1974;10:691-693. Gerweck LE, Gillette EL, Dewey WC. Effect of heat and radiation on synchronous Chinese hamster cells: killing and repair. Radiat Res 1975;64:611-623. Hahn GM, Braun J, Har-Kedar I. Thermochemotherapy: synergism between hyperthermia (42°-43°) and Adriamycin (or bleomycin) in mammalian cell inactivation. Proc Natl Acad Sci U S A 1975;72:937-940. Hall EJ, Roizin-Towle L. Biological effects of heat. Cancer Res 1984;44:4708-4713S. Perez C, Brady L, Cox J, et al. Randomized study to evaluate efficacy of levamisole in patients with unresectable non–oat cell carcinoma of the lung treated with radiation therapy. Int J Radiat Oncol Biol Phys 1984;10:97-98. Rubin P. The Franz Buschke Lecture. Late effects of chemotherapy and radiation therapy: a new hypothesis. Int J Radiat Oncol Biol Phys 1984;10:5-34. Stewart JR. Past clinical studies and future directions. Cancer Res 1984;44:4902-4904S. Tefft M, Chabora BMcC, Rosen G. Radiation in bone sarcomas. A reevaluation in the era of intensive systemic chemotherapy. Cancer 1977;39:806-816. Tubiana M. Les associations radiotherapiechemotherapie. J Eur Radiotherapy 1980;1:107-114. Westra A, Dewey WC. Variation in sensitivity to heat shock during the cell cycle of Chinese hamster cells in vitro. Int J Radiat Biol 1971;19:467-477.

Mutagenesis and Carcinogenesis Boice JD Jr, Land CE, Shore RE, et al. Risk of breast cancer ­following low-dose exposure. Radiology 1979;131:589-597. Bond VP, Thiessens JW (eds). Re-evaluation of Dosimetric Factors: Hiroshima and Nagasaki. Proceedings Symposium. Springfield, VA: US Department of Energy, US Department of Commerce, 1982. Cardis E, Gilbert ES, Carpenter L, et al. Effects of low doses and low dose rates of external ionizing radiation: cancer ­ mortality among ­ nuclear industry workers in three countries. Radiat Res 1995;142:117-132. Committee on the Biological Effects of Ionizing Radiation. BEIR III Committee Report: Effects on Populations of Exposure to Low Levels of Ionizing Radiation. Washington, DC: National Academy of Sciences, National Research Council, 1980. Court Brown WM, Doll R. Mortality from cancer and other causes after radiotherapy for ankylosing spondylitis. Br Med J 1965;2:  1327-1332.

49

Court- Brown WM, Doll R, Hill AB. The incidence of leukemia following exposure to diagnostic radiation in utero. Br Med J 1960;2:1539-1545. Doll R, Smith PG. The long-term effects of x-irradiation in patients treated for metropathia haemorrhagic. Br J Radiol 1968;41:  362-368. Doll R, Wakeford R. Risk of childhood cancer from fetal irradiation.  Br J Radiol 1997;70:130-139. Finkel AL, Miller CE, Hasterlik RJ: Radium-induced tumors in man. In Mays CW (ed): Delayed Effects of Bone-Seeking Radionuclides. Salt Lake City: University of Utah Press, 1969. Jablon S, Belsky JL, Tachikawa K, et al. Cancer in Japanese exposed as children to atomic bombs. Lancet 1971;1:927-932. Jablon S, Kato H. Childhood cancer in relation to prenatal exposure to atomic bomb radiation. Lancet 1970;2:1000-1003. Hempelmann LH. Risk of thyroid neoplasms after irradiation in childhood. Science 1968;160:159-163. Hempelmann LH, Pifer JW, Burke GJ, et al. Neoplasms in persons treated with x-rays in infancy for thymic enlargement: a report of third follow-up survey. J Natl Cancer Inst 1967;38:317-341. IARC Study Group on Cancer Risk Among Nuclear Industry ­Workers. Direct estimates of cancer mortality due to low doses of ionizing radiation: an international study. Lancet 1994;344:1039-1043. International Commission on Radiological Protection. ICRP 60: 1990 Recommendations of the International Commission on ­Radiological Protection. Oxford, England: Pergamon Press, 1990. MacKenzie I. Breast cancer following multiple fluoroscopies. Br J Cancer 1965;19:1-8. MacMahon B. Prenatal x-ray exposure and childhood cancer. J Natl Cancer Inst 1962;28:1173-1191. Modan B, Baidatz D, Mart H, et al. Radiation induced head and neck tumors. Lancet 1974;1:277. Muller HJ. The nature of the genetic effects produced by radiation. In Hollaender A (ed): Radiation Biology. New York: McGraw-Hill, 1954. Pierce DA, Shimizu Y, Preston DK, et al. Studies of the mortality of atomic bomb survivors, report 12. Part I. Cancer: 1950-1990. Radiat Res 1996;146:1-27. Rowland RE, Stehney AF, Lucas HF. Dose-response relationships for female radium dial workers. Radiat Res 1978;76:308. Russell WL. X-ray-induced mutations in mice. Cold Spring Harbor Symp Quant Biol 1951;16:327-336. Russell WL. The nature of the dose-rate effect of radiation on mutation in mice. Jpn J Genet 1965;40:128-140. Schull WL, Otake M, Neal JV. Genetic effects of the atomic bomb:  a reappraisal. Science 1981;213:1220-1227. Selby PB. Induced skeletal mutations. Genetics 92:127-133, 1979. Selby PB. The doubling dose for radiation, or for any other mutagen, is actually several times larger than has been previously thought if it is based on specific-locus mutation frequencies in mice. Environ Mol Mutagen 1996;27:61. Stewart A, Kneale GW. Changes in the cancer risk associated with obstetric radiography. Lancet 1968;1:104-107. Stewart A, Webb J, Hewitt D. A survey of childhood malignancies.  Br Med J 1958;1:1495-1508. United Nations Scientific Committee on the Effects of Atomic ­Radiation. Sources and Effects of Ionizing Radiation. New York: United Nations, 1977. United Nations Scientific Committee on the Effects of Atomic Radiation. Ionizing Radiation Sources and Biological Effects. New  York: United Nations, 2001. Upton AC. The dose response relation in radiation induced cancer. Cancer Res 1961;21:717-729. Wolff S. Radiation genetics. Annu Rev Genet 1967;1:221-244.

2

Clinical Radiation Oncology Physics George Starkschall, Lei Dong, Peter A. Balter, Almon S. Shiu, Firas Mourtada, Michael Gillin, And Radhe Mohan PLANNING FOR PHOTON BEAM RADIATION THERAPY Target Volume Delineation Imaging for Treatment Planning Immobilization Dose-Calculation Algorithms Inverse Planning Field-in-Field Planning DELIVERY OF PHOTON BEAM RADIATION THERAPY Multileaf Collimation Fixed-Beam Attenuation Helical Tomotherapy Quality Assurance of Intensity-Modulated Radiation Therapy Delivery STEREOTACTIC RADIATION THERAPY Image-Guided Stereotactic Radiosurgery or Radiation Therapy for Cranial Tumors Image-Guided Stereotactic Radiation Therapy for Extracranial Spinal Tumors Image-Guided Stereotactic Body Radiation Therapy for Thoracic Tumors ACCOUNTING FOR MOTION IN RADIATION THERAPY Intrafractional Motion: RespiratoryCorrelated Radiation Therapy Interfractional Motion: Image-Guided Radiation Therapy

The similarly named chapter in the eighth edition of this textbook provided insight into what was perceived to be the frontier of technologic advances in the planning and delivery of radiation in 2002. Probably the major ­advance at that time was the increasing use of ­intensity­modulated radiation therapy (IMRT) to complement three-­dimensional conformal radiation therapy. Cranial stereotactic radiation therapy was also becoming a standard mode of treatment, as was high-dose-rate brachytherapy. Intravascular brachytherapy showed significant promise, and respiratory gating was being proposed as a method to account for intrafractional respiratory-induced tumor ­motion. Major changes have taken place in radiation therapy since the previous edition. IMRT planning and delivery are now performed routinely for the treatment of tumors at many anatomic sites, although the definition of an appropriate objective function and constraints still remains more of an art than a science. As for frontier technologies,

50

EVOLUTION OF IN-ROOM IMAGING FOR RADIATION THERAPY On-Board Kilovoltage X-Rays Sonographically Guided Prostate Localization Three-Dimensional In-Room Volumetric Imaging Image-Guided Adaptive Radiation Therapy DEVELOPMENTS IN BRACHYTHERAPY Pulsed-Dose-Rate Brachytherapy for Cervical Cancer Liquid Source Brachytherapy for Brain Tumors Partial Breast Brachytherapy Beta Particle Plaques for Ocular Melanoma PROTON RADIATION THERAPY Production of Protons Passively Scattered Proton Beams Shaping Proton Dose Distributions Treatment Planning for Passively Scattered Protons Evaluating Proton Treatment Plans Discrete Spot Scanned Proton Beams Proton Radiobiologic Effectiveness FUTURE TECHNOLOGIC DEVELOPMENTS IN RADIATION THERAPY Image-Guided Radiation Therapy Biologic and Functional Imaging and Dose-Response Assessment and Modeling Proton Therapy

cranial stereotactic radiation therapy and high-dose-rate brachytherapy were replaced by extracranial stereotactic radiation therapy and pulsed-dose-rate (PDR) brachytherapy. Intravascular brachytherapy has not become a significant component of brachytherapy practice, having been replaced by medical intervention. Although gated beam delivery is still considered a promising method for reducing the size of treatment portals, it has not achieved widespread acceptance; instead, explicit determination of respiratory motion is used to customize the delineation of the treatment portal to the particular patient. The current buzzword is image-guided radiation therapy, in which imaging immediately before radiation delivery is used to modify the position of the patient or the parameters of the beam to reflect intrafractional motion and interfractional variations in the patient. Several institutions have opened or are soon to open proton radiation therapy facilities, with the promise of still more precise delivery of radiation to the target while sparing uninvolved tissue.

2  Clinical Radiation Oncology Physics

With these developments in mind, this chapter provides a current snapshot of clinical radiation physics, with particular emphasis on procedures in place or nearly so in the Radiation Oncology Clinic at The University of Texas M. D. Anderson Cancer Center. The first part of the chapter addresses the planning and delivery of high-precision photon beam radiation therapy. A section on stereotactic radiation therapy follows, with emphasis on extracranial hypofractionated treatment of the central nervous system and thorax. The handling of intrafractional motion and interfractional variations is the topic of the next section, ­focusing primarily on methods for imaging these variations. New approaches to brachytherapy are discussed, including PDR brachytherapy, liquid-source brachytherapy, and the use of novel beta-emitting isotopes. Some aspects of proton radiation therapy are presented (see Chapter 40). This chapter closes with a brief review of future technologic ­developments that are likely to take place in clinical radiation ­oncology physics, setting the stage for what may be the topics of interest in the next edition of this book.

Planning for Photon Beam Radiation Therapy The current practice of radiation therapy mandates great precision in the planning and the delivery of radiation. As doses and doses per fraction are escalated with the goal of increasing tumor control rates, the margin for uncertainty needs to be made smaller. Consequently, the need for more precise targeting of tumor volumes has increased, as has the need for accurate accounting of uncertainties in patient setup and tumor motion and for sparing critical uninvolved anatomic structures. Newer technologies have facilitated the achievement of these goals. The following section focuses on methods for high-precision planning for photon beam radiation therapy; this section is followed by aspects of highprecision delivery of photon beam radiation therapy.

A

51

Target Volume Delineation Planning and delivery of high-precision radiation therapy rely on the accurate definition and delineation of target volumes and critical anatomic structures. Although the definitions originally presented by the International Commission on Radiation Units and Measurements (ICRU) were designed to facilitate communication of information about radiation dose and volume, these definitions also provide a rational method for understanding the various factors involved in determining what regions of a patient need to be treated.1,2 Although the ICRU dose and volume definitions were specifically designed for photon radiation therapy, some of the concepts presented in these definitions can also be applied to therapy with other forms of ionizing radiation. Several of these definitions are applicable to oncology in general, independent of the therapeutic modality used. The ICRU1 defines gross tumor volume (GTV) as “the gross demonstrable extent and location of the malignant growth.” The basic rule of thumb for defining the GTV is that if it can be imaged and it is tumor, it is part of the GTV. What is not specified in the ICRU definition is the mode of imaging or the imaging parameters. The GTV may be different when it is determined by different ­ imaging ­modalities; the GTV may even be different for the same modality when different image viewing parameters are used (e.g., window width, window level settings). Consequently, consistency in assessing tumor location and ­extent for each tumor site is highly desirable. As an example, Figure 2-1 depicts two transverse computed tomography (CT) images of a patient’s thorax, one with the window and level settings normally used for displaying mediastinal structures (see Fig. 2-1A) and the other with the window and level settings normally used for displaying lung structures (see Fig. 2-1B). Comparing the two figures reveals that a lung tumor depicted with a mediastinal window may appear quite different from that depicted with a lung window. The GTV alone does not constitute the extent of tumor spread, because it does not include microscopic disease.

B

FIGURE 2-1.  Transverse CT images of a thorax using a mediastinal window and level (A) and a lung window and level (B) illustrate differences in the visualization of a lung tumor.

52

PART 1  Principles

According to the ICRU,1 the clinical target volume (CTV) is defined as “tissue volume that contains a demonstrable GTV and/or subclinical malignant disease that must be eliminated.” The CTV constitutes the volume of tumorbearing cells that must be irradiated to an appropriate dose to control the tumor. Failure to adequately irradiate the CTV to an appropriate dose may result in failure to achieve tumor control. It is possible to have a CTV without a GTV, as would be the case for a tumor that has been surgically resected. It is also possible to have multiple CTVs in a treatment scheme, such as when one dose is prescribed for the primary tumor and another for nodal disease. Exactly how the GTV should be expanded to generate the CTV is based partly on clinical experience and partly on specific pathology studies. For example, Giraud and colleagues3 assessed pathologic records from a series of patients with non–small cell lung tumors and established expansion margins of 8 mm for adenocarcinoma and 6 mm for squamous cell carcinoma. However, these expansion margins were based on a particular method for delineating the GTV. As imaging modalities are developed that can more precisely define the extent of tumor-bearing tissue, these expansion margins may need to be modified. For ­example, if some sort of in vivo pathologic assessment were possible by using extremely high-resolution CT, enabling larger amounts of tumor to be visualized, it might be possible to reduce the margins that generate the CTV from the demonstrable GTV. Because of the improvement in imaging, some portion of tumor that was originally considered CTV and not GTV would become part of the GTV. Most radiation treatment planning systems support software tools that generate the CTV from the GTV by means of uniform expansion of the GTV. After this expansion has been generated, however, the contours of the CTV must be reviewed and edited based on clinical experience. For example, if microscopic disease is limited to the tumorbearing organ, the CTV can be edited to avoid adjacent structures, provided the radiologist is confident that microscopic extension has not occurred. Radiation treatment has long been based on building a patient model based on acquisition of a CT scan several days before treatment and on the assumption that this model is an accurate reflection of the patient for the duration of the radiation treatment. Recognizing that tumors move, the ICRU2 defined another volume that accounts for tumor motion, the internal target volume (ITV). The ITV consists of the CTV plus an internal margin that ­accounts for intrafractional and interfractional motion. Margins ­ accounting for motion can be population-based or specifically measured for each patient. Methods to measure and mitigate the effects of various forms of motion that allow minimization of the internal margin are discussed under ­“Accounting for Motion in Radiation Therapy” in this chapter. The final uncertainty specified by the ICRU is the planning target volume (PTV), which accounts for variations in patient setup. The PTV is the ITV plus a setup margin and is defined as “a geometrical concept used for treatment planning, defined to select appropriate beam sizes and beam arrangements to ensure that the prescribed dose is actually delivered to the CTV.”1 The PTV is the entity that must be irradiated to the desired prescription dose to ensure that the entire CTV (i.e., tumor-bearing tissue) ­receives the prescribed radiation dose. Each CTV has a corresponding PTV. The setup

margin is determined by clinical ­experience and depends on the disease site and the type of patient ­immobilization used. Invasive forms of immobilization, such as stereotactic head frames, allow millimeter-level setup margins, whereas other forms of ­ immobilization may result in setup margins of 5 to 10 mm or more. A transverse CT image of a thorax from a treatment plan used to treat a lung tumor, illustrating a GTV, a CTV, and a PTV, is shown in Figure 2-2. The projection of the PTV in a beam’s eye view does not by itself define the radiation treatment portal; instead, an additional 5- to 7-mm margin must be added to the beam’s eye view projection of the PTV to account for the beam penumbra. The ICRU also addresses organs at risk, which are by definition site-specific. For example, the organs at risk in the treatment of lung tumors typically include the spinal cord, lungs, esophagus, and heart. Other organs, such as the brachial plexus or liver, may also be included, depending on the location of the tumor. In planning the radiation treatment of a lung tumor, the planner must remain cognizant of the dose received by these anatomic structures. ­Accounting for setup uncertainty and motion of the organs at risk ­requires placing a margin around those organs and introduces the concept of the planning organ-at-risk volume. The ostensible goal in planning for radiation treatments is to deliver the prescribed dose to the PTV while keeping the dose to specified volumes of the planning organs at risk within appropriate limits. In reality, compromises are often necessary. Sometimes, setup uncertainty and organ motion lead to overlap between portions of the PTV and the planning organ-at-risk volume such that delivery of high doses to the former may result in overdosing the latter. Because the CTV does not spend all of its time in all of the PTV, a typical criterion for PTV coverage is that 95% of the PTV receives the prescribed dose, and the entire CTV would receive the prescribed dose. However, if coverage of the CTV must be reduced to protect an organ at risk, the reduction is done with the recognition that doing so increases the probability that a portion of tumor cells may not receive a radiation dose considered adequate for tumor control. One of the goals of improving the quality of radiation therapy is to reduce the margin between the GTV and PTV. In an ideal situation, the imaging modality or modalities would

FIGURE 2-2.  Transverse CT image of a thorax from a treatment plan used to treat a lung tumor illustrating the gross tumor  volume (GTV; yellow line), clinical target volume (CTV; red line), and planning target volume (PTV; blue line).

2  Clinical Radiation Oncology Physics

53

1.8 1.6 1.4 Density

1.2 1.0 0.8 0.6 0.4 0.2 0.0 0

500

1000 1500 2000 2500 3000 3500 4000 4500 CT number

FIGURE 2-3.  CT number-to-density conversion table extracted from the radiation treatment planning system.

explicitly display all of the tumor-bearing tissue, and motion and setup uncertainties would be completely eliminated; what is imaged would be what is irradiated. ­Although this is often not the case in practice, it does demonstrate where improvements can be made in targeting. For example, better means of imaging the tumor with various imaging modalities and image-enhancing (contrast) agents may enhance the imaging of microscopic disease and ­reduce the margins used between the GTV and the CTV. (Some of these modalities are described under ­“Imaging for Treatment Planning.”) The magnitude of the expansion of the CTV to generate the ITV can be reduced in two ways: by explicitly imaging the tumor motion and by reducing the amount of intrafractional and interfractional motion. (These issues are addressed under “Accounting for Motion in Radiation Therapy.”) The magnitude of setup ­uncertainty can be ­reduced, allowing reductions in the margin used to expand the ITV to form the PTV. Setup uncertainty can be reduced by more effective methods of patient immobilization and by some of the techniques being used for ­ image-guided ­ radiation ­ therapy, which are discussed further in subsequent sections. The pretreatment GTV in a particular patient is assumed in the sense that the extent or location of the tumor cannot be controlled. The CTV is also assumed, but it could be delineated more precisely by improving the ability to detect microscopic disease and eliminate some uncertainty about its extent. The quantities that can be controlled are the ITV and the PTV. Logically, the potential for dose escalation would be based on the ability to safely reduce the size of the PTV by using better imaging and immobilization methods, as described in the sections that follow.

Imaging for Treatment Planning Imaging has a major role in radiation treatment planning. Effectively reducing the size of the region being irradiated requires methods for accurately determining the location and

extent of tumor. This can be achieved by improving existing imaging technologies and by the use of multimodality imaging. CT imaging, currently the standard for radiation treatment planning, is the most desirable imaging modality for this purpose because it most closely simulates the effects of ionizing radiation passing through the patient. However, CT images are typically acquired by using x-rays in the ­ kilovoltage energy range (120 to 140 kVp), whereas the ­radiation delivered as treatment is in the megavoltage ­energy range. In the kilovoltage energy range, a significant fraction of the photons interact by means of the photoelectric effect. In this energy range, the deposition of ­ energy and the attenuation of radiation depend greatly on the atomic number of the absorber. In the megavoltage energy range, almost all interactions are those of Compton scatter. Energy deposition and radiation dose depend almost entirely on electron density. Consequently, the CT voxel values (which reflect x-ray attenuation) do not exactly ­reproduce the extent to which megavoltage radiation interacts with tissue. Before radiation doses based on CT images can be calculated in a radiation treatment planning system, a table must be generated to convert CT numbers to electron densities. In these tables, each region of CT densities is assumed to represent a different type of tissue, with a ­different atomic number, and the CT-to-density relationship is slightly different for each of these types of tissue (Fig. 2-3). High-Z materials may also cause artifacts on CT ­images because of inadequate accounting for beam hardening due to the increased attenuation of the kilovoltage x-rays in the CT scanner.4 The technology of CT image acquisition has progressed substantially in recent years. The current form of CT technology, helical or spiral CT, is characterized by continuous table motion synchronized with continuous gantry ­rotation. Multislice helical CT scanners (Fig. 2-4) acquire

54

PART 1  Principles

A

B

FIGURE 2-4.  Two multislice, helical CT scanners are used to acquire images for radiation treatment planning in the Radiation Oncology Clinic at M. D. Anderson Cancer Center: the Philips Brilliance 64 CT scanner (A) and the GE Discovery ST CT scanner (B), which is a combined CT scanner with a PET scanner and uses the same patient support system.

2 to 64 image slices, corresponding to a region thickness of 2 to 4 cm, during a single gantry rotation. The ability to acquire patient information over a 2- to 4-cm-thick ­region coupled with gantry rotation times of less than 0.5 seconds has resulted in very short image acquisition times. For ­example, a CT image data set of the thorax extending from the mandible to the base of the liver can be acquired in 5 to 10 seconds. Although short CT image acquisition times have the ­advantage of reducing the magnitude of motion during ­image acquisition, they also introduce motion artifacts. For example, the diaphragm moves in an approximately supero­ inferior direction, parallel with that of the patient during CT image acquisition. Mismatches in the synchronization of the motion of the diaphragm with that of the patient during image acquisition can lead to artifacts in the diaphragm images (Fig. 2-5). Because the time scale for image acquisition is significantly shorter than that for radiation treatment, the CT images acquired on a multislice helical CT scanner may not reflect patient motion that is averaged during radiation treatment. An extension of the multislice CT concept is the cone-beam CT scanner. This type of scanner has no ­table ­translation; a flat-panel detector is placed opposite a ­kilovolt-energy radiation source, and an image is acquired during a single gantry rotation. Many cone-beam CT scanners are manufactured as components of linear accelerators (Fig. 2-6). The image quality of the cone-beam CT scanner may not be as high as that of a conventional CT scanner (Fig. 2-7), and cone-beam scanners therefore are used more for image-guided setup for therapy than for ­treatment ­planning. Although CT imaging provides the most appropriate representation of the interactions of radiation with the ­patient and for imaging anatomic regions of different mass densities, various aspects of tumors may be ­visualized ­better with other imaging modalities. Tumor metabolic activity, for example, is better visualized with positron emission tomography (PET) than with CT. PET is based on the annihilation of positrons with electrons, generating two 511-keV gamma rays emitted in opposite directions. Positron-emitting radionuclides such as fluorine 18 (18F)

FIGURE 2-5.  An artifact in a coronal CT image of a the thorax is caused by diaphragm motion during the acquisition of a fast CT scan. The red line delineates the gross tumor volume at end inspiration; the green line delineates the gross tumor volume at end expiration.

tend to be small atoms and typically bind to physiologically active molecules. Significant interest has developed during the past few years in the use of the compound 18F-fluorodeoxyglucose (FDG), the concentration of which is related to the degree of metabolic activity. Because ­ tumors tend to be more metabolically active than uninvolved tissue, the concentration of FDG, indicated by bright areas on PET scans, can be useful for identifying areas of tumor and nodal activity. A major problem in the use of PET imaging is registering the PET image with the CT image. Although PET may indicate areas of tumor activity better than CT, the radiation treatment planning still must be done with CT

2  Clinical Radiation Oncology Physics

images. The lack of commonality between the two types of images has made registration a significant problem. One solution has been to register the CT image with a transmission ­image, which is normally acquired along with the PET image to correct the PET image for the attenuation of the emitted gamma ray photons. The transmission image is usually ­obtained with a gamma emitter and depicts anatomy similarly to that on a CT image. Because the transmission image is acquired with the same apparatus as the PET image, it is automatically registered to the PET image, enabling the PET image to be registered to the CT image. Development of PET/CT scanner combinations has made the acquisition of the transmission image easier and the registration of the PET image with the planning CT

(d) (a)

(b)

(e)

(c)

FIGURE 2-6.  A kilovoltage source (a) and flat-panel imager (b) are attached to a linear accelerator gantry to produce cone-beam CT image data sets. The treatment table (c), linear ­accelerator source (d), and electronic portal imaging device (e) are shown.

A

55

i­mage automatic. These combined scanners consist of a dual gantry, a PET imager, and a CT imager (see Fig. 2-4B), with the patient table translated between the two ­imagers. Registration is no longer an issue, because the amount of table translation between the two imagers is a known quantity. When the CT image data from a PET/CT combination are used as a transmission correction, the difference in ­acquisition time can be problematic. Because the CT image is acquired over a much shorter time than the PET image, motion artifacts on the CT image data set may ­adversely affect the quality of the attenuation correction of the PET image. Pan and colleagues5 proposed that ­attenuation correction data for PET images be based on a respiration-­averaged CT scan rather than on a free-­breathing CT scan. In the example shown in Figure 2-8, use of the ­respiration-averaged CT data set (see Fig. 2-8B) led to a twofold ­increase in the specific uptake value of the tumor on the PET image (3.8 versus 1.9 for the free-breathing CT) (see Fig. 2-8A). Another imaging modality that can aid in treatment planning is single photon emission computed tomography (SPECT), which involves imaging single photons emitted from an internally applied gamma ray source. SPECT can be particularly useful for assessing lung perfusion and thereby the amount of functional lung tissue in a patient. Radiation treatment planning can then be designed to avoid irradiating well-functioning lung regions, with the idea that if a fraction of lung has to be sacrificed to achieve the prescription goals of the treatment plan, that region should be poorly functioning lung. Figure 2-9 illustrates a SPECT image combined with a CT image used for treatment planning for such a case. SPECT scanners can be combined with CT scanners to improve SPECT/CT image registration. Combined SPECT/CT scanners may not have the same advantages for applications involving only imaging as a combined PET/ CT scanner does, because the CT component of the PET scanner is also needed to obtain the transmission image for ­attenuation correction. Some of the issues related to the differences between photon interactions in CT scanners and those in linear ­accelerators are being addressed by the introduction of megavoltage CT imaging.6 High-Z materials, which produce significant artifacts in kilovoltage CT scans, have no such artifacts on megavoltage CT scans. Megavoltage CT scanners are constructed as cone-beam scanners that use the radiation beam from a linear accelerator or in conjunction with a helical tomotherapy unit. Megavoltage cone-beam

B

FIGURE 2-7.  Images of the same transverse slice of a thorax were obtained by conventional CT (A) and cone-beam CT (B).

56

PART 1  Principles

CT scanners typically are used for patient alignment and dose verification rather than for initial treatment planning.

Immobilization A key approach in reducing setup uncertainty and limiting some intrafractional motion is the use of appropriate immobilization techniques. The specific technique to be

A

used is based on the tolerance for setup uncertainty and patient motion for a given type of treatment (Table 2-1). Cranial stereotactic radiosurgery, for example, requires submillimeter tolerances, and invasive immobilization in the form of stereotactic head frames is used. For treatment of most head and neck tumors, the acceptable setup uncertainty is slightly greater than that for stereotactic

B

FIGURE 2-8.  PET images show an abdominal tumor with the attenuation correction based on a short-acquisition free-­breathing CT image (A) or a time-averaged CT image (B). The cursor is pointing to the tumor. (Courtesy of T. Pan, M. D. ­Anderson Cancer Center, Houston, TX.)

FIGURE 2-9.  Transverse, sagittal, and coronal views of SPECT images are combined with CT images to identify regions of high lung perfusion as critical avoidance structures for radiation treatment planning. The solid white and colored lines outline the various ­isodoses. Lighter areas in the lungs indicate regions of higher perfusion. (Courtesy of H. Liu, M. D. Anderson Cancer Center, Houston, TX.)

2  Clinical Radiation Oncology Physics

57

TABLE 2-1.   Immobilization Devices Used in Radiation Oncology Clinics and Their Associated Setup Uncertainties Immobilization Device

Application

Setup Uncertainty

Invasive head frame

Cranial stereotactic

0.5 mm

Thermoplastic mask and vacuum bag

Head and neck, CNS

3 mm

T bar and vacuum bag

Thorax

7 mm

T bar, extended vacuum bag (indexed to table), CT guidance

Thorax (hypofractionated)

3 mm

Vacuum bag, breast board indexed to table

Breast

2-5 mm

Knee and foot immobilization cushions indexed to table

Prostate

3 mm

CNS, central nervous system, CT, computed tomography.

FIGURE 2-10.  A customized thermoplastic mask is used to immobilize a patient being treated for a tumor of the head and neck region. The red lines are the localizing lasers in the simulation and treatment rooms.

FIGURE 2-11.  A vacuum bag and T-bar assembly is used to immobilize patients being treated for thoracic lesions. The red lines are the localizing lasers in the simulation and treatment rooms.

radiosurgery, and a thermoplastic mask (Fig. 2-10) is used. Thoracic treatments are delivered after the patient is set up with a T bar and vacuum bag system (Fig. 2-11). Patients are positioned on the vacuum bag, with their arms ­extended over their heads and their hands holding the T bar. Air is then drawn out of the vacuum bag, which ­resembles a beanbag filled with very small pellets that harden as the air is withdrawn, forming a template on which the patient can lie so that the patient’s position is reproducible from treatment to treatment. Setup marks are made on the ­patient. When greater precision in setup is required, such as for hypofractionated treatments of the thorax, a larger vacuum bag is used in combination with a table index to allow the bag to be placed in the same position on the table for each treatment. Markings in this case are made on the bag rather than on the patient.

use some form of the convolution algorithm,7,8 the current state of practice, for photon beam dose calculations. The concept behind the convolution algorithm is the existence of a dose-spread array that describes how unit photon fluence deposits dose in the vicinity of an interaction. The dose-spread array depends on only the positional relationship between the calculation point and the site of interaction, as well as the energy of the photons, and it needs to be calculated only once and then stored in the computer. Calculating photon fluence is relatively straightforward; as the photon beam passes through the patient, the fluence is determined by means of exponential attenuation. By modeling the photon fluence as a spectrum of energies and attenuating each spectral component of the photon beam based on its own energy-specific attenuation coefficient, hardening of the radiation beam as it passes through beam modifiers or the patient can be accurately modeled. The dose distribution is then calculated by convolving the dose-spread array with the photon fluence ­distribution. Commissioning the convolution model for a radiation treatment planning system involves modeling the photon fluence that is incident on the patient surface. Although this calculation needs to be done only once for a beam, the beam model may consist of many spectral components

Dose-Calculation Algorithms An important aspect of radiation treatment planning is the accurate simulation of the radiation dose distribution on a computer model of the patient. Various dose-calculation algorithms have been developed to balance the need for computational accuracy with time constraints. The most advanced radiation treatment planning systems typically

58

PART 1  Principles

and several empiric parameters describing the incident fluence distribution, and the calculations can be complex and time-consuming.9 Accuracy of computed dose distributions must then be verified against measured beam data. Still more accurate dose calculations can be achieved by using Monte Carlo models, which explicitly model specific photon interactions. These calculations require significantly more computational power than do convolution calculations, but they are also capable of accurately modeling regions of electronic disequilibrium, such as in the buildup region or at tissue interfaces. Monte Carlo calculations are used more for research than for clinical treatment planning, in part because their computational intensity requires high-speed parallel processors. This issue should be overcome in the near future, given the significant increases in computational power during the past few years. Increases in the accuracy and speed of dose calculations have raised interest in incorporating heterogeneity corrections into treatment planning. Although these corrections can more accurately depict the dose received in the delivery of radiation treatment, many radiation oncology practices have been reluctant to incorporate them into treatment planning, perhaps because the doses calculated by using heterogeneity-corrected calculations might be different from what radiation oncologists were accustomed to. For example, the clinical response to a dose of 60 Gy calculated for a “homogeneous” patient (i.e., the traditional technique, still in widespread use) may be different from the clinical response to a dose of 60 Gy calculated with the true heterogeneities present. Consequently, the ­ transition from calculations in homogeneous medium to calculations in a heterogeneous patient must be made carefully, ­because it is important not to modify the radiation dose that is ­delivered to the patient.10

Inverse Planning In traditional treatment planning, radiation was delivered with spatially uniform (“flat”) fields. Occasionally, wedges were used to modify the shape of the dose distribution to account for oblique beam configurations or for sloping surfaces, and compensating filters were used to modify the intensity of the incident beam to account for major surface irregularities, but the ultimate goal was a uniform dose distribution at some specified depth. Multiple fields with specified orientations were required to ensure a uniform dose distribution in the target volume and an acceptable dose distribution to uninvolved tissue outside of the target. The discovery that the conformality of the dose to desired volumes could be improved by using radiation beams of spatially varying intensities gave rise to IMRT and inverse planning, a process to determine the beam intensities needed to deliver the dose distribution that would achieve the desired clinical objectives. The inverse planning process involves identifying and optimizing a metric of plan quality called a cost function or objective function. (For the sake of generality, we ­assume that the objective function must be minimized; the ­inverse planning problem can be easily reformulated if the objective function has to be maximized.) This metric is a measure of the match between a desired dose distribution and the calculated dose distribution. Several issues are involved in selecting an objective function. The function

must be realistic, achievable, and clinically relevant. Early inverse planning efforts used the square of the difference ­between the calculated doses and the desired doses to target ­volumes and each of the normal critical structures as the objective function. Although mathematically this quantity is relatively easy to optimize, it does not realistically depict the true clinical goals. For example, the dose to a normal anatomic structure does not have to be equal to a desired dose for a plan to be optimal; instead, it should be less than some maximum allowable dose. Such inequalities can be incorporated into an optimization scheme as dose constraints; however, this condition is still not necessarily realistic because doses to individual points in a structure can often be exceeded as long as the dose to a specified volume of that structure is not exceeded. This condition gives rise to the specification of dose-volume constraints, stated as, for example, “No more than 35% of the total lung volume may receive more than 20 Gy.” The objective function can also be stated in biologic terms,11 such as a combination of the probability of tumor control and the probability of normal tissue complications. These objectives can be combined with the goal of maximizing the probability of uncomplicated local control.12 Alternatively, the objective can be stated in terms of equivalent uniform dose, which can be maximized to the target volume and minimized to uninvolved normal structures.13 Stating the objective function in biologic terms has several potential problems, however. For example, models for these biologic factors may accurately calculate these quantities but may not have a mechanistic basis or be valid for inverse planning. Verifying the validity of these models can be problematic. The models also increase the mathematical complexity of the objective function, requiring more sophisticated methods to optimize that function. Inverse planning is further complicated by the need to incorporate relative weighting of various components of the objective function. For example, the relative weighting of overdosing the spinal cord versus underdosing the target volume is a clinical decision that may be different among individual clinicians. In addition to uncertainties in weighting, specific values used in dose-volume constraints may be based on specific treatment techniques and may not be necessarily be generalized to other techniques. For ­example, Graham and colleagues14 established V20, the ­total volume of lung receiving 20 Gy, as the most significant risk factor for pneumonitis in the radiation treatment of non–small cell lung cancer. Because fixed beams were used for the treatment, it is possible that the V20 values correlated with the V10, V5, and V30 values, and one of these other dose-volume values may be a more accurate metric of potential lung toxicity. The use of intensitymodulated beams can significantly reduce the amount of lung receiving 20 Gy, but it does so at the expense of lung receiving 5 Gy, breaking the correlation in dose-volume levels. The objectives and constraints in an inverse planning problem may render the problem infeasible in that the dose-volume constraints cannot be met by any combination of beams and beam intensity patterns. In such cases, the constraints must be relaxed before the objective function can be optimized. Another possibility is that even ­after optimization, the optimal dose distribution may not

2  Clinical Radiation Oncology Physics

be clinically acceptable; in that case, it may be necessary to reformulate the constraints and objectives and reoptimize. After the constraints and objectives have been determined, the inverse planning problem must be solved by using one of a large number of optimization algorithms. Typically, the choice of algorithm is not up to the treatment planner but rather is a function of the particular treatment planning system. In summary, the process of setting up a treatment plan for inverse planning includes the explicit identification of target volumes and critical structures, followed by establishment of constraints and objectives for these regions of interest along with relative weightings for each constraint and objective. Because inverse planning is still somewhat of an art, it is often necessary to develop various tricks to aid in achieving inverse planning goals. For example, ­ensuring adequate coverage of target volumes or appropriate avoidance of critical structures may require adding various margins around those volumes and structures. Rings are sometimes defined around target volumes to ensure a high dose gradient outside the target volume. Ensuring that hot spots do not occur outside the specified critical structures may require that otherwise unspecified uninvolved tissue be considered a critical structure. Evaluating an IMRT plan requires distinguishing between structures that have been explicitly defined for the purpose of the inverse planning and the actual critical structures and target volumes.

Field-in-Field Planning Field-in-field planning, another technique used to generate the effect of intensity-modulated fields but based on forward treatment planning, is used extensively at M. D. Anderson Cancer Center for planning radiation treatments of the breast.15 The basic principle behind field-in-field planning is that of starting with a simple beam configuration, computing the dose distribution from this simple configuration, adding smaller fields to fill in cold spots to make the dose distribution more uniform. In the specific example of field-in-field planning of the breast, the planning begins with a pair of opposed tangential open fields. After the doses are calculated, the beam weights are adjusted until any hot spots are balanced ­between medial and lateral fields. The doses are displayed as a percentage of a point dose (typically the isocenter dose). The planner then decides on an isodose curve, corresponding to a hot spot, to block. Figure 2-12A displays an isodose distribution of a treatment plan of the breast on a transverse slice after the beam weights have been adjusted to balance the hot spots. In this case, the 114% isodose is selected for blocking. The medial field is then copied, and the treatment portal is adjusted to cover the identified isodose cloud (see Fig. 2-12B). The dose is calculated for this new field, and the weighting of this field is increased until the isodose cloud disappears. Figure 2-12C illustrates the new dose distribution with the added field and the disappearance of the 114% isodose. The planner then selects a new isodose curve to be blocked, in this case the 109% isodose. The lateral field is copied and the treatment portal is adjusted to cover the selected isodose cloud (Fig. 2-12D). The dose is calculated, and the weighting of the lateral field is increased until the isodose cloud disappears. The selected fields are alternated, modified to block a selected hot spot, the dose

59

r­ ecalculated, and the beam weighted until the dose distribution is sufficiently uniform. Figure 2-12E compares the original dose distribution with the more uniform dose distribution obtained by this field-in-field approach.

Delivery of Photon Beam Radiation Therapy Achieving the goals of high-precision radiation therapy requires that the shape of the radiation field conform precisely to the target volume such that the target volume is irradiated but the surrounding uninvolved anatomy is shielded. The previous section examined how multiple photon beams can be combined to generate a high-dose volume that is shaped to track the outline of the tumor and how modifying the fluence of the incident beams can render the shaping of the high-dose region even more precise. The following section addresses the delivery of these high-precision photon fields.

Multileaf Collimation Collimation, or shaping, of photon beams has progressed from the use of hand-placed blocks to customized blocks cast from low-melting-point alloy to multileaf collimators (MLC), which represent the current state of practice for shaping static radiation treatment fields. The advantage of using MLCs for shaping fields is that no manual intervention is necessary to change the shape of the fields. Field shapes, as determined by MLC locations, are generated by the treatment planning system and transferred directly to the record-and-verify system. The appropriate settings are then delivered to the linear accelerator for beam delivery. Perhaps even more important, the MLC settings can also be verified and recorded. The ability to generate the MLC settings in a treatment plan and transfer the settings directly to the linear accelerator has made feasible programming a sequence of MLC settings that results in the delivery of a spatially dependent fluence pattern, which has given rise to the practical implementation of IMRT delivery. Earlier forms of treatment planning for IMRT gave rise to an optimal fluence map calculated as a distribution of weights of beamlets, which was the solution of the inverse planning problem. One disadvantage of this solution was that the fluence map had to be converted into a series of MLC settings that could deliver the radiation fluence in a practical amount of time. Several algorithms were ­developed that converted the fluence map to a deliverable MLC pattern, but rendering the fluence pattern practical often required degrading the quality of the IMRT plan, ­occasionally to unacceptable levels. Many current radiation treatment planning systems provide algorithms for direct aperture optimization, determining the MLC settings as the specific output of the inverse planning algorithm.

Fixed-Beam Attenuation IMRT delivered by modifying MLC settings during beam delivery is not modulation of intensity but rather modulation of fluence. True modulation of the radiation beam intensity can be achieved by the use of compensating filters placed in the beam path.16 Calculating the various thicknesses of a compensating filter to generate a desired fluence distribution is relatively straightforward, and the

60

PART 1  Principles

A

C

B

D

E FIGURE 2-12.  There are several steps in the development of a field-in-field plan for treatment of a breast tumor. A, Isodose distribution on a transverse CT slice after beam weights have been adjusted to balance hot spots. B, Treatment portal for the additional medial field, blocking the 114% isodose. C, Isodose distribution with additional medial field added with sufficient weight to eliminate the 114% isodose. D, Treatment portal for the additional lateral field, blocking the 109% isodose. E, Comparison of the original dose distribution (left) and dose distribution for field-in-field treatment (right).

2  Clinical Radiation Oncology Physics

61

thickness information can easily be transferred to a milling machine that can mill that shape of a compensating filter out of an appropriate molding material such as Styrofoam blocks. Fine shot or a low-melting-point alloy can be used to fill the mold. Because the mold itself does not attenuate the radiation beam significantly, it can remain in place during treatment delivery. Although relatively straightforward, this technique has several drawbacks. In addition to the time required to fabricate the compensating filters, a reliable quality assurance procedure is needed to ensure that the filters have been fabricated correctly. Moreover, therapist intervention is required to change filters after each treatment field delivery. Consequently, this technique has been superseded in most institutions by the use of multileaf collimation.

Helical Tomotherapy Helical tomotherapy, a technique developed by Mackie and colleagues,17 combines inverse planning with the delivery of a helical CT–like megavoltage beam. Temporally modulated slits open and close across the beam opening to generate an intensity-modulated treatment beam. Using this beam delivery technique in combination with inverse planning allows the high-dose region to conform to highly irregular arbitrary target volumes. The need to support continuous rotation of the beam delivery gantry required the development of novel slip-ring–based technologies. The tomotherapy beam can also be used to generate megavoltage CT images. As discussed further under “Three-Dimensional In-Room Volumetric Imaging,” images acquired just before treatment can be used as the basis for adjustments in the patient setup and modifications in the treatment plan based on interfractional changes in the patient anatomy, making the possibility of adaptive beam delivery feasible.

Quality Assurance of IntensityModulated Radiation Therapy Delivery Quality assurance of beam delivery is even more important in IMRT than in conventional radiation beam delivery. Because of the need for spatially varying fluence patterns, dose distributions generated from intensity-modulated radiation beam delivery are significantly more complex than those obtained by conventional beam delivery. However, with the increased need for accuracy, the dose-calculation models for IMRT, based on small beamlets, may not be quite as accurate as those for conventional beam delivery. Moreover, because radiation fields are small, the accuracy of dose measurement of these fields may not be as good as those for large fields. Quality assurance of the delivery of IMRT treatments is required in the United States for third-party reimbursement for this procedure. Explicitly stated, the “… plan distribution must be verified for positional accuracy based on dosimetric verification of the intensity map with verification of treatment setup and interpretation of verification methodology.”18 Two components of patient-specific quality assurance of IMRT delivery go beyond the quality assurance needed for conventional conformal radiation therapy. The first is the need to verify that the MLC settings that are designed to deliver the beam fluence that solves the inverse planning problem will deliver the planned dose distribution. The

FIGURE 2-13.  A test phantom is used for quality assurance in intensity-modulated radiation therapy. The yellow sheet is the film on which the dose distribution is recorded for comparison with the planned dose distribution.

second is the need to verify that the calculation of monitor units as obtained from the treatment planning system will deliver the correct dose to the patient. In quality assurance of conventional conformal radiation therapy, commissioning of the beam model and the treatment planning system is sufficient to provide the user with the confidence that the calculated dose distribution is correct,19 and the comparison of the monitor unit calculation with a point dose calculation is routinely used as a quality assurance check on the monitor unit calculation. For IMRT delivery, however, relative dose distribution and absolute dose delivery must be verified with measurement. At M. D. Anderson, the quality assurance procedure for IMRT delivery involves delivery of the planned irradiation to a special IMRT quality assurance test phantom (Fig. 2-13) that contains an ion chamber and film. The film is placed in a transverse plane in the phantom, the ionization chamber is placed in an orifice at the center of the phantom, and the quality assurance procedure is carried out as follows. The fluence distributions and monitor units that comprise the IMRT plan are transferred from the treatment plan to a CT model of the test phantom on the treatment planning computer, and the dose distribution is calculated on the test phantom model. The MLC settings and monitor unit settings, which have been transferred to the recordand-verify system, are used to deliver the IMRT plan to the test phantom. The relative dose distribution indicated on the film is compared with the planned dose distribution, and percent dose difference and distances to agreement are determined (Fig. 2-14). The dose as recorded on the ionization chamber is compared with the calculated dose to the point on the treatment plan corresponding to the location of the ionization chamber and reported on a form. The agreement between calculated and measured dose distributions is evaluated by using the gamma index proposed by Low and colleagues,20 in which the goal is 5% dose

62

PART 1  Principles

Measured 0

5

5

10

10 A/P distance (cm)

A/P distance (cm)

Calculated 0

15 20

15 20

25

25

30

30

35

35

A

0

5

10

15 20 25 Lateral distance (cm)

30

35

B

0

5

10

15 20 25 Lateral distance (cm)

30

35

Calculated (solid) and measured (normalized data, dashed)contours. 0 5

A/P distance (cm)

10 15 20 25 30 35

C

0

5

10

15 20 25 Lateral distance (cm)

30

35

FIGURE 2-14.  Dose distributions using the beam configuration for an intensity-modulated radiation therapy plan are calculated from a test phantom (A) or measured from the same fluence pattern and monitor units on the test phantom (B). In the comparison of the two dose distributions (C), solid lines indicate the calculated dose distribution, and dashed lines indicate the measured dose distribution.

agreement in the regions of low dose gradient and 3 mm distance to agreement in the regions of high dose ­gradient. Another technique for evaluating agreement between calculated and measured dose distributions for IMRT quality assurance has been proposed by Palta and colleagues,21 who divide the dose comparison into four ­regions, each of which has its own separate criteria. Since January 2004, the gamma index method of quality assurance has been used for more than 1000 patients at M. D. Anderson, and acceptable gamma values (<1) have been determined approximately 93% of the time (Karl L. Prado, PhD, personal communication, 2008). In the event

of disagreement, an analysis is performed on a field-by-field basis in which the calculated dose in a plane perpendicular to the beam central axis at depth for each beam is compared with the measured dose distribution to identify the source of the discrepancy.

Stereotactic Radiation Therapy In 1951, Leksell22 introduced the term radiation surgery to describe a technique in which narrow beams are stereo­ tactically focused onto a target within the brain to achieve a high dose of localized irradiation. If the dose gradient is

2  Clinical Radiation Oncology Physics

very steep at the edge of the volume chosen for irradiation, a single dose sufficient to cause necrosis of the selected tissue volume can be administered. In radiation surgery, the most important factor is the physically determined concentration of the radiation on the target; this is based on differences in radiosensitivity between tumor cells and cells of the adjacent healthy tissue. Frame-based stereotactic radiosurgery (SRS) for the treatment of benign intracranial lesions has become widely accepted, and it produces excellent long-term outcomes and minimal toxicity.23 SRS has also become one of the major modalities for managing brain metastases. Three general approaches to intracranial SRS are used incurrent practice.24-27 The first uses positively charged particles such as protons or helium ions generated by large accelerators originally built for nuclear physics research, but only a few of these accelerators are in hospital-based facilities. The second approach, popularly referred to as the gamma knife, makes use of the gamma radiation emitted from a fixed array of small cobalt 60 (60Co) sources located within a large hemisphere that surrounds the patient’s head. The third approach uses high-energy x-ray radiation produced by a medical linear accelerator. The radiation oncology ­departments of most hospitals have linear accelerators that can be adapted for use in SRS.

Image-Guided Stereotactic Radiosurgery or Radiation Therapy for Cranial Tumors At M. D. Anderson, accelerator-based SRS has been under development since late 1989, and the first patient was treated with SRS in August 1991. The current SRS system (Fig. 2-15) consists of variable collimation apertures that are inserted at the end of a circular collimator, a table-mounted docking device; a laser target-localization frame is attached to the docking device. The acceleratorbased SRS system delivers a lethal, homogeneous dose

63

to an ap­proximately spherical treatment volume, which is sur­rounded by a region of rapid fall-off in dose. This form of treatment is best suited for small, spherical tumors.28 Tumors that are more irregularly shaped are treated with larger circular collimators. Because these larger collimators expose a greater proportion of healthy brain to high-dose irradiation, the doses that are to be delivered with large collimators are lower than those to be delivered with the smaller collimators; for this reason, tumors larger than about 4 cm in diameter should not be treated with SRS. Although the frame-based SRS is recognized as one of the most precise setups for aiming a narrow radiation beam to the target volume, a downside of this procedure is that the fixation of the head frame to the patient’s skull is invasive and painful. The patient also has to wear the head frame until the treatment is delivered, an interval that can be as long as several hours. Advances in imaging technology and treatment delivery systems have enabled the radiologists to develop a noninvasive intracranial SRS technique that has the same high degree of accuracy as that of the current, invasive head-frame SRS technique. This new technique involves use of image registration to correlate the daily CT images with the planning images and use of a Gill-Thomas-Cosman frame with a bite block assembly (Fig. 2-16) that allows the target isocenter to be coincident with the accelerator isocenter through the stereotactic frame coordinate system. A head phantom was first used to evaluate the feasibility of this method. Then, image-guided SRS was used for fractionated stereotactic radiation ­ therapy (SRT) to further demonstrate that this technique meets or exceeds the accuracy required for single-fraction SRS.

Image-Guided Stereotactic Radiation Therapy for Extracranial Spinal Tumors Palliative treatment of spinal bone metastases comprises a significant portion of the financial expenditures associated with cancer care. Radiation therapy is highly effective in relieving painful bone metastases; almost 90% of patients experience improvement, and about one half experience

Variable collimation apertures

Laser target localization frame

Table mounted docking device

FIGURE 2-15.  The linear accelerator–based stereotactic ­system is used at the M. D. Anderson Cancer Center.

FIGURE 2-16.  The patient setup is shown for stereotactic ­radiation therapy in a Gill-Thomas-Cosman head frame.

64

PART 1  Principles

complete pain relief.29 Unlike curative therapy, which uses high total doses of radiation to achieve tumor control, palliative therapy typically uses lower total doses of radiation given in a shorter schedule in fewer fractions. The large, collective experience of the Radiation Therapy Oncology Group indicated that low-dose, short-course treatment schedules were as effective as high-dose, protracted treatment programs.30 However, radiation oncologists are often called on to consider reirradiation for patients who have already received a palliative course of radiation treatment.30 Unfortunately, treatment options available for patients in this situation with spine metastases usually are limited to surgery and narcotics because reirradiation by conventional techniques carries a high risk of spinal cord damage and exposes excessive amounts of normal tissue to unnecessary radiation. Given the risks ­ associated with surgery and anesthesia and the time necessary for rehabilitation and recovery, we and others at M. D. Anderson developed a noninvasive alternative in which radiation can be precisely directed to spinal or paraspinal tumors and safely repeated if necessary.31 The key features of this technique are its use of true stereotactic localization methods enhanced by near-simultaneous CT image guidance and computerized imaging registration to tightly link patient anatomy to a stereotactic body-frame system. Our current system for treating spinal and paraspinal lesions comprises a stereotactic body-frame system (­BodyFix, Medical Intelligence, Schwabmünchen, Germany) that consists of a carbon fiber base plate, whole-body vacuum cushion, vacuum system, and plastic fixation sheet; the stereotactic localizer and target-positioning frame are from Tyco/Radionics (Burlington, MA). Treatment plans are generated with inverse planning IMRT software (Pinnacle,3 Philips Medical Systems, Milpitas, CA). Pretreatment CT imaging and IMRT treatment delivery are done with a targeting system that integrates a CT-on-rails scanner (Hi­Speed X/i, GE Medical Systems, Milwaukee, WI) that shares the same patient support assembly with a linear accelerator (21EX, Varian Medical Systems, Palo Alto, CA) equipped with a 120-leaf MLC (Fig. 2-17A). In March 2005, the original couch top was replaced with a sixdegrees-of-­freedom robotic couch top (see Fig. 2-17B) that allowed the stereotactic body frame system to be attached directly to the tabletop. The lack of a robust and direct spatial link between the patient anatomy and the localization frame in this system precluded our maintaining the necessary setup accuracy (no more than 2 mm) between treatments.32 This shortcoming led us to develop mutual information imaging registration software that directly compares a specified point of interest from a reference imaging set with that on the imaging sets that are obtained before each treatment session. In its current iteration, the registration process with this system takes only 2 to 3 minutes. Pretreatment CT scans are obtained, and SRT is delivered with the patient set up in the stereotactic body immobilization system on the accelerator/CT-on-rails unit. The pretreatment CT scans are obtained daily at the same slice thickness (2 or 1 mm) as that of the planning CT scans. The rotation of the isocenter point from the planning CT to the daily CT is determined by comparing the anteroposterior and right lateral digitally reconstructed radiographs from the daily CT scans with the planning radiographs on split

screens to test the accuracy of the alignment near the target areas, including the isocenter. The updated daily treatment isocenter is then input into the six-degrees-of-freedom robotic couch-top tracking and positioning software. The alignment function of the software allows the treatment isocenter to be positioned automatically and accurately based on the three infrared light–emitting ­ diode cameras and five reflected markers on the top of the body-position frame (Fig 2-18). The couch top is capable of correcting shifts and rotations in the roll, yaw, and pitch directions of up to ± 3 degrees. Our tolerance goal for the alignment accuracy between the treatment isocenter and the planned isocenter was within 1 mm or less. We published our findings on the relative accuracy of this system, which was determined by quantitatively comparing the distance of the daily treatment isocenter from the planning isocenter on portal film images, planning ­images, daily digitally reconstructed radiographs, and ­immediate posttreatment verification CT scans. Among 90 treatments delivered to 15 patients with this system, the daily setup deviation from the planned isocenter according to the pretreatment CT scans was unacceptably high, ranging from −6.0 mm to 4.9 mm in the lateral direction (mean,

Plastic fixation sheet

MLC

Stereotactic localizer Whole body vacuum cushion Carbon fiber base plate

A

Vacuum system

6D robotic couch top

B FIGURE 2-17.  A, CT-on-rails (middle left background) and linear accelerator (top right foreground) share the same patient support assembly. The patient is on a vacuum bag (purple) in a stereotactic body frame. B, CT-on-rails (left) and a linear accelerator assembly (top right) with a new six-degrees-of-freedom robotic tabletop (middle foreground).

2  Clinical Radiation Oncology Physics

65

DEVIATION FROM PLANNED ISOCENTER USING DAILY PRE-TREATMENT CT

Deviation (mm)

10.00 5.00 0.00 –5.00 –10.00 0

20

40

60

80

100

Treatment number LAT

A FIGURE 2-18.  Alignment of the isocenter for treating a patient with a paraspinal tumor by using five reflective markers on the body-positioning frame.

AP

SI

ASSESSMENT OF SETUP ACCURACY USING INTERMEDIATE POST-TREATMENT CT

0.19 mm; SD, 2.69 mm), from −5.5 mm to 5.0 mm in the anteroposterior direction (mean, −0.4; SD, 1.8 mm), and from −8.0 mm to 4.5 mm in the superoinferior direction (mean, −1.58; SD, 3.09 mm) (Fig. 2-19A). These results demonstrated that use of the stereotactic body frame alone, without the mutual registration program, cannot achieve the setup accuracy necessary for delivering a highly conformal dose of radiation to the target. However, with the nearreal-time CT image guidance provided by this program, the patient is aligned with the corrected daily isocenter, and the overall deviation from the planning isocenter for each treatment was within 1 mm (see Fig. 2-19B). These findings support the ability of our SRT system to accurately treat spinal and paraspinal tumors.

Image-Guided Stereotactic Radiation Therapy for Thoracic Tumors At M. D. Anderson, SRT, also referred to as hypofractionated image-guided radiation therapy, is also used to treat certain types of lung tumors. The rationale is that the dose needed to control lung cancer seems to be much greater than the 60 to 66 Gy delivered through standard 2-Gy fractionation schemes.33 Conventional fractionation is not practical because of normal tissue tolerances34; consequently, SRT, with its smaller fields and reduced margins for setup un­ certainty, is thought to be a feasible alternative. SRT may be particularly appropriate for patients with medically inoperable stage I or stage II lung tumors or for those with more advanced disease who may not be able to tolerate standard fractionation schemes because of mental condi­ tion, age, or other factors. As is true for cranial tumors, nodal disease or tumors larger than 4 cm generally should not be treated with SRT.

Fractionation The standard fractionation scheme used at M. D. Anderson for SRT for thoracic tumors is 50 Gy in 4 fractions, given on consecutive days, for primary tumors, and 40 Gy in 4 fractions, given on consecutive days, for recurrent tumors.35 Doses are prescribed and treatment portals designed in a manner analogous to dose prescriptions for conventionally fractionated radiation therapy, in that 95% of the PTV is targeted to receive the prescription dose.

Deviation (mm)

10.00 5.00 0.00 –5.00 –10.00 0

20

40

60

80

100

Treatment number

B

LAT

AP

SI

FIGURE 2-19.  Deviations of the treatment isocenter from the planned isocenter for 90 stereotactic body treatments are ­determined from CT images obtained before treatment (A) or immediately after treatment (B). AP, anteroposterior; LAT, ­lateral; SI, superoinferior.

Immobilization Even with image guidance, patient immobilization is an essential component of this procedure for several reasons. Patient motion must be minimized during the interval between daily imaging and treatment (approximately 5 to 10 minutes). The immobilization device must allow placement of imageable fiducials, because a stereotactic body frame is not used for treating thoracic tumors. The combination of immobilization and daily image guidance reduces the margins for setup uncertainty and interfrac­ tional variation to 3 mm. Several aspects of the immobilization process are important to consider. Immobilization and localization need not be linked; patients can be immobilized, locked in place, and then imaged. The immobilization process should ­allow the patient to hold a treatment position for 30 to 45 minutes, which is the typical maximum duration of an ­image-guided treatment session. Because image guidance is effected by the use of CT imaging before treatment, the immobilization method must allow the patient and immobilizers to fit through the bore of the CT scanner. If imageable external fiducials are used to aid in localization, the immobilization method should provide stable locations for these fiducials. Because beam delivery angles in SRT can be rather complex, care should be taken to ensure that the immobilization device does not interfere with the beam delivery

66

PART 1  Principles

FIGURE 2-20.  A vacuum bag is used for patient immobilization during hypofractionated treatments of lung lesions.

angles. Because SRT delivery can result in somewhat higher skin doses than conventional beam delivery, the immobilization method should not exacerbate this effect. The immobilization device used for SRT at M. D. ­Anderson is an extended vacuum bag (Medical Intelligence) (Fig. 2-20) that resembles conventional vacuum immobilization bags but extends from the patient’s head to the waist and can be fixed to a particular position on the table with registration tabs. Large amounts of material are added to the sides of the bag to help stabilize and ­immobilize the patient. The bag also allows a vacuum sheet to be placed on the anterior patient surface for additional immobilization, but we do not consider this necessary. The immobilization device is customized for each ­patient. The patient lies on top of the bag, which is then drawn by vacuum around the patient, more tightly than is done with conventional vacuum bags, to completely ­immobilize the patient. While the vacuum is being drawn, several staff members push on the bag around the patient to ensure a continually snug fit and to keep the bag sides flat and vertical so that they provide a stable position for the placement of external fiducials at the approximate vertical height of the center of the tumor. After the bag has been formed, the patient steps out of the bag and off the table. The patient is then set up again within the bag so that the patient’s position during the CT image acquisition is more representative of that of daily setup.

Treatment Planning Imaging for treatment planning is done with a multislice helical CT scanner (Discovery ST, General Electric Health­ care, Waukesha, WI). After the patient is placed in the immobilization bag on the table and a vacuum is partially drawn, images are obtained to determine the approximate tumor location. Based on images from a small number of transverse slices acquired in the vicinity of the tumor, the in-room lasers are moved so that they indicate a region close to the center of the tumor. After the lasers have been positioned, the full vacuum is drawn on the immobiliza­ tion bag. A fast helical scan is obtained under free-breathing conditions encompassing a region from slightly inferior to the orbits to the base of the liver. An extended region is imaged

to obtain a more complete digital model of the patient, ­ ecause non-coplanar beams are often used in the treatb ment plan. A four-dimensional scan36 is then obtained of the region around the tumor to determine the extent of ­tumor motion, which assists in targeting the tumor and surrounding critical structures. Because tumors treated by SRT often are completely surrounded by lower-density lung parenchyma, a maximum intensity projection (MIP) image, collected in addition to the 10 phases that constitute the four-dimensional data set, provides a good representation of the extent of tumor motion during respiration. If overlap is observed between the target volume and a critical structure, treatment with the patient under deep-inspiration breath hold37 may be necessary to ensure separation between the target volume and critical structure. If the patient is capable of deep-inspiration breath hold, we acquire several short helical scans near the tumor ­under that breath-hold condition and evaluate the residual ­motion and reproducibility of the tumor between breath holds. Patients are given visual feedback using virtual-­reality goggles to identify the target level for the breath hold and to assist in maintaining that breath hold. After the GTV has been delineated on the free-­breathing data set, based on the four-dimensional phases or the deepinspiration breath-hold images, it is expanded by 8 mm to generate a CTV.3 For the four-dimensional CT images, the CTVs generated over all phases of respiration are combined to generate the ITV; for the deep-­inspiration breathholds images, the CTV constitutes the ITV. An ­additional 3-mm margin is added for setup uncertainty to generate the PTV. When treatment portals are designed for thoracic SRT, the critical structures identified (i.e., lungs, spinal cord, esophagus, and heart) are the same as those for ­conventional radiation therapy. Because of the effects of hypofractionation, the major vessels and bronchial tree are also viewed as critical anatomic structures, as is the brachial plexus for upper lobe tumors. Table 2-2 lists the critical structures and the dose-volume constraints for treatment planning. Typically six to nine non-coplanar beams are used to ­irradiate the PTV. Non-coplanar beams are used so that the beams do not enter or exit through the contralateral ­(uninvolved) lung, thereby completely sparing it. Moreover, use of beams entering the patient from a more superior or ­inferior direction minimizes the effects of superoinferior respiratory-induced motion of the tumor, which tends to be the largest component of respiratory-induced motion. Use of a large number of beams is viewed as desirable because it improves the geometric fall-off of dose outside the target volume. Because large doses per fraction tend to increase the possibility of skin reactions, skin overlap of beams is avoided as much as possible. The plan should also be reviewed to ensure that the gantry does not collide with the patient. Photon beams with energies greater than 6 MV should be used with caution because of the loss of electronic equilibrium, which is more prevalent with ­ higher energy beams; high-energy (e.g., 18 MV) beams should be used only for radiation fields in which the tumor is distal to unit-density tissue. If the beam must pass through lung tissue, for example, before entering the tumor, the 18-MV beam should not be used. Even if only 6-MV beams are used, the issue of electronic disequilibrium in the vicinity of the tumor-lung interface remains an issue.

2  Clinical Radiation Oncology Physics

TABLE 2-2.   Dose-Volume Constraints for Stereotactic (Hypofractionated) Radiation Therapy for Lung Tumors Anatomic Structure

Dose

Volume

Total lung

20 Gy

≤20%

Total lung

10 Gy

≤30%

Total lung

5 Gy

≤40%

Cord

25 Gy

Max dose

Esophagus

40 Gy

≤1 cm3

Esophagus

36 Gy

≤10 cm3

Trachea

40 Gy

≤1 cm3

Trachea

36 Gy

≤10 cm3

Main bronchus

48 Gy

≤1 cm3

Main bronchus

40 Gy

≤10 cm3

Heart

48 Gy

≤1 cm3

Heart

40 Gy

≤10 cm3

Major vessels

48 Gy

≤1 cm3

Major vessels

40 Gy

≤10 cm3

Brachial plexus

40 Gy

≤1 cm3

Brachial plexus

35 Gy

≤10 cm3

Skin

40 Gy

≤1 cm3

Skin

35 Gy

≤10 cm3

Total dose is 50 Gy in four fractions. Data courtesy of J. Y. Chang, MD, PhD, M. D. Anderson ­ Cancer Center, Houston, TX.

Treatment planning for hypofractionated radiation therapy introduces some issues that are sometimes passed over in conventional treatment planning. The dose calculation algorithm used in the treatment planning system should be evaluated for its ability to predict dose to a small mass of tissue in a situation where electronic equilibrium has not been achieved. Although some modern treatment planning systems model the dose buildup empirically, the parameters used to model the buildup should be validated. Moreover, because only a few high-dose fractions are used, tissue responses based on conventional fractionation schemes may not be valid. Consequently, models of dose response, such as those that generate biologically equivalent doses, may be more appropriate for evaluating hypofractionated treatment plans.38

Treatment Delivery Delivery of SRT should be combined with some sort of real-time or almost real-time imaging. At M. D. Anderson, most of the thoracic SRT is delivered by the same linear accelerator combined with an in-room CT scanner, as is used for paraspinal SRT. The linear accelerator and CT scanner share the same patient support table, with lasers placed in the room to generate a consistent isocenter. In this system, the patient does not move through the CT scanner, but rather the CT scanner is mounted on a set of rails so that it moves past the patient (see Fig. 2-17A). Daily imaging is done on the CT scanner immediately before treatment.

67

The CT images obtained in this way are compared with the treatment planning CT image, and adjustments in the patient setup are made. After a confirmation CT scan is acquired, the table is rotated 180 degrees, placing the patient in the appropriate position for treatment from the linear accelerator. Other methods of image guidance can be used. A conventional CT scanner can be used if the patient is immobilized in a body frame that ensures no motion between scan acquisition and treatment delivery. Another image-guided technique perhaps more common than CT-on-rails involves use of a linear accelerator and kilovoltage or megavoltage cone-beam CT. Multiple fixed-position kilovoltage imaging units can be used in combination with a linear accelerator (e.g., Novalis Shaped Beam Surgery, BrainLAB, Feldkirchen, Germany) or with an accelerator with multiple degrees of freedom attached to a robotic arm (e.g., CyberKnife ­Robotic Radiosurgery System, Accuray, Sunnyvale CA). Regardless of which image guidance system is used, ­imaging should be done daily to accurately target the ­tumor. Ideally, the imaging is done in the treatment room on the treatment couch, but if the imaging is done in another room in a body frame, it is important that measurements be taken in the treatment room to ensure that the patient has not moved between imaging and treatment. Even better is imaging done during treatment delivery, which can be done with some fixed-position kilovoltage imaging systems, especially if done in combination with implanted fiducials that allow tracking of the target.

Accounting for Motion in Radiation Therapy The motion that must be accounted for in the planning and delivery of radiation therapy is of two kinds— intrafractional and interfractional. Intrafractional motion occurs during delivery of a fraction of radiation therapy; it is typically caused by physiologic movement and can be voluntary or involuntary. Examples of intrafractional motion include respiratory motion, cardiac motion, or peristalsis. Interfractional motion refers to motion or, perhaps more accurately, variation during the time between fractions of radiation therapy. Examples of interfractional variations include tumor regression, bladder filling, rectal filling, and weight loss. The next section reviews how motion can be measured and the steps that can be taken to reduce the size of the margins used to account for that motion.

Intrafractional Motion: RespiratoryCorrelated Radiation Therapy Respiratory motion has been the best-studied type of intrafractional motion, and much work has been done to measure and mitigate its effects. Some of the discussion that follows also is applicable to other forms of intrafractional motion. Respiration causes tumors in the lung and abdomen to move up to 2 cm or more, with the typical distance being about 0.5 to 1 cm.39-42 As described under “Target Volume Delineation,” the ITV represents the margin around the CTV that accounts for that motion; the PTV represents the ITV plus a margin to account for variations and uncertainties in the patient setup. Traditionally, the CTV was

68

PART 1  Principles

expanded uniformly; for lung tumors, the CTV-to-PTV expansion was typically in the range of 2.0 to 2.5 cm. The two assumptions inherent in this expansion were that the magnitude of tumor motion was consistent from patient to patient and that the tumor motion was isotropic (i.e., identical in all directions). Both assumptions are false, invalidating the concept of a “one size fits all,” population-based CTV expansion. Studies in which fiducials implanted in the vicinity of lung tumor were tracked have demonstrated that tumor motion is best represented by a highly eccentric ellipse with a curved major axis.43

Measuring Respiratory Motion The development of four-dimensional CT imaging has enabled the explicit measurement of respiratory motion. This technique involves acquiring phase-dependent CT image data sets based on periodic patient motion, such as respiratory motion. The motion is phase-dependent rather than time-dependent, because the general approach in four-dimensional CT imaging is to acquire a limited set of image data at specific phases during the respiratory cycle. The information is acquired from an anatomic thickness equal to the width of a multislice CT detector, typically 2.5 to 4.0 cm. Image data at the same phase in several respiratory cycles are combined to generate a full threedimensional image data set at each phase. Two approaches have been developed to acquire four­dimensional CT images; both require a multislice CT scanner and a respiratory monitor. Four-dimensional CT ­images can be acquired with a single-slice CT scanner, but the process is much more difficult. Of the various types of respiratory monitors available, one in common use is the RPM (Varian Medical Systems, Palo Alto, CA), which consists of a block with two reflective marks that is placed on the patient’s ­abdomen and an infrared light source. Light is reflected from the fiducial block and detected by a charge coupled device detector (Fig. 2-21). The position of the reflector as a function of time is stored as the respiratory signal. Other devices used to monitor the respiratory cycle include spirometers, which measure changes in airflow, and strain gauges, which measure changes in abdominal ­circumference. Regardless of the type of monitor, the end result is a data set containing information about position of the tumor during the respiratory cycle as a function of time. In one approach to acquiring four-dimensional CT ­images (image binning),36 the CT scanner is operated in cine mode. Projections are acquired during an interval at least as long as that of one respiratory cycle plus one gantry rotation. The table on which the patient lies is then indexed a distance equal to the width of the detector, and more projections are acquired. Images are reconstructed from the projections at specified time intervals (typically 0.25 to 0.50 seconds). Typically, approximately 2000 to 3000 axial images are reconstructed. Each image is identified with an acquisition time, and the acquisition times of the images are correlated to the times on the respiratory signal. Images are then binned based on the phase of the respiratory cycle, which is extracted from the respiratory signal. In a second approach (projection binning),44,45 the multislice helical CT scanner is operated at a very low pitch, which results in a file with a very large number of projections. The projections are tagged at a specified point in each

FIGURE 2-21.  The infrared camera, charge-coupled device detector, and video screen are used for monitoring respiratory motion.

respiratory cycle, typically at end ­ inspiration (i.e., 0% phase). The respiratory monitor provides the tag information. After projections are acquired, phases are determined for reconstruction, and projections acquired in the vicinity of the specified phases are binned before reconstruction. In other words, image binning reconstructs the images and then bins them, whereas projection binning bins the ­projections and then reconstructs the images. Both techniques have been successfully used at M. D. Anderson to obtain four-dimensional CT image data sets, which can then be used to explicitly delineate the ITV. Ideally, the GTV would be explicitly drawn on each phase of the data set and then expanded by the CTV ­margin on each phase. For non–small cell lung tumors, our current practice is to expand the GTV isotropically by 0.8 cm ­according to the pathology-based recommendations of ­Giraud and others.3 The CTVs are edited in each phase to remove regions in which no microscopic disease is thought to be present, such as into the chest wall or across fissures. The CTVs on all phases are then combined to ­generate the ITV. Because this technique is very time-consuming, measures have been proposed to facilitate ITV generation. One such method is to generate an MIP of the four-dimensional CT image data set, which involves taking the maximum CT voxel value of each corresponding voxel out of all of the phases in the data set. This approach is based on the observation that lung tumors tend to be more dense the surrounding parenchyma. An MIP-GTV is delineated on the MIP data set and then expanded to generate the ITV. This technique breaks down, however, when the tumor is near unit-density structures, such as mediastinal structures, that may interfere with the generation of the MIP. Moreover, the MIP-based technique does not account for the CTV being edited to exclude regions thought to lack microscopic spread. Consequently, an ITV solely based on the MIP-GTV, if no interfering structures are present, is a maximum approximation of the true ITV. In contrast to lung tumors, tumors in the liver tend to be less dense than the surrounding parenchyma, and a minimum intensity projection (sometimes called a least intensity projection) can be used instead of the MIP.

2  Clinical Radiation Oncology Physics

Another technique for generating an ITV is to manually delineate the GTV on a single phase of the image data set and then propagate the GTV from that one phase to all of the other phases. Some success has been achieved by using rigid body transformation46 or deformable image registration�47 for GTV propagation. Deformable image registration is discussed further under “Image-Guided Adaptive Radiation Therapy.” After the ITV has been generated, an additional margin is added to the ITV to produce the PTV. For lung tumors, we use a margin of 1.0 to 0.7 cm to account for setup uncertainty plus 0.3 cm to account for motion uncertainty. In the absence of four-dimensional CT, an ITV can be generated if CT images are acquired during breath hold at end inspiration and end expiration. The breath hold must be done under normal breathing conditions and not ­under deep inspiration or deep expiration. This technique should be used with care, however, because CT data sets acquired under breath-hold conditions may not accurately represent CT data sets at end phases under free-breathing ­conditions.48,49

Mitigating the Effects of Respiratory Motion: Gated Beam Delivery One technique for reducing the effects of respiratory motion is gated beam delivery. In this procedure, the delivery of radiation is synchronized with a specified condition of a periodic cycle (e.g., the respiratory cycle). Two methods of gating are used. In amplitude gating (i.e., displacement gating), radiation delivery is gated based on the actual position of the tumor or, more precisely, the tumor surrogate; in phase gating, the delivery of radiation is based on the phase of the respiratory cycle. Amplitude gating is easier to achieve, because a signal can be sent to the linear accelerator in real time based on the position of the respiratory monitor moving past a specified gating point. Phase gating is some­ what more difficult because generating a signal to the linear accelerator in real time requires predicting when a specified phase point will occur. This can be achieved by the use of a predictive algorithm in the respiratory monitor. Any type of respiratory monitor can be used to generate a gating signal; the most commonly used is the external fiducial. Shirato and colleagues50 have had substantial experience with the use of small gold fiducials implanted in the vicinity of lung tumors. Gating based on free breathing is typically done at end expiration because that seems to be the most reproducible point in the respiratory cycle; gating can also be effected in combination with breath holds at deep inspiration, end ­inspiration, and end expiration. Gated beam delivery can effectively reduce margins by reducing intrafractional ­motion, which results in the possibility of more lung sparing51 or higher tumor doses.52 Gated delivery also reduces the amount of lung tissue moving into and out of the ­radiation field, but the quantitative effect of this reduction on the dose distribution has yet to be determined. Gating has been shown to improve lung sparing compared with explicit delineation of the ITV,53 but only for small (<100 cm3) tumors with significant movement (>1 cm) and no residual motion. Studies of the movement of lung tumors marked with implanted fiducials during gated beam delivery have shown that residual motion is still significant,54 as illustrated in Figure 2-22. The potential does

69

FIGURE 2-22.  Residual motion of fiducials implanted in the ­vicinity of a lung tumor is determined by cine portal imaging. The red line indicates the extent of residual motion of the ­fiducials. (Courtesy of C Nelson, M. D. Anderson Cancer Center, Houston, TX.)

e­ xist, however, for reducing residual tumor motion by using feedback-guided breath holding, in which a display of the respiratory cycle is provided to the patient and the patient is trained to effect a breath hold based on the display.55

Interfractional Motion: Image-Guided Radiation Therapy Many studies have demonstrated that significant setup uncertainties and internal organ variations are present in external beam radiation therapy.56-63 The major reason for these uncertainties is the use of external (skin) marks for positioning, because the marks do not accurately represent the positions of the internal target or the surrounding normal tissue structures.64 Chandra and colleagues65 at M. D. Anderson reported the use of daily abdominal sonography to assess the position of the prostate relative to the initial skin marks for treatment setup. Data from 3328 images of 147 patients showed that the prostate shifted by as much as 2 cm from one fraction to another. Typical values of interfractional and intrafractional organ variations and setup uncertainties for gynecologic tumors, lung tumors, and movement of liver, diaphragm, prostate, seminal vesicles, bladder, and rectum have been reviewed by Langen and Jones58 and Booth and Zavgorodni.66 Both groups found significant variations for all organs studied. Even with careful immobilization and alignment of the patient, significant interfractional changes can occur because of anatomic nonrigidity, movement of gas in the bowel, and variable filling of the bladder.67-69 Even for relatively rigid anatomic sites such as in the head and neck region, soft tissues can change because of tumor shrinkage, weight loss, or other reasons.70 If not adequately accounted for, such changes may lead to the delivery of dose distributions that are significantly different from those that were planned. Advances in three-dimensional conformal radiation therapy71 and IMRT72,73 allow the generation of dose distributions that more precisely target tumor volumes and avoid normal tissues to a greater extent than was possible

70

PART 1  Principles

A

B

FIGURE 2-23.  Portal imaging is used for treatment verification. A, A lateral head-and-neck treatment field is acquired using a double-exposure technique. The shadowed area indicates the shape and location of the treatment area (shaped by a low-­meltingpoint alloy block) overlaid on the picture of the patient’s anatomy. B, The same portal film is digitally enhanced by using an adaptive histogram equalization technique.

with older techniques. However, even with these newer techniques, setup uncertainties and organ motion are usually accounted for by using wide margins around the target volume.63,74,75 Treatments based on wide margins may increase treatment-related toxicity or may inhibit the delivery of a radiation dose adequate to sterilize the disease. Moreover, highly conformal radiation therapy techniques such as IMRT are more susceptible to anatomic changes than three-dimensional conformal radiation therapy,76-78 and concern has been expressed that highly conformal, high-dose IMRT dose distributions designed based on a single CT data set (and, in the case of lung cancer, under free-breathing conditions) acquired for planning ­purposes may lead to unforeseen complications or to marginal misses of target volumes. This is one reason why imageguided radiation therapy is so important when conformal radiation therapy is used for cancer treatment, because significant underdosing or overdosing could happen if the ­target or the critical structures are not within the designed ­treatment margins.

Evolution of In-Room Imaging for Radiation Therapy In its original form, in-room imaging for verification of radiation treatment was done with pretreatment radio­ graphy.79 Portal films, exposed to the high-energy mega­ voltage x-rays used for therapy, are still the most commonly used method for radiation therapy quality assurance. In the late 1980s, the invention of various digital electronic portal imaging devices80-83 stimulated wide interest in the use of megavoltage digital x-ray images for treatment verification. The two major goals in portal imaging are to verify the treatment ports (i.e., shape and accuracy of the beam shaping device) and the patient’s anatomy relative to those ports. The portal films shown in Figure 2-23 were acquired using a double-exposure technique in which a large ­ anatomic

area is irradiated by opening the jaws of the collimator, after which the beam-shaping block used for treatment is inserted and irradiated with a small additional dose. The resulting portal image is an overlay of the treatment field (or portal) relative to the anatomy of the patient in the irradiated area. Physicians use portal images to verify that the treatment field was correctly manufactured or inserted into the field and that the patient was aligned accurately relative to the treatment field. This procedure remains an important part of quality assurance in radiation therapy, although motordriven MLCs are replacing the traditional custom blocks in many modern machines. Weekly portal filming still is a standard practice in many hospitals around the world. Unfortunately, given the physics of the interaction of megavoltage x-rays with matter, megavoltage portal images are of poor image quality for visualizing patient anatomy (see Fig. 2-23). Many factors contribute to this poor image quality, including the reduced differential attenuation between the bone and soft tissue with high-energy x-rays and the extra scattered radiation through the patient’s body. The dose from megavoltage imaging is not negligible; the typical radiation dose required to produce portal films of acceptable quality is 3 to 5 cGy. Digital imaging and improved detector efficiency can reduce this dose to 1 to 2 cGy per portal image. This exposure in addition to the need for multiple treatment ports for verification limits this form of portal imaging to weekly quality assurance instead of daily use for treatment verification or patient setup. The introduction of IMRT into routine clinical practice complicates the verification of treatment portals. With a segmental (i.e., “step-and-shoot”) IMRT field, multiple MLC-collimated treatment fields are used to modulate the intensity distribution in the beam’s eye view port. An IMRT field for a prostate cancer treatment at gantry angles of 225 degrees (right posterior oblique field) and 65 degrees (left anterior oblique field) is illustrated in Figure 2-24A and B. In addition to the difficulty of evaluating the shades of beam intensity for the IMRT field, the ­obliqueness of the

2  Clinical Radiation Oncology Physics

beam angle requires additional effort to judge the correct anatomy for the viewing angle. The current practice is to separate the quality assurance of the treatment field from the alignment of the patient’s anatomy. In most clinics, orthogonal x-ray images are acquired separately to verify if the patient is aligned correctly relative to the isocenter of the treatment machine (see Fig. 2-24C and D). In the example shown in Figure 2-24, a straight anteroposterior open field (at a gantry angle of 0 degrees in Fig. 2-24C) and a lateral open field (at 270 degrees [Fig. 2-24D] or at 90 degrees [not shown]) were used as a reference to verify if the patient was aligned accurately to the ­isocenter with

A

71

minimal rotation (i.e., body twist). Reference ­reticules (i.e., cross-hairs) are typically indicated on each image.

On-Board Kilovoltage X-Rays Other forms of two-dimensional imaging include onboard kilovoltage x-ray imaging, in which a kilovoltage x-ray tube is combined with a flat panel image detector mounted orthogonal to the therapeutic x-ray beam on a linear accelerator (Fig. 2-25). Imaging systems such as this provide diagnostic-quality x-ray images that can be used for patient setup; the images obtained are aligned with the refer­ ence digitally reconstructed radiographs computed from

B

C

D FIGURE 2-24.  Beam’s eye views of two intensity-modulated radiation therapy fields of a man with prostate cancer were obtained at gantry angles of 225 degrees (A) and 65 degrees (B). The modulated beam intensity was overlaid on the anatomy in the oblique view. Different radiation intensities were shown as different shades of red. Two orthogonal x-ray views of the patient were obtained at gantry angles of 0 degrees (C) and 270 degrees (D). In each beam’s eye view, the isocenter position is indicated by the intersection of the graticule.

72

PART 1  Principles

the treatment planning systems. In-room kilovoltage x-ray imaging represents a major upgrade from the traditional megavoltage x-ray imaging because of its superior image quality for visualizing bony structures (compare Fig. 2-26A and B) and its very low imaging dose (typically less than 0.05 cGy per kilovolt image). The kilovolt x-ray images also resemble traditional simulation images and digitally reconstructed radiographs. Most kilovoltage x-ray imaging systems also have a companion fluoroscopic mode, in which a continuous stream of kilovoltage images can be used to track the movement of implanted fiducial markers and possibly high-density targets in regions of low-density background (e.g., lung). Kilovoltage imaging represents a major shift away from using megavoltage imaging for beam’s eye port verification and toward using target localization or isocenter verification for patient setup. The orthogonal mounting of kilovoltage imaging systems, such as the one shown in

Robotic arm

KV X-ray source

Flat-panel Xray detector

FIGURE 2-25.  An integrated imaging and treatment unit ­(Trilogy, Varian Oncology Systems, Palo Alto, CA) is used to treat a patient with head and neck disease. The system has a kV x-ray imaging system orthogonal to the therapy beam line and can provide two-dimensional radiographic, fluoroscopic, and cone-beam CT modes for patient setup.

A

Figure 2-25, does not incorporate the beam-shaping device used for the treatment. With the introduction of on-board kilovoltage imaging, pretreatment stereoscopic imaging for patient setup, independent of the verification of treatment fields, is gradually becoming an accepted practice.

Sonographically Guided Prostate Localization Another noninvasive means of effectively imaging soft tissue targets in real time is sonography. The current clinical practice for prostate target localization at many institutions includes the use of B-mode transabdominal sonography.84-86 The ultrasound transducer, which is made of piezoelectric crystal, is a sound source and a sound detector. The transducer transmits brief pulses of ultrasound that propagate through the tissues. Changes in tissue density or elasticity at the interface of two tissue structures result in some of the ultrasound bouncing back to the transducer as echoes. The round-trip time from the pulse transmission to the reception of an echo is used to determine the transducerto-object distance. A scan line converter then constructs a two-dimensional image of the patient in which the amplitude of echoes varies along the depth direction. When the ultrasound waves sweep through a volume, a threedimensional image can be re­constructed. Because ultrasound echoes can be distorted by the presence of bones and air in the path, its use for soft-tissue imaging is limited primarily to tumors in abdominal and breast locations. Reviews of sonographic guidance for IMRT have been provided by Kuban and colleagues87 and Fung and colleagues.88 One image guidance system in wide use for IMRT of prostate cancer, the BAT system (NOMOS/North America Scientific, Chatsworth, CA), is shown in Figure 2-27. With regard to target localization, sonography can be used for direct targeting of soft tissues, whereas the traditional radiographic methods allow only indirect targeting through alignment of bony structures. Little and colleagues89 used both methods to estimate prostate movement in 35 patients undergoing IMRT for prostate cancer (243 pairs of orthogonal portal films plus sonographic measurements obtained on the same days). Their findings revealed that the mean extent of prostate ­motion

B FIGURE 2-26.  The kV (A) and MV (B) images show a patient being treated for cancer in the head and neck region.

2  Clinical Radiation Oncology Physics

73

Start BAT alignment

Setup

Register to gantry

Image patient

Align to plan

Proceed with treatment

No

Move required

Move couch

Yes

FIGURE 2-27.  In-room, ultrasound-guided prostate localization is used for intensity-modulated radiation therapy for prostate ­cancer. A BAT sonographic system (North America Scientific, Chatsworth, CA) is calibrated to the gantry of the linear accelerator. A therapist uses an ultrasound probe to interactively locate the position of the prostate just before treatment.

(± SD) was −0.89 ± 3.3 mm in the right-left (lateral) direction, −1.3 ± 5.7 mm in the anteroposterior direction, and −1.6 ± 6.4 mm in the superoinferior directions. These investigators concluded that without the use of the BAT device for localizing the prostate, organ motion would have caused the CTV to move outside the PTV margin in up to 42% of the treatments—an unacceptably high probability of missing the prostate despite verification of ­patient setup with megavoltage portal images. Chandra and colleagues65 reported still larger discrepancies when daily abdominal sonography was used to measure the position of the prostate relative to the initial skin marks for treatment setup. In that study, data from 3328 images of 147 patients showed that the prostate shifted by as much as 2 cm from 1 fraction to another. Because sonography is less expensive and easier to operate than CT imaging, sonographically guided procedures are widely used in clinical settings. However, the inherently poor image quality of sonography and the unfamiliar ­appearance of sonographic images to many radiation therapists have limited its potential for precise image guidance. As an example, Figure 2-28A shows the planning contours overlaid on CT-to-CT registered images acquired with an in-room CT-on-rails system90; Figure 2-28B shows the planning contours on a sonographic image of the same ­patient that was acquired within 5 minutes of the CT scan. The sonographic image is more difficult to interpret, even though the agreement between the sonographically guided alignment and the CT-guided alignment was within 2 mm in this case. Another shortcoming of sonography for patient setup is the substantial inter-user and intra-user variability in the operation of the devices.86,91 Sonographic alignment

has been associated with larger variations than implanted ­ ducial markers for prostate localization,91 leading to recfi ommendations for larger planning margins.92 The effectiveness of sonography for prostate localization can be further reduced by anatomic distortions resulting from the pressure of the probe on the abdomen.93-95

Three-Dimensional In-Room Volumetric Imaging The capacity for three-dimensional volumetric imaging in a treatment room just before that treatment is delivered represents the latest development in image-guided radia­ tion therapy. Obtaining a three-dimensional representation of a patient’s anatomy with a CT scanner or cone-beam CT system while that patient is immobilized on the treatment couch immediately before treatment has two advantages. It enables more accurate guidance for setting up the patient’s position relative to the treatment beams, and perhaps more important, it provides images that can be used to reconstruct dose distributions based on the anatomy visualized immediately before the treatment. In principle, in-room CT scanning may facilitate image-guided adaptive radiation therapy by providing the basis for treatment plans that are adapted to changes in patient anatomy over the course of the radiation therapy. In-room CT images provide more accurate soft tissue target information than can be obtained from typical twodimensional megavoltage or kilovoltage radiographs. An additional difference is the ability of CT to assess objects in three dimensions. As an example, identifying a bony ­landmark with stereoscopic imaging, which is accomplished by combining information from a pair of two-dimensional

74

PART 1  Principles

B

A FIGURE 2-28.  A sagittal CT image (A) and sagittal sonographic image (B, right) were obtained for a patient during one treatment session for prostate cancer. Images were acquired by an in-room CT-on-rails and BAT within minutes of each other in the treatment position. Although the images can appear quite different, the difference in alignment between this pair of CT and BAT images was only 2 mm. The orange contour represents the bladder, the yellow contour represents the prostate, and the green contour represents the rectum.

radiographic images, is difficult because other unwanted anatomical or structural objects (e.g., parts of the treatment couch) are piled into the viewing angle, which can make the visualization of the landmark problematic. Volumetric imaging techniques, by comparison, can determine the unique position and shape of the object in space, a capability that may well affect radiation therapy practices. As an example, a study by Zhang and colleagues96 using an inroom CT scanner to study setup based on bony anatomy in the head and neck region showed that several bony regions could give different alignment results because of bony flexion and rotations among different bony structures. A single couch translation may not be able to correct all shifts in the head and neck region, and the investigators found that the best compromise given this limitation was to use the C2 vertebral body for selection of a bony landmark for treating common head and neck cancers. A unique feature of the CT-on-rails system at M. D. Anderson (EXaCT, Varian Oncology Systems, Palo Alto, CA; see Fig. 2-17) is that the couch remains stationary during the CT scan while the entire CT gantry moves on rails across the patient. After the CT images are obtained, the couch is rotated 180 degrees to move the patient from the CT imaging position back to the normal treatment position. The mechanical precision of this system, assessed by analyzing different sources of uncertainties, was shown by Court and colleagues97 to be accurate to within 0.5 mm. Depending on the complexity of implementation, CT images acquired immediately before treatment can be used to reposition the patient to achieve the spatial alignment of the target volume with the planned radiation beams. This is the simplest form of image-guided radiation

t­ herapy, without the need to modify the original treatment plan. Typically, couch corrections are used to realign the ­patient. To reduce variations in manual alignment between operators, automatic CT-to-CT rigid-body image registration software has been developed by Court and others90 and extended by Zhang and others98 for CT-on-rails and cone-beam CT ­images. An in-house program developed at M. D. ­Anderson, the CT-Assisted Targeting program, has been used to guide the treatment of prostate cancer (Fig. 2-29) and lung cancer (Fig. 2-30). Figure 2-29A depicts the reference CT, the simulation CT image used for treatment planning; the contours of the prostate, rectum, pelvic bone, and femoral heads and isocenter position appearing on that image were obtained from the original treatment plan. Overlaying these contours on one of the daily CT images acquired during daily treatment (see Fig. 2-29B) revealed fairly good alignment with bony structures (i.e., femoral heads and pelvis) but inaccurate alignment with the prostate and rectum. The final alignment based on the prostate target volume is shown in Figure 2-29C. The prostate and the anterior rectal wall were aligned much more closely after the automatic CT-to-CT image registration of the prostate target. However, at this aligned position, the bony structures were significantly off, indicating substantial motion of the prostate target. In this case, the motion was caused by rectal filling, which is not visible in this axial slice. Similar assessments for cone-beam CT-guided hypofractionated radiation therapy (50 Gy at 12.5 Gy per fraction) for a medically inoperable non–small cell lung cancer are shown in Figure 2-30. Tumor motion was measured using four-dimensional CT imaging, and a free-breathing ITV was

2  Clinical Radiation Oncology Physics

A

75

B

C FIGURE 2-29.  A CT-assisted targeting software program is used to aid patient alignment. A, Reference CT image, with the prostate (red), rectum (green), femoral heads (magenta), and pelvis (violet) delineated. B, Daily CT image, with the same regions of interest delineated. C, Daily CT image after alignment was determined.

created as the treatment target. Use of daily cone-beam CT guidance for setup was sufficiently accurate to allow the use of a small PTV margin (3 mm) with the goal of reducing the toxicity of this high-dose treatment. The vertebral body closest to the tumor site was used as the alignment target for the treatment (see Fig. 2-30, lower panels). The PTV contours and CTV contours were overlaid on the CT images to verify soft tissue target coverage. These overlaid contours were used as a guide to judge the target alignment after bony registration.

Image-Guided Adaptive Radiation Therapy Anatomic variations can be significant, and the actual dose distributions delivered may be sufficiently different from the intended (planned) ones so as to negatively affect treatment outcome. Technologic developments in in-room volumetric imaging have made image-guided adaptive radiation therapy possible for clinical use. Many image-guided intervention strategies can be used to account for interfractional and intrafractional variations or mitigate the consequences of such variations. Although the dosimetric or clinical benefits of the different approaches are still unclear, one strategy— midcourse replanning and treatment—may be a simple and effective way of compensating for anatomic changes due to tumor shrinkage or weight loss. Adaptive radiation therapy requires more efficient tools for delineating the target volume and the normal structures to be avoided. One alternative to repeated manual

contouring on daily CT images, which can be impractical because of time constraints, is the use of deformable ­image registration, which can greatly facilitate image-guided adaptive radiation therapy.99-105 Deformable image registration also facilitates autosegmentation, which substantially ­improves the efficiency and consistency of anatomic definition over time for adaptive replanning.105 Figure 2-31 shows an ­example of autosegmentation of head and neck anatomy by deformable image registration. The top row of that figure represents two typical axial slices near the primary tumor, and the bottom row represents the nodal region for the same patient. The two scans at left (see Fig. 2-31A) are the original CT image, with the target volumes (i.e., GTV, primary CTV, and elective nodal CTV) and normal organs (i.e.,left and right parotids and spinal cord) delineated at the time of treatment planning. The scans in the middle (see Fig. 2-31B) obtained after 3 weeks of ­radiation treatments revealed significant anatomic changes due to tissue shrinkage and a change in the orientation of the mandible. The contours overlaid on those CT images, which are quite different from those in the scans shown in Figure 2-31A, had been determined automatically by ­image-based ­deformable image registration.47 The ­resulting deformation vectors from the CT image in Figure 2-31B to the original CT image in Figure 2-31A are plotted in Figure 2-31C. The comparison (see Fig. 2-31C) indicates that the automatically segmented target volumes and the parotid glands matched well with those on the CT images obtained after 3 weeks of treatment, allowing new treatment plans

76

PART 1  Principles

FIGURE 2-30.  Comparison of cone-beam CT images (top row and bottom left, Daily) and a reference conventional treatment planning CT image (bottom right, Ref) for image-guided radiation therapy. The red contour indicates the spine; the green contour, the gross tumor volume; the blue contour, the clinical target volume; and the yellow contour, the planning target volume.

to be designed based on those automatically modified targets. The new treatment plans are subject to review and approval by the treating physician. Briefly, the concept of image-guided adaptive radiation therapy is based on the following assumptions. Accurate imaging of tumor volume and repositioning of the target under daily image guidance can significantly reduce treatment margins and the amount of normal tissue exposed to high radiation doses. These enhancements in imaging may permit hypofractionated treatment of many cancers, which may have clinical benefits and be more convenient for the patients. Technologic revolutions in radiation therapy over the past few decades leading to the widespread implementation of IMRT have made it possible to produce highly conformal dose distributions. However, this improvement in dose conformality necessitates better means of (imageguided) target localization and ways of adapting to interfractional and intrafractional changes in the design and ­delivery of radiation treatments. The combination of IMRT with image-guided radiation therapy is a natural extension for highly precise, highly conformal radiation therapy for localized tumors.

Developments in Brachytherapy Several new clinical brachytherapy procedures have been implemented in the practice of radiation oncology during the past few years. In this section, we describe the implementa­­ tion of some of these state-of-the-art procedures, including

PDR technology for treating cervical cancer, liquid-filled balloon therapy for treating brain tumors, partial-breast irradiation that uses a balloon-based high-dose-rate brachy­ therapy source, and beta-particle plaque radiation therapy for ocular melanoma.

Pulsed-Dose-Rate Brachytherapy for Cervical Cancer Continuous low-dose-rate intracavitary brachytherapy has long been a standard treatment for locoregionally advanced cervical cancer. Most patients treated for cervical cancer receive external beam radiation therapy and intracavitary brachytherapy.106,107 At M. D. Anderson, a low-dose-rate remote afterloader unit (Selectron; Nucletron, Veenendaal, The Netherlands) has been used over the past few decades as an alternative to manually afterloaded radium 226 (226Ra) and cesium 137 (137Cs) tube sources. This device was chosen to retain the biologic advantages of continuous low-doserate treatment while reducing the radiation exposure for the treating personnel. Approximately 90% of our treatments that had been done with a gynecologic tandem applicator and the Fletcher-Suit-Delclos ovoids or a cylinder are currently done with the low-dose-rate remote afterloader unit. The remaining 10% of cases involve mostly manually loaded 137Cs tube sources, with occasional use of a highdose-rate remote afterloader unit.107 Discontinuation of the current low-dose-rate remote afterloader unit by the vendor in 2007 led to renewed ­interest in the use of PDR remote afterloader units for

2  Clinical Radiation Oncology Physics

A

B

77

C

FIGURE 2-31.  Autosegmentation of previously delineated anatomy in a patient being treated for head and neck disease shows slices near the primary tumor (top row) and slices near the nodal region (bottom row). A, Original (planning) CT image shows the target volumes (red, GTV; yellow, primary CTV; blue and green, elective nodal CTV) and normal organs (purple, left and right parotids; red, spinal cord) delineated at the time of treatment planning. B, CT image obtained after 3 weeks of radiation therapy. The contours were generated by an image-based deformable image registration program. C, Deformation vectors (green) illustrate the differences in the treatment CT (B) versus the planning CT (A), showing that the automatically segmented target volumes (red, green, yellow) and the parotid glands (purple) matched well with those on the CT images obtained after 3 weeks of treatment.

i­ntracavitary brachytherapy. Moreover, an amendment to the regulations of the Nuclear Regulatory Commission on PDR remote afterloader units (10 CFR part 35) in 2004 made this technology feasible for use in the United States. Previous studies in Europe108,109 had documented the safety of PDR techniques in hundreds of cases. The PDR remote afterloader unit currently used at M. D. Anderson (Fig. 2-32) is marketed by Nucletron and is based on the high-dose-rate remote afterloader unit manufactured by the same vendor. The physical model of the PDR source is virtually identical to the high-dose-rate source from the same vendor (mHDR v2); the only difference is the lower activity of the PDR source (37 to 74 GBq at installation, depending on the institution’s needs and the room shielding) compared with that of the high-dose-rate source (approximately 370 GBq at installation). The PDR remote afterloader unit uses a single stepping iridium 192 (192Ir) source to replace the traditional low-dose-rate 137Cs sources. The stepping 192Ir source is “pulsed” for 10 to 40 minutes per hour for every hour of treatment prescribed. A fraction of the total prescribed dose is delivered as a pulse every hour over a 48-hour period to mimic the biologic effectiveness of standard low-dose-rate ­treatments.110

FIGURE 2-32.  The pulsed-dose-rate (PDR) unit is used at the M. D. Anderson Cancer Center.

78

PART 1  Principles

FIGURE 2-33.  Optimized pulsed-dose-rate dose distribution for the treatment of cervical cancer is displayed by a brachytherapy treatment planning system.

Aspects of PDR technology have been reviewed by Horton and colleagues.111 One of the advantages of PDR brachytherapy over the older low-dose-rate brachytherapy technique for the treatment of cervical cancer is that PDR uses a computer-based dose-distribution-optimization treatment planning system (Plato version 14.1; Nucletron) that maximizes the tumor dose while sparing the rectum and bladder tissue (Fig. 2-33). Another advantage of PDR brachytherapy over low-doserate brachytherapy is that in PDR brachytherapy, nursing care can be given during the time between treatment pulses, which allows such care to be given as needed and allows the patient to have visitors without extending the overall treatment time, both of which reduce the stress of the treatment for the patient. The PDR remote afterloader unit also allows the dwell times at each source position during each treatment pulse to be varied, resulting in a more customized dose distribution. The facility shielding requirements for a PDR remote afterloader unit are similar to those for the Selectron lowdose-rate remote afterloader unit. The instantaneous dose equivalent rate is higher with the PDR remote afterloader unit than with the low-dose-rate remote afterloader unit, but the dose equivalent rate averaged over time should be the same as that for the low-dose-rate treatments. If ­­local regulations for shielding include limits for the instantaneous dose rate, shielding must be designed such that that these limits are not violated. One advantage of PDR brachytherapy over low-dose-rate brachytherapy is that the half-value layer in concrete for the 192Ir PDR sources can be less than that for 137Cs sources, because 192Ir has the lower average energy of the two.112 However, new PDR treatments are planned that may result in a higher average dose equivalent rate or higher total dose equivalent;

for example, large interstitial implants may result in higher average dose-equivalent rates than those for gynecologic PDR treatments. As is true for all radiation delivery devices, verification of the machine reliability is crucial for the safety of PDR treatments. A particularly important aspect of a PDR unit is its ability to transfer the source to and retract the source from the applicator multiple times without failure. The warranty supplied by the vendor (Nucletron) is that the PDR unit performs 25,000 transfers without failure. Tests should also be performed with various flexion angles of the transfer tubes to establish which flexion angle will cause failure. Procedures must be in place and training given for emergency response and machine operation for all treating personnel.

Liquid Source Brachytherapy for Brain Tumors Approximately 167,000 malignant brain tumors are dia­ gnosed in the United States each year; of those, about 17,000 are primary brain tumors, and more than 150,000 are brain metastases.113 The traditional standard of treatment for brain tumors has been surgery followed by external radiation therapy. However, most malignant brain tumors recur quickly, and treatment options for patients with recurrent tumors have been limited. In 2001, the U.S. Food and Drug Administration (FDA) approved a liquid source brachytherapy system (the GliaSite Radiation Therapy System; Cytyc Corporation, Marlborough, MA) for use in treating newly diagnosed ­metastatic and recurrent brain tumors. The device consists of a patented balloon catheter filled with Iotrex, a proprietary liquid iodine 125 (125I) radiation source. This system is ­designed to deliver a spherically uniform dose of radiation

2  Clinical Radiation Oncology Physics

to the tissue surrounding a brain tumor resection cavity, with the intent of improving tumor control and minimizing damage to healthy brain tissue. The delivery device consists of a double-walled balloon catheter that is implanted into the tumor cavity during the surgical resection of the tumor. The other end of the catheter extends outside of the skull and is concealed underneath the skin at the top of the head with a distal inflatable balloon (Fig 2-34), which is accessed through a subcutaneous injection port. The radioactive source, Iotrex, is a sterile, ­nonpyrogenic solution containing sodium 3-(125I) iodo-4-hydroxy-benzenesulfonate.114 It is delivered in nominal 1.0-mL unit dose conical vials, or for an additional cost, it is preloaded in a syringe that can be directly dispensed into the balloon during the procedure. Our experience and that of others115 indicates that with appropriate guidance and care, this procedure can be performed safely. To begin the procedure, the surgeon injects saline and iodine-based contrast (e.g., Omnipaque) into the ­ balloon to a total volume that fills the resected cavity. CT or magnetic resonance imaging is used to confirm balloon ­integrity and fluid volume.116 Doses for treatment ­planning are calculated from manufacturer-supplied tables that display dose as a function of balloon diameter and specified depth from the surface of the balloon. An Iotrex order contains a minimum of 150 mCi at the time of ­ calibration. Net ­afterloaded ­activity can range from 75 to 1200 mCi, ­although most common clinical loadings are in the range of 200 to 350 mCi. The required activity depends on the prescribed radiation dose and dose rate (typically 40 to 60 Gy at 40 to 60 cGy/hr), treatment depth (0.5 to 2 cm from the balloon ­surface), and balloon diameter (2 to 4 cm).117 Because conformance of the balloon to the resected ­cavity improves several days after surgery, the Iotrex is typically injected 3 to 4 days after surgery. The injection procedure is performed under the supervision of a radiation ­oncologist,

FIGURE 2-34.  Diagram shows the GliaSite Radiation Therapy System in place for treating a brain tumor.

79

a medical physicist, and a radiation safety officer. The Iotrex is injected into the balloon with a shielded syringe. Typically, the radiation is given for 72 hours at a dose rate of 50 cGy/hr to deliver doses on the order of 36 Gy. After ­completion of the radiation therapy, the Iotrex is ret­ri­ eved by the medical physicist and radiation safety ­officer. The typical loading or removal procedure takes 30 to 45 ­minutes. The Iotrex solution remains in a secured facility in the radiation oncology department or other designated area approved by the radiation safety officer until the actual procedure time, and the source is handled according to generally accepted safe uses of radiopharmaceutical practices. The safety and performance of the GliaSite system have been verified in a multicenter clinical study sponsored by the National Cancer Institute involving patients with recurrent brain tumors.118-120 The average survival time for patients treated with the GliaSite device was 387 days, and the 1-year survival rate was 52%. The GliaSite system is available in more than 150 centers throughout the United States, and as of 2006, more than 800 patients had been treated with it. Reimbursement for the treatment usually is available.

Partial Breast Brachytherapy More than 178,000 new cases of breast cancer are expected to be diagnosed among women in the United States in 2007, of which about 63% are diagnosed when the disease is still localized.121 Although some patients with earlystage disease choose to undergo mastectomy, many choose a less invasive course of treatment. This approach, called breast conservation therapy, usually involves tumor excision by lumpectomy followed by radiation therapy to reduce the likelihood of recurrence. Treatment options for radiation therapy after surgery for breast cancer include brachytherapy with devices such as the MammoSite Radiation Therapy System, which������� , like the GliaSite device ������������������������������������� is manufactured by Cytyc. MammoSite, approved by the FDA for the delivery of high-dose-rate brachytherapy in May 2002, may make it easier for more women to consider the choice of lumpectomy, and it provides physicians with an important tool for breast­conservation therapy.122-125 The brachytherapy device is designed to treat the tissue immediately surrounding the lumpectomy cavity. The treatment procedure is as follows. During or shortly after the lumpectomy, the deflated balloon is placed inside the tumor resection cavity. The applicator shaft, a tube connected to the balloon, remains outside the breast. Once in place, the balloon is inflated with saline to fill the cavity, the catheter site is dressed, and the patient can go home, returning for treatment on an outpatient basis. ­ Radiation treatment is delivered by connecting the balloon catheter to a computer-controlled, high-dose-rate remote afterloader (Fig. 2-35), a freestanding brachytherapy unit that ­houses the radiation source. The 192Ir source is then inserted into the balloon to deliver the prescribed dose of radiation in a 1- to 5-day sequence of treatments. When the therapy is concluded, the balloon is deflated, and the catheter is ­easily removed. The balloon, available in spherical and elliptical shapes in two sizes (4 to 5 cm or 5 to 6 cm), is inflated with radiopaque contrast solution to enhance identification of

80

PART 1  Principles

the balloon boundary on CT or conventional film radiography. Inflation of the balloon stretches the tissues and shapes them into an approximately spherical or elliptical shell conforming to the balloon surface. The high-doserate source is afterloaded into the balloon at its center for specified treatment durations to deliver the prescription dose fraction. Multiple dwell positions can be used with the spherical and elliptical balloons to optimize the dose distribution.126 A typical treatment prescription is 34 Gy delivered in fractions of 3.4 Gy twice each day for 5 days, with a minimum of 6 hours between fractions. The dose typically is prescribed to a distance of 1 cm from the balloon surface. This prescription line imposes some restrictions on the distance of skin from the balloon surface to limit the skin dose and avoid cosmetic damage. A minimum distance of 5 to 7 mm between the skin and the balloon surface is typically required for such treatments.127 General physical guidelines for determining whether this technique is appropriate for a specific patient include assessments of balloon conformance to the lumpectomy cavity to determine the adequacy of target coverage and dose distribution, minimum skin distance from balloon surface, and balloon diameter and symmetry. Image-based ­assessment (CT or radiography) and treatment planning are important for this purpose and for calculating the ­appropriate source dwelling time for the desired prescription. Accurate positioning (within 1 mm) of the high-doserate source at the center of the balloon or at some other specific dwell site is an essential part of the procedure. ­Error in the source position can result in significant variations in the dose at the prescription line. Verification x-ray films are used at each fraction to make sure the source is positioned at the exact location planned and to monitor any deformation or deflation of the balloon. The concentration of the contrast solution can also affect the dose delivered; according to Kassas and colleagues,128 the concentration of that solution in the balloon should be limited to 10% to ensure that reduction in the prescribed dose is limited to 3% or less regardless of balloon diameter. Other radiologists have reported similar conclusions.129-131

Beta Particle Plaques for Ocular Melanoma The incidence of ocular melanoma arising in the uveal tract (i.e., iris, ciliary body, and choroid) is about 6 per 1,000,000 people, and ocular melanoma is the most common primary

FIGURE 2-35.  Drawing shows placement of the MammoSite balloon and the 192Ir afterloading device.

eye cancer in adults. As an alternative to the traditional treatment (i.e., surgical enucleation), radiation therapy modalities, including brachytherapy with various radio­ active sources (e.g., 125I, ruthenium 106/rhodium 106 [106Ru/106Rh], 60Co, strontium 90/yttrium 90 (90Sr/90Y), 192Ir, palladium 103) or external beam therapy with protons or photons, have been tested in attempts to preserve the affected eye.132-134 The most commonly used procedure in the United States is 125I plaque brachytherapy. Local control rates are excellent (90% to 95%), and research is being conducted to reduce side effects to normal tissues, such as radiation-induced retinopathy or optic neuropathy, vitreous hemorrhage, cataract, neovascular glaucoma, and others.135-138 In 1999, Wilson and Hungerford138 conducted a retrospective, nonrandomized chart review of patients with choroidal melanoma treated in the United Kingdom with 125I or 106Ru/106Rh plaques or proton therapy and concluded that local control rates were better for those treated with the 125I plaques or protons (95%) than for those treated with 106Ru/106Rh plaques (89%). However, the 106Ru/106Rh plaques had fewer side effects. Comparative dosimetric studies of beta particle ­therapy have been confounded in the past by the large uncertainties (up to 30% at the 2s confidence interval) in the 106Ru/106Rh dose rate calibration. However, the primary calibration standard for beta-emitter brachytherapy has greatly improved, reducing the absolute dose calibration uncertainty to ± 7% at 2s.139 No prospective, randomized clinical trials have been done to compare clinical outcome among the various radiation therapy options for treating ocular tumors, although some anecdotal reports have been published.140-146 To conduct comparative trials, institutions must be able to predict radiation dose distributions in individual patients in an accurate and standardized manner, which has been difficult despite improvements in ­calibration. 106Ru/106Rh episcleral plaques were developed in ­BerlinBuch during the 1960s (BEBIG, Berlin, Germany) and have since been used successfully worldwide. 106Ru/106Rh is a parent-daughter beta emitter with maximum energy of 3.54 MeV, mean energy of 1.42 MeV emitted by 106Rh, and half-lives of 371.6 days (106Ru) and 29.8 ­seconds (106Rh). The plaques have a spherically concave silver backing with a 12-mm inner radius of curvature. 106Ru is electro­ deposited on the surface of a 0.2-mm-thick silver substrate mounted on 0.7-mm silver backing and covered with a 0.1-mm-thick silver window. Plaques can be made in a ­variety of shapes and sizes (Fig. 2-36) to treat tumors of different type, size, and location. 106Ru ophthalmic plaques are also used for follow-up radiation therapy after tumor resections and for the treatment of hemangioma, pterygiums, and conjunctival tumors. 106Ru plaque radiation therapy has been approved by the FDA for treating tumors with apical heights of 5 mm or less and tumor base diameters of up to 15 mm. A 2.5-mm margin between the edges of the tumor and the plaque is recommended. In the United States, tumors higher than 5 mm are not treated with plaque therapy so that the scleral dose can be limited to less than approximately 500 Gy, although scleral doses exceeding 1200 Gy have been ­ reported in Europe.147 The percent depth dose fall-off of beta emitters is steep at a distance from the source; at a depth of 7 mm, the percent depth dose of an

2  Clinical Radiation Oncology Physics 106Ru

plaque in tissue is approximately 5% to 10%, depending on plaque size and shape. Mourtada and colleagues148 compared dosimetry of 106Ru/106Rh plaque brachytherapy, external beam proton therapy (which has been proposed for ocular treatments at M. D. Anderson), and traditional 125I plaque brachytherapy as used in the large Collaborative Ocular Melanoma Study (COMS).136 The results indicated that the proton beams provided superior dose uniformity within the tumor volume, whereas the dose distribution from the 106Ru/106Rh plaques was quite heterogeneous. The dose distributions of 106Ru/106Rh plaques and proton therapy were better confined to the tumor volume, thereby sparing other critical ocular structures, than was that of the 125I plaques. For protons, only doses lower than the maximum dose were ­delivered outside the tumor volume. Depending on the clinical situation, use of 106Ru/106Rh brachytherapy or proton therapy may help to spare critical structures in the sclera and optic disc boundary. At M. D. Anderson, 106Ru plaques have been used as a complement to 125I plaques for the management of uveal melanoma since late 2003. More than 45 patients have been treated with 106Ru. For dosimetry quality assurance and plaque commissioning, we measure the absolute dose rate and relative dose uniformity by using radiochromic film and a hemispherical eye phantom. The films are digitized with a high-resolution charge-coupled detector densitometer (PeC 100, Photoelectron Corp., North Billerica, MA) to a 0.13-mm pixel resolution. A 90Sr/90Y source traceable to the National Institute of Standards and Technology (Gaithersburg, MD) is used for film calibration. A scaling function for converting the measured dose rate in plastic to that in water is estimated with Monte Carlo simulations (MCNPX code, Los Alamos National Laboratory, Los Alamos, NM). Our overall measurement uncertainty is ± 11% (2σ). The measured dose distributions are used for plaque commissioning and treatment planning. Treatment planning is done with an Excel spreadsheet and the SURFER program (Golden Software, Golden, CO) for isodose contour plotting. With these tests, we found use of the 106Ru ophthalmic plaque to be much more convenient than use of the 125I COMS plaque because the 106Ru plaques can be reused for about 18 months. We restrict the number of possible sterilization cycles to 50; after this period, the dose rate of

81

the plaque is reduced (because of the physical half-life of the 106Ru) to a point where the necessary treatment time is detrimental to the patient. The 106Ru ophthalmic plaques must be handled carefully to avoid scratching or deforming them and to avoid radioactive contamination of the patient and the environment.

Proton Radiation Therapy Like the rise of the proximal side of a pristine proton Bragg peak, interest in proton therapy is rising rapidly. Five proton therapy centers are in operation in the United States (Loma Linda University, Massachusetts General Hospital-Harvard Medical School, M. D. Anderson Cancer Center, the University of Florida, and the Midwest Proton Radiotherapy Institute in Bloomington, Indiana), and many more are in operation around the world. New centers are being built and more are likely in the not too distant future. Several characteristics of protons make them appealing for radiation therapy. The dose delivered in the regions proximal to the target volume is lower than that delivered by photons, and protons lose energy continuously, depositing increasing amounts of energy per unit path length as they penetrate tissues. Protons stop abruptly after penetrating a finite range; there is no exit dose compared with photons or electrons (Fig. 2-37). These properties allow the generation of target dose distribution patterns in which dose near the entrance is lower than that from photons and the dose at relatively small distances beyond the target volume drops rapidly to virtually zero.

Production of Protons Protons can be accelerated to energies of therapeutic range (typically, 70 to 250 MeV) by a cyclotron or a synchrotron. Both types of accelerators are used in radiation therapy, and each has its advantages and disadvantages. Cyclotrons are more compact and have higher beam intensity, whereas synchrotrons have greater energy flexibility, smaller en­ ergy spread, and lower power consumption. Cyclotrons produce continuous proton beams of fixed energy. The clinically desired energies are achieved by rapidly inserting or removing mechanical degraders. Synchrotrons accelerate

120.0

120 MeV 4 cm SOBP 20 MeV

100.0 80.0 60.0 40.0 20.0 0.0 – 20.0

FIGURE 2-36.  The 106Rb eye plaques can be made in a variety of shapes and sizes.

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

FIGURE 2-37.  Depth dose curves for protons (blue) and electrons (red). The proton beam energy is 120 MeV with a 4-cm, spread-out Bragg peak (SOBP), and the electron beam energy is 20 MeV. Curves are normalized to the distal 90% dose.

82

PART 1  Principles

rotating modulation wheels or by actively scanning with magnets and varying the energy of the protons before they enter the nozzle. This discussion focuses primarily on passively scattered protons, but a brief description of scanned protons is included under “Discrete Spot Scanned Proton Beams.”

Passively Scattered Proton Beams The narrow, essentially monoenergetic proton beam that en­ters the nozzle must be changed into a broad, polyenergetic beam to be clinically useful. A monoenergetic beam in water would deposit most of its energy in a very small volume near the end of its range, on the order of 1 cubic centimeter (see Fig. 2-39). To increase the size of this volume in the direction along the path of the beam (i.e. to spread the beam out), energy is degraded in SCANNED PROTON BEAM IN WATER, NOMINAL RANGE 20 CM

Y distance (pixels)

proton pulses to the desired energy levels. The different proton energies are obtained by having a different number of revolutions in the synchrotron, because the protons gain energy with each revolution. Because of the pulsed nature of the beam, scanning the proton beams produced by synchrotrons is more complex than scanning the beams produced by cyclotrons. The concept of beam scanning is discussed further under “Discrete Spot Scanned Proton Beams.” Typically, synchrotrons produce a narrow beam (1 to 10 mm in diameter, depending on the energy and accelerator type) of nearly monoenergetic protons with relatively small initial energy and intensity spread. This beam is ­directed magnetically from the synchrotron through the high-energy beamline to the gantry and into the treatment head, or nozzle. The term nozzle is used in proton therapy to describe the devices that the beam passes through after leaving the evacuated beam line; a schematic of the nozzle of the Hitachi proton therapy system used at M. D. ­Anderson is shown in Figure 2-38. The typical dose ­distribution in water from an unscattered beam exiting from the nozzle, a single spot, is illustrated in Figure 2-39. To be clinically useful, these dose distributions must be spread out laterally and longitudinally (i.e., in the direction of the beam). This can be accomplished by passively scattering protons with a combination of scattering foils and

50 100 150 200 250 300 0

100

Beam

Second scatterer

Dose monitor Flatness monitor Square collimator Snout

500

600

PROFILE

Profile monitor reference Dose monitor 2800 2600 X-ray tube

Range shifter

2400

3.2 m

Laser marker

Block collimator Irradiation head

Image intensity value

Range modulation wheel

200 300 400 X distance (pixels)

2200 2000 1800 1600 1400 1200

(Iso-center)

FIGURE 2-38.  Schematic drawing shows the components of the nozzle of the Hitachi proton therapy system that is used at the M. D. Anderson Cancer Center.

0

50

100

150

200

250

300

Y distance (pixels)

FIGURE 2-39.  A typical dose distribution in water is produced by an unscattered proton beam exiting from the nozzle. The graph is a profile measured along the dotted line.

2  Clinical Radiation Oncology Physics

83

Proton Beam

FIGURE 2-42.  A 2-cm-thick brass aperture is used to shape the proton beam. Three identical pieces are required for the  250-MeV proton beam. FIGURE 2-40.  A range-modulator wheel is used for 120-MeV protons for the Hitachi proton therapy system. The wheel ­rotates at 400 rpm and, in combination with a second scatterer, spreads the beam laterally to an approximately 25-cm diameter. The six groups of steps (three each rising and falling) mean that there are 2400 modulations per minute.

SOBP WIDTHS WITH BEAM GATING 120

Relative dose (%)

100 80 60 40

40% ON 70% ON 90% ON

20

60% ON 80% ON 100% ON

0 0

50

100

150

200

Depth in water (mm)

FIGURE 2-41.  Measured spread-out Bragg peaks (SOBPs) are provided for a series of beam widths. The proton beam is “gated on” at the thinnest part of range modulator wheel and is “gated off” when that wheel has rotated through a certain angle to achieve the SOBP of desired width. The key indicates the angles required to achieve the SOBPs as percentages of one full modulation.

predefined steps. This degradation is often achieved with the aid of range modulator wheels, which are shaped like a propeller with a series of steps of increasing thickness (Fig. 2-40). Range modulator wheels are placed with their axis of rotation parallel to the path of protons (see Fig. 2-38). As the wheel rotates, the monoenergetic proton beam passes through varying thicknesses of material that degrades it, thereby producing a series of pristine Bragg peaks of appropriate intensity. A typical passive scattering component of a proton therapy system includes an initial scattering foil, a range modulator wheel, and a second scattering foil to produce broad and flat proton beams. At the M. D. Anderson facility, for instance, each passive scattering proton treatment unit is equipped with eight different energies, 24 range modulator wheels, and nine

second scatters for three different snouts to provide maximum uncollimated field sizes of 25 × 25 cm, 18 × 18 cm, and 10 × 10 cm. The appropriate choice of range modulator wheel and its parameters and the second scatterer modulates the beam to the desired width and field size to adequately cover the target. The resulting dose distribution is called the spread-out Bragg peak (Fig. 2-41), which in the M. D. Anderson system can range from 2 cm to 16 cm wide and is defined as the distance between the 95% proximal dose point and the 90% distal dose point. Range, which is typically defined as the depth of the distal 90% dose, is an important parameter for clinical beams. The range depends on the beam energy entering the nozzle and the maximum field dimension, becoming smaller as the maximum field sizes become larger. Protons lose ­energy when they are scattered to cover a larger field size. For example, the maximum range for 250 MeV protons is approximately 32.5 cm when scattered to cover a 10 × 10 cm field, 28.5 cm to cover an 18 × 18 cm field, and 25.0 cm to cover a 25 × 25 cm field. Achieving the desired range to within 1 mm in water requires selecting an appropriate energy exiting from the accelerator and finetuning it as needed by using range shifters, which are slabs of water-equivalent material placed in the beam.

Shaping Proton Dose Distributions To create a dose distribution that conforms to the shape of the target, the spread-out Bragg peak of the passively scattered beam is further shaped lateral to the beam by using a field aperture and distal to the beam by using a range compensator. The aperture is typically made of brass of sufficient thickness (2 to 8 cm) to absorb incident protons of highest energy (Fig. 2-42). Some proton therapy delivery systems also include MLCs to shape the lateral field boundaries. The field aperture fits in an aperture holder in the nozzle and is upstream of the compensator. The compensator, which is made of nearly water-equivalent material such as wax or Lucite (Fig. 2-43), is used to degrade the beam energy on a point-by-point basis by variable amounts so that the distal edge of the beam conforms to the shape of the target plus a suitable margin. The compensator is the final element in the nozzle. The air gap between the patient and the compensator is minimized to reduce the penumbra by moving the snout

84

PART 1  Principles

FIGURE 2-43.  A custom-made acrylic compensator is ­designed to conform the distal edge of the beam to the target plus ­margin for an individual patient.

closer to the patient. Appropriate aperture openings and ­compensators are designed in the treatment planning process; ­individualized design information is sent to a ­machine shop, where a computer-controlled milling ­ machine constructs the devices. Because device construction may require several hours, adequate time must be ­ allowed between treatment planning and treatment delivery to construct the devices, review them for quality assurance, and collect ­patient-specific measurements before they are used.

Treatment Planning for Passively Scattered Protons Strategies for planning proton treatments must consider the unique characteristics of protons, especially with regard to their transport in tissues. The most important characteristic of protons is their limited range in the patient. The exact point at which protons stop in the patient is uncertain because of uncertainties in Hounsfield units and proton stopping powers (energy loss per centimeter); perturbation of proton beams due to interfractional and intrafractional variations in position and in the volume and shape of anatomic structures; and uncertainties in models that compute dose distributions, especially in the presence of complex heterogeneities. Proton dose distributions usually are more vulnerable to these uncertainties than are photon dose distributions because of the definite range of the protons. The PTV concept1 must be refined when planning proton treatments. Modest displacement of the patient in his or her entirety along the path of the proton beam has ­virtually no impact on proton dose distributions. However, deformations, rotations, and relative displacements of anatomic structures, whether along the beam direction or perpendicular to it, can affect the depth of penetration. The other sources of uncertainties described previously can also affect the range of protons. For this reason, field-specific

distal and proximal margins are assigned to the CTV to ensure coverage of the CTV along the path of each beam. The magnitudes of these margins are estimated based on clinical experience and measurements and depend on the depth of the target, heterogeneities in the path to the ­target, and image artifacts, among other factors. Lateral margins depend on the anatomic variations orthogonal to the beam direction and are determined in the same manner as those for photons. An additional strategy to ensure coverage of the target in the presence of variations of anatomic structures of ­significantly different densities perpendicular to the beam ­direction is “smearing” of the compensator.149 The smearing process essentially reduces the width of thicker regions of the compensator (i.e., smears the details of the compensator) to allow protons to penetrate more deeply even when adjacent higher-density tissues move into their path. This user-controlled parameter can be adjusted during treatment planning. Smearing and margins for range uncertainty ensure coverage of the target, albeit at the expense of higher dose to normal tissues distal to the target. Compensator design for a target that moves with respiration, if based on representation of the shape, size, and position of the target on conventional CT images, may lead to inadequate coverage of the target at the time of treatment. Four-dimensional CT data sets (comprising multiple three-dimensional CT data sets in a sequence of perhaps 10 phases of the respiratory cycle) are used for treatment planning for photons and protons to design the motionintegrated ITV enveloping the moving target.2 In ­designing compensators for proton therapy, the density at each point in the ITV may be set to the maximum density in any one of the phases.150 Alternatively, one compensator may be designed for each phase of the four-dimensional CT. The union of all the compensators represents the ­final composite compensator.151,152 Either strategy ensures that protons will penetrate sufficiently deeply to cover the target adequately regardless of its position. In a typical proton treatment planning system for a specified beam direction, the system determines (1) the maximum water-equivalent depth of penetration to the distal edge of the target plus the margin and (2) the minimum water-equivalent depth to the target minus the margin. These quantities are then used to determine the range and therefore the energy of incident protons and the width of the spread-out Bragg peak. The range and width values are eventually used to select the appropriate treatment machine parameters (e.g., energy, range modulator wheel, second scatterer, range shifters) required to deliver the beam. Planning systems generally use simple ray tracing to design the compensator. The thickness of the compensator at any point is assumed to be proportional to the difference ­between the maximum range and the water-equivalent path length to the distal edge of the target plus the ­ margin. This value is an approximation, because the variations in lateral transport of protons and secondary particles are ­ignored. More sophisticated iterative strategies are possible but typically are not used. The smearing process partially mitigates the consequences of this approximation. The dose distribution produced by compensated passively scattered proton beams is conformal only for the distal edge of the target, not for the proximal edge.

2  Clinical Radiation Oncology Physics

The current systems for dose calculations in planning passively scattered proton treatments generally use ­ semiempiric pencil beam models.153-155 Parameters of the ­analytic formalisms of the model are fitted with measured dose distributions. These models are quite accurate except for complex heterogeneities such as those in head and neck or thoracic regions or when intrafractional motion is ­present.

Discrete Spot Scanned Proton Beams Another way of shaping the dose distribution of protons in three dimensions is to use thin scanning pencil beams of a sequence of energies and varying intensities.156-157 Typically, the location of the pencil beam, often referred to as the spot at the end of its range, is controlled by two steering magnets, which are located upstream in the nozzle. One steering magnet controls the location of the beam in the x direction, and the other controls the location of the beam in the y direction. One or more scanned proton beams, directed at the target volume from different directions, may be used to deliver intensity- and energy-modulated proton therapy. The positions, energies, and intensities of individual pencil beams (or spots) composing the incident proton beams are determined with various optimization techniques. The optimization process minimizes an objective function so that the combined dose distributions from all pencil beams achieve the best possible approximation of the specified clinical objectives. Objective functions are typically specified in terms of dose and dose-volume constraints. Depending on the delivery device, intensity-modulated proton therapy can be delivered by using raster scanning or discrete spot scanning. In both cases, the treatment delivery is carried out in layers, with each layer corresponding to a ­given energy of incident protons. On completion of one layer, the energy is changed to the next one in sequence. In raster scanning, the pencil beam is moved continuously while its intensity is varied. In discrete spot scanning, the pencil beam is stopped, a specified dose is delivered, the radiation is then turned off, and the pencil beam is moved to the next spot. The widths of the pencil beam or spot in air and in ­water at the end of the range of protons are important ­parameters. Spot sizes are specified in terms of 1 × σ or full-width-at-half-maximum (2.35 × σ). Delivery systems are designed to minimize spot widths in air, and some systems have been able to achieve spots as small as σ = 3 mm. However, the increase in spot width due to scattering in the medium cannot be controlled. As an illustration, a discrete spot scanning system at M. D. Anderson was used to generate Figure 2-39, which shows the dose distribution of a single pencil beam in water for a proton beam with a range of 20 cm (energy of approximately 170 MeV). With that system, the full-width-at-half-maximum value of the spot in water as a function of energy ranges from approximately 12 mm for 220 MeV to 21 mm for 120 MeV. The intensity-­modulated proton therapy planning system determines the spot pattern (i.e., positions and intensities of spots) as a function of a ­ sequence of energy layers. Layers of spots are duplicated by the planning system to allow “repainting,” which is useful for mitigating the consequences of intrafractional motion. During the delivery of proton therapy, one energy layer is delivered at a time. Synchrotrons can accelerate protons

85

to different energies in approximately 2 seconds. Ranges from 4.0 to 30.6 cm in 94 energy steps are available, with field sizes up to 30 × 30 cm. The challenges associated with implementing scanning proton beam treatments are substantial, including the need to commission the treatment planning system and establish means of quality assurance for machine calibration and for ­patient-specific measurements. The treatment planning system provides the dose distribution and the spot position, ­energy, and intensity (the latter in numbers of monitor units).158 Because each location in the target volume receives dose from multiple spots, some of which may be many centimeters away, the spots must be modeled with great accuracy. An early evaluation of the planning system at M. D. ­Anderson revealed that a single Gaussian model was inadequate for modeling the in-air beam spot and that a more complex model (e.g., multiple-Gaussian model) is required. Verification of the actual dose distributed to the patient is more challenging for intensity-modulated proton ­therapy than for intensity-modulated photon therapy. Although the verification is a three-dimensional problem in terms of the dose delivered to a volume and the pattern of the spots, only two-dimensional dosimetry systems are available (e.g., film, two-dimensional ion chamber arrays, and a Lanex scintillator screen coupled with a charge-couple ­device camera). Because these systems are laborious and time-consuming to use and do not provide adequate resolution in depth, three-dimensional systems are needed. Another challenge is the independent verification of monitor units assigned by the treatment planning system. The approach at M. D. Anderson is to export the scan pattern produced by the planning system to an independent intensity-modulated proton therapy dose-computation system developed in house and compare the dose distributions computed by the two systems. A proton scanning beam approach in use at the Paul Scherrer Institute since 1996 involves one-dimensional scanning of proton pencil beams of different energies in the patient’s transverse plane.157 The other dimension is achieved by moving the couch along the patient’s longitudinal axis. Many facilities are at various stages of implementing two-dimensional scanning. The key advantage of scanned beams and intensity-modulated proton therapy is the opportunity to deliver more conformal dose distributions that optimally balance tumor dose compared with normal tissue doses. The third degree of freedom in intensity-modulated proton therapy relative to intensity-modulated photon therapy (i.e., the energy) allows greater flexibility to achieve superior dose distributions. The proximal dose can also be substantially reduced compared with passive scattering. Another important advantage is the reduction in the potentially carcinogenic neutron contaminants in the incident proton beams, given the absence of the various scattering components present in the scattering nozzle. Spot size is of concern in scanned beams because it ­ affects the width of the penumbra and limits fine adjustments of dose. Some institutions are installing MLCs to ­ reduce ­ penumbra width; however, this comes at the ­expense of some increase in neutron dose. For targets that move during irradiation (e.g., lung or liver tumors), hot and cold spots can arise through the interplay between

86

PART 1  Principles

spot ­ movement and anatomic motion. Repainting and ­respiratory-gated treatments can minimize these effects. High-speed raster scanning with fast energy change, available in some delivery systems, is another way of achieving similar goals. Ways of incorporating uncertainties in ­intensity-­modulated ­proton therapy, a significant challenge compared with passively scattered beams, have yet to be developed. Nevertheless, scanned beams and intensity-modulated proton therapy represent major advances in the field of proton therapy. As discussed further in Chapter 40, investigators from the Paul Scherrer Institute have used these techniques to treat ocular melanomas, brain tumors, head and neck cancers, chordomas, chondroscarcomas, prostate cancer, and pediatric cancers. At M. D. Anderson, ­potential clinical applications of this technology include the use of a single isocenter technique with precision couch movements to treat cranial or spinal tumors and the use of ­intensity-modulated proton therapy to treat cancers of the head and neck or breast.

Proton Radiobiologic Effectiveness Another factor that is important in proton treatment planning is the radiobiologic effectiveness (RBE) of protons. In routine clinical practice, the RBE for protons, as estimated from numerous experiments, is considered to be 1.1. Computed physical dose distribution is multiplied by the RBE to obtain the cobalt Gray equivalent (CGE) dose distribution. Dose is prescribed in units of dose equivalent by the oncologist and converted to physical dose for the purposes of monitor unit determination. However, the RBE has been shown to vary with linear energy transfer, which increases as the energy of protons decreases with increasing depth.159,160 This effect may be particularly significant near the end of the proton range. This variation in RBE is ignored, and the biologically effective dose near the end of the range is likely to be higher than what is seen on a treatment plan. For this reason and because of uncertainties in range, aiming the beams toward a normal critical structure is avoided, particularly when the number of beams in the treatment plan is small. In many ways, despite decades of clinical experience, proton therapy is still quite basic. Many opportunities exist to improve the proton dose distributions, and many questions about proton therapy remain unanswered. Even in its current state, proton therapy is expected to be superior to therapy with photons or electrons for specific diseases. The absence of an exit dose makes protons very desirable for pediatric patients. Developments in intensity- and energymodulated proton therapy, image-guided interventions for interfractional and intrafractional variations, reduction in the uncertainties discussed earlier, and mitigation of their consequences should lead to considerable improvement in the therapeutic potential of protons.

Future Technologic Developments in Radiation Therapy This chapter began with a look at earlier developments, and it closes with a look at what may be the major technologic improvements in radiation therapy in the second decade

of the 21st century. The rapid technologic advances that began in the mid-1980s with the development of threedimensional conformal radiation therapy and followed by IMRT are continuing at an increasing pace. Current research and its anticipated translation into clinical practice are discussed briefly in the following sections.

Image-Guided Radiation Therapy Image-guided radiation therapy is a major focus of the radiation oncology community. This technique is being introduced into clinical practice, although in a relatively rudimentary form. Many challenges remain, including a means of robust and accurate deformable registration and autosegmentation; adapting treatment planning, setup, and delivery to mini­mize the impact of and account for interfractional and in­trafractional variations, including optimal and efficient replanning to consider interfractional variations in anatomy, biology, and function (i.e., adaptive radiation therapy); im­proving the quality of respiratorycorrelated imaging; and enhancing the robustness of correlation between surrogates of respiratory motion and target motion. Although most techniques for addressing intrafractional motion focus on respiratory motion, other forms of motion, such as cardiac motion or swallowing, may also be addressed in the future. Numerous unanswered questions remain, including those on the magnitudes and patterns of variations; the consequences of current assumptions about the invariance of patient anatomy or the approximate manner with which anatomic variations are accounted for; the dependence of potential benefits of image-guided radiation therapy strategies of different complexities and sophistication on tumor and patient characteristics; and the gains in outcome expected with the implementation of these strategies, among others. Initially, many of the answers to these questions are likely to be obtained through virtual clinical trials (i.e., treatment planning studies). These studies will lead to the development and implementation of optimal image-­guided radiation therapy strategies and will allow the selection of appropriate cost-effective strategies that are customized for each patient. Image-guided interventions may become the standard of practice to account for interfractional variations in ­patient geometry. Various forms of real-time or almost-­real-time imaging methods, such as cone-beam CT or in-room CT, are currently in clinical use, and other techniques, such as limited projection imaging, may enhance the rapid acquisition of data on patient geometry. Real-time treatment planning techniques may allow the rapid acquisition of dose information with the ultimate capability of modifying the treatment plan to account for the interfractional variations in patient geometry. In addition to accounting for interfractional variations, image-guided interventions may eventually account for intrafractional variations in patient geometry as well. The ability to image implanted fiducials in real time radiographically or nonradiographically, along with the ability to correlate the position of these fiducials with internal anatomy, may provide a more accurate representation of the real-time variations in patient anatomy during delivery of radiation treatment, with the potential for real-time modifications in the beam delivery to account for these variations.

2  Clinical Radiation Oncology Physics

Biologic and Functional Imaging and Dose-Response Assessment and Modeling Another area of intense interest for research is biologic and functional imaging done before and after radiation therapy. The objectives of this research are to determine tumor extent, heterogeneity, and sensitivity; the sensitivity and function of the critical normal tissues that must be protected; and the tumor and normal tissue response to dose distributions. A related area of research is the analysis of dose-response data and the development of doseresponse models. Incorporation of clinically relevant doseresponse data and models into optimization of treatment plans is likely to lead to improved outcome. Functional imaging of physiologic processes,161 in conjunction with anatomic imaging, is likely to assume more significance in radiation treatment planning and delivery. These techniques may enable us to define more precisely the extent of active tumor, which would allow more precise targeting of spatially varying radiation doses. ­Newer applications of four-dimensional CT imaging during and after the course of radiation therapy162 may provide data that can be correlated with treatment planning CT images to obtain more accurate assessments of pulmonary function. Images obtained during the course of radiation therapy may allow pulmonary function to be incorporated into a conformal avoidance treatment planning schema.

Proton Therapy The upsurge of interest in proton therapy is leading to efforts to identify new areas of research. Proton therapy has been used for the past several decades, but the number of scientists and clinicians engaged in research and development of its technologies has been limited. This is likely to change rapidly. In theory, protons are superior to photons and electrons because of their ability to produce more conformal dose patterns. However, the physical and biologic uncertainties associated with proton dose distributions and the differences in the ways motion and positioning uncertainties manifest themselves in proton therapy present considerable opportunity for research and development to further improve proton therapy. The development of intensity- and energy-modulated proton therapy is expected to have a substantial effect on optimizing proton therapy. As the results of research and development become available and as the use of proton therapy increases in the coming years, technology to deliver proton therapy is expected to undergo major evolution. Advances such as these are necessary to demonstrate the true potential of protons compared with photons and electrons. Expected advances in proton therapy from the biologic and clinical perspectives are discussed further in Chapter 40.

REFERENCES   1. International Commission on Radiation Units and Measurements. Prescribing, Recording, and Reporting Photon Beam Therapy. ICRU Report 50. Bethesda, MD: International Commission on Radiation Units and Measurements, 1993.

87

  2. International Commission on Radiation Units and Measurements. Prescribing, Recording, and Reporting Photon Beam Therapy (supplement to ICRU report 50). ICRU Report 62. Bethesda, MD: International Commission on Radiation Units and Measurements, 1999.   3. Giraud P, Antoine M, Larrouy A, et al. Evaluation of microscopic tumor extension in non–small-cell lung cancer for three­dimensional conformal radiotherapy planning. Int J Radiat ­Oncol Biol Phys 2000;48:1015-1024.   4. Reft C, Alecu R, Gerbi BJ, et al. Dosimetric considerations for patients with hip prostheses undergoing pelvic irradiation. ­Report of the AAPM Radiation Therapy Committee Task Group 63. Med Phys 2003;30:1162-1182.   5. Pan T, Mawlawi O, Nehmeh SA, et al. Attenuation correction of PET images with respiration-averaged CT images in PET/CT. J Nucl Med 2005;46:1481-1487.   6. Pouliot J, Bani-Hashemi A, Chen J, et al. Low-dose megavoltage cone-beam CT for radiation therapy. Int J Radiat Oncol Biol Phys 2005;61:552-560.   7. Mackie TR, Scrimger JW, Battista JJ. A convolution ­ method of calculating dose for 15 MV x-rays. Med Phys 1985;12: 188-196.   8. Boyer AL, Mok EC. A photon dose distribution model employing convolution calculations. Med Phys 1985;12:169-177.   9. Starkschall G, Steadham RE, Popple RA, et al. Beam commissioning methodology for a 3-D convolution/superposition photon dose algorithm. J Appl Clin Med Phys 2000;1:8-27. 10. Frank SJ, Forster KM, Stevens CW, et al. Treatment planning for lung cancer: traditional homogeneous point-dose prescription compared with heterogeneity-corrected dose-volume prescription. Int J Radiat Oncol Biol Phys 2003;56:1308-1318. 11. Mohan R, Mageras GS, Baldwin B, et al. Clinically relevant optimization of 3-D conformal treatments. Med Phys 1992;19: 933-944. 12. Brahme A. Optimized radiation therapy based on radiobiological objectives. Semin Radiat Oncol 1999;9:35-47. 13. Wu Q, Mohan R, Niemierko A, et al. Optimization of intensitymodulated radiotherapy plans based on the equivalent uniform dose. Int J Radiat Oncol Biol Phys 2002;52:224-235. 14. Graham MV, Purdy JA, Emami B, et al. Clinical dose-volume histogram analysis for pneumonitis after 3D treatment for non– small cell lung cancer (NSCLC). Int J Radiat Oncol Biol Phys 1999;45:323-329. 15. Galvin JM, Chen XG, Smith RM. Combining multileaf fields to modulate fluence distributions. Int J Radiat Oncol Biol Phys 1993;27:697-705. 16. Chang SX, Cullip TJ, Deschesne KM. Intensity modulation delivery techniques: “step & shoot” MLC auto-sequence versus the use of a modulator. Med Phys 2000;27:948-959. 17. Mackie TR, Holmes T, Swerdloff S, et al. Tomotherapy: a new concept for the delivery of dynamic conformal radiotherapy. Med Phys 1993;20:1709-1719. 18. American Medical Association. CPT 2005 (Current Procedural Terminology). Chicago: American Medical Association Press, 2005. 19. Fraass BA, Doppke KP, Hunt MA, et al. Task Group 53: quality assurance for clinical radiotherapy treatment planning. Med Phys 1998;25:1773-1829. 20. Low DA, Harms WB, Mutic S, et al. A technique for the quantitative evaluation of dose distributions. Med Phys 1998;25: 656-661. 21. Palta JR, Kim S, Li JG, et al. Tolerance limits and action levels for planning and delivery of IMRT. In Palta JR, Mackie TR (eds): Intensity-Modulated Radiation Therapy: The State of the Art. Madison, WI: Medical Physics Publishing, 2003, pp 593-612. 22. Leksell L. The stereotactic method and radiosurgery of the brain. Acta Chir Scand 1951;102:316. 23. Karpinos M, The BS, Zeck O, et al. Treatment of acoustic ­neuroma: stereotactic radiosurgery vs microsurgery. Int J Radiat Oncol Biol Phys 2002;54:1410-1421.

88

PART 1  Principles

24. Kjellberg RN, Preston WM. The use of the Bragg peak of a proton beam for intracranial lesions. Proceedings of the ­ Second International Congress on Neurological Surgery, 1961, ­Washington, DC. 25. Leksell L. Stereotactic Radiosurgery: An Operative System. Springfield, IL: Charles C Thomas, 1971. 26. Lutz W, Winston KR, Maleki N. A system for stereotactic ­radiosurgery with a linear accelerator. Int J Radiat Oncol Biol Phys 1988;14:373-381. 27. Schell MC, Bova FJ, Larson DA, et al. Stereotactic Radiosurgery. Report of Task Group 42 Radiation Therapy Committee, AAPM Report #54, 1995. 28. Kooy H, Nedzi L, Loeffler JS, et al. Treatment planning for stereotactic radiosurgery of intracranial lesions. Int J Radiat Oncol Biol Phys 1991;21:683-693. 29. van Der Linden YM, Lok JJ, Steenland E, et al, for the Dutch Bone Metastasis Study Group. Single fraction radiotherapy is ­ efficacious: a further analysis of the Dutch Bone Metastasis Study controlling for the influence of retreatment. Int J Radiat Oncol Biol Phys 2004;59:528-537. 30. Mithal NP. Retreatment with radiotherapy for painful bone ­metastases. Int J Radiat Oncol Biol Phys 1994;29:1011-1014. 31. Shiu AS, Chang EL, Ye JS, et al. Near simultaneous computed tomography image-guided stereotactic spinal radiotherapy: an emerging paradigm for achieving true stereotaxy. Int J Radiat Oncol Biol Phys 2003;57:605-613. 32. Yenice KM, Lovelock DM, Hunt MA, et al. CT image-guided intensity-modulated therapy for paraspinal tumors using stereotactic immobilization. Int J Radiat Oncol Biol Phys 2003;55: 583-593. 33. Hayman JA, Martel MK, Ten Haken RK, et al. Dose escalation in non–small cell lung cancer using three-dimensional conformal radiation therapy: update of a phase I trial. J Clin Oncol 2001;19:127-136. 34. Bradley JD, Graham MV, Winter KW, et al. Acute and late toxicity results of RTOG 9311: a dose escalation study using 3D conformal radiation therapy in patients with inoperable non– small cell lung cancer [abstract]. Int J Radiat Oncol Biol Phys 2003;57:S137-S138. 35. Onishi H, Araki T, Shirato H, et al. Stereotactic ­hypofractionated high-dose irradiation for stage I nonsmall cell lung carcinoma: clinical outcomes in 245 subjects in a Japanese multi-­institutional study. Cancer 2004;101:1623-1631. 36. Pan T, Lee TY, Rietzel E, et al. 4D-CT imaging of a volume influenced by respiratory motion on multi-slice CT. Med Phys 2004;31:333-340. 37. Hanley J, Debois MM, Mah D, et al. Deep inspiration breathhold technique for lung tumors: the potential value of target immobilization and reduced lung density in dose escalation. Int J Radiat Oncol Biol Phys 1999;45:603-611. 38. Hall EJ. Radiobiology for the Radiologist. 4th ed. Philadelphia: JB Lippincott, 1994, p 224. 39. Ross CS, Hussey DH, Pennington EC, et al. Analysis of movement of intrathoracic neoplasms using ultrafast computerized tomography. Int J Radiat Oncol Biol Phys 1990;18:671-677. 40. Korin HW, Ehman RL, Riederer SJ, et al. Respiratory kinematics of the upper abdominal organs: a quantitative study. Magn Reson Med 1992;23:172-178. 41. Davies SC, Hill AL, Holmes RB, et al. Ultrasound quantitation of respiratory organ motion in the upper abdomen. Br J Radiol 1994;67:1096-1102. 42. Kubo HD, Hill BC. Respiration gated radiotherapy treatment: a technical study. Phys Med Biol 1996;41:83-91. 43. Seppenwoolde Y, Shirato H, Kitamura K, et al. Precise and realtime measurement of 3D tumor motion in lung due to breathing and heartbeat, measured during radiotherapy. Int J Radiat Oncol Biol Phys 2002;53:822-834. 44. Grass M, Manzke R, Nielsen T, et al. Helical cardiac cone beam reconstruction using retrospective ECG gating. Phys Med Biol 2003;48:3069-3084. 45. Keall PJ, Starkschall G, Shukla H, et al. Acquiring 4D ­thoracic CT scans using a multislice helical method. Phys Med Biol 2004;49:2053-2067.

46. Liu HH, Balter P, Tutt T, et al. Assessing respiration-induced ­tumor motion and internal target volume using four-dimensional computed tomography for radiotherapy of lung cancer. Int J ­Radiat Oncol Biol Phys 2007;68:531-540. 47. Wang H, Dong L, O’Daniel J, et al. Validation of an accelerated “demons” algorithm for deformable image registration in radiation therapy. Phys Med Biol 2005;50:2887-2905. 48. Starkschall G, Forster K, Fitzpatrick MJ, et al. Correspondence of respiratory-correlated CT data sets acquired using 4-D CT with data sets acquired using occlusion spirometry [abstract]. Med Phys 2004;31:1875. 49. Balter P, Luo D, Chi P, et al. A comparison of 4DCT with breath-hold CT for determination of tumor motion with respiration [abstract]. Med Phys 2005;32:1919. 50. Shirato H, Shizumi S, Kitamura K, et al. Four-dimensional treatment planning and fluoroscopic real-time tumor tracking radiotherapy for moving tumors. Int J Radiat Oncol Biol Phys 2000;48:435-442. 51. Butler LE, Forster KM, Stevens CW, et al. Dosimetric benefits of respiratory gating: a preliminary study. J Appl Clin Med Phys 2004;5:1-9. 52. Nelson C, Starkschall G, Chang JY. The potential for dose escalation resulting from reduction in margins used to generate the planning target volume. Int J Radiat Oncol Biol Phys 2006;65:573-586. 53. Starkschall G, Forster KM, Kitamura K, et al. Correlation of GTV excursion with the desirability of respiratory gating. Int J Radiat Oncol Biol Phys 2004;60:1291-1297. 54. Nelson C, Starkschall G, Balter P, et al. Assessment of lung ­tumor motion and setup uncertainties using implanted fiducials. Int J Radiat Oncol Biol Phys 2007;67:915-923. 55. Nelson C, Starkschall G, Balter P, et al. Respiration-correlated treatment delivery using feedback-guided breath hold: a technical study. Med Phys 2005;32:175-181. 56. Balter JM, Sandler HM, Lam K, et al. Measurement of prostate movement over the course of routine radiotherapy ­ using implanted markers. Int J Radiat Oncol Biol Phys 1995;31: 113-118. 57. Jaffray DA, Yan D, Wong JW. Managing geometric ­uncertainty in conformal intensity-modulated radiation therapy. Semin ­Radiat Oncol 1999;9:4-19. 58. Langen KM, Jones DT. Organ motion and its management. Int J Radiat Oncol Biol Phys 2001;50:265-278. 59. Yan D, Lockman D. Organ/patient geometric variation in external beam radiotherapy and its effects. Med Phys 2001;28: 593-602. 60. Stroom JC, Heijmen BJ. Geometrical uncertainties, ­radiotherapy planning margins, and the ICRU-62 report. Radiother Oncol 2002;64:75-83. 61. Mackie TR, Tome W, Welsh J, et al. Image guidance for precise conformal radiotherapy. Int J Radiat Oncol Biol Phys 2003;56:89-105. 62. Goitein M. Organ and tumor motion: an overview. Semin Radiat Oncol 2004;14:2-9. 63. van Herk M. Errors and margins in radiotherapy. Semin Radiat Oncol 2004;14:52-64. 64. MacKay RI, Graham PA, Logue JP, et al. Patient positioning ­using detailed three-dimensional surface data for patients undergoing conformal radiation therapy for carcinoma of the ­prostate: a feasibility study. Int J Radiat Oncol Biol Phys 2001;49: 225-230. 65. Chandra A, Dong L, Huang E, et al. Experience of ultrasoundbased daily prostate localization. Int J Radiat Oncol Biol Phys 2003;56:436-447. 66. Booth JT, Zavgorodni SF. Set-up error and organ motion uncertainty: a review. Australas Phys Eng Sci Med 1999;22: 29-47. 67. Nuyttens JJ, Robertson JM, Yan D, et al. The variability of the clinical target volume for rectal cancer due to internal organ ­motion during adjuvant treatment. Int J Radiat Oncol Biol Phys 2002;53:497-503.

2  Clinical Radiation Oncology Physics 68. Pos FJ, Koedooder K, Hulshof M, et al. Influence of bladder and rectal volume on spatial variability of a bladder tumor during radical radiotherapy. Int J Radiat Oncol Biol Phys 2003;55: 835-841. 69. Dawson LA, Balter JM. Interventions to reduce organ motion e­ffects in radiation delivery. Semin Radiat Oncol 2004;14: 76-80. 70. Barker JL Jr, Garden AS, Ang KK, et al. Quantification of volumetric and geometric changes occurring during fractionated radiotherapy for head-and-neck cancer using an ­integrated CT/linear accelerator system. Int J Radiat Oncol Biol Phys 2004;59:960-970. 71. Purdy JA. Advances in three-dimensional treatment planning and conformal dose delivery. Semin Radiat Oncol 1997;24: 655-671. 72. Zelefsky MJ, Fuks Z, Leibel SA. Intensity-modulated radiation therapy for prostate cancer. Semin Radiat Oncol 2002;12:229-237. 73. Ezzell GA, Galvin JM, Low D, et al. Guidance document on delivery, treatment planning, and clinical implementation of IMRT: report of the IMRT subcommittee of the AAPM radiation therapy committee. Med Phys 2003;30:2089-2115. 74. Stroom JC, de Boer HC, Huizenga H, et al. Inclusion of geometrical uncertainties in radiotherapy treatment planning by means of coverage probability. Int J Radiat Oncol Biol Phys 1999;43:905-919. 75. van Herk M, Remeijer P, Rasch C, et al. The probability of correct target dosage: dose-population histograms for deriving treatment margins in radiotherapy. Int J Radiat Oncol Biol Phys 2000;47:1121-1135. 76. Hector CL, Webb S, Evans PM. The dosimetric consequences of inter-fractional patient movement on conventional and ­intensity-modulated breast radiotherapy treatments. Radiother Oncol 2000;54:57-64. 77. Li JG, Xing L. Inverse planning incorporating organ motion. Med Phys 2000;27:1573-1578. 78. Samuelsson A, Mercke C, Johansson KA. Systematic set-up ­errors for IMRT in the head and neck region: effect on dose distribution. Radiother Oncol 2003;66:303-311. 79. Verhey LJ, Goitein M, McNulty P, et al. Precise positioning of patients for radiation therapy. Int J Radiat Oncol Biol Phys 1982;82:289-294. 80. Meertens H, van Herk M, Weeda J. A liquid ionisation detector for digital radiography of therapeutic megavoltage photon beams. Phys Med Biol 1985;30:313-321. 81. Lam KS, Partowmah M, Lam WC. An on-line electronic portal imaging system for external beam radiotherapy. Br J Radiol 1986;59:1007-1013. 82. Munro P, Rawlinson JA, Fenster A. Therapy imaging: a ­­signal-tonoise analysis of metal plate/film detectors. Med Phys 1987;14: 975-984. 83. van Herk M, Meertens H. A matrix ionisation chamber imaging device for on-line patient setup verification during radiotherapy. Radiother Oncol 1988;11:369-378. 84. Lattanzi J, McNeeley S, Donnelly S, et al. Ultrasound-based stereotactic guidance in prostate cancer—quantification of ­organ motion and set-up errors in external beam radiation therapy. Comput Aided Surg 2000;5:289-295. 85. Morr J, DiPetrillo T, Tsai JS, et al. Implementation and utility of a daily ultrasound-based localization system with intensitymodulated radiotherapy for prostate cancer. Int J Radiat Oncol Biol Phys 2002;53:1124-1129. 86. Serago CF, Chungbin SJ, Buskirk SJ, et al. Initial experience with ultrasound localization for positioning prostate cancer patients for external beam radiotherapy. Int J Radiat Oncol Biol Phys 2002;53:1130-1138. 87. Kuban DA, Dong L, Cheung R, et al. Ultrasound-based localization. Semin Radiat Oncol 2005;15:180-191. 88. Fung AYC, Enke CA, Ayyangar KM, et al. Prostate motion and isocenter adjustment from ultrasound-based localization during delivery of radiation therapy. Int J Radiat Oncol Biol Phys 2005;61:984-992.

89

89. Little DJ, Dong L, Levy LB, et al. Use of portal images and BAT ultrasonography to measure setup error and organ motion for prostate IMRT: implications for treatment margins. Int J Radiat Oncol Biol Phys 2003;56:1218-1224. 90. Court LE, Dong L. Automatic registration of the prostate for computed-tomography-guided radiotherapy. Med Phys 2003;30:2750-2757. 91. Langen KM, Pouliot J, Anezinos C, et al. Evaluation of ultrasound-based prostate localization for image-guided ­radiotherapy. Int J Radiat Oncol Biol Phys 2003;57:635-644. 92. Litzenberg D, Dawson LA, Sandler H, et al. Daily prostate targeting using implanted radiopaque markers. Int J Radiat Oncol Biol Phys 2002;52:699-703. 93. Trichter F, Ennis RD. Prostate localization using transabdominal ultrasound imaging. Int J Radiat Oncol Biol Phys 2003;56: 1225-1233. 94. Artignan X, Smitsmans MH, Lebesque JV, et al. Online ultrasound image guidance for radiotherapy of prostate cancer: impact of image acquisition on prostate displacement. Int J Radiat Oncol Biol Phys 2004;59:595-601. 95. Dobler B, Mai S, Ross C, et al. Evaluation of possible prostate displacement induced by pressure applied during transabdominal ultrasound image acquisition. Strahlenther Onkol 2006;182:240-246. 96. Zhang LF, Garden AS, Lo J, et al. Multiple regions-of-interest analysis of setup uncertainties for head-and-neck cancer radiotherapy. Int J Radiat Oncol Biol Phys 2006;64:1559-1569. 97. Court L, Rosen I, Mohan R, et al. Evaluation of mechanical precision and alignment uncertainties for an integrated CT/LINAC system. Med Phys 2003;30:1198-1210. 98. Zhang L, Dong L, Court L, et al. Validation of CT-assisted targeting (CAT) software for soft tissue and bony target localization [abstract]. Med Phys 2005;32:2106. 99. Yan D, Jaffray DA, Wong JW. A model to accumulate fractionated dose in a deforming organ. Int J Radiat Oncol Biol Phys 1999;44:665-675. 100. Lockman DM, Yan D, Wong J. Estimating the dose variation in a volume of interest with explicit consideration of patient geometric variation. Med Phys 2000;27:2100-2108. 101. Lu W, Chen ML, Olivera GH, et al. Fast free-form deformable registration via calculus of variations. Phys Med Biol 2004;49:3067-3087. 102. Wang H, Dong L, Mohan R, et al. Implementation and validation of a deformable image registration algorithm for in-room CT-guided radiotherapy [abstract]. Med Phys 2004;31:1791. 103. Wang P, Lovelock D, Joshi S, et al. Evaluation of an automated deformable registration algorithm for localizing the prostate in serial CT image sets [abstract]. Med Phys 2004;31:1791. 104. Brock KK, Sharpe MB, Dawson LA, et al. Accuracy of finite element model-based multi-organ deformable image registration. Med Phys 2005;32:1647-1659. 105. Wang H, Dong L, Lii MF, et al. Implementation and validation of a three-dimensional deformable registration algorithm for targeted prostate cancer radiotherapy. Int J Radiat Oncol Biol Phys 2005;61:725-735. 106. Eifel PJ, Morris M, Delclos L. Radiation therapy for cervical carcinoma. In Dilts P, Sciarra J (eds): Gynecology and Obstetrics, 4th ed. Philadelphia: JB Lippincott, 1993, pp 1-25. 107. Katz A, Eifel PJ. Quantification of intracavitary brachytherapy parameters and correlation with outcome in patients with carcinoma of the cervix. Int J Radiat Oncol Biol Phys 2000;48: 1417-1425. 108. Van Eijkeren M, Thienpont M, Boterberg T, et al. ­Implementation, uptime, and safety of a pulsed dose rate afterloading ­ machine. Int J Radiat Oncol Biol Phys 1996;36:1233-1237. 109. Koedooder K, Tiggelaar R, Van Der Grient H, et al. Evaluation of technical failures during treatment in pulsed dose rate brachytherapy [abstract]. Radiother Oncol 2004;71:S80-S81. 110. Brenner DJ, Hall EJ. Conditions for the equivalence of continuous to pulsed low dose rate brachytherapy. Int J Radiat Oncol Biol Phys 1991;20:181-190.

90

PART 1  Principles

111. Horton J, Lawyer A, Mourtada F. Continuous low dose rate and pulsed dose rate remote afterloader units. In Thomadsen B, ­Rivard M, Bulter W (eds): Brachytherapy Physics. Madison, WI: Medical Physics Publishing, 2005, pp 99-108. 112. National Council on Radiation Protection and Measurements. Protection Against Radiation from Brachytherapy Sources. NCRP report no. 40. Bethesda, MD: National Council on Radiation Protection and Measurements, 1972. 113. CBTRUS (2002-2003). Primary Brain Tumors in the United States Statistical Report 1995-1999. Central Brain Tumor Registry of the United States. Available at http://www.cbtrus.org ­(accessed January 2009). 114. Stubbs JB, Strickland AD, Frank RK, et al. Biodistribution and dosimetry of an aqueous solution containing sodium 3-(125I)­ iodo-4-hydroxybenzenesulfonate (Iotrex) for brachytherapy of resected malignant brain tumors. Cancer Biother ­ Radiopharm 2000;15:645-656. 115. Strzelczyk J, Safadi R. Radiation safety considerations in GliaSite 125I brain implant procedures. Health Phys 2004;86;(Suppl): S120-S123. 116. Matheus MG, Castillo M, Ewend M, et al. CT and MR imaging after placement of the GliaSite radiation therapy system to treat brain tumor: initial experience. AJNR Am J Neuroradiol 2004;25:1211-1217. 117. Monroe JI, Dempsey JF, Dorton JA, et al. Experimental validation of dose calculation algorithms for the GliaSite RTS, a novel 125I liquid-filled balloon brachytherapy applicator. Med Phys 2001;28:73-85. 118. Gabayan AJ, Green SB, Sanan A, et al. GliaSite brachytherapy for treatment of recurrent malignant gliomas: a retrospective multi-institutional analysis. Neurosurgery 2006;58:701-709. 119. Tatter SB, Shaw EG, Rosenblum ML etal, for The New ­Approaches to Brain Tumor Therapy Central Nervous System Consortium. An inflatable balloon catheter and liquid 125I radiation source (GliaSite Radiation Therapy System) for treatment of recurrent malignant glioma: multicenter safety and feasibility trial. J Neurosurg 2003;99:297-303. 120. Chan TA, Weingart JD, Parisi M, et al. Treatment of recurrent glioblastoma multiforme with GliaSite brachytherapy. Int J ­Radiat Oncol Biol Phys 2005;62:1133-1139. 121. American Cancer Society. Cancer Facts and Figures 2007. ­Atlanta, GA: American Cancer Society, 2007. Available at http://www.cancer.org/downloads/STT/CAFF2007PWSecured. pdf (accessed January 2009). 122. Keisch M, Vicini F, Kuske RR, et al. Initial clinical experience with the MammoSite breast brachytherapy applicator in ­women with early-stage breast cancer treated with breast-conserving therapy. Int J Radiat Oncol Biol Phys 2003;55:289-293. 123. Keisch M, Vicini F, Kuske RR, et al. Two-year outcome with the MammoSite breast brachytherapy applicator: factors associated with optimal cosmetic results when performing partial breast irradiation [abstract]. Int J Radiat Oncol Biol Phys 2003;57:S315. 124. Hogle WP, Quinn AE, Heron DE. Advances in brachytherapy: new approaches to target breast cancer. Clin J Oncol Nurs 2003;7:324-328. 125. Keisch M, Arthur DW. Current perspective on the MammoSite Radiation Therapy System—a balloon breast brachytherapy ­applicator. Brachytherapy 2005;4:177-180. 126. Astrahan MA, Jozsef G, Streeter OE Jr. Optimization of MammoSite therapy. Int J Radiat Oncol Biol Phys 2004;58:220-232. 127. Edmundson GK, Vicini FA, Chen PY, et al. Dosimetric characteristics of the MammoSite RTS, a new breast brachytherapy applicator. Int J Radiat Oncol Biol Phys 2002;52:1132-1139. 128. Kassas B, Mourtada F, Horton JL, et al. Contrast effects on dosimetry of a partial breast irradiation system. Med Phys 2004;31:1976-1979. 129. Ye SJ, Brezovich IA, Shen S, et al. Dose errors due to inhomogeneities in balloon catheter brachytherapy for breast cancer. Int J Radiat Oncol Biol Phys 2004;60:672-677. 130. Kirk MC, Hsi WC, Chu JC, et al. Dose perturbation induced by radiographic contrast inside brachytherapy balloon applicators. Med Phys 2004;31:1219-1224.

131. Cheng CW, Mitra R, Li XA, et al. Dose perturbations due to contrast medium and air in Mammosite treatment: an experimental and Monte Carlo study. Med Phys 2005;32:2279-2287. 132. Gragoudas ES. The Bragg peak of proton beams for treatment of uveal melanoma. Int Ophthalmol Clin 1980;20:123-133. 133. Brady LW, Hernandez JC. Brachytherapy of choroidal melanomas. Strahlenther Onkol 1992;168:61-65. 134. Marchini G, Gerosa M, Piovan E, et al. Gamma knife stereotactic radiosurgery for uveal melanoma: clinical results after 2 years. Stereotact Funct Neurosurg 1996;66:208-213. 135. Finger PT, Berson A, Ng T, et al. Palladium-103 plaque radiotherapy for choroidal melanoma: an 11-year study. Int J Radiat Oncol Biol Phys 2002;54:1438-1445. 136. Melia BM, Abramson DH, Albert DM, 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;108:348-366. 137. Lommatzsch PK, Werschnik C, Schuster E. Long-term follow-up of Ru-106/Rh-106 brachytherapy for posterior uveal melanoma. Graefes Arch Clin Exp Ophthalmol 2000;238:129-137. 138. Wilson MW, Hungerford JL. Comparison of episcleral plaque and proton beam radiation therapy for the treatment of choroidal melanoma. Ophthalmology 1999;106:1579-1587. 139. Soares CG, Bergstrom PJ. The divergence correction for extrapolation chambers measuring brachytherapy sources [abstract]. Med Phys 2003;30:1427. 140. Isager P, Ehlers N, Urbak SF, et al. Visual outcome, local tumour control, and eye preservation after 106Ru/Rh brachytherapy for choroidal melanoma. Acta Oncol 2006;45:285-293. 141. Hermann RM, Pradier O, Lauritzen K, et al. Does escalation of the apical dose change treatment outcome in beta-radiation of posterior choroidal melanomas with 106Ru plaques? Int J ­Radiat Oncol Biol Phys 2002;52:1360-1366. 142. Potter R, Janssen K, Prott FJ, et al. Ruthenium-106 eye plaque brachytherapy in the conservative treatment of uveal melanoma: evaluation of 175 patients treated with 150 Gy from 1981-1989. Front Radiat Ther Oncol 1997;30:143-149. 143. Damato B, Patel I, Campbell IR, et al. Local tumor control ­after 106Ru brachytherapy of choroidal melanoma. Int J Radiat ­Oncol Biol Phys 2005;63:385-391. 144. Damato B, Patel I, Campbell IR, et al. Visual acuity after Ruthenium(106) brachytherapy of choroidal melanomas. Int J Radiat Oncol Biol Phys 2005;63:392-400. 145. Bergman L, Nilsson B, Lundell G, et al. Ruthenium brachytherapy for uveal melanoma, 1979-2003: survival and functional outcomes in the Swedish population. Ophthalmology 2005;112:834-840. 146. Seregard S. Long-term survival after ruthenium plaque radiotherapy for uveal melanoma. A meta-analysis of studies including 1,066 patients. Acta Ophthalmol Scand 1999;77:414-417. 147. Summanen P, Immonen I, Kivela T, et al. Radiation related complications after ruthenium plaque radiotherapy of uveal melanoma. Br J Ophthalmol 1996;80:732-739. 148. Mourtada F, Koch N, Newhauser W. 106Ru/106Rh plaque and proton radiotherapy for ocular melanoma: a comparative dosimetric study. Radiat Prot Dosimetry 2005;116:454-460. 149. Urie M, Goitein M, Wagner M. Compensating for heterogeneities in proton radiation therapy. Phys Med Biol 1984;29: 553-566. 150. Kang Y, Zhang X, Chang JY, et al. 4D Proton treatment planning strategy for mobile lung tumors. Int J Radiat Oncol Biol Phys 2007;67:906-914. 151. Engelsman M, Kooy H. Target volume dose considerations in proton beam treatment planning for lung tumors. Med Phys 2005;32:3549-3557. 152. Engelsman M, Rietzel E, Kooy HM. Four-dimensional proton treatment planning for lung tumors. Int J Radiat Oncol Biol Phys 2006;64:1589-1595. 153. Gottschalk B, Koehler AM, Schneider RJ, et al. Multiple Coulomb scattering of 160 MeV protons. Nucl Instrum Methods Phys Res B 1993;74:467-490.

2  Clinical Radiation Oncology Physics 154. Hong L, Goitein M, Bucciolini M, et al. A pencil beam algorithm for proton dose calculations. Phys Med Biol 1996;41:1305-1330. 155. Schaffner B, Pedroni E, Lomax A. Dose calculation models for proton treatment planning using a dynamic beam delivery system: an attempt to include density heterogeneity effects in the analytical dose calculation. Phys Med Biol 1999;44:27-41. 156. Kanai T, Kawachi K, Kumamoto Y, et al. Spot scanning system for proton therapy. Med Phys 1980;7:365-369. 157. Pedroni E, Bacher R, Blattman H, et al. The 200-MeV proton therapy project at the Paul Scherrer Institute: conceptual design and practical realization. Med Phys 1995;22:37-53. 158. Pedroni E, Scheib S, Bohringer T, et al. Experimental characterization and physical modeling of the dose distribution of scanned proton pencil beams. Phys Med Biol 2005;50:541-561.

91

159. Paganetti H, Niemierko A, Ancukiewicz M, et al. Relative biological effectiveness (RBE) values for proton beam therapy. Int J Radiat Oncol Biol Phys 2002;53:407-421. 160. Paganetti H. Significance and implementation of RBE variations in proton beam therapy. Technol Cancer Res Treat 2003;2:413-426. 161. Apisarnthanarax S, Chao KSC. Current imaging paradigms in radiation oncology. Radiat Res 2005;163:1-25. 162. Guerrero T, Sanders K, Castillo E, et al. Dynamic ventilation ­imaging from four-dimensional computer tomography. Phys Med Biol 2006;51:777-791.

3

Principles of Combining Radiation Therapy and Surgery Gunar K. Zagars THE SIGNIFICANCE OF LOCOREGIONAL TUMOR CONTROL THEORETICAL BASIS FOR COMBINING SURGERY AND RADIATION THERAPY SUBCLINICAL DISEASE Mechanisms and Implications of Spread Radiation Biology of Subclinical Disease TOPOGRAPHIC RELATIONSHIPS BETWEEN SURGERY AND RADIATION THERAPY PREOPERATIVE VERSUS POSTOPERATIVE RADIATION THERAPY Preoperative Radiation Therapy Postoperative Radiation Therapy

The Significance of Locoregional Tumor Control Surgery and radiation therapy are locoregional treatment modalities. When they are successful, they permanently eradicate the primary tumor and regional nodal metastatic disease. Locoregional control of a tumor is an obvious ­­prerequisite for cure. Conversely, failure to achieve local control not only signifies failure to cure but also leaves the patient to face a variety of debilitating locoregional problems that ultimately may become insoluble and may be fatal if death from distant metastases or intercurrent disease does not occur first. These virtues of locoregional control are self-evident; less conclusive is the potential salutary effect of effective regional treatment on the subsequent evolution of metastatic disease. The potential presence of micrometastases complicates any evaluation of the effect of local treatment on the subsequent appearance of clinically evident metastases. The problem hinges on the time at which a primary tumor usually seeds its first metastasis—only in tumors that initiate the metastatic cascade beyond the time of diagnosis and treatment can locoregional therapy have any effect on metastatic evolution by removing the source of potential metastatic dissemination. The time at which tumors first successfully shed metastases is unknown but clearly is highly variable, as reflected in the clinical recognition of a spectrum of metastatic potential from highly metastasizing cancers such as small cell lung cancer to infrequently metastasizing tumors such as dermatofibrosarcoma protuberans. Moreover, the time of first metastasis varies among patients with the same histopathologic tumor; although most women with T1 breast cancer are free from clinically detectable metastases at diagnosis, some with ostensibly similar primary tumors already have overt metastatic disease. Despite these uncertainties, the high correlation between the size or extent of a primary tumor (T category) and the likelihood of subsequent metastatic relapse even

92

Time Interval between Treatment Modalities Intraoperative or Interstitial Radiation Therapy CLINICAL EXPERIENCES COMPLICATIONS GENERAL guidelines AND CONCLUSIONS Principles of Combining Radiation Therapy and Surgery

after successful local treatment is prima facie evidence that the shedding of a first metastasis often occurs at a time corresponding to the growth of a T1 cancer to become a T4 cancer. Radiobiologic evaluation of the response of hematogenous micrometastases to radiation also supports the idea that in a significant proportion of human cancer, the primary tumor seeds its first metastasis at or just before the time of its diagnosis.1 The continuing drive for early diagnosis and prompt ­local treatment is based on the premise that early intervention eradicates the primary tumor and removes the source for subsequent metastatic dissemination, achieving higher cure and survival rates than delayed treatment. This concept has been epidemiologically validated for breast,2 cervix,3 and colorectal4 cancer. In keeping with these observations are many retrospective after-treatment studies in which a significant positive correlation between local recurrence and subsequent metastatic relapse has been found.5-10 However, this finding has not been consistent, particularly among prospective randomized trials, in which improvement in local control often has not translated into significant improvements in freedom from metastasis or survival.11-15 Prospective randomized trials are reliable arbiters for many controversies in therapeutics, but care must be taken in drawing inferences about end points for which such trials were not designed. Virtually all trials conducted to evaluate locoregional treatment alternatives are designed to detect differences in locoregional control. It is almost certain that such trials will fail to detect real differences in metastatic outcome between the evaluated treatment groups. A simple calculation demonstrates this notion. Suppose that, for the tumor in question, local control is associated with a true metastatic incidence of 30% (metastases seeded before treatment) and local relapse results in a true metastatic rate of 60% ­(including the excess metastases seeded from the locally persistent tumor). If one local treatment achieves an 80% local control rate and another achieves only 60%, the ­difference

3  Principles of Combining Radiation Therapy and Surgery

will be apparent in a randomized trial of approximately 200 patients (100 in each treatment group). Simple calculation reveals that the superior treatment will be associated with a 36% true incidence of metastatic relapse compared with a 42% incidence for the poorer treatment; this difference would require some 2000 patients (1000 in each group) to have an 80% chance of being detected as significant, despite the fact that the patient whose tumor recurs has a twofold excess risk of metastatic relapse. Because it would be unethical to continue accruing patients to a study in which one treatment was demonstrably superior over another for locoregional disease control, randomized prospective trials provide no conclusive evidence on the relationship between local recurrence and metastatic dissemination. Retrospective analyses that compare the correlation between local recurrence and metastasis are bedeviled by the possibility that the tumors that recur may be those whose pathobiology also promotes early dissemination independently of any treatment effect. This is true to the extent that for many types of cancer, the pretreatment prognostic factors that correlate with locoregional relapse also correlate with metastatic dissemination. Nevertheless, given what is known about tumor biology and clinical outcomes, the onus of proof that local control has no effect on the likelihood of subsequent metastasis rests with those who uphold such a view. Common sense suggests that achievement of locoregional control avoids the problems of persistently growing local cancer and abrogates the metastatic process, resulting in a lower likelihood of metastatic relapse in some proportion of patients.

Theoretical Basis for Combining Surgery and Radiation Therapy Surgery and radiation can be combined in two fundamen­ tally different ways: in a salvage fashion, in which surgery is used after irradiation fails or irradiation is used after surgery fails, or in a planned, combined bimodality fashion. The use of these modalities in a salvage fashion is not discussed further because it amounts to repeated single-modality treatment. The use of planned combinations of surgery and irradiation evolved empirically as experience revealed the limitations of each modality when used alone in a variety of clinical circumstances and as the complementarity in their virtues became evident.16-19 In parallel with clinical experience, experimental work in radiation and tumor biology has advanced the theoretical foundations for this combination, and it is increasingly influencing clinical practice.

Subclinical Disease Mechanisms and Implications of Spread Locally invasive tumors, including all forms of cancer, infiltrate tissues around their primary site. Although local extension can occur in any direction, it typically follows the path of least resistance that is characteristic of different tumor types in different anatomic sites. Common pathways of local spread include those along fascial planes, muscle fibers, nerve sheaths, and periosteum. Important aspects of this spread are its microscopic character and potentially wide extent, creating a variable volume of grossly normal but tumor-bearing tissue around the primary tumor. This

93

growth pattern is repeated at regional nodes harboring metastatic disease, if extracapsular extension occurs there, and at sites of hematogenous metastasis. The larger a primary tumor, the more extensive and dense the volume of microscopic infiltration because of the longer time the tumor is likely to have been present and because large tumors tend to be inherently more aggressive. Peripheral tumor infiltrates can contain any number of clonogenic cells from a few to as many as 109 or more before they become clinically evident. For example, if 106 cells, which occupy a volume of approximately 1 mm3, were distributed in a tissue volume of 1 cm3, they might not be detected, even under a microscope; if 109 cells, which occupy approximately 1 cm3, were clustered, they could be radiologically detected. However, in anatomic sites such as the abdomen, considerably larger deposits can be occult. Any tumor cell deposits or extensions that cannot be detected by clinical-radiographic means are called subclinical disease. Although the terms subclinical and microscopic ­often are used interchangeably, it is important to understand that subclinical disease is not always microscopic, although advances in tumor imaging are bringing the two terms closer together. This distinction has implications for clinical practice because many of the basic concepts relating to subclinical disease and radiation therapy were developed from experience with accessible and evaluable anatomic sites such as the head and neck, breast, and axilla, and they may not be directly applicable to less accessible areas such as the chest, abdomen, and pelvis, where subclinical disease can occupy a more substantial volume.18,20-23 The therapeutic problems posed by subclinical disease around a primary tumor are illustrated schematically in Figure 3-1. Surgery is a macroscopic, visual modality, and its selective therapeutic effect is for macroscopic tumor, with the surgeon guiding the scalpel around grossly evident tumor. Surgical removal of subclinical, microscopic tumor entails resection of tissue, of which much is normal—a poor therapeutic ratio. Nevertheless, the importance of locoregional microscopic disease is reflected in the classical

FIGURE 3-1.  The schematic drawing shows subclinical disease around a primary tumor. A, A surgically undisturbed tumor is surrounded by a zone of microscopic disease of decreasing density with increasing distance from the tumor. Dotted lines outline potential surgical resection planes. B, Resection of the tumor shown in A passed through the inner ellipse with ­approximation of the excision margins. A significant amount of residual ­ microscopic disease remains; complete surgical ­excision would require resection through the outer ellipse. The density of ­tumor cells in situations such as the one on the right is usually so low that even careful microscopic examination fails to reveal the residual cancer. As the microscopic disease grows, however, local recurrence will become clinically apparent. This ­situation is ­compounded for the worse in many epithelial ­cancers by ­metastatic regional nodal disease, when the ­problem is ­reiterated as tumor extends beyond the nodal capsule.

94

PART 1  Principles

Radiation Biology of Subclinical Disease Radiation has a selective effect against cancer. Although the mechanisms for this selectivity are poorly understood, it is the cornerstone of radiation oncology. Radiation doses that have relatively little untoward effect on most normal tissues can eradicate significant deposits of most types of cancer cells. This ideal therapeutic ratio, however, depends on the amount of cancer present. Because of the random nature of cell killing by radiation, the curve for local control versus dose (i.e., dose-response relationship) for any tumor is sigmoid,24 and its location on the dose axis depends on many factors but principally on clonogen number (Fig. 3-2). The larger the clonogen number, the greater the dose that is needed to achieve any level of local 100 90

104

106

108

1010

5×1010

80

control and the closer the dose-control relationship is to the dose-complication relationship. For large aggregates of tumor cells, radiation loses its therapeutic selectivity and tumor eradication requires destruction of normal tissue. Alternative ways of looking at the dependence between dose, tumor control, and clonogen number are illustrated in Figures 3-3 and 3-4. These theoretical conclusions have been extensively verified in clinical practice25 (Table 3-1) and are the fundamental basis for the use of combined surgery and radiation therapy. Surgery effectively removes large, bulky tumors that even high doses of radiation often fail to control, whereas microscopic deposits around the primary tumor and in regional lymph nodes can be effectively eradicated by modest doses of radiation with minimal effects on normal tissue.

Number of 2-Gy fractions

radical surgical techniques designed to deal with this problem. The Halsted radical mastectomy, the Wertheim radical hysterectomy, the classic radical neck dissection often combined with en bloc resection of the primary tumor (the so-called commando procedure), the abdominoperineal resection of Miles for low rectal cancer, and the radical monobloc soft-part resection for soft tissue sarcoma were all developed in an attempt to deal with the primary tumor and its locoregional extensions. The targeting of apparently normal tissues for therapy inherent in these radical surgical approaches leads to the concept of elective, adjuvant, or prophylactic treatment—treatment directed at disease that is assumed to be present. This assumption is based on prior knowledge that a significant fraction of patients with similar tumors had such disease. The application of radical surgical techniques significantly reduces but does not eliminate the likelihood of locoregional recurrence. These gains in local control are achieved at the expense of significant anatomic, functional, and cosmetic deficits. Moreover, even radical surgical procedures often fail to achieve local control in large tumors (T3 and T4) at many anatomic sites.

34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 105

106

107

108

109

Number of clonogens

FIGURE 3-3.  The theoretical relationship is depicted for the number of 2-Gy fractions and the number of clonogens, for which the goal is to achieve 90% local control. The relationship was modeled by assuming a surviving fraction of 0.5 after each fraction (SF2). The tumor control probability (TCP) is equal to e−SN, in which S is the surviving fraction after F fractions (S = [SF2]F), and N is the clonogen number. The straight line is the relationship if the dose were a continuous variable.

60

100

50

90

40

80

Gy)

y)

30

s (60 ction

(50 G

Radiation dose

FIGURE 3-2.  The theoretical relationship between tumor control probability (TCP) and radiation dose is depicted for different numbers of clonogenic cells (colored lines). One possible relationship between normal tissue complications and dose (dashed line) is shown. As the number of clonogenic cells increases, the dose-control curve shifts to the right, and higher doses are required to achieve control; a favorable therapeutic ratio becomes progressively less favorable as the clonogen number increases. For the two largest clonogen numbers illustrated, no selectivity exists, and irradiation becomes essentially equivalent to surgery (i.e., radiosurgery).

40

Gy)

0

50

a 30 fr

10

60

ns actio 25 fr

20

70

0 ns (4 actio 20 fr

30

Percent local control

TCP (%)

70

20 10 0 104

105

106

107

108

109

1010

Number of clonogens

FIGURE 3-4.  The theoretical relationship between local control and clonogen number is depicted for different total doses delivered at 2 Gy per fraction. The parameters are the same as in Figure 3-3.

95

3  Principles of Combining Radiation Therapy and Surgery

Dose (Delivered at 10 Gy/wk)

Percent Control for Various Tumors

50 Gy

90% for subclinical neck nodes About 50% for 2- to 3-cm neck nodes

60 Gy

80% to 90% for T1 pharynx and larynx

1010 109

W

105

W

104

W

103 102

10–1

About 70% for 3- to 5-cm neck nodes

0

5

10

15

20

25

30

35

40

45

50

Days

FIGURE 3-5.  The consequences are shown for accelerated clonogen repopulation in a gross tumor (1010 cells) undergoing 7 weeks of radiation therapy (70 Gy given in 2-Gy fractions daily except on the weekends [W]). Each treatment reduces the surviving fraction by 50%. The red line shows the decline in clonogen number in the absence of repopulation; the approximately 0.3 cells left at the end of the treatment period corresponds to a control probability of approximately 75%. In the presence of repopulation (blue line), about 3000 clonogens remain at the end of treatment, precluding any chance for control. 106 105 Clonogen number

This combined approach has the potential to achieve high local control rates with minimal disfiguration or disruption of anatomy or function. When adjuvant radiation therapy was first introduced, it seemed that a dose of 45 to 50 Gy given at 1.8 to 2.0 Gy per fraction would eradicate about 90% of microscopic disease whether delivered before or after surgery.16-18 However, it soon became evident that postoperative irradiation delivered to surgically disturbed tissues had to be given in higher doses (approximately 60 Gy) to achieve equivalent control.22,26-28 The reasons for this differential dose effect remain unclear. It is unlikely that the classic explanation invoking radioresistance as a result of hypoxia in surgically scarred tissue is valid because radiobiologic hypoxia is extreme hypoxia that would be inconsistent with wound healing and tissue integrity. An alternative explanation invokes accelerated clonogen repopulation after surgical resection.29,30 Repopulation of clonogenic cells occurs during the irradiation of gross tumors, especially in head and neck cancer31,32 but also in other types of cancer.21,30 This accelerated repopulation typically begins after a lag period of approximately 3 to 4 weeks when gross tumors are treated and decreases the relative effectiveness of radiation delivered during the succeeding weeks (Fig. 3-5). Although relatively little is known about the repopulation kinetics of microscopic tumors after treatment, it is plausible to expect that they may repopulate quickly after cytoreductive therapy with surgery, irradiation, or chemotherapy. Some evidence exists to support acceleration of subclinical rectal adenocarcinoma after irradiation.21,30 The adverse consequences of early accelerated regrowth after surgery on the radiation response are illustrated in Figure 3-6. Adjuvant irradiation delivered preoperatively or delivered to surgically undisturbed tissues postoperatively is usually given to a dose of approximately 50 Gy over a ­period of 5 weeks; when surgically disturbed tissues are irradiated, the dose is typically increased to approximately 60 Gy given over 6 weeks. Because the density of clonogenic infestation decreases with distance from the primary tumor (see Fig. 3-1), the last 10 Gy of the planned postoperative 60 Gy are delivered to a cone-down volume

W

106

100

About 90% for 1- to 3-cm neck nodes

Modified from Fletcher GH, Shukovsky LJ. The interplay of radiocurability and tolerance in the irradiation of human cancers. J Radiol Electrol Med Nucl 1975;56:383-400.

W

107

101

About 50% for T3/T4 tonsil 70 Gy

W

108 Clonogen number

TABLE 3-1.   Local Control Rates According to Tumor Size and Radiation Dose for Squamous Cell Carcinoma of the Head and Neck

104 103 Postoperative (60 Gy in 6 weeks)

102 101

Preoperative (50 Gy in 5 weeks)

100 10–1

0

5

10

15

20

25

30

35

40

Days

FIGURE 3-6.  The graph shows the early accelerated repopulation from postoperative irradiation and lesser repopulation from preoperative irradiation. Postoperative conditions in the tumor bed and its surroundings are assumed to be particularly conducive to cell proliferation because of the local reparative tissue response and the release of microscopic tumor clonogens from restraining environmental conditions resulting from the preexisting gross tumor. In this example, 60 Gy given over a 6-week period after surgery (blue line) produces effects that are approximately equivalent to 50 Gy given over a 5-week period before surgery (red line). The surviving fraction after each dose is 0.5.

centered over the tumor bed itself—the shrinking field technique. Although the optimal radiation dose for subclinical disease is often presented as an all-or-none quantity, lesser or greater doses than 50 to 60 Gy may be optimal in some circumstances.1,20,25,27,29 The quantity of ­subclinical

96

PART 1  Principles

l­ocoregional disease present in any patient can range from a few cells to as many as 109 cells or more. The optimal radiation dose varies accordingly between patients. ­Unfortunately, no method has been found for quantifying clonogen density in individuals. Clinical judgment about factors such as the microscopic surgical margin and whether that margin is positive, negative, close, or wide is used to estimate likely clonogen density. Microscopically positive resection margins are a regularly documented predictor of increased recurrence rates in virtually all invasive tumors, and a dose of 65 Gy given by the shrinking field technique is widely used in this circumstance. Investigations of the response of subclinical disease to radiation have reaffirmed the great variation in clonogen density in different patients.1,21,29,30 The numbers of patients with different amounts of subclinical disease seem to be distributed across the spectrum of clonogen numbers from 1 to about 109, and the actual distribution is ­approximately log-normal (i.e., the fraction of patients with 10 to 102 cells is approximately the same as the fraction with 102 to 103 cells or 103 to 104 cells, and so on).1 Although knowing the exact distribution of patients across this spectrum is not crucial, it is important to understand that such a wide distribution of clonogen number results in dose­response curves for the whole population that do not have a dose threshold, at least for doses greater than 10 Gy. However, some patients have very few clonogens, and they can benefit even from relatively low radiation doses. This concept cannot be taken as an endorsement of lowdose adjuvant radiation therapy in general, but in some circumstances when tissue tolerance is low, relatively low doses can be justified. It is good to emphasize that the effectiveness of ­conventional-dose adjuvant radiation therapy is often not as impressive as initial inspection of clinical data suggests.1,29 For example, a typical outcome would be to find that combined surgery and irradiation results in a very acceptable 90% local control rate; however, if the control rate with surgery alone is 70%, only 20 of every 30 patients per 100 would have their subclinical disease eradicated by radiation, which is only a 60% local control rate by irradiation. In addition to observing overall local control rate, the radiation oncologist should pay close attention to the percentage reduction in recurrence rate, particularly the following: 100 × (Failure rate without radiation −Failure rate with radiation) ÷ Failure rate without radiation This calculation provides the clearest index of the success of this treatment because not every patient judged to have subclinical disease actually has it. Significantly, radiobiologic evaluation of the dose­response relationship for subclinical disease reveals that the curves are quite steep.21,29 This implies that relatively small increments in cytotoxicity should result in significant gains in control. Observations of the efficacy of combined chemotherapy and radiation (i.e., chemoradiation) as an adjuvant treatment may well reflect this observation.12,34-39 For many circumstances, chemoradiation is increasingly the preferred adjuvant treatment, as discussed in Chapter 4.

Topographic Relationships between Surgery and Radiation Therapy Surgery and radiation may be directed at the same area, or they can be used to treat different areas. When used to treat different areas, the plan typically is to use surgery for the primary tumor and adjuvant radiation therapy for the clinically negative nodal drainage area instead of elective lymphadenectomy, sparing the patient some morbidity. This approach is often used for treating head and neck cancer and is standard practice for testicular seminoma. When both modalities are directed at more or less the same area, the rationale is to improve control of large primary tumors (T3/T4) or to conserve organ anatomy and function and to minimize disfigurement for smaller tumors (T1/T2) by using conservative surgery for macroscopic tumor and irradiation for occult locoregional disease. Lumpectomy and irradiation for breast cancer is an example of this approach, as is the use of radiation therapy and sphinctersparing surgery for rectal cancer or the use of conservative resection and radiation therapy for soft tissue sarcoma. Large, primary tumors in many sites are treated with both modalities because locoregional control obtained with only one modality is often poor.

Preoperative versus Postoperative Radiation Therapy From general principles, it is possible to provide a rationale for preoperative and postoperative irradiation; therefore, it is not possible to definitively argue for one approach over the other. In practice, the decision to use one or the other of these sequences is dictated by clinical experience and tends to be site-specific. However, it is instructive to review the virtues and limitations of each approach and to consider some aspects of intraoperative radiation therapy and brachytherapy.

Preoperative Radiation Therapy Potential virtues of preoperative irradiation are as follows: 1. An inoperable or borderline operable tumor may be rendered operable by destruction of peripheral tumor extensions that invade adjacent critical organs. 2. Destruction of peripheral tumor extensions decreases the likelihood of disease-positive microscopic resection margins. 3. Inactivation of a large proportion of tumor clonogens decreases the likelihood of tumor cells being mechanically disseminated through vascular channels or being seeded throughout the surgical field. 4. Irradiation of a surgically undisturbed area allows the use of smaller radiation portals than would be possible postoperatively, when the whole surgical bed is potentially contaminated and requires irradiation.40,41 5. Irradiation of a surgically undisturbed tumor allows the use of a lower radiation dose (50 Gy) than that recommended postoperatively (60 Gy), potentially decreasing radiation complications.

3  Principles of Combining Radiation Therapy and Surgery

Potential limitations of preoperative irradiation are as follows: 1. Patients cannot be selected for adjuvant radiation therapy on the basis of accurate surgical staging; depending on initial selection criteria, a significant fraction of patients treated preoperatively may not need both modalities. 2. The achievement of significant downstaging may preclude optimal selection of further treatment such as adjuvant chemotherapy; for example, preoperative irradiation of the breast and axilla results in a high proportion of complete remissions within the axilla. 3. Because surgery is delayed until about 4 weeks after the completion of irradiation, a patient left with microscopically disease-positive resection margins is faced with “split-course” irradiation, which is known to be relatively ineffective. 4. Surgical complications are generally increased after preoperative irradiation.28,42-44 5. The primary treatment (surgery) is delayed. This common criticism is probably spurious, because radiation therapy is an effective modality, and tumors undergoing irradiation cannot reasonably be deemed to remain untreated during this time.

Postoperative Radiation Therapy Potential virtues of postoperative irradiation are as follows: 1. The extent of gross disease is accurately known, and radiation therapy can be specifically tailored to sites of special concern. 2. Microscopic margins are defined, and positive margins, if present, are identified, allowing targeted boost irradiation, ideally guided by surgical clips. 3. The likelihood of irradiating patients whose disease does not merit adjuvant treatment is reduced. 4. Increased surgical complications are avoided. Potential limitations of postoperative irradiation are as follows: 1. Delivery of radiation may be delayed by surgical complications, leading to significant clonogen repopulation and occasionally to gross recurrence. 2. Irradiation of a surgically disturbed tumor bed requires a higher dose than does preoperative irradiation. 3. The volume of tissue requiring irradiation is usually larger than for preoperative irradiation because the whole surgical bed and drain are potentially contaminated. 4. The surgical procedure may fix certain organs in or adjacent to the surgical bed, resulting in potentially significant radiation complications; fixation can be a particular problem for the small bowel after abdominopelvic surgery. Very few clinical trials have addressed the issues posed by the previous summary, and the conclusions have been contradictory. A major problem in interpretation is posed by the often idiosyncratic preoperative irradiation schedules, which emphasize the use of a few large fractions (e.g., 10 Gy given once; 3.7, 4, or 5 Gy given five times). Two prospective, randomized trials that used conventional fractionation seem to have had a significant influence on the pattern of combined-modality treatment for head and

97

neck cancer. One of these trials compared 55 Gy given over 5.5 weeks before or after surgery to patients with hypopharyngeal cancer and found a significant advantage for postoperative irradiation with regard to overall survival.45 Data on locoregional control were not clearly presented. The second trial comprised a much larger group of patients with advanced head and neck cancer and randomized them to undergo preoperative irradiation (50 Gy at 1.8 to 2.0 Gy per fraction) or postoperative irradiation (60 Gy at the same fraction size). Locoregional control was significantly better in the postoperative irradiation group.46 Since then, postoperative irradiation has become the standard for treating head and neck cancer. The only other disease for which preoperative irradiation has been directly compared with postoperative irradiation in randomized trials is rectal cancer, and in that case, preoperative irradiation emerged as the better sequence.44 Neither considerations from first principles nor clinical results allows global conclusions to be reached regarding the superiority of preoperative or postoperative adjuvant radiation therapy.

Time Interval between Treatment Modalities Regardless of whether radiation therapy is used before or after surgery, the time interval between the treatment modalities is of some significance in view of the possibility of tumor cell repopulation during that interval. Repopulation in the interval between preoperative irradiation and surgery would seem to be of little consequence unless the operation was to be unreasonably delayed. If preoperative treatment eradicates all microscopic disease, in situ repopulation is impossible, and reseeding of tissues by new clonogen outgrowth would be unlikely in the short term; if the treatment failed in this respect, even prompt surgery would be of little avail. The major determinant of the interval between modalities in this situation is the likelihood of surgical complications. When low-dose, high-doseper-fraction radiation therapy is used preoperatively, acute radiation reactions are insignificant, and experience confirms that surgery can be safely accomplished within 1 to 2 weeks of the completion of irradiation. Several prospective trials involving hypofractionated preoperative irradiation have shown no difference in postsurgical complications between patients treated with surgery alone and those treated with irradiation followed by surgery within 2 weeks of the completion of radiation therapy.12,44,47-49 With intermediate radiation doses (about 40 Gy in 2.0-Gy fractions given over 4 weeks), modest acute reactions may occur, and surgery is usually delayed for 2 to 4 weeks until the reactions have more or less resolved. In this circumstance, surgical complications do not seem to be increased.50-52 When 50 Gy is delivered over 5 weeks, acute radiation reactions are usual at the completion of therapy, and surgery is delayed to the 4- to 6-week interval; with this practice, surgical complications such as delayed wound healing and wound breakdown are increased for some tumor sites28,40,53,54 but not for others.55 Earlier studies found an increase in surgical complications if surgery was performed less than 4 weeks after irradiation,19 and it is therefore standard practice to delay surgery until 4 or 5 weeks after the completion of 50 Gy of

98

PART 1  Principles

preoperative radiation therapy. One advantage of delaying surgery for this time is to allow the tumor time to regress (i.e., downstaging). In low rectal tumors, downstaging increases the likelihood for sphincter-sparing surgery.51 It is a generally accepted principle that postoperative irradiation should be delayed until the surgical wound has healed. Depending on the type of surgery performed, radiation therapy is typically deferred for 3 to 6 weeks. Although accelerated clonogen repopulation may pose a problem even with these short time intervals, delays beyond 6 weeks raise the most concern. Some studies have reported poorer locoregional control in patients irradiated more than 6 weeks after surgery.56-58 However, this finding has not been universal,43 and it is possible that patients whose radiation therapy is inordinately delayed by postsurgical wound problems are those with more advanced tumors that needed more extensive surgery. This issue cannot be definitively resolved, but general principles suggest that irradiation should commence as soon as feasible after surgery and that surgery should be optimized to avoid potential postoperative wound problems.

Intraoperative or Interstitial Radiation Therapy Although preoperative or postoperative external beam radiation therapy is the conventional form of irradiation integrated with surgery, the use of intraoperative or interstitial irradiation also has virtues. Intraoperative electron beam irradiation or high-dose-rate brachytherapy allows patients to be selected on the basis of their tumor characteristics as defined at surgery, and the treatment can be directed to specific areas at high risk for harboring microscopic disease. The major disadvantage is that treatment is limited to one dose (typically 10 to 20 Gy), with all the attendant radiobiologic disadvantages. Intraoperative irradiation therefore is used as a boost to conventional external irradiation. Few randomized clinical trials have evaluated the utility of this treatment. In one such trial for treating gastric carcinoma, the use of intraoperative irradiation (20 Gy) as sole adjuvant was associated with a significant reduction of locoregional recurrence compared with that achieved with surgery and conventional postoperative external irradiation.59 For retroperitoneal sarcoma, the use of an intraoperative boost followed by external irradiation was superior to the use of postoperative irradiation with conventional boosting, but the gain in local control was small.60 Adjuvant brachytherapy has the potential advantage of delivering a high dose to a limited volume and of delivering the radiation therapy with minimal postoperative delay. Adjuvant brachytherapy has significantly reduced the incidence of local recurrence for completely resected soft tissue sarcoma.14 The use of brachytherapy as a boost form of irradiation combined with more wide-field external irradiation is an attractive strategy, but it has not been significantly explored as an adjuvant to surgery.

Clinical Experiences Although this chapter was not intended to be a comprehensive review of the clinical experience with combined irradiation and surgery, it is useful to point

out some clinical experiences that justify the theoretical arguments for combining these modalities. A large number of prospective trials have established that the use of irradiation as a surgical adjuvant can significantly decrease locoregional recurrence rates for many types of cancer. For example, the current standard treatment for early breast cancer is local excision (lumpectomy) and irradiation; the addition of irradiation to lumpectomy reduces the local recurrence rate from approximately 30% to approximately 10%.11,13,61 For rectal cancer, the addition of radiation therapy to surgery reduces local recurrence rates from approximately 30% to approximately 15%.62-67 Most of the rectal carcinoma trials showing positive results have used preoperative hypofractionated irradiation; in one trial of postoperative irradiation, no benefit was found for radiation therapy delivered in that sequence.68 For soft tissue sarcomas managed with conservation surgery, postoperative irradiation, whether external15 or inter­ stitial,14 reduces local recurrence rates from approximately 30% to approximately 10%. Node-positive non–small cell lung cancer has a locoregional recurrence rate of approximately 25% when managed surgically; the addition of postoperative irradiation reduces the rate to 8%.69 The risk of regional recurrence is reduced by a factor of 1.7 when preoperative irradiation is added to surgery for esophageal cancer.12 For gastric cancer, locoregional recurrences fall from 52% with surgery alone to 39% with adjuvant irradiation.50 Locoregional recurrence rates for patients with unfavorable cervical cancer fall from 19% with surgery alone to 13% when irradiation is added postoperatively.70 Survival duration of patients with malignant glioma increases from a median of 5 months with surgery alone to a median of 11 months with the addition of postoperative radiation therapy.71,72 Even metastatic disease can benefit from adjuvant irradiation. The incidence of recurrence of resectable brain metastases falls from 70% with surgery alone to 18% when postoperative irradiation is added.73 For certain types of cancer, prospective trials have failed to find benefit from adjuvant irradiation. For example, despite numerous studies, no conclusive evidence has been found that adjuvant preoperative radiation therapy influences the outcome of bladder carcinoma.48,49 Likewise, renal carcinoma does not seem to benefit from adjuvant irradiation.74,75 One trial found that mortality rates for ­patients with right-sided primary tumors who received ­adjuvant treatment was increased because of radiation hepatitis.74

Complications In general, the likelihood of complications increases when surgery and radiation therapy are combined over that expected with either modality alone. Hypofractionated preoperative irradiation does not seem to significantly increase surgical problems, but the incidence of immediate postsurgical wound complications is increased in many situations when 40 to 50 Gy is delivered preoperatively. Surgical procedures in irradiated areas should be modified to ensure healing, and appropriate surgical experience is important. In particular, suture lines should never be taut, and when hollow-organ reanastomosis is required,

3  Principles of Combining Radiation Therapy and Surgery

unirradiated portions should be used if possible. In the head and neck region, trifurcate skin incisions for neck dissection should not be used because they often place the trifurcation point over the carotid artery, risking skin necrosis and carotid “blowout.”45 Handling tissues gently during the operation, maintaining meticulous hemostasis, avoiding closure under tension, and eliminating dead space through the use of drains and rotation flaps may decrease wound complication rates.54 Similar attention should be paid to surgery for patients for whom postoperative irradiation is planned. The incidence of long-term sequelae is often increased among patients given postoperative radiation therapy. For example, long-term follow-up of women given adjuvant breast irradiation has revealed a significant incidence of cardiac morbidity that correlated directly with the proportion of the heart that was irradiated.76 Even preoperative, low-dose, hypofractionated radiation therapy has long-term effects on colon function.67 Adjuvant irradiation must be planned with all the attention to detail customarily used for definitive treatment because the lower doses used in the adjuvant setting do not preclude the development of significant long-term sequelae.

General Guidelines and Conclusions Principles of Combining Radiation Therapy and Surgery Not every patient with cancer needs combined-modality treatment. Specific indications for the use of adjuvant irradiation are discussed elsewhere in this textbook, but some general guidelines are presented here. Combined irradiation and surgery should be considered for the following circumstances: 1. Small tumors for which preservation of organs or functions are realistic treatment goals should be considered for conservative surgery and irradiation. The sequence of modalities often is dictated by patterns of current practice (e.g., breast cancer is usually managed with resection followed by irradiation, whereas sarcoma is usually managed with irradiation followed by resection). 2. Large tumors for which resection is expected to result in poor local control or the resectability of which is marginal should be considered for preoperative irradiation. 3. After resection, any one or more of the following adverse pathologic factors should lead to consideration of postoperative irradiation: microscopically diseasepositive resection margins; microscopically close ­margins (<5 mm); large primary tumors (T3/T4), regardless of margin status; tumors that invade beyond the capsule of their organ of origin; tumors with extensive lymph node involvement or extranodal extension; tumors with lymphatic permeation; tumors with perineural invasion, particularly of larger named nerves; and tumors with bone invasion. It is worth reiterating that irradiation in modest doses (45 to 60 Gy) is a highly effective adjuvant to surgical resection for many types of cancer. Its efficacy rests on

99

­ ersuasive radiobiologic tenets, and it has been confirmed p for a variety of tumor types in many prospective, randomized clinical trials.

REFERENCES   1. Withers HR, Peters LJ, Taylor JM. dose-response relationship for radiation therapy of subclinical disease. Int J Radiat Oncol Biol Phys 1995;31:353-359.   2. Fletcher S, Black W, Harris R, et al. Special article: report of the international workshop on screening for breast cancer. J Natl Cancer Inst 1993;85:1644-1656.   3. Hakama M. Potential contribution of screening to cancer mortality reduction. Cancer Detect Prev 1993;17:513-520.   4. Selby JV, Friedman GD, Quesenberry CP, et al. A case-control study of screening sigmoidoscopy and mortality from colorectal cancer. N Engl J Med 1992;326:653-657.   5. Fuks Z, Leibel SA, Wallner KE, et al. The effect of local control in metastatic dissemination in carcinoma of the prostate: longterm results in patients treated with 125I implantation. Int J Radiat ­Oncol Biol Phys 1991;21:537-548.   6. Zagars GK, von Eschenbach AC, Ayala AG, et al. The influence of local control on metastatic dissemination of prostate cancer treated by external beam megavoltage radiation therapy. Cancer 1992;6:2370-2377.   7. Vigliotti A, Rich TA, Romsdahl MM, et al. Postoperative adjuvant radiotherapy for adenocarcinoma of the rectum and rectosigmoid. Int J Radiat Oncol Biol Phys 1987;13:999-1006.   8. Trovik CS, Bauer HC. Local recurrence of soft tissue sarcoma a risk factor for late metastases: 379 patients followed for 0.5-20 years. Acta Orthop Scand 1994;65:553-558.   9. Chyle V, Zagars GK, Wheeler JT, et al. Definitive radiotherapy for carcinoma of the vagina: outcome and prognostic factors. Int J Radiat Oncol Biol Phys 1996;35:891-905. 10. Leibel SA, Scott CB, Mohiuddin M, et al. The effect of local­regional control on distant metastatic dissemination in carcinoma of the head and neck: results of an analysis from the RTOG head and neck database. Int J Radiat Oncol Biol Phys 1991;21:549-556. 11. Liljegren G, Holmberg L, Bergh J, et al. 10-Year results after sector resection with or without postoperative radiotherapy for stage I breast cancer: a randomized trial. J Clin Oncol 1999;17:2326-2333. 12. Bosset JF, Gignoux M, Triboulet JP, et al. Chemoradiotherapy followed by surgery compared with surgery alone in squamous-cell carcinoma of the esophagus. N Engl J Med 1997;337:161-167. 13. Fisher B, Anderson S, Redmond CK, et al. Reanalysis and results after 12 years of follow-up in a randomized clinical trial comparing total mastectomy with lumpectomy with or without irradiation in the treatment of breast cancer. N Engl J Med 1995;333:14561461. 14. Pisters PWT, Harrison LB, Leung DHY, et al. Long-term results of a prospective randomized trial of adjuvant brachytherapy in soft tissue sarcoma. J Clin Oncol 1996;14:859-868. 15. Yang JC, Chang AE, Baker AR, 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. 16. MacComb WS, Fletcher GH. Planned combination of surgery and radiation in treatment of advanced primary head and neck cancers. Ann Surg 1957;77:397-414. 17. Bagshaw MA, Thompson RW. Elective irradiation of neck in patients with primary carcinoma of head and neck. JAMA 1971;217:456-458. 18. Fletcher GH. The evolution of the basic concepts underlying the practice of radiotherapy from 1949 to 1977. Radiology 1978;127:3-19. 19. White EC, Fletcher GH, Clark RL. Surgical experience with preoperative irradiation for carcinoma of the breast. Ann Surg 1962;155:948-956. 20. Fletcher GH. Implications of the density of clonogenic infestation in radiotherapy. Int J Radiat Oncol Biol Phys 1986;12:16751680.

100

PART 1  Principles

21. Suwinski R, Taylor JM, Withers HR. Rapid growth of microscopic rectal cancer as a determinant of response to preoperative radiation therapy. Int J Radiat Oncol Biol Phys 1998;42:943-951. 22. Fletcher GH. Irradiation of subclinical disease in the draining lymphatics. Int J Radiat Oncol Biol Phys 1984;10:939-942. 23. Fletcher GH, Shukovsky LJ. The interplay of radiocurability and tolerance in the irradiation of human cancers. J Radiol Electrol Med Nucl 1975;56:383-400. 24. Zagars GK, Schultheiss TE, Peters LJ. Inter-tumor heterogeneity and radiation dose-control curves. Radiother Oncol 1987;8:353362. 25. Fletcher GH. Irradiation of subclinical disease in the draining lymphatics. Int J Radiat Oncol Biol Phys 1984;10:939-942. 26. Marcus RG, Million RR, Cassissi NJ, et al. Postoperative irradiation for squamous cell carcinomas of the head and neck: analysis of time-dose factors related to control above the clavicles. Int J Radiat Oncol Biol Phys 1979;5:1943-1949. 27. Amdur RJ, Parsons JT, Mendenhall WM. Postoperative irradiation for squamous cell carcinoma of the head and neck: an analysis of treatment results and complications. Int J Radiat Oncol Biol Phys 1989;16:25-36. 28. Pollack A, Zagars GK, Goswitz MS, et al. Preoperative radiotherapy in the treatment of soft tissue sarcomas: a matter of presentation. Int J Radiat Oncol Biol Phys 1998;42:563-572. 29. Peters LJ, Withers HR. Applying radiobiological principles to combined modality treatment of head and neck cancer—the time factor. Int J Radiat Oncol Biol Phys 1997;39:831-836. 30. Withers HR, Suwinski R. Radiation dose response for subclinical metastases. Semin Radiat Oncol 1998;8:224-228. 31. Withers HR, Taylor JMG, Maciejewski B. The hazard of accelerated tumor clonogen repopulation during radiotherapy. Acta ­Oncol 1988;27:131-146. 32. Awwad HK, Khafagy Y, Barsoum M, et al. Accelerated versus conventional fractionation in the postoperative irradiation of locally advanced head and neck cancer: influence of tumor proliferation. Radiother Oncol 1992;25:261-266. 33. Peters LJ, Ang KK, Byers RM, et al. Evaluation of the dose for postoperative radiation therapy of head and neck cancer: first report of a prospective randomized trial. Int J Radiat Oncol Biol Phys 1993;26:3-11. 34. El-Sayed S, Nelson N. Adjuvant and adjunctive chemotherapy in the management of squamous cell carcinoma of the head and neck region: a meta-analysis of prospective and randomized trials. J Clin Oncol 1996;14:838-847. 35. Munro AJ. An overview of randomized controlled trials of adjuvant chemotherapy in head and neck cancer. Br J Cancer 1995;71:83-91. 36. Heriot AG, Kumar D. Adjuvant therapy for resectable rectal and colonic cancer. Br J Surg 1998;85:300-309. 37. Klinkenbijl JH, Jeekel J, Sahmoud T, 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. Ann Surg 1999;230:776782. 38. Cox JD. Chemoradiation for malignant epithelial tumors. Cancer Radiother 1998;2:7-11. 39. Link KH, Staib L, Schatz M, et al. Adjuvant ­radiochemotherapywhat is the patient benefit?. Langenbecks Arch Surg 1998;383:416426. 40. O’Sullivan B, Davis A, Canadian Sarcoma Group, et al. Effect on radiotherapy field sizes in a recently completed Canadian Sarcoma Group and NCI Canada Clinical Trials Group randomized trial comparing pre-operative and post-operative radiotherapy in extremity soft tissue sarcoma [abstract 176]. Int J Radiat Oncol Biol Phys 1999;45;(Suppl):238-239. 41. Nielsen OS, Cummings B, O’Sullivan B, et al. Preoperative and postoperative irradiation of soft tissue sarcomas: effect on radiation field size. Int J Radiat Oncol Biol Phys 1991;21:1595-1599. 42. Suit HD, Makin HJ, Wood WC, et al. Preoperative, intraoperative, and postoperative radiation in the treatment of primary soft tissue sarcoma. Cancer 1985;55:2659-2667.

43. Taylor JMG, Mendenhall WM, Parson JT, et al. The influence of dose and time on wound complications following postradiation neck dissection. Int J Radiat Oncol Biol Phys 1992;23:41-46. 44. Frykholm GJ, Glimelius B, Pahlman L. Preoperative or postoperative irradiation in adenocarcinoma of the rectum: final treatment results of a randomised trial and an evaluation of secondary ­effects. Dis Colon Rectum 1993;36:564-572. 45. Vandenbrouck C, Sancho H, LeFur R, et al. Results of a randomized clinical trial of preoperative irradiation versus postoperative in treatment of tumors of the hypopharynx. Cancer 1977;39:1445-1449. 46. Tupchong L, Scott CB, Blitzer PH, et al. Randomized study of preoperative versus postoperative radiation therapy in advanced head and neck carcinoma: long-term follow-up of RTOG study 73-03. Int J Radiat Oncol Biol Phys 1991;20:21-28. 47. Hyams DH, Mamounas EP, Petrelli N, et al. A clinical trial to evaluate the worth of preoperative multimodality therapy in patients with operable carcinoma of the rectum: a progress report of National Adjuvant Breast and Bowel Protocol R-03. Dis Colon Rectum 1997;40:131-139. 48. Smith JA, Crawford ED, Paradelo JC, et al. Treatment of advanced bladder cancer with combined preoperative irradiation and radical cystectomy versus radical cystectomy alone: a phase III intergroup study. J Urol 1997;157:805-807. 49. Huncharek M, Muscat J, Geschwind JF. Planned preoperative radiation therapy in muscle invasive bladder cancer: results of a meta-analysis. Anticancer Res 1998;18:1931-1934. 50. Zhang ZX, Gu XZ, Yin WB, et al. Randomized clinical trial on the combination of preoperative irradiation and surgery in the treatment of adenocarcinoma of gastric cardia (AGC)—report on 370 patients. Int J Radiat Oncol Biol Phys 1998;42:929-934. 51. Francois Y, Nemoz CJ, Baulieux J, et al. Influence of the interval between preoperative radiation therapy and surgery on downstaging and on the rate of sphincter-sparing surgery for rectal cancer: the Lyon R90-01 trial. J Clin Oncol 1999;17:2396-2402. 52. Shipley WU, Cummings KB, Coombs LJ, et al. 4000 Rad preoperative irradiation followed by prompt radical cystectomy for invasive bladder carcinoma: a prospective study of patient tolerance and pathologic downstaging. J Urol 1982;127:448-451. 53. Cheng EY, Dusenbery KE, Winters MR, et al. Soft tissue sarcomas: preoperative versus postoperative radiotherapy. J Surg Oncol 1996;61:90-99. 54. Bujko K, Suit HD, Springfield DS, et al. Wound healing after preoperative radiation for sarcoma of soft tissues. Surg Gynecol Obstet 1993;176:124-134. 55. Pollack A, Zagars GK, Dinney CP, et al. Preoperative radiotherapy for muscle-invasive bladder cancer: long term follow-up and prognostic factors for 338 patients. Cancer 1994;74:2819-2827. 56. Vikram B, Strong EW, Shah JP, et al. Failure in the neck following multimodality treatment for advanced head and neck cancer. Head Neck Surg 1984;6:724-729. 57. Vikram B. Importance of time interval between surgery and postoperative radiation therapy in the combined management of head and neck cancer. Int J Radiat Oncol Biol Phys 1979;5:1837-1840. 58. Byers RM, Clayman GL, Guillamondegui OM, et al. Resection of advanced cervical metastasis prior to definitive radiotherapy for primary squamous carcinomas of the upper aerodigestive tract. Head Neck 1992;14:133-138. 59. Sindelar WF, Kinsella TJ, Tepper JE, et al. Randomized trial of intraoperative radiotherapy in carcinoma of the stomach. Am J Surg 1993;165:178-186. 60. Sindelar WF, Kinsella TJ, Chen PW, et al. Intraoperative radiotherapy in retroperitoneal sarcomas: final results of a prospective, randomized, clinical trial. Arch Surg 1993;128:402-410. 61. Recht A, Come SE, Gelman RS, et al. Integration of conservative surgery, radiotherapy and chemotherapy for the treatment of early stage node-positive breast cancer: sequencing, timing and outcome. J Clin Oncol 1991;9:1662. 62. Cohen AM. Has preoperative radiation therapy for resectable rectal cancer been proven effective. J Surg Oncol 1990;45:69-71. 63. Pahlman L, Glimelius B. Pre- or postoperative radiotherapy in rectal and rectosigmoid carcinoma: report from a randomized multicenter trial. Ann Surg 1990;211:187-192.

3  Principles of Combining Radiation Therapy and Surgery 64. Glimelius B, Isacsson U, Jung G, et al. Radiotherapy in addition to radical surgery in rectal cancer: evidence for a dose-response effect favoring preoperative treatment. Int J Radiat Oncol Biol Phys 1997;37:281-287. 65. Dahlberg M, Glimelius B, Pahlman L. Improved survival and reduction in local rates after preoperative radiotherapy: evidence for the generalizability of the results of Swedish Rectal Cancer Trial. Ann Surg 1999;229:493-497. 66. Hoskins RB, Gunderson LL, Dosoretz DE, et al. Adjuvant postoperative radiotherapy in carcinoma of the rectum and rectosigmoid. Cancer 1985;55:61. 67. Dahlberg M, Glimelius B, Graf W, et al. Preoperative irradiation affects functional results after surgery for rectal cancer: results from a randomized study. Dis Colon Rectum 1998;41:543-551. 68. Arnaud JP, Nordlinger B, Bosset JF, et al. Radical surgery and postoperative radiotherapy as combined treatment in rectal cancer: ­final results of a phase III study of the European Organization for Research and Treatment of Cancer. Br J Surg 1997;84:352-357. 69. Smolle-Juettner FM, Mayer R, Pinter H, et al. “Adjuvant” external radiation of the mediastinum in radically resected non–small cell lung cancer. Eur J Cardiothorac Surg 1996;10:947-950. 70. Sedlis A, Bundy BN, Rotman MZ, et al. A randomized trial of pelvic radiation therapy versus no further therapy in selected patients with stage IB carcinoma of the cervix after radical hysterectomy and pelvic lymphadenectomy. Gynecol Oncol 1999;73:177-183.

101

71. Walker MD, Alexander E, Hunt WE. Evaluation of BCNU and/or radiotherapy in the treatment of anaplastic glioma: a cooperative clinical trial. J Neurosurg 1978;49:333-343. 72. Kristiansen K, Hagen S, Kollevoid T, et al. Combined modality therapy of operated astrocytomas grade III and IV. Confirmation of the value of postoperative irradiation and lack of potentiation of bleomycin on survival time: a prospective multicenter trial of the Scandinavian Glioblastoma Study Group. Cancer 1981;47: 649-652. 73. Patchell RA, Tibbs PA, Regine WF, et al. Postoperative radiotherapy in the treatment of single metastases to the brain: a randomized trial. JAMA 1998;280:1485-1489. 74. Finney R. An evaluation of postoperative radiotherapy in hypernephroma treatment-a clinical trial. Cancer 1973;32: 1332-1340. 75. Kjaer M, Frederiksen PL, Engelholm SA. Postoperative radiotherapy in stage II and III renal adenocarcinoma: a randomized trial by the Copenhagen Renal Cancer Study Group. Int J Radiat Oncol Biol Phys 1987;13:665-672. 76. Gyenes G, Rutqvist LE, Liedberg A, et al. Long-term cardiac morbidity and mortality in a randomized trial of pre- and postoperative radiation therapy versus surgery alone in primary breast cancer. Radiother Oncol 1998;48:185-190.

4

Principles of Combining Radiation Therapy and Chemotherapy Luka Milas AND  James D. Cox THERAPEUTIC RATIO IMPROVING THE THERAPEUTIC RATIO Spatial Cooperation Independence of Toxicity Enhancement of Tumor Response Protection of Normal Tissues ADDITIVITY OF DRUG-RADIATION INTERACTIONS METHODS FOR ASSESSING DRUG-RADIATION EFFECTS MECHANISMS OF DRUG-RADIATION INTERACTIONS Initial Radiation Damage Inhibition of Cellular Repair Cell Cycle Effects Hypoxia Cell Regeneration Inhibition of Tumor Angiogenesis

In perhaps no other aspect of oncologic practice has a direction developed that is as clear and compelling as the use of chemotherapy in conjunction with radiation therapy or radiation therapy and surgery. The question of whether to use combined-modality treatment has changed to which sequence and which drugs are the best. Given the increasing availability of preclinical investigations that can be used to chart a course for combining radiation therapy with drugs, it is important to explore in some depth the preclinical findings that have pointed the way to the use of combined-modality therapy. It is equally important to clarify the essential steps by which clinical investigations establish the toxicity, efficacy, and proper place for chemotherapy in combined-modality therapies in standard practice. In this chapter, we define some basic concepts, outline mechanisms of interaction between irradiation and chemotherapy, present theoretical principles underlying sequencing strategies, and review clinical applications of the combined-treatment strategies.

Therapeutic Ratio Neither radiation nor chemotherapeutic drugs are parti­ cularly selective against tumors; both also damage normal tissues. This damage limits the dose of radiation or drugs, or both, that can be given to patients. To be therapeutically beneficial, individual agents or their com­bination must be more effective against tumors than against normal tissues. Both antitumor efficacy and normal tissue injury are positively related to the dose of radiation or cytotoxic drugs, and this relationship is commonly described by a sigmoid curve (Fig. 4-1). The plots shown in Figure 4-1 provide information on the therapeutic ratio or index,

102

SEQUENCING OF CHEMOTHERAPY AND RADIATION THERAPY Induction Chemotherapy Concurrent Chemotherapy Adjuvant Chemotherapy CLINICAL APPLICATIONS OF COMBINED RADIATION THERAPY AND CHEMOTHERAPY SEQUENTIAL CHEMOTHERAPY AND RADIATION THERAPY Induction Chemotherapy Adjuvant Chemotherapy Concurrent Chemotherapy and Radiation Therapy Chemoradiation plus Surgery Organ Conservation Toxicity Issues

which is defined as the ratio of the maximally tolerated drug (in this case, with regard to normal tissue damage, curve C) to the minimally curative or effective dose (in this case, the antitumor effect, curves A or B). Any level of probability can be used to calculate the therapeutic ratio; however, in experimental preclinical studies, 50% is the most commonly used. The objective of any cancer treatment is to achieve a positive therapeutic ratio (i.e., one in which the antitumor efficacy curve lies to the left of the curve representing normal tissue injury [see Fig. 4-1, curve A]). Treatments for which the therapeutic ratio is negative (i.e., the antitumor response curve lies to the right of that for normal tissue damage [see Fig. 4-1, curve B], and normal tissue damage is more likely than tumor damage) are contraindicated, ­except perhaps for palliation in some circumstances. In defining therapeutic ratio, the end points of tumor response and normal tissue injury must be appropriately categorized and quantified. The end points depend on factors such as whether the treatment is intended to be curative or palliative and which tissues limit the doses that can be given. Damage to normal tissue must be within acceptable limits in terms of the tissue type and the severity of the damage. The balance between complications arising from normal tissue damage and the probability of tumor control form the basis of the therapeutic ratio of a treatment.

Improving the Therapeutic Ratio The ultimate goal in combining chemotherapeutic drugs with radiation therapy is to improve the therapeutic ratio—to enhance the antitumor effect while minimizing the toxicity of the treatment to critical normal tissues. Steel

4  Principles of Combining Radiation Therapy and Chemotherapy

Probability of tumor control or normal tissue damage (%)

100

A

C

B

75

103

detail later in this chapter. Virtually all chemotherapeutic agents enhance radiation damage to normal tissues as well as malignancies; consequently, therapeutic benefit can be achieved if the susceptibility of tumors rather than that of critical normal tissues can be enhanced.

Protection of Normal Tissues

50

25

0 Dose

FIGURE 4-1.  The therapeutic ratio is the ratio of the doses of an agent or agents that produces the same probability of normal tissue damage (curve C) and antitumor effect (curves A and B).  Curve A illustrates an agent with a positive (beneficial) therapeutic ratio, one that produces more damage to tumor than to normal tissue; curve B illustrates a negative (undesirable) therapeutic ratio in which the damage to normal tissue exceeds the damage to the tumor.

and Peckham1 defined four principles by which therapeutic ratios of combined treatments can be improved: spatial cooperation, independence of toxicity, enhancement of tumor response, and protection of normal tissues.

Spatial Cooperation Treatments that act independently of each other at different anatomic sites are said to exhibit spatial coordination. For example, chemotherapy is commonly used to deal with tumor growth outside the radiation field, usually to eradicate metastases. Conversely, radiation can be used to treat neoplastic lesions in secluded sites that are poorly accessed by chemotherapeutic agents; an example of this strategy is the use of irradiation to supplement chemotherapeutic treatment of leukemia that has spread to the brain.

Independence of Toxicity Because toxicity to normal tissues is the principal doselimiting factor for chemotherapy and for radiation therapy, the antitumor efficacy of a combined treatment theoretically would be improved more readily and the treatment tolerated more easily through the use of drugs with toxic effects that do not overlap or enhance those induced by radiation therapy. Use of this principle requires thorough knowledge of the toxicity, pharmacology, and optimal administration of individual chemotherapeutic drugs to ensure that combining such drugs with radiation minimizes normal tissue damage while still retaining anti­tumor efficacy.

Enhancement of Tumor Response Some drugs can enhance radiation-induced tumor damage, or radiation can make some cells more sensitive to chemotherapy, presumably through interactions between the agents at the molecular, cellular, or pathophysiologic (microenvironmental or metabolic) levels. Some of the more important mechanisms of drug-radiation interactions that increase tumor susceptibility are considered in more

Some compounds are thought to increase the tolerance of normal tissues to radiation so that higher doses of radiation can be safely delivered to the tumor. Preclinical in vivo testing has shown that some chemicals can selectively or preferentially protect normal tissues.2-4 Among the commonly tested protectors are thiol compounds such as amitostine (WR-272), which may be useful in clinical settings, such as protecting salivary glands during radiation therapy delivered to the head and neck.5 A major problem with concurrent chemoradiation therapy is the increase in the toxic effects on tissues that is conferred by radiation. However, radioprotectants and technical improvements in radiation therapy, such as three-dimensional treatment planning, conformational radiation therapy, or the use of protons, are likely to minimize tissue toxicity and consequently enhance the effectiveness of chemoradiation.

Additivity of Drug-Radiation Interactions Interactions between chemotherapeutic agents and radia­ tion may result in additive, supra-additive, or subadditive effects, depending on whether the combination produces effects that are equal to, greater than, or less than, respectively, the sum of the activity produced by the individual agents. Evaluations of the additivity of a given combination must take into account the dose-response relationship for each individual agent. The calculations are simple when both agents have effects that are exponentially related to the dose. However, the dose-effect relationship is rarely ex­ponential. As described in Chapter 1, many radiation dose-cell survival curves have a shoulder of varying widths at lower doses of irradiation and an exponential relationship at higher doses. The presence of such a shoulder indicates that the cells can repair radiation damage. The dose-survival curves after chemotherapy tend to be more diverse; some have shoulders and some do not, and some show tails at higher drug doses. The presence of such tails implies the existence of cell subpopulations that are resistant to the chemotherapeutic agents. These nonlinear, dose-related characteristics in cell killing imply that when two agents are combined, additive effects can be produced over a range of doses. Determinations of additivity on the basis of a combination of only one dose of each agent are inappropriate and can be misleading. To overcome this problem, Steel and Peckham1 introduced the concept of isobologram analysis, in which an isoeffect plot is generated for the dose-­response pattern for the combination of two agents (Fig. 4-2). Dose-response curves are generated for each agent, and the curves are used to generate isobolograms, which ­illustrate the expected additive response. In the simplest situation, that in which each agent produces a linear dose response, the isobologram is also linear (see Fig. 4-2A). If the combination of the two agents has a supra-­additive

104

PART 1  Principles

Subadditive or protection

Dose of agent A

Supra-additive or synergy

C

Radiation dose (Gy)

FIGURE 4-3.  The effects of chemotherapeutic drugs on cell survival after irradiation are shown by blue lines indicating ­radiation only and by red indicating drug plus radiation. A, ­ Simple  displacement of radiation survival curve indicates additivity. B, Elimination of the shoulder indicates inhibition of sublethal damage repair. C, Change in the slope of the radiation curve indicates sensitization (displacement to the left) or protection (displacement to the right; green line).

A

En ve l

Subadditive or protection

op

e

of

ad

di

tiv

ity

Supra-additive or synergy

B

B

Cell survival

A

Dose of agent B

FIGURE 4-2.  Isobolograms are constructed for pairs of agents that have linear dose-response curves (A) or nonlinear dose­response curves (B). The envelope of additivity and regions of supra-additivity and subadditivity also are shown. (Modified from Steel GG, Peckham MJ. Exploitable mechanisms in combined radiotherapy-chemotherapy: the concept of additivity. Int J Radiat Oncol Biol Phys 1979;5:85-91.)

e­ ffect, the response appears to the left of the curve; if the combination has a subadditive effect, the response appears to the right of the curve. When the dose response is nonlinear, an envelope of additivity appears in the isobologram (see Fig. 4-2B), indicating a range of isoeffect curves that represent additive responses. In this situation, to obtain an additive effect, the doses of individual agents that produce the same biologic effect lie within the envelope. If the doses lie to the left of the isobologram envelope, the interaction is ­ supra-additive or synergistic—the effect is caused by doses of the two agents that are lower than would be predicted from the envelope of additivity. In contrast, if the doses lie to the right side of the envelope, the interaction is subadditive or antagonistic, meaning the effect requires higher doses of the two agents than predicted. The envelope of additivity becomes wider as the gap between the nonlinear dose responses to individual agents becomes greater.

Methods for Assessing Drug-Radiation Effects Many in vitro and in vivo assays are used to assess the biologic effects of drugs, radiation, and the combination of both.6,7 In vitro assessments most commonly involve cell

survival or clonogenic assays, which involve measuring the ability of cells to survive and produce colonies of progeny of a defined minimum size (i.e., ability to sustain reproductive integrity). Survival curves are usually designed by plotting the surviving fraction of cells on a logarithmic scale against the dose of radiation or drugs, which is plotted on a linear scale (Fig. 4-3). Established cell lines typically are used to assess clonogenic survival after treatment with drugs or radiation in vitro, although a variation of the clonogenic cell survival assay called the excision assay can also be used to study antitumor treatments given in vivo. In an excision assay, the tumor is excised at some specified time after treatment, and a single-cell suspension of tumor cells is prepared and plated for subsequent determination of cell survival. The addition of drugs to radiation can influence survival curves in several ways (see Fig. 4-3). First, the drugs may shift the curve downward without changing its shape. The magnitude of this displacement corresponds to the amount of cells killed by the drugs (see Fig. 4-3A). Second, the drugs may eliminate the shoulder of a radiation survival curve, which implies that the drugs are inhibiting the cellular radiation-repair process (see Fig. 4-3B). Third, drugs may change the slope of the exponential portion of the survival curve (see Fig. 4-3C); in this case, a steeper slope indicates sensitization and a shallower slope indicates ­protection. Many methods are available for assessing the effects of chemotherapeutic drugs on tumors and normal tissues in vivo.6,7 Two of the most common assays used to assess ­tumor response are the tumor growth delay assay8 and the tumor-control dose (TCD50) assay.9 When the treatment end point is a delay in tumor growth, the size of untreated and treated tumors is measured as a function of time and plotted in growth curves. The absolute growth delay ­represents the difference in time (usually in days) needed for untreated and treated tumors to reach a defined size (Fig. 4-4A). When a drug is given with radiation, the growth ­delay achieved by both modalities minus the delay caused by the drug alone is called the normalized growth delay. To determine whether the addition of drugs enhances or ­reduces the antitumor efficacy of radiation requires determining normalized growth delay after at least three different doses of radiation. The normalized growth delays after the combined treatment and the absolute growth delays

4  Principles of Combining Radiation Therapy and Chemotherapy Drug

Drug+xrt

Xrt

Tumor size

Control

0

A

B

A

C

D

Time after treatment 20

Normalized growth delay (days)

18 16 14 12

enhancement of the radiation response of a mouse mammary carcinoma by the administration of docetaxel before irradiation.10 In the TCD50 assay, the end point is tumor control or cure in 50% of the treated subjects. In this assay, the proportion of tumors cured by radiation only or by a drug-plusradiation combination is plotted as a function of radiation dose. If the combination displaces the radiation-only curve to the left, addition of the drug improves tumor curability; however, the displacement does not discriminate between improvement caused by independent cell kill by the drug and that caused by a supra-additive interaction between the drug and radiation. Many assays can be used to quantify the radiation doseresponse relationship for normal tissues and the influence of chemotherapeutic drugs on the radiation response ­(Table 4-1). Some of these assays are clonogenic; the end point ­ depends directly on the reproductive integrity of ­individual cells. More frequently, however, dose-response relationships for normal tissues are based on functional end points. These end points tend to reflect the minimum number of functional cells remaining in tissues or organs rather than the proportion of cells that retain reproductive integrity.

Mechanisms of Drug-Radiation Interactions

10 8 6 4 2 8

B

105

10

12

14

16

18

20

22

Radiation dose (Gy)

FIGURE 4-4.  The graph shows the delay in tumor growth after treatment with a drug, radiation, or a combination of both.  A, The points at B, C, and D are absolute growth delays ­after treatment with a drug (red line), radiation (Xrt; green line), or drug plus radiation (Drug + xrt; yellow line). Normalized growth delay is defined as D minus B. B, The interactions between the chemotherapeutic drug docetaxel and radiation in a murine mammary carcinoma are plotted. Addition of the drug displaced the radiation response curve to the left, indicating ­enhancement of the tumor radioresponse. Green circles represent the tumor response after radiation only; red circles and blue triangles ­represent the tumor radioresponse when ­docetaxel was given 9 or 48 hours before irradiation, respectively. (From Mason KA, Hunter NR, Milas M, et al. Docetaxel enhances tumor radioresponse in vivo. Clin Cancer Res 1997;3:2431-2438.)

after irradiation only are plotted as a function of radiation dose (see Fig. 4-4B). The displacement of the normalized growth delay curve in relation to the absolute growth ­delay curve indicates the effect of drugs on tumor radioresponsiveness. No displacement indicates a simple additive ­effect; a shift to the left (to lower radiation doses) indicates enhancement of tumor radioresponse; and a shift to the right (to higher radiation doses) indicates a reduction in tumor ­radioresponse. The results shown in Figure 4-4B ­illustrate

Clinical experience has shown that chemotherapy is most effective in improving local tumor control when it is given during the course of radiation therapy. This obser­ vation implies that chemotherapeutic drugs and radia­tion actively interact. The multiple mechanisms underlying that interaction, which occur at the cellular and tumormicroenvironment levels, are complex; six major mech­ anisms are considered in the following paragraphs.

Initial Radiation Damage The critical target for radiation injury is the DNA molecule (see Chapter 1). Radiation induces several types of lesions, including single-strand breaks, double-strand breaks, base damage, and DNA-DNA or DNA-protein cross-links. The most common type of lesion is the single-strand break, which is readily repaired. Double-strand breaks left unrepaired or misrepaired and associated chromosomal aberrations usually lead to cell death.11 Certain drugs, such as halogenated pyrimidines, that become incorporated into DNA make the DNA more susceptible to radiation damage.12 In addition to increasing the initial injury caused by radiation, halogenated pyrimidines can enhance cell radiosensitivity by interfering with cellular repair mechanisms.12,13

Inhibition of Cellular Repair Cells have the ability to repair sublethal14 and potentially lethal15 radiation damage. Both types of repair are defined in operational terms. Sublethal damage repair is defined as the increase in survival observed when a dose of radiation is split into two (or more) fractions with some time interval between. Sublethal damage repair manifests on cell survival curves as restoration of a shoulder on the curve after the second dose. Repair takes place quickly, typically

106

PART 1  Principles

TABLE 4-1.    In Vivo Assays for Normal Tissue Damage by Drug-Radiation Combinations Tissue

Clonogenic End Points

Functional End Points

Bone marrow

Exogenous and endogenous spleen colony assay

LD50, depletion of different cell types

Skin

In situ skin colonies

Skin erythema and desquamation, hair loss, ­telangiectasia, skin contraction

Gastrointestinal tract

Jejunum and colon microcolonies

LD50, weight loss, protein or electrolyte leakage

Kidney tubules

Urine output

LD50, breathing rate, CO uptake

Lung Kidney Bladder Testis

Urination frequency Spermatogonial stem cell assay

Central nervous system

Testis weight, sperm count, fertility LD50 (brain), paralysis (spinal cord)

CO, carbon monoxide; LD50, lethal dose in 50%.

with a half-time of about 1 hour and a completion time of 4 to 6 hours after irradiation, and it takes place in tumors, normal tissues, and cultured cells. Potentially lethal damage repair designates an increase in the proportion of cells that survived irradiation because of some modification of postirradiation conditions. Potentially lethal damage repair typically takes place under conditions that prevent cellular division for several hours, such as maintaining cultured cells in a plateau phase after irradiation or delaying the removal of cells from tumors for clonogenic survival assay. Conceptually, repair of potentially lethal damage can be explained in terms of successful repair of DNA lesions that would have been lethal had the DNA undergone replication within several hours of irradiation. Repair can be achieved through several mechanisms, such as restoration of damaged molecules by reducing species that donate electrons to oxidized substrates and through the involvement of a variety of repair enzymes. Possible repair mechanisms for the various DNA lesions are summarized in Table 4-2. Many chemotherapeutic drugs can interact with cellular repair mechanisms, and in inhibiting those mechanisms, the drugs may enhance the response of cells or tissues to radiation. Halogenated pyrimidines can enhance cell ­radiosensitivity by increasing initial radiation damage and by inhibiting cellular repair. Significant effort has focused on exploring the radioenhancing effects of nucleoside analogues such as fludarabine16,17 and gemcitabine16-18 that can inhibit the repair of radiation-induced DNA or chromosome damage. Fludarabine is a potent inhibitor of DNA primase, DNA polymerase-α, ε-DNA ligase, and ribonucleotide reductase, and it can cause DNA chain termination.19 Fludarabine phosphate is rapidly dephosphorylated in vivo to form fludarabine, an adenine nucleoside analogue that is transported by a carrier into cells, where it is converted to the active metabolite by the addition of adenosine triphosphate. Transport into cells and metabolic activation of fludarabine seem to vary among cell types, with malignant cells concentrating and metabolizing the drug more efficiently than normal cells do. Fludarabine is clinically active against leukemias and lymphomas, but it is not particularly effective against solid tumors. Gemcitabine (2′-deoxy-2′,2′-difluorocytidine) is a fluorine-­substituted cytarabine that requires intracellular

TABLE 4-2.   Radiation-Induced DNA Damage and its Repair Induced Lesions

Types of Repair

Single-strand breaks

Direct reversal

Double-strand breaks

Homologous and ­nonhomologous ­recombination

Base alterations

Base excision repair

DNA-protein cross-links

Nucleotide excision repair

DNA, deoxyribonucleic acid.

­ hosphorylation by deoxycytidine kinase to produce the p active metabolites gemcitabine diphosphate (dFdCDP) and gemcitabine triphosphate (dFdCTP).17,18,20 Compared with fludarabine, gemcitabine has greater membrane permeability and enzyme affinity and has longer intracellular retention.17,20 Gemcitabine diphosphate interferes with DNA synthesis by inhibiting ribonucleotide reductase and reducing the deoxynucleotide pools. Gemcitabine triphosphate competes with deoxycytidine triphosphate for incorporation into elongated DNA strands; only one additional ­deoxynucleotide can be incorporated after the insertion of dFdCTP, halting DNA polymerization.17,20 In contrast to fludarabine, gemcitabine is active against many common solid tumors in humans, including head and neck, lung, and pancreatic carcinoma.16,17 Nucleoside analogues such as fludarabine and gemcitabine typically are potent radiosensitizers of ­ mammalian cells in vitro16-18,21 and enhancers of tumor radioresponse in vivo.16,17,22-24 Investigations from our laboratory have demonstrated that fludarabine and gemcitabine can significantly enhance the radioresponse of several mouse solid ­tumors, as assessed by tumor growth delay or local tumor cure, after single-dose or fractionated-dose irradiation.16,17,22-24 Fludarabine in particular produced radioenhancement ranging from a factor of 1.2 to more than 2.0, depending on the timing of fludarabine administration relative to ­irradiation, the dose of fludarabine, the type of ­tumor, and schedule of irradiation. Although several mechanisms seem to have been responsible for the enhancement, inhibition of radiation damage repair seems to have been prominent.

107

4  Principles of Combining Radiation Therapy and Chemotherapy 100

100

Single dose

80 Relative survival

Control

60

40

10−1

1 nM paclitaxel

10 nM paclitaxel

10−2

Tumor cure (%)

20

0

10−3 0

100

Fractionated

80

60

40

20

0 20

40

60

80

100

120

Total radiation dose (Gy)

FIGURE 4-5.  Fludarabine-induced enhancement of the ­response of the FSA mouse fibrosarcoma line to single-dose or fractionated-dose (16 fractions with a 6-hour interval ­between fractions) irradiation. Fludarabine (400 mg/kg) was given 3 hours before single-dose irradiation (A) or daily during the ­  4-day fractionated treatment (B) starting 3 hours before the first fractional dose of radiation. Blue lines indicate radiation only; red lines indicate radiation plus fludarabine. (From Gregoire V, Hunter NR, Brock WA, et al. Improvement in the therapeutic ratio of radiotherapy for a murine sarcoma by indomethacin plus fludarabine. Radiat Res 1996;146:548-553.)

The effect of fludarabine on the radiocurability of a murine sarcoma after single-dose or fractionated ­irradiation is illustrated in Figure 4-5. The enhancement was much greater when fludarabine was combined with fractionated irradiation, suggesting that the inhibition of sublethal (or potentially lethal) damage repair was a significant mechanism of enhancement of tumor radioresponse.

Cell Cycle Effects The cytotoxicity of most chemotherapeutic agents and ionizing radiation depends greatly on the proliferative state of the cells, with most agents being more effective against proliferating cells than nonproliferating cells. Sensitivity to such agents also varies according to the phase of the cell cycle. The influence of cell cycle in response to radiation

2

4

6

8

Absorbed dose (Gy)

FIGURE 4-6.  Survival curves are plotted for astrocytoma cells treated with radiation and paclitaxel. Survival is presented at a particular radiation dose and expressed relative to that of nonirradiated controls given the same dose of paclitaxel. Points are means ± standard error. Symbols represent untreated control cells (blue circles), cells treated with dimethylsulfoxide (red circles), 1 nM paclitaxel (open green triangles), or 10 nM paclitaxel (closed green triangles). (From Tishler RB, Geard CR, Hall EJ, et al. Taxol sensitizes human astrocytoma cells to radiation. Cancer Res 1992;52:3495-3497.)

was first reported in 1963 by Terasima and Tolmach.25 In general, cells in the G2 and M phases are the most sensitive, and cells in the S phase are the most resistant. This variation in radiosensitivity over the cell cycle can be exploited for designing effective chemoradiation therapy strategies such as the use of drugs that accumulate cells in a radiosensitive phase or eliminate radioresistant S-phase cells. Two examples can illustrate these concepts. Taxanes, of which paclitaxel and docetaxel are prototypes, are potent mitotic spindle poisons. Taxanes inhibit tubulin depolymerization by binding to β-tubulin, increasing tubulin polymerization and promoting microtubule ­assembly and stability, which collectively leads to the cells becoming arrested in the radiosensitive G2 and M phases of the cell cycle.26 The ability of taxanes to block cells in G2/M was the biologic rationale that led to extensive testing of their radioenhancing properties. In the first report on the radioenhancing effect of taxanes,27 the radiosensitivity of a human astrocytoma cell line was reportedly increased by a factor of 1.8 (Fig. 4-6). Radiosensitization was achieved only when the cells were in phase G2 or M at the time of irradiation, a result subsequently confirmed in many in vitro and in vivo studies.28 Cells arrested in mitosis by taxanes often die by apoptosis (the underlying basis for the antitumor efficacy of these drugs), but they may overcome the mitotic block and continue with division. In that case, tumors are resistant to treatment with taxanes alone, but they can show an enhanced radioresponse if radiation is delivered at the time of mitotic arrest, usually between 6 and 12 hours after administration of the drug. (Tumor reoxygenation, another major mechanism of radioenhancement in tumors that respond to taxanes by mitotic arrest and apoptosis, is discussed in more detail under ­“Hypoxia.”) The histologic

108

PART 1  Principles

FIGURE 4-7.  The histologic appearance of mouse MCa-4 mammary tumors treated with 33 mg/kg of docetaxel features mitotically arrested cells (short arrows) and apoptotic cells (long arrows) 12 hours after treatment with docetaxel. Mitotically arrested cells show the characteristic coronal appearance (×1000).

appearance of mitotically arrested and apoptotic cells in murine MCa-4 tumors treated with docetaxel is shown in Figure 4-7. Another approach for exploiting the cell cycle effect in chemoradiation therapy is to use drugs that eliminate radioresistant S-phase cells. In addition to inhibiting radiation damage repair, nucleoside analogues such as fludarabine or gemcitabine become incorporated into S-phase cells, many of which then die by apoptosis.5,17,24,29 This preferential removal of the S-phase component from tumors leads to a parasynchronous movement of surviving cells through S phase and mitosis, resulting in accumulation of cells in the G2 and M phases of the cell cycle between 24 and 36 hours after administration of the drugs, when the highest radioenhancement values are observed. Elimination of radioresistant S-phase cells from the tumor cell population and redistribution of surviving cells into more radiosensitive compartments of the cell cycle seem to be involved in the radioenhancement.

Hypoxia The presence of oxygen has long been known to influence cell radiosensitivity; hypoxic cells are 2.5 to 3 times less sensitive to radiation than are well-oxygenated cells. Because tumors often include hypoxic areas but normal tissues usually do not, the radioresistant effect of hypoxia typically is restricted to tumors. Blood circulation within solid tumors tends to be poor because of abnormalities in the number, structure, and function of blood vessels within the tumor. Tumor blood vessels are commonly irregular and tortuous, and they have blind ends, arteriovenous shunts, and incomplete endothelial linings and basement membranes. All of these factors can limit intratumoral blood flow and the delivery of oxygen and nutrients, resulting in many tumor microregions becoming hypoxic, acidic, and eventually necrotic. Hypoxia develops in tissues when the distance from blood vessels exceeds about 100 to 150 μm; the proportion of hypoxic cells in tumors can often exceed 50%. Cells in hypoxic regions stop proliferating. In contrast, cells that are nearer to blood vessels are well oxygenated and actively proliferate.

Chemotherapeutic agents can improve the antitumor efficacy of radiation therapy by reducing or eliminating the negative (tumor-protective) effects of hypoxic cells. This improvement can be achieved by (1) eliminating ­ tumor cells in oxygenated regions, (2) selectively eliminating ­hypoxic cells, or (3) sensitizing hypoxic cells to radiation. Elimination of cells in oxygenated regions leads to tumor reoxygenation (see Chapter 1), and most chemotherapeutic drugs can achieve this effect because they preferentially kill well-oxygenated, proliferating cells. Eliminating oxygenated cells makes more oxygen available to the surviving cells by lowering the interstitial pressure on microvessels within the tumor, thereby reopening closed capillaries and increasing the intratumoral blood supply. Tumor shrinkage from cell loss and active migration of tumor cells also brings previously hypoxic microregions closer to blood vessels. Tumor reoxygenation has been shown to be a major mechanism by which taxanes enhance tumor radioresponsiveness (Fig. 4-8).30 In that study, giving paclitaxel to mice bearing an adenocarcinoma within 3 days before local irradiation enhanced the radiosensitivity of the tumors. However, this enhancing effect was reduced when the tumors were rendered hypoxic by clamping the tumor-bearing legs at the time of irradiation. Paclitaxel’s ability to increase ­tumor oxygenation was confirmed directly by measuring the intratumoral partial pressure of oxygen (Po2).30 Another chemotherapeutic approach that increases ­tumor radiosensitivity is to use drugs that selectively kill hypoxic cells. Several drugs with this property are available, including tirapazamine and mitomycin31; their cytotoxicity results from their reductive activation under hypoxic conditions. Some active metabolites can diffuse into surrounding oxygenated regions and kill oxygenated cells. Hypoxia is associated with other metabolic alterations, including the development of low pH from the production of lactic acid through hypoxia-driven anaerobic metabolism.32 The acidity is extracellular and does not seem to affect intracellular pH.32 The acidic state of tumors can be therapeutically exploited by using drugs that selectively accumulate in acidic compartments or become activated at low pH values.33 Another approach to reducing the negative influence of hypoxic cells is to use drugs that selectively radiosensitize the cells.34 These drugs mimic the effect of oxygen in increasing radiation damage. Misonidazole and other hypoxic cell radiosensitizers are highly effective in enhancing the radioresponsiveness of tumors in rodents. Among the many clinical trials conducted, only a few have shown improved outcome from the combination of hypoxic-cell radiosensitizers with radiation therapy. The major limitation of these combinations in humans has been neurotoxicity, which precludes the use of effective doses of the hypoxic-cell ­radiosensitizers.

Cell Regeneration In normal adult tissues, a balance is maintained between cell production and cell loss such that no net growth occurs. Radiation or chemotherapeutic drugs alter this balance by increasing cell loss; to restore the balance, the rate of cell production among the surviving cells increases. In clinical radiation therapy, which is typically given over several weeks, the cell loss that occurs after each fractional dose is counteracted by the compensatory cell regeneration (i.e., repopulation) that takes place between fractions. The

109

4  Principles of Combining Radiation Therapy and Chemotherapy 100

80

Air

Control 60 Gy-primed

70 80

TCD50 (Gy)

Tumor cure (%)

60 60

40

40 20

30

20

0

0

A

14

21

FIGURE 4-9.  Values for the radiation dose yielding local ­tumor control in 50% of the animals tested (TCD50) for mouse MCa-4 tumors are plotted with 95% confidence intervals as a ­function of time after tumor cell inoculation (blue squares) or after 8-mm ­ established tumors had been exposed to 60 Gy (red squares). The linear increase in TCD50 is consistent with exponential growth, which was significantly faster in the pre­ ir­radiated ­ tumors. (From Milas L, Yamada S, Hunter N, et al. Changes in TCD50 as a measure of clonogen doubling time in irradiated and unirradiated tumors. Int J Radiat Oncol Biol Phys 1991;21:1195-1202.)

Hypoxia

80

Tumor cure (%)

7 Days

100

60

40

20

0 30

B

50

40

50

60

70

80

90

Radiation dose (Gy)

FIGURE 4-8.  The effect of paclitaxel on radiation dose-­response curves is plotted for local control of murine MCa-4 tumors. Mice were irradiated under oxygenated (A) or hypoxic (B) conditions. Horizontal lines represent 95% confidence limits at the radiation dose yielding local tumor control in 50% of the animals tested  (TCD50). (From Milas L, Hunter NR, Mason KA, et al. Role of reoxygenation in induction of enhancement of tumor radioresponse by paclitaxel. Cancer Res 1995;55:3564-3568.)

extent of cell repopulation determines the tolerance of acutely responding tissues to radiation. In malignant tumors, the balance between cell production and cell loss is shifted in favor of cell production, ­resulting in uncontrolled growth. In tumors and in normal tissues, radiation and other cytotoxic treatments produce massive cell loss, which stimulates a compensatory regenerative ­response. Studies of cell regeneration in experimental tumors after single and fractionated radiation doses have shown that the rate of cell proliferation usually is higher in treated tumors than in untreated tumors, a phenomenon called ­accelerated cell proliferation.35-38 An ­example of ­accelerated proliferation after tumor irradiation is shown in Figure 4-9, for which the rate of cell proliferation was

­ easured by changes in TCD50 values. In this study, estabm lished 8-mm murine mammary carcinomas were irradiated with a single dose of 60 Gy, which allowed only a small number of tumor clonogens to survive. At different days thereafter, while the tumors regressed, transiently disappeared, or ­ regrew, the tumors were irradiated again with graded doses to establish TCD50 values. The later doses were given ­under hypoxic conditions to exclude the influence of oxygen on tumor radioresponse. The TCD50 ­values of tumors, regardless of size, that received no priming radiation dose served as controls. As shown in Figure 4-9, the slope of the TCD50 curve for the 60-Gy–primed tumors was significantly steeper than that for the control tumors, implying an increased rate of clonogen proliferation in the irradiated tumors. A similar accelerated repopulation was observed in murine tumors treated with cyclophosphamide.39 Accelerated proliferation of tumor clonogens seems to occur during clinical radiation therapy. Withers and associates40 reviewed the relationships among the probability of achieving local control of head and neck carcinoma, ­total radiation dose, and treatment duration. In that study, the TCD50 increased progressively over time if the treatment period was prolonged beyond 30 days (Fig. 4-10). The increase in curative dose was greater than anticipated from estimates of pretreatment tumor volume and clonogen doubling rates. In the study, the dose increase needed to offset accelerated clonogen repopulation was about 0.6 Gy/day; however, that increase can exceed 1 Gy/day.41 Acutely responding normal tissues and tumors respond to radiation- or chemotherapy-induced injury with increased clonogen production. Although accelerated cell proliferation can be beneficial in terms of sparing normal tissues,

Tumor dose (Gy) normalized to 2-Gy fractions

110

PART 1  Principles

70

60

50 HBO

Miso HBO

40 10

20

30

40

50

60

Treatment duration (days)

FIGURE 4-10.  Values for the radiation dose yielding local ­tumor control in 50% of the animals tested (TCD50) are ­ plotted as a function of overall treatment times for squamous cell carcinomas of the head and neck. Total doses were normalized to the dose equivalent of that of a regimen of 2-Gy fractions ­using an α/β value of 25 Gy. Doses and times are best estimates of ­median values. The dose and control rate reported in the ­literature from which the TCD50 values were calculated are shown as blue circles to indicate the extent of the extrapolation. The rate of increase in TCD50 was predicted from a 2-month clonogen-doubling rate (dashed line). Estimated increases in TCD50 (solid lines) over the radiation-treatment period are shown for T2 tumors (purple circles), T3 tumors (red squares), and tumors of mixed T stage (green triangles). HBO, hyperbaric oxygen; Miso, misonidazole. (From Withers HR, Taylor JMG, ­ Maciejewski B. The hazard of accelerated tumor clonogen repopulation ­during radiotherapy. Acta Oncol 1988;27:131-146.)

it also reduces the tumor-control efficacy of ­ irradiation. Any approach that reduces or eliminates ­accelerated clonogen proliferation in tumors would ­improve the efficacy of ­radiation therapy. This is likely one of the major mechanisms by which chemotherapeutic drugs improve ­local ­tumor control when given concurrently with radiation therapy. Even a small decrease in repopulation ­between ­radiation fractions can significantly improve ­tumor ­response to fractionated radiation therapy. ­However, care must be taken to select drugs that preferentially ­affect rapidly ­proliferating cells and preferentially localize in malignant tumors. ­Another possibility is to supplement chemoradiation therapy with agents that have antiproliferative properties, such as those that block membrane receptors for growth factors or interfere with signaling pathways involved in cell ­proliferation.42 Increased repopulation of tumor cells may also play an important role in induction (neoadjuvant) chemotherapy. The rationale for this form of therapy is to reduce the number of clonogenic cells, improve tumor oxygenation, and eradicate tumor cells that are intrinsically resistant to ­radiation. According to radiobiologic principles, all of these ­effects should improve tumor response to irradiation. Many clinical trials have been conducted in which induction chemotherapy has been combined with radiation therapy, but the results have not been greatly encouraging. Although large proportions of patients in these trials ­ responded to chemotherapy by showing complete or partial tumor ­regression, significant improvement in treatment outcome has yet to be achieved. Several possible explanations exist

for the failure of induction chemotherapy to ­improve radiation therapy. One possibility is that response to chemotherapy predicts the response to radiation therapy. According to this reasoning, tumor cells that are sensitive or resistant to chemotherapeutic drugs are also sensitive or resistant to radiation, and tumors that respond well to chemotherapy would be controlled with radiation therapy even if they had not been treated with chemotherapy. Another possibility is that chemotherapy induces or selects cell clones that are resistant to radiation; however, this explanation is not supported by the fact that most of the drugs that have been used for neoadjuvant chemotherapy do not produce crossresistant effects with radiation. The most likely explanation seems to be that chemotherapy induces accelerated repopulation of tumor cell clonogens, an explanation for which some experimental evidence has been found.39

Inhibition of Tumor Angiogenesis The antitumor effects of chemotherapeutic agents are not restricted to killing tumor cells. They also include the ability to affect normal tissue structures within the tumor, primarily by their cytotoxicity to endothelial cells of tumor blood vessels. Angiogenesis (i.e., formation of blood vessels) is a prerequisite to the progressive growth of malignant tumors beyond small microscopic aggregates of tumor cells. Endothelial cells in tumor blood vessels exhibit high rates of proliferation, making them sensitive to the cytotoxic actions of chemotherapeutic agents. Increasing evidence shows that chemotherapeutic agents have antiangiogenic effects.43-48 For example, docetaxel was reported to suppress endothelial cell proliferation and tubule assembly in vitro and vessel formation in an in vivo Matrigel plug assay.43 These antiangiogenic effects were partly blocked by the presence of the angiogenic factors vascular endothelial cell growth factor (VEGF) and basic fibroblast growth factor (bFGF). This protective effect, however, could be overcome by anti-VEGF antibody or by combining docetaxel with 2-methoxyestradiol, another antiangiogenic agent. In another study, docetaxel was shown to inhibit the in vivo formation of blood vessels in mice at the intradermal injection site of tumor cells.44 This antiangiogenic activity is probably another mechanism by which taxanes (and probably other chemotherapeutic agents) enhance tumor response to irradiation, as sug­ gested by increasing preclinical evidence that inhibitors of angiogenesis can enhance tumor radioresponse by a variety of mechanisms, including inhibition of VEGF and bFGF, which typically act as radioprotectants for tumor cells.49-51 VEGF is also a potent vessel permeability factor that causes fluid accumulation in extracapillary spaces and consequent impairment of blood flow and oxygen supply to tumor cells; inhibition of VEGF by chemotherapeutic agents would result in increased tumor oxygenation. Other preclinical evidence suggests that frequent administration of small doses of chemotherapeutic agents preferentially targets tumor endothelial cells and that this strategy has better antitumor efficacy and lower toxicity than high-dose chemotherapy given less frequently.45-48 It is reasonable to expect that this approach, referred to as antiangiogenic chemotherapy, low-dose chemotherapy, or metronomic chemotherapy, would enhance the effect of radiation therapy, perhaps even more so than high-dose chemotherapy, in vivo.

4  Principles of Combining Radiation Therapy and Chemotherapy

Sequencing of Chemotherapy and Radiation Therapy Depending on the principal aim of chemoradiation ther­ apy, drugs can be given before, during, or after the course of radiation therapy. According to this sequencing, chemo­ therapy is designated induction (neoadjuvant) chemo­ therapy when it is given before irradiation, concurrent chemotherapy when it is given during radiation therapy, and adjuvant chemotherapy when it is given after irradiation.

Induction Chemotherapy Induction or neoadjuvant chemotherapy has two main objectives. One is to eradicate micrometastases while they contain small numbers of tumor cells. For this reason, induction chemotherapy is begun soon after tumor diagnosis and can precede radiation therapy by weeks or months. The other objective of induction chemotherapy is to reduce the size of the primary tumor that is to be irradiated. Reducing the number of clonogenic cells in the tumor in this way increases the probability of tumor control by irradiation. Chemotherapy-induced tumor shrinkage may allow a smaller tissue volume to be irradiated, limiting damage to normal tissues. The latter strategy is particularly important in treating lymphomas and tumors in children. Induction chemotherapy generally has not provided the therapeutic benefits that might be expected from theoretical principles. One reason is that chemotherapy, like irradiation, may accelerate the proliferation of tumor cell clonogens. Another possibility is that induction chemotherapy may select for or induce drug-resistant cells that may be cross-resistant to radiation therapy. Development of drug resistance remains a significant problem in chemotherapy; nevertheless, evidence is lacking to convincingly demonstrate that cells that acquire drug resistance are also resistant to radiation. Chemotherapeutic drugs can, ­under ­certain conditions, increase the metastatic potential of ­tumor cells or damage normal tissues in such a way as to make them conducive to metastatic growth.

Concurrent Chemotherapy In this mode of chemoradiation therapy, chemotherapeutic drugs are given during a course of radiation treatment so that the drug or drugs exert their effects during several radiation fractions. Even though drugs can act on systemic and primary lesions, the main objective in concurrent chemotherapy is to use drug-radiation interactions to maximize tumor radioresponse. The optimal schedule for drug administration (e.g., once each week, daily, at some other interval) depends on many factors, primarily the mechanisms of radioenhancement and the particular drug and the conditions under which the enhancement is highest and the toxicity of the treatment to normal tissue. In this type of chemoradiation therapy, therapeutic benefit occurs only when the enhancement of tumor response is greater than the toxic effects on critical normal tissues. Concurrent chemoradiotherapy has provided the best clinical results in terms of local tumor control and patient survival. In one mode of concurrent chemoradiation therapy, chemotherapy and radiation therapy are alternated rapidly (e.g., radiation is given one week and chemotherapy during the subsequent week, or vice versa), and this pattern is

111

continued over the next few weeks. Alternating schedules such as these are given in an attempt to minimize excessive toxic effects on normal tissues (e.g., mucositis) and to enhance tumor response by exploiting some biologic effect of an agent that renders tumor cells more responsive to the subsequent agent (e.g., perturbing cell cycling, ­reoxygenation).

Adjuvant Chemotherapy Adjuvant chemotherapy is given after a course of radiation therapy is completed. The primary objective of adjuvant chemotherapy is to eradicate disseminated disease, although drugs also can eradicate surviving tumor cells in irradiated tumors.

Clinical Applications of Combined Radiation Therapy and Chemotherapy Laboratory investigations are providing important pre­ clinical data regarding the interactions between combin­ ations of cytotoxic drugs and radiation, but sequential clinical investigations are required to establish the toxicity, effectiveness, and value of combined chemotherapy and radiation therapy for the care of human patients. The effects of drugs, given as single agents or in combination with other drugs, have usually been investigated without the addition of radiation therapy. However, for diseases for which radiation therapy is the standard of care, the addition of chemotherapy before irradiation is common, even when the value of such an addition is unknown. Separating the administration of the drug and the radiation by 2 weeks or more usually does not increase toxic reactions, and induction chemotherapy often produces some degree of response. The situation is quite different for concurrent chemotherapy and radiation therapy, for which phase I clinical trials are needed to test a standard dose of radiation with increasing doses of chemotherapy or a standard dose of chemotherapy with increasing doses of radiation. After the maximum tolerated combination of chemotherapy and radiation therapy is established, phase II trials can be used to test the efficacy of the combination in one or more diseases. If phase II studies suggest that a particular combination is efficacious, phase III comparative trials are required to compare the standard treatment against the new combination. Rarely or never are phase II results sufficiently compelling to change practice throughout a nation or the world. Comparisons with historical control subjects are often undertaken, but known or unsuspected biases often influence the interpretation of such comparisons.

Sequential Chemotherapy and Radiation Therapy Induction Chemotherapy Induction chemotherapy relies heavily on the principle of spatial cooperation. For epithelial tumors, little evidence exists to suggest that the addition of chemotherapy before irradiation confers any benefit in controlling locoregional tumors; rather, the chemotherapy is directed at subclinical metastases. Only for malignant lymphomas does induction

112

PART 1  Principles

chemotherapy seem to be beneficial in terms of local tumor control. For example, for diffuse large cell lymphomas that are less than 3.5 cm in their greatest dimension, induction chemotherapy that produces a complete response coupled with subsequent irradiation can lead to successful local control at a low total dose of radiation (e.g., 30 Gy). However, tumors 10 cm in diameter or larger that respond completely to induction chemotherapy nonetheless require substantially higher total doses (at least 45 Gy) than do smaller tumors.52 In contrast to lymphomas, locoregional control of carcinomas is not improved by the addition of induction chemotherapy.53,54 The response of locoregional tumors to induction chemotherapy can be misleading. Many investigators have been convinced that initial shrinkage of the ­tumor must eventually lead to improvements in locoregional control. However, this has not been borne out for carcinomas of the upper aerodigestive tract54; moreover, phase III comparative trials of induction chemotherapy for cancer of the cervix have shown survival to be better after radiation therapy alone than after induction chemotherapy followed by pelvic irradiation.55,56 In some circumstances, the use of induction chemotherapy before irradiation can improve survival over that produced by radiation therapy alone. The best ­example is non–small cell carcinoma of the lung, for which ­cisplatinbased induction chemotherapy has been shown to ­improve survival rates over those produced by radiation therapy alone in three independent studies.57-59 ­ Despite the fact that induction chemotherapy has not been ­ particularly ­effective in other diseases, many oncologists feel compelled to use chemotherapy before radiation. The ­rationales for this practice include the convenience of ­being able to start chemotherapy quickly rather than having to wait to plan radiation therapy, the ability to give the maximum doses of chemotherapy if no radiation therapy is to be given, and the long-standing belief that the subclinical burden is smallest at the time of diagnosis. However, because induction chemotherapy does not affect local control of epithelial ­ tumors and because in some ­ circumstances (e.g., carcinoma of the cervix) starting chemotherapy can ­ delay effective local treatment and produce inferior ­results, ­ induction chemotherapy should be used only when a phase III comparative trial has ­ demonstrated a clear ­benefit. The toxicity associated with combining chemotherapy and radiation therapy can be minimized by separating them in time, another feature that can make induction chemotherapy more attractive. Occasionally, induction chemotherapy can even increase the protection of normal tissues. This is true for patients with Hodgkin disease who present with a large mediastinal mass; shrinking the mass with chemotherapy means that more lung can be protected in the course of subsequent radiation therapy.

Concurrent Chemotherapy and Radiation Therapy The concurrent use of chemotherapy and radiation ther­ apy tends to maximize the antitumor effect, even though it almost inevitably increases the acute toxicity of the treatment. Concurrent chemoradiation therapy has been shown to improve survival of patients with several types of carcinoma (Table 4-3). In a study of carcinoma of the esophagus, cisplatin and 5-fluorouracil (5-FU) given concurrently with modest-dose irradiation (50 Gy) pro­duced improved survival relative to high-dose radiation therapy (64 Gy) alone.62 In another study, the use of cisplatin concurrent with radiation therapy followed by cisplatin and 5-FU improved survival over radiation therapy alone for patients with carcinoma of the nasopharynx.63 Concurrent chemotherapy and radiation therapy also improved survival among patients with moderately advanced carcinoma of the cervix compared with radiation therapy alone.64 Other supporting studies clearly confirmed the value of the combined treatment.65 Although the findings for carcinoma of the nasopharynx are compelling, the role of concurrent chemotherapy and radiation therapy for other sites in the upper aerodigestive tract is less clear. In one meta-analysis of head and neck squamous cell carcinoma,54 an advantage was found for concurrent chemotherapy and radiation therapy but not for induction chemotherapy. In a trial comparing concurrent chemoradiation with hyperfractionated irradiation to a total dose of 75 Gy for patients with locally advanced head and neck cancer,66 concurrent chemotherapy and radiation therapy proved superior to irradiation alone in terms of local control and might have been beneficial for overall survival. The improvement in survival produced by ­ induction chemotherapy for unresectable non–small cell ­ carcinoma of the lung is based entirely on control of distant ­metastasis. Two studies67,68 have tested the hypothesis that TABLE 4-3.   Chemotherapy and Radiation Therapy for Carcinomas: Survival According to Therapy Sequence* Cancer Type

Induction

Concurrent

Head and neck54

±

+

Nasopharynx63

+

NSCLC53,54,59,67,68

+

+

SCLC54,69,79

+

+

Esophagus62,76,77

±S

+

Stomach80

S+

Rectum72-74

S+

Prostate60

Adjuvant Chemotherapy

Cervix64,65

With one exception, adjuvant chemotherapy (i.e., chemo­ therapy that is delivered after radiation therapy) is not particularly valuable. That exception is in the use of adjuvant hormonal therapy for prostate cancer. Diseasefree and overall survival rates have been improved by following radiation therapy to the prostate with 2 to 3 years of androgen-suppression therapy.60,61

Breast81,82 *Based

Adjuvant

+(H) −

+ S+

only on phase III comparative trials. H, hormone therapy; NSCLC, non–small cell lung cancer; SCLC, small cell lung cancer; S, surgical adjuvant radiation and chemotherapy; +, increased survival compared with ­radiation therapy; −, decreased survival compared with radiation therapy; ±, conflicting results (increased, decreased, and no difference).

4  Principles of Combining Radiation Therapy and Chemotherapy

c­ oncurrent ­ chemoradiation would be superior to induction ­chemotherapy by virtue of improving control of ­local ­disease and distant metastasis. In one, a comparative study,67 concurrent chemoradiation produced a statistically significant ­ improvement in overall survival over that produced by induction chemotherapy. Preliminary results from the ­Radiation Therapy Oncology Group (RTOG) protocol 9410 also support the benefit of concurrent chemoradiation over induction chemotherapy followed by standard ­irradiation.68 The value of radiation therapy in combination with chemotherapy for small cell carcinoma of the lung is ­unclear; this type of cancer has been so responsive to combination chemotherapy that radiation therapy is considered by some to be unnecessary. Several trials comparing chemotherapy alone versus chemotherapy with concurrent or adjuvant ­irradiation for small cell lung cancer have provided conflicting results. A meta-analysis suggested a small but statistically significant benefit to the use of radiation therapy over the use of chemotherapy alone. However, the optimal timing and sequencing of the treatments remain controversial. A study at the National Cancer Institute of ­Canada69 indicated that early use of chemoradiation therapy was ­superior to delayed irradiation of the thorax. The best ­results found with regard to limited-stage small cell carcinoma of the lung have come from a national ­intergroup trial comparing accelerated fractionation with standard fractionation, both of which were given concurrently with chemotherapy; however, that study did not address the difference between induction and concurrent chemoradiation. The only trial to have directly addressed that question indicated that the concurrent approach was superior.70 In contrast to the diseases previously described, carcinoma of the cervix has a relatively favorable prognosis after radiation therapy alone. The 5-year survival rate for patients with invasive cancer of the cervix (all stages combined) treated with radiation therapy is ­ approximately 60%. Induction chemotherapy does not improve this percentage and can have a deleterious effect on survival. However, a series of trials provided convincing evidence that concurrent chemoradiation therapy can improve on the results of radiation therapy. The RTOG protocol 9001 compared concurrent cisplatin and 5-FU with pelvic and intracavitary irradiation versus radiation therapy alone for cervical cancer.64 The concurrent chemoradiation strategy produced significantly better disease-free and overall survival rates and provided better local tumor control and freedom from distant metastasis. Other cooperative trials from the ­Gynecologic Oncology Group and the Southwest Oncology Group have emphasized the importance of concurrent cisplatin in treating ­cervical cancer.65

Chemoradiation plus Surgery Chemotherapy and radiation have been used in combination with surgery in attempts to produce better survival rates and organ conservation than can be obtained by surgical treatment alone. However, the results from trimodality treatments are difficult to compare and interpret by virtue of the complexities of treatment sequencing and timing, and the relative contributions of the modalities remain unclear. For example, the Gastrointestinal Tumor Study Group71 reported that use of combined chemoradiation therapy after surgery prolonged disease-free survival time among patients with rectal carcinoma compared with those who

113

had had only surgery. In another study of rectal cancer,72 postoperative chemotherapy delivered before, during, and after pelvic radiation therapy produced significantly better survival rates than did postoperative pelvic radiation therapy alone. The chemotherapy in that study included bolus-dose 5-FU; results from a later study indicated that protracted-infusion 5-FU produced better results than did bolus dosing.73 Another study by the National Surgical Adjuvant Breast and Bowel Project group found that among patients with rectal cancer, survival after postoperative chemoradiation was no better than that after postoperative chemotherapy alone.74 Controversies about the relative contributions of chemotherapy and radiation therapy for rectal cancer are not likely to be resolved any time soon, as new chemotherapeutic agents are tested in postoperative settings with or without pelvic irradiation and as more interest develops in the preoperative administration of various combined treatments. Issues regarding the com­ bination of radiation therapy with surgery are discussed in further detail in Chapter 3. Combinations of surgery, chemotherapy, and radiation therapy have been studied for potentially resectable carcinoma of the esophagus. No prospective randomized trial has shown that postoperative radiation therapy produces better results than surgical resection alone for carcinoma of the esophagus, nor has any evidence been found to suggest that preoperative irradiation or preoperative chemotherapy is beneficial in this disease.75 However, ­ considerable interest has been expressed with regard to induction chemotherapy given concurrently with radiation therapy followed by surgical resection. In one such study involving adenocarcinoma of the esophagus that included tumors of the gastric cardia, induction chemoradiation followed by surgery significantly improved ­survival rates compared with patients who underwent only surgery.76 However, the survival rates for the surgery-only group in that trial were substantially lower than those reported in a later study.75 A larger study of chemoradiation given before surgical resection for squamous cell carcinoma of the esophagus showed no evidence of benefit from this treatment.77 Tests of concurrent chemoradiation in the postoperative setting have shown evidence of benefit. A phase III comparative trial of concurrent cisplatin and external ­radiation therapy versus external radiation therapy alone was conducted in patients with high-risk squamous cell carcinoma of the head and neck after total resection of all evident disease. The rate of locoregional recurrence was significantly lower in the concurrent therapy group than in the ­radiation-only group. However, survival rates were no different between the two groups, and the incidence of severe acute adverse effects in the combined-modality group was more than double that in the irradiation-only group.78 In summary, the sequence of combination treatment that has produced the best results seems to be ­concurrent chemoradiation therapy (see Table 4-3).53,54,59-65,67-69,72-74,76,77,79-82 However, considerable variation is evident among disease sites and according to whether surgery was also used in the treatment.

Organ Conservation Conservation of organ structure and function remains an important goal even when survival may not be improved by treatment. One of the earliest demonstrations of this

114

PART 1  Principles

principle was the use of radiation therapy to avoid the need for a permanent colostomy in patients with advanced anal carcinoma. Chemotherapy combined with radiation therapy is now considered the treatment of choice for carcinoma of the anal canal.83 A phase III trial is being conducted to address whether mitomycin can improve outcome compared with the use of 5-FU and irradiation for rectal cancer (RTOG protocol 8704).84 Although some physicians prefer to use a combination of 5-FU and cisplatin with radiation therapy, the equivalence or superiority of these drugs to 5-FU and mitomycin has not been demonstrated.

Rectal Carcinoma The success of chemoradiation therapy in preserving function in patients with anal carcinoma led to widespread interest in the use of preoperative chemoradiation for carcinoma of the distal rectum (i.e., lesions occurring less than 6 cm from the anal verge). Traditional treatment for these lesions involves abdominal peritoneal resection and permanent colostomy. However, for some patients, preoperative chemoradiation (usually involving 5-FU, alone or in combination with other drugs) may reduce the tumor volume and peripheral extensions so that more limited surgical procedures that do not require a colostomy can be performed. No prospective studies have compared chemoradiation followed by limited resection versus limited resection alone versus abdominoperineal resection.

Laryngeal Carcinoma Total laryngectomy is accepted as being the single most effective treatment for laryngeal carcinoma. However, radiation therapy can be used to treat T1 and T2 lesions while allowing conservation of the larynx and laryngeal speech. Interest in laryngeal conservation for more advanced lesions was generated by results of a clinical trial within the U.S. Veterans Administration system.85 In that trial, patients were randomized to receive standard

laryngectomy followed by radiation therapy or to receive neoadjuvant therapy with cisplatin and 5-FU followed by radiation therapy for those who achieved a 50% or better tumor response to the chemotherapy. Survival among the first group (i.e., surgery plus radiation) was slightly but not significantly better than among those who had laryngectomy alone, but laryngeal function could be preserved in more than 60% of patients who underwent chemoradiation. These findings led to a large intergroup trial, coordinated by the RTOG, in which initial chemotherapy followed by radiation therapy was compared with radiation therapy alone and with a third arm in which radiation therapy was given concurrently with cisplatin. The results of this trial showed that concurrent chemotherapy and radiation therapy resulted in higher rates of locoregional control and laryngeal preservation than induction chemotherapy followed by radiation therapy or radiation therapy alone.86

Bladder Carcinoma Given sufficient motivation on the part of both patients and physicians, bladder conservation is possible in some cases of invasive bladder cancer through the use of chemotherapy, radiation therapy, and surgery. The most common sequence is to begin with transurethral resection of the bladder tumor, followed by two or more cycles of chemotherapy with methotrexate, cisplatin, and vinblastine, followed by pelvic irradiation with concomitant cisplatin. Careful follow-up using cystoscopy and biopsy is required, and cystectomy is reserved for persistent disease. In two studies of this approach, bladder conservation was possible for 40% to 50% of patients with T2 to T4a carcinomas of the bladder.87,88

Toxicity Issues All chemotherapeutic agents have clinically important toxic effects. An overview of the organ-specific toxicity of commonly used drugs or classes of drugs is provided in Table 4-4. Most chemotherapeutic agents are toxic to

TABLE 4-4.    Effects of Cytotoxic Drugs on Normal Tissues Drugs of Classes

Bone Marrow

GI

Liver

Kidney

Lung

Heart

Anthracycines

+++

+

+

+++

Alkylators

+++

++

Antifolates

+++

++

+

++

+

+

++

+

Antipyrimidines

+++

+++

+

+

Dacarbazine

+++

Taxanes

+++

Podophyllotoxins

PNS/CNS

++ +/+++

+

+

+

+

+++/+

++

+

+

Hydroxyurea

++

+

Mitomycin

++

Procarbazine

++

Vinca alkaloids

+

++

Cisplatin

+

+++

Bleomycin

+

+++

Gonads

++

+ + +++/+

+++

++/+ +++

CNS, central nervous system; GI, gastrointestinal; PNS, peripheral nervous system; +, some reports of toxicity, usually not major; ++, major toxicity not unusual; +++, principal toxicity expected.

4  Principles of Combining Radiation Therapy and Chemotherapy

bone marrow; however, several also are toxic to other organs, which must be taken into account when combining chemotherapy with radiation therapy. The sequential use of chemotherapeutic agents and ­radiation therapy usually does not increase toxicity. ­Although both modalities affect the bone marrow, the effects of radiation on bone tend to be localized unless the irradiated volume is very large. The sequential use of bleomycin and radiation therapy is an important exception, ­because the pulmonary toxicity of bleomycin can be greatly accentuated by subsequent radiation therapy. This effect is probably of greatest importance in the treatment of Hodgkin disease. Concurrent chemoradiation therapy is associated with much greater toxicity than is the sequential use of either modality. Drugs such as doxorubicin and gemcitabine, when given at the standard doses used to treat systemic disease, profoundly increase the effects of radiation therapy. These combinations can lead to life-threatening toxic effects at total doses of radiation that are usually well tolerated. Similarly, the concurrent use of chemotherapy and irradiation can greatly increase esophageal reactions when cisplatin and etoposide are given concurrently with irradiation for carcinoma of the lung. Two major strategies are available to enhance the ­effectiveness of concurrent treatment while avoiding its toxic effects. The first strategy is to use conformal radia­ tion therapy to specifically avoid irradiating normal tis­ sues that have the most severe reactions.89 The second strategy is to use cytoprotective agents such as amifos­ tine.90 These and other agents are being investigated for the degree of protection they produce during concurrent chemotherapy and radiation therapy. Newer cytoprotec­ tive drugs may become available within the next few years that may enhance the therapeutic ratio of concurrent che­ motherapy and radiation therapy, which remains the most compelling chemoradiation strategy for achieving tumor control.

REFERENCES   1. Steel GG, Peckham MJ. Exploitable mechanisms in combined ­radiotherapy-chemotherapy: the concept of additivity. Int J ­Radiat Oncol Biol Phys 1979;5:85-91.   2. Yuhas JM, Storer JB. Differential chemo-protection of normal and malignant tissues. J Natl Cancer Inst 1969;42:331-335.   3. Milas L, Hanson WR. Eicosanoids and radiation. Eur J Cancer 1995;31A:1580-1585.   4. Hahn SM, Krishna MC, Samuni A, et al. Potential use of nitroxides in radiation oncology. Cancer Res 1994;54:2006S-2010S.   5. Buntzel J, Kuttner K, Frohlich D, et al. Selective cytoprotection with amifostine in concurrent radiochemotherapy for head and neck cancer. Ann Oncol 1998;9:505-509.   6. Hall EJ. Radiobiology for the Radiologist, 4th ed. Philadelphia: JB Lippincott, 1993.   7. Kallman RF (ed). Rodent Tumor Models in Experimental Cancer Therapy. New York: Pergamon Press, 1987.   8. Begg AC. Principles and practices of the tumor growth delay assay. In Kallman RF (ed): Rodent Tumor Models in Experimental Cancer Therapy. New York: Pergamon Press, 1987, pp 114-121.   9. Suit HD, Sedlacek R, Thames HD Jr. Radiation dose-response ­assays of tumor control. In Kallman RF (ed): Rodent Tumor Models in Experimental Cancer Therapy. New York: Pergamon Press, 1987, pp 138-148. 10. Mason KA, Hunter NR, Milas M, et al. Docetaxel enhances tumor radioresponse in vivo. Clin Cancer Res 1997;3:2431-2438.

115

11. Radford IR. Evidence for a general relationship between the induced level of DNA double-strand breakage and cell-killing ­after x-irradiation of mammalian cells. Int J Radiat Biol 1986;49: 611-620. 12. Kinsella TJ, Dobson PP, Mitchell JB, et al. Enhancement of x-ray induced DNA damage by pretreatment with halogenated pyrimidine analogs. Int J Radiat Oncol Biol Phys 1987;13:733-739. 13. Wang Y, Pantelias GE, Iliakis G. Mechanism of radiosensitization by halogenated pyrimidines: the contribution of excess DNA and chromosome damage in BrdU radiosensitization may be minimal in plateau-phase cells. Int J Radiat Biol 1994;66:133-142. 14. Elkind MM, Sutton H. X-ray damage and recovery in mammalian cells in culture. Nature 1959;184:1293-1295. 15. Little JB. Repair of potentially-lethal radiation damage in mammalian cells: enhancement by conditioned medium from stationary cultures. Int J Radiat Biol 1971;20:87. 16. Gregoire V, Hittelman WN. Nucleoside analogs as radiosensitizing agents. In Cheson BD, Keating MJ, Plunkett W (eds): Nucleoside Analogs in Cancer Therapy. New York: Marcel Dekker, 1997, 315-358. 17. Gregoire V, Hittelman WN, Rosier JF, et al. Chemo-radiotherapy: radiosensitizing nucleoside analogues. Oncol Rep 1999;6:949-957. 18. Lawrence TS, Chang EY, Hahn TM, et al. Radiosensitization of pancreatic cancer cells by 2′,2′difluoro-2′-deoxycytidine. Int J ­Radiat Oncol Biol Phys 1996;34:867-872. 19. Plunkett W, Huang P, Gandhi V. Metabolism and action of fludarabine phosphate. Semin Oncol 1990;17:3-17. 20. Plunkett W, Huang P, Xu YZ, et al. Gemcitabine: metabolism, mechanisms of actions, and self-potentiation. Semin Oncol 1995;22:3-10. 21. Lawrence TX, Eisbruch A, Shewach DS. Gemcitabine-mediated radiosensitization. Semin Oncol 1997;24:24-28. 22. Gregoire V, Hunter N, Milas L, et al. Potentiation of radiationinduced regrowth delay in murine tumors by fludarabine. Cancer Res 1994;54:468-474. 23. Gregoire V, Hunter NR, Brock WA, et al. Improvement in the therapeutic ratio of radiotherapy for a murine sarcoma by indomethacin plus fludarabine. Radiat Res 1996;146:548-553. 24. Milas L, Fujii T, Hunter N, et al. Enhancement of tumor radioresponse in vivo by gemcitabine. Cancer Res 1999;59:107-114. 25. Terasima T, Tolmach LJ. Variations in several responses of HeLa cells to x-irradiation during the cell cycle. Biophys J 1963;3:11-33. 26. Schiff PB, Fant J, Horwitz SB. Promotion of microtubule assembly in vitro by taxol. Nature 1979;277:665-667. 27. Tishler RB, Geard CR, Hall EJ, et al. Taxol sensitizes human astrocytoma cells to radiation. Cancer Res 1992;52:3495-3497. 28. Milas L, Milas MM, Mason KA. Combination of taxanes with ­radiation: preclinical studies. Semin Radiat Oncol 1999;9:12-26. 29. Gregoire V, Van NT, Brock WA, et al. The role of fludarabine­induced apoptosis and cell cycle synchronization in ­ enhanced murine tumor radiation response in vivo. Cancer Res 1994;54: 6201-6209. 30. Milas L, Hunter NR, Mason KA, et al. Role of reoxygenation in induction of enhancement of tumor radioresponse by paclitaxel. Cancer Res 1995;55:3564-3568. 31. Brown JM, Giaccia AJ. The unique physiology of solid tumors: opportunities (and problems) for cancer therapy. Cancer Res 1998;58:1408-1416. 32. Vaupel P, Kallinowski F, Okunieff P. Blood flow, oxygen and nutrient supply, and metabolic microenvironment of human ­tumors: a review. Cancer Res 1989;49:6449-6465. 33. Tannock IF, Rotin D. Acid pH in tumors and its potential for therapeutic exploitation. Cancer Res 1989;49:4373-4384. 34. Brown JM, Biaglow JE, Hall EJ, et al. Sensitizers and protectors to radiation and chemotherapeutic drugs. Cancer Treat Symp 1984;1:85-101. 35. Hermens AF, Barendsen GW. The proliferative status and clonogenic capacity of tumour cells in a transplantable rhabdomyosarcoma of the rat before and after irradiation with 800 rad of x-rays. Cell Tissue Kinet 1978;11:83-100. 36. Stephens TC, Steel GG. Regeneration of tumors after cytotoxic treatment. In Meyn RE, Withers HR (eds): Radiation Biology in Cancer Research. New York: Raven Press, 1980, pp 385–295.

116

PART 1  Principles

37. Trott KR. Cell repopulation and overall treatment time. Int J ­Radiat Oncol Biol Phys 1990;19:1071-1075. 38. Milas L, Yamada S, Hunter N, et al. Changes in TCD50 as a ­measure of clonogen doubling time in irradiated and unirradiated ­tumors. Int J Radiat Oncol Biol Phys 1991;21:1195-1202. 39. Milas L, Nakayama T, Hunter N, et al. Dynamics of tumor cell clonogen repopulation in a murine sarcoma treated with cyclophosphamide. Radiother Oncol 1994;30:247-253. 40. Withers HR, Taylor JMG, Maciejewski B. The hazard of accelerated tumor clonogen repopulation during radiotherapy. Acta ­ Oncol 1988;27:131-146. 41. Taylor JMG, Withers HR, Mendenhall WM. Dose-time considerations of head and neck squamous cell carcinomas treated with irradiation. Radiother Oncol 1990;17:95-102. 42. Mendelsohn J, Fan Z. Epidermal growth factor receptor family and chemosensitization. J Natl Cancer Inst 1997;89:341-343. 43. Sweeney CJ, Miller KD, Sissons SE, et al. The antiangiogenic property of docetaxel is synergistic with a recombinant humanized monoclonal antibody against vascular endothelial growth factor or 2-methoxyestradiol but antagonized by endothelial growth factors. Cancer Res 2001;61:3369-3372. 44. Schimming R, Hunter NR, Mason KA, et al. Apoptosis and inhibition of neoangiogenesis as mechanisms of antitumor action by ­ docetaxel (Taxotere). Mund Kiefer Gesichtschir 1999;3: 210-212. 45. Browder T, Butterfield CE, Kräling BM, et al. Antiangiogenic scheduling of chemotherapy improves efficacy against experimental drug-resistant cancer. Cancer Res 2000;60:1878-1886. 46. Klement G, Baruchel S, Rak J, et al. Continuous low-dose therapy with vinblastine and VEGF receptor-2 antibody induces sustained tumor regression without overt toxicity. J Clin Invest 2000;105:R15-R24. 47. Hanahan D, Bergers G, Bergsland E. Less is more, regularly: metronomic dosing of cytotoxic drugs can target tumor angiogenesis in mice. J Clin Invest 2000;105:1045-1047. 48. Hahnfeldt P, Folkman J, Hlatky L. Minimizing long-term tumor burden: the logic for metronomic chemotherapeutic dosing and its antiangiogenic basis. J Theor Biol 2003;220:545-554. 49. Mason KA, Komaki R, Cox JD, et al. Biology-based combinedmodality radiotherapy: workshop report. Int J Radiat Oncol Biol Phys 2001;50:1079-1089. 50. Gorski DH, Beckett MA, Jaskowiak NT, et al. Blockage of the vascular endothelial growth factor stress response increases the antitumor effects of ionizing radiation. Cancer Res 1999;59: 3374-3378. 51. Haimovitz-Friedman A, Vlodavsky I, Chaudhur A, et al. Autocrine effects of fibroblast growth factor in repair of radiation damage in endothelial cells. Cancer Res 1991;51:2552-2558. 52. Wilder RB, Tucker SL, Ha CS, et al. Dose-response analysis for radiotherapy delivered to patients with intermediate grade and large-cell immunoblastic lymphomas that have completely responded to CHOP-based induction chemotherapy. Int J Radiat Oncol Biol Phys 2001;49:17-22. 53. Arriagada R, Le Chevalier T, Quoix E, et al. ASTRO (American Society for Therapeutic Radiology and Oncology) plenary: effect of chemotherapy on locally advanced non–small cell lung carcinoma: a randomized study of 353 patients. GETCB (Groupe d’Etude et Traitement des Cancers Bronchiques), FNCLCC (Féderation Nationale des Centres de Lutte contre le Cancer) and the CEBI trialists. Int J Radiat Oncol Biol Phys 1991;20: 1183-1190. 54. Pignon JP, Bourhis J, Domenge C, et al. Chemotherapy added to locoregional treatment for head and neck squamous-cell carcinoma: three meta-analyses of updated individual data. Lancet 2000;355:949-955. 55. Tattersall MH, Lorvidhaya V, Vootiprux V, et al. Randomized trial of epirubicin and cisplatin chemotherapy followed by pelvic radiation in locally advanced cervical cancer. J Clin Oncol 1995;13:444-451. 56. Souhami L, Gil RA, Allan SE, et al. A randomized trial of chemotherapy followed by pelvic radiation therapy in stage IIIB ­carcinoma of the cervix. J Clin Oncol 1991;9:970-977.

57. Dillman RO, Seagren SL, Propert KJ, et al. A randomized trial of induction chemotherapy plus high-dose radiation versus radiation alone in stage III non–small-cell lung cancer. N Engl J Med 1990;323:940-945. 58. Le Chevalier T, Arriagada R, Quoix E. Radiotherapy alone versus combined chemotherapy and radiotherapy in nonresectable non–small cell lung cancer. First analysis of a randomized trial in 353 patients. J Natl Cancer Inst 1991;83:417-423. 59. Sause WT, Scott C, Taylor S, et al. Radiation Therapy Oncology Group (RTOG) 88-08 and Eastern Cooperative Oncology Group (ECOG) 4588: preliminary results of a phase III trial in regionally advanced, unresectable non–small cell lung cancer. J Natl Cancer Inst 1995;87:198-205. 60. Pilepich MV, Caplan R, Byhardt RW, et al. Phase III trial of androgen suppression using goserelin in unfavorable-prognosis carcinoma of the prostate treated with definitive radiotherapy: report of Radiation Therapy Oncology Group Protocol 85-31. J Clin Oncol 1997;15:1013-1021. 61. Bolla M, Gonzalez D, Warde P, et al. Improved survival in patients with locally advanced prostate cancer treated with radiotherapy and goserelin. N Engl J Med 1997;337:295-300. 62. Herskovic A, Martz K, Al-Sarraf M, et al. Combined chemotherapy and radiotherapy compared with radiotherapy alone in patients with cancer of the esophagus. N Engl J Med 1992;326:1593-1598. 63. Al-Sarraf M, LeBlanc M, Giri PG, et al. Chemoradiotherapy versus radiotherapy in patients with advanced nasopharyngeal cancer: phase III randomized Intergroup study 0099. J Clin Oncol 1998;16:1310-1317. 64. Morris M, Eifel PJ, Lu J, et al. Pelvic radiation with concurrent chemotherapy compared with pelvic and para-aortic radiation for high-risk cervical cancer. N Engl J Med 1999;340:1137-1143. 65. Thomas GM. Improved treatment for cervical cancer: concurrent chemotherapy and radiotherapy. N Engl J Med 1999;340: 1198-1200. 66. Brizel DM, Albers ME, Fisher SR, et al. Hyperfractionated ­irradiation with or without concurrent chemotherapy for ­locally advanced head and neck cancer. N Engl J Med 1998;338: 1798-1804. 67. Furuse K, Fukuoka M, Kawahara M, et al. Phase III study of concurrent versus sequential thoracic radiotherapy in combination with mitomycin, vindesine, and cisplatin in unresectable stage III non–small-cell lung cancer. J Clin Oncol 1999;17:2692-2699. 68. Curran WJ, Scott C, Langer C, et al. Phase III comparison of ­sequential vs concurrent chemoradiation for PTS with unresected stage III non–small cell lung cancer (NSCLC): initial report of Radiation Therapy Oncology Group (RTOG) 9410 [abstract]. Proc Am Soc Clin Oncol 2000;19:484a. . Murray N, Coy P, Pater JL, et al. Importance of timing for thoracic irradiation in the combined modality treatment of limitedstage small-cell lung cancer. J Clin Oncol 1993;11:336-344. 70. Goto K, Nishiwaki Y, Takada M, et al. Final results of a phase III study of concurrent versus sequential thoracic radiotherapy (TRT) in combination with cisplatin (P) and etoposide (E) for limited-stage small cell lung cancer (LD-SCLC): the Japan Clinical Oncology Group (JCOG) study [abstract]. Proc Am Soc Clin Oncol 1999;18:468a. 71. Gastrointestinal Tumor Study Group. Prolongation of the diseasefree interval in surgically treated rectal carcinoma. N Eng J Med 1985;312:1465-1472. 72. Krook JE, Moertel CG, Gunderson LL, et al. Effective surgical adjuvant therapy for high-risk rectal carcinoma. N Engl J Med 1991;324:709-715. 73. O’Connell MJ, Martenson JA, Wieand HS, et al. Improving ­adjuvant therapy for rectal cancer by combining protractedinfusion fluorouracil with radiation therapy after curative surgery. N Engl J Med 1994;331:502-507. 74. Wolmark N, Wieand HS, Hyams DM, et al. Randomized trial of postoperative adjuvant chemotherapy with or without radiotherapy for carcinoma of the rectum: National Surgical Adjuvant Breast and Bowel Project protocol R-02. J Natl Cancer Inst 2000;92:388-396.

4  Principles of Combining Radiation Therapy and Chemotherapy 75. Kelsen DP, Ginsberg R, Pajak TF, et al. Chemotherapy followed by surgery compared with surgery alone for localized esophageal cancer. N Engl J Med 1998;339:1979-1984. 76. Walsh TN, Noonan N, Hollywood D, et al. A comparison of multimodal therapy and surgery for esophageal adenocarcinoma. N Engl J Med 1996;335:462-467. 77. Bosset J-F., Gignou M, Triboulet J-P, et al. Chemoradiotherapy followed by surgery compared with surgery alone in squamouscell cancer of the esophagus. N Engl J Med 1997;337:161-167. 78. Cooper JS, Pajak TF, Forastiere AA, et al. Postoperative concurrent radiotherapy and chemotherapy for high-risk squamous-cell carcinoma of the head and neck. N Engl J Med 2004;350:1937-1944. 79. Turrisi A, Kim K, Sause W, et al. Observations after 5 year followup of intergroup trial 0096: 4 cycles of cisplatin (P) etoposide (E) and concurrent 45 Gy thoracic radiotherapy (TRT) given in daily (QD) or twice-daily fractions followed by 25 Gy PCI. Survival differences and patterns of failure [abstract]. Proc Am Soc Clin Oncol 1998;17:457a. 80. Macdonald JS, Smalley S, Benedetti J, et al. Postoperative combined radiation and chemotherapy improves disease-free survival (DFS) and overall survival (OS) in resected adenocarcinoma of the stomach and GE junction. Results of intergroup study INT-0116 (SWOG 9008) [abstract]. Proc Am Soc Clin Oncol 2000;19:1a. 81. Overgaard M, Hansen PS, Overgaard J, et al. Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. N Engl J Med 1997;337: 949-955. 82. Ragaz J, Jackson SM, Le N, et al. Adjuvant radiotherapy and chemotherapy in node positive premenopausal women with breast cancer. N Engl J Med 1997;337:956-962. 83. Bartelink H, Roelofsen F, Eschwege F, et al. Concomitant radiotherapy and chemotherapy is superior to radiotherapy alone in the treatment of locally advanced anal cancer: results of a phase

117

III randomized trial of the European Organization for Research and Treatment of Cancer Radiotherapy Gastrointestinal Cooperative Groups. J Clin Oncol 1997;15:2040-2049. 84. Flam MS, Pajak TF, Haller D, et al. The role of mitomycin in combination with 5-fluorouracil and radiotherapy in the definitive nonsurgical treatment of epidermoid carcinoma of the anal canal: 5-year updated results of a phase III randomized study (RTOG 8704/ECOG 1289) [abstract]. Presented at the Japan/US Cancer Therapy Symposium, Hiroshima, Japan, April 1999. 85. Spaulding MB, Fischer SG, Wolf GT. Tumor response, toxicity, and survival after neoadjuvant organ-preserving chemotherapy for advanced laryngeal carcinoma. The Department of Veterans Affairs Cooperative Laryngeal Cancer Study Group. J Clin Oncol 1994;12:1592-1599. 86. Forastiere AA, Goepfert H, Maor M, et al. Concurrent chemotherapy and radiotherapy for organ preservation in advanced laryngeal cancer. N Engl J Med 2003;349:2091-2098. 87. Kaufman DS, Shipley WU, Griffin PP, et al. Selective bladder preservation by combination treatment of invasive bladder cancer. N Engl J Med 1993;329:1377-1382. 88. Tester W, Caplan R, Heaney J, et al. Neoadjuvant combined modality program with selective organ preservation for invasive bladder cancer: results of Radiation Therapy Oncology Group phase II trial 8802. J Clin Oncol 1996;14:119-126. 89. Fossella FV, Zinner RG, Komaki R, et al. Gemcitabine (G) with concurrent chest radiation (XRT) followed by consolidation chemotherapy with gemcitabine plus cisplatin (CDDP): a phase I trial for patients with stage III non–small cell lung cancer (NSCLC) [abstract]. J Clin Oncol 2001;20:312a. 90. Komaki R, Lee JS, Milas L, et al. Effects of amifostine on acute toxicity from concurrent chemotherapy and radiotherapy ­ for ­inoperable non–small cell lung cancer: report of a randomized comparative trial. Int J Radiat Oncol Biol Phys 2004;58:1369-1377.

5

Principles of Combining Radiation Therapy and Molecular Therapeutics K. Kian Ang, Michael S. O’Reilly, Raymond E. Meyn Jr., and Luka Milas EPIDERMAL GROWTH FACTOR RECEPTOR AS A TARGET IN CANCER THERAPY EGFR Structure and Function EGFR Signaling in Epithelial Neoplasms EGFR and Radiation Therapy Challenges for Anti-EGFR Therapy ANGIOGENESIS AS A TARGET IN CANCER THERAPY Antivascular versus Antiangiogenic Therapy Radiation Therapy Effects on the Vasculature Angiogenesis Inhibitors COMBINING ANTIANGIOGENIC AGENTS WITH RADIATION THERAPY OR RADIATION AND CHEMOTHERAPY Preclinical Findings Clinical Findings CYCLOOXYGENASES AS TARGETS IN CANCER THERAPY Expression and Function COX-2 Inhibitors

Numerous clinical trials have demonstrated that adding chemotherapy to radiation therapy, particularly when the two are given concurrently, can improve locoregional control and survival rates in several types of cancer. Rational modifications of radiation fractionation schedules can also improve outcome, mainly for patients with head and neck squamous cell carcinoma (HNSCC). These findings triggered a trend in which radiation-chemotherapy or radiation in altered fractionation schedules is used instead of radical surgery for the treatment of many types of cancer, as described elsewhere in this textbook. Although concurrent radiation-chemotherapy ­improves tumor control, the management of its acute toxic effects can be complex and labor-intensive, and its long-term morbidity can be severe. These observations, combined with advances in the knowledge of molecular radiation ­biology, provided the foundation for developing rational strategies for combining radiation therapy with molecular therapeutics. Research on the epidermal growth factor receptor (EGFR), a member of the ERBB receptor tyrosine kinase family, exemplifies this kind of coordinated research effort. Collectively, laboratory and clinical investigations have revealed mechanisms by which specific receptor tyrosine kinase pathways affect a cell’s response to radiation, vali­ dated the concept of selectively modulating tumor ­response to radiation therapy by targeting a perturbed signaling pathway, and established a novel combined-therapy strategy for the treatment of locally advanced HNSCC. This pivotal

118

CYCLINS AND CYCLIN-DEPENDENT KINASES AS TARGETS IN CANCER THERAPY Expression and Function CDK Inhibitors in Cancer Therapy RAS SIGNALING PATHWAY AS A TARGET IN CANCER THERAPY GENE THERAPY Delivering Radiosensitizing Drugs Gene Replacement Oncolytic Viruses Silencing Gene Expression IMMUNOTHERAPY Mechanisms Molecular Immunotherapy SUMMARY

success story inspired further investigations to explore potential uses of molecular therapeutics in the context of radiation therapy for the treatment of other types of cancer. This chapter summarizes findings for bench-to-bedside work on the targeting of EGFR, angiogenesis, cyclooxygenase (COX) enzymes, cyclins and cyclin-dependent ­kinases (CDKs), and RAS signaling pathways in the context of current and future investigations in radiation oncology. Relevant aspects of gene therapy and immunotherapy are presented in the context of their use in combination with radiation therapy.

Epidermal Growth Factor Receptor as a Target in Cancer Therapy EGFR Structure and Function The specific receptor for EGF was identified in 19751; later, its tyrosine-specific phosphorylation was found to activate intracellular signal transduction.2 This particular receptor (also called ERBB1 or HER1) was subsequently characterized as a 170-kd transmembrane receptor tyrosine kinase belonging to the ERBB family, members of which are essential for normal development. The cDNA sequence (3633 bp),3 chromosome localization (7p12-p22),4 genomic structure,5 and amino acid sequence of ERBB1 were subsequently identified.

5  Principles of Combining Radiation Therapy and Molecular Therapeutics

Structurally, the EGFR is composed of four extracellular domains: the ligand-binding regions (domains I and III); a hydrophobic transmembrane domain; a juxtamembrane domain, an intracellular protein tyrosine kinase (PTK) domain containing an ATP-binding pocket; and a regulatory carboxyl-terminal domain. In the absence of ligands, EGFR exists in monomer form; binding of a ligand to the extracellular domains I and III alters the spatial configuration of these domains, creating an extended and stabilized conformation that promotes homodimerization and heterodimerization6 and activates signal transduction. Activation of the EGFR initiates several layers of signaling amplification, diversity, and control of biologic responses. Four major cytoplasmic signaling routes that coregulate several biologic processes have been relatively well characterized. One, the RAS-RAF-MAPK pathway, leads to the ­activation of the mitogen-activated protein kinases (MAPKs) ERK1 and ERK2, and that activation regulates the transcription of molecules mediating cell proliferation, survival, and transformation.7 Phosphatidylinositol-3-kinase (PI3K) and its downstream protein, the serine/threonine kinase AKT (also called PKB or AKT1), govern cell growth, proliferation, survival, and motility. The JAK/STAT pathway translocates signal transducers and activators of transcription (STAT) molecules to the nucleus to directly induce the transcription of genes that mediate cell division, survival (or death), motility, invasion, adhesion, and DNA repair.8-11 The other pathway involves phospholipase Cγ, diacylglycerol, calcium/­ calmodulin, and protein kinase C interacting with PI3K to regulate cell cycle progression and cell ­motility.12,13 Nuclear EGFR pathways have been identified. Nuclear EGFR seems to originate from its cell surface counterparts and is present in an uncleaved, full-length, probably phosphorylated form.14,15 In the nucleus, EGFR seems to act as a transcription cofactor, upregulating the synthesis of promitogenic proteins.15 For example, it associates with STAT3 to induce inducible nitric oxide synthase (iNOS) or with E2F to induce MYBL2 (formerly designated BMYB).16 Nuclear EGFR also interacts with DNA-PK, a DNA-dependent kinase, to promote nonhomologous end-joining DNA repair, the predominant process by which radiation-induced DNA double-strand breaks are repaired.14,17 This function is discussed further in the context of the modulation of cellular radiation sensitivity later in this section. EGFR-mediated signaling is regulated through several mechanisms. The two main processes of signal attenuation are dephosphorylation of key residues by phosphatases and removal of the receptor by endocytosis.18-21 EGFR has been shown to localize in caveolae, lipid-rich membrane microdomains that are important for EGFR turnover.22

EGFR Signaling in Epithelial Neoplasms EGFR is highly expressed in most epithelial malignancies. In 80% to 90% of HNSCCs, for example, more EGFR is ex­ pressed than in the mucosa of individuals without cancer.23 The main mechanism of constitutive EGFR upregulation in HNSCC seems to be transcriptional activation, perhaps resulting from the autocrine production of transforming growth factor α (TGF-α).24,25 Other mechanisms include activating EGFR mutations,26 deletion of exons 2 to 7 in the extracellular domain,27 and polymorphisms in intron 128 that mediate cross-talk with other ERBB receptors in ways that lead to gain of function.29

119

ERBB receptors have been consistently detected in the nuclei of cancer cells.15 Nuclear EGFR expression has been correlated with poor outcomes for patients with HNSCC.30 Transcriptional targets of nuclear EGFR seem to be involved in tumor progression. An example is cyclin D1, which promotes the progression of cells through the G1/S phase checkpoint.31-33 Increased EGFR expression is often associated with more aggressive tumors, poor prognosis, and resistance to treatment with cytotoxic agents, including radiation therapy.

EGFR and Radiation Therapy Preclinical Findings Radiation-Induced Activation of EGFR and Tumor Clonogen Repopulation. One mechanism by which tumors can evade the cytotoxic effects of radiation therapy is through accelerated repopulation of tumor cells during a course of fractionated radiation therapy.34 Schmidt-Ullrich and colleagues found that cancer cells surviving fractionated irradiation could acquire a phenotype characterized by upregulation of EGFR and its principal ligand TGF-α35,36 and that radiation therapy in the therapeutic dose range could increase EGFR tyrosine phosphorylation.25,37,38 This EGFR response has been linked experimentally to several critical components of mitogenic or proliferative signaling pathways39 and has been interpreted as representing an adaptive mechanism favoring accelerated repopulation. EGFR and Cellular Radiation Sensitivity. The first clue that EGFR expression might affect intrinsic cellular radiation sensitivity in vivo emerged from a study of murine models by Akimoto and colleagues,40 who showed that exposure to a single radiation dose induced EGFR autophosphorylation and downstream signaling only in tumors with high EGFR expression and that this phenomenon was associated with the relative resistance of tumors to radiation (i.e., a higher median tumor control dose [TCD50]) (Fig. 5-1A). Because tumor clonogen repopulation usually has little effect on tumor response to single-dose irradiation in vivo,33 these results were interpreted as indicating that EGFR expression contributed to the intrinsic sensitivity of cells to irradiation. This interpretation is supported by the findings of a complementary correlative biomarker study involving human HNSCC specimens, as discussed later under “Clinical Findings.”41 A follow-up study by the same group revealed evidence of a causal relationship between cellular EGFR expression and relative radiation resistance. The findings from that study showed that transfection of human EGFR expression vectors into a low-EGFR-expressing murine ovarian carcinoma cell line, OCA-I, resulted in an EGFR level– dependent increase in resistance to ionizing radiation (see Fig. 5-1B).42 That study also demonstrated that exposing EGFR transfectants to cetuximab (C225, Erbitux), an antibody to EGFR, reduced the levels of the EGFR receptors and their downstream substrates (AKT and MAPK) and reversed cellular radiation resistance. Findings from a subsequent experiment elucidated the mechanism by which EGFR affects radiation sensitivity.14 That thorough study showed that unlike cytoplasmic signaling induced by the natural ligand EGF, ionizing radiation triggered EGFR translocation into the nucleus. That process was accompanied by a nuclear influx of the proteins

120

PART 1  Principles

-I

Ca

O

II G 29 35 I -V C-IV a-S K -4 a- Ca- CaCa CC Ca C C C A S M H M S M M

90

TCD50 (Gy)

80 70 60 50 40 0

5

A

10

15

20

25

EGFR densiometric value

Surviving fraction

100

Clinical Findings

10

1 OCA-I Neo clone-1 EGFR clone-5 0 0

B

XRCC6 (formerly called KU70) or XRCC5 ­ (formerly called KU80) and protein phosphatase 1 and consequently by increases in the activity of DNA-PK in the nucleus and the formation of DNA end-binding protein complexes containing DNA-PK, which are essential for the repair of DNA double-strand breaks. Blockade of EGFR by cetuximab abolished the import of EGFR into the nucleus, suppressed the radiation-induced activation of DNA-PK, inhibited DNA repair, and enhanced cellular radiation ­sensitivity. EGFR Inhibitors Enhance Tumor Response to Radiation Therapy. In the mid-1980s, Mendelsohn and colleagues were the first to show that monoclonal antibodies directed at the EGF-binding site on the EGFR had an antiproliferative effect.43,44 Over the next 2 decades, several types of EGFR inhibitors were designed and tested for a variety of effects in preclinical model systems, including their capacity to augment chemotherapy response. The potential for EGFR inhibitors to modulate ­radiation response came into focus in the late 1990s. Initial studies of the effect of cetuximab45-48 showed enhanced tumor response to single-dose and subsequently to fractionated radiation in tumor cell lines and in xenograft models in which regrowth delay45,47 and tumor control were used as end points.49 The magnitude of the radiation enhancement ­ varied among EGFR antagonists50 such as the low-­molecular-weight ATP-competitive inhibitors of the receptor’s tyrosine kinase activity.51-54

2

4

6

Radiation dose

FIGURE 5-1.  A, Western blots of nine murine carcinomas (top) show substantial variation in the expression of epidermal growth factor receptor (EGFR) protein. The relationship between EGFR expression and radiocurability (bottom) is measured by the radiation dose yielding 50% tumor cure (TCD50). Higher EGFR expression is associated with higher TCD50 values (blue line), suggesting increased radioresistance. B, Fractions of surviving cells are plotted against single radiation doses of 2, 4, or 6 Gy. Compared with the parental ovarian carcinoma cell line (OCA-1; green line), expression of EGFR after transfection of a human EGFR cDNA fragment (EGFR clone-5; blue line) increases the surviving fraction, whereas transfection of a control vector (Neo clone-1; red line) does not affect survival.   (From Liang K, Ang KK, Milas L, et al. The epidermal growth factor receptor mediates radioresistance. Int J Radiat Oncol Biol Phys 2003;57:246-254.)

EGFR Overexpression in Head and Neck Squamous Cell Carcinoma. A post hoc correlative biomarker analysis of specimens from patients enrolled in a phase III trial of radiation therapy fractionation for locally advanced HNSCC showed that high-level EGFR expression in tumor specimens (demonstrated by immunohistochemical staining [Fig. 5-2A and B]) was a strong independent predictor of locoregional control and overall survival after radiation therapy (see Fig. 5-2C and D).41 A concurrent pilot study demonstrated the safety of combining cetuximab with radiation therapy for locally advanced HNSCC in a regimen consisting of a 400 mg/m2 loading dose of cetuximab given over 120 minutes 1 week before radiation therapy, followed by seven weekly 250 mg/m2 doses commencing with the radiation therapy.55 The findings from this pilot study, in combination with the biomarker analysis and preclinical findings, were the impetus for launching a multinational phase III study to test the efficacy of the combination of cetuximab with radiation therapy relative to that of irradiation alone for patients with locally advanced HNSCC.56 The findings from this crucial trial demonstrated that the addition of only eight doses of cetuximab, starting 1 week before radiation therapy and continuing during the radiation therapy, significantly improved the locoregional control rate (3-year rate of 47% versus 34%, P = .005), median survival time (49 versus 29 months, P = .03), and 3-year survival rate (55% versus 45%, P = .05) (see Fig. 5-2E and F) without increasing the incidence or severity of radiationinduced side effects such as mucositis or dysphagia. The U.S. Food and Drug Administration (FDA) approved cetuximab for use in combination with radiation therapy as front-line treatment for patients with locally advanced HNSCC in February 2006.

5  Principles of Combining Radiation Therapy and Molecular Therapeutics

A

B 100

100

Local-regional relapse P = 0.0031

75

Overall survival P = 0.0006

75 >Median (n = 77)

% alive

% failed

121

50 ≤Median (n = 78) 25

≤Median (n = 78)

50 25

>Median (n = 77) 0

0 1

0

3

4

5

2

3

4

5

Years from randomization 100

100 80 Radiotherapy plus cetuximab

60 40 20

Radiotherapy

80 Radiotherapy plus cetuximab

60 40

Radiotherapy

20 0

0 0

E

1

0

D

Years from randomization

Overall survival (%)

Locoregional control (%)

C

2

10

20

30

40

50

60

70

Months

0

F

10

20

30

40

50

60

70

Months

FIGURE 5-2.  A and B, Immunohistochemical stains used for visualizing epidermal growth factor receptor (EGFR) in human head and neck squamous cell carcinoma specimens. (A, below median; B, above median). C and D, The relationship is shown for EGFR expression and locoregional relapse (C) or overall survival (D) rates for patients with locally advanced head and neck squamous cell carcinoma (HNSCC) treated with conventionally fractionated radiation therapy. E and F, Addition of cetuximab to radiation therapy results in significant improvements in locoregional control (E) and overall survival (F) rates for patients with locally advanced HNSCC. (A – D, From Ang KK, Berkey BA, Tu X, et al. Impact of epidermal growth factor receptor expression on survival and pattern of relapse in patients with advanced head and neck carcinoma. Cancer Res 2002;62:7350-7356; E and F, from Bonner JA, Harari PM, Giralt J, et al. Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. N Engl J Med 2006;354:567-578.)

Challenges for Anti-EGFR Therapy The findings from coordinated laboratory and clinical studies such as those described here generated considerable excitement; in addition to producing a new, less toxic option for therapy, these findings established the proof of principle that modulating a cellular signaling pathway that is expressed differently in tumors and in normal tissues can lead to selective sensitization of tumors to radiation therapy and hence improve the therapeutic index. Selective enhancement of tumor response has not been accomplished by combining radiation with traditional chemotherapeutic agents such as cisplatin, taxanes, or others.

The nonsurgical standard of care for locally advanced HNSCC was shifting from radiation therapy only to irradiation with concurrent chemotherapy (e.g., cisplatin) while the findings from the phase III cetuximab trial were maturing. Since then, additional trials have been ­undertaken to assess the feasibility of combining cetuximab with radiation therapy given with chemotherapy. In a phase II study led by the Memorial Sloan-Kettering Cancer Center, the combination of radiation therapy, cetuximab, and cisplatin was found to result in an overall survival rate of 76%, a progression-free survival rate of 56%, and a locoregional control rate of 71% at 3 years. Two early deaths were

122

PART 1  Principles

­attributed to pneumonia and unknown causes.57 This regimen is now being evaluated in a phase III trial led by the Radiation Therapy Oncology Group. Another challenge is that cetuximab has increased locoregional control and survival rates of patients with locally advanced HNSCC by only 10% to 15%, and more than 50% of patients given the combination of radiation therapy and cetuximab still developed locoregional relapse. Assays that can identify resistant tumors are urgently needed so that this rather expensive therapy can be given to those patients likely to benefit from it. Understanding the biologic basis of tumor resistance to EGFR antagonists is essential to extend this type of treatment to more patients and to develop alternative molecular-level treatment strategies for HNSCC. The search for predictive biomarkers for tumor response to EGFR antagonists has been rather disappointing. Contrary to expectations, the magnitude of tumor EGFR expression before treatment did not predict tumor response to EGFR antagonists in two clinical trials.58,59 Moreover, findings from colorectal cancer trials suggest that some patients whose tumors did not express EGFR (indicated by immunohistochemical staining) still benefited from cetuximab therapy.60,61 Notably, the assay methods used for correlative studies vary widely among investigators. In a standard immunohistochemical assay, for example, a cell is considered positive for EGFR if more than 30,000 receptors are present.62 Unfortunately, the number and perhaps the density of receptors required to confer a given biologic effect are poorly understood. This deficiency may account for some of the discrepancies among study findings. Several strategies for overcoming resistance to radiation sensitization by EGFR antagonists are being investigated. Cetuximab or EGFR tyrosine kinase inhibitors given as single-agent therapy in current dose regimens may not suppress EGFR-mediated signaling sufficiently to affect tumor survival. Other dose regimens, other antibodies and tyrosine kinase inhibitors, antisense nucleotides,63 and various combinations of these agents are being studied. Findings from a preclinical study, for example, showed that administering three additional doses of cetuximab after concurrent irradiation and cetuximab improved local tumor control compared with concurrent irradiation and cetuximab alone.64 Other antibodies being studied include hR3, which has a longer half-life and larger area under the curve than cetuximab, and ABX-EGF, which has greater affinity for EGFR. Another possibility being explored is that mutations in EGFR may result in aberrant signaling and diminished response to EGFR antagonists or ionizing radiation. For example, a form of EGFR with a truncated extracellular domain (EGFRvIII)27 detected in several tumors was different from wild-type EGFR (EGFRwt) in its preferential activation of signaling pathways.65,66 In one series, EGFRvIII expression was found in 14 (42%) of 33 tumors, and all 14 also expressed EGFRwt.67 The same study showed that transfecting HNSCC cell lines with EGFRvIII can ­reduce apoptosis in response to cisplatin or reduce growth inhibition by cetuximab, observations that form the basis for investigating the efficacy of an EGFRvIII-specific monoclonal antibody. Another possible explanation for resistance to EGFR inhibitors is that constitutive activation of signaling ­pathways

downstream of EGFR by mutation or upregulation of other ERBB family receptors or other receptor tyrosine kinase classes can promote cell survival (reviewed by ­Kalyankrishna and Grandis68). For example, high levels of activated AKT1 can occur downstream of EGFR inhibition through upstream activation of SRC, RAS, or mutated PTEN69; ­ amplification of the catalytic subunit of PI3K70; or loss of the phosphatase and tensin homologue tumor suppressor protein.71 Overexpression of six major ERBB family ligands can activate ERBB receptors such as ERBB2 and ERBB372 and the insulin-like growth factor receptor 1 (IGF1R),73 resulting in resistance to EGFR inhibition. Upregulation of IGF1R, for instance, was found to result in sustained signaling through the PI3K pathway, leading to potent antiapoptotic and proinvasion effects and resistance to an EGFR tyrosine kinase inhibitor in a glioblastoma model.74 In summary, discovery of the contribution of EGFR signaling in governing several key cellular regulatory processes, its perturbation in epithelial neoplasms, and its impact on tumor response to some therapeutic modalities led to the conception and completion of a clinical research program that produced a new front-line therapy option for locally advanced HNSCC. Although this pivotal phase III trial validated the concept that tumor cells could be selectively sensitized to radiation therapy through modulation of a perturbed signaling pathway, the clinical benefit resulting from this strategy has been rather modest, and many questions remained to be addressed. For example, the lack of correlation between the magnitude of EGFR expression and tumor response to EGFR antagonists is counterintuitive. We also need ways of identifying predictive biomarkers for response to EGFR inhibitors for clinical use and clarification of the mechanisms underlying the lack of cetuximab-mediated sensitization of most HNSCC tumors to radiation so that approaches to circumvent this obstacle can be formulated. This successful experience in translational research has led to many investigations of various combinations of radiation therapy with other novel agents and therapeutic strategies for HNSCC and other types of neoplasms. Although a comprehensive review of all such strategies is beyond the scope of this chapter, a few examples of other molecular-therapeutic strategies, including those based on tumor angiogenesis, are presented here.

Angiogenesis as a Target in Cancer Therapy Folkman first proposed that the growth and spread of malignancy depends on angiogenesis more than 3 decades ago.75 Since then, much work has been done in the field of angiogenesis, and it has become clear that the induction of adequate vasculature is critical for the growth of tumors, for the development of metastasis, and for several physiologic and pathophysiologic processes.76 Under physiologic con­ ditions that require neovascularization, angiogenesis is a tightly regulated process77 that begins with degradation of the basement membrane surrounding capillaries followed by stromal invasion by endothelial cells. Proliferation and migration of endothelial cells eventually lead to the formation of a network of new blood vessels as the

5  Principles of Combining Radiation Therapy and Molecular Therapeutics

invading endothelial cells organize into three-dimensional structures.78 During tumorigenesis, however, the process is sustained and requires the continued production of stimulators by tumor cells and stromal cells in excess of inhibitors. Several lines of evidence confirm the dependence of tumor growth on angiogenesis.76 A tumor mass that is smaller than about 0.4 mm in diameter can receive oxygen and nutrients by diffusion, but any further increase in tumor mass requires neovascularization.79 In malignant angiogenesis, new vessels are anomalous,80 characterized by a relative lack of pericytes and other support cells, impaired flow,81 the presence of plasma proteins that function as a neostroma,82 and increased vascular permeability. In addition to forming the new vessels that provide nutrients and oxygen and remove catabolites, ­ proliferating endothelial cells found in and around tumors produce several growth factors that can promote tumor cell growth, invasion, and survival.83-85 Angiogenesis therefore provides a perfusion effect and a paracrine effect for growing tumors,76 and tumor cells and endothelial cells can drive each other and perpetuate and amplify the malignant phenotype. These observations led Folkman to propose a two-compartment model of malignancy in which one compartment consists of endothelial cells and the other of tumor cells.86 Both compartments offer potential targets for anticancer strategies, and prudence dictates that the cancer cells and the angiogenic endothelium be targeted in such strategies.

Antivascular versus Antiangiogenic Therapy Vascular-targeted therapies consist of two broad types. They can prevent the formation of new blood vessels or revascularization, or they can disrupt the existing vasculature of a tumor. Antiangiogenic therapy is directed against the formation of neovasculature and does not significantly affect the resting vasculature. Antivascular therapy targets existing vasculature and can disrupt normal vessels. In practice, many agents, such as those that target vascular endothelial growth factor (VEGF), have both antiangiogenic and antivascular effects. One strategy to selectively target tumor vasculature is to use toxins conjugated to antibodies directed against antigens that are selectively expressed by tumor endothelial cells.87 Another approach involves using small-molecule agents that destabilize microtubulin, such as combretastatin A4 and ZD6126.88 Angiogenic endothelial cells can also be targeted by specific peptide sequences that selectively bind to proliferating endothelial cells but not to quiescent cells.89 Whether organ-specific peptides can home to sites of angiogenesis in selected tissues remains to be determined, but targeting the tumor vasculature is an attractive approach. Several studies have demonstrated enhanced antitumor effects from combining vascular-targeting agents with radiation therapy. In a study of KHT sarcoma xenografts, for example, ZD6126 administered 24 hours before, immediately after, or up to 4 hours after radiation therapy resulted in enhanced tumor cell killing compared with that achieved with irradiation alone, and that enhancement was optimal when the vascular-targeting agent was given within 1 hour of the radiation.90 ZD6126 was also effective when administered after fractionated radiation

123

therapy for A431 lung cancer xenografts,91 particularly when given with the anti-EGFR agent ZD1839. However, for U87 human glioblastoma xenografts in mice, ZD6126 given 1 hour before irradiation was associated with an acute increase in tumor-tissue hypoxia and faster tumor growth rates compared with irradiation alone. Giving this drug after irradiation did not affect hypoxia but did not enhance antitumor activity compared with radiation therapy alone.92 These findings seem to demonstrate the potential for antivascular agents to enhance tumor response to radiation therapy, but further studies are needed to optimize this approach. The acute disruption of tumor vasculature must be timed precisely to minimize the potential impairment of radiation effects by any ensuing hypoxia. Optimizing the timing and delivery of each modality may allow eradication of viable peripheral tumor tissue that may be resistant to the effects of vascular targeting agents. However, an antiangiogenic agent may also be required to maximize the endothelial apoptosis that can occur within hours of radiation therapy and to prevent the revascularization that can occur after radiation therapy.93

Radiation Effects on the Vasculature Radiation therapy can have systemic and local effects on angiogenesis, possibly through increasing the expression of proangiogenic factors such as VEGF.94 In the late 1940s, Kaplan and Murphy95 observed that some patients with epidermoid carcinoma of the lip developed widely metastatic disease after radiation therapy. Because that type of cancer rarely metastasizes in this way, they speculated that the irradiation might have been a contributing factor. To test this idea, they implanted mice subcutaneously with a mammary tumor cell line that rarely metastasizes to the lung, irradiated the mice, and found that more than 44% of the irradiated animals developed lung metastases compared with only 10% of nonirradiated controls. More than 20 years later, a report from Suit and colleagues96 observed that animals that developed disease recurrence at the primary tumor site after radiation therapy were more likely to develop metastatic disease. Later work with nude mice showed that the formation of recurrent tumors after radiation therapy was preceded by angiogenesis, most of which occurred within 20 days of the irradiation.97 Investigations of the mechanisms underlying these obser­ vations indicated that radiation therapy to a primary tumor can decrease the mobilization of angiogenesis inhibitors such as angiostatin; administration of an angiogenesis inhibitor can slow the formation of metastases in the lungs of mice after irradiation of primary Lewis lung carcinomas.98 These findings suggest that adding antiangiogenic therapy to radiation treatment may help to restore the balance between endogenous angiogenic and antiangiogenic factors so as to inhibit locoregional and metastatic tumor growth. Under some circumstances, radiation therapy can also impair angiogenesis. In studies of tumor growth delays caused by the irradiation of stroma, a phenomenon known as the tumor bed effect, Milas and colleagues99 confirmed that tumors grew more slowly in mice that received radiation to the tumor bed before tumor cell implantation; however, they later found an increased incidence of ­metastasis among those mice and concluded that a tumor’s angiogenic

124

PART 1  Principles

potential was an important factor in the development of the tumor bed effect.100 Evidence that suppression of angiogenesis by radiation therapy is not restricted to the primary tumor site comes from another study101 showing that irradiation of primary tumors in rodents was associated with inhibition of angiogenesis at a distant site (the cranial window). Moreover, plasma levels of the angiogenic agent endostatin in the irradiated mice were twice those found in the nonirradiated mice. Further study is needed to better define the dynamic interactions between the vasculature and irradiation.

Angiogenesis Inhibitors Several antiangiogenic agents have been described,76,102,103 and some have been approved or are awaiting approval for clinical use in the treatment of cancer and other angiogenesis-dependent diseases. However, given the complexity of the angiogenic process in various disease states, a better understanding of the process of angiogenesis and the interactions between endothelial cells and their microenvironment is necessary before antiangiogenic agents can reach their full potential for the treatment of such diseases in humans. Early experience with antiangiogenic agents in the treatment of human cancer, particularly as monotherapy, revealed some important limitations. An example is the delayed onset of activity of some antiangiogenic agents,104 which can allow significant tumor progression to occur before they become effective. Murine tumors tend to proliferate several times faster than tumors in humans, and this delay in onset of activity may translate to several months in patients. In a veterinary study of the use of the antiangiogenic thrombospondin peptide mimetics ABT-526 and ABT-510 to treat naturally occurring canine malignancies, objective tumor responses were observed only after at least 60 days of therapy.105 Therapy with the antiangiogenic cytokine interferon-α for life-­threatening pediatric hemangiomas106 or giant cell tumors107 ­required several months of continued therapy before significant clinical responses were observed. The delays in the onset of activity may make the use of these agents impractical for patients with locally advanced or metastatic disease. Perhaps a more significant limitation of antiangiogenic monotherapy is the inability of these agents to completely eradicate disease even after prolonged administration at high doses. For most angiogenesis inhibitors, tumor growth generally resumes within a short time after cessation of therapy.104 Although prolonged cyclical therapy with endostatin could induce sustained dormancy in a wide variety of preclinical tumor models that persisted after therapy had been stopped, residual microscopic disease was still found in all of the treated mice.108 Immunohistochemical staining of these residual microscopic tumors revealed little or no evidence of angiogenesis, but it did show a high proliferative index of the tumor cells that was offset by a high apoptotic index. These findings suggest that it may not be feasible or practical to produce the amounts of antiangiogenic agents that would be required for widespread prolonged use. As an alternative approach, combinations of angiogenesis inhibitors that target different angiogenic pathways may be more effective than monotherapy; a related strategy that

may be especially beneficial is to combine direct inhibitors of angiogenesis with agents that target stimulators. Experience with angiogenesis, like that with other molecularly targeted therapies, suggest that strategies that target each of the interdependent steps involved in the angiogenic cascade, perhaps in combination with other treatment modalities, will be necessary to effectively inhibit pathogenic angiogenesis. Strategies that target growth factor receptor tyrosine kinase signaling pathways involved in tumor angiogenesis, growth, and survival offer great potential for antitumor therapy. The most widely studied pathways are those involving VEGF and its receptors. VEGF induces vascular permeability and plays a pivotal role in vasculogenesis and angiogenesis.109,110 Although the effects of VEGFR signaling were initially thought to be specific to the vasculature, VEGF also participates in several other processes.112-115 In pregnant women, the presence of endogenous regulators of VEGF in the circulation has been linked with the risk of hypertension, proteinuria, and preeclampsia, effects thought to result from endothelial dysfunction.111 Hypertension and proteinuria have also been reported in patients with cancer who were treated with VEGF-signaling inhibitors.116,117 Other findings suggest that VEGFR1 expression by colon cancer cells contributes to the motility and invasiveness of those cells but has little direct effect on their proliferation.118 Several approaches to block VEGF activity are being evaluated, including anti-VEGF or VEGFR antibodies and VEGFR tyrosine kinase inhibitors.110 Bevacizumab (Avastin), a humanized antibody against VEGF, has shown great promise for extending survival among patients with metastatic colorectal or lung cancer when used in combination with chemotherapy.119,120 Nevertheless, there is ample room for improving the efficacy of this and other anti-VEGF agents beyond merely extending survival. Side effects are also an issue; an early clinical trial of multimodality therapy that included SU5416 (semaxanib), another VEGF inhibitor, linked its use with thromboembolic and hemorrhagic complications.121 A phase III clinical trial of bevacizumab with chemotherapy for women with previously treated breast cancer showed that the combination of bevacizumab and capecitabine increased response rates but did not improve progression-free or overall survival.122 The addition of PTK787 (vatalanib), a small-molecule inhibitor of VEGFR tyrosine kinase activation, to the chemotherapy regimen FOLFOX4 (5-fluorouracil, leucovorin, and oxaliplatin) did not improve overall survival among previously treated patients with metastatic colorectal cancer compared with chemotherapy only,123 although a ­ progression-free survival benefit might have been observed in a subgroup of patients with elevated lactate dehydrogenase levels.124 Therapy with these agents may have to be optimized for each patient. A further complication in anti-VEGF/VEGFR therapy is that the expression of receptor tyrosine kinases varies markedly among different tumor types, and most advanced malignant tumors produce several angiogenic factors.125 The ability to screen tumor specimens for the expression of angiogenesis ligands and receptors will be critical for appropriately tailoring therapy with specific protein tyrosine kinase inhibitors to individual patients. Many receptor tyrosine kinases have direct and indirect effects on the

5  Principles of Combining Radiation Therapy and Molecular Therapeutics

viability of tumor cells and tumor-associated endothelial cells. For example, in a mouse model of pancreatic cancer, blockade of EGFR led to apoptosis and tumor regression by targeting tumor cells and tumor-associated endothelial cells.126 The blockade of EGFRs works primarily by downregulating the expression of proangiogenic factors by tumor and stromal cells. However, antagonists of EGF and its receptor may also have direct effects on endothelial cells, because the expression of EGF or TGF-α by tumor cells can lead to increased expression of EGFR by endothelial cells in vivo.127 Another proangiogenic growth factor, platelet-derived growth factor (PDGF), may also have indirect effects on the tumor vasculature by increasing the expression of proangiogenic molecules such as VEGF and basic fibroblast growth factor (bFGF)128 and direct effects on endothelial cells by increasing their expression of PDGF receptor (PDGFR) in response to PDGF production by tumor cells.129 These agents offer the potential for indirect and direct effects on the tumor cell and endothelial cell compartments. An alternative to using multiple angiogenesis inhibitors to target multiple growth factors would be to design small molecules that can target several factors. Agents that target VEGFR and other receptor tyrosine kinases (e.g., EGFR, FGFR, PDGFR) are being developed to more broadly, but perhaps less selectively, target cancer growth.130 These agents are being studied as a component of combined­modality therapy for a wide variety of malignancies. One such agent is ZD6474 (vandetanib), an orally available receptor tyrosine kinase inhibitor that selectively targets VEGFR2 and, to a lesser degree, EGFR, with resultant prevention of tumor growth and angiogenesis.131 Another agent, SU11248 (sunitinib), targets VEGFRs, PDGFR, KIT, and FLT3 and has shown significant efficacy in preclinical models of cancer and in clinical trials of renal cancer,132 but it was also associated with significant toxicity that prevented continuous administration. Sorafenib (Nexavan), an orally available inhibitor of RAF1 kinase, PDGFR,VEGFR2, VEGFR3, and KIT, has improved progression-free survival for patients with renal cell carcinoma,133 but like sunitinib, it has been associated with significant toxic effects, including diarrhea, rash, fatigue, hypertension, and cardiac ­ischemia.

Combining Antiangiogenic Agents with Radiation Therapy or Radiation and Chemotherapy The difference in their toxicity profiles makes the concept of combining antiangiogenic agents with irradiation or chemotherapy appealing. However, additional work is needed to identify feasible combinations, to optimize them, and to identify potential problems that may arise from their use. Information available from preclinical and clinical studies is presented in the following paragraphs.

Preclinical Findings Several clinical trials have been designed to study the interaction of chemotherapy and antiangiogenic therapy, but relatively few trials have tested radiation therapy with

125

antiangiogenic therapy. The long-standing assumption was that angiogenesis inhibitors would impair the effect of ionizing radiation by inducing tumor hypoxia. However, in the mid-1990s, Teicher and colleagues134 observed that antiangiogenic therapy given in combination with the angiogenesis inhibitors TNP-470 and minocycline, a weak inhibitor of metalloproteinase, improved tumor oxygenation and enhanced the antitumor effect of radia­ tion therapy in mice. Subjecting mice with primary tumors known to produce proangiogenic factors to curative radiation therapy can lead to the rapid growth of previously dormant metastases that can kill the animals within 18 days after the completion of therapy.98 However, in the same study, the systemic administration of recombinant angiostatin, an angiogenesis inhibitor, prevented the growth of the metastases after irradiation of the primary tumor.98 The findings from these preclinical studies suggest that at least some patients can benefit from the addition of antiangiogenic therapy during or after radiation therapy to prevent the growth of distant metastases. However, the timing and duration of antiangiogenic therapy required to yield such an effect remain unknown. To clarify the possible involvement of tumor oxygenation in the observed benefit from combining antiangiogenic agents with radiation therapy, the Jain laboratory135 assessed tumor hypoxia by measuring the partial pressures of oxygen within hormone-dependent male mouse mammary carcinomas (intratumoral Po2) in mice during therapy with anti-VEGFR-2 antibodies. They found that ­intratumoral Po2 levels initially decreased, indicating hypoxia and vessel regression, but it then subsequently increased, suggesting a second wave of oxygenation and angiogenesis that would enhance the effectiveness of oxygen-dependent treatments such as radiation therapy.135 However, results of another study by Murata and colleagues136 indicated that concurrent treatment of mouse breast carcinoma xenografts with TNP-470 and fractionated radiation therapy led to decreases in tumor oxygenation and corresponding decreases in tumor control. Whether antiangiogenic therapy enhances or impedes the effect of radiation therapy may depend on the tumor microenvironment. To explore this idea, Lund and others137 implanted glioblastoma multiforme xenografts in the thighs or the craniums of mice and then treated the mice with TNP-470 or radiation, or both. In the mice with the thigh tumors, the combined-modality treatment significantly enhanced the antitumor effect of each modality; however, no such enhancement was observed in the mice with intracranial tumors. The reason for this difference was not determined, but it is tempting to speculate that differences in the capillary beds and microenvironment of the brain compared with those in the subcutaneous tissues of the thigh might have contributed to the differences in response. Further evidence of potential antitumor benefit from using antiangiogenic agents with radiation therapy comes from members of the Weichselbaum laboratory,138 who demonstrated synergistic effects from the addition of the specific antiangiogenic factor angiostatin to irradiation for a variety of transplantable tumors in mice; that study involved low doses of both modalities and large, established tumors. Other agents with antiangiogenic activity have been tested in combination with concurrent radiation therapy in preclinical models. Milas and colleagues139 found that a

126

PART 1  Principles

s­ elective inhibitor of cyclooxygenase 2 (COX-2) that also has antiangiogenic properties enhanced the antitumor effects of radiation therapy; those findings are ­discussed further under “Cyclooxygenases as Targets in Cancer Therapy.” Anti-VEGF antibodies also have enhanced the response to radiation therapy in mice bearing a variety of human tumor cell types.94 Investigators at Vanderbilt University found that blocking the VEGFR2 receptor on tumor-associated endothelium eliminated the radioresistance phenotype of glioblastoma multiforme and melanoma xenografts in vivo.140 That same group later found that AZD2171, a potent inhibitor of several receptor tyrosine kinases (including all three VEGFRs), enhanced the antitumor effects of radiation therapy and blocked the proliferation of H460 lung cancer xenografts in mice.141 The first published comparison of different sequencing of radiation therapy and antiangiogenic therapy142 showed that concurrent administration of radiation and angiostatin was more effective in a mouse model of Lewis lung carcinoma than was radiation therapy followed by antiangiogenic therapy. Later studies of another agent showed that the VEGFR antagonist PTK787 enhanced the radiation response in a human squamous cell carcinoma mouse model, but only when it was administered after radiation therapy.143 A similar effect was observed when the VEGFR and EGFR antagonist ZD6474 (vandetanib) was used in combination with radiation therapy; the most profound enhancement of radiation response was observed when the drug was given immediately after fractionated radiation therapy.144 Results of a study of an orthotopic xenograft model of human lung adenocarcinoma (H441) showed that vandetanib was better than paclitaxel in suppressing angiogenesis, diminishing metastatic tumor spread through the pleural cavity, and enhancing the effects of radiation on tumor vasculature and growth.145 Winkler and colleagues146 systematically evaluated several treatment schedules for the combination of DC101 (a VEGFR2-specific monoclonal antibody) and radiation therapy for orthotopic glioblastoma xenografts in mice. Findings from that study indicated that VEGFR2 blockade temporarily normalized tumor blood vessels (defined in terms of changes in vessel coverage by pericytes and basement membrane, found to be regulated by angiopoietin 1), resulting in improved tumor oxygenation and an enhanced response to radiation therapy. However, the period of normalization and the window of opportunity to exploit it were relatively brief, suggesting that these findings may be difficult to translate into clinical practice.

Clinical Findings Clinical studies testing the combination of radiation therapy with antiangiogenic agents have not gained momentum until recently because of the toxicity of some of the available agents and concerns that this type of treatment may induce tumor hypoxia and tumor resistance. Side effects of VEGF inhibitors observed in clinical trials have included hemorrhage, which has occasionally been fatal in patients with lung cancer,147 and thrombosis.121,148 Willett and colleagues149 designed a phase I study in which patients with locally advanced rectal adenocarcinoma were treated with bevacizumab for 2 weeks, followed by bevacizumab with 5-fluorouracil, external beam radiation therapy to the pelvis, and surgery 7 weeks later. The study design included the analysis of several noninvasive and

invasive surrogates for response to the combined therapy. Although the investigators concluded that the combination of bevacizumab with chemoradiation therapy was safe and potentially effective, subsequent studies have demons­ trated increased toxicity on escalation of the bevacizumab dose.150 In another study of patients with pancreatic cancer, concurrent bevacizumab did not significantly increase the acute toxicity of capecitabine-based chemoradiation therapy, and it might have enhanced antitumor efficacy.151 However, ulceration and bleeding occurred in the radiation field when the tumor involved the duodenal mucosa. The encouraging preclinical results for vandetanib summarized in the previous section have led to this agent being studied in several clinical protocols to evaluate its safety, efficacy, and optimal sequencing with other treatment modalities.131,144 In summary, the early clinical experience with antiangiogenic therapy combined with radiation therapy or chemoradiation therapy is sufficiently encouraging to warrant additional clinical trials, although they face several challenges. Antiangiogenic therapy needs to be integrated with existing and emerging anticancer therapies, and it may need to be optimized on a patient-specific basis. However, conventional strategies for monitoring the effectiveness of anticancer therapies may not apply to antiangiogenic agents, and novel surrogates of response must be developed and validated. Angiogenesis profiles based on circulating levels of proangiogenic and antiangiogenic factors; analysis of tumors before and after therapy for evidence of tumor and endothelial cell proliferation, apoptosis, and microvessel density; and means of imaging tumor blood flow, blood volume, and metabolism need to be developed further. The design of future clinical trials should include measures of safety and efficacy and a means of validating invasive and noninvasive surrogates of response.

Cyclooxygenases as Targets in Cancer Therapy Expression and Function The COX-1 and COX-2 enzymes mediate the production of prostaglandins from arachidonic acid (Fig. 5-3).152-154 COX-1 is a ubiquitous enzyme expressed constitutively in normal tissues; COX-2 is an inducible enzyme mainly expressed in pathologic states, primarily inflammatory processes and tumors. COX-2 is induced by several factors, including proinflammatory cytokines such as TNF-α, interleukin-1β, and platelet activity factors; growth factors such as EGF, PDGF, bFGF, and TGF-β; mitogenic substances; hypoxia; and oncogenes. In normal tissues, prostaglandins regulate various homeostatic physiologic functions such as vasomotility, platelet aggregation, and protection from different types of injury, including that inflicted by radiation. In tumors, the presence of COX-2, whether induced by prostaglandins or by mechanisms independent of prostaglandins, is conducive to progressive growth, invasion, and metastasis, and it has been linked with resistance to cytotoxic therapy. COX-2 is overexpressed in premalignant lesions and established tumors (reviewed by Liao and others154). A causal link between COX-2 and carcinogenesis has been demonstrated experimentally in studies of transgenic

5  Principles of Combining Radiation Therapy and Molecular Therapeutics Membrane phospholipids Phospholipase A2 Arachidonic acid

PGG2 NSAIDs

COX-1

NSAIDs COX-2 Cox-2 inhibitors

PGH2

Tissue specific isomerases

PGE2

PGF2α

PGD2

COX-1 • Ubiquitous • Normal tissue • Various physiological functions: – GI tract integrity – Vasodilation, vasoconstriction – Platelet aggregation – Kidney function

PGI2

TXA2

COX-2 • Inducible • Inflammatory states • Cancer

FIGURE 5-3.  Prostanoids are synthesized from arachidonic acid with the involvement of cyclooxygenase 1 and 2 (COX-1 and COX-2). Phospholipase A2 generates free arachidonic acid from membrane phospholipids. COX-1 or COX-2 activity converts arachidonic acid to prostaglandin (PG) G2, followed by conversion to PGH2 into other prostaglandin isoforms or thromboxane (TX) A2. Nonsteroidal anti-inflammatory drugs (NSAIDs) inhibit COX-1 and COX-2, whereas specific COX-2 inhibitors inhibit COX-2 only. (From Liao Z, Milas L. COX-2 and its inhibition as a molecular target in the prevention and treatment of lung cancer. Expert Rev Anticancer Ther 2004;4:543-560.)

mice lacking one or both alleles of COX-2. For example, the presence of COX-2 was found to be associated with higher numbers of premalignant colon cancer lesions in intestines155 and a higher incidence of premalignant breast cancer lesions in mammary glands.156 Conversely, treating rats157 or mice158 with selective COX-2 inhibitors, notably celecoxib, reduced the development of malignant lesions in those animals. COX-2 has been detected in a broad range of tumor types, including head and neck, lung, colon, breast, and pancreatic cancer.154,159,160 Upregulation of COX-2 also has been found in some stromal cells that infiltrate tumors, in endothelial cells of tumor vasculature, and even in nonneoplastic epithelium adjacent to the tumors. Increasing evidence suggests that the presence of COX-2 is associated with an increased propensity for metastatic spread and poor prognosis,149,159,160 as well as a poor response to chemotherapy or radiation therapy.154 In one study involving analysis of tissue samples of cervical carcinoma from women given radiation therapy, the absence of COX-2

127

e­ xpression was associated with better 5-year overall survival rates (75% versus 35% for those whose tumors expressed COX-2).161 COX-2 influences carcinogenesis, tumor growth, and resistance to therapy through several mechanisms, including stimulation of cell growth, angiogenesis, invasion, ­metastasis, resistance to apoptosis, and suppression of the immunosurveillance system.152,154 The radioprotective effect of prostaglandin E2, a major metabolite of COX-2, may be an additional mechanism by which COX-2–expressing tumors resist radiation therapy.153

COX-2 Inhibitors Nonsteroidal anti-inflammatory drugs (NSAIDs) have broad inhibitory effects on COX enzymes. Epidemiologic findings have suggested that nonspecific NSAIDs and more selective COX-2 inhibitors such as celecoxib can prevent the development of colorectal, esophageal, and lung cancer.154,162-167 Preclinical data have shown that selective COX-2 inhibitors can enhance tumor response to radiation and chemotherapeutic agents.152,154,168,169 The ability of the selective COX-2 inhibitor SC-236 to enhance the radiosensitivity of a murine NFSa fibrosarcoma, for example, is illustrated in Figure 5-4. Although this agent by itself had only minor effects on tumor growth, it greatly enhanced the response of the tumors to irradiation, as assessed by regrowth delay and tumor control assays.170 COX-2 inhibitors seem to enhance tumor radioresponse through several mechanisms, including direct ­interaction with tumor cells, thereby enhancing their ­radiosensitivity,171-174 and indirect actions by modulating the tumor cell microenvironment, primarily by suppressing angiogenesis and stimulating immune functions.152,154 In vitro studies showed that COX-2 inhibitors increased cellular radiosensitivity by inhibiting DNA repair,173,174 by increasing susceptibility to radiation-induced apoptosis,172 and by arresting cells in the radiosensitive phases of the cell cycle.173 Inspired by these preclinical findings, investigators have undertaken several clinical trials in which COX-2 inhibitors are combined with radiation therapy or chemotherapy, or both. One such study, a phase I trial done at The University of Texas M. D. Anderson Cancer Center, was designed to determine the maximum tolerated dose of celecoxib when combined with radiation therapy for patients with inoperable non–small cell lung cancer (NSCLC) and poor performance status.175 This trial included patients with locally advanced cancer with obstructive pneumonia or minimal metastatic disease; those with early-stage tumors who could not undergo surgery for medical reasons; and those who had been given induction chemotherapy before radiation therapy. Fortyseven patients received 45 to 66 Gy in 15 to 35 fractions combined with daily celecoxib doses of 200 to 800 mg. The local progression-free survival rate at 20 months was highly encouraging (66%) and similar to that observed after concurrent radiation therapy plus chemotherapy. Toxic effects included grade 1 or 2 nausea and esophagitis in 23 (49%) of 47 patients, grade 3 pneumonitis in two patients, and hypertension that did not resolve after discontinuation of celecoxib in one patient (on the 400-mg twice-daily regimen). Other trials have also shown encouraging results.154 Unfortunately, the discovery176 of

128

Growth delay (days)

PART 1  Principles

Tumor size (mm)

12

10

60 50 40 30 20 10 0

Cyclins and Cyclin-Dependent Kinases as Targets in Cancer Therapy Expression and Function

30 40 50 Radiation dose (Gy)

8

6

4 0

5

10

A

15

20

25

30

35

40

45

50

Days after irradiation 100

Tumor cure (%)

80 60 40 20 0 20

B

30

40

50

60

70

80

Radiation dose (Gy)

FIGURE 5-4.  The effects of the cyclooxygenase 2 (COX-2) inhibitor SC-236 on murine NFSa cells are shown for tumor growth delay (A) and tumor cure (B) after irradiation. Mice with NSFa fibrosarcoma cells implanted in the legs were treated with SC-236 (5 mg/kg) in their drinking water for 10 days, beginning when the tumors were 6 mm in diameter, and they were irradiated (30 Gy) when the tumors reached 8 mm in diameter. A, The graph shows the curves for the vehicle (red circles), SC-236 (red triangles), 30 Gy (blue circles), and SC-236 plus 30 Gy (blue triangles). Vertical bars represent the standard error. Inset, Tumor growth delay is plotted as a function of radiation dose, as shown by the curves for radiation only (red circles) and SC-236 plus radiation (blue circles). B, Tumor cure is plotted as a function of radiation dose, as shown by the curves for radiation only (red circles) and SC236 plus radiation (blue triangles). Horizontal bars represent 95% confidence intervals at the radiation dose level yielding 50% tumor cure (TCD50). (From Milas L, Kishi K, Hunter N, et al. Milas L, Kishi K, Hunter N J Natl Cancer Inst 1999;91:15011504.)

increased cardiovascular toxicity after long-term use of COX-2 inhibitors for chemoprevention of colorectal adenoma in 2005 prematurely dampened the enthusiasm of the industrial sector for further developing celecoxib as anticancer therapy. However, it is worth stressing that the benefit-risk ratio for COX-2 inhibitors may be considered favorable for patients with cancer.

Cyclin-dependent kinases (CDKs) are serine/threonine kinases that become active when associated with regulatory proteins, primarily cyclins. Nine CDKs have been identified in humans. Cyclin D in complex with CDK4 or CDK6 facilitates phosphorylation of the retinoblastoma protein (RB1) and enhances the transcription of E2F, which is important for the progression of cells through the G1/S checkpoint of the cell cycle. This signaling pathway is often dysregulated in neoplastic cells through overexpression of cyclins, downregulation of CDK inhibitors (e.g., CDKN2A [formerly designated p16], CDKN1A [formerly p21], IFI27 [formerly p27], CDKN1C [formerly p57]), or mutations in CDKs.177-179 Upregulation and overexpression of cyclin D1, for example, are common in Barrett’s esophagus and overt esophageal adenocarcinoma; moreover, cyclin D1 overexpression is associated with increased cancer risk from Barrett’s esophagus and with higher cancer stage.180,181 Alterations in the cell cycle cascade involving cyclin D, CDK4, CDK6, INK4, RB1, or E2F have been observed in more than 80% of malignant tumors in humans.178 Abnormalities in other cyclins (e.g., A, E), CDKs (e.g., 1, 2, 7), and the CDK inhibitors CDKN1A (p21) and IFI27 (p27) have been found in many types of cancer.182,183 RB1 can become inactivated through gene mutation, overexpression of cyclins or CDKs, or decreases in CDK inhibitors.183 Loss of the IFI27 (p27) protein has been associated with poor outcome in patients with lung, breast, colon, stomach, and prostate carcinomas, and loss of the CDKN2A (p16) protein with poor outcome in patients with lung carcinoma or melanoma.184 Increased expression of cyclins D and E has been associated with poor prognosis for several types of tumors.181,185 Several lines of evidence have implicated increased cyclin D1 expression with decreased tumor response to cytotoxic agents. Milas and colleagues33 demonstrated strong correlations between the level of cyclin D1 expression in tumors and the proliferation (according to labeling of proliferating cell nuclear antigen) and radioresistance of those tumor cells (according to a TCD50 assay). Reducing cyclin D expression by means of an antisense construct has been shown to inhibit tumor growth, reduce tumorigenicity, or induce cell death in various types of cancer.186-189 Suppression of cyclin D1 expression has been shown to enhance the sensitivity of several human gastrointestinal cancer cell lines to cisplatin and 5-­fluorouracil.186,187

CDK Inhibitors in Cancer Therapy Several CDK inhibitors have been investigated during the past decade; some modulate CDK activity directly and oth­ ers perturb CDK function indirectly by affecting the CDK regulatory proteins.178,183 One such inhibitor, flavopiridol [5,7-dihydroxy-8-(4-N-methyl-2-hydroxypyridyl)-6′chloroflavone hydrochloride], is a synthetic flavone that inhibits several phosphokinases in a dose-dependent manner. At lower doses, it inhibits primarily CDKs (CDK1 to CDK9), and at higher concentrations, it inhibits receptor tyrosine kinases, receptor-associated tyrosine kinases, and

5  Principles of Combining Radiation Therapy and Molecular Therapeutics

signal-transducing kinases.190 Flavopiridol can also repress transcriptional activity, possibly by inhibiting the activi­ ty of CDK9-dependent transcription elongation factor b (P-TEFb), which led to downregulation of several relevant genes, including cyclin D1, VEGF, and the DNA-repair protein KU70 (now designated XRCC6).190,191 At the cellular level, flavopiridol has shown cytostatic (cell cycle arrest) and cytotoxic (inducing apoptosis) effects.178,183,190,191 In various preclinical models, flavopiridol by itself led to modest tumor growth delays,192,193 but it interacted synergistically with several chemotherapeutic agents (e.g., paclitaxel, gemcitabine) in inducing tumor growth delay and apoptosis.192,193 Investigations at the M. D. Anderson Cancer Center demonstrated that flavopiridol potently enhanced radioresponse in cell culture systems191 and in vivo in mouse and human tumor models.192,194 For example, flavopiridol in combination with fractionated irradiation substantially suppressed the growth of MCa-29 murine adenocarcinoma tumor, as measured by assays of tumor growth delay and tumor curability (Fig. 5-5). Clinical studies of flavopiridol based on these preclini­ cal findings have yielded mixed results. An early phase I study showed only modest activity when flavopiridol alone

MCa 29 16

Tumor size (mm)

14 12 10 8 6 4 0

10

20

30

40

50

60

70

Days after initial radiation

FIGURE 5-5.  Effects are plotted for the response of MCa-29 adenocarcinoma tumors in mice treated with flavopiridol or fractionated radiation, or both. Mice bearing 8-mm tumors in the right hind legs were untreated (red circles) or treated with flavopiridol (blue circles), local tumor irradiation (red squares), or a combination of flavopiridol and irradiation (blue squares). Radiation was delivered in 4-Gy fractions daily for 20 consecutive days; flavopiridol was given at a dose of 2.5 mg/kg intraperitoneally twice daily for 10 days. When the agents were combined, flavopiridol was given immediately after each dose of fractionated radiation and given 6 to 8 h later for a total of  20 days. Each data point represents the mean size of eight or nine tumors; bars represent the standard error. Three of eight tumors treated with flavopiridol plus fractionated radiation permanently regressed and were excluded from the regrowth analysis. The dotted horizontal line represents leg thickness of 5 mm, below which it is not possible to measure tumor size accurately. (From Mason KA, Hunter NR, Raju U, et al. Flavopiridol increases therapeutic ratio of radiotherapy by preferentially enhancing tumor ­radioresponse. Int J Radiat Oncol Biol Phys 2004;59:1181-1189.)

129

was administered in a 72-hour continuous infusion every 14 days at a dose of 50 mg/m2/day,195 but more encouraging results were subsequently obtained with higher bolus doses.196 No results are available regarding the effects of combinations of flavopiridol and radiation, with or without chemotherapy.

RAS Signaling Pathway as a Target in Cancer Therapy The RAS oncogene is abnormally activated, usually by point mutations, in many types of cancer in humans. Much preclinical evidence linking activation of RAS with a radioresistant phenotype has led to its identification as a molecular target for restoring radiosensitivity in human tumors. In cultured cell models, for example, transformation of normal rodent fibroblasts with a mutated RAS construct confers radioresistance. One approach to circumventing the radioresistance conferred by RAS activation has involved inhibitors of farnesyltransferase, an enzyme that blocks the posttranslational farnesylation of RAS normally needed to mediate its downstream signaling. Consequently, inhibitors of the farnesyltransferase enzyme can block RAS function and, as shown in preclinical studies, can restore radiosensitivity to cells with mutated RAS genes.197 Two farnesyltransferase inhibitors have been tested in combination with radiation therapy in clinical trials. Investigators at the University of Pennsylvania have conducted phase I trials of the farnesyltransferase inhibitor L-778123 and radiation therapy for locally advanced NSCLC, for head and neck cancer,198 and for pancreatic cancer.199 Although both studies identified dose levels associated with acceptable toxicity and some responses, further clinical development of L-778123 has not been pursued by the manufacturer. The Radiation Therapy Oncology Group is conducting a phase II trial of another farnesyltransferase inhibitor, R115777 (tipifarnib), in combination with radiation therapy plus chemotherapy for the treatment of pancreatic cancer200; this compound is also being examined in combination with radiation therapy for glioblastoma multiforme.201

Gene Therapy Gene therapy continues to be an attractive strategy for treating cancer, and several methods are being tested in clinical trials. Although many studies have shown that viral vectors can effectively deliver therapeutic genes to some proportion of the cells in a solid tumor with mini­ mal systemic toxicity, the overall effectiveness of these approaches when used alone has been disappointing. Strategies in which gene therapy is combined with con­ ventional irradiation or chemotherapy are hoped to be more effective and are being investigated. Combinations of gene therapy with radiation therapy are usually in one of four categories: delivery of radiosensitizing drugs or agents, introduction of one or more replacement genes, use of oncolytic viruses, or suppression of the activity of genes that confer radioresistance.202 These approaches, the first of which has been explored to the greatest extent, are discussed further subsequently.

130

PART 1  Principles

Delivery of Radiosensitizing Drugs

Gene Replacement

The most thoroughly studied strategy for using gene therapy to deliver radiosensitizing drugs has been the use of replication-incompetent adenoviral vectors engineered to express the herpes simplex virus thymidine kinase (HSV-TK) gene.203 Preclinical studies have confirmed the antitumor activity of this approach in several tumor models. In its clinical application, patients are given the vector (in this case, Ad-TK), followed by systemic treatment with ganciclovir, a prodrug that when phosphorylated by the HSV-TK gene product, ultimately becomes a potent inhibitor of mammalian DNA polymerase and kills in­ fected cells by preventing DNA synthesis. This effect may also radiosensitize affected tumor cells through com­ promising their ability to repair radiation-induced DNA double-strand breaks. The safety of this approach has been demonstrated in a phase I/II clinical trial for men with prostate cancer,204 but additional trials are needed to demonstrate its efficacy. A similar strategy involves a more complex, ­replicationcompetent vector that encodes genes for HSV-TK and a prokaryotic cytosine deaminase.205 Cytosine deaminase, when expressed in mammalian cells, deaminates the relatively nontoxic prodrug 5-fluorocytosine, thereby creating the more toxic drug 5-fluorouracil. The advantages of this approach are that the toxic effects of the 5-­fluorouracil are confined to the infected tumor cells and that the 5­fluorouracil has its own radiosensitizing effects. In this “double suicide” gene therapy approach, patients are treated with the vector Ad-CD/TKrep in combination with 5fluorocytosine and valganciclovir. Because Ad-CD/TKrep is a replication-­competent vector, it also has oncolytic properties, an attribute that is discussed later under “Oncolytic Viruses.” The safety of this double-suicide gene therapy in combination with radiation therapy has been demonstrated in phase I testing for men with prostate cancer.206 Radiosensitizing agents other than drugs can also be delivered by gene therapy. The best example is the adenoviral delivery of the gene encoding human TNF-α, a soluble cytokine with potent antitumor and antiangiogenic properties that also mediates the cellular immune response. TNF-α has shown some antitumor activity as a single agent in clinical trials; however, it also has substantial toxicity on systemic administration. TNF-α has synergistic effects with radiation therapy, presumably because it can ­ damage DNA through reactive oxygen species.207 Weichselbaum and colleagues developed and tested a ­replication-­incompetent adenoviral vector that encodes the human TNF-α gene (TNF, formerly designated TNFA), the expression of which is under the control of the radiation-inducible promoter for the immediate early growth response 1 gene (EGR1), which is inserted in the vector upstream of the start site of the TNF cDNA.208 This vector, referred to as TNFerade, is then injected into the tumor, where it prompts infected tumor cells to express TNF-α on radiation. Intratumoral injection also avoids the toxicity associated with systemic administration of this cytokine. TNFerade has demonstrated synergistic effects in combination with radiation in preclinical models of human cancer (reviewed by Mezhir and colleagues209) and was well tolerated in a phase I trial.210 Several phase II trials of this strategy are being conducted.

The second gene therapy strategy that has received significant attention involves replacing a gene that is miss­ ing or abnormally deactivated in tumor cells. The beststudied example of such a gene is the tumor suppressor TP53 (formerly designated p53), which is mutated or otherwise functionally abnormal in most types of cancer in humans. In normal cells, TP53 participates in many cellular functions, including apoptosis, genomic stability, and cell cycle checkpoints. The use of gene therapy to restore TP53 in tumor cells in which it is deleted would be expected to enhance the cells’ response to cancer therapies that rely on the apoptosis pathway to kill tumor cells. Although no consensus has been reached about the importance of apoptosis as a cell death pathway in the clinical response of human tumors to radiation, numerous preclinical tests of adenoviral-mediated TP53 (Ad-p53) have shown restoration of radiation-induced apoptosis and an overall enhancement of tumor response in various tumor models.211,212 Based on these reports, a prospective, single-arm, phase II study was undertaken to evaluate the feasibility of Ad-p53 gene transfer and radiation therapy for patients with NSCLC.213 In that study, 19 patients with nonmetastatic NSCLC who were not eligible for chemotherapy or surgery were treated on an outpatient basis with 60 Gy over 6 weeks combined with three intratumoral injections of Ad-p53 on days 1, 18, and 32. The results showed that this combination was well tolerated and that some proportion of patients showed evidence of tumor regression. Although apoptosis was not assessed, serial biopsies of the tumors suggested that upregulation of the expression of the proapoptotic gene BAK correlated with Ad-p53 injection. These promising results are being followed up in larger trials.

Oncolytic Viruses Most adenoviral vectors that have been tested for gene therapy purposes have been engineered to disable their ability to replicate within the infected cells. This process normally involves splicing out regions of the viral genome required for its replication, the so-called early genes (E1-E4). However, some novel vectors have been developed in which the replication competency of the virus is left intact. When tumor cells are infected with these vectors, the ensuing viral replication eventually leads to oncolysis, which kills the infected cells. The lysed cells release viral progeny, which can infect neighboring, previously uninfected cells. This attribute of replication-competent vectors may overcome the major limitation of replication-incompetent vectors—their poor distribution throughout the injected tumor. Although most such vectors are designed for use primarily as single agents because of their cytotoxicity, some have been used in combination with irradiation. The best example is the combination of radiation therapy with administration of the replication-competent Ad-CD/TK vector for prostate cancer206 (discussed under “Delivery of Radiosensitizing Drugs”). Another type of oncolytic vector is the conditionally replicative vector, the best studied of which has been ONYX-15 (dl1520). This adenoviralbased vector lacks the E1B gene, the product of which binds to and inactivates wild-type TP53 in infected cells, a step that is required for viral replication. ONYX-15 can replicate only in tumor cells with mutant or deleted TP53

5  Principles of Combining Radiation Therapy and Molecular Therapeutics

and presumably not in normal cells, providing it with some tumor cell specificity. The safety of ONYX-15 as singleagent therapy has been demonstrated in several clinical trials (reviewed by Vecil and Lang214), and preclinical studies have shown at least additive effects when ONYX-15 is combined with irradiation.215,216 At least one other conditionally replicative vector is under development. Ad-Delta24-RGD has a deletion in the E1A region of the viral genome, which is responsible for binding to the cellular RB1 protein.214 In normal cells, the vector cannot bind to RB1 and therefore cannot replicate; however, in tumor cells with nonfunctional RB1, replication of this vector is supported. Preclinical studies have shown enhancement of xenograft tumor response when Ad-Delta24-RGD is combined with radiation.217,218 This strategy is expected to be tested in clinical trials in the near future.

Silencing Gene Expression Most of the gene therapy strategies discussed have been designed to directly or indirectly sensitize the targeted tumor cells to radiation by inducing the expression of exogenous genes. An alternative approach is to introduce agents that specifically suppress the expression of cellular genes whose protein products participate in mechanisms that mediate radioresistance. The two technologic means of silencing gene expression are antisense oligonucleotides and RNA interference (RNAi). Antisense oligonucleotides are short, single-stranded deoxynucleotide sequences designed to bind to mRNA that is complementary to the gene of interest in a sequence-specific manner.219 The resulting mRNA/oligonucleotide complex causes the targeted mRNA to be enzymatically destroyed by nucleases, leading to lack of transcription of the gene of interest. Antisense oligonucleotides have been widely tes­ted for the treatment of various diseases, including cancer. In preclinical studies, antisense oligonucleotides have been shown to radiosensitize human tumor cells by targeting different genes involved in the radiation response, especially genes whose protein products mediate blocks to apoptosis (e.g., BCL2) or are involved in the repair of radiation-induced DNA double-strand breaks (e.g., XRCC6 [formerly KU70]). Finding ways of effectively delivering antisense oligonucleotides to patients has been problematic and has required extensive development. However, encapsulating antisense oligonucleotides in liposomes for systemic delivery to patients seems to be a promising approach. A phase I study in which a liposomeencapsulated RAF antisense oligonucleotide was used in combination with radiation therapy for patients with advanced malignancies showed the combination to be well tolerated.220 Gene silencing by RNAi is accomplished through the use of small interfering RNAs (siRNAs) consisting of doublestranded RNA sequences complementary to the gene of interest.221 After being taken up by the recipient cell, the siRNA is processed by cellular enzymes to release the antisense strand, which then targets the respective gene’s mRNA for destruction, much like the process for antisense oligonucleotides described earlier. RNAi technology has become quite efficient and essentially results in a functional knockout of the targeted gene. Preclinical studies have described several examples in which siRNA directed

131

to genes whose products are critical to mediating a radioresistant phenotype has induced radiosensitization of tumor cells. Although no clinical studies of siRNA in combination with radiation therapy have been reported, such trials are expected in the near future.

Immunotherapy Mechanisms It has long been recognized that malignant tumors may express tumor-specific antigens that can elicit immuno­ logic responses that can restrict tumor growth. An effici­ ently functioning immune system may improve the efficacy of conventional treatment modalities, whereas immune deficiency may result in the opposite outcome.222 Tumor antigens evoke innate and adaptive immune res­ponses. Innate immunity involves the activation of phagocytic, natural killer, and antigen-presenting cells, and it is trig­ gered within hours to days after antigen presentation. Adaptive immunity involves functions of T and B lym­ phocytes and begins within days to months after antigen presentation.222 Both active immunization with heavily irradiated tumor cells and nonspecific immunotherapy with bacterial adjuvants to boost the immune response have been extensively investigated. One of the most thoroughly studied adjuvants is complete Freund’s adjuvant, which consists of a suspension of killed bacillus Calmette-Guérin (BCG) bacteria in mineral oil. Several types of bacteria or bacterial extracts, primarily BCG and Corynebacterium parvum, have been used to stimulate antitumor immunologic reactions as early as the 1970s.223,224 Although these agents are potent stimulators of many facets of immunologic reactions and possess substantial activity against murine tumors when given alone or in combination with chemotherapy or irradiation, they have shown little therapeutic benefit in human clinical trials.223,224 An important limiting factor in the clinical application of these agents is the toxicity ­associated with their repeated administration.223,224

Molecular Immunotherapy In the mid-1990s, it was discovered that the key immunostimulatory activity of bacteria resides in their DNA225—specifically in unmethylated cytosine-guanine (CpG) motifs, which are prevalent in bacteria but not in vertebrates.226 This discovery led to the synthesis of oligodeoxynucleotides containing unmethylated CpG motifs that showed strong immunostimulatory activity but little toxicity.226-228 CpG oligodeoxynucleotides are recognized by the innate immune cells of vertebrates, primarily plasmacytoid dendritic cells, through the toll-like receptor 9. These CpG oligodeoxynucleotides have been shown to have antitumor activity on their own229-232 and to enhance the antitumor activity of chemotherapeutic agents230 and that of radiation given as a single exposure232 or in fractionated regimens.231 Substantial increases in tumor cure rates from the addition of CpG oligodeoxynucleotides to fractionated irradiation for the treatment of a murine sarcoma model are illustrated in Figure 5-6. Tumors in mice treated with the CpG oligodeoxynucleotides and radiation were heavily infiltrated by host inflammatory cells, primarily lymphocytes232; moreover, the mice that were cured by the combined treatment remained resistant to re-inoculation

132

PART 1  Principles

100

CpG ODN 1826

Control

Tumor cure (%)

80 60 40 20 0 0

10

20

30

40

50

60

70

80

90

100

Total radiation dose (Gy)

FIGURE 5-6.  Effects on tumor curability are shown for the addition of the CpG motif–based oligodeoxynucleotide (ODN) 1826 to radiation therapy. The percentage of tumor cures is plotted as a function of radiation dose. Mice bearing FSa fibrosarcoma tumors in the legs were exposed to a range of fractionated doses when the tumors reached 8 mm in diameter. Mice were also treated with seven doses of active CpG ODN 1826 (blue circles) or inactive ODN 2138 (red circles) at a dose of 100 μg per mouse given subcutaneously peritumorally. The first dose was given when tumor diameters were 6 mm, the second dose at 8 mm, and then weekly for 5 additional weeks. The radiation dose yielding 50% tumor cure (TCD50) was determined at 100 days after irradiation. Horizontal bars represent 95% confidence intervals. (From Mason KA, Ariga H, Neal R, et al. Targeting toll-like receptor 9 with CpG oligodeoxynucleotides enhances tumor response to fractionated radiotherapy. Clin Cancer Res 2005;11:361-369.)

of tumor cells for many months after tumor regression.231 These results warrant testing of this therapeutic approach in carefully designed clinical trials, as was observed by Koski and Czerniecki233 and Sutherland.234

Summary Research on EGFR signaling exemplifies the development of novel cancer therapy strategies that are based on better understanding of tumor biology. The discovery of how perturbation of EGFR signaling affects malignant transformation and tumor response to therapy led to the conception and completion of a clinical research program that produced a new front-line therapy modality that improved the survival of patients with locally advanced HNSCC. Although this pivotal work established the proof of principle that tumors can be selectively sensitized to radiation therapy through modulation of a signaling pathway, the clinical benefit resulting from this strategy remains rather modest, and many questions remain to be addressed. Validation of this concept, however, has provided a robust foundation for expanding investigations aiming at augmenting tumor response to therapy by modulating other signaling pathways, individually or in combination. These efforts are anticipated to yield novel therapies for various types of cancer in the near future.

REFERENCES   1. Cohen S, Carpenter G, King L Jr. Epidermal growth factorreceptor-protein kinase interactions. Co-purification of receptor and epidermal growth factor-enhanced phosphorylation activity. J Biol Chem 1980;255:4834-4842.

  2. Cohen LF, Balow JE, Magrath IT. Acute tumor lysis syndrome: a review of 37 patients with Burkitt’s lymphoma. Am J Med 1980;68:486-491.   3. Downward J, Yarden Y, Mayes E, et al. Close similarity of epidermal growth factor receptor and v-erb-B oncogene protein sequences. Nature 1984;307:521-527.   4. Davies RL, Grosse VA, Kucherlapati R, et al. Genetic analysis of epidermal growth factor action: assignment of human epidermal growth factor receptor gene to chromosome 7. Proc Natl Acad Sci U S A 1980;77:4188-4192.   5. Haley J, Whittle N, Bennet P, et al. The human EGF receptor gene: structure of the 110 kb locus and identification of sequences regulating its transcription. Oncogene Res 1987;1:375-396.   6. Burgess AW, Cho HS, Eigenbrot C, et al. An open-and-shut case? Recent insights into the activation of EGF/ErbB receptors. Mol Cell 2003;12:541-552.   7. Nishinaka T, Yabe-Nishimura C. EGF receptor-ERK pathway is the major signaling pathway that mediates upregulation of aldose reductase expression under oxidative stress. Free Radic Biol Med 2001;31:205-216.   8. Darnell JE Jr, Kerr IM, Stark GR. Jak-STAT pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins. Science 1994;264:1415-1421.   9. Harari PM, Huang S. Radiation combined with EGFR signal inhibitors: head and neck cancer focus. Semin Radiat Oncol 2006;16:38-44. 10. Kijima T, Niwa H, Steinman RA, et al. STAT3 activation abrogates growth factor dependence and contributes to head and neck squamous cell carcinoma tumor growth in vivo. Cell Growth Differ 2002;13:355-362. 11. Xi S, Zhang Q, Gooding WE, et al. Constitutive activation of Stat5b contributes to carcinogenesis in vivo. Cancer Res 2003;63:6763-6771. 12. Hynes NE, Lane HA. ERBB receptors and cancer: the complexity of targeted inhibitors. Nat Rev Cancer 2005;5:341-354. 13. Thomas SM, Coppelli FM, Wells A, et al. Epidermal growth factor receptor-stimulated activation of phospholipase Cgamma-1 promotes invasion of head and neck squamous cell carcinoma. Cancer Res 2003;63:5629-5635. 14. Dittmann K, Mayer C, Fehrenbacher B, et al. Radiationinduced epidermal growth factor receptor nuclear import is linked to activation of DNA-dependent protein kinase. J Biol Chem 2005;280:31182-31189. 15. Lin SY, Makino K, Xia W, et al. Nuclear localization of EGF receptor and its potential new role as a transcription factor. Nat Cell Biol 2001;3:802-808. 16. Hanada N, Lo HW, Day CP, et al. Co-regulation of B-Myb expression by E2F1 and EGF receptor. Mol Carcinog 2006;45:10-17. 17. Khanna KK, Jackson SP. DNA double-strand breaks: signaling, repair and the cancer connection. Nat Genet 2001;27:247-254. 18. Bao J, Alroy I, Waterman H, et al. Threonine phosphorylation diverts internalized epidermal growth factor receptors from a degradative pathway to the recycling endosome. J Biol Chem 2000;275:26178-26186. 19. Hanover JA, Beguinot L, Willingham MC, et al. Transit of receptors for epidermal growth factor and transferrin through clathrincoated pits. Analysis of the kinetics of receptor entry. J Biol Chem 1985;260:15938-15945. 20. Herbst JJ, Opresko LK, Walsh BJ, et al. Regulation of postendocytic trafficking of the epidermal growth factor receptor through endosomal retention. J Biol Chem 1994;269:12865-12873. 21. Warren CM, Landgraf R. Signaling through ERBB receptors: multiple layers of diversity and control. Cell Signal 2006;18:923933. 22. Pike LJ, Miller JM. Cholesterol depletion delocalizes phosphatidylinositol bisphosphate and inhibits hormone-stimulated phosphatidylinositol turnover. J Biol Chem 1998;273:22298-22304. 23. Grandis J, Tweardy D. Elevated levels of transforming growth factor α and epidermal growth factor receptor messenger RNA are early markers of carcinogenesis in head and neck cancer. Cancer Res 1993;53:3579-3584.

5  Principles of Combining Radiation Therapy and Molecular Therapeutics 24. Grandis JR, Melhem MF, Barnes EL, et al. Quantitative immunohistochemical analysis of transforming growth factor- a and epidermal growth factor receptor in patients with squamous cell carcinoma of the head and neck. Cancer 1996;78:1284-1292. 25. Schmidt-Ullrich RK, Mikkelsen RB, Dent P, et al. Radiationinduced proliferation of the human A431 squamous carcinoma cells is dependent on EGFR tyrosine phosphorylation. Oncogene 1997;15:1191-1197. 26. Lynch TJ, Bell DW, Sordella R, 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;350:2129-2139. 27. Wikstrand CJ, McLendon RE, Friedman AH, et al. Cell surface localization and density of the tumor-associated variant of the epidermal growth factor receptor. EGFRvIII. Cancer Res 1997;57:4130-4140. 28. Etienne-Grimaldi MC, Pereira S, Magne N, et al. Analysis of the dinucleotide repeat polymorphism in the epidermal growth factor receptor (EGFR) gene in head and neck cancer patients. Ann Oncol 2005;16:934-941. 29. Cappuzzo F, Hirsch FR, Rossi E, et al. Epidermal growth factor receptor gene and protein and gefitinib sensitivity in non–smallcell lung cancer. J Natl Cancer Inst 2005;97:643-655. 30. Psyrri A, Yu Z, Weinberger PM, et al. Quantitative determination of nuclear and cytoplasmic epidermal growth factor receptor ­expression in oropharyngeal squamous cell cancer by using automated quantitative analysis. Clin Cancer Res 2005;11:5856-5862. 31. Lo HW, Hung MC. Nuclear EGFR signalling network in cancers: linking EGFR pathway to cell cycle progression, nitric oxide pathway and patient survival. Br J Cancer 2006;94:184-188. 32. Merlano M, Vitale V, Rosso R, et al. Treatment of advanced ­squamous-cell carcinoma of the head and neck with alternating chemotherapy and radiotherapy. N Engl J Med 1992;327: 1115-1121. 33. Milas L, Akimoto T, Hunter NR, et al. Relationship between cyclin D1 expression and poor radioresponse of murine carcinomas. Int J Radiat Oncol Biol Phys 2002;52:514-521. 34. Withers HR, Taylor JM, Maciejewski B. The hazard of accelerated tumor clonogen repopulation during radiotherapy. Acta Oncol 1988;27:131-146. 35. Schmidt-Ullrich RK, Johnson CR, Khandelwal S, et al. Concomitant boost-accelerated, superfractionated irradiation for the treatment of advanced carcinomas of the oral cavity and oropharynx. Recent Results Cancer Res 1994;134:69-79. 36. Schmidt-Ullrich RK, Valerie K, Chan W, et al. Expression of oestrogen receptor and transforming growth factor-alpha in MCF-7 cells after exposure to fractionated irradiation. Int J Radiat Biol 1992;61:405-415. 37. Kavanagh BD, Lin PS, Chen P, et al. Radiation-induced enhanced proliferation of human squamous cancer cells in vitro: a release from inhibition by epidermal growth factor. Clin Cancer Res 1995;1:1557-1562. 38. Schmidt-Ullrich RK, Valerie K, Fogleman PB, et al. Radiationinduced autophosphorylation of epidermal growth factor receptor in human malignant mammary and squamous epithelial cells. Radiat Res 1996;145:81-85. 39. Schmidt-Ullrich RK, Contessa JN, Lammering G, et al. ERBB receptor tyrosine kinases and cellular radiation responses. Oncogene 2003;22:5855-5865. 40. Akimoto T, Hunter NR, Buchmiller L, et al. Inverse relationship between epidermal growth factor receptor expression and radiocurability of murine carcinomas. Clin Cancer Res 1999;5:2884-2890. 41. Ang KK, Berkey BA, Tu X, et al. Impact of epidermal growth factor receptor expression on survival and pattern of relapse in patients with advanced head and neck carcinoma. Cancer Res 2002;62:7350-7356. 42. Liang K, Ang KK, Milas L, et al. The epidermal growth factor receptor mediates radioresistance. Int J Radiat Oncol Biol Phys 2003;57:246-254. 43. Mendelsohn J, Masui H, Goldenberg A. Anti-epidermal growth factor receptor monoclonal antibodies may inhibit A431 tumor cell proliferation by blocking an autocrine pathway. Trans Assoc Am Physicians 1987;100:173-178.

133

44. Sato JD, Kawamoto T, Le AD, et al. Biological effects in vitro of monoclonal antibodies to human epidermal growth factor receptors. Mol Biol Med 1983;1:511-529. 45. Huang SM, Bock JM, Harari PM. Epidermal growth factor receptor blockade with C225 modulates proliferation, apoptosis, and radiosensitivity in squamous cell carcinomas of the head and neck. Cancer Res 1999;59:1935-1940. 46. Huang SM, Harari PM. Epidermal growth factor receptor inhibition in cancer therapy: biology, rationale and preliminary clinical results. Invest New Drugs 1999;17:259-269. 47. Milas L. Chemoradiation interactions: potential of newer chemotherapeutic agents. In Perry MC (ed): American Society of Clinical Oncology Education Book, 2000, pp 207-213. 48. Saleh MN, Raisch KP, Stackhouse MA, et al. Combined modality therapy of A431 human epidermoid cancer using anti-EGFR antibody C225 and radiation. Cancer Biother Radiopharm 1999;14:451-463. 49. Nasu S, Ang KK, Fan Z, et al. C225 antiepidermal growth factor receptor antibody enhances tumor radiocurability. Int J Radiat Oncol Biol Phys 2001;51:474-477. 50. Krause M, Schutze C, Petersen C, et al. Different classes of EGFR inhibitors may have different potential to improve local tumour control after fractionated irradiation: a study on C225 in FaDu hSCC. Radiother Oncol 2005;74:109-115. 51. Bianco C, Tortora G, Bianco R, et al. Enhancement of antitumor activity of ionizing radiation by combined treatment with the selective epidermal growth factor receptor-tyrosine kinase inhibitor ZD1839 (Iressa). Clin Cancer Res 2002;8:3250-3258. 52. Chinnaiyan P, Huang S, Vallabhaneni G, et al. Mechanisms of enhanced radiation response following epidermal growth factor receptor signaling inhibition by erlotinib (Tarceva). Cancer Res 2005;65:3328-3335. 53. She Y, Lee F, Chen J, et al. The epidermal growth factor receptor tyrosine kinase inhibitor ZD1839 selectively potentiates radiation response of human tumors in nude mice, with a marked improvement in therapeutic index. Clin Cancer Res 2003;9:3773-3778. 54. Solomon B, Hagekyriakou J, Trivett MK, et al. EGFR blockade with ZD1839 (“Iressa”) potentiates the antitumor effects of single and multiple fractions of ionizing radiation in human A431 squamous cell carcinoma. Epidermal growth factor receptor. Int J Radiat Oncol Biol Phys 2003;55:713-723. 55. Robert F, Ezekiel MP, Spencer SA, et al. Phase I study of antiepidermal growth factor receptor antibody cetuximab in combination with radiation therapy in patients with advanced head and neck cancer. J Clin Oncol 2001;19:3234-3243. 56. Bonner JA, Harari PM, Giralt J, et al. Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. N Engl J Med 2006;354:567-578. 57. Pfister DG, Su YB, Kraus DH, et al. Concurrent cetuximab, cisplatin, and concomitant boost radiotherapy for locoregionally advanced, squamous cell head and neck cancer: a pilot phase II study of a new combined-modality paradigm. J Clin Oncol 2006;24:1072-1078. 58. Burtness B, Goldwasser MA, Flood W, et al. Phase III randomized trial of cisplatin plus placebo compared with cisplatin plus ­ cetuximab in metastatic/recurrent head and neck cancer: an Eastern Cooperative Oncology Group study J Clin Oncol 2005;23:8646-8654. 59. Soulieres D, Senzer NN, Vokes EE, et al. Multicenter phase II study of erlotinib, an oral epidermal growth factor receptor tyrosine ­kinase inhibitor, in patients with recurrent or metastatic squamous cell cancer of the head and neck. J Clin Oncol 2004;22:77-85. 60. Chung KY, Shia J, Kemeny NE, et al. Cetuximab shows activity in colorectal cancer patients with tumors that do not express the epidermal growth factor receptor by immunohistochemistry. J Clin Oncol 2005;23:1803-1810. 61. Saltz L. Epidermal growth factor receptor-negative colorectal cancer: is there truly such an entity? Clin Colorectal Cancer 2005;5;(Suppl 2):S98-S100. 62. Pal SK, Pegram M. Epidermal growth factor receptor and signal transduction: potential targets for anti-cancer therapy. Anticancer Drugs 2005;16:483-494.

134

PART 1  Principles

63. Choong NW, Cohen EE. Epidermal growth factor receptor directed therapy in head and neck cancer. Crit Rev Oncol Hematol 2006;57:25-43. 64. Milas L, Fang F, Mason KA, et al. Importance of maintenance therapy in C225-induced enhancement of tumor control by fractionated radiation. Int J Radiat Oncol Biol Phys 2007;67: 568-572. 65. Lammering G, Hewit TH, Valerie K, et al. EGFRvIII-mediated radioresistance through a strong cytoprotective response. Oncogene 2003;22:5545-5553. 66. Pedersen MW, Meltorn M, Damstrup L, et al. The type III epidermal growth factor receptor mutation. Biological significance and potential target for anti-cancer therapy. Ann Oncol 2001;12: 745-760. 67. Sok JC, Coppelli FM, Thomas SM, et al. Mutant epidermal growth factor receptor (EGFRvIII) contributes to head and neck cancer growth and resistance to EGFR targeting Clin Cancer Res 2006;12:5064-5073. 68. Kalyankrishna S, Grandis JR. Epidermal growth factor receptor biology in head and neck cancer. J Clin Oncol 2006;24:26662672. 69. Zhan M, Han ZC. Phosphatidylinositide 3-kinase/AKT in radiation responses. Histol Histopathol 2004;19:915-923. 70. Pedrero JM, Carracedo DG, Pinto CM, et al. Frequent genetic and biochemical alterations of the PI 3-K/AKT/PTEN pathway in head and neck squamous cell carcinoma. Int J Cancer 2005;114:242-248. 71. Okami K, Wu L, Riggins G, et al. Analysis of PTEN/MMAC1 ­alterations in aerodigestive tract tumors. Cancer Res 1998;58: 509-511. 72. Motoyama A, Hynes NE, Lane HA. The efficacy of ErbB ­receptortargeted anticancer therapeutics is influenced by the availability of epidermal growth factor-related peptides. Cancer Res 2002;62:3151-3158. 73. Hurbin A, Dubrez L, Coll JL, et al. Inhibition of apoptosis by amphiregulin via an insulin-like growth factor-1 receptor-­dependent pathway in non–small cell lung cancer cell lines. J Biol Chem 2002;277:49127-49133. 74. Chakravarti A, Loeffler JS, Dyson NJ. Insulin-like growth factor receptor I mediates resistance to anti-epidermal growth factor receptor therapy in primary human glioblastoma cells through continued activation of phosphoinositide 3-kinase signaling. Cancer Res 2002;62:200-207. 75. Folkman J. Tumor angiogenesis: therapeutic implications. N Engl J Med 1971;285:1182-1186. 76. O’Reilly MS. Antiangiogenesis: basic principles. In Rosenberg SA (ed): Principles and Practice of the Biologic Therapy of Cancer, 3rd ed. Philadelphia: Lippincott Williams & Wilkins, 2000, pp 827-843. 77. Folkman J. Angiogenesis in cancer, vascular, rheumatoid and other diseases. Nat Med 1995;1:27-31. 78. Fidler IJ, Langley RR, Kerbel RS, Ellis LM. Angiogenesis. In ­DeVita VT, Hellman S, Rosenberg SA (eds): Cancer Principles and Practice of Oncology, 7th ed. Philadelphia: Lippincott Williams & Wilkins, 2005. 79. Dark GG, Hill SA, Prise VE, et al. Combretastatin A-4, an agent that displays potent and selective toxicity toward tumor vasculature. Cancer Res 1997;57:1829-1834. 80. Jain RK. Normalizing tumor vasculature with anti-angiogenic therapy: a new paradigm for combination therapy. Nat Med 2001;7:987-989. 81. Konerding MA, Malkusch W, Klapthor B, et al. Evidence for characteristic vascular patterns in solid tumours: quantitative studies using corrosion casts. Br J Cancer 1999;80:724-732. 82. Dvorak HF. Tumors: wounds that do not heal. N Engl J Med 1986;315:1650-1659. 83. Hamada J, Cavanaugh PG, Lotan O, et al. Separable growth and migration factors for large-cell lymphoma cells secreted by microvascular endothelial cells derived from target organs for metastasis. Br J Cancer 1992;66:349-354. 84. Nicosia RF, Tchao R, Leighton J. Interactions between newly formed endothelial channels and carcinoma cells in plasma clot culture. Clin Exp Metastasis 1986;4:91-104.

  85. Rak J, Filmus J, Kerbel RS. Reciprocal paracrine interactions between tumour cells and endothelial cells: the ‘angiogenesis progression’ hypothesis. Eur J Cancer 1996;32A:2438-2450.   86. Folkman J. Tumor angiogenesis and tissue factor. Nat Med 1996;2:167-168.   87. Huang X, Molema G, King S, et al. Tumor infarction in mice by antibody-directed targeting of tissue factor to tumor vasculature. Science 1997;275:547-550.   88. Thorpe PE. Vascular targeting agents as cancer therapeutics. Clin Cancer Res 2004;10:415-427.   89. Arap W, Pasqualini R, Ruoslahti E. Cancer treatment by targeted drug delivery to tumor vasculature in a mouse model. Science 1998;279:377-380.   90. Siemann DW, Rojiani AM. Enhancement of radiation therapy by the novel vascular targeting agent ZD6126. Int J Radiat Oncol Biol Phys 2002;53:164-171.   91. Raben D, Bianco C, Damiano V, et al. Antitumor activity of ZD6126, a novel vascular-targeting agent, is enhanced when combined with ZD1839, an epidermal growth factor receptor tyrosine kinase inhibitor, and potentiates the effects of radiation in a human non–small cell lung cancer xenograft model. Mol Cancer Ther 2004;3:977-983.   92. Wachsberger PR, Burd R, Marero N, et al. Effect of the tumor vascular-damaging agent, ZD6126, on the radioresponse of U87 glioblastoma. Clin Cancer Res 2005;11:835-842.   93. Garcia-Barros M, Paris F, Cordon-Cardo C, et al. Tumor response to radiotherapy regulated by endothelial cell apoptosis. Science 2003;300:1155-1159.   94. Gorski DH, Beckett MA, Jaskowiak NT, et al. Blockade of the vascular endothelial growth factor stress response increases the antitumor effects of ionizing radiation. Cancer Res 1999;59:3374-3378.   95. Kaplan H, Murphy E. The effect of local roentgen irradiation on the biological behavior of a transplantable mouse carcinoma. Increased frequency of pulmonary metastasis. J Natl Cancer Inst 1949;9:407-413.   96. Suit H, Sedlacek R, Gillette E. Examination for a correlation between probabilities of development of distant metastasis and local recurrence. Radiology 1970;95:189-194.   97. Hast J, Schiffer IB, Neugebauer B, et al. Angiogenesis and ­fibroblast proliferation precede formation of recurrent tumors after radiation therapy in nude mice. Anticancer Res 2002;22: 677-688.   98. Camphausen K, Moses MA, Beecken WD, et al. Radiation therapy to a primary tumor accelerates metastatic growth in mice. Cancer Res 2001;61:2207-2211.   99. Milas L, Hirata H, Hunter N, et al. Effect of radiation-induced injury of tumor bed stroma on metastatic spread of murine sarcomas and carcinomas. Cancer Res 1988;48:2116-2210. 100. Milas L. Tumor bed effect in murine tumors: relationship to tumor take and tumor macrophage content. Radiat Res 1990;123:232-236. 101. Hartford AC, Gohongi T, Fukumura D, et al. Irradiation of a primary tumor, unlike surgical removal, enhances angiogenesis suppression at a distal site: potential role of host-tumor interaction. Cancer Res 2000;60:2128-2131. 102. Folkman J. Clinical applications of angiogenesis research. N Engl J Med 1995;333:1757-1763. 103. Libutti SK, Pluda JM. Antiangiogenesis: clinical applications. In Rosenberg SA (ed): Principles and Practice of the Biologic Therapy of Cancer, 3rd ed. Philadelphia: Lippincott Williams & Wilkins, 2000, pp 844-864. 104. O’Reilly MS, Holmgren L, Chen C, et al. Angiostatin induces and sustains dormancy of human primary tumors in mice. Nat Med 1996;2:689-692. 105. Rusk A, McKeegan E, Haviv F, et al. Preclinical evaluation of antiangiogenic thrombospondin-1 peptide mimetics, ABT-526 and ABT-510, in companion dogs with naturally occurring cancers. Clin Cancer Res 2006;12:7444-7455. 106. Ezekowitz RAB, Mulliken JB, Folkman J. Interferon alfa-2a therapy for life-threatening hemangiomas of infancy. N Engl J Med 1992;326:1456-1463.

5  Principles of Combining Radiation Therapy and Molecular Therapeutics 107. Kaban LB, Mulliken JB, Ezekowitz RA, et al. Antiangiogenic therapy of a recurrent giant cell tumor of the mandible with interferon alfa-2a. Pediatrics 1999;103:1145-1149. 108. Boehm T, Folkman J, Browder T, et al. Antiangiogenic therapy of experimental cancer does not induce acquired drug resistance. Nature 1997;390:404-407. 109. Alon T, Hemo I, Itin A, et al. VEGF acts as a survival factor for newly formed retinal vessels and has implications for retinopathy of prematurity. Nat Med 1995;1:1024-1028. 110. Ferrara N, Gerber HP, LeCouter J. The biology of VEGF and its receptors. Nat Med 2003;9:669-676. 111. Levine RJ, Maynard SE, Qian C, et al. Circulating angiogenic factors and the risk of preeclampsia. N Engl J Med 2004;350: 672-683. 112. Carmeliet P, Storkebaum E. Vascular and neuronal effects of VEGF in the nervous system: implications for neurological disorders. Semin Cell Dev Biol 2002;13:39-53. 113. Gerber HP, Malik AK, Solar GP, et al. VEGF regulates haematopoietic stem cell survival by an internal autocrine loop mechanism. Nature 2002;417:954-958. 114. LeCouter J, Moritz DR, Li B, et al. Angiogenesis-­independent endothelial protection of liver: role of VEGFR-1. Science 2003;299:890-893. 115. Ohm JE, Gabrilovich DI, Sempowski GD, et al. VEGF inhibits T-cell development and may contribute to tumor-induced immune suppression. Blood 2003;101:4878-4886. 116. Kabbinavar F, Hurwitz HI, Fehrenbacher L, et al. Phase II, randomized trial comparing bevacizumab plus fluorouracil (FU)/ leucovorin (LV) and FU/LV alone in patients with metastatic colorectal cancer. J Clin Oncol 2003;21:60-65. 117. Yang JC, Haworth L, Sherry RM, et al. A randomized trial of bevacizumab, an anti-vascular endothelial growth factor antibody, for metastatic renal cancer. N Engl J Med 2003;349: 427-434. 118. Fan F, Wey JS, McCarty MF, et al. Expression and function of vascular endothelial growth factor receptor-1 on human colorectal cancer cells. Oncogene 2005;24:2647-2653. 119. Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004;350:2335-2342. 120. Sandler A, Gray R, Perry MC, et al. Paclitaxel-carboplatin alone or with bevacizumab for non–small-cell lung cancer. N Engl J Med 2006;355:2542-2550. 121. Kuenen BC, Rosen L, Smit EF, et al. Dose-finding and pharmacokinetic study of cisplatin, gemcitabine, and SU5416 in patients with solid tumors. J Clin Oncol 2002;20:1657-1667. 122. Miller KD, Chap LI, Holmes FA, et al. Randomized phase III trial of capecitabine compared with bevacizumab plus capecitabine in patients with previously treated metastatic breast cancer. J Clin Oncol 2005;23:792-799. 123. Koehne C, Bajetta E, Lin E, et al. Results of an interim analysis of a multinational randomized, double-blind, phase III study in patients (pts) with previously treated metastatic colorectal cancer (mCRC) receiving FOLFOX4 and PTK787/ZK 222584 (PTK/ZK) or placebo (CONFIRM 2) [abstract]. J Clin Oncol 2006;24;(18S):3508. 124. Major P, Trarbach T, Lenz H, et al. A meta-analysis of two randomized, double-blind, placebo-controlled, phase III studies in patients (pts) with metastatic colorectal cancer (mCRC) receiving FOLFOX4 and PTK/ZK to determine clinical benefit on progression-free survival (PFS) in high LDH pts [abstract]. J Clin Oncol (Meeting Abstracts) 2006;24:3529. 125. Relf M, LeJeune S, Scott PA, et al. Expression of the angiogenic factors vascular endothelial cell growth factor, acidic and basic fibroblast growth factor, tumor growth factor beta-1, plateletderived endothelial cell growth factor, placenta growth factor, and pleiotrophin in human primary breast cancer and its relation to angiogenesis. Cancer Res 1997;57:963-969. 126. Solorzano CC, Baker CH, Tsan R, et al. Optimization for the blockade of epidermal growth factor receptor signaling for therapy of human pancreatic carcinoma. Clin Cancer Res 2001;7:2563-2572.

135

127. Baker CH, McCarty MF, Tsan R, et al. Phenotypic diversity of organ-specific endothelial cells [abstract]. Proc Am Assoc Cancer Res 2001;42:2186. 128. Brogi E, Wu T, Namiki A, et al. Indirect angiogenic cytokines upregulate VEGF and bFGF gene expression in vascular smooth muscle cells, whereas hypoxia upregulates VEGF expression only. Circulation 1994;90:649-652. 129. Uehara H, Kim SJ, Karashima T, et al. Blockade of the PDGF-R signaling by STI571 inhibits angiogenesis and growth of human prostate cancer cells in the bone of nude mice [abstract]. Proc Am Assoc Cancer Res 2001;42:2192. 130. Neeltje S, Johan WRN, Hans G. Small molecule tyrosine kinase inhibitors in the treatment of solid tumors: an update of recent developments. Ann Surg Oncol 2007;14:942-953. 131. Herbst RS, Heymach JV, O’Reilly MS, et al. Vandetanib (ZD6474): an orally available receptor tyrosine kinase inhibitor that selectively targets pathways critical for tumor growth and angiogenesis. Exp Opin Investig Drugs 2007;16:239-249. 132. Motzer RJ, Hutson TE, Tomczak P, et al. Sunitinib versus interferon alfa in metastatic renal-cell carcinoma. N Engl J Med 2007;356:115-124. 133. Escudier B, Eisen T, Stadler WM, et al. Sorafenib in advanced clear-cell renal-cell carcinoma. N Engl J Med 2007;356: 125-134. 134. Teicher B, Dupuis NP, Kusumoto T, et al. Anti-angiogenic agents can increase tumor oxygenation and response to radiation therapy. Radiat Oncol Investig 1995;2:269-276. 135. Hansen-Algenstaedt N, Stoll BR, Padera TP, et al. Tumor oxygenation in hormone-dependent tumors during vascular endothelial growth factor receptor-2 blockade, hormone ablation, and chemotherapy. Cancer Res 2000;60:4556-4560. 136. Murata R, Nishimura Y, Hiraoka M. An antiangiogenic agent (TNP-470) inhibited reoxygenation during fractionated radiotherapy of murine mammary carcinoma. Int J Radiat Oncol Biol Phys 1997;37:1107-1113. 137. Lund EL, Bastholm L, Kristjansen PE. Therapeutic synergy of TNP-470 and ionizing radiation: effects on tumor growth, vessel morphology, and angiogenesis in human glioblastoma multiforme xenografts. Clin Cancer Res 2000;6:971-978. 138. Mauceri HJ, Hanna NN, Beckett MA, et al. Combined effects of angiostatin and ionizing radiation in antitumour therapy. Nature 1998;394:287-291. 139. Milas L, Mason K, Hunter N, et al. In vivo enhancement of tumor radioresponse by C225 antiepidermal growth factor receptor antibody. Clin Cancer Res 2000;6:701-708. 140. Geng L, Donnelly E, McMahon G, et al. Inhibition of vascular endothelial growth factor receptor signaling leads to reversal of tumor resistance to radiotherapy. Cancer Res 2001;61:24132419. 141. Cao C, Albert JM, Geng L, et al. Vascular endothelial growth factor tyrosine kinase inhibitor AZD2171 and fractionated radiotherapy in mouse models of lung cancer. Cancer Res 2006;66:11409-11415. 142. Gorski DH, Mauceri HJ, Salloum RM, et al. Potentiation of the antitumor effect of ionizing radiation by brief concomitant exposures to angiostatin. Cancer Res 1998;58:5686-5689. 143. Zips D, Krause M, Hessel F, et al. Experimental study on different combination schedules of VEGF-receptor inhibitor PTK787/ZK222584 and fractionated irradiation. Anticancer Res 2003;23:3869-3876. 144. Williams KJ, Telfer BA, Brave S, et al. ZD6474, a potent inhibitor of vascular endothelial growth factor signaling, combined with radiotherapy: schedule-dependent enhancement of antitumor activity. Clin Cancer Res 2004;10:8587-8593. 145. Shibuya K, Komaki R, Shintani T, et al. Targeted therapy against VEGFR and EGFR with ZD6474 enhances the therapeutic efficacy of irradiation in an orthotopic model of human non–small-cell lung cancer. Int J Radiat Oncol Biol Phys 2007;69:1534-1543. 146. Winkler F, Kozin SV, Tong RT, et al. Kinetics of vascular normalization by VEGFR2 blockade governs brain tumor response to radiation: role of oxygenation, angiopoietin-1, and matrix metalloproteinases. Cancer Cell 2004;6:553-563.

136

PART 1  Principles

147. Johnson DH, Fehrenbacher L, Novotny WF, et al. Randomized phase II trial comparing bevacizumab plus carboplatin and paclitaxel with carboplatin and paclitaxel alone in previously untreated locally advanced or metastatic non–small-cell lung ­cancer. J Clin Oncol 2004;22:2184-2191. 148. Gordon MS, Margolin K, Talpaz M, et al. Phase I safety and pharmacokinetic study of recombinant human anti-vascular endothelial growth factor in patients with advanced cancer. J Clin Oncol 2001;19:843-850. 149. Willett CG, Boucher Y, di Tomaso E, et al. Direct evidence that the VEGF-specific antibody bevacizumab has antivascular effects in human rectal cancer. Nat Med 2004;10:145-147. 150. Willett CG, Boucher Y, Duda DG, et al. Surrogate markers for antiangiogenic therapy and dose-limiting toxicities for bevacizumab with radiation and chemotherapy: continued experience of a phase I trial in rectal cancer patients. J Clin Oncol 2005;23:8136-8139. 151. Crane CH, Varadhachary G, Wolff RA, et al. The argument for pre-operative chemoradiation for localized, radiographically resectable pancreatic cancer. Best Pract Res Clin Gastroenterol 2006;20:365-382. 152. Milas L. Cyclooxygenase-2 (COX-2) enzyme inhibitors as potential enhancers of tumor radioresponse. Semin Radiat Oncol 2001;11:290-299. 153. Milas L, Hanson WR. Eicosanoids and radiation [review]. Eur J Cancer 1995;31A:1580-1585. 154. Liao Z, Mason KA, Milas L. Cyclooxygenase-2 and its inhibition in cancer: is there a role? Drugs 2007;67:821-845. 155. Oshima M, Dinchuk JE, Kargman SL, et al. Suppression of intestinal polyposis in Apc716 knockout mice by inhibition of cyclooxygenase 2 (COX-2). Cell 1996;87:803-809. 156. Liu CH, Chang SH, Narko K, et al. Overexpression of ­cyclooxygenase-2 is sufficient to induce tumorigenesis in transgenic mice. J Biol Chem 2001;276:18563-18569. 157. Harris RE, Alshafie GA, Abou-Issa H, et al. Chemoprevention of breast cancer in rats by celecoxib, a cyclooxygenase 2 inhibitor. Cancer Res 2000;60:2101-2103. 158. Jacoby RF, Siebert K, Cole CE, et al. The cyclooxygenase-2 inhibitor celecoxib is a potent preventive and therapeutic agent in the min mouse model of adenomatous polyposis. Cancer Res 2000;60:5040-5044. 159. Khuri FR, Wu H, Lee JJ, et al. Cyclooxygenase-2 overexpression is a marker of poor prognosis in stage I non–small cell lung cancer. Clin Cancer Res 2001;7:861-867. 160. Sheehan KM, Sheahan K, O’Donoghue DP, et al. The relationship between cyclooxygenase-2 expression and colorectal cancer. JAMA 1999;282:1254-1257. 161. Gaffney DK, Holden J, Davis M, et al. Elevated cyclooxygenase-2 expression correlates with diminished survival in carcinoma of the cervix treated with radiotherapy. Int J Radiat Oncol Biol Phys 2001;49:1213-1217. 162. Arber N, Eagle CJ, Spicak J, et al. Celecoxib for the prevention of colorectal adenomatous polyps. N Engl J Med 2006;355: 885-895. 163. Stoner GD, Wang LS, Chen T. Chemoprevention of esophageal squamous cell carcinoma. Toxicol Appl Pharmacol 2007;224:337-349. 164. Sjödahl R. Nonsteroidal anti-inflammatory drugs and the gastrointestinal tract. Extent, mode, and dose dependence of anticancer effects [review]. Am J Med 2001;110:66S-69S. 165. Sandler AB, Dubinett SM. COX-2 inhibition and lung cancer. Semin Oncol 2004;31;(Suppl 7):45-52. 166. Steinbach G, Lynch PM, Phillips RK, et al. The effect of celecoxib, a cyclooxygenase-2 inhibitor, in familial adenomatous polyposis. N Engl J Med 2000;342:1946-1952. 167. Thun MJ, Namboodiri MM, Calle EE, et al. Aspirin use and risk of fatal cancer. Cancer Res 1993;53:1322-1327. 168. Koki A, Leahy KM, Masferrer JL. Potential utility of COX-2 inhibitors in chemoprevention and chemotherapy. Exp Opin Investig Drugs 1999;8:1623-1638. 169. Milas L, Fujii T, Hunter N, et al. Enhancement of tumor radioresponse in vivo by gemcitabine. Cancer Res 1999;59:107-114.

170. Milas L, Kishi K, Hunter N, et al. Enhancement of tumor ­response to gamma-radiation by an inhibitor of cyclooxygenase2 enzyme. J Natl Cancer Inst 1999;91:1501-1504. 171. Petersen C, Petersen S, Milas L, et al. Human glioma cell radiosensitization by a selective COX-2 inhibitor. Clin Cancer Res 2000;6:2513-2520. 172. Pyo H, Choy H, Amorino GP, et al. A selective cyclooxygenase2 inhibitor, NS-398, enhances the effect of radiation in vitro and in vivo preferentially on the cells that express cyclooxygenase-2. Clin Cancer Res 2001;7:2998-3005. 173. Raju U, Aroga H, Dittmann K. Inhibition of DNA repair as a mechanism of enhanced radioresponse of head and neck carcinoma cells by a selective cyclooxygenase-2 inhibitor, celecoxib. Int J Radiat Oncol Biol Phys 2005;63:520-528. 174. Raju U, Nakata E, Yang P, et al. In vitro enhancement of tumor cell radiosensitivity by a selective inhibitor of cyclooxygenase2 enzyme: mechanistic considerations. Int J Radiat Oncol Biol Phys 2002;54:886-894. 175. Liao Z, Komaki R, Milas L, et al. A phase I clinical trial of ­thoracic radiotherapy and concurrent celecoxib for patients with unfa­ vorable performance status inoperable/unresectable non–small cell lung cancer. Clin Cancer Res 2005;11:3342-3348. 176. Bresalier RS, Sandler RS, Quan H, et al. Cardiovascular events associated with rofecoxib in a colorectal adenoma chemoprevention trial. N Engl J Med 2005;352:1092-1102. 177. MacLachlan TK, Sang N, Giordano A. Cyclins, cyclin-dependent kinases and cdk inhibitors: implications in cell cycle control and cancer. Crit Rev Eukaryot Gene Expr 1995;5:127-156. 178. Senderowicz AM. Targeting cell cycle and apoptosis for the treatment of human malignancies. Curr Opin Cell Biol 2004;16: 670-678. 179. Zafonte BT, Hulit J, Amanatullah DF, et al. Cell-cycle dysregulation in breast cancer therapies targeting the cell cycle. Front Biosci 2000;5:D938-D961. 180. Arber N, Lightdale C, Rotterdam H, et al. Increased expression of the cyclin D1 gene in Barrett’s esophagus. Cancer Epidemiol Biomarkers Prev 1996;5:457-459. 181. Bani-Hani K, Martin IG, Hardie LJ, et al. Prospective study of cyclin D1 overexpression in Barrett’s esophagus: association with increased risk of adenocarcinoma. J Natl Cancer Inst 2000;92:1316-1321. 182. Grant WI, Woo SY. Clinical and financial issues for intensitymodulated radiation therapy delivery. Semin Radiat Oncol 1999;9:99-107. 183. Senderowicz AM. Small-molecule cyclin-dependent kinase modulators. Oncogene 2003;22:6609-6620. 184. Tsihlias J, Kapusta L, Slingerland J. The prognostic significance of altered cyclin-dependent kinase inhibitors in human cancer. Annu Rev Med 1999;50:401-423. 185. Keyomarsi K, Tucker SL, Buchholz TA, et al. Cyclin E and survival in patients with breast cancer. N Engl J Med 2002;347:15661575. 186. Arber N, Doki Y, Han EKH, et al. Antisense to cyclin D1 inhibits the growth and tumorigenicity of human colon cancer cells. Cancer Res 1997;57:1569-1574. 187. Kornmann M, Arber N, Korc M. Inhibition of basal and mitogenstimulated pancreatic cancer cell growth by cyclin D1 antisense is associated with loss of tumorigenicity and potentiation of cytotoxicity to cisplatinum. J Clin Invest 1998;101:344-352. 188. Sauter ER, Nesbit M, Litwin S, et al. Antisense cyclin D1 induces apoptosis and tumor shrinkage in human squamous carcinomas. Cancer Res 1999;59:4876-4881. 189. Zhou P, Jiang W, Zhang YJ, et al. Antisense to cyclin D1 inhibits growth and reverses the transformed phenotype of human esophageal cancer cells. Oncogene 1995;11:571-580. 190. Sedlacek HH. Mechanisms of action of flavopiridol. Crit Rev Oncol Hematol 2001;38:139-170. 191. Raju U, Nakata E, Mason KA, et al. Flavopiridol, a cyclin­dependent kinase inhibitor, enhances radiosensitivity of ovarian carcinoma cells. Cancer Res 2003;63:3263-3267. 192. Mason KA, Hunter NR, Raju U, et al. Flavopiridol increases therapeutic ratio of radiotherapy by preferentially enhancing ­tumor

5  Principles of Combining Radiation Therapy and Molecular Therapeutics radioresponse. Int J Radiat Oncol Biol Phys 2004;59:11811189. 193. Senderowicz AM, Sausville EA. Preclinical and clinical development of cyclin-dependent kinase modulators. J Natl Cancer Inst 2000;92:376-387. 194. Ariga H, Raju U, Liao Z, et al. Flavopiridol enhances radiosensitivity of human head and neck, and esophageal carcinoma cells [abstract 1357]. Proc Am Assoc Cancer Res 2004;45:311. 195. Senderowicz AM, Headlee D, Stinson SF, et al. Phase I trial of continuous infusion flavopiridol, a novel cyclin-dependent kinase inhibitor, in patients with refractory neoplasms. J Clin Oncol 1998;16:2986-2999. 196. Shapiro GI. Preclinical and clinical development of the ­cyclindependent kinase inhibitor flavopiridol. Clin Cancer Res 2004;10:4270S-4275. 197. Gupta AK, Bakanauskas VJ, Cerniglia GJ, et al. The Ras radiation resistance pathway. Cancer Res 2001;61:4278-4282. 198. Hahn SM, Bernhard EJ, Regine W, et al. A phase I trial of the farnesyltransferase inhibitor L-778,123 and radiotherapy for locally advanced lung and head and neck cancer. Clin Cancer Res 2002;8:1065-1072. 199. Martin NE, Brunner TB, Kiel KD, et al. A phase I trial of the dual farnesyltransferase and geranylgeranyltransferase inhibitor L-778,123 and radiotherapy for locally advanced pancreatic cancer. Clin Cancer Res 2004;10:5447-5454. 200. Willett CG, Safran H, Abrams RA, et al. Clinical research in pancreatic cancer: the Radiation Therapy Oncology Group trials. Int J Radiat Oncol Biol Phys 2003;56:31-37. 201. Moyal EC, Laprie A, Delannes M, et al. Phase I trial of tipifarnib (R115777) concurrent with radiotherapy in patients with glioblastoma multiforme. Int J Radiat Oncol Biol Phys 2007;68:13961401. 202. Munshi A, Meyn R. Combination of gene therapy with radiation. In Hunt K, Vorburger S, Swisher S (eds): Cancer Drug Discovery and Development: Gene Therapy for Cancer. Totowa, NJ: Humana Press, 2007, pp 243-256. 203. Collis SJ, Khater KDeWeese TL. Novel therapeutic strategies in prostate cancer management using gene therapy in combination with radiation therapy. World J Urol 2003;21:275-289. 204. Teh BS, Ayala G, Aguilar L, et al. Phase I-II trial evaluating combined intensity-modulated radiotherapy and in situ gene ­therapy with or without hormonal therapy in treatment of prostate ­cancer-interim report on PSA response and biopsy data. Int J Radiat Oncol Biol Phys 2004;58:1520-1529. 205. Freytag SO, Paielli D, Wing M, et al. Efficacy and toxicity of replication-competent adenovirus-mediated double suicide gene therapy in combination with radiation therapy in an orthotopic mouse prostate cancer model. Int J Radiat Oncol Biol Phys 2002;54:873-885. 206. Freytag SO, Stricker H, Pegg J, et al. Phase I study of ­replicationcompetent adenovirus-mediated double-suicide gene therapy in combination with conventional-dose three-dimensional conformal radiation therapy for the treatment of newly diagnosed, intermediate- to high-risk prostate cancer. Cancer Res 2003;63:7497-7506. 207. Yan B, Wang H, Rabbani ZN, et al. Tumor necrosis factor-alpha is a potent endogenous mutagen that promotes cellular transformation. Cancer Res 2006;66:11565-11570. 208. Weichselbaum RR, Kufe DW, Hellman S, et al. Radiation­induced tumour necrosis factor-alpha expression: clinical application of transcriptional and physical targeting of gene therapy. Lancet Oncol 2002;3:665-671. 209. Mezhir JJ, Smith KD, Posner MC, et al. Ionizing radiation: a genetic switch for cancer therapy. Cancer Gene Ther 2006;13:1-6. 210. Senzer N, Mani S, Rosemurgy A, et al. TNFerade biologic, an adenovector with a radiation-inducible promoter, carrying the human tumor necrosis factor alpha gene: a phase I study in patients with solid tumors. J Clin Oncol 2004;22:592-601. 211. Lang FF, Yung WK, Raju U, et al. Enhancement of radiosensitivity of wild-type p53 human glioma cells by adenovirus-­mediated delivery of the p53 gene. J Neurosurg 1998;89:125-132.

137

212. Sasaki R, Shirakawa T, Zhang ZJ, et al. Additional gene therapy with Ad5CMV-p53 enhanced the efficacy of radiotherapy in human prostate cancer cells. Int J Radiat Oncol Biol Phys 2001;51:1336-1345. 213. Swisher SG, Roth JA, Komaki R, et al. Induction of p53­regulated genes and tumor regression in lung cancer patients after intratumoral delivery of adenoviral p53 (INGN 201) and radiation therapy. Clin Cancer Res 2003;9:93-101. 214. Vecil GG, Lang FF. Clinical trials of adenoviruses in brain tumors: a review of Ad-p53 and oncolytic adenoviruses. J Neurooncol 2003;65:237-246. 215. Rogulski KR, Freytag SO, Zhang K, et al. In vivo antitumor activity of ONYX-015 is influenced by p53 status and is augmented by radiotherapy. Cancer Res 2000;60:1193-1196. 216. Geoerger B, Grill J, Opolon P, et al. Potentiation of radiation therapy by the oncolytic adenovirus dl1520 (ONYX-015) in human malignant glioma xenografts. Br J Cancer 2003;89:577-584. 217. Lamfers ML, Grill J, Dirven CM, et al. Potential of the conditionally replicative adenovirus Ad5-Delta24RGD in the ­ treatment of malignant gliomas and its enhanced effect with radiotherapy. Cancer Res 2002;62:5736-5742. 218. Lamfers ML, Idema S, Bosscher L, et al. Differential effects of combined Ad5-{Delta}24RGD and radiation therapy in in vitro versus in vivo models of malignant glioma. Clin Cancer Res 2007;13:7451-7458. 219. Vidal L, Blagden S, Attard G, et al. Making sense of antisense. Eur J Cancer 2005;41:2812-2818. 220. Dritschilo A, Huang CH, Rudin CM, et al. Phase I study of ­liposome-encapsulated c-raf antisense oligodeoxyribonucleotide infusion in combination with radiation therapy in patients with advanced malignancies. Clin Cancer Res 2006;12:1251-1259. 221. Tebes SJ, Kruk PA. The genesis of RNA interference, its potential clinical applications, and implications in gynecologic cancer. Gynecol Oncol 2005;99:736-741. 222. Dunn GP, Bruce AT, Ikeda H, et al. Cancer immunoediting: from immunosurveillance to tumor escape. Nat Immunol 2002;3:991-998. 223. Milas L, Scott MT. Antitumor activity of Corynebacterium parvum. Adv Cancer Res 1978;26:257-306. 224. Mihich E, Fefer A (eds): Biological Response Modifiers: Subcommittee Report. Monograph no. 63. Washington, DC: National Cancer Institute, 1983. 225. Tokunaga T, Yamamoto T, Yamamoto S. How BCG led to the discovery of immunostimulatory DNA. Jpn J Infect Dis 1999;52:1-11. 226. Krieg AM, Yi A, Matson S, et al. CpG motifs in bacterial DNA trigger direct B-cell activation. Nature 1995;374:546-549. 227. Sparwasser T, Vabulas RM, Villmow B, et al. Bacterial CpGDNA activates dendritic cells in vivo: T helper cell-independent cytotoxic T cell responses to soluble proteins. Eur J Immunol 2000;30:3591-3597. 228. Uhlmann E, Vollmer J. Recent advances in the development of immunostimulatory oligonucleotides. Curr Opin Drug Discov Devel 2003;6:204-217. 229. Baines J, Celis E. Immune-mediated tumor regression induced by CpG-containing oligodeoxynucleotides. Clin Cancer Res 2003;9:2693-2700. 230. Krieg AM. Antitumor applications of stimulating toll-like receptor 9 with CpG oligodeoxynucleotides. Curr Oncol Rep 2004;6:88-95. 231. Mason KA, Ariga H, Neal R, et al. Targeting toll-like receptor 9 with CpG oligodeoxynucleotides enhances tumor response to fractionated radiotherapy. Clin Cancer Res 2005;11:361-369. 232. Milas L, Mason KA, Ariga H, et al. CpG oligodeoxynucleotide enhances tumor response to radiation. Cancer Res 2004;64: 5074-5077. 233. Koski GK, Czerniecki BJ. Combining innate immunity with radiation therapy for cancer treatment.Clin Cancer Res 2005;11:7-11. 234. Sutherland RM. Radiation-enhanced immune response to cancer: Workshop, Anaheim, CA, April 17, 2005. Int J Radiat Oncol Biol Phys 2006;64:3-5.

The Skin

6

B. Ashleigh Guadagnolo, K. Kian Ang, and Matthew T. Ballo ANATOMY OF THE SKIN RESPONSE OF THE SKIN TO IRRADIATION Acute Reactions Chronic Reactions ETIOLOGY AND EPIDEMIOLOGY OF SKIN CANCER PREVENTION AND SCREENING CLINICAL MANIFESTATION, PATHOBIOLOGY, AND PATHWAYS OF SPREAD Basal Cell Carcinoma Squamous Cell Carcinoma Melanoma Merkel Cell Carcinoma Adnexal Carcinomas PATIENT EVALUATION AND STAGING Carcinoma Melanoma

PRIMARY THERAPY AND RESULTS Carcinoma Melanoma Merkel Cell Carcinoma Adnexal Carcinomas TREATMENT OF ADVANCED DISEASE AND PALLIATIVE THERAPY PRINCIPLES OF RADIATION THERAPY Carcinoma Melanoma Merkel Cell and Adnexal Carcinomas Patient Care during and after Radiation Therapy

Basal and squamous cell carcinomas of the skin are the most common types of cancer in the United States, with an estimated annual incidence of about 1 million cases.1 These carcinomas most often affect people with fair complexions or outdoor occupations, or both. Skin carcinomas may be disfiguring in certain locations but are rarely lethal. Although the exact figures are difficult to ascertain, the number of people who will die of cutaneous carcinomas in a given year is estimated to be between 2500 and 5000. Cutaneous melanoma is largely a disease of individuals with fair complexions who have a tendency to sunburn even after relatively brief exposure to sunlight.3,4 Approximately 59,940 cases of cutaneous melanoma were diagnosed in the United States in 2007.1 Despite progress in early detection, the incidence rates of malignant melanoma more than ­doubled between 1980 and 1990 and continue to increase rapidly.5,6 At present, 1 in 49 men and 1 in 73 women can be expected to develop malignant melanoma during their lifetime.1 Increasing exposure to sunlight during ­recreational activities and additional ultraviolet (UV) radiation reaching  Earth’s surface are considered strong contributing factors to the epidemic. Fortunately, because of better public ­awareness of the disease, most new cases of melanoma are diagnosed at early stages. Approximately 8110 people in the United States died of cutaneous melanoma in 2007.1 Less common types of skin cancer include Merkel cell carcinoma and adnexal carcinomas (i.e., sebaceous gland, eccrine, and apocrine carcinomas). Merkel cell carcinoma, also known as primary cutaneous neuroendocrine carcinoma or trabecular carcinoma of the skin, is a rare cancer originally described by Toker in 1972.7 Sebaceous gland carcinoma accounts for less than 1% of all cases of skin ­cancer.8 Eccrine (sweat gland) carcinoma is rare, ­accounting

for less than 0.01% of malignant epithelial neoplasms of the skin.9 Apocrine carcinoma is equally rare. Etiologic and epidemiologic data on Merkel cell and adnexal carcinomas are scarce. Likewise, information on prevention and early detection of these lesions is scanty.

Anatomy of the Skin The skin is composed of three layers: epidermis, dermis, and subcutaneous tissue. The epidermis is made up of stratified squamous epithelium that in most regions is only 0.05 to 0.15 mm thick. Cells in the basal layer of the epidermis regularly undergo mitotic division to replace epidermal cells lost from the surface. A thin basement membrane separates the epidermis from the dermis. The dermis, which is 1 to 2 mm thick, is composed of lymphatics, nerves, blood vessels, and sweat and sebaceous glands in a connective tissue stroma. The portion of the dermis in contact with the epidermis is referred to as the papillary layer. The deeper portion of the dermis, which contains bundles of collagenous and elastic fibers, is called the reticular layer. Subcutaneous tissue lies underneath the dermis and supports the lymphatics, blood vessels, and nerves in loose connective tissue containing variable amounts of fat.

Response of the Skin to Irradiation The extent of acute and chronic reactions of the skin to ­irradiation depends on several variables, including radiation dose and energy, number of days over which the dose is delivered, field size, and anatomic location.

141

142

PART 2  Skin

Acute Reactions Mitotic inhibition appears 30 minutes after a single dose of 4 to 5 Gy and is followed by a decrease in the proliferation of the germinal layer of the endothelium. The ­pathologic correlate of the erythema that begins to appear in the ­radiation therapy field within a few hours after irradiation is capillary dilatation and increased capillary ­permeability.10 Higher-energy radiation requires a larger total dose than lower-energy radiation to produce a given degree of  erythema.11 If radiation is given in several fractions (as ­opposed to a single fraction), a higher total dose is required to cause erythema because of repair of injury between fractions. As the radiation therapy continues, swelling and proliferation of capillary endothelial cells appear. In the arterioles, the tunica intima and tunica media are ­disrupted after 2 to 3 weeks of radiation therapy. The basal cells enlarge and show nuclear pyknosis.11 Maturing keratinocytes also show swelling, nuclear pyknosis, and cytoplasmic vacuolization. The moist desquamation phase of acute radiodermatitis begins 3 to 4 weeks after treatment begins and manifests as vascular dilatation and hyperemia, edema, and extravasation of erythrocytes and leukocytes. Approximately 4 weeks after treatment begins, small superficial blisters may form, coalesce, and rupture. Similar acute changes are seen in the adnexal structures of the skin and are caused by death of the rapidly dividing germinal layer of the epithelium. In hair follicles, germinal epithelium dies after a single-fraction dose of 4 to 5 Gy, but hair loss does not occur until approximately 3 weeks after treatment because the hairs adhere to the follicles. The more active the hair growth, the more sensitive the hair follicles; hair follicles in the chin in men and the scalp in both sexes are more sensitive to radiation than are hair follicles at other locations. The epithelial changes that develop in sebaceous and eccrine glands are similar to those in hair follicles, with sebaceous glands being about as sensitive as hair follicles and eccrine glands being slightly less so; these changes are responsible for the skin dryness that develops during treatment. Epidermal necrosis and ulceration can occur as early as 2 months after the start of radiation therapy. Ulcers rarely develop when standard radiation therapy techniques are used; when soft tissue necrosis does occur, it is typically associated with trauma or infection. If irradiation produces mild necrosis of the subepidermis, the new epidermis that forms will be thinner than normal. If the necrosis is sufficient to cause disappearance of the basement membrane beneath the basal cells, the new epidermis will be highly susceptible to trauma, infection, and other stresses. With more extensive necrosis associated with greater damage to blood vessels and connective tissue, repair may be largely by secondary intention (i.e., replacement fibrosis). When severe and deep necrosis of subepidermal supporting tissues and larger blood vessels occurs, an ulcer may take months or even years to heal.

Chronic Reactions After radiation therapy, the decrease in the number of small blood vessels and the increase in the density and amount of fibrous tissue in irradiated regions gradually progress over time.10 The damaged vasculature and connective tissue continue to deteriorate, partly because of the treatment

and partly because of aging and other acquired insults. The irradiated epidermis remains chronically dry because of permanent damage to the sebaceous and eccrine glands. Microscopic atrophy of the epidermis occurs, with loss of epidermal rete (interpapillary) ridges and dermal papillae, a condition known as poikiloderma.11 The basal layer shows few mitoses. Capillaries, venules, and lymphatic vessels are reduced in number and often show telangiectases. Telangiectases result from loss of microvascular endothelial cells and damage to the basement membrane, leading to contractions of multiple capillary loops into a single distorted, sinusoidal channel. Total blood flow in the irradiated skin is reduced.

Etiology and Epidemiology of Skin Cancer The main cause of cutaneous carcinoma is exposure to sunlight.1 Other contributory factors include exposure to chemical carcinogens (e.g., arsenic, tar, anthracene, crude paraffin oil), chronic irritation or inflammation, and ionizing radiation. Several genetic syndromes are also associated with a higher incidence of cutaneous carcinoma. Individuals with xeroderma pigmentosum are prone to the development of basal cell and squamous cell carcinomas because of defective repair of UV light–induced deoxyribonucleic acid (DNA) damage.12 The basal cell nevus syndrome is a genetic form of basal cell carcinoma inherited through an autosomal dominant gene.13 People with epidermodysplasia verruciformis tend to develop squamous cell carcinomas within large verrucous plaques in the third and fourth decades of life.14 Like cutaneous carcinomas, cutaneous melanomas tend to develop in sun-exposed areas in fair-skinned people who burn rather than tan when outdoors.3 ­Blistering sunburns during childhood increase the risk of melanoma later in life.15 A patient with melanoma has a 3% to 5% lifetime risk of developing a second lesion, which is 900 times the risk of the general population.16-18 A patient with multiple dysplastic nevi or the uncommon familial form of melanoma has an even higher lifetime risk (4% to 10%) of developing melanoma.3,16 Individuals with ­occupational exposure to polyvinyl chlorides19 or chemicals used in brewing and leather processing20 are at increased risk for developing melanoma. The incidence and aggressiveness of skin cancer are greatly increased in immunosuppressed individuals (e.g., organ transplant recipients, individuals with acquired immunodeficiency syndrome [AIDS]).

Prevention and Screening Skin cancer is an important target for preventive and screening measures because most cancer is caused by exposure to UV light and the impact of early detection and treatment on mortality and quality of life, particularly in melanoma, has been well documented. In the late 1990s, a panel of the American College of Preventive Medicine (ACPM) thoroughly reviewed the relevant medical literature and issued practice policy statements with regard to skin cancer prevention.21 Recommended preventive measures include avoiding exposure to sunlight, particularly by limiting time spent outdoors between 10 am and 4 pm, and wearing protective physical barriers such as hats and

6  The Skin

c­ lothing. If exposure to sunlight cannot be limited because of occupational, cultural, or other factors, the use of sunscreen that is opaque or blocks UVA and UVB is recommended. The ACPM recommends discussing sun avoidance and sun protection measures with children and teenagers. For high-risk individuals, such as those with fair skin, multiple nevi, a family history of skin cancer, or a personal history of skin cancer, the ACPM also recommends periodic screening, including 2- to 3-minute visual inspection of the entire integument, by an adequately trained physician.

Clinical Manifestations, Pathobiology, and Pathways of Spread Basal Cell Carcinoma Clinical Presentation and Pathology Basal cell carcinoma most commonly arises in the head and neck region, where it may manifest as an asymptomatic nodule, a pruritic plaque, or a bleeding sore that waxes and wanes.22,23 Multiple synchronous lesions may be present. Approximately one third of patients treated for basal cell carcinoma develop at least one additional basal cell carcinoma.24 Careful evaluation of sun-exposed skin should be part of the follow-up examination. Each of the several subtypes of basal cell carcinoma has distinctive histologic and clinical features. The main subtypes are summarized in the following sections. Nodulo-ulcerative Subtype. Nodulo-ulcerative basal cell carcinoma, also referred to as a rodent ulcer because of its deeply infiltrating or burrowing nature, is the most commonly observed subtype. It begins as a papule. Central ulceration develops as the lesion grows. The margins of the lesion appear waxy and contain enlarged capillaries. Histologically, large islands of monomorphous basaloid cells, varying in size and shape, are embedded in a fibroblastic stroma in the dermis. The basaloid cells have large, oval, hyperchromatic nuclei, scant cytoplasm, and no intercellular bridges. The peripheral cell layer of the aggregate often demonstrates palisading, whereas the central cells are less organized. Depending on the number of melanocytes in the lesion, nodulo-ulcerative basal cell carcinomas may be blue, brown, or black. Consequently, they may be difficult to differentiate clinically from malignant melanomas.25 ­Biopsy may be required to establish the correct diagnosis. Superficial Subtype. Superficial basal cell carcinomas present as red, scaly macules with an indistinct margin and are usually located on the trunk. They enlarge into a ­crusted, erythematous patch without induration and may be ­difficult to differentiate clinically from solar keratosis, psoriasis, squamous cell carcinoma in situ, or ­extramammary  Paget’s disease. Histologic examination shows multiple foci of neoplastic cells with peripheral palisading originating from the undersurface of the epidermis.25 Morphea-like Subtype. Morphea-like basal cell carcinomas, also referred to as sclerosing basal cell carcinomas, appear as flat, indurated, ill-defined macules. The lesions generally have a smooth, shiny surface that becomes depressed as the lesions grow. Histologically, morphea-like basal cell carcinomas are characterized by the presence of small groups of narrow strands of basaloid cells embedded

143

in a dense, fibrous connective tissue. Little peripheral palisading occurs.25 Infiltrative Subtype. Infiltrative basal cell carcinomas have an opaque, yellowish appearance and blend subtly into the surrounding skin.26 Histologically, the lesions are characterized by poorly circumscribed cell aggregates near the skin surface. Neoplastic cells infiltrate the reticular ­dermis and subcutis. Basosquamous Subtype. Basosquamous carcinomas, also known as metatypical or keratotic basal cell carcinomas, occur almost exclusively on the face and are uncommon. Histologically, they exhibit features of basal and squamous cell carcinomas. Terebrant Subtype. Terebrant carcinomas manifest as small, ulcerated lesions and typically arise in the triangular region of the cheeks extending from the lower eyelids to the nasolabial folds.27 Histologic examination usually reveals extensive infiltration of deeper structures by small nests of undifferentiated basal cells. This type of lesion may invade the orbit, nose, or maxilla, resulting in significant deformity. It should be treated at an early stage whenever possible.

Biology and Pattern of Spread Between weeks 4 and 8 of embryologic development, anatomic regions grow together, and the covering epithelium subsequently migrates across junctional sites known as embryologic fusion planes. It has commonly been accepted that epithelial carcinomas spread along these embryologic fusion planes such as the nasolabial groove or the infranasal depression.28,29 However, controversy has arisen as to whether embryologic fusion planes persist in adults and influence the spread of skin tumors.30 The perineural space, the cleavage plane between a nerve and its sheath, provides a route of spread for skin cancer.31 When perineural invasion is present at diagnosis,32,33 basal cell carcinomas are often advanced, and they tend to recur. Peripheral branches of the trigeminal and facial nerves are most commonly involved and provide a route for contiguous spread to the central nervous system.34-36 In such cases, skip areas, areas where no carcinoma is identified in the pathologic specimen, may be present along the nerve between the primary tumor and the site of central nervous system involvement.35 About two thirds of patients with cutaneous carcinoma with pathologic evidence of perineural invasion have no symptoms of this invasion.35,37-39   Compression of a nerve may develop over a period of years, causing numbness, formication, burning, stinging, facial weakness, ptosis, or blurred or double vision.35-38,40 Intraneural invasion is rare.31,37,41 The biologic behavior of basal cell carcinomas varies with the histologic subtype. Nodulo-ulcerative basal cell carcinomas grow slowly over many years in people with normal immunity. They spread by peripheral and deep invasion, particularly along the perineural space, perichondrium, and periosteum. Superficial basal cell carcinomas may remain stable for a long period or enlarge gradually over a number of years. In contrast, morphea-like, infiltrative, basosquamous, and terebrant carcinomas behave more aggressively. Because the epidermis lacks lymphatics and blood vessels, basal cell carcinomas rarely spread to regional or distant sites. The overall incidence of metastasis is less than

144

PART 2  Skin

0.01%.42 Two thirds of metastases are detected in the ­regional lymphatics, and up to 20% of metastases involve the bones, lungs, or liver.43 Metastases usually arise from large, ulcerated lesions in the head and neck region that have recurred after repeated surgery and radiation therapy.44

Squamous Cell Carcinoma Clinical Presentation and Pathology Squamous cell carcinomas arise from malignant transformation of keratinocytes in the epidermis. This type of skin cancer most commonly develops as a result of damage from sunlight.45 Much less often, it arises in an area of trauma or chronic irritation, such as a thermal burn scar (i.e., Marjolin’s ulcer)46 or a draining osteomyelitic sinus.47 The three main types of squamous cell carcinoma are squamous cell carcinoma in situ (i.e., Bowen’s disease), superficial squamous cell carcinoma, and infiltrating squamous cell carcinoma. Pathologically, squamous cell carcinoma in situ exhibits a verrucous or papillated border. Atypical  keratinocytes are confined to the epidermis. Clinically, squamous cell carcinoma in situ manifests as a soft, erythematous, scaly, circumscribed patch. Superficial squamous cell carcinoma represents atypical keratinocyte proliferation confined to the upper reticular dermis, whereas infiltrating squamous cell carcinoma denotes extension into and beyond the lower reticular epidermis. Clinically, infiltrating squamous cell carcinoma manifests as a firm, ulcerated mass with an elevated, nodular border. Histologically, squamous cell carcinoma is described as well, moderately, or poorly differentiated on the basis of the magnitude of cellular polymorphism, keratinization, and mitosis. Diagnosis of the spindle cell subtype may ­require immunoperoxidase studies.

Biology and Pattern of Spread Regional and distant metastases are uncommon in squamous cell carcinoma and are rare in basal cell carcinoma. At the time of diagnosis, metastatic spread is present in approximately 2% of patients with squamous cell carcinoma.48 Factors determining the likelihood of metastasis include degree of differentiation, prior treatment, greatest dimension of the lesion, depth of invasion, and anatomic location.49,50 Poorly differentiated or recurrent squamous cell carcinomas and those less than 3 cm in diameter, less than 4 mm thick, or arising on the lips are associated with lymph node metastases at diagnosis in approximately 10% of cases.49,51-56 Reports conflict about whether perineural spread by itself is associated with an increased risk of lymph node metastases.38,55,57-59 Squamous cell carcinomas that arise in burn scars or osteomyelitic foci are associated with lymph node metastases at diagnosis in 10% to 30% of cases.54

Melanoma Clinical Presentation and Pathology Few diseases have the extraordinary metastatic potential of malignant melanoma. Beginning as a superficial pigmented skin lesion, melanoma grows horizontally within the epidermal and superficial dermal layers. This rather innocuous phase of growth, referred to as the radial growth phase, is followed after some time by a vertical growth

phase, ­ characterized by extension to deeper dermal tissues and vessels. When the disease is caught during early growth, surgery is curative. With every millimeter of vertical growth, however, the rate of nodal spread increases, as does the likelihood of distant metastatic spread—a stage of the disease that is almost uniformly fatal. Primary cutaneous melanoma may develop from preinvasive lesions such as lentigo maligna melanomas or dysplastic nevi, or it may arise de novo in normal skin. The four major subtypes are described in the following sections.60 Superficial Spreading Melanoma. The most prevalent subtype is superficial spreading melanoma, which comprises  about 70% of cases.61,62 This subtype can occur in adults of any age and affects women more often than men. ­Superficial spreading melanoma often arises in a junctional nevus, where it first appears as a deeply pigmented lesion that ­ progresses gradually to a flat, indurated macule over a ­ period of years. Patches of amelanotic areas within the lesion are evident, apparently representing focal regression. As the lesion grows, the border may become irregular. Histologically, superficial spreading melanomas are characterized by a prominent intraepidermal proliferation of malignant melanocytes. The malignant cells resemble those of Paget’s disease, and this pattern is called pagetoid melanoma.63 The malignant cells may be confined to the lower portion of the epidermis or may spread up into the granular cell layer of the epidermis, which is often hyperplastic. As the lesion enlarges, clusters of malignant cells invade the dermis and subcutaneous tissue. Nodular Melanoma. Nodular melanoma is the second most common subtype, comprising 15% to 25% of ­cases.61,62 Nodular melanoma most often arises on the trunk or in the head and neck region and most often ­occurs in middle-aged individuals. In contrast to superficial spreading melanoma, nodular melanoma affects men more often than women. It manifests as a raised or dome-shaped, blue or black lesion. In general, nodular melanoma is darker than superficial spreading melanoma. About 5% of nodular melanomas manifest as nonpigmented, fleshy nodules and therefore are referred to as amelanotic melanomas. Histologically, nodular melanoma appears as expanding lesions involving the papillary dermis with little or no epidermal component. Lesions are composed of epithelioid cells. Spindle cells and small epithelioid cells may also be present. Deeper invasion of the dermis and subcutaneous tissue occurs as lesions grow. Lentigo Maligna Melanoma. The lentigo maligna subtype of melanoma comprises less than 10% of cases.62,64 This subtype occurs mainly on the face or neck of fairskinned women older than 50 years of age.65 It typically manifests as a relatively large (>3 cm), flat, tan-colored (with different shades of brown) lesion that has been present for longer than 5 years. The border becomes irregular as the lesion enlarges. Histologically, lentigo maligna lesions are usually composed of spindle-like cells, which may be embedded in a connective tissue stroma with a desmoplastic pattern or may form fascicles displaying neural features and infiltrating the perineural space.65,66 Acral Lentiginous Melanoma. Acral lentiginous melanoma arises characteristically on the palms or soles or ­beneath the nail beds. However, some volar and plantar lesions are superficial spreading or nodular melanomas.67-71 The relative frequency of acral lentiginous melanoma varies

10-year mortality rate (%)

100 80 60 40

Observed Predicted

20 0 0

1

2

3

4

5

6

7

8

9

10 11 12 13

Tumor thickness (mm)

FIGURE 6-1.  Observed (blue circles) and predicted (blue line) 10-year mortality rates of 15,320 patients with clinically localized melanoma are based on a mathematical model f(t)= 1�������� –������� 0.988e (–211t + 0.009t2) derived from the American Joint Committee on Cancer (AJCC) melanoma database. T is the measured thickness (mm) and f(t) is the 10-year melanoma-specific mortality rates (P < .0001). (From Balch CM, Soong S, Gershenwald JE, et al. Prognostic factors analysis of 17,600 melanoma patients: validation of the American Joint Committee on Cancer melanoma staging system. J Clin Oncol 2001;19:3622-3634.)

substantially with race. This subtype comprises only about 5% of melanomas in fair-skinned people but 35% to 60% in dark-skinned individuals.72-74 Most acral lentiginous melanomas are brown and occur on the soles in individuals older than 60 years of age. Lesions generally evolve over a period of years and reach an average diameter of 3 cm before the diagnosis is established.68,70 Histologically, early-stage acral lentiginous melanoma is composed of large, highly atypical, pigmented cells along the dermoepidermal junction in an area of hyperplasia. At the invasive stage, infiltrating cells may be epithelioid or spindle-shaped.66 Infiltration may be present at diagnosis; if so, it occurs predominantly through the eccrine ducts.63

Biology and Pattern of Spread Although no single model of tumor growth and dissemination applies in all cases of malignant melanoma, growth tends to be stepwise from primary to lymphatic to distant site. Superficial spreading and lentigo maligna melanomas generally grow slowly over a period of many years (i.e., radial growth phase). Left untreated, these tumors gradually invade the dermis and subcutaneous tissue (i.e., vertical growth phase). In contrast, acral lentiginous and nodular melanomas progress more rapidly to a vertical growth phase. The lymphatic system is particularly important in the dissemination of malignant melanoma. The rate of regional lymph node metastasis and the rate of distant metastasis can be predicted based on the thickness of the primary ­lesion. The risk of nodal spread is 4% for lesions up to 1 mm thick, 14% for lesions 1.01 to 2.0 mm thick, 29% for lesions 2.01 to 4.0 mm thick, and 45% for lesions more than 4.0 mm thick. The risk of distant metastases also increases with the thickness of the primary lesion (Fig. 6-1)75 and with the number of involved lymph nodes (Fig. 6-2).76 What thickness cannot predict is the exact site or extent of future spread. Melanoma has an almost unique ability to disseminate early to combinations of dermal, ­subcutaneous,

Probability of distant metastasis (PDM)

6  The Skin 1.0

145

1 PDM = –(0.450+1.232log(n)) 1+e

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0

5

10

15

20

25

30

35

Number of positive nodes (n)

FIGURE 6-2.  The observed (blue circles) and predicted (blue line) probabilities of developing distant metastases increase according to the number of involved lymph nodes. The best-fit logistic regression equation is shown. (From Ballo MT, Ross MI, Cormier JN, et al. Combined-modality therapy for patients with regional nodal metastases from melanoma. Int J Radiat Oncol Biol Phys 2006;64:106-113.)

lymphatic, and visceral sites that are not amenable to most forms of therapy.

Merkel Cell Carcinoma Clinical Presentation and Pathology Merkel cell carcinoma, which is also known as trabecular carcinoma or neuroendocrine carcinoma of the skin, is a rare, aggressive neoplasm that was first described by Toker7 in 1972. It typically presents as a firm, painless, red, pink, or blue cutaneous nodule involving the head and neck or extremities in people older than 60 years.77,78 The median size of primary lesions at diagnosis is less than 2 cm, and the incidence of nodal involvement at presentation is approximately 20%.79 The differential diagnosis includes poorly differentiated squamous cell carcinoma, melanoma, eccrine carcinoma, lymphoma, Langerhans’ cell histiocytosis, and metastatic cancer, including islet cell carcinoma, medullary carcinoma of the thyroid, and small cell carcinoma of the lung. The cell of origin is believed to be the dermal neurotactile Merkel cell, arising from the neural crest.80 The lesion usually involves the reticular dermis and subcutaneous tissue, with minimal extension to the papillary dermis. The neoplastic cells possess cytoplasmic extensions and small, uniform, membrane-bound neurosecretory granules.81 Several immunohistochemical markers have been identified in Merkel cell carcinoma. The neoplastic cells stain most consistently with polyclonal antisera to neuron-specific enolase and calcitonin; stains based on monoclonal antibodies bind to neurofilament and cytokeratin.82-84

Biology and Pattern of Spread Merkel cell carcinoma has an aggressive course characterized by a propensity for contiguous spread and a high incidence of lymphatic and hematogenous dissemination.78,79,85-87 The traditional treatment, wide local excision, results in locoregional recurrence in almost three fourths of patients88;

146

PART 2  Skin

in contrast, with surgery plus radiation therapy, the locoregional recurrence rate is only 15%.77,79,84,89-91 Assessment of patterns of failure in a series of 54 patients with Merkel cell carcinoma treated at The University of Texas M. D. Anderson Cancer Center revealed local and regional relapse rates of 35% and 67%, respectively.79 Distant metastases occurred in 32 patients (59%). The median time to recurrence after surgical resection was 5 months, and 91% of relapses occurred within the first year. One patient died of uncontrolled primary disease, and 32 patients died of distant metastases. The 5-year disease-free survival rate was 11% for patients treated with surgery alone and 56% for patients treated with surgery and postoperative radiation therapy (P = .002). Similarly, Meeuwissen and colleagues89 reported that the 3-year disease-free survival rate was 0% for 38 patients treated with surgery alone and 68% for 34 patients treated with surgery and postoperative radiation therapy.

Adnexal Carcinomas Sebaceous Carcinoma Clinical Presentation and Pathology. Sebaceous carcinomas arise most commonly from the sebaceous glands of the ocular adnexa, the meibomian glands, and the glands of Zeis in the lid margin and the caruncle. Sites of origin, in decreasing frequency, are the head and neck, trunk and extremities, and external genitalia.92-95 Histologically, sebaceous carcinomas are composed of dermal-based lobules and cords of tumor cells with various degrees of sebaceous differentiation and infiltration into the surrounding tissues. Differentiated cells contain foamy or vacuolated, slightly basophilic cytoplasm, whereas less differentiated cells contain more deeply basophilic cytoplasm. Clinical and pathologic features indicative of a poor prognosis include size greater than 1 cm (associated with a 5-year mortality rate of 50%), vascular and lymphatic invasion, poor sebaceous differentiation, highly infiltrative growth pattern, and carcinomatous changes in the overlying epithelium.96 Among the malignant neoplasms of the eyelids, sebaceous carcinomas are second only to basal cell ­carcinomas in frequency and second only to malignant ­ melanomas in lethality.96 The most common location of ocular ­sebaceous carcinomas is the upper eyelid; the second most common location is the lower eyelid and caruncle.97 ­Sebaceous ­carcinomas predominantly affect people older than 60 years, with a slight preponderance for women. Sebaceous carcinomas manifest as slowly growing, nontender, deep-seated, firm nodules. Often, they produce a chalazion or manifestations of inflammatory processes such as conjunctivitis, blepharitis, or tarsitis. Small primary tumors may be overlooked until they invade the orbit or spread to preauricular or cervical nodes. The differential diagnosis includes sebaceous adenoma and basal cell carcinoma with sebaceous differentiation (i.e., basosebaceous  epithelioma).97,98 Distinction between basosebaceous epithelioma and sebaceous carcinoma is important because of the difference in the biologic behavior of the two tumors. Biology and Pattern of Spread. Sebaceous carcinoma behaves more aggressively than squamous cell carcinoma, with a propensity for nodal and hematogenous spread. Of 104 patients with sebaceous carcinoma registered by the Ophthalmic Pathology Branch of the Armed Forces

I­ nstitute of Pathology, 23 patients (22%) experienced one local recurrence and 11 patients experienced two or more local recurrences. Twenty-three patients died of metastatic disease.96

Eccrine Carcinoma Clinical Presentation and Pathology. Eccrine (sweat gland) carcinomas arise most often in the skin of the head and neck or the extremities and usually manifest as slowgrowing, painless papules or nodules.99 Eccrine carcinomas are usually diagnosed in people between 50 and 70 years of age.100 Histologically, eccrine carcinomas resemble carcinomas of the breast, bronchus, and kidney and may be difficult to distinguish from cutaneous metastases.99,101 Histologic subtypes include ductal eccrine carcinoma, mucinous eccrine carcinoma, porocarcinoma, syringoid eccrine carcinoma, clear  cell carcinoma, and microcystic adnexal carcinoma.99-106 Biology and Pattern of Spread. Eccrine carcinomas behave more aggressively than do basal cell and squamous cell carcinomas. In a series of 14 patients with eccrine carcinoma who underwent surgical excision with negative margins, 11 had at least one local recurrence, and 5 had recurrence in regional nodes or distant sites. One patient died of an uncontrolled local relapse, and four patients died of distant metastases 2 months to 10 years after diagnosis.99

Apocrine Carcinoma Apocrine carcinoma arises most commonly from apocrine glands in the axilla. It can also originate from apocrine glands in the vulva or eyelids and from ceruminal glands in the external auditory canal.107-109 Apocrine carcinoma usually manifests as a reddish purple nodule in elderly individuals. Because of the rarity of this lesion, its natural history is not well understood. Nodal and distant spread have been reported.110

Patient Evaluation and Staging Clinical evaluation of skin cancer consists of inspection and palpation of the involved area and the regional lymph nodes. Imaging studies, such as chest radiographs, computed tomography scans, and magnetic resonance imaging scans, should be obtained when clinically indicated.

Carcinoma The 2002 American Joint Committee on Cancer (AJCC) tumor-nodes-metastasis (TNM) staging system for carcinoma of the skin (Box 6-1) is based primarily on clinical findings.111

Melanoma Two surgical microstaging systems are used for melanoma: the Breslow system and the Clark method. Both systems are based on the recognition that for patients with melanomas, the prognosis depends on the depth of invasion rather than the greatest dimension of the tumor. The Breslow system classifies lesions by vertical thickness between the granular layer of the epidermis and the deepest part of the lesion as measured with an ocular  micrometer. For ulcerated lesions, measurements are made from the surface of the skin to the deepest part of the primary tumor.112

6  The Skin

BOX 6-1.  American Joint Committee on Cancer Staging System for Carcinoma of the Skin (Excluding Eyelid, Vulva, and Penis) Primary Tumor (T) TX Primary tumor cannot be assessed T0 No evidence of primary tumor Tis Carcinoma in situ T1 Tumor 2 cm or less in greatest dimension T2 Tumor more than 2 cm, but not more than 5 cm, in greatest dimension T3 Tumor more than 5 cm in greatest ­dimension T4 Tumor invades deep extradermal structures (i.e., cartilage, skeletal muscle, or bone) Regional Lymph Nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Regional lymph node metastasis Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage Grouping Stage 0 Tis Stage I T1 Stage II T2 T3 Stage III T4 Any T Stage IV Any T

N0 N0 N0 N0 N0 N1 Any N

M0 M0 M0 M0 M0 M0 M1

Note: In case of multiple simultaneous tumors, the tumor with the highest T category is classified, and the number of separate tumors is indicated in parentheses, such as T2 (5). From Greene FL, Page DL, Fleming ID, et al (eds): AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 203-208.

The Clark method divides melanomas into five groups on the basis of the depth of invasion relative to normal tissue structures. Level I is confined to the epidermis, level II invades the papillary dermis, level III invades the papillaryreticular dermal interface, level IV invades the reticular dermis, and level V invades the subcutaneous tissue.61 The 2002 AJCC staging system for melanoma  (Box 6-2) includes a four-stage grouping based on the Breslow thickness and the presence or absence of primary ­tumor ulceration, nodal involvement, and satellitosis or distant metastasis to the lung or other viscera. The Clark level of invasion is significant only for lesions less than 1 mm thick. A fifth group—stage 0, melanoma in situ—was ­ added in 1997.113

Primary Therapy and Results Carcinoma Primary Tumor The general strategy for treatment of the primary tumor is the same for basal cell and squamous cell carcinomas. A variety of treatment approaches are available, including curettage and electrodesiccation (not recommended for

147

r­ ecurrent lesions), Mohs’ surgery, cryotherapy, surgical resection, and primary radiation therapy.114-117 Nearly all treatment approaches result in 5-year recurrence rates of 1% to 10% for previously untreated cutaneous carcinomas.118 The choice of treatment for individual patients is determined on the basis of factors such as the location and size of the lesion; expected functional and cosmetic results; treatment time and cost; and patient age, occupation, and general physical condition.114 Surgical modalities are preferred for most patients, particularly those with small lesions, for whom simple surgical procedures yield high cure rates. Primary radiation therapy is most often indicated for lesions on or around the nose, lower eyelids, and ears, where radiation therapy yields better functional and cosmetic results than surgery. Similarly, primary radiation therapy is indicated for extensive lesions of the cheek, lip, or oral commissure that would require full-thickness resection. In a series of more than 1000 patients with previously untreated and recurrent basal and squamous carcinomas of the eyelid who underwent primary radiation therapy, 5-year local control rates were more than 90% (Table 6-1).119 The size of the primary lesion is the major determinant of local control after radiation therapy. In a series of 646 patients with carcinomas of the eyelid, pinna, nose, and lip treated with ­radiation, the 10-year local control rates were 98% for tumors measuring 2 cm or smaller, 79% for 2- to 5-cm lesions, and 53% (at 8 years) for carcinomas larger than 5 cm.120 When results are adjusted for tumor size, local control rates seem to be slightly higher for basal cell carcinoma than for squamous cell carcinoma: 97% versus 91% for lesions smaller than 1 cm and 87% versus 76% for 1- to 5-cm lesions.121 Recurrent ­basal and squamous cell carcinomas are more difficult to control than are previously untreated lesions.89,117,122,123 Radiation therapy usually is not recommended for ­patients younger than 50 years because the late effects after radiation therapy increase over time, as does the risk of a second cancer developing within the radiation portal.124,125 Postoperative radiation therapy may be beneficial in several clinical situations, including positive surgical margins, perineural involvement, invasion of bone or cartilage, extensive skeletal muscle infiltration, a positive node measuring more than 3 cm in the greatest dimension, extranodal spread, and multiple positive nodes. Series published in the 1950s and 1960s reporting the use of orthovoltage x-rays and large daily doses of radiation126-128 suggested that radiation therapy for carcinomas involving bone or cartilage resulted in an excessive risk of osteoradionecrosis or radiochondritis. However, data from more recent series using more fractionated, higher-energy radiation indicate that radiation therapy may be the treatment of choice for selected patients with bone or cartilage involvement because the risk of complications is low when proper techniques are used.129,130

Regional Nodes Elective nodal treatment is not indicated for cutaneous basal cell carcinomas because of the low incidence of lymphatic spread. However, elective nodal treatment is ­indicated for large, infiltrative, ulcerative squamous cell carcinomas or recurrent squamous cell carcinomas.131 The choice of treatment for patients who present with nodal disease depends on the type of therapy selected for

148

PART 2  Skin

BOX 6-2.  American Joint Committee on Cancer Staging System for Melanoma of the Skin Primary Tumor (T) TX T0 Tis T1 T1a T1b T2 T2a T2b T3 T3a T3b T4 T4a T4b Regional Lymph Nodes (N) NX N0 N1 N1a N1b N2 N2a N2b N2c N3 Distant Metastasis (M) MX M0 M1 M1a M1b M1c Clinical Stage Grouping Stage 0 Stage IA Stage IB Stage IIA Stage IIB Stage IIC Stage III

Stage IV Pathologic Stage Grouping Stage 0 Stage IA Stage IB Stage IIA Stage IIB Stage IIC Stage IIIA

Primary tumor cannot be assessed (e.g., shave biopsy or regressed melanoma) No evidence of primary tumor Melanoma in situ Melanoma ≤ 1.0 mm in thickness with or without ulceration Melanoma ≤ 1.0 mm in thickness and level II or III, no ulceration Melanoma ≤ 1.0 mm in thickness and level IV or V or with ulceration Melanoma 1.0-2.0 mm in thickness with or without ulceration Melanoma 1.0-2.0 mm in thickness, no ulceration Melanoma 1.0-2.0 mm in thickness, with ulceration Melanoma 2.01-4.0 mm in thickness with or without ulceration Melanoma 2.01-4.0 mm in thickness, no ulceration Melanoma 2.01-4.0 mm in thickness, with ulceration Melanoma greater than 4.0 mm in thickness with or without ulceration Melanoma > 4.0 mm in thickness, no ulceration Melanoma > 4.0 mm in thickness, with ulceration Regional lymph nodes cannot be assessed No regional lymph node metastasis Metastasis in one lymph node Clinically occult (microscopic) metastasis Clinically apparent (macroscopic) metastasis Metastasis in two to three regional nodes or intralymphatic regional metastasis without nodal metastasis Clinically occult (microscopic) metastasis Clinically apparent (macroscopic) metastasis Satellite or in-transit metastasis without nodal metastasis Metastasis in four or more regional nodes, or matted metastatic nodes, or in-transit metastasis or satellites with metastasis in regional nodes Distant metastasis cannot be assessed No distant metastasis Distant metastasis Metastasis to skin, subcutaneous tissues or distant lymph nodes Metastasis to lung Metastasis to all other visceral sites or distant metastasis at any site associated with an ­elevated serum lactic dehydrogenase (LDH) Tis T1a T1b T2a T2b T3 T3b T4 T4b Any T Any T Any T Any T

N0 N0 N0 N0 N0 N0 N0 N0 N0 N1 N2 N3 Any N

M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M1

Tis T1a T1b T2a T2b T3a T3b T4a T4b T1-4a T1-4a

N0 N0 N0 N0 N0 N0 N0 N0 N0 N1a N2a

M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0

6  The Skin

149

BOX 6-2.  American Joint Committee on Cancer Staging System for Melanoma of the Skin—Cont’d Pathologic Stage Grouping—Cont’d Stage IIIB T1-4b T1-4b T1-4a T1-4a T1-4a/b Stage IIIC T1-4b T1-4b Any T Stage IV Any T

N1a N2a N1b N2b N2c N1b N2b N3 Any N

M0 M0 M0 M0 M0 M0 M0 M0 M1

From Greene FL, Page DL, Fleming ID, et al (eds): AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 209-220.

TABLE 6-1.    Control Rates for Carcinomas of the Eyelid Treated with Radiation Therapy Number of Cases Treated Histologic Subtype

Total

Previously Untreated

Recurrent

5-Year Local Control Rate

Basal cell carcinoma

1062

686

376

1009 of 1062 (95%)

Squamous cell ­carcinoma Total

104

62

42

97 of 104 (92%)

1166

748

418

1106 of 1166 (95%)

From Fitzpatrick PJ, Thompson GA, Easterbrook WM, et al. Basal and squamous cell carcinoma of the eyelids and their treatment by radiotherapy. Int J Radiat Oncol Biol Phys 1984;10:449-454.

the primary lesion and the size of the involved lymph nodes. In general, single-modality therapy (surgery or ­ radiation therapy) is sufficient for nodal disease up to 3 cm without evidence of extracapsular spread.131 Taylor and colleagues132 reported results from 36 ­patients who underwent surgery and postoperative radiation therapy for lymph node metastases to the parotid ­region from cutaneous carcinomas. The overall regional control rate was 89%. All four recurrences occurred in ­ patients with positive surgical margins and clinical evidence of facial nerve involvement. For a later series of 37 patients with parotid-area metastases treated in the same way, Audet and associates133 reported a locoregional control rate of 73% and a 3-year disease-specific survival rate of 72%. In that analysis, patients with facial nerve involvement and tumors larger than 6 cm experienced poorer outcomes. Veness and colleagues,56 reporting on 74 patients with nonparotid cervical metastases, found that surgery combined with postoperative radiation therapy led to lower relapse rates and better disease-free survival times than did single-modality treatment for cervical metastases.

Melanoma For patients with localized cutaneous melanoma (stage  I or II), wide local excision of the primary tumor is the standard treatment. For patients with known lymph node metastases at presentation (stage III), wide local excision plus therapeutic lymph node dissection is the accepted surgical approach. Use of adjuvant radiation therapy is based on the clinical and pathologic features of the disease.

Primary Tumor The standard treatment for localized cutaneous melanoma (stages I and II) is wide local excision. Pathologic examination of the specimen is essential for establishment of a ­tissue diagnosis, assessment of the margins, and ­microstaging. The

recommended skin margins are 1 cm for invasive lesions less than 1 mm thick and for lentigo maligna melanomas and 2 to 3 cm for melanomas at least 1 mm thick. Melanoma thickness is the most powerful determinant of local control after wide local excision. In a large multi­institutional trial, long-term local recurrence rates were 2.3% for melanomas 1 to 2 mm thick, 4.2% for melanomas 2 to 3 mm thick, and 11.7% for melanomas 3 to 4 mm thick.134 Local recurrence is a poor prognostic sign; most patients with local recurrence ultimately die of the disease. Radiation therapy is rarely indicated as the definitive treatment for primary malignant melanomas, except in the case of large facial lentigo maligna melanomas, for which wide local excision may necessitate extensive reconstruction. In a series of patients with lentigo maligna melanomas treated with primary radiation therapy at the Princess Margaret Hospital and followed for a period of 6 months to 8 years (median, 2 years), local control was achieved in 23 (92%) of 25 patients.135 The median time to complete regression of lesions was 8 months, although some lesions took 2 years to disappear. Radiation therapy was delivered with orthovoltage x-rays (100 to 250 keV) and consisted of 35 Gy in 5 fractions over 1 week for lesions smaller than  3 cm, 45 Gy in 10 fractions over 2 weeks for lesions ­measuring 3 to 4.9 cm, and 50 Gy in 15 to 20 fractions over 3 to 4 weeks for lesions measuring 5.0 cm or more.

Regional Nodes Elective Lymph Node Dissection. For many years, the most controversial aspect of melanoma management was the role of elective lymph node dissection (ELND), particularly the management of the clinically uninvolved nodal basin. After wide local excision alone, the risk of nodal failure for patients with intermediate-thickness melanoma was up to 50%, and it was believed that early removal of microscopic nodal disease would avoid nodal recurrence

150

PART 2  Skin

and prevent distant spread. In 1982, results of a prospective, nonrandomized study at the Sydney Melanoma Unit involving 1319 patients suggested that ELND improved the survival of patients with intermediate-thickness (0.76 to 4 mm) melanomas.136 Subsequent trials, however, have not confirmed these results. The World Health Organization (WHO) Melanoma Program conducted a prospective phase III trial in which 553 patients with stage I disease were randomly assigned to receive wide local excision and ELND or wide local excision and delayed therapeutic lymphadenectomy (i.e., lymph node dissection only if regional nodes became clinically detectable).137 No difference in the survival rates was found between the two groups. Two other multicenter trials, one sponsored by the National Cancer Institute of the United States (Intergroup Melanoma Surgical Program)138 and the other sponsored by the WHO Melanoma Program,139 were launched in an effort to resolve the controversy. The Intergroup Surgical trial included 740 patients with clinically localized melanomas 1.0 to 4.0 mm thick. No significant difference in 5-year overall survival was found between patients assigned to ELND and those assigned to observation (86% versus 82%, P = .25). The WHO Melanoma Program trial included 240 eligible patients with truncal melanomas at least 1.5 mm thick with no clinical evidence of regional or distant metastases. Multivariate analysis revealed that routine use of immediate ELND had no effect on 5-year overall survival (hazard ratio, 0.72; 95% confidence interval [CI], 0.5 to 1.02). Since that time, Morton and colleagues140 developed a minimally invasive intraoperative lymphatic mapping technique that allows identification of the sentinel node, defined as the first node in the regional nodal basin into which the primary tumor drains. With this procedure, patients with clinically occult nodal disease who may benefit from early lymph node dissection can be identified.141 The results of an international randomized study to assess the efficacy of selective lymphadenectomy became available in 2006.142 In that study, 1269 patients with cutaneous melanomas between 1.2 and 3.5 mm thick were randomly assigned to undergo wide local excision followed by nodal observation and delayed dissection if necessary or sentinel lymph node biopsy, with an immediate completion lymph node dissection performed if micrometastases were found in the sentinel lymph node. The 5-year survival rate was 72% for patients presenting with an involved node, compared with 90% in the patients presenting with an uninvolved node (P < .001). Although the melanoma-specific survival rates were similar for the two groups, the diseasefree survival rate was better in the sentinel lymph node group (78% versus 73%, P = .009). Some therapeutic value was evident for completion lymph node biopsy in patients with a positive sentinel node. Among the patients with nodal disease, the 5-year survival rate was higher in the group who underwent immediate lymphadenectomy than in the group who were observed but required subsequent lymphadenectomy for nodal recurrence (72% versus 52%, P = .004). The group requiring salvage dissection also had higher numbers of involved lymph nodes at the time of dissection (3.3 versus 1.4, P < .001). Sentinel lymph node biopsy is now the standard of care for patients with 1.2- to 3.5-mm thick melanoma. For patients with thinner lesions,

the procedure is performed if the primary tumor is ulcerated, classified as Clark level IV or V, or associated with elevated mitotic counts. Therapeutic Lymph Node Dissection. Therapeutic nodal dissection with wide local excision is the accepted surgical approach for patients with clinically apparent nodal metastases from melanoma.143 Surgery alone is sufficient for more than 85% of patients with stage III disease. However, for patients with nodal extracapsular extension (including matted lymphadenopathy), four or more involved lymph nodes, lymph nodes larger than 3 cm, or recurrent nodal disease after previous dissection, the risk of regional recurrence can range from 30% to 50%.144,145 Although most such patients die as a consequence of distant spread, which usually manifests less than 2 years after treatment,146 locoregional relapses are significant because they often do not respond to therapy and can confer substantial morbidity through pain, infection, and disfigurement. Locoregional relapse from cutaneous melanoma of the head and neck can be particularly distressing because of compression of vital organs. Prevention of locoregional failure is desirable from a quality-of-life standpoint, even though overall survival may not be significantly improved.

Radiation Therapy Melanoma has long had a reputation for being radioresistant, although the origins of this supposed characteristic remain somewhat obscure. Laboratory studies have confirmed that melanoma cell lines have an enhanced ability to repair sublethal damage from radiation. The labeling of melanoma as radioresistant, however, predates such studies. Most likely, early clinical experience with treating visibly and palpably advanced dermal or nodal metastases resulted in visibly and palpably progressive disease inside and outside the radiation therapy field. Whereas for most diseases, radiologic studies are the mainstay of surveillance for detecting progressing disease, melanoma tends to involve superficial sites easily observed on physical examination. Several retrospective studies have revealed response rates of about 80% when radiation is delivered in large doses per fraction (i.e., 4 Gy or more), suggesting that if treated appropriately, melanoma is responsive to radiation therapy.147-149 The Radiation Therapy Oncology Group launched a study in 1983 to compare the effects of four fractions of  8 Gy given at weekly intervals with those of 20 fractions of 2.5 Gy given daily.150 A total of 137 patients were enrolled; although metastatic disease at any site other than the abdomen or brain was allowed, most was in the soft tissue, skin, or lymph nodes. Approximately one half of the lesions in this trial were larger than 5 cm in the greatest diameter, and response was measured clinically and radiographically. Although no difference was found in complete or partial response rates according to fraction size, the overall response rate in both treatment groups was 60%. Investigators at M.  D. Anderson Cancer Center have long questioned the reputed radioresistance of melanoma and recommended radiation therapy for appropriately selected patients. Before the advent of routine sentinel lymph node ­biopsy, elective radiation therapy was the preferred approach for treating primary melanoma of the head and neck ­ region that was more than 1.5 mm thick or Clark level IV or V because such therapy effectively suppressed

6  The Skin

regional ­ recurrence and was associated with less morbidity than elective lymph node dissection. In a retrospective analysis from M.  D. Anderson, Bonnen and colleagues151 reviewed 157 patients with stage I or II cutaneous melanoma of the head and neck who received elective ­regional radiation instead of lymph node dissection after wide  local excision of the primary site. Indications for regional radiation included primary tumor thickness of 1.5 mm or more or Clark level IV or V disease. The results revealed 15 regional failures (for a regional control rate of 89% at 10 years), despite estimates that 33 to 40 patients actually had microscopically involved regional nodes. Six percent of patients required medical care for a clinically significant complication, with moderate hearing loss being the most common complaint (six patients). Although elective nodal radiation suffers from the same limitations as ELND, it can effectively provide regional control for patients at risk of regional recurrence while avoiding surgical dissection. We reserve elective radiation therapy for patients with significant medical comorbid conditions for whom detailed prognostic information is of little relevance and enrollment on clinical trials is unlikely. For patients with nodal extracapsular extension, four or more involved lymph nodes, lymph nodes measuring over 3 cm, or recurrent nodal disease after previous dissection, therapeutic radiation results in excellent regional control. In a review of 466 patients with nodal metastases treated with lymphadenectomy and radiation (with or without systemic therapy), the actuarial 5-year regional (in-basin) control rate for all patients was 89%.76 On multivariate analysis, no patient or disease characteristics were found to be associated with inferior regional control. For this same group of patients, the 5-year disease-specific, disease-free, and distant metastasis-free survival rates were 49%, 42%, and 44%, respectively. Multivariate analysis indicated that increasing numbers of involved lymph nodes and the presence of ulceration in the primary tumor were associated with inferior 5-year actuarial disease-specific and distant metastasis-free survival rates. The number of involved lymph nodes was also associated with the development of brain metastases. The risk of lymphedema was highest for those patients with groin lymph node metastases.76

Systemic Adjuvant Therapy Because distant metastasis is the main pattern of relapse in melanoma, numerous attempts have been made to ­develop effective systemic therapy for this disease. However, none of the regimens tested has been found to improve overall survival in a phase III setting, with the exception of ­interferonalpha (IFN-α). An Eastern Cooperative ­ Oncology Group trial enrolled 280 patients with the following disease characteristics: primary melanoma more than 4 mm thick without palpable nodes, metastasis detected at ELND, primary melanoma of any stage with clinically ­ palpable regional lymph nodes, or nodal recurrence at any interval after surgery for primary melanoma of any depth.152 Patients received IFN intravenously five times each week for 4 weeks and then subcutaneously three times each week for 48 weeks. This treatment significantly increased 5-year actuarial relapse-free survival rates (37% versus 26%, P = .0023) and 5-year actuarial overall survival rates  (46% ­versus 37%, P = .024). The overall benefit of treatment in this trial correlated with tumor burden and the presence

151

of palpable regional lymph node metastases. The benefit of IFN therapy was greatest among patients with palpable nodal metastases at presentation or nodal recurrence. A French multi-institutional trial accrued 489 eligible patients with no palpable lymph nodes and no evidence of distant metastasis who underwent wide local excision of melanomas more than 1.5 mm thick.153 This phase III trial showed that patients treated with subcutaneous IFN three times per week for 18 months had a significantly ­ longer ­relapse-free interval than that of patients in the control group (P = .035). The hazard ratio for relapse in treated patients was 0.75 (95% CI, 0.57 to 0.98), and the 5-year relapse rate was 43% in the IFN group and 51% in the group that did not receive IFN. There was a trend toward improved overall survival in IFN-treated patients compared with the control group (P = .059). IFN remains the only systemic therapy approved by the U.S. Food and Drug Administration for patients with high-risk disease.154,155 Cisplatin-based chemotherapy has resulted in response rates of 12% to 30% in randomized trials, with no improvement in the overall survival, among patients with ­metastatic melanoma.156 Numerous phase II trials have combined tamoxifen with different chemotherapeutic agents. Because results have been conflicting, the benefit of tamoxifen remains unclear.156 A large number of melanoma vaccines are in develop­ ment or clinical testing.157 Autologous and allogeneic ­tumor cell vaccines yielded interesting results in initial studies. However, subsequent randomized trials (e.g., trials of vaccinia melanoma oncolysate vaccine) did not show improved survival. Progress in immunology and molecular biology has led to a newer generation of vaccines that are undergoing clinical testing, including cell-, ganglioside-, peptide-, and DNA-based vaccines.157

Merkel Cell Carcinoma The initial treatment for Merkel cell carcinoma is usually surgery to establish a tissue diagnosis. The role of radiation therapy in the management of Merkel cell carcinoma has been evaluated since the early 1980s because of the poor locoregional control rates obtained with surgery alone. The radiosensitivity of Merkel cell carcinoma has been demonstrated in several studies.77,79,84,89,90,158 Pacella and colleagues84 reported that local control was achieved in 35 of 36 fields irradiated in 19 patients (gross disease was present in 23 of the fields). Similarly, in-field control was obtained in 30 of 31 patients treated in a series at M.  D. Anderson.79 However, marginal recurrences developed in three ­patients,79 indicating that the target volume should include a wide margin around gross disease. Between 1982 and 1987, as the natural history of Merkel cell carcinoma became more apparent, elective postoperative locoregional radiation therapy was given to five of six patients referred to M.  D. Anderson after local excision.79 The primary tumor bed and draining lymphatics were irradiated with an electron beam to cumulative doses of 50 to 55 Gy and 45 to 55 Gy, respectively, in daily 2.0- to 2.5-Gy fractions. At a minimum follow-up time of 3 years, none of these patients had experienced a local or regional recurrence. Although patient numbers in this study were small because of the rarity of Merkel cell carcinoma, many ­retrospective series support the use of surgery and postoperative radiation therapy for previously untreated or

152

PART 2  Skin

recurrent primary tumors.57,77,79,84,88,123,158-161 The available data are insufficient to permit detailed dose-response analysis. However, in-field recurrence seems to be rare after a total dose of 50 Gy to microscopic disease given in daily 2.0- to 2.5-Gy fractions.79,161 Evidence of the responsiveness of Merkel cell carcinoma to cytotoxic agents continues to emerge.162-164 The most active drugs are doxorubicin, cyclophosphamide, and vincristine. The Trans-Tasman Radiation Oncology Group reported the results of a phase II study of radiation therapy with concurrent carboplatin and etoposide for patients with localized, high-risk Merkel cell carcinoma. Three-year outcomes were promising, with an overall survival rate of 76%, a locoregional control rate of 75%, and a distant control rate of 76%.165 However, a comparison of the study cohort with a cohort of similar historical controls showed that chemotherapy did not significantly affect overall survival or disease-specific survival rates.166 Responses of ­metastatic deposits to multiagent chemotherapy can be dramatic, but unfortunately, such responses are often of short duration.164,167 The treatment policy for localized Merkel cell carcinoma at M.  D. Anderson can be summarized as follows. For patients presenting with no evidence of regional or distant metastases, surgical excision of the primary tumor should be followed in all cases by adjuvant radiation therapy to generous margins around the primary site. Although elective irradiation of the draining lymphatics had been recommended previously,117 current guidelines call for routine use of sentinel lymph node biopsy.168 If the sentinel node is involved, radiation therapy to the regional basin is preferred to surgical dissection. Given the radiosensitivity of Merkel cell carcinoma,79,84,158 patients who for medical reasons cannot undergo surgery are treated with primary radiation therapy to higher doses with elective nodal irradiation. For gross, unresected disease, 56 to 66 Gy is recommended.79 For patients presenting with regional lymph node metastases, our current practice is local excision of the primary tumor, therapeutic nodal dissection, and postoperative irradiation of the surgical bed, including the draining lymphatics. Chemotherapy is also considered for those patients with large primary tumors or adenopathy.

Adnexal Carcinomas The standard treatment for sebaceous carcinoma is wide local excision and therapeutic nodal dissection if lymphadenopathy is present. The role of adjuvant therapeutic modalities has not been explored systematically. As is the case for other types of skin cancer, indications for postoperative irradiation include positive surgical margins, perineural spread, invasion of bone or cartilage, extensive skeletal muscle infiltration, a positive node measuring more than  3 cm in the greatest dimension, extranodal spread, and multiple positive nodes. For eccrine and apocrine carcinomas, wide excision of the primary lesion is accepted as the standard treatment for patients with no lymphadenopathy at diagnosis. Postoperative radiation therapy to the surgical bed with 2- to 3-cm margins is indicated, particularly for eccrine carcinoma, because of the high incidence of local recurrence after surgery alone. The role of elective treatment of regional lymphatics is controversial. Treatment for patients with palpable lymphadenopathy consists of wide local excision,

lymph node dissection, and postoperative irradiation of the surgical bed.

Treatment of Advanced Disease and Palliative Therapy Radiation therapy administered to total doses of 60 to  70 Gy (depending on tumor size, location, and anticipated toxicity) can result in local control in some patients with advanced skin carcinomas. Lee and colleagues169 ­reported a 5-year local control rate of 67% in a group of 33 patients with previously untreated T4 skin carcinomas who underwent radiation therapy to doses of 60 to 75 Gy over 6 to 8 weeks. Bone invasion or perineural spread reduced the likelihood of local control in this small series of patients. Radiation therapy can reduce symptoms for most patients with locally advanced inoperable or metastatic skin cancer. The recommended dose-fractionation schedules depend on the tumor location and the patient’s life expectancy. The most commonly used regimens are 30 Gy in 10 fractions over 2 weeks, 45 to 50 Gy in 18 to 25 fractions over 4 to 5 weeks, and 36 Gy in 6 fractions over 3 weeks. The third of these regimens is used for the treatment of melanoma. For patients with one or two brain metastases, particularly from melanoma, and good performance status with stable extracranial disease, stereotactic radiosurgery (15 to 20 Gy in a single fraction) is delivered to treat the brain metastases.170

Principles of Radiation Therapy The general principles of radiation therapy for skin cancer are discussed in this section. The technical details of radiation treatment for skin cancer are described elsewhere.171

Carcinoma In radiation therapy for skin carcinoma, the target volume is determined mainly by the size of the lesion and the type of growth. Infiltrative lesions and those with ill-defined borders (e.g., morphea-like carcinomas) require more generous margins. Orthovoltage x-rays or an electron beam characteristically is used. With the electron beam, the field size at the skin surface is 0.5 cm larger because of constriction of the isodose lines in the depth.117 Typically, the portal is designed to encompass the visible or palpable tumor, along with 0.5 to 1.0 cm of surrounding normal-appearing skin in the case of carcinomas smaller than 1 cm and up to 2.0 cm of normal-appearing skin for larger or ill-defined carcinomas.172 The field size may be reduced after the delivery of a dose sufficient to sterilize microscopic disease. Margins may be smaller when areas close to the eye are treated. In some patients—almost exclusively those with squamous cell carcinoma who present with small (≤3 cm) involved lymph nodes or advanced primary tumors—the initial field is designed to encompass the primary tumor and regional lymphatics. The target volume is subsequently reduced to cover the areas of gross disease with adequate margins. Perineural invasion typically extends 1 to 2 cm along the nerve sheath173 but may rarely extend up to 14 cm.37

6  The Skin

Consequently, in the presence of major nerve invasion, the target volume should encompass the involved nerve to the skull base.57,58,174,175 In patients undergoing postoperative irradiation for pathologically proven perineural invasion or lymph node involvement, the initial target volume includes the primary tumor bed plus the potential route of perineural spread or the regional lymphatics. The target volume is then reduced to include the primary tumor and nodal beds. Several fractionation schedules have been used to treat skin carcinomas. In general, the more protracted schedules produce the best cosmetic results.120,176 The recommended radiation regimen for individual patients depends on the size and location of the carcinoma. For most cases of skin cancer, 40 Gy in 10 fractions, 45 Gy in 15 fractions, or 50 Gy in 20 fractions are appropriate and effective regimens.120 If the patient is elderly and in generally poor health, 32 Gy in 4 fractions or a single dose of 20 Gy may be used. For large lesions close to the eye, 60 to 70 Gy in 30 to 35 fractions is recommended.120 Patients with positive nodes usually receive 2-Gy fractions because of the large irradiated volume. Similarly, postoperative radiation therapy is delivered using 2-Gy fractions. The dose for orthovoltage x-rays is prescribed to Dmax (the maximum point dose), whereas the dose for electrons is prescribed to the 90% isodose line to take into account the difference in relative biologic effectiveness between the beams.177,178

Melanoma The target volume for patients with stage I to III melanoma depends on the site of disease. For adjuvant radiation therapy to the primary tumor site, the surgical bed is treated with margins of at least 2 cm. For cervical nodal metastases, the ipsilateral neck nodes down to the clavicle are included within an appositional electron beam field with the patient in an open-neck position (Fig. 6-3). For axillary disease, anterior and posterior photon beams should include level I through III axillary nodes. Typically, the cervical nodes are included only if they are involved or if more than 10 axillary nodes are involved. For nodal metastases to the groin, anteriorly weighted anterior and posterior photon beams are used to treat the groin, with electron fields matched above and below as needed to cover the full extent of the surgical scar. Elective adjuvant radiation therapy is administered in 6-Gy fractions prescribed to Dmax twice per week to a total dose of 30 Gy. If microscopic residual disease is present, an additional fraction may be given through a smaller portal for a cumulative dose of 36 Gy. Care is taken to ensure that the dose to the spinal cord does not exceed 24 Gy in  4 fractions over 2 weeks.

Merkel Cell and Adnexal Carcinomas In radiation therapy for Merkel cell carcinoma, the target volume consists of the primary tumor bed with 4- to  5-cm margins, except when the lesion is situated at or close to critical structures (e.g., optic apparatus). In the case of Merkel cell carcinoma of the head and neck region, the whole ipsilateral neck is irradiated. In radiation therapy for adnexal carcinoma, the target volume consists of the ­primary tumor bed with 2- to 3-cm margins and the draining lymphatics.

153

FIGURE 6-3.  The typical electron field arrangement is shown for a patient with nodal metastases from melanoma. The patient presented with a nodal recurrence of melanoma several years after a superficial spreading primary lesion on the temporal scalp had been removed. Limited neck dissection was followed by adjuvant radiation therapy. The treatment was simulated with the patient in the indicated open-neck position, and the field encompassed the low parotid down to the supraclavicular lymph nodes. A beveled tissue equivalent bolus was used over the larynx and within the ear canal. A beveled bolus may also be applied over the temporal lobe, represented by a line drawn between the lateral canthus and the mastoid tip (not shown). The radiation dose was prescribed to Dmax.

Radiation is administered, preferably with electrons, in 2-Gy daily fractions. For Merkel cell carcinoma, a dose of 46 to 50 Gy is delivered to the region of elective radiation therapy, followed by a boost consisting of 10 Gy in 5 fractions to the gross disease. At M.  D. Anderson, a cumulative dose of 66 Gy is given in the case of bulky primary tumors. For adnexal carcinoma, a dose of 60 Gy usually is administered in the postoperative setting.

Patient Care during and after Radiation Therapy A tumoricidal dose of radiation usually produces moist desquamation. The skin reaction usually reaches a peak 3 to 6 weeks after the start of radiation therapy and resolves spontaneously within 6 weeks. Irradiated skin should be pro­ tected from heat, cold, sunlight, friction, and other sources of irritation to avoid additional tissue injury. Patients are instructed to use sun blocks with a sun protection factor of at least 15 over the irradiated area and, in the case of facial lesions, to wear a hat when outdoors after the completion of treatment. After moist desquamation occurs, the area should be cleaned with aluminum subacetate (Domeboro) soaks or 1% hydrogen peroxide to prevent secondary infection. Patients should also be informed about the specific acute and late side effects of treatment, such as mucositis of the nasal passages and lips, nasal dryness, synechiae (when nostril lesions are treated), conjunctivitis, and loss of eyelashes (when eyelid lesions are treated). When the lacrimal canaliculi are treated, irrigation of the ducts during the healing phase after treatment helps to prevent synechiae. Late side effects include hypopigmentation, hyperpigmentation, telangiectasia, and skin atrophy, which may

154

PART 2  Skin

develop gradually over a period of years. Hair loss and loss of sweat gland function are usually permanent. Irradiated skin is more sensitive to trauma. Blood flow in the irradiated tissues may decrease. As a result, healing may be delayed after surgery in an irradiated region. Skin retraction at the lower eyelid may result in ectropion. Fibrosis of the lacrimal duct may result in epiphora. These side effects are rarely serious after properly administered radiation therapy. Cataracts can occur after irradiation of lower-eyelid lesions, but this side effect is uncommon when appropriate shielding is used. The long-term cosmetic results after radiation therapy depend on the fraction size, total dose, and volume of tissue irradiated. When appropriate techniques, such as those described here, are used, the risk of soft tissue necrosis is less than 3%, and the risk of osteoradionecrosis is approximately 1%. Radiochondritis is rare and is usually a sign of recurrent disease.179 Regular follow-up examinations include evaluation of the treated area for late complications and—because one third of treated patients will develop metachronous skin cancers—inspection for new lesions.

REFERENCES   1. American Cancer Society. Cancer Facts & Figures—2007.  Atlanta: American Cancer Society, 2007.   2. Brown RO, Osguthorpe JD. Management of the neck in nonmelanocytic cutaneous carcinomas. Otolaryngol Clin North Am 1998;31:841-856.   3. Wang SQ, Halpern AC. Management of cutaneous melanoma: a public health and individual patient care perspective. Adv  Dermatol 2007;23:81-98.   4. Goldberg MS, Doucette JT, Lim HW, et al. Risk factors for presumptive melanoma in skin cancer screening: American Academy of Dermatology National Melanoma/Skin Cancer Screening Program experience 2001-2005. J Am Acad Dermatol 2007;57:60-66.   5. Rigel DS, Kopf AW, Friedman RJ.The rate of malignant melanoma  in the United States: are we making an impact? J Am Acad Dermatol 1987;17:1050-1053.   6. Rigel DS, Friedman RJ, Kopf AW. The incidence of malignant melanoma in the United States: issues as we approach the 21st century. J Am Acad Dermatol 1996;34:839-847.   7. Toker C. Trabecular carcinoma of the skin. Arch Dermatol 1972;105:107-110.   8. Urban FH, Winkelmann RK. Sebaceous malignancy. Arch Dermatol 1961;84:63-72.   9. Tulenko JF, Conway H. An analysis of sweat gland tumors. Surg Gynecol Obstet 1965;121:343-348. 10. Rubin P, Casarett G. Clinical Radiation Pathology; vol 2. ­Philadelphia: WB Saunders, 1968, p 1057. 11. Fajardo LF. ������������������������������������ Skin�������������������������������� . In Fajardo LF, Berthrong M, Anderson RE (eds): Radiation Pathology. New York: Oxford University Press, 2001, pp 411-420. 12. Cleaver JE. Defective repair replication of DNA in xeroderma pigmentosum. Nature 1968;218:652-656. 13. Anderson DE, Taylor WB, Falls HF, et al. The nevoid basal cell carcinoma syndrome. Am J Hum Genet 1967;19:12-22. 14. Lutzner MA. Epidermodysplasia verruciformis: an autosomal recessive disease characterized by viral warts and skin cancer. A model for viral oncogenesis. Bull Cancer 1978;65:169-182. 15. Lew RA, Sober AJ, Cook N, et al. Sun exposure habits in patients with cutaneous melanoma: a case control study. J Dermatol Surg Oncol 1983;9:981-986. 16. Bellet RE, Vaisman I, Mastrangelo MJ, et al. Multiple primary malignancies in patients with cutaneous melanoma. Cancer 1977;40:1974-1981.

17. Scheibner A, Milton GW, McCarthy WH, et al. Multiple primary melanoma—a review of 90 cases. Australas J Dermatol 1982;23:1-8. 18. Veronesi U, Cascinelli N, Bufalino R. Evaluation of the risk of multiple primaries in malignant cutaneous melanoma. Tumori 1976;62:127-130. 19. Lundberg I, Gustavsson A, Holmberg B, et al. Mortality and cancer incidence among PVC-processing workers in Sweden. Am J Ind Med 1993;23:313-319. 20. Linet MS, Malker HS, Chow WH, et al. Occupational risks for cutaneous melanoma among men in Sweden. J Occup Environ Med 1995;37:1127-1135. 21. Ferrini RL, Perlman M, Hill L. American College of Preventive Medicine practice policy statement: skin protection from ultraviolet light exposure. Am J Prev Med 1998;14:83-86. 22. Kopf AW. Computer analysis of 3531 basal cell carcinomas of the skin. J Dermatol 1979;6:267-281. 23. Rahbari H, Mehregan AH. Basal cell epitheliomas in usual and unusual sites. J Cutan Pathol 1979;6:425-431. 24. Robinson JK. Risk of developing another basal cell carcinoma: a 5-year prospective study. Cancer 1987;60:118-120. 25. Aasi SZ, Leffell DJ. Cancer of the skin. In DeVita VT, Hellman S,  Rosenberg SA (eds): Cancer: Principles and Practice of Oncology, 5th ed. Philadelphia: Lippincott Williams & Wilkins, 2005, pp 1717-1744. 26. Siegle RJ, MacMillan J, Pollack S. Infiltrative basal cell carcinoma: a nonsclerosing subtype. J Dermatol Surg Oncol 1986;12:830836. 27. Battle RJ, Patterson TJ. The surgical treatment of basal-celled carcinoma. Br J Plast Surg 1960;13:118-135. 28. Mohs FE, Lathrop TG. Modes of spread of cancer of the skin. AMA Arch Derm Syphilol 1952;66:427-439. 29. Panje WR, Ceilley RI. The influence of embryology of the midface on the spread of epithelial malignancies. Laryngoscope 1979;89:1914-1920. 30. Wentzell JM, Robinson JK. Embryologic fusion planes and the spread of cutaneous carcinoma: a review and reassessment.  J Dermatol Surg Oncol 1990;16:1000-1006. 31. Dodd GD, Dolan PA, Ballantyne AJ, et al. The dissemination of tumors of the head and neck via the cranial nerves. Radiol Clin North Am 1970;8:445-461. 32. Binkley GW, Rauschkolb RR. Basal-cell epithelioma metastasizing to lymph nodes. Arch Dermatol 1962;86:332-342. 33. von Domarus H, Stevens PJ. Metastatic basal cell carcinoma: report of five cases and review of 170 cases in the literature.  J Am Acad Dermatol 1984;10:1043-1060. 34. Carlson KC, Roenigk RK. Know your anatomy: perineural involvement of basal and squamous cell carcinoma on the face.  J Dermatol Surg Oncol 1990;16:827-833. 35. Hanke CW, Wolf RL, Hochman SA, et al. Chemosurgical ­reports: perineural spread of basal-cell carcinoma. J Dermatol Surg Oncol 1983;9:742-747. 36. Smith JB, Bishop VL, Francis IC, et al. Ophthalmic manifestations of perineural spread of facial skin malignancy. Aust NZ J Ophthalmol 1990;18:197-205. 37. Ballantyne AJ, McCarten AB, Ibanez M. The extension of cancer of the head and neck through peripheral nerves. Am J Surg 1963;106:651-657. 38. Goepfert H, Dichtel WJ, Medina JE, et al. Perineural invasion in squamous cell skin carcinoma of the head and neck. Am J Surg 1984;148:542-547. 39. Mendenhall WM, Parsons JT, Mendenhall NP, et al. Carcinoma of the skin of the head and neck with perineural invasion. Head Neck 1989;11:301-308. 40. Clouston PD, Sharpe DM, Corbett AJ, et al. Perineural spread of cutaneous head and neck cancer. Its orbital and central neurologic complications. Arch Neurol 1990;47:73-77. 41. Mark GJ. Basal cell carcinoma with intraneural invasion. Cancer 1977;40:2181-2187. 42. Paver K, Poyzer K, Burry N, et al. The incidence of basal cell carcinomas and their metastases in Australia and New Zealand [letter]. Australas J Dermatol 1973;14:53.

6  The Skin 43. Cannon JR, Schneidman DW. Recent developments in adnexal pathology. In Moschella SL (ed): Dermatology Update: Reviews for Physicians. New York: Elsevier, 1982, pp 217-252. 44. Soffer D, Kaplan H, Weshler Z. Meningeal carcinomatosis due to basal cell carcinoma. Hum Pathol 1985;16:530-532. 45. Epstein JH. Photocarcinogenesis, skin cancer, and aging. J Am Acad Dermatol 1983;9:487-502. 46. Akguner M, Barutcu A, Yilmaz M, et al. Marjolin’s ulcer and chronic burn scarring. J Wound Care 1998;7:121-122. 47. Hejna WF. Squamous-cell carcinoma developing in the chronic draining sinuses of osteomyelitis. Cancer 1965;18:128-132. 48. Epstein E, Epstein NN, Bragg K, et al. Metastases from squamous cell carcinomas of the skin. Arch Dermatol 1968;97:245-251. 49. Friedman HI, Cooper PH, Wanebo HJ. Prognostic and therapeutic use of microstaging of cutaneous squamous cell carcinoma of the trunk and extremities. Cancer 1985;56:1099-1105. 50. Immerman SC, Scanlon EF, Christ M, et al. Recurrent squamous cell carcinoma of the skin. Cancer 1983;51:1537-1540. 51. Binder SC, Cady B, Catlin D. Epidermoid carcinoma of the skin of the nose. Am J Surg 1968;116:506-512. 52. Dinehart SM, Pollack SV. Metastases from squamous cell carcinoma of the skin and lip. An analysis of twenty-seven cases. J Am Acad Dermatol 1989;21:241-248. 53. Modlin JJ. Cancer of the skin: surgical treatment. Mol Med 1954;51:364-367. 54. Møller R, Reymann F, Hou-Jensen K. Metastases in dermatological patients with squamous cell carcinoma. Arch Dermatol 1979;115:703-705. 55. Moore BA, Weber RS, Prieto V, et al. Lymph node metastases from cutaneous squamous cell carcinoma of the head and neck. Laryngoscope 2005;115:1561-1567. 56. Veness MJ, Palme CE, Morgan GJ. High-risk cutaneous squamous cell carcinoma of the head and neck: results from 266 treated patients with metastatic lymph node disease. Cancer 2006;106:2389-2396. 57. Bourne RG. The Costello Memorial Lecture. The spread of squamous cell carcinoma of the skin via the cranial nerves. Australas Radiol 1980;24:106-114. 58. Loeffler JS, Larson DA, Clark JR, et al. Treatment of perineural metastasis from squamous carcinoma of the skin with aggressive combination chemotherapy and irradiation. J Surg Oncol 1985;29:181-183. 59. Mohs FE. Chemosurgery: Microscopically Controlled ­ Surgery for Skin Cancer. Springfield, IL: Charles C Thomas, 1978,  pp 261-267. 60. Gruber SB, Barnhill RL, Stenn KS, et al. Nevomelanocytic proliferations in association with cutaneous melanoma: a multivariate analysis. J Am Acad Dermatol 1989;21:773-780. 61. Clark WH, Ainsworth AM, Bernardino EA. The developmental biology of primary human malignant melanomas. Semin Oncol 1975;2:83-103. 62. McGovern VJ, Murad TM. Pathology of melanoma: an overview. In Balch CM, Milton GW (eds): Cutaneous Melanoma: Clinical Management and Treatment Results Worldwide. Philadelphia: JB Lippincott, 1985, pp 29-53. 63. Barnhill RL, Mihm MC. Histopathology of malignant melanoma and its precursor lesions. In Balch CM, Houghton AN, Milton GW, et al (eds): Cutaneous Melanoma, 2nd ed. Philadelphia: JB  Lippincott, 1992, pp 234-263. 64. Urist MM, Balch CM, Soong SJ. Head and neck melanoma in 534 clinical stage I patients: a prognostic factor analysis and results of surgical treatment. Ann Surg 1984;200:769-775. 65. Clark WH, Mihm MC. Lentigo maligna and lentigo-maligna melanoma. Am J Pathol 1969;55:39-67. 66. Reed RJ. The pathology of human cutaneous melanoma. In Costanzi JJ (ed): Malignant Melanoma I. The Hague: Martinus Nijhoff, 1983, pp 85-116. 67. Arrington JH, Reed RJ, Ichinose H, et al. Plantar lentiginous melanoma: a distinctive variant of human cutaneous malignant melanoma. Am J Surg Pathol 1977;1:131-143. 68. Coleman WP, Loria PR, Reed RJ, et al. Acral lentiginous melanoma. Arch Dermatol 1980;116:773-776.

155

69. Feibleman CE, Stoll H, Maize JC. Melanomas of the palm, sole, and nailbed: a clinicopathologic study. Cancer 1980;46:2492-2504. 70. Krementz ET, Feed RJ, Coleman WP, et al. Acral lentiginous melanoma. A clinicopathologic entity. Ann Surg 1982;195:  632-645. 71. Patterson RH, Helwig EB. Subungual malignant melanoma: a clinical-pathologic study. Cancer 1980;46:2074-2087. 72. Balch CM, Urist MM, Maddox WA, et al. Melanoma in the southern United States: experience at the University of Alabama in Birmingham. In Balch CM, Milton GW (eds): Cutaneous Melanoma: Clinical Management and Treatment Results Worldwide, 2nd ed. Philadelphia: JB Lippincott, 1985, pp 397-418. 73. Reintgen DS, McCarty KM, Cox E, et al. Malignant melanoma in black American and white American populations: a comparative review. JAMA 1982;248:1856-1859. 74. Seiji M, Takahashi M. Acral melanoma in Japan. Hum Pathol 1982;13:607-609. 75. Balch CM, Soong SJ, Gershenwald JE, et al. Prognostic factors analysis of 17,600 melanoma patients: validation of the American Joint Committee on Cancer melanoma staging system.  J Clin Oncol 2001;19:3622-3634. 76. Ballo MT, Ross MI, Cormier JN, et al. Combined-modality therapy for patients with regional nodal metastases from melanoma. Int J Radiat Oncol Biol Phys 2006;64:106-113. 77. Cotlar AM, Gates JO, Gibbs FA. Merkel cell carcinoma: combined surgery and radiation therapy. Am Surg 1986;52:  159-164. 78. Sibley RK, Dehner LP, Rosai J. Primary neuroendocrine (Merkel cell?) carcinoma of the skin. I.  A clinicopathologic and ultrastructural study of 43 cases. Am J Surg Pathol 1985;9:  95-108. 79. Morrison WH, Peters LJ, Silva EG, et al. The essential role of radiation therapy in securing locoregional control of Merkel cell carcinoma. Int J Radiat Oncol Biol Phys 1990;19:583-591. 80. Merkel F. Tastzellen und tastkoerperchen bei den haushieren und beim menschen. Arch Mikr Anat 1875;11:636-652. 81. Silva EG, Mackay B, Goepfert H, et al. Endocrine carcinoma of the skin (Merkel cell carcinoma). Pathol Annu 1984;19:1-30. 82. Bonfiglio TA. Neuroendocrine carcinoma of the skin: diagnostic and management considerations. Int J Radiat Oncol Biol Phys 1988;14:1321-1322. 83. Drijkoningen M, De Wolf-Peeters C, van Limbergen E, et al. Merkel cell tumor of the skin: an immunohistochemical study. Hum Pathol 1986;17:301-307. 84. Pacella J, Ashby M, Ainslie 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;14:1077-1084. 85. Bourne RG, O’Rourke MG. Management of Merkel cell tumour. Aust NZ J Surg 1988;58:971-974. 86. Hitchcock CL, Bland KI, Laney RG, et al. Neuroendocrine (Merkel cell) carcinoma of the skin. Its natural history, diagnosis and treatment. Ann Surg 1988;207:201-207. 87. Meland NB, Jackson IT. Merkel cell tumor: diagnosis, prognosis, and management. Plast Reconstr Surg 1986;77:632-638. 88. Rice RD Jr, Chonkich GD, Thompson KS, et al. Merkel cell tumor of the head and neck. Five new cases with literature review. Arch Otolaryngol Head Neck Surg 1993;119:782-786. 89. 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;31:325-331. 90. O’Brien PC, Denham JW, Leong AS. Merkel cell carcinoma: a review of behaviour patterns and management strategies.  Austral NZ J Surg 1987;57:847-850. 91. Wilder RB, Kittelson JM, Shimm DS. Basal cell carcinoma ­treated with radiation therapy. Cancer 1991;68:2134-2137. 92. Batsakis JG, Littler ER, Leahy MS. Sebaceous cell lesions of the head and neck. Arch Otolaryngol 1972;95:151-157. 93. Civatte J, Tsoitis G. Adnexal skin carcinomas. In Andrade R,  Gumport SL, Popkin GL, ��������������������������������������������� et al (�������������������������������������� eds): Cancer of the Skin,  vol 2. Philadelphia: WB Saunders, 1976, pp 1045-1068.

156

PART 2  Skin

94. Oppenheim AR. Sebaceous carcinoma of the penis. Arch Dermatol 1981;117:306-307. 95. Rulon DB, Helwig EB. Cutaneous sebaceous neoplasms. Cancer 1974;33:82-102. 96. Rao NA, Hidayat AA, McLean IW, et al. Sebaceous carcinomas of the ocular adnexa: a clinicopathologic study of 104 cases, with five-year follow-up data. Hum Pathol 1982;13:113-122. 97. Boniuk M, Zimmerman LE. Sebaceous carcinoma of the eyelid, eyebrow, caruncle and orbit. Trans Am Acad Ophthalmol Otolaryngol 1968;72:619-642. 98. Wick MR, Goellner JR, Wolfe JT, et al. Adnexal carcinomas of the skin. II. Extraocular sebaceous carcinomas. Cancer 1985;56:1163-1172. 99. Wick MR, Goellner JR, Wolfe JT III, et al. Adnexal carcinomas of the skin. I. Eccrine carcinomas. Cancer 1985;56:1147-1162. 100. Snow SN, Reizner GT, Dudley CM, et al. Eccrine mucinous carcinoma. In Miller SJ, Maloney ME (eds): Cutaneous ­Oncology: Pathophysiology, Diagnosis, and Management. Malden, MA: Blackwell Science, 1998, pp 758-762. 101. Berg JW, McDivitt RW. Pathology of sweat gland carcinomas. Pathol Annu 1968;3:123-144. 102. Lipper S, Peiper SC. Sweat gland carcinoma with syringomatous features: a light microscopic and ultrastructural study. Cancer 1979;44:157-163. 103. Liu Y. The histogenesis of clear cell papillary carcinoma of the skin. Am J Pathol 1949;25:93-99. 104. Mendoza S, Helwig EB. Mucinous (adenocystic) carcinoma of the skin. Arch Dermatol 1971;103:68-78. 105. Mishima Y, Morioka S. Oncogenic differentiation of the intraepidermal eccrine sweat duct: eccrine poroma, poroepithelioma and porocarcinoma. Dermatologica 1969;138:238-250. 106. Nickoloff BJ, Fleischmann HE, Carmel J, et al. Microcystic adnexal carcinoma: immunohistologic observations suggesting dual (pilar and eccrine) differentiation. Arch Dermatol 1986;122:290-294. 107. Ota Y, Fukasawa I, Shimizu K, et al. A case of T1N0M0 vulvar apocrine gland carcinoma with a positive outcome. Gynecol Oncol 2003;90:601-604. 108. Shintaku M, Tsuta K, Yoshida H, et al. Apocrine ­adenocarcinoma of the eyelid with aggressive biological behavior: report of a case. Pathol Int 2002;52:169-173. 109. Mansour P, George MK, Pahor AL, et al. Ceruminous gland ­tumours: a reappraisal. J Laryngol Otol 1992;106:727-732. 110. Roy S, Shafi NQ, Rose MG. Locally recurrent and metastatic apocrine-gland carcinoma in an elderly man. Nat Clin Pract ­Oncol 2007;4:56-59. 111. Greene FL, Page DL, Fleming ID, et al (eds): AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002. 112. Breslow A. Thickness, cross-sectional areas and depth of invasion in the prognosis of cutaneous melanoma. Ann Surg 1970;172:902-908. 113. Fleming I, Cooper JS, Henson DE, et al (eds). AJCC Cancer Staging Manual, 5th ed. Philadelphia: Lippincott Williams & Wilkins, 1997, pp 163-170. 114. Neville JA, Welch E, Leffell DJ. Management of nonmelanoma skin cancer in 2007. Nat Clin Pract Oncol 2007;4:462-469. 115. Rowe DE, Carroll RJ, Day CL Jr. Mohs surgery is the treatment of choice for recurrent (previously treated) basal cell carcinoma. J Dermatol Surg Oncol 1989;15:424-431. 116. Morrison WH, Garden AS, Ang KK. Radiation therapy for nonmelanoma skin carcinomas. Clin Plast Surg 1997;24:719-729. 117. Wilder RB, Margolis LW. Cancer of the skin. In Leibel SA, Phillips TL (eds): Textbook of Radiation Oncology, 2nd ed. ­Philadelphia: Lippincott Williams & Wilkins, 2004, pp ­1483-1501. 118. Rowe DE, Carroll RJ, Day CL Jr. Long-term recurrence rates in previously untreated (primary) basal cell carcinoma: implications for patient follow-up. J Dermatol Surg Oncol 1989;15:315-328. 119. Fitzpatrick PJ, Thompson GA, Easterbrook WM, et al. Basal and squamous cell carcinoma of the eyelids and their treatment by radiotherapy. Int J Radiat Oncol Biol Phys 1984;10:449-454. 120. Petrovich Z, Parker RG, Luxton G, et al. Carcinoma of the lip and selected sites of head and neck skin. A clinical study of 896 patients. Radiother Oncol 1987;8:11-17.

121. Lovett RD, Perez CA, Shapiro SJ, et al. External irradiation of epithelial skin cancer. Int J Radiat Oncol Biol Phys 1990;19:235-242. 122. Shimm DS, Wilder RB. Radiation therapy for squamous cell carcinoma of the skin. Am J Clin Oncol 191;14:383-386. 123. Wilder RB, Shimm DS, Kittelson JM, et al. Recurrent basal cell carcinoma treated with radiation therapy. Arch Dermatol 1991;127:1668-1672. 124. Janse GT, Westbrook KC. Cancer of the skin. In Suen JY, Myers EN (eds): Cancer of the Head and Neck. New York: Churchill Livingstone, 1981, pp 212-241. 125. Preston DS, Stern RS. Nonmelanoma cancers of the skin. N Engl J Med 1992;327:1649-1662. 126. Fredricks S. External ear malignancy. Br J Plast Surg 1956;9:  136-160. 127. Metcalf PB Jr. Carcinoma of the pinna. N Engl J Med 1954;  251:91-95. 128. Schewe EJ Jr, Pappalardo C. Cancer of the external ear. Am J Surg 1962;104:753-755. 129. Avila J, Bosch A, Aristizabal S, et al. Carcinoma of the pinna. Cancer 1977;40:2891-2895. 130. Petrovich Z, Kuisk H, Langholz B, et al. Treatment of carcinoma of the skin with bone and/or cartilage involvement. Am J Clin Oncol 1988;11:110-113. 131. Ang KK. Altered fractionation trials in head and neck cancer. Semin Radiat Oncol 1998;8:1-15. 132. Taylor BW Jr, Brant TA, Mendenhall NP, et al. Carcinoma of the skin metastatic to parotid area lymph nodes. Head Neck 1991;13:427-433. 133. Audet N, Palme CE, Gullane PJ, et al. Cutaneous metastatic squamous cell carcinoma to the parotid gland: analysis and outcome. Head Neck 2004;26:727-732. 134. Karakousis CP, Balch CM, Urist MM, et al. Local recurrence in malignant melanoma: long-term results of the multi-­institutional randomized surgical trial. Ann Surg Oncol 1996;3:446-452. 135. Harwood AR. Conventional fractionated radiotherapy for  51 patients with lentigo maligna and lentigo maligna melanoma. Int J Radiat Oncol Biol Phys 1983;9:1019-1021. 136. Milton GW, Shaw HM, McCarthy WH, et al. Prophylactic lymph node dissection in clinical stage I cutaneous malignant melanoma: results of surgical treatment in 1319 patients. Br J Surg 1982;69:108-111. 137. Veronesi U, Adamus J, Bandiera DC, et al. Stage I melanoma of the limbs: immediate versus delayed node dissection. Tumori 1980;66:373-396. 138. Balch CM, Soong SJ, Bartolucci AA, 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:  255-266. 139. Cascinelli N, Morabito A, Santinami M, et al. Immediate or delayed dissection of regional nodes in patients with melanoma of the trunk: a randomised trial. Lancet 1998;351:793-796. 140. Morton DL, Wen DR, Wong JH, et al. Technical details of intraoperative lymphatic mapping for early stage melanoma. Arch Surg 1992;127:392-399. 141. Lang PG Jr. Malignant melanoma. Med Clin North Am 1998;­  ­82:1325-1358. 142. Morton DL, Thompson JF, Cochran AJ, et al. Sentinel-node biopsy or nodal observation in melanoma. N Engl J Med 2006;355:1307-1317. 143. Singletary SE, Balch CM, Urist MM, et al. Surgical treatment of primary melanoma. In Balch CM, Houghton AN, Milton GW, et al (eds): Cutaneous Melanoma, 2nd ed. Philadelphia: JB Lippincott, 1992, pp 269-274. 144. Byers RM. The role of modified neck dissection in the treatment of cutaneous melanoma of the head and neck. Arch Surg 1986;12:1338-1341. 145. Cooper JS, Chang WS, Oratz R, et al. Elective radiation therapy for high-risk malignant melanomas. Cancer J 2001;7:498-502. 146. Balch CM, Soong SJ, Murad TM, et al. A multifactorial analysis of melanoma. IV. Prognostic factors in 200 melanoma patients with distant metastases (stage III). J Clin Oncol 1983;1:126-134. 147. Overgaard J. Radiation therapy of malignant melanoma. Int J Radiat Oncol Biol Phys 1980;6:41-44.

6  The Skin 148. Overgaard J, Overgaard M, Vejby Hansen P, et al. Some factors of importance in the radiation treatment of malignant melanoma.  Radiother Oncol 1986;5:183-192. 149. Strauss A, Dritschilo A, Nathanson L, et al. Radiation therapy of malignant melanoma: an evaluation of clinically used fractionation schemes. Cancer 1981;47:1262-1266. 150. Sause WT, Cooper JS, Rush S, et al. Fraction size in external beam radiation therapy in the treatment of melanoma. Int J Radiat Oncol Biol Phys 1991;20:429-432. 151. Bonnen MD, Ballo MT, Myers JN, et al. Elective radiotherapy provides regional control for patients with cutaneous melanoma of the head and neck. Cancer 2004;100:383-389. 152. Kirkwood JM, Strawderman MH, Ernstoff MS, et al. Interferon alfa-2b adjuvant therapy of high-risk resected cutaneous melanoma: the Eastern Cooperative Oncology Group trial EST 1684. J Clin Oncol 1996;14:7-17. 153. Grob JJ, Dreno B, de la Salmoniere P, et al. Randomised trial of interferon alpha-2a as adjuvant therapy in resected primary melanoma thicker than 1-5 mm without clinically detectable node metastases. Lancet 1998;351:1905-1910. 154. Mays SR, Nelson BR. Current therapy of cutaneous melanoma. Cutis 1999;63:293-298. 155. Tsao H, Atkins MB, Sober AJ. Management of cutaneous melanoma. N Engl J Med 2004;351:998-1012. 156. Rusthoven JJ. The evidence for tamoxifen and chemotherapy as treatment for metastatic melanoma. Eur J Cancer 1998;34:  S31-S36. 157. Wolchok JD, Weber JS, Houghton AN, et al. Melanoma ­vaccines. In Balch CM, Houghton AN, Sober AJ, eds: Cutaneous ­ Melanoma. 4th ed, St Louis: Quality Medical, 2003,  pp 645-656. 158. Wilder RB, Harari PM, Graham AR, et al. Merkel cell ­carcinoma. Improved locoregional control with postoperative radiation therapy. Cancer 1991;68:1004-1048. 159. Pilotti S, Rilke F, Bartoli C, et al. Clinicopathologic correlations of cutaneous neuroendocrine Merkel cell carcinoma. J Clin  Oncol 1988;6:1863-1873. 160. Kokoska GR, Kokoska MS, Collins BT, et al. Early aggressive treatment for Merkel cell carcinoma improves outcome. Am J Clin Oncol 1997;174:688-693. 161. Ott MJ, Tanabe KK, Gadd MA, et al. Multimodality management of Merkel cell carcinoma. Arch Surg 1999;134:388-392. 162. Feun LG, Savaraj N, Legha SS, et al. Chemotherapy for metastatic Merkel cell carcinoma. Review of the M.D. Anderson Hospital’s experience. Cancer 1988;62:683-685. 163. Wynne CJ, Kearsley JH. Merkel cell tumor: a chemosensitive skin cancer. Cancer 1988;62:28-31.

157

164. Grosh WW, Giannone L, Hande KR, et al. Disseminated Merkel cell tumor. Treatment with systemic chemotherapy. Am J Clin Oncol 1987;10:227-230. 165. Poulsen M, Rischin D, Walpole E, et al. High-risk Merkel cell carcinoma of the skin treated with synchronous carboplatin/ etoposide and radiation: Trans-Tasman Radiation Oncology Group Study TROG 96: 07. J Clin Oncol 2003;21:4371-4376. 166. Poulsen MG, Rischin D, Porter I, et al. Does chemotherapy improve survival in high-risk stage I and II Merkel cell carcinoma of the skin? Int J Radiat Oncol Biol Phys 2006;64:114-119. 167. Fenig E, Lurie H, Klein B, et al. The treatment of advanced Merkel cell carcinoma. A multimodality chemotherapy and radiation therapy treatment approach. J Dermatol Surg Oncol 1993;19:860-864. 168. Bichakjian CK, Lowe L, Lao CD, et al. Merkel cell carcinoma: critical review with guidelines for multidisciplinary management. Cancer 2007;110:1-12. 169. Lee WR, Mendelhall WM, Parsons JT, et al. Radical radiotherapy for T4 carcinoma of the skin of the head and neck: a multivariate analysis. Head Neck 1993;15:320-324. 170. Joseph J, Adler JR, Cox RS, et al. Linear accelerator-based stereotactic radiosurgery for brain metastases: the influence of number of lesions on survival. J Clin Oncol 1996;14:1085-1092. 171. Ang KK, Garden AS. Radiotherapy for Head and Neck Cancers: Indications and Techniques, 3rd ed. Philadelphia: Lippincott Williams & Wilkins, 2006. 172. Ang K. Altered fractionation in head and neck cancer. Semin Radiat Oncol 1998;8:230-236. 173. Carter RL, Foster CS, Dinsdale EA, et al. Perineural spread by squamous carcinomas of the head and neck: a morphological study using antiaxonal and antimyelin monoclonal antibodies.  J Clin Pathol 1983;36:269-275. 174. Barrett TL, Greenway HT Jr, Massullo V, et al. Treatment of ­basal cell carcinoma and squamous cell carcinoma with perineural ­invasion. In Callen JP, Dahl MV, Golitz LE, et al (eds): Advances in Dermatology. St Louis: CV Mosby, 1993, pp 277-306. 175. Cottel WI. Perineural invasion by squamous-cell carcinoma.  J Dermatol Surg Oncol 1982;8:589-600. 176. Traenkle HL, Mulay D. Further observations on late radiation necrosis following therapy of skin cancer.Arch Dermatol 1960;81:  908-913. 177. Williams PC, Hendry JH. The RBE of megavoltage photon and electron beams. Br J Radiol 1978;51:220. 178. Boland J. Human skin reaction. Br J Radiol 1957;30:351-352. 179. Wang CC. Cancer of the skin. In Wang CC (ed): Radiation Therapy for Head and Neck Neoplasms: Indications, Techniques, and Results. Chicago: Year Book Medical, 1990, p 85.

Advances in the Treatment of Head and Neck Cancer

7

K. Kian Ang ALTERED FRACTIONATION REGIMENS HIGH-PRECISION RADIATION THERAPY COMBINATION OF RADIATION THERAPY WITH CHEMOTHERAPY Primary Therapy for Locally Advanced Head and Neck Squamous Cell Cancer Adjuvant Treatment of Locally Advanced Head and Neck Squamous Cell Cancer

MOLECULAR BIOMARKERS AND THERAPEUTICS SUMMARY

A variety of neoplasms with widely different natural histories can arise in the head and neck area, a relatively small body region essential for basic physiologic functions and social interactions. Depending on the location, size, and pattern of spread, head and neck tumors can cause various degrees of structural deformities and functional handicaps. Treatment of head and neck tumors can also induce additional disfigurements and malfunctions, worsening quality of life. Consequently, head and neck cancers are challenging to manage. Refinement in surgical resection and reconstructive techniques and advances in radiation therapy planning and delivery technology yield good outcomes for most patients with early head and neck cancer. Unfortunately, for more locally advanced tumors, the results of radiation therapy alone or combinations of surgical resection and preoperative or postoperative radiation therapy are rather poor in terms of disease control or preservation of organ function, or both. Consequently, the search continues for better treatment approaches. This quest along with the simplicity of clinical evaluation and well-characterized patterns of relapse make head and neck carcinomas ideal models for testing the relative efficacy of novel therapy concepts and modalities. Scores of clinical trials have been conducted to test the relative efficacy of altered radiation fractionation regimens, combinations of irradiation with chemotherapy, and the combination of irradiation with molecular therapeutics (see Chapter 5). Most of the randomized trials conducted focus on patients with squamous cell carcinoma (SCC) of the oral cavity, oropharynx, larynx, and hypopharynx. Several treatment regimens have been shown to improve the success rate in achieving locoregional tumor control, survival, or organ preservation. Consequently, the standard of care for intermediate-stage to locally advanced head and neck squamous cell carcinoma (HNSCC) has gradually changed over the past decade. Because most randomized clinical trials have had rather broad eligibility criteria, their results may be applicable to a variety of primary tumor sites. To facilitate cross-referencing and minimize repetition in subsequent chapters in this section, some key findings are summarized in this introductory chapter.

Altered Fractionation Regimens Radiobiologic concepts derived from more than 2 decades of integrated laboratory and clinical investigations led to the conception of two classes of new radiation therapy ­regimens for the treatment of cancer. These altered fractionation regimens are referred to as hyperfractionated and accelerated fractionation schedules. The biologic rationale for modifying fractionation schedules is presented in ­Chapter 1. Briefly, in hyperfractionated schedules, the total dose is increased, the dose per fraction is significantly reduced, the number of dose fractions is increased, and overall time is relatively unchanged. Hyperfractionation exploits the differences in fractionation sensitivity between tumors and normal tissues manifesting late morbidity. Decreasing the fraction size reduces damage to late-responding normal tissues relative to the damage to tumors. In accelerated fractionation schedules, the overall ­treatment duration is reduced, and the number of dose fractions, total dose, and dose per fraction are unchanged or reduced, depending on the extent of overall time reduction. Although radiation therapy can be accelerated in a variety of permutations, for simplicity, the existing schedules can be conceptually grouped into two categories—accelerated fractionation with or without total dose reduction. The goal of accelerated fractionation is to reduce the impact of tumor proliferation as a cause of radiation therapy failure. Altered fractionation regimens have been extensively ­tested in patients with intermediate to locally advanced ­HNSCC. Bourhis and colleagues1 conducted a thorough meta-analysis that included updates of individual patient data. A total of 6515 patients enrolled in 15 phase III ­trials were included in the analysis. The main primary tumor sites were the oropharynx (3079 patients [44%]) and larynx (2377 patients [34%]), and most patients (5221 [74%]) had stage III or IV tumors. The length of follow-up ranged from 4 to 10 years, with a median of 6 years. The treatment regimens tested were considered in three categories: hyperfractionated irradiation and accelerated fractionation with or without dose reduction. Table 7-1 summarizes the benefit of altered fractionation compared with conventional fractionation schedules on different outcome end points.1 On the whole, altered fractionation yielded a significant survival

161

162

PART 3  Head and Neck

TABLE 7-1.   Absolute Benefit and Hazard Ratios for Altered Fractionation versus Conventional Fractionation for Intermediate-Stage to Locally Advanced Head and Neck Squamous Cell Cancer

End Point

Hyperfractionation

Accelerated ­ rac­tionation without F Dose Reduction

Accelerated ­Fractionation with Dose Reduction

Overall

P Value

Overall survival

+3.4%*

0.003

+8.2%

+2.0%

+1.7%

Locoregional control

+6.4%

<0.0001

+9.4%

+7.3%

+2.3%

Improvement at 5 Years

Hazard Ratios Total death

0.92 (0.86-0.97)

0.003

0.78 (0.69-0.89)

0.97 (0.89-1.05)

0.94 (0.84-1.05)

Cancer death

0.88 (0.83-0.94)

0.0002

0.78 (0.68-0.90)

0.91 (0.83-1.00)

0.93 (0.83-1.05)

Local relapse

0.77 (0.71-0.83)

<0.0001

0.75 (0.63-0.89)

0.74 (0.67-0.83)

0.83 (0.71-0.96)

Regional relapse

0.87 (0.79-0.97)

0.01

0.83 (0.66-1.03)

0.90 (0.77-1.04)

0.87 (0.72-1.06)

Locoregional ­relapse

0.82 (0.77-0.88)

<0.0001

0.76 (0.66-0.89)

0.79 (0.72-0.87)

0.90 (0.80-1.02)

Metastatic relapse

0.97 (0.82-1.15)

0.75

1.09 (0.76-1.58)

0.93 (0.74-1.19)

0.95 (0.68-1.32)

*An 8% reduction in the risk of dying. Modified from Bourhis J, Overgaard J, Audry H, et al. Hyperfractionated or accelerated radiotherapy in head and neck cancer: a meta­analysis. Lancet 2006;368:843-854.

­ enefit relative to conventional fractionation (P = .003). b Highly significant reductions were also observed in cancerrelated death (P = .0002) and local tumor failure (P < .0001). These benefits were significantly greater for younger patients than for older patients (P = .007 for test for trend). Overall, the magnitude of the survival benefit was greater in the hyperfractionated group than in the two accelerated-fractionation groups (P = .02). However, the investigators emphasized that the populations included in the three groups were dissimilar because more patients with early-stage or laryngeal cancer were included in the accelerated-without-dose-reduction group. Relapses in that group could be effectively salvaged, as was shown in the Radiation Therapy Oncology Group (RTOG) larynx preservation trial2,3 and therefore had little impact on the survival end point. In the only phase III trial in which hyperfractionation and accelerated fractionation by concomitant boost were tested simultaneously against conventional fractionation, both regimens were found to yield effects of similar magnitude.4 On the basis of evidence generated from numerous randomized trials, many centers have adopted altered fractionation as the current standard of care for the treatment of patients with intermediate-stage (i.e., T2N0-1 or exophytic T3N0-1) HNSCC or those with locally advanced HNSCC who are not eligible for or decline chemotherapy.

High-Precision Radiation Therapy Advances in computerized radiation therapy planning and delivery technology have opened the possibility to increasingly conform irradiation to irregular tumor target volumes.

Details of the technical development and the foundation for clinical application of these advances are presented in Chapter 2. Intensity-modulated radiation therapy (IMRT) is particularly interesting for the treatment of HNSCC because of its ability to reduce radiation doses to the many critical normal tissues that surround tumors, thereby diminishing morbidity. The efficacy of IMRT in sparing parotid glands, thereby diminishing the severity of radiation-induced permanent xerostomia in selected patients, has been amply demonstrated during the past decade.5 Early single-institution studies on the role of IMRT in the management of carcinoma of the nasopharynx and oropharynx have yielded exciting results. In one such study, patients with nasopharyngeal carcinoma were given IMRT alone or, for those with locally advanced carcinoma, in combination with concurrent cisplatin and adjuvant cisplatin plus fluorouracil.6 In that series of 67 patients, at a median follow-up interval of 31 months, the 4-year estimates of local progression-free, locoregional progression-free, distant metastasis-free, and overall survival rates were 98%, 97%, 66%, and, 88%, respectively. The worst acute toxicity experienced was grade 1 or 2 in 51 patients (76%), grade  3 in 15 patients (22%), and grade 4 in one patient (2%). The worst late effects were grade 1 in 20 patients (30%), grade 2 in 15 patients (22%), grade 3 in 7 patients  (10%), and grade 4 in 1 patient (2%). Xerostomia was less pronounced after IMRT than after conventional three­dimensional conformal radiation therapy and showed ­improvement over time. At 3 months after IMRT, 8% of patients had no dry mouth, 28% had grade 1 xerostomia, and 64% had grade 2 xerostomia. Of the 41 patients evaluated at 2 years, 66% had no dry mouth, 32% had grade 1 xerostomia, and only 1 patient had grade 2 xerostomia.

7  Advances in the Treatment of Head and Neck Cancer

In another series of 74 patients with oropharyngeal carcinoma reported by Chao and colleagues,7 14 received IMRT alone, 17 had IMRT combined with cisplatin-based chemotherapy, and 43 underwent surgery followed by postoperative IMRT. At a median follow-up time of 33 months, the 4-year estimated rates of locoregional control, distant metastasis-free survival, disease-free survival, and overall survival were 87%, 90%, 81%, and, 87%, respectively. Late xerostomia was reported in 41 patients (grade 1 in 32 and grade 2 in 9). Late skin toxicity was observed in three patients (two grade 1 and one grade 2), mucositis in three patients (all grade 1), and trismus in three patients. One report of an early series of 51 patients with oropharyngeal carcinoma treated with IMRT between October 2000 and June 2002 at The University of Texas  M. D. Anderson Cancer Center8 indicated that at a median follow-up time of 45 months, the 2-year rates of locoregional control, recurrence-free, and overall survival were 94%, 88%, and 94%, respectively. Although a substantial proportion of patients treated with IMRT required a gastric feeding tube (40%), its use was brief; only 10% of patients still required the feeding tube 6 months after treatment, and all patients were tube-free after 1 year. Only three ­patients had chronic difficulty swallowing. Inspired by these encouraging single-institution data, several multi-institutional trials were launched to address the role of IMRT in the treatment of head and neck carcinomas. Two such trials were introduced by the RTOG (i.e., trial 0022 for T1-2N0-1 oropharyngeal carcinoma and 0225 for nasopharyngeal carcinoma) after IMRT training, credentialing, and quality assessment procedures were established at qualifying treatment centers. The first analysis of the oropharyngeal carcinoma trial9 showed that only  3 (4.5%) of 67 patients developed locoregional recurrence after a median follow-up time of 1.6 years. At a median ­follow-up time of 13 months, 20% of patients reported grade 2 xerostomia and 1.5% grade 3 xerostomia, representing a significant reduction from the 89% calculated from the RTOG Head and Neck Database for similar ­patients not treated with IMRT.9 Increasing numbers of centers are implementing IMRT for the treatment of oropharyngeal carcinoma, nasopharyngeal carcinoma, sinonasal cancer, and thyroid neoplasms because of its ability to spare normal tissues (particularly the parotid glands and neural structures, including the optic apparatus) and to improve dose distribution to concave target volumes adjacent to critical organs. The available findings for IMRT for tumors at each anatomic site are presented in subsequent chapters in this textbook. Further progress is anticipated to more fully extend the benefit of this sophisticated technology. Refinements in target delineation will result from advances in topographic and biologic tumor imaging, quantification of intrafraction and interfraction anatomic variations due to motion and changes in tumor and normal tissue volume over the course of radiation therapy, and modifications of treatment plans to account for these variables. Although the results obtained with IMRT are encouraging, the observation that most recurrences originate from the high-dose irradiation region indicates that radiation dose escalation alone can improve outcome in only a subset of patients. Further advances in the treatment of solid tumors will likely come through application of new knowledge about

163

tumor biology, as exemplified by translational research addressing the role of the epidermal growth factor receptor (EGFR) in tumor progression and as a target for therapeutic intervention, as described subsequently and in Chapter 5.

Combining Radiation Therapy with Chemotherapy The strategy of combining chemotherapy with radiation therapy has been even more extensively investigated in patients with HNSCC than has altered fractionation. The biologic rationale for combining the two modalities is discussed in Chapter 4. Findings from clinical trials and their impact in changing the standard of care for HNSCC are summarized in the remainder of this section.

Primary Therapy for Locally Advanced Head and Neck Squamous Cell Cancer The Meta-Analysis of Chemotherapy in Head and Neck Cancer Collaborative Group undertook an extensive examination of the effects of various combined treatments on several end points.10 Project investigators obtained updated patient data from 63 randomized trials enrolling a total of 10,741 patients between January 1, 1965, and December 31, 1993. Most patients had carcinoma of the oral cavity, oropharynx, hypopharynx, or larynx. The median followup time was 5.9 years. In analyses of the whole study population, a small but highly significant benefit was found from adding chemotherapy to locoregional therapy, corresponding to a 4% improvement in absolute survival rates at 2 and 5 years (Table 7-2).10 No compelling evidence of survival benefit was detected from adding adjuvant or neoadjuvant chemotherapy, with the exception of neoadjuvant cisplatin plus fluorouracil, which yielded a significant improvement in survival (hazard ratio, 0.88; 95% confidence interval [CI], 0.79 to 0.97) (see Table 7-2). In contrast, concurrent chemotherapy conferred a highly significant benefit, resulting in an 8% improvement in absolute survival rate at 5 years or a 19% reduction in the risk of dying. However, heterogeneity was also greatest in the concurrent chemoradiotherapy category. These findings should be interpreted with caution, particularly when toxicity and cost-benefit ratio are considered in addition to survival. For example, the finding that chemotherapy had a decreasing effect on survival with increasing age should be kept in mind. Unfortunately, the data collected did not allow comprehensive analysis of the effect of chemotherapy on the patterns of treatment failure. Nevertheless, the general impression was that the benefit from chemotherapy resulted primarily from improving locoregional control. In that same meta-analysis,10 data were analyzed separately for patients with locally advanced laryngeal and hypopharyngeal carcinoma who had been randomly assigned to receive radical surgery and adjuvant radiation therapy (i.e., control group) or neoadjuvant cisplatin plus fluorouracil followed by radiation therapy (for responders) or radical surgery and adjuvant radiation therapy (for nonresponders) (i.e., chemotherapy group). Pooled data showed a trend  (P = .1) toward worse survival (6% lower absolute 5-year rate) in the chemotherapy group Notably, the types of first event were different between the two groups—twice as

164

PART 3  Head and Neck

TABLE 7-2.   Absolute Benefit and Hazard Ratios from Adding Chemotherapy to Locoregional Treatment for Locally Advanced Head and Neck Squamous Cell Cancer P Values

Absolute Benefit

Trial Category

Hazard Ratios (95% CI)

Chemo Effect

Heterogeneity

At 2 Years

At 5 Years

All adjuvant ­chemotherapy

0.98 (0.85-1.19)

0.74

0.35

+1%

+1%

All neoadjuvant chemotherapy

0.95 (0.88-1.01)

0.10

0.38

+2%

+2%

Neoadjuvant ­cisplatin + ­fluorouracil

0.88 (0.79-0.97)

0.05

Concurrent ­chemotherapy

0.81 (0.76-0.88)

<0.0001

<0.0001

+7%

+8%

Total

0.90 (0.85-0.94)

<0.0001

<0.0001

+4%

+4%

Modified from Pignon JP, Bourhis J, Domenge C, et al. Chemotherapy added to locoregional treatment for head and neck squamous-cell carcinoma: three meta-analyses of updated individual data. Lancet 2000;355:949-955.

many locoregional recurrences but fewer metastases were experienced in the chemotherapy groups. Death rates unrelated to cancer were similar in the two groups. Other published findings from phase III trials2,11-18 have confirmed the concept that chemotherapy given concurrently with radiation yields locoregional control and survival rates that are superior to those for radiation alone for patients with locally advanced HNSCC. On the basis of these findings, many centers have adopted concurrent radiation therapy and chemotherapy as the nonsurgical standard of care for the treatment of patients with locally advanced carcinoma of the oropharynx, larynx, and hypopharynx who are medically fit to receive chemotherapy. However, the best choice of radiation-chemotherapy regimen is still being debated. In 2008, the regimen that had been tested most extensively was the combination of conventionally fractionated radiation therapy (70 Gy in 35 fractions over 7 weeks) given with cisplatin. In early ­trials, the cisplatin was given in a dose of 100 mg/m2 during weeks 1, 4, and 7 of radiation therapy. Unfortunately, the systemic and mucosal toxicities of high-dose, intermittent cisplatin regimens such as this are rather severe; approximately one third of patients were not able to receive the final dose. Three more trials have shown a benefit in terms of locoregional control or survival, or both, from standard or altered fractionation combined with other concurrent single-agent cisplatin regimens (e.g., five doses of 20 mg/ m2 over 5 consecutive days, or four doses of 25 mg/m2 over 4 sequential days, during weeks 1, 4, and 716,19; or  6 mg/m2 per day, 5 days per week during the 7-week radiation period17). With the resurgence of interest in neoadjuvant chemotherapy with taxane-containing regimens, three randomized trials have been conducted for patients with locally advanced HNSCC. One phase III study assessed the efficacy of a neoadjuvant regimen consisting of three courses of paclitaxel, cisplatin, and fluorouracil (175 mg/m2 of paclitaxel on day 1, 100 mg/m2 of cisplatin on day 2, and 500 mg/m2 per day of fluorouracil for 5 days) compared with three courses of cisplatin and fluorouracil (100 mg/m2 of cisplatin on day 1 and 1000 mg/m2 per day of fluorouracil for 5 days), both followed by radiation therapy (70 Gy in

7 weeks) with concurrent cisplatin (100 mg/m2 on days 1, 22, and 43).20 The taxane-containing regimen produced higher complete response rates (33% versus 14%, P < .001) and a trend toward better overall survival (2-year rate of 66.5% versus 53.6%, P = .063). In terms of adverse events, neutropenia and alopecia were more severe in the paclitaxel group, whereas mucositis, nausea, and vomiting were more severe in the cisplatin-fluorouracil group. The efficacy of another taxane-based neoadjuvant regimen (docetaxel, cisplatin, and fluorouracil [TPF]) relative to that of cisplatin and fluorouracil has been tested in two phase III trials. The European Organisation for Research and Treatment of Cancer (EORTC) trial 2497121 compared four cycles of either docetaxel (75 mg/m2 on day 1), cisplatin (75 mg/m2 on day 1), and fluorouracil (750 mg/ m2 per day for 5 days) or cisplatin (100 mg/m2 on day 1) plus fluorouracil (1000 mg/m2 for 5 days), both followed by radiation therapy (in conventional or altered fractionations to a total dose of 66 to 74 Gy). The TAX 324 study22 evaluated three cycles of docetaxel (75 mg/m2 on day 1), cisplatin (100 mg/m2 on day 1), and fluorouracil (1000 mg/m2 per day for 4 days) against three cycles of cisplatin (100 mg/m2 on day 1) plus fluorouracil (1000 mg/m2 for 5 days), both followed by radiation therapy with weekly carboplatin (at a dose corresponding to an area under the curve of 1.5). Results are summarized in Table 7-3.21,22   The EORTC trial revealed more severe leukopenia, neutropenia, and alopecia in the taxane-containing group but more severe hearing loss and death from toxic effects in the cisplatin-fluorouracil group. However, because the control regimens studied are not the accepted standard of care for patients with locally advanced HNSCC, the value of neoadjuvant docetaxel, cisplatin, and fluorouracil needs further investigation. In summary, despite 3 decades of clinical research, many scientific questions related to combinations of radiation and chemotherapy remain unanswered, such as whether cisplatin has additional benefit when added to altered fractionation instead of conventional fractionation; whether newer cytotoxic agents are more effective when given concurrently with radiation therapy; and whether neoadjuvant docetaxel, cisplatin, and fluorouracil ­chemotherapy

7  Advances in the Treatment of Head and Neck Cancer

165

TABLE 7-3.   Efficacy of Neoadjuvant Docetaxel, Cisplatin, and Fluorouracil versus Cisplatin and Fluorouracil Followed by Chemoradiation for Locally Advanced Head and Neck Squamous Cell Cancer EORTC 24971 (TAX-323)

TAX-324

End Points

TPF

PF

P Value

Response rate (%)

68

54

0.006

Median progression-free survival (mo)

11.0

8.2

Time to disease progression (mo)

10.5

Median (mo)

18

Hazard ratio

0.73 (0.56-0.94)

TPF

PF

P Value

0.007

36

13

0.004

7.8

0.003

NR

14

0.002

14

0.005

71

30

0.006

Overall survival 0.70 (0.54-0.90)

Survival rate (%) 2-Year

43

32

67

55

3-Year

37

26

62

48

0.002

Distant metastasis — rate (%)

—

5

9

0.14

EORTC, European Organisation for Research and Treatment of Cancer; NR, not reached; PF, cisplatin and fluorouracil; TPF, docetaxel (­Taxotere), cisplatin, and fluorouracil. Data from Vermorken JB, Remenar E, van Herpen C, et al. Cisplatin, fluorouracil, and docetaxel in unresectable head and neck cancer. N Engl J Med 2007;357:1695-704; Posner MR, Hershock DM, Blajman CR, et al. Cisplatin and fluorouracil alone or with docetaxel in head and neck cancer. N Engl J Med 2007;357:1705-1715.

can further improve the outcome of concurrent radiation and chemotherapy. Several studies are being conducted to resolve some of these questions.

Adjuvant Treatment of Locally Advanced Head and Neck Squamous Cell Cancer The value of adding concurrent chemotherapy to postoperative adjuvant radiation therapy has also been investigated. In the first such trial, patients with histologically proven extracapsular extension of nodal disease were randomly assigned to receive postoperative radiation therapy (65 to 70 Gy  given in 1.7 to 2.0 Gy per fraction for 5 days per week) alone or combined with a bolus injection of 50 mg of cisplatin, given weekly during postoperative radiation therapy for a total of 7 to 9 doses.23 The results of this rather small trial showed significantly higher rates of overall survival  (P < .01), ­disease-free survival (P < .02), and locoregional control (P = .05) in favor of the radiation-chemotherapy regimen. Two larger trials of postoperative radiation and chemotherapy were undertaken by the RTOG24 and EORTC.25 Both studies randomly assigned patients with high-risk ­surgical-pathologic features to undergo surgery with conventionally fractionated postoperative radiation therapy (60 to 66 Gy in 2 Gy per fraction) alone or in combination with cisplatin (100 mg/m2 every 3 weeks for three cycles). In the RTOG trial,24 radiation and chemotherapy significantly reduced the risk of locoregional recurrence compared with radiation alone (hazard ratio, 0.61; P = .01) but did not improve overall survival. In contrast, the EORTC trial25 showed significant improvements in progression-free

s­ urvival (hazard ratio, 0.75; P = .04) and overall survival (hazard ratio, 0.70; P = .02) in addition to improvements in locoregional control. Both trials found that the addition of cisplatin did not affect the incidence of distant metastases but did increase the rate of grade 3 or higher adverse events (from 34% to 77% [P < .001] in the RTOG trial and from 21% to 41% [P = .001] in the EORTC trial). Pooled data from both trials were analyzed to determine which patients are most likely to benefit from intensive postoperative irradiation plus chemotherapy.26 This combined analysis revealed that only patients with extracapsular ­extension and positive surgical margins benefited from radiation and chemotherapy, a finding that further refined the criteria for high-risk disease and the corresponding treatment algorithm. An adjuvant ­ radiation-­chemotherapy regimen is now recommended only for patients with histologically proven extracapsular extension or positive surgical margins, or both. For patients with intermediate-risk disease features (e.g., positive nodes without extracapsular extension, perineural invasion, close margins), the current standard of care is postoperative radiation therapy without chemotherapy. Combinations of radiation therapy with molecular therapeutics are being tested in this subset of patients.

Molecular Biomarkers and Therapeutics Progress in searching for useful markers for early detection of tumors, estimations of tumor burden, predictions of response to therapy, and monitoring disease progression has

166

PART 3  Head and Neck

been slow. A prototypical marker is prostate-specific ­antigen, which has proven valuable for prostate cancer screening, for prognostic grouping, and for monitoring responses to therapy. Unfortunately, equivalent markers have not been discovered for most other solid tumors, but studies of head and neck carcinomas have generated optimism. The biologic rationale and emerging findings regarding molecular markers are described in detail in Chapter 5. Work leading to the establishment of a new therapy option for HNSCC is briefly summarized in the following paragraphs. Promising prognostic biomarkers for HNSCC studied during the past few years include TP53, the EGFR and one of its ligands, transforming growth factor-α (TGF-α), and cyclin D1.27-29 The prognostic value of EGFR in particular for HNSCC has been corroborated in several studies.30-35 Our own analysis of HNSCC specimens from patients enrolled in an RTOG phase III trial4 and randomly assigned to receive standard radiation therapy32 revealed no ­correlation between EGFR expression (measured with an image-analysis–based immunohistochemical assay) and  T category, N category, American Joint Committee on Cancer stage, and recursive partitioning analysis36 groupings  (r = −0.07 to +0.17). Rates of overall survival and disease-free survival for patients with high-EGFR-expressing carcinomas (i.e., tumors that expressed more than the median amount of EGFR) were much lower (P = .0006 and P = .0016, respectively) and the locoregional relapse rate was much higher  (P = .0031) than the rates for patients with low-EGFR­expressing carcinomas. However, rates of distant metastasis were no different in the two groups (P = .96). Multivariate analysis showed that EGFR expression was a strong, independent predictor of survival and of locoregional relapse. These observations, along with impressive results of preclinical studies, motivated many investigators to undertake a multinational randomized trial to assess the efficacy of adding an anti-EGFR antibody, cetuximab, to radiation therapy for the treatment of locally advanced HNSCC.37 The findings from this pivotal trial demonstrated that the addition of only eight doses of cetuximab, starting 1 week before and continuing during the radiation therapy period, led to significant improvements in locoregional control, progression-free survival, and overall survival (Table 7-4)37 without increasing the incidence or severity of common radiation-induced side effects (Table 7-5),37 particularly mucositis and dysphagia. However, cetuximab induced an acneiform rash needing attentive care and occasionally infusion reactions requiring discontinuation of the agent. Collectively, coordinated laboratory and clinical studies such as these have generated great excitement because in addition to producing a new, less toxic therapy option for locally advanced HNSCC, they established the proof of principle that modulating a signaling pathway that is perturbed in tumors can lead to selective tumor sensitization to radiation therapy, thereby improving the therapeutic index. Such selective enhancement of tumor response has not been accomplished by combining radiation with traditional chemotherapy agents such as cisplatin or taxanes. However, a weakness of this trial is that the control group received radiation therapy alone rather than concurrent radiation plus chemotherapy, which was recently established as the new standard of care. Based on tolerability, however, cetuximab plus radiation therapy may represent the treatment of choice for patients with intermediate-stage 

TABLE 7-4.   Efficacy of Adding Cetuximab to Radiation Therapy for Locally Advanced Head and Neck Squamous Cell Cancer RT + Ce- Hazard RT Alone tuximab Ratio (95% (n = 208) (n = 212) CI)

P Value

Median duration (mo)

14.9

24.4

0.68 (0.520.89)

0.005

Rate at 2 years (%)

41

50

Median duration (mo)

12.4

17.1

0.70 (0.540.90)

0.006

Rate at 2 years (%)

37

46

Median duration (mo)

29.3

49.0

0.74 (0.570.97)

0.03

Rate at 3 years (%)

45

55

Rate at 1 year (%)

10

8

Rate at 2 years (%)

17

16

End Points Locoregional Control

ProgressionFree Survival

Overall Survival

Distant Metastasis*

*Cumulative incidence. CI, confidence interval; RT, radiation therapy. Modified from Bonner JA, Harari PM, Giralt J, et al. Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. N Engl J Med 2006;354:567-578.

disease when the relatively good outcomes do not justify the toxicity of concurrent radiation plus chemotherapy. It may also provide an effective, better-tolerated alternative to radiation plus chemotherapy for patients with more extensive HNSCC who are not eligible for chemotherapy or are unlikely to complete radiation and chemotherapy as planned. Further investigations are seeking to elucidate the relative efficacy of cetuximab plus radiation versus radiation plus chemotherapy and to identify patient subpopulations that are appropriate for radiation therapy plus cetuximab relative to radiation plus chemotherapy. Studies are also ongoing to assess the efficacy of combining cetuximab with neoadjuvant therapy or with concurrent radiation therapy plus chemotherapy.

Summary The advances made in laboratory research and clinical trials on head and neck cancer during the past 2 decades have been exciting and gratifying. Progress in molecular ­biology

7  Advances in the Treatment of Head and Neck Cancer

167

TABLE 7-5.   Incidence of Common Adverse Events from Radiation Therapy with or without Cetuximab for Locally Advanced Head and Neck Squamous Cell Cancer Incidence (%) for All Grades*

Incidence (%) for Grades 3-5*

Side Effects

RT

RT + C

P Value

RT

RT + C

P Value

Mucositis

94

93

0.84

52

56

0.44

Dysphagia

63

65

0.68

30

26

0.45

Weight loss

72

84

0.005

7

11

0.12

Dehydration

19

25

0.16

8

6

0.57

Dermatitis

90

86

0.24

18

23

0.27

Xerostomia

71

72

0.83

3

5

0.32

Acneiform rash

10

87

<0.001

1

17

<0.001

Infusion ­reaction

2

15

<0.001

0

3

0.01

*For radiation therapy with cetuximab, n = 212; for radiation therapy without cetuximab, n = 208. C, cetuximab; RT, radiation therapy. Modified from Bonner JA, Harari PM, Giralt J, et al. Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. N Engl J Med 2006;354:567-578.

techniques has opened research avenues yielding new knowledge about the molecular basis of cellular behavior and mechanisms governing radiation response. Some of this wisdom has already found application in the development of novel therapy strategies that are being tested. The basic and translational research efforts have finally paid off in that the mortality rates for head and neck cancers in the United States have declined since the inception of record keeping. For example, the annual death rate among men from oral cavity and pharyngeal cancers in the United States decreased by an average of 1.9% from 1975 to 1993 and by 3% from 1993 to 2001.38 Clinical research on altered fractionation schedules and various combinations of radiation therapy with chemotherapy has generated firm data leading to changes in the standard of care for patients with intermediate-stage and locally advanced carcinoma of the oral cavity, oropharynx, larynx or hypopharynx, and nasopharynx. The concept of combining radiation therapy with molecular therapeutics has been validated, and the first generation of such therapy, radiation plus cetuximab, has become an option for dayto-day practice. The expertise in the application of IMRT for the treatment of some subsets of patients with HNSCC and in quality assurance and procedural improvements is improving, and proper indications for and the value of IMRT for combined-modality therapy are being assessed. The pace of discovery is anticipated to increase in the coming years. For example, it is reasonable to envision that new approaches for characterizing the molecular profile of various types of cancer will depict their individual virulence, predict their response to therapy, and guide the selection of treatment. The outlook for developing rational therapy strategies aimed at specific molecular targets to prevent malignant transformation or reverse a malignant phenotype is also increasingly optimistic. The insights and new technologies gained from further research should have considerable impact on reducing the mortality rates for head and neck cancers. The fast pace of new discoveries and the large number of research directions make determining what constitutes

“standard therapy” for a variety of patient subsets increasingly complex. When several treatment options can yield approximately the same locoregional tumor control rate, other determinants need to be taken into account in selecting the best treatment for individual patients. These determinants include the cosmetic and functional outcomes acute and long-term morbidity (i.e., quality of life), resource use (i.e., cost), physician expertise, and patient convenience.

REFERENCES   1. Bourhis J, Overgaard J, Audry H, et al. Hyperfractionated or accelerated radiotherapy in head and neck cancer: a meta-analysis. Lancet 2006;368:843-854.   2. Forastiere AA, Goepfert H, Maor M, et al. Concurrent chemotherapy and radiotherapy for organ preservation in advanced laryngeal cancer. N Engl J Med 2003;349:2091-2098.   3. Weber RS, Berkey BA, Forastiere A, et al. Outcome of salvage total laryngectomy following organ preservation therapy: the Radiation Therapy Oncology Group trial 91-11. Arch Otolaryngol Head Neck Surg 2003;129:44-49.   4. Fu KK, Pajak TF, Trotti A, et al. A Radiation Therapy Oncology Group (RTOG) phase III randomized study to compare hyperfractionation and two variants of accelerated fractionation to standard fractionation radiotherapy for head and neck squamous cell carcinomas: first report of RTOG 9003. Int J Radiat Oncol Biol Phys 2000;48:7-16.   5. Eisbruch A, Ten Haken RK, Kim HM, et al. Dose, volume, and function relationships in parotid salivary glands following conformal and intensity-modulated irradiation of head and neck cancer. Int J Radiat Oncol Biol Phys 1999;45:577-587.   6. Lee N, Xia P, Quivey JM, 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.   7. Chao KSC, Ozyigit G, Blanco AI, et al. Intensity-modulated radiation therapy for oropharyngeal carcinoma: impact of tumor volume. Int J Radiat Oncol Biol Phys 2004;59:43-50.   8. Garden AS, Morrison WH, Wong PF, et al. Disease-control rates following intensity-modulated radiation therapy for small primary oropharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2007;67:438-444.

168

PART 3  Head and Neck

  9. Eisbruch A, Harris J, Garden AS, et al. Phase II multi-­institutional study of IMRT for oropharyngeal cancer (RTOG 00-22): early results [abstract 79]. Presented at the 48th annual meeting of the American Society for Therapeutic and Radiation Oncology, Philadelphia, PA, November 5-9, 2006. Int J Radiat Oncol Biol Phys 2006;66(Suppl 1):S46-S47. 10. Pignon JP, Bourhis J, Domenge C, et al. Chemotherapy added to locoregional treatment for head and neck squamous-cell carcinoma: three meta-analyses of updated individual data. Lancet 2000;355:949-955. 11. Adelstein DJ, Li Y, Adams GL, et al. An Intergroup phase III comparison of standard radiation therapy and two schedules of concurrent chemoradiotherapy in patients with unresectable squamous cell head and neck cancer. J Clin Oncol 2003;21:  92-98. 12. Al-Sarraf M, LeBlance M, Shanker PG, et al. Chemoradiotherapy versus radiotherapy in patients with advanced nasopharyngeal cancer: phase III randomized intergroup study 0099. J Clin Oncol 1998;16:1310-1317. 13. Brizel DM, Albers ME, Fisher SR, et al. Hyperfractionated ­irradiation with or without concurrent chemotherapy for locally advanced head and neck cancer. N Engl J Med 1998;338:17981804. 14. Budach V, Stuschke M, Budach W, et al. Hyperfractionated accelerated chemoradiation with concurrent fluorouracil-­mitomycin is more effective than dose-escalated hyperfractionated accelerated radiation therapy alone in locally advanced head and neck cancer: final results of the Radiotherapy Cooperative Clinical Trials Group of the German Cancer Society 95-06 prospective randomized trial. J Clin Oncol 2005;23:1125-1135. 15. Denis F, Garaud P, Bardet E, et al. Late toxicity results of the GORTEC 94-01 randomized trial comparing radiotherapy with concomitant radiochemotherapy for advanced-stage oropharynx carcinoma: comparison of LENT/SOMA, RTOG/EORTC, and NCI-CTC scoring systems. Int J Radiat Oncol Biol Phys 2003;55:93-98. 16. Huguenin P, Beer KT, Allal A, et al. Concomitant cisplatin significantly improves locoregional control in advanced head and neck cancers treated with hyperfractionated radiotherapy. J Clin Oncol 2004;22:4665-4673. 17. Jeremic B, Shibamoto Y, Milicic B, et al. Hyperfractionated radiation therapy with or without concurrent low-dose daily cisplatin in locally advanced squamous cell carcinoma of the head and neck: a prospective randomized trial. J Clin Oncol 2000;18:1458-1464. 18. Wendt TG, Grabenbauer GG, Rodel CM, et al. Simultaneous radiochemotherapy versus radiotherapy alone in advanced head and neck cancer: a randomized multicenter study. J Clin Oncol 1998;16:1318-1324. 19. Wee J, Tan EH, Tai BC, 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;23:6730-6738. 20. Hitt R, Lopez-Pousa A, Martinez-Trufero J, et al. Phase III study comparing cisplatin plus fluorouracil to paclitaxel, cisplatin, and fluorouracil induction chemotherapy followed by chemoradiotherapy in locally advanced head and neck cancer. J Clin Oncol 2005;23:8636-8645. 21. Vermorken JB, Remenar E, van Herpen C, et al. Cisplatin, fluorouracil, and docetaxel in unresectable head and neck cancer.  N Engl J Med 2007;357:1695-1704. 22. Posner MR, Hershock DM, Blajman CR, et al. Cisplatin and fluorouracil alone or with docetaxel in head and neck cancer.  N Engl J Med 2007;357:1705-1715.

23. Bachaud JM, Cohen-Jonathan E, Alzieu C, et al. Combined postoperative radiotherapy and weekly cisplatin infusion for locally advanced head and neck carcinoma: final report of a randomized trial. Int J Radiat Oncol Biol Phys 1996;36:999-1004. 24. Cooper JS, Pajak TF, Forastiere AA, et al. Postoperative concurrent radiotherapy and chemotherapy for high-risk squamouscell carcinoma of the head and neck. N Engl J Med 2004;350:  1937-1944. 25. Bernier J, Domenge C, Ozsahin M, et al. Postoperative irradiation with or without concomitant chemotherapy for locally ­advanced head and neck cancer. N Engl J Med 2004;350:19451952. 26. Bernier J, Cooper JS, Pajak TF, et al. Defining risk levels in locally advanced head and neck cancers: a comparative analysis of concurrent postoperative radiation plus chemotherapy trials of the EORTC (#22931) and RTOG (#9501). Head Neck 2005;27:843-850. 27. Smith BD, Haffty BG. Molecular markers as prognostic factors for local recurrence and radioresistance in head and neck squamous cell carcinoma. Radiat Oncol Invest 1999;7:125-144. 28. Salesiotis AN, Cullen KJ. Molecular markers predictive of response and prognosis in the patient with advanced squamous cell carcinoma of the head and neck: evolution of a model ­beyond TNM staging. Curr Opin Oncol 2000;12:229-239. 29. Quon H, Liu FF, Cummings BJ. Potential molecular prognostic markers in head and neck squamous cell carcinomas. Head Neck 2001;23:147-159. 30. Maurizi M, Almadori G, Ferrandina G, et al. Prognostic significance of epidermal growth factor receptor in laryngeal squamous cell carcinoma. Br J Cancer 1996;74:1253-1257. 31. Grandis J, Melhem M, Gooding W, et al. Levels of TGF-α and EGFR protein in head and neck squamous cell carcinoma and patient survival. J Natl Cancer Inst 1998;90:824-832. 32. Ang KK, Berkey BA, Tu X, et al. Impact of epidermal growth factor receptor expression on survival and pattern of relapse in patients with advanced head and neck carcinoma. Cancer Res 2002;62:7350-7356. 33. Eriksen JG, Steiniche T, Askaa J, et al. The prognostic value of epidermal growth factor receptor is related to tumor differentiation and the overall treatment time of radiotherapy in squamous cell carcinomas of the head and neck. Int J Radiat Oncol Biol Phys 2004;58:561-566. 34. Bentzen SM, Atasoy BM, Daley FM, et al. Epidermal growth factor receptor expression in pretreatment biopsies from head and neck squamous cell carcinoma as a predictive factor for a benefit from accelerated radiation therapy in a randomized controlled trial. J Clin Oncol 2005;23:5560-5567. 35. Psyrri A, Yu Z, Weinberger PM, et al. Quantitative determination of nuclear and cytoplasmic epidermal growth factor receptor expression in oropharyngeal squamous cell cancer by using automated quantitative analysis. Clin Cancer Res 2005;11:58565862. 36. Cooper J, Farnum N, Asbell S, et al. Recursive partitioning analysis of 2,105 patients treated in RTOG studies of head and neck cancer. Cancer 1996;77:1905-1911. 37. Bonner JA, Harari PM, Giralt J, et al. Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. N Engl J Med 2006;354:567-578. 38. Jemal A, Clegg LX, Ward E, et al. Annual report to the nation on the status of cancer, 1975-2001, with a special feature regarding survival. Cancer 2004;101:3-27.

8

The Salivary Glands Adam S. Garden ANATOMY Parotid Glands Submandibular and Sublingual Glands RADIATION EFFECTS ON NORMAL SALIVARY GLANDS EPIDEMIOLOGY OF SALIVARY GLAND CANCER CLINICAL PRESENTATION Parotid Gland Tumors Submandibular Gland Tumors Minor Salivary Gland Tumors STAGING PATHOLOGY AND NATURAL HISTORY Benign Tumors Malignant Tumors

OUTCOMES TREATMENT OPTIONS Surgery Postoperative Radiation Therapy Primary Radiation Therapy RADIATION TECHNIQUES Parotid Gland Tumors Submandibular Gland Tumors Minor Salivary Gland Tumors Coverage in Cases of Spread along Nerves Locally Advanced and Inoperable Tumors

The salivary glands consist of major and minor glands. The major glands consist of paired parotid, submandibular, and sublingual glands. The minor salivary glands number in the hundreds and reside in the oral cavity, pharynx, and paranasal sinuses. The primary purpose of the major salivary glands is to secrete, store, and excrete saliva. Flow rates of saliva vary among individuals, but the average person excretes approximately 1 L of saliva per day. The salivary glands are highly radiosensitive, and many patients undergoing radiation treatment for head and neck tumors can develop xerostomia, a side effect that can lead to secondary oral complications and can significantly affect quality of life. Neoplasms of the salivary glands are uncommon, and most tumors are benign. Primary tumors of the parotid contribute the largest proportion of salivary gland tumors, representing approximately two thirds of all salivary gland neoplasms. The submandibular glands are the next most common site; together with the parotid glands, they account for 90% of salivary gland neoplasms. Sublingual primary malignancies are rare. In this chapter, the anatomy of the salivary glands and the effect of radiation on the normal glands are described first, followed by brief descriptions of the epidemiology, clinical presentation, staging, and pathology of salivary gland tumors. The treatment options, including radiation, are reviewed.

i­ntercalated ducts, and small striated ducts. The ductal system continues in the connective tissue as larger striated ducts, interlobular ducts, and a main excretory duct ­ (Fig. 8-1). The acini consist of serous cells, mucinous cells, or ­mixtures of both types of cells. The parotid glands are predominantly composed of serous cells, whereas the submandibular glands are mostly mucinous cells. The excretory ducts are lined with epithelial cells. The precise cell of origin of salivary gland neoplasms is not universally agreed on. According to one widely believed hypothesis, neoplasms arise from stem cells originating in the intercalated ducts or excretory ducts. Salivary gland carcinomas can be poorly differentiated or better differentiated, depending on where they develop along the differentiation pathway. Acinic cell carcinoma, adenoid cystic carcinoma, and mixed tumors (benign and malignant) mainly arise from stem cells of intercalated ducts, whereas mucoepidermoid carcinomas mainly arise from excretory ducts.

Myoepithelial cell Intercalated cells

Anatomy Embryologically, the major salivary glands are derived from the ectoderm. Epithelial buds originate from the oral cavity and eventually differentiate into the glands, which consist of ducts and secretory end pieces. On histologic examination, the salivary glands appear as lobules separated by ­connective tissue. The lobules consist primarily of acini,

Cells lining acinus

Striated cells

Excretory duct cells

FIGURE 8-1.  Tubuloalveolar structure of a salivary gland.

169

170

PART 3  Head and Neck

Parotid Glands Zygomatic arch Parotid gland

Accessory parotid gland Parotid duct Masseter muscle Submandibular gland

The parotid glands are pyramidal. Each gland lies below the zygomatic arch, in front of and below the external acoustic meatus, and in front of the mastoid process. The inferior border of the gland rests between the angle of the mandible and the anterior border of the sternocleidomastoid muscle. The anterior edge is somewhat C-shaped because of the gland’s relationship with the ascending ramus of the mandible. The gland abuts the ramus, extending anterolaterally around the edge of the bone and then onto the lateral edge of the masseter muscle to lie between this muscle and the skin. The parotid duct, which is 5 cm long, emerges from the parotid gland through the masseter muscle to open opposite the upper second molar. Anteromedially, the parotid gland extends between the medial edge of the ascending ramus and the medial pterygoid muscle. The posterior medial edge of the parotid gland abuts the styloid process. The medial edge of the gland lies close to the retropharyngeal space and may even come in contact with the lateral pharyngeal wall (Figs. 8-2 and 8-3). The parotid glands are divided into superficial and deep lobes by the facial nerve and its branches.

Submandibular and Sublingual Glands FIGURE 8-2.  Anatomy of the parotid gland and its relation to adjacent structures.

The submandibular glands lie mainly under the cover of the horizontal ramus of the mandible. Medially, they are related to the mylohyoid, hyoglossus, and digastric muscles. The deep portion of each gland lies inferior to the lingual nerve and superior to the hypoglossal nerve (Figs. 8-4 and 8-5). The sublingual glands lie in the floor of the mouth,

Anterior

#

Maxillary sinus

** *

Mandible

Parotid gland Mastoid bone

A

Posterior

B

FIGURE 8-3.  Axial drawing (A) and computed tomographic image (B) show the parotid gland curving around the posterior ascending ramus of the mandible. Notice the relationship of the gland with the masseter (hatch mark), the pterygoid (double asterisks) muscles, and the parapharyngeal space (single asterisk).

8  The Salivary Glands

below the mucous membrane and above the muscular floor, anterior to the submandibular glands.

Radiation Effects on Normal Salivary Glands The two main cells of the major salivary glands, serous and mucinous, differ in their sensitivity to radiation injury. Despite their slow turnover rate, the serous cells are highly radiosensitive. Stephens and colleagues1 studied the effects of radiation on the acinar serous cells of rhesus monkeys, which are quite similar to the salivary gland cells of humans. Injury to irradiated serous cells was expressed within 24 hours of irradiation, with cells exhibiting destruction resulting from apoptosis and an acute inflammatory response. In submandibular glands, the selective destruction of serous acini was particularly apparent within the normal mix of cells. Although acinar cells can recover and repopulate, the typical therapeutic radiation doses eventually damage the ducts, ductal stem cells, and the mucinous and supporting stromal and vascular cells. This damage often results in xerostomia and fibrosis. Swelling occurs and pain occasionally is experienced during the first week of radiation, particularly when doses per fraction are relatively high, but these symptoms are almost always self-limiting. The primary effect of radiation on the salivary glands is xerostomia. Because xerostomia affects quality of life and can increase the risk of dental caries and oral infections, early involvement of a dental oncologist is essential when patients are to undergo radiation to the head and neck that includes the major salivary glands. Defining a threshold dose (TD) in clinical situations is complicated because most tumoricidal doses exceed the radiation threshold of the salivary glands. Estimates of TD5/5 and TD50/5 (i.e., threshold doses at which 5% and 50% of the population, respectively, experience sequelae 5 years after treatment) vary widely in the literature and often have been based on anecdotal observations rather than

171

large prospective trials. This variability also results from inconsistent or unclear definitions of xerostomia or different ways of measuring salivary flow rates. Estimates of TD5/5 range from 30 to 50 Gy; estimates of TD50/5 range from 30 to 70 Gy.2-6 Evidence from the literature suggests that threshold doses are at the lower end of this range. Several investigators have found the TD100/5 to be 50 Gy, with all or nearly all patients who receive this or a higher dose to all major salivary glands experiencing xerostomia.2 Xerostomia develops relatively quickly and consistently. After treatment with 10 Gy (given in 5 fractions over a period of 1 week), dryness of the mouth is evident subjectively, and objective flow measurements demonstrate reductions in flow rate of about 50%. Additional radiation continues to diminish flow rate; at the completion of a full course of radiation therapy, flow rates are only 10% of their original baseline values. Dreizen and colleagues,7 documenting changes in salivary flow rates over time, suggested that salivary flow levels show little or no recovery after having reached this nadir. The late recovery process is controversial. Leslie and Dische8 argued that sparing part of the gland or maintaining the dose below threshold levels confers a reasonable expectation of recovery. In their studies, flow rates declined to less than 20% of their baseline values when more than 80% of the gland was in the radiation field; in contrast, flow rates could be maintained at 80% or more of baseline levels if less than 30% of the gland was in the radiation field.8 This observation led to exploratory use of conformal and intensity-modulated radiation therapy (IMRT) to partially spare the parotid glands and thereby reduce the risk of xerostomia. Eisbruch and colleagues9 used IMRT to define a

g Mandible Masseter muscle Pterygoid muscle

Parotid gland

v m

Tongue Submandibular gland

FIGURE 8-4.  Coronal drawing of the submandibular gland shows its relation to adjacent structures.

FIGURE 8-5.  Axial computed tomographic image of the submandibular gland (g) shows its relation in the neck to the sternocleidomastoid muscle (m) and vessels (v).

172

PART 3  Head and Neck

mean dose of 24 Gy or less to one gland as being sufficient to maintain salivary flow rates after treatment near the pretreatment baseline levels. Using a more complex model, Chao and colleagues10 determined that stimulated saliva secretion is reduced exponentially at a rate of 4%/Gy of mean dose to the parotid, with a threshold of 32 Gy. The use of amifostine, a radioprotectant preferentially taken up by the salivary glands, may also decrease the incidence of radiation-induced xerostomia. In one randomized trial, the incidence of grade 2 or higher xerostomia was reduced from 57% in controls to 35% in patients given amifostine.11 Amifostine is approved for postoperative treatment of squamous cell carcinoma of the head and neck. Xerostomia can be difficult to manage. After the symptom develops, effective therapies are limited. Saliva substitutes can be helpful for some patients; others respond well to induction or overstimulation of residual functioning glands by cholinergic agonists such as pilocarpine.12

Epidemiology of Salivary Gland Cancer Salivary gland neoplasms account for approximately 3% to 7% of head and neck tumors. Because most salivary gland neoplasms are benign, they most likely account for 1% to 2% of head and neck malignancies. Unlike most cases of head and neck cancer, salivary gland carcinomas are not associated with tobacco exposure. A strong association is evident between parotid carcinomas and prior radiation exposure. Alcohol use and exposure to hair dye are also linked to these carcinomas.13 No predilection is evident for a particular sex or age group, but the median age of patients with salivary gland neoplasms is younger than that of patients with more common malignancies of the head and neck. Younger patients (<40 years old) tend to be female, whereas older patients are more often male. A controversial association is the finding in some studies of a higher incidence of salivary gland tumors among women with breast cancer. This finding, considered together with the fact that salivary gland tumors and breast tumors arise in organs consisting of lobules and ducts, has led some investigators to use breast cancer therapies, mainly hormonal therapies, for patients with salivary gland ­malignancies. A general rule of thumb for salivary gland neoplasms is that the size of the gland is directly proportional to the likelihood that a neoplasm will develop but is inversely proportional to the likelihood that a neoplasm will be malignant. At most, one third of parotid gland tumors are malignant, but approximately two thirds of minor salivary gland tumors are malignant. The rate of malignancy for submandibular gland tumors is between the rates for the other two sites.

Clinical Presentation Parotid Gland Tumors More than 80% of parotid tumors develop in the superficial lobe (Fig. 8-6). The most typical presentation is a painless mass located over or behind the ascending ramus of the mandible. It is unusual for patients with parotid tumors to present with pain, and pain is a poor prognostic sign. Deep

FIGURE 8-6.  Axial computed tomographic image shows an adenoid cystic carcinoma of the superficial lobe of the left parotid gland.

lobe tumors (Fig. 8-7) can be indolent and can be present for a long time before symptoms appear. As they grow, they invade the retropharyngeal space. Deep lobe tumors are usually detected when the mass fills the pharyngeal space enough to interfere with swallowing or speech. Facial nerve paralysis is an uncommon finding that usually indicates a malignant tumor and is thought to correlate with a poorer prognosis. Other cranial nerve palsies are rare. Although deep lobe tumors can invade the retropharyngeal space, involvement of the cranial nerves that traverse this space is unusual. Lymph node metastases develop in 10% to 15% of patients with parotid malignancies (Fig. 8-8). Metastases are most common in patients with higher-grade tumors, particularly high-grade mucoepidermoid carcinomas or adenocarcinomas. The parotid glands drain first to lymph nodes within the superficial gland and overlying superficial fascia and then to the jugulodigastric lymph nodes in the neck.

Submandibular Gland Tumors Submandibular gland tumors present as painless masses in the submandibular triangle. Malignant submandibular gland tumors, particularly neurotropic adenoid cystic carcinoma, may manifest with nerve palsy and with ­sensory or motor changes in the tongue, depending on the nerve involved. Lymph node metastases develop in 10% to 15% of cases. The submandibular glands drain first to the submandibular nodes and then to nodes in the jugular chain.

Minor Salivary Gland Tumors Minor salivary gland tumors usually manifest as painless masses that may become symptomatic on further growth. Paranasal sinus tumors may manifest with obstructive breathing symptoms, and base-of-tongue and palatal

8  The Salivary Glands

A

173

B

FIGURE 8-7.  A, Axial computed tomographic image shows a benign pleomorphic adenoma of the deep lobe of the left parotid gland. B, Coronal magnetic resonance image shows a poorly differentiated carcinoma of the deep lobe of the right parotid gland.

t­ umors may interfere with swallowing. Because about two thirds of minor salivary gland tumors are malignant and because a very high proportion of those tumors are adenoid cystic carcinomas, symptoms related to cranial nerve deficits are not uncommon. Whether lymph node metastases occur depends greatly on the site of origin of the tumor and the richness of lymphatics at that site. Paranasal sinus tumors metastasize to lymph nodes in less than 5% of ­cases, whereas tongue and floor-of-mouth tumors metastasize in more than 35% of cases.14 Overall, neck metastases develop in 10% to 15% of patients with minor salivary gland tumors.

Staging Major salivary gland tumors are staged using the ­American Joint Committee on Cancer (AJCC) tumor-node-­metastasis (TNM) staging system, in which the primary tumor is staged on the basis of size and local extension (Box 8- 1).3,15 Larger size of the primary tumor and invasion into local tissues, particularly major nerves (i.e., facial, hypoglossal, or lingual), correlate with a poorer prognosis.16 The type of tissue invaded distinguishes T4a disease from T4b disease. Minor salivary gland tumors are staged by using the systems for squamous cell cancer of the respective sites. The nodal staging system is the same as that used for other head and neck mucosal sites (other than the nasopharynx).15

Pathology and Natural History Benign Tumors The most common benign tumor of the parotid gland is the benign mixed (or pleomorphic) adenoma, which accounts for 75% of benign parotid tumors and more than 50% of all parotid neoplasms. Most commonly, benign mixed adenomas develop in the superficial lobe of the parotid gland and consist of a mixture of ductal, myoepithelial, and mesenchymal cells. These tumors have a very low recurrence rate after surgery, but when they do recur, they can give

FIGURE 8-8.  High-grade mucoepidermoid carcinoma of the right parotid gland with lymph node metastases.

rise to carcinomas and must be examined very carefully microscopically. The second most common benign parotid tumor is Warthin’s tumor (papillary cystadenoma lymphomatosum), which represents 2% to 6% of all parotid tumors. It is rare for Warthin’s tumor to undergo malignant transformation. Less common benign tumors include lymphoepithelial lesions and monomorphic adenomas, such as basal cell adenomas, myoepitheliomas, oncocytomas, and salivary duct adenomas.

174

PART 3  Head and Neck

BOX 8-1.  American Joint Committee on Cancer Staging System for Salivary Gland Tumors Primary Tumor (T) TX Primary tumor cannot be assessed T0 No evidence of primary tumor T1 Tumor 2 cm or less in greatest dimension without extraparenchymal extension* T2 Tumor more than 2 cm but not more than 4 cm in greatest dimension without ­extraparenchymal extension* T3 Tumor more than 4 cm and/or having ­extraparenchymal extension* T4a Tumor invades skin, mandible, ear canal, and/or facial nerve T4b Tumor invades skull base and/or pterygoid plates and/or encases carotid artery Regional Lymph Nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Metastasis in a single ipsilateral lymph node, 3 cm or less in greatest dimension N2 Metastasis in a single ipsilateral lymph node, more than 3 cm but not more than 6 cm in greatest dimension; or in multiple ipsilateral lymph nodes, none more than 6 cm in greatest dimension; or in bilateral or contralateral lymph nodes, none more than 6 cm in greatest dimension N2a Metastasis in single ipsilateral lymph node more than 3 cm but not more than 6 cm in greatest dimension N2b Metastasis in multiple ipsilateral lymph nodes, none more than 6 cm in greatest dimension N2c Metastasis in bilateral or contralateral lymph nodes, none more than 6 cm in greatest dimension N3 Metastasis in a lymph node more than 6 cm in greatest dimension Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage Grouping Stage I T1 Stage II T2 Stage III T3 T1 T2 T3 Stage IVA T4a T4a T1 T2 T3 T4a Stage IVB T4b Any T Stage IVC Any T

N0 N0 N0 N1 N1 N1 N0 N1 N2 N2 N2 N2 Any N N3 Any N

M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M1

*Extraparenchymal extension is clinical or microscopic evidence of invasion of soft tissues. Microscopic evidence alone does not constitute extraparenchymal extension for classification purposes. From Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 69-75.

Malignant Tumors Malignant salivary gland tumors are a diverse group with regard to their histologic type, but their prognosis and behavior generally can be predicted by their grade. Highergrade tumors tend to be more invasive locally, are more prone to lymphatic and hematogenous metastases, and even in the case of appropriate surgical treatment, have high local recurrence rates. Mucoepidermoid carcinoma is the most common malignancy of the parotid glands but develops less often in other salivary glands. These tumors exhibit wide variation in their differentiation and behavior, although lower-grade lesions are most common. Mucoepidermoid carcinomas consist of a combination of mucin-producing and epidermoid cells. Mucin-producing cells are more prevalent in well-differentiated lesions than in higher-grade lesions, and high-grade tumors therefore are rarely classified as squamous cell carcinomas. The incidence of nodal and distant metastases and perineural invasion increases with increasing grade of mucoepidermoid carcinoma. Acinic cell carcinomas are the second most common malignant tumors of the parotid gland. These tumors rarely develop in other salivary glands. Acinic cell carcinomas are usually well-differentiated tumors. When treated appropriately, they are associated with an excellent prognosis. Adenoid cystic carcinoma is the most common malignancy of the major and minor salivary glands, with the exception of the parotid gland, for which it ranks in frequency behind mucoepidermoid and acinic cell carcinomas (Table 8-1).16,17 Adenoid cystic carcinomas represent 10% of parotid malignancies and 30% to 50% of malignancies of the other glands. Although histologically these tumors often have a low-grade appearance, they can behave similarly to high-grade lesions. They tend to grow slowly but are notorious for infiltrating adjacent tissue. They often spread along perineural spaces. Some pathologists have classified adenoid cystic carcinomas as having tubular, cribriform, or solid patterns. These patterns have some correlation with prognosis, with the tubular pattern associated with a favorable outcome and the solid pattern often associated with an aggressive course. Another important pathologic prognostic factor is tumor ploidy; aneuploid tumors often metastasize, whereas diploid tumors rarely do.18 Adenocarcinomas, although not rare, represent a minority of malignant salivary gland tumors. They are a heterogeneous group in terms of type and grade. They fall into two well-defined clinicopathologic groups: terminal duct adenocarcinomas, which are low-grade malignancies arising from intercalated ducts that most commonly are found in the palate, and salivary duct adenocarcinomas, which account for 5% to 10% of malignancies, typically metastasize early, and most commonly arise from major glands. Malignant mixed tumors also tend to behave aggressively and must be diligently sought in cases of recurrent benign tumors. The most common malignant mixed tumor is carcinoma ex pleomorphic adenoma, which arises from its benign counterpart and is a high-grade tumor. Less common are true mixed malignant tumors, which are a mix of carcinoma and sarcomatous elements. These are aggressive malignancies. Squamous cell carcinomas represent 1% to 2% of primary parotid malignancies. Histologically, squamous cell tumors of the parotid gland may be difficult to distinguish from metastatic squamous cell carcinoma within an ­intraparotid

8  The Salivary Glands

TABLE 8-1.   Histologic Findings for Primary Malignant Tumors of the Parotid and Submandibular Glands Number of Patients (%) Histologic ­Characteristics

Parotid Gland Tumors

Submandibular Gland Tumors

Mucoepidermoid

55 (36)

15 (18)

Acinic

17 (11)

1 (1)

Adenoid cystic

17 (11)

37 (44)

Adenocarcinoma

15 (10)

9 (10)

Unclassified/ undifferentiated

13 (8)

6 (7)

Salivary duct

13 (8)

0

Myoepithelial

9 (6)

0

Malignant mixed

8 (5)

9 (10)

Squamous/­ adenosquamous

6 (4)

9 (10)

Oncocytic/ basaloid

2 (1)

0

Total

155 (100)

86 (100)

Data from Frankenthaler RA, Luna MA, Lee SS, et al. ­ Prognostic variables in parotid gland cancer. Arch Otolaryngol Head Neck Surg 1991;117:1251-1256; Weber RS, Byers RM, Petit B, et al. ­Submandibular gland tumors. Adverse histologic factors and therapeutic ­ implications. Arch Otolaryngol Head Neck Surg 1990;116:1055-1060.

lymph node, a situation seen most often in the case of primary skin cancer. In the absence of a known skin tumor, squamous cell carcinoma in the parotid gland should be managed as a primary malignancy of the parotid gland.

Outcomes Ten-year survival rates for patients with major salivary gland tumors range from 50% to 70%. Most series16,17,19-22 have found that primary tumor stage (or size) and tumor grade are strong predictors of outcome, with low-grade and small tumors associated with low recurrence rates. Lymph node involvement and involvement of large or eponymous nerves are predictors of disease recurrence and poorer ­survival. Rates of distant metastasis vary from series to series. Metastases develop in approximately 25% to 30% of cases.16,23,24 Lung metastases are most common, accounting for more than one half of distant recurrences. The bones are the second most common site of distant metastases. Although most recurrences happen within 5 years of treatment, late recurrences are possible. In several long-term analyses, the disease-free survival curves flatten but do not reach a plateau past 10 years.23

Treatment Options Surgery Surgery is the treatment of choice for salivary gland tumors. For parotid tumors, surgery at a minimum consists of gross tumor removal through superficial or total parotidectomy,

175

depending on the tumor location and size. Surgery alone is often sufficient for low-grade malignancies. Higher-grade tumors necessitate removal of the lymph nodes in continuity with the gland. In the case of positive lymph nodes, a more complete neck dissection is required. Radiation therapy is effective in controlling microscopic disease, as described in the following sections. The entire facial nerve or its main trunk can often be spared, provided that the nerve is not invaded and that all gross tumor can be removed without removal of the nerve. In the case of very extensive lesions, sacrifice of the facial or auriculotemporal nerve or removal of skin or bone may be required to achieve a gross total removal. Surgical management of submandibular tumors consists of a regional dissection of the submandibular triangle, including the gland and the adjacent submandibular, submental, and facial lymph nodes. Extensive disease may require removal of additional structures, such as muscle, mandible, or a portion of the floor of the mouth. In the case of clinically apparent adenopathy, more extensive lymph node dissection, the details of which are dictated by the extent of nodal disease, is necessary. For minor salivary gland tumors, the location of the primary tumor and the extent of tumor involvement determine the extent of resection.

Postoperative Radiation Therapy The role of radiation in the treatment of salivary gland tumors is most often adjunctive. Before 1965, postoperative radiation therapy was rarely used because salivary gland tumors were thought to be radioresistant; unless inoperable, tumors were treated with surgery only. However, patients with high-grade malignancies were noted to have high local recurrence rates. On the basis of these observations, Fletcher started to use radiation therapy as an adjunctive treatment and found that the addition of postoperative radiation therapy improved local control relative to surgery alone.25 Moreover, control of subclinical disease was no different for epithelial salivary gland malignancies and squamous cell carcinomas.25 In 1975, Guillamondegui and colleagues26 reported the results of surgery alone for patients with parotid cancer and compared those results with outcomes in a later cohort treated with surgery followed by radiation. The investigators concluded that the addition of postoperative radiation therapy greatly improved local control of high-grade malignancies. This favorable experience with postoperative radiation therapy for parotid tumors was extrapolated to the treatment of all salivary gland cancer. In 1976, Tapley and Fletcher27 outlined the indications for postoperative radiation therapy. These have not significantly changed over the years and include gross residual disease, positive or close (<5 mm) margins, high tumor grade, metastasis to parotid lymph nodes, squamous cell subtype, extraglandular involvement, perineural invasion, recurrent disease, and lymph node involvement. Nevertheless, widespread use of adjuvant radiation therapy based on these indications has been only slowly accepted. Among 207 patients with major salivary gland malignancies treated surgically from 1960 through 2004,28 10-year locoregional control rates were found to range from 37% to 63% in cases that involved lymph node metastases, high tumor grade, positive margins, or T3 or T4 disease.

176

PART 3  Head and Neck

TABLE 8-2.   Locoregional Control Rates in a Series of Patients with Salivary Gland Tumors Treated with Surgery and Postoperative Radiation Therapy Study Fu et al, 197729 Tran et al,

198630

Harrison et al,

199031

199032

Glands

Number of Patients

Locoregional Control Rate*

Follow-Up Period

Major

29

86%

Minimum 3 yr

Major

57

75%

30 mo-10 yr

Major

46

73% (actuarial)

Median 5.8 yr

Major

64

92%

NS

199414

Minor

160

82%

Median 110 mo

Parsons et al, 199633

Minor

40

88%

Minimum 2 yr

Terhaard et al, 200434

Both

386

91%

Median 74 mo

Both

160

90%

Median 5 yr

North et al,

Garden et al,

Mendenhall et al,

200535

*All

rates shown are crude except as indicated. NS, not stated.

Outcomes with Surgery and Postoperative Radiation Therapy During the past 3 decades, numerous series have demonstrated improved local control rates in patients with salivary gland tumors treated with postoperative radiation compared with patients treated with surgery alone (Table 8-2).14,29-35 Fu and colleagues29 described a series of patients with major and minor salivary gland tumors treated with surgery alone or surgery and postoperative radiation. In patients with inadequate surgical margins, the addition of postoperative radiation therapy significantly improved local control rates. Simpson and colleagues36 reported improved locoregional control rates in patients with adenoid cystic carcinomas, from less than 15% without radiation therapy to more than 80% with radiation therapy and surgery. In a University of Texas M. D. Anderson Cancer Center series of 198 patients with adenoid cystic carcinoma, 70% of whom had inadequate surgical margins or perineural spread, postoperative radiation resulted in a 10-year local control rate of 86%.37 Locoregional control rates in patients with parotid malignancies treated with surgery and radiation are high— more than 80% in most series. Selected series are summarized in Table 8-3.38-43 Bissett and Fitzpatrick44 reported an improvement in locoregional control with the addition of postoperative radiation therapy to surgical resection of submandibular gland malignancies. The locoregional control rate was 30% with surgery alone, compared with 69% when radiation was added. Weber and colleagues17 similarly described an improvement in locoregional control with the addition of radiation therapy to surgery in patients whose tumors invaded the extraglandular soft tissues. Storey and colleagues45 found the 10-year locoregional control rate to be 88% for 83 patients with submandibular tumors treated with combined surgery and radiation therapy at M. D. ­ Anderson Cancer Center. My group at M. D. Anderson reported a crude local recurrence rate of 12% among patients treated with surgery and radiation for minor salivary gland cancer.14 Two thirds of these recurrences were diagnosed more than 5 years after treatment. Parsons and colleagues33 reported an ­ actuarial

TABLE 8-3.   Locoregional Control Rates in a Series of Patients with Parotid Gland Tumors Treated with Surgery and Postoperative Radiation Therapy Number of Patients

Locoregional Control Rate*

Median ­Follow-Up Period

Theriault et al, 198638

167

75%

10 yr

Spiro et al, 199239

57

87% (actuarial)

66 mo

Sykes et al, 199540

104

68% (actuarial)

51 mo

Garden et al, 199741

166

87%

155 mo

Leverstein et al, 199842

51

80%

90 mo

Renehan et al, 199943

66

85%

12 yr

Study

*All

rates shown are crude except as indicated.

15-year local control rate of 74%, which was similar to the rate in the M. D. Anderson series.

Postoperative Radiation Therapy for Benign ­Pleomorphic Adenomas Benign pleomorphic adenomas are the most common neoplasms of the parotid gland. Like their malignant counterparts, they are treated by surgical resection. The more thorough the surgical procedure, the lower the recurrence rate. A review of several series found that the recurrence rate was more than 20% among patients who underwent simple enucleation, compared with less than 0.5% in patients who underwent total parotidectomy.46 The role of postoperative radiation therapy in the treatment of benign pleomorphic adenomas is ­ controversial, and some investigators believe that radiation therapy

8  The Salivary Glands

TABLE 8-4.   Locoregional Control Rates in a Series of Patients with Benign Pleomorphic Adenomas of the Parotid Gland Treated with Surgery and Postoperative Radiation Therapy Number of Patients

Crude ­Locoregional Control Rate

Follow-Up Period

20

90%

1-26 yr

Samson et 21 al, 199147

90%

Mean 5.9 yr

Barton et 187 al, 199251

95%

Median 14 yr

Liu et al, 199548

55

95%

Median 21.5 yr

Renehan et 51 al, 199652

92%

Median 14 yr

Chen et al, 200653

94%

Median 101 mo

Study Dawson, 198950

34

should be avoided even in patients with incomplete resections and residual microscopic disease. This management algorithm involves waiting for recurrence and managing any recurrence with repeated and more aggressive surgical resections. Samson and colleagues47 have advocated radiation therapy for microscopic disease in patients with benign tumors adherent to but resectable from the ­facial nerve. Indications for postoperative irradiation in a group of 55 patients with benign pleomorphic adenomas treated at the Princess Margaret Hospital48 included multifocality, nerve involvement, tumor “spill,” residual disease (gross or microscopic), and recurrent disease (disease that recurred after the initial surgery and was treated with a second operation). The risk of a third recurrence is high in most series reporting the management of recurrent pleomorphic adenoma.49 In the Princess Margaret Hospital series, control rates among patients with recurrent disease were 81% for patients treated with postoperative radiation compared with 6% for patients treated with surgery alone. Overall control rates for patients with benign pleomorphic adenomas treated with surgery and postoperative radiation range from 90% to 95%.47,48,50-53 Results of several series are summarized in Table 8-4.47,48,50-53 One reason for the reluctance to use radiation therapy is the fear of inducing a second malignant tumor in the remaining gland or in other organs in the head and neck. However, series reviewing the outcome of patients undergoing radiation for benign pleomorphic adenoma have found that the incidence of second primary tumors is negligible.

Primary Radiation Therapy Occasionally, patients with salivary gland tumors present with unresectable disease or cannot tolerate surgery for medical reasons. These patients can be treated with radiation therapy alone. Wang and Goodman54 reported control rates exceeding 80% for unresectable parotid and minor salivary gland tumors treated with radiation alone, but later

177

series show lower rates of disease control with longer follow-up. In a 1996 article, Parsons and colleagues33 reported a control rate of 50% for minor salivary gland tumors treated with radiation alone; in a follow-up study, Mendenhall and others35 described a local control rate at 10 years of 42% for 64 patients with salivary gland ­cancer treated with radiation alone. In both series, these control rates were worse than those achieved after treatment with surgery and postoperative irradiation. However, the investigators argued that for some patients for whom the morbidity of surgery may be unacceptable, radiation is certainly a reasonable and sometimes preferable treatment option. Control rates among such patients depend on the radiation dose and the T category. Terhaard and others55 reported a 5-year control rate of 50% among patients given radiation in doses of 66 Gy or more, compared with 0% among patients given lower doses. Chen and colleagues56 reported similar results, with 10-year actuarial local control rates of 81% for patients receiving more than 66 Gy and 40% for patients receiving lower doses. T category was also important for local control; 10-year local control rates were 81% for patients with T1 or T2 tumors and only 39% for patients with T3 to T4 tumors.56

Radiation Techniques Techniques for the irradiation of salivary gland tumors vary depending on the location of the primary tumor. Because of their lateral locations, parotid gland and submandibular gland tumors often lend themselves to ipsilateral treatment. This approach allows normal tissues, particularly the contralateral salivary glands, to be spared and reduces the incidence of radiation-induced xerostomia. Treatment of minor salivary gland tumors is more individualized and depends greatly on the site of origin. Techniques tend to be similar to those used for squamous cell cancer arising from these sites, although issues regarding coverage of the base of the skull and the neck need to be addressed on a case-by-case basis.

Parotid Gland Tumors Two conventional techniques used for treating parotid tumors involve wedged pairs of two photon beams or a single ipsilateral field using a mixed beam of electrons and photons. Historically, the single ipsilateral field technique has been favored at M. D. Anderson because it treats less normal tissue (Fig. 8-9). However, both techniques can exceed spinal cord tolerance if care is not taken. The posterior field in the wedged pair must be off the spinal cord. The mixed-beam technique, despite the shallow range of electrons, can also exceed cord tolerance if the spinal cord is within a high-depth dose curve. This is particularly true for electron beams of very high energies (18 to 25 MeV) that have 90% dose curves at depths of 4 to 7 cm. A second concern with the wedged-pair technique is that the beam can exit through the contralateral eye if the proper precautions are not taken.

Submandibular Gland Tumors Submandibular gland tumors, like parotid gland tumors, can be treated with a unilateral technique. As we do with parotid tumors, we treat submandibular tumors (Fig. 8-10) by using a single appositional field with electrons only

178

PART 3  Head and Neck

FIGURE 8-9.  This 39-year-old man underwent a superficial parotidectomy in 1982. He had an acinic cell carcinoma that was dissected off the branches of the facial nerve. From October through November 1982, he received 60 Gy delivered to the parotid bed using a mixed-beam technique. He was alive without evidence of disease 15 years after treatment.

FIGURE 8-11.  This 60-year-old woman with adenoid cystic carcinoma of the left submandibular gland underwent resection of her tumor with neck dissection in 1986. Section margins were microscopically positive, and tumor had invaded into the strap muscles and lingual nerve. The initial postoperative portal arrangement was through parallel-opposed fields with a final dose to the base of the skull of 55 Gy. The submandibular bed received 60 Gy. The patient was alive without evidence of disease 13 years after treatment.

is ­particularly true if coverage of the base of the skull is a concern (see “Coverage in Cases of Spread along Nerves”).

Minor Salivary Gland Tumors

FIGURE 8-10.  This 46-year-old man presented with a submandibular mass. Excision in January 1985 revealed an adenoid cystic carcinoma with perineural and lymphatic space invasion. Findings at re-excision and selective neck dissection in February 1985 were negative for disease. From March through April 1985, radiation therapy was given as a 13-MeV electron-beam appositional field covering the submandibular gland bed and a 9-MeV electron-beam field covering the lower neck (i.e., operative bed). The glandular bed received 60 Gy dosed to the 90% isodose line, and the lower neck received 50 Gy dosed to the 100% line. The patient was alive without evidence of disease 14 years after treatment.

or a mixture of photons and electrons. Because the submandibular gland extends medially and sits behind bone, higher energies of electrons may be needed to provide adequate coverage of the glandular bed. Occasionally, even higher-energy electrons may not be sufficient, and then parallel-opposed fields may be desirable (Fig. 8-11). This

Techniques for irradiation of minor salivary gland tumors greatly depend on the primary site of origin. Techniques tend to be similar to those used for squamous cell tumors arising from the same sites: parallel-opposed fields for tongue (Fig. 8-12) and palate lesions, two or three wedged fields for sinus tumors, and ipsilateral fields for buccal ­tumors. Unlike squamous cell carcinoma, perineural spread of minor salivary gland tumors to the base of the skull is of greater concern than is lymphatic spread to the neck. In a review of 160 patients treated with postoperative irradiation for minor salivary gland malignancies,14 the incidence of recurrence in the neck was extremely low (<5%), even for patients who did not undergo elective irradiation of the neck. The incidence of nodal disease at presentation was also low (8%), and nodal disease was detected predominantly in patients with primary lesions in the tongue or the floor of the mouth. Adjunctive or elective treatment of the neck is recommended only if (1) nodal involvement is present, (2) the primary lesion arises from the tongue or the floor of the mouth, or (3) removal of the primary tumor required surgery on the neck. Coverage of the base of the skull is discussed in the next section.

Coverage in Cases of Spread along Nerves Salivary gland tumors, particularly adenoid cystic carcinomas, can spread along nerves. Discovery of such spread at pathologic examination raises the question of how far to

8  The Salivary Glands

179

FIGURE 8-13.  The intensity-modulated radiation therapy plan is shown for an extensive adenoid cystic carcinoma of the right parotid gland necessitating a parotidectomy, partial auriculectomy, and lateral temporal bone resection. Postoperative radiation was to be given to the tumor bed at doses of 60 Gy to the clinical target volume 1 (CTV1; red) and 54 Gy to the CTV2 (green). FIGURE 8-12.  This 51-year-old man presented in September 1994 with an adenoid cystic carcinoma that occupied the entire right base of the tongue and crossed the midline. He underwent resection of the primary tumor with an ipsilateral modified neck dissection. The tongue was reconstructed by using a soft tissue free-flap transfer. Findings at pathologic examination included involvement of the lingual and hypoglossal nerves and positive margins. He was treated with lateral fields to 44 Gy. Fields were reduced off the spinal cord to 50 Gy, and the tongue dose was boosted to 66 Gy. The initial portals were designed to include the base of the skull. The patient was alive without evidence of disease 6 years after treatment.

follow the nerve pathways with the radiation fields. Parotid tumors do not present much controversy; the fields need to be large enough to cover the facial nerve from the stylo­ mastoid foramen, but this does not involve a significant change in field size. The question is more difficult when the nerve involved or at risk is relatively distant from the gland, as in the common case of a submandibular gland tumor with lingual nerve involvement. Covering the perineural pathways to the base of skull in this case would result in a substantial increase in field size. At M. D. Anderson, the general policy is to cover the nerve pathways proximally to the skull base if major nerve involvement is present, as documented by clinical examination, findings at surgery, or findings on pathologic examination. Following these guidelines has resulted in a recurrence rate at the base of the skull of less than 2%.13 In the case of microscopic perineural invasion of small nerves, the lingual nerve is covered as it tracks proximally to join the other branches of the mandibular nerve. Conversely, in the case of major nerve involvement that is tracked during surgery to the respective foramina, both the skull base and the intracranial component of the nerve to the trigeminal ganglion are covered. In most situations, IMRT has replaced conventional radiation techniques. Although no clinical comparisons have

been done, IMRT planning for parotid neoplasms (Fig. 8-13) can reduce the dose to normal tissues to lower levels than is possible with conventional or three-dimensional approaches.57 The advantage of intensity-modulated techniques tends to be even greater for other salivary gland tumors (Figs. 8-14 and 8-15), particularly in reducing the dose to the brain when base-of-skull coverage is required. Depending on the risk of recurrence, we recommend radiation doses of 56 to 66 Gy to postoperative tumor beds. Doses at the lower end of this range may be appropriate for low-grade mucoepidermoid or acinic cell carcinoma, but doses at the higher end are recommended for lesions that are high grade or have close or positive margins. Treatment of intermediate- or high-grade tumors of the major salivary gland may include elective irradiation of level III and IV lymph nodes, with the dose depending on whether adenopathy is suspected or present. Level III or level IV nodes that are not involved and have not been dissected are treated to 50 to 54 Gy. Nodes that are not involved but have been dissected are treated to 56 to 57 Gy. Involved nodes are treated to 60 Gy. Techniques for treating the neck include a matched field that covers the ipsilateral neck. Level III and IV nodes can be covered as targets in IMRT treatment plans (see Fig. 8-13).

Locally Advanced and Inoperable Tumors Several trials58,59 have investigated the use of fast neutrons for the treatment of locally advanced and inoperable salivary gland tumors. Battermann and colleagues60 demonstrated that the relative biologic effectiveness of radiation with neutrons was higher in metastatic adenoid cystic carcinomas than in other metastatic tumors. It is believed that cells with long cell cycles are more sensitive to neutrons than photons, and salivary gland tumors with many cells remaining in the G0 phase of the cell cycle are selectively sensitive to neutrons.

180

PART 3  Head and Neck

A B

C FIGURE 8-14.  The intensity-modulated radiation therapy plan is shown for an unresectable, high-grade adenocarcinoma of the left submandibular gland that spread along the lingual nerve to involve the masticator space, the third division of the trigeminal nerve, and the skull base. The clinical target volume 1 (CTV1) included all sites thought to harbor gross disease and was treated to 70 Gy (red). Dose distributions are shown on axial slices through the submandibular bed (A), masticator space (B), and brain (C).

FIGURE 8-15.  The intensity-modulated radiation therapy plan is shown for adenoid cystic carcinoma after a palatectomy that left diffuse gross disease. The clinical target volume 1 (CTV1) is shown in red; the remaining palate and nasopharynx were included in the CTV1 and were to be treated to a higher dose.

The role of neutrons has been studied in a prospective, randomized cooperative trial61 conducted by the Radiation Therapy Oncology Group in the United States and the Medical Research Council in Great Britain. This study evaluated whether neutrons were more advantageous than photons for inoperable salivary gland tumors. Neutrons proved to be superior with respect to control of primary disease. The 2-year locoregional control rates were 67% for patients treated with neutrons and 17% for patients treated with photons. However, the high rate of distant metastases in the entire group canceled out any potential survival advantage. Ultimately, the study evaluated only 25 patients. The University of Washington has one of the largest experiences in using neutrons to treat salivary gland tumors. Investigators there report relatively high control rates for their patients with parotid gland and minor salivary gland tumors. Douglas and colleagues62 reported the use of neutrons for gross tumors of major salivary glands and described a 6-year actuarial locoregional control rate of 59%. A separate study evaluating the use of neutrons for adenoid cystic carcinomas revealed an overall 5-year actuarial ­local

8  The Salivary Glands

control rate of 57% among patients with gross residual disease.63 Doses to be delivered to lesions close to the brain may have to be limited because neutrons are more neurotoxic than photons; Douglas and colleagues62,63 found that control rates were substantially lower when the tumor involved the nasopharynx, cavernous sinus, or base of the skull. However, the development of improved conformal techniques may allow better dosing in these areas. Collectively, these studies suggest that in the uncommon situation of a patient with an inoperable salivary gland tumor, referral to a center with neutron capabilities may be appropriate. In the absence of a large randomized trial, however, it is difficult to interpret the substantial differences in the results of small single-institution series using photons or neutrons. As is true for other head and neck cancers, interest has been expressed in the use of concurrent chemoradiation therapy for inoperable salivary tumors, and preliminary results from Haddad and colleagues64 hint that this approach may yield more favorable outcomes.

REFERENCES   1. Stephens L, Schultheiss T, Price R, et al. Radiation apoptosis of serous acinar cells of salivary and lacrimal glands. Cancer 1991;67:1539-1543.   2. Emami B, Lyman J, Brown A, et al. Tolerance of normal tissue to therapeutic radiation. Int J Radiat Oncol Biol Phys 1991;21:109122.   3. Fajardo LF, Berthrong M, Anderson RE. Salivary glands. In Fajardo LF, Berthrong M, Anderson RE (eds): Radiation Pathology. New York: Oxford University Press, 2001, pp 265-270.   4. Marks J, Davis C, Gottsman V, et al. The effects of radiation on parotid salivary function. Int J Radiat Oncol Biol Phys 1981;7:1013-1019.   5. Mira J, Wescott W, Starcke E, et al. Some factors influencing salivary function when treating with radiotherapy. Int J Radiat Oncol Biol Phys 1981;7:535-541.   6. Rubin P, Casarett G. Clinical Radiation Pathology. Philadelphia: WB Saunders, 1968, pp 242-262.   7. Dreizen S, Brown L, Handler S, et al. Radiation-induced xerostomia in cancer patients. Cancer 1976;38:273-278.   8. Leslie MD, Dische S. The early changes in salivary gland function during and after radiotherapy given for head and neck cancer. Radiother Oncol 1994;30:26-32.   9. Eisbruch A, Ten Haken R, Kim H, et al. Dose, volume, and function relationships in parotid salivary glands following conformal and intensity-modulated irradiation of head and neck cancer. Int J Radiat Oncol Biol Phys 1999;45:577-587. 10. Chao K, Deasy J, Markman J, et al. A prospective study of salivary function sparing in patients with head-and-neck cancers receiving intensity-modulated or three-dimensional radiation therapy: initial results. Int J Radiat Oncol Biol Phys 2001;49:907-916. 11. Brizel D, Wasserman T, Henke M, et al. Phase III randomized trial of amifostine as a radioprotector in head and neck cancer. J Clin Oncol 2000;18:3339-3345. 12. LeVeque FG, Montgomery M, Potter D, et al. A multicenter, randomized, double-blind, placebo-controlled, dose-titration study of oral pilocarpine for treatment of radiation-induced xerostomia in head and neck cancer patients. J Clin Oncol 1993;11:1124-1131. 13. Spitz M, Fueger J, Goepfert H, et al. Salivary gland cancer. A case-control investigation of risk factors. Arch Otolaryngol Head Neck Surg 1990;116:1163-1166. 14. Garden AS, Weber RS, Ang KK, et al. Postoperative radiation therapy for malignant tumors of minor salivary glands. Cancer 1994;73:2563-2569. 15. Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 69-75.

181

16. Frankenthaler RA, Luna MA, Lee SS, et al. Prognostic variables in parotid gland cancer. Arch Otolaryngol Head Neck Surg 1991;117:1251-1256. 17. Weber RS, Byers RM, Petit B, et al. Submandibular gland tumors. Adverse histologic factors and therapeutic implications. Arch Otolaryngol Head Neck Surg 1990;116:1055-1060. 18. Luna M, El-Naggar A, Batsakis J, et al. Flow cytometric DNA content of adenoid cystic carcinoma of submandibular gland. Correlation of histologic features and prognosis. Arch Otolaryngol Head Neck Surg 1990;116:1291-1296. 19. Hicks MJ, El-Naggar AK, Flaitz CM, et al. Histocytologic grading of mucoepidermoid carcinoma of major salivary glands in prognosis and survival: a clinicopathologic and flow cytometric evaluation. Head Neck 1995;17:89-95. 20. Spiro RH, Thaler HT, Hicks WF, et al. The importance of clinical staging of minor salivary gland carcinoma. Am J Surg 1991;162:330-336. 21. Spiro RH, Huvos AG. Stage means more than grade in adenoid cystic carcinoma. Am J Surg 1992;164:623-628. 22. Vander Poorten VL, Balm AJ, Hilgers FJ, et al. Prognostic factors for long-term results of the treatment of patients with malignant submandibular gland tumors. Cancer 1999;85:2255-2264. 23. Fordice J, Kershaw C, El-Naggar A, et al. Adenoid cystic carcinoma of the head and neck: predictors of morbidity and mortality. Arch Otolaryngol Head Neck Surg 1999;125:149-152. 24. Spiro R. Salivary neoplasms: overview of a 35-year experience with 2,807 patients. Head Neck Surg 1986;8:177-184. 25. King J, Fletcher G. Malignant tumors of the major salivary glands. Radiology 1971;100:381-384. 26. Guillamondegui OM, Byers RM, Luna MA, et al. Aggressive surgery in treatment for parotid cancer: the role of adjunctive postoperative radiotherapy. AJR Am J Roentgenol 1975;123:49-54. 27. Tapley N, Fletcher G. Malignant tumors of the salivary gland. In Tapley N (ed): Clinical Applications of the Electron Beam. New York: John Wiley & Sons, 1976, pp 141-171. 28. Chen AM, Granchi PJ, Garcia J, et al. Local-regional recurrence after surgery without postoperative irradiation for carcinomas of the major salivary glands: implications for adjuvant therapy. Int J Radiat Oncol Biol Phys 2007;67:982-987. 29. Fu KK, Leibel SA, Levine ML, et al. Carcinoma of the major and minor salivary glands. Cancer 1977;40:2882-2890. 30. Tran L, Sadeghi A, Hanson D, et al. Major salivary gland tumors: treatment results and prognostic factors. Laryngoscope 1986;96:1139-1144. 31. Harrison LB, Armstrong JG, Spiro RH, et al. Postoperative radiation therapy for major salivary gland malignancies. J Surg Oncol 1990;45:52-55. 32. North C, Lee D, Piantadosi S, et al. Carcinoma of the major salivary glands treated by surgery or surgery plus postoperative radiotherapy. Int J Radiat Oncol Biol Phys 1990;18:13191326. 33. Parsons J, Mendenhall W, Stringer S, et al. Management of minor salivary gland carcinomas. Int J Radiat Oncol Biol Phys 1996;35:443-454. 34. Terhaard CH, Lubsen H, Van der Tweel I, 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;26:681-692. 35. Mendenhall WM, Morris CG, Amdur RJ, et al. Radiotherapy alone or combined with surgery for salivary gland carcinoma. Cancer 2005;103:2544-2550. 36. Simpson J, Thawley S, Matsuba H. Adenoid cystic salivary gland carcinoma: treatment with irradiation and surgery. Radiology 1984;151:509-512. 37. Garden A, Weber R, Morrison W, 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;32:619-626. 38. Theriault C, Fitzpatrick PJ. Malignant parotid tumors. Prognostic factors and optimum treatment. Am J Clin Oncol 1986;9:510516. 39. Spiro I, Wang C, Montgomery W. Carcinoma of the parotid gland. Cancer 1992;1:2699-2705.

182

PART 3  Head and Neck

40. Sykes A, Logue J, Slevin N, et al. An analysis of radiotherapy in the management of 104 patients with parotid carcinoma. Clin Oncol 1995;7:16-20. 41. Garden AS, El-Naggar A, Morrison W, et al. Postoperative radiotherapy for malignant tumors of the parotid gland. Int J Radiat Oncol Biol Phys 1997;37:79-85. 42. Leverstein H, van der Waal J, Tiwari R, et al. Malignant epithelial parotid gland tumours: analysis and results in 65 previously untreated patients. Br J Surg 1998;85:1267-1272. 43. Renehan A, Gleave E, Slevin N, et al. Clinico-pathological and treatment-related factors influencing survival in parotid cancer. Br J Cancer 1999;80:1296-1300. 44. Bissett RJ, Fitzpatrick PJ. Malignant submandibular gland tumors. A review of 91 patients. Am J Clin Oncol 1988;11:46-51. 45. Storey M, Garden A, Morrison W, et al. Postoperative radiotherapy for malignant tumors of the submandibular gland. Int J Radiat Biol Oncol Phys 2001;51:952-958. 46. Henriksson G, Westrin K, Carlsoo B, et al. Recurrent primary pleomorphic adenomas of salivary gland origin. Intrasurgical rupture, histopathologic features and pseudopodia. Cancer 1998;82:617-620. 47. Samson MJ, Metson R, Wang CC, et al. Preservation of the facial nerve in the management of recurrent pleomorphic adenoma. Laryngoscope 1991;101:1060-1062. 48. Liu F, Rotstein L, Davison A, et al. Benign pleomorphic adenomas: a review of the Princess Margaret Hospital experience. Head Neck 1995;17:177-183. 49. Piorkowski R, Guillamondegui O. Is aggressive surgical treatment indicated for recurrent benign mixed tumors of the parotid gland? Am J Surg 1981;142:434-436. 50. Dawson A. Radiation therapy in recurrent pleomorphic adenoma of the parotid. Int J Radiat Oncol Biol Phys 1989;16:819821. 51. Barton J, Slevin N, Gleave E. Radiotherapy for pleomorphic adenoma of the parotid gland. Int J Radiat Oncol Biol Phys 1992;22:925-928. 52. Renehan A, Gleave EN, McGurk M. An analysis of the treatment of 114 patients with recurrent pleomorphic adenomas of the parotid gland. Am J Surg 1996;172:710-714. 53. Chen AM, Garcia J, Bucci MK, et al. Recurrent pleomorphic adenoma of the parotid gland: long-term outcome of patients treated with radiation therapy. Int J Radiat Oncol Biol Phys 2006;66:1031-1035.

54. Wang C, Goodman M. Photon irradiation of unresectable carcinomas of salivary glands. Int J Radiat Oncol Biol Phys 1991;21:569-576. 55. Terhaard CH, Lubsen H, Rasch CR, et al. The role of radiotherapy in the treatment of malignant salivary gland tumors. Int J Radiat Oncol Biol Phys 2005;61:103-111. 56. Chen AM, Bucci MK, Quivey JM, et al. Long-term outcome of patients treated by radiation therapy alone for salivary gland carcinomas. Int J Radiat Oncol Biol Phys 2006;66:1044-1050. 57. Nutting CM, Rowbottom CG, Cosgrove VP, et al. Optimisation of radiotherapy for carcinoma of the parotid gland: a comparison of conventional, three-dimensional conformal, and intensity-modulated techniques. Radiother Oncol 2001;60:163-172. 58. Catterall M, Errington R. The implications of improved treatment of malignant salivary gland tumors by fast neutron radiotherapy. Int J Radiat Oncol Biol Phys 1987;13:1313-1318. 59. Griffin T, Pajak T, Laramore G, et al. Neutron vs photon irradiation of inoperable salivary gland tumors: results of an RTOGMRC cooperative randomized study. Int J Radiat Oncol Biol Phys 1988;15:1085-1090. 60. Battermann JJ, Breur K, Hart GA, et al. Observations on pulmonary metastases in patients after single doses and multiple fractions of fast neutrons and cobalt-60 gamma rays. Eur J Cancer 1981;17:539-548. 61. Laramore GE, Krall J, Griffin T, et al. Neutron versus photon irradiation for unresectable salivary gland tumors: final report of an RTOG-MRC randomized clinical trial. Int J Radiat Oncol Biol Phys 1993;27:235-240. 62. Douglas JG, Koh WJ, Austin-Seymour M, et al. Treatment of salivary gland neoplasms with fast neutron radiotherapy. Arch Otolaryngol Head Neck Surg 2003;129:944-948. 63. Douglas JG, Laramore GE, Austin-Seymour M, et al. Treatment of locally advanced adenoid cystic carcinoma of the head and neck with neutron radiotherapy. Int J Radiat Oncol Biol Phys 2000;46:551-557. 64. Haddad RI, Posner MR, Busse PM, et al. Chemoradiotherapy for adenoid cystic carcinoma: preliminary results of an organ­sparing approach. Am J Clin Oncol 2006;29:153-157.

The Nasal Cavity and Paranasal ­Sinuses

9

K. Kian Ang, Anesa Ahamad, and Adam S. Garden ANATOMY Nasal Cavity Paranasal Sinuses ETIOLOGY AND EPIDEMIOLOGY OF SINONASAL CANCER PREVENTION AND EARLY DETECTION NATURAL HISTORY Nasal Vestibule Tumors Nasal Cavity and Ethmoid Air Cell Tumors Maxillary Sinus Tumors CLINICAL MANIFESTATIONS, EVALUATION, AND STAGING Nasal Vestibule Tumors Nasal Cavity and Ethmoid Air Cell Tumors Maxillary Sinus Tumors PRIMARY THERAPY Nasal Vestibule Tumors Nasal Cavity and Ethmoid Air Cell Tumors Maxillary Sinus Tumors

RADIATION THERAPY TECHNIQUES Nasal Vestibule Tumors Nasal Cavity and Ethmoid Air Cell Tumors Maxillary Sinus Tumors Patient Care during and after Radiation Therapy New DIRECTIONS in THERAPY AND CLINICAL STUDIES Surgical Techniques Intensity-Modulated Radiation Therapy and Proton Beam Therapy Chemoradiation Therapy

Neoplasms arising in the nasal cavity and paranasal sinuses (i.e., sinonasal cancer) are relatively uncommon, accounting for less than 1% of all cases of cancer. The incidence of sinonasal cancer in the United States is about 0.75 cases per 100,000 individuals per year,1 with fewer than 4500 new cases diagnosed each year.2 The incidence is higher in some regions outside the United States, including Japan and South Africa.3 Tumors of the maxillary sinus are twice as common as those of the nasal cavity, and tumors of the ethmoid, sphenoid, and frontal sinuses are extremely rare. Sinonasal neoplasms are twice as common in men as in women. Sinonasal neoplasms present a difficult therapeutic challenge. These tumors have a distinctive pattern of spread, and most have an indolent clinical course. However, as a consequence of their rarity, clinical experience with sinonasal tumors is limited, and many tumors are diagnosed at advanced stages. The problem of advanced stage is compounded by the proximity of primary tumors to critical structures (e.g., eye, brain, cranial nerves) and by aesthetic considerations that complicate radical surgical resection and delivery of radiation treatment. The clinical presentation of epithelial neoplasms originating in the nasal cavity or paranasal sinuses, the natural history of these lesions, treatment options, and patient outcomes are summarized in this chapter. Other types of tumors that may arise in the nasal cavity or paranasal sinuses, such as soft tissue sarcomas, osteogenic sarcomas, lymphomas, and plasmacytomas, are discussed in the chapters devoted to those types of tumors.

Anatomy Nasal Cavity The nasal cavity is divided into three regions: the nasal vestibule, the nasal cavity proper or nasal fossa, and the olfactory region.4 The nasal vestibules are the two entry points into the nasal cavity. Each is a triangular space situated anterior to the limen nasi and defined laterally by the alae nasi, medially by the membranous septum—the distal end of the cartilaginous septum—and columella nasi, and inferiorly by the adjacent floor of the nasal cavity. The nasal vestibule is covered by skin, and therefore lesions at this location are essentially squamous cell skin cancer.5 The nasal fossa begins at the limen nasi anteriorly, ends at the choana posteriorly, and extends from the hard palate inferiorly to the base of the skull superiorly. The lateral walls consist of thin bony structures with three shell-shaped projections (i.e., superior, middle, and inferior conchae or turbinates) into the nasal cavity. The septum divides the nasal cavity into right and left halves. The nasal cavity is connected with a number of structures. The olfactory nerves penetrate the roof of the nasal cavity and the cribriform plate and innervate the superior nasal concha and the upper third of the septum. The superior nasal concha and the upper third of the septum are referred to as the olfactory region. The remainder of the cavity, the respiratory region, contains orifices connecting the nasal cavity with the paranasal sinuses. The superior meatus connects the nasal cavity with the posterior ethmoid air cells, the middle meatus connects the nasal cavity with the

183

184

PART 3  Head and Neck

a­ nterior and middle ethmoid air cells and the frontal sinus, and the inferior meatus connects the nasal cavity with the nasolacrimal duct. The sphenoid sinus drains into the nasal cavity through an opening in the anterior wall. Knowledge of nasal cavity anatomy facilitates comprehension of the pattern of spread of tumors of this region.

Paranasal Sinuses The paranasal sinuses are air-filled outgrowths of the nasal cavity into a number of cranial bones. They are named according to the bones in which they are located (i.e., maxillary, ethmoid, frontal, and sphenoid sinuses). Mucosal secretions from these sinuses drain to the nasal cavity through orifices situated at the lateral wall of the nasal cavity. The maxillary sinuses, the largest of the paranasal sinuses, are pyramid-shaped cavities located in the maxillae (Fig. 9-1).6 The base of each maxillary sinus forms the inferior part of the lateral wall of the nasal cavity. The roof of each maxillary sinus is formed by the floor of the orbit, which contains the infraorbital canal, and the floor of each maxillary sinus is composed of the alveolar process. The apex extends toward and often into the zygomatic bone. Maxillary sinus secretions drain into the middle meatus

Frontal lobe Crista galli Posterior ethmoidal cells Middle nasal meatus Maxillary ostium Nasal septum Inferior nasal turbinate Maxillary sinus (left)

through the hiatus semilunaris. Ohngren’s line divides the maxillary sinuses into the suprastructure and infrastructure (Fig. 9-2). The ethmoid sinuses consist of several small cavities called ethmoid air cells within the ethmoidal labyrinth, located below the anterior cranial fossa and between the nasal cavity and the orbit.6 The ethmoid air cells are separated from the orbital cavity by a thin, porous bone called the lamina papyracea and from the anterior cranial fossa by a portion of the frontal bone, the fovea ethmoidalis. The ethmoid sinuses are divided into anterior, middle, and posterior groups of air cells. The middle cells open directly into the middle meatus. The anterior cells may drain indirectly into the middle meatus through the infundibulum. The posterior cells open directly into the superior meatus.

Etiology and Epidemiology of Sinonasal Cancer In most cases of sinonasal carcinoma, the cause is unclear. Several types of nasal cavity and paranasal sinus cancer are relatively more common in individuals engaged in certain occupations or exposed to certain chemical compounds. Adenocarcinoma of the nasal cavity and ethmoid sinus, for instance, occurs more commonly in carpenters and sawmill workers who are exposed to wood dust, and squamous cell carcinoma of the nasal cavity develops more often in nickel workers. Maxillary sinus carcinoma has been associated with receipt of thorium oxide (Thorotrast), a radioactive thoriumcontaining contrast material used for radiographic study of the maxillary sinuses.3,7,8 Occupational exposure in the production of chromium, mustard gas, isopropyl alcohol, or radium also increases the risk of sinonasal carcinoma.1

Prevention and Early Detection

Alveolar process

FIGURE 9-1.  Diagram of a frontal section through the head shows the relation of the nasal fossa to the paranasal sinuses, orbit, and hard palate.

No specific recommendations regarding prevention and early detection of sinonasal tumors can be issued to the general population at present because the etiologic factors for these rare neoplasms are not well known. In general, carpenters, sawmill workers, and workers involved in the production of

Frontal bone Frontal surface

Mandibular angle

Mandibular notch

Anterior view

Lateral view

FIGURE 9-2.  Ohngren’s line, which divides the maxillary sinus into a superolateral suprastructure and an inferomedial infrastructure, passes through the medial canthus of the eye and the angle of the mandible.

9  The Nasal Cavity and Paranasal Sinuses

nickel, chromium, mustard gas, isopropyl alcohol, or radium should use appropriate face masks to reduce occupational exposure to potential carcinogens. Individuals in these professions who develop persistent nasal or sinus symptoms should see an otolaryngologist to rule out sinonasal cancer.

Natural History The clinical course of sinonasal cancer and the prognosis for affected patients depend on the anatomic site, histologic type, and stage of the tumor. The natural history of sinonasal cancer is described by disease site in the following paragraphs.

Nasal Vestibule Tumors Nasal vestibule carcinoma can spread by invasion of the upper lip, gingivolabial sulcus, premaxilla, or nasal cavity. ­Vertical invasion may result in septal (membranous or cartilaginous) perforation or destruction of alar cartilage. ­Lymphatic spread from nasal vestibule carcinoma is usually to the ipsilateral facial (buccinator and mandibular) and submandibular lymph nodes. Large lesions extending across the midline may spread to the contralateral facial or ­submandibular lymph nodes. In published series, the average incidence of nodal metastasis at diagnosis is about 5%.9-11 Without elective treatment of the lymphatics, about 15% of patients have nodal relapse.9-12 Hematogenous metastasis is rare.11

Nasal Cavity and Ethmoid Air Cell Tumors Respiratory Region and Ethmoid Air Cells Neoplasms originating in the respiratory region of the nasal cavity and the ethmoid air cells are squamous cell carcinoma, adenocarcinoma, and adenoid cystic carcinoma. ­Carcinoma may arise from inverted papilloma. The pattern of contiguous spread of carcinoma varies with the location of the primary lesion. Tumors arising in the upper nasal cavity and ethmoid air cells can extend to the orbit through the lamina papyracea and to the anterior cranial fossa through the cribriform plate or grow through the nasal bone to the subcutaneous tissue and skin. Lateral-wall primary tumors invade the maxillary antrum, ethmoid air cells, orbit, pterygopalatine fossa, and nasopharynx. Primary tumors of the floor of the nasal cavity and lower septum may invade the palate and maxillary antrum. Spreading can also occur through perineural spaces; this phenomenon is most often seen in adenoid cystic carcinoma. Lymphatic spread is rare in the case of primary tumors of the respiratory region. In a series at The University of Texas M. D. Anderson Cancer Center of 51 patients with such primary tumors, only 1 patient had palpable subdigastric lymphadenopathy at diagnosis.13 Of the 36 patients who did not undergo elective lymphatic irradiation, only 2 (6%) experienced subdigastric nodal relapse. Hematogenous dissemination is rare. In the M. D. Anderson Cancer Center series, for example, distant metastasis to bone, brain, or liver occurred in only 4 of 51 patients.13

185

by Berger and Luc14 in 1924 as “esthesioneuroepithelioma olfactif.” Since then, the tumor has been given a variety of names, including olfactory neuroblastoma, esthesioneurocytoma, and most commonly, esthesioneuroblastoma. This rare tumor constitutes approximately 3% of all intranasal neoplasms. About 250 cases were reported in the literature between 1924 and 1990.15 Esthesioneuroblastoma typically consists of round, oval, or fusiform cells containing neurofibrils with pseudorosette formation and diffusely increased microvascularity.16 Perivascular palisading of cells can be observed in the absence of pseudorosette formation. The route of contiguous spread of esthesioneuroblastoma is similar to that of ethmoidal carcinoma. In a 1979 review of 97 cases by Elkon and colleagues,17 the incidence of lymph node involvement at diagnosis was 11%, and the incidence of distant metastasis at diagnosis was 1%.

Maxillary Sinus Tumors The most common type of maxillary sinus cancer is squamous cell carcinoma, and the second most common is adenoid cystic carcinoma. Other histologic types arising in the maxillary sinus are adenocarcinoma, mucoepidermoid carcinoma, undifferentiated cancer, and occasionally, ­malignant melanoma. The pattern of spread of maxillary sinus cancer varies somewhat according to the site of origin within the antrum. Neoplasms of the suprastructure tend to spread into the nasal cavity, ethmoid air cells, orbit, pterygopalatine fossa, infratemporal fossa, and through the skull base into the middle cranial fossa (Fig. 9-3). Tumors originating from the infrastructure can be difficult to distinguish from those of the upper gum and hard palate. Neoplasms of the infrastructure tend to spread into the palate, alveolar process, gingivobuccal sulcus, soft tissue of the cheek, nasal ­cavity, masseter muscle, pterygopalatine space, and pterygoid fossa. Overall, clinical lymphadenopathy is relatively ­uncommon at diagnosis of maxillary sinus tumors. In an M. D. Anderson series, only 6 (8%) of 73 patients had palpable lymph­ adenopathy at diagnosis.18 The incidence of nodal spread, however, varied with the histologic type: 17% ­(5 of 29 patients) for squamous cell or poorly differentiated carcinoma and 4% (1 of 27 patients) for adenocarcinoma, adenoid cystic carcinoma, or mucoepidermoid carcinoma. The incidence of subclinical spread, reflected in the rate of nodal relapse without elective neck treatment, also varied with histologic type: 38% (9 of 24 patients) for squamous cell or poorly differentiated carcinoma and 8% (2 of 26 patients) for adenocarcinoma, adenoid cystic carcinoma, or mucoepidermoid carcinoma. The cumulative incidence of gross and microscopic nodal involvement for patients with squamous cell and poorly differentiated carcinoma was about 30%.18 Ipsilateral subdigastric and submandibular nodes are the most commonly involved. Hematogenous spread is uncommon.

Clinical Manifestations, Evaluation, and Staging

Olfactory Region

Nasal Vestibule Tumors

The olfactory region is the site of origin of esthesioneuroblastoma and, occasionally, adenocarcinoma. Esthesioneuroblastoma is a tumor of neural crest origin, first reported

Small carcinomas of the nasal vestibule usually manifest as asymptomatic plaques or nodules, often with crusting and scabbing. Advanced lesions may fill the vestibule or ­extend

186

PART 3  Head and Neck

A

B

C FIGURE 9-3.  Row A, Magnetic resonance imaging (MRI) shows the superior extension of a right maxillary sinus tumor through the floor of the right middle cranial fossa to the cavernous sinus (arrows) and compression of Meckel’s cave. Row B, MRI shows the difference between secretions (green arrows) and tumor (red arrows) in the sinus. Row C, Fusion of MRI with planning computed tomographic (CT) scans helps to delineate tumor in the cavernous sinus (orange arrows) that was not clearly visible on the CT images.

beyond it and may cause pain, bleeding, or ­ulceration. Large, ulcerated lesions may become infected, leading to severe tenderness that hinders complete clinical ­assessment. Contiguous spread is best detected by bimanual palpation and examination through a nasal speculum or fiberoptic telescope. Local anesthesia and decongestion of nasal mucosa facilitate the assessment of contiguous extension. Early buccinator (facial) and submandibular nodal involvement are best discovered by bimanual palpation. Nasal vestibule carcinoma is staged according to the American Joint

Committee on Cancer (AJCC) tumor-nodes-metastasis (TNM) classification system for the skin (see Chapter 6).

Nasal Cavity and Ethmoid Air Cell Tumors Carcinoma Clinical Manifestations. The presenting symptoms and signs of nasal cavity tumors usually consist of chronic ­unilateral nasal discharge, nasal ulcer, nasal ­ obstruction,

9  The Nasal Cavity and Paranasal Sinuses

a­ nterior headache, and intermittent epistaxis. The ­similarity between these symptoms and signs and those of nasal polyps often leads to a delay in diagnosis because many patients have a history of treatment for nasal polyps. Of 45 patients with nasal cavity tumors referred to M. D. Anderson between 1969 and 1985, the first symptom or sign was nasal obstruction in 23 patients (51%) and epistaxis in 7 (16%); the remaining 15 patients had a variety of symptoms, including nasal ulcer, discharge, tenderness, facial swelling, and pain.13 Additional symptoms and signs develop as the lesion enlarges, and their nature depends on the pattern of spread. Invasion of the orbital cavity, for instance, may produce a palpable medial orbital mass, proptosis, and diplopia. Obstruction of the nasolacrimal duct causes epiphora. Involvement of the olfactory region may produce anosmia (loss of the sense of smell) and expansion of the bridge of the nose. Extension through the cribriform plate results in frontal headache. The presenting symptoms and signs of ethmoid air cell tumors are sinus ache and referred pain to the nasal or retrobulbar region, subcutaneous nodule at the inner canthus, nasal obstruction, nasal discharge, diplopia, and proptosis. Evaluation and Staging. Physical assessment to determine the extent of nasal cavity and ethmoid air cell tumors should include fiberoptic examination of the nasal cavity and nasopharynx after mucosal decongestion and ­application of topical anesthesia; evaluation of cranial nerves (particularly cranial nerves III, IV, V, and VI); and palpation of the nose bridge, cheek, and medial orbital cavity. Computed tomography (CT) or, preferably, magnetic resonance imaging (MRI) is essential to determine the disease extent and invasion of adjacent structures (e.g., orbital cavity, anterior cranial fossa). In 2002, the AJCC introduced a staging system for nasal cavity and ethmoid sinus tumors in the sixth edition of their staging manual (Box 9-1).19 However, many groups continue to use the classification system proposed by the University of Florida.20 In this system, stage I represents lesions limited to the nasal fossa, stage II represents lesions extending to adjacent sites (i.e., paranasal sinuses, skin, orbit, pterygomaxillary fossa, and nasopharynx), and stage III represents lesions extending beyond these structures. These differences in staging systems are important when comparing results from various institutions. Tumors of the lateral wall and floor of the nasal cavity tend to be diagnosed at more advanced stages. Among patients with nasal cavity tumors referred to M. D. Anderson between 1969 and 1985, 3 (21%) of 14 patients with septal lesions and 20 (65%) of 31 patients with lateral wall or floor lesions had disease extension beyond the nasal cavity (i.e., stages II and III).13

Esthesioneuroblastoma Because of the rarity of esthesioneuroblastoma, the natural history of this disease was not well known until Elkon and colleagues17 conducted a thorough literature review of the characteristics of 97 patients reported between 1969 and 1977. The investigators found the age distribution to be bimodal, with a first peak in incidence occurring between 11 and 20 years old and a second, higher peak in incidence occurring between 51 and 60 years old. The incidence was slightly higher among women than among men.

187

The most common presenting symptoms of esthesio­ neuroblastoma are nasal obstruction and epistaxis. Spaulding and colleagues21 found that anosmia can precede diagnosis by many years. Other symptoms include symptoms related to contiguous disease extension into the orbit (i.e., proptosis, visual field defects, orbital pain, and epiphora), paranasal sinuses (i.e., medial canthus mass and facial swelling), and anterior cranial fossa (i.e., headache) and manifestations of inappropriate antidiuretic hormone secretion. The recommended physical evaluation and staging workup for esthesioneuroblastoma are similar to those for nasal cavity tumors. A commonly used staging system, proposed by Kadish and colleagues,16 is as follows: Stage A

Disease confined to the nasal cavity

Stage B

Disease confined to the nasal cavity and one or more paranasal sinuses

Stage C

Disease extending beyond the nasal cavity and paranasal sinuses that includes involvement of the orbit, base of skull or intracranial cavity, or cervical lymph nodes or distant metastasis

The stage distribution at diagnosis for the 97 patients reviewed by Elkon and colleagues17 was as follows: stage A, 30%; stage B, 42%; and stage C, 28%. Of the 22 patients with stage C disease, 9 had orbital involvement, 6 had invasion of the cribriform plate or cranial fossa, 11 had palpable neck nodes, and only 1 patient had distant metastasis.

Maxillary Sinus Tumors Maxillary sinus tumors tend to have an indolent course and therefore are usually diagnosed at advanced stages. Of the 73 patients referred to M. D. Anderson for treatment between 1969 and 1985,18 for example, 26 (36%) sought medical attention because of symptoms and signs induced by disease extension to the premaxillary region, such as facial swelling, pain, or paresthesia of the cheek. Twenty patients (27%) had symptoms related to tumor spread to the nasal cavity (e.g., epistaxis, nasal discharge, nasal obstruction), and 19 patients (26%) had symptoms related to tumor spread to the oral cavity (e.g., ill-fitting denture, alveolar or palatal mass, unhealed tooth socket after extraction). Five patients (7%) presented with symptoms and signs of orbital invasion, such as proptosis, diplopia, impaired vision, or orbital pain.18 Physical examination should include assessment of the oral and nasal cavities, palpation of the cheek, ­evaluation of eye movement and function, and assessment of the trigeminal nerve, particularly the branches of V2 (i.e., infraorbital nerve, anterior superior alveolar nerve, posterior superior alveolar nerve, and greater palatine nerve). Biopsy with a Caldwell-Luc operation should be avoided because this can result in numbness of the incisor teeth and upper lip. CT or preferably MRI is essential to determine the disease extent and invasion of adjacent structures (e.g., orbital cavity; infratemporal, pterygopalatine, and anterior cranial fossae). MRI can be particularly helpful for distinguishing tumor from accumulation of mucus (see Fig. 9-3).

188

PART 3  Head and Neck

BOX 9-1.  American Joint Committee on Cancer Staging System for Tumors of the Nasal Cavity and Paranasal Sinuses Primary Tumor (T) TX T0 Tis

Primary tumor cannot be assessed No evidence of primary tumor Carcinoma in situ

Maxillary Sinus T1 T2

Tumor limited to maxillary sinus mucosa with no erosion or destruction of bone Tumor causing bone erosion or destruction, including extension into the hard palate and/or middle nasal meatus, except extension to posterior wall of maxillary sinus and pterygoid plates Tumor invades any of the following: bone of the posterior wall of maxillary sinus, subcutaneous tissues, floor or medial wall of orbit, pterygoid fossa, ethmoid sinuses Tumor invades anterior orbital contents, skin of cheek, pterygoid plates, infratemporal fossa, cribriform plate, sphenoid or frontal sinuses Tumor invades any of the following: orbital apex, dura, brain, middle cranial fossa, cranial nerves other than maxillary division of trigeminal nerve (V2), nasopharynx, or clivus

T3 T4a T4b

Nasal Cavity and Ethmoid Sinus T1 T2

Tumor restricted to any one subsite, with or without bony invasion Tumor invading two subsites in a single region or extending to involve an adjacent region within the nasoethmoidal complex, with or without bony invasion Tumor extends to invade the medial wall or floor of the orbit, maxillary sinus, palate, or cribriform plate Tumor invades any of the following: anterior orbital contents, skin of nose or cheek, minimal extension to anterior cranial fossa, pterygoid plates, sphenoid or frontal sinuses Tumor invades any of the following: orbital apex, dura, brain, middle cranial fossa, cranial nerves other than (V2), nasopharynx, or clivus

T3 T4a T4b

Regional Lymph Nodes (N) NX N0 N1 N2

Regional lymph nodes cannot be assessed No regional lymph node metastasis Metastasis in a single ipsilateral lymph node, 3 cm or less in greatest dimension Metastasis in a single ipsilateral lymph node, more than 3 cm but not more than 6 cm in greatest dimension; or in multiple ipsilateral lymph nodes, none more than 6 cm in greatest dimension; or in bilateral or contralateral lymph nodes, none more than 6 cm in greatest dimension Metastasis in single ipsilateral lymph node more than 3 cm but not more than 6 cm in greatest dimension Metastasis in multiple ipsilateral lymph nodes, none more than 6 cm in greatest dimension Metastasis in bilateral or contralateral lymph nodes, none more than 6 cm in greatest dimension Metastasis in a lymph node more than 6 cm in greatest dimension

N2a N2b N2c N3

Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage Grouping Stage 0 Stage I Stage II Stage III Stage IVA Stage IVB Stage IVC

Tis T1 T2 T3 T1-T3 T4a T1-T3 Any T T4b Any T

N0 N0 N0 N0 N1 N0-N2 N2 N3 Any N Any N

M0 M0 M0 M0 M0 M0 M0 M0 M0 M1

From Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 59-67.

The T categories in the 1997 AJCC staging system22 were based on clinical and radiologic evidence of involvement of various adjoining anatomic structures: TX Primary tumor cannot be assessed T0

No evidence of primary tumor

Tis

Carcinoma in situ

T1

Tumor limited to the antral mucosa with no erosion or destruction of bone

T2

Tumor causing bone erosion or destruction, except for the posterior antral wall, including extension into the hard palate and/or the middle nasal meatus

9  The Nasal Cavity and Paranasal Sinuses

T3

Tumor invades any of the following: bone of the posterior wall of maxillary sinus, subcutaneous tissues, skin of cheek, floor or medial wall of orbit, infratemporal fossa, pterygoid plates, ethmoid sinuses

T4

Tumor invades orbital contents beyond the floor or medial wall, including any of the following: the orbital apex, cribriform plate, base of skull, nasopharynx, sphenoid, frontal sinuses

Of the 73 patients with maxillary sinus tumors referred to M. D. Anderson between 1969 and 1985, 3 had T1 tumors, 16 had T2 tumors, 32 had T3 tumors, and 22 had T4 tumors. Modifications to the 1997 categorization scheme in the 2002 edition of the staging manual (see Box 9-1)19 consist mostly of clarifications of T2, T3, and T4 disease.

Primary Therapy Nasal Vestibule Tumors Radiation therapy usually is preferred over surgery for carcinoma of the nasal vestibule because irradiation ­produces better cosmetic outcomes. Depending on the location and size of the primary lesion, irradiation can be accomplished by using external beam radiation therapy (EBRT), brachytherapy, or a combination of the two techniques. ­Cartilage invasion is not a contraindication for radiation therapy, because the risk of necrosis is low with fractionated ­irradiation.23 Surgery can yield high control rates with excellent cosmetic results for selected small superficial lesions. The rare cases of large primary tumors with extensive tissue destruction and distortion are best managed by resection in combination with preoperative or postoperative radiation therapy. Experienced prosthodontists can design custommade nasal prostheses. Of the 32 patients with newly diagnosed nasal vestibule carcinoma who received primary radiation therapy at M. D. Anderson between 1967 and 1984, 11 patients with small tumors were given interstitial brachytherapy, and the remaining 21 patients were given EBRT to the primary tumor12; 14 of the 21 patients also received irradiation to the draining lymphatics. Local control was achieved in all 11 of the 11 patients treated with brachytherapy and in 20 of the 21 patients treated with external beam therapy, resulting in an overall local primary tumor control rate of 97%. None of the 11 patients with well-delineated small lesions treated with brachytherapy experienced nodal recurrence. Four of nine patients who underwent EBRT to the primary lesion (including two patients who also had undergone radiation therapy to the facial lymphatics) experienced a neck node recurrence. In contrast, none of 12 patients who received ­local treatment with elective neck node irradiation had nodal failure. Table 9-1 summarizes the results of relatively large series of patients with nasal vestibule tumors treated with radiation therapy.11,12,24-26 Selected patients in the M. D. Anderson and University of Florida series received brachytherapy, whereas all patients in the Princess Margaret Hospital series underwent EBRT. Several conclusions can be reached from these findings. First, brachytherapy or EBRT cures, on average, about 90% of nasal vestibule lesions smaller than 2 cm. Second, given in adequate doses, EBRT can control

189

70% to 80% of tumors 2 cm or larger. Third, the rate of ­ ccult nodal spread is extremely low in lesions smaller o than 2 cm but can be as high as about 40% in the case of larger primary tumors. Third, the incidence of severe late complications is low with proper radiation technique and fractionation schedules. The Groupe Européen de Curiethérapie27 in France analyzed outcomes after radiation therapy for 1676 ­carcinomas of the skin of the nose and nasal vestibule and found the overall local control rate to be 93%. However, the local control rate varied according to tumor size (96% for tumors smaller than 2 cm, 88% for tumors 2 to 3.9 cm, and 81% for tumors 4 cm or larger), tumor site (94% for those on external surfaces versus 75% for those in the ­nasal vestibule), and whether the tumor was newly diagnosed (95%) or recurrent (88%).27 Local control was independent of histology for tumors smaller than 4 cm, but among tumors larger than 4 cm, basal cell carcinomas were more often controlled than were squamous cell ­ carcinomas. Few ­ complications were reported; the ­necrosis rate was 2%. The local control rate with surgery was ­approximately 90%.

Nasal Cavity and Ethmoid Air Cell Tumors Carcinoma Radiation therapy and surgery are equally effective in curing stage I lesions of the respiratory region. The choice of therapy depends on the size and location of the tumor and the expected cosmetic outcome. For example, posterior nasal septum lesions are easily treated with surgery ­without cosmetic impairment. In contrast, small (≤1.5 cm) ­anteroinferior septal lesions are best treated with interstitial brachytherapy with an iridium 192 (192Ir) wire implant to avoid deformity from surgical removal of the anterior nasal septum. For cosmetic reasons, EBRT is the preferred treatment for lateral wall lesions extending to the ala nasi. Stage II and operable stage III neoplasms of the nasal cavity are best treated with surgery and, in most cases, postoperative irradiation. Nasal Cavity Tumors: The M. D. Anderson ExperiBetween 1969 and 1985, 51 patients with ­carcinoma ence. ����������������������������������������������������� of the nasal cavity proper were referred to M. D. Anderson, of whom 45 received curative treatment and were followed for 2.8 to 16.8 years (median, 11 years).13 Eighteen ­ patients received radiation therapy only (interstitial brachytherapy in 5 and external beam therapy in 13), and 27 ­ underwent surgery and postoperative radiation therapy. The site of origin of the primary lesion was the nasal septum in 14 patients and the lateral wall or floor of the cavity in 31 patients. All 14 lesions of the nasal septum were ­squamous cell carcinoma. Of the 31 nonseptal nasal ­cavity tumors, 16 were squamous cell carcinoma, 9 were adenocarcinoma, 5 were adenoid cystic carcinoma, and 1 was undifferentiated cancer. Table 9-2 shows the pattern of failure in this series of patients.13 Disease was controlled in all 14 patients with carcinoma of the nasal septum (10 after initial therapy and 4 after salvage therapy). Nine of 31 patients with lesions of the lateral wall or floor died of disease—5 of uncontrolled local disease, 2 of distant metastases, and 2 of both. The ­disease-specific survival rates at 5 and 10 years were

190

PART 3  Head and Neck

TABLE 9-1.   Reported Local and Regional Control Rates for Nasal Vestibule Carcinomas Treated with Definitive Radiation Therapy Study

Local Control

Regional Control

Late Complications

M. D. Anderson Cancer Center12

Brachytherapy: 11/11 External-beam radiation therapy: 20/21 (95%) Total: 31/32 (97%)

Small lesions: 11/11 Large lesions: ELI: 12/12 (100%) No ELI: 5/9 (56%) Total: 28/32 (88%)

Osteonecrosis at area of hot spot: 1 Epistaxis: 1

University of Florida24

Brachytherapy: 9/9 External-beam radiation therapy: 8/11 (73%) Both: 12/16 (75%) Total: 26/30a (87%)

ELI: 3/3 No ELI: 23/27 (85%) (4 relapses were salvaged) Total: 29/36a (81%)

Transient soft tissue necrosis after implants: 4 Nasal stenosis: 1

Princess Margaret ­Hospital11

Tumors <2 cm (n = 34): 97%b Tumors ≥2 cm (n = 16) + size not recorded (n = 6): 57%b Dose <55 Gyc: 30%b Dose ≥55 Gyc: 82%b

No ELI: 51/54 (94%)b (2 patients presented with positive nodes)

Osteonecrosis: 2 (1 resulting in fistula) Nasal stenosis: 2 Massive epistaxis: 1

Rotterdam Radiotherapy Institute 25

Tumors ≤1.5 cm (n = 15): 72%b Tumors >1.5 cm (n = 17): 50%b Dose <54 Gyd: 37%b Dose ≥54 Gyd: 82%b

Details not presented

Details not presented

VU University Medical Center, Amsterdam26

Overall at 2 years: 79% (ultimate at 5 years [after salvage]: 95%) Tumors <1.5 cm (n = 32): 83% ­(ultimate: 94%) Tumors ≥1.5 cm (n = 24): 74% ­(ultimate: 96%)

Routine ELI to the mustache region Overall at 2 years: 87% (6 of 7 neck relapses were salvaged; ultimate at 5 years: 97%)

Rhinorrhea: 45% Nasal dryness: 39% Epistaxis: 15% Adhesions: 4% Skin necrosis: 3 patientse Nasal-vestibule sarcoma: 1 patient

ELI, elective lymphatic irradiation. aIncludes patients eligible for follow-up at 2 years. bFive-year actuarial rates. cRadiation delivered in 25 fractions over 5 weeks. dRadiation delivered in 2.5- to 3-Gy fractions. eAll three patients were in the intermediate-dose brachytherapy group (median rate of 2.7 Gy/h [range, 0.7 to 7.1 Gy/h]); median dose per fraction was 15 Gy [range, 3.2 to 26 Gy]; total dose was 16 Gy [range, 8 to 40 Gy]).

TABLE 9-2.    Patterns of Failure of Nasal Cavity Carcinomas Timing of Relapse and ­Location of Primary Tumor

Type of Relapse Number of ­Patients

Local

Nodal

Local and Distant

Distant Only

Septum

14

2

2

0

0

Lateral wall or floor

31

9

0

1

2

Septum

4

0

0

0

0

Lateral wall or floor

5

1

0

1

0

After primary therapy

After salvage therapy

Modified from Ang KK, Jiang GL, Frankenthaler RA, et al. Carcinomas of the nasal cavity. Radiother Oncol 1992;24:163-168.

83% and 80%, respectively, and the corresponding overall survival rates were 69% and 58%. Two patients underwent required orbital exenteration, and two others experienced radiation injury to the cornea and optic pathway. Other uncommon side effects were bone necrosis, dental decay, nasal stenosis, and septal perforation. However, these complications occurred predominantly in patients treated before the routine use of individual portal shaping. This study indicated that the prognosis for patients with nasal cavity carcinoma was better than that for patients

with maxillary sinus cancer treated during the same era (see “Maxillary Sinus Tumors”). The study also showed that carcinoma of the nasal septum was diagnosed at an earlier stage than carcinoma of the lateral wall and nasal floor and that excellent control could be achieved by primary radiation therapy. When the tumor site is accessible, interstitial brachytherapy may be the treatment of choice for ­early-stage carcinoma of the nasal septum. This conclusion was supported by the ­ findings of a systematic review of 154 articles published from 1960 to 1998 describing a ­total

9  The Nasal Cavity and Paranasal Sinuses

191

TABLE 9-3.    Outcomes for Patients with Nasal Cavity Tumors Treated with Radiation, Surgery, or Both Number of Patients Treated

M. D. Anderson Cancer Center13

5-year actuarial: 75% Radiation therapy only: 18 All 14 nasal septum lesions and Radiation therapy followed 22 of 31 tumors of lateral wall by surgery: 2 or floor were controlled. Surgery followed by ­radiation therapy: 25

Roswell Park Memorial ­Institute29*

Radiation therapy only: 30 Surgery only: 13 Surgery and radiation therapy: 14

5-year actuarial: 67% Patients with well-differentiated squamous cell carcinoma had the best outcomes.

Not presented

University of Puerto Rico30*

Radiation therapy only: 34 Surgery only: 6

5-year overall: 56%

Not presented

Mallinckrodt Institute of ­Radiology31*

5-year actuarial: 52% Radiation therapy only: 28 Survival not significantly Radiation therapy followed by surgery: 18 ­affected by histology Surgery followed by ­radiation Neck relapse occurred in 10 of 52 with N0 disease. therapy: 10

*Included

Survival

Late Complications: Number of Patients

Study

Radiation-induced blindness: 2 Surgical blindness: 2 Maxilla necrosis: 2 Stenosis/perforation: 3

Complications (soft tissue necrosis, cataract, synech­ iae, otitis media) requiring surgery: 5

patients with nasal vestibule carcinomas.

of 16,396 patients with sinonasal carcinoma.28 Survival rates for patients with nasal cavity cancer were consistently better for patients with maxillary sinus cancer; for example, overall survival rates for patients treated during the 1990s were 66% ± 15% for those with nasal cavity cancer and 45% ± 11% for those maxillary sinus cancer.28 Nasal Cavity Tumors: Outcomes in Other Large Series. Comparing findings in the literature on nasal cavity lesions is difficult because the patient populations vary among the reported series. Table 9-3 summarizes the ­results of relatively large series focusing on nasal cavity ­tumors.13,29-31 Locoregional control rates obtained with definitive irradiation for patients with tumors confined to the nasal cavity and with combined surgery and radiation therapy for patients with more advanced disease ranged from 60% to 80%. The control rate was higher for nasal cavity lesions than for maxillary sinus tumors (discussed later). The incidence of isolated regional recurrence in patients who had no palpable lymphadenopathy at diagnosis and did not receive elective nodal treatment was about 5% in the M. D. Anderson series. Treatment morbidity was proportional to the extent of disease, which dictated the magnitude of surgical resection and the target volume for radiation therapy. The most common complications were soft tissue necrosis, nasal stenosis, and visual impairment. A review of 783 patients with nasal cavity cancer in the Surveillance, Epidemiology, and End Results database from 1988 through 199832 showed the most common tumor type to be squamous cell carcinoma (49.3%), followed by esthesioneuroblastoma (13.2%). More than one half of the patients presented with a small primary tumor (T1), and only 5% had evidence of nodal disease at diagnosis. The overall mean survival time was 76 months (median, 81 months), and the 5-year overall survival rate was 56.7%. On multivariate analysis, male sex, older age, disease of higher T and N categories, and higher (poorer) tumor grade were found to have independent adverse effects on survival (P < .05). Radiation therapy was administered to 50.5% of patients, mainly those with poor prognostic features

(i.e., perineural invasion, positive margins, poor performance status, or medically unfit for surgery). Median survival times and 5-year survival rates according to tumor histology, T category, and grade are shown in Table 9-4.32 Ethmoid Air Cell Tumors: The M. D. Anderson ­Experience. Between 1969 and 1993, 34 patients with carcinoma of the ethmoid air cells were referred to M. D. Anderson for therapy.33 Thirteen patients underwent primary radiation therapy, 21 underwent surgery plus ­ radiation, and 9 underwent surgery plus adjuvant chemotherapy. ­Radiation doses, given at approximately 2 Gy/fraction, were 50 Gy before surgery, 52 to 66 Gy (median, 60 Gy) after surgery, and 50 to 70 Gy (median, 63 Gy) when no surgery was to be performed. None of the patients received elective nodal treatment. The 5-year actuarial rates of overall, disease-free, and disease-specific survival for these ­patients were 55%, 58%, and 63%, respectively. The actuarial 5-year local control rate for the entire group was 71%, (74% for those given surgery plus radiation and 64% for those given radiation alone). Local recurrence occurred in nine patients, nodal relapse in three patients, and distant metastasis in four patients. Histologically proven invasion of the dura mater was associated with poorer local control rates among patients undergoing surgery and radiation. A simple T-category system based on anatomic criteria was a good discriminator for local control, with 10-year control rates of 100% for those with T1 tumors, 79% for those with T2 tumors, and 53% for those with T3 tumors. Of the nine patients who received chemotherapy, three had complete responses, and four had partial responses; six of the seven responders had undifferentiated carcinoma. Severe complications of therapy occurred only among those patients treated before 1984 and consisted mainly of visual impairments and brain necrosis. The conclusions reached from this single-institution experience were that ethmoid sinus carcinomas show extensive local invasion but rarely lymphatic or hematogenous spread, making local recurrence the main cause of cancer-related death. The combination of surgery and postoperative radiation yielded the highest local control

192

PART 3  Head and Neck

TABLE 9-4.   Survival among 783 Patients Treated for Nasal Cavity Cancer from the Surveillance, Epidemiology, and End Results Database, 1988-1998 Tumor Characteristics

Number of Patients (%)

Median Survival Time (mo)

5-Year Survival Rate (%)

Histology All

783 (100)

81

56.7%

Adenocarcinoma

39 (5.0)

54

49.0%

Adenoid cystic carcinoma

38 (4.9)

106

59.1%

Squamous cell carcinoma

386 (49.3)

84

61.6%

Esthesioneuroblastoma

103 (13.2)

88

63.6%

13 (1.7)

42

49.5%

T1

431 (55.0)

94

66.4%

T2

64 (8.2)

66

51.8%

T3

222 (28.4)

51

45.6%

T4

66 (8.4)

39

40.2%

N0

356 (45.5)

88

62.3%

N1

39 (5.0)

25

28.4%

I (well differentiated)

121 (15.5)

105

75.3%

II (moderately differentiated)

163 (20.8)

NA

61.9%

III (poorly differentiated)

139 (17.8)

49

47.6%

62 (7.9)

46

36.8%

Sinonasal undifferentiated ­carcinoma T Category

N Category

Grade

Undifferentiated

NA, not applicable. Modified from Bhattacharyya N. Cancer of the nasal cavity. Survival and factors influencing prognosis. Arch Otolaryngol Head Neck Surg 2002;128:1079-1083.

rate, but because radiation therapy alone eradicated two thirds of primary tumors, it was considered a reasonable alternative for patients who could not undergo surgery. Among those patients who underwent surgery and radiation therapy, dura mater invasion predicted a higher ­local recurrence rate. Severe radiation complications were rare after the implementation of contemporary radiation therapy ­techniques. Chemotherapy induced excellent responses in ­ undifferentiated carcinoma, but its effect on overall disease control is unclear in this small series of ­patients.

Esthesioneuroblastoma Literature Overview. Because of the rarity of esthesioneuroblastoma, the optimal therapy cannot be defined on the basis of prospective studies. Elkon and colleagues,17 in reviewing case reports of patients treated before 1977 ­(described later), found that single-modality therapy (i.e., radiation or surgery) yielded ultimate locoregional control rates of more than 90% for Kadish stage A lesions (defined earlier), although several patients had to undergo salvage therapy for local or regional relapse. The optimal therapy for stage B disease is unclear, partly because this stage represents a heterogeneous group of tumors; nevertheless, combinations of surgery and radiation therapy seem to

have a slight advantage over other forms of treatment for patients with stage B or stage C disease. The available data do not justify routine elective nodal treatment because the incidence of isolated nodal relapse is less than 15%. Table 9-5 summarizes the Elkon literature review of ­patients with esthesioneuroblastoma treated by various modalities between 1969 and 1977.17 Treatment outcome for patients with stage A lesions was excellent. Although 42% of those patients developed local or regional recurrence after initial treatment, salvage therapy was ­successful in all nine patients in whom it was attempted. Consequently, only 1 of 24 patients with stage A tumors died of the disease (as a result of neck node recurrence and lung metastasis). The prognosis for patients with stage B disease was also relatively good. The locoregional recurrence rate in patients with stage B tumors after initial therapy was not higher than that in patients with stage A lesions, but salvage therapy was less successful. Overall, 30% of patients with stage B tumors died of the disease. About 60% of patients with stage C tumors died of the disease. Most of those patients succumbed to ­local tumor ­ progression. Distant metastasis was a rare ­failure pattern (≤10%), even in locoregionally advanced ­esthesioneuroblastoma.

9  The Nasal Cavity and Paranasal Sinuses

193

TABLE 9-5.    Outcomes for Patients with Esthesioneuroblastoma by Stage and Treatment Modality Patient Outcome (Number of Patients)

Nodal Relapse

Salvage with Subsequent RT or S

Distant ­ etastasis M

Died of Disease

Kadish Stage

Therapy

Local ­Recurrence

A (n = 24)

RT

2/5

1/5

3/3

0/5

0/5

4/9

1*/9

4/4

0/9

0/9 1/10

S S+RT

1/10

2/10

2/2

1†/10

Totals

All

7/24 (29%)

4/24 (17%)

9/9

1/24 (4%)

1/24 (4%)

B (n = 33)

RT

1/7

2/7

0/1

0/7

3/7

S

3/6

0/6

1/2

0/6

2/6

S+RT

5/20

1/20

3/4

3/20

5/20

Totals

All

9/33 (27%)

3/33 (9%)

4/7

3/33 (9%)

10/33 (30%)

C (n = 21)

RT

3/5

0/5

0/0

1/5

4/5

S

0/1

0/1

0/0

0/1

0/1

S+RT‡

6/15

1/15

1/1

1/15

8/15

All

9/21 (43%)

1/21 (5%)

1/1

2/21 (10%)

12/21 (57%)

Totals *Along

with local recurrence. with nodal relapse. ‡Site of relapse not specified in two patients. RT, radiation therapy; S, surgery. Modified from Elkon D, Hightower SI, Lim ML, et al. Esthesioneuroblastoma. Cancer 1979;44:1087-1094. †Along

Results of Individual Series. Several large series of patients with esthesioneuroblastoma have been reported. In one of largest single-institution series of patients with esthesioneuroblastoma, Spaulding and colleagues21 retrospectively reviewed the records of 25 patients treated at the University of Virginia Medical Center between 1959 and 1986. Twenty-four patients had stage B or C tumors and were followed for 2 or more years after treatment. Treatment policies and techniques had gradually evolved over the study period, with progressive introduction of cranio­ facial resections, use of complex field megavoltage radiation, and—for stage C disease—the addition of chemotherapy to radiation therapy and surgery. Patients were divided into two groups on the basis of treatment era for comparative analysis. Group I consisted of 10 patients (6 stage B and 4 stage C) treated between 1959 and 1975. Seven patients were treated with a combination of radiation therapy (40 to 60 Gy) and wide local excision, two received radiation therapy only (50 Gy and 55 Gy, respectively), and one received radiation therapy (23 Gy) with vincristine. Group II consisted of 15 patients (1 stage A, 6 stage B, and 8 stage C) treated between 1976 and 1985. Five ­patients with stage B disease underwent preoperative ­radiation therapy (50 Gy) and craniofacial resection, and the other patient with stage B disease, who had a synchronous lung cancer, received cyclophosphamide, doxorubicin, and vincristine followed by radiation therapy (63.3 Gy). The treatment of patients with stage C disease was more varied; two underwent preoperative radiation therapy with 50 Gy or 55 Gy and craniofacial resection, one had ­subtotal excision followed by radiation therapy (56 Gy), and five underwent two to four cycles of chemotherapy (vincristine

and cyclophosphamide) plus radiation therapy (50 Gy) and craniofacial resection. Among patients with stage B disease, those in groups I and II had identical local control rates (83%) and 2-year disease-specific survival rates (83%). Among patients with stage C disease, however, the local control rate was higher in group II (57% versus 25% for group I). Because salvage therapy for local recurrence was successful in several group II patients, the 5-year disease-specific survival rate was also higher in group II (≈70% versus 0%). Although this series was relatively small, two important conclusions can be drawn from this retrospective analysis. First, extensive craniofacial resection does not seem to ­ confer a major advantage over wide local excision for patients with stage B lesions. Second, the addition of chemotherapy to craniofacial resection and radiation ­therapy for patients with stage C tumors may yield a higher disease-specific survival rate than that achievable with wide local excision and radiation therapy. Two updates of this experience, the first including 35 patients treated from September 1976 through May 199834 and the second 50 patients treated between September 1976 and May 2004,35 confirmed that long-term disease control can be achieved with the combination of neoadjuvant chemotherapy and aggressive surgical resection. In a 2006 report, in which 32 patients presented with Kadish stage C lesions, ­diseasefree survival rates were 86.5% at 5 years and 82.6% at 15 years.35­ Local control and survival rates were also favorable, at 96.2% and 93.1%, respectively, at 5 years for patients with esthesioneuroblastoma in a series of 72 patients with sinonasal neuroendocrine tumors treated at M. D. Anderson

194

PART 3  Head and Neck

TABLE 9-6.   Survival and Disease Control Rates According to Tumor Type for 72 Patients with Sinonasal Neuroendocrine Tumors Treated at M. D. Anderson between 1982 and 2002 Rates at 5 Years (%) Tumor Histology

Survival

Local Control

Esthesioneuroblastoma (n = 31)

93.1

96.2

8.7

Sinonasal undifferentiated carcinoma (n = 16)

62.5

78.6

15.6

25.4

Neuroendocrine carcinoma (n = 18)

64.2

72.6

12.9

14.1

Small cell carcinoma (n = 7)

28.6

66.7

44.4

75.0

Regional Failure

Distant Metastasis 0

Modified from Rosenthal DI, Barker JL Jr, El-Naggar AK, et al. Sinonasal malignancies with neuroendocrine differentiation. Cancer 2004;101:2567-2573.

TABLE 9-7.   Outcomes for Patients with Advanced Paranasal Sinus Cancer Treated by Combined Surgery and Radiation Therapy Rates at 5 Years (%) Study Lavertu et al, Spiro et al,

198937

198941 198942

Number of Patients

Survival

Local Recurrence

Distant Metastasis

54

38

52

NR

105

38

49

15

149

36

43

NR

Jiang et al, 199118

67*

53†

24

27

Paulino et al, 199840

48

47

46

17

97

34

54

34

141

52

56

33

Zaharia et al,

Le et al,

199938

Myers et al,

200239

*All

had node-negative disease survival rate NR, not reported.

†Relapse-free

between 1982 and 2002 (Table 9-6).36 Moreover, both local and distant control of esthesioneuroblastoma were excellent with local therapy alone.36 These results contrast with those from the Surveillance, Epidemiology, and End Results analysis,32 which included 103 patients (13.2%) with esthesioneuroblastoma; the median survival time for those patients was 88 months, and the 5-year overall survival rate was 63.6%.

Maxillary Sinus Tumors For patients presenting with T1 or T2 maxillary sinus tumors of the infrastructure, surgery alone is often sufficient for local control. Depending on the location and size of the lesion, surgery can in most cases be limited to a partial maxillectomy. For patients with more advanced but resectable disease who are fit to undergo resection, the ­combination of surgery and radiation therapy, usually postoperative radiation therapy, is the treatment of choice. Radical maxillectomy with or without orbital exenteration is necessary for most such patients. Primary radiation therapy should be reserved for patients with medical conditions that preclude surgery or those who decline radical surgery. Treatment outcomes for patients with maxillary sinus carcinomas have improved over the past four decades.28 The Surveillance, Epidemiology, and End Results database shows that overall survival rates have increased progressively over time; the mean (± standard deviation) overall survival rates for patients treated during the 1960s were 26% ± 13%; for those treated during the 1970s, 31% ± 8%;

during the 1980s, 39% ± 14%; and during the 1990s, 45% ± 11% (P = .02). However, treatment outcomes for ­maxillary sinus tumors are still generally poor (Table 9-7),18,37-42 with less than 50% of patients surviving more than 5 years despite having radical surgery. Both local recurrence and distant metastasis are relatively common.

Postoperative Radiation Therapy According to published series, the combination of surgery and irradiation for maxillary sinus tumors yields 5-year local control rates ranging from 53% to 78% and 5-year survival rates from 39% to 64%.43-52 These rates are ­better than those achieved with either surgery or radiation ­therapy alone. Outcomes of 73 patients with maxillary sinus carcinoma who underwent postoperative radiation therapy at M. D. Anderson between 1969 and October 1985 were ­reported in 1989.18 Three patients had AJCC T1 disease, 16 had T2, 32 had T3, and 22 had T4 disease. Only six patients had palpable nodal disease at diagnosis. All patients underwent partial or radical maxillectomy followed by radiation therapy delivered using a wedged-pair or three-field technique. The radiation doses ranged from 50 to 60 Gy, except for three patients who received lower doses. Elective nodal irradiation was given to 17 patients. The results of this retrospective analysis are presented in Tables 9-8 through 9-11.18 Several conclusions can be drawn from this retrospective analysis. First, local recurrence and distant metastasis

9  The Nasal Cavity and Paranasal Sinuses

195

TABLE 9-8.    Patterns of Failure in 73 Patients with Maxillary Sinus Carcinoma Site of Relapse Timing of ­Relapse

Local

Nodal

Local and Nodal

Distant Metastasis

Local and Distant Metastasis

Nodal and ­Distant Metastasis

After initial therapy

10

10

1

8

3

1

After salvage therapy

11

0

1

11

4

1

Modified from Jiang GL, Ang KK, Peters LJ, et al. Maxillary sinus carcinomas: natural history and results of postoperative radiotherapy. ­Radiother Oncol 1991;21:193-200.

TABLE 9-9.    Incidence of Nodal Metastasis by Histologic Type in 73 Patients with Maxillary Sinus Carcinoma* Histologic Type

Patients Who Presented with Positive Neck Nodes (n/N)

Patients Who Developed Nodal ­ ecurrence (n/N) R

Squamous cell carcinoma

5/23

6/18

Undifferentiated tumor

0/6

3/6

Adenocarcinoma

1/6

1/5

Adenoid cystic carcinoma

0/19

1/19

Mucoepidermoid carcinoma

0/2

0/2

*Excludes

17 patients who received elective neck irradiation. Modified from Jiang GL, Ang KK, Peters LJ, et al. Maxillary sinus carcinomas: natural history and results of postoperative radiotherapy. ­Radiother Oncol 1991;21:193-200.

were the major causes of death (see Table 9-8). Second, the nodal relapse rate among patients with squamous cell and undifferentiated carcinoma who did not undergo elective lymphatic irradiation was more than 30% (see Table 9-9). Third, the local recurrence rate increased in parallel with T category and was significantly higher among the 42 ­patients with evidence of perineural invasion in the primary tumor (36%) than the 29 patients without it (10%) (see Table 9-10). Major complications occurred predominantly in the patients treated during the earlier part of the study period, when large wedged-pair portals were used, radiation was administered in fractions of greater than 2 Gy, and significant dose heterogeneity existed within the target volume (see Table 9-11). Based on these observations, the following modifications in M. D. Anderson’s treatment policy were introduced in August 1989. First, radiation portals were ­designed to more generously encompass the base of the skull, including the trigeminal (gasserian) ganglion, for patients with evidence of perineural invasion and patients with adenoid cystic carcinoma. Second, elective irradiation was delivered to the ipsilateral upper neck nodes of patients with T2 to T4 squamous cell or undifferentiated carcinoma. Third, measures were taken to improve dose distribution within the target volume; these included the use of a water-filled ­balloon to reduce the size of the air cavity. A reanalysis53 was undertaken in 2007 to assess how these changes in the treatment policy have affected patient outcomes. A total of 147 patients with maxillary sinus carcinoma had received postoperative radiation therapy at M. D. Anderson between 1969 and 2002, 90 before the implementation of the revised treatment policy and 57 afterward. The first group included proportionately ­fewer patients with squamous cell carcinoma than did the

s­ econd group (53% versus 72%), but the first group had more ­patients with adenoid cystic carcinoma (28% versus 14%) and undifferentiated carcinoma (10% versus 3.5%). The presence of perineural invasion was proportionately higher in the first group (53% versus 21%), but the second group had a higher proportion of patients with unknown perineural status (15% versus 2%). The two groups were balanced with regard to age, sex, T and N categories, margin status, radiation dose, and treatment time. At mean follow-up times of 97 months for the first group and 52 months for the second group, no ­statistically significant differences were found between groups in 5-year local control rate (76% versus 68%), regional control rate (83% versus 81%), distant metastases rate (29% versus 17%), recurrence-free survival rate (51% versus 54%), or overall survival rate (52% versus 61%). Recurrence at the primary tumor site and distant metastasis remained significant contributors to mortality, accounting for more than 50% of all deaths in the entire group. In accord with the change in treatment policy to cover the skull base in patients with evidence of perineural invasion, the association between local recurrence and the presence of perineural invasion in the first group was no longer evident for the second group (i.e., those treated ­after August 1989). Before this change, 77% of the ­patients who experienced recurrence at the primary site also had perineural invasion (P = .01), but after the radiation ­portals were enlarged, only 22% of patients with disease recurrence at the primary site also had perineural invasion. Why this difference did not translate into an improvement in local control remains unclear. In the entire subgroup of 83 patients with N0 squamous or undifferentiated carcinoma, patients given elective neck irradiation (n = 45) had significantly better 5-year regional control rates (92% versus 63%; P = .0004) than patients who

196

PART 3  Head and Neck

TABLE 9-10.   Influence of Disease and Treatment Variables on Treatment Outcome in 73 Patients with Maxillary Sinus Carcinoma Number of Patients

Variable

5-Year Local Control (%)

5-Year Nodal Control (%)*

5-Year Distant Metastasis (%)

5-Year RelapseFree Survival (%)

5-Year ­DiseaseSpecific Survival (%)

Pathologic T Category† T1+2

12

91

71

17

64

75

T3

28

77

80

25

57

69

T4

29

65

93

37

44

59

N0

67

74

84

27‡

53

66‡

N1-2

6

67

82

48§

33

33

Squamous cell carcinoma

36

62

86

17‡

49

53

Undifferentiated

9

89

67

50

53

42

Adenocarcinoma

6

80

67

39

42

60

Adenoid cystic carcinoma

20

82

94

31

60

84

Mucoepidermoid

2

(2/2)

(2/2)

(0/2)

(2/2)

(2/2)

29

90‡

86

20

74‡

78

42

64

80

36

37

57

N Category

Histologic Type

Nerve

Invasion†

No Yes Margin of

Resection†

Negative

54

74

86

17

58

66

Positive

17

77

76

32

38

66

Elective Nodal Radiation No

50

77

80

31‡

51

58

Yes

17

61

100

0

61

68

*Including

salvage treatment. patients for whom pathology reports were not complete. ‡Significant at 0.05 level. §Including 11 patients who did not receive elective neck treatment and developed nodal recurrence. Modified from Jiang GL, Ang KK, Peters LJ, et al. Maxillary sinus carcinomas: natural history and results of postoperative radiotherapy. ­Radiother Oncol 1991;21:193-200. †Excluding

were not given elective neck irradiation (n = 38). This effect translated into an improvement in 5-year recurrence-free survival (67% with elective treatment versus 42% ­without; P = .01). These findings held true when the 31 patients in the first group who had not received elective neck ­irradiation were compared with the 26 patients in the second group who had received ipsilateral neck treatment according to the new policy: 5-year regional control rates were 64% for the first group and 87% for the second group (P = .03), and 5-year recurrence-free survival rates were 41% for the first group and 67% for the second group (P = .03).53 The most striking difference between the two groups was the drop in the incidence of severe complications ­after the new treatment policy was implemented. Of the entire group of 147 patients, 58 (64.4%) of the 90 ­patients in the first group experienced late complications, but only 13 (22.8%) of the 57 patients in the second group did.53

Primary Radiation Therapy Radiation therapy alone is usually selected for patients who cannot undergo radical surgery. Compilation of data from the published literature reveals that definitive radiation therapy for maxillary sinus tumors yields 5-year local ­control rates of 22% to 39% and 5-year survival rates of 22% to 40%.28,43,44,54-56 The results of radiation therapy alone have also improved over the past 4 decades, with overall survival rates generally increasing from 21% ± 13% in the 1960s to 19% ± 17% in the 1970s, 28% ± 20% in the 1980s, and 33% ± 18% in the 1990s (P < .048).28

Chemotherapy for Locally Advanced Disease Studies are ongoing to investigate whether induction chemotherapy (e.g., intra-arterial administration of cisplatin and fluorouracil, combinations of newer agents), given before or during radiation therapy, can improve locoregional control

9  The Nasal Cavity and Paranasal Sinuses

197

TABLE 9-11.   Type and Incidence of Treatment Complications Occurring in 73 Patients with Maxillary Sinus Carcinoma Complication

Number of Patients (%)

Latent Period (mo)*

Ipsilateral blindness†

13 (30)

1-36

4 (5)

13-36

Bilateral blindness Brain necrosis

5 (7)

5-73

Bone necrosis requiring débridement

4 (5)

1-17

Soft tissue necrosis/fistula

2 (3)

17, 74

Trismus

9 (12)

0-4

Pituitary insufficiency

3 (4)

60-72

Hearing loss

3 (4)

3-48

*Range

of months after radiation therapy. †Assessed in 44 patients who did not undergo ipsilateral enucleation. Modified from Jiang GL, Ang KK, Peters LJ, et al. Maxillary sinus carcinomas: natural history and results of postoperative radiotherapy. ­Radiother Oncol 1991;21:193-200.

or reduce the extent of surgery and thereby improve functional and cosmetic outcomes in patients with maxillary ­sinus tumors. A pilot study of induction chemotherapy at M. D. Anderson for patients with mostly ­advanced ­maxillary sinus cancer did not have particularly successful results. Briefly, 23 patients (12 men and 11 women, 23 to 75 years old [median age, 57 years]) who would not or could not undergo surgery were enrolled between 1969 and 1985. For 18 patients, chemotherapy consisted of intra-arterial fluorouracil infusion delivered concurrently with radiation therapy at a dose of 4 to 6 mg/kg, 5 days per week, for 2 to 4 weeks depending on the severity of acute reactions. The remaining five patients received induction chemotherapy consisting of bleomycin, fluorouracil, and cyclophosphamide administered intravenously (four patients) or bleomycin and cisplatin administered intra-­arterially (one patient). Although the acute reactions were severe, 14 of 18 patients were able to tolerate radiation doses of 60 to 61 Gy. The remaining four patients received 50 to 55 Gy delivered in 2-Gy fractions. The 5-year overall survival and relapse-free survival rates in this pilot study were 18% and 35%, respectively, and the 5-year rate of disease control above the clavicle was 39%. Of the 15 patients whose entire ipsilateral orbital content was encompassed in the radiation portals, 2 had corneal injury leading to vision impairment, and 8 had blindness as a result of retinal or optic nerve injury. One patient developed bilateral blindness as a consequence of optic chiasm injury. Bone necrosis occurred in three patients, trismus occurred in two patients, and pituitary insufficiency occurred in one patient. The conclusion from this trial was that ­intra­arterial fluorouracil, given during radiation therapy as described here, did not improve the rate of disease control above the clavicles and was associated with serious adverse effects.

Radiation Therapy Techniques An overview of the radiation therapy policies and techniques in use at M. D. Anderson is presented here; more detailed information has been published elsewhere.57,58 Because of the geometric complexities involved in tumor extension and the proximity of critical organs, the use of

intensity-modulated radiation therapy (IMRT) is gaining popularity for the treatment of sinonasal cancers. Definitions of target volumes and dose prescriptions for IMRT are summarized in Table 9-12.

Nasal Vestibule Tumors Radiation Target Volume and Dose Small (≤1.5 cm), circumscribed, well-differentiated nasal vestibule tumors usually are treated with a therapeutic dose of 60 to 66 Gy to the primary tumor with a 1- to 2-cm margin. A small portal reduction is sometimes made after 50 Gy. Alternatively, accessible lesions can be treated with brachytherapy with 192Ir wire implants to 60 to 65 Gy delivered over 5 to 7 days (Fig. 9-4). Larger (>1.5 cm) and all poorly differentiated nasal vestibule tumors without palpable lymphadenopathy are treated with irradiation of the primary tumor with margins of at least 2 to 3 cm (wider margins for infiltrative tumor) and irradiation of the bilateral facial nodes (“mustache” area), submandibular nodes, and subdigastric nodes (Fig. 9-5). When upper neck lymph node involvement is present at diagnosis, the lower neck nodes are also treated with elective irradiation. After an elective radiation dose of 50 Gy is reached, the radiation volume is reduced to cover the original gross disease with 1- to 2-cm margins to deliver a boost dose of 16 to 20 Gy by external beam. Alternatively, when the tumor is accessible, it is reasonable to deliver a boost dose of 20 to 25 Gy over about 2 days by low-dose-rate brachytherapy (see Fig. 9-5).

External Beam Radiation Therapy Setup For EBRT, the patient is immobilized while lying supine with the head positioned slightly flexed so that the anterior surface of the maxilla is parallel to the top of the couch. This setup facilitates irradiation of the primary lesion through a vertical appositional field matching with appositional portals to cover the facial lymphatics and with opposed lateral photon fields to cover the upper neck lymph nodes. Lower neck nodes are irradiated, when necessary, with an anterior field matched to the lateral upper neck field (see Fig. 9-5). Radiation is usually administered by using a ­combination of electrons and photons in a ratio of 4:1. Skin ­collimation

198

PART 3  Head and Neck

TABLE 9-12.   Target Volumes for Treatment of Paranasal Sinuses with Intensity-Modulated Radiation Therapy Target

Description

Dose (Gy)

Postoperative Adjuvant Therapy CTVTB

Primary tumor bed with 1.0- to 1.5-cm margins

60*

CTVSB

Surgical bed with 1.0- to 1.5-cm margins

56*

CTVED

Tract of V2 if perineural invasion is present, additional skull base margin, elective nodal volume if indicated

54*

CTVHD, if applicable

Sites of suspected positive margins, gross macroscopic residual tumor

66-70*

Primary Radiation Therapy GTV

Gross tumor volume (represents prechemotherapy volume if applicable)

66-70†

CTVHD

GTV + 1.0- to 1.5-cm margins

66-70†

CTVID

CTVHD + 1.0- to 1.5-cm margins

59-63†

CTVED

Tract of V2 if perineural invasion is present, additional skull base margin, elective nodal volume if indicated

56-57†

*Dose

given in 30 fractions. given in 33 to 35 fractions. CTV, clinical target volume; ED, elective dose; HD, high dose; ID, intermediate dose; SB, surgical bed; TB, tumor bed.

†Dose

Tumor

A

B

C

D

E

F

FIGURE 9-4.  A, The photograph shows a 2.5-cm, well-differentiated squamous cell carcinoma (T2N0) located on the right side of the columella. B–D, Photographs show the placement of hollow stainless steel needles inserted for afterloading with low-dose-rate 192Ir interstitial brachytherapy. E and F, Computed tomography shows the needle positions. A dose of 60 Gy was administered at 68 cGy per hour, and the total treatment time was 90.6 hours.

is used to minimize scatter irradiation to the eye. For ­electron treatments, bolus material (i.e., beeswax) is ­applied to smooth the surface contour of the nose and fill the nares, improving dose homogeneity (see Figs. 9-5 and 9-6). The bolus is removed for photon treatments to achieve skin sparing.

Nasal Cavity and Ethmoid Air Cell Tumors Radiation Target Volume and Dose For primary radiation therapy for tumors of the nasal cavity or ethmoid air cells, the initial portals encompass the primary tumor with 2- to 3-cm margins. After a dose of 50 Gy

9  The Nasal Cavity and Paranasal Sinuses

Tumor

Tumor

Right nasal ala

A

199

Right nasal ala

B

Upper lip

C

D

E

G

F 50 Gy

50 Gy

H

I

K

J

L 85 cGy/hour

M

N

O

P

85 cGy/hour

Q

FIGURE 9-5.  Photographs (A) and coronal and transverse magnetic resonance images (B) show a T2N0, poorly differentiated carcinoma of the left and right nasal septum, vestibules, and columella that was treated with external beam radiation therapy and a brachytherapy boost. C, Setup at simulation shows the tentative field wired and a bolus in the nostrils. D, Beeswax bolus was fitted to a cut-out in the thermoplastic mask to allow wax contact with skin. E, “Mustache” fields are marked for irradiation of the facial lymph nodes. F, Lateral view of the mustache fields. G, Radiograph shows that the lateral photon field encompasses nodes of levels I and II. H, Radiograph shows the anterior 20-MeV field with superimposed skin collimation (white) and machine collimation of the jaws (red shaded area). I, Port film shows the patient setup for geometric verification. J–L, Isodose distribution of the external beam irradiation to 50 Gy in 25 fractions. M, The photograph shows the needle placement for a brachytherapy boost of an additional 25 Gy at 85 cGy/hr. N, The photograph shows the needle position with the left nasal ala brought in contact with the septum by two additional sutures to minimize the air gap of the left nostril for optimal dosimetry. O, Anteroposterior radiograph was taken in the operation room. P and Q, Sagittal and transverse sections of the final brachytherapy plan show that the 85-cGy/hr line adequately covers the tumor volume (green line).

is reached, the margins are reduced to 1 to 2 cm, and an additional 16 to 20 Gy is delivered, depending on the size of the tumor. For postoperative radiation therapy, the initial target volume consists of the surgical bed with 1- to 2-cm margins, depending on the surgical pathology findings and the proximity of critical structures, to a dose of 54 Gy. A boost of 6 to 12 Gy is then given (see Figs. 9-6 and 9-7). The boost volume consists of areas at higher risk for recurrence, such as the regions of close or positive section margins or extensive perineural invasion. Elective nodal irradiation is not routinely given. For carefully selected nasal cavity carcinomas, such as those of the anteroinferior nasal septum, low-dose-rate brachytherapy may be appropriate. The dose schedule is 60 to 65 Gy given over 5 to 7 days.

The optimal radiation therapy dose-fractionation s­ chedule for the treatment of esthesioneuroblastoma has not been well established. It seems reasonable to prescribe 60 Gy in 30 fractions for primary radiation therapy and 50 to 56 Gy in 25 to 28 fractions for postoperative radiation therapy.

External Beam Radiation Therapy Setup For EBRT, patients usually are immobilized in a supine position with the head positioned such that the hard ­palate is perpendicular to the couch. An intraoral stent is used to open the mouth and depress the tongue out of the radiation fields. The tumor location and extent determine the appropriate portal arrangement. In general, for tumors located in the anterior lower half of the nasal cavity, anterior oblique wedged-pair photon fields

200

PART 3  Head and Neck

Bolus

60

Intransal bolus

60 Gy

56 50

A

B

C

60 Gy

FIGURE 9-6.  A, The patient underwent a partial rhinectomy for a recurrent squamous cell carcinoma of the nasal cavity. Bone invasion was found, but surgical section margins were negative. Portals (tumor-surgical bed and facial lymphatics) were drawn at simulation before placement of a beeswax bolus and an intranasal bolus. B, Computed tomography shows a small amount of tissue with heterogeneous enhancement (arrows). The ipsilateral facial lymphatics (mustache field), the surgical bed with wide margin (including the residual portion of the right nostril and nasal septum), and the tumor bed with a 1.5-cm margin received 50 Gy, 56 Gy, and 60 Gy, respectively, all given in 2-Gy fractions. The nostril was treated using 20-MeV electrons, and the mustache field was treated with 9-MeV electrons. A second set of skin collimation was fabricated for the cone-down boost field. A wax bolus to compensate for surface obliquity was used for the whole treatment. An additional bolus was placed in the nostrils to minimize perturbation of the electron beam with the air cavity and the air-bone interfaces. C, The dose distribution is shown in the transverse, coronal, and sagittal planes.

are ­ appropriate. For tumors of the posterior nasal fossa without extension to ethmoid air cells, opposed lateral fields may be more appropriate because these can avoid the optic pathway. For irradiation of the upper nasal cavity and ethmoidal neoplasms, a three-field technique58 is usually preferred. This setup allows coverage of ethmoid air cells without the delivery of high doses to the optic apparatus. CT-based treatment planning is necessary to determine the appropriate intersection angle between the two beams, the ­appropriate wedge angle, and the relative loading of the fields. IMRT increasingly has been used for treating tumors in this region because of better target volume coverage and avoidance of the optic apparatus (Fig. 9-8). Because elective nodal irradiation need not be delivered, thereby eliminating concerns about matching radiation portals, the use of non-coplanar beam arrangements can improve dose distribution in some cases (Fig. 9-9).

Maxillary Sinus Tumors

the contralateral optic nerve and optic chiasm be excluded from the volume receiving more than 54 Gy to minimize the risk of bilateral blindness. For tumors extending through the posterior ethmoid air cells to abut the optic chiasm, the risks and benefits of delivering a dose of up to 60 Gy, which is associated with a 5% to 10% risk of optic nerve injury, should be discussed with the patient. The guidelines for primary radiation therapy for maxillary sinus carcinoma are similar to those for postoperative radiation therapy, and the total dose ranges from 66 to 70 Gy, depending on the stage (Fig. 9-11). Normal tissue doses to the eye and optic pathway are generally applied as noted in the previous paragraphs to minimize complications. Patients with squamous cell or poorly differentiated carcinoma presenting with no clinical evidence of lymphadenopathy receive elective irradiation of the ipsilateral submandibular and subdigastric nodes to a dose of 50 Gy. Patients in whom nodal dissection reveals multiple positive nodes or the presence of extracapsular ­ extension receive adjunctive postoperative radiation therapy to 60 Gy.

Radiation Target Volume and Dose

External Beam Radiation Therapy Setup

For radiation therapy given postoperatively for maxillary sinus carcinoma, the initial portals encompass the surgical bed with margins of 1 to 2 cm, depending on the surgical pathology findings and the proximity of critical structures, usually to a dose of 54 Gy. The normal tissues of particular concern are the eye and the optic pathways. The cornea usually can be shielded in patients with limited involvement of the medial or inferior orbital wall to avoid vision impairment resulting from iatrogenic keratitis. If the ­tumor invades the orbital cavity but does not necessitate ­orbital exenteration, the lacrimal gland is shielded ­ whenever ­possible to prevent the occurrence of painful dry eye. Cataract is no longer a major concern, because this side effect has become relatively easy to correct. The boost portals cover areas where the risk of recurrence is higher, such as regions of close or positive margins or extensive perineural invasion. The boost dose usually is an additional 6 to 12 Gy (Fig. 9-10). Prudence dictates that

For EBRT, patients are immobilized in a supine position, usually with the head slightly hyperextended to bring the floor of the orbit perpendicular to the couch. This position allows irradiation of the entire orbital floor while sparing the cornea and most of the globe. An intraoral stent helps to open the mouth and depress the tongue out of the radiation fields. Marking the lateral canthi, the medial and inferior parts of the limbus while the patient gazes straight ahead, and the oral commissures, external auditory canals, and external scar before simulation films are taken is helpful in designing radiation portals. When the surgical procedure includes palatectomy, the stent can be designed to hold a water-filled balloon, which occludes the surgical defect and minimizes the dose ­heterogeneity produced by a large air cavity. In the case of orbital exenteration, the defect is filled directly with the water-filled balloon to reduce the dose delivered to the temporal lobe.

9  The Nasal Cavity and Paranasal Sinuses

201

A

C

B

D

E TARGETS, STRUCTURES, DOSES CTVtumor bed 60Gy

L optic nerve<54 R optic nerve<54 60Gy

F

G

H

54Gy

FIGURE 9-7.  A, Preoperative magnetic resonance images show a large carcinoma of the left upper nasal cavity and ethmoid sinus (red arrows), which was resected with negative margins. B–E, The dose distribution is shown in transverse sections. The target volume depth precluded the use of an electron beam. The proximity of the target volume to the optic chiasm, brainstem, eyes, and optic nerves (arrows in B) necessitated the use of intensity-modulated radiation therapy. Dose limits for the chiasm and the contralateral optic nerve were set at 54 Gy, and that for the ipsilateral optic nerve was set at 56 Gy. F–H, The dose distribution is shown in the sagittal and coronal planes.

A three-field technique with an anterior portal and two lateral fields (usually with 60-degree wedges) is preferable for most patients, particularly for those with tumors involving the suprastructure or extending to the roof of the ­nasal cavity and ethmoid air cells; this technique is described elsewhere.58 IMRT is preferred because of its superior ability to conform the dose distribution to the irregular target volumes with proper dose constraints to the critical organs (see Figs. 9-10 and 9-11). Ipsilateral upper neck irradiation is accomplished through a lateral appositional electron field (usually 12 MeV) matched to the inferior border of the primary portals. A small triangle over the cheek is not irradiated, except when there is involvement of the soft tissues of the cheek, in which case a triangular abutting electron field is added. Use of IMRT allows the nodal region to be irradiated in continuity with the primary tumor bed. Bilateral neck treatment is indicated in patients presenting with extensive nodal involvement. A proper field

matching technique is essential to minimize dose inhomogeneity and to prevent overdose to the spinal cord and mandible due to beam divergence. This can be accomplished by treating the primary tumor bed and the upper neck with asymmetric jaw matching. The central axes of the primary fields and the upper neck fields are all placed in the plane of the inferior border of the maxillary fields (usually 1 cm below the floor of the maxillary sinus). Prudence dictates moving the junction line between the primary and neck fields during the course of treatment to minimize overdosing or underdosing. The middle neck and lower neck are irradiated with an anterior appositional photon field matched to the opposed upper neck fields.

Patient Care during and after Radiation Therapy Before treatment is begun, patients should be given specific information about the acute and late side effects of radiation therapy, such as skin desquamation, ­ mucositis, nasal

202

PART 3  Head and Neck

A Rejected

TARGETS AND STRUCTURES

Pre-op tumor 60 56

Rejected

B

Rejected

C

CTVtumor bed60Gy CTVSurg bed 60Gy

D Accepted

R optic nerve<54Gy L optic nerve<54Gy

60

60

Optic chiasm<54Gy

56

56

56 Accepted

E

Accepted

F

G

60Gy 56Gy 54Gy

FIGURE 9-8.  Row A, Diagnostic magnetic resonance images show an invasive, poorly differentiated carcinoma of the left naso­ ethmoid region (red arrows). The tumor measured 5 × 4 × 1.2 cm and invaded the skull base but not the overlying dura mater. After resection, the goal was to deliver a dose of 60 Gy in 30 fractions to the tumor bed and 54 Gy to the surgical bed using intensitymodulated radiation therapy. The plan shown in B, C, and D was rejected because the 54-Gy isodose line (pink) was too close to the optic nerves and chiasm (blue arrows). The re-optimized plan shown in E, F, and G was accepted. Because this information cannot be gleaned from the dose-volume histogram, the dosimetry on each image should be inspected carefully.

NON-COPLANAR BEAM ORIENTATION Beam

Couch

Gantry

Beam

Couch

Gantry

1

000

220

7

000

100

2

000

260

8

000

140

3

000

300

9

090

030

4

000

340

10

300

355

5

000

010

11

090

345

6

000

060

FIGURE 9-9.  Non-coplanar beam arrangement (beams 9, 10, and 11) improves coverage of a carcinoma of the ethmoid sinus while minimizing exposure of the ocular apparatus.

9  The Nasal Cavity and Paranasal Sinuses

203

A 56 WB

WB 60

B TARGETS AND STRUCTURES CTVtumor bed 60Gy WB

60

CTVSurg bed 56Gy

60

CTVNodal bed 54 56

C

56

R Parotid

54

L Parotid Water balloon Optic nerve<54Gy Optic chiasm<54Gy 60Gy 56Gy

D

54Gy

FIGURE 9-10.  Row A, A large, destructive squamous cell carcinoma is centered in the left hard palate and inferior maxilla. Red ­arrows depict extension into the base of the pterygoid plates, left pterygoid musculature, floor of the left maxillary sinus, left ­maxillary alveolar ridge, and left soft palate. The tumor extent was confirmed at resection, and histologic examination showed perineural invasion. Rows B–D, Transverse, sagittal, and coronal images show the dose distribution achieved with intensity-modulated radiation therapy. The 54-Gy target volume was designed to encompass the floor of orbit and foramen rotundum because of the perineural invasion. Pink arrows point to a water balloon used to fill the space to provide scatter and reduce dose heterogeneity.

dryness, and when applicable, the risk of visual impairment. It is also prudent to monitor the skin and mucosal reactions during treatment, particularly when radiation therapy is administered with a combination of electrons and photons with a surface bolus. Moist skin desquamation or confluent mucositis is treated with pain medication and measures to prevent infection. Saline nasal rinses are begun during the period of confluent mucositis to prevent crusting and secondary infection. A topical antibiotic (i.e., neomycin and polymyxin B [Neosporin ointment]) is prescribed when infection develops. Patients are instructed to use sun block over the irradiated area after the completion of treatment. Regular follow-up examinations include evaluation of the treated area for late complications. Table 9-13 summarizes

potential side effects that can result from tumor invasion or surgical or radiation therapy.

New Directions in Therapy and Clinical Studies Surgical Techniques Craniofacial surgical techniques have been refined and ­applied in the treatment of sinonasal tumors at increasing numbers of treatment centers. These techniques ­ provide better access for complete tumor resection, particularly resection of neoplasms extending to the skull base, while preserving some functionally important structures.

204

PART 3  Head and Neck

A

B

C

D

E TARGETS AND STRUCTURES

Tumor

E

F

G

H

GTV CTV70Gy CTV63Gy CTVNodal bed 57Gy

J

I

K

L

Optic nerve<54Gy Optic chiasm<54Gy 70Gy 63Gy 57Gy

M

N

O

P

FIGURE 9-11.  A–D, Magnetic resonance images show an extensive, partially enhancing, infiltrating, undifferentiated small cell ­carcinoma of the anteromedial walls of the maxillary sinus. The tumor invaded the subdermal layers of the skin, the hard palate, and the inferior portion of the right frontal sinus. The patient was given two cycles of induction chemotherapy (cisplatin and etoposide), with minimal response, followed by intensity-modulated radiation therapy to 70 Gy in 35 fractions with concurrent cisplatin administered during weeks 1, 4, and 7. E–L, Dose distribution is shown in transverse sections. M and N, The patient had brisk skin erythema over the high-dose region, with areas of moist desquamation and crusting. Prophylactic antibiotic eyedrops were prescribed. O and P, Appearance of the skin 6 weeks after completion of therapy.

TABLE 9-13.    Side Effects Caused by Sinonasal Disease, Surgery, or Radiation Therapy Organs Affected

Clinical Manifestations

Oral and connective tissues (oral mucosa, skin and ­subcutaneous tissues, major salivary glands, soft palate ­musculature, pharynx, larynx, bones)

Soft tissue necrosis, skin fibrosis, subcutaneous fibrosis, xerostomia, nasal dryness, choanal stenosis, dental caries, dysgeusia, swallowing and voice dysfunction, trismus, cartilage and bone exposure or necrosis, persistent lymphedema

Optic apparatus (lacrimal gland, eyes, lens, optic nerves and chiasm)

Xerophthalmia, cataracts, keratopathy, retinopathy, visual ­impairment, blindness

Neural tissues (brain, brain stem, spinal cord, temporal lobe)

Neurocognitive impairment, cranial neuropathy, myelopathy, brain necrosis

Vestibulo-cochlear tissues

Vestibular dysfunction, persistent otitis, tinnitus, hearing ­impairment

Endocrine tissues (pituitary gland, hypothalamus, thyroid gland)

Multiple endocrine dysfunction: hyperprolactinemia, syndromes associated with decreased GH, FSH, LH, thyroxine, TSH, ACTH, and their downstream hormones

ACTH, adrenocorticotropic hormone; FSH, follicle-stimulating hormone; GH, growth hormone; LH, luteinizing hormone; TSH, thyroid -stimulating hormone.

9  The Nasal Cavity and Paranasal Sinuses

­ ndoscopic resection, which offers better cosmetic outE come, is being refined, and robotic surgical methods are being ­introduced.

Intensity-Modulated Radiation Therapy and Proton Beam Therapy The experience with IMRT is encouraging in terms of reducing the morbidity associated with radiation therapy. Improvements in IMRT techniques should allow ­moderate dose escalation. Proton beam therapy may offer additional advantages for some carefully selected ­ patients; however, well-designed studies are needed to generate evidence in support of the use of proton therapy for this purpose.

Chemoradiation Therapy The efficacy of new cytotoxic drugs in the treatment of sinonasal cancer continues to be assessed. These agents are being used as induction therapy, and some are ­administered intra-arterially instead of intravenously to enhance ­tumor regression. Ongoing clinical studies are addressing questions such as whether the extent of surgery can be reduced when good tumor response is achieved with chemotherapy and whether surgery can be avoided altogether for those who achieve a complete response. Several investigators are assessing the role of concurrent ­chemoradiation therapy in the management of sinonasal carcinoma. The role of chemotherapy in the management of esthesioneuroblastoma is also being addressed. Unfortunately, because of the rarity of sinonasal cancer, it will take many years to assess the relative value of these new developments in improving the overall outcome of patients with this disease.

REFERENCES   1. Roush GC. Epidemiology of cancer of the nose and paranasal sinuses. Head Neck Surg 1979;2:3-11.   2. Greenlee RT, Hill-Harmon MB, Murray T, et al. Cancer statistics, ����������������������������������������������������������������������������� 2001 CA Cancer J Clin 2001; 51:15-36.   3. Torjussen W, Solberg LA, Hgetviet AC. Histopathological changes of nasal mucosa in nickel workers: a pilot study. Cancer 1979;44:963-974.   4. Fletcher GH. Nasal and paranasal sinus carcinoma. In Fletcher GH (ed:) Textbook of Radiotherapy. Philadelphia: Lea & Febiger, 1980, pp 408-425.   5. Goepfert H, Guillamondegui OM, Jesse RH, et al. Squamous cell carcinoma of nasal vestibule. Arch Otolaryngol Head Neck Surg 1974;100:8-10.   6. Moore KL, Dalley AF. Clinically Oriented Anatomy, 5th ed. Philadelphia: Lippincott Williams & Wilkins, 2006.   7. Acheson ED, Hadfield EH, Macbeth RG. Carcinoma of the nasal cavity and accessory sinuses in woodworkers. Lancet 1967;1:311-312.   8. Acheson ED, Cowdell RH, Hadfield E, et al. Nasal cancer in woodworkers in the furniture industry. Br Med J 1968;2:587596.   9. Bars G, Visser AG, van Andel JG. The treatment of squamous cell carcinoma of the nasal vestibule with interstitial iridium implantation. Radiother Oncol 1985;4:121-125. 10. McNeese MD, Chobe R, Weber RS, et al. Carcinoma of the ­nasal vestibule: treatment with radiotherapy. Cancer Bull 1989;41: 84-87. 11. Wong CS, Cummings BJ, Elhakim T, et al. External irradiation for squamous cell carcinoma of the nasal vestibule. Int J Radiat Oncol Biol Phys 1986;12:1943-1946. 12. Chobe R, McNeese MD, Weber R, et al. Radiation therapy for carcinoma of the nasal vestibule. Otolaryngol Head Neck Surg 1988;98:67-71.

205

13. Ang KK, Jiang GL, Frankenthaler RA, et al. Carcinomas of the nasal cavity. Radiother Oncol 1992;24:163-168. 14. Beitler JJ, Fass DE, Brenner HA, et al. Esthesioneuroblastoma: is there a role for elective neck treatment? Head Neck 1991;13:321-326. 15. Goldsweig HG, Sundaresan N. Chemotherapy of recurrent esthesioneuroblastoma: case report and review of the literature. Am J Clin Oncol 1990;13:139-143. 16. Kadish S, Goodman M, Wine CC. Olfactory neuroblastoma: a clinical analysis of 17 cases. Cancer 1976;37:1571-1576. 17. Elkon D, Hightower SI, Lim ML, et al. Esthesioneuroblastoma. Cancer 1979;44:1087-1094. 18. Jiang GL, Ang KK, Peters LJ, et al. Maxillary sinus carcinomas: natural history and results of postoperative radiotherapy. Radiother Oncol 1991;21:193-200. 19 Greene FL, Page DL, Fleming ID, et al (eds). AJCC ­ Cancer ­Staging Manual, 6th ed. New York: Springer, 2002, pp 59-67. 20. Parsons JT, Mendenhall WM, Mancuso AA, et al. Malignant tumors of the nasal cavity and ethmoid and sphenoid sinuses. Int J Radiat Oncol Biol Phys 1988;14:11-22. 21. Spaulding CA, Kranyak MS, Constable WC, et al. Esthesioneuroblastoma: a comparison of two treatment eras. Int J Radiat Oncol Biol Phys 1988;15:581-590. 22. Fleming I, Cooper JS, Henson DE, et al, eds. AJCC Cancer ­Staging Manual, 5th ed. Philadelphia: Lippincott-Raven, 1997, pp 47-52. 23. Million RR. The myth regarding bone or cartilage involvement by cancer and the likelihood of cure by radiotherapy. Head Neck Surg 1989;11:30-40. 24. McCollough WM, Mendenhall NP, Parsons JT, et al. Radiotherapy alone for squamous cell carcinoma of the nasal vestibule: management of the primary site and regional lymphatics. Int J Radiat Oncol Biol Phys 1993;26:73-79. 25. Mak ACA, van Andel J, Gvan Woerkom-Eijkenboom WMH. Radiation therapy of carcinoma of the nasal vestibule. Eur J Cancer 1980;16:81-85. 26. Langendijk JA, Poorter R, Leemans CR, et al. Radiotherapy of squamous cell carcinoma of the nasal vestibule. Int J Radiat Oncol Biol Phys 2004;59:1319-1325. 27. Mazeron JJ, Chassagne D, Crook J, et al. Radiation therapy of carcinomas of the skin of nose and nasal vestibule: a report of 1676 cases by the Groupe Européen de Curiethérapie. Radio­ ther Oncol 1988;13:165-173. 28. Dulguerov P, Jacobsen MS, Allal AS, et al. Nasal and paranasal sinus carcinoma: are we making progress? A series of 220 patients and a systematic review. Cancer 2001;92:3012-3029. 29. Badib AO, Kurohara SS, Webster JH, et al. Treatment of cancer of the nasal cavity. Am J Roentgenol 1969;106:824-830. 30. Bosch A, Vallecillo L, Frias Z. Cancer of the nasal cavity. Cancer 1976;37:1458-1463. 31. Hawkins RB, Wynstra JH, Pilepich MV, et al. Carcinoma of the nasal cavity—results of primary and adjuvant radiotherapy. Int J Radiat Oncol Biol Phys 1988;15:1129-1133. 32. Bhattacharyya N. Cancer of the nasal cavity: survival and factors influencing prognosis. Arch Otolaryngol Head Neck Surg 2002;128:1079-1083. 33. Jiang GL, Morrison WH, Garden AS, et al. Ethmoid sinus carcinomas: natural history and treatment results. Radiother Oncol 1998;49:21-27. 34. Levine PA, Gallagher R, Cantrell RW. Esthesioneuroblastoma: ­reflections of a 21-year experience. Laryngoscope 1999;109: 1539-1543. 35. Loy AH, Reibel JF, Read PW, et al. Esthesioneuroblastoma: continued follow-up of a single institution’s experience. Arch Otolaryngol Head Neck Surg 2006;132:134-138. 36. Rosenthal DI, Barker JL Jr, El-Naggar AK, et al. Sinonasal malignancies with neuroendocrine differentiation. Cancer 2004;101:2567-2573. 37. Lavertu P, Roberts JK, Kraus DH, et al. Squamous cell carcinoma of the paranasal sinuses: the Cleveland Clinic experience. Laryngoscope 1989;99:1130-1136.

206

PART 3  Head and Neck

38. Le Q-T, Fu KK, Kaplan MJ, et al. Lymph node metastasis in maxillary sinus carcinoma. Int J Radiat Oncol Biol Phys 2000; 46:541-549. 39. Myers LL, Nussenbaum B, Bradford CR, et al. Paranasal sinus malignancies: an 18-year single institution experience. Laryngoscope 2002;112:1964-1969. 40. Paulino AC, Fiosher SG, Marks JE. Is prophylactic neck ­irradiation indicated in patients with squamous cell carcinoma of the maxillary sinus? Int J Radiat Oncol Biol Phys 1997;39:283-289. 41. Spiro JD, Soo KC, Spiro RH. Squamous carcinoma of the nasal cavity and paranasal sinuses. Am J Surg 1989;158:328-332. 42. Zaharia M, Salem LE, Travezan R, et al. Postoperative radiotherapy in the management of cancer of the maxillary sinus. Int J Radiat Oncol Biol Phys 1989;17:967-971. 43. Beale FA, Garrett PG. Cancer of paranasal sinus with particular reference to maxillary sinus cancer. J Otolaryngol 1983;12: 377-382. 44. Bush SE, Bagshaw MA. Carcinoma of the paranasal sinuses. Cancer 1982;50:154-158. 45. Gadeberg CC, Hjelm HM, Sgaard H, et al. Malignant tumors of the paranasal sinuses and nasal cavity. Acta Radiol Oncol 1984;23:181-187. 46. Hordiji GJ, Brons EN. Carcinoma of maxillary sinus: a retrospective study. Clin Otolaryngol 1985;10:285-288. 47. Hu Y, Tu G, Qi Y, et al. Comparison of pre- and postoperative irradiation in the combined treatment of carcinoma of maxillary sinus. Int J Radiat Oncol Biol Phys 1982;8:1045-1049. 48. Lindeman P, Eklund U, Petruson B. Survival after surgical treatment in maxillary neoplasm of epithelial origin. J Laryngol Otol 1987;101:564-568.

49. Sakai S, Hohki A, Fuchihata H, et al. Multidisciplinary treatment of maxillary sinus carcinoma. Cancer 1983;52:1360-1364. 50. Schlappack OK, Dobrowsky W, Schratter M, et al. Radiotherapy for carcinoma of the paranasal sinuses. Strahlenther Onkol 1986;162:291-299. 51. St-Pierre S, Baker S. Squamous cell carcinoma of the maxillary sinus. Head Neck Surg 1983;5:508-513. 52. Tsujii H, Kamada T, Arimoto T, et al. The role of ­radiotherapy in the management of maxillary sinus carcinoma. Cancer 1986;57:2261-2266. 53. Bristol IJ, Ahamad A, Garden AS, et al. Postoperative ­radiotherapy for maxillary sinus cancer: long-term outcomes and toxicities of treatment. Int J Radiat Oncol Biol Phys 2007;68:719-730. 54. Amendola BE, Eisert D, Hazra TA, et al. Carcinoma of the maxillary antrum: surgery or radiation therapy? Int J Radiat Oncol Biol Phys 1981;7:743-746. 55. Frich JC. Treatment of advanced squamous carcinoma of the maxillary sinus by irradiation. Int J Radiat Oncol Biol Phys 1982;8:1453-1459. 56. Olmi P, Cellai E, Chiavacci A, et al. Paranasal sinuses and nasal cavity cancer: different radiotherapeutic options, results and late damages. Tumori 1986;72:589-595. 57. Ang KK, Garden AS. Radiotherapy for Head and Neck Cancers: Indications and Techniques, 2nd ed. Philadelphia: Lippincott Williams & Wilkins, 2002. 58. Ang KK, Garden AS. Radiotherapy for Head and Neck Cancers: Indications and Techniques. 3rd ed. Philadelphia: Lippincott Williams & Wilkins, 2006.

The Nasopharynx 10 ADAM S. GARDEN ANATOMY ROUTES OF DISEASE SPREAD STAGING HISTOPATHOLOGIC TYPES EPIDEMIOLOGY

THERAPEUTIC APPROACHES Radiation Therapy Chemotherapy Salvage Therapy for Local Recurrence NASOPHARYNGEAL CARCINOMAS IN CHILDREN

The most common type of malignancy arising in the ­nasopharynx, as in other head and neck sites, is ­carcinoma. However, malignancies in the nasopharynx are different from malignancies at other head and neck sites with respect to their epidemiology, histopathologic features, patterns of disease spread, and management. These differences prompted the American Joint Committee on ­Cancer (AJCC) to create a separate staging system for ­nasopharyngeal carcinoma (NPC) (see “Staging” later in this ­chapter).1

Anatomy The nasopharynx is superior to the other two subdivisions of the pharynx, the oropharynx and hypopharynx. The nasopharynx is a cuboidal cavity formed by muscle and fascia with an epithelial mucosal covering. The roof of the soft palate forms the inferior border of the nasopharynx and creates the junction between the nasopharynx and the oropharynx. Defining this junction clinically can be difficult because the palate slopes; the clinical and ­radiographic definitions of disease extent into the oropharynx are challenging. The nasopharynx is bound anteriorly by the posterior end of the nasal septum and the choanae; posteriorly by the vertebral bodies, the atlas, and the axis; and superiorly by the sphenoid sinus and the basisphenoid (Fig. 10-1). The muscular component of the nasopharynx is the superior pharyngeal constrictor, which emanates from the pharyngeal tubercle. Lateral to the nasopharynx are the parapharyngeal space and the masticator space (Fig. 10-2). These spaces are separated from the nasopharynx by the pharyngobasilar fascia. The fascia arises from the base of the skull and fans laterally just anterior to the carotid canal. The fascia then continues anteriorly and medially and inserts into the posterior aspect of the medial pterygoid plate. The eustachian tubes descend medially to enter the lateral nasopharynx. The posterior aspect of each tube’s orifice creates a protrusion called the torus tubarius (Fig. 10-3). Posterior to the torus is the lateral recess known as the pharyngeal recess, or Rosenmüller’s fossa. The lateral and posterior walls and the roof of the nasopharynx are covered by mucosa.

Routes of Disease Spread NPC most commonly arises in Rosenmüller’s fossa (Fig. 10-4A). Anterior extension of tumor along the lateral wall of the nasopharynx can partially or completely occlude the pharyngeal orifice of the eustachian tube, causing hearing loss or otitis media, or both. In the United States, where NPC is uncommon, these presenting symptoms can often go undiagnosed for months because of a low index of suspicion. Further anterior tumor extension or growth causes nasal obstruction or epistaxis (see Fig. 10-4B). Superior extension of NPC can result in bone erosion. Destruction of the clivus (see Fig. 10-4C) often results in headache. Further extension can involve the second or third division of the trigeminal nerve by infiltration into the superior maxillary foramen or the oval foramen of the sphenoid bone. Another mechanism of perineural involvement is invasion into the cavernous sinus (see Fig. 10-4D). In this scenario, the abducens nerve (cranial nerve VI) is usually damaged, resulting in diplopia. Extension of ­disease

Ethmoid sinuses Sphenoid sinus

Torus tubarius Rosenmüller's fossa and orifice Soft palate

FIGURE 10-1.  Diagram of a sagittal view of the nasopharynx shows its walls.

207

208

PART 3  Head and Neck

2 1

**

3

* 4

FIGURE 10-3.  Photograph of the nasopharynx as seen through a fiberoptic scope. 1, posterior wall; 2, fossa of Rosenmüller; 3, torus tubarius; 4, eustachian tube orifice. FIGURE 10-2.  Axial image of a normal nasopharynx and its relation to the masticator muscles. Single asterisk, nasopharynx; double asterisks, pterygoid muscle.

into the orbital apex (see Fig. 10-4E) can have ­ further ­effects on vision. Direct intracranial extension (see Fig. 10-4F) is seen in less than 10% of cases. Despite the lack of a barrier against spread, inferior extension of tumor into the oropharynx is uncommon. ­Occasionally, an examiner will notice a submucosal asymmetry or bulging of the posterior pharyngeal wall that may be a result of submucosal spread of tumor (Fig. 10-5A) or a clinical manifestation of extensive retropharyngeal ­adenopathy. Posterior and lateral infiltration of tumor beyond the pharyngobasilar fascia can result in involvement of cranial nerves IX to XII as they exit the jugular foramen and ­hypoglossal canal (see Fig. 10-5B), leading to cranial neuropathies. Direct lateral extension can result in invasion of the masticator space (see Fig. 10-5C), producing trismus as a presenting symptom.

Staging The staging system for NPC is the tumor-node-­metastasis (TNM) classification outlined in the sixth edition of the AJCC Cancer Staging Manual (Box 10-1).1 The current staging system, published in 2002, does not differ ­substantially from the previous staging system published in 1997; however, the 1997 version represented a significant change from the previous (1992) system, which must be borne in mind in reviewing the older literature. Because of reports suggesting that subsite within the nasopharynx was not prognostically relevant, the current staging system does not recognize subsite divisions. Tumors involving the ­oropharynx or nasal cavity are now considered T2 ­ rather than T3. Tumors that extend into the ­ parapharyngeal

space, defined as posterolateral infiltration beyond the ­pharyngobasilar fascia, are considered T2b. The prognostic significance of parapharyngeal extension is debatable, because some investigators have found it to be associated with an increase in local recurrence.2 The question is whether this apparent tendency for local recurrence reflects more advanced disease or the tumor more prone to being excluded from radiation portals because of its location. Bony erosion of the base of the skull without cranial nerve involvement is thought to be a more favorable presentation than perineural or intracranial involvement, and it has been downgraded from T4 to T3. The nasopharynx has a rich submucosal network of lymph­ atics. Clinical lymphadenopathy is common in NPC, observed in 70% of patients at presentation, and some suggest that if subclinical nodal metastases are included, as many as 90% of patients with NPC have nodal involvement at presentation. Involvement of retropharyngeal lymph nodes (Fig. 10-6A), posterior cervical nodes (level V) (see Fig. 10-6B), and jugular chain nodes (levels II to IV) is common. Clinical lymphadenopathy without other signs or symptoms is a common presentation of NPC. Because patterns of nodal spread in NPC are different from patterns of nodal spread in other head and neck cancers and because the implications of nodal involvement are different in NPC from those for other head and neck cancers, NPC has a unique nodal staging scheme (see Box 10-1). Notable differences between N staging for NPC and other head and neck cancers include the following: (1) Because the nasopharynx is considered a midline structure, the concept of contralateral nodes was eliminated in staging NPC; (2) because size differences are not as critical in NPC, only a 6-cm cutoff is used; and (3) because nodal levels are significant in NPC, ­staging based on involvement of the supraclavicular fossa is ­included.

10  The Nasopharynx

A

209

B

C

D

E

F FIGURE 10-4.  A, Magnetic resonance image of an early T1 nasopharyngeal carcinoma. B, Extension of tumor originating in the left fossa of Rosenmüller into the nasal cavity. C, Clival destruction by tumor. D, Invasion of the cavernous sinus by tumor. E, Orbital invasion by tumor. F, Intracranial extension of tumor.

210

PART 3  Head and Neck

A

B

C

FIGURE 10-5.  A, Extension of a nasopharyngeal carcinoma into the oropharynx. B, Posterior extension beyond the pharyngobasilar fascia that invades cranial nerves IX to XII. C, Invasion into the masticator space.

The validity of the 2002 staging system was tested through retrospective analysis of two large datasets from China.3,4 In both analyses, the investigators concluded that use of the 2002 system produced acceptable distributions of patient numbers and segregation of survival curves among the various stage groupings but that the prognostic accuracy of the staging system could be improved by refining the criteria for T, N, and stage groupings.

Histopathologic Types Carcinomas are the most common malignancies of the nasopharynx, accounting for about 90% of all cases. Because the nasopharynx is part of Waldeyer’s tonsillar ring and is rich in lymph tissue, lymphomas occasionally arise in the nasopharynx, and many carcinomas in the nasopharynx have a lymphoid stroma. A separate entity, lymphoepithelioma, has also been described in the nasopharynx; lymphoepithelioma is essentially a squamous carcinoma with a benign lymphoid component. Historically, compared with squamous carcinoma, lymphoepithelioma was thought to be associated with higher locoregional control rates ­ because of its pronounced radiosensitivity but with equivalent survival rates because of its higher rate of distant recurrence. However, the significance and the incidence of lymphoepithelioma are now thought to be less than previously ­described. According to the accepted World Health Organization (WHO) histopathologic classification,5 NPC occurs as one of three types: (1) keratinizing squamous cell carcinoma, (2) nonkeratinizing squamous cell carcinoma, and (3) undifferentiated carcinoma; lymphoepitheliomas are considered type 2 or type 3 lesions. Among populations in which NPC is endemic, type 3 accounts for almost 99% of cases of NPC. WHO type 1 is more common in the United States, where it accounts for approximately 20% of cases.

Epidemiology NPC is endemic in southeast China, where the incidence approaches 80 cases per 100,000 people per year. In contrast, in the United States, the incidence of NPC is less than

2 cases per 100,000 people per year. Other areas across Southeast Asia, North Africa, and Saudi Arabia have an intermediate incidence.6 NPC has a male predominance, although the male-to-female incidence ratio is lower for NPC than for other types of head and neck cancer. The disease also affects a younger population than other types of head and neck cancer; the median age at diagnosis is 50 to 60 years.7 A strong association exists between Epstein-Barr virus and endemic NPC,8,9 with immunoglobulin A (IgA) antiviral capsid antigen present in more than 80% of patients with NPC in endemic areas.10 Before and after treatment, circulating levels of Epstein-Barr virus DNA in the ­serum, measured by real-time polymerase chain reaction techniques, are associated with prognosis in endemic populations11 and may be useful in monitoring response to therapy.8,11 The high rates of NPC seen in southern China may also be the result of genetic predisposition. Several ­cytogenetic abnormalities have been described in NPC, including changes on the human leukocyte antigen major histocompatibility complex alleles.12 Alterations in the tumor ­suppressor gene CDKN2A (formerly designated p16) have been described in 35% of patients with primary undifferentiated NPC.13 However, the TP53 (formerly called p53) mutations seen in other types of head and neck cancer are less common in endemic NPC.14 Environmental factors may play a role in NPC. Various dietary components have been studied in populations at risk15 and have been implicated as direct carcinogens or Epstein-Barr virus activators. Further evidence of an environmental influence is the drastically lower incidence of NPC in United States–born descendants of Chinese individuals than the incidence of the disease in immigrants. This difference is thought to reflect the separation of the underlying genetic contribution to this disease from the environmental component. In the United States, approximately 20% of NPC cases are WHO type 1 (i.e., keratinizing squamous cell carcinoma), which is thought to be caused by alcohol and tobacco use16 and may behave like the tobacco-related squamous cell carcinomas at other head and neck sites.

10  The Nasopharynx

211

BOX 10-1.  American Joint Committee on Cancer Staging System for Nasopharyngeal Carcinoma Primary Tumor (T) TX T0 Tis T1 T2 T2a T2b T3 T4 Regional Lymph Nodes (N) NX N0 N1

Primary tumor cannot be assessed No evidence of primary tumor Carcinoma in situ Tumor confined to the nasopharynx Tumor extends to soft tissues of oropharynx Tumor extends to the oropharynx and/or nasal cavity without ­parapharyngeal extension* Any tumor with parapharyngeal extension Tumor involves bony structures and/or paranasal sinuses Tumor with intracranial extension and/or involvement of cranial nerves, infratemporal fossa, hypopharynx, orbit, or masticator space

N3 N3a N3b

Regional lymph nodes cannot be assessed No regional lymph node metastasis Unilateral metastasis in lymph node(s), 6 cm or less in greatest ­dimension, above the supraclavicular fossa† Bilateral metastasis in lymph node(s), 6 cm or less in greatest dimension, above the supraclavicular fossa† Metastasis in a lymph node(s)† >6 cm and/or to supraclavicular fossa Greater than 6 cm in dimension Extension to the supraclavicular fossa‡

Distant Metastasis (M) MX M0 M1

Distant metastasis cannot be assessed No distant metastasis Distant metastasis

N2

Stage Grouping: Nasopharynx Stage 0 Stage I Stage IIA Stage IIB

Stage III

Stage IVA

Stage IVB Stage IVC

Tis T1 T2a T1 T2 T2a T2b T2b T1 T2a T2b T3 T3 T3 T4 T4 T4 Any T Any T

N0 N0 N0 N1 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 M0 M1

*Parapharyngeal extension denotes posterolateral infiltration of tumor beyond the pharyngobasilar fascia. †Midline

nodes are considered ipsilateral nodes. zone or fossa is relevant to the staging of nasopharyngeal carcinoma and is the triangular region originally described by Ho. It is defined by three points: (1) the superior margin or the sternal end of the clavicle, (2) the superior margin of the lateral end of the clavicle, and (3) the point where the neck meets the shoulder. This includes caudal portions of levels IV and V. All cases with lymph nodes (whole or part) in the fossa are considered N3b. From Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 33-45. ‡Supraclavicular

Therapeutic Approaches The primary therapy for NPC is still largely radiation. Over the past several decades, the role of systemic therapies, given sequentially or in combination with radiation, in improving treatment outcome in NPC has been

i­nvestigated, and several meta-analyses have indicated that the addition of chemotherapy to conventionally fractionated radiation therapy confers a survival advantage in ­locoregionally advanced NPC. Improvements in radiation therapy techniques have shown promise for ­ improving

212

PART 3  Head and Neck

B

A

FIGURE 10-6.  A, Metastasis to a retropharyngeal node from a nasopharyngeal cancer. B, Posterior cervical node involvement by nasopharyngeal cancer.

A

B

FIGURE 10-7.  Lateral simulation films for a T2N0 nasopharyngeal cancer (A) and a T4 node-positive nasopharyngeal cancer (B).

local control rates and minimizing damage to normal ­tissues.

Radiation Therapy The goal of radiation therapy for NPC is to treat ­clinically apparent disease and sites of subclinical lymphatic spread. Defining these sites of subclinical spread remains a ­challenge even with the advent of sophisticated imaging technologies.

Treatment Planning In 1980, Fletcher and Million,17 who did not have access to the high-tech tools available today, used knowledge of patterns of metastasis in NPC to define targets to ensure adequate coverage of the sites of potential tumor spread. For early-stage tumors, the radiation portals encompassed the nasopharynx proper; the posterior 2 cm of the nasal cavity; the posterior ethmoid sinuses; the entire sphenoid sinus and basiocciput; the cavernous sinus; the base of the skull, including the oval foramen, spinous foramen, and carotid canal; the pterygoid fossae; the posterior third of the orbit;

the posterior third of the maxillary sinus; the lateral and the posterior oropharyngeal wall to the midtonsillar level; and the retropharyngeal nodes. In patients with base of skull or intracranial nerve involvement, the superior ­border was raised to include all of the pituitary fossa, the base of the brain in the suprasellar area, the adjacent middle cranial fossa, and the posterior portion of the anterior cranial fossa.17 The basic radiation therapy technique for NPC is a three-field technique: lateral opposed fields for the primary tumor and upper neck and a matching anterior low-neck portal (Fig. 10-7). The use of computed tomography (CT) and magnetic resonance imaging (MRI) has allowed better definition of the extent of disease. Fields are designed so that 2 cm of tissue around the gross disease outlined on imaging receives a subclinical dose of 50 Gy. A reduction is made at approximately 42 Gy off the posterior border to shield the spinal cord. These posterior tissues are then treated with a supplemental dose of electron beams that target the superficial node-bearing posterior tissues while limiting the dose to the deeper spinal cord. Boost fields are designed to deliver 66 to 70 Gy to gross disease

10  The Nasopharynx

with a 1- to 1.5-cm margin. These boost fields can continue through lateral fields, but the challenge is to adequately cover the tumor while limiting the dose to adjacent normal structures, particularly the normal neural structures, including the spinal cord, brainstem, and in advanced cases with significant superior extension, the optic chiasm. Treatment of NPC often presents technical challenges because of the proximity of the tumor to the brainstem and its often-concave shape. The latest approach to radiation therapy for NPC, intensity-modulated radiation ­therapy (IMRT), has largely replaced two-dimensional or three-dimensional conformal radiation therapy in treatment centers at which this technology is available. IMRT was shown by two groups at the University of California in San Francisco (UCSF) and the Memorial Sloan-­Kettering ­Cancer Center to have clear dosimetric advantages over more conventional approaches—tumor coverage improved, and doses to critical neural structures were reduced.18,19 Subsequently published clinical outcomes demonstrated that IMRT could produce high local and regional control rates in NPC.20,21 Most treatment centers that began IMRT programs in the late 1990s and early 2000s use an anterior low-neck field matched to the inferior border of the IMRT fields to treat the upper neck and nasopharyngeal disease. A block is placed centrally to shield the larynx and as a margin of safety to avoid accidental overlap on the spinal cord. If the treatment is directed at subclinical metastases only, 50 Gy is sufficient. If gross disease is present, that disease needs to receive a boost to 66 to 70 Gy. Because nodal disease is often present in the posterior cervical nodes (level V nodes), a posterior field is often required. At many centers, IMRT is used to treat the entire neck. This treatment requires defining avoidance structures in the lower neck to minimize the dose to the brachial plexus, cervical esophagus, and larynx.

Dose and Fractionation NPC usually is treated with conventionally fractionated ­radiation: 2-Gy fractions given once daily. The recommended total radiation dose for the primary tumor is 66 to 70 Gy, with the higher dose advised for most patients except those with T1 disease, delivered in 33 to 35 fractions. Dose recommendations for nodal disease are as follows: 50 Gy for subclinical nodal metastases, 66 Gy for nodes smaller than 3 cm, and 70 Gy for nodes 3 cm or larger. In NPC, as in other types of cancer of the head and neck, prolonging or interrupting radiation therapy can compromise the outcome. Kwong and colleagues22 reported a 3.3% decrease in the locoregional control rate for each day of treatment interruption in their patients treated with radiation therapy. However, unlike the case for other head and neck cancers, no firm evidence is available to support the use of altered fractionation for patients with NPC outside of the setting of clinical trials.23,24 Teo and colleagues25 conducted a randomized trial comparing hyperfractionated radiation (71 Gy in 40 fractions over 30 days) with conventional fractionation (60 Gy in 24 fractions over 30 days) for patients with NPC; no difference was observed in disease control rates, but the hyperfractionated schedule was associated with increased central nervous system toxicity.25 In a retrospective review, Lee and colleagues26 found that shorter overall treatment time had an adverse effect on the incidence of temporal lobe necrosis and that this effect

213

was exacerbated by twice-daily fractionation.26 However, both experiences involved the use of large conventional fields in which a significant volume of brain was potentially encompassed in the treatment portals, and the retrospective study involved the use of fractions larger than 2 Gy. A separate pilot study from Lee and colleagues27 suggested that six fractions of 2 Gy/fraction had a ­therapeutic ­advantage over five fractions per week (also at 2 Gy/ fraction)—an improvement in 3-year progression-free survival rate—with no excess late toxicity.27 These results led to the design of NPC-9002, a randomized trial to further evaluate the effect of accelerated fractionation for NPC.28 In this four-arm trial, patients were randomly assigned to receive conventionally fractionated or accelerated-fraction radiation therapy with or without concurrent chemotherapy. The trial was closed early because of slow accrual, but preliminary results suggested that patients treated with accelerated fractionation and concurrent chemotherapy had the best failure-free survival rates, albeit at the possible cost of increased late effects. Among patients who did not ­receive chemotherapy, outcomes were no ­different for those who underwent conventional or accelerated­fractionation ­radiation therapy. The total radiation dose used to treat NPC has gradually been escalated over time. In the 1950s and 1960s, patients treated at The University of Texas M. D. Anderson Cancer Center received an average dose of 61 Gy. Fletcher, recognizing that the location of the nasopharynx within typical portal arrangement created cold spots,17 recommended a makeup dose of 5 to 7.5 Gy. This approach resulted in an average dose to the primary tumor of 67 Gy during the 1970s and an average dose of 70 Gy more recently. The literature is mixed regarding the existence of a dose-effect relationship for treatment of NPC. Analysis of dose in 378 patients treated with radiation alone at M. D. Anderson Cancer Center revealed no clear trend for higher control rates with higher doses; the best control rates were seen in patients who received 60 to 65 Gy or more than 70 Gy. Kaasa and colleagues29 reported no correlation between radiation dose and survival time among 122 patients treated at the Norwegian Radium Hospital. In contrast, Perez and colleagues30 at the Mallinckrodt Institute of Radiology reported a clear benefit in local control when doses of greater than 66 Gy were used for T1 to T3 ­disease. However, no advantage was seen for higher doses in T4 ­disease (defined according to the 1988 AJCC staging ­system). A subsequent study by Willner and colleagues31 found that correcting the dose-response curve for tumor volume created a steep dose-response relationship and that tumors larger than 64 mm3 were unlikely to be controlled with 72 Gy. However, that study did not correct for the influence of histologic type on outcome. Most studies conducted during the past 10 years have used radiation doses of 66 to 70 Gy, given in combination with concurrent cisplatin, for NPC. Although many NPCs, especially those of WHO type 2 or 3 histology, are sensitive to chemotherapy and radiation therapy, exactly which patients can be safely treated with reduced-dose concurrent therapy remains unclear. Moreover, T3 or T4 disease, even with concurrent chemotherapy, still presents a challenge in terms of disease control, leading one group32 to explore the feasibility of using IMRT for dose escalation to 76 Gy.

214

PART 3  Head and Neck

Results of Radiation Therapy Outcomes for patients with NPCs have improved over the past 4 decades, in part because of improvements in treatment techniques but also because of progress in ­diagnostic imaging, which has enabled better disease staging and ­occasionally has resulted in stage migration. Survival rates are different among series (Table 10-1),30,33-44 and they depend on the stage distribution, the characteristics of the treated population, and the type of treatment delivered. Series from China and Hong Kong are composed almost exclusively of patients with undifferentiated NPC. In areas where NPC is endemic, a greater proportion of patients have earlier-stage disease because the symptoms and signs of NPC are more likely to be recognized in a timely manner. One of the largest series ­reported is from Queen Elizabeth Hospital in Hong Kong. This retrospective analysis of more than 5000 patients found a 10-year overall survival rate of 43%.33 At M. D. ­Anderson, by comparison, the 10-year actuarial survival rate among 378 patients treated between 1954 and 1992 was 34%.34 In another large study of more than 2600 ­patients ­treated in Hong Kong from 1996 through 2000, Lee and colleagues45 reported a 5-year overall survival rate of 75% (90% for patients with stage I disease and 58% for ­patients with stage IV disease). More than three fourths of the patients in that series had been treated with radiation alone. Local recurrence is the dominant relapse pattern in patients with NPC treated with radiation therapy alone. The 10-year local control rates in the Queen Elizabeth Hospital and the M. D. Anderson series were similar at 61% and 66%, respectively.33,34 Results from studies of chemoradiation therapy for NPC are encouraging, but large series with long-term follow-up are still needed. Numerous series have demonstrated the association between advanced tumor stage and poorer local control. Perez and colleagues30 reported 10-year primary tumor control rates of 85% for T1 disease and 40% for T4 disease (disease staged according to the 1988 AJCC staging system). These

findings are comparable to the 10-year primary tumor control rates at M. D. Anderson: 87% for T1 disease and 45% for T4 disease.34 In a later analysis from Hong Kong, the overall 5-year local control rates for all patients, with disease at any stage, was 85%,45 a considerable improvement that may have resulted from the use of MRI to assess the extent of local disease and a shift in the dominant pattern of failure from local to distant. Among patients with T4 disease, those with cranial nerve involvement have a lower probability of local control. Parapharyngeal space infiltration is also correlated with a poorer prognosis, although this relationship is not universally accepted. In series from areas where NPC is not endemic, the WHO type 1 subtype (i.e., keratinizing squamous cell carcinoma) is often regarded as having an unfavorable prognosis. In the M. D. Anderson series,34 the 10-year actuarial control rate for squamous cell carcinoma was only 54%, compared with 79% for lymphoepithelioma and 71% for unclassified carcinoma. With comprehensive nodal irradiation, the regional control rate for NPC is relatively high. The 10-year regional control rate of 64% reported from Queen Elizabeth Hospital33 resulted in part from a policy of not electively treating the neck in the absence of palpable lymph nodes; the regional failure rate in the subset of 906 patients who did not undergo irradiation of the neck was 40%.33 In the M. D. Anderson experience, systematic irradiation of the neck produced a 10-year regional control rate of 83%.34 Chow and colleagues46 at the Princess Margaret Hospital in Toronto compared regional control rates for patients with NPC and known lymph node metastases with those for patients with lymph node metastases from disease at other head and neck mucosal sites.46 The findings from that matched-case study indicated that, stage for stage, ­regional control was better for patients with NPC than for those with other types of cancer. Distant metastases are more common in NPC than in tumors at other head and neck sites. The bones and lungs are the most common sites. Nodal status is the strongest

TABLE 10-1.    Results of Selected Series of Radiation Therapy Only for Treatment of Nasopharyngeal Cancer Study and Reference Lee et al, Huang,

199233

198036

Qin et al, 198837 Zhang et al, 198938 Sanguineti et al,

199734

199735

Study Period

No. of Patients

Survival Rate (%)

Local Control Rate (%)

1976-1985

5037

43 (10 yr)

61 (10 yr)

1958-1973

1605

32 (5 yr)

84 (5 yr)

1958-1978

1379

41 (5 yr)

NS

1974

604

33 (10 yr)

NS

1954-1992

378

48 (5 yr)

71 (5 yr)

1970-1994

259

59 (5-yr DSS)

65 (5 yr)

Johansen et al,

199239

1975-1984

167

37 (10 yr)

54 (10 yr)

Laramore et al,

198840

1975-1985

166

23-39 (4 yr)

64 (crude rate)

1978-1991

165

53 (5-yr DSS)

72 (5 yr)

1956-1986

143

10-40 (10 yr)

40-85 (10 yr)

1971-1980

134

38 (5 yr)

NS

1970-1980

107

NS

69 (crude rate)

1955-1990

103

58 (5 yr) 30 (5-yr DFS)

68 (crude rate)

Wang,

Olmi et al, 199241 Perez et al, 199230 Dickson and Flores, Vikram et al, Bailet et al,

198543

199244

198542

DSS, disease-specific survival; DFS, disease-free survival; NS, not stated.

10  The Nasopharynx

predictor of distant failure. The highest rate of metastases (57%) reported by Lee and colleagues33 was found among patients with Ho’s N3 disease (i.e., nodes involving the supraclavicular region or the skin). In the M. D. ­Anderson series, the actuarial 10-year rate of distant disease was 32%, and the risk was greatly increased in the presence of ­advanced neck disease or low-neck disease.47 The improved outcomes reported with the use of IMRT, with or without concurrent chemotherapy, reflect improved local control; these treatment approaches have had little effect on the control of distant disease. In most modern series, distant failure has replaced local and regional failure as the dominant pattern of recurrence.

Complications of Radiation Therapy The large portals used in radiation therapy for NPC can induce acute reactions, such as dermatitis and mucositis, and late complications. Treating the masticator spaces can result in trismus. Radiation therapy for NPC often causes xerostomia, and in younger patients, who are less likely to have lost all of their natural teeth, this can lead to late osteoradionecrosis of the mandible and maxilla. Chronic ulceration of skin and soft tissue occurs infrequently. Endocrine sequelae should not be underestimated. Neck irradiation can result in hypothyroidism. Large fields encompassing the sphenoid sinus, base of skull, and cavernous sinus can result in hypopituitarism. The most severe sequelae of radiation therapy for NPC are neurologic. The presence of large tumors (T3 or T4) often necessitates coverage of the orbits and optic chiasm. If the dose to these structures is not kept within the limits of tolerance, blindness can result. Hearing loss is a common presenting symptom and can be a complication of radiation therapy. Hearing loss is likely to be more common in patients treated with the combination of cisplatin, which is ototoxic, and radiation.48 IMRT may be helpful in sparing the cochlea in some cases. Cranial nerve damage, particularly to nerves IX to XII, is possible, and bilateral damage can be devastating. The brainstem and spinal cord are potentially at risk, particularly in cases of bilateral parapharyngeal disease extension. Treating patients to 70 Gy or greater doses can lead to memory loss and potentially temporal lobe necrosis, resulting in seizure or strokelike presentations. Better tumor definition with more tailored fields, the use of conformal techniques, especially IMRT, and the use of a brachytherapy boost may help to reduce the incidence of these potentially severe complications. Few late complications have been reported after the use of IMRT for NPC.

Improving the Therapeutic Window Intensity-Modulated Radiation Therapy. At many centers, IMRT has become the standard external beam radiation therapy (EBRT) technique for the treatment of NPC. In two separate assessments, Xia and colleagues at UCSF19 and Hunt and colleagues at Memorial Sloan-­Kettering Cancer Center18 showed that IMRT provided better ­tumor coverage, with greater percentages of the target tumor volumes receiving the planned prescription dose than a conventional three-field treatment or a three-dimensional conformal plan. In addition to improved tumor ­coverage, IMRT can deliver lower doses to critical neural structures such as the spinal cord, brainstem, and temporal lobes ­

215

(Fig. 10-8). In a follow-up report from UCSF,20 67 patients with NPC had been treated with IMRT, 50 with concurrent chemotherapy and 26 with a brachytherapy boost. At a median follow-up time of 31 months, only two patients had had local or regional disease recurrence. Eight patients had grade 3 or 4 late toxicity. The experience was similar at Memorial Sloan Kettering21; at a median follow-up time: of 35 months, only 6 of 74 patients treated with IMRT had had local disease recurrence. Rates of severe xerostomia and ototoxicity were low. Several Asian series have also described high local control rates and low rates of severe xerostomia with the use of IMRT for NPC.49,50 Boost Approaches: Brachytherapy and Stereotactic ­ Radiosurgery. Because it is essentially a cavity lined by mucosa, the nasopharynx is a site amenable to brachytherapy. The appeal of brachytherapy for delivering boost doses within the nasopharynx is the ability to reduce EBRT doses to normal adjacent structures, particularly the brainstem and temporal lobes. Brachytherapy can be delivered by interstitial or intracavitary implant techniques. Interstitial approaches use gold 198 (198Au) grain implants or permanent iodine 125 (125I) or iridium 192 (192Ir) wires. Intracavitary approaches, which are preferred for their practicality, involve placing radioactive sources into tubes or manufactured applicators that can been inserted in the nasopharynx such that the radioactivity is placed against the nasopharyngeal mucosa. Historically, intracavitary techniques used low-­dose-rate sources. Wang35 described using a cesium boost ­technique as a component of therapy for patients with T1 to T3 disease (according to the 1988 staging system). In more recent years, low-dose-rate therapy has been replaced by highdose-rate techniques in which the use of easily ­ inserted, commercially available applicators and the rapidity of treatment minimize the discomfort associated with intracavitary therapy. Levendag and colleagues51 in Rotterdam reported a protocol in which patients received 60 to 70 Gy as EBRT followed by 11 to 18 Gy delivered as high-doserate brachytherapy in doses of 3 to 4 Gy per session. The 3-year local control rates for patients with T1 or T2 disease exceeded 90%.51 Severe complications were uncommon and consisted principally of nasal synechiae. Chang and colleagues52 described their experience treating 133 patients with T1 to T2 NPC with 64.8 to 68.4 Gy of EBRT followed by 1 to 3 fractions of 5 to 5.5 Gy of high-dose-rate brachytherapy. These investigators also reported improvement in local control with brachytherapy but found a moderate incidence of tissue perforation and necrosis, which correlated with the total dose, brachytherapy fraction number, and tumor size. They concluded that total doses should range from 72.5 to 75 Gy, with highdose-rate brachytherapy given in 2.5-Gy sessions. They prescribed their boost doses at 2 cm from the central axis, which resulted in a high mucosal surface dose and most likely contributed to the high incidence of complications. The advantage of brachytherapy—the ability to limit the dose to the mucosa with rapid fall-off—is a limitation for patients with disease that extends posteriorly into the parapharyngeal space or superiorly into the clivus or brain. In these patients, one potential means of delivering additional therapy is by a stereotactic radiosurgical boost. This method can deliver a boost to sites of disease remote from the nasopharynx while limiting the dose to normal nearby

216

PART 3  Head and Neck

A

B

C

D

FIGURE 10-8.  A, Intensity-modulated radiation therapy (IMRT) plan for treating a T1 N3 nasopharyngeal carcinoma. After induction chemotherapy, IMRT was used to treat the nasopharynx to a dose of 66 Gy while keeping the mean dose to the bilateral parotid glands below 26 Gy and avoiding the brainstem. Red are indicates the clinical target volume (CTV1). B, IMRT plan for treating a T2 nasopharyngeal carcinoma to 70 Gy with concurrent cisplatin chemotherapy. The CTV1 (red) was to be treated to 70 Gy and the CTV2 (blue area around CTV1) was to be treated to 60 Gy. C, IMRT plan for a nasopharyngeal cancer with gross right-sided retropharyngeal adenopathy to be treated with concurrent cisplatin chemotherapy. CTV1 (red) was to be treated to 70 Gy. D, IMRT plan for treating a nasopharyngeal carcinoma involving the posterior cervical nodes. After induction chemotherapy, the posterior nodal regions were delineated as CTV1 (red), which was to be treated to 70 Gy.

tissue. At Stanford University, patients have been treated with EBRT (median dose, 66 Gy) combined with chemotherapy followed by a stereotactic radiosurgical boost within 4 weeks of their EBRT.53 The boost was delivered as a single fraction of 7 to 15 Gy using one to four isocenters. The local control rate was 98% for the 82 patients treated with this approach.53 Complications included three cases of radiation retinopathy and 10 cases of temporal lobe ­necrosis.53

complete response rates ranging from 14% to 27%.54-57 Response rates are better when the chemotherapy ­includes ­platinum-based compounds. The perceived chemosensitivity of NPC has led to numerous trials ­ attempting to improve outcome by combining chemotherapy with radiation in the neoadjuvant (preoperative) or adjuvant (postoperative) ­settings, given in sequence or concurrently with the radiation. ­Results from these trials are reviewed in the next few paragraphs.

Chemotherapy

Neoadjuvant Chemotherapy

Chemotherapy has been used for NPC with hematogenous spread since the 1970s and has yielded fairly high overall response rates—almost 80% in some series—and

Theoretically, neoadjuvant chemotherapy has two potential benefits: eradication of subclinical distant metastases and reduction of local tumor volume, which may ­ potentially

10  The Nasopharynx

improve the efficacy of radiation therapy. Because NPCs are associated with high recurrence rates, and because these tumors are chemosensitive, neoadjuvant ­therapy is an attractive option for treatment of this disease. This ­strategy has been investigated in numerous phase II trials, the results of which have been encouraging. A phase II study conducted at M. D. Anderson of NPC that had been staged by CT showed that neoadjuvant chemotherapy with cisplatin and fluorouracil followed by ­ radiation therapy resulted in a 5-year survival rate of 68%, compared with 56% achieved with radiation alone.58 This statistically significant difference in survival seemed to have resulted from improvements in local control and reductions in distant metastases, primarily in a subgroup of patients with undifferentiated NPC. These results were subsequently substantiated in a matched-cohort study.59 These positive results led to the subsequent performance of two other phase II trials; one evaluated taxane-based induction therapy followed by radiation,60 and the other evaluated cisplatin plus fluorouracil followed by radiation or concurrent chemoradiation therapy, with T status determining which patients would receive concurrent therapy.61 The results of these studies suggested that consideration of tumor histology and T and N status, rather than ­disease stage, may be useful for determining which patients were likely to benefit from concurrent chemoradiation. ­However, neither protocol was demonstrably superior in terms of controlling disease recurrence. Unfortunately, randomized phase III trials have not borne out the promise suggested by the previous phase II trials (Table 10-2).62-66 Five such trials have been ­ conducted; four investigated neoadjuvant chemotherapy followed by radiation, and the fifth assessed neoadjuvant and adjuvant chemotherapy. The first reported trial was a single-institution trial performed at Prince of Wales Hospital in Hong Kong.62 All patients received radiation therapy; the experimental treatment consisted of two cycles of fluorouracil and cisplatin

before the radiation therapy plus four cycles of systemic therapy with the same drugs after the completion of radiation therapy. No difference in 2-year survival rates was detected between the chemotherapy-plus-radiation and radiation-alone treatment groups (about 80% for both groups).62 An analysis of pooled data from two trials, one from the Asian-Oceanic Clinical Oncology Association trial (334 patients)64 and the other from a study done in Guangzhou, China (456 patients),65 showed an improvement in disease-free survival, but not in overall survival, for patients who received neoadjuvant chemotherapy.67 Findings from two other meta-analyses of randomized trials ­ studying the influence of chemotherapy on treatment ­ outcome in NPC68,69 indicated that chemotherapy can modestly improve overall survival rates but that this ­benefit was seen principally when the chemotherapy was given ­concurrently with radiation. These results are discussed further under “Concurrent Chemotherapy and Radiation Therapy.”

Adjuvant Chemotherapy Adjuvant chemotherapy has been included in trials of neoadjuvant chemotherapy for NPC and in trials of concurrent chemotherapy and radiation therapy for NPC, but adjuvant chemotherapy has rarely been tested by itself in the treatment of NPCs. Rossi and colleagues70 treated patients with radiation therapy alone or radiation therapy plus adjuvant vincristine, cyclophosphamide, and doxorubicin and found that the adjuvant treatment produced no benefit in survival. A subsequent phase III trial conducted in Taiwan in which 157 patients with stage IV NPC were ­randomly assigned to undergo radiation therapy with or without adjuvant ­ cisplatin, fluorouracil, and leucovorin showed no benefit from adjuvant treatment.71 Another randomized trial from Hong Kong evaluating the ­addition of six cycles of a five-drug adjuvant therapy regimen to radiation therapy found no improvement in survival with ­adjuvant chemotherapy.72 Two other randomized trials have tested a combination of concurrent ­ cisplatin-based

TABLE 10-2.   Results of Randomized Trials of Chemotherapy and Radiation Therapy for Nasopharyngeal Cancer Study and Reference

No. of Patients

217

Approach

Drugs

Survival Rate*

Relapse-Free ­Survival Rate*

Prince of Wales ­Hospital, 199562

82

Neoadjuvant (2 cycles) plus adjuvant (4 cycles)

Cisplatin and ­fluorouracil

80% vs. 81% (P = NS)

68% vs. 72% (P = NS)

International ­Nasopharynx Cancer Study Group, 199663

339

Neoadjuvant (3 cycles)

Bleomycin, ­epirubicin, and cisplatin

63% vs. 60% (P = NS)

54% vs. 40% (P = .01)

Asian-Oceanian Clinical Oncology ­Association, 199864

334

Neoadjuvant (2-3 cycles)

Epirubicin and cisplatin

78% vs. 71% (P = .57)

48% vs. 42% (P = .45)

Sun Yat-Sen University of Medical Sciences, Guangzhou, 200165

456

Neoadjuvant (2-3 cycles)

Bleomycin, cisplatin, and fluorouracil

63% vs. 56% (P = .11)

59% vs. 49% (P = .05)

Japanese Hokkaido ­Radiotherapy ­Oncology Group, 200266

80

Neoadjuvant (2 cycles)

Cisplatin and ­fluorouracil

60% vs. 48% (P = NS)

55% vs. 43% (P = NS)

*Rates for experimental versus control. Control groups were treated with radiation only. NS, not significant.

218

PART 3  Head and Neck

c­ hemoradiation therapy and ­adjuvant cisplatin plus fluorouracil ­chemotherapy—­Intergroup 009973 and NPC-9901.74 Both trials revealed a survival benefit from the chemotherapy, and both trials attributed the benefit to the concurrent rather than the adjuvant approach. Nevertheless, because the adjuvant therapy was part of the successful treatment strategy and because ­distant disease is a ­principal mode of disease recurrence, adjuvant cisplatin-based chemotherapy is considered by many clinicians to be part of the standard of care for NPC.

Concurrent Chemotherapy and Radiation Therapy Results from randomized phase III trials testing the role of concurrent chemotherapy and radiation in NPC are summarized in Table 10-3.73-78 In brief, the use of concurrent chemotherapy and radiation has become the standard of care for virtually all patients with NPC (except for those with stage I disease). The main body of evidence in support of this approach comes from the Intergroup trial 0099.73 In that study, patients were randomly assigned to concurrent chemotherapy and radiation therapy followed by adjuvant chemotherapy or to radiation therapy alone. The experimental treatment called for delivery of 100 mg/m2 of cisplatin on days 1, 21, and 43 of radiation therapy, followed by three courses of adjuvant cisplatin (80 mg/m2 on day 1) and fluorouracil (1000 mg/m2/day on days 1 through 4). In both treatment groups, radiation therapy consisted of 70 Gy delivered over a period of 7 weeks. The trial was stopped early because of a highly significant survival benefit in the ­experimentaltherapy group. The 3-year disease-free survival rate was 69% in the chemotherapy group, compared with 24% in the radiation-only group. However, several criticisms have been raised regarding this trial. For example, only 147 of the 193 entered patients were eligible for primary survival and toxicity analyses. (The investigators performed a separate analysis to address this problem and concluded that it did

not influence the results.) The second criticism regards the poor results seen in the control group. Many argue that the control rates seen were 20% to 25% lower than ­expected for radiation therapy alone. In an attempt to address these criticisms, the Hong Kong Nasopharyngeal Cancer Study Group conducted a second, confirmatory trial of concurrent chemotherapy and radiation therapy for NPC, the NPC-9901 trial.74 The trial design was similar to that of the Intergroup trial. In the NPC-9901 trial, 348 patients with stage T1-4 N2-3 M0 NPC were randomly assigned to undergo irradiation only or concurrent chemoradiation therapy and adjuvant chemotherapy. Chemotherapy seemed to improve local control rates (92% versus 82% for radiation only), but the overall survival rates were identical for the two treatment groups, and the late toxicity rates were higher in the chemotherapy group (28% versus 13% at 3 years).74 Preliminary findings from a parallel trial conducted by the same group to evaluate altered radiation fractionation for T3-4 N0-1 M0 ­disease indicated that only accelerated radiation plus chemotherapy was beneficial; patients treated with ­ conventionally fractionated radiation plus chemotherapy fared no better than those treated with conventionally fractionated ­radiation therapy alone.28 Results from another randomized trial from Singapore, where NPC is endemic, were more positive. In that phase III study, Wee and colleagues75 compared a concurrentplus-adjuvant chemotherapy approach with radiation only among 221 randomized patients and found that chemotherapy was associated with improvements in survival, local control, and distant metastases. The final word on the effectiveness of concurrent chemoradiation therapy for NPC comes from two meta-­analyses, one based on published findings from 2450 patients in 10 randomized studies69 and the other on data from 1753 individual patients in eight trials.68 The conclusions from both meta-analyses were that adding chemotherapy

TABLE 10-3.   Results of Randomized Trials of Concurrent Chemotherapy and Radiation Therapy for Nasopharyngeal Cancer Approach

Drugs

Survival Rate*

Relapse-Free ­Survival Rate*

147

Concurrent (weeks 1, 4, and 7) Adjuvant (3 cycles)

Cisplatin

78% vs. 47% (P = .05)

69% vs. 24% (P < .001)

Prince of Wales Hospital and Queen Elizabeth ­Hospital, Hong Kong, 200576

350

Concurrent (every week)

Cisplatin

70% vs. 59% (P = .065)

60% vs. 54% (P = .16)

Taichung Veterans ­General Hospital, Taiwan, 200377

284

Concurrent (weeks 1 and 5)

Cisplatin and fluorouracil

72% vs. 53% (P = .002)

72% vs. 54% (P = .001)

Sun Yat-Sen ­University of Medical Sciences, Guangzhou, 200578

115

Concurrent (every week)

Oxaliplatin

96% vs. 83% (P = .02)

100% vs. 77% (P = .02)

National Cancer Center, Singapore, 200575

221

Concurrent (weeks 1, 4, and 7) Adjuvant (3 cycles)

Cisplatin

80% vs. 65% (P = .006)

72% vs. 53% (P = .009)

Hong Kong ­Nasopharyngeal Cancer Study Group 99-0174

354

78% vs. 78% (P = .97)

70% vs. 61% (P = .1)

Study and Reference

No. of Patients

Intergroup 0099, 199873

*Rates

Concurrent (weeks 1, 4, and 7) Adjuvant (3 cycles)

Cisplatin and ­fluorouracil

Cisplatin and ­fluorouracil Cisplatin Cisplatin and ­fluorouracil

for experimental versus control. Control groups were treated with radiation only.

10  The Nasopharynx

to radiation therapy conferred a survival advantage and that this advantage was principally seen when the chemotherapy was delivered concurrently with radiation. Moreover, further analyses of the individualized data that ­excluded patients from the Intergroup 0099 trial ­(because of the controversy surrounding this trial) or patients with WHO 1 disease histology showed that exclusion of ­either of these two groups did not affect the validity of the results.68 Concurrent chemoradiation therapy is considered the standard of care in NPC.

Salvage Therapy for Local Recurrence Despite the use of aggressive radiation therapy, 10% to 30% of patients will experience local recurrence of NPC. For patients with extensive local relapse, particularly in the skull base, treatment options may be limited to palliative chemotherapy or radiation therapy. However, in a subset of patients with small, superficial recurrences, disease can potentially still be cured with aggressive local therapy, and aggressive approaches have been attempted occasion­ ally for patients with deeper and bulkier recurrences. Table 10-435,79-91 summarizes the outcomes of ­ various ­salvage therapies for recurrent NPC. Although surgery is rarely used in the initial treatment of NPC, surgery can yield good results in some patients with local recurrence. Tumors have been resected through transpalatal, transcervical, and transmaxillary approaches. In a 2000 report, King and colleagues79 from the Prince of Wales Hospital in Hong Kong reported a 5-year survival rate of 42% among 31 patients with recurrent T1 to T3 tumors treated with nasopharyngectomy. Postoperative ­radiation was used for some patients and seemed to be beneficial. The investigators reported few complications. Wei92 ­reported a similar outcome; local control was ­ obtained

219

in 19 (43%) of 45 patients after resection of recurrent NPC. Brachytherapy was used for 22% of the patients in this series who had inadequate margins. Despite this relatively high control rate, the mortality rate was also high at 6%. Fee and colleagues80 also advocated surgery for ­selected cases of recurrent NPC but described a high rate of skull base osteomyelitis resulting from this procedure. Danesi and colleagues93 used the type C infratemporal fossa ­ approach to resect recurrent or residual NPC after primary radiation therapy in 11 patients and reported no mortality and a disease-free survival rate of 72%. Reirradiation is often recommended for treatment of relatively small recurrent nasopharyngeal tumors. No conclusive data exist regarding the value of chemotherapy in this clinical setting. Reported locoregional control rates with reirradiation range from 15% to 40%. Tumor classification at recurrence is often a predictor of outcome; for example, Wang reported a control rate of only 15% for ­recurrent T3 or T4 tumors, compared with 38% for recurrent T1 or T2 tumors.35 The wide range in results reflects differences in patient selection and radiation techniques among studies. The use of EBRT alone to doses of 60 Gy or more for recurrent disease produced a control rate of 15% in a series reported from the Prince of Wales Hospital.81 Reirradiation to these doses also resulted in a more than 20% rate of temporal lobe encephalopathy. The use of conformal techniques for the treatment of recurrent NPC, like that of untreated NPC, has increased over the past decade, with several groups ­ using stereotactic radiosurgery or radiation therapy for this purpose. Cmelak and colleagues94 at Stanford University ­reported local control in seven of 12 patients treated with single-fraction (median, 18 Gy) stereotactic radiosurgery

TABLE 10-4.    Results of Retreatment of Recurrent Nasopharyngeal Cancer Study and Reference

Approach

No. of Patients

5-Year Actuarial ­ urvival Rate (%) S

5-Year Actuarial Local Control Rate (%)

Lee et al, 199383

EBRT ± IRT

706

14

32

200084

EBRT ± SRT

186

12

NS

Teo et al, 199881

EBRT ± IRT

103

 8

15

Leung et al, 200085

Multiple approaches

  91

30

38

Chang et al,

200782

SRT

  90

64 (crude rate)

85 (crude rate)

Hwang et al,

199886

XRT

  74

37

40 (LRPF)

Pryzant et al,

199287

Wu et al,

EBRT ± IRT

  53

21

35

Wang, 199735

EBRT ± IRT

  38

59 (DSS)

40

Syed et al, 200088

EBRT + IRT*

  34

30

93

McLean et al, 199889

EBRT + IRT

  17

65 (crude rate)

65

SRT

  13

31 (3 yr)

NS

Orecchia et al,

199990

200079

Surgery

  31

47

43

199791

Surgery

  24

NS

46 (crude rate)

Fee et al, 199180

Surgery

  15

80 (crude rate)

46 (crude rate)

King et al, Hsu et al,

*Chemotherapy

and/or hyperthermia. DSS, disease-specific survival; EBRT, external beam radiation therapy; IRT, interstitial or intracavitary brachytherapy; LRPF, ­ locoregional progression-free; NS, not stated; SRT, stereotactic irradiation; XRT, external beam radiation therapy, brachytherapy, heavy-charged-­ ­particle-beam therapy, and/or gamma knife therapy.

220

PART 3  Head and Neck

for skull base recurrences of NPC. Wu and colleagues82 at Sun Yat-Sen University used stereotactic radiation therapy for 90 patients with recurrent or persistent NPC. Patients with recurrent disease were treated to 48 Gy in 6 fractions, and those with persistent disease received 18 Gy in 3 fractions. At a median follow-up time of 20 months, the 3-year local control rates were 75% for patients with recurrent disease and 89% for patients with persistent disease. Seventeen patients developed late complications, including two grade 5 events. Chua and colleagues at the Queen Mary Hospital in Hong Kong also used stereotactic radiosurgery to treat 48 patients with locally recurrent NPC.95 In that study, delivering a median dose of 12.5 Gy to the target periphery resulted in a 5-year local control rate of 47%. In addition to stereotactic radiosurgery, the group at Queen Mary Hospital has used other EBRT techniques for treating recurrent disease. In one report,96 31 patients with recurrent NPC were treated with IMRT (median dose, 54 Gy) in combination with induction chemotherapy (68% of cases) or a single-dose stereotactic boost (32%). The 1-year local control rate for all patients in that study was 56%; however, for patients with tumors that were T1, T2, or T3 at recurrence (rT1-3), the local control rate was 100%. Six patients (19%) experienced grade 3 late toxicities, mostly ototoxicity or cranial neuropathy, and the actuarial rate of grade 3 late toxicities at 1 year was 25%.96 The group at Queen Mary Hospital also reported their experience with gold grain implantation for selected ­patients with recurrent NPC.97 Local control rates in that study were equivalent for patients treated with stereotactic irradiation and for those treated with gold grain implants. The types of complications differed between groups, with the stereotactic treatment group experiencing more neuroendocrine problems and the gold implant group more headaches and fistulas.97

Nasopharyngeal Carcinomas in Children Epidermoid carcinoma, particularly of the head and neck, is rare in children and adolescents. The annual incidence in North America is estimated to be approximately one case per 100,000 people per year. NPC accounts for most of these head and neck neoplasms. Carcinomas account for 85% to 90% of NPCs in adults but only 20% to 50% of NPCs in children. Most childhood nasopharyngeal tumors are sarcomas and lymphomas. NPC in the young is closely associated with Epstein-Barr virus infection. Most patients present with advanced undifferentiated carcinoma (WHO type 3), with most ­series reporting nodal involvement in more than 90%. In a combined report from M. D. Anderson and Stanford University, only 4 of 57 patients with NPC between the ages of 4 to 21 years did not have clinically evident nodal involvement.98 In children, as in adults, radiation therapy is the primary treatment modality for NPC. However, because the disease is much less common in children, the appropriate radiation dose for the treatment of childhood NPC is unclear. ­Several investigators reported good control rates with doses in the range of 45 to 50 Gy, with or without chemotherapy.99 At M. D. Anderson, it is recommended that children older than 10 years be treated with doses similar to those used in

adults, with 66 Gy delivered for stage T3 to T4 undifferentiated carcinoma. A 10% dose reduction is recommended for children younger than 10 years. Locoregional control rates are high for children with NPC. In the M. D. Anderson and Stanford University series, only six patients (14%) had locoregional recurrences during a median follow-up period of 93 months (range, 6 to 178 months). In another series, Ayan and Altun100 reported a locoregional control rate of 79% among 50 patients with NPC, ages 5 to 16 years. Many of the recurrences in this series occurred during an early era of treatment, and the study authors suggested that these recurrences were in part the result of poor tumor definition. At the Princess ­Margaret Hospital, only 1 of 32 patients treated between 1975 and 1990 had an isolated locoregional recurrence.101 The primary failure pattern for children with NPC is distant metastasis. Five-year survival rates are similar to those for adults, ranging from 30% to 60%. Interest has been expressed in combining chemotherapy with radiation therapy in the treatment of NPC in children, but extrapolation from adult series is difficult, primarily because of the different patterns of recurrence. Concurrent ­chemotherapy and radiation is more difficult to rationalize in children than adults, but the question of whether a concurrent approach can lower the effective radiation dose needs to be addressed. Pao and colleagues99 achieved high local control rates in children with NPC with the use of 50 Gy plus ­adjuvant chemotherapy. Results of trials testing induction and adjuvant chemotherapy are mixed, and unfortunately, no ­ randomized trials and few prospective trials have been reported. Most investigators advocate chemotherapy because many ­ trials have suggested improvement in disease control with ­chemotherapy and because NPC in children is primarily systemic. Chronic side effects from radiation therapy are a ­problem for children with NPC. Late sequelae include hypopituitarism, hypothyroidism, trismus, advanced dental decay, arrested development (particularly of clavicular growth), skin and soft tissue fibrosis, and second malignancies.

REFERENCES   1. Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 17-88.   2. Teo P, Shiu W, Leung S, et al. Prognostic factors in nasopharyngeal carcinoma investigated by computer tomography—an analysis of 659 patients. Radiother Oncol 1992;23:79-93.   3. Lee AW, Au JS, Teo PM, et al. Staging of nasopharyngeal ­carcinoma: suggestions for improving the current UICC/ AJCC Staging System. Clin Oncol (R Coll Radiol) 2004;16: 269-276.   4. Liu MZ, Tang LL, Zong JF, et al. Evaluation of sixth ­ edition of AJCC staging system for nasopharyngeal carcinoma and ­proposed improvement. Int J Radiat Oncol Biol Phys 2008;70: 1115-1123.   5. Shanmugaratnam KS, Sobin LH. Histological Typing of Upper Respiratory Tract Tumors. Geneva: World Health Organization, 1978.   6. Yu MC. Nasopharyngeal carcinoma: epidemiology and dietary factors. IARC Sci Publ 1991;105:39-47.   7. Huang DP, Lo KW. Aetiological factors and ­ pathogenesis. In Van Hasselt CA, Gibb AG (eds): Nasopharyngeal ­ Carcinoma. 2nd ed. Hong Kong: Chinese University Press, 1999 pp 31–60.   8. Young LS, Rickinson AB. Epstein-Barr virus: 40 years on. Nat Rev Cancer 2004;4:757-768.

10  The Nasopharynx   9. Pathamathan R, Prasad U, Sadler R, et al. Clonal proliferations of cells infected with Epstein-Barr virus in preinvasive lesions related to NPC. N Engl J Med 1995;333:693-698. 10. Zeng Y. Seroepidemiological studies on nasopharyngeal carcinoma in China. Adv Cancer Res 1985;44:121-138. 11. Chan AT, Lo YM, Zee B, et al. Plasma Epstein-Barr virus DNA and residual disease after radiotherapy for undifferentiated nasopharyngeal carcinoma. J Natl Cancer Inst 2002;94: 1614-1619. 12. Burt R, Vaughan T, Nisperos B, et al. A prospective association between the HLA-A2 antigen and nasopharyngeal carcinoma in US Caucasians. Int J Cancer 1994;56:465-467. 13. Lo K, Huang D, Lau K. P16 gene alterations in nasopharyngeal carcinoma. Cancer Res 1995;15:2039-2043. 14. Effert P, McCoy R, Abdelmalid M. Alteration of the p53 gene in NPC. J Virol 1992;37:63-75. 15. Lee H, Gourley L, Duffy S, et al. Preserved foods and nasopharyngeal carcinoma: a case-control study among Singapore ­Chinese. Int J Cancer 1994;59:585-590. 16. Chow W, Mclaughlin J, Hrubec Z, et al. Tobacco use and nasopharyngeal carcinoma in a cohort of US veterans. Int J Cancer 1993;55:538-540. 17. Fletcher GH, Million RR. Nasopharynx. In Fletcher GH (ed): Textbook of Radiotherapy, 3rd ed. Philadelphia: Lea & Febiger, 1980, pp 364-383. 18. Hunt MA, Zelefsky MJ, Wolden S, et al. Treatment planning and delivery of intensity-modulated radiation therapy for ­ primary nasopharynx cancer. Int J Radiat Oncol Biol Phys 2001;49: 623-632. 19. Xia P, Fu KK, Wong GW, et al. Comparison of treatment plans involving intensity-modulated radiotherapy for nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2000;48:329-337. 20. Lee N, Xia P, Quivey JM, 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. 21. Wolden SL, Chen WC, Pfister DG, 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. 22. Kwong D, Sham J, Chua D, et al. The effect of interruptions and prolonged treatment time in radiotherapy for nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 1997;39:703-710. 23. Fu K, Pajak T, Trotti A, et al. A Radiation Therapy Oncology Group (RTOG) phase III randomized study to compare hyperfractionation and two variants of accelerated fractionation to standard fractionation radiotherapy for head and neck squamous cell carcinomas: first report of RTOG 9003. Int J Radiat Oncol Biol Phys 2000;48:7-16. 24. Horiot J, Le Fur R, N’Guyen T, et al. Hyperfractionation ­versus conventional fractionation in oropharyngeal carcinoma: final analysis of a randomized trial of the EORTC cooperative group of radiotherapy. Radiother Oncol 1992;25:231-241. 25. Teo PM, Leung SF, Chan AT, et al. Final report of a randomized trial on altered-fractionated radiotherapy in nasopharyngeal carcinoma prematurely terminated by significant increase in ­neurologic complications. Int J Radiat Oncol Biol Phys 2000;48: 1311-1322. 26. Lee AW, Kwong DL, Leung SF, et al. Factors affecting risk of symptomatic temporal lobe necrosis: significance of ­fractional dose and treatment time. Int J Radiat Oncol Biol Phys 2002;53:75-85. 27. Lee AW, Sze WM, Yau TK, et al. Retrospective analysis on treating nasopharyngeal carcinoma with accelerated fractionation (6 fractions per week) in comparison with conventional fractionation (5 fractions per week): report on 3-year tumor control and normal tissue toxicity. Radiother Oncol 2001;58:121-130. 28. Lee AW, Tung SY, Chan AT, et al. Preliminary results of a randomized study (NPC-9902 Trial) on therapeutic gain by concurrent chemotherapy and/or accelerated fractionation for locally advanced nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2006;66:142-151. 29. Kaasa S, Kragh-Jensen E, Bjordal K, et al. Prognostic factors in patients with nasopharyngeal carcinoma. Acta Oncologica 1993;32:531-536.

221

30. Perez C, Devineni V, Marcial-Vega V, et al. Carcinoma of the nasopharynx: factors affecting prognosis. Int J Radiat Oncol Biol Phys 1992;23:271-280. 31. Willner J, Baier K, Pfreundner L, et al. Tumor volume and local control in primary radiotherapy of nasopharyngeal carcinoma. Acta Oncol 1999;38:1031-1035. 32. Kwong DL, Sham JS, Leung LH, et al. Preliminary results of radiation dose escalation for locally advanced nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2006;64:374-381. 33. Lee A, Poon Y, Foo W, et al. Retrospective analysis of 5037 patients with nasopharyngeal carcinoma treated during 19761985: overall survival and patterns of failure. Int J Radiat Oncol Biol Phys 1992;23:261-270. 34. Sanguineti G, Geara FB, Garden AS, et al. Carcinoma of the ­nasopharynx treated by radiotherapy alone: determinants of ­local and regional control. Int J Radiat Oncol Biol Phys 1997;37: 985-996. 35. Wang CC Carcinoma of the nasopharynx. In Thawley SE, Panje WR, Batsakis JG, et al. (eds): Radiation Therapy for Head and Neck Neoplasms, 3rd ed. New York: Wiley-Liss, 1997, pp 257280. 36. Huang S. Nasopharyngeal cancer: a review of 1605 patients treated radically with cobalt 60. Int J Radiat Oncol Biol Phys 1980;6:401-407. 37. Qin D, Hu Y, Yan J, et al. Analysis of 1379 patients with nasopharyngeal carcinoma treated by radiation. Cancer 1988;61:11171124. 38. Zhang EP, Lian PG, Cai KL, et al. Radiation therapy of nasopharyngeal carcinoma: prognostic factors based on a 10-year ­follow-up of 1302 patients. Int J Radiat Oncol Biol Phys 1989;16:301-305. 39. Johansen L, Mesre M, Overgaard J. Carcinoma of the nasopharynx: analysis of treatment results in 167 consecutively admitted patients. Head Neck 1992:200-207. 40. Laramore G, Clubb B, Quick C, et al. Nasopharyngeal carcinoma in Saudi Arabia: a retrospective study of 166 cases treated with curative intent. Int J Radiat Oncol Biol Phys 1988;15: 1119-1127. 41. Olmi P, Cellai E, Fallai C. Nasopharyngeal carcinoma: an analysis of 308 patients treated with curative radiotherapy [abstract]. TV International IST Symposium. Interaction of Chemotherapy and Radiotherapy in Solid Tumors 1992. Geneva. 42. Dickson R, Flores A. Nasopharyngeal carcinoma: an evaluation of 134 patients treated between 1971-1980. Laryngoscope 1985;95:276-283. 43. Vikram B, Mishra UB, Strong E, et al. Patterns of failure in carcinoma of the nasopharynx. I. Failure at the primary site. Int J Radiat Oncol Biol Phys 1985;11:1455-1459. 44. Bailet J, Mark R, Abemayor E, et al. Nasopharyngeal carcinoma: treatment results with primary radiation therapy. Laryngoscope 1992;102:965-972. 45. Lee AW, Sze WM, Au JS, et al. Treatment results for nasopharyngeal carcinoma in the modern era: the Hong Kong ­experience. Int J Radiat Oncol Biol Phys 2005;61:1107-1116. 46. Chow E, Payne D, Keane T, et al. Enhanced control by radiotherapy of cervical lymph node metastases arising from nasopharyngeal carcinoma compared with nodal metastases from other head and neck squamous cell carcinomas. Int J Radiat Oncol Biol Phys 1997;39:149-154. 47. Geara F, Sanguineti G, Tucker S, et al. Carcinoma of the ­nasopharynx treated by radiotherapy alone: determinants of distant metastasis and survival. Radiother Oncol 1997;43:53-61. 48. Kwong D, Wei W, Sham J, et al. Sensorineural hearing loss in ­patients treated for nasopharyngeal carcinoma: a prospective study of the effect of radiation and cisplatin treatment. Int J Radiat Oncol Biol Phys 1996;36:281-289. 49. Kam MK, Leung SF, Zee B, et al. Prospective randomized study of intensity-modulated radiotherapy on salivary gland function in early-stage nasopharyngeal carcinoma patients. J Clin Oncol 2007;25:4873-4879. 50. Kwong DL, Pow EH, Sham JS, et al. Intensity-modulated ­radiotherapy for early-stage nasopharyngeal carcinoma: a prospective study on disease control and preservation of salivary function. Cancer 2004;101:1584-1593.

222

PART 3  Head and Neck

51. Levendag PC, Lagerwaard FJ, Noever I, et al. Role of endocavitary brachytherapy with or without chemotherapy in cancer of the nasopharynx. Int J Radiat Oncol Biol Phys 2002;52:755-768. 52. Chang J, See LC, Tang S, et al. The role of brachytherapy in early stage nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 1996;36:1019-1024. 53. Hara W, Loo BW Jr, Goffinet DR, et al. Excellent local control with stereotactic radiotherapy boost after external beam radiotherapy in patients with nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2008;71:393-400. 54. Cvitkovic E. Neoadjuvant chemotherapy (NACT) with epirubicin (EPI), cisplatin (CDDP), bleomycin (BLEO) (BEC) in undifferentiated nasopharyngeal cancer (UCNT): Preliminary results of an international (Int) phase (PH) III trial [abstract]. Proc Am Soc Clin Oncol 1994;13:283. 55. Decker D, Drelichman A, Al-Sarraf M, et al. Chemotherapy for nasopharyngeal carcinoma. A ten-year experience. Cancer 1983;52:602-605. 56. Choo R, Tannock I. Chemotherapy for recurrent or metastatic carcinoma of the nasopharynx. A review of the Princess Margaret Hospital experience. Cancer 1991;68:2120-2124. 57. Boussen H, Cvitkovic E, Wendling J, et al. Chemotherapy of metastatic and/or recurrent undifferentiated nasopharyngeal carcinoma with cisplatin, bleomycin, and fluorouracil. J Clin Oncol 1991;9:1675-1681. 58. Garden A, Lippman S, Morrison W, et al. Does induction chemotherapy have a role in the management of nasopharyngeal carcinoma? Results of treatment in the era of computerized ­ tomography. Int J Radiat Oncol Biol Phys 1996;36: 1005-1112. 59. Geara F, Glisson B, Sanguineti G, et al. Induction chemotherapy followed by radiotherapy versus radiotherapy alone in patients with advanced nasopharyngeal carcinoma. Cancer 1997;79:1279-1286. 60. Johnson FM, Garden A, Palmer JL, 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;100:991-998. 61. Johnson FM, Garden AS, Palmer JL, 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;63:717-724. 62. Chan A, Teo P, Leung T, et al. A prospective randomized study of chemotherapy adjunctive to definitive radiotherapy in advanced nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 1995;33:569-577. 63. International Nasopharynx Cancer Study Group. Preliminary results of a randomized trial comparing neoadjuvant chemotherapy (cisplatin, epirubicin, bleomycin) plus radiotherapy vs. radiotherapy alone in stage IV (≥N2, M0) undifferentiated nasopharyngeal carcinoma: a positive effect on progression-free survival. Int J Radiat Oncol Biol Phys 1996;35:463-469. 64. Chua D, Sham J, Choy D, et al. Preliminary report of the AsianOceanian Clinical Oncology Association randomized trial comparing cisplatin and epirubicin followed by radiotherapy versus radiotherapy alone in the treatment of patients with locoregionally advanced nasopharyngeal carcinoma. Cancer 1998;83:2270-2283. 65. Ma J, Mai HQ, Hong MH, et al. Results of a prospective randomized trial comparing neoadjuvant chemotherapy plus radiotherapy with radiotherapy alone in patients with locoregionally advanced nasopharyngeal carcinoma. J Clin Oncol 2001;19:1350-1357. 66. Hareyama M, Sakata K, Shirato H, et al. A prospective, randomized trial comparing neoadjuvant chemotherapy with radiotherapy alone in patients with advanced nasopharyngeal carcinoma. Cancer 2002;94:2217-2223. 67. Chua DT, Ma J, Sham JS, et al. Long-term survival after cisplatin-based induction chemotherapy and radiotherapy for nasopharyngeal carcinoma: a pooled data analysis of two phase III trials. J Clin Oncol 2005;23:1118-1124.

68. Baujat B, Audry H, Bourhis J, et al. Chemotherapy in locally advanced nasopharyngeal carcinoma: an individual patient data meta-analysis of eight randomized trials and 1753 patients. Int J Radiat Oncol Biol Phys 2006;64:47-56. 69. Langendijk J, Leemans C, Buter J, et al. The additional value of chemotherapy to radiotherapy in locally advanced nasopharyngeal carcinoma: a meta-analysis of the published literature. J Clin Oncol 2004;22:4604-4612. 70. Rossi A, Molinari R, Boracchi P, et al. Adjuvant ­chemotherapy with vincristine, cyclophosphamide, and doxorubicin after ­radiotherapy in local-regional nasopharyngeal cancer: results of a 4-year multicenter randomized trial. J Clin Oncol 1988;6: 1401-1410. 71. Chi KH, Chang YC, Guo WY, et al. A phase III study of adjuvant chemotherapy in advanced nasopharyngeal carcinoma patients. Int J Radiat Oncol Biol Phys 2002;52:1238-1244. 72. Kwong DL, Sham JS, Au GK, et al. Concurrent and adjuvant chemotherapy for nasopharyngeal carcinoma: a factorial study. J Clin Oncol 2004;22:2643-2653. 73. Al-Sarraf M, LeBlanc M, Giri PG, et al. Chemoradiotherapy versus radiotherapy in patients with advanced nasopharyngeal ­cancer: phase III randomized Intergroup study 0099. J Clin ­Oncol 1998;16:1310-1317. 74. Lee AW, Lau WH, Tung SY, et al. Preliminary results of a randomized study on therapeutic gain by concurrent chemotherapy for regionally advanced nasopharyngeal carcinoma: NPC-9901 Trial by the Hong Kong Nasopharyngeal Cancer Study Group. J Clin Oncol 2005;23:6966-6975. 75. Wee J, Tan EH, Tai BC, 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;23:6730-6738. 76. Chan AT, Leung SF, Ngan RK, et al. Overall survival after concurrent cisplatin-radiotherapy compared with radiotherapy alone in locoregionally advanced nasopharyngeal carcinoma. J Natl Cancer Inst 2005;97:536-539. 77. Lin J, Jan J, Hsu C, 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;21:631-637. 78. Zhang L, Zhao C, Peng P, et al. Phase III study comparing standard radiotherapy with or without weekly oxaliplatin in ­treatment of locoregionally advanced nasopharyngeal carcinoma: preliminary results. J Clin Oncol 2005;23:8461-8468. 79. King W, Ku P, Mok C, et al. Nasopharyngectomy in the ­treatment of recurrent nasopharyngeal carcinoma: a twelve-year experience. Head Neck 2000;22:215-222. 80. Fee W Jr, Roberson J, Goffinet D. Long-term survival ­ after surgical resection for recurrent nasopharyngeal cancer ­ after ­radiotherapy failure. Arch Otolaryngol Head Neck Surg 1991; 117:1233-1236. 81. Teo P, Kwan W, Chan A, et al. How successful is high-dose (≥60 Gy) reirradiation using mainly external beams in salvaging local failures of nasopharyngeal carcinoma? Int J Radiat Oncol Biol Phys 1998;40:897-913. 82. Wu SX, Chua DT, Deng ML, et al. Outcome of fractionated stereotactic radiotherapy for 90 patients with locally persistent and recurrent nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2007;69:761-769. 83. Lee A, Law S, Foo W, et al. Retrospective analysis of patients with nasopharyngeal carcinoma treated during 1976-1985: survival after local recurrence. Int J Radiat Oncol Biol Phys 1993;26:773-782. 84. Chang J, See L, Liao C, et al. Locally recurrent nasopharyngeal carcinoma. Radiother Oncol 2000;54:135-142. 85. Leung T, Tung S, Sze W, et al. Salvage radiation therapy for locally recurrent nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2000;48:1331-1338.

10  The Nasopharynx 86. Hwang J-M., Fu K, Phillips T. Results and prognostic factors in the retreatment of locally recurrent nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 1998;41:1099-1111. 87. Pryzant R, Wendt C, Delclos L, et al. Re-treatment of nasopharyngeal carcinoma in 53 patients. Int J Radiat Oncol Biol Phys 1992;22:941-947. 88. Syed A, Puthawala A, Damore S, et al. Brachytherapy for ­primary and recurrent nasopharyngeal carcinoma: 20 years’ ­experience at Long Beach Memorial. Int J Radiat Oncol Biol Phys 2000;47:1311-1321. 89. McLean M, Chow E, O’Sullivan B, et al. Re-irradiation for ­locally recurrent nasopharyngeal carcinoma. Radiother Oncol 1998;48:209-211. 90. Orecchia R, Redda M, Ragona R, et al. Results of hypofractionated stereotactic re-irradiation on 13 locally recurrent nasopharyngeal carcinomas. Radiother Oncol 1999;53:23-28. 91. Hsu MM, Ko JY, Sheen TS, et al. Salvage surgery for recurrent nasopharyngeal carcinoma. Arch Otolaryngol Head Neck Surg 1997;123:305-309. 92. Wei W. Salvage surgery for recurrent primary nasopharyngeal carcinoma. Crit Rev Oncol Hematol 2000; 33:91-98. 93. Danesi G, Zanoletti E, Mazzoni A. Salvage surgery for recurrent nasopharyngeal carcinoma. Skull Base 2007;17:173-180.

223

94. Cmelak A, Cox R, Adler J, et al. Radiosurgery for skull base malignancies and nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 1997;37:997-1003. 95. Chua DT, Sham JS, Hung KN, et al. Predictive factors of ­tumor control and survival after radiosurgery for local failures of nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2006;66:1415-1421. 96. Chua DT, Sham JS, Leung LH, et al. Re-irradiation of nasopharyngeal carcinoma with intensity-modulated radiotherapy. ­Radiother Oncol 2005;77:290-294. 97. Chua DT, Wei WI, Sham JS, et al. Stereotactic radiosurgery ­versus gold grain implantation in salvaging local failures of nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2007;69:469-474. 98. Ingersoll L, Woo S, Donaldson S, et al. Nasopharyngeal ­carcinoma in the young: a combined M. D. Anderson and Stanford experience. Int J Radiat Oncol Biol Phys 1990;19:881-887. 99. Pao W, Hustu O, Douglass E, et al. Pediatric nasopharyngeal ­carcinoma: long term follow-up of 29 patients. Int J Radiat ­Oncol Biol Phys 1989;17:299-305. 100. Ayan I, Altun M. Nasopharyngeal carcinoma in children: ­retrospective review of 50 patients. Int J Radiat Oncol Biol Phys 1996;35:485-492. 101. Ahern V, Jenkin D, Banerjee D, et al. Nasopharyngeal carcinoma in the young. Clin Oncol (Royal Coll Radiol) 1994;6:24-30.

11 The Oropharynx WILLIAM H. MORRISON, ADAM S. GARDEN, AND K. KIAN ANG PATHOLOGY ANATOMY AND PATHWAYS OF SPREAD Tongue Base Tonsillar Region Soft Palate Oropharyngeal Walls ETIOLOGY AND EPIDEMIOLOGY PREVENTION AND EARLY DETECTION MOLECULAR CHARACTERISTICS CLINICAL MANIFESTATIONS, EVALUATION, AND STAGING Patient Evaluation Staging PRIMARY THERAPY General Principles of Treatment Treatment Policies According to Disease Stage Treatment Policies According to Disease Subsite

The oropharynx consists of the tongue base, tonsillar ­region, soft palate, and lateral and posterior oropharyngeal walls (Fig. 11-1). These structures have essential roles in swallowing and speech. Neoplasms arising in the oropharynx can impede these functions by infiltrating the musculature, by obstructing the cavity, or by causing pain. Treatments administered to cure oropharyngeal tumors also can cause further structural deformity and functional impairment. Radiation therapy, in some cases enhanced with chemotherapy, is generally preferred over surgery for the ­ management of early-stage to moderately ­ advanced ­oropharyngeal tumors because tumoricidal doses can be given with a relatively high likelihood of preserving ­swallowing and speech. Squamous cell carcinoma of the oropharynx has served as a good model for investigating the radiobiology of ­ human tumors. Because oropharyngeal ­ carcinomas are relatively sensitive to radiation and are readily accessible to examination, disease staging and response to therapy can be evaluated accurately. A large body of time-dose fractionation data has been obtained from studying the response of oropharyngeal tumors to radiation. ­ Moreover, approximately 40% of patients in phase III trials comparing radiation plus chemotherapy with radiation alone had oropharyngeal tumors. Findings from several trials of various therapies are summarized in Chapter 7. The ­ oropharynx has been an important ­testing site for parotid-sparing radiation techniques such as ­intensity-­modulated radiation therapy (IMRT). Progress has also been made in understanding the role of

224

TREATMENT RESULTS Oropharyngeal Carcinoma Oropharyngeal Subsites Control of Nodal Disease Distant Metastasis Reirradiation for Recurrent or New Tumors after Radiation Therapy RADIATION THERAPY TECHNIQUES Primary Radiation Therapy Postoperative Radiation Therapy ONGOING STUDIES

human ­ papillomavirus (HPV) in the carcinogenesis of ­oropharyngeal carcinomas.

Pathology Approximately 95% of oropharyngeal tumors are squamous cell carcinomas. Prolonged, repeated exposure to carcinogenic insults such as tobacco and alcohol can cause widespread damage to the mucosa of the aerodigestive tract and increases the risk of developing multiple, independent premalignant and malignant foci. Premalignant foci can be clinically evident as leukoplakia or ­erythroplakia (Fig. 11-2). Eventually, field defects such as these can develop into additional carcinomas, which usually have a squamous histology. The annual rate of development of second squamous carcinomas in the head and neck is estimated at 4% to 7%.1 Malignant transformation is much more likely to develop from erythroplakic mucosal changes than from leukoplakia. Basaloid squamous carcinoma,2 a variant of squamous carcinoma with cytomorphologic resemblance to the solid variant of adenoid cystic carcinoma, can occur in the oropharynx. In most reported series, the clinical course of ­basaloid squamous carcinoma seems to be more aggressive that that of the more typical squamous carcinoma.3 Less common epithelial neoplasms originating in the oropharynx are minor salivary gland tumors4 and ­squamous carcinoma with lymphoid stroma (formerly known as ­lymphoepithelioma).5 Malignant lymphomas occasionally occur in Waldeyer’s ring of the oropharynx.

11  The Oropharynx

Anterior tonsillar pillar

Posterior wall of oropharynx

Soft palate Uvula

Uvula

Posterior tonsillar pillar

Soft palate

Palatine tonsil

Sulcus Posterior terminalis tonsillar pillar Anterior tonsillar pillar Palatine tonsil (in tonsillar fossa)

Glossopharyngeal sulcus

Retromolar trigone Circumvallate papilla

A Postcricoid space True vocal cord False vocal cord Pyriform sinus

225

Epiglottis Vallecula Lateral oropharyngeal wall

Base of tongue Foramen cecum

Median glossoepiglottic fold

Corniculate tubercle Cuneiform tubercle Aryepiglottic fold Pyriform sinus

Lateral pharyngeal wall Pharyngoepiglottic fold

Epiglottis Vallecula

Median glossoepiglottic fold

Base of tongue

B FIGURE 11-1.  Normal anatomy of the oropharynx is shown. A, Anterior view and lateral view of the oropharynx with the left lateral structures removed. B, Oropharynx as viewed through a fiberoptic endoscope.

Anatomy and Pathways of Spread The route of spread of oropharyngeal cancer varies somewhat according to the subsite of origin. Knowledge of the anatomy of the oropharyngeal subsites facilitates insight into the pathways of spread.

Tongue Base

FIGURE 11-2.  Superficial leukoplakia can be seen on the right soft palate of a patient presenting with a squamous cell carcinoma of the epiglottis. A biopsy of the leukoplakia showed severe squamous dysplasia but no invasive carcinoma. (From Gunderson LL, Tepper JE [eds]. Clinical Radiation Oncology, 2nd ed. Philadelphia: Elsevier Churchill Livingstone, 2007, p 675.)

The tongue base is the posterior, vertical component of the tongue and includes the valleculae. It starts behind the terminal sulcus of the tongue, borders the glossopharyngeal sulcus laterally, and ends at the junction between the ­valleculae and the base of the lingual epiglottis inferiorly. The tongue base contains numerous submucosal lymphoid follicles, which give its surface an irregular appearance. This lymphoid tissue can enhance on contrast-enhanced computed tomography (CT), producing a radiologic ­appearance that can be confused with tumor. The tongue base can also enhance on 18F-fluorodeoxyglucose positron

226

PART 3  Head and Neck

A

B

FIGURE 11-3.  A, CT scan shows an ulcerative carcinoma of the base of tongue. The ulcer is surrounded by a rim of enhancing tumor. B, CT scan shows an infiltrative tongue base carcinoma invading the extrinsic tongue muscles and fixing the tongue. (From ­Gunderson LL, Tepper JE [eds]. Clinical Radiation Oncology, 2nd ed. Philadelphia: Elsevier Churchill Livingstone, 2007, p 676.)

emission tomography (FDG-PET), particularly in young patients. The embryologic origin of the tongue base is different from that of the oral tongue. The main sensory innervation of the tongue base is from the lingual branch of the glossopharyngeal nerve. The hypoglossal nerve provides the motor innervation of the entire tongue, including the base. The main lymphatic drainage pathway of the tongue base is to the jugular chain of lymph nodes. Early-stage tongue base tumors manifest locally as an exophytic tumor growing out from the tongue or as a ­superficial mucosal tumor. More advanced tumors can have a prominent submucosal infiltrative or ulcerative component (Fig. 11-3A) that invades the intrinsic tongue muscles. At later stages, tongue base tumors can extend anteroinferiorly through the intrinsic tongue muscles to invade the extrinsic muscles, causing tongue deviation or impaired tongue ­ mobility (see Fig. 11-3B); they can spread laterally through the glossopharyngeal (glossopalatine) sulcus to the inferior tonsillar fossa and pharyngeal wall, or they can grow inferiorly into the larynx, hypopharynx, and pre-epiglottic space. Very advanced tongue base tumors can extend into the soft ­tissues of the neck or, rarely, through the mandibular cortex. The incidence of nodal involvement at diagnosis in patients with tongue base tumors referred to The University of Texas M. D. Anderson Cancer Center between 1948 and 1965 was 78%.6 Bilateral nodal disease was present in 29% of patients, as would be expected for a tumor arising in a midline structure with rich lymphatic networks. In a ­series of patients with tongue base tumors and no palpable lymphadenopathy who underwent neck dissection at the Mayo Clinic, the incidence of subclinical nodal involvement was 44%.7

Tonsillar Region The tonsillar region contains the palatine tonsil, which is located between the anterior and posterior tonsillar (faucial) pillars. The palatine tonsil has a surface mucosa of stratified squamous epithelium overlying lymphatic ­ nodules. A fibrous capsule forms the lateral surface of the tonsil, and underneath this capsule are the superior pharyngeal constrictor muscle, the parapharyngeal space, the pterygoid muscles, and the mandible. The internal carotid artery is ­located approximately 2 cm lateral and posterior to the tonsil in the parapharyngeal space. The sensory innervation of the tonsil is from the middle and posterior palatine branches of the maxillary nerve. Tumors originating in the tonsillar fossa can grow as an exophytic mass into the pharyngeal air space, particularly in early-stage presentations. More often, these tumors spread medially into the soft palate (Fig. 11-4), anteriorly into the anterior tonsillar pillar and retromolar trigone, across the glossopalatine sulcus into the tongue base, posteriorly to the posterior tonsillar pillar and lateral pharyngeal wall, laterally into the parapharyngeal space and pterygoid muscles, superiorly into the nasopharynx, and inferiorly into the pyriform sinus. Invasion of the tongue base can impair tongue mobility, and infiltration of the pterygoid muscles can cause trismus. Very advanced tonsillar tumors can extend directly into the neck, encase the carotid artery, or erode the mandibular cortex. The pattern of local disease extension among 160 patients with tonsillar carcinoma referred to M. D. Anderson Cancer Center between 1968 and 1979 is shown in Table 11-1.8 In that series, nodal disease was present at diagnosis in 69% of patients. In another series of patients with tonsillar tumors and no palpable lymphadenopathy who underwent neck dissection at the Mayo Clinic between 1970 and 1988, the incidence of subclinical nodal involvement was 32%.9

11  The Oropharynx

FIGURE 11-4.  An exophytic squamous cell carcinoma of the tonsil grows submucosally along the soft palate and into the pharyngeal airspace; no infiltrative or ulcerative features  are evident. (From Gunderson LL, Tepper JE [eds]. Clinical ­Radiation Oncology, 2nd ed. Philadelphia: Elsevier Churchill ­Livingstone, 2007, p 677.)

TABLE 11-1.   Sites of Local Extension of Carcinoma of the Tonsil in 160 Patients Seen at M. D. Anderson Cancer Center between 1968 and 1979 Site

Percent of Patients

Tonsil and pillars only

23

Tongue base

38

Lateral pharyngeal wall

31

Soft palate

23

Retromolar triangle

10

227

FIGURE 11-5.  A plaquelike squamous carcinoma of the left soft palate involves the entire uvula, with spread into the right soft palate and extension into the tonsillar fossa and tonsillar pillar. The tumor was not thick, was mobile on examination, and did not have highly infiltrative features. (From Gunderson LL, ­Tepper JE [eds]. Clinical Radiation Oncology, 2nd ed. ­ Philadelphia: Elsevier Churchill Livingstone, 2007, p 678.)

innervated by the contribution of the vagus nerve to the pharyngeal plexus. Almost all soft palate tumors arise on the oral surface. Early tumors tend to grow superficially along mucosal surfaces and often appear as regions of erythroplasia with indistinct borders. The erythroplasia may be multifocal. Common routes of spread for more advanced tumors are down the anterior tonsillar pillars (Fig. 11-5), into the tonsillar fossa and pharyngeal wall, and anteriorly onto the hard palate. Locally advanced soft palate tumors are not as common as are tumors in other oropharyngeal sites. Nodal involvement at diagnosis is present in about 40% of patients, a rate lower than that for tongue base and tonsillar region tumors.

Oral tongue

4

Oropharyngeal Walls

Posterior pharyngeal wall

3

Buccal mucosa

3

Lateral floor of mouth

2

Pterygoid muscles

2

Mandibular or maxillary bone

1

The lateral and posterior oropharyngeal walls consist of squamous epithelium overlying the superior and middle pharyngeal constrictor muscles. Located behind the epithelium of the posterior pharyngeal wall are the retropharyngeal space, the longus capitis and longus colli muscles, the prevertebral fascia, and the vertebral bodies. Tumors of the posterior oropharyngeal wall usually grow anteriorly into the pharyngeal cavity and posteriorly into the prevertebral muscles. However, invasion through the prevertebral fascia and into the vertebral bodies is rare. Tumors of the lateral pharyngeal wall extend anteriorly and posteriorly along the mucosa and parapharyngeal space, can grow as exophytic masses into the pharyngeal cavity, and in advanced stages, can directly invade the neck. A CT scan of an oropharyngeal wall tumor is shown in Figure 11-6. The incidence of nodal involvement at presentation among 164 patients referred to M. D. Anderson with oropharyngeal wall squamous carcinomas between 1954 and 1974 was 57%.10

From Remmler D, Medina JE, Byers RM, et al. Treatment of choice for squamous carcinoma of the tonsillar fossa. Head Neck Surg 1985;7:206-211.

Soft Palate The soft palate consists of five muscles (levator veli ­palatini, tensor veli palatini, uvulae, palatoglossus, and palatopharyngeus) covered with stratified squamous epithelium on the oral surface and on most of the nasal surface. The soft palate has an important role in speech and swallowing. It closes off the nasopharynx during swallowing to prevent nasal reflux and during phonation to produce certain sounds. The lesser palatine nerve, a branch of the maxillary nerve, provides sensory innervation to the soft palate. The tensor veli palatini muscle is innervated by the mandibular nerve through the branch to the medial pterygoid muscle. The other soft palate muscles are

Etiology and Epidemiology Oropharyngeal squamous cell carcinoma most often affects patients older than 50 years with a prolonged history of abusing tobacco or alcohol, or both. However, ­ approximately

228

PART 3  Head and Neck

­ rimary tumor. Patients who ingested more than 14 alcop holic drinks per week after treatment of an index cancer had a 50% increase in the risk of second primary tumors.1 Most carcinomas of the tonsil and soft palate are easily visualized. Early diagnosis requires identification of the neoplastic cause of a patient’s sore throat or other suspicious symptoms. Some soft palate carcinomas manifest as superficial erythema, with or without slightly raised nodules or plaques. Carcinomas of the tongue base generally are not visible by simple inspection. Patients who present with soreness in the region of the posterior tongue should be examined by indirect laryngoscopy or fiberoptic nasopharyngoscopy and palpation of the tongue base.

Molecular Characteristics

FIGURE 11-6.  CT shows a posterior oropharyngeal wall squamous carcinoma extending onto the right lateral oropharyngeal wall. No deep extension into the prevertebral muscles is evident. (From Gunderson LL, Tepper JE [eds]. Clinical Radiation Oncology, 2nd ed. Philadelphia: Elsevier Churchill Livingstone, 2007, p 679.)

20% of cases occur in people who are younger and do not smoke or abuse alcohol. Whether disease in those cases has a viral cause is the subject of considerable research. The association between HPV-16 and other oncogenic HPV viruses and oropharyngeal carcinoma, particularly tonsillar carcinoma, has been observed by several investigators. Strome and colleagues11 measured HPV products by polymerase chain reaction techniques in samples from 52 patients with tonsil cancer and found that HPV-associated carcinomas were more prevalent in younger patients than in older ones. Li and colleagues12 at the Sydney Head and Neck Cancer Institute also found HPV (mostly HPV-16) in 46% of 86 tonsil carcinoma specimens. They further found that patients with HPV-positive tumors were less likely to experience recurrence or die of disease than were patients with HPV-negative tumors.12 The consensus of opinion is that HPV-associated tonsil carcinomas occur more often in younger patients who do not have the typical risk profile of smoking or alcohol abuse and that disease in these individuals tends to respond better to therapy.13

Prevention and Early Detection Smoking and, to a lesser degree, heavy alcohol ingestion are known to promote the induction of oropharyngeal tumors. Most patients with oropharyngeal squamous cell carcinomas smoke or chew tobacco, and discontinuing these ­habits decreases the risk of developing a primary or secondary oropharyngeal carcinoma. Do and collaborators1 found that patients who continued to smoke after treatment of an index cancer had twice the risk of developing a second

Ang and others14 studied the influence of epidermal growth factor receptor (EGFR) expression on survival and pattern of relapse in 155 patients with squamous cell carcinoma of the head and neck treated as part of protocol 90-03 of the Radiation Therapy and Oncology Group (RTOG). Of the 155 patients studied, 82 (53%) had oropharyngeal carcinoma. EGFR expression was not correlated with disease stage at presentation or with any other known prognostic factors. Rates of overall survival and locoregional control for patients with tumors that expressed high levels of EGFR were significantly lower (P = .0006 and P = .003, respectively) than those rates for patients with tumors that expressed less EGFR. The rates of distant metastasis were similar in the two groups. These results were later confirmed by Ang in an as-yet unpublished analysis of patients treated with concomitant boost radiation therapy in RTOG 90-03.

Clinical Manifestations, Evaluation, and Staging In their early stages, oropharyngeal tumors are often asymptomatic. Pain, which can be local or radiate to the ear, is a common first symptom. The initial sign of disease is often an asymptomatic neck mass located in the jugulodigastric region. Masses in the tonsil and soft palate can be seen easily on direct examination. Tongue base carcinomas usually cannot be visualized directly, which may lead to a delay in diagnosis. Locally advanced disease can manifest with trismus, necrotic odor, odynophagia, dysphagia, and dysarthria.

Patient Evaluation Physical examination should include assessment of the ­local extent of the primary tumor and the presence and location of lymphadenopathy. The size of the primary tumor and nodal disease should be measured, because these measurements form the basis of the American Joint Committee on Cancer (AJCC) staging system (Box 11-1) for oropharyngeal cancer.15 Impairment of tongue protrusion in the anterior or lateral direction is a sign of involvement of the root of the tongue. Trismus, a consequence of infiltration of the pterygoid muscles, is documented by measuring the distance between the upper and lower incisors or alveolar ridges in edentulous patients. Tumor extension into the larynx or pyriform sinus should be assessed by ­fiberoptic ­nasopharyngoscopy, because this finding has

11  The Oropharynx

229

BOX 11-1.  American Joint Committee on Cancer Staging System for Tumors of the Oropharynx Primary Tumor (T) T1 Tumor 2 cm or less in the greatest dimension T2 Tumor more than 2 cm but not more than 4 cm in the greatest dimension T3 Tumor more than 4 cm in the greatest dimension T4a Tumor invades the larynx, deep/extrinsic muscle of tongue, medial pterygoid, hard palate, or mandible T4b Tumor invades lateral pterygoid muscle, pterygoid plates, lateral nasopharynx, or skull base or encases carotid artery Regional Lymph Nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Metastasis in a single ipsilateral lymph node, 3 cm or less in the greatest dimension N2 Metastasis in a single ipsilateral lymph node, more than 3 cm but not more than 6 cm in ­great dimension, or in multiple ipsilateral lymph nodes, none more than 6 cm in the greatest ­dimension, or in bilateral or contralateral lymph nodes, none more than 6 cm in the greatest dimension N2a Metastasis in a single ipsilateral lymph node more than 3 cm but not more than 6 cm in the greatest dimension N2b Metastasis in multiple ipsilateral lymph nodes, none more than 6 cm in the greatest dimension N2c Metastasis in bilateral or contralateral lymph nodes, none more than 6 cm in the greatest dimension N3 Metastasis in a lymph node more than 6 cm in the greatest dimension Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage Grouping: Oropharynx Stage 0 Tis N0 M0 Stage I T1 N0 M0 Stage II T2 N0 M0 Stage III T3 N0 M0 T1 N1 M0 T2 N1 M0 T3 N1 M0 Stage IVA T4a N0 M0 T4a N1 M0 T1 N2 M0 T2 N2 M0 T3 N2 M0 T4a N2 M0 Stage IVB T4b Any N M0 Any T N3 M0 Stage IVC Any T Any N M1 From Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2007, pp 33-45.

a crucial ­impact on the radiation portal design. Documentation of the anterior tumor border in relation to the position of radiopaque anatomic structures, such as the ­retromolar trigone, hard palate, and various teeth, also is useful for radiation portal design. All borders of the tumor must be ­determined precisely, particularly if IMRT is planned. In addition to a complete history and physical examination, the staging workup for oropharyngeal tumors includes a complete blood cell count, routine serum chemistry studies, and imaging studies, including chest radiography, fiberoptic endoscopy, and CT or magnetic resonance imaging (MRI) of the head and neck. Contrast-enhanced CT or MRI is essential for defining the anatomic extensions of the primary tumor and nodal disease; low-volume nodal disease that is not evident on physical examination often can be detected with these imaging techniques. Imaging can

demonstrate the presence of retropharyngeal lymphadenopathy (Fig 11-7A), invasion into the pterygoid muscles (see Fig. 11-7B), and the extent of tongue base invasion. At M. D. Anderson, oropharyngeal tumors are routinely staged with CT, and MRI is used occasionally to improve the visualization of fascial planes (Fig. 11-8). Patients who present with advanced nodal disease should undergo thoracic CT, which is more sensitive than chest radiography for detecting lung metastases. Other tests are performed as indicated by the clinical findings. The ability of PET scanning to detect occult systemic metastases that cannot be identified by standard CT scanning is being investigated.16 Many patients present with asymptomatic subdigastric lymphadenopathy, which should be evaluated if indicated by the clinical presentation. Malignant nodes associated with oropharyngeal tumors often appear cystic on CT scans

230

PART 3  Head and Neck

A

B

FIGURE 11-7.  A, MRI shows a large, right retropharyngeal lymph node in a patient with tongue base cancer. B, CT shows an infiltrative tonsil carcinoma directly invading the pterygoid muscles, thereby causing trismus. (From Gunderson LL, Tepper JE [eds]. Clinical Radiation Oncology, 2nd ed. Philadelphia: Elsevier Churchill Livingstone, 2007, p 680.)

A

B

FIGURE 11-8.  A, CT scan of a patient with a T4N2c tonsil carcinoma shows diffuse involvement of the tonsil, pterygoid muscles, parapharyngeal space, and prevertebral muscles. B, Magnetic resonance image of the same patient shows improved definition of the involved structures and tissue planes. A right-sided retropharyngeal node is visible, but the tumor does not seem to involve the prevertebral musculature. The lateral pterygoid muscle shows less diffuse infiltration on MRI than on CT. (From Gunderson LL, Tepper JE [eds]. Clinical Radiation Oncology, 2nd ed. Philadelphia: Elsevier Churchill Livingstone, 2007, p 680.)

and should not be confused with benign branchial cleft cysts. The preferred method for obtaining a biopsy sample of an oropharyngeal node is by fine-needle aspiration, with sonographic guidance used if the node is not palpable. Excisional node biopsy can be problematic because the associated scarring in the operative bed can be difficult to distinguish from tumor, thereby complicating treatment

planning. Advanced nodal involvement can manifest as matting, fixation, and extension through the skin. Patients who present with evidence of disease in the jugulodigastric or, less commonly, the midcervical nodes but no clear primary tumor are likely to have an oropharyngeal primary tumor. Examination in such cases should be performed under anesthesia, with biopsy of suspicious

11  The Oropharynx

mucosal sites of the head and neck. Early-stage tongue base cancers can appear as mucosal erythema, hypervascularity, or slight thickening of the mucosa without an obvious mass. Manipulation of the tongue base during the examination under anesthesia can cause bleeding, which can be a sentinel sign of an occult tongue base primary tumor. Small tonsillar tumors can lead to subtle asymmetry of the tonsils, and an ipsilateral tonsillectomy is often performed if a primary lesion is not found.

Staging The 2002 AJCC staging system for oropharyngeal carcinomas is shown in Box 11-1.15 All of the information obtained before definitive treatment is begun that could contribute to the accuracy of disease assessment, including findings from CT, MRI, and sonography, should be included in the staging process. The distinction between TI, T2, and T3 tumors is based on tumor size. In the 2002 staging system, T4 lesions can be T4a (i.e., resectable, meaning they invade the larynx, deep or extrinsic muscles of tongue, medial pterygoid muscle, hard palate, or mandible) or T4b (i.e., unresectable, meaning they invade the lateral pterygoid muscle, pterygoid plates, lateral nasopharynx, or skull base or encase the carotid artery). Stage IV disease has also been subdivided into IVA and IVB according to whether the disease is T4a or T4b or whether nodal disease is considered N2 or N3 (see Box 11-1). No staging system is ideal. For example, in the AJCC system, the T4 classification is based on “invasion” of local tumor into adjacent structures, meaning that a small tumor in the vallecula spreading into the lingual surface of the epiglottis or an extension of a soft palate tumor onto the adjacent hard palate would be considered T4. This shortcoming must be considered in formulating treatment ­recommendations.

Primary Therapy General Principles of Treatment The goal in treating oropharyngeal tumors, as is true for tumors at other sites, is to maximize tumor control while minimizing functional impairment and cosmetic deformities. The choice of treatment depends on the stage and site of the disease. As discussed in the remainder of this chapter, modifications of conventional radiation therapy techniques such as altered fractionation schedules or IMRT have been shown to improve outcome by reducing treatment-related morbidity.

Treatment Policies According to Disease Stage Early-Stage Carcinomas For early-stage oropharyngeal carcinomas (T1 or small T2 N0-1 disease), primary radiation therapy or surgery (often followed by postoperative radiation therapy as indicated by the pathologic findings) can produce similar ­local and regional control rates. Consequently, the choice of treatment is based on the anticipated functional and cosmetic outcome and the need for and extent of nodal therapy. For example, surgery is an option for treating small and superficial tumors of the anterior faucial pillars or tonsillar fossa,

231

particularly when resection can be done through a transoral approach and the likelihood of obtaining tumor-free resection margins is high. Ipsilateral radiation therapy is also suitable for the treatment of most T1 and early T2 N0-1 carcinomas of the tonsillar region with minimal extension onto the soft palate or other midline structures, because unilateral irradiation in this situation can result in high local and regional control rates without causing permanent xerostomia. In most cases, early-stage oropharyngeal carcinomas can be treated with IMRT, which improves the likelihood that adequate parotid flow will be retained. Radiation therapy is preferred for the treatment of small tonsil tumors that extend onto the soft palate and for almost all soft palate tumors because resection of portions of the soft palate would impair swallowing. Tumors involving midline oropharyngeal structures should be treated with bilateral irradiation. At M. D. Anderson, almost all patients with early-stage oropharyngeal carcinoma are treated with radiation therapy.

Intermediate-Stage Carcinomas The preferred treatment for intermediate-stage carcinomas (i.e., larger T2 or exophytic T3 N1-2a disease) is primary radiation therapy, given in an accelerated-fractionation schedule (e.g., concomitant boost or 6 fractions per week), with or without cetuximab, or in combination with concurrent chemotherapy, because most patients with intermediate-stage carcinomas treated with primary surgery end up requiring postoperative radiation therapy because of multiple positive nodes, extranodal extension, or close resection margins. Moreover, resection of intermediate-stage carcinomas often causes some loss of normal tissue function. Details of fractionated treatment regimens and IMRT are presented in Chapter 7. Approaches to the treatment of nodal disease are discussed under “Integration of Treatment of Primary Tumor and Nodal Disease” later in this chapter.

Advanced Infiltrative Primary Tumors Depending on the patient’s clinical condition, the expertise of the treating team, and the patient’s preference, advanced infiltrative primary tumors (i.e., endophytic T3 or T4a N23b disease) can be treated with radiation plus concurrent chemotherapy or cetuximab; with neoadjuvant (induction) chemotherapy with docetaxel, cisplatin, and fluorouracil followed by radiation therapy; or with composite primary resection, nodal dissection, and postoperative radiation, with or without chemotherapy. Large tissue deficits can be reconstructed with microvascular free flaps, which make these procedures more cosmetically acceptable. However, because of the severe functional losses associated with the resection of large oropharyngeal carcinomas, most patients (including those with T4b any N disease) are treated with radiation in combination with chemotherapy or cetuximab. Palliative treatment is occasionally recommended for patients with inoperable tumors who are in poor general condition.

Integration of Treatment of Primary Tumor and Nodal Disease Three general strategies are commonly used for patients with resectable nodal disease whose primary tumor is being treated with definitive radiation therapy (alone or with concurrent chemotherapy or cetuximab): radiation

232

PART 3  Head and Neck

A

B

FIGURE 11-9.  A, A 61-year-old man presented in August 2004 with a firm T2N0 squamous carcinoma in the right base of tongue. Intensity-modulated radiation therapy was given in a concomitant boost fractionation schedule to 72 Gy in 40 fractions over 6 weeks. A significant palpable tongue mass was evident 10 days after completing treatment, and an interstitial implant involving 12 catheters was applied 4 days later. Under direct palpation, the first three catheters were placed 1 cm apart in a coronal-plane row 1 cm behind the mass. Three more coronal-plane rows, also 1 cm apart, were placed progressively more anteriorly until the persistent mass was encompassed in three dimensions. B, The treatment goal was to deliver a 15-Gy boost by using iridium 192 at 70 cGy per hour. The red isodose cloud (generated with BrachyVision software) indicates the tissue volume receiving 15 Gy. The patient was without disease and functioning well without complications in March 2008. (From Gunderson LL, Tepper JE [eds]. Clinical Radiation Oncology, 2nd ed. Philadelphia: Elsevier Churchill Livingstone, 2007, p 684.)

therapy followed by neck dissection for patients with nodal masses that persist 6 to 12 weeks after the completion of radiation therapy; neck dissection followed by radiation therapy; or planned neck dissection after radiation therapy, even for patients in whom nodal disease regresses completely after irradiation. The strategy used in most cases at M. D. Anderson is the first of these three: initial irradiation (alone or with concurrent chemotherapy or cetuximab), with neck dissection reserved for residual disease that is present 6 to 8 weeks after the completion of therapy. However, if extensive dental extractions are necessary before radiation therapy, the neck dissection and the dental surgery can be done during the same period of general anesthesia (the second strategy). Patients selected for neck dissection before radiation therapy usually have nodes that are 5 cm or larger, which are more likely to require neck dissection even after radiation therapy. This strategy, the second of the three, is used less often now than in the past because of the high complete response rates that can be achieved with chemoradiation therapy and the emergence of effective induction chemotherapy regimens (e.g., docetaxel, cisplatin, and fluorouracil). The best treatment for large nodal disease located low in the level IV lymph nodes (i.e., below the cricoid cartilage) or in the supraclavicular fossa near the brachial plexus may be a planned neck dissection, regardless of the response to radiation. The brachial plexus, after traversing between the anterior and medial scalene muscles, is ­ relatively ­ superficial within the lateral supraclavicular

f­ ossa, and it is difficult to exclude from the high-dose ­volume. If the brachial plexus will receive doses higher than 64 to 66 Gy in the field required for nodal treatment, it is preferable to give a lower radiation dose (e.g., 60 Gy) followed by a neck dissection.

Treatment Policies According to Disease Subsite Tongue Base Carcinomas Many centers regard definitive IMRT (with concurrent cetuximab for endophytic lesions) as the treatment of choice for T1-2 N0-2a tongue base carcinomas. Induction chemotherapy with docetaxel, cisplatin, and fluorouracil followed by radiation therapy also is the preferred treatment for most patients with T1-2 N2b-3 tongue base carcinomas. For T3 or T4 tumors, chemotherapy (or cetuximab) is usually given concurrently with radiation therapy to enhance the tumor response. Lymph nodes on both sides of the neck are treated in all patients because tongue base tumors are situated at or close to the midline and have a high propensity for bilateral nodal spread. In highly selected cases, a boost dose to the primary tumor or a supplemental dose to a slowly regressing tumor can be delivered by interstitial therapy (Fig. 11-9). The traditional therapy for resectable T4 tongue base tumors is surgery and postoperative radiation therapy. The surgery often requires a laryngectomy to prevent aspiration and can lead to severe dysphagia from loss of the ­propulsive

11  The Oropharynx

force of the tongue base. In clinical trials, tumors at this stage are increasingly being treated with combinations of systemic therapy and radiation, with the goal of organ preservation.

Tonsillar Region Carcinomas Tumors of the tonsillar fossa are radiosensitive, and high local control rates can be achieved with radiation therapy alone. At many centers, radiation therapy is the preferred treatment for T1 or T2 tonsillar fossa carcinomas. To avoid radiation-induced permanent xerostomia, unilateral irradiation has been increasingly used in the treatment of T1 or early T2 N0-1 carcinomas of the tonsillar fossa with minimal extension onto the soft palate. IMRT techniques can effectively spare the contralateral parotid and submandibular glands. More advanced tumors are treated with fields that include the bilateral neck nodes at risk. In selected cases in which primary tumor extension into the tongue base persists after external beam radiation therapy (EBRT), a boost dose designed to eradicate the remaining tumor is administered by brachytherapy. An additional dose of 10 to 15 Gy can be given in this manner.17 In such cases, the implant should be placed within 2 weeks after the EBRT is completed. The preferred therapy for most T1-2 N3 tumors is induction chemotherapy with docetaxel, cisplatin, and fluorouracil followed by radiation therapy. The policy for treating T1-2 N2 disease is still evolving. Docetaxel, cisplatin, and fluorouracil chemotherapy is preferred for bulky N2 disease or disease that involves lower neck nodes (level III or IV). Radiation therapy with concurrent chemotherapy or cetuximab is recommended for most T3 and T4 carcinomas. Some T4 tumors are best treated by surgery and comprehensive postoperative radiation therapy, with or without chemotherapy, depending on the anatomic extent and infiltrative characteristics of the tumor. Involvement of unresectable structures such as the carotid, nasopharynx, and skull base should be ruled out before surgery is attempted. The amount of involvement of theoretically resectable tissues, such as the mandible, soft palate, pterygoid muscles, and tongue, also determines the feasibility of surgery. The role of chemoradiation therapy for organ preservation is being actively investigated.

Soft Palate Carcinomas The overall treatment strategy for soft palate carcinomas is similar to that for tongue base carcinomas. Early-stage soft palate tumors often have indistinct margins and are associated with other areas of erythroplakia. Consequently, negative surgical margins can be difficult to achieve with a limited excision. Surgical excision of all but the smallest soft palate carcinomas causes velopharyngeal dysfunction, nasal regurgitation, and speech dysfunction that cannot be adequately corrected with prosthetic devices. IMRT usually is considered to be the treatment of choice in these cases. For cases involving accessible, well-delineated T1 or T2 tumors, a boost dose of orthovoltage radiation can be delivered by means of an intraoral cylindrical applicator. This approach is most applicable for edentulous patients. The boost dose, usually in the range of 15 to 18 Gy in 5 or 6 fractions, is generally administered before EBRT when the tumor borders can be clearly visualized. Another reason for delivering the boost dose before EBRT is that patients can

233

tolerate the applicator better if it is used before mucositis develops. Induction chemotherapy with docetaxel, cisplatin, and fluorouracil followed by radiation therapy has become the preferred approach for most patients with T1-2 N2-3 disease. Radiation therapy with concurrent chemotherapy or cetuximab is the preferred treatment for T3 and T4 tumors, except for cases in which the tumor is judged to be T3 on the basis of its length but is superficial and of low volume.

Oropharyngeal Wall Carcinomas Radiation therapy usually is the preferred treatment for oropharyngeal wall tumors. Because the posterior pharyngeal wall is immediately adjacent to the prevertebral fascia and prevertebral muscles, it is usually not possible to resect tumors of the posterior oropharyngeal wall with wide margins, except perhaps for very small, superficial tumors. Carcinomas of the oropharyngeal wall are usually treated with radiation therapy alone (for T1-2 N0-1 disease); induction chemotherapy with docetaxel, cisplatin, and fluorouracil followed by radiation therapy (for T1-2 N2-3 disease); or radiation therapy given concurrently with chemotherapy or cetuximab (for more locally advanced disease).

Treatment Results The findings from relatively large series of radiation therapy for oropharyngeal carcinoma are summarized in this section. Experiences with IMRT and the experience at M. D. Anderson with standard-field radiation therapy for oropharyngeal carcinoma in general are described under “Oropharyngeal Carcinoma,” followed by a discussion of findings according to disease subsite. Additional findings on the combination of radiation therapy with cetuximab or induction chemotherapy with docetaxel, cisplatin, and fluorouracil, followed by radiation or combined radiationcarboplatin chemotherapy, for advanced head and neck cancer are summarized in Chapter 7.

Oropharyngeal Carcinoma Intensity-Modulated Radiation Therapy In most reports of outcomes after IMRT, results for all head and neck sites are grouped together, but the three reports discussed here focused on the oropharynx.18-20 Chao and colleagues18 described treatment outcomes for 74 patients after IMRT at Washington University for oropharyngeal carcinoma; 43 patients were given IMRT as adjuvant therapy after surgery, and 31 were given IMRT as definitive treatment. The primary tumors were T1 or T2 in 55% of cases; 84% of patients had nodal disease. The mean dose to patients in the definitive-treatment group was 70 Gy, given in 35 fractions over 7 weeks; the mean dose to the adjuvant-therapy group was 66 Gy, given in 2-Gy fractions over 6.6 weeks. The mean parotid dose was reported to be 18.6 Gy. Chemotherapy was given to 17 patients, mostly concurrently with the IMRT. Ten patients (13%) experienced locoregional recurrence at a median follow-up time of 13 months. The investigators’ conclusion was that IMRT was effective and had a favorable toxicity profile. Garden and others19 described outcomes for 54 patients with Tx, T1, or T2 oropharyngeal carcinoma treated with IMRT at M. D. Anderson up to June 2002; 45 of these

234

PART 3  Head and Neck

­ atients had nodal disease. Concurrent chemotherapy was p given to seven patients, four of whom had N3 disease. The most common fractionation schedule, used for 72% of patients, was 66 Gy given in 30 fractions over 6 weeks to the primary tumor and nodal disease and 54 Gy to subclinical sites. A concomitant boost schedule was used for four patients with larger T2 tumors (72 Gy in 40 fractions over 6 weeks). Acute toxicity was similar to that seen with the use of parallel-opposed fields and similar fractionation schedules, but as expected, the mean parotid dose was markedly reduced. At a median follow-up interval of 13 months, the locoregional control rate was 91%. The actuarial 2-year relapse-free and overall survival rates were 86% and 89%, respectively. Although parotid function was not specifically measured, it was thought to be better than what would be expected after parallel-opposed field treatment. Updated results were presented by Garden and others at the 2006 meeting of the American Society of Therapeutic Radiology and Oncology. Of the 259 patients treated up to June 2004, 85% had T1 or T2 tumors, and 87% had clinically involved nodes. Sixty-two patients who had T3 or T4 tumors and N2 or N3 disease were given chemotherapy in addition to IMRT. The 3-year actuarial locoregional control and overall survival rates were 94% and 88%, respectively. A full report of this series is being prepared. De Arruda and colleagues20 at Memorial SloanKettering Cancer Center reported results of IMRT for 50 patients with oropharyngeal carcinoma. Sixty-six percent of patients had T1 or T2 tumors, and 82% had clinical nodal involvement. Concurrent chemotherapy, usually singleagent cisplatin, was used for 86% of patients. The median follow-up time was 18 months. The most common fractionation schedule used was 70 Gy given in 33 fractions over 6.6 weeks, with doses of 59.4 Gy and 54 Gy given to regions of high-risk and low-risk subclinical disease in the same field. The 2-year actuarial local control rate was 98%, and the regional control rate was 88%. Ten patients underwent neck dissection for persistent nodal masses, and only one patient had pathologic nodal disease. The incidence of RTOG grade 2 late toxicity was 33%. Three patients had esophageal strictures that required dilatation. Saliva output improved with time after the completion of therapy. The investigators, who included some prominent surgeons at Memorial Sloan-Kettering, concluded that IMRT was ideal for oropharyngeal cancer in terms of effectiveness and tolerability. Acute toxic ­effects were manageable, and the focused sparing of the parotid glands was considered beneficial for patients’ quality of life after treatment. IMRT has become the standard of care for oropharyngeal cancer at M. D. Anderson because of the excellent locoregional control rates achieved and the improvement in parotid function over conventional radiation therapy techniques. Parallel-opposed fields continue to be used only for frail patients who cannot tolerate the longer duration of daily therapy required for IMRT and for the occasional patient with tumor so extensive that significant parotid sparing cannot be achieved.

Standard-Field Radiation Therapy: The M. D. Anderson Experience Results of radiation therapy for 175 patients with T1-2 N0 squamous cell carcinoma of the oropharynx treated at M. D. Anderson between 1970 and 1998 were reported

in 2004 by Selek and colleagues.21 Most of the patients in that series (124 [71%]) had T2 tumors. The two most common treatment schedules were standard daily fractionation to a median dose of 66 Gy in 33 fractions over 6.6 weeks (85 patients) and concomitant boost fractionation to 72 Gy in 42 fractions over 6 weeks (73 patients). The 5-year actuarial local control rate for patients with T1 or Tx disease was 92% and that for patients with T2 disease was 81%. The local control rate for each subsite within the oropharynx did not vary significantly. The 5-year actuarial regional control rate was 93%. Surgical salvage was successful in 11 of 34 patients who had a local or regional recurrence. The 5-year actuarial distant metastasis rate was 8%. Grade 4 complications included mucous membrane ulceration (nine patients) and bone necrosis and exposure (seven patients). Nine patients also had esophageal strictures that required dilatation. Actuarial overall survival rates were 70% at 5 years and 43% at 10 years; the corresponding disease-free survival rates were 85% and 79%. The low 10-year survival rate was explained in part by the appearance of second primary tumors, which developed in 29% of patients. Eighty-six percent of the second tumors were in the upper aerodigestive tract, and the development of second primary tumors correlated strongly with active smoking or heavy alcohol consumption. The high rate of esophageal stricture in this and other M. D. Anderson series led to a change in the standard supraclavicular fields used at that institution. Formerly, the supraclavicular field was treated to 50 Gy in 25 fractions specified at the point of maximum dose (Dmax) with no midline shielding below the laryngeal block. The more recent practice is to place a midline block in the supraclavicular field after 40 Gy to shield the esophagus. After this modification was introduced, esophageal stricture has essentially been eliminated as a complication of neck radiation in patients treated with this technique. Garden and colleagues22 analyzed a subset of 299 patients with T1 and T2 oropharyngeal carcinoma who presented to M. D. Anderson with regional adenopathy between 1975 and 1998. According to the 2002 AJCC staging system, these tumors would be considered stage III or IV, and the current treatment recommendation would be concurrent systemic therapy. However, during the period of this study, radiation alone was used. Nevertheless, the 5-year actuarial local control rates were 99% for patients with T1 or Tx tumors and 85% for patients with T2 tumors. The 5-year actuarial regional control rates for patients with Nx, N1, and N2 disease were 100%, 93%, and 91%, respectively; patients with N3 disease presentations had a 5-year regional control rate of 76%. Focusing on the N2 subclassifications, the 5-year actuarial regional control rates for patients with N2a, N2b, and N2c disease were 94%, 89%, and 89%, respectively. For patients in whom local control was maintained, the 5-year actuarial regional control rates for those with N1, N2a, N2b, N2c, and N3 disease were 97%, 95%, 88%, 95%, and 82%, respectively. The actuarial 5-year locoregional control rate for patients with T1 or Tx disease was 95%, and that for those with T2 disease was 79%. Patients in whom locoregional disease control was maintained had an actuarial 5-year distant metastasis rate of 17%. For patients who presented with Nx, N1, or N2a disease, the 5-year actuarial distant metastatic rate was 10%, in contrast to 26% for those with N2b

11  The Oropharynx

235

TABLE 11-2.    Local Control Rates for Squamous Carcinoma of the Tongue Base Study and Reference

Institution

T1

T2

T3

T4

Comments

M. D. Anderson ­Cancer Center

91%

71%

78%

52%

1954-1971, varied once-daily fractionation

Foote et al24

University of Florida

89%

90%

81%

36%

Once-daily fractionation

Wang25

Massachusetts ­General Hospital

87%

74%

28%

17%

Once-daily fractionation

Weber et al26

M. D. Anderson ­Cancer Center

100%

86%

59%

44%

Interstitial boosts in 14 patients (8%); once-daily fractionation

Wang25

Massachusetts ­General Hospital

90%

84%

64%

28%

Accelerated split-course ­fractionation

Mak et al 27*

M. D. Anderson Cancer Center

100%†

98%

76%

0%‡

Concomitant boost fractionation

Mendenhall et al28

University of Florida

96%

91%

81%

38%

Once-daily fractionation: 31% Hyperfractionation: 69%

Harrison et al29*

Memorial ­SloanKettering Cancer Center

87%

93%

82%

100%§

External beam radiation ­therapy plus interstitial ­radiation therapy

Spanos et

al23

*5-Year

actuarial. of three patients. ‡None of one patient. §Two of two patients. †Three

­ isease, 21% for those with N2c disease, and 43% for those d with N3 disease. The authors of this retrospective analysis22 concluded that patients with small primary tumors and Nx, N1, or N2a presentations would not benefit from further intensification of therapy with concurrent chemotherapy. More advanced nodal disease is more likely to be associated with distant metastasis and requires treatment designed to counteract this pattern of disease relapse. Radiation treatment techniques for regional disease have improved considerably since the first 15 years (i.e., 1975-1990) of the study described earlier.22 Before the early 1990s, M. D. Anderson radiation oncologists used a skin match technique for the neck junction that was geometrically imperfect. If the junction had been placed through an N3 node, some portion of the node would be underdosed. This may partially explain the high rate of regional recurrence among patients with N3 disease in that analysis.

Oropharyngeal Subsites Tongue Base Carcinomas External Beam Radiation Therapy. Reported ­ local control rates after radiation therapy for treatment of tongue base carcinomas are shown in Table 11-2.23-29 Radiation delivered in standard daily 2-Gy fractions yields a control rate of approximately 90% for T1 tongue base tumors. The dose per fraction and total dose used varied somewhat among series, but the results were similar. For T2 tumors, the local control rates ranged from 70% to 90%. A wide range of local control rates has been reported for T3 tumors; this may reflect heterogeneity, because there is no upper size limit for the T3 classification. Complications ­ after once-daily fractionation have included mandibular exposure, soft tissue necrosis, and dysphagia. The main complication in a large series reported from M. D.

­ nderson in 1976 was mandibular exposure and necroA sis.23 Foote and colleagues at the University of Florida24 reported a 4% incidence of moderate to severe bone exposure and a 7% incidence of moderate soft tissue necrosis. In a subsequent report from the University of Florida, Mendenhall and colleagues28 found a 4% rate of severe complications among patients treated predominantly with hyperfractionated radiation therapy; those complications included severe dysphagia ­requiring a permanent gastrostomy (2%) and osteonecrosis (1%). External Beam Radiation Therapy plus Brachytherapy. High rates of local and regional control of tongue base carcinomas can be achieved by using a combination of EBRT and brachytherapy. Harrison and colleagues29 described 68 patients treated with this approach at Memorial Sloan-Kettering. Forty-eight of those patients (71%) had tumors classified as T1 or T2. Patients were treated with 50 to 54 Gy of EBRT to the primary tumor site and upper neck followed by a 20- to 30-Gy boost to the primary tumor using an iridium 192 (192Ir) implant. The 5-year actuarial local control rates were 87% for those with T1 tumors, 93% for those with T2 tumors, and 82% for those with T3 tumors. The 5-year actuarial regional control rate was 96%. Complications, including soft tissue ulceration, mandibular osteonecrosis, and bleeding at the time of catheter removal, occurred in 19% of the patients. Similar results for patients with tongue base carcinomas treated with EBRT and brachytherapy were reported from Stanford University30 and from Long Beach Memorial Medical Center.31 With the high control rates that can be achieved with IMRT, altered fractionation schedules, and concurrent chemoradiation therapy, the use of brachytherapy for the management of tongue base carcinoma has declined, because it requires the patient to undergo significant surgery for no clear ­benefit in results.

236

PART 3  Head and Neck

TABLE 11-3.   Local Control Rates for Patients with Squamous Carcinoma of the Tonsil Treated with Standard 2-Gy Fractionation at M. D. Anderson between 1968 and 1983 EVALUABLE PATIENTS†

ALL PATIENTS* Tumor Stage

Local Control after XRT (%)

Local Control after XRT (%)

Salvage Surgery (n)

Local Control after Surgery (n)

Ultimate Local ­ Control (%)

T1

16/17 (94)

15/16 (94)

1

1/1

16/16 (100)

T2

48/59 (81)

41/52 (79)

6

3/6

44/52 (85)

T3

44/66 (67)

30/52 (58)

10

1/10

31/52 (60)

T4

5/8 (63)

3/6 (50)

1

0/1

3/6 (50)

Total

113/150 (75)

89/126 (71)

18

5/18

94/126 (75)

*Patients

were treated between July 1968 and December 1983. Analysis was performed December 1986. patients who died within 2 years after irradiation (XRT) without local failure were excluded from the analysis of local control. Length of follow-up after salvage therapy ranged from 26 to 162 months. Modified from Wong CS, Ang KK, Fletcher GH, et al. Definitive radiotherapy for squamous cell carcinoma of the tonsillar fossa. Int J Radiat Oncol Biol Phys 1989;16:657-662.

†Twenty-four

Surgery Alone for Early-Stage Tumors. Foote and colleagues7 reported a series of 55 patients with tongue base carcinomas who were treated with surgery alone at Mayo Clinic between 1971 and 1986. (During that era, the surgeons at Mayo Clinic were largely opposed to ­radiation therapy, including postoperative radiation, for such patients.) All 55 patients underwent partial glossectomy, and 11 of the patients also underwent partial or complete laryngectomy. The crude rates of local control were 77% for those with T1 tumors, 81% for T2 ­tumors, and 75% for T3 tumors. The crude rate of regional control was 56%, and the rate of disease control above the clavicles was 49%. Sixteen patients required further surgery to manage complications, which included fistula and ­abscess formation (eight patients), osteomyelitis with bone necrosis (six patients), pectoralis myocutaneous flap failure (two patients), and hemorrhage (two patients). ­Five ­patients required permanent gastrostomy tubes. The ­surgical mortality rate was 4%. In summary, the ­ locoregional control rate in this experience using surgery alone seems to be worse than locoregional control rates achieved with radiation therapy, and the ­complication rate was substantial. Surgery and Radiation Therapy in Combination. ­Extensive (>6 cm), infiltrative T3 and T4 tumors of the tongue base tend to recur locally when treated with radiation therapy alone. Historically, these tumors were treated with a combination of surgery and postoperative radiation therapy. De Los Santos and colleagues32 reported a series of 51 patients with advanced tongue base carcinoma who were treated with surgery and postoperative radiation therapy at M. D. Anderson between 1980 and 1997. Almost 90% of the patients had T3 tumors (21 patients) or T4 tumors (23 patients). All patients underwent glossectomy, and this was accompanied by laryngectomy in 51% of patients and mandibulectomy in 35%. The median radiation therapy dose was 63 Gy. The 5-year actuarial locoregional control rate was 74%, and the 5-year rate of distant metastases was 35%. Although swallowing function was not formally assessed, chronic swallowing deficits of various degrees occurred in 21 patients. Because of the significant functional problems seen after treatment with surgery plus radiation therapy, patients with advanced tongue base carcinomas are being treated in clinical trials of radiation therapy and chemotherapy.

Tonsillar Region Carcinomas External Beam Radiation Therapy: M. D. Anderson Series. Wong and colleagues33 reported on 150 patients with tonsillar fossa carcinoma treated with EBRT with 2-Gy fractionation at M. D. Anderson between 1968 and 1983. The mean tumor doses were as follows: 64 Gy to T1 tumors, 68 Gy to T2 tumors, and 70 Gy to T3 tumors. Boosts were usually given through ipsilateral reduced fields, most often with a combination of high-energy electrons and photons. In 26 patients, neck dissection was performed approximately 6 weeks after the completion of radiation therapy. The local control rates after radiation therapy were 94%, 79%, and 58%, respectively, for T1, T2, and T3 tumors (Table 11-3).33 The ultimate local control rates after surgical salvage therapy were 100% for T1 tumors, 85% for T2 tumors, and 60% for T3 tumors. The regional control rate, calculated after exclusion of patients who had a local recurrence, was 96%. Seventeen severe complications developed in 14 patients, including bone necrosis requiring mandibular resection (7 patients), soft tissue necrosis (4 patients), trismus (5 patients), and cranial nerve palsies (1 patient). All 14 patients received doses exceeding 67.5 Gy. Most of the treatments during this era consisted of onefield radiation therapy given once daily with 2:1 weighting, a technique that increases the dose per fraction received by the ipsilateral normal tissues. Before the era of IMRT, 98 patients with intermediatestage tonsillar fossa carcinoma (41 with T2 tumors and 57 with T3 tumors) had been treated at M. D. Anderson with a concomitant boost fractionation schedule that involved giving 54 Gy in 1.8-Gy fractions over a period of 6 weeks to lateral fields that included the primary tumor and the draining lymphatics. The 5-year actuarial local control rate for patients with T2 tumors treated with this schedule was 86%; that for patients with T3 tumors was 83% (WH Morrison, unpublished data). External-Beam Radiation Therapy: Non–M. D. ­Anderson Series. Over a 33-year interval, 400 patients with squamous carcinoma of the tonsillar region were treated at the University of Florida.34 Hyperfractionation to a median dose of 76.8 Gy was used for 60% of the patients; the other 40% were treated with standard fractionation. In addition to the EBRT, 27% of the patients received

11  The Oropharynx

237

TABLE 11-4.    Local Control of Soft Palate Tumors Treated with Radiation Therapy Tumor Stage Study and Reference

Institution

Keus et al41

Netherlands Cancer Institute

Lindberg and Fletcher42

M. D. Anderson Cancer Center

Wang25

Massachusetts General Hospital (1970–1994)

Fein et al43

University of Florida

Mazeron et al44

Henry Mondor Hospital

Fein et al43

University of Florida

Wang25

Massachusetts ­General Hospital

Morrison, ­unpublished data

M. D. Anderson Cancer Center

Erkal et al45

University of Florida

Comment

T1

T2

T3

T4

92%

67%

58%

37%

100%

88%

77%

83%

96%

81%

55%

24%

9/11 (81%)

13/20 (65%)

5/10 (50%)

1/4

13/14

20/23

—

—

Twice-daily ­ fractionation

1/1

10/10 (100%)

6/10 (60%)

0/3

Accelerated split course

91%

79%

68% (T3 + T4)

2/2

21/22 (95%)

13/20 (65%)

—

86%

91%

67%

36%

Once-daily ­fractionation

Once-daily ­fractionation

Once-daily ­fractionation: 53%; Twice-daily ­fractionation: 47%

a brachytherapy boost of 10 to 15 Gy, usually to the tongue base component of the tumor. Planned neck dissections were performed in 125 patients after the completion of radiation therapy. The local control rates were 83% for T1 tumors, 81% for T2 tumors, 74% for T3 tumors, and 60% for T4 tumors. When tumors of the anterior palatoglossal arches were excluded, the local control rates for the tonsillar fossa and posterior palatoglossal arch tumors that remained were 90% for T1 tumors, 88% for T2 tumors, 66% for T3 tumors, and 58% for T4 tumors. Nineteen patients experienced severe late complications, the most common of which were osteonecrosis necessitating a segmental mandibulectomy (eight patients) and swallowing difficulty necessitating a permanent gastrostomy (six patients). Bataini and colleagues35 reported on a series of 465 patients with squamous carcinoma of the tonsillar region who were treated almost exclusively with EBRT at the Institut Curie. Mean radiation doses were 65 Gy for T1 and T2 tumors, 67 Gy for T3 tumors, and 68 Gy for T4 tumors. The local control rates for T1, T2, T3, and T4 tumors were 89%, 84%, 63%, and 43%, respectively. The investigators reported that local control rates were higher for tumors of the tonsillar fossa than for tumors originating in other sites of the tonsillar region (P < .01). A 2001 report from Princess Margaret Hospital in ­Toronto described a series of 228 patients with tonsillar carcinoma treated with unilateral irradiation.36 The primary tumors were located in the tonsillar fossa in 150 patients, in the anterior tonsillar pillar in 73 patients, and in the posterior tonsillar pillar in 5 patients; 84% of the tumors were T1 or T2. Fifty-eight patients had N0 disease, 57 had N1 disease, and 39 had N2 or N3 disease. Radiation was delivered predominantly by a unilateral wedge-pair ­technique.

Only eight patients (3%) experienced contralateral nodal relapse. Among patients whose primary tumors were locally controlled, the contralateral nodal recurrence rate was 1.7%. No contralateral nodal recurrences occurred in patients who had T1 tumors or in patients who presented with N0 neck disease. The investigators recommended unilateral radiation therapy for tonsillar region carcinomas with at most minor extension (1 cm) onto the soft palate or the mucosa of the tongue base. In another report from Canada, the incidence of contralateral nodal recurrence was 2.6% in a series of 155 patients with tonsillar carcinoma treated with unilateral radiation at the Vancouver Cancer Center.37 Moderate-Dose External Beam Radiation Therapy with Brachytherapy Boost. Several investigators have reported good results treating tonsillar fossa tumors with EBRT to doses in the range of 45 to 50 Gy followed by an interstitial brachytherapy boost. Pernot and colleagues38 reported a series of 263 patients with velotonsillar carcinomas treated with EBRT to 50 Gy in 5 weeks followed by a brachytherapy boost of 20 to 30 Gy. Local control rates for T1, T2, and T3 tumors were 89%, 85%, and 67%, respectively. Mazeron and colleagues39,40 treated 69 patients with T1 and T2 tonsil tumors with EBRT to 45 Gy followed by a 30-Gy interstitial boost. At a minimum follow-up time of 2 years (median > 5 years), only 2 of the 69 patients had developed locoregional recurrence.

Soft Palate Carcinomas External Beam Radiation Therapy. Outcomes of various radiation therapy approaches to the treatment of soft palate carcinoma are shown in Table 11-4.25,41-45 At the Netherlands Cancer Institute,41 146 patients with soft

238

PART 3  Head and Neck

­ alate carcinoma were treated with EBRT alone (71%) or p with an intraoral applicator followed by EBRT (29%). The mean total dose was 68 Gy, and the mean treatment duration was 45 days. As would be predicted, patients who received boost doses with the intraoral cylinder had fewer normal tissue complications. For T3 and T4 tumors, control rates did not increase when doses were escalated from 60 to 75 Gy; however, doses were sometimes increased when persistent tumor was present at the completion of the initially planned dose. The investigators recommended limiting the EBRT dose to 70 Gy to minimize complications. Erkal and colleagues45 reported outcomes for 107 patients with soft palate carcinoma who underwent radiation therapy at the University of Florida between 1965 and 1996. Radiation boosts were given with orthovoltage x-rays through an intraoral applicator for 29 patients and with brachytherapy for 8 patients. Standard daily fractionation was used in 53% of the treatment courses and hyperfractionation in 47%. The median dose given with the intraoral applicator was 15 Gy in 3-Gy fractions. Median doses and treatment duration for once-daily fractionation were 65 Gy in 48 days, and those for hyperfractionation were 76.8 Gy in 46 days. Actuarial 5-year local control rates were 86% for T1 tumors, 91% for T2, 67% for T3, and 36% for T4 tumors. Local control rates were worse among patients with prolonged treatment time (P = .0002). Thirty-six patients developed 51 second primary tumors. Brachytherapy. Several groups, mostly from France, have achieved good results treating soft palate tumors with interstitial irradiation, usually in combination with EBRT. Mazeron and colleagues44 reported delivering a mean EBRT dose of 47 Gy and a mean brachytherapy dose of 31 Gy. The crude local control rates with this approach were 93% and 87% for T1 and T2 tumors, respectively. The incidence of soft tissue necrosis was low, and in most cases, complications could be managed conservatively. An additional six patients with small tumors were treated with interstitial therapy alone, and in all cases, local control was achieved. At the Institut Gustave-Roussy,46 43 patients with early-stage soft palate carcinoma were treated with interstitial radiation with or without EBRT. The local disease control rate was 100% for all 34 patients with T1 tumors, and the actuarial local control rate for all patients was 92%.

Oropharyngeal Wall Carcinomas Standard Fractionated and Concomitant Boost Radiation Therapy. Hull and colleagues47 reported outcomes for 148 patients treated between 1964 and 2000 at the University of Florida. Tumor epicenters were in the hypopharyngeal wall in 63% of cases and in the oropharyngeal wall in 37%. Most of the patients were treated with hyperfractionated (twice-daily) radiation therapy to 76.8 Gy. Eleven patients also received chemotherapy. Local control rates after radiation therapy were 93% for T1 tumors, 82% for T2 tumors, 59% for T3 tumors, and 50% for T4 tumors. The ultimate local control rate for T2 tumors, including those for which salvage treatment was successful, was 87%. On multivariate analysis, the locoregional control rates were significantly better for oropharyngeal carcinomas than for hypopharyngeal carcinomas (P = .02) and for patients given hyperfractionated rather than conventionally fractionated therapy (P = .0009). In another study of

26 patients with intermediate-stage oropharyngeal wall carcinoma (15 with T2 disease and 11 with T3 disease) treated with a concomitant boost schedule at M. D. Anderson, the 5-year actuarial local control rates were 93% for those with T2 tumors and 82% for those with T3 tumors (WH Morrison, unpublished data).

Control of Nodal Disease The treatment policies for regional disease from oropharyngeal cancer vary among institutions. At Memorial SloanKettering before the era of IMRT, patients with tongue base carcinoma and palpable nodes were routinely treated with neck dissection after radiation.29,48 The primary tumor and neck lymphatics were initially treated with EBRT to 54 Gy given in 1.8-Gy fractions. Nodal disease was then further irradiated with an electron beam to 60 Gy. The primary tumor was treated with a boost dose by using an interstitial implant, and patients underwent modified radical neck dissection during the same operation to place the implant. In a series of 50 patients treated with this approach, 70% of the patients had pathologically negative neck dissection specimens. Rates of complete pathologic response were similar for patients who presented with advanced nodal stage and for those with more limited nodal disease. The 5-year actuarial rate of regional control was 96%.29 Morbidity from neck dissection after receipt of 60 Gy was minimal. The investigators observed that their treatment policy might lead to overtreatment in some cases. However, the strategy of giving 60 Gy followed by neck dissection was considered particularly beneficial when significant nodal disease was present in the low neck because high radiation doses given to the region of the brachial plexus could cause brachial plexopathy. At M. D. Anderson, regional disease is treated using one of three strategies (see “Integration of Treatment of Primary Tumor and Nodal Disease”). Patients who will require general anesthesia for extensive dental extractions and who may require a neck dissection after radiation therapy (usually patients with nodal disease ≥ 5 cm at presentation) are sometimes treated with neck dissection before radiation therapy. For the uncommon presentation of extensive nodal disease caudal to the cricoid and overlying the brachial plexus, disease is treated to 60 Gy followed 6 weeks later by a planned neck dissection. Another option for treating extensive low neck disease is neoadjuvant chemotherapy; if the low neck nodes respond to that treatment, the radiation dose to the low neck can be reduced to 64 to 66 Gy. Most patients are treated initially with definitive radiation therapy, and disease is reassessed by clinical examination and CT scanning approximately 6 to 8 weeks after the completion of radiation therapy. Patients who show a complete clinical and radiologic response are observed thereafter. Similarly, patients with small (<1.5 cm) amounts of nonenhancing nodal tissue on CT imaging and negative findings on sonographically guided needle aspiration are usually observed. Patients who show persistent nodal masses larger than 1.5 cm or show persistent tumor on fine-needle aspiration undergo selective neck dissection in which the residual mass and immediately surrounding nodes are removed. The potential usefulness of PET scanning at 2 months after the completion of radiation therapy to see if persistent nodal masses contain viable tumor is an area of active study.16,49

11  The Oropharynx

An analysis of 92 patients treated with one concomitant boost schedule at M. D. Anderson50 revealed that the 17 patients who underwent neck dissection before irradiation had slightly larger nodes at presentation than did the 75 patients who were treated initially with radiation therapy (median node size, 4 versus 3 cm). Four (23%) of the 17 patients who were initially treated with neck dissection had a regional recurrence. Of the 75 patients initially treated with radiation therapy, 13 underwent a planned neck dissection for persistent nodal masses. Nodal size at presentation was larger than 3 cm in 12 of these 13 patients. None of the 13 patients experienced regional recurrence. The remaining 62 patients had a complete response in the neck and therefore did not undergo neck dissection after radiation therapy. Subsequent nodal recurrence occurred in 7 (11%) of those 62 patients, of whom 3 had an isolated regional relapse. In the series described previously,50 no differences in nodal control were found between the different oropharyngeal subsites, and no relationship was found between initial nodal size and the risk of subsequent recurrence in the neck. The actuarial 2-year probability of control of neck disease was 87% for patients presenting with nodes smaller than 3 cm and 85% for patients presenting with nodes larger than 3 cm. However, the probability of complete response to radiation therapy was lower in patients who initially presented with larger nodes. The conclusion from that analysis was that neck dissection could be omitted for patients whose nodes are included in the high-dose volume and who respond completely to treatment. Treatment delivery techniques have improved significantly since 1993 (when the study period in that report ended), and sonographically guided fine-needle aspiration allows more accurate assessment of nodal response after the completion of radiation. In a 2004 abstract, Garden and colleagues19 presented outcomes for 80 patients with oropharyngeal cancer who had undergone IMRT at M. D. Anderson. All patients presented with small primary tumors and nodal disease; 22 had N1 disease, 17 had N2a disease, 27 had N2b disease, 2 had N2c disease, and 12 had Nx disease. The median follow-up time was 17 months. Twenty-six patients had neck dissections after radiation therapy for persistent nodal masses, and only five patients (19%) had pathologic evidence of disease at that time. The other 54 patients experienced complete response and were observed after radiation therapy; only 3 of those patients experienced regional recurrence. The actuarial 2-year regional control rate was 93%. Phase III trials of the RTOG51 and the European Organisation for Research and Treatment of Cancer (EORTC)52 have shown that cisplatin, when given concurrently with postoperative radiation, can improve survival rates among patients with positive margins and extracapsular extension of nodal disease (see Chapter 7).

Distant Metastasis In analyzing the outcome of a series of 751 patients with oropharyngeal carcinoma treated with radiation therapy at M. D. Anderson between 1960 and 1974, Lindberg and Fletcher53 found that distant metastasis was the first site of relapse in 7.7% of the patients. The crude incidence of distant metastases was 2% for patients with stage I to III disease and 13% for those with stage IV ­disease. In the RTOG

239

90-03 study,54 in which 62% of the 1073 patients had oropharyngeal cancer and 31% had T4 primary tumors, the actuarial incidence of distant metastases at 2 years was 17%. Al-Othman and colleagues55 analyzed the rate of distant metastasis among 873 patients who underwent definitive radiation therapy for head and neck cancer at the University of Florida; 246 patients had tumors of the tonsil, 166 had tumors of the tongue base, and 42 had tumors of the soft palate. The 5-year rate of distant metastasis-free survival for the entire group was 86%. The corresponding rates for patients with tongue base tumors or tonsillar tumors were 86%, and that for patients with soft palate tumors was 98%. Factors that predicted significant increases in the rate of distant metastasis on multivariate analysis included increasing T stage, the presence of nodal disease in the lower neck, and recurrent head and neck disease (each P < .0001); other factors were increasing N stage (P < .006) and male sex (P = .03). The median time to the development of distant metastases was 12 months, and the median survival time after the diagnosis of distant metastases was 4 months.

Reirradiation for Recurrent or New Tumors after Radiation Therapy In the past, full-dose reirradiation of head and neck tumors was considered futile. However, investigations at some academic centers have demonstrated that between 10% and 20% of selected patients who undergo repeat radiation therapy can survive for 5 years or more after such treatment. A group at the University of Chicago56 published a report in 2006 describing long-term outcomes for 115 patients treated with chemotherapy and repeat irradiation for recurrent and second primary squamous carcinomas; 30 of those patients had oropharyngeal carcinoma. The reirradiation was done for clinically evident disease in 57% of cases and as planned adjuvant (postoperative) therapy in 43%. The median cumulative total radiation dose was 131 Gy, and the median reirradiation dose was 65 Gy. Patients were treated on seven consecutive chemotherapy protocols with different drugs. The 3-year actuarial survival rate was 22%. Multivariate analysis indicated that factors linked with overall survival were reirradiation dose (P < .001), having surgery before reirradiation (P < .002), and receiving cisplatin, paclitaxel, or gemcitabine (P < .017). Patients who received doses of 58 Gy or more had a 3-year overall survival rate of 30%, compared with only 6% among those who received less than 58 Gy. Grade 4 or 5 complications included carotid hemorrhage in 6 patients, osteoradionecrosis in 13 patients, myelopathy in 1 patient, and peripheral neuropathy in 1 patient. The conclusion from that study is that full-dose reirradiation with concurrent chemotherapy is a viable treatment option that can produce long-term survival among selected patients with recurrent or second primary head and neck cancer. Similar results were reported by De Crevoisier and colleagues57 at the Institut Gustave Roussy. In that series, 169 patients underwent repeat irradiation. The median cumulative dose was 120 Gy. The actuarial 5-year overall survival rate was only 9%. The incidence and severity of late toxicities were markedly higher after the second irradiation than after the first irradiation. Multivariate analyses indicated that the volume irradiated in the second irradiation

240

PART 3  Head and Neck

was the only factor linked with the risk of lethal toxicity (P = .01). These and other studies have shown that repeat irradiation given concurrently with chemotherapy does have some curative potential and is worth offering to patients who have no other therapeutic option. Results are better for patients whose disease can be treated with limited fields. Ideally, patients considering reirradiation should have good performance status and apparently healthy ­ irradiated tissues.

Palliation Patients with incurable tumors may benefit from receiving a short course of palliative radiation therapy. Oropharyngeal carcinoma is radiosensitive and responds to standard palliative regimens such as 30 Gy given in 10 fractions. Because this regimen causes significant mucositis, however, some patients with a short life expectancy are treated with 14 Gy given in 4 fractions over 2 days. This regimen can be repeated every 3 to 4 weeks for three cycles to a total dose of 42 Gy; the third course involves an off–spinal cord dose reduction. The 14-Gy regimen has the advantage of being below the dose threshold for mucositis.

Radiation Techniques Primary Radiation Therapy Treatment Setup and Portal Arrangement Essential steps to successful treatment include reproducible positioning of patients (usually immobilized in the supine position), accurate tumor localization through the integration of findings from physical examination with those from high-quality imaging, appropriate delineation of target volumes, and attention to dose distribution and fractionation. For patients with tongue base carcinoma, a tongue­depressing stent can be useful for separating the tongue base from the superior structures of the oropharynx. Tongue-depressing stents are also used for patients with soft palate carcinoma to displace the soft palate from the tongue base. Shoulder straps are sometimes necessary to depress the patient’s shoulders out of the horizontal upper neck fields. Use of extended thermoplastic masks or other techniques incorporating shoulder fixation is needed for immobilization of patients undergoing IMRT, which is the ­preferred technique for treating oropharyngeal carcinoma. The general approach is to deliver 66 to 70 Gy (depending on T category and fractionation) to the high-dose clinical target volume (CTVHD), which encompasses the primary tumor and involved nodes (if present) with margins of at least 1 to 1.5 cm, and an elective dose (50 to 59 Gy, depending on the fractionation) to the CTVED, which comprises tissues surrounding the tumor and clinically uninvolved nodes. Delineation of the CTVID, which is administered an intermediate dose (60 to 63 Gy), around the CTVHD can improve conformation of the dose distribution.

Intensity-Modulated Radiation Therapy The exquisitely precise localization of tumor required for planning IMRT fields requires input from the physical examination, which often includes fiberoptic endoscopy,

and imaging studies, which can include CT, MRI, PET (in investigational settings),16 and sonography. Sonographically guided fine-needle aspiration of lymph nodes can be helpful. Infiltrative tumors with diffuse borders require wider margins than exophytic tumors with distinct borders. Two logical options in IMRT planning are to treat the oropharynx and entire neck in one large field or to use a monoisocentric match technique.58 The monoisocentric setup requires that the junction between the oropharyngeal IMRT field and the supraclavicular field be placed at least 1.5 cm from the inferior border of the primary tumor. This technique is particularly appealing if the junction can be placed above the arytenoids, because the lower neck can be treated with an anterior portal with a laryngeal block to minimize the dose to the arytenoids, vocal cords, and cricopharyngeus muscle. With the current treatment software, the dose to the larynx, even when specified as an avoidance structure, is considerably higher when the whole neck is treated with IMRT. At M. D. Anderson, the monoisocentric approach has been used consistently if the junction between the upper neck and the supraclavicular fields can be placed above the shoulders. The junction is placed regardless of whether it transverses through involved nodes. Treatment planning software allows the IMRT and supraclavicular isodose distributions to be displayed on one plan and the dose through the junction to be observed; we have not observed nodal recurrences attributable to use of this technique. Outlining an oropharyngeal tumor is not as simple as delineating the gross tumor volume and assigning a standard margin in all directions around it; it also requires knowledge of the typical patterns of spread from the various anatomic subsites. For example, posterior pharyngeal wall tumors rarely involve the vertebral body, and a smaller margin can be assigned in that direction if imaging confirms that vertebral body invasion is not present. The following questions should be contemplated before the dose volumes are drawn: 1. Where exactly is the primary tumor and malignant adenopathy, and with what certainty can the tumor anatomy be defined? 2. What dose fractionation schedule, with or without chemotherapy, will optimize the therapeutic ratio in each specific case? 3. How close to the tumor can the margins be drawn in each specific case without risking missing known tumor? Can the descending dose volumes around the high-dose region covering the primary tumor be used to treat immediately adjacent tissues that may contain subclinical tumor extensions? 4. Which sites of potential subclinical spread will require a dose necessary to control subclinical disease? 5. What are the consequences to the normal tissues of delivering the specified dose? What are the consequences of having excessive dose in the target ­volumes to the underlying normal tissue? Drawing tumor volumes and normal tissue avoidances is such a subjective process that describing margin dimensions purely in centimeters underestimates the cognitive process required. The tumor dose volume must be sagaciously ­determined after all of the available information is considered, and the resultant margin around the tumor should be appropriate for the clinical presentation. ­Strategies used

11  The Oropharynx

FIGURE 11-10.  A 48-year-old man presented in February 2005 with a tonsillar tumor and a 4-cm ipsilateral subdigastric node. Unilateral intensity-modulated radiation therapy was used because the tumor was confined to the tonsillar fossa with no significant medial extension to the soft palate or base of tongue. (From Gunderson LL, Tepper JE [eds]. Clinical Radiation Oncology, 2nd ed. Philadelphia: Elsevier Churchill Livingstone, 2007, p 692.)

for this process for each oropharyngeal disease subsite are discussed in the following paragraphs. Tonsillar Region Carcinomas. The CTVHD for tumors that are centrally located in the tonsillar fossa should include at a minimum the lateral pharyngeal wall, the tonsillar fossa, and the anterior tonsillar pillar. The minimum field length extends from the lower aspect of the maxillary tuberosity to the tip of the epiglottis for early-stage tumors and to the hyoid or below for more advanced disease. Disease extension onto the soft palate can be localized with some precision by inspection and palpation. The anterior border of the CTVED should extend at least to the retromolar trigone, parapharyngeal space, and some of the adjacent pterygoid muscles. The amount of pterygoid muscle to be treated for subclinical spread increases with increasing T category. The maxillary tuberosity and the retromolar trigone are useful anatomic landmarks for planning because both can be precisely localized by physical examination and imaging. Unilateral irradiation encompassing the primary ­tumor and ipsilateral node for T1-2 N0-1 tumors of the tonsillar region is used at many centers based on the ­ results from studies done at the Princess Margaret Hospital,36 ­Vancouver Cancer Center,37 and Mayo Clinic.9 In the Mayo Clinic study, 56 patients (45 with T1 or T2 tumors) underwent surgery that did not include contralateral neck dissection, but only 6 patients (11%) experienced contralateral nodal recurrence. A typical primary tonsillar fossa tumor treated with unilateral irradiation is shown in Figure 11-10. The CTVHD encompasses the primary tumor and any nodal disease with margins of at least 1 to 1.5 cm. The CTVED encompasses the ipsilateral retromolar trigone and pterygoid muscles, the ipsilateral upper neck nodes, and the ipsilateral retropharyngeal nodes anterior to the C1 vertebra. An intraoral stent can be used to push the tongue contralaterally out of the high-dose region as much as possible. The dose to the contralateral submandibular and parotid glands and to the ipsilateral parotid gland can be minimized if IMRT is used.

241

Figure 11-11 shows the CT scan of and dose distributions for a small T1 tonsillar carcinoma treated with ipsilateral IMRT. Figure 11-12 shows the treatment of a more advanced tonsil carcinoma. Base of Tongue Carcinomas. Custom-fabricated tongue-depressing stents can be useful for immobilizing the tongue in the treatment of tongue base tumors, but some patients cannot tolerate these devices because of excessive gag or saliva buildup during long treatment sessions. For early-stage tongue base carcinomas, the temptation to draw an anterior margin that is too small should be avoided, especially given the mobility of the tongue. For superficial tongue base tumors that are only vaguely palpable or have been removed by excisional biopsy, the field should be specified such that the full dose is delivered 2 cm anterior to the mucosa. If the tumor is well localized on one side of the tongue base, a lower dose can be given to the contralateral side (Fig. 11-13). For more extensive tongue base tumors, the anterior and lateral borders can be palpated and observed on imaging; if the margins are difficult to pinpoint, delineation of a wider CTVHD may be necessary. The hypopharynx and larynx, including the pre-epiglottic space, should be assessed to see if the disease extends inferior to the hyoid. If not, the field junction can be placed above the arytenoid cartilages. If one pyriform sinus is involved, it may be possible to reduce the dose to the contralateral laryngohypopharynx. Soft Palate Carcinomas. For soft palate tumors, the boost dose is given by means of an intraoral applicator when feasible. The CTVHD includes the tumor with a 1.0- to 1.5-cm margin if the tumor borders are distinct on physical examination and the if tumor can be identified on the planning CT scan. The uvula is a useful reference point because it can be easily localized on examination and imaging. Soft palate carcinomas often show superficial spreading and have vague borders, in which case wider margins of coverage are necessary. The neck and the retropharyngeal nodes should be targeted in the IMRT fields. The course of treatment of a soft palate carcinoma is illustrated in Figure 11-14. Oropharyngeal Wall Carcinomas. Oropharyngeal wall tumors can extend posteriorly around the paravertebral muscles. These extensions can be easily encompassed in an IMRT field. The borders of the tumor can usually be determined by integrating the findings from physical examination with those from imaging studies. Given the midline nature of these tumors, both sides of the neck are irradiated. Treatment Planning and Dose Prescriptions for the Upper Neck and Retropharyngeal Lymph Nodes. The clinical target volume in head and neck cancer includes the tumor and various lymph nodes in the neck (see “Treatment Setup and Portal Arrangement”). Establishing guidelines for selective dissection or irradiation of neck nodes in various forms of head and neck cancer has been hampered in the past by inconsistencies in delineating the various nodal levels in the neck, which prevents the uniform delineation of elective-dose clinical target volumes (CTVED) in the neck by radiation oncologists. In an attempt to rectify this shortcoming, a consensus group consisting of the Danish Head and Neck Cancer Group, the EORTC, the French Head and Neck Cancer Group (GORTEC), the National Cancer Institute of Canada, and the RTOG published in

242

PART 3  Head and Neck

A

B

C FIGURE 11-11.  A 35-year-old woman presented in April 2004 with a T1 tumor of the right tonsil diagnosed by tonsillectomy.  A, Ipsilateral intensity-modulated radiation therapy was deemed appropriate when the CT scan revealed a 3-cm cystic jugulodigastric lymph node but showed no evidence of contralateral node involvement. B and C, Isodose distributions of planned treatment.  B, The tonsillar bed received 66 Gy given in 30 fractions; the mean dose to the left parotid gland and left submandibular gland was 10 Gy. C, A 68.5-Gy isodose line covers the involved node, which was contoured. A single-fraction boost dose of 1.5 Gy was given with a 9-MeV electron beam for a final nodal dose of 70 Gy. At 6 weeks, the nodal remnant was 8 mm in diameter. At 2 months after completion of treatment, fine-needle aspiration of the node and positron emission tomography showed no evidence of tumor, and the neck was observed thereafter. The patient was free of disease and had normal salivary output as of February 2008. (From Gunderson LL, Tepper JE [eds]. Clinical Radiation Oncology, 2nd ed. Philadelphia: Elsevier Churchill Livingstone, 2007, p 693.)

11  The Oropharynx

243

A

B

C

D

E FIGURE 11-12.  A, A 76-year-old man presented in July 2005 with advanced tonsil squamous carcinoma that displaced the uvula to the left. B, CT shows obliteration of the right parapharyngeal space and invasion of the right pterygoid muscles, which caused mild trismus. The tumor was classified as T3N0, and concurrent chemoradiation therapy with weekly carboplatin and paclitaxel was  prescribed. Radiation was given in a concomitant boost fractionation schedule to 72 Gy in 40 fractions over 6 weeks. C and  D, Isodose plans in the transverse and coronal planes. In the transverse plane (C), the tumor volume demonstrated on the diagnostic CT scan is covered by the 72-Gy isodose line. Regions of potential subclinical extension are encompassed by the 69-Gy line. In the coronal plane (D), approximately 70% of the parotid received less than 26 Gy. E, Delivery of 30.4 Gy in 16 fractions produced significant tumor response, with erythematous mucositis in the midline of the soft palate and some “tumoritis” in the clinical tumor.

244

PART 3  Head and Neck

F

G

FIGURE 11-12.—Cont’d  F, The oropharynx on the last day of treatment shows resolution of the tumor and diffuse, confluent mucositis. G, At 5 weeks after chemoradiation therapy, no tumor was visible, and the mucositis had mostly healed, with some moderate erythema. The patient was free of disease and doing well in June 2007. (From Gunderson LL, Tepper JE [eds]. Clinical Radiation Oncology, 2nd ed. Philadelphia: Elsevier Churchill Livingstone, 2007, p 696.)

2003 a set of guidelines that define the location of nodal beds at all levels of the neck for patients with head and neck squamous cell carcinoma with no clinical evidence of lymphadenopathy.59 For neck nodes in levels II and V, the consensus recommendation is to treat cranially up to the level of the caudal edge of the lateral process of C1. The retropharyngeal nodes should be treated to the level of the jugular fossa bilaterally, unless unilateral treatment is being delivered. The consensus group did not recommend that parapharyngeal nodes be included in the elective nodal coverage. Eisbruch and others60 at the University of Michigan used findings from their retrospective analysis to make recommendations about the superior extent of contralateral neck coverage that is required. In that review of 133 patients with head and neck cancer (80 of whom had oropharyngeal cancer), the contralateral neck was clinically node negative but was judged to be at high risk for subclinical disease; the question was where to place the cranial and caudal borders so that tumor control and parotid sparing would be greatest. The contralateral neck was uniformly treated superiorly to an anatomic landmark (where the digastric muscle crosses the jugular vein). There were no recurrences in the contralateral neck superior to this level, leading the group to conclude that the uninvolved contralateral neck need not be treated more cranially than their landmark, which is located slightly more caudal than the consensus recommendation. However, they did state that further studies were needed to clarify exactly how much superior coverage was needed in the contralateral neck. Other guidelines issued by the RTOG as part of its protocol 0522 for delineating target volumes around advanced tumors further illustrate the complexity of the IMRT planning process. That protocol stipulates defining a gross tumor volume (GTV) surrounded by a CTVHD (with a 1-cm margin) and a CTVED (with a 2-cm margin). The CTVED should also cover the nodal sites intended to receive the elective dose. For infiltrative tumors or those with ill-defined borders, it is desirable to add a CTVID that is slightly larger than the CTVED. Moreover, the protocol stipulates delineating planning target volumes (PTVs) that have a

­ inimum margin of 0.5 cm around each CTV to compenm sate for variability in daily treatment setup. The treatment goal in this protocol, in which IMRT is delivered with chemotherapy, is to deliver a dose of 70 Gy in 35 fractions over 6 weeks to the PTVHD. The subclinical dose to the PTVED is 56 Gy at 1.6 Gy per fraction. The dose prescribed to the PTVID could range from 59.5 Gy to 63 Gy. The mean dose to at least one parotid gland should be kept at or below 26 Gy; as an alternative, the dose to at least 20 cm3 of the combined volume of both parotid glands should be less than 20 Gy, or the dose to at least 50% of one gland should be less than 30 Gy. The dose to the larynx should be less than 45 Gy whenever feasible, and the dose to the brachial plexus should be no more than 60 Gy for patients who have evidence of lymphadenopathy in level IV nodes. Delineating GTVs, CTVs, and PTVs requires considerable judgment about the relative benefit and toxicity of delivering specific doses to malignant and normal tissues. The final treatment plan must be reviewed and approved by a radiation oncologist as covering the tumor with a reasonable but not excessive margin. This assessment is preferably performed with the tumor-outlining function turned off, so that only the image of the tumor itself and the radiation doses are evident. At this point in the planning process, difficult cases may require that the patient be re-examined to further clarify whether the dose coverage is appropriate. The mean dose must be considered in addition to the minimum and maximum doses. If the focus of the planning is to have the minimum dose cover the entire PTV, the mean dose is often considerably higher, which may lead to excessive complications. Dose-Fractionation Schedules. Three fractionation regimens have been tested for tumors of the head and neck by the RTOG. The first regimen, used for locally advanced carcinomas of the oropharynx, hypopharynx, or larynx (protocol 0522), consists of 70 Gy to the PTVHD, 59.5 to 63 Gy to the PTVID, and 56 Gy to PTVED, all delivered in 35 fractions over 6 weeks. This schedule requires delivery of twice-daily treatment once each week for 5 weeks; the fraction sizes are 2.0 Gy to the PTVHD, 1.8 Gy to the PTVID, and 1.6 Gy to the PTVED. Consequently, the normal tissues

11  The Oropharynx

A

245

B

D

C FIGURE 11-13.  A and B, A 77-year-old man presented with a poorly differentiated, 4-cm, squamous cell carcinoma that was visible on CT scans as occupying most of the right base of the tongue and extending from the superior aspect of the base of tongue to just above the vallecula. Multiple, small but necrotic ipsilateral nodes were also evident. Intensity-modulated radiation therapy was delivered in a concomitant boost-type fractionation schedule combined with weekly cetuximab. C and D, The primary tumor and involved nodes in the right neck were treated with 57 Gy in 30 fractions, and regions potentially harboring subclinical disease were treated to 54 Gy in the same treatment field. During the final 2 weeks of treatment, 15 Gy in 10 fractions was delivered to regions showing clinical disease as a second daily fraction; the interfractional interval was 6 hours. The final dose to the base of the tongue and neck was 72 Gy given in 40 fractions over 6 weeks. The supraclavicular fossa was treated with an isocentric match to the intensity-­modulated radiation fields to 50 Gy in 25 fractions, with a full midline block inserted after 40 Gy to protect the larynx and esophagus. The right level III neck nodes received an additional dose of 10 Gy. The neck was not dissected, because sonography showed no residual adenopathy and repeat positron emission tomography at 8 weeks after treatment showed a complete metabolic response in the primary tumor and neck. The patient was free of disease and had an excellent functional outcome 2.5 years after treatment.

246

PART 3  Head and Neck

A

B

C

D

FIGURE 11-14.  A, Ulcerative soft palate squamous carcinoma is seen in a 58-year-old man. The tumor involved the entire soft palate and extended into both tonsillar fossae (right more than left), with some necrosis of the right soft palate. B, Transpalatal involvement evident on CT led to classification of the tumor as T3N0, which was treated with concurrent cisplatin at 100 mg/m2 every 3 weeks and a concomitant boost fractionation schedule to 72 Gy in 40 fractions over 6 weeks. The tongue was excluded from the higher doses with a tongue-depressing stent. C, The treatment plan shows coverage of the clinical tumor by the 72-Gy isodose line, with sparing of the parotid. D, During the final week of treatment, tumor resolution is evident but so are perforations of the soft palate, which are to be reassessed at follow-up. (From Gunderson LL, Tepper JE [eds]. Clinical Radiation Oncology, 2nd ed. Philadelphia: Elsevier Churchill Livingstone, 2007, p 698.)

further away from the CTVHD receive progressively smaller doses per fraction, which is expected to induce fewer late complications. The second regimen (protocol 0022) is used for early-stage oropharyngeal carcinoma and consists of 66 Gy to the PTVHD, 60 Gy to the PTVID, and 54 Gy to the PTVED, all delivered in 30 fractions over 6 weeks. Investigators at M. D. Anderson treat locally advanced oropharyngeal cancer with a fractionation regimen similar to that of RTOG 0225, which itself was adapted from a regimen pilot-tested at the University of California at San Francisco. In the M. D. Anderson regimen, the doses are 70 Gy to the PTVHD and 56 or 57 Gy to the PTVED, given in 33 fractions over 6.5 weeks (corresponding to a fraction size of 1.70 to 1.73 Gy), and most patients also receive concurrent chemotherapy. Another regimen used

at M. D. Anderson, mostly for intermediate-stage oropharyngeal carcinoma, is a modification of the concomitant boost fractionation schedule. The rationale for using this schedule is the superior local control afforded by accelerated compared with standard fractionation in RTOG 90-03. The initial component of this regimen delivers 57 Gy to the PTVHD and 54 Gy to the PTVED in 30 fractions. The boost component, consisting of 15 Gy in 10 fractions to the PTVHD, is given during the last 2 weeks of treatment as a second daily fraction with a 6-hour interval between fractions. The PTVHD receives 72 Gy in 40 fractions over 6 weeks (daily treatment for 4 weeks and twice daily for 2 weeks). The rates of tumor control and acute mucositis are similar to those seen when the concomitant boost fractionation schedule is used to treat conventional fields.

11  The Oropharynx

Standard-Field Radiation Therapy Standard-field radiation therapy techniques can be effective for patients with advanced tumors or as postoperative therapy for patients who have had bilateral level II neck dissections and have operative beds that extend bilaterally to the mastoid tips. Standard-field techniques are also useful for patients who cannot tolerate the prolonged immobilization required for IMRT. Parallel-Opposed Lateral Fields for Primary Tumor and Upper Neck Nodes. Parallel-opposed lateral fields are used to treat the primary tumor with 2-cm margins, usually with a 6-MV photon beam. The superior border of the initial portals should be high enough to cover the retropharyngeal nodes and high jugular nodes approaching the jugular fossa, especially for patients with advanced nodal disease. The ipsilateral pterygoid muscles should be treated in patients with tonsil carcinoma. The anterior border is determined by the extent of the primary tumor, and for tonsil tumors, it always should be anterior to the retromolar trigone. The posterior border is just behind the ­spinous ­ processes or more posteriorly in the presence of large posterior cervical nodal masses. The inferior border, which matches the anterior lower neck field, is placed above the arytenoid cartilages, unless the primary tumor extends into the laryngeal or hypopharyngeal region. A monoisocentric match is generally placed just above the arytenoids, even in the presence of nodal disease at the junction, to allow shielding of the larynx and cricopharyngeal muscles by placing a block in the anterior middle to lower neck portal. Large upper jugular node disease often is located in the junction between the photon field and the posterior cervical electron portals if horizontal off-cord fields are used. Coverage of such disease may be improved by using oblique off-cord and boost fields, angled so as not to transect the nodal disease. The oblique angle is selected to provide a 1-cm margin behind the nodal mass. Parallel-opposed oblique beam angles of 20 to 35 degrees above and below the horizontal are often necessary to achieve this aim. Anterior Fields for Middle and Low Neck Nodes. Anterior photon fields are used to treat the middle and lower neck nodes and the supraclavicular fossa bilaterally. Nodal disease in these regions can be given a boost dose through an appositional electron portal or glancing anteroposterior fields, or both, depending on the size and location of the affected node or nodes. Dose-Fractionation Schedules for External Beam Therapy. For T1 carcinomas of the oropharynx, conventionally fractionated radiation is prescribed (66 Gy given in 2-Gy fractions for 5 days each week over 6.5 weeks). With this fractionation schedule, the recommended dose for elective nodal irradiation is 50 Gy in 25 fractions and that for therapeutic irradiation of palpable involved nodes ranges from 64 to 70 Gy in 32 to 35 fractions, depending on the size of the nodes. Patients with T2 oropharyngeal cancer being treated outside of a study protocol are treated with a concomitant boost fractionation schedule. The recommended dose to the primary tumor is 69 to 72 Gy, given in 40 to 42 fractions over a 6-week period. With proper supportive care,

247

the proportion of patients in whom treatment must be extended beyond 6 weeks because of acute mucosal reactions can be limited to no more than 2%. It is no longer considered ethical to introduce planned treatment interruptions to reduce acute toxicity. Patients with T3 oropharyngeal carcinoma usually are treated with radiation therapy with concurrent chemotherapy or cetuximab. The standard regimen is 70 Gy, given in 35 fractions over 7 weeks. The role of alteredfractionation radiation therapy in combination with chemotherapy is being investigated. The RTOG protocol 0129 was designed to address whether concomitant boost fractionation is superior to standard fractionation when used for chemoradiation therapy. Patients were randomly assigned to one of the two fractionation schedules while receiving cisplatin (100 mg/m2 every 3 weeks). This study was closed to accrual in 2005, and its results should be available in 2009.

Postoperative Radiation Therapy Treatment Setup and Portal Arrangement The general technique for postoperative radiation ­therapy is similar to that for primary EBRT. The initial target ­volume covers the entire surgical bed and all nodal areas of the neck at risk, and the boost volume encompasses areas that harbor known disease. The disease spread and extent of surgery (i.e., scar or flap) determine the portal borders. IMRT or lateral-opposed photon fields are used for treating the primary tumor and upper neck nodes, with a matching anterior photon field used to cover the middle and lower neck. The location of the lower border of the lateral fields is determined predominantly by the extent of the primary disease. This border should be placed above the arytenoids whenever possible to avoid irradiating the larynx; a ­midline shielding block can be placed in the anterior low neck field to exclude the larynx from the beam. The skin and subcutaneous tissue over the larynx, if in the surgical bed, are irradiated with low-energy electrons. A bolus should be placed over the surgical scars to ensure that the surface dose is adequate (Fig. 11-15).

Dose-Fractionation Schedules When postoperative radiation therapy is given in 2-Gy fractions, a dose of 56 Gy is administered to the low-risk components of the operative bed, and 60 to 66 Gy is administered to regions of potentially high clonogenic density, such as regions of positive margins, multiple positive nodes, tumor dissected off the carotid artery, or extracapsular extension. Higher doses may be needed for selected patients, such as those who have gross residual disease after surgery.

Ongoing Studies The results of the cetuximab trial61 (see Chapter 7) demonstrated that combining biological agents with radiation can improve the survival of patients with head and neck cancer. These results have sparked interest in clinical trials and ­basic research into biological therapies. The RTOG is investigating the role of cetuximab for the definitive and postoperative treatment of head and neck cancers. The Eastern Cooperative Oncology Group has initiated a phase II trial

248

PART 3  Head and Neck

A

B FIGURE 11-15.  A 49-year-old man presented in April 2004 to M. D. Anderson Cancer Center after a modified radical left neck dissection and tonsil biopsy for a T1N3 squamous carcinoma of the left tonsil. He refused further treatment. A and B, Scar and subcutaneous recurrences in the left neck are evident in June 2005. This case illustrates the importance of using a bolus in postoperative patients who are treated with high-energy photons (>4 MV) to increase the dose delivered to the surgical scar and subcutaneous tissues. (From Gunderson LL, Tepper JE [eds]. Clinical Radiation Oncology, 2nd ed. Philadelphia: Elsevier Churchill Livingstone, 2007, p 701.)

investigating the role of cetuximab in the context of neoadjuvant and concurrent chemotherapy. Other ­ biological agents are available and are being tested in clinical trials. During the past decade, remarkable progress has been made in radiation dose delivery, and research along this line continues. The technology of standard linear accelerators continues to improve. Immobilization devices are becoming more sophisticated. Proton therapy is becoming more widely available in the United States, which will facilitate still more precise treatment planning.

REFERENCES   1. Do KA, Johnson M, Doherty D, et al. Second primary tumors in patients with upper aerodigestive tract cancers: joint effects of smoking and alcohol (United States). Cancer Causes Control 2003;14:131-138.

  2. Klijanienko J, el-Naggar A, Ponzio-Prion A, et al. Basaloid squamous carcinoma of the head and neck. Immunohistochemical comparison with adenoid cystic carcinoma and squamous cell carcinoma. Arch Otolaryngol Head Neck Surg 1993;119:887890.   3. Luna M, el-Naggar A, Parichatikanond P, et al. Basaloid squamous carcinoma of the upper aerodigestive tract. Cancer 1990;66:537-542.   4. Garden AS, Weber RS, Ang KK, et al. Postoperative radiation therapy for malignant tumors of minor salivary glands. Cancer 1994;73:2563-2569.   5. Dubey P, Ha CS, Ang KK, et al. Nonnasopharyngeal lymphoepitheliomas of the head and neck. Cancer 1998;82:1556-1562.   6. Lindberg R. Distribution of cervical lymph node metastases from squamous cell carcinoma of the upper respiratory and digestive tracts. Cancer 1972;29:1446-1449.   7. Foote RL, Olsen KD, Davis DL, et al. Base of tongue carcinoma: patterns of failure and predictors of recurrence after surgery alone. Head Neck 1993;15:300-307.   8. Remmler D, Medina JE, Byers RM, et al. Treatment of choice for squamous carcinoma of the tonsillar fossa. Head Neck Surg 1985;7:206-211.   9. Foote RL, Schild SE, Thompson WM, et al. Tonsil cancer. Cancer 1994;73:2638-2647. 10. Meoz-Mendez RT, Fletcher GH, Guillamondegui OM, et al. Analysis of the results of irradiation in the treatment of squamous cell carcinomas of the pharyngeal walls. Int J Radiat Oncol Biol Phys 1978;4:579-584. 11. Strome SE, Savva A, Brissett A, et al. Squamous cell carcinoma of the tonsils: a molecular analysis of HPV associations. Clin Cancer Res 2002;8:1093-1100. 12. Li W, Thompson C, O’Brien C, et al. Human papillomavirus positivity predicts favourable outcome for squamous carcinoma of the tonsil. Int J Radiat Oncol Biol Phys 2003;106:553-558. 13. Ringstrom E, Peters E, Hasegawa M, et al. Human papillomavirus type 16 and squamous cell carcinoma of the head and neck. Clin Cancer Res 2002;8:3187-3192. 14. Ang KK, Berkey B, Tu X, et al. Impact of epidermal growth factor receptor expression on survival and pattern of relapse in patients with advanced head and neck carcinoma. Cancer Res 2002;62:7350-7356. 15. Greene FL, Page DL, Fleming ID, et al. Pharynx (including base of tongue, soft palate, and uvula). In Greene FL, Page DL, ­Fleming ID, et al (eds): AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 33-45. 16. Frank S, Chao K, Schwartz D, et al. Technology insight: PET and PET/CT in head and neck tumor staging and radiation therapy planning. Nat Clin Pract Oncol 2005;2:526-533. 17. Gwozdz JT, Morrison WH, Garden AS, et al. Concomitant boost radiotherapy for squamous carcinoma of the tonsillar fossa. Int J Radiat Oncol Biol Phys 1997;39:127-135. 18. Chao KSC, Ozyigit G, Blanco AI, et al. Intensity-modulated radiation therapy for oropharyngeal carcinoma: impact of tumor volume. Int J Radiat Oncol Biol Phys 2004;59:43-50. 19. Garden AS, Morrison W, Rosenthal DI, et al. Intensity modulated radiation therapy for metastatic cervical adenopathy from oropharynx carcinomas [abstract]. Int J Radiat Oncol Biol Phys 2004;60:S318. 20. de Arruda F, Puri D, Zhung J, et al. Intensity-modulated radiation therapy for the treatment of oropharyngeal carcinoma: the Memorial Sloan-Kettering Cancer Center experience. Int J Radiat Oncol Biol Phys 2006;64:363-373. 21. Selek U, Garden A, Morrison W, et al. Radiation therapy for early-stage carcinoma of the oropharynx. Int J Radiat Oncol Biol Phys 2004;59:743-751. 22. Garden A, Asper J, Morrison W, et al. Is concurrent chemoradiation the treatment of choice for all patients with stage III or IV head and neck carcinoma? Cancer 2004;100:1171-1178. 23. Spanos WJ, Shukovsky LJ, Fletcher GH. Time, dose, and tumor volume relationships in irradiation of squamous cell carcinomas of the base of the tongue. Cancer 1976;37:2591-2599.

11  The Oropharynx 24. Foote RL, Parsons JT, Mendenhall WM, et al. Is interstitial implantation essential for the successful radiotherapeutic treatment of base of tongue carcinoma? Int J Radiat Oncol Biol Phys 1990;18:1293-1298. 25. Wang CC. Radiation Therapy for Head and Neck Neoplasms, 3rd ed. New York: Wiley-Liss, 1997, pp 187-203. 26. Weber RS, Gidley P, Morrison WH, et al. Treatment selection for carcinoma of the base of the tongue Am J Surg 1990;160:415-419. 27. Mak AC, Morrison WM, Garden AS, et al. Base-of-tongue carcinoma: treatment results using concomitant boost radiotherapy. Int J Radiat Oncol Biol Phys 1995;33:289-296. 28. Mendenhall WM, Stringer SP, Amdur RJ, et al. Is radiation therapy a preferred alternative to surgery for squamous cell carcinoma of the base of tongue? J Clin Oncol 2000;18:35-42. 29. Harrison LB, Lee HJ, Pfister DG, et al. Long-term results of primary radiotherapy with/without neck dissection for squamous cell cancer of the base of tongue. Head Neck 1998;20:668-673. 30. Gibbs I, Le Q, Shah R, et al. Long-term outcomes after external beam irradiation and brachytherapy boost for base-of-tongue cancers. Int J Radiat Oncol Biol Phys 2003;57:489-494. 31. Puthawala A, Syed AMN, Eads DL, et al. Limited external beam and interstitial 192 iridium irradiation in the treatment of carcinoma of the base of the tongue: a ten year experience. Int J Radiat Oncol Biol Phys 1988;14:839-848. 32. De Los Santos JF, Morrison WH, Garden AS, et al. Surgery and postoperative radiation therapy for advanced squamous cell carcinoma of the base of tongue [abstract 53]. Cancer J 2000;6:411. 33. Wong CS, Ang KK, Fletcher GH, et al. Definitive radiotherapy for squamous cell carcinoma of the tonsillar fossa. Int J Radiat Oncol Biol Phys 1989;16:657-662. 34. Mendenhall WM, Stringer SP, Amdur RJ, et al. Radiation therapy for squamous cell carcinoma of the tonsillar region: a preferred alternative to surgery? J Clin Oncol 2000;18:2219-2225. 35. Bataini JP, Asselain B, Jaulerry C, et al. A multivariate primary tumour control analysis in 465 patients treated by radical radiotherapy for cancer of the tonsillar region: clinical and treatment parameters as prognostic factors. Radiother Oncol 1989;14:265-277. 36. O’Sullivan B, Warde P, Grice B, et al. The benefits and pitfalls of ipsilateral radiotherapy in carcinoma of the tonsillar region. Int J Radiat Oncol Biol Phys 2001;51:332-343. 37. Jackson SM, Hay JH, Flores AD, et al. Cancer of the tonsil: the results of ipsilateral radiation treatment. Radiother Oncol 1999;51:123-128. 38. Pernot M, Malissard L, Taghian A, et al. Velotonsillar squamous cell carcinoma: 277 cases treated by combined external irradiation and brachytherapy—results according to extension, localization, and dose rate. Int J Radiat Oncol Biol Phys 1992;23:715-723. 39. Mazeron JJ, Lusinchi A, Marinello G, et al. Interstitial radiation therapy for squamous cell carcinoma of the tonsillar region: the Creteil experience (1971–1981). Int J Radiat Oncol Biol Phys 1986;12:895-900. 40. Mazeron JJ, Noël G, Simon J. Head and neck brachytherapy. Semin Radiat Oncol 2002;12:95-108. 41. Keus RB, Pontvert D, Brunin F, et al. Results of irradiation in squamous cell carcinoma of the soft palate and uvula. Radiother Oncol 1988;11:311-317. 42. Lindberg RD, Fletcher GH. The role of irradiation in the management of head and neck cancer: analysis of results and causes of failure. Tumori 1978;64:313-325. 43. Fein DA, Lee WR, Amos WR, et al. Oropharyngeal carcinoma treated with radiotherapy: a 30-year experience. Int J Radiat Oncol Biol Phys 1996;34:289-296. 44. Mazeron JJ, Belkacemi Y, Simon JM, et al. Place of iridium 192 implantation in definitive irradiation of faucial arch squamous cell carcinomas. Int J Radiat Oncol Biol Phys 1993;27:251-257.

249

45. Erkal H, Serin M, Amdur RJ, et al. Squamous cell carcinomas of the soft palate treated with radiation therapy alone or followed by planned neck dissection. Int J Radiat Oncol Biol Phys 2001;50:359-366. 46. Esche BA, Haie CM, Gerbaulet AP, et al. Interstitial and external radiotherapy in carcinoma of the soft palate and uvula. Int J Radiat Oncol Biol Phys 1988;15:619-625. 47. Hull MC, Morris CG, Tannehill SP, et al. Definitive radiotherapy alone or combined with a planned neck dissection for squamous cell carcinoma of the pharyngeal wall. Cancer 2003;98:2224-2231. 48. Lee HJ, Zelefsky MJ, Kraus DH, et al. Long-term regional control after radiation therapy and neck dissection for base of tongue carcinoma. Int J Radiat Oncol Biol Phys 1997;38:995-1000. 49. Porceddu S, Jarmolowski E, Hicks R, et al. Utility of positron emission tomography for the detection of disease in residual neck nodes after (chemo) radiotherapy in head and neck cancer. Head Neck Surg 2005;27:175-181. 50. Peters L, Weber R, Morrison W, et al. Neck surgery in patients with primary oropharyngeal cancer treated by radiotherapy. Head Neck 1996;18:552-559. 51. Cooper JS, Ang KK. Concomitant chemotherapy and radiation therapy certainly improves local control. Int J Radiat Oncol Biol Phys 2005;61:7-9. 52. Bernier J, Domenge C, Eschwege F, et al. Chemo-radiotherapy, as compared to radiotherapy alone, significantly increases ­disease-free and overall survival in head and neck cancer patients after surgery: results of EORTC phase III trial 22931 [abstract]. Int J Radiat Oncol Biol Phys 2001;(3 suppl 1):1. 53. Lindberg R, Fletcher G. The role of irradiation in the management of head and neck cancer: analysis of results and causes of failure. Tumori 1978;64:313-325. 54. Fu KK, Pajak TF, Trotti A, et al. A Radiation Therapy Oncology Group (RTOG) phase III randomized study to compare hyperfractionation and two variants of accelerated fractionation to standard fractionation radiotherapy for head and neck squamous cell carcinomas: first report of RTOG 9003. Int J Radiat Oncol Biol Phys 2000;48:7-16. 55. Al-Othman M, Morris C, Hinerman R, et al. Distant metastases after definitive radiotherapy for squamous cell carcinoma of the head and neck. Head Neck Surg 2003;25:629-633. 56. Salama J, Vokes E, Chmura S, et al. Long-term outcome of concurrent chemotherapy and reirradiation for recurrent and second primary head-and-neck squamous cell carcinoma. Int J Radiat Oncol Biol Phys 2006;64:382-391. 57. De Crevoisier R, Bourhis J, Domenge C, et al. Full-dose reirradiation for unresectable head and neck carcinoma: experience at the Gustave-Roussy Institute in a series of 169 patients. J Clin Oncol 1998;16:3556-3562. 58. Dabaja B, Salehpour M, Rosen I, et al. Intensity-modulated radiation therapy (IMRT) of cancers of the head and neck: comparison of split-field and whole-field techniques. Int J Radiat Oncol Biol Phys 2005;63:1000-1005. 59. Grégoire V, Levendag P, Ang K, et al. CT-based delineation of lymph node levels and related CTVs in the node-negative neck: DAHANCA, EORTC, GORTEC, NCIC, RTOG consensus guidelines. Radiother Oncol 2003;73:383-384. 60. Eisbruch A, Marsh L, Dawson L, et al. Recurrences near base of skull after IMRT for head-and-neck cancer: implications for target delineation in high neck and for parotid gland sparing. Int J Radiat Oncol Biol Phys 2004;59:28-42. 61. Bonner JA, Harari PM, Giralt J, et al. Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. N Engl J Med 2006;354:567-578.

12 The Oral Cavity JAY S. COOPER ACUTE EFFECTS OF IRRADIATION ON NORMAL TISSUES ETIOLOGY ANATOMY WORKUP AND DIAGNOSIS Clinical Presentation Physical Examination Imaging Studies Ruling Out Second Malignancies PATTERNS OF SPREAD Lymphatic Metastases Hematogenous Metastases STAGING PATHOLOGY TREATMENT PRINCIPLES T1 and T2 Lesions Small T3 Lesions Larger T3 and T4 Lesions High-Risk Resected Disease Lesions of the Lips Lesions of the Buccal Mucosa Lesions of the Oral Tongue Lesions of the Floor of the Mouth Lesions of the Upper or Lower Gingiva or Retromolar Triangle

Carcinomas of the oral cavity were detected in approximately 22,550 U.S. individuals in 2007 (Table 12-1), and they occur approximately twice as often in men as in women.1 These tumors account for approximately 3% of all neoplasms (excluding basal and squamous cell carcinomas of the skin) in the United States.1 Only about 60% of patients with carcinoma of the oral cavity survive for more than 5 years after diagnosis. Despite ­improvements in imaging and therapy, the survival rate for ­patients with these tumors has not changed substantially for many years. The residual effects of smoking and alcohol consumption are likely to keep the incidence rate of these tumors at their current levels for the ­foreseeable future.

Acute Effects of Irradiation on Normal Tissues Changes in the salivary glands are usually the first sign of radiation-related damage in the region of the oral cavity. Salivary flow decreases markedly after administration of a few grays (Gy), and many patients complain of dryness of the mouth after receiving a single fraction of radiation therapy. As radiation therapy continues, salivary flow continues to decrease, and the saliva takes on a stickier quality.

250

Lesions of the Hard Palate Regional Lymph Nodes Recurrent Disease Chemoprevention RADIATION THERAPY TECHNIQUES Brachytherapy Intraoral Cone Therapy Conventional External Beam Therapy Three-Dimensional Conformal or ­ Intensity-Modulated Radiation Therapy RESULTS OF RADIATION THERAPY Factors Influencing the Likelihood of Local Control of Disease Patterns of Disease Regression Site-Specific Results LATE EFFECTS AND COMPLICATIONS OF RADIATION THERAPY Xerostomia Fungal, Viral, and Bacterial Infections Mucositis and Pain Nutritional Problems Osteonecrosis and Soft Tissue Necrosis Dental Decay FOLLOW-UP CARE CONCLUSIONS

Food and debris adhere tenaciously to the teeth and gums. A minority of patients experience self-limited swelling of the submandibular or parotid glands during the first few days of treatment. Most patients experience a temporary increase in serum amylase levels as a result of the effects of radiation on the salivary glands.2 Irradiation of the oral cavity affects the taste buds and decreases the ability to distinguish the taste of foods. As a result, patients often choose sweeter foods, which stick to the teeth and foster decay. Because cellular turnover is more rapid in mucosa than in skin, the mucosa tends to show changes before reactions are apparent in the skin. After 15 to 20 Gy, delivered in conventional fractions of 2 Gy/day, the mucosa becomes erythematous. After an additional 10 Gy, small white areas of mucositis begin to appear. In their early stages, these patches tend to be scattered throughout the treatment field, generating the term patchy mucositis. As treatment continues, the patches begin to coalesce, potentially leading to confluent mucositis, in which few areas of normal intact mucosa remain. While the mucosa progresses through the erythema– patchy mucositis–confluent mucositis continuum, the skin begins to demonstrate the effects of irradiation. Initially, erythema appears, and in some patients, the skin becomes

12  The Oral Cavity

251

TABLE 12-1.    Estimated Incidence of and Deaths from Oral Cavity Cancer in the United States in 2007 Estimated Incidence Tumor Site

Total

Estimated Deaths

Male

Female

Total

Male

Female

Tongue

9800

6390

2870

1830

1180

650

Mouth

10660

6480

4180

1860

1110

750

Other

2100

1460

640

1680

1270

410

Data from American Cancer Society. Cancer Facts and Figures, 2007. Atlanta, GA: American Cancer Society, 2007.

TABLE 12-2.    Radiation Therapy Oncology Group Criteria for Acute Toxic Effects Toxic Effects Site

Grade 0

Grade 1

Grade 2

Grade 3

Grade 4

Skin

No change from baseline

Follicular, faint, or dull erythema Epilation Dry desquamation Decreased sweating

Tender or bright erythema, patchy moist desquamation Moderate edema

Confluent, moist desquamation other than skin folds, pitting edema

Ulceration, ­hemorrhage, or necrosis

Mucous ­membranes

No change from baseline

Redness May experience mild pain not requiring analgesics

Confluent fibrinous Patchy mucositis mucositis that may produce May include severe an inflammatory pain requiring serosanguineous narcotic discharge May experience moderate pain requiring ­analgesia

Salivary gland

No change from baseline

Moderate to comMild mouth dryplete dryness ness; slightly Thick, sticky saliva thickened saliva Markedly altered May have slightly taste altered taste such as metallic taste Changes not ­reflected in altera­ tion in baseline feeding behavior, such as increased use of liquids with meals

Ulceration, ­hemorrhage, or necrosis

(RTOG classification Acute salivary gland necrosis does not describe grade 3 salivarygland effects.)

Modified from Cox JD, Stetz J, Pajak TF. Toxicity criteria of the Radiation Therapy Oncology Group (RTOG) and the European Organization for Research and Treatment of Cancer (EORTC). Int J Radiat Oncol Biol Phys 1995;31:1341-1346.

hyperpigmented. Many patients subsequently experience desquamation of the skin in which coarse flakes of skin are shed but the skin beneath remains intact (i.e., dry desquamation). Some patients have a more severe reaction in which the shedding of the overlying skin leaves a weeping, eroded site (i.e., moist desquamation). In some patients, moist desquamation is promoted by friction, such as that from a shirt collar against the skin of the neck. To facilitate comparisons among studies of the degree and type of side effects during treatment, the Radiation Therapy Oncology Group (RTOG) created a scale to describe acute toxic effects, defined as those experienced within 90 days of treatment (Table 12-2).3

Etiology Most tumors in the oral cavity are epithelially derived squamous cell carcinomas. The entire oral cavity can be visually inspected and palpated, and screening for tumors is simple and inexpensive because many invasive tumors are preceded by dysplastic or in situ lesions that are white (i.e., leukoplakia) or red (i.e., erythroplakia). However, comprehensive screening programs for oral cancer have not garnered the level of attention afforded screening programs for other types of cancer (e.g., prostate-specific antigen measurement for prostate cancer, mammography for breast cancer). Oral cancer screening gets so little attention that

252

PART 3  Head and Neck

Gingiva

Posterior pharynx

Hard palate

Retromolar trigone

A

Frenulum

Base of tongue

Border

Anterior pillar Tonsil

Tip

Root of tongue

Floor of mouth

Oral tongue

Buccal mucosa Uvula

Undersurface of tongue

Soft palate

Posterior pillar

Gingiva

Lower lip

B

FIGURE 12-1.  A, Front (open-mouth) view of the oral cavity. B, Front view of the floor of the mouth with the tongue raised.

Nose Soft palate Uvula

trigo

Upper lip

Nasopharynx palate

ne

Hard

olar

Buccal mucosa

rom

Anterior pillar Tonsillar fossa Posterior pillar Otopharynx

Ret

Gum

Vallecula

oph

Larynx

Hyp

Circumvallate papillae Tongue base

x aryn

igl

ott

Oral tongue

is

Lower lip

Ep

a recent Cochrane review could find “no evidence to recommend inclusion or exclusion of screening programs for oral cancer.”4 Perhaps as a consequence, many cases of oral cancer are not detected until they are advanced. The risk of developing an oral cancer is higher among men than women, although the male-to-female incidence ratio has decreased from approximately 6:1 several decades ago to about 2:1 now. The typical patient is a man in his 50s or 60s who smokes tobacco heavily or consumes large amounts of alcohol. The amount smoked correlates directly with the likelihood of dying of oral cancer, and cigarette, ­cigar, and pipe smoking seem to be equally dangerous.5 Smoking more than 20 cigarettes daily correlates with a sixfold increase in the risk of developing intraoral cancer compared with the risk in nonsmokers.6 Drinking more than 6 ounces of hard liquor daily increases the risk of intraoral cancer tenfold.7 Other kinds of smoking and alcohol consumption also seem to predispose to oral cancer; habitual use of ­marijuana8 and habitual use of alcoholcontaining mouthwashes9,10 have been implicated as risk factors for oral cavity tumors. Moreover, the deleterious effects of smoking and alcohol consumption may be more than additive.11 Several risk factors are associated with tumors in specific oral cavity subsites. Chewing a mixture of betel nut, tobacco, and slaked lime12; the use of smokeless chewing tobacco13; and snuff dipping13,14 are associated with ­carcinomas of the buccal mucosa, gingiva, lip, and floor of the mouth. Smoking cigars in a reversed position (lit end inside the mouth) is associated with tumors of the hard palate.15,16 Poor dental hygiene may contribute to the induction of intraoral tumors6; carcinomas of the oral tongue often arise directly adjacent to decayed or broken teeth. Carcinomas of the lip in large part result from sunlight exposure.17 Human papillomavirus (particularly HPV-16 and HPV-18) infection has been associated with oral squamous cell carcinoma; the recent development of a vaccine against these strains designed to prevent cervical cancer may soon provide another means of reducing the incidence of these tumors in the future.18,19 Host factors also seem to be important. Schantz and colleagues20 demonstrated that lymphocytes from young

FIGURE 12-2.  Lateral view of the oral cavity.

adults with squamous cell carcinoma have greater chromosomal fragility on exposure to bleomycin than do lymphocytes from age- and sex-matched controls. For unknown reasons, a previously unseen number of young patients (<40 years old) seem to be developing oral cancer, especially of the oral tongue.21 The epidemiology of oral cavity cancer is changing, as described in a review by Gillison.22 However, after an oral squamous cell carcinoma develops, the virulence of the disease is not related to the patient’s smoking or drinking history.23

Anatomy The oral cavity (Figs. 12-1 and 12-2) is composed of the mucosal portion of the lips, the buccal mucosa, the upper and lower gingivae, the floor of the mouth, the retromolar trigones, the hard palate, and the anterior two thirds of the tongue, up to the circumvallate papillae. Beneath the mucosa, the relationships among the muscles, nerves, and mandible are complex. At the insertion of the genioglossus and geniohyoid muscles on the ­mandible, bony

12  The Oral Cavity

253

TABLE 12-3.    Typical Workup for Oral Cancer Glossopharyngeal nerve Base of skull foramen Lingual nerve

Tympanic membrane

Petrosal ganglion Tympanic membrane Base of tongue

Epiglottis

Corda tympani Ganglion

Tonsillar fossa

Procedure

Purpose

Inspection

Evaluate extent of disease

Palpation

Evaluate extent of disease

Direct or indirect examination of entire head and neck mucosa

Evaluate extent of disease; rule out second primary tumor

Biopsy

Confirm diagnosis and ­histologic type

MRI or CT of oral cavity (multiplanar)

Evaluate extent of disease; rule out mandibular involvement

Dental evaluation with or without dental therapy

Reduce toxicity of treatment

CT of chest

Rule out metastases and ­second primary tumor

Esophagography or ­esophagoscopy

Rule out second primary tumor

CT, computed tomography; MRI, magnetic resonance imaging.

FIGURE 12-3.  Diagram of the nerve pathways that account for referred pain in the region of the ear from primary neoplasms of the oral cavity.

projections (i.e., genial tubercles) encroach on the floor of the mouth. The degree of encroachment varies from patient to patient, and in some patients, the tubercles are sufficiently prominent to impede brachytherapy. The submandibular ducts (i.e., Wharton ducts) course from the submandibular glands anteriorly and superiorly for approximately 5 cm to enter the floor of the mouth anteriorly near the midline. The lumen of these ducts can be obstructed by tumor, resulting in enlargement of the submandibular glands that clinically simulates tumor infiltration. The lingual nerve (i.e., cranial nerve V), which provides sensory fibers to the oral tongue, traverses the foramen ovale to meet the auriculotemporal nerve in the gasserian ganglion. Because the auriculotemporal nerve supplies sensory fibers to the helix and tragus of the ear as well as the anterior wall of the external auditory canal, lesions of the oral cavity can produce referred pain that is attributed to the ear (Fig. 12-3). Anatomic barriers induce tumors to grow locally in relatively predictable patterns. Tumors presumably take the paths of least resistance on the mucosal surfaces from which they arise and in the submucosal tissues that they invade. Spread of disease through the loose fibrofatty tissue planes that envelop muscles from their origin to insertion can often be detected with computed tomography (CT) or magnetic resonance imaging (MRI), even when such spread is not evident on physical examination. Tumors tend to ­ extend along neurovascular bundles, along peri­osteal surfaces, and rarely, if tumors penetrate the cortex of the mandible, through the marrow space.

Workup and Diagnosis Table 12-3 summarizes the typical workup for oral cancer and the rationale for each element of the workup.

TABLE 12-4.    C  ommon Signs and Symptoms of Oral Cancer Early Disease

Advanced Disease

Leukoplakia

Pain

Erythroplakia

Level I, II, or III adenopathy

Asymptomatic mass

Slurred speech

Persistent sore or ulceration

Difficulty chewing or ­swallowing

Bleeding mass

Clinical Presentation Because tumors in the oral cavity usually arise from the mucosal surfaces, they often are first detectable as mucosal irregularities. Common signs and symptoms of oral cavity tumors are listed in Table 12-4. Typically, tumors in the oral cavity initially appear as slightly discolored regions (white or red). As lesions progress, they can become thick and indurated. Eventually, the mucosa can become eroded or ulcerated, the growing edges of the visible lesion become more apparent, and the lesion may take on a whiter or a redder color. Some lesions develop deep central ulceration, whereas others take on a multilobulated, cauliflower‑like appearance. Bleeding may occur. Tumors of the floor of the mouth often appear as slitlike crevasses and are best appreciated by palpation of the involved area. When an oral tumor is suspected but is not obvious or when the borders of an oral tumor are indistinct, vital staining with tolonium chloride (toluidine blue) may help demarcate the lesion.24 Newer methods such as autofluorescence after excitation with a laser or xenon light, chemiluminescence, and oral brush biopsy, ­computer-scanned cytology may become more mainstream in the future.

254

PART 3  Head and Neck

Physical Examination All subregions of the oral cavity are accessible for digital examination in patients who are able to cooperate for examination. As tumors grow, the difference in their texture compared with the texture of adjacent normal tissues becomes more apparent on palpation. For lesions of the lip, floor of the mouth, or buccal mucosa, bimanual examination should be routine; a gloved finger is inserted in the mouth, and the tumor is gently pressed against the examiner’s other hand, which is held outside the mouth. In the case of tumors that have invaded into deeper tissues, infiltration into nerves can cause local pain, dysesthesias, paresthesias, and referred pain in the ipsilateral ear. More advanced lesions can cause loosening of the teeth, dysarthria, dysphagia, and trismus.

Imaging Studies For all oral cavity tumors except clearly superficial small lesions, CT and MRI have become standard elements of the diagnostic workup. Images of the primary tumor site, likely routes of spread, and regional nodal drainage pathways can demonstrate extensions of disease that are otherwise clinically imperceptible. In the era of cost-conscious medicine, physicians often must carefully consider whether to order CT or MRI, or both. Although CT and MRI can be viewed as alternative or even competitive technologies, they are complementary and supplement the physical examination for mapping and staging tumors. (Mapping refers to the spatial localization of all aspects of a tumor, whereas staging refers to the application of rules that group lesions of similar prognosis.) Both techniques have advantages (Table 12-5). Within the oral cavity, MRI usually is better than CT for delineating the size of primary tumors.25 Over the past few years, 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) has become more widely available and more commonly considered in the workup of squamous cell carcinomas arising in the oral cavity. However, the precise role of PET scanning for tumors of the oral cavity remains controversial. Evidence exists to support the accuracy, sensitivity, and specificity of PET; however, various reports differ somewhat in their interpretations. In one such report, Nakasone and colleagues26 studied 25 consecutive patients with oral cancer to assess the clinical value of FDG-PET for tumor staging. All primary tumors were said to be “accurately visualized” on FDGPET images, even in circumstances in which artifacts from nearby dental materials impaired the delineation of primary tumors on CT or MRI.26 The uptake of FDG in tumor metastatic to regional lymph nodes was also significantly greater than in uninvolved lymph nodes. The investigators conclude that “FDG-PET is an effective and convenient diagnostic tool for the evaluation of tumor staging in patients with oral cancer.” Goerres and others27 retrospectively evaluated the additional information obtained from PET scans from 34 consecutive patients with histologically confirmed squamous cell carcinomas of the oral cavity. Scans from four patients (12%) revealed a second independent malignancy, and scans from three other patients (6%) revealed distant metastases.27 The information obtained by PET scanning led

TABLE 12-5.    R  elative Advantages of CT and MRI for Mapping Tumors of the Oral Cavity Advantages of CT

Advantages of MRI

Metastatic nodal disease is more readily assessed.

Direct sagittal, coronal, and axial images are obtained, facilitating mapping.

Integrity of cortical bone (mandible) is more ­accurately assessed.

Intravenous contrast agents (e.g., gadolinium-DTPA) are safer than the iodinated agents used for CT.

Examination time is shorter.

Images of the oral cavity are less likely to be degraded by artifacts.

Cost is often less.

Perineural spread of tumors is more accurately assessed.

Suitable for patients who have pacemakers or ­aneurysm clips

Integrity of medullary bone (mandible) is more accurately assessed.

CT, Computed tomography; DTPA, diethylenetriaminepentaacetic acid; MRI, magnetic resonance imaging.

to a change in treatment for five patients (15%); however, one false-positive area of uptake was also seen. In a third study, Dammann and colleagues28 compared CT, MRI, and FDG-PET scans from 64 patients with oral cavity or oropharyngeal tumors that were later resected within 2 weeks after imaging. In terms of detecting or delineating the primary tumor, CT was inferior to MRI or PET; MRI and PET were relatively similar, except that MRI was more sensitive than PET (92% versus 87%).28 The authors of this report concluded that PET can be helpful when MRI or CT findings are equivocal but is not necessary for routine use. Ng and coworkers29 prospectively compared the ability of FDG-PET and CT or MRI to detect primary tumors and cervical lymph node metastases among 124 patients with oral cavity squamous cell carcinoma. The group concluded that FDG-PET is superior to CT or MRI for the detection of cervical lymph node metastases because of its greater sensitivity, although the specificity of the three techniques seemed similar.29 In contrast, Stoeckli and others,30 in their comparison of FDG-PET with sentinel lymph node biopsy and elective neck dissection in 12 patients with oral or oropharyngeal squamous cell carcinoma and no clinical evidence of lymph node metastases, concluded, “PET with FDG … [has] … poor sensitivity and specificity in revealing occult metastasis and has no role [in] the evaluation of otherwise clinically N0 necks.” Goerres and colleagues31 found that conventional contrast-enhanced CT scanning was as good as FDG-based PET/CT scanning for assessing cortical bone erosion. However, Kunkel and others32 found PET scanning to be “highly valuable” for detecting recurrent disease among 97 patients with previously resected oral squamous cell carcinoma. They further observed a significant association between greater FDG standardized uptake values and poorer 3-year survival rates. PET is still a relatively young technology that may complement more mature imaging techniques such as CT and MRI in some circumstances.

12  The Oral Cavity

FIGURE 12-4.  Axial CT image at the level of the mandible shows a destructive lesion predominantly involving the lingual cortex.

255

A

B

FIGURE 12-5.  Axial contrast-enhanced CT shows a ­pathologicappearing enhancing mass involving the oral and oropharyngeal portions of the right side of the tongue (asterisk). Notice the obliteration of the normal fatty lingual septum (arrowheads), indicating extension across the midline.

Evaluation of Tumor Extent There is no reason to use imaging to confirm a strictly mucosal abnormality that is obvious on visual examination. However, when tumors are sufficiently thick and invade below the mucosa, CT and MRI are useful because they provide detailed pictures of the submucosal tissues. This improved view of tumor extent can be a critical factor in determining the most appropriate treatment for a particular patient. For example, mandibular invasion (Fig. 12-4) from lesions of the floor of the mouth usually implies the need to resect the affected bone. In contrast, posterior lesions that cross the midline lingual septum (Fig. 12-5) typically cannot be resected with partial glossectomy, and CT or MRI evidence of such spread argues in favor of radiation therapy. Sometimes, the information obtained by imaging studies is unexpected and cannot be discerned by other means. For example, some lesions of the retromolar triangle invade the pterygomandibular raphe, a path of little resistance that courses from the mandible, just deep to the retromolar trigone, up to the pterygoids at the base of the skull. These lesions can extend to the base of the skull through the raphe in a manner that is radiographically obvious but undetectable by other means (Fig. 12-6). Knowledge of such spread

FIGURE 12-6.  A, Contrast-enhanced CT (same patient as in Fig. 12-4) shows an enhancing mass in the posterior left side of the floor of the mouth extending to the retromolar trigone (arrowheads). B, Contrast-enhanced CT (same patient as in Fig. 12-6A) at the level of the maxilla shows submucosal infiltration along the pterygomandibular raphe to the posterior aspect of the normally fat-filled buccinator space (asterisk). This clinically occult disease was identified only by imaging.

can prevent ill-advised surgery and influence the size and placement of radiation therapy portals.

Evaluation of Regional Adenopathy Imaging can also reveal regional adenopathy that cannot be detected by physical examination. Sectional imaging in particular can supplement the physical examination by ­ revealing the location of nonpalpable involved lymph nodes. A nodal classification system published in 199133 and updated in 200234 was designed to standardize the description of neck dissections, but it also applies to imaging.35 The system divides the cervical lymph nodes into six levels. Level I includes the submental (level IA) and submandibular (level IB) lymph nodes. Levels II, III, and IV correspond to the upper (jugulodigastric), middle (jugulo‑omohyoid), and lower (jugulocarotid) nodes, respectively, of the anterior deep jugular cervical chain. Level II is subdivided into the space anteromedial to the spinal accessory nerve (level IIA) and the space posterolateral to the spinal accessory nerve (level IIB). Level V designates nodes in the posterior triangle of the neck. Level VI refers to the anterior compartment and contains the precricoid, peritracheal, paratracheal, and perithyroidal nodes; level VI nodes are virtually never invaded by cancers that arise in the oral cavity.

256

PART 3  Head and Neck

FIGURE 12-9.  Axial contrast-enhanced CT at the level of the cricoid cartilage shows a single pathologic-appearing lymph node (asterisk) immediately posterior to the right internal jugular vein. This lymph node is at the boundary between levels III and IV. The left internal jugular vein (curved arrow) is markedly larger than the right. This is a normal anatomic variant. FIGURE 12-7.  Axial contrast-enhanced CT shows a single enlarged level I lymph node (asterisk) anterior to the left submandibular gland.

FIGURE 12-8.  Axial contrast-enhanced CT at the level of the hyoid bone shows a single pathologic-appearing lymph node (asterisk) lateral to the left common carotid artery. This lymph node is at the boundary between levels II and III.

Level I nodes are inferior and lateral to the ­mylohyoid muscle in proximity to the submandibular glands (Fig. 12-7). Nodes in levels II, III, and IV are located near the internal jugular vein, typically anterior or deep to the anterior aspect of the sternocleidomastoid muscle. Technically, levels II, III, and IV are demarcated by the level of the bifurcation of the carotid artery and the level of the intersection of the internal jugular vein and the ­omohyoid muscle. ­ Because the carotid bifurcation and omohyoid muscle can be difficult to find on axial images, the hyoid bone and cricoid cartilage, which lie at approximately the same levels as the carotid bifurcation and the omohyoid muscle, respectively, can be used as boundaries between levels II and III and levels III and IV (Figs. 12-8 and 12-9). Level V nodes are located predominantly posterior to the sternocleidomastoid muscle (Fig. 12-10). In general, uninvolved lymph nodes are 1 cm or less in diameter. It is prudent to consider any node longer than 1.5 cm along its long axis or 1.0 cm along its short axis to be abnormal.36,37 (Nodes smaller than 1.5 cm may well be invaded by tumor, but these nodes cannot be distinguished on the basis of size.) However, some nodes larger than 1.5 cm are enlarged

FIGURE 12-10.  Axial contrast-enhanced CT shows an abnormal mass deep to the sternocleidomastoid muscle. This is an example of a level V (spinal accessory chain, posterior triangle) lymph node.

because of hyperplasia rather than tumor. A better indicator of tumor is therefore a hypodense nodal center, commonly called a necrotic center even though histopathologic examination of such nodes often does not disclose necrosis of the cells. The appearance of central necrosis is sufficiently diagnostic of neoplastic invasion that nodes of any size with central lucency must be considered neoplastic (Fig. 12-11). Imaging can be used to assess the integrity of the nodal capsule. When lymph nodes are invaded by tumor, the fascial planes surrounding the node remain intact unless tumor ruptures the capsule. A node with central lucency and an indistinct periphery should therefore be interpreted as a tumor‑bearing node with extracapsular extension of disease (see Fig. 12-11). The presence of central necrosis and evidence of ­extracapsular extension of disease are readily detected

12  The Oral Cavity

FIGURE 12-11.  Axial contrast-enhanced CT at the level of the hyoid bone shows a 5-cm necrotic internal jugular chain lymph node (asterisk) with an extracapsular extension and a fistulous tract (curved arrow) extending to the skin. Notice the additional 1-cm necrotic lymph nodes posterior to the largest nodal mass (open arrow) and in the submental triangle (short solid arrow)

257

FIGURE 12-12.  The lymphatics of the upper lip drain to the buccal, parotid, upper cervical, and submandibular nodes. The lymphatics of the skin of the upper lip (dotted line) may cross the midline to terminate in submental and submandibular nodes on the contralateral side.

on ­ contrast-­enhanced CT. Contrast-enhanced MRI may also be a ­ reasonable alternative to CT for the evaluation of metastasis to level I and II nodes if appropriate pulse ­sequences are used.25

Ruling Out Second Malignancies The diagnostic workup for patients with oral cavity ­tumors should include a search for coincident second malignancies of the upper aerodigestive tract. Bronchoscopy, ­esophagoscopy, and imaging are useful for this purpose. In approximately 10% of cases, depending on the thoroughness of the search, a synchronous second malignancy is ­detected.38

Patterns of Spread Lymphatic Metastases Spread of disease by way of regional lymphatics is relatively common for carcinomas of the oral cavity. To a considerable degree, the likelihood of lymphatic metastases can be predicted on the basis of the primary tumor’s size, histologic grade, and location and the presence or absence of perineural and vascular invasion.39 The lymph drainage of the oral cavity filters predominantly through the submandibular and upper anterior cervical (jugulodigastric) nodes. However, some anatomic subsites have unique drainage patterns. The lymphatics of the upper lip drain into the submandibular and preauricular nodal beds (Fig. 12-12), and the lymphatics of the midlip and lower lip and the anterior floor of the mouth drain into the submental nodal group (Fig. 12-13). The lymphatics of the oral tongue drain into the anterior cervical chain; lesions located more anteriorly drain lower in the neck than lesions located more posteriorly (Fig. 12-14). Crossing lymphatics can lead to contralateral nodal metastases (Fig. 12-15). The consistency of nodal drainage from oral cavity carcinomas means that previously hidden primary tumors are sometimes discovered because of unusual patterns of nodal disease. The incidence and the location of lymph node metastases vary only slightly among the different tumor subsites

FIGURE 12-13.  The lymphatics of the lower lip drain to the submental and submandibular nodes. Sometimes disease involves facial nodes. The lymphatics of the skin of the lower lip (dotted line) may cross the midline to end in submental or submandibular nodes on the contralateral side.

in the oral cavity.40 Overall, the risk of clinically apparent lymph node involvement is approximately 35%. The ipsilateral ­upper jugular nodes (level II), submandibular nodes (level I), and midjugular nodes (level III) account for most sites of spread (Fig. 12-16). Contralateral spread is much less likely, but when it occurs, it involves the same nodal levels. The relative predictability of drainage pathways permits physicians to design treatment plans that rationally target clinically apparent and clinically undetectable (subclinical) disease. When regional lymph nodes are enlarged by tumor, they may be the deciding factor in determining the ­manner of treatment. The risk of subclinical nodal disease also must

258

PART 3  Head and Neck 40%

OT-I OT-C FOM-I FOM-C RMT-I RMT-C

30% 20% 10%

Supraclavicular

Posterior cervical

Low jugular

Mid jugular

Upper jugular

Submandibular

Submental

Preauricular

0%

FIGURE 12-16.  Incidence of ipsilateral and contralateral  adenopathy at presentation is plotted according to site of primary disease and nodal level. C, contralateral adenopathy; I, ipsilateral adenopathy; FOM, floor of mouth; OT, oral tongue; RMT, retromolar trigone. (From Lindberg RD. Distribution of ­ cervical lymph node metastases from squamous cell carcinoma of the upper respiratory and digestive tracts. Cancer 1972;29:1446-1450.) FIGURE 12-14.  The more anteriorly the lymphatics originate in the tongue, the lower in the neck their draining nodes may lie.

TABLE 12-6.    L  ikelihood of Subclinical Nodal Metastases from Oral Cavity Tumors by Nodal Group Nodal Group

Rate of Involvement (%)

Level I nodes

20

Level II nodes

17

Level III nodes

9

Level IV nodes

3

Level V nodes

1

Any regional nodes

34

Modified from Shah JP. Patterns of cervical lymph node metastasis from squamous carcinomas of the upper aerodigestive tract. Am J Surg 1990;160:405-409.

FIGURE 12-15.  Frontal section of the lymphatics of the tongue show ipsilateral and contralateral drainage in the submandibular and cervical lymph nodes.

be taken into account in decisions about treatment. As a general rule, elective nodal irradiation is justified when the risk of involvement by subclinical disease exceeds 10% (i.e., when the likelihood that treatment will benefit the patient exceeds the risk that the treatment will cause serious permanent damage to the patient). In most series, the risk of subclinical disease in the regional lymph nodes is approximately 30% overall.41 August and Gianetti42 found elective nodal irradiation to be beneficial for a group of 73 patients with surgically treated oral cancer. In that study, the recurrence rate was 3% among patients who underwent elective nodal ­ irradiation after surgery and 31% among patients who were only observed after surgery. The 5-year survival rates were 90% for the group that received elective nodal irradiation and 40% for the group that was observed. In a series of patients treated with radical neck dissection,

Shah found that levels I and II were the most likely to be involved but that all nodal levels had some degree of risk of involvement (Table 12-6).43 These findings should be interpreted with caution, because they reflect selection bias (no mention was made of why the “elective” dissections were done). The findings may also be out of date because newer imaging techniques may detect previously undetectable disease, rendering therapeutic a surgical procedure that would previously have been considered elective. Friedman and colleagues44 suggested that the rate of occult disease varies substantially by the method of evaluation. On the basis of physical examination only, the likelihood of occult disease in their patients was 31%, but when CT or MRI was added, the rate dropped to 12%. Because the size of the primary tumor strongly correlates with the incidence of clinically detectable nodal disease at presentation, it is reasonable to assume that a similar relationship exists between the size of the primary tumor and the likelihood of subclinical nodal disease at presentation. Clinical judgment must be exercised in decisions regarding elective irradiation of clinically unremarkable nodal drainage sites.

12  The Oral Cavity

Hematogenous Metastases Distant metastasis occurs relatively late in the development of carcinomas of the oral cavity. The risk of distant disease is best predicted by the degree of lymph node involvement. ­Patients who have no clinically appreciable adenopathy virtually never develop distant metastases as their first site of treatment failure. In contrast, for patients who present with N2 or N3 disease, the incidence of distant metastasis approaches the incidence of locoregional recurrence. In one such series, Merino and colleagues45 reviewed the records of 5019 patients with head and neck tumors who had no clinical ­evidence of hematogenous metastasis at presentation. Distant metastases subsequently occurred in 10% of these patients. The lungs and bones were the most commonly affected sites. The risk of distant metastasis seems to be substantially greater when the locoregional manifestations of tumor are not controlled.46 Somewhat paradoxically, as locoregional therapies become more effective and patients live longer, the rate of distant metastasis is likely to rise to 15% to 20% as more patients survive long enough to manifest distant disease.

Staging Staging of head and neck cancer is based on information obtained by clinical examination and imaging studies. The American Joint Committee on Cancer (AJCC) staging ­system47 is the standard classification for carcinomas of the oral cavity. In the AJCC system, disease stage is determined on the basis of the size of the primary tumor and whether it invades adjacent structures; the number, size, and location (unilateral or bilateral) of involved lymph nodes; and the presence or absence of distant metastases. The staging system published in 2002 is shown in Box 12-1.

Pathology Within the oral cavity, a spectrum of progression from seemingly normal cells to dysplastic regions to carcinoma in situ and to frankly invasive cancer can be observed, suggesting that oral cavity tumors are the result of a multistep process, including multiclonal field carcinogenesis, that ­occurs over many years. Preexistent or coincident leukoplakia (a clinical descriptor implying only that the tissues are white and cannot easily be scraped off the mucosa; histologically, leukoplakia is characterized by hyperplasia of keratinocytes that are associated with normal or neoplastic adjacent cells) can be seen adjacent to frank oral cavity tumors in approximately 20% of cases.48 The presence of ­erythroplakia (red macular changes analogous to leukoplakia) even more strongly suggests invasive tumor.49 Oral ­lichen planus, particularly the erosive variant, is associated with the subsequent appearance of frank malignancy in about 2% of cases50,51; however, it remains unclear whether this represents true transformation, coincidence, or two manifestations of changes in susceptible epithelium. Dysplasia is primarily defined by cytologic atypia such as cellular pleomorphism, increased nuclear size, hyperchromatism, dyskeratosis, and abnormal-appearing mitotic figures. When such changes are confined to the basal or parabasal epithelium, dysplasia is considered mild; involvement of the midspinous layer of the epithelium renders the dysplasia moderate. If such changes are evident in the

259

surface epithelium, the dysplasia is called severe, and if the full thickness of the epithelium is involved, the lesion is classified as carcinoma in situ. Whether such changes precede the development of all cases of oral cancer is unknown. However, certainly some forms of invasive oral cancer can arise without a clinically detectable premalignant ­precursor. The molecular events that correlate with premalignant and malignant changes in the oral cavity are becoming known. Biomarkers of genomic instability such as aneuploidy and allelic imbalance correlate with the risk of progression of premalignant lesions or intraepithelial neoplasia to invasive cancer.52 These molecular changes may provide targets for antitumor therapy or aid in individualizing treatment strategies.53 Alterations in the tumor suppressor gene TP53 (formerly designated p53) and mutations of chromosome 9 (9p21-22) are often associated with oral cancer.54-56 Nuclear expression of TP53 and cyclin D1 increases along the spectrum from normal epithelium to low-grade oral leukoplakia (i.e., mild to moderate dysplasia) to high-grade leukoplakia (i.e., severe dysplasia).57 As cells progress to frank carcinoma, a further increase in TP53 expression is ­apparent, often associated with a decrease in expression of the cyclin-dependent kinase inhibitors CDKN1A (formerly designated p21, Cip1, or WAF1) and CDKN1B (formerly designated p27 or Kip1).58-60 Overexpression of TP53 is detectable in approximately 40% to 50% of oral squamous cell carcinomas and in a substantial percentage of histologically normal-­appearing cells adjacent to these tumors.61 The S-phase protein cyclin A and the M-phase protein cyclin B1 are expressed at higher levels in squamous cell carcinomas of the floor of the mouth than in normal epithelium from the same site.62 The percentage of cells with detectable cyclin D1 is smaller in samples taken from the basal and suprabasal epithelium of oral malignancies than in samples of nontumorous basal and suprabasal epithelium.63 Expression of integrin subunits α2 and β1 and the nuclear retinoic acid receptor RAR-β also seems to be lower in oral squamous cell carcinomas than in normal oral epithelium.64,65 Epidermal growth factor receptor expression has been reported to be amplified in oral cancer,66 ­elevated in poorly differentiated tumors,67 associated with larger ­ primary tumors,68 and related to tumor progression69; downregulation of its membrane-associated form has been correlated with tumor recurrence.70 Oral cancer seems to be fostered by the effects of carcinogens such as tobacco smoke on normal epithelium, including genomic instability (e.g., 3p or 9p deletions early in the process; promoter methylation; cyclin D1 amplification; loss of heterozygosity at 4q, 8p, 11q, 13q, 14q, or 17p followed by TP53 inactivation; 13q or 18q deletion) in parallel with enhancement of proliferation signals (e.g., activation of cyclooxygenase-2, epidermal growth factor receptor, or cyclin D1), which leads to increased cell growth with the inherent risk of more mutations.71 Genetic changes related to angiogenesis72 and tumor invasion (related to dysfunctional intercellular adhesion molecules or the upregulation of certain integrins such as αvβ6)73,74 can also contribute to tumor progression. Other agents such as viral promoters may have a role in the development of oral cancer. HPV, for example, was detected in approximately one half of a series of 102 oral

260

PART 3  Head and Neck

BOX 12-1.  American Joint Committee on Cancer Staging System for Tumors of the Lip and Oral Cavity Primary Tumor (T) TX Primary tumor cannot be assessed T0 No evidence of primary tumor Tis Carcinoma in situ T1 Tumor 2 cm or less in the greatest dimension T2 Tumor more than 2 cm but not more than 4 cm in the greatest dimension T3 Tumor more than 4 cm in the greatest dimension T4 (lip) Tumor invades through cortical bone, inferior alveolar nerve, floor of mouth, or skin of face, i.e., chin or nose T4a Tumor invades adjacent structures (e.g., through cortical bone, into deep [extrinsic] muscle of tongue [genioglossus, hyoglossus, palatoglossus, and styloglossus], maxillary sinus, skin of face) T4b Tumor invades masticator space, pterygoid plates, or skull base and/or encases internal carotid artery Note: Superficial erosion alone of bone/tooth socket by gingival primary is not sufficient to classify as T4) Regional Lymph Nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Metastasis in a single ipsilateral lymph node, 3 cm or less in the greatest dimension N2 Metastasis in a single ipsilateral lymph node, more than 3 cm but not more than 6 cm in the greatest dimension; or in multiple ipsilateral lymph nodes, none more than 6 cm in the greatest dimension; or in bilateral or contralateral lymph nodes, none more than 6 cm in the greatest dimension N2a Metastasis in single ipsilateral lymph node more than 3 cm but not more than 6 cm in the greatest dimension N2b Metastasis in multiple ipsilateral lymph nodes, none more than 6 cm in the greatest dimension N2c Metastasis in bilateral or contralateral lymph nodes, none more than 6 cm in the greatest dimension N3 Metastasis in a lymph node more than 6 cm in the greatest dimension Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage Grouping Stage 0 Stage I Stage II Stage III Stage IVA Stage IVB Stage IVC

Tis T1 T2 T3 T1–T3 T4a T1-T3 Any T T4b Any T

N0 N0 N0 N0 N1 N0–N2 N2 N3 Any N Any N

M0 M0 M0 M0 M0 M0 M0 M0 M0 M1

From Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 23-32.

lesions (21 cases of oral hyperplasia and 81 cases of oral squamous cell carcinoma), raising the possibility that HPV is a cocarcinogen.75 HPV-16 and HPV-18 are the most common subtypes associated with oral cancer.75,76 Once formed, more than 90% of invasive tumors arising from structures within the oral cavity are squamous cell carcinomas (Figs. 12-17 and 12-18). Grossly, lesions range from exophytic masses (at any site) to deeply ulcerated, slitlike crevasses (typically in the floor of the mouth). Generally, tumors of the oral cavity are moderately to well differentiated. However, the degree of differentiation tends not to be a major independent determinant of a tumor’s biologic behavior. Vascular invasion portends a poorer outcome.77 Salivary gland tumors (typically mucoepidermoid carcinomas) arising from minor salivary glands located in the hard palate, tongue, or retromolar region account for a small percentage of oral cavity tumors.78 (Salivary gland malignancies are discussed further in Chapter 8.) The significance of salivary gland tumors lies in their penchant for perineural spread (particularly adenoid cystic carcinomas), often to relatively great distances from their site of origin.

The hard palate is unique among the sites in the oral cavity because it is the only site in which the predominant histopathologic tumor type is of salivary origin. Squamous cell carcinomas of the hard palate are sufficiently uncommon that when this diagnosis is being considered, special care should be taken to rule out the more common situation of a gingival tumor that has extended or a maxillary sinus tumor that has penetrated the floor of the antrum to manifest on the palate. Other malignancies rarely encountered in the oral cavity include those arising from soft tissue, bone, cartilage, melanocytes, and odontogenic structures.

Treatment Principles Carcinomas of the oral cavity can be treated with surgery, radiation therapy, or a combination of both. For early-stage lesions, either modality produces a high likelihood of cure, whereas for advanced-stage lesions, neither modality by itself is sufficient. Chemotherapy may also have a role in the treatment of advanced disease. Surgery is more essential

12  The Oral Cavity

A

B FIGURE 12-17.  A well-differentiated squamous cell carcinoma is seen on low power (A) and high power (B). (Photographs courtesy J. Jagadar, MD.)

for the curative management of oral cavity disease than for disease at other head and neck sites. When tumors of the oral cavity can be resected with acceptable functional loss, surgery (at least as part of therapy) is the preferred form of treatment. Unlike tumors that arise in the nearby oropharynx, oral cavity tumors tend to be more difficult to control with radiation therapy. Small oral tongue tumors, for example, can be treated effectively by a brachytherapy implant, but partial glossectomy often achieves similar control in a more efficient fashion. Because there may be more than one effective treatment plan for an individual patient, there are no absolute ­guidelines regarding the best treatment for a given presentation of disease. Lesions treated effectively with surgery at one institution may be treated effectively with radiation therapy at another and vice versa. To some degree, this choice of treatment reflects physicians’ preferences or ­biases, the degree of expertise available at a particular institution, and the nature of the patient population receiving care at an institution. The choice of treatment therefore depends on the complications, cosmetic changes, and functional con­ sequences that the treatment is expected to cause. Treatment options for each stage of disease at each anatomic site are summarized in Table 12-7. Table 12-8 provides representative local control and survival rates for the spectrum of available therapies for oral cancer. The greater likelihood of lesions arising from the lips to be localized at

261

A

B FIGURE 12-18.  A poorly differentiated squamous cell carcinoma is seen on low power (A) and high power (B). (Photographs courtesy J. Jagadar, MD.)

presentation (Fig. 12-19) translates into a commensurate benefit in overall survival (Fig. 12-20). Surgery permits the patient’s tumor to be treated most quickly, allows the pathologist to assess the likelihood that the entire tumor was excised (i.e., whether the margins of the specimen contain tumor), and causes few dental or salivary flow problems. Conceptually, surgery must ­encompass the gross disease and the presumed microscopic (subclinical) extensions of tumor. As tumors become larger, the ­applicability of surgery becomes limited by functional and cosmetic factors despite the availability of reconstructive techniques and prostheses. In the case of relatively medial lesions, in which the regional lymphatics in both sides of the neck are at risk of invasion and may therefore need to be treated, surgery tends to be associated with more morbidity than is radiation therapy. Radiation therapy takes somewhat longer (brachytherapy) or considerably longer (teletherapy) than surgery to complete. Although radiation therapy preserves the anatomy and the function of most irradiated tissues, teletherapy often imposes some degree of xerostomia and, at least temporarily, alters taste. Some patients treated with radiation therapy will develop necrosis of soft tissue or bone, particularly after brachytherapy, just as some patients treated with surgery will develop fistulous tracts. The selection of therapy must be based on advice from a head and neck surgical oncologist and a radiation ­oncologist

262

PART 3  Head and Neck

TABLE 12-7.    Overview of Treatment Options by Tumor Site Treatment Options Tumor Site

Stage I

Stage II

Stage III

Stage IV

Lip

S, R

S, R

SR, R, C

SR, R, C

Tongue

S, SR, R

R, SR

R, SR, C

SR, R, C

Buccal mucosa

S, R

R, S, SR

S, R, SR, C

S, R, SR, C

Floor of mouth

S, R

S, R, SR

S, R, C

S, R, C

Lower gingiva

S, R

S, R

RS, SR, C

S, R, SR, C

Retromolar trigone

S, R, RS

S, R, RS

SR, C, R

SR, C, R

Upper gingiva

S, SR, R

SR

R, SR, C

SR, C, R

C, chemotherapy with or without surgery, with or without radiation therapy as part of clinical trial; R, radiation therapy; RS, radiation therapy followed by surgery; S, surgery; SR, surgery followed by radiation therapy. Modified from the National Cancer Institute PDQ Database. Available at http://cancernet.nci.nih.gov/cancertopics/pdq/treatment/lip-andoral-cavity/HealthProfessional (accessed January 2009).

TABLE 12-8.    Local Control and 5-Year Survival Rates by Tumor Site and Disease Stage Local Control Rate (%)/5-Year Survival Rate (%) Tumor Site

Stage I

Stage II

Stage III

Stage IV

Lip

90/90

90/90

NS/NR

NS/NR

Anterior tongue

96-100/100

85-93/70

50-65/35

40-45/NS

Buccal mucosa

90+/NS

90/NS

NR

NR/NS

Floor of mouth

88-100/NS

70-90/NS

70/NS

30-50/NS

Retromolar trigone

90-100/NS

90/NS

90/NS

60-89/NS

Upper gingiva

NR

NR

NR

NR

Lower gingiva

90+/NS

90+/NS

<90/NS

30-55/NS

Modified from the National Cancer Institute PDQ Database, May 1992. Available at http://cancernet.nci.nih.gov/cancertopics/pdq/ treatment/lip-and-oral-cavity/HealthProfessional/ (accessed January 2008). NS, not stated; NR, not reported.

100% 100% Observed survival

80% 60% 40% 20%

80% 60% 40% 20% 0% 0

0% Lip

Floor of mouth

Tongue

Gum/other

In situ Localized Regional Distant

FIGURE 12-19.  The extent of tumor at presentation is graphed by anatomic site. (Data from SEER*Stat 2.0, National Cancer Institute, Bethesda, MD, April 1999.)

1

2

3

4

5

6

7

8

9

10

Years Lip Tongue FOM Other

FIGURE 12-20.  Overall survival is plotted by anatomic site of the primary disease. Notice the similarities between all sites except the lip. FOM, floor of mouth. (Data from SEER*Stat 2.0, National Cancer Institute, Bethesda, MD, April 1999.)

12  The Oral Cavity

and optimally requires input from a diagnostic ­radiologist, a surgical pathologist, a medical oncologist, a dentist, and the patient. A variety of tumor and host factors need to be considered. The anatomic location and size of the lesion, potential involvement of adjacent tissues, ­clinical ­evidence of regional lymph node spread, the histologic type and grade of the tumor, the nature of the tumor margins (“pushing” versus infiltrating), the patient’s Karnofsky performance status and estimated physical reserve, psychosocial factors that may influence care or outcome (e.g., alcohol abusers and smokers who cannot or will not change are relatively worse candidates for radiation therapy), any personal preferences that a patient may hold, and any prior therapy (e.g., surgery or radiation therapy) in the volume of interest may individually or in combination tip the decision toward a particular form of treatment. Such factors notwithstanding, some general concepts apply.

T1 and T2 Lesions Limited-size lesions (T1 and T2) usually can be controlled with single-modality therapy, and surgery and radiation therapy often produce equivalent results. For small tumors, brachytherapy can control 85% to 90% of tumors initially, and with salvage surgery for treatment of recurrences, an ultimate local control rate of 95% is feasible.79

Small T3 Lesions Some small T3 lesions that just cross the T2 to T3 border and are not associated with adenopathy (i.e., small T3N0M0 lesions) may also be suitable for single-modality treatment.

Larger T3 and T4 Lesions Relatively large lesions (most T3 and T4) usually are not well controlled with surgery or with radiation therapy (local control rates ≤ 50%). Combined-modality therapy represents the treatment of choice if surgery can extirpate all gross evidence of tumor and is reasonable given the patient’s overall status. If the tumor has invaded bone, ­surgical resection of the affected bone in conjunction with the primary tumor and regional lymph nodes is indicated. Despite prior debate about the relative merits of preoperative versus postoperative radiation therapy, in most situations, postoperative radiation is the preferred choice. A prospective randomized trial of preoperative (50 Gy) versus postoperative (60 Gy) radiation therapy for advanced operable tumors of the oral cavity, oropharynx, supraglottic larynx, or hypopharynx conducted by the RTOG80 showed that after a median follow‑up of 10 years, local control rates were superior for the group that received postoperative radiation therapy.

High-Risk Resected Disease Within the group of tumors that are better treated by ­combined-modality therapy is a subset of particularly high-risk tumors that can be identified by factors such as ­extracapsular spread of nodal disease or microscopically involved mucosal margins of resection, or both. Even with gross total resection and postoperative radiation therapy, these tumors are associated with particularly high rates of local recurrence (27% to 61%), distant metastasis (18% to 21%), and death (5-year survival rates of 27% to 34%).81 Several studies that have tested the concurrent addition of

263

c­ hemotherapy to postoperative radiation therapy (in essence, ­chemotherapy-enhanced radiation therapy)82-87 have established the proof of principle that more intense treatment improves outcome for patients with these tumors. The two largest trials conducted85,86 involved the administration of cisplatin (100 mg/m2 given intravenously on days 1, 22, and 43 of radiation therapy) and demonstrated a significant improvement in local control that translated into significantly improved disease-free survival. However, the two trials differed in their eligibility criteria, and a combined analysis of the merged databases87 suggests that concurrent cisplatin-based chemotherapy–enhanced postoperative radiation therapy provides the greatest benefit for patients who have extracapsular spread of nodal disease or microscopically involved mucosal margins of resection, or both.

Lesions of the Lips Lesions of the lips are often amenable to surgery or radiation therapy. Small lesions, particularly of the lower lip, often can be simply excised. In the case of larger lesions and those close to a commissure, more complex plastic surgery is generally required to permit patients to open their mouths relatively widely to eat or to insert their dentures; radiation therapy therefore is the simpler alternative. The choice between surgery and radiation therapy usually should be made on the basis of the anticipated cosmetic and functional result. Younger patients who are destined to receive considerable additional sun exposure should preferentially be treated with surgery. Radiation therapy for tumors of the lip can be delivered by any of several different techniques. Superficial lesions can be treated with superficial x-ray therapy, and somewhat larger or thicker lesions can be treated with en face electron beams. Custom-made coated lead shields are placed behind the lip to absorb transmitted radiation. Alternatively, brachytherapy (interstitial implants or surface molds) can be used for small lesions. When lesions are very thick or wrap sufficiently far around the lip that their shape makes treatment by en face techniques or brachytherapy less desirable, tangential megavoltage external beam techniques can be used. A compensator is built to fit over the lip to eliminate skin sparing and provide dose homogeneity (Fig. 12-21).

Lesions of the Buccal Mucosa Tumors of the buccal mucosa can be treated with surgical excision, radiation therapy, or a combination of both modalities. Very small lesions can be excised and closed primarily, whereas larger tumors typically require a splitthickness skin graft for closure. For larger lesions of the buccal mucosa, combined surgery and radiation therapy is the best treatment; however, locoregional recurrence remains the predominant mode of treatment failure despite combined therapy.88 Tumors that involve the commissure of the lips usually are not treated surgically because of the risk of subsequent microstomia unless reconstruction is done. These tumors and moderate-sized (i.e., small T2) ­ lesions tend to be well suited for brachytherapy.89 ­ Somewhat larger lesions generally require external beam radiation therapy (EBRT), preferably with a technique that limits the dose to the opposite side of the mouth (e.g., mixed electron/photon beam or wedged-pair photon techniques). Large lesions (i.e., T3 and T4) usually are best treated with surgery and postoperative irradiation.

264

PART 3  Head and Neck

Lesions of the Hard Palate

Compensator Cobalt beam

Tumor

Cobalt beam

Lip Mandible

Hard palate lesions are usually of salivary gland origin and should be treated surgically. Postoperative irradiation of the tumor bed and the most likely route of spread of salivary gland tumors, particularly those displaying perineural spread, is advisable.

Regional Lymph Nodes

FIGURE 12-21.  A tissue compensator is used to provide dose uniformity for tangential megavoltage irradiation of a large ­tumor of the lip.

Lesions of the Oral Tongue Small lesions of the oral tongue, particularly those of the anterior tip, are readily managed with transoral surgical excision and generally should be treated with surgery. However, as lesions increase in size, the increasing functional deficits that are imposed by surgical management limit the usefulness of surgery and argue for radiation therapy. Brachytherapy should be considered for relatively small lesions that cannot be excised without a substantial functional deficit.90 Some investigators91,92 advocate peroral cone therapy for small lesions. Still others advocate surgery followed by EBRT.93 Large lesions generally are best treated with resection followed by postoperative irradiation. When this approach is not feasible, definitive EBRT and an interstitial boost can sometimes be effective.94 Remote highdose-rate brachytherapy is emerging as another alternative. High-dose-rate brachytherapy, alone or as a supplement to surgery or conventional radiation therapy, seems to have some efficacy in the treatment of cancers of the oral tongue and floor of the mouth.95,96

Lesions of the Floor of the Mouth Treatment options for lesions of the floor of the mouth are the same as those for lesions of the oral tongue (described in the previous paragraph), although results of radiation therapy are better for tumors of the floor of the mouth. Surgery is favored for lesions that are adherent to the mandible, whereas radiation therapy should be considered for lesions that invade the tongue. Proponents of elective neck dissection point to data suggesting improved cure rates (compared with rates for patients who are observed97); however, elective radiation therapy most likely produces a similar benefit.

Lesions of the Upper or Lower Gingiva or Retromolar Triangle Small lesions of the upper or lower gingiva generally should be excised. The relative scarcity of tissue in these areas forces the surgeon to excise adjacent bone as lesions advance in size; however, this usually does not have functional or cosmetic sequelae. Radiation therapy provides an alternative for small lesions when surgery is not possible, but proxi­ mity to bone essentially limits radiation therapy options to EBRT techniques or molds. Advanced lesions generally are best treated with surgery and postoperative irradiation. Small tumors of the retromolar trigone can be treated with radiation therapy or surgery, but large lesions require both.

The mode of treatment selected for the primary oral cavity tumor often influences the treatment applied to the draining lymph nodes. However, the following principles of management for regional nodes apply to the extent that they do not conflict with the required management of the primary tumor. Nodal groups that have a high likelihood of containing only occult disease should be treated with radiation therapy when the primary tumor is treated with radiation therapy. Elective neck dissection is warranted only when the position of the primary tumor necessitates surgical manipulation of the regional nodes anyway or a limited dissection is sought for its prognostic value (e.g., a supraomohyoid neck dissection for carcinomas of the floor of the mouth). Caution is advised in response to overly enthusiastic advocates of “elective” neck dissections; when such procedures reveal no evidence of disease, they represent unnecessary surgery, and when they reveal metastatic disease, it is often considered an indication for subsequent radiation therapy (in which case the cancer is effectively treated twice). Relatively small involved nodes (i.e., 0.5 to 1.5 cm) can usually be effectively treated with surgery or radiation therapy. Often, the modality most appropriate for the primary tumor is used. Larger involved nodes tend to require combined-­modality treatment. After tumor cells have expanded a node to beyond 3 cm in diameter, the nodal capsule is often transgressed, and tumor in adjacent soft tissues usually cannot be removed surgically. Conversely, nodes larger than 3 cm typically contain too much viable disease to be controlled reliably with radiation therapy alone. Such nodes and multiple enlarged nodes represent indications for combined therapy.

Recurrent Disease The influence and effects of primary treatment preclude simple strategies for treatment of recurrent disease. In general, treatment of recurrence is most successful when it relies on a different modality from the primary modality (i.e., surgery for recurrence after radiation therapy and vice versa). However, attempts at surgical salvage of disease that recurs after radiation therapy are more likely to succeed than are attempts at radiotherapeutic salvage of disease that recurs after surgery. For selected subgroups for which postsurgical recurrent disease can be encompassed by using brachytherapy techniques, Vikram and colleagues98 documented a reasonable prospect of radiotherapeutic salvage. Aggressive combinations of surgery and radiation therapy or chemotherapy and radiation therapy typically offer the best prospects for control of advanced tumors.99

Chemoprevention With increasing progress in the treatment of oral cavity carcinomas, more patients will live long enough to develop second independent malignancies. Chemoprevention

12  The Oral Cavity

r­ emains a greater hope than reality, but chemoprevention may improve the survival of patients with oral cavity tumors in the next decade. The progression from normal to dysplastic to malignant tissue that is often seen for oral cavity tumors offers a target for early intervention. Administration of 13-cisretinoic acid (13-CRA, Isotretinoin), a synthetic analogue of vitamin A, has caused regression of oral leukoplakia. Hong and colleagues100 reported that 13-CRA, given orally in daily doses of 1 to 2 mg/kg for 3 months, reduced the size of leukoplakic patches in 67% of patients. Unfortunately, about two thirds of the patients experienced side effects (mostly dryness of the skin and mucous membranes and elevation of serum triglyceride levels), and the beneficial effects were only temporary (patch regrowth occurred in 50% of patients within 3 months of cessation of 13-CRA). Hong and colleagues101 have also used 13-CRA to try to prevent the appearance of second independent malignancies in patients already treated for a head and neck cancer. In that study, daily 50 to 100 mg/m2 doses of 13-CRA given for 1 year were associated with a significantly decreased rate of second malignancies at a median follow-up time of 32 months. However, only two thirds of the patients completed a full year of treatment, and second malignancies seemed to be suppressed and delayed, not eliminated, by treatment. Unfortunately, a successor trial (RTOG 91-15) of a smaller daily dose of 13-CRA given for a longer period (30 mg daily for 3 years) did not reduce the rates of second primary tumors, death, or smoking-related disease.102

Radiation Therapy Techniques Discussion of radiation therapy techniques for tumors arising in the oral cavity must include brachytherapy (implants and molds) and teletherapy (conventional EBRT, intraoral cones, and increasingly, three-dimensionally planned, ­intensity-modulated, conformal techniques). For small lesions, brachytherapy delivered by means of radioactive implants or, less often, by means of a surface mold is preferable. Intraoral cone irradiation, a unique form of teletherapy for small lesions, has its advocates. EBRT provides the most versatility and becomes mandatory for large tumors, for which it is usually given postoperatively. Selection among these modalities depends on the anatomic tumor site, the extent of tumor involvement, and the treating physician’s experience, expertise, and preferences.

Brachytherapy Ideally, as long as the tumor volume is relatively small (preferably < 2.5 cm), brachytherapy can deliver lethal doses of radiation to a target tumor volume relatively quickly (brachytherapy is a form of accelerated radiation therapy) while at the same time limiting the radiation dose to adjacent normal tissues. Local control rates of 85% to 90% are typical; however, they are accompanied by self-limited soft tissue ulceration during the first 18 months after treatment in approximately 30% of patients.103 The technical limitations of source implantation are such that for practical purposes, if a tumor has a maximum diameter of more than 2.5 cm, it is difficult to place ­radioactive sources without creating “cold spots” that are underdosed, although with current real-time ­computerized

265

dosimetry systems, the size limit is being stretched. Because of the rapid fall-off in dose from each radioactive source, the need for supplemental EBRT to enhance the dose homogeneity increases in proportion to the implant volume. (EBRT supplements the dose at each cold spot and thereby increases dose homogeneity.) As the implant volume increases, so does the risk of mandibular osteonecrosis.103 Simon and colleagues104 showed that the spacing between sources (9 to 14 mm versus 15 to 20 mm) was related to rates of local control at 5 years (86% versus 76%, P = .13) and necrosis at 5 years (33% versus 46%, P = .04). When low-dose-rate brachytherapy is used as the exclusive method of irradiation, doses of 65 to 75 Gy (delivered at approximately 0.5 Gy/hr, resulting in 10 to 15 Gy/day) are commonly used and are potentially combined with elective dissection of regional lymph nodes when the risk warrants. When lesions exceed 2.5 cm, typical combined treatment plans deliver 50 Gy over 5 weeks by teletherapy and 30 Gy by brachytherapy (relying on the external beam to eradicate adjacent subclinical disease) or 50 Gy by brachytherapy and 30 Gy by external beam (to fill in any cold spots). Although the rules of source placement for low-dose-rate brachytherapy established years ago for preloaded radium or cesium needles still apply, temporary iridium 192 (192Ir) afterloading techniques have become the standard. 192Ir sources in the United States consist of rice-sized iridium seeds that are manufactured at precisely spaced increments from each other within flexible nylon tubes. In Europe, solid iridium wires serve the same purpose. The ability to ­manufacture high-intensity iridium sources permits highdose-rate afterloaded brachytherapy by using computer­controlled temporary placement of an intense iridium source in many positions within a single or multiple tubes. Although the ­radiobiologic effect is different, the placement of sources and resultant distribution of dose can approximate low-dose-rate brachytherapy without the need for inpatient ­hospitalization or the exposure of medical personnel. There are many different ways to implant sources through submental or intraoral routes. Pierquin and colleagues105 ­described nearly a dozen afterloading techniques that can be used to place radioactive materials in desired geometric patterns. The techniques have in common the introduction of nonradioactive rigid or nonrigid guides into tissues in such a way that the subsequently placed iridium sources must conform to the same pattern as the guides. Without the time pressures imposed by antiquated preloaded radium or cesium needles, these guides can be slowly and accurately positioned, and the subsequently placed radioactive sources are more likely to yield more homogeneous dose patterns. There are also different philosophies of source arrangement. Initially, idealized plans were required because of the difficulty of manually calculating isodose distributions from multiple sources of radiation. The Manchester system106 provided practitioners with a set of rules for source position and source strength that permitted prediction of the distribution of isodose curves generated. An alternative approach, the Paris system,107,108 used sources of uniform linear activity placed in equidistant parallel lines and planes. This approach was well suited for 192Ir ribbons. Any of these historically derived systems can be viewed as a starting point; modern computerized dosimetry systems can calculate dose distributions from virtually any arrangement of sources to permit individualization of implants for individual patient’s needs.

266

PART 3  Head and Neck

Most brachytherapy source arrangements can be viewed as a series of parallel lines, often connected by looping from one line to the other the flexible iridium sources that make up each line (previously, separate radioactive needles, known as crossing sources, provided the same function as the top of the loop). Figure 12-22 illustrates several arrangements of radioactive sources that are suitable for treating oral tongue lesions of various sizes and in various locations. Although the sources are drawn as rigid needles, the concepts are equally applicable for flexible sources. A blind-end iridium ribbon implant is essentially equivalent to a noncrossed rigid needle implant; a looped iridium implant is functionally equivalent to a needle implant with one crossed end. Most of the sources in Figures 12-22A and 12-22F are oriented so that they are being viewed on end (i.e., they run from the dorsum of the tongue down through the substance of the tongue into the floor of the mouth). This orientation is illustrated in Figure 12-23 using cesium needles and in Figure 12-24 using iridium ribbons. Nag and others109 described a nonlooping technique that is particularly well suited for high-dose-rate sources. A protector (Fig. 12-25, see Fig. 12-22D) is placed over the mandible to shield it from radiation and decrease the risk of subsequent osteonecrosis. Shielding normal tissues with a high-density absorber or somewhat less effectively by gauze or some other spacer to keep the closest source 0.5 cm or further away from the periosteum decreases the risk of subsequent osteonecrosis.110,111 Tumors of the floor of the mouth, as long as they are not too close to the mandible, can be treated in a similar fashion. However, placement of radioactive implants too close to the mandible leads to an unacceptable rate of osteonecrosis. Patients who have protruding genial tubercles may not have sufficient space in their mouths to permit successful brachytherapy. Lesions of the lip or buccal mucosa usually are thin enough to be treated with single-plane implants. Typically, the parallel sources are spaced approximately 1 cm apart, and the sources physically encompass the entire gross extent of the tumor plus 1 cm to provide a margin that will encompass subclinical disease.

Intraoral Cone Therapy An intraoral cone permits treatment of small volumes with EBRT that is more concentrated and more like brachytherapy than are typical teletherapy techniques. In experienced hands, use of intraoral cones for selected lesions (e.g., T1 and T2 oral tongue lesions) can yield results that rival and perhaps exceed those achieved with interstitial implant techniques.91,92 Although cones can be built with a periscope to permit visualization of the tumor in the treatment position, cones require a cooperative patient who will keep his or her mouth open widely and work to keep the ­lesion immobile despite any inherent soreness. This ­ technique is not commonly available; however, ­ Cederbaum and ­colleagues described a simple home-built cone for interested practitioners.112

Conventional External Beam Therapy Definitive primary irradiation of oral cavity tumors with EBRT techniques is a difficult task because these tumors are less likely to respond to treatment than are their counterparts in most other head and neck sites. Although the

Protector Tongue

A

Mandible

D

B

E

C

F

FIGURE 12-22.  A, Diagram of a single-plane implant suitable for a carcinoma of the border of the tongue of 1-cm thickness and 4 cm in the anteroposterior dimension. Five needles are implanted vertically with 1-cm separation between needles; a single needle is used to cross the superior end of the vertical needles, implanting it 0.5 cm below the mucosa in an ­anteroposterior direction. B, Diagram of a single-plane implant suitable for a carcinoma of the dorsum of the tongue. The needles are placed 0.5 cm deep to the mucosa, and both ends are crossed. C, Diagram of a single-plane implant for a carcinoma of the dorsum of the tongue with both ends crossed. The needles are placed 0.5 cm deep to the mucosa. D, Diagram of a dental plastic protector to reduce the dose to the gingiva and mandible, as used in a single-plane implant for a carcinoma of the floor of the mouth. Notice that the needles are inserted through the tongue and that the ends are not crossed. E, Diagram of a twoplane implant suitable for an advanced, 2.5-cm-thick carcinoma of the tongue. F, Diagram of a volume implant suitable for a very advanced (more than 2.5-cm-thick) carcinoma of the tongue.

concept of a dose‑response relationship holds true in the oral cavity, the range of doses that can be used in this region is constrained. Even small lesions that for some reason cannot be excised or treated with brachytherapy typically require approximately 70 Gy, establishing the lower end of the range. Large lesions, which in theory would benefit from doses in excess of 70 Gy delivered by ­ conventional fractionation, cannot be treated to substantially larger doses because of the limitations imposed by surrounding normal tissues. Doses approximating 70 Gy delivered over 7 weeks, conventionally fractionated at 2 Gy per daily treatment session, have traditionally been used for ­unresectable disease. However, recent findings suggest that better ­control of advanced tumors can be achieved with ­altered­fractionation regimens.

12  The Oral Cavity

267

FIGURE 12-23.  Radiographs of rigid cesium needles placed in a single-plane configuration for treatment of a tongue lesion.

FIGURE 12-24.  Radiograph shows a radioactive iridium implant for an advanced tumor of the tongue and floor of the mouth. The iridium seeds are contained in flexible nylon threads that are inserted through hollow metal guides placed in a submental and submaxillary approach. After the iridium-containing threads are inserted (i.e., afterloading technique), the guides are removed. Hairpin-shaped iridium wire can be used as a substitute for needles or for flexible iridium-containing threads.

Altered-Fractionation Regimens The optimal altered-fractionation approach is a concomitant boost approach in which 72 Gy is delivered in 42 fractions over 6 weeks (1.8 Gy per fraction per day to a large field plus 1.5 Gy per fraction per day to a smaller boost field during the final 12 treatment days) or a ­hyperfractionated regimen in which 81.6 Gy is delivered in 68 fractions over 7 weeks (1.2 Gy per fraction, twice daily, 6 hours apart).113 At the 2-year mark, both of these altered-fractionation regimens seem to improve locoregional control and disease-free survival, at the cost of greater acute toxicity (approximately 50% more grade 3+ toxicity) but no increase in late toxicity.

FIGURE 12-25.  A plastic gum protector (i.e., separator) is prepared in anticipation of a needle implant for a carcinoma of the floor of the mouth. The needles will be implanted through the substance of the tongue. The protector increases the separation of the needles from the gum and decreases the risk of excessive irradiation of the adjoining gum and mandible.

The addition of concurrent chemotherapy may further improve outcome. For example, Brizel and colleagues114 compared hyperfractionated radiation therapy alone with hyperfractionated radiation therapy plus concurrent and adjuvant cisplatin and fluorouracil and observed that at 3 years of follow-up, the combined treatment was associated with improved locoregional control, improved ­disease-free survival, and improved overall survival. Equally ­ important, with the regimen used (70 Gy in 1.25-Gy ­fractions twice daily plus 12 mg/m2 per day of cisplatin and 600 mg/m2 per day of fluorouracil for 5 days during weeks 1 and 6 of irradiation), the rate of severe complications was not significantly different between the groups. Further

268

PART 3  Head and Neck

improvements in the integration of altered-­fractionation radiation therapy with chemotherapy will most likely provide the next major advance in the treatment of oral cavity tumors.115,116 Identification of precisely which tumors will respond optimally to combinations of radiation therapy and chemotherapy will most likely be one of the major advances in cancer care over the next decade.

Postoperative Regimens In the postoperative setting in which all grossly detectable disease has been removed, smaller radiation doses are adequate. Although 90% of undisturbed microscopic epithelial tumors in a variety of anatomic sites are controlled when treated with 45 to 50 Gy,117 subclinical disease in a bed that has been surgically altered is somewhat more difficult to control. Peters and colleagues,118 in a prospective, randomized, dose-seeking trial of postoperative EBRT for head and neck cancer, demonstrated that 52 to 54 Gy ­delivered in 1.8-Gy increments was insufficient even for tumors that were excised with uninvolved margins. In common practice, undissected tissues with a low risk of tumor involvement receive 50 Gy; dissected tissues with clean margins receive 60 Gy; and dissected tissues with involved margins receive 66 Gy. The necessity of adequate dose is particularly important for tumors with features that place the patient at a relatively higher risk for local ­recurrence— that is, two or more involved lymph nodes, extracapsular extension of disease, or microscopically involved mucosal margins.81 The presence of microscopic extension of disease beyond the capsule of an involved lymph node or the presence of microscopic amounts of tumor at the mucosal margin of resection also seem to warrant the concurrent addition of cisplatin chemotherapy to postoperative radiation therapy.85-87

Treatment Setup and Portal Arrangement Midline lesions of the oral cavity traditionally have been treated with a three-field technique consisting of lateral parallel-opposed photon fields and an anterior low-neck field. Far lateral tumors (e.g., tumors confined to the buccal mucosa) are better treated with wedged-pair, mixed electron-photon, or intensity-modulated techniques mated to an ipsilateral or bilateral low-neck field or by intensity-modulated therapy that spares contralateral normal tissues. The arrangement should encompass the primary tumor and the first echelon of clinically uninvolved draining nodes in a single portal. For most oral cavity subsites, these first-­echelon nodes are located in levels I and II. If lymph nodes are clinically involved, they should be included in the primary treatment portals whenever possible. In general, for tumors of the oral tongue or floor of the mouth, the lateral fields encompass the primary tumor, the submandibular nodes, and the upper jugular nodes. Irradiation of the posterior chain is not required unless the cervical nodes show clinical evidence of involvement. For tumors of the retromolar trigone, the ipsilateral posterior cervical nodes are usually irradiated only if the primary tumor is T3 or T4. For primary tumors at any site, the posterior cervical nodes generally should be ­irradiated if there are clinically involved nodes in the anterior chain. Figures 12-26 and 12-27 depict typical ­treatment ­portals.

Right lateral

Left lateral

Anteroposterior

FIGURE 12-26.  Schematic drawings show the anatomic sites included in radiation fields used for extensive tumors in the lower aspect of the oral cavity. For small, well-differentiated tumors, the field treating the low anterior neck should be omitted in the absence of palpable disease.

Tissue compensators may be needed to produce relative homogeneity of dose throughout the treatment volume. For lesions of the floor of the mouth, inferior gingiva, or oral tongue, a bite-block spacer (also known as a stent or lollypop) ideally should be placed inside the patient’s mouth so that the tongue is depressed and the maxilla is moved up and out of the fields. For all tumors except very small tumors with no demonstrable associated adenopathy, the primary tumor portals are mated to an anterior ­ low-neck field that irradiates the nodal stations that lie between the inferior edge of the primary treatment portals and the inferior edge of the clavicles. Philosophically, the intent is to encompass all nodal stations that have a probability of subclinical involvement (based on the characteristics of the primary tumor) while minimizing the risk of treatment complications. For lesions of the oral cavity that do not have bulky associated adenopathy, a midline block is generally used in the anterior neck field to shield the underlying larynx and spinal cord. To maximize the patient’s comfort, the primary portals should be as small as possible, and the ­anterior portal should be commensurately as large as possible without excluding gross disease from the lateral portals. As they traverse tissue, the inferior edges of the lateral treatment portals diverge downward, and the superior edge of the anterior portal diverges upward. If the lateral and anterior fields are abutted on the skin, there is a small area of overlap. Because the midline block in the anterior field protects the spinal cord from irradiation, the area of overlap usually is not clinically important. However, the match can be improved by using a half-beam block (in which the

12  The Oral Cavity

A

269

B

C

D FIGURE 12-27.  A, Representative volume included in the lateral treatment portals for postoperative irradiation of a tumor of the floor of the mouth (corresponding to Fig. 12-26). B, Anterior neck field for the same patient. C, Portal film of the lateral field.  D, Portal film of the anterior neck field.

lateral fields are treated with the upper, unblocked half of the portal and the anterior neck with the lower, unblocked half of a portal, using a common central axis) or by angling the anterior field toward the patient’s feet (which is easily done by turning the couch 90 degrees and angling the gantry approximately 3 degrees, depending on field sizes and source-to-skin distance, to eliminate cephalad ­divergence). If bulky adenopathy is present in the low-neck field, a halfbeam blocked combination of lateral opposed fields super­ iorly mated to anteroposterior-posteroanterior opposed fields inferiorly often produces an appropriate plan.

Before or during simulation, markers can be used to demarcate tumor for image guidance of daily setup. The patient’s head should be sufficiently immobilized (e.g., with an Aquaplast face mask) to prevent movement during treatment. As part of treatment planning, critical structures must be identified on the treatment plan, and steps must be taken to ensure that the doses to be received are within the tolerance of the normal tissues. Simulation films or digitally reconstructed radiographs and portal-image or beam films should be taken of all new fields to ensure accurate placement. Verification portal image films, beam films, or

270

PART 3  Head and Neck

TABLE 12-9.    Simplified Schema for Radiation Therapy of the Lip Disease Stage

Modality

Dose

Boost Dose

Neck Treatment

T1-T2 N0

Choice of XRT or brachytherapy

50 Gy, 25 fractions

6-10 Gy in 3-5 fractions None

No RT

T3 N0

XRT

50 Gy, 25 fractions

10-20 Gy in 5-10 fractions

Levels I and II

T1-T3, positive nodes

XRT

50 Gy, 25 fractions

16-20 Gy in 8-10 fractions

Levels I-IV

60-65 Gy, 5-7 days

No RT

RT, radiation therapy; XRT, external beam radiation therapy.

TABLE 12-10.    Simplified Schema for Radiation Therapy of the Oral Tongue and Floor of the Mouth Disease Stage

Modality

Dose

Boost Dose

Neck Treatment

T1 N0

Brachytherapy

65-75 Gy, 6-7 days

None

No XRT

T2 N0

XRT

50 Gy, 25 fractions

6-10 Gy in 3-5 fractions

Levels I and II

T3-T4 N0 or N+

Surgery + XRT or altered-fractionation XRT or chemotherapy + XRT

60-66 Gy, 30-33 ­fractions 72 Gy, 42 fractions, 6 weeks 70 Gy, 35 ­fractions

None

All levels

Included in 72 Gy None

XRT, external beam radiation therapy.

stereoscopic x-ray–guided images should be taken periodically throughout treatment to ensure that placement of the portals has remained accurate. Tables 12-9 and 12-10 present simplified radiation therapy schema for common sites and categories of oral cavity tumors.

Three-Dimensional Conformal or Intensity-Modulated Radiation Therapy The development of high-speed computers that can ­calculate dose distributions from non-coplanar fields in three dimensions within a time frame congruent with ­appropriate patient care has given birth to conformal radiation therapy and, subsequently, to intensity-modulated radiation therapy (IMRT). As the name suggests, the goal of conformal treatment is to shape the dose to the tumor as tightly as possible, sparing the normal tissues as much as possible. For such plans, the traditional concept of describing boundaries of treatment portals is replaced by definitions of three-­dimensional tumor volumes that need to receive specified dose minimums and three-dimensional normal tissue volumes that ideally should receive less than specified dose maximums. The adequacy of target coverage and assessment of the potential for injury of normal tissues can be quantified by dose-volume histograms; however, the intensity (fluence) of each beam is essentially uniform when viewed in cross section unless a modifier, such as a wedge, has been interposed. IMRT can subdivide each beam into smaller beamlets, creating beams of intentionally ­nonuniform intensity, thereby permitting greater shaping of the dose envelope than does three-dimensional conformal radiation therapy. IMRT also permits the simultaneous treatment of defined subvolumes to different dose levels.

Within the oral cavity, three-dimensional conformal therapy and IMRT protect normal tissues (most importantly, the salivary glands; see “Xerostomia” later in this chapter) and may result in better tumor control through escalation of the dose to small volumes. IMRT has not yet had the same impact in treating tumors of the oral cavity as it has in treating cancers of the oropharynx or nasopharynx. However, published evidence of its clinical benefit in some instances is growing.119,120 The general guidelines of Tables 12-9 and 12-10 also apply to three-dimensional conformal therapy and IMRT, with the understanding that biologically equivalent doses need to be substituted when intensity-modulated techniques deliver different daily doses to specified subvolumes.

Results of Radiation Therapy The success rate of radiation therapy for carcinomas of the oral cavity is best measured by local control of disease. The lifestyles that are often associated with oral cancer can also lead to other neoplastic and nonneoplastic causes of death that can obscure the efficacy of treatment.121 In addition to the cardiovascular consequences of smoking and drinking, second independent malignancies are commonly appreciated as important causes of morbidity and mortality in such patients. However, locoregional recurrence of tumor remains the principal cause of morbidity and mortality for them.116 Local recurrence after radiation therapy is more common for oral cavity tumors than for tumors in most other head and neck sites.122-124 Failure to control primary tumors becomes apparent relatively quickly; 80% to 90% of local recurrences are

12  The Oral Cavity

271

­ etected within 2 years after completion of treatment, and d virtually all other recurrences become apparent over the next year. Beyond 3 years, local recurrences are rare, and second independent malignancies begin to pose an increasingly greater problem. In each successive year, approximately 3% to 4% of patients who are cured of their first tumor develop a second independent tumor, most often arising in the upper aerodigestive tract.121 Three years of follow‑up is adequate to detect most distant metastases. Patients who are free of disease 3 years after treatment therefore can be considered cured of their disease.

TABLE 12-11.    P  ercentage of Patients in Whom Oral Cavity Tumors and Involved Lymph Nodes Were Cleared by Radiation Therapy According to Disease Stage

Factors Influencing the Likelihood of Local Control of Disease The probability of tumor control is influenced by various tumor and host factors. The best correlates of outcome are the anatomic measures of primary tumor, nodal, and distant metastatic involvement that form the basis of the AJCC staging system. The likelihood of control is also influenced, but to a lesser degree, by the anatomic subsite of disease (see Table 12-8). Vascular invasion seems to predict increased risk of local recurrence and diminished likelihood of survival.77 Suggestions have been made in some analyses that tumor control correlates with the primary tumor’s thickness53,125; the tumor’s histologic grade (with control less likely for well‑differentiated tumors than for moderately or poorly differentiated lesions); the patient’s hemoglobin level; and the patient’s sex, with control more likely in women than in men.126 In general, lesions that originate more anteriorly have a better prognosis than lesions that arise more posteriorly (i.e., moving from the lips toward the base of tongue). Local control is directly related to the total dose of radiation delivered and inversely related to the duration of treatment.126

Patterns of Disease Regression Primary tumors usually regress more rapidly than do involved regional lymph nodes. However, the rate of tumor response is only a weak predictor of eventual local control. Tumors that have not regressed completely by the completion of radiation therapy may resolve over the next few weeks. Sobel and colleagues127 suggested that the most accurate time to assess the degree of ­tumor clearance is 30 to 90 days after the completion of ­radiation therapy. Overall, complete regression of disease can be expected in 65% to 75% of all patients who present with tumors of the oral cavity. From 1977 to 1980, the RTOG prospectively registered all patients with tumors of the head and neck region who had been seen in member institutions.128 A total of 217 patients with cancer of the oral cavity (excluding tumors of the lips, which have a better prognosis) were followed, and for these patients, a complete response rate of 67% was observed. Salvage surgery eliminated residual disease in an additional 6% of patients, for a total complete response rate of 73%. It is instructive to examine disease clearance rates in the primary tumor and regional lymph nodes separately (Table 12-11). In the RTOG study, complete clearance of the primary tumor was observed in 97% of T1 lesions of the oral cavity, 87% of T2 lesions, 61% of T3 lesions, and 38% of T4 lesions.128 In similar fashion, rates of clearance of metastatic disease in regional lymph nodes decreased as

Patients with Clearance of Tumor (%) All T categories

76

T1

97

T2

87

T3

61

T4

38

Patients with Clearance of Lymph Node Involvement (%) All N categories

68

N1

83

N2

58

N3

55

Patients with Clearance of Primary Tumor and Lymph Node Involvement (%) Stage I

97

Stage II

87 (6)*

Stage III

51 (14)*

Stage IV

39 (5)*

*Clearance

achieved with radiation therapy and surgical salvage therapy is given within parentheses. Modified from Marcial V, Amato D, Pajak T. Patterns of failure after treatment for cancer of the upper respiratory and digestive tracts: a Radiation Therapy Oncology Group report. Cancer Treat Symp 1983;2:33-40.

nodal size increased. Disease was completely cleared in approximately 83% of patients with N1 disease, 58% with N2 disease, and 55% with N3 disease.128 Little or no difference was observed in the response of nodes based on the exact site of origin of the primary tumor. Data from the RTOG’s registry of patients with head and neck cancer treated with radiation therapy clearly demonstrate decreases in local control and survival rates with increases in disease stage (Tables 12-12 and 12-13). There is a greater difference between 3-year and 5-year survival rates than between 3-year and 5-year locoregional control rates, a pattern reflecting the relatively high rate of death from nonneoplastic causes and second independent malignancies in this patient population. Because of the diversity of tumors that are grouped within each stage category, even for single anatomic subsites, the expected outcome of treatment (ultimate local control or survival) can only be approximated. With this understanding, rates of local control of early-stage tumors of the oral cavity can be expected to be in the range of 80% to 90%.129 Moderately advanced lesions typically have a 60% to 70% likelihood of local control, whereas far­advanced lesions are controlled 35% to 45% of cases. It is perhaps surprising, but also instructive, to review the variations in reported outcome by stage and anatomic site and to notice for how many categories reliable (nonanecdotal) data are not available (see Table 12-8).

272

PART 3  Head and Neck

TABLE 12-12.    R  ates of Complete Response to Radiation Therapy and Locoregional Control at 3 and 5 Years for Patients with Oral Cavity Tumors in the Radiation Therapy Oncology Group Head and Neck Cancer Registry Local Control Rate (%)

Disease Stage

Number of Patients

Complete Response Rate (%)

3 Years

5 Years

I

36

97

74

65

II

71

87

49

49

III

51

51

11

*

IV

59

39

*

*

*Too

few patients for reliable estimate.

TABLE 12-13.    E  stimated Absolute Survival at 3 and 5 Years for Patients with Oral Cavity Cancer in the Radiation Therapy Oncology Group Head and Neck Cancer Registry Absolute Survival Rate (%) Disease Stage

Number of Patients

3 Years

5 Years

I

36

68

53

II

71

52

39

III

51

33

*

59

*

*

IV *Too

few patients for reliable estimate.

Site-Specific Results Lesions of the Lips Carcinomas of the lip have the most favorable outcome of all the oral cavity tumors. Most lesions are detected when they are relatively small, and cure rates exceeding 90% are possible with radiation therapy or surgery. Beauvois and colleagues130 treated 237 patients with squamous cell carcinoma of the lower lip (158 T1, 61 T2, 17 T3, and 1 T4 lesions) with low-dose-rate 192Ir brachytherapy alone. With 5 years of follow-up, the local control rate was 95%, the overall survival rate was 74%, and the cause-specific survival rate was 91%. Fongione and colleagues131 treated 69 patients with lip tumors (36 T1, 12 T2, 2 T3, and 19 recurrent tumors) with interstitial brachytherapy alone and reported only one persistent tumor and no local recurrences. At 5 years, the overall survival rate was 77%, and the disease-specific survival rate was 91%. Orecchia and colleagues132 treated 47 patients with lip tumors (21 T1 and 26 T2 lesions) and obtained local control in 94%. Tombolini and others133 retrospectively reviewed their results of low-dose-rate interstitial brachytherapy for patients with squamous cell carcinoma of the lower lip (27 T1, 20 T2, and 10 T3 lesions; eight patients had clinically involved regional nodes). The median tumor dose was 62 Gy (65 to 70 Gy to involved nodes), and all patients experienced complete remission. The actuarial local control rate was 90% at 5 years, increasing to 94% with salvage surgery; the locoregional control rate was 86%, increasing to 89% with salvage surgery; and the actuarial overall survival rate at 5 years was 76%. In a later study, Finestres-Zubeldia and colleagues134 used high-dose-rate afterloading applicators placed in surface molds to treat 28 carcinomas of the lip; doses ranged

from 60 to 65 Gy in 33 to 36 fractions of 1.8 Gy for tumors up to 4 cm and from 75 to 80 Gy in 43 to 46 fractions for larger lesions. That group reported no local recurrences and good to excellent cosmetic results 1 year later. In another study, Guinot and others135 used implanted needles to deliver high-dose-rate therapy (8 to 10 fractions of 5 to 5.5 Gy twice each day, for a total dose of 40.5 to 45 Gy) to 39 patients with carcinomas of the lip. The 88% local control rate at 3 years (95% for T1-2 versus 74% for T4) led the investigators to conclude that “[high-dose-rate] results are equivalent to [low-dose-rate results].”

Lesions of the Buccal Mucosa Relatively small carcinomas of the buccal mucosa generally are considered suitable for surgery or irradiation. However, Strome and others136 observed a startling 100% local recurrence rate after wide local excision alone for stage I and II tumors. Sakai and colleagues137 reported the outcome of a variety of treatments for 55 buccal carcinomas. As assessed by multivariate analysis, the local control rate was 85% in patients treated with preoperative radiation therapy followed by surgery, interstitial implant, or electron beam therapy. In contrast, the local control rate was approximately 30% for patients treated with external photon beam therapy or external photon beam therapy followed by mold therapy with a remote afterloading system. The investigators concluded that the success of interstitial brachytherapy rivals that of surgery for T1 to early T3 buccal carcinomas that do not invade the buccoalveolar sulci or retromolar trigone regions. Lapeyre and colleagues89 obtained local control in 91% of 22 buccal carcinomas treated with a looping (encircling) technique. Fang and colleagues88 reported a 3-year actuarial locoregional control rate of 64% and an overall survival rate of 55% after surgery and postoperative

12  The Oral Cavity

273

radiation therapy for six stage II, 21 stage III, and 30 stage IV buccal cancers. Mishra and colleagues125 found that tumor thickness predicted the risk of locoregional recurrence of buccal carcinomas.

should be considered for surgery or EBRT (alone or with a limited brachytherapy boost) or both ­modalities, depending on the tumor size and location.

Lesions of the Gingiva, Hard Palate, and ­Retromolar Triangle

For oral tongue tumors, surgery is the preferred form of treatment.145 Umeda and colleagues146 retrospectively compared low-dose-rate brachytherapy (78 patients), highdose-rate brachytherapy (26 patients), and surgery (71 patients) for stage I or stage II cancer of the tongue. Local recurrence was detected in 13 patients (17%) in the lowdose-rate group, 9 (35%) in the high-dose-rate group, and 4 (6%) in the surgery group. Although these investigators concluded, “These findings show that surgery is the optimal treatment method for patients with stage I-II tongue cancer,” it is difficult to know the precise role of case selection in that series. Other investigators have claimed that when small carcinomas of the oral tongue can be encompassed within the high-dose envelope produced by brachytherapy and the envelope avoids irradiation of the mandible, results comparable to surgery can be obtained.147 Bourgier and others148 at the French Centre Oscar Lambret treated 279 patients with T2 N0 mobile tongue carcinoma with low-dose-rate 192Ir brachytherapy (with or without neck dissection) according to the Paris system rules, with a median total dose of 60 Gy (median dose rate, 0.5 Gy/hr). The local control rate was 79% at 2 years, but occult nodal metastases were found in 45% of dissections, implying that more than local brachytherapy is needed for T2 N0 lesions. Pernot and colleagues149 treated 125 T1 and 186 T2 tumors with brachytherapy with or without EBRT and obtained 5-year local control rates of 93% and 65%, respectively. Five-year disease-specific survival rates were 78% and 43%, respectively, and overall survival rates were 69% and 41%, respectively, for T1 and T2 tumors. Mazeron and colleagues150 obtained a local control rate of 87% for 70 T1 N0 and 83 T2 N0 tumors treated with 192Ir. This corresponded to 5year actuarial survival rates of 52% for those with T1 N0 tumors and 44% for those with T2 N0 tumors. Yoshida and colleagues151 reported a 5-year local control rate of 78% for 498 T1 and T2 oral tongue tumors treated with brachytherapy alone or brachytherapy and EBRT. Similarly, in a series of 207 squamous carcinomas of the oral tongue treated with brachytherapy alone (127 tumors) or combined brachytherapy and EBRT (80 tumors), Fujita and colleagues152 reported 5-year local control rates of 93% for T1 tumors, 82% for T2a tumors, and 72% for T2b tumors. When low-dose-rate implants are combined with EBRT, several groups149,152,153 have reported better control when a greater percentage of the dose is delivered by brachytherapy. However, the retrospective nature of these series raises the possibility that larger lesions were more often treated with combined therapy and that the outcome reflects selection as much as it does modality. Nevertheless, this observation has been sufficiently common that it seems prudent to heed its implicit warning. Similarly, Pernot and colleagues149 compared outcomes in 152 T2 N0 tumors treated in a nonrandomized fashion with brachytherapy alone or brachytherapy and EBRT. Local control rates (90% versus 50%, P = .0000) and cause-specific survival rates (60% versus 31%, P = .01) were better after brachytherapy alone, although these data must be interpreted

Relatively little information is available about radiation therapy for carcinomas of the gingiva, hard palate, or retromolar trigone. It seems fortunate that relatively few squamous cell carcinomas arise on the upper gingiva or hard palate, because bone is often involved and the outcome after any type of therapy tends to be poor.138 However, Cengiz and colleagues139 reported two early-stage carcinomas of the upper gingiva that were locally controlled with high-dose-rate brachytherapy delivered by a surface mold. Fuchihata and colleagues140 treated 162 squamous cell carcinomas of the lower gingiva with EBRT and simultaneous bleomycin or peplomycin chemotherapy. Sixty-seven percent of T1 and T2 tumors responded completely, as opposed to only 35.5% of T3 and T4 tumors. Local control rates at 5 years (including salvage therapy) were 91% for T1 tumors, 89% for T2 tumors, 76% for T3 tumors, and 61% for T4 tumors. Five-year cause-specific survival rates were 75% for stage I, 87% for stage II, 71% for stage III, and 51% for stage IV tumors. Carcinomas of the retromolar trigones are often grouped with carcinomas of the anterior faucial pillar, and relatively few series clearly separate these two entities. However, Lo and colleagues141 reported that local control was achieved in 66% of a variety of retromolar trigone lesions (T1 through T4) treated essentially with 60 to 75 Gy of EBRT.

Lesions of the Floor of the Mouth Small, centrally located carcinomas of the floor of the mouth respond well to brachytherapy. In 2002, Marsiglia and ­colleagues142 reported a review of 160 cancers of the floor of the mouth treated at the Institute Gustave Roussy according to the Paris system rules, in which brachytherapy was delivered with 192Ir wires (for T1 tumors) or 192Ir wires plus neck dissection (for T2 and/or N1 tumors). The ­local control rates in that review were 93% for T1 tumors and 88% for T2 tumors, with follow-up times ranging from 9 to 19 years. Bone necrosis of any grade was seen in 18% of patients, and severe necrosis of any type was seen in less than 10% of patients. Pernot and colleagues143 treated 102 carcinomas of the floor of the mouth with brachytherapy only and treated 105 lesions with EBRT plus brachytherapy. At 5 years, local control rates were 97% for T1 lesions, 72% for T2 lesions, and 51% for T3 lesions. This translated into 5-year overall survival rates of 71%, 42%, and 35% and 5-year cause-specific survival rates of 88%, 47%, and 36%, respectively. Grade 2 or more severe complications occurred in 19% of patients. These ­ investigators concluded that brachytherapy alone is the preferable treatment for T1 N0 and T2 N0 tumors. ­ Mazeron and ­ colleagues144 ­reported primary tumor control rates of 93.5% and 74.5% for T1 N0 and T2 N0 ­ tumors, respectively, after 192Ir ­brachytherapy, which resulted in 5-year cause-specific survival rates of 94% and 62%. However, after lesions extended to the gingiva or exceeded 3 cm in ­ diameter, control rates were only 50% and 58%, respectively. Consequently, larger tumors and those located close to the mandible

Lesions of the Oral Tongue

274

PART 3  Head and Neck

with ­caution because of the hazards inherent in comparing nonrandomized groups. Definitive irradiation of oral tongue tumors is associated with soft tissue and bone complications. Pernot and colleagues149 reported grade 1 complications in 19% of patients (15% soft tissue and 3% bone), grade 2 complications in 6% of patients (2% soft tissue and 4% bone), and grade 3 complications in 3% of patients (1% soft tissue and 2% bone). Fein and colleagues145 reported some degree of soft tissue necrosis or bone exposure in approximately one third of patients and severe complications in 11% of patients with oral tongue tumors treated mostly with brachytherapy and EBRT. Studies of high-dose-rate brachytherapy have shown somewhat contradictory results. Leung and colleagues,154 reporting a 94.7% local control rate at 4 years after high-dose-rate therapy, described their results as “encouraging.” Inoue and others155 reported a prospective phase III trial comparing low-dose-rate brachytherapy (70 Gy given over 4 to 9 days; 26 patients) and high-dose-rate brachytherapy (60 Gy given in 10 fractions twice daily, spaced more than 6 hours apart over 1 week; 25 patients) for ­early-stage cancer of the oral tongue; the 5-year local control rates in that study were 84% for low-dose-rate and 87% for high-dose-rate brachytherapy. These findings stand in contrast to those of Umeda and colleagues,156 who treated 25 patients with high-dose-rate brachytherapy for stage I or stage II tongue cancer and compared the findings with those from 71 patients who had previously undergone low-dose-rate brachytherapy at the same institution. The investigators concluded that the high-dose-rate treatment produced lower local control and survival rates and higher rates of osteonecrosis, which also tended to appear earlier than that which followed low-dose-rate treatment.

Late Effects and Complications of Radiation Therapy Although many squamous cell carcinomas of the head and neck can be eradicated without inducing serious long‑term complications in surrounding normal tissues, the relative ­sensitivities of tumors and the surrounding normal tissues are such that complications (Box 12-2) are inevitable in a small percentage of patients. The development of validated instruments for measuring quality of life has allowed more precise estimates of the toxicity of conventional treatment. To facilitate comparison of the degree and type of side effects after treatment, the RTOG has created a scale to describe late toxicities (i.e., those appearing beyond 90 days after treatment) (Table 12-14), and the National Cancer Institute has developed the Common Terminology Criteria for Adverse Events.157

Xerostomia With conventional radiation therapy, virtually all patients experience some degree of persistent xerostomia if a substantial part of the parotid glands is included within the radiation portal. Scintigraphic evaluation of parotid gland function suggests that a dose of less than 33 Gy will result in subacute xerostomia (i.e., occurring within the first 6 months) in 50% of patients and that a dose of less than 52 Gy will result in chronic xerostomia (i.e., occurring after 12 months) in 50% of patients.158 In one study, most glands irradiated with doses of 40 to 50 Gy had recovered

BOX 12-2.  Side Effects and Complications of Treatment of Oral Cavity Tumors Xerostomia Mucositis Diminished taste acuity Sensitivity to spicy foods Enhanced dental decay Susceptibility to infection Impaired nutrition Surgically induced difficulty eating, speaking, and swallowing Relative stiffness of tissues Osteonecrosis

a­ pproximately 70% of their pretreatment salivary flow rate by 18 months after the completion of treatment.159 Epstein and colleagues160 documented complaints of dry mouth in 92% of patients, changes in taste in 75%, dysphagia in 63%, altered speech in 51%, difficulty using dentures in 49%, and increased dental decay in 31% of dentate patients. The advent of conformal IMRT techniques probably offers the best prospects for avoiding acute and subacute xerostomia if the mean dose delivered to one parotid gland can be kept below 26 Gy.161 Alternatively, pharmacologic agents may be able to moderate radiation-induced damage to some degree. Pilocarpine, a parasympathomimetic drug, may be able to degranulate secretory cells and possibly decrease their radiosensitivity.162 However, at the conventional radiation doses required for control of human tumors, the predominant mechanism of action by which pilocarpine restores saliva is stimulation of the parts of the salivary glands that are not encompassed within the treatment portals.163 In glands irradiated to tumoricidal doses in their entirety, the administration of 5 mg of pilocarpine four times each day during and after radiation therapy for a total of 3 months had no beneficial effect.163 The seminal work by Brizel and colleagues164 has shown amifostine, a free-radical scavenger, to be associated with a lower incidence of acute xerostomia (grade ≥ 2), a delay in the onset of xerostomia (i.e., requiring greater cumulative doses), and a diminution of chronic xerostomia.164 Nevertheless, xerostomia was still present in a substantial proportion of patients; the use of amifostine reduced the rate of acute xerostomia from 78% to 51% and that of chronic xerostomia from 57% to 35%. A Cochrane review165 recommended amifostine as an option for reducing the rates of acute and chronic xerostomia associated with conventionally fractionated radiation therapy given with curative intent that encompass at least 75% of the parotid glands. However, that review also found that the available data related to prevention of acute and chronic xerostomia contained “significant heterogeneity.” Consequently, many radiation oncologists are attempting to avoid xerostomia more often by dose shaping to avoid salivary tissue (e.g., by IMRT) than by pharmacologic stimulation or protection of such tissues. Xerostomia makes it more difficult for patients to chew and swallow their food. Changing diets to include softer foods and more gravies and sauces can be helpful. Artificial saliva improves the comfort level of some patients. Others simply carry a small bottle of water with them and rinse their mouths at frequent intervals. Other patients find that

12  The Oral Cavity

275

TABLE 12-14.    Radiation Therapy Oncology Group Criteria for Late Toxic Effects Toxic Effects Site

Grade 0

Grade 1

Grade 2

Grade 3

Grade 4

Skin

None

Patch atrophy Slight atrophy ­Pigmentation change Moderate Some hair loss ­telangiectasia Total hair loss

Marked atrophy Gross telangiectasia

Ulceration

Subcutaneous tissue

None

Slight induration (fibrosis) and loss of subcutaneous fat

Moderate fibrosis but Severe induration and loss of subcutanewithout symptoms ous tissue Slight field contracture Moderate field (<10% reduction in contracture (>10% linear measurement) reduction in linear measurement)

Mucous ­membranes

None

Slight atrophy and dryness

Moderate atrophy and telangiectasia Little mucus

Marked atrophy with complete dryness Severe telangiectasia

Ulceration

Salivary gland

None

Slight dryness of mouth; good ­response on ­stimulation

Moderate dryness of mouth; poor response on ­stimulation

Complete dryness of mouth; no response on stimulation

Fibrosis

Necrosis

Modified from Cox JD, Stetz J, Pajak TF. Toxicity criteria of the Radiation Therapy Oncology Group (RTOG) and the European Organization for Research and Treatment of Cancer (EORTC). Int J Radiat Oncol Biol Phys 1995;31:1341-1346.

c­ hewing gums specially formulated for dry mouths (e.g., Biotene) or sucking on sugar-free candy is soothing. Rinsing with a solution of baking soda in water (1 teaspoon baking soda, with or without some salt, dissolved in one-half glass of room-temperature water) is relatively bland and soothing and helps dissolve sticky saliva. Baking soda can also be used with a soft toothbrush to clean the teeth and tongue. Commercial mouthwashes (e.g., Biotene) have also been developed for ­patients who have undergone irradiation of the oral cavity.

Fungal, Viral, and Bacterial Infections Because the dry mouth is prone to opportunistic attack, patients may require specific therapy for fungal, viral, or bacterial infections. Candida albicans is the most common opportunistic organism. It classically appears as a fluffy, off‑white growth that can be partially scraped off the involved tissues by using gentle pressure with a tongue blade. Sometimes this organism is less obvious, appearing only as fissuring of the angles of the lips. Patients who demonstrate marked erythema or mucositis earlier than anticipated during the course of treatment should be examined particularly closely for such infection. In most instances, candidal infections can be managed with topical suspensions (e.g., nystatin) or, if necessary, with systemic therapy (e.g., fluconazole). When nystatin is chosen, patients must be instructed to swish the medication in the mouth vigorously for at least 5 minutes four times per day. Some patients are unwilling to do this, some patients are unable to do this because of soreness, and some patients cannot get the medication far enough back in the mouth to come in contact with the fungus. Theoretically, systemic therapy provides advantages over topical therapy. However, patients should be informed of the potential side effects of systemic therapy, including potentially severe hepatic toxicity, and should be allowed to participate in choosing their medication.

Herpes simplex is the most common viral infection. ­ lassically appearing as small vesicles, the ­infection ­typically C lasts approximately 1 week and is self-­limited. In patients who are more immunocompromised, the ­ lesions may ­ulcerate and can be extremely painful. ­Treatment ­typically involves an antiviral medication such as ­acyclovir. Overt bacterial infections are relatively uncommon. Despite the existence of gingivitis or periodontitis in many adults, clinically significant bacterial infections are uncommon. When they do occur, mild infections can be treated with a topical rinse such as chlorhexidine. However, chlorhexidine tends to stain dental plaque, and in some patients who are already experiencing severe soreness from treatment or tumor, an unacceptable burning sensation prevents the use of this drug. In cases of more severe infection, systemic antibiotics are required.

Mucositis and Pain Mucositis is the most common dose-limiting side effect of aggressive altered-fractionation and chemoradiation therapy regimens, and it occurs more often and persists longer after treatment with these regimens than with traditional regimens. Mucositis traditionally has been thought to reflect direct injury of the proliferating cells of the basal epithelium of the irradiated mucosa; however, to some degree, mucositis may be mediated by activation of nuclear factor-κB, which upregulates proinflammatory cytokines (e.g., tumor necrosis factor-α, interleukin-1β), prompting a signaling cascade by upregulation of mitogen-activated protein kinase, cyclooxygenase-2, and tyrosine kinases and activation of matrix metalloproteinases, all of which ultimately produce the injury we recognize as mucositis.166 Irradiation of the parotid glands can also foster mucositis. Christensen and colleagues167 reported a significant decrease in salivary levels of epidermal growth factor after radiation therapy and speculated that the drop in this

276

PART 3  Head and Neck

­ itogenic peptide, implicated in mucosal regeneration, m could contribute to the development of mucositis. Pain, usually caused by mucositis, is common by the midpoint of conventionally fractionated radiation therapy regimens and can last for many weeks after treatment. Some patients find that coating their mouths with olive oil or swallowing a teaspoon of olive oil before they eat and as necessary during meals makes them more comfortable. For mild pain, simple analgesics such as aspirin or acetaminophen crushed and dissolved in water or Aspergum provide relief. For more severe pain, stronger measures are required. Viscous lidocaine (Xylocaine) often provides effective pain relief, but patients tend to experience a burning sensation until the medication takes effect. To ameliorate this sensation, the drug can be dispensed in a mixture with soothing agents (e.g., mixture of Xylocaine, Mylanta, and Benadryl, commonly known as “magic mouthwash”). Systemic narcotics may be necessary.

Nutritional Problems Nutrition tends to suffer as treatment progresses. Patients require nutritional counseling and encouragement. Most patients can tolerate six half-sized meals per day better than three full meals. Nutritional supplements (e.g., Ensure) should be considered between meals. Capable patients can be instructed in the preparation of homemade nutritional supplements that they can vary to suit their taste. Patients who have advanced mucositis essentially have to rely on a soft, bland diet that is served at room temperature without added spices. Beverages should be similarly bland. Some patients fail to realize that oranges and tomatoes are relatively acidic and usually are best avoided during and for some time after treatment. When changing the diet and blending food are insufficient to permit the patient to consume an adequate diet, placement of a percutaneous gastrostomy tube provides an alternative means of supporting the patient’s nutritional requirements during and after radiation therapy. As treatment regimens have become more aggressive, the nutritional demands on patients have become more difficult to satisfy orally, and insertion of percutaneous gastrostomy tubes has become more common. Many chemoradiation therapy regimens are sufficiently aggressive that percutaneous gastrostomy tubes must be placed prophylactically. However, tube feeding may not be an effective functional substitute for good oral nutrition, and encouragement of oral intake to the greatest extent possible seems wise.168

Osteonecrosis and Soft Tissue Necrosis Osteonecrosis (i.e., osteoradionecrosis) and soft tissue necrosis typically involve the mandible and its adjacent tissues and represent a late-appearing manifestation of injury beyond the reparative capacity of the normal cells. Irradiation most likely produces hypocellular, hypovascular, hypoxic tissue, which leads to tissue breakdown and a chronic nonhealing wound.169 The role of infection in fostering ­osteonecrosis is subject to debate. Although the necrosis by itself is nonspecific, the development of periosteal thickening, mottled areas of resorption, osteoporosis, sclerosis, lysis, or fracture of bone occurring 6 months to several years after treatment is generally diagnostic. Osteonecrosis is typically considered a predictable (and to some degree acceptable) risk of brachytherapy

applied close to the mandible. Mazeron and colleagues144 treated 95 evaluable patients who had T1 or T2 cancer of the floor of the mouth with iridium brachytherapy alone. Grade 1 complications (spontaneously healing ­exposure that did not interfere with the patients’ sense of well-­being) ­ occurred in soft tissues 15 times and in bone seven times. Grade 2 complications requiring hospitalization and usually antibiotics and steroids occurred in soft tissues eight times and in bone three times. Grade 3 ­complications requiring surgery or leading to ­substantial disability or death occurred in soft tissues eight times and in bone seven times. However, osteonecrosis also can occur after EBRT. Niewald and colleagues170 reported an 8.6% rate of osteonecrosis after conventionally fractionated EBRT (60 to 70 Gy delivered in one 2-Gy fraction per day) and a 22.9% rate of osteonecrosis after more aggressive ­ hyperfractionated therapy (82.8 Gy delivered in two daily 1.2-Gy fractions given at least 4 hours apart). Particular care needs to be taken for IMRT because of the potential for unintentional “hot spots” within the mandible and the associated increase in risk of ­osteoradionecrosis.171 Although patients who are edentulous may have a more difficult time chewing their food during treatment, they are at a relative advantage in having a reduced risk of osteonecrosis. To some degree, having all patients undergo dental evaluation before radiation therapy is begun can lessen this difference. Healthy teeth should not be extracted even if they will be within the radiation portals; teeth that are irradiated are not markedly more damaged than adjacent nonirradiated teeth. Broken, badly decayed teeth or preexisting gingival disease needs to be corrected before radiation therapy begins. When teeth are so badly decayed that caries cannot be repaired quickly, extraction is appropriate. The socket should be rongeured smooth, and the mucosa overlying the socket should be sutured closed. Approximately 2 weeks should be allowed for healing before radiation therapy is begun. Ideally, the incidence of osteonecrosis could be decreased by waiting 3 weeks after extraction,172 but the need to start treatment as quickly as possible renders 2 weeks a suitable compromise for most patients.

Dental Decay Failure to protect teeth from the changes produced by radiation therapy over time leads to dental decay, which characteristically occurs along the gum line (Fig. 12-28A). Even when decay is relatively far advanced, appropriate intervention and restoration is worthwhile and can salvage some situations (see Fig. 12-28B). Aside from its salutary effect on the patient’s appearance, restoration removes one portal of entry for infection. On the other hand, dental manipulation by itself may be sufficiently traumatic to produce osteonecrosis in previously marginally viable mandibular bone. Prevention of decay is therefore preferable. If, despite all efforts, osteonecrosis does occur, antibiotic therapy and patience should be the first line of ­management. Hyperbaric oxygen arguably may help in some cases by stimulating angiogenesis, increasing neovascularization, and optimizing cellular oxygen levels to enhance osteoblast and fibroblast proliferation.173 Only when it becomes apparent that osteonecrosis will be progressive or persistent should surgical management be undertaken. One prospective, randomized trial has shown that if teeth

12  The Oral Cavity

A

277

B

FIGURE 12-28.  A, Characteristic radiation-induced dental decay occurs predominantly along the gum line. The patient placed fluoride gel in dental trays, but instead of placing the trays over his teeth, he dipped his toothbrush into the trays and then brushed. B, Same patient after dental restoration.

must be extracted from a previously irradiated mandible, hyperbaric oxygen is more successful than penicillin in preventing subsequent osteonecrosis (5% versus 30%).174 Unfortunately, a prospective, multicenter, randomized, double-blind, placebo-controlled trial of hyperbaric oxygen as an adjuvant to surgery (consisting of 30 hyperbaric sessions before surgery plus 10 sessions after surgery) had to be stopped early, after the accrual of 68 patients, because an interim analysis showed the 1-year rate of recovery from osteoradionecrosis to be worse in the hyperbaric oxygen group (19%) than in the placebo group (32%).175

Follow-up Care Head and neck tumors that are not controlled with initial treatment tend to recur within 2 years after radiation therapy. Follow-up care therefore needs to be offered to patients on an intensive basis for the first few years after treatment. For the typical patient who has no complicating factors that require earlier medical intervention, it is prudent to schedule the first follow-up appointment ­ approximately 2 weeks after the completion of treatment. At that time, the acute reaction from radiation therapy should have peaked, and the patient should verify and examination should demonstrate the onset of recovery. As soon as the radiation-­induced soreness subsides, daily fluoride therapy should be initiated to protect against future dental decay. The use of customshaped plastic trays that are fitted over the upper and lower teeth is preferable. A solution of approximately 1% neutral fluoride is prescribed. Patients place three or four drops of fluoride in each tray (enough to just fill the custom trays when fitted over the teeth) and then place the trays over their teeth for 10 min/day every day for the rest of their lives. During the first year of follow-up, patients tend to have concerns because of xerostomia, and many become frightened when a benign submental collection of lymphatic fluid (a dewlap) becomes evident 1 to 4 months after treatment. Patients should be offered palliative agents for xerostomia and should be reassured that the dewlap does not represent recurrent tumor and that its appearance will improve over time without intervention. Patients are seen every 1 to 2 months for the first year.

During the second year, patients tend to feel relatively well, and follow‑up appointments can be spaced to once every 2 to 3 months. However, the period around 18 months after completion of treatment constitutes the time of greatest risk of recurrence, and patients should be dissuaded from thinking that extending the interval between follow‑up examinations suggests that they should become lax in their care. By the third year, most recurrences have already become evident, and follow‑up can be extended to once every 3 to 4 months. After 3 years, the chance of recurrence is very small, and patients can be seen only once or twice per year. At that point, follow‑up care is justified by the need to detect second independent malignancies as early as possible.

Conclusions Radiation oncologists play a major role in the treatment of carcinomas of the oral cavity. The wide range in the size and behavior of lesions challenges physicians to use ­virtually all of their armamentarium to achieve optimal management of lesions. Fortunately, treatment of oral cavity lesions is often successful. The importance of multimodal care in the management of oral cavity carcinomas demands that radiation oncologists have a close working rapport with diagnostic radiologists, surgical pathologists, head and neck surgeons, medical oncologists, and dentists. The possibility of applying recently described principles of tumor biology and pioneered strategies of fractionation and sensitization raises the hope that we now stand on the threshold of better tumor control and decreased cosmetic and functional consequences for these lesions.

REFERENCES 1. American Cancer Society. Cancer Facts and Figures 2007. Atlanta: American Cancer Society, 2007. Available at http://www. cancer.org (accessed January 2009). 2. Borok TL, Cooper JS. Time course and significance of acute hyper­ amylasemia in patients receiving fractionated therapeutic radiation to the parotid gland region. Int J Radiat Oncol Biol Phys 1982;8:1449-1451.

278

PART 3  Head and Neck

  3. Cox JD, Stetz J, Pajak TF. Toxicity criteria of the Radiation Therapy Oncology Group (RTOG) and the European Organization for Research and Treatment of Cancer (EORTC). Int J Radiat Oncol Biol Phys 1995;31:1341-1346.   4. Kujan O, Glenny AM, Duxbury J, et al. Evaluation of screening strategies for improving oral cancer mortality: a Cochrane systematic review. J Dent Educ 2005;69:255-265.   5. US Department of Health and Human Services. The Health Consequences of Smoking: Cancer. A Report of the Surgeon General. DHHS publication no. PHS 82-50179. Rockville, MD: US Department of Health and Human Services, Public Health Service, 1982.   6. Graham S, Dayal H, Roher T, et al. Dentition, diet, tobacco and alcohol in the epidemiology of oral cancer. J Natl Cancer Inst 1977;59:1611-1618.   7. Wynder EL, Bross IJ, Feldman RM. A study of the etiological factors in cancer of the mouth. Cancer 1957;10:1300-1323.   8. Zhang Z-F, Morgenstern H, Spitz MR, et al. Marijuana use and increased risk of squamous cell carcinoma of the head and neck. Cancer Epidemiol Biomarkers Prev 1999;8:1071-1078.   9. Weaver A, Fleming SM, Smith DB, et al. Mouthwash and oral cancer: carcinogen or coincidence? J Oral Surg 1979; 37:250-253. 10. Winn DM, Blot WJ, McLaughlin JK, et al. Mouthwash use and oral conditions in the risks of oral and pharyngeal cancer. Cancer Res 1991;5:3044-3047. 11. Morse DE, Katz RV, Pendrys DG, et al. Smoking and drinking in relation to oral epithelial dysplasia. Cancer Epidemiol Biomarkers Prev 1996;5:769-777. 12. Hirayama T. An epidemiological study of oral and pharyngeal cancer in central and southeast Asia. Bull World Health Organ 1966;34:41-69. 13. Vogler WR, Lloyd WJ, Milmore BK. A retrospective study of aetiological factors in cancer of the mouth, pharynx and larynx. Cancer 1962;15:246-258. 14. Rosenfeld L, Callaway J. Snuff dippers’ cancer. Am J Surg 1963;106:840-844. 15. Pindborg JJ, Mehta FS, Gupta PC, et al. Reverse smoking in Andhra Pradesh, India: a study of palatal lesions among 10,169 villagers. Br J Cancer 1971;25:10-20. 16. Reddy DG, Rao VK. Cancer of the palate in coastal Andhra due to smoking cigars with the burning end inside the mouth. Indian J Med Sci 1957;11:791-798. 17. Dorn HF, Cutler SJ. Morbidity from Cancer in the United States. Parts I and II. Public Health monograph 56, Public Health Service publication no. 590. Washington, DC: Public Health Service, 1959. 18. Scully C. Oral squamous cell carcinoma: from an hypothesis about a virus to concern about possible sexual transmission. Oral Oncol 2002;38:227-234. 19. Smith EM, Ritchies JM, Summersgill KF, et al. Human papillomavirus in oral exfoliated cells and risk of head and neck cancer. J Natl Cancer Inst 2004;96:449-455. 20. Schantz SP, Hsu TC, Ainslie N, et al. Young adults with head and neck cancer express increased susceptibility to mutagen‑induced chromosome damage. JAMA 1989;262:3313-3315. 21. Shiboski CH, Shiboski SC, Silverman S Jr. Trends in oral cancer rates in the United States, 1973-1996. Community Dent Oral Epidemiol 2000;28:249-256. 22. Gillison ML. Current topics in the epidemiology of oral cavity and oropharyngeal cancers. Head Neck 2007;29:779-792. 23. Bundgaard T, Wildt J, Elbrond O. Oral squamous cell cancer in non-users of tobacco and alcohol. Clin Otolaryngol 1994;19: 320-326. 24. Silverman S Jr, Miglioratti C, Barbosa J. Toluidine blue staining in the detection of oral precancerous and malignant lesions. Oral Surg Oral Med Oral Pathol 1984;57:379-382. 25. Daura Sáez A, Rodríguez San Pedro F, Asenjo García B, et al. Magnetic resonance imaging in the diagnosis of oral carcinoma [in Spanish]. Acta Otorrinolaringol Esp 1999;50:283-290. 26. Nakasone Y, Inoue T, Oriuchi N, et al. The role of whole-body FDG-PET in preoperative assessment of tumor staging in oral cancers. Ann Nucl Med 2001;15:505-512.

27. Goerres GW, Schmid DT, Gratz KW, et al. Impact of whole body positron emission tomography on initial staging and therapy in patients with squamous cell carcinoma of the oral cavity. Oral Oncol 2003;39:547-551. 28. Dammann F, Horger M, Mueller-Berg M, et al. Rational diagnosis of squamous cell carcinoma of the head and neck region: comparative evaluation of CT, MRI, and 18FDG PET. AJR Am J Roentgenol 2005;184:1326-1331. 29. Ng SH, Yen TC, Liao CT, et al. 18F-FDG PET and CT/MRI in oral cavity squamous cell carcinoma: a prospective study of 124 patients with histologic correlation. J Nucl Med 2005;46:1136-1143. 30. Stoeckli SJ, Steinert H, Pfaltz M, et al. Is there a role for positron emission tomography with 18F-fluorodeoxyglucose in the initial staging of nodal negative oral and oropharyngeal squamous cell carcinoma. Head Neck 2002;24:345-349. 31. Goerres GW, Schmid DT, Schuknecht B, Eyrich GK. Bone invasion in patients with oral cavity cancer: comparison of conventional CT with PET/CT and SPECT/CT. Radiology 2005;237:281-287. 32. Kunkel M, Forster GJ, Reichert TE, et al. Detection of recurrent oral squamous cell carcinoma by [18F]-2-fluorodeoxyglucosepositron emission tomography: implications for prognosis and patient management. Cancer 2003;98:2257-2265. 33. Robbins KT, Medina JE, Wolfe GT, et al. Standardizing neck dissection terminology. Arch Otolaryngol Head Neck Surg 1991;117:601-605. 34. Robbins KT, Clayman G, Levine PA, et al. Neck dissection classification update: revisions proposed by the American Head and Neck Society and the American Academy of Otolaryngology–Head and Neck Surgery. Arch Otolaryngol Head Neck Surg 2002;128:747-748. 35. Levendag P, Gregoire V, Hamoir M, et al. Intraoperative validation of CT-based lymph nodal levels, sublevels IIa and IIb: is it of clinical relevance in selective radiation therapy? Int J Radiat Oncol Biol Phys 2005;62:690-699. 36. van den Brekel MWM, Castelijns JA. Radiologic evaluation of neck metastases: the otolaryngologist’s perspective. Semin ­Ultrasound CT MR 1999;20:162-174. 37. Som PM, Curtin HD, Mancuso AA. Imaging-based classification for the cervical lymph nodes designed as an adjunct to clinically based nodal classifications. Arch Otolaryngol Head Neck Surg 1999;125:388-396. 38. Black RJ, Gluckman JL, Shumrick DA. Multiple primary tumours of the upper aerodigestive tract. Clin Otolaryngol 1983;8:277-281. 39. Woolgar JA, Scott J. Prediction of cervical lymph node metastasis in squamous cell carcinoma of the tongue/floor of mouth. Head Neck 1995;17:463-472. 40. Lindberg RD. Distribution of cervical lymph node metastases from squamous cell carcinoma of the upper respiratory and digestive tracts. Cancer 1972;29:1446-1450. 41. Mendenhall WM, Million RR, Cassisi NJ. Elective neck irradiation in squamous‑cell carcinoma of the head and neck. Head Neck Surg 1980;3:15-20. 42. August M, Gianetti K. Elective neck irradiation versus observation of the clinically negative neck of patients with oral cancer. J Oral Maxillofac Surg 1996;54:1050-1055. 43. Shah JP. Patterns of cervical lymph node metastasis from squamous carcinomas of the upper aerodigestive tract. Am J Surg 1990;160:405-409. 44. Friedman M, Mafee MF, Pacella BL, et al. Rationale for elective neck dissection in 1990. Laryngoscope 1990;100:54-59. 45. Merino OR, Lindberg RD, Fletcher GH. An analysis of distant metastases from squamous cell carcinoma of the upper respiratory and digestive tracts. Cancer 1977;40:145-151. 46. Leibel SA, Scott CB, Mohiuddin M, et al. The effect of ­ local‑ regional control on distant metastatic dissemination in carcinoma of the head and neck: results of an analysis from the RTOG head and neck database. Int J Radiat Oncol Biol Phys 1991;21:549-556. 47. Greene FL, Page DL, Fleming ID, et al (eds): AJCC Cancer Staging Manual. 6th ed, New York: Springer, 2002, pp. 23-32. 48. Langdon JD, Harvey PW, Rapidis AD, et al. Oral cancer: the behaviour and response to treatment of 194 cases. J Maxillofac Surg 1977;5:221-237. 49. Shafer WG, Waldron CA. Erythroplakia of the oral cavity. Cancer 1975;36:1021-1028.

12  The Oral Cavity 50. Van der Meij EH, Schepman KP, van der Waal I. The possible premalignant character of oral lichen planus and oral lichenoid lesions: a prospective study. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2003;96:164-171. 51. Mignogna MD, Lo Russo L, Fedele S, et al. Clinical behaviour of malignant transforming oral lichen planus. Eur J Oral Surgical Oncol 2002;28:838-843. 52. Lippman SM, Sudbo J, Hong WK. Oral cancer prevention and the evolution of molecular-targeted drug development. J Clin Oncol 2005;23:346-356. 53. Bettendorf O, Piffko J, Bankfalvi A. Prognostic and predictive factors in oral squamous cell cancer: important tools for planning individual therapy? Oral Oncol 2004;40:110-119. 54. Shin DM, Kim J, Ro JY, et al. Activation of p53 gene expression in premalignant lesions during head and neck tumorigenesis. Cancer Res 1994;54:321-326. 55. Nawroz H, van der Riet P, Hruban RH, et al. Allelotype of head and neck squamous cell carcinoma. Cancer Res 1994;54:11521155. 56. van der Riet P, Nawroz H, Hruban RH, et al. Frequent loss of chromosome 9p21-22 early in head and neck cancer progression. Cancer Res 1994;54:1156-1158. 57. Liu SC, Hu Y, Sauter ER, et al. Image analysis of p53 and cyclin D1 expression in premalignant lesions of the oral mucosa. Anal Quant Cytol Histol 1999;21:166-173. 58. Kerdpon D, Rich AM, Reade PC. Expression of p53 in oral mucosal hyperplasia, dysplasia and squamous cell carcinoma. Oral Dis 1997;3:86-92. 59. Kudo Y, Takata T, Ogawa I, et al. Expression of p53 and p21CIP1/ WAF1 proteins in oral epithelial dysplasias and squamous cell carcinomas. Oncol Rep 1999;6:539-545. 60. Jordan RC, Bradley G, Slingerland J. Reduced levels of the cellcycle inhibitor p27Kip1 in epithelial dysplasia and carcinoma of the oral cavity. Am J Pathol 1998;152:585-590. 61. Shintani S, Yoshihama Y, Emilio AR, et al. Overexpression of p53 is an early event in the tumorigenesis of oral squamous cell carcinomas. Anticancer Res 1995;15:305-308. 62. Kushner J, Bradley G, Young B, et al. Aberrant expression of cyclin A and cyclin B1 proteins in oral carcinoma. J Oral Pathol Med 1999;28:77-81. 63. Kotelnikov VM, Coon Js IV, Mundle S, et al. Cyclin D1 expression in squamous cell carcinomas of the head and neck and in oral mucosa in relation to proliferation and apoptosis. Clin Cancer Res 1997;3:95-101. 64. Maragou P, Bazopoulou-Kyrkanidou E, Panotopoulou E, et al. Alteration of integrin expression in oral squamous cell carcinomas. Oral Dis 1999;5:20-26. 65. Xu XC, Ro JY, Lee JS, et al. Differential expression of nuclear retinoid receptors in normal, premalignant, and malignant head and neck tissues. Cancer Res 1994;54:3580-3587. 66. Yamamoto T, Kamata N, Kawano H, et al. High incidence of amplification of the epidermal growth factor receptor gene in human squamous carcinoma cell lines. Cancer Res 1986;46:414-416. 67. Scambia G, Panici PB, Battaglia F, et al. Receptors for epidermal growth factor and steroid hormones in primary laryngeal tumors. Cancer 1991;67:1347-1351. 68. Santini J, Formento JL, Francoual M, et al. Characterization, quantification, and potential clinical value of the epidermal growth factor receptor in head and neck squamous cell carcinomas. Head Neck 1991;13:132-139. 69. Kannan S, Balaram P, Chandran GJ, et al. Co-expression of ras p21 and epidermal growth factor receptor during various stages of tumour progression in oral mucosa. Tumour Biol 1994; 15:73-81. 70. Maiorano E, Favia G, Maisonneuve P, et al. Prognostic implications of epidermal growth factor receptor immunoreactivity in squamous cell carcinoma of the oral mucosa. J Pathol 1998;185:167174. 71. Sudbo J. Novel management of oral cancer: a paradigm of predictive oncology. Clin Med Res 2004;2:233-242. 72. Hasina R, Lingen MW. Angiogenesis in oral cancer. J Dent Educ 2001;65:1282-1290.

279

73. Thomas GJ, Speight PM. Cell adhesion molecules and oral cancer. Crit Rev Oral Biol Med 2001;12:479-498. 74. Ziober BL, Silverman SS Jr, Kramer RH. Adhesive mechanisms regulating invasion and metastasis in oral cancer. Crit Rev Oral Biol Med 2001;12:499-510. 75. Aggelopoulou EP, Skarlos D, Papadimitriou C, et al. Human papilloma virus DNA detection in oral lesions in the Greek population. Anticancer Res 199;19:1391-1395. 76. Sugerman PB, Shillitoe EJ. The high risk human papillomaviruses and oral cancer: evidence for and against a causal relationship. Oral Dis 1997;3:130-147. 77. Close LG, Brown PM, Vuitch MF, et al. Microvascular invasion and survival in cancer of the oral cavity and oropharynx. Arch Otolaryngol Head Neck Surg 1989;115:1304-1309. 78. Lopes MA, Kowalski LP, da Cunha Santos G, et al. A clinicopathologic study of 196 intraoral minor salivary gland tumours. J Oral Pathol Med 1999;28:264-267. 79. Henk JM. Treatment of oral cancer by interstitial irradiation ­ using iridium-192. Br J Oral Maxillofac Surg 1992;30: 355-359. 80. Tupchong L, Scott CB, Blitzer PH, et al. Randomized study of preoperative versus postoperative radiation therapy in advanced head and neck carcinoma: long‑term follow‑up of RTOG study 73‑03. Int J Radiat Oncol Biol Phys 1991;20:21-28. 81. Cooper JS, Pajak TF, Forastiere A, et al. Precisely defining highrisk operable head and neck tumors based on RTOG #85-03 and #88-24: targets for postoperative radiochemotherapy? Head Neck 1998;20:588-594. 82. Bachaud JM, Cohen-Jonathan E, Alzieu C, et al. Combined postoperative radiotherapy and weekly cisplatin infusion for locally advanced head and neck carcinoma: final report of a randomized trial. Int J Radiat Oncol Biol Phys 1996;36:999-1004. 83. Haffty B, Son Y, Sasaki C, et al. Mitomycin C as an adjunct to post-operative radiation therapy in squamous cell carcinoma of the head and neck: results from two randomized clinical trials. Int J Radiat Oncol Biol Phys 1993;27:241-250. 84. Smid L, Budihna M, Zakotnik B, et al. Postoperative concomitant irradiation and chemotherapy with mitomycin C and bleomycin for advanced head-and-neck carcinoma. Int J Radiat Oncol Biol Phys 2003;56:1055-1062. 85. Bernier J, Domenge C, Ozsahin M, et al. Postoperative irradiation with or without concomitant chemotherapy for locally advanced head and neck cancer. N Engl J Med 2004;350:1945-1952. 86. Cooper JS, Pajak TF, Forastiere AA, et al. Postoperative concurrent radiotherapy and chemotherapy for high-risk squamous-cell carcinoma of the head and neck. N Engl J Med 2004;350:19371944. 87. Bernier J, Cooper JS, Pajak TF, et al. Defining risk levels in locally advanced head and neck cancers: a comparative analysis of concurrent postoperative radiation plus chemotherapy trials of the EORTC (#22931) and RTOG (#9501). Head Neck 2005;27:843850. 88. Fang FM, Leung SW, Huang CC, et al. Combined-modality therapy for squamous carcinoma of the buccal mucosa: treatment results and prognostic factors. Head Neck 1997;19:506-512. 89. Lapeyre M, Peiffert D, Malissard LH. An original technique of brachytherapy in the treatment of epidermoid carcinomas of the buccal mucosa. Int J Radiat Oncol Biol Phys 1995;33: 447-454. 90. Lambin P, Haie‑Meder C, Gerbaulet A, et al. Curietherapy versus external irradiation combined with curietherapy in stage II squamous cell carcinoma of the mobile tongue. Radiother Oncol 1992;23:55-56. 91. Wang CC. Radiotherapeutic management and results of T1N0, T2N0 carcinoma of the oral tongue: evaluation of boost techniques. Int J Radiat Oncol Biol Phys 1989;17:287-291. 92. Wang CC. Intraoral cone for carcinoma of the oral cavity. Front Radiat Ther Oncol 1991;25:128-131. 93. Wendt CD, Peters LJ, Delclos L, et al. Primary radiotherapy in the treatment of stage I and II oral tongue cancers: importance of the proportion of therapy delivered with interstitial therapy. Int J Radiat Oncol Biol Phys 1990;18:1287-1292.

280

PART 3  Head and Neck

  94. Mazeron JJ, Grimard L, Benk V. Curietherapy versus external irradiation combined with curietherapy in stage II squamous cell carcinomas of mobile tongue and floor of mouth. Recent Results Cancer Res 1994;134:101-110.   95. Leung TW, Wong VY, Wong CM, et al. Technical hints for high dose rate interstitial tongue brachytherapy. Clin Oncol (R Coll Radiol) 1998;10:231-236.   96. Klein M, Menneking H, Langford A, et al. Treatment of squamous cell carcinomas of the floor of the mouth and tongue by interstitial high-dose-rate irradiation using iridium-192. Int J Oral Maxillofac Surg 1998;27:45-48.   97. McGuirt WF Jr, Johnson JT, Myers EN, et al. Floor of mouth carcinoma. The management of the clinically negative neck. Arch Otolaryngol Head Neck Surg 1995;121:278-282.   98. Vikram B, Strong EW, Shah JP, et al. Intraoperative radiotherapy in patients with recurrent head and neck cancer. Am J Surg 1985;150:485-487.   99. Spencer SA, Wheeler RH, Peters GE, et al. Concomitant chemotherapy and reirradiation as management for recurrent cancer of the head and neck. Am J Clin Oncol 1999;22:1-5. 100. Hong WK, Endicott J, Itri LM, et al. 13-Cis-retinoic acid in the treatment of oral leukoplakia. N Engl J Med 1986;315:1501-1505. 101. Hong WK, Lippman SM, Itri LM, et al. Prevention of second primary tumors with isotretinoin in squamous‑cell carcinoma of the head and neck. N Engl J Med 1990;323:795-801. 102. Khuri FR, Lee JJ, Lippman SM, et al. Randomized phase III trial of low-dose isotretinoin for prevention of second primary tumors in stage I and II head and neck cancer patients. J Natl Cancer Inst 2006;98:441-450. 103. Lozza L, Cerrotta A, Gardani G, et al. Analysis of risk factors for mandibular bone radionecrosis after exclusive low dose-rate brachytherapy for oral cancer. Radiother Oncol 1997;44:143-147. 104. Simon JM, Mazeron JJ, Pohar S, et al. Effect of intersource ­spacing on local control and complications in brachytherapy of mobile tongue and floor of mouth. Radiother Oncol 1993;26:19-25. 105. Pierquin B, Wilson JF, Chassagne D. Basic techniques of endocurietherapy. In: Modern Brachytherapy. New York: Masson, 1987. 106. Paterson R, Parker HM. A dosage system for gamma‑ray therapy. Br J Radiol 1934;7:592-633. 107. Pierquin B, Dutreix A. Pour une nouvelle methodologie en curietherapie: le Systeme de Paris (endo et plesio‑radiotherapie avec preparation non radio‑active). Ann Radiol 1966;9:757-760. 108. Pierquin B, Dutreix A, Paine CH, et al. The Paris system in interstitial radiation therapy. Acta Radiol Oncol 1978;17:33-48. 109. Nag S, Martinez-Monge R, Zhang H, et al. Simplified non­looping functional loop technique for HDR brachytherapy. ­Radiother Oncol 1998;48:339-341. 110. Yuasa K, Kawazu T, Morita M, et al. A new, simple method of making a spacer in interstitial brachytherapy for mobile tongue cancer. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2000;89:519-521. 111. Obinata K, Ohmori K, Tuchiya K, et al. Clinical study of a spacer to help prevent osteoradionecrosis resulting from brachytherapy for tongue cancer. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2003;95:246-250. 112. Cederbaum M, Zalik M, Rosenblatt E, et al. A simple oral cone attachment for the Varian linear accelerator. Med Dosim 1998;23:47-50. 113. Fu KK, Pajak TF, Trotti A, et al. A Radiation Therapy Oncology Group (RTOG) phase III randomized study to compare hyperfractionation and two variants of accelerated fractionation to standard fractionation radiotherapy for head and neck squamous cell carcinomas: preliminary results of RTOG 9003. Int J Radiat Oncol Biol Phys 2000;48:7-16. 114. Brizel DM, Albers ME, Fisher SR, et al. Hyperfractionated irradiation with or without concurrent chemotherapy for locally advanced head and neck cancer. N Engl J Med 1998;338: 1798-1804. 115. Milas L, Mason KA, Liao Z, et al. Chemoradiotherapy: emerging treatment improvement strategies. Head Neck 2003;25:152167.

116. Ang KK, Harris J, Garden AS, et al. Concomitant boost radiation plus concurrent cisplatin for advanced head and neck carcinomas: radiation therapy oncology group phase II trial 99-14. J Clin Oncol 2005;23:3008-3015. 117. Fletcher GH. Clinical dose‑response curves of human malignant epithelial tumours. Br J Radiol 1973;46:1-12. 118. Peters LJ, Goepfert H, Ang KK, et al. Evaluation of the dose for postoperative radiation therapy of head and neck cancer: first report of a prospective randomized trial. Int J Radiat Oncol Biol Phys 1993;26:3-11. 119. Claus F, Duthoy W, Boterberg T, et al. Intensity modulated radiation therapy for oropharyngeal and oral cavity tumors: clinical use and experience. Oral Oncol 2002;38:597-604. 120. Schwartz DL, Ford EC, Rajendran J, et al. FDG-PET/CT-guided intensity modulated head and neck radiotherapy: a pilot investigation. Head Neck 2005;27:478-487. 121. Cooper JS, Pajak TF, Rubin P, et al. Second malignancies in patients who have head and neck cancers: incidence, effect on survival and implications for preventive medicine based on the RTOG experience. Int J Radiat Oncol Biol Phys 1989; 17:449-456. 122. Kramer S, Gelber RD, Snow JB, et al. Combined radiation therapy and surgery in the management of advanced head and neck cancer: final report of study 73-03 of the Radiation Therapy Oncology Group. Head Neck Surg 1987;10:19-30. 123. Griffin TW, Pajak TF, Gillespie BW, et al. Predicting the response of head and neck cancers to radiation therapy with a multivariate modeling system: an analysis of the RTOG head and neck registry. Int J Radiat Oncol Biol Phys 1984;10:481-487. 124. Zelefsky MJ, Harrison LB, Armstrong JG. Long-term treatment results of postoperative radiation therapy for advanced stage oropharyngeal carcinoma. Cancer 1992;70:2388-2395. 125. Mishra RC, Parida G, Mishra TK, et al. Tumour thickness and relationship to locoregional failure in cancer of the buccal mucosa. Eur J Surg Oncol 1999;25:186-189. 126. Bentzen SM, Johansen LV, Overgaard J, et al. Clinical radiobiology of squamous cell carcinoma of the oropharynx. Int J Radiat Oncol Biol Phys 1991;20:1197-1206. 127. Sobel S, Rubin P, Keller B, et al. Tumor persistence as a predictor of outcome after radiation therapy of head and neck cancers. Int J Radiat Oncol Biol Phys 1976;1:873-880. 128. Marcial V, Amato D, Pajak T. Patterns of failure after treatment for cancer of the upper respiratory and digestive tracts: a Radiation Therapy Oncology Group report. Cancer Treat Symp 1983;2:33-40. 129. Wallner PE, Hanks GE, Kramer S. Patterns of care study: analysis of outcome survey data—anterior two‑thirds of tongue and floor of mouth. Am J Clin Oncol 1986;9:50-57. 130. Beauvois S, Hoffstetter S, Peiffert D, et al. Brachytherapy for lower lip epidermoid cancer: tumoral and treatment factors influencing recurrences and complications. Radiother Oncol 1994;33:195-203. 131. Fongione S, Signor M, Beorchia A. Interstitial brachytherapy in carcinoma of the lip. Case histories and results [in Italian]. Radiol Med (Torino) 1994;88:657-660. 132. Orecchia R, Rampino M, Gribaudo S, et al. Interstitial brachytherapy for carcinomas of the lower lip. Results of treatment. Tumori 1991;77:336-338. 133. Tombolini V, Bonanni A, Valeriani M, et al. Brachytherapy for squamous cell carcinoma of the lip. The experience of the Institute of Radiology of the University of Rome “La Sapienza.” Tumori 1998;84:478-482. 134. Finestres-Zubeldia F, Guix-Melcior B, Cloquell-Damian A, et al. Treatment of the carcinoma of the lip through high dose rate brachytherapy. Med Oral Pathol Oral Cir Bucal 2005;10:21-24. 135. Guinot JL, Arribas L, Chust ML, et al. Lip cancer treatment with high dose rate brachytherapy. Radiother Oncol 2003; 69:113-115. 136. Strome SE, To W, Strawderman M, et al. Squamous cell carcinoma of the buccal mucosa. Otolaryngol Head Neck Surg 1999;120:375-379.

12  The Oral Cavity 137. Sakai M, Hatano K, Sekiya Y, et al. Radiotherapy for carcinoma of the buccal mucosa: analysis of prognostic factors [in Japanese]. Nippon Igaku Hoshasen Gakkai Zasshi 1998;58:705-711. 138. Kryst L, Piekarczyk J, Miosek K, et al. Results of treatment of upper gingiva and palate carcinomas [in Polish]. Czas Stomatol 1990;43:62-67. 139. Cengiz M, Ozyar E, Ersu B, et al. High-dose-rate mold brachytherapy of early gingival carcinoma: a clinical report. J Prosthet Dent 1999;82:512-514. 140. Fuchihata H, Furukawa S, Murakami S, et al. Results of combined external irradiation and chemotherapy of bleomycin or peplomycin for squamous cell carcinomas of the lower gingiva. Int J Radiat Oncol Biol Phys 1994;29:705-709. 141. Lo K, Fletcher GH, Byers RM, et al. Results of irradiation in the squamous cell carcinomas of the anterior faucial pillar–­retromolar trigone. Int J Radiat Oncol Biol Phys 1987;13:969-974. 142. Marsiglia H, Haie-Meder C, Sasso G, et al. Brachytherapy for T1-T2 floor-of-the-mouth cancers: the Gustave-Roussy Institute experience. Int J Radiat Oncol Biol Phys 2002;52:1257-1263. 143. Pernot M, Hoffstetter S, Peiffert D, et al. Epidermoid carcinomas of the floor of mouth treated by exclusive irradiation: statistical study of a series of 207 cases. Radiother Oncol 1995;35:177185. 144. Mazeron JJ, Grimard L, Raynal M, et al. Iridium-192 curietherapy for T1 and T2 epidermoid carcinomas of the floor of mouth. Int J Radiat Oncol Biol Phys 1990;18:1299-1306. 145. Fein DA, Mendenhall WM, Parsons JT, et al. Carcinoma of the oral tongue: a comparison of results and complications of treatment with radiotherapy and/or surgery. Head Neck 1994;16:358365. 146. Umeda M, Komatsubara H, Ojima Y, et al. A comparison of brachytherapy and surgery for the treatment of stage I-II squamous cell carcinoma of the tongue. Int J Oral Maxillofac Surg 2005;34:739-744. 147. Shibuya H, Hoshina M, Takeda M, et al. Brachytherapy for stage I & II oral tongue cancer: an analysis of past cases focusing on control and complications. Int J Radiat Oncol Biol Phys 1993;26:51-58. 148. Bourgier C, Coche-Dequeant B, Fournier C, et al. Exclusive low-dose-rate brachytherapy in 279 patients with T2N0 mobile tongue carcinoma. Int J Radiat Oncol Biol Phys 2005; 63:434-440. 149. Pernot M, Malissard L, Hoffstetter S, et al. The study of tumoral, radiobiological, and general health factors that influence results and complications in a series of 448 oral tongue carcinomas treated exclusively by irradiation. Int J Radiat Oncol Biol Phys 1994;29:673-679. 150. Mazeron JJ, Crook JM, Benck V, et al. Iridium 192 implantation of T1 and T2 carcinomas of the mobile tongue. Int J Radiat Oncol Biol Phys 1990;19:1369-1376. 151. Yoshida K, Koizumi M, Inoue T, et al. Radiotherapy of early tongue cancer in patients less than 40 years old. Int J Radiat Oncol Biol Phys 1999;45:367-371. 152. Fujita M, Hirokawa Y, Kashiwado K, et al. Interstitial brachytherapy for stage I and II squamous cell carcinoma of the oral tongue: factors influencing local control and soft tissue complications. Int J Radiat Oncol Biol Phys 1999;44:767-775. 153. Mendenhall WM, Van Cise WS, Bova FJ, et al. Analysis of timedose factors in squamous cell carcinoma of the oral tongue and floor of mouth treated with radiation therapy alone. Int J Radiat Oncol Biol Phys 1981;7:1005-1011. 154. Leung TW, Wong VY, Kwan KH, et al. High dose rate brachytherapy for early stage oral tongue cancer. Head Neck 2002; 24:274-281. 155. Inoue T, Inoue T, Yoshida K, et al. Phase III trial of high- vs. low-dose-rate interstitial radiotherapy for early mobile tongue cancer. Int J Radiat Oncol Biol Phys 2001;51:171-175. 156. Umeda M, Komatsubara H, Nishimatsu N, et al. High-dose rate interstitial brachytherapy for stage I-II tongue cancer. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2000;90:667-670.

281

157. National Cancer Institute. Common Terminology Criteria for Adverse Events v3.0 (CTCAE). at http://ctep.cancer.gov/ protocol/Development/electronic_applications/docs/ ctcaev3.pdf (accessed April 2009). 158. Kaneko M, Shirato H, Nishioka T, et al. Scintigraphic evaluation of long-term salivary function after bilateral whole parotid gland irradiation in radiotherapy for head and neck tumour. Oral Oncol 1998;34:140-146. 159. Funegard U, Franzen L, Ericson T, et al. Parotid saliva composition during and after irradiation of head and neck cancer. Eur J Cancer B Oral Oncol 1994;30:230-233. 160. Epstein JB, Emerton S, Kolbinson DA, et al. Quality of life and oral function following radiotherapy for head and neck cancer. Head Neck 1999;21:1-11. 161. Eisbruch A, Ten Haken RK, Kim HM, et al. Dose, volume, and function relationships in parotid salivary glands following conformal and intensity-modulated irradiation of head and neck cancer. Int J Radiat Oncol Biol Phys 1999;45:577-587. 162. Beer KT. Campaign against radio-xerostomia [in German]. Ther Umsch 1998;55:453-455. 163. Valdez IH, Wolff A, Atkinson JC, et al. Use of pilocarpine during head and neck radiation therapy to reduce xerostomia and salivary dysfunction. Cancer 1993;71:1848-1851. 164. Brizel DM, Wasserman TH, Henke M, et al. Phase III randomized trial of amifostine as a radioprotector in head and neck cancer. J Clin Oncol 2000;18:3339-3345. 165. Head and Neck Cancer Disease Site Group. Hodson DI, Browman GP, Thephamongkhol K, et al. The role of amifostine as a radioprotectant in the management of patients with squamous cell head and neck cancer [full report]. Toronto (ON): Cancer Care Ontario (CCO); 2004 Mar [online update]. 22 p. (Practice quideline report: no. 5-8). [26 references] Available at http://www.guideline.gov/summary/pdf.aspx?doc_ id=5273&stat=1&string= (accessed April 2009). 166. Sonis ST. The pathobiology of mucositis. Nat Rev Cancer 2004;4:277-284. 167. Christensen ME, Hansen HS, Poulsen SS, et al. Immunohistochemical and quantitative changes in salivary EGF, amylase and haptocorrin following radiotherapy for oral cancer. Acta Otolaryngol 1996;116:137-143. 168. Rabinovitch R, Grant B, Berkey BA, et al. Impact of nutrition support on treatment outcome in patients with locally advanced head and neck squamous cell cancer treated with definitive radiotherapy: a secondary analysis of RTOG trial 90-03. Head Neck 2006;28:287-296. 169. Marx RE. Osteoradionecrosis: a new concept of its pathophysiology. J Oral Maxillofac Surg 1983;41:283-288. 170. Niewald M, Barbie O, Schnabel K, et al. Risk factors and doseeffect relationship for osteoradionecrosis after hyperfractionated and conventionally fractionated radiotherapy for oral cancer. Br J Radiol 1996;69:847-851. 171. Parliament M, Alidrisi M, Munroe M, et al. Implications of radiation dosimetry of the mandible in patients with carcinomas of the oral cavity and nasopharynx treated with intensity modulated radiation therapy. Int J Oral Maxillofac Surg 2005; 34:114-121. 172. Marx RE, Johnson RP. Studies in the radiobiology of osteoradionecrosis and their clinical significance. Oral Surg Oral Med Oral Pathol 1987;64:379-390. 173. Myers RA, Marx RE. Use of hyperbaric oxygen in postradiation head and neck surgery. NCI Monogr 1990;9:151-157. 174. Marx RE, Johnson RP, Kline SN. Prevention of osteoradionecrosis: a randomized prospective clinical trial of hyperbaric oxygen versus penicillin. J Am Dent Assoc 1985;111:49-54. 175. Annane D, Depondt J, Aubert P, et al. Hyperbaric oxygen therapy for radionecrosis of the jaw: a randomized, placebo­controlled, double-blind trial from the ORN96 study group. J Clin Oncol 2004;22:4893-4900.

13 The Larynx and Hypopharynx ADAM S. GARDEN ANATOMY Larynx Hypopharynx PATTERNS OF TUMOR SPREAD Tumors of the Glottis Tumors of the Supraglottis Tumors of the Subglottis Tumors of the Hypopharynx Lymphatic Metastases Distant Metastases PATHOLOGY CLINICAL MANIFESTATIONS Lesions of the Glottis Lesions of the Supraglottis Lesions of the Subglottis Lesions of the Hypopharynx PATIENT EVALUATION STAGING Larynx Hypopharynx

Anatomy Larynx The larynx is the organ responsible for speech and airway protection. It consists of cartilage, joints, ligaments, muscles, and mucosa, all of which interact to permit its important functions. Although the larynx is not critical for survival, its functions are important for good quality of life. The larynx is often the benchmark in oncologic studies examining the role of organ preservation. Most of the larynx is within and connected to the hypopharynx (Fig. 13-1). Inferiorly, the larynx is connected to the trachea at the same level at which the hypopharynx is connected to the esophagus. Posteriorly, the larynx lies anterior to the posterior pharyngeal wall, prevertebral fascia and muscles, and cervical vertebral bodies. Lateral to the larynx are the lateral walls of the pharynx. Anteriorly, the larynx is relatively superficial, and the anterior commissure can lie within 1 cm of the skin surface (Fig. 13-2). Inferolaterally, the larynx is very close to the thyroid gland. This can be of concern when managing diseases of the thyroid because the recurrent laryngeal nerve, which innervates most of the laryngeal muscles, lies adjacent to the medial portion of each lobe of the gland. The larynx is divided into three regions: the supraglottis, glottis, and subglottis. The natural history and optimal management of laryngeal tumors differ depending on the region in which the tumor arises. The supraglottis lies above the level where the mucosa of the upper surface of the true vocal cords turns upward to form the lateral wall of the

282

SELECTION OF TREATMENT Early-Stage Disease Advanced Disease Carcinoma of the Glottis Carcinoma of the Supraglottis Carcinoma of the Subglottis Carcinoma of the Hypopharynx RADIATION THERAPY TECHNIQUES Beam Energies Simulation and Field Arrangements Treatment Volumes and Radiation Fields Dose and Fractionation RESULTS OF RADIATION THERAPY Carcinoma of the Glottis Carcinoma of the Supraglottis Carcinoma of the Hypopharynx COMPLICATIONS

ventricle. The supraglottis consists of the epiglottis, aryepiglottic folds, arytenoids, and false vocal cords; essentially, the lateral walls of the aryepiglottic folds are the medial walls of the pyriform sinuses (Fig. 13-3). The epiglottis is a leafshaped structure. Superiorly, it sits in air within the inferior oropharynx and coordinates with the base of the tongue to permit proper swallowing. The midepiglottis is connected to the tongue by two membranes, the valleculae, one on either side. The inferior epiglottis becomes a stalk, the petiolus epiglottidis, which inserts into the anterior aspect of the thyroid cartilage. Just inferior to this is the anterior commissure, which is formed by the joining of the two vocal ligaments. For staging purposes, the epiglottis is often further divided into a suprahyoid and an infrahyoid epiglottis, although this is not a true anatomic division. The glottis consists of the true vocal cords and the region extending 0.5 cm inferiorly below the true vocal cords (Fig 13-4). The subglottis, the third subsite of the larynx, extends from the lower boundary of the glottis to the superior aspect of the trachea. The laryngeal framework consists of nine cartilages. The largest are the epiglottic, thyroid, and cricoid cartilages. Three pairs of smaller cartilages—the arytenoid, cuneiform, and corniculate cartilages—complete the cartilaginous composition of the larynx. The arytenoid, thyroid, and cricoid cartilages are composed of hyaline cartilage, which in most people undergoes calcification or ossification beginning around 20 years of age. This modification often aids in the radiographic localization of these structures. The pattern of ossification in the anterior aspect of the thyroid

13  The Larynx and Hypopharynx

283

3

1

4 2

5

FIGURE 13-1.  Photograph of the larynx and hypopharynx.

FIGURE 13-3.  Photograph of the structures of the supraglottic larynx: 1, true vocal cord; 2, false vocal cord; 3, arytenoid; 4, aryepiglottic fold; 5, infrahyoid epiglottis.

laminae. Posteriorly, the cartilage extends upward and downward to form the superior and inferior horns of the thyroid cartilage. Inferior to the thyroid cartilage lies the cricoid cartilage, a complete ring. The inferior border of the cricoid cartilage marks the inferior extent of the larynx and hypopharynx and the beginning of the trachea and esophagus.

Hypopharynx

FIGURE 13-2.  Axial CT view of the larynx at the true vocal cords.

cartilage is often referred to as figure-8 ossification because of its unique appearance on lateral radiographs. The thyrohyoid membrane connects the hyoid bone and thyroid cartilage. Schematically, on a sagittal view, the hyoid bone, infrahyoid epiglottis, and thyrohyoid membrane form a triangular space filled with fat that is referred to as the pre-epiglottic space. Because the pre-epiglottic space is filled only with fat, it offers little resistance to laryngeal tumors. The thyroid cartilage is a V-shaped structure formed by the fusion of two laminae. The anterior borders of these laminae diverge superiorly to form the thyroid notch. The inferior constrictors of the pharynx attach to the

The hypopharynx extends from the hyoid bone superiorly to the cricoid cartilage inferiorly. It is the most inferior of the anatomic subdivisions of the pharynx, beginning at the inferior border of the oropharynx and eventually connecting with the cervical esophagus at the cricopharyngeal muscle. The hypopharynx has three anatomic subsites: the pyriform sinuses, the pharyngeal walls, and the postcricoid region. The pyriform sinuses are formed by the invagination of the larynx into the hypopharynx. Each sinus consists of three walls: a lateral wall; a medial wall, which superiorly becomes the aryepiglottic fold; and an anterior wall (Fig. 13-5). Posteriorly, each pyriform sinus is open. Superior to each pyriform sinus is the vestibule formed by the rim of the three walls. Inferiorly, the three walls taper to form the apex. The pharyngeal walls of the hypopharynx are indistinguishable from the walls of the oropharynx and are defined by their relationship with neighboring anatomic structures. The separation between the oropharyngeal and hypopharyngeal walls is at the valleculae, pharyngoepiglottic folds, and lateral projections of the aryepiglottic folds. Tumors originating from the hypopharyngeal walls are uncommon. It is unusual to see a lesion confined to and clearly originating from the lateral or posterior hypopharyngeal wall. Such tumors often extend superiorly to involve the ­oropharyngeal walls, and the site of origin is often ­imprecise

284

PART 3  Head and Neck

B

A

FIGURE 13-4.  Photographs show direct views of the larynx. A, Partial abduction of the true vocal cords reveals the anterior subglottis and superior tracheal rings. B, Fully adducted and opposed true cords reveal the anterior commissure and infrahyoid epiglottis laterally and the false vocal cords and aryepiglottic folds laterally and superiorly. Posterior to the true cords are the arytenoids, and lateral to them are the medial walls of both pyriform sinuses.

*

A

B

FIGURE 13-5.  A, Photograph of a direct view of the pyriform sinus. B, Axial CT view of the pyriform sinus (asterisk).

and determined by the location of the bulk of disease. Posterior wall tumors can invade the prevertebral tissues or even more deeply to the bones of the cervical spine. The postcricoid region is the region of the hypopharynx posterior to the larynx, beginning at the level of the arytenoids superiorly and continuing until the transition into the esophagus. The anterior border of the postcricoid region is formed by the posterior larynx, and the posterior border of the postcricoid region is formed by the posterior pharyngeal wall.

Patterns of Tumor Spread Tumors of the Glottis Although the glottis is the smallest component of the larynx, most laryngeal cancer originates in this region on the true vocal cords. Most tumors of the true vocal cords arise

on the free margin and the upper surface of the anterior two thirds of the vocal cords (Fig. 13-6). Lesions often remain confined to the mucosa of the true cords, but untreated lesions can eventually spread into the supraglottic or subglottic larynx. Even small lesions can cause hoarseness by limiting the vibratory ability of the cord or the ability of the cord to completely adduct and oppose the contralateral cord. If lesions of the true vocal cords are left unattended, they can invade into the intrinsic musculature and joints of the larynx, resulting in decreased vocal cord mobility and, ultimately, vocal cord fixation. Exophytic lesions can cause airway obstruction (Fig. 13-7), necessitating a tracheostomy before definitive oncologic therapy. Anterior lesions can extend from the commissure directly into the thyroid cartilage and eventually into the neck (Fig. 13-8).

13  The Larynx and Hypopharynx

285

FIGURE 13-8.  CT scan of an advanced carcinoma of the glottis that is obstructing the airway and invading the neck.

FIGURE 13-6.  Photograph of a tumor of the true vocal cord.

FIGURE 13-7.  Photograph of an exophytic tumor of the true vocal cords that is causing obstruction of the airway.

Tumors of the Supraglottis Tumors of the supraglottic larynx account for approximately 25% to 33% of all laryngeal neoplasms. Suprahyoid epiglottic tumors can grow in an exophytic pattern off the tip of the epiglottis or spread inferiorly to infiltrate and ulcerate the epiglottic cartilage (Fig. 13-9). Because of their destructive nature, suprahyoid epiglottic tumors can, even with ­effective tumoricidal therapy, affect the ability of the epiglottis to protect the tracheal airway, resulting in aspiration. Infrahyoid epiglottic tumors can invade directly through the base of the epiglottis into the pre-epiglottic space, onto the lingual epiglottis. If they remain undetected, infrahyoid epiglottic tumors can invade the valleculae, pharyngoepiglottic folds, and tongue base. Sometimes they spread onto the aryepiglottic folds, where they can create a semicircular formation. They then spread inferiorly onto the false vocal cords or into the anterior commissure and glottic larynx.

Lesions originating on the aryepiglottic folds can spread onto the other components of the supraglottic larynx, including the epiglottis, arytenoids, and false vocal cords. These tumors can also spread laterally into the pyriform sinuses, and distinguishing the true epicenter (aryepiglottic fold versus medial pyriform sinus) can be difficult (Fig. 13-10). Lesions originating on the aryepiglottic folds are often infiltrative, and they have easy access to the paraglottic space, which lies between the external frame of the larynx and its internal components. The paraglottic space is filled with adipose tissue and offers little resistance to growing tumors. From this space, infiltrative tumors can break through ligaments above or below the thyroid cartilage into the neck or invade the cartilage directly. After invasion into the paraglottic space takes place, the mobility of the larynx often becomes impaired. Like lesions of the aryepiglottic folds, lesions of the false vocal cords face few barriers to their spreading. Extending inferiorly, they can invade the glottic larynx, and like aryepiglottic-fold lesions they can easily invade the paraglottic space, laryngeal musculature, and nerves to cause impaired or fixed mobility of the hemilarynx.

Tumors of the Subglottis Subglottic tumors are uncommon, accounting for less than 1% of all laryngeal cancer. These tumors can spread inferiorly to the trachea, extend through the cricothyroid membrane into the neck, or directly invade the cricoid cartilage. Distinguishing a tumor originating from the subglottis from a tumor originating from the true vocal cords can be difficult because subglottic tumors often involve the undersurface of the glottic larynx. Typically, laryngeal tumors are thought to originate from the subglottis if the bulk of disease is below the true vocal cords and the upper surfaces of the true cords are free of tumor. Identifying the subsite of origin is only of academic interest and should not affect management decisions.

Tumors of the Hypopharynx Pyriform sinus tumors account for 70% of cancer that originates in the hypopharynx. Unlike glottic tumors, pyriform sinus tumors often are diagnosed in advanced stages

286

PART 3  Head and Neck

A

B

FIGURE 13-9.  Photographs of tumors of the epiglottis show an exophytic tumor (A) and an infiltrative tumor (B).

FIGURE 13-10.  Photograph of large tumor presumed to originate from the supraglottis but with significant involvement of the hypopharynx.

FIGURE 13-11.  CT scan shows a carcinoma of the pyriform sinus  (arrow) with invasion of the larynx and neck.

(Fig. 13-11). Medial wall tumors have few barriers to tumor extension posteriorly along the sinus walls and into the postcricoid space. These tumors easily infiltrate the larynx, often resulting in fixation. Lateral wall tumors also have few barriers to growth. They can extend along the mucosa to involve the posterior hypopharyngeal wall. These tumors often fill the apex and extend deep to the sinus submucosally to involve other adjacent structures, including the laryngeal framework or the thyroid gland. Lateral wall tumors can eventually invade directly into the neck. Submucosal extension can continue inferiorly into the cervical esophagus, and it is sometimes difficult to identify precisely the extent of disease even with direct visualization and radiographic imaging. The patterns of spread can be similar to those of esophageal tumors, with extensive spread along lymphatic spaces and with skip lesions. Postcricoid tumors are rare and almost always are diagnosed in advanced stages. They can extend laterally, into the pyriform sinuses, or anteriorly, directly invading laryngeal structures.

The lymphatic drainage patterns differ significantly among the subsites of the larynx and the hypopharynx. The true vocal cords have a sparse lymphatic supply and therefore have a low propensity for lymphatic spread. The incidence of lymphadenopathy at diagnosis in patients with glottic carcinomas is low and probably increases only with increasing stage as a manifestation of the tumor gaining access to lymphatics beyond the true cords. Less than 2% of patients with T1 disease have involved lymph nodes. The incidence of nodal involvement is still relatively low in advanced glottic disease; only about 20% of patients who present with T3 or T4 disease have node-positive disease. The frequency of occult nodal involvement is also low; Byers and colleagues1 found microscopic nodal involvement in less than 20% of patients with T3 N0 or T4 N0 disease who underwent elective nodal dissection at the same time as laryngectomy. The subglottic lymphatic supply is also poor. The lymph­atic capillaries drain through the cricothyroid

Lymphatic Metastases

13  The Larynx and Hypopharynx

­ embrane to the Delphian node or the prelaryngeal node, m as well as to the midjugular and lower jugular nodes and the paratracheal and supraclavicular nodes. In contrast to the lack of lymphatic capillaries in the glottic larynx, the supraglottic larynx has a moderately rich supply of lymphatic drainage. Lymphatic vessels in the supraglottic larynx collect in channels that pass through the pyriform sinuses to drain to nodes along the jugular chain, particularly the upper jugular and midjugular lymph nodes. About 55% of patients with primary supraglottic cancer have clinically involved lymph nodes. In a study of patients undergoing neck dissection in addition to supraglottic laryngectomy at The University of Texas M. D. Anderson Cancer Center for treatment of T2 or T3 disease,2 almost two thirds of patients had lymph node involvement. One third of patients had palpable nodes at presentation, and almost another one third of patients had pathologic nodal involvement. Hypopharyngeal cancer commonly manifests with nodal metastases because the pharynx has an extensive lymphatic network. Lindberg3 reported an overall 75% incidence of nodal metastases in patients presenting to M. D. Anderson Cancer Center with hypopharyngeal tumors. In another study, the incidence of positive lymph nodes in a series of patients with early (T1 or T2) hypopharyngeal cancer treated with radiation at M. D. Anderson was 52%.4 Nodal metastases appear most often in the jugular chain, particularly the subdigastric and midjugular nodes. Bilateral lymphadenopathy occurs in 15% of patients. Hypopha­ ryngeal tumors also have access to deep jugular, posterior cervical, and retropharyngeal lymph nodes.

Distant Metastases Distant metastases from glottic cancer are rare. The hypopharynx represents the other end of the spectrum; of all the types of head and neck mucosal cancer, it is the cancer most likely to metastasize to distant sites. A study of more than 5000 patients at M. D. Anderson5 found that the incidence of distant failure was 1% for those with carcinomas of the true vocal cords, 13% for those with tumors of the supraglottic larynx, and 23% for patients with carcinomas of the hypopharynx. This pattern most likely reflects the extent of local and regional disease at presentation rather than the site of origin. Clinically evident metastases were present in approximately 20% of patients with stage IV disease.5 The lungs are the most common site for the development of distant metastases from laryngeal and hypopharyngeal cancers and are the first site of presentation of hematogenous metastases in almost 60% of patients. Bones are the next most common sites, with bone metastases developing in 20% of patients with distant disease. Liver metastases occur in 10% of patients with hematogenous spread of disease from the larynx and hypopharynx. Spread to mediastinal lymph nodes, the brain, or other organs is rare.

Pathology Most cases of laryngeal and hypopharyngeal cancer are squamous cell carcinomas. Laryngeal carcinomas, particularly those originating on the true vocal cords, are usually well differentiated. Hypopharyngeal cancer tends to be less well differentiated.

287

Pathologic tissue changes seen in the larynx range from premalignant atypia or dysplasia to carcinoma in situ to microinvasion. Carcinoma in situ is a pathologic entity in which the carcinomatous changes are confined to the thickened epithelium and do not penetrate the lamina propria. Clinically, carcinoma in situ appears as a whitish, thickened, irregular area on the mucosa. Pathologically, penetration through the basement membrane is necessary to make the diagnosis of invasive carcinoma. Invasion can be missed if biopsy samples are too small or too superficial for proper histologic evaluation. Variants of squamous cell carcinoma represent approximately 3% of all laryngeal malignancies. The most common variant is verrucous carcinoma, a slow-growing tumor with an exophytic, warty appearance. This tumor is often low grade and has a low metastatic potential. Another variant is squamous cell carcinoma with spindle cell features, also known as pseudosarcoma. On microscopic examination, squamous carcinoma cells are seen along with spindle cells. The nature of these spindle cells is controversial; they are believed to be benign or highly malignant. On clinical examination, squamous cell carcinomas with spindle cell features often manifest as large, polypoid lesions that sometimes act as ball valves in the larynx. Basaloid squamous cell carcinoma and lymphoepithelioma can also be seen in the larynx and hypopharynx. Neoplasms more commonly found in other locations are occasionally seen in the larynx and hypopharynx. These malignancies include minor types of salivary gland cancer, such as adenocarcinoma and adenoid cystic carcinoma; neuroendocrine carcinomas, including true small cell or oat cell carcinoma; sarcomas; and lymphomas.

Clinical Manifestations Lesions of the Glottis Hoarseness is the main presenting symptom in patients with glottic carcinoma. Pain, dysphagia, and ultimately airway obstruction can occur but not until the disease is advanced. Involvement of the recurrent laryngeal nerve can lead to referred otalgia.

Lesions of the Supraglottis Supraglottic cancer can manifest as voice changes, but the quality of those voice changes depends on the location of the tumor within the supraglottic larynx. Patients with false vocal cord lesions can present with hoarseness, whereas patients with epiglottic tumors can have voice changes often referred to as supraglottic voice or hot potato voice. Odynophagia and dysphagia are other common presenting symptoms of supraglottic carcinoma. Referred otalgia can be a presenting symptom. Rarely, patients may present with an asymptomatic neck metastasis. As tumors become more advanced, they can cause airway obstruction or aspiration as a result of dysfunction of the epiglottis. Advanced tumors can invade the tongue or extend directly into the neck.

Lesions of the Subglottis Early disease of the subglottis can manifest as hoarseness. However, because the subglottis is a “silent” region, often no symptoms are evident until tumors reach advanced stages. Symptoms of advanced subglottis tumors include

288

PART 3  Head and Neck

dyspnea and stridor because of respiratory obstruction and, occasionally, hemoptysis.

Lesions of the Hypopharynx The most common presenting symptom of hypopharyngeal cancer is throat pain with or without referred otalgia. Dysphagia and weight loss are common in patients with carcinoma of the hypopharynx, who often are malnourished or have poor performance status. Hoarseness can occur when invasion of the larynx results in impairment of vocal cord motion. Because the incidence of nodal involvement is very high, patients with hypopharyngeal cancer may present with asymptomatic neck masses.

Patient Evaluation Clinical evaluation of patients with laryngeal or hypopharyngeal cancer includes a complete medical history and thorough physical examination. The examination must include good visualization of the larynx and hypopharynx, which can be accomplished with indirect laryngoscopy with a mirror and fiberoptic endoscopy. The examination should include an assessment of the size of the lesion, its morphology, extension into adjacent structures, and vocal cord motion. The examination should also include assessment for indirect signs of pre-epiglottic space invasion, such as fullness of the valleculae or ulceration of the infrahyoid epiglottis. Examination of the neck is also important to detect any lymphadenopathy and to assess its extent in terms of mobility or fixation of nodal disease, the size of nodal disease, and the number of nodal groups involved. Palpation of the neck can reveal direct tumor extension, indicated by tenderness of the thyroid cartilage or the presence of a subcutaneous mass. Absence of the thyroid click may indicate a large posterior lesion displacing the larynx anteriorly. Distant metastases from laryngeal cancer are uncommon but are most likely to appear in the lungs. Examination of the chest is therefore a crucial part of the clinical examination for patients with laryngeal or hypopharyngeal cancer. At a minimum, a chest radiograph is required to rule out hematogenous spread and to assess for a synchronous second primary cancer involving the lungs. Additional screening with computed tomography (CT) of the chest has become routine, as has the use of positron emission tomography (PET) for staging head and neck cancer. In cases of advanced laryngeal or hypopharyngeal cancer, PET can provide additional information on the extent of locoregional and distant disease. A complete blood cell count and liver function tests should be obtained to assess for anemia and liver dysfunction. Although these tests have a low yield in identifying distant disease, they nevertheless are simple screening tests that should be part of the initial workup. Anemia is thought to be an indicator of poor outcome after radiation therapy,6 but it is unclear whether correcting this abnormality improves outcome or the presence of anemia should influence the choice of treatment modality. Radiologic studies supplement the clinical evaluation of the primary tumor site and the neck. CT and magnetic resonance imaging (MRI) are both used, and the modality of choice is controversial. Often, the choice of scan type is based on the experience of the evaluating team, particularly the diagnostic radiologist. Scans are useful for identifying

FIGURE 13-12.  CT scan shows retropharyngeal adenopathy (arrow) from a carcinoma of the pyriform sinus.

involved lymph nodes that were not detected by palpation. Scans also are useful for assessing the pre-epiglottic space, tongue base, paraglottic region, and subglottis, areas that are sometimes difficult to visualize on direct examination. Thyroid cartilage invasion, particularly if not through the full thickness of the cartilage, may be detected only by CT or MRI. The inferior extent of hypopharyngeal cancer is also difficult to detect on clinical examination, and radiographic scans are critical in determining disease extent. Evaluation of this area can be supplemented by a barium swallow study. For hypopharyngeal cancer, the scan must include the retropharyngeal region to rule out the possibility of metastatic disease in that region (Fig. 13-12). A tissue diagnosis is ultimately required, and this is obtained at direct laryngoscopy. This test also allows direct visualization of the disease and its extent. Some physicians still perform triple endoscopy (laryngoscopy, bronchoscopy, and esophagoscopy) as a component of the initial workup, but the yield in terms of finding a synchronous primary lesion in patients without symptoms that suggest such a lesion is low, and use of triple endoscopy in the initial workup is becoming rare.

Staging The staging system for cancer of the larynx and hypopharynx is the American Joint Committee on Cancer (AJCC) ­tumor-node-metastasis  (TNM)  system,7 which includes ­separate tumor classification systems for each of the three subsites of the larynx. The nodal classification system is the same for all three laryngeal subsites and the hypopharynx and is the same system used for head and neck cancer in general (except for the nasopharynx). The nodal classification system is based on the size, number, and laterality of

13  The Larynx and Hypopharynx

nodes. The current (2002) TNM system is shown in Box 13-1. A change from the previous (1997) staging system addressed the resectability of disease.7 In brief, T4 lesions are now classified as T4a (resectable) or T4b (unresectable); this change led to the division of stage IV disease into IVA, IVB, or IVC (see Box 13-1).

Larynx In the case of laryngeal cancer, primary tumor stage is determined on the basis of sites of extension and vocal cord mobility (a surrogate of disease extension) but not tumor size. Few modifications have been made to the staging

289

s­ ystem over the years. The latest system considers supraglottic tumors with mucosal involvement of the base of the tongue or involvement of the medial wall of the pyriform sinus as being more favorable disease, and such tumors are classified as T2. The 2002 AJCC staging system considers tumors with paraglottic space involvement as T3 disease, which can lead to disease upstaging. In the 2002 system, the presence of minor thyroid cartilage erosion denotes T3 disease, a change that represents a clarification from previous staging systems. In the previous system, T4 (now T4a) denoted disease that extends through the thyroid cartilage; however, tumors that showed any ­thyroid

BOX 13-1.  American Joint Committee on Cancer Staging System for Cancer of the Larynx and Hypopharynx Primary Tumor (T) TX T0 Tis Supraglottis T1 T2 T3 T4a T4b Glottis T1 T1a T1b T2 T3 T4a T4b Subglottis T1 T2 T3 T4a T4b Hypopharynx T1 T2 T3 T4a T4b

Primary tumor cannot be assessed No evidence of primary tumor Carcinoma in situ Tumor limited to one subsite of supraglottis with normal vocal cord mobility Tumor invades mucosa of more than one adjacent subsite of supraglottis or glottis or region outside the supraglottis (e.g., mucosa of base of tongue, vallecula, medial wall of pyriform sinus) without fixation of the larynx Tumor limited to larynx with vocal cord fixation and/or invades any of the following: postcricoid area, pre-epiglottic tissues, paraglottic space, and/or minor thyroid cartilage erosion (e.g., inner cortex) Tumor invades through the thyroid cartilage and/or invades tissues beyond the larynx (e.g., trachea, soft tissues of neck including deep intrinsic muscle of the tongue, strap muscles, thyroid, or esophagus) Tumor invades prevertebral space, encases carotid artery, or invades mediastinal structures Tumor limited to the vocal cord(s) (may involve anterior or posterior commissure) with normal mobility Tumor limited to one vocal cord Tumor involves both vocal cords Tumor extends to supraglottis and/or subglottis, and/or with impaired vocal cord mobility Tumor limited to the larynx with vocal cord fixation and/or invades paraglottic space, and/or minor thyroid cartilage erosion (e.g., inner cortex) Tumor invades through the thyroid cartilage and/or invades tissues beyond the larynx (e.g., trachea, soft tissues of neck including deep extrinsic muscle of the tongue, strap muscles, thyroid, or esophagus) Tumor invades prevertebral space, encases carotid artery, or invades mediastinal structures Tumor limited to the subglottis Tumor extends to vocal cord(s) with normal or impaired mobility Tumor limited to larynx with vocal cord fixation Tumor invades cricoid or thyroid cartilage and/or invades tissues beyond the larynx (e.g., trachea, soft tissues of neck including deep extrinsic muscles of the tongue, strap muscles, thyroid, or esophagus) Tumor invades prevertebral space, encases carotid artery, or invades mediastinal structures Tumor limited to one subsite of hypopharynx and 2 cm or less in the greatest dimension Tumor involves more than one subsite of hypopharynx or an adjacent site, or measures more than 2 cm but not more than 4 cm in the greatest diameter without fixation of hemilarynx Tumor measures more than 4 cm in the greatest dimension or with fixation of hemilarynx Tumor invades thyroid/cricoid cartilage, hyoid bone, thyroid gland, esophagus, or central compartment soft tissue* Tumor invades prevertebral fascia, encases prevertebral fascia, encases carotid artery, or involves mediastinal structures

Regional Lymph Nodes (N) NX N0 N1

Regional lymph nodes cannot be assessed No regional lymph node metastasis Metastasis in a single ipsilateral lymph node, 3 cm or less in the greatest dimension Continued

290

PART 3  Head and Neck

BOX 13-1.  American Joint Committee on Cancer Staging System for Cancer of the Larynx and Hypopharynx—cont’d N2 N2a N2b N2c N3

Metastasis in a single ipsilateral lymph node, more than 3 cm but not more than 6 cm in the greatest dimension, or in multiple ipsilateral lymph nodes, none more than 6 cm in the greatest dimension, or in bilateral or contralateral lymph nodes, none more than 6 cm in the greatest dimension Metastasis in single ipsilateral lymph node, more than 3 cm but not more than 6 cm in the greatest dimension Metastasis in multiple ipsilateral lymph nodes, none more than 6 cm in the greatest dimension Metastasis in bilateral or contralateral lymph nodes, none more than 6 cm in the greatest dimension Metastasis in a lymph node, more than 6 cm in the greatest dimension

Distant Metastasis (M) MX M0 M1

Distant metastasis cannot be assessed No distant metastasis Distant metastasis

Stage Grouping Stage 0

Tis

N0

M0

Stage I

T1

N0

M0

Stage II

T2

N0

M0

Stage III

T3

N0

M0

T1-T3

N1

M0

T4a

N0-N1

M0

T1-T4a

N2

M0

T4b

Any N

M0

Any T

N3

M0

Any T

Any N

M1

Stage IVA Stage IVB Stage IVC

*Central compartment soft tissue includes prelaryngeal strap muscles and subcutaneous fat. Modified from Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 33-45, 47-57.

cartilage invasion were often categorized as T4. The significance of this distinction is related to a fundamental difference between laryngeal cancer and cancer at other head and neck sites. At most head and neck sites, tumors that recur after nonsurgical therapy are difficult to treat successfully with salvage therapy. However, glottic or supraglottic cancer that is initially confined to the larynx and then recurs after radiation or chemoradiation therapy can often be salvaged successfully with surgery. Conversely, after a laryngeal tumor extends beyond the confines of the cartilages, surgical salvage after therapeutic failure is less likely to be successful. A distinction is made between tumors with minor erosion and tumors that extend through the cartilage and thereby into the neck.

Hypopharynx Staging of primary hypopharyngeal tumors is based on the sites of involvement, larynx motion, and the size of the tumor, although an accurate clinical assessment of tumor size can be difficult in this region. Aside from the division of T4 tumors into resectable and unresectable subcategories, the current staging system remains unchanged from the previous edition.

Selection of Treatment The primary goal in the treatment of cancer of the larynx and hypopharynx is to maximize the cure rate. However, preservation of speech and swallowing is also desirable.

Establishing algorithms that accomplish both goals is a challenge, and the focus of many studies of optimal treatments for these tumors has been on how best to maintain good survival rates while maximizing function. Different strategies have been pursued, including a greater reliance on radiation therapy instead of surgery for the treatment of local disease, improvements in partial laryngeal surgery to reduce the amount of laryngeal tissue removed and to preserve the natural voice, and exploration of the use of systemic therapies to enhance radiation therapy or identify patients for whom more limited or no surgical resection is required. Improvements in rehabilitative techniques, although beyond the scope of this chapter, also have been significant in improving the functional outcome of patients with laryngeal or hypopharyngeal tumors.

Early-Stage Disease Radiation therapy and surgery have been used in the management of laryngeal and hypopharyngeal cancer. Earlystage disease often can be cured with either modality, and single-modality therapy is preferred. The choice of modality is often based on the perception of the quality of function that can be achieved and is determined by the experience of the managing physicians. Limited surgery, including laser excision, stripping, laryngofissure, and cordectomy, may be successful for earlystage laryngeal cancer. Larger laryngeal lesions still considered early-stage disease (T2 tumors) can be managed with

13  The Larynx and Hypopharynx

291

TABLE 13-1.   Randomized Trials of Induction Chemotherapy followed by Surgery vs. Immediate Surgery for Advanced Laryngeal and Hypopharyngeal Carcinomas Study and ­Reference

Primary Tumor Site

Number of ­ atients P

T Category ­Distribution

Node ­Positive (%) 5-Year Survival Rates

VA18

Larynx

332

48

EORTC17

Hypopharynx

194

GETTEC16

Larynx

68

T1-2: 11% T3: 65% T4: 24% T1-2: 21% T3: 76% T4: 3% T3: 100%

Intergroup-RTOG 91-1119

Larynx

547

T2: 12% T3: 78% T4: 10%

50

65 23

Surgery first: 45% Induction chemotherapy: 42% (P = 0.35) Surgery first: 35% Induction chemotherapy: 30% (relative risk = .86) Surgery first: 62% Induction chemotherapy: 30% (P =.02) Induction chemotherapy: 55% Radiation only: 56% Concurrent chemotherapy: 54%

EORTC, European Organisation for Research and Treatment of Cancer; GETTEC, Groupe d’Etudes des Tumeurs de la Tete et du Cou; RTOG, Radiation Therapy Oncology Group; VA, U.S. Department of Veterans Affairs.

partial laryngeal surgery. A small hypopharyngeal tumor can sometimes be managed with a limited pharyngectomy. Radiation is an effective treatment for early laryngeal and hypopharyngeal cancer, particularly for early glottic tumors. Because the volume of tissue irradiated for these tumors is small, chronic side effects are minimal. For patients with early hypopharyngeal tumors located such that surgery would leave the patient with poor function, radiation therapy may be the best choice. Another advantage of radiation therapy for patients with early laryngeal or hypopharyngeal tumors is that it addresses subclinical disease, including potential retropharyngeal nodal disease. An important factor in the choice between surgery and radiation therapy is the quality of voice expected after therapy. Laser excision of a mid–vocal cord lesion often leaves patients with excellent voice quality, comparable to voice quality after radiation therapy.8 However, laser excision, particularly of lesions in the region of the anterior commissure, can result in permanent hoarseness.9 Harrison and colleagues10 studied voice quality in irradiated patients and concluded that the resultant voice quality was excellent. An additional factor in treatment decision-making is cost, particularly when other factors such as curability and resultant function are thought to be equivalent. Foote and colleagues11 evaluated the cost of radiation therapy for early laryngeal cancer and concluded that the cost was similar to that of limited surgery when salvage therapy was factored in. Other studies have suggested that radiation is cost-effective compared with surgery.12,13

Advanced Disease Advanced laryngeal and hypopharyngeal cancer often requires the use of combination therapies. Total laryngectomy (with partial pharyngectomy for hypopharyngeal cancer) has been the standard of care for most patients with T3 or T4 disease and for selected patients with bulky T2 disease. In the United States, total laryngectomy is still considered the standard for managing advanced laryngeal cancer, and it often represents the control condition for studies evaluating other therapies. Disease control rates are

moderately high, and use of tracheoesophageal punctures for voice restoration can allow patients to regain their verbal communication skills. Total laryngectomy remains the preferred treatment for patients with stage III disease for whom surveillance is difficult or for those who may not be fit for or have the social support needed to undergo more aggressive or protracted therapy. Several studies have indicated that radiation is an effective alternative therapy for advanced larynx cancer.14,15 Unfortunately, no randomized trials comparing radiation and surgery have been done, and comparisons are restricted to retrospective experiences. Nevertheless, radiation therapy may be an appropriate choice for patients whose therapeutic options are limited by comorbid conditions. Interest in laryngeal preservation led to studies testing the role of cytotoxic therapies, and chemotherapy has become an important part of managing advanced cancer of the larynx or hypopharynx. Three randomized studies16-18 investigated the use of induction chemotherapy as an initial treatment compared with immediate surgery consisting of total laryngectomy, and two of the studies concluded that induction chemotherapy did not compromise survival rates and allowed laryngeal preservation in some percentage of patients (Table 13-1).16-19 Systemic therapy has ­ particularly strong appeal in the management of hypopharyngeal cancer, which is associated with a high rate of distant metastasis. Results of an Intergroup larynx preservation trial (Radiation Therapy Oncology Group [RTOG] trial 91-11),19 which compared neoadjuvant chemotherapy with radiation alone or radiation concurrent with high-dose ­ cisplatin, indicated that concurrent therapy was more effective for laryngeal preservation and locoregional control compared with induction chemotherapy or radiation therapy alone. Moreover, survival rates were equivalent with all three approaches.19 Because of the rarity of hypopharyngeal cancer, randomized trials that specifically consider disease at this site are difficult to perform. However, many investigators have compared the effectiveness of concurrent chemotherapy and radiation with that of radiation alone for squamous cell

292

PART 3  Head and Neck

carcinomas of the head and neck.20,21 Even though these studies have focused on head and neck cancer in general as opposed to laryngeal and hypopharyngeal cancer, concurrent chemoradiation therapy for advanced resectable and unresectable laryngeal and hypopharyngeal cancer has become an accepted treatment of choice for these types of cancer. Ultimately, the patients must make the choice among the various treatment options available to them. In a survey of patient preferences with regard to treatment of laryngeal cancer, McNeil and colleagues22 found that 20% of respondents chose radiation therapy over surgery, even with a compromise in survival, to maintain their voices. Involving patients in treatment decisions and ensuring that patients understand possible function and rehabilitation after laryngectomy is critical. Van der Donk and colleagues23 found that although 83% of clinicians preferred radiation therapy over laryngectomy for managing T3 laryngeal cancer, only 50% of patients who had been treated for laryngeal cancer shared this preference. It is important to realize that larynx preservation is feasible and in most cases does not affect survival, but when laryngeal preservation is not prudent or desired, voice restoration can still be achieved, and the change in voice may not have as great an effect on patients’ quality of life as people may believe.

Carcinoma of the Glottis The optimal management strategy for all presentations of carcinoma of the glottis, including carcinoma in situ of the true vocal cords, remains controversial. Management practices for carcinoma in situ range from after-biopsy observation to vocal cord stripping, cordectomy, or open partial laryngectomy to primary radiation therapy.24 Results of conservative surgical approaches such as microexcision, laser ablation,25,26 and vocal cord stripping27 are often excellent. In the management of carcinoma in situ, we favor a limited surgical procedure as the initial treatment of choice. However, in the event of recurrence, further surgery, even if limited to stripping or laser excision, can cause significant degradation in voice quality, and radiation therapy is therefore an ideal treatment. Radiation therapy is also an effective initial treatment for carcinoma in situ and is recommended by some clinicians.

T1 and T2 Tumors T1 tumors of the glottis respond well to radiation therapy, and radiation therapy is the treatment of choice for these lesions at most centers. Other options include laser excision,28 laryngofissure,29 and partial laryngectomy,30,31 but these techniques often result in poorer voice quality. Many groups consider radiation therapy to be the treatment of choice for T1 and T2 lesions, although partial ­laryngeal surgery is also an option for T2 lesions. Results with partial vertical laryngectomy are similar to those achieved with radiation therapy, although voice quality is an issue. At the University of Florida, a review of patients with T2 disease revealed that more than 50% of the ­patients who underwent radiation therapy would have required total laryngectomy if a surgical approach had been chosen (i.e., voice-preserving surgery would not have been feasible).32 Surgical groups have reported using a horizontal partial laryngectomy with cricohyoidopexy to maintain voice and achieve high control rates.33 Some groups have

used induction chemotherapy followed by limited surgery for patients with T2 tumors, with promising results.34

T3 Tumors Historically, the standard therapy for T3 carcinoma of the glottis was total laryngectomy followed by postoperative irradiation therapy when high-risk pathologic features (e.g., poor differentiation, extensive subglottic extension, or cartilage invasion) are identified.35 Postoperative radiation was also advised for patients who require tracheostomy as a separate surgical procedure before laryngectomy and for patients with the uncommon finding of multiple involved nodes in the neck or extracapsular invasion. For some patients and for some practitioners, this traditional standard approach is still the treatment of choice. If high-risk features such as extranodal spread or inadequate margins are discovered at pathologic examination, the recommendation for medically fit patients is postoperative concurrent chemoradiation therapy.36,37 In many cases, nonsurgical therapy with attempted voice preservation can be considered for T3 glottic cancer. Radiation therapy alone can produce moderately high local ­control rates among patients with T3 disease and is the treatment of choice at some centers, although it is uncommon practice in the United States. The Intergroup larynx preservation trial19 showed that concurrent chemoradiation therapy produced higher laryngeal preservation rates and equivalent survival rates compared with radiation alone. Even though the addition of chemotherapy was associated with greater toxicity,19 concurrent chemoradiation therapy is considered by many to be the standard of care for patients with T3 laryngeal cancer.

T4 Tumors For patients with T4 glottic cancer, total laryngectomy remains the treatment of choice. Postoperative radiation is almost always required. Definitive radiation therapy is still an option for patients who cannot or are unwilling to undergo total laryngectomy.15 As the outcome of patients with T4 disease treated with induction chemotherapy and surgery seems poorer than expected for patients treated with surgery alone, induction chemotherapy is not recommended.38 Concurrent chemoradiation therapy may be beneficial for patients with T4 tumors, but this approach should be considered experimental and should be offered only in the context of a clinical trial.

Carcinoma of the Supraglottis Early-Stage Disease Radiation therapy and surgery can offer cure and voice preservation for patients with early supraglottic cancer. Endoscopic removal of early lesions is becoming a ­recommended procedure at some centers, particularly in Germany, where the procedure has been used for many years.39 Larger lesions can be removed by supraglottic laryngectomy. The success rate of supraglottic laryngectomy is high, but patient selection is critical. Patients must be motivated for the rehabilitation that is required, and they also must have sufficient pulmonary reserve to tolerate the procedure. ­ As­piration is a problem because the protection from the epiglottis is removed, and patients must be fit enough to combat this. The need for a permanent tracheostomy is

13  The Larynx and Hypopharynx

­ ncommon, although it takes months before decannulau tion is possible in some patients. Radiation therapy is often the treatment of choice for patients with early supraglottic carcinoma. Results of many series are comparable to results of series of patients treated with supraglottic laryngectomy.40 Most radiation therapy series include patients who were medically unable to undergo a voice-preserving treatment or whose tumors were not amenable to partial laryngectomy. In the treatment of supraglottic cancer, unlike glottic cancer, management of the neck is a significant issue that must be considered in addition to management of the primary tumor. In patients with clinically node-negative disease, the neck must be managed with the same modality as that chosen for management of the primary tumor. A neck dissection should accompany laryngectomy, because one third of patients without clinical adenopathy have subclinical nodal disease that is revealed during dissection. If multiple nodes or extracapsular nodal disease is found on pathologic review, postoperative radiation to reduce the risk of recurrence is recommended. Findings from surgical series41 suggest that patients with clinically nodenegative disease who are to undergo radiation as definitive therapy must have their necks included in the treatment fields. The nodal basins at risk are levels II and III. Patients with clinically evident adenopathy may require radiation therapy and surgery, because almost two thirds of these patients have pathologic nodal disease. Patients with N1 supraglottic disease often can be treated with radiation alone to the primary tumor and neck. A postirradiation neck dissection may be required if the nodal disease persists after therapy. For patients with more extensive nodal disease, many groups argue that a neck dissection after radiation should be mandatory, whereas others advocate observation in the event of complete disappearance of nodal disease after radiation.42

T3 Disease For patients with bulky T3 disease, supraglottic laryngectomy may be the treatment of choice, especially for disease with significant involvement of the pre-epiglottic space.43 This procedure is theoretically feasible if no significant subglottic involvement is present and one arytenoid cartilage is uninvolved and can be spared. The caveats regarding the patient being willing to undergo rehabilitation and having sufficient pulmonary reserve apply. The optimal management of supraglottic disease involving fixation of the hemilarynx remains unclear, and the questions that remain to be answered are similar to those regarding management of T3 glottic cancer. Concurrent chemoradiation therapy is becoming the standard treatment for many T3 supraglottic cancers.19 More than two thirds of the patients enrolled in the ­Intergroup larynx preservation trial had supraglottic cancer. Although many patients with T3 supraglottic cancer will undergo total laryngectomy, some will elect a partial laryngectomy. These patients often require postoperative radiation because of inadequate margins, perineural or vascular spread, cartilage involvement, or subglottic involvement of more than 1 cm. The finding of multiple nodes or extracapsular nodal spread also dictates the need for postoperative radiation, often given with concurrent ­chemotherapy.36,37

293

T4 Disease T4 supraglottic disease is usually best managed with total laryngectomy. Postoperative radiation is used in most cases because of cartilage invasion, subglottic spread, or inadequate margins. Some patients with disease classified as T4 because of involvement of the base of the tongue may do well with primary radiation therapy or, for those with very bulky disease, chemotherapy combined with radiation therapy. Tumors such as these tend to behave more like oropharyngeal carcinomas, which some groups speculate have better outcomes than laryngeal or hypopharyngeal cancer, stage for stage, in terms of local control.

Carcinoma of the Subglottis Carcinoma of the subglottis is uncommon and is often best managed with surgery. Because a partial laryngectomy is rarely possible, patients with subglottic cancer are ­usually treated with total laryngectomy. The finding of nodal disease, which necessitates adjuvant radiation therapy, is uncommon. Postoperative radiation is needed in cases of inadequate margins and cartilaginous extension. Small tumors may be treated with radiation therapy alone, but these are rare and most often are glottic tumors that originate from the undersurface of the true vocal cord.

Carcinoma of the Hypopharynx Early tumors of the hypopharynx are uncommon and can be treated with definitive radiation therapy. Partial pharyngectomy, with or without a partial laryngectomy, can be done, but radiation therapy often results in better function and addresses the lymphatics in patients with clinically negative nodes. Posterior wall lesions are particularly better suited for radiation therapy, because it is often difficult to obtain negative surgical margins in the region of the prevertebral fascia and muscles. Advanced tumors of the hypopharynx are often best managed with total laryngectomy and partial or total pharyngectomy. The feasibility of this extensive surgical procedure has been enhanced by the use of free tissue transfer in which tissue from the stomach or the intestine (jejunum) is used to fill the defect and allow return of swallowing. Postoperative radiation therapy usually is required to improve local and regional control. In the case of large hypopharyngeal tumors, generous margins are often difficult to obtain, and the patients, particularly those with advanced primary disease, usually have significant nodal disease. In a study similar to the Veterans Administration (VA) study of induction chemotherapy for patients with laryngeal cancer in the United States,18 the European Organisation for Research and Treatment of Cancer (EORTC) studied initial surgery and postoperative radiation versus induction chemotherapy for hypopharyngeal cancer.17 In the VA trial, patients with a complete or partial response to chemotherapy went on to undergo radiation, whereas in the EORTC trial, a complete response was required to avoid surgery and be treated with radiation alone. In the EORTC trial, the complete response rate was high at more than 50%. However, survival rates between the two study groups were thought to be equivalent, and laryngeal preservation was achieved in more than one third of the patients treated with induction chemotherapy. Most patients in the EORTC trial had T2 or T3 disease. Although no ­studies

294

PART 3  Head and Neck

have been conducted that specifically addressed hypopharyngeal cancer, findings from studies of head and neck cancer and meta-analyses strongly suggest that concurrent chemoradiation therapy produces survival advantages over radiation alone or induction chemotherapy.20,44 Concurrent chemoradiation therapy is often the treatment of choice for patients with T3 hypopharyngeal cancer. Few patients with T4 disease were treated in the EORTC study or in other trials of concurrent chemoradiation therapy, and surgery and postoperative chemoradiation therapy should be considered the appropriate treatment for these patients unless they cannot undergo surgery for medical reasons or have unresectable disease. For patients with unresectable disease, concurrent chemoradiation therapy seems to produce better outcomes than radiation therapy alone, although the treatment is more toxic and patients must be carefully selected for this approach. Little evidence is available regarding radiation therapy alone for unresectable T4 hypopharyngeal cancer, although the general impression has been that control with this approach is poor. An Intergroup trial comparing chemoradiation therapy with radiation alone for unresectable head and neck cancer demonstrated a survival ­advantage for patients treated with high-dose cisplatin ­delivered concurrently with radiation, although 5-year survival rates were low for all groups with this advanced ­disease ­presentation.45

Radiation Therapy Techniques Beam Energies Cobalt 60 (60Co) is often the ideal mode of radiation for treating laryngeal cancer. For primary treatment of laryngeal cancer, 60Co allows adequate dosing of the superficial tissues. Historically, great concern has been expressed that 6-MV photons have too large a buildup region and can potentially underdose superficial tissues. Most radiation oncology centers have only linear accelerators. If a 60Co unit is not available, 4-MV photon beams are adequate. However, the lowest energy of units at many centers is 6-MV photons. Even so, reviews of experiences with 6-MV photons indicate that control rates for early lesions remain high and use of this “higher” energy does not compromise treatment.46 A dosimetric comparison of 60Co and

A

6-MV beams ­conducted at the University of Florida47 led the investigators to conclude that the dose distribution for the two beams was similar except for the most anterior tissues (3 mm below the surface), where underdosing occurred with the 6-MV beam. On the basis of their results, the investigators cautioned against using 6-MV beams for lesions in the anterior commissure. However, when the anterior extent of disease is a concern, tissue-equivalent bolus material for buildup or a beam spoiler can be used. At M. D. Anderson, 60Co is still used for small laryngeal fields, and linear accelerators are ­increasingly being used for wider-field radiation. The treatment-planning advantages of a linear accelerator outweigh the small differences in beam properties. Advanced disease is increasingly being treated with intensity-­modulated radiation therapy (IMRT), in which low-energy photons are delivered by a linear accelerator. Most patients who present for postoperative radiation therapy at M. D. Anderson are treated with 6-MV photons generated by a linear accelerator, an approach that offers considerable flexibility for planning three-dimensional conformal or IMRT. Tissues that have been violated by surgery must be treated with a bolus material to avoid dermal or subcutaneous underdosing, which could result in a recurrence. Photon energies higher than 6 MV are rarely needed for postoperative therapy.

Simulation and Field Arrangements Treatment simulation is required before treatment is begun to allow establishment of a reproducible treatment setup. Simulation also allows delineation of fields before therapy and determination of the parameters necessary for treatment planning. Patients are treated in the supine position with the head extended, which often elevates the larynx slightly away from the shoulders and thoracic inlet. Straps are commonly used so that the patients’ shoulders are pulled to a more inferior position, thereby avoiding penetration of the shoulders by lateral beams (Fig. 13-13A). The head is immobilized with the use of individualized thermoplastic masks (see Fig. 13-13B). In many practices, treatment simulation is based on CT scans. In this process, images are obtained after the patient is immobilized, and fiducials are placed for isocenter localization. After the images are acquired, the target areas

B

FIGURE 13-13.  A patient with laryngeal cancer is being treated with radiation therapy. A, Use of straps to optimize shoulder ­placement. B, Immobilization of the head with a thermoplastic mask.

13  The Larynx and Hypopharynx

A

295

B

FIGURE 13-14.  Photographs of a three-field setup show two lateral fields (A) and an anterior field (B) matched on the skin.

are identified on the image set. These targets can then be used as guides for a “forward plan” that uses a conventional three-field approach or three-dimensional conformal therapy. Alternatively, the targets can be used to develop an “inverse” treatment plan to be delivered with IMRT. These concepts are discussed further in Chapter 2. Use of a linear accelerator offers the most flexibility for modern treatment planning.

Setup Techniques At M. D. Anderson, the setup technique for three-field matches uses a mono-isocenter technique in which the opposing lateral fields and anterior low-neck field are matched at the isocenter (Fig. 13-14). A small block, 1 cm long by 3 cm wide, is placed centrally at the superior aspect of the anterior field for additional protection of the spinal cord. We tend to avoid placing safety blocks on the inferoposterior aspect of the lateral fields over the spinal cord, although this may be the more prudent approach in some situations. In postoperative treatment, placement of safety blocks is often complicated because of the potential to block the operative bed. For hypopharyngeal cancer in particular, the posterior cervical chain is a significant lymphatic drainage site that should not be blocked. Anteriorly, the location of the stoma complicates placement of the isocenter and a potential anterior safety block. Treatment options include matching below the stoma, if possible, or blocking the stoma and then treating it with a small electron field. The dosimetry in either case can be quite challenging. Three-dimensional planning and conformal techniques allow more creative solutions for difficult problems. One such problem is the localization of safety blocks in a threefield setup. Sometimes, a safe block over the spinal cord cannot be identified because of concern that disease may be blocked. In such a case, more sophisticated planning is required. A simple solution is to use fields that are caudally oriented48 so that junctions can be avoided (Fig. 13-15). The ability to visualize these beams in a treatment ­planning system allows the beam weightings and distribution to be optimized. The ability to plan these treatments allows appropriate wedging or tissue compensation to be selected in areas that are anatomically heterogeneous. Another option for postoperative therapy is IMRT, which may be

FIGURE 13-15.  Digitally reconstructed radiograph shows a caudally tilted (angled down) field used to avoid a match when use of a safety block over the spinal cord with a three-field  technique could potentially block disease.

best for patients with node-positive disease in whom parotid ­sparing is desired.

Treatment Volumes and Radiation Fields Carcinoma of the Glottis T1 and T2 carcinomas of the true vocal cords can be treated with a pair of small lateral opposed photon fields that encompass only the larynx (Fig. 13-16). Typical field sizes range from 25 cm2 (5 × 5 cm) to 36 cm2 (6 × 6 cm). The fields are defined by anatomic landmarks. The isocenter is located at the middle of the true cord. Anteriorly, the fields

296

PART 3  Head and Neck

FIGURE 13-16.  Simulation film shows the opposed fields used to treat a patient with early glottic cancer.

should fall off the skin surface. The posterior border is the anterior edge of the vertebral bodies. The superior border is above the thyroid notch, and the inferior border is inferior to the cricoid cartilage. In patients with T2 lesions, the borders are increased to appropriately encompass the disease. Involvement of the supraglottis necessitates increasing the superior border, and conversely, subglottic disease extension requires increasing the inferior border, often to encompass the upper trachea. For T2 disease, unless extensive supraglottic involvement is present, we use small fields that do not encompass the draining lymphatics. Occasionally, other techniques are required for T1 or T2 glottic cancer, such as caudal orientation of the beams or using a pair of left and right anterior oblique fields (wedged pair), with or without a straight anterior field, which may provide a better arrangement for encompassing the larynx and providing an improved distribution. Wang49 described a four-field technique; other possible options are rotational techniques and IMRT. A dosimetric plan is obtained with CT scans to evaluate the dose distribution. Because of the triangular shape of the larynx and surrounding tissues, tissue compensation, usually in the form of wedges, is required to optimize the dose distribution. Treating without wedges results in a less homogeneous distribution, but this may be advantageous for anterior lesions where the built-in dose differential (5% to 10%) is desirable. Occasionally, lateralized lesions can be treated with a single appositional field, but the preferred approach is to treat lesions such as these with opposed beams, with unequal weighting of the fields (e.g., 3:2) to favor the side of the lesion. Because of the moderate incidence of subclinical lymphadenopathy associated with T3-4 N0 glottic carcinomas, the primary lesion and the draining lymphatics are treated. This is accomplished with parallel-opposed lateral

FIGURE 13-17.  Simulation film shows the opposed fields used to treat a patient with advanced laryngeal cancer without regional metastases.

portals covering the primary tumor and upper jugular and midjugular lymph node chains (Fig. 13-17). The superior border is located approximately 2 cm superior to the angle of the mandible. Posteriorly, the fields are more generous to encompass the jugular nodes, and the posterior border is placed behind the spinous processes. The inferior border is determined by the inferior extent of disease and is adjusted to encompass any subglottic extension. With the techniques described previously, a matching anterior portal encompassing the lower jugular and supraclavicular lymph nodes is also used. IMRT may be another option for T3-4 N0 glottic cancer, although the superior border of parallel-opposed fields in patients with N0 disease often does not encompass a significant amount of parotid gland. For patients with a low-lying larynx, intensity-modulating techniques may be superior to opposed fields that may not clear the shoulders. If a patient is to be treated with a three-field technique, the initial fields include the spinal cord, and a field ­reduction from the posterior border must be done before 45 Gy is delivered. Tissues excluded from the reduced fields may contain nodes at risk for subclinical metastases and can be supplemented with narrow electron beam fields to bring the dose to those regions to 50 Gy. After 50 Gy is delivered to the primary tumor with margin and the electively treated draining lymphatics, a boost dose is delivered to the primary site. That dose is delivered by using parallel-opposed fields with a 1- to 1.5-cm margin surrounding the gross disease. The field design for postoperative treatment of T3 and T4 glottic carcinoma is similar to that used for the ­definitive

13  The Larynx and Hypopharynx

297

FIGURE 13-19.  Intensity-modulated radiation therapy (IMRT) plans are shown for a pathologic T4 N0 squamous cell carcinoma of the larynx after total laryngectomy and bilateral neck dissection. An axial view of an isodose distribution generated with a nine-field IMRT plan shows the clinical tumor volume (CTV1, red) encompassing the neopharynx that harbored the original tumor volume and the CTV2 (blue) encompassing the dissected neck on both sides. FIGURE 13-18.  Simulation film shows the opposed fields used for postoperative radiation of a patient who had undergone laryngectomy for advanced glottic cancer.

treatment of such lesions (Fig. 13-18). A three-field technique is preferred, with the stoma included in the low-neck field. The operative bed and neopharynx are included in the opposing lateral fields, with fields adjusted to appropriately cover any clinical or pathologic adenopathy. The inferior border is set to ensure that all preclinical extensions of disease are encompassed in the lateral fields with a 1.5- to 2-cm margin. If this is not possible because of the patient’s anatomy, caudally oriented fields can be used. Areas of high risk are treated with boost doses through parallelopposed fields. If any area in the neck requires a higher dose than the neopharynx, the dose can be delivered by using an appositional electron field. As is true in definitive therapy, intensity-modulated techniques may produce conformal plans that are easier to deliver than trying to match conventional fields through areas that are challenging because of sharp anatomic changes or areas suspected of harboring disease. If IMRT is to be delivered, the clinical target volume should include the gross tumor and a 1- to 1.5-cm margin. At a minimum, the laryngeal contents, including the thyroid cartilage, are delineated. Gross nodal disease is also identified with a 1-cm margin. For patients with T3 or T4 disease, neck nodes at levels II, III, IV, and VI are delineated as subclinical target volumes. For patients with nodenegative disease, the superior border of the level II nodes is identified as the level where the subdigastric muscle crosses the jugular vein or the C1-C2 junction. For patients with node-positive disease, level II is delineated to the jugular fossa on the involved side, and levels IB and V are covered on that side as well. When planning radiation to be delivered postoperatively, all of the available clinical information is used to

identify the clinical target volume. Preoperative clinical and radiographic findings are combined with operative and pathologic evaluations to identify the tissues at highest risk. A secondary clinical target volume encompasses the operative bed, exclusive of the primary volume (Fig. 13-19). For patients with node-positive disease, a tertiary clinical target volume encompasses undissected areas that may harbor disease, principally the high level II nodes (Fig 13-20).

Carcinoma of the Supraglottis Supraglottic carcinoma requires treatment of the draining lymphatics. In patients with T1 N0 lesions, the low jugular and supraclavicular nodes need not be treated. Two parallel fields are used (Fig. 13-21). The superior border encompasses the subdigastric nodes. The posterior border is located at the posterior spinous processes to ensure coverage of the nodes along the jugular vein. The inferior border is set below the larynx. Fall-off may be required for anterior lesions. A reduction off the spinal cord is necessary before 45 Gy is delivered. The posterior cervical strips are supplemented with electrons to 50 Gy. After a subclinical dose of 50 Gy is delivered, fields are reduced to treat the primary lesion with a 1- to 1.5-cm margin. Boost fields for these early lesions can be small, but from a dosimetric standpoint, these fields should not be less than 4 cm in any dimension. The field design for T2 or T3 lesions is similar to the field design for T1 disease. The low neck is treated with a matching anterior neck field. The superior border is adjusted in patients with clinical adenopathy. The superior border must cover the entire jugular chain and follow the inner table of the skull posteriorly. The mastoid tip is ­included, and the superior border sweeps forward to encompass the anterior superior jugular nodes that are close to the angle of the mandible.

298

PART 3  Head and Neck

A

B

FIGURE 13-20.  Intensity-modulated radiation therapy plan is shown for a pathologic T4N2 squamous cell carcinoma of the larynx after laryngectomy and neck dissection. Three clinical target volumes (CTVs) were delineated. A, Axial view of CTV1 (red) encompassing the original laryngeal and nodal disease and CTV2 (blue) encompassing the adjacent posterior components of the neck that were violated by surgery but not involved by tumor. B, Coronal view indicates CTV3 (yellow) encompassing the undissected high jugular nodes.

FIGURE 13-21.  Simulation film shows the opposed fields used to treat a patient with early supraglottic cancer.

Nodes may lie at a level parallel and posterior to the spinal cord such that if true lateral fields were used for the off–spinal cord fields and boost fields, only a portion of each node would be treated. Although the posterior aspect of the nodes can potentially be encompassed in electron fields, we prefer, when possible, to use parallel oblique fields that fully encompass the nodes in the photon fields. Care must be taken with this field arrangement, however, to ensure that treatment of the primary tumor

is not compromised. When bilateral positive nodes are present, the decision becomes more difficult. If one side has deep adenopathy that would be difficult to encompass with electrons, using oblique fields to fully encompass the deep nodes and primary tumor in the photon fields while encompassing the smaller nodes in an appositional electron field is one alternative to best encompass all disease. These field designs are only guidelines; treatment in the complicated scenarios described here must be designed on an individual basis. Other treatment options include caudally oriented field arrangements or other three-dimensional techniques, particularly if field matching through disease or critical structures is a concern. The flexibility of IMRT in such complex clinical situations is leading to its becoming the treatment technique of choice for such situations. Radiation techniques to be used after a supraglottic laryngectomy are similar to those used for patients with an intact larynx. The techniques for postoperative radiation therapy after total laryngectomy for supraglottic carcinomas are similar to those for postoperative radiation therapy for glottic carcinoma, with the following exceptions. In the case of supraglottic carcinomas, the draining lymphatics may need to be covered more comprehensively because they are more likely to harbor disease. If nodes are involved, the superior aspect of the field must cover to the jugular foramen. Boost fields also may need to be more generous superiorly if the base of the tongue is involved. For IMRT, the delineation of clinical target volumes for supraglottic cancer is the same in principle as for glottic carcinoma. Neck nodes at levels II and III are delineated as a secondary clinical target volume for patients with T1 N0 disease, and levels IV and VI are included for patients with T2 N0 disease. Otherwise, the guidelines for treatment design are similar to those for disease at other stages (Fig. 13-22) or treatment delivered after surgery.

13  The Larynx and Hypopharynx

299

FIGURE 13-22.  Intensity-modulated radiation therapy plan is shown for an advanced supraglottic cancer of the right false vocal cord with right-sided nodal disease; treatment included concurrent cisplatin. Axial view shows the clinical target volume (CTV1) encompassing the gross disease and margin; the primary CTV1 (red) and the nodal CTV1 (green) were delineated separately. CTV2 (blue) represents an expansion of the margin to CTV1, and CTV3 (yellow) encompasses the node-negative left neck.

Carcinoma of the Hypopharynx Fields used to treat hypopharyngeal carcinomas are designed to encompass the primary tumor and draining lymphatics regardless of the primary tumor stage (Fig. 13-23). The nodal levels at risk include levels II through V and the retropharyngeal nodes. Large lateral portals cover the primary disease, the upper jugular nodes, and the midjugular nodes. High junctional nodes, posterior cervical nodes, and retropharyngeal nodes are all included. The classic retropharyngeal nodes lie anterior to C1 and C2. To treat all of these nodes, the superior border should extend to the base of the sphenoid sinus. For pyriform sinus tumors, the ­inferior border of the lateral fields must cover the ­ entire pyriform sinus, including the apex. The entire cricoid cartilage must be encompassed with a 1-cm margin in the lateral fields. For postcricoid tumors and posterior pharyngeal wall tumors, even more generous coverage is required, often including the upper cervical esophagus. This may require a different field arrangement if the lateral fields cannot be extended inferiorly to encompass this target without the shoulders’ interfering. The posterior border is drawn to cover the posterior cervical nodes. Anteriorly, a strip of skin can often be spared by pulling it forward out of the field with clothespins or another device. The anterior border must adequately cover the anterior jugular nodes and the larynx, including the pre-epiglottic space, but can exclude the oral cavity. A matching anterior portal encompassing the lower jugular and supraclavicular lymph nodes is also used. This match is commonly done by using a mono-isocentric technique. Using a true horizontal isoplane may be disadvantageous because it may make it more difficult to adequately cover the inferior aspect of the primary tumor without treating through the shoulder. One solution is to rotate the collimation on the lateral fields such that the posterior border is more superior than the anterior border, possibly allowing the shoulders to be avoided. The exact match is maintained by using a cephalad tilt on the anterior field

FIGURE 13-23.  Simulation film shows the opposed fields used to treat a patient with carcinoma of the pyriform sinus treated with definitive radiation.

with the exact angle as the collimation on the lateral fields. The challenge remains that of where best to put a safety block over the spinal cord, because even though the fields are “exactly” matched and theoretically there is no divergence, collimation of the individual units may create hot regions at the match at the spinal cord. If there are concerns that the safety block may shield disease, the technique may not be appropriate, and a pair of caudally oriented beams or IMRT may be better for ensuring disease coverage without overlap over the spinal cord. Field design for postoperative radiation in general follows the same concepts with the same targets—the ­primary tumor bed and all the lymphatics, including the ­retropharyngeal nodes (Fig. 13-24). The stoma is included in the lateral fields, if it can be encompassed in the generous inferior border of these fields, or in the anterior field. Target delineation for IMRT given as definitive treatment can be challenging, because the inferior border is often difficult to distinguish clinically and radiographically. Use of PET for defining targets in head and neck tumors is mostly considered experimental; however, PET may be particularly useful for defining the extent of hypopharyngeal tumors, as are findings from direct laryngoscopy and upper esophagoscopy. The gross tumor is identified and a 1- to 1.5-cm margin is added to define the clinical target volume (Fig. 13-25). Subclinical doses may be delivered to ­secondary clinical target volumes such as the postcricoid area and upper esophagus. Coverage of the neck is similar to that for supraglottic cancers, except that retropharyngeal and levels II to V lymph nodes are covered in all patients with node-negative disease. Postoperative target

300

PART 3  Head and Neck

FIGURE 13-25.  Intensity-modulated radiation therapy plan is shown for an advanced hypopharynx cancer treated definitively with chemoradiation. The gross tumor volume of the left  pyriform sinus is shown in red. The clinical target volume (CTV1, green) expands the margin around the gross tumor volume. CTV2 (blue) adds an additional margin to CTV1 and encompasses the anterior ipsilateral nodes. CTV3 (yellow) encompasses the remainder of the uninvolved neck on both sides.

FIGURE 13-24.  Simulation film shows the opposed fields used for postoperative radiation therapy for a patient who had undergone laryngopharyngectomy for advanced pyriform sinus cancer.

delineation is similar to that for laryngeal cancer, except that for hypopharyngeal cancer, the retropharyngeal nodes are covered as a tertiary clinical target volume.

Dose and Fractionation Carcinoma of the Glottis T1 tumors of true vocal cords are treated to doses of 60 to 66 Gy. The lower dose is used for patients whose diagnostic procedure has removed all visible tumor, whereas higher doses are used when visible gross disease is still present. Occasionally, doses as high as 70 Gy may be appropriate for very bulky lesions. Similar doses are used for carcinoma in situ. The precise dose per fraction is controversial, and the literature abounds with studies evaluating different dose schedules. It is clear from the literature that doses lower than 2 Gy/fraction result in lower control rates, and the fraction size should not be less than 2 Gy. The greater efficacy of higher doses per fraction is less clear, although some series recommend between 2.1 and 2.5 Gy/fraction, with total doses of 60 to 63 Gy.50-52 Investigators from Osaka Medical Center for Cancer and Cardiovascular Disease reported findings from 180 patients randomly assigned to receive radiation in 2- or 2.25-Gy fractions for T1 glottic cancer.53 The 5-year local control rates were significantly higher for the 2.25-Gy/fraction group (92% versus 77% for the 2-Gy group, P = .004). The standard dose schedule for T2 glottic carcinoma is 70 Gy in 35 fractions. The role of larger fraction sizes for that stage of disease is unclear. Several centers, including M. D. Anderson, have advocated hyperfractionation for that stage of disease. We treat T2 glottic carcinoma to 76.6 Gy given twice daily at 1.2 Gy/fraction, with a 6-hour interval

between treatments. The 5-year local control rates obtained with this approach were almost 80% and were higher for those patients treated with hyperfractionated therapy than for those treated with 2-Gy fractions.54 Complicating the analysis was a subgroup of patients treated with once-daily fractions of 2.06 to 2.26 Gy. Local control rates for this subgroup were comparable to those for the group treated with hyperfractionated therapy. The RTOG, in a randomized trial of 70 Gy in 35 fractions versus 79.2 Gy in 68 fractions for T2 glottic cancer,55 found 5-year control rates to be 70% for the first group given conventionally fractionated therapy and 79% for the group given hyperfractionated therapy.55 The 9% difference between groups was not statistically significant, although the investigators observed that this 9% difference was comparable to the differences in local control rates for conventional fractionation versus hyperfractionation described in other randomized trials of head and neck cancer. The full report of this trial is eagerly anticipated. The radiation dose schedule for T3 and T4 glottic cancer being treated with curative intent is usually 70 Gy in 35 fractions–the same as that used in the Intergroup larynx preservation trial.19 While the larynx preservation trial was underway, the RTOG conducted a separate trial investigating the effect of different fractionation schedules. Although patients with glottic cancer were excluded from that trial, the results for the remaining head and neck sites are compelling enough to suggest that hyperfractionation or accelerated fractionation with a concomitant boost schedule may enhance local control in patients with ­glottic cancer.56 Because it involves fewer fractions and completes the treatment sooner, we have begun to use accelerated fractionation with a concomitant boost for patients with advanced laryngeal cancer who are to be treated with ­radiation alone. The appropriate dose and fractionation to use when chemotherapy is involved is less clear. In the Intergroup larynx preservation trial, all patients received 70 Gy in

13  The Larynx and Hypopharynx

35 fractions. No data exist to support reducing the dose after induction therapy, even in the case of an excellent response. The role of altered radiation therapy schedules after induction therapy is less clear because the tumor kinetics pattern after chemotherapy is different from that for patients whose treatment starts with radiation. Although efficacy is expected to increase with concurrent chemoradiation therapy, the ability to reduce the total dose in this setting is unproved. If concurrent chemoradiation therapy is used, we favor standard fractionation of 70 Gy in 35 fractions. Studies are evaluating more efficacious schedules, such as concomitant boost with chemotherapy, recognizing that tolerance is a key end point of potentially toxic therapies. If IMRT is to be used, typically three clinical targets are identified, the first encompassing gross disease (primary and nodal) plus a margin and the second and third targeting subclinical disease. The dose to be delivered to subclinical disease by intensity-modulated therapy (with or without chemotherapy) is controversial, as is the need to identify separate targets to receive different doses. We deliver 59 to 63 Gy to subclinical targets deemed at high risk and 54 to 57 Gy to lower-risk subclinical targets. High-risk subclinical sites are those that are common sites of metastases or close to gross disease but outside the primary and nodal clinical target volume. Treatment delivered off-protocol is typically given once daily in 33 to 35 fractions. Our findings from a dose-seeking trial previously led us to recommend 63 Gy given over the course of 7 weeks as the optimal dose for postoperative radiation.57 Doses of 57.6 Gy were considered sufficient if the risk of recurrence as judged by pathologic findings was not considered high, but postoperative radiation was indicated nevertheless. The formulas for determining risk were complex but can be summarized as follows: high-risk disease was that involving multiple nodes, extracapsular spread of nodal disease, soft tissue extension of the primary tumor, or inadequate margins. We also attempted to shorten the schedule to 5 weeks by using twice-daily fractionation for 2 weeks. On observing that differences in the rates of acute complications were minimal between patients treated with 1.8-Gy fractions and those treated with 2-Gy fractions, we established our standard postoperative dose as being 60 Gy in 30 fractions. A secondary finding of this study was the importance of the overall treatment time. Patients whose therapy (from surgery to completion of radiation therapy) lasted 13 weeks or longer had a poorer outcome.58,59 Many groups advocate higher doses for patients with positive margins or extracapsular spread of disease, but few data are available to support this practice. We treat the operative bed to 56 Gy, reducing off the spinal cord at 42 or 44 Gy to ensure that the dose is less than 45 Gy. Additional boosts to the high-risk areas are given to the final desired dose. Postoperative IMRT is delivered in 30 fractions. Clinical target volumes that may harbor disease are identified and assigned different levels of risk, with the higher-risk volumes receiving the higher doses as follows. The dose prescribed to undissected subclinical sites deemed at risk is 54 Gy, that to dissected sites not thought to originally harbor disease is 57 Gy, and that to sites with known disease before surgery is 60 Gy. Small

301

volumes of positive margin or nodal extracapsular extension can be treated to 66 Gy.

Carcinoma of the Supraglottis Definitive radiation therapy for T1 supraglottic carcinoma consists of 66 Gy given in 33 fractions. Because this stage presentation is fairly uncommon, dose fractionation data are limited. Definitive radiation therapy for T2 and T3 lesions consists of 70 Gy in 35 fractions. The RTOG fractionation study,56 which included a cohort of patients with supraglottic tumors, showed that the hyperfractionation and concomitant boost schedules were equivalent in their efficacy and superior to standard fractionation. A ­retrospective review of our experience with hyperfractionation60 suggested that hyperfractionated schedules were superior to conventional once-daily schedules, and some groups also advocate hyperfractionation for T2 and T3 lesions.40 We continue to use hyperfractionation for patients with T2 disease. However, because the findings from the RTOG trial suggested equivalence between concomitant boost and hyperfractionated schedules, we prefer to use the concomitant boost fractionated schedule for patients with intermediate- or advanced-stage disease being treated with radiation only. However, in most cases, disease at these stages is treated with concurrent chemoradiation therapy, with radiation delivered in a conventionally fractionated schedule. When conventional techniques are applied, shrinkingfield techniques are used. An off–spinal cord reduction is made to keep the dose to the spinal cord below 45 Gy. The posterior cervical strips are supplemented to 50 to 55 Gy, depending on the fractionation. Boost fields to gross disease are treated after 50 to 55 Gy (the former dose is used for conventionally fractionated therapy and the latter for hyperfractionated therapy) to the final dose. The uninvolved low neck is treated to 50 Gy at 2 Gy/fraction, with the dose prescribed to a calculated point 2.5 to 3 cm deep. If IMRT is to be used, three clinical targets typically are identified, as described for carcinoma of the glottis. The dose to be delivered by intensity-modulated therapy to subclinical disease in supraglottic carcinoma (with or without chemotherapy) is controversial. Our practice is the same for supraglottic carcinomas as for glottic carcinomas; we deliver 59 to 63 Gy to high-risk subclinical targets and 54 to 57 Gy to lower-risk subclinical targets. Patients who undergo postoperative radiation after supraglottic laryngectomy receive 55 Gy in 6 weeks to the larynx. If significant nodal disease is present, the neck is given a boost dose to 63 Gy. Higher doses to the larynx after supraglottic laryngectomy are of great concern, even though few data are available to support this concern. Some clinicians believe that the risk of significant complications, particularly laryngeal necrosis, is increased after partial laryngectomy. For the rare cases in which positive margins are encountered, we recommend treating the larynx to 60 Gy. The dose and fractionation issues for postoperative radiation therapy after total laryngectomy for supraglottic carcinoma are similar to those described for glottic carcinomas. In a maximum of 6 weeks, 60 Gy is delivered with off–spinal cord reductions made at 42 Gy and boosts delivered after the surgical bed has received 56 Gy. If the

302

PART 3  Head and Neck

low neck is uninvolved, it is treated to 50 Gy, although if portions of the operative bed are within the low-neck field, these areas are supplemented to 56 Gy. Postoperative IMRT is given as described for glottic and supraglottic carcinoma.

Pene and Fletcher65 reported a local control rate of 86%, but several of the treated patients were thought to have had invasive carcinoma because the clinical appearance of their tumors suggested T2 disease although invasion was not seen on their biopsy specimens.65

Carcinoma of the Hypopharynx

T1 Disease

Dose and fractionation issues for cancer of the hypopharynx are similar to those for carcinoma of the supraglottis. T1 disease is treated to 66 Gy in 33 fractions. We favor hyperfractionated therapy for T2 disease and find the control rates to be significantly improved by using twice-daily treatment as opposed to conventional fractionation.4 Most patients with T3 disease are treated with chemoradiation therapy or are given postoperative radiation. Patients ­treated with radiation therapy alone can be treated with hyperfractionation or concomitant boost schedules. When conventional techniques are applied, shrinkingfield techniques are used as described under “Carcinoma of the Supraglottis.” An off–spinal cord reduction is made to keep the dose to the spinal cord below 45 Gy. The posterior cervical strips are supplemented to 50 to 55 Gy depending on the fractionation schedule. Boost fields to gross disease are treated to the final dose after 50 to 55 Gy has been delivered. The uninvolved low neck is treated to 50 Gy at 2 Gy/fraction with the dose prescribed to a calculated point 2.5 to 3 cm deep. If IMRT is to be used, three clinical targets are identified as described under “Carcinoma of the Glottis.” Our practice is to deliver 59 to 63 Gy to high-risk subclinical targets and 54 to 57 Gy to lower-risk subclinical targets. After total laryngectomy, partial pharyngectomy, or neck dissection, 60 Gy is delivered over a maximum period of 6 weeks. We do not change this dose when jejunal grafts are used for pharyngeal restoration, because we have found that if the graft has not been rejected before the radiation begins, the graft is not at significantly high risk despite the use of higher doses than would be used in the treatment of abdominal malignancies. Off–spinal cord reductions are made at 42 Gy, and boosts are delivered after the surgical bed has received 56 Gy. If the low neck is uninvolved, it is treated to 50 Gy, although if portions of the operative bed are within the low-neck field, these areas are supplemented to 56 Gy. Unless there is concern regarding disease in the peristomal tissues or the stoma is incorporated in the boost because of its proximity to the primary bed, we limit the stoma to 50 Gy. Postoperative intensity-modulated therapy is delivered as described for tumors of the glottis in 30 fractions. Clinical target volumes that may harbor disease are identified, with higher-risk volumes receiving higher doses as follows. The dose prescribed to undissected subclinical sites deemed at risk is 54 Gy, that to dissected sites not thought to originally harbor disease is 57 Gy, and that to sites with known disease before surgery is 60 Gy. Small volumes of positive margin or nodal extracapsular extension can be treated to 66 Gy.

The control rates with irradiation for T1 disease are similar to those achieved with radiation for carcinoma in situ. Most series describing radiation therapy in patients with T1 disease report control rates of 85% to 95% (Table 13-2).49,52,53,61,66-75 Although local control rates are equivalent for carcinoma in situ and T1 disease, the time to local failure is generally longer for carcinoma in situ; most series report recurrences developing between 2 and 5 years after radiation therapy for in situ disease.61,65,76-79 Salvage therapy is usually successful after local failure, and ultimate control rates are greater than 95%. For some patients, voice preservation is possible with partial laryngectomy. However, many patients who need salvage therapy for a recurrent tumor require a total laryngectomy. Wang and McIntyre80 reported a small series of reirradiation for recurrence of glottic tumors after radiation therapy. They reported a moderately high success rate, with a lower than anticipated rate of complications.

Results of Radiation Therapy Carcinoma of the Glottis Reported local control rates for carcinoma in situ of the glottis treated with primary radiation therapy are high, with most series reporting control rates of 90% or more.49,61-64

T2 Disease Local control rates for T2 glottic cancer treated with radiation range from 65% to 80%, and the ultimate local control rate is about 90% (Table 13-3).6,49,54,66-68,70-72,81-84 The wide range in local control rates is in part a result of the subtleties of the relatively imprecise staging system, which lead to a moderate amount of subjective variation with respect to identifying clinical impairment, but not fixation, of vocal cord motion. The staging system also does not account for tumor volume, which varies more as disease stage increases. The need for a separate substage based on vocal cord mobility within the classification of T2 disease is controversial. Numerous series demonstrate a decrease in control, ranging from 20% to 30% for T2 lesions with impaired vocal cord mobility compared with lesions associated with mobile vocal cords.66,67,85,86 However, equally numerous series have not demonstrated a difference in outcome based on impairment of vocal cord motion.49,81,84,87 In our analysis of T2 glottic cancer, mobility did not affect local control, but subglottic extension was a significant adverse factor.54 Hyperfractionated radiation therapy has been used as an attempt to improve control rates for T2 glottic carcinoma. Although this strategy has been shown to improve local control in oropharyngeal cancer and in advanced nonglottic head and neck cancer, data supporting the use of hyperfractionated radiation therapy for T2 glottic carcinoma are sparse. Trials from the University of Florida,68 M. D. Anderson,54 and the RTOG55 showed nonsignificant trends of improved local control for patients with T2 disease treated with hyperfractionation.

T3 Disease Not surprisingly, among patients treated with radiation alone, local control rates decrease with increasing disease stage. Reported local control rates for patients with T3 glottic carcinoma treated with radiation alone range from 45%

13  The Larynx and Hypopharynx

303

TABLE 13-2.    Results of Radiation Therapy for T1 Carcinomas of the True Vocal Cords Study and Reference

Number of Patients

Local Control Rate (%)*

Local Control Rate after Salvage (%)*

Wang, 199749

665

93 (5-yr actuarial)

98 (5-yr DS)

Hendrickson, 198570

364

90

ND

Lustig et al, 198471

342

90 (3-yr actuarial)

ND

333

86 (5-yr actuarial)

ND

332

89

84

315

84

83

197961

Harwood et al,

Fletcher and Goepfert, Le et al,

198069

199767 200168

291

93% (5-yr actuarial)

58

Schwaab et al, 199472

194

86

89

Yamazaki et al, 200653

180

86 (5-yr actuarial)

92

154

89

94

137

80

82

129

90

75

126

79

88

113

93

75

Mendenhall et al,

Akine et al,

199173

Olszewski et al,

198574

Klintenberg et al, Yu et al,

199666

199752

Pellitteri et al, 199175

*Rates are crude unless otherwise specified. DS, disease specific; ND, no data.

TABLE 13-3.    Results of Radiation Therapy for T2 N0 Glottic Carcinomas Study and Reference

Number of Patients

Local Control Rate (%)*

Local Control Rate after Salvage (%)*

Barton et al, 199282

327

69 (5-yr actuarial)

ND

286

69 (5-yr actuarial)

ND

Warde et al,

19986

199749*,†

237

71-77 (5-yr actuarial)

84-92 (5-yr DS)

Garden et al, 200354

230

72 (5-yr actuarial)

82

Mendenhall et al, 200168

Wang,

228

72-80 (5-yr actuarial)

80

Karim et al,

198784

156

78 (5-yr actuarial)

92‡

Lustig et al,

198471

109

78 (3-yr actuarial)

ND

94

74

38

83

76

65

83

67

74

76

73

50

71

76

65

65

66

55

Klintenberg et al,

199666

Van den Bogaert et al,

198383

Le et al, 199767 Hendrickson, 198570 Wiggenraad et al, Schwaab et al,

199081

199472

*Rates

are crude unless otherwise specified. given for T2a and T2b separately. ‡Ultimate local control probability after salvage surgery for failures after radiation therapy. DS, disease specific; ND, no data. †Data

to 70% (Table 13-4).14,49,70,71,88-91 The feasibility of successful surgical salvage therapy also decreases with increasing initial stage; only one half to two thirds of T3 ­glottic carcinomas that are not cured with radiation therapy are ultimately cured with surgery. In the Intergroup larynx preservation trial (91-11),19 79% of patients treated with radiation alone had T3 ­laryngeal cancer; the remainder had T2 or T4 disease. Of

the 31% of patients treated with radiation alone who later underwent salvage laryngectomy, the locoregional control rate was 90%. This rate seems to compare favorably to the 74% control rate for patients treated with chemotherapy, although this difference was not tested. Nevertheless, overall survival rates were no different among patients who required laryngectomy in any of the three groups. Despite the relatively high rates of successful salvage, particularly

304

PART 3  Head and Neck

TABLE 13-4.    Results of Radiation Therapy Alone for T3 Glottic Carcinomas Study and Reference

Number of Patients

Local Control Rate (%)*

Local Control Rate after Salvage (%)*

Lundgren et al, 198889

141

44

59

Terhaard et al, 199190

104†

53 (3-yr actuarial)

53

Hinerman et al, 200788

87

67

66

199749

65

57 (5-yr actuarial)

75

Bryant et al,

199514

55

55

ND

Lustig et al,

198471

47

65 (3-yr actuarial)

ND

39

56

47

30

70

66

Wang,

Hendrickson,

198570

Croll et al, 198991 *Rates

are crude unless otherwise specified. patients with cancer in any region of larynx. ND, no data.

†Includes

for the radiation-alone group, survival rates for patients requiring laryngectomy were lower than those for patients who did not undergo laryngectomy.92 Several groups have argued that when the end point for comparison is survival, expected results for patients treated with radiation therapy and surgical salvage therapy are similar to those that could be obtained with total laryngectomy and postoperative radiation therapy.14,93 However, a randomized study to answer this question has not been done, and it is unlikely that such a trial will ever be performed. The VA randomized study of advanced laryngeal cancer concluded that the strategy of induction chemotherapy followed by radiation therapy for those who responded did not compromise survival and resulted in substantial gains in voice preservation.18 A European trial studying the same question reported that induction therapy did seem to compromise survival outcome, but closure of that trial before its completion makes its results more questionable.16 A meta-analysis of three trials involving 602 patients suggested a nonsignificant trend toward better survival among patients treated with surgery rather than induction therapy.44 Because no trial has demonstrated an improvement in survival from the addition of induction chemotherapy to radiation therapy, the role of induction chemotherapy for patients with a low likelihood of distant failure is questioned by many groups. The Intergroup larynx preservation trial,19 in trying to resolve this issue, compared radiation therapy alone with radiation plus induction or concurrent chemotherapy. That trial showed no clear differences in laryngeal preservation rates between patients given induction therapy compared with radiation only, but those rates were higher for patients given concurrent cisplatin and radiation therapy.

T4 Disease Surgery is the treatment of choice for patients with T4 glottic disease. Although the VA study19 suggested that induction chemotherapy might be appropriate for ­advanced laryngeal cancer in general, the experience at M. D. ­Anderson38 suggests that patients with T4 disease do not benefit from this approach, and we recommend total laryngectomy as the standard of care.

Few radiation therapy series describe results for patients with T4 disease. Karim and colleagues94 reported a crude local control rate of 63% for 38 patients treated; similarly, Harwood and colleagues95 reported disease control in 56% of 39 patients with T4 disease treated with radiation therapy alone. Later reports have shown mixed results, albeit with very small numbers of patients. Hinerman and colleagues88 reported a local control rate of 82% for 22 patients with T4 glottic cancer treated with radiation alone, a rate that was higher than that reported for their patients with T3 disease. Those investigators did acknowledge that theirs was a highly selected population with low-volume disease, and their current recommendation is laryngectomy for “advanced disease” and concurrent chemoradiation therapy for selected patients with low-volume disease. A group from the University of Chicago96 reported favorable results from giving chemoradiation therapy to a highly selected subgroup of patients with T4 laryngeal cancer, but this experience remains to be duplicated in a nonacademic setting. Control rates for patients with T3 N0 or T4 N0 glottic cancer treated with total laryngectomy are high. At M. D. Anderson, local control rates for such patients treated with surgery and postoperative irradiation exceeded 90%. Similar control rates were achieved with surgery alone in the absence of the following adverse features: poor differentiation, subglottic extension of greater than 1 cm, or tracheotomy done emergently before the laryngectomy.35 When these features were present, the control rate was less than 75%, and the addition of postoperative radiation is recommended in these situations.

Carcinoma of the Supraglottis T1 tumors of the supraglottis are uncommon. Reported local control rates for patients with T1 tumors treated with radiation alone range from 84% to 100%.40,49,69 In 1974, Bataini and colleagues97 reported an overall locoregional control rate of 76% for a relatively large series of patients with T1-2 N0-2 supraglottic carcinomas treated with ­radiation therapy alone. Control rates for patients with T2 disease range from 61% to 90%.40,49,69 The worst control rates are from series describing experiences with conventional ­ fractionation.

13  The Larynx and Hypopharynx

In 1980, Fletcher and Goepfert69 demonstrated the ­importance of dose escalation to 70 Gy for T2 supraglottic tumors. Studies done at M. D. Anderson, the University of Florida, and Massachusetts General Hospital have suggested that twice-daily fractionation improves the control rate in this subgroup of patients.40,49,69 T3 disease comprises an extremely heterogeneous group of tumors. Control rates with radiation therapy alone range from 40% to 75%.40,49,69 The improvement in control with twice-daily fractionation seen in T2 disease also has been demonstrated for T3 disease. Several investigators have suggested that primary ­tumor stage and the presence of nodal disease are important ­determinants of outcome. Wall and colleagues98 at M. D. Anderson and Wang and colleagues99 at Massachusetts General Hospital have reported lower rates of local control in the presence of regional disease. However, these experiences were not replicated at the University of Florida.100 Most patients with large or infiltrative T3 supraglottic primary tumors, particularly those with vocal cord fixation, are treated with surgery. Some patients have lesions amenable to supraglottic laryngectomy, whereas others require total laryngectomy. Control rates for patients undergoing supraglottic laryngectomy range from 80% to 90%.2,41,101 Many such patients receive postoperative radiation therapy because of inadequate margins or the presence of nodal disease. However, results are rarely reported separately for this subgroup, and it is difficult to determine the value of radiation therapy for these patients. In patients with stage IV supraglottic cancer treated with total laryngectomy, Goepfert and colleagues102 reported a 2-year control rate of 37% for surgery alone and a rate of 63% for combined surgery and postoperative radiation therapy. However, salvage was possible in 50% of the unirradiated patients but was rare in patients treated initially with combined therapy. The VA study18 included many patients with supraglottic cancer, although a subsite analysis was not reported. The incidence of positive regional disease among patients was moderate, and those patients most likely had supraglottic rather than glottic tumors. The patients with supraglottic tumors might have accounted for most of the reduction in distant failure in the chemotherapy group of the study, but this interpretation is only speculative. A similar observation was made in the Intergroup larynx preservation trial.19

Carcinoma of the Hypopharynx Few series have reported outcome data for early hypopharyngeal carcinoma. Control rates range from 74% to 90% for T1 disease and 59% to 79% for T2 disease.4,103,104 Results from studies done in the 1990s have suggested that use of hyperfractionated radiation can improve control rates. At M. D. Anderson, the local control rate for patients with early hypopharyngeal carcinoma treated with hyperfractionated radiation therapy was 86%.4 At the ­University of Florida, Amdur and colleagues103 reported a local control rate of 94% for patients with T2 pyriform sinus cancer treated with twice-daily radiation therapy, a rate 16% higher than that for patients given once-daily treatment. Wang104 reported a 5-year actuarial local control rate of 76% for patients with T2 pyriform sinus cancer treated with split-course accelerated radiation therapy. The treatment of choice for patients with T3 or T4 hypopharyngeal cancer is surgery. Most patients are advised

305

to receive radiation therapy postoperatively. Vandenbrouck and colleagues105 performed a randomized trial comparing preoperative and postoperative radiation therapy in the treatment of hypopharyngeal tumors and demonstrated a significant improvement in survival when the radiation was delivered after surgery. Locoregional control rates in the ­reported series range from 51% to 89%.105-107 Despite this relatively wide range, the 5-year survival rates seem relatively consistent across series, ranging from 22% to 43%. Analyzing patients treated at M. D. Anderson between 1949 and 1976, El Badawi and colleagues106 found that patients who received surgery and radiation therapy had significantly higher local control and 5-year survival rates than did patients who underwent surgery alone. A later study by Frank and colleagues107 found that patients with hypopharyngeal cancer who received postoperative irradiation had only a 14% locoregional failure rate, compared with 57% for those treated with surgery only, despite the fact that patients treated with combined therapy had more advanced disease. Hinerman and colleagues108 reported a locoregional control rate of 77% for 80 patients treated with surgery and postoperative radiation. Not surprisingly, disease control rates were worse among a small subgroup of patients with positive margins than among patients with clear or close margins.108 The EORTC conducted a voice preservation study for patients with hypopharyngeal cancer17 that was similar to the larynx preservation trial conducted by the VA. In the EORTC trial, induction chemotherapy, followed by radiation therapy for complete responders or surgery for incomplete responders, was compared with immediate pharyngolaryngectomy and postoperative radiation. Survival rates were equivalent in the two treatment groups, but patients in the chemotherapy group had a voice preservation rate of 35% at 5 years. The complete response rate to cisplatin and fluorouracil was more than 50%. Even though the trial was for patients with stage III or IV cancer, only a minority had T4 disease. In summary, improvements in disease control and survival have been demonstrated in many trials of head and neck cancer comparing concurrent chemoradiation therapy with radiation alone. Many of the trials included small subsets of patients with hypopharyngeal cancer, but the numbers of such patients were typically small, and specific evidence-based recommendations of organ preservation strategies for this disease site are difficult to make. Nevertheless, the M. D. Anderson recommendation is to treat T3 hypopharyngeal cancer with concurrent chemoradiation therapy rather than surgery.

Complications Complications of radiation therapy depend on the radiation dose and the irradiated volume. Patients with T1 laryngeal disease treated with fields ranging from 25 cm2 to 36 cm2 and doses of 66 Gy have few late complications. Fletcher and Goepfert69 described a 1% crude incidence of late edema in patients with T1 disease treated between 1962 and 1979. Garden and colleagues54 at M. D. Anderson analyzed 230 patients with T2 glottic tumors and found that only 10 (4.3%) developed severe late complications (including dysfunctional larynges, laryngeal edema requiring hospitalization, and chondronecrosis). None of those 10 patients required

306

PART 3  Head and Neck

hospitalization to manage the complications. Complication rates are higher after radiation therapy for supraglottic and hypopharyngeal tumors than after radiation therapy for early-stage glottic carcinomas. Investigators at the University of Florida reported a 4% incidence of severe acute and late complications in patients receiving primary radiation therapy for supraglottic carcinomas.40 In a series of 236 patients with head and neck cancer treated with hyperfractionated radiation therapy at M. D. Anderson, the crude rate of severe late complications was 9%. Two deaths were attributed to complications, and two patients who had been rendered free of disease required total laryngectomy for management of the complications.60 Many of these severe complications occurred during the early days of using hyperfractionated radiation. These early experiences led to recognition of the importance of increasing the interfractional interval to at least 6 hours. Since that adjustment was made, complication rates have decreased. The total rates of severe toxic effects, including acute and late grade 3 to 5 complications, reported from the Intergroup larynx preservation trial19 ranged from 61% for patients treated with radiation only to 82% for patients treated with concurrent chemoradiation therapy. However, fewer severe late effects were experienced by the patients treated with concurrent therapy than by patients treated with radiation alone. Moreover, at 2 years no differences were found between groups in the proportion of patients reporting difficulty in swallowing.

REFERENCES   1. Byers RM, Wolf PF, Ballantyne AJ. Rationale for elective modified neck dissection. Head Neck Surg 1988;10:160-167.   2. Lee NK, Goepfert H, Wendt CD. Supraglottic laryngectomy for intermediate stage cancer: U.T.M. D. Cancer Center experience with combined therapy. Laryngoscope 1990;100:831-836.   3. Lindberg RD. Distribution of cervical lymph node metastases from squamous cell carcinoma of the upper respiratory and digestive tracts. Cancer 1972;29:1446-1449.   4. Garden AS, Morrison WH, Clayman GL, et al. Early squamous cell carcinoma of the hypopharynx: outcomes of treatment with radiation alone to the primary disease. Head Neck 1996;18:317322.   5. Merino OR, Lindberg RL, Fletcher GH. An analysis of distant metastases from squamous cell carcinoma of the upper respiratory and digestive tracts. Cancer 1977;40:145-151.   6. Warde P, O’Sullivan B, Bristow R, et al. T1/T2 glottic cancer managed by external beam radiotherapy: the influence of pretreatment hemoglobin on local control. Int J Radiat Oncol Biol Phys 1988;41:347-353.   7. Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 33-45, 47-57.   8. McGuirt WF, Blalock D, Koufman JA, et al. Comparative voice results after laser resection or irradiation of T1 vocal cord carcinoma. Arch Otolaryngol Head Neck Surg 1994;120:951-955.   9. Hirano M, Hirade Y, Kawasaki H. Vocal function following carbon dioxide laser surgery for glottic carcinoma. Ann Otol Rhinol Laryngol 1985;94:232-235. 10. Harrison LB, Solomon B, Miller S, et al. Prospective computerassisted voice analysis for patients with early stage glottic cancer: a preliminary report of the functional result of laryngeal irradiation. Int J Radiat Oncol Biol Phys 1990;19:123-127. 11. Foote R, Buskirk S, Grado G, et al. Has radiotherapy become too expensive to be considered a treatment option for early glottic cancer?. Head Neck Surg 1997;19:692-700. 12. Mittal B, Rao DV, Marks JE, et al. Comparative cost analysis of hemilaryngectomy and irradiation for early glottic carcinoma. Int J Radiat Oncol Biol Phys 1983;9:407-408.

13. Grégoire V, Hamoir M, Rosier JF, et al. Cost-minimization analysis of treatment options for T1N0 glottic squamous cell carcinoma: comparison between external radiotherapy, ­ laser microsurgery and partial laryngectomy. Radiother Oncol 1999;53:1-13. 14. Bryant GP, Poulsen MG, Tripcony L, et al. Treatment decision in T3N0M0 glottic carcinoma. Int J Radiat Oncol Biol Phys 1995;31:285-293. 15. Parsons J, Mendenhall W, Stringer S, et al. T4 laryngeal carcinoma: radiotherapy alone with surgery reserved for salvage. Int J Radiat Oncol Biol Phys 1998;40:549-552. 16. Richard J, Sancho-Garnier H, Pessey J, et al. Randomized trial of induction chemotherapy in larynx carcinoma. Oral Oncol 1998;34:224-228. 17. Lefebvre JL, Chevalier D, Luboinski B, et al. Larynx preservation in pyriform sinus cancer: preliminary results of a European Organization for Research and Treatment of Cancer phase III trial. EORTC Head and Neck Cancer Cooperative Group. J Natl Cancer Inst 1996;88:890-899. 18. Department of Veterans Affairs Laryngeal Cancer Study Group. Induction chemotherapy plus radiation compared with surgery plus radiation in patients with advanced laryngeal cancer. N Engl J Med 1992;324:1685-1690. 19. Forastiere AA, Goepfert H, Maor M, et al. Concurrent chemotherapy and radiotherapy for organ preservation in advanced laryngeal cancer. N Engl J Med 2003;349:2091-2098. 20. Brizel D, Albers M, Fisher S, et al. Hyperfractionated irradiation with or without concurrent chemotherapy for locally advanced head and neck cancer. N Engl J Med 1998;338:1798-1804. 21. Adelstein D, Lavertu P, Saxton J, et al. Mature results of a phase III randomized trial comparing concurrent chemoradiotherapy with radiation therapy alone in patients with stage III and IV squamous cell carcinoma of the head and neck. Cancer 2000;88:876-883. 22. McNeil B, Weichselbaum R, Pauker S. Speech and survival: tradeoffs between quality and quantity of life in laryngeal cancer. N Engl J Med 1981;305:982-987. 23. Van der Donk J, Levendag P, Kuijpers A, et al. Patient participation in clinical decision-making for treatment of T3 laryngeal cancer: a comparison of state and process utilities. J Clin Oncol 1995;13:2369-2378. 24. Ferlito A, Polidoro F, Rossi M. Pathological basis and clinical aspects of treatment policy in carcinoma in situ of the larynx. J Laryngol Otol 1981;95:141-154. 25. McGuirt WF, Browne JD. Management decisions in laryngeal carcinoma in situ. Laryngoscope 1991;101:125-129. 26. Wolfensberger M, Dort JC. Endoscopic laser surgery for early glottic carcinoma: a clinical and experimental study. Laryngoscope 1990;100:1100-1105. 27. Maran AG, Mackenzie IJ, Stanley RE. Carcinoma in situ of the larynx. Head Neck Surg 1984;7:28-31. 28. Strong MS. Laser excision of carcinoma of the larynx. Laryngoscope 1975;85:1286-1289. 29. Sessions D, Maness G, Mcswain B. Laryngofissure in the treatment of carcinoma of the vocal cord. Laryngoscope 1964;75:490502. 30. Soo KC, Shah JP, Gopinath KS, et al. Analysis of prognostic variables and results after supraglottic partial laryngectomy. Am J Surg 1988;156:301-305. 31. Laccourreye O, Weinstein G, Brasnu D, et al. Vertical partial laryngectomy: a critical analysis of local recurrence. Ann Otol Rhinol Laryngol 1991;100:68-71. 32. Mendenhall WM, Parsons JT, Stringer SP, et al. T1-T2 vocal cord carcinoma: a basis for comparing the results of radiotherapy and surgery. Head Neck Surg 1988;10:373-377. 33. Laccourreye H, Laccourreye O, Weinstein G, et al. Supracricoid laryngectomy with cricohyoidopexy: a partial laryngeal procedure for selected supraglottic and transglottic carcinomas. Laryngoscope 1990;100:735-741. 34. Laccourreye O, Brasnu D, Bassot V, et al. Cisplatin-fluorouracil exclusive chemotherapy for T1-T3N0 glottic squamous cell carcinoma complete clinical responders: five-year results. J Clin Oncol 1996;14:2331-2336.

13  The Larynx and Hypopharynx 35. Yuen A, Medina J, Goepfert H, et al. Management of stage T3 and T4 carcinomas. Am J Surg 1984;148:467-472. 36. Bernier J, Domenge C, Ozsahin M, et al. Postoperative irradiation with or without concomitant chemotherapy for locally advanced head and neck cancer. N Engl J Med 2004;350:19451952. 37. Cooper JS, Pajak TF, Forastiere AA, et al. Postoperative concurrent radiotherapy and chemotherapy for high-risk squamous-cell carcinoma of the head and neck. N Engl J Med 2004;350:19371944. 38. Shirinian MH, Weber RS, Lippman SM, et al. Laryngeal preservation by induction chemotherapy plus radiotherapy in locally advanced head and neck cancer: the M.D. Anderson Cancer Center experience. Head Neck 1994;16:39-44. 39. Steiner W. Results of curative laser microsurgery of laryngeal carcinomas. Am J Otolaryngol 1993;14:116-121. 40. Hinerman RW, Mendenhall WM, Amdur RJ, et al. Carcinoma of the supraglottic larynx: treatment results with radiotherapy alone or with planned neck dissection. Head Neck 2002;24:456-467. 41. Bocca E, Pignataro O, Oldini C. Supraglottic laryngectomy: 30 years of experience. Ann Otol Rhinol Laryngol 1983;92:14-18. 42. Corry J, Smith JG, Peters LJ. The concept of a planned neck dissection is obsolete. Cancer J 2001;7:472-474. 43. Robbins KT, Davidson W, Peters LJ, et al. Conservation surgery for T2 and T3 carcinomas of the supraglottic larynx. Arch Otolaryngol Head Neck Surg 1988;114:421-426. 44. Pignon J, Bourhis J, Domenge C, et al. Chemotherapy added to locoregional treatment for head and neck squamous-cell carcinoma: three meta-analyses of updated individual data. Lancet 2000;355:949-955. 45. Adelstein DJ, Li Y, Adams GL, et al. An intergroup phase III comparison of standard radiation therapy and two schedules of concurrent chemoradiotherapy in patients with unresectable squamous cell head and neck cancer. J Clin Oncol 2003;21:92-98. 46. Foote RL, Grado GL, Buskirk SJ, et al. Radiation therapy for glottic cancer using 6-MV photons. Cancer 1996;77:381-386. 47. Sombeck MD, Kalbaugh KJ, Mendenhall WM, et al. Radiotherapy for early vocal cord cancer: a dosimetric analysis of 60-Co versus 6 MV photons. Head Neck 1996;18:167-173. 48. Yom SS, Morrison WH, Ang KK, et al. Two-field versus threefield irradiation technique in the postoperative treatment of head-and-neck cancer. Int J Radiat Oncol Biol Phys 2006;66:469476. 49. Wang CC. Carcinoma of the larynx. In Wang CC (ed): Radiation Therapy for Head and Neck Neoplasms, 3rd ed. New York: Wiley-Liss, 1997, pp 221–255. 50. Burke LS, Greven KM, McGuirt WT, et al. Definitive radiotherapy for early glottic carcinoma: prognostic factors and implications for treatment. Int J Radiat Oncol Biol Phys 1997;38: 37-42. 51. Rudoltz MS, Benammar A, Mohiuddin M. Prognostic factors for local control and survival in T1 squamous cell carcinoma of the glottis. Int J Radiat Oncol Biol Phys 1993;26:767-772. 52. Yu E, Shenouda G, Beaudet MP, et al. Impact of radiation therapy fraction size on local control of early glottic carcinoma. Int J Radiat Oncol Biol Phys 1997;37:587-591. 53. Yamazaki H, Nishiyama K, Tanaka E, et al. Radiotherapy for early glottic carcinoma (T1N0M0): results of prospective randomized study of radiation fraction size and overall treatment time. Int J Radiat Oncol Biol Phys 2006;64:77-82. 54. Garden AS, Forster K, Wong PF, et al. Results of radiotherapy for T2N0 glottic carcinoma: does the “2” stand for twice-daily treatment? Int J Radiat Oncol Biol Phys 2003;55:322-328. 55. Trotti A, Pajak T, Emami B, et al. A randomized trial of hyperfractionation versus standard fractionation in T2 squamous cell carcinoma of the vocal cord. Int J Radiat Oncol Biol Phys 2006;66(Suppl 1):S15. 56. Fu KK, Pajak TF, Trotti A, et al. A Radiation Therapy Oncology Group (RTOG) phase III randomized study to compare hyperfractionation and two variants of accelerated fractionation to standard fractionation radiotherapy for head and neck squamous cell carcinomas: first report of RTOG 9003. Int J Radiat Oncol Biol Phys 2000;48:7-16.

307

57. Peters LJ, Goepfert H, Ang KK, et al. Evaluation of the dose for postoperative radiation therapy of head and neck cancer: first report of a prospective randomized trial. Int J Radiat Oncol Biol Phys 1993;26:3-12. 58. Ang KK, Trotti A, Brown BW, et al. Randomized trial addressing risk features and time factors of surgery plus radiotherapy in advanced head-and-neck cancer. Int J Radiat Oncol Biol Phys 2001;51:571-578. 59. Rosenthal DI, Liu L, Lee JH, et al. Importance of the treatment package time in surgery and postoperative radiation therapy for squamous carcinoma of the head and neck. Head Neck 2002;24:115-126. 60. Garden AS, Morrison WH, Ang KK, et al. Hyperfractionated radiation in the treatment of squamous cell carcinomas of the head and neck: a comparison of two fractionation schedules. Int J Radiat Oncol Biol Phys 1995;31:493-502. 61. Harwood AR, Hawkins NV, Rider WD, et al. Radiotherapy of early glottic cancer. I. Int J Radiat Oncol Biol Phys 1979;5: 473-476. 62. Kalter PO, Lubsen H, Delemarre JFM, et al. Squamous hyperplasia of the larynx (a clinical follow-up study). J Laryngol Otol 1987;101:579-588. 63. Fernberg JO, Ringborg U, Silfversward C, et al. Radiation therapy in early glottic cancer: analysis of 177 consecutive cases. Acta Otolaryngol 1989;108:478-481. 64. Smitt MC, Goffinet DR. Radiotherapy for carcinoma-in-situ of the glottic larynx. Int J Radiat Oncol Biol Phys 1994;28: 251-255. 65. Pene F, Fletcher GH. Results in irradiation of the in situ carcinoma of the vocal cord. Cancer 1976;37:2586-2590. 66. Klintenberg C, Lundgren J, Adell G, et al. Primary radiotherapy of T1 and T2 glottic carcinoma-analysis of treatment results and prognostic factors in 223 patients. Acta Oncol 1996;35(Suppl 8):81-86. 67. Le QX, Fu KK, Kroll S, et al. Influence of fraction size, total dose, and overall time on local control of T1-T2 glottic carcinoma. Int J Radiat Oncol Biol Phys 1997;39:115-126. 68. Mendenhall WM, Amdur RJ, Morris CG, et al. T1-T2N0 squamous cell carcinoma of the glottic larynx treated with radiation therapy. J Clin Oncol 2001;19:4029-4036. 69. Fletcher GH, Goepfert H. Larynx and pyriform sinus. In Fletcher G (ed): Textbook of Radiotherapy, 3rd ed. Philadelphia: Lea & Febiger, 1980, pp 330–363. 70. Hendrickson FR. Radiation therapy treatment of larynx cancers. Cancer 1985;55:2058-2061. 71. Lustig RA, MacLean CJ, Hanks GE, et al. The patterns of care outcome studies: results of the national practice in carcinoma of the larynx. Int J Radiat Oncol Biol Phys 1984;10:2357-2362. 72. Schwaab G, Mamelle G, Lartigau E, et al. Surgical salvage treatment of T1/T2 glottic carcinoma after failure of radiotherapy. Am J Surg 1994;168:474-475. 73. Akine Y, Tokit N, Ogino T, et al. Radiotherapy of T1 glottic cancer with 6 MeV x-rays. Int J Radiat Oncol Biol Phys 1991;20:1215-1218. 74. Olszewski SJ, Vaeth JM, Green JP, et al. The influence of field size, treatment modality, commissure involvement and histology in the treatment of early vocal cord cancer with irradiation. Int J Radiat Oncol Biol Phys 1985;11:1333-1337. 75. Pellitteri PK, Kennedy TL, Vrabec DP, et al. Radiotherapy—the mainstay in the treatment of early glottic carcinoma. Arch Otolargol Head Neck Surg 1991;117:297-301. 76. Hintz BL, Kagan AR, Nussbaum H, et al. A watchful waiting policy for in situ carcinoma of the vocal cords. Arch Otolaryngol 1981;107:746-751. 77. MacLeod PM, Daniel F. The role of radiotherapy in in-situ carcinoma of the larynx. Int J Radiat Oncol Biol Phys 1990;18: 113-117. 78. Sung DI, Chang CH, Harisiadis L, et al. Primary radiotherapy for carcinoma in situ and early invasive carcinoma of the glottic larynx. Int J Radiat Oncol Biol Phys 1979;15:467-472. 79. Elman AJ, Goodman M, Wang CC, et al. In situ carcinoma of the vocal cords. Cancer 1979;43:2422-2428.

308

PART 3  Head and Neck

80. Wang C, McIntyre J. Re-irradiation of laryngeal carcinomatechniques and results. Int J Radiat Oncol Biol Phys 1993;26: 783-785. 81. Wiggenraad RG, Terhaard CH, Horidjik GJ, et al. The importance of vocal cord mobility in T2 laryngeal cancer. Radiother Oncol 1990;18:321-327. 82. Barton MB, Keane TJ, Gadalla T, et al. The effect of treatment time and treatment interruption on tumor control following radical radiotherapy of laryngeal cancer. Radiother Oncol 1992;23:137-143. 83. Van den Bogaert W, Ostyn F, van der Schueren E. The significance of extension and impaired mobility in cancer of the vocal cord. Int J Radiat Oncol Biol Phys 1983;9:181-184. 84. Karim AB, Kralendonk JH, Yap LY, et al. Heterogeneity of stage II glottic carcinoma and its therapeutic implications. Int J Radiat Oncol Biol Phys 1987;13:313-317. 85. Harwood AR, Beale FA, Cummings BJ, et al. T2 glottic cancer: an analysis of dose-time-volume factors. Int J Radiat Oncol Biol Phys 1981;7:1501-1505. 86. Kelly MD, Hahn SS, Spaulding CA, et al. Definitive radiotherapy in the management of stage I and II carcinomas of the glottis. Ann Otol Rhinol Laryngol 1989;98:235-239. 87. Howell-Burke D, Peters LJ, Goepfert H, et al. T2 glottic cancer. Arch Otolaryngol Head Neck Surg 1990;116:830-835. 88. Hinerman R, Mendenhall W, Morris C, et al. T3 and T4 true vocal cord squamous carcinomas treated with external beam irradiation: a single institution’s 35-year experience. Am J Clin Oncol 2007;30:181-185. 89. Lundgren JAV, Gilbert RW, Van Nostrand AWP, et al. T3N0M0 glottic carcinoma—a failure analysis. Clin Otolaryngol 1988;13: 455-465. 90. Terhaard CHF, Karim ABMF, Hoogenraad WJ, et al. Local control in T3 laryngeal cancer treated with radical radiotherapy, time dose relationship: the concept of nominal standard dose and linear quadratic model. Int J Radiat Oncol Biol Phys 1991;20:1207-1214. 91. Croll GA, Gerritsen GJ, Tiwari RM, et al. Primary radiotherapy with surgery in reserve for advanced laryngeal carcinoma Results and complications. Eur J Surg Oncol 1989;15:350-356. 92. Weber RS, Berkey BA, Forastiere A, et al. Outcome of salvage total laryngectomy following organ preservation therapy: the Radiation Therapy Oncology Group trial 91-11. Arch Otolaryngol Head Neck Surg 2003;129:44-49. 93. Mendenhall W, Parsons J, Mancuso A, et al. Definitive radiotherapy for T3 squamous cell carcinoma of the glottic larynx. J Clin Oncol 1997;15:2394-2402. 94. Karim A, Kralendonk J, Njo K, et al. Radiation therapy for advanced (T3-T4N0-N3M0) laryngeal carcinoma: the need for a change of strategy. A radiotherapeutic viewpoint. Int J Radiat Oncol Biol Phys 1987;13:1625-1633.

95. Harwood A, Beale F, Cummings B, et al. T4N0M0 glottic cancer: an analysis of dose-time volume factors. Int J Radiat Oncol Biol Phys 1981;7:1507-1512. 96. Knab B, Salama J, Stenson K, et al. Definitive chemoradiotherapy for T4 laryngeal squamous cell carcinoma [abstract]. Int J Radiat Oncol Biol Phys 2006;66(Suppl 1):S14. 97. Bataini JP, Ennuyer A, Poncet P, et al. Treatment of supraglottic cancer by radical high dose radiotherapy. Cancer 1974;33: 1253-1262. 98. Wall TJ, Peters LJ, Brown BW, et al. Relationship between lymph nodal status and primary tumor control probability in tumors of the supraglottic larynx. Int J Radiat Oncol Biol Phys 1985;11:1895-1902. 99. Wang CC, Nakfoor BM, Spiro IJ, et al. Role of accelerated fractionated irradiation for supraglottic carcinoma: assessment of results. Cancer J Sci Am 1997;3:88-91. 100. Freeman DE, Mendenhall WM, Parsons JT, et al. Does neck stage influence local control in squamous cell carcinomas of the head and neck? Int J Radiat Oncol Biol Phys 1992;23:733-736. 101. Ogura JH, Marks JE, Freeman RB. Results of conservation surgery for cancers of the supraglottis and pyriform sinus. Laryngoscope 1980;90:591-600. 102. Goepfert H, Jesse RH, Fletcher GH, et al. Optimal treatment for the technically resectable squamous cell carcinoma of the supraglottic larynx. Laryngoscope 1975;85:14-32. 103. Amdur R, Mendenhall W, Stringer S, et al. Organ preservation with radiotherapy for T1-T2 carcinoma of the pyriform sinus. Head Neck 2001;23:353-362. 104. Wang CC. Carcinoma of the hypopharynx. In Wang CC (ed): Radiation Therapy for Head and Neck Neoplasms, 3rd ed. New York: Wiley-Liss, 1997, pp 205–220. 105. Vandenbrouck C, Sancho H, Le Fur R, et al. Results of a randomized clinical trial of preoperative irradiation versus postoperative in treatment of tumors of the hypopharynx. Cancer 1977;39:1445-1449. 106. El Badawi SA, Goepfert H, Fletcher GH, et al. Squamous cell carcinoma of the pyriform sinus. Laryngoscope 1982;92: 357-364. 107. Frank JL, Garb JL, Kay S, et al. Postoperative radiotherapy improves survival in squamous cell carcinoma of the hypopharynx. Am J Surg 1994;168:476-480. 108. Hinerman RW, Morris CG, Amdur RJ, et al. Surgery and postoperative radiotherapy for squamous cell carcinoma of the larynx and pharynx. Am J Clin Oncol 2006;29:613-621.

The Orbit 14 David L. Schwartz, Stella K. Kim, AND K. Kian Ang ANATOMY Response of Normal Orbital Structures to Irradiation Eyelashes and Eyelid Lacrimal Gland Cornea Lens Sclera Retina Optic Nerve and Chiasm

BENIGN ORBITAL DISEASES Graves’ Ophthalmopathy Orbital Pseudotumor Pterygium MALIGNANT ORBITAL DISEASES Ocular Melanoma Orbital Lymphomas Metastasis to the Orbit

The eye and surrounding orbit are subject to malignant and benign processes that exact steep tolls in quality of life, frequently outpacing the accompanying risk to survival. For such conditions, radiation is used to treat for cure while preserving orbital contents and function. Although orbital irradiation is typically safe, the chronic effects of ­treatmentrelated morbidity can have drastic effects on ­visual function. For benign orbital diseases, radiation therapy typically is used as an adjunct to surgery or is reserved for cases in which other treatments have failed. For malignant orbital diseases, however, radiation therapy has emerged as the initial treatment of choice given that surgical alternatives range from partial excision to formal enucleation. Technical advances in radiation delivery, such as particle beam treatment, hold promise for further improvements in safe and effective treatment. In all cases, optimal outcomes require careful patient selection.

Anatomy The outer covering of the eyeball is composed of three main tissue layers (Fig. 14-1). 1 The external fibrous coat consists of the sclera, cornea, and corneal ­ limbus. The middle vascular coat, the uvea, consists of the choroid, ciliary body, and iris. The inner neural coat, the ­retina, is the sensory region. Within the outer covering of the eyeball lie the crystalline lens, vitreous humor, and aqueous humor. The six extraocular muscles controlling eye movement insert into the sclera. The orbital contents that hold the eyeball in place consist of the extraocular muscles, the nerves and vessels that serve the globe, and in-filling fatty tissues. The orbital contents reside in a protective cone consisting of seven bones of the skull that together compose the bony orbit (Figs. 14-2 and 14-3).2 In the treatment of orbital diseases, detailed knowledge of the spatial relationships between ­anatomic structures allows radiation oncologists to spare as much normal tissue as possible without compromising coverage of the target volume. From topographic measurements of 66 patients made from computed tomography (CT) scans with eyelid markers, Karlsson and colleagues3 determined that the

mean distance from the anterior surface of the superior eyelid to the posterior surface of the lens was 10.1 ± 1 mm and that the mean distance from the posterior surface of the lens to the intercanthal line was 8.5 ± 2.6 mm. The median sagittal globe diameter was 24.6 mm.

Response of Normal Orbital Structures to Irradiation The various orbital tissues each respond differently to ­irradiation. In general, radiosensitivity decreases gradually from the anterior structures to the posterior structures. Rohrschneider, one of the first to address the topic, ­reported that the most radiosensitive structures, in order of decreasing sensitivity, were the lens, conjunctiva, cornea, uvea, and retina and that the least radiosensitive structure was the optic nerve.4 Later findings regarding the radiosensitivity of orbital structures5 confirm Rohrschneider’s early findings.

Lateral rectus muscle Schlemm’s canal Iris Lens Pupil Anterior chamber Cornea Posterior chamber Ciliary body Ora serrata Medial rectus muscle

Sclera Choroid Retina Fovea centralis Central retinal vein and artery Lamina cribrosa Optic nerve

FIGURE 14-1.  Horizontal section of the right eye. (From Saude T. The outer coats of the eye. In Saude T [ed]: Ocular Anatomy and Physiology. Oxford, UK: Blackwell Science, 1993, pp 11-25.)

309

310

Part 3  Head and Neck Lesser wing of sphenoid

Optic foramen

Supraorbital notch

Superciliary ridge

Orbital plate of great wing of sphenoid

Trochlear fossa Anterior ethmoidal foramen Maxillary process

Fossa for lacrimal gland Zygomatic process

Ethmoid

Superior orbital fissure Zygomatic tubercle Lateral orbital tubercle

Nasal bone Lacrimal bone and fossa Sutura notha

Zygomatic foramen

Lacrimal tubercle

Zygomaticofacial foramen

Orbital plate of maxilla

Zygomatic bone

Infraorbital suture

Inferior orbital Maxillary fissure process

Infraorbital groove

Infraorbital foramen

FIGURE 14-2.  The right orbit viewed along its axis. (From Bron AJ, Tripathi RC, Tripathi BJ. Wolff’s Anatomy of the Eye and Orbit, 8th ed. London: Arnold, 1997.)

Frontal sinus Trochlear fossa Ethmoid Lacrimal Lacrimal fossa Frontal process Nasal bone

Frontoethmoidal sinus Anterior ethmoidal foramen Frontal Posterior ethmoidal foramen Optic foramen Sphenoid Pituitary Orbital process of palatine

Infraorbital canal Lacrimal bone

Foramen rotundum Sphenoid process of palatine bone Lateral pterygoid plate

Uncinate process of ethmoid Inferior concha Antrum

Pterygoid hamulus

Vertical plate of palatine bone

FIGURE 14-3.  The medial wall of the orbit. (From Bron AJ, Tripathi RC, Tripathi BJ. Wolff’s Anatomy of the Eye and Orbit, 8th ed. London: Arnold, 1997.)

Eyelashes and Eyelid The eyelashes are end organs of touch and, on contact, initiate a blink reflex to protect the eye. Doses of 20 to 30 Gy or more given in 2- to 3-Gy fractions cause alopecia of the eyelashes and abolish this protective mechanism, resulting in irritation of the conjunctiva and corneal surfaces.6 ­Occasionally, regrowth of the lashes may produce trichiasis and entropion requiring intervention.7

Irradiation of the eyelid produces observable effects similar to those produced by irradiation of other skin and mucosa in the body, and with commonly prescribed radiation doses, healing is typically complete. However, the skin and mucosa of the eyelid are among the more delicate tissues in the body, and side effects occurring in this region are more uncomfortable for the patient than those arising in other regions.

14 The Orbit

Lacrimal Gland Radiation treatment fields that include the lacrimal gland can cause atrophy and loss of glandular function, leading to dry eye syndrome. Dry eye symptoms usually can be ­sufficiently treated with artificial tears or lubricants. Chronic irritation of the cornea by the eyelid, however, may lead to a painful eye with visual loss because of corneal laceration and subsequent ulceration with infection, opacification, or vascularization. Parsons and colleagues8 examined the dose-response relationship for dry eye complications using data from the University of Florida and from the literature. In that review, no complications occurred with doses up to 30 Gy. The incidence of injury was 5% to 25% in the range of 30 to 40 Gy and increased steeply for doses above 40 Gy, reaching 100% for doses of 57 Gy or more. With doses of radiation in the curative range, stenosis or even obstruction of the nasolacrimal duct may occur. The main symptom is epiphora, which may be remedied with recanalization of the duct.

Cornea It is difficult to distinguish direct radiation injury to the cornea from indirect injury resulting from dry eye syndrome because in most reported series, the entire eye is included in the treatment volume.9-11 Radiation-induced keratitis with various amounts of tearing may develop during treatment but may also manifest a few weeks or months after treatment.9 Radiation-induced corneal injury manifests as multiple, tiny ulcers in the corneal epithelium seen under fluorescein staining. This type of injury typically occurs after doses of 30 to 50 Gy given over 4 to 5 weeks12,13 and gradually resolves over 4 to 6 weeks with conservative treatment without sequelae. Larger corneal ulcerations may occur with doses greater than 50 Gy in 4 to 5 weeks. Merriam and colleagues9 reported three severe corneal ulcerations in 25 patients treated for rhabdomyosarcoma of the orbit with 60 Gy (22.5 MeV) given over 5 to 6 weeks. The cornea received the full dose, and perforation in two patients necessitated enucleation. Chan and Shukovsky10 reported that with a follow-up time of 2 years, corneal lesions occurred in 15% of patients treated with 60 Gy over 6 weeks to the entire eye for nasal cavity and paranasal sinus tumors. When intra-arterial fluorouracil was given along with radiation, all patients developed corneal lesions. Bessell and colleagues11 reported corneal ulceration in 2 of 13 patients treated with 40 to 49 Gy over 3 or 4 weeks for orbital lymphoma. However, both patients also had tear film instability, which most likely contributed to the ulcerative process. In both cases, the corneal ulcer eventually healed. Corneal stromal edema may be seen with doses of 40 to 50 Gy in 2 to 3 weeks.9 The edema is typically self-limiting, resolving over 2 to 4 weeks, and can be adequately treated with topical antibiotics and steroids. Prolonged edema of the stroma is uncommon but is more likely with doses greater than 50 Gy—usually on the order of 70 to 80 Gy. Reporting on a series of patients treated at The ­University of Texas M. D. Anderson Cancer Center, Jiang and colleagues14 suggested that visual impairment from corneal injury most likely resulted from a combination of lacrimal gland injury and direct effects of radiation on the cornea. In that series, no patients who had lacrimal and corneal

311

shielding during irradiation developed visual ­ impairment. In contrast, two of eight patients treated with lacrimal shielding only and two of 12 patients treated with cornealens shielding only developed corneal injury. The incidence of visual impairment at 2 years was 81% to 88% with doses of 56 to 74.5 Gy to the lacrimal gland (corneal dose, 31 to 41 Gy) with or without chemotherapy, compared with 17% for doses of 42 to 45 Gy (corneal dose, 23 to 30 Gy) without chemotherapy. The overall treatment time ranged from 29 to 52 days. The median time to manifestation of visual impairment was 9 months (range, 1 to 31 months). If the cornea and conjunctiva are not at risk for harboring subclinical disease, a practical strategy for decreasing the incidence of corneal complications is to irradiate patients with their eyelids open for the anterior field such that the cornea and conjunctiva are located in the buildup region of the megavoltage beams.

Lens The lens is the most radiosensitive organ in the body. The lens itself consists largely of fiber cells covered anteriorly by an epithelium and is enclosed in a capsule.15 Cell division in the lens continues throughout a person’s lifetime. With radiation injury to the dividing cells, the resulting abnormal fibers migrate toward the posterior pole, where they constitute the beginning of a cataract. Definitive studies from Merriam and colleagues9 shed light on the dose-response relationship for cataracts. The investigators analyzed the available dose estimates to the lens in 233 patients who received radiation therapy to the eye. Radiation cataracts developed in 128 ­patients. The minimum cataractogenic doses were 2 Gy for ­single-dose radiation, 4 Gy for multiple doses given over a ­period ranging from 3 weeks to 3 months, and 5.5 Gy for multiple doses given over a period longer than 3 months. Higher doses decreased the latent period and increased the severity of the cataracts. In patients who received 2.5 to 6.5 Gy, the latent period was 8 years and 7 months, and one third of the patients developed progressive cataracts. For patients who received 6.51 to 11.5 Gy, the latent ­period was 4 years and 4 months, and two thirds of the patients developed progressive cataracts.

Sclera The sclera is extremely radioresistant, as illustrated by its tolerance of the high doses of radiation used for the treatment of ocular melanomas. However, doses in the range of 200 to 300 Gy may cause atrophy after a latency period of several years.9

Retina Radiation damage to the retina is primarily seen in the endothelium of the retinal capillaries.16 Ophthalmoscopic findings include retinal hemorrhages, microaneurysms, hard exudates, cotton-wool spots, and telangiectases.17 In the M. D. Anderson Cancer Center series from Jiang and colleagues,14 the onset of visual impairment caused by retinopathy occurred 5 to 30 months (median, 11 months) after radiation treatment. With doses of 50 to 60 Gy given over 29 to 62 days, the overall actuarial incidence of retinopathy was 20%.14 However, this was most likely an ­ underestimation because not all patients underwent ­systematic ophthalmologic evaluation.

312

Part 3  Head and Neck

Published findings on the dose-response relationship for radiation retinopathy are limited. Parsons and colleagues18 reported no retinopathy in 33 eyes treated with doses less than 45 Gy. For higher doses, the incidences of retinopathy were as follows: 50% (6 of 12) for doses of 45 to 55 Gy; 83% (10 of 12) for doses of 55 to 65 Gy, and 100% (11 of 11) for doses above 65 Gy. Several anecdotal reports have suggested that patients with diabetes mellitus are at increased risk of developing radiation retinopathy.19-21 In our practice, we prefer to limit the retinal dose to 50 Gy when possible to minimize complications.

Optic Nerve and Chiasm In an M. D. Anderson series, Jiang and colleagues14 repor­ ted that visual impairment caused by damage to the optic nerve and chiasm occurred 7 to 50 months (median, 27 months) after radiation therapy. Optic neuropathy was not observed after doses of less than 56 Gy, and the incidence of this complication remained low (< 5% at 10 years) with doses up to 60 Gy in fractions not exceeding 2.5 Gy. However, with doses higher than 60 Gy, the incidence of optic neuropathy increased considerably, up to 34% at 10 years. In a University of Florida series from Parsons and colleagues,22 optic neuropathy was not observed in 106 optic nerves that received a total dose of less than 60 Gy. With doses of 60 Gy or more, the incidence of optic neuropathy increased, with the suggestion that the incidence depended on dose per fraction. With doses of 60 Gy or more, the 15-year actuarial risk of optic neuropathy was 11% with fractions of less than 1.9 Gy and 47% with fractions of 1.9 Gy or greater. Optic chiasm injury resulting in bilateral blindness is one of the most devastating complications of radiation therapy. In the series reported by Jiang and colleagues,14 none of 32 patients who received total doses of less than 50 Gy and only 4 of 110 patients (8% actuarial rate at 10 years) who received doses of 50 to 60 Gy in fractions of 2.6 Gy or less developed optic chiasm injury. Total doses of 61 to 76 Gy were associated with a higher complication rate (24% at 10 years). For treatment planning, we prefer to limit the dose to the optic nerve and chiasm to 54 Gy. When gross tumor is close and requires higher doses, we limit the dose to the optic nerve and chiasm to 60 Gy, and we ensure that the patient understands that the risk of blindness with this treatment approaches approximately 10%.

Benign Orbital Diseases Graves’ Ophthalmopathy Graves’ disease is an autoimmune disorder with three major manifestations: hyperthyroidism, ophthalmopathy, and dermopathy. Graves’ ophthalmopathy is an inflammatory process involving extraocular muscles and orbital tissues; occurring in 20% to 40% of patients with Graves’ disease, it manifests as exophthalmos with ophthalmoplegia and congestive oculopathy characterized by chemosis, conjunctivitis, periorbital swelling, and the potential complications of corneal ulceration, optic nerve compression, and optic atrophy.23 Pathologic examination reveals an inflammatory infiltrate of the orbital contents and enlargement of the extraocular muscles. Although the pathogenesis of Graves’ ophthalmopathy is still not fully understood, one proposed mechanism is

the development of antibodies against specific antigens in the extraocular muscles. In most patients, the disease has a benign course that does not follow the course of the hyperthyroidism. More severe cases of Graves’ ­ophthalmopathy require medical intervention. The first line of treatment is high-dose steroids (e.g., 100 to 120 mg of prednisone per day). Surgical decompression of the orbit can be performed to relieve severe proptosis or acute ­compression of the optic nerve and to reduce pain. Surgical correction of the eye muscles can ­ diminish diplopia, and correction of the eyelids can improve the function of the distorted edematous lids to lessen ­corneal damage. Orbital irradiation in the treatment of Graves’ ophthalmopathy is generally reserved for symptoms that are refractory to steroids or for patients in whom steroids are contraindicated (e.g., those with severe corneal abrasion or optic neuropathy). Because radiation effects can take up to several weeks to manifest and may transiently worsen local inflammation, steroid therapy can be continued for several weeks. Bilateral irradiation is preferred because of the high incidence of contralateral involvement detectable by diagnostic imaging, even in patients with minimal symptoms and signs.24 Moreover, when a one-sided lateral photon portal is used, subsequent treatment of the contralateral eye is problematic because of the relatively high exit dose received (approximately 60%).25 Opposed lateral fields with half-beam technique have been advocated.24,25 With this technique, the dose to the lens can be limited to only 4% of the prescribed dose by placing the anterior border at the lateral fleshy canthus. For more severe cases, however, the anterior border should be adjusted according to the degree of exophthalmos. Evidence of the efficacy of radiation therapy for Graves’ ophthalmopathy largely comes from reports of institutional experiences with patients receiving radiation as second-line treatment.24,29,30 With the recommended treatment course of 20 Gy in 10 fractions, steroid therapy can be discontinued after irradiation in 76% to 90% of patients. Two small, randomized studies comparing the combination of radiation therapy and systemic steroids with steroids alone or radiation therapy alone showed the combined treatment to be beneficial.28,31 Two randomized, double-blind trials directly compared sham irradiation with 20 Gy of irradiation. In one of these studies, Mourits and others32 enrolled a total of 60 patients with moderately severe ophthalmopathy; in the other, Prummel and colleagues33 studied 88 patients with milder problems. Although both series demonstrated an approximate doubling of response rates in irradiated patients (52% to 60% in the Mourits study versus 27% to 31% in the Prummel study), investigators in both series observed that these differences were mostly caused by improvements in eye motility rather improvements in proptosis, eyelid swelling, or overall quality of life. Nonetheless, radiation did seem to reduce the need for strabismus surgery and longterm follow-up and provided effective relief of symptomatic diplopia. Fortunately, long-term morbidity after irradiation for Graves’ ophthalmopathy is quite limited; nevertheless, the risk of secondary retinopathy after treatment may be higher in patients with diabetes.34 No secondary ­radiationrelated orbital malignancies in patients with Graves’ ­ophthalmopathy have been reported, although some

14 The Orbit

i­nvestigators have estimated a theoretical risk of tumor induction of 0.3% to 1.2%.35,36

Orbital Pseudotumor Orbital pseudotumor (idiopathic orbital inflammatory disease) is a nongranulomatous, acute or subacute inflammatory disease that usually occurs in the anterior orbit or midorbit and often involves the lacrimal gland.37 This condition is idiopathic by definition, and neoplastic, infectious, and systemic inflammatory or immunologic causes must be excluded.38 Contrast-enhanced magnetic resonance imaging (MRI) is usually the most helpful modality for gauging the extent of disease. The disease usually is unilateral, although it can be bilateral in some cases. Clinical symptoms consist of abrupt pain, conjunctival redness, chemosis, lid edema, exophthalmos, and restriction of eye movement. Pathologic examination reveals a gray, rubbery mass composed of a polymorphic infiltrate of lymphocytes, eosinophils, plasma cells, and polymorphonuclear leukocytes.39 The clinical course varies, and spontaneous regression may occur in a few patients. In severe cases, mass effects can cause secondary optic nerve atrophy, disc edema, and complete visual loss. Treatment is nonsurgical, with resection reserved only for cases that do not respond to medical management and radiation. First-line treatment consists of steroids, which produce dramatic responses in 30% to 60% of patients within 2 to 3 days.38,40 In refractory cases, radiation therapy is highly effective, yielding control rates of 65% to 75% after doses of 10 to 30 Gy, with or without systemic therapy.38,40-45 For anterior lesions, Donaldson and Findley46 described an electron beam technique with lens shielding that uses a centrally placed lead eye shield suspended 1 cm above the lens. The field size should be at least 4 × 4 cm to reduce dose inhomogeneity. The investigators ­recommend 9- to 16-MeV beams for superficial lesions. At M. D. ­Anderson, we prefer to mount the block on a contact lens to place the shield directly over the lens and cornea (see “Orbital Lymphomas”). Treatment planning should be individualized with the aid of three-dimensional CT ­guidance.

Pterygium A pterygium is a conjunctival growth over the sclera and onto the cornea; it occurs most often on the nasal side in the palpebral fissure area.47 Chronic exposure to ­ultraviolet light or dust may be causative.47 Treatment is indicated when pterygia cause visual or cosmetic impairment. First-line therapy is surgical excision of the pterygium. However, because the incidence of recurrence after excision alone is relatively high (20% to 39%,48,49 with most ­recurrences happening in the first year after excision50), surgery is usually accompanied by adjuvant treatment in an attempt to prevent recurrence. Nonirradiation treatment options ­ include mitomycin C, thiotepa, fluorouracil, and ­conjunctival autotransplantation. Extensive experience has been reported for treatment of pterygia with strontium 90/yttrium 90 (90Sr/90Y) ­betaray surface applicators. 49,51-55 The maximum surface dose and rapid dose fall-off make beta-rays ideally suited for ­ superficial disease coverage. The first application is ­typically given 12 to 24 hours after surgery. A ­variety of radiation schedules have been used, ranging from 20 Gy

313

in one ­fraction to 24 to 60 Gy given in weekly ­fractions of 8 to 10 Gy. With the addition of adjuvant radiation therapy to surgery, the control rate is excellent: the ­recurrence rate is less than 5%. The main concern with radiation therapy is the risk of late scleral necrosis and the attendant risks of corneoscleral ulceration and infection. However, with fractionated radiation therapy, the late complication rate is usually less than 1%.51,53,54,56 Higher complication rates have been associated with the use of large single fractions.49,57,58 To minimize late ­complications, plaques must be carefully calibrated to ensure an accurate dose rate, because measured rates can be 30% to 57% higher than the dose rates provided by the manufacturer.59,60 Paryani and colleagues53 from the North Florida Pterygium Study Group reported outcomes from the largest series of patients in North America treated with scleral plaques after surgery. Adjuvant beta-ray therapy was given within 24 hours after complete surgical excision of primary or recurrent pterygia in 690 patients. The source was calibrated by the National Bureau of Standards, and the regimen was 60 Gy in 6 weekly fractions of 10 Gy. At a median follow-up time of more than 8 years, the recurrence rate was 1.7%, with most of the recurrences occurring in patients who received fewer than five of the prescribed fractions. No major late complications were reported with this regimen. Primary radiation therapy can be considered for reducing the size of pterygia and preventing the future growth of symptomatic lesions. Pajic and colleagues55 reported a small series of 54 primary pterygia in 43 patients treated exclusively with 90Sr/90Y beta irradiation to a total dose of 50 Gy in 4 weekly fractions. Reduction in size occurred in all lesions. No patients developed recurrent growth, and no late side effects were reported.

Malignant Orbital Diseases Basal and squamous cell carcinoma and melanoma of the eyelids have a natural history similar to that of their counterparts in other skin locations. These neoplasms are discussed in Chapter 6.

Ocular Melanoma The most common primary malignant tumor of the eye is melanoma. Like its cutaneous counterpart, ocular melanoma originates from melanocytes derived from neural crest tissue. However, ocular melanomas are not related to sun exposure.61 Melanomas of the uveal tract (iris, ciliary body, and choroid) are 20 to 40 times more common than those of the conjunctiva.

Uveal Melanoma Although melanoma is the most common primary ­malignant tumor of the eye, the overall incidence of uveal melanoma is only five to seven cases per million people per year.61 The average age at diagnosis is 55 years.62 ­Tumors of the posterior uvea account for more than 90% of cases, with most ­ being choroidal melanomas. In the past, the misdiagnosis rate was high because of the difficulties in obtaining biopsy material. With the emergence of ancillary imaging techniques such as fluorescein angiography, sonography, color Doppler imaging, phosphorus 32 uptake

314

Part 3  Head and Neck

BOX 14-1.  American Joint Committee on Cancer Tumor Staging System for Melanomas of the Ciliary Body and Choroid Primary Tumor (T) TX T0 T1* T1a T1b T1c T2* T2a T2b T2c T3* T4 Stage Grouping Stage I

Stage II

Stage III Stage IV

Primary tumor cannot be assessed No evidence of primary tumor Tumor 10 mm or less in greatest diameter and 2.5 mm or less in greatest height (thickness) Tumor 10 mm or less in greatest diameter and 2.5 mm or less in greatest height (thickness) without ­microscopic extraocular extension Tumor 10 mm or less in greatest diameter and 2.5 mm or less in greatest height (thickness) with ­microscopic extraocular extension Tumor 10 mm or less in greatest diameter and 2.5 mm or less in greatest height (thickness) with ­macroscopic extraocular extension Tumor greater than 10 mm but not more than 16 mm in greatest basal diameter and between 2.5 and 10 mm in maximum height (thickness) Tumor 10 mm to 16 mm in greatest basal diameter and between 2.5 and 10 mm in maximum height (thickness) without microscopic extraocular extension Tumor 10 mm to 16 mm in greatest basal diameter and between 2.5 and 10 mm in maximum height (thickness) with microscopic extraocular extension Tumor 10 mm to 16 mm in greatest basal diameter and between 2.5 and 10 mm in maximum height (thickness) with macroscopic extraocular extension Tumor more than 16 mm in greatest diameter and/or greater than 10 mm in maximum height (thickness) ­without extraocular extension Tumor more than 16 mm in greatest diameter and/or greater than 10 mm in maximum height (thickness) with extraocular extension T1 T1a T1b T1c T2 T2a T2b T2c T3 T4 Any T Any T

N0 N0 N0 N0 N0 N0 N0 N0 N0 N0 N1 Any N

M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M1

*In clinical practice, the tumor base may be estimated in optic disc diameters (dd) (average: 1 dd = 1.5 mm). The elevation may be ­estimated in diopters (average: 3 diopters = 1 mm). Other techniques used, such as ultrasonography and computed stereometry, may provide a more accurate measurement. From Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 365-370.

testing, CT, and MRI, the error rate has been reduced to less than 1%, as reported by the multicenter Collaborative Ocular Melanoma Study (COMS).63 Combined positron emission tomography/CT may eventually be useful in the pretreatment assessment of uveal melanoma.64 Prognostic factors include age, cell type, tumor size, tumor location, ­integrity of Bruch’s membrane (the lamina between the choriocapillaris and the pigmented layer of the retina), scleral invasion, and pigmentation.65 As reflected by the American Joint Committee on Cancer staging system (Box 14-1),66 tumor size seems to be the most important ­ prognostic ­ factor, with tumors measuring more than 10 mm conferring a poor prognosis. For many years, all uveal melanomas were treated by enucleation, and little is known about the natural history of this disease. Most tumors arise from the pigmented cells in the choroid. Tumor spread is mainly local. With inward growth, the tumor may displace the retina or rupture Bruch’s membrane and spread over the retinal surface. Outward growth of the tumor may lead to extrascleral

infiltration into orbital soft tissues. Anterior choroidal and ciliary body tumors may involve the lens and seed the ­posterior chamber.6,67,68 Because the uvea has no lymphatic drainage, metastasis occurs hematogenously. The overall metastasis rate is approximately 50%, and metastases have a particular predilection for the liver. Doubling times range from a few months to several years. Sato and colleagues69 reported that the median time from treatment to systemic metastasis was 20 months for the group with the most unfavorable prognosis (i.e., age > 60 years, male sex, and tumor diameter > 10 mm), in contrast with 76 months for the group with the most favorable prognosis (i.e., age ≤ 60, female sex, and diameter ≤ 10 mm). The traditional practice of enucleation of the involved eye was questioned in the 1970s by Zimmerman and colleagues,70,71 who suggested that enucleation may worsen prognosis by promoting metastasis through the dissemination of tumor cells during the surgical procedure. Although this theory has been questioned by others,72-74

14 The Orbit

the trend at many centers has shifted toward preserving vision whenever possible. Shields and Shields75 ­ advocate ­reserving enucleation for large melanomas (those that are > 10 mm thick and > 15 mm in diameter), treatment of which with ­radiation therapy would be expected to cause considerable visual morbidity. However, if the melanoma is located in the patient’s only useful eye, it is reasonable to attempt control with radiation therapy despite the expected poor visual outcome. Brachytherapy. A variety of radiation therapy techniques are used for conservative management of uveal melanoma, including plaque therapy with radon 222 (222Rn), gold 198 (198Au), cobalt 60 (60Co), iridium 192 (192Ir), tantalum 182 (182Ta), ­ iodine 125 (125I), and ruthenium 106 (106Ru). In the ­United States, 125I has become widely used because its lower ­ energy ­ allows less shielding of normal tissues and medical personnel without compromising the dose distribution in the target volume.76 The continued controversy regarding the relative indications for enucleation and radiation therapy led to the launch of a multicenter COMS77 in the United States and Canada in 1986. In that trial, 1317 patients with medium-sized tumors (2.5 to 10 mm in apical height and ≤ 16 mm in basal diameter) were randomly assigned over a 12-year accrual period to treatment with 125I brachytherapy or enucleation. The COMS trial used 125I plaques, the size of which was chosen to cover the tumor base (defined to include all pigmented areas) plus a margin of 2 to 3 mm. Tumors contiguous with the optic disc were excluded because vision would be compromised because of radiation damage to the optic nerve. A dose of 100 Gy was delivered at a rate of 0.5 to 1.25 Gy/hr. Accrual ended in July 1998, and follow-up was expected to continue for up to an additional 10 years.78 Initial clinical outcomes were reported in 2001, when 81% of patients had been followed up for at least 5 years.79 No significant differences in unadjusted or adjusted mortality rates could be ­demonstrated between the two groups at that time. Five-year cumulative mortality rates were 19% for the ­enucleation group and 18% for the brachytherapy group. The risk of treatment failure at 5 years after brachytherapy was only 10.3%; however, ­ visual acuity was 20/200 or worse in 43% of treated eyes. In the absence of mature prospective data, Augsburger and colleagues80 published a matched-pair analysis of 448 ­patients (from an original group of 734 patients). The 15year survival rates were 57% for the ­enucleation group and 62% for the ­ plaque-radiation-­therapy group.80 The 15-year local relapse rate in the plaque-radiation group was 18.9%. ­Although this study did not have sufficient power to reveal small ­ differences in ­ survival, a large ­difference in ­ survival ­ between the ­ treatment groups would be unlikely. A second COMS randomized trial81 compared results after enucleation, preceded or not preceded by preoperative external beam radiation therapy (20 Gy in 5 daily fractions) for patients with large tumors (i.e., > 10 mm in apical height or > 16 mm in basal diameter). Mortality findings were published in 2004.81 With 10-year results known for 75% of enrolled patients, pre-enucleation radiation therapy did not yield a statistically significant improvement in ­survival. At a median survival time of approximately 7 years

315

for both treatment groups, actuarial 5-year overall survival rates were 56% for those who underwent enucleation only and 61% for those who had pre-enucleation radiation therapy. Serious complications were uncommon and were not increased by presurgical radiation.82 Given these findings, routine preoperative irradiation probably has no role in the treatment of large uveal melanomas. Particle Beam Therapy. During the past 3 decades, particle beam therapy has emerged as an effective alternative to brachytherapy. By taking advantage of the Bragg peak effect (described in Chapters 1 and 40), particle beam therapy can deposit radiation in the target while limiting the fall-off dose to adjacent noninvolved structures. Brachytherapy is effective and has potentially less associated morbidity for directly accessible lesions anterior to the equator of the globe, but particle beam therapy can be used to treat posterior lesions more homogeneously and completely, especially those approximating the optic disc. Several series have demonstrated equivalent outcomes for brachytherapy and particle beam therapy, although these findings remain somewhat limited at this time.83-85 Helium ions86 and protons have been investigated for uveal melanoma, but most of the work has been done with protons. Clinical results have been promising. Investigators from the Curie Institut-Orsay Proton Therapy Center published a review of their 20-year experience treating 1406 patients with protons.87 Patients were managed uniformly with 60Co Gy equivalent (CGE) in four consecutive once-daily fractions. At a median follow-up time of just over 6 years, the actuarial 5-year local disease control rate was 96%, the metastasis-free survival rate was 81%, and the overall survival rate was 79%. These results echo data from previous smaller series88-92; retrospective comparisons of protons with enucleation have not shown proton beam therapy to compromise survival.93,94 Complication rates were relatively low and mirrored published experiences. The enucleation rate for treatment-related toxicity (most often neovascular glaucoma and its attendant pain) was 7.7%. The risk of toxicity-related enucleation was associated with tumor thickness and the volume of lens ­receiving more than 30 CGE. Future efforts to reduce toxicity will focus on prospective trials of dose and treatment margin–reduction strategies. Other reports suggest that proton therapy can be useful for carefully selected, large melanomas of the uvea (i.e., tumors at least 10 mm thick or 20 mm in diameter and tumors located within 3 mm of the optic nerve that are at least 8 mm thick or 16 mm in diameter)95 or melanomas of the iris.96

Conjunctival Melanoma Conjunctival melanomas are rare, accounting for 2% of ocular melanomas. These tumors are treated with local excision or cryotherapy. The role of radiation therapy in this disease is unclear.97

Orbital Lymphomas Primary non-Hodgkin lymphoma of the orbit is a rare disease, accounting for 1% of cases of non-Hodgkin ­lymphoma98 and 10% of all orbital tumors.99 Primary non-­Hodgkin lymphomas of the orbit are distinct from ­intraocular lymphomas involving the vitreous humor, ­retina,

316

Part 3  Head and Neck

or choroid, which have a high rate of central ­nervous system ­involvement.100 Usually arising from the conjunctiva, ­non­Hodgkin ­lymphomas of the orbit ­manifest as smooth, shiny, salmon-­colored swellings that often are ­ asymptomatic.98 Retrobulbar ­ lesions restrict extraocular movement and cause exophthalmos, diplopia, edema, chemosis, and pain. Vision is usually preserved. In addition to CT or MRI to evaluate the extent of local disease, diagnostic evaluation should include the ancillary evaluations typically done for lymphoma (e.g., bone marrow biopsies and CT scanning of the chest, abdomen, and pelvis) to evaluate the extent of systemic disease. Although any subtype of non-Hodgkin lymphoma can affect the orbital adnexa, marginal zone Bcell mucosa-­associated lymphoid tissue (MALT) type and follicular histologies are the most ­common. The selection of treatment depends on the pathologic subtype and disease stage. Radiation therapy alone is an accepted treatment for low-grade lymphomas confined to the orbit. Because high-grade lymphomas are associated with high distant relapse rates (60%),101,102 treatment usually consists of initial chemotherapy followed by consolidative irradiation. Series published in the late 1980s reported excellent local control rates with radiation doses of 25 to 40 Gy given in the 1.5- to 2.0-Gy fractions typical for lymphoma (90% to 100%).101-103 Fung and others from Massachusetts General Hospital104 reported a lack of ­radiation 9-MeV Electron beam

0.0

1.0

3

Depth (cm)

2.0

dose-response relationship for follicular lymphoma of the ocular adnexa; all cases were controlled with doses in the mid-20-Gy range. However, MALT-type lymphomas seemed to require at least 30 Gy (30.6 to 32.4 Gy in 1.8-Gy fractions was recommended) to achieve 100% local control. Several other series support this finding,105-107 although doses between 25 and 30 Gy may also be effective.108,109 In a series from M. D. Anderson, none of the patients with low-grade lymphomas treated with 30 Gy given in 20 fractions had a local relapse.110 For high-grade lesions, we recommend initial chemotherapy followed by 30 Gy in 20 fractions for those tumors that show a complete response to chemotherapy or 36.0 to 39.6 Gy in 20 to 22 fractions if the tumor shows a partial response. The limited published experience with observation only after biopsy of MALT-type orbital lymphoma111,112 suggests that this approach requires careful patient selection and counseling and cannot be considered the standard of care given the expected efficacy and low morbidity of ­radiation. Electron beam therapy, typically 6 to 9 MeV with a 0.5to 1.0-cm bolus, is adequate for anterior lesions involving the eyelid or conjunctiva. A lens or eye shield can be used if its use does not compromise tumor coverage. For eyelid tumors, commercial lead eye shields, 1.7 mm thick, provide inadequate protection for electron beam energies of 6 MeV or greater, with 50% dose transmission on the surface (cornea) and 27% dose transmission at a depth of 6 mm (lens).113 In place of lead, we use a tungsten eye shield, 3 mm thick, which reduces the transmission dose to the cornea and lens to less than 5% for 6- to 9-MeV electrons (Fig. 14-4).114 Conjunctival tumors require irradiation of the palpebral and bulbar conjunctivae. In these cases, the lens may be protected by using a suspended lead eye bar, as outlined under “Orbital Pseudotumor.” With this technique, it is essential that the patient fix his or her gaze on the block while the treatment beam is on. However, we prefer to mount the block on a contact lens so that the shielding is placed directly over the lens and cornea (Fig. 14-5). The dose to the cornea with this technique has been calculated to be approximately 5% of the total dose. Posterior lesions usually require treatment with photon beams to achieve sufficiently deep dose distribution.

5 10

3.0

20 4.0

40

27

40

27 20 5.0

10 5

6.0

3 Long axis

FIGURE 14-4.  Isodose distribution beneath a tungsten eye shield for a 9-MeV electron beam from a linear accelerator. (Modified from Block R, Gartner S. The incidence of ocular metastatic carcinoma. Arch Ophthalmol 1971;85:673-675.)

FIGURE 14-5.  Field setup for an anteriorly located orbital lymphoma treated with an electron beam with shielding of the lens and cornea through the use of a brass shield mounted on a contact lens.

14 The Orbit

Metastasis to the Orbit Although metastasis to the orbit is uncommon, metastatic tumors are the most common intraocular malignancy. Metastases to the orbit most often originate from primary breast carcinoma or lung cancer.113,115-117 Given the increasing survival times of cancer patients, the incidence of orbital metastasis can be expected to increase. Metastases to the eye can involve any orbital structure. Intraocular metastases are typically located in the choroid. Presenting symptoms include decreased vision, exophthalmos, pain, heaviness, and double vision. Ophthalmoscopic examination often reveals retinal detachment.117 CT or MRI delineates the extent of the metastatic deposit. In most cases, a history of prior malignancy, multiple lesions in one eye, or bilateral tumors establishes the diagnosis. In other cases, metastatic choroid tumors can be distinguished from other tumors—mainly choroidal melanomas, benign lesions such as hemangiomas, and inflammatory granulomas118—by using ancillary studies as described earlier under “Uveal Melanoma.” Bilateral involvement is found in 20% to 25% of cases.116,117 The initial evaluation should include MRI of the brain because of the possibility of synchronous brain metastases. Unilateral radiation therapy with a lateral electron portal of sufficient energy is adequate to treat most ocular metastases. The anterior border should be placed just behind the anterior chamber of the eye, and a posterior tilt should be placed to avoid the lens. For bilateral metastases, posteriorly tilted opposing photon fields can be used. Radiation therapy is quite effective in relieving symptoms and controlling tumor growth. With doses of at least 30 Gy in daily fractions of 2 to 3 Gy, symptomatic improvement and tumor control are seen in approximately 90% of patients.117,119 Doses less than 30 Gy may be less effective. In a report by Maor and colleagues,119 none of the nine patients who received 30 Gy in 10 fractions had tumor regrowth after therapy, but 2 of 10 patients treated with 25 Gy in 10 fractions had tumor regrowth. In a series reported by Reddy and colleagues,116 30% of tumors did not respond to treatment with doses of 21 to 30 Gy. The median survival time after detection of choroidal metastases is 8.5 to 12 months, and for most patients, the benefit in vision produced by radiation therapy lasts for the remainder of their lives.

REFERENCES   1. Saude T. The outer coats of the eye. In Saude T (ed): Ocular Anatomy and Physiology. Oxford, UK: Blackwell Science, 1993, pp 11-25.   2. Bron AJ, Tripathi RC, Tripathi BJ. Wolff’s Anatomy of the Eye and Orbit, 8th ed. London: Arnold Publishers, 1997.   3. Karlsson U, Kirby T, Orrison W, et al. Ocular globe topography in radiotherapy. Int J Radiat Oncol Biol Phys 1995;33:705-712.   4. Rohrschneider W. Experimentelle Untersuchungen uber die Veranderungen normaler Augengewebe nach Rontgenbestrahlung. III. Mitteilung. Beranderungen der Linse der Netzhaut und des Sehnerven nach Rontgenbestrahlung. Graefes Arch Ophthal 1929;122:282.   5. Fajardo LF, Berthrong M, Anderson RE. Eye and its adnexa. In: Radiation Pathology. New York: Oxford University Press, 2001, pp 399-408.   6. Hungerford J. Current management of choroidal malignant melanoma. Br J Hosp Med 1985:34:287-293.

317

  7. Brady L, Shields J, Augusburger J, et al. Complications from ­radiation therapy to the eye. Front Radiat Ther Oncol 1989; 23:238-250.   8. Parsons J, Bova FJ, Fitzgerald CR, et al. Severe dry-eye syndrome following external beam irradiation. Int J Radiat Oncol Biol Phys 1994;30:775-780.   9. Merriam G, Szechter A, Focht E. The effects of ionizing ­radiations on the eye. In Vaeth JM (ed): Radiation Effect and ­Tolerance, Normal Tissue. Baltimore: University Park Press, 1972, pp 346-385. 10. Chan R, Shukovsky L. Effects of irradiation on the eye. ­Radiology 1976;120:673-675. 11. Bessell E, Henk JM, Whitelocke RA, et al. Ocular morbidity after radiotherapy of orbital and conjunctival lymphoma. Eye 1987;1:90-96. 12. Merriam G. Late effects of beta radiation on the eye. Arch Ophthalmol 1955;53:708. 13. Merriam G. The effects of β-radiation on the eye. Radiology 1956;66:240. 14. Jiang G, Tucker SL, Guttenberger R, et al. Radiation-induced injury to the visual pathway. Radiother Oncol 1994;30:17-25. 15. Hall EJ. Radiation cataractogenesis. In Hall EJ (ed): Radiobiology for the Radiologist. Philadelphia: JB Lippincott, 1994, pp ­379-384. 16. Archer D. Doyne lecture: responses of retinal and choroidal vessels to ionizing radiation. Eye 1993;7:1-13. 17. Brown G, Shields JA, Sanborn G, et al. Radiation retinopathy. Ophthalmology 1982;89:1494-1501. 18. Parsons J, Bova FJ, Fitzgerald CR, et al. Radiation retinopathy after external-beam irradiation: analysis of time-dose factors. Int J Radiat Oncol Biol Phys 1994;30:765-773. 19. Dhir S, Joshi A, Banerjee A. Radiation retinopathy in diabetes mellitus. Acta Radiol Oncol 1982;21:111-113. 20. Mewis L, Tang R, Salmonsen P. Radiation retinopathy ­ after “safe” levels of irradiation. Invest Ophthalmol Vis Sci 1982; 22:222. 21. Viebahn M, Barricks M, Osterloh M. Synergism between diabetic and radiation retinopathy: case report and review. Br J Ophthalmol 1991;75:629-632. 22. Parsons J, Bova FJ, Fitzgerald CR, et al. Radiation optic ­neuropathy after megavoltage external-beam irradiation: analysis of timedose factors. Int J Radiat Oncol Biol Phys 1994;30:755-763. 23. Wartofsky L. Diseases of the thyroid. In Isselbacher KJ, Braunwald E, Wilson JD, et al (eds): Harrison’s Principles of Internal Medicine, 13th ed. New York: McGraw-Hill, 1994, pp 19301953. 24. Olivotto I, Ludgate CM, Allen LH, et al. Supervoltage radiotherapy for Graves’ ophthalmopathy: CCABC technique and results. Int J Radiat Oncol Biol Phys 1985;11:2085-2090. 25. Donaldson S, McDougall I, Kriss J. Grave’s disease. In Alberti WE, Sagerman RH (eds): Medical Radiology: Radiotherapy of Intraocular and Orbital Tumors. Berlin: Springer-Verlag, 1993, pp 192-197. 26. Palmer D, Greenberg P, Cornell P, et al. Radiation therapy for Graves’ ophthalmopathy: a retrospective analysis. Int J Radiat Oncol Biol Phys 1987;13:1815-1820. 27. Petersen I, Kriss JP, McDougall IR, et al. Prognostic factors in the radiotherapy of Graves’ ophthalmopathy. Int J Radiat Oncol Biol Phys 1990;19:259-264. 28. Marcocci C, Bartalena L, Bogazzi F, et al. Orbital radiotherapy combined with high dose systemic glucocorticoids for Graves’ ophthalmopathy is more effective than radiotherapy alone: results of a prospective randomized study. J Endocrinol Invest 1991;14:853-860. 29. Pitz S, Kahaly G, Rosler HP, et al. Retrobulbar irradiation for Graves’ ophthalmopathy—long-term results [in German]. Klin Monatsbl Augenheilkd 2002;219:876-882. 30. Kulig G, Kazmierczyk-Puchalska A, Krzyzanowska-­Swiniarska B, et al. Effectiveness of treatment for thyroid orbitopathy in ­patients hospitalized at the Endocrinology Department of ­Pomeranian Medical University [in Polish]. Przegl Lek 2004; 61:852-854.

318

Part 3  Head and Neck

31. Bartalena L, Marcocci C, Chiovato L, et al. Orbital cobalt ­irradiation combined with systemic corticosteroids for Graves’ ophthalmopathy: comparison with systemic corticosteroids alone. J Clin Endocrinol Metab 1983;56:1139-1144. 32. Mourits MP, van Kempen-Harteveld ML, Garcia MB, et al. ­Radiotherapy for Graves’ orbitopathy: randomised placebo­controlled study. Lancet 2000;355:1505-1509. 33. Prummel MF, Terwee CB, Gerding MN, et al. A randomized controlled trial of orbital radiotherapy versus sham irradiation in patients with mild Graves’ ophthalmopathy. J Clin Endocrinol Metab 2004;89:15-20. 34. Wakelkamp IM, Tan H, Saeed P, et al. Orbital irradiation for Graves’ ophthalmopathy: is it safe? A long-term follow-up study. Ophthalmology 2004;111:1557-1562. 35. Snijders-Keilholz A, De Keizer RJ, Goslings BM, et al. Probable risk of tumour induction after retro-orbital irradiation for Graves’ ophthalmopathy. Radiother Oncol 1996;38:67-71. 36. Blank L, Barendsen GW, Prummel MF, et al. Probable risk of tumour induction after retro-orbital irradiation for Graves’ ­ophthalmopathy [letter]. Radiother Oncol 1996;40:187-188. 37. Dutton J. Orbital diseases. In Yanoff M, Duker JS (eds): Ophthalmology London: Mosby, 1999, pp 7.14.1-7.14.18. 38. Char D, Miller T. Orbital pseudotumor. Fine needle aspiration biopsy and response to therapy. Ophthalmology 1993;100:17021710. 39. Kennerdell J, Dresner S. The nonspecific orbital inflammatory syndromes. Surv Ophthalmol 1984;29:93-103. 40. Leone C, Lloyd W. Treatment protocol for orbital inflammatory disease. Ophthalmology 1985;92:1325-1331. 41. Sergott R, Glaser J, Charyulu K. Radiotherapy for ­ idiopathic inflammatory orbital pseudotumor. Arch Ophthalmol 1981; 99:853-856. 42. Orcutt J, Garner A, Henk JM, et al. Treatment of idiopathic inflammatory orbital pseudotumours by radiotherapy. Br J Ophthalmol 1983;67:570-574. 43. Gunalp I, Gunduz K, Yazar Z. Idiopathic orbital inflammatory disease. Acta Ophthalmol Scand 1996;74:191-193. 44. Lanciano R, Fowble B, Sergott RC, et al. The results of radiotherapy for orbital pseudotumor. Int J Radiat Oncol Biol Phys 1990;18:407-411. 45. Jacobs D, Galetta S. Diagnosis and management of orbital pseudotumor. Curr Opin Ophthalmol 2002;13:347-351. 46. Donaldson S, Findley D. Treatment of orbital lymphoid tumors with electron beams. Front Radiat Ther Oncol 1991;25:187200. 47. Sugar A. Conjunctival and corneal degenerations. In Yanoff M, Duker JS, (eds): Ophthalmology London: Mosby, 1999, pp 5.6.1-5.6.8. 48. Cameron M. Pterygium Throughout the World. Springfield, IL: Charles C Thomas, 1965. 49. Bahrassa F, Datta R. Postoperative beta radiation treatment of pterygium. Int J Radiat Oncol Biol Phys 1983;9:679-684. 50. Hirst L, Sebban A, Chant D. Pterygium recurrence time. Ophthalmology 1994;101:755-758. 51. Van Den Brenk H. Results of prophylactic postoperative irradiation in 1,300 cases of pterygium. AJR Am J Roentgenol 1968;103:723-733. 52. Wilder R, Buatti JM, Kittelson JM, et al. Pterygium treated with excision and postoperative beta irradiation. Int J Radiat Oncol Biol Phys 1992;23:533-537. 53. Paryani S, Scott WP, Wells JW, et al. Management of pterygium with surgery and radiation therapy. Int J Radiat Oncol Biol Phys 1994;28:101-103. 54. Monteiro-Grillo I, Gaspar L, Monteiro-Grillo M, et al. Postoperative irradiation of primary or recurrent pterygium: results and sequelae. Int J Radiat Oncol Biol Phys 2000;48:865-869. 55. Pajic B, Greiner RH. Long term results of non-surgical, exclusive strontium-/yttrium-90 beta-irradiation of pterygia. Radiother Oncol 2005;74:25-29. 56. Alaniz-Camino F. The use of postoperative beta radiation in the treatment of pterygia. Ophthalmic Surg 1982;13:10221025.

57. MacKenzie F, Hirst LW, Kynaston B, et al. Recurrence rate and complications after beta irradiation for pterygia. Ophthalmology 1991;8:1776-1781. 58. Moriarty A, Crawford GJ, McAllister IL, et al. Severe corneoscleral infection. A complication of beta irradiation scleral necrosis following pterygium necrosis. Arch Ophthalmol 1993;111: 947-951. 59. Dusenbery K, Alul IH, Holland EJ, et al. β-Irradiation of recurrent pterygia: results and complications. Int J Radiat Oncol Biol Phys 1992;24:315-320. 60. Goetsch S. Strontium-90 ophthalmic applicators: use with ­caution [letter]. Ophthalmic Surg 1988;19:827-828. 61. Scotto J, Fraumeni JJ, Lee J. Melanomas of the eye and noncutaneous sites: epidemiologic aspects. J Natl Cancer Inst 1976;56:489-491. 62. Egan K, Seddon JM, Glynn RJ, et al. Epidemiological aspects of uveal melanoma. Surv Ophthalmol 1988;32:239-251. 63. Collaborative Ocular Melanoma Study Group. Accuracy of diagnosis of choroidal melanoma in the Collaborative ­ Ocular Melanoma Study: COMS report no. 1. Arch Ophthalmol 1990;108:1268-1273. 64. Reddy S, Kurli M, Tena LB, et al. PET/CT imaging: detection of choroidal melanoma. Br J Ophthalmol 2005;89:1265-1269. 65. Shammas H, Blodi F. Prognostic factors in choroidal and ciliary body melanomas. Arch Ophthalmol 1977;95:63-69. 66. Greene FL, Page DL, Fleming ID, et al (eds): AJCC Cancer Staging Manual. 6th ed. New York: Springer, 2002, pp 365-370. 67. Moss W. The orbit. In Cox JD (ed): Moss’ Radiation Oncology: Rationale, Technique, Results, 7th ed. St Louis: CV Mosby, 1994, pp 246-259. 68. Sahel J, Earle J, Albert D. Intraocular melanomas. In DeVita VT Jr, Hellman S, Rosenberg SA (eds): Cancer: Principles and Practice of Oncology, 4th ed. Philadelphia: JB Lippincott, 1993, pp 1662. 69. Sato T, Babazono A, Shields JA, et al. Time to systemic metastases in patients with posterior uveal melanoma. Cancer Invest 1997;15:98-105. 70. Zimmerman L, McLean I. Changing concepts concerning the malignancy of ocular tumors. Arch Ophthalmol 1975;78: 487-494. 71. Zimmerman L, McLean I, Foster W. Does enucleation of the eye containing a malignant melanoma prevent or accelerate the dissemination of tumour cells? Br J Ophthalmol 1978;62: 420-425. 72. Seigel D, Myers M, Ferris F, et al. Survival rates after ­enucleation of eyes with malignant melanoma. Am J Ophthalmol 1979;87:761-765. 73. Manschot W, van Peperzeel H. Choroidal melanoma: ­enucleation or observation? A new approach. Arch Ophthalmol 1980; 98:71-77. 74. Manschot W, Van Strik R. Is irradiation a justifiable treatment of choroidal melanoma? Analysis of published results. Br J Ophthalmol 1987;71:348-352. 75. Shields J, Shields C. Management of posterior uveal melanoma. In Shields JA, Shields CL (eds): Intraocular Tumors: A Textbook and Atlas. Philadelphia: WB Saunders, 1992, pp 171-205. 76. Earle J, Kline R, Robertson D. Selection of iodine 125 for the Collaborative Ocular Melanoma Study. Arch Ophthalmol 1987;105:763-764. 77. Collaborative Ocular Melanoma Study Group. COMS Manual of Procedures. Springfield, VA: National Technical Information Service, 1995. 78. Earle JD. Results from the Collaborative Ocular Melanoma Study (COMS) of enucleation versus preoperative radiation therapy in the management of large ocular melanomas. Int J Radiat Oncol Biol Phys 1999;43:1168-1169. 79. Diener-West M, Earle JD, Fine SL, et al. The COMS randomized trial of iodine 125 brachytherapy for choroidal melanoma. III. initial mortality findings. COMS report no. 18. Arch Ophthalmol 2001;119:969-982.

14 The Orbit 80. Augsburger J, Schneider S, Freire J, et al. Survival following enucleation versus plaque radiotherapy in statistically matched subgroups of patients with choroidal melanomas: results in patients treated between 1980 and 1987. Graefes Arch Clin Exp Ophthalmol 1999;237:558-567. 81. Hawkins BS. The Collaborative Ocular Melanoma Study (COMS) randomized trial of pre-enucleation radiation of large choroidal melanoma. IV. Ten-year mortality findings and prognostic factors. COMS report no. 24. Am J Ophthalmol 2004;138:936-951. 82. Collaborative Ocular Melanoma Study Group. The Collaborative Ocular Melanoma Study (COMS) randomized trial of preenucleation radiation of large choroidal melanoma. III. Local complications and observations following enucleation. COMS report no. 11. Am J Ophthalmol 1998;126:362-372. 83. Wilson MW, Hungerford JL. Comparison of episcleral plaque and proton beam radiation therapy for the treatment of choroidal melanoma. Ophthalmology 1999;106:1579-1587. 84. Char DH, Quivey JM, Castro JR, et al. Helium ions versus ­iodine 125 brachytherapy in the management of uveal melanoma. A prospective, randomized, dynamically balanced trial. ­Ophthalmology 1993;100:1547-1554. 85. Desjardins L, Lumbroso L, Levy C, et al. [Treatment of uveal melanoma with iodine 125 plaques or proton beam therapy: ­indications and comparison of local recurrence rates]. J Fr Ophtalmol 2003;26:269-276. 86. Char DH, Kroll SM, Castro J. Ten-year follow-up of helium ion therapy for uveal melanoma. Am J Ophthalmol 1998;125: 81-89. 87. Dendale R, Lumbroso-Le Rouic L, Noel G, et al. Proton beam radiotherapy for uveal melanoma: results of Curie Institut­Orsay proton therapy center (ICPO). Int J Radiat Oncol Biol Phys 2006;65:780-787. 88. Munzenrider J. Proton therapy for uveal melanomas and other eye lesions. Strahlenther Onkol 1999;175(Suppl 2):68-73. 89. Egger E, Schalenbourg A, Zografos L, et al. Maximizing local tumor control and survival after proton beam radiotherapy of uveal melanoma. Int J Radiat Oncol Biol Phys 2001;51: 138-147. 90. Fuss M, Loredo LN, Blacharski PA, et al. Proton radiation therapy for medium and large choroidal melanoma: preservation of the eye and its functionality. Int J Radiat Oncol Biol Phys 2001;49:1053-1059. 91. Courdi A, Caujolle JP, Grange JD, et al. Results of proton ­therapy of uveal melanomas treated in Nice. Int J Radiat Oncol Biol Phys 1999;45:5-11. 92. Hocht S, Bechrakis NE, Nausner M, et al. Proton therapy of uveal melanomas in Berlin: 5 years of experience at the ­HahnMeitner Institute. Strahlenther Onkol 2004;180:419-424. 93. Seddon J, Gragoudas ES, Albert DM, et al. Comparison of ­survival rates for patients with uveal melanoma after treatment with proton beam irradiation or enucleation. Am J Ophthalmol 1985;99:282-290. 94. Seddon J, Gragoudas ES, Egan KM, et al. Relative survival rates after alternative therapies for uveal melanoma. Ophthalmology 1990;97:769-777. 95. Conway RM, Poothullil AM, Daftari IK, et al. Estimates of ­ocular and visual retention following treatment of extra-large uveal melanomas by proton beam radiotherapy. Arch ­ Ophthalmol 2006;124:838-843. 96. Damato B, Kacperek A, Chopra M, et al. Proton beam radiotherapy of iris melanoma. Int J Radiat Oncol Biol Phys 2005;63: 109-115. 97. Rennie I. Diagnosis and treatment of ocular melanomas. Br J Hosp Med 1991;46:144-156.

319

98. Fitzpatrick P, Macko S. Lymphoreticular tumors of the orbit. Int J Radiat Oncol Biol Phys 1984;10:33-340. 99. Reese A. Orbital neoplasms and lesions simulating them. In Hoeber PB (ed): Tumours of the Eye. New York: Harper & Row, 1976, pp 434-440. 100. Char D, Ljung BM, Miller T, et al. Primary intraocular lymphoma (ocular reticulum cell sarcoma) diagnosis and management. Ophthalmology 1988;95:625-630. 101. Bessell E, Henk JM, Wright JE, et al. Orbital and conjunctival lymphoma treatment and prognosis. Radiother Oncol 1988;13:237-244. 102. Bolek T, Moyses HM, Marcus RB, et al. Radiotherapy in the management of orbital lymphoma. Int J Radiat Oncol Biol Phys 1999;44:31-36. 103. Smitt M, Donaldson S. Radiotherapy is successful treatment for orbital lymphoma. Int J Radiat Oncol Biol Phys 1993;26:59-66. 104. Fung CY, Tarbell NJ, Lucarelli MJ, et al. Ocular adnexal lymphoma: clinical behavior of distinct World Health Organization classification subtypes. Int J Radiat Oncol Biol Phys 2003;57:13821391. 105. Le QT, Eulau SM, George TI, et al. Primary radiotherapy for localized orbital MALT lymphoma. Int J Radiat Oncol Biol Phys 2002;52:657-663. 106. Tsang RW, Gospodarowicz MK, Pintilie M, et al. Localized mucosa-associated lymphoid tissue lymphoma treated with radiation therapy has excellent clinical outcome. J Clin Oncol 2003;21:4157-4164. 107. Uno T, Isobe K, Shikama N, et al. Radiotherapy for extranodal, marginal zone, B-cell lymphoma of mucosa-associated lymphoid tissue originating in the ocular adnexa: a multiinstitutional, retrospective review of 50 patients. Cancer 2003;98:865-871. 108. Zhou P, Ng AK, Silver B, et al. Radiation therapy for orbital lymphoma. Int J Radiat Oncol Biol Phys 2005;63:866-871. 109. Suh CO, Shim SJ, Lee SW, et al. Orbital marginal zone B-cell lymphoma of MALT: radiotherapy results and clinical behavior. Int J Radiat Oncol Biol Phys 2006;65:228-233. 110. Pelloski CE, Wilder RB, Ha CS, et al. Clinical stage IEA-IIEA orbital lymphomas: outcomes in the era of modern staging and treatment. Radiother Oncol 2001;59:145-151. 111. Matsuo T, Yoshino T. Long-term follow-up results of observation or radiation for conjunctival malignant lymphoma. Ophthalmology 2004;111:1233-1237. 112. Tanimoto K, Kaneko A, Suzuki S, et al. Long-term follow-up results of no initial therapy for ocular adnexal MALT lymphoma. Ann Oncol 2006;17:135-140. 113. Shiu A, Tung SS, Gastorf RJ, et al. Dosimetric evaluation of lead and tungsten eye shields in electron beam treatment. Int J Radiat Oncol Biol Phys 1996;35:599-604. 114. Block R, Gartner S. The incidence of ocular metastatic carcinoma. Arch Ophthalmol 1971;85:673-675. 115. Ferry A, Font R. Carcinoma metastatic to the eye and orbit. I. A clinicopathologic study of 227 cases. Arch Ophthalmol 1974;92:276-286. 116. Reddy S, Saxena VS, Hendrickson F, et al. Malignant metastatic disease of the eye: management of an uncommon complication. Cancer 1981;47:810-812. 117. Glassburn J, Klionsky M, Brady L. Radiation therapy for metastatic disease involving the orbit. Am J Clin Oncol 1984;7:145148. 118. Gragoudas E. Current treatment of metastatic choroidal tumors. Oncology 1989;3:103-110. 119. Maor M, Chan R, Yound S. Radiotherapy of choroidal ­metastases. Cancer 1977;40:2081-2086.

15 The Temporal Bone, Ear, and Paraganglia David H. Hussey AND B-Chen Wen ANATOMY RESPONSE OF THE NORMAL EAR TO IRRADIATION Outer Ear Middle Ear Inner Ear Temporal Bone Necrosis MALIGNANT TUMORS OF THE EAR Pathology Routes of Spread Presenting Symptoms Clinical Evaluation

Primary tumors of the ear are rare, and they are seen only infrequently in most radiation oncology practices. Nevertheless, these tumors deserve special consideration because they involve an important sensory organ and lie close to critical structures, such as the dura mater, brain, cranial nerves, carotid artery, temporomandibular joint, and parotid gland. Any of these structures may be invaded by tumors of the ear. The functions of the ear—hearing and balance—can be disturbed by cancer or the treatment of cancer.

Anatomy Anatomists divide the ear into three parts: the outer ear, the middle ear, and the inner ear. The outer ear comprises the auricle and the external auditory canal (EAC) (Fig. 15-1). The EAC extends from the auricle laterally to the tympanic membrane medially and is approximately 2.5 cm long. The EAC has a cartilaginous lateral portion and a bony medial portion. The cartilaginous portion is pierced by fissures through which tumors can spread anteriorly into the temporomandibular joint or the parotid gland. The EAC is lined with skin that is closely adherent to the perichondrium and periosteum. The lymphatic drainage of the auricle and the EAC is to the preauricular, postauricular, and subdigastric lymph nodes. The middle ear is composed of the tympanic cavity, the eustachian tube, and the mastoid process of the temporal bone. The tympanic cavity is an air-containing chamber lined with predominantly squamous epithelium and bounded by the tympanic membrane laterally and the osseous labyrinth medially. The tympanic cavity contains the auditory ossicles (the malleus, incus, and stapes), the tensor muscle of the tympanum, the stapedius muscle, and the chorda tympani nerve. The tympanic cavity communicates posteriorly with the tympanic antrum and mastoid air cells. It is continuous anteromedially with the eustachian tube, which opens into the nasopharynx. The eustachian tube is approximately 3.5 cm long. It is lined with pseudostratified ciliated columnar epithelium. Nodal metastasis is

320

Staging Treatment Radiation Therapy Technique Results CHEMODECTOMAS Anatomy Pathology Presenting Symptoms Clinical Evaluation Clinical Staging Treatment Radiation Therapy Technique

­ ncommon for tumors of the middle ear, suggesting that u this region has few lymphatics. The inner ear comprises the osseous labyrinth, which is a series of cavities in the dense petrous part of the temporal bone, and the membranous labyrinth, which is an intricate series of membranous ducts suspended within the osseous labyrinth. The labyrinthine system has three parts: the cochlea, the vestibule (which contains the utricle and saccule), and the semicircular canals. Each of these structures has a specific function. The cochlea is responsible for sound detection, the utricle is responsible for static equilibrium, and the three semicircular canals are responsible for dynamic equilibrium. The cochlea is a spiral chamber surrounding a central hollow post, called the modiolus, through which passes the cochlear nerve (see Fig. 15-1B). This chamber is divided into three parts by the thin bony spiral lamina, the basilar membrane, and the vestibular membrane. On the basilar membrane lies the organ of Corti (i.e., spiral organ). The organ of Corti contains hair cells that make synaptic contact with the auditory nerve fibers. It is the structure that is responsible for converting fluid waves to nerve impulses. The tympanic membrane vibrates when it receives sound waves, and this vibration is transmitted through the chain of ossicles to the oval window. Vibration of the footplate of the stapes covering the oval window produces traveling waves within the perilymph. These traveling waves cause the basilar membrane to oscillate, leading to deformity of the hair cells of the organ of Corti. When the hair cells are deformed, the nerve endings are stimulated, producing nerve impulses that are transmitted to the brain by the eighth cranial nerve.

Response of the Normal Ear to Irradiation Outer Ear The skin of the outer ear responds to irradiation in a fashion similar to that of the skin in other parts of the body. However, skin reactions in the auricle tend to be more

15  The Temporal Bone, Ear, and Paraganglia

321

Subtotal temporal bone resection

Sleeve resection

Extended sleeve resection

A External auditory canal

Tympanic membrane

Mastoid antrum

Chorda tympani

Malleus

Eustachian tube

Cochlea

Cochlear nerve Scala vestibuli (perilymph) Vestibular membrane Scala media Tectorial (endolymph)membrane Hair cells Basilar membrane

B

Vestibular nerve

Cochlear nerve

Oval window Stapes Incus Facial nerve Semicircular Mastoid canal air cells

Scala tympani (perilymph)

FIGURE 15-1.  Anatomy of the ear and related structures is seen in an oblique coronal view (A) and an axial view (B). The organ of Corti is shown in detail.

s­ evere than skin reactions elsewhere in the body because the auricle has an irregular surface, which interferes with skin sparing, and has been sensitized by years of exposure to the sun. The most severe skin reactions are usually seen on the dorsum of the helix and in the sulci, where skin sparing is most impaired. The EAC commonly develops a moist desquamation during radiation therapy, and this can lead to considerable drainage or bleeding. Ceruminous gland function is usually depressed, and late reactions such as stenosis and even ­necrosis have been reported. However, these changes are uncommon and are usually associated with extensive ­tumors or prior surgery.1

Chondronecrosis is rare after fractionated megavoltage irradiation. When this side effect does occur, it usually appears within the first year after treatment.2 However, it may develop years later if the tissues are challenged by subsequent trauma. As is true for bone necrosis in the oral cavity, chondronecrosis usually occurs as a result of an overlying soft tissue necrosis.

Middle Ear Serous otitis media may develop after radiation therapy for tumors in the vicinity of the ear. This side effect is thought to be caused by swelling of the mucosa that leads to obstruction of the eustachian tube and transudation

PART 3  Head and Neck

of a ­sterile serous fluid.3 Serous otitis media is characterized clinically by ear discomfort, a sensation of fullness in the ear, a conductive hearing loss, and occasional tinnitus. This complication usually resolves with the administration of nasal decongestants but sometimes necessitates a myringotomy. The incidence of serous otitis media caused solely by irradiation is not clearly established, but this side effect has been reported in 15% to 50% of cases in some series.4,5 Many of these cases are caused by obstruction of the eustachian tube resulting from tumor or prior surgery.6 In general, the ossicles tolerate radiation therapy well, and necrosis of the ossicular chain after conventionally fractionated radiation therapy is rare. However, at least one case of osteoradionecrosis of the ossicles has been ­reported.7

Inner Ear In the past, sensorineural hearing loss resulting from radiation therapy for head and neck cancer was believed to be rare. However, this side effect is probably more common than previously thought, particularly when high doses have been delivered. Leach8 reported a 36% incidence of sensorineural hearing loss after fractionated doses of 30 to 80 Gy, and Grau and colleagues9 found a 29% incidence of this side effect after doses of 29 to 68 Gy. Moretti10 reported a higher incidence of sensorineural hearing loss (7 of 13 patients) in a small series of patients treated to doses of 60 to 240 Gy for nasopharyngeal cancer. However, these doses are much higher than those typically used clinically. There seems to be a dose-response relationship for ­radiation-induced sensorineural hearing loss. The threshold for this injury occurs at a dose of approximately 40 to 60 Gy in 5 to 6 weeks. Thibadoux and colleagues11 found no hearing loss in children with acute leukemia who received prophylactic cranial irradiation to a dose of 24 Gy in 12 fractions. Similarly, Evans and colleagues12 found no hearing impairment in the irradiated ear compared with the nonirradiated ear in patients with cancer of the parotid gland who received unilateral irradiation to doses of 55 to 60 Gy in 5 to 6 weeks. However, neither of those studies required baseline studies of auditory function before irradiation. Grau and colleagues9 tested patients’ hearing before and after irradiation and found an 8% (1 of 13) incidence of sensorineural hearing loss with fractionated doses up to 50 Gy and a 44% (8 of 18) incidence after doses of 59 Gy or more (Fig. 15-2). These findings correlate well with results of animal studies, which showed significant degeneration of the sensory and supporting cells of the organ of Corti and loss of fibers of the eighth cranial nerve after fractionated doses in excess of 60 Gy.13 Drugs such as cisplatin have ototoxic effects, and the ­ injury is greater in the setting of prior or concurrent irradiation.14 However, radiation therapy ­ delivered after chemotherapy does not worsen the side effects of ­ chemotherapy. This suggests that the synergistic effect of radiation may be caused by hyperemia or chemosensitization.15 Little is known about the effects of radiation on vestibular function, but vestibular damage seems to be rare after conventionally fractionated doses.16 Moskovskaya17

100 Percent with SNHL

322

80

≤ 50 Gy ≥ 59 Gy 8% 44% (1/13) (8/18)

60

2/4

40 20

6/14

1/8 0/3 0/2

0/12 0 0

10

20

30

40

50

60

70

80

Dose to ear (Gy)

FIGURE 15-2.  The incidence of sensorineural hearing loss (SNHL) is plotted as a function of radiation dose delivered to the ear. Only 8% of patients who received a dose of 50 Gy or less developed SNHL, compared with 44% of those who received a dose of 59 Gy or more. (Modified from Grau C, Moller K, Overgaard, M et al. Sensorineural hearing loss in patients treated with irradiation for nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 1991;21:723-728.)

reported increased excitability of the labyrinth during i­rradiation, and Gamble and colleagues18 found that the endolymphatic spaces of the semicircular canals became distended after large single doses of radiation. However, vestibular damage has not been reported with commonly used fractionated doses.3,8

Temporal Bone Necrosis Necrosis of the temporal bone occurs occasionally after high-dose radiation therapy for cancer of the ear. However, this side effect is rarely seen when the temporal bone is incidentally irradiated during the course of treatment for tumors at other head and neck sites. Osteoradionecrosis may be confined to the EAC, or it may involve the entire temporal bone diffusely. Focal ­osteoradionecrosis usually manifests as an area of exposed bone that ultimately forms a sequestrum. When necrosis is limited, the healing process may be slow, but the necrosis usually heals eventually. Diffuse osteoradionecrosis is more difficult to manage. Symptoms of necrosis of the temporal bone include severe boring pain, nausea, vertigo, and otorrhea. There is usually an area of exposed or dead bone, and there may be a fistula extending intracranially or to the skin. The risk of temporal bone radionecrosis is greater with orthovoltage irradiation than it is with megavoltage ­irradiation because of differences in the way kilovoltage and megavoltage beams are absorbed by bone. This complication is also more likely if high total doses have been delivered, if the treatment has not been well ­fractionated, or if there have been focal areas of overdosage, as can occur with poorly designed wedged portal field arrangements. Wang and Doppke19 reported a 60% incidence of ­osteoradionecrosis in patients with tumors of the ear who received doses greater than 2000 ret (rad equivalent therapeutics) (70 Gy). This would be approximately 74 Gy at 2 Gy/fraction if the treatment were given in a standard fashion. Other factors that increase the likelihood of ­temporal bone necrosis include prior surgery and a long history of ear ­infections.

15  The Temporal Bone, Ear, and Paraganglia

Malignant Tumors of the Ear

323

Malignant tumors of the EAC and middle ear are rare, with only about two cases per million people in the United States each year.1 However, the incidence of skin cancer involving the auricle is much greater. In one series,20 approximately 60% of tumors of the ear originated from the auricle, 28% originated in the EAC, and less than 12% were primary tumors of the middle ear. Tumors of the ear are much more common in the elderly than in the young and are slightly more common in women than in men. In most series, the ratio of EAC and middle ear neoplasms to benign ear disease is in the range of 1:5000 to 1:20,000, and most otolaryngologists will see only one or two cases of carcinoma of the ear during their entire careers.21 Because carcinomas of the ear are uncommon, they are usually diagnosed late in the course of the disease. Many patients are treated for chronic otitis for long periods before cancer is suspected. The location of these tumors—deep within the EAC or in the middle ear—also makes early diagnosis difficult, and many of these tumors are extensive when first diagnosed. The results of treatment in most ­series are understandably poor.

reaching the sigmoid sinus and posterior cranial fossa. The petrous part of the temporal bone resists invasion because it is very dense. However, the roof of the tympanic cavity (the tegmen tympani) is thin, and access to the middle cranial fossa through this route is relatively easy. Middle ear tumors can also erode through the floor of the hypotympanum into the tail of the parotid gland or the soft tissues of the upper neck. Middle ear tumors often spread to nearby structures by growing along the nerves and blood vessels that course through the temporal bone. Middle ear tumors may grow along the chorda tympani and the facial nerve to reach the posterior cranial fossa. They may also reach the posterior cranial fossa by eroding through the medial wall of the tympanic cavity and growing along the acoustic nerve. Middle ear tumors may spread to the middle cranial fossa by growing along the tympanic nerve (Jacobson’s nerve) or the lesser superficial petrosal nerve, both of which are branches of cranial nerve IX. Regional metastasis from middle ear tumors is uncommon in most series, occurring only late in the course of the disease.1 Boland and Paterson,25 for example, reported only a 12% (10 of 86) incidence of regional lymph node metastasis from middle ear tumors.

Pathology

Presenting Symptoms

Malignant tumors of the ear are predominantly squamous or basal cell carcinomas. However, the relative incidence of these two histopathologic types differs for various subsites. Approximately 55% of tumors involving the auricle are squamous cell carcinomas, and 45% are basal cell carcinomas,22 whereas 90% of the tumors involving the EAC and almost all of those involving the middle ear are squamous cell carcinomas. Other tumors that can originate in the ear include melanomas, adenoid cystic carcinomas, adenocarcinomas of the sebaceous and ceruminous glands, Ewing’s sarcomas, osteosarcomas, and rhabdomyosarcomas.

Patients with carcinomas of the EAC and the middle ear usually present with granulation tissue and a chronic draining ear or a polypoid mass visible in the EAC. The tympanic membrane is often perforated, and considerable purulent drainage is usually present. As the disease progresses, the patient may develop facial nerve paralysis, vertigo, tinnitus, or deep, boring pain. If the tumor invades the parotid gland or the soft tissues of the neck, it may be palpable as preauricular induration or an upper neck mass.

Routes of Spread External Auditory Canal Basal cell carcinomas of the EAC tend to originate at the entrance of the cartilaginous portion of the canal, whereas squamous cell carcinomas of the EAC can develop anywhere along the canal. Both types of tumors tend to fill the canal before invading the perichondrium or periosteum. After the cartilage is invaded, however, the tumor can spread anteriorly to the parotid gland or posteriorly into the posterior auricular area. Carcinomas of the bony portion of the canal usually spread longitudinally before infiltrating the middle ear. Squamous cell carcinomas of the EAC metastasize to regional lymph nodes in 10% to 20% of cases.23,24 The most common sites of regional metastasis are the preauricular, subdigastric, and postauricular lymph nodes. There are no reports of metastasis from basal cell carcinomas of the EAC.

Middle Ear Tumors of the middle ear can grow medially along the ­eustachian tube to the nasopharynx or erode through the wall of the eustachian tube to involve the petrous apex and carotid canal. They may extend posteriorly through the mastoid antrum to the mastoid air cells, ultimately

Clinical Evaluation Clinical evaluation of patients with tumors of the ear should include careful examination of the auricle, the EAC, and the tympanic membrane. A neurologic examination should be performed to detect facial nerve paralysis or other cranial nerve palsies. The facial nerve may be affected even if the tumor is confined within the temporal bone. However, involvement of cranial nerves IX to XI usually means that the tumor has spread beyond the temporal bone. The patient should be evaluated for trismus, the presence of which indicates the possibility of invasion of the pterygoid muscles or the temporomandibular joint. The neck should be examined for lymphadenopathy, especially in the subdigastric, preauricular, and postauricular areas. Hearing should be tested with a tuning fork, and baseline audiometry should be performed before treatment. In the past, most tumors of the ear were studied with plain roentgenograms and laminar tomograms of the skull and temporal bone. These studies have been replaced for the most part with high-resolution computed tomography (CT) and magnetic resonance imaging (MRI), which are better able to delineate local tumor extension. Often, both studies are necessary. CT is superior for detecting bone erosion, and MRI is better for assessing intracranial extension. Directly acquired coronal CT images (rather than reconstructed images) should be obtained for this evaluation, along with standard axial sections.

324

PART 3  Head and Neck

(Table 15-2).26,28-34 For example, Crabtree and colleagues28 reported a survival rate of 91% (20 of 22 patients) at 1 to 5 years after treatment for patients with tumors confined to the EAC. This rate is much better than the 30% to 40% 5-year survival rates usually reported for patients with tumors involving the middle ear.30,35,36 Survival correlates well with the extent of the disease. Goodwin and Jesse,26 for example, reported absolute 5-year survival rates of 57% for patients with cancer involving the cartilaginous portion of the EAC or the adjacent concha, 45% for patients with tumors involving the bony portion of the EAC or the mastoid cortex, and 29% for patients with tumors involving the deep structures of the temporal bone. Resectability is one of the most important prognostic indicators. In Goodwin and Jesse’s series,26 11 (58%) of 19

CT and MRI are usually adequate for assessing the s­ tatus of the carotid canal and the sigmoid and transverse sinuses. However, carotid angiography with balloon occlusion may be necessary if sacrifice of a carotid artery is being considered.

Staging The American Joint Committee on Cancer (AJCC) staging system has no classification for carcinoma of the ear, but several staging systems have been proposed that are based on a variety of prognostic factors.26,27 The staging system outlined in Table 15-1 is a modification of a classification proposed by Goodwin and Jesse.26 Carcinomas that are limited to the EAC usually have a better prognosis than those that involve the middle ear

TABLE 15-1.    General Policies for Treatment of Carcinomas of the External, Middle, or Inner Ear Tumor Extent

Description

Treatment

T1

Carcinoma of the external auditory canal with or without ­invasion of the cartilaginous portion of the canal or pinna (no involvement of the bony canal or tympanic membrane)

Sleeve resection* or local radiation therapy alone

T2

Tumor of the external auditory canal with minimal invasion of the bony canal or extension to tympanic membrane

Extended sleeve resection* with or without postoperative radiation therapy, depending on margins

T3

Technically resectable cancer that extends beyond the ­external auditory canal, including tumors of the middle ear and mastoid

Subtotal temporal bone resection* and ­postoperative radiation therapy

T4

Technically unresectable tumor (extension to apex of petrous bone, cavernous sinus, internal auditory canal, internal carotid artery, middle or posterior cranial fossa, Cl or C2, or base of occiput)

Radiation therapy or no treatment

*See Figure 15-3. Modified from Goodwin WJ, Jesse RH. Malignant neoplasms of the external auditory canal and temporal bone. Arch Otolaryngol 1980;106: 675-679.

TABLE 15-2.   Local Control Rates in the Treatment of Carcinoma of the External Auditory Canal and Middle Ear Limited Disease (T1-T2) Study and Reference Arriaga et Hahn et

al29

al30

Goodwin and

Jesse26

Lesser et al31 al28

Extensive Disease (T3-T4)

S

S + RT

S

S + RT

3/4

4/9

1/2

10/20

3/4

2/2

1/8

4/10

T1: 37/56* T2: 5/11

— 5/5

6/13 —

5/19 —

5/7†

—

3/17

—

20/22

3/3

0/5

3/9

Wang32

—

4/4

—

7/19

Gacek and Goodman33

3/3

7/8

—

8/18

Moody et al34

4/4

7/8

0/5

4/15

Subtotal

75/104 (72%)

37/46 (80%)

8/33 (24%)

44/127 (35%)

Crabtree et

Total

112/150 (75%)

*Includes five patients treated with radiation therapy alone. †Includes nine patients treated with radiation therapy alone. S, Surgery; RT, radiation therapy.

52/160 (33%)

15  The Temporal Bone, Ear, and Paraganglia

325

TABLE 15-3.   Comparison of Local Control Rates Achieved with Surgery Alone or Surgery Followed by Irradiation Stage

Surgery

Surgery + Irradiation

Total

Group I (confined to cartilaginous canal concha: T1)

—

—

37/56 (66%)*,†

Group II (involving bony canal or mastoid cortex: T2)

5/11 (45%)

5/5 (100%)

10/16 (62.5%) †

Group III (involving middle ear, facial canal, or base of skull: T3 and some T4) Complete resection Incomplete resection

6/13 (46%) —

5/6 (83%) 0/13 (0%)

11/19 (58%) 0/13 (0%)

*Mainly

treated with surgery. patients (groups I and II) classified as having persistent local tumor received salvage treatment. Modified from Goodwin WJ, Jesse RH. Malignant neoplasms of the external auditory canal and temporal bone. Arch Otolaryngol 1980;106: 675-679.

†Surgery

patients who had complete resection of their middle ear tumors (T3) had local tumor control, compared with none of 13 patients who had incomplete resection with or without radiation therapy (T4) (Table 15-3).26

Subtotal temporal bone resection

Treatment Treatment options by stage for carcinomas of the ear are shown in Table 15-1. Small tumors confined to the EAC can usually be treated satisfactorily with a single ­modality—surgery or radiation therapy. Larger EAC tumors or tumors originating in the middle ear are usually best treated with combined surgery and radiation therapy. Massive tumors are usually best managed palliatively with radiation ­therapy, although some patients with massive tumors do not benefit from any form of treatment. Patients with tumors of the ear should be evaluated for signs of technical inoperability such as evidence of disease extension to the internal auditory canal, the middle or posterior cranial fossa, the body of the atlas or axis, or the base of the occiput. Tumors of the ear are usually not technically resectable if cranial nerves IX to XI are involved, and although resection of the carotid artery is technically feasible in some cases, it usually results in undesirable long-term sequelae.24 Tumor extension to the parotid gland, the ascending ramus of the man­ dible, or the soft tissues of the upper neck is not a sign of ­ technical unresectability, nor is limited ­ involvement of the dura.24 The technical details of radiation therapy for carcinomas of the ear are described under “Radiation Therapy Technique” later in this chapter.

Sleeve resection

Extended sleeve resection

Sleeve resection

Extended sleeve resection

T1 Disease Tumors confined to the cartilaginous portion of the EAC and adjacent concha without extension to the bony portion of the EAC or tympanic membrane can be treated with surgery or radiation therapy alone. Surgery for these lesions usually consists of a sleeve resection of the auditory canal, which means removal of a core of skin, cartilage, and possibly some bone (Fig. 15-3). The defect is covered with a split-thickness skin graft. T1 tumors can also be treated with radiation therapy alone. The appropriate radiation dose is determined by the size of the tumor and ranges from 60 Gy in 6.5 weeks to 70 Gy in 7 to 7.5 weeks.

Subtotal temporal bone resection

FIGURE 15-3.  Surgical procedures are outlined for carcinomas of the ear.

T2 Disease Cancer that involves the bony portion of the EAC can be treated with an extended sleeve resection, which consists of removal of a core of skin, cartilage, and bone; the ­tympanic membrane; and the malleus or incus, or both

326

PART 3  Head and Neck

structures. The adjacent mastoid may also be removed (see Fig. 15-3). Postoperative radiation therapy is added if the resection margins are close or positive. The postoperative dose is 60 Gy in 6 to 6.5 weeks if the margins are negative and 65 to 70 Gy in 6.5 to 7 weeks if residual disease is present.

T3 Disease If the middle ear is involved, subtotal temporal bone resection (see Fig. 15-3) and postoperative radiation therapy are required. This treatment involves removal of the entire temporal bone en bloc except for the petrous apex. The tumor may have to be dissected off the dura of the middle and posterior cranial fossa, and the ipsilateral sigmoid sinus may have to be sacrificed. This can be accomplished if the contralateral sinus is patent. Tumor extension to the parotid gland, the mandible, or the neck may necessitate a parotidectomy, partial mandibulectomy, or radical neck dissection. Postoperative radiation therapy is almost always rec­ ommended for patients with T3 disease regardless of the resection margins because the local failure rate after treatment with surgery alone is high (see Table 15- 3).26 The radiation therapy doses are determined by the margins of resection, the volume of residual disease, and the tolerance of adjacent normal tissues and usually range from 60 Gy in 6 to 6.5 weeks to 70 Gy in 7 to 7.5 weeks.

T4 Disease Patients with technically unresectable tumors of the ear are usually treated with high doses of radiation, although the goal of treatment is often only palliation. The doses are usually in the range of 60 to 70 Gy or greater in 6.5 to 7.5 weeks.

Radiation Therapy Technique For temporal bone malignancies, a shrinking field technique is usually used. The initial treatment portals should include all clinically detectable tumor and areas at risk for subclinical disease. If the treatment is being given after surgery, the entire surgical bed must be encompassed in the treatment portals. After 45 to 50 Gy, the fields are reduced to include only the areas of gross tumor involvement with a small margin. Care is taken to limit the dose to critical normal structures, such as the optic chiasm, brainstem, temporal lobe, and spinal cord. If the tumor is confined to the cartilaginous portion of the EAC, the treatment fields should encompass the EAC with a 2- to 3-cm margin. A variety of field arrangements may be used depending on the clinical situation, including ipsilateral electron beams, mixed electron and photon beams, and wedged photon beams. The electron beam energy should be determined on the basis of the depth of the disease. Tumors involving the middle ear are usually more advanced, with a larger target volume. Patients with middle ear tumors may be treated with combined photon and electron beam portals if there is no bone invasion. ­ However, if the temporal bone is involved, wedged photon beams should be used rather than an electron-photon ­combination because the distribution of the electron beam dose in the dense temporal bone is uncertain.

Anteroposterior and superoinferior wedged portal arrangements have been used for tumors of the temporal bone (Fig. 15-4). An anteroposterior orientation is usually preferred because the isotope distribution conforms to the general shape of the temporal bone and makes it easier to exclude the brainstem (Figs. 15-5 and 15-6). However, the posterior oblique field must be angled inferiorly to avoid exit through the contralateral eye. The brainstem is usually included in the anterior oblique portal but excluded from the posterior oblique portal. This results in a triangular isodose distribution with the apex of the distribution situated anterior to the spinal cord.

Results Survival rates for tumors of the ear correlate well with the location and the extent of the disease. Five-year survival rates are usually in the range of 70% to 75% for patients with early cancer of the EAC (T1 and T2), compared with 40% to 45% for patients with technically resectable cancer involving the middle ear (T3). The results for patients with technically unresectable cancer (T4) are very poor, with only 0% to 20% of patients surviving 5 years. The results achieved with surgery alone, radiation therapy alone, and combined surgery and radiation therapy are difficult to compare because treatment methods are selected on the basis of the extent of the tumor. Surgery is not chosen unless the tumor is thought to be technically resectable and the patient able to tolerate the surgery. However, radiation therapy is often used for palliative treatment, and postoperative radiation therapy is usually added when there is a significant risk of residual disease after surgery (see Table 15-2). In general, patients who have been treated with surgery alone have had less extensive disease than those treated with radiation therapy or combined surgery and radiation therapy. In a series from The University of Texas M. D. Anderson Cancer Center,26 66% of patients with Tl tumors had local tumor control after treatment with a single modality (see Table 15-3). The overall local control rate for patients with T2 disease was similar (62%). However, patients who received surgery followed by radiation therapy had a much higher local control rate than those who were treated with surgery alone (5 of 5 versus 5 of 11). Similarly, the local control rate was better when postoperative radiation therapy was added for patients with T3 cancer of the middle ear (5 of 6 versus 6 of 13). Local tumor control was not obtained in any of the 13 patients with an incompletely resected middle ear tumor (T4), even with the addition of postoperative radiation (see Table 15-3).

Chemodectomas Chemodectomas are a group of rare tumors that arise from the chemoreceptor system. They are also called nonchromaffin paragangliomas and glomus tumors. Chemodectomas are three to four times more common in women than in men. The peak incidence is in the fifth decade of life, but tumors have been reported in patients as young as 22 months and as old as 85 years.37 Chemodectomas cause symptoms by progressive enlargement and compression or by infiltration of adjacent bone, nerves, and blood vessels. Although histologically benign, these tumors have been reported to metastasize in

15  The Temporal Bone, Ear, and Paraganglia

327

Cobalt 60 SSD 80 cm Portal Size Weight 1. Anterior oblique 45-degree wedge 8 × 7 cm 1 2. Posterior oblique 45-degree wedge 8 × 7 cm 1

Inferior

Posterior 110%

110%

100% 90% Midline

A

80%

Isodose distribution from anteroposterior 45-degree wedge portals

Cobalt 60 Portal 1. Superior 45-degree wedge 2. Inferior 45-degree wedge 3. Lateral open

Superior

SSD 80 cm Weight Size 1 9 × 7 cm 9 × 7 cm 1 8 × 7 cm 2

Inferior

110%

110% 100% 85%

Midline

B

65%

Isodose distribution using superoinferior pairs of 45-degree wedge and lateral open Cobalt 60 fields

FIGURE 15-4.  Isodose distributions are shown for two treatment plans. A, Anteroposterior 45-degree wedge filtered portals.  B, Three-field technique using superoinferior wedge filtered portals and a lateral open field.

2% to 6% of cases.38 They are the second most common benign neoplasm of the base of the skull, with acoustic neuromas being the most common.

Anatomy The normal glomus bodies, or paraganglia, are derived from neural crest cells. They are found in all parts of the body, but they are much more common in the head and neck and mediastinum than elsewhere. The glomus bodies of the head and neck and mediastinum are called the branchiomeric paraganglia because they are located near the great vessels and cranial nerves of the primitive branchial arches. The branchiomeric paraganglia include the jugulotympanic, intercarotid, subclavian, laryngeal, coronary, aortopulmonary, and pulmonary glomus bodies. Normal glomus bodies are very small, ranging from 0.1 to 1.5 mm in diameter.39 They are composed of two primary cell types: granule-storing chief cells and Schwannlike satellite cells. These cells are intermixed with a rich ­capillary network. The chief cells are thought to be the

FIGURE 15-5.  An angled wedged pair is used to deliver a high dose to the right temporal bone while sparing the underlying brain.

328

PART 3  Head and Neck

Approximately 10% of chemodectomas are multicentric, and they are highly associated with other neoplasms, especially other neural crest tumors.42,43 Chemodectomas can be familial, and the incidence of multicentric tumors is greater among patients with a familial history of the disease than in sporadic cases.44 In one series, 33% of 209 patients with familial glomus tumors had multicentric tumors.39 30

Presenting Symptoms

30

5050 50

30

90

30

The most common presenting symptoms of temporal bone chemodectoma are mild hearing loss, pulsatile tinnitus, and facial nerve paralysis. These symptoms are usually gradual in onset but become more severe as the disease progresses. Other common symptoms include bloody or purulent otorrhea, vertigo, and palsy of cranial nerves IX to XII. The differential diagnosis for a glomus tumor includes chronic otitis media, cholesterol granuloma, carcinoma of the middle ear, vascular anomalies, and primary brain ­tumors.37

Clinical Evaluation

FIGURE 15-6.  Isodose curves are shown for a patient treated with a portal arrangement similar to that shown in Figure 15-5.

functional cells. They contain dense granules and ­numerous organelles that contain acetylcholine, catecholamines, and serotonin. The glomus bodies are an important part of the chemoregulatory system. They respond to a variety of chemical changes within the body and help to keep the body in physiologic balance. The carotid bodies, for example, seem to serve as initiators of increased ventilation when oxygen tension is low and as suppressors of ventilation when oxygen tension is high.40,41 In addition to being hypoxia receptors, the paraganglia may be receptors for monitoring changes in pH, carbon dioxide concentration, blood flow, and temperature.39

Pathology The distribution of chemodectomas throughout the body corresponds to the anatomic distribution of the normal paraganglia. In the head and neck, chemodectomas are seen most commonly near the carotid bifurcation (carotid body tumors), the jugular bulb (glomus jugular tumors), the tympanic and auricular nerves (glomus tympanicum tumors), and the vagus nerve at the base of the skull (vagal body tumors). Glomus body tumors may also be found in the nasal cavity, orbit, pharynx, and larynx. Grossly, chemodectomas can appear as small, well­localized, reddish purple masses or as deeply infiltrating tumors destroying bone and adjacent vital structures. Histopathologically, chemodectomas are benign, similar in appearance to the normal paraganglia. They tend to form nests of 15 to 20 chief cells arranged haphazardly within a rich vascular stroma. Electron microscopy reveals ­neurosecretory-like granules in the chief cells.

Clinical evaluation of a patient with a suspected chemodectoma should include an otoscopic examination of the EAC and tympanic membrane. In most cases, a reddish purple discoloration of the tympanic membrane is evident. However, an aural polyp or tumor may obscure the tympanic membrane. The mass commonly pulsates, and it may blanch when external pressure is applied with a pneumatic otoscope (Brown’s sign).45 A complete neurologic examination should be performed to evaluate possible cranial nerve deficits. The facial nerve is the cranial nerve most commonly involved because of its location within the temporal bone. The incidence of paralysis of the facial nerve with glomus tumors ranges from 10% to 40%.39 Cranial nerves IX to XII are also commonly involved, and cranial nerves V and VI may be impaired if the tumor extends to the middle cranial fossa. This is usually seen only when the disease is advanced. Radiographic studies should include high-resolution CT to evaluate possible bone destruction and to delineate the soft tissue mass and MRI to determine whether intracranial extension is present (Figs. 15-7 and 15-8). Angiography should be performed to evaluate the arterial supply if the patient is being considered for surgery or embolization. Arteriography can be used to differentiate a glomus tumor from other vascular and nonvascular lesions of the temporal bone and may eliminate the need for a biopsy (Fig. 15-9).37

Clinical Staging The prognosis for a patient with a glomus tumor is closely related to the size of the tumor and its anatomic location. The following classification was proposed by McCabe and Fletcher46: Group I Tympanic tumors, characterized by absence of bone destruction on radiographs of the mastoid bone and jugular fossa, absence of facial nerve weakness, an intact eighth cranial nerve with conductive deafness only, and intact jugular foramen nerves (cranial nerves IX, X, and XI) Group II Tympanomastoid tumors, characterized by radiographic evidence of bone destruction confined to the mastoid bone and not involving the petrous

15  The Temporal Bone, Ear, and Paraganglia

FIGURE 15-7.  CT scan shows an enhancing destructive lesion  (arrow) in the region of the left jugular foramen with marked expansion and bony destruction in a 70-year-old woman.

329

FIGURE 15-9.  A 17-year-old woman presented with complaints of decreased hearing, a sensation of aural fullness, and pulsatile tinnitus. Otoscopic examination revealed a blue mass bulging behind the tympanic membrane. Plain films suggested a glomus jugulare tumor. Angiography revealed an aberrant carotid artery. Dashed lines indicate the normal course of the vessel. Biopsy of similar lesions has occasionally been fatal.

Treatment

FIGURE 15-8.  Magnetic resonance image of the same patient shown in Figure 15-7.

bone, a normal or paretic seventh cranial nerve, intact jugular foramen nerves, and no evidence of involvement of the superior bulb of the jugular vein on retrograde venography Group III Petrosal and extrapetrosal tumors, characterized by evidence of destruction of the petrous bone, jugular fossa, or occipital bone on radiography; positive findings on retrograde jugulography; evidence of destruction of the petrous occipital bone on arteriography; jugular foramen syndrome (paresis of cranial nerve IX, X, or XI); or presence of metastasis

The treatment for chemodectomas of the temporal bone is controversial. For limited chemodectomas, otolaryngologists often prefer surgery, and radiation oncologists usually prefer radiation therapy. Most agree that surgery is effective treatment for early tumors that can be completely excised. For example, surgery is usually adequate for small glomus tympanicum tumors of the middle ear (group I). These tumors can usually be excised through the ear canal or the mastoid with reasonable assurance of complete removal. However, surgery is less likely to be successful for more extensive glomus tympanicum tumors or for glomus jugulare tumors (groups II and III). Because these tumors are often resected in a piecemeal fashion, there is a significant risk of residual disease. There is also a significant risk of hemorrhage. The difficulties associated with surgery for glomus jugulare tumors are reflected in a poor initial local control rate with surgery alone (Table 15-4)47-60 and a high complication rate (up to 60% in the surgical literature).51 Complications of surgery include hemorrhage, cranial nerve deficits, dysphagia, wound infections, cerebrospinal fluid leaks, and meningitis. Group II and III chemodectomas of the temporal bone are usually better treated with radiation therapy because of its effectiveness in ameliorating symptoms and in halting the progression of the disease. However, a mass often persists clinically and angiographically after radiation therapy, and some surgeons have interpreted this as evidence that radiation therapy is not effective for chemodectomas. In most radiation therapy series, patients have been considered to have local tumor control if no clinical or radiographic evidence of disease progression is found at the time

330

PART 3  Head and Neck

TABLE 15-4.   Local Control with Surgery for Chemodectoma of the Temporal Bone: Literature Review

TABLE 15-5.   Local Control with Radiation Therapy for Chemodectomas of the Temporal Bone: Review of Literature

Study and Reference

Local Control Rate

Study and Reference

Local Control Rate*

Nominal Dosage Schedule

Royal Marsden Hospital47

0/4

Royal Marsden ­Hospital47

52/59

35-66 Gy in 4-6 wk

University of Virginia48

25/29

30-60 Gy¶

University of Iowa49

14/14†

29-67.5 Gy in 1.5-7 wk

University of Florida50

51/55

37.7-60 Gy in 4-5 wk

University of Kansas52

4/4

22-56 Gy in 3-6 wk

Geisinger Medical Center54

20/22

40-50 Gy¶

Arthus Municipal ­Hospital, Denmark55

13/15

50-60 Gy in 5-6 wk

University of California, San Francisco56

6/6

46-55 Gy 6-8 wk

Massachusetts General Hospital57

13/16

30-45 Gy¶

Washington University, St. Louis61

12/15

46-55 Gy¶

Princess Margaret ­Hospital, Toronto62

42/45‡

35 Gy in 3 wk

Rotterdamsch ­ Radio-Therapeutisch Institute, The Netherlands63

19/19

40-60 Gy in 4-6 wk

University of ­Washington64

10/13

28-65 Gy in 4-7 wk

University Hospital of Wales, Cardiff65

2/14

42.5-55 Gy in 12-30 fractions

Queen Elizabeth ­Hospital, UK66

22/23§

45-50 Gy in 4-5 wk

M.D. Anderson Cancer Center67

16/17

45-50 Gy in 4.5-5 wk

Mount Sinai Hospital (New York)68

6/6

40-50 Gy in 4-5 wk

University of ­Minnesota69

13/14

30-60 Gy in 3.5-7.5 wk

Mayo Clinic70

25/25

15 Gy with ­radiosurgery

Hospital Na Homolce, Prague71

47/66

20-60 Gy with ­radiosurgery#

Total

412/477 (86%)

University of

Virginia48

17/20

University of

Iowa49

6/13

University of Florida50

4/6

Mount Sinai Hospital51

16/17

University of Kansas Medical Center52

3/13

Washington University, St. Louis53

10/11* 35/45†

Geisinger Medical Center54

4/8

Arthus Municipal Hospital, Denmark55

4/6

University of California, San Francisco56

3/14

Massachusetts General

Hospital57

Mount Sinai Hospital (New

York)58

8/16 14/23‡

Nashville, TN59

73/80*

Rome, Italy60

43/53†

Total

240/329 (73%)

*Glomus

tympanicum. jugulare. ‡One patient lost to follow-up was excluded. †Glomus

of the analysis or until death from intercurrent disease. Using these criteria, the overall local control rate with radiation therapy is almost 90% (Table 15-5).47-50,52,54-57,61-71 The incidence of complications after radiation therapy is low because the radiation dose required for local tumor control is only 45 to 50 Gy in 4.5 to 5 weeks. Wang and colleagues49 compared the results achieved with radiation therapy with those attained with surgery alone. In that series, 13 patients were treated with surgery alone, 15 with radiation therapy alone, and 4 with combined surgery and radiation therapy. In general, the patients treated with radiation therapy or combined surgery and radiation therapy had more advanced disease than those treated with surgery alone. The initial local control rate with surgery was 46% (85% after salvage with additional treatment); 31% of patients developed complications, and 78% survived 10 years. The initial local control rate with radiation therapy was 84%; 11% of patients developed complications, and 76% survived 10 years. These results show that radiation therapy is an effective treatment for chemodectomas of the temporal bone.

Radiation Therapy Technique The treatment portals for chemodectomas of the temporal bone are similar to those outlined earlier in this chapter for carcinomas of the ear. Small portals can usually be used for relatively localized glomus tympanicum tumors (group I). However, larger portals must be used for the more advanced tympanomastoid, petrosal, and extrapetrosal tumors

*Includes patients treated with radiation therapy alone or a combination of surgery and radiation therapy. †Previously untreated patients. If patients treated for recurrent ­disease are included, the local control rate is 16 (84%) of 19. ‡Includes two patients for whom the initial treatment failed but salvage therapy was successful. §One patient for whom the initial treatment failed had successful salvage radiation therapy. ¶Fraction schedule not specified. #Maximum dose.

15  The Temporal Bone, Ear, and Paraganglia

(groups II and III). Most patients are treated with photons using a wedged field portal arrangement (see Fig. 15-4). Superoinferior or anteroposterior wedge portal arrangements can be used. Sometimes, a three-field portal arrangement, including two wedge fields and a lateral open field (­weighted according to the anatomy, field size, beam energy, and wedge thickness), is useful (see Fig. 15-4). The treatment is usually delivered with daily fraction sizes of 1.8 to 2 Gy to a total tumor dose of 45 Gy in 5 to 5.5 weeks.

REFERENCES   1. Stell PM. Carcinoma of the external auditory meatus and middle ear. Clin Otolaryngol 1984;9:281-299.   2. Parsons JT. The effect of radiation on normal tissues of the head and neck. In Million RR, Cassisi NJ (eds): Management of Head and Neck Cancer. Philadelphia: JB Lippincott, 1984, pp 173-207.   3. Borsanyi SJ, Blanchard CL. Ionizing radiation and the ear. JAMA 1962;181:958-961.   4. O’Neill JV, Katz AH, Skolnik EM. Otologic complications of radiation therapy. Otolaryngol Head Neck Surg 1979;87:359-363.   5. Brill AH, Martin MM, Fitz-Hugh GS, et al. Postoperative and postradiotherapeutic serous otitis media. Arch Otolaryngol 1974;99:406-408.   6. Myers EN, Beery QC, Bluestone CD, et al. Effect of certain head and neck tumors and their management on the ventilatory function of the eustachian tube. Ann Otol Rhinol Laryngol Suppl 1984;114:3-16.   7. Kveton JF, Sotelo-Avila C. Osteoradionecrosis of the ossicular chain. Am J Otol 1986;7:446-448.   8. Leach W. Irradiation of the ear. J Laryngol Otol 1965;79: 870-880.   9. Grau C, Moller K, Overgaard M, et al. Sensorineural hearing loss in patients treated with irradiation for nasopharyngeal ­carcinoma. Int J Radiat Oncol Biol Phys 1991;21:723-728. 10. Moretti JA. Sensorineural hearing loss following radiotherapy to the nasopharynx. Laryngoscope 1976;85:598-602. 11. Thibadoux GM, Pereira XW, Hodges JM, et al. Effects of cranial radiation on hearing in children with acute lymphocytic leukemia. J Pediatr 1980;96:403-406. 12. Evans RA, Liu KC, Azhar T, et al. Assessment of permanent hearing impairment following radical megavoltage radiotherapy. J Laryngol Otol 1988;102:588-589. 13. Bohne BA, Marks JE, Glasgow GP. Delayed effects of ionizing radiation on the ear. Laryngoscope 1985;195:818-828. 14. Kretschmar CS, Warren MP, Lavally BL, et al. Ototoxicity of preradiation cisplatin for children with central nervous system tumors. J Clin Oncol 1990;8:1191-1198. 15. Walker DA, Pillow J, Waters KD, et al. Enhanced cisplatinum ototoxicity in children with brain tumors who have received simultaneous or prior cranial irradiation. Med Pediatr Oncol 1989;17:48-52. 16. Berg NO, Lindgren M. Dose factors and morphology of delayed radiation lesions of the internal and middle ear in rabbits. Acta Radiol 1961;56:305-319. 17. Moskovskaya NV. Effect of ionizing radiation on function of vestibular analyzer. Otorinolaringologie 1960;22:43-49. 18. Gamble JE, Peterson EA, Chandler JR. Radiation effects on the inner ear. Arch Otolaryngol 1968;88:156-161. 19. Wang CC, Doppke K. Osteoradionecrosis of the temporal bone: consideration of nominal standard dose. Int J Radiat Oncol Biol Phys 1976;1:881-883. 20. Lewis JS. Cancer of the ear: a report of 150 cases. Laryngoscope 1960;50:551-579. 21. Kinney SE. Squamous cell carcinoma of the external auditory canal. Am J Otol 1989;10:111-116. 22. Hyams VJ. Pathology of tumors of the ear. In Thawley SE, Panje WR (eds): Comprehensive Management of Head and Neck Tumors. Philadelphia: WB Saunders, 1987, pp 168-180. 23. Lewis JS. Cancer of the ear. CA Cancer J Clin 1987;37:78-87.

331

24. Jesse RH. External auditory canal, middle ear, and mastoid. In MacComb WS, Fletcher GS (eds): Cancer of the Head and Neck. Baltimore: Williams & Wilkins, 1967, pp 412-427. 25. Boland J, Paterson R. Cancer of the middle ear and external auditory meatus. J Laryngol 1955;69:468-478. 26. Goodwin WJ, Jesse RH. Malignant neoplasms of the external auditory canal and temporal bone. Arch Otolaryngol 1980;106: 675-679. 27. Stell PM, McCormick MS. Carcinoma of the external auditory meatus and middle ear. J Laryngol Otol 1985;99:847-850. 28. Crabtree JA, Britton BH, Pierce MK. Carcinoma of the external auditory canal. Laryngoscope 1976;86:405-415. 29. Arriaga M, Hirsch BE, Kamerer DB, et al. Squamous cell carcinoma of the external auditory meatus (canal). Otolaryngol Head Neck Surg 1989;101:330-337. 30. Hahn SS, Kim JA, Goodchild N, et al. Carcinoma of the middle ear and external auditory canal. Int J Radiat Oncol Biol Phys 1983;9:1003-1007. 31. Lesser RW, Spector GJ, Devineni VR. Malignant tumors of the middle ear and external auditory canal: a 20-year review. Otolaryngol Head Neck Surg 1987;96:43-47. 32. Wang CC. Radiation therapy in the management of carcinoma of the external auditory canal, middle ear, or mastoid. Radiology 1975;116:713-715. 33. Gacek RR, Goodman M. Management of malignancy of the temporal bone. Laryngoscope 1977;87:1622-1634. 34. Moody SA, Hirsch BE, Myers EN. An evaluation of a staging system. Am J Otol 2000;21:582-588. 35. Sinha PP, Aziz HI. Treatment of carcinoma of the middle ear. Radiology 1978;126:485-487. 36. Lederman M. Malignant tumors of the ear. J Laryngol Otol 1965;79:185-199. 37. Zak FG, Lawson W. The Paraganglionic Chemoreceptor System: Physiology, Pathology, and Clinical Medicine. New York: ­Springer-Verlag, 1982. 38. Irons GB, Weiland LH, Brown WL. Paragangliomas of the neck: clinical and pathologic analysis of 116 cases. Surg Clin North Am 1977;57:575-583. 39. Smith PG, Schwaber MK, Goebel JA. Clinical evaluation of glomus tumors of the ear and the base of the skull. In Panje WR, Thawley SE (eds): Comprehensive Management of Head and Neck Tumors. Philadelphia: WB Saunders, 1987, pp 207-218. 40. Arias-Stella J, Valcarcel J. Chief cell hyperplasia in the human carotid body at high altitudes: physiologic and pathologic significance. Hum Pathol 1976;7:361-373. 41. Lack EE. Carotid body hypertrophy in patients with cystic fibrosis and cyanotic congenital heart disease. Hum Pathol 1977;8:39-51. 42. Hayes HM Jr, Fraumeni JF Jr. Chemodectomas in dogs: epidemiologic comparison with man. J Natl Cancer Inst 1974;52:14551458. 43. Spector GJ, Ciralsky R, Maisel RH, et al. Multiple glomus tumors in the head and neck. Laryngoscope 1975;85:1066-1075. 44. Cook PL. Bilateral chemodectomas in the neck. J Laryngol Otol 1977;91:611-618. 45. Brown LA. Glomus jugulare tumor of the middle ear: clinical aspects. Laryngoscope 1953;53:281-292. 46. McCabe BF, Fletcher M. Selection of therapy of glomus jugulare tumors. Arch Otolaryngol 1969;89:156-159. 47. Powell S, Peters N, Harmer C. Chemodectoma of the head and neck: results of treatment in 84 patients. Int J Radiat Oncol Biol Phys 1992;22:919-924. 48. Larner JM, Hahn SS, Spaulding CA, et al. Glomus jugulare tumors. Cancer 1992;69:1813-1817. 49. Wang ML, Hussey DH, Doornbos JF, et al. Chemodectoma of the temporal bone: a comparison of surgical and radiotherapeutic results. Int J Radiat Oncol Biol Phys 1988;14:643-648. 50. Hinerman RW, Mendenhall WM, Amdur RJ, et al. Definitive radiotherapy in the management of chemodectomas arising in the temporal bone, carotid body, and glomus vagale. Head Neck 2001;23:363-370. 51. Cece JA, Lawson W, Biller HF, et al. Complications in the management of large glomus jugulare tumors. Laryngoscope 1987;97:152-157.

332

PART 3  Head and Neck

52. Reddy EK, Mansfield CM, Hartman GV. Chemodectomas of glomus jugulare. Cancer 1983;52:337-340. 53. Spector GJ, Fierstein J, Ogura JH. A comparison of therapeutic modalities of glomus tumors in the temporal bone. Laryngoscope 1976;86:690-696. 54. Cole JM. Glomus jugulare tumor. Laryngoscope 1977;87: 1244-1258. 55. Thomsen K, Elbrond O, Andersen AP. Glomus jugulare tumors (a series of 21 cases). J Laryngol Otol 1975;89:1113-1121. 56. Newman H, Rowe JF Jr, Phillips TL. Radiation therapy of the glomus jugulare tumor. AJR Am J Roentgenol 1973;118: 663-669. 57. Hatfield PM, James AE, Schulz MD. Chemodectomas of the glomus jugulare. Cancer 1972;30:1164-1168. 58. Rosenwasser H. Glomus jugulare tumors: long-term tumors jugulare. Arch Otolaryngol 1969;89:186-192. 59. Forest JA 3rd, Jackson CG, McGrew BM. Long-term control of surgically treated glomus tympanicum tumor. Otol Neurotol 2001;22:232-236. 60. Sanna M, Jain Y, De Donato G, et al. Management of jugular paragangliomas: the Gruppo Otologico experience. Otol Neurotol 2004;25:797-804. 61. Konefal JB, Pilepich MV, Spector GJ, et al. Radiation therapy in the treatment of chemodectomas. Laryngoscope 1987;97: 1331-1335. 62. Cummings BJ, Beale FA, Garrett PG, et al. The treatment of glomus tumors in the temporal bone by megavoltage radiation. Cancer 1984;53:2635-2640. 63. Lybeert MLM, Van Andel JG, Eijkenboom WMH, et al. Radiotherapy of paragangliomas. Clin Otolaryngol 1984;9:105-109. 64. Simko TG, Griffin TW, Gerdes AJ, et al. The role of radiation therapy in the treatment of glomus jugulare tumors. Cancer 1978;42:104-106. 65. Gibbin KP, Henk JM. Glomus jugulare tumors in South Wales: a 20-year review. Clin Radiol 1978;29:607-609. 66. Arthur K. Radiotherapy in chemodectoma of the glomus jugulare. Clin Radiol 1977;28:415-417. 67. Tidwell TJ, Montague ED. Chemodectomas involving the temporal bone. Radiology 1975;116:147-149. 68. Silverstone SM. Radiation therapy of glomus jugulare tumors. Arch Otolaryngol 1973;97:43-48. 69. Maruyama Y. Radiotherapy of tympanojugular chemodectomas. Radiology 1972;105:659-663. 70. Foote RL, Pollock BE, Gorman DA, et al. Glomus jugulare tumor: tumor control and complications after stereotactic radiosurgery. Head Neck 2002;24:332-339. 71. Liscak A, Vladyka V, Wowra B, et al. Gamma knife radiosurgery of glomus jugulare tumour: a preliminary report—early multicentre experience. Acta Neurochir 1999;141:1141-1146.

ADDITIONAL READING Antoniades J. Uncommon Malignant Tumors. New York: Masson, 1982. Brown JS. Glomus jugulare tumors revisited: a 10-year statistical ­follow-up of 231 cases. Laryngoscope 1985;95:284-287. Burres SA, Wilner HL. Nonsurgical management of a large recurrent glomus jugulare tumor. Otolaryngol Head Neck Surg 1983;91:312-314.

Cannon CR, McLean WC. Adenoid cystic carcinoma of the middle ear and temporal bone. Otolaryngol Head Neck Surg 1983;91:96-98. Kabnick EM, Serchuk L. Metastatic chemodectoma. J Nat Med ­Assoc 1985;77:750-756. Lo WWM, Solti-Bohman LG. High-resolution CT in the evaluation of glomus tumors of the temporal bone. Radiology 1984;150:737-742. Lybeert MLM, Van Andel JG. Radiotherapy of paragangliomas. Clin Otolaryngol 1984;9:105-109. Mikhael MA, Wolff AP. Current concepts in neuroradiological diagnosis of acoustic neuromas. Laryngoscope 1987;97:471-476. Moss WT. Radiation Oncology: Rationale, Technique, Results, 6th ed. St Louis: CV Mosby, 1989. Olson LE, Cox JD. Chemodectomas. In Laramore GE (ed): Radiation Therapy of Head and Neck Cancer. Secaucus, NJ: SpringerVerlag, 1989, pp 171-179. Pallach JF, McDonald TJ. Adenocarcinoma and adenoma of the middle ear. Laryngoscope 1982;92:47-53. Pansky B. Review of Gross Anatomy, 5th ed. New York: Macmillan, 1984. Patel M, Cronin J: Primary adenocarcinoma of the middle ear. Ear Nose Throat J 1981;60:527-529. Perez CA, Brady LW (eds): Principles and Practice of Radiation Oncology Philadelphia: JB Lippincott, 1987. Phelps PD, Lloyd GA. Vascular masses in the middle ear. Clin Radiol 1986;37:359-364. Pritchett JW. Familial concurrence of carotid body tumor and pheochromocytoma. Cancer 1982;49:2578-2579. Reddy EK, Mansfield CM. Chemodectoma of glomus jugulare. Cancer 1983;52:337-340. Sataloff RT, Kemink JL. Total en bloc resection of the temporal bone and carotid artery for malignant tumors of the ear and temporal bone. Laryngoscope 1984;94:528-533. Schuller DE, Conley JJ. Primary adenocarcinoma of the middle ear. Otolaryngol Head Neck Surg 1983;91:280-283. Sinha PP, Aziz HI. Treatment of carcinoma of the middle ear. Radiology 1978;126:485-487. Strauss M, Nichols GG. Malignant catecholamine-secreting carotid body paraganglioma. Otolaryngol Head Neck Surg 1983;91:315-321. Thabet JH, Ali F. Unusual presentations of carotid body tumors. Ear Nose Throat J 1983;62:316-320. Tidwell TJ, Montague ED. Chemodectomas involving the temporal bone. Radiology 1975;116:147-149. Wang CC. Radiation Therapy for Head and Neck Neoplasms: Indications, Techniques and Results. 2nd ed. Chicago: Year Book Medical Publishers, 1990. Wang CC, Doppke K. Osteoradionecrosis of the temporal bone: consideration of nominal standard dose. Int J Radiat Oncol Biol Phys 1976;1:881-883.

The Thyroid 16 RICHARD W. TSANG AND JAMES D. BRIERLEY EFFECTS OF IRRADIATION ON THE NORMAL THYROID THYROID CANCER Incidence and Epidemiology Etiology and Causative Factors Pathologic Classification DIFFERENTIATED THYROID CANCER Clinical Manifestations and Evaluation Staging Prognostic Factors Principles of Management External Beam Radiation Therapy Local or Regional Recurrence of Thyroid Cancer Metastatic Disease Thyroglobulin-Positive, Scan-Negative Disease When Thyroid Cancer Does Not Take Up Iodine MEDULLARY THYROID CANCER Clinical Manifestations and Evaluation

Prognostic Factors Principles of Management Hypercalcitoninemia Without Clinically Detectable Metastasis Metastatic Disease ANAPLASTIC THYROID CARCINOMA Clinical Manifestations and Evaluation Prognostic Factors Principles of Management RARE TYPES OF THYROID CANCER RADIATION THERAPY TECHNIQUES Radioiodine External Beam Radiation Therapy Palliative Radiation Therapy for Metastatic Disease COMPLICATIONS OF RADIATION THERAPY Radioiodine External Beam Radiation Therapy

Effects of Irradiation on the Normal Thyroid The thyroid is incidentally irradiated during mantle radiation therapy and many head and neck treatment plans, which can result in thyroid dysfunction and neoplasia that is benign (e.g., goiter, nodular disease) or malignant (e.g., thyroid carcinoma). Sonographic examination of 47 patients treated in childhood for Hodgkin disease revealed that all had thyroid abnormalities; 95% had diffuse thyroid atrophy, 5% had a goiter, 42% had focal or multifocal abnormalities, and 5% had a carcinoma.1 The incidence of hypothyroidism after irradiation varies from 10% to 50%, depending on radiation dose, age at the time of irradiation, and length of follow-up.2-4 Patients who have had radiation therapy to the lower neck should have their thyroid­stimulating hormone (TSH) levels monitored regularly ­every 6 months for 5 years and annually thereafter.2 The risk of developing a second malignancy depends on age at the time of radiation exposure and the length of time since irradiation. The reported relative risk of developing thyroid cancer after irradiation for Hodgkin disease is 15.6 (95% confidence interval, 6.3 to 32.5).4 The tumors are usually well differentiated (the papillary type is more common than the follicular type, and anaplastic tumors are rare), and their behavior is similar to that of spontaneous thyroid cancer. Secondary thyroid cancer is often multifocal. Given the prolonged latency period from the time of radiation exposure to the development of thyroid dysfunction or malignancy, it is prudent to provide periodic ­clinical

examination and biochemical testing for patients who have undergone thyroid irradiation. Lesions suspected of harboring disease can be investigated by cytologic examination of fine-needle aspiration (FNA) samples. Radiation exposure is the predominant environmental factor linked to thyroid carcinogenesis.

Thyroid Cancer Thyroid diseases, including various inflammatory, auto­ immune, and benign neoplasms, are common in the ­general population, but cancer arising from the thyroid gland is rare, accounting for only 2.2% of all malignancies and 0.27% of cancer deaths in the United States.5 Thyroid cancer is important, however, because of its unique clinical features, the wide range in prognosis depending on its histology, the role of oncogenes in its pathogenesis, and the therapeutic effectiveness of radioactive iodine. Exposure to radiation is a significant etiologic factor in the development of the most common type of thyroid cancer, differentiated thyroid carcinoma. Differentiated thyroid cancer has a high cure rate; in contrast, anaplastic cancer commonly manifests as extensive localized disease and therefore poses a challenge of special interest to radiation oncologists. Anaplastic thyroid cancer (ATC) has such a poor prognosis that all cases are classified as stage IV by the American Joint Committee on Cancer (AJCC) and the International Union against Cancer (UICC).6,7 The management of thyroid cancer is primarily influenced by the histology of the tumor, the age of the patient,

333

334

PART 3  Head and Neck

and the extent of the disease. Because low-risk disease is characterized by a long natural history and a high survival rate, an important goal of therapy for low-risk disease is to minimize the toxicity of that therapy. Surgery, with or without radioiodine, remains the standard treatment for low-risk disease, and the involvement of experienced thyroid surgeons is essential for achieving the best results. High-risk disease often requires a coordinated multidisciplinary approach, with the participation of surgeons, endocrinologists, nuclear medicine physicians, and radiation oncologists, to optimally control the disease. Two caveats should be considered in discussions of the management of thyroid cancer. First, few randomized phase III studies have been done in thyroid oncology, and treatment decisions are therefore based on reports of historically treated patient populations without control groups. Second, because the prognosis for differentiated thyroid cancer is generally excellent, most retrospective reviews that include prolonged clinical follow-up are based on pathologic diagnoses made many years before. Changes in pathology classifications, as discussed later, make comparisons between previous and current classifications difficult. This chapter reviews the various types of thyroid cancer and their management, emphasizing the role of radioactive iodine and external beam radiation therapy (EBRT). The technical aspects of the delivery of radioiodine and EBRT and results of those treatments are discussed. Lymphomas of the thyroid gland are addressed in Chapter 34 rather than here, because the principles of their management are more in keeping with other non-Hodgkin lymphomas.

Incidence and Epidemiology The incidence of thyroid cancer in the United States has been increasing over the past 35 years. Virtually all of the 2.4-fold increase in thyroid cancer from 1973 to 2002 was attributable to the increase in papillary thyroid cancer, and 87% of the increase was attributable to tumors 2 cm in the greatest dimension or smaller. Despite this significant increase in incidence, mortality rates from thyroid cancer have remained stable because small papillary tumors are invariably curable. Although the cause of the increased incidence is uncertain, the fact that most tumors are small at diagnosis suggests improved detection of subclinical disease rather than any environmental cause.8 Thyroid cancer is more common in females, with a female-to-male ratio of 3:1, especially between the ages of 15 and 40 years, when the incidence among women peaks. Geographic variations are evident, with the highest incidence rate found in Hawaii (7.2 cases per 100,000 people during the period between 1993 and 1997). Papillary carcinoma is more common in iodine-rich areas, whereas follicular carcinomas predominate in iodine-deficient areas. Racial differences are also ­apparent, with lower incidence rates among blacks and Native Americans (2.9 cases per 100,000 people and 2.2 cases per 100,000 people, respectively, in 1990 to 1997).9,10

Etiology and Causative Factors Despite an increasing understanding of the genetic and phenotypic events underlying the development and progression of thyroid cancer, in most cases, the causative agent has not been identified. About 20% to 30% of cases of medullary thyroid cancer (MTC) are familial.11-13 Mutation in the germline proto-oncogene RET (formerly called

PTC) has been associated with familial MTC and multiple endocrine neoplasia (MEN) types 2A and 2B. The RET gene, located on chromosome 10q11.2, encodes a tyrosine kinase receptor. Mutation of codons 609, 611, 618, or 620 in exon 10 and, most commonly, codon 634 in exon 11 of the RET gene, usually by a single substitution of cysteine by another amino acid, is the causative event leading to constitutive activation of the tyrosine kinase receptor and giving rise to familial MTC and MEN2A.14,15 The classic MEN2A syndrome is associated with pheochromocytoma in 30% to 50% of cases and parathyroid diseases in 10% to 15%. Mutations of codon 918 are associated with MEN2B, a rare syndrome involving a marfanoid habitus in combination with ganglioneuromas of the mucosa and gastrointestinal tract. Somatic mutations of the RET proto-oncogene have been reported in 30% to 40% of cases of sporadic MTC.16-18 The importance of identifying the mutated gene in patients with MTC relates to the high penetrance of the gene and the need for genetic screening of family members; the lifetime risk of carriers developing MTC is approximately 90%. Treating identified carriers promptly with total thyroidectomy before the disease clinically manifests produces a cure rate of more than 90%.19,20 Although the papillary and follicular forms of thyroid cancer usually are not considered to be hereditary, several families have been documented to have a high incidence of the disease.21-23 Inherited syndromes associated with papillary thyroid carcinoma include Gardner’s syndrome (i.e., familial adenomatosis polyposis)24 and Cowden syndrome, which is characterized by multiple hamartomas and other benign or malignant lesions of the breast, ovary, and ­uterus.25 The main environmental factor linked to thyroid carcinogenesis is radiation exposure, especially in childhood.26 Until the early 1960s, radiation had been used to treat a variety of benign conditions (e.g., cervical tuberculous adenitis, acne, ringworm) in children, and a history of radiation exposure in childhood was not unusual. Since that time, the main iatrogenic cause of thyroid cancer is irradiation of the neck for childhood malignancies such as Hodgkin disease and, rarely, radioiodine for the treatment of thyrotoxicosis. The nuclear accident at Chernobyl, Ukraine, in 1986 attested to the powerful carcinogenic effects of radiation, particularly for young children.27,28 The incidence of childhood thyroid cancer in Belarus increased by 100-fold from 0.3 case per 100,000 in 1981 through 1985 to 30 cases per 100,000 in 1991 through 1994 after the Chernobyl accident in 1986.29 By 2065, the total number of excess thyroid cancers diagnosed in Europe because of this accident is expected to reach 16,000 cases.30 Radiation fallout after nuclear weapons testing on the Bikini atoll resulted in an increased relative risk of 4.5.26 Most radiation-induced thyroid cancer is of the papillary type and the latency period exceeds 10 years; however, the latency period in children, especially those exposed in the Chernobyl accident, can be much shorter.27 The association between low iodine intake and follicular carcinoma has suggested that iodine insufficiency results in chronic overproliferation of follicular cells. This over­ proliferation results in follicular adenomas, from which follicular carcinoma can develop. Mutations that affect RAS proteins, an important component of the signaling pathway

16  The Thyroid

for thyroid cell proliferation, have been found in 40% to 50% of follicular adenomas and carcinomas, suggesting that RAS mutations may be an early event in thyroid follicular tumorigenesis. In papillary thyroid cancer, RAS mutations are much less common, but gene rearrangements in RET are present in 20% to 50% of papillary thyroid cases and are even more common after radiation exposure.31-33 Mutations of the TP53 tumor-suppressor gene are also common features of ATC and may be a critical event that transforms a differentiated thyroid cancer into an undifferentiated one.34,35

Pathologic Classification The thyroid contains two main types of cells: follicular epithelial cells and C cells. Most cases of thyroid cancer develop from cells of follicular origin, and they are usually classified as differentiated (papillary and follicular carcinomas) or undifferentiated (anaplastic carcinoma). Insular carcinoma represents an intermediate point on the continuum between the differentiated and undifferentiated types.36 MTC originates from C cells and is clinically, pathologically, and etiologically distinct from differentiated and anaplastic cancer.

Papillary Carcinoma Papillary carcinoma manifests histologically as follicular differentiation with papillary or follicular architecture and characteristic nuclear features (Fig. 16-1), including enlargement, crowding, overlapping, and a hyperchromatic with finely dispersed chromatin (ground-glass) or a clear (Orphan Annie) appearance (Fig. 16-2). Extrafollicular psammoma bodies, caused by calcification arising from damaged papillae, are observed in about 50% of papillary tumors. Multicentricity, which may reflect multiple primary sites or lymphatic spread within the thyroid, is common. Immunohistochemical staining for thyroglobulin and cytokeratins is positive. The follicular variant of papillary carcinoma has a follicular architecture but has the nuclear features of papillary carcinoma, and its behavior is similar to that of pure papillary carcinoma.37 In the past, tumors with mostly follicular architecture have often been classified as follicular tumors; the change in classification has resulted in an apparent increase in the relative ratio of

FIGURE 16-1.  A papillary thyroid carcinoma contains papillary (right) and follicular (left) elements (×100).

335

papillary to follicular carcinomas.38 Vascular invasion, most commonly seen in follicular carcinomas, is associated with an increased incidence of disease recurrence and distant metastasis.39 In the tall-cell variant of papillary carcinoma, between 30% and 50% of cells are twice as high as they are wide. Tall-cell variants are more aggressive because patients usually present with a higher stage of disease and typically have more extensive local disease; but stage for stage, the prognosis associated with tall-cell disease seems to be similar to that for pure papillary carcinoma.40 A rare columnar cell variant has the distinctive feature of nuclear stratification, like tumors of the colon; columnar cell cases are more common in males than females and tend to be diagnosed at advanced stages.41,42 The diffuse sclerosing variant is characterized by diffuse involvement of one or both thyroid lobes with dense sclerosis and by scattered foci of papillary carcinoma. This type occurs in young patients, and although patients usually have advanced disease at diagnosis, the diffuse sclerosing variant is associated with a relatively low mortality rate.43

Follicular Neoplasms Follicular adenomas are common benign neoplasms of the thyroid. The adenoma cells resemble normal follicular cells; they stain positively for thyroglobulin and cytokeratins and are invariably encapsulated. Minimally invasive follicular carcinomas tend to have thicker capsules and are characterized by their invasion through the tumor capsule or into blood vessels. No cytologic features distinguish benign adenomas from minimally invasive carcinomas. In the past, many minimally invasive follicular carcinomas may have been misdiagnosed as benign neoplasms.38 In contrast, widely invasive follicular carcinomas are characterized by gross extension into the surrounding thyroid parenchyma, a poorly identified tumor capsule, necrosis, a high mitotic rate, and nuclear atypia (Fig. 16-3).

Hürthle Cell Tumors Hürthle probably described C cells; however, the name Hürthle (or Askanazy) is used to refer to oncocytic neoplasms, characterized by abundant granular cells rendered eosinophilic from numerous mitochondria. Hürthle cell

FIGURE 16-2.  Ground-glass or Orphan Annie nuclei are characteristic of papillary thyroid carcinoma in an area of follicular architecture (×250).

336

PART 3  Head and Neck

carcinomas, unlike Hürthle cell adenomas, show capsular or vascular invasion.44 Cases that exhibit papillary architecture throughout are classified as papillary Hürthle cell carcinomas.

Insular Carcinomas

FIGURE 16-3.  Follicular thyroid carcinoma has invaded the tumor capsule (×100).

Poorly differentiated or insular carcinomas are intermediate between well-differentiated and anaplastic carcinomas, and they represent part of a continuum from differentiated to undifferentiated (anaplastic) thyroid cancer.36,45 Insular carcinomas are composed of a population of homogenous small cells with round nuclei and areas of necrosis, and they stain positively for thyroglobulin and cytokeratins. Insular carcinoma is associated with poorer outcomes than differentiated thyroid cancer of similar size, and it rarely takes up radioiodine.46,47

Anaplastic Carcinomas ATCs are partially or completely composed of undifferentiated cells. The cells can have a sarcomatoid or spindle cell pattern or have a giant cell or squamoid pattern.48 All have high mitotic rates and often have areas of necrosis (Fig. 16-4A), lack of follicular formation, and intravascular tumor invasion. Positive staining for cytokeratins is typical, but positive staining for thyroglobulin is variable. Areas of well-differentiated tumor are occasionally present. The differential diagnosis includes thyroid lymphoma or MTC, and in the era before routine immunohistochemical staining, making an accurate diagnosis could be difficult.

Medullary Thyroid Carcinoma

A

The malignant cells in MTC are derived from C cells, the parafollicular cells of the thyroid gland. MTCs can exhibit a wide range of growth patterns; the most common is uniform nests of cells separated by fibrovascular stroma (Fig. 16-5). The appearance of the cells also varies widely. The most common is round or oval, but spindle cells are not uncommon. Amyloid, detected by Congo red staining, is present in at least 80% of specimens (see Fig. 16-5). The characteristic immunohistochemical findings are positive staining for calcitonin and carcinoembryonic antigen (CEA), but staining for cytokeratins and neuroendocrine markers (e.g., neuron-specific enolase) also can be positive.

B FIGURE 16-4.  A, An anaplastic thyroid cancer has spindle cells, bizarre giant cells, and mitoses (×250). B, Axial CT slice of an anaplastic thyroid carcinoma shows extensive extrathyroidal infiltration of neck tissues by tumor, with involvement of the laryngeal cartilage.

FIGURE 16-5.  Medullary thyroid carcinoma has uniform cells and large amounts of amyloid that stain with Congo red (×100).

16  The Thyroid

Differentiated Thyroid Cancer Clinical Manifestations and Evaluation Thyroid nodules are found at 50% of autopsies and in up to 70% of adults examined by ultrasonography, but the prevalence of palpable thyroid nodules in the United States ranges from 4% to 7%. Of these nodules, about 4% to 14% are malignant. Most thyroid nodules are asymptomatic. Local tumor growth can cause pain, and the presence of local compressive symptoms suggests extrathyroidal extension of disease. Hoarseness caused by recurrent laryngeal nerve paralysis is an important finding in more advanced cases. Rarely, carcinoma can present in a thyroglossal duct or cyst higher in the neck, usually close to the midline.49,50 Some patients present with cervical lymph node metastasis. The most commonly involved nodes are those in the central compartment (level XI) and the jugular chain (levels II, III, and IV). Occasionally, posterior cervical lymph nodes (level V) are affected, but level I nodes are rarely involved. Infrequently, the nodes are cystic. The most important diagnostic tool for a palpable thyroid nodule is cytologic analysis of FNA samples by experienced physicians and cytologists. Cytologic findings can be classified as benign, suspicious, malignant, or unsatisfactory or inadequate for diagnosis.51 Cellular yields can be higher when the FNA procedure is done under sonographic guidance. Because follicular carcinoma cannot be diagnosed cytologically, a suspicious or indeterminate finding in an FNA sample does not exclude a follicular carcinoma, and malignancy may be present in up to 30% of such cases.52 Although radioisotope scanning is highly sensitive, its specificity is low, and it is most useful as an adjunct rather than a primary diagnostic test. Radioisotope scanning can be useful in detecting a hot nodule (i.e., one that can take up radioiodine) if TSH levels are low. Sonographic examination of the thyroid can reveal nodules as small as 2 to 3 mm; although it cannot distinguish benign from malignant tumors, sonography is useful for aiding FNA biopsy of small lesions and for monitoring lesions after treatment.

Staging The UICC/AJCC staging classification for thyroid cancer6,7 (Box 16-1) is unique in that the classification depends on the age of the patient and the histology of the tumor. Because the prognosis of differentiated thyroid cancer for young patients is excellent, patients younger than 45 years are classified as having stage I disease unless distant metastasis is present, in which case the classification is stage II. The AJCC staging classification published in 2002 included changes in the definition of T1 and T2 categories and in the classification of extrathyroidal extension. Unfortunately for the sake of comparison across studies, many alternative staging and prognostic indices are used for thyroid cancer. The most common in North America are the AMES (age, metastasis [distant], extracapsular tumor, size)53 and the AGES (age, grade, extracapsular tumor, size), which subsequently evolved into the MACIS.54 The MACIS scale, based on a large series from the Mayo Clinic, may be the most valuable of the prognostic indices in terms of judging whether patients would benefit from adjuvant treatment with radioactive iodine. The MACIS system depends on metastasis (M), age (A), completeness of excision (C), invasion beyond the thyroid (I), and size (S) (Box 16-2).54

337

We and others55 analyzed 10 classification systems and found that the AJCC/UICC tumor-node-metastasis (TNM) classification was as effective in predicting outcome as the AGES and MACIS systems and better than the others.55 Because the TNM classification is universally available and widely accepted for other disease sites, we and the American Thyroid Association recommend that it be used for all reports of treatment and outcome in thyroid cancer.51

Prognostic Factors Although the cancer-specific survival rate for differentiated thyroid cancer is high (94% at 20 years in one series of 1353 patients56), several adverse prognostic factors have been identified.54,56-62 Age is the most important prognostic factor in determining survival, and older patients tend to present with additional high-risk factors. Other factors influencing cause-specific survival include tumor size, histologic grade, postoperative presence of macroscopic ­residual disease, and presence of distant metastases. For ­locoregional recurrence, the most significant risk factor is age, and other factors are tumor size, multifocality within the thyroid gland, lymph node involvement, surgical treatment consisting of a less-than-total thyroidectomy, and lack of use of radioiodine. Although recurrence rates are highest at extremes of age, survival rates with treatment are excellent for the young. Other factors reported as prognostic for recurrence include high thyroglobulin levels at the time of initial iodine therapy and within 12 months of initial iodine therapy,63 an aneuploid DNA profile (which may be prognostic in papillary but not follicular tumors), and perhaps the presence of RET rearrangements. RET mutations predicted aggressive behavior in radiation-induced papillary carcinomas in children after the Chernobyl disaster, but the reasons for its variable expression in sporadic cases and its role in influencing prognosis have yet to be fully determined. RAS mutations, CDKN1B (formerly designated p27 or Kip1) downregulation, ERBB2 expression, and NME1 (formerly designated NM23-H1) immunoreactivity may also be of prognostic value.61 Inactivating point mutations of TP53 are more common in poorly differentiated and anaplastic tumors than in differentiated tumors.

Principles of Management Many aspects of managing differentiated thyroid carcinoma are controversial. The minimum surgical treatment is thyroid lobectomy and isthmectomy,64 but many centers routinely perform total thyroidectomy for all patients ­after malignancy is confirmed by analysis of the FNA sample.52,65-67 If the findings from FNA and sonographic ­examination are not definitive for a malignancy or a benign lesion, a thyroid lobectomy and isthmectomy may be performed for diagnostic purposes; if malignancy is confirmed, a completion thyroidectomy can be done if indicated (i.e., if ablation with radioiodine is planned). The prognosis is excellent for patients younger than 45 years with tumors smaller than 4 cm, with virtually 100% cause-specific survival rates at 10 years after limited surgery without the use of radioiodine.60,65 However, a risk of locoregional recurrence remains, most commonly in cervical lymph nodes, for up to 10% of patients in whom disease is managed conservatively.68,69 This risk can be minimized by using as initial treatment total thyroid ablation: surgical total thyroidectomy followed by routine postoperative

338

PART 3  Head and Neck

BOX 16-1.  American Joint Committee on Cancer and the International Union against Cancer ­­ Tumor-Node-Metastasis Classification for Carcinoma of the Thyroid Gland Primary Tumor (T)* TX Primary tumor cannot be assessed T0 No evidence of primary tumor T1 Tumor 2 cm or less in the greatest dimension and limited to the thyroid T2 Tumor more than 2 cm but not more than 4 cm in the greatest dimension and limited to the thyroid T3 Tumor more than 4 cm in the greatest dimension and limited to the thyroid or T4a Tumor of any size extending beyond the thyroid capsule to invade subcutaneous soft tissues, larynx, trachea, esophagus, or recurrent laryngeal nerve T4b Tumor invades prevertebral fascia or encases carotid artery or mediastinal vessels Anaplastic Carcinoma† T4a Intrathyroidal anaplastic carcinoma—surgically resectable T4b

Intrathyroidal anaplastic carcinoma—surgically unresectable

Regional Lymph Nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Regional lymph node metastasis N1a Metastasis to level VI (pretracheal, paratracheal, and prelaryngeal/Delphian lymph nodes) N1b Metastasis to unilateral, bilateral, or contralateral cervical or superior mediastinal lymph nodes) Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis

Stage Grouping Papillary or Follicular Carcinoma Age < 45 Years Stage I Any T Any N M0 Stage II Any T Any N M1 Stage III

Stage IVA

Stage IVB Stage IVC Medullary Carcinoma Stage I Stage II Stage III

Stage IVA

Stage IVB Stage IVC Anaplastic Carcinoma‡ Stage IVA Stage IVB Stage IVC

Age ≥ 45 Years T1 N0 M0 T2 N0 M0 T3 N0 M0 T1 N1a M0 T2 N1a M0 T3 N1a M0 T4a N0 M0 T4a N1a M0 T1 N1b M0 T2 N1b M0 T3 N1b M0 T4a N1b M0 T4b Any N M0 Any T Any N M1 T1 T2 T3 T1 T2 T3 T4a T4a T1 T2 T3 T4a T4b Any T

N0 N0 N0 N1a N1a N1a N0 N1a N1b N1b N1b N1b Any N Any N

M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M1

T4a T4b Any T

Any N Any N Any N

M0 M0 M1

*All categories may be subdivided: (a) solitary tumor; (b) multifocal tumor (the largest determines the classification). †All anaplastic carcinomas are considered T4 tumors. ‡All anaplastic carcinomas are considered stage IV. From Greene FL, Page DL, Fleming, ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 77-87.

thyroid-remnant ablation with radioiodine.66,67 These two approaches—limited surgery without radioiodine and total thyroid ablation—have never been compared in a randomized trial. The concerns regarding the use of thyroidectomy for patients with low-risk disease are the small but not negligible risk of recurrent laryngeal nerve paralysis (1% to 4%, depending on surgical expertise) and the 5% to 10% risk of permanent hypoparathyroidism requiring lifelong calcium therapy. The risk associated with a single use of postoperative radioiodine is minimal; iodine 131 (131I) is typically given in doses of 1100 MBq (30 mCi) to 3700 MBq (100 mCi) for ablation, with higher doses used if there is evidence of persistent or metastatic disease. Successful ablation requires appropriate preparation, including thyroxine withdrawal (to achieve a high TSH ­level), a low-iodine diet, and a gamma camera scan (­before or after therapy) to document the location of abnormal areas of uptake and to screen for metastases (Fig. 16-6A). The availability of recombinant human TSH (rhTSH) ­allows administration of radioiodine without thyroxine

withdrawal, and several studies have confirmed the validity of ablation after rhTSH stimulation.70 There are three schools of thought about which patients require postoperative radioiodine ablation. One ­recommends total thyroidectomy and radioiodine ablation for all patients with tumors larger than 1 to 1.5 cm.56,67 The philosophy at the Mayo Clinic is that total thyroidectomy and TSH suppression without radioiodine is sufficient for patients with a MACIS score of less than 6.65 A third philosophy espoused by investigators at Memorial Sloan-Kettering is that lobectomy and isthmectomy is sufficient for patients with low-risk disease. Findings from a retrospective analysis of the use of radioiodine ablation for differentiated thyroid cancer64 suggested that such ablation might be beneficial in terms of reducing 10-year locoregional and distant recurrence-free rates, but the conclusion in that report was that the incremental benefit of remnant ablation for patients with low-risk disease was unclear. In the largest retrospective study of the management of differentiated thyroid cancer,65,71 investigators from the Mayo

16  The Thyroid

BOX 16-2.  The MACIS Scoring System for Carcinoma of the Thyroid Gland Characteristic Metastases Absent Present Age <40 years ≥40 years Extent of resection Complete Incomplete Extrathyroidal invasion Absent Present Tumor size

Score 0 3 3.1 0.08 × age 0 1 0 1 0.3 × size (in cm)

As an example, the MACIS score for a 59-year-old patient with a 4-cm papillary thyroid cancer with no extrathyroidal invasion and no metastases that was completely resected is < 6 (0 + 0.08 × 59 [4.72] + 0 + 0 + 0.3 × 4 [1.2] = 5.92). MACIS, metastasis, age, completeness of resection, invasion, and size. From Hay ID, Bergstralh EJ, Goellner JR, et al. Predicting outcome in papillary thyroid carcinoma: development of a reliable prognostic scoring system in a cohort of 1779 patients surgically treated at one institution during 1940 through 1989. Surgery 1993;114:1050-1057; discussion 1057-1058.

A

339

Clinic reported that radioiodine ablation was not beneficial in terms of reducing recurrence or mortality among 636 node-positive patients or among 527 node-negative patients with low-risk disease (defined as a MACIS score less than 6). This group did not recommend radioiodine ablation for patients with low-risk disease. Findings from a smaller series of patients seen at the Princess Margaret Hospital were similar.57,60 In our opinion, tumors smaller than 4 cm, especially in patients 45 years old or younger, are appropriately treated with total thyroidectomy and TSH suppression, and radioiodine ablation is not needed. The American Thyroid Association recommends that radioiodine ablation be used after total or near-total thyroidectomy for all patients with stage III or IV disease; for all patients younger than 45 years with stage II cancer (i.e., with distant metastatic disease); for most patients older than 45 years with stage II disease; and for selected patients with stage I cancer (especially those with ­multifocal ­disease, nodal involvement, extrathyroidal extension, vascular invasion, or aggressive histologies).51 After thyroidectomy, patients require lifelong thyroid replacement therapy to prevent hypothyroidism and to suppress TSH, which may stimulate tumor growth. Although suppressing TSH may improve survival,56,72 overaggressive suppression can lead to cardiac arrhythmias and osteopenia. In our practice, the aim is to suppress TSH to below-normal levels (in a sensitive assay) for patients with low-risk disease and to undetectable levels for those with

B

FIGURE 16-6.  A, Anterior-view radionuclide scan after total thyroidectomy, with the patient under hypothyroid conditions, was performed 7 days after the patient was given 5550 MBq (150 mCi) of 131I. Significant uptake is visible in both lungs and in the thyroid bed and upper mediastinum. Faint uptake is also visible in the submandibular salivary glands and urinary bladder. B, Axial CT scan through the lung of patient in A shows lung metastases.

340

PART 3  Head and Neck

high-risk disease. The level of suppression can be relaxed after 10 years of disease-free survival. Follow-up should consist of regular physical examinations; measurements of TSH, free triiodothyronine, and thyroglobulin; and sonography of the neck. After radioiodine therapy, we recommend that patients be scanned every 6 to 12 months until one scan is documented as being negative for iodine uptake. Others recommend scanning until two scans are negative73 or even scanning at regular intervals regardless of the number of negative scans obtained. The disadvantage of performing iodine scans as part of routine follow-up is the clinical hypothyroidism associated with thyroxine withdrawal, but this condition can be mitigated by using rhTSH and not stopping thyroxine. Regardless of whether thyroxine is withdrawn or rhTSH is used, thyroglobulin levels should be measured at the time of the scan. Some experts recommend followup with rhTSH and thyroglobulin measurement without radionuclide scanning.74 Thyroglobulin levels in the blood should be undetectable after complete thyroid ablation; a detectable or raised level, even in the absence of symptoms, indicates persistent or recurrent disease.74,75 Care is required in interpreting thyroglobulin levels, because the presence of thyroid antibodies interferes with some thyroglobulin assays, often producing a false-positive result.76 Some patients can have undetectable thyroglobulin levels despite the presence of disease,77 a finding that suggests that the tumor has dedifferentiated into a higher histologic grade or an anaplastic cancer.

who underwent EBRT for microscopic residual disease had a better 10-year actuarial local control rate than those who did not (93% versus 78%, P = .01).57 A follow-up to this review found a higher 10-year cause-specific mortality rate (81% versus 65%) and a higher locoregional relapse-free rate (86% versus 66%) among patients older than 60 years with extrathyroidal extension but no gross residual disease who had been treated with EBRT compared with those who had not.60 Many retrospective reviews have led to similar conclusions (Table 16-1).57,81-88 In some of these studies, not all patients were at high risk for recurrence in the thyroid bed or had had what would now be considered conventional treatment (i.e., total thyroidectomy and radioiodine) in addition to EBRT. One team from Germany studied 137 patients older than 40 years with thyroid cancer and extrathyroidal extension; all of these patients had had total thyroidectomy and radioiodine. The 85 patients who also received radiation therapy (50 to 60 Gy) to the thyroid bed and cervical and upper mediastinal lymph nodes had fewer local and regional recurrences than patients who had not had the radiation therapy (P = .004).81 Patients with postoperative gross residual disease or patients older than 45 years with extensive extrathyroidal extension (i.e., T4a or T4b lesions) and grossly complete resection are at high risk for local recurrence and may benefit from adjuvant EBRT after surgery and radioiodine treatment. Patients with minimal extrathyroidal extension, now defined as T3 disease, probably do not benefit from EBRT provided that the tumor is completely excised and that the patients undergo radioiodine therapy and TSH

External Beam Radiation Therapy Adjuvant Radiation to the Neck Most cases of differentiated thyroid cancer can be adequately treated with thyroidectomy and radioiodine. However, some locally invasive tumors do not concentrate radioiodine when they recur.78 Another approach, one involving EBRT, is needed to control local disease in patients judged to be at high risk for local recurrence, such as those with extrathyroidal extension of disease. Many investigators have reported the significance of extrathyroidal extension as a poor prognostic factor, especially among patients older than 45 years.39,79 Mazzaferri and Jhiang56 found that local tumor invasion into the neck structures (present at diagnosis in 8% of 1077 patients with papillary carcinoma and 12% of 278 patients with follicular carcinoma) was associated with higher 20-year recurrence rates (38% for those with tumor invasion versus 25% for those without, P < .001) and higher 20-year cancer mortality rates (20% for those with invasion versus 5% for those without, P = .001). Extrathyroidal extension at diagnosis or recurrence can be asymptomatic or can manifest as a neck mass, hoarseness, cough, dysphagia, hemoptysis, or dyspnea. A review of 262 patients with invasive papillary cancer found the site of invasion to be the muscle in 53% of cases, the laryngeal nerve in 47%, the trachea in 37%, the esophagus in 21%, and the larynx in 12%.80 Because radiation therapy for thyroid cancer has not been studied in randomized trials, its potential benefit can be assessed only from interpreting series of patients at adverse risk, comparing outcomes between patients who were treated with radiation therapy and those who were not. In one such study at the Princess Margaret Hospital, patients

TABLE 16-1.   Ten-Year Local Recurrence Rates after Adjuvant External Radiation Therapy for High-Risk Disease: Retrospective Studies Treatment

Study and Reference Tubiana et al, 198582 198883 *

Surgery with Radioactive Iodine (%)

Surgery and ­Radioactive Iodine with Radiation Therapy (%)

21

14

18

14

Phlips et al, 199385

21

3

Esik et al, 199484

60†

20†

50‡

10‡

Tsang et al, (papillary only)

22

7

Ford et al, 200386

63†,§

18†,§

Kim et al, (papillary only)

37.5‡,§

4.8‡,§

Keum et al, 200688

89

38

Simpson et al,

Farahati et al,

199681

199857

200387

*Local

recurrence rate after surgery alone was 74%. radioactive iodine; compared low-dose with higher-dose radiation therapy. ‡Includes distant failures. §Five-year local recurrence-free rate. †No

16  The Thyroid

suppression. Lymph node involvement by itself is not an ­indication for EBRT because regional control is usually achieved with initial neck dissection in combination with postoperative radioiodine. However, external irradiation is of value for patients with extracapsular extension of lymph node disease or recurrence after previous nodal dissection and radioiodine (discussed later). Radiation therapy, when recommended, should be given after the radioiodine to avoid the possibility of stunting (i.e., impairing the ability of the thyroid to take up radioiodine in subsequent therapy). However, the effect of stunting when external radiation is given before radioiodine has not been adequately studied.

Gross Residual Disease Grossly palpable disease occasionally remains in the thyroid bed after surgery. Radioiodine is unlikely to eradicate gross local disease, and patients with gross residual disease after thyroidectomy can benefit from planned combination therapy with radioiodine and external irradiation. In one series of 126 patients studied in Hong Kong, 10-year local control rates were significantly better among the 69 patients who had had EBRT than among those who had not (56% versus 24%, P = .002).89 Other investigators have reported disease control rates of 30% to 65%.2,57,82,90,91 The planned radiation therapy volume should cover the gross disease generously, using a dose of at least 60 Gy. The time to response can be as long as 8 to 12 months after completion of the radiation therapy.

Local or Regional Recurrence of Thyroid Cancer In most cases, differentiated thyroid cancer that recurs in the neck appears in the cervical lymph nodes (regional recurrence) rather than in the thyroid bed (local recurrence).57,92 Lymph node recurrence should be treated with neck dissection followed by additional radioiodine; this combination has been highly successful in tumor control.92 Occasionally, nodal disease recurs in the superior mediastinum, requiring dissection of the superior mediastinal nodes. In that case, radioiodine alone, without surgery, may be adequate if the tumor burden is low (e.g., nodes are smaller than 2 cm). EBRT is recommended only if the nodal disease persists or recurs despite a full modified neck dissection and radioiodine therapy. In this situation, the planned target volume should cover the bilateral cervical lymph nodes and the superior mediastinal lymph nodes to a total dose of 50 Gy, with consideration of a boost to the main areas of involvement to a further 10 to 16 Gy. Locally recurrent disease in the thyroid bed has a more ominous prognosis than does lymph node recurrence, largely because the disease is less likely to concentrate radioiodine.78 This condition usually occurs in older patients with T4 lesions at initial diagnosis that had not been adequately treated with radioiodine or adjuvant EBRT. Recurrence ­under these conditions is managed by a combination of surgical ­resection (if the disease is resectable), followed by postoperative radioiodine and EBRT. The prescribed radiation volume and dose is as described for the treatment of gross residual disease. Because survival can be prolonged even in the presence of metastatic disease,92 surgery may still be appropriate to prevent life-threatening complications from ­local disease, such as airway obstruction or hemorrhage.93

341

Metastatic Disease In a large French study of 2200 patients treated for differentiated thyroid cancer, 17.9% developed metastases to lung or bone at some stage of the disease; an additional 8 patients had metastases to the brain, 4 to the skin, and 3 to the liver.94 Survival of patients with metastatic disease depends on age, site of involvement, and tumor burden. Patients younger than 20 years have a 100% survival rate at 10 years, but survival among those older than 40 years falls to 20%. Metastases in older patients are less likely to concentrate iodine and therefore respond poorly to radioiodine therapy.94 Survival times are longest among patients with diffuse lung metastases that concentrate radioiodine (and can be seen on 131I imaging) but are too small to be seen on a chest radiograph.57,94,95 The mainstay of therapy is radioiodine, but this treatment is effective only if a total or neartotal thyroidectomy has been performed (see Fig. 16-6B). Failure to achieve complete remission in metastatic differentiated thyroid cancer can be predicted by failure to concentrate radioiodine, age older than 40 years, and the presence of bone metastases.94-96 In the large French study previously mentioned, patients with lung metastases had a complete response rate of 50% and a 10-year survival rate of 60%; however, the corresponding values for patients with bone metastases were only 10% and 20%.94 In view of this poor response to radioiodine and radiation therapy among patients with bone metastases, surgical resection of a solitary bone lesion may be appropriate.97 For patients expected to survive for long periods, the addition of highdose radiation therapy (e.g., 40 to 50 Gy in 1.8- to 2-Gy fractions) to radioiodine therapy can prolong local control. Even patients who present with distant metastatic disease can survive for prolonged periods. In one series of 44 such patients, the 20-year survival rate was 43%.96,98 Ideally, treatment should consist of total thyroidectomy to resect the primary lesion and to facilitate the use of ­radioiodine.

Thyroglobulin-Positive, Scan-Negative Disease When high thyroglobulin levels are present without evidence of disease on a radioisotope scan, other diagnostic tools can be used to locate the disease, including sonography, computed tomography (CT), magnetic resonance imaging, and 18F-fluorodeoxyglucose positron emission tomography (FDG-PET),with or without rhTSH stimulation.99,100 Measurable recurrence in the neck should be treated surgically. Because whole-body scanning after a therapeutic dose of radioactive iodine is more sensitive than diagnostic scanning, it has been argued that all patients with scan-negative, thyroglobulin-positive disease should be given a therapeutic dose of iodine. Others argue that if the diagnostic scan is negative, the uptake of a therapeutic dose will be too low to be of value. Negative findings on FDG-PET suggest that the residual or recurrent disease is metabolically inactive and the prognosis is therefore excellent.101

When Thyroid Cancer Does Not Take Up Iodine Little effective therapy is available for disease that has not responded to the combination of surgery, radioiodine, and EBRT. The most effective chemotherapy regimen is singleagent doxorubicin, which produces response rates of only

342

PART 3  Head and Neck

25% to 40%.102,103 Combination chemotherapy does not seem to be any more effective; in a small, randomized study that included advanced thyroid cancer of all histologic types, no difference in overall response rate was found when doxorubicin alone was compared with doxorubicin and cisplatin, although the complete remission rate was higher in the group given both drugs.102 Studies of newer biological agents, such as angiogenesis inhibitors, tyrosine kinase inhibitors, and proteosome inhibitors, have shown no evidence of their being any more effective than doxorubicin.

Medullary Thyroid Cancer

Prognostic Factors Survival among patients with familial MTC and MEN2A is better than that among patients with sporadic MTC, although this difference disappears when patients are matched for age, extent of tumor, and lymph node involvement.11,108,109 MEN2B manifests at earlier ages and in more advanced stages than does MEN2A and is associated with poorer survival.110 A multivariate analysis of 73 patients treated at the Princess Margaret Hospital revealed that extrathyroidal invasion and postoperative gross residual disease predicted a poor outcome.111

Clinical Manifestations and Evaluation

Principles of Management

Sporadic MTC usually manifests as a thyroid nodule. If cytologic analysis of an FNA sample yields a definitive diagnosis of MTC, a careful history and physical examination should be done, with particular attention to family history and any symptoms related to hypercalcitoninemia (e.g., watery diarrhea) or pheochromocytoma (e.g., episodic hypertension). Laboratory studies should include measurements of serum calcitonin and calcium, and catecholamines and their metabolites should be assayed in urine and serum to screen for pheochromocytoma. Blood lymphocytes should be tested for germline mutations of the RET oncogene. If biochemical test results suggest pheochromocytoma, abdominal imaging should be done and endocrinologists and anesthesiologists consulted to prepare the patient for surgical treatment of the pheochromocytoma first. Often, however, MTC is diagnosed only after definitive thyroid surgery from analysis of the histologic sections. In that case, the postoperative evaluation should include the same investigations as those described previously. If the basal calcitonin level is normal, a challenge test involving injection of pentagastrin (a gastric secretion stimulant) is a sensitive means of detecting residual disease.104,105 Pentagastrin is no longer available in North America; however, because the prognosis is generally excellent for patients with normal calcitonin levels after surgery, a watch-and-wait policy may be appropriate, with additional studies done if the calcitonin or CEA level rises. Significant numbers of patients have residual disease after surgical resection, as evidenced by high serum calcitonin and CEA levels or metastasis.100,106,107 Nodal involvement of the neck and superior mediastinum is relatively common, occurring in approximately 50% of cases, and it is often subclinical, with the only detected abnormality being an elevated serum calcitonin level.11,108 CT scans of the neck, chest, and abdomen should be obtained. A bone scan helps to rule out skeletal ­metastases. Scans with octreotide, a somatostatin analogue, are not particularly helpful unless therapy with octreotide is being considered. Distinguishing locoregional disease from distant metastasis is important because the prognoses and management strategies for the two conditions are ­different. Patients who show a germline mutation in the RET oncogene should be offered genetic counseling, with appropriate screening of first-degree relatives (i.e., siblings, parents, and children). The clinical probability that a RETmutation–positive person will develop MTC is very high (lifetime risk of more than 90%). Prophylactic thyroidectomy is recommended and has a high cure rate.13,19 Patients with MEN should also be screened regularly for pheochromocytoma and parathyroid hyperplasia.

The main treatment for MTC is surgical: total thyroidectomy and central neck dissection. Patients with node-positive disease should also undergo an ipsilateral modified neck dissection. Because up to 44% of patients with palpable unilateral nodes have contralateral nodal involvement, some investigators recommend that bilateral neck dissection should be part of the initial therapy.13,112 Radioiodine has no role in the management of MTC. Residual hypercalcitoninemia after thyroidectomy can be cured in some patients by meticulous nodal microdissection.107,112 Patients with gross or microscopic residual disease after surgery or extensive regional lymph node involvement are at high risk for recurrence; about 50% will experience recurrence. Postoperative adjuvant EBRT to the thyroid bed and regional nodal tissue can be considered for patients with high-risk disease. In one study, a radiation dose of 40 Gy given in 2-Gy fractions to lymph nodes in the neck and upper mediastinum and followed by a boost to the thyroid bed to a total dose of 50 Gy produced a locoregional control rate of 86% at 10 years.111 However, if nodal resection has been meticulous, EBRT may be of value only for patients with locally invasive disease. Although radiation treatment does not affect overall survival rates, locoregional control is important because relapse in the neck can negatively affect the patient’s quality of life.113-115 No role has been established for adjuvant chemotherapy.112 For patients with gross residual disease after surgery, the local control rate after radiation therapy is as low as 20%.111 Every attempt should be made to completely extirpate disease surgically, but when this is not possible, EBRT may result in long-term local control in a few patients.

Hypercalcitoninemia without Clinically Detectable Metastasis Even after adequate surgical therapy, high levels of calcitonin can persist with no clinical evidence of metastasis. Imaging with radionuclides such as ­metaiodobenzylguanidine or octreotide may reveal abnormalities and can help to ­direct the choice of therapy.112,116 Calcitonin and CEA levels correlate with tumor bulk; it is not unusual to see elevated but stable calcitonin or CEA levels for 5 to 10 years without obvious progression of disease.117 Chemotherapy for an asymptomatic patient is unnecessary even in the presence of measurable but nonprogressing disease. Diarrhea resulting from hypercalcitoninemia can be palliated with nonspecific drugs (e.g., diphenoxylate, loperamide) or a somatostatin analogue with or without interferon.112,118,119 A drop in the serum calcitonin level

16  The Thyroid

can indicate tumor dedifferentiation and herald a more aggressive clinical course.109

Metastatic Disease The most common sites of metastatic MTC are liver, lung, and bone. Treatment is palliative and includes supportive measures, analgesic drugs, and consideration of ­chemotherapy, hormonal therapy, and local radiation. ­Single-agent or combined therapy with dacarbazine, doxorubicin, cisplatin, or other drugs can produce response rates of 15% to 30%.13 Hormonal therapy with octreotide has been reported to lessen symptoms and reduce calcitonin levels but usually does not induce tumor shrinkage. A combination of octreotide and interferon-α was shown to improve response rates.13,112 Local radiation is best suited for symptomatic osseous metastases. Experimental therapies to target tumor cells by linking monoclonal antibodies (e.g., anti-CEA) to radionuclides (e.g., indium 111-­octreotide) and the use of tyrosine kinase inhibitors are being studied.112,120

Anaplastic Thyroid Carcinoma Clinical Manifestations and Evaluation ATC accounts for less than 2% of thyroid cancer cases.8,121 A review of the Surveillance, Epidemiology and End Results database revealed 516 patients diagnosed with ATC from 1973 to 2000: 7.5% with intrathyroidal ATC; 37.6% with extrathyroidal invasion or regional lymph node involvement, or both; and 43.0% with distant metastasis. The cancer-specific mortality rates at 6 and 12 months were 69.4% and 80.7%.122 The cardinal clinical features of ATC include a rapidly enlarging thyroid mass (typically over a few weeks) in an elderly person (see Fig. 16-4B). Some patients may have a history of goiter or thyroid mass consistent with prior undiagnosed differentiated thyroid cancer. Compressive symptoms of the upper aerodigestive tract are common in ATC. Some tumors invade and ulcerate through the skin of the neck. Metastases, typically in the lung, are found at diagnosis in 30% to 50% of cases.123 It is important to evaluate the integrity of the upper airway and the need for tracheostomy. Chest radiography or CT and bone scanning are useful for detecting metastatic disease.

Prognostic Factors All ATCs are classified as stage IV in the AJCC/UICC system.6,7 Many ATCs coexist with areas of differentiated thyroid carcinoma in the gland, suggesting that the anaplastic component arises from preexisting differentiated disease (i.e., dedifferentiation).124,125 An analysis by the thyroid cancer study group of the European Organisation for Research and Treatment of Cancer showed that the prognosis for patients with any element of anaplasia within a differentiated thyroid cancer was not different from that for patients with anaplastic tumor throughout, with a 1-year survival rate of only 10%. Nevertheless, the prognosis for patients with small tumors (<5 cm), no extrathyroidal extension, and no lymph node involvement is somewhat better.126 Occasionally, long-term survival has been observed among patients with completely resected tumors and extrathyroidal disease that extends only into the adjacent soft tissues.127

343

Principles of Management Complete surgical resection gives the best chance of cure, but unfortunately, this is possible in only a few cases.123,127- 129 Tumors that are most amenable to surgical resection are those that consist predominantly of differentiated thyroid carcinoma with only a small focus of ATC. Radical surgery is not warranted for patients with more extensive invasion of adjacent structures, because surgery has a significant negative effect on quality of life and does not seem to prolong survival. Surgery should be considered only when the extent of local disease allows thyroidectomy without undue morbidity.127 ATC in most patients is unresectable at diagnosis; the main role for surgery is biopsy to confirm the diagnosis and tracheostomy as needed to alleviate airway obstruction. Because ATC does not concentrate radioiodine, radiation is usually given in the form of EBRT with the goal of improving local control. Because the expected 5-year survival rate is only about 5% with any available treatment approach, the main goals of treatment are to achieve local control in the neck and to palliate any symptomatic metastatic disease to maintain quality of life. Local control remains a major problem even after EBRT; however, even patients in whom local control is achieved still die of distant metastatic disease. In one series of 134 cases, 98% had extrathyroidal extension at diagnosis, and the median survival time was only 3 months; the addition of EBRT extended the median survival time only to 5 months (P = .08).123 In another series, predictors of improved survival were found to be age younger than 75 years, absence of compressive symptoms, tumor smaller than 5 cm, and complete surgical resection.129 The ineffectiveness of radiation therapy alone for ATC prompted the development of novel fractionation schedules and concurrent chemotherapy regimens. Because ATC grows quickly, hyperfractionated and accelerated radiation schedules have been used, sometimes in combination with chemotherapy and surgical resection. The most successful regimen for obtaining local control, described by a Swedish group,130 involves combined preoperative hyperfractionated radiation, chemotherapy, and subsequent surgical resection. Although the local control rate in the study was 60%, the survival rate at 2 years was only 9%. In another study, the median survival time for a highly selected group of patients with little or no residual disease after resection treated with postoperative combined chemoradiation was 43 months.131 Even when local control is achieved, the metastatic rate for ATC remains high. Theoretically, giving ­chemotherapy beyond the course of radiation may improve survival if it eliminates micrometastases. Unfortunately, no known chemotherapy is effective against this type of cancer. ­Schlumberger and colleagues gave doxorubicin every 4 weeks for up to nine courses in addition to radiation, and they achieved only 15% survival at 20 months.132 At the Princess Margaret Hospital, patients with good performance status are prescribed a course of accelerated hyperfractionated radiation (60 Gy given in 40 fractions, 1.5 Gy/fraction, twice daily over 4 weeks) without chemotherapy. In the rare cases in which surgical resection is possible, the resection is done before the radiation therapy. To avoid the significant toxicity that others have encountered in

344

PART 3  Head and Neck

­ sing hyperfractionated radiation, no attempt is made to inu clude all of the regional lymph nodes. For patients with poor performance status or known metastases, radiation is given with palliative intent at 20 Gy in 5 fractions, with the option of a second course 4 weeks later for those who respond to this treatment. Because the prognosis of ATC is extremely poor and because current therapies are ineffective, innovative approaches are needed for the treatment of this disease.

Rare Types of Thyroid Cancer Rare primary tumors arising from the thyroid gland include squamous cell carcinomas, sarcomas, and germ cell tumors. Lymphomas are seen occasionally, typically in elderly women with a history of Hashimoto’s thyroiditis. The discovery of a rare tumor type involving the thyroid gland requires an expert pathology review to rule out tumors of nonthyroidal origin that may have infiltrated the thyroid gland (e.g., primary squamous cell carcinoma of the trachea or esophagus). Parathyroid carcinomas are seen occasionally, although they account for less than 1% of parathyroid neoplasms. Invasion into neck tissues or the thyroid gland, together with hypercalcemia from elevated parathyroid hormone production, should raise suspicion of a parathyroid carcinoma. For patients with parathyroid carcinoma and extensive local disease, a combination of surgery and postoperative local radiation therapy generally produces a favorable outcome.133 Primary squamous cell carcinoma of the thyroid, although equally uncommon, is highly malignant, with extensive locoregional invasion and a moderately high rate of metastatic spread. Treatment for this disease should consist of thyroidectomy with surgical extirpation of all gross disease when possible, followed by postoperative EBRT.134 Patients with unresectable disease can be treated with palliative radiation. Despite aggressive management, 70% to 80% of patients will succumb to the disease. Thyroidal lymphomas and sarcomas are treated according to the principles established for each histologic type. They are discussed in Chapters 34 and 36.

Radiation Therapy Techniques Radioiodine Radioactive iodine can be given as adjuvant treatment to ablate thyroid remnants after surgery or to treat residual, recurrent, or metastatic differentiated thyroid cancer. ­Ablative radioiodine can be given postoperatively (1) to ablate residual thyroid tissue that may contain microfoci of carcinoma (i.e., to treat multifocal disease); (2) to ablate residual thyroid tissue to remove the source of thyroglobulin so that the serum thyroglobulin assay can be used as a marker of thyroid cancer in follow-up (i.e., any rise in ­thyroglobulin can be assumed to be of malignant origin); and (3) to enable the detection of micrometastasis with iodine scanning that would otherwise be masked by absorption of radioiodine by the thyroid remnant. Many retrospective studies have indicated that postoperative iodine is beneficial in reducing the recurrence rate and improving the survival rate, particularly for patients with high-risk disease. No consensus exists about the best way of preparing patients for radioiodine therapy or the optimal radioactivity level to be given in such therapy. The approach at the

Princess Margaret Hospital is as follows. Although thyroid remnants can be ablated after a less-than-total or near-­total thyroidectomy, the success rate for this procedure is low (approximately 60%),135 and it can be associated with acute toxicity. We therefore recommend a completion thyroidectomy in the event of a previous thyroid lobectomy. The serum TSH level should be above 30 μU/mL to maximize the uptake of radioiodine by the remnant thyroid tissue. To achieve this level, thyroxine is stopped for 5 weeks and l-­triiodothyronine (liothyronine) is given for the first 3 weeks and then discontinued. The TSH level peaks at 2 weeks after cessation of the triiodothyronine. The resulting clinical hypothyroidism can be significant, and about 50% of patients may have to stop work for this reason.136 Reducing the length of time of thyroxine and triiodothyronine withdrawal, especially in elderly patients with cardiac disease, can lessen the effect. A preferable alternative, if available, is the use of rhTSH, which allows patients to continue taking their normal dose of thyroxine.70 Because consuming a low-iodine diet can increase 24-hour iodine uptake, a simplified low-iodine diet for 1 week before radioiodine therapy is recommended. However, many patients find the diet unpleasant, and some physicians do not routinely recommend a low-iodine diet. The use of iodinated contrast media in diagnostic imaging should be avoided for at least 2 months before radioiodine therapy. Medications containing high amounts of iodine such as amiodarone can also prevent successful iodine therapy. While the patient is under hypothyroid conditions or under rhTSH stimulation, a diagnostic scan (with 74 or 185 MBq [2 or 5 mCi] of 131I) can be useful for quantifying the radioiodine uptake in the neck before iodine therapy to assess the extent of thyroid remnants and the presence of any metastatic disease. The scan findings and extent of uptake are used to prescribe the dose of therapeutic iodine required. This policy results in an ­effective rate of ablation of 80%.137 However, for simplicity or because of concern about stunting (as experienced by a small proportion of patients exposed to 185 to 1100 MBq [5 to 30 mCi] of radioiodine138), some physicians use an empiric approach, prescribing a fixed amount of radio­iodine between 1100 MBq and 3700 MBq (30 to 100 mCi) without obtaining a scan before therapy. After radioiodine therapy, the usual practice is to scan 7 to 10 days later to locate the abnormal areas of radioiodine uptake. To radiation oncologists, it is more attractive to prescribe an absorbed dose to the target than to prescribe an empiric dose of radioiodine, but this practice requires ­estimating the mass of remnant thyroid tissue and the effective halflife of the retained radioactivity. The maximum tolerated activity is limited by the total blood dose, which should not exceed 2 Gy. Detailed dosimetric studies have de­ monstrated that 300 Gy resulted in ablation of 80% of remnants, and 100 ± 20 Gy resulted in 80% complete response of nodal or soft tissue metastases.139,140 However, no comparative studies have been done to verify the superiority of this approach in terms of improving recurrence rate or survival. The need to admit patients to a hospital facility depends on the recommendations of the local regulatory body. Although many centers require that patients given doses of 1100 MBq (30 mCi) or more be admitted until the retained activity drops below 1100 MBq, changes in the U.S. Nuclear Regulatory Commission Guidelines allow much larger

16  The Thyroid

345

a­ ctivity levels to be administered on an outpatient basis.141 This practice requires careful measurement and documentation of neck and whole-body retention, effective half-life, patients’ living conditions, and time spent with other people.

External Beam Radiation Therapy The main consideration regarding EBRT techniques for thyroid cancer is deciding whether to treat the thyroid bed alone or the thyroid bed plus the bilateral, cervical, and superior mediastinal lymph nodes. The thyroid bed volume curves around the vertebral body and includes the air column in the trachea. This U-shaped volume presents a challenge for adequately treating the thyroid bed while sparing the spinal cord. Before the era of intensity-­modulated radiation therapy (IMRT), many techniques were described to overcome these difficulties, but all required some degree of compromise.142 IMRT, however, is undoubtedly the best technique in terms of adequately covering the planned target volume and minimizing the dose to the spinal cord.143,144

A

Thyroid Bed Alone If only the thyroid bed is to be treated, the target volume extends from the hyoid superiorly to just below the suprasternal notch. A planning CT scan is obtained with the patient wearing an immobilization mask and in a neck-­extended position, and the scan is used to define the clinical target volume. The clinical target volume should cover the thyroid bed and the jugular and posterior cervical lymph nodes within the limits defined above (including level III, IV, VI, and partially level V nodal regions), with the volume adjusted according to the surgical and pathology findings. The IMRT plan is then generated to spare the spinal cord dose to the greatest extent possible. In the past, the characteristic rapid fall-off of a direct electron beam and the effect of the neck contour could produce an acceptable high-dose volume if 12- to 20-MeV beams or a mixed beam of two energies was used.57 An alternative technique was to use two anterolateral oblique wedged fields such as those used for laryngeal cancer,142 with the posterior border placed to exclude the spinal cord. The advantage of this technique over a single anterior electron beam is additional skin sparing, but it comes at the cost of underdosing the posterior lateral part of the volume. In the absence of gross disease, 40 to 50 Gy, given in 15 to 25 fractions over 3 to 5 weeks, is prescribed. If gross disease is present, a higher dose of 60 to 66 Gy is given in 30 to 33 fractions over 6 to 6.5 weeks.57,89

Thyroid Bed with the Cervical and Superior Mediastinal Lymph Nodes Large-volume treatment involving the cervical and superior mediastinal lymph nodes and the thyroid bed is typical for MTC and for differentiated thyroid cancer that persists after neck dissection and radioiodine therapy or has extensive nodal extracapsular extension. Such treatment also applies to ATC if regional nodes are involved in addition to the local thyroid disease. In the past, this technique involved at least two phases. In the first phase, the initial volume included the regional lymph nodes from the mastoid tip to the carina, including the thyroid bed, and consisted of parallel-opposed anteroposteriorposteroanterior fields treated to 40 to 46 Gy. Patients were treated while supine with the neck extended to

B FIGURE 16-7.  A, Axial CT scan shows a large thyroid mass deviating the trachea and esophagus in a 62-year-old man. No evidence of extrathyroidal extension was apparent on imaging, but at surgery, tumor found around the trachea was shaved off. Pathologic evaluation confirmed a papillary carcinoma with extrathyroidal extension. B, Intensity-modulated radiation therapy plan is shown for the patient in A. A dose of 60 Gy was delivered in 30 fractions to the thyroid bed and adjacent lymph nodes.

avoid the parotid glands and the oral cavity. The second phase involved a boost dose to a smaller volume to eliminate residual disease. IMRT has now become the preferred technique. The clinical target volume is defined to cover the thyroid bed and lymph nodes, including levels II, III, IV, V, and VI and the superior mediastinum (Fig. 16-7). The phase 1 dose is typically 50 Gy for microscopic disease, and the phase 2 reduced volume (shorter superoinferior extent) for gross or high-risk disease is treated to another 10 to 16 Gy. For single-phase IMRT, the dose to the gross tumor volume is typically 66 Gy, and the dose to the clinical target volume is 56 Gy, given in 33 fractions over 6.5 weeks.

346

PART 3  Head and Neck

Palliative Radiation Therapy for Metastatic Disease The principal sites of metastatic disease are bone and lung, with occasional involvement of soft tissues and lymph nodes. Liver metastasis is common in MTC. In general, for differentiated thyroid cancer, radioiodine should be considered first, and radiation therapy should be used only for disease in which iodine cannot be concentrated or is ineffective. The main role of EBRT is to palliate painful bone metastases. It is important to establish the goal of palliative therapy, particularly with regard to distinguishing whether it is offered for local control or purely to alleviate ­symptoms. Because the natural history of thyroid cancer is long, patients can live for many years despite the presence of metastatic disease. Patients for whom maximal control of the metastatic disease site is desirable should be offered a radical course of therapy to doses of 40 to 50 Gy. It is also important to consider the contributing dose from cumulative radioiodine therapy, particularly for vertebral lesions in direct contact with the spinal cord. Additional management considerations include the possibility of surgical resection of solitary metastases or resection and internal fixation of bones at risk of pathologic fracture. Pure symptom relief for patients whose life expectancy is short can be achieved with 8 Gy given as a single dose, 20 Gy given in 5 fractions, or 30 Gy given in 10 fractions, as is commonly used for other types of metastatic disease. Bisphosphonates may also be of value for patients with widespread bone metastases.

s­ tillbirths, or congenital defects were no higher than those among women given thyroid surgery without iodine therapy.148 Radiation pneumonitis or fibrosis is rare, having been reported in less than 2% of patients treated for lung metastases. Collectively, these risks make it important to limit the use of radioiodine to patients who will derive benefit from it.

External Beam Radiation Therapy Well-planned EBRT produces acceptable levels of acute toxicity, rarely produces serious complications, and does not preclude future surgical intervention. During the course of radiation therapy, moderate skin erythema develops, with higher doses causing dry or, rarely, moist desquamation. Mucositis of the esophagus, trachea, and larynx can occur toward the end of radiation and may require consumption of a soft diet and use of analgesics. Depending on the superior extent of the fields, changes in taste sensation and xerostomia are possible. Late toxicity is uncommon, consisting mostly of skin telangiectasias, skin pigmentation, soft tissue fibrosis, and mild lymphedema, usually appearing just below the chin. Esophageal or tracheal stenosis is rare. Tsang and colleagues reported no grade IV toxic effects (according to the Radiation Therapy Oncology Group scale) among patients given doses of 40 to 50 Gy in 2- to 2.5-Gy fractions to the neck and superior mediastinum.57 Farahati and colleagues observed no irreversible late toxic effects among 99 patients given doses of 50 to 60 Gy in 1.8- to 2.0-Gy fractions to a large volume.81

REFERENCES

Complications of Radiation Therapy Radioiodine Nausea, headache, and vomiting resulting from radiation sickness occur in 60% to 70% of patients given 7400 MBq (200 mCi) of radioiodine. Parotid swelling, altered sense of taste, and dry mouth can occur but are rare after doses of 3700 MBq (100 mCi) or less. Reversible bone marrow depression is most severe at 6 weeks. At cumulative activity levels of 18,500 to 37,000 MBq (500 to 1000 mCi), the incidence of chronic bone marrow depression is less than 1.5%; at cumulative activity levels exceeding 1000 mCi, the incidence rises to 25% or more.145 Leukemia has been associated with cumulative doses of more than 29,400 MBq (800 mCi), and an increased risk of breast and bladder cancer has been reported after ­average cumulative doses of about 37,000 MBq (1000 mCi).146 A large, pooled analysis of three European cohorts confirmed a 30% increase in the risk of a second malignancy among 6481 patients with thyroid cancer, with a linear dose-­response relationship between 131I administration for all cancers combined and for leukemia.147 That study also ­reported a strong relationship between cumulative 131I activity and cancers of the bone and soft tissue, colorectum, and salivary glands. Men can experience reversible or irreversible gonadal failure, especially after higher cumulative doses. In a study of 2113 pregnancies among women previously given radioiodine, the rates of miscarriages or preterm births,

  1. Shafford E, Kingston J, Healy J, et al. Thyroid nodular disease after radiotherapy to the neck for childhood Hodgkin’s disease. Br J Cancer 1999;80:808-814.   2. Garcia-Serra A, Amdur RJ, Morris CG, et al. Thyroid function should be monitored following radiotherapy to the low neck. Am J Clin Oncol 2005;28:255-258.   3. Grande C. Hypothyroidism following radiotherapy for head and neck cancer: multivariate analysis of risk factors. Radiother Oncol 1992;25:31-36.   4. Hancock SL, McDougall IR, Constine LS. Thyroid abnormalities after therapeutic radiation. Int J Radiat Oncol Biol Phys 1995;31:1165-1170.   5. Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2006. CA Cancer J Clin 2006;56:106-130.   6. Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer-Verlag, 2002.   7 Sobin LH, Wittekind C (eds). TNM Classification of Malignant Tumours, 6th ed. New York: John Wiley & Sons, 2002.   8. Davies L, Welch HG. Increasing incidence of thyroid cancer in the United States, 1973-2002. JAMA 2006;295:2164-2167.   9. Ries LAG, Eisner MP, Kosary CL, et al, (eds). SEER Statistics Review, 1973-1997. Bethesda, MD: National Cancer Institute, 2000. 10. Franceschi S, Boyle P, Maisonneuve P, et al. The epidemiology of thyroid carcinoma. Crit Rev Oncog 1993;4:25-52. 11. Raue F. German medullary thyroid carcinoma/multiple endocrine neoplasia registry. German MTC/MEN Study Group. Medullary thyroid carcinoma/multiple endocrine neoplasia type 2. Langenbecks Arch Surg 1998;383:334-336. 12. Bergholm U, Bergstrom R, Ekbom A. Long-term follow-up of patients with medullary carcinoma of the thyroid. Cancer 1997;79:132-138. 13. Randolph GW, Maniar D. Medullary carcinoma of the thyroid. Cancer Control 2000;7:253-261.

16  The Thyroid 14. Eng C. Seminars in medicine of the Beth Israel Hospital, Boston. The RET proto-oncogene in multiple endocrine neoplasia type 2 and Hirschsprung’s disease. N Engl J Med 1996;335:943–951. 15. Eng C. RET proto-oncogene in the development of human cancer. J Clin Oncol 1999;17:380-393. 16. Mulligan LM, Eng C, Healey CS, et al. Specific mutations of the RET proto-oncogene are related to disease phenotype in MEN 2A and FMTC. Nat Genet 1994;6:70-74. 17. Mulligan LM, Kwok JB, Healey CS, et al. Germ-line mutations of the RET proto-oncogene in multiple endocrine neoplasia type 2A. Nature 1993;363:458-460. 18. Eng C, Clayton D, Schuffenecker I, et al. The relationship between specific RET proto-oncogene mutations and disease phenotype in multiple endocrine neoplasia type 2. JAMA 1996;276:1575-1579. 19. Wells SA, Chi DD, Toshima K, et al. Predictive DNA testing and prophylactic thyroidectomy in patients at risk for multiple endocrine neoplasia type 2A. Ann Surg 1994;220:237-250. 20. Modigliani E, Cohen R, Campos JM, et al. Prognostic factors for survival and for biochemical cure in medullary thyroid carcinoma: results in 899 patients. The GETC Study Group. Groupe d’etude des tumeurs a calcitonine. Clin Endocrinol (Oxf) 1998;48:265-273. 21. Lupoli G, Vitale G, Caraglia M, et al. Familial papillary thyroid microcarcinoma: a new clinical entity. Lancet 1999;353:637-639. 22. Malchoff CD, Malchoff DM. Familial nonmedullary thyroid carcinoma. Semin Surg Oncol 1999;16:16-18. 23. Hemminki K, Dong C. Familial relationships in thyroid cancer by histo-pathological type. Int J Cancer 2000;85:201-205. 24. Cetta F, Toti P, Petracci M, et al. Thyroid carcinoma associated with familial adenomatous polyposis. Histopathology 1997;31:231-236. 25. Lynch ED, Ostermeyer EA, Lee MK, et al. Inherited mutations in PTEN that are associated with breast cancer, Cowden disease, and juvenile polyposis. Am J Hum Genet 1997;61:1254-1260. 26. Shore RE. Issues and epidemiological evidence regarding ­radiation-induced thyroid cancer. Radiat Res 1992;131:98-111. 27. Astakhova LN, Anspaugh LR, Beebe GW, et al. Chernobyl­related thyroid cancer in children of Belarus: a case-control study. Radiat Res 1998;150:349-356. 28. Thomas GA, Bunnell H, Cook HA, et al. High prevalence of RET/PTC rearrangements in Ukrainian and Belarussian postChernobyl thyroid papillary carcinomas: a strong correlation between RET/PTC3 and the solid-follicular variant. J Clin Endocrinol Metab 1999;84:4232-4238. 29. Stsjazhko VA, Tsyb AF, Tronko ND, et al. Childhood thyroid cancer since accident at Chernobyl. BMJ 1995;310:801. 30. Cardis E, Kesminiene A, Ivanov V, et al. Risk of thyroid cancer after exposure to 131I in childhood. J Natl Cancer Inst 2005;97:724-732. 31. Lemoine NR, Mayall ES, Wyllie FS, et al. High frequency of ras oncogene activation in all stages of human thyroid tumorigenesis. Oncogene 1989;4:159-164. 32. Moretti F, Nanni S, Pontecorvi A. Molecular pathogenesis of thyroid nodules and cancer. Baillieres Best Pract Res Clin Endocrinol Metab 2000;14:517-539. 33. Kondo T, Ezzat S, Asa SL. Pathogenetic mechanisms in thyroid follicular-cell neoplasia. Nat Rev Cancer 2006;6:292-306. 34. Fagin J, Matsuo K, Karmakar A, et al. High prevalence of mutations of the p53 gene in poorly differentiated human thyroid carcinomas. J Clin Invest 1993;91:179-184. 35. Zedenius J, Larsson C, Wallin G, et al. Alterations of p53 and expression of WAF1/p21 in human thyroid tumors. Thyroid 1996;6:1-9. 36. van den Brekel MW, Hekkenberg RJ, Asa SL, et al. Prognostic features in tall cell papillary carcinoma and insular thyroid carcinoma. Laryngoscope 1997;107:254-259. 37. Al-Brahim N, Asa SL. Papillary thyroid carcinoma: an overview. Arch Pathol Lab Med 2006;130:1057-1062. 38. LiVolsi V, Asa S. The demise of follicular carcinoma of the thyroid. Thyroid 1994;4:233-236. 39. LiVolsi VA, Fadda G, Baloch ZW. Prognostic factors in well­differentiated thyroid cancer. Rays 2000;25:163-175.

347

40. Johnson TL, Lloyd RV, Thompson NW, et al. Prognostic implications of the tall cell variant of papillary thyroid carcinoma. Am J Surg Pathol 1988;12:22-27. 41. Evans H. Columnar-cell carcinoma of the thyroid. A report of two cases of an aggressive variant of thyroid carcinoma. Am J Clin Pathol 1986;85:77-80. 42. Sobrinho-Simoes M, Nesland JM, Johannessen JV. Columnarcell carcinoma. Another variant of poorly differentiated carcinoma of the thyroid. Am J Clin Pathol 1988;89:264-267. 43. Carcangiu ML, Bianchi S. Diffuse sclerosing variant of papillary thyroid carcinoma. Clinicopathologic study of 15 cases. Am J Surg Pathol 1989;13:1041-1049. 44. Carcangiu ML, Bianchi S, Savino D, et al. Follicular Hurthle cell tumors of the thyroid gland. Cancer 1991;68:1944-1953. 45. Rodriguez JM, Parrilla P, Moreno A, et al. Insular carcinoma: an infrequent subtype of thyroid cancer. J Am Coll Surg 1998;187:503-508. 46. Pellegriti G, Giuffrida D, Scollo C, et al. Long-term outcome of patients with insular carcinoma of the thyroid: the insular histotype is an independent predictor of poor prognosis. Cancer 2002;95:2076-2085. 47. Volante M, Landolfi S, Chiusa L, et al. Poorly differentiated carcinomas of the thyroid with trabecular, insular, and solid patterns: a clinicopathologic study of 183 patients. Cancer 2004;100:950-957. 48. Carcangiu ML, Steeper T, Zampi G, et al. Anaplastic thyroid carcinoma. A study of 70 cases. Am J Clin Pathol 1985;83:135158. 49. Fernandez JF, Ordonez NG, Schultz PN, et al. Thyroglossal duct carcinoma. Surgery 1991;110:928-934; discussion 934-925. 50. O’ Connell M, Grixti M, Harmer C. Thyroglossal duct carcinoma: presentation and management, including eight cases reports. Clin Oncol (R Coll Radiol) 1998;10:186-190. 51. Cooper DS, Doherty GM, Haugen BR, et al. Management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 2006;16:109-142. 52. Yeh MW, Demircan O, Ituarte P, et al. False-negative fine­needle aspiration cytology results delay treatment and adversely affect outcome in patients with thyroid carcinoma. Thyroid 2004;14:207-215. 53. Cady B, Rossi R. An expanded view of risk-group definition in differentiated thyroid carcinoma. Surgery 1988;104:947-953. 54. Hay ID, Bergstralh EJ, Goellner JR, et al. Predicting outcome in papillary thyroid carcinoma: development of a reliable prognostic scoring system in a cohort of 1779 patients surgically treated at one institution during 1940 through 1989. Surgery 1993; 114:1050-1057; discussion 1057-1058. 55. Brierley JD, Panzarella T, Tsang RW, et al. A comparison of different staging systems predictability of patient outcome. Thyroid carcinoma as an example. Cancer 1997;79:2414-2423. 56. Mazzaferri EL, Jhiang SM. Long-term impact of initial surgical and medical therapy on papillary and follicular thyroid cancer. Am J Med 1994;97:418-428. 57. Tsang RW, Brierley JD, Simpson WJ, et al. The effects of surgery, radioiodine and external radiation therapy on the clinical outcome of patients with differentiated thyroid cancer. Cancer 1998;82:375-388. 58. Cady B. Papillary carcinoma of the thyroid gland: treatment based on risk group definition. Surg Oncol Clin N Am 1998;7:633-644. 59. DeGroot LJ, Kaplan EL, Straus FH, et al. Does the method of management of papillary thyroid carcinoma make a difference in outcome? World J Surg 1994;18:123-130. 60. Brierley J, Tsang R, Panzarella T, et al. Prognostic factors and the effect of treatment with radioactive iodine and external beam radiation on patients with differentiated thyroid cancer seen at a single institution over 40 years. Clin Endocrinol (Oxf) 2005;63:418-427. 61. Baloch ZW, LiVolsi VA. Prognostic factors in well-differentiated follicular-derived carcinoma and medullary thyroid carcinoma. Thyroid 2001;11:637-645. 62. Brierley JD, Asa SL. Thyroid Cancer. In Gospodarowicz MK, O’Sullivan B, Sobin LH (eds): Prognostic Factors in Cancer. Hoboken, NJ: John Wiley & Sons, 2006.

348

PART 3  Head and Neck

63. Toubeau M, Touzery C, Arveux P, et al. Predictive value for disease progression of serum thyroglobulin levels measured in the postoperative period and after (131)I ablation therapy in patients with differentiated thyroid cancer. J Nucl Med 2004;45:988-994. 64. Shaha AR. Implications of prognostic factors and risk groups in the management of differentiated thyroid cancer. Laryngoscope 2004;114:393-402. 65. Hay ID, McConahey WM, Goellner JR. Managing patients with papillary thyroid carcinoma: insights gained from the Mayo Clinic’s experience of treating 2,512 consecutive patients during 1940 through 2000. Trans Am Clin Climatol Assoc 2002;113:241-260. 66. Mazzaferri EL, Kloos RT. Clinical review 128: current approaches to primary therapy for papillary and follicular thyroid cancer. J Clin Endocrinol Metab 2001;86:1447-1463. 67. British Thyroid Association. Guidelines for the Management of Thyroid Cancer in Adults. London: Royal College of Physicians, 2002, pp 1-70. 68. Sanders LE, Cady B. Differentiated thyroid cancer: reexamination of risk groups and outcome of treatment. Arch Surg 1998;133:419-425. 69. Shaha AR, Shah JP, Loree TR. Patterns of failure in differentiated carcinoma of the thyroid based on risk groups. Head Neck 1998;20:26-30. 70. Pacini F, Landenson PW, Schlumberger M, et al. Radioiodine ablation of thyroid remnants after preparation with recombinant human thyrotropin in differentiated thyroid carcinoma: results of an international, randomized, controlled study. J Clin Endocrinol Metab 2006;91:926-932. 71. Hay ID, Thompson GB, Grant CS, et al. Papillary thyroid carcinoma managed at the Mayo Clinic during six decades (1940-1999): temporal trends in initial therapy and long-term outcome in 2444 consecutively treated patients. World J Surg 2002;26:879-885. 72. Cooper DS, Specker B, Ho M, et al. Thyrotropin suppression and disease progression in patients with differentiated thyroid cancer: results from the National Thyroid Cancer Treatment Cooperative Registry. Thyroid 1998;8:737-744. 73. Sherman S. Adjuvant therapy and long-term management of differentiated thyroid carcinoma. Semin Surg Oncol 1999;16: 30-33. 74. Schlumberger M, Berg G, Cohen O, et al. Follow-up of low-risk patients with differentiated thyroid carcinoma: a European perspective. Eur J Endocrinol 2004;150:105-112. 75. Pineda JD, Lee T, Ain K, et al. Iodine-131 therapy for thyroid cancer patients with elevated thyroglobulin and negative diagnostic scan. J Clin Endocrinol Metab 1995;80:1488-1492. 76. Hjiyiannakis P, Mundy J, Harmer C. Thyroglobulin antibodies in differentiated thyroid cancer. Clin Oncol (R Coll Radiol) 1999;11:240-244. 77. Westbury C, Vini L, Fisher C, et al. Recurrent differentiated thyroid cancer without elevation of serum thyroglobulin. Thyroid 2000;10:171-176. 78. Vassilopoulou-Sellin R, Schultz PN, Haynie TP. Clinical outcome of patients with papillary thyroid carcinoma who have recurrence after initial radioactive iodine therapy. Cancer 1996;78:493-501. 79. Anderson P, Kinsella J, Loree TR, et al. Differentiated carcinoma of the thyroid with extrathyroid extension. Am J Surg 1995;170:467-470. 80. McCaffrey TV, Bergstralh EJ, Hay ID. Locally invasive papillary thyroid carcinoma: 1940-1990. Head Neck 1994;16:165-172. 81. Farahati J, Reiners C, Stuschke M, et al. Differentiated thyroid cancer. Impact of adjuvant external radiotherapy in patients with perithyroidal tumor infiltration (stage pT4). Cancer 1996;77:172-180. 82. Tubiana M, Haddad E, Schlumberger M, et al. External radiotherapy in thyroid cancers. Cancer 1985;55:2062-2071. 83. Simpson WJ, Panzarella T, Carruthers JS, et al. Papillary and follicular thyroid cancer: impact of treatment in 1578 patients. Int J Radiat Oncol Biol Phys 1988;14:1063-1075.

84. Esik O, Németh G, Eller J. Prophylactic external irradiation in differentiated thyroid cancer: a retrospective study over a 30year observation period. Oncology 1994;51:372-379. 85. Phlips P, Hanzen C, Andry G, et al. Postoperative irradiation for thyroid cancer. Eur J Surg Oncol 1993;19:399-404. 86. Ford D, Giridharan S, McConkey C, et al. External beam radiotherapy in the management of differentiated thyroid cancer. Clin Oncol (R Coll Radiol) 2003;15:337-341. 87. Kim TH, Yang DS, Jung KY, et al. Value of external irradiation for locally advanced papillary thyroid cancer. Int J Radiat Oncol Biol Phys 2003;55:1006-1012. 88. Keum KC, Suh YG, Koom WS, et al. The role of postoperative external-beam radiotherapy in the management of patients with papillary thyroid cancer invading the trachea. Int J Radiat Oncol Biol Phys 2006;65:474-480. 89. Chow SM, Law SC, Mendenhall WM, et al. Papillary thyroid carcinoma: prognostic factors and the role of radioiodine and external radiotherapy. Int J Radiat Oncol Biol Phys 2002;52:784795. 90. Sheline GE, Galante M, Lindsay S. Radiation therapy in the control of persistent thyroid cancer. Am J Roentgenol Radium Ther Nucl Med 1966;97:923-930. 91. O’Connell ME, A’Hern RP, Harmer CL. Results of external beam radiotherapy in differentiated thyroid carcinoma: a retrospective study from the Royal Marsden Hospital. Eur J Cancer 1994;30A:733-739. 92. Coburn M, Teates D, Wanebo HJ. Recurrent thyroid cancer. Role of surgery versus radioactive iodine (I131). Ann Surg 1994;219:587-593; discussion 593-585. 93. Gillenwater AM, Goepfert H. Surgical management of laryngotracheal and esophageal involvement by locally advanced thyroid cancer. Semin Surg Oncol 1999;16:19-29. 94. Schlumberger M, Challeton C, De Vathaire F, et al. Radioactive iodine treatment and external radiotherapy for lung and bone metastases from thyroid carcinoma. J Nucl Med 1996;37:598605. 95. Casara D, Rubello D, Saladini G, et al. Distant metastases in differentiated thyroid cancer: long-term results of radioiodine treatment. Tumori 1991;77:432-436. 96. Shoup M, Stojadinovic A, Nissan A, et al. Prognostic indicators of outcomes in patients with distant metastases from differentiated thyroid carcinoma. J Am Coll Surg 2003;197:191-197. 97. Proye CA, Dromer DH, Carnaille BM, et al. Is it still worthwhile to treat bone metastases from differentiated thyroid carcinoma with radioactive iodine? World J Surg 1992;16:640-645; discussion 645-646. 98. Shaha AR, Shah JP, Loree TR. Differentiated thyroid cancer presenting initially with distant metastasis. Am J Surg 1997;174:474-476. 99. Lind P. 131I whole body scintigraphy in thyroid cancer patients. Q J Nucl Med 1999;43:188-194. 100. Grunwald F, Briele B, Biersack H-J. Non-131I scintigraphy in the treatment and follow-up of thyroid cancer. Q J Nucl Med 1999;43:195-206. 101. Robbins RJ, Wan Q, Grewal RK, et al. Real-time prognosis for metastatic thyroid carcinoma based on 2-[18F]fluoro-2-deoxyd-glucose-positron emission tomography scanning. J Clin Endocrinol Metab 2006;91:498-505. 102. Shimaoka K, Schoenfeld DA, DeWys WD, et al. A randomized trial of doxorubicin versus doxorubicin plus cisplatin in patients with advanced thyroid carcinoma. Cancer 1985;56:2155-2160. 103. Ekman ET, Lundell G, Tennvall J, et al. Chemotherapy and multimodality treatment in thyroid carcinoma. Otolaryngol Clin North Am 1990;23:523-527. 104. Wells SA, Baylin SB, Linehan WM, et al. Provocative agents and the diagnosis of medullary thyroid carcinoma. Ann Surg 1978;188:139-141. 105. Simpson WJ. Radioiodine and radiotherapy in the management of thyroid cancers. Otolaryngol Clin North Am 1990;23:509-521. 106. Tisell LE, Hansson G, Jansson S, et al. Reoperation in the treatment of asymptomatic metastasizing medullary thyroid carcinoma. Surgery 1986;99:60-66.

16  The Thyroid 107. Moley JF, Wells SA, Dilley WG, et al. Reoperation for recurrent or persistent medullary thyroid cancer. Surgery 1993;114: 1090-1095; discussion 1095-1096. 108. Samaan NA, Schultz PN, Hickey RC. Medullary thyroid carcinoma: prognosis of familial versus sporadic disease and the role of radiotherapy. J Clin Endocrinol Metab 1988;67:801-805. 109. Saad M, Orddonez N, Rashid R, et al. Medullary carcinoma: prognostic factors and treatment. Int J Radiat Oncol Biol Phys 1984;9:161-169. 110. O’Riordain D, O’Brien T, Crotty TB, et al. Multiple endocrine neoplasia type 2B: more than an endocrine disorder. Surgery 1995;118:936-942. 111. Brierley JD, Tsang RW, Simpson WJ, et al. Medullary thyroid cancer—analyses of survival and prognostic factors and the role of radiation therapy in local control. Thyroid 1996;6:305-310. 112. Moley JF. Medullary thyroid carcinoma. Curr Treat Options Oncol 2003;4:339-347. 113. Mak A, Morrison W, Garden A, et al. The value of postoperative radiotherapy for regional medullary carcinoma of the thyroid. Int J Radiat Oncol Biol Phys 1994;30(Suppl):234. 114. Fife KM, Bower M, Harmer CL. Medullary thyroid cancer: the role of radiotherapy in local control. Eur J Surg Oncol 1996;22:588-591. 115. Nguyen TD, Chassard JL, Lagarde P, 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;23:1-5. 116. Tisell LE, Ahlman H, Wangberg B, et al. Somatostatin receptor scintigraphy in medullary thyroid carcinoma. Br J Surg 1997;84:543-547. 117. Girelli ME, Nacamulli D, Pelizzo MR, et al. Medullary thyroid carcinoma: clinical features and long-term follow-up of ­seventy-eight patients treated between 1969 and 1986. Thyroid 1998;8:517-523. 118. Di Bartolomeo M, Bajetta E, Bochicchio AM, et al. A phase II trial of dacarbazine, fluorouracil and epirubicin in patients with neuroendocrine tumours. A study by the Italian Trials in Medical Oncology (I.T.M.O.). Group. Ann Oncol 1995;6:77-79. 119. Lupoli G, Cascone E, Arlotta F, et al. Treatment of advanced medullary thyroid carcinoma with a combination of recombinant interferon alpha-2b and octreotide. Cancer 1996;78:1114-1118. 120. Stein R, Juweid M, Zhang CH, et al. Assessment of combined radioimmunotherapy and chemotherapy for treatment of medullary thyroid cancer. Clin Cancer Res 1999;5:3199s-3206s. 121. Hundahl SA, Fleming ID, Fremgen AM, et al. A National Cancer Data Base report on 53,856 cases of thyroid carcinoma treated in the U.S., 1985-1995. Cancer 1998;83:2638-2648. 122. Kebebew E, Greenspan FS, Clark OH, et al. Anaplastic thyroid carcinoma. Treatment outcome and prognostic factors. Cancer 2005;103:1330-1335. 123. McIver B, Hay ID, Giuffrida DF, et al. Anaplastic thyroid carcinoma: a 50-year experience at a single institution. Surgery 2001;130:1028-1034. 124. Ain KB. Anaplastic thyroid carcinoma: behavior, biology, and therapeutic approaches. Thyroid 1998;8:715-726. 125. van der Laan BF, Freeman JL, Tsang RW, et al. The association of well-differentiated thyroid carcinoma with insular or anaplastic thyroid carcinoma: evidence for dedifferentiation in tumor progression. Endocr Pathol 1993;4:215-221. 126. Nel CJ, van Heerden JA, Goellner JR, et al. Anaplastic carcinoma of the thyroid: a clinicopathologic study of 82 cases. Mayo Clin Proc 1985;60:51-58.

349

127. Passler C, Scheuba C, Prager G, et al. Anaplastic (undifferentiated) thyroid carcinoma (ATC). A retrospective analysis. Langenbecks Arch Surg 1999;384:284-293. 128. Voutilainen PE, Multanen M, Haapiainen RK, et al. Anaplastic thyroid carcinoma survival. World J Surg 1999;23:975-978; discussion 978-979. 129. Pierie JP, Muzikansky A, Gaz RD, et al. The effect of surgery and radiotherapy on outcome of anaplastic thyroid carcinoma. Ann Surg Oncol 2002;9:57-64. 130. Tennvall J, Lundell G, Wahlberg P, et al. Anaplastic thyroid carcinoma: three protocols combining doxorubicin, hyperfractionated radiotherapy and surgery. Br J Cancer 2002;86:1848-1853. 131. Haigh PI, Ituarte PH, Wu HS, et al. Completely resected anaplastic thyroid carcinoma combined with adjuvant chemotherapy and irradiation is associated with prolonged survival. Cancer 2001;91:2335-2342. 132. Schlumberger M, Parmentier C, Delisle MJ, et al. Combination therapy for anaplastic giant cell thyroid carcinoma. Cancer 1991;67:564-566. 133. Chow E, Tsang RW, Brierley JD, et al. Parathyroid carcinoma— the Princess Margaret Hospital experience. Int J Radiat Oncol Biol Phys 1998;41:569-572. 134. Cook AM, Vini L, Harmer C. Squamous cell carcinoma of the thyroid: outcome of treatment in 16 patients. Eur J Surg Oncol 1999;25:606-609. 135. Logue JP, Tsang RW, Brierley JD, et al. Radioiodine ablation of residual thyroid tissue in thyroid cancer: relationship between administered activity, neck uptake, and outcome. Br J Radiol 1994;67:1127-1131. 136. Maxon HR. Detection of residual and recurrent thyroid cancer by radionuclide imaging. Thyroid 1999;9:443-446. 137. Hodgson DC, Brierley JD, Tsang RW, et al. Prescribing 131Iodine based on neck uptake produces effective thyroid ablation and reduced hospital stay. Radiother Oncol 1998;47:325-330. 138. Park HM, Perkins OW, Edmondson JW, et al. Influence of diagnostic radioiodines on the uptake of ablative dose of iodine-131. Thyroid 1994;4:49-54. 139. Maxon HR, Thomas SR, Hertzberg VS, et al. Relation between effective radiation dose and outcome of radioiodine therapy for thyroid cancer. N Engl J Med 1983;309:937-941. 140. Maxon HR. Quantitative radioiodine therapy in the treatment of differentiated thyroid cancer. Q J Nucl Med 1999;43:313-323. 141. U.S. Nuclear Regulatory Commission Regulatory Guide 8.39. Washington, DC: UNR, 1997. 142. Harmer C, Bidmead M, Shepherd S, et al. Radiotherapy planning techniques for thyroid cancer. Br J Radiol 1998;71:10691075. 143. Rosenbluth BD, Serrano V, Happersett L, et al. Intensity­modulated radiation therapy for the treatment of nonanaplastic thyroid cancer. Int J Radiat Oncol Biol Phys 2005;63:1419-1426. 144. Nutting CM, Convery DJ, Cosgrove VP, et al. Improvements in target coverage and reduced spinal cord irradiation using ­intensity-modulated radiotherapy (IMRT) in patients with carcinoma of the thyroid gland. Radiother Oncol 2001;60:173-180. 145. Grunwald F, Schomburg A, Menzel C, et al. Changes in the blood picture after radioiodine therapy of thyroid cancer. Med Klin 1994;10:522-528. 146. Maxon H. The role of radioiodine in the treatment of thyroid cancer. Thyroid Today 1993;16:1-9. 147. Rubino C, de Vathaire F, Dottorini ME, et al. Second primary malignancies in thyroid cancer patients. Br J Cancer 2003;89:1638-1644. 148. Dottorini M, Lomuscio G, Mazzucchelli L, et al. Assessment of female fertility and carcinogenesis after iodine-131 therapy for differentiated thyroid cancer. J Nucl Med 1995;36:21-27.

The Breast 17 Thomas A. Buchholz and Eric A. Strom ANATOMY OF THE BREAST AND DRAINING LYMPHATICS Breast Axillary Lymph Nodes Supraclavicular Lymph Nodes Internal Mammary Lymph Nodes Lymphatic Drainage to the Contralateral Breast and Axilla RESPONSE OF THE NORMAL BREAST TO IRRADIATION BREAST CANCER INCIDENCE AND EPIDEMIOLOGY Incidence Rates Mortality Rates Risk Factors Determining an Individual’s Risk BREAST CANCER PATHOGENESIS AND PREVENTION Breast Cancer Development Breast Cancer Prevention Strategies DIAGNOSIS AND STAGING Diagnosis Staging THERAPY FOR DUCTAL CARCINOMA IN SITU Common Presentations and Pathologic Subtypes Treatment THERAPY FOR EARLY-STAGE BREAST CANCER Equivalence of Breast-Conservation Therapy and Modified Radical Mastectomy Role of Radiation Therapy in Breast-Conservation Therapy

Breast-Conservation Therapy for Women Older Than 70 Years with Stage I Disease Contraindications for Breast-Conservation Therapy Factors Affecting Outcomes of Breast-Conservation Therapy Partial Breast Irradiation Axillary Treatment Systemic Therapy Sequencing of Treatment Modalities RADIATION THERAPY AFTER MASTECTOMY Patient Selection Radiation Therapy after Neoadjuvant Chemotherapy and Mastectomy BREAST CONSERVATION FOR ADVANCED DISEASE RADIATION THERAPY FOR ADENOCARCINOMA IN AXILLARY NODES WITH AN UNKNOWN PRIMARY TUMOR Initial Evaluation Treatment RADIATION THERAPY FOR INFLAMMATORY BREAST CANCER RADIATION THERAPY FOR LOCOREGIONAL RECURRENCE AFTER MASTECTOMY RADIATION THERAPY FOR PALLIATION RADIATION TECHNIQUES AND TREATMENT PLANNING Treatment Fields Boost Fields Dose Prescriptions Dosimetry

Anatomy of the Breast and Draining Lymphatics Designing radiation therapy fields that optimally cover targeted areas at risk for disease while sparing uninvolved normal structures requires that radiation oncologists have a comprehensive understanding of external and three­dimensional radiographic anatomy of the breast and chest wall, the axilla, and the ipsilateral supraclavicular and internal mammary lymph nodes. The importance of understanding anatomy is often underappreciated in breast cancer radiation oncology because in the past, fields were designed with standard external landmarks and dosages prescribed in standard templates. However, computed tomography (CT) treatment planning has allowed radiation treatments to be delivered in an optimized and more

customized fashion, underscoring the need to understand the anatomic landmarks of tissue at risk, particularly with respect to the breast’s lymphatic drainage pathways. The remainder of this section constitutes a brief review of the anatomy of the breast and surrounding lymph node areas.

Breast Mammary tissue extends medially to laterally from the midline to the midaxillary line or latissimus dorsi muscle. The cranial and caudal borders of the breast are ­typically the sixth anterior rib and the second anterior rib, respec­ tively, although the breast may extend higher in the  anterolateral region as it nears the axilla. This axillary extension of the breast is commonly known as the tail of Spence. Breast glandular tissue is supported by fibrous septa (Cooper’s suspensory ligaments) and contains blood vessels,

353

354

PART 4  Breast

nerves, lymphatics, and adipose tissue intermingled so that no surgical planes are within the gland itself. The ligamentous structures attach to the pectoralis major muscle, the overlying fascia of which serves as the deep boundary of the breast. Underneath the pectoralis major muscle lies the chest wall, which consists of the rib cage and interosseous muscles and fascia. The volume of the breast is not equally distributed ­between its boundaries. Primary breast cancers occur most often in areas with the largest volume of breast tissue, most commonly the upper outer quadrant. In carefully documented series reported by Haagensen1 and Donegan,2 40% to 48% of primary breast tumors were found to be located in the upper outer quadrant, 17% to 27% in the central region, 16% in the upper inner quadrant, 10% to 11% in the lower outer quadrant, and 6% to 7% in the lower inner quadrant.1,2 Microscopically, the breast parenchyma is composed of a series of ductal and lobular units. The ductal system of the breast is divided into segments, each of which has a ­series of branching ducts that converge proximally into major lactiferous ducts near the nipple-areola complex. The distal end of the ductal system interfaces with lobular breast parenchyma at the terminal ductal-lobular unit, the most common location for the development of breast cancer. The breast has a rich lymphatic plexus with primary drainage to the lymph nodes of the axilla, internal mammary chain, and supraclavicular fossa (Fig. 17-1).3 Lymph nodes draining the breast can also be identified within the substance of the breast (the intramammary nodes) and ­below the pectoralis major muscle (the subpectoral or ­interpectoral [Rotter’s] nodes).

Axillary Lymph Nodes The predominant lymphatic drainage of the breast flows to the axilla. The axilla is bordered medially by the serratus ­anterior muscle, anteriorly by the pectoralis major and ­minor muscles, and posteriorly by the converging muscles and fascia of the subscapular region. The apex of the ­ axilla lies ­below the midclavicle. Within this space, lymph nodes lie in proximity to the vascular structures. Axillary lymph nodes are customarily divided into three levels ­according to the following anatomic guidelines. Level I (low ­axillary) nodes lie below the inferolateral border of the ­ pectoralis ­ minor muscle, level II (midaxillary) nodes lie directly ­beneath the pectoralis minor muscle, and level III nodes lie superior to the pectoralis minor muscle. Level III lymph nodes are also called apical lymph nodes, subclavicular lymph nodes, or infraclavicular lymph nodes. As discussed later in this chapter, the presence of disease in the level III region increases the disease stage. Patients with advanced disease are also at risk for involvement of Rotter’s nodes. Figure 17-2 displays an axial CT image of a patient with breast cancer that presented with clinical and radiographic evidence of disease in level II axillary lymph nodes, in ­Rotter’s nodes, and in an ipsilateral internal mammary lymph node. Involvement of axillary lymph nodes remains the most important prognostic factor in breast cancer. Initially, axillary disease tends to appear in level I or level II nodes. Skip metastases (i.e., metastases in higher-level nodes in the ­absence of metastases in lower-level nodes) are rare; in an analysis of 539 patients who underwent complete axillary dissection, Veronesi4 reported skip metastases in only 3.8%

3

2

III B 4

A

C 5

1

6 7

10

II 9 8

D 11 E

G

I 12 F

13

FIGURE 17-1.  Anatomy of the breast and lymphatics: 1, internal mammary artery and vein; 2, substernal cross drainage to contralateral internal mammary lymphatic chain; 3, subclavius muscle and Halsted ligament; 4, lateral pectoral nerve (from the lateral cord); 5, pectoral branch from thoracoacromial vein; 6, pectoralis minor muscle; 7, pectoralis major muscle; 8, lateral thoracic vein; 9, medial pectoral nerve (from the medial cord); 10, pectoralis minor muscle; 11, median nerve; 12, subscapular vein; 13, thoracodorsal vein; A, internal mammary lymph nodes; B, apical lymph nodes; C, interpectoral (Rotter’s) lymph nodes; D, axillary vein lymph nodes; E, central lymph nodes; F, scapular lymph nodes; G, external mammary lymph nodes. Level I lymph nodes are lateral to the lateral border of the pectoralis minor muscle. Level II lymph nodes are behind the pectoralis minor muscle. Level III lymph nodes are medial to the medial border of the pectoralis minor muscle. (Modified from Osborne MP, Kinne DW. Breast development and anatomy. In Harris JR, Hellman S, Henderson KC (eds): Breast Disease. Philadelphia: JB Lippincott, 1987, pp 1-14.)

of patients. In that same series, when level I nodes were positive, higher-level nodes were also involved in 40% of cases.4 In another series of 1228 patients who underwent complete axillary dissection at Memorial Sloan-Kettering Cancer Center, Rosen and colleagues5 found skip metastases in 1.6% of patients overall and in 3% of those with lymph node metastases.

Supraclavicular Lymph Nodes The major route by which cancer spreads to the supraclavicular nodes is through the axillary chain. Large lymphatic trunks pass from the subclavicular nodes at the axillary apex medially through the small anatomic triangle formed by the subclavian vein, the subclavian muscle tendon, and the chest wall to terminate into the jugular-­subclavian ­venous confluence directly or by lymphatic ducts. As lymph nodes in this region become obstructed with tumor, retrograde spread can occur to other supraclavicular nodes located more laterally and to lymph nodes in the lower neck. The supraclavicular lymph nodes at greatest risk of involvement are often difficult to palpate because of their location behind the sternocleidomastoid muscle, whereas the secondary, more lateral nodes are easier to examine. Supraclavicular adenopathy without axillary involvement is rare in breast cancer, and in such cases, the disease may

17  The Breast

FIGURE 17-2.  Axial CT scan shows a locally advanced breast cancer that involves a level II lymph node (underneath the pectoralis minor), a Rotter’s lymph node (above the pectoralis minor and beneath the pectoralis major), and an ipsilateral internal mammary lymph node (arrows).

have arisen from another primary site. Involvement of  supraclavicular lymph nodes is associated with poor prognosis and represents the highest-risk category of nodal ­disease (pN3c).

Internal Mammary Lymph Nodes The internal mammary lymphatic trunks are located along the intercostal spaces and contain small nodes, usually 2 to 5 mm in diameter, and loose aggregates of lymphoid tissue that do not form clearly defined nodes. In most patients, these nodes and lymphatics lie within 3 cm of the sternal edge on the parietal pleura, separated from the pleura by a thin fascia. Nodes can usually be identified in the first three intercostal spaces; Urban and Marjani6 also identified nodes in the fourth and fifth interspaces in 47% and 12%, respectively, of patients undergoing extended radical mastectomy. The internal mammary lymphatic trunks empty into the great veins by way of the right lymphatic duct or, on the left, the thoracic duct or directly through the jugular vein–subclavian vein confluence. Nodes behind the manubrium connect the right and left lymphatic trunks at the first interspace in approximately 50% of patients.1 Despite this anatomic finding, clinical manifestations of contralateral internal mammary node involvement are rare. Insight into the anatomy of internal mammary lymph nodes has been gained through the use of lymphoscintigraphy to image lymphatic drainage before sentinel lymph node surgery. In a study of more than 1200 patients with clinically negative lymph nodes who underwent this procedure, only 1.6% were found to have drainage to the internal mammary lymph nodes alone, but an additional 19.8% had drainage to the axilla and internal mammary lymph nodes; 10.3% had no visible drainage at all, and the remaining patients had only axillary drainage. Among the patients with drainage to the internal mammary lymph nodes, only 36.7% had drainage limited to the upper half of the sternum, 30.8% had drainage to lymph nodes in the lower half of the sternum, and 32.5% had drainage to both sites.7-9 In a separate analysis of the probability of drainage to the internal mammary lymph nodes based on location of the primary tumor, the highest rates of drainage to the ­internal mammary lymph nodes were found for lower inner ­ 

355

FIGURE 17-3.  Axial CT scan shows a locally advanced breast cancer that involves an internal mammary lymph node and concurrent disease within the ipsilateral breast and level I axilla (arrows).

quadrant tumors (43%), lower outer quadrant tumors (32%), and central tumors (28%).8 An axial CT slice of a patient who presented with involvement of an internal mammary lymph node and concurrent disease within the ipsilateral breast and level I axilla is shown in Figure 17-3.

Lymphatic Drainage to the Contralateral Breast and Axilla The lymphatics of the breast and surrounding nodal ­areas generally follow the course of associated blood vessels. However, obstruction of normal paths can lead to retrograde lymphatic flow, resulting in tumor involvement of the opposite breast, chest wall, or axilla. Examination of the contralateral breast and nodal basins is therefore extremely important in disease staging, treatment, and ­follow-up care of patients with breast cancer.

Response of the Normal Breast to Irradiation In adult women, the breast is very tolerant of high-dose ­irradiation, and if standard techniques and dosages are used, radiation therapy for treatment of breast cancer is unlikely to cause serious injury to normal tissue. The most common complications after irradiation of the breast are excessive subcutaneous fibrosis and atrophy of breast tissue. However, complications such as these are unusual, with several series reporting good to excellent aesthetic outcomes in 80% to 95% of cases after breast irradiation to total doses of 45 to 50.4 Gy in daily fractions of 1.8 to 2.0 Gy.10-12 For the normal breast, a steep dose-response gradient for complications is evident between whole-breast doses of 50 and 60 Gy. In a series of 458 patients treated with breast-conserving surgery and breast irradiation, Taylor and Meltzer10 reported that the percentage of women with good to excellent aesthetic outcomes decreased with increasing radiation dose. Results were good to excellent for 85% of women who received 50 Gy to the breast, 73% of women who received 50 to 52 Gy, and 20% of ­women who received 52 to 62 Gy.10 Other investigators have

356

PART 4  Breast

r­ eported that the use of a tumor bed boost, which typically escalates the dose to a particular region of breast tissue to 60 Gy or greater, can have an adverse effect on aesthetic ­outcome.11,13 The daily dose per fraction seems to be as important as total dose in determining the response of the normal breast to irradiation. Van Limbergen and colleagues14 found that increasing the dose per fraction beyond 2 Gy negatively influenced the final cosmetic outcome. From their results, they estimated the alpha/beta ratio (defined in Chapter 1)   for breast tissue to be 2.5 Gy. These findings suggest that the breast is a late-responding normal tissue and that increases in the dose per fraction beyond 2 Gy will decrease the therapeutic ratio. Cosmetic complications of breast irradiation probably result from damage to the supportive connective tissue elements of the breast rather than damage to the breast parenchyma. Information on how therapeutic doses of radiation affect the breast parenchyma is derived largely from studies of mastectomy specimens from patients who received preoperative breast irradiation. Schnitt and colleagues15 compared 36 such breast specimens with  35 control breast specimens from women who had not had preoperative irradiation and found that the most common postirradiation change was atypical epithelial cells in the terminal ductal-lobular unit associated with lobular sclerosis and atrophy. Lactation and breast-feeding are possible after radiation therapy for breast cancer. In a survey of 53 patients who ­ became pregnant after a course of breast irradiation for breast cancer, Tralins16 reported that 34% of those women were able to lactate and 25% were able to successfully breast-feed from the treated breast. Moran and colleagues17 reported that lactation was possible in 56% of 22 breasts that had been irradiated for breast cancer but that lactation was diminished in most cases. Perhaps the most important potential negative effect of breast irradiation is an increase in the risk of subsequent breast cancer development. Some of the first evidence of the carcinogenic effect of radiation came from longitudinal studies of Japanese atomic bomb survivors. For women exposed to 1 Sv of radiation from the atomic bomb in Hiroshima, Tokunaga and colleagues18 reported a 13-fold increase in the relative risk of breast cancer for women younger than 35 years and a twofold increase for women older than 35 years at the time of exposure. The importance of radiation as a breast carcinogen was further confirmed by the discovery of increased breast cancer rates among women treated with radiation for nonmalignant conditions such as tuberculosis and enlargement of the thymus.18-20 The use of radiation to treat cancer has been shown to carry carcinogenic risks. In a large study of girls treated with mantle irradiation (i.e., treatment of the neck, axilla, and mediastinal lymph nodes) for Hodgkin disease, the risk of developing breast cancer was increased 75-fold compared with the risk in the general population, and the 30-year actuarial risk of developing breast cancer approached 35%.21 The lifetime risk was projected to be much higher than the risk for nonirradiated women and close to that for women with a germline mutation in the BRCA1 gene.21 The age of the patient at the time of radiation therapy and the total radiation dose were important contributors to breast cancer risk, with adolescent age 

(10 to 16 years) at exposure and radiation doses of 20 to 40 Gy conferring the highest relative risk. Whether radiation therapy for breast cancer increases the risk of new breast tumors is much less clear. This uncertainty arises from difficulties in distinguishing ipsilateral new primary breast tumors from local recurrence of the disease and in showing a relationship between new ­primary breast tumors and previous irradiation. Some evidence suggests that breast irradiation may increase the incidence of contralateral breast cancer in women 45 years old or younger. In an analysis based on the Connecticut cancer registry database of 41,109 breast cancer patients, Boice and colleagues22 reported that women who survived for at least 10 years after diagnosis and who had been treated with radiation had a relative risk of a second breast tumor of 1.33 compared with the risk for 10-year survivors who had not been treated with radiation. Although this finding has not been reproduced in clinical studies evaluating outcomes after breast-conservation therapy (BCT),23,24 a meta-analysis of the data from all radiation therapy trials in breast cancer showed a slight increase in the relative risk of developing breast cancer (1.18 for irradiated versus nonirradiated patients, P = .002).25 Irradiation of the breast bud before adolescence can prevent subsequent breast growth. Treatment of lung ­metastases from Wilms’ tumor and other pediatric malignancies with dosages as low as 12 Gy delivered over 2 weeks has resulted in impairment of breast development during adolescence.26 Radiation has been used therapeutically in the past for prevention of breast engorgement and discomfort in men with prostate cancer treated with estrogens.

Breast Cancer Incidence and Epidemiology Incidence Rates Breast cancer is the most common nondermatologic cancer among women in the United States. The American ­Cancer Society estimated that 184,450 new cases of invasive breast cancer, and an additional 67,770 new cases of in situ breast cancer are estimated to be diagnosed in 2008, accounting for 26% of all cancers diagnosed in women.27,28 The incidence of invasive breast cancer steadily increased between 1980 and 2000, but since 2001, the incidence has dropped by 3.5%, perhaps because of the decreased use of hormone replacement therapy during that time.27-29 ­However, over this same period, the incidence of noninvasive breast ­cancer has continued to increase.

Mortality Rates Even though the percentage of women with early-stage ­invasive disease has increased relative to the percentage of women with advanced disease, breast cancer remains the second leading cause of cancer death among women, with approximately 40,930 deaths from breast cancer occurring in 2008.27,28 However, the rates of breast cancer death have declined by a consistent 2% to 3% since 1990, probably from the combined effects of screening and treatment advances. Death rates from breast cancer vary significantly around the world. In general, Western industrialized nations have much higher breast cancer incidence and mortality rates

17  The Breast

TABLE 17-1.    Breast Cancer Risk Factors Significant Risk Factors

Moderate Risk Factors

Unlikely Risk Factors

Female sex

Higher alcohol consumption

Higher fat intake

Older age

Low amounts of exercise

Fibrocystic breast disease

Family history of breast cancer

Use of hormone replacement therapy

Fibroadenoma

Nulliparity

Lack of breast-feeding

Simple breast cysts

357

TABLE 17-2.   Estimated New Breast Cancer Cases and Deaths in Women by Age in the United States in 2007 Cases of Invasive Disease*

Deaths*

7,640

16,150

2,830

45 and older

54,390

162,330

37,630

Younger than 55

24,920

54,180

9,140

Age (yr) Younger than 45

Cases of In Situ Disease*

55 and older

37,110

124,300

31,320

Older age at ­menarche

Use of oral ­contraceptives

Younger than 65

40,520

105,960

16,950

Personal history of breast cancer

Use of tobacco

65 and older

21,510

72,520

23,510

All ages

62,030

178,480

40,460

Personal history of lobular ­carcinoma in situ Personal history of atypical ductal hyperplasia Germline mutation in BRCA1 or BRCA2

*Rounded

to the nearest 10. Data from American Cancer Society. Breast Cancer Facts and ­Figures 2007-2008. Atlanta: American Cancer Society, 2008.

women, younger women have a greater lifetime risk. Only 0.43% of women develop breast cancer before the age of 40 years, whereas 4% of women develop breast cancer  between the ages of 40 and 59 and 6.88% of women ­develop breast cancer between the ages of 60 and 79.30

Reproductive and Hormonal Factors than do nations of the Far East. For example, between 1994 and 1997, the age-adjusted death rate from breast cancer in China and Japan was approximately 25% that of the death rates in the United States, Western Europe, and the Scandinavian countries.30

Risk Factors Table 17-1 lists risk factors for developing breast cancer.

Female Sex The strongest risk factor for breast cancer is being female. Breast cancer in men accounts for only about 1% of all new breast cancer cases, with an estimated 1990 new cases in 2008.27

Older Age The second most important risk factor for breast cancer ­development is an individual’s age. Between 1998 and 2002, 95% of new breast cancer cases developed in women 40 years old or older.31 The annual risk increases exponentially up to the age of menopause, at which time the rate of increase in the annual risk slows significantly. After the age of 80 years, the annual incidence of breast cancer begins to show a slight decline.32 The risk of developing invasive or noninvasive breast cancer as a function of age is shown in Table 17-2.27 For women in their late 30s, the annual risk of developing breast cancer is approximately 0.07% per year. This increases to 0.44% per year for women in their late 70s. Although these percentages may seem low, they represent the risk for only a given year. The lifetime risk is a summation of the annual breast cancer risks. Because younger women have a longer life expectancy than older

The risk of breast cancer development is also determined by age at menarche, age at first pregnancy, and age at menopause. In a large study of Swedish women, the risk of breast cancer increased approximately 13% for every 5-year ­increase in age at first birth.33 Originally, it was thought that pregnancy early in life afforded protection against breast cancer by causing differentiation of the terminal ductallobular units of the breast. However, this mechanism has been questioned. The short-term high levels of circulating estrogen associated with pregnancy are protective against breast cancer later in life, but the mechanism of this protective effect is still unknown. Paradoxically, chronic exposure to low-dose estrogen increases breast cancer risk. The length of exposure of the breast to these lower estrogen levels is determined by a woman’s lifetime menstruation history. A woman with natural menopause at age 44 years has one half of the breast cancer risk of the general United States population, in which the median age at menopause is 52 years. Postmenopausal estrogen replacement therapy also modestly increases the risk of breast cancer. The Collaborative Group on Hormonal Factors in Breast Cancer performed a meta-analysis of 51 epidemiologic studies that included 52,705 women with breast cancer and 108,411 women without breast cancer.34 The investigators reported that postmenopausal hormone replacement therapy increased the annual relative risk of developing breast cancer by 2.3% for each year of hormonal therapy. A later study of 46,000 women reported that the combined use of estrogen and progesterone increased the relative risk of breast cancer by 8% compared with the risk in nonusers, whereas the use of estrogen alone increased the relative risk by only 1%.35

358

PART 4  Breast

Results from two large prospective studies conducted through the Women’s Health Initiative have provided additional insights concerning the risks and benefits associated with postmenopausal hormonal replacement therapy. The largest of these studies was a randomized, placebocontrolled study that investigated combined estrogen and progesterone replacement therapy for postmenopausal women and enrolled a total of 16,608 women.36 The ­results indicated that hormone replacement therapy was associated with an increased risk of breast cancer development (hazard ratio [HR] = 1.24); moreover, the tumors diagnosed in the group that received hormone replacement were larger and more often involved axillary lymph nodes. Hormone replacement therapy was also associated with a higher risk of coronary heart disease (HR = 1.24).37 In a second placebo-controlled, randomized study, this group evaluated estrogen-alone replacement therapy for postmenopausal women who had previously undergone hysterectomy.38 The data monitoring committee terminated that trial early when estrogen replacement therapy was found to be associated with an increased risk of stroke. The risk of coronary heart disease, however, was not increased, and the risk of breast cancer development seemed to decrease, although this apparent difference was not statistically significant (HR = 0.77, P = .06). When the data from the two trials were combined, hormone replacement therapy also seemed to increase the risk of dementia or mild cognitive impairment.39 The results of these two trials have significantly changed the use of hormone replacement therapy in the United States. Although many questions remain unanswered, these trials provided some initial evidence that the risk associated with hormone replacement therapy may outweigh the benefits. Understanding the effects of estrogen and progesterone on breast cancer risk may allow the development of strategies to minimize breast cancer risk. In an innovative research program, Guzman and colleagues40 explored chemoprevention of breast cancer through estrogen modulation. Using a rat model, they showed that induction of a false pregnancy with estradiol and progesterone early in life was quite effective in protecting against ­ carcinogeninduced mammary cancer.40 The implication of their work is that a similar strategy used in young adolescent girls may markedly reduce the risk of breast cancer in later life, although it is unlikely that modulation of hormonal levels in healthy female adolescents will be studied any time soon.

Family History of Breast Cancer Family history is another important determinant of an individual’s breast cancer risk. Women with a first-degree relative (i.e., mother or sister) with a history of breast cancer have an increased risk of developing the disease.41 This increased risk can be explained in part by inheritance of a genetic condition that predisposes an individual to breast cancer development (e.g., mutations in BRCA1 or BRCA2), shared lifestyle, and inheritance of genes that ­affect risk factors, such as body habitus and age at menarche. Between 20% and 25% of women diagnosed with breast cancer have a positive family history of the disease, and approximately 10% of women with breast cancer are from families that display an autosomal dominant pattern of breast cancer inheritance.

The actual risk that family history conveys depends on the number of relatives affected and their age at ­diagnosis; having a first-degree relative with premenopausal breast cancer conveys a greater risk than does having a first-­degree relative with postmenopausal cancer. Women with one firstdegree relative affected by the disease have an increased relative risk of developing breast cancer two to three times that of women with no family history.41 ­Women with two or more first-degree relatives with a diagnosis of breast ­cancer have a risk four to six times that of women with no family history (or two to three times that of the averagerisk population).

Environmental Factors Few environmental agents consistently demonstrate an association with breast cancer development. The clearest example of a breast cancer carcinogen is ionizing radiation. The carcinogenic effects of radiation on the normal breast are discussed under “Response of the Normal Breast to ­Irradiation.”

Dietary Factors It was once believed that increases in dietary fat intake could to help explain why rates of breast cancer are so much higher among women of Asian heritage who live in the ­United States than among Asian women living in Asia. However, carefully conducted studies, including a large study of 30,000 nurses that thoroughly evaluated diet,42 indicated that dietary fat does not affect breast cancer risk. Another randomized, prospective trial conducted by the Women’s Health Initiative to assess dietary ­ modification versus ­observation for 48,835 postmenopausal women ­ revealed that the dietary intervention was associated with a significantly lower fat intake and a nonsignificant trend (P = .07)  toward a lower risk of breast cancer ­development.43 An association between alcohol use and breast cancer has consistently been demonstrated in large studies correlating dietary habits with breast cancer risk. Alcohol increases the relative risk of developing breast cancer in a dose-dependent fashion, with three to four alcoholic drinks per day increasing the relative risk fourfold.44

Exercise and Body Weight Like dietary factors, exercise and body weight have a moderate impact on breast cancer risk and are risk ­factors that can be controlled by individuals. Relatively large studies have demonstrated that daily exercise, particularly ­ early in life, can decrease the risk of breast cancer for premenopausal and postmenopausal women.45,46 Exercise is thought to reduce the risk of breast cancer by decreasing circulating estrogen levels, although this association has not been definitively established. Obesity can increase circulating levels of estrogen, and associations have been found between abdominal body fat and adult weight gain and breast cancer risk.47 Some evidence suggests that exercise and body weight also can influence outcome after breast cancer treatment.48-50

Lactation Lactation may make a moderate contribution to the risk of breast cancer development. In a study of more than 3600 women with breast cancer who were age-matched with control subjects, women who breast-fed for at least

17  The Breast

2 weeks had a modest decrease in breast cancer risk compared with women who never lactated.51

Pathologic Factors In addition to patient and environmental factors, several pathologic factors increase breast cancer risk. Personal history of breast cancer and personal history of ductal carcinoma in situ (DCIS) are strong risk factors for the ­development of new breast cancer. Atypical ductal hyperplasia, atypical lobular neoplasia, and lobular carcinoma in situ (LCIS) also are associated with increased risk of breast cancer development. In contrast, a history of simple cysts or fibrocystic breast parenchyma does not seem to increase breast cancer risk.

Determining an Individual’s Risk It is important to consider the combination of risk factors when a generalized risk profile is determined. For example, combining age and family history indicates that a 40-yearold woman with two or more first-degree relatives with breast cancer has approximately the same annual risk of developing breast cancer as a 60-year-old woman with no risk factors.52,53 Despite this equivalence in annual risk, the 40-year-old woman in this example has a much ­higher overall lifetime risk. Gail and colleagues52,53 have used these epidemiologic risk factors to derive a model for predicting an individual’s annual and lifetime risks of breast cancer. In this model, an individual’s annual risk of breast cancer is based on her current age, number of first-degree relatives with breast cancer, age at first birth, age at menarche, number of breast biopsies, and history of atypical ductal hyperplasia. The use of exogenous hormones is not considered in this model.

Breast Cancer Pathogenesis and Prevention Breast Cancer Development All forms of breast cancer are thought to develop as a consequence of unregulated cell growth and the development of phenotypic changes such as the ability to invade, recruit a new blood supply, and metastasize. These changes in phenotypes result from the development of aberrations in genetic pathways. Some of these aberrations are inherited (germline mutations), whereas others develop during the life of a breast cell (somatic mutations). Most breast cancer is thought to be a consequence of a series of somatic mutations. Only 20% to 25% of breast cancer ­ patients have a history of breast cancer in a first-degree ­relative. However, it is possible that some women without a first-degree relative with breast cancer still inherit a ­ genetic background that predisposes to breast cancer. These mutations may be insufficient to cause breast cancer unless accompanied by other mutations and therefore would be predicted to have a low penetrance. Historically, it has been much more difficult to discover low-penetrance mutations than to discover germline mutations that result in an autosomal dominant pattern of breast cancer development. However,  newer molecular techniques, such as deoxyribonucleic acid (DNA)–array assays and single nucleotide polymorphism (SNP) arrays, may help in the identification of low­penetrance predisposing mutations.

359

Approximately 10% of breast cancer patients have f­ amilial breast cancer, typically defined as breast cancer showing an autosomal dominant inheritance pattern. During the 1990s, germline mutations in three important ­ tumor suppressor genes—TP53, BRCA1, and BRCA2—were discovered in family members of individuals with familial breast cancer. All three genes have been shown unequivocally to predispose to breast cancer.

Germline Mutations in TP53 Germline mutations in the TP53 gene (formerly called p53 or LFS1) are rare and result in Li-Fraumeni syndrome, named after two investigators who made significant contributions to the understanding of this condition.54 TP53 is one of the most important tumor suppressor genes and has been called the “guardian of the genome” because of its critical role in cellular pathways that recognize and ­direct a response to DNA injury. TP53 regulates the transcription of several important genes. Through its regulation of CDKN1A (formerly called p21, CIP1, or WAF1), GADD45A (formerly called GADD45), and other genes, TP53 can promote cell cycle arrest after DNA injury,  allowing cells sufficient time for DNA repair before S phase, G2, and mitosis.55 Alternatively, through regulation of the BCL2 family of genes, TP53 can direct a cell toward apoptosis to prevent propagation of damage to progeny cells.56 One consequence of a germline mutation in TP53 is cancer development. Individuals with Li-Fraumeni syndrome are at increased risk for a variety of types of cancer, including childhood sarcomas, gynecologic tumors, and breast cancer. Breast cancer is the most common malignancy in patients with Li-Fraumeni syndrome; the lifetime risk is estimated to be 90%.57

Germline Mutations in BRCA1 and BRCA2 Studies of patients with familial breast cancer led to the discovery of BRCA1 in 1990 and BRCA2 in 1996.58,59 Like TP53, BRCA1 and BRCA2 are tumor suppressor genes that contribute to the stability of the genome by ­mediating the effects of the cellular response to DNA injury. The BRCA1 and BRCA2 proteins have important roles in processing double-stranded DNA damage. The proteins encoded by these genes colocalize with RAD51 to form a complex important in the repair of double-strand breaks.60,61 BRCA1 also participates in ­ transcriptioncoupled repair, homologous repair, and end-rejoining.62,63 One consequence of loss of function of BRCA1 or BRCA2 is increased radiosensitivity,60-63 which is directly related to dysregulation of the double-stranded DNA damage pathway. Individuals with a germline mutation in BRCA1 have a lifetime risk of breast cancer of 65% to 85%.64 These ­individuals have a lifetime risk of ovarian cancer approaching 50%.64,65 Other types of cancer that develop more ­frequently in BRCA1 carriers include colon cancer and prostate cancer.64 The lifetime risk of breast cancer for women with germline BRCA2 mutations mirrors that for women with BRCA1 mutations.59 BRCA2 mutation carriers are also at increased risk for ovarian cancer compared with the general population, but their risk is much less than the risk for women with BRCA1 mutations. BRCA2 is also associated with male breast cancer.66,67

360

PART 4  Breast

TABLE 17-3.   Probability of BRCA1 Germline Mutations in Various Clinical Scenarios Scenario

Probability

Mother or father proven carrier

50%

A 40-year-old woman with breast cancer and a first-degree relative with breast cancer Ashkenazi Non-Ashkenazi

Breast Cancer Biology 20% 5%

A 60-year-old woman with bilateral breast cancer and a first-degree relative with breast cancer Ashkenazi Non-Ashkenazi

underwent screening with mammography, sonography, MRI, or some combination of these three modalities suggested that MRI screening had a higher sensitivity for ­  detecting breast cancers than the other radiographic ­modalities or physical examination alone.71 The American Cancer Society now recommends that MRI be used in conjunction with mammography for screening in women with a known BRCA mutation.72

20% 5%

A 30-year-old woman with breast cancer and a first-degree relative with ovarian cancer Ashkenazi

50%

Non-Ashkenazi

20%

Data from Shattuck-Eidens D, Oliphant A, McClure M, et al. BRCA1 sequence analysis in women at high risk for susceptibility mutations. Risk factor analysis and implications for genetic testing. JAMA 1997;278:1242-1250.

Breast cancer is a heterogeneous set of diseases. Advances in genomics and molecular biology have provided insights into the genetic heterogeneity of breast cancer and how this heterogeneity affects clinical outcome. Using DNA microarray technology, Perou and colleagues73-75 described a molecular classification scheme for breast cancer based on the genomic expression of the tumor, and several groups subsequently reproduced those results. The classifications, which largely reflect molecular signals in the estrogen ­receptor and ERBB2 (also called HER2/NEU) signaling pathways, represent four broad groups: luminal A and luminal B cancer (two predominantly estrogen receptor– positive classes with distinct biologic outcomes); basal-like cancer (which is predominantly estrogen receptor–negative and HER2/NEU–negative), HER2/NEU-positive cancer, and cancer with gene expression patterns that mimic those  of the normal breast.

Breast Cancer Prevention Strategies Tamoxifen

Genetic screening for germline mutations in BRCA1 and BRCA2 is possible and is increasing in frequency for women with strong family histories of breast cancer or ovarian cancer, or both. Testing should be performed only at centers that have genetic counseling programs designed to properly inform individuals of the social, economic, and legal consequences associated with genetic testing. Germline mutations in BRCA1 and BRCA2 are rare, occurring in fewer than 7% of patients with breast cancer.59,64 Only a few breast cancer patients with a family history of the disease are predicted to carry a mutation in one of these genes. Table 17-368 contains data concerning the probability of carrying a BRCA1 mutation based on an individual’s age at cancer diagnosis, personal cancer history, and family cancer history and whether the individual is of Ashkenazi Jewish descent.68 Treating physicians and patients have consistently been found to overestimate the probability of having a germline mutation in a BRCA gene.69 No definitive data exist on which to base screening recommendations for individuals with a proven germline mutation in a gene predisposing to the development of breast cancer. The American Society of Clinical Oncology published a consensus statement recommending that such individuals undergo annual mammography and clinical and self-breast examination beginning at the age of 25 to  35 years. Annual pelvic examinations with transvaginal ­sonography, color Doppler examinations of the ovaries,  and measurement of serum cancer antigen (CA) 125 levels also are recommended beginning at age 25 to 35 years.70 The use of magnetic resonance imaging (MRI) to screen for breast cancer in young women with a known BRCA mutation has gained interest. In one nonrandomized study of 236 women with a known BRCA mutation who 

One of the most exciting areas of breast cancer research is breast cancer prevention. Increasing knowledge of the role of estrogens and progesterones in breast cancer development has led to the development of pharmacologic strategies that could significantly reduce the incidence of breast cancer over the next 2 decades. Interest in tamoxifen as a chemopreventive agent arose after several randomized trials designed to test the ­efficacy of hormonal therapy for invasive breast cancer ­ reported that tamoxifen reduced the incidence of contralateral breast cancer.76-78 On the basis of these data, in 1992, the National Surgical Adjuvant Breast and Bowel Project (NSABP) began a randomized, placebo-controlled study (P-1 trial) to test the efficacy of 5 years of tamoxifen in the prevention of breast cancer.79 Between 1992 and 1997, 13,388 women with a 1.67% or greater predicted risk of developing breast cancer within 5 years were enrolled in this trial. Risk was assessed with a modification of the Gail model that allowed enrollment of any woman older than 60 years and of selected women younger than 60 years with additional risk factors that increased their annual risk to at least that of a 60-year-old woman. Women with a history of LCIS were also included because their risk was thought to exceed the cutoff risk of 1.67%. Women were not allowed to use estrogen replacement therapy during their participation in the trial. The results of the P-1 trial indicated that tamoxifen reduced the rates of invasive and noninvasive breast cancer by 49% and 50%, respectively.79 The benefit of tamoxifen was seen in all age groups (≤49 years, 50 to 59 years, and ≥ 60 years). Women with a history of atypical ductal hyperplasia had an 86% risk reduction, and women with a history of LCIS had a 56% risk reduction. The benefit

17  The Breast

was seen across all subgroups specified according to family ­history of breast cancer. Tamoxifen selectively ­reduced the incidence of estrogen receptor–positive tumors; ­ estrogen receptor–negative tumors developed at an equal rate in the tamoxifen and placebo groups. No evidence of a ­cardioprotective effect of tamoxifen was found in this trial, but the number of osteoporosis-related fractures was reduced in the tamoxifen-treated cohort. Tamoxifen ­increased the risk of developing stage I endometrial cancer (risk ratio = 2.53). Although this study indicated that 5 years of tamoxifen use decreased the 5-year risk of developing breast cancer by about 50%, whether this result warrants the widespread use of tamoxifen for breast cancer prevention is controversial. In the NSABP report, 5 years of tamoxifen therapy for 6576 women reduced the number of invasive or noninvasive breast cancer by 120 compared with the expected number.79 Because the follow-up period of this important study was relatively short, however, it was not possible to determine whether tamoxifen prevented this number of cases of breast cancer or instead delayed the onset of the disease.

Raloxifene and Other Newer Agents After completion of the P-1 study, the NSABP began a prospective, randomized prevention trial comparing tamoxifen and raloxifene, a drug originally approved for treatment of osteoporosis that was coincidentally found to be associated with a lower risk of breast cancer development. The Study of Tamoxifen and Raloxifene (STAR or P-2 trial), enrolled 19,474 postmenopausal women and randomly assigned them to receive 5 years of either drug.80 The initial results of this study indicated no differences in the rate of invasive breast cancer development between the two drugs. Tamoxifen was associated with higher risks of developing endometrial cancer and deep venous thrombosis. The number of DCIS cases was ­higher in the raloxifene group (80 cases) than in the tamoxifen group (57 cases). Based on the equivalence in breast ­cancer prevention and the more favorable side effect profile, the NSABP elected to pursue raloxifene as the new standard breast cancer preventive agent for subsequent clinical trials. The positive findings of the P-1 and P-2 trials proved that pharmacologic manipulation of estrogen signaling could effectively prevent breast cancer. In the near future, a host of more selective means of modifying breast cancer development risk is expected to be tested and approved. Preliminary findings from therapeutic trials of breast cancer have indicated that aromatase inhibitors also have significant chemopreventive activity, and some of these agents will be formally compared with raloxifene in the subsequent NSABP chemoprevention study.

361

expected to develop in this population, so the prophylactic surgery resulted in an 89.5% risk reduction (P < .001).81 Prophylactic mastectomy has also proved to be an ­effective risk reduction strategy for women with a ­germline mutation in BRCA1 or BRCA2. In a study of 483 BRCA carriers, the breast cancer development rate was only 1.9% in the 105 women who underwent ­ prophylactic mastectomy compared with 48.7% in those who did not.82 For BRCA carriers, prophylactic oophorectomy also can ­reduce the risk of breast cancer development. In one study, Rebbeck and colleagues83 reported that the rate of breast cancer among 99 BRCA carriers who had undergone ­oophorectomy but not mastectomy was 21%; the comparable rate among 60 matched controls was 42%.

Diagnosis and Staging Diagnosis Breast cancer most commonly manifests as an asymptomatic mammographic finding or an asymptomatic palpable breast mass. All women who present with either of these conditions require pathologic confirmation of the absence or presence of breast cancer. For women presenting with a new palpable breast mass, the initial step in evaluation should be a careful history and physical examination. The breast should be inspected with the patient seated and lying supine and with the pectoralis major muscle contracted and relaxed. The presence of breast asymmetry, skin retraction, erythema, or edema should be documented. Skin edema is a sign of advanced  disease and typically appears as thickened skin with prominent follicles (peau d’orange), particularly when the skin is compressed between two fingers (Fig. 17-4).  A palpable mass should be measured and its consistency described. The axilla and supraclavicular regions should be examined carefully for signs of adenopathy. If the clinical findings suggest breast cancer, bilateral diagnostic mammograms should be obtained. Sonographic examination is also helpful for confirming or ruling out the presence of a simple, fluid-filled cyst. In cases of a palpable mass, diagnostic biopsies are required to rule out malignancy even when radiographic findings are normal. Fine-needle ­ aspiration

Prophylactic Surgery An alternative strategy used to prevent breast cancer ­development is prophylactic surgery. Hartmann and colleagues81 analyzed outcomes in women with a family history of breast cancer who underwent bilateral prophylactic mastectomy at Mayo Clinic between 1960 and 1993. At a median follow-up time of 14 years, only 4 of the 639 ­treated patients had developed breast cancer. According to the Gail model, 37.4 cases of breast cancer would have been

FIGURE 17-4.  Classic appearance of skin edema is shown by thickened skin with prominent hair follicles on mild compression of the skin. This physical finding is called peau d’orange because the appearance is similar to that of the skin of an orange.

362

PART 4  Breast

or core-needle biopsy can be used for this purpose. Fineneedle aspiration is useful for ruling out simple cysts and establishing a diagnosis of malignancy, but because this method yields disaggregated cells rather than intact tissue, it does not allow pathologists to distinguish invasive disease from noninvasive breast cancer. For women presenting with mammographic abnormalities without a palpable mass, tissue can be obtained by means of stereotactic core biopsy or needle localization excisional biopsy. A stereotactic core biopsy is typically done in a radiology department under mammographic guidance. A needle localization biopsy also uses mammograms to localize the area of suspicion but requires an operative procedure. The advantages of stereotactic core biopsy over needle localization excisional biopsy are the avoidance of a breast scar and the avoidance of an additional operative procedure. In most cases in which cancer is diagnosed, excisional biopsy must be followed by re-excision to obtain adequate margins because surgeons are reluctant to perform wide excisions before the establishment of a cancer diagnosis.

Staging After a diagnosis of invasive cancer is established, several tests are helpful for determining the extent of disease. These tests include chest radiography, liver function tests, biochemical profile, and complete blood cell counts. Bone scans and liver imaging with sonography or CT are standard for women with lymph node–positive disease or large primary tumors. In 1960, the International Union against Cancer published the first tumor-node-metastasis (TNM) staging system for breast cancer, which was adopted, with modifications, by the American Joint Committee for Cancer (AJCC) Staging and End Results Reporting in 1962. The most recent revision of the AJCC staging system for breast cancer is the sixth edition, published in 2002 (Box 17-1).84,85 Significant changes incorporated into that revision included the following: • Considering the number of positive lymph nodes in assigning the pathologic N stage • Changing involvement of infraclavicular lymph nodes from N1 to N3 disease • Changing involvement of supraclavicular lymph nodes from M1 to N3 disease The reason for these changes was that each of these revisions allowed the incorporation of prognostically important items. For example, several studies have shown that patients with four or more positive lymph nodes (now considered N2) have a worse outcome than do those with one, two, or three positive lymph nodes (N1 disease).86 Another study indicated that patients with supraclavicular disease without systemic metastases could achieve 10-year survival rates of 31% after aggressive multimodality therapy.87 An interesting consequence of changing the staging system is the change in reported stage-specific survival rates, as seen in a study comparing stage-specific survival of a single group of patients classified under a previous staging system (the third edition, published in 1988) or the revised (2002) staging system. In that study, 45% of patients with stage II breast cancer (defined according to the 1988 staging system) had stage III disease according to the 2002 staging system. This shift resulted in the stage-specific ­survival

rates for patients with disease classified in the 2002 system to seem much better than the rates for patients with disease staged according to the 1988 system.86

Therapy for Ductal Carcinoma in Situ Noninvasive carcinoma of the breast is defined by the presence of malignant cells that are contained within the breast lobules or ducts and have not invaded through the basement membrane. The most common type of noninvasive breast cancer, DCIS, is considered a premalignant condition. The risk that DCIS will progress to invasive breast cancer is difficult to estimate accurately because few data are available on patients with DCIS who received no treatment after biopsy. The studies that address this issue have been hampered by small patient numbers and selection bias. It is likely that 30% to 50% of women with untreated DCIS will develop invasive disease within 10 years after diagnosis. DCIS is distinct from LCIS, the other type of noninvasive breast cancer, because DCIS is a true premalignant condition that can progress to invasive disease, but LCIS is considered a marker of increased risk of breast cancer rather than a true premalignant condition and is more appropriately called lobular neoplasia. Whereas DCIS increases the risk of invasive cancer in the ipsilateral breast, the classic type of LCIS increases the risk of invasive cancer equally in both breasts. The distinction between DCIS and LCIS has important therapeutic relevance. Because LCIS is not a premalignant condition, specific therapy to eradicate the LCIS is not routinely recommended. Rather, careful screening for breast cancer or discussion of prevention strategies such as chemoprevention or bilateral prophylactic mastectomy is appropriate. In contrast, for patients with DCIS, local therapies usually are recommended to eradicate the region of DCIS and minimize the risk of its evolving into an invasive breast cancer.

Common Presentations and Pathologic Subtypes The diagnosis of DCIS has become much more common over the past decade because of the increase in mammographic screening of asymptomatic women. The diagnosis of DCIS was uncommon before the advent of screening mammography, but since screening became routine DCIS has become the fourth most common cancer among ­women (the most common being invasive breast cancer, lung cancer, and colon cancer). Over the past 30 years, the incidence of DCIS has increased at a rate six times that of invasive breast cancer.31 Previously, the most common presentation of DCIS was a palpable mass, but most cases of DCIS seen today are nonpalpable lesions detected during mammographic screening. The most common mammographic findings are branching microcalcifications localized to a small region of the breast. The classic mammographic appearance of DCIS is shown in Figure 17-5. DCIS includes a spectrum of pathologic entities. Welldifferentiated DCIS closely resembles atypical ductal ­hyperplasia of the breast, whereas high-grade DCIS is ­cytologically indistinguishable from invasive breast cancer.

17  The Breast

363

BOX 17-1.  American Joint Committee on Cancer Staging System for Breast Cancer Primary Tumor (T) TX Primary tumor cannot be assessed T0 No evidence of primary tumor Tis Carcinoma in situ Tis (DCIS) Ductal carcinoma in situ Tis (LCIS) Lobular ductal carcinoma in situ Tis (Paget’s) Paget’s disease of the nipple with no tumora T1 Tumor 2 cm or less in the greatest dimension T1mic Microinvasion 0.1 cm or less in the greatest dimension T1a Tumor more than 0.1 but not more than 0.5 cm in the greatest dimension T1b Tumor more than 0.5 cm but not more than 1 cm in the greatest dimension T1c Tumor more than 1 cm but not more than 2 cm in the greatest dimension T2 Tumor more than 2 cm but not more than 5 cm in the greatest dimension T3 Tumor more than 5 cm in the greatest dimension T4 Tumor of any size with direct extension to (a) chest wall or (b) skin only, as described below T4a Extension to the chest wall T4b Edema (including peau d’orange) or ulceration of the skin of the breast, or satellite skin modules confined to the same breast T4c Both (T4a and T4b) T4d Inflammatory carcinoma Regional Lymph Nodes (N) Clinical NX Regional lymph node(s) cannot be assessed (e.g., previously removed) N0 No regional lymph node metastasis N1 Metastasis to movable ipsilateral axillary lymph node(s) N2 Metastases in ipsilateral axillary lymph nodes fixed or matted, or in clinically apparentb ipsilateral internal ­mammary nodes in the absence of clinically evident axillary lymph node metastasis N2a Metastasis in ipsilateral axillary lymph nodes fixed to one another (matted) or to other structures N2b Metastasis only in clinically apparentb ipsilateral internal mammary lymph nodes and in the absence of ­clinically evident axillary lymph node metastasis N3 Metastasis in ipsilateral infraclavicular lymph node(s) with or without axillary lymph node involvement, or in clinically apparentb ipsilateral internal mammary node(s) and in the presence of clinically evident axillary lymph node metastasis; or metastasis in ipsilateral supraclavicular lymph nodes(s) with or without axillary or internal mammary lymph node involvement N3a Metastasis in ipsilateral infraclavicular lymph node(s) N3b Metastasis in ipsilateral internal mammary lymph node(s) and axillary lymph node(s) N3c Pathologic (pN)c pNX pN0 pN0(i−) pN0(i+) pN0(mol−) pN0(mol+) pN1 pN1mi pN1a pN1b pN1c pN2 pN2a pN2b pN3

Metastasis in ipsilateral supraclavicular lymph node(s) Regional lymph nodes cannot be assessed (e.g., previously removed or not removed for pathologic study) No regional lymph node metastasis histologically, no additional examination for isolated tumor cellsd No regional lymph node metastasis histologically, negative immunohistochemical staining Isolated tumor cells identified histologically or by positive immunohistochemical staining, no cluster > 0.2 mme No regional lymph node metastasis histologically, negative molecular findings (RT-PCR)f No regional lymph-node metastasis histologically, positive molecular findings (RT-PCR)f Metastasis in 1 to 3 axillary lymph nodes, and/or in internal mammary nodes with microscopic disease ­detected by sentinel lymph node dissection but not clinically apparentg Micrometastasis (> 0.2 mm, none > 2.0 mm) Metastasis in 1 to 3 axillary lymph nodes Metastasis in internal mammary nodes with microscopic disease detected by sentinel lymph node dissection but not clinically apparentg Metastasis in 1 to 3 axillary lymph nodesh and in internal mammary lymph nodes with microscopic disease detected by sentinel lymph node dissection but not clinically apparentg Metastasis in 1 to 3 axillary lymph nodes, and/or in internal mammary nodes with microscopic disease ­detected by sentinel lymph node dissection but not clinically apparentg Metastasis in 4 to 9 axillary lymph nodes (at least one tumor deposit greater than 2.0 mm) Metastasis in clinically apparent internal mammary lymph nodes in the absence of axillary lymph node metastasis Metastases in 10 or more axillary lymph nodes, or in infraclavicular lymph nodes, or in clinically apparentb ipsilateral internal mammary lymph nodes in the presence of 1 or more positive axillary lymph nodes; or in more than 3 axillary lymph nodes with clinically negative microscopic metastasis in internal mammary lymph nodes; or in ipsilateral supraclavicular lymph nodes continued

364

PART 4  Breast

BOX 17-1.  American Joint Committee on Cancer Staging System for Breast Cancer—cont’d pN3a pN3b pN3c

Metastasis in 10 or more axillary lymph nodes (at least one tumor deposit greater than 2.0 mm), or metastasis to the infraclavicular lymph nodes Metastasis in clinically apparentb ipsilateral internal mammary lymph nodes in the presence of 1 or more ­positive axillary lymph nodes; or in more than 3 axillary lymph nodes and an internal mammary lymph nodes with microscopic disease detected by sentinel lymph node dissection but not clinically apparentg Metastasis in ipsilateral supraclavicular lymph nodes

Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage Grouping Stage 0 Stage I Stage IIA Stage IIB Stage IIIA

Stage IIIB Stage IIIC Stage IV

Tis T1i T0 T1i T2 T2 T3 T0 T1i T2 T3 T3 T4 T4 T4 Any T Any T

N0 N0 N1 N1 N0 N1 N0 N2 N2 N2 N1 N2 N0 N1 N2 N3 Any N

M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M1

aPaget’s

disease associated with a tumor is classified according to the size of the tumor. apparent is defined as detected by imaging studies (excluding lymphoscintigraphy) or by clinical examination or grossly visible pathologically. cClassification is based on axillary lymph node dissection with or without sentinel lymph node dissection. Classification based solely on sentinel lymph node dissection without subsequent axillary lymph node dissection is designated (sn) for sentinel node, such as pN0(i+)(sn). dIsolated tumor cells are defined as single tumor cells or small cell clusters ≤ 0.2 mm, usually detected only by immunohistochemical or molecular methods but that may be verified on hematoxylin and eosin stains. Isolated tumor cells do not usually show evidence of metastatic activity (e.g., proliferation, stromal reaction). eDefinition of (i+) was adapted in 2003 to be consistent with the updated International Union against Cancer (UICC) classification. fThe reverse transcriptase–polymerase chain reaction (RT-PCR) is used. gNot clinically apparent is defined as not detected by imaging studies (excluding lymphoscintigraphy) or by clinical examination. hIf associated with more than three positive axillary lymph nodes, the internal mammary nodes are classified as pN3b to reflect the increased tumor burden. iT1 includes T1mic. Modified from Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 223-240; Singletary SE, Connolly JL. Breast cancer staging: working with the sixth edition of the AJCC Cancer Staging Manual. CA Cancer J Clin 2006;56:37-47. bClinically

A pathology consensus conference in 1997 recommended that all cases of DCIS be categorized to reflect the biologic potential of the disease.88 Accordingly, information on nuclear grade, the presence or absence of necrosis, tissue architecture, and polarization or radial orientation should be provided. Tissue architecture can range from the more indolent papillary, micropapillary, and cribriform patterns to the more biologically aggressive solid and comedo subtypes. Comedo necrosis represents necrotic debris centrally located in the duct. An example of high-grade DCIS with comedo necrosis is shown in Figure 17-6.

Treatment BCT has become the standard treatment for women with DCIS. Mastectomy is reserved for the less common cases that manifest with diffuse microcalcifications throughout

the entire breast or large volumes of palpable disease or for cases in which surgical margins free of carcinoma cannot be obtained with breast-conserving surgery. A randomized clinical trial has never been designed to directly compare mastectomy with BCT for patients with DCIS. In the NSABP B-06 trial, which compared mastectomy and BCT for women with invasive breast cancer, a small percentage of women enrolled were later recognized to have DCIS. Within this relatively small subset of ­patients, an overall survival rate of 96% was the same for the conservatively treated patients and for the patients treated with mastectomy.89 This very low probability of breast cancer death after BCT for DCIS has been confirmed in several subsequent publications, and a ­ randomized trial comparing this treatment against mastectomy is therefore ­unnecessary.

17  The Breast

365

TABLE 17-4.   Rates of Recurrence in the Ipsilateral Breast in the NSABP B-17, EORTC 10853, UK ANZ, and SweDICS Trials of Treatment for Ductal Carcinoma In Situ of the Breast Recurrence Rate (%) Trial NSABP

No Radiation B-1792

Overall

Radiation

P Value

16

<.000005

(12-Year Results) 32

Invasive

17

8

<.0001

Noninvasive

15

8

.001

EORTC 1085393 (10-Year Results) FIGURE 17-5.  In the typical mammographic appearance of untreated ductal carcinoma in situ of the breast, the microcalcifications can be described as clustered, pleomorphic, and branching.

Overall

26

15

<.0001

Invasive

13

8

.0065

Noninvasive

14

7

.0011

UK ANZ94 (5-Year Results) Overall

14

6

<.0001

Invasive

6

3

.01

Noninvasive

7

3

.0004

7

<.0001

SweDCIS95 Overall

(5-Year Results) 22

EORTC, European Organisation for Research and Treatment of ­Cancer; NSABP, National Surgical Adjuvant Breast and Bowel Project.

Randomized Trials

FIGURE 17-6.  Histologic section shows a high-grade ductal carcinoma in situ with extensive comedo necrosis.

Role of Radiation Therapy in Breast-Conservation Therapy With the historical change from mastectomy to BCT as the preferred therapeutic approach for DCIS, controversy has arisen over the need for radiation therapy and tamoxifen as standard components of treatment. Although data from randomized trials are available that address both of these questions, these data have not led to resolution of the current controversies. For example, despite clear and convincing evidence from phase III randomized studies, which indicate a benefit from breast radiation therapy ­after surgery, patterns of care data suggest that radiation is often not used. Specifically, an analysis of the Surveillance, Epidemiology, and End Results (SEER) database found that in 1999, years after the publication of the  B-17 study, only 54% of patients with DCIS received ­radiation therapy after breast conservation surgery.90 Similar practice patterns are seen elsewhere in the world. A survey of Australian and New Zealand physicians found that only 22% of patients treated with breast-conserving surgery for DCIS were referred for radiation oncology consultations.91

Findings from the large randomized studies conducted on the need for radiation in the treatment of DCIS are summarized in Table 17-4,92-95 and their salient points are discussed briefly in the following paragraphs. NSABP B-06 Trial. The first data defining the role of radiation therapy in the management of DCIS came from the NSABP B-06 trial. The B-06 trial was designed for patients with invasive disease, but 76 patients were subsequently found to have pure DCIS with no evidence of invasion. At a median follow-up time of 83 months, the local recurrence rates were 7% for patients treated with lumpectomy and irradiation and 43% for patients treated with lumpectomy alone (P = .01).89 NSABP B-17 Trial. Although data from the B-06 trial provided the first suggestion that radiation has a critically important role in BCT for DCIS, that trial was specifically designed to study invasive disease. The NSABP therefore began a new trial in 1985 that was designed to define the role of radiation therapy in the treatment of DCIS. The B-17 trial was a dual-arm prospective trial that enrolled 818 women who were randomly assigned to receive or not receive radiation therapy after breast-conserving surgery.96 Negative surgical margins were a requirement for eligibility, but postoperative mammograms or specimen radiographs were not. The overall survival rates at 12 years were equivalent in the two treatment groups. Radiation therapy reduced the 12-year rates of ipsilateral breast cancer recurrence from 32% to 16% (P < .000005).92 More than one half of the patients who developed recurrent

366

PART 4  Breast

disease in the ipsilateral breast after surgery alone had invasive disease. The NSABP B-17 study included a wide spectrum of pathologic subtypes of DCIS. It has been argued that the results of this trial are most relevant to patients with ­disease at the more biologically aggressive end of the spectrum. To address this criticism, the NSABP carefully analyzed the patient and pathologic characteristics of the trial to determine whether the trial was representative of the types of cases currently being diagnosed.91,97 With respect to presenting characteristics, 83% of the patients presented with nonpalpable disease, and less than 5% of patients had disease larger than 2 cm, which was detected by gross assessment of the pathology specimen or by mammography. Approximately 50% of the study participants had DCIS diagnosed from a 1-cm or smaller cluster of microcalcifications found on routine mammography. The benefit of ­radiation therapy was significant for tumors measuring  1.0 cm or smaller and tumors larger than 1.0 cm.96 In the 623 cases for which central pathology review was possible, the presence and degree of comedo necrosis was the strongest predictor of recurrence in both treatment groups. Most importantly, radiation led to reduction in disease recurrence in the breast for the patients with high-risk pathologic features and the patients with low-risk pathologic features.97 The B-17 trial also has been criticized for the fact that even though negative margins were required, margins were considered clear even if a single normal cell separated the disease from the margin. EORTC 10853 Trial. The European Organisation for Research and Treatment of Cancer (EORTC) completed an important randomized, prospective clinical trial that compared breast-conserving surgery with or without radiation therapy for patients with DCIS (trial 10853).98 More than 1000 patients with DCIS smaller than 5 cm that was treated with breast-conserving surgery were randomly ­assigned to radiation therapy or to no further therapy. In this trial, re-excision was required for the 34% of patients with initial positive margins. The final margin status was positive in only seven cases, although the definition of what constituted a negative margin was not provided. ­Approximately 20% of the patients presented with palpable disease, with a mean tumor diameter of 2.1 cm. The results from this study confirmed those of the B-17 trial. The use of radiation therapy resulted in an improvement in the 10-year rates of breast recurrence from 26% to 15% (P < .0001) and the rates of invasive recurrences from 13% to 8%  (P = .0011).93 UK ANZ Trial. A third large randomized trial that evaluated the benefit of radiation therapy and tamoxifen after breast conservation surgery for patients with DCIS was conducted in the United Kingdom, Australia, and New Zealand.94 That trial, which enrolled 1701 patients, further demonstrated that radiation therapy reduced the risk of overall breast cancer recurrence (HR = 0.38, P < .0001) and invasive breast cancer recurrence (HR = 0.45, P = .01). The use of tamoxifen alone did not confer a benefit. SweDCIS Trial. The fifth and perhaps final randomized study of the role of radiation in DCIS was conducted by the Swedish Breast Cancer Group.95 Of the 1067 enrolled patients, 80% had mammographically detected DCIS, and 90% had negative margins. At a median follow-up time of 5 years, the cumulative incidence of breast recurrence

was 22% for the group that underwent surgery only and. 7% for the group that received surgery plus radiation  (P < .0001).95 Meaningful benefit from radiation therapy was evident in all clinicopathologic subgroups.99

Retrospective Studies Van Nuys Data. In 1995, Silverstein and colleagues,100 with the hope of defining the optimal treatment strategy for DCIS according to clinical and pathologic characteristics of the disease, developed the Van Nuys Prognostic Index. Their classification was based on retrospective analysis of 333 patients treated at two institutions between 1979 and 1995. The index incorporated tumor grade and the presence or absence of necrosis (combined into one category), tumor size, and margin status. At a median follow-up time of 8 years, the probability of local recurrence after surgery alone for the 76 patients with a prognostic index score of 3 to 4 was very low. However, radiation therapy clearly ­reduced the local recurrence rates for patients with index scores of 5 or higher. A subsequent update of this study that included 469 patients with DCIS treated with breastconserving surgery with or without radiation therapy101 showed that margin status was highly predictive of outcome. For patients with tumor margins of 10 mm or more, the local recurrence rates at 8 years were only 4% for the  40 patients who received radiation therapy and 3% for the 93 patients treated with surgery alone. However, for ­women who received surgery only, the 8-year local recurrence rates rose to 20% when margins were 1 to 10 mm and 58% when margins were less than 1 mm. When radiation was used, the corresponding recurrence rates were 12% and 30%.101 An update102 of this influential study indicated that at 12 years, a benefit from radiation was evident even for patients with the widest surgical margins (≥10 mm),  with local recurrence rates of 1% for the 60 patients given irradiation and 14% for the 212 patients not given radiation.102 The latest Van Nuys index includes patient age after this group’s discovery that young age independently predicts for breast recurrence after surgery without radiation therapy.103 Although these findings are a valuable contribution to the literature on DCIS, they must be interpreted within the limitations of the studies that generated them. First, the  data were obtained from two institutions that used ­special pathologic handling procedures that are not routinely used at all United States hospitals. The studies were retrospective and the use of radiation was not randomized. The numbers of patients in the comparative subcategories were small, and the confidence intervals for those data were large.

Single-Arm Prospective Trials In another effort to identify some subgroup of patients with DCIS who could be safely treated with surgery alone, investigators at the Dana-Farber/Harvard Cancer ­Center104 conducted a single-arm prospective trial in which the patients had mammographically detected low- or ­ intermediategrade DCIS that had been treated with wide excision with margins of at least 1 cm or a re-excision without evidence of residual disease (78% of the participants were in the ­latter group).104 The original accrual goal was 200 patients, but the study was stopped after accrual of 158 ­ patients because the predefined stopping boundary for breast 

17  The Breast

recurrences had been crossed. Specifically, the rate of ­ipsilateral local recurrence was 2.4% per patient-year, for an estimated 5-year rate of recurrence of 12%. This rate was much higher than that reported by the Van Nuys group and more closely approximated the rates of recurrence in the surgery-alone groups of the randomized, prospective trials described in the previous section. The Eastern Oncology Collaborative Group completed a much larger single-arm prospective trial investigating surgery alone for DCIS with favorable disease characteristics. In that study,105 the 5-year breast recurrence rate for 580 women with low- or intermediate-grade DCIS that was smaller than 2.5 cm and had negative margins at least 3 mm wide was 6.8%; 50% of these recurrences were ­invasive. By comparison, the 5-year ipsilateral breast recurrence rate in 102 patients with high-grade DCIS that was smaller than 1 cm and had at least 3-mm-wide negative margins was 13.7%.105 These findings further underscore the importance of including radiation therapy in breast­conserving treatment for DCIS.

Role of Hormonal Therapy After the B-17 trial, the NSABP began the B-24 trial to investigate the role of tamoxifen therapy for patients with DCIS.106 Patients eligible for this trial were women who had undergone breast-conserving surgery for DCIS. Unlike the B-17 trial, the B-24 trial included patients with involved surgical margins and patients with suspicious scattered calcifications. Approximately 25% of the patients in the study had positive or unknown margin status. All women received radiation therapy consisting of 50 Gy to   the breast. More than 1800 patients were then randomly assigned to receive tamoxifen or placebo for 5 years. The primary end point of the trial was breast events, which ­included the development of invasive and noninvasive ­ipsilateral and contralateral disease. The 5-year results of the study indicated that the tamoxifen group had fewer breast cancer events (8.2% versus 13.4% for the placebo group, P = .0009). The rate at which invasive breast cancer developed was also reduced in the tamoxifen group, as was the risk of ipsilateral breast cancer recurrence.106 After publication of the B-24 results, investigators from the NSABP retrospectively investigated available tissue blocks (n = 628) for the presence or absence of the estrogen receptor. That analysis revealed that the benefits of tamoxifen were limited to patients with DCIS that was ­estrogen receptor–positive.107 Specifically, the relative risk of breast events for patients with estrogen receptor–­positive DCIS who were treated with tamoxifen was 0.41 compared with placebo (P = .0002), whereas the relative risk of breast events for patients with estrogen receptor–negative disease who were treated with tamoxifen was 0.80 compared with placebo (P = .51). Staining tumor specimens for the presence of estrogen or progesterone receptors has since become routine and an is an accepted component of the standard of care in DCIS. Unlike the B-24 trial, the UK ANZ trial did not find that tamoxifen reduced invasive ipsilateral breast recurrences among patients treated with lumpectomy and radiation,94 but it did seem to reduce the risk of DCIS ­ recurrence in such patients. The potential benefit of tamoxifen as a function of estrogen receptor status was not tested in that study.

367

The use of tamoxifen for patients with DCIS has been sporadic. In an analysis of patients treated after publication of the B-24 study results, investigators from The University of Texas M. D. Anderson Cancer Center found that only 23% of those patients had been treated with adjuvant tamoxifen. The reasons tamoxifen was not used were physicians not offering the option (40%), patients choosing to not use tamoxifen after it had been offered (27%), and discontinuation because of side effects (10%).108 Tamoxifen has therapeutic and chemopreventive value in DCIS. However, in terms of its therapeutic value, the inclusion of patients with unknown or positive margins after surgery in the B-24 trial may make the findings of that trial less relevant for patients with DCIS and negative surgical margins. In the subset of women with negative surgical margins in that trial, tamoxifen treatment prevented or delayed the onset of only two recurrences in the ipsilateral breast and 1.4 cases of contralateral breast cancer over a 5-year period for every 100 women treated. Fewer than 3.5% of patients with DCIS and negative surgical margins therefore would be expected to benefit from tamoxifen use during a 5-year period. Because the clinical benefit seems to be low, the use of tamoxifen for women with DCIS and negative margins remains controversial. Newer forms of hormonal therapy under investigation for DCIS include aromatase inhibitors such as anastrozole (Arimidex), exemestane (Aromasin), and letrozole ­(Femara). These agents are discussed under “Hormonal Therapy.” In the B-35 trial, the NSABP compared 5 years of tamoxifen with 5 years of anastrozole for postmenopausal women with DCIS treated with lumpectomy and radiation. The findings from this trial are eagerly awaited.

Therapy for Early-Stage Breast Cancer Several significant changes over the past four decades have significantly improved the management of early-stage (stage I and II) breast cancer. The increased use of mammographic screening and heightened efforts in public education about breast cancer have dramatically increased the percentage of cases of breast cancer diagnosed at noninvasive or early stages of disease.27 Simultaneously, limited breast surgery followed by radiation therapy was found to result in outcomes equivalent to the historical treatment of a modified radical mastectomy for women with earlystage disease. This organ-preserving approach to treatment was found to have a positive impact on patient well-being and quality of life.109 Advances in radiation techniques and axillary surgery have reduced the morbidity of treatment of axillary lymph nodes. Perhaps even more important has been the introduction of systemic treatment for the majority of women with early-stage breast cancer. It is clear that systemic treatment reduces the odds of cancer recurrence for all women with invasive breast tumors larger than 1 cm in diameter, independent of lymph node status.110,111

Equivalence of Breast-Conservation Therapy and Modified Radical Mastectomy After single-institution reports were published describing the successful use of breast-conserving surgery and radiation therapy as an alternative to mastectomy for early-stage

368

PART 4  Breast

breast cancer, several phase III clinical trials were developed and conducted to directly compare these two approaches. Between 1976 and 1984, the NSABP conducted the largest of these trials, the B-06 trial. In that trial, a total of 1600 women were randomly assigned to receive one of three locoregional treatments: modified radical mastectomy, lumpectomy plus axillary dissection plus breast irradiation, or lumpectomy plus axillary dissection alone.23 All patients treated with breast-conserving surgery were required to have negative surgical margins after lumpectomy, defined in this study as no tumor being present at the inked margin of resection. Breast irradiation consisted of 50 Gy in 2-Gy fractions delivered to the whole breast with no tumor bed boost. The trial was open to women with primary tumors up to 4 cm in diameter. Women with positive lymph nodes were treated with chemotherapy, whereas women with negative lymph nodes did not receive systemic treatment. The 20-year outcome data conclusively demonstrated that treatment with lumpectomy and radiation therapy was equivalent to modified radical mastectomy in terms of disease-free and overall survival. The trial did not statistically compare the 20-year rates of breast failure (14.3%) or lymph node failure (2.7%) for the patients treated with lumpectomy and radiation against the rates of chest wall failure (10.2%) or lymph node failure (4.6%) in the patients treated with mastectomy.112 The Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) performed a meta-analysis of the data from NSABP B-06 plus six other randomized trials comparing modified radical mastectomy to breast-conserving surgery plus radiation therapy for early-stage breast cancer.113 Their analysis, which included data from 3100 women, revealed no difference in overall survival according to treatment type, with both treatment groups having a 10-year survival rate of 71%.113 A second meta-analysis by a different group that focused only on the outcome of breast-conserving surgery and radiation therapy versus modified radical mastectomy for patients with lymph node–positive disease114 found the two treatments to be equivalent with respect to overall survival. The interesting finding from this second meta-analysis was that BCT offered a survival advantage over mastectomy for patients with lymph node–positive disease in the combined analysis of trials in which postmastectomy radiation was not used (odds ratio [favoring BCT] of 0.69). In the studies that compared BCT against modified radical mastectomy with postoperative irradiation, the two treatments achieved equivalent outcomes. These findings suggest that radiation therapy should be a component of care for women with high-risk lymph node–positive breast cancer.

Role of Radiation Therapy in Breast-Conservation Therapy After it was established that limited breast surgery combined with a course of breast irradiation achieved outcomes equivalent to those of treatment with mastectomy, several trials studied whether radiation therapy was needed as a component of BCT. In the NSABP B-06 trial, patients treated with lumpectomy and axillary dissection alone were found to have a significantly higher rate of recurrence in the ipsilateral breast than those treated with lumpectomy and axillary dissection followed by radiation. The  20-year actuarial local recurrence data from this trial

r­ evealed an ipsilateral breast recurrence rate of 39% in the lumpectomy-alone group compared with 14% in the lumpectomy plus radiation therapy group (P < .001).112 These data were also included and confirmed in the metaanalysis compiled by the EBCTCG.25 A subsequently published update of the results of this meta-analysis, which included more than 7300 women treated in randomized clinical trials, showed that breast radiation after lumpectomy reduced the 10-year rates of breast recurrence from 29% to 10% for patients with negative lymph nodes and from 47% to 13% for patients with positive lymph nodes. Most importantly, these benefits led to a statistically significant decrease in the 15-year risk of dying from breast cancer. For patients with negative lymph nodes, the breast cancer mortality rate was reduced from 31% to 26%, and for patients with positive lymph nodes, the breast cancer mortality rate was reduced from 55% to 48%. The data from these analyses are shown in Figure 17-7.25 These findings demonstrate conclusively that an improved locoregional treatment approach that eradicates persistent locoregional disease can lead to long-term overall improvements in survival, implying that persistent locoregional disease can be a cause of and a marker for subsequent distant metastases. The data suggest that one of every four patients for whom locoregional recurrence is avoided through the use of radiation therapy will avoid a breast cancer death. The other three fourths of patients will be successfully treated for locoregional recurrence or die of metastatic disease that was not prevented by eradicating persistent locoregional disease. Most subsets of patients treated with breast conservation experience benefit from breast radiation therapy. Some important results of randomized trials that have evaluated specific subsets of women treated with BCT are summarized here. In one such study, a single-arm prospective trial of 82 patients at Harvard, the attempt to define a favorable subset of patients for whom breast radiation  could be omitted was stopped early because the breast recurrence rate exceeded the predefined stopping rules. The local recurrence rate after a median follow-up ­interval of 86 months was 23%.115 The question of whether ­radiation therapy is needed after excision of a small tumor with widely negative margins was addressed in a trial from ­Milan that included women with tumors 2.5 cm or smaller. The comparison in that trial was between quadrantectomy plus axillary dissection and quadrantectomy plus axillary dissection plus breast irradiation. In that trial, radiation provided a highly significant reduction in the 10-year risk of breast recurrence despite the extensive nature of the surgical procedure (6% for the radiation group versus 24% for the no-radiation group, P < .001).24 Randomized trials from Sweden and Finland have specifically addressed whether patients with stage I disease require radiation therapy after breast-conserving surgery. Both studies indicated that  radiation led to highly significant reductions in breast ­recurrence.116,117 Systemic treatment with chemotherapy or tamoxifen does not obviate the need for breast radiation. In the NSABP B-06 trial, chemotherapy was used for patients with lymph node–positive disease; for those patients, the  20-year risk of breast recurrence without radiation was 44%, compared with only 9% for patients with positive lymph nodes who were treated with lumpectomy, ­ 

17  The Breast

369

6097 WOMEN WITH BCS AND NODE-NEGATIVE DISEASE 60

5-year gain 16.1% (SE 1.0)

50

BCS BCS+RT

Breast cancer mortality (%)

Isolated local recurrence (%)

60

40 29.2

30

22.9

20 10 6.7

0 0

5

15-year gain 5.1% (SE 1.9) Logrank 2p=0.006

50

BCS BCS+RT

40 31.2% 30 20.3 20 8.9

10

10.0

26.1%

17.4

8.0 0

10

15

0

5

Time (years)

10

15

Time (years) 1214 WOMEN WITH BCS AND NODE-POSITIVE DISEASE 60

5-year gain 30.1% (SE 2.8) 46.5

50

Breast cancer mortality (%)

Isolated local recurrence (%)

60

41.1 40 BCS BCS+RT

30 20 10

11.0

13.1

15-year gain 7.1%(SE 3.6) Logrank 2p=0.01

50

55.0%

45.2 47.9%

40 36.5

30

24.3

20

20.9 BCS BCS+RT

10 0

0 0

5

10

15

Time (years)

0

5

10

15

Time (years)

FIGURE 17-7.  The Oxford meta-analysis of trials investigating breast-conserving surgery (BCS) with or without breast radiation shows that radiation therapy (RT) significantly reduces isolated local recurrence and reduces breast cancer mortality for patients with lymph node–negative disease and for those with lymph node–positive disease. (From Early Breast Cancer Trialists’ Collaborative Group. Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomised trials. Lancet 2006;366:2087-2106.)

radiation, and chemotherapy.112 In a Scottish randomized trial in which all patients received systemic treatment (with most receiving tamoxifen as the sole systemic treatment), radiation also reduced breast recurrences.118 A similar ­finding was observed in a Canadian trial that included women older than 50 years with T1 N0 disease treated with breast conservation and tamoxifen; in that study, the 8-year recurrence rate in the radiation group was 4%, compared with 12% in the no-radiation group.119 In the NSABP  B-21 trial, women with lymph node–­negative breast ­tumors measuring less than 1.0 cm that were treated with lumpectomy and axillary lymph node dissection were randomly assigned to receive tamoxifen alone, radiation ­therapy alone, or tamoxifen and radiation.120 Despite the very favorable disease characteristics required for ­inclusion

in this study, radiation reduced the risk of breast recurrence over tamoxifen treatment alone. These findings ­ indicate that systemic treatments should not be considered alternatives to radiation therapy but rather may have synergistic effects with radiation therapy in further reducing the rate of ipsilateral breast recurrences.

Breast-Conservation Therapy for Women Older than 70 Years with Stage I Disease The one group of patients for whom breast conservation without radiation may be an appropriate option are elderly women with estrogen receptor–positive T1 N0 disease. An Intergroup trial randomly assigned women 70 years old or older with those disease characteristics to undergo breastconserving surgery plus tamoxifen or breast-­conserving

370

PART 4  Breast

surgery, tamoxifen, and breast irradiation. The 5-year  results of this trial indicated a small but significant benefit from radiation, but the recurrence rate in the tamoxifenonly group was only 4.4%.121 Notably, 16% of the patients enrolled in this study died of intercurrent disease despite the relatively short follow-up period of 5 years; how these findings should be applied to women older than 70 years with a longer life expectancy is less clear. An update of this trial indicated that at a median follow-up time of 7.9 years, radiation had reduced the locoregional recurrence rate from 7% to 1%.122 To further investigate this question, investigators from Yale University reviewed the SEER-Medicare database and identified 8724 patients who met the eligibility criteria for this trial. The outcomes at 5 years were similar to those reported in the Intergroup trial.123 However, for patients 70 to 79 years old who had no comorbid conditions, the benefits of radiation were more pronounced and of a clinically relevant magnitude. In conclusion, all of the clinical studies have indicated that breast irradiation reduces the risk of local recurrence after breast-conserving surgery and therefore should be considered a standard component of treatment for all women with early-stage invasive disease. All attempts to identify subsets of breast cancer patients with favorable early-stage disease who may not require radiation therapy have been unsuccessful.

Contraindications for Breast-Conservation Therapy Not all cases of early-stage breast cancer can be treated with BCT. BCT is contraindicated when the primary ­tumor cannot successfully be removed with a procedure less ­extensive than mastectomy, such as cases in which mammography reveals diffuse suspicious microcalcifications throughout the breast or cases in which repeated attempts at breast-conserving surgery fail to achieve negative surgical margins. BCT is relatively contraindicated when a more optimal aesthetic result can be achieved with mastectomy and breast reconstruction, such as when the ratio of the primary tumor size to the breast size is large or a central segmentectomy with resection of the nipple-areola complex is required to achieve negative margins. BCT is also contraindicated for patients in whom the potential for ­severe injury from radiation therapy is high. Examples are women who are pregnant and women with certain connective tissue diseases. For women early in the course of their pregnancy, internal radiation scatter from irradiation of the intact breast can reach lethal and teratogenic dose levels.124 For this reason, it is critical that all women of childbearing age be counseled against becoming pregnant during the course of radiation therapy. Women with connective tissue disorders may be at increased risk for significant late radiation complications, including significant breast fibrosis and pain, chest wall necrosis, and brachial plexopathy. The collagen vascular diseases associated with the greatest risk for radiation injury are systemic scleroderma, lupus erythematosus, polymyositis or dermatomyositis, and mixed connective tissue disorders. In a study of 209 patients from Massachusetts General Hospital in Boston, 131 patients with rheumatoid arthritis seemed not to have elevated rates of late toxic

e­ ffects,125 suggesting that BCT may not be contraindicated for patients with rheumatoid arthritis. A case-controlled matched analysis from Yale University also showed that scleroderma was the only connective tissue disorder that seemed to be associated with an increased risk of severe complications after radiation therapy delivered as part of BCT.126

Factors Affecting Outcomes of Breast-Conservation Therapy Several patient-, tumor-, and treatment-related factors have been found to influence local recurrence rates after BCT.

Patient Factors Age. The clearest example of a patient factor that influences local and distant recurrence rates after BCT is patient age. Several single-institution studies have reported that young patient age, usually defined as an age younger than 30, 35, or 40 years, is associated with an increased risk  for distant metastases and reduced disease-specific sur­ vival.92,97,98 Young age has been shown to increase the risk of recurrence in the ipsilateral breast after BCT.127-132 A trial  of 5387 patients that investigated the benefit of a radiation boost after whole-breast irradiation reported patient age as being an independent predictor of breast recurrence. The 5-year rate of breast recurrence for patients 40 years or younger was 15%, compared with 7% for patients 41 to 50, 4% for those 51 to 60, and 3% for patients more than 60 years old.133 Another study evaluated the outcome of 196 women younger than 40 years old treated with BCT and found that an age younger than 35 was independently predictive of ipsilateral breast recurrence.134 Because age may also affect local recurrence rates after mastectomy, investigators from M. D. Anderson Cancer Center retrospectively evaluated the locoregional treatment outcome of 668 breast cancers in patients  35 years old or younger.135 Patients with stage I disease who were treated with chemotherapy had acceptable loco­regional disease control with BCT or mastectomy, but for patients with stage II disease, the 10-year locoregional recurrence rates were lower for those treated with mastectomy followed by radiation (6%) than for those treated with breast conservation (18%) or mastectomy without  radiation (23%). At the other end of the age spectrum, breast cancer is considered by some to be a more indolent disease in the elderly population (65 years or older), and the use of ­ radiation therapy and systemic treatment after BCT ­decreases as ­ patient age increases.136,137 However, ­ Solin and colleagues138 suggest that breast cancer is not an ­indolent disease in the elderly population and that decisions concerning treatment recommendations should not be based on the patient’s chronologic age. These investigators compared the 10-year outcomes of BCT in women at least 65 years old and women 50 to 65 years old and reported that the two groups had equally virulent disease (determined by stage and prognostic factors) and equivalent rates of breast cancer death.138 Race. Data are limited regarding the influence of race on the success of BCT. The data that do exist mostly concern African Americans; virtually no data are available concerning Hispanics, Asian Americans, and other ethnic groups. One of the larger series to consider the success of

17  The Breast

l­ocoregional therapy in African American women was reported by Newman and colleagues139 from M. D. ­Anderson. In that series of 210 women with primary ­tumors ­smaller than 5 cm, local recurrence rates were similar for the 42 women treated with BCT (9.8%) and the 168 women treated with mastectomy (8.9%).139 Family History and Genetic Predisposition. Several retrospective studies have investigated the influence of family history or a genetic predisposition for breast cancer on the success of BCT. Investigators from Harvard’s Joint Center for Radiation Therapy (JCRT) compared local control rates for 29 women 36 years old or younger who had a first-degree relative with breast cancer before age 50 or a family history of ovarian cancer and for 172 women, also 36 years old or younger, who did not have these family history criteria.140 The 5-year crude local recurrence rates were statistically similar for the patients with a positive family history and those with a negative family history (3% versus 14%). However, the patients with positive family histories had a 5.7-fold increased risk of developing contralateral breast cancer compared with patients with a negative family history.140 The impact of family history and young age was also investigated in a series of 1021 patients treated at the University of Pennsylvania.141 As was the case in the JCRT experience, no difference in local control was found based on family history of breast cancer in young patients. For patients 40 years old or younger, the 5-year local failure rate was 8% for those with a first-degree relative with breast cancer, 2% for those with a non–first-degree relative with breast cancer, and 12% for those with a negative family history (P ≥ .18). Less is known of the outcomes after BCT for ­individuals with known predisposing germline mutations. ­Investigators at Yale University sequenced the BRCA gene in 127 ­patients aged 42 or younger who were treated with lumpectomy and radiation and found that 22 had deleterious mutations.142 At 12 years after treatment, the rates of ipsilateral breast recurrence (49% versus 21%, P = .007) and contralateral events (42% versus 9%, P = .001) were both significantly higher among the patients with BRCA mutations.142 ­However, Pierce and colleagues at the ­University of ­ Michigan at Ann Arbor143 compared the outcome of 160 carriers with 445 matched controls and found no significant differences in breast recurrence rates between the carriers and the controls. Those investigators suggested that the reason for the low rate of breast recurrences was ­because a large number of women in that series had undergone oophorectomy. Among the carriers who had not had a bilateral oophorectomies, the rate of breast recurrences was increased (HR = 1.99, P = .04).143 These results are consistent with an earlier finding that oophorectomy is protective against breast cancer development in BRCA carriers.144

Tumor and Pathology Factors Surgical Margins. One of the most important pathologic factors that affect local control after breast­conserving surgery and radiation therapy is the status of the surgical ­margins. It has become standard practice to ink surgical specimens and report margin status quantitatively  in all ­ cases of invasive breast cancer treated with breast­conserving surgery. Recognition of the importance of margin status in predicting local recurrence has led to increased rates of re-excision to obtain clear surgical margins.

371

Although margin status is clearly predictive for breast tumor recurrence, controversy remains regarding how a “negative” margin should be defined. The NSABP B-06 trial defined a negative margin as the absence of tumor cells on the ink-defined margin; when pathologist-reviewers agreed that the surgical margins were involved, residual disease was found at subsequent mastectomy in 62% of cases.145 Other groups have defined surgical margins more precisely by measuring the total distance from the closest malignant cell to the ink and by further distinguishing focally positive margins from margin involvement over a wider area. No consensus has been reached regarding the optimal definition of close margins. In one analysis of 1021 patients,146 investigators at the University of Pennsylvania reported that local failure rates were not significantly different for patients with negative margins (n = 518), focally close margins (n = 96), focally positive margins (n = 124), or unknown margins (n = 283). The local failure rates for those groups at 8 years were as follows: negative, 8%; focally positive, 10%; unknown margin status, 16%; and focally close, 17% (P = .21).146 The ­actuarial curves were nearly identical over the first 4 years and began to separate thereafter. A shortcoming of this study, the limited number of patients who had close or positive margins and were followed for more than 5 years, renders uncertain whether patients with positive or close margins have higher rates of late local recurrence. A later article that also addressed the extent of margin involvement reviewed slides from 607 breast cancer cases treated with BCT and reclassified the margins according to the distance of the margin and volume of disease at that margin. The results revealed that both factors were related to outcome, with the range of recurrence at 12 years ranging from 9% (negative), to 24% (large volume that was close to margin) to 30% (positive margin).147 For this reason, the recommended treatment policy is re-excision for all patients with positive or unknown margin status and consideration of reexcision for women younger than 40 years who have close margins (<2 mm). For patients with positive margins, the extent of margin involvement probably influences the total risk of recurrence. Wazer and colleagues148 at the New England Medical Center in Boston evaluated 105 patients with ­positive surgical margins after breast-conserving surgery. The 73 patients with focal or minimal margin involvement had a crude local recurrence rate of only 2.8% (median followup time, 75 months), compared with a rate of 22.2% for the 27 patients with moderate or extensive margin involvement (median follow-up time, 88 months).148 The median interval to recurrence in this series was 64 months, and the number of patients at risk at 10 years was very small. Margin status may also be related to the extent of intraductal disease within or adjacent to invasive carcinoma. In 1990, investigators at the JCRT reported that women with tumors that had an extensive intraductal component (EIC), defined as tumors that are predominantly noninvasive or tumors with a DCIS component comprising at least 25% and with DCIS present in surrounding normal breast tissue, are at increased risk of local recurrence.149 However, in a subsequent analysis,150 the increased risk ­associated with an EIC was found to be limited to patients with positive margins. None of the 18 patients with an EIC who had negative, close, or focally positive margins

372

PART 4  Breast

experienced a breast tumor recurrence, whereas 6 of the 12 patients with more than a focally positive margin and an EIC had a recurrence.150 These findings suggest that an EIC increases the difficulty of achieving negative margins and, in selected cases, may represent a more diffuse disease process. However, patients with an EIC in whom negative margins are achieved should be considered appropriate candidates for BCT. Several cofactors interact with margin status in deter­ mining outcome. Park and colleagues151 reported that patients with focally positive margins who were treated with systemic therapy had a local recurrence rate of 7% at 8 years, whereas higher rates were seen for those not treated with systemic therapy or those with more diffuse positive margins.151 However, other findings from Fox Chase Cancer Center in Philadelphia suggest that systemic therapy delays rather than prevents local recurrences that are associated with close or positive margins.152 A group from the Netherlands showed that margin status was of particular importance for women 40 years old or younger. In that group, the risk of breast recurrence according to margin status was 37% (positive) versus 8% (negative).153 A group at Harvard also found that patients with close or positive margins had a poor outcome if radiation was ­delayed until after adjuvant chemotherapy, but if negative margins were achieved, delaying radiation had no adverse effect on local control.115 Wazer and colleagues154 reported the results of a treatment policy in which the boost radiation dose to the tumor bed was chosen according to the margin status. After a dose of 50.4 Gy to the breast, the tumor bed was boosted as follows: negative margins after re-excision, 0 Gy; negative margin greater than 5 mm, 10 Gy; negative margin of 2 to 5 mm, 14 Gy; and negative margin less than 2 mm, 20 Gy. For women 45 years old or older, this policy resulted in ­excellent local control. However, for women younger than 45 years, the failure rates remained proportional to the ­final margin status.154 Collectively, these findings suggest that margin status correlates with long-term local control for patients treated with BCT. Recurrences may be delayed by the use of higher radiation dosages and systemic therapy. In general, we recommend re-excision for patients with positive margins and individualizing treatment recommendations for ­ patients with close margins. We also recommend re-excision for young patients with margins smaller than 2 mm if re­excision can be done with an acceptable aesthetic outcome. For older women and women in whom re-excision would result in an unacceptable cosmetic outcome, increasing the tumor bed dose to 64 to 66 Gy is a reasonable strategy, particularly if the women also receive systemic treatment. Multicentric and Multifocal Disease. A second tumor feature associated with an increased recurrence risk after BCT is the presence of gross multicentric disease, defined as separate foci of disease in different quadrants of the breast, or gross multifocal disease, defined as two or more foci of disease in the same quadrant of the breast. These terms ­refer to clinically evident disease and do not refer to cases in which pathologic examination reveals two areas of disease separated by an area of normal tissue. Data are limited concerning the success of BCT for multicentric or multifocal disease. In one of the largest published series, Kurtz and colleagues130 reported a breast

recurrence rate of 25% in 61 patients treated for gross multicentric or multifocal disease compared with a rate of 11% in patients with unifocal disease. Most of these 61 patients had multifocal disease, with only 10 patients requiring a separate incision or a generous partial mastectomy. No increased risk of recurrence was evident for patients with multifocal disease if the disease was detected only on microscopic examination of the specimen, nor was an increased risk evident among the 22 patients for whom the final resection margin status was negative. A small series from the JCRT also associated multicentric disease with an increased recurrence risk. For 10 patients with multiple synchronous tumors in one breast, the local failure rate ­after BCT was 40%.155 No data concerning the relationship between recurrence and margin status were reported. We recommend mastectomy for women who present with gross multicentric disease that would require separate surgical incisions. For women with multifocal disease within a sphere of 4 cm that is amenable to complete resection with a single segmental mastectomy, the option of BCT is offered, provided that negative surgical margins (<2 mm) can be achieved. The limited data available for multifocal disease that is completely resected with adequate margins suggest that patients with such disease are not at increased risk of recurrence. Disease Stage. Although tumor size and nodal status are powerful prognostic factors for disease-free and overall survival of patients with early-stage breast cancer, these factors have much less impact on local recurrence rates. Among patients in the NSABP B-06 trial who were treated with lumpectomy and radiation therapy, the recurrence rate was 10% for patients with T2 disease and 7% for patients with T1 disease.23 Still more striking was that patients with positive lymph nodes had only a 9% risk of local recurrence at 20 years compared with a 17% risk for patients with lymph node–negative disease.112 These recurrence patterns may have been influenced by the selective use of chemotherapy for patients with lymph node–positive disease. Histologic Subtype. Most of the information concerning the success of BCT has been obtained from women with infiltrating ductal carcinoma of the breast. The randomized prospective trials that directly compared BCT and modified radical mastectomy tend to have insufficient data regarding other histologic subtypes of breast cancer. However, several institutions have reported that BCT is an appropriate option for women with invasive lobular carcinoma or the favorable histologic subtypes of medullary, colloid, and ­tubular carcinoma. Veronesi and colleagues156 reported that the breast tumor recurrence rate after BCT for 431 cases of lobular cancer was 6% and was equivalent to the rate for infiltrating ductal carcinoma. These findings were confirmed in a smaller series from Poen and colleagues,157 who reported an ipsilateral recurrence rate of only 2% for 60 women treated with BCT for lobular carcinoma, and by investigators at  M. D. Anderson, who found a 10-year recurrence rate of 7% for patients with lobular carcinoma, compared with 9% for those with invasive ductal carcinoma.158 Evidence suggests that histologic grade and estrogen and progesterone receptor status do not seem to be independent predictors of local recurrence after BCT. In the NSABP  B-06 trial, grade and estrogen and progesterone receptor status affected local recurrence rates for patients treated with lumpectomy alone but had no influence on recurrence

17  The Breast

patterns for patients treated with lumpectomy and radiation therapy.97 HER2/NEU status does not seem to correlate with an increased risk of breast cancer recurrence.159

Treatment-Related Factors Systemic Therapy. In addition to reducing the probability of development of metastatic disease, systemic ­therapy has been shown to reduce the risk of recurrence in the ipsilateral breast. Randomized trials have clearly shown that chemotherapy and tamoxifen are not appropriate substitutes for radiation therapy.23,120,160 However, for patients who are treated with radiation therapy, the use of systemic therapy improves local control. In the NSABP B-06 ­trial, patients with lymph node–positive disease who were ­treated with radiation therapy and chemotherapy had an 20-year local recurrence rate of 9%, compared with a local recurrence rate of 17% for lymph node–negative patients treated with surgery and radiation therapy alone.112 In the NSABP B-21 trial, tamoxifen similarly improved ­local ­control rates among patients with lymph node–­negative breast tumors smaller than 1 cm.120 In that trial, the crude rate of breast tumor recurrence was only 3% for women randomly assigned to undergo lumpectomy, radiation therapy, and tamoxifen, compared with 7% for women treated with lumpectomy and radiation therapy alone.120 A retrospective analysis from M. D. Anderson investigating the impact of systemic therapy on local control after BCT in patients with lymph node–negative breast cancer confirmed the findings of the NSABP trial.132 In that study, 277 patients treated with systemic therapy had improved 5-year local control rates (97.5% versus 89.8%) and 10-year local control rates (95.6% versus 85.2%) compared with 207 patients who received no systemic treatment. No ­statistically significant difference in local control was seen between patients treated with chemotherapy and those treated with tamoxifen alone (P = .219). In a Cox regression analysis, the use of systemic therapy was the most ­powerful clinical, pathologic, or treatment predictor of ­local control, reducing the risk of local recurrence by a ­factor of 3.3.132 Tumor Bed Boost. The use of a tumor bed boost after irradiation of the intact breast may influence the risk of ­local recurrence. The first randomized trial investigating the effect of a 10-Gy boost after 50 Gy of breast irradiation was performed in Lyon, France. The use of a boost led to a small but statistically significant reduction in the rate of local recurrence at 5 years (3.6% versus 4.5%, P = .04).13 The EORTC subsequently published the results of a much larger randomized trial133 in which patients were ­randomly assigned to receive or not receive a 16 Gy boost after  50 Gy of whole breast radiation treatment. The use of a boost reduced the risk of a breast recurrence at 5 years by a factor of two (P < .001), and the absolute reduction in risk at 5 years was approximately 4%.133 This benefit was observed in all patient age subgroups. An update of this trial found that use of a tumor bed boost reduced the risk of breast recurrence at 10 years from 10.2% to 6.2%.161

Partial Breast Irradiation Accelerated partial breast irradiation (APBI) is a relatively new treatment strategy that is gaining interest for the treatment early-stage breast cancer. The theoretical justification for treating less than the whole breast is that less than 5%

373

of early tumor recurrences occur in quadrants other than that containing the index tumor,162-164 suggesting that treating the distant normal breast adds relatively little benefit for small unicentric tumors. Advocates of APBI argue that this approach increases access to and compliance with care (because of the shortened overall treatment time) and theorize that the volume reductions afforded by APBI may sufficiently reduce normal tissue complications from  hypofractionated radiation to allow therapy to be completed  within 1 week. APBI can be delivered by numerous techniques, but three methods are primarily in current use in North America: multiplane and multicatheter brachytherapy, point source brachytherapy delivered by balloon applicator (MammoSite), and conformal external beam radiation therapy (EBRT). Strictly speaking, these three techniques are not dosimetrically interchangeable,165 but the clinical effects, if any, of these dosimetric differences have not been studied. Alternative techniques that use intraoperative electrons,166,167 intraoperative orthovoltage applicators,168 and postoperative proton therapy169 have also been advocated at a few centers. Evidence from phase I/II trials of APBI for early-stage breast cancer conducted by the Radiation Therapy Oncology Group (RTOG) is promising,170-172 but only a single long-term direct comparison of “modern” APBI with conventional whole-breast radiation therapy has been published.173 That experience has produced single-arm outcome data at 5 years or longer primarily for multicatheter brachytherapy, and results have been mixed with regard to local control and toxicity.174-178 A thorough review of APBI has been provided by Arthur and Vicini.179 Because of the limitations of the available data and ­because the standard treatment (i.e., whole-breast radiation therapy) has had excellent efficacy and tolerability at very long follow-up times, APBI is offered at M. D. ­Anderson primarily in the context of an ongoing phase III trial (RTOG 04-13, cosponsored by the NSABP as trial B-39) or on a highly selected basis for patients who are not eligible to participate in the clinical trial. The primary target population includes patients with DCIS or stage I or II breast cancer who are otherwise appropriate candidates for standard wide local excision (i.e., those with unicentric tumors no larger than 3 cm with histologically negative margins and radiographic evidence of complete removal). The patient’s anatomy must be favorable for APBI (i.e., postoperative CT scan demonstrating a lumpectomy cavity–to–whole-breast ratio for which APBI is technically feasible within the protocol dose constraints). Additional requirements for each of the APBI techniques are outlined subsequently. Patients participating in the clinical trial are randomly assigned to  receive whole-breast irradiation (45 to 50 Gy in 1.8- to 2.0-Gy fractions, followed by an optional boost to at least 60 Gy) or to APBI delivered by a method of the physician’s choice (conformal EBRT, with 38.5 Gy in 10 twice-daily fractions, or multicatheter or MammoSite brachytherapy, with 3.4 Gy in 10 twice-daily fractions). The doses are calculated to be biologically equivalent; the apparent ­difference between techniques results from different dose specification conventions for brachytherapy versus EBRT. The primary aim of the study is to determine whether APBI provides local control equivalent to that seen after conventional whole-breast irradiation; secondary aims 

374

PART 4  Breast

i­nclude comparisons of cosmesis, quality of life, and acute and late toxicity. The trial opened in 2005; long-term ­results are unlikely to be available until 2016.

Techniques Standard Brachytherapy. Interstitial brachytherapy has been delivered with low-dose-rate (LDR) or highdose-rate (HDR) radiation sources. When LDR implants are used, a dose of 45 to 50 Gy is usually delivered to a target volume at a rate of 30 to 70 cGy/hr. Because LDR approaches require that the radioactive implant remain in place during the course of therapy, the patients must be admitted to the hospital for 3 to 5 days and stay in specially shielded rooms designed specifically for this purpose. HDR implants have become increasingly popular because they can be used on an outpatient basis, and this convenience and safety has made HDR implants the de facto standard in most settings. For HDR therapy, a dose of  34 Gy is delivered as twice-daily fractions of 3.4 Gy over a total of 5 days. Multiplane, Multineedle Implant. Definitive PBI using a needle- or catheter-based approach requires a largervolume implant than that usually used for boost purposes. Definitive catheter-based brachytherapy, although the best validated of the PBI techniques, is also technically quite challenging. Typically, a minimum of two planes is required, and more are needed if the surgical volume is extensive. The deep plane is placed at the level of and parallel to the bottom of the excision cavity. The superficial plane is at the level of the superficial border of the cavity. Within a given plane, the catheters are ideally spaced 1.0 to 1.5 cm apart and should extend at least 2 cm beyond the edge of the clips or seroma. The number of catheters is chosen so that a least one catheter is placed 1 cm or more beyond the edge of the target. As is true for all planar implants, the catheters are parallel and as straight as possible. Achieving adequate target volume coverage typically requires at least two LDR seeds or 1.5 to 2.0 cm of HDR dwell positions beyond the edge of the target volume. This implies that three to four LDR seeds or 3.5 to 4.0 cm of HDR dwell positions are required beyond the surgical clips. Sources should not protrude or be located near the skin entry points to prevent permanent marking of the skin. An acceptable implant results in a dose homogeneity index (the ratio of peripheral dose to mean central dose) of at least 0.85. For LDR implants, 45 Gy is specified at  1 to 1.5 cm outside the implant volume by using an isodose line that delivers a dose rate of 30 to 60 cGy/hr. For HDR implants, 34 Gy is prescribed to the target volume and ­delivered in 2 fractions per day, separated by at least 6 hours, for a total of 10 fractions of 3.4 Gy each. MammoSite. In 2002, the U.S. Food and Drug Administration’s approval of the MammoSite applicator device began an era of greatly increased interest in breast brachytherapy. The MammoSite applicator is a double-lumen balloon that can be placed directly into the lumpectomy cavity during the local excision procedure or a few weeks later. The type of applicator should be selected to conform to the seroma cavity without large air pockets or residual seroma fluid. Ideally, the applicator is 1 cm or more from the skin surface, because focal skin necrosis has occurred when the separation between the surface of the skin and the surface of the implant is less than 7 mm. After

FIGURE 17-8.  The isodose distribution is shown for MammoSite brachytherapy. The dose is specified at 1 cm from the surface of the balloon.

v­ erification that no significant distortion of the applicator has occurred and that the source travels to the geometric center of the applicator, 34 Gy is prescribed at 1 cm from the balloon surface (Fig. 17-8) and delivered in 2 fractions per day, separated by at least 6 hours, for a total of 10 fractions of 3.4 Gy each. Conformal External Beam Therapy. PBI has been ­delivered by using three to five non-coplanar photon beams. For this technique, the clinical target volume is ­defined by uniformly expanding the excision cavity volume by 10 to 15 mm and then limiting it to at least 5 mm from the skin surface and lung-chest wall interface. The clinical target volume is further expanded an additional 10 mm (or less if breathing motion studies have been performed) to generate a planning target volume. A total of 38.5 Gy   is prescribed to the isocenter and given in 10 fractions of 3.85 Gy delivered twice each day with at least 6 hours between fractions. The planning goal is to treat 90% of the planning target to no less than 90% of the prescribed dose while simultaneously treating no more than one fourth of the whole breast to the reference dose and no more than one half of the breast to more than one half of the reference dose. This approach is dosimetrically similar to that of brachytherapy—in which the dose is specified to a volume—because 90% of 38.5 Gy is approximately  35 Gy. CT scans illustrating a dosimetry plan for a multibeam conformal PBI are shown in Figure 17-9.

Phase III Trials Christie Hospital Trial. The only prospective randomized trial of PBI that has been completed and has long ­ follow-up is one conducted at the Christie Hospital

17  The Breast

375

recurrence rate seen may reflect the difficulty in identifying the target rather than a failure of the concept. Hungarian National Institute of Oncology Trial. Beginning in 1998, patients at the Hungarian National Institute of Oncology were offered the opportunity to participate in a prospective phase III trial of PBI delivered by multicatheter, multiplane interstitial implant or electron beam versus standard whole-breast irradiation. The latest report of this trial173 indicated that 258 patients had been randomized and an interim analysis had been completed. At a median follow-up time of 66 months, no statistically significant differences in local recurrence rates have been found.

Phase I/II Trials A

B FIGURE 17-9.  CT scans for planning multibeam conformal partial breast irradiation are shown. A, Axial slice (actual size). B, Coronal slice (×2) shows the isodose distribution through the center of the tumor bed. Pink indicates the tumor bed contour. The planning target volume (green) is expanded to include a 2.5-cm radial margin around the tumor bed but does not extend into the chest wall, lung, or skin. The dose prescribed is 38.5 Gy to the isocenter (red line) with the goal of covering 90% or more of the volume planning target volume, indicated by the 90% isodose curve (yellow line), to 34.65 Gy.

(­ Manchester, U.K.) between 1982 and 1987.180 In that study, 708 patients were treated with standard wholebreast tangential fields to a dose of 40 Gy in 15 fractions over 21 days or with local electrons to a dose of 40 to 42.5 Gy in 8 fractions over 10 days. The last update, published at a median follow-up time of 8 years, indicated no difference in overall survival between the two groups, but locoregional recurrence rates were 25% for the patients treated with limited-field (electron beam) irradiation and 13% for those treated with whole-breast irradiation. Several features of this trial limit its applicability to current clinical settings. The most important is that the limitedfield radiation was planned without CT-based information about the tumor bed location. Instead, the electron beam fields were designed by centering the field in the surgical entry site—a technique that we now know can miss substantial portions of the lumpectomy cavity. The high local

Interstitial Brachytherapy. Interstitial brachytherapy for breast cancer has a history going back to the dawn of radiation therapy; one of the very first applications of ­radiation to treat cancer was the use of radium needles for inoperable breast cancer. Like now, the technique required very specialized expertise to be performed safely. After several well-established proponents of interstitial brachytherapy showed favorable results from its use in BCT,176,178 the RTOG opened a multicenter phase I/II trial to determine whether these technically challenging procedures could be performed in a standardized and consistent manner. The 11 participating institutions had to undergo a credentialing process before patient enrollment could  begin; in addition, implant dosimetry had to be evaluated by the trial’s principal investigators in a 24-hour rapid review process before patient treatment was begun. The target ­patient population had unicentric breast cancer less than 3 cm in diameter that had been excised with pathologically clear margins. Surgical clips were used to define the cavity and treatment target, because CT-based brachytherapy dosimetry was in its infancy at the time. For the 100 patients treated in this trial, all but one treatment conformed to the treatment guidelines, and the estimated local failure rate for the 99 evaluable patients was 4% at 5 years.170 Acute toxic effects attributed to the implants included erythema, edema, tenderness, pain, and infection. Of the 66 patients treated with HDR implants, 2 (3%) had grade 3 or 4 toxicity. Of the 33 patients treated with LDR implants, 3 (9%) had grade 3 or 4 toxicity during the brachytherapy. Late toxic effects included skin thickening, fibrosis, breast tenderness, and telangiectasia. No patient experienced late grade 4 toxicity; late grade 3 toxicity was experienced by 18% of the LDR group and 4% of the HDR group.170 Conformal Three-Dimensional Therapy. After the initial exploratory single-institution experience with using conformal EBRT for PBI,181 the RTOG opened a multiinstitutional phase I/II trial to assess this technique.172 Like the RTOG brachytherapy trial, reproducibility, as measured by technical feasibility, was the primary end point. Patients received 38.5 Gy in 3.85 Gy/fraction delivered twice daily to a clinical target volume that included the lumpectomy cavity plus a 10- to 15-mm margin bounded by 5 mm within the skin surface and the lung-chest wall interface. CT planning was required for this technique. Fifty-eight patients were enrolled in this study from August 2003 through April 2004; eligibility requirements were similar to those for the brachytherapy trial. The trial achieved its reproducibility goal, and analysis of the 51 evaluable

376

PART 4  Breast

patients demonstrated 90% compliance with tumor bed and normal tissue dosimetry. Tumor control and treatment complication rates were not intended to be evaluated in the initial report. MammoSite. The MammoSite system was designed to ease some of the technical difficulties associated with multicatheter, multiplane implants. Clearance as a medical ­device by the U.S. Food and Drug Administration required an initial assessment of the physical performance of the ­applicator during radiation therapy and documentation of any acute effects experienced during and immediately ­after the treatment.182 The MammoSite rapidly came into popular use, in large part because of the manufacturer’s effective marketing plan, and several thousand patients have been treated with this modality. The system involves specifying a radiation dose 1 cm away from the surgical margin, and given the large dose gradient inherent in use of a point source implant, it was important to assess the applicator’s safety and efficacy. The 5-year results of this initial assessment174 indicate that none of the 36 evaluable patients have had a local recurrence, toxic effects have been minimal, and 83% of patients have had good to excellent cosmetic results. A voluntary treatment registry, initiated by the manufacturer and subsequently overseen by the American Society of Breast Surgeons, was designed to provide more extensive information on the use of this ­device and on the complications and outcomes experienced by patients treated with it; it likely will be the largest study of its kind. Enrollment closed in 2004, and the study will continue until patients have completed 7 years of followup.183 The full follow-up data must be critically reviewed, because the voluntary nature of participation in the registry may lead to underestimating unfavorable outcomes. ­Several small short-term studies have been reported, ­typically with ­follow-up ­intervals of 2 to 3 years.184-187

Axillary Treatment Traditional Axillary Dissection Until recently, a level I and II axillary lymph node dissection was part of the standard treatment for all women with invasive breast cancer. This procedure effectively treated involved lymph nodes, provided important prognostic information, and provided information that often influenced decisions about the use of systemic treatment. However, the positive aspects of axillary dissection are often offset by its associated morbidity. Although serious long-term complications after axillary dissection are rare, the procedure universally has a negative effect on patients’ quality of life. All patients undergoing this procedure experience axillary pain and numbness and decreased range of motion in the arm, and they are at risk for chronic lymphedema. These concerns have led to investigation of alternatives to axillary dissection. Another impetus for seeking alternatives to axillary dissection is the increased use of systemic treatment for patients with lymph node–negative breast cancer, a trend that makes the prognostic information gained from axillary dissection less important. The two alternative strategies in use are nodal radiation and sentinel lymph node surgery.

Axillary Nodal Irradiation Whether radiation therapy and surgery have equal efficacy in the therapeutic management of the clinically negative axilla was first studied in the 1970s in the NSABP B-04

t­ rial.188 This trial was a three-arm prospective trial in which patients with clinically lymph node–negative disease were randomly assigned to treatment with radical mastectomy, simple mastectomy, or simple mastectomy plus irradiation of the chest wall and regional lymph nodes, including the supraclavicular and infraclavicular areas.188 This trial directly compared surgical treatment of the axilla (i.e., radical mastectomy) with no treatment of the axilla (i.e., simple mastectomy) and primary irradiation of the axilla. Thirty percent of the patients with clinically negative lymph nodes who were treated with radical mastectomy had pathologically involved axillary lymph nodes. The lymph node failure rates were 2% for the radical mastectomy group, 18% for the simple mastectomy group, and 3% for the irradiated group.188 These results indicated that the rate of pathologic involvement of lymph nodes may be almost double the rate of axillary recurrence without therapy. The results also indicated that radiation therapy and surgery were equivalent in preventing axillary recurrences. The investigators of that study did not report a comparison of axillary treatment–related morbidity with the three approaches. Findings from this B-04 study also suggested that patients with clinically positive lymph nodes at presentation had lower axillary recurrence rates after surgery than after definitive irradiation.188 However, at 25 years of ­follow-up, no differences were evident in the rates of distant recurrence or overall survival between the three treatment groups in this trial.189 A second randomized trial that more directly compared axillary lymph node dissection with radiation treatment of the axilla for patients treated with BCT was conducted in France. In that study, after 15 years of follow-up, the axillary recurrence rates were 3% in the radiation group and 1% in the surgical group (P = .04).190 On the basis of these data, several institutions began to omit axillary dissection in patients for whom knowledge of lymph node status would not affect the recommendation for systemic treatment. The patients most often were ­elderly women who would be given tamoxifen regardless of their lymph nodes status. In a prospective, single-arm study at New England Medical Center, 73 women 64 years old or older with clinically negative lymph nodes were treated with axillary irradiation instead of axillary dissection.191 Radiation therapy was delivered with a four-field technique that involved opposed breast tangents, a matched anterior oblique field to include the remaining axilla and supraclavicular lymph nodes, and a posterior field to supplement the dose to the midaxilla. Only one patient experienced a regional recurrence, which occurred 103 months after treatment.191 The treatment-related morbidity from the axillary irradiation was minimal. Similar results were found in a study done at Yale University, where the regional ­failure rate among 327 women treated with primary irradiation of the breast and regional lymphatics was only 3% at 5 years.192 The disadvantages of using these comprehensive fields instead of standard breast tangent fields are the increased amount of lung that is irradiated and the increased risks of arm edema and of fibrosis at the match line of the fields. In an attempt to avoid these disadvantages, the JCRT developed a prospective policy of primary irradiation of the axilla using high tangent fields. This technique used only two radiation fields (i.e., breast tangents) and raised the cranial edge of the treatment field to include the majority of level I

17  The Breast

and a portion of the level II lymph nodes. In a highly selected group of elderly patients with low-risk disease, only 1 of 92 patients developed a regional nodal failure, which was also associated with a primary recurrence in the breast.193

Sentinel Lymph Node Surgery A second strategy for minimizing the morbidity of axillary treatment is sentinel lymph node surgery, which attempts to remove only the first-echelon lymph nodes to which the primary tumor drains—the sentinel nodes. The strategy is based on the assumption that the lymphatic drainage of a primary breast tumor progresses first through one or a small number of lymph nodes. Correspondingly, if these lymph nodes do not contain disease, the probability of disease in any other axillary lymph node should be very small. Two strategies are used to identify sentinel lymph nodes: injection of blue dye and injection of a radioactive colloid. The blue-dye technique involves injection of isosulfan blue dye into the tumor or the dermal lymphatics overlying the tumor bed approximately 3 to 10 minutes before surgery. After massaging the breast, the surgeon makes a small incision in the axilla so that the lymph nodes and lymphatic channels that take up the dye can be visualized. These bluestained lymph nodes are then excised and examined histologically to determine whether a formal axillary dissection is warranted. This technique is relatively simple and is the least expensive sentinel lymph node technique in that it involves only the breast surgeon and can be done entirely within the operating room. The radioactive colloid technique uses technetium 99m to visualize drainage patterns by lymphoscintigraphy before surgery. During the surgery, the sentinel lymph nodes are identified with a handheld gamma probe. The radiocolloid technique offers the advantage of preoperative assessment of lymphatic drainage and allows identification of secondary or primary drainage to the internal mammary lymph node chain. The primary disadvantage of the radiocolloid localization technique is the added schedule coordination and cost. The radioactivity from the primary tumor can mask that of a sentinel lymph node. Advantages of Sentinel Lymph Node Surgery over Axillary Dissection. Sentinel lymph node surgery is associated with significantly less morbidity than complete axillary dissection. Axillary drains are rarely required after surgery and recovery of the range of motion in the arm is more rapid. A prospective trial of 5327 patients reported the following types and rates of morbidity after sentinel lymph node surgery without completion dissection: anaphylaxis to blue dye, 0.1%; axillary wound infection, 1.0%; axillary seroma, 7.1%; axillary hematoma, 1.4%; axillary paresthesias, 8.6%; decrease in upper extremity range of motion, 3.8%; and proximal upper extremity lymphedema (defined as an increase in arm circumference of more than 2 cm from baseline), 6.9%.194 In addition to minimizing the morbidity of axillary treatment, sentinel lymph node surgery may allow more thorough pathologic investigation of the nodes. Because the sentinel lymph nodes are thought to represent the primary echelon of nodes at risk, it is more feasible for pathologists to serially section these lymph nodes and more carefully examine them for micrometastases. The sentinel nodes can be stained with cytokeratin, a nonspecific epithelial marker that is more sensitive for detecting metastases than routine histopathologic examination. The prognostic significance

377

of lymph node metastases detected with such staining, however, has yet to be determined. Initial Experiences with Sentinel Lymph Node Surgery. When sentinel lymph node surgery was first developed, several centers reported the results of sentinel lymph node surgery followed by formal axillary dissection so that false-negative rates (the probability of having a positive nonsentinel lymph node when the sentinel lymph node is negative) could be calculated. In these reports, the success rate in identifying a sentinel lymph node ranged from 91% to 99% and seemed to be independent of the localization technique used.195-197 The false-negative rates (determined by using all patients with identified sentinel lymph nodes who had histologically confirmed lymph node metastases as the denominator) ranged from 0% to 11.4%.195-197 In an important ­contribution, Krag and colleagues196 evaluated the results obtained by a series of community surgeons and found that the false-negative rate depended in part on the skill and experience of the surgeon and could be as high as 29%. The current recommendation is that all surgeons, ­before beginning to use sentinel lymph node surgery as a sole modality of axillary treatment, first establish their personal sentinel lymph node identification rate and false-negative rate by performing sentinel lymph node surgery followed by axillary dissection. The safety and efficacy of sentinel lymph node dissection compared with level I and II lymph node dissection for patients with clinically and pathologically negative nodes have been evaluated in two prospective randomized trials. In one study conducted in Italy, 516 patients with clinically negative lymph nodes were randomly assigned to undergo sentinel lymph node surgery with axillary dissection if the sentinel lymph node was positive or to undergo sentinel lymph node surgery followed by axillary dissection for all patients.198 No difference were found in the rates of breast cancer events between the two groups, and no axillary ­ recurrences occurred in the sentinel lymph node surgery–only group.198 A similar but larger study, NSABP B-32, has been completed, and the results are pending.

Surgical Treatment for Positive Sentinel Lymph Nodes An area of controversy is whether all patients with positive sentinel lymph nodes require a completion axillary dissection. This approach remains the standard of care, because Veronesi and colleagues197 reported that 60% of patients with positive sentinel lymph nodes also had ­additional disease that was found at the time of axillary dissection. However, Chu and colleagues199 demonstrated that the risk of additional positive lymph nodes depends on the extent of sentinel lymph node involvement. In their series, none of the 33 patients with metastases detected solely with cytokeratin staining had additional nodal disease, whereas 14% of the 36 patients with sentinel lymph node disease measuring less than 2 mm and 55% of the 88 ­ patients with disease measuring 2 mm or more had ­ additional nodal disease.199 A randomized trial of com­pleted dissection versus no additional axillary treatment was attempted by the American College of Surgeons, but that trial was closed after insufficient accrual. Several nomograms200-202 have been developed to help surgeons and patients predict the probability of additional axillary disease when the sentinel lymph node is positive. Important factors considered in these nomograms are the

378

PART 4  Breast

number of positive sentinel lymph node, the size of the ­metastasis, the presence of lymphovascular space invasion, the primary tumor size, and the presence of extracapsular disease extension.

Radiation Therapy for Positive Sentinel Lymph Nodes Radiation therapy complements sentinel lymph node surgery. With careful design of the tangential breast fields, most level I and II axillary nodes are included in the radiation field. In theory, axillary recurrence rates in cases with false-negative results on sentinel lymph node surgery should be very low when proper axillary radiation therapy is delivered. However, tangential radiation alone may be inadequate for patients with extensive sentinel lymph node disease. For patients who undergo axillary dissection, the current standard is to add a third radiation field to cover the supraclavicular fossa and axillary apex for all patients who have four or more positive lymph nodes. Three studies have evaluated factors that predict having four or more positive lymph nodes when a sentinel lymph node is positive.9,203,204 Important factors include the number of positive sentinel lymph nodes, evidence of drainage (or lack of drainage) according to lymphoscintigraphy, the presence or absence of lymphovascular invasion, the size of the metastasis, and the size of the tumor bed.9,203,204

Systemic Therapy Systemic therapy has an important role in the multidisciplinary management of early-stage and locally advanced breast cancer, and advances in systemic treatments have contributed to the decrease in breast cancer death rates. The EBCTCG has performed several meta-analyses of all breast cancer trials investigating the use of chemotherapy and hormonal therapy for breast cancer. Their 2005 publication205 reported data from almost 150,000 patients treated on 194 randomized trials. Because only trials that had begun by 1995 were included in the meta-analysis, no information is yet available on the efficacy of taxanes, ­aromatase inhibitors, or trastuzumab.

Chemotherapy Chemotherapy was found to produce similar proportional reductions in the probability of disease recurrence and the risk of death for patients with lymph node–positive disease and those with lymph node–negative disease.205 Because patients with positive lymph nodes are at greater risk for recurrence, the absolute percentage of patients who benefit from chemotherapy is greatest in this group. Other findings from the meta-analysis were that use of multiagent chemotherapy resulted in better outcomes than single-agent ­chemotherapy; that inclusion of an anthracycline in the multiagent chemotherapy regimen (e.g., fluorouracil, doxorubicin, and cyclophosphamide [FAC regimen] or fluorouracil, epirubicin, and cyclophosphamide [FEC regimen]) led to better outcomes than the previous standard chemotherapy regimen (i.e., cyclophosphamide, methotrexate, and fluorouracil [CMF regimen]); and that younger patients derive a larger benefit from chemotherapy than do older patients. Specifically, the proportional reduction in the risk of recurrence with chemotherapy was 36% for patients younger than 50 and 29% for those 50 to 69 years old. The introduction of taxanes has further improved outcomes beyond those that could be achieved with

a­ nthracycline-based chemotherapy. Three key trials that have shown a benefit from the use of adjuvant taxanes after anthracycline chemotherapy have been the Cancer and Leukemia Group B (CALGB) 9344/Intergroup 0148 trial,206 the NSABP B-28 trial,207 and the Breast Cancer International Research Group (BCIRG) 001 trial.208 Collectively, these trials have indicated that the addition of ­taxanes to anthracycline-based chemotherapy further reduces the risk of recurrence in the adjuvant treatment of breast cancer. This benefit was confirmed in a 2006 metaanalysis of trials testing the use of adjuvant taxanes for breast cancer. That analysis revealed improvements in disease-free survival rates (an estimated absolute benefit of 3% to 5% for patients with lymph node–positive disease at 5 years) and overall survival rates (an estimated absolute benefit of 2% to 3% for patients with lymph node–positive disease at 5 years) from the inclusion of taxanes relative to outcomes after anthracycline chemotherapy without taxanes.209 After taxanes became established as an important component of adjuvant therapy for breast cancer, new trials began investigating different administration schedules. The CALGB 9741/Intergroup trial randomly assigned ­patients who were receiving four cycles of doxorubicin and ­cyclophosphamide followed by four cycles of paclitaxel or doxorubicin, four cycles of paclitaxel, and four cycles of cyclophosphamide to one of two schedules in which drugs were delivered every 3 weeks or in a dose-dense schedule in which the chemotherapy was given (with growth factor support) every 2 weeks.210 Findings from that trial indicated that dose-dense therapy led to better disease-free survival rates at 4 years (82% for the every-2-weeks schedule versus 75% for the every-3-weeks schedule).

Hormonal Therapy The EBCTCG meta-analysis demonstrated that adjuvant tamoxifen therapy led to a 31% proportional reduction in the annual risk of breast cancer death for patients with ­estrogen receptor–positive disease.205 This benefit was seen for premenopausal and postmenopausal women. The metaanalysis further indicated that 5 years of tamoxifen therapy provided better outcomes than did 1 or 2 years of treatment. The meta-analysis found that the benefits of chemotherapy and tamoxifen therapy were additive for ­patients with estrogen receptor–positive disease; the estimated ­cumulative reduction in the risk of mortality from combined anthracycline chemotherapy followed by tamoxifen for young women with estrogen receptor–positive disease was estimated to be as high as 57%.205 Aromatase inhibitors represent another significant advance in hormonal therapy for breast cancer. Unlike tamoxifen, which directly blocks the estrogen receptor on tumor cells, aromatase inhibitors act by decreasing circulating estrogen levels by inhibiting the conversion of adrenal testosterone-like hormones into estrogen. Accordingly, aromatase inhibitors do not have some of the pro-­estrogenic effects associated with tamoxifen. Aromatase inhibitors do not have tamoxifen’s beneficial preventive effect against osteoporosis, but they also do not have tamoxifen’s potentially harmful effect of stimulating the endometrial lining. Because aromatase inhibitors do not block the production of estrogen by the ovaries, tamoxifen remains the preferred hormonal therapy for all premenopausal women. Postmenopausal women may be ­appropriately treated with tamoxifen or an aromatase inhibitor.

17  The Breast

Several randomized trials indicate that aromatase inhibitors used instead of or in sequence with tamoxifen can improve the outcome for postmenopausal women with estrogen receptor–positive disease relative to tamoxifen alone. The Arimidex, Tamoxifen, Alone or in Combination (ATAC) trial211 and the Breast International Group (BIG) 1-98 trial212 showed that 5 years’ therapy with anastrozole (ATAC) or letrozole (BIG 1-98) produced better outcomes than did 5 years of tamoxifen. The Intergroup Exemestane Trial found that patients who took tamoxifen for 2 or 3 years had improved outcomes if they switched to exemestane at that time compared with continuing tamoxifen for 5 years.213 The National Cancer Institute of Canada’s MA.17 trial found that adding 5 years of letrozole after 5 years of tamoxifen improved outcome compared with just 5 years of tamoxifen.214 Other ongoing trials are underway to further define the best way of incorporating aromatase inhibitors into the adjuvant treatment of estrogen receptor–positive breast cancer in postmenopausal women.

Biologic Therapy The introduction of trastuzumab, a humanized monoclonal antibody that binds to HER2, into adjuvant therapy for tumors with amplified expression of HER2/NEU represents a new era in breast cancer treatment. Prospective randomized trials showed that the addition of trastuzumab increased the efficacy of adjuvant anthracycline and taxane chemotherapy for patients with tumors that overexpressed HER2/NEU.215-218 In the United States, the combined results of two similarly designed trials from the NSABP and the North Central Cancer Treatment Group indicated that the addition of trastuzumab led to a 52% reduction in the relative risk of recurrence compared with anthracycline and taxane chemotherapy alone.217 In those trials, the trastuzumab was started after four cycles of chemotherapy and was given concurrently with four cycles of paclitaxel and then continued as a single agent for 1 year. Because trastuzumab increases the risk of decreased cardiac ejection fraction, patients who are treated with this agent require careful cardiac supervision.

Sequencing of Treatment Modalities Radiation Therapy and Chemotherapy Most patients with early-stage breast cancer are treated with surgery, systemic therapy, and radiation, and determination of the best way to integrate these modalities remains an important issue. Most patients who are candidates for breast conservation at the time of diagnosis undergo surgery first. For patients who are to receive adjuvant chemotherapy in addition to radiation, most practices administer the chemotherapy before the radiation therapy. The JCRT conducted a randomized trial to investigate the sequencing of radiation therapy and chemotherapy for 244 patients treated with breast-conserving surgery. Patients were randomly assigned to four cycles of doxorubicin-based combination chemotherapy followed by radiation therapy or to radiation therapy followed by four cycles of the same chemotherapy.219 The initial report from this trial suggested that the patients who received chemotherapy first had a significantly lower rate of distant metastasis than patients who received radiation first (25% versus 36% at 5 years, 

379

P = .05), but this difference disappeared with longer  follow-up.220 Moreover, because close surgical margins also confer an increased risk of local recurrence,115 re-excision in such cases may be a better approach than delaying radiation so that chemotherapy can be given first. Although the JCRT study provided important data concerning treatment sequencing, that study predominantly focused on patients with lymph node–positive disease. Investigators at M. D. Anderson retrospectively ­analyzed the sequencing of chemotherapy and radiation for 124 patients with lymph node–negative disease treated with BCT. In that series, 79% of the patients had negative margins. The 5-year actuarial rates of ­local control were 100% for the chemotherapy-first group (most commonly six cycles of doxorubicin-based chemotherapy) and 94% for the radiation-first group (P = .351).  The 5-year ­ recurrence-free survival rates for the ­chemotherapy-first and radiation-first groups were 92% and 77% (P = .083), respectively.221 An important study from the CALGB addressed whether delaying radiation so that taxanes could be given first would increase the risk of local recurrence. That study compared patterns of ­local disease recurrence among patients who were ­treated with radiation therapy after four cycles of ­doxorubicin­cyclophosphamide chemotherapy with those who ­received radiation after four cycles of doxorubicin and cyclophosphamide followed by four cycles of paclitaxel. The findings revealed that patients given the ­ additional paclitaxel treatment had lower risks of isolated locoregional recurrence (3.7% versus 9.7%, P = .04).222

Radiation Therapy and Hormonal Therapy The best way to sequence radiation and adjuvant ­hormonal therapy has been somewhat controversial. Initially, the concern was that concurrent tamoxifen use might reduce the effectiveness of radiation therapy by arresting cells in radioresistant phases of the cell cycle. Tamoxifen can also increase levels of transforming growth factor β (TGF-β), which may enhance the risk of radiation-induced fibrosis of the lung. Bentzen and colleagues223 reported that the risk of lung fibrosis associated with postmastectomy radiation was doubled for patients treated with concurrent tamoxifen compared with those treated with radiation alone.223 A randomized trial from the Southwest Oncology Group that compared concurrent tamoxifen and chemotherapy with chemotherapy followed by tamoxifen showed that the concurrent treatment was associated with a worse ­outcome.224 No randomized trials have been conducted to directly compare concurrent tamoxifen and radiation with radiation followed by tamoxifen. Three retrospective studies reported no difference in outcome according to the sequencing of radiation and hormonal therapy. Pierce and colleagues225 examined this question in 309 patients treated in the Southwest Oncology Group study and found 10-year recurrence rates of 7% with concurrent treatment and 5% with sequential therapies (P = .54).225 Ahn and colleagues226 at Yale University examined this issue in 495 patients treated with breast conservation and found no difference in local control, development of distant metastases, or overall survival at 10 years of follow-up. A similar study reported by Harris and colleagues227 at the University of Pennsylvania showed very similar results.

380

PART 4  Breast

The institutional practice at M. D. Anderson is to give tamoxifen in sequence with radiation for patients with tumor features that place them at relatively low risk of metastatic disease. For other patients, such as those with an estrogen receptor–positive tumor with multiple lymph nodes that remain positive after the completion of neoadjuvant chemotherapy, hormonal therapy can reasonably be started as early as possible. No data are available concerning the safety and efficacy of concurrent administration of radiation and aromatase inhibitors.

Radiation Therapy and Trastuzumab When the phase III trials of adjuvant trastuzumab were ­designed, the decision was made to sequence the trastuzumab concurrently with radiation. Investigators were hesitant to disrupt the course of trastuzumab to deliver the radiation, and because trastuzumab was given for a minimum of a year in these trials, a sequential approach for these treatments was not a potential option. Before these trials, little was understood about the normal tissue ­interactions between trastuzumab and radiation therapy. Fortunately, two studies with 3 to 4 years of follow-up suggest that giving trastuzumab and radiation concurrently is not associated with an increased risk of normal tissue or heart injury. Findings from the NSABP B-31 trial showed a rate of congestive heart failure of 3.2% for patients treated with trastuzumab and left-sided radiation compared with a rate of 4% for those treated with trastuzumab and no left­sided radiation (P = .80).228 These findings were ­supported by those of the North Central Cancer Treatment Group N9831 trial, which indicated that concurrent trastuzumab and radiation did not increase the rate of cardiac events relative to that in patients who were treated with trastuzumab but not radiation.229 Although the information available suggests that concurrent administration of trastuzumab and radiation is safe, the possibility of long-term adverse effects remains. Typically, the cardiac toxicity associated with radiation therapy manifests only after a decade of clinical follow-up, so it is too early to conclude that these two treatments have no adverse interaction with respect to late cardiac effects. One way of ensuring the long-term safety of this approach is to use modern radiation treatment techniques that spare the heart as much as possible. If radiation therapy for breast cancer involves little or no exposure of the heart, adverse cardiac interactions between trastuzumab and radiation are highly unlikely.

Chemotherapy and Surgery Sequencing chemotherapy before surgery has become increasingly common for treating stage II and III breast cancer. One potential advantage of giving chemotherapy first is that breast conservation can be offered to appropriately selected patients with T3 or T4 primary tumors. This sequence also allows assessment of the response of the disease to a particular chemotherapy regimen, which can be important for patients with resistant disease because it can provide an opportunity to change to a different treatment regimen. Being able to assess response also is extremely valuable for clinical research. For example, the ability to assess disease responsiveness directly by comparing rates of complete pathologic response in a  randomized phase III trial comparing two ­chemotherapy

regimens allows relatively small study populations and short follow-up times to be used relative to ­studies that compare chemotherapy regimens in an adjuvant setting. Neoadjuvant chemotherapy can also facilitate translational research to investigate the mechanisms of ­ chemotherapy­induced cell death and the development of ­chemotherapy resistance. Two large randomized trials have compared the efficacy of neoadjuvant chemotherapy with that of adjuvant chemotherapy for operable breast cancer. The NSABP B-18 study randomly assigned 1523 patients to receive four cycles of doxorubicin and cyclophosphamide before or after surgical treatment.230,231 At 9 years of follow-up, the rates of overall survival and disease-free survival were nearly identical for the two groups. A second randomized, prospective trial, which was conducted by the EORTC, confirmed these results and again found rates of survival and distant metastases to be equivalent between the neoadjuvant chemotherapy and adjuvant chemotherapy groups.232 A meta-analysis of data from 3946 patients treated in nine randomized trials that compared neoadjuvant with adjuvant chemotherapy for breast cancer also found no statistical differences in the risk of death, disease progression, or distant disease recurrence between the two treatment strategies.233 In the NSABP and EORTC studies, the proportion of women who underwent BCT was higher in the neoadjuvant chemotherapy groups. Breast conservation after neoadjuvant chemotherapy for patients with advanced primary tumors is discussed under “Breast Conservation for Advanced Disease.”

Radiation Therapy after Mastectomy The history of postmastectomy radiation therapy can be characterized as one of conflicting data and passionate debate. When few or no systemic therapy options were available for the treatment of localized breast cancer,  radiation therapy was used alone or as an adjunct to surgery to prevent locoregional recurrence.234-238 High-dose radiation therapy alone (without surgical intervention) was effective in the short term but was associated with excessive long-term morbidity.239 Mastectomy followed by radiation therapy could result in long-term control of disease, even for some patients with locally advanced breast cancer.240 Although many studies showed that postmastectomy radiation could reduce the risk of locoregional recurrence, none showed clear improvements in survival. After the publication in 1987 of a meta-analysis by Cuzick and colleagues241 that reported increased mortality among women who had received postmastectomy radiation therapy as a component of their treatment, use of postmastectomy radiation was intensely challenged.  Although these and subsequent meta-analyses were strongly influenced by the results of a few large early trials that were hampered by poor radiation techniques, selection of low-risk patients, and the infrequent use of adjuvant systemic therapy, the attitude toward using radiation in all but the most advanced cases of breast cancer ranged from indifference to opposition. When effective ­systemic therapy became available that produced unequivocal improvements in overall survival, many thought that 

17  The Breast

chemotherapy would make the routine use of radiation therapy after mastectomy obsolete. In 1997, two reports from large prospective trials reversed the prejudice against the use of postmastectomy radiation therapy. Those reports showed that in lymph node–positive breast cancer, mastectomy followed by both chemotherapy and comprehensive irradiation resulted in superior locoregional disease control, disease-free survival, and overall survival. The Danish 82b trial evaluated the use of postoperative radiation therapy for 1708 premenopausal women with operable node-positive breast cancer treated with mastectomy and methotrexate-based chemotherapy.242,243 At 10 years of follow-up, the patients who had received radiation therapy to the chest wall and regional lymphatics had a lower rate of locoregional recurrence (9% versus 32%) and improved cancer-specific survival rates (48% versus 34%) and overall survival rates (54% versus 45%) compared with women who had not received radiation therapy. These results confirmed that patients with larger tumors or extensive nodal involvement benefited from postmastectomy irradiation, but surprisingly, it also revealed significant improvements in all other subgroups of patients, including patients with one to three involved nodes. A smaller trial from British Columbia reported that the addition of radiation therapy after mastectomy and chemotherapy resulted in improved locoregional control rates (87% versus 67%) and improved cancer-specific survival rates (50% versus 33%) in premenopausal women with lymph node–positive disease.244 A separate report from the Danish Breast Cancer Cooperative Group demonstrated that postmenopausal women with node-positive breast cancer also benefited from radiation therapy after mastectomy and adjuvant tamoxifen.245 Subsequent updates of these critical trials continue to support their initial conclusions, although no additional trials have yet been reported. At 20 years of follow-up, the British Columbia group continued to see a reduction of two thirds in locoregional recurrence and a 10% absolute survival advantage for women who received radiation after mastectomy and chemotherapy.246 The Danish Group, analyzing their outcomes at 15 years, showed similar findings.247

Patient Selection Together, the trials of postmastectomy irradiation have demonstrated that adjuvant radiation therapy can improve overall survival in circumstances in which locoregional failure rates can be substantially reduced. It is imperative to identify patients at significant risk of recurrence, because they are the most likely to benefit from adjuvant ­irradiation. Postmastectomy irradiation is used to treat subclinical tumor in the remaining tissue of the anterior chest wall and the regional lymphatic basins. Because this subclinical disease is by definition undetectable, the probability of its presence and its likely distribution must be inferred from a variety of sources. These sources include histologic data from surgical specimens, pattern-of-failure analyses, and prospective trials in which defined groups of patients (e.g., those with lymph node–positive disease) or specific target volumes are allotted to treatment or to observation. In the United States, the standard procedure for modified radical mastectomy is the procedure described by

381

Madden and colleagues.248,249 This procedure involves removing all visible breast tissue along with a complete level I and II axillary dissection. All of the axillary tissue is removed in a single piece to the level of the pectoralis minor muscle. In contrast to the classic Halsted mastectomy, the skin flaps are not completely undermined, and both pectoral muscles are spared. Given the presumption that only patients with residual subclinical disease in the remaining tissue will benefit from radiation therapy, only patients thought to have an intermediate to high risk of persistent disease after mastectomy should be considered for radiation. At M. D. Anderson, postmastectomy radiation therapy is routinely offered to all patients when their cumulative (lifetime) risk of locoregional recurrence exceeds 20% and to selected patients who are at lower but still significant risk of locoregional recurrence. The challenge is to accurately interpret the clinical factors that predict these levels of risk. Most clinicians agree that patients with locally advanced primary breast cancer and extensive lymph node involvement (i.e., large nodal metastases with gross extranodal extension or four or more lymph nodes containing tumor)250 or those for whom surgery was incomplete251 fit this description. The lower end of this risk range is less well defined, as subsequently described.

Evidence from Clinical Trials Few studies have been published concerning patterns of failure in patients treated with modified radical mastectomy and modern chemotherapy regimens. Until recently, most of the large series reported included patients who were treated with CMF regimens.252,253 Given the mounting evidence that anthracycline-based chemotherapy results in superior disease-free survival for women with lymph node–positive breast cancer,254 data about outcomes after anthracycline-containing regimens are probably more relevant to current practice. Several retrospective series have analyzed locoregional recurrence rates after mastectomy and anthracycline-containing regimens.255-259 The results of those series seem quantitatively different from those of the CMF-based studies, even the prospective randomized trials assessing the role of radiation after mastectomy and CMF, as subsequently discussed. British Columbia Trial. The British Columbia trial244,246 included 318 patients with lymph node–positive disease who underwent mastectomy by their local providers and then were referred to the provincial cancer center for adjuvant therapy. There, they received CMF chemotherapy and were randomly assigned to radiation or to observation. The median follow-up time at the final analysis was 249 months. At that time, radiation was found to have significantly decreased the locoregional recurrence rates (10% versus 25%) and to have increased breast cancer–specific survival rates (48% versus 30%). Locoregional recurrence was the first site of failure for all patients. These findings may not be applicable to other clinical circumstances for several reasons. First, the patients were given CMF chemotherapy, which is less effective than current doxorubicin- or taxane-containing regimens (especially for HER2-amplified tumors). Second, 44% of patients who had positive nodes had nodes ranging from 2 to 5 cm in diameter, with an additional 3% of patients having nodes

382

PART 4  Breast

larger than 5 cm, leading many to speculate that a large proportion of patients might have had clinical N2 (stage III) disease. Third, the radiation technique used resulted in the internal mammary chain being treated with cobalt 60 photons (including irradiation of some of the ­mediastinum) at a fairly high dose per fraction (2.34 Gy), even though the total dose was 37.5 Gy. Late cardiac toxicity in this group might have compromised the overall survival benefit from radiation therapy. Fourth, the local recurrence rates for patients with one to three positive nodes and no ­radiation therapy was substantially higher than the rates reported in several U.S. studies that used ­anthracycline-based regimens.260,261 Danish Breast Cancer Cooperative Group Trials. The Danish Breast Cancer Cooperative Group designed a set of masterful research trials in which very large numbers of women with breast cancer have been treated in a quite consistent manner.242,243,245,247,262 The randomized postmastectomy trials from Denmark are especially ­important because each report showed statistically significant ­ improvement in survival in the groups given combined-­modality therapy. Those trials were directed at premenopausal women treated with CMF polychemotherapy (trial 82b) and postmenopausal women treated with tamoxifen (trial 82c). The 82b trial included 1708 women who underwent mastectomy and CMF chemotherapy and were randomly assigned to radiation therapy or observation. The first findings from this trial were published when the ­ median follow-up time was 114 months. The surgical intent in that trial was a total glandular mastectomy plus axillary lymph node sampling; the median number of nodes examined was seven, and 76% of patients had fewer than 10 lymph nodes removed. Sixty-two percent of the patients had one to three positive lymph nodes, but the size of the largest node and the presence or absence of extranodal extension were not reported. At 10 years, the ­addition of radiation had reduced the local recurrence rates from 32% to 9%, and the overall survival rates were 54% for the CMF plus radiation therapy group and 45% for the chemotherapy-only group.

Although these trials were well designed and conducted, their findings are not necessarily applicable in all circumstances. First, the chemotherapy regimen was CMF given every 28 days. (The British Columbia trial and the Danish Breast Cancer Cooperative Group 82b trial sandwiched the radiation between courses of CMF.) Second, the limited axillary sampling is a source of significant concern, particularly because the local recurrence rate in the chemotherapy-only group was much higher than the rates reported in most other trials; moreover, almost one half of these recurrences occurred in the axilla, which is quite unusual. In a report given at the 1999 meeting of the Gilbert A. Fletcher Society, patients with T1 or T2 tumors and three positive lymph nodes of 10 nodes sampled had an 80% to 85% chance of having additional positive nodes if the axillary dissection was completed. In other words, many critics of the Danish trial think that the group of patients with one to three positive nodes would have included patients with four or more positive nodes if patients had undergone a more complete axillary dissection. Although Overgaard and others247 subsequently reanalyzed the subset of 1152 patients in whom eight or more nodes were removed and found a similar outcome, the previous argument still holds because patients were not stratified at protocol entry by the number of lymph nodes removed. The M. D. Anderson Studies. At M. D. Anderson, outcomes after mastectomy and chemotherapy have been studied in 1031 patients treated with mastectomy and doxorubicin-based chemotherapy but without radiation in five prospective clinical trials spanning a 20-year period.  The median follow-up time was almost 10 years.260,263 In these M. D. Anderson studies, only patients with four or more involved nodes were consistently found to have locoregional recurrence rates in excess of 20% (Table 17-5).260 Among patients with one to three involved nodes, recurrence rates were higher for patients who also had tumors measuring 4 cm or larger, gross extranodal extension, or inadequate axillary dissection compared with patients without these factors. In our practice, most patients with stage II breast

TABLE 17-5.   Ten-Year Risk of Locoregional Recurrence after Mastectomy and Doxorubicin-Containing Chemotherapy at M. D. Anderson Cancer Center According to Tumor Stage and Number of Involved Lymph Nodes Risk of Locoregional Recurrence (%)* Tumor Stage

0 Nodes Involved

1-3 Nodes Involved

4-9 Nodes Involved

≥10 Nodes Involved

Isolated Locoregional

Recurrence*

T1

6

7

T2

11

12

23

17

T3

29

9

31

29

9

17

Cumulative Locoregional Recurrence* T1

11

9

17

17

T2

14

16

27

34

T3

29

23

40

29

*Isolated

locoregional recurrence indicates locoregional recurrence alone as the first site of failure; cumulative locoregional recurrence i­ndicates locoregional recurrence occurring at any time. Modified from Katz A, Strom EA, Buchholz TA, et al. Locoregional recurrence patterns after mastectomy and doxorubicin-based chemotherapy: implications for postoperative irradiation. J Clin Oncol 2000;18:2817-2827.

17  The Breast

cancer and one to three involved nodes have had a substantially lower risk of recurrence than the risk seen in the Danish and British Columbia trials, so it is unclear whether these patients with one to three positive nodes would have benefited from radiation therapy. An additional study of patients treated with mastectomy and anthracycline-based chemotherapy was reported by Taghian and colleagues.261 In that analysis, the records of 5758 patients enrolled in five NSABP trials were examined retrospectively to assess patterns and risk of locoregional failure among women with lymph node–positive breast cancer, 90% of whom had received doxorubicin-based chemotherapy. The cumulative incidence of locoregional failure as first event, with or without distant failure, for patients with one to three positive nodes was 13.0%; that for patients with four to nine positive nodes was 24.4%, and that for patients with 10 or more positive nodes was 31.9%. These local failure results are consistent with those in the M. D. Anderson experience and represent considerably lower locoregional recurrence risks than those seen in the Danish or British Columbia trials for patients with one to three positive nodes. Presumably, patients with a lower risk of persistent local disease would be less likely to benefit from additional local therapy, and it is uncertain if any survival benefit can be attributed to the addition of radiation therapy. Why are the M. D. Anderson and NSABP results different from those of the British Columbia or Danish trials? The difference may reflect the use of anthracycline-based chemotherapy rather than CMF, but because ­chemotherapy by itself has never been shown to influence the risk of ­local recurrence, that explanation seems unlikely. Another ­explanation proposed by Truong and others264 centers on differences in the numbers of excised nodes (i.e., median of 7 in the Danish studies, 11 in the British Columbia study, and 15 or more in the U.S. series). When locoregional recurrence is considered in terms of the absolute numbers of involved lymph nodes, the risk for patients with one, two, or three involved nodes was 21.5% for the patients  in the British Columbia trial but only 12.6% in the  M. D. Anderson trial (P = .02). However, when nodal ratios (the number of positive nodes divided by the number of ­excised nodes) are used instead of absolute numbers, the

383

rates were more comparable (<20% for patients with one, two, or three positive nodes and nodal ratios between 0.0 and 0.19 in both series). In both series, nodal ratios that exceeded 0.20 were associated with locoregional recurrence rates in excess of 20%, and radiation therapy would be recommended for those patients. The pattern of locoregional recurrence is instructive for the practice of radiation oncology. In all published series, most locoregional recurrences occur in the chest wall skin and subcutaneous tissues, often leading postmastectomy irradiation to be dubbed “chest wall irradiation,” even though other targets usually are included in the radiation treatment plan. The second most common location for recurrence is the anterior structures of the undissected (level III) axilla, infraclavicular fossa, and supraclavicular fossa. In the M. D. Anderson studies, consistent with numerous other reports (Tables 17-6260 and 17-7252,265-271), failure in the dissected portion of the axilla (levels I and II) was rare (3%). Because the risk of failure in the midaxilla at M. D. Anderson was not significantly higher for patients with ­increasing numbers of involved nodes (or increasing percentages of involved nodes),260 specific irradiation of the axilla is rarely indicated if the axilla has undergone full compartment dissection.

Radiation Therapy after Neoadjuvant Chemotherapy and Mastectomy All of the clinical trial findings discussed in the previous section reflect mastectomy being used as the first component of treatment. This strategy has therapeutic and diagnostic benefits. When chemotherapy is used before the surgery, the pathologic features that identify populations at high risk for locoregional recurrence—and serve as indications for postmastectomy radiation therapy—are likely to be irretrievably altered. The initial response to neoadjuvant chemotherapy is an important predictor for enhanced patient survival (see “Chemotherapy and Surgery”); however, this very response to systemic therapy makes locoregional treatment planning more complex. First, patients who are at high risk of locoregional recurrence because of the clinical findings present at diagnosis remain at risk for locoregional recurrence and should receive postmastectomy radiation therapy.272-274 These findings include having T3 and T4 primary tumors, having advanced regional nodal presentations

TABLE 17-6.   Patterns of Locoregional Recurrence after Mastectomy and Doxorubicin-Containing Chemotherapy at M. D. Anderson Cancer Center Isolated Locoregional Recurrence* Percent†

Cumulative Locoregional Recurrence Percent†

Location of Recurrence

Number

Chest wall

122

98

122

68

Supraclavicular nodes

41

33

71

40

Axilla

21

17

25

14

Number

Infraclavicular nodes

10

8

12

7

Internal mammary chain

—

—

15

8

124

100

179

100

Any site *Isolated

locoregional recurrence indicates recurrence alone as the first site of failure; cumulative locoregional recurrence indicates locoregional recurrence at any time. †Some patients experienced more than one site of failure, resulting in cumulative percentages greater than 100%. Modified from Katz A, Strom EA, Buchholz TA, et al. Locoregional recurrence patterns after mastectomy and doxorubicin-based chemotherapy: implications for postoperative irradiation. J Clin Oncol 2000;18:2817-2827.

384

PART 4  Breast

TABLE 17-7.    Locoregional Recurrence Patterns after Mastectomy by Site Site of Recurrence*

Chest Wall

Clavicular Nodes

Internal Mammary Glands

Axilla

209

142 (68)

52+ (25)

NS

15 (7)

M. D. Anderson ­ Cancer ­Center267†

148

89 (60)

19+ (13)

5+ (3)

11+ (7)

Mallinckrodt269

129

77 (60)

42 (33)

14 (11)

23 (18)

University of ­Pennsylvania270

128

106 (83)

30 (24)

4 (3)

14 (11)

Institut Jules Bordet266

128

99 (77)

32 (25)

NS

13 (10)

Sinai-Miami271

26 (21)

Number of ­ Patients

University Hospital of Cleveland268†

Study and ­Reference

124

95 (77)

14 (11)

10 (8)

ECOG252

70

37 (53)

17+ (24)

NS

DBCG265

214

137 (64)

15 (7)

NS

Mt.

8+ (11) 73 (34)

*Values

are numbers (percentages), unless otherwise indicated. Because of inclusion of patients with multiple failure sites, totals for some studies may be greater than 100%. †Details regarding multiple regional sites were not provided. DBCG, Danish Breast Cancer Cooperative Group; ECOG, Eastern Cooperative Oncology Group; NS, not stated.

(N2 or N3), and having four or more positive lymph nodes. Second, a complete pathologic response does not eliminate the need for radiation therapy for patients who present with advanced disease.275 Patients at intermediate risk for locoregional recurrence are those whose initial clinical stage and postchemotherapy pathologic findings indicate T1 or T2 tumors and one to three positive lymph nodes. Other factors, such as the presence of lymphovascular invasion, lack of estrogen receptor expression, the presence of gross extranodal extension, or patient age, may also influence risk of locoregional recurrence,274 but these factors await validation in large groups of patients.

Current Practice at M. D. Anderson Presumably, only patients who have residual tumor in the chest wall and regional nodes can benefit from radiation therapy to these sites. Because radiation therapy itself confers some risk to the patient who undergoes it, clinical decisions ideally should be based on a risk-benefit analysis for each clinical scenario. In many day-to-day practices, any  patient whose risk of local recurrence is considered to be 15% to 20% or more is thought to benefit from comprehensive postmastectomy irradiation. However, this recommendation comes with some caveats. Given the inherent variance in radiation therapy plans that are generated by different people with different experience for individual ­patients, it is reasonable to assume that some treatment plans and treatment delivery will be better than others. Blanket ­recommendations—especially for patients with marginal indications for treatment—should not be universally ­adopted. The recommendations given subsequently arise from a specific multidisciplinary practice in a highly specialized ­tertiary cancer center, and radiologists should assess the relevance of these recommendations to their own situations. At M. D. Anderson, the recommendation is that all patients with locally advanced (stage III) breast cancer receive postmastectomy irradiation regardless of their response to chemotherapy (if chemotherapy is given in the

neoadjuvant setting). This includes patients with tumors larger than 5 cm, skin involvement, or extensive lymph node involvement, as indicated by four or more positive lymph nodes, fixed or matted nodes, or involvement of the internal mammary, infraclavicular, or supraclavicular regions. Because patients with clinical stage II breast cancer (including those with one, two, or three involved nodes) are generally at low risk for local recurrence, only some of the patients should be given postmastectomy irradiation— specifically, those with gross extranodal tumor extension, lymphovascular space invasion in the primary tumor, positive margins of resection, pathologic involvement of the skin, or fewer than 10 lymph nodes removed in the axillary dissection or those who are young. Patients with earlystage, lymph node–negative breast cancer are not typically offered postmastectomy irradiation.

Treatment of the Internal Mammary Chain Whether the internal mammary chain should be treated in the course of postmastectomy radiation therapy continues to be hotly debated. The controversy arises because the available data are confusing and seem to conflict. In early series, histologic sampling of the internal mammary chain in patients with axillary node–positive breast cancer revealed occult disease in 20% to 50% of cases (Table 17-8).6,276-278 This is not just a historic curiosity. A review of 1679 patients who underwent extended radical mastectomy (i.e., mastectomy with internal mammary dissection) between 1956 and 2003 revealed pathologic involvement of the internal mammary chain in up to 40% of some patient groups.279 In that study, patients whose risk of internal mammary node involvement exceeded 20% were those with four or more positive axillary lymph nodes; those with medial tumors and positive axillary lymph nodes; those with T3 tumors who were younger than 35 years; those with T2 tumors and positive axillary lymph nodes; and those with medial T2 tumors.279 However, clinical  recurrence in the internal mammary chain is rare.

17  The Breast

385

TABLE 17-8.   Probability of Pathologically Involved Internal Mammary Nodes in Patients with Positive Axillary Nodes Probability of Involved Internal Mammary Nodes Study and Reference

Number of Patients

Outer Quadrant Tumors (%)

Inner Quadrant Tumors (%)

Urban and Marjani6

341

42

53

Bucalossi et al276

553

Handley277

535

21

48

635

25

35

Ki and

Shen278

Any Quadrant Tumors (%)

29%

Only one prospective trial has assessed the value of various forms of treatment of the internal mammary chain.280,281 In that trial, conducted between 1958 and 1978, patients with histologically positive axillary nodes ­undergoing mastectomy for operable breast cancer were prospectively treated by one of four methods: internal mammary chain dissection, internal mammary chain irradiation, both, or neither. A total of 1159 women were included in the analysis. Eligibility criteria included age younger than 70 years, tumor size less than 7 cm, and histologically positive nodes. Patients with T4 lesions were excluded. From 1958 to 1963 and from 1968 to 1972, all patients with histologic evidence of axillary node metastasis, regardless of tumor site, were treated with internal mammary chain dissection and postoperative radiation therapy. Some patients with medial quadrant lesions and negative axilla nodes underwent internal mammary chain dissection and radiation therapy. From 1963 to 1968, internal mammary chain dissection without postoperative irradiation was performed according to randomization in a multicenter trial. From 1972 to 1978, patients with lymph node–positive disease were randomly assigned to peripheral lymphatic irradiation or no further treatment. The 10-year risk of death from all causes was decreased (P = .06) for all patients who received internal mammary chain treatment of any kind compared with those who did not, but a statistically significant benefit was seen only in patients with medial quadrant lesions who had improved survival (P = .001) and lowered risk of metastasis (P = .005). Further adding fuel to the argument, each of the randomized trials showing survival advantages for postmastectomy irradiation included the internal mammary chain in the intended treatment volume. Although a more complete exploration of this issue has been published elsewhere,282 the basic philosophy at M. D. Anderson is to attempt to include the internal mammary chain in the treatment volume, provided that doing so does not compromise other aspects of treatment. For most patients, multifield chest wall irradiation allows better conformance of the radiation to the region at risk for tumor and lessens the volume of critical structures incidentally irradiated. This philosophy is consistent with the approaches used in later randomized trials evaluating postmastectomy irradiation (Table 17-9).237,243,244,259,283 The EORTC is conducting a trial (trial 22922) to assess the role of internal mammary and medial supraclavicular lymph node chain irradiation in stage I to III breast cancer. The National Cancer Institute of Canada recently completed accrual to its MA.20 trial, which was designed to assess the role

of regional nodal irradiation, including the internal mammary chain, in the context of breast conservation.

Target Volume Considerations At M. D. Anderson, the choice of treatment techniques and volumes for postmastectomy irradiation is the result of the synthesis of our own experience with important studies from other institutions. A substantial portion of the patient population has locally advanced breast cancer, which clearly influences our perceptions about the disease and, in particular, the techniques used for postmastectomy irradiation for patients with intermediate-stage disease. In our practice, metastases to the internal mammary chain, interpectoral region, and supraclavicular nodes are common. This results in a treatment philosophy that tends to be inclusive of all locoregional volumes at risk, even if encompassing these regions is technically difficult. Multiple adjacent fields are often used to optimize treatment and minimize irradiation of uninvolved adjacent structures. The quality of the plan for postmastectomy irradiation of the chest wall can be assessed by how well the plan (1) covers the breadth of the chest wall, (2) covers the ipsilateral internal mammary chain region, (3) minimizes lung exposure, and (4) avoids the heart.284 Further details of treatment techniques are discussed later under “Radiation Techniques and Treatment Planning.” One of the keys to the successful treatment of ­intermediate-stage and locally advanced breast cancer is obtaining a detailed and accurate definition of the extent of disease before the initiation of therapy. Because most ­patients with intermediate-stage and locally advanced breast cancer will be treated with neoadjuvant (preoperative) chemotherapy, the initial pathologic description of the disease (i.e., extent of disease in the breast and the lymph nodes) is not helpful to radiation oncologists in the subsequent ­decision-making process. The decision about whether radiation therapy is appropriate therefore may be based on the initial breast examination. Subtle skin involvement, attachment of the tumor to the underlying chest wall structures, and the presence of satellite lesions and multicentric tumors can ­affect local recurrence rates and determine whether BCT is possible if the tumor responds well. Radiologic or clinical ­evidence of tumor in the internal mammary, axillary apical, or supraclavicular nodal basins is important for correctly staging the disease and for planning local therapy. Areas of putative disease extent that are identified on radiologic studies but will not be surgically addressed (e.g., internal mammary chain or infraclavicular nodes) should be subjected to needle biopsy before the initiation of therapy. The systemic staging evaluation for patients with intermediate-stage and

386

PART 4  Breast

TABLE 17-9.    Radiation Sites and Doses in Selected Trials of Postmastectomy Radiation Therapy Study and Reference

Dose per Fraction

Chest Wall

Subclavicular Nodes

Axillary Nodes

IMC Supplement

DBCG243

2.0 Gy

Electrons 50 Gy

Photons ≥50 Gy

Selected patients

Electrons 50 Gy

British Columbia244

2.34 Gy

Tangents 27.5 Gy

Photons ≥35 Gy

Yes 35 Gy

Photons 35 Gy

Stockholm237

1.8-2.0 Gy

Electrons or tangents >45 Gy

Photons 45 Gy

Yes 45 Gy

Electrons or wide tangents 45 Gy

Oslo283

2.5 Gy

Not treated

Photons 50 Gy

Not treated

Photons 50 Gy

Dana-Farber259

1.8 Gy

Tangents 43 Gy

Photons 45 Gy

Selected patients

Not treated

DBCG, Danish Breast Cancer Cooperative Group; IMC, internal mammary chain.

locally advanced breast cancer is similar to that for patients  with early-stage disease except that a bone scan and abdominal CT or sonography are obtained even in the absence of clinical symptoms or biochemical abnormalities.

Breast Conservation for Advanced Disease A particularly exciting development in breast cancer treatment over the past 10 years has been the use of neoadjuvant chemotherapy for patients who present with advanced disease in the hope of sufficiently shrinking the tumor so that these patients can be offered BCT.284-288 T2 tumors larger than 4 cm (stage IIA disease) or T3 tumors without fixed or matted axillary nodes (stage IIB and IIIA disease) are technically operable by classic criteria. Although total mastectomy with axillary lymph node dissection may be an acceptable initial treatment choice for patients with significant comorbid conditions, neoadjuvant doxorubicin-based or taxanebased chemotherapy is becoming the preferred option for initial treatment. This strategy allows the responsiveness of the tumor to the chosen regimen to be assessed and allows some patients to subsequently undergo BCT rather than mastectomy. Patients considering BCT should undergo percutaneous placement of radiopaque markers in the tumor bed (typically under sonographic guidance) to facilitate future surgical resection.289 Patients who are to be treated initially with mastectomy should undergo adjuvant therapy with a doxorubicin-based or paclitaxel-based regimen ­before locoregional radiation if they are medically fit to do so. Patients being considered for BCT after neoadjuvant systemic therapy are first assessed by a multidisciplinary team that includes a surgeon, medical oncologist, diagnostic radiologist, and radiation oncologist. Most patients show objective responses to neoadjuvant chemotherapy, but only one fourth of patients with advanced disease at M. D. Anderson have become candidates for BCT on the basis of the following objective criteria290: complete resolution of skin edema, residual tumor size of less than 5 cm, and absence of known tumor multicentricity or extensive intramammary lymphatic invasion. If the response of the tumor to neoadjuvant chemotherapy is sufficient to fulfill these criteria for BCT, the surgeon attempts a wide local excision of the residual abnormality and an axillary dissection. Removal of all mammographically abnormal tissue (although not necessarily the i­nitial extent of disease) and histologically clear margins are

r­ equired. After surgery, patients undergo the remainder of the chemotherapy regimen followed by radiation therapy to the breast and regional lymphatics. The choice of radiation therapy technique used for BCT in the context of locally advanced breast cancer can be challenging. The target volume, which for early breast cancer is limited to the glandular breast mound, is extended in locally advanced cases to include the regional lymphatics and can include the axillary, interpectoral, and supraclavicular and internal mammary chain lymph nodes. Classically, treatment of the target volume requires the use of multiple adjacent fields, resulting in greater complexity of setup and reproducibility. Imprecise matching of adjacent fields can result in underdosing or overdosing the target volume, the latter of which results in normal tissue complications. Ideally, non-coplanar beams with precise matching techniques such as those recommended by Chu and colleagues291 are used when photon fields abut one another. Use of adjacent electron beam fields with shaped, matching field edges may be useful when part of the target area, such as the internal mammary chain, can be adequately covered with electrons. Typically, the breast and undissected lymphatics are treated to a dose of 50 Gy in 25 fractions over 5 weeks, followed by a 10-Gy boost to the tumor bed. Two important studies that support the use of breast conservation after neoadjuvant chemotherapy are the NSABP B-18 and EORTC 10902 trials, which compared neoadjuvant chemotherapy with adjuvant chemotherapy for patients with stage II or stage III breast cancer. Both studies reported that rates of BCT were higher for patients treated with neoadjuvant chemotherapy and that the overall breast recurrence risk for patients treated with neoadjuvant chemotherapy was not statistically different from that for patients treated with surgery first.230-232 In contrast, a meta-analysis of nine randomized studies comparing neoadjuvant and adjuvant chemotherapy reported that the use of neoadjuvant chemotherapy was associated with an increase in the relative risk of locoregional recurrence compared with adjuvant chemotherapy (relative risk = 1.22; 95% confidence interval, 1.04 to 1.43).233 This difference was largely influenced by the trials in which surgery was not performed and breast conservation after neoadjuvant chemotherapy was achieved with the use of radiation therapy alone. These findings suggest that even patients whose tumors show a complete clinical response would still benefit from surgery in addition to radiation. An analysis of the patients treated at M. D. Anderson with breast conservation after neoadjuvant chemotherapy

17  The Breast

has shown encouraging results. In that study of 340 carefully selected patients, the local recurrence rates were 5% at 5 years and 10% at 10 years despite the fact that 72% of the patients had clinical stage IIB or III disease.292 Four factors were found to be independently associated with breast cancer recurrence and locoregional recurrence: the presence of clinical N2 or N3 disease, lymphovascular space invasion, a multifocal pattern of residual disease, and residual disease larger than 2 cm in diameter. For the 84% of patients who had none or just one of these factors, the 10-year recurrence rate was only 4%.293 In contrast, for the 4% of patients who had three of these factors, the 10-year recurrence rate was 45%. Women with primary clinical T3 or T4 disease were at very low risk for recurrence if the tumor shrank to a solitary nidus or showed a pathologic complete response, but patients with T3 or T4 tumors that broke up and left a multifocal pattern of residual disease had a recurrence rate of 20%.292

Radiation Therapy for Adenocarcinoma in Axillary Nodes with an Unknown Primary Tumor Although adenocarcinoma presenting in axillary lymph nodes without an identifiable primary tumor site is an uncommon clinical entity, oncologists must be aware of its predictable clinical course. This presentation is usually the result of an occult primary breast tumor, although carcinomas of the thyroid, lung, stomach, and colorectum can metastasize to axillary nodes. The true incidence of this rare phenomenon is unknown. Reports from major referral centers indicate that it accounts for 0.3% to 1% of breast cancer cases.294-296

Initial Evaluation In most female patients presenting with isolated axillary disease, the source of the tumor seems to be the ipsilateral breast,297 but it is difficult to determine how extensive the investigation of other possible tumor sites should be. Reviews by Patel and colleagues298 and Fortunato and colleagues299 have asserted that exhaustive investigations are unlikely to yield additional information. Consequently, the following tests are probably adequate for the initial evaluation: a full history and physical examination, hemogram,

387

biochemical profile, bilateral mammography, chest radiography, and bone scan. Sonography of the breast and ­regional lymphatics may help to delineate the extent of disease or to detect mammographically occult lesions. Nonetheless, mammography remains the starting place for assessing the breast in this context. Other investigational tools that can be useful in the search for an occult primary tumor include MRI300,301and color Doppler sonography.302 A biopsy guided by MRI is possible but technically difficult; secondlook sonography with a sonographically guided biopsy may make the workup more efficient. If mammography, breast sonography, and breast MRI fail to disclose an obvious primary lesion, the subsequent therapeutic plan is based on the assumption that a microscopic cancer is present in the ipsilateral breast.

Treatment Since Halsted303 observed that three of his patients with adenocarcinoma in axillary nodes with an unknown primary tumor subsequently presented with an overt primary breast tumor, the classic treatment for patients with this presentation has included ipsilateral mastectomy. Careful evaluation of the mastectomy specimen often reveals a primary lesion in the breast, with the large range of specimen positivity (8% to 67%)294,298,304-308 reflecting variation in pathologists’ diligence in examining specimens  (Table 17-10).294,298,305-307,309 Tumors, when found, are usually no larger than a few millimeters. In most cases, axillary lymph nodes must reach a substantial size before the patient notices any abnormality, and the most common disease stage at presentation is T0 N2 (stage IIIA). After careful assessment of the extent and distribution of the regional disease—usually with a combination of physical and sonographic examinations of the axilla, infraclavicular fossa, and supraclavicular fossa—treatment is initiated in a manner consistent with the treatment of any other stage III breast tumor. Neoadjuvant chemotherapy is the norm, and typically six to eight cycles are given preoperatively. Locoregional therapy for most patients with adenocarcinoma in axillary nodes with unknown primary tumor consists of surgical excision of all gross disease (usually an axillary dissection) followed by radiation therapy to the breast and regional lymphatics. A few studies have investigated simple observation of the ipsilateral breast, but the existing data

TABLE 17-10.   Results of Mammography and Pathologic Evaluation of the Breast in Patients with Axillary Adenocarcinoma with Unknown Primary Tumor Number of Patients

Normal Mammograms/Total Mammograms

Tumor Identified in Mastectomy Specimen/ Breasts Examined

Study and Reference

Year of Publication

Roswell Park298

1981

29

9/17

16/29

Hope306

1986

20

19/20

5/11

Rush Presbyterian–St. Luke’s309

1987

11

11/11

11/11

M. D. Anderson Cancer Center305

1990

42

37/40

2/17

1991

35

23/32

22/32

1991

56

55/55

27/33

154/175

83/133 (62%)

City of

Memorial Milan307 Total

Sloan-Kettering294

193

388

PART 4  Breast

TABLE 17-11.    Axillary Adenocarcinoma with Unknown Primary Tumor: Results of Breast Observation Study and Reference

Year of Publication

Axillary Treatment*

Breast Failures/ Breasts Observed

Average Follow-up Period (yr)

Nottingham310

1987

Dissection

2/4

NS

M. D. Anderson Cancer Center305

1990

Biopsy and dissection

7/13

NS

Milan307

1992

Biopsy and dissection

9/17

6

Rotterdam313

1993

Dissection and biopsy

2/15

7.7

University of Mississippi311

1995

Dissection

7/8

3.5

Netherlands312

1995

Dissection

1/5

2.2

Total

28/62 (45%)

*In

each study, some patients were treated with lymphatic radiation, and others were not. NS, not stated.

TABLE 17-12.   Axillary Adenocarcinoma with Unknown Primary Tumor: Results of Breast Conservation Therapy Study and Reference

Year of Publication

Breast Failures/ Breasts Treated

Disease-Free Survival Rate (%)

Institut Curie314

1989

Ax Diss + RT RT alone

6/30

76

Western General Edinburgh296

1990

Ax Diss + RT

3/20

66

Milan307

1996

Ax Bx + RT Ax Diss + RT

2/6

NS

M. D. Anderson Cancer Center315

1996

Ax Bx + RT Ax Diss + RT

2/27

69

13/83

16

Treatment

Total

Ax Diss, axillary dissection; Ax Bx, axillary biopsy; NS, not stated; RT, radiation therapy to breast and lymphatics.

suggest that this approach is associated with a high probability of local failure in the observed breast, which often necessitates salvage mastectomy (Table 17-11).305,307,310-313 In most cases, surgical removal of the apparently normal breast has no therapeutic advantage because comprehensive irradiation will be required in any case. Definitive irradiation of the breast at the same time as ­adjuvant irradiation of the regional nodes is a logical combination. Whillis and colleagues296 treated 12 patients with  radiation therapy to the breast and lymphatics and ­reported no local recurrences at follow-up times ranging from 4 months to 89 months. Campana and colleagues314 also used conservative treatment; in their series of 30 patients, eight ­locoregional recurrences were evident, four in the breast and four in the axilla, with a median time to recurrence of  112 months in the breast and 23 months in the axilla. The survival rate at 10 years was 71%. These results are very similar to the ones obtained in M. D. Anderson ­series.315,316 All of these studies show that long follow-up times are necessary to determine the true local control rates because many relapses may not occur until several years after treatment. Evidently, local control rates approximating 85% are possible from using radiation therapy to the

breast. Most patients can retain their breasts and the few patients for whom BCT fails can usually be successfully treated with salvage mastectomy at the time of recurrence (Table 17-12).296,307,314-316 If regional nodes are irradiated without the breast being treated, it is difficult to irradiate the breast (or the chest wall if a salvage mastectomy is needed) because of uncertainties regarding junctions and overlap into the previously treated fields. One analysis of 27 patients with axillary node adenocarcinoma with an unknown primary tumor showed high rates of local and regional control after treatment with BCT.315 The 5-year actuarial local control rate in the breast was 100%, and the regional control rate at 5 years was 92.6%. In a separate analysis, no difference in locoregional control or survival rates could be detected between a mastectomybased approach and a breast-conservation approach.316 The ipsilateral breast is treated with tangential fields of megavoltage photons to a dose of 50 Gy in 25 fractions. The supraclavicular fossa and axillary apex are also treated to 50 Gy in 25 fractions, and because most patients have N2 disease, the midaxilla may be supplemented by using a posterior field to a dose of 40 to 50 Gy, depending on the pathologic extent of disease.

17  The Breast

Radiation Therapy for Inflammatory Breast Cancer Inflammatory breast cancer is a rare but virulent form of breast cancer, accounting for approximately 3% of all breast cancer cases.317 It is characterized by a clinical triad of ridging (i.e., distended lines in the breast), peau d’orange (i.e., orange-peel appearance of the breast), and diffuse erythema of the skin. Rapid onset of symptoms (i.e., within  3 months) is used to distinguish inflammatory breast ­cancer from locally advanced breast cancer that has secondarily involved the skin. The latter is a more indolent disease and is typically associated with longer survival.317,318 The skin changes associated with inflammatory breast cancer result from extensive involvement of dermal lymphatics, and a skin biopsy is usually positive for dermal lymphatic invasion, although positive findings on skin biopsy are not considered necessary to establish the diagnosis. The rare patients who present without the clinical triad of symptoms but have evidence of dermal lymphatic invasion on biopsy also seem to have a poor prognosis, although slightly better than that of patients with more obvious diffuse breast involvement.319 The AJCC TNM staging system designates inflammatory breast cancer as T4, automatically placing localized disease in the stage IIIB category. Physical examination of patients with inflammatory breast cancer reveals a swollen, tender breast that is warm to the touch and often associated with ipsilateral axillary and even supraclavicular lymphadenopathy. Approximately one fourth of the patients with inflammatory breast cancer present with distant metastases, and a careful staging workup is mandatory. Approximately one half of the patients with inflammatory breast cancer at M. D. Anderson present with a dominant mass, known as the common presentation; the remaining patients have no visible or palpable discrete mass, known as the classic presentation.320 Often, patients without a well-defined tumor mass have been treated with antibiotics for several weeks before the appropriate diagnosis is established. Because conventional imaging such as mammography may not be helpful in establishing the diagnosis, MRI is being investigated for this purpose.321 Compared with other breast tumors, inflammatory breast tumors tend to be poorly differentiated, are more often estrogen receptor– and progesterone receptor–negative, and are more likely to have HER2/NEU overexpression, TP53 mutations, and a high thymidine-labeling index.322-324 Before the availability of systemic chemotherapy combinations, inflammatory breast cancer was almost always ­fatal. Because local treatment with surgery, radiation ­therapy, or both modalities produced 5-year survival rates of 5% or less and median survival times of less than 15 months, a major goal of locoregional therapy was to provide palliation.325 The introduction of doxorubicin-based chemotherapy into clinical practice in the mid-1970s led to significant improvements in survival, with 5-year overall survival rates of 30% to 40%.326 With the introduction of additional active systemic agents, the outcome for patients with inflammatory breast cancer has improved still further.327 Management of inflammatory breast cancer requires carefully integrated care by a multidisciplinary team. Ideally, patients with inflammatory breast cancer should be evaluated in a multidisciplinary center at the time of

389

­ resentation to confirm the diagnosis of inflammatory disp ease, document the extent of disease (including photographs of areas of involved skin), and agree on a treatment plan. Inflammatory breast cancer is considered inoperable at presentation and should be treated initially with neoadjuvant chemotherapy; trastuzumab may be appropriate if the disease is HER2/NEU–positive. The optimal chemotherapy regimen for inflammatory breast cancer includes anthracyclines and taxanes.327 Approximately 80% of patients with inflammatory breast cancer will achieve a clinical response such that the disease becomes operable.327 Patients with operable disease should undergo a modified radical mastectomy, and all patients should receive postmastectomy radiation therapy delivered to the chest wall and the draining lymphatics. In an important study that helped to define appropriate locoregional treatment, Panades and others328 evaluated 308 patients with inflammatory breast cancer treated with chemotherapy and found that the 10-year local recurrence–free survival rates were significantly better for patients who underwent mastectomy than for those who did not (about 60% versus 34%,  P = .0001). The 10-year breast cancer–specific survival rates were also improved (34% with mastectomy versus 23% without, P = .005).328 A multivariate analysis that considered other potential prognostic factors found that the use of mastectomy remained a significant factor for improved local recurrence–free survival (P = .04).328 Results from an M. D. Anderson study confirmed these results, with multivariate analysis revealing that a complete or partial response to neoadjuvant chemotherapy, the use of radiation therapy, and the addition of mastectomy to the therapeutic regimen all significantly improved disease-specific survival rates.329 Later studies have focused on whether breast conservation can be safely performed for patients who achieved a complete clinical response to neoadjuvant chemotherapy.330-333 Some series have reported favorable control rates. However, Swain and Lippman344 reported a local recurrence rate of 30% despite a clinical complete response to neoadjuvant chemotherapy, often documented by several negative biopsies before irradiation. Low and colleagues332 also reported a 40% local recurrence rate in 15 patients with inflammatory breast cancer treated with radiation therapy alone after a biopsy-proven complete response. Brun and colleagues331 reported a 54% local failure rate with attempts at breast conservation, and Chevallier and colleagues335 reported a 61% local failure rate in patients treated with breast conservation after they had achieved a complete response to neoadjuvant chemotherapy. These findings lead us to consider mastectomy a standard component of locoregional therapy for all patients with inflammatory breast cancer who are being treated with curative intent. Radiation therapy has an important role in the management of inflammatory breast cancer. Use of radiation therapy after neoadjuvant chemotherapy and mastectomy has led to locoregional control rates of more than 80%.329 The optimal delivery schedule for that radiation therapy was explored at M. D. Anderson, where an accelerated hyperfractionated radiation delivery schedule was used in which 51 Gy was delivered to the chest wall and draining lymphatic fields in twice-daily 1.5-Gy fractions, followed by a boost to the chest wall to an additional 15 Gy, given in 10 1.5-Gy fractions twice daily. In an analysis of 192 patients treated with neoadjuvant chemotherapy, mastectomy, and

390

PART 4  Breast

postmastectomy radiation (most with this fractionation schedule), the 5-year locoregional control rate was 84%.329 Factors associated with higher rates of locoregional control included complete response to chemotherapy, age older than 45 years, achievement of negative margins, having three or fewer positive lymph nodes after chemotherapy, and the use of taxane chemotherapy. A multivariate ­model indicated that response to chemotherapy and surgical margin status were the two most important predictors of locoregional control. In terms of the extent of disease response, the 5-year locoregional control rates were 95% for patients with a clinical complete response, 86% for patients with a clinical partial response (>50% reduction in tumor size), and 51% for patients with less than a partial response. In terms of margin status, the 5-year locoregional control rates were 91% for patients with negative margins and 68% for patients with positive margins. The risk of complications was radiation dose–related, with 15% of patients treated to a dose of 60 Gy or less experiencing a late grade 3 or 4 complication compared with 25% of patients treated to a median dose of 66 Gy.329 The results of this analysis led us to recommend that doses of 66 Gy or higher be reserved for patients at higher risk of locoregional recurrence—  specifically, those with less than a partial response to neoadjuvant chemotherapy, positive surgical margins, significant residual disease after chemotherapy, or age younger than 45 years. However, no randomized trials have compared accelerated hyperfractionation with conventional fractionation for such patients, and many centers continue to use conventionally fractionated radiation therapy.

Radiation Therapy for Locoregional Recurrence after Mastectomy Locoregional recurrence of breast cancer after mastectomy represents a heterogeneous mix of entities resulting from a variety of causes, including inadequate initial treatment and virulent tumor biology. The prognosis for patients with an isolated chest wall nodule after a long disease-free interval is different from the prognosis for patients with an extensive, inflammatory-type recurrence involving the entire chest wall. Despite the wide variety of presentations of locoregional recurrence after mastectomy, a few principles can be applied in the evaluation and treatment of this phenomenon: • Treat with curative intent using aggressive local and systemic therapy. • Use initial chemotherapy or surgical excision to achieve resolution of gross disease. • Use large-field radiation techniques to include the site of the recurrence, the entire chest wall, and undissected regional lymphatics. • Treat to doses of at least 66 Gy in the boost volume. The most important principle is the first one—patients with locoregional recurrence should be treated with curative intent, not automatically treated with palliative intent as though they had visceral metastases. Although up to one half of patients with locoregional recurrence also manifest visceral metastases at presentation, a subgroup of patients who are free of documented metastasis can be long-term survivors when treated with curative intent.

Treatment of local or regionally recurrent breast cancer after initial mastectomy has evolved from simple surgical excision and localized irradiation to comprehensive multimodality therapy. Complete surgical excision, comprehensive irradiation, and systemic chemotherapy are now considered the standard of care. Occasionally, the recurrent disease can initially be completely resected with clear margins. In these circumstances, systemic therapy and locoregional radiation can be delivered in the adjuvant setting. More commonly, induction chemotherapy is required to render the disease resectable, and radiation therapy is delivered after surgery. Overall 5-year survival rates in the range of 35% to 50% have been reported, although 5-year disease-free survival rates are generally only about  25%.269-271,336 Locoregional control rates ranging from 42% to 57% have been reported with the use of modern surgical and radiation therapy techniques.269 The need for radiation therapy after adequate surgical excision is well supported by clinical experience; recurrence rates after surgical excision alone range from 67% to 76%.269,337 Early studies established the importance of radiation therapy technique in the treatment of locoregionally recurrent disease. In 1981, Bedwinek and colleagues reported the following local control rates: surgical excision alone, 24%; radiation therapy alone, 38%; and surgical excision plus radiation therapy, 40%. When the radiation therapy results were analyzed, adequate irradiation in combination with surgical excision was found to produce local control rates of 72%, compared with 28% when inadequate irradiation was used.269 This study established the criteria for optimal irradiation of locoregionally recurrent disease after initial mastectomy, including dose and volume recommendations that have continued to hold true to this day. The chest wall has consistently been the most common site of locoregional recurrence after mastectomy (see Tables 17-6 and 17-7). To prevent uncontrolled locoregional disease, which severely reduces the quality of life of ­individuals with recurrence, all sites of potential disease should be ­treated. At a minimum, the anatomic site or sites of disease and the entire ipsilateral chest wall and undissected lymphatic ­basins should be included within the target volume. Another reason that locoregional disease control is important is the suggestion that improved locoregional control may improve overall survival rates.270,338 This is inferred from the survival improvement seen after the use of postmastectomy radiation therapy (discussed earlier). Schwaibold and colleagues270 reported a 5-year survival rate of 64% for patients with locally controlled disease, compared with 35% for patients with locally uncontrolled disease (P = .001). The influence of local control on overall survival also was highly significant in a study from M. D. Anderson.338 Although the increased complexity of treatment has improved long-term survival for patients experiencing locoregional disease recurrence after mastectomy, the rates of disease-free survival and locoregional control of recurrent disease have changed little in the past 20 years. The locoregional control rates of 68% at 5 years and 55% at 10 years reported for patients treated with combined-modality therapy at M. D. Anderson267 are consistent with those of previous studies and represent, for reasons outlined previously, the benchmarks for locoregional control at M. D. ­Anderson. Factors that predict a poor outcome have been identified, with a short interval between initial mastectomy and ­recurrence

17  The Breast

and the presence of residual gross disease being particularly important, but more aggressive local therapy beyond comprehensive irradiation after complete surgical excision of the recurrence has done little to improve outcome, and hyperfractionated accelerated radiation therapy did not seem to substantially improve local control rates.267 Patients with locally recurrent disease that does not respond to initial chemotherapy have a particularly guarded prognosis. Treatment of macroscopic recurrences with radiation therapy alone is unlikely to be successful. In our series of patients treated with multimodal therapy, the presence of gross disease at the initiation of radiation therapy was associated with a 36% locoregional control rate.267 Inevitably, these individuals also developed distant metastases. Because of the typically massive amount of tumor present, it is not surprising that reasonable doses of radiation failed to control the disease. Although dose escalation may have some value for this population, concurrent chemotherapy and radiation therapy as an alternative is being investigated. Locoregionally recurrent breast cancer after mastectomy is a challenging but potentially curable condition. Adequate locoregional control is a prerequisite for improved overall survival. The appropriate treatment volume for these patients is similar to that in the postmastectomy setting. The entire chest wall and the undissected lymphatics of the axillary apex and supraclavicular fossa comprise the minimum treatment volume. Commonly, the internal mammary chain and the midaxilla are also treated. Because of the substantial number of treatment failures seen with the use of doses comparable with those used for postmastectomy irradiation, a 10% dose escalation (compared with the dose in other postmastectomy settings) is used when treating locoregionally recurrent disease. All areas treated prophylactically are treated to 54 Gy in 27 fractions, and all areas to receive boost doses because of microscopic disease receive an additional 12 Gy in 6 fractions.

Radiation Therapy for Palliation Radiation therapy is an effective palliative tool for patients with symptomatic breast cancer. The two main indications for palliative radiation therapy are relief of symptoms such as pain and malodorous discharge and control of local tumor growth to prevent complications resulting from increasing pressure or structural compromise. Examples of such ­complications are paralysis from spinal cord infringement and fractures resulting from extensive cortical destruction of weight-bearing bones. Optimal palliation is achieved by balancing the potential benefits of palliative treatment with the side effects, costs, and inconvenience that the patient must bear. The radiation therapy scheme used depends on the volume to be irradiated, the tolerance of the surrounding tissues, and comorbid conditions. Generally, higher cumulative doses and smaller fraction sizes are used for patients with a relatively good long-term outlook, whereas ultrarapid schedules are preferred for patients with rapidly evolving disease. Bone metastases respond particularly well to radiation therapy. Relief of pain and reversal of disability can be expected in most patients with bone metastases. Although some pain relief can be expected during the radiation ­therapy

391

itself, most of the benefit occurs after completion of treatment. After treatment, restoration of bone can be ­expected, but radiographic evidence of normalization is uncommon. It is important to keep patients physically active even though they have metastases. Effective use of nonsteroidal anti-­inflammatory agents, narcotic analgesics, and physical ­therapy support are crucial to good long-term function. Because patients with breast cancer that has metastasized only to the bone (i.e., bone-only disease) are likely to live for years or even decades, treatment must be designed to minimize late effects, and the possibility of future ­courses of radiation therapy must be anticipated. The careful selection of field borders and the use of an appropriate fractionation scheme are factors that contribute to these goals. Typically, 30 Gy is delivered in 10 fractions over 2 weeks, although higher doses may be appropriate for ­patients with indolent disease and 8 Gy in a single fraction can be used for patients with limited life expectancy. Patients with extensive soft tissue disease involving the breast, chest wall, or brachial plexus often require intensive radiation therapy to minimize pain, neuropathy, and woundcare problems. In patients with unresectable disease that is unresponsive to systemic agents, radiation therapy may provide substantial palliation. Because of the large volume of disease, high doses are required to achieve the desired effect. At a minimum, the region is treated to 45 Gy in 15 fractions with vigorous use of bolus doses over cutaneous nodules. Because it is common to see confluent moist desquamation and exfoliation of thin layers of normal skin overlying bulk tumor, a treatment break is usually required after this initial treatment course. After this break, an additional 15 to 30 Gy can be delivered to any residual tumor provided that no dose-limiting structures are located in the boost volume. Pilot studies of radiation therapy delivered concurrently with chemosensitizing agents such as paclitaxel, docetaxel, and capecitabine have produced encouraging response rates. Further study is required, however, before this approach can be recommended outside the setting of a clinical trial. The development of brain metastases usually represents the beginning of an accelerated phase of breast cancer. Whenever possible, symptomatic brain lesions should be excised to achieve the most rapid response. Although whole-brain irradiation is probably effective in controlling microscopic and small-volume macroscopic disease, these benefits must be balanced against the potential for late effects on normal brain parenchyma. Total dose and fraction size contribute to the risk of late effects in this setting. Fractions of 2 or 2.5 Gy are preferable to larger fractions. A dose of 30 Gy is commonly delivered to the whole brain, and boost doses may be directed at individual tumor nodules by using reduced-field or radiosurgical techniques. Spinal cord compression represents a true radiotherapeutic emergency. For patients with tumor-induced symptomatic compression of the cord, high-dose steroids should be begun immediately, followed by neurosurgical decompression and radiation therapy to be delivered as quickly as possible to prevent irreversible myelopathy. The diagnostic dilemma is to distinguish this clinical scenario from cord impingement by retropulsed bony fragments from a ­collapsed vertebra. In the latter case, surgical intervention is preferred because radiation therapy cannot reverse the mechanical abnormality. Typically, 30 Gy is delivered in 3- Gy fractions using parallel-opposed fields.

392

PART 4  Breast

Radiation Techniques and Treatment Planning

in this chapter is not an adequate substitute for the experience and careful attention to detail that are required for optimal patient care.

The objective of radiation therapy in the management of breast cancer is to minimize the risk of locoregional ­recurrence while simultaneously minimizing the risk of ­treatment-related normal tissue injury. To accomplish this goal, careful treatment planning is critical. Principles of ­ radiation biology, radiation physics, mathematics, and physical and radiographic anatomy all must be understood for optimal treatment to be delivered. Radiation fields used for the treatment of breast cancer are relatively standard, but each field must be individualized according to patient anatomy, volumes at risk, and the acceptable risk of normal tissue injury. The description of treatment fields provided

Treatment Fields Intact Breast The intact breast is most commonly irradiated by using opposed tangential fields. These two matched fields travel obliquely across the thorax on the side of the affected breast and include the entire ipsilateral breast but the smallest possible volume of the lung and heart. Several equally effective techniques are available for designing tangential radiation fields. Patients are usually treated in a supine position with their ipsilateral arm abducted and externally rotated (Fig. 17-10). For patients with very large breasts, a prone radiation technique can be considered (Fig.17-11).

Tumor bed

Incision

A B Tumor bed

C FIGURE 17-10.  Tangential fields used for breast preservation are shown. A, Projection of fields on the skin. B, Digitally reconstructed radiograph shows contours of the tumor bed (pink) and incision site (blue). C, In the isodose distribution, the 100% isodose line is located at the pectoralis surface.

17  The Breast

FIGURE 17-11.  CT scan of a patient in the prone position shows the treatment plan and isodose distribution for breastconserving radiation therapy.

Immobilization devices such as a vacuum-sealed beaded cradle that conforms to the patient’s individual anatomy are effective methods for precisely reproducing the patient position for daily treatments. In the past few years, CT simulation has become the de facto standard for designing radiation fields for the treatment of breast cancer. This technology allows breast tangential fields to be designed virtually on the basis of CT data sets from individual patients. Because the breast volume is difficult to distinguish from the surrounding connective tissue on CT, the results of physical examination also must be taken into account in the design of treatment fields. At M. D. Anderson, the practice is to mark the anatomic borders of the breast tissue with a thin radiopaque wire before scanning. After the CT data are acquired, a field isocenter is placed, and marks are placed on the patient. To improve reproducibility, the isocenter should not be placed in an area of the breast or chest wall with significant slope. The fields are then designed with treatment-planning software that uses the three­dimensional anatomic display available from the scans, and the plans are verified with reconstructed radiographs. The tangential fields are designed with a nondivergent deep (intrathoracic) field border. This can be accomplished with additional gantry rotation to compensate for the beam divergence or with the use of a half-beam block. The tangential fields are also collimated to match the slope of the chest wall. At a minimum, the tumor bed and clips are contoured during virtual simulation; other critical structures, such as the internal mammary chain or heart, may be selectively contoured. The use of CT simulation makes clear the extent of lung and heart volume in the proposed treatment fields. For cancer of the left breast, care should be taken to avoid irradiation of the heart, particularly the coronary arteries. ­Unfortunately, the left anterior descending artery often

393

lies very close to the deep edge of the tangential radiation fields. Most tangential fields include a portion of the low axilla as well as the breast. For patients at risk of having residual disease in axillary lymph nodes, the deep and cranial corner of the tangential field should remain within the thorax to ensure inclusion of the entire axillary target. A greater percentage of the axilla can be treated within the tangential fields by increasing the cranial edge of the field and making this field edge nondivergent, a technique often referred to as the high tangent technique. To accomplish this technique, the treatment couch is rotated such that the feet of the patient are moved away from the gantry. Mathematical formulas can be used to calculate the required rotation of the couch, but a simpler method is to use a device known as a rod and chain. This device consists of a radiopaque rod, which is placed at the nondivergent cranial edge of the tangent fields, and a chain, which hangs freely at a 90-degree angle from the end of the rod, just below the abducted ipsilateral arm.291 During CT simulation, the chain is contoured, and the couch is then virtually rotated until the rod and chain are superimposed. As the couch is rotated, the gantry and collimator angles are readjusted to ensure alignment of the medial and lateral edge wires and proper coverage of the chest wall. Areas of the treatment field cranial to the rod and chain are then blocked with a customized mounted cerrobend block or a multileaf collimator. The high tangent technique is also an effective strategy for treatment of tumors located high in the breast or in the tail of Spence. Selected patients with stage II disease treated with breast-conserving surgery should also be treated with radiation of the axillary apex and supraclavicular fossa. In those patients, this region is treated with a third field that is exactly matched to be cranial to the breast tangential fields. This third field is indicated for patients with involvement of four or more axillary lymph nodes. The use of a third field for patients with one to three positive nodes who have undergone an adequate axillary lymph node dissection (>15 lymph nodes examined) is controversial because the failure rate in the supraclavicular fossa is low.339 Treatment recommendations are individualized for these patients on the basis of their age, the presence and degree of extracapsular extension of disease (≥2 mm considered an indication to treat), the size of the lymph node metastasis, and the degree to which the patient is concerned about arm lymphedema. The technique for matching an axillary apex–supraclavicular fossa field to the tangents is described in the next section.

Chest Wall Postmastectomy irradiation always includes treatment of the chest wall. Tangential 6-MV photons or, occasionally, photons of higher energy are used. An adjacent, matching electron beam field is routinely added to treat the most medial portion of the chest wall and encompass the lymph nodes of the ipsilateral internal mammary chain. In addition to covering both targets, this treatment plan gives the added benefit of broad coverage of the chest wall while avoiding irradiation of excessive amounts of the intrathoracic contents. With this technique, the left ventricle can be completely excluded from the ­treatment volume for cancer of the left breast. ­Inclusion

394

PART 4  Breast

B

A

IMC in first 3 intercostal spaces

C FIGURE 17-12.  A, Postmastectomy radiation fields are shown. The blue medial field represents the medial electron field that is matched to the lateral photon tangent field (red) and the photon supraclavicular fossa field (purple). B, Digitally reconstructed radiograph shows the tangent fields. C, Digitally reconstructed radiograph shows the internal mammary chain (IMC) (arrow).

of a ­ maximum of 2 cm of lung usually is acceptable  (Figs. 17-12 and 17-13). Alternatively, electrons can be used to treat the chest wall and internal mammary nodes. Although this technique has the advantage of improved conformation of the treatment volume to the target volume, the risks of geographic miss of tumor or excess transmission into lung are higher. Variations in tissue thickness make treatment planning difficult, particularly for patients with irregular surface contours. Regardless of the technique used, the entire mastectomy flaps must be included in the treatment volume. This includes the entire length of the mastectomy scar and any clips and drain sites. Typically, the fields extend from the midsternum to at least the midaxillary line. Depending on the extent of original disease and patient anatomy, the fields may need to be extended even to the posterior axillary line. A common error in treatment planning is to skimp on the inferior border. This border should be placed approximately 2 cm caudad to the previous location of the inframammary sulcus.

The superior border of the chest wall fields abuts the supraclavicular field. To avoid junctional hot spots, one of several techniques must be used to eliminate a divergent edge. We routinely use the rod-and-chain technique.291 This technique has several advantages; the border can be placed in any desired orientation, the full length of the tangential beam is used, and the match line can be confirmed visually during treatment. The other alternative, use of a mono-  isocentric technique, has the disadvantage of allowing use of only one half of the available beam length (usually 20 cm) for chest wall treatment. For patients with long torsos, this results in an excessively large supraclavicular field and, as a consequence, treatment of a larger volume of lung. CT-based treatment planning is important because it allows precise visualization of the intrathoracic contents encompassed by the proposed treatment fields. Treatment of any cardiac volume should be actively avoided because such treatment is associated with late morbidity and mortality. Although most reports evaluating the cardiac morbidity of breast irradiation have studied the portion of myocardium

17  The Breast

395

100% isodose line, consistent with the recommendations of the International Commission on Radiation Units and  Measurements.340

Supraclavicular Fossa

A

B FIGURE 17-13.  The isodose distributions are shown for postmastectomy irradiation of the upper chest wall (A) and the lower chest wall (B).

treated, it is likely that the sensitive target structures are the epicardial vessels. The implication is that no safe volume of heart can be included in the high-dose volume. Tissue heterogeneity corrections are routinely used in the planning process to minimize the “hot spots” caused by increased lung transmission. The dose of 50 Gy in 25 fractions is specified at an isodose that encompasses the entire target and represents the target minimum dose. Moderate bolus schedules allow the treatment to be completed without interruption. Typically, a 3-mm or 5-mm bolus is used every other day for the first 2 to 3 weeks. Additional bolus can be used during the boost if the patient’s skin reaction allows it. Moderate erythema and dry desquamation are the desired end results. All patients receive a boost dose to the chest wall flaps. These fields are only slightly smaller than the initial chest wall fields; treatment is delivered with electrons, and two fields are often required. The skin and subcutaneous tissues are the primary targets and are included within the distal 90% line. For patients with negative surgical margins, 10 Gy is given in 5 fractions. For patients with close or positive margins, 14 to 16 Gy is the usual boost dose. Electron doses are all specified at the

The second obligate treatment target for postmastectomy radiation therapy and a component of treatment for selected patients after breast-conserving surgery are the lymphatics of the supraclavicular fossa and undissected axillary apex. After a classic level I and II axillary dissection, the undissected portion of the axillary apex lies directly beneath the pectoralis major muscle. The supraclavicular field extends from the jugular groove medially to the lateral edge of the pectoralis major muscle. Careful placement of the inferior border, which is also the superior border of the chest wall fields, to be as cephalad as possible can substantially reduce the volume of lung included in the aggregate treatment fields. Although photons are most commonly used to treat the supraclavicular and axillary apical nodes, the use of electrons may be considered for thin patients because all the structures are anterior. The dose for microscopic disease is 50 Gy in 25 fractions. To ensure a geometric match of this third field with the tangents, the tangential fields must have a nondivergent superior edge. Collimator rotation of the tangential fields leads to an angled cranial edge relative to the matched third field, and this angle must be corrected. The use of a rod and chain (described under “Intact Breast”) and a mounted block of the tissue above the rod and chain corrects the divergence and angulation problems. A half-beam block is used for the caudal edge of the supraclavicular ­fossa–­axillary apex field to eliminate divergence into the tangential fields. The ­ supraclavicular fossa–axillary apex field is anterior in its ­orientation; a 15-degree gantry rotation is used to ­exclude the spinal cord while including the supraclavicular fossa. The upper border of the supraclavicular field should be above the acromioclavicular joint. The medial border should be set on the pedicles of the vertebral bodies. The lateral border can be set at the coracoid process of the scapula to exclude the dissected axilla or lateral to the ­humeral head to include a portion of the dissected axilla. The targeted lymph node chain lateral to the coracoid becomes progressively deeper from the anterior surface, ending at a point deeper than the supraclavicular lymph nodes. Rarely, a posterior field that covers only the axilla above the rod and chain is added to supplement the dosage to this deeper anatomic region. Because axillary recurrences are rare after level I and II axillary dissection and because increases in the dose to the axilla increase the risk of arm edema, these fields (often called posterior ­ axillary boost fields) are used less commonly. A posterior axillary boost field is used only when concern exists about the ­adequacy of the axillary dissection or when foci of disease are present in the axillary soft tissues.

Axilla Specific radiation treatment of the dissected axilla is rarely required. When irradiation of the axilla is desired, most of the axillary dose is delivered through the anterior supraclavicular field by using photons. It is important to verify that all operative clips left by the surgeon in the axilla are included in the chest wall fields or the lateral portion of the supraclavicular field. In a typical patient, delivering 

396

PART 4  Breast

Undissected portion of axilla IMC in first 3 intercostal spaces

A

correction is mandatory to ensure selection of an electron beam that covers the nodes without excess exit into the lung. Ideally, the nodes are located within the distal 90% isodose line, and the dose of 50 Gy in 25 fractions is specified at the 100% isodose line. Some patients have anatomic conformations that are not well suited to the use of a separate electron beam field to treat the internal mammary nodes. This is especially true for obese patients, patients with a protuberant intact breast, and patients who have undergone breast reconstruction after mastectomy. The use of “half-wide” or “partly deep” tangent fields can be considered in these cases. The tangent fields are made extra deep at the superior border to encompass the contoured first three intercostal spaces. Additional field blocking is placed on the lower portion of the tangents, similar to the field blocking used in a standard conformation. This technique has several disadvantages. The volume of lung treated is invariably greater than the volume treated with other techniques. The upper field edges often extend well beyond the midline, resulting in treatment of a part of the contralateral breast. Moreover, immediate breast reconstruction can significantly impair radiation therapy planning. For these reasons, immediate breast reconstruction is strongly discouraged for patients with advanced breast cancer for whom postoperative irradiation is definitely to be delivered when success of that therapy substantially reduces locoregional failure rates and improves overall survival.341

Boost Fields Breast

B FIGURE 17-14.  Supraclavicular fields are shown for postmastectomy irradiation. A, Digitally reconstructed radiograph illustrates the undissected axilla (arrow) and internal mammary chain (IMC) (arrow). B, CT scan shows the supraclavicular field isodose distribution for a patient treated with a mixed electron and photon technique.

50 Gy to the supraclavicular fossa–axillary apical field ­results in 35 to 37 Gy at the midline axillary structures (Fig. 17-14). A posterior photon field is then used to supplement the midaxilla to the desired target dose. Because of the morbidity of axillary irradiation and the rarity of ­axillary failure, the usual dose is 40 Gy. Higher doses are used only for patients with extensive axillary soft tissue disease or an undissected axilla.

Internal Mammary Lymph Nodes When the decision has been made to treat the internal mammary nodes, either of two techniques can be used. The preferred technique is to use a separate electron beam field on the most medial chest wall. Because the nodes typically lie 3 to 3.5 cm from the midline, this field must extend even further laterally to compensate for beam constriction. The usual field extends 5 to 6 cm away from the midline and is shaped to precisely match the edge of the tangent fields. A small lateral tilt often results in better field matching at depth. CT-based treatment planning with ­heterogeneity

The excision cavity is routinely treated with a boost dose in all but the most exceptional cases. Use of a boost clearly reduces ipsilateral breast tumor recurrence rates, particularly for patients who are young or have close surgical margins13,161; however, boost dosing may have some adverse aesthetic effects.342 The tumor cavity and incision site are completely contoured, including any operative clips. The projected volume is then expanded by 2 cm to create the aperture for the electron beam. The electron energy is then chosen so that the entire contour is encompassed by the distal 90% isodose line, with the dose specified at the 100% isodose line.

Regional Nodes Tumors in the internal mammary, infraclavicular, or supraclavicular regions are typically not completely resected, and boost doses are delivered to these sites in patients who present with adenopathy in these regions. The target area is contoured using any diagnostic images available from initial and subsequent staging evaluations. An aperture is then developed by expanding the projected contour by 2 cm for the electron beam. The electron energy is then chosen so that the entire contour is encompassed by the distal 90% isodose line, with the dose specified at 100%. Figure 17-15 shows an example of plans for an infraclavicular boost; Figure 17-16   shows plans for an internal mammary chain boost; and  Figure 17-17 shows plans for a supraclavicular boost.

Dose Prescriptions Even though breast cancer is one of the most common types of cancer treated with radiation in the private and academic sectors, little consistency is evident in

17  The Breast

397

A

A

Undissected portion of axilla

B

B Undissected portion of axilla

C FIGURE 17-15.  Fields are shown for an infraclavicular boost. A, Projection of an infraclavicular boost field on the skin. B, Digitally reconstructed radiograph shows an infraclavicular boost with a contoured target. C, CT scan shows the isodose distribution of an infraclavicular boost.

dose prescriptions across centers in the United States. At  M. D. Anderson, the standard dosage for the initial fields is 50 Gy in 25 fractions, given over 5 weeks. This dose is typically prescribed at the pectoralis surface, making the breast “minimum” dose 50 Gy. For patients treated after 

FIGURE 17-16.  Fields are outlined for a boost dose to the internal mammary chain. A, Digitally reconstructed radiograph shows the boost area. B, The isodose distribution is shown for the boost dose. The green contoured target in the internal mammary region is covered by the 90% electron isodose curve shown in yellow.

breast-conserving surgery, the tumor bed dose is then boosted with electrons to an additional 10 Gy for patients with negative margins, 14 Gy for patients with margins of 2 mm or less, or 16 Gy for patients with focally positive margins. CT treatment planning is used to ensure that the targeted boost volume is encompassed by the 90% isodose curve. For patients treated with chest wall tangents after mastectomy, a similar prescription point is chosen just anterior to the pectoralis major muscle. After the initial  50 Gy is delivered, most of the chest wall is then given a boost with electron fields to a total dose of 60 Gy. All electron fields are prescribed to the Dmax point, and an electron beam energy is chosen with CT planning to ensure that 90% of the dose is delivered to the deep portion of the pectoralis major muscle.

Dosimetry Careful dosimetry is a critical component of effective radiation therapy for breast cancer. A CT-based, three­dimensional treatment planning system is used to calculate the dose in off-axis planes and to incorporate the

398

PART 4  Breast

A

B

C FIGURE 17-17.  Fields are shown for a supraclavicular boost. A, Projection of the supraclavicular boost dose on the skin. B, Digitally reconstructed radiograph shows the supraclavicular boost area. C, The isodose distribution is outlined.

c­ ontribution of scatter into the dose calculation ­algorithm. This dose modeling is accurate under most treatment planning circumstances. The goal of treatment ­ planning is to achieve the most homogeneous distribution of dose throughout the target volume. This can be achieved through the use of three-dimensional compensation using a field-in-field intensity-modulated approach for patients with intact breasts or for patients with large amounts of residual tissue after mastectomy. This technique has been shown to improve dose homogeneity within the volume and to reduce the incidence of acute reactions.343-346 The policy at M. D. Anderson since 2001 has been to use the field-in-field intensity-modulated technique to treat the intact breast. This technique involves modulation of beams by dynamic multileaf collimators that can selectively attenuate dosages to small areas of the field with respect to each other. The precision of this beam attenuation requires that the position of the breast be reproduced exactly on a daily basis and that small movements during normal respiration be minimized. The dose at any hot spots usually can be limited to less than 105% of the dose at the prescription point located in the central deep breast. The most common reason for higher doses at hot spots is a separation size (i.e.,

length of tissue the beam traverses in traveling through the field) that is too large for a beam energy of 6 MV. Under these circumstances, a mixture of higher-energy beams can be incorporated to reduce the dose at the hot spots relative to the deep central dose. However, the increased beam energy comes at the expense of a lower dose to the superficial regions of the breast. It is important to individualize these choices on the basis of the location of the tumor bed and the specifics of the individual case. The field-in-field technique usually is avoided for postmastectomy photon irradiation. Many patients undergoing such treatment have only a thin amount of tissue between air and lung, and the dose-modeling algorithms used for radiation planning may result in significant errors relative to the measured data. To avoid accentuating these dosimetric errors, standard wedged photon fields are typically preferred.

REFERENCES    1. Haagensen C. Diseases of the Breast, 2nd ed. Philadelphia: WB Saunders, 1971, pp 576-584.    2. Donegan WL. Diagnosis of mammary cancer. Major Probl Clin Surg 1967;5:34-69.

17  The Breast    3. Osborne MP, Kinne DW. Breast development and anatomy. In Harris JR, Hellman S, Henderson KC, (eds): Breast Disease. ­Philadelphia: JB Lippincott, 1987, pp 1-14.    4. Veronesi U. Rationale and indications for limited surgery in breast cancer: current data. World J Surg 1987;11:493-498.    5. Rosen PP, Lesser ML, Kinne DW, et al. Discontinuous or “skip” metastases in breast carcinoma. Analysis of 1228 axillary dissections. Ann Surg 1983;197:276-283.    6. Urban JA, Marjani MA. Significance of internal mammary lymph node metastases in breast cancer. Am J Roentgenol Radium Ther Nucl Med 1971;111:130-136.    7. Kawase K, Gayed IW, Hunt KK, et al. Use of lymphoscintigraphy defines lymphatic drainage patterns before sentinel lymph node biopsy for breast cancer. J Am Coll Surg 2006;203:  64-72.    8. Shahar KH, Buchholz TA, Delpassand E, et al. Lower and central tumor location correlates with lymphoscintigraphy drainage to the internal mammary lymph nodes in breast carcinoma. Cancer 2005;103:1323-1329.    9. Shahar KH, Hunt KK, Thames HD, et al. Factors predictive of having four or more positive axillary lymph nodes in patients with positive sentinel lymph nodes: implications for selection of radiation fields. Int J Radiat Oncol Biol Phys 2004;59:10741079.   10. Taylor G, Meltzer A. Inflammatory carcinoma of the breast. Am J Cancer 1995;33:33-49.   11. Wazer D, DiPetrillo T, Schmidt-Ullrich R, et al. Factors influencing cosmetic outcome and complication risk after conservative surgery and radiotherapy for early-stage breast carcinoma. J Clin Oncol 1992;10:356-363.   12. Rose M, Olivotto I, Cady B, et al. Conservative surgery and radiation therapy for early breast cancer. Long-term cosmetic results. Arch Surg 1989;124:153-157.   13. Romestaing P, Lehingue Y, Carrie C, et al. Role of a 10-Gy boost in the conservative treatment of early breast cancer. J Clin Oncol 1997;15:963-968.   14. Van Limbergen E, Rijnders A, van der Schueren E, et al. Cosmetic evaluation of breast conserving treatment for mammary cancer. 2. A quantitative analysis of the influence of radiation dose, fractionation schedules and surgical treatment techniques on cosmetic results. Radiother Oncol 1989;16:  253-257.   15. Schnitt S, Connolly J, Harris J, et al. Radiation-induced changes in the breast. Hum Pathol 1984;15:545-550.   16. Tralins A. Lactation after conservative breast surgery combined with radiation therapy. Am J Clin Oncol 1995;18:40-43.   17. Moran MS, Colasanto JM, Haffty BG, et al. Effects of breastconserving therapy on lactation after pregnancy. Cancer J 2005;11:399-403.   18. Tokunaga M, Land C, Tokuoka S, et al. Incidence of female breast cancer among atomic bomb survivors, 1950-1985. Radiat Res 1994;138:209-233.   19. Hildreth N, Shore R, Dvoretsky P. The risk of breast cancer after irradiation of the thymus in infancy. N Engl J Med 1989;321:1281-1284.   20. Hrubec Z, Boice J, Monson R, et al. Breast cancer after multiple chest flouroscopies: second follow-up of Massachusetts women with tuberculosis. Cancer Res 1989;49:229-234.   21. Bhatia S, Robison L, Oberlin O, et al. Breast cancer and other second neoplasms after childhood Hodgkin’s disease. N Engl J Med 1996;334:745-751.   22. Boice JD Jr, Harvey EB, Blettner M, et al. Cancer in the contralateral breast after radiotherapy for breast cancer. N Engl J Med 1992;326:781-785.   23. Fisher B, Anderson S, Redmond CK, et al. Reanalysis and results after 12 years of follow-up in a randomized clinical trial comparing total mastectomy with lumpectomy with or without irradiation in the treatment of breast cancer. N Engl J Med 1995;333:1456-1461.   24. Veronesi U, Luini A, Del Vecchio M, et al. Radiotherapy after breast-preserving surgery in women with localized cancer of the breast. N Engl J Med 1993;328:1587-1591.

399

  25. Clarke M, Collins R, Darby S, et al, for the Early Breast Cancer Trialists’ Group. Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomised trials. Lancet 2005;366:2087-2106.   26. Magrini SM, Papi MG, Bagnoli R, et al. [Late sequelae involving the breast after radiotherapy for lung metastasis of Wilms’ tumor. Experience of the Florence Radiotherapy and analysis of the literature]. Radiol Med (Torino) 1995;89:865-869.   27. American Cancer Society. Breast Cancer Facts & Figures 20072008. Atlanta: American Cancer Society, 2008, pp 2-8.   28. American Cancer Society. Cancer Facts & Figures 2008. Atlanta: American Cancer Society, 2008, pp 3-11.   29. Ravdin PM Cronin KA, Howlader N, et al. The decrease in breast cancer incidence in 2003 in the United States. N Engl J Med 2007;356:1670-1674.   30. Greenlee RT, Hill-Harmon MB, Murray T, et al. Cancer statistics, 2001. CA Cancer J Clin 2001;51:15-36.   31. American Cancer Society. Breast Cancer Facts & Figures 20052006. Atlanta: American Cancer Society, 2006, pp 1-25.   32. American Cancer Society. Breast Cancer Facts and Figures 19992000. Atlanta: American Cancer Society, 2000.   33. Lambe M, Hsieh C, Tsaih S, et al. Maternal risk of breast cancer following multiple births: a nationwide study in Sweden. Cancer Causes Control 1996;7:533-538.   34. Breast cancer and hormone replacement therapy: collaborative reanalysis of data from 51 epidemiological studies of 52,705 women with breast cancer and 108,411 women without breast cancer. Collaborative Group on Hormonal Factors in Breast Cancer. Lancet 1997;350:1047-1059.   35. Schairer C, Lubin J, Troisi R, et al. Menopausal estrogen and estrogen-progestin replacement therapy and breast cancer risk. JAMA 2000;283:485-491.   36. Chlebowski RT, Hendrix SL, Langer RD, et al. Influence of estrogen plus progestin on breast cancer and mammography in healthy postmenopausal women: the Women’s Health Initiative randomized trial. JAMA 2003;289:3243-3253.   37. Manson JE, Hsia J, Johnson KC, et al. Estrogen plus progestin and the risk of coronary heart disease. N Engl J Med 2003;349:  523-534.   38. Anderson GL, Limacher M, Assaf AR, et al. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women’s Health Initiative randomized controlled trial. JAMA 2004;291:1701-1712.   39. Shumaker SA, Legault C, Kuller L, et al. Conjugated equine estrogens and incidence of probable dementia and mild cognitive impairment in postmenopausal women: Women’s Health Initiative memory study. JAMA 2004;291:2947-2958.   40. Guzman RC, Yang J, Rajkumar L, et al. Hormonal prevention of breast cancer: mimicking the protective effect of pregnancy. Proc Natl Acad Sci U S A 1999;96:2520-2525.   41. Burke W, Daly M, Garber J, et al. Recommendations for followup care of individuals with an inherited predisposition to cancer. II. BRCA1 and BRCA2. Cancer Genetics Studies Consortium. JAMA 1997;277:997-1003.   42. Holmes MD, Hunter DJ, Colditz GA, et al. Association of dietary intake of fat and fatty acids with risk of breast cancer. JAMA 1999;281:914-920.   43. Prentice RL, Caan B, Chlebowski RT, et al. Low-fat dietary pattern and risk of invasive breast cancer: the Women’s Health Initiative Randomized Controlled Dietary Modification Trial. JAMA 2006;295:629-642.   44. Smith-Warner SA, Spiegelman D, Yaun SS, et al. Alcohol and breast cancer in women: a pooled analysis of cohort studies. JAMA 1998;279:535-540.   45. Carpenter CL, Ross RK, Paganini-Hill A, et al. Lifetime exercise activity and breast cancer risk among post-menopausal women. Br J Cancer 1999;80:1852-1858.   46. Shoff SM, Newcomb PA, Trentham-Dietz A, et al. Early-life physical activity and postmenopausal breast cancer: effect of body size and weight change. Cancer Epidemiol Biomarkers Prev 2000;9:591-595.

400

PART 4  Breast

  47. Kumar NB, Cantor A, Allen K, et al. Android obesity at diagnosis and breast carcinoma survival: evaluation of the effects of anthropometric variables at diagnosis, including body composition and body fat distribution and weight gain during life span, and survival from breast carcinoma. Cancer 2000;88:2751-2757.   48. Daley AJ, Crank H, Saxton JM, et al. Randomized trial of exercise therapy in women treated for breast cancer. J Clin Oncol 2007;25:1713-1721.   49. Milne HM, Wallman KE, Gordon S, et al. Effects of a combined aerobic and resistance exercise program in breast cancer survivors: a randomized controlled trial. Breast Cancer Res Treat 2008;108:279-288.   50. Markes M, Brockow T, Resch KL. Exercise for women receiving adjuvant therapy for breast cancer. Cochrane Database Syst Rev 2006;(4): CD005001.   51. Newcomb PA, Egan KM, Titus-Ernstoff L, et al. Lactation in relation to postmenopausal breast cancer. Am J Epidemiol 1999;150:174-182.   52. Benichou J, Gail MH, Mulvihill JJ. Graphs to estimate an individualized risk of breast cancer. J Clin Oncol 1996;14:103-110.   53. Gail MH, Brinton LA, Byar DP, et al. Projecting individualized probabilities of developing breast cancer for white females who are being examined annually. J Natl Cancer Inst 1989;81:  1879-1886.   54. Malkin D, Li FP, Strong LC, et al. Germline p53 mutations in a familial syndrome of breast cancer, sarcomas, and other neoplasms. Science 1990;250:1233-1238.   55. Velculescu VE, El-Deiry WS. Biological and clinical importance of the p53 tumor suppressor gene. Clin Chem 1996;42:858868.   56. Miyashita T, Krajewski S, Krajewska M, et al. Tumor suppressor p53 is a regulator of bcl-2 and bax gene expression in vitro and in vivo. Oncogene 1994;9:1799-1805.   57. Hollstein M, Soussi T, Thomas G, et al. P53 gene alterations in human tumors: perspectives for cancer control. Recent Results Cancer Res 1997;143:369-389.   58. Hall JM, Lee MK, Newman B, et al. Linkage of early-onset ­familial breast cancer to chromosome 17q21. Science 1990;250:  1684-1689.   59. Wooster R, Bignell G, Lancaster J, et al. Identification of the breast cancer susceptibility gene BRCA2. Nature 1995;378:  789-792.   60. Scully R, Chen J, Plug A, et al. Association of BRCA1 with Rad51 in mitotic and meiotic cells. Cell 1997;88:265-275.   61. Sharan SK, Morimatsu M, Albrecht U, et al. Embryonic lethality and radiation hypersensitivity mediated by Rad51 in mice lacking Brca2. Nature 1997;386:804-810.   62. Gowen LC, Avrutskaya AV, Latour AM, et al. BRCA1 required for transcription-coupled repair of oxidative DNA damage. Science 1998;281:1009-1012.   63. Zhong Q, Chen CF, Li S, et al. Association of BRCA1 with the hRad50-hMre11-p95 complex and the DNA damage response. Science 1999;285:747-750.   64. Ford D, Easton DF, Bishop DT, et al. Risks of cancer in BRCA1mutation carriers. Breast Cancer Linkage Consortium. Lancet 1994;343:692-695.   65. Struewing JP, Hartge P, Wacholder S, et al. The risk of cancer associated with specific mutations of BRCA1 and BRCA2 among Ashkenazi Jews. N Engl J Med 1997;336:1401-1408.   66. Friedman LS, Gayther SA, Kurosaki T, et al. Mutation analysis of BRCA1 and BRCA2 in a male breast cancer population. Am J Hum Genet 1997;60:313-319.   67. Mavraki E, Gray IC, Bishop DT, et al. Germline BRCA2 ­mutations in men with breast cancer. Br J Cancer 1997;76:  1428-1431.   68. Shattuck-Eidens D, Oliphant A, McClure M, et al. BRCA1 sequence analysis in women at high risk for susceptibility mutations. Risk factor analysis and implications for genetic testing. JAMA 1997;278:1242-1250.   69. Burke W, Culver JO, Bowen D, et al. Genetic counseling for women with an intermediate family history of breast cancer. Am J Med Genet 2000;90:361-368.

  70. American Society of Clinical Oncology. Statement of the American Society of Clinical Oncology: genetic testing for cancer susceptibility, adopted on February 20, 1996. J Clin Oncol 1996;14:1730-1736; discussion 1737-1740.   71. Warner E, Plewes DB, Hill KA, et al. Surveillance of BRCA1 and BRCA2 mutation carriers with magnetic resonance imaging, ultrasound, mammography, and clinical breast examination. JAMA 2004;292:1317-1325.   72. Saslow D, Boetes C, Burke W, et al. American Cancer Society guidelines for breast screening with MRI as an adjunct to mammography. CA Cancer J Clin 2007;57:75-89.   73. Perou CM, Sorlie T, Eisen MB, et al. Molecular portraits of ­human breast tumours. Nature 2000;406:747-752.   74. Sorlie T, Perou CM, Tibshirani R, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses  with clinical implications. Proc Natl Acad Sci U S A 2001;98:  10869-10874.   75. Sorlie T, Tibshirani R, Parker J, et al. Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc Natl Acad Sci U S A 2003;100:8418-8423.   76. Fisher B, Costantino J, Redmond C, et al. A randomized clinical trial evaluating tamoxifen in the treatment of patients with node-negative breast cancer who have estrogen-receptor-­positive tumors. N Engl J Med 1989;320:479-484.   77. Adjuvant tamoxifen in the management of operable breast cancer: the Scottish Trial. Report from the Breast Cancer Trials Committee, Scottish Cancer Trials Office (MRC), Edinburgh. Lancet 1987;2:171-175.   78. Fornander T, Rutqvist LE, Cedermark B, et al. Adjuvant tamoxifen in early breast cancer: occurrence of new primary cancers. Lancet 1989;1:117-120.   79. Fisher B, Costantino JP, Wickerham DL, et al. Tamoxifen for prevention of breast cancer: report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. J Natl Cancer Inst 1998;90:1371-1388.   80. Vogel VG, Costantino JP, Wickerham DL, et al. Effects of tamoxifen vs raloxifene on the risk of developing invasive breast cancer and other disease outcomes: the NSABP Study of Tamoxifen and Raloxifene (STAR) P-2 trial. JAMA 2006;295:2727-2741.   81. Hartmann LC, Schaid DJ, Woods JE, et al. Efficacy of bilateral prophylactic mastectomy in women with a family history of breast cancer. N Engl J Med 1999;340:77-84.   82. Rebbeck TR, Friebel T, Lynch HT, et al. Bilateral prophylactic mastectomy reduces breast cancer risk in BRCA1 and BRCA2 mutation carriers: the PROSE Study Group. J Clin Oncol 2004;22:1055-1062.   83. Rebbeck TR, Lynch HT, Neuhausen SL, et al. Prophylactic oophorectomy in carriers of BRCA1 or BRCA2 mutations. N Engl J Med 2002;346:1616-1622.   84. Greene FL, Page DL, Fleming ID, et al (eds): AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 223-240.   85. Singletary SE, Connolly JL. Breast cancer staging: working with the sixth edition of the AJCC Cancer Staging Manual. CA Cancer J Clin 2006;56:37-47: quiz 50-31.   86. Woodward WA, Strom EA, Tucker SL, et al. Changes in the 2003 American Joint Committee on Cancer staging for breast cancer dramatically affect stage-specific survival. J Clin Oncol 2003;21:3244-3248.   87. Brito RA, Valero V, Buzdar AU, et al. Long-term results of combined-modality therapy for locally advanced breast cancer with ipsilateral supraclavicular metastases: The University of Texas M. D. Anderson Cancer Center experience. J Clin Oncol 2001;19:628-633.   88. Consensus conference on the classification of ductal carcinoma in situ. Hum Pathol 1997;28:1221-1225.   89. Fisher ER, Leeming R, Anderson S, et al. Conservative management of intraductal carcinoma (DCIS) of the breast. Collaborating NSABP investigators. J Surg Oncol 1991;47:139-147.   90. Baxter NN, Virnig BA, Durham SB, et al. Trends in the treatment of ductal carcinoma in situ of the breast. J Natl Cancer Inst 2004;96:443-448.

17  The Breast   91. Shugg D White VM, Kitchen PR, et al. Surgical management of ductal carcinoma in situ in Australia in 1995. ANZ J Surg 2002;72:708-715.   92. Fisher B, Land S, Mamounas E, et al. Prevention of invasive breast cancer in women with ductal carcinoma in situ: an ­update of the National Surgical Adjuvant Breast and Bowel Project experience. Semin Oncol 2001;28:400-418.   93. Bijker N, Meijnen P, Peterse JL, et al. Breast-conserving treatment with or without radiotherapy in ductal carcinoma-in-situ: ten-year results of European Organisation for Research and Treatment of Cancer randomized phase III trial 10853—a study by the EORTC Breast Cancer Cooperative Group and EORTC Radiotherapy Group. J Clin Oncol 2006;24:3381-3387.   94. Houghton J, George WD, Cuzick J, et al. Radiotherapy and tamoxifen in women with completely excised ductal carcinoma in situ of the breast in the UK, Australia, and New Zealand: randomised controlled trial. Lancet 2003;362:95-102.   95. Emdin SO, Granstrand B, Ringberg A, et al. SweDCIS: radiotherapy after sector resection for ductal carcinoma in situ of the breast. Results of a randomised trial in a population offered mammography screening. Acta Oncol 2006;45:536-543.   96. Fisher B, Dignam J, Wolmark N, et al. Lumpectomy and radiation therapy for the treatment of intraductal breast cancer: findings from National Surgical Adjuvant Breast and Bowel Project B-17. J Clin Oncol 1998;16:441-452.   97. Fisher ER, Dignam J, Tan-Chiu E, et al. Pathologic findings from the National Surgical Adjuvant Breast Project (NSABP) eightyear update of protocol B-17: intraductal carcinoma. Cancer 1999;86:429-438.   98. Julien JP, Bijker N, Fentiman IS, et al. Radiotherapy in breastconserving treatment for ductal carcinoma in situ: first results of the EORTC randomised phase III trial 10853. EORTC Breast Cancer Cooperative Group and EORTC Radiotherapy Group. Lancet 2000;355:528-533.   99. Ringberg A, Nordgren H, Thorstensson S, et al. Histopathological risk factors for ipsilateral breast events after breast conserving treatment for ductal carcinoma in situ of the breast—results from the Swedish randomised trial. Eur J Cancer 2007;43:291-298. 100. Silverstein MJ, Lagios MD, Craig PH, et al. A prognostic ­index for ductal carcinoma in situ of the breast. Cancer 1996;77:  2267-2274. 101. Silverstein MJ, Lagios MD, Groshen S, et al. The influence of margin width on local control of ductal carcinoma in situ of the breast. N Engl J Med 1999;340:1455-1461. 102. MacDonald HR, Silverstein MJ, Lee LA, et al. Margin width as the sole determinant of local recurrence after breast conservation in patients with ductal carcinoma in situ of the breast. Am J Surg 2006;192:420-422. 103. Silverstein MJ. The University of Southern California/Van Nuys prognostic index for ductal carcinoma in situ of the breast. Am J Surg 2003;186:337-343. 104. Wong JS, Kaelin CM, Troyan SL, et al. Prospective study of wide excision alone for ductal carcinoma in situ of the breast. J Clin Oncol 2006;24:1031-1036. 105. Hughes L, Wang M, Page D, et al. Five-year results of intergroup study E5194: local excision alone (without radiation treatment) for selected patients with ductal carcinoma in situ (DCIS) ­[abstract]. Breast Cancer Res Treat 2006;100:A29. 106. Fisher B, Dignam J, Wolmark N, et al. Tamoxifen in treatment of intraductal breast cancer: National Surgical Adjuvant Breast and Bowel Project B-24 randomised controlled trial. Lancet 1999;353:1993-2000. 107. Allred D, Bryant J, Land S, et al. Estrogen receptor expression as a predictive marker of the effectiveness of tamoxifen in the treatment of DCIS: findings from NSABP protocol B-24 ­[abstract]. Breast Cancer Res Treat 2002;76(Suppl 1):A30. 108. Yen TW, Hunt KK, Mirza NQ, et al. Physician recommendations regarding tamoxifen and patient utilization of tamoxifen after surgery for ductal carcinoma in situ. Cancer 2004;100:942-949. 109. Ganz PA, Kwan L, Stanton AL, et al. Quality of life at the end of primary treatment of breast cancer: first results from the moving beyond cancer randomized trial. J Natl Cancer Inst 2004;96:376-387.

401

110. Polychemotherapy for early breast cancer: an overview of the randomised trials. Early Breast Cancer Trialists’ Collaborative Group. Lancet 1998;352:930-942. 111. Tamoxifen for early breast cancer: an overview of the randomised trials. Early Breast Cancer Trialists’ Collaborative Group. Lancet 1998;351:1451-1467. 112. Fisher B, Anderson S, Bryant J, et al. Twenty-year follow-up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy plus irradiation for the treatment of invasive breast cancer. N Engl J Med 2002;347:1233-1241. 113. Early Breast Cancer Trialists’ Collaborative Group. Favourable and unfavourable effects on long-term survival of radiotherapy for early breast cancer: an overview of the randomised trials. Lancet 2000;355:1757-1770. 114. Morris AD, Morris RD, Wilson JF, et al. Breast-conserving thera­ py vs mastectomy in early-stage breast cancer: a meta-analysis of 10-year survival. Cancer J Sci Am 1997;3:6-12. 115. Lim M, Bellon JR, Gelman R, et al. A prospective study of conservative surgery without radiation therapy in select patients with stage I breast cancer. Int J Radiat Oncol Biol Phys 2006;65:1149-1154. 116. Liljegren G, Lindgren A, Bergh J, et al. Risk factors for local recurrence after conservative treatment in stage I breast cancer. Definition of a subgroup not requiring radiotherapy. Ann Oncol 1997;8:235-241. 117. Holli K, Saaristo R, Isola J, et al. Lumpectomy with or without postoperative radiotherapy for breast cancer with favourable prognostic features: results of a randomized study. Br J Cancer 2001;84:164-169. 118. Forrest AP, Stewart HJ, Everington D, et al. Randomised control­ led trial of conservation therapy for breast cancer: 6-year analysis of the Scottish trial. Scottish Cancer Trials Breast Group. Lancet 1996;348:708-713. 119. Fyles AW, Manchul L, McCready D, et al. Long-term results of a randomized trial of tamoxifen with or without radiation in women over 50 years of age with T1/2 N0 breast cancer ­[abstract]. Int J Radiat Oncol Biol Phys 2006;66:S4-S5. 120. Fisher B, Bryant J, Dignam JJ, for the National Surgical Adjuvant Breast and Bowel Project. Tamoxifen, radiation therapy, or both for prevention of ipsilateral breast tumor recurrence after lumpectomy in women with invasive breast cancers of one centimeter or less. J Clin Oncol 2002;20:4141-4149. 121. Hughes KS, Schnaper LA, Berry D, et al. Lumpectomy plus tamoxifen with or without irradiation in women 70 years of age or older with early breast cancer. N Engl J Med 2004;351:971-977. 122. Hughes KS, Schnaper LA, Berry D, et al. Lumpectomy plus tamoxifen with or without irradiation in women 70 years of age or older with early breast cancer: a report of further follow-up [abstract 11]. Breast Cancer Res Treat 2006;100:S6. 123. Smith BD, Gross CP, Smith GL, et al. Effectiveness of radiation therapy for older women with early breast cancer. J Natl Cancer Inst 2006;98:681-690. 124. Stovall M, Blackwell CR, Cundiff J, et al. Fetal dose from radiotherapy with photon beams: report of AAPM Radiation Therapy Committee Task Group No. 36. Med Phys 1995;22:63-82. 125. Morris MM, Powell SN. Irradiation in the setting of collagen vascular disease: acute and late complications. J Clin Oncol 1997;15:2728-2735. 126. Chen AM, Obedian E, Haffty BG. Breast-conserving therapy in the setting of collagen vascular disease. Cancer J 2001;7:480491. 127. Fowble BL, Schultz DJ, Overmoyer B, et al. The influence of young age on outcome in early stage breast cancer. Int J Radiat Oncol Biol Phys 1994;30:23-33. 128. Haffty BG, Fischer D, Rose M, et al. Prognostic factors for ­local recurrence in the conservatively treated breast cancer patient: a cautious interpretation of the data. J Clin Oncol 1991;9:  997-1003. 129. Halverson KJ, Perez CA, Taylor ME, et al. Age as a prognostic factor for breast and regional nodal recurrence following breast conserving surgery and irradiation in stage I and II breast cancer. Int J Radiat Oncol Biol Phys 1993;27:1045-1050.

402

PART 4  Breast

130. Kurtz JM, Jacquemier J, Amalric R, et al. Breast-conserving therapy for macroscopically multiple cancers. Ann Surg 1990;212:38-44. 131. Fisher ER, Anderson S, Redmond C, et al. Ipsilateral breast tumor recurrence and survival following lumpectomy and irradiation: pathological findings from NSABP protocol B-06. Semin Surg Oncol 1992;8:161-166. 132. Buchholz TA, Tucker SL, Erwin J, et al. Impact of systemic treatment on local control for patients with lymph node-negative breast cancer treated with breast-conservation therapy. J Clin Oncol 2001;19:2240-2246. 133. Antonini N, Jones H, Horiot JC, et al. Effect of age and radiation dose on local control after breast conserving treatment: EORTC trial 22881-10882. Radiother Oncol 2006;82:265-271. 134. Oh JL, Bonnen M, Outlaw ED, et al. The impact of young age on locoregional recurrence after doxorubicin-based breast conservation therapy in patients 40 years old or younger: how young is “young”? Int J Radiat Oncol Biol Phys 2006;65:1345-1352. 135. Beadle BM, Woodward WA, Allen PK, et al. Locoregional recurrence rates in young women with breast cancer by treatment approach [abstract]. Int J Radiat Oncol Biol Phys 2007;69:S212. 136. Joslyn SA. Radiation therapy and patient age in the survival from early-stage breast cancer. Int J Radiat Oncol Biol Phys 1999;44:821-826. 137. Hebert-Croteua N, Brisson J, Latreille J, et al. Compliance with consensus recommendations for the treatment of early-stage breast cancer in elderly women. Cancer 1999;85:1104-1113. 138. Solin LJ, Schultz DJ, Fowble BL. Ten-year results of the treatment of early-stage breast carcinoma in elderly women using breast-conserving surgery and definitive breast irradiation. Int J Radiat Oncol Biol Phys 1995;33:45-51. 139. Newman LA, Kuerer HM, Hunt KK, et al. Local recurrence and survival among black women with early-stage breast cancer treated with breast-conservation therapy or mastectomy. Ann Surg Oncol 1999;6:241-248. 140. Chabner E, Nixon A, Gelman R, et al. Family history and treatment outcome in young women after breast-conserving surgery and radiation therapy for early-stage breast cancer. J Clin Oncol 1998;16:2045-2051. 141. Haas JA, Schultz DJ, Peterson ME, et al. An analysis of age and family history on outcome after breast-conservation treatment: the University of Pennsylvania experience. Cancer J Sci Am 1998;4:308-315. 142. Haffty BG, Harrold E, Khan AJ, et al. Outcome of ­conservatively managed early-onset breast cancer by BRCA1/2 status. Lancet 2002;359:1471-1477. 143. Pierce LJ, Levin AM, Rebbeck TR, et al. Ten-year multi­institutional results of breast-conserving surgery and ­radiotherapy in BRCA1/2-associated stage I/II breast cancer. J Clin Oncol 2006;24:2437-2443. 144. Kauff ND, Satagopan JM, Robson ME, et al. Risk-reducing ­salpingo-oophorectomy in women with a BRCA1 or BRCA2 mutation. N Engl J Med 2002;346:1609-1615. 145. Fisher ER. Lumpectomy margins and much more. Cancer 1997;79:1453-1458. discussion 1459-1460. 146. Peterson ME, Schultz DJ, Reynolds C, et al. Outcomes in breast cancer patients relative to margin status after treatment with breast-conserving surgery and radiation therapy: the University of Pennsylvania experience. Int J Radiat Oncol Biol Phys 1999;43:1029-1035. 147. Vicini FA, Goldstein NS, Pass H, et al. Use of pathologic factors to assist in establishing adequacy of excision before radiotherapy in patients treated with breast-conserving therapy. Int J Radiat Oncol Biol Phys 2004;60:86-94. 148. Wazer DE, Jabro G, Ruthazer R, et al. Extent of margin positivity as a predictor for local recurrence after breast conserving irradiation. Radiat Oncol Investig 1999;7:111-117. 149. Boyages J, Recht A, Connolly JL, et al. Early breast cancer: pre­ dictors of breast recurrence for patients treated with conservative surgery and radiation therapy. Radiother Oncol 1990;19:29-41. 150. Schnitt SJ, Abner A, Gelman R, et al. The relationship between microscopic margins of resection and the risk of local recurrence in patients with breast cancer treated with breast-conserving surgery and radiation therapy. Cancer 1994;74:1746-1751.

151. Park CC, Mitsumori M, Nixon A, et al. Outcome at 8 years after breast-conserving surgery and radiation therapy for invasive breast cancer: influence of margin status and systemic therapy on local recurrence. J Clin Oncol 2000;18:1668-1675. 152. Freedman G, Fowble B, Hanlon A, et al. Patients with early stage invasive cancer with close or positive margins treated with conservative surgery and radiation have an increased risk of breast recurrence that is delayed by adjuvant systemic therapy. Int J Radiat Oncol Biol Phys 1999;44:1005-1015. 153. Jobsen JJ, van der Palen J, Ong F, et al. The value of a positive margin for invasive carcinoma in breast-conservative treatment in relation to local recurrence is limited to young women only. Int J Radiat Oncol Biol Phys 2003;57:724-731. 154. Wazer DE, Schmidt-Ullrich RK, Ruthazer R, et al. Factors determining outcome for breast-conserving irradiation with margindirected dose escalation to the tumor bed. Int J Radiat Oncol Biol Phys 1998;40:851-858. 155. Leopold KA, Recht A, Schnitt SJ, et al. Results of conservative surgery and radiation therapy for multiple synchronous cancers of one breast. Int J Radiat Oncol Biol Phys 1989;16:11-16. 156. Veronesi U, Salvadori B, Luini A, et al. Conservative treatment of early breast cancer. Long-term results of 1232 cases treated with quadrantectomy, axillary dissection, and radiotherapy. Ann Surg 1990;211:250-259. 157. Poen JC, Tran L, Juillard G, et al. Conservation therapy for invasive lobular carcinoma of the breast. Cancer 1992;69:  2789-2795. 158. Vo TN, Meric-Bernstam F, Yi M, et al. Outcomes of breast­conservation therapy for invasive lobular carcinoma are equivalent to those for invasive ductal carcinoma. Am J Surg 2006;192:  552-555. 159. Harris EE, Hwang WT, Lee EA, et al. The impact of HER-2 status on local recurrence in women with stage I-II breast cancer treated with breast-conserving therapy. Breast J 2006;12:  431-436. 160. Schnitt SJ, Hayman J, Gelman R, et al. A prospective study of conservative surgery alone in the treatment of selected patients with stage I breast cancer. Cancer 1996;77:1094-1100. 161. Bartelink H, Horiot JC, Poortmans P, et al. Impact of radiation dose on local control, fibrosis and survival after breast ­conserving treatment: 10 years results of the EORTC trial 22881-10882 [abstract]. Breast Cancer Res Treat 2006;100:A10. 162. Clark RM, Whelan T, Levine M, et al. Randomized clinical trial of breast irradiation following lumpectomy and axillary dissection for node-negative breast cancer: an update. Ontario Clinical Oncology Group. J Natl Cancer Inst 1996;88:1659-1664. 163. Liljegren G, Holmberg L, Bergh J, et al. 10-Year results after sector resection with or without postoperative ­ radiotherapy for stage I breast cancer: a randomized trial. J Clin Oncol 1999;17:2326-2333. 164. Veronesi U, Marubini E, Mariani L, et al. Radiotherapy after breast-conserving surgery in small breast carcinoma: long-term results of a randomized trial. Ann Oncol 2001;12:997-1003. 165. Weed DW, Edmundson GK, Vicini FA, et al. Accelerated partial breast irradiation: a dosimetric comparison of three different techniques. Brachytherapy 2005;4:121-129. 166. Orecchia R, Ciocca M, Lazzari R, et al. Intraoperative radiation therapy with electrons (ELIOT) in early-stage breast cancer. Breast 2003;12:483-490. 167. Veronesi U, Orecchia R, Luini A, et al. A preliminary report of intraoperative radiotherapy (IORT) in limited-stage breast cancers that are conservatively treated. Eur J Cancer 2001;37:21782183. 168. Vaidya JS, Tobias JS, Baum M, et al. TARGeted Intraoperative radiotherapy (TARGIT): an innovative approach to partialbreast irradiation. Semin Radiat Oncol 2005;15:84-91. 169. Kozak KR, Smith BL, Adams J, et al. Accelerated partial-breast irradiation using proton beams: initial clinical experience. Int J Radiat Oncol Biol Phys 2006;66:691-698. 170. Arthur DW, Winter K, Kuske RR, et al. A phase II trial of brachytherapy alone after lumpectomy for select breast cancer: tumor control and survival outcomes of RTOG 95-17. Int J Radiat Oncol Biol Phys 2008;72:467-473.

17  The Breast 171. Kuske RR, Winter K, Arthur DW, et al. Phase II trial of brachytherapy alone after lumpectomy for select breast cancer: toxicity analysis of RTOG 95-17. Int J Radiat Oncol Biol Phys 2006;65:45-51. 172. Vicini F, Winter K, Straube W, et al. A phase I/II trial to evaluate three-dimensional conformal radiation therapy confined to the region of the lumpectomy cavity for stage I/II breast carcinoma: initial report of feasibility and reproducibility of Radiation Therapy Oncology Group (RTOG) study 0319. Int J Radiat Oncol Biol Phys 2005;63:1531-1537. 173. Polgar C, Fodor J, Major T, et al. Breast-conserving treatment with partial or whole breast irradiation for low-risk invasive breast carcinoma—5-year results of a randomized trial. Int J Radiat Oncol Biol Phys 2007;69:694-702. 174. Benitez PR, Keisch ME, Vicini F, et al. Five-year results: the initial clinical trial of MammoSite balloon brachytherapy for partial breast irradiation in early-stage breast cancer. Am J Surg 2007;194:456-462. 175. Fentiman IS, Poole C, Tong D, et al. Inadequacy of iridium implant as sole radiation treatment for operable breast cancer. Eur J Cancer 1996;32A:608-611. 176. King TA, Bolton JS, Kuske RR, et al. Long-term results of widefield brachytherapy as the sole method of radiation therapy after segmental mastectomy for T(is,1,2) breast cancer. Am J Surg 2000;180:299-304. 177. Perera F, Yu E, Engel J, et al. Patterns of breast recurrence in a pilot study of brachytherapy confined to the lumpectomy site for early breast cancer with six years’ minimum follow-up. Int J Radiat Oncol Biol Phys 2003;57:1239-1246. 178. Vicini FA, Horwitz EM, Lacerna MD, et al. Long-term outcome with interstitial brachytherapy in the management of patients with early-stage breast cancer treated with breast-conserving therapy. Int J Radiat Oncol Biol Phys 1997;37:845-852. 179. Arthur DW, Vicini FA. Accelerated partial breast irradiation as a part of breast conservation therapy. J Clin Oncol 2005;23:  1726-1735. 180. Magee B, Swindell R, Harris M, et al. Prognostic factors for breast recurrence after conservative breast surgery and radiotherapy: results from a randomised trial. Radiother Oncol 1996;39:  223-227. 181. Baglan KL, Sharpe MB, Jaffray D, et al. Accelerated partial breast irradiation using 3D conformal radiation therapy (3D-CRT). Int J Radiat Oncol Biol Phys 2003;55:302-311. 182. Keisch M, Vicini F, Kuske RR, et al. Initial clinical experience with the MammoSite breast brachytherapy applicator in ­women with early-stage breast cancer treated with breast-conserving therapy. Int J Radiat Oncol Biol Phys 2003;55:289-293. 183. Vicini F, Beitsch PD, Quiet CA, et al. Three-year analysis of treatment efficacy, cosmesis, and toxicity by the American Society of Breast Surgeons MammoSite Breast Brachytherapy Registry Trial in patients treated with accelerated partial breast ­irradiation (APBI). Cancer 2008;112:758-766. 184. Chao KK, Vicini FA, Wallace M, et al. Analysis of treatment efficacy, cosmesis, and toxicity using the MammoSite breast brachytherapy catheter to deliver accelerated partial-breast irradiation: the William Beaumont Hospital experience. Int J Radiat Oncol Biol Phys 2007;69:32-40. 185. Chen S, Dickler A, Kirk M, et al. Patterns of failure after MammoSite brachytherapy partial breast irradiation: a detailed analysis. Int J Radiat Oncol Biol Phys 2007;69:25-31. 186. Cuttino LW, Keisch M, Jenrette JM, et al. Multi-institutional experience using the MammoSite radiation therapy system in the treatment of early-stage breast cancer: 2-year results. Int J Radiat Oncol Biol Phys 2008;71:107-114. 187. Tsai PI, Ryan M, Meek K, et al. Accelerated partial breast irradiation using the MammoSite device: early technical experience and short-term clinical follow-up. Am Surg 2006;72:929-934. 188. Fisher B, Montague E, Redmond C, et al. Findings from NSABP protocol no. B-04—comparison of radical mastectomy with alternative treatments for primary breast cancer. I. Radiation compliance and its relation to treatment outcome. Cancer 1980;46:1-13.

403

189. Fisher B, Jeong JH, Anderson S, et al. Twenty-five-year follow-up of a randomized trial comparing radical mastectomy, total mastectomy, and total mastectomy followed by irradiation. N Engl J Med 2002;347:567-575. 190. Louis-Sylvestre C, Clough K, Asselain B, et al. Axillary treatment in conservative management of operable breast cancer: dissection or radiotherapy? Results of a randomized study with 15 years of follow-up. J Clin Oncol 2004;22:97-101. 191. Wazer DE, Erban JK, Robert NJ, et al. Breast conservation in elderly women for clinically negative axillary lymph nodes without axillary dissection. Cancer 1994;74:878-883. 192. Haffty BG, McKhann C, Beinfield M, et al. Breast conservation therapy without axillary dissection. A rational treatment strategy in selected patients. Arch Surg 1993;128:1315-1319; discussion 1319. 193. Wong JS, Recht A, Beard CJ, et al. Treatment outcome after tangential radiation therapy without axillary dissection in patients with early-stage breast cancer and clinically negative axillary nodes. Int J Radiat Oncol Biol Phys 1997;39:915920. 194. Wilke LG, McCall LM, Posther KE, et al. Surgical complications associated with sentinel lymph node biopsy: results from a prospective international cooperative group trial. Ann Surg Oncol 2006;13:491-500. 195. Giuliano AE, Jones RC, Brennan M, et al. Sentinel lymphadenectomy in breast cancer. J Clin Oncol 1997;15:2345-2350. 196. Krag D, Weaver D, Ashikaga T, et al. The sentinel node in breast cancer—a multicenter validation study. N Engl J Med 1998;339:941-946. 197. Veronesi U, Paganelli G, Viale G, et al. Sentinel lymph node biopsy and axillary dissection in breast cancer: results in a large series. J Natl Cancer Inst 1999;91:368-373. 198. Veronesi U, Paganelli G, Viale G, et al. A randomized comparison of sentinel-node biopsy with routine axillary dissection in breast cancer. N Engl J Med 2003;349:546-553. 199. Chu KU, Turner RR, Hansen NM, et al. Do all patients with sentinel node metastasis from breast carcinoma need complete axillary node dissection? Ann Surg 1999;229:536-541. 200. Van Zee KJ, Manasseh DM, Bevilacqua JL, et al. A nomogram for predicting the likelihood of additional nodal metastases in breast cancer patients with a positive sentinel node biopsy. Ann Surg Oncol 2003;10:1140-1151. 201. Katz A, Smith BL, Golshan M, et al. Nomogram for the prediction of having four or more involved nodes for sentinel lymph node-positive breast cancer. J Clin Oncol 2008;26:2093-2098. 202. Kohrt HE, Olshen RA, Bermas HR, et al. New models and ­online calculator for predicting non-sentinel lymph node status in sentinel lymph node positive breast cancer patients. BMC Cancer 2008;8:66-71. 203. Chagpar AB. Predicting extensive nodal disease in women with breast cancer. Ann Surg Oncol 2006;13:3-4. 204. Katz A, Niemierko A, Gage I, et al. Factors associated with involvement of four or more axillary nodes for sentinel lymph node-positive patients. Int J Radiat Oncol Biol Phys 2006;65:  40-44. 205. Early Breast Cancer Trialists’ Group. Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet 2005;365:1687-1717. 206. Henderson IC, Berry DA, Demetri GD, et al. Improved outcomes from adding sequential paclitaxel but not from escalating doxorubicin dose in an adjuvant chemotherapy regimen for patients with node-positive primary breast cancer. J Clin Oncol 2003;21:976-983. 207. Mamounas EP, Bryant J, Lembersky BC, et al. Paclitaxel following doxorubicin / cyclophosphamide as adjuvant ­chemotherapy for node-positive breast cancer: results from NSABP B-28 ­[abstract]. Proc Am Soc Clin Oncol 2003;22:A12. 208. Martin M, Peienkowski T, Mackey JR, et al. TAC improves disease free survival and overall survival over FAC in node positive early breast cancer patients, BCIRG 001: 55 months follow-up [abstract]. Breast Cancer Res Treat 2003;82:A43.

404

PART 4  Breast

209. Bria E, Nistico C, Cuppone F, et al. Benefit of taxanes as adjuvant chemotherapy for early breast cancer: pooled analysis of 15,500 patients. Cancer 2006;106:2337-2344. 210. Citron ML, Berry DA, Cirrincione C, et al. Randomized trial of dose-dense versus conventionally scheduled and sequential versus concurrent combination chemotherapy as postoperative adjuvant treatment of node-positive primary breast cancer: first report of Intergroup Trial C9741/Cancer and Leukemia Group B Trial 9741. J Clin Oncol 2003;21:1431-1439. 211. Howell A, Cuzick J, Baum M, et al. Results of the ATAC (Arimidex, Tamoxifen, Alone or in Combination) trial after completion of 5 years’ adjuvant treatment for breast cancer. Lancet 2005;365:60-62. 212. Thurlimann B, Keshaviah A, Coates AS, et al. A comparison of letrozole and tamoxifen in postmenopausal women with early breast cancer. N Engl J Med 2005;353:2747-2757. 213. Coombes RC, Hall E, Gibson LJ, 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;350:1081-1092. 214. Goss PE, Ingle JN, Martino S, 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:  1793-1802. 215. Joensuu H, Kellokumpu-Lehtinen PL, Bono P, et al. Adjuvant docetaxel or vinorelbine with or without trastuzumab for breast cancer. N Engl J Med 2006;354:809-820. 216. Piccart-Gebhart MJ, Procter M, Leyland-Jones B, et al. Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N Engl J Med 2005;353:1659-1672. 217. Romond EH, Perez EA, Bryant J, et al. Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med 2005;353:1673-1684. 218. Slamon D, Eiermann W, Robert N, et al. Phase III randomized trial comparing doxorubicin and cyclophosphamide followed by docetaxel (ACT) with doxorubicin and cyclophosphamide followed by docetaxel and trastuzumab (ACTH) with docetaxel, carboplatin and trastuzumab (TCH) in HER2 positive early breast cancer patients: BCIRG 006 study [abstract]. Breast Cancer Res Treat 2005;94;(Suppl 1):1. 219. Recht A, Come SE, Henderson IC, et al. The sequencing of chemotherapy and radiation therapy after conservative surgery for early-stage breast cancer. N Engl J Med 1996;334:  1356-1361. 220. Bellon JR, Come SE, Gelman RS, et al. Sequencing of chemotherapy and radiation therapy in early-stage breast cancer: updated results of a prospective randomized trial. J Clin Oncol 2005;23:1934-1940. 221. Buchholz TA, Hunt KK, Amosson CM, et al. Sequencing of chemotherapy and radiation in lymph node-negative breast cancer. Cancer J Sci Am 1999;5:159-164. 222. Sartor CI, Peterson BL, Woolf S, et al. Effect of addition of adjuvant paclitaxel on radiotherapy delivery and locoregional control of node-positive breast cancer: Cancer and Leukemia Group B 9344. J Clin Oncol 2005;23:30-40. 223. Bentzen SM, Skoczylas JZ, Overgaard M, et al. Radiotherapyrelated lung fibrosis enhanced by tamoxifen. J Natl Cancer Inst 1996;88:918-922. 224. Albain K, Barlow W, O’Malley F, et al. Concurrent (CAFT) versus sequential (CAF-T) chemohormonal therapy (cyclophosphamide, doxorubicin, 5-fluorouracil, tamoxifen) versus T alone for postmenopausal, node-positive, estrogen (ER) and/ or progesterone (PgR) receptor-positive breast cancer: mature outcomes and new biologic correlates on phase III intergroup trial 0100 (SWOG-8814) [abstract]. Breast Cancer Res Treat 2005;89:A37. 225. Pierce LJ, Hutchins LF, Green SR, et al. Sequencing of tamoxifen and radiotherapy after breast-conserving surgery in early-stage breast cancer. J Clin Oncol 2005;23:24-29. 226. Ahn PH, Vu HT, Lannin D, et al. Sequence of radiotherapy with tamoxifen in conservatively managed breast cancer does not ­affect local relapse rates. J Clin Oncol 2005;23:17-23.

227. Harris EE, Christensen VJ, Hwang WT, et al. Impact of concurrent versus sequential tamoxifen with radiation therapy in early-stage breast cancer patients undergoing breast conservation treatment. J Clin Oncol 2005;23:11-16. 228. Tan-Chiu E, Yothers G, Romond E, et al. Assessment of cardiac dysfunction in a randomized trial comparing doxorubicin and cyclophosphamide followed by paclitaxel, with or without trastuzumab as adjuvant therapy in node-positive, human epidermal growth factor receptor 2-overexpressing breast cancer: NSABP B-31. J Clin Oncol 2005;23:7811-7819. 229. Halyard MY, Pisansky TM, Solin LJ, et al. Adjuvant radiotherapy (RT) and trastuzumab in stage I-IIA breast cancer: toxicity data from North Central Cancer Treatment Group phase III trial N9831 [abstract]. Proc Am Soc Clin Oncol 2006;24;­(Suppl):523. 230. Fisher B, Brown A, Mamounas E, et al. Effect of preoperative chemotherapy on local-regional disease in women with operable breast cancer: findings from National Surgical Adjuvant Breast and Bowel Project B-18. J Clin Oncol 1997;15:2483-2493. 231. Wolmark N, Wang J, Mamounas E, et al. Preoperative chemotherapy in patients with operable breast cancer: nine-year results from National Surgical Adjuvant Breast and Bowel Project B-18. Cancer 2001;30:96-102. 232. van der Hage JA, Cornelis JH, van de Velde CJ, et al. Preoperative chemotherapy in primary operable breast cancer: results from the European Organization for Research and Treatment of Cancer trial 10902. J Clin Oncol 2001;19:4224-4237. 233. Mauri D, Pavlidis N, Ioannidis JP. Neoadjuvant versus adjuvant systemic treatment in breast cancer: a meta-analysis. J Natl Cancer Inst 2005;97:188-194. 234. Early Breast Cancer Trialists’ Group. Effects of radiotherapy and surgery in early breast cancer. An overview of the randomized trials. Early Breast Cancer Trialists’ Collaborative Group. N Engl J Med 1995;333:1444-1455. 235. Host H, Brennhovd IO, Loeb M. Postoperative radiotherapy in breast cancer—long-term results from the Oslo study. Int J Radiat Oncol Biol Phys 1986;12:727-732. 236. McArdle CS, Crawford D, Dykes EH, et al. Adjuvant radiotherapy and chemotherapy in breast cancer. Br J Surg 1986;73:264-266. 237. Rutqvist LE, Pettersson D, Johansson H. Adjuvant radiation therapy versus surgery alone in operable breast cancer: longterm follow-up of a randomized clinical trial. Radiother Oncol 1993;26:104-110. 238. Velez-Garcia E, Carpenter JT Jr, Moore M, et al. Postsurgical adjuvant chemotherapy with or without radiotherapy in women with breast cancer and positive axillary nodes: a South-Eastern Cancer Study Group (SEG) Trial. Eur J Cancer 1992;28A:1833-1837. 239. Spanos WJ, Montague ED, Fletcher GH. Late complications of radiation only for advanced breast cancer. Int J Radiat Oncol Biol Phys 1980;6:1473-1476. 240. Strom EA, McNeese MD, Fletcher GH, et al. Results of mastectomy and postoperative irradiation in the management of locoregionally advanced carcinoma of the breast. Int J Radiat Oncol Biol Phys 1991;21:319-323. 241. Cuzick J, Stewart H, Peto R, et al. Overview of randomized trials of postoperative adjuvant radiotherapy in breast cancer. Cancer Treat Rep 1987;71:15-29. 242. Overgaard M, Christensen JJ, Johansen H, et al. Evaluation of radiotherapy in high-risk breast cancer patients: report from the Danish Breast Cancer Cooperative Group (DBCG 82) Trial. Int J Radiat Oncol Biol Phys 1990;19:1121-1124. 243. Overgaard M, Hansen PS, Overgaard J, et al. Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. Danish Breast Cancer Cooperative Group 82b Trial. N Engl J Med 1997;337:949-955. 244. Ragaz J, Jackson SM, Le N, et al. Adjuvant radiotherapy and chemotherapy in node-positive premenopausal women with breast cancer. N Engl J Med 1997;337:956-962. 245. Overgaard M, Jensen MB, Overgaard J, et al. Postoperative radiotherapy in high-risk postmenopausal breast-cancer patients given adjuvant tamoxifen: Danish Breast Cancer Cooperative Group DBCG 82c randomised trial. Lancet 1999;353:1641-1648.

17  The Breast 246. Ragaz J, Olivotto IA, Spinelli JJ, et al. Locoregional radiation therapy in patients with high-risk breast cancer receiving adju­ vant chemotherapy: 20-year results of the British Columbia randomized trial. J Natl Cancer Inst 2005;97:116-126. 247. Overgaard M, Nielsen HM, Overgaard J. Is the benefit of postmastectomy irradiation limited to patients with four or more positive nodes, as recommended in international consensus reports? A subgroup analysis of the DBCG 82 b&c randomized trials. Radiother Oncol 2007;82:247-253. 248. Madden JL. Modified radical mastectomy. Surg Gynecol Obstet 1965;121:1221-1230. 249. Madden JL, Kandalaft S, Bourque RA. Modified radical mastectomy. Ann Surg 1972;175:624-634. 250. Pierce LJ, Oberman HA, Strawderman MH, et al. Microscopic extracapsular extension in the axilla: is this an indication for axillary radiotherapy? Int J Radiat Oncol Biol Phys 1995;33:  253-259. 251. Harris JR, Halpin-Murphy P, McNeese M, et al. Consensus statement on postmastectomy radiation therapy. Int J Radiat Oncol Biol Phys 1999;44:989-990. 252. Fowble B, Gray R, Gilchrist K, et al. Identification of a subgroup of patients with breast cancer and histologically positive axillary nodes receiving adjuvant chemotherapy who may benefit from postoperative radiotherapy. J Clin Oncol 1988;6:1107-1117. 253. Recht A, Gray R, Davidson NE, et al. Locoregional failure 10 years after mastectomy and adjuvant chemotherapy with or without tamoxifen without irradiation: experience of the Eastern Cooperative Oncology Group. J Clin Oncol 1999;17:  1689-1700. 254. Levine MN, Bramwell VH, Pritchard KI, et al. Randomized ­trial of intensive cyclophosphamide, epirubicin, and fluorouracil chemotherapy compared with cyclophosphamide, methotrexate, and fluorouracil in premenopausal women with node-­positive breast cancer. National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 1998;16:2651-2658. 255. Blomqvist C, Tiusanen K, Elomaa I, et al. The combination of radiotherapy, adjuvant chemotherapy (cyclophosphamidedoxorubicin-ftorafur) and tamoxifen in stage II breast cancer. Long-term follow-up results of a randomised trial. Br J Cancer 1992;66:1171-1176. 256. Buzdar AU, Blumenschein GR, Smith TL, et al. Adjuvant chemotherapy with fluorouracil, doxorubicin, and cyclophosphamide, with or without bacillus Calmette-Guérin and with or without irradiation in operable breast cancer. A prospective randomized trial. Cancer 1984;53:384-389. 257. Buzzoni R, Bonadonna G, Valagussa P, et al. Adjuvant chemotherapy with doxorubicin plus cyclophosphamide, methotrexate, and fluorouracil in the treatment of resectable breast ­cancer with more than three positive axillary nodes. J Clin Oncol 1991;9:2134-2140. 258. Fisher B, Redmond C, Wickerham DL, et al. Doxorubicin­containing regimens for the treatment of stage II breast cancer: The National Surgical Adjuvant Breast and Bowel Project experience. J Clin Oncol 1989;7:572-582. 259. Griem KL, Henderson IC, Gelman R, et al. The 5-year results of a randomized trial of adjuvant radiation therapy after chemo­ therapy in breast cancer patients treated with mastectomy. J Clin Oncol 1987;5:1546-1555. 260. Katz A, Strom EA, Buchholz TA, et al. Locoregional recurrence patterns after mastectomy and doxorubicin-based chemotherapy: implications for postoperative irradiation. J Clin Oncol 2000;18:2817-2827. 261. Taghian A, Jeong JH, Mamounas E, et al. Patterns of locoregional failure in patients with operable breast cancer treated by mastectomy and adjuvant chemotherapy with or without tamoxifen and without radiotherapy: results from five National Surgical Adjuvant Breast and Bowel Project randomized clinical trials. J Clin Oncol 2004;22:4247-4254. 262. Overgaard M. Overview of randomized trials in high risk breast cancer patients treated with adjuvant systemic therapy with or without postmastectomy irradiation. Semin Radiat Oncol 1999;9:292-299.

405

263. Katz A, Strom EA, Buchholz TA, et al. The influence of pathologic tumor characteristics on locoregional recurrence rates following mastectomy. Int J Radiat Oncol Biol Phys 2000;50:  735-742. 264. Truong PT, Woodward WA, Thames HD, et al. The ratio of positive to excised nodes identifies high-risk subsets and reduces inter-institutional differences in locoregional recurrence risk estimates in breast cancer patients with 1-3 positive nodes: an analysis of prospective data from British Columbia and the M. D. Anderson Cancer Center. Int J Radiat Oncol Biol Phys 2007;68:59-65. 265. Overgaard M, Christensen JJ, Johansen H, et al. Postmastectomy irradiation in high-risk breast cancer patients. Present status of the Danish Breast Cancer Cooperative Group trials. Acta Oncol 1988;27:707-714. 266. Andry G, Suciu S, Vico P, et al. Locoregional recurrences after 649 modified radical mastectomies: incidence and significance. Eur J Surg Oncol 1989;15:476-485. 267. Ballo MT, Strom EA, Prost H, et al. Local-regional control of recurrent breast carcinoma after mastectomy: does hyperfractionated accelerated radiotherapy improve local control? Int J Radiat Oncol Biol Phys 1999;44:105-112. 268. Crowe JP Jr., Gordon NH, Antunez AR, et al. Local-regional breast cancer recurrence following mastectomy. Arch Surg 1991;126:429-432. 269. Bedwinek JM, Fineberg B, Lee J, et al. Analysis of failures following local treatment of isolated local-regional recurrence of breast cancer. Int J Radiat Oncol Biol Phys 1981;7:581-585. 270. Schwaibold F, Fowble BL, Solin LJ, et al. The results of radiation therapy for isolated local regional recurrence after mastectomy. Int J Radiat Oncol Biol Phys 1991;21:299-310. 271. Toonkel LM, Fix I, Jacobson LH, et al. The significance of local recurrence of carcinoma of the breast. Int J Radiat Oncol Biol Phys 1983;9:33-39. 272. Buchholz TA, Katz A, Strom EA, et al. Pathologic tumor size and lymph node status predict for different rates of locoregional recurrence after mastectomy for breast cancer patients treated with neoadjuvant versus adjuvant chemotherapy. Int J Radiat Oncol Biol Phys 2002;53:880-888. 273. Buchholz TA, Tucker SL, Masullo L, et al. Predictors of localregional recurrence after neoadjuvant chemotherapy and mastectomy without radiation. J Clin Oncol 2002;20:17-23. 274. Huang EH, Tucker SL, Strom EA, et al. Postmastectomy radiation improves local-regional control and survival for selected patients with locally advanced breast cancer treated with neoadjuvant chemotherapy and mastectomy. J Clin Oncol 2004;22:  4691-4699. 275. McGuire SE, Gonzalez-Angulo AM, Huang EH, et al. Postmastectomy radiation improves the outcome of patients with ­locally advanced breast cancer who achieve a pathologic complete response to neoadjuvant chemotherapy. Int J Radiat Oncol Biol Phys 2007;68:1004-1009. 276. Bucalossi P, Veronesi U, Zingo L, et al. Enlarged mastectomy for breast cancer. Review of 1,213 cases. Am J Roentgenol Radium Ther Nucl Med 1971;111:119-122. 277. Handley RS. Carcinoma of the breast. Ann R Coll Surg Engl 1975;57:59-66. 278. Ki KY, Shen ZZ. An analysis of 1242 cases of extended radical mastectomy. Breast 1984;10:10-19. 279. Huang O, Wang L, Shen K, et al. Breast cancer subpopulation with high risk of internal mammary lymph nodes metastasis: analysis of 2,269 Chinese breast cancer patients treated with extended radical mastectomy. Breast Cancer Res Treat 2008;107:379-387. 280. Arriagada R, Lê MG, Mouriesse H, et al. Long-term effect of internal mammary chain treatment. Results of a multivariate analysis of 1195 patients with operable breast cancer and positive axillary nodes. Radiother Oncol 1988;11:213-222. 281. Lê MG, Arriagada R, de Vathaire F, et al. Can internal mam­mary chain treatment decrease the risk of death for patients with medical breast cancer and positive axillary lymph nodes? Cancer 1990;66:2313-2318.

406

PART 4  Breast

282. Strom EA, McNeese MD. Postmastectomy irradiation: rationale for treatment field selection. Semin Radiat Oncol 1999;9:247-253. 283. Auquier A, Rutqvist LE, Host H, et al. Post-mastectomy megavoltage radiotherapy: the Oslo and Stockholm trials. Eur J Cancer 1992;28:433-437. 284. Schechter NR, Strom EA, Perkins GH, et al. Immediate breast reconstruction can impact postmastectomy irradiation. Am J Clin Oncol 2005;28:485-494. 284. Hortobagyi GN, Ames FC, Buzdar AU, et al. Management of stage III primary breast cancer with primary chemotherapy, surgery, and radiation therapy. Cancer 1988;62:2507-2516. 285. Hortobagyi GN, Buzdar AU, Strom EA, et al. Primary chemotherapy for early and advanced breast cancer. Cancer Lett 1995;90:103-109. 286. Jacquillat C, Weil M, Baillet F, et al. Results of neoadjuvant chemotherapy and radiation therapy in the breast-conserving treatment of 250 patients with all stages of infiltrative breast cancer. Cancer 1990;66:119-129. 287. Lippman ME, Sorace RA, Bagley CS, et al. Treatment of locally advanced breast cancer using primary induction chemotherapy with hormonal synchronization followed by radiation therapy with or without debulking surgery. NCI Monogr 1986;1:153-159. 288. Mauriac L, MacGrogan G, Avril A, et al. Neoadjuvant chemotherapy for operable breast carcinoma larger than 3 cm: a unicentre randomized trial with a 124-month median follow-up. Institut Bergonie Bordeaux Groupe Sein (IBBGS). Ann Oncol 1999;10:47-52. 289. Edeiken BS, Fornage BD, Bedi DG, et al. US-guided implantation of metallic markers for permanent localization of the tumor bed in patients with breast cancer who undergo preoperative chemotherapy. Radiology 1999;213:895-900. 290. Singletary SE, McNeese MD, Hortobagyi GN. Feasibility of breast-conservation surgery after induction chemotherapy for locally advanced breast carcinoma. Cancer 1992;69:2849-2852. 291. Chu JC, Solin LJ, Hwang CC, et al. A nondivergent three field matching technique for breast irradiation. Int J Radiat Oncol Biol Phys 1990;19:1037-1040. 292. Chen AM, Meric-Bernstam F, Hunt KK, et al. Breast conservation after neoadjuvant chemotherapy: the M D Anderson cancer center experience. J Clin Oncol 2004;22:2303-2312. 293. Chen AM, Meric-Bernstam F, Hunt KK, et al. Breast conservation after neoadjuvant chemotherapy. Cancer 2005;103:  689-695. 294. Knapper WH. Management of occult breast cancer presenting as an axillary metastasis. Semin Surg Oncol 1991;7:311-313. 295. Vezzoni P, Balestrazzi A, Bignami P, et al. Axillary lymph node metastases from occult carcinoma of the breast. Tumori 1979;65:87-91. 296. Whillis D, Brown PW, Rodger A. Adenocarcinoma from an unknown primary presenting in women with an axillary mass. Clin Oncol (R Coll Radiol) 1990;2:189-192. 297. Copeland EM, McBride CM. Axillary metastases from unknown primary sites. Ann Surg 1973;178:25-27. 298. Patel J, Nemoto T, Rosner D, et al. Axillary lymph node metastasis from an occult breast cancer. Cancer 1981;47:2923-2927. 299. Fortunato L, Sorrento JJ, Golub RA, et al. Occult breast cancer. A case report and review of the literature. N Y State J Med 1992;92:555-557. 300. Davis PL, Julian TB, Staiger M, et al. Magnetic resonance imaging detection and wire localization of an ‘occult’ breast cancer. Breast Cancer Res Treat 1994;32:327-330. 301. Orel SG, Weinstein SP, Schnall MD, et al. Breast MR imaging in patients with axillary node metastases and unknown primary malignancy. Radiology 1999;212:543-549. 302. Lee WJ, Chu JS, Chung MF, et al. The use of color Doppler in the diagnosis of occult breast cancer. J Clin Ultrasound 1995;23:192-194. 303. Halsted WS. The results of radical operations for the cure of carcinoma of the breast. Ann Surg 1907;46:1-19. 304. Baron PL, Moore MP, Kinne DW, et al. Occult breast cancer presenting with axillary metastases. Updated management. Arch Surg 1990;125:210-214.

305. Ellerbroek N, Holmes F, Singletary SE, et al. Treatment of patients with isolated axillary nodal metastases from an occult primary carcinoma consistent with breast origin. Cancer 1990;66:1461-1467. 306. Kemeny MM. Mastectomy: is it necessary for occult breast cancer? N Y State J Med 1992;92:516-517. 307. Merson M, Andreola S, Galimberti V, et al. Breast carcinoma presenting as axillary metastases without evidence of a primary tumor. Cancer 1992;70:504-508. 308. Rosen PP, Kimmel M. Occult breast carcinoma presenting with axillary lymph node metastases: a follow-up study of 48 patients. Hum Pathol 1990;21:518-523. 309. Bhatia SK, Saclarides TJ, Witt TR, et al. Hormone receptor studies in axillary metastases from occult breast cancers. Cancer 1987;59:117-1172. 310. Griffith CD, Christmas TJ, Elston CW, et al. Axillary lymph node metastases from clinically occult breast carcinoma. J R Coll Surg Edinb 1987;32:237-238. 311. Jackson B, Scott-Conner C, Moulder J. Axillary metastasis from occult breast carcinoma: diagnosis and management. Am Surg 1995;61:431-434. 312. van de Weijer GH, van Ooijen B, Hesp WL, et al. [Expectant management concerning the breast in 5 patients with ­occult breast carcinoma]. Ned Tijdschr Geneeskd 1995;139:  1648-1650. 313. van Ooijen B, Bontenbal M, Henzen-Logmans SC, et al. Axillary nodal metastases from an occult primary consistent with breast carcinoma. Br J Surg 1993;80:1299-1300. 314. Campana F, Fourquet A, Ashby MA, et al. Presentation of axillary lymphadenopathy without detectable breast primary (T0N1b breast cancer): experience at Institut Curie. Radiother Oncol 1989;15:321-325. 315. Read NE, Strom EA, McNeese MD. Carcinoma in axillary nodes in women with unknown primary site: results of breast­conserving therapy. Breast J 1996;2:403-409. 316. Vlastos G, Jean ME, Mirza AN, et al. Feasibility of breast preservation in the treatment of occult primary carcinoma presenting with axillary metastases. Ann Surg Oncol 2001;8:425-431. 317. Chang S, Parker SL, Pham T, et al. Inflammatory breast carcinoma incidence and survival. The Surveillance Epidemiology and End Results (SEER) Program of the National Cancer Institute, 1975-1992. Cancer 1998;82:2366-2372. 318. Rubens R, Armitage P, Winter P, et al. Prognosis in inoperable stage III carcinoma of the breast. Eur J Cancer 1977;13:805-811. 319. Ellis DL, Teitelbaum SL. Inflammatory carcinoma of the breast. A pathologic definition. Cancer 1974;33:1045-1047. 320. Thoms WJ, McNeese MD, Fletcher GH, et al. Multimodal treatment for inflammatory breast cancer. Int J Radiat Oncol Biol Phys 1989;17:739-745. 321. Le-Petross CH, Bidaut L, Yang WT. Evolving role of imaging modalities in inflammatory breast cancer. Semin Oncol 2008;35:  51-63. 322. Dawood S, Broglio K, Gong Y, et al. Prognostic significance of HER-2 status in women with inflammatory breast cancer. Cancer 2008;112:1905-1911. 323. Gonzalez-Angulo AM, Sneige N, Buzdar AU, et al. P53 expression as a prognostic marker in inflammatory breast cancer. Clin Cancer Res 2004;10:6215-6221. 324. Resetkova E, Gonzalez-Angulo AM, Sneige N, et al. Prognostic value of P53, MDM-2, and MUC-1 for patients with inflammatory breast carcinoma. Cancer 2004;101:913-917. 325. Barker JL, Nelson AJ, Montague ED. Inflammatory carcinoma of the breast. Radiology 1976;121:173-176. 326. Fields JN, Perez CA, Kuske RR, et al. Inflammatory carcinoma of the breast: treatment results on 107 patients. Int J Radiat Oncol Biol Phys 1989;17:249-255. 327. Cristofanilli M, Buzdar AU, Sneige N, et al. Paclitaxel in the multimodality treatment for inflammatory breast carcinoma. Cancer 2001;92:1775-1782. 328. Panades M, Olivotto IA, Speers CH, et al. Evolving treatment strategies for inflammatory breast cancer: a population-based survival analysis. J Clin Oncol 2005;23:1941-1950.

17  The Breast 329. Bristol IJ, Woodward WA, Strom EA, et al. Locoregional treatment outcomes after multimodality management of inflammatory breast cancer. Int J Radiat Oncol Biol Phys 2008;72:  474-484. 330. Arthur DW, Schmidt-Ullrich RK, Friedman RB, et al. Accelerated superfractionated radiotherapy for inflammatory breast carcinoma: complete response predicts outcome and allows for breast conservation. Int J Radiat Oncol Biol Phys 1999;44:  289-296. 331. Brun B, Otmezguine Y, Feuilhade F, et al. Treatment of inflammatory breast cancer with combination chemotherapy and mastectomy versus breast conservation. Cancer 1988;61:  1096-1103. 332. Low JA, Berman AW, Steinberg SM, et al. Long-term followup for locally advanced and inflammatory breast cancer patients treated with multimodality therapy. J Clin Oncol 2004;22:  4067-4074. 333. Perez CA, Fields JN, Fracasso PM, et al. Management of locally advanced carcinoma of the breast. II. Inflammatory carcinoma. Cancer 1994;74:466-476. 334. Swain SM, Lippman ME. Treatment of patients with inflammatory breast cancer. Important Adv Oncol 1989:129-150. 335. Chevallier B, Asselain B, Kunlin A, et al. Inflammatory breast cancer. Determination of prognostic factors by univariate and multivariate analysis. Cancer 1987;60:897-902. 336. Halverson KJ, Perez CA, Kuske RR, et al. Isolated local-regional recurrence of breast cancer following mastectomy: radiotherapeutic management. Int J Radiat Oncol Biol Phys 1990;19:  851-858. 337. Beck TM, Hart NE, Woodard DA, et al. Local or regionally recurrent carcinoma of the breast: results of therapy in 121 ­patients. J Clin Oncol 1983;1:400-405. 338. Janjan NA, McNeese MD, Buzdar AU, et al. Management of locoregional recurrent breast cancer. Cancer 1986;58:1552-1556.

407

339. Strom EA, Woodward WA, Katz A, et al. Clinical investigation: regional nodal failure patterns in breast cancer patients treated with mastectomy without radiotherapy. Int J Radiat Oncol Biol Phys 2005;63:1508-1513. 340. International Commission on Radiation Units and Measurements. ICRU Report 50: Prescribing, Recording, and Reporting ­Photon Beam Therapy. Bethesda: ICRU, 1993. 341. Motwani SB, Strom EA, Schechter NR, et al. The impact of immediate breast reconstruction on the technical delivery of postmastectomy radiotherapy. Int J Radiat Oncol Biol Phys 2006;66:76-82. 342. Vrieling C, Collette L, Fourquet A, et al. The influence of patient, tumor and treatment factors on the cosmetic results after breast-conserving therapy in the EORTC ‘boost vs. no boost’ trial. EORTC Radiotherapy and Breast Cancer Cooperative Groups. Radiother Oncol 2000;55:219-232. 343. Borghero YO, Salehpour M, McNeese MD, et al. Multileaf fieldin-field forward-planned intensity-modulated dose compensation for whole-breast irradiation is associated with ­ reduced ­contralateral breast dose: a phantom model comparison. ­Radiother Oncol 2007;82:324-328. 344. Donovan E, Bleakley N, Denholm E, et al. Randomised trial of standard 2D radiotherapy (RT) versus intensity modulated radiotherapy (IMRT) in patients prescribed breast radiotherapy. Radiother Oncol 2007;82:254-264. 345. Evans PM, Donovan EM, Partridge M, et al. The delivery of intensity modulated radiotherapy to the breast using multiple static fields. Radiother Oncol 2000;57:79-89. 346. Pignol JP, Olivotto I, Rakovitch E, et al. A multicenter randomized trial of breast intensity-modulated radiation therapy to reduce acute radiation dermatitis. J Clin Oncol 2008;26:  2085-2092.

The Blood Vessels and Heart 18 ROGER W. BYHARDT BLOOD VESSELS Anatomy Small Blood Vessels Large Blood Vessels HEART Anatomy Methods of Assessing Radiation-Induced Cardiac Effects Radiation Factors Related to Cardiac Injury Effects of Irradiation on the Pericardium Effects of Irradiation on the Myocardium Effects of Irradiation on the Conduction System and Heart Valves

Effects of Irradiation on the Coronary Arteries Irradiation for Esophageal Cancer Irradiation for Breast Cancer Irradiation for Hodgkin Disease Irradiation for Seminoma Summary of Studies Concerning the Cardiac Effects of Irradiation Coronary Irradiation to Prevent Restenosis Prevention of Radiation-Induced Heart Damage Management Strategies for Patients at Risk for Radiation-Induced Heart Damage

Blood Vessels Understanding the effects of radiation on blood vessels is key to understanding many of its normal tissue effects. Many of the late effects and, to a lesser extent, the acute effects of radiation on different tissues and organs are mediated by a continuum of vascular effects. Throughout the body, cellular viability and organ function depend on blood vessel integrity. Probably the most critical component of the blood vessel structure, especially that of capillaries and small arterioles, is the monolayer of endothelium lining the interior. Because endothelial cells proliferate relatively slowly, cell loss and the often-abnormal proliferative response of surviving cells begin months after radiation therapy has been completed. In smaller-caliber vessels, the aberrant endothelial cell proliferative response or the development of thromboses in the affected area can partially or completely obstruct blood flow, leading to downstream tissue necrosis. In larger vessels with smooth muscle walls, radiation damage to the muscle cells leading to fibrosis and diminished blood flow occurs years later. Endothelial cell proliferation is less likely to restrict blood flow in these larger-caliber vessels but can still provide sites for the formation of thrombi. Although the acute effect of radiation on blood vessels is an important trigger, it is the late effect of radiation on small vessels leading to impairment of circulation that has the greatest clinical significance. An impaired blood supply leads to local tissue changes in nutrition, electrolyte concentrations, and oxygenation. These changes impair the ability of tissues to respond to other types of injury. The early and late vascular effects of radiation depend on vessel size and location, the presence of vascular disease, the dose of radiation, the time over which radiation is delivered, the volume irradiated, and the stresses to which the irradiated volume is subjected. With conventional daily doses of 1.8 to 2.0 Gy, the risk of arterial damage rises after

a total dose of 50 Gy, or about 40 Gy for damage to capillaries. The radiation tolerance of veins is somewhat higher. Although some of the cardiac effects of radiation are similar to the vascular effects, especially the effects on the coronary arteries, the results are more complex and multifactorial. Several decades ago, many thought the heart was relatively resistant to radiation injury; however, the emergence of more effective imaging and functional assessments has provided credence to the old adage that “the more one looks, the more one finds.”

Anatomy Walls of blood vessels consist of three concentric layers (­tunicae). The inner layer (intima) consists of endothelium; the outer layer (adventitia) is made up of connective tissue, bracketing a middle layer (media) consisting of smooth muscle.

Small Blood Vessels Acute Vascular Changes Erythema. The first and most noticeable gross vascular change after irradiation is cutaneous or mucosal erythema. The intensity of the erythema increases in direct proportion to the radiation dose and the area irradiated. Erythema of the skin appears in waves that occur at predictable times during the course of radiation therapy—on the first day, during the second or third week, and at the end of the first month. The mechanism of radiation-induced erythema is not known, but the first period of erythema is thought to be a vascular response to local extracapillary cell injury, whereas the later periods of erythema are presumed to be caused by direct capillary damage.1 Radiation-induced erythema is not a result of nerve injury. It has been suggested that a ­histamine-like substance is produced by the radiation injury and that this substance diffuses through tissues to produce

411

412

PART 5  Thorax

erythema within and slightly beyond the margins of the irradiated volume. However, this mechanism is questionable because radiation-induced erythema is not decreased by administration of antihistamines. Moreover, during the first period of erythema, the vessels maintain their ability to constrict in response to a scratch test, and histamine does not produce wheal formation. Skin temperature rises several degrees with the onset of radiation-induced erythema, suggesting that vasodilation, not merely vasocongestion, occurs. Local administration of epinephrine or acetylcholine reduces but does not eliminate vasoconstrictive and vasocongestive responses. Direct observation of radiation-induced changes in arterioles and capillaries in bats’ wings, in the webs of frogs’ feet, in human nail folds, and through experimentally placed transparent “windows” confirms that acute radiation-induced damage to small blood vessels is nonspecific. During the acute phase of erythema, damage of the vessel endothelium is not striking. With the radiation doses used clinically, however, some endothelial cells are killed, and thrombi form and narrow or obliterate the vessel ­lumen. Increased Vessel Permeability and Vessel Dilation. The functions of small blood vessels are to deliver oxygen and nutrients to cells and to carry waste products away from cells. These functions depend on blood flow rates and on the transmission and transport of nutrients and waste products through the capillary wall. The relative contributions of the two layers of the capillary wall—the endothelium and the basement membrane—to capillary permeability remain uncertain. A few hours to a few weeks after treatment with moderate doses of radiation, the permeability of the capillary wall increases, as manifested by associated edema. As expected, the rapidly proliferating endothelial cells of newly developing capillaries are more sensitive to radiation than are endothelial cells in older capillaries.2,3 This fact explains some of the inhibitory effects of irradiation on wound ­healing. One of the most consistent early effects seen in the capillaries and prearterioles after irradiation is vessel dilation, which can be accompanied by endothelial cell swelling, degeneration, and necrosis and the presence of a cellular inflammatory infiltrate. Increased vascular permeability with resulting tissue edema is a common early manifestation (Fig. 18-1). The pathogenesis of radiation-induced vascular dilation and increased capillary permeability is not clear. Endothelial cell damage can be minimal or absent even in the case of severe edema. It is possible that irradiation interferes, at least temporarily, with vasomotor regulatory mechanisms and results in hemodynamic alterations conducive to tissue edema. Increased capillary permeability manifested as local edema can be seen after doses of 5 Gy or more. Various substances, including labeled plasma, erythrocytes, Evans blue dye, and colloidal gold, permeate the capillary walls more rapidly after irradiation. The peak change develops 2 weeks after a single exposure. The contribution of thrombi and vasodilation to the increased permeability is unknown. The capillaries exhibit increased fragility. Bleeding occurs more readily even though the clotting mechanism is normal. The combination of thrombus formation, vasodilation, and increased capillary fragility and bleeding

FIGURE 18-1.  Arteries from a 34-year-old woman irradiated 5 months earlier with 65 Gy for carcinoma of the cervix. Fatty deposits in the intima have caused extensive narrowing of the lumina. Notice the edema surrounding the blood vessels.

resembles an inflammatory response.4 As is true for an inflammatory response, these effects subside as the chronic vascular changes manifest.

Chronic Vascular Changes After the acute reaction to radiation subsides and before major narrowing of the vessel lumen develops, diffusion of material through capillary walls decreases. The vessel wall may not appear to be seriously damaged; however, the ultrafiltration ability of the endothelial lining is decreased.4 The basement membrane of the capillary wall is thickened, and this thickening is presumed to contribute to decreased capillary permeability. The connective tissue barrier between the capillaries and the dependent tissues becomes thicker as a result of extracapillary fibrosis. The cause of this extracapillary fibrosis is unknown, although Rubin and Casarett5 related it to a connective tissue reaction after “leakage” of plasma constituents from capillaries. Still later, the number of small vessels is decreased through the process of vessel occlusion (Fig. 18-2). Arteriolar capillary intimal hyalinosis proceeds as a discontinuous process. Tumoricidal doses of radiation inhibit capillary sprouting and vascular remodeling,6 which probably contributes to delays in wound healing after irradiation. However, vascular endothelium seems to recover rapidly, and it is possible that migrating unirradiated endothelial cells account for this apparently rapid recovery. The most common chronic radiation-induced changes seen in vessels of small and medium caliber are swelling and vacuolation of the endothelial cells of the tunica intima vasorum. Later, proliferation of endothelial cells and lipid deposits may occur. These lesions resemble atheromatous plaques but differ in their location because atheromatous plaques rarely occur spontaneously in small arteries. These lesions are not specific for radiation damage; any physical or chemical damage to the vessel, especially the tunica adventitia vasorum, results in similar lesions. This is particularly true in experimental situations in which animals are rendered hyperlipemic by dietary means. In humans, no evidence has been found of any relationship between serum cholesterol levels and development of the foamy

18  The Blood Vessels and Heart

413

reactions, sclerotic changes in aged patients, hypertension, associated trauma such as surgery, and fractionation schedule. The recognition of these factors and the exploitation of the advantages of increased fractionation are the basis for individualization of radiation therapy techniques.

Large Blood Vessels In large blood vessels, high doses of radiation produce few recognizable acute changes. Veins are little affected, and the changes affecting large, elastic arteries usually occur as late manifestations.

Mechanisms of Radiation-Induced Injury

FIGURE 18-2.  Small blood vessels 5 months after irradiation with 65 Gy in 7 weeks. The vessels have strikingly thickened walls and narrowed lumina.

l­esions in the intima after irradiation. These lesions can be seen a few days after radiation therapy and can persist for many years. Concomitant with these changes in the intima, vacuolation and degeneration of the smooth muscle cells of the tunica media vasorum may occur. Repair with fibrosis of the tunica media is seen in chronic lesions. Narrowing of the lumen occurs mainly as a result of concentric fibrosis and loss of vascular elasticity, but thrombosis and intimal proliferation also play important roles. In the superficial part of the cutis, extensive dilation of capillaries, recognized as telangiectasia, is usually seen after radiation therapy. The telangiectatic vessels arise from existing capillaries and are presumed to be a result of greater blood flow through the few remaining damaged small blood vessels. The sequelae of small blood vessel obliteration vary with the organ in question. The vasculoconnective tissue of the skin can tolerate extensive damage before necrosis occurs. However, rather minor defects in the vasculature of the brain, myocardium, kidney, or lungs may result in seriously limited function. In addition to the changes in small blood vessels that directly result from irradiation, changes caused by tumor shrinkage also can be seen. Quantitative studies show improved vascular filling after irradiation. After irradiation, the appearance of vessels in large tumors reverts to the appearance of vessels in small tumors.7,8 These ­ vascular changes are fundamental to the process of radiation-­ induced reoxygenation of the tumor. The popularity of radiation doses of about 60 Gy in 6 weeks is a recognition of the tolerance of most normal ­vasculoconnective tissues. The incidence of necrosis of normal tissues increases rapidly with increasing dose when doses higher than 60 Gy administered in 6 weeks (or the equivalent) are delivered to large volumes. A variety of factors modify the response of small vessels to radiation, including tissue oxygen concentration, associated ­ inflammatory

The mechanisms of chronic radiation injury to large blood vessels are not known. Although a direct effect of radiation on the cellular elements of the tunica media cannot be excluded, the changes seen are similar to those observed after damage of the adventitial vessels nourishing most of the vascular wall-changes that also can be produced ­experimentally by burning or freezing the adventitia. ­Extensive dissection around the vessels with destruction of their blood supply is likely to be a predisposing factor but is by no means a prerequisite for such radiation-induced changes. The changes themselves basically consist of degeneration of muscle cells of the tunica media. The muscle cell degeneration can be a patchy, localized process or can manifest as a wide zone of cystic medionecrosis. Weakening of the vessel wall may lead to rupture, but more often, fibrosis of the tunica media is observed. The intima overlying the areas of medial degeneration usually develops changes indistinguishable from those of ordinary atherosclerosis.8 Similar changes can be produced experimentally by damaging the adventitia by chemical or physical means. The pathogenesis of these experimental lesions seems to be interference with nutrition of the arterial wall through damage of the adventitial vasa vasorum. It seems likely that a similar mechanism operates in radiation-induced damage to large arteries; damage is mediated through narrowing and occlusion of the vasa vasorum caused by irradiation. In clinical situations, the presence of tumor cells and the tissue reactions they elicit also may contribute to the decreased blood supply of the vessels. The decreased strength of the vessel walls after irradiation is particularly important when radical neck dissection is performed after high radiation doses have been given to the carotid arteries. In this situation, rupture of the carotid arteries may occur.9 Arteriosclerosis and Atherosclerosis. After irradiation, changes within the vessel lumen are more obvious than changes in the vessel walls. Single, large doses of 15 to 20 Gy produce arteriosclerosis and atherosclerosis within 30 to 40 weeks. This sclerosis is confined to the ­irradiated zone.8 Fractionated doses of 30 to 50 Gy ­produce less ­severe changes than large single doses. Hypertensive Vascular Damage. Asscher10 found that radiation sensitizes blood vessels to hypertension. Hypertension from any cause produces early profound hypertensive vascular damage in irradiated blood vessels, whereas unirradiated blood vessels subjected to the same hypertension may appear relatively normal for months.10 When ­hypertension develops years after high-dose irradiation, the patient can develop a localized vascular insufficiency leading to necrosis of the irradiated volume.

414

PART 5  Thorax

Radiation-Induced Carotid Artery Disease. Silverberg and colleagues11 compared a variety of variables in 9 patients with atherosclerotic carotid artery disease associated with neck irradiation and 40 unirradiated control ­patients. The 9 irradiated patients with occlusive carotid artery disease were younger than the controls, were less ­likely to have peripheral vascular disease and coronary artery disease, and were less likely to have hyperlipemia or hypercholesterolemia.11 These findings support the recognition of radiation-induced carotid artery disease as a clinical entity. Reconstructive carotid surgery in patients with carotid artery damage resulting from radiation therapy should be approached as if radiation were not a factor, even though there may be some periarterial fibrosis and increased difficulty in separating the plaques from the tunica media.11 Arrest of Growth of Vessel Diameter. During childhood, the aorta and other large vessels grow in proportion to increasing body surface area, not in relation to chronologic age.12 In children, doses of 25 to 28 Gy given to large vessels over 2 to 3 weeks arrest growth in vessel diameter.13 The inhibitory effect of irradiation on the growth of vessel diameter is greatest when irradiation is delivered during infancy and decreases with increasing age at irradiation.

Heart Despite dosimetric efforts to exclude the heart from the radiation field, it often receives radiation during the course of treatment for several malignancies. For example, treatment of the chest wall in patients with left-sided breast cancer often involves some exposure of the heart, including the coronary vessels. Larger heart volumes are included during treatment of the mediastinum in patients with lung cancer, esophageal cancer, lymphoma, or seminoma. The variety of radiation doses and dose distributions used to treat each tumor type have provided useful clinical data regarding dose-effect relationships for cardiac irradiation, including fraction-size effects. The most useful data on the late effects of cardiac irradiation have come from studies of patients with breast cancer, Hodgkin disease, and seminoma, because many of these patients survive for long periods after the treatment, long enough for late effects to become manifest. Conversely, the relatively short survival times after definitive irradiation for inoperable lung or esophageal cancer have resulted in much less information on late cardiac effects. The radiation treatment plans for these types of cancer usually result in a gradient of dose through the heart, making it a challenge to relate subclinical and clinical cardiac effects to a specific dose of radiation. A clear understanding of the relationship of radiation-related heart injury to partial-volume cardiac ­doses has been limited by a scarcity of dose-volume dosimetric information until the middle to late 1990s. The widespread use of computed tomography (CT)–based treatment planning systems has provided more detailed information on cardiac dose-volume relationships. By using animal models and multifraction techniques similar to those used in clinical practice, investigators have shown that late cardiac injury from radiation is multifactorial. Although damage to the cardiac microvasculature is a significant risk,14,15 all of the components of the heart are affected by radiation, including the pericardium, myocardium, endocardium, valves, and coronary vessels. Clinically,

more rigorous dosimetric information coupled with improvements in cardiac functional assessment has provided a clearer picture of radiation dose-effect relationships. Pericarditis is often reported as being the most common radiation-induced heart injury, but subtle functional changes may be present that are detectable only by specialized testing. Acute pericarditis usually appears during the first year after treatment and has a benign, self-limited course; it tends to manifest as acute chest pain, dyspnea, and low-grade fever, with or without progression to fibrosis and tamponade. The heart does not function in isolation. Inclusion of heart and lung in the irradiated volume results in a complex interplay of cardiac and pulmonary effects. In addition to the total radiation dose and fraction size and the volume of heart incidentally irradiated, the extent of radiation­induced cardiotoxicity may be affected by adjacent tumor affecting the lymphatic drainage from the heart. Heart structure and function are also intimately linked to the integrity of the coronary arteries. Numerous clinical studies have shown a relationship between cardiac irradiation, accelerated coronary vessel atherosclerosis, and increased risk of ischemic coronary artery disease, which increase the risk of myocardial infarction.8,16-26 These effects are not inevitable; they depend on dose and volume and on the individual’s genetic and lifestyle-associated risks for coronary artery disease. The available clinical findings should be interpreted with care, because some studies did not ­include a control group with which to distinguish the risk of radiation-induced coronary injury from the expected incidence of coronary disease in appropriately age- and riskmatched populations.

Anatomy The wall of the heart is composed of three layers: the epicardium (visceral layer of the pericardium), the ­myocardium, and the endocardium. The epicardium and pericardium are thin, relatively avascular layers that function to minimize friction and provide damping of cardiac contractions. These layers are particularly susceptible to radiation damage, and the response is not unlike that produced by an infection. The myocardium varies in thickness according to the site within the heart wall. The endocardium is continuous with the intima of the blood vessels, and the two layers seem to respond to irradiation in similar ways. Radiation-induced injuries of the heart trigger a doserelated thickening of one or more layers of the heart wall. The heart valves and conduction system also may show dose-related radiation effects. The microvasculature and the larger blood vessels of the heart (e.g., coronary arteries) may undergo radiation dose–related changes. The effects of radiation on the microvasculature of the heart often form the basis for radiation injury of the cardiac components. Irradiation of the coronary arteries may lead to the most serious side effects.

Methods of Assessing Radiation-Induced Cardiac Effects A spectrum of tests are now available to evaluate potential radiation-induced heart damage, including cardiac ­magnetic resonance imaging (MRI), fast-scan cardiac CT, ventriculography, gated single photon emission CT, and multigated acquisition (MUGA) scanning. Many tests are capable of

18  The Blood Vessels and Heart

detecting subtle, subclinical changes in cardiac function. Electrocardiography can reveal conduction abnormalities. Echocardiography can reveal changes in the thickness of the myocardium. Radionuclide angiocardiography and single photon emission CT can reveal changes in myocardial perfusion and in cardiac output through measurements of the left ventricular ejection fraction (LVEF).27-29 Exercise testing on a treadmill or bicycle ergometer with thallium 201 scintigraphy can demonstrate subclinical changes in cardiac output. Various radiographic imaging techniques, including 64-slice CT and MRI, can more precisely characterize pericardial abnormalities. Positron emission tomography can also demonstrate changes in the myocardium corresponding to areas of high-dose irradiation.30

Radiation Factors Related to Cardiac Injury Factors that influence the risk of cardiac injury resulting from radiation therapy include total dose, exposed volume, and fraction size.

Total Dose and Exposed Volume A probit analysis of 318 patients who underwent radiation therapy for Hodgkin disease yielded a sigmoid doseresponse curve for radiation-induced heart disease (Fig. 18-3).31 The curve shows a steep rise in radiation-induced cardiac injury above a total dose of 40 Gy. These findings have since been confirmed by other clinical and animal studies.14,32-36 The dose-response curve also depends on the volume of heart irradiated. When large cardiac volumes are irradiated, as takes place in the treatment of Hodgkin disease, doses lower than 40 Gy given over a 4-week period may induce cardiac injury. However, when small cardiac volumes are treated, as is the case in the postoperative treatment of breast cancer, doses higher than 40 Gy in 4 weeks may be well tolerated.37 The tolerance dose (TD), which is the total dose causing a specific level of risk of radiation-related cardiac injury, varies with the volume of heart irradiated. Small heart volumes may tolerate 60 Gy given at 1.8 to

Percent carditis (RIHD)

100 80 60 40 20 0 0

20

40

60

80

100

Dose (Gy)

FIGURE 18-3.  Probit analysis of the dose-response relationship for radiation-induced heart disease. The curve is derived by replotting by probit of the analysis of 318 determinant cases of Hodgkin disease. LD5 is the dose causing heart disease in 5% of patients, and for 43.3 Gy, the 95% confidence interval is 40.3 to 46.4 Gy. Squares, data points; blue line, fitted probit curve; red and green lines, 95% confidence limits. (Modified from Stewart JR. Normal tissue tolerance to irradiation of the cardiovascular system. Front Radiat Ther Oncol 1989;23:302-309.)

415

2.0 Gy/day for 5 days/week, whereas large volumes may tolerate only 40 Gy at 1.8 to 2.0 Gy/day. In 1991, based on clinical observations available at that time, volume-based estimates were published for the TD5/5 (i.e., dose associated with a 5% risk of complications within 5 years) and the TD50/5 (i.e., dose associated with a 50% risk of complication within 5 years) for the heart.38 When pericarditis was used as the endpoint, the TD5/5 for conventionally fractioned radiation was estimated at 60 Gy for one third of the heart, 45 Gy for two thirds of the heart, and 40 Gy for the entire heart. The corresponding TD50/5 doses for the same volumes were 70 Gy, 55 Gy, and 50 Gy. Values such as these have been used as cardiac dose limits in several clinical studies of radiation therapy for anatomic sites near the heart. Plots of normal tissue complication probabilities for late cardiac mortality versus dose have yielded sigmoid curves of similar shape for 33%, 66%, and 100% of heart volume, suggesting that the volume dependence for this end point is small.39 For breast cancer or other types of cancer associated with long-term survival that may involve incidental radiation to the heart, the TD5/5 values may be set much lower (e.g., 25 Gy).

Fraction Size Although the TD5/5 values discussed in the previous section suggest that the heart has a high dose tolerance, the threshold for mild changes may be 20 Gy when exposed heart volumes exceed 50%. Theoretically, an alpha/beta ratio (defined in Chapter 1) of about 1 Gy suggests that fractionation of the radiation dose should spare the heart. However, use of concurrent chemotherapy, particularly with cardiotoxic agents such as doxorubicin, can compound radiation damage. Fraction size is an important determinant of the risk of radiation-induced cardiac injury. Larger doses per fraction tend to cause more cardiac damage.37 Some of the highest rates of pericarditis have been reported among patients with Hodgkin disease treated with anteriorly weighted fields in which the anterior cardiac structures received a larger dose per fraction than the midline tumor.32 Use of equally weighted fields, lower total doses, daily treatments of each field, and when possible, cardiac shielding has reduced the risk of cardiac damage in patients with Hodgkin disease.31 Analyses of the effect of fraction size on radiation­induced heart damage are complicated because of the recent trends toward lower total doses and lower doses per fraction. Nevertheless, reviews of the available clinical data and observations of cardiac injury after fractionated irradiation of canine hearts confirm a fraction-size effect.3,40 A study of the effect of fractionated irradiation on the number of adrenergic receptors, myocardial norepinephrine concentration, and cardiac function in rats also showed that cardiac damage depends on radiation fraction size.33

Effects of Irradiation on the Pericardium Acute Pericardial Changes Radiation-induced acute pericarditis may appear a few weeks or several years after irradiation. If pericarditis is clinically overt, the patient may present with fever, tachycardia, substernal pain, and pericardial friction rub. Pericardial effusion is common, and cardiac tamponade may

416

PART 5  Thorax

develop as the effusion increases. The cardiac silhouette is widened. The echocardiogram may show inversion and flattening of the T waves, elevation of the ST segment, and decrease of the QRS segment. The course of acute pericarditis is variable. The condition is self-limited in about one half of the patients who present with clinical symptoms; the remaining patients experience recurrence or progression to chronic constrictive pericarditis. The dose-time relationship for acute pericarditis has been clarified by Stewart and Fajardo.41 Most patients with symptoms of acute pericarditis have received at least 40 Gy in 4 weeks to a major portion of the heart, but the observation that acute pericarditis is uncommon even after 50 Gy given in 5 to 6 weeks has led to a search for possible contributing factors. Cardiac venous and lymphatic pathways drain to the mediastinal lymphatics, and tumor-­related changes in these pathways may contribute to the formation of pericardial effusion. This may partially ­account for the correlation between larger mediastinal tumor size and higher risk of pericardial effusion after radiation therapy to the chest. The incidence of pericardial effusion increases rapidly as doses increase above 50 Gy in 5 weeks.34,42 Some cases of postirradiation pericardial effusion are linked to myxedema, which may develop after thyroid ­irradiation. Prolonged high iodide levels after lymphangiography provoke thyroid hyperplasia and increase the radiosensitivity of the thyroid gland.43 Patients with pericardial effusion should be evaluated for thyroid function before the effusion is attributed solely to the effects of heart irradiation. Nevertheless, studies of humans and animal models confirm that most cardiac effects of radiation are a result of direct cellular damage to the pericardium, the epicardium, and the blood vessels of these layers rather than indirect mechanisms.

Chronic Constrictive Pericarditis Chronic constrictive pericarditis usually manifests as chronic constrictive changes associated with myocardial and endocardial fibrosis or with varying degrees of effusion. The latent period from the beginning of radiation therapy to the onset of chronic pericarditis varies from 6 months to several years. Chronic pericarditis may or may not be preceded by acute pericarditis. Symptoms and signs of chronic pericarditis can include dyspnea, chest pain, venous distention, and pleural ­effusion. Paradoxical pulse and fever may be present. The electrocardiogram may show decreased QRS voltage, flat or inverted T waves, and elevation of the ST segment. The cardiac silhouette is enlarged. The morphologic changes are not specific for radiation damage. The pericardium is thickened as a result of deposition of collagen. The myocardium and endocardium are often fused or adherent to each other, with a fibrinous exudate. The pericardium may be adherent to the heart and pleura. The blood vessels in the pericardium show characteristic subendothelial connective tissue proliferation. The underlying myocardium usually appears normal on gross examination but may not be microscopically or functionally normal. Although the cause of chronic constrictive pericarditis is the same as that of acute pericarditis, fewer than half of patients with chronic constrictive pericarditis have a history of previous acute pericarditis. Overall, about 5% of ­patients who receive more than 40 Gy to at least 50%

of the heart will develop chronic pericarditis.44 The treatment for chronic constrictive pericarditis depends on the severity of the condition and consists of antipyretics in the case of fever, pericardiocentesis in the case of tamponade, and ­pericardiectomy in the case of severe constrictive ­symptoms.

Effects of Irradiation on the Myocardium Three components of the myocardium are important in the response of the myocardium to irradiation: the vascular component, the connective tissue component, and the muscle component. Gillette and colleagues2,3 found that all fractionation schedules tested with a variety of total radiation doses led to diminishment of the cardiac vasculature. The higher the fraction size and the higher the total dose, the greater the decrease in the vasculature. The amount of connective tissue in the heart, which is distributed between the muscle bundles throughout the heart, increased with increasing dose up to a maximum of about 44 Gy at 4 Gy/fraction, after which further increases in dose produced smaller increases in connective tissue at 3 months and at 6 months after irradiation. Subsequent studies of fractionated irradiation of the beagle heart confirmed early and late myocardial damage, which manifested as increased heart rates, increased heart wall thickness, conduction abnormalities (early effects), and thinning of the cardiac wall and fibrosis with functional abnormalities (late effects).40 Cardiac and pulmonary functions are highly interdependent. When the heart and the lung are included in the radiation field, subclinical levels of radiation-induced myocardial injury may clinically manifest as a consequence of radiation-induced changes in the pulmonary vasculature. Studies of rat and canine models have shown radiationinduced reduction of pulmonary capillary volume, which causes an increase in pulmonary artery pressure.14,45 The radiation-damaged heart may not be able to overcome this increased resistance. Some of the volume effects discussed previously under “Total Dose and Exposed Volume” may be partially related to the amount of lung irradiated at the same time as the heart.46 Studies of rats in which autoradiography with tritiumlabeled thymidine was used have shown that radiation also damages the myocardial capillary network; this damage is followed by compensatory proliferation after a latency period of 6 to 8 months.15,45 These changes precede myocardial degeneration, but the findings cannot be linked to hypoxia or radiation-induced mitotic death of endothelial cells alone. Early reductions in α- and β-adrenergic receptors followed by upregulation of those receptors, which are ­located in endothelial cells, have been observed.15,47 In the early period after radiation therapy, catecholamines may be released from sympathetic nerve endings, leading to depletion of catecholamines and downregulation of the receptors. The subsequent upregulation of α- and β-­adrenergic receptors seen in later stages may be caused by a prolonged decrease in sympathetic stimulation. The release of catecholamines may be a direct effect of radiation on the sympathetic nerve endings or may be mediated by radiation damage to the capillary network and resultant local hypoxia. The relationships among the vasculature, connective tissue, and muscle of the myocardium at various times after irradiation are quite complex, and additional sequential

18  The Blood Vessels and Heart

studies are necessary to better understand the mechanism of radiation-induced myocardial injury. The lack of mitotic activity in the cardiac muscle makes the myocardium one of the more radioresistant tissues and suggests that most of the changes seen in heavily ­irradiated myocardium result from radiation-induced vascular changes. Direct radiation injury of the muscle fiber is presumed to be a minor factor, but the degree of fibrosis between the muscle bundles does change with dose and with fractionation.3,40 This connective tissue that forms in response to radiation is diffuse and does not resemble the fibrosis seen after an infarct. Radiation-induced fibrosis may not necessarily result from small vessel occlusion and may not parallel the severity of vascular occlusion.48 In rhesus monkeys, total body irradiation led to high plasma levels of atrial natriuretic peptide that were associated with a change in cardiac dimensions and predicted cardiac damage.49 Serial measurements of atrial natriuretic peptide levels after cardiac radiation exposure in humans may serve as a surrogate marker for cardiac toxicity. Procollagen I and III levels also may become elevated about 6 months after heart irradiation, corresponding to the appearance of fibrosis in myocardial connective tissue.50 The existence of subclinical abnormalities of heart function suggests that radiation-induced cardiac injury is not a threshold phenomenon. Studies published in the early 1980s indicated that LVEF in clinically asymptomatic ­ patients who had received more than 35 Gy to the heart during mediastinal irradiation for Hodgkin disease was lower at 5 to 15 years after treatment than that in patients who had not undergone mediastinal irradiation.35,36 ­Abnormalities on radionuclide angiography have been found years after treatment of Hodgkin disease with 40 Gy in 4 weeks with a mantle field in one fourth of patients who were 25 to 35 years old at the time of treatment.16 Electrocardiography studies of patients with breast cancer treated with left-sided internal mammary node irradiation have shown subclinical T-wave changes, and bicycle ergometry stress testing with technitium 99m sestamibi scintigraphy has detected radiation-induced microvascular myocardial ­damage.51-53 Radiation given to patients previously treated with doxorubicin can worsen the drug’s cardiac side effects. The cardiac effects of doxorubicin, which are expressed in some patients but not others, appear about 6 months after treatment and consist of a diffuse cardiomyopathy that can lead to congestive heart failure. The cardiac vessels are usually normal. The myocardium shows profound degeneration of the muscle cells, combined with interstitial edema and fibrosis. These effects can be worsened by radiation therapy. Care should be taken to limit the cardiac dose and the volume irradiated after doxorubicin has been given.

Effects of Irradiation on the Conduction System and Heart Valves Conduction abnormalities, such as atrioventricular block, occur in up to 5% of patients after mediastinal irradiation that includes the heart; these abnormalities are thought to be related to late myocardial fibrosis.44,54 Valvular defects, often subclinical, may be found in 15% to 20% of patients who undergo mediastinal irradiation that includes the heart. Autopsy studies of hearts exhibiting radiation injury (e.g., after mediastinal irradiation for Hodgkin

417

­ isease) have shown that any of the four heart valves may d be affected, but clinical evidence of valvular dysfunction is present in only about one half of the cases, with the damage identified only at autopsy.55 Successful replacement of radiation-damaged valves is possible.56 As is the case for radiation-induced injury to the other components of the heart, valvular radiation damage is related to dose and is rare when doses are less than 30 Gy.

Effects of Irradiation on the Coronary Arteries The effects of irradiation on the cardiac blood vessels are the same as the effects of irradiation on blood vessels ­elsewhere in the body (see “Blood Vessels”). However, ­radiation-induced injury of cardiac vessels has more serious consequences than radiation-induced vascular injury in most other sites and is therefore worthy of special ­attention. No specific vessels are affected more often than ­others. The consequence of radiation-induced vessel occlusion is the same as vessel occlusion from other causes. The understanding of the relationship between coronary artery disease and prior cardiac irradiation continues to evolve. Although the issue remains somewhat controversial, findings from several studies of large numbers of patients that often included appropriately matched control subjects and cardiac function testing suggest that the risk of coronary vessel damage is reduced when heart-sparing radiation therapy techniques are used.23,36,51-60 Earlier reviews summarized case reports of precocious coronary artery disease in long-term survivors who had undergone mediastinal irradiation for Hodgkin disease and other malignancies.17,18 The number of affected patients in those studies was small, and many long-term survivors were without apparent coronary artery disease. However, the observation of reproducible postirradiation coronary artery disease in animals17,45 suggested that the risk of coronary artery disease in humans may increase with longer-term follow-up.61 The early case reports showed that the mechanism of coronary artery disease was accelerated coronary atherosclerosis and fibrosis. Some cases involved older patients with predisposing metabolic factors or familial risk for coronary disease. However, other cases involved young patients with no predisposing risk factors. Angiographic and pathologic findings suggested a localized process limited to the portion of the heart included in the radiation field. Concern was raised that occult early effects would predispose patients to earlier or more severe coronary artery disease. Reviews of long-term follow-up information on patients who underwent irradiation of the mediastinum for treatment of Hodgkin disease, breast cancer, or seminoma have described the risk of coronary artery disease and death from acute myocardial infarction. Several factors make interpretation and comparison of these reports difficult, however. In some studies, the risk of coronary artery disease was compared with the risk in a control population that had not received mediastinal irradiation. Some studies reported the total number of deaths from acute myocardial infarction in the treated population, and others reported the percentage of total deaths attributable to acute myocardial infarction. The risk of subclinical or nonfatal coronary artery disease could not always be discerned. Some reviews were based on information gleaned from death certificates, which may be misleading. Additional details of studies of

418

PART 5  Thorax

radiation-induced coronary artery disease are reviewed in the following sections.

Irradiation for Esophageal Cancer Improvements in combined-modality treatment for distal esophageal cancer have increased the number of long-term survivors. These patients provide a model of the effects of high-dose radiation to large heart volumes and of the potential additive effects of concurrent chemotherapy on acute and chronic cardiotoxicity. One study of 78 patients with distal esophageal cancer in complete remission after chemoradiation showed that at a median follow-up time of 53 months, 16 (20%) had grade 2 to 4 pericarditis and 2 had died of myocardial infarction.62 Another study involving MUGA scanning before and after platinum-based chemoradiation (50.4 Gy) for 20 patients with distal esophageal cancer showed a statistically significant drop in LVEF from 59% to 54%, although no clinical evidence of functional change was observed.63 Dose-volume histograms showed the median volume of heart that had received more than 40 Gy to be 53%. In a similar study,64 findings from MUGA scans and dose-volume histograms revealed a significant reduction in LVEF from 63% to 58%. Moreover, use of three-dimensional conformal therapy with shielding was associated with a median cardiac dose of 27 Gy, compared with 35 Gy if conventional planning without shielding blocks had been used. Use of a three-field treatment technique further reduced the median cardiac dose to 22 Gy and the volume of the heart receiving 70% of the total dose from 64% to 25%.64

Irradiation for Breast Cancer In older series, postmastectomy radiation therapy for tumors of the left breast was associated with an increased risk of death from heart disease.19,20 Much of this experience was based on radiation given after radical mastectomy. A 1994 meta-analysis of the effects of postmastectomy ­irradiation showed a significantly increased risk of death from heart disease among patients who survived at least 10 years after the treatment ended.21 Since that time, several reviews of large numbers of patients with long-term follow-up after radiation therapy for breast cancer have been published describing the risks of cardiac morbidity and mortality associated with incidental cardiac irradiation.22,53,54,57,58,60,65,66 Some report increases in the risk of cardiac events between 10 years and 20 years after treatment, although not all studies included an age-matched control population for comparison. Increased cardiac risk has been associated with use of older radiation dosimetric techniques, and presumably the use of newer, more conformal techniques would be associated with less risk, even though the follow-up period for many patients treated with the new techniques has yet to reach 10 years. The risk of radiation-induced cardiac injury was increased among women undergoing left chest wall irradiation after mastectomy because large heart volumes were included; the more conservative approach of lumpectomy followed by postoperative partial-breast irradiation reduces exposure of the heart.23,67 Even with newer techniques, treatment of left-sided breast cancer seems to cause more cardiac morbidity than treatment of right-sided breast cancer. A 2006 review of 961 patients with stage I or stage II breast cancer ­treated

with radiation after lumpectomy from 1977 to 1994 found no differences in the numbers of cardiac deaths in patients treated for left-sided compared with right-sided tumors for the first 20 years after treatment, but the risks of coronary artery disease and myocardial infarction for the patients with left-sided disease were twice those of the patients with right-sided disease.53 An analysis of more than 300,000 patients with breast cancer registered in the Surveillance, Epidemiology, and End Results (SEER) database found consistent increases in cardiac mortality ratios among patients undergoing radiation for left-sided tumors (compared with those with right-sided tumors), with the greatest difference among those diagnosed from 1973 and 1982 and treated with older radiation techniques.57 A similar analysis of 8363 patients with left-sided breast cancer and 7909 with right-sided breast cancer diagnosed from 1986 to 1993 showed no differences in cardiac morbidity rates during the first 15 years after irradiation.66 A multi­ national study of 2128 patients, the Multinational Monitor­ ing of Trends and Determinants in Cardiovascular Disease (MONICA) project, also showed no differences in the incidence of cardiac morbidity and mortality at 10 years after lumpectomy and breast irradiation among patients with right- or left-sided breast cancer diagnosed from 1982 to 1989.60 Echocardiography and myocardial perfusion scintigraphy, especially newer techniques that incorporate molecular radioligands, may be useful for detecting and evaluating the severity of subclinical cardiotoxicity after irradiation.29 Findings from functional imaging of the heart after postlumpectomy irradiation to the left breast have varied. One study with thallium 201 angioscintigraphy failed to show perfusion defects.68 However, a subsequent study with technetium 99m sestamibi or tetrofosmin imaging revealed heart perfusion defects in about 40% of the patients studied, with the incidence varying from 10% to 20% among those with less than 5% of the left ventricle in the radiation field to as high as 50% to 60% when 75% of the left ventricle was in the field.69 Some wall motion abnormalities were observed, although functional effects were not evident.69 A smaller study of myocardial perfusion after “modern technique radiotherapy” for early-stage breast cancer showed abnormalities in 17 (71%) of 24 patients with left breast cancer who had had at least 1 cm of heart in the treatment field, but LVEF was normal in all patients, and the perfusion defects were not thought to require treatment. By comparison, perfusion defects were seen in only 2 (17%) of 12 patients after treatment for right-sided breast cancer (P = .002).28

Irradiation for Hodgkin Disease Compared with radiation therapy for breast cancer, radiation therapy for Hodgkin disease typically involves a somewhat lower cardiac dose (35 to 45 Gy in 4 to 5 weeks) but often includes a larger cardiac volume. In older reports, the cardiac doses often exceeded 50 Gy.32 More modern techniques that include partial transmission block shielding of the heart, shrinking fields, or interrupted therapy ­generally expose smaller heart volumes, and the doses are often 35 Gy or less. Accordingly, follow-up of these patients suggests a lower incidence of cardiac injury than in patients treated previously. For example, the relative risks of myocardial infarction or cardiac death for patients ­ treated in

18  The Blood Vessels and Heart

the 1970s and 1980s ranged from 1.8 to 3.1.24,26,70 In contrast, two later reviews found no differences in risk of coronary artery disease, abnormalities in myocardial perfusion, or LVEF in patients with Hodgkin disease who were or were not given mediastinal radiation (median cardiac dose, 35 Gy).52,71 However, a subsequent study that ­included ­electrocardiography, echocardiography, and exercise stress testing ­revealed that 47 of 48 long-term survivors of Hodgkin disease (median mediastinal dose, 40 Gy) showed evidence of cardiac abnormalities in patients with no symptomatic heart disease at a median 14.3 years after treatment.72 Clinically significant conduction defects were present in 75% of patients, valvular disease in 42%, and reduced peak oxygen uptake in 30%.72 Similarly, a group of 294 patients who had received at least 35 Gy to the mediastinum between 1964 and 1994 were shown to have a higher than expected incidence of valvular disease, especially in the aortic valve, on echocardiography; moreover, the incidence increased over time, being higher among patients treated 20 years previously than among those treated 10 years previously.73 These 294 patients were also found to be at greater risk for diastolic dysfunction,74 stress­induced signs of ischemia, and significant coronary artery disease.73-75 In another study of 286 patients with bulky mediastinal disease given radiation therapy after induction chemotherapy (median follow-up time, 24 years), the 15-year actuarial risk of coronary artery disease requiring surgical intervention was 4.1%.59 A variety of cardiac abnormalities can appear years after mediastinal radiation therapy, even though the risk is probably lower with newer radiation techniques that reduce the cardiac dose and treatment volume. The relative risk of cardiac morbidity increases with younger age at treatment, longer follow-up, and higher cardiac dose-volumes.58 Moreover, these changes may be progressive. Despite the apparent lack of clinical symptoms, special attention should be given to screening long-term survivors of Hodgkin disease for cardiovascular disease.

Irradiation for Seminoma The experience with seminoma provides some information about the effects of cardiac doses less than 30 Gy. Lederman and colleagues76 reported a 3.4% incidence of death resulting from acute myocardial infarction in a group of 58 patients with seminoma treated with mediastinal irradiation to 24 Gy compared with no cases of infarction in 61 patients with seminoma treated to the abdomen only. The relative risk of infarction was 1.97 (P = .019); however, this risk was not different from the risk in the large Framingham normal population study.77 In a comparison of risk of cardiovascular disease among more than 2500 patients with seminoma who had received mediastinal radiation after multiagent chemotherapy versus that in the general population,78 the standardized incidence ratio for coronary artery disease was higher among the irradiated patients at a median follow-up time of 18.4 years. Mediastinal radiation was associated with a 3.7-fold increase in risk of myocardial infarction compared with surgery alone or infradiaphragmatic radiation (neither of which was associated with an increased risk of myocardial infarction). However, the possible additive effect of the chemotherapy regimen given (cisplatin, vinblastine, and bleomycin) was not clear.78 Another study of 453 long-term survivors of

419

seminoma who had been treated with mediastinal radiation showed an increase in the cardiac standardized mortality ratio, but only beyond 15 years after the irradiation (P < .01).79

Summary of Studies Concerning the Cardiac Effects of Irradiation Some older series show a modest but significant increase in the risk of acute myocardial infarction after mediastinal irradiation, a risk that may be related to dose and volume. Newer reviews report lower risk levels of cardiac damage and less common functional effects among patients treated for breast cancer or Hodgkin disease. For Hodgkin disease, it seems that the previously reported increased risk of death resulting from heart disease may relate to the use of older radiation therapy techniques involving higher total doses and no cardiac shielding. The use of death certificates to determine end points and the absence of clear information about radiation techniques render these findings inconclusive. Treatment with adequate heart sparing yields minor subclinical functional effects. Because acute myocardial infarction has been reported at radiation doses of less than 40 Gy, consideration of reducing the cardiac dose or the volume irradiated should be a routine part of treatment planning for all patients receiving mediastinal irradiation, especially young patients with a good probability of longterm survival.

Coronary Irradiation to Prevent Restenosis An in-depth discussion of coronary irradiation to prevent restenosis is beyond the scope of this chapter. However, the emergence of intravascular brachytherapy as an effective means of reducing the risk of coronary artery restenosis after angioplasty or stenting for atherosclerotic narrowing of the coronary arteries warrants discussion of the relevant radiobiologic mechanisms of action. Findings from animal studies suggest that there is no reason to believe that irradiation reverses established coronary stenosis; however, irradiation is thought to prevent the overexuberant healing response associated with balloon injury or stent implantation. Studies of animals suggest that activated adventitial macrophages or monocytes are responsible for initiating the arterial neointimal hyperplasia and vascular remodeling that occur after angioplasty.80 This reaction is the prerequisite for the initiation and perpetuation of atheromatous thickening, whether incited by balloon injury or occurring as a natural process. A cytokine cascade, including platelet-derived growth factor, is likely responsible for initiating the macrophage response.81 Smooth muscle cells also are stimulated to migrate and proliferate. Radiation delivered immediately after balloon injury probably is effective because the macrophage population is very radiosensitive; irradiation short-circuits this process. Several large trials showed that intravascular brachytherapy reduced the rates of coronary restenosis after the placement of stents.82-87 However, clinical tests have shown that medicated stents can also reduce ­ restenosis rates after angioplasty to less than 10%. These stents elute antiproliferative agents such as sirolimus or paclitaxel and have the advantage of not requiring the use of ­radiation-protection measures during their insertion.88-90

420

PART 5  Thorax

As a consequence, the use of intravascular brachytherapy for treating coronary vessels may be declining. Intravascular brachytherapy also has been used to prevent restenosis after angioplasty for atherosclerotic peripheral vascular disease and to prevent stenosis of peripheral vascular ­access vessels.91,92

Prevention of Radiation-Induced Heart Damage Numerous studies have linked improvements in dosimetry and treatment planning with decreased dose-volume irradiation of the heart. These improvements include three-dimensional conformal radiation therapy, intensity­modulated radiation therapy, and the use of breath ­control or gating for respiratory and cardiac motion during treatment.93-103 Most of these studies compared treatment plans generated by these newer techniques with plans generated with conventional dosimetry, and only a few have reported preliminary, correlative data on cardiac toxicity. In addition to dosimetric efforts to reduce radiation exposure of the heart, radioprotectors may be useful for reducing the toxic effects of radiation when high doses to the heart are unavoidable. In rat models, cardioprotection with antioxidants such as the aminothiol amifostine has been shown to reduce both myocardial and coronary damage from ­irradiation,104,105 and findings from some early clinical ­trials have been promising.106,107 Dosimetric and radioprotective techniques such as these may improve sparing of the heart during radiation therapy and further reduce the incidence and severity of cardiac damage. Regarding appropriate TD5/5 and TD50/5 values, current findings suggest that the values published by Emami and colleagues in 199138 are a good starting point but that those values should be adjusted downward and additional doseeffect end points should be added to reflect more ­ subtle forms of cardiac injury (e.g., changes in perfusion and function). Increasing awareness of the need to reduce the risk of cardiotoxicity from radiation in combination with the ability to detect subclinical cardiac injury may lead to the publication of revised cardiac TD5/5 and TD50/5 levels over the next several years.

Management Strategies for Patients at Risk for Radiation-Induced Heart Damage In view of the subtle functional effects of irradiation on the heart that can occur after conventional radiation doses, techniques to limit the dose to the heart and the volume of heart irradiated are especially important in patients expected to have long-term survival, such as those with Hodgkin disease or breast cancer. Careful ­follow-up of patients who undergo cardiac irradiation in any form is mandatory. These evaluations should include a careful clinical history, with special attention to signs and symptoms that may be transient and illusory, and attention to cardiac diameter and silhouette on ­ followup chest radiography, electrocardiography, echocardiography, and periodic radionuclide angiography. In the case of transient episodes of overt pericardial effusion, pericardiocentesis and treatment with an anti-inflammatory agent may be sufficient. However, chronic recurrent effusions, especially with the development of constrictive changes and tamponade, may require therapeutic pericardiectomy.108,109

REFERENCES    1. Fajardo LF. Pathology of Radiation Injury. New York: Masson, 1982, pp 186-200.    2. Gillette EL, Maurer GD, Severin GA. Endothelial repair of ­radiation damage following beta irradiation. Radiology 1975;116: 175-177.    3. Gillette EL, McChesney SL, Hoopes PJ. Isoeffect curves for ­radiation-induced cardiomyopathy in the dog. Int J Radiat Oncol Biol Phys 1985;11:2091-2097.    4. Reinhold HS, Keyeux A, Dunjic A, et al. The influence of radiation on blood vessels and circulation. XII. Discussion and conclu­ sions. Curr Top Radiat Res Q 1974;10:185-198.    5. Rubin P, Casarett GW. Clinical Radiation Pathology. Philadelphia: WB Saunders; 1968.    6. Van den Brenk HAS. The effect of ionizing radiations on capillary sprouting and vascular remodeling in the regenerating repair blastema observed in the rabbit ear chamber. Radiology 1959;81:859-884.    7. Hilmas DE, Gillette EL. Tumor microvasculature following fractionated x irradiation. Radiology 1975;116:165-169.    8. Lindsay S, Kohn HI, Dakin RL, et al. Aortic arteriosclerosis in the dog after localized aortic x irradiation. Circ Res 1962;10:51-60.    9. Roscher AA, Steele BC, Woodward JS. Carotid artery rupture after irradiation of the larynx. Arch Otolaryngol 1966;83: 472-476.   10. Asscher AW. The delayed effects of renal irradiation. Clin Radiol 1964;15:320-325.   11. Silverberg GD, Britt RH, Goffinet DR. Radiation-induced ­carotid artery disease. Cancer 1978;41:130-137.   12. Taber P, Kosobkin MT, Gooding LT, et al. Growth of the abdominal aorta and renal arteries in childhood. Radiology 1972;102:129-134.   13. Colguhoun J. Hypoplasia of the abdominal aorta following therapeutic irradiation in infancy. Radiology 1966;86:454-456.   14. McChesney SL, Gillette EL, Powers BE, et al. Early radiation response of the canine heart and lung. Radiat Res 1991;125:34-40.   15. Lauk S, Trott KR. Endothelial cell proliferation in the rat heart following local heart irradiation. Int J Radiat Biol 1990;57: 1017-1030.   16. Morgan G, Freeman A, McLean R, et al. Late cardiac, thyroid, and pulmonary sequelae of mantle radiotherapy for Hodgkin’s disease. Int J Radiat Oncol Biol Phys 1985;11:1925-1931.   17. Fajardo LF. Radiation-induced coronary artery disease. Chest 1977;71:563-564.   18. Kopelson G, Herwig KJ: The etiologies of coronary artery disease in cancer patients. Int J Radiat Oncol Biol Phys 1978;4: 895-906.   19. Rutqvist LE, Lax I, Fornander T, et al. Cardiovascular mortality in a randomized trial of adjuvant radiation therapy versus surgery alone in primary breast cancer. Int J Radiat Oncol Biol Phys 1992;22:887-896.   20. Host H, Brennhovd IO, Loeb M. Postoperative radiotherapy in breast cancer: long-term results from the Oslo study. Int J Radiat Oncol Biol Phys 1986;12:727-732.   21. Cuzick J, Stewart H, Rutqvist L, et al. Cause-specific mortality in long-term survivors of breast cancer who participated in trials of radiotherapy. J Clin Oncol 1994;12:447-453.   22. Early Breast Cancer Trialists’ Collaborative Group. Effects of radiotherapy and surgery in early breast cancer: an overview of the randomized trials. N Engl J Med 1995;333:1444-1455.   23. Nixon AJ, Manola J, Gelman R, et al. No long-term increase in cardiac-related mortality after breast-conserving surgery and radiation therapy using modern techniques. J Clin Oncol 1998;16:1374-1379.   24. Boivin JF, Hutchinson GB, Lubin JH, et al. Coronary artery disease mortality in patients treated for Hodgkin’s disease. Cancer 1992;69:1241-1247.   25. Boivin J, Hutchinson GB. Coronary heart disease mortality after irradiation for Hodgkin’s disease. Cancer 1982;49:2470-2475.   26. Hancock S, Donaldson S, Hoppe R. Cardiac disease following treatment of Hodgkin’s disease in children and adolescents. J Clin Oncol 1993;11:1208-1215.

18  The Blood Vessels and Heart   27. Goethals I, Dierckx R, De Meerleer G, et al. The role of nuclear medicine in the prediction and detection of radiation­associated normal pulmonary and cardiac damage. J Nucl Med 2003;44:1531-1539.   28. Seddon B, Cook A, Gothard L, et al. Detection of defects in myocardial perfusion imaging in patients with early breast cancer treated with radiotherapy. Radiother Oncol 2002;64:53-63.   29. Goethals I, De Winter O, De Bondt P, et al. The clinical value of nuclear medicine in the assessment of irradiation-induced and anthracycline-associated cardiac damage. Ann Oncol 2002;13: 1331-1339.   30. Jingu K, Kaneta T, Nemoto K, et al. The utility of 18F-fluorodeoxyglucose positron emission tomography for early diagnosis of radiation-induced myocardial damage. Int J Radiat Oncol Biol Phys 2006;66:845-851.   31. Stewart JR. Normal tissue tolerance to irradiation of the cardiovascular system. Front Radiat Ther Oncol 1989;23:302-309.   32. Byhardt RW, Brace K, Ruckdeschel JC, et al. Dose and treatment factors in radiation related pericardial effusion associated with a mantle technique for Hodgkin’s disease. Cancer 1975;35: 795-802.   33. Wondergem J, Franken NA, van der Laarse A, et al. Changes in cardiac performance and sympathetic stimulation during and after fractionated radiotherapy in a rat model. Radiother Oncol 1996;38:33-40.   34. Mill WB, Baglan RJ, Kurichety P, et al. Symptomatic radiationinduced pericarditis in Hodgkin’s disease. Int J Radiat Oncol Biol Phys 1984;10:2061-2065.   35. Gomez GA, Park JJ, Panahon AM, et al. Heart size and function after radiation therapy to the mediastinum in patients with Hodgkin’s disease. Cancer Treat Rep 1983;67:1099-1103.   36. Gottdiener JS, Katin MJ, Borer JS, et al. Late cardiac effects of therapeutic mediastinal irradiation: assessment by echocardiography and radionuclide angiography. N Engl J Med 1983;308: 569-572.   37. Stewart JR, Fajardo L. Radiation-induced heart disease: an update. Prog Cardiovasc Dis 1984;27:173-194.   38. Emami B, Lyman J, Brown A, et al. Tolerance of normal tissue to therapeutic irradiation. Int J Radiat Oncol Biol Phys 1991;21:109-122.   39. Gagliardi G, Lax I, Rutquist LE. Partial irradiation of the heart. Semin Radiat Oncol 2000;11:224-233.   40. McChesney SL, Gillette EL, Orton EC. Canine cardiomyopathy after whole heart and partial lung irradiation. Int J Radiat Oncol Biol Phys 1988;14:1169-1174.   41. Stewart R, Fajardo LF. Dose response in human and experimental radiation-induced heart disease. Radiology 1971;99:403-408.   42. Applefeld MM, Cole JF, Pollock SH, et al. The late appearance of chronic pericardial disease in patients treated by radiation therapy for Hodgkin’s disease. Ann Intern Med 1981;94:338-341.   43. Rogoway WM, Finkelstein S, Rosenberg SA, et al. Myxedema developing after lymphangiography and neck irradiation. Clin Res 1966;14:133.   44. Girinsky T, Cosset JM. Pulmonary and cardiac late effects of ionizing radiations alone or combined with chemotherapy. Cancer Radiother 1997;1:735-743.   45. Geist BJ, Lauk S, Bornhausen M, et al. Physiologic consequences of local heart irradiation in rats. Int J Radiat Oncol Biol Phys 1990;18:1107-1113.   46. van Luijk P, Novakova-Jiresova A, Faber H, et al. Radiation damage to the heart enhances early radiation-induced lung function loss. Cancer Res 2005;65:6509-6511.   47. Lauk S, Bohm M, Feiler G, et al. Increased number of cardiac adrenergic receptors following local heart irradiation. Radiat Res 1989;119:157-165.   48. Boerma M, Zurcher C, Esveldt I, et al. Histopathology of ventricles, coronary arteries and mast cell accumulation in transverse and longitudinal sections of the rat heart after irradiation. Oncol Rep 2004;12:213-219.   49. Wondergem J, Persons K, Zurcher C, et al. Changes in circulating atrial natriuretic peptide in relation to the cardiac status of Rhesus monkeys after total-body irradiation. Radiother Oncol 1999;53:67-75.

421

  50. Kruse JJ, Bart CI, Visser A, et al. Changes in transforming growth factor-beta (TGF-beta 1), procollagen types I and II mRNA in the rat heart after irradiation. Int J Radiat Biol 1999;75: 1429-1436.   51. Gyenes G, Fornander T, Carlens P, et al. Detection of radiationinduced myocardial damage by technetium-99m sestamibi scintigraphy. Eur J Nucl Med 1997;24:286-292.   52. Constine LS, Schwartz RG, Savage DE, et al. Cardiac function, perfusion, and morbidity in irradiated long-term survivors of Hodgkin’s disease. Int J Radiat Oncol Biol Phys 1997;39: 897-906.   53. Harris EE, Correa C, Hwang WT, et al. Late cardiac mortality and morbidity in early-stage breast cancer patients after breastconservation treatment. J Clin Oncol 2006;24:4100-4106.   54. Kaplan BM, Miller AJ, Bharati S, et al. Complete AV block following mediastinal radiation therapy: electrocardiographic and pathologic correlation and review of the world literature. J Interv Card Electrophysiol 1997;1:175-188.   55. Veinot JP, Edwards WD. Pathology of radiation-induced heart disease: a surgical and autopsy study of 27 cases. Hum Pathol 1996;27:766-773.   56. Mittal S, Berko B, Bavaria J, et al. Radiation-induced cardiovascular dysfunction. Am J Cardiol 1996;78:114-115.   57. Darby SC, McGale P, Taylor CW, et al. Long-term mortality from heart disease and lung cancer after radiotherapy for early breast cancer: prospective cohort study of about 300,000 women in US SEER cancer registries. Lancet Oncol 2005;6:557-565.   58. Adams MJ, Lipshultz SE, Schwartz C, et al. Radiation-associated cardiovascular disease: manifestations and management. Semin Radiat Oncol 2003;13:346-356.   59. Chronowski GM, Wilder RB, Tucker SL, et al. Analysis of in-field control and late toxicity for adults with early-stage Hodgkin’s disease treated with chemotherapy followed by radiotherapy. Int J Radiat Oncol Biol Phys 2003;55:36-43.   60. Vallis KA, Pintilie M, Chong N, et al. Assessment of coronary heart disease morbidity and mortality after radiation therapy for early breast cancer. J Clin Oncol 2002;20:1036-1042.   61. Bouyer-Dalloz F, Maingon P, Benderitter M, et al. Effect of in vivo heart irradiation on coronary reactivity in the rat. ���������� Cell Mol Biol ���������������������������������������������������������������������� (Noisy-le-grand) 2003; 49 Online Pub:OL435-42.   62. Ishikura S, Nihei K, Ohtsu A, et al. Long-term toxicity after definitive chemoradiotherapy for squamous cell carcinoma of the thoracic esophagus. J Clin Oncol 2003;21:2697-2702.   63. Tripp P, Malhotra HK, Javle M, et al. Cardiac function after chemoradiation for esophageal cancer: comparison of heart dose-volume histogram parameters to multiple gated acquisition scan changes. Dis Esophagus 2005;18:400-405.   64. Mukherjee S, Aston D, Minett M, et al. The significance of cardiac doses received during chemoradiation of oesophageal and gastro-oesophageal junctional cancers. Clin Oncol (R Coll Radiol) 2003;15:115-120.   65. Taylor CW, McGale P, Darby SC. Cardiac risks of breast-cancer radiotherapy: a contemporary view. Clin Oncol (R Coll Radiol) 2006;18:236-246.   66. Patt DA, Goodwin JS, Kuo YF, et al. Cardiac morbidity of adjuvant radiotherapy for breast cancer. J Clin Oncol 2005;23: 7475-7482.   67. Fuller SA, Haybittle JL, Smith REA, et al. Cardiac doses in postoperative breast irradiation. Radiother Oncol 1992;25:19-24.   68. Cowen D, Gonzague-Casabianca L, Brenot-Rossi I, et al. ­Thallium-201 perfusion scintigraphy in the evaluation of late myocardial damage in left-side breast cancer treated with adjuvant radiotherapy. Int J Radiat Oncol Biol Phys 1998;41:809-815.   69. Marks LB, Yu X, Prosnitz RG, et al. The incidence and functional consequences of RT-associated cardiac perfusion defects. Int J Radiat Oncol Biol Phys 2005;63:214-223.   70. Cosset JM, Henry-Amar M, Pellae-Cosset B, et al. Pericarditis and myocardial infarctions after Hodgkin’s disease therapy. Int J Radiat Oncol Biol Phys 1991;21:447-449.   71. Vlachaki MT, Ha CS, Hagemeister FB, et al. Long-term outcome of treatment for Ann Arbor stage 1 Hodgkin’s disease: patterns of failure, late toxicity and second malignancies. Int J Radiat Oncol Biol Phys 1997;39:609-616.

422

PART 5  Thorax

  72. Adams MJ, Lipsitz SR, Colan SD, et al. Cardiovascular status in long-term survivors of Hodgkin’s disease treated with chest radiotherapy. J Clin Oncol 2004;22:3139-3148.   73. Heidenreich PA, Hancock SL, Vagelos RH, et al. Diastolic dysfunction after mediastinal irradiation. Am Heart J 2005;150: 977-82.   74. Heidenreich PA, Hancock SL, Lee BK, et al. Asymptomatic cardiac disease following mediastinal irradiation. J Am Coll Cardiol 2003;42:743-749.   75. Heidenreich PA, Schnittger I, Strauss HW, et al. Screening for coronary artery disease after mediastinal irradiation for Hodgkin’s disease. J Clin Oncol 2007;25:43-49.   76. Lederman GS, Sheldon TA, Chaffey JT, et al. Cardiac disease after mediastinal irradiation for seminoma. Cancer 1987;60: 772-776.   77. Kannel WB, McGee D, Gordon T. A general cardiovascular risk profile: the Framingham study. Am J Cardiol 1976;38:46-51.   78. van den Belt-Dusebout AW, Nuver J, de Wit R, et al. Long-term risk of cardiovascular disease in 5-year survivors of testicular cancer. J Clin Oncol 2006;24:467-475.   79. Zagars GK, Ballo MT, Lee AK, et al. Mortality after cure of testicular seminoma. J Clin Oncol 2004;22:640-647.   80. Rubin P, Williams JP, Riggs PN, et al. Cellular and molecular mechanisms of radiation inhibition of restenosis. Part I: role of the macrophage and platelet-derived growth factor. Int J Radiat Oncol Biol Phys 1998;40:929-941.   81. Libby P, Schwartz D, Brogi E, et al. A cascade model for restenosis. A special case of atherosclerosis progression. Circulation 1992;86(Suppl III):47-52.   82. Teirstein PS, Massullo V, Jani S, et al. Catheter-based radiotherapy to inhibit restenosis after coronary stenting. N Engl J Med 1997;336:1697-1703.   83. Uchida T, Bakhai A, Almonacid A, et al. A meta-analysis of ­randomized controlled trials of intracoronary gamma- and betaradiation therapy for in-stent restenosis. Heart Vessels 2006;21: 368-374.   84. Geiger MH, Ludwig J, Burckhard R, et al. High dose intracoronary irradiation after de novo stent implantation results of the EVEREST (Evaluation of Endoluminal Radiation in Elective Stenting) trial. Strahlenther Onkol 2006;182:9-15.   85. Lee SW, Park SW, Hong MK, et al. Long-term outcomes after treatment of diffuse in-stent restenosis with rotational atherectomy followed by beta-radiation therapy with a 188Re-MAG3filled balloon. Int J Cardiol 2005;99:201-205.   86. Rha SW, Kuchulakanti PK, Pakala R, et al. Real-world clinical practice of intracoronary radiation therapy as compared to investigational trials. Catheter Cardiovasc Interv 2005;64:61-66.   87. Schwartz RS. Ionizing radiation and the coronary arteries: is the plot thickening? Catheter Cardiovasc Interv 2004;63:282-283.   88. Stone GW, Ellis SG, O’Shaughnessy CD, et al, for the TAXUS V ISR Investigators. Paclitaxel-eluting stents vs vascular brachytherapy for in-stent restenosis within bare-metal stents: the TAXUS V ISR randomized trial. JAMA 2006;295:1253-1263.   89. Feres F, Munoz JS, Abizaid A, et al. Comparison between sirolimus-eluting stents and intracoronary catheter-based beta radiation for the treatment of in-stent restenosis. Am J Cardiol 2005;96:1656-1662.   90. Rau M, Maikowski C, Weber M, et al. Treatment of in-stent restenosis with sirolimus-eluting-stents—a six month clinical and angiographic follow-up. Z Kardiol 2005;94:742-747.   91. Tripuraneni P, Giap H, Jani S. Endovascular brachytherapy for peripheral vascular disease. Semin Radiat Oncol 1999;9: 190-202.   92. Parikh S, Nori D. Radiation therapy to prevent stenosis of peripheral vascular accesses. Semin Radiat Oncol 1999;9:144154.   93. Ragazzi G, Cattaneo GM, Fiorino C, et al. Use of dose-volume histograms and biophysical models to compare 2D and 3D irradiation techniques for non-small cell lung cancer. Br J Radiol 1999;72:279-288.   94. Smitt MC, Li SD, Shostak CA, et al. Breast conserving radiation therapy: potential of inverse planning with intensity modulation. Radiology 1997;203:871-876.

  95. Chen MH, Chuang ML, Bornstein BA, et al. Impact of respiratory maneuvers on cardiac volume within left-breast radiation portals. Circulation 1997;96:3269-3272.   96. Hiatt JR, Evans SB, Price LL, et al. Dose-modeling study to compare external beam techniques from protocol NSABP B-39/RTOG 0413 for patients with highly unfavorable cardiac anatomy. Int J Radiat Oncol Biol Phys 2006;65:1368-1374.   97. Chang JY, Zhang X, Wang X, et al. Significant reduction of normal tissue dose by proton radiotherapy compared with three-dimensional conformal or intensity-modulated radiation therapy in stage I or stage III non-small-cell lung cancer. Int J Radiat Oncol Biol Phys 2006;65:1087-1096.   98. Cominos M, Mosleh-Shirazi MA, Tait D, et al. Quantification and reduction of cardiac dose in radical radiotherapy for oesophageal cancer. Br J Radiol 2005;78:1069-1074.   99. Fogliata A, Nicolini G, Alber M, et al. IMRT for breast. a planning study. Radiother Oncol 2005;76:300-310. 100. Cozzi L, Fogliata A, Nicolini G, et al. Clinical experience in breast irradiation with intensity modulated photon beams. Acta Oncol 2005;44:467-74. 101. Gladstone DJ, Flanagan MF, Southworth JB, et al. Radiationinduced cardiomyopathy as a function of radiation beam gating to the cardiac cycle. Phys Med Biol 2004;49:1475-1484. 102. Liu HH, Wang X, Dong L, et al. Feasibility of sparing lung and other thoracic structures with intensity-modulated radiotherapy for non–small-cell lung cancer. Int J Radiat Oncol Biol Phys 2004;58:1268-1279. 103. Lomax AJ, Cella L, Weber D, et al. Potential role of intensity­modulated photons and protons in the treatment of the breast and regional nodes. Int J Radiat Oncol Biol Phys 2003;55:785-792. 104. Tokatli F, Uzal C, Doganay L, et al. The potential cardioprotective effects of amifostine in irradiated rats. Int J Radiat Oncol Biol Phys 2004;58:1228-1234. 105. Kruse JJ, Strootman EG, Wondergem J. Effects of amifostine on radiation-induced cardiac damage. Acta Oncol 2003;42:4-9. 106. Wagdi P, Fluri M, Aeschbacher B, et al. Cardioprotection in patients undergoing chemo- and/or radiotherapy for neoplastic disease. A pilot study. Jpn Heart J 1996;37:353-359. 107. Capizzi RL. The preclinical basis for broad-spectrum selective cytoprotection of normal tissues from cytotoxic therapies by amifostine (Ethyol). Eur J Cancer 1996;32A(Suppl 4):S5-S16. 108. Morton DL, Glancy DO, Joseph WL. Management of ­patients with radiation-induced pericarditis with effusion. Chest 1973;64: 291-297. 109. Steinberg I. Effusive constrictive pericarditis. Am J Cardiol 1967;19:434-439.

ADDITIONAL READING Adams MJ, Hardenbergh PH, Constine LS, et al. Radiation-associated cardiovascular disease. Crit Rev Oncol Hematol 2003;45:55-75. Cohn KE, Stewart JR, Fajardo LF, et al. Heart disease following radiation. Medicine (Baltimore) 1967;46:281-298. Crocker I. Radiation therapy to prevent coronary artery restenosis. Semin Radiat Oncol 1999;9:134-143. De Caro E, Fioredda F, Calevo MG, et al. Exercise capacity in apparently healthy survivors of cancer. Arch Dis Child 2006;91:47-51. Eltringham JR, Fajardo LJ, Stewart JR. Adriamycin cardiomyopathy: enhanced cardiac damage in rabbits and combined drug and cardiac irradiation. Radiology 1975;115:471-472. Hafeez N. Prevention of coronary restenosis with radiation therapy: a review. Clin Cardiol 2002;25:313-322. Hall EJ. Radiobiology for the Radiologist. Philadelphia: Lippincott Williams & Wilkins, 2000, p 356. Hurkmans CW, Cho BC, Damen E, et al. Reduction of cardiac and lung complication probabilities after breast irradiation using conformal radiotherapy with or without intensity modulation. Radiother Oncol 2002;62:163-171. Korreman SS, Pedersen AN, Aarup LR, et al. Reduction of cardiac and pulmonary complication probabilities after breathing adapted radiotherapy for breast cancer. Int J Radiat Oncol Biol Phys 2006;65:1375-1380.

18  The Blood Vessels and Heart Mantini G, Smaniotto D, Balducci M, et al. Radiation-induced ­cardiovascular disease: impact of dose and volume. Rays 2005;30:157-168. Miller AJ. Some observations concerning pericardial effusions and the relationship to the venous and lymphatic circulation of the heart. Lymphology 1970;2:76-78. Poussin-Rosillo H, Nisce LZ, Lee BJ. Complications of total nodal irradiation of Hodgkin’s disease, stages III and IV. Cancer 1978;42:437-441. Prosnitz RG, Chen YH, Marks LB. Cardiac toxicity following thoracic radiation. Semin Oncol 2005;32(Suppl 3):S71-S80. Prosnitz RG, Marks LB. Radiation-induced heart disease: vigilance is still required. J Clin Oncol 2005;23:7391-7394. Heidenreich PA, Schnittger I, Strauss HW, et al. Screening for coronary artery disease after mediastinal irradiation for Hodgkin’s disease. J Clin Oncol 2007;25:43-49. Raj KA, Marks LB, Prosnitz RG. Late effects of breast radiotherapy in young women. Breast Dis 2005-2006;23:53-65. Recht A. Which breast cancer patients should really worry about radiation-induced heart disease—and how much? J Clin Oncol 2006;24:4059-4061.

423

Rosen EM, Fan S, Goldberg ID, et al. Biological basis of radiation sensitivity. Part I. Factors governing radiation tolerance. Oncology 2000;14:543-550. Rostock R, Siegelman S, Lenhard R, et al. Thoracic CT scanning for mediastinal Hodgkin’s disease: results and therapeutic implications. Int J Radiat Oncol Biol Phys 1983;9:1451-1457. Rubenstein JH. Spare the heart: how much effort should we expend to ensure cardiac tissue is not needlessly irradiated? Am J Clin Oncol 2006;29:217-218. Rubin P, Soni A, Williams JP. The molecular and cellular biologic basis for the radiation treatment of benign proliferative diseases. Semin Radiat Oncol 1999;9:203-214. Stewart JR, Cohn KE, Fajardo LF, et al. Radiation-induced heart disease. Radiology 1967;89:302-310. Stewart R, Fajardo LF. Radiation-induced heart disease. In Vaeth JM (ed): Radiation Effect and Tolerance, Human Tissue. Baltimore: University Park Press, 1972.

19   The Lung, Pleura, and Thymus Ritsuko Komaki, Elizabeth L. Travis, and James D. Cox RESPONSE OF THE TRACHEA AND LUNG TO IRRADIATION Normal Lung Response of the Lung to Irradiation Molecular Mechanisms of Lung Damage Volume of Lung Irradiated Clinical Changes Radiologic Changes Functional Effects Treatment of Radiation Pneumonitis RADIATION THERAPY FOR LUNG CARCINOMA Indications for Radiation Therapy Surgical Adjuvant Radiation Therapy

Response of the Trachea and Lung to Irradiation Thoracic irradiation is an important part of the therapy for many types of cancer, including esophageal cancer, breast cancer, and lung cancer. More than 50% of patients with these types of cancer will receive radiation therapy at some point during the management of the disease. Undiseased parts of the lung inevitably will receive some exposure as part of these treatments, and the lung is the dose-­limiting tissue in the treatment of these malignant neoplasms. ­Optimal treatment planning requires knowledge of the effects of radiation on the normal lung and other normal structures within the treatment field. Although nearly a century has passed since radiation pneumonitis and fibrosis were first described by Evans and Leucutia,1 we remain unable to circumvent or effectively treat these two side effects of pulmonary irradiation, and they continue to limit the effective treatment of lung cancer.

Normal Lung The normal lung can be divided into two major functional sections: the non–gas exchange unit and the gas exchange, or respiratory, unit. The first is a system of branching and bifurcating conducting airways, the tracheobronchial tree, which consists of all of the conducting airways from the trachea to the nonconducting bronchioles. The main func-­ tion of these airways is to transport air into and out of the lung. The respiratory unit begins at the ramification of the terminal bronchioles and includes the respiratory bronchi-­ oles and the alveolar ducts, which open into the alveolar sacs, each of which bears numerous alveoli, the actual sites of gas exchange. The number of alveoli increases along the whole length of the respiratory unit, and gas exchange ­occurs along the entire length of the respiratory unit as well. The alveoli are thin-walled structures bound on one side by the alveolar epithelium and on the other by capillaries.

424

Curative Therapy for Inoperable Non–Small Cell Lung Carcinoma Radiation Therapy for Small Cell Lung Carcinoma Radiation Therapy for Apical Sulcus Tumors Radiation Therapy for Superior Vena Cava Obstruction INOPERABLE CARCINOMA OF THE LUNG Prognosis Palliative Irradiation THE THYMUS Thymomas Treatment Results

The alveolar epithelium consists of two cell types—type I and type II—which together form a continuous surface covering the basement membrane; this surface forms the interface with the air spaces of the lung. Type I cells are squamous, flat, and normally devoid of organelles. Their attenuated cytoplasmic extensions allow them to cover ­several adjacent alveoli, providing efficient gas exchange. Type I cells have little capacity to respond to injury, and when they are damaged, they must be replaced by the second type of surface epithelial cell, the type II cell. Unlike type I cells, which are difficult to identify under the light microscope because of their ­ attenuated cyto-­ plasm, type II cells are easily identified in normal lung tissue by their distinct cuboidal shape, their vacuolated cytoplasm, and their localization at the corners of the alveolar walls. Type II cells synthesize and secrete surfactant, the ­lipid­protein complex that prevents atelectasis at low lung vol-­ umes by reducing surface tension at the air-­alveolar ­interface. In addition to their major functional role in ­ pulmonary function, type II cells are the progenitor cells during normal cell turnover and after injury. The type II cells can divide and differentiate into type I cells or proliferate to replenish their own population, restoring normal tissue structure and function. Type II cells are actively involved in the transport of sodium ions to control fluid balance in the alveolar space. The alveolar epithelium rests on a basement membrane, on the other side of which are the endothelial cells that make up the capillaries of the pulmonary vasculature. Other cell types normally present in the lung are pul-­ monary macrophages and fibroblasts. Both are important in pulmonary repair after injury.

Response of the Lung to Irradiation The lower respiratory tract is composed of many types of tissues; some tolerate radiation well, and others are rela-­ tively sensitive. The trachea and bronchi are lined with

19  The Lung, Pleura, and Thymus

425

these cells. Alveolar macrophages and proteinaceous mate-­ rials fill the alveolar spaces, the latter as a result of capillary leakage from the depopulated alveolar epithelium. The interstitium becomes thickened, further decreasing gas exchange and pulmonary perfusion. As the inflammation wanes, alveolar exudate becomes organized, and fibroblasts proliferate. The incorporation of the fibroproliferative process into the alveolar walls signals the initiation of the fibrotic phase. Interstitial fibrosis, with scarring and altered function of the alveolar unit, occurs in concert with thickening of the alveolar septa and capillary walls. This process leads to airspace and vascular consolidation and obliteration, with remodeling of the pulmonary architecture and attendant loss of pulmonary function (Fig. 19-2). The final phase of pulmonary fibrosis is considered chronic, progressive, and irreversible.

a ciliated columnar epithelium that is shed only at high radiation doses. This reaction was emphasized when La­ cassagne2 directed a beam sufficient to produce a loss of esophageal epithelium to a rabbit’s mediastinum, produc-­ ing noticeable early macroscopic change in the tracheal lining. Microscopic changes consist of thickening of the epithelium and an increased number of goblet cells. These changes are accompanied by a decrease in ciliary function and a dryness that may result in coughing. Bronchoscopy after therapeutic doses (50 to 60 Gy) reveals a reddened, hyperemic mucosa and thickened secretions that tend to accumulate and obstruct the lumen. In contrast to the tracheobronchial tree, the lung is relatively radiosensitive. Irradiation of the whole lung or a major portion of it results in distinct phases of damage characterized by differences in the time of expression ­after irradiation and distinct histologic changes or molecular changes, or both. Two of these phases of damage, radiation pneumonitis and fibrosis, have been well characterized,3 although their pathogenesis remains unclear. Traditionally, the lung’s response to irradiation is de-­ scribed as having a latent phase, occurring immediately and up to 3 months after irradiation; an exudative phase, also known as radiation pneumonitis, occurring from 3 to 6 months after irradiation; and a final fibrotic phase, occur-­ ring from 6 months after irradiation onward. Many cell types are involved in this process, but those thought to be the most critical to the process are type II cells, macro-­ phages, and fibroblasts. During the latent phase of injury, the lung appears ­histologically and radiographically normal, although the biologic processes that culminate in the later outward manifestations of pulmonary irradiation begin during this time. The exudative phase, classic radiation pneumonitis, is characterized by an inflammatory cell infiltrate of mac-­ rophages and lymphocytes accompanied by edema in the air and interstitial spaces (Fig. 19-1). Type I and type II cells are desquamated into the alveolar spaces, removing the barrier between the interstitial capillaries and the airalveolar wall interface. The loss of type II cells results in alterations in surfactant composition and production by

The phases of radiation-induced lung damage are a cul-­ mination of changes in the pulmonary parenchyma and vasculature arising from a dynamic interaction among ­different cell types, including resident cells such as type II cells, ­ endothelial cells, and fibroblasts and nonresi-­ dent cells, notably macrophages and lymphocytes. How-­ ever, the idea that killing and depletion of one or more “target cells” is responsible for the later manifestation of ­pulmonary ­injury has been replaced by a model in which complex ­interactions among these cells and the messages produced by them in response to injury culminate months later in the signs and symptoms of radiation pneumonitis and fibrosis.4,5 At the time of irradiation, events initiated at the molecular level set in motion a persistent cascade of molecular processes that continue and persist through the fibrotic phase. Soluble mediators stimulate cells to overproduce those extracellular matrix components that characterize fibrosis. After their initial contact with the cell surface, these molecules activate intracellular signaling pathways, which activate the expression of specific genes that determine the phenotype of cells involved in lung in-­ jury. Although the factors ­modulating these events are not

FIGURE 19-1.  Histologic section from the lung of a C3Hf/Kam mouse irradiated to the whole thorax and killed 4 months later shows classic radiation pneumonitis, with inflammatory cell infiltrate and edema in the air spaces and interstitium.

FIGURE 19-2.  Histologic section from the lung of a C57B1/6J  mouse killed 5 months after the administration of 16 Gy to the whole thorax shows organizing alveolitis and collagen ­deposition.

Molecular Mechanisms of Lung Damage

426

PART 5  Thorax

completely defined, information about some pathways is beginning to emerge (Fig. 19-3).5 Rabbit models have shown that immediate damage from irradiation results in membrane, intracellular, or DNA strand damage that leads to the altered expression and im-­ mediate release of such growth factors as transforming growth factor β (TGF-β), platelet-derived growth factor (PDGF), and interleukin-1 (IL-1), which activate growth factor receptors, resulting in activation of collagen genes and increased production of collagen.5 These changes are thought to begin and persist throughout the latent phase and throughout the overt expression of damage. Intercel-­ lular interactions occur through cytokine pathways that produce changes in the genetic expression of collagen pre-­ cursors. The intricate pathways that influence fibrocyte and fibroblast maturation are affected in an orderly fashion, ­resulting in an altered fibrocyte-to-fibroblast ratio condu-­ cive to collagen production and fibrosis.6 TGF-β is an important mediator of tissue damage in a variety of conditions characterized by excessive collagen production and accumulation.7-10 TGF-β is a pleiotropic cytokine that stimulates fibroblast and lymphocyte recruit-­ ment to the site of injury, promotes fibroblast proliferation, and stimulates the production of collagen and fibronectin, all of which results in a net increase in extracellular matrix material. In addition to stimulating the overproduction of connective tissue, TGF-β downregulates the degradation of

extracellular matrix, which ultimately results in matura-­ tion of the connective tissue to form a scar. Experimental studies have confirmed the expected increase in TGF-β in postirradiation lung fibrosis.7-11 H­owever, TGF-β levels are also elevated during the latent phase after irradiation, long before the onset of pneumo-­ nitis or fibrosis. In a chemical model of pulmonary fibro-­ sis, elevations in TGF-β have been observed 1 week after bleomycin exposure in mice that developed pulmonary fibrosis 3 weeks later.12 Antibodies to this cytokine and a receptor antagonist reduce collagen accumulation in the lung after bleomycin exposure. Even if this cytokine is not the initial trigger in the fibrotic process, it clearly is a downstream effector in this process, and as such, it is an attractive ­ molecular target for interventional strat-­ egies. Martin and colleagues13 suggested that TGF-β has a dual function in the fibrotic process, acting as a master switch for the process and a specific target for antifibrotic agents. In the latter case, pentoxifylline and vitamin E have been shown to reduce radiation-induced subcutaneous ­fibrosis in ­ humans,14 and antibodies to the leukocyte in-­ tegrin CD11 reduce ­radiation-induced fibrosis in a mouse model.15 These findings challenge the long-held view that tissue fibrosis is ­irreversible. A third role for TGF-β in the treatment of lung cancer with radiation is as a predictor of radiation pneumonitis and fibrosis, as shown by Anscher and colleagues.16 ­Anscher and

IFN-γ BCGF

IL-1ra MIF

Immune cells

IL-6 IL-8 MCP-1

TGF-β M-CSF G-CSF

Fibroblasts

IFN-α

TNF-α

IL-1α

IL-1β

Endothelial cells

Other cells

Alveolar macrophage

Stimulus

TGF-α TGF-β MPGF

Fibronection IGF-1 TNF-α TGF-α IL-1 TGF-β IL-6 PDGF

PDGF IL-8

Proliferation

IL-9 IL-10 TNF-β LIF OncoM GM-CSF

IFN-β

IL-11

IL-12

EPO

Proliferation

Fibroblast

Differentiation TGF-α TGF-β EGF

IL-2 IL-3 IL-4 IL-5 IL-7

Fibronectin GF-1 IL-6 IL1

Matrix protein synthesis (fibrosis)

Type II cell

FIGURE 19-3.  Schema of cell-to-cell interactions and control of gene expression by growth factors in lung injury. CSF, colony-stimulating factor; BCGF, B-cell growth factor; EGF, endothelial growth factor; EPO, erythropoietin; IFN, interferon; IL, interleukin; IL-1ra, interleukin-1 receptor antagonis; GM-CSF, granulocyte-macrophage colony-stimulating factor; IGF-1, insulin-like growth factor; LIF, leukocyte inhibitory factor; MIF, macrophage inhibitory factor; MPGF, macrophage-phagocyte growth factor; OncoM, oncostatin M, PDGF, platelet-derived growth factor; TGF, tumor growth factor; TNF, tumor necrosis factor. (From Rubin P, Johnston CJ, Williams JP, et al. A perpetual cascade of cytokines postirradiation leads to pulmonary fibrosis. Int J Radiat Oncol Biol Phys 1995;33:99-109.)

19  The Lung, Pleura, and Thymus

Volume of Lung Irradiated An effective method of reducing the damage and morbid-­ ity from pulmonary irradiation is to limit the volume of lung irradiated, which is the rationale for the use of tech-­ niques such as conformal radiation therapy (CRT) and intensity-modulated radiation therapy (IMRT). Radiation oncologists routinely limit the volume of nondiseased tis-­ sue irradiated while still encompassing the entire tumor volume. However, the relationships among the volume of lung irradiated, the dose, and the oncologic and functional outcomes remain unclear. A question that has yet to be answered, even in the era of CRT or IMRT, is whether it is better to give a large dose to a small volume than a small dose to a large volume. In an attempt to address this question, Gopal and ­colleagues18 studied the relationship between radiation dose and loss of lung function in 26 patients by measuring the ability of each patient’s lungs to diffuse carbon mon-­ oxide before and after therapy. This group found a drastic ­decrease in this capacity when the radiation dose exceeded 13 Gy (95% confidence interval [CI], 11 to 15 Gy); how-­ ever, this effect was mitigated by amifostine, which post-­ poned the loss to a dose level above 36 Gy (95% CI, 25 to 48 Gy). These investigators also confirmed that treating a small volume of normal lung with a high dose of radia-­ tion had fewer deleterious effects than treating a high vol-­ ume of normal lung with a low dose. In another attempt to define the relationships of dose, volume, and radiation ­pneumonitis, Liao, Tucker, and Travis19-21 undertook a largescale mouse study that included two assays of total lung function—breathing rate and lethality. In that study, a range of lung volumes from 20% to 100% was irradiated with a range of doses to produce dose-response curves for each volume. The effect of the location within the lung of the irradiated volume on morbidity was also determined by irradiating similar volumes in the lung’s base and apex. These experiments tested the hypothesis that the respira-­ tory unit, the functional subunit in the lung, is randomly and homogeneously distributed throughout the lung. Figure 19-4 shows the proportions of dead mice after the irradiation of two matched volumes (50% and 75%) in the apex and base of the lung.22 These findings indicate that the base of the lung is more sensitive than the apex, at least for these two irradiated volumes. This relationship was subsequently found to hold for any volume of lung ­irradiated (Fig. 19-5)21,22; the isoeffect dose calculated from the dose-response curves for lethality from radiation pneumonitis was always lower if the irradiated volume was in the base of the lung rather than the apex. The base of

100 B 70%

100%

50 A 75%

Deaths (%)

others17 showed that the plasma level of TGF-β at the end of treatment, combined with the volume of lung ­irradiated, can be used to reliably predict the outcome of radiation therapy, enabling the identification of patients who can ­tolerate relatively high radiation doses. Just as the concept that the killing and depletion of a target cell may be responsible for radiation pneumonitis and fibrosis is now known to be simplistic, it is likely that TGF-β, although a critical factor in fibrosis, is not the only one involved in the complex fibrotic process. Tumor necro-­ sis factor α (TNF-α), FAS and FAS ligand, basic fibroblast growth factor (bFGF), and other molecules have been sug-­ gested as being critical to fibrosis.

427

0 100 B 50%

100%

50

A 50%

0 8

12

16

20

24

Radiation dose (Gy)

FIGURE 19-4.  Dose-response curves of lethality from radiation pneumonitis after irradiation of two partial volumes of lung, 70% to 75% (top) and 50% (bottom), in the apex (A) or base (B) of C3H mice. Irradiation of 100% of the lung is shown for comparison. These data show a clear volume effect for radiation pneumonitis in the lung and dependence on the location of the partial volume irradiated. (From Travis EL, Komaki R. Treatmentrelated lung damage. In Pass HI, Mitchell JB, Johnson DH, et al [eds]: Lung Cancer: Principles and Practice, 3rd ed. Philadelphia: Lippincott Williams & Wilkins, 2005.)

the lung is more sensitive than the apex, at least in this mouse model. The mechanism underlying this spatial heterogeneity in the response of the mouse lung to irradiation is unknown. Liao, Tucker, and Travis19-21 proposed that the heteroge-­ neous distribution of functional units in the lung is a func-­ tion of pulmonary anatomy; in other words, larger portions of the apex and middle regions of the lung are occupied by major conducting airways and fewer gas exchange units than the base of the lung, where the ratio of gas exchange units to conducting airways would be higher. Because the conducting airways are not involved in gas exchange, more of the critical functional units are in the base than in the apex of the lung. A given dose of radiation has less effect on the same volume irradiated in the apex than in the base of the lung simply because more of the structures critical to lung function are present in this area than in others. It has been suggested that regional differences in the incidence of pneumonitis could be accounted for by differ-­ ences in the functionality of cells in the apex and the base

428

PART 5  Thorax 24

LD50 (Gy)

20

Probability of pneumonitis (%)

100

Apex

16

Base

80 60 40 20 0

12

0

20

40

60

80

100

Size of irradiated subvolume (%) 20

40

60

80

100

Volume irradiated (%)

FIGURE 19-5.  Lethality as measured by the dose causing 50% of experimental subjects to die (LD50) from radiation pneumonitis is plotted as a function of partial volume irradiated, showing a clear location-dependent response for all partial volumes irradiated. (From Travis EL, Liao ZX, Tucker SL. Spatial heterogeneity of the volume effect for radiation pneumonitis in mouse lung. Int J Radiat Oncol Biol Phys 1997;38:1045-1054.)

of the lung.23 In a series of experiments designed to deter-­ mine whether factors released by cells in one part of the lung influence the effect in another, Khan and colleagues24 found that irradiated fibroblasts in the base of the lung sus-­ tained more DNA damage than did irradiated fibroblasts residing in the apex of the lung and that this effect was sustained independent of which region was irradiated. This effect disappeared when the entire lung was irradi-­ ated. Although the mechanism of the regional differences in radiation response remains unclear, findings from two independent laboratories have shown that the morbidity of lung irradiation is related to the volume of lung irradiated and to the location of this irradiated subvolume in the lung. The two groups agree that the base is the subvolume most sensitive to radiation. What remains unanswered by these experiments, how-­ ever, is the question of whether it is better to irradiate a large volume with a small dose or a small volume with a large dose. Using the findings from the mouse studies, Tucker and colleagues20 calculated the incidence of pneu-­ monitis for subvolumes of the lung irradiated with a fixed dose. As shown in Figure 19-6, the incidence of complica-­ tions varied from 100% after irradiation of the base to 0% after irradiation of the middle region of the lung, with an intermediate number after irradiation of the apex.20 These findings indicate that relatively small changes in the loca-­ tion of the irradiated subvolume in the lung correspond to significantly greater changes in the incidence of pneumoni-­ tis than do moderate changes in the dose or the size of the irradiated subvolume. On the basis of these data, Tucker and colleagues20 found that the risk of pneumonitis after a constant integrated dose of 12.7 Gy varied enormously depending on the site of the irradiated subvolume. This analysis of a constant integrated dose (calculated as the product of volume and dose) dem-­ onstrated that for the apex and the middle region of the

FIGURE 19-6.  Predicted incidence of pneumonitis in mouse lung as a function of irradiated volume when the integrated dose (dose × volume) is fixed at 12.7 Gy, the single-dose LD50. The upper, middle, and lower curves represent volumes in the base, apex, and middle regions of the lung, respectively. In the base, the risk of pneumonitis is greatest after irradiation of an intermediate subvolume with an intermediate dose, whereas in the apex, a little radiation to a lot of lung usually is more damaging than a lot to a little. (From Tucker SL, Liao ZX, Travis EL. Spatial heterogeneity of the volume effect for radiation pneumonitis in mouse lung. Int J Radiat Oncol Biol Phys 1997;38: 1055-1066.)

lung, a little radiation to a large volume of lung was worse than a lot to a little (i.e., the tails of both curves increased at large volumes). However, in the base of the lung, the peak probability of pneumonitis occurred between irra-­ diated volumes of 50% and 70%; the probability fell off dramatically for volumes outside this range. Although this seems counterintuitive for the large volumes, for which the probability of pneumonitis would be expected to increase with the volume of lung irradiated, this decrease actually reflects use of the constant integrated dose. As the volume gets larger, the dose necessarily becomes smaller to main-­ tain constancy. A third factor that can be defined from these curves is the volume below which radiation is expected to cause no detectable injury, even after very high doses (i.e., threshold volume). Like the radiation doses associated with lethality and complications, the threshold volume depends on the site of irradiation, ranging from 10% in the more sensitive base of the lung to 30% to 40% in the more resistant ­middle region and apex. Regardless of the parameter ­ examined, these studies indicate that the site of the irradiated volume within the lung is an important consideration when choos-­ ing a treatment plan and that the base of the lung is more sensitive than other regions. Similar findings have been reported in humans. Yamada and colleagues25 reported that the risk of pneumonitis in patients varied with the location of the irradiated volume. They also found that inclusion of volumes in the lower lung increased the incidence of pneumonitis to 70% from 20% when sites outside the lower lung were included (P = .01).25 Multivariate analysis revealed a significant ­relationship between radiation site and the risk of pneu-­ monitis (P = .0096). These patient data are consistent with the mouse data and indicate that the risk of pneumonitis

19  The Lung, Pleura, and Thymus

is significantly higher when irradiated volumes are in the lower part of the lung than when they are in the upper part of the lung. The site of the irradiated subvolume must be a consideration in treatment planning for all patients with lung cancer.

Clinical Changes The clinical symptoms related to the effects of irradiation on the tracheobronchial tree rarely dominate the picture. Excess sputum production may be relieved by agents that shrink the mucosa or reduce secretions; dryness and non-­ productive cough may be helped by cool-mist humidifiers. The clinical manifestations of radiation effects on the lung can be separated into an early degenerative “pneu-­ monitic” phase and a later regenerative phase of scarring or fibrosis. During the degenerative phase, dyspnea and coughing that occasionally produce thick, white sputum may develop. Patients may have fever and night sweats. Percussion and auscultation during this phase are often remarkably unrevealing. Usually, the symptoms subside in 2 to 3 months, and no further symptoms are noticed, ­despite abnormal radiographic findings. If the volume of lung irradiated to a dose above 25 to 30 Gy is very large, the clinical signs and symptoms of radiation pneumonitis ­appear in 3 to 6 weeks. The patient may become progres-­ sively more dyspneic at rest, febrile, and develop night sweats and cyanosis. Alveolar-capillary block is the most disabling physiologic change. Symptoms of cough, dyspnea, or fever may range from mild to extreme, depending on the volume of lung irradiated and the dose-fractionation regimen. With careful attention to keeping the high-dose

A

429

volume as small as possible, symptoms from pulmonary ­irradiation are the exception rather than the rule. If the volume of lung irradiated has been extensive, ­permanent scarring that produces chronic debilitation because of respiratory compromise may develop. Dys-­ pnea and cough may be severe. Clubbing of the fingers may ­ develop. Secondary infection with abscess forma-­ tion may lead to death from sepsis. Pulmonary hyperten-­ sion with right-sided heart failure has been reported.26,27 ­Arteriovenous shunting may be a major cause of dyspnea and cyanosis; pneumonectomy has rarely proved beneficial in severe cases.28

Radiologic Changes Often the only clinical manifestations of radiation pneu-­ monitis are radiographic changes: infiltrates may become evident on chest images conforming to the treatment field (Fig. 19-7). In the past, when most treatment fields were square or rectangular, diagnostic radiologists recognized “square pneumonia” as a hallmark of radiation pneumoni-­ tis. The phenomenon is now much more subtle because of the use of individualized custom blocks that have ­irregular, rounded margins defining the treatment portal. The ­changes remain confined to the treated volume29 unless some other process supervenes; no dosimetric or pathophysiologic ba-­ sis exists for radiation pneumonitis to occur in the normal lung outside the treated volume. Several conditions, such as bacterial and viral pneumo-­ nias, fungal and protozoal infections, and lymphangitic car-­ cinomatosis, can give rise to diffuse pulmonary infiltrates. When such infiltrates follow a course of radiation therapy,

B

D C FIGURE 19-7.  A, Posteroanterior chest radiograph and B, corresponding CT scan of a patient with lung cancer. C, Chest radiograph  6 months after radiation therapy shows radiation pneumonitis and indeterminate findings relative to the tumor. D, CT scan shows that the tumor is no longer evident and linear infiltrates correspond to the radiation.

PART 5  Thorax

they may be incorrectly attributed to radiation pneumo-­ nitis. Adult respiratory distress syndrome may sporadically follow thoracic irradiation30; diffuse bilateral infiltrates are not confined to the irradiated volume, and rapidly progres-­ sive respiratory failure with unresponsive hypoxemia is seen. Although high-dose corticosteroids are often admin-­ istered, they probably do not affect outcome,31 and they may have catastrophic effects if the infiltrates are the result of supervening infection. The relationship of the distribution of the pulmonary infiltrates to the fields of irradiation can be difficult to as-­ sess with standard posteroanterior and lateral radiographs. Computed tomography (CT) of the thorax can be help-­ ful for this purpose; if the infiltrates seen on CT scans do not correlate with the irradiated volume, and specifically if infiltrates appear in portions of the lungs that received no irradiation, radiation pneumonitis can be ruled out and the actual cause can be sought (see Fig. 19-7). The advantages of giving chemotherapy ­ concurrently with radiation therapy—no delay in initiating either mo-­ dality, shorter overall treatment time, and potential sensi-­ tization of the tumor—are paired with the ­ disadvantages of increased toxicity and the inability to assess the ef-­ fects of each modality separately. A typical toxic effect of concurrent radiation and chemotherapy is grade 3 pneu­ monitis, which is symptomatic, interferes with activities of daily living, and can require administration of oxygen. In a retrospective review of 223 patients with non–small cell lung cancer (NSCLC) treated with three-­dimensional conformal radiation therapy (3D CRT), Wang and col-­ leagues32 found that grade 3 or higher treatment-related pneumonitis affected 22% of patients at 6 months and 32% at 1 year. Univariate analyses suggested that lung volume, gross tumor volume, mean lung dose, and rV5 (the rela-­ tive volume of lung receiving 5 Gy) through rV65 values, in 5-Gy increments, were linked with grade 3 or higher treatment-related pneumonitis. Multivariate regression, however, revealed that an rV5 cut point was the only fac-­ tor significantly related to grade 3 or higher treatment­related pneumonitis (1-year actuarial incidence was 3% for rV5 ≤ 42% and 38% for rV5 > 42%, P = .001). The recom-­ mendations from this analysis were that physicians consider all of these factors when considering concurrent radiation and chemotherapy for lung disease, inasmuch as all dose levels were highly correlated with each other.32 In another retrospective study, Yom and colleagues33 evaluated concurrent combinations of chemotherapy and radiation therapy for patients with NSCLC and found that combining the chemotherapy with IMRT (68 patients) re-­ sulted in fewer cases of high-grade pneumonitis than did combining concurrent chemotherapy with 3D CRT (222 patients) (Fig. 19-8).33 Twenty patients in the intensitymodulated group (29%) and 131 (59%) from the 3D CRT group had had induction chemotherapy (P < .0001). Over time, 26 patients (38%) in the IMRT group received adju-­ vant chemotherapy, and 6 (9%) who had disease progres-­ sion also had further chemotherapy. Overall, 50 (74%) of the IMRT group and 154 (69%) of the 3D CRT group had received concurrent chemotherapy with carboplatin and a taxane. The median follow-up times were 8 months (range, 0 to 27 months) for the IMRT group and 9 months (range, 0 to 56 months) for the 3D CRT group. The rate of radiation pneumonitis at 12 months was significantly

lower among those who had IMRT (9%) than among those given 3D CRT (32%) (P = .002).33 These findings showed that IMRT techniques were beneficial in terms of reducing the incidence of high-grade pneumonitis among patients treated with concurrent radiation and chemotherapy for stage III NSCLC.

Functional Effects The pathophysiology of radiation pneumonitis and scarring and the associated functional effects are related to vascular, epithelial, and interstitial injury. It is often difficult to deter-­ mine how much of a given functional outcome is attribut-­ able to vascular abnormality and how much is attributable to epithelial and interstitial disease. Carbon monoxide diffus-­ ing capacity decreases strikingly, as would be expected with hyaline membrane formation, interstitial fibrosis of septa, and thrombosis or progressive vascular sclerosis. Compli-­ ance is markedly decreased by interstitial sclerosis.34,35 It follows that lung volumes are diminished. Poor ventilation of the irradiated segment and incomplete oxygenation of the blood contribute to the patient’s disability in propor-­ tion to the ratio of damage to the normal lung. Choi and associates36 found that the functional changes produced by irradiation for cancer of the lung vary accord-­ ing to pulmonary reserve before treatment. Paradoxically, patients with less functional compromise before radia-­ tion therapy (e.g., a forced expiratory volume in less than 1 second [FEV1] of more than 50% of the predicted value) lost more pulmonary function and showed higher airway resistance after radiation therapy than did patients with less pretreatment reserve. Those with an FEV1 of less than 50% of the predicted value before therapy had improved IMRT (N = 68) vs 3D (N = 222) Freedom from grade >= 3TRP

430

1.00

IMRT

p = 0.002

0.75 3D 0.50 0.25 0.00 0

12

24

36

48

60

Months from start of RT

FIGURE 19-8.  Intensity-modulated radiation therapy (IMRT) can reduce the incidence of treatment-related pneumonitis in patients being treated for non–small cell lung cancer. The time to the development of grade 3 or higher treatment-­related pneumonitis (TRP) is shown for 68 patients who underwent concurrent chemotherapy and IMRT and for 222 patients who underwent concurrent chemotherapy with three-dimensional conformal radiation therapy (3D). At 12 months, 8% of the IMRT group (95% CI, 4% to 19%) and 32% of the 3D group (95% CI, 26% to 40%) had pneumonitis of grade 3 or higher  (P = .002). (From Yom SS, Liao Z, Liu H, et al. Initial evaluation of treatment-related pneumonitis in advanced-stage non-small-cell lung cancer patients treated with concurrent chemotherapy and intensity-modulated radiotherapy. Int J Radiat Oncol Biol Phys 2007;68:94-102.)

19  The Lung, Pleura, and Thymus

or minimally reduced pulmonary function after therapy. A study by Brady and colleagues37 confirmed the relative safety of giving effective doses to modest treatment vol-­ umes and the hazards of giving high doses to large treat-­ ment volumes. The dose-escalation studies of the Radiation Therapy Oncology Group (RTOG) have confirmed the much greater importance of the volume irradiated than the dose given when the radiation is in the therapeutic range (>50 Gy); in their studies, increasing the total doses from 60 to 74.4 Gy did not increase the frequency of acute or late pulmonary toxicity.38-40 Advances in imaging technology are extending the tools with which to study lung function, with the ultimate goal of manipulating radiation therapy fields to avoid well­ventilated lung tissue. Guerrero and others41 reported the use of high-resolution CT for quantifying lung ventilation in a murine model of lung cancer and found that direct assessment of pulmonary compliance with high-resolution CT demonstrated a 60% reduction in compliance in irra-­ diated mice as compared with nonirradiated control mice (Fig. 19-9).41

A

Another adverse consequence of concurrent radiation and chemotherapy for lung cancer is acute esophagitis, which can be sufficiently debilitating and painful as to prompt interruption of therapy. To identify risk factors for acute esophagitis, Wei and colleagues42 retrospec-­ tively studied 215 patients with NSCLC who underwent 3D CRT (median dose of 63.5 Gy in 35 fractions) and ­concurrent chemotherapy. More than one half of these patients (127) had also had induction chemotherapy. A total of 93% of patients experienced acute esophagi-­ tis during and up to 1 month after therapy, and 20.5% had cases considered grade 3 or higher. Univariate anal-­ ysis suggested that three dosimetric factors—the mean ­radiation dose to the esophagus, the absolute esophageal volume treated to 15 to 45 Gy (aV15 to aV45), and the relative esophageal volume treated to 10 to 45 Gy (rV10 to rV45)—were related to acute esophagitis of grade 3 or higher. However, in multivariate analysis, only one factor—rV20—remained a significant predictor of acute esophagitis.42 These findings confirm that the most im-­ portant factor in preventing acute severe esophagitis in

B

C 0.0

D

E

431

0.0

0.3

0.3

0.6

0.6

F

0.0

0.3

0.6

0.0

0.3

0.6

FIGURE 19-9.  Pulmonary compliance coronal section CT images for one control mouse (A-C) and one irradiated mouse (D-F) for 6 and 14 cm H2O of pressure. The first image of each row, referred to as the compliance image at 6 cm H2O, is from two CT images, one obtained at 2 cm H2O and the other at 10 cm H2O. Each is from two constant-pressure images with an 8-cm H2O pressure difference. Each records the fractional increase in volume of each voxel between the two parent images. At center on each row is the compliance image at 6 cm H2O of pressure, and at right on each two is the compliance image at 14 cm H2O of pressure. The second and third images on each row employ the National Institutes of Health color scale (i.e., 0 to 0.6 scale underneath the compliance images). Investigators calculated compliance from the values in the images multiplied by the voxel volume and divided by the pressure difference (two constant values). (From Guerrero T, Castillo R, Noyola-Martinez J, et al. Reduction of pulmonary compliance found with high-resolution computed tomography in irradiated mice. Int J Radiat Oncol Biol Phys 2007;67:879-887.)

432

PART 5  Thorax

patients treated with concurrent radiation therapy and chemotherapy is to limit the relative volume of ­esophagus treated to 20 Gy.

Treatment of Radiation Pneumonitis Radiation pneumonitis is an inevitable consequence of ­aggressive therapy for malignant conditions that arise in the thorax. In most patients, radiographic manifestations of the changes are the first and only evidence of this condi-­ tion. Usually radiographic opacity decreases or disappears without symptoms or treatment. When symptoms are sufficiently severe to justify treat-­ ment, corticosteroids may help. About one half of patients so treated experience prompt, marked relief.26,35 Cortico-­ steroids may reverse the symptoms of radiation pneumonitis within hours, but they do not prevent or reverse the fibrotic phase. They may be contraindicated in disease processes in which they have been shown to have a deleterious effect on outcome, such as carcinoma of the lung.43 It is important in cases of radiation pneumonitis to withdraw steroids very slowly to avoid recrudescence. Unless the symptoms are se-­ vere, it is preferable to provide supportive care with bron-­ chodilators, expectorants, antibiotics, bed rest, and oxygen and to avoid the long-term use of corticosteroids. Prophylactic administration of cortisone to animals irra-­ diated to the entire thorax has been shown to decrease the frequency of severe acute reactions.35 However, prophylactic

administration of steroids to humans has not been tested. The best prophylaxis is careful treatment planning that minimizes the volume irradiated to high doses. Komaki and colleagues44 found that amifostine signifi-­ cantly reduced severe pneumonitis after combined irra-­ diation and chemotherapy for inoperable, nonmetastatic NSCLC; their results also dispelled the notion of amifos-­ tine’s having a tumor protective effect. Five (16%) of the 31 patients who did not receive amifostine (experienced severe pneumonitis, whereas none of the 31 patients given amifostine (500 mg intravenously before any treatment on days 1 and 2 of any treatment week) had it (P = .02). At a median follow-up time of 31 months for living patients, the median survival time differed between the groups by only 1 month. Esophageal toxicity was less severe among those given amifostine (P = .021), and neutropenic fever was less common among those given amifostine (16% versus 39%, P = .046) during combined radiation and chemotherapy. However, amifostine was also associated with an increased frequency of mild hypotension, dysgeusia, and sneezing.44 Amifostine may be particularly helpful for patients who are likely to experience severe pulmonary complications because of irradiation of large tumors and surrounding nor-­ mal lung tissue. An example of such a case, a 54-year-old man with stage IIIB (T3 N3 M0) squamous cell carcinoma of the left lower lobe with left hilar and right paratracheal nodal involvement, is illustrated in Figure 19-10.45 Even

Norm. volume

A 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0

CTV PTV Esophagus Heart Total lung

0

B

1000 2000 3000 4000 5000 6000 7000 8000 Dose (cGy)

FIGURE 19-10.  Isodose distributions (A) and dose-volume histogram (B) for an intensity-modulated radiation therapy plan for a 54-year-old man with non–small cell lung cancer who experienced severe pulmonary complications after therapy. The broad scope of disease, from the apex of the right lung to the lower lobe of the left lung plus mediastinal involvement, meant exposing a large volume of normal lung to radiation, which probably contributed to the onset of complications. CTV, clinical target volume; PTV, planning target volume. (From Liu HH. The physics aspects of intensity-modulated radiotherapy for lung cancers. In Cox JD, Chang JY, Komaki R [eds]: Image-guided Radiotherapy for Lung Cancer. New York: Informa Health Care, 2008, pp 103-126.)

19  The Lung, Pleura, and Thymus

433

TABLE 19-1.    Dose-Volume Constraints for Radiation Therapy Alone or in Combination   with Chemotherapy or Surgery Organ or   Structure

  Radiation Therapy Alone

  Radiation-Chemotherapy

Radiation-Chemotherapy   plus Surgery

Spinal cord

50 Gy

45 Gy

45 Gy

Lung

20 Gy

Heart

(<40%)

40 Gy

(<100%)

50 Gy

(<50%)

Esophagus

60 Gy

(<50%)

Liver

30 Gy

(<40%)

Kidney

20 Gy

(<50% or 25%)*

20 Gy

(<35%)

40 Gy

(<50%)

55 Gy

(<50%)

20 Gy

(<20%)

15 Gy

(<30%)

10 Gy

(<40%)

40 Gy

(<50%)

*When

20 Gy is used, volume constraints are less than 50% when both kidneys are exposed and less than 25% when one kidney is exposed and the other kidney is not functioning. Data courtesy of Xiong Wei, MD, Department of Radiation Oncology, The University of Texas M. D. Anderson Cancer Center, 2007.

with an IMRT plan, the lung V20 was 40%, which ex-­ ceeds the dose-volume constraint of V20 less than 35% for ­concurrent radiation and chemotherapy (Table 19-1). The esophageal V20 (see Fig. 19-10),45 at 78%, also exceeded the dose-volume limit of 50% (see Table 19-1). This ­patient might have benefited from a course of subcutaneous ami-­ fostine, given as 500-mg subcutaneous doses daily 1 hour before each radiation-chemotherapy session for 7 weeks, to prevent acute pneumonitis and esophagitis.

Radiation Therapy for Lung Carcinoma Carcinoma of the lung is the most common malignant dis-­ ease in the United States and in several countries in north-­ ern and western Europe. It accounts for one of every six new cancer cases diagnosed in the United States, and it is responsible for one of every four cancer deaths. The striking rise in incidence and the lack of any major improvement in treatment have led to a steep rise in annual mortality dur-­ ing the second half of the 20th century. Since that time, however, the incidence among men in the United States has declined from 102 cases per 100,000 in 1984 to 78.5 cases per 100,000 in 2003, and the incidence among U.S. women is approaching a plateau.46,47 Decreases in smok-­ ing during this period have been reflected in decreased death rates among men and a near plateau in death rates for women. Nevertheless, lung cancer remains the most common cause of cancer death among U.S. men and has surpassed cancer of the breast as the number one cause of cancer death among U.S. women. The single most important cause of the lung cancer is the inhalation of tobacco smoke. Many other agents also have been shown to increase the risk of bronchopulmonary cancer, including asbestos, iron ore, radioactive ores, and isopropyl oil.48 The mainstay of treatment for patients with small, ­localized carcinomas of the lung continues to be surgical resection. With current imaging procedures, augmented

by cervical mediastinoscopy, anterior mediastinotomy, and transtracheal or transbronchial biopsy, only 20% to 25% of all patients are considered candidates for definitive resec-­ tion. The remaining patients, except for those with distant metastases, are appropriately treated with radiation alone or, increasingly, with cytotoxic drugs. The choice of therapy is largely based on the histo-­ pathologic type of lung cancer. One classification origi-­ nally proposed by the World Health Organization and later modified49 consists of squamous cell carcinoma, small cell carcinoma, adenocarcinoma, large cell carcinoma, and ­adenosquamous carcinoma. Squamous cell carcinoma is the single most common histopathologic type, but adenocarcinoma is only slightly less common.50 Each constitutes somewhat less than one third of all cases of carcinoma of the lung. Included within adenocarcinoma is bronchioal-­ veolar carcinoma, an uncommon lung cancer but one of the few types that is not associated with tobacco inhala-­ tion. Squamous cell carcinomas and adenocarcinomas are subject to histologic grading, but most are moderately to poorly differentiated. Small cell carcinoma includes the lymphocyte-like type also known as oat cell carcinoma. A variant of this, previ-­ ously thought to represent an intermediate cell type within the small cell grouping, is now recognized as being a mix-­ ture of small cell and large cell carcinoma. Large cell carcinoma is quite possibly the most undifferentiated form of ­ adenocarcinoma, and as such large cell carcinoma and adenocarcinoma have been indistinguishable in a variety of clinicopathologic studies. Giant cell carcinoma and clear cell carcinoma are considered variants of large cell ­carcinoma. Carcinoid tumors51 are relatively indolent but unques-­ tionably malignant tumors that tend to occur in younger individuals and are unrelated to the inhalation of tobacco smoke. This type of tumor shows neuroendocrine differen-­ tiation, and its relationship to small cell carcinomas is con-­ troversial. Carcinoid tumors are parabronchial tumors that often have a small portion extending into the lumen of the bronchus. They occasionally produce vasoactive substances that can cause flushing, diarrhea, and hypotension.

434

PART 5  Thorax

TABLE 19-2.    Cause of Death by Cell Type Number of Patients (%) Dying of the Cancer* Cause of Death

Squamous Cell Carcinoma

Small Cell ­ arcinoma C

Adenocarcinoma

Large Cell ­ arcinoma C

Combined

Carcinomatosis

21 (25)

49 (70)

49 (48)

23 (47)

2

9 (9)

4 (8)

—

Central nervous system metastasis Infection

2 (2)

—

30 (36)

6 (11)

19 (19)

13 (27)

6

Hemorrhage

7 (8)

4 (7)

4 (4)

1 (2)

—

Respiratory failure

5 (6)

2 (3)

7 (7)

4 (8)

—

Heart failure

17 (20)

5 (9)

8 (8)

3 (6)

—

Pulmonary emboli

—

—

4 (4)

1 (2)

—

—

2 (2)

Other cancer

1 (1)

—

—

*Samples

taken from 300 consecutive Veterans Administration Lung Group autopsies. From Cox JD, Yesner RA. Causes of treatment failure and death in carcinoma of the lung. Yale J Biol Med 1981;54:201-207.

Squamous cell carcinoma is the most likely to remain confined to the thorax, whereas almost every patient with small cell carcinoma will have distant metastasis. Studies of causes of death (Table 19-2)52 and patterns of failure show that squamous cell carcinoma most often kills by lo-­ cal tumor progression, whereas small cell carcinoma tends to cause death from widespread dissemination. Adenocar-­ cinoma and large cell carcinoma occupy an intermediate place in the spectrum from squamous cell carcinoma to small cell carcinoma; adenocarcinoma and large cell carci-­ noma are less likely to cause death from intrathoracic tu-­ mor progression than is squamous cell carcinoma, but they are more likely to spread beyond the thorax. The lymphatic drainage of the lungs is shown in Fig-­ ure 19-11.53,54 In surgical series, one fourth to one half of patients who undergo resection have regional lymph node metastasis (Table 19-3).55-57 At autopsy, more than 90% of patients are found to have involvement of hilar or mediasti-­ nal lymph nodes. However, carcinoma of the lung does not spread in such a predictable manner. Mediastinal lymph nodes and blood vessels are often involved by the time the diagnosis is established. The highly vascular composition of the lung, with its dual circulation of bronchial and pulmo-­ nary vessels, provides a ready anatomic explanation for the high frequency of extrathoracic dissemination. Small cell carcinoma, adenocarcinoma, and large cell carcinoma have a notable propensity to spread to the brain (Table 19-4).50 Even with the relatively high rate of sur-­ vival among patients with small cell carcinoma (because of ­effective combinations of chemotherapy), 50% to 80% of patients eventually have metastasis to the brain.58-60 Clini-­ cal assessments of the frequency of brain metastasis from adenocarcinoma and large cell carcinoma have confirmed the high rate of involvement found in autopsy ­studies.60 Carcinoma of the lung is most often diagnosed from symptoms resulting from the intrathoracic tumor. Cough, dyspnea, chest pain, and hemoptysis are among the most frequent symptoms and signs. However, a large proportion of patients have unexplained weight loss. Radiographic studies reveal that squamous cell carci-­ noma and small cell carcinoma tend to occur as central

11R

10R

7

4

1

11L

10L

5

2

SC

FIGURE 19-11.  Lymphatic drainage of the lobes of the lungs related to the schema commonly used for classifying regional lymph nodes. Notice the crossover from the lingula of the left upper lobe and the superior portion of the left lower lobe to the right paratracheal nodes. Lymph node drainage does not parallel arteries or veins, nor does it remain distinct for each lobe. Node colors correspond to stations in the AJCC Cancer Staging Manual. (Modified from Nohl HC. An investigation into the lymphatic and vascular spread of carcinoma of the bronchus. Thorax 1956;11:172-185; Greene FL, Page DL, Fleming ID, et al [eds]. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 167-177.)

lesions (Table 19-5),61 whereas adenocarcinoma and large cell carcinoma occur more often as peripheral parenchymal tumors. Small cell carcinoma and large cell carcinoma are often associated with mediastinal abnormalities. Any of the histopathologic types of lung carcinoma may occur in the

19  The Lung, Pleura, and Thymus

435

TABLE 19-3.    Distribution of Mediastinal and Supraclavicular Lymph Node Metastases by Site of Primary Carcinoma of the Lung Diseased Nodes In The Region (%) Lobe

Subcarinal

Pararight Tracheal

Supraright Clavicular

Paraleft Tracheal

Right upper

32

64

30

 4

 1

Right lower

46

42

 3

12

—

Left upper

42

16

 3

42

39

Left lower

50

21

 4

29

 4

Supraleft Clavicular

Data from Carlens E. Appraisal of choice and results of treatment for bronchogenic carcinoma. Chest 1974;65:442-445; Goldberg EM, Glickman AS, Kahan FR. Mediastinoscopy for assessing mediastinal spread in clinical staging of carcinoma of the lung. Cancer 1970;25: 347-353; Palumbo LT, Sharpe WS. Correlation with other diagnostic procedures in 550 cases Arch Surg 1969;98:90-93.

TABLE 19-4.    Frequency of Brain Metastasis at Autopsy Primary Tumor ­  (Histopathologic Type)

Number of Patients

Patients with Metastasis ­Including the Brain (%)

Patients with Brain ­Metastasis Only (%)

Squamous cell carcinoma

123

13

 4

Small cell carcinoma

  82

45

 6

Adenocarcinoma

129

54

12

Large cell carcinoma

  54

52

 9

Combined

  12

25

 0

totals

400

38

 8

Modified from Cox JD, Yesner RA. Adenocarcinoma of the lung: recent results from the Veterans Administration Lung Group. Am Rev Respir Dis 1979;120:1025-1029.

TABLE 19-5.    Common Abnormalities Observed in Roentgenograms of Various Types of Lung Carcinoma Proportion of Cases with the Abnormality (%) Mass Characteristics

Squamous Cell Carcinoma (n = 263)

Small Cell Carcinoma (n = 114)

Adenocarcinoma (n = 126)

Large Cell Carcinoma (n = 97)

Hilar or perihilar mass

40

78

17

32

≤4 cm

 9

21

45

18

Parenchymal mass ≥4 cm

19

 8

26

41

Apical mass

 3

 3

 1

 4

Consolidation or collapse

53

38

25

33

Mediastinal mass

 1

13

 2

10

Pleural effusion

 3

 5

 1

 2

Modified from Byrd RB, Carr DT, Miller WE. Radiographic abnormalities in carcinoma of the lung as related to histological cell type. Thorax 1969;24:573-575.

extreme apex of either lung, with symptoms of shoulder pain, ulnar dysesthesias, Horner’s syndrome, and evidence of chest wall invasion and destruction of the adjacent ribs or vertebrae (i.e., Pancoast syndrome). Obstruction of the superior vena cava can lead to swelling of the face, neck, and arms and evidence of collateral circulation over the anterior chest wall; approximately 4% of all patients with cancer of the lung present with this syndrome. Paraneoplastic phenomena include clubbing, ectopic hormone production (e.g., adrenocorticotropic hormone,

calcitonin, human chorionic gonadotropin, antidiuretic hormone, or parathyroid hormone), and a variety of neu-­ romuscular syndromes. These may disappear with effective treatment of the primary tumor. All but two of the para-­ neoplastic phenomena are seen more often in patients with small cell carcinoma than in patients with other types of lung cancer—hypercalcemia and clubbing are rare in small cell carcinoma. Patients who have cancer of the lung should undergo a careful physical examination and selected laboratory and

436

PART 5  Thorax

imaging studies. A complete blood count is always appro-­ priate, because anemia or a leukoerythroblastic picture may suggest bone marrow involvement. A biochemical survey, including liver function tests such as alkaline phos-­ phatase and aspartate aminotransferase, can be conducted to help detect hepatic involvement. Carcinoembryonic antigen and lactate dehydrogenase levels may be elevated and may serve as a means of monitoring the effectiveness of therapy.62 In addition to plain posteroanterior and lateral chest radiography, CT of the thorax contributes to an apprecia-­ tion of the extent of the primary tumor and regional lymph node metastasis,63 and it has an important function in treatment planning if radiation therapy is used. CT of the upper abdomen can be invaluable in the search for occult distant metastasis. Hepatic metastasis can be identified, and the more common adrenal metastasis can be seen.64 Renal and retroperitoneal lymph node metastases are sometimes identified. CT of the brain is necessary in patients with small cell carcinoma, adenocarcinoma, and large cell carci-­ noma because as many as 10% of patients without neuro-­ logic symptoms or signs may have occult brain metastasis. Whole-body positron emission tomography (PET) with 18F-fluorodeoxyglucose can improve the rate of detection of regional and distant metastasis in patients with NSCLC. Pieterman and colleagues65 found PET to be more sensitive than CT (91% versus 75%) and more specific than CT (86% versus 66%). In that study, PET identified distant metasta-­ ses in 11 of 102 patients considered to be free of metastasis by all other pretreatment staging methods. Although PET correctly detected mediastinal lymph node metastasis in 90% of patients with histopathologic evidence of involved lymph nodes, it gave false-positive results for lymph nodes in patients with reactive hyperplasia and silicoanthraco-­ sis. PET is complementary to CT for radiation treatment planning.66 When the complete clinical evaluation is ac-­ complished, more than one half of all patients with lung carcinoma are found to have distant metastasis. Clinical staging classifications for carcinomas of the lung have been used widely for more than 20 years. The cur-­ rent version of the system advocated by the American Joint Committee on Cancer (AJCC)54 is shown in Box 19-1. Further illustration of a schema that represents a combina-­ tion of two previous classification systems for lymph node involvement is given in Figure 19-1267 and in Box 19-2.67 This combined system was generated to solve problems caused by inconsistent staging; it incorporates anatomic landmarks to identify N1 and N2 nodes.

Indications for Radiation Therapy There are many indications for radiation therapy for cancer of the lung. Radiation therapy can be used as an adjunct to surgery, in place of surgery for patients with small tumors who cannot undergo surgery, with chemotherapy as a cure for unresectable but nonmetastatic tumors of all cell types, as prophylactic treatment of subclinical metastasis, and as palliative therapy for distressing symptoms. These indica-­ tions are described further in the following sections.

Surgical Adjuvant Radiation Therapy Radiation therapy is indicated only for selected patients with operable carcinoma of the lung. Prospective studies have never shown that preoperative irradiation improves results

over resection alone.68 However, preoperative ­ irradiation may have conferred no benefit in these studies because without surgery, physicians cannot identify patients who have the greatest risk of locoregional recurrence. ­Curiously, preoperative irradiation is still considered by many surgeons and radiation oncologists to be standard treatment in the management of apical sulcus (Pancoast) tumors.69 Postoperative irradiation does not seem to improve results in patients with no evidence of metastasis to the hilar or mediastinal lymph nodes.70 Postoperative irradia-­ tion may have a net deleterious effect on survival. Shields71 and members of the PORT Meta-analysis Trialists Group72 observed that local recurrence was uncommon in cases of pathologically complete resection and no evidence of regional lymph node metastasis, suggesting that irradia-­ tion would accomplish little for such patients. Moreover, radiation therapy may place an additional burden on pul-­ monary function, especially in patients who have under-­ gone ­ pneumonectomy. Findings from a meta-analysis72 reiter­ated the long-recognized potential for adverse effects of radiation therapy when rigorous CT-based treatment ­planning is not done. Several retrospective studies73-75 have suggested that postoperative irradiation can improve outcome when ­mediastinal nodal metastases are present. An analysis72 of postoperative radiation therapy in available prospec-­ tive ­trials, most of which was not based on CT planning, showed neither a detrimental nor a beneficial effect. On the basis of the striking improvement in local control of squamous cell carcinoma after postoperative radiation ther-­ apy shown by the Lung Cancer Study Group,76 the East-­ ern Cooperative Oncology Group (ECOG) coordinated a large intergroup trial to compare postoperative radiation therapy plus concurrent chemotherapy with cisplatin and etoposide to postoperative radiation therapy alone.77 Initial results for the 373 patients eligible for analysis at a median follow-up of 44 months indicated median survival times of 42 months for patients given only radiation and 38 months for those given combined-modality treatment (P = .48). No significant differences were found in failure patterns between the two study groups. Although postoperative ­irradiation can clearly reduce the risk of local recurrence in patients with mediastinal node metastasis,76,78 its effect on survival (with or without adjuvant chemotherapy) is still under ­investigation. Using a different approach, Sawyer and colleagues79 studied patients with stage IIIA (N2) NSCLC by using ­regression tree analysis to identify patients at low, inter-­ mediate, and high risk of local recurrence. Numbers of in-­ volved lymph nodes, locations of mediastinal lymph nodes relative to the location of the primary tumor, and T stage were used to identify intermediate-risk and high-risk groups. Excluding the low-risk group, postoperative radia-­ tion therapy conferred highly significant improvements in local recurrence-free and overall survival rates. These find-­ ings may well form the basis for future studies of the value of postoperative radiation and chemotherapy. Controversy continues regarding the most effective treatment for patients with marginally resectable NSCLC. A retrospective review of patients treated at The ­University of Texas M. D. Anderson Cancer Center80 confirmed the impression that patients with more favorable disease are selected for surgical intervention. Two small ­ prospective

19  The Lung, Pleura, and Thymus

437

BOX 19-1.  American Joint Committee on Cancer Staging System for Cancer of the Lung Primary Tumor (T) TX Primary tumor cannot be assessed, or tumor proven by the presence of malignant cells in sputum or bronchial ­washings but not visualized by imaging or bronchoscopy T0 No evidence of primary tumor Tis Carcinoma in situ T1 Tumor 3 cm or less in greatest dimension, surrounded by lung or visceral pleura, without bronchoscopic evidence of invasion more proximal than the lobar bronchus* (i.e., not in the main bronchus) T2 Tumor with any of the following features of size or extent: more than 3 cm in greatest dimension; involves main bronchus, 2 cm or more distal to the carina; invades the visceral pleura; associated with atelectasis or obstructive pneumonitis that extends to the hilar region but does not involve the entire lung T3 Tumor of any size that directly invades any of the following: chest wall (including superior sulcus tumors), ­diaphragm, mediastinal pleura, parietal pericardium; or tumor in the main bronchus less than 2 cm distal to the carina, but without involvement of the carina; or associated atelectasis or obstructive pneumonitis of the entire lung T4 Tumor of any size that invades any of the following: mediastinum, heart, great vessels, trachea, esophagus, vertebral body, carina; or separate tumor nodules in the same lobe; or tumor with malignant pleural effusion� Regional Lymph Nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Metastasis to ipsilateral peribronchial and/or ipsilateral hilar lymph nodes, and intrapulmonary nodes including involvement by direct extension of the primary tumor N2 Metastasis to ipsilateral mediastinal and/or subcarinal lymph node(s) N3 Metastasis to contralateral mediastinal, contralateral hilar, ipsilateral or contralateral scalene, or supraclavicular lymph node(s) Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis present� Stage Grouping Occult carcinoma Stage 0 Stage IA Stage IB Stage IIA Stage IIB Stage IIIA

Stage IIIB Stage IV

TX Tis T1 T2 T3 T1 T2 T3 T1 T2 T3 T3 Any T T4 Any T

N0 N0 N0 N0 N0 N1 N1 N0 N2 N2 N1 N2 N3 Any N Any N

M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M1

*The uncommon superficial tumor of any size with its invasive component limited to the bronchial wall, which may extend proximal to the main bronchus, is also classified T1. �Most pleural effusions associated with lung cancer are caused by tumor. However, there are a few patients in whom multiple cytopathologic examinations of pleural fluid are negative for tumor. In these cases, fluid is not bloody and is not an exudate. These patients may be further evaluated by videothoracoscopy and direct pleural biopsies. When these elements and clinical judgment dictate that the effusion is not related to the tumor, the effusion should be excluded as a staging element, and the patient should be staged T1, T2, or T3. �M1 includes separate tumor nodule(s) in a different lobe (ipsilateral or contralateral). From Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 167-177.

studies81,82 provided quite similar results, suggesting that induction chemotherapy improves survival in patients with resectable NSCLC. The potential advantages of che-­ motherapy concurrent with radiation therapy delivered preoperatively were studied by the Southwest Oncology Group (SWOG). Albain and colleagues83 reported the re-­ sults for 126 patients treated in a phase II trial of cisplatin and etoposide concurrent with radiation therapy (45 Gy at 1.8 Gy/fraction given 5 days/week). After an interval of 2 to 4 weeks, thoracotomy and resection were ­attempted.

A high resection rate (76%) and a 3-year survival rate of 27% were reported, and the outcome was the same for ­patients with stage IIIB and IIIA disease. These findings led to the initiation of a randomized comparative trial ­coordinated by the RTOG to determine the role of resec-­ tion compared with chemotherapy and radiation therapy in ­patients with confirmed mediastinal node metastasis. In that study, ­ randomization was performed before patients were enrolled, and patients were assigned to undergo in-­ duction chemotherapy and radiation therapy for 5 weeks

438

PART 5  Thorax

Brachiocephalic (innominate) artery

2R Ao

4R

Azygos vein

4L 10R

PA 7

11R

11L 10L

8

9

12, 13, 14R

12, 13, 14L Inferior pulmonary ligament

3 Ligamentum arteriosum 6 Phrenic nerve Ao

5 PA

Left pulmonary artery

FIGURE 19-12.  Regional lymph node stations for lung cancer staging. Nodes are divided into four groups: superior ­mediastinal, aortic, inferior mediastinal, and N1. The superior mediastinal nodes (top) include the (1) highest mediastinal nodes (red), (2) upper paratracheal nodes (blue), (3) prevascular and retrotracheal nodes (inset, red), and (4) lower paratracheal (including azygos) nodes (orange). The aortic nodes (bottom) include the (5) subaortic nodes (anteroposterior window, dark green) and (6) para-aortic (ascending aorta or phrenic) nodes (purple). The inferior mediastinal nodes include the (7) subcarinal nodes (green), (8) paraesophageal nodes (below the carina, gray), and (9) pulmonary ligament nodes (brown). The N1 nodes include the (10) hilar nodes (yellow), (11) interlobar nodes (tan), (12) lobar nodes (pink), (13) segmental nodes (pink), and subsegmental nodes (pink). Box 19-2 provides lymph node ­definitions. Ao, aortic; L, left; PA, pulmonary artery; R, right. (From Mountain CF, Dresler CM. Regional lymph node classification for lung cancer staging. Chest 1997;111:1718-1723.)

and then thoracotomy and resection or a continuation of chemoradiation to a total dose of 61.0 Gy in 33 fractions. The study was closed to enrollment in 2001.84

Curative Therapy for Inoperable Non–Small Cell Lung Carcinoma Curative treatment for inoperable NSCLC was considered so uncommon that attempts at such treatment were not al-­ lowed in many developed countries seeking to limit health care resources. However, cure is an increasing reality in clinical investigations and standard practice. Two distinct

groups of patients can be identified for such treatment— those with stage I or II tumors who cannot undergo surgery because of comorbid medical conditions and those with more advanced, technically unresectable tumors. The ability of definitive radiation therapy to control small tumors and provide a cure for patients who cannot undergo surgery is widely recognized. At least 10% of ­patients with stage I NSCLC cannot undergo surgery for medical rea-­ sons. In 1996, Dosoretz and colleagues85 reviewed ­several series showing that 5-year survival rates ranged from 6% to 32% after radiation therapy. Findings from four subse-­ quent series of patients treated with radiation are shown in Table 19-6.86-89 Although the 2-year survival rates in the later series ranged from 34% to 46%, the 5-year survival rates fell within the same range as those in the earlier series. The poor overall survival rates reflect death from intercur-­ rent disease. In an M. D. Anderson ­Cancer Center series of 117 patients,90 51% died of intercurrent disease. However, local control was achieved in only one third of patients treated. Because local control and survival are still far from sat-­ isfactory, there is great interest in the use of 3D CRT to deliver higher doses while sparing normal lung tissue for patients with inoperable NSCLC, many of whom have seriously impaired pulmonary reserve. Preliminary results from dose-escalation studies with these patients reported by Rosenzweig and colleagues91 showed that much higher total doses could be achieved than the usually accepted tolerance dose of 60 to 65 Gy given over 6 weeks. These investigators recognized the volume-related morbidity of pulmonary irradiation and the need to restrict the volume that receives 20 Gy or more to 35%. One third of all patients with NSCLC have tumors that cannot be resected but have no demonstrable distant ­metastasis. The advent of PET scanning for pretreatment evaluation65 will affect this proportion by its superior ability to detect extrathoracic metastasis compared with other imaging studies such as CT scanning and magnetic resonance imaging (MRI). PET studies will also be useful in monitoring the response of the disease to treatment, ­particularly with regard to local tumor control. Local control of bulky primary tumors and nodal dis-­ ease within the thorax is a major determinant of survival for squamous cell carcinoma in particular and, to a lesser extent, for adenocarcinoma and large cell carcinoma of the lung. This concept was suggested by studies of patterns of failure and causes of death (see Table 19-2). Studies of the effect of local control by radiation therapy have confirmed that control of the primary tumor and its regional metas-­ tases is associated with much better survival than progres-­ sion of locoregional disease.92 This belies the myth that all ­patients die of distant dissemination regardless of what happens to the primary disease. The local tumor, if not eradicated, is the cause of death of most patients with inoperable NSCLC. Table 19-252 shows that death after treatment with palliative irra-­ diation or single-agent chemotherapy results more often from intrathoracic disease than extrathoracic metastasis. Saunders and colleagues93 studied causes of death among ­patients with localized unresectable NSCLC treated with a few large fractions of radiation; most were followed un-­ til ­autopsy. These investigators found that 72% of patients died as the result of the intrathoracic tumor and only 15%

19  The Lung, Pleura, and Thymus

439

BOX 19-2.  Lymph Node Map Definitions Nodal Stations N2 nodes 1 Highest mediastinal nodes 2 Upper paratracheal nodes 3 Prevascular and ­retrotracheal nodes 4 Lower paratracheal nodes

5 Subaortic (aorto-­pulmonary window) 6 Para-aortic nodes (ascending aorta or phrenic) 7 Subcarinal nodes 8 Paraesophageal nodes (below the carina) 9 Pulmonary ligament nodes

Anatomic Landmarks All N2 nodes lie within the mediastinal pleural envelope Nodes lying above a horizontal line at the upper rim of the bracheocephalic (left innominate) vein where it extends to the left, crossing in front of the trachea at its midline Nodes lying above a horizontal line drawn tangential to the upper margin of the aortic arch and below the inferior boundary of No. 1 nodes Prevascular and retrotracheal nodes may be designated 3A and 3P; midline nodes are ­considered to be ipsilateral The lower paratracheal nodes on the right lie to the right of the midline of the trachea ­between a horizontal line drawn tangential to the upper margin of the aortic arch and a line extending across the right main bronchus at the upper margin of the upper lobe bronchus, and contained within the mediastinal pleural envelope; the lower paratracheal nodes on the left lie to the left of the midline of the trachea between a horizontal line drawn tangential to the upper margin of the aortic arch and a line extending across the left main bronchus at the level of the upper margin of the left upper lobe bronchus, medial to the ligamentum arteriosum and contained within the mediastinal pleural envelope Researchers may wish to designate the lower parabracheal nodes as No. 4s (superior) and No. 4i (inferior) subsets for study purposes: the No. 4s nodes may be defined by a horizontal line extending across the trachea and drawn tangential to the cephalic border of the azygos vein; the No. 4i nodes may be defined by the lower boundary of the No. 4s and the lower boundary of No. 4, as described above Subaortic nodes are lateral to the ligamentum arteriosum or the aorta or left pulmonary artery and proximal to the first branch of the left pulmonary artery and lie within the mediastinal pleural envelope Nodes lying anterior and lateral to the ascending aorta and the aortic arch or the innominate artery, beneath a line tangential to the upper margin of the aortic arch Nodes lying caudal to the carina of the trachea, but not associated with the lower lobe bronchi or arteries within the lung Nodes lying adjacent to the wall of the esophagus and to the right or left of the midline, excluding subcarinal nodes Nodes lying within the pulmonary ligament, including those in the posterior wall and lower part of the inferior pulmonary vein

N1 nodes

All N1 nodes lie distal to the mediastinal pleural reflection and within the visceral pleura

10 Hilar nodes

The proximal lobal nodes, distal to the mediastinal pleural reflection and the nodes adjacent to the bronchus intermedius on the right; radiographically, the hilar shadow may be created by enlargement of both hilar and interlobar nodes Nodes lying between the lobar bronchi Nodes adjacent to the distal lobar bronchi Nodes adjacent to the segmental bronchi Nodes around the subsegmental bronchi

11 Interlobar nodes 12 Lobar nodes 13 Segmental nodes 14 Subsegmental nodes

From Mountain CF, Dressler CM. Regional lymph node classification of lung cancer staging. Chest 1997;111:1718-1723.

TABLE 19-6.    Medically Inoperable Stage I Non–Small Cell Lung Cancer Treated with Radiation Overall Actuarial Survival (%) Study and Reference

Number of Patients

Dose (Gy)

2 Years

5 Years

Morita et al, 199786

149

55-75

34

22

199887

141

50-80

39

13

199988

  36

60-81

42 (3 yr)

23

M. D. Anderson Cancer Center, 200089

117

20-83.9*

46

18

Sibley et al,

Hayakawa et al,

*Neutrons.

440

PART 5  Thorax

had extrathoracic metastasis. In a subsequent randomized trial, these investigators showed that continuous hyper-­ fractionated accelerated radiation therapy (1.5 Gy given 3 times/day for 12 consecutive days) improved local con-­ trol and survival rates for patients with inoperable NSCLC compared with standard fractionation radiation ­ therapy (a total dose of 60 Gy given in 2.0 Gy/fraction).94 A ­prospective trial95 of low-dose cisplatin given concurrently with radiation ­therapy daily or weekly as a radiosensitizer improved ­ local control and survival relative to radiation therapy alone. ­Clonogenic tumor cells evidently continue to proliferate during treatment of NSCLC,96 which ex-­ plains the success of the ­hyperfractionated regimen relative to a 6-week course of irradiation and the value of concur-­ rent ­chemotherapy.

Techniques of External Irradiation Radiation therapy is much less effective for carcinoma of the lung than for tumors of similar histopathologic type presenting in other locations. Reasons for this are becom-­ ing clearer. Squamous cell carcinomas and adenocarcino-­ mas of the lung may have greater proliferative potential than ­tumors at other sites, and they may be associated with more unrecognized extensions of disease. Many tumors in the parenchyma of the lung move in unpredictable direc-­ tions during respiration. Careful immobilization of patients is important, and CT planning of therapy is critical. A margin of 1.5 to 2 cm around the intrathoracic tumor, including its movement during respiration, is required. Studies are underway to characterize tumor motion and to stop respiration briefly during treatment to achieve better tumor coverage and better sparing of normal lung (Fig. 19-13).97 Debate continues about the value of encompassing clinically uninvolved regional lymph node drainage sites. Studies of failure patterns after irradiation of only the ­demonstrable tumor suggest that failure in the nodes is uncommon.98 However, even relatively small tumors not associated with node enlargement on chest CT have been shown to have a high risk of regional node involvement.79 Preliminary evidence suggests that PET scanning66 can ­reveal such subclinical node involvement and indicate when lymph nodes must be included or when they can be omitted.

FIGURE 19-13.  Coronal reconstruction of a four-dimensional thoracic CT data set. Gross tumor volume based on an end­expiration data set is shown in green; gross tumor volume based on an end-inspiration data set is shown in red. (From Starkschall G. Respiratory gated radiation therapy. In Cox JD, Chang JY, ­Komaki R [eds]: Image-guided Radiotherapy of Lung Cancer. New York: Informa Health Care, 2008, pp 83-92.)

Local control rates have been unacceptably low with standard fractionation and total doses of 60 to 65 Gy. In the first large study (353 patients) in which local control was evaluated with fiberoptic bronchoscopy and biopsy of the original tumor site, Arriagada and colleagues99 ­reported ­local control in less than 20% of patients who received 65 Gy in 26 fractions over 6.5 to 7 weeks. A total dose of 60 Gy, given in 30 fractions of 2 Gy each, had been the standard in comparative trials of the RTOG for sev-­ eral years. However, a report40 of a possible dose-response ­effect on survival when the radiation dose was between 60 Gy and 69.6 Gy, given in fractions of 1.2 Gy twice daily separated by 4 hours or more, led to a phase III random-­ ized trial of 60 Gy given as 2 Gy/day for 5 days per week, compared with. 69.6 Gy given as 1.2 Gy twice daily (with a 6-hour interfraction interval) for 5 days per week.100,101 In that trial, the 69.6-Gy fractionation regimen produced no improvement in survival over the 60-Gy regimen, with corresponding median survival times of 12 months and 11.4 months and 5-year survival rates of 6% and 5%, ­respectively.101 The markedly different densities of thoracic tissues and the dose-response relationships for normal tissues and ­tumors indicate a need for inhomogeneity corrections.102 Uncertainties still exist concerning the best way to make such corrections. Advanced treatment planning systems can make corrections automatically; in that case, it is ­important to anticipate the need to increase total doses. Total doses of 60 Gy or more (without inhomogeneity corrections) are necessary,103 and much higher total doses may be ben-­ eficial. Dose-escalation studies are underway, but they are limited by the fact that larger tumors necessarily require irradiation of larger volumes of normal lung. Treatment planning using three-dimensional systems provides very important information, especially dose-volume histograms. However, advanced treatment planning, even including IMRT, only partially solves the problem. Proton therapy could permit delivery of very high doses without increasing the volume of lung treated early in the treatment process, with subsequent treatments adjusted as needed to reduce the dose as the tumor shrinks (Fig. 19-14).104 These and other ­advanced approaches require further study.

Chemotherapy and Radiation Therapy for Non–Small Cell Lung Cancer According to standard staging procedures, one half of all patients with NSCLC have evidence of distant metasta-­ ses at diagnosis. Those without demonstrable extrathoracic spread have a high probability of occult metastasis. Che-­ motherapy is an important component of treatment for most patients with inoperable NSCLC. Numerous studies have shown the benefit of combining chemotherapy with radiation therapy for inoperable NSCLC. Induction (neoadjuvant) chemotherapy followed by ­radiation therapy has many attractive features. Such a plan allows the most immediate treatment of all components of the tumor, both clinically evident and subclinical. The demonstration of response to chemotherapy provides justification for its continuation during or after radiation therapy. Three prospective trials that compared induc-­ tion chemotherapy and subsequent radiation therapy with high-total-dose radiation therapy alone provide compelling evidence of the value of this combination.

19  The Lung, Pleura, and Thymus

441

100

Living patients (%)

80

A

Induction chemotherapy 60 40 Radiation therapy only 20

Hot skin dose

0 0

6

12

18

24

30

Time on study (months) Hot lung dose

B

C FIGURE 19-14.  Isodose distributions for proton therapy for a patient with stage III non–small cell lung cancer. The original plan (A) would not be suitable for a subsequent treatment 3 weeks later (B) because the tumor shrinkage would result in unacceptably high skin and lung doses. An adapted treatment plan based on the smaller tumor (C) reduces the skin and lung doses accordingly. (From Chang JY, Mohan R, Cox JD. Imageguided proton radiotherapy in lung cancer. In Cox JD, Chang JY, Komaki R [eds]: Image-guided Radiotherapy of Lung Cancer. New York: Informa Health Care, 2008, pp 103-126.)

The first of these trials was published by Dillman and colleagues105 in 1990. The Cancer and Leukemia Group B (CALGB) 8433 trial compared induction therapy with cisplatin and vinblastine for 5 weeks plus radiation therapy (60 Gy given as 2.0 Gy/fraction for 5 days/week) beginning on day 50 of treatment with immediate radiation therapy but no chemotherapy. Although survival was significantly improved among patients in the induction-­chemotherapy group, a high death rate was observed during the first 100 days in the radiation-alone group that was eventually explained by an imbalance in performance status ­between the two groups.106 A replication of the CALGB trial by the RTOG and ECOG studied a much larger number of patients but yielded nearly identical results—induction chemotherapy caused a small but statistically significant improvement in survival (Fig. 19-15).100

FIGURE 19-15.  Survival according to treatment group in the Radiation Therapy Oncology Group 88-08 and Eastern Cooperative Oncology Group 4588 trials. Induction chemotherapy plus irradiation (blue line) resulted in improved outcome over radiation therapy only (red line). (Data from Sause WT, Scott C, Taylor S, et al. Radiation Therapy Oncology Group (RTOG) 88-08 and Eastern Cooperative Oncology Group (ECOG) 4588: ­preliminary results of a phase III trial in regionally advanced, unresectable non-small-cell lung cancer. J Natl Cancer Inst 1995;87:198-205.)

The third positive trial was reported by Le Chevalier and colleagues.107 A cooperative group in France enrolled 353 patients in a study that compared induction chemotherapy (three monthly cycles of vindesine, cyclophosphamide, cis-­ platin, and lomustine followed by radiation therapy [65 Gy given as 2.5 Gy/fraction for 4 days/week] beginning on day 75 to 80) with radiation therapy given in the same way but beginning on day 1. This group also found that induction chemotherapy improved survival.108 Perhaps more impor-­ tant, these investigators showed a statistically significant re-­ duction in the incidence of distant metastasis with induction chemotherapy but no improvement in local tumor control.99 Their systematic use of fiberoptic bronchoscopy and biopsy at the site of the original tumor 3 months after the start of treatment allowed these investigators to show that tumor persisted in more than 80% of patients and that induction chemotherapy conferred no advantage in local control. The rationale for adding a single chemotherapeutic agent to radiation therapy is to increase local tumor con-­ trol. Such therapy would be expected to have little or no effect on distant metastasis. An important study from the European Organisation for Research and Treatment of Cancer95 (EORTC) established the value of concurrent ­chemotherapy, with cisplatin given daily or weekly at the same time as radiation therapy. Local tumor ­ control im-­ proved (Fig. 19-16), and survival was better (Fig. 19-17).95 These findings corroborated the evidence that control of the intrathoracic tumor is an important determinant of ­ survival for patients with locally advanced NSCLC, ­especially the squamous cell type, and that improving out-­ come for patients with NSCLC depends on improved local ­tumor control. Sause and colleagues100,101 reported a small series of patients treated with induction cisplatin and vinblastine followed by radiation therapy with concurrent cisplatin.

442

PART 5  Thorax 100

Group 1 RT Group 2 RT + cisplatin weekly Group 3 RT + cisplatin daily

80

Group 1 RT Group 2 RT + cisplatin weekly Group 3 RT + cisplatin daily

80 Survival (%)

Patients without recurrence (%)

100

60 40

60 40 20

20

0

0 0

1

2

3

0

4

1

2

FIGURE 19-17.  Overall survival for patients with non–small cell lung cancer. Comparison of treatment group 1 (blue line) with groups 2 (red line) and 3 (green line) showed significant differences (P = .04). (From Schaake-Koning C, van den Bogaert W, Dalesio O, et al. Effects of concomitant cisplatin and radiotherapy on inoperable non-small-cell lung cancer. N Engl J Med 1992;326:524-530.)

Survival rate (%)

­Komaki and colleagues106 reported experience with this regimen with larger numbers of patients and found it to be well tolerated and encouraging with regard to median sur-­ vival time (15.5 months) and 1-year rate of survival (65%). The question of whether induction or concurrent che-­ motherapy is better is a question of whether control of distant metastases is more important than control of ­local tumor and distant metastases. Two similar trials provide compelling evidence in answer to this question. Furuse and colleagues109 conducted a phase III trial of induction ­chemotherapy with cisplatin, vindesine, and mitomycin fol-­ lowed by radiation therapy (56 Gy in 2.0-Gy fractions for 5 fractions/week for 28 fractions) starting approximately 7 to 8 weeks after the start of chemotherapy. This regimen was compared with one using the same chemotherapy plus radiation therapy beginning on the second day; the radia-­ tion therapy was modified by permitting a 10-day inter-­ ruption after administration of the first 28 Gy, but the total dose was the same. Among the 320 patients enrolled in this trial, the median survival time for patients receiving con-­ current radiation therapy was 16.5 months, compared with 13.3 months for those receiving sequential treatment. The overall survival rate was significantly better among those given concurrent therapy (P = .03998) (Fig. 19-18).109 Curran and colleagues110,111 reported a study of the RTOG comparing three treatments. The first group was given vinblastine and cisplatin followed by radiation ther-­ apy to 60 Gy, given as 2.0 Gy per fraction for 5 fractions per week beginning on day 50. Patients in the second group were given the same regimen, but with radiation therapy beginning on day 1. A third group was given concurrent cis-­ platin and etoposide and hyperfractionated irradiation to a total dose of 69.6 Gy, given as 1.2 Gy twice daily separated by 6 or more hours. Preliminary results showed the second treatment to be superior to the first, with median patient survival of 17 months in the second group and 14.5 months in the first. Survival is now statistically ­ superior in cases of concurrent (rather than sequential) chemotherapy and ­radiation therapy.

4

Year of study

Year of study

FIGURE 19-16.  Survival without local recurrence for patients with non–small cell lung cancer. Comparison of treatment group 1 (blue line) with groups 2 (red line) and 3 (green line) showed significant differences (P = .009). (From Schaake-Koning C, van den Bogaert W, Dalesio O, et al. Effects of concomitant cisplatin and radiotherapy on inoperable non-small-cell lung cancer. N Engl J Med 1992;326:524-530.)

3

100 90 80 70 60 50 40 30 20 10 0

P = .03998 Concurrent Sequential 0

1

2

3

4

5

6

7

Time on study (years) No. of patients alive Concurrent 156 100 Sequential 158 86

54 43

34 23

25 16

11 7

FIGURE 19-18.  Survival by treatment group in the West Japan Lung Cancer Group trial of induction (sequential, red line) or concurrent (blue line) thoracic irradiation for inoperable non– small cell lung cancer. (From Furuse K, Fukuoka M, Kawahara M, et al. Phase III study of concurrent versus sequential thoracic radiotherapy in combination with mitomycin, vindesine, and cisplatin in unresectable stage III non-small-cell lung cancer. J Clin Oncol 1999;17:2692-2699.)

The very similar findings of the Furuse and the RTOG studies suggest that concurrent chemoradiation should be considered the standard treatment for unresectable NSCLC as long as the patients have the same favorable performance status and weight loss characteristics of the patients in those trials.

Targeted Molecular Therapy in Non–Small Cell Lung Cancer The difficulty in successfully treating unresectable NSCLC, particularly in patients with poor performance status, has led to study of several potential targets for molecular-level therapy. Those targets include cyclooxygenase 2 (COX-2),

19  The Lung, Pleura, and Thymus

443

PA SVC MB

SVC

PA MB

FIGURE 19-19.  CT scans of a patient given celecoxib with concurrent radiation therapy for unresectable non–small cell lung cancer as part of a phase I trial. Top left, Aorta-pulmonary window view shows partial obstruction of the superior vena cava (SVC) at presentation. Top right, Subcarinal view shows evidence of compression of the pulmonary artery (PA) and right main bronchus (MB), as well as subcarinal node enlargement. Bottom left, The patent SVC is evident after treatment with 45 Gy in 15 fractions. Bottom right, Patent MB and PA and regression of subcarinal nodes are evident at 6 months’ follow-up. (Photo courtesy of Zhongxing Liao, MD)

an enzyme involved in prostanoid synthesis, and two re-­ ceptor tyrosine kinases, epidermal growth factor receptor (EGFR) and vascular endothelial growth factor recep-­ tor (VEGFR). In a phase I study conducted by Liao and colleagues,112 48 patients with unfavorable NSCLC were given radiation therapy in combination with the COX-2 inhibitor celecoxib. The plan was to escalate the celecoxib dose from 200 mg/day to 400 mg/day, 600 mg/day, and ultimately 800 mg/day (the highest dose approved by the U.S. Food and Drug Administration). Among the 37 evaluable patients, this combination regimen produced 14 complete responses and 13 partial responses (which in one case included resolution of a partial obstruction of the superior vena cava and compression of other structures [Fig. 19-19]); the other 10 patients had stable or progres-­ sive disease. Actuarial local progression-free survival rates were encouraging at 66% at 1 year after the initiation of radiation therapy and 42% at 2 years. The maximum toler-­ ated dose could not be determined.112 Other targeted-therapy agents being tested include ­cetuximab, an inhibitor of EGFR activity, and ZD6474 (van-­ detanib), an inhibitor of EGFR and VEGFR that targets a variety of molecular signaling pathways involved in ­tumor cell proliferation and survival (Fig. 19-20).113-116 Clini-­ cal trials of the combination of tyrosine kinase inhibitors with radiation or combined radiation and chemotherapy for NSCLC were prompted by encouraging results from its use for head and neck cancer (see Chapters 5 and 7); however, several questions remain unanswered about this

strategy, including which such inhibitor is most ­ effective in combination with irradiation, whether combinations of such inhibitors are more effective than just one, and how to identify patients who are most likely to benefit from such therapy.117 Other questions to be addressed ­focus on the mechanism of action of tyrosine kinase ­ inhibitors— ­whether they act by radiosensitizing tumor cells, by block-­ ing tumor cell regrowth between radiation fractions, or by being cytotoxic to tumor cells.117 Preliminary findings from RTOG 0324 suggest that ­cetuximab results in relatively low rates of esophagitis and grade 3 or higher pneumonitis (Table 19-7). Laboratory and phase III studies are comparing the relative effectiveness of single- or multiple-agent tyrosine kinase inhibition, alone or in combination with chemotherapy, for NSCLC.118 The optimal doses and timing of such combinations also need to be determined, preferably by additional preclinical stud-­ ies and by small clinical trials that include biopsy.117 The combination of vandetanib and radiation therapy has shown therapeutic effectiveness in laboratory studies of lung cancer xenograft models (Fig. 19-21).119 Scheduledependent interactions between vandetanib dosing and ra-­ diation therapy are being investigated in laboratory studies of head and neck cancer xenograft models.120 In mechanis-­ tic studies of another targeted-therapy compound, Itasaka and colleagues121 showed that the angiogenesis inhibitor endostatin significantly enhanced the radioresponsiveness of A431 human epidermoid carcinoma xenografts in mice; this combination affected the expression of interleukin-8

444

PART 5  Thorax EGFR

VEGFR-2

Tumor cell

Inhibition

Metastasis

Invasion

Angiogenesis

Endothelial cell

Vascular permeability

Inhibition of apoptosis

Migration

Protease production Proliferation

Proliferation

FIGURE 19-20.  Consequences of inhibiting the tyrosine kinase activity of the epithelial growth factor receptor (EGFR) and the vascular endothelial growth factor receptor-2 (VEGFR-2) by ZD6474 (vandetanib). Blockade or inhibition of a variety of signaling pathways leads to suppression of invasion and metastasis in tumor cells and vascular permeability and protease production in endothelial cells. (From Frederick B, Gustafson D, Bianco C, et al. ZD6474, an inhibitor of VEGFR and EGFR tyrosine kinase activity in combination with radiotherapy. Int J Radiat Oncol Biol Phys 2006;64:33-7.)

TABLE 19-7.    Preliminary Toxicity Findings from Trials of Chemoradiation Therapy for Unresectable   Non–Small Cell Lung Cancer Pneumonitis-Free   ­Survival Rate at 1 Year

Grade ≥3 ­Esophagitis

Grade ≥3   Pneumonitis

Sequential CRT (RT starts on day 50)

33%

  3%

  7%

Concurrent CRT (RT starts on day 1)

43%

24%

  3%

Concurrent CRT (hyperfractionated RT starts on day 1)

36%

45%

  3%

Concurrent CRT (hyperfractionated RT starts on day 43) with amifostine

53%

26%

18%

Concurrent CRT (hyperfractionated RT starts on day 43) without amifostine

53%

30%

13%

52% (30.1%-73.4%)

12%

10%

Conformal RT

33%

19%

  3%

Chemotherapy + conformal RT

35%

17%

  2%

Trial RTOG 94-10

RTOG 98-01

RTOG 0324 CALGB 39801

CALGB, Cancer and Leukemia Group B; CRT, chemoradiation therapy; RT, radiation therapy; RTOG, Radiation Therapy Oncology Group.

and matrix metalloproteinase-2 markedly but had only minimal effects on VEGF expression.121 Fundamental questions remain to be answered regarding whether EGFR and VEGFR inhibitors do inhibit EGFR or VEGFR in study subjects and, if so, how the signaling molecules downstream of these targets are affected by the combination of tyrosine kinase inhibitors and irradiation.117

Prophylactic Cranial Irradiation for Non–Small Cell Lung Cancer Patients with inoperable adenocarcinoma or large cell carci-­ noma of the lung have a high probability of developing brain metastases (see Table 19-4); the same is true of ­ patients with resectable tumors who have metastases to regional lymph nodes.71 Tables 19-2 and 19-4 show that 8% to 12%

19  The Lung, Pleura, and Thymus

445

FIGURE 19-21.  Images of thoracic chest wall and lymph node metastases illustrate the effectiveness of ZD6474 and radiation therapy in a lung adenocarcinoma mouse model. Top left, control (no treatment). Top right, radiation therapy only. Bottom left, ZD6474 only. Bottom right, ZD6474 plus radiation therapy. (From Shibuya K, Komaki R, Shintani T, et al. Targeted therapy against VEGFR and EGFR with ZD6474 enhances the therapeutic efficacy of irradiation in an orthotopic model of human non-small cell lung cancer. Int J Radiat Oncol Biol Phys 2007;69:1534-1543.)

of patients who die of these types of cancer of the lung have metastasis only to the brain. Corroborating data come from experiences with patients who have developed clinical brain metastases, have undergone irradiation, and have lived 5 to 10 years with no further evidence of cancer.60 In a randomized prospective trial of the Veterans ­Administration Lung Group,122 patients with NSCLC ­received prophylactic cranial irradiation (PCI) to a total dose of 20 Gy given in 10 fractions over 2 weeks. Brain metastases were present less often in patients who had had normal findings on brain radionuclide scans before treat-­ ment. Although the numbers were too small for correla-­ tion with each specific cell type, patients with squamous cell carcinoma, large cell carcinoma, or adenocarcinoma had a statistically significantly lower frequency of brain metastasis than did those with other types of lung cancer (29% versus 0%, P < .05).122 No difference in survival was found between patients who received cranial irradiation and those who did not. According to autopsy findings relating to single-organ brain metastasis (see Table 19-4), PCI is likely to affect

survival among patients with adenocarcinoma or large cell carcinoma. Russell and colleagues123 reported the results of a phase III study of PCI for patients at high risk for ce-­ rebral metastasis. These investigators found a trend toward reduction of intracranial spread with PCI but no effect on survival. They concluded that PCI had no role, even for those with high-risk disease.

Radiation Therapy for Small Cell Lung Carcinoma No doubt remains about the importance of radiation ­therapy delivered to the thoracic tumor for SCLC limited to the chest. Two meta-analyses124,125 showed the value of tho-­ racic irradiation for controlling disease within the chest, and a small but statistically significant improvement in survival also was observed. Combining chemotherapy with radiation therapy for SCLC can be particularly effective126 for shrink-­ ing the primary tumor (Fig. 19-22) and for reducing bulky lymph node involvement (Fig. 19-23). The optimal combi-­ nation of these two modalities for SCLC is not clear. The use of radiation and ­chemotherapy ­simultaneously seems to be

446

PART 5  Thorax

FIGURE 19-22.  Before (left) and after (right) CT images of limited small cell lung cancer treated with chemotherapy and boost-field irradiation to avoid toxicity to normal tissues. Considerable shrinkage of the primary tumor (right) occurred after 3 weeks of therapy, when the radiation therapy plan was recalculated.

Before

After

FIGURE 19-23.  Before (left) and after (right) CT images of the chest of a 71-year-old woman with small cell lung cancer. Bulky and extensive lymph node involvement is evident before treatment (left), and an excellent response to chemoradiation therapy is evident after treatment (right). (From Komaki R. Radiotherapy guidelines in small cell lung cancer. In: Cox JD, Chang JY, Komaki R, eds. Imageguided Radiotherapy of Lung Cancer. New York: Informa Health Care, 2008, pp 39-62.)

associated with the highest complete response rates, but it is unquestionably associated with increased morbidity, espe-­ cially esophagitis and bone marrow suppression. Adminis-­ tering radiation therapy after the initiation of chemotherapy may be the most effective use of systemic therapy. The maximum tolerated dose of thoracic radiation in combination radiation and chemotherapy for limitedstage SCLC in a multisite phase I study was found to be 61.2 Gy.127 In that trial, 64 patients were given cisplatin and etoposide chemotherapy in conjunction with acceler-­ ated thoracic radiation therapy for four cycles.127 During the first two cycles, radiation was delivered at 1.8 Gy/­ fraction daily to the clinical target volume, beginning on day 1; during the last two cycles, radiation was delivered to the gross tumor volume twice a day for the last 3, 5, 7, 9, or 11 treatment days, for corresponding total doses of 50.4, 57.6, 61.2, or 64.8 Gy. The maximum tolerated dose, that which produced grade 3 to 4 nonhematologic ­toxicity in more than 50% of treated patients, was found to

be 61.2 Gy. The 18-month survival rate for patients given that dose (61.2 Gy) was 82% as compared with 25% for those given 50.4 Gy.127 Murray and colleagues128 in Canada addressed the ques-­ tion of timing in a prospective randomized comparison of early and late thoracic irradiation (Fig. 19-24). Cycles of cyclophosphamide, doxorubicin, and vincristine were alter-­ nated with cisplatin and etoposide. Thoracic irradiation was given with the cisplatin and etoposide beginning on week 3 or week 15. The total dose was 40 Gy, given in 15 fractions over 3 weeks. Patients who showed a complete response to chemotherapy and thoracic irradiation were offered PCI to 25 Gy given in 10 fractions over 2 weeks. Survival was significantly better (P = .008) among patients given early thoracic irradiation (Fig. 19-25).128 Patients who received early thoracic irradiation also had a significant reduction (P = .006) in the incidence of brain metastasis. Another study that addressed the question of timing of thoracic irradiation was conducted by the Japanese ­Clinical

447

19  The Lung, Pleura, and Thymus

Early thoracic irradiation

C A V

0

E P

C A V

3

7

E P

C A V

E P

10

13

16

19

Surviving patients (%)

100

21

Weeks

Median survival (months) Early TI 21.2 Late TI 16.0

80 60

P = 0.008

40 20

R

0

0

E P

3

C A V

6

E P

9

C A V

12

0.0 E P

15

18

21

Weeks

FIGURE 19-24.  Schema of the National Cancer Institute of Canada trial of early versus late thoracic irradiation for small cell cancer of the lung. CAV, cyclophosphamide, doxorubicin, and vincristine; EP, etoposide and cisplatin. (From Murray N, Coy P, Pater JL, et al. Importance of timing for thoracic irradiation in the combined modality treatment of limited-stage small-cell lung cancer. The National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 1993;11:336-344.)

Oncology Group.129 Investigators randomly ­ assigned 228 patients to receive induction chemotherapy followed by thoracic irradiation or concurrent treatment. The chemo-­ therapy consisted of cisplatin and etoposide and the ­total radiation dose was 45 Gy given in 1.5-Gy, twice-daily frac-­ tions for 5 days per week for 3 weeks. The 2-year survival rates favored concurrent (55%) over sequential (35%) treatment (P = .057). The fractionation question has been addressed in two phase II studies and one phase III study. McCracken and colleagues130 from the SWOG combined intravenous cis-­ platin, etoposide, and vincristine in two 4-week cycles with concurrent radiation therapy (45 Gy given in 1.8-Gy frac-­ tions daily for 5 days per week); PCI was given with a third cycle of cisplatin and etoposide, and additional chemother-­ apy including methotrexate, doxorubicin, and cyclophos-­ phamide was given for 12 weeks. Among the 154 patients treated, this therapy produced relatively high survival rates (42% at 2 years and 30% at 4 years).130 In the other phase II study, Turrisi and Glover131 gave 32 patients cisplatin and etoposide concurrently with accelerated fractionation (45 Gy given as 1.5 Gy twice daily for 5 days/week). Two cycles of chemotherapy were given at 3-week intervals dur-­ ing irradiation, and six additional cycles alternated cisplatinetoposide with ­cyclophosphamide-­doxorubicin-­vincristine. PCI was given to those who showed complete response after chemotherapy. The 2-year disease-free survival rate was 45%, and no deaths from cancer were observed ­after 2 years.130 Updated results of this combined-modality ­approach using twice-daily radiation showed 2-year and 4-year survival rates of 54% and 36%, respectively, with a local tumor control rate of 96%.132 An intergroup phase III study coordinated by the ECOG, with participation by the RTOG, compared the two approaches (i.e., once-daily versus accelerated ­

1.0

2.0

3.0

4.0

5.0

Time (years)

No. at risk (early) No. at risk (late)

155

122

64

40

23

11

153

99

53

32

18

6

FIGURE 19-25.  Survival by timing of thoracic irradiation (TI) in the National Cancer Institute of Canada trial. (From Murray N, Coy P, Pater JL, et al. Importance of timing for thoracic irradiation in the combined modality treatment of limited-stage smallcell lung cancer. The National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 1993;11:336-344.)

1.0 Probability of survival

Late thoracic irradiation

C A V

P = .04 by log-rank rest

0.8 0.6 0.4

Twice-daily radiotherapy

0.2 Once-daily radiotherapy

0.0 0

10

20

30

40

50

60

70

80

90 100

Months Treatment group

0–20 mo

20–40 mo

40–60 mo

60–80 mo

80–100 mo

No. of deaths/No. at risk Once daily Twice daily

108/206 100/211

48/96 47/109

15/47 7/62

4/21 5/42

0/5 1/14

FIGURE 19-26.  Survival by treatment group: accelerated fractionation (twice-daily radiation therapy) versus standard treatment (once-daily radiation therapy). (From Turrisi AT III, Kim K, Blum R, et al. Twice-daily compared with once-daily thoracic radiotherapy in limited small-cell lung cancer treated concurrently with cisplatin and etoposide. N Engl J Med 1999;340:265-271.)

twice-daily radiation therapy). Both treatments involved concurrent administration of cisplatin and etoposide with the ­ radiation. The results, reported by Turrisi and colleagues,133 showed a significant survival advantage (P = .04) for the 211 patients treated with twice-daily ­irradiation relative to the 206 patients treated once daily (Fig. 19-26).133 Their 2-year survival rates are the best ­reported to date in a cooperative group trial (40%). The 5-year survival rates were 28% for patients given accelerated fractionation and 21% for those given once-daily treatment. Nevertheless, the ­intrathoracic failure rates were far too high; one half of the patients treated once daily and one third of the patients treated twice daily had local recurrence. Efforts

448

PART 5  Thorax

are underway to improve local control by increasing the total dose above the 45 Gy used in the ECOG-RTOG intergroup trial, while keeping the overall treatment time as short as possible. At least some patients with disseminated SCLC may also benefit from thoracic irradiation. Livingston and col-­ leagues134 observed a high proportion of patients (29 of 43 [67%]) with disseminated disease that responded com-­ pletely to chemotherapy but had treatment failure first in the chest without other evidence of disease. It seems ­appropriate to consider thoracic irradiation for patients with disseminated disease if the tumor proves to be highly responsive to systemic chemotherapy.

Prophylactic Cranial Irradiation for Small Cell Lung Carcinoma PCI is an important part of treatment for many patients with SCLC. As is true for thoracic irradiation, questions still remain about when in the course of treatment the PCI should be given and how to identify which patients would benefit from it. Even with effective systemic chemotherapy, the risk of brain metastasis from SCLC increases at the rate of 2% to 3% per month for 18 to 24 months.50,135,136 If pa-­ tients who develop neurologic symptoms and signs and are found to have brain metastasis could be treated success-­ fully with radiation therapy, PCI would be less important in the overall management plan. In one study of 40 patients with brain metastasis from SCLC (confirmed by radionu-­ clide scans or CT), 20 died as a direct result of progression of the intracranial lesions.137 Baglan and Marks138 reported more success in treating patients with overt intracranial metastasis from SCLC, but they eliminated patients who had such rapidly progressive disease that effective symp-­ tomatic treatment could not be instituted; those patients might have benefited from PCI. PCI has been considered unnecessary in the treatment of patients with SCLC because it has had no effect on survival in several studies. However, the end point in such studies has usually been median survival duration; as is true for evaluat-­ ing the effectiveness of thoracic irradiation, this end point is relatively meaningless for evaluating PCI. Rosen and col-­ leagues139 assessed long-term survival rates in three groups of patients given systemic chemotherapy for SCLC at the National Cancer Institute between 1970 and 1980. In group 1, PCI was administered from the first day of treatment; in group 2, PCI was given between the 12th and 24th weeks; in group 3, PCI was omitted. These investigators found a significant improvement in long-term survival rates among patients who received PCI: the 30-month survival rates were 27% in group 1, 40% in group 2, and 14% in group 3.

A comparison of two sequential treatment policies at the Memorial Sloan-Kettering Cancer Center by Rosenstein and colleagues140 suggested that PCI could be important for long-term survival. Of 36 patients with limited SCLC whose treatment included PCI (from 1979 to 1982), the brain was the first site of failure in 18% of the 22 patients who showed a complete response to treatment. Between 1985 and 1989, a group of 26 patients treated without PCI had a 45% rate of disease failure in the brain. The 2-year survival rates were 42% for the PCI-treated patients as op-­ posed to 13% for those not receiving PCI; moreover, 38% of those given PCI were alive at 5 years compared with none of those who were not given cranial irradiation.140 A meta-analysis141 of seven prospectively randomized clinical trials showed higher rates of disease-free and over-­ all survival for patients who received PCI. However, this analysis has been criticized because more than one half of the trials had fewer than 100 patients, the frequency of neurotoxicity was not addressed in patients who ­received PCI compared with those who did not, and the dose­fractionation regimens used were not consistent. Late neuropsychological sequelae and CT abnormalities have been found in patients given PCI and combination chemo-­ therapy for SCLC.142,143 Although the PCI has consistently been implicated in these findings, it is also likely that certain chemotherapeutic agents—the nitrosoureas, intravenous methotrexate, and the epipodophyllotoxins—interact with cranial radiation to produce late sequelae. In most studies, the individual fraction size for PCI has ranged from 3 to 4 Gy. In one nonrandomized comparison of the results of PCI given at 2.5 Gy and 3 Gy per fraction combined with systemic therapy with cyclophosphamide, doxorubicin, and vincristine (Table 19-8),136 the smaller fraction seemed as effective as the larger one.136 Although long-term followup of these patients was lacking, as were detailed neuro-­ psychological evaluations, the lowest effective fraction size and the lowest total dose of PCI seem the least likely to be associated with late neuropsychological sequelae. In a prospective study of patients undergoing PCI for SCLC,144 detailed neuropsychological assessments before and after PCI showed a strikingly high frequency of neu-­ rocognitive dysfunction before PCI, and deficits were no greater after PCI. Much of the late neurocognitive deficits seen after treatment for SCLC may be related to the disease itself, a paraneoplastic phenomenon, or the ­chemotherapy. Lee and others145 used decision analysis to evaluate quality-adjusted life expectancy for patients with limited SCLC given or not given PCI, assuming certain survival rates and certain levels of PCI neurotoxicity. PCI was found to offer higher quality-adjusted life expectancy ­values

TABLE 19-8.    Frequency of Brain Metastasis in Small Cell Carcinoma by Treatment Group Treatment Group No PCI

Number of Patients

Number with Brain Metastasis

82

21 (26%)

30 Gy (10 fractions of 3.0 Gy)

51

5 (10%)

25 Gy (10 fractions of 2.5 Gy)

194

6 (3%)

PCI

PCI, prophylactic cranial irradiation. From Komaki R, Cox JD, Hartz AJ, et al. Characteristics of long-term survivors after treatment for inoperable carcinoma of the lung. Am J Clin Oncol 1985;8:362-370.

19  The Lung, Pleura, and Thymus

(compared with no PCI) for patients with SCLC who achieved complete response to chemotherapy—but only when the assumed neurotoxicity level was low as opposed to substantial. The authors of this study observed that as survival rates for patients with SCLC continue to improve, the incidence and severity of neurotoxicity must be con-­ trolled to maintain a favorable benefit-risk ratio in terms of recommending PCI.145

Radiation Therapy for Apical Sulcus Tumors Any histopathologic type of carcinoma of the lung can occur in the extreme apex or superior sulcus of the lung (see Table 19-5)61 and cause Pancoast syndrome, which manifests as shoulder or arm pain, often accompanied by dysesthesia and atrophy along the distribution of the ­ulnar nerve. Patients also may have Horner’s syndrome and ­radiographic evidence of destruction of ribs or vertebrae. The diagnosis of apical sulcus tumors often can be estab-­ lished by percutaneous needle aspiration; sputum cytology studies and bronchoscopy are usually unrevealing. Preoperative radiation therapy followed by exploratory thoracotomy and resection, if possible, has been considered standard treatment for apical sulcus tumors for the past 30 years. Although there was little objective justification for this practice other than the results of a small, highly ­selective

series reported by Paulson,146,147 it was a reasonable ­approach, and no alternative treatments were apparent at the time. Sev-­ eral subsequent series have convincingly demonstrated that apical sulcus tumors can be cured with aggressive radiation therapy.148-150 This information, combined with the recogni-­ tion that split-course radiation therapy is disadvantageous in the local control of cancer of the lung, justifies a completely different approach to ­ patients with potentially resectable carcinoma in the apical sulcus of the lung. Although tumors that present in the apical sulcus ­often cannot be diagnosed by bronchoscopy, fine-needle aspi-­ ration under CT guidance usually can yield a diagnosis. ­Pretreatment evaluation by MRI is essential because MRI more accurately displays the anatomy of superior sulcus tumors than does CT.151 A sagittal section obtained by MRI can show extension of such a tumor to the poste-­ rior wall of the subclavian artery, the vertebral bodies, and mediastinal lymph nodes. MRI is essential to evaluate the resectability of superior sulcus tumors. MRI of the brain is also important because the propensity of superior sulcus tumors to spread to the brain is greater than that of tumors at other locations in the lung.152 The radiation dose distri-­ bution must be carefully controlled, given the proximity of apical tumors to critical structures (Fig. 19-27).45 In a study of the effectiveness of neoadjuvant radia-­ tion and chemotherapy followed by surgical resection of

B

A

449

C

Norm. volume

DOSE VOLUME HISTOGRAM 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0

GTV CTV PTV Total lung Esophagus Heart Cord

0

D

1000 2000 3000 4000 5000 6000 7000 8000 Dose (cGy)

E

FIGURE 19-27.  Image-guided intensity-modulated radiation therapy is delivered with concurrent chemotherapy for a superior sulcus tumor. In this case, 69.6 Gy was prescribed to the gross tumor volume plus a 3-mm setup uncertainty margin (white lines); 60 Gy was delivered to the clinical tumor volume (green lines), and 50 Gy was delivered to the planning target volume (blue lines). The dose to the brachial plexus was limited to less than 66 Gy, and the dose to the spinal cord was limited to less than 45 Gy. A, Transverse view. B, Sagittal view. C, Coronal view. D, Dose-volume histogram. E, Onboard cone beam CT used before delivery of each fraction. (From Chang JY. Guidelines and techniques for image-guided radiotherapy for non-small cell lung cancer. In Cox JD, Chang JY, Komaki R [eds]: Image-guided Radiotherapy of Lung Cancer. New York: Informa Health Care, 2008, pp 19-37.)

450

PART 5  Thorax

T3 or T4 NSCLC tumors in the superior sulcus, Rusch and associates153 reported that 5-year survival rates were 44% for all 110 eligible patients and 54% for those who under-­ went surgery (i.e., those with disease that remained stable or responded to the neoadjuvant chemotherapy). These rates are an improvement over previous survival rates of about 30% achieved with protocols that did not include chemotherapy. Survival rates were better among patients whose tumors showed pathologic complete response than among those with residual disease after radiation and chemotherapy (P = .02). Assembling this group of patients required the cooperative efforts of 76 surgeons represent-­ ing all of the North American cooperative study groups. Although interest in preoperative radiation therapy and combined radiation and chemotherapy continues,154 this sequence is disadvantageous for the 20% to 50% of patients who prove to have unresectable tumors. It is not possible to pursue a successful course of treatment after an uncon-­ trolled interruption of radiation therapy or chemotherapy for the operative procedure. If mediastinoscopy shows that no lymph nodes are involved, thoracotomy can be per-­ formed first and some tumors can be resected completely; these patients can be spared all other treatments unless they are participating in an investigational protocol. If the tumors are unresectable, aggressive management with con-­ current chemotherapy and radiation therapy can be pur-­ sued with curative intent.

Radiation Therapy for Superior Vena Cava Obstruction Approximately 5% of all patients with cancer of the lung have signs and symptoms suggesting obstruction of the superior vena cava. In the United States, 95% of patients with these manifestations have a malignant tumor; in three fourths of these patients, it is cancer of the lung. The ob-­ struction is usually the result of extrinsic compression by the tumor, although direct invasion of the vessel wall can occur. Thrombosis can further impede venous return. Although superior vena cava obstruction is considered a radiation therapy emergency, the onset of symptoms and signs can be acute or subacute. A histologic or cytologic diag-­ nosis can usually be made by means of sputum cytology tests, fiberoptic bronchoscopy, or percutaneous needle aspiration. If small cell carcinoma is confirmed, it is usually desirable to initiate systemic combination chemotherapy and concomi-­ tant thoracic irradiation if the disease is limited to the chest. The most effective palliation is achieved by radiation therapy consisting of first large fractions and later more conventional fractionation.155,156 Fractions of 3.5 or 4 Gy for the first 3 or 4 days usually relieve the symptoms of dyspnea within a few days; 85% to 90% of patients156,157 experience relief of symptoms and signs within 3 weeks. Corticosteroids are not needed, because they have not been shown to speed the relief of symptoms.

Inoperable Carcinoma of the Lung Prognosis The most important prognostic factors for patients with inoperable carcinoma of the lung, even those who will ul-­ timately die of the disease, are pretreatment performance

status158 and weight loss.159 Despite the fact that patients with minimal symptoms and no weight loss live longer than those who are more symptomatic, virtually every ­ patient with cancer of the lung who is not effectively treated dies within a relatively short time. Before the demonstration of the effectiveness of radia-­ tion therapy for cancer of the lung and the existence of the dose-response relationships, an argument could be made for withholding treatment from patients who were rela-­ tively asymptomatic. It was thought that radiation therapy would have marginal or no benefit and might produce symptoms. This idea has persisted into the megavoltage era as a general therapeutic nihilism about lung cancer.160 A review of studies conducted by the Veterans Adminis-­ tration Lung Group in which patients with inoperable lung cancer were given placebos and “best-available supportive care” consisting of antibiotics, expectorants, and oxygen showed that only 4% of patients were alive at 2 years.38 Of the 82 patients who had the highest performance sta-­ tus (80 to 100 score on the Karnofsky scale, meaning they were minimally symptomatic) and who were treated only with supportive measures, none survived 5 years. How-­ ever, ­ Komaki and associates136 reported that 45 (11%) of 410 patients were alive 3 years and 32 (8%) were alive 5 years after therapy for cancer of the lung. The 3-year sur-­ vival rate for 197 patients treated between 1971 and 1975 was 7%, compared with 14% for patients treated from 1975 through 1978. The improvement in survival was not because of selecting more patients with more favorable disease ­stages for treatment but rather was because of the availability of more effective treatments for patients with advanced-stage cancer, especially those with metastases to the ­mediastinal and supraclavicular lymph nodes (Table 19-9).136 All long-term survivors had a performance status score of 80 or above on the Karnofsky scale before treatment (Table 19-10).136 Of the patients who were alive at 36 months, 80% lived 5 or more years, and two thirds lived 10 years. The observation that large cell carcinoma is associ-­ ated with the greatest probability of long-term survival prompted a subsequent study of patients who were treated more recently than the 1970s.161 In that study, large cell carcinoma and adenocarcinoma were combined because failure pattern analyses had not shown any differences between these two groups. Compared with patients with squamous cell carcinoma, patients with adenocarcinoma or large cell carcinoma had a higher probability of long-term ­survival.161

Palliative Irradiation Tumors that recur postoperatively in the bronchial stump or mediastinum can be reduced by radiation, although long-term control is rare. Postirradiation recurrences that have an endobronchial component may benefit from brachytherapy. Remote afterloading of bronchoscopically placed catheters with radioactive sources, usually iridium 192 (192Ir), can be used to relieve symptoms and reverse atelectasis.162 Most patients with cancer of the lung develop distant metastases that produce distressing symptoms and even-­ tually threaten life. Radiation is the most effective means of alleviating symptoms when they are clearly localized. The most common sites of distant metastasis are the liver,

19  The Lung, Pleura, and Thymus

TABLE 19-9.    Effect of Tumor Extent on 3-Year Survival Rates Survival Rate (%) during the Specified Treatment Period Tumor Extent

1971-1975

1975-1978

I

33.3 (36)*

20.8 (48)

II

4.8 (21)

18.2 (33)

III

1.4 (140)

10.6 (132)

Stage

Node involvement None

18.2 (72)

8.0 (81)

Hilar

14.3 (21)

15.2 (33)

Mediastinal

1.2 (86)

12.2 (82)

Supraclavicular

5.6 (18)

17.6 (17)

*Numbers

in parentheses are numbers of patients. Modified from Komaki R, Cox JD, Hartz AJ, et al. Characteristics of long-term survivors after treatment for inoperable carcinoma of the lung. Am J Clin Oncol 1985;8:362-370.

a­ drenal glands, bones, and brain.49 The only symptom from hepatic metastases that can be relieved by radiation therapy is pain.163 Adrenal metastasis can cause pain in the back, flank, or upper abdomen, and irradiation usually confers prompt relief. Prospective studies of radiation therapy for skeletal metastasis have been carried out by the RTOG.164 The ­patients treated in these trials had metastases from many primary sites, but cancer of the lung was common. The re-­ sults of treatment of approximately 1000 patients showed that 90% experienced some relief, and more than one half had complete disappearance of pain. The RTOG also conducted prospective trials of radia-­ tion therapy for the palliative treatment of metastasis to the brain.165 The frequency with which symptoms were relieved was related to the types of neurologic symptoms and signs. Complete relief was achieved in 35% to 72% of patients, and some benefit was observed in 65% to 85%. Major motor and focal seizures and headache were the symptoms most consistently relieved (60% to 70% of cases). Symptoms disappeared completely less often in patients with impaired mentation, cerebellar dysfunction, motor loss, and cranial nerve abnormalities than in patients without these impairments, but fully two thirds of such patients do experience some benefit. Back pain and neurologic symptoms and signs may herald the onset of paraplegia caused by metastases from cancer of the lung, especially small cell carcinoma. Myelog-­ raphy or MRI can demonstrate the level of the block, and CT or MRI usually shows whether the spinal cord is being impinged upon as the result of an extradural tumor or ver-­ tebral collapse with mechanical compromise. Widening of the spinal cord may represent intramedullary metastasis.166 It is important to identify the cause as quickly as possible; the sooner dexamethasone is started and palliative irradia-­ tion is begun, the greater the probability of stabilizing or reversing symptoms and signs.

451

TABLE 19-10.    Effect of Pretreatment Status   on 3-Year Survival Rates Status

Survival Rate (%)

Performance Status (Karnofsky Scale) 80

0.0 (193)*

80-89

6.6 (76)

90-99

25.5 (131)

100

70.0 (10)

Tumor Histology Squamous cell Small cell

9.6 (228) 9.0 (67)

Adenocarcinoma

16.2 (37)

Large cell

28.6 (35)

Adenosquamous

10.0 (5)

Unspecified

0.0 (38)

*Numbers

in parentheses are numbers of patients. Modified from Komaki R, Cox JD, Hartz AJ, et al. Characteristics of long-term survivors after treatment for inoperable carcinoma of the lung. Am J Clin Oncol 1985;8:362-370.

The Thymus The thymus plays a critical role in the development of cellmediated immunity. It acts on primitive cells that origi-­ nate in the bone marrow to render them immunologically competent. Surgical removal of the thymus in the newborn decreases circulating lymphocytes and severely impairs cel-­ lular immunity. Thymectomy in older children and adults does not result in immune deficiencies, although the circu-­ lating lymphocytes may still decrease. Similarly, irradiation of the adult thymus may decrease the number of circulat-­ ing lymphocytes; however, vast experience with incidental irradiation of the thymus during treatment of Hodgkin dis-­ ease, malignant lymphoma, and cancer of the breast, lung, and esophagus has not led to evidence of immune altera-­ tion independent of the diseases themselves.

Thymomas Tumors of the thymus are uncommon, and most are ­benign. Limited experiences therefore have led to disparate views of the natural history and treatment of invasive thymus neoplasms. Thymomas are neoplasms of the thymic epi-­ thelial cells; tumors that had been grouped with them in the past (i.e., “granulomatous thymomas”) are discussed with tumors of hematopoietic tissues (see Chapter 34) and are now properly identified as Hodgkin disease with nodu-­ lar sclerosis. Seminomas and related tumors are discussed along with extragonadal malignant germinal tumors so that they can be compared and contrasted with their gonadal counterparts in Chapters 27 and 32. Thymomas account for at least 10% of all mediastinal tumors.48 They arise in the anterior mediastinum, although they may rarely be found in the posterior mediastinum or in the neck.167 Although the median patient age at the time of diagnosis is 50 years, thymomas can occur at any age. They constitute about 10% of mediastinal tumors in children and are seen with equal frequency in males and females.

452

PART 5  Thorax

TABLE 19-11.    Invasive Thymoma Recurrence after Complete Resection by Treatment Group Number of Recurrences in Number of Patients Treated* Study and Reference al175

No Postoperative Irradiation

Postoperative Irradiation

1 in 1

9 in 19

Cohen et

al176

4 in 5

5 in 8

Batata et

al174

Nordstrom et

8 in 8

6 in 11

Ariaratnam et al177

—

3 in 8

Gerein et al178

—

2 in 6

1 in 1

0 in 3

9 in 21

14 in 68

Penn and

Hope-Stone179

Monden et

al180

al181

—

0 in 3

Arriagada et al182

—

1 in 6

Pollack et al183

1 in 3

1 in 6

Krueger et al184

Marks et

—

0 in 1

Urgesi et

al185

—

3 in 33

Curran et

al186

8 in 19

0 in 5

32 in 58 (57%)

44 in 177 (25%)

Totals *Recurrence

indicates a new manifestation at any site, local or distant.

Thymomas have been grouped according to their histo-­ pathologic features. Most investigators recognize predomi-­ nantly epithelial, predominantly lymphoid, and spindle cell types. There is little correlation between natural history and subtype. Correlation is poor between microscopic cri-­ teria for malignancy and the behavior of these neoplasms. Suster and Rosai168 studied histologic material in which they found malignant cytologic features that justified the designation thymic carcinoma from 60 patients, 48 of whom had undergone resection. They found that patients who had high-grade tumors that were poorly circumscribed, lacked lobular growth patterns, and had a high mitotic index had a rapidly fatal outcome. However, the terms benign and ­malignant can be quite misleading when refer-­ ring to thymomas. Pollack and colleagues169 performed flow ­cytometric analysis of paraffin-imbedded tissues from 25 patients with thymomas, specifically excluding thy-­ mic carcinomas; they found no relationship between the percentage of cells in S phase and patient outcome, but survival was significantly better for the 16 patients with diploid tumors than for the 9 patients whose tumors were aneuploid (P = .002). Even tumors that are considered relatively benign have a propensity to recur locally, but distant metastasis is un-­ common. Metastasis has been reported in 5% of all thy-­ momas.170,171 At least one fourth of patients with invasive thymomas develop metastases, usually to supraclavicular lymph nodes, liver, or bone. The most important features that determine treatment and outcome are whether a thymoma is encapsulated or in-­ vasive and whether myasthenia gravis is present. One third to one half of all thymomas are invasive. Although only 15% of all patients with myasthenia gravis have a thymic tumor, one third to one half of all patients with thymomas have myasthenia gravis.

A useful staging classification, based on surgical and pathologic findings, was originally proposed by Bergh and colleagues172 and modified by Masaoka and others173: Stage I Macroscopically completely encapsulated; no microscopic capsular invasion Stage II Microscopic invasion into capsule; macroscopic invasion into surrounding fatty tissue or mediastinal pleura Stage III Macroscopic invasion into neighboring organs Stage IVA Pleural or pericardial implants Stage IVB Lymphogenous or hematogenous metastasis

Treatment Complete resection of the tumor should be accomplished if at all possible. Encapsulated (stage I) and minimally invasive (stage II) thymomas are virtually always resectable unless there are medical contraindications to thoracotomy. At least 20% of obviously invasive (stage III) thymomas are unre-­ sectable.173,174 Contemporary imaging procedures, ­primarily high-resolution CT with intravenous contrast or MRI, can provide information to clarify the resectability of a tumor. A stage I thymoma that is completely resected can recur locally or by pleural implantation. Whether radiation ther-­ apy also should be given for encapsulated tumors is unclear. No evidence supports the systematic use of postoperative irradiation for completely resected stage I tumors. If total gross removal can be achieved and the tumor proves to be invasive, postoperative irradiation is well jus-­ tified. Batata and colleagues,174 reporting the experience from Memorial Hospital in New York City, found that no patient with invasive thymoma was cured by resection alone. A review of the very limited data available (Table 19-11)174-186 strongly suggests that adjuvant radiation therapy after surgery reduces the probability of tumor ­recurrence in patients who have invasive thymomas. These

19  The Lung, Pleura, and Thymus

453

TABLE 19-12.    Local Control of Unresectable Invasive Thymoma According to Total Radiation Dose Total Dose (Gy)* Study and Reference Gerein et Marks et

al178

al181 al190

≤48

49-59

≥60

2 in 9

—

1 in 2

6 in 6

—

—

—

0 in 3

—

Ariaratnam et al177

6 in 10

1 in 1

—

Arriagada et al182

7 in 12

11 in 17

3 in 3

1 in 1

5 in 8

—

1 in 3

6 in 8

—

23 in 47 (49%)

26 in 41 (63%)

4 in 5 (80%)

Chahinian et

Kersh et

al191

Krueger et Totals *Number

al184

of cases controlled in the number of cases treated.

data were based on all recurrences, not just recurrences within the irradiated volume. It is possible to increase the resectability of invasive thy-­ momas. Unresectable thymomas can be rendered resectable by preoperative radiation therapy. However, encouraging results with chemotherapy187 have led us to pursue initial chemotherapy for unresectable thymomas, reserving radia-­ tion therapy for use postoperatively or as definitive therapy if unresectability persists. Our current regimen consists of a combination of cyclophosphamide, doxorubicin, cisplatin, and prednisone.188 For every patient with an invasive thymoma that cannot be completely resected, definitive radiation therapy is necessary. These tumors have a considerable propensity to disseminate within the pleural cavity and to spread beyond the thorax. The entire mediastinum and both hila should be encom-­ passed within the irradiated volume, as well as any sites of pleural implantation identified by the thoracic surgeon. The supraclavicular region also should be included if ad-­ enopathy is present. Prophylactic irradiation of one or both supraclavicular fossae can be recommended, but its contribution to survival remains to be determined. Simi-­ larly, it may be desirable to irradiate the entire hemithorax if pleural implantation has been documented, although ­experience with this approach is limited.189 The dose required to control invasive thymoma is not clear. Reports by Gerein and colleagues178 and Marks and colleagues180 indicate that patients treated with a total dose of 45 Gy or less after complete tumor resec-­ tion had ­ recurrent disease, whereas those who received a higher total dose remained free of cancer. When only a partial resection or biopsy is performed, higher total doses are advisable. Available data concerning the dose response of unresectable thymoma are summarized in Table 19-12.177,178,181,182,184,190,191 Although the published experience is not sufficient to know the control rate at ­total doses of 60 Gy or higher, the substantial failure rate at ­lower doses suggests that the total dose should be as high as can be tolerated by the surrounding normal tissues.

Results Myasthenia gravis has long been considered to have an unfavorable effect on the prognosis of patients with thy-­ moma. However, the studies of Shamji and colleagues,192 Verley and Hollmann,193 Wilkins and Castleman,194 Maggi

and colleagues,195 and Urgesi and colleagues185 have shown that myasthenia does not adversely affect prognosis. The role of radiation therapy for invasive thymoma is clearly established, although many uncertainties still ex-­ ist about the optimal treatment volume and dose-time relationships. Of 24 patients reported by Batata and col-­ leagues,174 5 (21%) were alive and well 5 years after irradi-­ ation, 7 (29%) were alive with disease, and 12 (50%) were dead of disease. No patient who underwent only tumor resection was cured. Nordstrom and colleagues175 reported 20 patients with stage III thymoma, all but one of whom received radiation therapy: no patients were alive and well 5 or more years after treatment. Kersh and colleagues191 reviewed the cases of 10 patients with invasive thymo-­ ma that had not been completely resected; local control was achieved in 6 patients, but 3 of them died of distant metastatic disease. Arriagada and colleagues182 reported a retrospective study of 56 cases from four hospitals in or near Paris; 6 patients underwent complete tumor resection, 22 had incomplete resections, and 28 had only biopsies. All received radiation therapy. The local recurrence rate at 2 years was 31%, and 21 patients (37.5%) developed ­metastasis beyond the thorax. The actuarial survival rate at 5 years was 46%. With more systematic use of radia-­ tion therapy as soon as the diagnosis of invasive thymoma is established, the prognosis for these patients might be ­substantially improved. The initial stage of the tumor is an important prognostic factor. Disease stage is closely related to the achievement of complete resection. The outlook for patients with stage I or II tumors that have been completely resected is excel-­ lent.186 Patients who have complete or subtotal resection and then postoperative radiation have a far more favorable prognosis if their cancer can be classified as stage III rather than stage IV.185

REFERENCES    1. Evans W, Leucutia T. Intrathoracic changes induced by heavy radiation. Am J Roentgenol 1925;13:203-220.    2. Lacassagne A. Action des rayons du radium sur les muqueueses de l’oesophage et de la trachee chez le lapin. CR Soc Biol 1921;84:26-27.    3. Fajardo L, Berthrong M, Anderson R. Respiratory tract. In: Ra-­ diation Pathology. New York: Oxford University Press, 2001, pp 198-208.

454

PART 5  Thorax

   4. King R, Jones M, Minoo P. Regulation of lung cell proliferation by polypeptide growth factors. Am J Physiol 1989;257:23-38.    5. Rubin P, Johnston CJ, Williams JP, et al. A perpetual cascade of cytokines postirradiation leads to pulmonary fibrosis. Int J Ra-­ diat Oncol Biol Phys 1995;33:99-109.    6. Rodemann HP, Binder A, Burger A, et al. The underlying cel-­ lular mechanism of fibrosis. Kidney Int Suppl 1996;54:S32-36.    7. Franklin TJ. Therapeutic approaches to organ fibrosis. Int J Bio-­ chem Cell Biol 1997;29:79-89.    8. Grande JP. Role of transforming growth factor-beta in tissue in-­ jury and repair. Proc Soc Exp Biol Med 1997;214:27-40.    9. Anscher MS, Murase T, Prescott DM, et al. Changes in plasma TGF beta levels during pulmonary radiotherapy as a predictor of the risk of developing radiation pneumonitis. Int J Radiat Oncol Biol Phys 1994;30:671-676.   10. Beck LS, DeGuzman L, Lee WP, et al. One systemic admin-­ istration of transforming growth factor-beta 1 reverses age- or glucocorticoid-impaired wound healing. J Clin Invest 1993;92: 2841-2849.   11. McDonald S, Rubin P, Constine L, et al. Biochemical markers as predictors for pulmonary effects of radiation. Rad Oncol Invest 1995;3:56-63.   12. Finkelstein JN, Johnston CJ, Baggs R, et al. Early alterations in extracellular matrix and transforming growth factor beta gene expression in mouse lung indicative of late radiation fibrosis. Int J Radiat Oncol Biol Phys 1994;28:621-631.   13. Martin M, Lefaix J, Delanian S. TGF-beta1 and radiation fibro-­ sis: a master switch and a specific therapeutic target? Int J Radiat Oncol Biol Phys 2000;47:277-290.   14. Delanian S. Striking regression of radiation-induced fibrosis by a combination of pentoxifylline and tocopherol. Br J Radiol 1998;71(848):892-894.   15. Piguet PF, Rosen H, Vesin C, et al. Effective treatment of the pulmonary fibrosis elicited in mice by bleomycin or silica with anti-CD-11 antibodies. Am Rev Respir Dis 1993;147:435-441.   16. Anscher MS, Kong FM, Andrews K, et al. Plasma transforming growth factor beta1 as a predictor of radiation pneumonitis. Int J Radiat Oncol Biol Phys 1998;41:1029-1035.   17. Anscher MS, Thrasher B, Rabbani Z, et al. Antitransforming growth factor-beta antibody 1D11 ameliorates normal tissue damage caused by high-dose radiation. Int J Radiat Oncol Biol Phys 2006;65:876-881.   18. Gopal R, Tucker S, Komaki R, et al. The relationship between local dose and loss of function for irradiated lung. Int J Radiat Oncol Biol Phys 2003;56:106-113.   19. Liao ZX, Travis EL, Tucker SL. Damage and morbidity from pneumonitis after irradiation of partial volumes of mouse lung. Int J Radiat Oncol Biol Phys 1995;32:1359-1370.   20. Tucker SL, Liao ZX, Travis EL. Estimation of the spatial distri-­ bution of target cells for radiation pneumonitis in mouse lung. Int J Radiat Oncol Biol Phys 1997;38:1055-1066.   21. Travis EL, Liao ZX, Tucker SL. Spatial heterogeneity of the vol-­ ume effect for radiation pneumonitis in mouse lung. Int J Radiat Oncol Biol Phys 1997;38:1045-1054.   22. Travis E, Komaki R. Treatment-related lung damage. In Pass HI, Mitchell JB, Johnson DH, et al (eds): Lung Cancer: Princi-­ ples and Practice. 3rd ed, Philadelphia: Lippincott Williams & Wilkins, 2005.   23. Boersma LJ, Theuws JC, Kwa SL, et al. Regional variation in functional subunit density in the lung: regarding Liao et al. IJROBP 32:1359-1370; 1995. Int J Radiat Oncol Biol Phys 1996;34:1187-1188.   24. Khan MA, Hill RP, Van Dyk J. Partial volume rat lung irradia-­ tion: an evaluation of early DNA damage. Int J Radiat Oncol Biol Phys 1998;40:467-476.   25. Yamada M, Kudoh S, Hirata K, et al. Risk factors of pneumoni-­ tis following chemoradiotherapy for lung cancer. Eur J Cancer 1998;34:71-75.   26. Whitfield AG, Bond WH, Kunkler PB. Radiation damage to tho-­ racic tissues. Thorax 1963;18:371-380.

  27. Stone D, Schwarz M, Green R. Fatal pulmonary insufficiency due to radiation effect upon the lung. Am J Med 1956;21: 211-226.   28. Bergmann M, Graham EA. Pneumonectomy for severe irradia-­ tion damage of the lung. J Thorac Surg 1951;22:549-567.   29. Lipshitz H, Shuman L. Radiation-induced pulmonary change: CT findings. J Comput Assist Tomogr 1984;8:15-19.   30. Byhardt RW, Abrams R, Almagro U. The association of adult respiratory distress syndrome (ARDS) with thoracic irradiation (RT). Int J Radiat Oncol Biol Phys 1988;15:1441-1446.   31. Bernard GR, Luce JM, Sprung CL, et al. High-dose corticos-­ teroids in patients with the adult respiratory distress syndrome. N Engl J Med 1987;317:1565-1570.   32. Wang S, Liao Z, Wei X, et al. Analysis of clinical and dosimetric factors associated with treatment-related pneumonitis (TRP) in patients with non-small-cell lung cancer (NSCLC) treated with concurrent chemotherapy and three-dimensional con-­ formal radiotherapy (3D-CRT). Int J Radiat Oncol Biol Phys 2006;66:1399-1407.   33. Yom SS, Liao Z, Liu HH, et al. Initial evaluation of treatmentrelated pneumonitis in advanced-stage non-small-cell lung cancer patients treated with concurrent chemotherapy and ­intensity-modulated radiotherapy. Int J Radiat Oncol Biol Phys 2007;68:94-102.   34. Sweany SK, Moss WT, Haddy FJ. The effects of chest irradiation on pulmonary function. J Clin Invest 1959;38:587-593.   35. Moss W, Haddy FJ, Sweany SK. Some factors altering the se-­ verity of acute radiation pneumonitis: variation with cortisone, heparin, and antibiotics. Radiology 1960;75:50-54.   36. Choi N, Kanarek D, Kazemi H. Physiologic changes in pul-­ monary function after thoracic radiotherapy for patients with lung cancer and role of regional pulmonary function studies in predicting postradiotherapy pulmonary function before radio-­ therapy. Cancer Treat Symp 1985;2:119-130.   37. Brady LW, Germon PA, Cander L. The effects of radiation ther-­ apy on pulmonary function in carcinoma of the lung. Radiology 1965;85:130-134.   38. Cox JD, Komaki R, Byhardt RW. Is immediate chest radio­ therapy obligatory for any or all patients with limited-stage non– small cell carcinoma of the lung? Yes. Cancer Treat Rep 1983;67: 327-331.   39. Cox JD. Fractionation: a paradigm for clinical research in ra-­ diation oncology. Int J Radiat Oncol Biol Phys 1987;13: 1271-1281.   40. Cox JD, Azarnia N, Byhardt RW, et al. A randomized phase I/II trial of hyperfractionated radiation therapy with total doses of 60.0 Gy to 79.2 Gy: possible survival benefit with greater than or equal to 69.6 Gy in favorable patients with Radiation Therapy Oncology Group stage III non–small-cell lung carci-­ noma: report of Radiation Therapy Oncology Group 83-11. J Clin Oncol 1990;8:1543-1555.   41. Guerrero T, Castillo R, Noyola-Martinez J, et al. Reduction of pulmonary compliance found with high-resolution computed tomography in irradiated mice. Int J Radiat Oncol Biol Phys 2007;67:879-887.   42. Wei X, Liu HH, Tucker SL, et al. Risk factors for acute esophagi-­ tis in non–small-cell lung cancer patients treated with ­concurrent chemotherapy and three-dimensional conformal radiotherapy. Int J Radiat Oncol Biol Phys 2006;66:100-107.   43. Wolf J, Spear P, Yesner R. Nitrogen mustard and the steroid hor-­ mones in the treatment of inoperable bronchogenic carcinoma. Am J Med 1960;29:1008-1016.   44. Komaki R, Lee JS, Milas L, et al. Effects of amifostine on acute toxicity from concurrent chemotherapy and radiotherapy for in-­ operable non-small-cell lung cancer: report of a randomized com-­ parative trial. Int J Radiat Oncol Biol Phys 2004;58:1369-1377.   45. Liu HH. The physics aspects of intensity-modulated radiother-­ apy for lung cancers. In Cox JD, Chang JY, Komaki R (eds): Image-guided Radiotherapy of Lung Cancer. New York: Informa Healthcare, 2008, pp 103-126.   46. Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2007. CA Cancer J Clin 2007;57:43-66.

19  The Lung, Pleura, and Thymus   47. American Cancer Society. Cancer Facts and Figures 2007. ­Atlanta: American Cancer Society; 2007.   48. Del Regato J, Spjut H, Cox JD. Cancer: Diagnosis, Treatment and Prognosis, 6th ed. St Louis: Mosby, 1985.   49. Yesner R, Carter D. Pathology of carcinoma of the lung. Chang-­ ing patterns. Clin Chest Med 1982;3:257-289.   50. Cox JD, Yesner RA. Adenocarcinoma of the lung: recent results from the Veterans Administration Lung Group. Am Rev Respir Dis 1979;120:1025-1029.   51. Kuhn CI, Askin F. Lung and mediastinum. In: Kissane J, ed: ­ Anderson’s Pathology, 9th ed. St Louis: Mosby, 1990, pp 920-1046.   52. Cox JD, Yesner RA. Causes of treatment failure and death in carcinoma of the lung. Yale J Biol Med 1981;54:201-207.   53. Nohl HC. An investigation into the lymphatic and vascular spread of carcinoma of the bronchus. Thorax 1956;11:172-185.   54. Greene FL, Page DL, Fleming ID, et al (eds): AJCC Cancer Stag-­ ing Manual, 6th ed. New York: Springer, 2002, pp 167-177.   55. Carlens E. Appraisal of choice and results of treatment for bron-­ chogenic carcinoma. Chest 1974;65:442-445.   56. Goldberg EM, Glicksman AS, Khan FR, et al. Mediastinoscopy for assessing mediastinal spread in clinical staging of carcinoma of the lung. Cancer 1970;25:347-353.   57. Palumbo LT, Sharpe WS. Scalene node biopsy. Correlation with other diagnostic procedures in 550 cases. Arch Surg 1969;98: 90-93.   58. Nugent JL, Bunn PA Jr, Matthews MJ, et al. CNS metastases in small cell bronchogenic carcinoma: increasing frequen-­ cy and changing pattern with lengthening survival. Cancer 1979;44:1885-1893.   59. Komaki R, Cox JD, Whitson W. Risk of brain metastasis from small cell carcinoma of the lung related to length of survival and prophylactic irradiation. Cancer Treat Rep 1981;65:811-814.   60. Komaki R, Cox JD, Stark R. Frequency of brain metastasis in ad-­ enocarcinoma and large cell carcinoma of the lung: correlation with survival. Int J Radiat Oncol Biol Phys 1983;9:1467-1470.   61. Byrd RB, Carr DT, Miller WE, et al. Radiographic abnormali-­ ties in carcinoma of the lung as related to histological cell type. Thorax 1969;24:573-575.   62. Byhardt RW, Hartz A, Libnoch JA, et al. Prognostic influence of TNM staging and LDH levels in small cell carcinoma of the lung (SCCL). Int J Radiat Oncol Biol Phys 1986;12:771-777.   63. McLoud TC, Bourgouin PM, Greenberg RW, et al. Broncho-­ genic carcinoma: analysis of staging in the mediastinum with CT by correlative lymph node mapping and sampling. Radiol-­ ogy 1992;182:319-323.   64. Norlund JD, Byhardt RW, Foley WD, et al. Computed tom-­ ography in the staging of small cell lung cancer: implications for combined modality therapy. Int J Radiat Oncol Biol Phys 1985;11:1081-1084.   65. Pieterman RM, van Putten JW, Meuzelaar JJ, et al. ­Preoperative staging of non–small-cell lung cancer with positron-emission ­tomography. N Engl J Med 2000;343:254-261.   66. MacManus M. Influence of PET scanning on radiotherapeutic management of NSCLC. Lung Cancer 2000;29:180-181.   67. Mountain CF, Dressler CM. Regional lymph node classification of lung cancer staging. Chest 1997;111:1718-1723.   68. Komaki R. Preoperative and postoperative irradiation for cancer of the lung. J Belge Radiol 1985;68:195-198.   69. Komaki R. Preoperative radiation therapy for superior sulcus ­lesions. Chest Surg Clin N Am 1991;1:13-35.   70. van Houtte P. Postoperative radiotherapy for lung cancer. Lung Cancer 1991;7:57-64.   71. Shields T. Treatment failures after surgical resection of thoracic tumors. Cancer Treat Symp 1983;2:69.   72. PORT Meta-analysis Trialists Group. Postoperative ­radiotherapy in non-small-cell lung cancer: systematic review and meta­analysis of individual patient data from nine randomised con-­ trolled trials. Lancet 1998;352:257-263.   73. Chung CK, Stryker JA, O’Neill M Jr, et al. Evaluation of ­adjuvant postoperative radiotherapy for lung cancer. Int J Radiat Oncol Biol Phys 1982;8:1877-1880.

455

  74. Green N, Kurohara SS, George FW III, et al. Postresection ­irradiation for primary lung cancer. Radiology 1975;116:405407.   75. Kirsh MM, Kahn DR, Gago O, et al. Treatment of broncho-­ genic carcinoma with mediastinal metastases. Ann Thorac Surg 1971;12:11-21.   76. The Lung Cancer Study Group. Effects of postoperative me-­ diastinal radiation on completely resected stage II and stage III epidermoid cancer of the lung. N Engl J Med 1986;315: 1377-1381.   77. Keller SM, Adak S, Wagner H, et al. A randomized trial of post-­ operative adjuvant therapy in patients with completely resected stage II or IIIA non-small-cell lung cancer. Eastern Cooperative Oncology Group. N Engl J Med 2000;343:1217-1222.   78. Stephens RJ, Girling DJ, Bleehen NM, et al. The role of post­operative radiotherapy in non-small-cell lung cancer: a mul-­ ticentre randomised trial in patients with pathologically staged T1-2, N1-2, M0 disease. Medical Research Council Lung Cancer Working Party. Br J Cancer 1996;74:632-639.   79. Sawyer TE, Bonner JA, Gould PM, et al. Effectiveness of post-­ operative irradiation in stage IIIA non-small cell lung cancer according to regression tree analyses of recurrence risks. Ann ­Thorac Surg 1997;64:1402-1407; discussion 1407–1408.   80. Durci ML, Komaki R, Oswald MJ, et al. Comparison of sur-­ gery and radiation therapy for non-small cell carcinoma of the lung with mediastinal metastasis. Int J Radiat Oncol Biol Phys 1991;21:629-636.   81. Rossell R, Gomez-Codina J, Camps C. Preresectional chemother-­ apy in stage IIIA non-small cell lung cancer: a 7-year ­assessment of a randomized controlled trial. Lung Cancer 1999;26:7-14.   82. Roth JA, Fossella F, Komaki R, et al. A randomized trial compar-­ ing perioperative chemotherapy and surgery with surgery alone in resectable stage IIIA non-small-cell lung cancer. J Natl Cancer Inst 1994;86:673-680.   83. Albain KS, Rusch VW, Crowley JJ, et al. Concurrent cisplatin/ etoposide plus chest radiotherapy followed by surgery for stages IIIA (N2) and IIIB non-small-cell lung cancer: mature results of Southwest Oncology Group phase II study 8805. J Clin Oncol 1995;13:1880-1892.   84. Albain KS, Swann RS, Rusch VR, et al. Phase III study of con-­ current chemotherapy and radiotherapy (CT/RT) vs CT/RT fol-­ lowed by surgical resection for stage IIIA(pN2) non-small cell lung cancer (NSCLC): outcomes update of North American Intergroup 0139 (RTOG 9309) [abstract]. Proc Am Soc Clin Oncol 2005; 23(165):7014.   85. Dosoretz DE, Katin MJ, Blitzer PH, et al. Medically inoperable lung carcinoma: the role of radiation therapy. Semin Radiat ­Oncol 1996;6:98-104.   86. Morita K, Fuwa N, Suzuki Y, et al. Radical radiotherapy for medically inoperable non-small cell lung cancer in clinical stage I: a retrospective analysis of 149 patients. Radiother Oncol 1997; 42:31-36.   87. Sibley GS, Jamieson TA, Marks LB, et al. Radiotherapy alone for medically inoperable stage I non-small-cell lung cancer: the Duke experience. Int J Radiat Oncol Biol Phys 1998;40:149154.   88. Hayakawa K, Mitsuhashi N, Saito Y, et al. Limited field irra-­ diation for medically inoperable patients with peripheral stage I non-small cell lung cancer. Lung Cancer 1999;26:137-142.   89. Arzu J, Komaki R, Allen P, et al. Outcomes of patients with medically inoperable stage I non-small cell lung cancer (NSCLC) treated by radiation therapy (RT) alone [abstract 261]. Presented at the Eighty-sixth Scientific. Assembly and Annual Meeting of the Radiological Society of North America, 2000, Chicago, IL.   90. Komaki R, Arzu I, Allen P, et al. Radiation therapy for clini-­ cal stage I non-small cell lung cancer. Lung Cancer 2000; 29(Suppl 2):153-154.   91. Rosenzweig KE, Mychalczak B, Fuks Z, et al. Final report of the 70.2-Gy and 75.6-Gy dose levels of a phase I dose escala-­ tion study using three-dimensional conformal radiotherapy in the treatment of inoperable non-small cell lung cancer. Cancer J 2000;6:82-87.

456

PART 5  Thorax

  92. Eisert DR, Cox JD, Komaki R. Irradiation for bronchial car-­ cinoma: reasons for failure. I. Analysis of local control as a function of dose, time, and fractionation. Cancer 1976;37: 2665-2670.   93. Saunders MI, Bennett MH, Dische S, et al. Primary tumor control after radiotherapy for carcinoma of the bronchus. Int J ­Radiat Oncol Biol Phys 1984;10:499-501.   94. Saunders MI. The implications of the CHART trial for the treatment of non-small cell lung cancer. Lung Cancer 2000;2: 177-178.   95. Schaake-Koning C, van den Bogaert W, Dalesio O, et al. Effects of concomitant cisplatin and radiotherapy on inoperable nonsmall-cell lung cancer. N Engl J Med 1992;326:524-530.   96. Cox JD, Pajak TF, Marcial VA, et al. Interruptions adversely ­affect local control and survival with hyperfractionated radiation therapy of carcinomas of the upper respiratory and digestive tracts. New evidence for accelerated proliferation from Radia-­ tion Therapy Oncology Group Protocol 8313. Cancer 1992;69: 2744-2748.   97. Starkschall G. Respiratory gated radiation therapy. In Cox JD, Chang JY, Komaki R (eds): Image-guided Radiotherapy of Lung Cancer. New York: Informa Healthcare, 2008, pp 83-92.   98. Graham MV. Conformal three-dimensional radition therapy for early lung cancer [abstract]. Lung Cancer 2000;29(Suppl 2):179.   99. Arriagada R, Le Chevalier T, Quoix E, et al. ASTRO (Ameri-­ can Society for Therapeutic Radiology and Oncology) plenary: Effect of chemotherapy on locally advanced non-small cell lung carcinoma: a randomized study of 353 patients. GETCB (Groupe d’Etude et Traitement des Cancers Bronchiques), ­FNCLCC (Federation Nationale des Centres de Lutte contre le Cancer) and the CEBI trialists. Int J Radiat Oncol Biol Phys 1991;20:1183-1190. 100. Sause WT, Scott C, Taylor S, et al. Radiation Therapy Oncol-­ ogy Group (RTOG) 88-08 and Eastern Cooperative Oncology Group (ECOG) 4588: preliminary results of a phase III trial in regionally advanced, unresectable non-small-cell lung cancer. J Natl Cancer Inst 1995;87:198-205. 101. Sause W, Kolesar P, Taylor SIV, et al. Final results of phase III trial in regionally advanced unresectable non-small cell lung cancer. Radiation Therapy Oncology Group, Eastern Coopera-­ tive Oncology Group, and Southwest Oncology Group. Chest 2000;117:358-364. 102. Mah K, Van Dyk J. On the impact of tissue inhomogeneity cor-­ rections in clinical thoracic radiation therapy. Int J Radiat Oncol Biol Phys 1991;21:1257-1267. 103. Perez CA, Stanley K, Grundy G, et al. Impact of irradiation tech-­ nique and tumor extent in tumor control and survival of patients with unresectable non-oat cell carcinoma of the lung: report by the Radiation Therapy Oncology Group. Cancer 1982;50:10911099. 104. Chang JY, Mohan R, Cox JD. Image-guided proton radio-­ therapy in lung cancer. In Cox JD, Chang JY, Komaki R, (eds): ­Image-guided Radiotherapy of Lung Cancer. New York: Informa Healthcare, 2008, pp 127-140. 105. Dillman RO, Seagren SL, Propert KJ, et al. A randomized trial of induction chemotherapy plus high-dose radiation versus radia-­ tion alone in stage III non-small-cell lung cancer. N Engl J Med 1990;323:940-945. 106. Komaki R, Pajak TF, Byhardt RW, et al. Analysis of early and late deaths on RTOG non-small cell carcinoma of the lung tri-­ als: comparison with CALGB 8433. Lung Cancer 1993;10 (3-4):189-197. 107. Le Chevalier T, Arriagada R, Quoix E, et al. Radiotherapy alone versus combined chemotherapy and radiotherapy in nonresect-­ able non-small-cell lung cancer: first analysis of a randomized trial in 353 patients. J Natl Cancer Inst 1991;83:417-423. 108. Le Chevalier T, Arriagada R, Tarayre M, et al. Significant effect of adjuvant chemotherapy on survival in locally advanced nonsmall-cell lung carcinoma. J Natl Cancer Inst 1992;84:58. 109. Furuse K, Fukuoka M, Kawahara M, et al. Phase III study of con-­ current versus sequential thoracic radiotherapy in combination with mitomycin, vindesine, and cisplatin in unresectable stage III non-small-cell lung cancer. J Clin Oncol 1999;17:2692-2699.

110. Curran WJ, Scott C, Langer C, et al. Phase III comparison of se-­ quential vs concurrent chemoradiation for PTS with unresected stage III non-small cell lung cancer (NSCLC): initial report of Radiation Therapy Oncology Group (RTOG) 9410 [abstract]. Proc Am Soc Clin Oncol 2000;19:484a. 111. Curran WJ, Scott C, Langer C, et al. Long-term benefit is ob-­ served in a phase III comparison of sequential vs concurrent chemoradiation for patients with unresectable NSCLC: RTOG 9410 [abstract]. Proc Am Soc Clin Oncol 2003;22:621a. 112. Liao Z, Komaki R, Milas L, et al. A phase I clinical trial of tho-­ racic radiotherapy and concurrent celecoxib for patients with unfavorable performance status inoperable/unresectable nonsmall cell lung cancer. Clin Cancer Res 2005;11:3342-3348. 113. Wedge SR, Ogilvie DJ, Dukes M, et al. ZD6474 inhibits vascu-­ lar endothelial growth factor signaling, angiogenesis, and ­tumor growth following oral administration. Cancer Res 2002;62: 4645-4655. 114. Heymach JV, Johnson BE, Prager D, et al. A phase II trial of ZD6474 plus docetaxel in patients with previously treated NSCLC: follow-up results [abstract]. J Clin Oncol 2006;24 (Suppl):7016. 115. Herbst RS, Heymach JV, O’Reilly MS, et al. Vandetanib (ZD6474): an orally available receptor tyrosine kinase inhibitor that selectively targets pathways critical for tumor growth and angiogenesis. Expert Opin Investig Drugs 2007;16:239-249. 116. Frederick B, Gustafson D, Bianco C, et al. ZD6474, an inhibitor of VEGFR and EGFR tyrosine kinase activity in combination with radiotherapy. Int J Radiat Oncol Biol Phys 2006;64:33-37. 117. Nyati MK, Morgan MA, Feng FY, et al. Integration of EGFR in-­ hibitors with radiochemotherapy. Nat Rev Cancer 2006;6:876885. 118. Byers LA, Heymach JV. Dual targeting of the vascular endothe-­ lial growth factor and epidermal growth factor receptor path-­ ways: rationale and clinical applications for non-small-cell lung cancer. Clin Lung Cancer 2007;8(Suppl 2):S79-S85. 119. Shibuya K, Komaki R, Shintani T, et al. Targeted therapy against VEGFR and EGFR with ZD6474 enhances the therapeutic effi-­ cacy of irradiation in an orthotopic model of human non-small cell lung cancer. Int J Radiat Oncol Biol Phys 2007;69:1534-1543. 120. Gustafson DL, Frederick B, Merz AL, et al. Dose scheduling of the dual VEGFR and EGFR tyrosine kinase inhibitor vande-­ tanib (ZD6474, Zactima®) in combination with radiotherapy in EGFR-positive and EGFR-null human head and neck tumor xenografts. Cancer Chemother Pharmacol 2008;61:179-188. 121. Itasaka S, Komaki R, Herbst RS, et al. Endostatin improves ­radioresponse and blocks tumor revascularization after radiation therapy for A431 xenografts in mice. Int J Radiat Oncol Biol Phys 2007;67:870-878. 122. Cox JD, Stanley K, Petrovich Z, et al. Cranial irradiation in can-­ cer of the lung of all cell types. JAMA 1981;245:469-472. 123. Russell AH, Pajak TE, Selim HM, et al. Prophylactic cranial ir-­ radiation for lung cancer patients at high risk for development of cerebral metastasis: results of a prospective randomized trial conducted by the Radiation Therapy Oncology Group. Int J ­Radiat Oncol Biol Phys 1991;21:637-643. 124. Pignon JP, Arriagada R, Ihde DC, et al. A meta-analysis of tho-­ racic radiotherapy for small-cell lung cancer. N Engl J Med 1992;327:1618-1624. 125. Warde P, Payne D. Does thoracic irradiation improve survival and local control in limited-stage small-cell carcinoma of the lung? A meta-analysis. J Clin Oncol 1992;10:890-895. 126. Komaki R. Radiotherapy guidelines in small cell lung cancer. In Cox JD, Chang JY, Komaki R (eds): Image-guided Radiotherapy of Lung Cancer. New York: Informa Healthcare, 2008, pp 39-62. 127. Komaki R, Swann RS, Ettinger DS, et al. Phase I study of thorac-­ ic radiation dose escalation with concurrent chemotherapy for patients with limited small-cell lung cancer: report of Radiation Therapy Oncology Group (RTOG) protocol 97-12. Int J Radiat Oncol Biol Phys 2005;62:342-350. 128. Murray N, Coy P, Pater JL, et al. Importance of timing for tho-­ racic irradiation in the combined modality treatment of limitedstage small-cell lung cancer. The National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 1993;11:336-344.

19  The Lung, Pleura, and Thymus 129. Goto K, Nishiwaki Y, Takada M, et al. Final results of a phase III study of concurrent versus sequential thoracic radiotherapy (TRT) in combination with cisplatin (P) and etoposide (E) for limited-stage small cell lung cancer (LD-SCLC): the Japan Clin-­ ical Oncology Group (JCOG) study [abstract]. Proc Am Soc Clin Oncol 1999;18:468a. 130. McCracken JD, Janaki LM, Crowley JJ, et al. Concurrent chem-­ otherapy/radiotherapy for limited small-cell lung carcinoma: a Southwest Oncology Group Study. J Clin Oncol 1990;8: 892-898. 131. Turrisi AT III, Glover DJ. Thoracic radiotherapy variables: influ-­ ence on local control in small cell lung cancer limited disease. Int J Radiat Oncol Biol Phys 1990;19:1473-1479. 132. Turrisi A, Glover D, Mason B. Long-term results of platinum etoposide (PE) thoracic radiotherapy (TRT) for limited small cell lung cancer: results on 32 patients with 48-month minimum follow-up [abstract]. Proc Am Soc Clin Oncol 1992;11:975. 133. Turrisi AT III, Kim K, Blum R, et al. Twice-daily compared with once-daily thoracic radiotherapy in limited small-cell lung can-­ cer treated concurrently with cisplatin and etoposide. N Engl J Med 1999;340:265-271. 134. Livingston RB, Mira JG, Chen TT, et al. Combined modal-­ ity treatment of extensive small cell lung cancer: a Southwest ­Oncology Group study. J Clin Oncol 1984;2:585-590. 135. Komaki R, Byhardt RW, Anderson T, et al. What is the lowest effective biologic dose for prophylactic cranial irradiation? Am J Clin Oncol 1985;8:523-527. 136. Komaki R, Cox JD, Hartz AJ, et al. Characteristics of long-term survivors after treatment for inoperable carcinoma of the lung. Am J Clin Oncol 1985;8:362-370. 137. Cox JD, Komaki R, Byhardt RW, et al. Results of whole-brain irradiation for metastases from small cell carcinoma of the lung. Cancer Treat Rep 1980;64:957-961. 138. Baglan RJ, Marks JE. Comparison of symptomatic and prophy-­ lactic irradiation of brain metastases from oat cell carcinoma of the lung. Cancer 1981;47:41-45. 139. Rosen ST, Makuch RW, Lichter AS, et al. Role of prophylactic cranial irradiation in prevention of central nervous system me-­ tastases in small cell lung cancer. Potential benefit restricted to patients with complete response. Am J Med 1983;74:615-624. 140. Rosenstein M, Armstrong J, Kris M, et al. A reappraisal of the role of prophylactic cranial irradiation in limited small cell lung cancer. Int J Radiat Oncol Biol Phys 1992;24:43-48. 141. Aupérin A, Arriagada R, Pignon JP, et al. Prophylactic cranial irradiation for patients with small-cell lung cancer in complete remission. Prophylactic Cranial Irradiation Overview Collabora-­ tive Group. N Engl J Med 1999;341:476-484. 142. Johnson BE, Becker B, Goff WB II, et al. Neurologic, neuropsy-­ chologic, and computed cranial tomography scan abnormalities in 2- to 10-year survivors of small-cell lung cancer. J Clin Oncol 1985;3:1659-1667. 143. Chak LY, Zatz LM, Wasserstein P, et al. Neurologic dysfunc-­ tion in patients treated for small cell carcinoma of the lung: a clinical and radiological study. Int J Radiat Oncol Biol Phys 1986;12:385-389. 144. Komaki R, Meyers CA, Shin DM, et al. Evaluation of cognitive function in patients with limited small cell lung cancer prior to and shortly following prophylactic cranial irradiation. Int J ­Radiat Oncol Biol Phys 1995;33:179-182. 145. Lee JJ, Bekele BN, Zhou X, et al. Decision analysis for prophy-­ lactic cranial irradiation for patients with small-cell lung cancer. J Clin Oncol 2006;24:3597-3603. 146. Paulson DL. Carcinoma in the superior pulmonary sulcus. Ann Thorac Surg 1979;28:3-4. 147. Paulson DL. The survival rate in superior sulcus tumors treated by presurgical irradiation. JAMA 1966;196:342. 148. Komaki R, Roh J, Cox JD, et al. Superior sulcus tumors: results of irradiation of 36 patients. Cancer 1981;48:1563-1568. 149. Van Houtte P, MacLennan I, Poulter C, et al. External radiation in the management of superior sulcus tumor. Cancer 1984;54: 223-227. 150. Ahmad K, Fayos JV, Kirsh MM. Apical lung carcinoma. Cancer 1984;54:913-917.

457

151. Heelan RT, Demas BE, Caravelli JF, et al. Superior sulcus tumors: CT and MR imaging. Radiology 1989;170;(Pt 1): 637-641. 152. Komaki R, Derus SB, Perez-Tamayo C, Byhardt RW, et al. Brain metastasis in patients with superior sulcus tumors. Cancer 1987;59:1649-1653. 153. Rusch VW, Giroux DJ, Kraut MJ, et al. Induction ­chemoradiation and surgical resection for superior sulcus non-­small-cell lung carcinomas: long-term results of Southwest Oncology Group ­Trial 9416 (Intergroup Trial 0160). J Clin Oncol 2007;25: 313-318. 154. Kraut MJ, Rusch VW, Crowley JJ. Induction chemoradiation plus surgical resection is a feasible and highly effective treatment for Pancoast tumors: initial results of SWOG 9416 (Intergroup 0160) trial [abstract]. Proc Am Soc Clin Oncol 2000;19:487a. 155. Scarantino C, Salazar OM, Rubin P, et al. The optimum radia-­ tion schedule in treatment of superior vena caval obstruction: importance of 99mTc scintiangiograms. Int J Radiat Oncol Biol Phys 1979;5:1987-1995. 156. Slawson RG, Scott RM. Radiation therapy in bronchogenic car-­ cinoma. Radiology 1979;132:175-176. 157. Davenport D, Ferree C, Blake D, et al. Response of superior vena cava syndrome to radiation therapy. Cancer 1976;38:15771580. 158. Karnofsky DA, Burchenal JH. The clinical evaluation of chemo-­ therapeutic agents in cancer. In Macleod CM (ed): Evaluation of Chemotherapeutic Agents. New York: Columbia University Press, 1949, pp 191–205. 159. Bauer M, Birch R, Pajak TF. Prognostic factors in cancer of the lung. In Cox JD (ed): Lung Cancer: A Categorical Course in ­Radiation Therapy. Oak Brook, IL: Radiological Society of North America, 1985, pp 87-112. 160. Cohen MH. Is immediate radiation therapy indicated for ­patients with unresectable non-small cell lung cancer? No. Cancer Treat Rep 1983;67:333-336. 161. Cox JD, Barber-Derus S, Hartz AJ, et al. Is adenocarcinoma/ large cell carcinoma the most radiocurable type of cancer of the lung? Int J Radiat Oncol Biol Phys 1986;12:1801-1805. 162. Komaki R, Garden AS, Cundiff JH. Endobronchial Radiothera-­ py. Cambridge, MA: Blackwell Scientific, 1993. 163. Leibel SA, Guse C, Order SE, et al. Accelerated fractionation radiation therapy for liver metastases: selection of an optimal patient population for the evaluation of late hepatic injury in RTOG studies. Int J Radiat Oncol Biol Phys 1990;18:523-528. 164. Tong D, Gillick L, Hendrickson FR. The palliation of sympto-­ matic osseous metastases: final results of the Study by the Radia-­ tion Therapy Oncology Group. Cancer 1982;50:893-899. 165. Borgelt B, Gelber R, Kramer S, et al. The palliation of brain metastases: final results of the first two studies by the Radia-­ tion Therapy Oncology Group. Int J Radiat Oncol Biol Phys 1980;6:1-9. 166. Murphy KC, Feld R, Evans WK, et al. Intramedullary spinal cord metastases from small cell carcinoma of the lung. J Clin Oncol 1983;1:99-106. 167. Salyer WR, Eggleston JC. Thymoma: a clinical and pathological study of 65 cases. Cancer 1976;37:229-249. 168. Suster S, Rosai J. Thymic carcinoma. A clinicopathologic study of 60 cases. Cancer 1991;67:1025-1032. 169. Pollack A, el-Naggar AK, Cox JD, et al. Thymoma. The prog-­ nostic significance of flow cytometric DNA analysis. Cancer 1992;69:1702-1709. 170. Jose B, Yu AT, Morgan TF, et al. Malignant thymoma with extrathoracic metastasis: a case report and review of literature. J Surg Oncol 1980;15:259-263. 171. Wick MR, Scheithauer BW, Weiland LH, et al. Primary thymic carcinomas. Am J Surg Pathol 1982;6:613-630. 172. Bergh NP, Gatzinsky P, Larsson S, et al. Tumors of the thymus and thymic region. I. Clinicopathological studies on thymomas. Ann Thorac Surg 1978;25:91-98. 173. Masaoka A, Monden Y, Nakahara K, et al. Follow-up study of thymomas with special reference to their clinical stages. Cancer 1981;48:2485-2492.

458

PART 5  Thorax

174. Batata MA, Martini N, Huvos AG, et al. Thymomas: clinico-­ pathologic features, therapy, and prognosis. Cancer 1974;34: 389-396. 175. Nordstrom DG, Tewfik HH, Latourette HB. Thymoma: therapy and prognosis as related to operative staging. Int J Radiat Oncol Biol Phys 1979;5:2059-2062. 176. Cohen DJ, Ronnigen LD, Graeber GM, et al. Management of patients with malignant thymoma. J Thorac Cardiovasc Surg 1984;87:301-307. 177. Ariaratnam LS, Kalnicki S, Mincer F, et al. The management of malignant thymona with radiation therapy. Int J Radiat Oncol Biol Phys 1979;5:77-80. 178. Gerein AN, Srivastava SP, Burgess J. Thymoma: a ten year ­review.. Am J Surg 1978;136:49-53. 179. Penn CR, Hope-Stone HF. The role of radiotherapy in the man-­ agement of malignant thymoma. Br J Surg 1972;59:533-539. 180. Monden Y, Nakahara K, Nanjo S, et al. Invasive thymoma with myasthenia gravis. Cancer 1984;54:2513-2518. 181. Marks RD Jr, Wallace KM, Pettit HS. Radition therapy control of nine patients with malignant thymoma. Cancer 1978;41: 117-119. 182. Arriagada R, Bretel JJ, Caillaud JM, et al. Invasive carcinoma of the thymus A multicenter retrospective review of 56 cases. Eur J Cancer Clin Oncol 1984;20:69-74. 183. Pollack A, Komaki R, Cox JD, et al. Thymoma: treatment and prognosis. Int J Radiat Oncol Biol Phys 1992;23:1037-1043. 184. Krueger JB, Sagerman RH, King GA. Stage III thymoma: ­results of postoperative radiation therapy. Radiology 1988;168: 855-858. 185. Urgesi A, Monetti U, Rossi G, et al. Role of radiation therapy in locally advanced thymoma. Radiother Oncol 1990;19:273-280.

186. Curran WJ Jr., Kornstein MJ, Brooks JJ, et al. Invasive thymoma: the role of mediastinal irradiation following complete or incom-­ plete surgical resection. J Clin Oncol 1988;6:1722-1727. 187. Goldel N, Boning L, Fredrik A, et al. Chemotherapy of in-­ vasive thymoma. A retrospective study of 22 cases. Cancer 1989;63:1493-1500. 188. Shin DM, El-Naggar A, Putnam JB. Pathologic remission of invasive thymoma after induction chemotherapy. Cancer Bull 1992;44:346-348. 189. Uematsu M, Kondo M. A proposal for treatment of invasive thy-­ moma. Cancer 1986;58:1979-1984. 190. Chahinian AP, Bhardwaj S, Meyer RJ, et al. Treatment of in-­ vasive or metastatic thymoma: report of eleven cases. Cancer 1981;47:1752-1761. 191. Kersh CR, Eisert DR, Hazra TA. Malignant thymoma: role of ­ radiation therapy in management. Radiology 1985;156: 207-209. 192. Shamji F, Pearson FG, Todd TR, et al. Results of surgical treat-­ ment for thymoma. J Thorac Cardiovasc Surg 1984;87:43-47. 193. Verley JM, Hollmann KH. Thymoma: a comparative study of clinical stages, histologic features, and survival in 200 cases. ­Cancer 1985;55:1074-1086. 194. Wilkins EW Jr, Castleman B. Thymoma: a continuing survey at the Massachusetts General Hospital. Ann Thorac Surg 1979;28: 252-256. 195. Maggi G, Giaccone G, Donadio M, et al. Thymomas: a review of 169 cases, with particular reference to results of surgical treat-­ ment. Cancer 1986;58:765-776.

The Esophagus 20 Zhongxing Liao, Ritsuko Komaki, and James D. Cox ANATOMY RESPONSE OF THE NORMAL ESOPHAGUS TO IRRADIATION Acute Effects Late Effects EPIDEMIOLOGY OF ESOPHAGEAL CANCER CLINICAL EVOLUTION PRETREATMENT EVALUATION Esophagography Endoscopic Sonography Computed Tomography Positron Emission Tomography Bronchoscopy STAGE CLASSIFICATION TREATMENT Surgery

Radiation Therapy PALLIATION Adverse Effects of Treatment for Esophageal Cancer Prognostic Factors in Esophageal Cancer Involvement of Celiac Axis Nodes Pathologic Response Tumor Metabolic Activity on Position Emission Tomography Molecular Factors TECHNIQUES OF IRRADIATION Technical Advances Proton Therapy Target Volume Considerations SUMMARY

Anatomy The esophagus is a hollow, muscular tube that is approximately 25 cm long. It extends from the lower border of the cricoid cartilage at the level of vertebra C7 to the stomach. The esophagogastric junction is near the lower border of vertebra T11. Locations within the esophagus are often described in terms of their endoscopic distance from the upper incisor teeth. For classifying, staging, and reporting esophageal malignancies, the American Joint Committee on Cancer (AJCC)1 recommends dividing the esophagus into four ­ regions: the cervical esophagus, the intrathoracic/­ abdominal esophagus (the upper and midthoracic portions), and the lower thoracic/abdominal esophagus (Fig. 20-1). The cervical esophagus extends from the lower edge of the cricoid cartilage to the thoracic inlet, approximately  18 cm from the incisors. The intrathoracic esophagus consists of the upper thoracic esophagus, which extends from the thoracic ­inlet to the tracheal bifurcation (approximately 24 cm from the incisors), and the midthoracic esophagus, which lies between the tracheal bifurcation to the ­level of the distal esophagus just above the gastroesophageal junction (32 cm from the incisors). The most distal, lower thoracic/abdominal ­ esophagus is approximately  3 cm long and includes the gastroesophageal junction (40 to  43 cm from the ­incisors).2 The anterior surface of the intrathoracic esophagus is ­immediately adjacent to the trachea and left atrium (Fig. 20-2). The esophagus relates to the azygous vein on the right and the pleura on the left and is adjacent to the recurrent laryngeal nerve, common carotid artery, and subclavian artery. The cross-sectional anatomy

of the esophagus as seen on computed tomography (CT) is shown in Figure 20-3. The esophageal wall, normally less than 5 mm thick, has four layers: the innermost mucosa, the submucosa, the ­  muscularis propria, and the adventitia. The mucosal ­layer itself consists of three sublayers: the nonkeratinizing, ­ stratified squamous epithelium; the fibrovascular ­connective-­tissue lamina propria; and the smooth muscle cell–containing muscularis mucosa. The submucosal layer consists of loose connective tissue that contains vessels, nerve fibers, lymphatics, and submucosal glands. The muscularis propria consists of an inner circular layer and an outer longitudinal layer of skeletal muscle. Unlike the rest of the gastrointestinal tract, the esophagus has no continuous serosa; the outermost adventitial layer consists largely of loose connective tissue.2 The epithelium, submucosa, and muscularis propria form longitudinal folds. The blood supply to and from the esophagus is illustrated in Figure 20-4. The cervical esophagus is supplied ­mainly by branches of the inferior thyroid artery; the thoracic esophagus is supplied by branches of the bronchial arteries, intercostal arteries, and aorta; and the abdominal esophagus is supplied by branches of the left gastric and inferior phrenic arteries (see Fig. 20-4A). Branches from these arteries run within the muscularis propria and give rise to branches that course within the submucosa with extensive anastomoses.2 The venous system of the esophagus (see Fig. 20-4B) is distributed largely within the folds of the esophageal mucosa. The venous supply of the lower esophagus consists of four layers3 in which radially ­arranged epithelial channels communicate with the superficial venous plexus in the upper submucosa. The superficial venous plexus

461

462

PART 6  Gastrointestinal Tract

C15.0 18 cm

24 cm

Thoracic inlet C15.3 Tracheal bifurcartion C15.4

32 cm C15.5 40 cm

Oesophagogastric junction

FIGURE 20-1.  Anatomic regions of the hypopharynx, esophagus, and gastric cardia. (From Haustermans K, Lerut A. Esophageal tumors. In Brady L, Heilmann H, Molls M [eds]: Clinical Target Volumes in Conformal and Intensity-Modulated Radiation Therapy: A Clinical Guide to Cancer Treatment. New York: Springer, 2003, pp 107-119.)

then communicates with the deep intrinsic veins in the lower submucosa. From there, perforating veins pierce the muscularis propria and connect with the adventitial veins on the esophageal surface.2,3 Branches from the upper two thirds of the esophagus lead into the inferior thyroidal vein and the azygous system, eventually draining into the superior vena cava. The lower segment drains into the systemic circulation through the branches of the azygous vein, the left inferior phrenic vein, and the short gastric veins. The caval and portal systems communicate through the veins within the submucosa.2 The lymphatic networks extend from the submucosal and muscular layers of the esophagus and form three groups: the upper, middle, and lower lymphatic trunks (Fig. 20-5).4 All three groups of lymphatics drain to the paraesophageal lymph nodes located immediately adjacent to the esophagus. The upper lymphatic trunks, which drain the esophagus above the aortic arch, drain to the cervical and supraclavicular lymph nodes; the middle lymphatic trunks end in the posteromediastinal and retrotracheal lymph nodes; and the lower lymphatic trunks drain to the lymph nodes along the gastric cardia and the lesser curvature of the stomach. The extensive communication between the submucosal and muscular lymphatic networks is most marked and most complex in the midthoracic part of the esophagus. Drainage from the thoracic part of the esophagus may occur in a superior direction but usually occurs in a caudal direction. The mostly longitudinal orientation of the esophageal lymphatic system and the extensive communication between layers facilitate extensive tumor spread, both intramucosal and submucosal, beyond any grossly visible tumor. This characteristic becomes important when delineating the clinical target volume (CTV) for radiation treatment of esophageal cancer.

Response of the Normal Esophagus to Irradiation The esophageal epithelium, like that of the oral cavity and oropharynx, is moderately sensitive to radiation.5 The timing of onset of symptoms from irradiation is predictable, and the duration of symptoms varies with the intensity of the irradiation.

Acute Effects The acute reaction of the esophagus to irradiation consists of pseudomembranous inflammation. In rats, a single dose of 30 Gy produces submucosal edema at 4 days, basal epithelial cell death at 6 days, and denudation of the mucosa with a pseudomembrane consisting of necrotic debris by  10 days. Re-epithelialization begins by 14 days, and complete regeneration of the epithelium is usually seen by 20 days. In humans, the response of the esophagus to irradiation follows a similar pattern: odynophagia begins between  10 and 12 days after the first treatment, peaks in severity at 12 to 14 days, and then lessens over the next week. This decrease in symptoms while the treatment continues is an excellent example of accelerated proliferation in response to irradiation. Giving cytotoxic chemotherapy concurrently with ­irradiation can accelerate the onset of the esophageal ­reaction and profoundly increase its length and ­magnitude. Typically, concurrent radiation and ­chemotherapy causes esophagitis that increases over time and peaks at the end of the treatment6; clinical symptoms ­persist for another  1 to 2 weeks until normal tissues heal. ­Esophageal reactions are one of the most important ­dose-­limiting side effects in patients treated with combinations of ­ chemotherapy and radiation therapy for ­cancer of the thorax.7

20  The Esophagus

463

Lateral view Inferior pharyngeal constrictor muscle

C4 Thyroid cartilage Cricoid cartilage Trachea

C6

Cricopharyngeus muscle (part of inferior pharyngeal constrictor)

T1

Esophagus T3

Aorta T5

Sternum T7

Heart in pericardium

T9 Diaphragm T11

L1

L3

FIGURE 20-2.  Relative positions of the esophagus, heart, great vessels, trachea, and diaphragm. (Netter medical illustration, with permission from Elsevier.)

Late Effects The acute effects of esophageal irradiation usually resolve completely, and prolonged reactions leading to permanent effects are rare. Berthrong and Fajardo8 found histologic evidence of complete epithelial regeneration as early as  3 months after irradiation and as late as 2 years afterward. Late effects can include thickening of the epithelium, ­telangiectasis underlying an intact epithelium, and thickening of the tunica mucosa and surrounding connective tissue. Narrowing of the lumen resulting from submucosal scarring is uncommon.

Epidemiology of Esophageal Cancer Esophageal cancer is the eighth most common cancer worldwide.9 Overall incidence rates are twice as high in less-developed than in better-developed geographic regions, and the highest rates are found in Asia.9-11 The incidence and mortality rates for esophageal cancer are two to three times higher among men than among women.9,12 In the United States, cancer of the esophagus accounts for

less than 2% of all new cancer cases, although it causes 4% of cancer deaths among men.13 The two forms of esophageal cancer are squamous cell carcinoma of the esophagus and adenocarcinoma of the esophagus. Worldwide, the squamous cell type is the more prevalent of the two, and its incidence has remained high in less-developed countries.9,14-16 In the United States, the incidence of adenocarcinoma of the esophagus has risen steadily, whereas the incidence of squamous cell carcinoma has remained relatively constant or decreased slightly. The incidence of both types of cancer was nearly equivalent in North America and in The Netherlands in the early 2000s.17-19 Most adenocarcinomas manifest in the lower third of the esophagus. A high proportion of the population of the United States has symptomatic gastroesophageal reflux, and the relationship between that condition and adenocarcinoma of the esophagus is strong enough to constitute presumptive evidence that gastroesophageal reflux is a causative factor in this neoplastic disease.12,20-22 Long-standing gastroesophageal reflux leads to Barrett’s esophagus, a form of columnar epithelial metaplasia in which the squamous epithelium of the distal esophagus is gradually replaced by columnar epithelium like that in

464

PART 6  Gastrointestinal Tract

T E

A

A B

B

A

E

B

E

C

FIGURE 20-3.  CT scans of the esophagus and surrounding normal structures. A, Upper thoracic structures (approximately 15 cm from the upper incisors). B, Midthoracic structures at the level of the carina (approximately 25 cm from the upper incisors). C, Lower thoracic structures (approximately 36 cm from the upper incisors). A, aorta; B, bronchus; E, esophagus; T, trachea. (Courtesy of Ronelle A. DuBrow, MD, Houston, TX.)

the lining of the stomach. Barrett’s esophagus is considered a major contributor to the rising incidence of esophageal  adenocarcinoma. In the United States, the incidence and mortality rates for squamous cell carcinoma of the esophagus are markedly higher among black men than white men,12 whereas the incidence of adenocarcinoma is lower among black men than white men.18 In addition to having a higher frequency of squamous cell carcinoma, black patients also present with larger tumors and more weight loss.17,23 Age-adjusted death rates from esophageal cancer among men in the United States are 11.2 deaths per 100,000 African American men and 7.5 deaths per 100,000 white men.9,13 The major etiologic factors for squamous cell carcinoma seem to be alcohol consumption (especially whiskey and drinks with high concentrations of spirits) and tobacco use21,24-27; dietary factors28 and consumption of hot drinks also may contribute.29,30 The Plummer-Vinson syndrome has been associated with an increased risk for squamous cell carcinoma of the esophagus; the proposed mechanism is deficiency of iron and vitamins.31,32

Clinical Evolution Carcinoma of the esophagus usually manifests with ­ bstruction caused by a fungating mass that obliterates o the lumen. Less often, diffusely infiltrating tumors may cause stenosis. A compensatory dilatation proximal to the ­obstructing tumor is common. Because the esophagus lacks a serosa, tumors can readily spread into the ­adjacent mediastinum. Tumors of the middle third of the ­esophagus

may invade the left main stem bronchus, the ­thoracic duct, and even the aortic arch. Tumors of the lower third of the esophagus may invade the pericardium and ­ descending aorta. The rich interconnecting lymphatic network contributes to extension of disease superiorly and inferiorly well beyond the gross tumor. Although tumors of the upper third of the esophagus may spread to lymph nodes in the anterior jugular chain or the supraclavicular region, tumors of the middle and lower third preferentially spread to the subdiaphragmatic lymph nodes. Patterns of lymph node metastasis from esophageal cancer in a representative study are illustrated in Figure 20-6.33 Distant metastases are found in more than one half of patients at autopsy, most commonly in the lung and liver but also in the adrenal glands, bones, and brain.34 Progressive dysphagia is the most common early symptom, and weight loss is almost inevitable. Pain in the chest or abdomen is present in approximately one half of all patients. More advanced disease can result in cough, sialorrhea, and laryngeal paralysis from involvement of the recurrent laryngeal nerve. Progression of any symptom as reflected by deteriorating performance status scores and weight loss is associated with a worse prognosis. Carcinoma of the esophagus has a propensity to spread regionally and distantly. Table 20-1 shows a typical distribution of metastases to regional and distant sites.34 The pattern of distribution differs according to the level of the esophagus at which the carcinoma originated. In the case of carcinomas of the upper third of the esophagus, metastasis to the paratracheal and paraesophageal lymph nodes is most common; metastasis to the liver

20  The Esophagus

465

Arteries of esophagus Esophageal branch of inferior thyroid artery

Esophageal branch of Inferior thyroid artery Cervical part of esophagus Subclavian artery

Subclavian artery

Thyrocervical trunk

Common carotid artery

Internal thoracic artery

Brachiocephalic trunk

Vertebral artery

Trachea

Arch of aorta 3rd right posterior intercostal artery Right bronchial artery

Esophageal branch of right bronchial artery

Superior left bronchial artery

Thoracic (descending) aorta

Inferior left bronchial artery and esophageal branch

Esophageal branches of thoracic aorta

Abdominal part of esophagus Diaphragm

Thoracic part of esophagus

Stomach

Esophageal branch of left gastric artery

Inferior phrenic arteries Common hepatic artery (cut)

Left gastric artery Splenic artery (cut)

Celiac trunk

Common variations: Esophageal branches may originate from left inferior phrenic artery and/or directly from celiac trunk. Branches to abdominal esophagus may also come from splenic or short gastric arteries.

A FIGURE 20-4.  Illustrations of arteries (A) and veins

and lungs is less common. Middle-third lesions tend to spread to mediastinal and abdominal lymph nodes and more often involve the liver. Tumors arising in the lower third of the esophagus also have a high probability of spread to mediastinal and abdominal lymph nodes but involve the liver more often than the more proximal  lesions. Involvement of the lungs and pleura occurs with similar frequency regardless of the site of origin within the esophagus. Advanced cancer of the esophagus causes death through complications associated with the locoregional tumor—pneumonia associated with inanition, formation of tracheoesophageal fistulas, or erosion of the aorta. More effective control of locoregional tumors may be associated with a higher risk for death from distant metastasis. Failure-pattern studies show approximately equal risks of local failure and distant metastasis regardless of whether

the primary treatment is surgery35 or radiation therapy and chemotherapy.36

Pretreatment Evaluation For patients with esophageal cancer, a complete medical history and a physical examination with associated ­hematologic and biochemical studies are essential. Chest radiographs, CT of the chest and upper abdomen, ­positron emission tomography (PET) imaging using a fluorine  18 (18F) –tagged version of the glucose analogue fluorodeoxyglucose (18F-fluorodeoxyglucose [FDG-PET]), barium contrast imaging of the upper gastrointestinal tract, and endoscopic sonography contribute to the evaluation of the extent of disease. Esophagoscopy with biopsy is essential. Bronchoscopy is required for tumors of the upper third of the esophagus.

466

PART 6  Gastrointestinal Tract Veins of esophagus Internal jugular vein External jugular vein

Inferior thyroid vein Inferior Internal thyroid vein jugular vein

Vertebral vein

Subclavian vein

Subclavian vein

Thoracic duct

Right brachiocephalic vein

Left brachiocephalic vein

Superior vena cava

Left superior intercostal vein

Right superior intercostal vein

Submucous venous plexus

Esophageal veins (plexus)

Esophagus 6th right posterior intercostal vein

Accessory hemiazygos vein Venae comitantes of vagus nerve

Junction of hemiazygos and azygos veins

Hemiazygos vein Left inferior phrenic vein

Inferior vena cava (cut)

Short gastric veins

Diaphragm Hepatic vein

Left suprarenal vein

Inferior vena cava

Splenic vein

Liver Hepatic portal vein

Left renal vein

Left gastric vein

Omental (epiploic) veins

Right gastric vein Right renal vein

Left gastro-omental (gastroepiploic) vein Esophageal branches of left gastric vein Right gastro-omental (gastroepiploic) vein

Inferior mesenteric vein Superior mesenteric vein

B FIGURE 20-4.  cont’d  (B) associated with the esophagus. (Netter medical illustrations, with permission from Elsevier.)

Esophagography Double-contrast esophagography with fluoroscopy is ­important in localizing a suspected tumor and characterizing its intraluminal growth pattern (Fig. 20-7). It may be difficult to determine the length of the tumor by ­contrast esophagography because of the variable degree of ­obstruction.

Endoscopic Sonography Endoscopic sonography contributes to the evaluation of the endoluminal tumor, the area of involvement on the wall of the esophagus, and the immediately adjacent paraesophageal lymph nodes (Fig. 20-8). However, in approximately one third of patients, the transducer cannot be passed into the esophagus at the level of the tumor. Endoscopic sonography is superior to any imaging modality currently available for determining the depth of penetration of the tumor into or through the wall of the esophagus. Endoscopic sonography is also superior to other imaging modalities for identifying enlarged lymph nodes immediately adjacent to the esophagus. Use of endoscopic sonography to guide

fine-needle aspiration can increase the accuracy of nodal staging by providing cytologic confirmation of malignancy from peritumoral and celiac lymph nodes; evidence of such disease has a considerable effect on management and ­  outcome.37

Computed Tomography CT of the thorax is the diagnostic modality most often used for preclinical evaluation and target delineation. CT is particularly valuable for establishing the degree of ­extraesophageal extension of the tumor (Fig. 20-9). CT is more effective than endoscopic sonography in detecting more distant involved mediastinal lymph nodes, enlarged nodes in the celiac axis, and hepatic metastases. Endoscopic sonography and CT are complementary in assessing the ­ local and regional extensions of esophageal cancer. An asymmetric thickening of the esophageal wall on CT scans is the principal nonspecific finding of esophageal ­ carcinoma.38 The normal wall typically appears on 3 mm thick or less on CT scans; anything thicker than  5 mm is considered ­abnormal.39 The apparent thickening

20  The Esophagus

of the esophageal wall in combination with proximal dilation of the lumen from obstruction is a cardinal sign of an esophageal ­malignancy on CT. CT is not useful for establishing T status in esophageal cancer, although it can be used to exclude T4 cancers by visualizing the preservation of fat planes between an esophageal tumor and adjacent Jugulocervical Superior mediastinum Nodes on innominate artery Paraesophageal Paratracheal Nodes around ligamentum alteriosum Middle mediastinum Tracheobronchial Paraesophageal Pulmonary hilar Lower mediastinum Paraesophageal Diaphragmatic

Common hepatic artery

Splenic artery

FIGURE 20-5.  Lymph node groups that drain the esophagus. (Data from Thompson WM. Esophageal Cancer. Int J Radiat ­Oncol Biol Phys 1983;9:1533-1565, Akiyama H, Tsurumaru M, ­Kawamura T, et al. Principles of surgical treatment for carcinoma of the esophagus: analysis of lymph node involvement. Ann Surg 1981;194:438-446.) 26.7%

21.1%

13.3%

46.7%

s­ tructures.38 The cephalad extent of the tumor often can be identified by its interface with a dilated air- or fluidfilled proximal esophageal ­lumen, whereas the caudal tumor margin may be difficult to distinguish on CT scans; CT can underestimate tumor length by up to 3 cm compared with length as shown on a barium swallow test.40 Regional lymph nodes are readily visualized in paraesophageal and retroperitoneal fat by CT. Enlarged lymph nodes or clusters of multiple lymph nodes are considered abnormal. The length of the short axis of a lymph node usually determines whether that node is normal or enlarged. For intrathoracic or intra-abdominal nodes, the cutoff is 1 cm for enlarged margins, 0.5 cm for supraclavicular nodes, and 0.6 cm for posterior diaphragmatic (retrocrural) nodes.39 Accuracy rates for CT in detecting enlarged cervical, mediastinal, and abdominal nodes range from 61% to 90%; sensitivity rates range from 8% to 75%; and specificity rates range from 60% to 98%.41

Positron Emission Tomography

Superior gastric Paracardiac Lesser curvature Nodes on L gastric artery

Celiac axis

467

Use of FDG-PET is becoming increasingly more common in preclinical evaluations of suspected malignancies, including esophageal cancer. FDG-PET, which visualizes the uptake and decay of FDG by metabolically active tumors, is not helpful for identifying microscopic disease. However,  through tomographic imaging and coregistration with CT scans, whole-body PET scans can provide relatively detailed anatomic images (Fig. 20-10).38 Esophageal malignancies can be visualized by PET in 92% and 100% to cases.42,43 A meta-analysis of 12 studies showed FDG-PET to have moderate sensitivity and specificity for the detection of locoregional metastases (0.51 [95% confidence interval {CI}, 0.34 to 0.69] and 0.84 [95% CI, 0.76 to 0.91], 17.5%

16.7%

33.3%

42.1%

18.7%

18.4% 25.7% 16.1%

16.3% 28.6% 2.0%

34.1% 6.9% 5.0%

Upper esophageal cancer (n* = 15) (n = 49)

Middle esophageal cancer (n* = 57) (n = 261)

24.6% 66.4% 17.2% 5.3%

Lower esophageal cancer (n* = 24) (n = 134)

FIGURE 20-6.  Frequency of lymph node metastasis from esophageal cancer per number of cases resected according to tumor location (red ovals). n*, number of cases for analysis of cervical nodes. (From Akiyama H, Tsurumaru M, Kawamura T, et al. Principles of surgical treatment for carcinoma of the esophagus: analysis of lymph node involvement. Ann Surg 1981;194:438-446.)

468

PART 6  Gastrointestinal Tract

TABLE 20-1.   Anatomic Sites of Metastases from Esophageal Carcinoma in 67 Male Patients with Metastatic Disease Site

Number of Patients

Percentage (%)

Lymph nodes

58

73

Lung

41

52

Liver

37

47

Adrenals

16

20

Thyroid

5

6

Gastrointestinal serosa

5

6

Aorta

4

5

Peritoneum

4

5

Modified from Anderson L, Lad T. Autopsy findings in ­squamous-cell carcinoma of the esophagus. Cancer 1982;50:1587-1590.

18 CM

LN

FIGURE 20-8.  Endoscopic sonogram shows a thickened ­esophageal wall and an adjacent enlarged lymph node (LN). (Courtesy of Ronelle A. DuBrow, MD, Houston, TX.)

FIGURE 20-9.  CT scan of esophageal squamous cell carcinoma. The patient later developed a fistula between the esophagus and the left main stem of the bronchus during radiation therapy.

Bronchoscopy

FIGURE 20-7.  Esophagogram shows a typical fungating mass obliterating the lumen and proximal esophageal dilatation. (Courtesy of Ronelle A. DuBrow, MD, Houston, TX.)

respectively) and reasonable sensitivity and specificity for the detection of distant lymphatic or hematogenous metastases [0.67 (95% CI, 0.58 to 0.76) and 0.97 (95% CI, 0.90 to 1.0), respectively].44 Mounting evidence also supports the use of FDG-PET for planning radiation treatments45 and for predicting tumor response after chemotherapy or chemoradiation therapy (see Fig. 20-10).46-53 The use of FDG-PET findings for predicting survival in esophageal cancer is discussed further under “Tumor ­ Metabolic Activity on Positron Emission ­Tomography” later in this ­chapter.

Bronchoscopy is important as an adjunct to careful ­examination of the upper aerodigestive tract because of the ­relatively high risk of second primary malignant tumors associated with esophageal cancer. Thoracic esophageal ­tumors may involve the tracheal bronchial tree directly, and such involvement can lead to the ominous ­complication of fistula formation.

Stage Classification Earlier classifications of esophageal cancer were based on surgical findings alone. In contrast, the 1997 and 2002 classification systems by the AJCC were designed to reflect all regions of the esophagus and both clinical and pathologic findings. The clinical findings include those from endoscopic sonography and CT; pathologic staging is based on examination of the resected specimen. The 2002 AJCC  tumor-node-metastasis (TNM) classification is shown in Box 20-1.1 Although the stage classification is considered essential in reporting results of clinical investigations, another

20  The Esophagus

469

FIGURE 20-10.  Coregistered FDG-PET and CT scans demonstrate skip lesions in a 47-year-old woman with invasive squamous cell carcinoma of the cervical esophagus and adenocarcinoma of the midthoracic esophagus. Upper panels, transverse PET/CT images of the cervical (left) and midthoracic (right) tumors. Lower panels, coronal views show the normal esophagus between the two tumors.

i­mportant prognostic factor is weight loss. Patients who have lost more than 10% of their body weight are significantly more likely to die,35 regardless of whether treatment is by surgical resection or nonsurgical means. Ellis and colleagues54 also found that disease stage, particularly the number of involved lymph nodes (0, 1 to 4, or 5 or more), profoundly affected the outcome after surgical resection for esophageal cancer.

Treatment Until the early 1980s, the literature on treatment for ­esophageal cancer dealt predominantly with squamous cell carcinoma. When other histopathologic types were considered, they were often denoted as anaplastic rather than ­being identified as adenocarcinomas. Although many reports have been published on adenocarcinoma of the esophagogastric junction, these tumors are most appropriately ­ considered synonymous with distal esophageal ­adenocarcinomas ­unless they arise in the gastric cardia.55

Surgery Surgical resection continues to be the mainstay of treatment for carcinoma of the esophagus. An extensive review by Earlam and Cunha-Melo56 of the surgical literature ­published before 1980 concluded that the mortality rate for esophageal resection (29%) was the highest of the mortality rates for frequently performed surgical procedures. That analysis involved 122 reports including 83,783 patients. Surgical techniques have changed since that 1980 report, and a national Intergroup trial published in 1998 involving institutions throughout the United States and southern Canada revealed an operative mortality rate

of only 6%.35 The operative procedures currently in use ­include transhiatal esophagectomy; the Ivor-Lewis esophagectomy, which includes a right thoracotomy and a celiotomy; and left thoracotomy, which can be combined with a celiotomy (Table 20-2).57 The transhiatal esophagectomy has been popularized and widely used by Orringer.58 This procedure permits a subtotal esophagectomy with anastomosis in the cervical region and avoids the need for thoracotomy in most patients. Although Orringer’s indications for transhiatal esophagectomy include tumors of the entire esophagus and gastric cardia,57,58 many surgeons consider the transhiatal esophagectomy to be most applicable for lesions located below the carina. Thoracoabdominal approaches are preferred by many thoracic surgeons. Left thoracoabdominal approaches are used most often for distal esophageal lesions. If Barrett’s (intestinal) metaplasia extending superiorly into the middle third of the esophagus is evident, the Ivor-Lewis procedure is usually preferred. In a large Intergroup study reported by Kelsen and colleagues35 in which the specific operative procedure performed was left to the discretion of the participating surgeon, the overall operative mortality rate was 6%. The relative advantages and disadvantages of these various operative procedures have been summarized by Whyte and Orringer57 and by Hulscher and colleagues.59 Other factors that influence mortality rates for esophagectomy are the number of cases performed at a particular treatment center (i.e., hospital volume) and the experience of the surgeon performing them (i.e., case volume). Mortality rates at hospitals with high volumes of esophagectomies are usually less than 5%, although these rates tend to be higher at centers that perform fewer such operations.60-64  

470

PART 6  Gastrointestinal Tract

BOX 20-1.  American Joint Committee on Cancer Staging System for Esophageal Cancer 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 Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Regional lymph node metastasis Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Tumors of the Lower Thoracic Esophagus M1a Metastasis in celiac lymph nodes M1b Other distant metastasis Tumors of the Midthoracic Esophagus M1a Not applicable M1b Nonregional lymph nodes and/or other distant metastasis Tumors of the Upper Thoracic Esophagus M1a Metastasis in cervical nodes M1b Other distant metastasis Stage Grouping Stage 0 Tis Stage I T1 Stage IIA T2 T3 Stage IIB T1 T2 Stage III T3 T4 Stage IV Any T Stage IVA Any T Stage IVB Any T

N0 N0 N0 N0 N1 N1 N1 Any N Any N Any N Any N

M0 M0 M0 M0 M0 M0 M0 M0 M1 M1a M1b

From Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 91-98.

Several groups have found that both the experience of the surgeon and hospital volume were inversely correlated with adverse events, including mortality.65-67 This is especially true of complex procedures such as esophagectomy, for which differences in mortality rate can be as high as 10 to 13 percentage points. These differences may not reflect simply differences in technical skills or experience; active multidisciplinary, disease-focused clinics associated with high-volume practices also tend to be associated with improved outcomes.60,68 This finding was confirmed by a Patterns of Care Survey analysis that was reported by Suntharalingam and colleagues.69 They showed that patients with esophageal cancer treated from 1996 to 1999 at small centers were at a higher risk for death (hazard ratio = 1.32; P = .03) than were patients treated at larger institutions. Although long-term survival rates for selected patients have ranged from 10%56,70 to 40%,70 a large North American Intergroup trial35 showed the median survival time to be 16.1 months and the 3-year survival rate to be 26%. These results seem representative of the current state of the art with surgical resection alone. These less than satisfactory results have led to waning enthusiasm in many quarters for the use of surgical resection alone, and several studies have been conducted to test the effect of adding neoadjuvant (preoperative) chemotherapy. Two large, randomized studies of preoperative chemotherapy followed by surgery have shown conflicting results, with one finding no improvement (Fig. 20-11A)35 and the other showing improvements in overall and disease-free survival (see Fig. 20-11B).71 A meta-analysis of 11 randomized controlled  trials of chemotherapy plus surgery versus surgery alone for esophageal cancer72 showed somewhat mixed findings. In that study, patients who underwent combination therapy were more likely to have had a pathologically complete resection than those undergoing surgery alone (odds ratio = 0.71; 95% CI, 0.58 to 0.87; P = .001), but the combination offered no survival advantage. Possible reasons for this disappointing finding offered by the authors of this meta-analysis included the potential ineffectiveness of presurgical chemotherapy to eradicate micrometastases or to facilitate resection by locally downstaging tumors or, conversely, the effectiveness of the chemotherapy being offset by treatment-related mortality. Although the strategy of combining chemotherapy with surgery may not be

TABLE 20-2.    Techniques of Esophagectomy Location of ­Anastomosis

Approach

Location of Tumor

Advantages

Disadvantages

Transhiatal ­esophagectomy

Entire esophagus and gastric cardia

Cervical region

Low-morbidity ­cervical ­anastomosis, few ­respiratory side ­effects

Lack of exposure of ­middle esophagus

Right thoracotomy (Ivor-Lewis ­procedure)

Middle to upper ­esophagus

Thoracic or cervical region

Good exposure of middle esophagus

Intrathoracic ­anastomosis, ­thoracotomy

Left thoracotomy

Lower third of esophagus or gastric cardia

Low to middle ­thoracotomy

Good exposure of lower esophagus

Intrathoracic anastomosis, thoracotomy

Radical (en bloc) ­resection

Entire esophagus and gastric cardia

Cervical or thoracic region

More extensive ­lymphadenectomy, ­potentially better ­survival

Increased operative risk with no proven increase in survival

From Whyte R, Orringer M. Surgery for carcinoma of the esophagus: the case for transhiatal esophagectomy. Semin Radiat Oncol 1994; 4:146-156.

20  The Esophagus

ideal, these findings have raised considerable interest in ­using chemotherapy in combination with radiation therapy ­before ­surgical treatment.

­ iameter. However, for those with larger tumors, particud larly tumors considered unresectable, the results with radiation therapy alone were less favorable. Long-term results for the radiation-only treatment group in an Intergroup trial conducted by the Radiation Therapy Oncology Group (RTOG)36 showed that none of 60 patients treated with radiation therapy alone lived beyond 3 years.74

Radiation Therapy Radiation Therapy Alone In their comprehensive and critical review of 8500 patients with esophageal cancer treated with radiation therapy alone, Earlam and Cunha-Melo56 found a 5-year survival rate of 5%. Although patients with small (<5-cm) tumors might be candidates for resection, radiation therapy for some of these patients can also provide favorable results. For example, Yang and colleagues73 analyzed 1136 patients in China who lived 5 years or longer after radiation therapy. These investigators found a higher proportion of long-term survivors among patients with tumors less than 5 cm in

Postoperative Radiation Therapy Because 20% to 25% of recurrences after surgical ­treatment consist of local recurrence alone,35 postoperative irradiation theoretically could improve outcome. Postoperative irradiation has not been studied in comparative trials in the United States. However, three separate prospective comparative trials throughout the world showed no ­improvement in long-term survival with the addition of postoperative irradiation.75-77

80 60 Surgery (n = 234)

20 Chemotherapy plus surgery (n = 233)

0 0

1

2

3

4

100

Disease-free survival (%)

Overall survival (%)

100

40

80 Chemotherapy plus surgery (n = 213)

60 40 20

Surgery (n = 227)

0

0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

5

Years after randomization

Years after randomization No. of patients at risk

No. of patients at risk Chemotherapy plus surgery

136

73

42

28

15

Chemotherapy plus surgery

80

44

30

19

11

Surgery

138

81

45

27

16

Surgery

67

47

30

17

11

100

100

90

90

80

80

Disease-free survival (%)

A

Survival (%)

70 60 50 40

P = .004

30 20

CS S

10

70 60 50 40 30

P = .0014

20

CS S

10

0

0 0

1

2

3

4

5

0

Years

B

471

1

2

3

4

5

Years, from 6 months after randomization

Patients at risk (events)

Patients at risk (events)

CS 400 (164) 231 (73) 143 (26) 81 (13) 36 (2) 14

CS 400 (237)156 (43) 96 (16) 48 (4) 25

(2)

9

S

S

(2)

7

402 (185) 212 (76) 124 (32) 70 (18) 28 (5) 10

402 (279)120 (35) 74 (13) 36 (6) 17

FIGURE 20-11.  A, Kaplan-Meier curves show overall survival (left) and disease-free survival (right) among patients in a U.S. multi-institutional randomized trial of chemotherapy followed by surgery or surgery alone for localized esophageal cancer. Left, median survival time was 14.9 months for patients given chemotherapy before surgery and 16.1 months for those who underwent surgery only (P = .53, log-rank test). Right, disease-free survival time at a landmark date of 6 months after randomization did not differ between the two treatment groups (P = .50, log-rank test [Sposto’s modification]). B, Kaplan-Meier curves show overall survival from the date of randomization (left) and disease-free survival from 6 months after randomization (right) in the British Medical Research Council trial of chemotherapy followed by surgery or surgery alone for resectable esophageal cancer. (A, Kelsen DP, Ginsberg R, Pajak TF, et al. Chemotherapy followed by surgery compared with surgery alone for localized esophageal cancer. N Engl J Med 1998;339:19791984; �B, From Medical Research Council Oesophageal Cancer Working Party. Surgical resection with or without preoperative chemotherapy in oesophageal cancer: a randomised controlled trial. Lancet 2002;359:1727-1733.�)

472

PART 6  Gastrointestinal Tract

Preoperative Radiation Therapy In principle, preoperative radiation therapy could improve resectability, decrease the risk for regional or distant metastasis, and improve survival for patients with operable carcinoma of the esophagus. This approach has not been investigated in prospective comparative trials in the United States, and most of the available findings come from trials conducted in other countries. Those trials tended to have unusual fractionation schemes compared with those widely  used in the United States. An early trial78 from France enrolled a total of 124 patients, who were randomly assigned to preoperative radiation therapy with 39 to 45 Gy (average, 40 Gy) in 8 to 12 days or to immediate surgical exploration and resection if possible. The results showed no difference between groups in rates of resectability or operative mortality, the latter of which was 23% within 30 days for both groups. The median survival time was 12 months for the 57 patients who had surgery alone compared with 11 months for the 67 patients who had preoperative irradiation. No differences in 5-year survival rates were evident either; those rates were 11.5% for patients who had immediate operation and 9.5% for patients who underwent preoperative irradiation. Investigators from the European Organisation for Research and Treatment of Cancer (EORTC) also used a rapid fractionation approach in which 33 Gy was given in 10 fractions over 12 days followed by exploration and esophagectomy 8 days later.79 A total of 229 patients with potentially resectable esophageal carcinoma were randomly assigned to preoperative irradiation or immediate exploration and resection; 208 patients were evaluable. Five of the 102 irradiated patients did not undergo surgery; the operative mortality rate for the other 97 was 25%. The operative mortality rate among the 106 patients who underwent immediate exploration was 18%. The median survival time was 11 months for both groups in the study, and no difference was found between the two groups in the 5-year survival rate (10%). A third trial was reported by Wang and colleagues80 from China. Patients were randomly assigned to undergo preoperative irradiation with 40 Gy in 8 fractions or immediate resection. No difference was found in the local failure rate between the two treatment groups (50%). The 5-year survival rate was 30% for the preoperative irradiation group and 35% for the immediate surgery group. More conventional fractionation schedules have been studied in two other trials. Arnott and colleagues81 reported a randomized study of preoperative irradiation with 20 Gy given in 10 fractions over 2 weeks compared with surgical resection alone. Only 129 of the 176 patients who were randomized underwent radical resection. The 5-year survival rate for the 62 patients randomized to immediate surgery was 17% as compared with 9% for the 67 patients who received preoperative irradiation. The overall operative mortality rate for the 129 patients was 14%, with no difference between the two groups. The only trial to show an advantage for preoperative irradiation came from Scandinavia.82 In that study, patients were randomly assigned to preoperative irradiation with  35 Gy given in 20 fractions or surgery alone. The 3-year survival rate for the 41 patients who had surgery alone was 9% compared with 21% for the 48 patients who had preoperative radiation therapy (P = .08). This study included two

additional treatment groups: preoperative chemotherapy followed by surgery, and chemotherapy plus concurrent radiation (details on this approach are given under “Definitive Radiation Therapy with Concurrent Chemotherapy” later in this chapter). The 3-year survival rate was 19% for patients who received any radiation therapy and 6% for patients who received none (P = .0009). Taken together, the findings from these studies do not suggest a benefit for preoperative irradiation when it is used as the only adjuvant to surgery.

Preoperative Radiation Therapy with Concurrent Chemotherapy As has been the case with all single- and multiple-modality treatments for esophageal cancer studied, single-institution studies that have not included concurrent control groups have suggested benefit from concurrent chemotherapy and radiation therapy (i.e., chemoradiation therapy) before surgical resection (Table 20-3).82-90 Urba and colleagues84 reported findings from a randomized trial conducted between 1989 and 1994 at the University of Michigan. A total of 100 patients (25% had squamous cell carcinoma, and 75% had adenocarcinoma) were equally divided between surgery alone (transhiatal esophagectomy) and chemoradiation therapy before surgery (cisplatin, vinblastine, and fluorouracil [5-FU] plus radiation therapy with 1.5 Gy given twice daily for 5 days per week for 3 weeks). The median survival time was 17 months in both groups. A significant reduction in the rate of local failure was observed among those given chemoradiation therapy (19% versus 39% in the surgery-alone arm, P = .039). The 3-year survival rate was 32% for the combined-modality group and 15% for the surgery-alone group.84 Two other small studies that each included fewer than 100 patients provided no support for the use of chemoradiation therapy before surgical resection. Nygaard and colleagues82 randomly assigned patients with squamous cell carcinoma to receive cisplatin, bleomycin, and concurrent irradiation (35 Gy in 20 fractions) followed by resection or immediate resection alone. The 41 patients in the surgeryonly group had a 3-year survival rate of 9%, compared with 21% for the 47 patients in the combined-modality group (P = .3). Le Prise and colleagues85 randomly assigned 86 patients with squamous cell carcinoma of the esophagus to immediate resection or to preoperative cisplatin and 5-FU plus 20 Gy given in 10 fractions over 2 weeks. The median survival time was 12 months in both groups, and the 3-year survival rates were similar: 19% for combined-­modality therapy and 14% for surgery. The studies reported in which preoperative chemoradiation therapy was tested in multi-institutional trials have produced conflicting results. In the first of these studies, Walsh and colleagues from Ireland86 randomly assigned 113 patients with adenocarcinoma of the esophagus to surgical resection alone or preoperative 5-FU for 5 days beginning on day 1, cisplatin on day 7, and radiation therapy (40 Gy in 15 fractions) beginning on day 1. The chemotherapy was repeated during week 6, and surgery was carried out 8 weeks after the start of treatment. CT scans were obtained from some patients, but the selection was not done systematically. Abdominal sonography was performed for all patients. The median survival time for the

20  The Esophagus

473

TABLE 20-3.    Randomized Trials Comparing Preoperative Chemoradiation Therapy with Surgery No. of ­ Patients

Chemotherapy Drugs

Radiation Therapy Dose (Gy)

Median ­ urvival S Time (mo)

3-Year Survival Rate (%)

Surgery alone Preoperative RCT

41 47

C, B

35

NA NA

 9 17

NS

Le Prise et al, 199485

Surgery alone Preoperative RCT

41 45

C, F

20

10 10

14 19

NS

Apinop et al, 199489

Surgery alone Preoperative RCT

34 35

C, F

40

 7 10

20 26

NS

Walsh et al, 199686

Surgery alone Preoperative RCT

55 58

C, F

40

11 16

 6 32

<.01

Bosset et al, 199783

Surgery alone Preoperative RCT

139 143

C

37

19 19

37 39

NS

Law et al, 199790

Surgery alone Preoperative RCT

30 30

C, F

40

27 26

NA NA

NS

Urba et al, 199784

Surgery alone Preoperative RCT

50 50

C, F, V

45

18 17

16 30

.15

Burmeister et al, 200287

Surgery alone Preoperative RCT

128 128

C, F

35

22 19

NA NA

.38

Bosset et al, 200588

Surgery alone Preoperative RCT

380 planned 121 enrolled

C, F

45

NA

NA

Ongoing

Study and ­Reference Nygaard et al,

Therapy 199282

P Value

B, bleomycin; C, cisplatin; E, etoposide; F, 5-fluorouracil; L, leucovorin; NA, not available; NS, not significant; RCT, chemoradiation therapy; V, vinblastine.

1.0 0.9 Probability of survival

patients treated with multiple modalities was 16 months, compared with 11 months for the patients who had resection alone. In an intent-to-treat analysis, the 3-year survival rates were 32% for the induction chemoradiation ­therapy group and 6% for the resection-only group (P = .01)  (Fig. 20-12).86 A much larger study, conducted by a group of French investigators led by Bosset with support from the EORTC Data Center in Brussels,83 randomly assigned 282 patients with squamous cell carcinoma of the thoracic esophagus to undergo immediate resection or induction chemoradiation therapy followed by resection. Although esophageal sonography was not available, patients were evaluated with CT of the chest and upper abdomen plus sonography of the liver. The chemotherapy consisted of cisplatin alone 0 to 2 days before each course of radiation. Two courses of radiation were administered, each consisting of 18.5 Gy given in 5 fractions in 1 week, separated by 2 weeks. The total dose of radiation was 37 Gy in  10 fractions in 4 weeks with a 2-week interruption. The median survival time for both treatment groups was 18.6 months, and the numbers of people who survived for long ­periods were no different between the groups. However, local ­ tumor ­ control was ­ significantly better (P = .01) in

0.8 0.7 0.6

Multimodal therapy

0.5 0.4 0.3

P = .01 by the log-rank test

0.2 Surgery alone

0.1 0.0 0

10

20

30

40

50

60

Survival (months)

FIGURE 20-12.  Survival according to treatment (intent-to-treat analysis) in a trial of induction radiation therapy, cisplatin, and fluorouracil followed by surgery compared with surgery alone for adenocarcinoma of the esophagus. (From Walsh TN, Noonan N, Hollywood D, et al. A comparison of multimodal therapy and surgery for esophageal adenocarcinoma. N Engl J Med 1996;335:462-467.)

474

PART 6  Gastrointestinal Tract

the combined-modality group, as was disease-free survival  (P = .003) (Fig. 20-13).83 In a third study, this one from Australia, Burmeister and others87 evaluated 256 patients (157 with adenocarcinoma). Combined-modality treatment consisted of 35 Gy given over 3 weeks and one course of 5-FU plus cisplatin at week 1. The median survival times were 21.7 months in the combined-modality group and 18.5 months in the surgery-only group (P = .38). The French and Australian studies had more patients and better survival rates in the surgery-only control groups than did the Irish study.86 Nevertheless, in all three studies, the survival rates for patients treated with preoperative chemoradiation therapy plus surgery ranged from 25% to 35%,83,84,86 rates similar to those of 25% to 27% after chemoradiation therapy only in two other studies.91,92 The

100

Disease-free survival (%)

Combined treatment Surgery alone 80

discovery of a survival benefit from preoperative chemoradiation therapy plus surgery compared with surgery alone in a 2003 meta-analysis93 led to a clinical trial that is being conducted by the Fondation Française de Cancérologie Digestive (FFCD) and the EORTC to address the ­question of whether preoperative chemoradiation therapy plus surgery is superior to surgery alone.88 In that trial, patients with stage IIA or IIB esophageal cancer of squamous cell carcinoma or adenocarcinoma histology are randomly assigned to undergo preoperative chemoradiation therapy (45 Gy over 5 weeks and two courses of 5-FU and cisplatin) or surgery only. The trial was designed to enroll 380 patients to be able to detect an increase in the 3-year survival rate from 35% with surgery only to 50% with preoperative chemoradiation therapy. The trial has enrolled  121 patients.88 The fact that there was a significant difference in disease-free survival in the French trial83 supports the possible benefit of induction chemoradiation therapy. It is hoped that the findings from the ongoing FFCD-EORTC trial will clarify the value of preoperative chemoradiation therapy.

Three-Step Strategies

60 P = .003 by the log-rank test 40

20

0 0

1

2

3

4

5

6

Years after randomization

FIGURE 20-13.  Disease-free survival among patients with squamous cell carcinoma of the esophagus treated with induction radiation therapy and concurrent cisplatin followed by surgery or with surgery alone. Overall survival rates did not differ for the two treatment groups. (From Bosset JF, Gignou M, Triboulet JP, et al. Chemoradiotherapy followed by surgery compared with surgery alone in squamous-cell cancer of the esophagus. N Engl J Med 1997;337:161-167.)

Preoperative chemoradiation therapy seems to confer a benefit in terms of locoregional control over surgery only or chemoradiation therapy only, but 5-year overall survival rates are still low, ranging from 25% to 36% in randomized trials.83,84,86 The use of preoperative chemoradiation therapy for esophageal cancer seems to have shifted the pattern of recurrence from locoregional to distant metastasis.94 To overcome the problem of distant metastasis, a three-step strategy consisting of induction chemotherapy, preoperative chemoradiation therapy, and surgery was proposed. Several small, mostly single-institution experiences with this approach have been reported (Table 20-4).95-100 Jin and colleagues96 retrospectively studied patients who had been treated between January 1990 and December 1998 with additional systemic chemotherapy before preoperative chemoradiation therapy or with preoperative chemoradiation therapy alone for surgically resectable esophageal cancer. That analysis showed that patients who received induction chemotherapy before chemoradiation therapy had higher rates of complete pathologic response,

TABLE 20-4.    Results of Three-Step Treatment for Locally Advanced Esophageal Cancer Concurrent Chemo. Drugs

Radiation Dose (Gy)

Pathologic Complete Response Rate (%)

3-Year Overall Survival Rate (%)

5-Year Overall Survival Rate (%)

Median ­ urvival S Time (mo)

Study and Reference

No. of Patients

Induction Chemo. Drugs

Jin et al, 200496

42

F, C, Pac

F, C

45

37

74*

Bains et al, 200298

41

C, Pac

C, Pac

50.4

22

NA

NA

NA

Swisher et al, 200397

38

C, F, Pac

C, F

45

21

63

39

NA

NA

22.1

Ajani et al,

200495

43

Irin

F, Pac

50.4

28

42*

Ilson et al,

200399

19

C, Irin

C, Irin

50.4

27

NA

NA

25.6† 18.5‡

47

Doc, C, F

Doc, C, F

50

29.8

NA

NA

NA

Pasini et al, 2005100 *Rate

at 2 years those showing pathologic complete or partial response ‡For those showing less than pathologic partial response C, cisplatin; Chemo., chemotherapy; Doc, docetaxel; F, 5-fluorouracil; Irin, irinotecan; NA, not available; Pac, paclitaxel. †For

20  The Esophagus

locoregional control, distant-metastasis-free survival, ­disease-free survival, and overall survival. Other phase  I studies99-101 suggested that paclitaxel, docetaxel, and irinotecan, all of which increase cellular radiosensitivity by increasing the proportion of cells in the G2/M phase of the cell cycle and are active against metastatic disease, might be particularly promising for the three-step strategy. Swisher and colleagues from The University of Texas M. D. Anderson Cancer Center97 reported the long-term outcome of a phase II trial of induction chemotherapy, chemoradiation therapy, and surgery for 38 patients with locoregionally advanced esophageal cancer. Patients were initially given two cycles of paclitaxel (200 mg/m2),  5-FU (750 mg/m2/day for 5 days), and cisplatin (15 mg/ m2/day for 5 days). They then received chemoradiation therapy consisting of 45 Gy given over 5 weeks with concurrent 5-FU (300 mg/m2/day continuous infusion) and cisplatin (15 mg/m2/day for 5 days). Surgical resection was performed 4 to 6 weeks after completion of the chemoradiation therapy. Esophageal sonography revealed that most patients (87%) had T3 tumors and 66% had N1 disease; 84% had adenocarcinoma. Thirty-seven patients (97%) completed the planned induction chemotherapy and chemoradiation therapy, and 35 patients (92%) underwent surgery. The 30-day mortality rate after surgery was 6% (2 of 35 patients died). Twenty-five (71%) of the 35 patients who underwent surgery had a pathologic complete response or microscopic residual carcinoma (<10% viable cells) after chemotherapy, and a pathologic complete response was associated with disease-free survival rates of 72% at 3 years and 51% at 5 years. On the basis of an intention-to-treat analysis and a median potential follow-up of 58 months, the 3-year and 5-year overall survival rates for all 38 patients were 63% and 39%, ­respectively.97 The same group from M. D. Anderson Cancer Center95 also reported results from 43 patients with esophageal cancer who were given two cycles of induction chemotherapy with irinotecan and cisplatin followed by chemoradiation therapy consisting of 45 Gy concurrent with 5-FU and paclitaxel followed by surgery. With this strategy, 50% of patients experienced relief of dysphagia after the induction chemotherapy. Of the 43 patients, 39 (91%) underwent a complete resection; 2 patients died after surgery. The  2-year survival rate was 42%. Of the 25 patients who died of recurrent disease, 20 (80%) had distant recurrence.95 A randomized phase II trial comparing chemoradiation  therapy followed by esophagectomy with the three-step approach is being conducted. Similar results with a three-step strategy were reported by Bains and colleagues at Memorial Sloan-Kettering.98 That study involved treating 41 patients with two cycles of induction paclitaxel and cisplatin followed by 50.4 Gy of radiation combined with cisplatin and paclitaxel followed by surgery. Ninety-two percent of patients experienced improvement of dysphagia after induction chemotherapy, and only 5% of patients experienced grade 3 esophagitis. In summary, the addition of induction chemotherapy to preoperative chemoradiation therapy is a fairly recent ­innovation in the treatment of esophageal cancer. Early ­studies of this approach suggest that if the local effect of preoperative chemoradiation therapy is sufficiently ­pronounced, a survival benefit may be observed.

475

Brachytherapy Intraluminal insertion of radioactive isotopes for the treatment of esophageal cancer has a long history. Interest in this approach continues to be considerable, especially in countries such as Japan, where a high index of suspicion for this relatively common disease often leads to detection of superficial esophageal cancer at early stages. Larger and more infiltrating carcinomas cannot be ­adequately treated with brachytherapy alone but may be treated with brachytherapy in conjunction with ­external irradiation.102 However, this approach does carry some risk. The RTOG conducted a study of esophageal brachytherapy103 combined with external radiation and concurrent chemotherapy with cisplatin and 5-FU. The chemotherapy was given during weeks 1, 5, 8, and 11. The external radiation dose was 50 Gy, given in  25 fractions over 5 weeks beginning at the same time as the first ­ chemotherapy administration. Two weeks after the completion of external irradiation, esophageal brachytherapy was administered—high-dose-rate therapy with 5 Gy delivered in weeks 8, 9, and 10 for a total dose of  15 Gy or low-dose-rate therapy with 20 Gy delivered during week 8. More than 90% of the 49 patients treated had squamous cell carcinoma. Five patients (10%) died of complications of treatment, and another 12 patients (24%) experienced life-threatening toxic effects. Esophageal fistulas seemingly related to treatment occurred in 6 patients (12%) within 7 months of brachytherapy. ­Although brachytherapy may be indicated as the sole treatment for the unusual patient with superficial ­esophageal cancer, great caution must be exercised when esophageal brachytherapy is combined with external radiation and ­concurrent chemotherapy.

Definitive Radiation Therapy with Concurrent Chemotherapy The standard of care for patients with inoperable carcinoma of the esophagus is concurrent radiation therapy and chemotherapy. The most appropriate chemotherapy regimen is 5-FU and cisplatin unless the use of either drug is contraindicated. The basis for this treatment standard is a relatively small but highly significant Intergroup trial conducted by the RTOG (protocol 85-01).91 The rationale for this trial was the suggestion of improvement in local tumor control and survival in several studies of concurrent radiation therapy and chemotherapy reported in the early 1980s.104-106 The RTOG trial was designed to compare radiation therapy alone (total dose of 64 Gy in 32 fractions, with 5 fractions per week for 6.5 weeks) with concurrent chemotherapy and radiation therapy to a total dose of 50 Gy in 25 fractions, with 5 fractions per week. The chemotherapy consisted of 75 mg/m2 cisplatin given intravenously on the first day and 1 g/m2 5-FU given by continuous infusion for the first 4 days of weeks 1, 5, 8, and 11. Although 150 patients were to be enrolled in this study, a planned interim analysis was conducted when 90 patients were evaluable. This analysis showed a marked improvement in survival in the combined-modality group (P = .005) and led to cessation of randomization. Subsequently enrolled patients were allocated only to the combined-modality group. The original report of this trial included only the randomized patients (Fig. 20-14).36

476

100

100

75

75

50

Combined

25

Radiation

Percent alive

Survival (%)

PART 6  Gastrointestinal Tract

50

25

0 0

6

12

18

24

30

36

0

Months since randomization

1

2

3

4

5

Years from randomization

Patients at risk Combined therapy Radiation therapy

0

61

45

28

18

10

9

7

60

35

17

7

4

4

0

FIGURE 20-14.  Kaplan-Meier plot of survival of patients with esophageal carcinoma treated with radiation therapy alone or with radiation therapy and chemotherapy combined. Bars indicate 95% confidence intervals at 24 months. (From Herskovic A, Martz K, Al-Sarraf M, et al. Combined chemotherapy and radiotherapy compared with radiotherapy alone in patients with cancer of the esophagus. N Engl J Med 1992;326:1593-1598.)

It was evident that combined-modality therapy produced improved local control and freedom from distant metastasis. The rate of local tumor persistence or recurrence was reduced from 67% to 50%. The long-term results74,91 showed that 26% of the patients treated with combined-modality therapy survived for 5 years. The outcomes for the 69 nonrandomized patients enrolled in the combined-­modality group were quite similar to those of the 61 patients treated on the randomized combined­modality group (Fig. 20-15),74 providing corroboration of the results. None of the patients in the randomized combined­-modality group had died of esophageal cancer after 5 years, indicating that these patients with esophageal cancer were cured.91 Although the number of patients with adenocarcinoma treated with combined-modality therapy in this ­ series was small (23 [18%] of 130), no significant difference was evident in outcome between these patients and those with squamous cell carcinoma. Several additional studies support the value of definitive radiation therapy and concurrent chemotherapy for patients with inoperable esophageal cancer. Results of a study begun in 1980 at the Fox Chase Cancer Center that used concurrent 5-FU, mitomycin-C, and radiation therapy with 60 Gy in 6 to 7 weeks were similar to those of the RTOG study. Coia and colleagues106,107 reported 3- and 5-year survival rates of 29% and 18%, respectively, for 57 patients with clinical stage I or stage II tumors. On the basis of the Fox Chase Cancer Center study, the Eastern Cooperative Oncology Group (ECOG) conducted a randomized trial between 1982 and 1988.108 In this trial, patients were randomly assigned to concurrent 5-FU, mitomycin, and radiation therapy with 60 Gy in 6 to 7 weeks or radiation therapy alone. An option for esophagectomy after a total dose of 40 Gy was available. Survival for the 59 evaluable patients treated with combined modalities was superior to that for the 60 patients who received radiation 

FIGURE 20-15.  Long-term survival among patients with esophageal cancer treated with radiation therapy alone (blue line; median survival, 9.3 months) or radiation therapy with concurrent cisplatin and fluorouracil. When an interim analysis showed a marked improvement in survival for the patients randomized to receive chemoradiation (red line; median survival, 17.2 months), all of the remaining patients enrolled in the trial were treated with the combined-modality therapy (purple line; median survival, 14.1 months). (Modified from Al-Sarraf M, Martz K, Herskovic A, et al. Progress report of combined chemoradiotherapy versus radiotherapy alone in patients with esophageal cancer: an intergroup study. J Clin Oncol 1997;15:277-284.)

therapy alone (median survival, 14.8 months versus 9.2 months; 2-year survival, 27% versus 12%), but the uncontrolled use of resection confounds interpretation of the results (Fig. 20-16).108 No difference was evident in median survival between the patients who underwent resection and those who received chemotherapy and radiation therapy without esophagectomy. Richmond and colleagues109 also provided information about the contribution of esophagectomy to outcome ­after chemotherapy and radiation therapy. They reported a nonrandomized comparison of radiation therapy alone ­compared with concurrent radiation therapy, 5-FU, and cisplatin with or without surgical resection. No patient ­treated  with radiation therapy alone lived for 2 years, whereas 37% of patients treated with combined modalities lived for 2 years. No significant difference was evident in survival ­between patients treated with chemotherapy and radiation therapy and patients treated with chemotherapy, radiation therapy, and resection. These findings suggested that outcome in patients with esophageal cancer could be improved further by adding induction chemotherapy before chemoradiation therapy in an effort to further reduce the incidence of distant metastasis or by increasing the total dose of radiation therapy to more than 50 Gy in an effort to reduce the incidence of local failure. The former approach was studied in Intergroup trial 0122 (ECOG trial PE-289; RTOG trial 9012).110 Patients received three monthly cycles of cisplatin (day 1) and 5-FU (days 1 to 5) followed by concurrent cisplatin, 5-FU, and radiation therapy. The results in the 38 evaluable patients did not suggest improvement over concurrent chemoradiation therapy alone, and 9% of patients died of complications of treatment. A randomized Intergroup trial (INT 0123) comparing a higher total dose of radiation therapy (64 Gy) to the now standard total dose of 50 Gy with concurrent cisplatin and 5-FU was terminated

Survival probability

20  The Esophagus 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0

Treatment Total Dead Alive Median RT alone 60 58 2 0.7 RT + chemo 59 54 5 1.2

0

1

2

3

4

5

6

7

8

9

10 11 12

Survival year

FIGURE 20-16.  Survival in the Eastern Cooperative Oncology Group trial EST-1282 according to assigned treatment group (radiation therapy alone versus radiation therapy and concurrent mitomycin-C and fluorouracil). These data represent eligible patients only. (From Smith TJ, Ryan LM, Douglass HO Jr, et al. Combined chemoradiotherapy vs. radiotherapy alone for early stage squamous cell carcinoma of the esophagus: a study of the Eastern Cooperative Oncology Group. Int J Radiat Oncol Biol Phys 1998;42:269-276.)

early when it became evident that no improvement was provided by the higher total dose.111 Evidence supports the idea that concurrent chemotherapy with radiation therapy is superior to radiation therapy alone. The important question of whether concurrent therapy is equal to or better than surgery is discussed in the next section.

Definitive Radiation Therapy and Concurrent Chemotherapy plus Surgery The fact that more than 50% of the patients in the RTOG 85-01 and 95-04 trials experienced locoregional failure after treatment was evidence of the need for better locoregional disease control. Several groups have tried to improve outcomes with concurrent radiation and chemotherapy followed by surgery, with the rationale that surgical resection would remove any residual disease and thereby increase locoregional control. The evidence is conflicting, but the results of the studies done so far suggest that adding surgery after concurrent radiation therapy and chemotherapy may improve local control but probably does not improve overall survival. In one retrospective study, esophagectomy was associated with higher locoregional control, diseasefree survival, distant metastasis-free survival, and overall survival112; however, no difference in survival was demonstrated in a Pattern of Care Survey study of patients with locally advanced esophageal cancer given definitive radiation therapy and concurrent chemotherapy.113 Findings from two randomized phase III trials from Europe comparing chemoradiation therapy alone with chemoradiation therapy followed by esophagectomy were reported.114,115 The FFCD 9102 trial and a trial by the German Esophageal Cancer Study Group enrolled patients with locally advanced but resectable esophageal cancer. The hypothesis was that the two treatments were equivalent in terms of overall survival, and the results of both trials supported this hypothesis. The FFCD 9102 trial114 included patients with thoracic epidermoid or glandular T3-4 N0-1 M0 esophageal ­cancer.

477

Patients underwent induction chemoradiation therapy consisting of two 3-week cycles of 5-FU and cisplatin (administered on days 1 through 5 and 22 through 26) plus radiation therapy, either protracted (46 Gy over 4.5 weeks) or split-course (two 15-Gy fractions on days 1 through 5 and 22 through 26). Patients who experienced at least a partial response to chemoradiation therapy and had no contraindications to subsequent esophagectomy or chemoradiation therapy were randomly assigned to undergo surgery or additional chemoradiation therapy (three cycles of 5-FU and cisplatin plus protracted [20 Gy] or split-course [15 Gy] radiation therapy). Between January 1993 and November 2000, 455 eligible patients had started induction chemoradiation therapy, 259 of whom were randomly assigned to surgery or further chemoradiation therapy. No differences were found between the two groups in the 2year survival rate (34% for surgery versus 40% for definitive chemoradiation therapy, adjusted odds ratio = 0.91;  P = .56) or median survival time (17.7 months for surgery versus 19.3 months for definitive chemoradiation therapy). Among those who underwent surgery, the mortality rate within 3 months after the start of induction chemoradiation therapy was somewhat higher than that in a previous report35 (9% versus 1%, P = .002). Patients in the definitive chemoradiation therapy group had shorter hospital stays and better performance status, but they needed esophageal dilation and stent placement more often than did patients in the surgery group. The results from this study support the contention that patients with locally advanced but operable esophageal cancer that responds to induction chemoradiation therapy have the option of continuing the chemoradiation therapy as an alternative to surgery and that continued chemoradiation therapy is associated with equivalent overall survival, a lower early mortality rate, shorter hospital stays, and equivalent or better posttreatment performance status, albeit with a more frequent need for palliative procedures. Shortcomings of this trial were its short follow-up time, unconventional radiation schema, and accrual of some patients at centers with a low surgical volume, which could have contributed to higher operative morbidity and mortality. The German Esophageal Cancer Study Group trial115 included only patients with high-risk disease (i.e., sonographically staged T3-4 N0-1 M0 squamous cell carcinoma of the esophagus). Patients were stratified according to treatment center, tumor and lymph node status, completeness of endoscopic sonography, sex, and extent of weight loss within the 8 weeks before randomization.115 Of 189 patients from 11 German institutions enrolled, 5 were found to be ineligible and 12 refused randomization, leaving 172 patients for analysis. Patients were randomly assigned to receive three cycles of 5-FU, leucovorin, etoposide, and cisplatin, followed by chemoradiation therapy (cisplatin and etoposide and 40 Gy) and then transthoracic esophagectomy, with two-field lymphadenectomy as the standard procedure, or to receive the same chemotherapy followed by definitive chemoradiation therapy (cisplatin and etoposide and radiation to a dose of at least 65 Gy). Treatment was completed by 62% of patients in the surgery group and 85% of patients in the definitive chemoradiation therapy group. Treatment-related mortality rates were 10% in the surgery group and 3.5% in the definitive chemoradiation therapy group. The postoperative mortality rate ­decreased

478

PART 6  Gastrointestinal Tract

from 10.7% to 4.2% during the last 3 years of the study as surgeons gained experience. The median observation time was 6 years. Analysis of the 172 eligible, randomly assigned patients (86 per group) showed that overall survival was equivalent in the two treatment groups (log-rank test for equivalence, P < .05). The 2-year local-progression–free survival rate was better in the surgery group (64.3%; 95% CI, 52.1% to 76.5%) than in the definitive chemoradiation therapy group (40.7%; 95% CI, 28.9% to 52.5%; the hazard ratio for definitive chemoradiation therapy versus surgery was 2.1 [95% CI, 1.3 to 3.5; P < .003). The treatment-related mortality rate was significantly higher in the surgery group (12.8% versus 3.5%, P < .03).115 In the German trial, clinical tumor response to induction chemotherapy was the single independent prognostic factor for overall survival (hazard ratio = 0.30; 95% CI, 0.19 to 0.47; P = .0001)—a means of identifying patients with favorable prognosis despite the high-risk disease. Complete pathologic response to induction chemotherapy was experienced by 35% of the patients who underwent esophagectomy. A trend toward improved local tumor control was observed in the surgery group: local tumor progression was the first mode of treatment failure in 64% of patients in this group compared with 81% of patients in the definitive chemoradiation therapy group (P = .08). No between-group differences were found in median survival time (16 months versus 15 months) or 3-year survival rate (28% versus 20%) (log-rank P =.22). The 3-year survival rates among patients who responded to induction chemotherapy were 45% in the surgery group and 44% in the definitive chemoradiation therapy group, and the 3-year survival rates among patients who did not respond to induction chemotherapy were 18% in the surgery group and 11% in the definitive chemoradiation therapy group. The 3-year survival rate for nonresponders who underwent complete tumor resection after chemoradiation therapy was 35%.115 The results from these two trials suggest that adding surgery to chemoradiation therapy improved local tumor control but did not improve survival among patients with locally advanced squamous cell carcinoma of the ­esophagus.115 A meta-analysis of six randomized, controlled trials of the value of chemoradiation therapy plus surgery versus surgery alone116 found that the combined-modality treatment reduced 3-year mortality compared with surgery alone (odds ratio = 0.53; 95% CI, 0.31 to 0.93; P = .03), but at the cost of increased postoperative mortality (odds ratio = 2.10; 95% CI, 1.18 to 3.73; P = .01). Chemoradiation therapy was also found to reduce the pathologic disease stage at surgery compared with surgery alone (odds ratio = 0.43; 95% CI, 0.26 to 0.72; P = .001). A Patterns of Care Survey of 59 U.S. institutions69 confirmed the potential value of chemoradiation therapy followed by surgery, revealing a decreased chance of death for patients undergoing presurgical chemoradiation therapy compared with chemotherapy alone (hazard ratio = 0.32; P < .0001). Additional randomized trials are required to resolve the differences in these study findings. In summary, definitive chemoradiation therapy seems to result in long-term overall survival that is equivalent to that in contemporary surgical series and is therefore an acceptable alternative to surgery.

Palliation Another benefit of concurrent chemotherapy and radiation therapy in esophageal cancer is its ability to induce rapid and long-term palliation of dysphagia. Whether undertaken as a preoperative measure or as definitive treatment, concurrent therapy achieves symptomatic benefit for most patients. Coia and colleagues106,107 reported improvement in swallowing for 90% of patients treated in their pilot study. Patients had improvement within 2 weeks. Moreover, patients with stage I and II tumors had durable palliation. Of the 25 patients who were free of disease more than 1 year after treatment, 68% were asymptomatic, 25% had dysphagia only with solid foods, and 12% had some dysphagia with soft foods. Most patients with esophageal carcinoma have distressing symptoms ranging from slight to marked dysphagia and odynophagia to complete obstruction of the esophagus with inability to handle natural secretions or intractable cough from tracheoesophageal fistula. The most common treatment in the past was dilatation and placement of a Celestin or Atkinson intraluminal tube. Among patients who are to be treated with curative intent, the definitive treatment, whether esophagectomy or concurrent radiation therapy and chemotherapy, also produces palliation. Among patients for whom definitive treatment with curative intent is not possible, palliation can be achieved with radiation therapy in approximately 75%.117 Although a wide variety of procedures, including surgical resection, dilation, laser debulking, intraluminal stent or tube placement, and endoesophageal brachytherapy, can produce relief of symptoms,118 external radiation therapy is the least invasive and usually produces some benefit within 2 weeks. It is better to pursue immediate irradiation for palliation than to await the development of more severe symptoms, because immediate irradiation produces a greater and more lasting palliation than the alternatives. The single relative contraindication to external irradiation is fistula formation between the esophagus and the tracheobronchial tree. Irradiation can increase symptoms from fistulas when tumor regression leads to a larger communication between the esophagus and respiratory structures. Placement of an intraluminal tube followed by external irradiation may allow healing, but results are not predictable.

Adverse Effects of Treatment for Esophageal Cancer Preoperative chemoradiation therapy has toxic effects that can render patients unable to undergo surgery and increase the risk for perioperative death. In current clinical practice, postoperative mortality rates after esophageal resection in high-volume treatment centers range from about 2% to 5%.68,119 Typical postoperative mortality rates after preoperative chemoradiation therapy and esophagectomy range from 2% to 8%.120-122 The toxicity of preoperative chemoradiation therapy results from its biologic effects on rapidly dividing cells both within the irradiated volume (e.g., esophagogastric mucosa, lung) and outside the irradiated volume (e.g., bone marrow). The severity of the toxic effects depends on the volume of critical tissues irradiated, the dose and fractionation of the radiation therapy (accelerated schemes are more toxic than conventional ones), and the choice of chemotherapeutic agents, including dose and method of delivery. Acute toxic effects that result in

20  The Esophagus

Lung Preoperative chemoradiation therapy followed by esophagectomy has been associated with high rates of morbidity and mortality from pulmonary complications, with deaths attributed to sepsis, pneumonia, and acute respiratory distress syndrome.124 Pulmonary complications are the most common serious adverse event after esophagectomy and the leading cause of postoperative mortality among patients treated with surgery for esophageal cancer. The incidence of pulmonary complications after esophagectomy can exceed 50%,83,124 and the associated in-hospital mortality rate in patients with pulmonary complications is 55%.125 Poor performance status, poor lung function, and advanced age have been identified as risk factors for poor outcomes after surgery,125-127 as have long duration of surgery and excessive blood loss.128 Excessive radiation doses or unconventional fractionation schemes may account for the higher mortality rates observed in selected series.129-131 In one study in which patients received unconventional split-course radiation therapy in relatively large fractions of 3.5 Gy, 11 of 18 patients who survived surgical resection experienced major radiation-induced pleural and pericardial complications.83   A randomized trial of preoperative chemoradiation therapy for esophageal squamous cell cancer83 showed increased disease-free survival but no overall survival benefit, probably because of an excessive number of postoperative deaths, which usually resulted from pulmonary complications, in those given preoperative chemoradiation therapy. A contributing factor might have been the unconventional radiation therapy scheme, which included fractional doses of 3.7 Gy. Postoperative mortality was also related to the number of patients treated at each participating center, with lower patient volumes associated with higher postoperative mortality rates.83 Lee and colleagues132 were the first to report that postoperative pulmonary complications increased significantly when more than 40% of the lung received radiation doses higher than 10 Gy. Postoperative pulmonary complications, defined as pneumonia or adult respiratory distress syndrome that developed within 30 days after surgery and before discharge,133 was the clinical end point in that study. Additional evidence from another retrospective study of a larger sample that included the same patients, reported by Wang and others,134 suggested that the volume of lung spared from doses of 5 Gy or more was the only independent predictive factor associated with pulmonary complications in

0.6 Incidence of pulmonary complications

treatment interruptions can compromise the efficacy of preoperative chemoradiation therapy. Late-responding tissues with slowly dividing cells (e.g., connective tissues, capillary endothelium) can exhibit toxic effects months or years after the completion of treatment. Late toxic effects also depend on the radiation dose, the fractional dose (risk is increased for doses larger than 2 Gy), and the choice of chemotherapeutic agent.123 A meta-analysis116 showed that postoperative mortality rates were higher in patients treated with chemoradiation therapy plus surgery than in those treated with surgery alone, although that conclusion followed mainly from the higher mortality in the chemoradiation therapy group of the largest randomized trial in the analysis, in which both high radiation dose fractions and high cisplatin doses were used.83

479

0.5 0.4 0.3 0.2 0.1 0 0

1000

2000

3000

4000

5000

Volume of lung exposed to <5 Gy (cc)

FIGURE 20-17.  Incidence of postoperative pulmonary complications versus lung volume spared from doses of 5 Gy or higher (VS5). The blue curve represents the fit of the logistic model to the data (incidence of pulmonary complications versus VS5). The solid circles represent the observed incidence of pulmonary complications in each of five equal (n = 22) patient subgroups plotted at the mean value of VS5 for the group. The horizontal error bars represent ± 1 standard deviation of the mean VS5 in each group. The vertical error bars represent ± 1 standard deviation calculated from the observed incidence assuming binomial statistics. (From Wang SL, Liao Z, Vaporciyin AA, et al. Investigation of clinical and dosimetric factors associated with postoperative pulmonary complications in esophageal cancer patients treated with concurrent chemoradiotherapy followed by surgery. Int J Radiat Oncol Biol Phys 2006;64:692-699.)

patients with esophageal cancer treated with chemoradiation therapy followed by surgery—the smaller the volume of lung spared from doses of 5 Gy or higher, the higher the incidence of postoperative pulmonary complications (Fig. 20-17).134 Moreover, this conclusion was reached with an alpha/beta ratio of 10 Gy, which was selected because surgical pulmonary complications were considered an acute event (unlike radiation pneumonitis, for which an alpha/ beta ratio of 3 Gy is often used). The same conclusion was reached when lung dose-volume histograms were computed  with an alpha/beta ratio of 3 Gy or direct physical dose, indicating that the results were insensitive to the use of the alpha/beta ratios and the use of biologically equivalent or physical doses. Further analyses of that data set with a normal tissue probability model135 confirmed that the risk for postoperative pulmonary complications was strongly associated with small absolute volumes of lung spared from doses of at least 5 Gy. However, bootstrap analyses suggested that other factors in addition to the volume of lung receiving 5 Gy or more, such as mean lung dose and effective dose, were also linked with the probability of postoperative pulmonary complications.135 Given the shortcomings of the latter study (retrospective, with small numbers of patients and events), these conclusions need to be substantiated with additional, preferably prospectively collected data.

Esophagus Accurate assessment of the incidence and severity of esophagitis in patients with esophageal cancer is difficult because the symptoms that are used to define esophagitis

480

PART 6  Gastrointestinal Tract

TABLE 20-5.    Prognostic Factors for Overall Survival in Esophageal Cancer Prognostic Factors and Reference

3-Year Overall Survival Rates

P Value

Negative: 54% to 56% Positive: 12%

.03

Number of positive nodes after chemoradiation143

0-1 positive node: 5-yr OS of 34% to 38% 2 or more positive nodes: 5-yr OS of 6%

.02

Standardized uptake values on FDG-PET after ­chemoradiation53

SUV < 4: 18-month survival 77% SUV ≥ 4: 18-month survival 34%

.01

Extent of disease resection144

R0: 59% R1: 37% R2: 4%

<.001

Complete pathologic response142

0% residual: OS 74% 1% to 50% residual: OS 54% >50% residual: OS 24%

<.001

Clinical tumor response to induction chemotherapy115

50% vs. 17.9%

<.0001

Celiac node status before

treatment141

FDG-PET, 18F-fluorodeoxyglucose positron emission tomography; OS, overall survival; SUV, standardized uptake values.

(e.g., dysphagia, odynophagia) are nonspecific and can be caused by the disease or by the treatment. Nevertheless, it is clear that 5-FU–based chemotherapy regimens are associated with adverse effects on the esophagus. Several independent studies have shown that severe esophagitis occurs in up to 80% of patients treated with 5-FU and that esophagitis necessitating nutritional support measures and unplanned hospitalizations occur in about one half of all patients treated with 5-FU.88,121,122,136 Substituting pacli­ taxel for 5-FU may reduce the rates of esophagitis but does not affect rates of complete resection or pathologic response.129,131,137,138

Heart Chemoradiation therapy for esophageal cancer can have adverse effects on the heart. The most extensive evidence in support of this comes from a study by Liu and colleagues,139 who retrospectively reviewed and analyzed the records of 614 patients with esophageal cancer treated at M. D. Anderson between 1998 and 2005. In that analysis, 101 patients (nonsurgical group) had been treated with definitive chemoradiation therapy without surgery  between 1998 and 2003, and 513 (surgical group) underwent esophagogastrectomy with radiation-chemotherapy (n = 169) or without radiation-chemotherapy (n = 344) between 1998 and 2005. In the nonsurgical group, 22% had stage I or II disease, and 73% had stage III or IV disease; in the surgical group, 36% had stage I or II disease, and 64% had stage III or IV disease. In the nonsurgical group, 35% of patients had upper or midesophageal tumors, and 60% had distal tumors; in the surgical group, 11% had upper or midesophageal tumors, and 88% had distal tumors. Threedimensional conformal radiation therapy was the primary technique, with prescribed doses of 45 or 50.4 Gy. Patient characteristics and cardiac risk factors were matched between the nonsurgical and surgical groups. Cardiac abnormality was evaluated on the basis of clinical records and perfusion imaging studies, if they were available. In the nonsurgical group, the most common cardiac ­ adverse effects were pericardial effusion, which occurred in 28% of patients, and atrial fibrillation, which ­occurred in 8%. These cardiac events developed within 

15 months ­after radiation therapy (median time to onset, 5.3 months for pericardial effusion and 4.6 months for atrial ­fibrillation). The risk for pericardial effusion was associated with the mean pericardial dose (P = .002) and an array of dose-volume points for the pericardium from 5 to 45 Gy,  possibly because of correlation among the dosimetric factors. In the surgical group, atrial fibrillation occurred in 15% of patients within 30 days after surgery. In this surgical group, 49 patients had distal esophageal tumor located behind the heart and underwent gated myocardial perfusion single-photon emission computed tomography (SPECT) before surgery.140 Within this subgroup, 25 patients had undergone chemoradiation therapy and 24 patients surgery only. SPECT imaging revealed myocardium perfusion defects, including ischemia or scarring, in 11 (42%) of the  25 patients in the chemoradiation group but in only 1 (4%) of the 24 patients treated with surgery only (P = .001). The time to onset of myocardial perfusion defect ranged from  1 to 48 months (mean, 7.5 months; median, 3 months).140

Prognostic Factors in Esophageal Cancer Clinical experience indicates that tumors usually start responding to chemoradiation therapy by 2 to 3 weeks after the treatment is begun and that tumor response is usually indicated by improved ability to swallow. Differences in response to chemoradiation therapy are thought to reflect differences in the biologic behavior of the tumors, and ideally the treatment would be chosen on the basis of these predicted behaviors. For example, if a ­ tumor is predicted to be relatively radioresistant, the ­radiation dose may need to be increased to encourage early tumor shrinkage and improve the potential for complete response and avoidance of esophagectomy. The ability to predict tumor response to chemoradiation therapy before treatment is begun would allow truly ­individualized treatments on the basis of the intensity and modalities of the treatment to be delivered. The best studied of the prognostic factors found are listed in Table 20-5.53,115,141-144

Percentage alive

20  The Esophagus 100 90 80 70 60 50 40 30 20 10 0

*P = .03

EUS-cN1M0 EUS-cN0M0

EUS-cM1A (celiac)*

0

6

12

18

24

30

36

42

48

54

60

Months N0M0: 65 N1M0: 96 M1A: 18

37 46 8

16 30 2

14 18 1

9 8 1

4 4 0

FIGURE 20-18.  Overall survival of patients with adenocarcinoma of the distal esophagus according to node status identified by endoscopic sonography before treatment. P = .03 for cM1A versus cN0 M0, cN1 M0. (From Malaisrie S, Hofstetter W, Correa A, et al. Endoscopic ultrasonography-identified celiac adenopathy remains a poor prognostic factor despite preoperative chemoradiotherapy in esophageal adenocarcinoma. J Thorac Cardiovasc Surg 2006;131:65-72.)

Involvement of Celiac Axis Nodes Celiac adenopathy is a poor prognostic factor in esophageal cancer, especially for patients treated with surgery only or with preoperative chemoradiation therapy and surgery.141,145,146 Because the risk for relapse is high for patients with celiac axis involvement treated with surgery alone, the 2002 AJCC1 stage classification considers celiac axis involvement a form of metastatic disease (M1a) rather than a form of nodal involvement. In one study, Malaisrie and colleagues141 retrospectively examined the impact of celiac adenopathy in a single-institution experience that included 186 patients with adenocarcinoma of the distal esophagus who were treated with preoperative chemoradiation therapy and esophagectomy or induction chemotherapy followed by preoperative chemoradiation therapy and esophagec­ tomy. The identification of celiac adenopathy on pretreatment sonography was a significant predictor of decreased long-term survival. For the 65 patients with cN0 M0 disease, median survival time was 49 months, and the 3-year survival rate was 54%; for the 96 patients in the cN1 M0 group, 45 months and 56%; and for the 18 patients with cM1a disease (celiac adenopathy), 19 months and 12% (P = .03). Celiac adenopathy might have been associated with an increased risk for systemic relapse (44% versus 22% for patients without celiac node disease, P = .07).  The only factor associated with increased survival in ­patients with celiac adenopathy was the addition of induction chemotherapy before chemoradiation therapy and surgical intervention (27 versus 15 months, P = .02). ­Celiac adenopathy might have been associated with an increased risk for systemic rather than locoregional ­relapse. Systemic relapse rates were 44% for patients with celiac adenopathy, 18% for patients with no lymph node involvement, and 26% for patients with periesophageal node involvement (P = .03) (Fig. 20-18).141

481

Locoregional relapse rates were similar among the three groups, and receipt of induction chemotherapy before chemoradiation therapy was associated with long-term survival, suggesting that additional improvements in survival for patients with celiac lymph node disease will come from strategies designed to achieve better systemic control.141 These findings indicate that it is important to identify patients with celiac adenopathy before treatment and to evaluate the survival benefits of induction chemotherapy before chemoradiation therapy in this high-risk group.141 In another study, Catalano and coworkers147 found that the use of endoscopic sonography for the preoperative detection of celiac lymph node involvement, with nodes characterized in terms of size, shape, echogenicity, and borders, had a sensitivity of 83% and a specificity of 98%. False-positive results occurred for only 2 of 145 patients.147 Another group146 showed that the mere detection of celiac axis lymph nodes was highly predictive of malignant involvement146 and that virtually 100% of celiac lymph nodes that are larger than 10 mm contain metastatic disease.148

Pathologic Response Complete eradication of cancer cells in the surgical specimen has consistently been recognized as a favorable prognostic factor for both locoregional tumor control and overall  survival in esophageal cancer.149,150 Pathologic complete response rates are 2.5% to 5% for patients given induction chemotherapy alone before surgery,35,151 11% to 41% for those given preoperative chemoradiation therapy with 5-FU and cisplatin, 17.7% to 46% for those given preoperative chemoradiation therapy with paclitaxel, and up to 71% for those given induction chemotherapy followed by preoperative chemoradiation therapy.97,152-155 Trials investigating preoperative chemoradiation therapy have found marked advantages in both locoregional control and overall survival for patients who experienced pathologic complete response to such therapy.121,156-158 In an attempt to clarify whether pretreatment clinical disease or posttreatment pathologic disease stage was the better predictor of survival in esophageal cancer, Chirieac and others159 studied 235 consecutive patients with pretreatment clinical stage II, III, or IVA carcinoma of the esophagus or gastroesophageal junction who were then treated with chemoradiation therapy followed by esophagectomy. Posttreatment status was classified according to pathologic stage and semiquantitative assessment of residual carcinoma. Clinicopathologic features, residual carcinoma status, and disease stage before and after therapy were considered in relationship to disease-free and overall survival. Chemoradiation therapy led to disease downstaging in 56% of patients in this study, and posttherapy pathologic stage was the strongest independent predictor of disease-free and overall survival (both P = .02), followed by extent of residual carcinoma (P = .04). On the basis of this and an expanded analysis that included 354 patients who did not have chemoradiation therapy before surgery, Swisher and colleagues142 proposed that the esophageal cancer staging system be revised to include pathologic response to preoperative chemoradiation therapy (Fig.  20-19).142 These ­ findings support the concept that disease can be downstaged by preoperative chemoradiation therapy and ­confirm the findings of previous studies86,87,160

482

PART 6  Gastrointestinal Tract 100 90

Percentage alive

80

P0 (0% residual)

70

P1 (1–50% residual)

60 50 40

P2 (>50% residual)

30 20

P <.001

10 0 0

6

12

18

24

30

36

42

48

Months P0: 86 P1: 98 P2: 53

76 80 34

53 45 17

40 29 10

30 13 4

FIGURE 20-19.  Overall survival of patients with resectable esophageal cancer treated with preoperative chemoradiation and surgery according to the pathologic response of the primary tumor (pP). At 3 years, the survival rate for those with P0 disease (i.e., 0% residual disease in the pathology specimen) was 74%; for those with P1 (1% to 50% residual disease), 54%; and for those with P2 disease (≥50% residual disease), 24% (P < .001). (From Swisher S, Hofstetter W, Wu T, et al. Proposed revision of the esophageal cancer staging system to accommodate pathologic response (pP) following preoperative chemoradiation (CRT). Ann Surg 2005;241:810-817.)

that ­ preoperative chemoradiation therapy for locoregionally ­ advanced esophageal cancer can produce complete or nearly complete pathologic response in at least some ­patients.

Tumor Metabolic Activity on Positron Emission Tomography FDG-PET imaging is widely used for anatomic imaging and staging esophageal cancer.161,162 FDG uptake on PET images reflects tumor cell metabolism, which supports cell growth and cellular maintenance and repair, and it is therefore used as a measure of tumor responsiveness to various treatments. In one study, the number of abnormalities visualized on pretreatment PET was predictive of overall survival and disease-free survival.163 FDG uptake on PET after preoperative chemotherapy or chemoradiation therapy has been shown to correlate with patient survival and pathologic response.49 In that study, FDG-PET scans obtained after radiation therapy but before surgery showed that PET had acceptable sensitivity for tumor viability when 50% to 100% of the tumor cells were viable. Disappointingly, FDG-PET did not have acceptable sensitivity for detecting residual tumor that consisted of less than 50% viable tumor cells.49 Downey and colleagues50 reported that PET could detect metastatic disease better than conventional imaging at initial evaluations but not at later ones, and they called for further studies to correlate FDG-PET with survival. Flamen and others51 used FDG-PET before and after concurrent chemoradiation therapy for locally advanced esophageal cancer and found that tumor response as visualized on the scans correlated strongly with pathologic response and survival. The value of reduction in FDG uptake by tumors on PET images early in the course of

preoperative chemoradiation therapy as a prognostic sign also warrants further investigation, as it may correlate with tumor response to preoperative chemotherapy.52 Although FDG-PET scanning seems to accurately predict pathologic complete response,53 at least 20% of patients with normal preoperative PET scans can still have residual disease in the resected specimen.49

Molecular Factors Significant advances have been made in elucidating the fundamental biology of cancer during the past decade. Many molecular processes and signaling pathways that regulate the survival, proliferation, and function of normal cells have been found to be dysregulated and dysfunctional in cancer cells. Many of these processes and pathways have been implicated in the development and growth of malignant tumors and in the development of resistance to standard cancer treatments. Because of these properties, the abnormal molecular processes and pathways in cancer cells have increasingly been investigated as potential predictors of outcome after specific therapies and as targets for cancer therapy. Targeted therapy directed against cancer cells is particularly appealing in combination with radiation therapy, chemotherapy, or chemoradiation therapy because targeted therapy theoretically will not damage normal cells and will not be associated with additional normal-tissue damage. Potential molecular markers implicated in esophageal cancer identified are shown in Table 20-6).164-187 These molecules are involved in growth regulation (e.g., epidermal growth factor receptor, HER2/NEU [also called ERBB2], Ki-67), angiogenesis (vascular endothelial growth factor), inflammation (cyclooxygenase 2, nuclear factor КB), cellcycle control (e.g., CDKN2A [formerly designated p16], CDKN1A [formerly designated p21], cyclin D1, flavopiridol), apoptosis (e.g., TP53, BAX, FAS, BCL2), metastasis (e.g., tissue inhibitor of metalloproteinases, E-cadherin), and intrinsic sensitivity to ­chemotherapy (e.g., P-­glycoprotein, thymidylate synthase, glutathione S-transferase, metallothionein, excision repair ­ cross-­complementing 1 gene [ERCC1]).164,188-191 Several of these molecules have repeatedly been shown to predict clinical outcomes.188,190,192-195 For example, overexpression of the epidermal growth factor receptor predicts a poor prognosis for both squamous cell carcinoma and adenocarcinoma of the esophagus.196-199 However, the prognostic value of other markers, ­particularly TP53 and cyclooxygenase 2, ­remains unclear.190,193,200

Techniques of Irradiation Cancer of the esophagus can be successfully treated with radiation therapy through the use of modern linear­accelerated photons or proton beams that exclude or minimize the exposure of normal lung, heart, and spinal cord tissue.

Technical Advances Traditional radiation treatment for esophageal cancer ­involved external beam irradiation and two-dimensional treatment planning. This approach, however, was ­associated with great uncertainty about dose distributions and a high probability of toxicity to normal tissues. During the past decade, three-dimensional conformal radiation therapy has

TABLE 20-6.    Potential Molecular Targets in Esophageal Cancer Biomarker

Alteration

Mechanism

Function

Percentage

References

Cyclin D1

Gene amplification

Promotes entry into S phase

Growth self-sufficiency

30%-70%

165-167

Retinoblastoma

Mutation, LOH

Tumor suppressor, blocks passage to S phase

Growth self-sufficiency

50%

166,168,169

Epidermal growth factor receptor

Gene amplification

Growth factor receptor

Growth self-sufficiency

30%-90%

164

Her2/NEU (ERBB2)

Gene amplification

Growth factor receptor

Growth self-sufficiency

23%-52%

170

Ki-67

75% Gene amplification

Growth factor

Growth self-sufficiency

Unknown

171

Vascular endothelial growth factor (VEGF); VEGF ­receptor

Overexpression

Endothelial cell growth factor; receptor

Angiogenesis

30%-64%

172-174

Cyclooxygenase 2 (COX-2)

Overexpression

Angiogenesis, inhibition of ­ apoptotic pathway, etc.

Angiogenesis, avoidance of apoptosis

75%

175,176

CDKN2B (formerly called p15)

Mutation, LOH at 9p21, ­promoter of ­hypermethylation

Tumor suppressor; inhibits cyclin-CDK complex

Resistance to antigrowth signal

80%-90%

177,178

CDKN2A (formerly called p16)

Mutation, LOH at 9p21, ­promoter of ­hypermethylation

Tumor suppressor; inhibits cyclin-CDK complex

Resistance to antigrowth signal

52.9%-81%

179

TP53

Mutation, LOH

Tumor suppressor; causes G1 arrest of DNA-damaged cells

Resistance to antigrowth signal

40%-75%

176,179,180

Human ribosomal protein L14

LOH and downregulation

Tumor suppressor

Resistance to antigrowth signal

43%-63%

181

Telomerase

Upregulation

Maintains telomere length

Unlimited DNA replication

100%

182,183

FAS; FAS ligands

Underexpression or ­overexpression

Death receptor and death receptor ligands

Inhibition of apoptosis in ­ cancer cells; immune ­surveillance

89.7%

184

E-cadherin

Downregulation

Cell adhesion protein

Invasion and metastasis

Periplakin

Downregulation

Cell adhesion protein

Invasion and metastasis

44.6%

185

Glut1 (human erythrocyte glucose transporter)

Overexpression

Increased glucose transportation to cancer cell

Cell growth

100%

186

Chemokine and bone ­ marrow-homing receptor (CXCR4)

Overexpression

Increased lymph node and bone ­marrow microinvolvement

Metastatic dissemination

55%

187

483

LOH, loss of heterozygosity

20  The Esophagus

Transforming growth factor α (TGF-α)

PART 6  Gastrointestinal Tract

become the standard of care for esophageal ­cancer. The dose conformality achievable with this technique ­ minimizes ­irradiation of critical structures and other normal tissues, thereby minimizing the toxicity of the treatment. Dose conformality and sparing of normal tissues can be improved still further by using intensity-modulated ­radiation therapy, a novel approach to the planning and delivery of therapeutic radiation. Unlike conventional threedimensional conformal techniques, intensity-­modulated radiation therapy usually involves inverse planning, in which dose-volume constraints for targets and normal tissues are defined a priori and then optimized with the use of a computer algorithm. Targets and normal tissues are first delineated on a CT scan. The algorithm then identifies beam orientations and patterns of intensity that optimize conformality of the prescription dose to the shape of the target in three dimensions while sparing surrounding normal tissues. Treatment is typically delivered with the help of multileaf collimators, which consist of individual motorized leaves that can move in and out of the beam’s path during treatment, thereby modulating the beam’s intensity.201 Numerous investigators have demonstrated the benefits of intensity-modulated radiation therapy and planning for sparing normal tissues202-204 and for the ability to deliver radiation doses higher than the conventional ones to a variety of tumor sites.205,206 Preliminary clinical reports have been promising; rates of adverse events have been low,207-209 even when higher-than-conventional doses are delivered.210,211

100 80 Percent volumn

484

60 40 20 0 0

10

20

30

40

50

60

Dose (Gy) CTV Heart RT kidney LT kidney Liver Spinal cord Total lung

CTV Heart RT kidney LT kidney Liver Spinal cord Total lung

FIGURE 20-20.  Dose-volume histogram comparing dose exposures to various organs from proton beam (solid lines) radiation therapy and from photon intensity-modulated radiation therapy (dashed lines). CTV, clinical target volume; LT, left; RT, right. (Courtesy of Ronald Zhu, PhD, Houston, TX.)

Proton Therapy The most cutting-edge form of radiation therapy being used to treat esophageal cancer is proton therapy. Protons are positively charged particles with a unique property that makes them ideal for delivering therapeutic radiation to sensitive structures. Protons deposit relatively little energy until they reach the end of their path, where their ionization peaks.212 After this peak, the ionization drops to virtually zero; there is no exit dose. The length of the path of a proton beam depends on its initial energy, which can be tailored to the specific clinical situation. Proton beams therefore deposit less radiation to the normal tissue in front of the tumor and hardly any to the normal tissue behind it. This unique property of protons allows proton therapy to spare more normal tissues than photon therapy can and therefore increase the therapeutic ratio of the irradiation. Developments in active scanning of proton beams are facilitating the development of intensity-modulated proton therapy, in which the dose conforms even more exquisitely to the tumor. Tumors that have been treated with proton therapy include esophageal cancer, chordomas, chondrosarcomas, meningiomas, pituitary adenomas, uveal ­melanomas, nasopharyngeal carcinomas, paranasal sinus cancers, non–small cell lung cancers, hepatocellular carcinomas, prostate cancers, medulloblastomas, retinoblastomas, ­sarcomas in children, and metastatic tumors to the brain and liver. Proton therapy has proved superior to photon beam therapy for control of skull base chordomas and chondrosarcomas.213 The largest study of proton therapy for esophageal cancer was reported by Sugahara and colleagues,214 who treated 46 patients with esophageal cancer with proton

beams, with or without photon irradiation, between 1985 and 1998. Forty patients received a combination of photon therapy (median dose, 48 Gy) and a boost dose with proton therapy (median dose, 31.7 Gy). The median combined dose from photon and proton therapy was 76 Gy (range, 69.1 to 87.4 Gy). The other six patients received only proton therapy (median dose, 82 Gy; range, 75 to 89.5 Gy). The median fraction dose was 3.0 Gy (range, 2.5 to 3.7 Gy), which is much higher than the median dose that can be delivered via conventional fractionation. The 5-year actuarial overall survival rates for all 46 patients was 34%; that for the 23 patients with T1 tumors was 55%, and that for the 23 patients with T2-4 tumors was 13%.214 Corresponding 5-year actuarial disease-specific survival rates were 67% for all patients, 95% for those with T1 tumors, and 33% for T2-4 tumors; 5-year actuarial locoregional control rates were 57%, 83%, and 29%, respectively. Of the 18 cases of disease recurrence, 16 were locoregional, and the other two were distant. The major adverse event associated with this treatment was esophageal ulcers, which occurred in 22 patients. ­Seventy-five percent of the patients who received the equivalent of 80 Gy or more developed esophageal ulcers, compared with 37% of those who received the equivalent of less than 80 Gy. The high incidence of ulcers in this study may have resulted from the large fractions and high total dose. Figure 20-20 illustrates the differences in dose to various tissues for a proton beam radiation plan (solid lines) and a photon intensity-modulated treatment plan (dotted lines) for a patient with esophageal cancer. The proton beam plan significantly reduced the volume of normal lung and heart in the treatment fields.

20  The Esophagus

485

FIGURE 20-21.  Coregistered CT and PET images for target delineation of esophageal cancer. Red lines denote the gross tumor volume; yellow denotes the clinical target volume.

Target Volume Considerations For radiation therapy to be successful, the gross tumor and the microscopic disease must be encompassed in the treatment portals. Information from whichever test yields positive results—endoscopy, endoscopic sonography, CT, or PET—should be used to determine the gross tumor volume (GTV). An esophageal wall thickness of more than  0.5 cm should be included in that volume. A barium swallow procedure that images the lumen of the esophagus may be helpful in delineating the length of a primary tumor. Coregistered PET-CT provides anatomic and functional information in a single imaging session, and the accurate registration of PET and CT data can facilitate tumor staging and treatment planning.215-222 The impact of coregistered PET-CT imaging on conformal radiation treatment planning for thoracic cancer, and, to a lesser extent, head and neck cancer,223 has been extensively studied and reviewed.224-227 In a 2005 study of esophageal cancer,228 target delineation was initially done on CT scans, and findings from PET scans were overlaid on those images to define the target volumes. FDG-PET visualized previously undetected distant metastatic disease in two patients and led

to increasing or decreasing the GTV in others. Modifications in the delineation of the GTV led to displacement of the isocenter of the planning target volume (PTV) in 18 patients and affected the percentage of total lung volume receiving more than 20 Gy in 25 patients.228 The value of coregistered CT-PET images for delineating the GTV for a large locally advanced adenocarcinoma of the gastroesophageal junction is shown in Figure 20-21. It would have been very difficult to identify the GTV without PET in this case. Accurate delineation of the CTV is more difficult than accurate delineation of the GTV in esophageal cancer, in part because of the tendency of esophageal cancer for multicentric disease or submucosal skip metastases, which can be found at considerable distances from the primary tumor.229,230 This fact supports the use of generous proximal and distal margins for treating the primary tumor.231 Analyses of patterns of failure after radiation therapy showed that marginal failure or failure outside the radiation field in the esophagus occurred when 5-cm longitudinal margins232,233 were used, leading some to advocate including the entire esophagus in the initial treatment.233,234  

486

PART 6  Gastrointestinal Tract

FIGURE 20-22.  Typical treatment plan for three-dimensional conformal radiation therapy for esophageal cancer.

In RTOG 85-01, the entire esophagus was included in the radiation portals. However, toxicity was severe, especially when chemotherapy was administered. In the follow-up RTOG trial 94-05, 5-cm proximal and distal margins and a 2-cm lateral margin from the lateral border of the GTV were recommended. Because esophageal cancer carries a high risk of nodal involvement, even apparently uninvolved regional lymph nodes should be included in the CTV. If the primary tumor is above the carina (i.e., in the proximal esophagus), the supraclavicular nodes should be included in the CTV. For tumors of the lower two thirds of the esophagus, treatment of the celiac nodes should be considered. For cervical esophageal cancer, the treatment fields usually extend from the laryngopharynx to the upper two thirds of the esophagus and encompass the lower cervical, supraclavicular, and superior mediastinal nodes. Including the nodal basin of the upper jugular, subdigastric region in the treatment field is still debated. The fact that these nodes are routinely ­included in the CTV in treatment of oral pharyngeal or hypopharyngeal cancer supports the inclusion of these nodes in the treatment of cervical esophageal cancer. In theory, instead of a uniform expansion of CTV from GTV, it may be reasonable to exclude uninvolved vertebral bodies, pericardium, and lung tissue from CTV. However, the effect of this reduction on CTV in terms of reduction of normal tissue toxicity and locoregional control needs to be investigated. The PTV accounts for daily setup uncertainty and motion of the target, another critical consideration in determining target volume and treatment planning for esophageal cancer. Esophageal tumor motion can be caused by respiration and peristalsis, especially in cases of gastroesophageal junction tumors. Unfortunately, motion of esophageal cancer is severely understudied, although longitudinal respiratory motion of 2 to 3 cm has been reported.235 Bosset and colleagues88 proposed the definition of an initial target volume, CTV1, which includes lymph node stations with high risk (>25%) of involvement, and a second target volume, CTV2, incorporating 2-cm margins in the cephalad and caudal directions and a 1-cm margin extending radially from the GTV. The CTV1 and CTV2 on the CT scans can be delineated from nodal atlases. The CTV1 dose is limited to 30 Gy over 3 weeks, and the CTV2 dose is 45 Gy.88 Cervical esophageal carcinomas can be treated with the patient in a supine position similar to that used for

i­rradiation of hypopharyngeal cancer. If the lesion is  located at or near the esophagogastric junction, it is also important to limit irradiation of the heart, liver, and kidneys. In the case of lesions of the middle and lower esophagus, positioning the patient in the prone position sometimes permits a greater distance between the spinal cord and the PTV. CT with the patient in the prone or the supine position facilitates treatment planning with three-­dimensional conformal or intensity-modulated radiation therapy. Tumors of the upper and middle thirds of the esophagus ­require treatment of the supraclavicular regions. Irradiation of the celiac lymph node region is appropriate for tumors of the lower third of the esophagus. A minimal field width for the PTV is usually 2 cm beyond the GTV, but the most appropriate radial margin around tumors of various locations and sizes is not known. Esophagography with thinbarium swallowing during simulation and associated CT helps in determining the most appropriate radial margin. Fractions of 1.8 Gy or 2.0 Gy are appropriate for treatment with curative intent. Larger fractions may be indicated if palliation is the only goal. Although total doses of 60 to 70 Gy are indicated if radiation therapy is the only treatment, a total dose of 50 Gy in 5 weeks is appropriate if concurrent chemotherapy is used. No data have been ­reported to support the use of higher total doses of radiation therapy with concurrent chemotherapy. With or without chemotherapy, the total dose to the spinal cord should be limited to 45 Gy. The dose to the normal lungs should be limited to 20 Gy and that to the heart to 40 Gy. Advanced treatment planning with systems that provide dose-volume histograms permits the most judicious balance of dose to the CTV and exposure of the surrounding normal tissues. It is never appropriate to place a posterior spinal cord block that simultaneously blocks the CTV. Although total doses of 50 Gy in 5 or 6 weeks are considered necessary to control subclinical extensions of disease, lower total doses may be used with concurrent chemotherapy. In the chemotherapy treatment group of RTOG 85-01,91 a total dose of 30 Gy was used for the initial large field; the smaller “boost” field that still had a minimal margin of 5 cm above and below the gross tumor received an additional 20 Gy for a total dose of 50 Gy to the known tumor. An example of an appropriate three-dimensional radiation therapy plan with isodose distribution for carcinoma of the thoracic esophagus is shown in Figure 20-22.

20  The Esophagus

Summary During the past decade, treatment outcomes for patients with cancer of the esophagus have improved greatly. Palliation of dysphagia and odynophagia can be achieved with radiation therapy alone, but palliation may be more consistent and durable if radiation is combined with chemotherapy. Patients with inoperable tumors can be cured with concurrent chemotherapy and moderate-dose radiation therapy. The addition of esophageal brachytherapy to ­external irradiation and concurrent chemotherapy is associated with a high risk for complications. Surgical techniques and support have improved such that mortality rates after esophagectomy have decreased dramatically. However, results of surgical treatment for esophageal cancer remain disappointing. It may be advantageous to use concurrent chemoradiation therapy before esophageal resection, but further studies are needed. Most of the studies comparing chemoradiation therapy with and without esophageal resection have not shown a survival benefit from the addition of surgery. Advances in radiation therapy techniques have, however, provided a means to increase the therapeutic ­ratio in the multimodality treatment of esophageal cancer.

REFERENCES   1. Greene FL, Page DL, Fleming ID, et al, eds. American Joint Committee on Cancer Staging Manual, 6th ed. New York: SpringerVerlag, 2002, pp 91-98.   2. DeNardi FG, Riddell RH. The normal esophagus. Am J Surg Pathol 1991;15:296-309.   3. Kitano S, Terblanche J, Kahn D, et al. The venous anatomy of the lower esophagus in portal hypertension: practical implications. Br J Surg 1986;73:525-531.   4. Thompson W. Esophageal cancer. Int J Radiat Oncol Biol Phys 1983;9:1533-1565.   5. Fajardo F, Berthroung M, Anderson R. Esophagus. Radiation Pathology. New York: Oxford University Press, 2001, pp 214-220.   6. Wei X, Liu HH, Tucker SL, et al. Risk factors for acute esophagitis in non-small-cell lung cancer patients treated with concurrent chemotherapy and three-dimensional conformal radiotherapy. Int J Radiat Oncol Biol Phys 2006;66:100-107.   7. Komaki R, Scott C, Ettinger D, et al. Randomized study of chemotherapy/radiation therapy combinations for favorable patients with locally advanced inoperable nonsmall cell lung cancer: Radiation Therapy Oncology Group (RTOG) 92-04. Int J Radiat Oncol Biol Phys 1997;38:149-155.   8. Berthrong M, Fajardo L. Radiation injury in surgical pathology. Part II. Alimentary tract. Am J Surg Pathol 1981;5:149-155.   9. Kamangar F, Dores GM, Anderson WF. Patterns of cancer incidence, mortality, and prevalence across five continents: defining priorities to reduce cancer disparities in different geographic regions of the world. J Clin Oncol 2006;24:2137-2150. 10. Blot W, Li J, Taylor P, et al. Nutrition intervention trials in Linxian, China: supplementation with specific vitamin/mineral combinations, cancer incidence, and disease-specific mortality in the general population. J Natl Cancer Inst 1993;85:1483-1492. 11. Somdyala N, Marasas W, Venter F, et al. Cancer patterns in four districts of the Transkei region—1991-1995. S Afr Med J 2003;93:144-148. 12. Holmes RS, Vaughan TL. Epidemiology and pathogenesis of esophageal cancer. Semin Radiat Oncol 2007;17:2-9. 13. American Cancer Society. Cancer Facts & Figures. Atlanta: American Cancer Society, 2008. 2008. 14. Stewart B, Kleihues P. World Cancer Report. Lyon: IARC Press, 2003. 15. Islami F, Kamangar F, Aghcheli K, et al. Epidemiologic features of upper gastrointestinal tract cancers in Northeastern Iran. Br J Cancer 2004;90:1402-1406.

487

16. Mahboubi E, Kmet J, Cook P, et al. Oesophageal cancer studies in the Caspian Littoral of Iran: the Caspian cancer registry. Br J Cancer 1973;28:197-214. 17. Brown L, Devesa S. Epidemiologic trends in esophageal and gastric cancer in the United States. Surg Oncol Clin N Am 2002;11:235-256. 18. Devesa S, Blot W, Fraumeni JF. Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer 1998;83:2049-2053. 19. Wijnhoven B, Louwman M, Tilanus H, et al. Increased incidence of adenocarcinomas at the gastro-oesophageal junction in Dutch males since the 1990s. Eur J Gastroenterol Hepatol 2002;14:  115-122. 20. Lagergren J, Bergstrom R, Lindgren A, et al. Symptomatic gastroesophageal reflux as a risk factor for esophageal adenocarcinoma. N Engl J Med 1999;340:825-831. 21. Crane SJ, Locke GR 3rd, Harmsen WS, et al. Subsite-specific risk factors for esophageal and gastric adenocarcinoma. Am J Gastroenterol 2007;102:1596-1602. 22. De Jonge PJ, Wolters LM, Steyerberg EW, et al. Environmental risk factors in the development of adenocarcinoma of the oesophagus or gastric cardia: a cross-sectional study in a Dutch cohort. Aliment Pharmacol Ther 2007;26:31-39. 23. Streeter OE Jr, Martz KL, Gaspar LE, et al. Does race influence survival for esophageal cancer patients treated on the radiation and chemotherapy arm of RTOG #85-01? Int J Radiat Oncol Biol Phys 1999;44:1047-1052. 24. Brown L, Hoover R, Greenberg R, et al. Are racial differences in squamous cell esophageal cancer explained by alcohol and tobacco use? J Natl Cancer Inst 1994;86:1340-1345. 25. Brown L, Hoover R, Silverman D, et al. Excess incidence of squamous cell esophageal cancer among US black men: role of social class and other risk factors. Am J Epidemiol 2001;153:114-122. 26. Tuyns A. Oesophageal cancer in non-smoking drinkers and in non-drinking smokers. Int J Cancer 1983;32:443-444. 27. Tran G, Sun X, Abnet C, et al. Prospective study of risk factors for esophageal and gastric cancers in the Linxian general population trial cohort in China. Int J Cancer 2005;113:456-463. 28. Brown L, Swanson C, Gridley G, et al. Dietary factors and the risk of squamous cell esophageal cancer among black and white men in the United States. Cancer Causes Control 1998;9:467-474. 29. Launoy G, Milan C, Day NE, et al. Oesophageal cancer in France: potential importance of hot alcoholic drinks. Int J Cancer 1997;71:917-923. 30. Sharp L, Chilvers CE, Cheng KK, et al. Risk factors for squamous cell carcinoma of the oesophagus in women: a case-­control study. Br J Cancer 2001;85:1667-1670. 31. Novacek G. Plummer-Vinson syndrome. Orphanet J Rare Dis 2006;1:36. 32. Messmann H. Squamous cell cancer of the oesophagus. Best Pract Res Clin Gastroenterol 2001;15:249-265. 33. Akiyama H, Tsurumaru M, Kawamura T, et al. Principles of surgical treatment for carcinoma of the esophagus: analysis of lymph node involvement. Ann Surg 1981;194:438-446. 34. Anderson L, Lad T. Autopsy findings in squamous-cell carcinoma of the esophagus. Cancer 1982;50:1587-1590. 35. Kelsen D, Ginsberg R, Pajak T, et al. Chemotherapy followed by surgery compared with surgery alone for localized esophageal cancer. N Engl J Med 1998;339:1979-1984. 36. Herskovic A, Martz K, al-Sarraf M, et al. Combined chemotherapy and radiotherapy compared with radiotherapy alone in patients with cancer of the esophagus. N Engl J Med 1992;326:15931598. 37. Familiari P, Marchese M, Larghi A, et al. Staging of esophageal carcinoma: endoscopic ultrasonography. Rays 2005;30:357-362. 38. Rice T. Clinical staging of esophageal carcinoma CT, EUS, and PET. Chest Surg Clin N Am 2000;10:471-485. 39. Van Overhagen H, Bechker C. Diagnosis and staging of carcinoma of the esophagus and gastroesophageal junction, and detection of postoperative recurrence, by computer tomography. In Meyers MA (ed): Neoplasms of the Digestive Tract: Imaging, Staging and Management. Philadelphia: Lippincott-Raven, 1998, pp 31-48.

488

PART 6  Gastrointestinal Tract

40. Saunders HS, Wolfman NT, Ott DJ. Esophageal cancer. Radiologic staging. Radiol Clin North Am 1997;35:281-294. 41. Chandawarkar RY, Kakegawa T, Fujita H, et al. Comparative analysis of imaging modalities in the preoperative assessment of nodal metastasis in esophageal cancer. J Surg Oncol 1996;61:214-217. 42. Flanagan FL, Dehdashti F, Siegel BA, et al. Staging of esophageal cancer with 18F-fluorodeoxyglucose positron emission tomography. AJR Am J Roentgenol 1997;168:417. 43. Rankin SC, Taylor H, Cook GJ, et al. Computed tomography and positron emission tomography in the preoperative staging of oesophageal carcinoma. Clin Radiol 1998;53:659-665. 44. van Westreenen HL, Westerterp M, Bossuyt PMM, et al. Systematic review of the staging performance of 18F-fluorodeoxyglucose positron emission tomography in esophageal cancer. J Clin Oncol 2004;22:3805-3812. 45. Vrieze O, Haustermans K, De Wever W, et al. Is there a role for FGD-PET in radiotherapy planning in esophageal carcinoma? Radiother Oncol 2004;73:269-275. 46. Levine EA, Farmer MR, Clark P, et al. Predictive value of 18fluoro-deoxy-glucose-positron emission tomography (18FFDG-PET) in the identification of responders to chemoradiation therapy for the treatment of locally advanced esophageal cancer. Ann Surg 2006;243:472-478. 47. Blackstock AW, Farmer MR, Lovato J, et al. A prospective evaluation of the impact of 18-F-fluoro-deoxy-D-glucose positron emission tomography staging on survival for patients with locally advanced esophageal cancer. Int J Radiat Oncol Biol Phys 2006;64:455-460. 48. Westerterp M, van Westreenen HL, Reitsma JB, et al. Esophageal cancer: CT, endoscopic US, and FDG PET for assessment of response to neoadjuvant therapy—systematic review. Radiology 2005;236:841-851. 49. Swisher S, Erasmus J, Maish M, et al. 2-Fluoro-2-deoxy-d-glucose  positron emission tomography imaging is predictive of pathologic response and survival after preoperative chemoradiation in patients with esophageal carcinoma. Cancer 2004;101:  1776-1785. 50. Downey R, Akhurst T, Ilson D, et al. Whole body 18FDG-PET and the response of esophageal cancer to induction therapy: results of a prospective trial. J Clin Oncol 2003;21:429-432. 51. Flamen P, Van Cutsem E, Lerut A, et al. Positron emission tomography for assessment of the response to induction radiochemotherapy in locally advanced oesophageal cancer. Ann Oncol 2002;13:361-368. 52. Weber WA, Ott K, Becker K, et al. Prediction of response to ­preoperative chemotherapy in adenocarcinomas of the esophagogastric junction by metabolic imaging. J Clin Oncol 2001;19:  3058-3065. 53. Swisher S, Maish M, Erasmus J, et al. Utility of PET, CT, and EUS to identify pathologic responders in esophageal cancer. Ann Thorac Surg 2004;78:1152-1160. 54. Ellis FH Jr, Heatley G, Krasna M, et al. Esophagogastrectomy for carcinoma of the esophagus and cardia: a comparison of findings and results after standard resection in three consecutive eightyear intervals with improved staging criteria. J Thorac Cardiovasc Surg 1997;113:836-846. 55. Swisher S, Pisters P, Komaki R, et al. Gastroesophageal junction adenocarcinoma. Curr Treat Options Oncol 2000;1:387-398. 56. Earlam R, Cunha-Melo J. Oesophageal squamous cell carcinoma. I. A critical review of surgery. Br J Surg 1980;67:381-390. 57. Whyte R, Orringer M. Surgery for carcinoma of the esophagus: the case for transhiatal esophagectomy. Semin Radiat Oncol 1994;4:146-156. 58. Orringer M. Resection of the esophagus. In Shields T, LoCicero J, Ponn R (eds): General Thoracic Surgery, 5th ed. Philadelphia: Lippincott Williams & Wilkins, 2000, pp 1697-1722. 59. Hulscher JBF, van Sandick JW, de Boer AGEM, et al. Extended transthoracic resection compared with limited transhiatal resection for adenocarcinoma of the esophagus. N Engl J Med 2002;347:1662-1669. 60. Swisher SG, Deford L, Merriman KW, et al. Effect of operative volume on morbidity, mortality, and hospital use after esophagectomy for cancer. J Thorac Cardiovasc Surg 2000;119:1126-1132.

61. Kuo EY, Chang Y, Wright CD. Impact of hospital volume on clinical and economic outcomes for esophagectomy. Ann Thorac Surg 2001;72:1118-1124. 62. Wouters MW, Wijnhoven BP, Karim-Kos HE, et al. High-­volume versus low-volume for esophageal resections for cancer: the essential role of case-mix adjustments based on clinical data. Ann Surg Oncol 2008;15:80-87. 63. Wenner J, Zilling T, Bladström A, et al. The influence of surgical volume on hospital mortality and 5-year survival for carcinoma of the oesophagus and gastric cardia. Anticancer Res 2005;25:  419-424. 64. Metzger R, Bollschweiler E, Vallböhmer D, et al. High volume centers for esophagectomy: what is the number needed to achieve low postoperative mortality? Dis Esophagus 2004;17:310-314. 65. Verhoef C, van de Weyer R, Schaapveld M, et al. Better survival in patients with esophageal cancer after surgical treatment in university hospitals: a plea for performance by surgical oncologists. Ann Surg Oncol 2007;14:1678-1687. 66. Migliore M, Choong CK, Lim E, et al. A surgeon’s case volume of oesophagectomy for cancer strongly influences the operative mortality rate. Eur J Cardiothorac Surg 2007;32:375-380. 67. McCulloch P, Ward J, Tekkis PP. for the British Oesophago­Gastric Cancer Group. Mortality and morbidity in gastrooesophageal cancer surgery: initial results of ASCOT ­multicenter prospective cohort study. BMJ 2003;327:1192-1197. 68. Rodgers M, Jobe BA, O’Rourke RW, et al. Case volume as a predictor of inpatient mortality after esophagectomy. Arch Surg 2007;142:829-839. 69. Suntharalingam M, Moughan J, Coia LR. Outcome results of the 1996-1999 Patterns of Care Survey of the national practice for patients receiving radiation therapy for carcinoma of the esophagus. J Clin Oncol 2005;23:2325-2331. 70. Whittington R. Controversies in the management of adenocarcinoma of the esophagus and esophagogastric junction. Semin Radiat Oncol 1994;4:170-178. 71. Medical Research Council (MRC) Oesophageal Cancer Working Party. Surgical resection with or without preoperative chemotherapy in oesophageal cancer: a randomised controlled trial. Lancet 2002;359:1727-1733. 72. Urschel J, Vasan H. A meta-analysis of randomized controlled trials that compared neoadjuvant chemoradiation and surgery to surgery alone for resectable esophageal cancer. Am J Surg 2003;185:538-543. 73. Yang Z, Gu X, Zhao S, et al. Long term survival of radiotherapy for esophageal cancer: analysis of 1136 patients surviving for more than 5 years. Int J Radiat Oncol Biol Phys 1983;9:1769-1773. 74. al-Sarraf M, Martz K, Herskovic A, et al. Progress report of combined chemoradiotherapy versus radiotherapy alone in patients with esophageal cancer: an intergroup study. J Clin Oncol 1997;15:277-284. 75. Teniere P, Hay J, Fingerhut A, et al. Postoperative radiation therapy does not increase survival after curative resection for squamous cell carcinoma of the middle and lower esophagus as shown by a multicenter controlled trial. French University Association for Surgical Research. Surg Gynecol Obstet 1991; 173:123-130. 76. Fok M, Sham J, Choy D, et al. Postoperative radiotherapy for carcinoma of the esophagus: a prospective, randomized controlled study. Surgery 1993;113:138-147. 77. Zieren H, Muller J, Jacobi C, et al. Adjuvant postoperative radiation therapy after curative resection of squamous cell carcinoma of the thoracic esophagus: a prospective randomized study. World J Surg 1995;19:444-449. 78. Launois B, Delarue D, Campion J, et al. Preoperative radiotherapy for carcinoma of the esophagus. Surg Gynecol Obstet 1981;153:690-692. 79. Gignoux M, Roussel A, Paillot B, et al. The value of preoperative radiotherapy in esophageal cancer: results of a study of the E.O.R.T.C. World J Surg 1987;11:426-432. 80. Wang M, Gu X, Yin W, et al. Randomized clinical trial on the combination of preoperative irradiation and surgery in the treatment of esophageal carcinoma: report on 206 patients. Int J Radiat Oncol Biol Phys 1989;16:325-327.

20  The Esophagus 81. Arnott SJ, Duncan W, Kerr GR, et al. Low dose preoperative radiotherapy for carcinoma of the oesophagus: results of a randomized clinical trial. Radiother Oncol 1992;24:108-113. 82. Nygaard K, Hagen S, Hansen H, et al. Preoperative radiotherapy prolongs survival in operable esophageal carcinoma: a randomized, multicenter study of preoperative radiotherapy and chemotherapy. The second Scandinavian trial in esophageal cancer. World J Surg 1992;16:1104-1110. 83. Bosset J, Gignoux M, Triboulet J, et al. Chemoradiotherapy followed by surgery compared with surgery alone in squamous-cell cancer of the esophagus. N Engl J Med 1997;337:161-167. 84. Urba S, Orringer M, Turrisi A, et al. A randomized trial comparing surgery (s) to preoperative concomitant chemoradiation plus surgery in patients (pts) with resectable esophageal cancer (CA): updated analysis [abstract]. Proc Am Soc Clin Oncol 1997;16:277a. 85. Le Prise E, Etienne P, Meunier B, et al. A randomized study of chemotherapy, radiation therapy, and surgery versus surgery for localized squamous cell carcinoma of the esophagus. Cancer 1994;73:1779-1784. 86. Walsh T, Noonan N, Hollywood D, et al. A comparison of multimodal therapy and surgery for esophageal adenocarcinoma. N Engl J Med 1996;335:462-467. 87. Burmeister B, Smithers B, Fitzgerald L. A randomized phase III trial of preoperative chemoradiation followed by surgery (CS-S) versus surgery alone (S) for localized resectable cancer of the esophagus [abstract]. Proc Am Soc Clin Oncol 2002;21:130a. 88. Bosset J, Lorchel F, Mantion G, et al. Radiation and chemoradiation therapy for esophageal adenocarcinoma. J Surg Oncol 2005;92:239-245. 89. Apinop C, Puttisak P, Preecha N. A prospective study of combined therapy in esophageal cancer. Hepatogastroenterology 1994;41:391-393. 90. Law S, Fok M, Chow S, et al. Preoperative chemotherapy versus surgical therapy alone for squamous cell carcinoma of the esophagus: a prospective randomized trial. J Thorac Cardiovasc Surg 1997;114:210-217. 91. Cooper J, Guo M, Herskovic A, et al. Chemoradiotherapy of locally advanced esophageal cancer: long-term follow-up of a prospective randomized trial (RTOG 85-01). Radiation Therapy Oncology Group. JAMA 1999;281:1623-1627. 92. Minsky B, Pajak T, Ginsberg R, et al. INT 0123 (Radiation Ther­ apy Oncology Group 94-05) phase III trial of combined-­modality therapy for esophageal cancer: high-dose versus ­ standard-dose radiation therapy. J Clin Oncol 2002;20:1167-1174. 93. Kaklamanos IG, Walker GR, Ferry K, et al. Neoadjuvant treatment for resectable cancer of the esophagus and the gastroesophageal junction: a meta-analysis of randomized clinical ­trials. Ann Surg Oncol 2003;10:754-761. 94. Liao Z, Jin J, Zhang Z, et al. Surgical resection may increase survival of patients with clinical stage II and III esophageal cancer treated with concurrent radiation [abstract]. Proceedings of the 84th annual meeting of the American Radium Society, Puerto Rico, 2002. Cancer J 2002;8:486. 95. Ajani J, Walsh G, Komaki R, et al. Preoperative induction of CPT-11 and cisplatin chemotherapy followed by chemoradiotherapy in patients with locoregional carcinoma of the esophagus or gastroesophageal junction. Cancer 2004;100:2347-2354. 96. Jin J, Liao Z, Zhang Z, et al. Induction chemotherapy improved outcomes of patients with resectable esophageal cancer who received chemoradiotherapy followed by surgery. Int J Radiat Oncol Biol Phys 2004;60:427-436. 97. Swisher S, Ajani J, Komaki R, et al. Long-term outcome of phase II trial evaluating chemotherapy, chemoradiotherapy, and surgery for locoregionally advanced esophageal cancer. Int J Radiat Oncol Biol Phys 2003;57:120-127. 98. Bains M, Stojadinovic A, Minsky B, et al. A phase II trial of preoperative combined-modality therapy for localized esophageal carcinoma: initial results. J Thorac Cardiovasc Surg 2002;124:270-277. 99. Ilson DH, Bains M, Kelsen DP, et al. Phase I trial of escalatingdose irinotecan given weekly with cisplatin and concurrent radiotherapy in locally advanced esophageal cancer. J Clin Oncol 2003;21:2926-2932.

489

100. Pasini F, de Manzoni G, Pedrazzani C, et al. High pathological response rate in locally advanced esophageal cancer after neoadjuvant combined modality therapy: dose finding of a weekly chemotherapy schedule with protracted venous infusion of  5-fluorouracil and dose escalation of cisplatin, docetaxel and concurrent radiotherapy. Ann Oncol 2005;16:1133-1139. 101. Komaki R, Janjan N, Ajani J, et al. Phase I study of irinotecan and concurrent radiation therapy for upper GI tumors. Oncology 2000;14:34-37. 102. Sur R, Singh D, Sharma S, et al. Radiation therapy of esophageal cancer: role of high dose rate brachytherapy. Int J Radiat Oncol Biol Phys 1992;22:1043-1046. 103. Gaspar L, Winter K, Kocha W, et al. A phase I/II study of external beam radiation, brachytherapy, and concurrent chemotherapy for patients with localized carcinoma of the esophagus (Radiation Therapy Oncology Group Study 9207). Cancer 2000;88:988-995. 104. Kelsen D, Bains M, Hilaris B, et al. Combination chemotherapy of esophageal carcinoma using cisplatin, vindesine, and bleomycin. Cancer 1982;49:1174-1177. 105. Leichman L, Steiger Z, Seydel H, et al. Combined preoperative chemotherapy and radiation therapy for cancer of the esophagus: the Wayne State University, Southwest ­ Oncology Group and Radiation Therapy Oncology Group experience. Semin  Oncol 1984;11:178-185. 106. Coia L, Engstrom P, Paul A, et al. A pilot study of combined radiotherapy and chemotherapy for esophageal carcinoma. Am J Clin Oncol 1984;7:653-659. 107. Coia L, Engstrom P, Paul A. Long term results of infusional  5-FU, mitomycin-C and radiation as primary management of esophageal cancer. Int J Radiat Oncol Biol Phys 1991;20:29-36. 108. Smith TJ, Ryan LM, Douglass HO Jr, et al. Combined chemoradiotherapy vs. radiotherapy alone for early stage squamous cell ­carcinoma of the esophagus: a study of the Eastern ­Cooperative Oncology Group. Int J Radiat Oncol Biol Phys 1998;42:  269-276. 109. Richmond J, Seydel H, Bae Y, et al. Comparison of three treatment strategies for esophageal cancer within a single institution. Int J Radiat Oncol Biol Phys 1987;13:1617-1620. 110. Minsky B, Neuberg D, Kelsen D, et al. Final report of Intergroup Trial 0122 (ECOG PE-289, RTOG 90-12): phase II trial of neoadjuvant chemotherapy plus concurrent chemotherapy and high-dose radiation for squamous cell carcinoma of the esophagus. Int J Radiat Oncol Biol Phys 1999;43:517-523. 111. Minsky BD, Pajak TF, Ginsberg RJ, et al. INT 0123 (Radiation Therapy Oncology Group 94-05) phase III trial of ­ combined­modality therapy for esophageal cancer: high-dose versus ­standarddose radiation therapy. J Clin Oncol 2002;20:1167-1174. 112. Liao Z, Zhang Z, Jin J, et al. Esophagectomy after concurrent chemoradiotherapy improves locoregional control in clinical stage II or III esophageal cancer patients. Int J Radiat Oncol Biol Phys 2004;60:1484-1493. 113. Coia LR, Minsky BD, Berkey BA, et al. Outcome of patients receiving radiation for cancer of the esophagus: results of the 1992-1994 patterns of care study. J Clin Oncol 2000;18:  455-462. 114. Bedenne L, Michel P, Bouché O, et al. Chemoradiation followed by surgery compared with chemoradiation alone in squamous cancer of the esophagus: FFCD 9102. J Clin Oncol 2007;25:1160-1168. 115. Stahl M, Stuschke M, Lehmann N, et al. Chemoradiation with and without surgery in patients with locally advanced squamous cell carcinoma of the esophagus. J Clin Oncol 2005;23:  2310-2317. 116. Fiorica F, Di Bona D, Schepis F, et al. Preoperative chemoradiotherapy for oesophageal cancer: a systematic review and metaanalysis. Gut 2004;53:925-930. 117. Caspers R, Welvaart K, Verkes R, et al. The effect of radiotherapy on dysphagia and survival in patients with esophageal cancer. Radiother Oncol 1988;12:15-23. 118. Brown S. Palliation of malignant dysphagia: surgery, ­radiotherapy, laser, intubation alone or in combination? Gut 1991;32:  841-844.

490

PART 6  Gastrointestinal Tract

119. Siewert J, Feith M, Werner M, et al. Adenocarcinoma of the esophagogastric junction: results of surgical therapy based on anatomical/topographic classification in 1,002 consecutive patients. Ann Surg 2000;232:353-361. 120. Hofstetter W, Swisher S, Correa A, et al. Treatment outcome of resected esophageal cancer. Ann Surg 2002;236:1-10. 121. Heath E, Burtness B, Heitmiller R, et al. Phase II evaluation of preoperative chemoradiation and postoperative adjuvant chemotherapy for squamous cell and adenocarcinoma of the esophagus. J Clin Oncol 2000;18:868-876. 122. Forastiere A, Orringer M, Perez-Tamayo C, et al. Preoperative chemoradiation followed by transhiatal esophagectomy for ­carcinoma of the esophagus: final report. J Clin Oncol 1993;11:  1118-1123. 123. Pavy JJ, Bosset JF. Late effects of radiations on the esophagus [in French]. Cancer Radiother 1997;1:732-734. 124. Abou-Jawde RM, Mekhail T, Adelstein DJ, et al. Impact of induction concurrent chemoradiotherapy on pulmonary function and postoperative acute respiratory complications in esophageal cancer. Chest 2005;128:250-255. 125. Law S, Wong K, Kwok K, et al. Predictive factors for postoperative pulmonary complications and mortality after esophagectomy for cancer. Ann Surg 2004;240:791-800. 126. Ferguson MK, Durkin AE. Preoperative prediction of the risk of pulmonary complications after esophagectomy for cancer.  J Thorac Cardiovasc Surg 2002;123:661-669. 127. Avendano CE, Flume PA, Silvestri GA, et al. Pulmonary complications after esophagectomy. Ann Thorac Surg 2002;73:  922-926. 128. Fang W, Kato H, Tachimori Y, et al. Analysis of pulmonary complications after three-field lymph node dissection for esophageal cancer. Ann Thorac Surg 2003;76:903-908. 129. Adelstein D, Rice T, Rybicki L, et al. Does paclitaxel improve the chemoradiotherapy of locoregionally advanced esophageal cancer? A nonrandomized comparison with fluorouracil-based therapy. J Clin Oncol 2000;18:2032-2039. 130. Keller S, Ryan L, Coia L, et al. High dose chemoradiotherapy ­ followed by esophagectomy for adenocarcinoma of the ­esophagus and gastroesophageal junction: results of a phase II study of the Eastern Cooperative Oncology Group. Cancer 1998;83:1908-1916. 131. Safran H, Gaissert H, Akerman P, et al. Paclitaxel, cisplatin, and concurrent radiation for esophageal cancer. Cancer Invest 2001;19:1-7. 132. Lee H, Vaporciyan A, Cox J, et al. Postoperative pulmonary complications after preoperative chemoradiation for esophageal carcinoma: correlation with pulmonary dose-volume histogram parameters. Int J Radiat Oncol Biol Phys 2003;57:1317-1322. 133. CDC. Definitions for nosocomical infections. Am Rev Respir Dis 1989;139:1058-1059. 134. Wang SL, Liao Z, Vaporciyan AA, et al. Investigation of clinical and dosimetric factors associated with postoperative pulmonary complications in esophageal cancer patients treated with concurrent chemoradiotherapy followed by surgery. Int J Radiat Oncol Biol Phys 2006;64:692-699. 135. Tucker S, Liu H, Wang S, et al. Dose-volume modeling of the risk of postoperative pulmonary complications among esophageal cancer patients treated with concurrent chemoradiotherapy followed by surgery. Int J Radiat Oncol Biol Phys 2006;66:  754-761. 136. Choi N, Park SD, Lynch T, et al. Twice-daily radiotherapy as concurrent boost technique during two chemotherapy cycles in neoadjuvant chemoradiotherapy for resectable esophageal carcinoma: mature results of phase II study. Int J Radiat Oncol Biol Phys 2004;60:111-122. 137. Urba S, Orringer M, Iannettoni M, et al. A phase II trial of preoperative cisplatin, paclitaxel, and radiation therapy (XRT) before trans-hiatal esophagectomy (THE) in patients (pts) with locoregional esophageal cancer (CA) [abstract]. Proc Am Soc Clin Oncol 2000;19:248a. 138. Blanke C, Chiappori A, Epstein B. A phase II trial of neoadjuvant paclitaxel (T) and cisplatin (P) with radiotherapy, followed by surgery (S) and postoperative T with 5-fluorouracil

(F) and leucovorin (L) in patients (pts) with locally advanced esophageal cancer (LAEC) [abstract]. Proc Am Soc Clin Oncol 1997;16:283. 139. Liu HH, Blackmon S, Gayed I, et al. Cardiac toxicity from chemoradiation with or without surgery for patients with ­esophageal cancer [abstract]. Int J Radiat Oncol Biol Phys 2006;66:  S285-S286. 140. Gayed I, Liu H, Wei X. The prevalence of myocardial ischemia after concurrent chemoradiation therapy as detected by gated myocardial perfusion imaging in patients with esophageal cancer. J Nucl Med 2006;47:1756-1762. 141. Malaisrie S, Hofstetter W, Correa A, et al. Endoscopic ­ultrasonography-identified celiac adenopathy remains a poor prognostic factor despite preoperative chemoradiotherapy in esophageal adenocarcinoma. J Thorac Cardiovasc Surg 2006;131:65-72. 142. Swisher S, Hofstetter W, Wu T, et al. Proposed revision of the esophageal cancer staging system to accommodate pathologic response (pP) following preoperative chemoradiation (CRT). Ann Surg 2005;241:810-817. 143. Gu Y, Swisher S, Ajani J, et al. The number of lymph nodes with metastasis predicts survival in patients with esophageal or esophagogastric junction adenocarcinoma who receive preoperative chemoradiation. Cancer 2006;106:1017-1025. 144. Mariette C, Taillier G, Van Seuningen I, et al. Factors affecting postoperative course and survival after en bloc resection for esophageal carcinoma. Ann Thorac Surg 2004;78:1177-1183. 145. Christie N, Rice T, DeCamp M, et al. M1a/M1b esophageal carcinoma: clinical relevance. J Thorac Cardiovasc Surg 1999;118:900-907. 146. Eloubeidi M, Wallace M, Hoffman B, et al. Predictors of survival for esophageal cancer patients with and without celiac axis lymphadenopathy: impact of staging endosonography. Ann Thorac Surg 2001;72. 212-120. 147. Catalano M, Alcocer E, Chak A, et al. Evaluation of metastatic celiac axis lymph nodes in patients with esophageal carcinoma: accuracy of EUS. Gastrointest Endosc 1999;50:352-356. 148. Eloubeidi MA, Wallace MB, Reed CE, et al. The utility of EUS and EUS-guided fine needle aspiration in detecting celiac lymph node metastasis in patients with esophageal cancer: a singlecenter experience. Gastrointest Endosc 2001;54:714-719. 149. Geh J, Crellin A, Glynne-Jones R. Preoperative (neoadjuvant) chemoradiotherapy in esophageal cancer. Br J Surg 2001;88:338356. 150. Hennequin C, Gayet B, Sauvanet A, et al. Impact on survival of surgery after concomitant chemoradiotherapy for locally advanced cancers of the esophagus. Int J Radiat Oncol Biol Phys 2001;49:657-664. 151. Hoffman PC, Haraf DJ, Ferguson MK, et al. Induction chemotherapy, surgery, and concomitant chemoradiotherapy for carcinoma of the esophagus: a long-term analysis. Ann Oncol 1998;9:647-651. 152. Luketich J, Nguyen N, Ramanathan R, et al. Induction and postoperative chemotherapy with paclitaxel, 5-fluorourcial and cisplatin regimen for carcinoma of the esophagus [abstract]. Proc Am Soc Clin Oncol 1998;17:295a. 153. Lynch T, Choi N, Wright M, et al. A phase I/II trial of preoperative taxol (t), 5-FU (f) and concurrent boost radiation (XRT) in esophageal cancer [abstract]. Proc Am Soc Clin Oncol 1997;16:261a. 154. Meluch A, Hainsworth J, Gray J, et al. Preoperative combined modality therapy with paclitaxel, carboplatin, prolonged infusion 5-fluorouracil, and radiation therapy in localized esophageal cancer: preliminary results of a Minnie Pearl Cancer Research Network phase II trial. Cancer J Sci Am 1999;5:84-91. 155. Weiner L, Colarusso P, Goldberg M, et al. Combined modality therapy for esophageal cancer: phase I trial of escalating doses of paclitaxel in combination with cisplatin, 5-fluorouracil, and high-dose radiation before esophagectomy. Semin Oncol 1997;24(Suppl 19):93-95. 156. Stahl M, Wilke H, Fink U, et al. Combined preoperative chemo­ therapy and radiotherapy in patients with locally advanced ­esophageal cancer: interim analysis of a phase III trial. J Clin Oncol 1996;14:829-837.

20  The Esophagus 157. Ganem G, Dubray B, Raoul Y, et al. Concomitant chemoradiotherapy followed, where feasible, by surgery for cancer of the esophagus. J Clin Oncol 1997;15:701-711. 158. Forastiere A, Heitmiller R, Lee D, et al. Intensive chemoradiation followed by esophagectomy for squamous cell and adenocarcinoma of the esophagus. Cancer J Sci Am 1997;3:144-152. 159. Chirieac L, Swisher S, Ajani J, et al. Posttherapy pathologic stage predicts survival in patients with esophageal carcinoma receiving preoperative chemoradiation. Int J Radiat Oncol Biol Phys 2005;103:1347-1355. 160. Urba S, Orringer M, Turrisi A, et al. Randomized trial of ­preoperative chemoradiation versus surgery alone in patients with locoregional esophageal carcinoma. J Clin Oncol 2001;19:  305-313. 161. Flamen P, Lerut A, Van Cutsem E, et al. Utility of positron emission tomography for the staging of patients with potentially operable esophageal carcinoma. J Clin Oncol 2000;18:3202-3210. 162. Luketich JD, Friedman DM, Weigel TL, et al. Evaluation of distant metastases in esophageal cancer: 100 consecutive ­positron emission tomography scans. Ann Thorac Surg 1999;68:  1133-1136. 163. Hong D, Lunagomez S, Kim E, et al. Value of baseline positron emission tomography for predicting overall survival in patient with nonmetastatic esophageal or gastroesophageal junction carcinoma. Cancer 2005;104:1620-1626. 164. Wilkinson N, Black J, Roukhadze E, et al. Epidermal growth factor receptor expression correlates with histologic grade in resected esophageal adenocarcinoma. J Gastrointest Surg 2004;8. 448-443. 165. Arber N, Gammon M, Hibshoosh H, et al. Overexpression of cyclin D1 occurs in both squamous carcinomas and adenocarcinomas of the esophagus and in adenocarcinomas of the stomach. Hum Pathol 1999;30:1087-1092. 166. Sarbia M, Stahl M, Fink U, et al. Prognostic significance of cyclin D1 in esophageal squamous cell carcinoma patients treated with surgery alone or combined therapy modalities. Int J Cancer 1999;84:86-91. 167. Izzo J, Malhotra U, Wu T, et al. Impact of cyclin D1 A870G polymorphism in esophageal adenocarcinoma tumorigenesis. Semin Oncol 2005;32(Suppl 9):S11-S15. 168. Roncalli M, Bosari S, Marchetti A, et al. Cell cycle-related gene abnormalities and product expression in esophageal carcinoma. Lab Invest 1998;78:1049-1057. 169. Boynton RF, Huang Y, Blount PL, et al. Frequent loss of heterozygosity at the retinoblastoma locus in human esophageal cancers. Cancer Res 1991;51:5766-5769. 170. Safran H, DiPetrillo T, Nadeem A, et al. Trastuzumab, paclitaxel, cisplatin, and radiation for adenocarcinoma of the esophagus: a phase I study. Cancer Invest 2004;22:670-677. 171. Yoshida K, Kuniyasu H, Yasui W, et al. Expression of growth ­factors and their receptors in human esophageal carcinomas: regulation of expression by epidermal growth factor and ­transforming growth factor alpha. J Cancer Res Clin Oncol 1993;119:  401-407. 172. Saad RS, El-Gohary Y, Memari E, et al. Endoglin (CD105) and vascular endothelial growth factor as prognostic markers in esophageal adenocarcinoma. Hum Pathol 2005;36:955-961. 173. Kleespies A, Bruns C, Jauch K. Clinical significance of VEGFA, -C, and -D expression in esophageal malignancies. Onkologie 2005;28:281-288. 174. Driessen A, Landuyt W, Pastorekova S, et al. Expression of carbonic anhydrase IX (CA IX), a hypoxia-related protein, rather than vascular-endothelial growth factor (VEGF), a pro-angiogenic  factor, correlates with an extremely poor prognosis in esophageal and gastric adenocarcinomas. Ann Surg 2006;243:334-340. 175. Sivula A, Buskens C, van Rees B, et al. Prognostic role of  cyclooxygenase-2 in neoadjuvant-treated patients with ­squamous cell carcinoma of the esophagus. Int J Cancer 2005;116:  903-908. 176. Kaur T, Khanduja K, Kaushik T, et al. P53, COX-2, iNOS protein expression changes and their relationship with anti-oxidant enzymes in surgically and multi-modality treated esophageal carcinoma patients. J Chemother 2006;18:74-84.

491

177. Wong D, Barrett M, Stoger R, et al. P16ink4a promoter is hypermethylated at a high frequency in esophageal adenocarcinomas. Cancer Res 1997;57:2619-2622. 178. Klump B, Hsieh C, Holzmann K, et al. Hypermethylation of the CDKN2/p16 promoter during neoplastic progression in Barrett’s esophagus. Gastroenterology 1998;115:1381-1386. 179. Chang M, Lee H, Lee B, et al. Differential protein expression between esophageal squamous cell carcinoma and dysplasia, and prognostic significance of protein markers. Pathol Res Pract 2005;201:417-425. 180. Huang H, Wang L, Tian H, et al. Expression of retinoic acid ­receptor-beta mRNA and p16, p53, Ki67 proteins in esophageal carcinoma and its precursor lesions. Zhonghua Zhong Liu Za Zhi 2005;27:152-155. 181. Huang XP, Zhao CX, Li QJ, et al. Alteration of RPL14 in squamous cell carcinomas and preneoplastic lesions of the esophagus. Gene 2006;366:161-168. 182. Koyanagi K, Ozawa S, Ando N, et al. Clinical significance of telomerase activity in the non-cancerous epithelial region of oesophageal squamous cell carcinoma. Br J Surg 1999;86:674-679. 183. Morales C, Lee C, Shay J. In situ hybridization for the detection of telomerase RNA in the progression from Barrett’s esophagus to esophageal adenocarcinoma. Cancer 1998;83:652-659. 184. Chan KW, Lee PY, Lam AKY, et al. Clinical relevance of Fas expression in esophageal squamous cell carcinoma. J Clin Pathol 2006;59:101-104. 185. Nishimori T, Tomonaga T, Matsushita K, et al. Proteomic analysis of primary esophageal squamous cell carcinoma reveals downregulation of a cell adhesion protein, periplakin. Proteomics 2006;6:1011-1018. 186. Tohma T, Okazumi S, Makino H, et al. Overexpression of glucose transporter 1 in esophageal squamous cell carcinomas: a marker for poor prognosis. Dis Esophagus 2005;18:185-189. 187. Kaifi JT, Yekebas EF, Schurr P, et al. Tumor-cell homing to lymph nodes and bone marrow and CXCR4 expression in esophageal cancer. J Natl Cancer Inst 2005;97:1840-1847. 188. Harpole DH Jr, Moore MB, Herndon JE II, et al. The prognostic value of molecular marker analysis in patients treated with trimodality therapy for esophageal cancer. Clin Cancer Res 2001;7:562-569. 189. Duhaylongsod F, Gottfried M, Iglehart J, et al. The significance of c-erb B-2 and p53 immunoreactivity in patients with adenocarcinoma of the esophagus. Ann Surg 1995;221:677-684. 190. Gibson MK, Abraham SC, Wu TT, et al. Epidermal growth factor receptor, p53 mutation, and pathological response predict survival in patients with locally advanced esophageal cancer treated with preoperative chemoradiotherapy. Clin Cancer Res 2003;9:6461-6468. 191. Heeren P, Plukker J, van Dullemen H, et al. Prognostic role of cyclooxygenase-2 expression in esophageal carcinoma. Cancer Lett 2005;225:283-289. 192. Walsh T, Grannell M, Mansoor S. Predictive factors for success of neo-adjuvant therapy in upper gastrointestinal cancer. J Gastroenterol Hepatol 2002;17(Suppl):S172-S175. 193. Ribeiro UJ, Finkelstein S, Safatle-Ribeiro A, et al. P53 sequence analysis predicts treatment response and outcome of patients with esophageal carcinoma. Cancer 1998;83:7-18. 194. Kanamoto A, Kato H, Tachimori Y, et al. No prognostic significance of p53 expression in esophageal squamous cell carcinoma. J Surg Oncol 1999;72:94-98. 195. Hickey K, Grehan D, Reid I, et al. Expression of epidermal growth factor receptor and proliferating cell nuclear antigen predicts response of esophageal squamous cell carcinoma to chemoradiotherapy. Cancer 1994;74:1693-1698. 196. Nicholson R, Gee J, Harper M. EGFR and cancer prognosis. Eur J Cancer 2001;37(Suppl 4):S9-S15. 197. Hirai T, Kuwahara M, Yoshida K, et al. Clinical results of transhiatal esophagectomy for carcinoma of the lower thoracic esophagus according to biological markers. Dis Esophagus 1998;11:221-225. 198. Itakura Y, Sasano H, Shiga C, et al. Epidermal growth factor receptor overexpression in esophageal carcinoma. An immunohistochemical study correlated with clinicopathologic findings and DNA amplification. Cancer 1994;74:795-804.

492

PART 6  Gastrointestinal Tract

199. Kitagawa Y, Ueda M, Ando N, et al. Further evidence for prognostic significance of epidermal growth factor receptor gene amplification in patients with esophageal squamous cell carcinoma. Clin Cancer Res 1996;2:909-914. 200. Ito T, Kaneko K, Makino R, et al. Prognostic value of p53 mutations in patients with locally advanced esophageal carcinoma treated with definitive chemoradiotherapy. J Gastroenterol 2001;36:303-311. 201. Leibel S, Fuks Z, Zelefsky M, et al. Intensity-modulated radiotherapy. Cancer J 2002;8:164-176. 202. Liu H, Wang X, Dong L, et al. Feasibility of sparing lung and other thoracic structures with intensity-modulated radiotherapy for non-small-cell lung cancer. Int J Radiat Oncol Biol Phys 2004;58:1268-1279. 203. Lujan A, Mundt A, Yamada S, et al. Intensity-modulated radiotherapy as a means of reducing dose to bone marrow in gynecologic patients receiving whole pelvic radiotherapy. Int J Radiat Oncol Biol Phys 2003;57:516-521. 204. Murshed H, Liu H, Liao Z, et al. Dose and volume reduction for normal lung using intensity-modulated radiotherapy for advanced-stage non-small-cell lung cancer. Int J Radiat Oncol Biol Phys 2004;58:1258-1267. 205. Ahmed R, Kim R, Duan J, et al. IMRT dose escalation for positive para-aortic lymph nodes in patients with locally advanced cervical cancer while reducing dose to bone marrow and other organs at risk. Int J Radiat Oncol Biol Phys 2004;60:505-512. 206. Nutting CM, Corbishley CM, Sanchez-Nieto B, et al. Potential improvements in the therapeutic ratio of prostate cancer irradiation: dose escalation of pathologically identified tumour nodules using intensity modulated radiotherapy. Br J Radiol 2002;75:151-161. 207. Chao K, Majhail N, Huang C, et al. Intensity-modulated radiation therapy reduces late salivary toxicity without compromising tumor control in patients with oropharyngeal carcinoma: a comparison with conventional techniques. Radiother Oncol 2001;61:275-280. 208. Eisbruch A, Kim H, Terrell J, et al. Xerostomia and its predictors following parotid-sparing irradiation of head-and-neck cancer. Int J Radiat Oncol Biol Phys 2001;50:695-704. 209. Mundt A, Mell L, Roeske J. Preliminary analysis of chronic gastrointestinal toxicity in gynecology patients treated with ­intensity-modulated whole pelvic radiation therapy. Int J Radiat Oncol Biol Phys 2003;56:1354-1360. 210. Kupelian P, Reddy C, Carlson T, et al. Preliminary observations on biochemical relapse-free survival rates after short-course intensity-modulated radiotherapy (70 Gy at 2.5 Gy/fraction) for localized prostate cancer. Int J Radiat Oncol Biol Phys 2002;53:904-912. 211. Zelefsky M, Fuks Z, Hunt M, et al. High-dose intensity modulated radiation therapy for prostate cancer: early toxicity and biochemical outcome in 772 patients. Int J Radiat Oncol Biol Phys 2002;52:1111-1116. 212. Hall E. Radiobiology for the Radiologist, 4th ed. New York: Lippincott Williams & Wilkins, 1993. 213. Hug E, Slater J. Proton radiation therapy for chordomas and chondrosarcomas of the skull base. Neurosurg Clin N Am 2000;11:627-638. 214. Sugahara S, Tokuuye K, Okumura T, et al. Clinical results of proton beam therapy for cancer of the esophagus. Int J Radiat Oncol Biol Phys 2005;61:76-84. 215. Beyer T, Townsend D, Brun T, et al. A combined PET/CT scanner for clinical oncology. J Nucl Med 2000;41:1369-1379. 216. Kluetz P, Meltzer C, Villemagne V, et al. Combined PET/CT imaging in oncology. Impact on patient management. Clin Positron Imaging 2000;3:223-230.

217. Bar-Shalom R, Yefremov N, Guralnik L, et al. Clinical performance of PET/CT in evaluation of cancer: additional value for diagnostic imaging and patient management. J Nucl Med 2003;44:1200-1209. 218. Hany TF, Steinert HC, Goerres GW, et al. PET diagnostic accuracy: improvement with in-line PET-CT system: initial results. Radiology 2002;225:575-581. 219. Charron M, Beyer T, Bohnen N, et al. Image analysis in patients with cancer studied with a combined PET and CT scanner. Clin Nucl Med 2000;25:905-910. 220. Lardinois D, Weder W, Hany TF, et al. Staging of non-small-cell lung cancer with integrated positron-emission tomography and computed tomography. N Engl J Med 2003;348:2500-2507. 221. Schoder H, Erdi Y, Larson S, et al. PET/CT: a new imaging technology in nuclear medicine. Eur J Nucl Med Mol Imaging 2003;30:1419-1437. 222. Townsend DW, Carney JPJ, Yap JT, et al. PET/CT today and tomorrow. J Nucl Med 2004;45:4S-14. 223. Scarfone C, Lavely WC, Cmelak AJ, et al. Prospective feasibility trial of radiotherapy target definition for head and neck cancer using 3-dimensional PET and CT imaging. J Nucl Med 2004;45:543-552. 224. Ashamalla H, Rafla S, Parikh K, et al. The contribution of integrated PET/CT to the evolving definition of treatment volumes in radiation treatment planning in lung cancer. Int J Radiat Oncol Biol Phys 2005;63:1016-1023. 225. Brahme A. Recent advances in light ion radiation therapy. Int J Radiat Oncol Biol Phys 2004;58:603-613. 226. Grills I, Yan D, Martinez A, et al. Potential for reduced toxicity and dose escalation in the treatment of inoperable non-smallcell lung cancer: a comparison of intensity-modulated radiation therapy (IMRT), 3D conformal radiation, and elective nodal irradiation. Int J Radiat Oncol Biol Phys 2003;57:875-890. 227. Grosu A-L, Piert M, Weber WA, et al. Positron emission tomography for radiation treatment planning. Strahlenther Onkol 2005;181:483-499. 228. Moureau-Zabotto L, Touboul E, Lerouge D, et al. Impact of CT and 18F-deoxyglucose positron emission tomography image fusion for conformal radiotherapy in esophageal carcinoma. Int J Radiat Oncol Biol Phys 2005;63:340-345. 229. Mandard AM, Marnay J, Gignoux M, et al. Cancer of the esophagus and associated lesions: detailed pathologic study of 100 esophagectomy specimens. Hum Pathol 1984;15:660-669. 230. Kato H, Tachimori Y, Watanabe H, et al. Intramural metastases of thoracic esophageal carcinoma. Int J Cancer 1992;50:49-52. 231. Mauer AM, Weichselbaum RR. Multimodality therapy for carcinoma of the esophagus. In Posner MC, Vokes EE, Weischelbaum RR (eds): Cancer of the Upper Gastrointestinal Tract. In Steele GD, Phillips TL, Chabner BA (series eds): American Cancer ­Society Atlas of Clinical Oncology. London: BC Decker, 2002, pp 157-183. 232. Miller C. Carcinoma of thoracic oesophagus and cardia. A review of 405 cases. Br J Surg 1962;49:507-522. 233. Elkon D, Lee MS, Hendrickson FR. Carcinoma of the esophagus: sites of recurrence and palliative benefits after definitive radiotherapy. Int J Radiat Oncol Biol Phys 1978;4:615-620. 234. Herskovic A, Leichman L, Lattin P, et al. Chemo/radiation with and without surgery in the thoracic esophagus: the Wayne State experience. Int J Radiat Oncol Biol Phys 1988;15:655-662. 235. Konski A. Preliminary analysis of incorporating PET and MRI scans into the treatment planning process for esophageal carcinoma [abstract]. Proceedings of the 85th annual meeting of the American Radium Society, Houston, Texas; 2003. Cancer J 2003:9:496.

The Stomach and Small Intestine 21 Christopher H. Crane and Sunil Krishnan STOMACH Normal Anatomy and Histology Acute Reactions to Radiation Therapy Late Effects of Radiation Therapy EPIDEMIOLOGY OF GASTRIC CANCER ETIOLOGY OF GASTRIC CANCER Environmental Factors Host-Related Factors Infectious Factors SCREENING PATTERNS OF SPREAD PRESENTATION AND WORKUP STAGING PATHOLOGY PROGNOSTIC FACTORS SURGICAL THERAPY ADJUVANT THERAPY Historical Perspective Postoperative Chemotherapy, Radiation Therapy, or Chemoradiation Therapy Postoperative Chemotherapy Preoperative Chemotherapy

Preoperative Chemotherapy, Radiation Therapy, and Chemoradiation Therapy Chemoradiation Therapy for Unresectable Gastric Cancer Supportive Care Palliative Radiation Therapy Palliative Chemotherapy RADIATION TECHNIQUES Current Techniques Ongoing Investigations SMALL INTESTINE Anatomy and Histology Acute Effects of Radiation Late Effects of Radiation Therapy EPIDEMIOLOGY AND ETIOLOGY OF SMALL BOWEL CANCER STAGING THE ROLE OF RADIATION THERAPY Adenocarcinomas Sarcomas Carcinoid Tumors Choice of Radiation Techniques Summary of Radiation Therapy for Small Bowel Tumors

Stomach Normal Anatomy and Histology The stomach is divided into four major anatomic regions: cardia, fundus, body, and pylorus (Fig. 21-1). The gastroesophageal junction is located just to the left of midline at the level of the 11th thoracic vertebral body, and the pylorus is located to the right of midline at the level of the first lumbar vertebral body. The pylorus empties into the first portion of the duodenum. The greater and lesser curvatures form attachments with the greater omentum and lesser omentum, respectively. The undistended stomach relaxes into longitudinal folds known as rugae, which flatten out when the stomach is filled with food. The stomach has three histologically distinct regions: cardia, body or fundus, and pylorus or antrum. Throughout the stomach, the mucosa consists of surface epithelium that invaginates to various degrees into the lamina propria to form gastric pits, which end blindly in glands with neck and base regions (see Fig. 21-1). The undifferentiated cells of the neck region of the glands migrate to the surface to form mucus-producing cells or deeper into the glands to differentiate into mucous neck cells, parietal cells, enteroendocrine cells, or chief cells. The structure of the pits is distinct in each region of the stomach. In the cardia region, mucus-producing glands predominate, whereas the fundus

and body are rich in parietal and chief cells. Parietal cells are eosinophilic and tend to be pyramidal. They produce and secrete hydrochloric acid and intrinsic factor, the ­latter of which is essential for vitamin B12 absorption. The chief cells secrete pepsinogen, a proteolytic proenzyme that is contained in characteristic zymogen granules. The ­antrum is rich in gastrin-secreting G cells and somatostatin­secreting D cells, which have stimulatory and ­ inhibitory effects, ­respectively, on gastric acid secretion.

Acute Reactions to Radiation Therapy Before modern pharmacology, radiation therapy was used to treat peptic ulcer disease. The rationale for treatment was based on the empiric observation that radiation therapy reduces gastric acid secretion. This practice stimulated investigations of the physiologic effects of gastric irradiation.1 Important observations were recorded that allowed insight into the acute reactions of the stomach to radiation therapy. To understand the histologic changes that occur with irradiation, Goldgraber and colleagues2 conducted serial biopsies in three patients treated with 1600 rad (16 Gy) in 10 equal fractions to the entire stomach through anteroposterior-posteroanterior portals with 250-kV x-rays. The biopsies showed a loss of cytoplasmic detail and a disappearance of cytoplasmic granules in the parietal and chief cells as early as 8 days after initial radiation ­treatment.

493

494

PART 6  Gastrointestinal Tract

Esophagus Pit

Pit

Fundus

Junction of gland and base of pit

Cardia

Lamina propria Glands Muscularis mucosae

Body Cardia

P y l o r u s

Pyloric canal Duodenum

Pyloric sphincter

Pit

Pit Neck Neck Gland Gland Gland Lamina propria Lymph nodule Pylorus

Muscularis mucosae Submucosa

Base Body

FIGURE 21-1.  The four major anatomic regions of the stomach are the cardia, the fundus, the body, and the pylorus (antrum). The structure of the gastric pits is distinct in each region.

The cytoplasm became glassy and eosinophilic, and the nuclei of these cells became elongated and pyknotic. These changes were called coagulation necrosis, and they occurred with a relative sparing of the mucosal surface. By day 16 after the initiation of irradiation, the architecture of the mucosal epithelium began to change. The glandular pits began to deepen, many of the parietal and chief cells sloughed into the gastric lumen, and cells at the glandular neck started to proliferate. At this point, other than some cuboidal change, changes in the surface columnar epithelium were still minimal. Three weeks after initiation of therapy, marked architectural disorganization, a patchy loss of glandular architecture, and an overall loss of mucosal thickness were seen. The surface epithelial cells varied in

appearance from flat to columnar (Fig. 21-2).2 Regeneration began in the glandular neck cells. At 5 weeks, the mucosal and glandular architecture began to revert to normal. By week 6, nearly all radiation changes had resolved, with only focal areas of glandular coagulation necrosis being evident. Although the same dose was used in each of the three patients treated, the time of onset of the earliest changes varied between 5 and 16 days. The severity and duration of the changes also varied. Palmer and Templeton3 studied the gastric secretory response in 88 patients treated at the University of Chicago with about 1500 rad (15 Gy, a relatively low dose of radiation) in 10 equal fractions for peptic ulcer disease. Basal gastric acid production and histamine-induced gastric acid

21  The Stomach and Small Intestine

A

495

B

FIGURE 21-2.  A, Endoscopic biopsy sample of normal gastric mucosa. (×110). B, Sample of gastric mucosa 3 weeks after treatment with 15 Gy given over 10 days. (×100). (From Goldgraber M, Rubin C, Palmer W, et al. The early gastric response to irradiation: a serial biopsy study. Gastroenterology 1954;27:1-20.)

r­ esponse were measured. Achlorhydria was documented in 35 patients and varied in duration from a few days to 6 months. Unable to find a correlation of the inhibitor effect with sex or body weight, the investigators hypothesized the existence of differences in parietal cell radiation sensitivity among individuals. Overall, 1500 patients were given similar doses in the University of Chicago series, and 40% experienced a 50% reduction in gastric acid secretion that lasted for a year or more.4 Gastric secretory function tended to parallel the observed glandular changes. Gastric acid was present at fasting concentrations 2 days after the start of therapy but undetectable thereafter. Secretory studies using histamine showed achlorhydria at 2 weeks and an attenuated induction of the peak response of gastric acid by histamine at 12 weeks, although by then the mucosa had reverted to normal.2 The critical structures in a typical radiation field for gastric carcinoma include the stomach, small intestine, liver, kidneys, and spinal cord. The acute reactions that occur with irradiation of these structures include gastritis (epigastric pain), nausea, anorexia, diarrhea, abdominal cramping, fatigue, and hematologic suppression. Most patients experience some degree of nausea and fatigue and a moderate to severe amount of hematologic suppression depending on the type of concurrent chemotherapy used. Acute reactions usually peak during the fourth week of treatment and subside 1 to 2 weeks after completion of a 5-week course of treatment.

Late Effects of Radiation Therapy The effects of fractionated radiation therapy (up to 60 Gy) on the canine stomach have been demonstrated. Perforating gastric ulcers are seen within 4 weeks of irradiation.5 In humans, gastritis, ulceration, and obstruction are unusual at doses of less than 40 Gy but have been documented with a

sharp increase in incidence above 45 Gy in patients irradiated for gastric ulcer using low-energy photons and suboptimal technique. At Walter Reed Army Medical Center, 35 (14%) of 256 patients who underwent prophylactic irradiation for malignant germ cell tumor experienced gastric ulceration 2 to 3 months after completion of treatment.6 The rates of ulceration, perforation, or obstruction in this setting have been observed to be 25% to 50% at 55 Gy and 63% at doses between 55 and 64 Gy, with perforation occurring in one fourth of the patients. Most of these patients were treated with 200-kV to 1-MV photons, short source-to-skin distances, large fractions, and only one field per day. Although these data illustrate the sensitivity of the upper abdomen to irradiation, they should be interpreted in the context of the technical inadequacies. The use of modern linear accelerators and techniques has produced a very low risk of gastric injury with doses of less than 50 Gy.1,7-11 A dose-response relationship does seem to exist for radiation injury between 50 and 60 Gy. Morbidity risks of less than 10% are expected for 50 to 55 Gy and 5% to 15% for doses in excess of 60 Gy.7,8,12-23 For the complications that do not require gastrectomy,24 healing of radiation-induced lesions is characterized by re-epithelialization with fibrosis and contraction in 2 to 3 years.23 The other critical structures in the region include the kidneys, liver, small intestine, and spinal cord. Mean decreases in creatinine clearance of 10% and 24% have been observed in patients when 50% and 90% to 100%, respectively, of a single kidney is irradiated with doses higher than 26 Gy.25 However, clinically relevant compromise of kidney function rarely develops if at least one kidney is spared from the field. Radiation-induced liver disease (RIND), often called radiation hepatitis, is characterized by the development of

496

PART 6  Gastrointestinal Tract

anicteric ascites approximately 2 weeks to 4 months after hepatic irradiation.26 The whole liver has been treated safely to doses of more than 20 Gy in patients with pancreatic cancer,27 but whole-liver doses of more than 35 Gy have resulted in a significant risk of RIND.28 Although the risk of RIND can be predicted from dose and volume considerations,29,30 it is clear that limited volumes can tolerate high doses of irradiation.31 Radiation effects on the small bowel are discussed under “Small Intestine” later in this chapter; effects on the spinal cord are discussed in Chapter 33.

Epidemiology of Gastric Cancer Since the 1930s, when gastric cancer was the leading cause of cancer-specific mortality among men in the United States,32 the overall incidence rates have dropped dramatically, and the location of presenting lesions has changed from noncardia to cardia sites.33 In 2008, an estimated 21,500 people in the United States were diagnosed with gastric cancer, with a male-to-female ratio of 1.67:1.34 Blacks, Hispanics, and Native Americans are 1.5 to 2.5 times more likely to develop gastric cancer than are whites in the United States.35 Although the same trend of declining incidence is seen throughout the world, gastric cancer is still the second leading cause of cancer-specific mortality worldwide.36 It is the most common malignancy in ­Japan, and it is also common in China, South America, and ­Eastern Europe.37 The decline in the incidence of gastric cancer from 1930 to 1976 has paralleled a decline in the incidence of distal gastric lesions. This phenomenon lags 20 to 30 years behind the widespread availability of refrigeration, and this factor has been proposed to account for the sharp decline in gastric carcinoma.38-40 Since 1976, however, the incidence of proximal cardia and gastroesophageal junction adenocarcinoma in the United States and Europe have increased at a rate that exceeds that of any other malignancy.41 The rise in incidence of gastroesophageal junction and cardia lesions suggests a common pathogenesis distinct from that of distal gastric lesions, but the cause remains unclear. Gastric cancer is of two general histologic types—an intestinal type in which cells are closely adherent and tend to form glands and a diffuse type that tends to infiltrate and thicken the gastric wall without forming a discrete mass.42 The intestinal type is more common, is more often distal, and predominates in endemic areas. The diffuse type has a worse prognosis, tends to occur in younger patients, and can occur anywhere in the stomach but is most common in the cardia.43,44 A separate, similar classification, Borrmann’s classification, distinguishes five types. The first three types are analogous to the intestinal type with increasing degrees of ulceration, the fourth is diffusely infiltrating, and the fifth is unclassifiable.45

Etiology of Gastric Cancer The shift in anatomic site of origin in gastric cancer has stimulated much epidemiologic research. Chronic atrophic gastritis and the resulting intestinal metaplasia seem to be precursor conditions to the intestinal type of gastric ­cancer.46 Mucosal changes brought about by a variety of

environmental insults can eventually lead to atrophic gastritis. Environmental, host-related, and infectious causes may all play roles.

Environmental Factors Environmental factors, especially early in life, probably determine the risk of stomach cancer.47,48 The widespread availability of refrigeration has paralleled the reduction in the incidence of gastric carcinoma.38 Refrigeration may have resulted in reduced exposure to various carcinogens such as nitrates and nitrites used to prevent bacterial and fungal contamination of food. Another possibility is that refrigeration has allowed an increase in dietary intake of fresh fruit and a reduction in the consumption of smoked, cured, and salted foods. These dietary factors have correlated with lower gastric carcinoma rates in epidemiologic studies.39,49-61 In a prospective study of Japanese residents of Hawaii, consumption of pickled, processed, or salted foods was not associated with gastric cancer, but frequent consumption of fresh fruit and raw vegetables was inversely related to the incidence of gastric cancer.62 Serum vitamin C levels did not correlate with gastric cancer in a multicenter study,63 although high or low intake of other micronutrients (e.g., beta-carotene, ascorbic acid, methionine, ferritin) could play a role.64,65 Dietary diversity, especially with regard to fruits and vegetables, has been reported to be protective.66 A positive correlation with the extent of cooking red meat has also been reported in gastric cancer patients ­interviewed for food preparation preference.67 Another study found no ­relationship between nitrate concentrations in ­ drinking ­ water and the incidence of stomach or ­esophageal ­cancer.68 An interaction between dietary methionine and salt has been suggested.69 Other environmental factors such as smoking54,55,63,70 and industrial dust exposure70 may have an association, but convincing evidence is still lacking to demonstrate a role for alcohol.50,58,62,70-74 The most compelling evidence for an environmental cause of the intestinal type of gastric cancer comes from studies of migrating populations.74 Japanese,75 South American,47 and Eastern European migrants assume a lower risk of developing gastric cancer after two to three generations spent in a lowerrisk environment with a lower-risk diet.47,74,75

Host-Related Factors Host factors such as low serum ferritin levels76 and pernicious anemia are associated with a twofold to threefold excess risk for gastric cancer.77-79 An association is evident between gastric ulcer and an increase in the risk of gastric cancer.80 Medically treated peptic ulcer disease does not seem to be associated with an increased risk, but patients who have undergone distal gastrectomy for benign peptic ulcer disease have been reported to have up to a fivefold increase in risk after a long latency period.81-83 An association between adenomatous polyps and gastric cancer has been suggested,84 and clear associations exist between Barrett’s esophagus and adenocarcinoma of the distal esophagus and gastric cardia,85,86 with the risk estimated to be 0.8% per annum. Evidence is also clear for a heritable risk of gastric cancer in a small proportion of patients. Familial predisposition74,75,87-89 seems to be independent of environmental factors.89 Controlling for dietary risk, patients with hereditary nonpolyposis colorectal cancer, but not those with familial adenomatous polyposis, are at increased risk for gastric cancer.90

21  The Stomach and Small Intestine

Infectious Factors Although an association between Helicobacter pylori infection and gastric cancer has been reported in epidemiologic studies,91 no reduction in risk after treatment of H. pylori infection has been demonstrated. Prospective serologic studies have shown that patients with H. pylori infection have a threefold to fivefold higher risk.92-94 The evidence points to a specific strain, cytotoxin-associated gene A– ­positive H. pylori,95,96 and to infection early in life.97 The increased risk has been reported to be for noncardia lesions only.98 An analysis concluded that screening is substantially more cost-effective in endemic areas (outside the United States).99 A cohort study looking at stored serum suggests that chronic Epstein-Barr virus (EBV) infection may play an etiologic role in a subset of patients.100 Because only a small proportion of patients infected with H. pylori or EBV develop malignancy, additional environmental cofactors undoubtedly contribute to the carcinogenic process.101,102

Screening Countries in which gastric cancer is endemic have implemented endoscopic screening programs. Japan, in particular, has successfully increased detection of early-stage disease, and some have suggested that such programs have reduced gastric cancer–specific mortality rates.103 Widespread implementation of screening programs has resulted in 34.5% of cases being diagnosed while disease is confined to the mucosa.103,104 The Cancer Institute Hospital of Tokyo has also reported an increase in the rate of curative resections from 48% in the 1940s to 98% during the 1980s.105 Without prospective evaluation, however, this contention cannot be confirmed because of confounding environmental variables that have also changed, such as westernization of the diet. Validation of a role for screening radiography is also lacking. A case-control study evaluating the value of screening radiography in Venezuela failed to demonstrate a positive effect.106

Patterns of Spread Gastric cancer can spread by direct extension, lymphatic or hematologic invasion, or transperitoneal dissemination. Critical organs such as the pancreas, diaphragm, transverse colon, duodenum, spleen, jejunum, liver, left kidney, left adrenal gland, and the celiac axis are in close proximity and are frequently involved. Lymphatic spread initially occurs to the perigastric nodes along the lesser and greater curvatures and then to lymphatics that accompany the branches of the celiac axis (the hepatic, subpyloric, gastroduodenal, and pancreaticosplenic nodes). Remote lymphatic spread can occur to the hepatoduodenal, peripancreatic, superior mesenteric, and para-aortic nodal chains. Similar to cancer of the esophagus and duodenum, clinically occult spread of gastric cancer beyond the gross lesion can occur through the abundant subserosal and submucosal lymphatics. Distant metastasis occurs most commonly through portal venous drainage to the liver. Systemic spread occurs less commonly to the lungs and even less often to other sites. Lesions of the gastroesophageal junction have a stronger tendency to spread to the lungs. Transperitoneal spread occurs in 23% to 43% of patients who have undergone gastrectomy.107-110

497

Clinical,108,110 reoperative,109 and autopsy series111-115 have been used to define patterns of failure in gastric ­cancer. The report of Gunderson and Sosin109 on ­Wangensteen’s second-look laparotomy series116 defined the pattern of failure for surgically treated gastric cancer. At 1 year ­after the initial surgical treatment, failures were seen in the ­anastomoses, gastric bed, gastric remnant, duodenal stump, and regional nodes; the isolated locoregional failure rate was 29%, and a locoregional component of failure was present in 88% of patients with initially localized disease (Fig. 21-3).109 A similar pattern of failure was reported in surgically treated patients from the Massachusetts General Hospital, with a locoregional component of relapse in 38% of these patients and an isolated locoregional relapse in 16%.108 The true incidence of failure was probably underestimated ­because of the limitations of imaging disease in the gastric bed, regional nodes, and peritoneum and the lack of uniformity in restaging relapsed disease. Autopsy studies have correlated well with these findings.111-115 In summary, surgically treated gastric cancer is associated with a significant isolated locoregional component of failure. Effective adjuvant or neoadjuvant therapy could be significantly beneficial.

Presentation and Workup The most common presenting symptoms of gastric cancer are upper abdominal discomfort, weakness from anemia, and weight loss.117 Early satiety, hematemesis, or melena occur less commonly. Obstructing lesions in the antrum or cardia can cause vomiting or dysphagia, respectively. Typically, presenting symptoms are nonspecific and unfortunately do not usually occur with superficially invasive disease; rather, disease usually has progressed to the point of transmural invasion or distant metastasis at the time of presentation. Palpable disease is usually indicative of advanced disease. Careful physical examination can detect metastases in the left supraclavicular fossa or those forming a Blumer’s (rectal) shelf, indicative of peritoneal dissemination. Initial workup usually includes a double-contrast upper gastrointestinal study, chest radiograph, and blood work that includes tests of liver enzymes and a differential blood cell count. Upper endoscopy is sensitive118-120 and allows tissues to be sampled but is invasive and costly. Biopsies should be multiple and deep,120 and fresh tissue samples should undergo flow cytometric analysis if lymphoma is suspected. When the diagnosis of gastric cancer is established, computed tomography (CT) scans of the abdomen with intravenous phase contrast medium are indicated. Evaluation of CT findings in comparison with those of laparotomy reveal that CT has many limitations, including poor characterization of primary disease and a lack of sensitivity in detecting peritoneal or liver metastases.121,122 Staging laparoscopy and endoscopic ultrasonography have sometimes been used in attempts to improve the ability to identify patients who can benefit from curative therapies.123,124 Accuracy rates for tumor staging of between 80% and 85% have been reported for endoscopic ­sonography.125,126 In comparing modern and historical series, it is crucial to realize the selection bias introduced by improvements in preoperative imaging, the use of peritoneal staging, and the increased use of endoscopic sonography.

498

PART 6  Gastrointestinal Tract

A

B

FIGURE 21-3.  Patterns of failure in gastric cancer were documented in a landmark report of second-look laparotomy 1 year after surgical treatment. Isolated locoregional failure was found in 29% of patients and a locoregional component of failure was found in 88% of patients. A hypothetical radiation therapy field is included in the figure, but it was not intended to be used as a standard radiotherapy port. A, Locoregional disease. B, Distant disease. Solid circles, local failures in surrounding organs or tissues; open circles, lymph node failures; asterisks, lung metastases; plus signs, liver metastases. (From Gunderson LL, Sosin H. Adenocarcinoma of the stomach: areas of failure in a re-operation series (second or symptomatic look) clinicopathologic correlation and implications for adjuvant therapy. Int J Radiat Oncol Biol Phys 1982;8:1-11.)

Staging Gastric cancer staging is based on the tumor-­node-­metastasis (TNM) system (Box 21-1).127 Although pathologic staging has traditionally been used, clinical staging has become more common with the use of endoscopic sonography and ­laparoscopic peritoneal exploration for patients undergoing neoadjuvant treatment.

Pathology Between 90% and 95% of malignant gastric tumors are adenocarcinomas, with non-Hodgkin lymphoma, leiomyosarcoma, carcinoid, adenoacanthoma, squamous cell, and neuroendocrine undifferentiated tumors also occasionally occurring. Tumor cells may contain mucin vacuoles that can distend the cell and compress the nucleus, creating the characteristic signet-ring appearance. Diffusely infiltrative tumors often stimulate a marked desmoplastic response, leading to a diffusely thickened gastric wall; this particular form of infiltrative tumor is called linitis plastica.

Prognostic Factors The most important prognostic factor for patients with gastric cancer is the TNM stage.128,129 Gross hematogenous or transperitoneal disease is incurable. The degree of transmural penetration beyond the gastric wall109,128,130-135 and the

presence, number, and location of involved lymph nodes correlate with survival.128,130-136 Patients with transmural penetration and nodal involvement have a particularly poor prognosis.109,131,134,136 Peritoneal cytology analysis has shown evidence of ­disease in 16% to 35% of patients at the time of resection.137-139 Such results have been consistently ­ reported to have a negative influence on survival in univariate ­analyses,137-143 but not all patients with cytologic evidence of disease will develop peritoneal carcinomatosis.137,138,143 Intestinal-type lesions are associated with a better prognosis than diffusely infiltrative lesions.42-44 Linitis plastica carries an extremely poor prognosis.16,17,144,145 The presence of receptors for estrogen146 on tumor cells is associated with higher rates of tumor infiltration, poorer ­differentiation, and a worse prognosis.

Surgical Therapy Since Theodore Billroth successfully performed the first gastrectomy for gastric cancer in 1881, surgery has been the cornerstone of curative therapy for this disease. The initial surgical approach for malignant disease was a partial gastrectomy that was not radically different from gastric surgery for benign disease. A more aggressive approach involving resection of the entire stomach, spleen, and distal pancreas followed a landmark autopsy series from Memorial Sloan-Kettering Cancer Center published in 1951,

21  The Stomach and Small Intestine

499

BOX 21-1.  American Joint Committee on Cancer Staging System for Carcinoma of the Stomach Primary Tumor (T) TX T0 Tis T1 T2 T2a T2b T3 T4

Primary tumor cannot be assessed No evidence of primary tumor Carcinoma in situ; intraepithelial tumor without invasion of the lamina propria Tumor invades lamina propria or submucosa Tumor invades muscularis propria or subserosa* Tumor invades muscularis propria Tumor invades subserosa Tumor penetrates serosa (visceral peritoneum) without invasion of adjacent structures†,‡ Tumor invades adjacent structures†,‡

Regional Lymph Nodes (N) NX Regional lymph node(s) cannot be assessed N0 No regional lymph node metastasis N1 Metastasis in 1 to 6 regional lymph nodes N2 Metastasis in 7 to 15 regional lymph nodes N3 Metastasis in more than 15 regional lymph nodes Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage Grouping Stage 0 Stage IA Stage IB Stage II Stage IIIA Stage IIIB Stage IV

Tis T1 T1 T2a/b T1 T2a/b T3 T2a/b T3 T4 T3 T4 T1 T2 T3 T4 T4 Any T

N0 N0 N1 N0 N2 N1 N0 N2 N1 N0 N2 N1 N3 N3 N3 N2 N3 Any N

M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M1

*A tumor may penetrate the muscularis propria with extension into the gastrocolic or gastrohepatic ligaments or into the greater or lesser omentum without perforation of the visceral peritoneum covering these structures. In this case, the tumor is classified T2. If there is perforation of the visceral peritoneum covering the gastric ligaments or the omentum, the tumor should be classified T3. †The adjacent structures of the stomach include the spleen, transverse colon, liver, diaphragm, pancreas, abdominal wall, adrenal gland, kidney, small intestine, and retroperitoneum. ‡Intramural extension to the duodenum or esophagus is classified by the depth of greatest invasion in any of these sites, including the ­stomach. From Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 99-106.

which reported 80% local failure and 50% involvement of the gastric remnant.107 Wangensteen’s initial findings at second-look laparotomy116 also prompted extensive lymphadenectomy. A prospective randomized study from France comparing total and subtotal gastrectomy found no survival difference and similar perioperative morbidity for the two procedures.147 The optimal surgical procedure has not been defined and remains controversial. For lesions arising in the gastric body and antrum, a radical subtotal gastrectomy is preferred. In this procedure, 80% of the stomach and the

perigastric lymphatics, the gastrohepatic and gastrocolic omenta, and the first portion of the duodenum are resected. Retrospective studies from Memorial Sloan-Kettering and Japan have not shown a survival benefit from the addition of splenectomy.148-151 More proximal lesions require a total gastrectomy. Rates of positive margins vary from 30% with a 2-cm margin to 0% with a 4- to 6-cm margin beyond the gross lesion.152 Mucosal margins should be at least 5 to 6 cm. Overall, the reported frequency of positive margins in patients undergoing curative resection is approximately 25%.153-157 Although this rate is similar to those reported

500

PART 6  Gastrointestinal Tract

for anastomotic and gastric remnant failures,108,109 anastomotic failure has been reported to occur in only 23% of patients with positive margins.152 Extensive lymph node dissection has been advocated by some Japanese investigators and has been reported to improve survival.110,158-160 Other investigators have found no improvement in locoregional control or survival.109,158 The impressive results in the Japanese studies, even in patients with advanced disease, seem to be improving with time and have stimulated much debate in the literature. Extended lymphadenectomy has not been embraced by Western surgeons largely because of the lack of a clear benefit and high morbidity associated with the procedure.161-166 A D0 lymphadenectomy typically involves less than complete dissection of N1 lymph nodes in the perigastric region; a D1 lymphadenectomy involves removal of all N1 nodes (i.e., perigastric lymph nodes along the lesser and greater curvature); and an extended (D2) lymphadenectomy involves removal of N2 nodes (i.e., nodes beyond the ­ perigastric area, along the branches of the celiac axis), often requiring a distal pancreatectomy and splenectomy. A Dutch randomized trial showed an increased risk of complications, postoperative death, and longer hospital stays for patients who underwent extended (D2) lymphadenectomy compared with those who had a D1 lymphadenectomy.163 Part of the explanation for the difference in morbidity rates between the Western trials and those from Japan lies in the high rates of morbid obesity in Western countries; obesity increases operative risk and creates a more technically demanding operation. The high morbidity rates likely obscure any benefit. Another explanation for the apparent difference in results could lie in the concept of stage migration. More extensive lymphadenectomies and more compulsive pathologic evaluation give rise to an apparent improvement in stage-specific survival without an overall benefit. The overall operative mortality rate in the United States has been reported to be as high as 7% in some centers117 but is typically 1% to 3% in more experienced hands.164 Modern surgery is highly successful for patients with early-stage disease, with 5-year survival rates of 80% to 90% for those with T1 disease and 50% to 60% for those with T2 disease.103,128,129,167 Unfortunately, in the United States, less than 5% of patients have T1 or T2 disease. Potentially curative resection is possible in only 25% to 40% of cases that are judged to be resectable by clinical staging.168 Complete gross resection without adjuvant therapy has resulted in median survival durations of 15 to 18 months and 5-year survival rates of 20% to 30% in Western populations.117,128,130,155,169,170

Adjuvant Therapy Historical Perspective With the advent of modern surgical techniques and the recognition of surgery’s limitations, adjuvant therapies began to be evaluated. Radiation therapy was a key component of initial investigations. Unfortunately, during that early era, adequate technique, technology, and especially supportive care were not available or not used. This led to widespread toxic effects and consequent failure to define a role for radiation therapy in gastric cancer. This left an opportunity for the investigation of adjuvant chemotherapy. Over

the next decade, adjuvant chemotherapy was exhaustively evaluated. From that experience, substantial toxicity was demonstrated with intensive chemotherapy, but a survival benefit could not be established. This has led some centers to evaluate induction (neoadjuvant) chemotherapy. With intensive and toxic regimens, single-arm, prospective studies have shown modest response rates and unimpressive pathologic complete response rates. These findings led to the inclusion of radiation, the single most active agent against gastric cancer, in the most recent single-arm prospective studies. With vital supportive care, very large radiation fields with potent concurrent radiosensitizers are very well tolerated, and a substantial increase in the pathologic complete response rate has been observed. In the future, phase III studies are imperative, but the relevant questions to be addressed by such trials will be under considerable debate. The relative roles of neoadjuvant chemotherapy and concurrent chemoradiation therapy must be established.

Postoperative Chemotherapy, Radiation Therapy, or Chemoradiation Therapy The patterns of failure discussed previously (see Fig. 21-3) served as the rational basis for the use of postoperative ­ radiation therapy in gastric cancer. Several single­institution experiences suggest that postoperative ­radiation confers an advantage with or without concurrent chemotherapy.154,157,171-173 One randomized study from Japan174 showed that intraoperative radiation therapy with 28 to 35 Gy conferred a survival advantage for patients with stage II to IV gastric cancer but not for those with stage I gastric cancer. Although methodologically flawed, this study offered proof of principle for intraoperative ­ radiation therapy in cancer treatment. A smaller subsequent study also showed that intraoperative radiation conferred a significant local control advantage.175 However, results of another trial by the British Stomach Cancer Group shifted the direction of subsequent investigations toward the use of chemotherapy for gastric cancer. In that phase III trial, 436 patients were stratified by age, symptom duration, and disease stage and then randomized to undergo surgery alone, surgery followed by chemotherapy (with agents such as 5-fluorouracil [5-FU], doxorubicin, or mitomycin), or surgery followed by radiation therapy (45 Gy in 25 fractions with or without a 5-Gy boost).155,176 A locoregional control advantage was demonstrated in the radiation therapy group (90% versus 73% for surgery alone and 81% for ­ surgery ­followed by chemotherapy, P < .01), but cause-­specific and overall survival rates were not statistically different among the groups. Despite several confounding factors and flaws in study design (e.g., one third of patients had residual disease, 18% had positive margins, and one third received 40 Gy or less) and despite the lack of inclusion of a chemoradiation treatment group, the results of this trial influenced subsequent investigations of strategies that did not include radiation therapy for gastric cancer. The use of postoperative 5-FU–based chemoradiation therapy has been investigated since the 1960s. In one trial,10 patients were randomly assigned to undergo resection with or without postoperative chemoradiation (35 to 40 Gy with bolus 5-FU at 15 mg/kg for 3 days). Patients were not stratified for known prognostic factors, and 10 of 39 patients refused adjuvant therapy after randomization. An

21  The Stomach and Small Intestine

overall and relapse-free survival advantage was seen in the adjuvant therapy group (23% versus 4%, P = .05). The locoregional recurrence rate was also diminished (39% versus 54%). The overall survival rate was highest (30%) for the group who refused treatment after randomization to chemoradiation. However, when the results were analyzed by treatment actually received, survival was not statistically different between groups. The survival difference seen was therefore the result of an imbalance of prognostic features between the treatment groups.10 These results illustrate the problems that can arise from using small numbers of patients in clinical trials, the importance of stratification for known prognostic variables, and the importance of obtaining consent for treatment before randomization. The largest randomized trial conducted to evaluate the role of postoperative 5-FU–based chemoradiation therapy is the Intergroup 0116 trial (INT-0116), a well-designed and well-conducted trial that compared surgery alone with surgery followed sequentially by one cycle of 5-FU/leucovorin chemotherapy, concurrent chemoradiation therapy (45 Gy with bolus 5-FU/leucovorin during the first and last week of irradiation), and two cycles of 5-FU/leucovorin chemotherapy.177 At a median follow-up time of 5 years for the 556 patients with stage IB through IV (T1 N1, T2-4 any N; M0) gastric or gastroesophageal cancer randomized in this trial, the median overall survival time was 27 months for the surgery-only group and 36 months for those in the chemoradiation therapy group, and the 3-year overall survival rate was 41% for the surgery-alone group and 50% for the chemoradiation therapy group (P = .005). The median relapse-free survival time was 30 months for the chemoradiation therapy group and 19 months for the surgeryonly group, and the 3-year relapse-free survival rates were 48% for the chemoradiation therapy group and 31% for the surgery-only group. The results of this trial established postoperative adjuvant 5-FU–based chemoradiation as the standard of care for resected gastric cancer. Some issues that were raised by this trial are worthy of further discussion. First, a centralized quality ­assurance effort documented major or minor technical errors in as many as 35% of treatment plans; most of those errors were corrected before the start of radiation therapy. This observation underscores the need for strict adherence to treatment guidelines that have since been formalized in a monograph that lists important clinical and anatomic issues related to radiation therapy in commonly occurring clinical presentations.178 Second, 36% of patients in the chemoradiation therapy group were unable to complete protocol treatment as planned, mostly because of significant rates of gastrointestinal (33%) and hematologic (54%) toxic effects and disease progression (5%). Chemoradiation therapy is best delivered by a multidisciplinary team that has experience with managing these complications and with supportive care of the specific challenges faced by these patients, such as deficiencies of vitamin B12, iron, and calcium; ­ reactive hypoglycemia; and dumping syndrome. The inclusion criteria for this trial specifically required adequate caloric intake before patients could be enrolled on the trial. Third, chemoradiation seemed to reduce the incidence of local failure; the first site of failure in those who experienced treatment failure was local in 19% of patients in the chemoradiation therapy group and in 29% of those in the surgeryonly group. However, because failure ­information was not

501

collected prospectively and because definitions of regional failure included disease in lymph nodes, peritoneum, and liver, the absolute effect of chemoradiation on the patterns of failure is difficult to gauge from this study The trial had some unique patient and treatment characteristics. Most of the patients had distal gastric cancer (only 20% had cardia tumors). Almost two thirds of all patients had stage T3 or T4 tumors, and 85% had nodal metastases, meaning that these patients had more advanced disease than did those in the Dutch D1 versus D2 lymphadenectomy trial. A D2 lymphadenectomy was recommended, but patients were not excluded on the basis of the extent of lymphadenectomy. Only 10% of patients underwent a D2 dissection, 36% had a D1 dissection, and 54% had a D0 lymphadenectomy, perhaps because of a lack of strict quality assurance in pathology reviews, selection bias in favor of D1 dissection imposed by the caloric intake requirements for enrollment, and the relative lack of surgical expertise because of the rarity of gastric cancer in the United States compared with Japan, where D2 lymphadenectomies are performed routinely. This last difference is highlighted by a Japanese trial comparing D2 with more extended (paraaortic) lymphadenectomies; in this trial, eligible surgeons were required to have personally performed more than 100 gastric resections or to be at an institution that performed more than 80 gastric resections annually.179 In an editorial accompanying the publication of these findings, Mansfield180 observed that only two hospitals in the state of Texas performed more than 15 gastric resections annually, and neither would have qualified for participation in the Japanese study. It may therefore be reasonable to assume that the Intergroup trial was representative of the standard practice in most medical centers in the United States and that its results are more generalizable to U.S. patient populations. Another argument that is often raised in discussions of this trial is that the chemoradiation may not have achieved the survival benefits reported in this study if the patients had undergone a D2 lymphadenectomy. However, this is unlikely; D2 lymphadenectomy failed to improve overall survival in the Dutch randomized D1 versus D2 ­lymphadenectomy trial, and the observed trend toward a lower risk for relapse after D2 dissection was not significant statistically and may be attributable to stage migration.163 Moreover, a large, retrospective analysis from Korea found the median overall survival time to be significantly longer among patients who received 5-FU–based chemoradiation therapy after D2 dissection than among patients who underwent only D2 dissection (95 versus 63 months, P = .02).181 These results are, however, subject to selection bias from patients with better prognosis having been selected for chemoradiation treatment. Similarly, an analysis of the Surveillance, Epidemiology, and End Results database for patients treated between 1988 and 2003 demonstrated a survival benefit for adjuvant therapy compared with surgery alone.182 Taken together, however, the results of the pivotal Intergroup trial point to a clear role for postoperative chemoradiation therapy for resected gastric cancer and established postoperative chemoradiation therapy as a standard of care for these patients, at least in the United States. Two ongoing trials are evaluating more intensive therapies building on the foundation of postoperative

502

PART 6  Gastrointestinal Tract

c­ hemoradiation therapy defined by the Intergroup trial. The Cancer and Leukemia Group B 80101 trial is a randomized phase III trial for patients with resected gastric or gastroesophageal cancer that compares two sequential regimens, one involving one cycle of 5-FU/leucovorin, infusional 5FU/radiation therapy, and two cycles of 5-FU/leucovorin and the other consisting of one cycle of epirubicin/cisplatin/5-FU (ECF), infusional 5-FU/radiation therapy, and two cycles of ECF. As of October 2005, 91 patients had been accrued, and the frequency of grade 3 adverse events was comparable in the two treatment groups, although grade 4 adverse events were less common in the ECF group. The Radiation Therapy Oncology Group (RTOG) G-0114 trial is a randomized phase II trial for patients with resected gastric or gastroesophageal cancer comparing two treatment regimens: a 5-FU–containing regimen, in which two cycles of cisplatin/paclitaxel/infusional 5-FU were followed by paclitaxel/infusional 5-FU/radiation therapy, and a non-5FU–containing regimen, in which two cycles of cisplatin/ paclitaxel were followed by cisplatin/paclitaxel/radiation therapy. The part of the trial involving the 5-FU–containing regimen had to be closed early because of toxicity. Results of the analysis for the primary objective of improvement in 2-year disease-free survival rates after the non-5-FU–based treatment are awaited.

Postoperative Chemotherapy After initial trials failed to establish a role for radiation therapy in resected gastric cancer, adjuvant chemotherapy was evaluated for its role after curative resection. The results of two small trials suggested a survival benefit,170,183 but numerous other studies have not.176,184-187 Authors of a meta-analysis188 concluded that despite a probable publication bias in favor of positive trials, chemotherapy after curative surgery did not offer a survival benefit. Other meta-analyses have found some benefit from adjuvant chemotherapy, but this benefit has been modest and of borderline statistical significance.189-192 No basis for the routine recommendation of adjuvant chemotherapy is evident at this time.

Preoperative Chemotherapy Neoadjuvant chemotherapy has been investigated for patients with potentially resectable gastric cancer.193-198 Although the criteria for resectability have varied and sometimes have not been stated, this approach has been reported to increase the rate of curative resection. Analysis of a series of patients treated at The University of Texas M. D. Anderson Cancer Center on separate single-arm, prospective studies of intensive chemotherapy regimens revealed overall response rates of 24% to 38%, a resectability rate of 72%, a modest pathologic complete response rate of 4%, and high rates of toxic effects (40% of patients required hospitalization).195 The median survival time of 15 months is comparable to that for historical surgical controls, but such a comparison is biased in favor of the surgical series because such series report only patients who underwent resection. A multivariate analysis found that response to induction chemotherapy was the only independent prognostic factor. Limitations of the preoperative chemotherapy approach are the high rates of locoregional and peritoneal failure. To address these shortcomings, postoperative and preoperative radiation

therapy were added to a multi-­institutional trial recently completed.198

Preoperative Chemotherapy, Radiation Therapy, and Chemoradiation Therapy Preoperative radiation therapy, with or without chemotherapy, has been reported in single-arm, single-institution experiences to improve outcome.193-203 Prospective randomized trials conducted in Russia204-206 have shown improvements in resectability and survival without increased morbidity from the use of suboptimal radiation doses without concurrent chemotherapy. A randomized trial from China of preoperative radiation therapy (40 Gy given 2 to 4 weeks before surgery) demonstrated a survival benefit from the addition of preoperative radiation therapy (without chemotherapy) and improvements in resectability, disease downstaging (T and N), and locoregional control rates.207 Several phase II trials at M. D. Anderson Cancer Center and two multi-institutional trials have evaluated preoperative irradiation (45 Gy) with concurrent 5-FU–based chemotherapy.206,208-210 In an initial pilot experience, 24 patients received 45 Gy with concurrent continuous-infusion 5-FU, after which disease was restaged 4 to 6 weeks later followed by surgical resection and 10 Gy of intraoperative radiation therapy. Four patients (17%) had progressive disease and did not undergo resection. Two patients (11% of patients who had resection) had pathologic complete responses, and 12 (63%) had pathologic evidence of a significant treatment effect. The morbidity rate was 32%, and the mortality rate was 5%. The realization that neoadjuvant chemoradiation therapy could downstage disease or produce pathologic complete responses and margin-negative resections prompted further trials. In a subsequent trial, 41 patients received two cycles of induction chemotherapy with 5-FU, paclitaxel, and cisplatin followed by 45 Gy of radiation plus concurrent 5-FU and paclitaxel before disease restaging and resection.208 The rates of complete resection (R0; 78%), pathologic complete response (20%), and pathologic partial response (<10% residual cancer cells in the resected specimen; 15%) were encouraging. The pathologic response and T and N classifications after treatment were more predictive of overall and disease-free survival than were pretreatment variables. Toxicity was manageable. In a multi-institutional trial of a similar strategy that did not include paclitaxel, the complete resection rate was 70%, the pathologic complete response rate was 30%, and the pathologic partial response rate was 24% among 33 patients.209 Those who showed a pathologic response (complete or partial) had a median survival time of 64 months, compared with 12.6 months among those who did not show a pathologic response (P = .03). Two patients died of treatment-related effects. A retrospective analysis of 149 patients treated prospectively with preoperative chemoradiation at M. D. Anderson showed that 105 patients ultimately had R0 resections; the pathologic complete response rate was 23%, and the 3-year actuarial incidence of locoregional recurrence was 13%. Most of the recurrences occurred in the treated volume (Fig. 21-4).211 The RTOG conducted a phase II trial of two cycles of induction 5-FU, leucovorin, and cisplatin followed by concurrent chemoradiation therapy (45 Gy with infusional 5-FU and weekly paclitaxel).210 Among 43 assessable patients, 26% experienced a pathologic complete response, and 77% had a complete (R0) resection. Despite high rates

21  The Stomach and Small Intestine

FIGURE 21-4.  Location of 14 locoregional recurrences among 105 patients treated with neoadjuvant chemoradiation followed by R0 resection at M.  D. Anderson Cancer Center. Two marginal recurrences were attributed to inadequate coverage of regional nodal areas when oblique fields were used as part of a threedimensional radiation therapy plan. (From Reed VK, Krishnan S, Mansfield PF, et al. Incidence, natural history, and patterns of locoregional recurrence in gastric cancer patients treated with preoperative chemoradiotherapy. Int J Radiat Oncol Biol Phys 2008;71:741-747.)

of grade 4 toxicity (21%) and radiation-therapy protocol variations (56%), the ability to achieve a pathologic complete response rate higher than 20% in a study conducted across 40 institutions and the ability to achieve D2 dissections in 50% of patients suggest that this strategy of preoperative chemotherapy followed by chemoradiation therapy is worthy of further investigation. Current practice involves clinical disease staging with CT and endoscopic sonography and laparoscopic peritoneal staging, which includes visual inspection and cytologic evaluation of wash samples. A jejunostomy feeding tube is placed at the time of laparoscopy and has a critical role in supportive care during neoadjuvant therapy and in preparation for and recovery from surgery. Publication of the results of the Medical Research Council Adjuvant Gastric Infusional Chemotherapy (MAGIC) trial has further changed the landscape of treatment options for patients with locally advanced gastric cancer.212 This randomized phase III trial compared surgery alone with surgery plus three cycles of preoperative and postoperative ECF. Among the 503 patients accrued to this trial, the toxicity of ECF was manageable, and the rates of postoperative morbidity and mortality were similar in the two treatment groups. Tumors in the ECF group were also smaller at ­ resection (3 versus 5 cm, P < .001) and of lower T category (T1/T2, 51.7% ­versus

503

36.8%, P = .002) and N category (N0/N1, 84.4% versus 70.5%, P = .01). At a median follow-up time of 4 years, patients in the ECF group had significantly better 5-year overall survival rates (36% versus 23%, P = .009) and ­progressionfree survival rates (hazard ratio = 0.66; 95% confidence ­interval, 0.53 to 0.81; P < .001). For patients with stomach cancer who present before gastrectomy, these results clearly establish this regimen as a new treatment option. A few points worth mentioning with regard to the MAGIC trial include the fact that only 42% of patients completed the prescribed protocol, and 34% underwent surgery after preoperative chemotherapy but did not complete postoperative chemotherapy. No information on patterns of failure was reported, but it is worth noting that pathologic complete responses did not occur after preoperative chemotherapy alone. Endoscopic sonography was not routinely used in this study; rather, the information on reductions in tumor size was based on endoscopic measurement of the maximal tumor diameter, and the information on T and N downstaging was based on the assumption of a random distribution of T and N stage at diagnosis, because CT scans cannot provide accurate detail for disease staging. The United Kingdom National Cancer Research Institute Upper Gastrointestinal Clinical Studies Group ST03 trial (i.e., MAGIC-B) is addressing this concern by prospectively evaluating T and N classifications with endoscopic sonography. In a trend seen across all gastrointestinal malignancy trials and a logical extension of the MAGIC trial, MAGICB is randomly assigning patients with resectable gastric cancer to receive perioperative chemotherapy with epirubicin, cisplatin, and capecitabine with or without the antivascular endothelial growth factor antibody bevacizumab.

Chemoradiation Therapy for Unresectable Gastric Cancer Studies of unresectable or partially resected gastric cancer indicate that chemoradiation therapy can be curative in a small percentage of cases.16,17,213 Initial strategies for the nonoperative management of gastric cancer demonstrated an advantage for combining 5-FU with radiation. A randomized trial was conducted at Mayo Clinic that involved 35 to 40 Gy with or without bolus 5-FU for 3 days for a variety of unresected gastrointestinal malignancies.10 A small group of patients with gastric cancer were found to have superior 5-year survival rates after the combination treatment (12% versus 0% without chemotherapy), verifying the efficacy of a combination that has endured for 3 decades. Further evidence of the importance of 5-FU administration came from a randomized study conducted by the European Organisation for Research and Treatment of Cancer153 that showed a survival advantage for patients treated with chemoradiation compared with those who were treated with radiation alone to 55.5 Gy. The few patients who had residual disease and who ­remained free of progressive disease for at least 19 months after treatment had been given concurrent chemoradiation therapy and long-term (18 months) treatment with 5-FU ­thereafter.153 In a trial conducted by the Gastrointestinal Tumor Study Group with patients with incompletely resected gastric cancer, chemotherapy with 5-FU and methyl-lomustine was compared with a split course of radiation to a total dose of 50 Gy and bolus of 5-FU given at the beginning

504

PART 6  Gastrointestinal Tract

of each course of radiation therapy.213 The trial was discontinued early because of at least six treatment-related deaths and six disease-related deaths in the combined-­modality group within the first 6 months of therapy. Nevertheless, a statistically significant survival benefit was seen in the ­combined-therapy group at 5 years (20% versus 6%, P < .01). Modern supportive care, conventional treatment planning, and infusional chemotherapy schedules probably would have enhanced the difference in favor of the combinedmodality group. Seeking to confirm these results, the Study Group designed a second trial214 with similar entry criteria comparing 5-FU, methyl-lomustine, and doxorubicin with a schedule of the same chemotherapy followed on day 57 by chemoradiation (43.2 Gy with bolus 5-FU). ­Unfortunately, the trial was not well designed (radiation doses were low and delayed) and problems were encountered with patient accrual and compliance (11 of the 59 ­ patients in the radiation group did not receive the ­prescribed dose, and 7 of those 11 patients received no ­radiation at all). These difficulties may explain why no ­survival difference was found between treatment groups.

Supportive Care The importance of aggressive supportive care, as illustrated by the experience of the Gastrointestinal Tumor Study Group,213,214 cannot be overemphasized. Although the outcome for the chemoradiation therapy group was better than that for the chemotherapy group, 6 of the 45 patients in the chemoradiation therapy group died of sepsis or nutritional inadequacy. In administering radiation for gastric cancer, the routine use of prophylactic antiemetics and jejunostomytube feeding vastly improves treatment tolerance. Feeding tubes can obviate the need for intravenous fluid replacement or hyperalimentation and reduce the incidence of hospitalization for acute reactions. Nutritionists can monitor patients’ energy intake and weight, provide energy-intake goals, and serve as a resource for other ­nutrition-related issues.

Palliative Radiation Therapy radiation effectively relieves the symptoms associated with locally advanced gastric cancer. Several series have indicated that between 50% and 75% of patients experience improvement of bleeding, gastric outlet obstruction, and pain.7,215-217

Palliative Chemotherapy The chemotherapy drugs that are most active as single agents in gastric cancer are 5-FU, mitomycin, and doxorubicin.11,218-220 Randomized studies that have compared multiagent chemotherapy with supportive care have consistently shown an advantage in terms of median survival time for chemotherapy but not in long-term survival.221-223 The advantage of multiagent regimens over single-agent treatment has not been established.224 The median survival duration is relatively short (5 to 7 months) in most studies. A phase III trial reported from Europe225 evaluated highdose methotrexate, 5-FU, and doxorubicin versus etoposide, leucovorin, and 5-FU versus infusional 5-FU and cisplatin for advanced gastric cancer. The most remarkable finding in that trial was the lack of efficacy of any of the three treatments. The investigators concluded that none of these regimens should be considered standard for advanced or metastatic gastric cancer.225 Given the high morbidity, the cost, and the lack of an established benefit for multiagent regimens,

­ atients should be offered participation in clinical trials that p evaluate new agents or new combinations.

Radiation Techniques Current Techniques Typical anteroposterior and lateral simulation films are shown in Figure 21-5. Simulation films should be obtained when patients’ stomachs are empty (i.e., no oral intake for 3 hours), and patients should be treated in the morning ­before eating to minimize organ motion resulting from stomach distention. Oral contrast medium should be given 15 to 30 minutes before simulation, and arm-elevation and stabilization devices should be used. CT-based dosimetry is helpful in planning field arrangements. Patients can be treated with a four-field technique or through opposed anterior and posterior portals. The use of four or more radiation fields can significantly reduce the incidence of grade 4 or 5 toxicity compared with two fields.226 However, lateral fields are not always advisable for patients with an intact stomach because the gastric fundus often extends too far posteriorly. In such instances, the kidneys and spinal cord cannot be shielded without shielding the target, and opposed fields are preferred. To keep the spinal cord dose below 45 Gy, the anterior field can be more heavily weighted (usually 3:2) or an off-cord boost can be delivered with oblique fields after the initial 39.6 Gy. If preoperative imaging or fluoroscopy indicates that the tumor is located anteriorly, lateral fields can be used with beam weighting that limits their contribution to 20 Gy. Customized blocking is used to shield the critical structures in the field, including the small bowel, liver, spinal cord, and kidneys. CT-based treatment planning and dose-volume histograms can help objectively evaluate various technical treatment options for individual patients. An example of such a plan is illustrated in Figure 21-6. The target volume should include the entire stomach or gastric bed and the draining lymphatics in most cases. All available imaging (CT and barium swallowing) findings, endoscopic findings, and intraoperative findings should be reviewed and correlated before simulation. According to the findings from surgical studies discussed previously under “Surgical Therapy,” surgical margins larger than 4 to 6 cm are associated with low recurrence rates in the gastric remnant in patients undergoing partial gastrectomy. By analogy, radiation therapy fields could be limited to a partial gastric volume if the gross tumor volume is well defined by endoscopy. When radiation fields are designed for patients with an intact stomach, it is important to examine the CT scans for evidence of perigastric extension, which is not apparent when viewing the contrast-filled stomach at the time of simulation. Endoscopic and surgical reports should be correlated with imaging findings. Anterior and posterior fields should include the gastric, gastro-omental, celiac, hepatic, subpyloric, gastroduodenal, pancreaticosplenic, and retropancreaticoduodenal nodes. The splenic hilum, porta hepatis, and celiac trunk can be located on CT scans. The celiac axis usually arises at the bottom of the T12 vertebra, and the porta hepatis is 4 to 5 cm to the right of the T11 and L1 vertebral bodies. The remaining nodal tissue can be encompassed by treating the stomach and the medial portion of the C-loop of the duodenum or the stomach and duodenal bed. The lower border usually extends to the inferior aspect of L3 to encompass the entire C-loop.

21  The Stomach and Small Intestine

A

Beam’s eye view for “AP eq wght”

B

505

Beam’s eye view for “Rt eq wght”

C FIGURE 21-5.  Typical anteroposterior (A) and lateral (B) fields for the postoperative treatment of gastric cancer, with central axis dosimetry (C). The preoperative tumor volume is contoured along with the entire nodal region at risk (i.e., clinical target volume).

For proximal lesions involving the cardia or gastroesophageal junction (defined by endoscopy, upper gastrointestinal radiography, or CT), the paraesophageal nodes should be included, and a 5-cm margin should be included in the cephalad field margin. For distal lesions at or near the gastroduodenal junction, a 5-cm margin of duodenum should be included.

Ongoing Investigations Altered Fractionation The safe administration of conventionally fractionated radiation is limited by the tolerance of surrounding organs and that of the stomach itself. Pilot studies evaluating the

tolerance of hyperfractionated radiation have demonstrated the feasibility of this approach.227-229 Small fractions seem to be better tolerated.

New Radiosensitizing Agents Radiosensitizers such as irinotecan (CPT-11), oxaliplatin, and paclitaxel are being investigated in combination with radiation therapy at M. D. Anderson for advanced, unresectable gastric cancer. Preliminary experience indicates that the concurrent administration of these agents with radiation therapy is well tolerated when appropriate supportive care is given. Future investigations center on optimizing the dose, timing, and scheduling of these

PART 6  Gastrointestinal Tract

Volume

506

Dose Volume Histogram

1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0

1000

2000

3000

4000

5000

6000

Dose (cGy) Liver Left kidney Stomach Right kidney Spinal cord

FIGURE 21-6.  Example of a CT-based treatment plan with dose-volume histograms. CT planning is critical to objectively evaluate various technical treatment options for individual patients.

agents in combination with radiation therapy. Newer strategies should be evaluated in comparative trials to establish their superiority over 5-FU–based chemoradiation therapy.

Conformal Therapy The current technical limitations of conformal therapy in gastric cancer include the limited ability to identify gross tumor and problems associated with organ motion (i.e., respiratory motion and gastric distention). The RTOG is beginning to assess the feasibility of conformal therapy in its emerging gastric cancer protocols. Charged particle therapy with helium ions or protons also offers improved dose distribution but has limitations similar to those of conformal therapy. Several ways of addressing respiratory motion are being addressed, including active and passive respiratory gating. These techniques must be refined to fully realize the potential of conformal therapy for tumors in the upper ­abdomen.

Neutron Therapy Evaluations of the toxicity and efficacy of neutron therapy have indicated that this treatment holds little promise for gastric cancer. Severe late effects indicate a narrow therapeutic index,230 which is referable to the lack of an opportunity for sublethal damage repair between fractions. In the stomach, an organ with limited radiotolerance, the use

of neutrons eliminates the most important radiotherapeutic advantage.

Hepatic and Peritoneal-Directed Approaches Disease in the liver and peritoneum represents a significant problem in treating gastric cancer.108,109,217-231 Prospective evaluations of regimens that address peritoneal failure have been conducted by the University of Southern ­ California194 and Memorial Sloan-Kettering.232 Both involved preoperative chemotherapy followed by postoperative intraperitoneal chemotherapy in patients undergoing curative resection. Although the median survival times were not drastically different from those of studies that did not include intraperitoneal chemotherapy, the peritoneal failure rate of 16% in the Memorial Sloan-Kettering study232 was lower than expected. Intraperitoneal chemotherapy combined with hyperthermia also has had ­ documented activity.233-235 Although intraperitoneal ­ chemotherapy is conceptually provocative, controlled validation of these approaches is lacking. Whole-abdomen radiation therapy has been investigated for ovarian and endometrial cancer. However, the maximum tolerated dose of whole-abdomen irradiation is limited to 20 to 25 Gy, which makes this approach unlikely to be of benefit in gastric cancer. The role of prophylactic hepatic irradiation in pancreatic cancer has been evaluated by the RTOG,27 M. D. Anderson,236 and Johns

21  The Stomach and Small Intestine

­Hopkins University.237 All studies used at least 23.4 Gy to the whole liver. The first site of failure was the liver in 13% of patients in the first study.27 The M. D. Anderson study was closed early because of toxicity after only 11 patients had been treated,236 and no survival benefit was detected in the Johns Hopkins study.237 Although extended-field radiation could hypothetically reduce hepatic or peritoneal failure, it does not currently have a role in the treatment of gastric cancer.

Small Intestine Anatomy and Histology The small bowel extends from the gastric pylorus to the ileocecal valve and is divided into three parts: duodenum, jejunum, and ileum. Although its length varies, the average small bowel is 6 to 7 m long. The first section of small bowel, the duodenum, is about 12 fingerbreadths long and forms a C-shaped loop around the head of the pancreas. The ampulla of Vater is located in the second or descending portion of the duodenum. The jejunum and ileum compose the remainder of the small bowel and are attached to the posterior abdominal wall by a fan-shaped mesentery. The root of this mesentery runs obliquely from the left upper abdominal quadrant to the right lower abdominal quadrant. The processes of digestion are completed and the products absorbed in the small intestine. Plicae circulares (longitudinal folds that are often visible on radiographs), ­innumerable intestinal villi (0.5 to 1.5 mm in length), and microvilli greatly increase the luminal surface area for absorption. Intestinal crypts are found between the villi. Undifferentiated cells normally migrate and differentiate upward from the lower half of the crypts to the tips of the villi as they mature. Most will form absorptive cells, but others become mucus-secreting goblet cells or enteroendocrine cells. All three cell types live only 3 to 6 days.

Acute Effects of Radiation The acute effects of radiation on the small intestine reflect the sensitivity of the rapidly dividing undifferentiated intestinal crypt cells. These immature cells are the most sensitive to radiation and are preferentially lost after treatment with radiation, resulting in an eventual loss of mature replacement cells at the surface of the villi. The mature mucosal cells then become denuded, and a loss of absorptive function follows. This loss of function results in fluid and nutrient wasting, diarrhea, and sometimes dehydration.238 In a murine model, re-epithelialization typically occurs within 96 hours, starting with the recovery of the rapidly dividing crypt cells.239 The clinical symptoms of acute small bowel radiation injury may include diarrhea, cramps with abdominal pain, and nausea. The incidence and severity of small bowel injury are related to the dose, volume, and dose rate of ­radiation delivery.240 These acute effects can be managed with appropriate supportive care, such as dietary modification, antimotility agents, narcotics, and antiemetic agents. Diets low in fiber can decrease transit time and increase absorption in patients with acute radiation enteritis. The liberal use of antimotility agents such as diphenoxylate hydrochloride and atropine (Lomotil) or loperamide and the judicious use of narcotics can alleviate most cases of ­radiation-induced diarrhea.

507

Late Effects of Radiation Therapy Possible late manifestations of radiation therapy include chronic diarrhea, abdominal cramping, nausea, vomiting, malabsorption with weight loss, and chronic blood loss. Progressive fibrosis, shortening, constriction, and stenosis of the irradiated portion of bowel can occur in radiationinjured patients. Adhesions related to surgical and radiation injury can form between loops of bowel. Predisposing conditions to small bowel injury include fixation of the bowel at the time of treatment, pelvic inflammatory disease, diabetes, hypertension, and thin body habitus. Irradiation of the small bowel to a dose of 45 Gy given in 25 fractions would be expected to produce less than a 5% risk of late complications. In freely mobile sections of small intestine, a boost dose of 5.4 Gy in 3 fractions is routinely given safely. The dose to any segment of small bowel should not exceed 50.4 Gy, although boost doses are often delivered if the small bowel can be completely excluded from the ­treatment field.

Epidemiology and Etiology of Small Bowel Cancer Primary small bowel malignancies are rare, representing 1% to 3% of all gastrointestinal malignancies.241,242 Approximately 2100 to 2400 new cases are diagnosed each year in the United States.243 Most small bowel tumors occur in the duodenum or proximal jejunum.244 Several hypotheses have been proposed to explain the low incidence of small bowel malignancies. Transit time is shorter in the small bowel than in the large bowel, the large volume of secretions in the small bowel dilutes potential carcinogens, and the small bowel is colonized with fewer and comparatively more inert bacteria.245 The liquid contents of the small bowel are also less mechanically irritating than are feces in the large bowel. Many plasma cells and B cells secrete immunoglobulin A in the distal ileum. Immunodeficient patients are more prone to develop small bowel lymphoma than are immunocompetent patients. Adenocarcinomas occur predominantly in the proximal small bowel, correlating well with the duration of contact of bile salts and pancreatic enzymes with the mucosa.246 In animal studies, biliary diversion has been shown to decrease the incidence of experimentally induced small bowel malignancy.247 The risk of small bowel cancer has been reported to be increased with consumption of red meat and salt-cured food.248 The per capita consumption of ­dietary fat in different countries correlates well with small bowel cancer risk.244 Patients with a history of large bowel cancer also have an increased risk of developing small bowel cancer, indicating some overlap in predisposing factors.241,249 Patients with Crohn’s disease,250-253 celiac disease,254 Peutz-Jeghers syndrome, or von Recklinghausen’s disease (neurofibromatosis type 1)255 have been reported to have a higher incidence of small bowel malignancy.

Staging Because primary cancer of the small bowel is rare, a staging system has been published only recently by the American Joint Committee on Cancer (Box 21-2).256 The primary

508

PART 6  Gastrointestinal Tract

BOX 21-2.  American Joint Committee on Cancer Staging System for Carcinoma of the Small Intestine 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 through muscularis propria into the subserosa or into the nonperitonealized perimuscular tissue (mesentery or retroperitoneum) with extension 2 cm or less* T4 Tumor perforates the visceral peritoneum, or directly invades other organs or structures (includes other loops of small intestine, mesentery, or retroperitoneum more than 2 cm, and abdominal wall by way of serosa; for ­duodenum only, invasion of pancreas) Regional Lymph Nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Regional lymph node metastasis Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage Grouping Stage 0 Tis Stage I T1 T2 Stage II T3 T4 Stage III Any T Stage IV Any T

N0 N0 N0 N0 N0 N1 Any N

M0 M0 M0 M0 M0 M0 M1

*The

nonperitonealized perimuscular tissue is, for jejunum and ileum, part of the mesentery and, for duodenum in areas where serosa is l­acking, part of the retroperitoneum. From Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 107-112.

tumor is staged according to its depth of penetration and the involvement of adjacent structures or distant sites. Lateral spread within the duodenum, jejunum, or ileum is not considered in this classification.

The Role of Radiation Therapy Adenocarcinomas Radiation therapy can be used as an adjuvant to surgery or as primary palliative treatment of small bowel adenocarcinoma. Anatomic constraints and the intrinsic sensitivity of the normal small bowel to radiation limit the role of radiation as a locoregional treatment. Localized segments of the small bowel can tolerate doses of 45 to 50 Gy with conventional fractionation of 1.8 to 2.0 Gy per fraction; however, tolerance to radiation is significantly less when large volumes of bowel are in the radiation field. Because the small bowel and surrounding mesenteric lymphatics are mobile, regional treatment for malignancies of the jejunum and ileum has to include very large fields. The tolerance of the whole abdomen to radiation257 is less than the dose that would reliably sterilize microscopic disease. Because most of the small intestine is suspended by mesentery, surgical resection is usually adequate ­local therapy for adenocarcinoma of the small bowel. Given these limitations and the rarity of these tumors, the role of adjuvant radiation has not been clearly defined for ­adenocarcinoma of the small bowel. For lesions that

i­nvolve contiguous structures, ­preoperative or postoperative radiation can theoretically improve resectability and local control over results that can be achieved by surgical ­resection alone. Forty-five percent of small bowel adenocarcinomas arise in the duodenum,258-262 and for these tumors, the treatment is surgical resection. For lesions arising in the first and second portions of the duodenum, pancreaticoduodenectomy is usually required, but for more distal lesions, a segmental duodenectomy is often possible. The 5-year survival rates after surgery alone range from 20% to 60%.258,261-267 Johns Hopkins and Memorial Sloan­Kettering have reported 5-year survival rates of 60%268 and 53%,267 respectively, in patients who have undergone ­resection. In duodenal cancer, regional lymphatics are often ­involved, and the competing risk of distant failure is modest. The spread pattern provides a rationale for preoperative or postoperative chemoradiation therapy. Although some investigations of small numbers of patients have not found adjuvant therapy to be beneficial,267 no study conducted has included enough patients to adequately detect even large differences in survival according to type of therapy. The role of adjuvant therapy has therefore not been defined. In pancreatic cancer, which also has a significant locoregional pattern of spread, a survival benefit has been demonstrated with postoperative adjuvant therapy.269

21  The Stomach and Small Intestine

The primary tumor site and regional lymphatic drainage can be safely treated with 45 to 50 Gy, a dose that reliably eliminates microscopic residual disease. Radiation portals should cover the entire duodenum and the hepatoduodenal, pancreatic, superior mesenteric, periportal, subpyloric, pyloric, and para-aortic nodal chains. Mucosal margins of at least 5 cm proximally and distally are recommended because of the potential for submucosal spread. Localized boost doses of 10 Gy at most to areas thought to be at highest risk can be given with external beam or intraoperative radiation therapy. The risk of duodenal ulceration or perforation increases sharply at total doses higher than 55 Gy. Extrapolation from treating pancreatic cancer suggests that preoperative therapy is likely to be better ­tolerated than postoperative therapy.270 Palliative radiation can be beneficial in cases of unresectable or recurrent tumors. When the goal is palliation, tumor regression and symptomatic improvement can be achieved with 30 Gy, given in 10 fractions. Goals of palliative radiation include controlling blood loss, decreasing pain, and relieving obstructive symptoms.

Sarcomas Similar principles can be applied for evaluating the role of radiation in treating sarcomas of the small bowel. Experience with sarcoma of the extremities suggests that preoperative or postoperative radiation therapy may reduce the risk for local recurrence if wide surgical margins cannot be obtained. When locally advanced disease is recognized at staging, preoperative therapy to a dose of 45 to 50 Gy is preferred.

Carcinoid Tumors The specific indications for radiation therapy for small bowel carcinoid tumors are lacking, given the rarity of this clinical presentation. However, radiation has been shown to be beneficial in palliation of unresected disease, with perhaps less benefit for patients with carcinoid ­syndrome.271

Choice of Radiation Techniques The choice of radiation techniques for tumors of the small bowel depends on the tumor location and individual anatomic variations. For duodenal primary tumors, a four-field technique is usually preferred. The dose contribution from the lateral fields should be limited to 20 Gy, and appropriate shielding should be used wherever possible to limit the dose to critical adjacent structures such as the liver, kidneys, stomach, spinal cord, and uninvolved small bowel.

Summary of Radiation Therapy for Small Bowel Tumors The role of radiation therapy in the treatment of small bowel tumors is largely undefined because of the rarity of these tumors, anatomic constraints, and the wide range of clinical presentations. Problems specific to irradiation of the small bowel include the mobility of the primary tumor and the limited radiation tolerance of uninvolved tissues. Although a role for adjuvant therapy has yet to be defined, radiation therapy remains an important modality for selected cases in the adjuvant setting and for palliation of small bowel tumors of all histologic types.

509

REFERENCES    1. Fajardo LF, Berthrong M, Anderson RE. Stomach. In ­ Radiation Pathology. New York: Oxford University Press, 2001, pp 221-227.    2. Goldgraber MB, Rubin CE, Palmer WL, et al. The early gastric response to irradiation; a serial biopsy study. Gastroenterology 1954;27:1-20.    3. Palmer WL, Templeton F. The effect of radiation therapy on gastric secretion. JAMA 1939;112:1429-1434.    4. Palmer W. Gastric Irradiation in Peptic Ulcer. Chicago: University of Chicago Press, 1974.    5. Hueper WC, DeCarvajal-Forero J. The effect of repeated irradiation of the gastric region with small doses of roentgen rays upon the stomach and blood of dogs. AJR Am J Roentgenol 1944;40:1938.    6. Hamilton F. Gastric ulcer following radiation. Arch Surg 1947;55:394-399.    7. Klaassen DJ, MacIntyre JM, Catton GE, et al. Treatment of locally unresectable cancer of the stomach and pancreas: a randomized comparison of 5-fluorouracil alone with radiation plus concurrent and maintenance 5-fluorouracil—an Eastern Cooperative Oncology Group study. J Clin Oncol 1985;3:373-378.    8. Caudry M, Escarmant P, Maire JP, et al. Radiotherapy of gastric cancer with a three field combination: feasibility, tolerance, and survival. Int J Radiat Oncol Biol Phys 1987;13:1821-1827.    9. Haas CD, Mansfield CM, Leichman LP, et al. Combined nonsimultaneous radiation therapy and chemotherapy with 5-FU, doxorubicin, and mitomycin for residual localized gastric adenocarcinoma: a Southwest Oncology Group pilot study. Cancer Treat Rep 1983;67:421-424.   10. Moertel CG, Childs DS Jr, Reitemeier RJ, et al. Combined 5fluorouracil and supervoltage radiation therapy of locally unresectable gastrointestinal cancer. Lancet 1969;2:865-867.   11. Schein PS, Smith FP, Woolley PV, et al. Current management of advanced and locally unresectable gastric carcinoma. Cancer 1982;50:2590-2596.   12. Brookland RK, Rubin S, Danoff BF. Extended field irradiation in the treatment of patients with cervical carcinoma involving biopsy proven para-aortic nodes. Int J Radiat Oncol Biol Phys 1984;10:1875-1879.   13. Ballon SC, Berman ML, Lagasse LD, et al. Survival after ­extraperitoneal pelvic and para-aortic lymphadenectomy and ­radiation therapy in cervical carcinoma. Obstet Gynecol 1981;57: 90-95.   14. Hughes RR, Brewington KC, Hanjani P, et al. Extended field irradiation for cervical cancer based on surgical staging. Gynecol Oncol 1980;9:153-161.   15. Fletcher GH, Rutledge FN. Extended field technique in the management of the cancers of the uterine cervix. Am J Roentgenol Radium Ther Nucl Med 1972;114:116-122.   16. Asakawa H, Otawa H, Yamada S, et al. Combination therapy of gastric carcinoma with radiation and chemotherapy. Tohoku J Exp Med 1982;137:445-452.   17. Asakawa H, Takeda T. High energy x-ray therapy of gastric carcinoma. Nippon Gan Chiryo Gakkai Shi 1973;8:362-371.   18. Emami B, Watring WG, Tak W, et al. Para-aortic lymph node radiation in advanced cervical cancer. Int J Radiat Oncol Biol Phys 1980;6:1237-1241.   19. Macdonald JS, Schnall SF. Adjuvant treatment of gastric cancer. World J Surg 1995;19:221-225.   20. Roswit B, Malsky SJ, Reid CB. Severe radiation injuries of the stomach, small intestine, colon and rectum. Am J Roentgenol Radium Ther Nucl Med 1972;114:460-475.   21. Rubin SC, Brookland R, Mikuta JJ, et al. Para-aortic nodal metastases in early cervical carcinoma: long-term survival following extended-field radiotherapy. Gynecol Oncol 1984;18:213-217.   22. Welander CE, Pierce VK, Nori D, et al. Pretreatment laparotomy in carcinoma of the cervix. Gynecol Oncol 1981;12:336-347.   23. Roswitt B, Malsky S, Reid C. Radiation tolerance of the gastrointestinal tract. Front Radiat Ther Oncol 1971;5:160-180.   24. Bowers R, Brick I. Surgery in radiation injury of the stomach. Surgery 1947;22:20-40.

510

PART 6  Gastrointestinal Tract

  25. Willett CG, Tepper JE, Orlow EL, et al. Renal complications secondary to radiation treatment of upper abdominal malignancies. Int J Radiat Oncol Biol Phys 1986;12:1601-1604.   26. Lawrence TS, Robertson JM, Anscher MS, et al. Hepatic toxicity resulting from cancer treatment. Int J Radiat Oncol Biol Phys 1995;31:1237-1248.   27. Komaki R, Wadler S, Peters T, et al. High-dose local irradiation plus prophylactic hepatic irradiation and chemotherapy for inoperable adenocarcinoma of the pancreas. A preliminary report of a multi-institutional trial (Radiation Therapy Oncology Group Protocol 8801). Cancer 1992;69:2807-2812.   28. Jirtle RL, Anscher MS, Alati T. Radiation Sensitivity of the liver. Adv Radiat Biol 1990;14:269-311.   29. Lawrence TS, Ten Haken RK, Kessler ML, et al. The use of 3-D dose volume analysis to predict radiation hepatitis. Int J Radiat Oncol Biol Phys 1992;23:781-788.   30. McGinn CJ, Ten Haken RK, Ensminger WD, et al. Treatment of intrahepatic cancers with radiation doses based on a normal tissue complication probability model. J Clin Oncol 1998;16:2246-2252.   31. Dawson LA, McGinn CJ, Normolle D, et al. Escalated focal liver radiation and concurrent hepatic artery fluorodeoxyuridine for unresectable intrahepatic malignancies. J Clin Oncol 2000;18:2210-2218.   32. Haenszel W. Variation in incidence of and mortality from stomach cancer, with particular reference to the United States. J Natl Cancer Inst 1958;21:213-262.   33. Devesa SS, Blot WJ, Fraumeni JF Jr. Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer 1998;83:2049-2053.   34. American Cancer Society. Cancer Facts & Figures 2008. Atlanta: American Cancer Society, 2008.   35. Wiggins CL, Becker TM, Key CR, et al. Stomach cancer among New Mexico’s American Indians, Hispanic whites, and non­Hispanic whites. Cancer Res 1989;49:1595-1599.   36. Pisani P, Parkin DM, Bray F, et al. Estimates of the worldwide mortality from 25 cancers in 1990. Int J Cancer 1999;83:18-29.   37. Parkin DM, Bray F, Ferlay J, et al. Global cancer statistics. CA Cancer J Clin 2002;2005;55:74-108.   38. La Vecchia C, Negri E, D’Avanzo B, et al. Electric refrigerator use and gastric cancer risk. Br J Cancer 1990;62:136-137.   39. Coggon D, Barker DJ, Cole RB, et al. Stomach cancer and food storage. J Natl Cancer Inst 1989;81:1178-1182.   40. Howson CP, Hiyama T, Wynder EL. The decline in gastric cancer: epidemiology of an unplanned triumph. Epidemiol Rev 1986;8:1-27.   41. Blot WJ, Devesa SS, Kneller RW, et al. Rising incidence of adenocarcinoma of the esophagus and gastric cardia. JAMA 1991;265:1287-1289.   42. Lauren P. The two histological main types of gastric carcinoma: diffuse and so-called intestinal-type carcinoma. An attempt at a histo-clinical classification. Acta Pathol Microbiol Scand 1965;64:31-49.   43. Lauren PA, Nevalainen TJ. Epidemiology of intestinal and diffuse types of gastric carcinoma. A time-trend study in Finland with comparison between studies from high- and low-risk areas. Cancer 1993;71:2926-2933.   44. Munoz N, Asvall J. Time trends of intestinal and diffuse types of gastric cancer in Norway. Int J Cancer 1971;8:144-157.   45. Borrmann R. Geschwlste des Mangens und Duodenums, vol. 4. Berlin: Julius Springer, 1926.   46. Morson BC. Carcinoma arising from areas of intestinal metaplasia in the gastric mucosa. Br J Cancer 1955;9:377-385.   47. Correa P, Cuello C, Duque E, et al. Gastric cancer in Colombia. III. Natural history of precursor lesions. J Natl Cancer Inst 1976;57:1027-1035.   48. Coggon D, Osmond C, Barker DJ. Stomach cancer and migration within England and Wales. Br J Cancer 1990;61:573-574.   49. Buiatti E, Palli D, Decarli A, et al. A case-control study of gastric cancer and diet in Italy. II. Association with nutrients. Int J Cancer 1990;45:896-901.   50. Chyou PH, Nomura AM, Hankin JH, et al. A case-cohort study of diet and stomach cancer. Cancer Res 1990;50:7501-7504.

  51. Boeing H, Frentzel-Beyme R, Berger M, et al. Case-control study on stomach cancer in Germany. Int J Cancer 1991;47:858-864.   52. Correa P, Fontham E, Pickle LW, et al. Dietary determinants of gastric cancer in south Louisiana inhabitants. J Natl Cancer Inst 1985;75:645-654.   53. Demirer T, Icli F, Uzunalimoglu O, et al. Diet and stomach cancer incidence. A case-control study in Turkey. Cancer 1990;65:23442348.   54. Gonzalez CA, Sanz JM, Marcos G, et al. Dietary factors and stomach cancer in Spain: a multi-centre case-control study. Int J Cancer 1991;49:513-519.   55. Nomura A, Grove JS, Stemmermann GN, et al. Cigarette smoking and stomach cancer. Cancer Res 1990;50:7084.   56. Ramon JM, Serra L, Cerdo C, et al. Dietary factors and gastric cancer risk. A case-control study in Spain. Cancer 1993;71: 1731-1735.   57. Risch HA, Jain M, Choi NW, et al. Dietary factors and the incidence of cancer of the stomach. Am J Epidemiol 1985;122:947-959.   58. Trichopoulos D, Ouranos G, Day NE, et al. Diet and cancer of the stomach: a case-control study in Greece. Int J Cancer 1985;36:291-297.   59. Wu-Williams AH, Yu MC, Mack TM. Life-style, workplace, and stomach cancer by subsite in young men of Los Angeles County. Cancer Res 1990;50:2569-2576.   60. You WC, Chang YS, Yang ZT, et al. Etiological research on gastric cancer and its precursor lesions in Shandong, China. IARC Sci Publ 1991;105:33-38.   61. Gonzalez CA, Riboli E, Badosa J, et al. Nutritional factors and gastric cancer in Spain. Am J Epidemiol 1994;139:466-473.   62. Galanis DJ, Kolonel LN, Lee J, et al. Intakes of selected foods and beverages and the incidence of gastric cancer among the Japanese residents of Hawaii: a prospective study. Int J Epidemiol 1998;27:173-180.   63. Webb PM, Bates CJ, Palli D, et al. Gastric cancer, gastritis and plasma vitamin C: results from an international correlation and cross-sectional study. The Eurogast Study Group. Int J Cancer 1997;73:684-689.   64. La Vecchia C, Ferraroni M, D’Avanzo B, et al. Selected micronutrient intake and the risk of gastric cancer. Cancer Epidemiol Biomarkers Prev 1994;3:393-398.   65. Zhang L, Blot WJ, You WC, et al. Serum micronutrients in relation to pre-cancerous gastric lesions. Int J Cancer 1994;56: 650-654.   66. La Vecchia C, Munoz SE, Braga C, et al. Diet diversity and gastric cancer. Int J Cancer 1997;72:255-257.   67. Ward MH, Sinha R, Heineman EF, et al. Risk of adenocarcinoma of the stomach and esophagus with meat cooking method and doneness preference. Int J Cancer 1997;71:14-19.   68. Barrett JH, Parslow RC, McKinney PA, et al. Nitrate in drinking water and the incidence of gastric, esophageal, and brain cancer in Yorkshire, England. Cancer Causes Control 1998;9:153-159.   69. La Vecchia C, Negri E, Franceschi S, et al. Case-control study on influence of methionine, nitrite, and salt on gastric carcinogenesis in northern Italy. Nutr Cancer 1997;27:65-68.   70. Coggon D, Barker DJ, Cole RB. Stomach cancer and work in dusty industries. Br J Ind Med 1990;47:298-301.   71. Nomura A, Grove JS, Stemmermann GN, et al. A prospective study of stomach cancer and its relation to diet, cigarettes, and alcohol consumption. Cancer Res 1990;50:627-631.   72. Kneller RW, McLaughlin JK, Bjelke E, et al. A cohort study of stomach cancer in a high-risk American population. Cancer 1991;68:672-678.   73. Jedrychowski W, Boeing H, Wahrendorf J, et al. Vodka consumption, tobacco smoking and risk of gastric cancer in Poland. Int J Epidemiol 1993;22:606-613.   74. Staszewski J. Migrant studies in alimentary tract cancer. Recent Results Cancer Res 1972;39:85-97.   75. Haenszel W, Kurihara M, Locke FB, et al. Stomach cancer in Japan. J Natl Cancer Inst 1976;56:265-274.   76. Nomura A, Chyou PH, Stemmermann GN. Association of serum ferritin levels with the risk of stomach cancer. Cancer Epidemiol Biomarkers Prev 1992;1:547-550.

21  The Stomach and Small Intestine   77. Blackburn EK, Callender ST, Dacie JV, et al. Possible association between pernicious anaemia and leukaemia: a prospective study of 1,625 patients with a note on the very high incidence of stomach cancer. Int J Cancer 1968;3:163-170.   78. Brinton LA, Gridley G, Hrubec Z, et al. Cancer risk following pernicious anaemia. Br J Cancer 1989;59:810-813.   79. Hsing AW, Hansson LE, McLaughlin JK, et al. Pernicious anemia and subsequent cancer. A population-based cohort study. Cancer 1993;71:745-750.   80. La Vecchia C, Braga C, Negri E, et al. Risk of stomach cancer in patients with gastric or duodenal ulcer. Eur J Cancer Prev 1997;6:20-23.   81. Stalnikowicz R, Benbassat J. Risk of gastric cancer after ­gastric surgery for benign disorders. Arch Intern Med 1990;150: 2022-2026.   82. Tersmette AC, Offerhaus GJ, Tersmette KW, et al. Meta-analysis of the risk of gastric stump cancer: detection of high risk patient subsets for stomach cancer after remote partial gastrectomy for benign conditions. Cancer Res 1990;50:6486-6489.   83. Dubrow R. Gastric cancer following peptic ulcer surgery. J Natl Cancer Inst 1993;85:1268-1270.   84. Nakamura T, Nakano G. Histopathological classification and malignant change in gastric polyps. J Clin Pathol 1985;38: 754-764.   85. Spechler SJ. The role of gastric carditis in metaplasia and neoplasia at the gastroesophageal junction. Gastroenterology 1999;117:218-228.   86. Spechler SJ. Barrett’s esophagus. Semin Oncol 1994;21: 431-437.   87. La Vecchia C, Negri E, Franceschi S, et al. Family history and the risk of stomach and colorectal cancer. Cancer 1992;70:50-55.   88. Barker DJ, Coggon D, Osmond C, et al. Poor housing in childhood and high rates of stomach cancer in England and Wales. Br J Cancer 1990;61:575-578.   89. Munoz SE, Ferraroni M, La Vecchia C, et al. Gastric cancer risk factors in subjects with family history. Cancer Epidemiol Biomarkers Prev 1997;6:137-140.   90. Offerhaus GJ, Giardiello FM, Krush AJ, et al. The risk of upper gastrointestinal cancer in familial adenomatous polyposis. Gastroenterology 1992;102:1980-1982.   91. Zhang L, Blot WJ, You WC, et al. Helicobacter pylori antibodies in relation to precancerous gastric lesions in a high-risk ­Chinese population. Cancer Epidemiol Biomarkers Prev 1996;5: 627-630.   92. Nomura A, Stemmermann GN, Chyou PH, et al. Helicobacter pylori infection and gastric carcinoma among Japanese Americans in Hawaii. N Engl J Med 1991;325:1132-1136.   93. Parsonnet J, Harris RA, Hack HM, et al. Should we treat H. pylori infection to prevent gastric cancer? Gastroenterology 1997;112:1044-1045.   94. Forman D, Newell DG, Fullerton F, et al. Association between infection with Helicobacter pylori and risk of gastric cancer: evidence from a prospective investigation. BMJ 1991;302: 1302-1305.   95. Webb PM, Crabtree JE, Forman D. Gastric cancer, cytotoxin­associated gene A-positive Helicobacter pylori, and serum pepsinogens: an international study. The Eurogast Study Group. Gastroenterology 1999;116:269-276.   96. Blaser MJ, Perez-Perez GI, Kleanthous H, et al. Infection with Helicobacter pylori strains possessing cagA is associated with an increased risk of developing adenocarcinoma of the stomach. Cancer Res 1995;55:2111-2115.   97. Blaser MJ, Chyou PH, Nomura A. Age at establishment of Helicobacter pylori infection and gastric carcinoma, gastric ulcer, and duodenal ulcer risk. Cancer Res 1995;55:562-565.   98. Chow WH, Blaser MJ, Blot WJ, et al. An inverse relation between cagA+ strains of Helicobacter pylori infection and risk of esophageal and gastric cardia adenocarcinoma. Cancer Res 1998;58:588-590.   99. Harris RA, Owens DK, Witherell H, et al. Helicobacter pylori and gastric cancer: what are the benefits of screening only for the CagA phenotype of H. pylori? Helicobacter 1999;4:69-76.

511

100. Levine PH, Stemmermann G, Lennette ET, et al. Elevated antibody titers to Epstein-Barr virus prior to the diagnosis of ­Epstein-Barr-virus–associated gastric adenocarcinoma. Int J ­Cancer 1995;60:642-644. 101. Forman D. Helicobacter pylori infection and cancer. Br Med Bull 1998;54:71-78. 102. Forman D. Helicobacter pylori: the gastric cancer problem. Gut 1998;43;Suppl 1:S33-34. 103. Maruyama K, Okabayashi K, Kinoshita T. Progress in gastric cancer surgery in Japan and its limits of radicality. World J Surg 1987;11:418-425. 104. Prolla JC, Kobayashi S, Kirsner JB. Gastric cancer. Some recent improvements in diagnosis based upon the Japanese experience. Arch Intern Med 1969;124:238-246. 105. Nakajima T, Nishi M, Kajitani T. Improvement in treatment results of gastric cancer with surgery and chemotherapy: experience of 9,700 cases in the Cancer Institute Hospital, Tokyo. Semin Surg Oncol 1991;7:365-372. 106. Pisani P, Oliver WE, Parkin DM, et al. Case-control study of gastric cancer screening in Venezuela. Br J Cancer 1994;69: 1102-1105. 107. McNeer G, Vandenberg H Jr, Donn FY, et al. A critical evaluation of subtotal gastrectomy for the cure of cancer of the stomach. Ann Surg 1951;134:2-7. 108. Landry J, Tepper JE, Wood WC, et al. Patterns of failure following curative resection of gastric carcinoma. Int J Radiat Oncol Biol Phys 1990;19:1357-1362. 109. Gunderson LL, Sosin H. Adenocarcinoma of the stomach: areas of failure in a re-operation series (second or symptomatic look) clinicopathologic correlation and implications for adjuvant therapy. Int J Radiat Oncol Biol Phys 1982;8:1-11. 110. Papachristou DN, Fortner JG. Local recurrence of gastric adenocarcinomas after gastrectomy. J Surg Oncol 1981;18:47-53. 111. Horn RC Jr. Carcinoma of the stomach; autopsy findings in ­untreated cases. Gastroenterology 1955;29:515-523; discussion, 523-515. 112. Stout A. Pathology of carcinoma of the stomach. Arch Surg 1943;46:807-822. 113. Thomson F, Robbins R. Local recurrence following subtotal resection for gastric carcinoma. Surg Gynecol Obstet 1952;95: 341-344. 114. Wisbeck WM, Becher EM, Russell AH. Adenocarcinoma of the stomach: autopsy observations with therapeutic implications for the radiation oncologist. Radiother Oncol 1986;7:13-18. 115. McNeer G, Vandenberg H, Donn F. A critical evaluation of subtotal gastrectomy for the cure of cancer of the stomach. Arch Surg 1957;134:2-7. 116. Wangensteen OH, Lewis FJ, Arhelger SW, et al. An interim report upon the second look procedure for cancer of the stomach, colon, and rectum and for limited intraperitoneal carcinosis. Surg Gynecol Obstet 1954;99:257-267. 117. Wanebo HJ, Kennedy BJ, Chmiel J, et al. Cancer of the stomach. A patient care study by the American College of Surgeons. Ann Surg 1993;218:583-592. 118. Kurihara M, Shirakabe H, Yarita T, et al. Diagnosis of small early gastric cancer by x-ray, endoscopy, and biopsy. Cancer Detect Prev 1981;4:377-383. 119. Kurtz RC, Sherlock P. The diagnosis of gastric cancer. Semin Oncol 1985;12:11-18. 120. Graham DY, Schwartz JT, Cain GD, et al. Prospective evaluation of biopsy number in the diagnosis of esophageal and gastric carcinoma. Gastroenterology 1982;82:228-231. 121. Cook AO, Levine BA, Sirinek KR, et al. Evaluation of gastric adenocarcinoma. Abdominal computed tomography does not replace celiotomy. Arch Surg 1986;121:603-606. 122. Sussman SK, Halvorsen RA Jr., Illescas FF, et al. Gastric ­adenocarcinoma: CT versus surgical staging. Radiology 1988;167: 335-340. 123. Ajani JA, Mansfield PF, Ota DM. Potentially resectable gastric carcinoma: current approaches to staging and preoperative therapy. World J Surg 1995;19:216-220.

512

PART 6  Gastrointestinal Tract

124. Smith JW, Brennan MF, Botet JF, et al. Preoperative endoscopic ultrasound can predict the risk of recurrence after operation for gastric carcinoma. J Clin Oncol 1993;11:2380-2385. 125. Botet JF, Lightdale CJ, Zauber AG, et al. Preoperative staging of gastric cancer: comparison of endoscopic US and dynamic CT. Radiology 1991;181:426-432. 126. Tio LT, Blank LE, Wijers OB, et al. Staging and prognosis using endosonography in patients with inoperable esophageal carcinoma treated with combined intraluminal and external irradiation. Gastrointest Endosc 1994;40:304-310. 127. Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 99-106. 128. Shiu MH, Perrotti M, Brennan MF. Adenocarcinoma of the stomach: a multivariate analysis of clinical, pathologic and treatment factors. Hepatogastroenterology 1989;36:7-12. 129. Baba H, Korenaga D, Okamura T, et al. Prognostic factors in gastric cancer with serosal invasion. Univariate and multivariate analyses. Arch Surg 1989;124:1061-1064. 130. Serlin O, Keehn RJ, Higgins GA Jr, et al. Factors related to survival following resection for gastric carcinoma: analysis of 903 cases. Cancer 1977;40:1318-1329. 131. Hartley LC, Evans E, Windsor CJ. Factors influencing prognosis in gastric cancer. Aust N Z J Surg 1987;57:5-9. 132. Kennedy BJ. TNM classification for stomach cancer. Cancer 1970;26:971-983. 133. Bedikian AY, Chen TT, Khankhanian N, et al. The natural history of gastric cancer and prognostic factors influencing survival. J Clin Oncol 1984;2:305-310. 134. Dockerty MB. Pathologic aspects of primary malignant ­neoplasms of the stomach. In ReMine W, Priestly J, Berkson J, (eds): Cancer of the Stomach. Philadelphia: WB Saunders, 1964, p 173. 135. Baba H, Korenaga D, Haraguchi M, et al. Width of serosal invasion and prognosis in advanced human gastric cancer with ­special reference to the mode of tumor invasion. Cancer 1989;64: 2482-2486. 136. Nagatomo T, Murakami E, Kondo K. Histologic criteria of ­serosal rupture and prognosis in gastric carcinoma. Cancer 1972;29: 180-190. 137. Hayes N, Wayman J, Wadehra V, et al. Peritoneal cytology in the surgical evaluation of gastric carcinoma. Br J Cancer 1999;79:520-524. 138. Benevolo M, Mottolese M, Cosimelli M, et al. Diagnostic and prognostic value of peritoneal immunocytology in gastric cancer. J Clin Oncol 1998;16:3406-3411. 139. Nakajima T, Harashima S, Hirata M, et al. Prognostic and therapeutic values of peritoneal cytology in gastric cancer. Acta Cytol 1978;22:225-229. 140. Kitamura M, Arai K, Iwasaki Y. [Evaluation of intraoperative peritoneal cytology and intraperitoneal chemotherapy using CDDP combined with MMC for gastric carcinomatous peritonitis]. Gan To Kagaku Ryoho 1995;22:1523-1526. 141. Shirakawa T, Onda M, Tokunaga A, et al. [Role of peritoneal lavage smears and gastric wall brushing smears in gastric cancer surgery]. Gan To Kagaku Ryoho 1996;23:1472-1475. 142. Teramoto T, Onda M, Tokunaga A, et al. [Prognostic evaluation of peritoneal lavage smears and gastric wall brushing smears in gastric cancer surgery]. Gan To Kagaku Ryoho 1995;22: 1606-1609. 143. Koga S, Kaibara N, Iitsuka Y, et al. Prognostic significance of intraperitoneal free cancer cells in gastric cancer patients. J Cancer Res Clin Oncol 1984;108:236-238. 144. Maccomb WS, Fletcher GH. Planned combination of surgery and radiation in treatment of advanced primary head and neck cancers. Am J Roentgenol Radium Ther Nucl Med 1957;77: 397-414. 145. Maruta K, Shida H. Some factors which influence prognosis after surgery for advanced gastric cancer. Ann Surg 1968;167: 313-318. 146. Harrison JD, Morris DL, Ellis IO, et al. The effect of tamoxifen and estrogen receptor status on survival in gastric carcinoma. Cancer 1989;64:1007-1010.

147. Gouzi JL, Huguier M, Fagniez PL, et al. [Total gastrectomy versus partial gastrectomy of adenocarcinoma of the antrum. A French prospective controlled study]. Ann Chir 1989;43: 356-360. 148. Sugimachi K, Kodama Y, Kumashiro R, et al. Critical evaluation of prophylactic splenectomy in total gastrectomy for the stomach cancer. Gann 1980;71:704-709. 149. Brady MS, Rogatko A, Dent LL, et al. Effect of splenectomy on morbidity and survival following curative gastrectomy for carcinoma. Arch Surg 1991;126:359-364. 150. Otsuji E, Yamaguchi T, Sawai K, et al. Total gastrectomy with simultaneous pancreaticosplenectomy or splenectomy in ­patients with advanced gastric carcinoma. Br J Cancer 1999;79: 1789-1793. 151. Otsuji E, Yamaguchi T, Sawai K, et al. End results of simultaneous splenectomy in patients undergoing total gastrectomy for gastric carcinoma. Surgery 1996;120:40-44. 152. Papachristou DN, Agnanti N, D’Agostino H, et al. Histologically positive esophageal margin in the surgical treatment of gastric cancer. Am J Surg 1980;139:711-713. 153. Bleiberg H, Goffin JC, Dalesio O, et al. Adjuvant radiotherapy and chemotherapy in resectable gastric cancer. A randomized trial of the gastro-intestinal tract cancer cooperative group of the EORTC. Eur J Surg Oncol 1989;15:535-543. 154. Slot A, Meerwaldt JH, van Putten WL, et al. Adjuvant postoperative radiotherapy for gastric carcinoma with poor prognostic signs. Radiother Oncol 1989;16:269-274. 155. Allum WH, Hallissey MT, Ward LC, et al. A controlled, prospective, randomised trial of adjuvant chemotherapy or radiotherapy in resectable gastric cancer: interim report. British Stomach Cancer Group. Br J Cancer 1989;60:739-744. 156. Gill PG, Jamieson GG, Denham J, et al. Treatment of adenocarcinoma of the cardia with synchronous chemotherapy and radiotherapy. Br J Surg 1990;77:1020-1023. 157. Gez E, Sulkes A, Yablonsky-Peretz T, et al. Combined 5-­fluorouracil (5-FU) and radiation therapy following resection of locally advanced gastric carcinoma. J Surg Oncol 1986;31: 139-142. 158. Kern KA. Gastric cancer: a neoplastic enigma. J Surg Oncol Suppl 1989;1:34-39. 159. Kodama Y, Sugimachi K, Soejima K, et al. Evaluation of extensive lymph node dissection for carcinoma of the stomach. World J Surg 1981;5:241-248. 160. Okajima K. Surgical treatment of gastric cancer with special reference to lymph node removal. Acta Med Okayama 1977;31:369-382. 161. Bonenkamp JJ, Songun I, Hermans J, et al. Randomised comparison of morbidity after D1 and D2 dissection for gastric cancer in 996 Dutch patients. Lancet 1995;345:745-748. 162. Bonenkamp JJ, van de Velde CJ, Sasako M, et al. R2 compared with R1 resection for gastric cancer: morbidity and mortality in a prospective, randomised trial. Eur J Surg 1992;158:413-418. 163. Bonenkamp JJ, Hermans J, Sasako M, et al. Extended lymphnode dissection for gastric cancer. N Engl J Med 1999;340: 908-914. 164. Smith JW, Shiu MH, Kelsey L, et al. Morbidity of radical lymphadenectomy in the curative resection of gastric carcinoma. Arch Surg 1991;126:1469-1473. 165. Robertson CS, Chung SC, Woods SD, et al. A prospective randomized trial comparing R1 subtotal gastrectomy with R3 total gastrectomy for antral cancer. Ann Surg 1994;220:176-182. 166. Dent DM, Madden MV, Price SK. Randomized comparison of R1 and R2 gastrectomy for gastric carcinoma. Br J Surg 1988;75:110-112. 167. Noguchi Y, Imada T, Matsumoto A, et al. Radical surgery for gastric cancer. A review of the Japanese experience. Cancer 1989;64:2053-2062. 168. Adashek K, Sanger J, Longmire WP Jr. Cancer of the stomach. Review of consecutive ten year intervals. Ann Surg 1979;189:6-10. 169. Buchholtz TW, Welch CE, Malt RA. Clinical correlates of ­resectability and survival in gastric carcinoma. Ann Surg 1978;188:711-715.

21  The Stomach and Small Intestine 170. Gastrointestinal Tumor Study Group. Controlled trial of adjuvant chemotherapy following curative resection for gastric cancer. Cancer 1982;49:1116-1122. 171. Robinson E, Cohen Y. The combination of surgery, radiotherapy and chemotherapy in the treatment of gastric cancer. Recent Results Cancer Res 1977:177-180. 172. Regine WF, Mohiuddin M. Impact of adjuvant therapy on locally advanced adenocarcinoma of the stomach. Int J Radiat Oncol Biol Phys 1992;24:921-927. 173. Martinez-Monge R, Calvo FA, Azinovic I, et al. Patterns of failure and long-term results in high-risk resected gastric cancer treated with postoperative radiotherapy with or without intraoperative electron boost. J Surg Oncol 1997;66:24-29. 174. Takahashi M, Abe M. Intra-operative radiotherapy for carcinoma of the stomach. Eur J Surg Oncol 1986;12:247-250. 175. Sindelar WF, Kinsella TJ, Tepper JE, et al. Randomized trial of intraoperative radiotherapy in carcinoma of the stomach. Am J Surg 1993;165:178-186; discussion 186–177. 176. Hallissey MT, Dunn JA, Ward LC, et al. The second British Stomach Cancer Group trial of adjuvant radiotherapy or chemotherapy in resectable gastric cancer: five-year follow-up. Lancet 1994;343:1309-1312. 177. Macdonald JS, Smalley SR, Benedetti J, et al. Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach or gastroesophageal junction. N Engl J Med 2001;345:725-730. 178. Malley SR, Gunderson L, Tepper J, et al. Gastric surgical adjuvant radiotherapy consensus report: rationale and treatment implementation. Int J Radiat Oncol Biol Phys 2002;52:283-293. 179. Sano T, Sasako M, Yamamoto S, et al. Gastric cancer surgery: morbidity and mortality results from a prospective ­randomized controlled trial comparing D2 and extended para-aortic ­lymphadenectomy—Japan Clinical Oncology Group study 9501. J Clin Oncol 2004;22:2767-2773. 180. Mansfield PF. Lymphadenectomy for gastric cancer. J Clin Oncol 2004;22:2759-2761. 181. Kim S, Lim DH, Lee J, et al. An observational study suggesting clinical benefit for adjuvant postoperative chemoradiation in a population of over 500 cases after gastric resection with D2 nodal dissection for adenocarcinoma of the stomach. Int J Radiat Oncol Biol Phys 2005;63:1279-1285. 182. Moody JS. Post-operative radiotherapy for gastric carcinoma: a SEER analysis [abstract]. Proceedings of the American Society of Clinical Oncologists Gastrointestinal Symposium. Proc Am Soc Clin Oncol 2007;2635a. 183. Alcobendas F, Milla A, Estape J, et al. Mitomycin C as an adjuvant in resected gastric cancer. Ann Surg 1983;198:13-17. 184. Engstrom PF, Lavin PT, Douglass HO Jr, et al. Postoperative adjuvant 5-fluorouracil plus methyl-CCNU therapy for gastric cancer patients. Eastern Cooperative Oncology Group study (EST 3275). Cancer 1985;55:1868-1873. 185. Higgins GA, Humphrey EW, Dwight RW, et al. Preoperative radiation and surgery for cancer of the rectum. Veterans Administration Surgical Oncology Group Trial II. Cancer 1986;58:352-359. 186. Coombes RC, Schein PS, Chilvers CE, et al. A randomized trial comparing adjuvant fluorouracil, doxorubicin, and mitomycin with no treatment in operable gastric cancer. International Collaborative Cancer Group. J Clin Oncol 1990;8:1362-1369. 187. Krook JE, O’Connell MJ, Wieand HS, et al. A prospective, randomized evaluation of intensive-course 5-fluorouracil plus doxorubicin as surgical adjuvant chemotherapy for resected gastric cancer. Cancer 1991;67:2454-2458. 188. Hermans J, Bonenkamp JJ, Boon MC, et al. Adjuvant therapy after curative resection for gastric cancer: meta-analysis of randomized trials. J Clin Oncol 1993;11:1441-1447. 189. Earle CC, Maroun JA. Adjuvant chemotherapy after curative resection for gastric cancer in non-Asian patients: revisiting a metaanalysis of randomised trials. Eur J Cancer 1999;35:1059-1064. 190. Mari E, Floriani I, Tinazzi A, et al. Efficacy of adjuvant chemotherapy after curative resection for gastric cancer: a meta-analysis of published randomised trials. A study of the GISCAD (Gruppo Italiano per lo Studio dei Carcinomi dell’Apparato Digerente). Ann Oncol 2000;11:837-843.

513

191. Panzini I, Gianni L, Fattori PP, et al. Adjuvant chemotherapy in gastric cancer: a meta-analysis of randomized trials and a comparison with previous meta-analyses. Tumori 2002;88:21-27. 192. Janunger KG, Hafstrom L, Glimelius B. Chemotherapy in gastric cancer: a review and updated meta-analysis. Eur J Surg 2002;168:597-608. 193. Ajani JA, Ota DM, Jessup JM, et al. Resectable gastric carcinoma. An evaluation of preoperative and postoperative chemotherapy. Cancer 1991;68:1501-1506. 194. Leichman L, Silberman H, Leichman CG, et al. Preoperative systemic chemotherapy followed by adjuvant postoperative intraperitoneal therapy for gastric cancer: a University of Southern California pilot program. J Clin Oncol 1992;10:1933-1942. 195. Lerner A, Gonin R, Steele GD Jr, et al. Etoposide, doxorubicin, and cisplatin chemotherapy for advanced gastric adenocarcinoma: results of a phase II trial. J Clin Oncol 1992;10:536-540. 196. Lowy AM, Mansfield PF, Leach SD, et al. Response to neoadjuvant chemotherapy best predicts survival after curative resection of gastric cancer. Ann Surg 1999;229:303-308. 197. Plukker JT, Mulder NH, Sleijfer DT, et al. Chemotherapy and surgery for locally advanced cancer of the cardia and fundus: phase II study with methotrexate and 5-fluorouracil. Br J Surg 1991;78:955-958. 198. Groves LK, Rodriguez-Antunez A. Treatment of carcinoma of the esophagus and gastric cardia with concentrated preoperative irradiation followed by early operation. A progress report. Ann Thorac Surg 1973;15:333-338. 199. Hoshi H. Histologic study on the effect of preoperative irradiation of gastric cancer. Tohoku J Exp Med 1968;96:293-311. 200. Kamimura S, Mikuriya S, Hara O, et al. [Preoperative radiotherapy of carcinoma of the stomach and breast]. Gan To Kagaku Ryoho 1987;14:1558-1563. 201. Mikuriya S. Oh’ami H. Radiotherapy and cellular infiltration of tumor nests. Radiat Med 1983;1:248-254. 202. Stoliarov VI, Volkov ON, Kanaev SV, et al. [Comparative evaluation of surgical and combined treatment of cancer of the proximal region of the stomach]. Vopr Onkol 1984;30:31-35. 203. Kosse VA. Combined treatment of gastric cancer using hypoxic radiotherapy]. Vopr Onkol 1990;36:1349-1353. 204. Shchepotin IB, Evans SR, Chorny V, et al. Intensive preoperative radiotherapy with local hyperthermia for the treatment of gastric carcinoma. Surg Oncol 1994;3:37-44. 205. Talaev MI, Starinskii VV, Kovalev BN, et al. [Results of combined treatment of cancer of the gastric antrum and gastric body]. Vopr Onkol 1990;36:1485-1488. 206. Lowy AM, Feig BW, Janjan N, et al. A pilot study of preoperative chemoradiotherapy for resectable gastric cancer. Ann Surg Oncol 2001;8:519-524. 207. Zhang ZX, Gu XZ, Yin WB, et al. Randomized clinical trial on the combination of preoperative irradiation and surgery in the treatment of adenocarcinoma of gastric cardia (AGC)—report on 370 patients. Int J Radiat Oncol Biol Phys 1998;42:929-934. 208. Ajani JA, Mansfield PF, Crane CH, et al. Paclitaxel-based chemoradiotherapy in localized gastric carcinoma: degree of pathologic response and not clinical parameters dictated patient outcome. J Clin Oncol 2005;23:1237-1244. 209. Ajani JA, Mansfield PF, Janjan N, et al. Multi-institutional trial of preoperative chemoradiotherapy in patients with potentially resectable gastric carcinoma. J Clin Oncol 2004;22:2774-2780. 210. Ajani JA, Winter K, Okawara GS, et al. Phase II trial of preoperative chemoradiation in patients with localized gastric adenocarcinoma (RTOG 9904): quality of combined modality therapy and pathologic response. J Clin Oncol 2006;24:3953-3958. 211. Reed VK, Krishnan S, Mansfield PF, et al. Incidence, natural history, and patterns of locoregional recurrence in gastric cancer patients treated with preoperative chemoradiotherapy. Int J Radiat Oncol Biol Phys 2008;71:741-747. 212. Cunningham D, Allum WH, Stenning SP, et al. Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer. N Engl J Med 2006;355:11-20. 213. Gastrointestinal Tumor Study Group. A comparison of combination chemotherapy and combined modality therapy for locally advanced gastric carcinoma. Cancer 1982;49:1771-1777.

514

PART 6  Gastrointestinal Tract

214. Gastrointestinal Tumor Study Group. The concept of locally advanced gastric cancer. Effect of treatment on outcome. Cancer 1990;66:2324-2330. 215. Buffet C, Turner K, Pelletier G, et al. Palliative radiotherapy. Br J Surg 1983;70:131. 216. Falkson G, van Eden EB. A controlled clinical trial of fluorouracil plus imidazole carboxamide dimethyl triazeno plus vincristine plus bis-chloroethyl nitrosourea plus radiotherapy in stomach cancer. Med Pediatr Oncol 1976;2:111-117. 217. Gunderson LL, Hoskins RB, Cohen AC, et al. Combined modality treatment of gastric cancer. Int J Radiat Oncol Biol Phys 1983;9:965-975. 218. Gastrointestinal Tumor Study Group. Phase II-III chemotherapy studies in advanced gastric cancer. Cancer Treat Rep 1979;63:1871-1876. 219. Comis RL, Carter SK. A review of chemotherapy in gastric cancer. Cancer 1974;34:1576-1586. 220. Moertel CG. Chemotherapy of gastrointestinal cancer. Clin Gastroenterol 1976;5:777-793. 221. Pyrhonen S, Kuitunen T, Nyandoto P, et al. Randomised comparison of fluorouracil, epidoxorubicin and methotrexate (FEMTX) plus supportive care with supportive care alone in patients with non-resectable gastric cancer. Br J Cancer 1995;71:587-591. 222. Glimelius B, Ekstrom K, Hoffman K, et al. Randomized comparison between chemotherapy plus best supportive care with best supportive care in advanced gastric cancer. Ann Oncol 1997;8:163-168. 223. Murad AM, Santiago FF, Petroianu A, et al. Modified therapy with 5-fluorouracil, doxorubicin, and methotrexate in advanced gastric cancer. Cancer 1993;72:37-41. 224. Cullinan SA, Moertel CG, Fleming TR, et al. A comparison of three chemotherapeutic regimens in the treatment of advanced pancreatic and gastric carcinoma. Fluorouracil vs fluorouracil and doxorubicin vs fluorouracil, doxorubicin, and mitomycin. JAMA 1985;253:2061-2067. 225. Vanhoefer U, Rougier P, Wilke H, et al. Final results of a randomized phase III trial of sequential high-dose methotrexate, fluorouracil, and doxorubicin versus etoposide, leucovorin, and fluorouracil versus infusional fluorouracil and cisplatin in advanced gastric cancer: a trial of the European Organization for Research and Treatment of Cancer Gastrointestinal Tract Cancer Cooperative Group. J Clin Oncol 2000;18:2648-2657. 226. Henning GT, Schild SE, Stafford SL, et al. Results of irradiation or chemoirradiation following resection of gastric adenocarcinoma. Int J Radiat Oncol Biol Phys 2000;46:589-598. 227. Schuster-Uitterhoeve AL, Gonzalez Gonzalez D, Blank LE. Radiotherapy with multiple fractions per day in pancreatic and bile duct cancer. Radiother Oncol 1986;7:205-213. 228. Seydel HG, Stablein DM, Leichman LP, et al. Hyperfractionated radiation and chemotherapy for unresectable localized adenocarcinoma of the pancreas. The Gastrointestinal Tumor Study Group experience. Cancer 1990;65:1478-1482. 229. O’Connell MJ, Gunderson LL, Moertel CG, et al. A pilot study to determine clinical tolerability of intensive combined modality therapy for locally unresectable gastric cancer. Int J Radiat Oncol Biol Phys 1985;11:1827-1831. 230. Ellis F, Weatherburn H. RBE and clinical response in radiotherapy with neutron beams. Br J Radiol 1984;57:817-822. 231. Galandiuk S, Wieand HS, Moertel CG, et al. Patterns of recurrence after curative resection of carcinoma of the colon and rectum. Surg Gynecol Obstet 1992;174:27-32. 232. Kelsen D, Karpeh M, Schwartz G, et al. Neoadjuvant therapy of high-risk gastric cancer: a phase II trial of preoperative FAMTX and postoperative intraperitoneal fluorouracil-cisplatin plus intravenous fluorouracil. J Clin Oncol 1996;14:1818-1828. 233. Fujimoto S, Shrestha RD, Kokubun M, et al. Intraperitoneal hyperthermic perfusion combined with surgery effective for gastric cancer patients with peritoneal seeding. Ann Surg 1988;208:36-41. 234. Fujimoto S, Shrestha RD, Kobayashi K, et al. Combined treatment with surgery and intraperitoneal hyperthermic perfusion (IPHP) for far-advanced gastrointestinal cancer with peritoneal seeding. Acta Med Austriaca 1989;16:76-80.

235. Koga S, Hamazoe R, Maeta M, et al. Prophylactic therapy for peritoneal recurrence of gastric cancer by continuous hyperthermic peritoneal perfusion with mitomycin C. Cancer 1988;61:232-237. 236. Evans DB, Abbruzzese JL, Cleary KR, et al. Preoperative chemoradiation for adenocarcinoma of the pancreas: excessive toxicity of prophylactic hepatic irradiation. Int J Radiat Oncol Biol Phys 1995;33:913-918. 237. Abrams RA, Grochow LB, Chakravarthy A, et al. Intensified adjuvant therapy for pancreatic and periampullary adenocarcinoma: survival results and observations regarding patterns of failure, radiotherapy dose and CA19-9 levels. Int J Radiat Oncol Biol Phys 1999;44:1039-1046. 238. Fajardo LF, Berthrong M, Anderson RE. Small intestine. ­In: Radiation Pathology. New York: Oxford University Press, 2001, pp 228-229. 239. Montagna W, Wilson JW. A cytologic study of the intestinal epithelium of the mouse after total-body X-irradiation. J Natl Cancer Inst 1955;15:1703-1735. 240. Kinsella TJ, Bloomer WD. Tolerance of the intestine to radiation therapy. Surg Gynecol Obstet 1980;151:273-284. 241. Neugut AI, Jacobson JS, Suh S, et al. The epidemiology of cancer of the small bowel. Cancer Epidemiol Biomarkers Prev 1998;7:243-251. 242. Ellis H. Tumours of the small intestine. Semin Surg Oncol 1987;3:12-21. 243. Ciccarelli O, Welch JP, Kent GG. Primary malignant tumors of the small bowel. The Hartford Hospital experience, 1969-1983. Am J Surg 1987;153:350-354. 244. Lowenfels AB, Sonni A. Distribution of small bowel tumors. Cancer Lett 1977;3:83-86. 245. Lowenfels AB. Why are small-bowel tumours so rare? Lancet 1973;1:24-26. 246. Ross RK, Hartnett NM, Bernstein L, et al. Epidemiology of adenocarcinomas of the small intestine: is bile a small bowel carcinogen? Br J Cancer 1991;63:143-145. 247. Scudamore CH, Freeman HJ. Effects of small bowel transection, resection, or bypass in 1,2-dimethylhydrazine-induced rat intestinal neoplasia. Gastroenterology 1983;84:725-731. 248. Chow WH, Linet MS, McLaughlin JK, et al. Risk factors for small intestine cancer. Cancer Causes Control 1993;4:163-169. 249. Neugut AI, Santos J. The association between cancers of the small and large bowel. Cancer Epidemiol Biomarkers Prev 1993;2:551-553. 250. Frank JD, Shorey BA. Adenocarcinoma of the small bowel as a complication of Crohn’s disease. Gut 1973;14:120-124. 251. Hawker PC, Gyde SN, Thompson H, et al. Adenocarcinoma of the small intestine complicating Crohn’s disease. Gut 1982;23:188-193. 252. Morowitz DA, Block GE, Kirsner JB. Adenocarcinoma of the ileum complicating chronic regional enteritis. Gastroenterology 1968;55:397-402. 253. Michelassi F, Testa G, Pomidor WJ, et al. Adenocarcinoma complicating Crohn’s disease. Dis Colon Rectum 1993;36:654-661. 254. Swinson CM, Slavin G, Coles EC, et al. Coeliac disease and malignancy. Lancet 1983;1:111-115. 255. Hochberg FH, Dasilva AB, Galdabini J, et al. Gastrointestinal involvement in von Recklinghausen’s neurofibromatosis. Neurology 1974;24:1144-1151. 256. Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 107-112. 257. Whelan TJ, Dembo AJ, Bush RS, et al. Complications of whole abdominal and pelvic radiotherapy following chemotherapy for advanced ovarian cancer. Int J Radiat Oncol Biol Phys 1992;22:853-858. 258. Lai EC, Doty JE, Irving C, et al. Primary adenocarcinoma of the duodenum: analysis of survival. World J Surg 1988;12:695-699. 259. Kerremans RP, Lerut J, Penninckx FM. Primary malignant duodenal tumors. Ann Surg 1979;190:179-182. 260. Spira IA, Ghazi A, Wolff WI. Primary adenocarcinoma of the duodenum. Cancer 1977;39:1721-1726. 261. Joesting DR, Beart RW Jr, van Heerden JA, et al. Improving survival in adenocarcinoma of the duodenum. Am J Surg 1981;141:228-231.

21  The Stomach and Small Intestine 262. Alwmark A, Andersson A, Lasson A. Primary carcinoma of the duodenum. Ann Surg 1980;191:13-18. 263. Lillemoe K, Imbembo AL. Malignant neoplasms of the duodenum. Surg Gynecol Obstet 1980;150:822-826. 264. Lillemoe KD. Primary duodenal adenocarcinoma: role for aggressive resection. J Am Coll Surg 1996;183:155-156. 265. Herter FP, Cooperman AM, Ahlborn TN, et al. Surgical experience with pancreatic and periampullary cancer. Ann Surg 1982;195:274-281. 266. Williamson RC, Welch CE, Malt RA. Adenocarcinoma and lymphoma of the small intestine. Distribution and etiologic associations. Ann Surg 1983;197:172-178. 267. Sohn TA, Lillemoe KD, Cameron JL, et al. Adenocarcinoma of the duodenum: factors influencing long-term survival. J Gastrointest Surg 1998;2:79-87.

515

268. Rose DM, Hochwald SN, Klimstra DS, et al. Primary duodenal adenocarcinoma: a ten-year experience with 79 patients. J Am Coll Surg 1996;183:89-96. 269. Gastrointestinal Tumor Study Group. Further evidence of effective adjuvant combined radiation and chemotherapy ­ following curative resection of pancreatic cancer. Cancer 1987;59: 2006-2010. 270. Spitz FR, Abbruzzese JL, Lee JE, et al. Preoperative and postoperative chemoradiation strategies in patients treated with pancreaticoduodenectomy for adenocarcinoma of the pancreas. J Clin Oncol 1997;15:928-937. 271. Order SE, Donaldson SS. Carcinoids. Berlin: Springer-Verlag, 1998.

22 The Pancreas Christopher H. Crane, Gauri R. Varadhachary, Douglas B. Evans, and Robert A. Wolff ANATOMY EFFECTS OF IRRADIATION Acute Effects Late Effects EPIDEMIOLOGY AND ETIOLOGY OF PANCREATIC CANCER PRESENTING SYMPTOMS AND DIAGNOSIS TREATMENT Pancreaticoduodenectomy Staging Prognostic Factors

Anatomy The pancreas is located in the upper retroperitoneum and is closely surrounded by blood vessels that supply the ­upper abdominal viscera and critical radiosensitive structures, ­including the liver, kidneys, stomach, spinal cord, jejunum, and duodenum. For descriptive purposes, the pancreas is divided into four parts: head, neck, body, and tail. The tail is located cranial to the pancreatic head and abuts the hilum of the spleen. The head and neck are located anterior and to the right lateral aspect of the superior mesenteric artery, and the uncinate process extends around posterior to this artery. The anterior surface of the pancreas is covered with peritoneum; the posterior surface is in direct contact with the preaortic soft tissues and close to the superior mesenteric artery and splenic vein. Early tumor invasion often occurs by direct extension from the primary tumor to the surrounding critical vessels, such as the superior mesenteric vessels, the portal vein, and the celiac artery and its branches. The presence or absence of invasion of these critical vascular structures determines whether the tumor is resectable. The pancreas is drained by an abundant supply of lymphatics. Primary drainage is to the pancreaticoduodenal, suprapancreatic, pyloric, and pancreaticosplenic nodal ­regions, which all drain to the celiac lymph nodes and the superior mesenteric nodes. Nodes in the porta hepatis ­region also can be involved, especially in advanced disease. The pancreatic duct and common bile duct lie close to one another as they travel through the head of the pancreas on their way to the ampulla of Vater. Invasion or compression of the common bile duct and main pancreatic duct is responsible for the presenting symptoms of jaundice and exocrine pancreatic insufficiency. The divisions of the vagus and splanchnic nerves form the celiac and superior mesenteric plexuses. Nerve fibers reach the pancreas and other abdominal organs by ­traveling

516

Patterns of Failure Radiation Therapy Techniques Chemoradiation Therapy for Potentially Resectable Disease Downstaging Locally Advanced Disease: Defining Resectability Chemoradiation Therapy for Locally Advanced Disease Novel Molecular Targeted Treatment Approaches PROPHYLACTIC HEPATIC IRRADIATION

along the celiac and superior mesenteric artery and their branches. Arterial resection and repair with an arterial graft often results in denervation of the midgut, subsequently causing disabling diarrhea. Direct extension of tumor posteriorly to the first and second celiac ganglia leads to ­characteristic sharp pain radiating to the back.

Effects of Irradiation Acute Effects The acute effects of irradiation to the upper abdomen result from depletion of the rapidly dividing mucosal lining of the stomach (in postoperatively treated patients, that of the gastric remnant and jejunal reconstruction), which can lead to nausea and vomiting. Severe acute reactions can lead to ulceration of the stomach or duodenum. Mild nausea can result when large volumes of the liver are irradiated. Diarrhea is less common because of the relatively small volume of jejunum that is typically in the radiation field; exocrine insufficiency should be considered in patients with diarrhea. Antiemetics, pancreatic enzyme replacement, and a proton pump inhibitor or an H2-blocker should be considered for prophylaxis of acute mucosa injury in patients undergoing upper abdominal radiation therapy.

Late Effects With regard to late effects, the dose-limiting structures surrounding the pancreas include the stomach, small bowel (particularly the duodenum), kidneys, spinal cord, and liver. The dose of conventional external beam radiation therapy (EBRT) that can safely be used in the upper abdomen depends on the tolerance of these organs; these doses have been established for patients given radiation therapy by itself1-4 or in combination with 5-fluorouracil (5-FU).5 ­Radiation doses up to 50 Gy, given in 2-Gy fractions, are well tolerated even when large volumes of tissue are ­irradiated. Doses of 50 to 60 Gy can injure the gastric

22  The Pancreas

and ­ duodenal mucosa, depending on the irradiated volume. Doses up to 68 Gy are well tolerated with or without 5-FU–based chemotherapy if care is taken to limit the volume of irradiated mucosa to the smallest possible amount.6 The most common late radiation injury is ulceration of the gastric or duodenal mucosal surface, which can lead to perforation. Ulceration usually can be managed medically. Perforations can be addressed by placement of an external drain and use of antibiotics for patients with limited life expectancy or surgical repair for patients with controlled disease. With regard to kidney tolerance, mean decreases in creatinine clearance of 10% and 24% have been observed in patients when 50% and 90% to 100%, respectively, of a single kidney is irradiated with doses higher than 26 Gy.7 Despite these physiologic changes, clinically relevant compromise of kidney function is rare if at least one kidney is spared from the field. Radiation-induced liver disease (RILD), often called radiation hepatitis, is characterized by the development of anicteric ascites approximately 2 weeks to 4 months after hepatic irradiation.8 Among patients with pancreatic cancer, the whole liver has been treated safely to doses greater than 20 Gy,9 but whole-liver doses that exceed 35 Gy carry a significant risk of RILD.10 Although the risk of RILD can be predicted on the basis of radiation dose and volume,11,12 it is clear that limited volumes of liver can tolerate high radiation doses.13

Epidemiology and Etiology of Pancreatic Cancer Pancreatic cancer is the 10th most common malignancy in men and women in the United States, but it is the fourth most common cause of cancer death.14 The incidence of and absolute mortality from pancreatic cancer increased dramatically from 1920 to 1970.15 Pancreatic cancer is found primarily in Western countries, paralleling the epidemiology of colorectal cancer.16 Since 1970, the incidence has decreased among white men and increased among women and black men,17 resulting in a stabilization of rates overall. The risk factors for the development of pancreatic cancer have yet to be clearly defined. Convincing evidence indicates a causal relationship between pancreatic cancer and smoking18,19; the relative risk of developing pancreatic cancer is 1.6 to 3.1 times higher for smokers than for nonsmokers. Although some investigators have proposed associations between pancreatic cancer and alcohol and coffee consumption, the evidence supporting these associations is not as strong as for smoking.20 Studies evaluating dietary factors indicate that consumption of fruits and vegetables may be protective, but diets high in animal fat may increase risk.21,22 Exposure to toxic chemicals such as dichlorodiphenyltrichloroethane (DDT) and some petroleum ­derivatives also may increase risk.23,24 Exposure to ionizing radiation16 or chemotherapy,25 both known carcinogens, probably accounts for a small number of cases. The relationship between pancreatic cancer and diabetes mellitus is unclear. Although results from a 1995 meta-analysis26 suggested an association, results from a later study showed that the association was weaker if patients with recentonset diabetes were excluded.27 Chronic pancreatitis has been implicated, but this remains controversial.28 A very

517

small proportion (about 3%) of cases of pancreatic cancer seem to be clustered in some families, suggesting a genetic predisposition.29,30

Presenting Symptoms and Diagnosis The most common presenting symptoms of pancreatic cancer are jaundice, weight loss caused by anorexia and exocrine insufficiency, and abdominal pain. Jaundice is usually a presenting symptom in patients with pancreatic head lesions. Lesions arising in the body or tail of the pancreas more often manifest with pain, typically described as sharp and knifelike, in the middle epigastric region and radiating to the back. Pain is a symptom of locally advanced disease and often indicates unresectable disease. The initial goals in the evaluation and treatment of the symptomatic patient are to reestablish biliary tract ­patency, determine the resectability of the primary tumor, and ­establish a histologic diagnosis. Diagnosis, clinical evaluation, definitions of resectability, and management of pancreatic cancer differ among medical centers in the United States. Diagnostic approaches to determine ­ resectability may include preoperative imaging, laparoscopy, or laparotomy with attempted resection. Among the diagnostic ­imaging techniques, abdominal computed tomography (CT) scanning is the most common for confirming suspected pancreatic malignancy. Although standard CT techniques are relatively insensitive for the assessment of ­resectability, newer techniques can more successfully assess vascular involvement and liver metastases.31 Helical CT can be used for rapid scanning at thin-slice collimation during various phases of intravenous contrast administration. Sequential imaging during of contrast ­administration is critical for delineating the primary pancreatic tumor and liver metastases and for defining the relationship of the primary tumor to the adjacent vessels. Primary adenocarcinomas of the pancreas typically can be best seen as hypodense masses that contrast with the ­enhancing normal pancreatic parenchyma. Magnetic resonance imaging (MRI) was initially limited by long scanning times that resulted in the introduction of artifact caused by organ motion. Dynamic MRI with rapid scanning sequences and bolus intravenous contrast enhancement is reported to have sensitivity and specificity comparable to those of helical CT.32 Historically, histologic confirmation has been most commonly obtained at surgical exploration or by means of percutaneous CT-guided fine-needle aspiration. However, both techniques have their shortcomings; intraoperative biopsy has a relatively low yield and carries a risk of pancreatitis, and CT-guided fine-needle aspiration depends on the experience of the operator and is not possible if the lesion cannot be seen on CT. Brush samples collected during endoscopic retrograde cholangiopancreatography also have a relatively low yield.33 Laparoscopic exploration can be useful for obtaining a tissue diagnosis and has the added value of visualizing the peritoneal surfaces in patients with suspected metastases. Endoscopy of the upper gastrointestinal tract is the most valuable initial step in disease management for patients who present with obstructive jaundice from presumed

518

PART 6  Gastrointestinal Tract

pancreatic cancer. During this procedure, biliary outflow can be reestablished by the placement of an endobiliary stent and endoscopic sonography can be used to facilitate fine-needle biopsy, which adds sensitivity and specificity to lesion detection.34,35 Endoscopic sonography–directed fine-needle biopsy of the pancreas is the procedure of choice for the diagnosis of pancreatic malignancy at many centers. The use of endobiliary stents has largely replaced the more traditional palliative surgical bypass; however, ­because biliary tract patency with stents is less durable than a bypass, ­patients with potentially resectable disease who are found to have unresectable or metastatic disease at surgical ­ exploration are best treated with a palliative ­surgical bypass. Assessment of tumor markers such as carcinoembryonic antigen and CA 19-9 is complementary to a complete workup. Because these tests lack sensitivity and specificity, they should not be used as screening tests for pancreatic cancer. However, elevated levels of these tumor markers in combination with positive results on imaging studies ­increase the positive predictive value of the diagnosis of pancreatic cancer,36 and serial measurements of CA 19-9 levels can be useful for monitoring response to therapy.

Treatment Pancreaticoduodenectomy Whipple and colleagues37 first described pancreaticoduodenectomy in 1935. More than 70 years later, this ­procedure remains the mainstay of curative treatment for pancreatic cancer. Nevertheless, 80% to 95% of tumors are unresectable at diagnosis because of local or regional extension or metastatic spread. Moreover, of the 5% to 20% of patients with resectable disease, obtaining surgical margins that are free of disease is possible for only 16% to 30%. ­Unfortunately, the median survival times for ­patients with ­ disease-positive margins after surgery38-40 are similar to those for patients treated with chemoradiation therapy alone5,41-45 (9 to 11 months). Accurate preoperative ­radiographic ­assessment of resectability and use of ­objective definitions of resectability are critical for ­avoiding incomplete ­resections. Lesions of the pancreatic body and tail often manifest with signs and symptoms referable to invasion of surrounding structures, and most are inoperable because of arterial encasement or distant metastasis. In contrast, tumors arising in the pancreatic head often produce painless jaundice and symptoms referable to exocrine insufficiency; these symptoms more often lead to the disease being diagnosed when it is still resectable. The best surgical series report 5-year survival rates of 20% to 25%.46-48 The use of modern surgical techniques by experienced surgeons can produce operative mortality rates of less than 5% for patients who undergo pancreaticoduodenectomy.46,47

Staging Determining whether a tumor is resectable is the most important aspect of clinical staging because surgical resection offers the only chance of cure. Three criteria are necessary for tumors to be considered resectable: no metastasis, no encasement of the celiac artery or superior mesenteric artery, and patency of the superior mesenteric and portal veins.

BOX 22-1.  American Joint Committee on Cancer Staging System for Pancreatic Cancer Primary Tumor (T) TX Primary tumor cannot be assessed T0 No evidence of primary tumor Tis Carcinoma in situ* T1 Tumor limited to the pancreas, 2 cm or less in greatest dimension T2 Tumor limited to the pancreas, more than 2 cm in greatest dimension T3 Tumor extends beyond the pancreas but without involvement of the celiac axis or the superior mesenteric artery T4 Tumor involves the celiac axis or the superior ­mesenteric artery (unresectable primary tumor) Regional Lymph Nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Regional lymph node metastasis Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage Grouping Stage 0 Stage IA Stage IB Stage IIA Stage IIB Stage III Stage IV

Tis T1 T2 T3 T1 T2 T3 T4 Any T

N0 N0 N0 N0 N1 N1 N1 Any N Any N

M0 M0 M0 M0 M0 M0 M0 M0 M1

*This

includes the PanInIII classification. From Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 157-164.

Not all surgeons recommend resection if venous involvement is present, but venous resections can be performed safely with excellent oncologic outcomes. The outcome for patients requiring venous resection has been similar to that for patients who do not require resection.49 Changes in the sixth edition of American Joint Committee on Cancer staging system (Box 22-1)50 for cancer of the exocrine pancreas reflect a clinical definition of resectability based on CT assessment. Tumors designated T1, T2, and T3 are considered potentially resectable, and T4 tumors are considered locally advanced (i.e., unresectable). Tumors that involve the superior mesenteric artery or celiac artery are classified as T4; however, tumors that involve the superior mesenteric, splenic, or portal veins are classified as T3 because those veins can be resected and reconstructed if they are patent.49

Prognostic Factors Tumor size (<2 cm), the presence or absence of perineural invasion, and lymph node status are all considered significant prognostic pathologic variables.51-53 Treatment-­related variables include the use of adjuvant chemoradiation ­therapy and, most importantly, complete surgical resection

22  The Pancreas

of the primary tumor. Resections that result in microscopic or gross residual disease after pancreaticoduodenectomy are not curative. The median survival time for patients with positive margins after surgery (9 to 11 months)38-40 is the same as that after chemoradiation therapy without surgery.5,41-45 The sole exception is when the chemoradiation therapy is given before surgery (i.e., neoadjuvant or induction therapy). In a series from The University of Texas M. D. Anderson Cancer Center, patients with microscopically positive margins after neoadjuvant chemoradiation therapy had a median survival time of 20 months, a ­period statistically similar to that for patients with negative ­margins.49

Patterns of Failure Pancreatic cancer has high rates of local and distant failure even after “curative” resection. Even when locoregional control is optimized, distant recurrence leads to a mortality rate of up to 80%. Randomized trials of postoperative chemoradiation therapy have shown local failure rates ranging from 39% to 62%.54-57 The most common distant sites of failure are the liver and peritoneum58; the lungs and bone are less commonly involved.

Radiation Therapy Techniques Adjuvant Radiation Therapy Postoperative (adjuvant) radiation therapy for pancreatic cancer typically consists of 50.4 Gy given in 1.8-Gy fractions. Field reductions are often made after 45 Gy has been delivered. The most important target to be treated is the superior mesenteric artery margin, which is often close, is often positive, and is a common site of local ­recurrence. The boost volume should therefore include the superior mesenteric vessels, the tumor bed, and the celiac axis. Patients are positioned for CT simulation in the supine position, with their arms overhead and their upper bodies immobilized. Intravenous contrast administration with a 45-second delay before imaging or fusion of the planning CT with the diagnostic CT should be used to correctly identify and contour the vascular structures and organs at risk. A four-field technique using anterior, posterior, and opposed lateral fields allows sparing of critical tissues such as the liver, kidneys, stomach, spinal cord, and small bowel. Beam angles and weighting are customized to achieve dose constraints to the organs at risk. Typically, weighting that ­restricts the dose contribution from the lateral fields to 20 Gy is a good starting point in the planning process. This can be accomplished by weighting the anteroposterior– ­posteroanterior fields at twice that of the lateral fields. This approach prevents the liver and kidney tissue in the lateral fields that are not also in the anteroposterior and posteroanterior fields from receiving doses beyond their tolerance. When the right kidney would receive too high a dose, the contribution from the posterior field can be omitted or minimized. Customized oblique fields also can be used in these situations. Intensity-modulated radiation therapy is not likely to be less toxic than standard or three­dimensional conformal radiation, but it should be considered for providing higher focal doses to grossly enlarged lymph nodes or positive margins after surgery (except

519

when the positive margins are at the pancreatic remnant). In such cases, a nested gross tumor volume can be used to deliver a higher dose to those nodes and margins. A simultaneous dose of 63 Gy in 28 fractions can be delivered to these high-risk areas and 50.4 Gy to the microscopic areas at risk, provided the jejunal reconstruction can be avoided. The oral contrast agent given during CT does not fill jejunal reconstructions but is always visualized in the pancreatic remnant after pancreaticoduodenectomy. It is critical to correctly identify this reconstruction and limit the dose to it to 50.4 Gy. Although findings from the treatment of ­tumors at other sites indicate that positive margins require doses higher than 60 Gy, no studies have established that higher radiation doses can compensate for positive margins in pancreatic cancer. For lesions located in the pancreatic head, the anteroposterior and posteroanterior fields typically cover the T11 through L3 vertebrae (Fig. 22-1). However, the ­location of the tumor bed and lymphatics must be evaluated carefully when designing a treatment plan because the relation of the visceral and vascular structures to the bony anatomy varies considerably. The celiac axis, which is most commonly located at T12, should be covered with a 2-cm margin superiorly. That superior border is usually enough to cover the porta hepatis as well. Inferiorly, the goal is to cover the tumor and duodenal bed with a 2-cm margin. The left border is usually placed 2 cm to the left of the vertebral body edge, as long as coverage of the preoperative tumor volume is adequate. The preoperative tumor volume and preoperative location of the duodenum define the right field border and the anterior extent of the lateral fields. The porta hepatis should be identified on the planning CT scans and covered; useful landmarks are the bifurcation of the portal vein and the hepatic artery. This usually means that the upper right border of the field is located 4 to 5 cm to the right of the vertebral body edge and the anterior aspect of the lateral field is 5 to 6 cm from the anterior vertebral body edge. A block is placed over the inferior pole of the right kidney in the anteroposterior and posteroanterior fields, and another is placed bisecting the vertebral bodies in the lateral fields. Corner blocks should be placed in lateral fields. Care should be taken not to block the preoperative tumor volume or porta hepatis. Dose-volume histograms can help to verify the dose to be delivered to the target and to limit the dose to critical structures. For lesions of the pancreatic body and tail, similar fields are used except that the splenic hilum is covered and the porta hepatis and duodenal bed are not covered. The right field border is typically located 2 cm from the right vertebral body edge.

Radiation Therapy for Locally Advanced Disease Because patients with locally advanced tumors probably do not benefit from regional lymph node irradiation and because adequate coverage of the locoregional nodes usually increases gastrointestinal toxicity, radiation therapy fields should be confined to the gross tumor. (This ­strategy also reduces the gastrointestinal toxicity of chemoradiation therapy.) The importance of accurate characterization of pancreatic tumors cannot be overemphasized. On ­contrast-enhanced CT scans, pancreatic tumors are typically hypodense compared with the surrounding pancreatic

520

PART 6  Gastrointestinal Tract Blocks

Blocks

A

Beam’s eye view DRR for “ap”

B

Beam’s eye view DRR for “rt lat”

FIGURE 22-1.  Typical anteroposterior (A) and lateral (B) simulation films for postoperative treatment of pancreatic cancer. ­Preoperative gross tumor volume and kidneys have been contoured and are shown.

­ arenchyma. When the location of the primary tumor is in p doubt, the CT images should be reviewed with a diagnostic radiologist. Administration of intravenous and oral contrast agents at the time of simulation helps to visualize the duodenal “C-loop” and vascular structures. Alternatively, the contrast diagnostic CT images can be fused to the simulation CT. Endobiliary biliary stents are useful for identifying the common bile duct (Fig. 22-2). The pancreas and duodenum move a median of 1 cm with respiratory excursion.59 If only the gross tumor is to be treated, respiratory motion must be controlled or accounted for in radiation therapy planning. The most common way to do so is by adding an additional margin to the planned radiation fields in the cranial and caudal directions. However, because axial tumor motion is negligible, an additional margin for motion in the axial directions is not necessary. Radiation treatment that is gated to the respiratory cycle (i.e., respiratory gating)60 is a necessary component of radiation dose-escalation studies that seek to deliver more than 60 Gy to the primary tumor while sparing the duodenum. Radiation fields designed to spare the duodenum that are tightly confined to the primary tumor without correction for organ motion can lead to underdosing of the tumor target or a “marginal miss.” A four-field technique is recommended with equally weighted anterior, posterior, and opposed lateral fields. A 2-cm block margin is used in the radial directions, and a 3-cm block margin is used in the cranial and caudal directions to account for respiratory motion.

Intensity-Modulated Radiation Therapy Intensity-modulated radiation therapy allows precise delivery of radiation to a target that is not moving. If the target is not near a radiosensitive structure, the dose usually can be escalated safely, and the escalation may or may not ­improve

1

2

FIGURE 22-2.  Typical anteroposterior simulation film for the treatment of an intact pancreatic head. Contrast medium fills the duodenal C-loop. Blocking is placed to shield the right kidney. Arrow 1 indicates refluxed contrast in the biliary system, demonstrating the location of the porta hepatis. Arrow 2 indicates an endobiliary stent in the intrapancreatic portion of the common bile duct.

22  The Pancreas

outcome. However, pancreatic tumors move considerably with respiration and are virtually always surrounded by the highly radiosensitive duodenum. Moreover, because dose escalation beyond 50.4 Gy has never been convincingly demonstrated to improve outcome in pancreatic cancer, we abandoned the investigation of intensity-modulated radiation therapy for the purpose of toxicity reduction after our phase I trial demonstrated that the toxicity of gemcitabine-based chemoradiation therapy was not reduced despite improved dose precision at the target.61 However, the careful use of intensity-modulated radiation therapy to deliver focal high doses of radiation to areas of gross disease that are at least 1 or 2 cm away from visceral structures can be considered for boost doses to the lymph nodes or positive margins. In some very uncommon cases, boost doses of 63 to 68 Gy can be delivered to small tumors in the body or neck of the pancreas if the tumors are remote from the stomach, duodenum, and jejunum.

Stereotactic Radiation Therapy A relatively recent innovation in the treatment of pancreatic cancer is stereotactic radiation therapy using systems such as the CyberKnife (Accuray, Sunnyvale, CA), in which a combination of image guidance and computercontrolled robotics is used to continuously track, detect, and correct for tumor and patient movements throughout the treatment. This system was evaluated at Stanford University in a phase I trial of dose escalation of single-­fraction radiation therapy,62 in which metallic markers placed at laparoscopy or laparotomy or under CT guidance were used to account for organ motion. The final dose of 25 Gy was well tolerated and has been recommended for further study. The median time to disease progression in that study was 2 months. In a follow-up study,63 45 Gy of intensitymodulated radiation therapy was given in 25 fractions with concurrent 5-FU, followed by 25 Gy in 1 fraction given as a stereotactic boost. Although this treatment was reasonably well tolerated, the median survival time was a disappointingly short 7.6 months.63 In a similar experience, a group from Denmark investigated stereotactic radiosurgery in a phase II trial. The median survival time among 22 patients treated with 3 fractions of 15 Gy was only 5.7 months, and 4 patients experienced severe mucositis or ulceration of the stomach or duodenum. The investigators concluded that stereotactic radiation therapy “was associated with poor outcome, unacceptable toxicity, and questionable palliative effect and cannot be recommended for patients with advanced pancreatic cancer.”64 These studies were conducted at a center with highly specialized capabilities, and the treatment techniques are not yet considered ready for use outside of clinical trials. Moreover, the investigators acknowledged that it was impossible to avoid treating the duodenum to a high dose in most cases.62 Given the lack of convincing prospective findings, the constraints of organ motion, and the lack of evidence that dose escalation results in any meaningful improvement in outcome, we believe that neither intensitymodulated nor stereotactic radiation therapy has a role in treating pancreatic cancer outside of clinical trials. Treatment of locally advanced pancreatic cancer is well tolerated, and modest survival benefits can be obtained when the treatment volumes are confined to the gross tumor and

521

clinically enlarged lymph nodes, as can be accomplished with a standard four-field conformal plan. The dose to the spinal cord, kidneys, and liver can easily be kept within tolerance limits by using this straightforward approach.

Radiation Dose-Response and Dose Escalation Effective local control of locally advanced (i.e., unresectable) pancreatic cancer is an important issue from the standpoint of symptomatic palliation, and it is a prerequisite to the development of curative therapies. Radiation dose escalation using advanced radiation technology could be valuable for some patients. Several single-institution studies have addressed the role of radiation dose escalation in attempts to improve local disease control in patients with unresectable pancreatic cancer. Some studies showed that use of an intraoperative boost after preoperative radiation could substantially improve local control but only modestly improved median survival times.65-69 The use of brachytherapy by means of implants has produced similar results with regard to enhancing local control.70,71 Although selection bias is probably responsible for the modest improvements observed in median survival time in these studies, a radiation dose-response relationship does seem to exist for local tumor control. In one study, use of an aggressive radiation-based ­approach was shown for the first time to have curative ­potential for some patients with locally advanced disease.72 In that study, 150 patients with unresectable or locally advanced pancreatic cancer were treated at Massachusetts General Hospital with a combination of intraoperative electron beam radiation therapy (15 to 20 Gy in 1 fraction) and EBRT (typically 50 Gy in 5 weeks). The 1-, 2-, and 3-year actuarial survival rates for all 150 patients were 54%, 15%, and 7%, respectively. Moreover, eight patients have survived for prolonged periods—five beyond 5 years and three others between 3 and 4 years.72 Even if improved local control does improve median survival somewhat, high rates of intra-abdominal failure in unirradiated areas continue to prevent significant improvements in long-term survival for most patients. The poor results with systemic therapy for patients with measurable metastatic disease and the inability to prevent the development of distant metastatic disease with adjuvant chemotherapy have led to the investigation of new systemic agents for pancreatic cancer. As discussed in the following sections on chemoradiation therapy, radiosensitizers such as gemcitabine are being investigated for their possible systemic effect, but dramatic improvements in median survival time have yet to be demonstrated.

Chemoradiation Therapy for Potentially Resectable Disease Adjuvant Chemoradiation Therapy The superiority of the combination of 5-FU and radiation over radiation alone in a Gastrointestinal Tumor Study Group (GITSG) trial for patients with locally ­advanced pancreatic cancer in the 1970s41,55,73 ­prompted investigation of the role of chemoradiation therapy for ­patients who had undergone resection of pancreatic ­cancer (Table 22-1).54-57,74-77 This same group initiated a ­prospective, randomized study of postoperative adjuvant chemoradiation therapy56 that eventually accrued

522

PART 6  Gastrointestinal Tract

TABLE 22-1.    Postoperative Adjuvant Chemoradiation Studies in Patients with Resectable Pancreatic Cancer Study and Reference

No. of Patients*

EBRT Dose (Gy)

Chemotherapy Agent

Median Survival (mo)

21

40

5-FU

20

22

—

—

11

Randomized Chemoradiation Trials Kalser and Ellenberg, 198555 Surgery alone 199954

60

40

5-FU

17.1

54

—

—

12.6

145

—

5-FU

17.9

144

—

—

15.9

179

—

—

6.9

177

—

—

13.4

Gemcitabine†

187

50.4

5-FU

20.6

5-FU†

194

50.4

5-FU

16.9

Klinkenbijl et al,

Surgery alone Randomized Chemotherapy Trials Neoptolemos et al, 200457 Surgery alone Oettle et

al,77

2007 (DFS)

Gemcitabine† Regine et al, 200876‡

Nonrandomized Trials GITSG, 198756

30

40

5-FU

18

Foo et al,

199374

29

45+

±5-FU

22.8

Yeo et al,

199775

120

>45

5-FU

20

Surgery alone

53

—

—

14

*All

patients underwent pancreatectomy with curative intent. †Component of systemic therapy without radiation. ‡Pancreatic head lesions only. DFS, disease-free survival; EBRT, external beam radiation therapy; 5-FU, 5-fluorouracil; GITSG, Gastrointestinal Tumor Study Group.

49 ­patients over 8 years. Disease in all patients had been surgically staged; only patients with disease confined to the pancreas and peripancreatic organs, regional lymph nodes, and regional peritoneum were eligible. After pancreaticoduodenectomy, radiation therapy was delivered to the primary site of disease and regional lymphatics in a split course of 40 Gy given over 6 weeks with concurrent 5-FU (500 mg/m2/day as an intravenous bolus on days 1 to 3 of the beginning of each 2-week course of radiation). Despite the use of what is now considered to be suboptimal chemotherapy (in dose and schedule) and inadequate radiation (in dose, equipment [supervoltage], and schedule [split course]), that trial also demonstrated a survival ­advantage from multimodality therapy relative to resection alone (survival rates of 43% versus 18% at 2 years and 19% versus 0% at 5 years, P < .05). These results were duplicated when an additional 30 patients were subsequently enrolled in the experimental group.56 Approximately 50% of patients undergoing postoperative therapy experienced local recurrence, which indicates that the results might have been more striking if better chemoradiation doses and techniques had been used. Similar findings were reported more than 10 years later by the European Organisation for Research and ­Treatment of Cancer.54 Between 1987 and 1995, 218 ­patients who had undergone pancreaticoduodenectomy for ­adenocarcinoma of the pancreas or periampullary region were ­ randomly assigned to receive chemoradiation therapy (40 Gy in a split course of EBRT and 5-FU, given

as a ­ continuous infusion at a dose of 25 mg/kg/day during irradiation) or no further treatment. Eleven patients were deemed ineligible for analysis because of incomplete resection of ­extensive local disease. Of the remaining 207 patients, 114 (55%) had cancer of the pancreas. The median survival duration for the entire group was 24.5 months for patients given adjuvant therapy and 19 months for patients given surgery alone (log-rank test, P = 0.21). For patients with pancreatic cancer, the median survival time was 17.1 months for those given adjuvant therapy and 12.6 months for those given surgery alone (P = .099).54 As was true in the GITSG trial, patients in the European trial were considered for ­enrollment after they had recovered from ­ pancreaticoduodenectomy. Even so, 21 (20%) of 104 ­evaluable patients assigned to undergo chemoradiation therapy did not ­receive the ­intended therapy because of patient refusal, comorbid medical conditions, or rapid tumor progression. In this intent-to-treat analysis, the low level of compliance might have been responsible for the lack of a clear benefit. In addition to these randomized studies, single-institution reports also have shown that postoperative chemoradiation therapy can improve overall survival (see Table 22-1).74,75,78 The Radiation Therapy Oncology Group (RTOG) conducted a randomized trial (RTOG 97-04)76 in which patients who had undergone pancreaticoduodenectomy were given adjuvant systemic therapy consisting of gemcitabine (1000 mg/m2) or infused 5-FU (250 mg/m2) followed by concurrent chemoradiation therapy, in which both groups

22  The Pancreas

received 50.4 Gy in 28 fractions to regional fields plus concurrent 5-FU (250 mg/m2/day) followed by gemcitabine or 5-FU. A planned subset analysis revealed an improvement in median overall survival time in the group given gemcitabine (20.6 versus 16.9 months, P = .033) among patients with lesions arising in the pancreatic head. No difference was found between the use of gemcitabine and the use of 5-FU for patients with lesions of the pancreatic body.76 The Chariti Onkologie (CONKO-001) trial,77 conducted in Germany, evaluated 6 months of treatment with gemcitabine alone versus observation only after pancreaticoduodenectomy. Chemoradiation therapy was not used. Disease-free survival time was different in the two groups (median, 13.4 months for the gemcitabine group versus 6.9 months for the observation group, P < .001) but overall survival was not. The 5-year survival rates were 20.5% for the gemcitabine group and 11.5% for the observation group (P = .6).77 In that study, like all of the other randomized trials of adjuvant therapy conducted, there was no central surgical or pathologic quality control, and CT scans were not required before study entry. Unrecognized incomplete surgical resections may well have confounded interpretation of these results. The controversial European Study Group of Pancreatic Cancer-1 (ESPAC-1) study has called into question the benefits of chemoradiation therapy.57,79 The ESPAC-1 trial was a four-arm study with a 2 × 2 factorial design that compared the effects of adjuvant chemoradiation therapy (40 Gy in a split course with concurrent 5-FU), adjuvant chemotherapy (5-FU and leucovorin), chemoradiation therapy followed by chemotherapy, and observation alone after pancreaticoduodenectomy for pancreatic or periampullary carcinoma. Accrual for this study began in 1994, and medical centers in 11 countries enrolled 530 patients. Most patients were entered into the randomized 2 × 2 factorial design. However, because of lack of access to radiation therapy or because of specific institutional bias, 188 patients were randomly assigned to receive only chemotherapy or no chemotherapy, and 68 patients were ­randomly assigned to receive chemoradiation or no chemoradiation therapy. In the latter two nonfactorial groups, patients could receive nonstandard therapy at the discretion of their treating physicians. For example, patients who were assigned in the nonfactorial design to chemotherapy or no chemotherapy could receive radiation therapy. Moreover, the nonrandom treatments were not standardized, and 25% of living patients had less than 1 month of follow-up. The initial analysis of all patients in this study suggested no benefit from being given postoperative chemoradiation therapy but a survival benefit from receiving adjuvant 5-FU and leucovorin chemotherapy. Data from the 2 × 2 factorial analysis suggested that postoperative chemoradiation therapy had a harmful effect.79 The nonfactorial and 2 × 2 factorial arms of the study produced contradictory results. Numerically improved survival that did not reach statistical significance was seen among the 33 patients who were randomized to undergo postoperative adjuvant chemoradiation therapy in the nonfactorial arm. For the 285 patients randomized in the 2 × 2 factorial design, postoperative chemotherapy showed no survival benefit over surgery alone, but improved survival was found among the 92 patients who received chemotherapy in the nonfactorial arm. The finding of improved survival from

523

systemic therapy that is known to have minimal ­ activity in metastatic disease was surprising. The inconsistencies in these results make accurate analysis of the findings quite difficult and raise questions about the reliability of any conclusions drawn. The subsequent final report of the 2 × 2 factorial component of the study57 has provided additional insight into recurrence patterns: 62% of patients had a component of local disease as the initial site of recurrence. Local disease recurrence rates this high can only mean that significant numbers of patients were enrolled in this trial after undergoing incomplete resections. Although only 18% of patients were reported to have had positive margins, this is probably an underestimate because, as is true for most trials, standardized specimen processing was not required. CT scans to exclude gross residual disease would have helped to exclude gross residual disease but were not required before admission to the trial. The distribution of local recurrence between the treatment groups was not reported, but imbalances in noncurative resections probably led to the higher rates of local recurrence in the chemoradiation therapy groups and influenced the outcome of the trial. Commonly cited criticisms of this trial include its lack of quality control, use of nonstandardized off-study therapies, poor protocol compliance, and the use of outdated doses and schedules of therapy. In summary, results from the randomized trials of adjuvant therapy for pancreatic cancer are difficult to interpret because of the potential inclusion of patients with incompletely resected tumors, the use of outdated chemotherapy and chemoradiation therapy regimens in the older trials, and the lack of surgical, pathologic, radiographic, or radiotherapeutic quality control in many of the trials.80 The results seem to support the inclusion of both 5-FU–based chemoradiation therapy and systemic gemcitabine (as was given to the experimental group in RTOG 97-04), but ­other conclusions are possible. Patterns of recurrence indicate high rates of local and distant disease in all of these studies. Current investigations include the United States Eastern Cooperative Oncology Group (ECOG)/Intergroup 2204 trial, in which chemoradiation therapy is to be given to both treatment groups; the European Organisation for Research and Treatment of Cancer 40013 trial, in which one group is to be given chemoradiation therapy; and the ESPAC-3 trial, in which one group is to be given chemotherapy alone (Box 22-2).

Neoadjuvant Chemoradiation Therapy Preoperative chemoradiation therapy is increasingly being used to improve the chances of obtaining negative surgical margins, to increase tumor control rates among patients who have resectable disease at the time of clinical evaluation, and to convert borderline-resectable disease into resectable disease. The rationale for the use of preoperative therapy, as opposed to postoperative adjuvant therapy, has been discussed in detail elsewhere.81-83 During the past 20 years, investigators at M. D. Anderson Cancer Center have conducted a series of clinical trials to evaluate preoperative chemoradiation therapy.82,84-87 Eligibility criteria were identical in these trials and included use of a CT-based definition of resectable disease, a requirement for biopsyproven adenocarcinoma of the pancreatic head, a uniform pancreaticoduodenectomy technique, and a standardized

524

PART 6  Gastrointestinal Tract

BOX 22-2.  Ongoing Multi-Institutional Trials of Adjuvant Treatment for Pancreatic Adenocarcinoma Resected with Curative Intent Study

Design

Evaluation or Comparison

American College of Surgeons Oncology Group Z5031*

Single-arm phase II study of interferon-based adjuvant chemoradiation therapy

Surgery followed by chemoRT (50.4 Gy/5.5 weeks) + weekly 5-FU + interferon + cisplatin

European Organisation for ­Research and Treatment of Cancer 40013

Randomized phase III study of adjuvant gemcitabine followed by gemcitabine plus concomitant radiation vs. ­gemcitabine alone

Surgery followed by four cycles of ­gemcitabine (1000 mg/m2 once weekly for 3 weeks ­followed by 1 week rest per cycle) vs. Surgery followed by two cycles of gemcitabine (1000 mg/m2 once weekly for 3 weeks followed by 1-week rest per cycle) followed by chemoRT (gemcitabine 300 mg/m2 4 hours before RT [50.4 Gy in 28 1.8-Gy fractions])

European Study Group for ­Pancreatic Cancer 3

Phase III study of adjuvant chemotherapy

Surgery followed by 5-FU + leucovorin vs. Surgery followed by gemcitabine

Eastern Cooperative Oncology Group/GI Intergroup 2204

Randomized phase II study of ­adjuvant bevacizumab or cetuximab with ­gemcitabine followed by capecitabine and bevacizumab with RT

Surgery followed by gemcitabine + cetuximab followed by capecitabine + bevacizumab + RT (50.4 Gy/5.5 weeks) followed by ­gemcitabine + cetuximab vs. Surgery followed by gemcitabine + bevacizumab followed by capecitabine + bevacizumab + RT (50.4 Gy/5.5 weeks) followed by gemcitabine + bevacizumab

*Closed to new patient accrual. chemoRT, chemoradiation therapy; GI, gastrointestinal; RT, radiation therapy.

system for pathologic evaluation of surgical specimens including resection margins, as described elsewhere.81 The initial preoperative regimen at M. D. Anderson reported in 1992 consisted of 50.4 Gy given over 5.5 weeks in standard fractionation to the tumor and regional lymph nodes with concurrent protracted-infusion 5-FU (300 mg/m2/day, Monday through Friday). Intraoperative ­radiation therapy also was used in selected cases. Because ­approximately one third of patients required hospitalization for significant acute gastrointestinal toxic effects (nausea, vomiting, and dehydration), the radiation dose was subsequently modified in favor of a short course of “rapid fractionation” radiation therapy (30 Gy given in 10 fractions over 2 weeks to the tumor and regional lymph nodes), with the same chemotherapy, and a supplemental 10-Gy dose of intraoperative radiation therapy delivered during surgical resection for all patients. (Linear quadratic modeling indicated that the effective dose delivered to the ­tumor bed with the latter approach is comparable to that ­delivered with standard fractionation.88) Efficacy measures in these studies included eventual resection, the degree of pathologic cell kill according to a standardized ­evaluation,82 and median survival time. Although the pancreatic tumors ­remained technically resectable after chemoradiation therapy, approximately 40% were not resected because metastatic disease was identified at the time of radiographic restaging 4 to 5 weeks after the neoadjuvant therapy.84 A pathologic partial response to therapy (defined as >50% of the ­ tumor cells in the surgical specimen being nonviable) was seen in ­approximately 40% of cases. Local tumor

control and median survival times were similar with the ­standard-fractionation regimen (50.4 Gy in 5.5 weeks) and the ­ rapid-fractionation regimen and were comparable to ­results after postoperative chemoradiation therapy (median survival times of 18 months for the standard regimen and 25 months for the rapid-fractionation regimen). The M. D. Anderson group also investigated the use of paclitaxel as a preoperative radiosensitizer in patients with resectable pancreatic cancer. Unfortunately, paclitaxel was no more advantageous than 5-FU–based chemoradiation therapy in terms of toxicity, resection rate, local treatment effect, or overall survival.84,85 Gemcitabine was incorporated into the same rapid-fractionation radiation therapy schedule and technique as that used for the 5-FU–based regimen.86 The dosage, 400 mg/m2 given weekly for 7 weeks, was chosen based on findings from a phase I trial for patients with locally advanced disease.87 Intraoperative radiation therapy was not used because of concerns about potential toxicity. Among the 86 patients enrolled in this clinical trial,86 the toxic effects were similar to those observed in the initial phase I trial. Even though the time from enrollment to surgery (pancreaticoduodenectomy) was longer (11 to 12 weeks) in this study than in previous trials of 5-FU and EBRT (7 to 9 weeks), 74% of ­patients underwent successful pancreaticoduodenectomy, compared with 57% to 60% after 5-FU–based chemoradiation therapy. Pathologic partial responses were observed in more than one half of the surgical specimens, and two ­patients showed pathologic complete responses—something not observed in the 5-FU studies. Patients with potentially

22  The Pancreas

resectable pancreatic cancer at M. D. Anderson are given systemic therapy with a combination of gemcitabine and cisplatin followed by gemcitabine-based chemoradiation therapy before planned pancreaticoduodenectomy. Unfortunately, many studies of preoperative therapy for pancreatic cancer have included heterogeneous patient populations, enrolling patients with locally advanced or marginally resectable pancreatic cancer. Conversely, many studies of chemoradiation therapy intended for locally advanced disease have included patients with radiographically resectable disease, leading some to conclude that chemoradiation therapy can downstage tumors from unresectable to resectable (see “Chemoradiation Therapy for Downstaging Locally Advanced Disease”). Few investigations have included clear anatomic definitions of locally advanced disease; many studies were based on the extent of local tumor as assessed during surgery, observations that are subjective and not reproducible. In general, patients with locally advanced pancreatic cancer should not be included in studies of preoperative therapy for clearly resectable disease, and patients with resectable tumors should not be included in studies intended for locally advanced disease. Doing so confounds reports of resection rates and the downstaging potential of chemoradiation therapy and complicates interstudy comparisons. Therein lies the importance of using accurate, reproducible anatomic definitions for resectability that are based on high-quality CT imaging.

Barriers to Neoadjuvant Therapy Difficulties associated with the delivery of preoperative therapy for pancreatic cancer include the need for a pretreatment biopsy and often for intermediate-term palliation of jaundice with endoscopically placed biliary stents.89,90 Preoperative chemoradiation therapy before a planned pancreaticoduodenectomy usually should be ­administered only on cytologic confirmation of malignancy. Fine-needle aspiration guided by endoscopic sonography is the procedure of choice for obtaining a cytologic diagnosis of adenocarcinoma. Reports of this procedure have shown it to be accurate and safe for pancreatic ­lesions.31,35,91,92 When fine-needle aspiration specimens are interpreted by an ­experienced cytopathologist, false-positive results are rare, and false-negative results become less common as physician experience increases.83,93 Most patients with adenocarcinoma of the pancreatic head have biliary obstruction at the time of diagnosis due to obstruction of the intrapancreatic portion of the common bile duct. The delivery of chemoradiation therapy ­before pancreaticoduodenectomy delays surgery and therefore requires that patients with biliary obstruction receive some form of nonoperative biliary decompression. This is usually done with an endoscopically placed polyethylene stent. Some investigators have suggested that the use of prosthetic biliary drains increases the risk of chemoradiation-related morbidity and subsequent operative morbidity and mortality.92,94,95 In an ECOG study, stent-related recurrent biliary obstruction with cholangitis was thought to be the cause of 38% of hospital admissions for ­treatmentrelated complications before pancreaticoduodenectomy.95 This experience prompted a review of the risk of stent­related morbidity at M. D. Anderson.89 Among 154 patients treated with a 3- to 5-week course of preoperative ­chemotherapy and ­ concurrent EBRT (30 or 50.4 Gy),

525

nonoperative ­ biliary decompression was performed in 101 patients (66%): endobiliary stent placement in 77 and percutaneous transhepatic catheter placement in 24. Stent-­related complications (i.e., occlusion or migration) occurred in 15 patients, 7 of whom required hospitalization for antibiotics and stent exchange (median hospital stay, 3 days). No patient experienced uncontrolled biliary sepsis, hepatic abscess, or stent-related death. The overall risk for biliary stent occlusion (with or without cholangitis) among patients receiving chemoradiation therapy was ­approximately 15%.89 This experience was not reproduced, however, when the preoperative interval was increased from 6 to 8 weeks to 3 to 4 months. As the preoperative ­interval increased, so did the incidence of plastic stent ­ occlusion. For this reason, we began placing covered metal stents in patients whose preoperative interval was anticipated to be greater than 6 to 8 weeks.96

Downstaging Locally Advanced Disease: Defining Resectability Because the only curative treatment for pancreatic cancer available is surgical resection, an objective clinical definition of resectability is crucial in the assessment of localized pancreatic cancer. There is a widespread perception that chemoradiation therapy can render unresectable pancreatic tumors resectable. The interpretation of whether this downstaging actually occurs is limited by inconsistent and subjective definitions of resectability and by inadequate preoperative radiologic assessments of resectability. Probably the most variable factor in determining resectability and in interpreting whether a tumor has been converted to resectable from unresectable is the type and extent of vascular involvement. Although most surgeons would agree that tumors that encase the celiac artery or the superior mesenteric artery constitute unresectable disease, opinions vary with regard to more limited arterial involvement; theoretically, active cytotoxic therapy could lead to downstaging in such cases. At M. D. Anderson, locally advanced tumors resected with negative margins typically have had very limited arterial involvement (less than one third the circumference of the artery and less than 1 cm along the length of the artery) and responded to chemoradiation therapy.97 Such cases are sometimes referred to as marginally resectable. Another factor affecting the determination of resectability and whether resectability can be increased is the significance of tumor involvement of the superior mesenteric and portal venous confluence. Tumor extension to a venous structure without occlusion is not an absolute contraindication to resection. Veins can be successfully resected and reconstructed at the time of pancreaticoduodenectomy. However, many surgeons would consider this type of tumor extension, seen during surgery or on preoperative imaging, as evidence of unresectability. The attribution of increased resectability to chemoradiation therapy in some studies may result from differences in surgical opinion and practice. The existence of broader definitions of locally ­advanced pancreatic cancer gives the impression that resectability is increased more commonly than it actually is. Another potential confounding variable in the interpretation of resectability is the lack of reproducible imaging before and after chemoradiation therapy. Imaging that is not designed to address the issue of vascular involvement

526

PART 6  Gastrointestinal Tract

may not have sufficient resolution to allow such an assessment. CT scans taken of the same patient at different times may result in different interpretations, even before therapy is administered. Moreover, evaluating whether a particular therapy has converted an unresectable tumor to a resectable one requires the use of reproducible imaging with adequate resolution during phase contrast administration before and after the therapy.31 Although many studies have reported that chemoradiation therapy can render at least some advanced tumors resectable (Table 22-2),98-110 fulfillment of the criteria discussed here is the exception rather than the rule. Even with loose criteria and inconsistent imaging, disease downstaging after 5-FU–based chemoradiation therapy is uncommon, occurring in only about 8% to 16% cases.98-104 The use of newer radiosensitizers such as gemcitabine and capecitabine in combination with biological agents and radiation therapy may result in increased local tumor response and possibly increased resectability in patients with locally advanced disease, but this has yet to be clearly demonstrated. Ideal­ ly, all studies involving the use of novel chemoradiation regimens should adhere to a strict CT-based definition of locally advanced pancreatic cancer that includes arterial involvement (e.g., low-density tumor inseparable from the superior mesenteric artery or celiac axis on contrast­enhanced CT) or occlusion of the superior mesenteric and portal venous confluence when the issue of converting an unresectable tumor to a resectable one is addressed.

Chemoradiation Therapy for Locally Advanced Disease Locally advanced pancreatic cancer usually is incurable, and all therapies have significant limitations. Early clinical trials established that the concurrent use of 5-FU with radiation therapy afforded superior median survival compared with radiation alone5,41 or chemotherapy alone,43 but these studies included small numbers of patients and may not be relevant in the context of newer chemotherapeutic agents and modern radiation therapy techniques. In reality, patients probably benefit modestly from systemic therapy and chemoradiation therapy; these approaches are complementary, and both should be considered for patients with locally advanced disease. Although no new phase III trials of chemoradiation therapy have been conducted during the past 20 years, the development of novel cytotoxic and targeted therapeutic agents during that time has stimulated the investigation of novel chemoradiation combinations in phase I and phase II trials. Because the impact of standard therapies is so limited, all patients with locally advanced pancreatic cancer should be considered for protocol-based therapy. Patients with locally advanced disease are usually eligible for clinical trials evaluating systemic therapy alone as well as trials evaluating novel chemoradiation regimens. If they refuse or are not eligible, patients should be given a relatively well-tolerated chemoradiation regimen, preceded or followed by systemic gemcitabine-based chemotherapy. The primary purpose of concurrent radiation­chemotherapy for several types of cancer (e.g., anal cancer, head and neck cancer, cervical cancer) is the enhancement of local disease control through radiosensitization. The ­ideal concurrent chemotherapeutic agent in the treatment of pancreatic cancer, in contrast, should have systemic ­activity as well as radiosensitizing properties. Because the

median survival time for patients with locally advanced disease usually is less than 1 year and the benefit of treatment is limited, the combination of agent and radiation should be well tolerated. Acute toxicity can be drastically reduced by confining the radiation fields to the gross primary tumor and clinically enlarged lymph nodes, regardless of which radiosensitizing agent is used. Treating uninvolved regional lymph nodes involves exposure of larger amounts of gastric and duodenal mucosa, which can lead to high rates of gastrointestinal toxicity and is not likely to improve median survival.

5-Fluorouracil–Based Trials Almost 20 years ago, EBRT and concomitant ­chemotherapy with 5-FU were shown in a landmark randomized trial conducted by the GITSG to prolong survival of patients with locally advanced adenocarcinoma of the pancreas.5,41 In that trial, use of a split course of radiation therapy with bolus 5-FU prolonged the median survival time over that obtained after radiation alone. However, no survival benefit was found for groups given concurrent 5-FU with highdose (60 Gy) radiation relative to those given low-dose (40 Gy) radiation.5,41 Because this study was conducted before CT scanning was available, patterns of failure could not be documented. Nevertheless, local disease was a component of the first clinical site of failure in more of the ­patients treated with 40 Gy than in those treated with 60 Gy. This suggestion of improved local control with the 60-Gy dose over that of the 40-Gy dose was offset by the problem of competing risk of distant failure. Somewhat different results were seen in an ECOG study44 in which patients were randomly assigned to receive 5-FU (600 mg/m2 in a weekly intravenous bolus) or radiation therapy (40 Gy) with concurrent 5-FU (600 mg/m2, days 1 to 3) followed by maintenance 5-FU (600 mg/m2). No difference was found between groups in ­median survival time, which could be explained by inadequacies in the chemoradiation dose or schedule or by the fact that 22% of patients did not follow the protocol or could not be evaluated.44 Nevertheless, additional randomized studies from the Mayo Clinic and the GITSG support the superiority of combined-modality therapy42,43,73 over ­radiation therapy or chemotherapy alone, as do other ­single-­institution ­studies.45 The randomized trial presented in abstract form in 2006 by the French Fondation Francophone de Cancérologie Digestive and Société Française de Radiothérapie ­Oncologique (FFCD-SSRO) is the only modern randomized trial to compare chemotherapy with chemoradiation therapy for locally advanced pancreatic cancer.111 Its findings have been a topic of debate. Patients were assigned to receive gemcitabine or concurrent 5-FU, cisplatin, and radiation followed by gemcitabine. The median survival time in the gemcitabine-only group was unusually long (14.3 versus 9.1 months in ECOG 6201, a contemporaneous randomized trial of chemotherapy alone112), and the median survival time in the chemoradiation group was unusually short (8.4 months versus a more typical 10 to 12 months). Median survival times of more than 10 months are rare in multi-institutional phase II or phase III trials of chemotherapy for locally advanced, unresectable pancreatic cancer. Most studies of chemotherapy for metastatic pancreatic cancer have included variable numbers of

TABLE 22-2.   Experience with Neoadjuvant Chemoradiation Therapy to Allow Surgical Resection in Patients with Locally Advanced Pancreatic Cancer

Chemotherapy

Number Showing Radiographic Response

Number Undergoing Surgical Resection

Percent with Margin-Negative Resections

Median ­ urvival (mo) S

55

5-FU + LV + cisplatin

4 CR 14 PR

3

8

14

50

5-DFUR

7 PR

5

16

9

8

NS

Number of Patients

EBRT Dose (Gy)

Kornek et al, 200098

38

Bajetta et al, 199999

32

Study and Reference 5-FU–Based Chemoradiation

25

45

5-FU + Mito (12) or ­cisplatin (10)

6 of 22 showed shrinkage of > 1 cm

5*

Todd et al, 1998101

38

0

5-FU + LV + Mito + ­dipyridamole

1 CR 14 PR

4 (1 CR†)

11

15.5

Kamthan et al, 1997102

35

54

5-FU + STZ + cisplatin

6 CR 9 PR

5 (2 CR†)

14

15

16

45

5-FU

NS

2

13

9.6‡

20

50 (split course)

5-FU

NS

2

10

10

White et al,

1999100

Jessup et al,

1993103

Jeekel and Treurniet- Donker, 1991104

(1

CR†)

Gemcitabine-Based Chemoradiation McGinn et al, 2001105

22

24-50.4

Gem

8

5

23

NS

2000106

24

55.8

Gem + cisplatin

NS

7 of 13

54

NS

20

50.4

Gem

PR, 4

2

10

12

40

NS

Brunner et al,

Epelbaum et al,

2000107

Wilkowski et al,

2000108

10

45

Gem + 5-FU

7

5§

40

50.4

Paclitaxel

PR: 11 of 38

3

8

8.5

24

50.4

Paclitaxel

PR: 6 of 18

2

8

NS

Paclitaxel-Based Chemoradiation DiPetrillo, 2000109 Safran et al,

of five had positive resection margins. complete response. ‡Mean. §One of five had positive resection margins. CR, complete response; 5-DFUR, doxifluridine; EBRT, external beam radiation therapy; 5-FU, 5-fluorouracil; Gem, gemcitabine; LV, leucovorin; Mito, mitomycin C; NS, not stated; PR, partial response; STZ, streptozocin. †Histologic

22  The Pancreas

*Three

1999110

527

528

PART 6  Gastrointestinal Tract

­ atients with locally advanced disease, and the 9.1 months p for patients with locally advanced disease in ECOG 6201 is more realistic. Most studies evaluating adjuvant therapy after curative resection report median survival rates of between 14 and 20 months. In the FFCD-SSRO study, the acute toxicity rates in the chemoradiation therapy group were also very high, and the resultant poor compliance undoubtedly contributed to the poor outcome in that group. Factors that probably contributed to the poor tolerance were the use of cisplatin, high-dose radiation (60 Gy), and regional nodal fields. As discussed under “Radiation Therapy Techniques,” radiation fields for locally advanced, unresectable disease should be confined to the gross tumor and enlarged lymph nodes; moreover, the standard concurrent chemotherapeutic agent should be infusional 5-FU or capecitabine. Another factor that could explain the results is a possible imbalance of patients who went on to undergo resection between the two treatment groups. These and other details are awaited in the full report of this trial.

Gemcitabine-Based Trials The introduction of gemcitabine was a modest step forward in the treatment of pancreatic cancer. As a nucleoside analogue, its value as a systemic agent in pancreatic cancer and the recognition of its radiosensitizing properties stimulated the study of combinations of gemcitabine with EBRT for patients with localized pancreatic cancer.87,105,113,114 Strategies investigated have included 7 weeks of weekly intravenous gemcitabine during short-course radiation therapy (30 Gy), weekly or twice-weekly gemcitabine with 50.4 Gy of radiation therapy, and full-dose weekly gemcitabine with escalating doses of radiation. Most of these studies indicated that gastrointestinal toxicity was a dose-limiting factor, but hematologic toxicity has also been observed. No standard approach has been established for combining gemcitabine and radiation, but several variables seem to be important in predicting the maximum tolerated dose, including the size of the radiation portal, the total radiation dose, possibly the dose of radiation per fraction, and whether gemcitabine is administered once or twice weekly.115 Three multi-institutional studies of gemcitabine-based chemoradiation therapy have been completed. A small study performed in Taiwan116 involved randomly assigning 34 patients with locally advanced pancreatic cancer to receive 5-FU–based chemoradiation therapy (500 mg/m2 daily for 3 days every 14 days in combination with radiation to a total dose of 50.4 to 61.2 Gy) or gemcitabine and radiation (600 mg/m2 weekly with equivalent doses of radiation). The objective response rates were 50% for gemcitabine and radiation therapy and only 13% for 5-FU chemoradiation therapy. Median survival time was also substantially better in the gemcitabine group (14.5 versus 6.7 months for the 5-FU group, P = .027). However, these efficacy results must be interpreted cautiously because of the limited accrual (34 patients) and the poor results in the control group. Although the investigators in this trial concluded that the gemcitabine-based treatment was no more toxic than the 5-FU–based treatment, therapy was actually poorly tolerated in both groups. Only 75% of patients were able to complete the full dose of radiation therapy, and roughly one third of the patients in both groups were hospitalized for 2 to 6 weeks because of acute toxicity.

A phase II study of locally advanced pancreatic cancer conducted by the Cancer and Leukemia Group B ­(CALGB)117 found that gemcitabine given at 40 mg/m2 twice weekly produced high rates of grade 3 or 4 gastrointestinal toxicity (35%) and grade 3 or 4 hematologic toxi­ city (50%) and a median survival time of only 8.5 months. Not sur­prisingly, the CALGB abandoned that approach for treating locally advanced pancreatic cancer. Both studies ­included irradiation of regional nodal fields, which likely contributed to the significant gastrointestinal toxicity. In ­contrast, an approach developed at the University of Michigan105 delivers the manufacturer’s recommended dose of gemcitabine (1000 mg/m2) and a slightly lower radiation dose (36 Gy in 15 fractions over 3 weeks), with con­formal ­radiation fields encompassing the gross tumor volume alone. That approach was well tolerated, presumably because a smaller volume of normal tissue was ­irradiated.105 Investigators have since embarked on a multi-institutional phase II study evaluating the same regimen. Early results with 39 treated patients indicate that approximately 26% ­experienced grade 3 or 4 nonhematologic toxicity.118 Several points about gemcitabine-based chemoradiation therapy are worth emphasizing. Although of some value as a systemic agent,119 gemcitabine is probably only modestly better than 5-FU when used in combination with radiation therapy, but it is not tolerated as well.120 The gastrointestinal toxicity reported in the three multi-institutional studies of gemcitabine demonstrate that the therapeutic ratio for the combination of gemcitabine with radiation is narrow. Typically, the radiation dose or the gemcitabine dose must be attenuated if the combination is to be given safely. Compared with radiation therapy fields that target only the gross tumor, elective regional nodal irradiation results in increased gastrointestinal toxicity. If gemcitabine is used in combination with irradiation of esophageal, gastric, or duodenal mucosa, the volume of mucosa being treated should be minimized to minimize the significant risk of severe acute toxicity. Although it is reasonable for investigators and clinicians with experience using any of these regimens to use them, a dose and schedule that they find to be well tolerated should be used, particularly if the goal is to build on the results of these and other studies of novel chemotherapeutic agents, such as those that target tumor-specific molecular pathways.

Paclitaxel-Based Trials Paclitaxel, like gemcitabine, has radiosensitizing properties as well as systemic activity in pancreatic cancer. In one phase I study of paclitaxel and radiation that included ­patients with locally advanced pancreatic cancer,56 the dose-limiting toxicities were related to gastrointestinal effects in the radiation field (nausea, anorexia, and ­abdominal pain). At the dose recommended for further study, therapy was well tolerated. Subsequently, a phase II study conducted by the RTOG121 revealed a median survival time of 11.3 months, which compares favorably with that of historical control subjects but came at the cost of significant but manageable acute toxicity. In an effort to build on this regimen, the RTOG PA-0020 trial, which ­recently reached full ­accrual, is testing gemcitabine in combination with ­paclitaxel.122 Findings from a phase II study at M. D. ­Anderson of neoadjuvant chemoradiation therapy for ­patients with potentially resectable tumors showed that

22  The Pancreas

­ aclitaxel ­resulted in increased acute toxicity but histologic p responses similar to those in historical controls.85 These findings led to the abandonment of paclitaxel in favor of concurrent gemcitabine-based chemoradiation therapy. In summary, concurrent paclitaxel and radiation therapy can cause significant acute treatment-related morbidity. However, targeting only the gross tumor rather than the regional lymphatics is likely to improve the tolerability of this treatment.

Capecitabine-Based Trials The efficacy of capecitabine seems similar to that of intravenously administered 5-FU,123 and capecitabine therefore is an appropriate substitute for infusional or bolus 5-FU when used with radiation therapy. An orally administered agent, capecitabine has a clinical benefit similar to that of gemcitabine for patients with locally advanced or metastatic pancreatic cancer,124 and unlike gemcitabine, it has a favorable toxicity profile when given at systemic doses with regional nodal irradiation. A phase I trial showed 800 mg/m2 to be the recommended dose when capecitabine was given only on days of radiation.125 Another phase I trial conducted at M. D. Anderson126 indicated that the combination of bevacizumab (an inhibitor of vascular endothelial growth factor), capecitabine, and radiation therapy was well tolerated in that only 2 of 47 patients developed grade 3 gastrointestinal toxicity, and no patient experienced significant hematologic toxicity.

Novel Molecular Targeted Treatment Approaches Oncology has entered an era of molecular therapy, and molecular agents are increasingly being incorporated into anticancer therapy. Agents such as bevacizumab and cetuximab have begun to change the standard of care for patients with advanced gastrointestinal tumors, most notably colon cancer, and erlotinib has shown some benefit in pancreatic cancer.127-129 Most investigations of these and other drugs have focused on their benefits as components of systemic therapy for advanced disease, and their usefulness as adjuvant therapy is being studied in colon cancer. These molecular agents may also have important roles as radiosensitizers. Bevacizumab has been combined with gemcitabine for the treatment of advanced pancreatic cancer, and the drug’s radiosensitizing properties have become appreciated clinically. The possible mechanisms of radiosensitization are not clear but may include enhancing the lethality of ionizing radiation to endothelial cells130 or tumor cells131 or inducing vascular normalization, thereby alleviating tumor hypoxia.132 The first clinical evidence of the radiosensitizing effects of bevacizumab was reported by Willett and colleagues,133,134 who found that administering bevacizumab before the initiation of radiation to 11 patients with advanced rectal cancer led to rapid changes in tumor perfusion, interstitial intratumor pressure, and numbers of circulating endothelial cells. In a phase I dose escalation study, we and others at M. D. Anderson135 found that the addition of bevacizumab to capecitabine and radiation (50.4 Gy) did not enhance acute toxicity among 47 patients with locally advanced pancreatic cancer. Acute toxic effects were minimal and easily managed by adjusting the capecitabine dose, without interruption or attenuation of

529

the bevacizumab or radiation. Twenty of the 47 patients experienced grade 2 gastrointestinal toxicity, and 11 developed grade 2 hand-foot syndrome; however, reducing the oral capecitabine dose partway through the trial successfully reduced the incidence of grade 2 toxicity and was sufficient to avoid grade 3 toxicity in most cases. Limiting the radiation therapy fields to the gross tumor volume probably contributed to minimizing the incidence of grade 3 acute toxicity to 4.3%. Although bevacizumab did not seem to enhance acute toxicity, patients with tumors that had invaded the duodenum seemed to be at higher risk for bleeding or perforation. (There were no bleeding events in the final 18 patients accrued after a protocol modification excluded patients with duodenal invasion.) The tumors in 9 (20%) of 46 evaluable patients showed an objective partial response to initial therapy, including 6 of 12 tumors treated to a bevacizumab dose of 5 mg/kg.135 Four patients experienced ulceration with bleeding or perforation in the radiation field; these events occurred between 3 and 20 weeks after the completion of chemoradiation therapy and were observed at each bevacizumab dose level. Because tumor-associated duodenal bleeding can occur in patients with locally advanced disease treated with chemoradiation therapy alone,136 it is not certain that these events were directly related to the addition of bevacizumab to the chemoradiation therapy. This potential risk should be assessed within the context of the possible improvement in local tumor control, as evidenced by radiographic tumor response in 6 patients and durable stabilization of local disease in 28 patients. Four patients in this study were able to undergo pancreaticoduodenectomy between 4 and 20 weeks after the last dose of bevacizumab without perioperative complications.135 Other trials evaluating capecitabine-based chemoradiation therapy with bevacizumab are being conducted by ECOG (trial 2204), the American College of Surgeons Oncology Group (Z5031 and Z5041), and the CALGB (80303). Preliminary toxicity analysis in the ECOG trial revealed low rates of gastrointestinal toxicity, with only one episode of a grade 3 or higher bleeding event among 82 patients accrued; however, that study specifically excluded patients with tumor invasion of the duodenum. The Z5041 trial was proposed as a phase II neoadjuvant study for patients with radiographically resectable tumors that will evaluate concurrent bevacizumab and gemcitabine before pancreaticoduodenectomy and capecitabine, radiation therapy, and bevacizumab after pancreaticoduodenectomy. The recommended dose of bevacizumab for further study is 5 mg/kg every 2 weeks with radiation therapy (50.4 Gy in 28 fractions) and concurrent capecitabine (825 mg/m2 twice daily Monday through Friday). The phase III 80303 trial is evaluating a 10 mg/kg bevacizumab dose with gemcitabine against gemcitabine alone for patients with metastatic disease. Inhibitors of the epidermal growth factor receptor (EGFR) hold promise as radiosensitizers, although none of the available inhibitors (gefitinib, erlotinib, or cetuximab) has been evaluated in multi-institutional trials in combination with radiation for locally advanced pancreatic cancer. A phase I dose escalation study at Brown University tested a combination of gemcitabine (75 mg/m2), paclitaxel (40 mg/m2), and weekly or daily erlotinib with radiation (50.4 Gy to the primary tumor and regional lymphatics)

530

PART 6  Gastrointestinal Tract

for 13 patients with locally advanced pancreatic cancer. The maximum tolerated dose of erlotinib was found to be 50 mg/m2. The median survival time was 14.0 months, and six patients (46%) experienced a partial response, indicating that erlotinib-based chemoradiation regimens may be worthy of further study.127 The intolerance of erlotinib at doses higher than 75 mg/day during chemoradiation ­therapy might have been caused by the concurrent gemcitabine and paclitaxel and by the use of regional nodal irradiation. Another phase I study at Memorial Sloan­Kettering is evaluating gemcitabine-based chemotherapy in combination with erlotinib. Cetuximab, another EGFR inhibitor, has been shown to improve local tumor control and overall survival in combination with EBRT alone for locally advanced head and neck cancer.137 That study constitutes the first phase III evidence that adding EGFR inhibition to radiation therapy can improve outcome. Cooperative group trials are planned to evaluate cetuximab and bevacizumab for their radiosensitizing properties in adjuvant (see Table 22-1) and neoadjuvant therapy (see Table 22-2). Another study in development by the Southwest Oncology Group (S0527) involves preoperative gemcitabine, oxaliplatin, and radiation for resectable pancreatic cancer. A phase II study at M. D. Anderson is evaluating gemcitabine, oxaliplatin, and cetuximab, followed by capecitabine, radiation therapy (50.4 Gy) and cetuximab, to be followed by gemcitabine and cetuximab until disease progression. Other trials that combine cetuximab or erlotinib with bevacizumab and EBRT for the treatment of pancreatic cancer are anticipated in the near future, as are studies that combine bevacizumab with gemcitabine and radiation therapy.

Prophylactic Hepatic Irradiation Two single institutions and the RTOG have prospectively evaluated prophylactic hepatic irradiation. At M. D. Anderson, a phase I trial of preoperative radiation therapy for patients with resectable disease used 50.4 Gy to the pancreatic head and regional lymphatics and 23.4 Gy to the whole liver with the concurrent continuous infusion of 5-FU (300 mg/m2/day, Monday to Friday). That study was abandoned after two early treatment-related deaths among 11 patients.138 At Johns Hopkins University, a prospective study of patients who had undergone resection evaluated 50.4 to 57.6 Gy of EBRT to the operative bed and regional lymphatics and 23.4 to 27.0 Gy to the whole liver with concurrent continuous-infusion 5-FU (200 mg/m2/day) and leucovorin (5 mg/m2) given Monday through Friday. That treatment was followed by four cycles of the same chemotherapy 1 month after completion of the radiation. Although the toxicity of this regimen was judged acceptable, the site of first failure was the liver in 9 of the 21 patients so treated.78 In a study by the RTOG of localized unresectable pancreatic cancer, 61.2 Gy was delivered to the primary disease site and 23.4 Gy to the whole liver with a concurrent 5-day continuous infusion of 5-FU (1000 mg/m2/day) on days 1 and 30, followed by a weekly bolus of 600 mg/m2 for 6 months beginning 3 weeks after the completion of therapy. More than one half of the patients had grade 3 or worse acute reactions, and the median survival time was

only 8.4 months. The first site of failure was the liver in 13% of the patients, which is a lower rate than what would be expected from historical controls; however, survival remained poor because of local failure and intra-abdominal metastasis.9

REFERENCES    1. Coia LR, Myerson RJ, Tepper JE. Late effects of radiation therapy on the gastrointestinal tract. Int J Radiat Oncol Biol Phys 1995;31:1213-1236.    2. Goldgraber M, Rubin C, Palmer W, et al. The early gastric response to irradiation, a serial biopsy study. Gastroenterology 1954;27:1-20.    3. Hamilton F. Gastric ulcer following radiation. Arch Surg 1947;55:394-399.    4. Roswitt B, Malsky S, Reid C. Radiation tolerance of the gastrointestinal tract. In Vaeth J (ed): Frontiers of Radiation Oncology and Therapy. Baltimore: University Park Press, 1972, pp 160-181.    5. Moertel CG, Frytak S, Hahn RG, 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;48:17051710.    6. Dawson LA, McGinn CJ, Normolle D, et al. Escalated focal liver radiation and concurrent hepatic artery fluorodeoxyuridine for unresectable intrahepatic malignancies. J Clin Oncol 2000;18:2210-2218.    7. Willett CG, Tepper JE, Orlow EL, et al. Renal complications secondary to radiation treatment of upper abdominal malignancies. Int J Radiat Oncol Biol Phys 1986;12:1601-1604.    8. Lawrence TS, Robertson JM, Anscher MS, et al. Hepatic toxicity resulting from cancer treatment. Int J Radiat Oncol Biol Phys 1995;31:1237-1248.    9. Komaki R, Wadler S, Peters T, et al. High-dose local irradiation plus prophylactic hepatic irradiation and chemotherapy for inoperable adenocarcinoma of the pancreas. A preliminary report of a multi-institutional trial (Radiation Therapy Oncology Group Protocol 8801). Cancer 1992;69:2807-2812.   10. Ingold J, Reed G, Kaplan H, et al. Radiation hepatitis. AJR Am J Roentgenol 1965;93:200-208.   11. Dawson LA, Normolle D, Balter JM, et al. Analysis of radiationinduced liver disease using the Lyman NTCP model. Int J Radiat Oncol Biol Phys 2002;53:810-821.   12. Lawrence TS, Ten Haken RK, Kessler ML, et al. The use of 3-D dose volume analysis to predict radiation hepatitis. Int J Radiat Oncol Biol Phys 1992;23:781-788.   13. Dawson LA, McGinn CJ, Lawrence TS. Conformal chemoradiation for primary and metastatic liver malignancies. Semin Surg Oncol 2003;21:249-255.   14. American Cancer Society. Cancer Facts & Figures 2008. Atlanta: American Cancer Society, 2008.   15. Gordis L, Gold EB. Epidemiology of pancreatic cancer. World J Surg 1984;8:808-821.   16. Ahlgren JD. Epidemiology and risk factors in pancreatic cancer. Semin Oncol 1996;23:241-250.   17. Howe GR. Pancreatic cancer. Cancer Surv 1994;19-20:139158.   18. Best EW, Walker CB, Baker PM, et al. Summary of a Canadian study of smoking and health. Can Med Assoc J 1967;96:11041108.   19. Whittemore AS, Paffenbarger RS Jr, Anderson K, et al. Early precursors of pancreatic cancer in college men. J Chronic Dis 1983;36:251-256.   20. Velema JP, Walker AM, Gold EB. Alcohol and pancreatic cancer. Insufficient epidemiologic evidence for a causal relationship. Epidemiol Rev 1986;8:28-41.   21. Gold EB. Epidemiology of and risk factors for pancreatic cancer. Surg Clin North Am 1995;75:819-843.

22  The Pancreas   22. Howe GR, Burch JD. Nutrition and pancreatic cancer. Cancer Causes Control 1996;7:69-82.   23. Garabrant DH, Held J, Langholz B, et al. DDT and related compounds and risk of pancreatic cancer. J Natl Cancer Inst 1992;84:764-771.   24. Mancuso TF, el-Attar AA. Cohort study of workers exposed to betanaphthylamine and benzidine. J Occup Med 1967;9:277285.   25. Krause JR, Ayuyang HQ, Ellis LD. Secondary non-hematopoietic cancers arising following treatment of hematopoietic disorders. Cancer 1985;55:512-515.   26. Everhart J, Wright D. Diabetes mellitus as a risk factor for pancreatic cancer. A meta-analysis. JAMA 1995;273:1605-1609.   27. Gullo L, Pezzilli R, Morselli-Labate AM. Diabetes and the risk of pancreatic cancer. Italian Pancreatic Cancer Study Group. N Engl J Med 1994;331:81-84.   28. Gold EB, Cameron JL. Chronic pancreatitis and pancreatic cancer. N Engl J Med 1993;328:1485-1486.   29. Fernandez E, La Vecchia C, D’Avanzo B, et al. Family history and the risk of liver, gallbladder, and pancreatic cancer. Cancer Epidemiol Biomarkers Prev 1994;3:209-212.   30. Hruban RH, Petersen GM, Goggins M, et al. Familial pancreatic cancer. Ann Oncol 1999;10(Suppl 4):69-73.   31. Tamm EP, Silverman PM, Charnsangavej C, et al. Diagnosis, staging, and surveillance of pancreatic cancer. AJR Am J Roentgenol 2003;180:1311-1323.   32. Ichikawa T, Haradome H, Hachiya J, et al. Pancreatic ductal adenocarcinoma: preoperative assessment with helical CT versus dynamic MR imaging. Radiology 1997;202:655-662.   33. Ferrari AP Jr, Lichtenstein DR, Slivka A, et al. Brush cytology during ERCP for the diagnosis of biliary and pancreatic malignancies. Gastrointest Endosc 1994;40:140-145.   34. Rosch T, Lorenz R, Braig C, et al. Endoscopic ultrasonography in diagnosis and staging of pancreatic and biliary tumors. Endoscopy 1992;24(Suppl 1):304-308.   35. Shin HJ, Lahoti S, Sneige N. Endoscopic ultrasound-guided fine-needle aspiration in 179 cases: the M. D. Anderson Cancer Center experience. Cancer 2002;96:174-180.   36. Ritts RE Jr, Nagorney DM, Jacobsen DJ, et al. Comparison of ­preoperative serum CA19-9 levels with results of diagnostic ­imaging modalities in patients undergoing laparotomy for suspected pancreatic or gallbladder disease. Pancreas 1994;9:707-716.   37. Whipple A, Parson W, Mullin C. Treatment of carcinoma of the ampulla of Vater. Ann Surg 1935:763-779.   38. Makary MA, Warshaw AL, Centeno BA, et al. Implications of peritoneal cytology for pancreatic cancer management. Arch Surg 1998;133:361-365.   39. Willett CG, Lewandrowski K, Warshaw AL, et al. Resection margins in carcinoma of the head of the pancreas. Implications for radiation therapy. Ann Surg 1993;217:144-148.   40. Yeo CJ, Cameron JL, Lillemoe KD, et al. Pancreaticoduodenectomy for cancer of the head of the pancreas. 201 patients. Ann Surg 1995;221:721-731.   41. A multi-institutional comparative trial of radiation therapy alone and in combination with 5-fluorouracil for locally unresectable pancreatic carcinoma. The Gastrointestinal Tumor Study Group. Ann Surg 1979;189:205-208.   42. Radiation therapy combined with Adriamycin or 5-fluorouracil for the treatment of locally unresectable pancreatic carcinoma. Gastrointestinal Tumor Study Group. Cancer 1985;56:25632568.   43. Treatment of locally unresectable carcinoma of the pancreas: comparison of combined-modality therapy (chemotherapy plus radiotherapy) to chemotherapy alone. Gastrointestinal Tumor Study Group. J Natl Cancer Inst 1988;80:751-755.   44. Klaassen DJ, MacIntyre JM, Catton GE, et al. Treatment of locally unresectable cancer of the stomach and pancreas: a randomized comparison of 5-fluorouracil alone with radiation plus concurrent and maintenance 5-fluorouracil—an Eastern Cooperative Oncology Group study. J Clin Oncol 1985;3:373-378.   45. Mohiuddin M, Cantor RJ, Biermann W, et al Combined modality treatment of localized unresectable adenocarcinoma of the pancreas. Int J Radiat Oncol Biol Phys 1988;14:79-84.

531

  46. Trede M. The surgical treatment of pancreatic carcinoma. Surgery 1985;97:28-35.   47. Warshaw AL, Fernandez-del Castillo C. Pancreatic carcinoma. N Engl J Med 1992;326:455-465.   48. Yeo CJ, Abrams RA, Grochow LB, et al. Pancreaticoduodenectomy for pancreatic adenocarcinoma: postoperative adjuvant chemoradiation improves survival. A prospective, single-institution ­experience. Ann Surg 1997;225:621-633; discussion 633–636.   49. Tseng JF, Raut CP, Lee JE, et al. Pancreaticoduodenectomy with vascular resection: margin status and survival duration. J Gastrointest Surg 2004;8:935-950.   50. Greene FL, Page DL, Fleming ID et al, eds. AJCC Cancer Staging Manual. 6th ed. New York: Springer, 2002, pp 157-164.   51. Birk D, Fortnagel G, Formentini A, et al. Small carcinoma of the pancreas. Factors of prognostic relevance. J Hepatobiliary Pancreat Surg 1998;5:450-454.   52. Koong AC, Christofferson E, Le QT, et al. Phase II study to assess the efficacy of conventionally fractionated radiotherapy followed by a stereotactic radiosurgery boost in patients with locally advanced pancreatic cancer. Int J Radiat Oncol Biol Phys 2005;63:320-323.   53. Ozaki H, Hiraoka T, Mizumoto R, et al. The prognostic significance of lymph node metastasis and intrapancreatic perineural invasion in pancreatic cancer after curative resection. Surg Today 1999;29:16-22.   54. Klinkenbijl JH, Jeekel J, Sahmoud T, 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. Ann Surg 1999;230:776-782; discussion 782–784.   55. Kalser MH, Ellenberg SS. for the Gastrointestinal Tumor Study Group. Pancreatic cancer: adjuvant combined radiation and chemotherapy following curative resection. Arch Surg 1985;120:899-903.   56. Gastrointestinal Tumor Study Group. Further evidence of effective adjuvant combined radiation and chemotherapy following curative resection of pancreatic cancer. Cancer 1987;59:20062010.   57. Neoptolemos JP, Stocken DD, Friess H, et al. A randomized trial of chemoradiotherapy and chemotherapy after resection of pancreatic cancer. N Engl J Med 2004;350:1200-1210.   58. Griffin JF, Smalley SR, Jewell W, et al. Patterns of failure after curative resection of pancreatic carcinoma. Cancer 1990;66: 56-61.   59. Bussels B, Goethals L, Feron M, et al. Respiration-induced movement of the upper abdominal organs: a pitfall for the three­dimensional conformal radiation treatment of pancreatic cancer. Radiother Oncol 2003;68:69-74.   60. Balter JM, Lam KL, McGinn CJ, et al. Improvement of CT-based treatment-planning models of abdominal targets using static exhale imaging. Int J Radiat Oncol Biol Phys 1998;41:939-943.   61. Crane CH, Antolak JA, Rosen II, et al. Phase I study of concomitant gemcitabine and IMRT for patients with unresectable adenocarcinoma of the pancreatic head. Int J Gastrointest Cancer 2001;30:123-132.   62. Koong AC, Le QT, Ho A, et al. Phase I study of stereotactic radiosurgery in patients with locally advanced pancreatic cancer. Int J Radiat Oncol Biol Phys 2004;58:1017-1021.   63. Koong AC, Christofferson E, Le QT, et al. Phase II study to assess the efficacy of conventionally fractionated radiotherapy followed by a stereotactic radiosurgery boost in patients with locally advanced pancreatic cancer. Int J Radiat Oncol Biol Phys 2005;63:320-323.   64. Hoyer M, Roed H, Sengelov L, et al. Phase II study on stereotactic radiotherapy of locally advanced pancreatic carcinoma. Radiother Oncol 2005;76:48-53.   65. Garton GR, Gunderson LL, Nagorney DM, et al. High-dose preoperative external beam and intraoperative irradiation for locally advanced pancreatic cancer. Int J Radiat Oncol Biol Phys 1993;27:1153-1157.   66. Gunderson LL, Martin JK, Kvols LK, et al. Intraoperative and external beam irradiation +/- 5-FU for locally advanced pancreatic cancer. Int J Radiat Oncol Biol Phys 1987;13:319-329.

532

PART 6  Gastrointestinal Tract

  67. Mohiuddin M, Regine WF, Stevens J, et al. Combined intraoperative radiation and perioperative chemotherapy for unresectable cancers of the pancreas. J Clin Oncol 1995;13:2764-2768.   68. Roldan GE, Gunderson LL, Nagorney DM, et al. External beam versus intraoperative and external beam irradiation for locally advanced pancreatic cancer. Cancer 1988;61:1110-1116.   69. Shipley WU, Tepper JE, Warshaw AL, et al. Intraoperative radiation therapy for patients with pancreatic carcinoma. World J Surg 1984;8:929-934.   70. Mohiuddin M, Rosato F, Barbot D, et al. Long-term results of combined modality treatment with I-125 implantation for carcinoma of the pancreas. Int J Radiat Oncol Biol Phys 1992;23:305311.   71. Shipley WU, Nardi GL, Cohen AM, et al. Iodine-125 implant and external beam irradiation in patients with localized pancreatic carcinoma: a comparative study to surgical resection. Cancer 1980;45:709-714.   72. Willett CG, Del Castillo CF, Shih HA, et al. Long-term results of intraoperative electron beam irradiation (IOERT) for patients with unresectable pancreatic cancer. Ann Surg 2005;241:295-299.   73. Moertel CG, Childs DS Jr, Reitemeier RJ, et al. Combined 5fluorouracil and supervoltage radiation therapy of locally unresectable gastrointestinal cancer. Lancet 1969;2:865-867.   74. Foo ML, Gunderson LL, Nagorney DM, et al. Patterns of failure in grossly resected pancreatic ductal adenocarcinoma treated with adjuvant irradiation +/− 5 fluorouracil. Int J Radiat Oncol Biol Phys 1993;26:483-489.   75. Yeo CJ, Abrams RA, Grochow LB, et al. Pancreaticoduodenectomy for pancreatic adenocarcinoma: postoperative ­ adjuvant chemoradiation improves survival: a prospective, single­institution experience. Ann Surg 1997;225:621-633; discussion 633–636.   76. Regine WF, Winter KA, Abrams RA, et al. Fluorouracil vs gemcitabine chemotherapy before and after fluorouracil-based chemoradiation following resection of pancreatic adenocarcinoma: a randomized controlled trial. JAMA 2008;299:1019-1026.   77. Oettle H, Post S, Neuhaus P, et al. Adjuvant chemotherapy with gemcitabine vs observation in patients undergoing curative­intent resection of pancreatic cancer: a randomized controlled trial. JAMA 2007;297:267-277.   78. Abrams RA, Grochow LB, Chakravarthy A, et al. Intensified adjuvant therapy for pancreatic and periampullary adenocarcinoma: survival results and observations regarding patterns of failure, radiotherapy dose and CA19-9 levels. Int J Radiat Oncol Biol Phys 1999;44:1039-1046.   79. Neoptolemos JP, Dunn JA, Stocken DD, et al. Adjuvant chemoradiotherapy and chemotherapy in resectable pancreatic cancer: a randomised controlled trial. Lancet 2002;358:1576-1585.   80. Picozzi VJ, Pisters PWT, Vickers SM, et al. Strength of the evidence: adjuvant therapy for resected pancreatic cancer. J Gastrointest Surg 2008;12:657-661.   81. Pisters PW, Wolff RA, Crane CH, et al. Combined-modality treatment for operable pancreatic adenocarcinoma. Oncology 2005;19:393-404.   82. Evans DB, Rich TA, Byrd DR, et al. Preoperative chemoradiation and pancreaticoduodenectomy for adenocarcinoma of the pancreas. Arch Surg 1992;127:1335-1339.   83. Crane CH, Varadhachary G, Wolff RA, Pisters PWT, Evans DB. The argument for pre-operative chemoradiation for localized, radiographically resectable pancreatic cancer. Best Pract Res Clin Gastroenterol 2006;20:365-382.   84. Pisters PWT, Abbruzzese JL, Janjan NA, et al. Rapid-­fractionation preoperative chemoradiation, pancreaticoduodenectomy, and intraoperative radiation therapy for resectable pancreatic adenocarcinoma. J Clin Oncol 1998;16:3843-3850.   85. Pisters PWT, Wolff RA, Janjan NA, et al. Preoperative paclitaxel and concurrent rapid-fractionation radiation for resectable pancreatic adenocarcinoma: toxicities, histologic response rates, and event-free outcome. J Clin Oncol 2002;20:2537-2544.   86. Wolff RA, Evans DB, Crane CH, et al. Initial results of a preoperative gemcitabine (GEM)-based chemoradiation for resectable pancreatic adenocarcinoma [abstract]. Proc Am Soc Clin Oncol 2002;21:130a.

  87. Wolff RA, Evans DB, Gravel DM, et al. Phase I trial of gemcitabine combined with radiation for the treatment of locally advanced pancreatic adenocarcinoma. Clin Cancer Res 2001;7:2246-2253.   88. Rich TA, Janjan NA, Abbruzzese JL, et al. Preoperative and postoperative chemoradiation strategies in patients treated with pancreaticoduodenectomy for adenocarcinoma of the pancreas [response]. J Clin Oncol 1997;15:3292-3293.   89. Pisters PWT, Hudec WA, Hess KR, et al. Effect of preoperative biliary decompression on pancreaticoduodenectomy-­associated morbidity in 300 consecutive patients. Ann Surg 2001;234: 47-55.   90. Pisters PWT, Hudec WA, Lee JE, et al. Preoperative chemoradiation for patients with pancreatic cancer: toxicity of endobiliary stents. J Clin Oncol 2000;18:860-867.   91. Faigel DO, Ginsberg GG, Bentz JS, et al. Endoscopic ultrasound-guided real-time fine-needle aspiration biopsy of the pancreas in cancer patients with pancreatic lesions. J Clin Oncol 1997;15:1439-1443.   92. Suits J, Frazee R, Erickson RA. Endoscopic ultrasound and fine needle aspiration for the valuation of pancreatic masses. Arch Surg 1999;134:639-643.   93. Harewood GC, Wiersema LM, Halling AC, et al. Influence of EUS training and pathology interpretation on accuracy of EUSguided fine needle aspiration of pancreatic masses. Gastrointest Endosc 2002;55:669-673.   94. Povoski SP, Karpeh MS, Conlon KC et al. Association of preoperative biliary drainage with postoperative outcome following pancreaticoduodenectomy. Ann Surg 1999;230:131-142.   95. Hoffman JP, Lipsitz S, Pisansky T, et al. Phase II trial of preoperative radiation therapy and chemotherapy for patients with localized, resectable adenocarcinoma of the pancreas: an Eastern Cooperative Oncology Group Study. J Clin Oncol 1998;16:317323.   96. Mullen JT, Lee JH, Gomez HF, et al. Pancreaticoduodenectomy after placement of endobiliary metal stents. J Gastrointest Surg 2005;9:1094-1104; discussion 1104–1105.   97. Crane CH, Janjan NA, Evans DB, et al. Toxicity and efficacy of concurrent gemcitabine and radiotherapy for locally advanced pancreatic cancer. Int J Pancreatol 2001;29:9-18.   98. Kornek GV, Schratter-Sehn A, Marczell A, et al. Treatment of unresectable, locally advanced pancreatic adenocarcinoma with combined radiochemotherapy with 5-fluorouracil, leucovorin and cisplatin. Br J Cancer 2000;82:98-103.   99. Bajetta E, Di Bartolomeo M, Stani SC, et al. Chemoradiotherapy as preoperative treatment in locally advanced unresectable pancreatic cancer patients: results of a feasibility study. Int J Radiat Oncol Biol Phys 1999;45:285-289. 100. White R, Lee C, Anscher M, et al. Preoperative chemoradiation for patients with locally advanced adenocarcinoma of the pancreas. Ann Surg Oncol 1999;6:38-45. 101. Todd KE, Gloor B, Lane JS, et al. Resection of locally advanced pancreatic cancer after downstaging with continuous-infusion 5fluorouracil, mitomycin-C, leucovorin, and dipyridamole. J Gastrointest Surg 1998;2:159-166. 102. Kamthan AG, Morris JC, Dalton J, et al. Combined modality therapy for stage II and stage III pancreatic carcinoma. J Clin Oncol 1997;15:2920-2927. 103. Jessup JM, Steele G Jr, Mayer RJ, et al. Neoadjuvant therapy for unresectable pancreatic adenocarcinoma. Arch Surg 1993;128:559-564. 104. Jeekel J, Treurniet-Donker AD. Treatment perspectives in locally advanced unresectable pancreatic cancer. Br J Surg 1991;78:1332-1334. 105. McGinn CJ, Zalupski MM, Shureiqi I, et al. Phase I trial of radiation dose escalation with concurrent weekly full-dose gemcitabine in patients with advanced pancreatic cancer. J Clin Oncol 2001;19:4202-4208. 106. Brunner T, Grabenbauer G, Kasti S, et al. A phase I trial of simultaneous gemcitabine/cisplatin and radiotherapy for patients with locally advanced pancreatic cancer [abstract]. Proc Am Soc Clin Oncol 2000;19:1109a.

22  The Pancreas 107. Epelbaum R, Rosenblatt E, Nasrallah S, et al. Phase II study of gemcitabine combined with radiation therapy in localized, unresectable pancreatic cancer [abstract]. Proc Am Soc Clin Oncol 2000;19:1029a. 108. Wilkowski R, Heinemann V, Rau H. Radiochemotherapy including gemcitabine and 5-fluorouracil for treatment of locally advanced pancreatic cancer [abstract]. Proc Am Soc Clin Oncol 2000;19:1078a. 109. DiPetrillo T. Paclitaxel and concurrent radiation for locally advanced pancreatic cancer [abstract 1152]. Proc Am Soc Clin Oncol 2000;19:294a. 110. Safran H, Akerman P, Cioffi W, et al. Paclitaxel and concurrent radiation therapy for locally advanced adenocarcinomas of the pancreas, stomach, and gastroesophageal junction. Semin Radiat Oncol 1999;9:53-57. 111. Chauffert B, Mornex F, Bonnetain F, et al. Phase III trial comparing initial chemoradiotherapy (intermittent cisplatin and infusional 5-FU) followed by gemcitabine vs. gemcitabine alone in patients with locally advanced non metastatic pancreatic cancer: an FFCD-SFRO study [abstract]. Proc Am Soc Clin Oncol 2006;24:4008a. 112. Poplin E, Levy D, Berlin J, et al. Phase III trial of gemcitabine (30-minute infusion) versus gemcitabine (fixed-dose-rate ­infusion [FDR]) versus gemcitabine + oxaliplatin (GEMOX) in ­patients with advanced pancreatic cancer (E6201) [abstract]. Proc Am Soc Clin Oncol 2006;24:LBA4004. 113. Pipas JM, Mitchell SE, Barth RJ, et al. Phase I study of twice-weekly gemcitabine and concomitant external-beam ­radiotherapy in patients with adenocarcinoma of the pancreas. Int J Radiat Oncol Biol Phys 2001;50:1317-1322. 114. Blackstock AW, Bernard SA, Richards F, et al. Phase I trial of twice-weekly gemcitabine and concurrent radiation in patients with advanced pancreatic cancer. J Clin Oncol 1999;17:22082212. 115. Crane CH, Wolff RA, Abbruzzese JL, et al: Combining gemcitabine with radiation in pancreatic cancer: understanding important variables influencing the therapeutic index. Semin Oncol 2001;28:25-33. 116. Li CP, Chao Y, Chi KH, et al. Concurrent chemoradiotherapy treatment of locally advanced pancreatic cancer: gemcitabine versus 5-fluorouracil, a randomized controlled study. Int J Radiat Oncol Biol Phys 2003;57:98-104. 117. Blackstock A, Tempero M, Niedwiecki D, et al. Cancer and Leukemia Group B (CALGB) 89805: phase II chemoradiation trial using gemcitabine in patients with localoregional adenocarcinoma of the pancreas [abstract 49]. Int J Radiat Oncol Biol Phys 2001;51:31. 118. Small W Jr, Berlin J, Freedman GM, et al. Full-dose gemcitabine with concurrent radiation therapy in patients with nonmeta­ static pancreatic cancer: a multicenter phase II trial. J Clin Oncol 2008;26: 942-947. 119. Burris HA III, Moore MJ, Andersen J, et al. Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreas cancer: a randomized trial. J Clin Oncol 1997;15:2403-2413. 120. Crane CH, Abbruzzese JL, Evans DB, et al. Is the therapeutic index better with gemcitabine-based chemoradiation than with 5-fluorouracil-based chemoradiation in locally advanced pancreatic cancer? Int J Radiat Oncol Biol Phys 2002;52:1293-1302. 121. Rich T, Harris J, Abrams R, et al. Phase II study of external irradiation and weekly paclitaxel for nonmetastatic, unresectable pancreatic cancer: RTOG-98-12. Am J Clin Oncol 2004;27: 51-56.

533

122. Willett CG, Safran H, Abrams RA, et al. Clinical research in pancreatic cancer: the Radiation Therapy Oncology Group trials. Int J Radiat Oncol Biol Phys 2003;56:31-37. 123. Cassidy J, Scheithauer W, McKendrick J, et al. Capecitabine (X) vs bolus 5-FU/leucovorin (LV) as adjuvant therapy for colon cancer (the X-ACT study): efficacy results of a phase III trial [abstract]. J Clin Oncol 2004;23:3509. 124. Cartwright TH, Cohn A, Varkey JA, et al. Phase II study of oral capecitabine in patients with advanced or metastatic pancreatic cancer. J Clin Oncol 2002;20:160-164. 125. Saif MW, Eloubeidi MA, Russo S, et al. Phase I study of capecitabine with concomitant radiotherapy for patients with locally advanced pancreatic cancer: expression analysis of genes related to outcome. J Clin Oncol 2005;23:8679-8687. 126. Crane CH, Ellis LM, Abbruzzese JL, et al. Phase I trial evaluating the safety of bevacizumab with concurrent radiotherapy and capecitabine in locally advanced pancreatic cancer. J Clin Oncol 2006;24:1145-1151. 127. Iannitti D, Dipetrillo T, Akerman P, et al. Erlotinib and chemoradiation followed by maintenance erlotinib for locally advanced pancreatic cancer: a phase I study. Am J Clin Oncol 2005;28:570575. 128. Dragovich T, Huberman M, Von Hoff DD, et al. Erlotinib plus gemcitabine in patients with unresectable pancreatic cancer and other solid tumors: phase IB trial. Cancer Chemother Pharmacol 2007;60:295-303. 129. Moore MJ, Goldstein D, Hamm J, et al. Erlotinib plus gemcitabine compared with gemcitabine alone in patients with advanced pancreatic cancer: a phase III trial of the National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 2007;25:1960-1966. 130. Gorski DH, Beckett MA, Jaskowiak NT, et al Blockage of the vascular endothelial growth factor stress response increases the antitumor effects of ionizing radiation. Cancer Res 1999;59:33743378. 131. Wey J, Fan F, Gray M, et al. Vascular endothelial growth factor receptor-1 promotes migration and invasion in pancreatic carcinoma cell lines. Cancer 2005;104:427-438. 132. Jain RK. Normalization of tumor vasculature: an emerging concept in antiangiogenic therapy. Science 2005;307:58-62. 133. Willett CG, Boucher Y, di Tomaso E, et al. Direct evidence that the VEGF-specific antibody bevacizumab has antivascular effects in human rectal cancer. Nat Med 2004;10:145-147. 134. Willett CG, Kozin SV, Duda DG, et al. Combined vascular endothelial growth factor-targeted therapy and radiotherapy for rectal cancer: theory and clinical practice. Semin Oncol 2006;33;(Suppl 10):S35-S40. 135. Crane CH, Ellis LM, Abbruzzese JL, et al. Phase I trial evaluating the safety of bevacizumab with concurrent radiotherapy and capecitabine in locally advanced pancreatic cancer. J Clin Oncol 2006;24:1145-1151. 136. Saif M, Moody J, Russo S. Gastrointestinal bleeding associated with concurrent capecitabine and radiotherapy in patients with pancreatic cancer. J Pancr 2005;6:325-333. 137. Bonner JA, Harari PM, Giralt J, et al. Radiotherapy plus cetuximab for squamous cell carcinoma of the head and neck. N Engl J Med 2006;354:567-578. 138. Evans DB, Abbruzzese JL, Cleary KR, et al. ­ Preoperative ­chemoradiation for adenocarcinoma of the pancreas: excessive toxicity of prophylactic hepatic irradiation. Int J Radiat Oncol Biol Phys 1995;33:913-918.

23 The Liver and Biliary System Sunil Krishnan, Prajnan Das, and Christopher H. Crane NORMAL ANATOMY AND HISTOLOGY REACTIONS TO RADIATION THERAPY Acute Effects Late Effects The Hepatic Normal-Tissue Complication Probability Model ETIOLOGY, PATHOLOGIC FEATURES, AND PROGNOSIS OF HEPATOBILIARY CANCER EVALUATION AND STAGING HEPATOCELLULAR CARCINOMA: TREATMENT AND RESULTS Surgery Liver-Directed Therapies Radiation Therapy Summary of Approach to Hepatocellular Carcinoma

Normal Anatomy and Histology The liver is divided into four lobes based on surface features. The falciform ligament forms the demarcation between the anatomic left and right lobes anteriorly. The two additional lobes, the caudate and the quadrate, are visible posteriorly between the right and left lobes. The porta hepatis, which transmits the portal vein and lymphatics, hepatic artery and nerves, and hepatic duct to the liver, marks the separation of these two lobes. At the porta hepatis, the portal vein divides into its left and right branches, and the left and right hepatic ducts join to form the common hepatic duct. The cystic duct, which empties the gallbladder, joins the common hepatic duct to form the common bile duct. The liver parenchyma receives most of its blood supply from the portal vein, which carries deoxygenated blood and nutrients from the gastrointestinal tract, and a small percentage (≈25%) from the hepatic artery, which carries oxygenated blood. The architecture of cells within the liver is dominated by the presence of lobules, which are polygonal functional units with branching, spokelike radii of hepatocytes converging at the central vein. At the corners of these polygons are the portal triads, branches of the portal vein, hepatic artery, and bile duct. Sinusoids, small blood vessels between the radiating rows of hepatocytes, convey oxygen-rich hepatic arterial blood and nutrient-rich portal venous blood to the hepatocytes and eventually drain into the central vein, which drains into the hepatic vein. Three hepatic veins (right, middle, and left) empty into the inferior vena cava. Understanding the functional anatomy of the liver allows the anatomic segmentation of the liver to be further

534

BILIARY TRACT CARCINOMA: TREATMENT AND RESULTS Surgery Adjuvant Radiation Therapy Neoadjuvant Radiation Therapy Palliative Radiation Therapy for Unresectable Biliary Tract Carcinoma Radiation Therapy Techniques Summary of Approach to Biliary Tract Carcinoma

refined for the purposes of surgical resection based on independent functional units supplied by a single portal triad. Cantlie’s line, which runs along the gallbladder fossa and inferior vena cava, provides the first division of the liver into a functional left and right lobe. Couinaud further refined this classification by dividing the liver along three vertical planes defined by the three hepatic veins and a transverse plane defined by the left and right portal vein branches. The resulting eight segments are numbered clockwise in the frontal plane; the caudate lobe is segment I. Surgical resection along the plane of a functional segment of the liver provides landmarks for segmental resection of the liver with minimal blood loss. In addition to the benefit of this nomenclature for surgeons during resections, radiologists use this system to define the location of tumors within the liver and their relationship to major vascular structures. The use of intravenous contrast during computed tomography (CT) scanning or magnetic resonance imaging (MRI) of the liver aids radiologists in defining the relationship of tumors to ­major vascular structures and permits characterization of the biology of tumors based on their contrast uptake characteristics. Traditional liver imaging studies involve multiple sequences of images obtained after rapid injection of a bolus of intravenous contrast. In the normal liver, contrast enhancement proceeds through three phases: the he­patic arterial phase, the parenchymal phase, and the portal venous phase. Tumors may be more visible during certain phases of this sequence and can be missed if imaging ­sequences are not obtained during that phase. A variety of classification schemes have been proposed for the biliary tree; the anatomic schemes are most commonly used in practice and are clinically relevant. ­According

23  The Liver and Biliary System

to this classification, the extrahepatic biliary tree is anatomically divided into three regions. The upper third extends from the left and right hepatic ducts to the common hepatic duct at the level of the cystic duct. The middle third extends from the common bile duct at this level to the intrapancreatic portion of the common bile duct. Tumors arising at the bifurcation of the left and right hepatic ducts are often called hilar or Klatskin tumors. The lower third is entirely within the pancreas and empties through the wall of the duodenum into the ampulla of Vater.

Reactions to Radiation Therapy Acute Effects Early studies of radiation therapy to the abdomen demonstrated that irradiation of the liver could produce signs and symptoms of liver disease within 3 months of initiation of therapy in roughly one third of all patients.1-3 These patients developed rapid weight gain, increase in abdominal girth, enlargement (often with tenderness) of the liver, and occasionally jaundice.1 The classic laboratory abnormality was a marked elevation in serum alkaline phosphatase, with less frequent transaminasemia.1 The clinical picture resembled veno-occlusive disease as seen in Budd-Chiari syndrome or after conditioning regimens used for bone marrow transplantation. Although most patients recovered, a few died of this condition. This form of subacute liver injury has been called radiation-induced liver disease (RILD).4 The classic description of the pathologic hallmarks of this injury, as recorded by Reed and Cox,2 include marked congestion, hyperemia, and hemorrhage of the central portion of the lobules and atrophy of the hepatocytes surrounding the central vein with minimal congestion of the portal triad. The central venous congestion extends to the sublobular veins (between the central vein and the hepatic vein) but not the hepatic vein.4,5 Later in the subacute course of RILD, small portal vein branches become obstructed because of collagen deposition.5 The lack of a predominantly inflammatory component to this injury led to the choice of nomenclature as RILD and not radiation hepatitis. The pathogenesis of this typical radiation injury pattern seen in the subacute period after radiation therapy is not fully understood. Fajardo and Colby5 proposed that RILD is characterized by endothelial lesions with fibrinous deposits that trap erythrocytes. Collagen condenses and replaces the fibrin, leading to venous obstruction.6 The deposition of fibrin in the centrilobar region, possibly after activation of the coagulation cascade, and the clinical and biochemical profile of presentation that mimics veno-­occlusive disease strongly suggest that the primary cause of this injury is the endothelial cell rather than the normal hepatocyte. Although the cause of the fibrin deposition remains unclear, a substantial body of evidence suggests that transforming growth factor β plays a key role by activating fibroblasts that migrate to areas of injury.4 The clinical picture of classic RILD is that of a patient with anicteric ascites, tender hepatomegaly, and a disproportionately elevated alkaline phosphatase level in relation to transaminase levels, typically appearing between 2 weeks and 3 months after the completion of radiation treatment given in standard fractions. To confirm that this presentation results from RILD rather than disease progression, it is customary to obtain a CT scan and to perform ­paracentesis

535

to exclude the likelihood of progression of the original tumor. The CT scan may demonstrate a band of decreased density of liver parenchyma corresponding to the radiation portals, although the band may be less apparent after the use of multiple-beam or intensity-modulated radiation therapy and more prominent after proton therapy.4,7,8 The area of low attenuation seen on noncontrast CT scans may show enhancement with dynamic contrast studies.7 No standard therapy has been established for subacute RILD. Over the course of 1 or 2 months, most patients respond to conservative measures such as supportive care, therapy with diuretics (e.g., spironolactone), administration of vitamin K for coagulopathy, and repeated paracentesis. Few patients develop jaundice (characterized by markedly elevated bilirubin and alkaline phosphatase levels and prothrombin time with moderately elevated transaminase levels), progressive ascites refractory to paracentesis and diuretics, and coagulopathies. Although some of these patients may recover, a substantial fraction will die of liver failure.4 The appearance of this unpredictable toxicity in about one third of patients treated with whole-liver irradiation in early studies led to waning interest in its use as treatment for primary and metastatic liver disease. Even doses inadequate for tumor eradication, such as 30 Gy delivered over 3 weeks, to the whole liver carry a 5% risk of RILD, and that risk increases sharply at doses exceeding 33 Gy. Technologic advances in radiation delivery and the recognition that partial-liver irradiation permits adequate dose escalation to liver tumors while sparing the surrounding normal liver have led to a resurgence of interest in the role of radiation therapy in the management of unresectable liver tumors. Nevertheless, even with partial-liver irradiation or focal dose-escalated radiation therapy, the risk of RILD needs to be minimized. Liver tolerance depends on the radiation dose, the relative volume of liver that is irradiated, the baseline functional capacity of the liver, and the use of concurrent chemo­ therapy. Preirradiation estimates of liver radiation tolerance are difficult to make when liver function is damaged or when the total liver volume has been reduced by cancer, chemotherapy, or previous disease. No accurate measure of liver reserve has been found, although liver function tests coupled with CT or isotope scanning, angiography, and sonography can be quite helpful. One method of determining the functional capacity of the liver used by surgical groups, mostly in Japan, is an indocyanine-green retention test.9,10 Another method that provides functional and anatomic information is technetium 99m (99mTc) galactosyl serum albumin scintigraphy, which provides volumetric information as well as kinetic distribution curves.11 For patients who are to undergo preoperative portal vein embolization to induce hypertrophy of the planned remnant liver, scintigraphy offers the opportunity to document this compensatory hypertrophy and, possibly, a shift of functional reserve to the planned remnant liver. The challenge for radiation oncologists is in balancing the desire to administer a tumoricidal dose of radiation to the liver tumor while minimizing the severity of parenchymal destruction produced by irradiating a volume of liver. Maintaining this balance becomes increasingly difficult as the ­volume to be subjected to high-dose radiation increases in size. To aid radiation oncologists in making these ­treatment

536

PART 6  Gastrointestinal Tract

decisions, a rough guide has been ­ developed based on a model of normal tissue complication probability (NTCP) originally developed at the University of Michigan (see “The Hepatic Normal-Tissue Complication Probability Model”). Another liver-related acute toxic effect observed in patients with hepatocellular carcinoma (HCC) treated with focal radiation therapy is that of reactivation of viral hepatitis and precipitation of underlying liver disease.12 These findings led to the incorporation of biologic factors into the NTCP model, thereby defining separate strata of risk of RILD for those who carry the hepatitis B virus and those who do not.13 Another complication that is unique to patients with portal hypertension and splenomegaly or baseline liver dysfunction is an increased risk of coagulopathies due to decreases in platelet counts or function or increases in prothrombin time. Because these patients are at risk of bleeding complications (often from stomach and duodenal mucosae within the treatment field), it may be advisable to treat them prophylactically with proton pump inhibitors and consider discontinuing the radiation therapy if platelet counts drop below an arbitrary level of 50,000 per μL. For biliary tract tumors, radiation-induced hepatic dysfunction has been rare when radiation fields are limited to the biliary tumor bed and adjacent nodal areas. Gastrointestinal bleeding has been reported in 25% to 30% of patients given external irradiation plus transluminal therapy with iridium 192 (192Ir). Many institutions have reported a low risk of severe acute toxicity (e.g., infection, cholangitis, nausea, weight loss, emesis, diarrhea) among patients undergoing palliative radiation therapy.14-20 Presumably, not all of these complications resulted from the radiation exposure, because many of these patients had tumor relapses. The other critical structures in a typical radiation field for hepatobiliary cancer include the stomach, small intestine, kidneys, spinal cord, heart, and lung. The acute reactions that occur after irradiation of these structures include gastritis or epigastric pain, nausea, anorexia, diarrhea, abdominal cramping, fatigue, and hematologic suppression. Most patients undergoing chemoradiation therapy experience some degree of nausea and fatigue and moderate to severe hematologic suppression depending on the type of concurrent chemotherapy used. Acute reactions usually peak during the fourth week of treatment and subside 1 to 2 weeks after completion of a 5-week course of treatment.

Late Effects Chronic radiation liver injury is not well characterized in humans. Zook and colleagues21 evaluated late hepatic complications in beagles after a course of fractionated radiation therapy with fast neutrons or photons to the right hemithorax. The clinical signs included anorexia, weight loss, lethargy, abdominal tenderness, icterus, and hematochezia; biochemical abnormalities included elevation of bilirubin, transaminases, and alkaline phosphatase levels. At necropsy, the irradiated right lobes of the dogs’ livers were atrophic, and the nonirradiated left lobes had undergone compensatory hypertrophy. No qualitative differences were observed be­ tween hepatic lesions in neutron- versus photon-irradiated dogs. The predominant forms of injury were arteriolar and biliary ductal narrowing due to periarteriolar and periductal fibrosis at the microscopic level and subcapsular thickening, parenchymal atrophy, and liver surface nodularity on gross anatomic evaluation.21 Proliferation of bile

ductules was observed in the unirradiated portions of the liver. Similar histopathologic changes were seen in another canine model in which late complications were assessed after intraoperative radiation therapy to the liver.22,23 A significant dose effect was demonstrated at 1 to 2 years after intraoperative irradiation at a dose of 30 Gy, but these atrophic changes were largely reversed through regeneration over longer-term follow-up. In this model, there was a 40% reduction in the uptake of 99mTc-sulfur colloid in the irradiated animals.24 Other major late complications from radiation include upper gastrointestinal bleeding, duodenal ulceration, and hemorrhagic antral gastritis. The effects of radiation on the stomach and small intestines are discussed in Chapter 21. Two studies of patients with cholangiocarcinoma showed that about 5% of irradiated patients experienced severe gastrointestinal bleeding that required transfusion or hospitalization or resulted in death.17,18 However, the attribution of these complications to the radiation is complicated by the persistence of tumor in many instances. Kopelson and colleagues25 have reported asymptomatic, nonobstructing biliary fibrosis after 60 Gy of external irradiation. Such fibrosis is often amenable to endoscopic or percutaneous transhepatic bile duct stenting. However, a couple of studies suggest that aggressive bile duct stenting combined with irradiation can increase the likelihood of developing stent-related infections.16,26 In another study of 63 patients with hilar cholangiocarcinoma, Mogavero and colleagues27 reported that 7 of those patients developed gastric and duodenal obstruction after undergoing radiation therapy. Because this complication was seen only after radiation therapy, it was presumed to be caused by radiation fibrosis, although the investigators acknowledged that the dense adhesions and fibrosis identified during palliative bypass surgery might have resulted from tumor. Use of modern radiation planning and delivery techniques and established dose guidelines helps to ensure that organs at risk are spared from high-dose radiation, which is particularly relevant when external beam radiation therapy (EBRT) is combined with intraluminal brachytherapy. The other critical structures near the liver include the kidneys, liver, small intestine, and spinal cord. Mean decreases in creatinine clearance of 10% and 24% have been observed in patients when 50% and 90% to 100%, respectively, of a single kidney is irradiated with doses higher than 26 Gy.28 However, clinically relevant compromise of kidney function is rare if at least one kidney is spared from the field. Radiation effects on the spinal cord are discussed in Chapter 33.

The Hepatic Normal-Tissue Complication Probability Model Efforts to safely escalate the radiation dose for intrahepatic cancer require some means of predicting which doses to which volumes are most likely to cause liver injury. Much of the work with a Lyman NTCP model for this purpose has been done by investigators at the University of Michigan Department of Radiation Oncology.29,30 This model was first proposed by Lyman as a means of fitting radiation dose-volume histogram data to the likelihood of clinical treatment complications.31 The liver NTCP model uses volume and radiation dose parameters that became available in the early 1990s through the use of three-­dimensional

23  The Liver and Biliary System

537

LAO

RPO

A 120

Percent normal liver

100 80 60 40 20 0 0

B

20

40

60

80

100

120

Dose (%)

FIGURE 23-1.  Radiation treatment planning for an unresectable primary biliary cancer. A, Solid surface reconstruction of the target. The primary tumor (red), liver (blue), kidneys (orange), and spinal cord (green) are shown at left, and beam arrangements are shown at right. B, Dose-volume histogram of the normal liver treated in A. This patient would have received more than 50% of the isocenter dose. The calculated effective volume (Veff) for this plan was 38%, which led to a prescription dose of 69 Gy. LAO, left anterior oblique; RPO, right posterior oblique. (From McGinn CJ, Ten Haken RK, Ensminger WD, et al. Treatment of intrahepatic cancers with radiation doses based on a normal tissue complication probability model. J Clin Oncol 1998;16:2246-2252.)

c­ onformal radiation therapy planning techniques. The model assumes a normal distribution of radiation tolerance and a sigmoid dose-response relationship between fractional dose to a given volume of uniformly irradiated liver and the probability of complications within that organ. The three parameters in the model are TD50, n, and m. TD50(i) is the 50% tolerance dose for uniform irradiation of fractional volume i, n represents the volume effect, and m represents the slope of the dose-complication curve. For an organ such as the liver, in which a small fraction can be injured without significantly affecting the function of the whole organ, n is large and approaches 1. When this model is fit with clinical data, a nonuniform conformal three-dimensional radiation treatment plan is converted to an equivalent dose and volume pair (for all available dose

bins) for uniform partial organ irradiation. The KutcherBurman algorithm32 is then used to calculate the effective volume (Veff) according to the following equation: Veff = ∑ Vi (Di /Dmax )1/n In the equation, Veff is the portion of the liver volume that, if treated uniformly to the prescribed isocenter dose, would be predicted to produce the same risk of complications as the nonuniform irradiation that was actually given. When n equals 1, Veff equals the fractional volume receiving the mean dose. When dose-volume data are fit to estimate a 10% probability of RILD using the NTCP model, the point at which Veff intersects the 10% isocomplication probability contour is used to determine the isocenter prescribed dose (Figs. 23-1 and 23-2).29

538

PART 6  Gastrointestinal Tract 1.00

Veff

0.80

0.60

0.40

0.20 30

40

50

60

70

80

90

100

Dose (Gy)

FIGURE 23-2.  Isocomplication probability contours for the normal tissue complication probability model based on prior clinical data; blue line, 10%; red line, 20%. After the effective dose was determined, the point at which that value intersected the 10% isocomplication probability contour was used to determine the dose prescribed. (From McGinn CJ, Ten Haken RK, Ensminger WD, et al. Treatment of intrahepatic cancers with radiation doses based on a normal tissue complication probability model. J Clin Oncol 1998;16:2246-2252.)

Using this model, McGinn and colleagues have been able to safely escalate the radiation dose delivered to intrahepatic tumors. In one study,29,33 they treated 21 patients with radiation, given in 1.5-Gy fractions twice daily for 11 fractions per week, with concurrent hepatic arterial continuous infusion of fluorodeoxyuridine for up to a maximum of 14 days. The total radiation dose was delivered in two parts, with a 2-week break between treatment blocks. The prescribed dose ranged from 40.5 Gy (for a Veff of 0.831) to 81 Gy (for a Veff of 0.224). RILD occurred in the sole patient with the Veff of 0.831. That patient’s liver injury manifested as a sevenfold elevation of alkaline phosphatase level and ascites that lasted for 1 month; however, the patient did well until the tumor progressed 6 months after treatment. In a subsequent University of Michigan series of 36 patients with unresectable liver malignancies,34 the median survival time free of in-field recurrence was 10.8 months among patients ­treated with a median dose of 58.5 Gy (range, 28.5 to 90 Gy). In the latest update of this experience,35 primary and metastatic liver tumors have been treated safely to doses as high as 90 Gy. Similar but not identical NTCP models have been developed for predicting RILD in Asian patients with HCC, who tend to have a higher prevalence of chronic hepatitis B and compromised hepatic reserve. A Taiwanese model initially identified a lower value for the volume effect parameter n and later incorporated a correction for hepatitis B carrier status.12,13,36 The fitted estimates for n were 0.26 for hepatitis B carriers and 0.86 for noncarriers.12 A similar report from China suggests that the NTCP model may need to be tailored to the population being studied.37 The tolerance of the liver to irradiation is also modified by chemotherapy. Fatal hepatopathy has been reported ­after a dose of 24 Gy given in 17 fractions over 28 days to

the whole liver with preceding and concurrent doxorubicin therapy.38 In a study of liver complications in 20 patients with lymphoma treated with combination chemotherapy and whole- or upper-abdomen irradiation,39 the right liver lobe received 20 Gy in all patients, and the left lobe received 20 Gy in 3 patients, 25 Gy in 10 patients, and 30 to 40 Gy in 7 patients. Doxorubicin was given to 16 patients. The rate of hepatic complications was higher among patients given 30 to 40 Gy to the left lobe, with subclinical liver damage evident in two patients and clinical damage in four patients. The authors of this report concluded that chemotherapy potentiated the effect of radiation and that the tolerance of liver may be reduced to 20 to 25 Gy when high-dose doxorubicin is given during the weeks before the radiation therapy.39

Etiology, Pathologic Features, and Prognosis of Hepatobiliary Cancer Liver and intrahepatic bile duct cancer is the most rapidly increasing type of cancer in the United States, with 21,370 estimated new cases in 2008.40 This type of cancer also accounts for an estimated 18,410 deaths yearly and ranks fifth as a cause of cancer mortality in men.40 HCC is particularly common in sub-Saharan Africa and Southeast Asia, where endemic hepatitis B and aflatoxin exposure account for this high prevalence. Worldwide, HCC is the sixth most common type of cancer (>600,000 new cases per year) and the third most common cause of cancer-related death (about 600,000 deaths per year).41 In the Western world, HCC is often associated with a history of hepatitis C and cirrhosis. Approximately 75% of patients acutely infected with hepatitis C develop chronic hepatitis, and almost one third of those patients go on to develop progressive fibrosis and cirrhosis. After cirrhosis develops, the incidence of HCC is roughly 1% to 4% per year. Another major cause of cirrhosis in addition to hepatitis is chronic alcohol consumption. The insulin resistance syndrome (i.e., metabolic syndrome), manifesting as obesity and diabetes, is also emerging as a risk factor for HCC in the United States.42 In the United States, HCC is almost two to three times more common among Asians, Pacific Islanders, and Native Americans than among African Americans, who have a two to three times higher rate of HCC than whites. Men are two to three times more often affected than women. HCC is one of the few types of cancer that is increasing in both frequency and mortality in the United States, with a 40% increase in mortality between 1990 and 2003.40 The increase in the incidence of HCC in Western countries has not been matched by improvements in survival for this type of cancer. The median survival time remains less than 12 months, and the 5-year survival rate is less than 5%.42 Less than 20% of patients with HCC are candidates for resection or liver transplantation, the traditional curative treatment modalities used for this disease.43,44 Even among patients with favorable tumor features, potentially curative therapy is underused.43 Successful treatment of HCC poses a unique challenge because HCC is essentially two complex diseases in one patient—an aggressive malignant tumor with a propensity to be multifocal and to

23  The Liver and Biliary System

invade vascular structures and a diseased cirrhotic liver that harbors the malignant tumor or tumors and often other premalignant areas that are at risk for progression to overt malignant tumor. Cancers of the biliary tract are rare and have varied presentations. These cancers include intrahepatic and extrahepatic cholangiocarcinoma, gallbladder carcinoma, and carcinoma of the ampulla of Vater. The ­incidence of biliary tract tumors is highest in Southeast Asia, Eastern Europe, South America, and among the Maoris of New Zealand. Risk factors for the development of cholangiocarcinoma include sclerosing cholangitis, Caroli disease, ulcerative colitis, liver fluke ­ infestation, hepatolithiasis, exposure to thorium dioxide (Thorotrast), and the presence of choledochal cysts. Gallbladder carcinomas are more common in patients with long-standing cholelithiasis or calcified gallbladders, and they typically occur in people between the ages of 50 and 70 years, with minimal difference in incidence between men and women. The only potentially curative treatment for cancers of the biliary tract is complete surgical resection, the feasibility of which is often dictated by the location of the tumor. The relative incidence of tumors in the upper, middle, and lower thirds of the extrahepatic biliary tree are 56%, 13%, and 18%, respectively,45 with corresponding resectability rates of 32%, 47%, and 51%.45 Cholangiocarcinomas can also be classified as intrahepatic, perihilar, or distal (Fig. 23-3), with corresponding incidences of 6%, 67%, and 27% and resectability rates of 50%, 60%, and 90%.46 Unfortunately, 60% to 80% of patients are not candidates for surgical resection at presentation. Consequently, the overall prognosis of most hepatobiliary tumors is very poor, with a median survival time of less than 12 months and a 5-year survival rate of approximately 5%.

Evaluation and Staging Primary HCC often manifests with right upper quadrant abdominal pain and worsening of hepatic synthetic functions. However, depending on the manifestations of the preexisting cirrhosis or the malignancy, or both, myriad presentations with nonspecific clinical features are equally likely. Asymptomatic tumors are being diagnosed more often because of rigorous screening programs and improved imaging techniques. Because most patients with primary HCC have underlying cirrhosis, which dictates their overall prognosis, assessment of the severity of the cirrhosis is integral to risk stratification. The most commonly used system for this purpose is the Child-Pugh score, which was originally used to predict surgical mortality but is now used as a guide for evaluating prognosis, suitability for a liver transplantation, and life expectancy.47 The five clinical and biochemical parameters that constitute this scoring system are ­ascites, ­hepatic encephalopathy, total bilirubin level, albumin level, and prothrombin time. A maximum of three points are assigned for each of these parameters based on severity. ­Estimated 2-year survival rate for patients with Child class A disease (5 or 6 points on the Child-Pugh scale) is 85%; that for those with Child class B disease (7 to 9 points) is 55%; and that for those with Child class C disease (more than 9 points) is only about 35%.

539

Intrahepatic 6%

Perihilar 67%

Distal 27%

FIGURE 23-3.  Distribution of 294 cholangiocarcinomas into intrahepatic, perihilar, and distal subgroups. (From Nakeeb A, Pitt HA, Sohn TA, et al. Cholangiocarcinoma. A spectrum of intrahepatic, perihilar, and distal tumors. Ann Surg 1996;224:463-475.)

Serum alpha-fetoprotein levels and characteristic imaging features are often sufficient to make a preoperative diagnosis. The classic imaging feature of a well-differentiated HCC is that of rapid uptake of contrast in the hepatic arterial phase, early washout in the hepatic parenchymal phase, and persisting hypointensity in the portal venous phase of a triple-phase contrast-enhanced CT scan. However, poorly differentiated tumors are often difficult to identify, and early lesions share imaging features similar to those of dysplastic nodules. In these instances, diagnosis can be confirmed by biopsy or repeat imaging. Most cholangiocarcinomas are slow-growing, appear most commonly in the upper biliary tract, and typically manifest with signs or symptoms of obstructive jaundice. Other clinical features include weight loss, right upper quadrant pain (possibly radiating to the right shoulder), cholangitis, and a right upper quadrant mass. The diagnosis is most often suggested by ultrasonographic, CT, or MRI detection of a mass with or without hepatic biliary ductal dilatation. Sonography is the most sensitive noninvasive test for distinguishing between obstructive and nonobstructive jaundice. The diagnosis is then confirmed by analyzing brushing or tissue samples obtained by means of percutaneous transhepatic cholangiography or endoscopic retrograde cholangiopancreatography. Both procedures usually can define the site and nature of the obstruction and permit placement of a stent to relieve obstruction. Endoscopic sonography also provides high-resolution images of the biliary duct for guided biopsies.48 Cholangiocarcinoma can be difficult to distinguish from cholangitis because of the desmoplastic reaction that produces fibrotic tissue with a few clumps of malignant cells. Concentrations of cancer antigen (CA) 19-9 of more than 100 U/mL with associated malignant-appearing strictures have been used to establish a presumptive diagnosis of cholangiocarcinoma in some instances.49 The 2002 staging classification systems of the American Joint Committee on

540

PART 6  Gastrointestinal Tract

BOX 23-1.  American Joint Committee on Cancer Staging System for Cancer of the Liver, Including Intrahepatic Bile Ducts

BOX 23-2.  American Joint Committee on Cancer Staging System for Cancer of the Gallbladder

Primary Tumor (T) TX Primary tumor cannot be assessed T0 No evidence of primary tumor T1 Solitary tumor without vascular invasion T2 Solitary tumor with vascular invasion or multiple tumors none more than 5 cm in greatest dimension T3 Multiple tumors more than 5 cm or tumor involving a major branch of the portal or hepatic vein(s) T4 Tumor(s) with direct invasion of adjacent organs other than the gallbladder or with perforation of visceral peritoneum

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 muscle layer T1a Tumor invades lamina propria T1b Tumor invades muscle layer T2 Tumor invades perimuscular connective tissue; no extension beyond serosa or into liver T3 Tumor perforates the serosa (visceral peritoneum) or directly invades the liver and/or one other adjacent organ or structure, such as the stomach, duodenum, colon, or pancreas, omentum or extrahepatic bile ducts T4 Tumor invades main portal vein or hepatic artery or invades multiple extrahepatic organs or structures

Regional Lymph Nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Regional lymph node metastasis Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage Grouping Stage I T1 Stage II T2 Stage IIIA T3 Stage IIIB T4 Stage IIIC Any T Stage IV Any T

N0 N0 N0 N0 N1 Any N

M0 M0 M0 M0 M0 M1

From Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 131-138.

Cancer (AJCC) for staging cancer of the liver, gallbladder, extrahepatic bile ducts, and ampulla of Vater are shown in Boxes 23-1 to 23-4.50-53

Hepatocellular Carcinoma: Treatment and Results Surgery Liver resection or transplantation offers the best chance of 5-year survival for carefully selected patients with HCC. Resection is usually reserved for patients with solitary HCCs without satellite lesions or major vascular invasion who have either no cirrhosis or Child class A cirrhosis with adequate hepatic functional reserve. In modern series reporting as-treated (as opposed to intent-to-treat) analyses, the 5-year survival rate among such patients approaches 50%, with perioperative mortality rates of less than 5%.54 Recurrences can occur locally (because of understaging of the primary lesion), metachronously (because of persistence of the carcinogenic injury within the residual liver), or ­distantly.55 Liver transplantation can offer excellent outcomes (5-year survival rate of 70%) for patients with cirrhosis and inadequate functional reserve who satisfy the original Milan criteria (a single HCC ≤5 cm in diameter or up

Regional Lymph Nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N Regional lymph node metastasis Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage Grouping Stage 0 Tis Stage IA T1 Stage IB T2 Stage IIA T3 Stage IIB T1 T2 T3 Stage III T4 Stage IV Any T

N0 N0 N0 N0 N1 N1 N1 Any N Any N

M0 M0 M0 M0 M0 M0 M0 M0 M1

From Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 139-144.

to three nodules £3 cm each).56 The high survival rates have led some to advocate expanding the Milan criteria to include additional patients; however, from an intention-­­ to-treat perspective, doing so would lead to declines in overall survival rates as patients drop off the waiting list because of death or disease progression. Adjuvant therapies during the waiting period have not affected outcome. Moreover, the demand for livers for transplantation has already outstripped their availability.54 Living-donor transplantation is an option for some patients and has the potential to offset the imbalance between the supply and demand.

Liver-Directed Therapies The term liver-directed therapies usually refers to percutaneous ablative procedures and transarterial embolization procedures. Percutaneous ethanol injection is a safe and effective option for patients with small tumors and preserved liver function, with 5-year survival rates ­ rivaling

23  The Liver and Biliary System

541

BOX 23-3.  American Joint Committee on Cancer Staging System for Cancer of the Extrahepatic Bile Ducts

BOX 23-4.  American Joint Committee on Cancer Staging System for Cancer of the Ampulla of Vater

Primary Tumor (T) TX Primary tumor cannot be assessed T0 No evidence of primary tumor Tis Carcinoma in situ T1 Tumor confined to the bile duct histologically T2 Tumor invades beyond the wall of the bile duct T3 Tumor invades the liver, gallbladder, pancreas, and/ or unilateral branches of the portal vein (right or left) or hepatic artery (right or left) T4 Tumor invades any of the following: main portal vein or its branches bilaterally, common hepatic artery, or other adjacent structures, such as the colon, stomach, duodenum, or abdominal wall

Primary Tumor (T) TX Primary tumor cannot be assessed T0 No evidence of primary tumor Tis Carcinoma in situ T1 Tumor limited to ampulla of Vater or ­sphincter of Oddi T2 Tumor invades duodenal wall T3 Tumor invades pancreas T4 Tumor invades peripancreatic soft tissues or adjacent organs or structures

Regional Lymph Nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Regional lymph node metastasis Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage Grouping Stage 0 Tis Stage IA T1 Stage IB T2 Stage IIA T3 Stage IIB T1 T2 T3 Stage III T4 Stage IV Any T

N0 N0 N0 N0 N1 N1 N1 Any N Any N

M0 M0 M0 M0 M0 M0 M0 M0 M1

From Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 145-150.

those of resection.57 However, use of this technique for patients with larger tumors (≤5 cm in diameter) or multiple tumors is associated with recurrence rates of 33% to 43%.57 The other drawback of this procedure is that it requires repeated injections to achieve tumor necrosis. In recent years, radiofrequency ablation has largely replaced percutaneous ethanol injection because it requires fewer treatment sessions and achieves better rates of necrosis, recurrence, and survival.57,58 However, radiofrequency ablation is not feasible for tumors located close to the gastrointestinal ­ mucosa (because of ulceration), diaphragm (because of pleural effusions), large biliary radicles (because of bile leaks), and large vessels (because ablative temperatures cannot be achieved because of the heat-sink effect). Other techniques that have been used for liverdirected therapy include cryotherapy and thermal ablation with microwaves, lasers, and high-intensity focused ultrasound.57,59 Transarterial embolization procedures capitalize on the preferential vascularization of HCCs by the hepatic artery rather than the portal vein. Typically, these procedures involve injecting Lipiodol, an iodinated viscous contrast

Regional Lymph Nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Regional lymph node metastasis Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage Grouping Stage 0 Tis Stage IA T1 Stage IB T2 Stage IIA T3 Stage IIB T1 T2 T3 Stage III T4 Stage IV Any T

N0 N0 N0 N0 N1 N1 N1 Any N Any N

M0 M0 M0 M0 M0 M0 M0 M0 M1

From Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 151-156.

agent that stays within the tumor, and an embolizing agent with or without chemotherapeutic agents (e.g., 5-­fluorouracil [5-FU], cisplatin, doxorubicin, mitomycin C, epirubicin) into the hepatic arterial branch supplying the tumor. This treatment combines selective chemotherapeutic and ischemic approaches. Two randomized studies and a meta-analysis have confirmed the superiority of transarterial chemoembolization (TACE) over supportive or symptomatic care.60-62 The 2-year survival rates in one of these studies were 63% for the chemoembolization group and 27% for the control group (P = .009)61; the corresponding rates in the other study were 31% and 11% (P = .002).60 Some notable aspects of these studies were the stringent criteria for patient selection,61 the discontinuation of ther­ apy in 78% of patients in one study,61 and the average 3 to 4.5 TACE procedures performed per patient.60,61 In the larger of the two studies, 903 patients were assessed before 310 underwent curative therapy and 112 were enrolled in the study, highlighting the need for caution before extrapolating these results to patients with vascular invasion, large tumors, and minimal hepatic reserve.61 The 30% objective response rate, the low 5-year survival rate (<10%), and the persistence of viable tumor cells within the TACE field (especially around the capsule because of shunting of ­ collateral blood supply from the portal vein or recanalization of the embolized artery) indicate that

542

PART 6  Gastrointestinal Tract

f­ urther improvements are needed if this approach is to significantly improve outcomes.

Radiation Therapy Advances in the ability to quantify the radiation tolerance of the liver and parallel advances in diagnostic imaging, radiation therapy planning, techniques to account for the respiratory motion of the liver, techniques to target liver tumors by using image-guided therapy, and the increasing availability of stereotactic and charged-particle therapy have allowed the doses of focal radiation therapy to liver tumors to be escalated without increasing toxicity. This improved ability to deliver tumoricidal doses of radiation without increased toxicity has led to a resurgence of interest in using radiation therapy for the curative treatment of HCC, especially for tumors that cannot be resected and are not suitable for traditional liver-directed therapies.

Technologic Advances Accurate delivery of high, highly focused doses of radiation to the liver while sparing the non–tumor-bearing surrounding liver parenchyma requires a confluence of optimal parameters for imaging the tumor, accounting for respiratory movement of the liver, refinements in treatment planning, and the use of more precise radiation beams. Newer imaging modalities and more refined contrast-enhancement protocols have improved the ability to delineate tumors and to integrate these imaging techniques into the radiation therapy planning process. Nevertheless, accurate delivery of therapeutic radiation doses to liver cancer requires accurate delineation of the tumor and a means of compensating for respiratory motion (typically 1 to 2 cm, mostly in the superior-inferior direction)63-67 and for between-treatment differences in patient setup, which remain a challenge even with on-board kilovoltage imaging because most liver tumors appear isointense with the liver without the use of intravenous contrast agents. Many techniques have been developed to account for a priori respiratory movement of liver tumors during radiation therapy, including use of the diaphragm as a surrogate for liver position during breath hold,68 active breathing control to regulate breathing during the delivery of the radiation,69,70 treatment of a volume that encompasses the entire trajectory of the liver tumor (visualized with contrast) on four-dimensional CT (i.e., CT image sets covering a complete breathing cycle traced by the movement of the abdominal wall),71 and use of end-expiratory respiratory gating with the same device that tracks breathing.72,73 Some centers correct errors in patient setup by aligning deformably registered wholeliver contours obtained by cone-beam CT, which also provides volumetric information, at the time of treatment with simulation contours to facilitate image-guided radiation therapy.74 The ability of an external fiducial to serve as a surrogate for the tumor has been evaluated by using implanted internal fiducials, allowing real-time correction of patient setup and offline verification. Initial results suggest that interfractional errors can be reduced to less than 3 mm and intrafractional errors to less than 2 mm during end-expiratory gating.74 These findings suggest that a relatively small expansion in the contour of the tumor volume would be adequate to encompass both errors while maintaining precision during gating

(­ unpublished data). Implanted fiducials can also be used to create a Cartesian coordinate system for image-guided radiation therapy with real-time tracking of tumor motion and gating or robotic control of the treatment beam that can permit dynamic track-and-treat options.75-77

Three-Dimensional Conformal Radiation Therapy The updated results of the University of Michigan experience with conformal radiation therapy in which an isotoxicity model was used for dose escalation provide some of the only mature data on outcomes of radiation therapy for HCC in the United States. In that study, 35 patients with Child class A disease underwent twice-daily radiation therapy (1.5 Gy per fraction) to doses of 40 to 90 Gy with concurrent hepatic arterial floxuridine. The median survival time was 15.2 months, and survival was improved among patients given the higher radiation doses compared with those given the lower doses (P < .01).35 Dawson and colleagues78 used a similar individualized isotoxicity-based prescription dose for 33 patients with HCC treated with stereotactic body radiation therapy to a maximum permitted dose of 60 Gy in 6 fractions. Mornex and colleagues79 published the results of a prospective phase II study evaluating the role of standardfraction radiation therapy (66 Gy in 33 fractions) for small HCCs (one lesion ≤5 cm or two lesions ≤3 cm). Among 27 patients with Child class A or B cirrhosis, the objective response rate was 92%, including an 80% complete response rate. The local recurrence rate was 22%, and the out-of-field liver recurrence rate was 41%. Treatment was tolerated well, with grade 4 toxicities limited to those with Child class B disease. Treatment results from a representative patient in this study are illustrated in Figure 23-4. The largest experience with using radiation therapy for HCC is from Asia. In one study from Korea, Seong and ­colleagues80,81 treated 158 patients with large HCCs (mean tumor size, 9 cm) to a dose of 25.2 to 59.4 Gy (at 1.8 Gy per fraction) and achieved a response rate of 67.1%, a 2-year survival rate of 19.9%, and a 5-year survival rate of 4.7%. Higher radiation doses led to higher objective response rates.81 Other studies have documented objective response rates ranging from 48% to 91% and survival rates from 42% to 94% at 1 year and from 6% to 19% at 5 years, depending on the treatment strategy adopted, the intent of the treatment, the underlying liver function, the assessment of response, and time from which survival was calculated.12,82-89 The treatment strategies in these studies fall into two broad categories: radiation therapy after failure of TACE or planned radiation therapy integrated with TACE. Rates of grade 3 or higher toxicity in these series are less than 27%.12,83-89 Outcomes after hypofractionated regimens seem to be similar to those after standard­fractionation regimens.87,88 A dose-response relationship was observed in some series, but it is difficult to discern whether this reflected the fact that patients with smaller tumors and better liver function were given the higher ­radiation doses. The feasibility and efficacy of using radiation therapy, alone or in combination with TACE, to treat portal venous tumor thrombi are being tested.90-94 When the only radiation target is the tumor thrombus, objective local response rates are about 50%, and median survival time is 5 to 10 months. Although recanalization of major portal venous

23  The Liver and Biliary System

A

B

C

D

543

FIGURE 23-4.  Serial CT scans of a patient with hepatocellular carcinoma treated with radiation alone to 66 Gy in 33 fractions.  A, Before treatment. B, Scan obtained at 1 month after completion of radiation therapy shows acute blush of contrast enhancement. C, Scan at 3 months after completion of therapy shows attenuation of the blush of contrast enhancement. D, Scan at 1 year after completion of therapy shows durable complete response. (From Mornex F, Girard N, Beziat C, et al. Feasibility and efficacy of highdose three-dimensional-conformal radiotherapy in cirrhotic patients with small-size hepatocellular carcinoma non-eligible for curative therapies-mature results of the French Phase II RTF-1 trial. Int J Radiat Oncol Biol Phys 2006;66:1152-1158.)

branches is rare, in most cases, response to this therapy is associated with retraction of the portal venous tumor thrombus from the main portal trunk.93

Proton and Heavy Ion Radiation Therapy Conformal radiation therapy techniques can deliver treatment safely to the liver, but the dose that can be delivered is limited by how much radiation exits the target area to affect other areas of the normal liver and surrounding critical structures (primarily the stomach, bowel, and spinal cord). Radiation therapy using charged particles such as protons or heavy ions exploits the unique characteristic of charged particle beams to deposit most of their energy at the end of their range (i.e., the Bragg peak effect), with steep dose gradients after the fall-off. Consequently, there is virtually no exit dose beyond the target volume. Charged-particle radiation also has sharp lateral margins, a slightly greater biologically effective dose than photons, and can be more precisely targeted to tumor through the routine use of image-guided radiation therapy. A modulated spread-out Bragg peak is used to fully encompass the target volume. The end result is a smaller integral irradiated volume, less irradiation of surrounding tissues, and the potential for dose escalation. The largest experience with proton therapy for HCC comes from Tsukuba, Japan. Chiba and colleagues95 treated 162 patients (192 unresectable HCCs) with proton beam therapy with or without transarterial embolization or percutaneous ethanol injection.95 The median dose of 72 Gy, given in 16 fractions over 29 days, was well tolerated. The 5-year local control rate was 87%, and the 5-year overall survival rate was 23.5%. Among a subset of 50 patients with solitary tumors and Child class A cirrhosis, an ­impressive

5-year survival rate of 53.5% was achieved. Similarly favorable outcomes were observed for patients with unfavorable disease features such as portal venous thrombosis,96 Child class C cirrhosis,97 and limited treatment options.98 This same group reported the safety, feasibility, and efficacy of repeated courses of proton therapy in a series of 27 patients with 68 lesions.99 The median interval between the first and second courses of therapy was 24 months, and the median dose of the second treatment was 66 Gy in 16 fractions. The 5-year overall survival rate in this analysis was 56%, and the 5-year local control rate was 88%. Results of a phase II trial of proton therapy in which 34 patients were treated with 63 cobalt Gray equivalent (CGE) in 15 fractions over 3 weeks indicated a 2-year local control rate of 75% and an overall survival rate of 55%.100 Gastrointestinal bleeding from tumors near the bowel occurred in 9% of the patients in that trial. The next year, Kawashima and colleagues101 reported a 66% overall survival rate at 2 years among 30 patients treated on a phase II protocol with 72 CGE given in 20 fractions over 5 weeks. In the first reported phase I/II trial in which carbon ions were used to treat HCC, Kato and colleagues102 treated 24 patients with 49.5 to 79.5 CGE delivered in 15 fractions over 5 weeks with stepwise escalation of dose per fraction. The 5-year local control rate in that study was 81%, and the 5-year overall survival rate was 25%. The only grade 3 toxicity experienced was radiation dermatitis. Consistent themes in these reports of conformal and charged-particle radiation therapy are the relative safety of administering high radiation doses to a portion of the liver, the excellent local control rates, the possibility of cure for selected patients, and the persisting likelihood of disease

544

PART 6  Gastrointestinal Tract

recurrence outside the high-dose region. As discussed further in the section that follows, this pattern of treatment outcomes suggests that the relative benefits of radiation therapy and other treatment modalities could be maximized by combination strategies that include one or more of these types of therapy.

Radiation Therapy Combined with Systemic, ­Regional, and Local Treatment Modalities In the 1980s, several studies of whole-liver radiation therapy combined with chemotherapy were conducted by the Radiation Therapy Oncology Group (RTOG). The wholeliver irradiation regimens included 21 Gy in 7 fractions and 24 Gy in 10 fractions (administered as 1.2 Gy twice daily, separated by 4 hours), and the chemotherapy regimens included 5-FU and doxorubicin.103 The conclusions from these studies were that the investigated regimen of hyperfractionated radiation therapy had higher acute toxicity and did not demonstrate a significant benefit over standard daily radiation therapy regimens and that alpha-­fetoprotein status was an independent predictor of outcome.103,104 This trial was followed by a randomized study that evaluated induction chemoradiation therapy followed by high-dose chemotherapy against therapy with an iodine 131 (131I) anti-ferritin antibody; crossover between treatment groups was allowed at disease progression.105 The median survival time was 6 months in both treatment groups. The 131Ianti-ferritin antibody therapy salvaged some chemotherapy failures and downstaged disease in some patients to the point at which it could be surgically resected. Using a similar induction chemoradiation strategy, investigators from Johns Hopkins University treated 76 patients with unresectable HCC with whole-liver radiation therapy (21 Gy given in 7 fractions over 10 days) and cisplatin followed by monthly hepatic arterial infusion of cisplatin.106 The overall objective response rate of 43% led to another RTOG randomized study of this induction regimen followed by ­randomization to hepatic arterial cisplatin with or without 131I-­anti-­ferritin antibody therapy.107 Patients treated with hepatic arterial cisplatin had significantly longer median survival times than those receiving intravenous chemotherapy (9.1 versus 3.6 months, P = .0001). 131I-anti-­ferritin antibody therapy increased hematologic toxicity and did not improve survival. Since the completion of those trials and with the advent of three-dimensional conformal radiation therapy, most radiation therapy trials for HCC have abandoned whole-liver irradiation in favor of partial liver irradiation. However, interest continues in combining partial-liver irradiation with chemotherapy, because improvements in local in-field control rates from the use of focal radiation therapy have shifted the pattern of failure to out-of-field intrahepatic and extrahepatic failures. The strategy of combining localized chemotherapy (i.e., TACE) with radiation therapy to improve local control has been extensively investigated in Asian countries. Another strategy being used at the University of Michigan involves regional chemotherapy administered as hepatic arterial infusion of the radiosensitizer floxuridine. A similar strategy involving regional chemotherapy plus irradiation has been used with acceptable toxicity rates for the treatment of advanced HCC in Korea. Systemic chemotherapy continues to be combined with focal radiation therapy in many studies.108

Future trials of combination therapies will rely on an improved understanding of the biology of the underlying liver disease and the nature of the tumor response to radiation. For instance, the recognition that the in-field local control benefit from radiation was often offset by rapid out-of-field intrahepatic and extrahepatic tumor progression led researchers in Taiwan to investigate the role of sublethal radiation-induced paracrine factors in this phenomenon.109 One factor that was consistently elevated in a time- and dose-dependent manner after sublethal irradiation of multiple HCC cells in vitro was vascular endothelial growth factor (VEGF). This elevation translated into enhanced intratumor angiogenesis in vivo, which correlated well with elevated levels of serum VEGF. Surgical series have documented the independent prognostic significance of serum VEGF levels on overall and progression-free survival in HCC.110 Activation of another radiationinduced prosurvival signal transduction pathway in HCC cells, phosphoinositol-3-kinase/atypical kinase/nuclear factor kappa B (PI3K/AKT/NF-ƘB), has been shown to promote invasiveness through expression of matrix metalloproteinases.111 With the availability of targeted agents that specifically inhibit components of these pathways and signaling molecules, future trials are likely to specifically evaluate rational combination strategies.112 Increasing understanding of how cytokines mediate RILD will open doors for the use of some of these targeted agents to mitigate RILD while simultaneously radiosensitizing the tumor.113

Local, Regional, and Systemic Radionuclide ­Therapies Although the results of the early RTOG studies of systemic radionuclide therapies were somewhat disappointing,105,107 newer local and regional radionuclide therapeutic options are becoming available for use in HCC. Delivery of highly localized radiation by means of intratumoral injection of yttrium 90 (90Y) glass microspheres under sonographic guidance has been reported by Tian and colleagues.114 90Y, a pure beta emitter, decays to stable zirconium 90 with a physical half-life of 2.7 days. The beta rays from 90Y embedded in 20- to 30-μm glass microspheres have an average penetration range of 2.5 mm and a maximum range of 11 mm in tissue. One gigabecquerel (27 mCi) of 90Y delivers a total absorbed radiation dose of 50 Gy/kg.115 This product has been approved by the U.S. Food and Drug Administration for treatment of unresectable HCC by segmental, subsegmental, regional, and whole-liver delivery by appropriately placed hepatic arterial catheters. In most inves­ tigations, the 90Y microspheres have been delivered by catheter rather than by intratumoral injection. Two concerns about local and regional therapy with 90Y microspheres were identified during the early experiences— the need for dosimetric guidelines and the potential toxicity from hepatopulmonary shunting resulting from pathologic arteriovenous channels (which are common in HCC) and reflux into the gastroduodenal or right gastric arteries.115 The actual dose delivered is based on estimates of nominal target dose and liver mass calculated from CT scans. Use of 99mTc-macroaggregated albumin as a surrogate for the microspheres before treatment allows gastroesophageal reflux and hepatopulmonary shunt functions to be assessed by angiography and scintigraphy. The subsequent

23  The Liver and Biliary System

delivery procedure with the 90Y microspheres is not very different from chemoembolization. Early trials evaluated the use of the microspheres in the treatment of colorectal metastases to the liver.116,117 Safety trials later conducted for the treatment of HCC have since demonstrated that doses of up to 100 Gy can be safely administered.118-121 Findings from a subsequent phase II trial with a planned dose of 100 Gy suggested a dose-response relationship and a survival benefit among patients receiving more than 104 Gy.122 Survival times were longer among patients with better-preserved liver function and earlierstage disease.123 In a risk-stratification analysis of factors contributing to liver toxicity, Goin and colleagues124 documented a 90-day mortality rate of 18% after treatment with 90Y microspheres. Patients were then stratified into high- or low-risk cohorts based on the presence or absence of any of seven pretreatment and treatment variables related to liver reserve, tumor histology, and radiation dose. The significantly higher rate of 90-day mortality (49% versus 7%) and significantly shorter median survival time (108 versus 466 days) among the high-risk group compared with the low-risk group led the investigators to conclude that the use of 90Y microspheres should be limited to patients with low-risk characteristics.125 Another form of 90Y microspheres consists of biocompatible resin microspheres (rather than glass), with a diameter of 20 to 40 μm. Dose calculations for these microspheres assume a nonuniform distribution in the liver and use an empiric method based on partition modeling of the 99mTc-macroaggregated albumin scan or a formula that incorporates body surface area and fraction of liver volume involved by tumor. Most of the clinical experience of using these resin microspheres to treat unresectable HCC comes from Hong Kong, where an initial phase I/II trial documented their safety and efficacy during intraoperative delivery and monitoring.126 Tumor regression and survival were found to be dose related; patients receiving a dose of less than 120 Gy were more likely to have stable or progressive disease and lower median survival times (26 versus 56 weeks) than did those who received higher doses. A larger trial of 71 patients given an estimated median tumor dose of 225 Gy confirmed a 27% partial response rate, with two of four patients who later underwent resection experiencing a complete pathologic response.127 The treatment was well tolerated, and the median survival time was 9.4 months. These resin microspheres are approved for use in the United States only for the treatment of metastatic colorectal cancer. Another means of delivering regional radiation internally is by hepatic arterial injection of 131I-Lipiodol. The Lipiodol is selectively retained and concentrated within the tumor, and the 131I delivers highly localized radiation. 131I decays with a half-life of 8 days by beta emission, but it also produces gamma rays that can be used for scintigraphy. Two randomized trials of this treatment technique have been conducted in France by Raoul and colleagues.128,129 In the first trial, 27 patients who could not undergo TACE because of the presence of a portal venous tumor thrombus were randomly assigned to receive intrahepatic artery injection of 131I-Lipiodol or supportive medical care with steroids, 5-FU, tamoxifen, and nonsteroidal anti-inflammatory agents.128 The treatment was well tolerated and produced a significant improvement in the 6-month overall survival

545

rate (48% versus 0%, P < .01). In the subsequent trial, 129 patients with unresectable HCC were randomly assigned to receive TACE or 131I-Lipiodol.129 Neither overall survival nor response as gauged by alpha-fetoprotein levels or tumor size differed between the two groups. However, the 131I-Lipiodol therapy was better tolerated clinically and was associated with fewer arterial thromboses identified on angiography during planned repeat procedures. The technical challenges associated with this treatment remain difficulty with achieving accurate dosimetry,130 the small risk that free 131I would be released and trapped by the thyroid (which can be prevented by saturating the thyroid with potassium iodide),131 and the need for hospitalization for protection from the 364-keV gamma rays. The final mode of administering radionuclides is ­systemic, accomplished by parenteral infusion of radiolabeled antibodies. Early attempts at using monoclonal 131Ianti-ferritin antibody were disappointing in that this ­procedure was no better than chemotherapy.105 Monoclonal antibodies against putative HCC-specific antigens have been ­ conjugated to 131I, including Hepama-1, isoferritin, alpha-fetoprotein, and carcinoembryonic antigen.132,133 Achieving adequate tumor specificity and tumor retention remain vexing issues in radioimmunotherapy.

Summary of the Approach to Hepatocellular Carcinoma HCC is a global public health problem and has become an increasing problem in the Western world. Undoubtedly, prevention is the best strategy to confront this problem. In tandem, an emphasis on early detection through surveillance strategies for high-risk populations, identification of prognostic markers (including advances in imaging techniques), and delineation of the molecular etiopathogenic mechanisms should remain a priority. Curative resection, liver transplantation, or ablation should be pursued whenever possible. Radiation therapy definitely has a role in the management of many cases. Close collaborations with colleagues in surgical oncology, interventional radiology, diagnostic radiology, hepatology, and medical oncology and collaboration among radiation oncologists across continents will be required to design robust trials to identify the optimal strategies for integrating radiation therapy with other therapies for this disease.

Biliary Tract Carcinoma: Treatment and Results Surgery Complete surgical resection with negative margins results in the best chance for long-term survival and ultimate cure of biliary duct or gallbladder carcinoma. Unfortunately, however, 60% to 80% of patients present with unresectable disease,134,135 mostly because of extensive comorbid conditions or the presence of metastases, lobar atrophy, early radial extension of disease to adjacent critical vascular structures (portal vein and common hepatic artery) or to hepatic ligaments and parenchyma, or luminal extension involving both biliary ducts. In a review of 294 patients with cholangiocarcinoma who underwent surgical exploration at Johns Hopkins Hospital, 6% were found to have intrahepatic tumors, 67% to have perihilar tumors, and 27%

546

PART 6  Gastrointestinal Tract

to have distal tumors (see Fig. 23-3).46 The corresponding rates of resectability were 50% for intrahepatic tumors, 56% for perihilar tumors, and 91% for distal tumors. The respective 5-year actuarial survival rates for the 191 patients with resected intrahepatic, perihilar, and distal tumors were 44%, 11%, and 28%, and the median survival times were 26, 19, and 22 months. Five-year actuarial survival rates for patients with negative surgical margins were 57% for those with intrahepatic tumors, 19% for those with perihilar tumors, and 29% for those with distal tumors. Patients with positive lymph nodes or with poorly differentiated tumors had shorter survival times. Survival curves were similar for patients with resected tumors who did or did not undergo radiation therapy, but no details were given on the tumor characteristics of these two groups. The median survival time for patients with tumors that were not resected was 8 months. Surgical resection for intrahepatic cholangiocarcinomas requires liver resection; for perihilar tumors, resection of the hepatic duct bifurcation with possible liver resection; and for distal tumors, pancreaticoduodenectomy. Even when disease is deemed resectable by aggressive surgical techniques, optimal margins are difficult to attain. ­Miyazaki and colleagues, reporting 76 patients with hilar cholangiocarcinoma treated with surgery only,136 found that resection was curative in 5 of 11 patients undergoing local resection and in 49 of 65 patients undergoing hepatic resection. Five-year survival rates were 26% for all resections, 40% for curative resections, and 0% for noncurative resections. These investigators stressed the importance of using aggressive surgical approaches to obtain curative resection. It is increasingly evident that a margin-negative resection, often requiring hepatic resection, is the only potentially curative procedure.134,135,137,138 Even with optimal margins after surgery, two of three patients will experience recurrence within 2 years, and 60% of the recurrences of hilar cholangiocarcinoma will be local or regional.139 This pattern of locoregional recurrence provides a rationale for adjuvant treatment with a locoregional modality such as radiation therapy, with or without radiosensitizing ­chemotherapy.

Adjuvant Radiation Therapy Cholangiocarcinomas have a propensity to spread along the bile duct system (proximally and distally); locally by direct extension into adjacent structures (more common for intrahepatic and perihilar tumors); regionally to pericholedochal, peripancreatic, hilar, and cystic lymph nodes (30% to 50% of patients); and distantly to the liver and peritoneum. The usual pattern of failure after resection of extrahepatic cholangiocarcinoma is locoregional recurrence within the liver, tumor bed, and regional lymph nodes.139 In their study of 16 patients with recurrent or residual tumor, Kopelson and colleagues reported regional failure in 13 patients (81%) and distant failure in 9 patients (56%).140 Penetration of tumor through the ductal wall was more predictive of locoregional recurrence than was tumor grade, lymph node status, or perineural or lymphovascular invasion. An autopsy analysis141 of patients who died of recurrent cancer of the main hepatic duct junction after extensive initial resection showed that 93% of recurrences were in the liver hilum. The authors of that analysis

proposed that cancer cells infiltrating and remaining in the connective tissue of the hepatoduodenal ligament might explain the frequent failures in the liver hilum. Nodal recurrences were observed in 29% of cases, peritoneal spread in 14%, and distant metastases in 57%.141 To facilitate comparisons with the surgical literature and to group patients according to whether they have resectable or unresectable disease, the following discussion is based on an anatomic-surgical classification system used to evaluate treatment outcomes from adjuvant therapy.

Perihilar Cholangiocarcinomas Perihilar cholangiocarcinomas are resectable in approximately 14% to 40% of cases, with median survival times of 18 to 35 months and 5-year survival rates of 10% to 30%.46,134,135,142-144 Surgery involves resection, reconstruction with a Roux-en-Y hepaticojejunostomy, and often a liver resection. The likelihood of obtaining negative margins has improved in recent years with the increasing use of hepatic resection, and median survival times are becoming longer at up to 42 months. Even so, in a large contemporary series, about one fifth of all patients had positive margins, and about one fourth of all patients had hepatoduodenal nodal metastases.135 Nodal metastases were found to be of prognostic significance in some series46,138,145 but not in others.135 Even with optimal margins after surgery, two of three patients with hilar cholangiocarcinoma will experience recurrence within 2 years, and 60% of those recurrences will be locoregional.139 The likelihood of close or positive margins after resection, lymph node metastases being present at the time of surgery, and the locoregional nature of recurrences after surgery alone suggest that adjuvant locoregional therapy may be helpful. The role of adjuvant radiation therapy, with or without chemotherapy, has not been conclusively established in prospective randomized trials; nevertheless, most reported series suggest a potential benefit from the addition of adjuvant treatment to surgical resection.17,18,146-151 The European Organisation for Research and Treatment of Cancer (EORTC) reported a survival benefit from the use of EBRT plus intraluminal brachytherapy (ILBT) after curative resection of Klatskin tumors in 38 patients relative to survival among 17 patients who underwent only surgery.151 Significant differences were found in median survival times (19 versus 8.25 months), 1-year survival rates (85% versus 36%), and 3-year survival rates (31% versus 10%) (P = .0005).151 Gerhards and colleagues150 reported a significant improvement in survival time for 71 patients with resectable hilar cholangiocarcinoma who received adjuvant radiation therapy (EBRT with or without ILBT) compared with 20 patients who underwent resection only (24 versus 8 months). No significant survival benefit was obtained from the addition of ILBT to EBRT; moreover, ILBT produced more complications such as abdominal pain and cholangitis (63%) than did no irradiation (40%) or EBRT only (32%) (P = .03). Japanese investigators similarly reported improved 5-year survival rates in a series of 28 patients with stage IV Klatskin tumors treated with intraoperative radiation therapy, postoperative EBRT, or both (33.9%) compared with 19 patients who underwent resection only (13.5%) (P = .0141).149 This difference was attributed to an improvement in the locoregional control rate from 31.3%

23  The Liver and Biliary System

in the surgery-only group to 79.7% in the postoperative radiation therapy group. Schoenthaler and colleagues,148 evaluating treatment outcomes for 129 patients with extrahepatic bile duct adenocarcinoma, found that for patients treated with ­curative intent, no significant difference was evident in median survival time between patients who received surgery only (16 months), surgery with conventional irradiation (16 months), or surgery with charged helium or neon particles (23 months). Patients with microscopic negative margins had a survival advantage over those with positive surgical margins. Patients with microscopic or gross residual tumor had improved survival times when they were treated with adjuvant irradiation. Another analysis of the benefit of adjuvant chemoradiation therapy after resection of cholangiocarcinoma was done at The University of Texas M. D. Anderson Cancer Center.152 In that study, patients with resected extrahepatic cholangiocarcinoma with microscopically positive resection margins or pathologic evidence of locoregional nodal involvement received adjuvant chemoradiation therapy, and those with negative resection margins and pathologically negative nodes did not receive adjuvant treatment. Patients in the adjuvant treatment group and the surgeryonly group had similar 5-year rates of overall survival (36% versus 42%, P = .6) and locoregional recurrence (5-year rates 38% versus 37%, P = .13). Three patients in the ­adjuvant treatment group (7%) experienced acute gastrointestinal side effects of grade 3 or higher. The lack of a locoregional or overall survival difference between the low-risk surgery-only group and the high-risk group given adjuvant treatment suggests that adjuvant chemoradiation therapy could have improved overall survival by reducing the locoregional recurrence rate. However, not all series have shown consistent improvements in survival from the use of adjuvant radiation therapy. In a series reported by Pitt and colleagues at Johns Hopkins,153 no survival difference was seen between patients with perihilar cholangiocarcinoma treated with postoperative EBRT and those treated with surgery alone. Ten of 31 resections were partial resections, and 14 of the 31 patients who underwent surgery were given radiation therapy after surgery.153 Even though the extent of resection was reported as a positive survival factor, the number of patients with partial resection who also received radiation therapy was not reported, suggesting a potential source of bias. Irradiated patients were also more likely to have had hepatic artery encasement than those treated with surgery alone. A subsequent update from the same group showed similar results.154 In another study of 40 resected tumors reported by McMasters and colleagues,155 no difference in survival was found among patients who underwent no radiation therapy, those who underwent preoperative chemoradiation therapy, or those who underwent postoperative chemoradiation therapy. In the absence of randomized, controlled clinical trials evaluating the benefit of adjuvant therapy, it is difficult to state conclusively that adjuvant treatment improves outcomes in hilar cholangiocarcinoma. It seems reasonable to assume that the patients who received radiation therapy in these nonprotocol settings would have had unfavorable disease features, including closer margins, more positive nodes, and more locally invasive tumors (vascular or

547

­ epatic parenchymal). Nevertheless, patients receiving h chemoradiation therapy had outcomes comparable to patients receiving only surgery, suggesting that adjuvant therapy had some benefit.

Distal Cholangiocarcinomas Distal cholangiocarcinomas are more readily resectable in about 90% of cases, with median survival times of about 22 months and 5-year overall survival rates of 30% to 40%. Surgery generally consists of a Whipple procedure that includes pancreaticoduodenectomy with pancreaticojejunostomy, choledochojejunostomy, and gastrojejunostomy. Negative margins are achieved more often than for perihilar cholangiocarcinomas. The role of adjuvant radiation therapy for distal cholangiocarcinoma, like that for perihilar disease, is poorly defined. Serafini and colleagues156 compared the mean survival rates for 34 patients who underwent adjuvant chemoradiation therapy after resection of cholangiocarcinomas or resection only. Mean survival time was not statistically different between these groups (42 versus 29 months, P = .07). However, a subset analysis of patients with distal tumors showed an improved mean survival time among patients receiving chemoradiation therapy (41 versus 25 months for surgery only, P = .04). The investigators concluded that distal location of cholangiocarcinomas was a greater factor in support of adjuvant chemoradiation therapy than was tumor stage. Patients with positive margins also benefited from adjuvant chemoradiation therapy in that series. In an update of that analysis,157 a survival benefit was reported from use of an aggressive adjuvant chemoradiation therapy regimen after resection compared with resection alone (40.5 versus 24 months, P = .05). One criticism of these results is that postoperative complications that prevented some patients from receiving adjuvant therapy may account for the better outcomes among the patients given adjuvant therapy. Other groups have reached a different conclusion, suggesting that adjuvant therapy does not improve outcomes after resection of distal cholangiocarcinomas. One such group26 found that the addition of adjuvant chemoradiation therapy to resection did not improve the median survival time among patients with distal cholangiocarcinomas. Another study, reported by Nakeeb and others,46 also concluded that outcomes for patients given adjuvant radiation therapy after resection of distal cholangiocarcinomas were similar to those for patients not given adjuvant irradiation. The best predictors of outcome in that study were nodal status and degree of differentiation of the tumor. However, patients were not given concurrent chemotherapy with the radiation in this series, even though such therapy has been the recognized standard adjuvant treatment for most gastrointestinal malignancies. This group later revised their institutional policy to offer adjuvant concurrent chemoradiation therapy to all patients with adenocarcinoma of the distal common bile duct who underwent ­pancreaticoduodenectomy. In a subsequent study in which 34 patients were given radiation therapy (median dose, 50.4 Gy) with concurrent 5-FU chemotherapy followed by maintenance chemotherapy, the median overall survival time was significantly longer than that of historical controls who underwent resection alone (36.9 versus 22 months; P < .05).158 The excellent

548

PART 6  Gastrointestinal Tract

actuarial 5-year local control rate of 70% shifted the pattern of failure in that study to predominantly distant metastatic disease. Collectively, these findings indicate that adjuvant chemo­ radiation therapy may have a role in resected distal cholangiocarcinomas in terms of improving local control and possibly survival.

Carcinomas of the Ampulla of Vater Adenocarcinomas of the ampulla of Vater are often difficult to distinguish from peri-ampullary adenocarcinomas. When the distinction is clearly stated in the literature, carcinomas of the ampulla of Vater seem to have a more favorable prognosis than pancreatic cancer. However, one study found that even after microscopically radical resection, 43 of 86 patients who did not undergo adjuvant therapy experienced local or distant metastatic disease recurrence.159 A few studies suggest that adjuvant chemoradiation therapy can be well tolerated and can improve disease control in patients with high-risk features, although this supposition has not been definitively proven.160,161 Willett and colleagues160 first reported their experience with 41 patients with ampullary carcinomas, 12 of whom were considered to have high-risk features (i.e., pancreatic invasion, high tumor grade, positive nodes, or positive margins) and received adjuvant radiation therapy (40 to 50.4 Gy), some with radiosensitizing 5-FU chemotherapy. The 5-year actuarial local control rates were better among these 12 patients given adjuvant therapy than among 17 patients with high-risk features who underwent resection only (83% versus 50%), although this apparent difference was not statistically significant. Distant failure in the liver, peritoneum, or pleura was the predominant pattern. No difference in overall survival was observed. Patients with low-risk features had a 5-year actuarial local control rate of 100% after surgery only. These findings imply that adjuvant therapy improved local control in the high-risk group to make the local control rate more equivalent to that of the low-risk group. Lee and colleagues161 reported a series of 39 patients with resected ampullary cancer, 13 of whom had received adjuvant chemoradiation therapy. Radiation was delivered to the surgical bed and regional nodes to a median dose of 48.6 Gy with a concurrent bolus or continuous-­infusion 5-FU. The overall 3-year survival rate in this series was 55%, but patients with tumor invasion into the pancreas or node-positive disease had worse overall survival rates (roughly 25%) on multivariate analysis. In a prospective single-arm study at Stanford Univer­ sity,162 12 patients with ampullary carcinoma with a variety of unfavorable risk features underwent pancreaticoduodenectomy followed by EBRT (45 Gy) and concurrent protracted venous-infusion 5-FU chemotherapy. Unfavorable features included positive lymph nodes, positive surgical margins, high tumor grade, large tumor, and neurovascular invasion. The 2-year actuarial overall survival rate was 89%, and the median survival time was 34 months, which was better than the median survival time of 14 months in a parallel cohort of 26 patients with resected high-risk pancreatic cancer who underwent the same adjuvant treatment regimen (P < .004). In a large single-institution experience at the Mayo ­Clinic with 125 patients who underwent potentially curative

pancreaticoduodenectomies, 29 of whom received adjuvant chemoradiation therapy, Bhatia and colleagues163 reported an improvement in overall survival among patients with pathologically node-positive disease who received adjuvant 5-FU-based chemoradiation therapy (50.4 Gy). In our own series of 96 patients with ampullary carcinoma who underwent margin-negative pancreaticoduodenectomy at M. D. Anderson Cancer Center,164 54 patients received neoadjuvant or adjuvant fluoro­ pyrimidine-based chemoradiation therapy. Actuarial 5-year rates of local control, distant control, and overall survival were 76%, 68%, and 63%, respectively. On multivariate analysis, T3/T4 tumor classification was ­ independently prognostic for lower overall survival rate (P = .002). Among the 34 patients with T3/T4 tumors, overall survival seemed better for those who received adjuvant chemoradiation therapy (median survival time, 35.2 months versus 16.5 months for those who did not have adjuvant therapy, P = .063). These results suggest that ampullary cancers have a distinctly better treatment outcome than pancreatic adenocarcinomas. A consistent finding across these studies is that adjuvant therapy is primarily beneficial for patients with tumors that have high-risk features. However, the definition of high risk varies across studies. In the absence of clinical trials, recommendations for adjuvant treatment could reasonably be made based on risk stratification. For patients with highrisk features, including pancreatic invasion (T3), positive nodes (N1), positive margins (R1 or R2), and poor histologic grade, resection followed by adjuvant chemoradiation therapy with concurrent fluoropyrimidine-based agents may be justified. If these risk factors are identified or can be predicted in the preoperative setting, preoperative chemoradiation therapy can be considered. These guidelines seem reasonable for a select subset of patients with high-risk disease features given the excellent long-term outcomes observed, the increasing ability to perform chemoradiation therapy safely by using better treatment techniques, and the ability to address local failure, an important cause of disease failure.

Gallbladder Carcinomas Gallbladder carcinomas often manifest as locally advanced disease at diagnosis, with a rapid clinical course and poor overall prognosis. Surgery remains the only curative option; however, only 10% to 30% of patients present with surgically resectable disease.165 For patients with tumors confined to the gallbladder wall, 5-year survival rates after surgery alone range from 10% to 30%.166-168 However, these tumors have a tendency to directly invade the liver; spread regionally to pericholedochal, hilar and cystic lymph nodes; and metastasize distantly to the liver and peritoneum.166,169 Liver invasion and nodal involvement portend a particularly poor prognosis.168 Gallbladder carcinomas detected incidentally after a cholecystectomy usually require a completion radical cholecystectomy with resection of the adjacent liver segment, draining lymph nodes, and incision sites. Radical resection has improved survival, even for patients with advanced-stage disease.170,171 A report by Jarnagin and coworkers139 suggests that gallbladder carcinoma treated with potentially curative surgical resection tends to recur at distant sites. However, local recurrences in the gallbladder fossa are difficult to

23  The Liver and Biliary System

­ etect radiographically, and occult recurrences or persisd tent ­ disease may serve as a source of distant metastatic disease. Locoregional recurrences were a component of all recurrences in 45% of patients experiencing relapse in this large series. Although improvements in survival are more likely to be achieved with systemic therapies, a few reports suggest that locoregional modalities such as radiation therapy can confer a survival benefit. Todoroki and colleagues172 reported a series of 85 patients with stage IV gallbladder carcinoma who showed significant improvement in 5-year survival rates after adjuvant radiation therapy given after aggressive resection (n = 47) compared with aggressive resection alone (8.9% versus 2.9%, P = .002). The local control rate was also significantly better in the group that received radiation therapy (59.1% versus 36.1%, P = .05). A long-term survival advantage was evident among those patients with only microscopic residual disease after surgery who had been treated with radiation (P = .003). Kresl and others173 reported an improvement in the 5-year survival rate among 12 patients with negative surgical margins who were given concurrent 5-FU-based ­chemotherapy with radiation relative to the rate among 9 patients with residual disease given adjuvant chemoradiation therapy (64% versus 0%, P = .002). Five-year survival rates for patients with a complete resection were superior to survival rates for historical surgical controls (64% versus 33%). Median survival times also depended on the extent of resection: 0.6 years for patients with gross residual disease, 1.4 years for those with microscopic disease, and 5.1 years for those with no residual disease (P = .02). Mahe and colleagues174 treated 19 patients with gallbladder cancer with 30 to 50 Gy of postoperative EBRT. Of these patients, 11 had a complete resection, 6 had an incomplete gross resection, and 2 had only percutaneous transhepatic biliary drainage. The overall survival rate at 5 years was 36% for the 11 patients who had undergone complete resection. Czito and others175 reported findings from 22 patients with resected gallbladder cancer given adjuvant radiation therapy (median dose, 45 Gy), 18 of whom also underwent concurrent 5-FU chemotherapy. The median survival time for all 22 patients in that study was 1.9 years, and the local control rate was 65%. In a large National Cancer Data Base review of 2500 patients with gallbladder cancer treated between 1989 and 1990,176 overall survival was found to be better among patients who underwent surgery followed by chemoradiation therapy than among patients who underwent surgery only (hazard ratio = 0.625; 95% confidence interval, 0.52 to 0.75). However, these results might have been related to selection bias that favored multimodality therapy for patients with better prognoses. In summary, the available data are insufficient to reach definitive conclusions regarding the role of adjuvant chemoradiation therapy after resection of gallbladder cancer. One strategy worth considering, given the propensity of these tumors to recur at distant sites after surgery, is to use adjuvant chemotherapy followed by consolidative chemoradiation therapy. A strategy in which chemother­ apy is used first may “select out” patients at risk for distant metastatic relapse and reserve chemoradiation therapy for

549

patients most likely to benefit from a locoregional treatment modality.

Neoadjuvant Radiation Therapy Preoperative chemoradiation therapy has been incorporated into the standard management of gastrointestinal cancers such as rectal cancer and esophageal cancer, and it is being explored for the treatment of gastric cancer. In the case of cholangiocarcinomas, in which close or positive margins are common after surgery, compelling reasons for adopting such an approach include the greater possibility of obtaining clear margins at resection for patients with borderline resectable disease, the possibility of converting unresectable tumors to resectable ones by downstaging, the greater radiobiologic efficacy of such treatment against tumors with intact vasculature, the lower probability of tumor seeding at the time of surgery, the greater likelihood of accomplishing multimodality therapy without delays in postoperative chemoradiation therapy due to prolonged recovery time after major surgery, and the ability to avoid an unnecessary major resection in patients whose disease progresses rapidly during neoadjuvant therapy. From a practical standpoint, the treatment volume in neoadjuvant therapy is likely to be smaller because neither the whole resection edge of the liver nor the bowel loops that fill up the space once occupied by the resected liver segments need to be treated, potentially translating to better tolerability as well. With postoperative radiation therapy, the jejunal loop of the choledochojejunostomy (often the dose-limiting structure) is difficult to access endoscopically in case of complications, whereas in the preoperative setting, the jejunal loop brought up for the Roux-en-Y anastomosis is likely to have not been exposed to radiation and is not a dose-limiting structure. However, nonsurgical definition of tumor can be quite complicated, often requiring the use of more than one imaging modality to accurately define the extent of tumor involvement. A few studies have reported promising results from using preoperative radiation therapy or chemoradiation therapy for gallbladder cancer. McMasters and colleagues155 reported a series of nine patients who underwent preoperative chemoradiation therapy before curative resection of extrahepatic cholangiocarcinoma. The preoperative chemoradiation therapy course was well tolerated. All nine patients treated preoperatively underwent margin-negative resections, compared with 54% of those who did not receive preoperative therapy (P < .01). Disease in six of the nine patients was considered unresectable initially but was rendered resectable after preoperative therapy. Three of the nine patients achieved a pathologic complete response, a 30% rate that parallels that found after preoperative treatment of esophageal, gastric, and rectal cancers. In a similar study from Chile,177 18 patients with potentially resectable gallbladder carcinoma underwent a course of preoperative radiation therapy with concurrent 5-FU chemotherapy. Thirteen patients underwent subsequent gross resection, seven were alive at a median follow-up of 24 months, and only one had documented disease recurrence. Urego and others178 reported findings from 61 patients with cholangiocarcinoma who underwent preoperative or postoperative EBRT (median dose, 49.5 Gy). Thirty patients in the study also were given chemotherapy with 5-FU, leucovorin, and interferon-α or paclitaxel. Twenty-three

550

PART 6  Gastrointestinal Tract

patients had complete resections (including 17 with orthotopic liver transplantation), 4 had partial resections, and 34 had biopsy only. The median survival time was 20 months, and the 5-year actuarial survival rate was 24%. Patients who had undergone complete resections had an impressive 5year actuarial survival rate of 53%. The authors of that report concluded that complete surgical resection with negative surgical margins was the most important factor for achieving cure or long-term survival in cholangiocarcinoma. In an effort to reduce the risk of implantation metastases after endoscopic biliary drainage, Gerhards and colleagues179 evaluated preoperative radiation therapy (given as 3 fractions of 3.5 Gy) for 21 patients who underwent resection of proximal cholangiocarcinomas. The investigators reported that preoperative radiation therapy could be administered without complications and that none of the patients so treated developed implantation metastases within a follow-up time of 2 to 79 months. Although the experience with preoperative radiation therapy for biliary tract tumors has been limited, this approach seems to be feasible and promising. The lack of enthusiasm for neoadjuvant radiation therapy for this purpose may be related to concerns that preoperative irradiation of the proximal portal triad may complicate surgical dissection of that area in patients with resectable disease and that precise delineation of tumor extent is difficult before surgery and may impede the probability of obtaining ­ margin-negative resections. With close cooperation between ­diagnostic radiologists, radiation oncologists, and surgeons, a preoperative radiation approach is still worth exploring, especially for patients with borderline resectable disease. The results of EORTC 05991, a phase II study of preoperative chronomodulated 5-FU and radiation therapy for biliary cancer, should be forthcoming shortly. That study completed accrual in November 2003, when 47 ­ patients had been accrued by centers in Europe and selected centers in the United States.

Palliative Radiation Therapy for Unresectable Biliary Tract Carcinoma When curative surgical resection is not possible, options for palliation include operative bypass procedures, nonoperative drainage by stenting (percutaneous or ­endoscopic), photodynamic therapy, chemoradiation therapy, and brachytherapy. The options involving radiation therapy are reviewed briefly in the following sections.

External Beam Radiation Therapy Few studies have reported the efficacy of radiation therapy for the palliative treatment of gallbladder cancer. In 1970, Vaittinen and others180 showed that the addition of radiation therapy might have extended survival time for patients in whom complete resection was not feasible (13 versus 8 months, P < .10). Mahe and colleagues174 reported that patients at the French Centre René Gauducheau with unresectable gallbladder cancer who were given palliative radiation therapy experienced some benefit in survival and symptom palliation. The role of radiation, with or without chemotherapy, in the treatment of unresectable cholangiocarcinoma has been addressed in several single-institution series.146,154,181-183 Several studies have shown that in addition to palliation of symptoms, radiation therapy can improve outcomes for

­ atients with unresectable cancer of the biliary tract. In one p series of 28 patients with localized, nonmetastatic extrahepatic bile duct cancer, Grove and colleagues181 reported significantly longer median survival times for patients receiving radiation therapy after transtumoral dilatation and intubation than for those who underwent only surgical decompression (12.2 versus 2.2 months, P = .05). Tollenaar and others182 also reported an increase in median survival time from the use of EBRT after surgical decompression compared with surgical decompression alone (16 versus 3 months, P < .001) in a group of 55 patients. In a retrospective report by Crane and colleagues,183 median survival times and 1-year and 2-year survival rates for 52 patients with unresectable disease undergoing primarily EBRT were 10 months, 44%, and 13%, respectively. A suggestion of improved local control was seen with higher radiation doses. In a similar analysis, Ben-David and others184 reviewed their experience with EBRT for 81 patients with extrahepatic cholangiocarcinomas including gallbladder carcinoma; 52 of these tumors were unresectable or had persisting gross residual disease (R2) after attempted resection. The median radiation dose, adjusted for differences in fraction size, was about 60 Gy for the 52 patients. About one half of all patients received concomitant chemotherapy. The median overall survival and progression-free survival times were 13.1 and 7.9 months, respectively. Complete resection (R0) was the only predictive factor significantly associated with an increase in overall and progression-free survival. No difference in outcome was found between those who had R2 resections and those who had resections with microscopic residual disease (R1). The first site of failure was predominantly locoregional (68.8% of all failures). All of these studies are limited in their ability to provide information that can be used for general treatment recommendations because of the small numbers of patients involved. No randomized studies have been conducted to compare patients with similar tumor characteristics who have or have not been treated with radiation. It seems reasonable to assume that a standard 50-Gy dose of radiation therapy (even with radiosensitizing chemotherapy) is unlikely to cure an adenocarcinoma. Local failure remains a challenge in such cases, arguing for more intense radiation schedules and better radiosensitizing strategies.184 The ability to safely escalate the radiation dose in the treatment of cholangiocarcinomas is limited by the presence of radiosensitive gastrointestinal mucosa near the tumor. Distal cholangiocarcinomas are less likely to be amenable to radiation boosts because of their proximity to duodenal mucosa. Similarly, proximity to the hepatic flexure and duodenal mucosa often limits the ability to escalate the radiation dose to gallbladder cancers. However, dose escalation for unresectable hilar and intrahepatic cholangiocarcinomas is technically more feasible, particularly given the recent technologic advances in radiation delivery. Some of the radiation dose-intensification strategies used in the treatment of cholangiocarcinomas are described in the sections that follow.

Charged Particle Radiation Therapy Although some work has been done with charged-particle radiation for the treatment of HCC (see “Proton and Heavy Ion Radiation Therapy”), little information is available on its use for cholangiocarcinomas. In one of the few

23  The Liver and Biliary System

published studies, Schoenthaler and colleagues148 reported treating 129 patients with extrahepatic bile duct cancers with surgery alone (62 patients) or surgery followed by conventional radiation therapy (67 patients) or helium/neon ion therapy (22 patients). Median overall survival times for these three groups of patients were 6.5 months, 11 months, and 14 months, respectively. Overall survival times were significantly better among patients who underwent gross total resection than among those who underwent subtotal resection or biopsy. Adjuvant charged-particle therapy extended median survival times among patients with microscopic or gross residual disease to a greater extent than conventional radiation therapy. In another study, Zurlo and coworkers185 compared several photon treatment plans (“forward-planned” coplanar or non-coplanar conformal fields or “inverse-planned” intensity-modulated fields) with proton treatment plans by using spot scanning for largevolume cholangiocarcinomas. The investigators were able to deliver target doses of 70 Gy to inoperable cholangiocarcinomas while still respecting the dose-volume ­constraints of critical organs such as kidneys, liver, and bowel.185 However, the largest experience with dose escalation for the treatment of biliary tract cancers, predating the use of conformal, intensity-modulated, image-guided, or proton radiation therapy, comes from the use of ILBT.

Intraluminal Brachytherapy ILBT sources for the treatment of biliary tract cancer can be placed within the ductal lumen through percutaneous transhepatic catheters, through retrograde catheters placed endoscopically, or through an access loop jejunostomy. A few retrospective reports suggest that ILBT, with or without EBRT, can improve symptom palliation and median survival times for patients with unresectable biliary tract cancer compared with historical controls or patients who undergo only stenting.14,15,17,19,186-194 Fields and Emami186 showed that treatment with external radiation plus an 192Ir implant improved median survival time for patients with extrahepatic biliary carcinoma over that for patients treated with external radiation only (15 versus 7 months). Alden and Mohiuddin,15 reporting an experience with 24 patients with extrahepatic bile duct cancer, observed that increasing the radiation dose had a favorable effect on survival. The 2-year survival rates in that study were 40% for 13 patients treated with EBRT and ILBT with or without chemotherapy, 25% for 8 patients treated with EBRT and chemotherapy, and 0% for 3 patients treated with ILBT. The median survival times according to radiation dose administered were 4.5 months for less than 45 Gy, 9 months for 45 to 54 Gy, 18 months for 55 to 65 Gy, 25 months for 66 to 70 Gy, and 24 months for doses exceeding 70 Gy. The investigators recommended the use of EBRT to 44 to 46 Gy plus 25 Gy at 1 cm from 192Ir ILBT as an effective and well-tolerated treatment strategy. Another study reported by Kuvshinoff and others19 showed that treating 12 patients with unresectable or recurrent hilar cholangiocarcinoma with internal biliary drainage followed by intraluminal 192Ir and EBRT effectively relieved jaundice in 10 patients; moreover, all patients survived for at least 6 months, and 5 patients were still alive at 16 to 36 months after treatment. Foo and colleagues194 reviewed 24 patients with extrahepatic bile duct carcinoma (22 with locally unresectable disease and 2 with

551

residual disease) treated with EBRT and transcatheter 192Ir. The median survival time was 12.8 months. The overall survival rates in that study were 19% at 2 years and 14% at 5 years, and three patients survived free of disease for 5 or more years. Two of nine patients who were also given 5FU survived for 5 or more years, suggesting a trend toward improved survival among patients given 5-FU in addition to radiation therapy. Morganti and colleagues14 reported 20 patients with unresectable (n = 16) or nonradically resected (n = 4) extrahepatic bile tumors (1 with gallbladder cancer) treated with ILBT and EBRT. The median survival time was 21.2 months for all patients and 13 months for patients who did not undergo resection. Jaundice was controlled in all patients who did not undergo resection, all of whom received biliary stent placement, and pain was controlled in 83% of patients who presented with pain before treatment. Milella and others188 also reported a survival advantage from palliative radiation therapy among a group of patients with unresectable disease. Of the 111 patients who underwent percutaneous transhepatic biliary drainage in that study, 89 received no further treatment, 10 underwent radiation therapy, and 12 received a course of EBRT and ILBT. The median survival time was improved for the two groups given radiation therapy over that for the drainageonly group (11.5 to 14 versus 3.6 months). Five of the 10 patients who received ILBT had complete responses and required no further biliary drainage. Radiation therapy also improved quality of life in that study.188 Golfieri and colleagues190 reported an improvement in mean survival time for nine patients with hepatic hilar carcinoma treated with percutaneous intervention with ILBT followed by EBRT and 5-FU compared with that for seven patients treated with ILBT alone (10 versus 6 months). The investigators found that patients who underwent multimodality treatment achieved mean survival times comparable to those experienced by patients who underwent resection, that all patients achieved remission of jaundice after ILBT, and that no acute complications occurred after radiation therapy; however, one patient developed gastrointestinal bleeding and hepatic abscesses 5 months after completing multimodality therapy. Montemaggi and colleagues189 treated 31 patients with unresectable cholangiocarcinomas (n = 18) or pancreatic cancers (n = 13) by ILBT to 50 Gy or by ILBT to 30 Gy plus EBRT to 45 Gy given with continuous-infusion 5-FU. Although three patients had early cholangitis and three had late gastrointestinal bleeding, jaundice was palliated in all patients, and pain was alleviated in 11 of 13 patients. The 2-year survival rate of 62% for patients with unresectable cholangiocarcinoma led the investigators to conclude that ILBT and EBRT could provide effective palliation and might be “curative.” However, another retrospective analysis195 of 56 patients treated with endoscopic biliary stenting followed by EBRT and ILBT or observation showed no difference in median overall survival times (10 months for the radiation group and 7 months for the control group). The 1-year survival rates for both groups were identical. More patients in the radiation group required stent exchange. Eschelman and colleagues196 evaluated the effect of combined-modality therapy (including intraluminal 192Ir) on the patency of Gianturco stents and survival among 22 patients with malignant biliary obstruction. Eleven of

552

PART 6  Gastrointestinal Tract

the patients had cholangiocarcinoma, and 11 had secondary extrahepatic bile duct malignancies. All patients were given intraluminal 192Ir brachytherapy, with a mean dose of 25 Gy at a 1-cm radius. Of the 11 patients with cholangiocarcinoma, 8 were also given EBRT to a mean dose of 45 Gy. Among the 11 patients with cholangiocarcinoma, the mean duration of stent patency was 19.5 months, and the mean survival time was 22.6 months. In a retrospective analysis, Shin and colleagues197 found a significant improvement in 2-year actuarial survival rate among patients given EBRT followed by ILBT compared with EBRT alone (21% versus 0%, P = .015). The addition of ILBT did not significantly increase treatment morbidity in that analysis. Despite these apparently positive results, the use of ILBT has been associated with an increased risk of complications in some reports. In one report of 16 patients with cholangiocarcinoma treated with intraluminal 192Ir from Leung and colleagues,198 prior treatments had included complete resection in 3 patients, partial resection in 1, bypass procedures in 8, endoscopic stents in 5, external biliary drainage in 15, and external irradiation in 1. The median survival time was 23 months, and 61% of patients survived for 1 year. The most common acute complication was cholangitis in four patients; the most common late complications were duodenal ulcer, seen in two patients, and cholangitis, also seen in two patients. González González and colleagues18 reported 38 patients with unresectable disease treated with EBRT with or without ILBT. In that study, the brachytherapy boost did not influence outcome but was associated with late complications, including duodenal stenosis, upper digestive tract bleeding, and cholangitis. The investigators thought that the complications might not have been entirely caused by the radiation, because most patients also had tumor relapses. In evaluating the role of brachytherapy for patients with resected or unresectable disease, this group found that high radiation doses could be associated with significant toxicity and were potentially detrimental for overall treatment outcomes. With modern radiation therapy techniques, the use of metal stents, and fastidious follow-up, these complications should be less common. If ILBT is to be considered for the treatment of cholangiocarcinoma, it probably has the most value for unresectable hilar cholangiocarcinomas in which the irradiated segment is at least 1 or 2 cm from the closest duodenal mucosa; minimal tumor remains after EBRT; and a metal stent can be placed across the obstructed segment to prevent scarring and consequent cholangitis.

Intraoperative Radiation Therapy Use of intraoperative radiation therapy can spare normal tissue structures because they can be physically moved out of the treatment field and shielded from radiation during the surgery. For unresectable disease, 10 to 25 Gy is usually administered as a single dose to the tumor bed, followed by postoperative EBRT. Buskirk and colleagues16 documented local failure in 9 of 17 patients treated with EBRT (45 to 60 Gy) alone or with 5-FU, in 3 of 10 patients given an 192Ir transcatheter boost, and in 2 of 6 patients given an intraoperative boost with curative intent. Median survival time was 12 months. The only patients surviving longer than 18 months had undergone gross total or subtotal

r­ esection before external irradiation or had been given 192Ir or intraoperative boosts. Monson and others199 subsequently reported a median survival time of 16.5 months in an updated series of 13 patients with unresectable cholangiocarcinoma treated with EBRT and intraoperative radiation after biliary stenting. Similar survival rates for small series of patients have been reported by other groups.200,201 Among 16 patients with unresectable, incompletely resected, or locally recurrent bile duct cancers treated on an RTOG protocol with EBRT (45 to 50 Gy) and intraoperative radiation (14 to 22 Gy), all patients with residual disease larger than 2 cm at the time of intraoperative radiation therapy died of disease.202 In a contrasting opinion, Kurosaki and colleagues203 suggested that intraoperative radiation and EBRT could be useful for patients with bile duct cancer and lymph node metastases or those who underwent noncurative resection, but they thought that these treatment modalities might not extend survival for patients with no nodal metastases or those who undergo curative resection. In an analysis of patients with microscopically positive surgical margins (without macroscopic residual disease), Todoroki and others149 reported that intraoperative radiation and postoperative EBRT improved survival among patients with locally advanced perihilar cholangiocarcinomas. This approach, however, requires intraoperative pathologic assessment of margin status in all patients. In summary, the use of combinations of EBRT and ILBT or intraoperative radiation therapy for patients with unresectable cholangiocarcinoma results in median survival times of 12 to 14 months and 1-year survival rates of 50% to 70% in most studies.14,15,191,199,200 Long-term survival is possible in roughly 10% to 20% of highly selected ­patients.17,187,193,194

Preoperative Treatment Followed by ­Transplantation An attractive option for patients with unresectable perihilar cholangiocarcinomas without nodal involvement is preoperative chemoradiation therapy followed by transplantation, a strategy adopted by groups at the University of Pittsburgh and the Mayo Clinic. Urego and coworkers178 reported a 5-year survival rate of 53.5% for 23 patients who underwent margin-negative resections (17 of whom then went on to undergo liver transplantation) after preoperative chemoradiation therapy, compared with 5% for those who did not undergo margin-negative resections. In a subset analysis, the investigators reported an improved 5-year survival rate for 9 of 25 patients with lymph node–negative hilar cholangiocarcinomas who underwent liver transplantation (>60% versus 0% for those who did not undergo transplantation, P = .009).178 Rea and colleagues204 analyzed the outcomes of 38 patients with unresectable hilar cholangiocarcinoma who received neoadjuvant EBRT and ILBT with 5-FU or capecitabine chemotherapy before orthotopic liver transplantation. These outcomes were compared with those of 26 patients with resectable disease who underwent R0 resections without neoadjuvant therapy. The 5-year survival rate after transplantation was 82% and that after resection was 21%. The investigators concluded that liver transplantation with neoadjuvant chemoradiation therapy achieved better survival with less recurrence than conventional

23  The Liver and Biliary System

resection and should be considered as an alternative to resection for patients with localized, node-negative hilar cholangiocarcinoma.204 This study had biases from the use of nonrandomized comparisons and as-treated analyses, but one feature that stands out in this report is the finding that 16 explanted livers had no residual carcinoma despite the fact that 9 of the 16 patients had unequivocal histologic or cytologic confirmation of tumor before administration of neoadjuvant therapy. This ability to achieve pathologic complete responses with chemoradiation therapy in 29% to 42% of patients attests to the efficacy of this treatment.

Radiation Therapy Techniques For adjuvant treatment, the radiation therapy volume should include the tumor bed as identified by surgical description and the visualization of clips on CT images. The need to treat regional nodes in the porta hepatis and celiac

553

axis is not clear; inclusion of these nodes should be based on surgical information (e.g., nodal positivity, adequacy of nodal sampling, lymphovascular invasion) and the technical feasibility of conformal avoidance of adjacent critical normal structures. For treatment of primary tumors, the radiation volume should include the tumor as defined on CT or MRI with contrast or the extent of tumor as defined on cholangiography. Three-dimensional treatment planning provides greater precision in identifying and treating the appropriate tumor volume. Movement of the liver with respiration needs to be considered in planning the radiation therapy volume, which is readily achieved by using techniques such as four­dimensional CT simulation, breath holds, fluoroscopic assessment of respiratory excursion of the diaphragm, or respiratory gating. If these techniques are not available, diaphragmatic movement of intrahepatic and extrahepatic

A: 59.1 mm

B: 54.9 mm

A

B

A: 42.5 mm B: 40.1 mm

C

D

FIGURE 23-5.  Diagnosis, treatment, and follow-up of a locally advanced hilar cholangiocarcinoma. A, Reconstruction of CT scans shows encasement of left portal vein by the tumor in the left hilar fissure. B, Axial view of the mass in the hilum causing ductal obstruction and left lobar or caudate atrophy. C, Plan for three-dimensional conformal radiation treatment using a three-field technique. The dose to the primary tumor is 50.4 Gy. Red stippled area indicates the internal target volume; yellow, the clinical target volume; and green, the planning target volume. D, Follow-up CT scan at 8 months shows tumor regression with resolution of the biliary ductal dilatation.

554

PART 6  Gastrointestinal Tract

tumors can be accounted for by including in the conformal radiation therapy plan a clinical target volume (gross tumor volume plus at least 5 mm) plus a 1-cm margin in crosssectional directions and a 1.5- to 2.5-cm margin in the cephalocaudad direction. Representative images ­ obtained during radiation therapy for a cholangiocarcinoma are shown in Figure 23-5. Four-field (i.e., anteroposterior, posteroanterior, and lateral/oblique) techniques can provide good distributions for doses of 45 to 50.4 Gy given in fractions of 1.8 to 2 Gy/day. These doses are well tolerated. Doses to the gastric and duodenal mucosae should be limited to 50 Gy, and further escalation of dose should be limited to intrahepatic cholangiocarcinomas. Treatment for these tumors is similar to that for HCCs. The position of the spinal cord and kidneys relative to the radiation therapy volume should be identified. When four-field EBRT is used, the spinal cord and most of the kidney volume should be posterior to the lateral/oblique radiation beams. It is important that most of the liver in the volume does not receive more than 25 to 30 Gy. Transhepatic or internal biliary drainage should be maintained during radiation therapy. Use of innovative nonaxial convergent beams may allow greater ­sparing

B

A

C

of the kidney and other normal structures. The choice of whether to use intensity-modulated radiation therapy should be made on an individual basis for selected patients with predictable tumor motion and with critical structures adjacent to the target volume. Intraluminal irradiation with a linear arrangement of the 192Ir source can be given through percutaneous transhepatic biliary drainage tubes that traverse the tumor. This approach is most effective when combined with 45 to 50 Gy of external beam radiation. With the rapid falloff in radiation dose at increasing distance from the line of sources, the effectively irradiated cylindric volume is small. The usual radiation dose from the implant is 15 to 20 Gy at 1 cm from the axis of the central source line. To prevent the radioactive strand from going through the sidehole perforations in the biliary drainage tube, an internal catheter without side holes is first placed over a guidewire within the biliary tube, and the radioactive strand is placed within the internal catheter. To minimize the likelihood of overdosing the region at the intersection of the two biliary tubes, the radioactive source strands are separated by about 1 cm, and only one is used to treat the common hepatic duct. To minimize the dose to the duodenal mucosa, the

D

FIGURE 23-6.  Postoperative irradiation for cholangiocarcinoma at the bifurcation of hepatic ducts with positive hepatic surgical margins after central hepatic resection and right and left hepaticojejunostomy. A, Isodose plan of the four-field (right anterior  oblique–left posterior oblique, right and left lateral) radiation technique, with a total external dose of 50.4 Gy in 28 fractions. B, Simulation film of the right anterior oblique radiation field shows a preliminary plan for 192Ir seeds to be placed in the right and left hepatic duct catheters, which extend from the skin to the hepaticojejunostomy. C, Anteroposterior radiograph of intraluminal 192Ir seeds,  with 10 seeds in the right catheter and 5 seeds in left catheter. The total activity was 234 mg radium equivalent, and the duration was 44 hours for an implant dose of 25 Gy at 1 cm from the axis of the 192Ir strands. D, Cholangiogram 3.5 years after treatment shows some narrowing of the hepatic ducts. The patient was alive 4.5 years after treatment.

23  The Liver and Biliary System

lowest source in the strand treating the common hepatic duct should be at least 1 to 2 cm away from the duodenum. A similar arrangement of sources can be accomplished by endoscopic placement of 10-Fr catheters in the biliary tree. A representative image of a cholangiocarcinoma treated by ILBT is illustrated in Figure 23-6. For tumors of the lower biliary tract, radiation doses can be boosted with intraoperative electron beam radiation therapy or intraoperative high-dose-rate brachytherapy to give 12.5 to 20 Gy in a single treatment, preferably after 45 to 50 Gy of EBRT given before surgery. As a rule of thumb, 12.5 Gy is used for close margins, 15 Gy for microscopically positive margins, and 20 Gy for gross residual disease.

Summary of the Approach to Biliary Tract Carcinoma The literature suggests that biliary tract cancers have a poor prognosis. Complete surgical resection with negative margins is the only potentially curative treatment for these tumors. The patterns of failure and the results of small treatment series suggest that adjuvant chemoradiation therapy may improve local control and overall survival rates for patients with resected perihilar cholangiocarcinoma, distal cholangiocarcinoma, and high-risk ampullary carcinoma. The patterns of failure in resected gallbladder cancer may indicate a role for selective chemoradiation therapy in patients with resected gallbladder cancer. Patients with unresectable tumors may benefit from radiation therapy with or without chemotherapy. Radiation dose escalation should be considered whenever it is technically possible without increasing toxicity. Radiosensitization strategies that include novel agents should be pursued in prospective clinical trials.

REFERENCES    1. Ingold JA, Reed GB, Kaplan HS, et al. Radiation hepatitis. Am J Roentgenol Radium Ther Nucl Med 1965;93:200-208.    2. Reed GB Jr, Cox AJ Jr. The human liver after radiation injury. A form of veno-occlusive disease. Am J Pathol 1966;48:597-611.    3. Ogata K, Hizawa K, Yoshida M, et al. Hepatic injury following irradiation—a morphologic study. Tokushima J Exp Med 1963;43:240-251.    4. Lawrence TS, Robertson JM, Anscher MS, et al. Hepatic toxicity resulting from cancer treatment. Int J Radiat Oncol Biol Phys 1995;31:1237-1248.    5. Fajardo LF, Colby TV. Pathogenesis of venoocclusive liver disease after radiation. Arch Pathol Lab Med 1980;104:584-588.    6. Fajardo LF, Berthrong M, Anderson RE. Liver. In Radiation Pathology. New York: Oxford University Press, 2001, pp 249257.    7. Ahmadi T, Itai Y, Onaya H, et al. CT evaluation of hepatic injury following proton beam irradiation: appearance, enhancement, and 3D size reduction pattern. J Comput Assist Tomogr 1999;23:655-663.    8. Yamasaki SA, Marn CS, Francis IR, et al. High-dose localized radiation therapy for treatment of hepatic malignant tumors: CT findings and their relation to radiation hepatitis. AJR Am J Roentgenol 1995;165:79-84.    9. Yamanaka N, Okamoto E, Toyosaka A, et al. Prognostic factors after hepatectomy for hepatocellular carcinomas. A univariate and multivariate analysis. Cancer 1990;65:1104-1110.   10. Hemming AW, Scudamore CH, Shackleton CR, et al. Indocyanine green clearance as a predictor of successful hepatic resection in cirrhotic patients. Am J Surg 1992;163:515-518.   11. Schneider PD. Preoperative assessment of liver function. Surg Clin North Am 2004;84:355-373.

555

  12. Cheng JC, Wu JK, Lee PC, et al. Biologic susceptibility of hepatocellular carcinoma patients treated with radiotherapy to radiation-induced liver disease. Int J Radiat Oncol Biol Phys 2004;60:1502-1509.   13. Cheng JC, Liu HS, Wu JK, et al. Inclusion of biological factors in parallel-architecture normal-tissue complication probability model for radiation-induced liver disease. Int J Radiat Oncol Biol Phys 2005;62:1150-1156.   14. Morganti AG, Trodella L, Valentini V, et al. Combined modality treatment in unresectable extrahepatic biliary carcinoma. Int J Radiat Oncol Biol Phys 2000;46:913-919.   15. Alden ME, Mohiuddin M. The impact of radiation dose in ­combined external beam and intraluminal Ir-192 brachytherapy for bile duct cancer. Int J Radiat Oncol Biol Phys 1994;28: 945-951.   16. Buskirk SJ, Gunderson LL, Adson MA, et al. Analysis of ­failure following curative irradiation of gallbladder and extrahepatic bile duct carcinoma. Int J Radiat Oncol Biol Phys 1984;10: 2013-2023.   17. Kamada T, Saitou H, Takamura A, et al. The role of radiotherapy in the management of extrahepatic bile duct cancer: an analysis of 145 consecutive patients treated with intraluminal and/ or external beam radiotherapy. Int J Radiat Oncol Biol Phys 1996;34:767-774.   18. González González D, Gouma DJ, Rauws EA, et al. Role of radiotherapy, in particular intraluminal brachytherapy, in the treatment of proximal bile duct carcinoma. Ann Oncol 1999;10 (Suppl 4):215-220.   19. Kuvshinoff BW, Armstrong JG, Fong Y, et al. Palliation of irresectable hilar cholangiocarcinoma with biliary drainage and radiotherapy. Br J Surg 1995;82:1522-1525.   20. Montemaggi P, Morganti AG, Smaniotto D, et al. Carcinoma of the pancreatic head area. Therapy: intraluminal brachytherapy. Rays 1995;20:338-349.   21. Zook BC, Bradley EW, Casarett GW, et al. The pathologic effects of fractionated fast neutrons or photons on canine liver. Cancer Clin Trials 1981;4:47-65.   22. Cromheecke M, Meijer D, Hietkamp J, et al. Beagle model used in a tissue tolerance study of the response of normal and surgically manipulated liver to single high-dose intraoperative radiotherapy. Lab Anim Sci 1996;46:640-647.   23. Cromheecke M, Grond AJ, Szabo BG, et al. Short- and longterm histopathological changes in the canine liver following single high-dose intraoperative radiation therapy (IORT). Int J Radiat Biol 1999;75:1437-1448.   24. Cromheecke M, Piers BA, Beekhuis H, et al. Tissue damage after single high-dose intraoperative irradiation of the canine liver: evaluation in time by means of radionuclide imaging and light microscopy. Radiat Res 2000;154:537-546.   25. Kopelson G, Harisiadis L, Tretter P, et al. The role of radiation therapy in cancer of the extra-hepatic biliary system: an analysis of thirteen patients and a review of the literature of the effectiveness of surgery, chemotherapy and radiotherapy. Int J Radiat Oncol Biol Phys 1977;2:883-894.   26. Alden ME, Waterman FM, Topham AK, et al. Cholangiocarcinoma: clinical significance of tumor location along the extrahepatic bile duct. Radiology 1995;197:511-516.   27. Mogavero GT, Jones B, Cameron JL, et al. Gastric and duodenal obstruction in patients with cholangiocarcinoma in the porta hepatis: increased prevalence after radiation therapy. AJR Am J Roentgenol 1992;159:1001-1003.   28. Willett CG, Tepper JE, Orlow EL, et al. Renal complications secondary to radiation treatment of upper abdominal malignancies. Int J Radiat Oncol Biol Phys 1986;12:1601-1604.   29. McGinn CJ, Ten Haken RK, Ensminger WD, et al. Treatment of intrahepatic cancers with radiation doses based on a normal tissue complication probability model. J Clin Oncol 1998;16: 2246-2252.   30. Dawson LA, Lawrence TS, Ten Haken RK. Partial liver irradiation. Semin Radiat Oncol 2001;11:240-246.   31. Lyman JT. Complication probability as assessed from dosevolume histograms. Radiat Res Suppl 1985;8:S13-S19.

556

PART 6  Gastrointestinal Tract

  32. Kutcher GJ, Burman C. Calculation of complication ­probability factors for non-uniform normal tissue irradiation: the effective volume method. Int J Radiat Oncol Biol Phys 1989;16: 1623-1630.   33. McGinn CJ, Lawrence TS. Clinical results of the combination of radiation and fluoropyrimidines in the treatment of intrahepatic cancer. Semin Radiat Oncol 1997;7:313-323.   34. Dawson LA, McGinn CJ, Normolle D, et al. Escalated focal liver radiation and concurrent hepatic artery fluorodeoxyuridine for unresectable intrahepatic malignancies. J Clin Oncol 2000;18:2210-2218.   35. Ben-Josef E, Normolle D, Ensminger WD, et al. Phase II trial of high-dose conformal radiation therapy with concurrent hepatic artery floxuridine for unresectable intrahepatic malignancies. J Clin Oncol 2005;23:8739-8747.   36. Cheng JC, Wu JK, Huang CM, et al. Radiation-induced liver disease after three-dimensional conformal radiotherapy for patients with hepatocellular carcinoma: dosimetric analysis and implication. Int J Radiat Oncol Biol Phys 2002;54:156-162.   37. Xu ZY, Liang SX, Zhu J, et al. Prediction of radiation-induced liver disease by Lyman normal-tissue complication probability model in three-dimensional conformal radiation therapy for primary liver carcinoma. Int J Radiat Oncol Biol Phys 2006;65:189-195.   38. Kun LE, Camitta BM. Hepatopathy following irradiation and adriamycin. Cancer 1978;42:81-84.   39. Haddad E, Le Bourgeois JP, Kuentz M, et al. Liver complications in lymphomas treated with a combination of chemotherapy and radiotherapy: preliminary results. Int J Radiat Oncol Biol Phys 1983;9:1313-1319.   40. American Cancer Society. Cancer Facts & Figures 2008. Atlanta: American Cancer Society, 2008.   41. Parkin DM, Bray F, Ferlay J, et al. Global cancer statistics, 2002. CA Cancer J Clin 2005;55:74-108.   42. El-Serag HB. Hepatocellular carcinoma: recent trends in the United States. Gastroenterology 2004;127:S27-S34.   43. El-Serag HB, Siegel AB, Davila JA, et al. Treatment and outcomes of treating of hepatocellular carcinoma among Medicare recipients in the United States: a population-based study. J Hepatol 2006;44:158-166.   44. El-Serag HB, Mason AC. Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med 1999;340:745-750.   45. Launois B, Reding R, Lebeau G, et al. Surgery for hilar cholangiocarcinoma: French experience in a collective survey of 552 extrahepatic bile duct cancers. J Hepatobiliary Pancreat Surg 2000;7:128-134.   46. Nakeeb A, Pitt HA, Sohn TA, et al. Cholangiocarcinoma. A spectrum of intrahepatic, perihilar, and distal tumors. Ann Surg 1996;224:463-473; discussion 473-465.   47. Pugh RN, Murray-Lyon IM, Dawson JL, et al. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973;60:646-649.   48. Brugge WR, Van Dam J. Pancreatic and biliary endoscopy. N Engl J Med 1999;341:1808-1816.   49. de Groen PC, Gores GJ, LaRusso NF, et al. Biliary tract cancers. N Engl J Med 1999;341:1368-1378.   50. Greene FL, Page DL, Fleming ID, et al, eds. Liver (including intrahepatic bile ducts). In AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 131-138.   51. Greene FL, Page DL, Fleming ID, et al, eds. Gallbladder. In AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 139-144.   52. Greene FL, Page DL, Fleming ID, et al, eds. Extrahepatic bile ducts. In AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 145-150.   53. Greene FL, Page DL, Fleming ID, et al, eds. Ampulla of Vater. In AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 151-156.   54. Llovet JM, Schwartz M, Mazzaferro V. Resection and liver transplantation for hepatocellular carcinoma. Semin Liver Dis 2005;25:181-200.   55. Bruix J, Hessheimer AJ, Forner A, et al. New aspects of diagnosis and therapy of hepatocellular carcinoma. Oncogene 2006;25:3848-3856.

  56. Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996;334:693-699.   57. Lencioni R, Cioni D, Crocetti L, et al. Percutaneous ablation of hepatocellular carcinoma: state-of-the-art. Liver Transpl 2004;10;(Suppl 1):S91-S97.   58. Lin SM, Lin CJ, Lin CC, et al. Radiofrequency ablation improves prognosis compared with ethanol injection for hepatocellular carcinoma £4 cm. Gastroenterology 2004;127:1714-1723.   59. Shibata T, Iimuro Y, Yamamoto Y, et al. Small hepatocellular carcinoma: comparison of radio-frequency ablation and percutaneous microwave coagulation therapy. Radiology 2002;223: 331-337.   60. Lo CM, Ngan H, Tso WK, et al. Randomized controlled trial of transarterial lipiodol chemoembolization for unresectable hepatocellular carcinoma. Hepatology 2002;35:1164-1171.   61. Llovet JM, Real MI, Montana X, et al. Arterial embolisation or chemoembolisation versus symptomatic treatment in patients with unresectable hepatocellular carcinoma: a randomised controlled trial. Lancet 2002;359:1734-1739.   62. Llovet JM, Bruix J. Systematic review of randomized trials for unresectable hepatocellular carcinoma: chemoembolization improves survival. Hepatology 2003;37:429-442.   63. Beddar AS, Kainz K, Briere TM, et al. Correlation between internal fiducial tumor motion and external marker motion for liver tumors imaged with 4D-CT. Int J Radiat Oncol Biol Phys 2007;67:630-638.   64. Brandner ED, Wu A, Chen H, et al. Abdominal organ motion measured using 4D CT. Int J Radiat Oncol Biol Phys 2006;65:554-560.   65. Gierga DP, Brewer J, Sharp GC, et al. The correlation between internal and external markers for abdominal tumors: implications for respiratory gating. Int J Radiat Oncol Biol Phys 2005;61:1551-1558.   66. Minohara S, Kanai T, Endo M, et al. Respiratory gated irradiation system for heavy-ion radiotherapy. Int J Radiat Oncol Biol Phys 2000;47:1097-1103.   67. Shirato H, Seppenwoolde Y, Kitamura K, et al. Intrafractional tumor motion: lung and liver. Semin Radiat Oncol 2004;14:1018.   68. Balter JM, Dawson LA, Kazanjian S, et al. Determination of ventilatory liver movement via radiographic evaluation of diaphragm position. Int J Radiat Oncol Biol Phys 2001;51:267-270.   69. Dawson LA, Eccles C, Bissonnette JP, et al. Accuracy of daily image guidance for hypofractionated liver radiotherapy with active breathing control. Int J Radiat Oncol Biol Phys 2005;62:12471252.   70. Eccles C, Brock KK, Bissonnette JP, et al. Reproducibility of liver position using active breathing coordinator for liver cancer radiotherapy. Int J Radiat Oncol Biol Phys 2006;64:751-759.   71. Rietzel E, Chen GT, Choi NC, et al. Four-dimensional imagebased treatment planning: target volume segmentation and dose calculation in the presence of respiratory motion. Int J Radiat Oncol Biol Phys 2005;61:1535-1550.   72. Wagman R, Yorke E, Ford E, et al. Respiratory gating for liver tumors: use in dose escalation. Int J Radiat Oncol Biol Phys 2003;55:659-668.   73. Mageras GS, Yorke E. Deep inspiration breath hold and respiratory gating strategies for reducing organ motion in radiation treatment. Semin Radiat Oncol 2004;14:65-75.   74. Brock KK, Dawson LA, Sharpe MB, et al. Feasibility of a novel deformable image registration technique to facilitate classification, targeting, and monitoring of tumor and normal tissue. Int J Radiat Oncol Biol Phys 2006;64:1245-1254.   75. Shirato H, Shimizu S, Kunieda T, et al. Physical aspects of a realtime tumor-tracking system for gated radiotherapy. Int J Radiat Oncol Biol Phys 2000;48:1187-1195.   76. Murphy MJ. Tracking moving organs in real time. Semin Radiat Oncol 2004;14:91-100.   77. Wurm RE, Gum F, Erbel S, et al. Image guided respiratory gated hypofractionated stereotactic body radiation therapy (HSBRT) for liver and lung tumors: initial experience. Acta Oncol 2006;45:881-889.

23  The Liver and Biliary System   78. Dawson LA, Eccles C, Craig T. Individualized image guided isoNTCP based liver cancer SBRT. Acta Oncol 2006;45:856-864.   79. Mornex F, Girard N, Beziat C, et al. Feasibility and efficacy of high-dose three-dimensional-conformal radiotherapy in cirrhotic patients with small-size hepatocellular carcinoma non-eligible for curative therapies-mature results of the French Phase II RTF-1 trial. Int J Radiat Oncol Biol Phys 2006;66:1152-1158.   80. Seong J, Park HC, Han KH, et al. Clinical results and prognostic factors in radiotherapy for unresectable hepatocellular carcinoma: a retrospective study of 158 patients. Int J Radiat Oncol Biol Phys 2003;55:329-336.   81. Park HC, Seong J, Han KH, et al. Dose-response relationship in local radiotherapy for hepatocellular carcinoma. Int J Radiat Oncol Biol Phys 2002;54:150-155.   82. Hawkins MA, Dawson LA. Radiation therapy for hepatocellular carcinoma: from palliation to cure. Cancer 2006;106: 1653-1663.   83. Guo WJ, Yu EX, Liu LM, et al. Comparison between chemoembolization combined with radiotherapy and chemoembolization alone for large hepatocellular carcinoma. World J Gastroenterol 2003;9:1697-1701.   84. Li B, Yu J, Wang L, Li C, et al. Study of local three-dimensional conformal radiotherapy combined with transcatheter arterial chemoembolization for patients with stage III hepatocellular carcinoma. Am J Clin Oncol 2003;26:e92-99.   85. Liu MT, Li SH, Chu TC, et al. Three-dimensional conformal radiation therapy for unresectable hepatocellular carcinoma patients who had failed with or were unsuited for transcatheter arterial chemoembolization. Jpn J Clin Oncol 2004;34:532-539.   86. Zeng ZC, Tang ZY, Fan J, et al. A comparison of chemoembolization combination with and without radiotherapy for unresectable hepatocellular carcinoma. Cancer J 2004;10:307-316.   87. Wu DH, Liu L, Chen LH. Therapeutic effects and prognostic factors in three-dimensional conformal radiotherapy combined with transcatheter arterial chemoembolization for hepatocellular carcinoma. World J Gastroenterol 2004;10:2184-2189.   88. Zhou ZH, Liu LM, Chen WW, et al. Combined therapy of transcatheter arterial chemoembolization and three-dimensional conformal radiotherapy for hepatocellular carcinoma. Br J Radiol 2007;80:194-201.   89. Park W, Lim DH, Paik SW, et al. Local radiotherapy for patients with unresectable hepatocellular carcinoma. Int J Radiat Oncol Biol Phys 2005;61:1143-1150.   90. Zeng ZC, Fan J, Tang ZY, et al. A comparison of treatment combinations with and without radiotherapy for hepatocellular carcinoma with portal vein and/or inferior vena cava tumor thrombus. Int J Radiat Oncol Biol Phys 2005;61:432-443.   91. Kim DY, Park W, Lim DH, et al. Three-dimensional conformal radiotherapy for portal vein thrombosis of hepatocellular carcinoma. Cancer 2005;103:2419-2426.   92. Tazawa J, Maeda M, Sakai Y, et al. Radiation therapy in combination with transcatheter arterial chemoembolization for hepatocellular carcinoma with extensive portal vein involvement. J Gastroenterol Hepatol 2001;16:660-665.   93. Yamada K, Izaki K, Sugimoto K, et al. Prospective trial of ­combined transcatheter arterial chemoembolization and three­dimensional conformal radiotherapy for portal vein tumor throm­ bus in patients with unresectable hepatocellular carcinoma. Int J Radiat Oncol Biol Phys 2003;57:113-119.   94. Ishikura S, Ogino T, Furuse J, et al. Radiotherapy after transcatheter arterial chemoembolization for patients with hepatocellular carcinoma and portal vein tumor thrombus. Am J Clin Oncol 2002;25:189-193.   95. Chiba T, Tokuuye K, Matsuzaki Y, et al. Proton beam therapy for hepatocellular carcinoma: a retrospective review of 162 patients. Clin Cancer Res 2005;11:3799-3805.   96. Hata M, Tokuuye K, Sugahara S, et al. Proton beam therapy for hepatocellular carcinoma with portal vein tumor thrombus. Cancer 2005;104:794-801.   97. Hata M, Tokuuye K, Sugahara S, et al. Proton beam therapy for hepatocellular carcinoma patients with severe cirrhosis. Strahlenther Onkol 2006;182:713-720.

557

  98. Hata M, Tokuuye K, Sugahara S, et al. Proton beam therapy for hepatocellular carcinoma with limited treatment options. Cancer 2006;107:591-598.   99. Hashimoto T, Tokuuye K, Fukumitsu N, et al. Repeated proton beam therapy for hepatocellular carcinoma. Int J Radiat Oncol Biol Phys 2006;65:196-202. 100. Bush DA, Hillebrand DJ, Slater JM, et al. High-dose proton beam radiotherapy of hepatocellular carcinoma: preliminary results of a phase II trial. Gastroenterology 2004;127:S189-193. 101. Kawashima M, Furuse J, Nishio T, et al. Phase II study of radiotherapy employing proton beam for hepatocellular carcinoma. J Clin Oncol 2005;23:1839-1846. 102. Kato H, Tsujii H, Miyamoto T, et al. for the Liver Cancer Working Group. Results of the first prospective study of carbon ion radiotherapy for hepatocellular carcinoma with liver cirrhosis. Int J Radiat Oncol Biol Phys 2004;59:1468-1476. 103. Stillwagon GB, Order SE, Guse C, et al. 194 hepatocellular cancers treated by radiation and chemotherapy combinations: toxicity and response: a Radiation Therapy Oncology Group Study. Int J Radiat Oncol Biol Phys 1989;17:1223-1229. 104. Stillwagon GB, Order SE, Guse C, et al. Prognostic factors in unresectable hepatocellular cancer: Radiation Therapy Oncology Group Study 83-01. Int J Radiat Oncol Biol Phys 1991;20: 65-71. 105. Order S, Pajak T, Leibel S, et al. A randomized prospective trial comparing full dose chemotherapy to 131I antiferritin: an RTOG study. Int J Radiat Oncol Biol Phys 1991;20:953-963. 106. Abrams RA, Cardinale RM, Enger C, et al. Influence of prognostic groupings and treatment results in the management of unresectable hepatoma: experience with cisplatinum-based chemoradiotherapy in 76 patients. Int J Radiat Oncol Biol Phys 1997;39:1077-1085. 107. Abrams RA, Pajak TF, Haulk TL, et al. Survival results among patients with alpha-fetoprotein-positive, unresectable hepatocellular carcinoma: analysis of three sequential treatments of the RTOG and Johns Hopkins Oncology Center. Cancer J Sci Am 1998;4:178-184. 108. Hsu WC, Chan SC, Ting LL, et al. Results of three-dimensional conformal radiotherapy and thalidomide for advanced hepatocellular carcinoma. Jpn J Clin Oncol 2006;36:93-99. 109. Chung YL, Jian JJ, Cheng SH, et al. Sublethal irradiation induces vascular endothelial growth factor and promotes growth of hepatoma cells: implications for radiotherapy of hepatocellular carcinoma. Clin Cancer Res 2006;12:2706-2715. 110. Poon RT, Ho JW, Tong CS, et al. Prognostic significance of serum vascular endothelial growth factor and endostatin in patients with hepatocellular carcinoma. Br J Surg 2004;91:1354-1360. 111. Cheng JC, Chou CH, Kuo ML, et al. Radiation-enhanced hepatocellular carcinoma cell invasion with MMP-9 expression through PI3K/Akt/NF-kappaB signal transduction pathway. Oncogene 2006;25:7009-7018. 112. Ueda S, Basaki Y, Yoshie M, et al. PTEN/Akt signaling through epidermal growth factor receptor is prerequisite for angiogenesis by hepatocellular carcinoma cells that is susceptible to inhibition by gefitinib. Cancer Res 2006;66:5346-5353. 113. Christiansen H, Sheikh N, Saile B, et al. x-Irradiation in rat liver: consequent upregulation of hepcidin and downregulation of hemojuvelin and ferroportin-1 gene expression. Radiology 2007;242:189-197. 114. Tian JH, Xu BX, Zhang JM, et al. Ultrasound-guided internal radiotherapy using yttrium-90-glass microspheres for liver malignancies. J Nucl Med 1996;37:958-963. 115. Murthy R, Nunez R, Szklaruk J, et al. Yttrium-90 microsphere therapy for hepatic malignancy: devices, indications, technical considerations, and potential complications. Radiographics 2005;25 (Suppl 1):S41-S55. 116. Herba MJ, Illescas FF, Thirlwell MP, et al. Hepatic malignancies: improved treatment with intraarterial Y-90. Radiology 1988;169:311-314. 117. Blanchard RJ, Morrow IM, Sutherland JB. Treatment of liver tumors with yttrium-90 microspheres alone. Can Assoc Radiol J 1989;40:206-210.

558

PART 6  Gastrointestinal Tract

118. Grady ED. Intrahepatic arterial 90-yttrium resin spheres to treat liver cancer. Int J Nucl Med Biol 1978;5:253-254. 119. Houle S, Yip TK, Shepherd FA, et al. Hepatocellular carcinoma: pilot trial of treatment with Y-90 microspheres. Radiology 1989;172:857-860. 120. Shepherd FA, Rotstein LE, Houle S, et al. A phase I dose escalation trial of yttrium-90 microspheres in the treatment of primary hepatocellular carcinoma. Cancer 1992;70:2250-2254. 121. Yan ZP, Lin G, Zhao HY, et al. An experimental study and clinical pilot trials on yttrium-90 glass microspheres through the hepatic artery for treatment of primary liver cancer. Cancer 1993;72:3210-3215. 122. Dancey JE, Shepherd FA, Paul K, et al. Treatment of nonresectable hepatocellular carcinoma with intrahepatic 90Y­microspheres. J Nucl Med 2000;41:1673-1681. 123. Geschwind JF, Salem R, Carr BI, et al. Yttrium-90 microspheres for the treatment of hepatocellular carcinoma. Gastroenterology 2004;127:S194-205. 124. Goin JE, Salem R, Carr BI, et al. Treatment of unresectable hepatocellular carcinoma with intrahepatic yttrium 90 microspheres: a risk-stratification analysis. J Vasc Interv Radiol 2005;16:195-203. 125. Goin JE, Salem R, Carr BI, et al. Treatment of unresectable hepatocellular carcinoma with intrahepatic yttrium 90 microspheres: factors associated with liver toxicities. J Vasc Interv Radiol 2005;16:205-213. 126. Lau WY, Leung WT, Ho S, et al. Treatment of inoperable hepatocellular carcinoma with intrahepatic arterial yttrium-90 microspheres: a phase I and II study. Br J Cancer 1994;70:994-999. 127. Lau WY, Ho S, Leung TW, et al. Selective internal radiation therapy for nonresectable hepatocellular carcinoma with intraarterial infusion of 90-yttrium microspheres. Int J Radiat Oncol Biol Phys 1998;40:583-592. 128. Raoul JL, Guyader D, Bretagne JF, et al. Randomized controlled trial for hepatocellular carcinoma with portal vein thrombosis: intra-arterial iodine-131-iodized oil versus medical support. J Nucl Med 1994;35:1782-1787. 129. Raoul JL, Guyader D, Bretagne JF, et al. Prospective randomized trial of chemoembolization versus intra-arterial injection of 131I-labeled-iodized oil in the treatment of hepatocellular carcinoma. Hepatology 1997;26:1156-1161. 130. Monsieurs MA, Bacher K, Brans B, et al. Patient dosimetry for 131I-lipiodol therapy. Eur J Nucl Med Mol Imaging 2003;30: 554-561. 131. Bacher K, Brans B, Monsieurs M, et al. Thyroid uptake and radiation dose after (131)I-lipiodol treatment: is thyroid blocking by potassium iodide necessary? Eur J Nucl Med Mol Imaging 2002;29:1311-1316. 132. Lambert B, Van de Wiele C. Treatment of hepatocellular carcinoma by means of radiopharmaceuticals. Eur J Nucl Med Mol Imaging 2005;32:980-989. 133. Ho S, Lau WY, Leung TW, et al. Internal radiation therapy for patients with primary or metastatic hepatic cancer: a review. Cancer 1998;83:1894-1907. 134. Burke EC, Jarnagin WR, Hochwald SN, et al. Hilar cholangiocarcinoma: patterns of spread, the importance of hepatic resection for curative operation, and a presurgical clinical staging system. Ann Surg 1998;228:385-394. 135. Jarnagin WR, Fong Y, DeMatteo RP, et al. Staging, resectability, and outcome in 225 patients with hilar cholangiocarcinoma. Ann Surg 2001;234:507-517; discussion 517–509. 136. Miyazaki M, Ito H, Nakagawa K, et al. Aggressive surgical approaches to hilar cholangiocarcinoma: hepatic or local resection? Surgery 1998;123:131-136. 137. Pichlmayr R, Weimann A, Klempnauer J, et al. Surgical treatment in proximal bile duct cancer. A single-center experience. Ann Surg 1996;224:628-638. 138. Nimura Y, Kamiya J, Nagino M, et al. Aggressive surgical treatment of hilar cholangiocarcinoma. J Hepatobiliary Pancreat Surg 1998;5:52-61. 139. Jarnagin WR, Ruo L, Little SA, et al. Patterns of initial disease recurrence after resection of gallbladder carcinoma and hilar cholangiocarcinoma: implications for adjuvant therapeutic strategies. Cancer 2003;98:1689-1700.

140. Kopelson G, Galdabini J, Warshaw AL, et al. Patterns of failure after curative surgery for extra-hepatic biliary tract carcinoma: implications for adjuvant therapy. Int J Radiat Oncol Biol Phys 1981;7:413-417. 141. Tsuzuki T, Kuramochi S, Sugioka A, et al. Postresection autopsy findings in patients with cancer of the main hepatic duct junction. Cancer 1991;67:3010-3013. 142. Klempnauer J, Ridder GJ, Werner M, et al. What constitutes long-term survival after surgery for hilar cholangiocarcinoma? Cancer 1997;79:26-34. 143. Lillemoe KD. Current status of surgery for Klatskin tumors. Curr Opin Gen Surg 1994:161-167. 144. Chao TC, Greager JA. Carcinoma of the extrahepatic bile ducts. J Surg Oncol 1991;46:145-150. 145. Klempnauer J, Ridder GJ, von Wasielewski R, et al. Resectional surgery of hilar cholangiocarcinoma: a multivariate analysis of prognostic factors. J Clin Oncol 1997;15:947-954. 146. Mahe M, Romestaing P, Talon B, et al. Radiation therapy in extrahepatic bile duct carcinoma. Radiother Oncol 1991;21:121127. 147. Kraybill WG, Lee H, Picus J, et al. Multidisciplinary treatment of biliary tract cancers. J Surg Oncol 1994;55:239-245. 148. Schoenthaler R, Phillips TL, Castro J, et al. Carcinoma of the extrahepatic bile ducts. The University of California at San Francisco experience. Ann Surg 1994;219:267-274. 149. Todoroki T, Ohara K, Kawamoto 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: 581-587. 150. Gerhards MF, van Gulik TM, González González D, et al. Results of postoperative radiotherapy for resectable hilar cholangiocarcinoma. World J Surg 2003;27:173-179. 151. González González D, Gerard JP, Maners AW, et al. Results of radiation therapy in carcinoma of the proximal bile duct (Klatskin tumor). Semin Liver Dis 1990;10:131-141. 152. Borghero Y, Crane C, Szklaruk J, et al. Extrahepatic bile duct adenocarcinoma: patients at high-risk for local recurrence treated with surgery and adjuvant chemoradiation have an equivalent overall survival to patients with standard-risk treated with surgery alone. Ann Surg Oncol 2008;15:3147-3156. 153. Pitt HA, Nakeeb A, Abrams RA, et al. Perihilar cholangiocarcinoma. Postoperative radiotherapy does not improve survival. Ann Surg 1995;221:788-797; discussion 797-788. 154. Cameron JL, Pitt HA, Zinner MJ, et al. Management of proximal cholangiocarcinomas by surgical resection and radiotherapy. Am J Surg 1990;159:91-97; discussion 97-98. 155. McMasters KM, Tuttle TM, Leach SD, et al. Neoadjuvant chemoradiation for extrahepatic cholangiocarcinoma. Am J Surg 1997;174:605-608; discussion 608-609. 156. Serafini FM, Sachs D, Bloomston M, et al. Location, not staging, of cholangiocarcinoma determines the role for adjuvant chemoradiation therapy. Am Surg 2001;67:839-843; discussion 843-834. 157. Kelley ST, Bloomston M, Serafini F, et al. Cholangiocarcinoma: advocate an aggressive operative approach with adjuvant chemotherapy. Am Surg 2004;70:743-748; discussion 748-749. 158. Hughes MA, Frassica DA, Yeo CJ, et al. Adjuvant concurrent chemoradiation for adenocarcinoma of the distal common bile duct. Int J Radiat Oncol Biol Phys 2007;68:178-182. 159. De Castro SM, Kuhlmann KF, van Heek NT, et al. Recurrent disease after microscopically radical (R0) resection of periampullary adenocarcinoma in patients without adjuvant therapy. J Gastrointest Surg 2004;8:775-784. 160. Willett CG, Warshaw AL, Convery K, et al. Patterns of failure after pancreaticoduodenectomy for ampullary carcinoma. Surg Gynecol Obstet 1993;176:33-38. 161. Lee JH, Whittington R, Williams NN, et al. Outcome of ­pancreaticoduodenectomy and impact of adjuvant therapy for ­ampullary carcinomas. Int J Radiat Oncol Biol Phys 2000;47: 945-953. 162. Mehta VK, Fisher GA, Ford JM, et al. Adjuvant chemoradiotherapy for “unfavorable” carcinoma of the ampulla of Vater: preliminary report. Arch Surg 2001;136:65-69.

23  The Liver and Biliary System 163. Bhatia S, Miller RC, Haddock MG, et al. Adjuvant therapy for ampullary carcinomas: the Mayo Clinic experience. Int J Radiat Oncol Biol Phys 2006;66:514-519. 164. Krishnan S, Rana V, Evans DB, et al. Role of adjuvant chemoradiation therapy in adenocarcinomas of the ampulla of Vater. Int J Radiat Oncol Biol Phys 2008;70:735-743. 165. Fong Y, Malhotra S. Gallbladder cancer: recent advances and current guidelines for surgical therapy. Adv Surg 2001;35:1-20. 166. Piehler JM, Crichlow RW. Primary carcinoma of the gallbladder. Surg Gynecol Obstet 1978;147:929-942. 167. Houry S, Haccart V, Huguier M, et al. Gallbladder cancer: role of radiation therapy. Hepatogastroenterology 1999;46:15781584. 168. Arnaud JP, Casa C, Georgeac C, et al. Primary carcinoma of the gallbladder—review of 143 cases. Hepatogastroenterology 1995;42:811-815. 169. Sons HU, Borchard F, Joel BS. Carcinoma of the gallbladder: autopsy findings in 287 cases and review of the literature. J Surg Oncol 1985;28:199-206. 170. Todoroki T, Kawamoto T, Takahashi H, et al. Treatment of gallbladder cancer by radical resection. Br J Surg 1999;86:622-627. 171. Shirai Y, Yoshida K, Tsukada K, et al. Radical surgery for gallbladder carcinoma. Long-term results. Ann Surg 1992;216:565568. 172. Todoroki T, Kawamoto T, Otsuka M, et al. Benefits of combining radiotherapy with aggressive resection for stage IV gallbladder cancer. Hepatogastroenterology 1999;46:1585-1591. 173. Kresl JJ, Schild SE, Henning GT, et al. Adjuvant external beam radiation therapy with concurrent chemotherapy in the management of gallbladder carcinoma. Int J Radiat Oncol Biol Phys 2002;52:167-175. 174. Mahe M, Stampfli C, Romestaing P, et al. Primary carcinoma of the gall-bladder: potential for external radiation therapy. ­Radiother Oncol 1994;33:204-208. 175. Czito BG, Hurwitz HI, Clough RW, et al. Adjuvant externalbeam radiotherapy with concurrent chemotherapy after resection of primary gallbladder carcinoma: a 23-year experience. Int J Radiat Oncol Biol Phys 2005;62:1030-1034. 176. Donohue JH, Stewart AK, Menck HR. The National Cancer Data Base report on carcinoma of the gallbladder, 1989-1995. Cancer 1998;83:2618-2628. 177. de Aretxabala X, Roa I, Burgos L, et al. Preoperative chemoradiotherapy in the treatment of gallbladder cancer. Am Surg 1999;65:241-246. 178. Urego M, Flickinger JC, Carr BI. Radiotherapy and multimodality management of cholangiocarcinoma. Int J Radiat Oncol Biol Phys 1999;44:121-126. 179. Gerhards MF, González DG, ten Hoopen-Neumann H, et al. Prevention of implantation metastases after resection of proximal bile duct tumours with pre-operative low dose radiation therapy. Eur J Surg Oncol 2000;26:480-485. 180. Vaittinen E. Carcinoma of the gall-bladder. A study of 390 cases diagnosed in Finland 1953-1967. Ann Chir Gynaecol Fenn Suppl 1970;168:1-81. 181. Grove MK, Hermann RE, Vogt DP, et al. Role of radiation after operative palliation in cancer of the proximal bile ducts. Am J Surg 1991;161:454-458. 182. Tollenaar RA, van de Velde CJ, Taat CW, et al. External radiotherapy and extrahepatic bile duct cancer. Eur J Surg 1991;157: 587-589. 183. Crane CH, Macdonald KO, Vauthey JN, et al. Limitations of conventional doses of chemoradiation for unresectable biliary cancer. Int J Radiat Oncol Biol Phys 2002;53:969-974. 184. Ben-David MA, Griffith KA, Abu-Isa E, et al. External-beam radiotherapy for localized extrahepatic cholangiocarcinoma. Int J Radiat Oncol Biol Phys 2006;66:772-779.

559

185. Zurlo A, Lomax A, Hoess A, et al. The role of proton therapy in the treatment of large irradiation volumes: a comparative planning study of pancreatic and biliary tumors. Int J Radiat Oncol Biol Phys 2000;48:277-288. 186. Fields JN, Emami B. Carcinoma of the extrahepatic biliary system—results of primary and adjuvant radiotherapy. Int J Radiat Oncol Biol Phys 1987;13:331-338. 187. Minsky BD, Kemeny N, Armstrong JG, et al. Extrahepatic biliary system cancer: an update of a combined modality approach. Am J Clin Oncol 1991;14:433-437. 188. Milella M, Salvetti M, Cerrotta A, et al. Interventional radiology and radiotherapy for inoperable cholangiocarcinoma of the extrahepatic bile ducts. Tumori 1998;84:467-471. 189. Montemaggi P, Morganti AG, Dobelbower RR Jr, et al. Role of intraluminal brachytherapy in extrahepatic bile duct and pancreatic cancers: is it just for palliation? Radiology 1996;199:861-866. 190. Golfieri R, Giampalma E, Renzulli M, et al. Unresectable hilar cholangiocarcinoma: multimodality approach with percutaneous treatment associated with radiotherapy and chemotherapy. In Vivo 2006;20:757-760. 191. Buskirk SJ, Gunderson LL, Schild SE, et al. Analysis of failure after curative irradiation of extrahepatic bile duct carcinoma. Ann Surg 1992;215:125-131. 192. Tsujino K, Landry JC, Smith RG, et al. Definitive radiation therapy for extrahepatic bile duct carcinoma. Radiology 1995;196: 275-280. 193. Lee CK, Barrios BR, Bjarnason H. Biliary tree malignancies: the University of Minnesota experience. J Surg Oncol 1997;65:298305. 194. Foo ML, Gunderson LL, Bender CE, et al. External radiation therapy and transcatheter iridium in the treatment of extrahepatic bile duct carcinoma. Int J Radiat Oncol Biol Phys 1997;39:929-935. 195. Bowling TE, Galbraith SM, Hatfield AR, et al. A retrospective comparison of endoscopic stenting alone with stenting and radiotherapy in non-resectable cholangiocarcinoma. Gut 1996;39: 852-855. 196. Eschelman DJ, Shapiro MJ, Bonn J, et al. Malignant biliary duct obstruction: long-term experience with Gianturco stents and combined-modality radiation therapy. Radiology 1996;200: 717-724. 197. Shin HS, Seong J, Kim WC, et al. Combination of external beam irradiation and high-dose-rate intraluminal brachytherapy for inoperable carcinoma of the extrahepatic bile ducts. Int J Radiat Oncol Biol Phys 2003;57:105-112. 198. Leung J, Guiney M, Das R. Intraluminal brachytherapy in bile duct carcinomas. Aust NZ J Surg 1996;66:74-77. 199. Monson JR, Donohue JH, Gunderson LL, et al. Intraoperative radiotherapy for unresectable cholangiocarcinoma—the Mayo Clinic experience. Surg Oncol 1992;1:283-290. 200. Busse PM, Stone MD, Sheldon TA, et al. Intraoperative radiation therapy for biliary tract carcinoma: results of a 5-year experience. Surgery 1989;105:724-733. 201. Todoroki T, Iwasaki Y, Orii K, et al. Resection combined with intraoperative radiation therapy (IORT) for stage IV (TNM) gallbladder carcinoma. World J Surg 1991;15:357-366. 202. Wolkov HB, Graves GM, Won M, et al. Intraoperative radiation therapy of extrahepatic biliary carcinoma: a report of RTOG8506. Am J Clin Oncol 1992;15:323-327. 203. Kurosaki H, Karasawa K, Kaizu T, et al. Intraoperative radiotherapy for resectable extrahepatic bile duct cancer. Int J Radiat Oncol Biol Phys 1999;45:635-638. 204. Rea DJ, Heimbach JK, Rosen CB, et al. Liver transplantation with neoadjuvant chemoradiation is more effective than resection for hilar cholangiocarcinoma. Ann Surg 2005;242:451-458; discussion 458-461.

24 The Colon and Rectum Nora A. Janjan, Marc E. Delclos, Christopher H. Crane, Sunil Krishnan, and Prajnan Das ANATOMY TOLERANCE OF THE COLON AND RECTUM TO IRRADIATION Acute Effects Late Effects COLORECTAL CANCER Etiology and Risk Factors Potential Molecular Targets Screening and Surveillance STAGING CONSIDERATIONS Tumor Classification with Imaging Modalities Endorectal Sonography TREATMENT STRATEGIES FOR RECTAL CANCER Surgical Approaches Combined-Modality Therapy for Locally Advanced Disease

Anatomy The large intestine originates at the cecum in the right iliac fossa and continues in the retroperitoneum as the ascending colon. At the hepatic flexure, it becomes intraperitoneal and crosses the abdomen to again become ­retroperitoneal at the splenic flexure. In the left iliac fossa, it again ­becomes intraperitoneal, curving on itself as the sigmoid colon. The rectum begins at the sacrum, with the anterior portion ­covered by the peritoneum, and terminates at the anus. Given the proximity of the sigmoid colon and rectum to the uterus and prostate, these structures are especially ­subject to acute and late effects of radiation treatment. Within the pelvis, anatomic areas of interest include the mesorectum, the presacral space, and the lateral, inferior, and anterior pelvic spaces. The mesorectum is a cylindrical structure consisting of adipose tissue that ­ contains ­lymphovascular and neural structures, and it is encapsulated by the mesorectal fascia. With cone-shaped tips in the cranial-caudal direction, the mesorectum starts at the level of the sacral promontory at the origin of the superior rectal artery and ends where the levator ani muscle inserts into the rectal wall. The presacral space is a triangular area enclosed posteriorly by the presacral (Waldeyer’s) ­ fascia and anteriorly by the mesorectal fascia. The ­ presacral space contains the median and lateral sacral vessels, the presacral lymphatic chains, the anterior branches of the sacral nerves, and the inferior hypogastric plexus. The lateral pelvic space encompasses the lateral aspect of the mesorectal fascia and includes the lateral pelvic ­sidewalls. The inferior pelvic space consists of the anal triangle of the perineum and contains the anal sphincter complex

560

Radiation Therapy Other Treatment Approaches MANAGEMENT OF TREATMENT-RELATED SIDE EFFECTS Controlling Treatment-Related Diarrhea Late Effects MANAGEMENT OF RECURRENT RECTAL CANCER Salvage Surgery Reirradiation Intraoperative Radiation Therapy Palliative Radiation Therapy MANAGEMENT OF PROXIMAL COLON CANCER Adjuvant Radiation Therapy Adjuvant Chemotherapy SUMMARY

with the surrounding perianal and ischiorectal space. The ­ nterior pelvic space contains all the pelvic organs located a ventrally from the mesorectal space.1

Tolerance of the Colon and Rectum to Irradiation The rectum and rectosigmoid colon are usually considered more tolerant of radiation than the small intestine. However, the fixation of these structures in the pelvis and the high radiation doses often delivered in curative treatment of carcinomas arising in pelvic structures may be expected to lead to more consistent adverse effects than are ­ usually reported. Studies of prostate carcinoma suggest that the volume of rectum irradiated is as important as the ­total doses received.2,3 Two factors are important in establishing the radiation tolerance dose: the rate of repair of ­nonlethally injured cells between fraction doses and the rate of regeneration and maturation of the remaining undifferentiated cells in the crypt bases.4 Small portions of the ­colon and rectum can tolerate total doses in excess of 70 Gy if delivered in fractions of 1.8 to 2.0 Gy or at low dose rates (<1.0 Gy/hr). However, large fractions (10 Gy) can ­produce ­severe effects even if the total dose is as low as 30 Gy.5

Acute Effects Although the threshold dose for early colorectal ­ injury seems to be 10 to 20 Gy (given in daily fractions of 1.5 to 2.0 Gy, 5 days per week),4 doses as small as 5 Gy can produce significant changes in the epithelial ultrastructure

24  The Colon and Rectum

of the small bowel.6 Extending the dose period to 25 days (0.2 Gy/day) allows changes in the villous enterocytes, goblet cells, and lamina propria to return to normal by 3 days after the irradiation is terminated.6 The pathogenesis of early lesions consists of direct injury to the rapidly responding mucosal cells. Mast cell hyperplasia is characteristic of inflammatory and fibrotic intestinal radiation injury. Mast cells allow transforming growth factor β to stimulate a fibrotic reaction.7 Protecting the intestinal mucosa during the early phase of radiation enteropathy, mast cells promote intestinal fibrosis after breakdown of the mucosal barrier, a process that begins with early lesions and continues throughout the appearance of delayed complications.4 Radiation also produces a dose-dependent decrease in the absorption of water and sodium and chloride ions and causes a twofold increase in the secretion of potassium 4 days after exposure.8 Acute effects of radiation on the bowel wall have included reductions of 40% to 70% in tensile strength 3 days after surgery when 10 to 25 Gy was given to a colonic anastomosis in a rat model.9 Although the anastomosis remained patent 7 days after surgery, the adjacent area of colon was at increased risk for perforation. Anastomotic regions seem to be at increased risk during the acute period after intraoperative irradiation.

Late Effects Late radiation effects on the colon have also been evaluated in humans and in animal models. In humans, the threshold dose for delayed lesions is about 50 Gy given in 1.5- to 2.0-Gy fractions.4 In a mouse model, a dose of 32 Gy given to various lengths of the colorectum produced a threshold effect at a length of 10 to 15 mm.10 Re-epithelization ­resulting from the proliferation of epithelial cells outside the crypt on the mucosal surface was complete after a dose of 32 Gy when the irradiated length of colorectum was less than 20 mm. Irradiating more than 20 mm of colorectum, however, led to obstruction from secondary fibrosis. A single dose of 10 Gy to 20 Gy, administered as an intraoperative dose of radiation to the distal limb of a bowel anastomosis, did not threaten anastomotic integrity or bowel function; however, dose-related changes, such as fibrosis, were ­observed between 6 and 12 months after surgery.11

Colorectal Cancer Colorectal cancer is the third most commonly diagnosed cancer in the United States, and its incidence continues to rise over time. In 2001, 98,200 new cases of colon cancer and more than 37,200 new cases of rectal cancer were diagnosed12; the corresponding numbers in 2008 were 108,070 colon cancers and 40,740 rectal cancers.13 Epidemiologically, colorectal cancer has been linked to specific genetic syndromes, breast and ovarian cancer, and dietary consumption patterns. Widespread screening for colorectal cancer is based on a defined risk that benign lesions will be transformed into cancerous ones. Clarification of the influence of these and other possible risk factors will help in establishing and refining guidelines for early detection.

Etiology and Risk Factors Colorectal cancer usually arises from adenomatous polyps that form in a field of epithelial cell hyperproliferation and crypt dysplasia. The multistep process that occurs ­involves

561

several suppressor genes that result in abnormalities of cell regulation. Results from the National Polyp Study demonstrated that two thirds of all polyps removed were ­adenomatous with the potential for malignant transformation.14 That study also indicated that the transformation of a polyp into a cancer, called the adenoma-adenocarcinoma sequence, takes about 5.5 years for an adenomatous polyp larger than 1 cm and about 10 years for smaller polyps. Removing polyps sharply decreases the expected incidence of colorectal cancer.14 Techniques for removing polyps and early carcinomas are described further in the section on “Risk-Reduction Strategies.” Known risk factors for colorectal tumors include a family history of such tumors, consumption of meat, smoking, and consumption of alcohol. Inverse associations have been observed between risk and consumption of a high-fiber diet, use of nonsteroidal anti-inflammatory drugs (NSAIDs), hormone replacement therapy, and physical activity. Incidence rates for colorectal cancer vary by a factor of 20 around the world.15 The risk of colorectal cancer is highly reflective of the environment; incidence rates among immigrants and their descendants can reach those of the country in which they reside within a single generation. This effect is particularly evident among the Hawaiian Japanese, who have the highest rates of colon cancer in the world, even though the rates of colon cancer in Japan are among the lowest. Age, race, and sex were found to be independent risk factors for proximal and distal colorectal cancer in a study of 38,931 cases of colorectal cancer in Illinois.16 In that study, whites were more likely to develop distal colorectal cancer than blacks, and blacks were more likely to develop proximal colorectal cancer than whites.16 A similar study of three geographic areas indicated that age had a greater effect on the development of proximal rather than distal colorectal cancer.17 Many factors, including socioeconomic issues, contribute to the generally poorer survival rates among blacks than whites with colorectal cancer. In a study conducted by the National Surgical Adjuvant Breast and Bowel Project (NSABP), disease stage, number of involved lymph nodes, rate of local recurrence, and other factors were similar between blacks and whites, but blacks had a 21% higher mortality rate.18 According to data from the U.S. Surveillance, ­Epidemiology, and End Results (SEER) program and the ­National Cancer Data Base, most cases of colorectal cancer arise in the proximal colon; the distribution across anatomic sites in 1997 was 15% in the cecum or appendix, 14% in the ascending colon or hepatic flexure, 14% in the transverse colon or splenic flexure and descending colon, 24% in the sigmoid, and 28% in the rectosigmoid and rectum.19 The anatomic distribution of adenomatous polyps differs somewhat, totaling 40% in the proximal colon, 45% in the distal colon, and 15% in the rectum. Approximately 45% of the adenomatous polyps in the colon (but only 12% of those in the rectum) are at least 1 cm in diameter. Although 30% to 50% of individuals have adenomatous polyps, only 6% develop clinically ­diagnosed colorectal cancer.20 Studies of the genetic and molecular basis of colorectal cancer (discussed later) are beginning to reveal why only some polyps undergo ­malignant ­transformation.

562

PART 6  Gastrointestinal Tract

Genetic Factors As a group, familial cancer syndromes make up only about 5% of all types of cancer. A small proportion of common types of cancer, including colorectal, breast, ovarian, and endometrial cancer, are caused by highly penetrant autosomal dominant genes.21 Ethnic differences in cancer ­ incidence and mortality are the result of genetic and environmental factors. Germline mutations in cancer susceptibility genes have been identified in individuals of all races and ethnic groups. The differences in cancer risk among ethnic groups are thought to be the result of founder mutations within cancer-predisposing genes; studies of these genes are expected to help create novel therapies that are based on ­specific mutations.22 Several molecular pathways exist by which colorectal cancer can develop. Among them are the adenomatous ­polyposis coli (APC)–β-catenin–T-cell factor (TCF, a ­ transcriptional activator) pathway and pathways that involve abnormalities of deoxyribonucleic acid (DNA) mismatch repair.15 ­Expression of key genes in any of these pathways can be lost through acquired mutation or hypermethylation, which can account for sporadic cases, or through inheritance, as seems to be the case in inherited syndromes such as familial adenomatous polyposis (FAP) and hereditary nonpolyposis colorectal cancer (HNPCC). Mutations in APC have been identified in up to 80% of families of patients with FAP. In one study of 79 family members of patients with attenuated FAP at the Hereditary Cancer Institute at Creighton University, the specific mutations in APC associated with this syndrome were present in 39 people.23 The same study identified mutations associated with HNPCC in 123 of 316 family members of patients with HNPCC.23 Familial Adenomatous Polyposis. FAP is an autosomal dominant condition caused by a mutation in the APC gene on chromosome 5q. Characterized by multiple polyps (typically more than 100) in the colon and rectum, the lifetime risk of malignancy associated with FAP is 100%, with malignant tumors usually appearing before the age of 40 years. The British Society of Gastroenterology recommends prophylactic colectomy before the age of 25 years, with annual endoscopic screening of the stump.24 The cumulative risk of developing rectal cancer after the conventional prophylactic procedure (i.e., subtotal colectomy with ileorectal anastomosis) is 4% at 5 years after the surgery, 5.6% at 10 years, 7.9% at 15 years, and 25% at 20 years.25 A more extensive procedure, proctocolectomy and ileal pouch–anal anastomosis,26 has the advantage of completely removing the rectal mucosa while maintaining bowel continuity. The mean risk of polyp formation at the pouch-anal anastomosis after this surgery is 18% at 7 years (31% for a stapled anastomosis versus 10% for a hand-sewn anastomosis).27 The rectal stump or pouch should be assessed endoscopically at least once each year; patients older than 45 years who have undergone an ileorectal anastomosis should be screened every 4 months and should consider prophylactic proctectomy, because the cumulative risk of rectal cancer among these patients is 29% by the age of 60 years.28 Hereditary Nonpolyposis Colorectal Cancer. A distinct autosomal dominant syndrome caused by inherited mutations in the mismatch repair genes MLH1, MSH2, MSH6, and PMS2, HNPCC accounts for 15% to 20% of the total number of cases of colorectal cancer.29,30 ­Individuals

with the HNPCC syndrome have a 70% risk of developing colorectal cancer by the age of 70 years. The median age at diagnosis is 45 years. HNPCC is characterized by right-sided lesions, excess synchronous and metachronous lesions, and an increased risk for extracolonic neoplasms.30 People with HNPCC are at increased risk for developing lesions of the endometrium, small bowel, stomach, ­ renal pelvis and ureter, ovary, and skin. Skin lesions such as ­sebaceous ­ adenomas, carcinomas, and keratoacanthomas are also common among patients with HNPCC. However, no specific histopathologic characteristics have been identified that can distinguish HNPCC from sporadic colorectal cancer, with the possible exception of tumors that are right-sided, undifferentiated, or of the signet ring cell type occurring in young people. Because of the early difficulty in distinguishing ­genetic patterns indicative of HNPCC from those associated with sporadic colorectal cancer, the International Collaborative Group on Hereditary Nonpolyposis Colorectal ­Cancer31 in 1991 established minimal clinical criteria for recruiting ­patients with HNPCC for collaborative studies. These criteria, known as the Amsterdam criteria, consist of (1) having at least three relatives with histologically verified colorectal cancer, one of whom must be a first-­degree ­relative of the other two; (2) having colorectal cancer ­ occurring in at least two successive generations; and (3) having colorectal ­cancer diagnosed in at least one individual before the age of 50 years. Although these criteria were valuable in ­identifying kindreds, they do not account for small kindreds or tumors at extracolonic sites. Moreover, only about one half of the patients with colorectal cancer who meet the Amsterdam criteria have mutations ­associated with HNPCC; conversely, mutations typical of HNPCC are identified in only 8% of families with a strong positive history of colon cancer but do not meet the Amsterdam criteria. In an effort to identify clinicopathologic criteria that may aid in the early identification of individuals with HNPCC, a working group convened by the ­National Cancer Institute in 1996 expanded the Amsterdam criteria to create the Bethesda Guidelines (Box 24-1).30 Screening recommendations for individuals with HNPCC syndrome include biennial total colonoscopy beginning at age 25 years or 5 years before the youngest family member was diagnosed with cancer. Although some investigators advocate prophylactic colectomy for patients with HNPCC, evidence in support of this practice is lacking. Subtotal colectomy remains the treatment of choice for patients with HNPCC presenting with a cancer, but this has not been proved to be superior to segmental ­resection and colonoscopic surveillance.21 The risk of developing metachronous cancer after segmental resection has been estimated at 40% within 7 years32; in one study, 11% of patients with rectal cancer who had undergone subtotal colectomy developed a metachronous cancer, diagnosed at a mean of 13 years after surgery.33

Nonsteroidal Anti-inflammatory Drugs Use of NSAIDs has been reported to decrease the risk of developing colorectal cancer through the inhibition of ­cyclooxygenase 2 (COX-2).34,35 However, NSAIDs also have deleterious effects on the small bowel and colonic mucosa, including ulceration and strictures of the jejunum, ileum, and colon. Clinically, NSAIDs can cause diarrhea,

24  The Colon and Rectum

BOX 24-1.  The Bethesda Guidelines* 1. Individuals who meet the Amsterdam Criteria 2. Individuals with two HNPCC-related cancerous ­tumors, including synchronous and metachronous ­colorectal cancer or associated extracolonic ­cancer (i.e., ­endometrial, ovarian, gastric, ­hepatobiliary, small bowel, or transitional cell cancer of the renal pelvis or ureter) 3. Individuals with colorectal cancer and a first-degree relative with colorectal cancer and/or ­HNPCC-related extracolonic cancer and/or a colorectal adenoma a. One of the cancers was diagnosed at <45 years b. One of the adenomas was diagnosed at <40 years 4. Individuals with colorectal cancer or endometrial cancer diagnosed at <45 years 5. Individuals with right-sided colorectal cancer with an undifferentiated histopathologic pattern ­diagnosed at <45 years 6. Individuals with a signet ring cell type of colorectal cancer (i.e., tumor consists of at least 50% signet ring cells) diagnosed at age <45 years 7. Individuals with adenomas diagnosed at age <40 years HNPCC, hereditary nonpolyposis colorectal cancer. *Tumors from individuals who meet these criteria should be tested for microsatellite instability. Modified from Rodriguez-Bigas MA, Boland CR, Hamilton SR, et al. A National Cancer Institute Workshop on Hereditary Nonpolyposis Colorectal Cancer Syndrome: meeting highlights and Bethesda guidelines. J Natl Cancer Inst 1997;89:1758-1762.

abdominal pain, blood and protein loss, colitis, and perforation,36 and NSAIDs should not be used by people with ulcerative colitis or Crohn’s disease.

Hormone Replacement Therapy The use of hormone replacement therapy seems to be ­associated with a reduced risk of colon cancer.15,37,38 A meta-analysis of 18 epidemiologic studies of hormone ­replacement therapy in postmenopausal women showed a 20% reduction in the risk of colorectal cancer.37 ­ Several hypotheses have been proposed to explain this effect. Issa and colleagues39 found that colon tumors had abnormal methylation patterns in the estrogen receptor gene and postulated that hypermethylation of and a reduction in expression of that gene, both of which ­increase with age, result in deregulated growth of the colonic mucosa.39 ­Alternatively, exogenous ­ estrogens decrease bile acid synthesis by up to 15%; this is important because bile acids, which are absorbed in the proximal ­ colon, are thought to promote malignant changes in the colonic epithelium. High fecal bile acid concentrations have been linked to recurrence of colon polyps; according to one study, having bile acid concentrations in the upper quartile was associated with a threefold increase in the risk of polyp formation.37 Estrogen may also ­reduce levels of the mitogen insulin-like growth factor 1,15 which is required by some cells to progress from G1 to S phase.

Dietary Factors Evidence from case-control and cohort studies has supported strong links between diet and risk of colorectal cancer.15 In a case-controlled study of 402 cases of colorectal cancer

563

and 668 controls, high intake of total fat and saturated fat and low intake of dietary fiber (especially from vegetable sources), calcium, and vitamins A and E were associated with the development of colorectal cancer.40 Results from the Australian Polyp Prevention Project and other studies have shown that wheat bran fiber, which reduces fecal bile acid concentrations, reduces the incidence of large colorectal adenomas, which is thought to be a critical step in the development of malignancies.41-43 In another study, dietary changes involving increased intake of calcium and decreased consumption of fat produced reductions in colonic epithelial cell proliferation and the size of the proliferative compartment, and it restored markers of normal cellular differentiation such as acidic mucin, cytokeratin AE1 distribution, and nuclear size.44 Vitamins A and D, fatty acids, and zinc may affect colon cancer progression through their direct effects on gene transcription.45 Nitrosamines and heterocyclic amines, bloodborne carcinogens from tobacco smoke and meat, have been linked to mutations in APC or KRAS.15 Consumption of alcohol may contribute to carcinogenesis by reducing folate levels, which can result in DNA damage. Consumption of vegetables has a protective effect, because vegetables are a source of folate, antioxidants, and inducers of detoxifying enzymes.15,46 Folate is important for DNA integrity, antioxidants reduce DNA damage, and fiber affects short-chain fatty acids, which themselves influence apoptosis. It is also possible that polymorphisms in the genes coding for methylene-tetrahydrofolate reductase and N-acetyltransferases 1 and 2 affect the metabolism of folate and heterocyclic amines.15

Prior Irradiation Previous radiation exposure can increase the risk of developing colorectal cancer. Among men treated with radiation for prostate cancer, the overall risk of a second malignancy was 1 in 290, and that risk increased to 1 in 70 for those surviving at least 10 years.47 The increase in relative risk in that study was 15% at 5 or more years and 34% among those who survived 10 years or longer.47

Inflammatory Bowel Disease: Ulcerative Colitis and Crohn’s Disease The risk of colorectal cancer is increased and the radiation tolerance level reduced among patients with inflammatory bowel diseases such as ulcerative colitis or colonic Crohn’s disease. Ulcerative colitis and Crohn’s disease, both considered forms of autoimmune disease, have been linked to ­abnormalities in chromosomes 3, 6, 7, 12, and 16 that result in a dysregulated mucosal response that leads to chronic mucosal inflammation. Ulcerative colitis causes superficial inflammation of the mucosa that can cause ulcer formation in the colon. Most commonly, ulcerative colitis affects the rectal and sigmoid areas and progresses proximally.48 Pathologically, ulcerative colitis is characterized by acute inflammatory events with neutrophils and crypt abscesses. A meta-analysis published in 2001 revealed that the risk of developing colorectal cancer among individuals with ­ulcerative colitis was 2% at 10 years, 8% at 20 years, and 18% at 30 years.49 Although no conclusive evidence has been found to indicate that surveillance colonoscopy ­reduces colorectal cancer mortality in patients with extensive colitis,50 surveillance does allow cancer to be detected

564

PART 6  Gastrointestinal Tract

at earlier stages, which might be expected to confer a better prognosis. In one study, the 5-year cancer-free survival rates for 41 patients with colitis were 77% for the 19 people who had been under colonoscopic surveillance and 36% for the 22 patients who had not.51 Crohn’s disease can occur anywhere in the gastrointestinal tract, but it is particularly common in the ­ terminal ­ileum and ascending colon.48 Often, Crohn’s disease affects discontinuous areas, with areas of inflammation dispersed between normal mucosa throughout the gastrointestinal tract. Lymphocytes and macrophages accumulate within the intestinal crypts, and protogranulomas are seen. Crohn’s disease produces transmural inflammation of the bowel resulting in the formation of strictures, fistulas, and abscesses. Crohn’s disease affects about 400,000 people in the United States.

Potential Molecular Targets The molecular basis of colorectal cancer, like that of many other forms of cancer, has been the object of intense study during the past 10 years. The expression or lack of expression of various molecules in tumors and perturbations in associated cellular signaling pathways are being assessed for their potential as diagnostic aids for identifying ­preneoplastic lesions, malignant tumors, and metastases; as biomarkers of treatment effectiveness; and as markers of prognosis. Polymorphisms in the gene for thymidylate synthase (TYMS), for example, have been linked with susceptibility to the fluoropyrimidine 5-fluorouracil (5-FU); tumors that express low levels of this enzyme are more likely to ­respond to preoperative treatment with 5-FU than tumors that express high levels of TYMS.52 At least one group has found that specific polymorphisms in TYMS are predictive of tumor downstaging and improved disease-free survival in patients with rectal cancer treated with preoperative 5-FU–based chemoradiation therapy and surgery.53 Other pathways being studied for their role in cancer development and progression include those of the COX enzymes, the epidermal growth factor receptor (EGFR), and the ­vascular endothelial growth factors (VEGFs) and their ­receptors. Additional information on these and other ­molecular-level studies is included in Chapters 5 and 7; studies germane to colorectal cancer are reviewed briefly in the following sections.

Cyclooxygenase 2 The gene for COX-2 (PTGS2) is overexpressed in some cases of sporadic and HNPCC-type colonic adenomas.34,54 COX enzymes participate in the synthesis of prostaglandins from arachidonic acid; their peroxidase activity also catalyzes the conversion of procarcinogens to proximate carcinogens.35 The increased production of prostaglandins in inflammation and possibly by tumor cells in part results from the cytokine-mediated induction of COX-2. COX-2 is thought to contribute to tumorigenesis through several mechanisms, such as inhibiting apoptosis, increasing angiogenesis, modulating inflammation and immune suppression, and converting procarcinogens to carcinogens. Experimental evidence also suggests that ­ chemotherapy and radiation therapy can induce the ­expression of COX-2 by tumor cells and that cross-talk takes place between the EGFR and COX-2 signaling pathways. One group of investigators sought to analyze the expression of COX-2 and

EGFR in rectal tumors before and after chemoradiation therapy and to correlate the results with evidence of tumor downstaging and survival.55 Archived biopsy and surgical specimens were analyzed by immunohistochemical staining for COX-2 and EGFR. COX-2 expression was found in 19 of 20 pretreatment samples and in all 17 surgical specimens; EGFR expression, in contrast, was observed in only one half of the samples before treatment and was not induced by chemoradiation therapy. The investigators concluded that some evidence of COX-2 induction by chemoradiation therapy had been found, but the prognostic significance of this induction remains to be defined in a larger cohort, because no differences in survival were observed between groups that did or did not show COX-2 induction.55

Epidermal Growth Factor Receptor EGFR is a transmembrane protein, activation of which triggers an intracellular cascade of signaling that controls cell division, migration, and differentiation. Major signaling pathways interacting with the EGFR pathways include the RAS-RAF mitogen-activated protein kinase (MAPK) pathway, the phosphatidylinositol-3-kinase (PI3K)/AKT pathway, and the signal transducer and activator of transcription (STAT) pathway. Activation of these pathways regulates tumor cell survival, proliferation, and apoptosis. EGFR expression has been associated with radioresistance, aggressive tumor phenotypes, and lower locoregional control and survival rates, especially in head and neck cancer.56,57 In one study of rectal cancer, 92 pretreatment biopsy specimens were assessed by immunohistochemical staining for EGFR expression, and the prognostic value was assessed in terms of the level of that expression. EGFR was expressed in 65 specimens (71%); expression levels were low to none in 83 cases and high in 9. High levels of EGFR expression was predictive of shorter overall survival (P = .013), disease-free survival (P = .002), and distant ­ metastasis–free survival (P = .003).58 However, EGFR ­ expression did not predict whether a particular tumor would or would not respond to preoperative ­chemotherapy.58 In a different study, the presence of a particular ­polymorphism in the SP1 promoter site for the EGFR gene, −216 G/T, was found to correlate with response to preoperative ­chemoradiation therapy for locally advanced rectal ­cancer.59 The apparent involvement of EGFR in many cancer­related processes has led to efforts to develop agents that block or circumvent its actions. One such agent, cetuximab, works by competitively inhibiting the binding of EGFR ligands (e.g., EGF, transforming growth factor α), which blocks downstream signaling so as to inhibit cell proliferation and angiogenesis, induce apoptosis, and presumably prevent metastasis. Cetuximab may also influence antibody-dependent cell-mediated cytotoxicity, an immune effector mechanism in which antigen-specific antibodies direct immune effector cells to kill antigen-expressing cancer cells. One group has suggested that polymorphisms in two immunoglobulin receptors may be useful for predicting whether particular cases of metastatic colorectal ­cancer will respond to cetuximab.60 A molecular marker such as this would be quite useful because only about 10% of colorectal cancers respond to cetuximab as single-agent therapy. Moreover, the expression of such markers may

24  The Colon and Rectum WOMEN

MEN 70

70

White-incidence

60

60 Black-incidence

40 Black-mortality 30

White-mortality

20

50 Rate per 100,000

50 Rate per 100,000

565

40 White-incidence 30 Black-mortality 20

10

10

Black-incidence

White-mortality

0

0 73 75 77 79 81 83 85 87 89 91 93 95 97

73 75 77 79 81 83 85 87 89 91 93 95 97

Year of diagnosis/death

Year of diagnosis/death

FIGURE 24-1.  Cancer incidence by sex and ethnic origin. (From Ries LAG, Wingo PA, Miller DS, et al. The annual report to the nation on the status of cancer, 1973-1997, with a special section on colorectal cancer. Cancer 2000;88:2398-2424.)

not be limited to tumor tissue; another group showed in a retrospective analysis of 67 patients that disease recurrence after adjuvant chemoradiation therapy could be predicted by expression levels of VEGF and EGFR in normal tissue adjacent to the tumor and by the expression of a DNA-­repair factor, RAD51, in tumor tissue. These findings further suggest that the tumor microenvironment may be more important in the development of recurrence of rectal ­tumors than previously thought.61

Antiangiogenic Factors Several members of the VEGF family have been found to have different effects on angiogenesis, a process essen­ tial for tumor growth and progression. A recombinant ­humanized monoclonal antibody with high binding specificity for VEGF, bevacizumab (Avastin, Genentech, South San Francisco, CA), has been evaluated for its antiangiogenic properties in a variety of preclinical and clinical studies. ­ Bevacizumab has demonstrated efficacy against several solid tumors, including colorectal cancer.62,63 A ­combined analysis of three randomized clinical studies of ­ bevacizumab in combination with 5-FU and leucovorin for ­ previously ­ untreated metastatic colorectal cancer indicated that the addition of bevacizumab extended the duration of ­progression-free and overall survival over 5-FU and leucovorin alone.62 Another large trial by the Eastern ­Cooperative Oncology Group (ECOG E3200) showed that adding ­bevacizumab to oxaliplatin, 5-FU, and leucovorin improved survival duration for patients with previously treated metastatic colorectal cancer.63 Unfortunately, markers that predict before ­ treatment who will respond to bevacizumab have yet to be ­identified. In one such attempt, tissue samples from 278 ­ patients who received first-line therapy with irinotecan, 5-FU, and

l­eucovorin, with or without bevacizumab, were ­examined for their expression of the vascular biomarkers VEGF, thrombospondin 2, and microvessel density. None of these markers predicted response, although the addition of ­ bevacizumab did have a positive influence on overall survival.64

Screening and Surveillance Screening procedures have had a significant effect on the detection of colorectal cancer. Over the second half of the 1900s, the incidence of colon cancer rose sharply, especially among men and blacks. Trends in the rates of colorectal cancer were specifically evaluated between 1975 and 1994 from more than 220,000 cases of colorectal cancer in the United States using data from the U.S. SEER program. In the late 1970s, the incidence of metastatic ­disease at ­diagnosis began to decline among whites; 10 years later, ­declines were also seen in the incidence of regional disease.65 These trends correlated with increased rates of screening for colorectal cancer during the mid1980s. The numbers of cases of cancer found in the ­distal colon and rectum decreased more significantly because screening initiatives ­ focused on the use of the least invasive techniques.15,65 Consequently, numbers of proximal colon cancer cases were higher than distal colon and rectal cancer cases throughout the study period.65 Most cases of cancer arose in the proximal colon (39%), followed by the distal colon (30%) and rectum (29%). Localized disease was ­ diagnosed in 37%, regional disease in 37%, and distant ­disease in 20%. More commonly, localized disease was diagnosed in the distal colon and rectum, and regional disease was found in proximal colon tumors. The incidence of cancer in the proximal colon was significantly higher among blacks than whites (Fig. 24-1).66

566

PART 6  Gastrointestinal Tract

Despite its success in cancer detection, screening, especially genetic testing, is fraught with difficulties, including patients’ concerns about insurability, employment discrimination, denial, and fear.23,67 In a study of 208 members of four extended families with HNPCC, 43% elected to undergo testing, and 57% declined. Of the 139 subjects (67%) who consented to a telephone interview, 55 (40%) did not wish to receive their test results. Among the 84 people who did accept the test results, 35 (42%) found out that they had HNPCC-associated mutations.68 Those who wished to receive the test information were better educated and were more likely to have participated in a previous genetic linkage study than those who declined to receive their test results. The presence of depressive symptoms resulted in a fourfold decrease in acceptance of test results among women and a slight decrease among men, although equal numbers of men and women underwent the testing.68 A survey among first-degree relatives of patients with colorectal cancer revealed that perceived barriers to screening included lack of public awareness, lack of physician recommendation for screening procedures, questionable efficacy of the screening procedures, fear of positive results, embarrassment, and discomfort and uncleanliness associated with the test.69 For these reasons, only 14% of those individuals had ever been tested for fecal occult blood; 59% had never had a digital rectal examination, 73% had never had a flexible sigmoidoscopy, and 64% had never had a colonoscopy. Nevertheless, this survey group understood that screening would reduce the chance of dying of colorectal cancer and that undergoing screening procedures would reassure them of being free of cancer. In another study, 199 family members of patients with HNPCC underwent genetic testing for the presence of known mutations.54 Before their mutation status was disclosed, subjects were asked about prior screening tests; only 55% had undergone at least one colonoscopy, and 47% had undergone at least one other screening test for colon cancer.54 After their mutation status was revealed, 95% of those who carried the mutation were willing to undergo lifelong surveillance (annual or semiannual colonoscopy beginning at age 20 years),70 and 70% were willing to undergo prophylactic colectomy. A cross-sectional study of 98 patients with colorectal cancer showed that 61% reported being worried about their relatives having colorectal cancer, and 64% were concerned about the risk of genetic transmission.71 Although 81% were unaware that genetic testing for colorectal cancer was available, 72% reported being willing to take such a test if it were available to them.71 From the physician’s point of view, the difficulties associated with screening for inherited cancer can be classified in terms of confidentiality, coordination, and communication. It is often difficult for physicians to distinguish ­between cases that are sporadic, familial (i.e., at least one affected relative but no clear mendelian pattern or known mutation), or inherited (i.e., single gene mutation playing a causative role).72 However, if an inherited syndrome is suspected, referral should be made to a center for genetic assessment. Another resource is the Online Mendelian Inheritance in Man (OMIM) Database (http://www.ncbi. nlm.nih.gov/sites/entrez?db=omim).73 Of the 35 familial cancer-­predisposing disorders identified,74 tumors of the

colon or rectum have been reported in six: FAP, Bloom’s syndrome, breast/ovarian cancer (BRCA1), breast cancer/ other (BRCA2), HNPCC, and Peutz-Jeghers syndrome. At least 10 types of hereditary colon cancer have been ­identified.73

Obligation to Inform All of the physicians involved in the care of a patient ­assume the responsibility of documenting that the possibility of a genetic link to the disease was discussed with that patient.75 Without this knowledge, patients cannot inform family members regarding their potential risk and the need to seek medical evaluation. Various court cases involving familial cases of breast, ovarian, and colon cancer have set precedents regarding the obligation to inform. The legal precedents are based on the experience with multiple endocrine neoplasia (MEN) type 2 syndrome, in which identification of the RET proto-oncogene can lead to a precise diagnosis of susceptibility to medullary thyroid carcinoma. It is now considered mandatory to offer testing to at-risk family members, including children as young as age 3 years, so that they can undergo prophylactic thyroidectomy.23

Risk-Reduction Strategies Risk-reduction strategies can yield great improvements in life expectancy for people with premalignant or malignant colorectal lesions. One decision-analysis model76 ­suggested that prophylactic proctocolectomy at age 25 years for patients with FAP could increase their life expectancy by 14 years over that of patients with FAP without such an ­intervention; however, colectomy performed at the time of cancer diagnosis produced only minimal benefit. As the age of the patient increased, surveillance proved as ­beneficial as colectomy. However, when health-related quality of life was considered, surveillance led to the greatest quality­adjusted life expectancy benefit.76 Other effective risk-reduction strategies are not as drastic as partial or total colectomy. For example, removal of polyps during screening or surveillance procedures can greatly reduce the expected incidence of colorectal cancer14 and can be done by using minimally invasive procedures under endoscopic guidance. Techniques for removing large rectal adenomas or early carcinomas include transanal endoscopic microsurgery, endoscopic mucosal resection, and snare polypectomy.77 Endoscopic microsurgery is a minimally invasive surgical technique in which the plane of dissection extends through the muscularis propria into the surrounding adipose tissue; the remaining defect is typically closed by using a continuous suture. Transanal endoscopic microsurgery is most commonly used to resect large, benign rectal adenomas. ­Microscopic involvement of the resection margin in such tumors is significantly associated with recurrence, and overall recurrence rates range from 2% to 11% at 24 months.78 Endoscopic mucosal resection is used to excise large sessile polyps that are not amenable to conventional snare polypectomy. In this technique, a methylene blue solution injected into the submucosa around the polyp causes the lesion to lift away from the muscularis propria; the lesion is then resected, wholly or in pieces, with a conventional polypectomy snare. Endoscopic mucosal resection should be used only for benign polyps and not for lesions that are indurated, ulcerated, friable, or fail to “lift” in response to a

24  The Colon and Rectum

submucosal dye injection, because these characteristics are associated with malignancy. Recurrence rates at a median 12 to 24 months after resection of benign sessile polyps in the colon and rectum usually range from 17% to 21%, and invasive adenocarcinoma is found in 12% to 18% of such lesions.79,80 In another study in which endoscopic mucosal resection was used for lateral spreading tumors of the rectum, recurrence rates varied with the anatomic site of the lesion: 6% for upper rectal lesions, 17% for lower rectal lesions, and 80% for very low lesions involving the pectinate line.81 Follow-up colonoscopy is recommended at 3 months and 6 months after the procedure, because early regrowth can be treated with argon plasma coagulation or conventional surgery.79 Snare polypectomy is used for polyps with narrow stalks that can be completely excised. Risk stratification and recommended follow-up depend on the number and size of the polyps. People at high risk are those who have five or more adenomas or have more than three adenomas when one or more is larger than 1 cm. A follow-up colonoscopy in 12 months is recommended for these individuals,82 for whom the cumulative risk of developing invasive cancer without polypectomy is estimated as 24% at 20 years.83 Patients at low risk are those who have undergone removal of one or two adenomas that are smaller than 1 cm; these patients should undergo repeat colonoscopy in 5 years.82 Patients at intermediate risk are those who have three to four adenomas or have one adenoma larger than 1 cm. These patients should be offered follow-up colonoscopy at 3 years.82

Screening Procedures Procedures used for colorectal cancer screening include the fecal occult blood test, colonoscopy, sigmoidoscopy, and proctoscopy. Each technique has advantages and disadvantages; for example, fecal occult blood tests are the least invasive, but their sensitivity and specificity are questionable. Endoscopic procedures are invasive, require preparation, and are often costly. Proctoscopy, although beneficial for detecting rectal tumors, cannot detect tumors in the ­colon. Sigmoidoscopy allows examination of a larger portion of the colon than does proctoscopy, but results from one prospective trial of 114 patients indicated that sigmoidoscopy detected abnormalities in only 28% of ­ patients, ­ although 41% of these patients were found to have a proximal ­adenoma.84 Computed tomography (CT) scanning of the colon, or virtual colonoscopy, may be an alternative to endoscopic colonoscopy, but the procedure still involves insufflation of air and the attendant discomfort. More ­experience is needed with the technique to resolve issues ­regarding accuracy.85,86 The need to educate primary care physicians in cancer screening is critical. In a survey of 168 internal medicine residents in four accredited programs who completed questionnaires regarding colorectal cancer screening and the use of fecal occult blood tests, 71% correctly identified 50 years as the age at which an asymptomatic patient at average risk should begin colorectal cancer screening as recommended in national guidelines.87 Only 64% routinely performed fecal occult blood testing at outpatient clinic visits, 79% performed a digital rectal examination, and only 29% recommended colonoscopy to evaluate the appearance of blood in the feces. The residents’ level of training

567

influenced their use of annual fecal occult blood tests and flexible sigmoidoscopy (58% among those in their first or second year of training versus 80% among those in their third or fourth years of training) and the practice of following a positive fecal occult blood test with colonoscopy (19% versus 48%). The Minnesota Colon Cancer Control Study saw a ­decrease in mortality over a 13-year period that was attributed to early detection of colorectal cancer.88 In that study, patients underwent an expanded version of the annual ­fecal occult blood test each year during that 13-year period. In that test, two fecal samples from each of three consecutive stools were placed on a test card. If any one sample was positive, a diagnostic workup that included a colonoscopy was performed. The sensitivity of this procedure was found to be about 90%; it produced a cure rate for early colorectal cancer of about 90%, and it reduced the mortality from colorectal cancer by one third among patients older than 50 years. Because the incidence of colorectal cancer did not change over the 13-year study period, the decrease in mortality was attributed to its early detection. The number of slides that were positive among the six slides in the set was linked to the presence of polyps and colorectal cancer.88 Screening specifically benefits patients with known colorectal abnormalities. In one study, 204 patients with a history of adenomatous polyps underwent screening and two follow-up colonoscopies.89 During the median total surveillance period of 55 months, 493 adenomas and one cancer were detected. The first follow-up colonoscopy was normal in 91 patients (45%); among these patients, the second colonoscopy was also normal in 57 (63%), but ­ adenomas were identified in 34 (37%). Of the 113 ­patients (55%) who were found to have adenomas on the first ­follow-up colonoscopy, only 46 (40%) had normal results on a subsequent colonoscopy; the other 67 (60%) had ­adenomas on the second follow-up colonoscopy. The risk of subsequent adenomas was significantly reduced if the findings at interim colonoscopy were normal; at 3 years, only 15% of those whose interim colonoscopy ­ result was normal had adenomas, compared with 40% of those for whom the ­interim colonoscopy result was positive. ­However, by 4 years of follow-up, the risk of adenoma was the same ­regardless of the findings from the interim colonoscopy. Adenomas found after a colonoscopy with normal results were dispersed throughout the colon in 28 cases and ­limited to the rectosigmoid in only 6 cases. A significant reduction in the risk of a subsequent diagnosis of colon cancer was seen when the follow-up colonoscopy result was normal. These findings led the authors of this study to recommend surveillance colonoscopy at 4- to 5-year intervals. The importance of screening for colorectal cancer in ­asymptomatic individuals cannot be overemphasized. An autopsy series of 1105 cases between 1986 and 1995 ­revealed 250 cases of cancer, of which 111 tumors in 100 patients had been undiagnosed or misdiagnosed.90 The immediate cause of death was attributable to the cancer in 57 patients. In this study, as in other reports, the discordance between clinical and autopsy diagnoses of cancer was 44%. Of the undiagnosed tumors, only 33% were clinically suspected, and 23% of undiagnosed cancerous tumors were in the gastrointestinal tract. In another study in which 186 premenopausal women with iron-deficiency anemia underwent colonic endoscopy,91 a “clinically important”

568

PART 6  Gastrointestinal Tract

lesion was found in 23 cases, of which six were colon cancer.91 Independent predictors for having a gastrointestinal lesion identified by endoscopy included having positive findings on the fecal occult blood test, a hemoglobin level of less than 10 g/dL, and abdominal symptoms.91

Screening Guidelines and Cost-Benefit ­Considerations Evidence-based guidelines92 recommend that asymptomatic individuals at average risk for colorectal cancer ­undergo screening beginning at age 50 years and that patients at higher risk undergo more intensive screening and follow-up. Adopting these principles could reduce the mortality from colorectal cancer by more than 50%. The best technique or techniques to use for that screening, however, ­remains controversial. According to one estimate, using biennial guaiac fecal occult blood testing to screen populations at average risk for colorectal cancer can reduce cancer-specific mortality rates by 15%; use of flexible sigmoidoscopy can reduce those rates by 50% to 80%.77 In the Minnesota trial mentioned earlier, annual fecal occult blood testing found 260 of the first 281 cancer cases, for a sensitivity rate of 92.5%.88 Reductions in mortality rates have been attributed in part to the use of colonoscopy to evaluate positive findings from fecal occult blood tests; only 20% of the observed ­benefit resulted from colonoscopy after false-positive screening tests.88,93 Helm and colleagues,94 in analyzing projected outcomes from three large randomized ­ controlled trials, found that of the more than 1 million colorectal cancers ­expected over a 10-year period, 60% would be fatal ­without screening. If 60% to 67% of people eligible for screening actually were screened, a fecal occult blood ­ testing program would detect 30% of known colorectal cancers and save more than 100,000 lives over that 10-year period. The total cost of such screening would be $3 billion to $4 billion over 10 years, representing a cost of $2500 per ­life-year saved. The authors of this study concluded that the effectiveness of fecal occult blood testing would be limited to saving the lives of roughly 15% of people who would die from colorectal cancer in the first 10 years ­after beginning mass screening. Although fecal occult blood testing can be cost-effective, its limitations underscore the need for more effective noninvasive screening tools.94 The cost-effectiveness of fecal occult blood ­ testing, flexible sigmoidoscopy, and colonoscopy for screening has been evaluated by using Markov modeling.95 At less than $20,000 per life-year saved, all strategies were found to be cost-effective compared with no screening. The most ­effective strategies were twice-lifetime colonoscopy or ­flexible sigmoidoscopy combined with fecal occult blood testing. Assuming perfect compliance, flexible sigmoidoscopy combined with fecal occult blood testing would be slightly more effective than twice-lifetime colonoscopy at the ages of 50 and 60 years but substantially more ­expensive, with an incremental cost-effectiveness of $390,000 per additional life-year saved. However, compliance with the primary screening test or follow-up colonoscopy does ­affect screening decisions. If the rate of follow-up colonoscopy for ­ polyps is less than 75%, even once-in-a-lifetime ­colonoscopy is preferred over most combinations of flexible sigmoidoscopy and fecal occult blood testing. These investigators concluded that two colonoscopies, one at age 50 and

one at age 60, is the preferred option, but they did observe that screening recommendations should be tailored to the compliance levels that can be achieved in various practice settings.95 As for the selection of sigmoidoscopy or colonoscopy, cohort and case-control studies have shown that use of flexible sigmoidoscopy as the only screening procedure reduces the risk of death from distal colorectal cancer by about 70%.93 Although the “miss” rate for flexible sigmoidoscopy can be as high as 50% because of incomplete bowel preparations and technical limitations, combining this test with fecal occult blood testing overcomes many of the ­limitations of both tests used separately.93 Although the number of screening colonoscopies performed in the United States has been increasing as the baby boom generation ages, relatively little information is available on the benefit of colon cancer screening for ­patients between 70 and 94 years old, particularly those with ­significant comorbid conditions. In one study that explicitly examined the probability of colorectal cancer screening–­related benefits and harms in older ­patients of various health states and life expectancies,96 the ­investigators developed a model to estimate the risk of dying of colorectal cancer, the potential reduction in mortality with colorectal cancer screening, the number of colonoscopies needed, and the potential for colonoscopy-related complications in ­ elderly patients at average risk for colorectal cancer based on their underlying health status. The three strategies assessed were annual fecal ­occult blood tests, flexible sigmoidoscopy ­ every 5 years, and colonoscopy ­ every 10 years. Each strategy was assessed in terms of the number needed to screen to prevent one cancer-related death and the number needed to encounter one ­ screening-related ­complication. The potential benefit from screening was found to vary widely with age, life expectancy, and screening modality. One cancer-related death would be prevented by screening 42 healthy men between 70 and 74 years old with colonoscopy; 178 healthy women between 70 and 74 years old with fecal occult blood tests; 431 women between 75 and 79 years old in poor health with colonoscopy; or 945 men between 80 and 84 years in average health with fecal ­ occult blood tests. Colonoscopy screening had the greatest benefit but the highest risk of complications. Nevertheless, at all ages and life expectancies, the potential ­reduction in mortality from screening outweighed the risk of colonoscopy-related death.96 However, given the ­assumption in this study that a reduction in cancer-related mortality would not be seen for at least 5 years for any screening modality, colorectal cancer screening is unlikely to benefit individuals whose life expectancy is less than 5 years. Fecal occult blood testing and flexible sigmoidoscopy in the United States are considered cost-effective because of the corresponding savings in the cost of treating cancer. In the University of Minnesota study, the cost of screening was calculated to be between $15,000 and $25,000 per year of life saved; this cost is well below the $40,000 ­considered by the federal government to be cost-effective. For comparison, the cost of screening mammography is about $30,000 per year of life saved. The cost of treating colorectal cancer is second only to that of breast cancer, totaling more than $6.5 billion per

24  The Colon and Rectum

year. The National Institutes of Health estimated that the total direct medical cost of cancer care in the ­United States in 2005 was $74 billion; that cost increased to $209.9 billion when indirect morbidity and mortality costs (lost productivity) were accounted for.97 The average cost for treatment of colorectal cancer is about $50,000 per ­patient. As the results of screening initiatives become more mature, cost-effectiveness analysis, particularly for the more invasive screening procedures, will become increasingly ­important. The American College of Gastroenterology updated the Agency for Health Care Policy and Research Guidelines for colorectal cancer screening in 2000. For the averagerisk individual, defined as 50 years old or older with no risk factors for colorectal cancer except for age, the preferred screening strategy is a colonoscopy every 10 years.98 Colonoscopy confers several advantages over flexible sigmoidoscopy, particularly the ability to evaluate the entire colon and the ability to remove adenomas during the same procedure. Roughly one half of all patents with colon cancer have disease proximal to the splenic flexure, and the incidence of proximal colon cancer has increased over the past 30 years.99 Nevertheless, when resources are limited, the recommendation is to use flexible sigmoidoscopy every 5 years with annual fecal occult blood testing. However, patients whose fecal occult blood test results are positive should undergo colonoscopy. Although reports of virtual (CT-based) colonoscopy have indicated that polyps as small as 6 mm can be detected with this technique, more experience is needed before virtual colonoscopy can be recommended as an alternative to colonoscopy. Individuals considered to be at high risk for developing colorectal cancer include those with FAP, HNPCC, or a family history of colon polyps or cancer that does not meet the clinical criteria for these syndromes. According to the American College of Gastroenterology, about 10% of people in the United States will eventually have an immediate family member with colon cancer, and having a family history of colon cancer doubles a person’s risk of this cancer.98 Other groups at high risk include those with a history of adenomatous polyps, ulcerative colitis, or Crohn’s disease. Screening for FAP should begin in childhood, with a sigmoidoscopy every 1 to 2 years beginning at 10 to 12 years of age. After polyps appear, a colectomy should be planned. Individuals diagnosed at a later age and ­unscreened individuals and relatives should have a full colonoscopy; screening should continue until 40 years of age, when standard-risk screening procedures can begin. Even if the criteria for HNPCC are not met, individuals with a strong family ­ history of colon polyps or cancer should begin screening at age 40 or at 10 years earlier than the age of the youngest affected relative, whichever is first. Colonoscopy should be performed every 10 years if a single first-­degree ­relative with colorectal cancer is diagnosed at or after 60 years of age. If the first-degree relative was younger than 60 at diagnosis, colonoscopy should be performed every 3 to 5 years.

Surveillance Guidelines Because patients with resected colorectal cancer are at risk for recurrent cancer and metachronous neoplasms in the colon, screening strategies have been adapted for surveillance purposes after colorectal cancer has been diagnosed

569

and treated. The recommendations for surveillance depend on the anatomic location and natural history of the disease. Colon cancer usually metastasizes to the liver first, but rectal cancer can spread through paravertebral venous and lymphatic channels to the lungs without involving the liver. Most recurrences appear within 3 years of surgery, and almost all appear within 5 years. The American Society of Clinical Oncology used ­evidence-based criteria to create the Guidelines for Colorectal Cancer Surveillance, which were first published in 2000100 and updated in 2005.101 Standard follow-up includes a regular history, physical examination, and diagnostic studies (i.e., complete blood cell count, liver function studies, carcinoembryonic antigen [CEA] levels, and fecal occult blood tests), which should be done every 3 to 6 months for the first 3 years, every 6 months during years 4 and 5, and subsequently at the discretion of the physician. Annual radiography of the chest; CT scans of the chest, abdomen, and pelvis; and colonoscopy are recommended for 3 years after primary therapy. If the colonoscopy at 3 years after the operative procedure shows normal results, colonoscopy should be continued every 5 years thereafter. A 2006 update of guidelines from the American Cancer Society and the U.S. Multi-Society Task Force on Colorectal Cancer102 addressed the use of endoscopy for surveillance after resection for colorectal cancer. According to those guidelines, patients with endoscopically resected stage I colorectal cancer, surgically resected stage II or III cancers, or stage IV cancer resected with curative intent (with isolated hepatic or pulmonary metastasis) are candidates for endoscopic surveillance. After the colorectum is declared free of synchronous neoplasia (by preoperative colonos­copy if the colon is not obstructed or by preoperative double­contrast barium enema or CT-based colonography if the colon is ­obstructed, followed by another colonoscopy 3 to 6 months after surgery), colonoscopy should be performed at 1 year to look for metachronous lesions. If that examination ­result is normal, the interval before the next subsequent examination should be 3 years. If that colonoscopy result is ­normal, the interval before the next subsequent examination should be 5 years. Shorter intervals may be indicated if adenoma is detected or if the patient’s age, family ­history, or tumor testing indicate definite or probable HNPCC. ­Endoscopic sonography or flexible sigmoidoscopy at 3- to 6-month intervals for the first 2 years after resection can be considered for the purpose of detecting surgically curable recurrence of the original rectal cancer.102 Little consensus exists regarding the usefulness of the various components of follow-up evaluations. Routine CEA testing is a case in point. One report suggests that routine CEA testing can detect metastatic disease an average of 5 months before it becomes apparent on routine clinical evaluations.103 However, about 30% of recurrent tumors do not produce CEA, and a false-negative CEA ­ level is common in patients with poorly differentiated tumors. Nevertheless, the American Society of Clinical Oncology recommends that CEA levels be measured at diagnosis and every 3 months for at least 3 years for those with stage II or III disease if the patient is a potential candidate for surgery or chemotherapy for metastatic disease.101 CEA was also considered the marker of choice to monitor the response of metastatic disease to systemic therapy. Information on other potential tumor markers (e.g., TP53, RAS, thymidine

570

PART 6  Gastrointestinal Tract

synthase, dihydropyrimidine dehydrogenase, thymidine phosphorylase, microsatellite instability, 18q loss of heterozygosity, or deleted in colon cancer [DCC] protein) is insufficient to recommend their use for colorectal cancer surveillance.104 In a large cooperative group trial of 1247 patients with colon cancer,105 548 patients had recurrent disease; salvage surgery was attempted in 222 of these patients. and another 109 underwent surgery with curative intent.105 Among those for whom salvage was attempted, surgery was prompted because of abnormal CEA test results (38%), symptoms (25%), or abnormalities found on CT scanning (33%) or other tests (5%). Only 5% of these patients were free of disease 5 years after the recurrence. Among the 109 patients who underwent curative-attempt surgery (20% of those with recurrent disease), the presence of a solitary ­lesion at recurrence was the most favorable prognostic factor. At a median follow-up period exceeding 5 years, the estimated 5-year disease-free survival rate among those treated with curative intent was 23%. The limited potential for benefit must be balanced with the substantial morbidity associated with progressive disease. The authors of this report concluded that follow-up testing and early evaluation of symptoms could often prompt surgery, which could extend disease-free survival. In another study, intensive compared with limited follow-up was evaluated prospectively among 216 patients who had undergone resection of rectal cancer. Reoperation was attempted in 30% of the intensive and limited followup groups; the median survival of 19 months in the intensive group was not statistically different than the 8-month median survival for the limited follow-up group.106 ­Median survival duration for asymptomatic and symptomatic ­patients was 15 months, even though 44% of asymptomatic patients and 27% of symptomatic patients underwent ­reoperation. With regard to laboratory studies, of 8749 total laboratory values obtained in follow-up for 105 patients with colorectal cancer,107 5894 values (67%) were normal. Of the 2004 (23%) values that were low, 5% elicited a therapeutic response.107 Of the 851 (10%) values that were high, 2.5% elicited a response. These findings led the ­ authors of this study to recommend that laboratory tests be conducted for patients with coexisting medical problems such as diabetes and cardiovascular disease requiring ­potassium supplementation, patients for whom laboratory values were likely to be abnormal.

Imaging Modalities for Surveillance Until recently, the usefulness of routine CT scanning for surveillance had been considered unproven; however, the American Society of Clinical Oncology now recommends that CT scanning of the chest, abdomen, and pelvis be ­included in colon and rectal cancer surveillance programs. Patients who are at high risk for recurrence who could undergo surgery with curative intent should ­undergo ­annual CT of the chest and abdomen for 3 years after primary therapy for colorectal cancer. Pelvic CT scanning should also be considered for rectal cancer surveillance, especially for patients who have not undergone radiation therapy. This recommendation, which represents a significant change from previous policies, was based on evidence from three meta-analyses108-110 reviewed for the 2005 ­update of

the Society’s practice guidelines.101 Those analyses showed that CT scanning, particularly of the liver, conferred a survival benefit; mortality rates were 25% lower for ­patients who underwent liver scans than for those ­ followed with ­nonimaging strategies. This benefit was ­ directly related to the effectiveness of resection of limited hepatic ­metastases. The findings were further corroborated by Chau and ­colleagues,111 who described patterns of relapse and ­survival among 530 patients undergoing surveillance after participating in a randomized trial of adjuvant chemoradiation therapy for stage II or III colon cancer. These ­patients underwent CEA testing and CT scans of the chest, ­abdomen, and pelvis. CEA testing detected almost the same number of relapses as did CT scanning (45 and 49, respectively), and 14 were detected by both tests. Survival was better among the patients whose relapse was detected by CT compared with the 65 patients whose relapse was detected by symptoms (P = .0046). Patients who were able to undergo potentially curative surgery had improved survival, and relapses were most often detected in this group by CT scanning (26.5%) or by CEA tests (17.8%) and least often by symptoms (3.1%) (CT versus symptoms, P < .001; CEA versus symptoms, P <.015).111 Magnetic resonance imaging (MRI) may be better able than CT to visualize the extent of disease in the ­peritoneum, bowel, and mesentery. In one study, MRI was able to identify 90% of 154 surgically confirmed extrahepatic tumor sites, whereas CT revealed only 66%.112 Because the additional information from the MRI affected the therapeutic recommendations in that study, the investigators ­concluded that MRI might be beneficial for identifying resectable disease in some patients. In another study, findings from MRI agreed with those at histopathologic examination 84% of the time among patients undergoing surgery after preoperative chemoradiation therapy. The authors of that study also concluded that high-spatial-resolution MRI was useful for accurately staging tumors of the distal sigmoid, rectosigmoid, and upper rectum.113 Whole-body positron emission tomography (PET) scanning has been used to identify recurrent disease ­before it becomes evident on other imaging studies. In this ­technique, a glucose analogue labeled with fluorine 18 (2-[18F]fluoro-2-deoxy-d-glucose [FDG]) is used to detect malignancy by making visible the increase in ­ glucose uptake and metabolism by cancer cells. The agent is ­transported into cells by epithelial glucose-transporter proteins; ­ because FDG cannot be converted to the fructose analogue, it accumulates in the cells.114,115 A Belgian study114 found a discordance rate of 10% between FDGPET and conventional diagnostic imaging, but FDG-PET was determined to have additional diagnostic value in 14 of 16 locoregional, 6 of 7 hepatic, 7 of 8 abdominal, and 8 of 9 extra-abdominal ­ recurrences. In a patientbased analysis, FDG-PET was shown to have additional diagnostic value in terms of ­defining resectability in 20% of patients. An analysis at the Mallinckrodt Institute of Radiology showed that ­ FDG-PET had a positive predictive value of 89% and a ­ negative predictive value of 100% among patients with rising CEA levels after surgery for colorectal cancer.116 ­Another group found the sensitivity and specificity of FDG-PET to be 93% and 98%, ­respectively, compared with 69% and 98% for CT scans,

24  The Colon and Rectum

for ­ detecting recurrent colorectal cancer.117 In the latter study, the ­potential savings resulting from demonstration of unresectable disease by FDG-PET was calculated as being more than $3000 per preoperative study. FDG-PET has been especially useful in evaluating the retroperitoneum for adenopathy because PET can detect very small lesions and lesions in postoperative surgical fields that are difficult to assess with CT.118 In terms of histologic features, the cellularity and amount of mucin in the tumor mass were predictive of FDG-PET results, with FDG-PET being of limited value for evaluating mucinous tumors, particularly hypocellular lesions with abundant mucin.119

Summary of Surveillance Recommendations The current recommendations for surveillance of patients with colorectal cancer101 include a history and physical ­examination every 3 to 6 months for the first 3 years, every 6 months during years 4 and 5, and subsequently at the discretion of the physician, and CEA measurement every 3 months after surgery for at least 3 years after diagnosis, if the patient is a candidate for surgery or systemic therapy. In terms of imaging, annual CT scanning of the chest and ­abdomen is recommended for 3 years after primary ­therapy for patients who are at high risk for recurrence and who could be candidates for surgery with curative intent; pelvic CT is recommended for patients with rectal cancer, especially with several poor prognostic factors, including those who have not been treated with radiation. A colonoscopy should be done at 3 years after operative ­ treatment, and if results are normal, it should be done every 5 years there­after. Flexible sigmoidoscopy should be done every 6 months for 5 years for patients with rectal cancer that has not been treated with pelvic radiation. Chest radiographs, complete blood cell counts, and liver function tests are not recommended. Based on the evidence available, findings from tests of molecular or ­cellular markers should not influence the surveillance strategy. Future studies should target screening modalities that are most likely to have the greatest influence on survival. Nevertheless, because the morbidity of progressive symptoms is ­substantial, close clinical follow-up is necessary, and diagnostic studies should be undertaken promptly when symptoms appear to determine their cause and to initiate appropriate measures for their control. Because treatment-related effects and ­ intercurrent medical problems such as infection and inflammation can also produce symptoms and other diagnostic abnormalities, scanning can provide important information regarding treatable causes of ­intercurrent disease.120

Staging Considerations Tumor Classification with Imaging Modalities The tumor-node-metastasis (TNM) system from the American Joint Committee on Cancer (AJCC), used for staging tumors of the colon and rectum, is outlined in Box 24-2.121 T classification is determined through a combination of physical examination, including proctoscopy, and imaging by transrectal ultrasonography, CT, or MRI with an intrarectal coil. Patients with T2 or higher tumors should also undergo chest radiography and CT scanning of the ­abdomen and pelvis to exclude the presence of occult

571

­ etastases. Specialized techniques for the administration m of contrast agents and scan time should be used in CT scanning to increase its sensitivity and specificity for detecting liver metastases. Although CT and MRI are more than 90% accurate for identifying metastatic spread before treatment, both are limited with respect to their ability to reveal local disease extension. The ability of CT to visualize a primary tumor depends on the size and location of the tumor; in one study of 158 patients, CT identified the primary tumor in only 75% of cases.122 The sensitivity and specificity of CT in detecting perirectal fat involvement range from about 50% to 75%; tumors in the lower rectum are difficult to evaluate with CT because of the paucity of fat in this region.123,124 The accuracy of transrectal MRI for detecting tumor involvement of the perirectal fat has been higher than 85% in several series.123 The success rate of CT and MRI for identifying perirectal lymph node involvement is about 60%.124

Endorectal Sonography Endorectal sonography can be useful for accurately staging primary rectal tumors before surgery or preoperative chemoradiation therapy. Preparations for the procedure are minimal, requiring only an enema. The normal rectal wall appears as concentric alternating hyperechoic and hypoechoic bands (Fig. 24-2). Drawbacks of this technique are its inability to visualize normal perirectal lymph nodes, which are less than 3 mm in diameter, and the ease with which lesions at or near the anal verge can be missed because the balloon on the transducer tip will not pass through a normal sphincter without being deflated. In the latter situation, replacing the balloon with a hard plastic cap improves the resolution of the images.125 Because the choice of surgical technique is determined by whether tumor involves the sphincteric muscles, other sonographic techniques can be useful for evaluating muscle involvement. In women, transvaginal sonography is one such option. For detecting internal sphincter defects, the sensitivity of this technique is 44%, and its specificity is 96%; for external sphincter defects, the corresponding rates are 48% and 88%.126 Transperineal sonography of the rectum also can be useful for evaluating the distal anorectum, particularly when tumors in the distal rectum render that region stenotic and painful.127 Lymph nodes that are visible on endorectal sonography may contain metastatic disease, or they may be inflamed. The sonographic criteria for distinguishing the two conditions overlap somewhat,125 but metastatic disease generally is characterized by the appearance of discrete margins and is of uniform hypoechogenicity or the same echotexture as the primary tumor. The overall rate of accuracy of endorectal sonography was reported in one study128 to be 70%, with a positive predictive value of 67% and a negative predictive value of 75%. Overstaging of the primary tumor occurred for only 11% of patients who had not undergone irradiation, but overstaging occurred for 21% of patients who had undergone preoperative radiation therapy because of fibrosis in the tumor bed. Estimation of lymph node involvement in this study was accurate in 71% of patients, with a ­ positive predictive value of 47% and a negative predictive value of 76%. No false-positive results were obtained for unirradiated patients, but 8 of 54 patients who underwent preoperative

572

PART 6  Gastrointestinal Tract

BOX 24-2.  American Joint Committee on Cancer Staging System for Cancer of the Colon and Rectum Primary Tumor (T) TX Primary tumor cannot be assessed T0 No evidence of primary tumor Tis Carcinoma in situ: intraepithelial or invasion of lamina propria* T1 Tumor invades submucosa T2 Tumor invades muscularis propria T3 Tumor invades through the muscularis propria into the subserosa or into nonperitonealized pericolic or perirectal tissues T4 Tumor directly invades other organs or structures and/or perforates visceral peritoneum†‡ Regional Lymph Nodes (N)§ NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Metastasis in 1 to 3 regional lymph nodes N2 Metastasis in 4 or more regional lymph nodes Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage Grouping AJCC/UICC Stage 0 Stage I Stage IIA Stage IIB Stage IIIA Stage IIIB Stage IIIC Stage IV

Tis T1 T2 T3 T4 T1-T2 T3-T4 Any T Any T

N0 N0 N0 N0 N0 N1 N1 N2 Any N

M0 M0 M0 M0 M0 M0 M0 M0 M1

Dukes¶ — A — B B C C C —

MAC¶ — A B1 B2 B3 C1 C2/C3 C1/C2/C3 D

*Tis

includes cancer cells confined within the glandular basement membrane (intraepithelial) or lamina propria (intramucosal) with no ­extension through the muscularis mucosae into the submucosa. †Direct invasion in T4 includes invasion of other segments of the colorectum by way of the serosa (e.g., invasion of the sigmoid colon by a carcinoma of the cecum). ‡Tumor that is adherent to other organs or structures macroscopically is classified T4. However, if no tumor is present in the adhesion ­microscopically, the classification should be pT3. The V and L substaging should be used to identify the presence or absence of vascular or lymphatic invasion. §A tumor nodule in the pericolorectal adipose tissue of a primary carcinoma without histologic evidence of residual lymph nodes in the nodule is classified in the pN category as a regional lymph node metastasis if the nodule has the form and smooth contour of a lymph node. If the nodule has an irregular contour, it should be classified in the T category and coded as V1 (microscopic venous invasion) or as V2 (if grossly evident), because there is a strong likelihood that it represents venous invasion. ¶Dukes B is a composite of better (T3 N0 M0) and worse (T4 N0 M0) prognostic groups, as is Dukes C (any T N1 M0 and any T N2 M0). AJCC, American Joint Committee on Cancer; UICC, International Union against Cancer. From Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 113-123.

radiation had false-positive results. False-negative results occurred for 6 of 36 unirradiated patients and for 12 of 54 patients who underwent preoperative radiation.128 The overall accuracy of endorectal sonography in determining the depth of penetration of the pelvic wall during preoperative staging ranges from 72% to 97%, with up to 12% of tumors overstaged and 9% understaged.125 The lack of specific imaging criteria is reflected in the rates of ­accuracy of predicting perirectal lymph node invasion, which range from 60% to 85%, with a sensitivity of 59% to 89% and a specificity of 64% to 87%. Other sources of error in endorectal sonographic staging include tumor location and stenosis, peritumoral inflammation, changes after biopsy or surgery, hemorrhage, and pedunculated or villous tumors.125 When used to evaluate tumor recurrence after completion of all therapy, sonography compares favorably with

other follow-up techniques. A German group129 used ­sonography to prospectively follow 338 patients with rectal cancer, 116 of whom developed recurrent disease. Among the follow-up tests performed, the accuracy of tumor markers was 58% (sensitivity of 36%, specificity of 88%), the accuracy of digital examination was 23% (sensitivity of 21%, specificity of 99%), the accuracy of proctoscopy was 35% (sensitivity of 31%, specificity of 98%), and the accuracy of endorectal sonography was 79% (sensitivity of 79%, specificity of 100%).129 In this study, follow-up endorectal sonography had the highest rates of sensitivity and specificity, and it had a positive predictive value of 100% and a negative predictive value of 90%. In contrast, the positive predictive value for tumor markers was 62%, and that for a digital examination with proctoscopy was 90%; the negative predictive value for these procedures was 70%. The authors of this study concluded that assessment of local

24  The Colon and Rectum

Balloon/mucosa interface

A

1

Muscularis mucosa

2

Submucosa

3

Muscularis propria

4 5

Perirectal fat

C

573

B

D

FIGURE 24-2.  Endorectal sonographic images. A and B, The layers of the normal rectal wall. 1, balloon-mucosa interface; 2, muscularis mucosa; 3, submucosa; 4, muscularis propria; 5, perirectal fat. C, Perirectal lymph node involvement is indicated by a hypo­ echoic lesion (arrows) outside the rectal wall. D, A tumor extends through the bowel wall into the perirectal fat (arrows).

recurrence might be best evaluated by the cost-effective option of endorectal sonography.129

patients who developed recurrence had persistent positive lymph nodes after preoperative 5-FU–based chemoradiation therapy, and those who responded to the preoperative therapy had longer disease-free and overall survival intervals than those who did not.131

Treatment Strategies for Rectal Cancer

Surgical Approaches

The anatomic location of and the structures involved by tumors in the rectal area often determine therapeutic recommendations. The rectum is classically divided into three levels: low rectum, middle rectum, and upper or proximal rectum (Fig. 24-3). However, variations in anatomy and body habitus make it difficult to precisely delineate each of these segments. The American College of Surgeons considers tumors located from the anal verge to 15 cm proximally to be rectal tumors; others contend that tumors located 12 to 15 cm proximal to the anal verge act more like colon cancer than rectal cancer. A more stringent definition of rectal tumors includes tumors found during rigid proctoscopy, with the patient in the left lateral (Sims’) position, to be within 12 cm of the anal verge. With regard to choice of treatment, tumor and lymph node status are critically important prognostic factors. In one analysis of pooled data from 3791 patients from five randomized phase III trials of adjuvant therapy for rectal cancer in North America,130 T category significantly influ­ enced 5-year overall survival rates even among patients with N2 disease: The survival rates were 67% for T1 or T2 tumors, 44% for T3 tumors, and 37% for T4 tumors. Responding to neoadjuvant chemoradiation therapy has been associated with longer disease-free survival and perhaps overall survival. According to one study,131 all of the

The success of combined-modality therapy (discussed ­later) means that abdominoperineal resection is now reserved for tumors located close to the anal verge or those from the distal aspect of the coccyx to the levator ani muscles (see Chapter 25). Other surgical procedures in current use are local excision, standard low anterior resection, low anterior resection with coloanal anastomosis, and low anterior resection or coloanal anastomosis with the creation of a J-pouch colonic reservoir; the success rates and outcomes after each type of surgery are discussed throughout this section. The surgical management of colorectal cancer has four goals. The first goal is cure, defined as the prevention of ­systemic spread; the second is local control; the third is preservation of the anal sphincter or anorectal function; and the fourth is the preservation of sexual and urinary functions by preserving the integrity of the autonomic ­nervous system.132 In one study,133 resection of rectal ­cancer that spared the pelvic autonomic nerves and ­ included lateral lymph node dissection resulted in local control rates of 100% in patients with Dukes stage A or B disease (see Box 24-2) at a median follow-up time of 53 months. In the same study, the 5-year survival rate for patients with Dukes stage A disease was 96%, that for patients with stage B ­disease was 84%, and that for those with stage C

574

PART 6  Gastrointestinal Tract Measurements from anal verge

Rectal divisions

16 cm

Upper 1/3 12 cm

Middle 1/3 8 cm Levator ani muscle

Lower 1/3

4 cm D Dentate line

2 cm S Sc

Anal verge

External sphincter

FIGURE 24-3.  Anatomy of the rectum. D, deep; S, superficial; Sc, subcutaneous.

­ isease was 67%.133 Urinary function was preserved in 94% d of ­ patients undergoing dissection of the unilateral pelvic plexus. When the complete pelvic plexus was preserved, sexual function could be preserved in 70% of men; if the hypogastric nerves had to be removed but the pelvic plexus could be ­ preserved, 67% of men could maintain the ability to have an erection and intercourse without normal ­ejaculation.133

Importance of Total Mesorectal Resection Complete resection of the mesorectum is critical for achieving local control of colorectal cancer. Incomplete ­resection of the mesorectal fat consistently results in local failure rates as high as 90% within 18 months because of residual tumor in the lymph nodes attached to the pelvic sidewall.132 The visceral fascia covers the rectum and the mesorectum. The parietal fascia covers the musculoskeletal and vascular structures of the pelvic sidewalls, including the pelvic autonomic plexus. When the mesorectal lymph nodes are positive, one third of patients also have lateral nodal involvement. Lateral lymph node involvement is ­associated with transmural and extraperitoneal disease. For patients treated with surgery only, the overall recurrence rates are 65% when the lateral pelvic lymph nodes are positive and 38% when those nodes are negative; the corresponding local recurrence rates are 26% and 10%, and the corresponding 5-year overall survival rates are 49% and 74%. In contrast, the addition of pelvic radiation and concurrent chemotherapy to surgical treatment can clear residual microscopic tumor in pelvic lymph nodes. Combined-modality therapy that includes surgical resection with total mesorectal excision can produce local control rates in excess of 90% for locally advanced rectal cancer. Another detailed study involved assessing the local depth of invasion of colon cancer in terms of the site and extent of regional lymph node metastases.134 A total of

12,496 lymph nodes from 164 patients (about 76 nodes per patient) were evaluated, and metastatic involvement of the mesenteric lymph nodes was documented in 59% of cases. Of these 724 metastatic lymph nodes, 33% had tumor deposits that were less than 4 mm in diameter. The pattern and extent of pericolonic lymph node spread led the investigators to suggest that proximal and distal margins of at least 7 cm, with excision of the regional mesentery and the intermediate and central lymph nodes, were needed to prevent recurrence.134 Further information regarding the effect of nodal involvement is available from a study of 1664 patients with T3, T4, or node-positive rectal cancer treated with radiation and chemotherapy.135 In that study, the number of nodes examined in the pathologic specimen correlated withww ­relapse and survival rates among patients who were considered node-negative after therapy. Among node-­negative patients who had up to 4 nodes examined, the 5-year ­relapse and survival rates were 27% and 68%, respectively; the corresponding rates for patients who had 5 to 8 nodes examined were 34% and 73%: for patients who had 9 to 13 nodes examined, 26% and 72%; and for patients who had 14 or more nodes examined, 19% and 82%.135 No such associations were found among patients with node-positive disease. Advances in surgical technique have greatly improved the local control of rectal cancer, as demonstrated in a retrospective review of 1581 patients who underwent curative resection for rectal cancer between 1974 and 1991.136 (Total mesorectal excision was introduced in 1985.) The median follow-up time for this study was 13 years; none of these patients was given adjuvant chemotherapy or radiation therapy.136 The local recurrence rate decreased from 39% to 10% during the study period, and the ­ observed 5-year survival rate increased from 50% to 71%. The overall survival rate was significantly lower among the

24  The Colon and Rectum

306 patients who experienced local recurrence. Distant disease developed in 275 of 1285 patients who had no local recurrence, 8% of whom had stage I disease, 16% of whom had stage II disease, and 40% of whom had stage III disease. Because the type of the surgery had no effect on the development of distant metastases, the authors of the study concluded that improvements in surgical technique independently improved survival by improving local control of the disease.136 Another variable identified in the late 1980s as being critical to the success of local control is the radial or circumferential resection margin. The Dutch Colorectal Cancer Group, in a review published in 2008 of more than 17,500 cases of rectal cancer,137 indicated that a positive circumferential margin, defined as tumor cells being present within 2 mm of the resection margin, was a powerful predictor of the development of local recurrence, distant metastasis, and survival. Tumor characteristics associated with positive circumferential margins were being large and ulcerative or having a stenosing growth pattern, infiltrating margin, poor differentiation, or vascular invasion. Lower rectal tumors were more likely to have circumferential ­ involvement than middle tumors.138 Moreover, positive radial margins were more common after abdominoperineal resection than after anterior resection for low and middle rectal cancer.138 The authors of the study ascribed the poor prognosis of the patients who underwent abdominoperineal resection to the resection plane of the operation, which was often ­associated with margin involvement and perforation.138

Sphincter Preservation The three major types of sphincter-preserving operations are a standard low anterior resection, a low anterior resection with coloanal anastomosis, and a low anterior ­resection or coloanal anastomosis with the creation of a J-pouch ­colonic reservoir. All low anterior resections involve resection and an anastomosis between a serosalized colon and the extraperitoneal nonserosalized rectum. The standard low anterior resection involves an intrapelvic anastomosis within the sacral hollow; the length of the remaining distal segment of rectum varies.139,140 By comparison, a coloanal anastomosis is an extrapelvic anastomosis at the apex of the anal canal or at the dentate line, and no distal rectum remains. A J-pouch reconstruction can be combined with a low anterior resection or a coloanal resection; the reconstruction recreates the reservoir function of the rectum and improves long-term bowel function. Extensive experience with this technique has confirmed excellent rates of ­ tumor control and functional outcome, including fecal ­continence.141,142 Other innovations in rectal surgery ­include ­intersphincteric resection, with excision of the internal anal sphincter, after preoperative radiation therapy.143 Early results with this technique suggest that it produces rates of local control and continence that are comparable to those of other techniques.143 Whether neoadjuvant therapy could improve the chance of being able to undergo sphincter-preserving surgery for low rectal cancer was addressed by a group in Italy.144 In that study, 90 patients with locally advanced cancer of the lower rectum underwent endorectal sonography before and after preoperative chemoradiation therapy (50 Gy in 25 fractions with concomitant continuous-infusion 5-FU at 225 mg/m2, 7 days per week), and the findings were

575

c­ ompared with the histopathologic findings after surgery done 6 to 8 weeks later. Of the 90 cases, 82.5% showed ­tumor downsizing, and the distance between the tumor and the anal sphincter increased in 60.2% of cases. Sphincterpreserving surgery was performed in 61% of cases, which was twice the proportion anticipated from the results of the preoperative staging.144 Body mass index may also need to be considered as an independent factor in sphincter preservation. A report from the Intergroup 0114 trial of stage II or stage III rectal cancer145 showed that overweight or obese men and women experienced less toxicity from the chemoradiation therapy, but men with high body mass indices were less likely to be able to undergo sphincter-preserving surgery and more likely to experience local recurrence.

Local Excision Local excision, although technically a type of sphincterpreserving operation, can cure only those tumors confined to the bowel wall because it does not include a ­mesorectal resection. The risk of lymph node involvement in ­ rectal cancer is linked to the tumor classification; for T1 tumors, the risk is 0% to 12%; for T2 tumors, 12% to 28%; and for T3 and T4 tumors, the risk ranges from 36% to 79%.146 ­Local failure rates among patients who undergo local excision would be expected to be higher than those among patients who undergo more comprehensive surgery, particularly for patients with higher-stage tumors. A group at the University of Minnesota compared outcome between 108 patients with T1 or T2 tumors who underwent transanal excision and 153 patients with similar characteristics who were treated with standard radical surgery. The local recurrence rate was higher in the transanal excision group (28% versus 4% in the radical surgery group).146 Moreover, this effect was stage-dependent: 18% of T1 and 47% of T2 tumors removed by transanal excision recurred, compared with 0% of T1 and 6% of T2 tumors removed by comprehensive surgical resection. This increase in local recurrence rate also resulted in an increased overall recurrence rate of 21% for T1 tumors and 47% for T2 tumors in the local excision group, compared with 9% and 16%, respectively, in the radical surgery group. The 5-year survival rate also was lower among those undergoing transanal excision (72% for those with T1 tumors and 65% for those with T2 tumors) than among those undergoing radical surgery (80% for those with T1 tumors and 81% for those with T2 tumors).146 Other series involving local excision have produced similar results. A 1998 review revealed that the local recurrence rate for 234 patients with T1 or T2 tumors treated by local excision was 27% and that the overall survival rate at 5 years was 65%.147 In 404 such cases at 2 to 3 years of follow-up, the local recurrence rate was 5%, and the overall survival rate 97% for those with T1 tumors, with corresponding rates of 18% and 92% for those with T2 tumors and 22% and 80% for those with T3 tumors. The addition of postoperative radiation improved outcome despite the radiation therapy group having had more adverse features such as higher T category (T2 tumors were present in 15% of those treated with local excision and in 70% of those treated with local excision plus radiation). In another study,148 the 5-year actuarial results for 52 cases of local excision and for 47 cases of local excision plus radiation

576

PART 6  Gastrointestinal Tract

showed better local control rates (72% versus 90%) and better relapse-free survival rates (66% versus 74%) among those given postoperative radiation. Adverse findings on histologic examination such as poor differentiation and the presence of lymphovascular invasion were associated with higher rates of local recurrence after local excision without radiation.148 The only prospective multi-institutional trial performed evaluating the use of local excision for T1 or T2 rectal ­tumors was conducted by the Cancer and Leukemia Group B (CALGB).149 This group studied 59 patients with T1 ­tumors and 51 with T2 tumors; the tumors had to be smaller than 4 cm, encompass less than 40% of the bowel wall circumference, and be located less than 10 cm from the dentate line (see Fig. 24-3). T1 tumors were treated with local excision only, and T2 tumors were treated with local excision followed by postoperative radiation with 54 Gy (given in 30 fractions) and chemotherapy (an intermittent infusion of 5-FU at 500 mg/m2).149 Significantly, of the 161 patients for whom a full-thickness local excision was attempted, 51 had positive margins or tumors higher than T2 or lower than T1. Among the 59 patients with T1 disease, 4 had a local recurrence, and 3 of those 4 patients died with disseminated disease. Among the 51 patients with T2 disease, 20% had a local recurrence at a median followup of 4 years; moreover, one half of these local recurrences also involved distant disease and could not be treated with salvage surgery. This study provided four conclusions.149 First, sphincter preservation was not easy to accomplish in a uniformly defined manner; 51% of the patients evaluated proved to be ineligible after surgical resection. Second, sphincter-preserving surgery was associated with less morbidity than was radical surgery. Third, because of the few recurrences, predictors of outcome were not available. Fourth, the effectiveness of salvage therapy remains to be determined.149 Another important study addressing the question of the oncologic adequacy of local excision for rectal cancer was published in 2007 by You and colleagues.150 It analyzed longitudinal trends in the use of surgical therapy (local ­excision versus standard resection) among 35,179 patients with stage I rectal cancer diagnosed during the period from 1989 through 2003. A second analysis focused on perioperative morbidity, local tumor control, and survival among 2124 patients diagnosed between 1994 and 1996; 765 of these patients (601 with T1 tumors and 164 with T2 tumors) underwent local excision, and 1359 patients (493 with T1 tumors and 866 with T2 tumors) underwent standard resection. The use of local excision was found to increase over time from 1989 through 2003. Perioperative morbidity (i.e., that experienced within 30 days of surgery) was less common after local excision (5.6%) than after standard resection (14.6%, P < .001). After adjustments were made for patient and tumor characteristics, the local recurrence rate at 5 years was found to be higher after local excision than after standard resection (for T1 tumors, 12.5% versus 6.9%, P = .003; for T2 tumors, 22.1% versus. 15.1%, P = .01).150 For those with T1 tumors, the 5-year overall survival rates were influenced by age and the presence of comorbid conditions but not by the type of surgery; however, for those with T2 tumors, 5-year survival rates were lower after local excision (67.6%) than after standard ­ resection (76.5%; P = .01). The authors of this ­ report ­ emphasized

the need to balance the benefits of local ­excision against the increased risk of local failure for each patient, ­particularly those with T2 tumors.150 At this point, the role of local excision in the treatment of rectal cancer depends on disease stage and other tumor characteristics and is still largely undefined.151 Although interpretation of most studies of transanal resection is complicated by the presence of comorbid conditions, diseasefree survival usually is worse among patients undergoing transanal excision. Local recurrence after transanal excision of rectal cancer may be amenable to radical resection, but not all local recurrences can be cured by radical surgery, and long-term survival rates in such cases are only about 50% to 60%.152 Attempts to maximize sphincter preservation must not be compromised by inferior rates of tumor control or functional outcome.

Surgery for Locally Advanced Disease At the other end of the surgical spectrum is the use of limited resection and organ preservation for disease that has infiltrated adjacent organs. Often the bladder can be spared during resection of a locally advanced rectal cancer without sacrificing local control. In one study,153 partial ­resection of the bladder was achieved with negative pathologic margins in 94% of cases. Moreover, the incidence of postoperative urinary complications and the rates of local control (about 85%) and survival at 3 years (about 45%) were the same for those who underwent bladder-sparing surgery as for those who underwent standard total pelvic exenteration.153

Predictors of Surgical Outcome The success of curative surgery for rectal carcinoma ­depends on the disease presentation, patient-related factors, the experience of the treating physicians, and the surgical volume of the institutions at which treatments take place. Considerable variability has been found among surgeons and among institutions in the rates of locoregional recurrence and survival. In one comparison of seven institutions, the local recurrence rates varied from 10% to 37%, and the cancer-related survival rate varied from 54% and 75%.154 Among the 14 surgeons surveyed, local recurrence rates ranged from 4% to 55%, and cancer-specific survival rates from 73% to 85%; among surgeons who performed colorectal cancer surgery infrequently, the mean local recurrence rate was 23%, and the mean cancer-specific survival rate was 69%.154 A clear correlation was apparent between ­local recurrence and survival. In another study,155 the only factors related to local recurrence rate were the length of the resection specimen obtained and whether the surgeon specialized in colorectal surgery; the local recurrence rate in that study was 3.42 times higher if general surgeons did the resection than if colorectal surgeons did. Advanced disease stage, the presence of vascular invasion, and having a surgeon who had not specialized in colorectal surgery were independent predictive factors for local recurrence and overall survival.155 The hospital volume is also an important factor in oncologic outcome after surgery for low rectal cancer. In an observational study within a prospective multicenter trial of 1557 patients with low rectal cancer undergoing low anterior resection or abdominoperineal resection, univariate analyses suggested that abdominoperineal resection was

24  The Colon and Rectum

577

TABLE 24-1.   Combined-Modality Treatment for Colorectal Cancer: the NSABP R-01 Trial Outcome Measures at 5 Years (%)* Treatment Group

Overall ­Survival Rate

Disease-­Free Survival Rate

Local Failure Rate

Surgery

43

30

25

Surgery + chemotherapy

53

42

21

Surgery + radiation therapy

39

34

16

*The

trial included 528 patients, the total radiation dose was 46 to 47 Gy, and chemotherapy consisted of fluorouracil, semustine, and vincristine. Data from Fisher B, Wolmark N, Rockette H, et al. Postoperative adjuvant chemotherapy or radiation therapy for rectal cancer: results from NSABP R-01. J Natl Cancer Inst 1988;80:21-29.

a­ ssociated with higher local recurrence rates and shorter survival time; hospital volume may have affected local recurrence rates (p = .06) but not disease-free survival rates. In multivariate analyses, however, local recurrence was influenced by hospital volume and disease-free survival by the type of surgical procedure performed.156 Independent predictors of major perioperative complications after resection, with or without ­ preoperative chemoradiation therapy, have been evaluated in ­ several studies.157,158 Such predictors include the American ­Society of Anesthesiologists score,159 use of temporary ­ ileostomy, type of surgical procedure, preoperative hemoglobin level, and extent of intraoperative blood loss. Failure to use a temporary diverting stoma was an especially strong ­predictor of major complications.158 The use of preoperative chemoradiation therapy did not influence the risk of perioperative complications,158 at least when the surgeons were experienced157; rather, length of hospital stay, readmission rates, and mortality were related to the severity of the ­illness rather than to any preoperative treatment.160

Combined-Modality Therapy for Locally Advanced Disease The concept that chemotherapy can improve the therapeutic ratio of radiation therapy was first demonstrated in studies of these modalities for rectal cancer, and these seminal studies provided a model for combined-modality therapy for cancer at other sites. As described in Chapter 4, radiosensitization usually is accomplished by limiting tumor cell repopulation and reducing the efficiency of radiation repair ­ during treatment.161-163 The benefit of adjuvant radiation and chemotherapy for locally advanced rectal cancer (T3 or T4 tumors, with or without positive regional lymph nodes) has been clearly demonstrated in several prospective randomized trials.164-168 Treatment with surgery produces an overall 5-year disease-free survival rate of about 55%. The local failure rate after surgery ranges from 35% to 50%; this rate is reduced to 10% to 20% by the addition of postoperative radiation.169,170 However, in the absence of systemic therapy, the risk of distant metastasis after surgery plus radiation is ­approximately 20% in stage II disease and 40% to 60% in stage III disease. Without chemotherapy, the 5-year survival rates are approximately 70% to 90% for stage II disease, 40% for T3 N1 disease, and 15% to 20% for T4 N1 disease.

Adjuvant Therapy Several prospective randomized studies have been conducted to evaluate the influence of adjuvant chemotherapy and radiation therapy on rectal cancer. In the late 1980s,

the NSABP compared outcomes among patients treated with surgery only, surgery plus postoperative radiation, and surgery plus postoperative chemotherapy.168 The radiation dose was 46 Gy in 25 fractions; however, only 86% of the patients received the total prescribed dose. The group given surgery plus postoperative radiation had a lower local recurrence rate than the other groups, but this difference did not affect survival (Table 24-1).168 The addition of adjuvant chemotherapy to surgery significantly improved disease-free and overall survival rates over those for the surgery-only group, but the rates of local failure and distant metastasis were no different between the two treatment groups. Two other randomized studies, one conducted by the Gastrointestinal Tumor Study Group (GITSG) and the other by the Mayo/North Central Cancer Treatment Group (NCCTG), showed reductions in local recurrence rates and improvements in disease-free and overall survival rates when postoperative radiation and chemotherapy were ­administered to patients with Dukes stage B2 or C rectal cancer. The GITSG trial compared four treatments: surgery alone, surgery plus postoperative radiation (40 or 48 Gy), surgery plus postoperative chemotherapy, and surgery plus postoperative radiation (40 or 48 Gy) plus chemotherapy. The addition of any form of adjuvant therapy improved local control, disease-free, and overall survival rates over those produced by surgery only (Table 24-2).165-167 The addition of radiation therapy decreased the incidence of local failure as the initial form of disease recurrence and, in combination with chemotherapy, reduced the risk of ­local failure to 10%. The addition of chemotherapy to surgery did not improve the risk of local recurrence (see Table 24-2).165-167,171 The Mayo/NCCTG trial,164 which compared outcome after postoperative radiation (50.4 Gy in 28 fractions) with or without chemotherapy (5-FU and semustine), also showed an advantage for combined-modality therapy. The addition of chemotherapy reduced the rate of local failure from 25% to 13.5% and reduced the rate of distant metastasis from 46% to 29%. These reductions resulted in improvements in the disease-free survival rate (from 37% to 59%) and overall survival rate (from 48% to 58%). The combined-modality treatment reduced local and distant recurrences by 34% and reduced the overall death rate by 29%.164 On the basis of these findings, the National Cancer ­Institute announced in 1991 that adjuvant radiation therapy and chemotherapy were to be the standard of care for locally advanced rectal cancer.172 Long-term findings

578

PART 6  Gastrointestinal Tract

TABLE 24-2.    Combined-Modality Treatment for Colorectal Cancer: the GITSG 7175 Trial* Outcome Measures (%)* 10-Year Overall ­Survival Rate

10-Year Disease-Free Survival Rate

Recurrence Rate

Local Failure Rate

Surgery

26

44

55

25

Surgery + chemotherapy

41

51

46

27

Surgery + radiation therapy

33

50

46

20

Surgery + chemoradiation therapy

45

65

35

10

Treatment Group

*The

trial included 202 patients, the total radiation dose was 40 to 48 Gy, and chemotherapy consisted of fluorouracil and semustine. Data from Gastrointestinal Tumor Study Group. Prolongation of the disease-free interval in surgically treated rectal carcinoma. N EngI J Med 1985;312:1465-1472; Thomas PR, Lindblad AS. Adjuvant postoperative radiotherapy and chemotherapy in rectal carcinoma: a review of the Gastrointestinal Tumor Study Group experience. Radiother Oncol 1988;13:245-252.

r­ eported in 2006 from the Gastrointestinal Intergroup trial INT 0144 confirmed the success of postoperative adjuvant chemoradiation therapy for 1917 patients with high-risk rectal cancer (T3-4 N0 or T1-4 N1-3).173 Some evidence suggests that pT3 N0 rectal cancer with favorable pathologic features has such low rates of local failure after surgery that postoperative chemoradiation therapy may not be necessary.174-176 Findings from a pooled analysis reported by Gunderson and colleagues130 indicated that 5-year overall survival rates for patients with pT3 N0 rectal cancer were similar regardless of whether they were treated with surgery followed by chemotherapy alone (84%) or surgery followed by chemoradiation ther­ apy (74% to 80%), suggesting that trimodality therapy may be excessive for some patients with T3 N0 disease. Another group177 found that 22% of patients with rectal cancer that was staged as cT3 N0 by endorectal sonography or MRI had pathologically positive lymph nodes even after preoperative chemoradiation therapy. Moreover, because preoperative chemoradiation therapy may reduce the total number of involved lymph nodes and sterilize ­mesorectal lymph nodes,178-180 the true rate of unidentified pathologically involved lymph nodes may be even higher. The investigators concluded that reliance on endorectal sonography or MRI for rectal cancer staging results in a substantial number of cT3 N1 rectal cancers being understaged and therefore potentially not receiving preoperative chemoradiation therapy, and they contend that preoperative chemoradiation therapy should remain the standard of care for patients with T3 rectal cancer staged in this way.174-178

Neoadjuvant Therapy Little doubt remains about the value of adjuvant therapy compared with surgery alone for locally advanced colorectal cancer. However, evidence from several large trials ­indicates that the benefit is enhanced still further by the use of neoadjuvant therapy—that is, by giving radiation or chemoradiation therapy before surgery. In their 2004 report, the German Rectal Cancer Study Group established that treatment of locally advanced rectal cancer with preoperative chemoradiation therapy led to significant improvements in local control, toxicity, and sphincter preservation compared with postoperative chemoradiation therapy.181 In that phase III trial, 823 patients with clinical T3 or T4 or node-positive rectal cancer (i.e., stage II or III) were randomly assigned to receive

chemoradiation therapy (50.4 Gy with concurrent continuous-infusion 5-FU during the first and fifth weeks of radiation therapy) before or after surgery. The postoperative therapy group was given an additional radiation boost of 5.4 Gy. Patients in both groups also were given four cycles of adjuvant chemotherapy with bolus 5-FU. The study showed that preoperative chemoradiation therapy produced fewer locoregional recurrences (6%) than postoperative chemoradiation therapy (13%) and less acute or long-term toxicity. The neoadjuvant group experienced fewer episodes of grade 3 or 4 diarrhea (12% versus 18% in the adjuvant group) and fewer anastomotic strictures (4% versus 12%) and were more likely to be able to undergo sphincter-preserving surgery (among patients thought to require an abdominoperineal resection, 39% in the neoadjuvant group underwent sphincter-sparing surgery as ­opposed to 19% of those in the adjuvant group).181 These results and those from trials conducted by the European Organisation for the Research and Treatment of Cancer (EORTC)182,183 and the Fondation Francophone Cancérologie Digestive (FFCD)184 (described in the following paragraph) indicate that preoperative chemoradiation therapy should be considered the current standard of treatment for stage II or III rectal cancer. The question of whether neoadjuvant chemoradiation therapy is superior to neoadjuvant radiation therapy for ­ locally advanced rectal cancer was addressed in two randomized trials. EORTC 22921182,183 assigned 1011 ­patients with resectable T3 or T4 rectal cancer to one of the following four treatments: preoperative radiation; preoperative chemoradiation therapy; preoperative radiation with postoperative chemotherapy; or preoperative chemoradiation therapy with postoperative chemotherapy.182,183 ­ Patients treated with preoperative chemoradiation therapy had significantly lower pathologic disease stage and smaller tumors at surgery than did patients treated with preoperative radiation. Patients in the preoperative chemoradiation therapy group also had a higher rate of pathologic complete response (14% versus 5%) and higher rates of conservative resection (56% versus 52%) than did patients treated with preoperative radiation. The rates of overall survival and progression-free survival were not different among the four treatment groups, but local relapse rates were significantly lower in the three groups given chemotherapy. The FFCD 9203 trial showed similar results.184 In that phase III trial, 762 patients with

24  The Colon and Rectum

resectable T3 or T4 rectal cancer were given preoperative radiation or preoperative chemoradiation therapy. The addition of chemotherapy to preoperative radiation ­increased the pathologic complete response rate from 4% to 12% and decreased the local failure rate from 17% to 8%. However, preoperative chemoradiation therapy in this trial did not increase the rates of sphincter-sparing surgery over that in the preoperative radiation group, nor did it improve event-free or overall survival rates compared with radiation only. Preoperative chemoradiation therapy was associated with a higher rate of grade 3 or 4 toxicity (15% versus 3%).184 Although these two randomized trials did not show a survival benefit from the addition of chemotherapy to radiation therapy, both trials demonstrated that adding chemotherapy to standard-fractionation radiation therapy in the preoperative setting increased the rates of local control and pathologic complete response.

Changes in Treatment Trends over Time Analysis of the National Cancer Data Base revealed the ­following four trends in the patterns of care for rectal ­adenocarcinoma between 1985 and 1995.185 First, stage I disease was diagnosed at decreasing frequencies over time and was found in only one third of cases at the end of the study period. Second, local excision was being performed with increasing frequency for stage I disease over time. Third, stage for stage, a decline in the use of abdominoperineal ­resection occurred over time. Fourth, multimodal treatments were being used with greater frequency, ­especially for stage II and III disease, over the course of the study ­period.185 In 1985, the combination of radiation, chemotherapy, and surgery was used for less than 1% of patients with stage I disease, 6% of patients with stage II disease, 13% of patients with stage III disease, and 8% of patients with stage IV disease; by 1995, the percentages had increased to 11%, 40%, 52%, and 15%, respectively. For patients diagnosed with stage II disease in 1989 to 1990, survival rates at 5 years were 62% for those treated with chemotherapy and surgery, with or without radiation, compared with 55% for those treated with surgery alone or in combination with radiation.168 For those with stage III rectal cancer, the 5-year survival rates were 42% for those treated with chemotherapy and surgery, with or without radiation; 39% for those treated with surgery and radiation; and 31% for those treated with surgery alone.168 The influence of the 1991 National Cancer Institute statement was evident in a Patterns of Care study186 in which treatment of rectal cancer at 57 institutions in the United States between 1992 and 1994 was evaluated. Radiation was given postoperatively to 75% of the ­ patients, preoperatively to 22%, and preoperatively and postoperatively to 2%. Among the 507 patients in this study, 243 had T3 or N1-2 M0 disease. Although only 7% of those with locally advanced disease were enrolled in clinical trials, 90% were given chemotherapy, which lasted a median of 21 weeks. With regard to surgery, 54% of those with locally advanced disease underwent a low anterior resection, and 46% had an abdominoperineal resection. Modern radiation therapy techniques with high-energy photons were used for most patients, and 93% were treated with three or four radiation treatment portals.186 Ninety percent of the ­ patients treated in the United States from 1992 to 1994 were given ­combined-modality therapy.

579

Long-term follow-up analysis of patients treated in another Patterns of Care study performed between 1988 and 1989 confirmed that only stage and use of chemotherapy were significant with regard to survival.187 The 5-year survival rate was 69% for those treated with postoperative chemoradiation therapy and 50% for those treated with postoperative radiation. Preoperative radiation resulted in a higher survival rate (85%) at 5 years than did postoperative radiation (50%) but was not statistically better than postoperative chemoradiation therapy.187 Survival also depended on disease stage; 5-year overall survival rates were 85% for those with stage I disease, 69% for stage II disease, and 54% for stage III disease.187 The 5-year survival rate also dropped from 89% for those with Dukes stage C1 disease to 48% for those with stage C2 or C3 disease.187

Chemotherapy Agents Neoadjuvant chemoradiation therapy, because of its demonstrated benefits in terms of reduced local relapse and ­reduced toxicity, has become the preferred adjuvant modality for the treatment of stage II or III rectal cancer.181 The chemotherapy agents used most often for this purpose have been 5-FU, given in bolus doses or as continuous infusions, with or without leucovorin. Capecitabine, an oral prodrug that is metabolized to 5-FU, has been investigated for the neoadjuvant treatment of rectal carcinoma, and retrospective studies have suggested that capecitabine and 5-FU are equally effective in this setting.188,189 Postoperatively, the administration of adjuvant chemotherapy in the form of 5-FU with or without leucovorin is considered necessary to reduce the risk of systemic relapse.163,165,168,173,190 The addition of oxaliplatin to fluoropyrimidine adjuvant therapy (5-FU with or without leucovorin) may also improve disease-free survival and overall survival in stage III colon cancer.191,192 The agents used in adjuvant chemotherapy for locally advanced rectal cancer are reviewed briefly in the following paragraphs; further discussion of agents used in combination with radiation therapy for rectal cancer ­appears in the section on “Other Treatment Approaches.” 5-Fluorouracil. The type, dose, means of administration, and timing of the various components of combined-­modality therapy are critical issues in optimizing ­treatment. A comprehensive review of the antitumor ­effect and ­toxicity of 5-FU–based chemotherapy regimens, for example, supported the idea that 5-FU can be considered two distinct chemotherapeutic agents on the basis of its mode of delivery—bolus or continuous infusion.193 (A third mode, oral delivery, is discussed under “Other Treatment Approaches”). First, the two modes of intravenous delivery produce different toxic effects; bolus schedules cause leukopenia, mucositis, and diarrhea, ­ whereas continuous-infusion schedules cause stomatitis and ­dermatitis.193 Second, continuous infusion of 5-FU allows much higher dose-intensities to be achieved (1625 to 2875 mg/m2/week) than are possible with bolus dosing (500 to 750 mg/m2/week). Moreover, in most North American trials, bolus 5-FU is given over a 1- to 2-minute period, a practice that produces severe mucositis or diarrhea in about 30% of patients treated. In European trials, where bolus 5-FU is administered over a 15-minute period, these effects occurred in only 10% of patients.193 The effectiveness of biomodulation of 5-FU with other agents (e.g., leucovorin, semustine, irradiation) depends on

580

PART 6  Gastrointestinal Tract

the schedule of administration.193 In a large trial of 1696 patients with rectal cancer,194 the effectiveness and toxicity of three types of chemotherapy were compared: bolus 5-FU alone, 5-FU with leucovorin or levamisole, or 5-FU with leucovorin and levamisole. All patients were to be given two cycles of chemotherapy followed by pelvic ­ irradiation with chemotherapy and two more cycles of ­ chemotherapy.194 At a median follow-up time of 48 months, no advantage was apparent from the use of ­biomodulation; however, more gastrointestinal toxicity was experienced in the group that took all three drugs than in the group that took only ­ bolus 5-FU.194 A subsequent report at 7.4 years190 essentially confirmed these findings, leading the authors of the study to conclude that leucovorin- or ­ levamisole-containing regimens had no advantage over bolus 5-FU alone in the adjuvant treatment of rectal cancer when combined with irradiation. They also reported that the local and distant ­recurrence rates were still high, especially among those with T3N+ and T4 disease, even with full adjuvant chemoradiation therapy.190 Differences in effectiveness and toxicity between ­bolus and infusional 5-FU were evaluated in a study of 660 ­patients with stage II or stage III rectal cancer.195 Adjuvant 5-FU was given at a dose of 500 mg/m2 on days 1 to 5 and days 36 to 40, and 450 mg/m2 was given on days 134 to 138 and 169 to 173. During pelvic irradiation, which began on day 64, patients received 500 mg/m2 bolus 5-FU for 3 consecutive days during weeks 1 and 5 of irradiation or received infusional 5-FU given at a rate of 225 mg/m2, 7 days per week.195 The infusional 5-FU schedule produced significant improvements in the time to relapse and overall survival rates over those of bolus 5-FU, and local ­control rates were similar in the two treatment groups. The improvements in time to relapse and overall survival rate were attributed to the systemic effect of 5-FU during radiation therapy. The average dose of 5-FU during irradiation was 6546 mg/m2 when given by infusion and 2499 mg/m2 when given by bolus technique. The higher total doses and the more prolonged exposure to 5-FU were thought to ­enhance cytotoxicity. The toxicity profiles were also ­different; bolus 5-FU was associated with more ­hematologic effects, and infusional 5-FU was associated with more gastrointestinal effects.195 Outcomes based on schedule of chemotherapy administration were evaluated by meta-analysis for 1219 patients given 5-FU by bolus or continuous infusion. The tumor ­response and overall survival rates were significantly higher among patients who received the drug by continuous infusion.196 Only performance status and the type of 5-FU administration predicted tumor response; these factors and the primary tumor site (rectum or colon) predicted survival. Grade 3 or 4 hematologic toxicity occurred more often in the bolus-infusion group than in the infusion group (31% versus 4%). In contrast, hand-foot syndrome occurred more often with infusional 5-FU (34% of the ­infusion group versus 13% of the bolus group). Capecitabine. The pharmacokinetics and practical ­ advantages of the orally bioavailable 5-FU prodrug capecitabine (Xeloda) have led many to shift practice from the use of infused 5-FU to that of orally administered capecitabine during radiation therapy for locally ­advanced rectal cancer. The tumor selectivity of capecitabine has been demonstrated in preclinical studies in which

5-FU concentrations were higher in tumor cells and ­lower in normal tissues after capecitabine than after bolus or continuous-infusion 5-FU.197 Although the initial acquisition costs are higher for capecitabine than for 5-FU, the higher administration costs associated with 5-FU result in capecitabine having a 57% lower overall cost.198 Adverse events associated with capecitabine administration also resulted in fewer medications and hospitalizations than those associated with bolus or infusional 5-FU, and societal costs, such as travel and time, were reduced by 75% with capecitabine.198 Findings from two prospective trials199,200 and a retrospective evaluation189 showed no significant differences in rates of pathologic response, sphincter preservation, local or distant recurrence, and overall survival for patients treated with preoperative capecitabine compared with preoperative 5-FU; rates of grade 3 or 4 acute toxicity were also low and similar among these studies.189,199,200 Other prospective phase II studies have shown that clinical outcomes after capecitabine are comparable to those after infusional 5-FU.201 Oxaliplatin. Oxaliplatin (trans-l-diaminocyclohexane oxalate platinum II) is a third-generation platinum derivative first synthesized in Japan and later developed in ­Europe as a less toxic and more effective alternative to cisplatin. Preclinical studies indicate that oxaliplatin is a potent ­radiosensitizing agent.202,203 Oxaliplatin was ­approved in Europe in 1996 and introduced into the standard therapy for metastatic colorectal cancer, where it showed ­convincing disease responses when added to 5-FU–based therapy.204 When combined with fluorouracil and leucovorin, oxaliplatin is among the most effective chemotherapies for metastatic colorectal cancer.205-208 Results from two studies have shown that the addition of oxaliplatin to 5-FU and leucovorin can improve disease-free survival rates of ­patients with stage II or III colon ­cancer.191,209 Other Biological Agents. Attempts to enhance the likelihood of response to chemoradiation therapy have ­included combining various systemic agents with different mechanisms of action. Although combining therapeutic approaches is theoretically promising, this strategy requires careful evaluation of the potential for additive toxicity relative to clinical outcomes. For example, preclinical ­studies have shown enhanced cytotoxicity from the combination of gefitinib, an EGFR inhibitor, with radiation and chemotherapy,210,211 particularly chemotherapy that ­ includes capecitabine. Significant toxicity (mostly grade 3 or 4 diarrhea) was observed in a phase I trial of pelvic irradiation (50.4 Gy) with capecitabine and gefitinib for ­ locally advanced rectal cancer, and a phase II dose level could not be determined because of that toxicity.212 In summary, combined-modality therapy has significantly improved the rates of local and distant disease control and has become the standard of care for locally advanced rectal cancer. However, identifying the optimal components or schedules for such therapy remains a challenge.

Radiation Therapy Radiation therapy has three general roles in rectal cancer management. First, radiation is used to enhance locoregional control by eliminating microscopic residual disease around the primary tumor and in the draining ­lymphatics. Second, significant tumor regression can result when

24  The Colon and Rectum

r­ adiation is administered preoperatively to locally advanced primary tumors. In these cases, an inoperable lesion can ­ become resectable or amenable to a more conservative surgical approach that involves sphincter preservation. Third, radiation therapy can be used to palliate symptoms arising from the infiltration of pelvic structures or metastatic disease. The prognostic factors that have been identified for ­rectal cancer treated by surgery alone reflect the risk for residual microscopic disease; these factors include stage, tumor location, and the presence of serosal, lymphatic, and neurovascular invasion (see “Surgical Approaches”).213,214 Adjuvant radiation eradicates microscopic residual ­tumor and eliminates local recurrence in approximately 90% of patients with adverse prognostic factors. Extension of ­tumor into the perirectal fat or adjacent viscera (stage II, T3-4 N0) raises the rate of local recurrence to approximately 30% and that of tumor in the regional lymphatics to more than 50% in patients treated with surgery only. Adjuvant ­ radiation can reduce these recurrence rates ­substantially. Adjuvant radiation can be administered before or after surgery. Regardless of the sequence, the primary goal is to improve locoregional control by eradicating microscopic residual disease. Each approach has advantages and disadvantages. Prospective randomized assessment of preoperative versus postoperative adjuvant radiation has been attempted, but because of low patient accrual, the studies had to be closed before they were completed. Emphasis has shifted to other issues, such as the combination of newer systemic agents with radiation and studies of fractionation schedules.162 The following paragraphs review radiation techniques and describe available information on the use of radiation with surgery, with or without chemotherapy, for the treatment of rectal cancer.

Radiation-Field Arrangement Careful attention to radiation technique is important regardless of whether the radiation is delivered before or ­after surgery. Whenever possible, the small bowel should be displaced from the treatment portal because the volume

581

of small bowel in the radiation portal correlates directly with radiation toxicity.215 Use of a belly board to displace the small bowel from the radiation portal helps to reduce gastrointestinal side effects (Fig. 24-4).171 Comparisons of CT treatment plans made when the patient is supine and those made when the patient is prone and the belly board is used have shown the median reduction in exposed small-bowel volume to be 54% and that of the bladder to be 62%.216 The corresponding median doses to the small bowel were 24 Gy (supine) and 15 Gy (prone) with the belly board. When the belly board cannot be used because of the size of the patient or extension of the tumor to anterior structures, treatment while the patient is prone is still preferred over having the patient supine. Another study showed that the belly board could reduce the irradiated small-bowel volume by as much as 70%, regardless of the patient’s weight, age, or sex or whether the radiation was given preoperatively or postoperatively.217 Exclusion of small bowel from the radiation field is particularly important because diarrhea can result from ­radiation or chemotherapy. Maximizing the doses of radiation therapy and chemotherapy that can be given without treatment interruption requires as much small bowel to be excluded as possible and aggressive supportive care to be provided during treatment. In most cases, the radiation treatment portals encompass the entire pelvis through a posterior and a right and left lateral treatment portal. High-energy photons (e.g., 18 MV) should be used to reduce the integral dose. Unlike the treatment of other pelvic malignancies, the external iliac lymph node chain need not be routinely included in the treatment portal.218 Anterior fields should be considered when disease extends anteriorly to the urogenital regions. In some cases, anterior-posterior fields or a four-field arrangement may prove necessary because of ­tumor extension or anatomic constraints. Adequate coverage of the tumor volume should be confirmed with treatment-planning imaging. If an anterior portal is used, the dose to the small bowel must be determined. The volume of small bowel in the field can be determined by using contrast at the time of the simulation or by CT-based treatment planning.

Inset: Demonstrating open table top

MODIFIED OPEN TABLE TOP: Face mask and arm support increase patient comfort and setup reproducibility

Duodenum

Rectum Lateral XRT portal avoids small bowel

A

Small bowel shift with patient in prone position

B

FIGURE 24-4.  Use of a belly board in the treatment of colorectal cancer, A, The patient lies prone on a modified table top.  B, A treatment planning CT scan confirms displacement of the small bowel. XRT, radiation therapy. (A, From Rich T, Ajani JA, ­Morrison WH, et al. Chemoradiation therapy for anal cancer. Radiation plus continuous infusion of 5-fluorouracil with or without ­cisplatin. Radiother Oncol 1993;27:209-215.)

582

PART 6  Gastrointestinal Tract

Even in cases of distal rectal involvement, including i­nfiltration of the anal canal, the risk of inguinal node ­involvement is limited, and radiation treatment portals need not include the inguinal region. The risk of recurrence in the inguinal region for distal rectal tumors is less than 5%.219 However, radiation techniques should be modified if inguinal lymph nodes are evident clinically and confirmed pathologically to be involved with metastatic disease. Metastatic involvement of these nodes portends an especially poor prognosis with regard to dissemination of disease.220 The superior border of the radiation field usually should be placed at the L5-S1 interspace. However, if the rectosigmoid is involved, individual anatomy may dictate that the superior border be placed at the L4-L5 interspace (Fig. 24-5). The inferior border of the field is dictated by the inferior extent of the tumor; the inferior aspect of the field usually should be placed 2 cm inferior to the lowest aspect of the tumor. The anus may need to be included in the treatment field in some cases when the tumor is located in the distal rectum. Whenever possible, however, the anus should be blocked from the radiation portal to reduce treatmentrelated morbidity. For tumors located in the middle and proximal rectum, the inferior border of the radiation field is placed at the bottom of the obturator foramen. The ­lateral border should include the sacroiliac joints, and a 2-cm margin should be placed around the pelvic brim to include the internal iliac lymph node chains in the ­posterior treatment portal. For the lateral treatment fields, the superior and inferior borders should be consistent with the posterior portal. In routine simulations, contrast should be placed in the rectum to ensure that margins around the primary tumor are adequate. The anterior aspect of the field should be placed 2 cm in front of the most anterior aspect of the sacral

promontory and anterior to the femoral heads to include the obturator (internal iliac) nodes. Every effort should be made to exclude small bowel. To reduce morbidity to the genitalia, a block should be placed anterior to the femur. Posteriorly, the entire sacrum should be included. This is important to avoid the penumbra of the beam and ensure the adequacy of the radiation dose to the radial margin of resection in the presacral region. Care should be taken to exclude any chemotherapy infusion pumps from the radiation field.221

Patterns of Failure Few studies have been published that offer guidelines for defining and delineating clinical target volumes in treating rectal cancer. In an attempt to fill this gap, Roels and colleagues1 analyzed data from 17 studies of the location and frequency of local recurrences and lymph node involvement in rectal cancer. Of the five pelvic subsites identified (i.e., mesorectum, posterior pelvis, lateral pelvis, inferior pelvis, and anterior pelvis), the areas at greatest risk for ­local recurrence were the mesorectum, the posterior pelvis, and the inferior pelvis.1 In terms of lymphatic involvement, tumors spread mostly in three directions: caudally to the inferior mesenteric nodes, laterally to the internal iliac nodes, and inferiorly to the external iliac and inguinal nodes. Although the incidence of local tumor recurrence in the mesorectum has declined drastically with the advent of total mesorectal excision, the risk of lymphatic recurrence remains. The risk for inferior pelvic region involvement was highest among patients with a tumor at or within 6 cm from the anal margin and in patients who had undergone an abdominopelvic resection.1 On the basis of these results, the authors of this ­review recommended that the clinical target volume should ­encompass the tumor, the mesorectum, and the posterior

Prone PA Right

A

B

FIGURE 24-5.  Routine radiation treatment portals for rectal cancer. A, Posterior field. B, Lateral fields, showing displacement of the bowel.

24  The Colon and Rectum

pelvis in all cases. The inferior pelvic subsite is also at risk if the tumor is located within 6 cm from the anal margin and the patient is to undergo sphincter-sparing surgery or if the tumor invades the anal sphincter and an abdominoperineal resection is needed. In terms of nodal coverage, the mesorectal and lateral nodes should be included in the clinical target volume for all patients.1 Patterns of failure for individual patients should be outlined on serial CT slices and incorporated into the CT-based treatment plan. This practice facilitates maximizing the conformality of the treatment, which reduces the volume of small bowel in the treatment field and the treatment-related toxicity.

Endocavitary Radiation and Brachytherapy Brachytherapy and endocavitary radiation can be used alone or, more commonly, in conjunction with external beam radiation therapy (EBRT) to treat rectal cancer. The delivery of high doses of radiation directly to well-defined, circumscribed volumes can be effective for local control. These approaches are most often used for patients who cannot undergo surgical resection. Endocavitary Treatment. In contact or endocavitary radiation therapy, soft (50-kV) x-rays are produced by a generator that delivers radiation through a tube containing an anode and ring filament.222 Of the two types of ­aluminum filters available, the 0.5-mm filter is used for most applications. Using this filter at a focal distance of 4 cm, an output of 20 Gy per minute, and a 30-mm circular field, the percentage depth doses are 100% at 0 mm depth, 44% at 5 mm, 23% at 10 mm, and 9% at 20 mm. This depth distribution precludes the use of contact therapy to treat highly infiltrative tumors. The tumor must be within 12 cm of the anal verge and accessible; posterior wall tumors can ­occasionally be difficult to localize with the cone. The typical dose in the first contact treatment is ­between 30 and 40 Gy, given in 2 to 4 minutes with a 0.5-mm aluminum filter. Between 1 and 2 weeks later, a second application of contact radiation therapy is given. Three weeks after the initiation of therapy, the third contact application is given; at this time, a smaller (2-cm) cone typically is used to account for tumor regression. For 3-cm or larger tumors, 90 to 120 Gy is given in 4 to 5 fractions over 5 to 8 weeks. For tumors smaller than 2 cm that regress completely after two sessions, a total dose of 80 Gy in 4 fractions (30 Gy on day 1, 20 Gy on day 7, 15 Gy on day 21, and 15 Gy on day 36) is sufficient.222 The normal mucosa of the rectal wall should not receive more than 15 to 20 Gy per fraction. As is true for local excision, patient selection is critical for the success of contact therapy. Patients with lymph node involvement are not candidates for contact therapy.222 Factors that increase the risk of lymph node involvement, such as high histologic grade, perineural or lymphovascular invasion, or ulceration, usually are contraindications for contact therapy. The response to contact therapy predicts therapeutic success. The probability of local control is good if the tumor regresses by more than 80% of its original volume by 2 weeks after the second contact session.222 Endocavitary therapy can produce complete response rates of more than 90% in carefully selected patients.223 Although the local failure rate can be as high as 28%, the addition of EBRT in cases involving poor prognostic factors (e.g., tumors larger than 3 cm or tumors that are partially fixed to the pelvic sidewall) can improve local control. In

583

one study of patients so treated,223 the local control rate (after salvage therapy) was 82%, and sphincter preservation was possible in 82% of patients. The incidence of ­severe late effects was 4%. Brachytherapy. The choice of brachytherapy technique depends on anatomic considerations. The four techniques available involve the use of metallic needles through a perineal template, an iridium “fork,” a plastic loop, or a remote afterloading device.222 To localize the radiation dose to the lesion and to avoid irradiating the opposite rectal wall, ­obturators are often used with the implants to distend the rectum. Lesions smaller than 3 cm can be treated with brachytherapy alone, but EBRT should be added for larger ­tumors. Using this approach, local control rates of 90% can be achieved without significant complications.224 ­Although continuous low-dose-rate brachytherapy typically is used, pulsed low-dose-rate brachytherapy results in similar late effects and can be considered as an alternative ­approach.225 Endocavitary therapy and brachytherapy can be effective means of administering localized radiation, especially for patients with limited treatment options. Further integration of these techniques should be considered for use with conservative surgical approaches.

Postoperative Radiation Therapy The advantages of administering radiation therapy postoperatively include the ability to plan treatment on the basis of the pathologic response of the primary tumor and, theoretically, a reduction in perioperative morbidity because surgery is performed in an unirradiated field. Pathologic ­tumor staging can confirm the need for adjuvant radiation and allows precise definition of the regions that are to be included in the radiation treatment portal.226 The disadvantages of postoperative radiation include the relative hypoxic state within the operative bed and the associated potential for growth of residual disease during postoperative healing. Hypoxia enhances resistance to radiation; approximately three times the radiation dose is required to kill hypoxic tumor cells as to kill the same number of well-oxygenated cells. The total dose of radiation that is prescribed usually is higher when it is given after surgery (53 to 55 Gy) than before surgery (45 to 50 Gy). However, the risk of short- and long-term gastrointestinal toxicity increases precipitously as the total dose of radiation increases.227 The potential risk of radiation morbidity from postoperative ­radiation also increases because the small bowel is less mobile because of the formation of adhesions after surgery.

Preoperative Radiation Therapy The key advantage of preoperative radiation is the ­potential for shrinking the tumor, thereby improving the chance that a sphincter-preserving surgical procedure can be used. This advantage is especially important for ­lesions located in the distal rectum. Preoperative radia­ tion also offers the opportunity for sterilizing potential sites of residual disease, such as the radial surgical margin and ­ regional ­ lymphatics.228-231 Another advantage is that the effects of preoperative radiation and chemotherapy may be enhanced because the blood supply in the ­affected area has not been disrupted by surgery. A lower total dose

584

PART 6  Gastrointestinal Tract

of ­ radiation can be given, which reduces the risk of radiation morbidity. That risk is also less after preoperative ­radiation than postoperative radiation because the small bowel is more mobile before surgery, allowing it to be more easily displaced from the radiation field.232 The ability to ­assess response to induction chemotherapy and radiation may have prognostic implications. Techniques for staging ­rectal cancer before and after neoadjuvant chemoradiation therapy have included endorectal sonography, CT, and MRI.110,233-235 The disadvantages of preoperative radiation ­include not having information from pathologic tumor ­ staging,228,229 the need for particular surgical expertise to minimize the ­ perioperative morbidity associated with ­ resecting an ­irradiated field,157,236 and the need for a two-stage surgical approach (i.e., resection with fecal diversion followed by re-establishment of intestinal continuity) to allow the anastomosis to heal. Concerns that a hypofractionated course of radiation (5 fractions of 5 Gy) given before ­total ­ mesorectal excision would hinder postoperative recovery were not borne out in a review of 1530 patients in a ­ prospective randomized trial by the Dutch Colorectal ­ Cancer Group.237 In that study, essentially no acute ­radiation ­ effects were experienced. ­ Patients who underwent radiation ­therapy before abdominoperineal resection had more perineal complications than did those who had ­radiation before a low anterior ­ resection, and the overall total number of complications was slightly higher in the irradiated group. No differences were identified, however, in the number of surgical reinterventions needed or in ­postoperative ­mortality rates.237 The Dutch Colorectal Cancer Group trial also included a study of the cost-effectiveness of preoperative radiation.238 Investigators used a Markov model to estimate the societal costs and quality-adjusted life expectancy of 1861 patients with resectable rectal cancer treated with total mesorectal resection with or without 5 fractions of 5 Gy of radiation therapy before the surgery. The loss in quality of life from the preoperative radiation therapy was found to be outweighed by the gain in life expectancy; specifically, life expectancy increased by 0.67 years, quality-adjusted life expectancy by 0.39 years, and costs by $9800 per patient. The corresponding cost-effectiveness ratio, $25,100 per quality-adjusted life-year, remained at acceptable amounts under a wide range of assumptions regarding the extent of reduction in local recurrence and improvement in survival, leading the authors of the study to conclude that this strategy was both effective and cost-effective.238 A meta-analysis of 14 randomized controlled trials of preoperative radiation therapy for resectable rectal ­cancer (6426 patients) indicated that preoperative radiation did not increase postoperative mortality, although it did increase morbidity in some studies.239 Moreover, preoperative radiation was found to reduce the 5-year overall mortality rate, cancer-related mortality rate, and local ­recurrence rate, but no reduction was seen in the incidence of distant metastases.239

Local Control and Survival Among the three randomized trials that evaluated the use of postoperative radiation without chemotherapy (conducted by the NSABP,168 the GITSG,166 and the NCCTG164), only the NSABP trial showed a modest improvement in the

rates of local failure over surgery alone (see Table 24-1).168 However, the total doses of radiation used were different in these trials; the NSABP trial involved doses of 46 to 47 Gy,168 the GITSG trial used 40 or 48 Gy,166 and the Mayo/NCCTG trial used 45 to 50.4 Gy.164 The NSABP trial was the only one in which the radiation was given as a continuous course and using modern techniques.226 The GITSG166 and NCCTG164 studies demonstrated a benefit for combined-modality therapy over surgery only, surgery plus radiation, or surgery plus chemotherapy. The results of the treatment arms for these studies are shown in Table 24-3.164,166,195,240 The effect of adjuvant radiation on local control and survival was also evaluated in the Uppsala and the ­ Swedish Rectal Cancer Trials. In those trials, the addition of radiation to surgery decreased the risk of local failure over the risk conferred by treatment with surgery only.236,241,242 The use of preoperative radiation was found to improve survival rates over those conferred by surgery only.236 Comparisons of preoperative and postoperative radiation were difficult to interpret because of significant differences in the fractionation schedules used.243,244 The issue of ­whether sphincter preservation could be used after preoperative radiation was not addressed because not enough time was allowed for downstaging between the completion of radiation and surgery. A prospective randomized trial by the Stockholm Colorectal Cancer Study Group compared the use of preoperative radiation with surgery for 285 patients with rectal cancer at any stage.241,242 Local recurrence occurred in 16% of the group given preoperative radiation and in 30% who underwent surgery only (P < .001). The advantage was confined to patients with Dukes stage B disease, although this also might have been true for patients with Dukes stage C disease (P = .068). At a median follow-up of 50 months, the rate of distant metastases was 19% after preoperative radiation and surgery and 26% after surgery alone (P = .02), a difference that translated into a survival advantage for irradiated patients (P = .02). Although the disease-free interval was initially longer in the group given preoperative radiation, the lower rate of distant metastases and overall survival ­advantage after preoperative radiation did not last when the follow-up interval extended to 107 months. This loss in survival advantage, however, was accounted for by the higher perioperative mortality rate in the preoperative radiation group, because radiation therapy continued to confer an advantage in terms of disease-specific survival (P < .01). Advantages of preoperative radiation therapy were shown in the Swedish Rectal Cancer Trial, in which patients were randomized to undergo surgery only or preoperative radiation followed by surgery.236 Consistent with the previously reported Stockholm experience,242 preoperative therapy totaled 25 Gy, given in 5 fractions over 1 week, followed by surgery within 1 week after radiation. After 5 years of follow-up, the local recurrence rate was 11% for those given preoperative radiation and 27% for those given surgery only (P < .001).236 Improvements in local control were observed for all stages of disease. The overall survival rate at 5 years was also significantly improved by the addition of preoperative radiation (58% versus 48%, P = .004). The cancer-specific survival rates at 9 years were 74% for

24  The Colon and Rectum

585

TABLE 24-3.    Outcomes in Three Large Trials of Treatment for Colorectal Cancer Local Relapse Rate (%)

Distant Relapse Rate (%)

Disease-Free ­Survival Rate (%)

Overall Survival Rate (%)

P Value

Surgery

24

34

45

32

0.005

Surgery + chemotherapy*

27

27

54

48

Surgery + radiation

Trial and Treatment Regimen GITSG (N =

202)166

20

30

52

45

chemotherapy*

11

26

67

57

Surgery + radiation

25

46

38

38

Surgery + radiation + chemotherapy*

14

29

58

53

Surgery + radiation + bolus

NS

NS

53

60

Surgery + radiation + PVI fluorouracil

NS

NS

63

70

Surgery + radiation + NCCTG (N =

204)164† 0.025

INT (N = 660) 195‡ 0.01§

*Chemotherapy

was fluorouracil and semustine. follow-up. ‡Four-year follow-up. §Fluorouracil. From Vaughn DJ, Haller DG. Adjuvant therapy for colorectal cancer: past accomplishments, future directions. Cancer Invest 1997;15: 435-447. INT, intergroup; GITSG, Gastrointestinal Tumor Study Group; NCCTG, North Central Cancer Treatment Group; NS, not stated; PVI, ­protracted venous infusion. †Seven-year

the preoperative radiation group and 65% for the surgeryonly group (P = .002). Updated results at a ­median of 13 years of follow-up continued to demonstrate the benefit from preoperative radiation: overall survival rates were 38% in the radiation-plus-surgery group and 30% in the surgery-only group; the corresponding cancer-specific survival rates were 72% and 62%, and the local recurrence rates were 9% and 26%.245 Outcomes after preoperative or postoperative radiation were compared in a trial conducted at Uppsala University.243 In that study, 471 patients with resectable rectal cancer were assigned to undergo a short course of preoperative radiation or, for those who had Dukes B or C disease, high-dose postoperative radiation. For the 236 patients given preoperative radiation, radiation was delivered in 5 fractions of 5.1 Gy (total dose of 25.5 Gy) over 5 to 7 days, and resection was performed 1 week later. For the 235 patients given postoperative radiation, radiation treatment was initiated 4 to 6 weeks after surgery. ­ Patients were given 40 Gy in 2-Gy fractions over a period of 4 weeks; at that point, radiation was discontinued for 10 to 14 days, after which an additional 10 Gy was administered to the entire pelvis and another 10 Gy was administered to a ­reduced treatment volume. (In this regimen, the pelvis was treated to 50 Gy and the tumor bed to 60 Gy.) Only patients who had evidence of transmural extension of disease or positive pelvic nodes at resection received postoperative chemotherapy. Although the radiation doses and the effective biologic doses were substantially different among treatment groups in this trial, a local control benefit was observed for preoperative radiation over postoperative radiation; local failure rates at 5 years were 13% and 22%, respectively. No difference was evident in survival rate ­between the groups.

Toxicity In the trials described in the previous section, moderate to mild acute radiation effects were observed in virtually all patients ­given postoperative radiation; in contrast, preoperative radiation led only infrequently to acute radiation effects such as diarrhea and cystitis.243,244,246 Perioperative complications included small-bowel obstruction, diagnosed radiographically and requiring surgical intervention, in 6% after surgery only, 5% after preoperative therapy, and 11% after postoperative radiation. However, perioperative complications were more common after preoperative therapy. For example, perineal wound sepsis occurred after abdominoperineal resection in 33% of those given preoperative radiation and in 18% of those given postoperative radiation.243,244,246 Because of perioperative complications, one half of the patients in the postoperative group could not start radiation therapy within the recommended 6 weeks after surgery. Although the Stockholm trials did identify advantages for combined-modality therapy, significant late morbidity was also observed. Peripheral nerves are relatively resistant to the late effects of radiation therapy, but lumbosacral plexopathy occurred after the preoperative radiation regimen used in that trial.243 Other late adverse effects of that preoperative radiation regimen included thromboembolism (P = .01), femoral neck and pelvic fractures (P = .03), intestinal obstruction (P = .02), and postoperative fistulas (P = .01). No difference in genitourinary complications was observed between groups. The radiation techniques that were used significantly influenced the risk of complications and mortality. The incidence of complications resulting from small bowel toxicity was high because the superior border was placed at L2 in many of the studies rather than at the L5-S1

586

PART 6  Gastrointestinal Tract

­interspace.243,244,246 Mortality rates were higher among patients treated with anterior-posterior opposed portals than among those treated with three- or four-field techniques (15% versus 3%, P < .001). Because of these complications, the overall mortality rates were the same for those treated with surgery only and those treated with preoperative radiation. Although the addition of radiation therapy significantly improved the local control and disease-specific survival rates, the substantial morbidity of radiation therapy was enough to negate the effect of these improvements. In another experience,247 giving hypofractionated ­radiation preoperatively to 83 patients did not result in significant treatment-related morbidity. The dose given was similar to those given in the Swedish trials236 (25 Gy in 5 fractions over 1 week followed immediately by surgery), but anterior-posterior portals were avoided, and the superior border of the radiation field was limited to the L5-S1 interspace. At a mean follow-up of about 5 years, the actuarial local control rate was 95%, and the diseasespecific survival rate was 77%.247 The incidence of grade 3 or higher perioperative or late toxic reactions was only 13%; the incidence of bowel obstruction was 3.5%. No significant difference in toxic reactions was observed among the 16 patients who were also given chemotherapy during the radiation. Although not a prospective randomized trial, results from a 1988-1989 Patterns of Care study187 allow comparison of survival on the basis of whether radiation was given preoperatively or postoperatively and whether chemotherapy was given during radiation therapy. Local failure rates were approximately 10% for all stages of disease and in all treatment groups. Survival rates were significantly higher after preoperative radiation or postoperative chemoradiation therapy than after postoperative radiation only.187 At 5 years, the overall survival rates were 50% in the postoperative radiation group, 85% in the preoperative radiation group, and 69% in the postoperative chemoradiation therapy group.187 A significant dose-volume effect has been observed between radiation dose and the volume of small bowel that is irradiated with regard to treatment toxicity. In one study, 41 patients were treated to 45 Gy over 5 weeks, administered in three or four fields; 5-FU (350 mg/m2) was given as a 1-hour infusion with a folinic acid bolus on week 1 and week 5 of the radiation treatment. The severity of diarrhea correlated strongly with the volume of irradiated small bowel at every dose level, with the strongest correlation at the lowest doses. Highly significant differences in the median volume of irradiated small bowel were identified between patients experiencing grade 0 to 1 versus grade 2 to 4 diarrhea; in contrast, no correlation was found for anorexia, nausea, vomiting, abdominal cramps, age, body mass index, sex, tumor position, or number of fields used.248 At higher radiation doses, small increases in the small bowel volume in the radiation field resulted in more severe treatment-related diarrhea. Assessment of inverse treatment plans generated for eight patients with the highest volumes of irradiated small bowel showed that in all eight cases, use of seven-field intensity-modulated plans would have been superior to five-field plans in terms of planning target volumes and normal tissue dosimetry; this type of planning would have reduced the median dose to the small bowel by 5.1 Gy and reduced the probability of late effects on normal tissues by 67%.248

Another group of investigators,249 studying a larger group of 96 patients, confirmed a highly significant association between the small bowel dose-volume effect and acute grade 3 diarrhea in patients receiving preoperative or postoperative pelvic radiation therapy for rectal cancer. An apparent difference in the risk of grade 3 diarrhea between patients who received preoperative radiation (18%) and those who received postoperative radiation (28%) was not statistically significant.249 Collectively, these studies demonstrate the benefit of radiation therapy in terms of reducing the risk of local failure and possibly affecting overall survival. Variations in radiation techniques and in whether chemotherapy was administered significantly affect morbidity and outcome. Although it is difficult to interpret outcome when radiation was given with planned treatment interruptions, the treatment interruptions might have improved tolerance to therapy. Some trials included chemotherapy agents that provided no benefit but might have contributed to the overall toxicity of therapy. The routes and schedules of ­administration of chemotherapy also significantly influence the patterns of failure.

Sphincter Preservation Sphincter preservation is a major goal of preoperative radiation therapy. The current use of preoperative chemoradiation therapy builds on the experience with postoperative combined-modality therapy and the data that demonstrate the efficacy of preoperative radiation alone.250,251 ­Tumors located less than 6 cm from the anal verge usually require an abdominoperineal resection. However, variation ­ exists in surgical practices regardless of whether preoperative radiation is used. Among 18,695 cases of operable rectal cancer in Ontario, wide geographic variations were demonstrated in rates of abdominoperineal resection and referrals for postoperative radiation.252 Sphincter preservation rates after preoperative radiation or chemoradiation therapy range between 65% and 85%.228,229,253 These variations probably are related to details of the clinical presentation, including tumor size, circumference, and extent of invasion of the sphincter muscles. However, an interim analysis of the NASBP R-03 trial showed that preoperative therapy allowed only 27% of patients to undergo a sphincter-preserving procedure rather than an abdominoperineal resection.254 The reasons for this were unclear. With the goals of preserving sphincter function, minimizing the total radiation dose, and maximizing tumor oxygenation, many centers have evaluated the use of preoperative radiation therapy, with the most recent series also using chemotherapy in conjunction with the radiation therapy. The results of several of these series are summarized in Table 24-4.157,255-263 Preoperative radiation produces pathologic complete response rates of about 15%. The addition of bolus or infusional 5-FU during the first and last week of radiation increases the rate to about 20%, and use of a continuous infusion of 5-FU throughout the course of preoperative radiation further increases the rate to 27%.157 Tumor regression or downstaging is evident in about 65% of cases. The experience at The University of Texas M.  D. ­Anderson Cancer Center suggests that sphincter preservation after preoperative chemoradiation therapy is possible in more than two thirds of patients with low rectal

TABLE 24-4.    Preoperative Radiation Therapy and Sphincter Preservation, Pathologic Response, and Tumor Downstaging in Colorectal Cancer Study and Reference

Number of Patients

Preoperative ­ Radiation Dose

Chemotherapy during XRT

Time between XRT and Surgery

Complete Response Rate (%)

Downstaging Rate (%)

Sphincter Preservation Rate (%)

Janjan et al, 2000256

42

45 Gy to pelvis + 7.5-Gy CB

CI 5-FU 300 mg/m2 5 days/week during pelvic XRT

4-6 weeks

31

86

79

Bosset et al, 2000257

60

45 Gy

Bolus 5-FU week 1, 350-450 mg/m2 week 5, 350-370 mg/m2

3 weeks (range, 13-111 days)

15.6

30

58

Pucciarelli et al, 2000258

51

45 Gy

5-day 5-FU infusion 350 mg/m2 days 1-5 and days 29-33

4-5 weeks (range, 19-53 days)

15.7

59

84

Valentini et al, 1999259

40

45 Gy to pelvis + 5.4-Gy boost

5-day 5-FU infusion 1000 mg/m2 days 1-4 and days 29-32; cisplatin 60 mg/ m2 days 1, 29

6-8 weeks

23

68

85

Movsas et al, 1998261

23

45 Gy to pelvis + hyperfx boost to total doses of 54.6, 57, or 61.8 Gy

5-day 5-FU infusion 1000 mg/m2 days 1-4 and days 29-32

4-6 weeks

17.4

57

30

Janjan et al, 1998157

117

45 Gy to pelvis

CI 5-FU 300 mg/m2; 5 days/week ­ during pelvic XRT

4-6 weeks

27

62

59

35

46.8 Gy to pelvis + 3.6-Gy boost

No

4-5 weeks

14

63

77

Mohiuddin et al, 1998263

70

40 to 45 Gy + 10- to 15-Gy boost

No

5-10 weeks

NR

30

86

CB, concomitant boost; CI, continuous infusion; hyperfx, hyperfractionated; NR, not reported; XRT, radiation therapy.

24  The Colon and Rectum

Wagman et al, 1998262

587

588

PART 6  Gastrointestinal Tract

­cancer.157 In an update of that experience,264 238 cases in which T3 or T4 rectal tumors located less than 6 cm from the anal verge treated with preoperative chemoradiation therapy between 1989 and 2000 were reviewed. Logistic regression analysis was used to analyze the influence of objective tumor response (defined as complete clinical response) and other prognostic factors on sphincter preservation. Of the 238 patients, 117 (49%) had sphincter-preserving surgery after preoperative chemoradiation therapy, and 47% had a clinical complete response. Independent predictors of sphincter preservation included the year of surgery, tumor distance from the anal verge, circumferential tumor involvement, and response to chemoradiation therapy. The sphincter preservation rate increased over time, from 28% between December 1989 and December 1992 to 44% between January 1993 and December 1996 to 67% between January 1997 and December 2000. The difference in the rates of sphincter preservation according to response to preoperative chemoradiation therapy was most striking among patients with tumors less than 3 cm from the anal verge (44% for those who showed a response versus 22% for those who did not, P = .01). Moreover, the incidence of pelvic disease recurrence (typically less than 10%) did not increase as the use of sphincter-preserving surgery became more common. The authors of the study concluded that a clinical response to preoperative chemoradiation therapy is not necessary for sphincter preservation, but such a response contributes to the likelihood of sphincter-preserving surgery, particularly for patients with low rectal tumors.264 Factors that can enhance tumor regression and increase the possibility of undergoing sphincter-sparing surgery ­include allowing an appropriate interval between completion of radiation therapy and surgery to maximize tumor regression, administration of higher radiation doses, and administration of chemotherapy.265 Issues specific to chemotherapy during preoperative radiation include the type of agent and the schedule of infusion. These topics are discussed briefly in the sections that follow. Radiation-Surgery Interval. The interval between completion of preoperative radiation and surgery is critical for allowing maximal tumor regression. In one study, when the interval between preoperative radiation (39 Gy in 3 weeks) and surgery was increased from 2 weeks to 6 to 8 weeks, the downstaging rate increased from 10% to 26%.266 Although it was intended that surgery be performed within 10 days of completion of radiation in the Swedish Rectal Cancer Trial, the interval between radiation and surgery varied widely (5 to 155 days).236 In that trial, the downstaging rate was found to be only 4% among patients who underwent resection within 10 days of completing radiation therapy; in contrast, the downstaging rate was 45% when more than 10 days elapsed between completion of radiation therapy and surgery. Another trial ­ specifically evaluated the interval between completion of radiation (39 Gy in 13 fractions) and pathologic downstaging after surgery among 201 patients.266 Pathologic ­ evidence of downstaging was evident in only 11% when surgery was performed within 2 weeks of completing ­ radiation therapy, but that rate increased to 26% when the interval after radiation was extended to 6 to 8 weeks.266 Radiation Doses. A possible disadvantage of increasing the total preoperative radiation dose is the potential

increase in treatment-related toxicity. Although this was not reported in a dose-escalation trial that included an intermittent infusion of 5-FU during radiation, no significant improvements in the rates of pathologic complete response, downstaging, or sphincter preservation were ­observed.261 In another study, a dose-response relationship for downstaging with preoperative radiation was identified at 44 Gy, given in conventional fractions.265 However, this relationship held for Dukes stage A and B disease but not for stage C disease. Administration of Chemotherapy. The influence of combined-modality therapy for locally advanced rectal cancer was evaluated by comparing the types of surgical procedures performed between 1975 and 1990 with those performed between 1991 and 1997 at the University of Florida. Of the 328 patients who underwent preoperative radiation therapy, 219 were treated in the early period and 109 were treated in the late period; none of the patients in the early era but 64 in the late era were given preoperative chemotherapy.267 The rate at which sphincter-sparing procedures were performed increased from 13% to 52%, and the rate of pathologic tumor downstaging increased from 42% to 58%. However, the incidence of lymph node ­metastases and 5-year local recurrence rates remained similar between groups. Adjustments for other independent prognostic factors, such as T category, nodal status, and age, led to significant improvement in the overall 5-year survival rate among the group treated in the late period (87% versus 58% for those treated in the early period).267 In this study, the addition of chemotherapy to radiation significantly improved local control and overall survival. Administration of 5-FU as a continuous infusion rather than a bolus during radiation seems to confer benefit in terms of disease-free and overall survival. This benefit has been shown in studies of postoperative chemoradiation therapy195 and indirectly in studies of preoperative radiation using conventional fractionation and a concomitant boost.157,256,268,269 Using a lower total dose of preoperative radiation (45 Gy) with continuous-infusion 5-FU produced downstaging comparable with higher total radiation doses.270 When accelerated fractionation (concomitant boost) was used to deliver a total radiation dose of 52.5 Gy with a continuous infusion of 5-FU, the pathologic complete response rates were nearly twice those of other series using the same or higher doses of radiation and bolus or intermittent infusion of 5-FU.256 The pathologic downstaging rates were approximately 20% higher than those achieved in other series that used bolus or intermittent infusion of 5-FU.256 Aside from its potential influence on sphincter preservation, tumor response to preoperative therapy may also be of prognostic importance. The level of response to preoperative radiation for locally advanced rectal cancer has been reported to influence survival rates, with improved survival rates observed among patients who showed pathologic ­evidence of downstaging.265 The 5-year overall survival rate after preoperative radiation was 92% for patients whose tumors were downstaged to Dukes stage A; the corresponding 5-year survival rate for Dukes B disease was 67%, and that for Dukes C disease was 26%. The corresponding diseasefree survival rates were 87%, 56%, and 28%.265 In another study, patients who had pathologic evidence of tumor downstaging after preoperative radiation

24  The Colon and Rectum

and bolus 5-FU had much higher cancer-specific survival rates (100%) than those whose tumors did not respond (45%).271 In that same study, the corresponding 5-year recurrence-free survival rates were 94% and 50%.271 A similar relationship was seen when 5-FU was given as a continuous infusion in combination with preoperative ­radiation.272 Any response to preoperative chemoradiation therapy resulted in reduced risk of distant metastases and in improved disease-free survival rates. Higher rates of tumor regression have been observed for tumors that shrink and have higher Ki-67 levels and higher mitotic activity in response to radiation.273 In that same study, having tumors that had elevated proliferative activity after preoperative radiation strongly correlated with improved survival.273 FDG-PET, which allows imaging of tumor function, is being used in investigational settings to evaluate response to preoperative chemoradiation therapy.274 In one study of 88 patients with locally advanced rectal cancer,275 FDG-PET findings correlated strongly with outcome after preoperative chemoradiation therapy. Five-year overall and disease-free survival rates for the entire group were 83% and 73%, respectively. Only two factors were found to be independently predictive of disease-free and overall survival: pathologic stage and FDGPET findings. Five-year survival rates for patients with no evidence of disease on PET were 91%, compared with 72% for patients with positive PET findings (P = .024). The corresponding disease-free survival rates were 81% and 62% (P = .003).275 In addition to total radiation dose, the risk of pelvic relapse is associated with overall radiation treatment time. A review of the literature on the incidence of pelvic relapse as a function of radiation dose and overall treatment time revealed that the dose-response curves are steep if the effect of overall treatment time is accounted for.276 The time-related displacement of these dose-response curves was consistent with a median tumor doubling time of about 4 to 5 days—in other words, more rapidly than the doubling time of the primary tumor. The authors of this study concluded that growth of subclinical tumor deposits in the pelvis accelerated rapidly during preoperative radiation, and strategies that shorten overall treatment time or provide uninterrupted administration of antineoplastic therapy should be investigated.276

Other Treatment Approaches Radiation Therapy with Oral Fluoropyrimidine Analogues The use of oral chemotherapy confers several advantages over the use of intravenous chemotherapy, among which are ease of administration and avoidance of the inconvenience, discomfort, and cost associated with intravenous 5-FU.274,277,278 Biologic advantages of the oral agents ­include the ability to concentrate the active fluoropyrimidine in tumor cells to a greater extent than infused 5-FU can and the ability to completely inactivate dihydropyrimidine dehydrogenase activity.279 These advantages have led to oral fluoropyrimidines being quickly adopted into routine clinical practice. Capecitabine. The most commonly used oral fluoropyrimidine is capecitabine, which is given during preoperative chemoradiation therapy at doses of 825 mg/m2 twice

589

daily throughout the course of radiation.280 Capecitabine is a prodrug that undergoes a three-step metabolic activation; the final step is its enzymatic conversion to 5-FU by thymidine phosphorylase, which some have found to be upregulated in colorectal cancer.281 The pharmacokinetic profile of capecitabine is comparable to that of infusional 5-FU, and prolonged exposure to 5-FU results in depletion of pyrimidine nucleotides and disruption of DNA synthesis through inhibition of thymidylate synthase.198 Capecitabine and 5-FU produce comparable rates of pathologic response, sphincter preservation, local or distant recurrence, and overall survival when used as preoperative therapy for rectal cancer (see “Chemotherapy Agents”).189,199,200 A retrospective review280 compared findings from 542 patients with locally advanced rectal cancer who were given continuous-infusion 5-FU (197 ­patients) or capecitabine (345 patients) concurrently during EBRT. Both regimens were generally well tolerated, and no grade 4 toxic effects were reported. Capecitabine produced a higher rate of pathologic complete response (25% versus 13% with 5-FU). The toxicity profiles differed in that hand-foot syndrome was more common with capecitabine and grade 3 diarrhea was more common with continuousinfusion 5-FU. The authors of this study concluded that the apparent superiority in efficacy and reasonable toxicity of capecitabine made it a reasonable choice for treating locally advanced rectal cancer.280

Radiation Therapy with Oxaliplatin In addition to having significant single-agent activity in colorectal cancer, oxaliplatin is a reasonable candidate for combined-modality programs because of its relative lack of acute dose-limiting adverse effects when added to radiation and 5-FU or capecitabine. As discussed under “Chemotherapy Agents,” oxaliplatin was added to fluoropyrimidine adjuvant therapy because of evidence of improved diseasefree survival and overall survival in patients with stage III colon cancer.191,192 Presumably, adding oxaliplatin to a fluo­ ropyrimidine-plus-radiation regimen for the neoadjuvant treatment of rectal cancer can enhance preoperative tumor downstaging and may reduce the duration of postoperative oxaliplatin-based adjuvant chemotherapy. The safety and efficacy of adding oxaliplatin to fluoropyrimidines and ­radiation for the treatment of locally advanced rectal ­cancer has been evaluated in several phase I/II trials. In one dose-seeking phase I trial of oxaliplatin plus capecitabine to be given during preoperative radiation therapy (50.4 Gy in 28 fractions) for stage II or III rectal cancer, diarrhea was the dose-limiting toxicity.282 The dose combination of capecitabine (725 mg/m2 given twice daily) with oxaliplatin (50 mg/m2/week) was recommended based on the toxicity profile. Endorectal tumor biopsy samples were obtained before and on the third day of treatment to ­explore the effects of treatment on thymidine phosphorylase, thymidylate synthase, dihydropyrimidine dehydrogenase, DNA repair, and apoptosis. No differences were seen between the baseline and day-3 values, but the apoptosis index was high at both measurement times among patients who achieved a complete pathologic response to treatment.282 These findings were later confirmed in a phase II study conducted by the same group283 to evaluate the ­efficacy of the capecitabine, oxaliplatin, and radiation combination as neoadjuvant therapy for stage II or III rectal

590

PART 6  Gastrointestinal Tract

cancer. Of the 25 patients enrolled, 6 (24%) had a complete pathologic response; 52% experienced T downstaging, and 53% experienced N downstaging. Grade 3 diarrhea was the most common grade 3 or 4 toxicity, occurring in 20% of patients. At a median follow-up time of 20 months, only two patients had experienced disease recurrence. Levels of potential biomarkers did not vary significantly between tumor biopsies taken before treatment and on the third day of treatment, nor did they predict for T downstaging or complete pathologic response. The investigators concluded that this regimen was an effective neoadjuvant treatment for stage II or III rectal cancer and led to more complete pathologic responses than have been seen in historical ­fluoropyrimidine-plus-radiation controls.283 In another phase I trial,284 preoperative therapy with capecitabine, oxaliplatin, and radiation (XELOX-RT) was given to 18 patients with clinically unresectable ­rectal cancer. Capecitabine was given at 500 to 825 mg/m2 twice daily for 7 days per week; oxaliplatin as a 2-hour intravenous infusion of 130 mg/m2 on days 1 and 29; and ­ pelvic radiation Monday through Friday for 5 weeks, for a total dose of 45 Gy in 25 daily 1.8Gy fractions. The maximum ­ tolerated dose, defined as the capecitabine dose causing dose-limiting toxicity in one third or more of patients ­ treated per dose ­ level, was found to be 825 mg/m2; the recommended dose for further study was 650 mg/m2 twice daily. Although this was primarily conceived as a dose-­finding study, the ­activity of this combination was promising—14 patients had ­ histologically confirmed R0 resections, and 5 had a pathologic complete response.284 Seeking to enhance the effectiveness of this therapy, ­another group285 tested a slightly different combination in which neoadjuvant capecitabine and oxaliplatin chemotherapy (XELOX) was given first, followed by concomitant capecitabine and pelvic radiation, then total mesorectal excision, and then 12 weeks of capecitabine. Among 77 patients with clinically inoperable disease, the radio­logic response rate was 88% after neoadjuvant XELOX and 97% after chemoradiation therapy. Of the 67 patients who went on to undergo total mesorectal excision, an R0 resection was accomplished in 66. Pathologic complete response was observed in 16 patients (24%), and an additional 32 patients (48%) had only microscopic tumor foci ­remaining in the surgical specimen.285 The CALGB286 reported their findings from a multiinstitutional phase I/II study of weekly oxaliplatin with continuously infused 5-FU (200 mg/m2) during preoperative radiation (50.4 Gy in 1.8-Gy fractions) for clinical T3 or T4 rectal cancer with no evidence of metastasis. The maximum tolerated dose of oxaliplatin was determined to be 60 mg/m2; this dose level produced grade 3 or 4 diarrhea in 12 patients, grade 3 neutropenia in 2 patients, and grade 3 thrombocytopenia in 1 patient. Fifty-six percent of patients entered at a dose of 60 mg/m2 completed all 6 weeks of oxaliplatin. Eight (25%) of 32 patients enrolled at the phase II dose experienced a pathologic complete response.286 A regimen of weekly oxaliplatin, continuousinfusion 5-FU, and radiation therapy is being evaluated by the NSABP. After reporting a single-center experience with neoadjuvant capecitabine and oxaliplatin plus radiation for rectal cancer,287 Rödel and colleagues undertook a multicenter

European trial to evaluate capecitabine (1650 mg/m2 on days 1 to 14 and 22 to 35) and oxaliplatin (50 mg/m2 on days 1, 8, 22, and 29) administered during 50.4 Gy of preoperative radiation therapy for 110 patients with rectal cancer.287 Surgery was scheduled for 4 to 6 weeks after the completion of the neoadjuvant regimen and was to be followed by four cycles of adjuvant XELOX (capecitabine 1000 mg/m2 twice daily on days 1 to 14 and oxaliplatin 130 mg/m2 on day 1). The overall goal of this trial was to establish a regimen of preoperative XELOX plus radiation followed by surgery plus adjuvant chemotherapy with XELOX that could be evaluated against standard 5-FU chemoradiation therapy plus adjuvant 5-FU in a phase III trial. After the preoperative therapy, 103 of 104 eligible patients underwent surgery, and 17 patients (16%) showed a pathologic complete response. R0 resections were achieved in 95% of patients, and sphincter preservation was accomplished in 77%. Grade 3 or 4 diarrhea occurred in 12% of patients, and postoperative complications occurred in 43% of patients. Although 96% of patients received fulldose preoperative therapy, only 60% of patients received all four cycles of adjuvant capecitabine and oxaliplatin; for them, sensory neuropathy (18%) and diarrhea (12%) were the main grade 3 or 4 toxic effects.288 In summary, the optimal doses and schedules of ­administration of oxaliplatin with radiation have yet to be determined. Although administration is clinically feasible, the incremental benefit relative to the added toxicity from adding oxaliplatin to radiation for locally advanced rectal cancer must be determined.

Radiation Therapy with Local Excision As part of the trend to maintain sphincter function whenever possible, radiation therapy has been used in combination with local excision of early-stage tumors that involve the distal rectum. The Radiation Therapy Oncology Group (RTOG) was among the first to report using such an approach.289 In that trial, 65 patients who had undergone local excision were observed or were given adjuvant treatment with 5-FU and one of two radiation dose levels according to tumor size and stage, tumor grade, and surgical margins.289 The 14 patients who were followed with observation had microscopic confirmation of a total excision, a well- or moderately well-differentiated T1 tumor that was smaller than 3 cm, and no evidence of vascular or lymphatic permeation on pathologic examination. At a median follow-up of 5 years, 11 of 65 patients had recurrent tumor. The incidence of local failure correlated with T category, with tumors recurring in 4% of T1 cases, 16% of T2 cases, and 23% of T3 cases. Other factors that affected ­local failure were the extent of circumferential involvement; ­local failure rate was only 6% for tumors with less than 20% circumferential involvement but was 18% when 20% to 40% of the circumference was involved. Selection of patients is extremely important in deciding whether to perform a local excision. Because lymph nodes are not resected in this procedure, local excision is contraindicated for patients with evident lymph node metastases.290 The depth of tumor penetration and the tumor grade are directly related to the risk of perirectal node involvement. The risk of lymph node involvement for a low-grade T2 tumor is about 12%, but the risk is 55% for a high-grade T2 tumor.291 Other studies have shown

24  The Colon and Rectum

that tumor ulceration and histologic evidence of lymphatic, vascular, or perineural invasion are also associated with an increased risk of nodal metastases.290,291 Tumors that are to be locally excised should be 3 cm or smaller and involve less than one half of the rectal circumference. Fewer complications are associated with a transanal approach than with a Kraske approach to local excision and postoperative radiation.292 Radiation therapy after local excision is ­ unnecessary for T1 tumors with favorable clinical and histologic ­characteristics because the risk of local failure is less than 10%.253 Postoperative radiation is recommended for T1 tumors with unfavorable characteristics or for T2 ­tumors because the risk of local failure is about 20%253,291-293; however, some studies have argued against the use of local excision for T2 tumors.150 A complete surgical ­resection with mesorectal excision is recommended for patients with T3 tumors; in one study, the local ­recurrence rate was ­ approximately 30% among patients with T3 ­tumors ­ treated with local excision and chemoradiation ­therapy.253 Techniques for delivering radiation in association with local excision are similar to those used for postoperative therapy.291 The entire pelvis is treated to 45 Gy, with the boost volume treated to a total dose of 54 to 65 Gy, depending on the resection margins and the presence of aggressive histologic features.291 Small bowel must be excluded from the radiation boost volume. The use of preoperative chemoradiation therapy ­followed by local excision for patients with T1 rectal tumors, especially tumors with favorable characteristics, is controversial because no information on tumor pathology is available, and these patients may be able to avoid ­radiation therapy. Preoperative chemoradiation therapy or radiation followed by local excision traditionally has been used for patients who decline or are medically unfit for a radical resection.294 The opportunity for pathologic downstaging and the theoretical advantages of preoperative chemoradiation therapy have led to this approach ­ being judiciously applied in a broader population of patients with rectal cancer.

Management of ­TreatmentRelated Side Effects Attempts to minimize the hematologic toxic effects of ­chemotherapy have included prolonging the infusion of 5-FU so that higher total doses can be administered with fewer hematologic side effects.195 Although the use of protracted infusions or alternative agents (e.g., oxaliplatin) can improve disease-free and overall survival rates, they carry the risk of gastrointestinal side effects, including profound diarrhea, and other toxic effects, such as mucositis and hand-foot syndrome.295 The possibility of superinfection, especially with Candida species, must be closely monitored; clinical signs such as fever and blood in the stool may suggest underlying sepsis and should initiate prompt evaluation.295 The occasional occurrence of mild nausea dictates evaluation of nutritional and fluid balance as needed. Gastrointestinal symptoms that occur during the first 2 weeks of radiation typically are attributable to chemotherapy. However, management of treatment-related diarrhea is the

591

same, regardless of whether it is caused by the radiation therapy or the chemotherapy.

Controlling Treatment-Related Diarrhea The Three-Step Bowel Management Program The University of Texas M. D. Anderson Cancer Center has instituted a three-step bowel management program for controlling diarrhea associated with chemoradiation therapy for locally advanced rectal cancer. Preliminary results of a prospective evaluation of this program have shown no difference between the presenting symptoms and the symptoms reported at the completion of chemoradiation therapy.296 The goal of the three-step program is to prevent ­rather than treat symptoms such as frequent stooling (i.e., more than three bowel movements per day), fluid and ­electrolyte imbalances, and skin irritation. All patients are given ­written instructions at the treatment-­simulation session and a ­prescription for an antidiarrheal agent (e.g., ­ loperamide) and an antiemetic agent. The strategy ­ underlying the ­program is proactive; at the first sign of ­ symptoms, the ­patient should proceed to the next step in the ­program.296 In step 1, initial symptoms, which usually occur in the third week of therapy, are treated with loperamide as needed. When that need exceeds two tablets per day, the patient proceeds to step 2, in which loperamide is administered according to a schedule (usually four times per day, 30 minutes before meals and before bedtime) to prevent frequent stooling. If that procedure fails to control symptoms, the patient proceeds to step 3, in which opioid analgesics are added to relieve the abdominal cramping and provide a constipating effect. The doses of analgesic are titrated to effect in accordance with established principles for pain management; the doses of opioids required in step 3 are usually modest. Because the treatment-related changes in the bowel do not resolve while therapy continues, sustained-release analgesics are preferred in step 3 to avoid the need for multiple drug doses during the day and the possible development of cyclical symptoms. The use of fentanyl patches should be avoided because of the difficulty in titrating the dose and because the need for analgesic does not remain stable over the course of radiation therapy. Immediate-release analgesics can be used for breakthrough abdominal cramping or bleeding. Skin reactions usually are restricted to the perianal region. At the first signs of dry desquamation, a barrier­emollient cream such as Skin Protectant (Lantiseptic, Marietta, GA) is applied and used consistently except during the radiation. Use of such a cream can ease the irritation and discomfort arising from dry desquamation of the ­perianal skin; it can also help prevent moist desquamation and the attendant risk of secondary infections from fecal bacteria, a particular concern when chemotherapy is given with radiation. Patients who develop candidiasis in the perianal region can benefit significantly from use of a mixture of creams or ointments containing zinc oxide (e.g., Desitin), nystatin, and lidocaine. Patients who develop localized moist desquamation in the perianal area during treatment of distal rectal tumors can maintain hygiene and relieve symptoms by using sitz baths or warm compresses with an aluminum acetate solution such as Domeboro (Bayer Corporation, Pittsburgh, PA).

592

PART 6  Gastrointestinal Tract

Other Approaches Other approaches to treating chemotherapy-induced diarrhea have included varying the dose-intensity of the chemotherapy agent. In one such study, pharmacokinetic evaluations showed broad variation in the metabolism of 5-FU among patients; acute toxic effects correlated with blood 5-FU levels that were higher than 3000 μg/L but not with the 5-FU dose given.297 Because metabolism of most cytotoxic drugs varies considerably among patients, at least one group has proposed basing chemotherapy dose calculations on body surface area.298 An alternative strategy for controlling secretory diarrhea is the use of loperamide with an enkephalinase ­ inhibitor such as acetorphan to treat the secretory diarrhea induced by irinotecan. Octreotide, a somatostatin analogue that prolongs intestinal transit time, promotes absorption of electrolytes in the intestine, decreases mesenteric blood flow, and decreases the secretion of fluids and electrolytes, is also effective against secretory diarrhea. Octreotide has been used to treat diarrhea associated with 5-FU or bone marrow transplants.295,299 The optimal dose and route of administration of octreotide are unknown.295 Assessments of the cost-effectiveness of octreotide or alternative strategies should reflect the cost of preventing hospitalization or infusional therapies for treatment-related complications.

Late Effects The success of sphincter-preserving strategies is defined largely in terms of late treatment-related complications. One complication, proctitis, may represent a continuum of symptoms. Despite the radiobiologic dissociation between acute and late toxic effects, acute proctitis has been linked in at least one study300 to three other late symptoms (i.e., tenesmus, frequency, and diarrhea) and the late morbidity score on a scale developed by the EORTC and the RTOG. Although rectal bleeding and discharge greatly affected the late morbidity score, tenesmus and bleeding had the greatest effect on daily living.300 Moreover, concordance was low between the patient’s assessment of symptoms and the late morbidity score.300 In a study of 386 patients, Miller and colleagues301 ­observed a relationship between the development of symptomatic acute enteritis and chronic bowel injury. Only 13 patients developed acute radiation-induced enteritis; three of these patients also experienced chronic bowel injury. A total of 18 patients developed chronic treatment-related symptoms, and reoperation was required in 17.301 Chronic proctitis had developed in 19% of the men in this study at 5 years. Factors related to an increased risk of complications were transanal excision, high radiation dose, and old age. Other quality-of-life issues relate to fertility. Given the risk of sexual dysfunctions such as retrograde ejaculation associated with surgical treatment of rectal cancer, men can consider preoperative sperm preservation. According to one cost-effectiveness analysis, the cost of sperm preservation was 2.2 to 5 times higher than the cost of one microsurgical epididymal sperm-aspiration procedure.302 Because the prostate is included in the radiation portal for rectal cancer, men undergoing radiation therapy for rectal cancer should consider sperm preservation. Assessments of quality of life should include consideration of how the intervention will affect survival and how

it will affect the health-related quality of that life. Factors to be considered in analyses of rectal cancer treatment include the risk of local recurrence, having to live with a permanent colostomy, the acute and late side effects of treatment, and the costs associated with the recommended therapies.303 Moreover, long-term follow-up is required to account for the completion of therapy, especially for ­locally advanced rectal cancer. Combined-modality therapy can involve 1 to 2 months of radiation, approximately 1 month between radiation therapy and surgery, and about 4 months of adjuvant chemotherapy after recovery from surgery. If a sphincter-sparing procedure is performed, a second operation is necessary to remove the temporary ­ileostomy. The total treatment time can be as long as 8 to 12 months. Quality-of-life ratings predictably improve after 3 years of follow-up, and bowel function is closely linked to quality of life.304-307

Management of Recurrent Rectal Cancer Recurrent rectal cancer can vary significantly in its clinical presentation, and it requires a broad range of therapeutic approaches. When a recurrence is identified early, the patient is usually asymptomatic, and resection is possible. Combined PET/CT scanning has been shown to be ­accurate for detecting rectal cancer recurrence, with 100% sensitivity and 96% specificity.308 Regrettably, recurrent rectal cancer is often diagnosed in the course of evaluating progressive symptoms, and at that time, the disease is often inoperable. Given the profound morbidity associated with recurrent disease, aggressive therapeutic attempts are often undertaken to achieve locoregional control.

Salvage Surgery The prevalence and influence of salvage therapy among ­patients with recurrent rectal cancer was assessed in a subset analysis of the large Intergroup 0114 trial.309 At a median 8.9 years of follow-up, 715 patients (42% of 1696 assessable patients) had experienced tumor recurrence, and 10% had died without evidence of disease. Of the 500 patients for whom follow-up information was ­ available and recurrence had appeared in a single organ or at a ­single site, 171 (34%) were treated by potentially curative ­resection. Among those 500 patients, the probability of 5-year ­ survival was 27% for the resected group and 6% for patients who did not undergo resection (P < .0001), and controlling for performance status did not affect this ­relationship.309 These results support the use of salvage surgery when ­possible. Pain at presentation for recurrent rectal cancer is an ­independent predictor of pain after treatment, and it warrants early involvement of cancer pain management ­specialists. Recurrent disease that manifests as pelvic or ­sciatic pain or necessitates total pelvic exenteration or bony resection usually is associated with higher rates of ­moderate to severe pain after treatment.310

Reradiation Management of recurrent disease in patients who have ­previously been treated with radiation therapy poses ­ significant challenges. Approaches have included

24  The Colon and Rectum

r­ eradiation with EBRT and intraoperative radiation therapy. The dose of EBRT given to these patients varies, especially when given in conjunction with intraoperative radiation. In one study, reradiation relieved bleeding in 100% of cases, pain in 65%, and mass effect in 24% of ­patients; moreover, for most responding patients, palliation lasted until death.311,312 Acute grade 3 toxicity appeared in 31% of cases and late grade 3 and 4 toxicity in 23% and 10% of cases, respectively, including small-bowel obstruction in 17%. The only factor that reduced late toxicity was hyperfractionation. Factors that influenced median survival were Karnofsky performance scale score, initial stage of disease, and the reradiation dose, with up to 30.6 Gy leading to a median survival time of 22 months and more than 30.6 Gy leading to a median survival time of 9.5 months.311 A Danish study313 showed that the risk of pelvic recurrence after the diagnosis of recurrent rectal cancer was no different for patients initially treated with an abdominoperineal resection (18%) than for those treated with a low anterior resection (24%).313 Of the 175 patients evaluated, 25 had a pelvic recurrence without evidence of distant ­metastasis313; among them, the patients given radiation therapy survived longer (15.9 months) than those given surgery or analgesics (2.4 months). Among 519 patients treated with radiation for ­locally recurrent rectal cancer at the Princess Margaret Hospital in Toronto,314 326 had undergone an abdominoperineal resection, 151 a low anterior resection, and 42 a local excision. None had been given radiation as part of the initial treatment.314 The mean time between the initial surgery and salvage radiation was 18 months (range, 3 to 138 months). The relapse was confined to the pelvis in 355 patients, but distant metastases were found with local recurrence in 164 (32%). The median survival time was 14 months, and the median time to ­ local disease progression was 5 months; the pelvic ­ disease progression-free rate was 7%, and the 5-year ­ survival rate was 5%.314 Multivariate analysis showed that overall survival was associated with performance status (measured on an ECOG scale), the absence of extrapelvic ­ metastases, a long interval between the initial surgery and ­salvage radiation, total radiation dose, and the ­absence of ­obstructive uropathy.314 Review of the published retrospective series suggests that there is no dose-response relationship between radiation and recurrence of rectal cancer.315 No significant differences were observed in initial response or durability of the response to therapy.315 However, a dose-response relationship might have been present when patients with primary untreated cancer and postoperative residual disease were evaluated separately.315 At the Peter MacCallum Cancer Institute, three dose regimens were used to treat incompletely resected, locally recurrent rectal cancer316,317: radical radiation dose (50 to 60 Gy), high-dose radiation (45 Gy, given in conventional fractionation), and palliative radiation dose (<45 Gy). No significant difference was observed in symptomatic relief between the radical- and high-dose groups (about 80%), but only 33% of those given palliative therapy achieved even partial relief of symptoms.316,317 The estimated ­median survival was 16 months for all patients; for patients in the radical-dose group, the median survival was 26 months, and for those in the high-dose group, it was

593

16 months. Survival was compromised when macroscopic residual disease was present; the median survival time in that case was 14 months, compared with 31 months if only microscopic residual tumor was present. The Italian Study Group for Therapies of Rectal ­Malignancies (STORM)318 reported results of a multicenter phase II trial to evaluate the response rate, resectability rate, local control, and treatment-related toxicity of preoperative hyperfractionated chemoradiation therapy for locally recurrent rectal cancer in patients who had previously received pelvic radiation. Among the 59 patients enrolled, the previous radiation doses ranged from 30 Gy to 55 Gy; 76% had had a prior anterior ­resection, and the other 24% had had abdominoperineal resection; and 75% of patients had received adjuvant or concurrent chemotherapy. The median interval between prior radiation therapy and the onset of reradiation was 27 months (range, 9 to 106 months). The preoperative ­radiation dose for recurrent disease was 30 Gy, given as twice-daily 1.2-Gy fractions separated by at least 6 hours. That dose was delivered to the gross tumor volume plus a 4-cm margin, followed by a 10.8-Gy boost (with the same fractionation) to the gross tumor volume plus a 2-cm margin. Concurrent chemotherapy consisted of continuous-infusion 5-FU (225 mg/m2/day, 7 days per week). Surgical resection, when feasible, was attempted 4 to 6 weeks after completion of the radiation therapy. ­Adjuvant chemotherapy with the thymidylate synthase ­inhibitor raltitrexed was subsequently given at 3 mg/m2 every 3 weeks for five cycles. Fifty-one (86%) of the 59 patients completed chemoradiation therapy without treatment interruption. Grade 3 acute toxicity was experienced by only 5% of patients, and no patient developed grade 4 acute toxicity. Late toxicities occurred in seven patients, including one smallbowel fistula requiring surgical diversion and two urinary complications requiring nephrostomy. After chemoradiation therapy, 5 patients (9%) had a complete response, 21 (36%) had a partial response, 31 (53%) had no change, and only 2 patients (3%) showed disease progression. Twenty of the 24 patients presenting with pelvic pain had symptomatic relief. Surgical resection was accomplished in 30 (51%) of 59 patients, and included 18 abdominoperineal resections, 4 anterior resections, 2 local excisions. and 6 other procedures; R0 resection was accomplished in 21 patients (36%). However, at a median follow-up time of 36 months (range, 9 to 69 months), 28 patients (48%) had developed local tumor recurrence or tumor progression, and 18 (31%) had developed distant metastases. The median survival time was 42 months, and the overall survival rate was 39% at 5 years. Local control and disease-free survival correlated with the interval between resection of the primary tumor (P = .028) and local recurrence (P = .003). Radical resection significantly influenced local control (P = .01), disease-free survival (P = .01), and overall ­survival (P = .05).318 The investigators concluded that hyperfractionated chemoradiation therapy was beneficial in terms of low treatment-related toxicity; improved quality of life, particularly pain control; and improved survival. These benefits extended to the one third of the study population who could undergo an R0 resection, the more than 80% of patients who experienced relief of pelvic pain,

594

PART 6  Gastrointestinal Tract

and the two thirds of patients who remained alive at 5 years.318

Intraoperative Radiation Therapy Intraoperative radiation has been used as a supplement to EBRT or as the only therapy when further EBRT is not possible. Outcome after such therapy was evaluated among 73 patients with recurrent rectal cancer, 86% of whom had locally recurrent disease without metastasis.319 EBRT had been given previously to 52% of these patients, and the recurrences were located in the presacral region (radial margin) in 55%. Surgical resection resulted in complete macroscopic resection of disease in 57% of patients; partial resection with gross residual disease was performed in 29%, and no resection was performed for 14% of the recurrences. The intraoperative radiation dose ranged from 10 Gy to 25 Gy; 30 patients also received preoperative or postoperative EBRT. Actuarial overall survival rates were 72% at 1 year, 45% at 2 years, and 31% at 3 years; comparable actuarial local control rates were 71%, 48%, and 31%, and actuarial disease-free survival rates were 58%, 27%, and 18%. Controlling symptoms of recurrent disease is a critical factor for quality of life because of the relatively long survival times that can be achieved. In the study described previously,319 actuarial survival rates were improved after complete or partial resection. Among the 21 instances of local failure, 8 appeared in the intraoperative radiation field, 8 in the EBRT field, and 3 in areas that received ­intraoperative radiation and EBRT. For patients in whom the recurrence appeared in the intraoperative radiation field, local control rates were 87% at 1 year, 72% at 2 years, and 65% at 3 years. At 3 years, actuarial local control rates in the intraoperative radiation field were better for those without gross residual disease (77%) than for those with gross residual disease (51%). Among patients with an isolated local recurrence who underwent partial or complete resection, the local control rates at 3 years were 61% for those given intraoperative radiation and EBRT and 0% for those given only intraoperative radiation. Of the four cases of long-term morbidity, one involved sciatic plexopathy that resolved with further follow-up (in a person given 12 Gy intraoperatively plus 45 Gy postoperatively), two involved the appearance of sacral necrosis at 6 months (in a person given 39 Gy preoperatively plus 25 Gy intraoperatively) and at 13 months (in a person given 40 Gy preoperatively plus 20 Gy intraoperatively), and one involved necrosis of the L5 vertebra after 51 months of follow-up (in a person given 28 Gy preoperatively plus 15 Gy ­intraoperatively).319 In contrast to this study, which showed that adding EBRT to intraoperative radiation therapy conferred an advantage,319 another study showed that the advantage resulted from the addition of intraoperative radiation to EBRT for locally advanced or recurrent rectal cancer.320 Local control depended on the amount of residual disease; for all patients, the 2-year actuarial local control rates were 77% (64% among those with recurrent disease) if no gross disease remained and 10% if gross residual disease was present after surgery. For all patients, the overall survival rate was improved if no gross disease remained after surgery (88% versus 46% for those with gross residual disease). The 2-year actuarial risk of major complications attributable to intraoperative radiation was 16%. Intraoperative radiation

therapy seems to be feasible in all clinical settings and adds less than 1 hour to the surgical procedure.321 Chemoradiation therapy and surgery can improve ­resectability, local control, and survival in patients with ­recurrent rectal cancer. In a study of 47 such patients given a preoperative dose of 45 to 47 Gy (34 patients) or 23.4 Gy (13 preirradiated patients) with concomitant 5-FU and mitomycin chemotherapy,260 more than one half showed an objective response, and 45% were able to undergo radical resection. Pain relief was ­accomplished in 86%, with higher radiation doses producing better pain control. After radical resection, the overall 5-year survival and local control rates were 22% and 32%, respectively; when intraoperative radiation was added, the corresponding 5-year rates were 79% and 41%. A modified Suzuki classification system, based on the extent to which the recurrence was fixated along the pelvic sidewall, was found to be useful for predicting level of pain, disease-free survival, and overall survival.260 A comparison of three intraoperative radiation ­approaches for recurrent colorectal cancer (i.e., intraoperative radiation therapy, high-dose-rate brachytherapy, and intraoperative placement of iodine 125 [125I] seeds) showed no difference in local control rates among treatment groups; at 5 years, the overall 5-year local control rate in that study was 26%.322 Although patients with tumors in the para-aortic region had better rates of local control than did patients with tumors in the pelvis (65% versus 19% at 5 years), this difference did not lead to an improvement in overall survival rate. The 5-year overall survival rate in the study322 was 4%; only the presence of residual disease (30% for those with microscopic disease and 7% for those with macroscopic disease at 3 years) and the use of postoperative EBRT (48% versus 12% at 3 years and 24% versus 0% at 5 years) influenced overall survival rates. Because of the difficulty in detecting residual disease, radioimmunoguided surgery has been advocated by some323 as a way of identifying occult tumor deposits. Combined-modality therapy that includes intraoperative radiation for recurrent colorectal cancer can ­produce problems in the healing of wounds in 46% to 65% of ­patients so treated; however, the introduction of unirradiated tissue by means of myocutaneous flap reconstruction has reduced the risk of major complications to 12%.324,325 Regardless of whether intraoperative radiation is used, resection of recurrent disease often requires extensive ­surgery for tumor clearance, a situation that is often complicated by previous radiation. In one study of 131 patients, resection of recurrent disease was possible in 103 (79%), and this resulted in a 5-year survival rate of 31%. Favorable prognostic factors included having a normal CEA level before surgery and the recurrent tumor being limited to the bowel wall.326 Aggressive surgical resection, including abdominosacral resection for fixed tumors, can produce long-term survival rates of about 31%.327 These aggressive surgical approaches are possible with reconstruction procedures that include placement of a myocutaneous graft. In summary, long-term survival is possible for patients with recurrent disease if the disease is detected early and if total surgical resection is possible. Radiation therapy is needed to clear potential nests of residual microscopic tumor. Prior therapies such as surgery and radiation can create hypoxic regions that require higher radiation doses for sterilization of residual tumor. Reradiation, especially

24  The Colon and Rectum

595

with hyperfractionated treatment schedules, can be well tolerated.

Palliative Radiation Therapy Achieving locoregional control is important in the management of rectal cancer regardless of stage. Locoregional failure, occurring in approximately 75% of all patients who die of rectal cancer, often produces significant symptoms such as pain, bleeding, and obstruction that are difficult to control.328-330 Although palliative radiation can relieve pain and bleeding in about 70% of cases, high-grade obstructive symptoms may require a diverting colostomy. Occasionally, urinary diversion may also be necessary. Palliative ­radiation relieves symptoms by causing partial regression of the ­tumor. The goal is to provide palliation of symptoms for the duration of the patient’s life. At the Princess Margaret Hospital, the radiation schedule most often used for palliation involves giving 52 Gy in 20 fractions over 4 weeks in a four-field technique. The 5-year survival rate after such treatment depends on the extent of the tumor; 48% of patients with mobile tumors, 27% with partially fixed tumors, and only 4% of patients with fixed tumors were alive at 5 years.331,332 Tumor ­extent also predicted response to radiation; 50% of those with mobile tumors, 30% of those with partially fixed tumors, and only 9% of those with fixed tumors (Fig. 24-6) achieved a complete clinical response to radiation. The rate of ­tumor regression was slow; only 60% achieved a complete ­response by 4 months, and 9 months was required for 90% of those who had a complete response. Of the complete responders, approximately one half of those with mobile or partially fixed tumors and more than 70% of those with fixed tumors developed progressive disease. However, salvage surgery to relieve symptoms was accomplished without significant complications in more than 90% of patients who developed progressive or recurrent disease.331,332 At M. D. Anderson Cancer Center, hypofractionated radiation therapy with concurrent chemotherapy produced symptomatic relief in 75 of 80 patients with locally advanced rectal cancer presenting with synchronous metastases.255,333 The hypofractionated regimen in that study consisted of 35 Gy, given in 2.5-Gy fractions; when the tumor extended anteriorly, an anterior-posterior treatment portal was used. When a belly board could be used, 30 Gy was given in 6 fractions over 3 weeks. These treatments produced endoscopic evidence of complete response in 36% of patients.255,333 Twenty-five patients underwent primary ­tumor resection. Although the 2-year survival rate was greater for the group of patients who underwent resection than for those who did not (46% versus 11%), the colostomy-free survival rate was higher for the group of patients who did not undergo surgery (79% versus 51%). Predictors of worse prognosis included pelvic pain at presentation, biologic equivalent dose of less than 35 Gy at 2 Gy per fraction, and poorly differentiated tumor histology. By comparison, the median survival time in another study of patients with locally advanced rectal cancer who had liver metastases at presentation was 17 months when treated with palliative radiation, with or without ­chemotherapy.334 Radiation is often given when tumors are inoperable because of extensive infiltration. Radiation may produce sufficient tumor regression to render such disease ­operable

A

B FIGURE 24-6.  A, Magnetic resonance image shows a ­presacral recurrence infiltrating into the bone. B, Axial section shows ­extensive infiltration of the fixed tumor into the sacrum and nerve roots.

and thereby offer the patient a chance to be free of disease. In one such study,335 29 patients with inoperative ­locally advanced tumors were given preoperative radiation to a total dose of 54 Gy and continuous-infusion 5-FU. ­Resection was undertaken in 23 of these cases, 18 of which were done with curative intent (13 abdominoperineal ­resections, 3 low anterior resections, and 2 local excisions). Pathologic findings included a complete response rate of 13% and a tumor downstaging rate of 90%. At a median ­follow-up of 28 months, 15 patients were free of disease. The ­authors of this study concluded that an aggressive surgical ­ approach could render patients free of disease and might improve overall survival. Another group found that multivisceral ­ resection for patients with locally advanced rectal cancer resulted in approximately a 50% survival rate at 5 years.336 Tumor infiltration that leads to acute colonic obstruction can be relieved through placement of a stent; in ­addition to relief of symptoms, this procedure can facilitate the administration of radiation therapy or cleansing of the colon before surgery.329,337,338 Stent placement relieves bowel obstruction in up to 96% of cases, with a risk of colonic

596

PART 6  Gastrointestinal Tract

perforation of less than 5%. The stents can remain patent for long periods; in one study, 91% of stents were patent 6 months after placement.337 When surgical resection of extensively infiltrating ­tumors is undertaken, radiation therapy must be administered because of the high risk of microscopic residual disease and subsequent recurrence—a risk that can exceed 80% after pelvic exenteration if radiation therapy is not administered.339 This risk is substantially higher than the risk of local recurrence for locally advanced but operable rectal cancer. Radiation-based approaches have included the use of intraoperative radiation therapy in conjunction with EBRT. In the Mayo Clinic experience, for example, factors related to improvements in overall survival were the use of preoperative EBRT with chemotherapy plus intraoperative radiation.340 The presence of gross residual disease after maximum tumor resection resulted in a 25% risk of progressive disease; gross total excision and the use of preoperative chemoradiation therapy decreased the risk of distant metastases.340 The addition of intraoperative radiation can improve local control rates for patients who present with inoperable rectal cancer. In one study, the addition of intraoperative radiation improved the local control rate after preoperative EBRT from 33% to 92%.324 Others have found that surgery preceded and followed by chemoradiation therapy (the sandwich technique) improved 4-year local failure rates over the use of preoperative chemoradiation therapy followed by surgery (0% versus 23%).341 The distant failure rate in this study decreased from 47% for those in the preoperative protocol to 28% for those in the sandwich protocol, which corresponded to overall survival rates of 41% and 72%.341 The use of continuous, chronobiologically shaped 5-FU infusions with radiation (45 or 55.8 Gy) has reportedly improved the response of initially unresectable tumors (pathologic complete response rates of 28%342 versus 11% for preoperative radiation with 50.4 Gy and bolus 5-FU with leucovorin343). Another sandwich protocol involving preoperative and postoperative 5-FU with leucovorin for patients with unresectable disease produced a local failure rate of 30% and a distant metastasis rate of 40%,344 with an actuarial disease-free survival rate at 4 years of 67% and an overall survival rate of 76%.343 In summary, substantial progress has been made in the treatment of rectal cancer, especially with the advent of combined-modality therapy. Each component treatment contributes to the local and systemic control of disease, and in combination, they have enabled the development of function-preserving surgical approaches.345

Management of Proximal Colon Cancer In contrast to its role in rectal cancer, the role of chemoradiation therapy in colon cancer is controversial. By the early 1990s, the benefit of adjuvant 5-FU chemotherapy for stage III colon cancer was well established, and no benefit had been observed from the addition of levamisole to 5-FU and leucovorin. As is true for other forms of cancer, the ability to achieve complete resection is important for local control and

s­ urvival. When primary colon cancer invades other viscera, such as the pancreas or duodenum, en bloc resection can improve overall survival without increasing perioperative morbidity.346 In the study reported by Koea and coworkers, 346 perioperative morbidity occurred in 28% of 5853 patients after colectomy for colon cancer. Preoperative factors that predicted a high risk of mortality included ascites, serum sodium level of more than 145 mg/dL, low albumin level, and high American Society of Anesthesiologists score for operative complications159 (classes III, IV, or V).347

Adjuvant Radiation Therapy The use of adjuvant postoperative radiation was evaluated for 78 cases of locally advanced (Dukes stage B2 to C) colon cancer.348 In that study, no patients were lost to follow-up, and all were followed for a minimum of 3 years. The overall local control rate was 88%. Multivariate analysis showed that the postoperative radiation dose was the only factor that influenced local control rates; the 5-year local control rate was 96% when the dose was 50 to 55 Gy, but that rate decreased to 76% when the postoperative dose was less than 45 Gy.348 These improvements in ­local control were achieved without significant radiation-­induced morbidity. Only 50% of stage C3 tumors, ­compared with 75% of stage C2 and B2 tumors, were locally controlled by 45 Gy.348,349 Disease stage was the only factor that influenced cause-specific survival.348 The efficacy of radiation in treating node-positive colon cancer was demonstrated in a pilot trial of 14 patients conducted by the ECOG. In that trial, 45 Gy was delivered to the tumor bed and periaortic lymph nodes and 30 Gy to the liver; the 10-year survival rate was 58%.350 A retrospective analysis of 152 patients with T4 N0 or T4 N1 colon cancer who were given postoperative radiation showed improved actuarial rates of local control (88%) and recurrence-free survival (58%).351 The local control rate decreased to 66% and the recurrence-free survival rate decreased to 27% for patients with T4 N2-3 disease. Disease could not be controlled in any patient with four or more positive nodes. Given the variability in the use of concurrent and adjuvant chemotherapy in this series, no benefit could be identified by the addition of 5-FU during radiation.351 In the early 1990s, a landmark phase III study provided the first evidence that systemic therapy with 5-FU and levamisole improved survival for patients with completely resected lymph node–positive colon cancer.352,353 After that study was published, adjuvant 5-FU and levamisole became the standard of care for such patients, and some oncologists began to give adjuvant chemotherapy to patients with node-negative colon cancer with other high-risk features, such as tumors that were nondiploid, had high proliferative indices, or were associated with adherence, invasion, or obstruction. At that time, the RTOG, Southwest Oncology Group, ECOG, CALGB, and National Cancer Institute of Canada jointly designed a randomized study (Intergroup 0130) to compare postoperative 5-FU and levamisole therapy with 5-FU, levamisole, and radiation therapy for high-risk colon cancer.354 Even though accrual to that study was slow and did not reach its objective, the trial was nevertheless one of the largest studies of adjuvant radiation in colon cancer. No differences were found ­between treatment groups in the rates of disease-free ­ survival

24  The Colon and Rectum

(51% for both groups) or overall survival (62% for the chemotherapy group versus 58% for the chemoradiation therapy group) at 5 years, but grade 3 or higher toxicity rates were higher for the chemoradiation therapy group (54% versus 42%), mostly because of the high rate of leukopenia in the chemoradiation therapy group (21% versus 10% in the chemotherapy group, P = .02). The investigators urged caution in interpreting the results of this study because of its lack of statistical power to detect meaningful differences in outcome, the lack of preoperative imaging, and the large number of ineligible patients.354

Adjuvant Chemotherapy Adjuvant chemotherapy has an established role in the treatment of colon cancer. In one study of 317 patients with high-risk stage II or stage III colon cancer,355 six cycles of 5-FU (425 mg/m2) and leucovorin (20 mg/m2) administered postoperatively for 5 consecutive days every 4 to 5 weeks significantly improved time to relapse and overall survival relative to surgery only.355 In an analysis356 of randomized trials of adjuvant chemotherapy for colorectal cancer, use of 5-FU and leucovorin was found to improve disease-free and overall survival rates over those produced by surgery only. Disease-free survival rates ranged from 71% to 77% after adjuvant chemotherapy and from 58% to 64% after surgery only; overall survival rates were 75% to 84% for adjuvant chemotherapy and 63% to 77% for surgery only.356 The International Union against Cancer compared the efficacy of adjuvant chemotherapy with 5-FU and leucovorin or 5-FU and levamisole for 680 patients with curatively resected stage III colon cancer.357 In that trial, biomodulation with leucovorin significantly improved disease-free and overall survival rates over those produced by 5-FU and levamisole.357 The NSABP C-04 trial found no improvement from the addition of levamisole to 5-FU alone or in combination with leucovorin in any aspect of outcome.358 The comparative efficacy of adjuvant chemotherapy was evaluated in four NSABP trials (C-01, C-02, C-03, and C-04). In all four studies, adjuvant chemotherapy improved the overall, disease-free, and recurrence-free survival rates improved among patients with Dukes stage B or C disease.359 In combined analyses, the mortality reduction was 30% for patients with Dukes stage B disease and 18% for those with Dukes stage C, regardless of adverse prognostic factors. Several other agents have been developed for the treatment of colon cancer. A randomized trial showed that the addition of oxaliplatin to 5-FU and leucovorin improved outcomes for patients receiving adjuvant therapy for colon cancer.191 Emerging evidence indicates that the addition of the antiangiogenic agent bevacizumab to 5-FU, leucovorin, and irinotecan can extend progression-free and overall survival for patients with metastatic colorectal cancer. The addition of bevacizumab to 5-FU and leucovorin provided a statistically significant and clinically relevant benefit in three independent studies of patients with previously untreated metastatic colorectal cancer.62 In another study,63 the ECOG showed that adding bevacizumab to the FOLFOX4 regimen (oxaliplatin at 85 mg/m2 given intravenously over 2 hours on day 1; leucovorin at 200 mg/m2 given intravenously over 2 hours on days 1 and 2; and a fluoropyrimidine [5-FU] at 400 mg/m2 given by intravenous bolus and followed by 600 mg/m2 given intravenously over 22 hours on days 1 and 2) was effective for patients

597

with metastatic colorectal cancer that had been previously treated with a fluoropyrimidine and irinotecan.63 Although bevacizumab was associated with some hypertension (6%) and bleeding (3%), it nevertheless extended the duration of survival in heavily pretreated patients.63

Summary Optimal treatment of colorectal cancer requires a ­multimodality approach. For maximum therapeutic benefit, the biologic basis of all therapeutic components in all treatment disciplines should be understood and exploited. ­ Adjuvant chemotherapy has significantly improved ­disease-free and overall survival rates, and adjuvant radiation has provided tremendous benefit by improving ­localcontrol rates and avoiding the morbidity associated with recurrent disease. Tumor regression in response to preoperative chemoradiation therapy has allowed increased use of sphincter-preserving surgical procedures with excellent functional outcomes, even for locally advanced tumors of the distal rectum. Sphincter preservation may be ­possible in even more cases as the ability to clear microscopic disease ­ improves, ­ especially disease in the mesorectum. Newer chemotherapeutic agents hold promise for further improvements because of their increased systemic activity and their synergistic effects with radiation.

REFERENCES    1. Roels S, Duthoy W, Haustermans K, et al. Definition and delineation of the clinical target volume for rectal cancer. Int J Radiat Oncol Biol Phys 2006;65:1129-1142.    2. Michalski JM, Purdy JA, Winter K, et al. Preliminary report of toxicity following 3D radiation therapy for prostate cancer on 3DOG/RTOG 9406. Int J Radiat Oncol Biol Phys 2000;46: 391-402.    3. Pollack A, Zagars GK, Starkschall G, et al. Conventional vs. conformal radiotherapy for prostate cancer: preliminary results of dosimetry and acute toxicity. Int J Radiat Oncol Biol Phys 1996;34:555-564.    4. Fajardo LF, Berthrong M, Anderson RE. Large intestine and anal canal. In Radiation Pathology. New York: Oxford University Press, 2001, pp 239-247.    5. Spanos WJ Jr, Wasserman T, Meoz R, et al. Palliation of ­advanced pelvic malignant disease with large fraction pelvic radiation and misonidazole: final report of RTOG phase I/II study. Int J Radiat Oncol Biol Phys 1987;13:1479-1482.    6. Brennan PC, Carr KE, Seed T, et al. Acute and protracted radiation effects on small intestinal morphological parameters. Int J Radiat Biol 1998;73:691-698.    7. Zheng H, Wang J, Hauer-Jensen M. Role of mast cells in early and delayed radiation injury in rat intestine. Radiat Res 2000;153:533-539.    8. Dublineau I, Ksas B, Griffiths NM. Functional changes in the rat distal colon after whole-body radiation: dose-response and temporal relationships. Radiat Res 2000;154:187-195.    9. Hendricks T, Wobbes T, deMan BM, et al. Moderate doses of intraoperative radiation severely suppress early strength of anastomoses in the rat colon. Radiat Res 1998;150:431-435.   10. Skwarchuk MW, Travis EL. Volume effects and epithelial regeneration in irradiated mouse colorectum. Radiat Res 1998;149:1-10.   11. Seifert WF, Biert J, Wobbes T, et al. Late effects of intraoperative radiation therapy in anastomotic rat colon. Int J Radiat Oncol Biol Phys 1998;42:623-629.   12. American Cancer Society. Cancer Facts & Figures, 2001. Atlanta: American Cancer Society, 2001.   13. American Cancer Society. Cancer Facts & Figures, 2008. Atlanta: American Cancer Society, 2008.

598

PART 6  Gastrointestinal Tract

  14. Winawer SJ. Natural history of colorectal cancer. Am J Med 1999;106:3S-6S.   15. Potter JD. Colorectal cancer: molecules and populations. J Natl Cancer Inst 1999;91:916-932.   16. Nelson RL, Dollear T, Freels S, et al. The relation of age, race, and gender to the subsite location of colorectal carcinoma. ­Cancer 1997;80:193-197.   17. Slattery ML, Friedman GD, Potter JD, et al. A description of age, sex, and site distributions of colon carcinoma in three geographic areas. Cancer 1996;78:1666-1670.   18. Dignam JJ, Colangelo L, Tian W, et al. Outcomes among ­AfricanAmericans and Caucasians in colon cancer adjuvant therapy trials: findings from the National Surgical Adjuvant Breast and Bowel Project. J Natl Cancer Inst 1999;91:1933-1940.   19. Mettlin CJ, Menck HR, Winchester DP, et al. A comparison of breast, colorectal, lung, and prostate cancers reported to the National Cancer Data Base and the Surveillance, Epidemiology, and End Results Program. Cancer 1997;79:2052-2061.   20. Ikeda Y, Mori M, Yoshizumi T, et al. Cancer and adenomatous polyp distribution in the colorectum. Am J Gastroenterol 1999;94:191-193.   21. Weitzel JN. Genetic cancer risk assessment—putting it all ­together. Cancer 1999;86:2483-2492.   22. Neuhausen SL. Ethnic differences in cancer risk resulting from genetic variation. Cancer 1999;86:2575-2582.   23. Lynch HT, Watson P, Shaw TG, et al. Clinical impact of ­molecular genetic diagnosis, genetic counseling, and management of hereditary cancer. I. Studies of cancer in families. ­Cancer 1999;86:2449-2456.   24. Dunlop MG. Guidance on gastrointestinal surveillance for hereditary non-polyposis colorectal cancer, familial adenomatous polyposis, juvenile polyposis, and Peutz-Jeghers syndrome. Gut 2002;51;(Suppl 5):V21-V27.   25. Heiskanen I, Jarvinen HJ. Fate of the rectal stump after colectomy and ileorectal anastomosis for familial adenomatous ­polyposis. Int J Colorectal Dis 1997;12:9-13.   26. Church J, Burke C, McGannon E, et al. Risk of rectal cancer in patients after colectomy and ileorectal anastomosis for familial adenomatous polyposis: a function of the available surgical ­options. Dis Colon Rectum 2003;46:1175-1181.   27. Van Duijvendijk P, Vasen HF, Bertario L, et al. Cumulative risk of developing polyps or malignancy at the ileal pouch-anal anastomosis in patients with familial adenomatous polyposis. J ­Gastrointest Surg 1999;3:325-330.   28. Nugent KP, Phillips RK. Rectal cancer risk in older patients with familial adenomatous polyposis and an ileorectal anastomosis: a cause for concern. Br J Surg 1992;79:1204-1206.   29. Annie Yu HJ, Lin KM, Ota DM, et al. Hereditary nonpolyposis colorectal cancer: preventive management. Cancer Treat Rev 2003;29:461-470.   30. Rodriguez-Bigas MA, Boland CR, Hamilton SR, et al. A ­National Cancer Institute Workshop on Hereditary Nonpolyposis ­Colorectal Cancer Syndrome: meeting highlights and Bethesda guidelines. J Natl Cancer Inst 1997;89:1758-1762.   31. Vasen HF, Mecklin JP, Khan PM, Lynch HT. The International Collaborative Group on Hereditary Non-polyposis Colorectal Cancer (ICGHNPCC). Dis Colon Rectum 1991;34:424-425.   32. Lynch HT. Is there a role for prophylactic subtotal colectomy among hereditary nonpolyposis colorectal cancer germline ­mutation carriers? Dis Colon Rectum 1996;39:109-110.   33. Rodriguez-Bigas MA, Vasen HF, Pekka-Mecklin J. Rectal cancer risk in hereditary nonpolyposis colorectal cancer after abdominal colectomy. International Collaborative Group on HNPCC. Ann Surg 1997;225:202-207.   34. Hawk E, Lubet R, Limburg P. Chemoprevention in hereditary colorectal cancer syndromes. Cancer 1999;86:2551-2563.   35. Dannenberg AJ, Zakim D. Chemoprevention of colorectal cancer through inhibition of cyclooxygenase-2. Semin Oncol 1999;26:499-504.   36. Felder JB, Korelitz BI, Rajapakse R, et al. Effects of nonsteroidal anti-inflammatory drugs on inflammatory bowel disease: a casecontrol study. Am J Gastroenterol 2000;95. 1949-.

  37. Grodstein F, Newcomb PA, Stampfer MJ. Postmenopausal hormone therapy and the risk of colorectal cancer: a review and meta-analysis. Am J Med 1999;106:574-582.   38. Paganini-Hill A. Estrogen replacement therapy and colorectal cancer risk in elderly women. Dis Colon Rectum 1999;42: 1300-1305.   39. Issa JP, Ottaviano YL, Celano P, et al. Methylation of the oestrogen receptor CpG island linking ageing and neoplasia in human colon. Nat Genet 1994;7:536-540.   40. Ghadirian P, Lacroix A, Maisonneuve P, et al. Nutritional factors and colon carcinoma—a case-control study involving French Canadians in Montreal, Quebec, Canada. Cancer 1997;80: 858-864.   41. Macrae F. Wheat bran fiber and development of adenomatous polyps: evidence from randomized, controlled clinical trials. Am J Med 1999;106:38S-42S.   42. Earnest DL, Einspahr JG, Alberts DS. Protective role of wheat bran fiber: data from marker trials. Am J Med 1999;106: 32S-37S.   43. Reddy BS. Role of dietary fiber in colon cancer: an overview. Am J Med 1999;106:16S-19S.   44. Holt PR, Atillasoy EO, Gilman J, et al. Modulation of abnormal colonic epithelial cell proliferation and differentiation by low-fat dairy foods—a randomized controlled trial. JAMA 1998;280:1074-1079.   45. Cousins RJ. Nutritional regulation of gene expression. Am J Med 1999;106:20S-23S.   46. Shike M. Diet and lifestyle in the prevention of colorectal cancer: an overview. Am J Med 1999;106:11S-15S.   47. Brenner DJ, Curtis RE, Hall EJ, et al. Second malignancies in prostate carcinoma patients after radiotherapy compared with surgery. Cancer 2000;88:398-406.   48. Strober W, Ludviksson BR, Fuss IJ. The pathogenesis of mucosal inflammation in murine models of inflammatory bowel disease and Crohn’s disease. Ann Intern Med 1998;128:848-856.   49. Eaden JA, Abrams KR, Mayberry JF. The risk of colorectal cancer in ulcerative colitis: a meta-analysis. Gut 2001;48:526-535.   50. Collins PD, Mpofu C, Watson AJ, et al. Strategies for detecting colon cancer and/or dysplasia in patients with inflammatory bowel disease. Cochrane Database Syst Rev 2006;(2): CD000279.   51. Choi PM, Nugent FW, Schoetz DJ Jr., et al. Colonoscopic surveillance reduces mortality from colorectal cancer in ulcerative colitis. Gastroenterology 1993;105:418-424.   52. Iba T, Yagi Y, Kidokoro A, et al. The intratumoral levels of thymidylate synthetase and folate in gastric and colon cancer. Gan To Kagaku Ryoho 1998;25:1165-1171.   53. Villafranca E, Okruzhnov Y, Dominguez MA, et al. Polymorphisms of the repeated sequences in the enhancer region of the thymidylate synthase gene promoter may predict downstaging after preoperative chemoradiation of rectal cancer. J Clin Oncol 2001;19:1779-1786.   54. Lynch HT, Watson P, Shaw TG, et al. Clinical impact of molecular genetic diagnosis, genetic counseling, and management of hereditary cancer. II. Hereditary nonpolyposis colorectal carcinoma as a model. Cancer 1999;86:2457-2463.   55. Watwe V, Javle M, Lawrence D, et al. Cyclooxygenase-2 [COX-2] levels before and after chemotherapy: a study in rectal cancer. Am J Clin Oncol 2005;28:560-564.   56. Ang KK, Berkey BA, Tu X, et al. Impact of epidermal growth factor receptor expression on survival and pattern of relapse in patients with advanced head and neck carcinoma. Cancer Res 2002;62:7350-7356.   57. Chua DT, Nicholls JM, Sham JS, 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.   58. Li S, Kim JS, Kim JM, et al. Epidermal growth factor receptor as a prognostic factor in locally advanced rectal cancer patients treated with preoperative chemoradiation. Int J Radiat Oncol Biol Phys 2006;65:705-712.

24  The Colon and Rectum   59. Spindler KL, Nielsen JN, Lindebjerg J, et al. Prediction of ­response to chemoradiation in rectal cancer by a gene polymorphism in the epidermal growth factor receptor promoter region. Int J Radiat Oncol Biol Phys 2006;66:500-504.   60. Zhang W, Gordon M, Schultheis AM, et al. FCGR2A and FCGR3A polymorphisms associated with clinical outcome of epidermal growth factor receptor-expressing metastatic colorectal cancer patients treated with single agent cetuximab. J Clin Oncol 2007;25:3712-3718.   61. Schneider S, Park DJ, Yang D, et al. Gene expression in tumoradjacent normal tissue was associated with recurrence in patients with rectal cancer treated with adjuvant chemoradiation. Pharmacogenet Genomics 2006;16:555-563.   62. Kabbinavar FF, Hambleton J, Mass RD, et al. Combined analysis of efficacy: the addition of bevacizumab to fluorouracil/leucovorin improves survival for patients with metastatic colorectal cancer. J Clin Oncol 2005;23:3706-3712.   63. Giantonio BJ, Catalano PJ, Meropol NJ, et al. Bevacizumab in combination with oxaliplatin, fluorouracil, and leucovorin [FOLFOX 4] for previously treated metastatic colorectal cancer: results from the Eastern Cooperative Oncology Group study E3200. J Clin Oncol 2007;25:1539-1544.   64. Jubb AM, Hurwitz HI, Bai W, et al. Impact of vascular endothelial growth factor-A expression, thrombospondin-2 expression, and microvessel density on the treatment effect of bevacizumab in metastatic colorectal cancer. J Clin Oncol 2006;24:217-227.   65. Troisi RJ, Freedman AN, Devesa SS. Incidence of colorectal carcinoma in the U.S.—an update of trends by gender, race, age, subsite, and stage, 1975-1994. Cancer 1999;85:1670-1676.   66. Ries LAG, Wingo PA, Miller DS, et al. The annual report to the nation on the status of cancer, 1973-1997, with a special section on colorectal cancer. Cancer 2000;88:2398-2424.   67. Lynch PM. New issues in genetic counseling of hereditary colon cancer. Clin Cancer Res 2007;13;(22 Pt 2):s6857-6861s.   68. Lerman C, Hughes C, Trock BJ, et al. Genetic testing in families with hereditary nonpolyposis colon cancer. JAMA 1999;17:1618-1622.   69. Rawl SM, Menon U, Champion RL, et al. Colorectal ­ cancer screening beliefs—focus groups with first-degree relatives. ­Cancer Practice 2000;8:32-37.   70. Lynch PM. Clinical challenges in management of familial adenomatous polyposis and hereditary nonpolyposis colorectal cancer. Cancer 1999;86:2533-2550.   71. Kinney AY, Choi YA, DeVellis B, et al. Attitudes toward ­genetic testing in patients with colorectal cancer. Cancer Practice 2000;8:178-186.   72. Worthen HG. Inherited cancer and the primary care physicianbarriers and strategies. Cancer 1999;86:2583-2588.   73. Lynch HT, Smyrk TC. Hereditary colorectal cancer. Semin ­Oncol 1999;26:478-484.   74. Lindor NM, Greene MH. The concise handbook of family cancer syndromes. J Natl Cancer Inst 1998;90:1039-1071. Mayo Familial Cancer Program.   75. Severin MJ. Genetic susceptibility for specific cancers: medical liability of the clinician. Cancer 1999;86:2564-2569.   76. Syngal S, Weeks JC, Schrag D, et al. Benefits of colonoscopic surveillance and prophylactic colectomy in patients with hereditary nonpolyposis colorectal cancer mutations. Ann Intern Med 1998;129:787-796.   77. Tweedle EM, Rooney PS, Watson AJM. Screening for rectal cancer—will it improve cure rates? Clin Oncol (R Coll Radiol) 2007;19:639-648.   78. Whitehouse PA, Tilney HS, Armitage JN, et al. Transanal endoscopic microsurgery: risk factors for local recurrence of benign rectal adenomas. Colorectal Dis 2006;8:795-799.   79. Conio M, Repici A, Demarquay JF, et al. EMR of large sessile colorectal polyps. Gastrointest Endosc 2004;60:234-241.   80. Hurlstone DP, Sanders DS, Cross SS, et al. Colonoscopic ­resection of lateral spreading tumours: a prospective analysis of endoscopic mucosal resection. Gut 2004;53:1334-1339.   81. Tamura S, Nakajo K, Yokoyama Y, et al. Evaluation of endoscopic mucosal resection for laterally spreading rectal tumors. Endoscopy 2004;36:306-312.

599

  82. Atkin WS, Saunders BP. Surveillance guidelines after removal of colorectal adenomatous polyps. Gut 2002;51;(Suppl 5):V6-V9.   83. Stryker SJ, Wolff BG, Culp CE, et al. Natural history of untreated colonic polyps. Gastroenterology 1987;93:1009-1013.   84. Stern MA, Fendrick AM, McDonnell WM, et al. A randomized, controlled trial to assess a novel colorectal cancer screening strategy: the conversion strategy—a comparison of sequential ­sigmoidoscopy and colonoscopy with immediate ­ conversion from sigmoidoscopy to colonoscopy in patients with an ­abnormal screening sigmoidoscopy. Am J Gastroenterol 2000;95: 2074-2079.   85. Karadi C, Beaulieu CF, Jeffrey RB, et al. Display modes for CT colonography. I. Synthesis and insertion of polyps into patient CT data. Radiology 1999;212:195-201.   86. Schwartz LH. Advances in cross-sectional imaging of colorectal cancer. Semin Oncol 1999;26:569-576.   87. Sharma VK, Corder FA, Raufman JP, et al. Survey of internal medicine residents’ use of the fecal occult blood test and their understanding of colorectal cancer screening and surveillance. Am J Gastroenterol 2000;95:2068-2073.   88. Church TR, Ederer F, Mandel JS. Fecal occult blood screening in the Minnesota study: sensitivity of the screening test. J Natl Cancer Inst 1997;89:1440-1448.   89. Blumberg D, Opelka FG, Hicks TC, et al. Significance of a normal surveillance colonoscopy in patients with a history of ­adenomatous polyps. Dis Colon Rectum 2000;43:1084-1092.   90. Burton EC, Troxclair DA, Newman WP. Autopsy diagnoses of malignant neoplasms—how often are clinical diagnoses incorrect? JAMA 1998;280:1245-1248.   91. Bini EJ, Micale PL, Weinshel EH. Evaluation of the gastrointestinal tract in premenopausal women with iron deficiency anemia. Am J Med 1998;105:281-286.   92. Markowitz AJ, Winawer SJ. Screening and surveillance for colorectal cancer. Semin Oncol 1999;26:485-498.   93. Bond JH. Screening guidelines for colorectal cancer. Am J Med 1999;106:7S-10S.   94. Helm JF, Russo MW, Biddle AK, et al. Effectiveness and economic impact of screening for colorectal cancer by mass fecal occult blood test. Am J Gastroenterol 2000;95:3250-3258.   95. Vijan S, Hwang EW, Hoffer TP, et al. Which colon cancer screening test? A comparison of costs, effectiveness. compliance. Am J Med 2001;111:593-601.   96. Ko CW, Sonnenberg A. Comparing risks and benefits of colorectal cancer screening in elderly patients. Gastroenterology 2005;129:1163-1170.   97. American Cancer Society. Cancer Facts & Figures, 2006. Atlanta: American Cancer Society, 2006.   98. Rex DK, Johnson DA, Lieberman DA, et al. Colorectal cancer prevention 2000: screening recommendations of the ­American College of Gastroenterology. Am J Gastroenterol 2000;95: 868-877.   99. Nelson RL, Persky V, Turyk M. Time trends in distal colorectal cancer subsite location related to age and how it affects choice of screening modality. J Surg Oncol 1998;69:235-238. 100. Desch CE, Benson AB, Smith TJ, et al. Recommended colorectal cancer surveillance guidelines by the American Society of Clinical Oncology. J Clin Oncol 1999;17:1312-1321. 101. Desch CE, Benson AB, Somerfield MR, et al. Colorectal cancer surveillance: 2005 update of an American Society of Clinical Oncology Practice Guideline. J Clin Oncol 2005;23:8512-8519. 102. Rex DK, Kahi CJ, Levin B. Guidelines for colonoscopy surveillance after cancer resection: a consensus update by the American Cancer Society and US Multi-Society Task Force on Colorectal Cancer. CA Cancer J Clin 2006;56:160-167. 103. Macdonald JS. Carcinoembryonic antigen screening: pros and cons. Semin Oncol 1999;26:556-560. 104. Locker GY, Hamilton S, Harris J, et al. ASCO 2006 update of recommendations for the use of tumor markers in gastrointestinal cancer. J Clin Oncol 2006;24:5313-5327. 105. Goldberg RM, Fleming TR, Tangen CM, et al. Surgery for recurrent colon cancer: strategies for identifying resectable recurrence and success rates after resection. Ann Intern Med 1998;129: 27-35.

600

PART 6  Gastrointestinal Tract

106. Secco GB, Fardelli R, Rovida S, et al. Is intensive follow-up really able to improve prognosis of patients with local recurrence after curative surgery for rectal cancer? Ann Surg Oncol 2000;7:32-37. 107. Skenderis BS, Rodriguez-Bigas M, Weber TK, et al. Utility of routine postoperative laboratory studies in patients undergoing potentially curative resection for adenocarcinoma of the colon and rectum. Cancer Invest 1999;17:102-109. 108. Figueredo A, Rumble RB, Maroun J, et al. Follow-up of patients with curatively resected colorectal cancer: a practice guideline. BMC Cancer 2003;3:26. 109. Renehan AG, Egger M, Saaunders MP, et al. Impact on survival of intensive follow-up after curative resection for colorectal cancer: systematic review and meta-analysis of randomized trials. BMJ 2002;324:813. 110. Jeffery GM, Hikey BE, Hider P. Follow-up strategies for patients treated with nonmetastatic colorectal cancer. Cochrane Database Syst Rev 2002;(1). CD002200. 111. Chau I, Allen MJ, Cunningham D, et al. The value of routine serum carcinoembryonic antigen measurement and computed tomography in the surveillance of patients after adjuvant chemotherapy for colorectal cancer. J Clin Oncol 2004;22:1420-1429. 112. Low RN, Semelka RC, Worawattanakul S, et al. Extrahepatic abdominal imaging in patients with malignancy: comparison of MR imaging and helical CT with subsequent surgical correlation. Radiology 1999;210:625-632. 113. Burton S, Brown G, Daniels I, et al. MRI identified prognostic features of tumors in distal sigmoid, rectosigmoid, and upper rectum: treatment with radiotherapy and chemotherapy. Int J Radiat Oncol Biol Phys 2006;65:445-451. 114. Flamen P, Stroobants S, Van Cutsem E, et al. Additional value of whole-body positron emission tomography with fluorine-182-fluoro-2-deoxy-d-glucose in recurrent colorectal cancer. J Clin Oncol 1999;17:894-901. 115. Akhurst T, Larson SM. Positron emission tomography and imaging of colorectal cancer. Semin Oncol 1999;26:577-583. 116. Flanagan FL, Dehdashti F, Ogunbiyi O, et al. Utility of FDGPET for investigating unexplained plasma CEA elevation in patients with colorectal cancer. Ann Surg 1998;227:319-323. 117. Valk PE, Abella-Columna E, Haseman MK, et al. Whole-body PET imaging with (18F) fluorodeoxyglucose in management of recurrent colorectal cancer. Arch Surg 1999;134:503-511. 118. Vesselle HJ, Miraldi FD. FDG PET of the retroperitoneum: normal anatomy, variants, pathologic conditions, and strategies to avoid diagnostic pitfalls. Radiographics 1998;18:805-823. 119. Berger KL, Nicholson SA, Dehdashti F, et al. FDG PET evaluation of mucinous neoplasms: correlation of FDG uptake with histopathologic features. AJR Am J Roentgenol 2000;174:10051008. 120. Kawamoto S, Soyer PA, Fishman EK, et al. Nonneoplastic liver disease: evaluation with CT and MR imaging. Radiographics 1998;18:827-848. 121. Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer ­Staging Manual, 6th ed. New York: Springer, 2002, pp 113-123. 122. Horton KM, Abrams RA, Fishman EK. Spiral CT of colon cancer: imaging features and role in management. Radiographics 2000;20:419-430. 123. Vogl TJ, Pegios W, Mack MG, et al. Accuracy of staging rectal ­tumors with contrast-enhanced transrectal MR imaging. AJR Am J Roentgenol 1997;168:1427-1434. 124. Zerhouni EA, Rutter C, Hamilton SR, et al. CT and MR imaging in the staging of colorectal carcinoma: report of the Radiology Diagnostic Oncology Group II. Radiology 1996;200:443-451. 125. Kruskal JB, Kane RA, Sentovich SM, et al. Pitfalls and sources of error in staging rectal cancer with endorectal US. Radiographics 1997;17:609-626. 126. Frudinger A, Bartram CI, Kamm MA. Transvaginal versus anal endosonography for detecting damage to the anal sphincter. AJR Am J Roentgenol 1997;168:1435-1438. 127. Rubens DJ, Strang JG, Bogineni-Misra S, et al. Transperineal sonography of the rectum: anatomy and pathology revealed by sonography compared with CT and MR imaging. AJR Am J Roentgenol 1998;170:637-642.

128. Lindmark GE, Kraaz WG, Elvin PAB, et al. Rectal cancer: evaluation of staging with endosonography. Radiology 1997;204: 533-538. 129. Lohnert MSS, Doniec JM, Henne-Bruns D. Effectiveness of endoluminal sonography in the identification of occult local rectal cancer recurrences. Dis Colon Rectum 2000;43:483-491. 130. Gunderson LL, Sargent DJ, Tepper JE, et al. Impact of T and N stage and treatment on survival and relapse in adjuvant rectal cancer: a pooled analysis. J Clin Oncol 2004;22:1785-1796. 131. Liersch T, Langer C, Ghadimi BM, et al. Lymph node status and TS gene expression are prognostic markers in stage I/II ­ rectal cancer after neoadjuvant fluorouracil based chemotherapy. J Clin Oncol 2006;24:4062-4068. 132. Enker WE. Designing the optimal surgery for rectal carcinoma. Cancer 1996;78:1847-1850. 133. Sugihara K, Moriya Y, Akasu T, et al. Pelvic autonomic nerve preservation for patients with rectal carcinoma. Oncologic and functional outcome. Cancer 1996;78:1871-1880. 134. Hida J, Yasutomi M, Maruyama T, et al. The extent of lymph node dissection for colon carcinoma: the potential impact on laparoscopic surgery. Cancer 1997;80:188-192. 135. Tepper JE, O’Connell MJ, Niedzwiecki D, et al. Impact on number of nodes retrieved on outcome in patients with rectal cancer. J Clin Oncol 2001;19:157-163. 136. Kockerling F, Reymond MA, Altendorf-Hofmann A, et al. Influence of surgery on metachronous distant metastases and survival in rectal cancer. J Clin Oncol 1988;16:324-329. 137. Nagtegaal ID, Quirke P. What is the role for the circumferential margin in the modern treatment of rectal cancer? J Clin Oncol 2008;26:303-312. 138. Nagtegaal ID, van de Velde CJH, Marijnen CAM, et al. Low rectal cancer: a call for a change of approach in abdominoperineal resection. J Clin Oncol 2005;23:9257-9264. 139. Enker WE. Sphincter-preserving operations for rectal cancer. Oncology 1996;10:1673-1689. 140. Paty PB, Cohen AM. Technical considerations for coloanal anastomosis and J-pouch. Semin Radiat Oncol 1998;8:48-53. 141. Berger A, Tiret E, Cunningham C, et al. Rectal excision and colonic pouch—anal anastomosis for rectal cancer-oncologic ­results at five years. Dis Colon Rectum 1999;42:1265-1271. 142. Renner K, Rosen HR, Novi G, et al. Quality of life after surgery for rectal cancer—do we still need a permanent colostomy? Dis Colon Rectum 1999;42:1160-1167. 143. Rullier E, Zerbib F, Laurent C, et al. Intersphincteric resection with excision of internal anal sphincter for conservative treatment of very low rectal cancer. Dis Colon Rectum 1999;42: 1168-1175. 144. Gavioli M, Losi L, Luppi G, et al. Preoperative therapy for lower rectal cancer and modifications in distance from anal sphincter. Int J Radiat Oncol Biol Phys 2007;69:370-375. 145. Meyerhardt JA, Tepper JE, Niedzwiecki D, et al. Impact of body mass index on outcomes and treatment-related toxicity in patients with stage II and III rectal cancer: findings from Intergroup Trial 0114. J Clin Oncol 2004;22:648-657. 146. Mellgren A, Sirivongs P, Rothenberger DA, et al. Is local excision adequate therapy for early rectal cancer? Dis Colon Rectum 2000;43:2064-2074. 147. Weber TK, Petrelli NJ. Local excision for rectal cancer: an uncertain future. Oncology 1998;12:933-943. 148. Chakravarti A, Compton CC, Shellito PC, et al. Long-term followup of patients with rectal cancer managed by local excision with and without adjuvant radiation. Ann Surg 1999;230:49-54. 149. Steele GD, Herndon JE, Bleday R, et al. Sphincter-sparing treatment for distal rectal adenocarcinoma. Ann Surg Oncol 1999;6:433-441. 150. You YN, Baxter NN, Stewart A, et al. Is the increasing rate of local excision for stage I rectal cancer in the United States justified? A nationwide cohort study from the National Cancer Database. Ann Surg 2007;245:726-733. 151. Petrelli NJ, Weber TK. Sphincter preservation for distal rectal cancer: paradise lost? Ann Surg Oncol 1999;6:413-415. 152. Baxter NN, Garcia-Aguilar J. Organ preservation for rectal ­cancer. J Clin Oncol 2007;25:1014-1020.

24  The Colon and Rectum 153. Balbay MD, Slaton JW, Trane N, et al. Rationale for bladdersparing surgery in patients with locally advanced colorectal carcinoma. Cancer 1999;86:2212-2216. 154. Hermanek P. Impact of surgeon’s technique on outcome after treatment of rectal carcinoma. Dis Colon Rectum 1999;42: 559-562. 155. Dorrance HR, Docherty GM, O’Dwyer PJ. Effect of surgeon specialty interest on patient outcome after potentially curative colorectal cancer surgery. Dis Colon Rectum 2000;43:492-498. 156. Ptok H, Marusch F, Kuhn R, et al. Influence of hospital volume on the frequency of abdominoperineal resections and long-term oncological outcomes in low rectal cancer. Eur J Surg Oncol 2007;33:854-861. 157. Janjan NA, Khoo VS, Rich TA, et al. Locally advanced rectal cancer: surgical complications after infusional chemotherapy and radiation therapy. Radiology 1998;206:131-136. 158. Pucciarelli S, Toppan P, Friso ML, et al. Preoperative combined radiotherapy and chemotherapy for rectal cancer does not ­affect early postoperative morbidity and mortality in low anterior resection. Dis Colon Rectum 1999;42:1276-1284. 159. American Society of Anesthesiologists. New classification of physical status. Anesthesiology 1963;24:111. 160. Gorksi T, Rosen L, Lawrence S, et al. Usefulness of a state­legislated comparative database to evaluate quality in colorectal surgery. Dis Colon Rectum 1999;42:1381-1387. 161. Kinsella TJ. An approach to the radiosensitization of human ­tumors. Cancer J Sci Am 1996;2:184-193. 162. Milas L. Chemoradiation interactions: potential of newer chemotherapeutic agents. In American Society of Clinical Oncology Educational Handbook. Alexandria, VA: American Society of Clinical Oncology, 2000, pp 207-213. 163. Koutcher JA, Alfieri AA, Thaler H, et al. Radiation enhancement by biochemical modulation and 5-fluorouracil. Int J Radiat ­Oncol Biol Phys 1997;39:1145-1152. 164. Krook JE, Moertel CG, Gunderson LL, et al. Effective surgical adjuvant therapy for high-risk rectal carcinoma. N Engl J Med 1991;324:709-715. 165. Douglas HO, Moertel CG, Mayer RJ, et al. Survival after postoperative combination treatment of rectal cancer. N Engl J Med 1986;315:1294-1295. 166. Gastrointestinal Tumor Study Group. Prolongation of the disease-free interval in surgically treated rectal carcinoma. N EngI J Med 1985;312:1465-1472. 167. Thomas PR, Lindblad AS. Adjuvant postoperative radiotherapy and chemotherapy in rectal carcinoma: a review of the Gastrointestinal Tumor Study Group experience. Radiother Oncol 1988;13:245-252. 168. Fisher B, Wolmark N, Rockette H, et al. Postoperative adjuvant chemotherapy or radiation therapy for rectal cancer: results from NSABP R-01. J Natl Cancer Inst 1988;80:21-29. 169. Freedman GM, Coia LR. Adjuvant and neoadjuvant treatment of rectal cancer. Semin Oncol 1995;22:611-624. 170. Gunderson LL, Martenson JA. Postoperative adjuvant radiation with or without chemotherapy for rectal carcinoma. Semin Radiat Oncol 1993;3:55-63. 171. Rich T, Ajani JA, Morrison WH, et al. Chemoradiation ­therapy for anal cancer. Radiation plus continuous infusion of 5-­fluorouracil with or without cisplatin. Radiother Oncol 1993;27:209-215. 172. National Cancer Institute Clinical Announcement. Adjuvant Therapy of Rectal Cancer. Bethesda: National Cancer Institute, 1991. 173. Smalley SR, Benedetti JK, Williamson SK, et al. Phase III trial of fluorouracil based chemotherapy regimens plus radiotherapy in postoperative adjuvant rectal cancer: GI INT 0144. J Clin Oncol 2006;24:3542-3547. 174. Merchant NB, Guillem JG, Paty PB, et al. T3N0 rectal cancer: results following sharp mesorectal excision and no adjuvant therapy. J Gastrointest Surg 1999;3:642-647. 175. Picon AI, Moore HG, Sternberg SS, et al. Prognostic significance of depth of gross or microscopic perirectal fat invasion in T3 N0 M0 rectal cancers following sharp mesorectal excision and no adjuvant therapy. Int J Colorectal Dis 2003;18:487-492.

601

176. Willett CG, Badizadegan K, Ancukiewicz M, et al. Prognostic factors in stage T3N0 rectal cancer: do all patients require postoperative pelvic radiation and chemotherapy? Dis Colon Rectum 1999;42:167-173. 177. Guillem JG, Diaz-Gonzalez JA, Minsky BD, et al. cT3N0 rectal cancer: potential overtreatment with preoperative chemoradiotherapy is warranted. J Clin Oncol 2008;26:368-373. 178. Baxter NN, Morris AM, Rothenberger DA, et al. Impact of ­preoperative radiation for rectal cancer on subsequent lymph node evaluation: a population-based analysis. Int J Radiat Oncol Biol Phys 2005;61:426-431. 179. Wichmann MW, Muller C, Meyer G, et al. Effect of preoperative radiochemotherapy on lymph node retrieval after resection of rectal cancer. Arch Surg 2002;137:206-210. 180. Schaffzin DM, Wong WD. Endorectal ultrasound in the ­preoperative evaluation of rectal cancer. Clin Colorectal Cancer 2004;4:124-132. 181. Sauer R, Becker H, Hohenberger W, et al. Preoperative versus postoperative chemoradiotherapy for rectal cancer. N Engl J Med 2004;351:1731-1740. 182. Bosset JF, Calasi G, Mineur L, et al. Enhanced tumoricidal effect of chemotherapy with preoperative radiotherapy for rectal cancer: preliminary results—EORTC 22921. J Clin Oncol 2005;23:5620-5627. 183. Bosset JF, Collette L, Calais G, et al. Chemotherapy with preoperative radiotherapy in rectal cancer. N Engl J Med 2006;355:1114-1123. 184. Gérard JP, Conroy T, Bonnetain F, et al. Preoperative radiotherapy with or without concurrent fluorouracil and leucovorin in T3-4 rectal cancers: results of FFCD 9203. J Clin Oncol 2006;24:4620-4625. 185. Jessup JM, Stewart AK, Menck HR. The National Cancer Data Base Report on patterns of care for adenocarcinoma of the rectum, 1985-1995. Cancer 1998;83:2408-2418. 186. Minsky BD, Coia L, Haller DG, et al. Radiation therapy for rectosigmoid and rectal cancer: results of the 1992-1994 Patterns of Care Process Survey. J Clin Oncol 1998;16:2542-2547. 187. Coia LR, Gunderson LL, Haller D, et al. Outcomes of ­patients receiving radiation for carcinoma of the rectum. Cancer 1999;86:1952-1958. 188. Kim DY, Jung KH, Kim TH, et al. Comparison of 5-fluorouracil/ leucovorin and capecitabine in preoperative chemoradiotherapy for locally advanced rectal cancer. Int J Radiat Oncol Biol Phys 2007;67:378-384. 189. Das P, Lin EH, Bhatia S, et al. Preoperative chemoradiotherapy with capecitabine versus protracted infusion 5-fluorouracil for rectal cancer: a matched-pair analysis. Int J Radiat Oncol Biol Phys 2006;66:1378-1383. 190. Tepper JE, O’Connell M, Niedzwiecki D, et al. Adjuvant ­therapy in rectal cancer: analysis of stage, sex, and local control—final report of Intergroup 0114. J Clin Oncol 2002;20:1744-1750. 191. Andre T, Boni C, Mounedji-Boudiaf L, et al. Oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment for colon cancer. N Engl J Med 2004;350:2343-2351. 192. de Gramont A, Boni C, Navarro M, et al. Oxaliplatin/5FU/LV in adjuvant colon cancer: updated efficacy results of the MOSAIC trial, including survival, with a median follow-up of six years [abstract]. J Clin Oncol 2007;25:4007. 193. Sobrero AF, Aschele C, Bertino JR. Fluorouracil in colorectal cancer—a tale of two drugs: implications for biochemical modulation. J Clin Oncol 1997;15:368-381. 194. Tepper JE, O’Connell MJ, Petroni GR, et al. Adjuvant postoperative fluorouracil modulated chemotherapy combined with pelvic radiation therapy for rectal cancer: initial results of intergroup 0114. J Clin Oncol 1997;15:2030-2039. 195. O’Connell MJ, Martenson JA, Wieand HS, et al. Improving ­adjuvant therapy for rectal cancer by combining protracted­infusion fluorouracil with radiation therapy after curative surgery. N Engl J Med 1994;331:502-507. 196. Meta-Analysis Group in Cancer. Efficacy of intravenous ­continuous infusion of fluorouracil compared with bolus administration in advanced colorectal cancer. J Clin Oncol 1998;16:301-308.

602

PART 6  Gastrointestinal Tract

197. Kovach JS, Beart RW Jr. Cellular pharmacology of fluorinated pyrimidines in vivo in man. Invest New Drugs 1989;7:13-25. 198. Cassidy J, Douillard JY, Twelves C, et al. Pharmacoeconomic analysis of adjuvant oral capecitabine vs. intravenous 5FU/ LV in Dukes C colon cancer: the X-ACT trial. Br J Cancer 2006;94:1122-1129. 199. Desai SP, El-Rayes BF, Ben-Josef E, et al. A phase II study of preoperative capecitabine and radiation therapy in patients with rectal cancer. Am J Clin Oncol 2007;30:340-345. 200. Craven I, Crellin A, Cooper R, et al. Preoperative radiotherapy combined with 5 days per week capecitabine chemotherapy in locally advanced rectal cancer. Br J Cancer 2007;97:1333-1337. 201. Ben-Josef E. Capecitabine and radiotherapy as neoadjuvant treatment for rectal cancer. Am J Clin Oncol 2007;30:649-655. 202. Cividalli A, Ceciarelli F, Livdi E, et al. Radiosensitization by oxaliplatin in a mouse adenocarcinoma: influence of treatment schedule. Int J Radiat Oncol Biol Phys 2002;52:1092-1098. 203. Magne N, Fischel JL, Formento P, et al. Oxaliplatin-5-fluorouracil and ionizing radiation: importance of the sequence and ­influence of p53 status. Oncology 2003;64:280-287. 204. Bleiberg H, de Gramont A. Oxaliplatin plus 5-fluorouracil: clinical experience in patients with advanced colorectal cancer. Semin Oncol 1988;25:32-39. 205. Andre T, Bensmaine MA, Louvet C, et al. Multicenter phase II study of bimonthly high-dose leucovorin fluorouracil infusion, and oxaliplatin for metastatic colorectal cancer resistant to the same leucovorin and fluorouracil regimen. J Clin Oncol 1999;17:3560-3568. 206. Giacchetti S, Perpoint B, Zidani R, et al. Phase III multicenter randomized trial of oxaliplatin added to chronomodulated fluorouracil-leucovorin as first-line treatment of metastatic colorectal cancer. J Clin Oncol 2000;18:136-147. 207. de Gramont A, Figer A, Seymour M, et al. Leucovorin and fluorouracil with or without oxaliplatin as first-line treatment in advanced colorectal cancer. J Clin Oncol 2000;18:2938-2947. 208. Goldberg RM, Sargent DJ, Morton RF, et al. A randomized controlled trial of fluorouracil plus leucovorin, irinotecan and oxaliplatin combinations in patients with previously untreated metastatic colorectal cancer. J Clin Oncol 2004;22:23-30. 209. Wolmark N, Wienand S, Kuebler JP, et al. A phase III trial comparing FULV + oxaliplatin in stage II or III carcinoma of the colon: results of NSABP Protocol C-07 [abstract]. J Clin Oncol 2005;23:246s. 210. Williams K, Telfer B, Stratford I, et al. ZD1839 (‘Iressa’), a specific oral epidermal growth factor receptor-tyrosine kinase inhibitor, potentiates radiotherapy in a human colorectal cancers xenograft model. Br J Cancer 2002;86:1157-1161. 211. Raben D, Helfrich B, Chan DC, et al. ZD 1839, a selective epidermal growth factor tyrosine kinase inhibitor, alone and in combination with radiation and chemotherapy as a new therapeutic strategy in non-small cell lung cancer. Semin Oncol 2002;29: 37-46. 212. Czito BG, Willett CG, Bendell JC, et al. Increased toxicity with gefitinib, capecitabine and radiation therapy in pancreatic and rectal cancer: phase I trial results. J Clin Oncol 2006;24: 656-662. 213. Tominaga T, Sakabe T, Koyama Y, et al. Prognostic factors for patients with colon or rectal carcinoma treated with resection only—five year follow-up report. Cancer 1996;78:403-408. 214. Phillips RKS, Hittinger R, Blesovsky L, et al. Local recurrence following “curative” surgery for large bowel cancer. 1. The overall picture. Br J Surg 1984;71:12-16. 215. Minsky BD, Conti JA, Huang Y, et al. Relationship of acute gastrointestinal toxicity and the volume of irradiated small bowel in patients receiving combined modality therapy for rectal cancer. J Clin Oncol 1995;13:1409-1416. 216. Koelbl O, Richter S, Flentje M. Influence of patient positioning on dose-volume histogram and normal tissue complication probability for small bowel and bladder in patients receiving pelvic radiation: a prospective study using a 3D planning system and a radiobiological model. Int J Radiat Oncol Biol Phys 1999;45:1193-1198.

217. Das IJ, Lanciano RM, Movsas B, et al. Efficacy of a belly board device with CT-simulation in reducing small bowel volume within pelvic radiation fields. Int J Radiat Oncol Biol Phys 1997;39:67-76. 218. Chun M, Timmerman RD, Mayer R, et al. Radiation therapy of external iliac lymph nodes with lateral pelvic portals: identification of patients at risk for inadequate regional coverage. Radiology 1995;194:147-150. 219. Taylor N, Crane C, Skibber J, et al. Elective groin radiation is not indicated among patients with adenocarcinoma of the rectum extending to the anal canal. Int J Radiat Oncol Biol Phys 2001;51:741-747. 220. Tocchi A, Lepre L, Costa G, et al. Rectal cancer and inguinal ­metastases—prognostic role and therapeutic indications. Dis Colon Rectum 1999;42:1464-1466. 221. Lacerna MD, Sharpe MB, Robertson JM. The effect of radiation on an ambulatory chemotherapy infusion pump. Cancer 1999;86:2150-2153. 222. Gérard JP, Romestaing P, Ardiet JM, et al. Endocavitary radiation therapy. Semin Radiat Oncol 1998;8:13-23. 223. Maingon P, Guerif S, Darsouni R, et al. Conservative management of rectal adenocarcinoma by radiotherapy. Int J Radiat ­Oncol Biol Phys 1998;40:1077-1085. 224. Salembier C, Battermann JJ. Interstitial brachytherapy in the conservative treatment of anal and rectal cancers. Endocuriether Hyperthermia Oncol 1996;12:183-189. 225. Armour EP, White JR, Armin AR, et al. Pulsed low dose rate brachytherapy in a rat model: dependence of late rectal injury on radiation pulse size. Int J Radiat Oncol Biol Phys 1997;38: 825-834. 226. Minsky BD. Multidisciplinary management of resectable rectal cancer. Oncology 1996;10:1701-1714. 227. Minsky BD, Cohen AM, Enker WE, et al. Efficacy of postoperative 5-FU, high-dose leucovorin, and sequential radiation therapy for clinically resectable rectal cancer. Cancer Invest 1995;13:1-7. 228. Minsky BD. Adjuvant therapy of rectal cancer. Semin Oncol 1999;26:540-544. 229. Minsky BD. Preoperative radiation therapy followed by low ­anterior resection with coloanal anastomosis. Semin Radiat Oncol 1998;8:30-35. 230. Bosset JF, Pelissier EP, Manion G, et al. Plea for a preoperative adjuvant approach in the management of rectal cancer. Int J ­Radiat Oncol Biol Phys 1994;29:205-208. 231. Rich TA. Adjuvant radiotherapy for rectal cancer: more on pre- or postoperative therapy. Int J Radiat Oncol Biol Phys 1995;32:547-548. 232. Coucke PA, Sartorelli B, Cuttat JF, et al. The rationale to switch from postoperative hyperfractionated accelerated radiotherapy to preoperative hyperfractionated accelerated radiotherapy in rectal cancer. Int J Radiat Oncol Biol Phys 1995;32:181-188. 233. Burton S, Brown G, Daniels I, et al. MRI identified prognostic features of tumors in distal sigmoid, rectosigmoid, and upper rectum: treatment with radiotherapy and chemotherapy. Int J Radiat Oncol Biol Phys 2006;65:445-451. 234. Dinter DJ, Hofheinz RD, Hartel M, et al. Preoperative staging of rectal tumors: comparison of endorectal ultrasound, hydroCT, and high-resolution endorectal MRI. Onkologie 2008;31: 230-235. 235. Muthusamy VR, Chang KJ. Optimal methods for staging rectal cancer. Clin Cancer Res 2007;13(Pt 2):6877s-6884s. 236. Improved survival with preoperative radiotherapy in resectable rectal cancer. Swedish Rectal Cancer Trial. N Engl J Med 1997;336:980-987. 237. Marijnen CAM, Kapiteijn E, van de Velde CJH, et al. Acute side effects and complications after short term preoperative radiotherapy combined with total mesorectal excision in primary rectal cancer: report of a multicenter randomized trial. J Clin Oncol 2002;20:817-825. 238. van den Brink M, van den Hout WB, Stiggelbout AM, et al. Cost-utility analysis of preoperative radiotherapy in patients with rectal cancer undergoing total mesorectal excision: a study of the Dutch Colorectal Cancer Group. J Clin Oncol 2004;22: 244-253.

24  The Colon and Rectum 239. Camma C, Giunta M, Fiorica F, et al. Preoperative radiotherapy for resectable rectal cancer: a meta-analysis. JAMA 2000;284:1008-1015. 240. Vaughn DJ, Haller DG. Adjuvant therapy for colorectal cancer: past accomplishments, future directions. Cancer Invest 1997;15:435-447. 241. Cedermark B, Johansson H, Rutqvist LE, et al. The Stockholm I trial of preoperative short term radiotherapy in operable rectal carcinoma: a prospective randomized trial. Cancer 1995;75:2269-2275. 242. Stockholm Colorectal Cancer Study Group. Randomized study on preoperative radiotherapy in rectal carcinoma. Ann Surg ­Oncol 1996;3:423-430. 243. Frykholm GJ, Glimelius B, Pahlman L. Preoperative or postoperative radiation in adenocarcinoma of the rectum: final ­treatment results of a randomized trial and an evaluation of late secondary effects. Dis Colon Rectum 1993;36:564-572. 244. Pahlman L, Glimelius B. Pre- or postoperative radiotherapy in rectal and rectosigmoid carcinoma: a report from a randomized multicenter trial. Ann Surg 1990;211:187-195. 245. Folkesson J, Birgisson H, Pahlman L, et al. Swedish rectal cancer trial: long lasting benefits from radiotherapy on survival and ­local recurrence rate. J Clin Oncol 2005;23:5644-5650. 246. Frykholm GJ, Isacsson U, Nygard K, et al. Preoperative radiotherapy in rectal carcinoma—aspects of acute adverse effects and radiation technique. Int J Radiat Oncol Biol Phys 1996;35:1039-1048. 247. Myerson RJ, Genovesi D, Lockett MA, et al. Five fractions of preoperative radiotherapy for selected cases of rectal carcinoma: long-term tumor control and tolerance to treatment. Int J Radiat Oncol Biol Phys 1999;43:537-543. 248. Tho LM, Glegg M, Paterson J, et al. Acute small bowel toxicity and preoperative chemoradiotherapy for rectal cancer: investigating dose-volume relationships and role for inverse planning. Int J Radiat Oncol Biol Phys 2006;66:505-513. 249. Robertson JM, Lockman D, Yan D, et al. The dose-volume ­relationship of small bowel radiation and acute grade 3 ­diarrhea during chemoradiotherapy for rectal cancer. Int J Radiat Oncol Biol Phys 2008;70:413-418. 250. Minsky BD, Cohen AM, Enker WE, et al. Sphincter preservation with preoperative radiation therapy and coloanal anastomosis. Int J Radiat Oncol Biol Phys 1995;31:553-559. 251. Marks G, Mohiuddin M, Masoni L, et al. High-dose preoperative radiation therapy as the key to extending sphincter-preservation surgery for cancer of the distal rectum. Surg Oncol Clin North Am 1992;1:71-86. 252. Paszat LF, Brundage MD, Groome PA, et al. A population-based study of rectal cancer: permanent colostomy as an outcome. Int J Radiat Oncol Biol Phys 1999;45:1185-1191. 253. Ng AK, Recht A, Busse PM. Sphincter preservation therapy for distal rectal carcinoma: a review. Cancer 1997;79:671-683. 254. Hyams DM, Mamounas EP, Petrelli N, et al. A clinical trial to evaluate the worth of preoperative multimodal therapy in patients with operable carcinoma of the rectum: a progress report of the National Surgical Adjuvant Breast and Bowel Project protocol R-03. Dis Colon Rectum 1997;40:131-139. 255. Janjan NA, Breslin T, Lenzi R, et al. Avoidance of colostomy placement in advanced colorectal cancer with twice weekly ­hypofractionated radiation plus continuous infusion 5-fluorouracil. J Pain Symptom Manage 2000;20:266-272. 256. Janjan NA, Crane CH, Feig BW, et al. Prospective trial of preoperative concomitant boost radiotherapy with continuous infusion 5-fluorouracil for locally advanced rectal cancer. Int J Radiat Oncol Biol Phys 2000;47:713-718. 257. Bosset JF, Magnin V, Maingon P, et al. Preoperative radiochemotherapy in rectal cancer: long-term results of a phase II trial. Int J Radiat Oncol Biol Phys 2000;46:323-327. 258. Pucciarelli S, Friso ML, Toppan P, et al. Preoperative combined radiotherapy and chemotherapy for middle and lower rectal cancer: preliminary results. Ann Surg Oncol 2000;7:38-44. 259. Valentini V, Coco C, Cellini N, et al. Preoperative chemoradiation with cisplatin and 5-fluorouracil for extraperitoneal T3 rectal cancer: acute toxicity, tumor response, sphincter preservation. Int J Radiat Oncol Biol Phys 1999;45:1175-1184.

603

260. Valentini V, Morganti AG, De Franco A, et al. Chemoradiation with or without intraoperative radiation therapy in patients with locally recurrent rectal carcinoma—prognostic factors and long term outcome. Cancer 1999;86:2612-2624. 261. Movsas B, Hanlon AL, Lanciano R, et al. Phase I dose escalating trial of hyperfractionated pre-operative chemoradiation for locally advanced rectal cancer. Int J Radiat Oncol Biol Phys 1998;42:43-50. 262. Wagman R, Minsky BD, Cohen AM, et al. Sphincter preservation in rectal cancer with preoperative radiation therapy and coloanal anastomosis: long term follow-up. Int J Radiat Oncol Biol Physics 1998;42:51-57. 263. Mohiuddin M, Regine WF, Marks GJ, et al. High-dose preoperative radiation and the challenge of sphincter-preservation surgery for cancer of the distal 2 cm of the rectum. Int J Radiat Oncol Biol Phys 1998;40:569-574. 264. Crane CH, Skibber JM, Feig BW, et al. Response to preoperative chemoradiation increases the use of sphincter-preserving surgery in patients with locally advanced low rectal carcinoma. Cancer 2003;97:517-524. 265. Berger C, de Muret A, Garaud P, et al. Preoperative radiotherapy [RT] for rectal cancer: predictive factors of tumor downstaging and residual tumor cell density (RTCD): prognostic implications. Int J Radiat Oncol Biol Phys 1997;37:619-627. 266. Francois Y, Nemoz CJ, Baulieux J, et al. Influence of the ­interval between preoperative radiation therapy and surgery on downstaging and on the rate of sphincter-sparing surgery for ­rectal cancer: the Lyon R90-01 randomized trial. J Clin Oncol 1999;17:2396-2402. 267. Vauthey JN, Marsh RW, Zlotecki RA, et al. Recent advances in the treatment and outcome of locally advanced rectal cancer. Ann Surg 1999;229:745-752. 268. Rich TA, Skibber JM, Ajani JA, et al. Preoperative infusional chemoradiation therapy for stage T3 rectal cancer. Int J Radiat Oncol Biol Phys 1995;32:1025-1029. 269. Weinstein GD, Rich TA, Shumate CR, et al. Preoperative infusional chemoradiation and surgery with or without an electron beam intraoperative boost for advanced primary rectal cancer. Int J Radiat Oncol Biol Phys 1995;32:197-204. 270. Janjan NA, Khoo VS, Abbruzzese J, et al. Tumor downstaging and sphincter preservation with preoperative chemoradiation in locally advanced rectal cancer: the M.D. Anderson Cancer Center experience. Int J Radiat Oncol Biol Phys 1999;44: 1027-1038. 271. Kaminsky-Forrett MC, Conroy T, Luporsi E, et al. Prognostic implications of downstaging following preoperative radiation therapy for operable T3-T4 rectal cancer. Int J Radiat Oncol Biol Phys 1998;42:935-941. 272. Janjan NA, Abbruzzese J, Pazdur R, et al. Prognostic implications of response to preoperative infusional chemoradiation in locally advanced rectal cancer. Radiother Oncol 1999;51: 153-160. 273. Willett CG, Warland G, Hagan MP, et al. Tumor proliferation in rectal cancer following preoperative radiation. J Clin Oncol 1995;13:1417-1424. 274. DeMario MD, Ratain MJ. Oral chemotherapy: rationale and ­future directions. J Clin Oncol 1998;16:2557-2567. 275. Capirci C, Rubello D, Chierichetti F, et al. Long-term prognostic value of 18 F-FDG PET in patients with locally advanced rectal cancer previously treated with neoadjuvant radiochemotherapy. AJR Am J Roentgenol 2006;187:W202-W208. 276. Suwinski R, Taylor JMG, Withers HR. Rapid growth of microscopic rectal cancer as a determinant of response to preoperative radiation therapy. Int J Radiat Oncol Biol Phys 1998;42: 943-951. 277. Hoff PM, Lassere Y, Pazdur R, et al. Preoperative UFT and calcium folinate and radiotherapy in rectal cancer. Oncology 1999;13(Suppl 3):129-131. 278. Sharma S, Saltz LB. Oral chemotherapeutic agents for colorectal cancer. Oncologist 2000;5:99-107. 279. Ahmed FY, Johnston SJ, Cassidy J, et al. Eniluracil treatment completely inactivates dihydropyrimidine dehydrogenase in colorectal tumors. J Clin Oncol 1999;17:2439-2445.

604

PART 6  Gastrointestinal Tract

280. Saif MW, Hashmi S, Zelterman D, et al. Capecitabine vs. continuous infusion 5FU in neoadjuvant treatment of rectal cancer: a retrospective review. Int J Colorectal Dis 2008;32:139-145. 281. Matsusaka S, Yamasaki H, Fukushima M, et al. Upregulation of enzymes metabolizing 5-fluorouracil in colorectal cancer. Chemotherapy 2007;53:36-41. 282. Fakih MG, Rajput A, Yang GY, et al. A phase I study of weekly intravenous oxaliplatin in combination with oral daily capecitabine and radiation therapy in the neoadjuvant treatment of rectal adenocarcinoma. Int J Radiat Oncol Biol Phys 2006;65: 1462-1470. 283. Fakih MG, Bullarddunn K, Yang GY, et al. Phase II study of weekly intravenous oxaliplatin combined with oral daily capecitabine and radiotherapy with biologic correlates in neoadjuvant treatment of rectal adenocarcinoma. Int J Radiat Oncol Biol Phys 2008;72:650-657. 284. Glynne-Jones R, Sebag-Montefiore D, Maughan TS, et al. A phase I dose escalation study of continuous oral capecitabine in combination with oxaliplatin and pelvic radiation [XELOXRT] in patients with locally advanced rectal cancer. Ann Oncol 2006;17:50-56. 285. Chau I, Brown G, Cunningham D, et al. Neoadjuvant capecitabine and oxaliplatin followed by synchronous chemoradiation and total mesorectal excision in magnetic resonance imagingdefined poor-risk rectal cancer. J Clin Oncol 2006;24:668-674. 286. Ryan DP, Niedzwiecki D, Hollis D, et al. Phase I/II study of preoperative oxaliplatin, fluorouracil, and external-beam radiation therapy in patients with locally advanced rectal cancer: Cancer and Leukemia Group B 89901. J Clin Oncol 2006;24: 2557-2562. 287. Rödel C, Grabenbauer GG, Papadopoulos T, et al. Phase I/II trial of capecitabine, oxaliplatin, and radiation for rectal cancer. J Clin Oncol 2003;21:3098-3104. 288. Rödel C, Liersch T, Hermann RM, et al. Multicenter phase II trial of chemoradiation with oxaliplatin for rectal cancer. J Clin Oncol 2007;25:110-117. 289. Russell AH, Harris J, Rosenberg PJ, et al. Anal sphincter conservation for patients with adenocarcinoma of the distal rectum: long-term results of Radiation Therapy Oncology Group Protocol 89-02. Int J Radiat Oncol Biol Phys 2000;46:313-322. 290. Johnson DE, Hoffman JP. Surgical considerations for local excision. Semin Radiat Oncol 1998;8:39-47. 291. Willett CG. Local excision followed by postoperative radiation therapy. Semin Radiat Oncol 1998;8:24-29. 292. Bouvet M, Milas M, Giacco GG, et al. Predictors of recurrence after local excision and postoperative chemoradiation therapy of adenocarcinoma of the rectum. Ann Surg Oncol 1999;6:26-32. 293. Jessup JM, Loda M, Bleday R. Clinical and molecular prognostic factors in sphincter-preserving surgery for rectal cancer. Semin Radiat Oncol 1998;8:54-69. 294. Ahmad NR, Nagle DA. Preoperative radiation therapy followed by local excision. Semin Radiat Oncol 1998;8:36-38. 295. Wadler S, Benson AB, Engelking C, et al. Recommended guidelines for the treatment of chemotherapy-induced diarrhea. J Clin Oncol 1998;16:3169-3178. 296. Callister M, Janjan N, Crane C, et al. Effective management of symptoms during preoperative chemoradiation for locally advanced rectal cancer [abstract 60]. Int J Radiat Oncol Biol Phys 2000;48(Suppl):177. 297. Gamelin EC, Danquechin-Dorval EM, Dumesnil YF, et al. Relationship between 5-fluorouracil (5-FU) dose intensity and therapeutic response in patients with advanced colorectal cancer receiving infusional therapy containing 5-FU. Cancer 1996;77:441-451. 298. Gurney H. Dose calculation of anticancer drugs: a review of the current practice and introduction of an alternative. J Clin Oncol 1996;14:2590-2611. 299. Kornblau S, Benson AB, Catalano R, et al. Management of cancer treatment related diarrhea: issues and therapeutic strategies. J Pain Symptom Manage 2000;19:118-129. 300. Denham JW, O’Brien PC, Dunstan RH, et al. Is there more than one late radiation proctitis syndrome? Radiother Oncol 1999;51:43-53.

301. Miller AR, Martenson JA, Nelson H, et al. The incidence and clinical consequences of treatment-related bowel injury. Int J Radiat Oncol Biol Phys 1999;43:817-825. 302. Van Duijvendijk P, Slors JFM, Taat CW, et al. What is the benefit of preoperative sperm preservation for patients who undergo restorative proctocolectomy for benign disease? Dis Colon Rectum 2000;43:838-842. 303. Porter GA, Skibber JM. Outcomes research in surgical oncology. Ann Surg Oncol 2000;7:367-375. 304. Ramsey SD, Andersen MR, Etzioni R, et al. Quality of life in survivors of colorectal carcinoma. Cancer 2000;88:1294-1303. 305. Ko CY, Rusin LC, Schoetz DJ, et al. Does better functional ­result equate with better quality of life? Implications for surgical treatment in familial adenomatous polyposis. Dis Colon Rectum 2000;43:829-837. 306. Soffer EE, Hull T. Fecal incontinence: a practical approach to evaluation and treatment. Am J Gastroenterol 2000;95: 1873-1880. 307. Ulander K, Jeppsson B, Grahn G. Quality of life and independence in activities of daily living preoperatively and at followup in patients with colorectal cancer. Support Care Cancer 1997;8:402-409. 308. Even-Sapir E, Parag Y, Lerman H, et al. Detection of recurrence in patients with rectal cancer: PET/CT after abdominoperineal or anterior resection. Radiology 2004;232:815-822. 309. Tepper JE, O’Connell M, Hollis D, et al. Analysis of surgical salvage after failure of primary therapy in rectal cancer: results from Intergroup study 0114. J Clin Oncol 2003;21:3623-3628. 310. Esnaola NF, Cantor SB, Johnson ML, et al. Pain and quality of life after treatment in patients with locally recurrent rectal cancer. J Clin Oncol 2002;20:4361-4367. 311. Lingareddy V, Ahmad NR, Mohiuddin M. Palliative reradiation for recurrent rectal cancer. Int J Radiat Oncol Biol Phys 1997;38:785-790. 312. Mohiuddin M, Marks GM, Lingareddy V, et al. Curative surgical resection following reradiation for recurrent rectal cancer. Int J Radiat Oncol Biol Phys 1997;39:643-649. 313. Nymann T, Jess P, Christiansen J. Rate and treatment of ­pelvic recurrence after abdominoperineal resection and low ­anterior resection for rectal cancer. Dis Colon Rectum 1995;38: 799-802. 314. Wong CS, Cummings BJ, Brierley JD, et al. Treatment of locally recurrent rectal carcinoma—results and prognostic factors. Int J Radiat Oncol Biol Phys 1998;40:427-435. 315. Wong R, Thomas G, Cummings B, et al. In search of a doseresponse relationship with radiotherapy in the management of recurrent rectal carcinoma in the pelvis: a systematic review. Int J Radiat Oncol Biol Phys 1998;40:437-446. 316. Guiney MJ, Smith JG, Worotniuk V, et al. Results of external beam radiotherapy alone for incompletely resected carcinoma of rectosigmoid or rectum: Peter MacCallum Cancer Institute Experience 1981-1990. Int J Radiat Oncol Biol Phys 1999;43: 531-536. 317. Guiney MJ, Smith JG, Worotniuk V, et al. Radiotherapy treatment for isolated loco-regional recurrence of rectosigmoid cancer following definitive surgery: Peter MacCallum Cancer Institute experience. 1981-1990. Int J Radiat Oncol Biol Phys 1997;38:1019-1025. 318. Valentini V, Morganti AG, Gambacorta MA, et al. Preoperative hyperfractionated chemoradiation for locally recurrent rectal cancer in patients previously irradiated to the pelvis: a multicentric phase II study. Int J Radiat Oncol Biol Phys 2006;64: 1129-1139. 319. Bussieres E, Gilly FN, Rouanet P, et al. Recurrences of rectal cancers: results of a multimodal approach with intraoperative radiation therapy. Int J Radiat Oncol Biol Phys 1996;34:49-56. 320. Lanciano RM, Calkins AR, Wolkov HB, et al. A phase I/II study of intraoperative radiotherapy in advanced unresectable or ­recurrent carcinoma of the rectum: a Radiation Therapy Oncology Group (RTOG) study. J Surg Oncol 1993;53:20-29. 321. Sofo L, Ratto C, Doglietto GB, et al. Intraoperative radiation therapy in integrated treatment of rectal cancers—results of a phase II study. Dis Colon Rectum 1996;39:1396-1403.

24  The Colon and Rectum 322. Martinez-Monge R, Nag S, Martin EW. Three different intraoperative radiation modalities (electron beam, high-dose-rate brachytherapy, and Iodine-125 brachytherapy) in the adjuvant treatment of patients with recurrent colorectal adenocarcinoma. Cancer 1999;86:236-247. 323. Schneebaum S, Papo J, Fraif M, et al. Radioimmunoguided surgery benefits for recurrent colorectal cancer. Ann Surg Oncol 1997;4:371-376. 324. Kim HK, Jessup JM, Beard CJ, et al. Locally advanced rectal carcinoma: pelvic control and morbidity following preoperative radiation therapy, resection, and intraoperative radiation therapy. Int J Radiat Oncol Biol Physics 1997;38:777-783. 325. Shibata D, Hyland W, Busse P, et al. Immediate reconstruction of the perineal wound with gracilis muscle flaps following abdominoperineal resections and intraoperative radiation therapy for recurrent carcinoma of the rectum. Ann Surg Oncol 1999;6:33-37. 326. Salo JC, Paty PB, Guillem J, et al. Surgical salvage of recurrent rectal carcinoma after curative resection: a 10-year experience. Ann Surg Oncol 1999;6:171-177. 327. Wanebo HJ, Antoniuk P, Koness RJ, et al. Pelvic resection of recurrent rectal cancer—technical considerations and outcomes. Dis Colon Rectum 1999;42:1438-1448. 328. Wesselmann U, Burnett AL, Heinberg LJ. The urogenital and rectal pain syndromes. Pain 1997;73:269-294. 329. Mancini I, Bruera E. Constipation in advanced cancer patients. Support Care Cancer 1998;6:356-364. 330. Rousseau P. Management of malignant bowel obstruction in ­advanced cancer: a brief review. J Palliat Med 1998;1:65-72. 331. Wong CS, Brierley JD. External beam radiation therapy alone. Semin Radiat Oncol 1998;8:3-12. 332. Brierley JD, Cummings BJ, Wong CS, et al. Adenocarcinoma of the rectum treated by radical external radiation therapy. Int J Radiat Oncol Biol Phys 1995;31:255-259. 333. Crane CH, Janjan NA, Abbruzzese JL, et al. Effective pelvic symptom control using initial chemoradiation without colostomy in metastatic rectal cancer. Int J Radiat Oncol Biol Phys 2001;49:107-116. 334. Scoggins CR, Meszoely IM, Blanke CD, et al. Nonoperative management of primary colorectal cancer in patients with stage IV disease. Ann Surg Oncol 1999;6:651-657. 335. Videtic GM, Fisher BJ, Perera FE, et al. Preoperative radiation with concurrent 5-fluorouracil continuous infusion for locally advanced unresectable rectal cancer. Int J Radiat Oncol Biol Phys 1998;42:319-324. 336. Rowe VL, Frost DB, Huang S. Extended resection for locally advanced colorectal carcinoma. Ann Surg Oncol 1997;4:131-136. 337. Camunez F, Echenagusia A, Simo G, et al. Malignant colorectal obstruction treated by means of self-expanding metallic stents: effectiveness before surgery and in palliation. Radiology 2000;216:492-497. 338. Binkert CA, Ledermann H, Jost R, et al. Acute colonic obstruction: clinical aspects and cost-effectiveness of preoperative and palliative treatment with self-expanding metallic stents—a preliminary report. Radiology 1998;206:199-204. 339. Luna-Perez P, Delgado S, Labastida S, et al. Patterns of recurrence following pelvic exenteration and external radiotherapy for locally advanced primary rectal adenocarcinoma. Ann Surg Oncol 1996;3:526-533. 340 Gunderson LL, Nelson H, Martenson JA, et al. Locally advanced primary colorectal cancer: intraoperative electron beam and external beam radiation +/− 5-FU. Int J Radiat Oncol Biol Phys 1997;37:601-614. 341. Chan AKP, Wong AO, Langevin JM, et al. “Sandwich” preoperative and postoperative combined chemotherapy and radiation in tethered and fixed rectal cancer: impact of treatment intensity on local control and survival. Int J Radiat Oncol Biol Phys 1997;37:629-637.

605

342. Marsh RD, Chu NM, Vauthey JN, et al. Preoperative treatment of patients with locally advanced unresectable rectal adenocarcinoma utilizing continuous chronobiologically shaped 5-­fluorouracil infusion and radiation therapy. Cancer 1996;78:217-225. 343. Minsky BD, Cohen AM, Enker WE, et al. Preoperative 5-FU, low-dose leucovorin, and radiation therapy for locally advanced and unresectable rectal cancer. Int J Radiat Oncol Biol Phys 1997;37:289-295. 344. Minsky BD, Cohen AM, Enker WE, et al. Preoperative 5-FU, low dose leucovorin, and concurrent radiation therapy for rectal cancer. Cancer 1993;73:273-278. 345. Ky AJ, Sung MW, Milsom JW. Research in colon and rectal cancer, with an emphasis on surgical progress. Dis Colon Rectum 1999;42:1369-1380. 346. Koea JB, Conlon K, Paty PB, et al. Pancreatic or duodenal resection or both for advanced carcinoma of the right colon—is it justified? Dis Colon Rectum 2000;43:460-465. 347. Longo WE, Virgo KS, Johnson FE, et al. Risk factors for morbidity and mortality after colectomy for colon cancer. Dis Colon Rectum 2000;43:83-91. 348. Amos EH, Mendenhall WM, McCarty PJ, et al. Postoperative radiotherapy for locally advanced colon cancer. Ann Surg Oncol 1996;3:431-436. 349. Janjan NA. Postoperative radiotherapy for locally advanced ­colon cancer. Ann Surg Oncol 1996;3:421-422. 350. Merrick HW, Turner SS, Dobelbower RR, et al. Large-field ­external beam radiation as a surgical adjuvant for node-­positive colon carcinoma: an Eastern Cooperative Oncology Group Pilot Study (PA285). Am J Clin Oncol 2000;23:337-340. 351. Willett CG, Goldberg S, Shellito PC, et al. Does postoperative radiation play a role in the adjuvant therapy of T4 colon cancer? Cancer J Sci Am 1999;5:242-247. 352. Moertel CG, Fleming TR, Macdonald JS, et al. Levamisole and fluorouracil for adjuvant therapy of resected colon carcinoma. N Engl J Med 1990;322:352-358. 353. Moertel CG, Fleming TR, Macdonald JS, et al. Fluorouracil plus levamisole as effective adjuvant therapy after resection of stage III colon carcinoma: a final report. Ann Intern Med 1995;122:321-326. 354. Martenson JA Jr, Willett CG, Sargent DJ, et al. Phase III study of adjuvant chemotherapy and radiation therapy compared with chemotherapy alone in the surgical adjuvant treatment of colon cancer: results of Intergroup protocol 0130. J Clin Oncol 2004;22:3277-3283. 355. O’Connell MJ, Mailliard JA, Kahn MJ, et al. Controlled trial of fluorouracil and low-dose leucovorin given for 6 months as postoperative adjuvant therapy for colon cancer. J Clin Oncol 1997;15:246-250. 356. Zaniboni A. Adjuvant chemotherapy in colorectal cancer with high-dose leucovorin and fluorouracil: impact on disease-free survival and overall survival. J Clin Oncol 1997;15:2432-2441. 357. Porschen R, Bermann A, Loffler T, 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;19:1787-1794. 358. Wolmark N, Rockette H, Mamounas E, et al. Clinical trial to ­ assess the relative efficacy of fluorouracil and leucovorin, fluorouracil and levamisole, and fluorouracil, leucovorin, and ­levamisole in patients with Dukes B and C carcinoma of the colon: ­results from National Surgical Adjuvant Breast and Bowel Project C-04. J Clin Oncol 1999;17:3553-3559. 359. Mamounas E, Wieand S, Wolmark N, et al. Comparative efficacy of adjuvant chemotherapy in patients with Dukes B versus Dukes C colon cancer: results from four National Adjuvant Breast and Bowel Project Adjuvant Studies (C-01, C-02, C-03, and C-04). J Clin Oncol 1999;17:1349-1355.

25 The Anal Region Prajnan Das, Michelle S. Ludwig, Nora A. Janjan, And Christopher H. Crane ANATOMY TOLERANCE TO RADIATION TUMORS OF THE ANAL CANAL Epidemiology Pathology Clinical Presentation Staging Prognostic Factors TREATMENTS FOR TUMORS OF THE ANAL CANAL Combined-Modality Therapy Treatment of Patients with Human Immunodeficiency Virus Infection Summary of the Approach to Anal Canal Tumors

Although anal cancer is uncommon, innovations in its treatment have made it a model for the use of organ preservation and combined-modality therapy. Previously, abdominoperineal resection was the only therapeutic option for anal cancer, but advances in radiation techniques and seminal studies combining chemotherapy and radiation have led to their use as definitive treatment for anal cancer, reserving surgery for recurrent or persistent disease. Functional ­outcome and quality of life are key components of anal cancer treatment. Although the total number of patients with anal cancer is small, innovations in the treatment of this disease have led to the establishment of therapeutic principles that have been applied in the ­ treatment of ­almost every type of cancer.

Anatomy The anatomy of the anal region is shown in Figure 25-1.1 The anal canal is about 3 to 4 cm long. The pectinate or dentate line is a histologic transition zone between squamous and columnar epithelium at the level of the anal valves. According to the American Joint Commission on Cancer (AJCC) and the Union International Contre le Cancer (UICC) clinical staging guidelines, the distal end of the anal canal is defined as the level at which the walls come in contact in their normal resting state,2 at the level of the anorectal ring. The anorectal ring is the palpable muscle bundle formed by the upper portion of the internal sphincter, the deep or subcutaneous part of the external sphincter, the puborectalis muscle, and the distal longitudinal muscle from the large bowel.1,3 In the AJCC system, perianal ­tumors are located within a 5- to 6-cm radius from the anal verge in the buttock and perineal region, and the tumors are staged according to the system used for cancer of the skin.

606

TREATMENTS FOR TUMORS OF THE ANAL MARGIN RADIATION THERAPY TECHNIQUES Balancing Efficacy with Toxicity: Dose-Time Considerations External Beam Radiation Therapy with an Interstitial Implant MEDICAL MANAGEMENT OF TREATMENT-RELATED TOXIC EFFECTS SURGERY AS SALVAGE TREATMENT SUMMARY

Local and lymphatic extensions are the most common routes of spread. Hematogenous dissemination is less common, although the risk of disease spreading in this way is higher among patients with advanced disease. Blood is supplied to the area above the pectinate line by the superior and middle rectal arteries, which are branches of the inferior mesenteric and hypogastric arteries, respectively. Venous drainage is to the portal system.3 Below the pectinate line, the arterial supply is to the middle and inferior rectal arteries, and venous drainage is to the inferior rectal vein. Direct invasion into the sphincteric muscles and perianal connective tissue occurs early in the course of the disease, and about one half of patients present with ­ tumor invasion of the rectum or perianal region, or both.4,5 ­Extensive tumors can infiltrate the sacrum or the pelvic sidewalls. ­Extension to the vagina in women is common, but invasion of the prostate gland in men is uncommon. Only 10% of patients are found to have distant metastases at the time of diagnosis. When distant metastases ­occur, they are most commonly found in the liver and lungs. However, local ­ relapse is more common than the development of ­extrapelvic disease. Tumors proximal to the pectinate line drain to the perirectal, external iliac, obturator, hypogastric, and para-aortic lymph nodes. At abdominoperineal resection, about 30% of patients with tumors in the anal canal are found to have pelvic lymph node metastases, and 16% have inguinal node metastases.1,4 Tumors in the distal canal drain to the ­inguinal-femoral node region and the external and common iliac nodal regions. About 15% to 20% of patients have clinical evidence of inguinal lymph node involvement at presentation, and involvement is usually unilateral.6 Inguinal node metastases are evident in 30% of superficial tumors and 63% of deeply infiltrating or poorly differentiated tumors.7

25  The Anal Region

607

ASIS

Rectum Anorectal ring Pectinate line

Transitional zone

Anal anatomy conventions

B 1 cm

Anal canal Anal canal

Pecten Intersphincteric groove

D

Anal margin Anal verge

PT

Perianal

Perianal skin

FIGURE 25-1.  Anatomy of the anal region. (Modified from Cummings BJ. Anal canal. From Perez CA, Brady LW, eds. Principles and Practice of Radiation Oncology, 3rd ed. Philadelphia: Lippincott-Raven, 1998, pp 1511-1524.)

The inguinal nodes are located in an anatomic region bounded by defined landmarks. The most medial location for the inguinal lymph nodes is 3 cm from the symphysis pubis or midline; from there, the inguinal nodes extend to the lateral aspect of the femoral head.8 The most inferior location is 2.5 cm caudal to the inferior pubic ramus, and the most superior location is the superior aspect of the femoral head (Fig. 25-2).

Tolerance to Radiation Radiation therapy, with or without chemotherapy, provides good local tumor control, usually with preservation of anal sphincter function. The acute effects of radiation or chemoradiation therapy, typically arising 1 or 2 weeks after therapy has begun, result largely from direct injury to the anal epithelial and endothelial cells. Management of these effects, which can include desquamation, pain, and diarrhea, is discussed later under “Medical Management of Treatment-Related Toxic Effects.” Delayed injury to fibroblasts and myofibroblasts can lead to stromal fibrosis and intimal vascular lesions. Late complications can include skin or soft tissue necrosis, anal stenosis, anorectal bleeding, diarrhea, bladder bleeding, leg edema, and injury to the small bowel.9 Complications such as anal canal fibrosis or stenosis or sphincter incontinence requiring abdominoperineal resection are rare10 and typically do not occur at doses lower than 60 Gy.9 Female patients commonly develop vaginal fibrosis and dryness that can result in dyspareunia. Careful use of specialized techniques such as intensitymodulated radiation therapy (IMRT) and three-dimensional conformal planning can avoid unnecessary irradiation of the vulva and vagina (discussed later). If the female urethra is within the radiation field, urethral stricture is possible, but uncommon.10-12 Anal sphincter incompetence results from radiation injury to the smooth and striated muscle sphincter apparatus and perhaps from damage to the neurosensory component of the mucosa.9 Reversible intermittent sphincter incompetence also can result from internal or external hemorrhoids.

3 cm

C

A

2.5 cm

FIGURE 25-2.  X-ray image shows the anatomic location of the inguinal nodes; 86% of the lymph nodes lie within the ­defined rectangle. ASIS, anterior superior iliac spine; PT, pubic ­tubercle.

BOX 25-1.  Risk Factors for Anal Cancer Strong Supporting Evidence Human papillomavirus infection (anogenital warts) History of receptive anal intercourse History of sexually transmitted disease More than 10 sexual partners History of cervical, vulvar, or vaginal cancer Immunosuppression after solid-organ transplantation Moderately Strong Supporting Evidence Human immunodeficiency virus infection Long-term use of corticosteroids Cigarette smoking

Tumors of the Anal Canal Epidemiology Anal cancer is uncommon, representing less than 2% of all types of cancer of the digestive tract. However, risk factors for anal cancer share some similarities with those for other types of cancer (Box 25-1), and anal cancer tends to be more common in groups that share those factors. For example, high rates of infection with human papillomavirus (HPV) have been observed in cases of cervical cancer, vulvar cancer, and anal cancer. Tissue specimens from more than 85% of patients with anal cancer have been found to contain HPV deoxyribonucleic acid (DNA).3,13-15 Among women, a history of seropositivity for herpes simplex type 2 or Chlamydia trachomatis is associated with an increased risk of anal cancer. Men who have sexual

608

PART 6  Gastrointestinal Tract

i­ntercourse with men are at an increased risk for HPV­induced squamous ­intraepithelial lesions and anal cancer.16-19 ­Cost-­effectiveness models have shown that anal Papanicolaou screening of men who have sex with men every 2 to 3 years would provide life-expectancy benefits comparable to those of other screening procedures.20 Regardless of their sexual practices, individuals who are seropositive for the human immunodeficiency virus (HIV) have twice to three times the risk for anal HPV infection compared with HIV-seronegative individuals. Virus­induced immunosuppression is considered to increase the risk of anal HPV infection and subsequent anal cancer3,21; an inverse relationship has been shown between CD4+ T-cell count and HPV infection.19 Immunosuppression from other causes, such as organ transplantation, is also ­associated with a 100-fold increase in the risk of vulvar and anal cancer because of the high prevalence of HPV infection.22,23 Unlike HPV or HIV infection, however, no direct relationship has been found between the acquired immune deficiency syndrome (AIDS) and the development of anal cancer.18,24 Cigarette smoking is strongly associated with anal cancer and cervical cancer, raising the risk of anal cancer by a factor of 2 to 5.21 A prior diagnosis of anal cancer also increases the risk of lung cancer by a factor of 2.5.25-27 No association has been found between anal cancer and inflammatory bowel disease. Among 9602 patients with Crohn’s disease or ulcerative colitis who were followed for up to 18 years, only two cases of anal cancer were identified during 99,229 person-years of observation.28 Other inflammatory conditions and trauma have not been found to result in a higher incidence of anal cancer.29

Pathology The rectum and the anus have separate embryonic ­origins, and different types of cells line the two structures. ­Glandular mucosa lines the rectum, and lieberkühnian (glandular) adenocarcinoma predominates there; squamous cell ­mucosa lines the anal canal, and squamous epithelial carcinoma predominates in the anus.30 Between the glandular mucosa of the rectum and the squamous mucosa of the anal region lies the transitional epithelium (see Fig. 25-1), which extends for about 6 to 20 mm and incorporates rectal, urothelial, and squamous elements. Another area of transitional epithelium exists between the squamous epithelium in the anal canal and the skin around the anus; this region, called the pecten,1 is bordered superiorly by the pectinate line (see Fig. 25-1). On the inferior side of the pecten, the perianal region is lined by skin similar to that located elsewhere and contains apocrine glands. Keratinizing squamous cell carcinoma is the most common type of anal cancer arising distal to the pectinate line; cancer that develops in the transition zone, between the squamous and columnar epithelium around the pectinate line, is usually nonkeratinizing squamous cell carcinoma and is often referred to as cloacogenic cancer.1,3 Squamous cell carcinoma of either histologic type should be treated with definitive chemoradiation therapy.31 Other histologic types of anal cancer include adenocarcinoma and ­melanoma. A National Cancer Data Base report32 on carcinoma of the anus reflects the substantial improvement achieved in the treatment of anal cancer between 1988 and 1993.

At both assessment times, the distribution of histologic subtypes was similar: approximately 75% squamous cell carcinomas and about 19% adenocarcinomas. Among ­patients with squamous cell carcinoma of the anal canal, 18% were treated with surgery alone in 1988 and 15% in 1993; chemoradiation therapy was given as definitive treatment to 43% of patients in 1988 and 47% in 1993. Of the ­patients with adenocarcinoma, about 37% were treated with surgery alone at both evaluation times, and only about 8% were treated with definitive chemoradiation therapy (Table 25-1).32 The patterns of treatment failure among the patients surveyed in this report are shown in Table 25-2.32 The rates of local and regional recurrence were twice as high for squamous cell tumors as for adenocarcinomas; however, the risk of distant metastasis was twice as high for adenocarcinoma as for squamous cell cancer. These findings suggest that more than 25% of patients with anal adenocarcinoma will develop distant metastases.32,33 In a review from the University of Minnesota of 192 cases of anal cancer (119 women and 73 men; median age, 58 years),34 squamous cell carcinoma accounted for 74%, adenocarcinoma for 19%, and melanoma for 4% of all tumors. The remaining 3% were neuroendocrine tumors (two cases), carcinoid (one case), Kaposi’s sarcoma (one case), leiomyosarcoma (one case), and lymphoma (one case). In terms of T classification, 3% of patients had T1 disease, 46% had T2, 28% had T3, and 12% had T4 disease. The 5-year survival rates according to tumor histology were 57% for patients with squamous cell carcinoma, 63% for those with adenocarcinoma, and 33% for those with melanoma. The 5-year survival rates according to T category were 62% for those with T1 disease, 57% for T2 disease, 45% for T3 disease, and 17% for T4 disease.32 Although anal adenocarcinoma is considered to be more aggressive than squamous cell carcinoma, the outcome for the 36 patients with adenocarcinoma in the University of Minnesota series34 was comparable to that for patients with squamous cell carcinoma. Of those patients, 22 (61%) underwent surgery and the other 14 (39%) underwent surgery and adjuvant chemoradiation therapy. The 5-year survival rate was 63%, and the recurrence rate was 21%, rates similar to those among patients with squamous cell carcinoma (who had been treated with surgery or chemoradiation therapy). These studies probably included some cases in which adenocarcinoma of the rectum extending into the anal canal had been classified as adenocarcinoma of the anal canal. Studies in which adenocarcinoma is confirmed of being of true anal ductal origin by pathology review tend to have considerably fewer patients.33 In one such study, a ­ retrospective review of 16 patients with confirmed anal adenocarcinoma from The University of Texas M.  D. ­Anderson Cancer Center, definitive treatment with radiation (median dose, 55 Gy) and concurrent chemotherapy resulted in an actuarial local tumor control rate of only 46% at 5 years, with a 66% rate of distant recurrence.33 In contrast, chemoradiation therapy provided good locoregional control in a smaller study of anal adenocarcinoma at the Peter ­MacCallum Cancer Institute in Melbourne, Australia.35 In that study, locoregional control was achieved for all six of the patients given chemoradiation or radiation therapy with curative intent; none of the nine patients treated with palliative intent survived the disease.

25  The Anal Region

609

TABLE 25-1.    National Cancer Data Base Comparison of Treatment Modalities for Anal Cancer 1988 Treatment

Number of cases

1993

Percentage (%)

Number of Cases

Percentage (%)

All Cases Surgery only

252

24.0

258

20.0

Radiation therapy only

40

3.8

61

4.7

Chemotherapy only

22

2.1

11

0.9

104

9.9

64

5.0

Surgery and radiation therapy Surgery and chemotherapy

17

1.6

14

1.1

Radiation therapy and chemotherapy

357

34.0

511

39.6

Surgery, radiation therapy, and chemotherapy

207

19.7

299

23.2

No treatment

34

3.3

52

4.0

Unknown

17

1.6

19

1.5

1050

100.0

1289

100.0

141

18.1

157

15.4

Radiation therapy only

27

3.5

53

5.2

Chemotherapy only

18

2.3

9

0.9

Surgery and radiation therapy

48

6.2

44

4.3

Surgery and chemotherapy

12

1.5

4

0.4

Radiation therapy and chemotherapy

336

43.1

483

47.4

Surgery, radiation therapy, and chemotherapy

174

22.3

230

22.5

15

1.9

25

2.5

9

1.2

15

1.5

780

100.0

1020

100.0

Surgery only

89

38.7

78

36.8

Radiation therapy only

13

5.7

8

3.8

3

1.3

0

0.0

53

23.0

19

9.0

4

1.7

8

3.8

Radiation therapy and chemotherapy

17

7.4

19

9.0

Surgery, radiation therapy, and chemotherapy

30

13.0

59

27.8

No treatment

14

6.1

19

9.0

7

3.0

2

0.9

230

100.0

212

100.0

Total Epidermoid Carcinoma Surgery only

No treatment Unknown Total Adenocarcinoma

Chemotherapy only Surgery and radiation therapy Surgery and chemotherapy

Unknown Total

From Myerson RJ, Karnell LH, Menck HR. The National Cancer Data Base report on carcinoma of the anus. Cancer 1997;80:805-815.

Clinical Presentation Bleeding and anal discomfort are the most common symptoms of anal cancer, occurring in about one half of the ­patients with this disease. Less common symptoms include the sensation of a mass in the anus, pruritus, and anal discharge.1 Proximal tumors can cause obstructive symptoms, but distal tumors rarely do. Less than 5% of patients have sphincteric destruction resulting in fecal incontinence.

­ istulas between the anus and the vaginal or other strucF tures are uncommon. In metastatic disease, metastases involve primarily the lymph nodes, liver, and lung. In one study of 19 cases of metastatic anal cancer,36 6 were considered synchronous and 13 metachronous. Nodal metastases involved the paraaortic nodes in five cases, the iliac region in four cases, and the inguinal nodes in two cases; nine of the cases involved

610

PART 6  Gastrointestinal Tract

TABLE 25-2.    Tumor Recurrences According to Histologic Type and Disease Stage Type of Recurrence (%)* Tumor All cases

No Recurrence (%)*

Local

Regional

Distant

9.2

6.4

16.4

48.7

Never DiseaseFree (%)* Total (%)

Total Number of Cases

19.3

100.0

347

Histology Epidermoid carcinoma

10.5

7.7

11.8

51.8

18.2

100.0

247

5.6

3.4

28.1

41.6

21.3

100.0

89

I

9.8

3.3

8.2

68.9

9.8

100.0

61

II

9.8

16.4

16.4

42.6

14.8

100.0

61

III

4.5

11.1

33.3

20.0

31.1

100.0

45

IV

3.2

0.0

16.1

42.0

38.7

100.0

31

Adenocarcinoma Stage†

*Recurrence

status was unknown for 703 cases, reducing the total number of cases from 1050 to 347. stage, augmented by clinical stage when pathologic stage was not available. From Myerson RJ, Karnell LH, Menck HR. The National Cancer Data Base report on carcinoma of the anus. Cancer 1997;80:805-815.

†Pathologic

solitary metastases. Metastases to the liver were ­documented in 10 cases (of which 7 were solitary) and metastases to the lung in 3 cases. With local and systemic therapy, the median survival time was 35 months.36

Staging The AJCC staging system (Box 25-2)2 is the most commonly used scheme for staging carcinomas of the anal canal. Anal canal tumors are staged mostly according to their size and local extension.2,3 Tumors that arise in the anal margin are staged according to the system used for skin cancer.

Prognostic Factors Tumor size and nodal status are the most significant prognostic factors in anal cancer. Among 132 patients who ­underwent abdominoperineal resection for anal cancer, the 5-year ­survival rates were 78% for those with 1- to 2-cm ­tumors, 55% for those with 3- to 5-cm tumors, and 40% for 6-cm or larger tumors.7 The relationship between tumor size and ­survival rate holds regardless of whether the treatment is ­ radiation alone10 or chemoradiation therapy.37,38 The ­ presence of nodal disease also influences survival, with another study showing 5-year survival rates of 44% among ­node-positive patients and 74% among node-negative ­patients.7 Disease stage at presentation can be used to predict outcome. Among patients with T1 or T2 disease, disease-free and colostomy-free survival rates were significantly better (87% and 94%, respectively) than those among patients who presented with T3 or T4 disease (59% disease-free survival and 73% colostomy-free survival). Patients with node-negative disease at presentation also had significantly better cancer-related survival rates (85%, compared with 58% among node-positive patients) and disease-free survival rates (80%, compared with 53% among node-positive patients).39 A study at M.  D. Anderson Cancer Center evaluated prognostic factors among 167 patients with nonmetastatic squamous cell anal carcinoma that had been treated with definitive chemoradiation therapy.40 Higher T category and higher N category independently predicted higher rates of

locoregional failure: 3-year locoregional control rates were 90% for those with T1 or Tx tumors, 86% for T2 tumors, 77% for T3 tumors, and 63% for those with T4 tumors; the corresponding 3-year locoregional control rates in terms of nodal involvement were 84% to 88% for those with N0 to N2 tumors and 39% for those with N3 tumors. Higher N category independently predicted lower rates of distant control and overall survival: 3-year distant control rates were 94% for those with N0 disease, 79% for N1 disease, 75% for N2 disease, and 76% for those with N3 disease. The corresponding 3-year overall survival rates were 93% for those with N0 disease, 74% for N1 disease, 74% for N2 disease, and 56% for those with N3 disease.40

Molecular Characteristics Efforts are under way to assess possible associations between molecular characteristics and clinical outcome. A study of patients treated in the Radiation Therapy Oncology Group (RTOG) 87-04 trial showed trends toward lower locoregional control and overall survival among patients with tumors that overexpressed the tumor suppressor TP53 (formerly called p53 or P53).41 A study from the Princess Margaret Hospital also showed that increasing TP53 expression was associated with worse disease-free survival and locoregional control.42 Conversely, high expression of the cell cycle regulator protein cyclin A has been associated with higher rates of overall survival and lower rates of locoregional failure.43 Decreased CDKN1A (formerly called p21) expression has been linked with lower overall survival.44 Most studies on molecular and biologic markers have been based on limited numbers of patients, with limited concordance between studies regarding the role of molecular markers.

Treatments for Tumors of the Anal Canal Combined-Modality Therapy The superiority of radiation or chemoradiation treatment over surgery alone in achieving local tumor control while preserving anorectal function without sacrificing survival

25  The Anal Region

611

BOX 25-2.  American Joint Committee on Cancer Staging System for Cancer of the Anal Canal Primary Tumor (T) TX Primary tumor cannot be assessed T0 No evidence of primary tumor Tis Carcinoma in situ T1 Tumor 2 cm or less in greatest dimension T2 Tumor more than 2 cm but not more than 5 cm in greatest dimension T3 Tumor more than 5 cm in greatest dimension T4 Tumor of any size that invades adjacent organ(s), e.g., vagina, urethra, bladder (direct invasion of the rectal wall, perirectal skin, or the sphincter muscle(s) is not classified as T4) Regional Lymph Nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Metastasis in perirectal lymph node(s) N2 Metastasis in unilateral internal iliac and/or inguinal lymph node(s) N3 Metastasis in perirectal and inguinal lymph nodes and/or bilateral internal iliac and/or inguinal lymph nodes Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage Grouping Stage 0 Tis Stage I T1 Stage II T2 T3 Stage IIIA T1 T2 T3 T4 Stage IIIB T4 Any T Any T Stage IV Any T

N0 N0 N0 N0 N1 N1 N1 N0 N1 N2 N3 Any N

M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M1

From Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 125-130.

has served as a model for organ preservation in other types of cancer. The use of radiation alone (to total doses of 60 to 65 Gy) has produced local control rates of about 70% and 5-year survival rates of about 60%.37,45 However, these rates have varied considerably among studies, ranging from 56% to 100% for local control and from 50% to 94% for 5-year overall survival.45,46 Exploration of combined-modality therapy for anal cancer began in 1974, when Nigro and colleagues47 gave three patients radiation therapy in combination with 5-fluorouracil (5-FU) and mitomycin C before abdominoperineal resection. Pathologic examination of surgical specimens from two of those patients revealed complete response with no evidence of residual tumor; the third patient declined surgery and remained free of disease. Subsequent trials conducted by the United Kingdom Coordinating Committee for Cancer Research48 and the European Organisation for Research and Treatment of Cancer (EORTC)49,50 have shown significant improvements from the addition of chemotherapy to radiation in terms of primary tumor control and colostomy-free survival rate but not in terms of overall survival rate (Table 25-3).3,48-52 These trials are described in more detail in the following sections.

Radiation Therapy versus Chemoradiation Therapy with 5-Fluorouracil and Mitomycin C In a trial conducted by the United Kingdom Coordinating Committee for Cancer Research,48 577 patients were randomized to receive radiation alone or combinedmodality therapy. In all cases, the total radiation dose was 45 Gy, given in 20 to 25 fractions over 4 or 5 weeks. The combined-modality group was given radiation in combination with 5-FU (1000 mg/m2 for 4 days or 750 mg/m2 for 5 days) and mitomycin C (12 mg/m2 as an intravenous bolus injection) on the first day of each course of chemotherapy (see Table 25-3).48 If the tumor had not regressed by at least 50% at 6 weeks after the completion of chemoradiation therapy, surgery or further radiation (15 Gy given in 6 fractions to a perineal field or an interstitial implant of 25 Gy over 2 to 3 days) was considered. The local recurrence rates were 36% in the combined-modality group and 59% in the radiation-only group (see Table 25-3). Except for hematologic effects, which were more common in the combined-modality group, the side effects of treatment were comparable among the groups.

612

PART 6  Gastrointestinal Tract

TABLE 25-3.    Randomized Trials of Chemoradiation Therapy for Squamous Cell Anal Carcinoma Study and ­References UK Coordinating ­ Committee for Cancer ­Research48

European Organisation for Research and Treatment of Cancer49,50

No. of Patients

Variable

585

RT only

Early toxicity

48%

39%

0.03

Local failure

36%

59%

<0.0001

Anal cancer mortality

28%

39%

0.02

Overall survival

65%

58%

0.25

RT + FU + MitoC

RT only

Complete remission

80%

54%

Locoregional control*

68%

50%

0.02

Colostomy-free rate*

72%

40%

0.002

58%

54%

0.17

RT + FU + MitoC

RT + FU

Grade 4-5 toxicity

23%

7%

<0.001

Colostomy rate

9%

22%

0.002

Colostomy-free survival

71%

59%

0.014

Disease-free survival

73%

51%

0.0003 0.31

110

survival*

310

survival*

75%

68%

FU + MitoC + RT

FU + CP + RT

Grade 3-4 ­ hematotoxicity

61%

42%

<0.001

Colostomy rate

10%

19%

0.02

Disease-free survival

60%

54%

0.17

Overall survival

75%

70%

0.10

Overall Radiation Therapy Oncology Group 98-1172

P Value

RT + FU + MitoC

Overall Radiation Therapy Oncology Group 87-0451

Type of Treatment and Results

682

*Estimated

from figures Modified from Das P, Crane CH, Ajani JA. Current treatment for localized anal carcinoma. Curr Opin Oncol 2007;19:396-400. CP, cisplatin; FU, fluorouracil; MitoC, mitomycin C; RT, radiation therapy

Another comparison of radiation alone versus chemoradiation therapy by the EORTC49,50 included 103 patients and had a design similar to that of the United Kingdom trial. Radiation was given to a total dose of 45 Gy over 5 weeks; chemotherapy included 5-FU (750 mg/m2 for 5 days), given as a continuous infusion during the first and fifth weeks of radiation, and mitomycin C (15 mg/m2), ­given as an intravenous bolus on the first day of each course of chemotherapy. After 6 weeks, a boost dose of radiation by external beam radiation therapy (EBRT) or interstitial techniques was given, depending on the level of the response; 15 Gy was given if the response was complete, and 20 Gy was given if the response was partial. Findings from this study, which showed the benefit of combined-modality therapy over radiation alone, are summarized in Table 25-3. Tumor regression rates were significantly higher in

the combined-modality group.49,50 The colostomy-free survival rate at 5 years was 70% for the combined-­modality therapy group and only 52% for the group treated with radiation alone (P = .002). As was true in the United Kingdom study, combined-modality treatment resulted in better local control and colostomy-free survival rates but did not affect overall survival rate (see Table 25-3).

Chemoradiation Therapy Protocols with 5-Fluorouracil and Mitomycin C Several other trials have shown combinations of radiation plus chemotherapy with 5-FU and mitomycin C to be ­effective in anal cancer. In one such study, the Eastern Cooperative Oncology Group (ECOG)53 treated 50 patients with 40 Gy to the pelvis and a 10- to 13-Gy boost to the tumor; 5-FU and mitomycin C were given on days 1 and 28 of

25  The Anal Region

radiation. An abdominoperineal resection was performed if a biopsy conducted 6 to 8 weeks after the completion of the chemoradiation protocol showed signs of disease. This protocol produced a 74% complete response rate and a 24% partial response rate; the local control rate was 80% and the overall survival rate was 58%. However, this protocol was found to be highly toxic; 37% of patients experienced severe toxic effects, including life-threatening hematologic toxicity in 4%. In another trial of chemoradiation therapy,54 the RTOG evaluated 79 patients who had been treated between 1983 and 1987. The radiation dose to the pelvis and perineum was 40.8 Gy (given in 24 fractions over 35 days), and chemotherapy consisted of 5-FU (1000 mg/m2 as a continuous intravenous infusion on days 2 through 5 and 28 through 31 of radiation) and mitomycin C (given as a bolus dose of 10 mg/m2 on the first day of radiation). In that study, the colostomy-free survival rate was 90%, and the local control and survival rates at 3 years were 71% and 73%, respectively. Acute toxic effects of the treatment included mild to moderate diarrhea in 67% of patients, skin reactions in 86%, moist desquamation in 19%, and neutropenia in 18% (severe in 3%). Late toxic effects were uncommon. A group at Memorial Sloan-Kettering Cancer Center treated 42 patients with one cycle of 5-FU and mitomycin C and a 30-Gy dose of radiation.55 This protocol produced a 5-year survival rate of 82%; moreover, only one half of the 18 patients in whom disease persisted after therapy had progressive disease.55 The conclusion from this study was that evidence of disease at biopsy shortly after completing chemoradiation therapy does not represent clinically viable tumor nor should it prompt an abdominoperineal resection. Issues of toxicity and biopsy confirmation after combination chemotherapy were addressed in a joint RTOG/ ECOG study in which chemoradiation therapy was prospectively compared with 5-FU alone or in combination with mitomycin C.51 Radiation therapy consisted of 45 Gy given in 25 fractions or 50.4 Gy given in 28 fractions over 5 weeks. In all cases, 5-FU (1000 mg/m2) was given as a continuous infusion over 4 days on the first and fifth week of radiation. One group was given mitomycin C (10 mg/m2) on the first day of each course of ­chemotherapy. Grade 4 or 5 toxic effects were experienced by 23% of the patients given combined-modality therapy, compared with 7% of those given radiation alone. Mitomycin C was associated with more severe hematologic toxic effects, with fatal neutropenia occurring in 2% of cases. At 4 years of follow-up, the combination-chemotherapy group showed better colostomy rates (9% versus 23%), colostomy-free survival rates (71% versus 59%), and disease-free survival rates (73% versus 51%) relative to the radiation-plus5-FU–only group.51 However, no difference was found in terms of overall survival between the two groups (see Table 25-3). These findings are consistent with the results of a subsequent retrospective review of 287 cases of anal cancer treated with radiation alone or in combination with chemotherapy. In that review, the 5-year overall survival rate with combined-modality therapy was 78%, and with radiation alone, it was 60%.56 Other aspects of the RTOG/ECOG study51 that must be considered involved the timing of biopsies and the use of salvage therapy. Biopsies were required 6 weeks after the completion of the radiation. Patients with positive

613

­ iopsy findings (15% in the group receiving radiation plus b 5-FU only versus 8% in the group receiving radiation plus combination chemotherapy) were given an additional 9 Gy of radiation (in 5 fractions) with a 4-day infusion of 5-FU (1000 mg/m2) and a single injection of cisplatin (100 mg/m2). Of the 24 patients who underwent salvage therapy, 12 were rendered disease-free. However, in light of the Memorial Sloan-Kettering findings,55 it is unclear how many of these patients actually derived benefit from the additional chemoradiation therapy. Moreover, repeated biopsies traumatize irradiated tissues, increasing the risk of infection and delayed wound healing. Despite many trials having been conducted on the topic (Table 25-4),1,5,46,54,57-61 the optimal radiation dose and schedule for radiation therapy given in combination with 5-FU and mitomycin C remain unknown. Local control rates have ranged between 70% and 95%, and the incidence of treatment-related complications leading to colostomy have ranged from 5% to 10%. For treating tumors up to about 3 cm in diameter, schedules that give 30 Gy in 15 fractions over 3 weeks or 40.8 Gy in 24 fractions over 5 weeks have shown excellent control rates.1,54,57 Higher radiation doses are used to treat larger tumors, and they usually produce local control rates exceeding 75%.1,5,46,54,57-61 The total doses to the pelvis range from 40 to 45 Gy and are given over 5 weeks; additional localized boosts to the area of gross tumor involvement typically bring the total dose to 50 to 65 Gy. In contrast to the variations in radiation scheduling, the schedule for administering 5-FU and mitomycin C has tended to be more consistent among the trials (see Table 25-4).

Other Drug Combinations Concerns over the hematologic toxicity of mitomycin C and its long-term effects on the lungs, kidneys, and bone marrow have led to the use of cisplatin with 5-FU for the treatment of anal cancer (Table 25-5).1,63-68 This combination has been shown to yield complete response rates of 90% to 95% and colostomy-free survival rates of 86% at 3 years.60,64,65 Moreover, the toxic effects of cisplatin and 5-FU, particularly on the hematologic system, seem to be more limited. For example, early results from the E4292 trial of the ECOG59 indicated that use of bolus cisplatin with 5-FU led to nine cases of grade 3 leukopenia, one case of grade 3 anemia, one case of grade 4 thrombocytopenia, and no cases of grade 5 hematologic toxicity.64 In similar studies at the Istituto Nazionale Tumori in Milan60 and M.  D. Anderson,69 no cases of grade 3 or more severe hematologic toxicity were reported, regardless of whether the cisplatin was given as a bolus60 or an infusion.69 Induction chemotherapy followed by EBRT has been evaluated for toxicity and effectiveness in anal cancer. In one study conducted in Sweden,70 induction chemotherapy consisting of carboplatin and 5-FU, followed by radiation, produced complete responses in 29 of 31 patients.70 At 5 years, the overall survival rate in this study was 67%, the local control rate was 80%, and the sphincter preservation rate was 69%. The Cancer and Leukemia Group B (CALGB) reported results from a phase II trial of induction chemotherapy for patients with T3 to T4 or bulky N2 to N3 anal cancer.71 Patients were treated with two cycles of infusional 5-FU and cisplatin followed by radiation ­therapy with concurrent 5-FU and mitomycin C, with

614

Irradiation (Dose/Fractions/ Time)

Primary Tumor

Control

15 mg/m2 IVB D1

30 Gy/15/ D1-21

31/34 (91%) (≤5 cm)

1000 mg/m2/24 hr IVI D2-5, D28-31

10 mg/m2 IVB D1

40.8 Gy/24/ D1-35

Cummings et al, 199146

1000 mg/m2/24 hr IVI D1-4, D43-46

10 mg/m2 IVB D1 and D43

Cummings et al, 19845

1000 mg/m2/24 hr IVI D1-4, D43-46

Schneider et al, 199258

Serious Complications†

5-Year Survival Rate

7/10 (70%) (>5 cm)

NS

80% (crude)

22/26 (85%) (<3 cm)

32/50 (85%) (≥3 cm)

2/79 (3%)

73% (3-year actuarial)

48 Gy/24/D1-58 (split course)

15/17 (88%) (≤5 cm)

13/16 (81%) (>5 cm or T4)

1/33 (3%)

65% (actuarial)

10 mg/m2 IVB D1 and D43

50 Gy/20/D1-56 (split course)

10/10 (≤5 cm)

3/4 (>5 cm or T4)

5/14 (36%)

65% (actuarial)

1000 mg/m2/24 hr IVI D1-4, D29-32

10 mg/m2 IVB D1 and D29

50 Gy/25-28 D1-35 ± boost

21/22 (95%) (≤5 cm)

14/19 (74%) (>5 cm or T4)

3/41 (7%)

77% (actuarial)

Tanum et al, 199159

1000 mg/m2/24 hr IVI D1-4

10-15 mg/m2 IVB D1

50-54 Gy/25-27/ D1-35

28/30 (93%) (≤5 cm)

42/56 (75%) (>5 cm or T4)

14/89 (16%)

72% (actuarial)

Cummings et al, 19845

1000 mg/m2/24 hr IVI D1-4

10 mg/m2 IVB D1

50 Gy/20/D1-28

3/3 (≤5 cm)

11/13 (85%) (>5 cm or T4)

10/16 (63%)

75% (actuarial)

Doci et al, 199660

750 mg/m2/24 hr IVI (120 hr) D1-5, D43-47, D85-89

15 mg/m2 IVB D1, D43, D85

54-60 Gy/30-33/ D1-53 (split course)

28/38 (74%) (≤ 5 cm)

9/17 (53%) (>5 cm)

2/56 (4%)

81% (actuarial)

Papillon et al, 198761

600 mg/m2/24 hr IVI (120 hr) D1-5

12 mg/m2 IVB D1

42 Gy/10/D1-19 DPF + interstitial boost 20 Gy D78

—

57/70 (81%) (≥4 cm)

<5%

NS

Study and Reference

5-FU*

Mitomycin C

Leichman et al, 198557

1000 mg/m2/24 hr IVI D1-4, D29-32

Sischy et al, 198954

*All

infusions were administered for 96 hours except where shown. complications required surgery or were grade 3 or greater. D, day; DPF, direct perineal field (all other series used pelvic fields tangential to perineum); IVB, intravenous bolus injection; IVI, continuous intravenous infusion; NS, not stated; T4, invading adjacent organs. From Cummings BJ. Anal canal. In Perez CW, Brady LW (eds): Principles and Practice of Radiation Oncology, 3rd ed. Philadelphia: Lippincott-Raven, 1998, pp 1511-1524.

†Serious

PART 6  Gastrointestinal Tract

TABLE 25-4.    Selected Studies of Mitomycin C and 5-Flourouracil for Anal Cancer

25  The Anal Region

615

TABLE 25-5.    Trials of 5-Flourouracil and Cisplatin with Radiation for Primary Anal Cancer Chemotherapy Study and ­Reference

Complete Response Radiation Therapy (Dose/Fractions/ Time)

Rates/Total Number of ­Patients (%)

5-FU

Cisplatin

Wagner et al, 199463

1000 mg/m2/24 hr IVI (96 hr) D1-4

25 mg/m2 IVB D2-5

42 Gy/10/D1-19 + interstitial boost D63-64

47/51 (92%)

Approximately 75% (5-year)

Martenson et al, 199664

1000 mg/m2/24 hr IVI (96 h) D43-46

75 mg/m2 IVB D1, D43

59.4 Gy/33/D1-59 (split after 36 Gy)

14/17 (82%)

Not stated

Rich et al, 199365

300 mg/m2/24 hr IVI 5 days (120 hr)/wk D1-42

4 mg/m2 IVI D1-42, 5 days/wk

45-54 Gy/25-30/ D1-42

20/21 (95%)

91% (2-year actuarial)

Brunet et al, 199066

1000 mg/m2/24 hr IVI (120 hr) D2-6; repeat cycles D22, D43

100 mg/m2 45 Gy/25/D64-69 + IVB D1, D22, D43 boost

17/19 (89%)

No cancer deaths, 10-40 months

Svensson et al, 199367

1000 mg/m2/24 hr IVI (120 hr) D1-5; repeat cycles D29, D57

Carboplatin 300350 mg/m2 IVB D1, D29, D57

66 Gy/33/D79-125

6/6 (100%)

No recurrences 8-21 months

50 mg/m2 IVB D1-2, D22-23

20 Gy/10/D8-21, D29-42 + boost

18/25 (72%)

87% (5-year ­actuarial)

Survival Rates

Concurrent Therapy

Induction Therapy

Alternating Therapy Roca et al, 199068

750 mg/m2/24 hr IVI (120 hr) D2-7, D23-28

D, day; 5-FU, 5-Fluorouracil; IVB, intravenous bolus injection; IVI, continuous intravenous infusion. From Cummings BJ. Anal canal. In Perez CW, Brady LW (eds): Principles and Practice of Radiation Oncology, 3rd ed. Philadelphia: ­Lippincott-Raven, 1998, pp 1511-1524.

selected ­patients also receiving a radiation boost with concurrent 5-FU and cisplatin. Among 45 patients, induction chemotherapy yielded 8 complete and 21 partial responses. At 4 years of follow-up, the overall and disease-free survival rates were 68% and 61%, respectively. A phase III Gastrointestinal Intergroup study (RTOG 98-11) compared radiation therapy with concurrent 5-FU and mitomycin C versus two courses of induction 5-FU and cisplatin followed by concurrent radiation with 5-FU and cisplatin.72 Radiation doses up to 59 Gy were allowed for patients with T3-4 or node-positive disease. A total of 682 patients were randomized and stratified based on sex, nodal status, and tumor diameter. Rates of acute nonhematologic toxicity and long-term toxicity were similar in both treatment groups. Severe (grade 3 or 4) acute hematologic toxicity was significantly worse in the mitomycin C group (61% versus 42%). No significant difference was seen between the two groups in terms of locoregional control, disease-free survival, or overall survival. However, the cumulative colostomy rate was lower in the mitomycin C group than in the cisplatin group (10% versus 19%). It remains unclear whether this difference in the rate of colostomy resulted from the difference in the concurrent chemotherapy regimens or the use of induction therapy in the cisplatin group.72 A randomized trial in the United Kingdom, the Anal Cancer Trial II (ACT II), compared 5-FU and cisplatin with 5-FU and mitomycin C.73 Patients were randomly assigned to undergo one of four treatments: (1) concurrent 5-FU and mitomycin C, (2) concurrent 5-FU and mitomycin C followed by adjuvant chemotherapy with 5-FU and ­cisplatin,

(3) concurrent 5-FU and cisplatin, or (4) concurrent 5-FU and cisplatin followed by adjuvant chemotherapy with 5FU and cisplatin. This trial has completed accrual, and results are awaited. The EORTC conducted a phase II study of radiation therapy with concurrent cisplatin and mitomycin C for 21 patients with anal carcinoma.74 The treatment was well tolerated, and 19 patients (90%) experienced a complete response (the other 2 patients experienced a partial response). The EORTC is conducting a randomized phase II/III trial (EORTC 22011-40014) to compare concurrent 5-FU and mitomycin C with concurrent cisplatin and mitomycin C.

Treatment of Patients with Human Immunodeficiency Virus Infection Tolerance of therapy is a significant consideration among patients with HIV and invasive anal cancer. Excessive mucosal reactions from radiation and significant cytopenia have been reported during treatment, especially when the CD4 count is less than 200 cells/μL.75-77 Most studies of HIV-positive patients treated with chemoradiation therapy show rates of tumor response and locoregional control that are comparable to those among HIV-negative patients.75-81 However, toxicity rates are higher and treatment-compliance rates lower among HIV-positive patients than HIVnegative patients, which can affect outcomes.78,82,83 Overall survival rates for patients with HIV and anal cancer seem to be lower than those for HIV-negative patients with anal cancer, but this difference probably reflects underlying diseases and comorbid conditions.40,78

616

PART 6  Gastrointestinal Tract Initial anterior field

Posterior field

Anterior field after 30.6 Gy

FIGURE 25-3.  These treatment portals were used for high-dose radiation therapy in the Radiation Therapy Oncology Group trials. (Modified from Martenson JA, Lipsitz SR, Wagner H, et al. Initial results of a phase II trial of high dose radiation therapy, 5-fluorouracil and cisplatin for patients with anal cancer (E4292): an Eastern Cooperative Oncology Group study. Int J Radiat Oncol Biol Phys 1996;35:745-749.)

Summary of the Approach to Anal Canal Tumors The toxicity of therapy can be profound in anal cancer, but the consequence of local failure (or colostomy) is also profound. Local control of anal canal tumors is necessary for three reasons: (1) to eliminate the life-threatening complications of the disease, (2) to decrease the risk of distant metastases, and (3) to preserve sphincter function.84 Although radiation is effective in achieving local tumor control, the radiation-sensitizing effects of chemotherapy can substantially improve local control rates and reduce the need for aggressive surgical procedures. Although radiation administered with concurrent 5-FU and mitomycin C represents the current standard of care, questions remain regarding the optimal chemotherapy agents and schedule and the radiation dose and schedule for treating anal cancer.

Treatments For Tumors Of The Anal Margin The anal margin, considered by some to be anatomically distinct from the anal canal, involves the anal verge and includes the perianal tissues around the anal canal. Squamous cell carcinomas in this region are staged like other tumors of the skin and generally have a good prognosis.3 Most are well differentiated and keratinized. Visceral metastases are rare, and inguinal node metastases are uncommon. Disease management depends on tumor location and extent and patient factors. Anal margin tumors can be treated by local excision or radiation; abdominoperineal resection should be reserved for patients with fecal incontinence at diagnosis or for those with locally recurrent disease. Local excision is sufficient for T1 lesions, but pelvic and inguinal node radiation is indicated for T2 lesions. Chemotherapy should be included with radiation for T3 and T4 tumors.12,85

Radiation Therapy Techniques Radiation therapy approaches for anal cancer vary among centers, but some basic principles are common to all reported techniques. First, a dose-response relationship based on clonogenic burden has been demonstrated.31,86 Second, computed tomography (CT)–based treatment planning is

imperative, particularly because of the significant anatomic variation in the depth of the inguinal nodes. Larger tumor volumes should be treated with higher total doses of radiation. However, potential improvements in local control rates and disease-free and overall survival rates from increasing the dose of radiation31,86 must be balanced against the increased risk of toxic effects caused by the higher doses. Experience from M.  D. Anderson suggests that 30 Gy, given in conventional fractionation schedules, is sufficient to control microscopic disease.87,88 In our experience, patients with clinically uninvolved inguinal nodes given this dose are at minimal risk for inguinal failure.40,87,88 The following technique has been widely used for anal cancer and has been incorporated into RTOG protocols (Fig. 25-3). Patients are treated with anterior and posterior pelvic fields to a dose of 30.6 Gy, with the superior border of the field at the L5-S1 interspace, the inferior border at least 2.5 cm below the inferior extent of the anal canal, the lateral borders of the anterior field including the inguinal nodes, and the lateral borders of the posterior field 2 cm outside the pelvic brim. Subsequently, the superior border is placed at the bottom of the sacroiliac joints, and the reduced pelvic field is treated to 45 Gy by using anterior and posterior fields. Boost fields are used to administer an additional 10 to 14 Gy to the primary tumor, depending on the T classification. Electron fields are used to supplement the dose to the inguinal regions to 36 to 59 Gy depending on the N classification. M.  D. Anderson and other institutions have used an alternative radiation therapy technique (Fig. 25-4). Patients are treated with anterior and posterior fields to a dose of 30.6 Gy, with inferior and superior borders similar to those in the RTOG protocol described previously; however, the lateral borders of the anterior and posterior fields encompass the inguinal nodes. Subsequently, the superior border is placed at the bottom of the sacroiliac joints, and the reduced pelvic field is treated to 45 Gy by using a posterior and two lateral fields. Boost fields are used to administer an additional 5.4 to 14 Gy to the primary tumor, typically with a three-field technique. Clinically involved inguinal nodes receive a cumulative dose of 50.4 to 59 Gy, using supplemental electron fields that begins simultaneously with treatment of the reduced pelvic field. The electron energy is selected based on the depth of the grossly enlarged lymph nodes, and coverage is verified on the composite

25  The Anal Region

617

L5-S1 30.6 Gy

PA

+14.4GY GY Inguinal nodes Primary tumor Anal marker

+ 18-20 Gy with electrons for N+

B

A LAT

+14.4GY GY

PA + 18-20 Gy with electrons for N+

+10-14Gy

C

D

LAT

E FIGURE 25-4.  In this technique, used at M.  D. Anderson, the initial 30.6 Gy is administered through anteroposterior-posteroanterior portals (A). An additional 14.4 Gy is given to the pelvis, for a total dose of 45 Gy, in a three-field approach with posteroanterior (B) and two lateral fields (C). Clinically involved inguinal nodes simultaneously receive a boost of 18 to 20 Gy with electrons. A localized boost field is given an additional 10 to 14 Gy, for a total dose of 55 to 59 Gy, in a three-field approach with posteroanterior (D) and two lateral fields (E).

plan. This technique allows ­sparing of genitalia and small bowel in the reduced pelvic field. With this alternative technique, inguinal electron fields are necessary only if the inguinal regions are involved. At M.  D. Anderson, patients are placed in a frog-leg position to reduce skin toxicity.

Historically, the placement of the superior border of the radiation therapy field has varied, ranging from the bottom of the sacroiliac joints to the L5 to S1 interspace. An analysis of patterns of failure after chemoradiation therapy illustrated the importance of placing the superior border

618

PART 6  Gastrointestinal Tract

at the L5 to S1 interspace.40 Among 167 patients treated with radiation and concurrent chemotherapy, 5 had pelvic recurrences in the presacral and iliac regions, and all 5 of the recurrences occurred in patients for whom the superior border had been placed initially at the bottom of the sacroiliac joints. The rate of pelvic regional recurrence among patients with the superior border at the bottom of the sacroiliac joints was 7%, compared with 0% among those with the superior border at the L5 to S1 interspace.40 There is considerable interest in the use of IMRT for anal cancer.89-92 A dosimetric comparison of IMRT and conventional radiation-treatment plans showed that using IMRT resulted in a reduction of dose to the femoral head and neck and to the external genitalia, with comparable coverage of tumor volume.90 In the largest clinical study reported, 53 patients were treated with IMRT and concurrent chemotherapy.93 The primary tumor and involved nodes were treated with a median dose of 51.5 Gy, and regional nodes were treated with a median dose of 45 Gy. The rate of grade 3 acute gastrointestinal toxicity was 15%, and the rate of grade 3 acute dermatologic toxicity was 38%. Ninety-three patients experienced a complete response; at 18 months, the rates of ­colostomyfree survival and overall survival were 93% and 84%, respectively.93 The RTOG (rtog. org) recently completed a phase II trial of IMRT with concurrent 5-FU and mitomycin C. In that study, the primary tumor was treated with a dose of 50.4 Gy in 28 ­fractions for T2 N0 anal cancer and 54 Gy in 30 fractions for T3 to T4 or node-positive anal cancer. Involved nodes 3 cm or smaller in diameter were treated with 50.4 Gy, and larger involved were treated with 54 Gy. The elective nodal regions were treated with 42 Gy in 28 fractions for T2 N0 disease or 45 Gy in 30 fractions for T3 to T4 or node-­positive disease. Specific dose constraints were prescribed for the small bowel, femoral heads, iliac crests, external genitalia, bladder, and large bowel. Results from this trial are pending. In the meantime, an RTOG consensus panel has formulated specific contouring guidelines for treating anal cancer, and a contouring atlas is available at the RTOG Web site (www.rtog.org). IMRT has been used to treat anal cancer at M.  D. Anderson since 2005. Patients are first immobilized with a customized immobilization device. The primary tumor, involved nodes, elective nodal regions, and avoidance structures are carefully contoured on CT images. The primary tumor and any enlarged lymph nodes are treated to 50 to 59 Gy, depending on the size of the tumor and lymph nodes. The elective nodal regions, specifically the internal iliac, external iliac, presacral, perirectal, and inguinal regions, are treated to 43 to 46 Gy, given in 25 to 28 fractions. The femoral heads, genitalia, bladder, and small bowel are outlined as “avoidance structures” to minimize toxicity.

Balancing Efficacy with Toxicity: Dose-Time Considerations The timing of radiation delivery, in addition to total dose, significantly affects the response to therapy for anal cancer. Higher radiation doses given in an uninterrupted course are considered necessary to improve local control rates, but concern exists about treatment-related toxic effects.5,46,94 Results of sequential studies are shown in Table 25-4.

In early RTOG protocols involving the use of 5-FU, mitomycin C, and a 59.4-Gy radiation dose, instituting a planned 2-week treatment interruption after 36 Gy of radiation had been administered produced toxic effects at levels similar to those of previous protocols but led to unacceptably high rates (23%) of local failure and colostomy.94 When the planned 2-week interruption was eliminated, the local failure rate dropped to 11%, but the appearance of toxic reactions mandated treatment interruptions (median duration, 11 days) for one half of the patients treated in this manner. Investigators at the Princess Margaret Hospital in Toronto reported a similar experience. Early protocols in which radiation (50 Gy in 20 fractions without interruption) was given with 5-FU and mitomycin C produced good rates of local control but also produced high rates of acute and late toxicity.5 Reduction of the dose per fraction to 2 Gy and splitting the course of therapy significantly decreased treatment-related toxic effects without appearing to affect local control rates.46 An attempt to minimize hematologic complications by omitting mitomycin C from the chemotherapy was successful in that regard but compromised local control rates.46 Another study of time-dose considerations86 verified that local control rates were improved when radiation doses were 54 Gy or more and when overall treatment time was less than 40 days. Multivariate analyses in that study showed that radiation dose and hemoglobin level independently influenced local control; that radiation dose influenced disease-free survival; and that radiation dose, hemoglobin level, T category, and HIV status all influenced overall survival.86 Outcomes therefore depend on optimizing the radiation dose and chemotherapy regimen and minimizing the treatment-related toxicity.

Toxicity in the Elderly Special concerns exist about treatment-related toxic e­ ffects among the elderly. In a Swiss study95 of patients 75 years old or older, 21 were given radiation therapy, and 21 were given radiation therapy with chemotherapy. The ­total dose of radiation to the pelvis equaled 39.6 Gy; after a median treatment interruption of 43 days, a boost dose (median, 20 Gy) was given by brachytherapy or perineal EBRT. Chemotherapy consisted of mitomycin C (9.5 mg/m2) and a 96-hour infusion of 5-FU (600 mg/m2/day). Of the 40 patients able to complete therapy (95%), acute toxic effects led to reduction in the total radiation dose administered to two patients (one in the radiation-only group and one in the chemoradiation group) and led to interruptions in treatment for 11 patients (four in the radiation-only group and seven in the chemoradiation group). Grade 3 toxic effects were observed in 54% of patients (32% in the radiation-only group and 68% in the chemoradiation group). Among the chemoradiation group, grade 2 or 3 effects were leukopenia (25% of patients) and fatigue (58% of patients). At 5 years, the overall survival rate was 54% (49% for the ­radiation group and 59% for the chemoradiation group), and the actuarial local control rate was 79% (73% for the radiation group and 83% for the chemoradiation group). The authors of this study emphasized that particular attention should be paid to acute reactions to avoid treatment interruptions.95

25  The Anal Region

External Beam Radiation Therapy with an Interstitial Implant Other attempts to reduce treatment-related morbidity have included the use of lower-dose EBRT with interstitial iridium 192 (192I) implants. One report of such an approach96 involved giving 40 Gy by EBRT plus 20 Gy by means of an implant to 69 patients; 45 of these patients were also given 5-FU and mitomycin C. This approach resulted in a complete response rate of 81% despite relatively low actuarial local control rates at 2 years (65%) and 5 years (59%).96 However, the colostomy-free survival rates were 61% at 2 years, 47% at 5 years, and 37% at 10 years. EBRT administered with 5-FU and cisplatin and an interstitial implant has been effective in treating anal cancer. In a long-term study conducted in France,97 this strategy resulted in colostomy-free survival rates of 71% at 5 years and 67% at 8 years; the corresponding overall response rates were 84% and 77%. Among the 78 patients whose sphincters were preserved surgically, sphincter function was judged excellent or good in 72 (92%).97 Careful assessment of pararectal lymph node involvement at staging is imperative; the 5-year overall survival rates in the French study were 76% among node-positive patients and 88% among node-negative patients.97 Results from a German study suggest that the success rate of interstitial implants can be further improved by using sonography for three-dimensional tumor reconstruction and brachytherapy simulation.98 Results of a retrospective review at the Beatson Oncology Centre in Glasgow, Scotland, show that EBRT used with interstitial implants but without chemotherapy can achieve excellent local control of small tumors (100% for T1 tumors and 90% for T2 tumors).99 In that study, the dose of EBRT was 40 to 50 Gy, given at 2 Gy per fraction, and about 25 Gy was administered through the implant.99 Citing an increased risk of acute toxicity and a 1% to 2% incidence of chemotherapy-related death from the use of 5-FU and mitomycin C, Mitchell and colleagues at the ­University of Florida advocate the use of radiation alone for T1 and T2 tumors.100

Medical Management Of Treatment-related Toxic Effects Expert medical management and an aggressive approach are needed to control the side effects of chemoradiation therapy for anal cancer. Adverse effects that can be prevented and treated include those on the skin (i.e., desquamation), blood cell counts, fluid and electrolyte levels, pain, nutritional status, and fatigue. Supportive care measures can help prevent treatment interruption and thereby improve therapeutic outcome; however, aggressive supportive care is required during and after the completion of therapy regardless of whether therapy is given in a continuous or interrupted protocol. With attentive medical management, less than 5% of patients at M.  D. Anderson needed treatment interruption because of adverse effects.101 At the first signs of dry desquamation, which usually occur during the end of the second week of radiation, emollient ointments such as Aquaphor (Beiersdorf AG) should be used in the

619

i­nguinal region, and barrier creams (e.g., Lantiseptic’s Skin ­Protectant, Marietta, GA) should be used in the perianal region. The barrier cream is especially important for preventing secondary infection from fecal bacteria. Moist desquamation often occurs in the intergluteal region and along the inguinal folds. Products that act as a skin, such as the hydrogel sheet dressing from Vigilon (Bard, Murray Hill, NJ), can be placed over the area to prevent secondary infection and trauma and to provide a soothing, protective dressing. Secondary infections, especially those from Candida, can occur and must be treated. Difficulties with hygiene can be addressed by using sitz baths with Domeboro (Bayer Corporation, Pittsburgh, PA), an aluminum acetate solution; such products should also be used to clean and soak the inguinal region. Pain during urination from the contact of the acidic urine with the skin can be minimized by using barrier creams and spraying warm water over the perineum during urination. Significant pain from perianal and inguinal skin reactions can manifest during urination, defecation, and movement. Opioid analgesics can be helpful in controlling pain and, through their constipating effect, the frequent stooling often experienced as a result of pelvic radiation and chemotherapy. Bowel management programs have been established to address frequent stooling and pain associated with treatment for colorectal or anal cancer. The three-step program used at M.  D. Anderson involves the use of antidiarrheal agents such as diphenoxylate with atropine (Lomotil) or loperamide hydrochloride (Imodium) and opioid analgesics. The goal of this program (described in Chapter 24) is to maintain the number of stools produced at no more than three per day to avoid fluid and electrolyte imbalances and to reduce trauma and possible infection in the perianal region.102,103 Initial symptoms, which usually occur in the third week of therapy, are treated with antidiarrheal agents as needed. When that need reaches three times per day, the agents are then given according to a schedule (one to two tablets three or four times per day). If symptoms persist, opioid analgesics are given; a long-acting preparation is preferred for the convenience of the patient and for avoiding episodic diarrhea. Pain-management principles dictate that the analgesics be titrated to effect and that short-acting analgesics be made available until the appropriate dose of long-acting analgesic is identified. Blood cell counts and perhaps cytokine levels should be closely monitored, especially if mitomycin C is given, to avoid the need to interrupt treatment because of toxic effects. Administration of cisplatin also warrants close monitoring of magnesium levels; in the event of magnesium wasting, supplemental magnesium is required. The profound fatigue occasionally experienced during chemoradiation therapy can contribute to poor oral intake and can lead to nutritional and electrolytic imbalance. Fatigue can improve with correction of anemia, fluid, and nutritional status. Close follow-up is necessary after treatment until reepithelialization of the anal region is complete. During this time, symptoms, evidence of infection, blood cell counts, and fluid and electrolyte status must be evaluated and treated as necessary. Response to therapy can be monitored during this time, but biopsy of a residual abnormality is discouraged when tumor regression remains evident. Patients

620

PART 6  Gastrointestinal Tract

TABLE 25-6.    Incidence of Late Toxic Effects in 34 Patients with Anal Cancer after Chemoradiation Therapy Site of Toxic Effect*

Grade 0

Grade 1

Grade 2

Grade 3

Grade 4

Small or large bowel

17 (50%)

  9 (26%)

  7 (20%)

1 (3%)

0

Skin

24 (70%)

  4 (12%)

  3 (9%)

1 (3%)

Anal canal

18 (53%)

  5 (15%)

10 (29%)

1 (3%)

Worst cytotoxicity (per patient)

10 (29%)

10 (29%)

10 (29%)

2 (6%)

2 (6%) 0 2 (6%)

From John M, Flam M, Palma N. Ten-year results of chemoradiation for anal cancer: focus on late morbidity. Int J Radiat Oncol Biol Phys 1996;34:65-69. *Late was defined as an effect occurring more than 90 days after treatment was begun; all incidents were included, even those that were transient and resolved spontaneously.

may have fatigue for several months after completion of therapy; treatable causes such as anemia and hypothyroidism should be sought, treated, or ruled out as appropriate. Long-term bowel management strategies, especially the scheduled use of antidiarrheal agents and fluid management, should be pursued to optimize bowel control and improve quality of life. Excellent functional outcomes can be achieved after definitive chemoradiation therapy, particularly among patients who had limited symptoms at presentation and are rendered free of disease. In a long-term study in Geneva,104 quality of life was assessed in terms of symptoms and functioning among 41 disease-free patients (35 women and 6 men, with a median age of 71 years) who had completed treatment for anal cancer at least 3 years earlier. The median follow-up time was 116 months. In terms of function (best score is 100), the global quality of life rating was 71, the role-function score was 85, and the body image score was 74. As for symptoms (best score is 0), the scores were 28 for persistent diarrhea and 66 for sexual dysfunction among men. Patients who presented with anal dysfunction resulting from extensive tumor involvement had poorer scores and poorer outcomes in general.104 Types of long-term morbidity related to the use of chemoradiation therapy for anal cancer in a study of 34 patients is illustrated in Table 25-6.105 Only four patients (12%) experienced treatment toxicity of grade 3 or 4105; no long-term sequelae were identified in 29%, and 58% experienced grade 1 or 2 effects. Authors of a study of anorectal dysfunction after pelvic radiation for cervical cancer attributed late proctitis to sensory and motor abnormalities caused by the radiation.106

Surgery as Salvage Treatment Abdominoperineal resection remains essential as a salvage treatment for recurrent anal cancer. About one half of patients with isolated disease recurrence can be rendered disease free after this procedure, as demonstrated in one study by a subset of 42 of 185 consecutive patients with anal cancer.107 Of these 42 patients, 27 had only local failure, and the other 15 had regional or distant disease in addition to local failure. Supportive care was given to 16 patients (38%), and curative resection was undertaken in 26 patients (local excision in 3 and abdominoperineal resection in 23). At a median follow-up of about 22 months, all 11 patients given supportive care had died, and 11 of the 26 patients who had undergone resection were free of

­ isease. The survival rate for the entire group after the d ­diagnosis of local failure was 28%, but that for the resected group was almost 45%.107 In another series,34 the outcome for 192 patients was evaluated in terms of primary tumor histology, primary therapy, and salvage surgery. Among the 143 patients with squamous cell cancer, an abdominoperineal resection was performed as the primary therapy in 21; at 5 years, the survival rate in this subgroup was 60%, and the recurrence rate was 23%.34 Among the 122 patients who underwent chemoradiation therapy as the definitive treatment, the 5-year survival and recurrence rates were 55% and 34%, respectively. Forty-two patients experienced recurrent disease after chemoradiation; 13 underwent salvage surgery, and 8 (62%) of them were alive at a mean follow-up of 32 months. Another group108 evaluated outcome after abdominoperineal resection as salvage treatment for residual disease (21 patients, 11 of whom had evidence of disease on biopsy within 6 months of completing chemoradiation therapy) or recurrent disease (10 patients, in all of whom recurrence appeared more than 6 months after completing therapy [mean, 15 months; range, 9 to 41 months]).108 No patient had evidence of nodal or distant metastases at diagnosis or at abdominoperineal resection, although lymph node metastases were found at pathologic evaluation in two cases. Of the 11 patients who lived, 9 were disease-free at a mean follow-up of 40 months. The overall survival rate at 3 years after salvage treatment was 58%. Overall survival rates at 3 years were 72% for patients with residual disease and 29% for those with recurrent disease; at 5 years, the corresponding survival rates for these groups were 60% and 0% (P = .06).108

Summary Significant advances have been made in the treatment of anal cancer. Anal sphincter function can be preserved in most patients by using chemoradiation therapy as definitive treatment. Close clinical follow-up is imperative for supportive care and assessment of response. Salvage surgery should not be pursued unless clinical evidence of disease progression is shown. Routine biopsies are discouraged, because positive findings on biopsy after chemoradiation therapy in the absence of clinical evidence of disease progression are not predictive of persistent disease. A dose-response relationship has been observed for radiation and chemotherapy. However, treatment-related toxic

25  The Anal Region

effects can be significant, and toxic effects that mandate interruptions in treatment can compromise local control. Aggressive supportive care and clinical strategies that further reduce treatment-related toxic effects are necessary for further therapeutic advances. Strategies to minimize toxic effects and improve rates of local control are being investigated. They include the development of therapeutic approaches that are specific for tumor stage and that optimize the effectiveness of chemoradiation therapy.

REFERENCES    1. Cummings BJ. Anal canal. In Perez CW, Brady LW, eds: Principles and Practice of Radiation Oncology. 3rd ed, Philadelphia: Lippincott-Raven, 1998, pp 1511-1524.    2. Greene FL, Page DL, Fleming ID, et al, eds: AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 125-130.    3. Chawla AK, Willett CG. Squamous cell carcinoma of the anal canal and anal margin. Hematol Oncol Clin North Am 2001;15:321-344.    4. Boman BM, Moertel CG, O’Connell MJ, et al. Carcinoma of the anal canal. A clinical and pathological study of 188 cases. Cancer 1984;54:114-125.    5. Cummings BJ, Keane TJ, Thomas GM, et al. Results and toxicity of the treatment of anal canal carcinoma by radiation therapy or radiation therapy and chemotherapy. Cancer 1984;54:20622068.    6. Stearns MW Jr, Urmacher C, Sternberg SS, et al. Cancer of the anal canal. Curr Probl Cancer 1980;4:1-44.    7. Frost DB, Richards PC, Montague ED, et al. Epidermoid cancer of the anorectum. Cancer 1984;53:1285-1293.    8. Wang CJ, Chin YY, Leung SW, et al. Topographic distribution of inguinal lymph nodes metastasis: significance in determination of treatment margin for elective inguinal lymph nodes ­irradiation of low pelvic tumors. Int J Radiat Oncol Biol Phys 1996;35: 133-136.    9. Fajardo LF, Berthrong M, Anderson RE. Large intestine and anal canal. In Radiation Pathology. New York: Oxford University Press, 2001, pp 239-247.   10. Touboul E, Schlienger M, Buffat L, et al. Epidermoid carcinoma of the anal canal. Results of curative intent radiation therapy in a series of 270 patients. Cancer 1994;73:1569-1579.   11. Martenson JA Jr, Gunderson LL. External radiation therapy without chemotherapy in the management of anal cancer. Cancer 1993;71:1736-1740.   12. Peiffert D, Bey P, Pernot M, et al. Conservative treatment by irradiation of epidermoid carcinomas of the anal margin. Int J Radiat Oncol Biol Phys 1997;39:57-66.   13. Frisch M, Glimelius B, van den Brule AJ, et al. Sexually transmitted infection as a cause of anal cancer. N Engl J Med 1997;337:1350-1358.   14. Palefsky JM, Holly EA, Gonzales J, et al. Detection of human papillomavirus DNA in anal intraepithelial neoplasia and anal cancer. Cancer Res 1991;51:1014-1019.   15. Zaki SR, Judd R, Coffield LM, et al. Human papillomavirus infection and anal carcinoma. Retrospective analysis by in situ hybridization and the polymerase chain reaction. Am J Pathol 1992;140:1345-1355.   16. Daling JR, Weiss NS, Hislop TG, et al. Sexual practices, sexually transmitted diseases, and the incidence of anal cancer. N Engl J Med 1987;317:973-977.   17. Daling JR, Weiss NS, Klopfenstein LL, et al. Correlates of homosexual behavior and the incidence of anal cancer. JAMA 1982;247:1988-1990.   18. Frisch M, Melbye M, Moller H. Trends in incidence of anal cancer in Denmark. Br Med J 1993;306:419-422.   19. Palefsky JM, Holly EA, Ralston ML, et al. Anal squamous intraepithelial lesions in HIV-positive and HIV-negative homosexual and bisexual men: prevalence and risk factors. J Acquir Immune Defic Syndr Hum Retrovirol 1998;17:320-326.

621

  20. Goldie SJ, Kuntz KM, Weinstein MC, et al. Cost-effectiveness of screening for anal squamous intraepithelial lesions and anal cancer in human immunodeficiency virus-negative homosexual and bisexual men. Am J Med 2000;108:634-641.   21. Ryan DP, Compton CC, Mayer RJ. Carcinoma of the anal canal. N Engl J Med 2000;342:792-800.   22. Penn I. Cancers of the anogenital region in renal transplant recipients. Analysis of 65 cases. Cancer 1986;58:611-616.   23. Arends MJ, Benton EC, McLaren KM, et al. Renal allograft recipients with high susceptibility to cutaneous malignancy have an increased prevalence of human papillomavirus DNA in skin tumours and a greater risk of anogenital malignancy. Br J Cancer 1997;75:722-728.   24. Biggar RJ, Horm J, Goedert JJ, et al. Cancer in a group at risk of acquired immunodeficiency syndrome (AIDS) through 1984. Am J Epidemiol 1987;126:578-586.   25. Daling JR, Sherman KJ, Hislop TG, et al. Cigarette smoking and the risk of anogenital cancer. Am J Epidemiol 1992;135: 180-189.   26. Rabkin CS, Biggar RJ, Melbye M, et al. Second primary cancers following anal and cervical carcinoma: evidence of shared etiologic factors. Am J Epidemiol 1992;136:54-58.   27. Winkelstein W Jr. Smoking and cervical cancer—current status: a review. Am J Epidemiol 1990;131:945-957.   28. Frisch M, Johansen C. Anal carcinoma in inflammatory bowel disease. Br J Cancer 2000;83:89-90.   29. Frisch M, Olsen JH, Bautz, et al. Benign anal lesions and the risk of anal cancer. N Engl J Med 1994;331:300-302.   30. Godlewski G, Prudhomme M. Embryology and anatomy of the anorectum. Basis of surgery. Surg Clin North Am 2000;80: 319-343.   31. Hughes LL, Rich TA, Delclos L, et al. Radiotherapy for anal cancer: experience from 1979-1987. Int J Radiat Oncol Biol Phys 1989;17:1153-1160.   32. Myerson RJ, Karnell LH, Menck HR. The National Cancer Data Base report on carcinoma of the anus. Cancer 1997;80: 805-815.   33. Papagikos M, Crane CH, Skibber J, et al. Chemoradiation for adenocarcinoma of the anus. Int J Radiat Oncol Biol Phys 2003;55:669-678.   34. Klas JV, Rothenberger DA, Wong WD, et al. Malignant tumors of the anal canal: the spectrum of disease, treatment, and outcomes. Cancer 1999;85:1686-1693.   35. Joon DL, Chao MWT, Ngan SYK, et al. Primary adenocarcinoma of the anus: a retrospective analysis. Int J Radiat Oncol Biol Phys 1999;45:1199-1205.   36. Faivre C, Rougier P, Ducreux M, et al. 5-Fluorouracile and cisplatinum combination chemotherapy for metastatic squamouscell anal cancer [in French]. Bull Cancer 1999;86:861-865.   37. Touboul E, Schlienger M, Buffat L, et al. Conservative versus nonconservative treatment of epidermoid carcinoma of the anal canal for tumors longer than or equal to 5 centimeters. A retrospective comparison. Cancer 1995;75:786-793.   38. Zelnick RS, Haas PA, Ajlouni M, et al. Results of abdominoperineal resections for failures after combination chemotherapy and radiation therapy for anal canal cancers. Dis Colon Rectum 1992;35:574-577.   39. Grabenbauer GG, Matzel KE, Schneider IH, et al. Sphincter preservation with chemoradiation in anal carcinoma: abdomi­ noperineal resection in selected cases? Dis Colon Rectum 1998;41:441-450.   40. Das P, Bhatia S, Eng C, et al. Predictors and patterns of recurrence after definitive chemoradiation for anal cancer. Int J Radiat Oncol Biol Phys 2007;68:794-800.   41. Bonin SR, Pajak TF, Russell AH, et al. Overexpression of p53 protein and outcome of patients treated with chemoradiation for carcinoma of the anal canal: a report of randomized trial [Radiation Therapy Oncology Group] RTOG 87-04. Cancer 1999;85:1226-1233.   42. Wong CS, Tsao MS, Sharma V, et al. Prognostic role of p53 protein expression in epidermoid carcinoma of the anal canal. Int J Radiat Oncol Biol Phys 1999;45:309-314.

622

PART 6  Gastrointestinal Tract

  43. Nilsson PJ, Lenander C, Rubio C, et al. Prognostic significance of cyclin A in epidermoid anal cancer. Oncol Rep 2006;16: 443-449.   44. Holm R, Skovlund E, Skomedal H, et al. Reduced expression of p21WAF1 is an indicator of malignant behaviour in anal carcinomas. Histopathology 2001;39:43-49.   45. Salmon RJ, Fenton J, Asselain B, et al. Treatment of epidermoid anal canal cancer. Am J Surg 1984;147:43-48.   46. Cummings BJ, Keane TJ, O’Sullivan B, et al. Epidermoid anal cancer: treatment by radiation alone or by radiation and 5-fluorouracil with and without mitomycin-C. Int J Radiat Oncol Biol Phys 1991;21:1115-1125.   47. Nigro N, Vaitkevicius V, Considine B. Combined therapy for cancer of the anal canal: a preliminary report. Dis Colon Rectum 1974;15:354-356.   48. Epidermoid anal cancer. results from the [UK Co-ordinating Committee on Cancer Research] UKCCCR [Anal Cancer Trial Working Party] randomized trial of radiotherapy alone versus radiotherapy, 5-fluorouracil, and mitomycin. Lancet 1996;348:1049-1054.   49. Roelofsen F, Bosset JF, Eschwege F, et al. Concomitant radiotherapy and chemotherapy are superior to radiotherapy alone in the treatment of locally advanced anal cancer: results of a phase III randomized trial of the EORTC Radiotherapy and Gastrointestinal Cooperative Group [abstract]. Proc Am Soc Clin Oncol 1995;14:194.   50. Bartelink H, Roelofsen F, Eschwege F, 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;15:2040-2049.   51. Flam M, John M, Pajak TF, et al. Role of mitomycin in combination with fluorouracil and radiotherapy and of salvage chemoradiation in the definitive non-surgical treatment of epidermoid carcinoma of the anal canal: results of a phase III randomized intergroup study. J Clin Oncol 1996;14:2527-2539.   52. Das P, Crane CH, Ajani JA. Current treatment for localized anal carcinoma. Curr Opin Oncol 2007;19:396-400.   53. Martenson JA, Lipsitz SR, Lefkopoulou M, et al. Results of combined modality therapy for patients with anal cancer (E7283): an Eastern Cooperative Oncology Group study. Cancer 1995;76:1731-1736.   54. Sischy B, Doggett RL, Krall JM, et al. Definitive irradiation and chemotherapy for radiosensitization in management of anal carcinoma: interim report on Radiation Therapy Oncology Group study no. 8314. J Natl Cancer Inst 1989;81:850-856.   55. Miller EJ, Quan SH, Thaler T. Treatment of squamous cell carcinoma of the anal canal. Cancer 1991;67:2038-2041.   56. Gerard JP, Morignat E, Romestaing P, et al. Radiotherapy of cancer of the anal canal. 15 years experience at the Lyon civil hospices and brief review of the literature. Ann Chir 1998;52: 17-23.   57. Leichman L, Nigro N, Vaitkevicius VK, et al. Cancer of the anal canal. Model for preoperative adjuvant combined modality therapy. Am J Med 1985;78:211-215.   58. Schneider IHF, Grabenbauer GG, Reck T, et al. Combined radiation and chemotherapy for epidermoid carcinoma of the anal canal. Int J Colorectal Dis 1992;7:192-196.   59. Tanum G, Tveit K, Karlsen KO, et al. Chemoradiotherapy and radiation therapy for anal carcinoma. Survival and late morbidity. Cancer 1991;67:2462-2466.   60. Doci R, Zucali R, LaMonica G, et al. Primary chemoradiation therapy with fluorouracil and cisplatin for cancer of the anus. Results in 35 consecutive patients. J Clin Oncol 1996;14: 3121-3125.   61. Papillon J, Montbarbon JF. Epidermoid carcinoma of the anal canal. A series of 276 cases. Dis Colon Rectum 1987;30:324-333.   62. Gerard JP. Chemotherapy in the treatment of cancer of the anal canal. Cancer Radiother 1998;2:708-712.   63. Wagner JP, Mahe MA, Romestaing P, et al. Radiation therapy in the conservative treatment of carcinoma of the anal canal. Int J Radiat Oncol Biol Phys 1994;29:17-23.

  64. Martenson JA, Lipsitz SR, Wagner H, et al. Initial results of a phase II trial of high dose radiation therapy, 5-fluorouracil and cisplatin for patients with anal cancer (E4292): an Eastern Cooperative Oncology Group study. Int J Radiat Oncol Biol Phys 1996;35:745-749.   65. Rich T, Ajani JA, Morrison WH, et al. Chemoradiation therapy for anal cancer. Radiation plus continuous infusion of 5-­fluorouracil with or without cisplatin. Radiother Oncol 1993;27:209-215.   66. Brunet R, Becouarn Y, Pigneux J, et al. Cisplatine et fluorouracile en chimiotherapie neoadjuvante des carcinomas epidemoides du canal anal [in French]. Lyon Chir 1990;87:77.   67. Svensson C, Goldman S, Friberg B. Radiation treatment of epidermoid cancer of the anus. Int J Radiat Oncol Biol Phys 1993;27:67-73.   68. Roca E, De Simone G, Barugel M, et al. A phase II study of alternating chemoradiotherapy including cisplatin in anal canal carcinoma [abstract]. Proc Am Soc Clin Oncol 1990;9:128.   69. Paleologo F, Goswitz M, Rich TA, et al. Results of radiation and concomitant continuous infusion 5-fluorouracil and cisplatin for anal cancer [abstract 135]. Int J Radiat Oncol Biol Phys 1997;39(Suppl):202.   70. Svensson C, Goldman S, Friberg B, et al. Induction chemotherapy and radiotherapy in loco-regionally advanced epidermoid carcinoma of the anal canal. Int J Radiat Oncol Biol Phys 1998;41:863-867.   71. Meropol NJ, Niedzwiecki D, Shank B, et al. Induction therapy for poor-prognosis anal canal carcinoma: a phase II study of the cancer and Leukemia Group B (CALGB 9281). J Clin Oncol 2008;26:3229-3234.   72. Ajani JA, Winter KA, Gunderson LL, et al. Fluorouracil, mitomycin, and radiotherapy vs fluorouracil, cisplatin, and radiotherapy for carcinoma of the anal canal: a randomized controlled trial. JAMA 2008;299:1914-1921.   73. James R, Meadows H, WanACT S II. the second UK phase III anal cancer trial. Clin Oncol (R Coll Radiol) 2005;17:364-366.   74. Créhange G, Bosset M, Lorchel F, et al. Combining cisplatin and mitomycin with radiotherapy in anal carcinoma. Dis Colon Rectum 2007;50:43-49.   75. Hoffman R, Welton ML, Klencke B, et al. The significance of pretreatment CD4 count on the outcome and treatment tolerance of HIV-positive patients with anal cancer. Int J Radiat Oncol Biol Phys 1999;44:127-131.   76. Chadha M, Rosenblatt EA, Malamud S, et al. Squamous-cell carcinoma of the anus in HIV-positive patients. Dis Colon Rectum 1994;37:861-865.   77. Peddada AV, Smith DE, Rao AR, et al. Chemotherapy and lowdose radiotherapy in the treatment of HIV-infected patients with carcinoma of the anal canal. Int J Radiat Oncol Biol Phys 1997;37:1101-1105.   78. Edelman S, Johnstone PA. Combined modality therapy for HIVinfected patients with squamous cell carcinoma of the anus: outcomes and toxicities. Int J Radiat Oncol Biol Phys 2006;66: 206-211.   79. Cleator S, Fife K, Nelson M, et al. Treatment of HIV-associated invasive anal cancer with combined chemoradiation. Eur J Cancer 2000;36:754-758.   80. Blazy A, Hennequin C, Gornet JM, et al. Anal carcinomas in HIV-positive patients: high-dose chemoradiotherapy is feasible in the era of highly active antiretroviral therapy. Dis Colon Rectum 2005;48:1176-1181.   81. Kim JH, Sarani B, Orkin BA, et al. HIV-positive patients with anal carcinoma have poorer treatment tolerance and outcome than HIV-negative patients. Dis Colon Rectum 2001;44: 1496-1502.   82. Oehler-Jänne C, Seifert B, Lütolf UM, et al. Local tumor control and toxicity in HIV-associated anal carcinoma treated with radiotherapy in the era of antiretroviral therapy. Radiat Oncol 2006;1:29.   83. Kauh J, Koshy M, Gunthel C, et al. Management of anal cancer in the HIV-positive population. Oncology (Williston Park) 2005;19:1634-1638. discussion 1638-1640, 1645 passim.   84. Cox JD. Chemoradiation for malignant epithelial tumors. ­Cancer Radiother 1998;2:7-11.

25  The Anal Region   85. Mendenhall WM, Zlotecki RA, Vauthey JN, et al. Squamous cell carcinoma of the anal margin. Oncology 1996;10:1843-1854.   86. Constantinou EC, Daly W, Fung CY, et al. Time-dose considerations in the treatment of anal cancer. Int J Radiat Oncol Biol Phys 1997;39:651-657.   87. Janjan NA, Crane CH, Ajani J, et al. 30 Gy is a sufficient dose to control micrometastases from anal cancer-dosimetric evaluation of N0 patients undergoing chemoradiation [abstract 11]. Cancer J 1999;5:117.   88. Hung AV, Ballo MT, Crane C, et al. Platinum-based combined modality therapy for anal carcinoma [abstract]. Proc Am Soc Clin Oncol 2001;19:A679.   89. Chen YJ, Liu A, Tsai PT, et al. Organ sparing by conformal avoidance intensity-modulated radiation therapy for anal cancer: dosimetric evaluation of coverage of pelvis and inguinal/femoral nodes. Int J Radiat Oncol Biol Phys 2005;63:274-281.   90. Milano MT, Jani AB, Farrey KJ, et al. Intensity-modulated radiation therapy (IMRT) in the treatment of anal cancer: ­ toxicity and clinical outcome. Int J Radiat Oncol Biol Phys 2005;63: 354-361.   91. Kachnic L, Tsai HK, Willins J, et al. Dose-painted intensity modulated radiation therapy for anal cancer: dosimetric comparison and acute toxicity [abstract 2126]. Int J Radiat Oncol Biol Phys 2006;66:S280.   92. Devisetty K, Mell LK, Salama JK, et al. Anal cancer treated with intensity modulated radiation therapy (IMRT) and concurrent chemotherapy: a dosimetric analysis of acute gastrointestinal toxicity [abstract 2138]. Int J Radiat Oncol Biol Phys 2006;66:S287.   93. Salama JK, Mell LK, Schomas DA, et al. Concurrent chemotherapy and intensity-modulated radiation therapy for anal canal cancer patients: a multicenter experience. J Clin Oncol 2007;25:4581-4586.   94. John M, Pajak TJ, Flam MS, et al. Dose escalation in chemoradiation for anal cancer. Preliminary results of RTOG 92-08. Cancer J Sci Am 1996;2:205-211.   95. Allal AS, Obradovic M, Laurencet F, et al. Treatment of anal carcinoma in the elderly: feasibility and outcome of radical radiotherapy with or without concomitant chemotherapy. Cancer 1999;85:26-31.   96. Berger C, Felix-Faure C, Chauvet B, et al. Conservative treatment of anal canal carcinoma with external radiotherapy and interstitial brachytherapy, with or without chemotherapy: longterm results. Cancer Radiother 1999;3:461-467.

623

  97. Gerard JP, Ayzac L, Hun D, et al. Treatment of anal canal carcinoma with high dose radiation therapy and concomitant fluorouracil-cisplatinum. Long term results in 95 patients. Radiother Oncol 1998;46:249-256.   98. Lohnert M, Doniec JM, Kovacs G, et al. New method of radiotherapy for anal cancer with three-dimensional tumor reconstruction based on endoanal ultrasound and ultrasound-guided afterloading therapy. Dis Colon Rectum 1998;41:169-176.   99. Sandhu AP, Symonds RP, Robertson AG, et al. Interstitial iridium-192 implantation combined with external radiotherapy in anal cancer: ten years experience. Int J Radiat Oncol Biol Phys 1998;40:575-581. 100. Mitchell SE, Mendenhall WM, Zlotecki RA, et al. Squamous cell carcinoma of the anal canal. Int J Radiat Oncol Biol Phys 2001;49:1007-1013. 101. Hung A, Crane C, Delclos M, et al. Cisplatin-based combined modality therapy for anal carcinoma: a wider therapeutic index. Cancer 2003;97:1195-1202. 102. Callister M, Janjan N, Brown T, et al. Effective management of treatment-related enteritis during preoperative chemoradiation for locally advanced rectal cancer [abstract 228]. Int J Radiat Oncol Biol Phys 1997;48(Suppl):225. 103. Throckmorton T, Janjan NA, Bisanz A, et al. Development of a bowel function assessment tool to measure bowel function in patients receiving radiation and/or chemotherapy [abstract 1040]. Int J Radiat Oncol Biol Phys 1997;39(Suppl):235. 104. Allal AS, Sprangers MA, Laurencet F, et al. Assessment of longterm quality of life in patients with anal carcinomas treated by radiotherapy with or without chemotherapy. Br J Cancer 1999;80:1588-1594. 105. John M, Flam M, Palma N. Ten-year results of chemoradiation for anal cancer: focus on late morbidity. Int J Radiat Oncol Biol Phys 1996;34:65-69. 106. Kim GE, Lim JJ, Park W, et al. Sensory and motor dysfunction assessed by anorectal manometry in uterine cervical carcinoma patients with radiation induced late rectal complications. Int J Radiat Oncol Biol Phys 1998;41:835-841. 107. Allal AS, Laurencet FM, Reymond MA, et al. Effectiveness of surgical salvage therapy for patients with locally uncontrolled anal carcinoma after sphincter-conserving treatment. Cancer 1999;86:405-409. 108. Pocard M, Tiret E, Nugent K, et al. Results of salvage abdominoperineal resection for anal cancer after radiotherapy. Dis ­Colon Rectum 1998;41:1488-1493.

The Urinary Bladder 26 Michael F. Milosevic and Mary K. Gospodarowicz ANATOMY OF THE NORMAL URINARY BLADDER EFFECTS OF RADIATION ON THE NORMAL URINARY BLADDER Early Radiation Effects Late Radiation Effects Influences of Patient, Tumor, and Treatment Factors Management of Radiation-Induced Bladder Side Effects BLADDER CANCER Epidemiology and Risk Factors Pathology

Presentation, Natural History, and Staging Management of Ta, T1, and Tis Bladder Cancer Management of Muscle-Invasive (T2-T4) Bladder Cancer Palliative Management of Bladder Cancer Radiation Treatment Planning and Delivery Complications of Radiation Therapy Assessment of Radiation Response and Management of Recurrence SUMMARY

Anatomy of the Normal Urinary Bladder The anatomy of the normal urinary bladder is described in detail elsewhere,1 and only the aspects relevant to the diagnosis and treatment of bladder cancer with radiation therapy are reviewed here. The normal empty bladder is a pyramidal organ that lies in the low pelvis behind the symphysis pubis. The base of the bladder is situated posteriorly in close relation to the vagina in women and the rectum in men. It is bounded superiorly by the ureteric orifices and inferiorly by the urethra. The apex is located opposite the base immediately behind the upper symphysis. The medial umbilical ligament and urachus extend from the bladder apex through the extraperitoneal fat to the umbilicus. The superior surface of the bladder is almost completely covered by peritoneum in men, and it is closely related to the sigmoid colon and loops of small bowel. In women, only the anterosuperior aspect of the bladder is covered by peritoneum, which is reflected at the uterus to form the uterovesical pouch. The uterus displaces the normal abdominal contents away from the bladder and helps to minimize the radiation dose to these organs and the risk of radiation complications during treatment of bladder cancer or other pelvic malignancies. Inferiorly, the bladder neck is contiguous with the prostate gland in men and the urogenital diaphragm in women. The bladder wall comprises three main layers: the mucosa, muscularis propria, and serosa. The mucosa covers the internal surface of the bladder and consists of the urothelium, which is several cell layers thick and is in direct contact with the urine. The lamina propria is a zone of loose connective tissue that lies immediately beneath the urothelium. It contains blood vessels, lymphatic channels, and small incomplete bundles of smooth muscle called the muscularis mucosa. The mucosa of the empty bladder is folded on itself except at the trigone, the inner surface of the bladder at the base, where it is smooth. The muscularis

propria is situated immediately beneath the lamina propria and is a layer of interlaced smooth muscle arranged in a complex mesh. Distinguishing muscularis propria from muscularis mucosa in a biopsy or cystectomy specimen is an important aspect of bladder cancer staging. The superior aspect of the bladder is covered by serosa. The normal bladder has a urine capacity of 150 to  300 mL. As the bladder fills, it rises out of the pelvis into the upper abdomen, and the shape becomes ovoid. The superior surface is the most mobile, whereas the bladder neck and base (trigone) remain relatively fixed. Accounting for this motion during fractionated radiation therapy for bladder cancer is important to ensure adequate tumor coverage throughout treatment. The lymphatic drainage of the bladder2 begins in the microscopic lymphatic channels of the lamina propria and muscularis propria. These channels terminate in collecting ducts that emerge from the bladder near the trigone and along the anterior and posterior walls. The collecting ducts travel superiorly and laterally and terminate in the medial external iliac, obturator, and other internal iliac lymph nodes situated along the pelvic sidewall medial to the acetabulum. Drainage also can occur directly through the anterior and lateral external iliac lymph nodes, common iliac nodes, and presacral lymph nodes. Small paravesical lymph nodes  often lie close to the bladder wall, and intercalating lymph nodes may be found along the course of the collecting ducts between the bladder and the pelvic sidewall.

Effects of Radiation on the Normal Urinary Bladder The urinary bladder, by virtue of its central location in the pelvis and its relative radiosensitivity, strongly influences the radiation dose that can safely be used to treat a ­variety of pelvic malignancies. Irradiation of tumors arising

627

628

PART 7  Urinary Tract

in the bladder itself or in adjacent pelvic organs, including the prostate, cervix, uterus, and rectum, can result in the ­ delivery of doses to the normal bladder or portions of the bladder that are sufficient to produce side effects. Detailed understanding of the radiation response of the normal bladder as a function of total dose, fraction size, and irradiated bladder volume is essential when planning treatment. Information about the radiation response of the normal bladder has largely been derived from animal studies and from the treatment of patients with bladder, cervix, or prostate cancer. Although the patient data are more relevant to clinical practice, they are limited by the difficulty associated with accurately determining the dose to the bladder during a course of fractionated radiation therapy. The bladder dose can be easily and reliably calculated when the entire organ is irradiated uniformly. However, in many clinical scenarios, the dose to the bladder is heterogeneous. Careful correlations of normal tissue dose and volume with clinical outcome are now technically feasible and are essential for accurately defining the radiation tolerance of the bladder and other organs. Variations in bladder filling during a course of fractionated radiation therapy, differences in bladder contouring (solid organ versus bladder wall only), and underreporting of side effects are persistent problems that hinder comparison of results among treatment centers and the establishment of robust radiation dose constraints. The effects of ionizing radiation on the bladder can be classified broadly as early (occurring during or immediately after irradiation) or late (occurring 6 months to many years after irradiation). Both types of effects are described further in the following sections.

Early Radiation Effects The bladder is lined by urothelium, which is characterized by proliferating basal cells and more superficial differentiated cells. The acute effects of radiation on the normal bladder are thought to arise mainly from injury to the rapidly dividing basal cells.3-5 Cellular changes, including swelling, vacuolization of the cytoplasm, and alterations in the appearance of the plasma membrane and nucleus, may be seen within hours of irradiation. Epithelial desquamation, focal ulceration, and interstitial edema are common. Disruption of the tight epithelial cell junctions and loss of the polysaccharide mucosal lining compromises the normal barrier between the hypertonic urine and the isotonic bladder wall tissues.6 These changes correlate with symptoms of acute radiation cystitis, including dysuria, urinary frequency, nocturia, and microscopic or gross hematuria. Numerous studies of adjuvant pelvic radiation therapy for endometrial cancer and rectal cancer have described severe acute cystitis in fewer than 5% of patients after whole-pelvis doses of 45 to 50 Gy in 1.8- to 2-Gy daily fractions.7,8 Further details of the effects of partial bladder irradiation were later reported by Storey and colleagues,9 who described severe irritative voiding symptoms or bleeding in approximately 20% of patients who received 70 Gy or more to at least 30% of the bladder but in only 5% of patients who received 70 Gy or more to at least 15% of the bladder. Michalski and others10 reported a twofold increase in the risk of acute bladder effects when 65 Gy or more was administered to at least 30% of the bladder wall.

Late Radiation Effects Late radiation-induced bladder injury arises because of microvascular changes leading to ischemia and fibrosis.3-5   Irradiation produces microvascular dilation followed by endothelial cell swelling and vacuolization. Endothelial proliferation and thickening of the vessel media leading to luminal narrowing or complete occlusion may occur later. Dilated vessels are often visible on cystoscopy months to years after irradiation and can spontaneously rupture, leading to microscopic or gross hematuria. A cystoscopic image of a previously irradiated bladder wall demonstrating vascular dilation and telangiectasia is shown in Figure 26-1A. In extreme cases, a fibrotic, thickened, and relatively avascular bladder wall may result. Fistulas may arise between the bladder and adjacent organs. Chronic bladder contracture produces symptoms of urinary frequency, dysuria, and hematuria. Evidence is mounting from animal and clinical studies that the severity of acute radiation-induced bladder reactions correlates with the severity of late reactions8,11-14 and may be mediated through altered expression of genes that influence vascular and fibrotic responses. The interval between irradiation and the onset of late bladder complications is typically 2 to 3 years, although it can be substantially longer. Late bladder complications usually occur later than rectosigmoid complications.6 The dose-volume relationship for late radiation-induced bladder complications was reviewed by Emami and colleagues,15 with symptomatic bladder contracture and volume loss as the end point. The investigators estimated that receiving 65 Gy in 1.8- to 2-Gy fractions to the entire bladder resulted in a 5% risk of late toxicity and that a total dose of 80 Gy was tolerable if only two thirds of the bladder was irradiated. Marks and colleagues,6 in a separate thorough review of the clinical literature, estimated the whole-bladder  tolerance dose for late effects to be only 50 Gy. A dose of 67 Gy to less than 30% of the bladder volume also ­seemed to be tolerable, as was a dose of 75 Gy to less than 10% of the bladder. Later, the proportion of the bladder irradiated to more than 65 Gy was shown to correlate with the risk of late complications in patients who underwent conformal radiation therapy for prostate cancer.16 Most estimates of the radiation tolerance of the bladder have been derived from series of patients who received pelvic radiation therapy in daily fractions of 1.8 to 2.5 Gy. Fraction size is an important determinant of late radiation complications.17 Animal studies have suggested that the α/β ratio for late bladder toxicity may be in the range of 5 to 7 Gy,11,18,19 compared with 2 to 3 Gy for most normal tissues.17 This raises the possibility that the influence of fraction size on late bladder complications may be less than for other tissues, although clinical evidence to support this is lacking. Overall, the available data suggest that a radiation dose of 50 to 60 Gy, given in 1.8- to 2-Gy daily fractions to the entire bladder, is well tolerated by most  patients and that smaller bladder volumes can tolerate doses in the range of 65 to 70 Gy.

Influence of Patient, Tumor, and Treatment Factors Several patient-, tumor-, and treatment-related factors may increase the risk of early and late bladder-related side effects in patients receiving pelvic radiation therapy. ­ Preexisting

26  The Urinary Bladder

A

B

C

D

629

FIGURE 26-1.  Photographs obtained during cystoscopy show radiation cystitis in a patient who had received high-dose pelvic radiation therapy for cervical cancer several years earlier (A); carcinoma in situ (B); a low-grade papillary bladder tumor (C); and a muscle-invasive bladder tumor (D).

bladder symptoms arising from coexisting infection or long-standing urinary outflow obstruction may be aggravated by irradiation. Multiple prior transurethral resections can lead to bladder wall fibrosis, decreased bladder compliance, and irritative voiding symptoms that similarly may worsen with treatment.20 Systemic chemotherapy with high-dose cyclophosphamide can cause hemorrhagic cystitis and late bladder contracture21 and reduce the tolerance of the bladder to subsequent irradiation.22 There is no evidence that cisplatin administered before, during, or after pelvic radiation therapy increases the risk of late radiationinduced bladder complications.23,24

bladder or at other sites in the pelvis, which can be difficult to distinguish from late radiation-induced side effects. Cystoscopic coagulation of bleeding vessels may help ­alleviate hematuria caused by radiation-induced vascular dilation and fragility. Patients with persistent hematuria may benefit from treatment with hyperbaric oxygen.25-27 Late bladder symptoms will resolve within 2 to 3 years of onset in 70% to 80% of patients.14 In extreme cases, cystectomy may be necessary to alleviate severe, persistent, or disabling bladder symptoms. Whether early aggressive management of acute radiation-induced bladder side effects reduces the risk of severe late reactions remains unclear.

Management of Radiation-Induced Bladder Side Effects

Bladder Cancer

The management of radiation-induced bladder side effects is primarily aimed at controlling symptoms. Urinary tract infections should be treated. Bladder spasms may be ­relieved with topical anesthetics or anticholinergic agents. Acute irritative voiding symptoms usually resolve within  1 month after the completion of radiation therapy. Patients who present with voiding symptoms or hematuria months to years after completing pelvic radiation therapy should be examined carefully to exclude tumor recurrence in the

Malignant tumors of the urinary bladder are common and cause significant morbidity and mortality. An estimated 68,810 new cases will be diagnosed in the United States in 2008, and 14,100 people will die of the disease.28 The age-­adjusted incidence has been relatively constant over the past 20 years at 20 to 25 cases per 100,000 people. Men are affected three times as often as women. The incidence of bladder cancer is lower among

Epidemiology and Risk Factors

630

PART 7  Urinary Tract

blacks and people of ­ Hispanic, Japanese, or Indian descent than among whites.29 Most cases are diagnosed in people 50 to 80 years old. The strongest risk factor for the development of bladder cancer is a history of cigarette smoking, and the incidence of this disease generally parallels the incidence of lung cancer and other smoking-related illnesses. Between 50% and 80% of all cases of bladder cancer are attributable to smoking.30   Brennan and colleagues,31 in a meta-analysis of 11 casecontrol studies, found the relative risk of developing bladder cancer to increase from 2.0 after 20 years of smoking to 5.6 after 60 years. People who stopped smoking realized a 30% reduction in their risk of bladder cancer within 1 to 4 years and a 60% reduction by 25 years. However, the risk of bladder cancer never returned to the lower level of lifelong nonsmokers.31 Patients with bladder cancer who are current or former smokers may have other serious medical problems, such as cardiovascular or pulmonary disease, that can limit their ability to tolerate aggressive therapies, including surgery. Occupational exposure to agents such as aromatic amines and heavy metals accounts for about 20% of bladder cancer cases.32 Individuals working in the textile, dye, tire, or chemical industries are at the greatest risk.29,33 In one study, exposure to paints or dyes, herbicides, or other chemicals was associated with two to five times the risk of developing bladder cancer relative to the risk for unexposed control subjects.33 The latency period from ­exposure to the manifestation of occupation-related bladder cancer usually exceeds 15 years.32 Studies of occupational exposure are often difficult to interpret because of the confounding ­effect of cigarette smoking. Recurrent urinary tract infections increase the risk of squamous cell bladder cancer, especially in paraplegics who have permanent indwelling catheters.34-36 Urinary schistosomiasis is an important etiologic factor in areas of Africa and the Eastern Mediterranean. The infective agent, Schistosoma haematobium, is a waterborne fluke that deposits ova in the bladder wall. The ova incite a chronic inflammatory reaction that leads to neoplastic changes. Approximately 70% of schistosomiasis-associated  tumors are squamous cell carcinomas, and the remaining cases are urothelial carcinomas.30 Drugs, including longterm use of phenacetin and exposure to high-dose cyclophosphamide, are uncommon causes of bladder cancer. Several reports suggest that the risk of bladder cancer is increased after pelvic irradiation for prostate cancer or other tumors.37-40

Pathology The consensus classification of the World Health Organization (WHO) and International Society of Urological  Pathology (ISUP) for primary tumors of the urinary bladder is summarized in Box 26-1.41 Most tumors are of ­epithelial origin, and only 5% develop from the mesenchymal components of the bladder wall.30 Of the epithelial tumors, 90% are urothelial carcinomas that arise from the urothelium of the bladder mucosa. The remaining cases are squamous cell carcinomas, adenocarcinomas, neuroendocrine tumors, and mixed tumors. Adenocarcinomas can be of urachal or nonurachal origin. Urothelial carcinomas can be divided into two broad groups according to the presence or absence of muscularis propria invasion; these two types

BOX 26-1.  Classification of Primary Neoplasms of the Urinary Bladder Categories of Urothelial Neoplasms Hyperplasia Flat hyperplasia Papillary hyperplasia Flat Lesions with Atypia Reactive (inflammatory) atypia Atypia of unknown sifnificance Dysplasia (low-grade intraurothelial neoplasia) Carcinoma in situ (high-grade intraurothelial neoplasia)* Papillary Neoplasms Papilloma Inverted papilloma Papillary urothelial neoplasm of low malignant potential Papillary carcinoma, low grade Papillary carcinoma, high grade† Invasive Neoplasms Lamina propria invasion Muscularis propria (detrusor muscle) invasion *Includes cases with “severe dysplasia” †Option exists to add comment as to the presence of marked anaplasia Adapted from Epstein JI, Amin MB, Reuter VR, et al. The World Health Organization and International Society of Urological ­Pathology consensus classification of urothelial tumors of the ­bladder. Am J Surg Pathol 1998; 22:1435-1448.

of tumor have different clinical behaviors, prognoses, and forms of management.

Urothelial Neoplasms without Muscularis Propria Invasion Urothelial neoplasms without muscularis propria invasion account for about 75% of all bladder tumors. Carcinoma in situ (CIS), noninvasive papillary carcinomas, and tumors that invade only the lamina propria of the bladder wall are commonly called superficial bladder tumors, although the WHO/ISUP consensus statement advises against this practice because it ignores important biologic differences among these types of tumors.41 Cystoscopic photographs of bladder CIS and a low-grade papillary carcinoma are shown in Figure 26-1. CIS is a flat lesion of the urothelium characterized by cytologic and nuclear atypia; by definition, CIS is always classified as high grade.41 It can be present focally or diffusely in the bladder and can occur alone or in association with other noninvasive or invasive neoplasms. Its presence predicts a high rate of recurrence and the development of muscle-invasive disease. Cohesion among the atypical urothelial cells is reduced, which may lead to sloughing of these cells from the mucosa into the urine. Full-­thickness epithelial denudation in a biopsy specimen implies the ­existence of CIS. Contact between urine and the denuded mucosa leads to inflammation and a characteristic velvety red appearance at cystoscopy. The 1998 WHO/ISUP consensus statement introduced several changes to the classification of noninvasive papillary tumors to improve the uniformity of pathologic reporting.41 Papillary tumors were divided into five subgroups: papilloma, inverted papilloma, papillary ­urothelial

26  The Urinary Bladder

neoplasm of low malignant potential (PUNLMP), lowgrade papillary carcinoma, and high-grade papillary carcinoma. The ­ PUNLMP subgroup was added to reduce interobserver variability in the classification of papilloma versus low-grade carcinoma. Like papillomas, PUNLMPs are characterized by thickened urothelium, but they lack the cytologic atypia of low-grade papillary carcinomas, and the risk of recurrence after resection is intermediate between these two other subtypes.42 Most noninvasive papillary carcinomas are low grade, with only 3% to 18% being high grade.43 Tumors that invade the lamina propria can appear as papillary or sessile. Histologically, they are characterized by isolated or contiguous clusters of malignant cells that often are surrounded by a prominent inflammatory stromal reaction. Malignant cells can involve the muscularis mucosa, an incomplete thin layer of smooth muscle at the midlevel of the lamina propria that is discernable in about 50% of transurethral bladder resection specimens.41 Muscularis mucosa involvement has little prognostic significance, but it is important to distinguish it from muscularis propria involvement. Tumor invasion of the small vessels and lymphatic channels of the lamina propria is uncommon.

Urothelial Neoplasms with Muscularis Propria Invasion Only about 25% of bladder cancer cases involve the muscularis propria. In most cases, no history of bladder cancer is found, and the muscle-invasive tumor is the index lesion. Muscle-invasive cancers can be papillary but often appear sessile (see Fig. 26-1D). Most have high-grade features and appear as solitary lesions in the bladder, but multiple ­invasive tumors can be present, particularly in the setting of prior or coexisting superficial disease or CIS. The natural history of muscle-invasive bladder cancer is progressive extension through the muscular and serosal layers of the bladder wall to involve perivesical fat. Adjacent organs and tissues, including the prostate, uterus, vagina, rectum, and pelvic bones, may be invaded. Tumors arising in the region of the trigone may occlude the ureteric orifices, leading to hydronephrosis. Lymph node metastases and distant ­metastases often develop, making muscle-invasive bladder cancer a life-threatening illness.

Acquired Genetic and Molecular Changes An inherited genetic predisposition to bladder cancer is rare and accounts for only a very small percentage of such tumors.44 Nevertheless, bladder cancer is characterized by a variety of acquired genetic changes that are promoted at least in part by exposure to environmental carcinogens. The progression from normal bladder mucosa to intraepithelial malignancy to invasive bladder cancer requires a series of molecular changes that involve the activation of specific proto-oncogenes and the inactivation of tumor suppressor genes. These acquired genetic changes lead to alterations in important processes that determine the balance between cell proliferation and cell death, including signaling, cell cycle regulation, and apoptosis.45 Numeric and structural abnormalities of chromosomes 9, 11, and 17 are common in bladder cancer.29 Complete or partial deletions of chromosome 9 occur in a large percentage of tumors with or without invasion of the ­muscularis propria and are the most common cytogenetic ­abnormality.

631

Deletions involving the short arm of chromosome 9 lead to inactivation of the CDKN2A (previously designated p16) tumor suppressor gene, which is an important cell cycle regulator.46 Changes in the expression of other genes involved in apoptosis and DNA repair, ­including CDKN1A (formerly called p21), TP53 (formerly designated p53), RB1 (formerly called Rb and identified in retinoblastoma), BAX, and BCL2 (formerly designated bcl-2), have also been described. Several studies have associated altered TP53 and RB1 expression with clinical and histopathologic indicators of aggressive disease and an increased risk of progression from superficial to muscle-invasive cancer.47-52 Alterations in these genes may predict decreased survival among patients with muscle-invasive tumors after treatment with cystectomy53 or radiation therapy.54-62

Presentation, Natural History, and Staging Patients with bladder cancer typically present with hematuria, with or without irritative urinary symptoms. Urinary frequency, nocturia, and dysuria are common as a result of reduced bladder compliance, urinary outflow obstruction, or superimposed infection. Hematuria may be grossly evident or detectable only by microscopic examination of the urine. Patients with locally advanced or metastatic muscle-invasive tumors often have symptoms that reflect the pattern of disease spread. Ureteric obstruction may ­result in costovertebral angle pain, particularly if the obstruction develops rapidly, and places patients at risk for urinary sepsis. Direct extension of tumor to adjacent pelvic organs may produce pain, a change in bowel habits, rectal bleeding, or vaginal bleeding. Extensive pelvic lymphadenopathy is a cause of leg edema. Distant metastases most often involve liver, lung, and bone. All patients with a suspected or confirmed new diagnosis of bladder cancer should undergo a thorough ­history and physical examination, cystoscopy, and examination under anesthesia. Multiple tumors may be found in the bladder at diagnosis, and disease recurrence or the development of new tumors is common despite treatment. These characteristics imply a generalized susceptibility to carcinogenesis that affects the entire mucosa. Transurethral ­resection or a deep biopsy of the bladder tumor is required to ­ establish the diagnosis and determine important prognostic characteristics, including histologic type, grade, and the presence or absence of muscularis propria invasion. The rest of the bladder should be examined, and other abnormal-looking areas should be biopsied. Random biopsies of  normal-­appearing mucosa are required to exclude occult CIS, particularly if radiation therapy is being considered for the treatment of muscle-invasive disease.63 The location, size, mobility, and extravesical extension of palpable disease should be carefully documented. Patients with muscle-invasive bladder cancer require chest radiography, computed tomography (CT) of the abdomen and pelvis, and bone scanning to exclude lymph node and distant metastases. Magnetic resonance imaging (MRI) provides better soft tissue resolution than CT and more accurate identification of the extent of local disease for staging and treatment planning.64 Axial, sagittal, and coronal views of T2-weighted MRI of a muscle-invasive tumor with extensive involvement of the anterior wall and dome of the bladder are shown in Figure 26-2.

632

PART 7  Urinary Tract

UC

A

UC

B

UC

C FIGURE 26-2.  T2-weighted axial (A), sagittal (B), and coronal (C) magnetic resonance images of a urothelial carcinoma (UC) show extensive involvement of the anterior and superior walls of the bladder.

The 2002 American Joint Committee on Cancer (AJCC) and Union Internationale Contre le Cancer (UICC) tumornode-metastasis (TNM) staging system for bladder cancer is outlined in Box 26-2.65,66 The T categories reflect the progressively worse prognosis associated with penetration of tumor through the anatomic layers of the bladder wall

into the perivesical tissues and organs. Transurethral resection may be sufficient to detect the presence or absence of muscle invasion, but it is not adequate for determining the extent of muscle invasion, which is required to  distinguish T2a, T2b, and T3 tumors.41 Detailed pathologic information from a complete cystectomy specimen is ­ necessary to properly categorize muscle-invasive tumors ­ using this system, which limits its applicability when ­patients ­undergo bladder-conserving treatment with ­radiation therapy. Nevertheless, all cases of bladder cancer should be staged ­according to the current TNM classification, taking ­ account of all available clinical, radiographic, and ­pathologic ­information. In men, bladder cancer may involve the prostate gland. Tumor that extends directly from the bladder into the prostate carries a worse prognosis than a separate focus of urothelial carcinoma in the prostate or superficial involvement of the mucosa of the prostatic urethra by bladder cancer.67,68 The latter two situations reflect the generally more favorable prognosis associated with primary urothelial carcinoma of the prostate. Patients with tumor involving the bladder neck should undergo random biopsies of the prostatic urethra as a component of routine staging. It is important to exclude occult CIS of the prostatic urethra, particularly if cystectomy and continent urinary diversion with an orthotopic bladder are contemplated.69 Pelvic lymph node metastases are present at diagnosis in 20% to 30% of patients with bladder cancer. The depth of bladder wall invasion, as reflected in the pathologic T category, is the strongest surgical-pathologic predictor of lymph node involvement. That risk is less than 10% for patients with Ta, Tis, or T1 tumors; 15% to 20% for those with muscle-invasive disease confined to the bladder wall (T2); and 30% to 50% for patients with extravesical ­ disease.70-75 In the current TNM staging classification (see Box 26-2), increasing N category reflects the increasing bulk of external and internal iliac lymph node disease, independent of nodal laterality in the pelvis in relation to the primary tumor.65,66 Involvement of the common iliac and para-aortic lymph nodes is classified as distant metastatic disease and carries an M1 designation. In most cases, patients with lymph nodes found to be positive at ­lymphadenectomy have microscopic involvement of only one or two nodes.72,76 CT, MRI, and positron emission ­tomography (PET) can be used to assess lymph node status in patients being considered for primary radiation therapy, but none is particularly sensitive for the detection of small  (<1 cm) metastases.77-80

Management of Ta, T1, and Tis Bladder Cancer Surgery Superficial bladder cancer is typically managed with complete transurethral resection of intravesical disease. Careful pathologic examination of the resected tissue is required to exclude invasion of the muscularis propria. Given the high likelihood of disease recurrence or the development of new tumors, patients should undergo urinary cytologic  and cystoscopic examinations every 4 to 6 months for  5 years after resection. Persistently abnormal findings on urine cytology after normal findings on cystoscopy should prompt evaluation of the prostatic urethra in men and the upper urinary tract in men and women.81-83 Risk factors

26  The Urinary Bladder

633

BOX 26-2.  American Joint Committee on Cancer Staging System for Bladder Cancer Primary Tumor (T) TX Primary tumor cannot be assessed T0 No evidence of primary tumor Ta Noninvasive papillary carcinoma Tis Carcinoma in situ: “flat tumor” T1 Tumor invades subepithelial connective tissue T2 Tumor invades muscle pT2a Tumor invades superficial muscle (inner half) pT2b Tumor invades deep muscle (outer half) T3 Tumor invades perivesical tissue pT3a Tumor invades microscopically pT3b Tumor invades macroscopically (extravesical mass) T4 Tumor invades any of the following: prostate, uterus, vagina, pelvic wall, abdominal wall T4a Tumor invades prostate, uterus, vagina T4b Tumor invades pelvic wall, abdominal wall Regional Lymph Nodes (N)* NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Metastasis in a single lymph node, 2 cm or less in greatest dimension N2 Metastasis in a single lymph node, more than 2 cm but not more than 5 cm in greatest dimension; or multiple lymph nodes, none more than 5 cm in greatest dimension N3 Metastasis in a lymph node, more than 5 cm in greatest dimension Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage Grouping Stage 0a Ta N0 M0 Stage 0is Tis N0 M0 Stage I T1 N0 M0 Stage II T2a N0 M0 T2b N0 M0 Stage III T3a N0 M0 T3b N0 M0 T4a N0 M0 Stage IV T4b N0 M0 Any T N1-3 M0 Any T Any N M1 *Regional

lymph nodes are situated in the true pelvis (below the bifurcation of the common iliac vessels, corresponding to the internal and external iliac lymph nodes); metastases involving other lymph node groups are classified as distant metastases (M1). From Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 335-340.

for tumor ­recurrence after transurethral resection include high histologic grade, involvement of the lamina propria (T1), multifocal disease, and the presence of CIS.84 About 15% of noninvasive cases eventually progress to involve the muscularis propria, and most of these are poorly differentiated, sessile, T1 tumors with diffuse CIS.29 Radical cystectomy should be considered for patients with extensive recurrent Ta, T1, or Tis disease that is ­refractory to intravesical bacillus Calmette-Guérin (BCG); those with risk factors for progression to muscle invasion; and those with severe, irreversible impairment of bladder function due to recurrent disease or prior treatment.

Intravesical Immunotherapy or Chemotherapy Intravesical immunotherapy or chemotherapy is used to treat superficial bladder cancer in two settings: adjuvant therapy after complete transurethral resection with the aim of preventing disease recurrence and progression to

muscle invasion and primary treatment of existing intravesical disease that is too extensive to be completely  resected, including extensive CIS. The many agents used for this purpose have included BCG, thiotepa, ­doxorubicin, and mitomycin C. Cumulative evidence suggests that BCG is the most ­effective form of intravesical therapy.85,86 BCG is typically administered into the bladder weekly for 6 weeks, where it incites an intense inflammatory response in the bladder wall that is responsible for the therapeutic effect. Complete regression of existing superficial disease occurs in up to 70% of patients treated with BCG, including those with extensive CIS.85,87 A meta-analysis of 24 randomized studies comparing the combination of transurethral resection plus BCG or resection plus another form of intravesical treatment with resection alone demonstrated a 27% reduction in the odds of progression in favor of BCG.88 However, the follow-up interval in many of the studies included in

634

PART 7  Urinary Tract

this analysis was short, and the recurrence rates in the BCG and non-BCG groups were quite low, raising concern about the long-term robustness of this finding. Nevertheless, in one randomized study with a minimum follow-up of 10 years, having BCG after complete transurethral resection significantly reduced the rate of local tumor recurrence in the bladder and the development of invasive or metastatic disease.89 The 10-year progression-free survival rate in this trial was 62% for patients treated with BCG and 37% for those who had only transurethral surgery. The corresponding 10-year disease-free survival rates were 75% and 55%. Overall, the evidence strongly supports the use of intravesical BCG after transurethral resection for patients who are at high risk for disease recurrence or progression.

Radiation Therapy Radiation therapy is not commonly used to treat superficial bladder cancer. Nevertheless, many of the largest ­radiation therapy series include patients with T1 tumors.90-94 ­  Complete response rates after radiation therapy with or without concurrent chemotherapy range from 50% to 80%, depending on the tumor’s histologic grade, configuration (papillary versus solid), and multiplicity.91,95-97 ­ Sustained local control is achieved in 30% to 60% of cases. Long-term cause-specific survival rates vary between 30% and 70%. Radiation therapy for T1 disease has never been ­compared in a randomized study with more conservative treatment consisting of transurethral resection with or without ­intravesical BCG or chemotherapy. Because of the limited evidence supporting the use of radiation therapy for superficial bladder cancer and the effectiveness and convenience of other management strategies, radiation should be reserved for circumstances in which these ­other options have been exhausted or are inappropriate for ­ patient-­specific reasons.

Management of Muscle-Invasive (T2-T4) Bladder Cancer Muscle-invasive bladder cancer is most often treated with radical cystectomy or external beam radiation therapy (EBRT). The outcomes to be expected with these different strategies are well known from numerous reports of large numbers of patients treated in relatively uniform fashions at single institutions and followed for extended periods. However, the therapeutic efficacy of cystectomy versus that of radiation therapy continues to be debated. A randomized trial comparing these two approaches has never been done, and the results of single-institution studies are difficult to compare because of inherent differences in how patients are selected for one treatment or the other. Young, otherwise healthy patients more often undergo cystectomy, whereas those who are older or have other serious medical problems are more likely to undergo radiation therapy. The resulting selection bias can be adjusted, if not completely corrected for, by accounting for obvious imbalances in known prognostic factors. Other therapeutic approaches that have been investigated include partial cystectomy, planned preoperative ­radiation therapy, planned postoperative radiation therapy, interstitial brachytherapy, and various combinations of cystectomy or radiation therapy with systemic chemotherapy. These approaches are discussed in the following sections.

Radical Cystectomy Radical cystectomy is the most common treatment for ­patients with muscle-invasive bladder cancer. In men, radical cystectomy involves the en bloc removal of the bladder, perivesical tissues, prostate, and seminal vesicles; in women, the procedure involves the en bloc removal of the bladder, uterus, fallopian tubes, ovaries, anterior vaginal wall, and urethra.98 Bilateral pelvic lymphadenectomy is usually performed at the same time. The extent of lymphadenectomy and the techniques used vary from center to center. Some clinicians have advocated thorough removal of all lymph nodes lying between the bifurcation of the aorta superiorly, the inguinal ligament inferiorly, and the genitofemoral nerve (overlying the psoas muscle) laterally70,76 to encompass all of the common iliac, external iliac, and internal iliac lymph nodes. Because the lymph node specimen is often removed en bloc with the cystectomy specimen, the paravesical and intercalating lymph nodes along the course of the collecting ducts often are included. Urinary diversion is required, usually with an ileal conduit; however, continent urinary diversion with a stomal or orthotopic neobladder eliminates the need for an external appliance and is now standard practice for some patients.99 The major prognostic factors for survival after cystectomy are the extent of local disease in the bladder and the presence or absence of pelvic lymph node metastases. ­Local recurrence is expected in less than 10% of patients with primary tumors confined to the bladder wall but in 30% to 50% of patients with direct extravesical extension of disease.74,100 In two large contemporary series,70,101 the 5-year recurrence-free survival rates were 75% to 85% for patients with organ-confined disease (pT1-2  N0), 55% to 60% when tumor involved extravesical tissues but not pelvic lymph nodes (pT3-4  N0), and 30% to 35% for those with nodal metastases (pN1-3). Pelvic lymphadenectomy provides important prognostic information for patients with bladder cancer. The risk of distant metastases and disease recurrence outside the pelvis is significantly higher if pelvic nodes are involved, and adjuvant chemotherapy should be considered in such cases.102-105  The therapeutic benefit of pelvic lymphadenectomy has never been tested in a randomized study. However, several retrospective series have suggested that the combination of cystectomy and bilateral pelvic lymph node ­dissection, with or without chemotherapy, cures between 15% and 30% of patients with nodal metastases, particularly those with nonbulky disease confined to the first-echelon ­pelvic nodes.70,73-75,106-110 In the largest of these studies,70 ­comprising almost 250 patients with positive nodes, the 10-year recurrence-free and overall survival rates were 34% and 23%, respectively. Survival rates were higher among ­ patients with four or fewer involved nodes (40% versus 25%) and among those in whom the primary tumor was confined to the bladder (43% versus 30%). Distant ­metastases developed in 52% of patients with node-positive  disease. Advances in surgical techniques and perioperative care have resulted in reduced surgical morbidity and mortality rates. The major-complication rate after radical cystectomy  is typically less than 20%, and surgical mortality rates are less than 5%.111,112 Radical cystoprostatectomy with preservation of sexual function can be performed in some men.

26  The Urinary Bladder

635

TABLE 26-1.    Effectiveness of External Beam Radiation Therapy Alone for Muscle-Invasive Bladder Cancer Study and Reference Goffinet et al, Yu et al,

1975115

1985116 1981117

5-Year Survival by T Category (%)

Number of Patients

T2

T3 (T3a/T3b)

T4

Overall

384

35-42

20

—

—

356

42

(35/23)

—

—

470

—

—

—

38

Duncan and Quilty, 198690

963

40

26

12

30

Blandy et al, 1988118

614

27

38

9

—

182

46

35

—

40

355

50

(38/280)

—

46

Goodman et al,

Jenkins et al,

1988119 *

Gospodarowicz et al,

199191 *

1991121 *

319

31

16

6

28

Davidson et al, 1990122 *

709

49

28

2

25

Greven et al, 1990123

116

59

10

0

—

Smaaland et al, 1991124

146

26

10†

—

—

308

38‡

14§

—

24

—

—

Jahnson et al,

Fossa et al,

1993125

1993126

60

38

12

1994127

135

42

20

0

26

Moonen et al, 199893

379

25

17

—

22

Borgaonkar et al, 2002128 *

163

48

26

0

45

Vale,

Pollack et al,

*Cause-specific

survival.

†T3/T4

‡T2/T3a. §T3b/T4.

Partial Cystectomy Partial cystectomy has a limited role in the management of bladder cancer.113,114 It is appropriate only for patients with small, solitary tumors that are situated near the dome and well separated from the trigone and bladder neck. Less than 5% of patients meet these criteria. Low-dose preoperative EBRT and postoperative brachytherapy have been used in conjunction with partial cystectomy to reduce the risk of recurrence in the abdominal wall scar and bladder, respectively.

External Beam Radiation Therapy The goal of radiation therapy for bladder cancer is to eradicate the tumor while preserving the structure and function of the bladder and other surrounding normal tissues. Optimal delivery of definitive radiation therapy requires a detailed understanding of the multiple biologic and treatment variables that influence local tumor control and the development of radiation-induced side effects. The published experience with EBRT alone for the treatment of bladder cancer is summarized in Table  26-1.90,91,93,115-128 Radical EBRT produces complete regression of muscle-invasive disease in approximately 70% of patients,90,118-120,123,124,126 and 30% to 50% experience sustained local control (i.e., complete tumor regression without subsequent recurrence in the bladder).91,121,123 Nevertheless, distant metastases develop in more than 50% of patients,120 and long-term overall survival rates are only about 25% to 40%.90,93,120,125,127 Salvage cystectomy for progressive or ­ recurrent disease after radiation therapy,

which is an important part of the integrated management plan, has historically been undertaken in less than 30% of patients.120,123,125,129 Overall, approximately 80% of longterm survivors after EBRT have intact, well-functioning bladders.94 Sustained local control of muscle-invasive bladder cancer and long-term patient survival are strongly influenced by the bulk of local disease at the time of the radiation therapy. Advanced T category, large tumor, and the presence of an extravesical mass have been associated with incomplete response to radiation, local disease recurrence, or death from bladder cancer.90-94,120,123,126-128,130 ­ Transurethral ­resection of all gross intravesical tumor before radiation may improve local control and survival rates,94,131 although this has not been a consistent observation.124,126,132 ­Overall survival rates are in the range of 30% to 50% for patients with T2 tumors and 15% to 35% for those with more ­advanced disease.90,115-128 Several factors other than local tumor bulk influence the outcome of patients with muscle-invasive bladder cancer undergoing radiation therapy. Multivariate analyses have identified advanced age at diagnosis, high histologic grade, lymphatic vessel involvement, and hydronephrosis to be strong predictors of reduced survival, independent of ­ tumor bulk.91,92,94,133 Molecular markers of apoptosis, DNA repair capacity, and cell cycle regulation, including ­altered expression of CDKN1A, TP53, RB1, BAX, and BCL2 genes, have been associated with worse clinical outcome in patients with bladder cancer who undergo radiation therapy.54-62 Hoskin and colleagues134 evaluated tumor

636

PART 7  Urinary Tract

TABLE 26-2.   Randomized Studies of Combination Chemotherapy and Radiation Therapy for Patients with Bladder Cancer

Control Group

Overall Survival Rates at 3-5 Years

Methotrexate

RT alone or PreRT + cyst

39% vs. 37%, NS

T2-T4

Cisplatin

RT alone

NS

123

T2-T4

CMV

RT + concurrent cisplatin Early cyst if no CR

48% vs. 49%, NS

976

T2-T4

CMV

RT or cyst

55% vs. 50%, NS

99

T2-T4

Cisplatin

RT alone or PreRT + cyst

47% vs. 33%, NS (improved pelvic control)

T3

Doxorubicin + fluorouracil

RT

NS

Number of Patients

Disease Stage

Chemotherapy Regimen

Shearer et al, 1988139

423

T3

Wallace et al, 1991140

255

Shipley et al (RTOG 89-03), 1998141 Ghersi et al (international collaboration), 199923

Study and Reference Neoadjuvant Therapy

Concurrent Therapy Coppin et al, 199624

Adjuvant Therapy Richards et al, 1982142

129

CMV, cisplatin, methotrexate, and vinblastine; CR, complete response; Cyst, cystectomy; EORTC, European Organisation for Research and Treatment of Cancer; NS, not significant; PreRT, planned preoperative radiation therapy; RT, radiation therapy; RTOG, Radiation Therapy Oncology Group.

hypoxia in 64 patients with bladder cancer by measuring the intrinsic hypoxic marker proteins carbonic anhydrase IX (CA9) and glucose transporter 1 (GLUT1). No association was found between hypoxia and radiation-induced tumor control. However, hypoxia strongly predicted causespecific and overall survival rates independent of clinical prognostic factors. These findings are consistent with those from studies of cervical cancer and other types of tumors, in which hypoxia had only a small effect on radiation­induced local control but had a profound effect on the development of lymph node and distant metastases.135 Radiation therapy for bladder cancer is associated with a long-term risk of new tumor development in the preserved bladder. For example, Gospodarowicz and others91 reported  a continuous decline in the local relapse-free and causespecific cystectomy-free survival rates in a large cohort of patients with long follow-up, in part because of new tumor development. The patients at greatest risk for developing new tumors usually are those with multifocal, superficial disease at presentation.91-93 Extensive CIS, when associated with muscle-invasive bladder cancer, is a particularly strong adverse prognostic factor for sustained local control with radiation therapy.120,136-138 In most cases, superficial disease that recurs after radiation can be salvaged with transurethral resection and intravesical therapy with no compromise in overall survival.138 A minority of patients will require cystectomy, typically those with BCG-­resistant disease, extensive superficial recurrence, or impaired bladder function. Although CIS is associated with reduced long-term disease control in the bladder, these results indicate that it is not an absolute contraindication to the use of radiation therapy for patients with muscle-invasive disease, and it should be considered in the context of other 

tumor- and patient-related factors in deciding among treatment options. Any decision to treat initially with radiation therapy in the hope of maintaining normal bladder function must be part of a larger treatment plan that includes salvage cystectomy in the case of persistent or residual tumor after radiation therapy. Many patients referred for definitive ­radiation therapy have bulky, unresectable tumors and are at high risk for occult micrometastases or have concurrent medical problems that make them unsuitable for initial surgery. These patients often are not considered for salvage cystectomy for the same reasons. However, in cases of local failure after radiation therapy, surgery remains the only ­option for prolonged survival, and a higher risk of complications may therefore be acceptable in some circumstances. All ­patients with residual or recurrent bladder cancer after radiation therapy should be evaluated for cystectomy.

Neoadjuvant Chemotherapy Followed by Radiation Therapy Tumor recurrence in the pelvis and the development of distant metastases are common problems after radiation therapy for advanced bladder cancer. Combinations of chemotherapy and radiation therapy have been explored in an effort to improve outcome and preserve normal bladder function, and several phase III studies of this approach have been conducted (Table 26-2).23,24,139-142 Neoadjuvant chemotherapy has the potential to reduce the size of the primary tumor before radiation, thereby reducing the number of clonogenic cells that need to be killed by radiation to control the disease. A reduction  in tumor size may be associated with improved oxygenation, making the remaining clonogenic cells more ­sensitive

26  The Urinary Bladder

to radiation. Neoadjuvant chemotherapy also offers the important advantage of early treatment of occult ­metastatic disease. Initial phase II clinical studies suggested that neoadjuvant chemotherapy increased local control and ­prolonged survival among patients with bladder cancer.143-145 However, the results of several phase III studies have been disappointing. The largest of these studies included 976 patients with T2 to T4 urothelial carcinoma who were randomly assigned to receive cystectomy or radiation therapy alone or three cycles of neoadjuvant chemotherapy with cisplatin, methotrexate, and vinblastine (CMV) followed by cystectomy or radiation therapy.23 No differences were found in local control, overall survival, or toxicity between the chemotherapy and no-chemotherapy groups in that study. Even though individual phase III studies of neoadjuvant chemotherapy have generated less than encouraging results, a meta-analysis of data pooled from 2688 patients from  10 studies in which the primary treatment was ­cystectomy or radiation therapy demonstrated a survival advantage in favor of neoadjuvant treatment with platinum-based chemotherapy.146 Neoadjuvant therapy led to a 13% relative reduction in the risk of death and an improvement in  5-year survival from 45% to 50%. Local disease-free survival and metastasis-free survival rates were also improved (relative risk reductions of 13% and 18%, respectively). Chemotherapy seemed to be equally effective regardless of the treatment that followed (cystectomy or radiation therapy).

Radiation Therapy with Concurrent Chemotherapy Theoretically, concurrent administration of radiation and chemotherapy would enhance local tumor control by sensitizing cells to radiation, killing cells that are resistant to radiation, and offsetting the adverse affect of tumor proliferation during treatment. Chemotherapy may also kill cells outside the irradiated volume, thereby reducing the risk of developing recurrent disease at distant sites. Several reports have suggested a benefit from combined treatment with radiation therapy and concurrent chemotherapy for patients with bladder cancer.94,147 ­Rodel and colleagues94 updated their large experience with 415 ­patients with T1, T2, T3, or T4 bladder cancer, 289 of whom underwent radiation therapy and concurrent chemotherapy. More than 50% of the patients in that study had T3 or T4 disease. Complete tumor regression was observed in 72% of patients, 50% of whom were free of any relapse in the bladder at 10 years after treatment. Distant metastases developed in 35% of patients. The  10-year cause-specific survival rate was 42%. Shipley and colleagues147 reported similar results after radiation therapy and concurrent chemotherapy, with early cystectomy for patients who failed to respond to induction treatment. The long-term cause-specific survival rates were 60% overall and 45% for those with an intact bladder. Eapen and ­colleagues148,149 reported excellent results from using concurrent intra-arterial cisplatin for patients with T3 or T4 bladder cancer. Complete clearance of pelvic tumor was seen in 96% of patients, and the 2-year actuarial survival rate was 90%. However, 46% of patients developed sacral sensory neuropathy. The risk of neuropathy was significantly lower in a subsequent cohort of patients who received a lower dose of cisplatin.

637

The only phase III study of concurrent chemoradiation therapy for bladder cancer was conducted by the ­National Cancer Institute of Canada.24 Patients with T2 to T4b ­tumors were randomly assigned to receive locoregional ­therapy (definitive radiation therapy or ­preoperative radiation therapy and cystectomy) with or ­without ­concurrent intravenous cisplatin. No differences were found in overall survival or distant relapse-free survival rates between the treatment groups. However, patients who ­received concurrent chemotherapy had a significantly lower rate of pelvic recurrence than those treated with ­radiation alone. ­Serious side effects occurred in similar proportions of patients in both groups. These findings provide the strongest ­evidence for using cisplatin concurrently with radiation for bladder cancer as a means of enhancing local tumor control and minimizing the likelihood of future tumor-related ­morbidity.

Radiation Therapy Followed by Adjuvant Chemotherapy Only one phase III clinical study of adjuvant chemotherapy after radiation therapy for high-risk bladder cancer has been conducted. This study was done during the precisplatin era and showed no difference in overall survival.142 In contrast, several randomized studies of platinum-based chemotherapy after cystectomy for patients with locally extensive primary tumors or involved pelvic lymph nodes102-105 have shown that adjuvant chemotherapy improved diseasefree survival but did not affect overall survival. This result probably reflects the high likelihood of response to chemotherapy administered at the time of recurrence, particularly among chemotherapy-naïve patients, such that the beneficial effect of adjuvant treatment on overall survival was diminished. Adjuvant chemotherapy may have some benefit for ­patients undergoing radiation therapy. However, the indications for adjuvant chemotherapy after radiation therapy are less well defined, mainly because the detailed surgicalpathologic prognostic information provided by cystectomy and pelvic lymphadenectomy is not usually available. Further study is needed to identify clinical, surgical-­pathologic, microenvironmental, and molecular factors predictive of disease recurrence after radiation therapy to aid in ­selecting patients for additional treatment.

Planned Preoperative Radiation Therapy Planned preoperative radiation therapy for the management of bladder cancer, although not often used in modern practice because of the availability of effective systemic chemotherapy, was advocated in the past as a means of ­enhancing tumor control and preventing recurrence in the pelvis. The patients most likely to benefit from preoperative radiation therapy are those with locally extensive primary tumors (pT3b or pT4), which are associated with a 30% to 50% risk of lymph node metastases and a similar risk of pelvic recurrence after cystectomy alone.70-75,100 Preoperative radiation therapy is unlikely to yield significant improvements in overall survival in such cases because the clinical and surgical-pathologic prognostic factors that predict local recurrence also predict distant recurrence. The largest single-institution experience with preoperative radiation therapy for bladder cancer is from The University of Texas M. D. Anderson Cancer Center.150 Planned

638

PART 7  Urinary Tract

TABLE 26-3.   Randomized Studies of Planned Preoperative Radiation Therapy and Cystectomy for Patients with Bladder Cancer Study and Reference

Number of Patients

Disease Stage

Preoperative Radiation Dose

Bloom et al, 1982152

189

T3

40 Gy

Anderstrom et al, 1983153

44

T1-T3

1985154 *

Overall Survival Rates at 3-5 Years

RT + cystectomy for salvage

38% vs. 29%, NS

Cystectomy alone

75% vs. 61%

20 Gy

Cystectomy alone

Sell et al, 1991155

183

T2-T4a

40 Gy

RT + cystectomy for salvage

NS

Smith et al (SWOG), 1997156

140

Tis-T3†

20 Gy

Cystectomy alone

43% vs. 53%, NS

Ghoneim et al,

92

Control Group

*Patients

with schistosomiasis. †Sixty-four percent were T3 tumors. NS, not significant; RT, radiation therapy; SWOG, Southwest Oncology Group.

preoperative radiation therapy for 338 patients with T2 to T4 disease resulted in pathologic downstaging in 65% of patients, and no tumor was found in the surgical specimen in 42%. The overall survival rate at 5 years was 44%. The rates of pelvic and distant recurrence were 16% and 43%, respectively. Preoperative radiation therapy seemed to improve pelvic control among patients with T3b disease (91%) relative to another cohort treated with cystectomy alone (72%).151 Several randomized trials have compared planned preoperative radiation therapy plus cystectomy with cystectomy alone or radical radiation therapy alone followed by cystectomy for salvage (Table 26-3).152-156 Many of these studies included patients who had early-stage disease and were unlikely to benefit from preoperative radiation. Other problems included the use of relatively low doses of radiation and infrequent reporting of local control, which is the end point most likely to be affected by preoperative treatment. Two studies showed a survival advantage in favor of preoperative radiation therapy, but no between-group differences were found in the other studies. The preoperative treatment was well tolerated and did not increase surgical morbidity or mortality.154,155 Overall, the indications for planned preoperative ­radiation therapy in the management of bladder cancer are limited. Patients with high-risk disease are better treated with neoadjuvant chemotherapy followed by cystectomy or radiation therapy, with the aim of improving local control and reducing the likelihood of developing distant ­metastases.

Postoperative Radiation Therapy Postoperative pelvic radiation therapy, like preoperative ­radiation therapy, is most likely to benefit patients who are at high risk for pelvic recurrence. This conclusion is supported by several reports of low pelvic recurrence rates (10% to 20%) despite the presence of high-risk surgicalpathologic prognostic factors among patients with ­bladder cancer.157-160 However, the results in many studies are difficult to interpret because patients also often received planned preoperative pelvic radiation therapy.157,158 The only randomized study of radical cystectomy alone versus cystectomy and postoperative radiation for high-risk bladder cancer was reported by Zaghloul and colleagues159 and

accrued patients with schistosomiasis. The 5-year pelvic recurrence rate was 50% in the cystectomy-only group but only 7% with the addition of radiation therapy. Postoperative radiation therapy also reduced the risk of metastases and improved disease-free survival rates.159 Notwithstanding these results, which may not necessarily be translatable to patients with urothelial carcinoma and without schistosomiasis, the benefit of pelvic radiation on overall survival is likely to be small. Postoperative pelvic radiation therapy might be valuable as a means of reducing potentially morbid pelvic ­recurrences in patients with high-risk bladder cancer if it were not for the unacceptably high rate of serious late gastrointestinal toxicity, which ranges from 20% to 40%.157,158,160-162 Many factors probably contribute to this increased complication rate, but the most important may be the larger volume of small bowel occupying the pelvis after cystectomy. Patients who are identified after cystectomy as being at high risk for recurrence usually are better treated with chemotherapy, which affects local and distant relapse and is usually well tolerated by otherwise healthy patients.

Interstitial Radiation Therapy Interstitial radiation therapy for bladder cancer has been practiced in some parts of Europe for many years. This approach allows a high dose of radiation to be delivered ­focally to a small area of the bladder with relative sparing of surrounding normal tissues. However, only a few selected bladder tumors are suitable for treatment with interstitial radiation therapy.163-168 These tumors are solitary, less than 5 cm in diameter, and preferably confined to the bladder wall (pT1 or pT2) with no extravesical ­extension. Random biopsies of uninvolved bladder ­mucosa should be negative for CIS. The patient must be medically fit for surgery. These criteria are similar to the selection criteria for partial cystectomy, and interstitial ­radiation therapy often has been used in conjunction with partial cystectomy to minimize the risk of recurrence in the bladder wall. Several techniques for interstitial bladder irradiation have been described that involve the temporary or permanent implantation of active or remotely afterloaded radiation sources.163-165,169-173 Pelvic EBRT with doses of 10 to 40 Gy before implantation of the brachytherapy source has been advocated to reduce the risk of recurrence at

26  The Urinary Bladder

the ­operative site.163,164,167,168,171 The interstitial radiation dose is typically 30 to 50 Gy, depending on the preoperative dose.163-165,172 The largest experience with interstitial ­radiation therapy for bladder cancer was described by ­Rozan and colleagues165 and was based on 205 patients treated at eight radiation therapy centers in France. Most patients had solitary T1 or T2 tumors, and 58% underwent partial cystectomy before interstitial radiation therapy. ­Tumor recurred locally in 17% of patients, and this was often the only site of recurrence. The long-term overall survival rate was 67%, which is higher than in most series of cystectomy or EBRT, a finding that undoubtedly reflects the generally more favorable characteristics of patients who are candidates for interstitial treatment. Moderate to severe late side effects developed in 14% of patients. Other investigators have described chronic mucosal ulceration in 15% to 30% of patients, although this condition is usually asymptomatic and heals spontaneously.164,168,172 The limited clinical evidence suggests that interstitial  radiation therapy may contribute to enhanced pelvic control and bladder preservation in selected patients. ­However, no phase III clinical studies have been conducted to compare interstitial radiation therapy with conservative surgery alone or high-dose EBRT. The role of interstitial brachytherapy  in routine clinical practice therefore remains ­unclear.

Palliative Management of Bladder Cancer It is important to place palliative radiation therapy for bladder cancer in the context of overall palliative care. The Canadian Palliative Care Standards Task Force defined palliative care as “the combination of active and compassionate therapies intended to comfort and support the patient and family who are living with a life-threatening illness.”174 Unlike definitive treatment, the objective of which is long-term tumor control, palliative therapy has as its goals symptom relief and preservation of quality of life.

Radiation Therapy in Palliative Care Radiation therapy has an important role in the palliative management of bladder cancer symptoms caused by ­local tumor progression in the pelvis or metastatic disease  (Box 26-3). Fractionation schemes commonly used for palliative pelvic radiation therapy include 30 to 35 Gy given in 10 daily fractions, 20 Gy in 5 daily fractions, and 21 Gy in  3 fractions every other day. Many case series have described reduced hematuria and pelvic pain in a large proportion of patients treated with these regimens but less consistent relief of urinary frequency, urgency, and nocturia and high rates of gastrointestinal toxicity (20% to 30%).175-178 The Medical Research Council in the United Kingdom completed a prospective, randomized trial of two dose prescriptions for palliative pelvic radiation therapy for bladder cancer, comparing 35 Gy in 10 fractions over 2 weeks with 21 Gy in 3 fractions given every second day.179 The study accrued 500 patients, but only 272 were fully assessable 3 months after completing radiation therapy, reflecting the high mortality rate in this population. No differences in symptom relief or side effects were observed between the two treatment groups. Overall, hematuria improved in 88% of patients, dysuria improved in 72%, and urinary frequency improved in 82% of patients in that study. Case series and small studies highlight several issues that need to be considered in prescribing palliative pelvic

639

BOX 26-3.  Indications for Palliative Radiation Therapy for Patients with Bladder Cancer Local Disease Hematuria Pelvic pain Dysuria Urinary frequency Urgency Metastatic Disease Spinal cord compression Bone pain Nerve entrapment Nodal metastases causing pain or lymphedema Brain metastases Bronchial obstruction or hemoptysis

r­ adiation therapy for patients with bladder cancer. Short palliative fractionation schemes are convenient for patients and often temporarily alleviate hematuria and pain. However, symptoms are likely to recur given sufficient time. Short-course treatment is therefore most appropriate for patients with pelvic pain or hematuria from bladder cancer who also have rapidly worsening performance status, metastases, or limited life expectancy for other reasons. However, if sustained local tumor control (not just palliation of symptoms) is the objective of treatment, a higher, more conventional radiation dose with smaller daily fractions should be prescribed. Severe urinary symptoms are rarely adequately controlled with radiation therapy, and urinary diversion may be more appropriate in such cases.

Chemotherapy in Palliative Care Many patients with bladder cancer present with distant metastases at diagnosis or develop metastatic disease during the course of their illness. The survival time for patients with untreated metastatic bladder cancer is usually less than 6 months, and a variety of disabling symptoms can develop during this interval that disrupt quality of life.180,181 Chemotherapy may be effective in ameliorating these symptoms. Urothelial carcinoma is very sensitive to chemotherapy. Cisplatin is the most active single agent, and response rates in the range of 15% to 30% have been documented in several studies.180 Cisplatin has been combined with other ­active agents to yield regimens with higher response rates, usually at the cost of greater toxicity. The most commonly used regimens are methotrexate, vinblastine, doxorubicin, and cisplatin (M-VAC); cisplatin, methotrexate, and vinblastine (CMV); and gemcitabine plus cisplatin. M-VAC, compared with cisplatin alone in a cooperative-group randomized study, was found to improve the response rate and to extend progression-free and overall survival times.182,183 However, only 3.7% of patients randomly assigned to ­receive M-VAC were alive at 6 years. Good performance status at the time of treatment and the presence of only nodal or soft tissue metastases but no visceral metastases were associated with prolonged survival. Patients receiving M-VAC experienced greater nausea and vomiting, mucositis, and hematologic toxicity, and 4% died of drug-related complications.182,183 Other randomized studies showed

640

PART 7  Urinary Tract

M-VAC to be superior to the combination of cyclophosphamide, doxorubicin, and cisplatin and showed CMV to be superior to methotrexate and vinblastine alone.184,185 Although M-VAC and CMV can extend survival time compared with expectant management or treatment with cisplatin alone, these regimens are associated with significant toxicity that may detract from the goal of palliation for patients with metastatic disease. Other regimens comprising active agents with nonoverlapping toxic effects are being investigated with the aim of enhancing the therapeutic ratio. They include combinations of cisplatin with gemcitabine or paclitaxel.181 An update of a large international randomized study that compared M-VAC with gemcitabine plus cisplatin for patients with metastatic bladder cancer186,187 revealed no differences in response rate, survival rate, or quality of life between the treatment groups. However, gemcitabine plus cisplatin was significantly less toxic; more patients completed the full course of treatment, fewer had significant neutropenia  or sepsis, and the rate of death from side effects was lower. These findings have led to gemcitabine plus cisplatin  becoming the standard of care for patients with metastatic bladder cancer.181 Because advanced bladder cancer is characterized by abnormalities in many genes and molecular signaling pathways, molecular therapies are being developed for use alone or in combination with conventional cytotoxic chemotherapies. Therapies in development include those targeted to the epidermal growth factor receptor or other components of the mitogen-activated protein kinase (MAPK) pathway, mutant TP53, the angiogenic process, and DNA ­hypermethylation.45,181

Radiation Treatment Planning and Delivery Patient Selection The goal of definitive radiation therapy for bladder cancer is cure with maintenance of normal bladder function. For carefully selected patients, this approach has the potential benefits of a high probability of durable local control, the ability to deliver treatment in a reproducible fashion, and a low risk of serious complications. However, bladder cancer is a heterogeneous disease, and for an unselected population, the outcome is influenced more by clinical and pathologic prognostic factors than by therapy. All patients with bladder cancer for whom ­ definitive ­radiation therapy is being contemplated should be ­examined thoroughly to document the location and size of the tumor in the bladder, the degree of muscle invasion, the extent of coexisting CIS, and the presence or absence of lymph node or distant metastases. Intravesical tumors should be completely resected to facilitate clinical ­staging and to ­improve the likelihood of achieving a complete ­response to radiation therapy.94,131 Residual tumor masses larger than 5 cm that involve extravesical tissues or organs are unlikely to be controlled by radiation therapy. Gross pelvic lymphadenopathy also implies a low probability of success after radiation.94 A better management strategy for patients in this situation who desire bladder conservation is to use a multimodality approach that incorporates initial chemotherapy followed by consolidative pelvic ­radiation therapy for those showing a complete response to ­chemotherapy.

Patients with extensive bladder CIS are at high risk for developing new or recurrent tumors after radiation therapy, and cystectomy should be considered for these ­patients instead.120,136-138 However, the presence of focal CIS does not preclude the use of radiation therapy. Superficial bladder cancer, including CIS, that recurs after radiation usually can be managed effectively with transurethral resection and intravesical BCG, provided that BCG has not already been used.138 The potential benefit to be derived from definitive ­radiation therapy and bladder preservation is limited in the presence of severe impairment of bladder function, anatomic and technical factors that limit the accuracy and reproducibility of treatment delivery, and risk factors for severe acute and late complications. Patients with severe irritable bladder symptoms at presentation as a result of long-standing outflow obstruction, chronic infection, or prior transurethral resection often experience aggravation of symptoms during treatment that persists afterward; these patients are better served by cystectomy. Pelvic ­ radiation therapy is not appropriate for patients with active inflammatory bowel disease, previous pelvic radiation, extensive prior pelvic surgery, or chronic pelvic infections because of the high risk for serious late bowel complications. The use of definitive radiation therapy for patients with large bladder diverticula or atonic bladders also should be questioned, because treatment reproducibility for these patients is likely to be poor and the possible need for larger-thannormal fields can increase the risk of serious late radiation complications. Overall, the ideal candidates for definitive radiation therapy with bladder preservation have small cT2 tumors with no associated CIS, no significant bladder symptoms, no other serious medical problems, a desire to maintain normal bladder function, and a willingness to commit to a prolonged course of therapy and regular follow-up cystoscopies thereafter.

Treatment Volume Definition The successful treatment of bladder cancer with radiation requires accurate and reproducible delivery of multiple ­radiation fractions to the gross tumor and to regions of subclinical disease extension. Microscopic tumor is likely to be present in the lamina propria, muscularis propria, and lymphovascular spaces adjacent to the primary tumor, although the degree of extension beyond gross disease has not been studied and is likely to vary. Disease also may extend into paravesical tissues. Evidence of the location and extent of gross tumor from diagnostic and staging tests, including cystoscopy, pelvic CT, and pelvic MRI, should be integrated with CT imaging of the patient in the treatment position to develop a comprehensive three-dimensional radiation therapy plan. CT-based treatment planning is more accurate than conventional planning techniques that rely on cystography alone.128,188-190 However, use of CT for defining the extent of gross tumor, particularly tumors near the trigone or bladder neck, is associated with significant interobserver variability.191,192 This variability may be reduced in the future by using MRI, which has superior soft tissue resolution for treatment planning (see Fig. 26-2), and by registering the CT and MR images with each other. Another source of variability during radiation therapy for bladder cancer is changes in bladder size and position

26  The Urinary Bladder

with the volume of urine. To some extent, this ­variability can be minimized by asking patients to completely ­empty their bladders immediately before imaging done for ­treatment planning purposes and before delivery of each radiation fraction. However, significant changes in bladder volume are still possible, depending on the patient’s ­ability to comply with this request, the interval between voiding and treatment delivery, the patient’s state of hydration, and the use of diuretic medications or beverages. Extrinsic pressure on the bladder, such as can arise from ­differences in rectal filling or from tumor characteristics such as the size and degree of extravesical extension, can also affect its position. Turner and colleagues193 demonstrated interfractional bladder wall movement of at least 1.5 cm in 18 of 30 patients, with the greatest movement among those with large initial bladder volumes. The changes in bladder position resulted in inadequate coverage of the clinical target volume in 10 patients.193 The authors of that study recommended using an isotropic field margin of 2 cm around the gross tumor (clinical target volume plus planning target volume) to account for random internal movement. ­Others have advocated anisotropic bladder margins of  1 cm inferiorly and laterally and 2 to 2.5 cm superiorly near the bladder dome.191,194 However, the use of wider margins at the dome may be associated with an increased risk of ­treatment-related intestinal complications.195 The problem of bladder motion may be best dealt with by using daily imaging immediately before treatment and correcting the individual radiation beams accordingly to ensure appropriate target coverage. Fiducial markers can be

641

implanted in the tumor or adjacent bladder wall at cystoscopy before treatment and visualized daily by means of online portal imaging, or the tumor can be visualized ­directly by using a CT scanner integrated with the ­ treatment unit. Examples of diagnostic CT scans with corresponding kilovoltage cone-beam CT images acquired during treatment of a patient with bladder cancer are shown in  Figure 26-3.196 Pelvic lymph node metastases are present in 20% to 30% of patients with muscle-invasive bladder cancer. For this reason, radiation treatment volumes are usually designed to encompass the internal and external iliac lymph nodes, regardless of whether there is radiographic evidence of gross nodal metastases. However, the benefit of nodal irradiation in bladder cancer has not been documented. Several surgical series have suggested long-term survival rates of 15% to 30% among patients with nodal metastases who undergo thorough lymph node dissection at the time of cystectomy.70,73-75,106-110 By extrapolation, aggressive nodal treatment may also benefit patients whose disease is managed primarily with radiation therapy. Treatment of pelvic nodes with doses sufficient to control microscopic disease generally is well tolerated and does not increase the risk of serious late radiation complications.

Treatment Techniques Radiation therapy for bladder cancer is most commonly delivered by using a two-phase, four-field technique. The initial phase of treatment encompasses the primary tumor and internal and external iliac lymph nodes, followed by

A

B

C

D

E

F

FIGURE 26-3.  Diagnostic axial (A), sagittal (B), and coronal (C) CT images and the corresponding kilovoltage cone-beam CT images (D-F) of a patient with bladder cancer. Daily imaging during fractionated radiation therapy with online correction for organ motion has the potential to increase the accuracy of treatment delivery.

642

PART 7  Urinary Tract

treatment of the bladder tumor alone. The study protocols of the Radiation Therapy Oncology Group provide useful benchmarks for conventional treatment planning.197 Radiation is typically delivered with an isocentric technique using anterior, posterior, right lateral, and left lateral 10- to 25-MV fields. Simulation and treatment are often done while the patient has an empty bladder to minimize the potential for day-to-day variability in bladder position. However, a full or partially full bladder may allow better visualization with CT or MRI for treatment planning and verification.198 Radiation therapy planning should take account of all available information from clinical examination, cystoscopy, and imaging to document the location of the primary tumor and lymph nodes (the clinical target volume) as accurately as possible. As a general guide, the cranial aspect of the planning target volume for the first phase of treatment (which encompasses the pelvic lymph nodes) should be positioned at the level of the midsacrum, the caudal aspect at the lower pole of the obturator foramen, and the lateral aspects at least 1 cm beyond the bony margins of the true pelvis at its widest point. The anterior aspect of the planning target volume usually is dictated by the bladder wall, which may

extend 1 to 2 cm anterior to the pubic symphysis in patients who are treated in the supine position. The anterior bladder wall is best documented by planning CT scans or by the position of the bladder air bubble on air-contrast cystography. The anterior margin of the planning target volume should be positioned 1 cm anterior to the bladder wall or at the anterior aspect of the symphysis, whichever is further forward. Posteriorly, the planning target volume should be 1.5 cm behind the most posterior aspect of the bladder or bladder tumor. Shielding should be used in the corners of the fields to exclude small bowel and the femoral heads. The planning target volume for the second phase of treatment should encompass the gross bladder tumor alone with a 1.5-cm margin. It is not necessary to treat the entire bladder if the tumor can be localized accurately. Digitally reconstructed radiographs for both phases of treatment for a patient with bladder cancer are shown in Figure 26-4. Conventional Radiation Dose and Fractionation. With conventional radiation techniques, the dose that can be delivered safely to pelvic lymph nodes is limited by the radiation tolerance of the surrounding normal tissues (typically 40 to 50 Gy in 1.8- to 2.0-Gy daily fractions). Several dose-fractionation schemes have been used to treat

A

B

C

D

FIGURE 26-4.  Digitally reconstructed radiographs show the typical anteroposterior (A) and lateral (B) pelvic radiation fields for the treatment of bladder cancer. Typical small-volume anteroposterior (C) and lateral (D) bladder tumor boost fields are also shown. The solid red area indicates the bladder tumor, with the external bladder wall contour outlined in red; the rectum is shown in orange; and the sigmoid colon is shown in beige.

26  The Urinary Bladder

­ rimary bladder tumors. In North America, the primary tup mor dose is typically 60 to 66 Gy in daily fractions of 1.8 to  2 Gy given over 6 to 7 weeks. In the United Kingdom, doses of 50 to 55 Gy in 2.5-Gy fractions have been described. Evidence to support the existence of a dose-response relationship for bladder cancer is lacking. At least four case series have suggested improved local tumor control with doses greater than 55 to 60 Gy,93,123,124,199 but others have found no evidence to support such a relationship.127,130 Intensity-Modulated Radiation Therapy. The potential for radiation therapy to eradicate bladder cancer has probably been limited over the years by technologic factors that prevented the accurate and reproducible delivery of radiation at sufficient doses to the primary tumor and pelvic lymph nodes while simultaneously sparing important adjacent normal tissues and minimizing toxicity. Advances in radiation therapy planning and delivery have the potential to overcome these limitations, thereby improving the likelihood of curing bladder cancer while preserving normal bladder function and minimizing side effects. One such advance, intensity-modulated radiation

A

643

therapy, allows the dose to the primary bladder tumor and pelvic lymph nodes to be escalated because of its tight conformation to the target volume (Fig. 26-5), with the expectation of improved tumor control with equal or lesser toxicity to normal tissues.2,200 The narrow planning target margins and steep dose gradients emphasize the importance of daily online imaging to ensure accurate treatment delivery. Hyperfractionated Radiation Therapy. Hyperfractionated radiation therapy, defined as the delivery of a larger number of smaller fractions (typically less than 1.5 Gy) with no increase in overall treatment time, has been explored for bladder cancer as a means of exploiting biologic differences in the DNA repair capacity of tumor cells and normal tissue cells. In theory, hyperfractionated radiation therapy should allow the radiation dose to be increased without increasing the risk of late radiation complications. At least two randomized studies have been conducted comparing hyperfractionated radiation therapy (60 to 84 Gy in 1- to 1.2-Gy fractions two or three times per day) and conventional radiation therapy (60 to 66 Gy in 2-Gy daily

B RECTAL DVH 1.0 0.9 0.8

IMRT

4 FB

0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

C

0

1000 2000

3000 4000 5000 6000

Dose (cGy)

FIGURE 26-5.  A-C, An intensity-modulated radiation therapy (IMRT) plan encompasses the entire bladder and pelvic lymph nodes in a patient with bladder cancer and illustrates the potential for reduced dose to the rectum and other normal tissues. Daily image guidance with online correction for setup variability and internal organ motion is essential for the success of treatment because of the close margins and high dose gradients. The solid red area indicates the bladder tumor, with the external bladder wall contour outlined in red; the rectum is shown in brown; the sigmoid colon is shown in yellow; and the iliac and presacral lymph node clinical target volume is shown in yellow. The 95% isodose volume is outlined in blue. DVH, dose-volume histogram; 4FB, four-field box.

644

PART 7  Urinary Tract

fractions) for patients with bladder cancer.201,202 Both trials included a 2-week rest period midway through treatment to allow recovery from acute side effects. Clinical complete response and sustained local control were improved in the hyperfractionated groups relative to the conventional fractionation groups in both studies. A meta-analysis based on the pooled data from these studies indicated a significant improvement in overall survival from the hyperfractionated treatment.203 Hypofractionated Radiation Therapy. Hypofractionated radiation schedules, in which larger doses in the range of 2.5 to 7 Gy are given daily, have been used for the treatment of bladder cancer, usually in the palliative setting. Only one small phase III study has been done of curative hypofractionated radiation therapy for bladder cancer. In that study,204 patients were randomly assigned to receive partially hypofractionated treatment (30 Gy in 3-Gy daily fractions followed by a 4-week break and then 30 Gy in 1.5-Gy daily fractions) or hyperfractionated radiation therapy (60 Gy in 1.5-Gy daily fractions). Survival rates were inferior in the hypofractionated-therapy group (39% versus 52% in the hyperfractionated group), possibly because of accelerated repopulation during the 4-week break.204 Total radiation doses of 30 to 40 Gy delivered in large (5to 6-Gy) fractions have been reported to provide effective symptom palliation for elderly patients with poor performance status.205,206 However, hypofractionated regimens involving doses in this range have also been associated with an unacceptably high risk of late radiation complications affecting the bladder, small bowel, and rectum.176,207 These results imply that the benefit of hypofractionated radiation therapy in terms of patient convenience must be balanced against the possibility of increased side effects. Careful attention must be given to treatment planning, with the aim of excluding as much normal bladder, small bowel, and rectum as possible from the high-dose volume. Highly conformal treatment plans delivered by intensity-modulated techniques may be helpful in achieving this goal. Accelerated Radiation Therapy. Laboratory and clinical studies have shown that tumor growth can increase during a course of fractionated radiation therapy as a result of changes in the balance between cellular ­proliferation and cell loss.17 Such an increase in tumor growth may offset the beneficial effects of radiation and decrease the probability of tumor eradication. Poorly differentiated urothelial carcinoma has been reported to have a high labeling index and short potential doubling time, both indicators of rapid tumor growth.208-210 Theoretically, accelerated radiation therapy regimens (i.e., those designed to deliver the same total dose in a shorter interval) may improve the outcome for patients with this type of bladder cancer. Care must be taken to allow sufficient time between fractions to ensure complete repair of sublethal radiation damage by normal tissues, an important factor in minimizing the risk of late radiation side effects. At least six phase II studies of accelerated radiation therapy for patients with bladder cancer have suggested that this treatment can improve locoregional control and that it has acceptable toxicity.211-216 Horwich and colleagues217 reported the results of a prospective study in which 229 patients with bladder cancer were randomly assigned to receive accelerated radiation therapy consisting of 60.8 Gy in 32 fractions over 26 days or conventional radiation therapy

with 64 Gy in 32 fractions over 45 days. No differences were found in local control, time to metastasis, or overall survival between the groups, but patients who received ­accelerated therapy had a significantly higher rate of acute bowel complications.

Complications of Radiation Therapy The risk and severity of radiation-induced side effects during and after treatment of bladder cancer depend on the condition of the bladder and other normal tissues before the radiation therapy is begun, the location and extent  of the tumor, the volume of normal bowel and bladder within the irradiated volume, the treatment technique, the dose and fraction size, the presence of concurrent medical  problems, and individual intrinsic radiosensitivity. Most ­patients with bladder cancer who undergo EBRT experience acute intestinal and lower urinary tract side effects. Abdominal cramping, diarrhea, and general rectal and bladder irritation are common and are usually easily controlled with dietary modification and medications. Attention should be paid to the diagnosis and eradication of ­urinary tract infections, which can be difficult to distinguish clinically from acute radiation cystitis. Late complications of radiation therapy for bladder cancer primarily affect the bowel and normal bladder and ­typically arise 1 to 4 years after the treatment is completed.  Approximately 75% of late complications appear by  3 years.90,121 Symptoms of severe bladder damage include urinary frequency, dysuria, and intermittent hematuria. Extreme cases may involve uncontrollable bleeding or a contracted bladder with little useful functional capacity. Estimates of the risk of severe late bladder side effects range from 8% to 15%.90,121,155 Treatment with large doses  per fraction and a history of multiple prior transurethral bladder resections or intravesical chemotherapy can  increase the risk of late effects.20 Mounting evidence indicates that severe acute bladder side effects during radiation therapy may predispose to the development of late side effects.8,11-14 Severe intestinal damage, manifested by persistent diarrhea, rectal bleeding, bowel obstruction, or fistula formation, has been reported in 6% to 17% of patients.90,121 General quality of life seems to be preserved after radiation therapy for bladder cancer, although the data are sparse.218-220 Sexual function remains normal in 30% to 40% of patients.218,221 The mortality rate for patients with locally advanced bladder cancer is high, and many patients do not live long enough to manifest late side effects. Nevertheless, recognition of the factors that contribute to the development of side effects and careful attention to radiation treatment planning and delivery are essential. Current practice with doses in the range of 60 to 65 Gy in 1.8- to 2-Gy daily fractions should produce less than a 5% risk of serious late bladder and intestinal complications. Patients who develop bladder or bowel symptoms after radiation therapy need to be examined carefully to distinguish tumor recurrence from radiation-related side effects.

Assessment of Radiation Response and Management of Recurrence Bladder cancer fails to regress completely in about 30% of patients who undergo radiation therapy. Even those who initially show a complete response are at risk for ­developing

26  The Urinary Bladder

a bladder recurrence or distant metastases, which underscores the importance of long-term bladder surveillance.138 Cystoscopy should be performed 1 to 3 months after the completion of treatment and at 3- to 6-month intervals thereafter. Other tests should be done as warranted by the symptoms. Three distinct clinical scenarios are often grouped together as “local recurrence” after radical radiation therapy for bladder cancer: (1) the failure of radiation therapy to control the index tumor, (2) the development of a new superficial or muscle-invasive bladder tumor, and (3) the presence of residual or recurrent CIS. Failure to control the index tumor and new muscle-invasive disease are indications for prompt salvage cystectomy. However, a new noninvasive papillary tumor or CIS that appears after radiation therapy can often be managed with transurethral resection, intravesical BCG, and continued surveillance.138 Salvage cystectomy should be considered for recurrent CIS after radiation therapy for cases in which the original muscleinvasive tumor developed against a background of BCGresistant disease. Cystectomy should also be considered for patients with severe, irreversible impairment of bladder function at the time of recurrence. Salvage cystectomy for local recurrence of muscle-  invasive bladder cancer after radiation therapy ­historically has been undertaken in only 20% to 30% of patients.94,120,123,125,129,133,152,155 This reflects the generally poor prognosis for patients treated with radiation, who often have contraindications to surgery. Quilty and colleagues129 reported a long-term local control rate of 86% among 111 patients who underwent salvage cystectomy; long-term survival rates in several other series have ranged from 40% to 60%.94,106,120,222-227 The strongest predictor of survival after salvage cystectomy is the extent of tumor at the time of the procedure (Ta/Tis versus T1/T2 versus T3).129,224-226,228 Postoperative morbidity rates from 25% to 35%, and mortality rates from 5% to 8%, have been described.106,223,224 Other reports suggest that salvage cystectomy can be done safely, with a risk of complications that is not significantly different from the risk associated with cystectomy alone or planned preoperative radiation therapy and cystectomy.120,227,229 Prior high-dose pelvic radiation therapy does not preclude continent urinary diversion at the time of cystectomy.229,230 All patients with residual or recurrent muscle-invasive bladder cancer after radiation therapy should be evaluated for cystectomy.

Summary Radiation therapy is an effective treatment for muscle­invasive bladder cancer. It can produce complete eradication of tumor, sustained local control, and prolonged ­survival for selected patients who would otherwise die of the disease. All patients with muscle-invasive bladder cancer should be evaluated for cystectomy and radiation therapy, and the most appropriate treatment should be selected based on individual patient and tumor characteristics to yield the highest probability of cure, the lowest toxicity, and the highest possible quality of life. Advances in ­image-guided intensity-modulated radiation therapy and improved understanding of the radiation biology of tumors and normal tissues will further these goals and ­result in greater proportions of cured patients with intact, ­functional bladders.

645

REFERENCES    1. Williams PL (ed). Gray’s Anatomy, 38th ed. New York: ­Churchill Livingstone, 1995.    2. Milosevic M, Gospodarowicz M, Jewett M, et al. Intensity­modulated radiation therapy (IMRT) for lymph node metastases in bladder cancer. In Gregoire V, Scalliet P, Ang K (eds): Clinical Target Volumes in Conformal and Intensity-Modulated Radiation Therapy. New York: Springer-Verlag, 2003, pp 157-169.    3. Fajardo LF, Berthrong M, Anderson RE. Ureters and urinary bladder. In Radiation Pathology. New York: Oxford University Press, 2001, pp 281-287.    4. Mettler FA, Upton AC. Medical Effects of Ionizing Radiation. Philadelphia: WB Saunders, 1995.    5. Rubin P, Casarett GW. Clinical Radiation Pathology. ­Philadelphia: WB Saunders, 1968.    6. Marks LB, Carroll PR, Dugan TC, et al. The response of the urinary bladder, urethra and ureter to radiation and chemotherapy. Int J Radiat Oncol Biol Phys 1995;31:1257-1280.    7. Minsky BD. Primary treatment of rectal cancer: present and ­future. Crit Rev Oncol Hematol 1999;32:19-30.    8. Greven KM, Lanciano RM, Herbert SH, et al. Analysis of complications in patients with endometrial carcinoma receiving adjuvant irradiation. Int J Radiat Oncol Biol Phys 1991;21:  919-923.    9. Storey MR, Pollack A, Zagars G, et al. Complications from radiotherapy dose escalation in prostate cancer: preliminary results of a randomized trial. Int J Radiat Oncol Biol Phys 2000;48:  635-642.   10. Michalski JM, Winter K, Purdy JA, et al. Toxicity after three­dimensional radiotherapy for prostate cancer on RTOG 9406 dose level V. Int J Radiat Oncol Biol Phys 2005;62:706-713.   11. Dorr W, Bentzen SM. Late functional response of the mouse urinary bladder to fractionated x-irradiation. Int J Radiat Biol 1999;75:1307-1315.   12. Dorr W, Beck-Bornholdt HP. Radiation-induced impairment of urinary bladder function in mice: fine structure of the acute response and consequences on late effects. Radiat Res 1999;151:461-467.   13. Schultheiss TE, Lee WR, Hunt MA, et al. Late GI and GU complications in the treatment of prostate cancer. Int J Radiat Oncol Biol Phys 1997;37:3-11.   14. Zelefsky MJ, Cowen D, Fuks Z, et al. Long-term tolerance of high-dose three-dimensional conformal radiotherapy in patients with localized prostate carcinoma. Cancer 1999;85:2460-2468.   15. Emami B, Lyman J, Brown A, et al. Tolerance of normal tissue to therapeutic radiation. Int J Radiat Oncol Biol Phys 1991;21:  109-122.   16. Michalski JM, Purdy JA, Winter K, et al. Preliminary report of toxicity following 3D radiation therapy for prostate cancer on 3DOG/RTOG 9406. Int J Radiat Oncol Biol Phys 2000;46:  391-402.   17. Thames HD, Hendry JH. Fractionation in Radiotherapy. London: Taylor & Francis, 1987.   18. Stewart FA, Randhawa VS, Michael BD. Multifraction irradiation of mouse bladders. Radiother Oncol 1984;2:131-140.   19. Bentzen SM, Lundbeck F, Christensen LL, et al. Fractionation sensitivity and latency of late irradiation injury to the mouse urinary bladder. Radiother Oncol 1992;25:301-307.   20. Kagan AR. Bladder, testicle, and prostate irradiation injury. In Vaeth JM, Meyer JL (eds): Radiation Tolerance of Normal Tissues. Frontiers in Radiation Therapy and Oncology, vol 23. Basel: Karger, 1989, pp 323-337.   21. Levine L, Richie JP. Urological complications of cyclophosphamide. J Urol 1989;141:1063-1069.   22. Edrees G, Luts A, Stewart F. Bladder damage in mice after combined treatment with cyclophosphamide and x-rays. The influence of timing and sequence. Radiother Oncol 1988;11:  349-360.   23. Ghersi D, Stewart LA, Parmar MKB, et al. Neoadjuvant cisplatin, methotrexate, and vinblastine chemotherapy for muscleinvasive bladder cancer: a randomised controlled trial. Lancet 1999;354:533-540.

646

PART 7  Urinary Tract

  24. Coppin C, Gospodarowicz M, James K, et al. Improved ­ local control of invasive bladder cancer by concurrent cisplatin and preoperative or definitive radiation. J Clin Oncol 1996;14:  2901-2907.   25. Pizzo JJD, Chew BH, Jacobs SC, et al. Treatment of radiation induced hemorrhagic cystitis with hyperbaric oxygen: long-term results. J Urol 1998;160:731-733.   26. Mathews R, Rajan N, Josefson L, et al. Hyperbaric oxygen therapy for radiation induced hemorrhagic cystitis. J Urol 1999;161:  435-437.   27. Bui QC, Lieber M, Withers HR, et al. The efficacy of hyperbaric oxygen therapy in the treatment of radiation-induced late side effects. Int J Radiat Oncol Biol Phys 2004;60:871-878.   28. American Cancer Society. Cancer Facts & Figures, 2008. Atlanta: American Cancer Society, 2008.   29. Brauers A, Jakse G. Epidemiology and biology of human urinary bladder cancer. J Cancer Res Clin Oncol 2000;126:575-583.   30. Cottran RS, Kumar V, Collins T. Pathologic Basis of Disease. Philadelphia: WB Saunders, 1999.   31. Brennan P, Bogillot O, Cordier S, et al. Cigarette smoking and bladder cancer in men: a pooled analysis of 11 case-controlled studies. Int J Cancer 2000;86:289-294.   32. Vineis P, Simonato L. Proportion of lung and bladder cancers in males resulting from occupation: a systematic approach. Arch Environ Health 1991;46:6-15.   33. la Vecchia C, Negri E, D’Avanzo B, et al. Occupation and the risk of bladder cancer. Int J Epidemiol 1990;19:264-268.   34. Kantor AF, Hartge P, Hoover RN, et al. Urinary tract infection and risk of bladder cancer. Am J Epidemiol 1984;119:510-515.   35. Locke JR, Hill DE, Walzer Y. Incidence of squamous cell carcinoma in patients with long-term catheter drainage. J Urol 1985;133:1034-1035.   36. Dolin PJ, Darby SC, Beral V. Paraplegia and squamous cell carcinoma of the bladder in young women: findings from a casecontrol study. Br J Cancer 1994;70:167-168.   37. Liauw SL, Sylvester JE, Morris CG, et al. Second malignancies after prostate brachytherapy: incidence of bladder and colorectal cancers in patients with 15 years of potential follow-up. Int J Radiat Oncol Biol Phys 2006;66:669-673.   38. Neugut AL, Ahsan H, Robinson E, et al. Bladder carcinoma and other second malignancies after radiotherapy for prostate carcinoma. Cancer 1997;79:1600-1604.   39. Brenner DJ, Curtis RE, Hall EJ, et al. Second malignancies in prostate carcinoma patients after radiotherapy compared with surgery. Cancer 2000;88:398-406.   40. Moon K, Stukenborg GJ, Keim J, et al. Cancer incidence after ­ localized therapy for prostate cancer. Cancer 2006;107:  991-998.   41. Epstein JI, Amin MB, Reuter VR, et al. The World Health ­Organization/International Society of Urological Pathology consensus classification of urothelial (transitional cell) neoplasms of the urinary bladder. Am J Surg Pathol 1998;22:1435-1448.   42. Holmang S, Andius P, Hedelin H, et al. Stage progression in Ta papillary urothelial tumors: relationship to grade, immunohistochemical expression of tumor markers, mitotic frequency, and DNA ploidy. J Urol 2001;165:1124-1128.   43. Sylvester RJ, Meijden Avd, Witjes JA, et al. High-grade Ta urothelial carcinoma and carcinoma in situ of the bladder.  Urology 2005;66:90-107.   44. Kirkali Z, Chan T, Manoharan M, et al. Bladder cancer: epidemiology, staging and grading, and diagnosis. Urology 2005;66:4-34.   45. Mitra AP, Datar RH, Cote RJ. Molecular pathways in invasive bladder cancer: new insights into mechanisms, progression, and target identification. J Clin Oncol 2006;24:5552-5564.   46. Nobori T, Miura K,Wu DJ, et al. Deletions of the cyclin-dependent  kinase-4 inhibitor gene in multiple human tumors. Nature 1994;368:753-756.   47. Fujimoto K, Yamada Y, Okajima E, et al. Frequent association of p53 gene mutation in invasive bladder cancer. Cancer Res 1992;52:1393-1398.   48. Sidransky D, von Eschenbach A, Tsai YC, et al. Identification of p53 gene mutations in bladder cancers and urine samples.  Science 1991;252:706-709.

  49. Sarkis AS, Dalbagni G, Cordon-Cardo C, et al. Nuclear overexpression of p53 protein in transitional cell bladder carcinoma: a marker for disease progression. J Natl Cancer Inst 1993;85:  53-59.   50. Sarkis AS, Dalbagni G, Cordon-Cardo C, et al. Association of p53 nuclear overexpression and tumor progression in carcinoma in situ of the bladder. J Urol 1994;152:388-392.   51. Sarkis A, Bajorin D, Reuter V, et al. Prognostic value of p53 nuclear overexpression in patients with invasive bladder cancer treated with neoadjuvant MVAC. J Clin Oncol 1995;13:  1384-1390.   52. Grossman HB, Liebert M, Antelo M, et al. P53 and RB expression predict progression in T1 bladder cancer. Clin Cancer Res 1998;4:829-834.   53. Chatterjee SJ, Datar R, Youssefzadeh D, et al. Combined effects of p53, p21, and pRb expression in the progression of bladder transitional cell carcinoma. J Clin Oncol 2004;22:1007-1013.   54. Moonen L, Ong F, Gallee M, et al. Apoptosis, proliferation and p53, cyclin D1, and retinoblastoma gene expression in relation to radiation response in transitional cell carcinoma of the bladder. Int J Radiat Oncol Biol Phys 2001;49:1305-1310.   55. Hussain SA, Ganesan R, Hiller L, et al. BCL2 expression predicts survival in patients receiving synchronous chemoradiotherapy in advanced transitional cell carcinoma of the bladder. Oncol Rep 2003;10:571-576.   56. Matsumoto H, Wada T, Fukunaga K, et al. Bax to Bcl-2 ratio and Ki-67 index are useful predictors of neoadjuvant chemoradiation therapy in bladder cancer. Jpn J Clin Oncol 2004;34:  124-130.   57. Ong F, Moonen LM, Gallee MP, et al. Prognostic factors in transitional cell cancer of the bladder: an emerging role for Bcl-2 and p53. Radiother Oncol 2001;61:169-175.   58. Qureshi KN, Griffiths TR, Robinson MC, et al. Combined p21WAF1/CIP1 and p53 overexpression predict improved survival in muscle-invasive bladder cancer treated by radical radiotherapy. Int J Radiat Oncol Biol Phys 2001;51:1234-1240.   59. Rotterud R, Berner A, Holm R, et al. P53, p21 and mdm2 expression vs the response to radiotherapy in transitional cell carcinoma of the bladder. BJU Int 2001;88:202-208.   60. Pollack A, Wu CS, Czerniak B, et al. Abnormal bcl-2 and pRb expression are independent correlates of radiation response in muscle-invasive bladder cancer. Clin Cancer Res 1997;3:  1823-1829.   61. Rodel C, Grabenbauer GG, Rodel F, et al. Apoptosis, p53, bcl-2, and Ki-67 in invasive bladder carcinoma: possible predictors for response to radiochemotherapy and successful bladder preservation. Int J Radiat Oncol Biol Phys 2000;46:1213-1221.   62. Wu CS, Pollack A, Czerniak B, et al. Prognostic value of p53 in muscle-invasive bladder cancer treated with preoperative radiotherapy. Urology 1996;47:305-310.   63. Quilty PM, Hargreave TB, Smith G, et al. Do normal mucosal biopsies predict prognosis in patients with transitional cell carcinoma of bladder treated by radical radiotherapy? Br J Urol 1987;59:242-247.   64. Barentsz JO, Jager GL, Witjes JA. MR imaging of the urinary bladder. Magn Reson Imaging Clin North Am 2000;8:853-867.   65. Sobin LH, Wiitekind CH. UICC International Union Against Cancer TNM Classification of Malignant Tumors. 6th ed, New York: Wiley-Liss, 2002.   66. Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 335-340.   67. Pagano F, Bassi P, Ferrante GL, 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;155:244-247.   68. Esrig D, Freeman JA, Elmajian DA, et al. Transitional cell carcinoma involving the prostate with a proposed staging classification for stromal invasion. J Urol 1996;156:1071-1076.   69. Freeman JA, Esrig D, Stein JP, et al. Management of the patient with bladder cancer. Urethral recurrence. Urol Clin North Am 1994;21:645-651.

26  The Urinary Bladder   70. Stein JP, Lieskovsky G, Cote R, et al. Radical cystectomy in the treatment of invasive bladder cancer: long-term results in 1,054 patients. J Clin Oncol 2001;19:666-675.   71. Smith JA, Whitmore WF. Regional lymph node metastases from bladder cancer. J Urol 1981;126:591-593.   72. Wishnow KI, Johnson DE, Ro JY, et al. Incidence, extent and location of unsuspected pelvic lymph node metastases in patients undergoing radical cystectomy for bladder cancer. J Urol 1987;137:408-410.   73. Poulsen AL, Horn T, Steven K. Radical cystectomy: extending the limits of pelvic lymph node dissection improves survival for patients with bladder cancer confined to the bladder wall. J Urol 1998;160:2015-2019.   74. Frazier HA, Robertson JE, Dodge RK, et al. The value of pathologic factors in predicting cancer-specific survival among patients treated with radical cystectomy for transitional cell carcinoma of the bladder and prostate. Cancer 1993;71:3993-4001.   75. Bassi P, Ferrante GD, Piazza N, et al. Prognostic factors of outcome after radical cystectomy for bladder cancer: a retrospective study of a homogeneous patient cohort. J Urol 1999;161:  1494-1497.   76. Lerner SP, Skinner DG, Lieskovsky G, et al. The rationale for en bloc pelvic lymph node dissection for bladder cancer patients with nodal metastases: long-term results. J Urol 1993;149:  758-764.   77. Barentsz JO, Witjes JA, Ruijs JH. What is new in bladder cancer imaging. Urol Clin North Am 1997;24:583-602.   78. Roy C, Le Bras Y, Mangold L, et al. Small pelvic lymph node metastases: evaluation with MR imaging. Clin Radiol 1997;52:  437-440.   79. Heicappell R, Muller-Mattheis V, Reinhardt M, et al. Staging of pelvic lymph nodes in neoplasms of the bladder and prostate by positron emission tomography with 2-[(18)F]-2-deoxy-d­glucose. Eur Urol 1999;36:582-587.   80. Kosuda S, Kison PV, Greenough R, et al. Preliminary assessment of fluorine-18 fluorodeoxyglucose positron emission tomography in patients with bladder cancer. Eur J Nucl Med 1997;24:  615-620.   81. Rabbani F, Perrotti M, Russo P, et al. Upper-tract tumors after an initial diagnosis of bladder cancer: argument for long-term surveillance. J Clin Oncol 2001;19:94-100.   82. Herr HW, Cookson MS, Soloway SM. Upper tract tumors in patients with primary bladder cancer followed for 15 years. J Urol 1996;156:1286-1287.   83. Schwalb DM, Herr HW, Fair WR. The management of clinically unconfirmed positive urinary cytology. J Urol 1993;150:  1751-1756.   84. Millan-Rodriguez F, Chechile-Toniolo G, Salvador-Bayarri J, et al. Primary superficial bladder cancer risk groups according to progression, mortality and recurrence. J Urol 2000;164:  680-684.   85. Herr HW, Laudone VP, Whitmore WF. An overview of intravesical therapy for superficial bladder tumors. J Urol 1987;138:  1363-1368.   86. Bohle A, Jocham D, Bock PR. Intravesical bacillus CalmetteGuérin versus mitomycin C for superficial bladder cancer: a formal meta-analysis of comparative studies on recurrence and toxicity. J Urol 2003;169:90-95.   87. Herr HW, Pinsky CM, Whitmore WF, et al. Long-term effect of intravesical bacillus Calmette-Guérin on flat carcinoma in situ of the bladder. J Urol 1986;135:265-267.   88. Sylvester RJ, van der Meijden AP, Lamm DL. Intravesical ­bacillus Calmette-Guérin reduces the risk of progression in patients with superficial bladder cancer: a meta-analysis of the published results of randomized clinical trials. J Urol 2002;168:1964-1970.   89. Herr HW, Schwalb DM, Zhang ZF, et al. Intravesical bacillus Calmette-Guérin therapy prevents tumor progression and death from superficial bladder cancer: ten-year follow-up of a prospective randomized trial. J Clin Oncol 1995;13:1404-1408.   90. Duncan W, Quilty PM. The results of a series of 963 patients with transitional cell carcinoma of the urinary bladder primarily treated by radical megavoltage x-ray therapy. Radiother Oncol 1986;7:299-310.

647

  91. Gospodarowicz MK, Rider WD, Keen CW, et al. Bladder cancer: long term follow-up results of patients treated with radical radiation. Clin Oncol 1991;3:155-161.   92. Mameghan H, Fisher R, Mameghan J, et al. Analysis of failure following definitive radiotherapy for invasive transitional cell carcinoma of the bladder. Int J Radiat Oncol Biol Phys 1995;31:  247-254.   93. Moonen LM, vd Voet H, de Nijs R, et al. Muscle-invasive bladder cancer treated with external beam radiation: influence of total dose, overall treatment time, and treatment interruption on local control. Int J Radiat Oncol Biol Phys 1998;42:525-530.   94. Rodel C, Grabenbauer G, Kuhn R, et al. Combined-modality treatment and selective organ preservation in invasive bladder cancer: long-term results. J Clin Oncol 2002;20:3061-3071.   95. Rodel C, Dunst J, Grabenbauer G, et al. Radiotherapy is an ­effective treatment for high-risk T1-bladder cancer. Strahlenther Onkol 2001;177:82-88.   96. Quilty PM, Duncan W. Treatment of superficial (T1) tumours of the bladder by radical radiotherapy. Br J Urol 1986;58:147-152.   97. Weiss C, Wolze C, Engehausen DG, et al. Radiochemotherapy after transurethral resection for high-risk T1 bladder cancer: an alternative to intravesical therapy or early cystectomy? J Clin Oncol 2006;24:2318-2324.   98. Richie JP. Surgery for invasive bladder cancer. Hematol Oncol Clin North Am 1992;6:129-145.   99. Wu SD, Simma-Chang V, Stein JP. Pathologic guidelines for orthotopic urinary diversion in women with bladder cancer: a review of the literature. Rev Urol 2006;8:54-60. 100. Greven KM, Spera JA, Solin LJ, et al. Local recurrence after cystectomy alone for bladder carcinoma. Cancer 1992;69:  2767-2770. 101. Madersbacher S, Hochreiter W, Burkhard F, et al. Radical cystectomy for bladder cancer today—a homogeneous series without neoadjuvant therapy. J Clin Oncol 2003;21:690-696. 102. Freiha F, Reese J, Torti FM. A randomized trial of radical cystectomy versus radical cystectomy plus cisplatin, vinblastine and methotrexate chemotherapy for muscle invasive bladder cancer. J Urol 1996;155:495-499. 103. Skinner DG, Daniels JR, Russell CA, et al. The role of adjuvant chemotherapy following cystectomy for invasive bladder cancer: a prospective comparative trial. J Urol 1991;145:459-464. 104. Studer UE, Bacchi M, Biedermann C, et al. Adjuvant cisplatin chemotherapy following cystectomy for bladder cancer: results of a prospective randomized trial. J Urol 1994;152:81-84. 105. Stockle M, Meyenburg W, Wellek S, et al. Adjuvant polychemotherapy of nonorgan-confined bladder cancer after radical cystectomy revisited: long-term results of a controlled prospective study and further clinical experience. J Urol 1995;153:47-52. 106. Smith JA, Whitmore WF. Salvage cystectomy for bladder cancer after failure of definitive irradiation. J Urol 1981;125:643-645. 107. Turner WH, Markwalder R, Perrig S, et al. Meticulous pelvic lymphadenectomy in surgical treatment of the invasive bladder cancer: an option or a must? Eur Urol 1998;33(Suppl 4):21-22. 108. Vieweg J, Gschwend JE, Herr HW, et al. Pelvic lymph node dissection can be curative in patients with node positive bladder cancer. J Urol 1999;161:449-454. 109. Zincke H, Patterson DE, Utz DC, et al. Pelvic lymphadenectomy and radical cystectomy for transitional cell carcinoma of the bladder with pelvic lymph node disease. Br J Urol. 1985;57:156-159. 110. Bretheau D, Ponthieu A. Results of radical cystectomy and pelvic lymphadenectomy for bladder cancer with pelvic node metastases. Urol Int 1996;57:27-31. 111. Pagano F, Bassi P, Galetti TP, et al. Results of contemporary radical cystectomy for invasive bladder cancer: a clinicopathological study with an emphasis on the inadequacy of the tumor, nodes and metastases classification. J Urol 1991;145:45-50. 112. Amling CL, Thrasher JB, Frazier HA, et al. Radical cystectomy for stages Ta, Tis and T1 transitional cell carcinoma of the bladder. J Urol 1994;151:31-36. 113. Holzbeierlein JM, Lopez-Corona E, Bochner BH, et al. Partial cystectomy: a contemporary review of the Memorial Sloan­Kettering Cancer Center experience and recommendations for patient selection. J Urol 2004;172:878-881.

648

PART 7  Urinary Tract

114. Kassouf W, Swanson D, Kamat AM, et al. Partial cystectomy for muscle invasive urothelial carcinoma of the bladder: a contemporary review of the M. D. Anderson Cancer Center experience. J Urol 2006;175:2058-2062. 115. Goffinet DR, Schneider MJ, Glatstein EJ, et al. Bladder cancer: results of radiation therapy in 384 patients. Radiology 1975;117:149-153. 116. Yu WS, Sagerman RH, Chung CT, et al. Bladder carcinoma. Experience with radical and preoperative radiotherapy in 421 patients. Cancer 1985;56:1293-1299. 117. Goodman GB, Hislop TG, Elwood JM, et al. Conservation of bladder function in patients with invasive bladder cancer treated by definitive irradiation and selective cystectomy. Int J Radiat Oncol Biol Phys 1981;7:569-573. 118. Blandy JP, Jenkins BJ, Fowler CG, et al. Radical radiotherapy and salvage cystectomy for T2/3 cancer of the bladder. Prog Clin Biol Res 1988;260:447-451. 119. Jenkins BJ, Caulfield MJ, Fowler CG, et al. Reappraisal of the role of radical radiotherapy and salvage cystectomy in the treatment of invasive (T2/T3) bladder cancer. Br J Urol 1988;62:  342-346. 120. Gospodarowicz MK, Hawkins NV, Rawlings GA, et al. Radical radiotherapy for the muscle invasive transitional cell carcinoma of the bladder: failure analysis. J Urol 1989;142:1448-1454. 121. Jahnson S, Pedersen J, Westman G. Bladder carcinoma—a 20year review of radical irradiation therapy. Radiother Oncol 1991;22:111-117. 122. Davidson SE, Symonds RP, Snee MP, et al. Assessment of factors influencing the outcome of radiotherapy for bladder cancer.  Br J Urol 1990;66:288-293. 123. Greven KM, Solin LJ, Hanks GE. Prognostic factors in patients with bladder carcinoma treated with definitive irradiation. Cancer 1990;65:908-912. 124. Smaaland R, Akslen L, Tonder B, et al. Radical radiation treatment of invasive and locally advanced bladder cancer in elderly patients. Br J Urol 1991;67:61-69. 125. Fossa SD, Waehre H, Aass N, et al. Bladder cancer definitive radiation therapy of muscle-invasive bladder cancer. A retrospective analysis of 317 patients. Cancer 1993;72:3036-3043. 126. Vale JA, A’Hern RP, Liu K, et al. Predicting the outcome of radical radiotherapy for invasive bladder cancer. Eur Urol 1993;24:48-51. 127. Pollack A, Zagars GK, Swanson DA. Muscle-invasive bladder cancer treated with external beam radiotherapy: prognostic factors. Int J Radiat Oncol Biol Phys 1994;30:267-277. 128. Borgaonkar S, Jain A, Bollina P, et al. Radical radiotherapy and salvage cystectomy as the primary management of transitional cell carcinoma of the bladder. Results following the introduction of a CT planning technique. Clin Oncol 2002;14:141-147. 129. Quilty PM, Duncan W, Chisholm GD, et al. Results of surgery following radical radiotherapy for invasive bladder cancer. Br J Urol 1986;58:396-405. 130. Shipley WU, Rose MA, Perrone TL, et al. Full-dose irradiation for patients with invasive bladder carcinoma: clinical and histological factors prognostic of improved survival. J Urol 1985;134:679-683. 131. Shipley WU, Prout GR, Kaufman SD, et al. Invasive bladder carcinoma. The importance of initial transurethral surgery and other significant prognostic factors for improved survival with full-dose irradiation. Cancer 1987;60:514-520. 132. Timmer PR, Harlief HA, Hooijkaas JA. Bladder cancer: pattern of recurrence in 142 patients. Int J Radiat Oncol Biol Phys 1985;11:899-905. 133. Fossa SD, Berner AA, Jacobsen AB, et al. Clinical significance of DNA ploidy and S-phase fraction and their relation to p53 protein, c-erbB-2 protein and HCG in operable muscle-invasive bladder cancer. Br J Cancer 1993;68:572-578. 134. Hoskin P, Sibtain A, Daley F, et al. GLUT1 and CAIX as intrinsic markers of hypoxia in bladder cancer: relationship with vascularity and proliferation as predictors of outcome of ARCON. Br J Cancer 2003;89:1290-1297.

135. Milosevic M, Fyles A, Hedley D, et al. The human tumor microenvironment: invasive (needle) measurements of oxygen and interstitial fluid pressure. Semin Radiat Oncol 2004;14:249-258. 136. Fung CY, Shipley WU, Young RH, et al. Prognostic factors in invasive bladder carcinoma in a prospective trial of preoperative adjuvant chemotherapy and radiotherapy. J Clin Oncol 1991;9:1533-1542. 137. Wolf H, Olsen PR, Hojgaard K. Urothelial dysplasia concomitant with bladder tumours: a determinant for future new occurrences in patients treated by full-course radiotherapy. Lancet 1985;1:1005-1008. 138. Zietman A, Grocela J, Zehr E, et al. Selective bladder conservation using transurethral resection, chemotherapy, and radiation: management and consequences of Ta, T1, and Tis recurrence within the retained bladder. Urology 2001;58:380-385. 139. Shearer RJ, Chilvers CED, Bloom HJG, et al. Adjuvant chemotherapy in T3 carcinoma of the bladder. A prospective trial and preliminary report. Br J Urol 1988;62:558-564. 140. Wallace DMA, Raghavan D, Kelly KA, et al. Neo-adjuvant (preemptive) cisplatin therapy in invasive transitional cell carcinoma of the bladder. Br J Urol 1991;67:608-615. 141. Shipley WU, Winter KA, Kaufman DS, et al. Phase III trial of neoadjuvant chemotherapy in patients with invasive bladder cancer treated with selective bladder preservation by combined radiation therapy and chemotherapy: initial results of Radiation Therapy Oncology Group 89-03. J Clin Oncol 1998;16:  3576-3583. 142. Richards B, Bastable JRG, Freedman L, et al. Adjuvant chemotherapy with doxorubicin (Adriamycin) and 5-fluorouracil in T3, NX, M0 bladder cancer treated with radiotherapy. Br J Urol 1982;55:386-391. 143. Tester W, Caplan R, Heaney J, et al. Neoadjuvant combined modality program with selective organ preservation for invasive bladder cancer: results of Radiation Therapy Oncology Group phase II trial 8802. J Clin Oncol 1996;14:119-126. 144. Farah R, Chodak GW, Vogelzang NJ, et al. Curative radiotherapy following chemotherapy for invasive bladder carcinoma (a preliminary report). Int J Radiat Oncol Biol Phys 1991;20:  413-417. 145. Wajsman Z, Marino R, Parsons J, et al. Bladder-sparing approach in the treatment of invasive bladder cancer. Semin Urol 1990;8:210215. 146. Neoadjuvant chemotherapy in invasive bladder cancer. a systematic review and meta-analysis. Advanced Bladder Cancer (ABC) Meta-analysis Collaboration. Lancet 2003;361:1927-1934. 147. Shipley W, Kaufman D, Zehr E, et al. Selective bladder preservation by combined modality protocol treatment: long-term outcomes of 190 patients with invasive bladder cancer. Urology 2002;60:62-67. 148. Eapen L, Stewart D, Danjoux C, et al. Intraarterial cisplatin and concurrent radiation for locally advanced bladder cancer. J Clin Oncol 1989;7:230-235. 149. Eapen L, Stewart D, Crook J, et al. Intraarterial cisplatin and concurrent pelvic radiation in the management of muscle invasive bladder cancer. Int J Radiat Oncol Biol Phys 1992;24  (Suppl 1):211. 150. Pollack A, Zagars GK, Dinney CP, et al. Preoperative radiotherapy for muscle-invasive bladder carcinoma. Long term follow-up and prognostic factors for 338 patients. Cancer 1994;74:  2819-2827. 151. Cole CJ, Pollack A, Zagars GK, et al. Local control of muscleinvasive bladder cancer: preoperative radiotherapy and cystectomy versus cystectomy alone. Int J Radiat Oncol Biol Phys 1995;32:331-340. 152. Bloom HJ, Hendry WF, Wallace DM, et al. Treatment of T3 bladder cancer: controlled trial of pre-operative radiotherapy and radical cystectomy versus radical radiotherapy. Br J Urol 1982;54:136-151. 153. Anderstrom C, Johansson S, Nilsson S, et al. A prospective randomized study of preoperative irradiation with cystectomy or cystectomy alone for invasive bladder carcinoma. Eur Urol 1983;9:142-147.

26  The Urinary Bladder 154. Ghoneim MA, Ashamallah AK, Awaad HK, et al. Randomized trial of cystectomy with or without preoperative radiotherapy for carcinoma of the bilharzial bladder. J Urol 1985;134:  266-268. 155. Sell A, Jakobsen A, Nerstrom B, et al. Treatment of advanced bladder cancer category T2, T3 and T4a. Scand J Urol Nephrol 1991;138:193-201. 156. Smith J, Crawford E, Paradelo J, et al. Treatment of advanced bladder cancer with combined preoperative irradiation and radical cystectomy versus radical cystectomy alone: a phase III intergroup study. J Urol 1997;157:805-808. 157. Spera JA, Whittington R, Littman P, et al. A comparison of preoperative radiotherapy regimens for bladder carcinoma: the University of Pennsylvania experience. Cancer 1988;61:  255-262. 158. Reisinger SA, Mohiuddin M, Mulholland SG. Combined preand postoperative adjuvant radiation therapy for bladder cancer—a ten year experience. Int J Radiat Oncol Biol Phys 1992;24:  463-468. 159. Zaghloul MS, Awwad HK, Akoush HH, et al. Postoperative radiotherapy of carcinoma in bilharzial bladder: improved disease free survival through improving local control. Int J Radiat Oncol Biol Phys 1992;23:511-517. 160. Zaghloul MS. Radiation as adjunctive therapy to cystectomy for bladder cancer: is there a difference for bilharzial association? [letter]. Int J Radiat Oncol Biol Phys 1994;28:783. 161. Skinner DG, Tift JP, Kaufman JJ. High dose, short course pre­ operative radiation therapy and immediate single stage radical cystectomy with pelvic node dissection in the management of bladder cancer. J Urol 1982;127:671-674. 162. Kopelson G, Heaney JA. Postoperative radiation therapy for muscle-invading bladder carcinoma. J Surg Oncol 1983;23:  263-268. 163. De Neve W, Lybeert ML, Goor C, et al. Radiotherapy for T2 and T3 carcinoma of the bladder: the influence of overall treatment time. Radiother Oncol 1995;36:183-188. 164. Wijnmaalen A, Helle PA, Koper PCM, et al. Muscle invasive bladder cancer treated by transurethral resection, followed by external beam radiation and interstitial iridium-192. Int J Radiat Oncol Biol Phys 1997;39:1043-1052. 165. Rozan R, Albuisson E, Donnarieix D, et al. Interstitial iridium192 for bladder cancer (a multicentric survey: 205 patients). Int J Radiat Oncol Biol Phys 1992;24:469-477. 166. van der Werf-Messing BHP, Hop WCL. Carcinoma of the urinary bladder (category T1 NX M0) treated either by radium implant or by transurethral resection only. Int J Radiat Oncol Biol Phys 1981;7:299-303. 167. van der Werf-Messing B, Menon RS, Hop WCJ. Cancer of the urinary bladder category T2, T3, (NXM0) treated by interstitial radium implant: second report. Int J Radiat Oncol Biol Phys 1983;9:481-485. 168. Moonen LM, Horenblas S, van der Voet JC, et al. Bladder conservation in selected T1G3 and muscle-invasive T2-T3a bladder carcinoma using combination therapy of surgery and iridium192 implantation. Br J Urol 1994;74:322-327. 169. Barringer BS. Twenty-five years of radon treatment of cancer of the bladder. JAMA 1947;135:616-618. 170. Munro AI. The results of using radioactive gold grains in the treatment of bladder growths. Br J Urol 1964;36:541-548. 171. van der Werf-Messing BHP. Cancer of the urinary bladder treated by interstitial radium implant. Int J Radiat Oncol Biol Phys 1978;4:373-378. 172. Pernot M, Hubert J, Guillemin F, et al. Combined surgery and brachytherapy in the treatment of some cancers of the bladder (partial cystectomy and interstitial iridium-192). Radiother Oncol 1996;38:115-120. 173. Van Poppel H, Lievens Y, Limbergen EV, et al. Brachytherapy with iridium-192 for bladder cancer. Eur Urol 2000;37:  605-608. 174. Mackillop WJ. The principles of palliative radiotherapy: a radiation oncologist’s perspective. Can J Oncol 1996;6(Suppl):5-11.

649

175. Fossa SD, Hosbach G. Short-term moderate-dose pelvic radiotherapy of advanced bladder carcinoma. A questionnaire-based evaluation of its symptomatic effect. Acta Oncol 1991;30:  735-738. 176. Salminen E. Unconventional fractionation for palliative radiotherapy of urinary bladder cancer. A retrospective review of 94 patients. Acta Oncol 1992;31:449-454. 177. Holmang S, Borghede G. Early complications and survival following short-term palliative radiotherapy in invasive bladder carcinoma. J Urol 1996;155:100-102. 178. Wijkstrom H, Naslund I, Ekman P, et al. Short-term radiotherapy as palliative treatment in patients with transitional cell bladder cancer. Br J Urol 1991;67:74-78. 179. Duchesne GM, Bolger JJ, Griffiths GO, et al. A randomized trial of hypofractionated schedules of palliative radiotherapy in the management of bladder carcinoma: results of medical research council trial BA09. Int J Radiat Oncol Biol Phys 2000;47:  379-388. 180. Raghaven D. Systemic chemotherapy for metastatic cancer of the uro-epithelial tract. In Hall RR, ed: Clinical Management of Bladder Cancer. London: Arnold Publishing, 1999. 181. Garcia JA, Dreicer R. Systemic chemotherapy for advanced bladder cancer: update and controveries. J Clin Oncol 2006;24:  5545-5551. 182. Loehrer PJ, Einhorn LH, Elson PJ, et al. A randomized comparison of cisplatin alone or in combination with methotrexate, vinblastine, and doxorubicin in patients with metastatic urothelial carcinoma: a cooperative group study. J Clin Oncol 1992;10:1066-1073. 183. Saxman SB, Propert KJ, Einhorn LH, et al. Long-term follow-up of a phase III intergroup study of cisplatin alone or in combination with methotrexate, vinblastine, and doxorubicin in patients with metastatic urothelial carcinoma: a cooperative group study. J Clin Oncol 1997;15:2564-2569. 184. Mead GM, Russel M, Clark P, et al. A randomized trial comparing methotrexate and vinblastine (MV) with cisplatin, methotrexate and vinblastine (CMV) in advanced transitional cell carcinoma: results and a report on prognostic factors in a Medical Research Council study. MRC Advanced Bladder Cancer Working Party. Br J Cancer 1998;78:1067-1075. 185. Logothetis CJ, Dexus FH, Finn L, et al. A prospective randomized trial comparing MVAC and CISCA chemotherapy for patients with metastatic urothelial tumors. J Clin Oncol 1990;8:  1050-1056. 186. von der Maase H, Hansen SW, Roberts JT, 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;18:3068-3077. 187. von der Maase H, Sengelov L, Roberts JT, et al. Long-term survival results of a randomized trial comparing gemcitabine plus cisplatin, with methotrexate, vinblastine, doxorubicin, plus cisplatin in patients with bladder cancer. J Clin Oncol 2005;23:  4602-4608. 188. Larsen LE, Engelholm SA. The value of three-dimensional radiotherapy planning in advanced carcinoma of the urinary bladder based on computed tomography. Acta Oncol 1994;33:655-659. 189. Bentzen SM, Jessen KA, Jorgensen J, et al. Impact of CT-based treatment planning on radiation therapy of carcinoma of the bladder. Acta Radiol Oncol 1984;23:199-203. 190. Rothwell RI, Ash DV, Jones WG. Radiation treatment planning for bladder cancer: a comparison of cystogram localisation with computed tomography. Clin Radiol 1983;34:103-111. 191. Meijer G, Rasch C, Remeijer P, et al. Three-dimensional analysis of delineation errors, setup errors, and organ motion during radiotherapy of bladder cancer. Int J Radiat Oncol Biol Phys 2003;55:1277-1287. 192. Logue JP, Sharrock CL, Cowan RA, et al. Clinical variability of target volume description in conformal radiotherapy planning. Int J Radiat Oncol Biol Phys 1998;41:929-931.

650

PART 7  Urinary Tract

193. Turner SL, Swindell SL, Bowl N, et al. Bladder movement during radiation therapy for bladder cancer: implications for treatment planning. Int J Radiat Oncol Biol Phys 1997;39:355-360. 194. Muren L, Smaaland R, Dahl O. Organ motion, set-up variation and treatment margins in radical radiotherapy of urinary bladder cancer. Radiother Oncol 2003;69:291-304. 195. Muren L, Redpath A, McLaren D. Treatment margins and treatment fractionation in conformal radiotherapy of muscle-­invading urinary bladder cancer. Radiother Oncol 2004;71:65-71. 196. Jaffray DA, Siewerdsen JH, Wong JW, et al. Flat-panel conebeam computed tomography for image-guided radiation therapy. Int J Radiat Oncol Biol Phys 2002;53:1337-1349. 197. Hagan MP, Winter KA, Kaufman DS, et al. RTOG 97-06: initial report of a phase I-II trial of selective bladder conservation using TURBT, twice-daily accelerated irradiation sensitized with cisplatin, and adjuvant MCV combination chemotherapy. Int J Radiat Oncol Biol Phys 2003;57:665-672. 198. Pos FJ, Hulshof M, Lebesque J, et al. Adaptive radiotherapy for invasive bladder cancer: a feasibility study. Int J Radiat Oncol Biol Phys 2006;64:862-868. 199. Quilty PM, Kerr GR, Duncan W. Prognostic indices for bladder cancer: an analysis of patients with transitional cell carcinoma of the bladder primarily treated by radical megavoltage x-ray therapy. Radiother Oncol 1986;7:311-321. 200. Muren LP, Redpath AT, McLaren D, et al. A concomitant tumour boost in bladder irradiation: patient suitability and the potential of intensity-modulated radiotherapy. Radiother Oncol 2006;80:98-105. 201. Naslund I, Nilsson B, Littbrand B. Hyperfractionated radiotherapy of bladder cancer. A ten-year follow-up of a randomized clinical trial. Acta Oncol 1994;33:397-402. 202. Goldobenko G, Matveev B, Shipilov V, et al. Radiation treatment of bladder cancer using different fractionation regimens. Med Radiol (Mosk) 1991;36:14-16. 203. Stuschke M, Thames HD. Hyperfractionated radiotherapy of human tumors: overview of the randomized clinical trials. Int J Radiat Oncol Biol Phys 1997;37:259-267. 204. Kob D, Arndt J, Kriester A, et al. Results of percutaneous radiotherapy of bladder cancer using 1 and 2 series of irradiation. Strahlentherapie 1985;161:673-677. 205. Rostom A, Tahir S, Gershuny A, et al. Once weekly irradiation for carcinoma of the bladder. Int J Radiat Oncol Biol Phys 1996;35:289-292. 206. McLaren DB, Morrey D, Mason MD. Hypofractionated radiotherapy for muscle invasive bladder cancer in the elderly. Radiother Oncol 1997;43:171-174. 207. Scholten AN, Leer JW, Collins CD, et al. Hypofractionated radiotherapy for invasive bladder cancer. Radiother Oncol 1997;43:163-169. 208. Trott KR, Kummermehr J. What is known about tumour proliferation rates to choose between accelerated fractionation or hyperfractionation. Radiother Oncol 1985;3:1-9. 209. Rew DA, Thomas DJ, Coptcoat M, et al. Measurement of in vivo urological tumour cell kinetics using multiplanar flow cytometry. Preliminary study. Br J Urol 1991;68:44-48. 210. Hainau B, Dombernowsky P. Histology and cell proliferation in human bladder tumors. Cancer 1974;33:115-126. 211. Cole DJ, Durrant KR, Roberts JT, et al. A pilot study of accelerated fractionation in the radiotherapy of invasive carcinoma of the bladder. Br J Radiol 1992;65:792-798.

212. Moonen L, van der Voet H, Horenblas S, et al. A feasibility study of accelerated fractionation in radiotherapy of carcinoma of the urinary bladder. Int J Radiat Oncol Biol Phys 1997;37:537-542. 213. Plataniotis G, Michalopoulos E, Kouvaris J, et al. A feasibility study of partially accelerated radiotherapy for invasive bladder cancer. Radiother Oncol 1994;33:84-87. 214. Pos F, Tienhoven GV, Hulshof M, et al. Concomitant boost radiotherapy for muscle invasive bladder cancer. Radiother Oncol 2003;68:75-80. 215. Yavuz A, Yavuz M, Ozgur G, et al. Accelerated superfractionated radiotherapy with concomitant boost for invasive bladder cancer. Int J Radiat Oncol Biol Phys 2003;56:734-745. 216. Zouhair A, Ozsahin M, Schneider D, et al. Invasive bladder carcinoma: a pilot study of conservative treatment with accelerated radiotherapy and concomitant cisplatin. Int J Cancer 2001;96:350-355. 217. Horwich A, Dearnaley D, Huddart R, et al. A randomised trial of accelerated radiotherapy for localised invasive bladder cancer. Radiother Oncol 2005;75:34-43. 218. Mommsen S, Jakobsen A, Sell A. Quality of life in patients with advanced bladder cancer. A randomized study comparing cystectomy and irradiation—the Danish Bladder Cancer Study Group (DAVECA protocol 8201). Scand J Urol Nephrol 1989;125:  115-120. 219. Lynch WJ, Jenkins BJ, Fowler CG, et al. The quality of life after radical radiotherapy for bladder cancer. Br J Urol 1992;70:  519-521. 220. Caffo O, Fellin G, Graffer U, et al. Assessment of quality of life after cystectomy or conservative therapy for patients with infiltrating bladder carcinoma. A survey by a self-administered questionnaire. Cancer 1996;78:1089-1097. 221. Little FA, Howard GC. Sexual function following radical radiotherapy for bladder cancer. Radiother Oncol 1998;49:157-161. 222. Blandy JP, England HR, Evans SJ, et al. T3 bladder cancer—the case for salvage cystectomy. Br J Urol 1980;52:506-510. 223. Abratt RP, Wilson JA, Pontin AR, et al. Salvage cystectomy after radical irradiation for bladder cancer—prognostic factors and complications. Br J Urol 1993;72:756-760. 224. Crawford ED, Skinner DG. Salvage cystectomy after irradiation failure. J Urol 1980;123:32-34. 225. Konnak JW, Grossman HB. Salvage cystectomy following failed definitive radiation therapy for transitional cell carcinoma of bladder. Urology 1985;26:550-553. 226. Nurmi M, Valavaara R, Puntala P, et al. Single-stage salvage cystectomy: results and complications in 20 patients. Eur Urol 1989;16:89-91. 227. Swanson DA, von Eschenbach AC, Bracken RB, et al. Salvage cystectomy for bladder carcinoma. Cancer 1981;1;47:  2275-2279. 228. Osborn DE, Honan RP, Palmer MK, et al. Factors influencing salvage cystectomy results. Br J Urol 1982;54:122-125. 229. Bochner BH, Figueroa AJ, Skinner EC, et al. Salvage radical cystoprostatectomy and orthotopic urinary diversion following ­radiation failure. J Urol 1998;160:29-33. 230. Mannel RS, Manetta A, Buller RE, et al. Use of ileocecal continent urinary reservoir in patients with previous pelvic irradiation. Gynecol Oncol 1995;59:376-378.

The Testicle 27 Deborah A. Kuban, Marvin L. Meistrich, Min Rex Cheung, and David H. Hussey RESPONSE OF THE NORMAL TESTICLE TO IRRADIATION ANATOMY DEVELOPMENT AND DEVELOPMENTAL DEFECTS SPERMATOGENESIS Radiation Effects on Spermatogenesis Genetic Effects of Irradiation Effects of Irradiation on Testicular Endocrine and Paracrine Function Radiation Damage to the Somatic Environment within the Testis CANCER OF THE TESTICLE Incidence Pathology Histopathologic Classification

Differential Diagnosis Routes of Spread Staging Pretreatment Evaluation TREATMENT OF TESTICULAR CANCER Controversies Regarding Therapy for Pure Seminoma Recommended Treatment for Pure Seminoma Treatment for Nonseminomatous Testicular Tumors EXTRAGONADAL GERM CELL TUMORS TUMORS OF THE GONADAL STROMA MALIGNANT LYMPHOMAS

Response of the Normal Testicle to Irradiation The testis has long been an important area of study for ­radiation biology. This important organ has a self-renewing stem cell component, is readily accessible for local irradiation, and is composed of a variety of cell types with a wide range of radiosensitivities. The radiobiology of the testis has been studied from the perspectives of histology, sperm production, endocrine function, induction of heritable mutations, and fertility.1 Initial information regarding the radiosensitivity of the cells that make up the testis came from studies of experimental animals. Knowledge of the effects of radiation on human testes was limited because the testis is rarely directly irradiated in the clinical setting. However, the testes do receive a significant amount of scattered radiation when the pelvis, abdomen, or thigh is irradiated, and this is sometimes sufficient to cause infertility. Measurements of the functional response of the testis to measured levels of scattered dose have provided much of the information regarding the radiation response of the human testis. In this chapter, aspects of findings from animal studies are presented to illustrate basic principles that apply to humans, but the quantitative differences between animal and human testes should be borne in mind.

Anatomy The testis (Fig. 27-1) is an ovoid gland weighing 15 to 19 g, and it is surrounded by a thick fibrous capsule, the tunica albuginea. The testis and epididymis are enclosed in the tunica vaginalis, a double-layered capsule derived from

the peritoneum. The two layers of the tunica vaginalis slide freely on one another, enabling the testis to move easily within the scrotal sac. The testis is composed of a mass of convoluted seminiferous tubules1 that are arranged in wedge-shaped compartments or lobules formed by incomplete connective tissue septa. At the apex of the lobules, the seminiferous tubules unite to form the straight tubules that anastomose to form a network, the rete testis. In this region, 12 to 15 efferent ducts from the rete testis perforate the tunica albuginea to carry spermatozoa and secreted fluid to the head (caput) of the epididymis. The efferent ducts join to form the single duct of the epididymis, which abruptly bends at the tail (cauda) of the epididymis and continues as the vas deferens (see Fig. 27-1). The seminiferous tubules are 35 to 70 cm long and 170 μm in diameter. They contain multiple layers of germ cells (Fig. 27-2),2,3 all of which are in contact with the ­Sertoli cells, which serve nutritional, paracrine regulatory, and supportive functions for the developing germ cells. The seminiferous tubules lie within a stroma of connective tissue containing blood vessels and clumps of Leydig cells, which are responsible for the production of testosterone.

Development and Developmental Defects During the early phases of embryonic development, embryonic stem cells produce primordial germ cells. Those cells migrate to the genital ridge at 5 to 6 weeks after fertilization,4 and at about 8 weeks of fetal age, they become enclosed by the developing seminiferous cords and ­transform

653

654

PART 8  Male Genital Tract

into gonocytes.5 Later, at about 20 weeks’ gestation but sometimes continuing into the first year of postnatal development,6 these gonocytes differentiate into prespermatogonia. At that time, the germ cells lose their embryonic characteristics while acquiring the features necessary for entry into meiosis, and at that point, they are committed to producing cells that enter terminal meiotic divisions. Primordial germ cells or gonocytes that remain undifferentiated, because of direct effects on the embryonic germ cells themselves or disturbances in the microenvironment of the developing gonad, are identified as carcinoma in situ cells.

These embryonic-derived cells give rise to testicular germ cell tumors in young adults.

Spermatogenesis During puberty, the prespermatogonia mature into adult stem spermatogonia, which lie next to the basement membrane of the seminiferous tubule and are responsible for the production of cells involved in the complex process of spermatogenesis (see Fig. 27-2).2,7 The spermatogonia are the source of spermatocytes, which divide to become ­spermatids (Fig. 27-3).8 The spermatids transform into spermatozoa, which are released into the lumen of the ­tubule. Humans and nonhuman primates have two types of spermatogonia: A and B. Type A spermatogonia consist of two subpopulations7,9: type A dark spermatogonia, which under normal conditions are in a resting, nondividing phase and are considered reserve stem cells, and type A pale spermatogonia, which are more actively proliferating and are normally considered to be the self-renewing stem cells of the testis.10 When type A pale cells divide, their daughter cells remain as type A pale spermatogonia or divide again to become type B spermatogonia. When the population of type A pale spermatogonia cells is diminished, the type A dark spermatogonia can replenish the supply of type A pale spermatogonia. In rodents, the stem cell population seems to be different in that the cells are slowly proliferating, isolated, undifferentiated type A spermatogonia.10,11 These cells are self-renewing; some of their daughter cells remain as isolated stem cells, but others are committed to differentiation, when they form pairs and eventually chains of type A spermatogonia. These chains of spermatogonia morphologically differentiate into type A1 spermatogonia, which undergo five cell divisions to eventually produce type B spermatogonia.

Efferent ducts Epididymis Tunica albuginea

Ductus deferens

Rete testis

Seminiferous tubules

Straight tubules

Epididymis

Tunica vaginalis

FIGURE 27-1.  Diagram of vertical section of a testis and the reproductive tract.

Lumen

Elongated spermatid

Round spermatid Pachytene primary spermatocyte

Sertoli cell Stem spermatoganium Sertoli cell junction

Differentiating spermatogonium

Lymphati c

Leydig cell

space

Vasculature

Extracellular matrix Myoid cell

Lymphatic endothelium

FIGURE 27-2.  Diagram of a normal seminiferous tubule and major elements of the interstitial area. (From Meistrich ML. Relationship between spermatogonial stem cell survival and testis function after cytotoxic therapy. Br J Cancer 1986; 53:89-101.)

ra ds

Sp er m Spermatozoa > 60,000 rads

40

es

Ais Stem cell

yt

480 rads Pachytene

A2

< 100 rads Aal

oc

Zygotene

A4 A3

200–300 rads

at m

In

er

B

200 rads prelepto- Leptotene tene 520 rads

655

Sp

at og on ia

27  The Testicle

7.4 days

A1

560 rads

13.7 days

5.3 days

16

Diplotene

14.1 days 15 14 13

12

11 10 9 8 7 6

5

4

Secondary Step 1 2 3

>1 5 Spe 00 rad rma s tids

Sperm heads LD50 rads

FIGURE 27-3.  Kinetics and radiation sensitivities of spermatogenic cells in the mouse. Radiation sensitivity is given for most cells as the median lethal dose to 50% of the cells (LD50); the value for the stem cells is the threshold dose on the linear high-dose part of the survival curve (D0), which is the dose necessary to further reduce cell survival by 73%. (Modified from Meistrich ML, Hunter NR, Suzuki N, et al. Gradual regeneration of mouse testicular stem cells after exposure to ionizing radiation. Radiat Res 1978;74:349-362.)

In humans and rodents, the type B spermatogonia undergo one more division to give rise to primary spermatocytes, which replicate their deoxyribonucleic acid (DNA) and move toward the lumen of the seminiferous tubule. The DNA in the primary spermatocytes undergoes genetic recombination in preparation for the meiotic divisions. The first meiotic division yields secondary spermatocytes, and shortly thereafter, the second meiotic division yields spermatids, which have a haploid chromosome number and a haploid DNA content. The spermatids then undergo a process of metamorphosis in which the nuclear material condenses and elongates and the tail forms, thereby yielding mature spermatozoa. The spermatozoa then pass through the rete testis, efferent ducts, epididymis, and vas deferens, where they are stored for ejaculation. In a human male, the time required for type A spermatogonia to develop into mature spermatozoa in the ejaculate is about 67 days.12 Type B spermatogonia require 58 to 67 days to become spermatozoa, spermatocytes between 34 and 58 days, and spermatids between 12 and 34 days.13

Radiation Effects on Spermatogenesis Radiosensitivity of Germ Cells The sensitivity of germ cells to a given dose of radiation is related to the stage they are in at the time they are irradiated (see Fig. 27-3).8,14,15 In rodents and humans, type B spermatogonia are the most radiosensitive, type A spermatogonia are somewhat less sensitive, and primary and secondary spermatocytes and spermatids are much less sensitive (i.e., more resistant) than either of the spermatogonial cell types.14-16 The extreme radiosensitivity of the more differentiated type A and type B spermatogonia

r­ esults from radiation-induced apoptosis.17 Radiosensitivity generally diminishes as the cells progress along the maturation pathway, except for an increase in sensitivity as type A spermatogonial cells differentiate into other types of spermatogonia.8 The radiosensitivity of the spermatogonia was revealed by studies of the effects of single-dose irradiation of human testes14; these cells displayed morphologic and quantitative signs of injury after single doses as low as 0.1 Gy.16 Although spermatocytes had no visible signs of injury even after single doses of 2 to 3 Gy, they did suffer damage because they were unable to complete their meiotic divisions and produce spermatids. Single doses of 4 to 6 Gy did damage spermatocytes morphologically and affected the ability of spermatids to produce spermatozoa.

Effects of Fractionation Spermatogonial stem cells have long been known to be more sensitive to fractionated irradiation than they are to single doses. This seemingly paradoxical observation was first observed by Regaud in his classic study of rams’ testicles.18 Regaud found that suppression of spermatogenesis was more severe when the radiation dose was delivered over 28 days than when similar or even greater doses were given over 30 to 42 hours. However, the more fractionated schedule resulted in less damage to scrotal skin, as is typical of all other tissues. Similar results have been obtained in a variety of different species.19 Single exposures as large as 1000 roentgen (R, representing an absorbed dose of approximately 10 Gy) failed to sterilize canine testicles, whereas total cumulative exposures as low as 475 R caused permanent azoospermia in dogs when delivered in small daily fractions over 32 weeks.20 The recovery of

656

PART 8  Male Genital Tract

s­ permatogenesis from surviving stem cells in some strains of mice is also slower when the same dose is delivered in two daily fractions instead of a single fraction.7 Although the responses are qualitatively similar, the doses necessary to produce specific levels of damage to the human testis seem to be greater when the radiation is fractionated over 3 to 7 weeks than when it is given as a single dose.22 We discuss results from different dose schedules in separate sections in this chapter. The mechanism underlying the sensitizing effect of fractionation is still not completely understood. During fractionated irradiation, spermatogonial stem cells may be recruited into a more radiosensitive phase, although radiosensitivity seems to be inversely related to the proliferative state of these cells.23 The sensitizing effect of fractionation must exceed the capacity for repair of sublethal damage21,23 and repopulation.8,24

Depletion of Germ Cells Irradiation does not affect the kinetics of spermatogenic cell differentiation, and the surviving germ cells differentiate with kinetics identical to cells in the normal testis.25 As a result, sperm production is maintained from surviving, more radioresistant cells until the radiosensitive cells would have become sperm. The sperm counts of mice26 and men20 are not affected at first (until 30 and 50 days, respectively) by single doses of about 1 Gy, which would kill many of the differentiating spermatogonia but few of the other germ cells. The magnitude of the decline in sperm count is related to radiation dose. Single doses of less than 0.1 Gy do not significantly decrease sperm count,16,27 whereas doses between 0.1 and 0.7 Gy reduce sperm counts. Minimum sperm counts are about 1 million/mL after 0.5 Gy. Doses of 0.8 Gy or more lead to total azoospermia. Similar results have been reported after radiation given over 3- to 7-week periods during fractionated ­ radiation therapy.28 Cumulative doses to the testis of less than 0.15 Gy have little effect on sperm count. Sperm counts are significantly reduced at doses between 0.15 and 0.35 Gy, and 50% of men may experience transient azoospermia after 0.35 Gy.22 Almost all men become azoospermic after 1 Gy to the testis.

Recovery of Spermatogenesis after Irradiation Recovery of spermatogenesis after irradiation involves regeneration of stem cell numbers from the surviving stem cells and repopulation of the seminiferous tubules with maturing cells. In the mouse, stem cells initially undergo only self-renewing divisions, but within 10 days, the combination of continued increase in stem cell numbers and differentiation29 results in increases in sperm counts within 6 weeks after irradiation.26 Sperm count can recover to 80% of control values within 20 weeks after a dose of 6 Gy. Some recovery is observed even after 16 Gy. The recovery of spermatogenesis from stem cells is affected more by radiation in the rat than it is in the mouse. In certain strains, doses as low as 3.5 Gy produce azoospermia lasting more than 1 year.30 In this situation, stem cells survive and proliferate but undergo apoptosis when they attempt to differentiate beyond the type A spermatogonial stage because of damage to the stromal environment within the testis.31,32 Transient suppression of testosterone

results in restoration of the ability of the testis to support spermatogenic differentiation, recovery of sperm production, and restoration of fertility by a mechanism that is ­unknown.33 In humans, the stem spermatogonia seem to be more sensitive to killing than in rodents, and recovery of spermatogenesis is delayed.27 After single doses of radiation, the numbers of stem and differentiated cells in humans progressively decrease for 7 months after doses of 1 to 6 Gy,16 after which recovery of germ cell numbers begins.22,27 Although sperm may be present in the semen of some men at 7 months after irradiation with 1 Gy, 24 months are required for the appearance of sperm after 6 Gy.16 The causes of this block in sperm production and the trigger for the apparently spontaneous recovery of spermatogenesis are unknown. Whether transient hormone changes can stimulate recovery of spermatogenesis after moderate doses of radiation in humans, as they can in rats, remains to be tested,34 although negative results have been reported in a nonhuman primate model.35 The estimated single dose required for permanent sterilization in normal humans is between 6 and 9.5 Gy.20 Spermatogenesis occurred within 5 years (although it can occur at as late as 9 years) in about 15% of men who received a 10-Gy single-dose total-body exposure (testicular dose, 8 Gy) or 12- to 16-Gy exposures (testicular doses, 10 to 13 Gy) fractionated over 6 or 7 days as preparation for bone marrow transplantation.36,37 In these studies, the testis did not seem to be any more sensitive to fractionation over 6 or 7 days than to single-dose irradiation. When the radiation is fractionated over periods of 3 to 7 weeks, recovery to maximal counts usually requires ­between 1 and 2 years after gonadal doses of less than 1 Gy, although in some individuals, recovery is more gradual. After doses between 1 and 2 Gy, 2 to 6 years are ­required for full recovery.28,38 After higher doses of fractionated irradiation, recovery is sporadic but can occur after ­doses in the range of 2.3 to 3 Gy.39,40 The upper dose limit given in these fractionation regimens, above which permanent azoospermia is probable, is considered to be about 2.5 Gy, compared with more than 6 Gy for single-dose ­ irradiation.22 The sperm count usually recovers, at least after lower doses, to normospermic levels (>20 million cells/mL) or to levels normal for an individual before irradiation.28 However, after higher doses, recovery in some men is limited to oligospermic levels (<20 million cells/mL).41,42 Although no absolute boundary of sperm count (aside from azoospermia) separates fertile from infertile men, the risk of infertility in the general population progressively increases as sperm counts decline below 20 million cells/mL.43 However, fertility in patients whose sperm counts recover after cancer therapy has been observed for men with counts as low as 1 million cells/mL, probably because sperm motility returns to normal despite increases in numbers of sperm with morphologic abnormalities.37

Testicular Doses from Radiotherapeutic Procedures In the typical patient being treated with megavoltage irradiation with a “hockey-stick” field to irradiate the periaortic and ipsilateral iliac lymph nodes for pure seminoma, the dose to the remaining testis, outside the radiation field, is in the range of 0.3 to 1.8 Gy.40,44 An even greater dose

27  The Testicle

328 160

T

120

P 285

Tumor dose = 3650 RAD

FIGURE 27-4.  Individually designed Cerrobend testicular shield with a balanced lid to prevent pinching of scrotum as it drops over the lip of the base. Total testicular dose was 120 to 160 cGy, for a total dose of 36.5 Gy to the midpelvis and inguinal nodes. Doses measured on the perineum outside the scrotal shield were 3.28 Gy and 2.85 Gy. (From Million RR. The lymphomatous diseases. In Fletcher GH [ed]: Textbook of Radiotherapy, 3rd ed. Philadelphia: Lea & Febiger, 1980.)

may be delivered to the contralateral testis if the hemiscrotum is irradiated. In the treatment of Hodgkin disease, use of periaortic-spleen or abdominal fields results in doses of 0.09 to 0.32 Gy to the testes.28,44 Doses from pelvic fields are about 2 Gy.45 Whenever possible, a testicular shield should be used in men undergoing pelvic irradiation (Fig. 27-4).46 This apparatus works principally by shielding the testicles from radiation scattered within the patient. The use of shielding devices can reduce testicular exposure to approximately 1% of the dose delivered to the midpelvis. A device to position and shield the remaining testicle should be used for hemiscrotal irradiation.47

Genetic Effects of Irradiation Anyone working with radiation should be aware that ionizing radiation produces genetic mutations in germ cells of experimental animals and in somatic cells in humans. The frequency of the mutations depends on the radiation dose delivered, and the relationship is often a straight line without a threshold. Radiation-induced mutations are almost always harmful. Although no conclusive data have been found to indicate that radiation can induce heritable mutations in human germ cells,48 such mutations are most likely produced, but at a frequency that falls below the level that can be detected in epidemiologic or molecular studies with present techniques. In male rodents, although meiotic and postmeiotic cells are resistant to killing by radiation, they are susceptible to the induction of mutations. The induction of mutations, which can be lethal to embryos and transmitted to offspring, is higher in later-stage germ cells than in stem spermatogonia.49 Because mutations induced in spermatocytes and spermatids are eliminated through the differentiation and maturation processes, the mutations produce mutation-carrying sperm for only a brief time. However, mutations induced in stem cells may result in the production of abnormal spermatozoa for a long time. Fortunately,

657

the frequency of transmissible mutations is much lower in spermatozoa arising from these stem cells. Patients should also be advised against having children starting when the irradiation treatments begin (especially during the first 2 months, when fertility is likely to be maintained) until 6 months after the end of all cytotoxic therapy. By then, the body has eliminated spermatozoa derived from cells that were beyond the stem cell stage during radiation therapy. No evidence exists to suggest any significant increase in birth defects or other genetically related abnormalities in offspring conceived 6 months or more after radiation therapy.50

Effects of Irradiation on Testicular Endocrine and Paracrine Function Testicular tissue is complex, consisting of germ cells, vascular cells, Leydig cells, macrophages in the interstitial regions, and peritubular myoid and Sertoli cells, which comprise the seminiferous tubules (see Fig. 27-2). These cells interact by paracrine mechanisms and have endocrine functions, secreting hormonally active molecules into the circulation.

Sertoli Cells The Sertoli cells are directly involved in the support of spermatogenesis by providing paracrine signals to all germ cells. Sertoli cells also provide nutrients to developing spermatocytes and spermatids, isolate them from bloodborne molecules, and facilitate their movement toward the ­lumen of the seminiferous tubule. Optimal Sertoli cell function depends on the action of testosterone and folliclestimulating hormone from the pituitary. Sertoli cells also produce and secrete androgen-binding protein, which acts as a reservoir and transport molecule for testosterone; inhibin, which suppresses the release of follicle-stimulating hormone; and stem cell factor, which stimulates spermatogonial differentiation. Although Sertoli cells are postmitotic cells and do not seem to be killed (in adults) by radiation doses of up to 20 Gy, their function is altered by the loss of germ cells and possible direct radiation effects. One manifestation of altered Sertoli cell function is increased follicle-stimulating hormone levels, which result from decreased inhibin production by the Sertoli cells, but this is a secondary consequence of the loss of germ cells.16,51 Direct effects of radiation on Sertoli cells have been demonstrated in vitro; secretion of interleukin-6 by Sertoli cells is enhanced by radiation doses as low as 3 Gy.52

Leydig Cells Leydig cells are the major source of testosterone. Luteinizing hormone (LH) from the anterior pituitary stimulates Leydig cells to form testosterone, which diffuses into the general circulation and into the seminiferous tubules to stimulate spermatogenesis. High testosterone levels suppress LH release through a negative-feedback mechanism. Increased serum LH levels, sometimes accompanied by decreased serum testosterone levels, have been observed in humans after doses of 2 Gy and in rats after doses of 5 Gy.53,54 Although such increases in LH have been considered to indicate Leydig cell dysfunction after irradiation, that belief is not necessarily valid. These doses cause germinal aplasia, which reduces testis size, and a concomitant

658

PART 8  Male Genital Tract

reduction in testicular blood flow.55 With reduced blood flow, less testosterone is distributed into the circulation, and the Leydig cells are exposed to less LH in the serum. These changes in hemodynamics reset the pituitary-gonadal feedback loop and elevate the levels of serum LH necessary to maintain serum testosterone in the normal range. The active production of testosterone by Leydig cells after these radiation doses has been demonstrated by high intratesticular testosterone concentrations in irradiated rats and nonhuman primates.35,36 Radiation therapy reduces testosterone levels to the low-normal range in some men,57 but most men whose testosterone levels are low to normal after cancer therapy show little benefit from androgen replacement.58 High doses of radiation (20 to 30 Gy), such as those used for treatment or prophylaxis in testicular leukemia or testicular carcinoma in situ, damage Leydig cells and markedly reduce testosterone levels even in adult men.59,60 In rats, the Leydig cells are much more sensitive to irradiation during puberty than during adulthood.53 Similarly, an inverse age dependence of Leydig cell damage is evident in humans; in boys, doses as low as 12 Gy cause a high incidence of failure of testicular function and of normal progression through puberty, resulting in extremely low testosterone levels.61,62

Radiation Damage to the Somatic Environment within the Testis Although it has been difficult to pinpoint the target for damage, the failure of recovery of spermatogenesis from surviving stem cells in irradiated inbred rats has been shown to be caused by changes in the somatic environment, as demonstrated by reciprocal spermatogonial transplantation experiments.32 When spermatogonia from irradiated rats were transplanted into the testicular tubules of nude mice depleted of endogenous germ cells, the spermatogonia differentiated normally; however, when spermatogonia from normal rats were transplanted into the tubules of the irradiated rats, they colonized along the basement membrane but failed to differentiate to later stages. These results are important in the context of attempts to use spermatogonial stem cell transplantation to restore male fertility after cytotoxic treatments that deplete the testis of germ cells,63 particularly when prepubertal boys are treated and sperm banking is not an option. Preclinical tests of this approach have had promising results; in one study, transplantation of cryopreserved spermatogonia stem cells from an immature mouse to the testes of a mouse of the same strain whose germ cells had been eliminated by a cancer chemotherapeutic drug was successful in restoring fertility in the recipient mouse.64 Preliminary evidence suggests that autologous transplantation of spermatogonial stem cells can also restore spermatogenesis in the irradiated testes of nonhuman primates.65 However, damage to the somatic environment may prevent the transplanted stem cells from developing into sperm. Other hormonal approaches and transplantation of testicular somatic cells are being explored in the hope of overcoming such damage.32

Cancer of the Testicle Significant changes in the treatment of cancer of the testis have occurred over the past several decades. These changes are mainly the result of the development of effective

c­ hemotherapy regimens for nonseminomatous cancer and new imaging techniques and tumor markers for cancer detection and staging. These advances have made testicular cancer one of the most curable of all cancer types, with disease-specific 5-year survival rates of about 95%.66 The discovery of effective chemotherapeutic agents for germ cell tumors has also affected the indications for radiation therapy in the management of cancer of the testis. Although radiation therapy still plays a major role in the treatment of patients with pure seminoma, its role in the management of nonseminomatous tumors has diminished considerably. This chapter focuses on the current status of radiation therapy in the management of the primary tumors of the testis, including seminomas, nonseminomas, stromal cell tumors, and lymphomas.

Incidence Testicular cancer is an uncommon tumor, with only about 7000 new cases diagnosed each year in the United States.66 The number of new cases, however, has increased significantly over the past 30 years, from 3.2 cases per 100,000 in 1976 to 4.6 cases per 100,000 in 1996 and to 5.7 cases per 100,000 in 2004. Testis cancer is the most common cancer in males between the ages of 15 and 35. However, the incidence peaks somewhat later for seminomas (35 to 40 years) than for nonseminomas (25 to 30 years).67 Testis cancer is much more common in whites (7.0 cases per 100,000 in 2004) than in blacks (1.0 cases per 100,000 in 2004).66 Considerable variation is evident in the incidence of testis cancer geographically. The Scandinavian countries, for example, report almost 10 times as many new cases per capita as do Japan or Puerto Rico. The populations that have the highest incidence of seminoma also have the highest incidence of nonseminoma and vice versa.68 A slight familial predilection is evident for this disease; testicular cancer is somewhat more common in identical twins or family members of a patient with a testis tumor than it is in men without a family history of this disease. Approximately 10% of patients with testicular cancer have a history of cryptorchidism,69,70 and patients with cryptorchidism have about 35 times the risk of testicular cancer as patients with normally descended testicles.71 The risk of cancer in a cryptorchid testis is directly related to the degree of maldescent. The incidence is approximately 1 in 20 if the testis is located intra-abdominally and 1 in 80 if it is located in the inguinal canal. Testicular cancer is slightly more common in the right testis than the left, and this differential correlates with the increased incidence of cryptorchidism in the right testis.72 Most investigators believe that the high incidence of germ cell malignancies in patients with cryptorchidism is related to a developmental defect.73 Evidence for this opinion includes the observations that a significant percentage (5% to 10%) of the cases of cancer associated with cryptorchidism occur in the contralateral, normally descended testis, and that approximately one fourth of patients who have a testicular cancer in a cryptorchid testis have carcinoma in situ in the other testis.74 Overall, approximately 1% to 3% of testicular cancer cases are bilateral, and more than one half of patients with bilateral tumors have a history of cryptorchidism.75 Approximately 5% of patients with germ cell tumors of the

27  The Testicle

testis have carcinoma in situ in the contralateral testis, and in many of them, invasive cancer eventually develops in the second testis.76,77 Bilateral testicular cancer may occur synchronously or metachronously, with the second cancer usually appearing within 2 years of the original diagnosis. However, cancer in the second testis has occurred up to 15 years after the first cancer.

Pathology About 95% of cases of primary cancer of the testis are germ cell tumors. They are usually divided into seminomas and nonseminomatous categories for the purpose of treatment decisions. Of these, approximately one third are seminomas, and two thirds are nonseminomatous tumors. Less than 5% of testicular malignancies originate from the gonadal stroma (e.g., Sertoli cell tumors, Leydig cell tumors, gonadoblastomas). Sertoli cell tumors and Leydig cell tumors are usually benign and are most often treated surgically. Other tumors that can originate in the testis include lymphomas (usually diffuse large cell lymphomas), melanomas, and rhabdomyosarcomas. The tumors that are of the most interest to radiation oncologists are the germ cell tumors and, to a lesser extent, lymphomas and rhabdomyosarcomas.

Histopathologic Classification The histopathologic classifications of germ cell tumors are different in Great Britain and the United States. Each is based on a theory of germ cell tumor development, and several theories focus on the histogenesis of germ cell tumors. The British histopathologic classification is based on the model of Willis,78 who proposed that all nonseminomatous tumors are teratomas that have descended from displaced blastomeres. The classification used in the United States is based on the model proposed by Friedman and Moore,79 Dixon and Moore,80 and Mostofi81 (Fig. 27-5). These investigators proposed that primordial germ cells divide into two groups of cells when they are fertilized: extraembryonic or trophoblastic cells and embryonic or somatic cells (see

Germ cell Embryonal (totipotential) Somatic

Trophoblastic

Seminoma Carcinoma

Ectoderm Endoderm Mesoderm

Teratoma

Choriocarcinoma

FIGURE 27-5.  Diagram illustrating the histogenesis of the germ cell tumors of the testis.

659

Fig. 27-5). The trophoblastic cells can implant into the maternal uterus just as the placenta does during normal ­development. The embryonic cells, however, may take one of several paths. One cell becomes identifiable as the germ cell early in embryogenesis, and the others differentiate into somatic cells (i.e., ectoderm, mesoderm, and endoderm), which form the various organs of the embryo. Primordial germ cells have the potential to differentiate into aggressive cells (i.e., trophoblasts) or into somatic cells, which later develop into mature tissues and organs. Germ cell tumors are believed to be able to develop anywhere along this pathway. Seminomas are thought to originate from the germinal epithelium of the seminiferous tubules. Evidence that supports this hypothesis includes the observation that seminoma cells are morphologically similar to spermatogonia and that carcinoma in situ is often found within the seminiferous vessels in the early stage of development of a seminoma. According to this theory, nonseminomatous germ cell tumors arise from the primordial cells or their progeny.82,83 Embryonal carcinomas are believed to originate from totipotential primordial germ cells, whereas the other germ cell tumors are thought to develop further along the pathway (see Fig. 27-5). If the embryonal cells develop intraembryonically along somatic lines, teratomas or teratocarcinomas may develop, and if they develop along extraembryonic pathways, choriocarcinomas or yolk sac tumors may be formed. The U.S. and British histopathologic classification systems are not easily compared. Seminomas and spermatocytic seminomas are identical in the two classifications, but differences exist in the way the nonseminomatous germ cell tumors are organized.

Seminoma: Dixon-Moore Category I Seminomas make up 35% of all germ cell tumors of the testis. They tend to occur in a slightly older age group than do nonseminomatous tumors. Grossly, seminomas manifest as pale gray to yellow nodules of a uniform or slightly lobulated consistency. Three histologic subtypes have been described: classic, anaplastic, and spermatocytic. Classic Seminomas. Classic seminomas are composed of a monotonous sheet of seminoma cells, large cells containing a clear cytoplasm and a hyperchromatic nucleus. Syncytiotrophoblasts or foreign body giant cells may be seen, but mitoses are infrequent. The syncytiotrophoblastic elements are seen in 10% to 15% of cases, an incidence that corresponds with the incidence of β-human chorionic gonadotropin (β-HCG) elevations in seminomas. Approximately 85% of pure seminomas are of the classic type. Anaplastic Seminomas. Anaplastic seminomas are characterized by marked nuclear pleomorphism and higher mitotic rates than classic seminomas. The diagnosis is made by finding three or more mitotic figures per high-power microscopic field. The radiosensitivity of anaplastic seminomas is similar to that of classic seminomas, but patients with anaplastic seminomas usually present with disease of a higher clinical stage.84 Approximately 10% to 12% of seminomas are anaplastic. Spermatocytic Seminomas. Spermatocytic seminomas contain cells resembling secondary spermatocytes and spermatids. They tend to occur in elderly men and

660

PART 8  Male Genital Tract

are characterized by an excellent prognosis. Spermatocytic seminomas account for 4% to 5% of all seminomas.

Embryonal Carcinoma: Dixon-Moore Category II Embryonal carcinomas have a grayish white, slightly variegated appearance with areas of hemorrhage and necrosis. Histologically, they contain cells that resemble malignant epithelial cells, with considerable pleomorphism. The nuclei vary in shape, and mitotic figures are common. The infantile variety is known as orchioblastoma, yolk sac tumor, or endodermal sinus tumor. Approximately 20% of germ cell tumors of the testis are embryonal carcinomas.

Lumbar nodes

Teratoma: Dixon-Moore Category III Teratomas contain more than one germ cell layer in various stages of maturation. Grossly, they may contain cysts filled with clear, gelatinous or mucinous material along with various amounts of solid tissue. They may contain squamous or neuronal tissue of ectodermal origin, gastrointestinal or respiratory tissue of endodermal origin, and cartilage, muscle, or bone of mesodermal origin. Teratomas are benign in infants, but they may metastasize in adults. They account for approximately 5% of all germ cell tumors.

Choriocarcinoma: Dixon-Moore Category V Choriocarcinomas are highly malignant tumors that tend to metastasize early in the course of the disease. They are composed entirely of syncytiotrophoblasts and ­ cytotrophoblasts, compared with mixed tumors with foci of choriocarcinoma, which have only scattered ­syncytiotrophoblasts. Choriocarcinomas are rare, making up less than 1% of all germ cell tumors.

Mixed Tumors Almost 40% of germ cell tumors contain more than one histopathologic type. The most common type of mixed tumor is teratocarcinoma (Dixon-Moore category IV), which is a combination of teratoma and embryonal carcinoma, with or without seminoma.80

Differential Diagnosis The differential diagnosis of a testis tumor includes epididymitis, hydrocele, spermatocele, hematocele, granulomatous orchitis, and varicocele. Epididymitis usually manifests as an enlarged tender epididymis that is clearly separable from the testis, and it usually has an acute onset of symptoms such as fever and urethral discharge. It is usually identifiable on ultrasonography. A hydrocele can usually be transilluminated and is easily distinguished as a fluid-filled mass on scrotal sonography.

Routes of Spread Local Extension A testicular cancer usually manifests as a painless mass within the testis. It may spread locally to the rete testis, epididymis, and spermatic cord. However, this spread occurs in only 10% to 15% of patients.85 Testis cancer can also extend through the tunica albuginea into the scrotum, but this is even more uncommon (<5%) because the tunica albuginea forms a natural barrier to direct extension.85,86 This is important with regard to management, because the scrotum rarely requires treatment.

External iliac nodes

FIGURE 27-6.  The lymphatics of the testis ascend and anastomose along the spermatic cord. Most collecting trunks accompany the internal spermatic artery and vein to the point where these blood vessels cross the ureter and then separate from the internal spermatic vessels by turning medially to terminate in the lumbar lymph nodes. The nodal connections are different on the right and left. On the right, most channels end in the lateral aortic, precaval, or preaortic nodes situated between the renal vein and the aortic bifurcation. In about 10% of patients, some of the trunks end in a node situated at the junction of the renal vein and inferior vena cava. On the left, most collecting trunks drain into the lateral aortic nodes situated below the left renal vein, usually terminating in the superior nodes of this group. However, they may also end in the preaortic nodes or in nodes at the level of the aortic bifurcation. On both sides, some of the lymphatic channels abandon the spermatic vessels after they reach the vesical peritoneum, ending in nodes lying along the external iliac vein. (From Weiss L, Gilbert HA, Ballon SC [eds]. Lymphatic System Metastasis, vol III. Boston: GK Hall, 1980.)

Lymphatic Metastasis Embryologically, the testes originate from the genital ridge near the second lumbar vertebra before migrating into the scrotum. As they do so, they carry along their own vascular and lymphatic supply. The primary collecting lymphatic trunks follow the internal spermatic vessels to the periaortic area (Fig. 27-6).87 The channels swing out wide laterally, particularly on the left (Fig. 27-7). The primary drainage from the right and left testis is different. The lymphatics from the right testis tend to drain to lymph nodes located along the inferior vena cava from L2 to L5. However, these vessels can cross over directly to terminate in the contralateral left periaortic lymph nodes. The lymphatics from the left testis tend to drain to nodes located just below the renal vein (Fig. 27-8).88 However, they can also drain to lymph nodes located elsewhere in the left periaortic area. Unlike the lymphatic vessels from the right testis, the vessels from the left testis do not cross

27  The Testicle

A

661

B

FIGURE 27-7.  Direct filling of periaortic lymph nodes draining the testicle is demonstrated by testicular lymphangiograms of the right (A) and left (B) sides. The iliac nodes are bypassed. (Courtesy F. M. Busch, MD, San Diego, CA.) Right testis

Solitary metastasis Multiple metastases

Left testis

FIGURE 27-8.  Distribution of lymph node metastases found at lymphadenectomy for nonseminomatous tumors. (Data from Ray B, Hajdu SI, Whitmore W Jr. Proceedings: distribution of retroperitoneal lymph node metastases in testicular germinal tumors. Cancer 1974;33:340-348.)

over directly to the contralateral side. The lymphatics from either testis can extend to periaortic nodes located above the renal hila or to nodes situated at the level of the aortic bifurcation. On both sides, some of the lymphatic channels abandon the spermatic vessels after they reach the vesicle peritoneum, ending in nodes lying along the external iliac vein (see Fig. 27-6). This lymphatic drainage is usually ipsilateral unless the patient has had groin surgery or extensive retroperitoneal metastasis leading to retrograde spread. Metastasis to the lymph nodes above the diaphragm is rare in the absence of involvement of the periaortic lymph

nodes. However, supraclavicular metastasis may occur in the absence of mediastinal disease because the main route of lymphatic spread from the periaortic area is through the thoracic duct. Supraclavicular metastases are usually located on the left side, near the junction of the thoracic duct and the left subclavian vein.

Hematogenous Metastasis The regional lymph nodes are almost always the first site of metastasis for a pure seminoma, and hematogenous metastasis is uncommon for this disease. Embryonal carcinomas and teratocarcinomas also tend to spread to the lymph

662

PART 8  Male Genital Tract

nodes first, but in those tumors, hematogenous metastasis is not uncommon. Choriocarcinomas almost always have hematogenous metastasis by the time the diagnosis is established. The lungs are usually the first site of hematogenous spread from a testicular tumor.

Staging The results achieved with radiation therapy are reported in terms of clinical staging because the pathologic status of the retroperitoneal lymph nodes is not known in those cases. In general, the results are better, stage for stage, when a series is reported in terms of pathologic staging than when they are reported in terms of clinical staging. This is true for stage I and stage II cancer because patients whose disease is clinically stage I but pathologically stage II have less extensive retroperitoneal disease than those who have both clinically and pathologically stage II disease. These differences must be considered when comparing the results of surgical series with those obtained with radiation therapy. Several staging systems have been used for testis cancer. The classification that was used for testis cancer for many years was first proposed by Boden and Gibb in 1951, and a modification of this classification is shown here: Stage I Tumor clinically limited to the testis and spermatic cord Stage II Clinical or radiographic evidence of tumor spread beyond the testis but limited to the regional nodes below the diaphragm Stage IIA Moderate-size retroperitoneal metastasis Stage IIB Massive retroperitoneal metastasis Stage III Extension above the diaphragm Stage IIIA Extension above the diaphragm but still confined to the mediastinal or supraclavicular lymphatics Stage IIIB Extranodal metastasis The stage distributions are different for seminomas and nonseminomas because the patterns of spread are different.89-91 In a series of patients seen at The University of Texas M. D. Anderson Cancer Center, 32% of patients with nonseminomatous tumors had clinical evidence of hematogenous spread (Royal Marsden stage IIIB) at the time of diagnosis, compared with only 6% of those with pure seminomas.90 However, the percentage of patients with bulky retroperitoneal disease (Royal Marsden stage IID) was significantly greater for those with pure seminomas than it was for those with nonseminomas (16% versus 6%). This is because nonseminomatous tumors are more likely to have metastasized hematogenously by the time the retroperitoneal disease becomes large. Several other staging systems have been proposed.89,91 They describe the extent of the tumor at the primary site, in the regional lymphatics, and in distant organs in more detail. The two most widely used staging systems are those of the American Joint Committee on Cancer (AJCC) and the Royal Marsden Hospital. The AJCC system uses the tumor-node-metastasis (TNM) notation, whereas the Royal Marsden system closely resembles the original Boden and Gibb classification. The AJCC and Royal Marsden classifications are compared in Box 27-1. The Royal Marsden classification is used in this chapter whenever possible. This staging system was selected because it subdivides stage II disease into four subcategories on the basis of the size of the retroperitoneal lymph node metastases: smaller than 2 cm, 2 to 5 cm, 5 to 10 cm, and

larger than 10 cm. This makes it easy to convert to other staging systems that may use different size criteria to distinguish between moderate-size and bulky disease in the regional lymph nodes. For pure seminomas, 10 cm typically is used to distinguish between moderately advanced and massive retroperitoneal metastasis, whereas 5 cm is used to distinguish between moderately advanced and massive retroperitoneal disease for nonseminomatous testicular cancer.91,92

Pretreatment Evaluation Clinical evaluation of a patient suspected of having a testicular tumor should begin with a complete history and physical examination. This should include careful palpation of both testicles and transillumination to distinguish a solid mass from a hydrocele or spermatocele. The abdomen should be examined carefully to evaluate for retroperitoneal lymphadenopathy and hepatosplenomegaly. The neck should be palpated because the supraclavicular area is the most likely site of lymphatic spread above the diaphragm. The breasts should be evaluated for gynecomastia, which commonly occurs with choriocarcinomas. The initial laboratory studies should include a complete blood count, urinalysis, liver function studies, and blood urea nitrogen assay. If a testis tumor is suspected, testicular sonography should be performed, and serum markers should be drawn for β-HCG and alpha-fetoprotein (AFP). Semen analysis and sperm banking should be considered when treatment is likely to compromise fertility. If the initial clinical evaluation remains consistent with testicular tumor, the primary tumor should be removed by radical orchiectomy through an inguinal incision and the diagnosis confirmed by histopathologic examination. Further studies must then be performed to accurately stage the extent of the disease. Serum marker measurements should be repeated and compared with the results obtained before orchiectomy. A computed tomography (CT) scan should be obtained to establish whether pulmonary metastases are present. Retroperitoneal lymph nodes should be evaluated by pedal lymphangiography and abdominopelvic CT ­scanning.

Abdominal Imaging Lymphangiography. Pedal lymphangiography is useful for detecting regional lymph node metastasis, as an aid to setting up treatment portals, and for following the response to treatment (the nodes often remain opacified for months after the procedure). Lymphangiography can detect lymph node metastases as small as 5 mm in diameter. These lesions are usually seen as filling defects beneath the capsule of the node. Other findings suggesting nodal metastasis include deviation or obstruction of the lymph vessels or nonvisualization of a lymph node group. The accuracy of lymphangiography depends on the experience of the diagnostic radiologist, but it is said to have an overall accuracy of 80% in testicular cancer, with falsenegative rates of 15% to 20% and false-positive rates of 5% to 10%93,94 Computed Tomography. CT is less sensitive than lymphangiography because the nodes must be 1.5 to 2 cm in diameter or larger to be detected as abnormal. CT scans have been reported to have an overall accuracy of 76%,95 but the incidence of false negatives is greater with CT than

27  The Testicle

663

BOX 27-1.  Staging Classification Systems for Germ Cell Tumors of the Testis American Joint Committee on Cancer Tumor-Node-Metastasis Staging System* Primary Tumor (T) pTx Primary tumor cannot be assessed pT0 No evidence of primary tumor (e.g., histologic scar in testis) pTis Intratubular germ cell neoplasia (carcinoma in situ) pT1 Tumor limited to the testis and epididymis ­without vascular or lymphatic invasion; tumor may invade into the tunica albuginea but not the tunica vaginalis pT2 Tumor limited to the testis and epididymis with vascular or lymphatic invasion, or tumor extending hrough the tunica albuginea with involvement of the tunica vaginalis pT3 Tumor invades the spermatic cord with or ­ without vascular or lymphatic invasion pT4 Tumor invades the scrotum with or without ­vascular or lymphatic invasion‡

Royal Marsden Staging System� I II IIB IIC IID III IV

Tumor confined to the testis and spermatic cord Metastasis to lymph nodes below the diaphragm Maximum diameter 2-5 cm Maximum diameter 5-10 cm Maximum diameter > 10 cm Extension to nodes above the diaphragm ­(abdominal status as listed above: A, B, C, or D) Extranodal metastasis

Regional Lymph Nodes, Clinical (cN) and Pathologic (pN) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Metastasis with a lymph node mass 2 cm or less in greatest dimension; or multiple lymph nodes, none more than 2 cm in greatest dimension N2 Metastasis with a lymph node mass more than 2 cm but not more than 5 cm in greatest dimension; or multiple lymph nodes, any one mass greater than 2 cm but not more than 5 cm in greatest dimension N3 Metastasis with a lymph node mass more than 5 cm in greatest dimension Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis M1a Nonregional nodal or pulmonary metastasis M1b Distant metastasis other than to nonregional lymph nodes and lungs Serum Tumor Markers (S) SX Marker studies not available or not performed S0 Marker studies within normal limits S1 Lactate dehydrogenase (LDH) < 1.5 × N§ and human chorionic gonadotropin (hCG) < 5000 mIU/mL and alpha ­fetoprotein (AFP) < 1000 ng/mL S2 LDH 1.5-10 × N or hCG 5000-50,000 mIU/mL or AFP 1000-10,000 ng/mL S3 LDH > 10 × N or hCG > 50,000 mIU/mL or AFP > 10,000 ng/mL Stage Grouping Stage 0 Stage I Stage IA Stage IB

Stage IS Stage II Stage IIA Stage IIB Stage IIC Stage III Stage IIIA Stage IIIB

pTis pT1-4 pT1 pT2 pT3 pT4 Any pT/Tx Any pT/Tx Any pT/Tx Any pT/Tx Any pT/Tx Any pT/Tx Any pT/Tx Any pT/Tx Any pT/Tx Any pT/Tx Any pT/Tx Any pT/Tx Any pT/Tx

N0 N0 N0 N0 N0 N0 N0 N1-3 N1 N1 N2 N2 N3 N3 Any N Any N Any N N1-3 Any N

M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M1 M1a M1a M0 M1a

S0 SX S0 S0 S0 S0 S1-3 SX S0 S1 S1 S1 S0 S1 SX S0 S1 S2 S2 Continued

664

PART 8  Male Genital Tract

BOX 27-1.  Staging Classification Systems for Germ Cell Tumors of the Testis—cont’d Stage IIIC

Any pT/Tx Any pT/Tx Any pT/Tx

N1-3 Any N Any N

M0 M1a M1b

S3 S3 Any S

*Modified from Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 317-322. †Modified from Thomas GM. Consensus statement on the investigation and management of testicular seminoma. EORTC Genito-Urinary Group, monograph 7. In Newling DW, Jones WG (eds): Prostate Cancer and Testicular Cancer. New York: Wiley-Liss, 1990. ‡Except for pTis and pT4, extent of primary tumor is classified by radical orchiectomy. TX may be used for other categories in the absence of radical orchiectomy. §N indicates the upper limit of normal for the LDH assay.

it is with lymphangiography. However, CT scans are useful for evaluating nodes that do not opacify with pedal lymphangiography, such as the sentinel nodes near the renal hila and those located above the renal pedicles. Considerable controversy exists about whether ­testicular cancer should be staged initially by pedal lymphangiography or by CT. Some oncologists prefer pedal lymphangiography because it is more accurate than CT and has fewer false negatives.90,96,97 Others recommend CT, particularly if the retroperitoneal nodes are to be irradiated anyway. The argument is that any nodal metastasis from seminoma that is not detectable by CT is small enough to be easily controlled with the radiation therapy doses used electively.95,98,99 Lymphangiography is not readily available at many institutions, and most radiation oncologists rely on CT scans to identify the presence and extent of retroperitoneal spread.

Tumor Markers The development of tumor markers for germ cell tumors has had a profound effect on the management of testicular cancer. Tumor markers are useful for disease staging, for monitoring the effectiveness of therapy, and for the early detection of recurrent cancer. Tumor markers improve the accuracy of clinical staging for patients with testis cancer. Scardino and colleagues,100 for example, found that the addition of AFP and β-HCG assays to the initial workup reduced the false-negative rate from 29% to 13% in a series of patients with nonseminomatous testis cancer. Human Chorionic Gonadotropin. HCG is a glycoprotein normally secreted by syncytiotrophoblastic cells in the placenta. It is composed of two polypeptide chains, an α-HCG and a β-HCG component. The radioimmunoassay for cancer detection is directed to the β-HCG chain because antibodies to the α-HCG chain can cross-react with a variety of other hormones (e.g., LH, follicle-stimulating hormone, thyroid-stimulating hormone). β-HCG is found in many patients with germ cell tumors, usually those with trophoblastic elements. Abnormal titers are found in about 40% to 60% of patients with nonseminomatous testicular tumors and in 7% to 10% of patients with pure seminomas.101,102 However, the elevations seen with pure seminoma are almost always moderate. β-HCG is not found in normal men or boys. However, the level can be elevated with a variety of nontesticular malignancies, such as breast, stomach, pancreas, and liver cancer. Alpha-Fetoprotein.  AFP is a glycoprotein that is normally produced by the fetal yolk sac, liver, and gastrointestinal tract. Abnormal AFP levels can be found in newborns but not in adults. AFP titers may be elevated in patients

with hepatomas or other gastrointestinal malignancies or those with yolk sac tumors, embryonal carcinomas, or teratocarcinomas of the testis. Yolk sac tumors tend to produce AFP, just as the normal yolk sac does during fetal development. Approximately 70% of patients with teratocarcinomas or embryonal carcinomas have elevated AFP titers.101 Elevated AFP values are not found in patients with pure seminomas or choriocarcinomas. Any patient with a histopathologic diagnosis of pure seminoma and an elevated AFP should be considered to have a mixed tumor. Other Markers.  Several other markers have been proposed for detecting testicular tumors. Placental alkaline phosphatase and lactate dehydrogenase have been useful in the management of pure seminoma. Placental alkaline phosphatase is the most sensitive for detecting metastatic disease and has a sensitivity of 50% to 55%. However, false positives do occur.103 Clinical Applications of Tumor Markers. Tumor markers should be assayed before and after orchiectomy. However, postorchiectomy specimens should be obtained after sufficient time has elapsed to allow for metabolism of markers present in the serum at the time of orchiectomy. β-HCG has a metabolic half-life of 24 hours, and AFP has a metabolic half-life of 5 days. If the markers remain elevated after orchiectomy, metastatic disease is present. However, normal findings on postorchiectomy marker studies do not guarantee that metastasis is not present because false­negative results occur in 15% to 30% of cases.101

Treatment of Testicular Cancer Controversies Regarding Therapy for Pure Seminoma Pure seminomas seem to be ideally suited for treatment with radiation therapy because they are exquisitely radiosensitive and are characterized by an orderly lymphatic spread. Distant metastasis is rare. A review of the literature shows that the 5-year relapse-free survival rates with orchiectomy and regional lymphatic irradiation were more than 96% for men with stage I tumors and 82% for men with stage II tumors (Table 27-1).70,104-124 In a series reported from M. D. Anderson Cancer Center,92,125 the 5-year cause-specific survival rate was 97% for patients with no evidence of metastasis beyond the testis (Royal Marsden stage I) and 95% for patients with small to moderate-size regional lymph node metastasis (nodes smaller than 10 cm in ­diameter; Royal Marsden stage IIA-C). Even patients with massive nodal disease (>10 cm in diameter,

27  The Testicle

665

TABLE 27-1.    Survival Achieved with Orchiectomy and Radiation Therapy for Pure Seminoma Stage I Institution and Reference

Number of Patients

Stage II

Results (%)*

Number of Patients

Results (%)*

End Points

Antoni van Leeuwenhoek Hospital104

78

95

25

72

3-year RFS

Brooke General Hospital119

64

97

—

—

3-year survival

University117

33

94

19

80

2.5-year survival

42

93

19

89

5-year RFS

79

92

—

—

10-year survival

Indiana

Johns Hopkins

University109

Joint Center for Radiation

Therapy116

Roussy111

184

96

—

—

Act. 5-year survival

London Regional Cancer Centre105

169

97

43

88

2- to 10-year RFS

M. D. Anderson Hospital106

282

97

68

84

5-year RFS

135

95

25

84

5-year RFS

329

94

—

—

10-year survival

Institute Gustave

Massachusetts General Norwegian Radium

Hospital107

Hospital110

Hospital120

338

94

86

74

Act. 5-year survival

Rotterdam Institute122

153

95

74

82

2-year RFS

Royal Marsden Hospital71

121

97

54

81

2-year RFS

Princess Margaret

Stanford University108

71

100

27

85

Act. 5-year survival

State University of New

York123

104

94

—

—

5-year survival

State University of New

York124

—

—

45

89

5-year RFS

U.S. Naval Hospital San

Diego114

52

94

—

—

3- to 5-year survival

23

96

11

91

3-year survival

284

97

34

76

Act. 10-year survival

95

100

—

—

5-year survival

185

96

53

90

5-year RFS

61

98

18

100

5-year DSS

2882

96

556

82

University of Wisconsin113 Walter Reed General ­Hospital118 Washington

University115

Western General Yale

University112

Totals

Hospital121

*Whenever

possible, the results were reported in terms of 2- to 5-year relapse-free survival or absolute 5- to 10-year survival rates. Act., actuarial; DSS, disease-specific survival; RFS, relapse-free survival.

Royal Marsden stage IID) had a 5-year cause-specific survival rate of 67%. Recurrence within the irradiated area is rare in any series, even for patients with bulky disease.

Elective Irradiation versus Surveillance for Stage I Tumors Despite the excellent results achieved with radiation therapy, several controversies have arisen regarding its use in selected clinical situations. Standard treatment for stage I pure seminoma has been radical orchiectomy and elective irradiation of the regional lymphatics. However, several investigators have proposed treating these patients with orchiectomy alone and following them closely for disease progression. The arguments for surveillance in clinical stage I pure seminoma are that (1) only 15% to 20% of these patients actually have retroperitoneal metastasis, and therefore 80% to 85% are being irradiated needlessly and that (2) most patients who experience relapse after treatment with orchiectomy alone can be treated with salvage

r­ adiation therapy or chemotherapy when the disease becomes apparent clinically. Surveillance for stage I pure seminoma has been tested in clinical trials in Denmark, Britain, and Canada.126-129 Each of these studies showed relapse rates in the range of 16% to 20%, which correlates well with the expected incidence of retroperitoneal metastasis in clinical stage I seminoma. Moreover, almost all of the patients who developed recurrent disease were successfully treated with additional therapy, and the ultimate survival rates were not compromised when the patients were followed closely in a protocol situation. A Danish group,127 for example, reported a 20% actuarial 4-year relapse rate for 261 patients treated by orchiectomy and close surveillance. Twenty-two percent of 49 patients who experienced disease relapse had bulky abdominal disease at the time of progression, and 16% of these patients had metastasis beyond the periaortic area (5 in pelvic lymph nodes, 2 in inguinal lymph nodes, and

666

PART 8  Male Genital Tract

1 in the lungs). Similarly, only four patients (8% of those treated for salvage) experienced a second relapse, and only three (1.1% of the total group) died of seminoma-related events (one of progressive disease and two of complications related to salvage treatment). A similar relapse rate was reported in the British study from the Royal Marsden Hospital, but the disease was more advanced when it was detected.130 The authors of that study reported a 16% actuarial 3-year relapse rate for 133 patients treated with surveillance. However, 8 of 13 patients in whom progressive disease developed had bulky abdominal disease at relapse, and 1 had distant metastasis. Only one third of the patients who experienced relapse had early disease when the tumor recurred, and 5 of the 13 had a second relapse. In the Canadian study from the Princess Margaret Hospital, 172 patients were managed by orchiectomy and surveillance.128 Twenty-seven of them experienced relapse, for an actuarial 5-year relapse rate of 18%. However, 5 of the 27 patients who experienced relapse did so more than 4 years after diagnosis, including 1 patient after 9 years. These results support the notion that patients undergoing surveillance for pure seminoma must be followed for a long time because late recurrences are more common with seminomas than they are with nonseminomatous cancer.131 These studies show that surveillance is a satisfactory alternative for some patients with clinical stage I disease. However, it is probably not appropriate for most patients because aggressive follow-up is necessary if relapses are to be detected early. In the Canadian study,128 tumor markers were obtained at 2-month intervals for 2 years, at 4month intervals for the third year, and at 6-month intervals thereafter. Abdominal CT scans were obtained at 4-month intervals for the first 3 years, at 6-month intervals for years 4 to 7, and at yearly intervals thereafter. Patients undergoing surveillance must be followed much more closely than patients receiving elective radiation therapy. Not all patients comply with treatment recommendations, and a significant number of those who experience relapse will have bulky disease and require treatment with chemotherapy. For these reasons, most oncologists continue to favor elective irradiation for patients with stage I pure seminoma.

Treatment Portals for Stage I or II Seminoma The standard treatment portals for stage I or II pure seminoma have in the past been directed toward the periaortic and ipsilateral iliac areas. These fields have usually included the inguinal orchiectomy scar and the hemiscrotum if the tumor has extended through the tunica albuginea. How­ ever, inclusion of the inguinal nodes and hemiscrotum results in a significant radiation dose to the contralateral testis, and the risk of tumor seeding in the scrotum and inguinal region is low. Because of this, the International Consensus Conference in Leeds in 1989 recommended that treatment of the inguinal area and scrotum be omitted, even if the orchiectomy incision extends into the scrotum.91,132 Several investigators have proposed treating an even smaller volume. These investigators have advocated limiting the irradiation to the periaortic region and omitting treatment of the iliac lymph nodes.133-135 One of the main reasons for limiting the treatment to only the periaortic area is to decrease the risk of second malignancies.

Several investigators134,136-139 have reported an increased incidence of second cancer in patients with seminomas, and it has been proposed that some of these cases of a second cancer are the result of the abdominal radiation therapy received by these patients. However, not all of the second primary cancer cases reported were located in or near the radiation therapy treatment fields, and some investigators found a greater-than-expected incidence of second malignancies, even in men with testicular cancer who received no radiation therapy.140 Consequently, the high incidence of a second cancer may simply represent a predilection for men with seminoma to develop second primary malignant tumors regardless of the mode of treatment. Nevertheless, the risk of metastatic disease in the pelvis of patients with clinical stage I seminoma is quite small. Only 15% to 20% of patients with clinical stage I seminoma have occult lymphatic metastases, and almost all of these are located in the periaortic area (see Fig. 27-8). Occult metastasis to the pelvic area is much less common (about 1% to 3%) because the primary lymphatic drainage is to the periaortic nodes. This projection correlates well with pelvic failure rates when the iliac lymph nodes have not been irradiated. The reported follow-up times for most patients treated with only para-aortic irradiation are still relatively short, but Kiricuta and others141 reported only a 2.3% incidence of failure in the inguinal-iliac region in a group of patients followed for 1 to 13 years. Similarly, Sedlmayer and colleagues142 found a 3.5% failure rate in the pelvis in a series of patients followed for a period of 30 to 210 months in Salzburg, Austria. In a randomized study with a median follow-up time of 4.5 years, Fosså and colleagues133 reported a pelvic relapse rate of 1.7% among 236 patients treated with para-aortic radiation only and no pelvic relapses in patients who also had ipsilateral iliac treatment.

Elective Irradiation of the Mediastinal and ­Supraclavicular Nodes Before 1985, most patients with stage II pure seminoma were treated electively with mediastinal and supraclavicular irradiation. The rationale for this policy was that pure seminomas metastasize in a stepwise fashion, first to the retroperitoneal lymph nodes, then to the mediastinum and supraclavicular nodes, and subsequently to distant organs. The policy was to electively treat the next echelon of uninvolved nodes on the supposition that these areas may harbor occult disease. Most patients with stage II disease are no longer given elective radiation therapy above the diaphragm, mainly because effective chemotherapy for pure seminoma has been developed. There are four arguments against elective irradiation of the mediastinal and supraclavicular nodes: (1) the risk of subclinical disease in these nodes is small if the disease in the retroperitoneal nodes is limited; (2) elective irradiation of the mediastinum increases the risk of subsequent coronary artery disease; (3) irradiation of this area would make it more difficult to deliver effective chemotherapy later if progressive disease develops; and (4) patients with occult metastasis in the mediastinum and supraclavicular area can be treated successfully with further radiation therapy or chemotherapy when the metastases become clinically apparent.143 It is generally agreed that elective irradiation to the mediastinum is not indicated for

27  The Testicle

TABLE 27-2.   Incidence of Mediastinal or Supraclavicular Failure in Patients with Stage II Disease Who Did Not Undergo Elective Mediastinal and Supraclavicular Irradiation Royal Marsden Stage

Mediastinal or Supraclavicular Recurrences (%)

IIA (<2 cm)

9 (2 of 22)

IIB (2-5 cm)

16 (12 of 76)

IIC (>5 cm or bulky)

23 (38 of 164)

IID (≥10 cm or palpable)

16 (33 of 127)

Modified from Speer TW, Sombeck MD, Parsons JT, et al. Testicular seminoma: a failure analysis and literature review. Int J Radiat Oncol Biol Phys 1995;33:89-97.

patients with clinical stage I disease, and patients with stage II disease are being treated primarily with radiation below the diaphragm. Dosmann and Zagars106 found a 20% actuarial failure rate in the left supraclavicular area in a group of patients with clinical stage IIA-C disease (Royal Marsden stage, nodes smaller than 10 cm) who did not receive elective irradiation in this area. However, none of those who did receive elective supraclavicular irradiation had failures in this area. Similarly, Thomas and colleagues132 reported a 22% (10 of 46) relapse rate in the lymphatics above the diaphragm in patients with stage IIB pure seminoma (2- to 5-cm nodes) treated to only the retroperitoneal nodes. However, Herman and others143 reported only a 4% incidence of mediastinal and supraclavicular failures in patients with stage IIA disease (nodes smaller than 2 cm), an observation that suggests that elective irradiation is not necessary if the retroperitoneal lesions are smaller than 2 cm. Speer’s group144 found that the relapse rate in the mediastinum and supraclavicular area correlated well with the volume of disease in the periaortic lymph nodes (Table 27-2). In a review of 35 reports involving 389 patients with stage II disease, the investigators found a 9% failure rate in the mediastinal and supraclavicular areas in stage IIA, 16% in stage IIB, 23% in stage IIC, and 26% in stage IID disease (Royal Marsden classification).144 In view of the limited morbidity associated with supraclavicular irradiation, elective irradiation of the supraclavicular area seems to be appropriate for patients with stage IIB-C disease. However, irradiation of the mediastinum is not advocated because treatment of this area increases the risk of subsequent cardiac disease. In the Dosmann and Zagars series,106 most relapses above the diaphragm ­occurred in the supraclavicular area, indicating that tumor emboli follow normal anatomic channels, bypassing the mediastinum by way of the thoracic duct. If mediastinal irradiation is omitted, the risk of cardiac injury and the compromise of subsequent treatment with chemotherapy should be minimized.

Bulky Abdominal Disease The cure rates with radiation therapy alone for patients with massive retroperitoneal metastasis (stage IID) are poorer than they are for patients with small to moderatesize retroperitoneal disease (stage IIA-C). However, most

667

of these failures are the result of distant metastasis. Even massive seminomas can be eradicated with radiation therapy because seminomas are very radiosensitive. Anscher and others145 reported a 38% relapse rate for patients with bulky intra­abdominal disease. However, only 15% had recurrences within the treated area (Table 27-3).92,109,112,116,117,120,145-153 Zagars and Babaian92 also reported a significantly greater relapse rate for patients with bulky retroperitoneal metastasis than for those with less extensive disease, but none of the patients in their series developed in-field recurrences. These results show how exquisitely radiosensitive pure seminomas are, even when they are very large. Most patients with massive retroperitoneal metastases (Royal Marsden stage IID) are treated initially with chemotherapy. These are patients with nodal metastases larger than 10 cm. The use of chemotherapy has led to a modest improvement in survival for men with stage IID disease. Babaian and Zagars,92 for example, reported a 78% diseasefree survival rate for patients with stage IID disease after treatment with chemotherapy, compared with 68% for patients treated initially with radiation therapy alone. One reason for not using radiation therapy as the initial treatment for stage IIB disease is that it can be difficult to detect a relapse in these patients because of residual ­fibrotic masses. This can also be a problem after treatment with chemotherapy. A greater concern is that salvage with chemotherapy can be difficult because of bone marrow suppression after abdominal irradiation. In a review by Anscher and colleagues,145 only one fourth of the patients who had a relapse after radiation therapy for extensive intraabdominal metastases were successfully treated with chemotherapy (see Table 27-3).

Chemotherapy for Stage I Seminoma Chemotherapy as treatment for stage I pure seminoma has gained popularity in Europe over the past 2 decades.154,155 Typical regimens involve one or two cycles of carboplatin.154 Higher doses of one-cycle carboplatin or two cycles of lower doses are associated with recurrence rates of approximately 3%.154,155Some oncologists have advocated using one cycle of high-dose carboplatin chemotherapy for prophylaxis in stage I seminoma instead of the standard radiation therapy. In a randomized trial to compare radiation therapy with single-dose carboplatin after orchiectomy,155 the relapse-free survival rates at 2 years were 96.7% for ­radiation and 97.7% for carboplatin.155 Because treatment outcomes for stage I seminoma are excellent, the argument about whether to replace ­radiation therapy with chemotherapy is based mainly on ­toxicity considerations. However, findings on long-term toxicity and mortality take many years to collect. In one study of 453 men, the standardized mortality ratio for patients treated with radiation was not excessive compared with that of the general population for the first 15 years after treatment, but beyond 15 years, the ratio increased to 1.85.156 The cardiac-specific standardized mortality ratio in that study was 1.61, and the secondary cancer– ­specific ratio was 1.91, but both were significantly elevated only beyond 15 years after treatment.156 In another very large study of data from 40,576 patients, secondary solid tumors were associated with radiation therapy (relative risk = 2) and with cisplatin-based chemotherapy (relative risk = 1.8)157 among 10-year survivors. Moreover,

668

PART 8  Male Genital Tract

TABLE 27-3.   Patterns of Relapse after Treatment with Radiation Therapy Alone for Stage IIB Pure Seminoma (Bulky Abdominal Disease) Sites of Relapse Study and Reference

Number with Relapse

al146

Local Only

Local and Distant

Distant Only

1/4

0

0

1

Ball et al147

9/23

2

4

3

Dosoretz et al107

4/7

1

2

1

1/3

0

0

1

4/14

1

2

1

0/3

0

0

0

Andrews et

Epstein et

al109

Gregory and Hunter and

Peckham148

Peschel112 Halme150

0/2

0

0

0

Lederman et al116

6/9

2

0

4

Lester et al117

1/4

0

0

1

6/24

0

1

5

Kellokumpu-Lehtinen and

Mason and

Kearsley151 al152

1/11

0

0

1

Thomas et

al120

24/26

10

0

14

Willan and

McGowan153

Sagerman et

6/14

0

0

6

Zagars and Babain92

4/11

2

0

2

Totals

67/175

18

9

40

(38%)

(10%)

(5%)

(23%)

(Local failure rate = 15%) (Distant metastasis rate = 28%) Modified from Anscher MS, Marks LB, Shipley WU. The role of radiotherapy in patients with advanced seminomatous germ cell tumors. Controversies in management. Part 2. Oncology (Williston Park) 1992;6:97-104.

both ­ radiation therapy and chemotherapy are associated with leukemia158 and an elevated risk of myocardial infarction.159 In a study of the latter, at a median followup time of 18.4 years for seminoma survivors, receipt of radiation therapy was associated with a 3.7-fold increase and receipt of ­ cisplatin-based chemotherapy with a 1.9-fold increase in the risk of myocardial infarction.159 No evidence has been found that chemotherapy causes significantly less long-term toxicity in seminoma survivors than does radiation therapy. Data on long-term toxicity for carboplatin are lacking.

Radiation Therapy for Managing Residual Masses after Chemotherapy for Seminoma Residual masses after chemotherapy for bulky seminoma are detected on CT scans in about 80% of treated patients.160 Some of these postchemotherapy masses can take more than a year to resolve.161 The chance of harboring viable tumor in such a mass seems to be related to the size of the mass seen on CT; at 1 month after the completion of chemotherapy, residual masses larger than 3 cm have a 25% risk of harboring viable tumor cells.161-163 The best management strategy for postchemotherapy residual masses is controversial.160 Some oncologists advocate surgical resection of masses that are larger than 3 cm on CT,164 and others advocate observation.165-166 Because most such masses are fibrotic and can be difficult to ­resect,

radiation therapy to the residual mass is an option. Most postchemotherapy masses, including ones larger than 3 cm, do not harbor viable tumor cells; the use of 18F-fluorodeoxy-d-glucose positron emission tomography (FDGPET) scanning for detecting viable tumor cells has been tested. When interpreted in conjunction with CT results, PET scans can be highly accurate for this purpose.162,167,168 One prospective multicenter study showed that PET had 100% specificity for detecting viable tumor cells after chemotherapy, including those in masses larger than 3 cm.162,167 Observation seems reasonable for patients with postchemotherapy masses larger than 3 cm but negative findings on PET scanning.

Recommended Treatment for Pure Seminoma The recommended treatment policy for pure seminoma is outlined in Figures 27-9 and 27-10. The primary tumor in the testis should be managed by inguinal orchiectomy with high ligation of the spermatic cord, and the regional lymphatics should be treated with radiation therapy. If an incisional biopsy or scrotal orchiectomy has been performed, the spermatic cord should be resected before proceeding with treatment of the regional lymphatics. In keeping with the recommendations of the Leeds Consensus Conference,91 the inguinal scar and hemiscrotum do not need to be included in the radiation treatment portals.

27  The Testicle

669

D

C

B

A

FIGURE 27-9.  Typical field arrangement (A through D) for treating a seminoma of the left testis using separate periaortic and iliac fields. (From Hussey DH. Testicular cancer. In Levitt SH, Kahn FM, Potish RA, et al [eds]: Levitt & Tapley’s Technological Basis of Radiation Therapy: Practical Clinical Applications, 3rd ed. Philadelphia: Lippincott Williams & Wilkins, 1998.)

C

B

A

FIGURE 27-10.  Typical field arrangement (A through C) for treating a seminoma of the left testis using a hockey-stick-shaped periaortic and iliac field arrangement. (From Hussey DH. Testicular cancer. In Levitt SH, Kahn FM, Potish RA, et al [eds}: Levitt & Tapley’s Technological Basis of Radiation Therapy: Practical Clinical Applications, 3rd ed. Philadelphia: Lippincott Williams & Wilkins, 1998.)

Stage I If the tumor is clinically limited to the testis, the ­regional lymphatics below the diaphragm should be electively irradiated because 15% to 20% of these patients have occult metastasis in the retroperitoneal lymph nodes. Surveillance after orchiectomy may be used in selected patients, but it is not recommended routinely, because patients undergoing surveillance must be followed very closely with CT scans and tumor markers for a long time. The recommended tumor dose has been 25 Gy in 15 fractions over 3 weeks. However, the results of a randomized trial of two other radiation dose levels for microscopic pure seminoma indicate that at a median follow-up time of 61 months, tumor control rates were similar after 20 Gy (given in 2-Gy fractions) and after 30 Gy (also given in 2-Gy fractions), and the lower dose was associated with less toxicity.169 A dose of 20 Gy in 10 fractions may also be acceptable for microscopic disease. Periaortic and ipsilateral iliac portals (see Fig. 27-9A to C) or periaortic portals only (see Fig. 27-9B and C) are considered appropriate because the periaortic nodes are the ones that are most likely to be involved.133 The mediastinum and supraclavicular areas do not need to be irradiated electively in stage I disease because occult

metastasis above the diaphragm is present in less than 2% of such cases.143

Stage IIA In patients with retroperitoneal metastases measuring less than 2 cm in diameter, the periaortic and ipsilateral iliac areas should be treated through portals similar to those used for stage I disease. A dose of 25 Gy in 15 fractions over 3 weeks or 20 Gy in 10 fractions169 can be used (see Fig. 27-9A to C), followed by an additional 5 to 10 Gy in 2 to 5 fractions through reduced fields directed toward the area initially involved by cancer. Elective irradiation of the mediastinum and supraclavicular area is no longer recommended for stage IIA disease, because less than 10% of such cases have occult metastasis in the mediastinal or supraclavicular nodes, and in most cases, relapse in this region can be salvaged with chemotherapy or additional radiation (see Table 27-2).120,132,144

Stage IIB or IIIC In patients with moderate-size or large retroperitoneal metastases (2 to 10 cm in diameter), the periaortic and ipsilateral iliac lymph nodes can be irradiated to 25 Gy in

670

PART 8  Male Genital Tract

15 fractions over 3 weeks or to 20 Gy in 10 fractions.169 The fields should then be reduced and the total dose to the initially involved areas taken to 35 to 40 Gy by using a shrinking field technique. Nodal masses of up to 5 cm can also be treated with radiation, with chemotherapy reserved for more bulky disease. Elective irradiation of the left supraclavicular area (see Fig. 27-9D) is recommended for stage IIB-C disease because approximately 20% of patients with disease at this stage will experience failure in the supraclavicular nodes if this area is not electively treated (see Table 27-2).106,132,144 The mediastinum is not routinely irradiated because treatment of this area can increase the risk of subsequent coronary artery disease and make it more difficult to deliver effective chemotherapy should the need arise. Supraclavicular radiation is given to a dose of 25 Gy over 3 weeks (or 20 Gy in 2 weeks).169 The supraclavicular field adds little morbidity, and it should not prolong the treatment because it can be delivered concurrently with the retroperitoneal lymphatic irradiation.

Stage IID In the 1980s, most patients with bulky intra-abdominal disease (stage IID) were treated with radiation therapy. Currently, patients with stage IID disease are treated with chemotherapy, and radiation therapy is reserved for salvage if chemotherapy fails.

Stage III or IV Patients with stage III or IV pure seminomas are treated initially with chemotherapy, and radiation therapy is used to eradicate residual disease or to palliate local problems. A significant number of patients with stage III or IV disease can be cured with aggressive treatment.

Treatment Portals The retroperitoneal lymph nodes can be irradiated through parallel-opposing hockey stick–shaped portals (see Fig. 27-10) or separate periaortic and iliac fields (see Fig. 27-9). The advantage of using the hockey-stick arrangement is that it avoids a field junction between the ­periaortic and iliac fields. The advantages of using separate periaortic and iliac fields are that the blocks are smaller and easier to handle, the dose transmitted to the bowel is less, and the periaortic and iliac fields can be weighted differently, which may be appropriate because the iliac nodes are situated anterior to the periaortic nodes. The periaortic nodes are treated through parallel­opposed, equally weighted anterior and posterior portals, and the tumor dose is prescribed at midplane (see Fig. 27-10). Alternately, CT planning can be used and the ­prescription written to the appropriate isodose line. The periaortic fields are approximately 10 cm wide and extend from T10 to the L5-S1 junction. The nodes in the renal hila are included, especially on the left, where the left spermatic vein joins the left renal vein. The lateral two thirds of the kidney is excluded if possible. If separate periaortic and iliac fields are used, the portals are usually matched at the L5-S1 junction. The iliac nodes are usually treated with a single anterior field, and the dose is calculated 3 cm anterior to midplane. The upper field border varies from T10 to T12 among institutions. It is important that this border is superior enough to include the level of the renal hila.

The periaortic and iliac fields are treated by using 6- to 18-MV x-rays. However, if the patient has a large anteroposterior diameter, 10- to 18-MV x-rays should be used, and the iliac area should be treated with parallel-opposed portals to reduce the risk of subcutaneous fibrosis and other late side effects of irradiation. The supraclavicular area should be treated with 4- to 6-MV x-rays. The medial margin of the supraclavicular field should include the sternocleidomastoid muscle with a 1-cm margin; the superior margin should be at the cricoid cartilage; and the inferior margin at the lower edge of the clavicle.

Treatment for Nonseminomatous Testicular Tumors Radiation therapy played a role in the management of nonseminomatous germ cell tumors of the testis before the development of effective chemotherapy.79,149 However, it has not been a major part of the treatment of nonseminomas since the early 1980s. Nonseminomatous testicular tumors are not as radiosensitive as pure seminomas, and consequently, 45 Gy or more is required to control them. These tumors also have a greater tendency to spread hematogenously, and nonseminomas are not as easily cured by any form of locoregional treatment. The recommended treatment policy for nonseminomatous testicular tumors is outlined by disease stage. As is true for other testicular tumors, the primary tumor is treated by an inguinal orchiectomy with high ligation of the spermatic cord. Management after this depends on the stage of the disease.

Stage I Patients with no evidence of metastases are usually treated initially with retroperitoneal lymphadenectomy. Those who have nodal metastases at surgery or have persistently elevated tumor markers after lymphadenectomy are treated with chemotherapy. In some institutions, a policy of surveillance is used for selected patients with low-risk disease. At Memorial Sloan-Kettering Cancer Center,72 patients are considered to be candidates for surveillance if (1) the tumor is confined within the tunica albuginea, (2) no vascular invasion is evident, (3) the tumor marker levels normalize after orchiectomy, (4) radiographic imaging shows no evidence of metastatic disease, and (5) the patient is considered likely to return for the necessary follow-up visits. Patients who are being managed by surveillance must be followed closely for at least 3 years. Patients are followed monthly for the first 2 years and bimonthly in the third year. Tumor markers are measured after each visit, and a chest radiograph and abdominal CT scan are obtained every 3 to 4 months. Most relapses of nonseminomatous testicular tumors occur within the first year.

Stage II Patients with limited stage II disease (nodes smaller than 3 cm) are treated initially with inguinal orchiectomy and retroperitoneal lymphadenectomy in a manner similar to that described for stage I disease. However, adjuvant chemotherapy is usually given postoperatively because the rate of recurrence is high when the patients are treated with lymphadenectomy alone.

27  The Testicle

Patients with moderate-size to bulky retroperitoneal disease (stage IIB-D, nodes larger than 3 cm) are treated with aggressive chemotherapy primarily in a manner similar to those who present with stage III or IV disease. These patients are treated with orchiectomy followed by primary platinum-based combination chemotherapy. Later resection is required if a residual mass is found on imaging studies. Approximately 20% of these masses contain residual cancer, usually embryonal carcinoma. In another 40%, the mass is composed of only teratoma, and in 40%, it represents fibrosis. If the tumor marker levels fail to normalize, additional salvage chemotherapy is required.

Stage III or IV Patients with stage III or IV disease are treated initially with multiagent chemotherapy. Radiation therapy or surgery, or both, may be used if the response to drug therapy is incomplete.

Extragonadal Germ Cell Tumors Approximately 1% of malignant germ cell tumors appear outside the testes.170 The most common sites are the anterior mediastinum, retroperitoneum, and brain. Less common sites include the prostate, bladder, stomach, and thymus. Extragonadal germ cell tumors are identical histopathologically to germ cell tumors originating in the testis,170 and all types of germ cell tumors have been reported in extragonadal sites.171 They are thought to arise from primordial germ cells that have been displaced during their migration from the yolk sac endoderm to the genital ridge to the testes.171 The treatment for an extragonadal germ cell tumor is similar to that used for a testicular tumor of the same histopathologic type. Most patients with extragonadal pure seminomas are treated with radiation therapy alone, and the dose is determined by tumor size.172 Although Bush and colleagues173 have suggested that primary mediastinal seminomas require a higher radiation dose than that used for testicular seminomas, few data exist to suggest that ­ extragonadal seminomas are any less radiosensitive than testicular seminomas of a similar size. Nevertheless, multiagent chemotherapy may be indicated if the tumor is quite large.174,175 For nonseminomatous extragonadal germ cell tumors, chemotherapy similar to that used for nonseminomatous tumors of the testis is recommended. This is usually followed by resection of the residual mass if the tumor mass persists and is technically resectable. The prognosis for a patient with an extragonadal germ cell tumor is somewhat poorer than it is for a patient with a testicular tumor of the same histopathologic type. For ­example, disease-free survival rates with radiation therapy for extragonadal seminomas are approximately 65% (range, 40% to 80%),145 probably because many of these tumors are much larger than the typical testicular seminoma. Cure rates of about 85% can be expected for patients with small to moderate-size extragonadal seminomas. The results for nonseminomatous extragonadal germ cell tumors are poorer than those that can be achieved for testicular cancer of the same histopathologic type. In a

671

­review of the literature, Cox176 found reports of only two patients with embryonal cell carcinomas of the mediastinum that had been cured, and not a single patient with teratocarcinoma or choriocarcinoma of the mediastinum was a long-term survivor. Somewhat higher cure rates have been achieved with more effective chemotherapy, but the cure rates for extragonadal tumors are still not as high as those that can be achieved with chemotherapy for nonseminomatous tumors of the testis.171,175,177

Tumors of the Gonadal Stroma A variety of gonadal stromal tumors can be found in the testes, including Leydig cell tumors, Sertoli cell tumors, undifferentiated tumors, and combinations of these histopathologic types. Gonadoblastomas contain a mixture of germ cell and gonadal stromal cell elements. All stromal cell tumors of the testis are rare.88,178 Leydig cell tumors can occur at any age, but they are most common in men 30 to 50 years old. The presenting symptoms in children include macrogenitosomia, hirsutism, early musculoskeletal development, and sexual precocity. Adults may have excessive genital development, gynecomastia, or other signs of feminization. Patients with Leydig cell tumors typically have elevated 17-ketosteroid levels in the urine and blood. Levels of estrogen or other steroidal products may also be elevated. Sertoli cell tumors are rare neoplasms. About one third of patients with these tumors have gynecomastia, and some of them experience decreased libido. Patients with Sertoli cell tumors may have elevated levels of androgens, estrogens, and pregnanediol in the urine. However, the 17-ketosteroid levels are normal. Gonadoblastomas occur in patients with an underlying gonadal disorder and abnormal secondary sex characteristics. The typical picture histologically is a mixture of proliferating seminoma-like germ cells, Sertoli cells, granulosa cells, and Leydig cells. The standard treatment for patients with any of these gonadal stromal tumors is radical orchiectomy. Only about 10% of these tumors are malignant. The most common sites of metastasis from malignant stromal cell tumors are the lymph nodes, liver, and bone. Radiation therapy does not usually play a role in the management of these types of tumors.

Malignant Lymphomas Testicular lymphomas usually occur in older men and are considered the most common testicular tumor in men older than 50 years.73 They account for between 1% and 7% of testicular tumors in various series.179 Testicular lymphomas are usually diffuse large cell lymphomas, but other varieties of lymphomas in the testes have been reported.180 Lymphomas of the testis are usually a manifestation of metastatic disease. However, it is fairly common for one to be the only evidence of disease at the time of presentation. Painless testicular enlargement is the most common presenting symptom. However, patients may have nodal or extranodal disease elsewhere in the body or generalized symptoms such as anorexia, weakness, or weight loss. A patient with a diagnosis of lymphoma should be carefully evaluated for systemic disease. Special studies that are

672

PART 8  Male Genital Tract

indicated, in addition to those usually obtained to evaluate a patient with a testicular malignancy, include a bone marrow biopsy; CT scanning of the head and neck, chest, abdomen, and pelvis; and lumbar puncture for cytologic examination of the cerebrospinal fluid.181 If the disease is confined to the testis, radical orchiectomy should be performed, and patients should be treated with doxorubicin-based combination chemotherapy supplemented with intrathecal chemotherapy.182 The contralateral testis is considered an isolated structure in that it is unlikely to receive the full effect of combination chemotherapy. After combination systemic chemotherapy and intrathecal drug injections, irradiation of the scrotum is recommended. A total dose of 30 Gy, given in 15 to 17 fractions, is considered sufficient to eliminate subclinical lymphoma. Most patients with testicular lymphoma have systemic disease, even those who have no clinical evidence of tumor elsewhere, and the prognosis has been poor. Preliminary results of the treatment policies previously outlined suggest an improvement in outcome for what has heretofore been a highly lethal form of testicular cancer.

REFERENCES    1. Fajardo LF, Berthrong M, Anderson RE. Testis. In Radiation Pathology. New York: Oxford University Press, 2001, pp 312-318.    2. Russell LD, Ettlin RA, Hikim APS, et al. Histological and Histopathological Evaluation of the Testis. Clearwater, FL: Cache River Press, 1990.    3. Meistrich ML. Relationship between spermatogonial stem cell survival and testis function after cytotoxic therapy. Br J Cancer 1986;53:89-101.    4. Wartenberg H. Differentiation and development of the testes. In Burger HG, de Kretser D (eds): The Testis, 2nd ed. New York: Raven Press����������������������� , 1989, pp 67–118.    5. Rajpert-De Meyts E. Developmental model for the pathogenesis of testicular carcinoma in situ: genetic and environmental aspects. Hum Reprod Update 2006;12:303-323.    6. Almstrup K, Sonne SB, Hoei-Hansen CE, et al. From embryonic stem cells to testicular germ cell cancer—should we be concerned? Int J Androl 2006;29:211-218.    7. Clermont Y. Renewal of spermatogonia in man. Am J Anat 1966;118:509-524.    8. Meistrich ML, Hunter NR, Suzuki N, et al. Gradual regeneration of mouse testicular stem cells after exposure to ionizing radiation. Radiat Res 1978;74:349-362.    9. Ehmcke J, Simorangkir DR, Schlatt S. Identification of the starting point for spermatogenesis and characterization of the testicular stem cell in adult male rhesus monkeys. Hum Reprod 2005;20:1185-1193.   10. Meistrich ML, van Beek MEAB. Spermatogonial stem cells. In Ewing LL (ed): Cell and Molecular Biology of the Testis. New York: Oxford University Press, 1993, pp 266-295.   11. Huckins C. The spermatogonial stem cell population in adult rats. I. Their morphology, proliferation and maturation. Anat Rec 1971;169:533-557.   12. Heller CH, Clermont Y. Kinetics of the germinal epithelium in man. Recent Prog Horm Res 1964;20:545-575.   13. da Cunha MF, Meistrich ML, Haq MM, et al. Temporary ­effects of AMSA (4′-(9-acridinylamino) methanesulfonm-anisidide) chemotherapy on spermatogenesis. Cancer 1982;49: 2459-2462.   14. Oakberg EF. Sensitivity and time of degeneration of spermatogenic cells irradiated in various stages of maturation in the mouse. Radiat Res 1955;2:369-391.   15. Oakberg EF. Degeneration of spermatogonia of the mouse following exposure to x-rays, and stages in the mitotic cycle at which cell death occurs. J Morphol 1955;97:39-54.

  16. Rowley MJ, Leach DR, Warner GA, et al. Effect of graded doses of ionizing radiation on the human testis. Radiat Res 1974;59: 665-678.   17. Hasegawa M, Wilson G, Russell LD, et al. Radiation-induced cell death in the mouse testis: relationship to apoptosis. Radiat Res 1997;147:457-467.   18. Regaud C. Influence de la duree d’irradiation sur les effects determines dans le testicule par le radium. C R Seances Soc Biol Fil 1922;86:787-790.   19. Ferroux R, Regaud C, Samssonaw N. Effects des rayons de roentgen administres sans fractionnement de la dose sur les testicules du rat au point de vue de la sterilisation de l’epithelium seminal. C R Seances Soc Biol Fil 1938;128:170-173.   20. Lushbaugh CC, Casarett GW. The effects of gonadal irradiation in clinical radiation therapy: a review. Cancer 1976;37: 1111-1125.   21. Meistrich ML, Finch M, Lu CC, et al. Strain differences in the response of mouse testicular stem cells to fractionated radiation. Radiat Res 1984;97:478-487.   22. Meistrich ML, van Beek MEAB. Radiation sensitivity of the human testis. Adv Radiat Biol 1990;14:227-268.   23. van der Meer Y, Huiskamp R, Davids JA, et al. The sensitivity of quiescent and proliferating mouse spermatogonial stem cells to x irradiation. Radiat Res 1992;130:289-295.   24. Withers HR, Hunter N, Barkley HT Jr., et al. Radiation survival and regeneration characteristics of spermatogenic stem cells of mouse testis. Radiat Res 1974;57:88-103.   25. Edwards RG, Sirlin JL. The effect of 200 r of x-rays on the rate of spermatogenesis and spermiogenesis in the mouse. Exp Cell Res 1958;15:522-528.   26. Meistrich ML, Samuels RC. Reduction in sperm levels after testicular irradiation of the mouse: a comparison with man. Radiat Res 1985;102:138-147.   27. Clifton DK, Bremner WJ. The effect of testicular x-irradiation on spermatogenesis in man. A comparison with the mouse. J Androl 1983;4:387-392.   28. Dubey P, Wilson G, Mathur KK, et al. Recovery of sperm production following radiation therapy for Hodgkin’s disease after induction chemotherapy with mitoxantrone, vincristine, vinblastine, and prednisone (NOVP). Int J Radiat Oncol Biol Phys 2000;46:609-617.   29. van Beek ME, Meistrich ML, de Rooij DG. Probability of selfrenewing divisions of spermatogonial stem cells in colonies, formed after fission neutron irradiation. Cell Tissue Kinet 1990;23:1-16.   30. Kangasniemi M, Huhtaniemi I, Meistrich ML. Failure of spermatogenesis to recover despite the presence of a spermatogonia in the irradiated LBNF1 rat. Biol Reprod 1996;54:1200-1208.   31. Shuttlesworth GA, de Rooij DG, Huhtaniemi I, et al. Enhancement of A spermatogonial proliferation and differentiation in irradiated rats by GnRH antagonist administration. Endocrinology 2000;141:37-49.   32. Zhang Z, Shao S, Meistrich M. The radiation-induced block in spermatogonial differentiation is due to damage to the somatic environment, not the germ cells. J Cell Physiol 2007;211: 149-158.   33. Meistrich ML, Wilson G, Shuttlesworth G, et al. GnRH agonists and antagonists stimulate recovery of fertility in irradiated LBNF1 rats. J Androl 2001;22:809-817.   34. Thomson AB, Anderson RA, Irvine DS, et al. Investigation of suppression of the hypothalamic-pituitary-gonadal axis to restore spermatogenesis in azoospermic men treated for childhood cancer. Hum Reprod 2002;17:1715-1723.   35. Boekelheide K, Schoenfeld H, Hall SJ, et al. Gonadotropinreleasing hormone antagonist (cetrorelix) therapy fails to protect non-human primates (Macaca arctoides) from radiation-induced spermatogenic failure. J Androl 2005;26:222-234.   36. Anserini P, Chiodi S, Spinelli S, et al. Semen analysis following allogeneic bone marrow transplantation. Additional data for evidence-based counseling. Bone Marrow Transplant 2002;30: 447-451.   37. Jacob A, Barker H, Goodman A, et al. Recovery of spermatogenesis following bone marrow transplantation. Bone Marrow Transplant 1998;22:277-279.

27  The Testicle   38. Hansen PV, Trykker H, Svennekjaer IL, et al. Long-term recovery of spermatogenesis after radiotherapy in patients with testicular cancer. Radiother Oncol 1990;18:117-125.   39. Ash P. The influence of radiation on fertility in man. Br J Radiol 1980;53:271-278.   40. Hahn EW, Feingold SM, Simpson L, et al. Recovery from aspermia induced by low-dose radiation in seminoma patients. Cancer 1982;50:337-340.   41. Hahn EW, Feingold SM, Nisce L. Aspermia and recovery of spermatogenesis in cancer patients following incidental gonadal irradiation during treatment: a progress report. Radiology 1976;119:223-225.   42. Sandeman TF. The effects of X irradiation on male human fertility. Br J Radiol 1966;39:901-907.   43. Meistrich ML, Brown CC. Estimation of the increased risk of human infertility from alterations in semen characteristics. Fertil Steril 1983;40:220-230.   44. Jacobsen KD, Olsen DR, Fosså K, et al. External beam abdominal radiotherapy in patients with seminoma stage I: field type, testicular dose, and spermatogenesis. Int J Radiat Oncol Biol Phys 1997;38:95-102.   45. Speiser B, Rubin P, Casarett G. Aspermia following lower truncal irradiation in Hodgkin’s disease. Cancer 1973;32: 692-698.   46. Million RR. The lymphomatous diseases. In Fletcher GH (ed): Textbook of Radiotherapy, 3rd ed. Philadelphia: Lea & Febiger, 1980.   47. Harter DJ, Hussey DH, Delclos L, et al. Device to position and shield the testicle during irradiation. Int J Radiat Oncol Biol Phys 1976;1:361-364.   48. Neel JV, Schull WJ, Awa AA, et al. The children of parents exposed to atomic bombs: estimates of the genetic doubling dose of radiation for humans. Am J Hum Genet 1990;46: 1053-1072.   49. Joshi DS, Yick J, Murray D, et al. Stage-dependent variation in the radiosensitivity of DNA in developing male germ cells. Radiat Res 1990;121:274-281.   50. Boice JD Jr, Tawn EJ, Winther JF, et al. Genetic effects of radiotherapy for childhood cancer. Health Phys 2003;85:65-80.   51. Delic JI, Hendry JH, Morris ID, et al. Dose and time relationships in the endocrine response of the irradiated adult rat testis. J Androl 1986;7:32-41.   52. Guitton N, Brouazin-Jousseaume V, Dupaix A, et al. Radiation effect on rat Sertoli cell function in vitro and in vivo. Int J Radiat Biol 1999;75:327-333.   53. Delic JI, Hendry JH, Morris ID, et al. Leydig cell function in the pubertal rat following local testicular irradiation. Radiother Oncol 1986;5:29-37.   54. Shapiro E, Kinsella TJ, Makuch RW, et al. Effects of fractionated irradiation of endocrine aspects of testicular function. J Clin Oncol 1985;3:1232-1239.   55. Wang J, Galil KA, Setchell BP. Changes in testicular blood flow and testosterone production during aspermatogenesis after irradiation. J Endocrinol 1983;98:35-46.   56. Shetty G, Wilson G, Hardy MP, et al. Inhibition of recovery of spermatogenesis in irradiated rats by different androgens. Endocrinology 2002;143:3385-3396.   57. Nader S, Schultz PN, Cundiff JH, et al. Endocrine profiles of patients with testicular tumors treated with radiotherapy. Int J Radiat Oncol Biol Phys 1983;9:1723-1726.   58. Howell SJ, Radford JA, Adams JE, et al. Randomized placebocontrolled trial of testosterone replacement in men with mild Leydig cell insufficiency following cytotoxic chemotherapy. Clin Endocrinol 2001;55:315-324.   59. Petersen PM, Giwercman A, Daugaard G, et al. Effect of graded testicular doses of radiotherapy in patients treated for carcinoma-in-situ in the testis. J Clin Oncol 2002;20:1537-1543.   60. Shalet SM. Effect of irradiation treatment on gonadal function in men treated for germ cell cancer. Eur Urol 1993;23:148-151. discussion 152.   61. Brauner R, Czernichow P, Cramer P, et al. Leydig-cell function in children after direct testicular irradiation for acute lympho­ blastic leukemia. N Engl J Med 1983;309:25-28.

673

  62. Shalet SM, Tsatsoulis A, Whitehead E, et al. Vulnerability of the human Leydig cell to radiation damage is dependent upon age. J Endocrinol 1989;120:161-165.   63. Radford J. Restoration of fertility after treatment for cancer. Horm Res 2003;59(Suppl 1):21-23.   64. Avarbock MR, Brinster CJ, Brinster RL. Reconstitution of spermatogenesis from frozen spermatogonial stem cells. Nat Med 1996;2:693-696.   65. Schlatt S, Foppiani L, Rolf C, et al. Germ cell transplantation into x-irradiated monkey testes. Human Reprod 2002;17:55-62.   66. Ries LAG, Melbert D, Krapcho M, et al (eds). SEER Cancer Statistics Review, 1975-2004. Bethesda, MD: National Cancer Institute. Available at http://seer.cancer.gov/csr/1975_2004/(accessed April 2009), based on November 2006 SEER data submission, posted in 2007.   67. Schottenfeld D, Warshauer ME, Sherlock S, et al. The epidemiology of testicular cancer in young adults. Am J Epidemiol 1980;112:232-246.   68. Ekbom A, Akre O. Increasing incidence of testicular cancer— birth cohort effects. APMIS 1998;106:225-229. discussion 229–231.   69. Batata MA, Whitmore WF Jr, Chu FC, et al. Cryptorchidism and testicular cancer. J Urol 1980;124:382-387.   70. Peckham MJ, McElwain TJ, Hendry WF. Testis and epididymis. In Halnan KE (ed): Treatment of Cancer. New York: Igaku-­Shoin Medical Publishers, 1982.   71. Peckham MJ. Testicular cancer. Rev Oncol 1988;1:439-453.   72. Presti JC, Herr HW. Genital tumors. In Tanagho MOA, McAnnich W (eds): Smith’s General Urology. 13th ed. Norwalk, CT: Appleton & Lange, 1992.   73. Mostofi FK, Price EB. Tumors of the Male Genital System. Washing­ton, DC: Armed Forces Institute of Pathology, 1973.   74. Skakkebaek NE. Carcinoma in situ of the testis: frequency and relationship to invasive germ cell tumours in infertile men. [Comment on] N.E. Skakkebaek. Histopathology 1978;2: 157-170. Histopathology 2002;41:2–3.   75. Sokal M, Peckham MJ, Hendry WF. Bilateral germ cell tumours of the testis. Br J Urol 1980;52:158-162.   76. Dieckmann KP, Loy V. Management of contralateral testicular intraepithelial neoplasia in patients with testicular germ-cell tumor. World J Urol 1994;12:131-135.   77. von der Maase H, Giwercman A, Muller J, et al. Management of carcinoma-in-situ of the testis. Int J Androl 1987;10:209-220.   78. Willis RA. Pathology of Tumors. New York: Appleton-­CenturyCrofts, 1967.   79. Friedman NB. Tumors of the testes. Milit Surg 1946;99: 573-593.   80. Dixon FJ, Moore RA. Tumors of the Male Sex Organs.. Washington, DC: Armed Forces Institute of Pathology, 1952.   81. Mostofi FK. Testicular tumors. epidemiologic, etiologic, and pathologic features. Cancer 1973;32:1186-1201.   82. Melicow MM. The New ‘British’ classification of testicular tumors: a correlation, analysis and critique. J Urol 1965;94:64-68.   83. Stevens LC. Origin of testicular teratomas from primordial germ cells in mice. J Natl Cancer Inst 1967;38:549-552.   84. Johnson DE, Gomez JJ, Ayala AG. Anaplastic seminoma. J Urol 1975;114:80-82.   85. Babaian RJ, Zagars GK. Testicular seminoma: the M. D. Anderson experience. An analysis of pathological and patient characteristics, and treatment recommendations. J Urol 1988;139:311-314.   86. Sandeman TF, Matthews JP. The staging of testicular tumors. Cancer 1979;43:2514-2524.   87. Weiss L, Gilbert HA, Bullon SC (eds). Lymphatic System Metastasis. Boston: GK Hall, 1980.   88. Ray B, Hajdu SI, Whitmore WF Jr. Proceedings: distribution of retroperitoneal lymph node metastases in testicular germinal tumors. Cancer 1974;33:340-348.   89. Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual. 6th ed. New York: Springer, 2002, pp 317-322.   90. Hussey DH. Testicular cancer. In Levitt SH, Khan FM, Potish RA, et al (eds): Levitt & Tapley’s Technological Basis of Radiation Therapy: Practical Clinical Applications, 3rd ed. Philadelphia: Williams & Wilkins, 1998.

674

PART 8  Male Genital Tract

  91. Thomas G, Jones W, Van Oosterom A, et al. Consensus statement on the investigation and management of testicular seminoma 1989. Prog Clin Biol Res 1990;357:285-294.   92. Zagars GK, Babaian RJ. The role of radiation in stage II testicular seminoma. Int J Radiat Oncol Biol Phys 1987;13:163-170.   93. Jing B, Wallace S, Zornoza J. Metastases to retroperitoneal and pelvic lymph nodes: computed tomography and lymphangiography. Radiol Clin North Am 1982;20:511-530.   94. Maier JG, Schamber DT. The role of lymphangiography in the diagnosis and treatment of malignant testicular tumors. Am J Roentgenol Radium Ther Nucl Med 1972;114:482-491.   95. Heiken JP, Schamber DT. Testicular tumors. In Bragg DG, Rubin P, Youker JE (eds): Oncologic Imaging. New York: Pergamon Press, 1985.   96. Skinner DG, Scardino PT, Daniels JR. Testicular cancer. Annu Rev Med 1981;32:543-557.   97. Morton GC, Thomas GM. Testis. In Perez CA, Brady LW (eds): Principles and Practice of Radiation Oncology, 5th ed. Philadelphia: Lippincott Williams & Wilkins, 2008, pp 1503-1518.   98. Marks LB, Anscher MS, Shipley WU. Radiation therapy for testicular seminoma: controversies in the management of earlystage disease. Oncology (Williston Park) 1992;6:43-48; discussion 51–42.   99. Zagars GK. Management of stage I seminoma: radiotherapy. In Horwich A (ed): Testicular Cancer: Investigation and Management. Baltimore: Williams & Wilkins, 1991. 100. Scardino PT, Cox HD, Waldmann TA, et al. The value of serum tumor markers in the staging and prognosis of germ cell tumors of the testis. J Urol 1977;118:994-999. 101. Javadpour N. The role of biologic tumor markers in testicular cancer. Cancer 1980;45:1755-1761. 102. Lange PH, Nochomovitz LE, Rosai J, et al. Serum alpha­fetoprotein and human chorionic gonadotropin in patients with seminoma. J Urol 1980;124:472-478. 103. Weissbach L, Bussar-Maatz R, Mann K. The value of tumor markers in testicular seminomas. Results of a prospective multicenter study. Eur Urol 1997;32:16-22. 104. Battermann JJ, Delemarre JF, Hart AA, et al. Testicular tumors. A retrospective study. Arch Chir Neerl 1973;25:457-469. 105. Bauman GS, Venkatesan VM, Ago CT, et al. Postoperative radiotherapy for stage I/II seminoma: results for 212 patients. Int J Radiat Oncol Biol Phys 1998;42:313-317. 106. Dosmann MA, Zagars GK. Post-orchiectomy radiotherapy for stages I and II testicular seminoma. Int J Radiat Oncol Biol Phys 1993;26:381-390. 107. Dosoretz DE, Shipley WU, Blitzer PH, et al. Megavoltage irradiation for pure testicular seminoma: results and patterns of failure. Cancer 1981;48:2184-2190. 108. Earle JD, Bagshaw MA, Kaplan HS. Supervoltage radiation therapy of the testicular tumors. Am J Roentgenol Radium Ther Nucl Med 1973;117:653-661. 109. Epstein BE, Order SE, Zinreich ES. Staging, treatment, and results in testicular seminoma. A 12-year report. Cancer 1990;65: 405-411. 110. Fosså SD, Aass N, Kaalhus O. Radiotherapy for testicular seminoma stage I: treatment results and long-term post-­irradiation morbidity in 365 patients. Int J Radiat Oncol Biol Phys 1989;16:383-388. 111. Giacchetti S, Raoul Y, Wibault P, et al. Treatment of stage I testis seminoma by radiotherapy: long-term results—a 30-year experience. Int J Radiat Oncol Biol Phys 1993;27:3-9. 112. Hunter M, Peschel RE. Testicular seminoma. Results of the Yale University experience, 1964-1984. Cancer 1989;64:1608-1611. 113. Kademian MT, Bosch A, Caldwell WL. Seminoma: results of treatment with megavoltage irradiation. Int J Radiat Oncol Biol Phys 1976;1:1075-1079. 114. Kurohara SS, George FW 3rd, Dykhuisen RF, et al. Testicular tumors. Analysis of 196 cases treated at the U.S. Naval Hospital in San Diego. Cancer 1967;20:1089-1098. 115. Lai PP, Bernstein MJ, Kim H, et al. Radiation therapy for stage I and IIA testicular seminoma. Int J Radiat Oncol Biol Phys 1994;28:373-379.

116. Lederman GS, Herman TS, Jochelson M, et al. Radiation therapy of seminoma: 17-year experience at the Joint Center for Radiation Therapy. Radiother Oncol 1989;14:203-208. 117. Lester SG, Morphis JG 2nd, Hornback NB. Testicular seminoma: analysis of treatment results and failures. Int J Radiat Oncol Biol Phys 1986;12:353-358. 118. Maier JG, Van Buskirk K. Treatment of testicular germ cell malignancies. JAMA 1970;213:97-98. 119. Saxena VS. Seminoma of the testis. Am J Roentgenol Radium Ther Nucl Med 1973;117:643-652. 120. Thomas GM, Rider WD, Dembo AJ, et al. Seminoma of the testis: results of treatment and patterns of failure after radiation therapy. Int J Radiat Oncol Biol Phys 1982;8:165-174. 121. Vallis KA, Howard GC, Duncan W, et al. Radiotherapy for stages I and II testicular seminoma: results and morbidity in 238 patients. Br J Radiol 1995;68:400-405. 122. van der Werf-Messing B. Radiotherapeutic treatment of testicular tumors. Int J Radiat Oncol Biol Phys 1976;1:235-248. 123. van Rooy EM, Sagerman RH. Long-term evaluation of postorchiectomy irradiation for stage I seminoma. Radiology 1994;191:857-861. 124. Whipple GL, Sagerman RH, van Rooy EM. Long-term evaluation of postorchiectomy radiotherapy for stage II seminoma. Am J Clin Oncol 1997;20:196-201. 125. Zagars GK, Babaian RJ. Stage I testicular seminoma: rationale for postorchiectomy radiation therapy. Int J Radiat Oncol Biol Phys 1987;13:155-162. 126. Horwich A, Alsanjari N, A’Hern R, et al. Surveillance following orchidectomy for stage I testicular seminoma. Br J Cancer 1992;65:775-778. 127. von der Maase H, Specht L, Jacobsen GK, et al. Surveillance following orchidectomy for stage I seminoma of the testis. Eur J Cancer 1993;29A:1931-1934. 128. Warde P, Gospodarowicz MK, Panzarella T, et al. Stage I testicular seminoma: results of adjuvant irradiation and surveillance. J Clin Oncol 1995;13:2255-2262. 129. Warde PR, Gospodarowicz MK, Goodman PJ, et al. Results of a policy of surveillance in stage I testicular seminoma. Int J Radiat Oncol Biol Phys 1993;27:11-15. 130. Duchesne GM, Horwich A, Dearnaley DP, et al. Orchidectomy alone for stage I seminoma of the testis. Cancer 1990;65: 1115-1118. 131. Whitmore WF. The Marks et al article review ed. Oncology 1992;(6):51-52. 132. Thomas GM. Controversies in the management of testicular seminoma. Cancer 1985;55:2296-2302. 133. Fosså SD, Horwich A, Russell JM, et al. Optimal planning target volume for stage I testicular seminoma: a Medical Research Council randomized trial. Medical Research Council Testicular Tumor Working Group. J Clin Oncol 1999;17:1146. 134. Hanks GE, Peters T, Owen J. Seminoma of the testis: long-term beneficial and deleterious results of radiation. Int J Radiat Oncol Biol Phys 1992;24:913-919. 135. Read G, Johnston RJ. Short duration radiotherapy in stage I seminoma of the testis: preliminary results of a prospective study. Clin Oncol (R Coll Radiol) 1993;5:364-366. 136. Hay JH, Duncan W, Kerr GR. Subsequent malignancies in patients irradiated for testicular tumours. Br J Radiol 1984;57: 597-602. 137. Moller H, Mellemgaard A, Jacobsen GK, et al. Incidence of second primary cancer following testicular cancer. Eur J Cancer 1993;29A:672-676. 138. Stein ME, Kessel I, Luberant N, et al. Testicular seminoma stage I: treatment results and long-term follow-up (1968-1988). J Surg Oncol 1993;53:175-179. 139. van Leeuwen FE, Stiggelbout AM, van den Belt-Dusebout AW, et al. Second cancer risk following testicular cancer: a follow-up study of 1,909 patients. J Clin Oncol 1993;11:415-424. 140. Kleinerman RA, Liebermann JV, Li FP. Second cancer following cancer of the male genital system in Connecticut, 1935-82. Natl Cancer Inst Monogr 1985;68:139-147.

27  The Testicle 141. Kiricuta IC, Sauer J, Bohndorf W. Omission of the pelvic irradiation in stage I testicular seminoma: a study of postorchiectomy paraaortic radiotherapy. Int J Radiat Oncol Biol Phys 1996;35:293-298. 142. Sedlmayer F, Joos H, Deutschmann H, et al. [Long-term tumor control and fertility after para-aortic limited radiotherapy of stage I seminoma]. Strahlenther Onkol 1999;175:320-324. 143. Herman JG, Sturgeon J, Thomas GM. Mediastinal prophylactic irradiation in seminoma [abstract]. Proc Am Soc Clin Oncol 1983;2:133. 144. Speer TW, Sombeck MD, Parsons JT, et al. Testicular seminoma: a failure analysis and literature review. Int J Radiat Oncol Biol Phys 1995;33:89-97. 145. Anscher MS, Marks LB, Shipley WU. The role of radiotherapy in patients with advanced seminomatous germ cell tumors. Controversies in management. Part 2. Oncology (Williston Park) 1992;6:97-104; discussion 107-108, 110. 146. Andrews CF, Micaily B, Brady LW. Testicular seminoma. Results of a program of megavoltage irradiation. Am J Clin Oncol 1987;10:491-495. 147. Ball D, Barrett A, Peckham MJ. The management of metastatic seminoma testis. Cancer 1982;50:2289-2294. 148. Gregory C, Peckham MJ. Results of radiotherapy for stage II testicular seminoma. Radiother Oncol 1986;6:285-292. 149. Hussey DH, Luk KH, Johnson DE. The role of radiation therapy in the treatment of germinal cell tumors of the testis other than pure seminoma. Radiology 1977;123:175-180. 150. Kellokumpu-Lehtinen P, Halme A. Results of treatment in irradiated testicular seminoma patients. Radiother Oncol 1990;18: 1-7. 151. Mason BR, Kearsley JH. Radiotherapy for stage 2 testicular seminoma: the prognostic influence of tumor bulk. J Clin Oncol 1988;6:1856-1862. 152. Sagerman RH, Kotlove DJ, Regine WF, et al. Stage II seminoma: results of postorchiectomy irradiation. Radiology 1989;172: 565-568. 153. Willan BD, McGowan DG. Seminoma of the testis: a 22-year experience with radiation therapy. Int J Radiat Oncol Biol Phys 1985;11:1769-1775. 154. Horwich A, Shipley J, Huddart R. Testicular germ-cell cancer. Lancet 2006;367:754-765. 155. Oliver RTD, Mason MD, Mead GM, et al. Radiotherapy versus single-dose carboplatin in adjuvant treatment of stage I seminoma: a randomised trial Lancet 2005;366:293-300. 156. Zagars GK, Ballo MT, Lee AK, et al. Mortality after cure of testicular seminoma. J Clin Oncol 2004;22:640-647. 157. Travis LB, Fosså SD, Schonfeld SJ, et al. Second cancers among 40,576 testicular cancer patients: focus on long-term survivors. J Natl Cancer Inst 2005;97:1354-1365. 158. Travis LB, Andersson M, Gospodarowicz M, et al. Treatment­associated leukemia following testicular cancer. J Natl Cancer Inst 2000;92:1165-1171. 159. van den Belt-Dusebout AW, Nuver J, de Wit R, et al. Long-term risk of cardiovascular disease in 5-year survivors of testicular cancer. J Clin Oncol 2006;24:467-475. 160. Quek ML, Simma-Chiang V, Stein JP, et al. Postchemotherapy residual masses in advanced seminoma: current management and outcomes. Expert Rev Anticancer Ther 2005;5:869-874. 161. Flechon A, Bompas E, Biron P, et al. Management of postchemotherapy residual masses in advanced seminoma. J Urol 2002;168:1975-1979. 162. De Santis M, Becherer A, Bokemeyer C, et al. 2-18Fluoro-deoxyd-glucose positron emission tomography is a reliable predictor for viable tumor in postchemotherapy seminoma: an update of the prospective multicentric SEMPET trial. J Clin Oncol 2004;22:1034-1039.

675

163. Puc HS, Heelan R, Mazumdar M, et al. Management of residual mass in advanced seminoma: results and recommendations from the Memorial Sloan-Kettering Cancer Center. J Clin Oncol 1996;14:454-460. 164. Herr HW, Sheinfeld J, Puc HS, et al. Surgery for a post­chemotherapy residual mass in seminoma. J Urol 1997;157: 860-862. 165. Schultz SM, Einhorn LH, Conces DJJ, et al. Management of postchemotherapy residual mass in patients with advanced seminoma: Indiana University experience. J Clin Oncol 1989;7: 1497-1503. 166. Stomper PC, Jochelson MS, Friedman EL, et al. CT evaluation of advanced seminoma treated with chemotherapy. AJR Am J Roentgenol 1986;146:745-748. 167. Becherer A, De Santis M, Karanikas G, et al. FDG PET is superior to CT in the prediction of viable tumour in post-­chemotherapy seminoma residuals. Eur J Radiol 2005;54:284-288. 168. Johns Putra L, Lawrentschuk N, Ballok Z, et al. 18F-fluorodeoxyglucose positron emission tomography in evaluation of germ cell tumor after chemotherapy. Urology 2004;64:1202-1207. 169. Jones WG, Fosså SD, Mead GM, et al. Randomized trial of 30 versus 20 Gy in the adjuvant treatment of stage I testicular seminoma: a report on Medical Research Council trial TE18, European Organisation for the Research and Treatment of Cancer Trial 30942 (ISRCTN18525328). J Clin Oncol 2005;23:1200-1208. 170. Luna MA. Extragonadal germ cell tumors. In Johnson DE (ed): Testicular Tumors. 2nd ed. Flushing, NY: Medical Examination Publishing, 1976. 171. Raghavan D. Malignant extragonadal germ cell tumors in adults. In Horwich A (ed): Testicular Cancer: Investigation and Management. Baltimore: Williams & Wilkins, 1991. 172. Buskirk SJ, Evans RG, Farrow GM, et al. Primary retroperitoneal seminoma. Cancer 1982;49:1934-1936. 173. Bush SE, Martinez A, Bagshaw MA. Primary mediastinal seminoma. Cancer 1981;48:1877-1882. 174. Jain KK, Bosl GJ, Bains MS, et al. The treatment of extragonadal seminoma. J Clin Oncol 1984;2:820-827. 175. Logothetis CJ, Samuels ML, Selig DE, et al. Chemotherapy of extragonadal germ cell tumors. J Clin Oncol 1985;3:316-325. 176. Cox JD. Primary malignant germinal tumors of the mediastinum. A study of 24 cases. Cancer 1975;36:1162-1168. 177. Dueland S, Stenwig AE, Heilo A, et al. Treatment and outcome of patients with extragonadal germ cell tumours—the Norwegian Radium Hospital’s experience 1979-94. Br J Cancer 1998;77:329-335. 178. Donohue JP. Metastatic pathways of nonseminomatous germ cell tumors. Semin Urol 1984;2:217-229. 179. Johnson DE, Butler JJ. Malignant lymphoma of the testis. In Johnson DE (ed): Testicular Tumors. 2nd ed. Flushing, NY: Medical Examination Publishing, 1976. 180. Turner RR, Colby TV, MacKintosh FR. Testicular lymphomas: a clinicopathologic study of 35 cases. Cancer 1981;48:20952102. 181. Touroutoglou N, Dimopoulos MA, Younes A, et al. Testicular lymphoma: late relapses and poor outcome despite doxorubicinbased therapy. J Clin Oncol 1995;13:1361-1367. 182. Zouhair A, Weber D, Belkacémi Y, et al. Outcome and patterns of failure in testicular lymphoma: a multicenter Rare Cancer Network study. Int J Radiat Oncol Biol Phys. 2002;52:652-656.

28 The Prostate Andrew K. Lee and Alan Pollack NORMAL ANATOMY AND RESPONSE TO IRRADIATION Anatomy Radiation Effects on the Prostate and Surrounding Normal Tissues FEATURES OF PROSTATE CANCER Epidemiology Etiology Detection Histology Grading WORKUP AND STAGING OF PROSTATE CANCER Role of Rectal Examination and Transrectal Ultrasonography Controversies in Prostate Cancer Staging Role of Magnetic Resonance Imaging Biopsy Variables Imaging Studies Lymph Node Dissection PROGNOSTIC FACTORS AND MODELS Biochemical Evidence of Treatment Failure Prostate-Specific Antigen Nadir Level after Radiation Therapy Prostate-Specific Antigen Level, Gleason Score, and Clinical Stage Transurethral Resection for T3 to T4 Disease Cell Proliferation and DNA Content Characteristics Family History Pretreatment Serum Testosterone Level Molecular Markers

Normal Anatomy and Response to Irradiation Anatomy The prostate is a glandular structure that is bounded superiorly by the bladder, inferiorly by the urogenital diaphragm, anteriorly by the pubic symphysis, and posteriorly by the rectum (Fig. 28-1). Typically the size of a walnut in younger men, the prostate enlarges progressively over time in men older than 40 years. The prostate is somewhat ­ triangular, with the apex inferior at the urogenital diaphragm and the base superior at the urinary bladder neck. Anatomically, the prostate is divided into the peripheral zone, transition zone, central zone, and anterior fibromuscular stroma (Fig. 28-2).1,2 The peripheral zone constitutes 70% of the prostate gland and is readily palpated through

676

EXTERNAL BEAM RADIATION THERAPY: TREATMENT PLANNING AND TECHNIQUES Evolution of Techniques Three-Dimensional Treatment Planning and Conformal Radiation Therapy Intensity-Modulated Radiation Therapy Setup Error and Prostate Movement Considerations Localization of the Prostate Apex BRACHYTHERAPY: TREATMENT PLANNING AND TECHNIQUES Evolution of Techniques Planning Methods Evaluation of the Pretreatment Plan Evaluation after Treatment Permanent Implantation of Iodine 125 or Palladium 103 Radioactive Seeds plus External Beam Radiation Therapy Temporary Iridium 192 Implants RADIATION THERAPY RESULTS Radiation Therapy Alone for Favorable Disease Radiation Therapy for Intermediate- to High-Risk Disease Salvage Procedures after Radiation Therapy RADICAL PROSTATECTOMY Clinical and Biochemical End Points after Radical Prostatectomy Surgery for Favorable Disease Surgery for Intermediate-Risk to High-Risk Disease Adjuvant and Salvage Radiation Therapy CRYOTHERAPY

the rectal wall. The ejaculatory ducts course through the central zone to the verumontanum at the prostatic ­urethra. The central zone contains 20% to 25% of the prostate glandular tissue. The transition zone consists of two lateral ­opposing glandular regions around the urethra that tend to enlarge with age because of benign prostatic hypertrophy, causing urinary outlet obstruction. The transition zone normally makes up 5% to 10% of the prostate glandular tissue. The anterior fibromuscular stroma contains very little glandular tissue. The neurovascular bundles that control ­erectile function3 are located posterolaterally to the prostate. Approximately 75% of prostate malignancies originate in the peripheral zone, 20% in the transition zone, and 5% in the central zone. The major routes of disease spread are direct extension, lymphatic drainage, and hematogenous spread. The prostate is encompassed by a capsule except

28  The Prostate

Trigone of bladder Apex of bladder Retropubic space

Prostatic urethra

Symphysis pubis Penile urethra Corpus cavernosum penis Urogenital Bulb diaphragm of penis

Corpus spongiosum penis

A

Rectum (ampulla) Prostate gland

B

677

Orifice of Base of ureter bladder Ductus deferens

Seminal vesicle Ampulla of ductus deferens Prostate gland Membranous urethra

FIGURE 28-1.  Sagittal section through the pelvis shows the relationship of the prostate to the surrounding pelvic structures (A) and enlargement at the region of the prostate and bladder neck (B).

SV

inferiorly at the apex, where it is continuous with the ­superior fascia of the urogenital diaphragm. ­ Penetration of cancer through the capsule is a prognostic factor and is incorporated into the staging system (category T3). ­Direct extension to the anterior rectal wall is rare because Denonvilliers’ fascia acts as a protective barrier. ­Lymphatic ­drainage flows primarily to the periprostatic (obturator) and internal and external iliac lymph nodes and, to a lesser degree, the presacral lymph nodes. Hematogenous spread commonly appears first in the axial skeleton ­ (vertebral bodies and ribs). Batson’s plexus may have a role in ­hematogenous spread.

ED BN AFM

PZ

TZ

Radiation Effects on the Prostate and Surrounding Normal Tissues

A SV

ED

BN PZ TZ AFM

B FIGURE 28-2.  Transparent, three-dimensional images of prostate show the two zones of cancer origin. A representative ­ cancer in each zone is related to a transverse section through the midprostate (curved arrows). A, Transition zone (TZ) ­ cancer is viewed from the rectal surface and base. The cancer conforms to the Z boundary but invades anteriorly into the anterior ­ fibromuscular stroma (arrow). B, Nontransition zone cancer is viewed from the anterior surface. The cancer conforms to the TZ boundary but invades posterolaterally through the capsule (arrow) in the region where nerve penetration of the capsule is common. AFM, anterior fibromuscular stroma; BN, bladder neck; ED, ejaculatory duct; PZ, peripheral zone; SV, seminal vesicle. (Modified from McNeal JE, Villers AA, ­Redwine EA, et al. Capsular penetration in prostate cancer. Significance for natural history and treatment. Am J Surg Pathol 1990;14:240-247.)

The prostate is relatively resistant to radiation. Pelvic irra­ diation of the prostate in the absence of prostate cancer to doses of 45 to 65 Gy reduces the serum prostate-­specific antigen (PSA) level to a median value of 0.5 ng/mL.4,5 Pathologic response to irradiation includes atrophy of glandular tissue, fibrosis, and typical radiation-induced vascular changes.6-9 Significant prostate shrinkage in response to high-dose radiation is seen in roughly 50% of patients. The prostate also may become firmer because of fibrosis. These changes are usually not accompanied by significant changes in urinary function over the long term, although improvement or worsening (possibly requiring a transurethral resection of the prostate [TURP]) may occur. Stricture of the urethra is uncommon and appears mostly in patients who have undergone TURP before radiation therapy.10,11 TURP after radiation therapy should be used with caution because it has been associated with increased rates of ­incontinence.12 Side effects from high-dose radiation therapy for prostate cancer are primarily related to the bladder and rectum, although the dose to the femoral heads also should be considered in treatment planning. Acute bladder side effects during treatment and for 1 to 6 months afterward include increased nocturia, urinary frequency, dysuria, ­hematuria, urgency, bladder spasm, and urinary incontinence. Acute rectal side effects include increased bowel movement ­frequency, pain on defecation, hematochezia, spasm, and fecal incontinence. Ejaculation during the latter half of ­radiation therapy is painful. Symptoms lessen considerably by 1 month after external beam radiation ­therapy (EBRT),

678

PART 8  Male Genital Tract

but chronic symptoms may persist. Urinary ­ symptoms persist for longer periods after brachytherapy, sometimes for more than a year. Acute and late morbidities are traditionally graded on a scale of 0 to 5 using Radiation Therapy ­ Oncology Group (RTOG) criteria13 or modifications thereof (Table 28-1).14,15 Late complications are usually mild to moderate, along the same lines as acute symptoms. Hemorrhagic cystitis or pronounced rectal bleeding from telangiectatic vessels may require coagulation therapy for control. Severe complications such as ulceration and fistula leading to urinary diversion or colostomy occur in less than 1 of 1000 cases. The risk of impotence over the long term is about 50%.16-18

Estimates of normal tissue complication probabilities have improved considerably as a result of dose-volume ­histogram (DVH)–based correlates of these complications. These ­relationships are summarized later in this chapter.

Features of Prostate Cancer Epidemiology Except for skin cancer, prostate cancer is the most common malignancy among men in the United States. An ­estimated 186,320 men were diagnosed with prostate cancer in

TABLE 28-1.    Acute and Delayed Radiation Toxicity Grading Scales System Toxicity Acute

Grade 1

Grade 2

Grade 3

Grade 4

Toxicity* Diarrhea needing parenteral Acute or subacute support; severe mucous obstruction; ­fistula ­discharge requiring extended or perforation; use of sanitary pads; abdominal ­gastrointestinal distention; rectal pain requiring bleeding ­requiring frequent narcotics; gastroinmore than one testinal bleeding requiring one ­transfusion; transfusion ­abdominal pain or tenesmus requiring bowel diversion

Lower gastrointestinal

Increased frequency Diarrhea needing parasymor change in patholytic drugs (e.g. quality of bowel diphenoxylate-atropine habits not needing [Lomotil]); mucous discharge infrequently ­medication; rectal requiring sanitary pads; discomfort not rectal pain needing ­requiring analgesics analgesics or occasional narcotics; mild rectal bleeding

Urinary

Frequency or nocturia hourly or Frequency or nocturia Frequency or nocturia less more; dysuria, pain, or spasm frequent than hourly; twice pretreatment needing frequent narcotics; dysuria, bladder spasm baseline levels; gross hematuria requiring one needing local anesthetic dysuria not needing transfusion; prolonged urinary (e.g., phenazopyridine medication obstruction caused by prostate hydrochloride [Pyridium]) inflammation or clots requiring or occasional narcotics; incatheterization (including frequent gross hematuria; suprapubic) temporary catheterization

Hematuria needing more than one transfusion; hospitalization for sepsis caused by obstruction and/or ulceration or necrosis of the bladder

Late Toxicity Up to two antidiarrheals/wk; More than two antidiarrheals/day; Dysfunction requiring up to two coagulations surgery; perforaat least one blood transfusion or tion; life-threatening more than two coagulations for for bleeding; ­occasional bleeding bleeding; prolonged steroids steroids for ­ulceration; for enemas; hyperbaric oxygen occasional dilation; for bleeding or ulceration; intermittent use of regular dilation; persistent use ­incontinence pads; of incontinence pads; regular regular nonnarcotic or ocnarcotic for pain casional narcotic for pain

Lower gastrointestinal

Excess bowel movements twice baseline level; slight rectal discharge or blood

Urinary

Moderate frequency; Nocturia twice Severe frequency and dysuria; nocturia more than twice baseline level; nocturia more frequent than baseline levels; genermicroscopic hemaonce every hour; reduction alized telangiectasia; turia; slight mucosal in bladder capacity (150 mL); intermittent macroscopic atrophy and minor frequent hematuria requiring hematuria; up to two telangiectasia at least one transfusion; more than two coagulations for ­coagulations; ­intermittent use of ­incontinence pads; ­hematuria; hyperbaric oxygen regular ­nonnarcotic or for bleeding or ulceration; persistent use of incontinence occasional narcotic for pads; regular narcotic for pain pain

*The

Severe ­hemorrhagic cystitis or ­ulceration requiring ­urinary ­diversion or ­cystectomy, or both

acute scale was based on the modified Radiation Therapy Oncology Group criteria. late scale was based on criteria from the Radiation Therapy Oncology Group and the Late Effects Normal Tissue Task Force. Modified from Storey MR, Pollack A, Zagars GK, et al. Complications from dose escalation in prostate cancer: preliminary results of a randomized trial. Int J Radiat Oncol Biol Phys 2000;48:635-642.

†The

28  The Prostate

2008, with an estimated 28,660 deaths.19 Prostate cancer is second to lung cancer as a cause of cancer-related death among men in the United States. Mortality rates from prostate cancer are the highest among African American men worldwide. New developments in screening for prostate cancer, such as the use of serum PSA measurements, the free-to-total PSA ratio, and prostate biopsy guided by transrectal ultrasonography (TRUS), have contributed to a shift to earlier detection of the disease.20-24 The introduction of these improvements produced a surge in the number of cases of prostate cancer diagnosed between 1983 and 1991.25,26 During this period, cases were diagnosed earlier (i.e., at lower disease stages), the age at diagnosis dropped, and the use of radical prostatectomy increased dramatically. As a consequence, a unique problem faces physicians treating this disease: determining which tumors are clinically relevant and should be treated and which may be latent and should be observed. Autopsy studies performed before the use of PSA testing documented that at least 40% of men in their 70s and higher percentages of men in their 80s had occult, latent prostate cancer that might not have needed treatment.27-30 In later series, Sakr and colleagues31,32 found that more than one third of men in their fifth decade of life harbor malignant cells in their prostates. These astonishing statistics suggest that some cases that are diagnosed by means of aggressive staging may not require treatment.

Etiology Numerous theories have been proposed concerning the cause of prostate cancer, and progress has been made in identifying possible genetic factors. Familial clustering has been documented by several investigators.33-35 Inherited risk has been linked to loci on the X chromosome36 and on chromosome 1.37 Mutations in BRCA1 or BRCA2 also may be involved.38 However, genetic factors are believed to account for a relatively small proportion of prostate cancer. Diet may account for at least 25% of all cases of prostate cancer. Migration from low-risk, low-fat-intake countries such as Japan to high-risk, high-fat-intake countries such as the United States is associated with a substantial increase in the incidence of prostate cancer.39,40 Animal studies ­suggest that the higher risk is not attributable solely to fat intake and that the combination of ­greater dietary fat ­ intake accompanied by other dietary factors (e.g., ­fiber ­intake, protein intake) may contribute. Diet may affect prostate cancer development by altering circulating ­hormone levels.40,41 Another possible etiologic factor is ­vasectomy; some studies have maintained and others have refuted vasectomy as a causal factor. Other data suggest that a small subset of prostate cancers has a viral cause.42 Endogenous androgen and estrogen levels have long been hypothesized to influence prostate cancer development and progression. Androgens are responsible for sustaining the equilibrium between cell loss and cell gain in the normal prostate. When androgen is withdrawn in rats, 90% of the prostate epithelial cells die by apoptosis over a 10-day period.43,44 Although the administration of androgens to castrated male rats causes prostate cell proliferation, equilibrium eventually is reestablished when cell numbers are replenished. The data concerning endogenous androgen levels and the risk of prostate cancer are not clear, but some relationships have emerged. Serum testosterone

679

c­ oncentrations have been higher in patients at the time of prostate cancer diagnosis than in age-matched ­ control subjects in some studies. However, African American men have higher serum testosterone levels until about age 35 years,45 which may be a predisposing factor for prostate cancer.46 Findings from Ross and colleagues47 suggest a relationship between prostate cancer and dihydroxytestosterone metabolites. Androgens do seem to play a role in prostate carcinogenesis. Administration of androgens to rats in vivo causes adenocarcinoma of the prostate.48,49 Androgens also act as a promoter when administered in conjunction with chemical carcinogens that target the prostate,50 with cadmium,51 or with ionizing radiation.52 The promoter action of androgens may be important in human carcinogenesis.

Detection The use of the PSA assay in the clinic and the refinements in TRUS-directed biopsy of the prostate have vastly improved the ability to detect prostate cancer. PSA is a 26-kd glycoprotein that is produced by epithelial cells in the prostate. A serine protease, PSA has chymotrypsin-like activity and causes liquefaction of semen. The serum PSA level is proportional to prostate volume, both of which increase with age and are strong determinants of outcome.53-56 Lepor and colleagues57 calculated that for every 1 cm3 of normal prostate epithelium, the amount of PSA in the serum is 0.33 ng/mL. Because about 20% of the prostate consists of epithelial cells,58 the serum PSA level for men with benign prostate tissue is estimated at about 0.066 ng/mL, an amount similar to estimates by Oesterling and colleagues20 and Collins and colleagues.59 The serum PSA concentration is elevated in the presence of prostate cancer because of the disruption of the normal glandular pattern by the tumor. Even though high-grade, poorly differentiated tumors are associated with higher serum PSA values, the serum PSA amount contributed per tumor volume is inversely proportional to the Gleason score.60 Higher-grade tumors are larger and disrupt more of the gland, thereby compensating for the lower serum PSA per tumor volume such that serum PSA levels are generally higher overall. The serum PSA level has been roughly correlated with prostate tumor volume.53-56 Stamey and colleagues61 estimated that for every 1 cm3 of prostate cancer, the serum PSA level is elevated by 3.5 ng/mL. Because normal prostate volume and inflammation increase with age, Oesterling and colleagues21 proposed age-specific PSA cut points for the detection of prostate cancer. For example, a serum PSA level of more than 4 ng/mL is considered ­abnormal, but for a 70-year-old man, a better cut point may be 6.5 ng/mL, and for a 50-year-old man, the better cut point may be 3.5 ng/mL. Modifications to the age-specific reference ranges based on race have been proposed, with lower PSA cut points by age proposed for African American men than for white men.23,62 Multiple forms of PSA coexist in the serum, and all ­except the free PSA form are in complex with protease inhibitors. For unknown reasons, the concentration of free PSA in patients with cancer is reduced. For men older than 50 years, when the free-to-total PSA ratio is less than 25% and the total PSA level is between 4 and 10 ng/mL, the risk of prostate cancer is increased.23,63,64 Prostate cancer detection using free PSA measurements may be better

680

PART 8  Male Genital Tract

than ­using the age-specific PSA reference ranges.64 The use of free PSA in screening has gained acceptance, and it is a valuable aid in deciding whether to perform a prostate biopsy. TRUS-guided prostate biopsy for diagnostic purposes has evolved considerably over the past 15 years. At a minimum, sextant (six-core) biopsies should be performed, with additional samples taken from hypoechoic areas. The sextant biopsy scheme, however, is associated with a high rate of subsequent positive biopsy findings after a negative result for the first sextant biopsy.65 Repeat sextant biopsies have become common practice for men with persistently elevated PSA levels, a high PSA velocity (i.e., rate of PSA rise),66 or a low free-to-total PSA ratio. Later studies have indicated that diagnostic precision is improved by obtaining 8 to 12 biopsy samples. The increase in the number of prostate needle biopsy samples from six was predicated on the finding that cancer was diagnosed in approximately 25% of patients at a second biopsy. Through tumor modeling based on radical prostatectomy specimens,67,68 prostate areas such as the anterior horns, transition zone, and periurethral region have been added to the sampling sites, and their inclusion has reduced the false-negative rate.24,69,70

Histology More than 95% of malignant prostatic neoplasms are ­adenocarcinomas arising from the acinar cells. One of the more common secondary histologic entities is ductal ­(endometrioid) adenocarcinoma, which classically has been described as arising from the large periurethral prostatic ducts or verumontanum. Ductal adenocarcinomas typically display a papillary or cribriform pattern and were initially thought to be less aggressive,71 although later ­evidence suggests the opposite.72 Ductal carcinomas are very often associated with acinar adenocarcinomas; they are responsive to androgen ablation (AA), they secrete PSA, and they have been observed in the peripheral zone of the prostate, suggesting that they may be variants of acinar ­adenocarcinoma.73 Neuroendocrine cells are prevalent in the prostate, particularly in the prostatic ducts, and are part of the amine precursor uptake and decarboxylation system.74 The presence of focal neuroendocrine differentiation in prostatic adenocarcinomas is common and of unclear prognostic significance. Small cell carcinomas of the prostate are highly aggressive anaplastic tumors that tend to metastasize early, do not secrete PSA, are rarely associated with paraneoplastic syndromes, and sometimes occur in the setting of acinar carcinoma.75,76 Other primary prostate cancer variants are rare and include transitional cell carcinoma, sarcomatoid carcinoma, mucinous adenocarcinoma, signet ring carcinoma, lymphoepithelioma-like carcinoma, squamous and adenosquamous carcinoma, adenocystic (basaloid) carcinoma, and mesenchymal tumors such as rhabdomyosarcomas (in children) and leiomyosarcomas (in adults).77

Grading The Gleason scoring system78,79 is the most ­ commonly used grading system for adenocarcinomas of the ­prostate, ­although more than 30 other systems have been described.80 The Gleason score consists of the sum of the major and minor glandular patterns, graded on a scale of 1 to 5. An anaplastic appearance, with the absence of glandular structures, is consistent with a Gleason grade of 5.

Discrete glandular formation with slight disorganization is ­ consistent with a Gleason grade of 1. The combined ­Gleason score ranges from 2 to 10 and has been shown to correlate strongly with patient outcome in terms of ­biochemical, clinical, and survival end points. Because the prognostic differences between each of the nine Gleason scoring categories are slight and require large patient numbers to observe, the scores are usually lumped into four or fewer groups. One such four-tier system81 ­consists of Gleason scores 2 and 3, 4 to 6, 7, and 8 to 10. A Gleason score of 7 has been established as connoting a significantly worse prognosis than a Gleason score of 6 or less and a significantly better prognosis than a ­Gleason score of 8 or more. It is common for Gleason 7 to be ­categorized separately and included in the criteria for ­classifying ­ intermediate-risk disease. Cases of Gleason 7 disease can be further subdivided according to the major plus minor patterns of Gleason 3 + 4 versus 4 + 3. The latter pattern is associated with a worse outcome,82 and its presence has been used as a basis for treatment decisions. Along these lines, the proportion of Gleason 4 disease present in the specimen and the presence of tertiary Gleason grade 5 ­disease are of prognostic value.56,83,84

Workup and Staging of Prostate Cancer Role of Rectal Examination and Transrectal Ultrasonography Fundamental to the staging of prostate cancer is the digital rectal examination, the classification of which has changed minimally since the Whitmore-Jewett system.85,86 The 1992 version of the American Joint Committee on Cancer (AJCC) staging system87 (Table 28-2) incorporated findings from TRUS, adding a new dimension that has resulted in stage migration. For example, a palpable nodule limited to the prostate on digital rectal examination that would be classified as category T2 could be classified as T3 if the sonographic examination indicated extracapsular extension. The adoption of TRUS for staging might have been premature, because several studies, including significant randomized trials,88,89 have shown it to be inaccurate in staging prostate cancer. Prostate tumors are hypoechoic in 65%, isoechoic in 30%, and hyperechoic in 5% of cases.90 Biopsy sampling of hypoechoic areas does not always yield tumor, and the diagnostic accuracy may be as low as 50%. A classification of T1c assigned by palpation and TRUS implies that the tumor was isoechoic; if hypoechoic or hyperechoic abnormalities were present, the tumor would be classified as T2 or T3. The application of TRUS in this manner has been nonuniform and confusing. The most significant inaccuracy with regard to the use of TRUS for staging comes from the designation of T3 disease on the basis of inhomogeneity of the prostate boundary echo or seminal vesicle invasion. Although some studies have supported the use of TRUS for the detection of T3 disease,91-95 others have not.88,89,96 TRUS is less accurate for detecting seminal vesicle invasion than it is for detecting extracapsular extension.88,96 Perhaps more important is the lack of clinically relevant correlation of stage—or upstaging or downstaging—as assessed by TRUS with clinical outcome.97,98 Palpatory staging by

28  The Prostate

TABLE 28-2.  Differences Between the 1992 and 1997 American Joint Committee on Cancer Staging Systems with Regard to Tumor Palpability Tumor Attributes

1992 System

1997 System

<5% of resected tissue involved

T1a

T1a

>5% of resected tissue involved

T1b

T1b

PSA-directed biopsy

T1c

T1c

Less than one half of one lobe involved

T2a

T2a

More than one half of one lobe involved

T2b

T2a

Both lobes involved

T2c

T2b

T3a

T3a

Nonpalpable

Palpable without ECE

Palpable with ECE Unilateral involvement Bilateral involvement

T3b

T3a

Seminal vesicle involvement

T3c

T3b

ECE, extracapsular extension; PSA, prostate-specific antigen.

digital rectal examination is recommended and should be defined as such in all documentation.

Controversies in Prostate Cancer Staging Two other aspects of staging are controversial: the ­absence of PSA levels in the staging criteria and the ­ changes to those criteria that were made between 1992 and 1997 and between 1997 and 2002. Staging criteria are based on relationships to survival. Although a high PSA level before treatment is the most significant predictor of a rising PSA level and local recurrence after treatment (i.e., biochemical failure), an association between pretreatment PSA level and survival81,99,100 has yet to be established. Nonetheless, the pretreatment PSA level has been shown to be a strong correlate of distant metastasis101,102 and is always considered in case management along with the Gleason score and stage. Eventually, the pretreatment PSA level will be included in prostate cancer ­staging systems. The implementation of the 1997 AJCC staging system brought about a consolidation of some of the categories (see ­ Table 28-2). However, comparative studies indicate that the 1992 system should not have been abandoned,83,103-106 and this has resulted in re-expansion of the T2 category in the 2002 AJCC staging system (Box 28-1).107 For this reason, when disease stage is ­reported, the year of the AJCC staging ­ system used should be designated. Another drawback of the 1997 and 2002 staging systems lies in the accompanying statement that all available information can be used in the staging process.107 This statement implies, but does not specify, that biopsy information may be included in the staging; for example, a palpable T2a lesion may be upstaged to T2b if bilateral positive ­biopsy samples are obtained. The consequent stage migration

681

makes comparisons of results with prior studies that used prior staging systems impossible. Use of strict staging criteria for palpable and nonpalpable tumors is preferable.

Role of Magnetic Resonance Imaging Endorectal coil magnetic resonance imaging (MRI) is gradually gaining a foothold in staging prostate cancer. The accuracy of endorectal coil MRI in determining extraprostatic extension is far greater than that of TRUS.88 D’Amico and colleagues108,109 reported that endorectal coil MRI findings correlate with surgical outcome. Given the improved ­accuracy of endorectal coil MRI in defining prostate anatomy (e.g., apex of the prostate110) and in identifying suspicious intraprostatic areas,111 extracapsular extension, and seminal vesicle invasion,112-114 treatment planning using three-dimensional conformal radiation therapy (CRT) or intensity-modulated radiation therapy (IMRT) should be enhanced. Magnetic resonance spectroscopy permits the visualization of choline-rich areas, which are related to bulky tumor regions identified at pathology examination.115 Dose-painting techniques that use IMRT to ­focally escalate the dose per fraction in choline-rich areas are ­being developed.116,117 Functional imaging methods have tremendous promise in radiation treatment planning for prostate cancer and will undoubtedly become more commonplace over time.

Biopsy Variables The systematic sampling of six or more regions from the prostate for diagnosis has led to the inspection of biopsy tissue variables. The number of biopsy samples positive for tumor and the proportion of tumor in the samples118-124 are usually predictive of adverse pathologic features and patient outcome independent of pretreatment PSA level, tumor grade, and disease stage. Biopsies can, however, give inaccurate or misleading results. For example, the finding of a small amount of tumor in one of six biopsy samples may not equate to a low tumor burden.125,126 Nevertheless, the presence of multiple positive tumor cores with extensive tumor involvement is commensurate with a large tumor burden. Caution should be used in basing treatment decisions on biopsy findings, although the predictive value of acquiring more than six core samples may enhance accuracy. Qualitatively, estimates of tumor extent from biopsy samples have been used for years, albeit subjectively, to identify candidates for brachytherapy or other radiation therapy options. Additional studies leading to a consensus about how to apply such data are needed. Processing and grading of diagnostic material from prostate needle biopsies vary greatly. Because segregation of individual biopsy cores raises cost, biopsy material is occasionally pooled according to the side of the gland it was removed from (right or left). In most major cancer treatment centers, each core is processed individually. Segregation of the biopsy specimens allows quantification of the percentage of tumor in each core, the number of positive cores, and the Gleason score of each specimen. The generation of a Gleason score for each biopsy specimen has led to the use of the highest Gleason score as a prognostic variable. Conceptually, however, the Gleason score is a composite assessment of all of the tumor tissue obtained. If one biopsy sample shows a focus of Gleason 8 and three samples show substantial Gleason 6 tumor content, the global

682

PART 8  Male Genital Tract

BOX 28-1.  2002 American Joint Committee on Cancer Staging System for Prostate Cancer Primary Tumor, Clinical (T) TX Primary tumor cannot be assessed T0 No evidence of primary tumor T1 Clinically inapparent tumor not palpable nor visible by imaging T1a Tumor incidental histologic finding in 5% or less of tissue resected T1b Tumor incidental finding in more than 5% of tissue resected T1c Tumor identified by needle biopsy (e.g., because of elevated PSA) T2 Tumor confined within prostate * T2a Tumor involves one-half of one lobe or less T2b Tumor involves more than one-half of one lobe but not both lobes T2c Tumor involves both lobes T3 Tumor extends through the prostate capsule † T3a Extracapsular extension (unilateral or bilateral) T3b Tumor invades seminal vesicle(s) T4 Tumor is fixed or invades adjacent structures other than seminal vesicles: bladder neck, external sphincter, rectum, levator muscles, and/or pelvic wall Primary Tumor, Pathologic (pT) pT2 ‡ Organ confined pT2a Unilateral, involving one-half of one lobe or less pT2b Unilateral, involving more than one-half of one lobe but not both lobes pT2c Bilateral pT3 Extraprostatic extension pT3a Extraprostatic extension pT3b Seminal vesicle invasion pT4 Invasion of bladder, rectum Regional Lymph Nodes (N) NX Regional lymph node(s) cannot be assessed (e.g., previously removed) N0 No regional lymph node metastasis N1 Metastasis in regional lymph node or nodes Distant Metastasis (M) $ MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis M1a Nonregional lymph node(s) M1b Bone(s) M1c Other site(s) Histopathologic Grade (G) ¶ GX Grade cannot be assessed G1 Well differentiated (slight anaplasia) (Gleason 2-4) G2 Moderately differentiated (moderate anaplasia) (Gleason 5-6) G3-4 Poorly differentiated or undifferentiated (marked anaplasia) (Gleason 7-10) Stage Grouping Stage I T1a Stage II T1a T1b T1c T1 T2 Stage III T3 Stage IV T4 Any T Any T

N0 N0 N0 N0 N0 N0 N0 N0 N1 Any N

M0 M0 M0 M0 M0 M0 M0 M0 M0 M1

G1 G2-4 Any G Any G Any G Any G Any G Any G Any G Any G

* Tumor found in one or both lobes by needle biopsy, but not palpable or reliably visible by imaging, is classified as T1c. † Invasion into the prostatic apex or into (but not beyond) the prostatic capsule is not classified as T3, but as T2. ‡ There is no pathologic T1 classification. $ When more than one site of metastasis is present, the most advanced category is used; pM1c is the most advanced. ¶ Gleason score is considered to be the optimal method of grading. From Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer-Verlag, 2002.

28  The Prostate

Gleason score would be less than Gleason 8. Because the percentage of Gleason grade 4 to 556,83 and designation of a tertiary Gleason score may have prognostic value,84,127 the scoring of individual biopsy specimens may prove to be important. Few studies have compared the relationship of the highest biopsy Gleason score with the global biopsy Gleason score.84,128

Imaging Studies The use of the PSA assay in screening and the use of more systematic TRUS-guided prostate biopsy sampling have resulted in a dramatic reduction in the locoregional extent of disease at diagnosis.83,129,130 As a consequence, the incidence of positive findings on bone scans or pelvic computed tomography (CT) scans has lessened considerably. Detection of metastasis on bone scans is related to the pretreatment PSA value, Gleason score, and disease stage.131 The association with the pretreatment PSA level is pronounced. Bone scanning is recommended if the PSA value is higher than 10 ng/mL,132 though some would argue that a PSA level of more than 20 ng/mL should be used.20 Even though pelvic CT scanning is recommended for patients with a high-risk feature, such as a PSA level of more than 20 ng/mL, Gleason score of 8 or more, or category T3 or higher, lymph node enlargement is rarely found. Another imaging test that has yet to gain wide acceptance is the ProstaScint scan (Cytogen Corp., Princeton, NJ) for imaging the prostate or prostate cancer cells. The test uses an indium 111–labeled antibody (capromab ­pendetide [ProstaScint]) to prostate-specific membrane antigen, a transmembrane glycoprotein that is expressed mainly on cells of prostate origin. The ProstaScint scan has been used clinically as an adjunctive test for staging high-risk disease, for identifying lymph node or hematogenous spread, and for characterizing sites of recurrence after ­definitive therapy.133-137 Imaging of prostate-specific membrane antigen has been used to visualize lymph node ­involvement in patients with high-risk disease who are ­being considered for definitive local therapy; to visualize regional and distant spread in patients who have experienced biochemical failure after radiation therapy and are being considered for salvage prostatectomy or cryosurgery; and to detect isolated recurrences in the prostate bed in patients who show rising PSA levels and are good candidates for salvage radiation therapy. Although the ProstaScint scan has produced generally encouraging results, it does have a significant rate of false-positive findings. Improvements in single photon emission computed tomography (SPECT) and CT fusion techniques should reduce these errors. Newer imaging techniques involving the administration of lymphotropic superparamagnetic nanoparticles followed by MRI have shown promise in early studies. These techniques may represent a noninvasive method for detecting metastases in pelvic lymph nodes that are not visibly ­enlarged.138

Lymph Node Dissection The routine use of lymph node dissection as a staging tool for patients desiring definitive radiation therapy has largely been abandoned. For patients with favorable risk factors, the risk of lymph node involvement is less than 5%.83,103,130,139 The incidence of lymph node involvement has decreased in the PSA era.83,130 For those with ­intermediate-risk features,

683

the incidence of pelvic lymph node spread is about 10% to 15%. Patients with high-risk features, particularly category T3 disease, have a 40% to 50% risk of regional spread.139 Because the standard of care for patients with high-risk prostate cancer is radiation therapy in combination with AA and because use of this combination seems to be effective for patients with lymph node involvement,139 lymph node dissection offers little benefit. The only justification for performing a lymph node dissection is if the result would alter treatment. For example, lymph node dissection may influence the decision to use permanent rather than extended temporary AA if lymph nodes are positive for disease. Otherwise, the radiation therapy management of regionally lymph node–positive (stage D1) disease and of clinically defined high-risk disease (i.e., PSA value less than 20 ng/mL, Gleason score 8 or higher, or T3 or higher disease) would be the same.

Prognostic Factors and Models Biochemical Evidence of Treatment Failure Since the rising PSA profile was identified as a surrogate end point, exploration of potential predictors of prostate cancer outcome has expanded rapidly. Pretreatment PSA levels, Gleason score, and disease stage are the three core factors widely considered to be essential in any evaluation of therapeutic options. Terms such as locally advanced will always be associated with an increased risk of treatment failure, but these terms describe only the staging portion of risk assessment. The preferred designations of risk assign patients to low-, intermediate-, or high-risk groups on the basis of PSA level, Gleason score, and disease stage. Several models have been proposed to approximate the risk of adverse pathologic features after radical prostatectomy and the risk of treatment failure after radiation therapy and radical prostatectomy.101,139-143 Central to modeling is the use of biochemical failure (i.e., rising PSA levels) as an end point. Biochemical failure is a harbinger of eventual clinical relapse, with an average delay of 2 to 5 years.144,145 For most men with a rising PSA level after radiation therapy, biopsy of the prostate indicates residual tumor, suggesting that biochemical failure most often represents persistence of local disease.146,147 Intensive biopsy of the vesicourethral anastomosis after prostatectomy when evidence of biochemical failure is present also yields evidence of local persistence in 30% to 50% of cases.148,149 The question that remains is whether biochemical failure is a surrogate for eventual distant metastasis and death. Crucial to this determination is a follow-up period that is long enough to allow disease to progress from local to distant relapse to death for patients without micrometastatic disease at diagnosis. The relationship between rising PSA levels and the development of distant metastasis is incontrovertible. Pound and colleagues150 followed a cohort of patients treated solely with radical prostatectomy who did not undergo early salvage treatment after biochemical failure. In that study, 34% of men who experienced biochemical failure developed distant metastasis during the study period. The 8-year actuarial rate of distant metastasis, calculated from the time that the PSA level rose, was 50%.

684

PART 8  Male Genital Tract

In that study, all patients who developed distant metastasis had biochemical failure first. Many other groups have reported that biochemical failure is a precursor to distant metastasis in all but a few cases.107,151-149 Further complicating the use of biochemical failure as an end point are inconsistencies in its definition. ­According to the American Society for Therapeutic Radiology and ­Oncology (ASTRO) consensus guidelines, three consecutive increases in the PSA level constitute treatment ­failure.154 Although this definition has helped to standardize research efforts, some situations can lead to misclassification. For example, PSA levels have been observed to transiently rise 1 to 3 years after radiation therapy and then fall back to a stable low level for no apparent reason. This ­so-called PSA bounce phenomenon occurs most often after brachytherapy with permanent radioactive seeds,155 but it also has been documented after EBRT.156 Another consideration in defining biochemical failure is that after AA plus radiation therapy, the PSA level usually reaches a nadir in the undetectable range and then rises, after some delay, after the AA is discontinued. It is not uncommon to see three consecutive increases and then a leveling off or even a fall in the PSA level. Characterization of treatment failure in patients treated with AA may best be accomplished by setting a PSA cut point (e.g., 1.5 ng/mL) and declaring that PSA levels in excess of this amount indicate biochemical failure.157 Because of these shortcomings, a second consensus conference, sponsored by the ASTRO and RTOG, was convened in 2005 to recommend a new definition of biochemical (PSA) failure. After examining several definitions of biochemical failure and their correlation to clinical outcomes, the Consensus Conference committee proposed the so-called Phoenix definition: a PSA rise of 2 ng/mL or more above the absolute nadir PSA level after radiation therapy, with or without hormonal therapy.158 They also recommended that the date of failure be recorded at the time of the PSA increase and not backdated (as had been the case in the prior definition). The primary reasons for this new definition of PSA failure were to improve the overall accuracy (sensitivity and specificity) of the previous definition, to allow comparisons between several types of radiation modalities (e.g., EBRT versus EBRT with hormonal ­therapy versus brachytherapy), to account for the PSA bounce phenomenon, to eliminate the bias introduced by backdating, and to have a definition that was linked to actual clinical outcomes. The Consensus Conference committee also allowed the continued use of the prior Consensus definition of biochemical failure (three consecutive increases in PSA level after radiation therapy alone [i.e., radiation without hormone therapy]) as long as the stated date of control was 2 years short of the median follow-up time. The committee emphasized that neither of these definitions should be used for treatments that include nonradiation modalities (e.g., surgery) and that they should not be used as the sole basis for making clinical decisions.

Prostate-Specific Antigen Nadir Level after Radiation Therapy Some investigators159 have advocated using a strict PSA nadir cut point as an indicator of freedom from treatment failure after radiation therapy. The ASTRO consensus154 was that the nadir PSA level after radiation therapy is not the

sole determinant of outcome after that therapy, although it can be used along with pretreatment PSA level, Gleason score, and stage as a prognostic factor (Fig. 28-3).160-162 The lower the PSA nadir level after radiation therapy, the lower the likelihood of a rising PSA profile. The PSA nadir value can vary according to the type of radiation therapy administered (e.g., brachytherapy versus EBRT) because of differences in the destruction of normal prostate cells. PSA nadir levels are usually established at more than 3 years after radiation therapy and are typically grouped as being less than or equal to 0.5 ng/mL, more than 0.5 to 1.0 ng/mL, more than 1.0 to 2.0 ng/mL, and more than 2.0 ng/mL. As shown in Figure 28-3, the associated 5-year rates of freedom from biochemical failure (according to ­rising PSA levels) are 81%, 58%, 44%, and 6%, respectively. The earliest and most robust determinant of eventual clinical relapse is a rise in the PSA level, and the PSA nadir level is a strong posttreatment prognostic factor that is independent of pretreatment PSA, Gleason score, and stage. The nadir PSA measurement approximates the risk of eventual biochemical relapse, but use of this value by itself without considering other prognostic factors would result in the misclassification of failure in many cases. More importantly, the nadir PSA level is a research variable and should not be applied clinically. Many patients with nadir PSA levels between 0.5 and 2.0 ng/mL have no biochemical evidence of treatment failure for 5 to 10 years.163

Prostate-Specific Antigen Level, Gleason Score, and Clinical Stage Pretreatment PSA, Gleason score, and clinical disease stage are independent correlates of treatment success, which for prostate cancer is defined as having no biochemical evidence of disease (bNED), also referred to as freedom from biochemical failure. Mathematical modeling that incorporates biochemical (i.e., PSA) evidence of success after ­radiation therapy has allowed more precise definition of risk groupings.101,141,143,164-167 Most models assign patients to prognostic bins such as those defined by the AJCC staging classification. Movsas and colleagues168 compared several of these bin models using a cohort of 421 patients treated at Fox Chase Cancer Center in the PSA era and found that the ­differences among the models in ability to predict rates of freedom from biochemical failure were minor. The most common prognostic grouping assigns patients to favorable-risk, intermediate-risk, and high-risk groups.165,166 Patients in the favorable-risk group have a pretreatment PSA level of 10 ng/mL or less, a Gleason score of less than 7, and category T1 to T2a disease (according to the 1992 AJCC staging system). Patients with a pretreatment PSA level of more than 20 ng/mL, a Gleason score of 8 to 10, or ­category T3 to T4 disease are considered to be at high risk for biochemical failure. Intermediate-risk patients fall ­ between these two groups and include cases with no high-risk features but a pretreatment PSA level of 10 to 20 ng/mL, a Gleason score of 7 or less, or category T2b to T2c disease. The KaplanMeier curves for freedom from biochemical failure among these risk groups in a cohort from The University of Texas M. D. Anderson Cancer Center are shown in Figure 28-4. Refinements to this grouping will be forthcoming, and in the future, bin modeling may give way to more precise, continuous-function estimates.

28  The Prostate 1.0

1.0

Grouped Gleason score

PSA > 4–10 (n = 525) 0.6 PSA > 10–20 (n = 335)

0.4 0.2

Fraction free of failure

Fraction free of failure

PSA ≤ 4 (n = 200) 0.8

685

0.8

2–4 (n = 197)

0.6 7 (n = 319)

0.4

5&6 (n = 319)

8–10 (n = 145) 0.2

PSA > 20 (n = 153) 0.0

0.0 0

20

A

40

60

0

80 100 120 140 160

Months after radiotherapy

40

60

80 100 120 140 160

Months after radiotherapy 1.0

0.8

T1–T2a (n = 601)

0.6

T2b–T2c (n = 292)

0.4 T3–T4 (n = 320)

0.2

Fraction free of failure

1.0 Fraction free of failure

20

B

0.0

0.8

Nadir ≤ 0.5

0.6

Nadir > 0.5–1.0

0.4

Nadir > 1.0–2.0

0.2

Nadir > 2.0

0.0 0

20

C

40

60

0

80 100 120 140 160

Months after radiotherapy

D

20

40

60

80 100 120 140 160

Months after radiotherapy

FIGURE 28-3.  Curves of freedom from biochemical treatment failure for patients with prostate cancer treated with definitive ­radiation therapy, without androgen ablation, according to the pretreatment prostate-specific antigen (PSA) level (A); Gleason score (B); clinical stage according to digital rectal examination (C); and nadir PSA level after treatment (D).

Fraction free of failure

1.0 0.9

Low risk (n = 281)

0.8 0.7

Intermediate risk (n = 470)

0.6 0.5

High risk (n = 462)

0.4 0.3 0.2 0.1 0.0 0

20

40

60

80

100 120 140 160

Months after radiotherapy

FIGURE 28-4.  Curves of freedom from biochemical treatment failure curves for 1213 patients with prostate cancer treated at M. D. Anderson with definitive radiation therapy and without androgen ablation according to risk level. Patients with favorable (low-risk) disease had a pretreatment prostate-specific antigen (PSA) level of no more than 10 ng/mL, a Gleason score of no more than 6, and a T1 to T2a tumor according to the 1992 American Joint Committee on Cancer staging system. Intermediate-risk factors were a pretreatment PSA level of more than 10 to 20 ng/mL, clinical stage T2b to T2c disease, or a Gleason score of no more than 7, with no high-risk factors. High-risk factors were a pretreatment PSA level of more than 20 ng/mL, a Gleason score of 8 to 10, or clinical stage T3 to T4 disease.

Pretreatment PSA level, Gleason score, and stage also are strong correlates of distant metastasis,101,102 and Gleason score and stage are predictive of survival. No ­association between pretreatment PSA level and survival has been ­documented.

Transurethral Resection for T3 to T4 Disease Patients with category T3 or T4 disease who have undergone TURP for symptoms of urinary obstruction have been identified as being at greater risk for developing distant ­metastasis.169-172 The independence of TURP as a determinant of outcome has been called into question by the ­results of multivariate analysis.173,174 Concern that use of the TURP procedure in such cases would cause dissemination of ­ tumor cells has diminished considerably, because urinary obstruction can be managed medically in most cases by ­using 5α-reductase inhibitors, antiandrogens, luteinizing hormone-releasing hormone agonists, or α-adrenergic blocking agents.

Cell Proliferation and DNA Content Characteristics The growth rate of prostate cancer is notably slow. Analyses of cell cycle kinetics have been limited but have produced intriguing findings. The labeling index for prostate tumors, obtained by injecting nucleoside analogue tracers into patients, is less than 6%,175-177 and the potential doubling time (Tpot) is more than 24 days. Staining prostate cancer specimens with monoclonal antibodies to the proliferation marker Ki-67 confirmed that only a small proportion of prostate cancer cells are proliferating at any given time.178,179 These findings suggest that most of the tumor cells are in a resting state. The time it takes for the serum PSA concentration to double in patients with prostate cancer is often more than 12 months,180,181 a period similar to estimates of tumor doubling time from measurements of cell cycle kinetics.182 These kinetic attributes have been

686

PART 8  Male Genital Tract

likened to those of late-reacting tissues and have stimulated interest in the radiobiologic characteristics of prostate cancer. The α/β ratio for prostate cancer seems to be between 1.5 and 5.0,183-186 suggesting that hypofractionated radiation may be desirable for the treatment of these tumors.187 DNA content analysis by flow cytometry or image analysis supports the findings from studies of cell cycle kinetics. Most malignant prostate tumors are diploid,188 with less than 5% of cells in S phase.189,190 DNA ploidy has been shown repeatedly to predict outcome after radical prostatectomy.102 The few studies of patients undergoing radiation therapy also have indicated that DNA ploidy has predictive value for outcome after this treatment. DNA diploidy is associated with a more favorable prognosis than is nondiploidy (including tetraploidy and aneuploidy). However, Khoo and colleagues178 have suggested that the proliferation marker Ki-67 may be a more significant correlate of biochemical failure than DNA ploidy. In that study, a low Ki-67 labeling index correlated with improved freedom from biochemical failure. The radiation therapy series that have investigated DNA ploidy and Ki-67 labeling have included few patients, however, and larger, prospective studies are needed.

Family History Family history has been reported by Kupelian and c­ olleagues191,192 to be a determinant of treatment ­success after radiation therapy. However, Hanus and ­coworkers193 were unable to corroborate these findings. Likewise, Bova and others194 observed that ­patients with sporadic and ­ familial prostate cancer who were treated with radical prostatectomy had the same prognosis.

Pretreatment Serum Testosterone Level Serum testosterone concentration in men with prostate cancer has been associated with outcome. Zagars and Pollack195 found that a serum testosterone level of more than 500 ng/dL was associated with distant metastasis in patients with clinical T1-T3 Nx/N0 prostate cancer treated with radiation therapy. They further found that similarly high serum testosterone concentrations were predictive of biochemical failure in patients with regionally lymph node–positive (stage D1) disease treated with AA.196 For patients with metastatic disease, however, a high serum testosterone concentration has been associated with a better response to AA and prolonged survival.197-199 The reason for these differences is not clear, but it seems that the serum testosterone level does affect outcome.

Molecular Markers The literature on biomarkers for prostate cancer has the same problem as the reports on cell cycle kinetics and proliferation measurements: small numbers of patients and promising but inconclusive findings. Some of the markers that have potential as pretreatment predictors of prostate cancer outcome are microvessel density and TP53, CDKN1A (formerly designated p21), CDKN1B (formerly p27), BCL2, BAX, CDKN2A (formerly p16), and RB1 (formerly retinoblastoma protein).102

External Beam Radiation Therapy: Treatment Planning and Techniques Evolution of Techniques Techniques for treating prostate cancer have evolved considerably over the past 10 years. The conventional fourfield approach has yielded to more complex conformal ­arrangements (e.g., three-dimensional CRT) and recently to IMRT. Simulations involving fluoroscopy have given way to simulations with CT. The considerations surrounding the application of new technology are complex, and adoption of new methods should be approached systematically. Three-dimensional CRT and the still more precise method of IMRT allow more accurate visualization of the prostate in relation to the surrounding normal structures. The rationale and support for the use of conformal techniques to treat prostate cancer are described in this section. Results from dose-response analyses are described under “Radiation Therapy Results.”

Three-Dimensional Treatment Planning and Conformal Radiation Therapy Three-dimensional CT-based reconstructions of the prostate, bladder, rectum, and pelvic bony anatomy are readily performed with commercial treatment planning systems. An example of a digitally reconstructed radiograph from a pelvic CT scan is shown in Figure 28-5.200 The fields superimposed on this reconstructed radiograph represent typical conventional anteroposterior and lateral fields that were used in the past. The prostate and seminal vesicles are shown on both views, with the bladder and rectum in the lateral display. Rather large margins surround the prostate in all dimensions except posteriorly, where a block shields the rectum. Although these fields are generous, they are not considered to be whole-pelvis fields, which would include 1.5 to 2 cm of the pelvic brim. Three-dimensional treatment planning illustrates the ­excessiveness of the margins used in conventional treatment planning, except for the posterior (rectal) ­margin. Figure 28-6 illustrates examples of oblique three-­dimensional CRT fields with 1.5-cm margins from the prostate to the block edge (a 1.0-cm planning target volume [PTV] margin) except posteriorly, where the margin is 1.0 cm (a 0.5-cm PTV margin). As detailed in Chapter 2, three-­dimensional treatment planning involves design of the gross tumor volume (GTV), the clinical target volume (CTV), and the PTV. For prostate treatments, the CTV is usually ­assumed to be the same as the GTV and is visualized on CT scans as the prostate itself. The PTV margin around the CTV accounts for uncertainties in setup and motion of the target. The PTV margin in this example is 1.0 cm in all dimensions except posteriorly, where it is 0.5 cm. Another 0.5 cm from the PTV to the block edge is added for ­penumbra. No clear-cut definitions for the CTV in terms of when and how much of the seminal vesicles to include have been determined; the amount depends on the risk of their ­being involved. In a single-institution analysis, Kestin and ­colleagues201 examined the extent of seminal vesicle involvement in 344 radical prostatectomy specimens from men with clinically localized (T1c to T2c) prostate cancer.

28  The Prostate

A

687

B

FIGURE 28-5.  Digitally reconstructed radiographs of conventional anteroposterior (A) and right lateral (B) fields (as part of the four-field technique) used at M.  D. Anderson. The prostate is shown in red, seminal vesicles in blue, rectum in green, and bladder in yellow.

Hetero

A

Hetero

Beam’s eye view for “RAO”

B

Beam’s eye view for “RPO”

FIGURE 28-6.  Digitally reconstructed right anterior oblique (RAO) (A) and right posterior oblique (RPO) (B) field radiographs used for conformal radiation therapy. The prostate is shown in red, seminal vesicles in purple, and planning target volume as a blue wireframe. These fields were taken from a seven-field plan with one anterior, two lateral, and four oblique fields. Oblique fields were at 35 degrees above and below the lateral positions.

They found that only 1% of patients with low-risk characteristics (i.e., PSA level < 10 ng/mL, Gleason score ≥ 6, and clinical stage ≤ T2a) had any seminal vesicle involvement, whereas up to 58% of patients with high-risk characteristics (i.e., PSA level ≥ 10 ng/mL, biopsy Gleason score ≥ 7, or clinical T category ≥ T2b) had positive seminal vesicles. The median length of seminal vesicle involvement was 1 cm (range, 0.2 to 3.8 cm), and less than 4% of all subgroups had more than 2 cm of vesicle (60% of the median vesicle length) involved. The investigators recommended including a portion of the seminal vesicles in the radiation field for some patients but stated that few men with clinically localized prostate cancer would require coverage of the entire length of the seminal vesicles.201 Common practice is that little or none of the seminal vesicles is included in the CTV for patients with favorable risk features. For patients at intermediate risk, some or all of the seminal vesicles are included for some or all of the treatment duration. For high-risk prostate cancer, most oncologists include the seminal vesicles in their entirety for the duration of treatment. However, compromises are often made in terms of ­seminal

vesicle coverage because of excessive overlap with the rectum. The seminal vesicles often fall around the rectum when simulations are done while the patients are supine. The definitions of the bladder and rectum volumes used for DVH analysis vary among treatment centers. At ­Memorial Sloan-Kettering Cancer Center, the outline consists of the walls of the bladder and rectum, rather than the walls plus the inner contents.202 Initially, at M. D. Anderson ­Cancer Center, the entire rectal contents were outlined, with the length of rectum extending from the ischial tuberosities 11 cm superiorly.15,200,203 However, an RTOG doseescalation trial (94-06)204 defined the length of rectum as starting at the ischial tuberosities extending superiorly to the anterior flexion of the rectosigmoid. These differences in the definition of structures for radiation planning could affect the variables used to determine the adequacy of a plan. Care should be taken to adopt the definitions of structure and DVH variables as specified by the group ­being emulated. Conformal field arrangements also vary widely. At Fox Chase Cancer Center, a four-field technique was used

688

PART 8  Male Genital Tract

throughout treatment, with a field reduction posteriorly from 1.5 to 0.5 cm or less for the last 10 Gy to reduce the risk of rectal bleeding.205 Others have used six or seven fields167,203,206, from the beginning of treatment or as a boost after an initial four-field setup. The six- to seven-field placements include four oblique fields at 30 to 40 degrees above and below the lateral positions, two lateral fields, and in some cases, an anterior field for quality assurance (the dose from the anterior field is usually less than that for the other fields). The lateral fields are classically weighted more heavily to spare more of the rectum. Assessment of three-dimensional CRT plans relies on examination of isodose lines on transverse CT scan images and DVHs. Visual inspection of the isodose lines on each CT scan slice, taken at 3- to 5-mm intervals, is used to verify that the PTV is covered adequately and that the rectum and bladder are spared from high doses. The DVHs are also essential in determining the suitability of the plan. The DVH variables for determining acceptability vary among treatment centers. In an analysis of patients treated at M.  D. Anderson, the proportion of rectum receiving at least 70 Gy strongly correlated with the 5-year actuarial rate of rectal complications of grade 2 or higher.15 In a detailed multivariate analysis, Skwarchuk and colleagues202 found that factors such as a higher maximum dose to the rectum, enclosure of the rectal contour by the 50% isodose line, and a smaller anatomically defined length of rectum were associated with a higher risk of late rectal bleeding. Kupelian and colleagues208 found in multivariate analyses that absolute rectal volume (15-mL cut point), rather than a percentage of rectal volume, correlated with late rectal complications. Figure 28-7A displays the isodose curves for a treatment planned with the conformal boost technique used in a randomized dose-­escalation trial at M.  D. Anderson or with IMRT. The IMRT plan has ­greater conformality, and the dose fall-off across the rectum is sharper. The dose specification should be to the PTV, ­because the minimum or average dose is more of a determinant of outcome than the isocenter dose. This recommendation is in accordance with the 1993 specifications of the International Commission on Radiation Units and Measurements.209

Intensity-Modulated Radiation Therapy The fluence of photon beams produced by linear accelerators is even across their length, and that fluence can be modulated to create more ideal dose distributions. Although simply placing a wedge in a beam modulates its intensity, the concept underlying IMRT is that the fluence is altered through the division of the radiation field into multiple beamlets of various intensity, the sum of which ­constitutes

a highly modulated beam. Manually cut transmission blocks or compensating filters have been used, but the procedures are laborious,210 and results can be duplicated or surpassed by using other conceptual IMRT approaches that are available commercially. Systems using tomotherapy or multileaf collimation (MLC) for IMRT are expanding the use of IMRT in community practice, much the way three-­dimensional CRT entered into use 10 years ago. Tomotherapy is the rotational delivery of radiation therapy by using several relatively small collimator vanes that open and close during treatment to modulate the intensity of the fan beam.211 The Peacock apparatus (Nomos Corp., Sewickley, PA) delivers the dose in a slice-by-slice fashion by means of multiple arcs, much as radiation is used in CT scanning to reconstruct three-dimensional ­images. ­Delivery of IMRT by MLC can be static116,212 or dynamic.213-215 Findings from M.  D. Anderson indicate that identical results in the treatment of prostate cancer can be achieved with tomotherapy (using the Peacock multileaf intensity-­modulating collimator [MIMiC]) and with segmental MLC delivery systems.216 Rectal DVH analysis of 20 patients with prostate ­cancer whose treatment was planned by using several methods is shown in Figure 28-7B. These methods (four ­ conformal and two intensity-modulated arrangements) include a conformal four-field technique with a field reduction after 46 Gy; a 10-field conformal boost, as was used in the M.  D. Anderson randomized dose-escalation trial200; a seven-field conformal arrangement throughout with 40% of the dose from the lateral fields; a seven-field conformal arrangement throughout with 50% of the dose from the lateral fields; an IMRT plan using the Peacock MIMiC system; and a 10-field sliding-window MLC plan. The conformal plans used 18-MV photons, whereas the intensity-­modulated plans used 6-MV photons. The dose was 75.6 Gy in 42 fractions prescribed to the PTV, except in the 10-field plan, in which 78 Gy in 39 fractions was prescribed to the isocenter. The proportion of the rectum that received 70 Gy or more was significantly lower in the IMRT plans. No difference was observed between the Peacock MIMiC and the MLC plans. The benefit from IMRT was apparent despite the choice of patients with favorable-risk disease and very little outlining of the seminal vesicles. The advantages of IMRT over three-dimensional CRT are seen in two principal areas. First, IMRT allows ­greater conformality of the radiation beam than does three­dimensional CRT. Second, the dose fall-off from the ­prescription isodose line is much greater with IMRT than with three-­dimensional CRT. Another benefit for the treating physician is that after the planning parameters are established, the planning process is much faster than that of

FIGURE 28-7.  A, Dosimetry for a patient with favorable-risk prostate cancer generated by six contemporary conformal-based planning methods. Top row: left, four-field plan with lateral field reduction posteriorly at 46 Gy; right, 10-field conformal boost plan (as used in the M.  D. Anderson dose-escalation trial). Middle row: left, seven-field conformal plan with 40% weighting of lateral fields; right, plan using the Peacock multileaf intensity-modulating collimator (MIMiC) to deliver intensity-modulated radiation therapy. Bottom: left, a seven-field conformal plan with 50% weighting of lateral fields; right, plan using multileaf collimation (MLC) for ­  intensity-modulated radiation therapy. The prescribed dose was 75.6 Gy to the planning target volume, except in the 10-field conformal boost plan, in which the dose was prescribed to the isocenter (as done in the M.  D. Anderson protocol). B, Rectal dosevolume histogram analyses from the six planning methods shown in A. Bars indicate the standard error; asterisks indicate that the percentage of rectum treated to more than 70 Gy was significantly less when intensity-modulated radiation therapy methods were used. DMLC, dynamic multileaf collimator.

28  The Prostate

4-field

10-field

7-field-40%

MIMiC

7-field-50%

MLC

A

79.6 75.6

Percent of volume (%)

30 Isodose lines (in Gy)

70.0

20 * * 10

% rectum >70 Gy % bladder >70 Gy % femur >50 Gy

45.0 30.0 20.0

 

0

60.0

Figure 28-7.  See legend on opposite page.

689

 B

4-field 7F-40% 7F-50% 10-field MIMiC DMLC

PART 8  Male Genital Tract

three-dimensional CRT. There is no need to draw blocks when IMRT is used, and inverse planning methods facilitate planning optimization. As described in Chapter 2, inverse planning is a computer planning process wherein the structures of interest and structure dose limits are specified and the computer determines the optimal beam weighting, segmentation, and segment intensity to achieve the stated goals. In tomotherapy, this process is fully automated. In MLC-based IMRT, the number and angles of the beams must be empirically determined first. In the future, the selection of optimal beam angles will be incorporated into commercial MLC inverse-planning software.

Setup Error and Prostate Movement Considerations By definition, the PTV is the margin placed around a target organ or site to account for setup errors and changes in organ position relative to the reference planning CT scan. In ­treating prostate cancer, the extent to which setup error can be minimized by immobilization and the best method for immobilization are still somewhat controversial.217-220 At Memorial Sloan-Kettering, patients are treated while prone, with a thermoplastic body cast used for stabilization.217 The advantage of this approach is that the ­prostate and ­seminal vesicles fall anteriorly away from the rectum, thereby ­ reducing rectal doses. However, this treatment has not been consistently reproducible elsewhere. Later studies demonstrated that prostate movement with respiration was more pronounced when thermoplastic ­ abdominopelvic shells were used221,222; less respiratory movement is ­observed when patients are treated while supine, without abdominopelvic casts. At M.  D. Anderson, a standard combination knee-and-foot cradle indexed to the treatment table is used for all patients with prostate cancer. Controlling the position of the knees and ankles also ­allows reproducible positioning of the hips while daily ­images of the prostate are obtained by sonography or by fiducialbased onboard kilovoltage or CT imaging. Prostate position in the pelvis is affected by rectal and bladder filling; rectal filling seems to have the greater influence. Serial CT scanning during radiation therapy has documented that the changes in prostate position are not inconsequential.223-232 Positional changes are greatest in the anteroposterior and superoinferior directions, probably along the angle of the pubic symphysis. Antolak and colleagues231 calculated that to include all of the CTV in the PTV 95% of the treatment time, the PTV would have to be 0.7 cm in the lateral and superoinferior directions and 1.01 cm in the anteroposterior direction. Adding 0.5 cm for penumbra would make the posterior margin at the rectum larger than 1.5 cm, which would mean that in many patients the entire circumference of the rectum would be treated to a full dose. Although larger margins may be necessary for some patients, attempting dose escalation to more than 70 Gy using these recommendations would produce unacceptable rectal complications. Measures to reduce the error associated with changes in prostate position are needed for three-dimensional CRT and IMRT. Antolak and colleagues231 found that rectal and bladder volumes were greater on the pretreatment planning CT scans than on subsequent scans obtained during treatment (Fig. 28-8).231 One way of minimizing the ­effects of this difference is to prepare another plan after 2 to 3 weeks of treatment; however, this practice is ­ laborious and still

350 300 Bladder

250 Volume (cm3)

690

200 150

Rectum

100 50 0 Pretreatment

On treatment 1

On treatment 2

On treatment 3

FIGURE 28-8.  Bladder and rectal volume changes as determined from serial CT scans during radiation therapy. Scans were obtained once before treatment and three times during treatment, with 2- to 3-week intervals between scans. (Modified from Antolak JA, Rosen II, Childress CH, et al. Prostate target volume variations during a course of radiotherapy. Int J Radiat Oncol Biol Phys 1998;42:661-672.)

does not account for day-to-day deviations. ­ Methods are available for monitoring prostate position at each treatment session232; these methods are becoming an integral part of ensuring that the prostate is centered in the PTV for three-dimensional CRT and IMRT. The Nomos Corporation has developed the B-mode ­Acquisition and Targeting (BAT) system, an ultrasound ­device that localizes the prostate in three-dimensional space, ­ compares prostate position to that on the pretreatment CT scan, and calculates the shift necessary to place the prostate in the middle of the PTV (Fig. 28-9). ­Lattanzi and colleagues233 reported that shifts calculated in this way match those ­derived from daily CT scanning. Several other ­ ultrasound-based ­ localization systems have become ­ commercially available. Use of these systems requires careful contouring of the prostate-bladder interface for ­ reference purposes and a patient who can cooperate with having at least a semifull bladder to help visualize the prostate. ­Regardless of the system used, a PTV of ­approximately 5 mm is still recommended based on correlative CT ­studies.234,235 Prostate position can be documented by placing metallic fiducial markers in the prostate before radiation therapy is begun and using those markers to facilitate daily localization by electronic portal imaging.223 Typically, three or four markers are inserted at the base, in the middle of the gland, and at the apex of the prostate to allow sufficient image registration. Inserting the markers at least 5 to 7 days before the simulation minimizes the possibility of their subsequent migration. Amorphous silicone detectors can be used for recording marker position daily and calculating shifts to position the prostate properly within the PTV, but if megavoltage imaging is to be used, the additional radiation dose should be taken into account. Newer treatment units with dedicated on-board kilovoltage imaging capability have improved image quality, allowed use of ­ smaller fiducial markers, and obviated the need for ­ additional ­radiation doses.

28  The Prostate

691

Before shift

A

After shift

B FIGURE 28-9.  Sonograms obtained with the Nomos Corporation’s BAT system show the outlines of the bladder, rectum, and prostate from planning CT scans superimposed over sonogram taken before alignment (A) and after alignment (B).

Localization of the Prostate Apex Use of CT scanning to localize the apex of the prostate is inexact because the apex itself is not well visualized; ­rather, the bulb of the penis, below the urogenital ­diaphragm, is the best landmark. Retrograde urethrography has been ­advocated as a means for more precisely determining the position of the prostate apex.236,237 In this technique, about 10 mL of contrast medium is injected into the urethra, which is then blocked with a Cunningham clamp or a partially inflated Foley (balloon) catheter. The distance from the superior tip of the contrast to the prostate apex is about 1.3 cm. However, the manipulations associated with this technique can introduce error; Malone and colleagues238 found that instilling contrast under pressure displaced the prostate apex superiorly by 5 to 7 mm. Sandler and colleagues239 also reported that when radiodense markers were implanted at the prostate apex under sonographic guidance, CT localization of the prostate apex was more accurate than was retrograde urethrography. The safest approach is to use the bulb of the penis on CT scans as a guide for apex position or to use MRI, which visualizes the apex more clearly.240 The prostate apex is on average 3 mm (range, 0 to 6 mm) superior to the penile bulb.241 The sagittal images on the planning CT scans may also be useful for defining the prostate apex. Accurate definition of the prostate apex is critical for radiation therapy planning. Overtreating the apex may lead to higher doses being delivered to the penile bulb, which has been shown in some series to be associated with higher rates of erectile dysfunction after treatment.242 However, this correlation has not been uniformly confirmed in prospective studies.243,244 Underdefining the prostate apex may result in marginal misses, and pathologic analysis of salvage radical prostatectomy specimens has shown cancer foci in the apex in more than 90% of cases.245

Brachytherapy: Treatment Planning and Techniques Evolution of Techniques Implantation of permanent radioactive seeds in the prostate (PRSIP) has a long history. The largest series of patients treated in this way with a retropubic approach came from Memorial Sloan-Kettering.246 Investigators developed a nomogram for the use of iodine 125 (125I), which has a half-life of 60 days, as the primary treatment for prostate cancer. The other isotope commonly used in the past has been gold 198 (198Au), which has a half-life of 2.7 days and was used in combination with EBRT.247,248 In these early procedures, an open retropubic approach was used that involved the needles being placed retropubically into the prostate while the prostate was palpated with a finger in the rectum (Fig. 28-10).247,249 The resulting distribution of seeds in the prostate was uneven, making the dosimetry was substandard by current criteria. ­ Local control rates in these early series were very low.250,251 Scardino and colleagues252 documented a very high rate of positive biopsy results after therapy with gold seeds and EBRT, but current sonographically guided transperineal approaches have improved the success of PRSIP considerably (Fig. ­28-11).253,254 Techniques for PRSIP continue to evolve with the use of intraoperative treatment planning.

Planning Methods Two principal methods for brachytherapy treatment planning are in use today. The most common approach ­involves pretreatment planning with TRUS,255,256 ­although ­intraoperative planning is gaining in popularity.257,258 ­Combinations of preplanning and intraoperative ­planning are also in use. Preplanning involves acquiring transrectal, 0.5-cm step-section sonographic images of the prostate

692

PART 8  Male Genital Tract

within a month of beginning the treatment. The step-section ­images are loaded onto a treatment planning computer, and a plan is formulated through an interactive process with the dosimetrist and the attending physician. The ­availability of inverse planning programs has greatly ­simplified this procedure. The finalized plan designates ­ needle position and the location of each seed in each ­needle. The seeds are delivered to the predetermined locations in the prostate by one of two methods. The first method

FIGURE 28-10.  Needle placement for prostate seed implant using the open retropubic approach. (Modified from Chan RC, Gutierrez AE. Carcinoma of the prostate: its treatment by a combination of radioactive gold-grain implant and external ­irradiation. Cancer 1976;37:2749.)

involves the use of preloaded needles; the seeds are ­loaded into the needles in locations specified by the plan. The second method involves the Mick applicator system (see Fig. 28-11), which allows seeds to be deposited one at a time. The Mick applicator procedure takes longer but offers more control in terms of adjusting seed spacing, which is monitored during the procedure with fluoroscopy. A drawback of the preplanning approach is that the patient must be placed in precisely the same position as for the preplanning TRUS procedure. Sometimes exact positioning is not possible. Moreover, the position, shape, and size of the prostate when the seeds are placed may differ slightly from their appearance on preplanning images, which can reduce the possibility of achieving acceptable dosimetry. The use of preplanning and preloaded needles requires considerably less operating time than the alternative of real-time intraoperative planning. Use of the Mick applicator system also may involve a steeper learning curve than use of preloaded needles. However, intraoperative computer planning provides valuable feedback on dosimetry during the implantation procedure. Seed placement is typically monitored with fluoroscopy, and the seed positions are observed on the planning computer display. The profile developed is an approximation of the position of the seeds in three-dimensional space. The use of MRI259 or CT260 for imaging during implantation allows more precise real-time three-dimensional reconstructions of dose to be generated; however, these procedures take considerably longer than does sonography. From a clinical standpoint, if more than about 90% of the target volume receives 100% of the dose, it may be difficult to show a difference between three-dimensional intraoperative reconstructions from MRI or CT compared with those from sonography. The same argument can be applied to intraoperative planning; use of intraoperative

FIGURE 28-11.  Transperineal implantation of permanent radioactive seeds in the prostate. The needles are arranged in a peripheral pattern (left), and the Mick applicator is used to direct implantation (right).

28  The Prostate

computer planning techniques requires more time in the operating room, which is costly. If the differences between preplanning reconstructions and intraoperative planning reconstructions are insignificant, preplanning is probably preferable. Initial evaluations have favored intraoperative planning in terms of improved dosimetric variables such as the volume of the target that receives 100% of the prescribed dose (V100) and the dose received by 90% of the target volume (D90). Intraoperative planning also may be beneficial in terms of the homogeneity of the dose distribution. On balance, the findings support the adoption of intraoperative planning, and there is every indication that this technique will become more standard over time.

Evaluation of the Pretreatment Plan Development of an acceptable plan, whether generated by preplanning or intraoperative planning methods, requires consideration of possible errors from seed migration and postoperative edema. The goal is to encompass the prostate with the prescription isodose line as uniformly as possible. A 3- to 5-mm margin is recommended, with the tighter margin posteriorly at the rectum. Davis and colleagues261 and Sohayda and coworkers262 demonstrated that such margins would encompass extracapsular extension in more than 95% of cases of favorable disease treated with ­brachytherapy.

Evaluation after Treatment It is customary to obtain a CT scan within 24 hours after the implant and another approximately 1 month later. The postplan evaluation is critical to quality assurance monitoring (Fig. 28-12). A CT scan performed immediately after the implant represents the worst case scenario because edema is pronounced, causing increased spacing of the seeds relative to one another.263 Much of the swelling is gone by 1 month after implantation, and a CT scan performed 1 to 2 months after implantation, although inconvenient for some patients, provides a better estimate of dosimetry.263,264 In either case, postimplantation DVH analyses of prostate V100 and D90 and other similar measures have been shown to correlate with clinical outcome. Figure 28-13

> 200.0 Gy 200.0 Gy 180.0 Gy 160.0 Gy 140.0 Gy 120.0 Gy 100.0 Gy 80.0 Gy 60.0 Gy 40.0 Gy 20.0 Gy 0.0 Gy < 0.0 Gy

FIGURE 28-12.  Color wash dose display after implantation of needles in the prostate. Notice that the posterior apex of the prostate receives less than the prescribed dose.

693

displays four images from a postimplantation CT plan and the DVHs for the prostate, bladder, and rectum. Stock and colleagues265 found that when the D90 approximated the prescribed dose of radiation from 125I, the freedom from biochemical failure rate was significantly better than in cases in which the D90 was below this level. The ­American Brachytherapy Society266 recommends that the D90 be ­reported along with the D80, D100, V80, V90, V100, V150, and V200. For definitive prostate cancer treatment, the prescription dose from 125I is 144 Gy and that from palladium 103 (103Pd) is 125 Gy. The prescription dose for prostate boost treatment when PRSIP is used in conjunction with EBRT is 110 Gy for 125I and 90 Gy for 103Pd. Before publication of the American Association of Physicists in Medicine Task Group’s report 43,267 the standard reference dose for 125I prostate implants was 160 Gy. The recommendations of Task Group 43 were based on a new National Institute of Standards and Technology (NIST) air-kerma standard for 125I seeds, which was adjusted by about 10%.268 The result is that the prescription dose for 125I implants was reduced to 144 Gy.269 A similar adjustment has been made for 103Pd.270,271 Before the NIST 1999 standard, the prescribed dose for 103Pd was 115 Gy; that dose should be changed to 125 Gy to achieve the same dosimetry.

Permanent Implantation of Iodine 125 or Palladium 103 Radioactive Seeds plus External Beam Radiation Therapy Although no standard method has been established for the use of EBRT in combination with PRSIP, most investigators have given the EBRT first, followed approximately 2 weeks later with PRSIP. The rationale for substituting 103Pd for 125I is based on the hypothesis that poorly differentiated prostate tumors have a more rapid proliferation rate and consequently a shortened Tpot; 103Pd, with its shorter halflife, would be more effective.272,273 However, no clinical data exist to suggest that 103Pd is any better than 125I for the treatment of poorly differentiated or other high-risk forms of prostate cancer. Critz and colleagues274 have advocated administering 125I before beginning a protracted course of EBRT given at 1.5 Gy/day during the active decay of the isotope. Others have advocated using 103Pd and waiting about 2 months after the implant before giving EBRT. The typical sequence, however, is for EBRT to be given first, ­followed about 2 weeks later by PRSIP.256,275 The usual external beam radiation dose is 45 to 50 Gy given over 5 to 5.5 weeks.

Temporary Iridium 192 Implants Temporary implants with high-dose-rate iridium 192 (192Ir) are usually combined with EBRT for the treatment of patients with intermediate- to high-risk prostate cancer. The procedure is similar to that for permanent implants in that temporary afterloading needles are placed transperineally into the prostate under sonographic guidance.276,277 A pretreatment plan may be used, but real-time planning in the operating room has been used to optimize needle placement.278 High-dose-rate 192Ir implants are more forgiving than are 125I or 103Pd permanent implants; CT

694

PART 8  Male Genital Tract Image # 11 Position: 8.00cm

Image # 12 Position: 7.50cm

Image # 13 Position: 7.00cm

Image # 14 Position: 6.50cm

CUMULATIVE DVH 100

% Structure volume

90 80 70 60 50 40 30 20 10 0 0

50 100 150 200 250 300 Dose (Gy) Prostate Bladder Urethra Rectum sv

FIGURE 28-13.  Transverse CT images of the pelvis with implant dosimetry superimposed. Contours show the bladder (yellow), prostate (red), and rectum (blue); the 144-Gy isodose line (yellow) around the prostate can be seen in the top two images. The dosevolume histogram (DVH) analyses are shown (left) for the prostate, bladder, urethra, and rectum.

Radiation Therapy Results Radiation Therapy Alone for Favorable Disease Treatment Options Treatment outcome for patients with pretreatment PSA levels of 10 ng/mL or less, a Gleason score of less than 7, and category T1 to T2a disease (according to the 1992 staging system) is very favorable. Outcomes after radical prostatectomy, brachytherapy alone, EBRT alone, and the combination of EBRT plus brachytherapy have been comparable. Direct comparisons among single-institution series are complicated by imbalances in prognostic factors and length of follow-up; however, little difference among institutions seems to be found for patients with favorable risk factors. Poorer results are evident when EBRT doses are low.281-283 In a study of 1213 patients treated at M.  D. Anderson with only EBRT through 1998 (i.e., during the PSA era), the Kaplan-Meier freedom from biochemical failure rates for the 281 patients with favorable risk ­factors

1.0

Fraction free of failure

planning for the high-dose-rate treatment is done before the sources are placed, and the time at which the source dwells at particular points can be adjusted to optimize the dose distribution. The dose of each high-dose-rate fraction, the intervals between those fractions, and the sequencing with EBRT have varied greatly among investigators. Dosing schemes have ranged from 3.0 to 4.0 Gy in 4 fractions over 40 hours276 to 9.5 Gy in 2 fractions over 2 weeks,278 and the high-dose-rate 192Ir seeds have been implanted before, during, and after the EBRT.276,278-280

Low risk (n = 199)

0.8

Intermediate risk (n = 300)

0.6 High risk (n = 214)

0.4 0.2 0.0 0

20

40

60

80

100

Months after radiotherapy

FIGURE 28-14.  Freedom from biochemical failure rates for 713 patients with prostate cancer treated at the M.  D. Anderson Cancer Center with external beam radiation to a dose of more than 67 Gy.

were 88% at 5 years and 84% at 10 years (see Fig. 28-4). The freedom from failure rates for the 713 patients in this study who were treated to more than 67 Gy are shown in Figure 28-14; among the 199 favorable-risk ­patients in that subgroup, the freedom from failure rate at 5 years was 92%, with no treatment failures observed after 5 years. These results are similar to those of other series evaluating EBRT (Table 28-3),168,282,284-286 radical prostatectomy (Table 28-4),103,105,152,287,288 or brachytherapy (Table 28-5)255,256,285,289-291 for patients with prostate cancer.

28  The Prostate

695

TABLE 28-3.  External Beam Radiation Series According to Pretreatment Prostate-Specific Antigen Level Study and Reference Variable

M. D. Anderson Cancer Center, 2000

Hanks et al, 2000286

Brachman et al, 2000285

Shipley et al, 1999166

Number of patients

713*

377†

1527

1765‡

Median follow-up

46 mo

53 mo

41 mo

> 60 mo

Time of analysis

5 yr

5 yr

5 yr

5 yr

71%

—

—

66%

88%

—

90%

81%

Patients with PSA > 4-10

80%

—

74%

68%

Patients with PSA > 10-20

63%

≈ 84%

70%

51%

Patients with PSA > 20

32%

≈ 40%

51%

31%

Patients with PSA ≤ 10

81%

≈ 90%

—

78%

Patients with PSA > 10

56%

≈ 52%

—

—

Freedom from bio­ chemical failure rates All patients Patients with PSA ≤

4§

*Patients

treated in the PSA era (i.e., after 1988) were given doses of more than 67 Gy. group. ‡Pooled multi-institutional analysis. §PSA levels are expressed in ng/mL. PSA, prostate-specific antigen. †Dose-escalated

TABLE 28-4.  Surgical Series According to Pretreatment Prostate-Specific Antigen Level Study and Reference Kupelian et al, 1996103

Dillioglugil et al, 1997152

Catalona et al, 1998288

1354

337

536

1615

Median follow-up

60 mo

36 mo

24 mo

—*

Time of analysis

5 yr

5 yr

5 yr

5 yr

80%

61%

78%

—

94%

89%

93%

≈ 95%

Patients with PSA > 4-10

82%

63%

76%

≈ 85%

Patients with PSA > 10-20

72%

57%

69%

—

Patients with PSA > 20

54%

27%

46%

—

Patients with PSA ≤ 10

—

71%

—

—

Patients with PSA > 10

—

42%

—

≈ 58%

Variable

Pound et al,

Number of patients

1997287

Freedom from bio­ chemical failure rates All patients Patients with PSA ≤

4†

*Follow-up

for patients with postoperative PSA determinations was not stated. levels are expressed in ng/mL. PSA, prostate-specific antigen.

†PSA

Subsequent post hoc analyses of patients with low-risk disease treated in two different randomized dose-­escalation trials suggested a benefit in biochemical control from doses higher than 70 Gy. Updated results from the M. D. Anderson randomized trial200 showed that the 8-year freedom from failure rate for patients with low-risk disease treated to 78 Gy was 88%, compared with 63% for those treated

to 70 Gy (P = .042).292 Another randomized trial conducted at the Loma Linda University Medical Center and Massachusetts General Hospital compared doses of 79.2 and 70.2 gray equivalents (GyE), delivered with combination proton and photon therapy. That study also found a benefit in terms of biochemical control from giving men with low-risk disease the higher radiation doses; the 5-year

696

PART 8  Male Genital Tract

TABLE 28-5.  125I Implant Series According to Pretreatment Prostate-Specific Antigen Level Study and Reference Variable

Blasko et al, 1995255

Wallner et al, 1996289

Grado et al, 1998290

Brachman et al, 2000285

Number of patients

197

92

490

695

Median follow-up

36 mo

36 mo

27 mo

51 mo

Time of analysis

4 yr

4 yr

5 yr

5 yr

Freedom from biochemical failure rates All patients

93%

63%

79%

—

Patients with PSA ≤ 4*

98%

100%

—

87%

Patients with PSA >4-10

90%

80%

—

76%

Patients with PSA >10-20

89%

45%

72%

53%

Patients with PSA > 20

80%

38%

57%

49%

Patients with PSA ≤ 10

—-

—

88%

—

Patients with PSA > 10

—

57%

—

—

*PSA

levels are expressed in ng/mL PSA, prostate-specific antigen.

freedom from biochemical failure rate was 97.3% for the higher-dose group and 82.6% for the lower-dose group (P < .001).293 Direct comparisons of results from the use of EBRT alone, brachytherapy alone, and radical prostatectomy have been attempted. Zelefsky and colleagues294 compared patients undergoing EBRT only and those treated with 125I PRSIP at Memorial Sloan-Kettering and found essentially no difference in freedom from biochemical failure rates (88% and 82%, P = .09). They did report that protracted urinary symptoms of grade 2 or higher were more prevalent in the implantation group. In a study of 2222 men with prostate cancer, Brachman and coworkers285 found that EBRT produced results similar to those of PRSIP among patients with favorable risk factors. D’Amico and colleagues165 compared outcomes for patients from selected institutions who were given EBRT, PRSIP, or radical prostatectomy. At a median follow-up time of 38 months, no statistically significant difference in freedom from failure was observed for the patients with favorable-risk disease. Stokes and colleagues295 reached a similar conclusion. However, when Polascik and coworkers296 and Ramos and colleagues297 tried to match their radical prostatectomy patients to the 125I-PRSIP patients described by Ragde and colleagues,298 the two groups came to different conclusions. Polascik and colleagues296 found that the 7-year freedom from biochemical failure rates were 97.8% for the surgery group and 79% for the brachytherapy group in the study by Ragde and coworkers,298 suggesting that surgery was the treatment of choice. Ramos and colleagues297 quoted a 7-year freedom from biochemical failure rate of 84% and concluded that surgery and brachytherapy were equally effective. These different conclusions underscore the problems in trying to match patient prognostic factors between institutions. Keyser and coworkers284 and Kupelian and colleagues299 found no difference in outcome between EBRT and radical prostatectomy for men with favorable-risk disease,

despite their later finding of a significant dose-response effect among those patients.283

Brachytherapy Alone The use of brachytherapy for the treatment of prostate cancer has shown a resurgence over the past 20 years, in large part because of the introduction of TRUS for treatment planning and needle placement by a group at the Northwest Prostate Institute in Seattle.255,256 Several variants of brachytherapy are in use, but most involve the methods previously described for permanent radioactive seed-­implant monotherapy. 125I and 103Pd seeds have been used interchangeably, and no clear-cut evidence has been found that either isotope is more effective than the other. The half-life of 103Pd is 17 days, which is shorter than the 60-day half-life of 125I. Some believe that the shorter half-life of 103Pd is preferable for treating high-grade tumors, which may have a shorter doubling time.272,273 This hypothesis, although seemingly sound, has not been borne out in clinical studies. The Tpot of prostate cancer may be long enough to favor 125I, even in poorly differentiated tumors.175-177 The effectiveness of PRSIP monotherapy has been established in several studies255,256,285,289-291 (see Table 28-5). This approach has been shown repeatedly to be most effective for men with a pretreatment serum PSA concentration of no more than 10 ng/mL, a Gleason score of no more than 6, and stage T1 to T2a disease according to the 1992 AJCC staging system. The 1992 AJCC staging system is used here because of the suggested difference between palpable T2a, T2b, and T2c tumors (see Fig. 28-3C).83,103,106 D’Amico and colleagues165 observed in a retrospective review that the freedom from biochemical failure rates for patients with intermediate- or high-risk prognostic characteristics were significantly lower when they were treated with PRSIP than when similar patients were treated with EBRT or radical prostatectomy. No difference between these modalities was observed for patients with favorable prognostic characteristics.

28  The Prostate

The best results with PRSIP monotherapy have resulted from strict adherence to selection criteria. Most investigators consider such criteria to be the proportion of biopsy specimens that are positive for tumor, the amount of tumor in the specimens, and whether tumor is present bilaterally. These rather subjective criteria are used routinely in the clinic, but they have not been rigorously tested. For example, a patient with otherwise favorable risk factors and a small 1- to 2-mm focus of Gleason grade 7 disease may be considered for PRSIP over someone with three of six biopsy cores containing Gleason 6 disease that involves more than one half of each positive core. No distinct guidelines have been developed for the use of biopsy tissue findings such as these, and further work is needed in this area. The rationale for selecting a treatment strategy on the basis of biopsy findings is that the incidence of extracapsular extension increases as the number of positive biopsy specimens and the extent of tumor in the specimens ­increase.120,300,301 The accuracy of the biopsy findings mentioned here depends on the adequacy of the sampling. Sextant biopsies may not be enough to appreciate the extent of disease, particularly if a small focus of Gleason 7 disease is present. Repeat biopsy often reveals a larger tumor burden in these cases. Other considerations in the selection of patients for PRSIP are baseline urinary function status, whether the patient has previously undergone TURP, and baseline prostate volume. PRSIP causes significant urinary obstructive symptoms that seem to be greater than those resulting from EBRT. Difficulty emptying the bladder may be experienced for prolonged periods after PRSIP. Urinary side effects tend to resolve 6 to 12 months after the implantation, but improvements may occur even beyond a year. α-Adrenergic blockers such as doxazosin (Cardura), terazosin (Hytrin), or tamsulosin (Flomax) are used liberally during this period to reduce the frequency of nocturia and other symptoms of urinary obstruction. The American Urological Association’s symptom index questionnaire or a similar scoring system is useful for identifying patients who may be predisposed to urinary obstruction after PRSIP.302,303 However, the decision cut points on the scale are arbitrary. A high score (≥15) implies additional risk of prolonged urinary obstructive symptoms and the possible need for catheterization or TURP after implantation. Intermediate scores above 10 probably also reflect a higher risk, and clinical judgment based on experience is helpful in such cases. In our experience, the use of TURP after seed implantation should be avoided because of the risk of incontinence and poor healing of the urethral mucosa. Patients who develop prolonged urinary obstructive symptoms after seed implantation may be better served by more conservative measures such as α-adrenergic blockers and intermittent self-catheterization as needed. (Consideration should be given to whether the patient is already taking an α-blocker.) Because urinary symptoms may take more than a year to resolve, conservative measures such as these may spare patients a potentially lifechanging procedure. A history of TURP before PRSIP has been associated in the past with an increased risk of urinary incontinence after the implantation,255,289 but this might have been related to the use of older, uniform seed distribution methods. Later

697

studies indicate that peripheral loading techniques can be helpful in avoiding urinary complications in patients who have undergone TURP.303,304 Another concern is prostate volume and the possibility of interference from the pubic arch. The prostate sits behind the pubic symphysis, and if the gland is large, its lateral aspect may be blocked for ­needle placement. Placing the patient in an extended lithotomy position minimizes the possibility of pubic arch interference. Although the frequency of pubic arch interference increases as prostate volume exceeds 50 cm3, TRUS and pelvic CT scanning have been helpful in estimating the risk of this condition before the implant is placed,305,306 and some patients with large prostates have undergone successful implantation with excellent dosimetry and low morbidity.307,308 Patients with large prostates, however, may be predisposed to urinary symptoms after implantation.309 The isotope used also can influence the incidence of urinary side effects; urinary morbidity after 103Pd implantation may not last as long as morbidity after 125I implantation.310 Late rectal complications after PRSIP may be fewer and less severe than those seen after EBRT.311 Impotence rates after PRSIP312-314 do not seem much different from those after EBRT (about 50% overall).16,17,315 The overall rates of side effects associated with PRSIP after urinary symptoms resolve have been low316 and are comparable to those associated with EBRT.

Radiation Therapy for Intermediate- to High-Risk Disease Brachytherapy in Combination with External Beam Radiation Therapy The options available for the treatment of intermediate- to high-risk prostate cancer include an expanding list of techniques that combine EBRT and brachytherapy. These options include PRSIP with 125I or 103Pd given before or after EBRT,159,256,275,317 PRSIP with 198Au plus EBRT (with the implant usually given first248), and temporary high-doserate 192Ir given before, during, or after EBRT.276,278-280 Each method has technical advantages and disadvantages, but comparisons of them are not available. However, results from small, retrospective series have been promising. Ragde and colleagues256 have shown that after long followup, the results after PRSIP plus EBRT were slightly, but not significantly, better than results with PRSIP alone, even though the patients treated with PRSIP alone had more favorable risk features. Findings from a later retrospective analysis318 suggested that EBRT followed by a brachytherapy boost could achieve reasonable biochemical control rates (80.3%) among patients with intermediate-risk disease. These results suggest that combination therapy is another reasonable method for escalating the total radiation dose. As has been found for EBRT monotherapy167,200,283,319,320 and PRSIP monotherapy,265 a dose-response relationship seems to exist for the combination therapy. In one study of combination therapy, Martinez and colleagues278 gradually escalated the biologically equivalent dose from highdose-rate 192Ir implants given during EBRT to 46 Gy. Their ­preliminary results suggest that the higher biologically equivalent doses were associated with improved outcome, but the follow-up period in this study was short.

698

PART 8  Male Genital Tract

External Beam Radiation Therapy Alone: Dose-Response Evaluations Figure 28-4 shows a Kaplan-Meier analysis of freedom from biochemical failure for patients with favorable-, ­intermediate-, or high-risk disease treated with EBRT alone at M.  D. Anderson between 1987 and 1998. The rates at 5 years and 8 years were 70% and 67%, respectively, for the intermediate-risk group and 39% and 34% for the high-risk group. The evidence in support of three-dimensional CRT and IMRT for the treatment of prostate cancer has been decisive. Improvements in treatment planning and delivery techniques and in image guidance have allowed the ­margins around the prostate to be reduced, and as a consequence, the volume of normal tissue that is treated to higher doses and the side effects have been reduced without sacrificing tumor control. These treatment techniques have allowed the dose to be escalated significantly. Patients with intermediate- to high-risk disease features have benefited greatly from these approaches. Retrospective dose-response analyses in pre-PSA era studies have been promising. Hanks and colleagues321,322 and Perez and coworkers323,324 reported that local control rates were higher for patients with locally advanced disease when the dose was increased. Conventional radiation techniques were used, although more severe side effects were observed when the doses exceeded 70 Gy. In a single-­institution trial at Massachusetts General Hospital, a proton boost was used to escalate the dose from 67.2 Gy to 75.6 Gy.325 In that trial, the higher dose afforded by the boost did not have a statistically significant benefit in terms of disease freedom or survival, although the local control rate at 8 years was improved from 19% to 84% (P = .0014). Among the subset of patients with a Gleason score of 8 to 10, the rates for freedom from disease were better for those given the higher dose. The patients in this study had locally advanced disease and are not representative of patients seen in the PSA era. Nevertheless, several ­retrospective studies in the PSA era strongly support the

conviction that radiation dose is an independent correlate of patient outcome.167,283,286,319,320 A positive effect of dose escalation above 70 Gy on freedom from biochemical failure rates has consistently been described for patients with intermediate-risk features such as a PSA level of 10 to 20 ng/mL. Dramatic gains in freedom from biochemical failure can be achieved with a modest elevation in dose (Table 28-6).167,283,286 The caveat is that the dose in these studies was raised sequentially over time, and during that time, stage migration was also evident. Randomized comparisons are needed to substantiate observations made in the retrospective analyses. A randomized dose-escalation trial conducted in the PSA era at M.  D. Anderson was originally reported in 2000200 and then updated in 2008292 with the definition of PSA failure set at the nadir level plus 2 ng/mL. Patients were randomly assigned to receive 70 Gy delivered with conventional fields or 78 Gy with a conformal boost technique. Stratification of the 301 evaluable patients was based on pretreatment PSA values. Dose was specified to the isocenter, as was the standard practice at M.   D. Anderson when the trial was begun in 1993. As discussed in “Three-­Dimensional Treatment Planning and Conformal Radiation Therapy,” the current recommendation is to prescribe to the PTV, which is consistent with the criteria of the International Commission on Radiation Units and Measurements.209 The freedom from biochemical failure rates for the two treatment groups are illustrated in Figure 28-15.195 A statistically significant difference between groups was apparent at 8 years (78% for the 78-Gy group versus 59% for the 70-Gy group, P = .004). The greatest gain was seen among patients with intermediate-risk or high-risk disease who had pretreatment PSA levels of more than 10 ng/mL; the freedom from failure rates were 78% for the 78-Gy group and 39% for the 70-Gy group (P = .001) (Fig. 28-16).292 In that trial, however, only 16 patients had pretreatment PSA levels higher than 20 ng/mL, which means that few men in that study had high-risk disease by PSA criteria. No significant difference in freedom from failure rates

TABLE 28-6.  External Beam Radiation Therapy: Retrospective Dose-Escalation Studies Percent Free From Biochemical Failure At 5 Years Study and Reference

Number of Patients

Risk

Dose 1

Dose 2

P Value

743

Low Intermediate High

87 (≤70 Gy) 56 (≤70 Gy) 37 (≤70 Gy)

96 (≥75.6 Gy) 78 (≥75.6 Gy) 53 (≥75.6 Gy)

NS .04 .03

Lyons et al, 2000283

738

Low High

81 (<72 Gy) 41 (<72 Gy)

98 (≥72 Gy) 75 (≥72 Gy)

.02 .001

Hanks et al, 2000286

618

Low† Intermediate High

86 (<70 Gy) 29 (<71.5 Gy) 8 (<71.5 Gy)

80 (≥70 Gy) 66 (≥71.5 Gy) 29 (≥71.5 Gy)

NS <.05 <.05

M. D. Anderson Cancer Center Study, 2000

1213

Low Low Intermediate Intermediate High High

84 (≤67 Gy) 91 (>67-77 Gy) 55 (≤67 Gy) 79 (>67-77 Gy) 27 (≤67 Gy) 47 (>67-77)

91 (>67-70 Gy) 100 (>77 Gy) 79 (>67-70 Gy) 89 (>77 Gy) 47 (>67-77 Gy) 67 (>77 Gy)

NS NS .0001 NS .0001 .016

Zelefsky et al,

*Four-year

1998167 *

results. on pretreatment prostate-specific antigen (PSA) level. NS, not significant.

†Based

Freedom from clinical and/or biochemical failure

1.00

0.75 70 Gy 78 Gy

0.50

P = .004

Year 1 2 3 4 5 6 70 Gy # pts 141 131 117 104 88 69 78 Gy # pts 146 134 120 102 86 77

0.25

7 53 65

8 33 42

9 24 28

10 18 13

0.00 5

0

10

15

Years after end of RT

FIGURE 28-15.  Freedom from biochemical failure rates after external beam irradiation to 70 Gy (by conventional radiation therapy) or 78 Gy (by conformal boost radiation therapy) in a randomized, dose-escalation trial at the M.  D. Anderson Cancer Center. The 10-year freedom from biochemical failure rates were 50% in the group treated to 70 Gy and 73% in the group treated to 78 Gy. (From Kuban DA, Tucker SL, Dong L, et al. Long-term results of the M.  D. Anderson randomized dose-escalation trial for prostate cancer. Int J Radiat Oncol Biol Phys 2008;70:67-74.)

Freedom from clinical and/or biochemical failure

1.00

Percent free of biochemical evidence of disease for 4 years bNED

28  The Prostate

699

100 R 2 = 0.969 P < 0.001

90

(n = 35)

80 70

(n = 92)

60

(n = 24)

50 40 (n = 61)

30

n = 286

(n = 74)

20 60

64

68

72

76

80

Dose (Gy)

FIGURE 28-17.  Evidence of a radiation dose-response ­ effect for patients with a pretreatment prostate-specific antigen (PSA) level of more than 10 ng/mL and category T1-T2 disease treated at M.  D. Anderson Cancer Center with radiation therapy only. Freedom from biochemical failure at 4 years is plotted against the mean dose at five dose levels: ≤ 65 Gy (n = 74), 65 to 67 Gy (n = 61), > 67 to 69 Gy (n = 24), > 69 to 77 Gy (n = 92), and > 77 Gy (n = 35). A highly significant linear relationship was found between mean dose at each level and freedom from failure. (From Pollack A, Smith LG, von Eschenbach AC. External beam radiotherapy dose-response characteristics of 1127 men with prostate cancer treated in the PSA era. Int J Radiat Oncol Biol Phys 2000;48:507-512.)

0.75

0.50 PSA > 10 ng/mL 0.25 70 Gy 78 Gy

P = .001

0.00 0

5

10

15

Years after end of RT

FIGURE 28-16.  Differences in freedom from biochemical failure rates after external beam irradiation to 70 Gy (by conventional radiation therapy) or 78 Gy (by conformal boost radiation therapy) among men with pretreatment prostate-specific antigen (PSA) levels of more than 10 ng/mL. (From Kuban DA, Tucker SL, Dong L, et al. Long-term results of the M.  D. Anderson randomized dose-escalation trial for prostate cancer. Int J Radiat Oncol Biol Phys 2008;70:67-74.)

was found among the relatively favorable-risk subgroup of patients who had pretreatment PSA levels less than 10 ng/mL (78% for the higher-dose group and 66% for the lower-dose group, P = 0.237). Another observation that may be important for future trials was that patients with the relatively favorable but still intermediate-risk features of category T1 to T2 disease and pretreatment PSA level higher than 10 ng/mL ­demonstrated the greatest difference in freedom from failure rates: 94% for the 78-Gy group and 65% for the 70-Gy group; however, at P = .076, this apparent difference was not statistically significant. This observation prompted a review of the dose-response relationship among all patients treated in the PSA era (between 1987 and 1997) who had category T1 to T2 disease and pretreatment PSA values of

less than 10 ng/mL.319 The clear dose-response relationship in this subgroup of patients is shown in Figure 28-17. Two additional prospective randomized trials have shown similar results. A multi-institutional study ­conducted at Massachusetts General Hospital and Loma Linda University Medical Center showed a significant benefit from 79.2 GyE compared with 70.2 GyE delivered at 1.8 GyE per fraction using proton therapy and followed by threedimensional conformal photon therapy.293 In that study, 393 men were randomly assigned to receive 19.8 GyE or 28.8 GyE using protons directed only to the prostate, followed by 50.4 Gy to the prostate and seminal vesicles using three-dimensional conformal photon therapy. The 5-year biochemical failure rate (according to the ASTRO definition of three consecutive increases in PSA level) was 91.3% versus 78.8% (P < .001) in favor of the high-dose treatment. Moreover, this benefit was seen in patients with low-risk disease (97.3% versus 82.6%, P < .001) and in those with intermediate- or high-risk disease (87.4% versus 74.5%, P = .02). However, no difference was seen in overall survival rates, which were 96% to 97% in both treatment groups. A Dutch multicenter randomized trial also showed a benefit from 78 Gy compared with 68 Gy in 669 men with T1b to T4 prostate cancer.326 All men were treated with three-dimensional CRT with the dose defined at the isocenter; the CTVs varied according to clinical T category, PSA levels, and Gleason score. At a median follow-up time of 51 months, the 5-year freedom from failure rates were 64% for the 78-Gy group and 54% for the 68-Gy group (P = .02). This benefit was most pronounced among patients with intermediate- and high-risk disease features. Unlike the previous randomized trials, the Dutch study did not demonstrate a significant difference in late ­gastrointestinal

700

PART 8  Male Genital Tract

TABLE 28-7. External Beam Radiation Therapy: Doses That Produce Rectal Reactions of Grade 2 or Higher

TABLE 28-8. External Beam Radiation Therapy: Rectal Volume Irradiated and Rectal Reactions of Grade 2 or Higher

Study and Reference

Dose

Incidence of Rectal Toxicity

Time at Analysis

Study and Reference

≤70 Gy 70-75 Gy >75 Gy

22% 20% 60%

2 yr

Shipley et al, 1995325

67.2 Gy 75.6 CGE

12%*

Lee et al, 1996205

<72 Gy 72-76 Gy >76 Gy

Zelefsky et al, 1998167 Storey et al, 200015

Rectal Volume Rate of Rectal Irradiated Toxic Effects

Benk et al, 1993329

75 CGE

<40% ARW

75 CGE

≥40% ARW

10 yr

Lee et al, 1996205

74-76 Gy

Rectal block

74-76 Gy

No block

10% (18-month actuarial) 19%

7% 16% 23%

1.5 yr

Dearnaley et al, 1999327

64 Gy 64 Gy

3DCRT Conven-RT

8% (5-yr actuarial) 18%

≤70.2 >75.6

6% 17%

5 yr

Boersma et al, 1998330

70 Gy 70 Gy

≤ 30% >30%

0% (crude)† 9%

70 Gy 78 Gy

14% 21%

5 yr

Storey et al, 200015

70 Gy 70 Gy

≤25% >25%

13% (5-yr actuarial) 37%

32%

*Grade 1 and 2 rectal reactions. CGE, cobalt gray equivalent.

toxicity between the two treatment groups (rates of late grade 2 or higher effects were 32% for the 78-Gy dose group and 27% for the 68-Gy dose group, P = 0.2), perhaps because of rectal dose constraints specified by the protocol and routine review of treatment plans for quality control. Another prospective randomized trial327 and several ­retrospective series have shown that three-dimensional CRT produces fewer side effects than standard techniques do when similar doses are used. Adverse side effects from dose escalation have primarily been rectal. Although significant bladder toxicity has not been reported, longer ­followup may be needed for the effects to become evident. Late rectal bleeding is the most common complication resulting from increasing the radiation dose to the prostate. In the M.  D. Anderson randomized trial, the rates of grade 2 or higher rectal reactions at 5 years were 14% in the 70-Gy group and 21% in the 78-Gy group. The corresponding rates at 10 years were 13% in the 70-Gy group and 26% in the 78-Gy group (P = .013). The finding that more than 20% of patients in the high-dose group experienced such reactions indicates that 78 Gy to the isocenter is near the limit for this technique. Rectal reactions are a function of dose and volume. The relationship of prescribed dose to the prostate and the incidence of rectal complications of grade 2 or higher is shown in Table 28-7.15,167,205,325,328 The irradiated volume is also an important determinant of rectal toxicity (Table 28-8).15,205,327,329,330 For example, Lee and colleagues205 observed that restricting the dose to the anterior rectal wall to 72 Gy or less reduced the incidence of rectal bleeding from 23% to 7%. In the M.  D. Anderson trial, a DVH analysis was performed for patients given the 78-Gy conformal boost treatment to examine possible predictors of rectal bleeding.15 When more than 25% of the rectum received 70 Gy or more, the 5-year actuarial incidence of grade 2 or higher rectal bleeding was 37%, but it was 13% when less than 25% of the rectum received 70 Gy or more (Fig. 28-18).15 The current M.  D. Anderson ­ recommendations

19% (40-month actuarial)* 71%

*Any

rectal bleeding. rectal bleeding. ARW, anterior rectal wall; CGE, cobalt gray equivalent; Conven-RT, conventional radiation therapy; 3DCRT, three-dimensional conformal radiation therapy.

†Severe

100

≤ 25%

90 Percent complication-free

Smit et al, 1990328

Dose

80 70 60

> 25%

50 40 30 20

P = 0.05

10 0 0

12

24

36

48

60

Time (months)

FIGURE 28-18.  Kaplan-Meier analysis of the incidence of rectal reactions (grade 2 or higher) among patients in the M.  D. Anderson Cancer Center dose-escalation trial divided by the percentage of rectum that had been treated to 70 Gy or more. The  incidence of rectal side effects was significantly reduced when the amount of rectum receiving 70 Gy or more was less than 25%. (From Storey MR, Pollack A, Zagars GK, et al. Complications from dose escalation in prostate cancer: preliminary results of a randomized trial. Int J Radiat Oncol Biol Phys 2000;48:635-642.)

for rectal dose constraints include multiple thresholds, specifically less than 20% of the total rectum volume at 70 Gy, less than 35% at 60 Gy, and less than 50% at 45 Gy. Various other dose-volume constraints are used by other institutions and generally perform well in predicting late rectal side effects regardless of whether the entire rectum volume is included in the calculations or only the rectal wall.331

28  The Prostate

Risk of Lymph Node Involvement and Whole-Pelvis Treatment Between 30% and 50% of patients with high-risk prostate cancer have lymph node involvement at diagnosis.139,140 An unresolved issue is whether to irradiate the whole pelvis in these patients. Several randomized and retrospective studies have been done, but no conclusive evidence exists to support this practice. RTOG protocol 77-06 examined the use of elective pelvic irradiation for patients with stage A2 or B (T1b or T2) prostate carcinoma without clinical evidence of lymph node involvement. This population was estimated to have about a 25% risk of nodal involvement. Elective lymph node treatment did not significantly affect outcome.332 The reports of retrospective data are about evenly split between those that support this treatment333-337 and those that do not.246,338-340 There is obvious reluctance to limit coverage of areas known to harbor tumor in almost 40% of patients with prostate cancer. The RTOG attempted to answer the question of whether to use whole-pelvis irradiation in protocol 94-13.341 This prospective trial randomly assigned men with a greater than 15% risk of nodal involvement to undergo whole-pelvis radiation therapy (to a maximum central axis dose of 50.4 Gy) followed by cone-down boost to the prostate (an additional 19.8 Gy) or 70.2 Gy (prescribed to the isocenter) to the prostate and seminal vesicles; men were further randomized to receive 4 months of total androgen suppression beginning 2 months before radiation or at the completion of radiation. The study used a 2 × 2 factorial design and was not intended to be an independent four-arm trial. The primary end point, ­progression-free survival, was significantly better in the whole-pelvis radiation plus neoadjuvant hormone therapy group in the original report.341 However, an updated report with a median follow-up time of almost 7 years342 showed no difference in progression-free survival or overall survival between the whole-pelvis versus prostate-only ­ radiation groups or between the neoadjuvant versus adjuvant hormone therapy groups (the 5-year progression-free survival rates were approximately 50% in all four groups).342

External Beam Radiation Therapy with Androgen Ablation for Clinically High-Risk Disease The rationale for combining EBRT with AA is that the combination is expected to be more effective in eradicating prostate cancer than either treatment given separately. ­Because both modalities cause apoptosis, the suggestion has been made that the combination of the two may ­result in additive or supra-additive cell killing.343-345 From an ­additive perspective, neoadjuvant AA would reduce the number of clonogens that must be eradicated by radiation, and the combination should result in improved local control. This hypothesis was initially tested by the RTOG in phase II clinical trials346,347 and later in phase III trials.348,349 The advantage of giving AA before radiation, as opposed to starting both modalities together, has never been established. The distinction between neoadjuvant and adjuvant AA is blurred in clinical trials. Neoadjuvant AA is defined as that begun before radiation, but the AA is then continued during radiation therapy and for various periods afterward. Adjuvant AA is that begun during radiation; however, when adjuvant AA is started at the beginning of

701

a protracted course of radiation therapy, the tumor shrinks during radiation such that fewer clonogens remain at the end of the radiation period. There is a component of adjuvant AA in neoadjuvant treatment and vice versa. No evidence has been found to indicate that the results from EBRT preceded by neoadjuvant AA are different from those achieved with adjuvant AA when the AA is begun at the same time as the radiation. Another reason for starting AA before radiation is to shrink exceptionally large prostates so that the volume of rectum and bladder receiving high doses of radiation is lessened when conformal radiation therapy is applied. This philosophy is espoused by several groups (Table 28-9).350-353 The prostate shrinks by 30% to 40% in ­ response to 3 months of neoadjuvant AA, and DVH analyses have confirmed significant reductions in the proportions of bladder and rectum that receive 80% to 90% of the prescribed dose. Although these gains should translate into reductions in morbidity, the effect of shrinking the PTV with neoadjuvant AA before radiation in patients with locally advanced disease and extracapsular extension is uncertain. Microscopic foci may be left in the regions where the prostate was originally, regions that after neoadjuvant AA become filled by rectum or bladder. The published randomized trials that have incorporated neoadjuvant AA348,349,354-362 (Table 28-10) used whole-pelvis fields initially, with more generous coverage of the periprostatic tissues than can usually be accomplished with three-dimensional CRT. Interest in combined EBRT and AA has increased ­during the past several years, primarily in response to encouraging results from randomized trials. Results from randomized studies, most of which compared EBRT used alone and with AA, are summarized in Table 28-10.348,349,354-362 Only one trial355 failed to demonstrate superior rates of freedom from failure (i.e., disease or biochemical evidence of disease when available) for treatment with both modalities. A survival advantage also was ascribed to the EBRT plus AA combination in many of these trials. Brief reviews of each trial are provided in subsequent sections. The M.  D. Anderson Diethylstilbestrol Trial. The earliest multi-institutional randomized trial of EBRT with or without AA was championed by Del Regato363 in the 1960s TABLE 28-9. External Beam Radiation Therapy Given after 3 Months of Androgen Ablation: Effect on Target Volume Study and Reference

Reduction in Volume (%) Number of Patients Prostate* Rectum† Bladder†

Shearer et al, 1992350

22

43%

—

—

Yang et al, 1995351

7

43%

28 (D80)

46 (D80)

Forman et al, 20 1995352

37%

34 (D93)

20 (D93)

Zelefsky et al, 22 1994353

25%*

25 (D95)

50 (D95)

*Reduction

in the target volume. in the percent volume of the rectum or bladder receiving 80% (D80), 93% (D93), or 95% (D95) of the prescribed dose.

†Reduction

702

PART 8  Male Genital Tract

TABLE 28-10.  External Beam Radiation Therapy Plus Androgen Ablation: Results of Randomized Trails 5-Year Freedom from Failure Rate, % (n)

5-Year Overall ­Survival Rate, % (n)

5-Year Overall Survival Rate for Subgroups, %

RT Alone

RT + AA

RT Alone

RT + AA

Subgroup

RT Alone

RT + AA

15 (200)

36 (196)*

≈ 56 (230)

≈ 56 (226) GL 2-6

52 (NS)

70 (NS)*

39 (46)

69 (45)*

39 (46)

62 (45)*

LN+

0 (20)

50 (19)*

14.5 yr

49 (40)

71 (38)*

68 (40)

73 (38)

—

—

—

4.5 yr

≈ 30 (88)

≈ 44 (99)

≈ 43 (88)

≈ 58 (99)

—

—

—

(438)*

71 (68)

75 (477)

GL 8-10

55 (137)

66 (139)*†

62 (208)

79 (207)*

—

—

—

78

88*

—

—

—

RT + AA4

RT + AA28

71 (NS)

81 (NS)*

Early

Late

50 (256)

44 (244)*

Length or Type of Androgen Ablation

Follow-up Period

RTOG 8610348,359

≈ 4 mo (N)

4.5 yr

Umea

Perm (N)

9.3 yr

M. D. Anderson Cancer Center354

Perm (A)

MRC355

Perm (A)

Study and Reference

University358

Perm (A) RTOG 85-31356,361

4.5 yr

20 (429)

53

EORTC357

45 mo

48 (208)

85 (207)*

Harvard362 RTOG 92-02349 MRC360

3 yr (A) 6 mo (N)

4.52 y

4 mo (N) vs. 5.8 y 28 mo (N) Early vs. late

NS

(NS)*

80 (NS)

60

RT + AA4

RT + AA28 RT + AA4 (NS)*

28 (NS)

46

Early†

Late†

68 (256)

32

(201)*

RT + AA28

78 (NS)

80 (NS)

Early†

Late†

37 (469)

43

(465)*

GL 8-10

M0

*P

< .05, log-rank test. reviewed Gleason score. ‡Freedom from distant metastasis rate for patients with M0 (nonmetastatic) disease. A, adjuvant; AA, androgen ablation; EORTC, European Organisation for Research and Treatment of Cancer; GL, Gleason score; MRC, Medical Research Council (United Kingdom); N, neoadjuvant; NS, not stated; Perm, permanent; LN+, lymph node–positive; RT, radiation therapy; RTOG, Radiation Therapy Oncology Group. Modified from Pollack A, Kuban D, Zagars GK. Impact of androgen deprivation therapy on survival in men treated with radiation for ­prostate cancer. Urology 2002;60(3Suppl 1):22-30. †Centrally

for locally advanced, category T3 to T4 prostate cancer. AA was accomplished by using diethylstilbestrol, a ­ synthetic estrogen, beginning at the end of the radiation ­ period. The trial design was to continue the diethylstilbestrol indefinitely. Results from this study were never published in their entirety, but Zagars and colleagues354 published the results for patients who were randomized and treated at M.  D. Anderson. At a median follow-up time of 14.5 years, a highly significant and sustained improvement in freedom from disease was observed for those treated with EBRT plus AA (see Table 28-10).354 No survival difference was detected, a finding that might have been attributable to the small number of patients in each group (approxi­mately 40) and excess intercurrent deaths as a consequence of the diethylstilbestrol. The high dose of diethylstilbestrol (5 mg/day) used when the trial was begun was later reduced in light of evidence that this dose caused increased cardiac mortality from thromboembolic events.364 RTOG 86-10. In RTOG protocol 86-10,348,359 patients in the combined-modality group were given a short course of neoadjuvant AA starting 2 months before EBRT and continuing throughout the radiation period. Of the 471 patients entered, 456 were evaluable. Eligibility criteria included having locoregionally advanced disease, with T2 to T4 tumors (>25 cm2 on digital rectal examination). Smaller tumors with evidence of lymph node involvement

below the common iliac vessels also were permitted. The median follow-up time was 4.5 years. Standard radiation doses were used, with 45 Gy given to the pelvis plus a 20to 25-Gy boost to the prostate. Extended fields were used to treat the lymph nodes when they were involved. Total AA was accomplished by using goserelin and flutamide. Statistically significant improvements in local control rate and freedom from biochemical failure rate were observed (see Table 28-10),348 along with a borderline improvement in the rate of freedom from distant metastasis. No difference in overall survival was seen. In an update of this trial,359 these relationships were strengthened, and a survival benefit was seen for patients with Gleason 2 to 6 disease. RTOG 85-31. The RTOG 85-31 protocol356,361 also involved patients with locoregionally advanced disease, including those who had undergone primary definitive treatment with radiation and those who had undergone radical prostatectomy and were identified at pathologic examination as being at high risk. The definitively irradiated patients were men with clinical category T3 (<25 cm2) or T1 to T2 disease with radiographic or histologic lymph node involvement, including the para-aortic nodes. Patients were eligible after prostatectomy if capsular penetration with positive margins or seminal vesicle involvement had been found. Of 977 patients enrolled, 945 were evaluable. Conventional radiation therapy techniques were used to

28  The Prostate

deliver a total dose of 65 to 70 Gy, with the whole pelvis (and para-aortic lymph nodes as indicated) treated to 44 to 46 Gy. Men who had had radical prostatectomy were given a total dose of 65 to 66 Gy. AA was accomplished by monthly injections of goserelin acetate (Zoladex) beginning the last week of radiation therapy. Combined-modality treatment reduced the 5-year rates of local failure (16% versus 28%), distant metastasis (17% versus 30%), and biochemical failure (47% versus 80%). All of these differences were highly statistically significant. Initially, no difference was observed in overall survival for the entire cohort; however, the survival rate of patients with Gleason 8 to 10 disease was better among those given combined-modality therapy (see Table 28-10). In an update at a median follow-up time of more than 7.5 years, an overall survival benefit was seen in favor of the adjuvant AA treatment (10-year overall survival rates of 49% versus 39%, P = .002), with continued benefit in other clinical outcomes.365 Umea Trial. In 1986, a trial at Umea University in Sweden was opened to patients with T1-4 pN0-3 disease. Patients were randomly assigned to receive radiation therapy, given to relatively standard doses, or combined orchiectomy and radiation therapy. AA was started 5 weeks before the start of radiation therapy. The trial was closed early in 1991, when a survival benefit was observed favoring the combined-treatment group. An update of that trial358 at a median follow-up of 9.3 years showed that the combinedmodality group continued to have higher rates of freedom from progression and overall survival. The survival benefit was attributed mainly to the effects on men with lymph node–positive disease; 50% of these men treated with radiation and AA remained alive at 10 years, compared with no survivors in the group given only radiation therapy. In this trial, early, permanent AA in combination with EBRT had a dramatic effect on survival. European Organisation for Research and Treatment of Cancer Trial. Perhaps the most convincing study to document the benefit of combining AA plus radiation comes from the European Organisation for Research and Treatment of Cancer (EORTC). Bolla and colleagues357 reported the results of this study, in which 415 men with locally advanced disease were randomly assigned to undergo EBRT alone or in combination with an extended but temporary course of AA. Most patients had stage T3 to T4 disease, although some had high-grade T1 to T2 disease. AA was accomplished by giving patients goserelin acetate for 3 years. Cyproterone acetate was given during the first month to counter the surge in androgens that occurs with the first injection of a luteinizing hormone–releasing hormone agonist. The radiation therapy dose was 70 Gy. Median follow-up was 45 months. The local control rates were 97% in the combined-treatment group and 77% in the radiation-only group. Overall rates of freedom from disease (85% versus 48%) and survival (79% versus 62%) (see Table 28-10) were significantly higher in the group ­given radiation therapy plus AA. The finding of improvement in overall survival rate at only 45 months of follow-up was surprising. The investigators showed that the difference in survival was mainly related to deaths from cancer. A reanalysis of the data published in 2002, at a median follow-up time of 66 months,366 confirmed that the survival advantage from radiation therapy plus AA had been sustained.

703

RTOG 92-02. The randomized RTOG 92-02 trial349 compared the efficacy of short-course AA, as was administered in RTOG 86-10, with that of long-course AA, similar in principle to the European study described previously. The short-course group received 4 months of AA beginning 2 months before EBRT. The long-course group received the same 4 months of AA before and during radiation, followed by an additional 24 months of AA. There were 1554 patients enrolled, and median follow-up was 48 months. Standard radiation doses of 65 to 70 Gy were used, with the pelvic lymph nodes receiving 44 to 50 Gy. Patients had locally advanced T2c to T4 disease, with PSA levels less than 30 ng/mL. This trial was important because it helped to clarify the benefit of long-course AA and because it was truly a contemporary PSA-era trial. Long-course AA conferred improvement in every end point measured except overall survival; improvements were seen in 5-year rates of freedom from biochemical failure (46% versus 28%, P < .001), freedom from distant metastasis (89% versus 83%, P = .001), and cause-specific survival (94.6% versus 91.2%, P = .003). In a subset analysis of patients with Gleason 8 to 10 disease, the overall survival rate was higher for those taking long-course AA (81% versus 70.7%, P = .044), demonstrating, as was the case in RTOG protocol 85-31, that patients with high-grade disease require prolonged AA for the maximum benefit. Androgen Ablation Alone: The Medical Research Council Trials. The need for an AA-alone control group in these trials cannot be overstated. Except for the Medical Research Council trial,355 none of the other studies presented in Table 28-10 included an AA-alone control group because AA alone is considered palliative treatment and few patients in the United States would agree to participate. However, some evidence suggests that AA alone prolongs survival, which raises the question of what EBRT contributes to AA, especially when long-course AA is used. The older Veterans Administration studies364 compared early AA with delayed AA given at relapse. Although most of these studies did not seem to show any survival advantage for early AA, two factors might have played a role. One factor is that diethylstilbestrol, the synthetic estrogen used in most of these studies, predisposes its recipients to thromboembolic and cardiovascular events, resulting in early deaths. Another factor is that intercurrent deaths in older men make it more difficult to detect a survival advantage. When younger men were analyzed separately from older men,367 early AA was found to confer a survival advantage over delayed AA. A large, multi-institutional study sponsored by the Medical Research Council in the United Kingdom360 reexamined this question using less toxic AA methods. Early AA was achieved by a luteinizing hormone–releasing hormone agonist or orchiectomy. Patients with locally advanced disease (category T2 to T4) or limited asymptomatic ­metastasis were enrolled. Early AA resulted in lower rates of disease progression and higher survival rates than did late AA. Mortality was most significantly reduced among patients who had locally advanced disease with no evidence of distant metastasis; these are the patients who are typical of the candidates for the radiation therapy trials summarized in Table 28-10. The findings emphasize the need for AA-only control groups so that the contribution of each modality to outcome and survival can be deciphered.

704

PART 8  Male Genital Tract

Only one randomized trial that examined the utility of EBRT plus AA has included an AA-only group.355 It has been criticized for being underpowered, with only about 90 patients per group, and for not specifying the radiation dose or techniques used. The local and distant failure rates were high (about 70% and 75%, respectively), and the overall survival rate was less than 40% at 7 years. These patients must have had exceedingly advanced disease. That no differences were observed between the AA-only group and the combined-treatment group is not surprising. This trial is not representative of patients with high-risk disease receiving EBRT plus AA today. The findings from these Medical Research Council trials force caution in the interpretation of radiation therapy ­series, particularly because long-course AA suppresses androgens beyond the time of drug administration. Oefelein and colleagues368 have shown that after a single depot injection, the median time for serum androgen levels to recover to baseline levels is more than 6 months. Still longer androgen-recovery periods have been reported after multiple injections.369 The survival advantage from long-course AA plus EBRT may be attributable in part to the AA. Short-Course Androgen Ablation. The concept that short-course AA might be a substitute for radiation in patients with intermediate-risk prostate cancer has been brought to the forefront. In a retrospective analysis, D’Amico and colleagues370 compared the effects of 6 months of AA plus EBRT to about 70 Gy with those of EBRT alone for patients with favorable-, intermediate, and high-risk prostate cancer. The radiation-only group consisted of 1310 patients, and the combined-therapy group had 276 patients. The median follow-up interval was only 28 months and was calculated from the date of diagnosis; follow-up from the completion of treatment was short. These investigators found that among the men with intermediate- or high-risk disease, combined-modality treatment improved the rates of freedom from biochemical failure. However, the actual differences in the KaplanMeier curves were slight, and very few patients were at risk for failure at 5 years in the combined-modality group. The slight advantage reported for the short-course AA plus radiation therapy is not as great as the benefit from dose escalation. Moreover, the use of freedom from biochemical failure rates when the follow-up period is short may be misleading, because AA may delay failure. Substituting short-course AA for radiation is not advisable until longer follow-up results are available. In a subsequent prospective randomized study, D’Amico and colleagues362 showed benefits in progression-free and overall survival from the addition of 6 months of hormone therapy to radiation (see Table 28-10). In that trial, 206 men with primarily intermediate- to high-risk localized (T1-2b) prostate cancer were randomly assigned to receive 70 Gy to the prostate or the same radiation dose with 6 months of hormone therapy beginning 2 months before the start of radiation therapy. Most patients were at intermediate risk for biochemical failure, but approximately 15% had Gleason scores of 8 to 10, and 12% to 13% had pretreatment PSA levels of 20 to 40 ng/mL. After a median follow-up time of 4.52 years, the progression-free survival rate in the combination-therapy group was 80% and 60% in the radiation-alone group (P = .002). Surprisingly, an overall survival benefit at 5 years was also documented in the

combination-therapy group (88% versus 78% for ­radiation only, P = .04). This trial represents one of the only studies that showed an overall survival benefit from the addition of short-course hormone therapy (that lasting less than 8 months) to radiation. Before initiating hormone therapy for all patients with intermediate-risk features on the basis of these findings, several points are important to consider. This trial included a proportion of patients with high-risk disease who might have preferentially benefited from hormone therapy. However, the results of RTOG 86-10 suggested that only ­patients with Gleason scores of less than 7 would see a survival benefit. Moreover, the total dose of radiation used in this study was low by modern standards; however, none of the randomized dose-escalation trials has shown an overall survival benefit. The discovery in the RTOG 86-10 trial (see Table 28-10) of a survival advantage from EBRT plus AA for patients with Gleason scores of 2 to 6 suggests that short-course AA is beneficial. However, because the freedom from biochemical failure rates were extremely low, even for patients with ­favorable prognostic features, these findings could be explained by slower tumor growth, even after testosterone levels had returned to normal. Animal model experiments have proved that this is a possibility.371 Although administering short-course AA in combination with EBRT may be beneficial in that tumor growth is slowed far beyond the time when AA is withdrawn, the main objective is to enhance survival by curing patients of the disease. The other confounding factor is that the subgroup analysis in RTOG 86-10359 was based on Gleason scores that were centrally reviewed rather than the Gleason scores that were used for stratification. An uneven distribution of prognostic factors in the treatment groups is therefore possible. Multivariate analysis was not reported. Dose escalation will be a necessary component of future trials to establish the role of short-course AA. RTOG protocol 92-02, in which short-course AA plus EBRT was compared with long-course AA plus EBRT, ­revealed a survival advantage only for patients who had Gleason scores of 8 to 10 (see Table 28-10).349 In light of the findings described previously, the RTOG 92-02 observation suggests that patients with Gleason scores of 2 to 6 should not be treated with extended AA plus radiation, whereas patients with a Gleason score of 8 to 10 should. Although this may be true, evidence from studies with longer follow-up would be more convincing. Longer follow-up in protocol 92-02 may ultimately show an overall survival advantage from long-term hormone therapy such as that observed in 85-31; however, the freedom from biochemical failure and survival results must be examined together to distinguish between a survival advantage from cure and a survival advantage from a slowing of tumor growth. Moreover, the use of salvage hormone therapy and the timing of that use may affect subsequent survival beyond the effects of the initial (randomized) treatment.

External Beam Radiation Therapy plus Androgen Ablation for Lymph Node–Positive Disease The results of single-modality treatment for regional lymph node–positive (stage D1) disease have been disappointing. The freedom from failure rates at 5 years produced by radiation therapy alone246,372-377 (Table 28-11) and by radical prostatectomy alone378-383 (Table 28-12) have been ­similar

705

28  The Prostate

TABLE 28-11.    External Beam Radiation Therapy for Lymph Node–Positive (Stage D1) Disease Study and Reference Bagshaw 1984373 Smith et al,

1984372

Gervasi et al,

1989374

al,1992375

Follow-Up Period (mo)

5-Year Disease-Free Survival Rate

5-Year Overall Survival Rate

60

>60

≈ 15%

≈ 45%

46

>60

17%

60%

152

103

22%

65%

56

108

Number of Patients

61%

76%

Hanks et al, 1993376

—

—

≈ 30%

64%

Leibel et al, 1994246

345

105

≈ 35%*

≈ 70%

75

59

11%

65%

Lawton et

Lawton et al, *Free

1997377

of distant metastasis.

TABLE 28-12.  Radical Prostatectomy for Lymph Node–Positive (Stage D1) Disease Study and Reference

Number of Patients

Follow-Up Period (mo)

Disease-Free Survival Rate

Overall Survival Rate ≈ 55% (5 yr)

deKernion et al, 1990378

35

72

≈ 30%

Myers et al, 1992379

34

54

29%

1992380

77

60

24%

Zincke et al,

— 71% (10 yr)*

Sgrignoli et al, 1994381

113

—

≈ 20%

—

Messing et al, 1999382

51

85

—

≈ 58% (10 yr)

survival rate.

at about 20% to 30%. A slightly greater proportion of ­patients treated with AA alone have been free of failure at 5 years, although few patients remain free of disease progression by 10 years.384 Standard treatment for regional lymph node–positive (stage D1) disease has been (and usually still is) AA. Over the past 15 years, interest has been increasing in adding local therapy to AA to prolong freedom from progression. Zagars and colleagues384 found that local failure was the dominant site of disease progression in patients with lymph node–positive disease, with a median time to local failure of about 8 years. This observation prompted the policy to treat all such patients with permanent AA plus radiation therapy. The preliminary results were positive,139 and a subsequent update385 revealed that biochemical and clinical failure rates for patients treated with EBRT plus AA were far lower than those for men treated with AA alone (Fig. 28-19).384 The AA-alone group consisted of 183 patients (median follow-up time, 9.4 years) and the radiation plus AA group had 72 patients (median follow-up time, 5.9 years). The median radiation dose was 68 Gy, and although generous fields were used, the whole pelvis was not treated. The differences observed between ­combined treatment and AA-only treatment in univariate analyses were sustained in multivariate analyses: combined­modality treatment produced superior rates of freedom from biochemical failure, freedom from distant metastasis, and survival. Radical prostatectomy may be equally effective at prolonging freedom from failure and survival when combined with AA for lymph node–positive disease,382,386 indicating that the addition of a local treatment, such as

100 AA + RT

90 80 70 Percent

*Cause-specific

60 50 AA alone

40 30 20

Significant in multivariate analysis

10 0 0

1

2

3

4

5

6

7

8

Years after lymphadenectomy

FIGURE 28-19.  Freedom from relapse or from a rising prostatespecific antigen level among patients with regional lymph node– positive (stage D1) prostate cancer treated with permanent  ­androgen ablation (AA) and external beam radiation therapy (RT) or with AA alone. (From Zagars GK, Sands ME, Pollack A, et al. Early androgen ablation for stage D1 (N1-N3, M0) prostate cancer. Prognostic variables and outcome. J Urol 1994;151:1330-1333.)

r­ adiation therapy or radical prostatectomy, to AA would improve disease-free and overall survival rates. The RTOG attempted to contrast EBRT plus AA with AA alone in a randomized trial; however, accrual was too slow, and the trial was closed. It is doubtful that a randomized trial will

706

PART 8  Male Genital Tract

be successfully completed. The recommendation is to use radiation therapy plus permanent or at least extended AA for the treatment of lymph node–positive disease.

Particle Beam Therapy Particle beam therapy for prostate cancer has consisted mainly of treatment with neutrons or protons. In two randomized trials, the efficacy of neutron or neutron-photon therapy was reportedly better than the efficacy of photons, but the side effects of neutrons were significantly ­greater.387-389 Some groups continue to use conformal neutrons, and the results remain encouraging.390,391 It is impossible to assess from existing data the equivalence of neutrons or neutron-photons with dose-escalated photons. Given the discovery that prostate cancer may behave as a late-reacting tissue,183,184 interest may be renewed in directly comparing neutrons and photons in the PSA era. Conformal proton therapy has tremendous potential in the treatment of prostate cancer. The Bragg peak and the similarity in the relative biologic effectiveness of protons and photons create the potential for precipitous dose fall-off beyond the CTV that is much greater for protons than for photons. With photons, the posterior aspect of the rectum usually receives about 50% of the prescribed dose. With protons, the posterior rectum presumably receives a much smaller amount. All of the available results on dosevolume relationships for rectal morbidity indicate that the complication rates should be lower with protons than they have been with photons. A randomized trial of proton boost done at Massachusetts General Hospital revealed increased late rectal bleeding in the proton-boost group of patients, who received 75.6 GyE compared with only 67.2 Gy for the photon group. A suggestion of an improvement in the 5-year local control rate for the high-dose treatment (92% versus 80%, P = .089) came with significantly increased rectal toxicity, which might have been related to the use of a perineal proton boost field, limited proton beam energy, and suboptimal CT data for planning.325,329 Results from studies conducted at Loma Linda392 have been promising, but definitive clinical evidence that protons are superior to photons is still lacking. A multicenter prospective randomized study comparing 79.2 GyE with 70.2 GyE given as combination proton and photon therapy did show a benefit in terms of biochemical control (when failure was defined as three increases in PSA level) for the higher radiation dose among men with localized prostate cancer (5-year freedom from biochemical failure rates of 80.4% versus 61.4% for the lower-dose group, P < .001).293 Although it was later discovered that the original report published in 2005 involved inaccurate calculations of PSA failure (defining failure as any three PSA level increases rather than three consecutive increases), a subsequent reanalysis in 2008 using the correct definition also showed a significant difference in PSA outcome between treatment groups (5-year biochemical control rates of 91.3% for the higher-dose group and 78.8% for the lower-dose group, P < .001).293 In that study, 393 men with 1992 AJCC clinical stage T1b-2b disease and PSA levels of less than 15 ng/mL were randomly assigned to receive an upfront proton boost of 19.8 GyE or 28.8 GyE, followed by three-dimensional CRT to 50.4 Gy to the prostate and seminal vesicles. Benefits in biochemical and local control were observed in the high-dose group, but overall survival

was not different for the two groups (97% versus 96%). A post hoc analysis also showed a biochemical-control benefit from the higher dose in the subgroups with low- and combined intermediate- and high-risk disease. This trial was not a randomized comparison between protons and photons, and a relatively small portion of the dose was administered with protons. However, the rates of grade 2 or higher late rectal toxicity in this study were slightly lower (9% in the low-dose group and 18% in the high-dose group, P = .005) than those in the M.  D. Anderson randomized trial of three-dimensional CRT (13% versus 26%, P = .001).292

Prostate Biopsies after Radiation Therapy Much debate has ensued over the significance of positive findings on prostate biopsy after definitive radiation therapy. Several groups have found high rates of positivity after ­radiation therapy,393,394 even in patients whose PSA levels do not rise over time. The average biopsy-positivity rate is about 40%. The assumption that all patients with positive biopsy findings will experience treatment failure is unjustified given the findings of Cox and Stoffel395 and others.396-399 These investigators found that carcinoma cells are cleared from the prostate over time, suggesting that most of the treated ­carcinoma cells identified at biopsy eventually die. In the M. D. Anderson randomized trial of conventional versus conformal boost dose-escalation therapy,200 prostate biopsy samples were obtained at 2 years after treatment. A significant proportion of the biopsy samples contained “atypical cells consistent with or suspicious for treated carcinoma” or “carcinoma with treatment effect.” The clonogenicity of such cells is unclear. The preliminary analysis of patients in the trial who did not have evidence of biochemical failure at the time of prostate biopsy revealed that the presence of carcinoma cells showing the treatment effect was associated with the same outcome as the presence of frank carcinoma cells.200 Investigators at Memorial Sloan-Kettering have advocated that prostate biopsies be taken at 3 years after treatment to allow clearance and the demonstration of a dose response.167,320 Staining with proliferation markers such as Ki-67 or proliferating-cell nuclear antigen may aid in establishing the true rate of residual cancer clonogenicity. Crook and colleagues398,399 found that the presence of staining for proliferating-cell nuclear antigen in biopsy samples classified as “indeterminant” correlated with an increased frequency of relapse.

Salvage Procedures after Radiation Therapy Biochemical failure precedes clinical manifestation of disease by several years, allowing much earlier identification of local relapse in the absence of detectable distant spread. Factors associated with slow progression of disease after biochemical failure include a delay in treatment failure of more than 2 years, a PSA doubling time of more than 1 year, and a Gleason score of less than 8.144,150,400-402 These factors and others such as a history or current evidence of extracapsular disease, serum PSA level (the lower the better; salvage therapy when PSA is >10 ng/mL is not ­advised), DNA ploidy,403 and possibly the presence of ­ molecular markers404 should be considered when evaluating patients for salvage procedures. With appropriate selection of patients, salvage procedures are reasonable for young,

28  The Prostate

healthy men with a life expectancy of at least 10 years, although complication rates are invariably high.

Salvage Prostatectomy All of the reports on salvage prostatectomy suffer from relatively small patient numbers. Careful selection of ­patients is the key to obtaining 5-year freedom from failure rates of roughly 50% in the PSA era.403-409 The rates of urinary incontinence and reoperation for complications are high. Historically, androgen deprivation therapy was used as salvage therapy after radiation therapy failure in most cases.410 Surgical resection after radiation therapy can be challenging because of fibrotic and ischemic changes around the ­irradiated prostate. Radiation-induced changes can obliterate normal surgical planes and result in aberrant vasculature, which makes blunt dissection and postoperative healing more difficult. Careful patient selection of men who are younger and have good performance status, favorable PS-level kinetics, and Gleason scores of less than 8 after ­radiation failure is critical for optimizing outcomes, and salvage prostatectomy should only be performed by surgeons with extensive experience with this procedure.

Salvage Cryotherapy Cryotherapy for prostate cancer that recurs after radiation therapy is associated with high rates of urinary incontinence and severe complications requiring reoperation. ­ Severe complications seem to occur more often and the freedom from failure rates are lower after salvage cryotherapy than after salvage prostatectomy.411-415 Complications have ­included urinary incontinence, urinary obstruction, rectal or perineal pain, scrotal edema, hematuria, osteitis pubis, and rectourethral fistulas. Modern cryotherapy procedures and the use of urethral and rectal warming catheters may improve these side effects. At M.  D. Anderson, salvage radical prostatectomy is considered for younger, healthier patients, and salvage cryotherapy is reserved for patients who are older or have lower performance status.

Salvage Radioactive Seed Implantation after ­External Beam Radiation Therapy Another option for addressing locally persistent prostate cancer after EBRT is brachytherapy by means of PRSIP.416,417 Although published results in the PSA era have been limited, it seems that PRSIP under these circumstances produces long-term biochemical control in about 30% to 50% of cases. The results may be better with the use of stricter selection criteria. The rates of side effects such as urinary incontinence have been lower after PRSIP than those after salvage prostatectomy or salvage cryotherapy. Severe complications have occurred infrequently. The side effects of salvage brachytherapy after radiation failures in the modern dose-escalation era will require further study. The RTOG is conducting a phase II study (05-26) examining the role of salvage brachytherapy after radiation failure, with eligibility limited to patients who received 72 Gy or less in 2-Gy fractions or 75.6 Gy or less in 1.8-Gy fractions.

Radical Prostatectomy Radical prostatectomy was first described more than 140 years ago. The initial perineal approach has given way to the retropubic approach now used by most surgeons. In

707

the late 1970s and early 1980s, Walsh and colleagues3,418 defined the anatomy of the dorsal vein and periprostatic venous complex and the neurovascular bundles controlling erectile function. These advancements led to nerve-sparing prostatectomy. The nerves that control erectile function are spared in patients for whom the risk of extracapsular ­extension is thought to be low. Sometimes, the neurovascular bundle is sacrificed on one side where the risk of extracapsular extension is deemed to be high. When both nerves are left intact, the risk of impotence is reportedly as low as 15%.419,420 When one nerve is sacrificed, the potency rate is further reduced by at least one half. On average, the incidence of impotence after radical prostatectomy ranges from 40% to 60%,421-425 with the higher rates typically being reported when questionnaires are used rather than face-to-face interviews. Overall potency rates for a given population may be ­related to age, history of diabetes and coronary artery disease, use of medications (e.g., antihypertensives), and other factors. The timing of the assessment is also critical, because erectile function returns gradually after radical prostatectomy.17 In contrast, erectile function decreases after radiation therapy.17 Assessment should be made at least 1 year after surgery. Impotence rates after radical prostatectomy, when measured beyond 1 year after the surgery, are not much different from those after radiation therapy. Sural nerve grafts may decrease impotence rates after radical prostatectomy.426 Urinary incontinence after radical prostatectomy is a major drawback to the procedure. The most objective measure of incontinence is the need for pads after at least 1 year of follow-up. Questionnaire series have shown that rates of incontinence range from 13% to 47%.423-425 In contrast, the percentage of patients with urinary incontinence requiring pads after radiation therapy is less than 5%.16,427,428 However, the differences in overall urinary function between men who have undergone radical prostatectomy and those who have undergone radiation therapy are not as great as may be anticipated when urinary incontinence is considered as an isolated factor.429

Clinical and Biochemical End Points after Radical Prostatectomy Radical prostatectomy is an effective treatment for prostate cancer. Before the PSA era, clinical local relapse rates were extremely low, suggesting that clinical failure was a consequence of occult regional and distant metastatic disease that had been present at the time of surgery. ­Although most urologists probably still believe that a detectable PSA level after prostatectomy does not reflect failure to extirpate all prostate cancer cells from the surgical bed, increasing evidence suggests that residual prostate cancer cells remain in the prostatic fossa much more often than previously believed. Biochemical failure occurs approximately 3 to 5 years before clinical relapse.144,145 Even though PSA is secreted in small amounts by nonprostatic tissues,430 a detectable PSA value of 0.2 to 0.4 ng/mL after prostatectomy is a strong correlate of eventual clinical relapse.150 Earlier recognition of treatment failure through the use of biochemical criteria has prompted more aggressive biopsies of the prostatic fossa and urethrovesicular anastomosis. In some series, biochemical failure has been associated with the

708

PART 8  Male Genital Tract

presence of residual prostate cancer cells in the surgical bed in more than 50% of cases.148,149

Surgery for Favorable Disease Most patients undergoing radical prostatectomy in the United States have favorable pretreatment clinical characteristics, such as a PSA level of 10 ng/mL or less, a ­Gleason score of 6 of less, and a 1992 AJCC clinical T classification of T2a or less. Results from some large studies of outcome after radical prostatectomy are displayed in Table 28-13.152,287,288,431,432 Because the patients in radical prostatectomy series may have more favorable characteristics on average than those treated with EBRT, patients should be stratified as much as possible to compare results between modalities. The most important predictor of biochemical failure is pretreatment PSA level, which has been the main factor used to compare different treatment modalities. At 5 years, Kaplan-Meier survival estimates of freedom from biochemical failure for patients with a pretreatment PSA level of 10 ng/mL or less are 80% to 90%, estimates similar to those from trials of EBRT or PRSIP (see Tables 28-3 and 28-5). Comparing patients on the basis of pretreatment PSA level may not be sufficient, however, because subtle differences in Gleason score and clinical stage may be present. Other factors that could independently predict outcome may not have been randomly allocated between groups. The number of pretreatment biopsy samples involved with cancer and the proportion of cancer in the biopsy specimens, for example, are factors to be considered in recommending radical prostatectomy or PRSIP. Biopsy of the seminal vesicles and extraprostatic tissue may be recommended for patients considered to be at high risk.433,434 Patients with palpable T2 disease and biopsy-proven ­extracapsular extension are more often referred for EBRT than radical prostatectomy. In series that included intraoperative examination of lymph nodes by frozen section analysis, a limited number of patients presenting for radical prostatectomy may have been excluded from subsequent analyses on the discovery of positive lymph nodes during surgery. Retrospective surgical series also typically do not include patients who received adjuvant radiation therapy

TABLE 28-13. Outcome after Radical Prostatectomy Percent Showing No Biochemical Evidence of Disease

Study and Reference

Number of Patients

5 Years

10 Years

Trapasso et al, 1994431

601

69

47

Catalona et al, 1998288

1778

78

65

Dillioglugil et al, 611 1997152

78

76

Pound et al, 1997287

1623

80

68

Amling et al, 2000432

2782

76

60

for adverse pathologic features (e.g., positive margins), and this is another source of selection bias. A few studies in the PSA era have considered pretreatment PSA level, Gleason score, and T category in direct comparisons of radical prostatectomy and EBRT.165,299 The Kaplan-Meier freedom from biochemical failure curves for patients treated with radical prostatectomy or EBRT have been superimposable when patients are grouped according to favorable, intermediate, and unfavorable risk features. These findings are convincing evidence of the equivalence of radical prostatectomy and EBRT across the full range of prostate cancer, at least when disease is localized to the pelvis. The evidence also indicates that PRSIP should be limited to patients with favorable risk features, because the results of radical prostatectomy and EBRT have been found to be superior to those for PRSIP alone among men with intermediate- to high-risk disease characteristics.165,285

Surgery for Intermediate-Risk to High-Risk Disease Radical Prostatectomy Alone Patients with intermediate- to high-risk prostate cancer fare similarly when treated with radical prostatectomy or EBRT. D’Amico and colleagues165 and Kupelian and colleagues299 found that the freedom from biochemical failure rates at 5 years were approximately 60% for patients at intermediate risk and 30% for patients at high risk. Radiation doses in these studies were about 70 Gy, a dose that has since been shown to be inadequate for this patient group.166,200,316 There is every reason to believe that because the results after treatment with 70 Gy are similar to those after radical prostatectomy, higher radiation doses will be more efficacious than radical prostatectomy. Further evidence in support of EBRT is that the freedom from biochemical failure curves after radiation for patients treated in the PSA era seem to reach a plateau, suggesting a cure fraction. Although the literature is ­replete with examples of plateaus in Kaplan-Meier survival curves that fall after longer follow-up periods, the maintenance of a plateau with extended follow-up beyond 5 years is convincing. In all but one152 of the major radical prostatectomy series, freedom from biochemical failure rates have inexorably decreased between 5 and 10 years, with no hint of a plateau. Results of trials of radical prostatectomy152,287,288,431,432 (see Table 28-13) show a decrease between 5 and 10 years of 15% to 20% on average. These data are for all men studied, and the decrease is more substantial for those with intermediate- or high-risk disease features. For example, Amling and colleagues432 reported a decrease in freedom from biochemical failure survival rates from 76% at 5 years to 59% at 10 years for the entire study population; the corresponding rates were 82% and 68% for those with pathologic confirmation that disease was confined to the prostate. The hazard function continued to show a persistent and significant risk of failure at 10 years. The 1-year hazard rates at 9 and 10 years after surgery were on the order of 4 for the entire cohort and 5 to 10 for those with a Gleason score of 8 to 10, cT3 tumors, a PSA level of more than 10 ng/mL, or positive margins. The annual hazard rates changed little between 5 and 10 years. These rates seem to be higher than the annual hazard rates beyond

28  The Prostate

5 years reported for men with prostate cancer treated ­definitively with radiation therapy. A group at Fox Chase has described annual hazard rates of less than 1 between 5 and 10 years.435,436 Other EBRT series have reported a plateau in freedom from biochemical failure survival curves for patients at risk beyond 5 years. Longer follow-up is needed to establish more conclusively that EBRT results in a lower risk of relapse than radical prostatectomy when follow-up extends into the 10-year range.

Neoadjuvant Androgen Ablation plus Prostatectomy Among patients with palpable T2b to T2c disease, the incidence of extracapsular extension approaches 65%, and that of margin positivity approaches 50%.437-444 Although these rates may be declining as a consequence of stage migration, new strategies are needed to reduce the risk of failure ­after radical prostatectomy. To this end, several randomized

709

studies have investigated the effect of neoadjuvant AA given for 3 months before radical prostatectomy versus radical prostatectomy alone. Table 28-14437-445 summarizes the results of these trials, showing that the reduction in margin positivity was almost 50% when neoadjuvant AA was given for 3 months. Unfortunately, with longer follow-up, no difference was seen in freedom from biochemical failure rates between these groups (Table 28-15).438,441-448 Gleave and colleagues449 hypothesized that extending the duration of neoadjuvant AA to 8 months might affect long-term freedom from biochemical failure rates after radical prostatectomy because they found two slopes in the PSA response curve to AA, suggesting that the ­nadir of PSA was maximal at about 8 months. A follow-up report of patients treated in this manner was encouraging,450 and a phase III trial will investigate this strategy. Other investigators are examining the effect of prolonging neoadjuvant AA by comparing 3 months and 6 months of

TABLE 28-14.  Neoadjuvant Androgen Ablation plus Radical Prostatectomy and Margin Status Percent with Positive Margins Study and Reference

Disease Stage

Number of Patients

Hormones Only

Surgery Only

Hormones +   Surgery

Van Poppel et al, 1995437

T2b-T3

127

6 wk Estr

47

40

T2b

303

3 mo TAB

48

18*

Goldenberg et al, 1996439

T1b-T2c

213

12 wk CyA

65

28*

Labrie et al, 1997440

T1-T3

161

3 mo TAB

34

8*

T1-T2

114

3 mo TAB

36

17

T2-T3

354

3 mo TAB

46

26*

T1b-T3a

122

3 mo TAB

46

24*

Soloway et al,

Fair et al,

1996438

1997441

Witjes et al,

1997442 †

Aus et al, 1998443 *P

< .05.

†Benefit

in T2 disease was greater than in T3 disease. CyA, cyproterone acetate; Estr, estramustine; TAB, total androgen blockade. Modified from Pollack A, Zagars GK, Rosen I. Androgen ablation in addition to radiotherapy for prostate cancer: is there true benefit? Semin Radiat Oncol 1998;8:95-106; Pollack A. Does radiotherapy add anything to androgen ablation? In Taylor J (ed): American Society of Clinical Oncology Spring Educational Book. Baltimore: Lippincott Williams & Wilkins, 2000, pp 320-326.

TABLE 28-15. Neoadjuvant Androgen Ablation plus Radical Prostatectomy and Biochemical (Prostate-Specific Antigen) Failure Percent Showing Biochemical Evidence of Treatment Failure Study and Reference

Number of Patients

Disease Stage

Follow-Up Period

Hormonal Treatment

Surgery Only

Hormone + Surgery

Soloway et al, 1996438

263

T2b

≥12 mo

3 mo TAB

12

16

194

T1-T3

29 mo

3 mo TAB

11

16

215

T2-T3

15 mo

3 mo TAB

23

22

Fair et al,

1997441 *

Witjes et al, 1997442 1998443

122

T1b-T3a

38 mo

3 mo TAB

41

35

Klotz et al,

1999444

213

T1b-T2c

36 mo

3 mo CyA

30

40

Baert et al,

1998445

161

T2b-T3

40 mo

6 mo Estr

680

T1-T3

38 mo

≥ mo Mix

Aus et al,

Meyer et al, 1999446 *Pooled

Similar 36

29

data from phase II and III studies. CyA, cyproterone acetate; Estr, estramustine; Mix, mixed; TAB, total androgen blockade. Modified from Pollack A, Zagars GK, Rosen I. Androgen ablation in addition to radiotherapy for prostate cancer: is there true benefit? Semin Radiat Oncol 1998;8:95-106; Pollack A. Does radiotherapy add anything to androgen ablation? In Taylor J (ed): American Society of Clinical Oncology Spring Educational Book. Baltimore: Lippincott Williams & Wilkins, 2000, pp 320-326.

710

PART 8  Male Genital Tract

such treatment.451,452 These studies indicate that further reductions in margin positivity and tumor volume can be achieved by extending the duration of neoadjuvant AA from 3 to 6 months. Possible reasons for the lack of improvement in biochemical or clinical end points from the reduction in positive margins in these studies may be related to distant disease failures or variations in the pathologic assessment of margin status after neoadjuvant hormone therapy.

Adjuvant and Salvage Radiation Therapy Classification of Adjuvant and Salvage Treatment The division in the classification of adjuvant versus salvage radiation therapy after prostatectomy is based on serum PSA level. Adjuvant radiation therapy applies to patients who, during the first month after surgery, develop an ­undetectable PSA level. Salvage radiation therapy applies to situations in which PSA levels remain detectable after surgery or rise from undetectable levels at a later time. Although some may consider the administration of radiation therapy for a persistently elevated PSA level to be adjuvant treatment, these patients have a higher risk of regional and distant metastasis and perhaps a greater local residual tumor burden than patients with an undetectable PSA level after prostatectomy. The available data suggest that salvage radiation therapy for a persistently elevated PSA after prostatectomy is slightly less effective than salvage radiation therapy for delayed biochemical failure.453-456 However, these two populations are inherently different; patients undergoing salvage radiation therapy have already experienced treatment failure and may therefore represent a group with more aggressive disease. The decision to deliver adjuvant EBRT within 6 months of prostatectomy depends on the risk of local recurrence according to adverse pathologic features in the prostatectomy specimen. The hypothesis is that adverse findings are

associated with incomplete removal of the tumor, leaving microscopic residual cells behind. Because 1 cm3 of prostate cancer (109 cells) is thought to produce a serum PSA level of 3.5 ng/mL,61 most current PSA assays cannot detect 106 to 107 residual tumor cells.457 The correlation between high-risk pathologic features after prostatectomy and biochemical failure is incontrovertible.56,103,152,287,432

High-Risk Pathologic Features after Prostatectomy Increased rates of clinical or pathologic failure in patients with clinically organ-confined (cT1 to T2) prostate cancer have been associated with the high-risk pathologic features of extracapsular extension, Gleason score of more than 6, seminal vesicle involvement, margin positivity, and lymph node involvement. Of these factors, lymph node involvement is a strong predictor of distant spread and should be considered separately. Although seminal vesicle involvement also has been associated with a relatively high incidence of regional disease or distant metastasis, modifying factors exist.458 A subset of patients with seminal vesicle involvement has a reasonable prognosis; local adjuvant therapy may be beneficial for such patients. The frequency of seminal vesicle involvement and lymph node involvement ranges from 5% to 10% (Table 28-16).103,105,129,152,287,459-466 ­Extracapsular extension is seen in 32% to 61% of patients and ­ margin positivity in 15% to 43%. Stage migration in the PSA era has been considerable; reductions in the proportion of patients presenting with extracapsular extension,83,129,130 margin positivity,467 and lymph node ­positivity83,130 have been significant. A Gleason score of more than 6 is another modifying factor to be considered in assessing the need for adjuvant radiation therapy after radical prostatectomy. An isolated finding of a Gleason score of more than 6 is insufficient grounds for adjuvant radiation therapy. The decision of whether to deliver adjuvant radiation therapy is based

TABLE 28-16.  Adverse Pathologic Features in Prostatectomy Specimens from Patients with cT1 to T2 Disease Study and Reference

Number of Patients

Extracapsular Extension (%)

Seminal Vesicle Involvement (%)

Lymph Node Involvement (%)

Positive Margins (%)

Ohori et al, 1995459

478

43

12

5

16

Kupelian et al, 1996103

337

61

18

8

43

Epstein et al, 1996460

721

58

7

8

23

583

35

4

3

15

Lowe et al,

1997461 1997152

611

39

11

6

—

Pound et al, 1997287

1623

52

6

7

—

Bauer et al, 1998462

Dillioglugil et al,

Ramos et al,

378

57

—

—

—

1999105 *

1620

32

7

1

24

1999463

1395

47

7

3

33

Jhaveri et al, 1999129

731

54

—

—

—

Cheng et al, 2000464

339

35

14

6

24

2475

32

15

7

—

41%

9%

5%

24%

Gilliland et al,

Blute et al,

2000465

Weighted averages *T1

to T2b (T2c excluded). Modified from Pollack A, Smith LG. Adjuvant external beam radiotherapy postprostatectomy. In D’Amico AV, Carroll P, Kantoff PW (eds): Prostate Cancer. Philadelphia: Lippincott Williams & Wilkins, 2000.

28  The Prostate

­ rimarily on the presence of extracapsular extension p and margin positivity. The incidence of failure, however, is clearly higher among patients who have extracapsular extension and a Gleason score of more than 6.145,460,467 Moreover, failure rates correlate with the extent of extracapsular extension and margin positivity (focal versus ­extensive or single versus multiple) and are affected by the Gleason score.145,287,460,467 Another correlate of extracapsular extension and margin positivity is clinical stage. The 1992 AJCC staging system is a better discriminator of the risk of adverse pathologic findings than the 1997 staging system (see Table 28-2). For category cT2a disease (according to the 1992 system), Han and colleagues106 found that the rate of extracapsular ­extension was 47.1%, that of seminal vesicle involvement was 5.5%, and that of lymph node involvement was 3.5%. The corresponding rates for category cT2b disease were 54.1%, 7.8%, and 11.4%. Other investigators also have reported higher rates of pT3 disease in category cT2b cases than in cT2a tumors.83,103,105 In contrast, the differences between cT2b and cT2c tumor characteristics were not significant. As shown in Table 28-17,145,152,460,466-471 margin positivity is a much stronger determinant of biochemical failure at 5 and 10 years after radical prostatectomy than is extracapsular extension. Approximately one half of the ­patients with margin positivity will experience biochemical failure by 5 years. With longer follow-up, the incidence of failure is higher (see Table 28-17), but not all patients with margin positivity experience biochemical failure. Among the possible explanations for this are that (1) cells may ­extend to the resection margins but not beyond; (2) a small focus of cells remaining beyond the resection margins may eventually die; (3) the cells may remain dormant; and (4) the follow-up period, even at 10 years, may be too short. For cancer at almost every other anatomic site, surgeons enlist radiation oncologists to assist in reducing the risk of treatment failure when that risk approaches 50%. However, for prostate cancer, other concerns have been raised, not the least of which is that adjuvant radiation therapy has never been shown to prolong survival, perhaps because of the long natural history of the disease and deaths caused by

711

intercurrent illness. Another question is whether the timing of radiation therapy—adjuvant versus salvage—makes a difference. It is hard to imagine that leaving tumor behind in more than 50% of patients with high-risk disease after radical prostatectomy would not have an adverse effect on outcome if left untreated until relapse, particularly among younger men with prostate cancer.

Adjuvant Radiation Therapy for High-Risk ­Pathologic Features after Prostatectomy Studies done before the PSA era have shown that adjuvant radiation therapy after prostatectomy effectively ­ reduces local failure in the prostatic fossa.472-477 Differences in local control rates between irradiated and nonirradiated treatment groups are much greater at 10 years than at 5 years, attesting to the prolonged natural history of the disease. Anscher and colleagues457,475 found that the local control rate at 10 years was 92% for patients given adjuvant radiation and 60% for those not given adjuvant radiation. Most series have shown that this local control effect translates into an overall increase in disease freedom, but consistent significant reductions in distant metastasis or death rates have not been observed. The argument against adjuvant radiation therapy has centered on these findings and on the belief of some that salvage radiation therapy is as ­effective.478,479 The relationship between local control and distant ­metastasis (or survival) has never been adequately examined because studies have been small, retrospective, and grossly underpowered. If 30 of 100 patients in a cohort are destined to develop local failure in a 10-year period and the local failure rate after adjuvant radiation therapy is reduced to 10%, the potential improvement affects only 20 patients of the 100 studied. Even if a high percentage of those 20 men were to develop distant metastasis before dying of intercurrent disease, those men would be mixed in with patients who had distant metastasis without local failure. If 30% of the original cohort had distant disease at 10 years and if as many as a third of these cases were attributable to local failure, the numbers would be far too small to detect a difference statistically. Because almost 30,000 men die of

TABLE 28-17. Relationships of Extracapsular Extension and Margin Positivity to Freedom from Biochemical Failure Percent Freedom from Biochemical Failure 5 Years Study and Reference

Extracapsular Extension

Paulson, 1994145

—

Zietman et al, 1994469 Watson et al,

1996470

Epstein et al,

1996460,467

Dillioglugil et al, 1997152 Kupelian et al,

1997468

Grossfeld et al, *Confirmed

2000471

27 — 78* † 79 ≈ 43 —

10 Years

Positive Margins

Extracapsular Extension

Positive Margins

42

—

38

26

—

—

66

—

—

64

68

55

—

79

—

≈ 35

—

—

52

—

—

extracapsular extension. rate. Modified from Pollack A, Smith LG. Adjuvant external beam radiotherapy postprostatectomy. In D’Amico AV, Carroll P, Kantoff PW (eds): Prostate Cancer. Philadelphia: Lippincott Williams & Wilkins, 2000.

†Four-year

712

PART 8  Male Genital Tract

prostate cancer every year in the United States, it behooves us to reduce failure rates when possible, particularly among men for whom the life expectancy is at least 10 years, ­because the morbidity of adjuvant treatment is low. The results of adjuvant postoperative radiation therapy for patients with high-risk pathologic characteristics are provided in Table 28-18.454,480-487 The freedom from biochemical failure rate is about 70% on average, which seems to be higher than the 50% rate for patients treated with radical prostatectomy alone (see Table 28-17). Single­institution comparisons of radical prostatectomy alone versus radical prostatectomy plus adjuvant radiation therapy are summarized in Table 28-19.456,477,483,484,486 Two of these studies involved matched-pair analyses. In all but one

r­ eport, the addition of adjuvant radiation therapy significantly improved freedom from biochemical failure rates. The lack of a significant effect in the University of Southern California cohort486 may be related to the use of lower radiation doses (about 48 Gy); the other studies used doses higher than 60 Gy. Data on the dose requirements for adjuvant radiation therapy are limited. Valicenti and colleagues488 found evidence of a dose-response relationship among patients with pT3 N0 disease given adjuvant radiation. The 3-year freedom from biochemical failure rate for patients who ­received 61.2 Gy or less (n = 14) was 64%, compared with 90% (P = .015) for those who received more than 61.2 Gy (n = 38). Small numbers obviated multivariate analysis,

TABLE 28-18.  Freedom from Biochemical Failure after Adjuvant External Beam Radiation Therapy

Study and Reference Zietman et al, 1993480 Schild et al,

1996481 1996454

Number of Patients

Pathologic Tumor Stage

Positive Margins

Freedom from Biochemical Failure Rate

Follow-Up Period (mo)

Dose (Gy)

NS

60-64

64% (5 yr)

68*

T3 N0

94%

60

T3 N0

82%

32

62

57% (5 yr)

30

T2-T3c

100%

33

66-70

50% (4 yr)

Morris et al, 1997482

40

T3 N0

85%

31

60-62

88% (3 yr)

Valicenti et al, 1998483

15

T3 c N0

20%

≈ 38

64.8

86% (3 yr)

1999484

52

T3 N0

83%

39

64.8

82% (5 yr)

Coetzee et al,

Valicenti et al, Vicini et al,

1999485

Petrovich et al, 1999486 Nudell et al, 1999487 Leibovich et al,

2000477

38

T2c-T4

49

59.4

67% (5 yr)

201

T3 N0

41%

68

48

67% (5 yr)

36

T2-T3

97%

≈ 23

68

≈ 58% (5 yr)

T2 N0

100%†

29

63

76

—

88%

*Only

27 were confirmed as having undetectable prostate-specific antigen (PSA) levels. positive margin. NS, not stated. Modified from Pollack A, Smith LG. Adjuvant external beam radiotherapy postprostatectomy. In D’Amico AV, Carroll P, Kantoff PW (eds): Prostate Cancer. Philadelphia: Lippincott Williams & Wilkins, 2000.

†Single

TABLE 28-19.  Adjuvant External Beam Radiation Therapy after Prostatectomy: Comparative Studies Study and Reference Leibovich et al,

2000477 *

Petrovich et al, 1999486

Dose (Gy)

Freedom from Biochemical Failure Rate

P Value

76

None

59% (5 yr)

.005

76

63

88% (5 yr)

40

None

69% (5 yr)

48

68% (5 yr)

36

None

66% (3 yr)

36

64.8

93% (3 yr)

≈ 228

None

40% (5 yr)

≈ 60

60-67

90% (5 yr)

20

None

48% (3 yr)

15

64.8

86% (3 yr)

Number of Patients

201 Valicenti et al,

1998483 *

Schild, 1998456 Valicenti et al, *Matched-pair

1999484 †

NS NA .0003 .01

analysis. pT3c N0 patients. NA, not available; NS, not significant. Modified from Pollack A, Smith LG. Adjuvant external beam radiotherapy postprostatectomy. In D’Amico AV, Carroll P, Kantoff PW (eds): Prostate Cancer. Philadelphia: Lippincott Williams & Wilkins, 2000.

†For

28  The Prostate

making it unclear whether prognostic factors were evenly distributed between the two groups. Nonetheless, these ­results, combined with the lesser success of Petrovich and colleagues486 from using 48 Gy, indicate that doses higher than 60 Gy should be used until further information is available. Seminal vesicle involvement has been equated with distant spread. No doubt exists that involvement of the seminal vesicles is an adverse prognostic factor; patients with this feature are at increased risk for regional and distant metastasis, but they are also at increased risk for ­local failure.472,489 In the absence of concurrent lymph node involvement, subgroups of patients with seminal vesicle involvement exist who can benefit from adjuvant radiation therapy. Valicenti and colleagues483 directly addressed this question in a study of 35 men with pT3 cN0 disease who had undetectable PSA levels after prostatectomy. Of these patients, 15 were given adjuvant radiation therapy. The 3-year freedom from biochemical failure rates were 86% for the group given adjuvant radiation and 48% for the ­observation group.

Randomized Trials of Adjuvant Radiation Therapy Two large, randomized trials, RTOG/Southwest Oncology Group (SWOG) 8794 and EORTC 22911, have shown a benefit in progression-free survival from adjuvant radiation therapy after radical prostatectomy for men with pathologic stage T3 disease or positive surgical margins, or both (Table 28-20).490,491 In the SWOG trial,490 425 men with pathologic extraprostatic extension (including seminal vesicle involvement) or positive surgical margins, or both, after prostatectomy were randomly assigned to receive 60 to 64 Gy to the prostatic fossa or observation. Pelvic lymphadenectomy was required, and patients with nodepositive disease were not eligible, although an undetectable postoperative PSA level was not required. The primary end point was metastasis-free survival. At a median follow-up time of 10.6 years in the SWOG trial, a trend toward improvement in 10-year metastasisfree survival was evident in the adjuvant radiation therapy group (about 73% versus 63% in the observation group, hazard ratio = 0.75, P = .06), but no difference in overall survival was observed. Significant benefits were seen in terms

713

of clinical progression-free and biochemical ­ relapse–free survival and in freedom from salvage hormonal treatment for the adjuvant therapy group. The median metastasis-free survival time in the observation group was significantly longer than estimated (actual 13.2 years versus an estimated 6 years), which may be partially responsible for the lack of statistical significance. This longer-than-expected median metastasis-free survival time might have resulted from treatment crossover (approximately one third of patients in the observation group received pelvic radiation therapy), the effectiveness of salvage hormone therapy, or stage ­migration. The EORTC trial491 also examined patients with pT3 N0 M0 disease or positive surgical margins, or both, and ­randomly assigned them to observation or 60 Gy (prescribed to the isocenter) of adjuvant radiation to the ­surgical bed. There were no significant differences in ­metastasis-free or overall survival between the treatment groups, but as in the SWOG trial, significant benefits were seen in terms of clinical progression-free and PSA failure–free survival with the relatively low dose of 60 Gy. Locoregional failure was the most common site of clinical progression in both groups, and only 59 distant failures ­occurred. ­Approximately 22.5% of patients in the observation group ultimately received salvage pelvic radiation. In both studies, the radiation group had more mild to moderate late side effects, but the incidence of severe (grade 3 or higher) side effects was low (<5%) and, like the incidence of total urinary incontinence, was not different in the treatment groups. When pathologic factors were analyzed in the SWOG study, adjuvant radiation therapy seemed to confer benefits in biochemical and local control for men with positive surgical margins, extracapsular extension, seminal vesicle involvement, or combinations of these factors. Overall, the median time to biochemical relapse for patients with seminal vesicle involvement was significantly shorter (2.1 years) than for patients with positive margins or extracapsular ­extension (6.1 years).490 In the EORTC trial, a benefit from adjuvant radiation was evident for all patients, regardless of whether they had extracapsular extension, seminal vesicle involvement, or positive surgical margins.491 However, in a subsequent analysis of 552 men that included

TABLE 27-20. Randomized Trials of Adjuvant External Beam Radiation Therapy versus Observation after Prostatectomy ProgressionFree Survival Rates

Hazard Ratio (95% CI)

Freedom from Biochemical Relapse Rate

Hazard Ratio (95% CI)

60 to 64 Gy (n = 214)

≈ 44% (at 10 yr)

0.62 (0.46-0.82)

≈ 55% (at 10 yr)

0.43 (0.31-0.58)

Observation (n = 211)

≈ 70% (at 10 yr)

Study and Reference

Number of Patients

Treatment Groups

Thompson et al, 2006490

425

≈ 30% (at 10 yr) P < .001

(P = .001) Bolla et al, 2006491

CI, confidence interval.

1005

60 Gy (n = 502)

85% (at 5 yr)

0.61 (0.43-0.87)

74% (at 5 yr)

Observation (n=503)

77.5% (at 5 yr)

52.6% (at 5 yr)

(P < .001)

(P < .001)

0.48 (0.37-0.62)

714

PART 8  Male Genital Tract

c­ entralized pathology review, positive surgical margin status was the strongest predictor of long-term biochemical control among patients undergoing adjuvant radiation therapy, with a hazard ratio of 0.38 (95% confidence interval, 0.26 to 0.54) compared with a hazard ratio of 0.88 (95% confidence interval, 0.53 to 1.46) among patients with negative margins who received radiation therapy.492 The prognostic significance of postoperative PSA level before adjuvant radiation therapy has been established in prior retrospective studies and confirmed in these prospective trials. In the EORTC trial, patients who underwent adjuvant radiation had better rates of 5-year biochemical failure–free survival if their PSA level before radiation was 0.2 ng/mL or less (78.8%) than if the PSA level was more than 0.2 ng/mL (62.6%). A post hoc analysis of the SWOG trial493 also showed a difference in biochemical failure–free survival on the basis of PSA level before radiation. In that analysis, 5-year estimates of freedom from biochemical failure were 77% for patients with preradiation PSA levels of 0.2 ng/mL or less, 34% for those with levels between 0.2 and 1.0 ng/mL, and 0% for those with levels higher than 1.0 ng/mL.493 In the same study, patients who received initial adjuvant radiation therapy were found to have better PSA control rates than patients who received radiation therapy at the time of failure.493 This benefit was evident primarily among patients with postoperative PSA levels of less than 1.0 ng/mL. The prognostic significance of a lower PSA level after surgery but before radiation may be related to a decreased local tumor burden or a decreased risk of occult distant metastasis. A lower PSA level may also reflect the inherent biology of

the disease; several studies have indicated that not all patients with detectable PSA levels after prostatectomy experience subsequent increases.494,495 This observation has led some institutions to define PSA failure as a level of 0.4 ng/mL or higher after prostatectomy. Ideally, adjuvant radiation would be used for patients who have undetectable PSA levels after surgery but have high-risk pathologic features; however, the previous studies suggest that salvage therapy can still be successful in some proportion of men with detectable postoperative PSA levels. The decision to use postoperative radiation therapy for these patients should consider clinical and pathologic factors in addition to PSA level.

Salvage External Beam Radiation Therapy after Prostatectomy Salvage radiation therapy is given after radical prostatectomy for two distinct indications: a persistently elevated postoperative PSA level and a delayed increase in the PSA level occurring 6 months or more after surgery. Differences in prognosis between these groups have not been firmly established, mainly because of small patient numbers, ­relatively short follow-up periods, and imbalances in prognostic factors. Although some investigators consider the prognosis to be the same for those two groups,478 patients with a delay in the appearance of biochemical failure ­usually have had better outcomes.453-456 The success of salvage radiation therapy depends on various patient- and treatment-related factors. Presurgical characteristics such as PSA level, Gleason score, disease stage, and margin status seem to be less important than the PSA level before radiation. Table 28-21 shows the

TABLE 27-21. Salvage External Beam Radiation Therapy after Prostatectomy: Importance of Prostate-Specific Antigen Level before Radiation Study and Reference Wu et al,

1995496

Schild et al,

1996481

Morris et al, 1997482 Forman et al,

1997478

Zelefsky et al, 1997497 Rogers et al,

1998498 †

Anscher et al, 2000499 Pisansky et al, *Includes

2000500

Median Dose, Gy (range)

Number of Patients

61.2 (55-66)

53

64 (60-67) NS (60-64) 66 (50-74) 65 (38-70) 70 (61-70) 66 (55-70) 64 (54-72)

some patients given adjuvant radiation. had biopsy-confirmed local recurrence. NS, not stated; PSA, prostate-specific antigen.

†All

46 48 47 28 34 89 166

PSA Cut Points

Freedom from Biochemical Failure Rate

P Value

<2.5 ng/mL (n = 27)

52% (crude)

.001

≥2.5 ng/mL (n = 26)

8% (crude)

≤1.1 ng/mL (NS)

76% (3 yr)

>1.1 ng/mL (NS)

26% (3 yr)

≤1.7 ng/mL (n = 24)

66% (3 yr)

>1.7 ng/mL (n = 24)

29% (3 yr)

≤2 ng/mL (n = 29)

44% (4 yr)

>2 ng/mL (n = 18)

13% (4 yr)

≤1.0 ng/mL (n = 25)*

75% (2 yr)

>1.0 ng/mL (n = 13)

26% (2 yr)

≤4 ng/mL (NS)

60% (3.5 yr)

>4 ng/mL (NS)

0% (3.5 yr)

<2.5 ng/mL (NS)

4.8 yr (median)

≥2.5 ng/mL (NS)

1.5 yr (median)

≤1.0 ng/mL (n = 96)

53% (5 yr)

>1.0 ng/mL (n = 70)

38% (5 yr)

.02 .03 .005 .008 .03 .03 .08

28  The Prostate

r­ elationship between the preradiation PSA level and biochemical failure rates, indicating that the earlier treatment is begun, the greater the chance for control. 478,481,482,496-500 Salvage therapy is successful in more than 50% of patients if the PSA level before radiation is low (<1 ng/mL), whereas the success rate is less than 30% if the PSA level is high. The relationship between the preradiation PSA level and biochemical failure is probably a continuous function. Younger men with a long life expectancy should be treated as soon as biochemical failure and the absence of metastasis are established. Further delay in treatment allows time for the disease to progress. Ideally, salvage radiation therapy should be considered for cases in which biochemical failure is thought to reflect residual disease in the surgical bed. A history of a positive margin, extracapsular extension, or seminal vesicle involvement provides a rationale for local treatment. Negative ­results from biopsies of the urethrovesical anastomosis and prostatic bed should not discourage plans for radiation therapy management, because the volume of residual disease may be small. The conundrum arises when there is no identifiable reason to administer local therapy (e.g., organconfined disease with negative margins) but the PSA level is rising slowly after a delay of several years. In cases such as

715

this, locally persistent disease may still be present because pathologic assessment of margins is far from perfect. This dilemma should be discussed in detail with the patient, with consideration of the advantages and disadvantages of salvage radiation therapy. Radiation dose has been shown to be important in several studies of salvage radiation therapy.456,481,488,499 The best results are obtained when the surgical bed receives more than 65 Gy. Deciding who should be treated with salvage radiation therapy can be a perplexing clinical problem, especially because a rising PSA level can precede radiographically or clinically detectable disease by several years. Establishing the site of failure in such cases can be difficult; even a positive biopsy result in the surgical bed does not eliminate the possibility of distant failure in addition to local disease. These cases require clinicians to consider all available factors, not the least of which is the estimated longevity of the patient to see a clinical benefit. The ideal candidate for salvage radiation therapy has (1) longevity expected to be sufficient to see a benefit (e.g., relatively young, good performance status, few comorbid conditions), (2) favorable clinical presentation at initial diagnosis (e.g., low PSA ­level, low Gleason score), (3) pathologic features ­ suggesting

Pre-RT PSA ≤2.0

>2.0

4-year PFP (Cl) 52% (49–56) 980 patients

4-year PFP (Cl) 19% (15–24) 346 patients

Gleason score 4–7

8–10

4-year PFP (Cl) 58% (54–62) 689 patients

4-year PFP (Cl) 27% (19–35) 151 patients

Surgical margins

Surgical margins

Positive

Negative

Positive

4-year PFP (Cl) 61% (55–67) 360 patients

4-year PFP (Cl) 55% (49–61) 314 patients

4-year PFP (Cl) 33% (22–43) 151 patients

PSADT

PSADT

PSADT

>10

≤10

>10

≤10

>10

Negative 4-year PFP (Cl) 18% (7–29) 63 patients

≤10

4-year PFP (Cl) 4-year PFP (Cl) 4-year PFP (Cl) 4-year PFP (Cl) 4-year PFP (Cl) 4-year PFP (Cl) 50% (16–86) 30% (17–44) 61% (50–72) 48% (40–56) 69% (59–79) 57% (49–66) 14 patients 50 patients 96 patients 175 patients 110 patients 192 patients

FIGURE 28-20.  Four-year progression-free probability (PFP) after salvage radiation therapy for 1326 men stratified by prostatespecific antigen (PSA) levels, Gleason score, surgical margins, and PSA doubling time (PSADT, in months) before radiation therapy (RT). None of these men underwent androgen deprivation before or during radiation therapy. CI, 95% confidence interval. (From Stephenson AJ, Scardino PT, Kattan MW, et al. Predicting the outcome of salvage radiation therapy for recurrent prostate cancer after radical prostatectomy. J Clin Oncol 2007;25:2035-2041.)

716

PART 8  Male Genital Tract

l­ocal (as opposed to distant) recurrence (e.g., positive surgical margins, extraprostatic extension), and (4) favorable postoperative PSA kinetics (e.g., long interval to PSA failure, long PSA doubling time, low PSA level at ­radiation).501,502,503 Stephenson and colleagues504 published an analysis of outcomes after salvage radiation therapy using data from 17 tertiary care centers (Fig. 28-20). They found that men with a preradiation PSA level of 2.0 ng/mL or less, a Gleason score of 7 or less, positive surgical margins, and a PSA doubling time of more than 10 months had an estimated 4-year progression-free probability of 69% after salvage radiation therapy. In contrast, men with a preradiation PSA level of more than 2.0 ng/mL, a Gleason score of 8 to 10, and negative surgical margins had an estimated ­progressionfree probability of only 18%. Before applying these findings and the accompanying nomogram for predicting progression after salvage radiation, the retrospective nature of the analysis must be understood. Although these tools are helpful, they must be considered in the appropriate clinical context (e.g., an oncologist might be more willing to treat a patient with less favorable disease features if the patient were relatively young and healthy).

Adjuvant versus Salvage External Beam Radiation Therapy Younger men with a long life expectancy and high-risk pathologic features after radical prostatectomy should be treated with adjuvant radiation therapy until further data can be generated to support the alternative practice of watchful waiting and salvage radiation therapy at a later time. The findings presented in Table 28-22 show that salvage radiation therapy produces freedom from biochemical failure rates that are roughly one half or less

than one half of those produced by adjuvant radiation therapy. 453,454,456,482,485,488 However, these single-institution analyses are fraught with problems because they are retrospective, involve small numbers of patients with short follow-up, and have imbalances in the prognostic features of the populations being compared. The physician must also consider that perhaps 50% of men given adjuvant treatment may never experience relapse. Nonetheless, without conclusive evidence that salvage treatment is equivalent to ­adjuvant treatment, adjuvant radiation therapy is preferred. Earlier treatment has consistently produced better results at lower doses for diseases at other anatomic sites. The side effects of adjuvant radiation therapy for prostate cancer are usually mild to moderate, with few severe complications. The side effects can be reduced further by using conformal radiation therapy.

Cryotherapy TRUS-guided cryotherapy is briefly mentioned here ­because of a resurgence in its use since its introduction in the late 1980s. The procedure involves the percutaneous, transperineal placement of multiple cryoprobes and thermosensors to monitor development of the ice ball. Single and double freeze-thaw cycles are used, with the latter more effectively ablating prostate tumors. The urethra is commonly warmed to prevent urethral sloughing. ­Improvements in the technique have included the use of more cryprobes,505 miniaturization of the freezing probes,506 and the use of argon gas.507 The results of cryosurgery have been preliminary for the most part, with relatively short follow-up periods. Published series505,507-510 indicate that long-term failure rates after cryosurgery are higher than those after radical prostatectomy or radiation therapy.

TABLE 27-22. Adjuvant versus Salvage External Beam Radiation Therapy after Prostatectomy: Comparative Studies Study and Reference

Number of Patients

Treatment Type*

Median Dose, Gy (range)

Freedom from Biochemical Failure Rate

P Value

McCarthy et al, 1994453

64

Adj (n = 27)

NS (< 60-65)

66 (3 yr)

.008

Salv (n = 22)

NS (< 60-65)

68 (3 yr)

NS (< 60-65)

32 (3 yr)

Salv (n = Coetzee et al,

1996454

Morris et al, 1997482 Valicenti et al,

1998488

Schild, 1998456 Vicini et al, *Number

1999485

45 88 79 NS 61

Adj (n = 30)

NS (66-70)

50 (5 yr)

Salv (n = 15)†

NS (66-70)

0 (5 yr)

Adj (n = 40)

NS (60-64)

81 (3 yr)

Salv (n = 48)

NS (60-64)

48 (3 yr)

Adj (n = 52)

65 (55-70)

93 (3 yr)

Salv (n = 27)

65 (55-70)

44 (3 yr)

Adj (NS)

62 (57-68)

90 (5 yr)

Salv (n = 55)

64 (60-68)

49 (5 yr)

Adj (n = 38)

59.4 (50-61)

67 (5 yr)

Salv (n = 23)

61 (59-68)

16 (5 yr)

(n) per treatment group. elevated prostate-specific antigen level. Adj, adjuvant; NS, not stated; Salv, salvage.

†Persistently

15)†

NS .0006 <.0001 NS <.001

28  The Prostate

REFERENCES    1. McNeal JE. The zonal anatomy of the prostate. Prostate 1981;2:35-49.    2. McNeal JE, Villers AA, Redwine EA, et al. Capsular penetration in prostate cancer. Significance for natural history and treatment. Am J Surg Pathol 1990;14:240-247.    3. Walsh PC, Donker PJ. Impotence following radical prostatectomy: insight into etiology and prevention. J Urol 1982;128: 492-497.    4. Willett CG, Ooi CJ, Zietman AL, et al. Acute and late toxicity of patients with inflammatory bowel disease undergoing irradiation for abdominal and pelvic neoplasms. Int J Radiat Oncol Biol Phys 2000;46:995-998.    5. Zietman AL, Zehr EM, Shipley WU. The long-term effect on PSA values of incidental prostatic irradiation in patients with pelvic malignancies other than prostate cancer. Int J Radiat ­Oncol Biol Phys 1999;43:715-718.    6. Bostwick DG, Egbert BM, Fajardo LF. Radiation injury of the normal and neoplastic prostate. Am J Surg Pathol 1982;6: 541-551.    7. Sheaff MT, Baithun SI. Effects of radiation on the normal prostate gland. Histopathology 1997;30:341-348.    8. Gaudin PB, Zelefsky MJ, Leibel SA, et al. Histopathologic effects of three-dimensional conformal external beam radiation therapy on benign and malignant prostate tissues. Am J Surg Pathol 1999;23:1021-1031.    9. Fajardo LF, Berthrong M, Anderson RE. Prostate. In Radiation Pathology. New York: Oxford University Press, 2001, pp 319-325.   10. Seymore CH, el-Mahdi AM, Schellhammer PF. The effect of prior transurethral resection of the prostate on postradiation urethral strictures and bladder neck contractures. Int J Radiat Oncol Biol Phys 1986;12:1597-1600.   11. Sandhu AS, Zelefsky MJ, Lee HJ, et al. Long-term urinary toxicity after 3-dimensional conformal radiotherapy for prostate cancer in patients with prior history of transurethral resection. Int J Radiat Oncol Biol Phys 2000;48:643-647.   12. Liu M, Pickles T, Berthelet E, et al. Urinary incontinence in prostate cancer patients treated with external beam radiotherapy. Radiother Oncol 2005;74:197-201.   13. Cox JD, Stetz J, Pajak TF. Toxicity criteria of the Radiation Therapy Oncology Group (RTOG) and the European Organization for Research and Treatment of Cancer (EORTC). Int J Radiat Oncol Biol Phys 1995;31:1341-1346.   14. Hanlon AL, Schultheiss TE, Hunt MA, et al. Chronic rectal bleeding after high-dose conformal treatment of prostate cancer warrants modification of existing morbidity scales. Int J Radiat Oncol Biol Phys 1997;38:59-63.   15. Storey MR, Pollack A, Zagars GK, et al. Complications from dose escalation in prostate cancer: preliminary results of a randomized trial. Int J Radiat Oncol Biol Phys 2000;48:635-642.   16. Nguyen LN, Pollack A, Zagars GK. Late effects after radiotherapy for prostate cancer in a randomized dose-response study: results of a self-assessment questionnaire. Urology 1998;51: 991-997.   17. Litwin MS, Flanders SC, Pasta DJ, et al. Sexual function and bother after radical prostatectomy or radiation for prostate cancer: multivariate quality-of-life analysis from CaPSURE. Cancer of the Prostate Strategic Urologic Research Endeavor. Urology 1999;54:503-508.   18. Wilder RB, Chou RH, Ryu JK, et al. Potency preservation after three-dimensional conformal radiotherapy for prostate cancer: preliminary results. Am J Clin Oncol 2000;23:330-333.   19. American Cancer Society. Cancer Facts & Figures 2008. Atlanta: American Cancer Society, 2008.   20. Oesterling JE, Martin SK, Bergstralh EJ, et al. The use of ­prostate-specific antigen in staging patients with newly diagnosed prostate cancer. JAMA 1993;269:57-60.   21. Oesterling JE, Jacobsen SJ, Chute CG, et al. Serum prostatespecific antigen in a community-based population of healthy men. Establishment of age-specific reference ranges. JAMA 1993;270:860-864.

717

  22. Catalona WJ, Partin AW, Slawin KM, et al. Use of the percentage of free prostate-specific antigen to enhance differentiation of prostate cancer from benign prostatic disease: a prospective multicenter clinical trial. JAMA 1998;279:1542-1547.   23. Morgan TO, Jacobsen SJ, McCarthy WF, et al. Age-specific reference ranges for prostate-specific antigen in black men. N Engl J Med 1996;335:304-310.   24. Babaian RJ, Toi A, Kamoi K, et al. A comparative analysis of sextant and an extended 11-core multisite directed biopsy strategy. J Urol 2000;163:152-157.   25. Potosky AL, Miller BA, Albertsen PC. The role of increasing detection in the rising incidence of prostate cancer. JAMA 1995;273:548-552.   26. Merrill RM, Potosky AL, Feuer EJ. Changing trends in U.S. prostate cancer incidence rates. J Natl Cancer Inst 1996;88: 1683-1685.   27. Andrews GS. Latent carcinoma of the prostate. J Clin Pathol 1949;2:197-208.   28. Edwards CN, Steinthorsson E, Nicholson D. An autopsy study of latent prostatic cancer. Cancer 1953;6:531-554.   29. Franks LM. Latent carcinoma of the prostate. J Pathol Bacteriol 1954;68:603-616.   30. Breslow N, Chan CW, Dhom Q, et al. Latent carcinoma of the prostate at autopsy in seven areas. Int J Cancer 1977; 20:680-688.   31. Sakr WA, Haas GP, Cassin BF, et al. The frequency of carcinoma and intraepithelial neoplasia of the prostate in young male patients. J Urol 1993;150:379-385.   32. Sakr WA, Grignon DJ, Crissman JD, et al. High grade prostatic intraepithelial neoplasia (HGPIN) and prostatic adenocarcinoma between the ages of 20-69: an autopsy study of 249 cases. In Vivo 1994;8:439-444.   33. Steinberg GD, Carter BS, Beaty TH, et al. Family history and risk of prostate cancer. Prostate 1990;17:337-347.   34. Carter BS, Beaty TH, Steinberg GD, et al. Mendelian inheritance of familial prostate cancer. Proc Natl Acad Sci U S A 1992;89:3367-3371.   35. Shibata A, Whittemore AS. Genetic predisposition to prostate cancer: possible explanation for ethnic differences in risk. Prostate 1997;32:65-72.   36. Xu J, Meyers DA, Freije D, et al. Evidence for a prostate cancer susceptibility locus on the X chromosome. Nat Genet 1998;20:175-179.   37. Smith JR, Freije D, Carpten JD, et al. Major susceptibility locus for prostate cancer on chromosome 1 suggested by a genomewide search. Science 1996;274:1371-1374.   38. Lehrer S, Fodor F, Stock RG, et al. Absence of 185delAG mutation of the BRCA1 gene and 6174delT mutation of the BRCA2 gene in Ashkenazi Jewish men with prostate cancer. Br J Cancer 1998;78:771-773.   39. Bosland MC. The etiopathogenesis of prostate cancer with special reference to environmental factors. Adv Cancer Res 1988;51:1-106.   40. Fair WR, Fleshner NE, Heston W. Cancer of the prostate: a nutritional disease? Urology 1997;50:840-848.   41. Dorgan JF, Judd JT, Longcope C, et al. Effects of dietary fat and fiber on plasma and urine androgens and estrogens in men: a controlled feeding study. Am J Clin Nutr 1996;64:850-855.   42. Urisman A, Molinaro R, Fischer N, et al. Identification of a novel Gammaretrovirus in prostate tumors of patients homozygous for R462QRNASEL variant. PLoS Pathogens 2006;2:211-225.   43. Kyprianou N, Isaacs JT. Activation of programmed cell death in the rat ventral prostate after castration. Endocrinology 1988;122:552-562.   44. English HF, Kyprianou N, Isaacs JT. Relationship between DNA fragmentation and apoptosis in the programmed cell death in the rat prostate following castration. Prostate 1989;15:233-250.   45. Ellis L, Nyborg H. Racial/ethnic variations in male testosterone levels: a probable contributor to group differences in health. Steroids 1992;57:72-75.   46. Pettaway CA. Racial differences in the androgen/androgen receptor pathway in prostate cancer. J Natl Med Assoc 1999;91: 653-660.

718

PART 8  Male Genital Tract

  47. Ross RK, Bernstein L, Lobo RA, et al. 5-Alpha-reductase activity and risk of prostate cancer among Japanese and U.S. white and black males. Lancet 1992;339:887-889.   48. Noble RL. Production of Nb rat carcinoma of the dorsal prostate and response of estrogen-dependent transplants to sex hormones and tamoxifen. Cancer Res 1980;40:3547-3550.   49. Pollard M, Luckert PH, Schmidt MA. Induction of prostate adenocarcinomas in Lobund Wister rats by testosterone. Prostate 1982;3:563-568.   50. Bosland MC. Chemical and hormonal induction of prostate cancer in animal models. Urol Oncol 1996;2:103-110.   51. Waalkes MP, Rehm S, Perantoni AO, et al. Cadmium exposure in rats and tumours of the prostate. In Nordberg GF, ­ Herber RFM, Alessio L (eds): Cadmium in the Human Environment: Toxicity and Carcinogenicity. Lyon, France: IARC, 1992, pp 391-400.   52. Takizawa S, Hirose F. Role of testosterone in the development of radiation-induced prostate carcinoma in rats. Gann 1978;69: 723-726.   53. McNeal JE, Bostwick DG, Kindrachuk RA, et al. Patterns of progression in prostate cancer. Lancet 1986;1:60-63.   54. Stamey TA, Freiha FS, McNeal JE, et al. Localized prostate cancer. Relationship of tumor volume to clinical significance for treatment of prostate cancer. Cancer 1993;71:933-938.   55. Chan LW, Stamey TA. Calculating prostate cancer volume preoperatively: the D’Amico equation and some other observations. J Urol 1998;159:1998-2003.   56. Stamey TA, McNeal JE, Yemoto CM, et al. Biological determinants of cancer progression in men with prostate cancer. JAMA 1999;281:1395-1400.   57. Lepor H, Wang B, Shapiro E. Relationship between prostatic epithelial volume and serum prostate-specific antigen levels. Urology 1994;44:199-205.   58. Marks LS, Treiger B, Dorey FJ, et al. Morphometry of the prostate. I. Distribution of tissue components in hyperplastic glands. Urology 1994;44:486-492.   59. Collins GN, Lee RJ, McKelvie GB, et al. Relationship between prostate specific antigen, prostate volume and age in the benign prostate. Br J Urol 1993;71:445-450.   60. Aihara M, Lebovitz RM, Wheeler TM, et al. Prostate specific antigen and Gleason grade: an immunohistochemical study of prostate cancer. J Urol 1994;151:1558-1564.   61. Stamey TA, Kabalin JN, McNeal JE, et al. Prostate-specific antigen in the diagnosis and treatment of adenocarcinoma of the prostate. II. Radical prostatectomy treated patients. J Urol 1989;141:1076-1083.   62. Weinrich MC, Jacobsen SJ, Weinrich SP, et al. Reference ranges for serum prostate-specific antigen in black and white men without cancer. Urology 1998;52:967-973.   63. Catalona WJ, Partin AW, Slawin KM, et al. Percentage of free PSA in black versus white men for detection and staging of prostate cancer: a prospective multicenter clinical trial. Urology 2000;55:372-376.   64. Catalona WJ, Southwick PC, Slawin KM, et al. Comparison of percent free PSA, PSA density, and age-specific PSA cutoffs for prostate cancer detection and staging. Urology 2000;56: 255-260.   65. Fowler JE Jr, Bigler SA, Miles D, et al. Predictors of first repeat biopsy cancer detection with suspected local stage prostate cancer. J Urol 2000;163:813-818.   66. Carter HB, Morrell CH, Pearson JD, et al. Estimation of prostatic growth using serial prostate-specific antigen measurements in men with and without prostate disease. Cancer Res 1992;52:3323-3328.   67. Chen ME, Troncoso P, Tang K, et al. Comparison of prostate biopsy schemes by computer simulation. Urology 1999;53: 951-960.   68. Egevad L, Frimmel H, Norberg M, et al. Three-dimensional ­computer reconstruction of prostate cancer from radical prostatectomy specimens: evaluation of the model by core biopsy simulation. Urology 1999;53:192-198.

  69. Presti JC Jr, Chang JJ, Bhargava V, et al. The optimal systematic prostate biopsy scheme should include 8 rather than 6 biopsies: results of a prospective clinical trial. J Urol 2000;163:163-166.   70. Ravery V, Goldblatt L, Royer B, et al. Extensive biopsy protocol improves the detection rate of prostate cancer. J Urol 2000;164:393-396.   71. Sufrin G, Gaeta J, Staubitz WJ, et al. Endometrial carcinoma of prostate. Urology 1986;27:18-23.   72. Brinker DA, Potter SR, Epstein JI. Ductal adenocarcinoma of the prostate diagnosed on needle biopsy: correlation with clinical and radical prostatectomy findings and progression. Am J Surg Pathol 1999;23:1471-1479.   73. Bock BJ, Bostwick DG. Does prostatic ductal adenocarcinoma exist? Am J Surg Pathol 1999;23:781-785.   74. di Sant’Agnese PA, de Mesy Jensen KL, Churukian CJ, et al. ­Human prostatic endocrine-paracrine (APUD) cells. Distributional analysis with a comparison of serotonin and neuron­specific enolase immunoreactivity and silver stains. Arch Pathol Lab Med 1985;109:607-612.   75. Ro JY, Tetu B, Ayala AG, et al. Small cell carcinoma of the prostate. Part II. Immunohistochemical and electron microscopic study of 18 cases. Cancer 1987;59:977-982.   76. Oesterling JE, Hauzeur CG, Farrow GM. Small cell anaplastic carcinoma of the prostate: a clinical, pathological and immunohistological study of 27 patients. J Urol 1992;147:804-807.   77. Randolph TL, Amin MB, Ro JY, et al. Histologic variants of ­adenocarcinoma and other carcinomas of prostate: pathologic criteria and clinical significance. Mod Pathol 1997;10:612-629.   78. Gleason DF. Classification of prostatic carcinomas. Cancer Chemother Rep 1966;50:125-128.   79. Gleason DF. Histologic grading of prostate cancer: a perspective. Hum Pathol 1992;23:273-279.   80. Mostofi FK, Sesterhenn IA, Davis CJ Jr. A pathologist’s view of prostatic carcinoma. Cancer 1993;71:906-932.   81. Zagars GK, Ayala AG, von Eschenbach AC, et al. The prognostic importance of Gleason grade in prostatic adenocarcinoma: a long-term follow-up study of 648 patients treated with radiation therapy. Int J Radiat Oncol Biol Phys 1995;31:237-245.   82. Chan TY, Partin AW, Walsh PC, et al. Prognostic significance of Gleason score 3+4 versus Gleason score 4+3 tumor at radical prostatectomy. Urology 2000;56:823-827.   83. Stamey TA, Sozen TS, Yemoto CM, et al. Classification of localized untreated prostate cancer based on 791 men treated only with radical prostatectomy: common ground for therapeutic trials and TNM subgroups. J Urol 1998;159:2009-2012.   84. Patel AA, Chen MH, Renshaw AA, et al. PSA failure following definitive treatment of prostate cancer having biopsy Gleason score 7 with tertiary grade 5. JAMA 2007;298:1533-1538.   85. Whitmore WF. Hormone therapy in prostatic cancer. Am J Med 1956;21:697-713.   86. Jewett HJ. The present status of radical prostatectomy for stages A and B prostatic cancer. Urol Clin North Am 1975;2:105-124.   87. Bears OH, Henson DE, Hutter RVP, et al (eds). AJCC Manual for Staging of Cancer, 4th ed. Philadelphia: JB Lippincott, 1997, pp 181-186.   88. Rifkin MD, Zerhouni EA, Gatsonis CA, et al. Comparison of magnetic resonance imaging and ultrasonography in staging early prostate cancer: results of a multi-institutional cooperative trial. N Engl J Med 1990;323:621-626.   89. Smith JA, Scardino PT, Resnick MI, et al. Transrectal ultrasound versus digital rectal examination for the staging of carcinoma of the prostate: results of a prospective, multi-institutional trial. J Urol 1997;157:902-906.   90. Shinohara K, Wheeler TM, Scardino PT. The appearance of prostate cancer on transrectal ultrasonography: correlation of imaging and pathological examinations. J Urol 1989;142:76-82.   91. Pontes JE, Eisenkraft S, Watanabe H. Preoperative evaluation of localized prostatic carcinoma by transrectal ultrasonography. J Urol 1985;134:289-291.   92. Perrapato SD, Carothers GG, Maatman TJ, et al. Comparing clinical staging plus transrectal ultrasound with surgical-­pathologic staging of prostate cancer. Urology 1989;33:103-105.

28  The Prostate   93. Hamper UM, Sheth S, Walsh PC, et al. Capsular transgression of prostatic carcinoma: evaluation with transrectal US with pathologic correlation. Radiology 1991;178:791-795.   94. Vijverberg PM, Giessen MA, Kurth KH, et al. Is preoperative transrectal ultrasonography of value in localized prostatic ­carcinoma? A blind comparative study between preoperative transrectal ultrasonography and the histopathological radical prostatectomy specimen. Eur J Surg Oncol 1992;18:449-455.   95. Vapnek JM, Hricak H, Shinohara K, et al. Staging accuracy of magnetic resonance imaging versus transrectal ultrasound in stages A and B prostatic cancer. Urol Int 1994;53:191-195.   96. Ohori M, Egawa S, Shinohara K, et al. Detection of microscopic extracapsular extension prior to radical prostatectomy for clinically localized prostate cancer. Br J Urol 1994;74:72-79.   97. Pinover WH, Hanlon A, Lee WR, et al. Prostate carcinoma ­patients upstaged by imaging and treated with irradiation. Cancer 1996;77:1334-1341.   98. Liebross RH, Pollack A, Lankford SP, et al. Transrectal ultrasound for staging prostate cancer prior to radiation therapy: an evaluation based on disease outcome. Cancer 1999;85:1577-1585.   99. Jhaveri FM, Zippe CD, Klein EA, et al. Biochemical failure does not predict overall survival after radical prostatectomy for localized prostate cancer: 10-year results. Urology 1999;54:884-890. 100. Sandler HM, Dunn RL, McLaughlin PW, et al. Overall survival after prostate-specific-antigen–detected recurrence following conformal radiation therapy. Int J Radiat Oncol Biol Phys 2000;48:629-633. 101. Zagars GK, Pollack A, von Eschenbach AC. Prognostic factors for clinically localized prostate carcinoma: analysis of 938 patients irradiated in the prostate specific antigen era. Cancer 1997;79:1370-1380. 102. Pollack A, Zagars GK. Pretreatment prognostic factors for prostate cancer patients treated with external beam radiotherapy. In Greco C, Zelefsky MJ, (eds): Radiotherapy of Prostate Cancer. Singapore: Harwood Academic Publishers, 2000, pp 113-128. 103. Kupelian P, Katcher J, Levin H, et al. Correlation of clinical and pathologic factors with rising prostate-specific antigen profiles after radical prostatectomy alone for clinically localized prostate cancer. Urology 1996;48:249-260. 104. Iyer RV, Hanlon AL, Pinover WH, et al. Outcome evaluation of the 1997 American Joint Committee on Cancer Staging System for prostate carcinoma treated by radiation therapy. Cancer 1999;85:1816-1821. 105. Ramos CG, Carvalhal GF, Smith DS, et al. Clinical and pathological characteristics, and recurrence rates of stage T1C versus T2A or T2B prostate cancer. J Urol 1999;161:1212-1215. 106. Han M, Walsh PC, Partin AW, et al. Ability of the 1992 and 1997 American Joint Committee on Cancer staging systems for prostate cancer to predict progression-free survival after radical prostatectomy for stage T2 disease. J Urol 2000;164:89-92. 107. Greene FL, Page DL, Fleming ID, et al (eds): AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 309-316. 108. D’Amico AV, Whittington R, Malkowicz SB, et al. Combination of the preoperative PSA level, biopsy Gleason score, percentage of positive biopsies, and MRI-T-stage to predict early PSA failure in men with clinically localized prostate cancer. Urology 2000;55:572-577. 109. D’Amico AV, Whittington R, Malkowicz SB, et al. Endorectal magnetic resonance imaging as a predictor of biochemical outcome after radical prostatectomy in men with clinically localized prostate cancer. J Urol 2000;164:759-763. 110. Debois M, Oyen R, Maes F, et al. The contribution of magnetic resonance imaging to the three-dimensional treatment planning of localized prostate cancer. Int J Radiat Oncol Biol Phys 1999;45:857-865. 111. Perrotti M, Han KR, Epstein RE, et al. Prospective evaluation of endorectal magnetic resonance imaging to detect tumor foci in men with prior negative prostatic biopsy: a pilot study. J Urol 1999;162:1314-1317. 112. Scheidler J, Hricak H, Vigneron DB, et al. Prostate cancer: localization with three-dimensional proton MR spectroscopic ­imaging—clinicopathologic study. Radiology 1999;213:473-480.

719

113. Yu KK, Hricak H, Alagappan R, et al. Detection of extracapsular extension of prostate carcinoma with endorectal and phasedarray coil MR imaging: multivariate feature analysis. Radiology 1997;202:697-702. 114. Yu KK, Scheidler J, Hricak H, et al. Prostate cancer: prediction of extracapsular extension with endorectal MR imaging and three-dimensional proton MR spectroscopic imaging. Radiology 1999;213:481-488. 115. Koutcher JA, Zakian K, Hricak H. Magnetic resonance spectroscopic studies of the prostate. Mol Urol 2000;4:143-152. 116. Pickett B, Vigneault E, Kurhanewicz J, et al. Static field intensity modulation to treat a dominant intra-prostatic lesion to 90 Gy compared to seven field 3-dimensional radiotherapy. Int J Radiat Oncol Biol Phys 1999;44:921-929. 117. Ling CC, Humm J, Larson S, et al. Towards multidimensional radiotherapy (MD-CRT): biological imaging and biological conformality. Int J Radiat Oncol Biol Phys 2000;47:551-560. 118. Ravery V, Boccon-Gibod LA, Dauge-Geffroy MC, et al. Systematic biopsies accurately predict extracapsular extension of prostate cancer and persistent/recurrent detectable PSA after radical prostatectomy. Urology 1994;44:371-376. 119. Ravery V, Chastang C, Toublanc M, et al. Is the percentage of cancer biopsy cores predictive of extracapsular disease in T1-T2 prostate carcinoma? Cancer 1996;78:1079-1084. 120. Borirakchanyavat S, Bhargava V, Shinohara K, et al. Systematic sextant biopsies in the prediction of extracapsular extension at radical prostatectomy. Urology 1997;50:373-378. 121. D’Amico AV, Whittington R, Malkowicz SB, et al. Clinical utility of the percentage of positive prostate biopsies in defining biochemical outcome after radical prostatectomy for patients with clinically localized prostate cancer. J Clin Oncol 2000;18: 1164-1172. 122. D’Amico AV, Whittington R, Malkowicz SB, et al. Clinical ­utility of percent-positive prostate biopsies in predicting biochemical outcome after radical prostatectomy or external beam radiation therapy for patients with clinically localized prostate cancer. Mol Urol 2000;4:171-176. 123. Gao X, Mohidden N, Flanigan RC, et al. The extent of biopsy ­involvement as an independent predictor of extraprostatic extension and surgical margin status in low risk prostate cancer: implications for treatment selection. J Urol 2000;164:1982-1986. 124. Ravery V, Chastang C, Toublanc M, et al. Percentage of cancer on biopsy cores accurately predicts extracapsular extension and biochemical relapse after radical prostatectomy for T1-T2 prostate cancer. Eur Urol 2000;37:449-455. 125. Ravery V, Szabo J, Toublanc M, et al. A single positive prostate biopsy in six does not predict a low-volume prostate tumour. Br J Urol 1996;77:724-728. 126. Wang X, Brannigan RE, Rademaker AW, et al. One core positive prostate biopsy is a poor predictor of cancer volume in the radical prostatectomy specimen. J Urol 1997;158: 1431-1435. 127. Pan CC, Potter SR, Partin AW, et al. The prognostic significance of tertiary Gleason patterns of higher grade in radical prostatectomy specimens: a proposal to modify the Gleason grading system. Am J Surg Pathol 2000;24:563-569. 128. Forman JD, DeYoung C, Tekyi-Menash S, et al. The prognostic significance of the worst versus overall Gleason score in patients with multiple positive needle biopsies. Int J Radiat Oncol Biol Phys 2000;48(Suppl):206. 129. Jhaveri FM, Klein EA, Kupelian PA, et al. Declining rates of extracapsular extension after radical prostatectomy: evidence for continued stage migration. J Clin Oncol 1999;17: 3167-3172. 130. Amling CL, Blute ML, Lerner SE, et al. Influence of prostatespecific antigen testing on the spectrum of patients with prostate cancer undergoing radical prostatectomy at a large referral practice. Mayo Clin Proc 1998;73:401-406. 131. Lee N, Fawaaz R, Olsson CA, et al. Which patients with newly diagnosed prostate cancer need a radionuclide bone scan? An analysis based on 631 patients. Int J Radiat Oncol Biol Phys 2000;48:1443-1446.

720

PART 8  Male Genital Tract

132. Lee CT, Oesterling JE. Using prostate-specific antigen to eliminate the staging radionuclide bone scan. Urol Clin North Am 1997;24:389-394. 133. Polascik TJ, Manyak MJ, Haseman MK, et al. Comparison of clinical staging algorithms and 111indium-capromab pendetide immunoscintigraphy in the prediction of lymph node involvement in high risk prostate carcinoma patients. Cancer 1999;85: 1586-1592. 134. Manyak MJ, Hinkle GH, Olsen JO, et al. Immunoscintigraphy with indium-111-capromab pendetide: evaluation before definitive therapy in patients with prostate cancer. Urology 1999;54:1058-1063. 135. Sodee DB, Malguria N, Faulhaber P, et al. Multicenter Prosta­ Scint imaging findings in 2154 patients with prostate cancer. Urology 2000;56:988-993. 136. Fang DX, Stock RG, Stone NN, et al. Use of radioimmunoscintigraphy with indium-111–labeled CYT-356 (ProstaScint) scan for evaluation of patients for salvage brachytherapy. Tech Urol 2000;6:146-150. 137. Murphy GP, Elgamal AA, Troychak MJ, et al. Follow-up ProstaScint scans verify detection of occult soft-tissue recurrence after failure of primary prostate cancer therapy. Prostate 2000;42:315-317. 138. Harisinghani MG, Barentsz J, Hahn PF, et al. Noninvasive ­detection of clinically occult lymph-node metastases in prostate cancer. N Engl J Med 2003;348:2491-2499. 139. Zagars GK, Pollack A, von Eschenbach AC. Unfavorable localregional prostate cancer management with radiation and androgen ablation. Cancer 1997;80:764-775. 140. Roach M 3rd, Marquez C, You HS, et al. Predicting the risk of lymph node involvement using the pre-treatment prostate specific antigen and Gleason score in men with clinically localized prostate cancer. Int J Radiat Oncol Biol Phys 1994;28:33-37. 141. Pisansky TM, Kahn MJ, Rasp GM, et al. A multiple prognostic index predictive of disease outcome following irradiation for clinically localized prostatic carcinoma. Cancer 1997;79: 337-344. 142. Partin AW, Kattan MW, Subong EN, et al. Combination of prostate specific antigen, clinical stage, and Gleason score to predict pathological stage of localized prostate cancer. A multi­institutional update. JAMA 1997;277:1445-1451. 143. Roach M 3rd, Chen A, Song J, et al. Pretreatment prostate­specific antigen and Gleason score predict the risk of extracapsular extension and the risk of failure following radiotherapy in patients with clinically localized prostate cancer. Semin Urol Oncol 2000;18:108-114. 144. Pollack A, Zagars GK, Kavadi VS. Prostate specific antigen doubling time and disease relapse after radiotherapy for prostate cancer. Cancer 1994;74:670-678. 145. Paulson DF. Impact of radical prostatectomy in the management of clinically localized disease. J Urol 1994;152:1826-1830. 146. Zagars GK, Kavadi VS, Pollack A, et al. The source of pretreatment serum prostate-specific antigen in clinically localized prostate cancer—T, N, or M? Int J Radiat Oncol Biol Phys 1995;32:21-32. 147. Zagars GK, Pollack A, von Eschenbach AC. Prostate cancer and radiation therapy—the message conveyed by serum prostatespecific antigen. Int J Radiat Oncol Biol Phys 1995;33:23-25. 148. Connolly JA, Shinohara K, Presti JC Jr, et al. Local recurrence after radical prostatectomy: characteristics in size, location, and relationship to prostate-specific antigen and surgical margins. Urology 1996;47:225-231. 149. Shekarriz B, Upadhyay J, Wood DP Jr, et al. Vesicourethral anastomosis biopsy after radical prostatectomy: predictive ­value of prostate-specific antigen and pathologic stage. Urology 1999;54:1044-1048. 150. Pound CR, Partin AW, Eisenberger MA, et al. Natural history of progression after PSA elevation following radical prostatectomy. JAMA 1999;281:1591-1597. 151. Lee WR, Hanks GE, Hanlon A. Increasing prostate-specific antigen profile following definitive radiation therapy for localized prostate cancer: clinical observations. J Clin Oncol 1997;15:230-238.

152. Dillioglugil O, Leibman BD, Kattan MW, et al. Hazard rates for progression after radical prostatectomy for clinically localized prostate cancer. Urology 1997;50:93-99. 153. Smith LG, Pollack A, Zagars GK. Predictors of distant metastasis 7 years after a rising PSA in prostate cancer patients treated with external beam radiotherapy. Int J Radiat Oncol Biol Phys 1999;45(Suppl):138. 154. American Society for Therapeutic Radiology and Oncology Consensus Panel. Consensus statement: guidelines for PSA following radiation therapy. Int J Radiat Oncol Biol Phys 1997;37: 1035-1041. 155. Critz FA, Williams WH, Benton JB, et al. Prostate specific ­antigen bounce after radioactive seed implantation followed by external beam radiation for prostate cancer. J Urol 2000;163: 1085-1089. 156. Hanlon AL, Pinover WH, Horwitz EM, et al. Patterns and fate of PSA bouncing following 3D-CRT. Int J Radiat Oncol Biol Phys 2000;48(Suppl):207. 157. Slivjak AM, Pinover WH, Hanlon AL, et al. The ASTRO consensus definition of bNED failure is an inappropriate endpoint for prostate cancer patients receiving conformal radiation therapy and androgen deprivation (CRT+AD) [abstract]. Int J Radiat Oncol Biol Phys 1998;42(Suppl 1):176. 158. Roach M 3rd, Hanks G, Thames H, et al. Defining biochemical failure following radiotherapy with or without hormonal therapy in men with clinically localized prostate cancer: recommendations of the RTOG-ASTRO Phoenix Consensus Conference. Int J Radiat Oncol Biol Phys 2006;65:965-974. 159. Critz FA, Williams H, Holladay CT, et al. Post-treatment PSA ≤ 0.2 ng/ml defines disease freedom after radiotherapy for prostate cancer using modern techniques. Urology 1999;54: 968-971. 160. Kavadi VS, Zagars GK, Pollack A. Serum prostate-specific ­antigen after radiation therapy for clinically localized prostate cancer: prognostic implications. Int J Radiat Oncol Biol Phys 1994;30:279-287. 161. Lee WR, Hanlon AL, Hanks GE. Prostate specific antigen nadir following external beam radiation therapy for clinically localized prostate cancer: the relationship between nadir level and disease-free survival. J Urol 1996;156:450-453. 162. Zietman AL, Tibbs MK, Dallow KC, et al. Use of PSA nadir to predict subsequent biochemical outcome following external beam radiation therapy for T1-2 adenocarcinoma of the prostate. Radiother Oncol 1996;40:159-162. 163. Yock TI, Thakral HK, Zietman AL, et al. The durability of PSA failure free survival from 5 years to 13 years following radiotherapy in patients with localized prostate cancer [abstract]. Int J Radiat Oncol Biol Phys 2000;48(Suppl):227. 164. D’Amico A, Propert K. Prostate cancer volume adds significantly to prostate specific antigen in the prediction of early biochemical failure after external beam radiation therapy. Int J Radiat ­Oncol Biol Phys 1996;35:273-279. 165. D’Amico AV, Whittington R, Malkowicz SB, et al. Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA 1998;280:969-974. 166. Shipley WU, Thames HD, Sandler HM, et al. Radiation therapy for clinically localized prostate cancer: a multi-institutional pooled analysis. JAMA 1999;281:1598-1604. 167. Zelefsky MJ, Leibel SA, Baudin PB, et al. Dose escalation with three-dimensional conformal radiation therapy affects the outcome in prostate cancer. Int J Radiat Oncol Biol Phys 1998;41:491-500. 168. Movsas B, Hanlon AL, Teshima T, et al. Analyzing predictive models following definitive radiotherapy for prostate carcinoma. Cancer 1997;80:1093-1102. 169. Hanks GE, Leibel S, Kramer S. The dissemination of cancer by transurethral resection of locally advanced prostate cancer. J Urol 1983;129:309-311. 170. Perez CA, Garcia D, Simpson JR, et al. Factors influencing outcome of definitive radiotherapy for localized carcinoma of the prostate. Radiother Oncol 1989;16:1-21.

28  The Prostate 171. Anscher MS, Prosnitz LR. Transurethral resection of prostate prior to definitive irradiation for prostate cancer. Urology 1991;38:206-211. 172. Zagars GK, von Eschenbach AC, Ayala AG. Prognostic factors in prostate cancer: analysis of 874 patients treated with radiation therapy. Cancer 1993;72:1709-1725. 173. Zelefsky MJ, Whitmore WF Jr, Leibel SA, et al. Impact of transurethral resection on the long-term outcome of patients with prostatic carcinoma. J Urol 1993;150:1860-1864. 174. Duncan W, Catton CN, Warde P, et al. The influence of transurethral resection of prostate on prognosis of patients with ­adenocarcinoma of the prostate treated by radical radiotherapy. Radiother Oncol 1994;31:41-50. 175. Nemoto R, Uchida K, Shimazui T, et al. Immunocytochemical demonstration of S phase cells by anti-bromodeoxyuridine monoclonal antibody in human prostate adenocarcinoma. J Urol 1989;141:337-340. 176. Haustermans KMG, Hofland I, Poppel HV, et al. Cell kinetic measurements in prostate cancer. Int J Radiat Oncol Biol Phys 1997;37:1067-1070. 177. Borre M, Høyer M, Sørensen FB, et al. Association between in vivo iododeoxyuridine labeling, MIB-1 expression, malignancy grade and clinical stage in human prostate cancer. APMIS 1998;106:389-395. 178. Khoo VS, Pollack A, Cowen D, et al. Relationship of Ki-67 ­labeling index to DNA-ploidy, S-phase fraction, and outcome in prostate cancer treated with radiotherapy. Prostate 1999;41: 166-172. 179. Cowen D, Troncoso P, Khoo VS, et al. Ki-67 staining is an ­independent correlate of biochemical failure in prostate cancer treated with radiotherapy. Clin Cancer Res 2002;8:1148-1154. 180. Schmid HP, McNeal JE, Stamey TA. Observations on the doubling time of prostate cancer: the use of serial prostate-­specific antigen in patients with untreated disease as a measure of ­increasing cancer volume. Cancer 1993;71:2031-2040. 181. Hanks GE, Hanlon AL, Lee WR, et al. Pretreatment prostate specific antigen doubling times: clinical utility of this predictor of prostate cancer behavior. Int J Radiat Oncol Biol Phys 1996;34:549-553. 182. Berges RR, Vukanovic J, Epstein JI, et al. Implication of cell ­kinetic changes during progression of human prostatic cancer. Clin Cancer Res 1995;1:473-480. 183. Brenner DJ, Hall EJ. Fractionation and protraction for radiotherapy of prostate carcinoma. Int J Radiat Oncol Biol Phys 1999;43:1095-1101. 184. Duschesne GM, Peters LJ. What is the α/β ratio for prostate cancer? Rationale for hypofractionated high dose-rate brachytherapy. Int J Radiat Oncol Biol Phys 1999;44:747-748. 185. King CR, Mayo CS. Is the α/β ratio of 1.5 from Brenner and Hall a modelling artifact? Int J Radiat Oncol Biol Phys 1999;47: 536-538. 186. King CR. What is the Tpot for prostate cancer? Radiobiological implications of the equivalent outcome with 125I or 103Pd. Int J Radiat Oncol Biol Phys 2000;47:1165-1167. 187. Mohan DS, Kupelian PA, Willoughby TR. Short-course ­intensitymodulated radiotherapy for localized prostate cancer with daily transabdominal ultrasound localization of the prostate gland. Int J Radiat Oncol Biol Phys 2000;46:575-580. 188. Gauwitz MD, Pollack A, El-Naggar AK, et al. The prognostic significance of DNA ploidy in clinically localized prostate cancer treated with radiation therapy. Int J Radiat Oncol Biol Phys 1994;28:821-828. 189. Centeno BA, Zietman AL, Shipley WU, et al. Flow cytometric analysis of DNA ploidy, percent S-phase fraction, and total proliferative fraction as prognostic indicators of local control and survival following radiation therapy for prostate carcinoma. Int J Radiat Oncol Biol Phys 1994;30:309-315. 190. Pollack A, Troncoso P, Zagars GK, et al. The significance of DNA-ploidy and S-phase fraction in node-positive (stage D1) prostate cancer treated with androgen ablation. Prostate 1997;1: 21-28. 191. Kupelian P, Klien EA, Witte JS, et al. Familial prostate cancer: a different disease? J Urol 1997;158:2197-2201.

721

192. Kupelian P, Kupelian VA, Witte JS, et al. Family history of prostate cancer in patients with localized prostate cancer: an independent predictor of treatment outcome. J Clin Oncol 1997;15:1478-1480. 193. Hanus MC, Zagars GK, Pollack A. Familial prostate cancer: outcome following radiation therapy with or without adjuvant ­androgen ablation. Int J Radiat Oncol Biol Phys 1999;43:379-383. 194. Bova GS, Isaacs SD, Partin AW, et al. Biological aggressiveness of hereditary prostate cancer (HPC): long term evaluation following radical prostatectomy (RRP) [abstract]. J Urol 1995;153:505A. 195. Zagars GK, Pollack A. Serum testosterone levels after external beam radiation for clinically localized prostate cancer. Int J ­Radiat Oncol Biol Phys 1997;39:85-89. 196. Kelly J, Pollack A, Zagars GK. Serum testosterone is not a correlate of prostate cancer lymph node involvement, but does predict biochemical failure for lymph-node-positive disease. Urol Oncol 2000;5:78-84. 197. Harper ME, Pierrepoint CG, Griffiths K. Carcinoma of the prostate: relationship of pre-treatment hormone levels to survival. Eur J Cancer Clin Oncol 1984;20:477-482. 198. Chodak GW, Vogelzang NJ, Caplan RJ, et al. Independent prognostic factors in patients with metastatic (stage D2) prostate cancer. JAMA 1991;265:618-621. 199. Iversen P, Rasmussen F, Christensen IJ. Serum testosterone as a prognostic factor in patients with advanced prostatic carcinoma. Scand J Urol Nephrol 1994;157(Suppl):41-47. 200. Pollack A, Zagars GK, Smith LG, et al. Preliminary results of a randomized radiotherapy dose-escalation study comparing 70 Gy with 78 Gy for prostate cancer. J Clin Oncol 2000;18: 3904-3911. 201. Kestin L, Goldstein N, Vicini F, et al. Treatment of prostate cancer with radiotherapy: should the entire seminal vesicles be included in the clinical target volume? Int J Radiat Oncol Biol Phys 2002;54:686-697. 202. Skwarchuk MW, Jackson A, Zelefsky MJ, et al. Late rectal toxicity after conformal radiotherapy of prostate cancer (I): multivariate analysis and dose-response. Int J Radiat Oncol Biol Phys 2000;47:103-113. 203. Pollack A, Zagars GK, Starkschall G, et al. Conventional versus conformal radiotherapy for prostate cancer: preliminary results of dosimetry and acute toxicity. Int J Radiat Oncol Biol Phys 1996;34:555-564. 204. Michalski JM, Purdy JA, Winter K, et al. Preliminary report of toxicity following 3D radiation therapy for prostate cancer on 3DOG/RTOG 9406. Int J Radiat Oncol Biol Phys 2000;46: 391-402. 205. Lee WR, Hanks GE, Hanlon AL, et al. Lateral rectal shielding reduces late rectal morbidity following high dose three­dimensional conformal radiation therapy for clinically localized prostate cancer: further evidence for a significant dose effect. Int J Radiat Oncol Biol Phys 1996;35:251-257. 206. Pickett B, Roach M 3rd, Horine P, et al. Optimization of the ­oblique angles in the treatment of prostate cancer during sixfield conformal radiotherapy. Med Dosim 1994;19:237-254. 207. Perez CA, Michalski JM, Purdy JA, et al. Three-dimensional conformal therapy or standard irradiation in localized carcinoma of prostate: preliminary results of a nonrandomized comparison. Int J Radiat Oncol Biol Phys 2000;47:629-637. 208. Kupelian PA, Willoughby TR. Short course intensity modulated radiotherapy (70 Gy at 2.5 Gy per fraction) versus 3D­conformal radiotherapy (78 Gy at 2.0 Gy per fraction) for the treatment of localized prostate cancer: comparable toxicity. Int J Radiat Oncol Biol Phys 2000;48(Suppl 1):250. 209. ICRU. Report 50. Prescribing, recording, and reporting photon beam therapy. International Commission on Radiation Units and Measures, Bethesda, MD. ICRU Rep 1993;9:1-72. 210. Sultanem K, Shu H, Xia P, et al. Three-dimensional intensitymodulated radiotherapy in the treatment of nasopharyngeal carcinoma: the University of California-San Francisco experience. Int J Radiat Oncol Biol Phys 2000;48:711-722. 211. Mackie TR, Holmes T, Swerdloff S, et al. Tomotherapy: a new concept for the delivery of dynamic conformal radiotherapy. Med Phys 1993;20:1709-1719.

722

PART 8  Male Genital Tract

212. Verhey LJ. Comparison of three-dimensional conformal radiation therapy and intensity-modulated radiation therapy systems. Semin Radiat Oncol 1999;9:78-98. 213. Boyer AL, Yu CX. Intensity-modulated radiation therapy with dynamic multileaf collimators. Semin Radiat Oncol 1999;9: 48-59. 214. Zelefsky MJ, Fuks Z, Happersett L, et al. Clinical experience with intensity modulated radiation therapy (IMRT) in prostate cancer. Radiother Oncol 2000;55:241-249. 215. Klein EE, Low DA, Sohn JW, et al. Differential dosing of prostate and seminal vesicles using dynamic multileaf collimation. Int J Radiat Oncol Biol Phys 2000;48:1447-1456. 216. Dong L, O’Daniel JC, Smith LG, et al. Comparison of 3D conformal and intensity-modulated radiation therapy for early-stage prostate cancer [abstract 2183]. Int J Radiat Oncol Biol Phys 2001;51(Suppl 1):320. 217. Zelefsky MJ, Happersett L, Leibel SA, et al. The effect of treatment positioning on normal tissue dose in patients with prostate cancer treated with three-dimensional conformal radiotherapy. Int J Radiat Oncol Biol Phys 1997;37:13-19. 218. Fiorino C, Reni M, Bolognesi A, et al. Set-up error in supine­positioned patients immobilized with two different modalities during conformal radiotherapy of prostate cancer. Radiother Oncol 1998;49:133-141. 219. Malone S, Szanto J, Perry G, et al. A prospective comparison of three systems of patient immobilization for prostate radiotherapy. Int J Radiat Oncol Biol Phys 2000;48:657-665. 220. Weber DC, Nouet P, Rouzaud M, et al. Patient positioning in prostate radiotherapy: is prone better than supine? Int J Radiat Oncol Biol Phys 2000;47:365-371. 221. Dawson LA, Litzenberg DW, Brock KK, et al. A comparison of ventilatory prostate movement in four treatment positions. Int J Radiat Oncol Biol Phys 2000;48:319-323. 222. Malone S, Crook JM, Kendal WS, et al. Respiratory-induced prostate motion: quantification and characterization. Int J Radiat Oncol Biol Phys 2000;48:105-109. 223. Crook JM, Raymond Y, Salhani D, et al. Prostate motion during standard radiotherapy as assessed by fiducial markers. Radiother Oncol 1995;37:35-42. 224. Roeske JC, Forman JD, Mesina CF, et al. Evaluation of changes in the size and location of the prostate, seminal vesicles, bladder, and rectum during a course of external beam radiation therapy. Int J Radiat Oncol Biol Phys 1995;33:1321-1329. 225. Rudat V, Schraube P, Oetzel D, et al. Combined error of patient positioning variability and prostate motion uncertainty in 3D conformal radiotherapy of localized prostate cancer. Int J Radiat Oncol Biol Phys 1996;35:1027-1034. 226. Van Herk M, Bruce AM, Kroes APG, et al. Quantification of ­organ motion during conformal radiotherapy of the prostate by three dimensional image registration. Int J Radiat Oncol Biol Phys 1995;33:1311-1320. 227. Balter JM, Sandler HM, Lam K, et al. Measurement of prostate movement over the course of routine radiotherapy using implanted markers. Int J Radiat Oncol Biol Phys 1995;31: 113-118. 228. Althof VG, Hoekstra CJ, te Loo HJ. Variation in prostate position relative to adjacent bony anatomy. Int J Radiat Oncol Biol Phys 1996;34:709-715. 229. Melian E, Mageras GS, Fuks Z, et al. Variation in prostate position quantitations and implications for three-dimensional conformal treatment planning. Int J Radiat Oncol Biol Phys 1997;38: 73-81. 230. Vigneault E, Pouliot J, Laverdière J, et al. Electronic portal imaging device detection of radioopaque markers for the evaluation of prostate position during megavoltage irradiation: a clinical study. Int J Radiat Oncol Biol Phys 1997;37:205-212. 231. Antolak JA, Rosen II, Childress CH, et al. Prostate target volume variations during a course of radiotherapy. Int J Radiat ­Oncol Biol Phys 1998;42:661-672. 232. Zelefsky MJ, Crean D, Mageras GS, et al. Quantification and predictors of prostate position variability in 50 patients evaluated with multiple CT scans during conformal radiotherapy. ­Radiother Oncol 1999;50:225-234.

233. Lattanzi J, McNeeley S, Pinover W, et al. A comparison of daily CT localization to a daily ultrasound-based system in prostate cancer. Int J Radiat Oncol Biol Phys 1999;43:719-725. 234. O’Daniel J, Dong L, Zhang L, et al. Dosimetric comparison of four target alignment methods for prostate cancer radiotherapy. Int J Radiat Oncol Biol Phys 2006;66:883-897. 235. Feigenberg S, Paskalev K, McNeeley S, et al. Comparing computed tomography localization with daily ultrasound during image-guided radiation therapy for the treatment of prostate cancer: a prospective evaluation. J Appl Clin Med Phys 2007;8: 99-110. 236. Chen LM, Lubich L, Chiru P, et al. Localization of the prostatic apex for radiotherapy planning: a comparison of two techniques. Br J Radiol 1996;69:821-829. 237. Wilder RB, Fone PD, Rademacher DE, et al. Localization of the prostatic apex for radiotherapy treatment planning using urethroscopy. Int J Radiat Oncol Biol Phys 1997;38:737-741. 238. Malone S, Donker R, Broader M, et al. Effects of urethrography on prostate position: considerations for radiotherapy treatment planning of prostate carcinoma. Int J Radiat Oncol Biol Phys 2000;46:89-93. 239. Sandler HM, Bree BL, McLaughlin PW, et al. Localization of the prostatic apex for radiation therapy using implanted markers. Int J Radiat Oncol Biol Phys 1993;27:915-919. 240. Milsosevic M, Voruganti S, Blend R, et al. Magnetic resonance imaging (MRI) for localization of the prostatic apex: comparison to computed tomography (CT) and urethrography. Radiother Oncol 1998;47:277-284. 241. Plants B, Chen D, Fiveash J, et al. Bulb of penis as a marker for prostatic apex in external beam radiotherapy of prostate cancer. Int J Radiat Oncol Biol Phys 2003;56:1079-1084. 242. Fisch BM, Pickett B, Weinberg V, et al. Dose of radiation ­received by the bulb of the penis correlates with risk of impotence after three-dimensional conformal radiotherapy for prostate cancer. Urology 2001;57:955-959. 243. Selek U, Cheung R, Lii M, et al. Erectile dysfunction and radiation dose to penile base structures: a lack of correlation. Int J Radiat Oncol Biol Phys 2004;59:1039-1046. 244. van der Wielen GJ, Hoogeman MS, Dohle GR, et al. Dose­volume parameters of the corpora cavernosa do not correlate with erectile dysfunction after external beam radiotherapy for prostate cancer: results from a dose-escalation trial. Int J Radiat Oncol Biol Phys 2008;71:795-800. 245. Huang WC, Kuroiwa K, Serio AM, et al. The anatomical and pathological characteristics of irradiated prostate cancers may influence the oncological efficacy of salvage ablative therapies. J Urol 2007;177:1324-1329. 246. Liebel SA, Fuks Z, Zelefsky MJ, et al. The effects of local and regional treatment on the metastatic outcome in prostatic carcinoma with pelvic lymph node involvement. Int J Radiat Oncol Biol Phys 1994;28:7-16. 247. Chan RC, Gutierrez AE. Carcinoma of the prostate: its treatment by a combination of radioactive gold-grain implant and external irradiation. Cancer 1976;37:2749-2754. 248. Butler EB, Scardino PT, The BS, et al. The Baylor College of Medicine experience with gold seed implantation. Semin Surg Oncol 1997;13:406-418. 249. Levitt SH, Tapley N. Technological Basis of Radiation Therapy: Practical Clinical Applications. Philadelphia: Lea & Febiger, 1984. 250. Zelefsky MJ, Whitmore WF Jr. Long-term results of retropubic permanent 125iodine implantation of the prostate for clinically localized prostate cancer. J Urol 1997;158:23-29. 251. Eastham JA, Kattan MW, Groshen S, et al. Fifteen-year survival and recurrence rates after radiotherapy for localized prostate cancer. J Clin Oncol 1997;15:3214-3222. 252. Scardino PT, Frankel JM, Wheeler TM, et al. The prognostic significance of post-irradiation biopsy results in patients with prostatic cancer. J Urol 1986;135:510-516. 253. Prestidge BR, Hoak DC, Grimm PD, et al. Posttreatment biopsy results following interstitial brachytherapy in earlystage prostate cancer. Int J Radiat Oncol Biol Phys 1997;37: 31-39.

28  The Prostate 254. Stock RG, Stone NN, Kao J, et al. The effects of disease and treatment-related factors on biopsy results after prostate brachytherapy. Cancer 2000;89:1829-1834. 255. Blasko JC, Wallner K, Grimm PD, et al. Prostate specific ­antigen based disease control following ultrasound guided 125iodine implantation for stage T1/T2 prostatic carcinoma. J Urol 1995;154:1096-1099. 256. Ragde H, Korb LJ, Elgamal AA, et al. Modern prostate ­brachytherapy-prostate specific antigen results in 219 patients with up to 12 years of observed follow-up. Cancer 2000;89: 135-141. 257. Stock RG, Stone NN, Lo YC. Intraoperative dosimetric representation of the real-time ultrasound-guided prostate implant. Tech Urol 2000;6:95-98. 258. Zelefsky MJ, Yamada Y, Cohen G, et al. Postimplantation ­dosimetric analysis of permanent transperineal prostate implantation: improved dose distributions with an intraoperative computer-optimized conformal planning technique. Int J Radiat Oncol Biol Phys 2000;48:601-608. 259. D’Amico A, Cormack R, Kumar S, et al. Real-time magnetic resonance imaging-guided brachytherapy in the treatment of selected patients with clinically localized prostate cancer. J ­Endourol 2000;14:367-370. 260. Koutrouvelis P, Lailas N, Katz S, et al. High- and low-risk prostate cancer treated with 3D CT-guided brachytherapy: 1- to 5-year follow-up. J Endourol 2000;14:357-366. 261. Davis BJ, Pisansky TM, Wilson TM, et al. The radial distance of extraprostatic extension of prostate carcinoma. Cancer 1999;85:2630-2637. 262. Sohayda C, Kupelian PA, Levin HS, et al. Extent of extracapsular extension in localized prostate cancer. Urology 2000;55: 382-386. 263. Prestidge BR, Bice WS, Kiefer EJ, et al. Timing of computed ­tomography-based postimplant assessment following permanent transperineal prostate brachytherapy. Int J Radiat Oncol Biol Phys 1998;40:1111-1115. 264. Yue N, Chen Z, Peschel R, et al. Optimum timing for imagebased dose evaluation of 1251 and 103Pd prostate seed implants. Int J Radiat Oncol Biol Phys 1999;45:1063-1072. 265. Stock RG, Stone NN, Tabert A, et al. A dose-response study for I-125 prostate implants. Int J Radiat Oncol Biol Phys 1998;41:101-108. 266. Nag S, Bice W, DeWyngaert K, et al. The American Brachytherapy Society recommendations for permanent prostate brachytherapy postimplant dosimetric analysis. Int J Radiat Oncol Biol Phys 2000;46:221-230. 267. American Association of Physicists in Medicine. AAPM report no. 43. Dosimetry of interstitial brachytherapy sources: recommendations of the AAPM Radiation Therapy Committee Task Group No. 43. Med Phys 1995;22:209-234. 268. Kubo HD, Coursey BM, Hanson WF, et al. Report of the Ad Hoc Committee of the AAPM Radiation Therapy Committee on 125I sealed source dosimetry. Int J Radiat Oncol Biol Phys 1998;40:697-702. 269. Bice WS, Prestidge BR, Prete JJ, et al. Clinical impact of ­implementing the recommendations of AAPM Task Group 43 on permanent prostate brachytherapy using 125I. American Association of Physicists in Medicine. Int J Radiat Oncol Biol Phys 1998;40:1237-1241. 270. Beyer D, Nath R, Butler W, et al. American Brachytherapy ­Society recommendations for clinical implementation of NIST1999 standards for (103) palladium brachytherapy. The clinical research committee of the American Brachytherapy Society. Int J Radiat Oncol Biol Phys 2000;47:273-275. 271. Williamson JF, Coursey BM, DeWerd LA, et al. Recommendations of the American Association of Physicists in Medicine on 103Pd interstitial source calibration and dosimetry: implication for dose specification and prescription. Med Phys 2000;27: 634-642. 272. Dicker AP, Lin CC, Leeper DB, et al. Isotope selection for permanent prostate implants? An evaluation of 103Pd versus 125I based on radiobiological effectiveness and dosimetry. Semin Urol Oncol 2000;18:152-159.

723

273. Ling CC. Permanent implants using Au-198, Pd-103 and I-125: radiobiological considerations based on the linear quadratic model. Int J Radiat Oncol Biol Phys 1992;23:81-87. 274. Critz FA, Levinson AK, Williams WH, et al. Simultaneous ­radiotherapy for prostate cancer: 125I prostate implant followed by external beam radiation. Cancer J Sci Am 1998;4:359-363. 275. Dattoli M, Wallner K, Sorace R, et al. 103Pd brachytherapy and external beam irradiation for clinically localized, high-risk prostatic carcinoma. Int J Radiat Oncol Biol Phys 1996;35:875-879. 276. Mate TP, Gottesman JE, Hatton J, et al. High dose-rate afterloading 192iridium prostate brachytherapy: feasibility report. Int J Radiat Oncol Biol Phys 1998;41:525-533. 277. Vicini FA, Kestin LL, Stromberg JS, et al. Brachytherapy boost techniques for locally advanced prostate cancer. Oncology 1999;13:491-509. 278. Martinez AA, Kestin LL, Stromberg JS, et al. Interim report of image-guided conformal high-dose-rate brachytherapy for patients with unfavorable prostate cancer: the William Beaumont phase II dose-escalating trial. Int J Radiat Oncol Biol Phys 2000;47:343-352. 279. Borghede G, Hedelin H, Holmäng S, et al. Combined treatment with temporary short-term high dose rate iridium-192 brachytherapy and external beam radiotherapy for irradiation of localized prostatic carcinoma. Radiother Oncol 1997;44:237-244. 280. Deger S, Dinges S, Roigas J, et al. High-dose rate iridium192 afterloading therapy in combination with external beam irradiation for localized prostate cancer. Tech Urol 1997;3:190-194. 281. Pollack A, Zagars GK. External beam radiotherapy dose response of prostate cancer. Int J Radiat Oncol Biol Phys 1997;39: 1011-1018. 282. Pinover WH, Hanlon AL, Horwitz EM, et al. Defining the ­appropriate radiation dose for pretreatment PSA � ≤ 10 ng/ml prostate cancer. Int J Radiat Oncol Biol Phys 2000;47:649-654. 283. Lyons JA, Kupelian PA, Mohan DS, et al. Importance of high radiation doses (72 Gy or greater) in the treatment of stage T1-T3 adenocarcinoma of the prostate. Urology 2000;55:85-90. 284. Keyser D, Kupelian PA, Zippe CD, et al. Stage T1-2 prostate cancer with pretreatment prostate-specific antigen level ≤ 10 ng/ml: radiation therapy or surgery? Int J Radiat Oncol Biol Phys 1997;38:723-729. 285. Brachman DG, Thomas T, Hilbe J, et al. Failure-free survival following brachytherapy alone or external beam irradiation alone for T1-2 prostate tumors in 2222 patients: results from a single practice. Int J Radiat Oncol Biol Phys 2000;48:111-117. 286. Hanks GE, Hanlon AL, Pinover WH, et al. Dose escalation for prostate cancer patients based on dose comparison and dose ­response studies. Int J Radiat Oncol Biol Phys 2000;46:823-832. 287. Pound CR, Partin AW, Epstein JI, et al. Prostate-specific antigen after anatomic radical retropubic prostatectomy. Urol Clin North Am 1997;24:395-406. 288. Catalona WJ, Smith DS. Cancer recurrence and survival rates after anatomic radical retropubic prostatectomy for prostate cancer: intermediate-term results. J Urol 1998;160:2428-2434. 289. Wallner K, Roy J, Harrison L. Tumor control and morbidity following transperineal iodine 125 implantation for stage T1/T2 prostate carcinoma. J Clin Oncol 1996;14:449-453. 290. Grado GL, Larson TR, Balch CS, et al. Actuarial disease-free survival after prostate cancer brachytherapy using interactive techniques with biplane ultrasound and fluoroscopic guidance. Int J Radiat Oncol Biol Phys 1998;42:289-298. 291. Stock RG, Stone NN, DeWyngaert JK, et al. Prostate specific ­antigen findings and biopsy results following interactive ultrasound guided transperineal brachytherapy for early stage prostate carcinoma. Cancer 1996;77:2386-2392. 292. Kuban D, Tucker S, Dong L, et al. Long-term results of the M. D. Anderson randomized dose-escalation trial for prostate cancer. Int J Radiat Oncol Biol Phys 2008;70:67-74. 293. Zietman AL. Correction: inaccurate analysis and results in a study of radiation therapy in adenocarcinoma of the prostate. JAMA 2008;299:898-899. [Comment on Zietman A, DeSilvio M, Slater J, et al. Comparison of conventional-dose vs. high-dose conformal radiation therapy in clinically localized adenocarcinoma of the prostate. JAMA 2005;294:1233-1239.]

724

PART 8  Male Genital Tract

294. Zelefsky MJ, Wallner KE, Ling C, et al. Comparison of the 5-year outcome and morbidity of three-dimensional conformal radiotherapy versus transperineal permanent iodine-125 implantation for early-stage prostatic cancer. J Clin Oncol 1999;17:517-522. 295. Stokes SH. Comparison of biochemical disease-free survival of patients with localized carcinoma of the prostate undergoing radical prostatectomy, transperineal ultrasound-guided radioactive seed implantation, or definitive external beam irradiation. Int J Radiat Oncol Biol Phys 2000;47:129-136. 296. Polascik TJ, Pound CR, De Weese TL, et al. Comparison of radical prostatectomy and iodine 125 interstitial radiotherapy for the treatment of clinically localized prostate cancer: a 7-year biochemical (PSA) progression analysis. Urology 1998;51:884-890. 297. Ramos CG, Carvalhal GF, Smith DS, et al. Retrospective ­comparison of radical retropubic prostatectomy and 125iodine brachytherapy for localized prostate cancer. J Urol 1999;161:1212-1215. 298. Ragde H, Blasko JC, Grimm PD, et al. Interstitial iodine 125 radiation without adjuvant therapy in the treatment of clinically localized prostate carcinoma. Cancer 1997;80:442-453. 299. Kupelian P, Katcher J, Levin H, et al. External beam radiotherapy versus radical prostatectomy for clinical stage T1-2 prostate cancer: therapeutic implications of stratification by pretreatment PSA levels and biopsy Gleason scores. Cancer J Sci Am 1997;3:78-87. 300. Lee F, Bahn DK, Siders DB, et al. The role of TRUS-guided biopsies for determination of internal and external spread of prostate cancer. Semin Urol Oncol 1998;16:129-136. 301. Taneja SS, Penson DF, Epelbaum A, et al. Does site-specific ­labeling of sextant biopsy cores predict the site of extracapsular extension in radical prostatectomy surgical specimen? J Urol 1999;162:1352-1357. 302. Terk MD, Stock RG, Stone NN. Identification of patients at ­increased risk for prolonged urinary retention following radioactive seed implantation of the prostate. J Urol 1998;160:1379-1382. 303. Gelblum DY, Potters L, Ashley R, et al. Urinary morbidity following ultrasound-guided transperineal prostate seed implantation. Int J Radiat Oncol Biol Phys 1999;45:59-67. 304. Stone NN, Ratnow ER, Stock RG. Prior transurethral resection does not increase morbidity following real-time ultrasoundguided prostate seed implantation. Tech Urol 2000;6:123-127. 305. Bellon J, Wallner K, Ellis W, et al. Use of pelvic CT scanning to evaluate pubic arch interference of transperineal prostate brachy­ therapy. Int J Radiat Oncol Biol Phys 1999;43:579-581. 306. Wallner K, Ellis W, Russell K, et al. Use of TRUS to predict pubic arch interference of prostate brachytherapy. Int J Radiat Oncol Biol Phys 1999;43:583-585. 307. Stone NN, Stock RG. Prostate brachytherapy in patients with prostate volumes ≥ 50 cm (3): dosimetic analysis of implant quality. Int J Radiat Oncol Biol Phys 2000;46:1199-1204. 308. Wang H, Wallner K, Sutlief S, et al. Transperineal brachytherapy in patients with large prostate glands. Int J Cancer 2000;90:199-205. 309. Lee N, Wuu C, Brody R, et al. Factors predicting for postimplantation urinary retention after permanent prostate brachytherapy. Int J Radiat Oncol Biol Phys 2000;48:1457-1460. 310. Peschel RE, Chen Z, Roberts K. Long-term complications with prostate implants: iodine-125 vs palladium-103. Radiat Oncol Investig 1999;7:278-288. 311. Gelblum DY, Potters L. Rectal complications associated with transperineal interstitial brachytherapy for prostate cancer. Int J Radiat Oncol Biol Phys 2000;48:119-124. 312. Sánchez-Ortiz RF, Broderick GA, Rovner ES, et al. Erectile function and quality of life after interstitial radiation therapy for prostate cancer. Int J Impot Res 2000;12:S18-24. 313. Zelefsky MJ, Hollister T, Raben A, et al. Five-year biochemical outcome and toxicity with transperineal CT-planned permanent I-125 prostate implantation for patients with localized prostate cancer. Int J Radiat Oncol Biol Phys 2000;47:1261-1266. 314. Zelefsky MJ, Wallner KE, Ling CC, et al. Comparison of the 5-year outcome and morbidity of three-dimensional conformal radiotherapy versus transperineal permanent iodine-125 implantation for early-stage prostatic cancer. J Clin Oncol 1999;17:517-522.

315. Wilder RB, Chou RH, Ryu JK, et al. Potency preservation after three-dimensional conformal radiotherapy for prostate cancer: preliminary results. Am J Clin Oncol 2000;23:330-333. 316. Brandeis JM, Litwin MS, Burnison CM, et al. Quality of life outcomes after brachytherapy for early stage prostate cancer. J Urol 2000;163:851-857. 317. Zeitlin SI, Sherman J, Raboy A, et al. High dose combination ­radiotherapy for the treatment of localized prostate cancer. J Urol 1998;160:91-96. 318. Sylvester J, Grimm P, Blasko J, et al. 15-Year biochemical relapse free survival in clinical stage T1-T3 prostate cancer following combined external beam radiotherapy and brachytherapy: ­Seattle experience. Int J Radiat Oncol Biol Phys 2007;67:57-64. 319. Pollack A, Smith LG, von Eschenbach AC. External beam radiotherapy dose-response characteristics of 1127 men with prostate cancer treated in the PSA era. Int J Radiat Oncol Biol Phys 2000;48:507-512. 320. Zelefsky MJ, Fuks Z, Venkatraman ES, et al. Predictors of posttreatment prostatic biopsy outcome after 3-dimensional conformal radiotherapy for patients with clinically localized prostate cancer. Int J Radiat Oncol Biol Phys 2000;48(Suppl 1):231. 321. Hanks GE, Leibel S, Kramer S. The dissemination of cancer by transurethral resection of locally advanced prostate cancer. J Urol 1983;129:309-311. 322. Hanks GE, Martz KL, Diamond JJ. The effect of dose on ­local control of prostate cancer. Int J Radiat Oncol Biol Phys 1988;15:1299-1305. 323. Perez CA, Walz BJ, Zivnuska FR, et al. Irradiation of carcinoma of the prostate localized to the pelvis: analysis of tumor response and prognosis. Int J Radiat Oncol Biol Phys 1980;6:555-563. 324. Perez CA, Pilepich MV, Garcia D, et al. Definitive radiation therapy in carcinoma of the prostate localized to the pelvis: ­experience at the Mallinckrodt Institute of Radiology. NCI Monogr 1988;7:85-94. 325. Shipley WU, Verhey LJ, Munzenrider JE, et al. Advanced prostate cancer: the results of a randomized comparative trial of high dose irradiation boosting with conformal protons compared with conventional dose irradiation using photons alone. Int J Radiat Oncol Biol Phys 1995;32:13-20. 326. Peeters S, Heemsbergen W, Koper P, et al. Dose-response in ­radiotherapy for localized prostate cancer: results of the Dutch multicenter randomized phase III trial comparing 68 Gy of ­radiotherapy with 78 Gy. J Clin Oncol 2006;24:1990-1996. 327. Dearnaley DP, Khoo VS, Norman AR, et al. Comparison of radiation side-effects of conformal and conventional radiotherapy in prostate cancer: a randomized trial. Lancet 1999;353:267-272. 328. Smit WGJM, Helle PA, van Putten WL, et al. Late radiation damage in prostate cancer patients treated by high dose external radiotherapy in relation to rectal dose. Int J Radiat Oncol Biol Phys 1990;18:23-29. 329. Benk VA, Adams JA, Shipley WU, et al. Late rectal bleeding following combined x-ray and proton high dose irradiation for patients with stages T3-T4 prostate carcinoma. Int J Radiat ­Oncol Biol Phys 1993;26:551-557. 330. Boersma LJ, van den Brink M, Bruce AM, et al. Estimation of the incidence of late bladder and rectum complications after high-dose (70-78 Gy) conformal radiotherapy for prostate cancer, using dose-volume histograms. Int J Radiat Oncol Biol Phys 1998;41:83-92. 331. Tucker S, Dong L, Cheung R, et al. Comparison of rectal dosewall histogram versus dose-volume histogram for modeling the incidence of late rectal bleeding after radiotherapy. Int J Radiat Oncol Biol Phys 2004;60:589-601. 332. Asbell SO, Krall JM, Pilepich MV, et al. Elective pelvic irradiation in stage A2, B carcinoma of the prostate: analysis of RTOG 77-06. Int J Radiat Oncol Biol Phys 1988;15:1307-1316. 333. McGowan DG. The value of extended field radiation therapy in carcinoma of the prostate. Int J Radiat Oncol Biol Phys 1981;7:1333-1339. 334. Ploysongsang S, Scott RM, Aron BS, et al. Comparison of whole pelvic versus small-field radiation therapy for carcinoma of the prostate. Urology 1986;27:10-16.

28  The Prostate 335. Lawton CA, Cox JD, Glisch C, et al. Is long-term survival possible with external beam irradiation for stage D1 adenocarcinoma of the prostate? Cancer 1992;69:2761-2766. 336. Seaward SA, Weinberg V, Lewis P, et al. Identification of a highrisk clinically localized prostate cancer subgroup receiving maximum benefit from whole-pelvic irradiation. Cancer J Sci Am 1998;4:370-377. 337. Seaward SA, Weinberg V, Lewis P, et al. Improved freedom from PSA failure with whole pelvic irradiation for high-risk prostate cancer. Int J Radiat Oncol Biol Phys 1998;42:1055-1062. 338. Zagars GK, von Eschenbach AC, Johnson DE, et al. Stage C ­adenocarcinoma of the prostate. An analysis of 551 patients treated with external beam radiation. Cancer 1987;60:1489-1499. 339. Rasp GM, Pisansky TM, Haddock MG, et al. Elective pelvic nodal irradiation in patients with clinically localized prostate cancer at high risk for pelvic nodal involvement [abstract]. Int J Radiat Oncol Biol Phys 1996;36(Suppl):245. 340. Perez CA, Michalski J, Brown KC, et al. Nonrandomized evaluation of pelvic lymph node irradiation in localized carcinoma of the prostate. Int J Radiat Oncol Biol Phys 1996;36:573-584. 341. Roach M 3rd, DeSilvio M, Lawton C, et al. Phase III trial comparing whole-pelvic versus prostate-only radiotherapy and ­neoadjuvant versus adjuvant combined androgen suppression: Radiation Therapy Oncology Group 9413. J Clin Oncol 2003;21: 1904-1911. 342. Lawton C, DeSilvio M, Roach M 3rd, et al. An update of the phase III trial comparing whole pelvic to prostate only ­radiotherapy and neoadjuvant to adjuvant total androgen suppression: updated analysis of RTOG 94-13, with emphasis on unexpected hormone/radiation interactions. Int J Radiat Oncol Biol Phys 2007;69:646-655. 343. Lim Joon D, Hasegawa M, Sikes C, et al. Supra-additive apoptotic response of R3327-G rat prostate tumors to androgen ablation and radiation. Int J Radiat Oncol Biol Phys 1997;38:1071-1077. 344. Zietman AL, Prince E, Nakfoor BM, et al. Androgen deprivation and radiation therapy: sequencing studies using the shionogi in vivo tumor system. Int J Radiat Oncol Biol Oncol 1997;38: 1067-1070. 345. Pollack A, Ashoori F, Sikess C, et al. The early supra-additive apoptotic response of R3327-G prostate tumors to androgen ­ablation and radiation is not sustained with multiple fractions. Int J Radiat Oncol Biol Phys 1999;46:153-158. 346. Pilepich MV, Krall JM, John MJ, et al. Hormonal cytoreduction in locally advanced carcinoma of the prostate treated with ­definitive radiotherapy: preliminary results of RTOG 83-07. Int J Radiat Oncol Biol Phys 1989;16:813-817. 347. Pilepich MV, John MJ, Krall JM, et al. Phase II Radiation Therapy Oncology Group study of hormonal cytoreduction with flutamide and Zoladex in locally advanced carcinoma of the prostate treated with definitive radiotherapy. Am J Clin Oncol 1990;13:461-464. 348. Pilepich MV, Krall JM, Al-Sarraf M, et al. Androgen deprivation with radiation therapy compared with radiation therapy alone for locally advanced prostatic carcinoma: a randomized comparative trial of the Radiation Therapy Oncology Group. Urology 1995;45:616-623. 349. Hanks GE, Pajak TF, Porter A, et al. Phase III trial of long-term adjuvant androgen deprivation after neoadjuvant hormonal ­cytoreduction and radiotherapy in locally advanced carcinoma of the prostate: the Radiation Therapy Oncology Group protocol 92-02. J Clin Oncol 2003;21:3972-3978. 350. Shearer RJ, Davies JH, Gelister JS, et al. Hormonal cytoreduction and radiotherapy for carcinoma of the prostate. Br J Urol 1992;69:521-524. 351. Yang FE, Chen GT, Ray P, et al. The potential for normal tissue dose reduction with neoadjuvant hormonal therapy in conformal treatment planning for stage C prostate cancer. Int J Radiat Oncol Biol Phys 1995;33:1009-1017. 352. Forman JD, Kumar R, Haas G, et al. Neo-adjuvant hormonal downsizing of localized carcinoma of the prostate: effects on the volume of normal tissue irradiation. Cancer Invest 1995;13: 132-133.

725

353. Zelefsky MJ, Leibel SA, Burman CM, et al. Neoadjuvant hormonal therapy improves the therapeutic ratio in patients with bulky prostatic cancer treated with three-dimensional conformal ­radiation therapy. Int J Radiat Oncol Biol Phys 1994;29:755-761. 354. Zagars GK, Johnson DE, von Eschenbach AC, et al. Adjuvant estrogen following radiation therapy for stage C adenocarcinoma of the prostate: long term results of a prospective randomized study. Int J Radiat Oncol Biol Phys 1988;14:1085-1091. 355. Fellows GJ, Clark PB, Beynon LL, et al. Treatment of advanced localized prostatic cancer by orchiectomy, radiotherapy, or combined treatment. A Medical Research Council study. Br J Urol 1992;70:304-309. 356. Pilepich MV, Caplan R, Byhardt RW, et al. Phase III trial of androgen suppression using goserelin in unfavorable-prognosis carcinoma of the prostate treated with definitive radiotherapy: report of the Radiation Therapy Oncology Group protocol 85-31. J Clin Oncol 1997;15:1013-1021. 357. Bolla M, Gonzalez D, Warde P, et al. Improved survival in ­patients with locally advanced prostate cancer treated with radiotherapy and goserelin. N Engl J Med 1997;337:295-300. 358. Granfors T, Modig H, Damber JE, et al. Combined orchiectomy and external radiotherapy versus radiotherapy alone for nonmetastatic prostate cancer with or without pelvic lymph node involvement: a prospective randomized trial. J Urol 1998;159:2030-2034. 359. Pilepich MV, Winter K, John MJ, et al. Phase III Radiation Therapy Oncology Group trial 86-10 of androgen deprivation adjuvant to definitive radiotherapy in locally advanced carcinoma of the prostate. Int J Radiat Oncol Biol Phys 2001;50:1243-1252. 360. Medical Research Council Prostate Cancer Working Investigators Group. Immediate versus deferred treatment for advanced prostatic cancer: initial results of the Medical Research Council trial. Br J Urol 1997;79:235-246. 361. Lawton CA, Winter K, Murray K, et al. Updated results of the phase III Radiation Therapy Oncology Group (RTOG) trial 85-31 evaluating the potential benefit of androgen suppression following standard radiation therapy for unfavorable carcinoma of the prostate. Int J Radiat Oncol Biol Phys 2001;49:937-946. 362. D’Amico A, Manola J, Loffredo M, et al. 6-month androgen suppression plus radiation therapy vs. radiation therapy alone for patients with clinically localized prostate cancer: a randomized controlled trial. JAMA 2004;292:821-827. 363. Del Regato J. Long term curative results of radiotherapy for patients with inoperable prostatic carcinoma. Radiology 1979;131:291-297. 364. Byar DP. The Veterans Administrative Cooperative Research Group’s studies of conservative treatment. Scand J Urol 1980;55(Suppl):99-102. 365. Pilepich M, Winter K, Lawton C, et al. Androgen suppression adjuvant to definitive radiotherapy in prostate carcinoma—longterm results of phase III RTOG 85-31. Int J Radiat Oncol Biol Phys 2005;61:1285-1290. 366. Bolla M, Collette L, Blank L, et al. Long-term results of immediate androgen suppression and external irradiation in patients with locally advanced prostate cancer (an EORTC study): a phase III randomized trial. Lancet 2002;360:103-108. 367. Byar DP, Green SB. The choice of treatment for cancer patients based on covariate information: application to prostate cancer. Bull Cancer 1988;67:477-490. 368. Oefelein MG, Feng A, Scolieri MJ, et al. Reassessment of the definition of castrate levels of testosterone: implications for clinical decision making. Urology 2000;56:1021-1024. 369. Nejat RJ, Rashid HH, Bagiella E, et al. A prospective analysis of time to normalization of serum testosterone after withdrawal of androgen deprivation therapy. J Urol 2000;164:1891-1894. 370. D’Amico AV, Schultz D, Loffredo M, et al. Biochemical outcome following external beam radiation therapy with or ­without androgen suppression therapy for clinically localized prostate cancer. JAMA 2000;284:1280-1283. 371. Granfors T, Tomic R, Bergh A, et al. After radiotherapy testosterone stimulation is unable to increase growth in the dunning R3327-PAP prostate tumour. Urol Res 1999;27:357-361.

726

PART 8  Male Genital Tract

372. Smith JA Jr, Haynes TH, Middleton RG. Impact of external irradiation on local symptoms and survival free of disease in patients with pelvic lymph node metastasis from adenocarcinoma of the prostate. J Urol 1984;131:705-707. 373. Bagshaw MA. Radiotherapeutic treatment of prostatic carcinoma with pelvic node involvement. Urol Clin North Am 1984;11:297-304. 374. Gervasi LA, Mata J, Easley JD, et al. Prognostic significance of lymph nodal metastases in prostate cancer. J Urol 1989;142: 332-336. 375. Lawton CA, Cox JD, Glisch C, et al. Is long-term survival possible with external beam irradiation for stage D1 adenocarcinoma of the prostate? Cancer 1992;69:2761-2766. 376. Hanks GE. The challenge of treating node-positive prostate cancer. An approach to resolving the questions. Cancer 1993;71:1014-1018. 377. Lawton CA, Winter K, Byhardt R, et al. Androgen suppression plus radiation versus radiation alone for patients with D1 (pN+) adenocarcinoma of the prostate (results based on a national prospective randomized trial, RTOG 85-31). Radiation Therapy Oncology Group. Int J Radiat Oncol Biol Phys 1997;38:931939. 378. deKernion JB, Neuwirth H, Stein A, et al. Prognosis of patients with stage D1 prostate carcinoma following radical prostatectomy with and without early endocrine therapy. J Urol 1990;144:700-703. 379. Myers RP, Larson-Keller JJ, Bergstralh EJ, et al. Hormonal treatment at time of radical retropubic prostatectomy for stage D1 prostate cancer: results of long-term followup. J Urol 1992;147:910-915. 380. Zincke H, Bergstralh EJ, Larson-Keller JJ, et al. Stage D1 prostate cancer treated by radical prostatectomy and adjuvant hormonal treatment: evidence for favorable survival in patients with DNA diploid tumors. Cancer 1992;70:311-323. 381. Sgrignoli AR, Walsh PC, Steinberg GD, et al. Prognostic factors in men with stage D1 prostate cancer: identification of patients less likely to have prolonged survival after radical prostatectomy. J Urol 1994;152:1077-1081. 382. Messing EM, Manola J, Sarosdy M, et al. Immediate hormonal therapy compared with observation after radical prostatectomy and pelvic lymphadenectomy in men with node-positive prostate cancer. N Engl J Med 1999;341:1781-1788. 383. Seay TM, Blute ML, Zincke H. Long-term outcome in patients with pTxN+ adenocarcinoma of prostate treated with radical prostatectomy and early androgen ablation. J Urol 1998;159: 357-364. 384. Zagars GK, Sands ME, Pollack A, et al. Early androgen ablation for stage D1 (N1-N3, M0) prostate cancer. Prognostic variables and outcome. J Urol 1994;151:1330-1333. 385. Zagars G, Pollack A, von Eschenbach A. Addition of radiotherapy to androgen ablation improves outcome for subclinically node-positive prostate cancer. Urology 2001;52:233-239. 386. Ghavamian R, Bergstralh EJ, Blute ML, et al. Radical retropubic prostatectomy plus orchiectomy versus orchiectomy alone for pTxN+ prostate cancer: a matched comparison. J Urol 1999;16:1223-1227. 387. Laramore GE, Krall JM, Thomas FJ, et al. Fast neutron radiotherapy for locally advanced prostate cancer. Am J Clin Oncol 1993;16:164-167. 388. Russell KJ, Caplan RJ, Laramore GE, et al. Photon versus fast neutron external beam radiotherapy in the treatment of locally advanced prostate cancer: results of a randomized prospective trial. Int J Radiat Oncol Biol Phys 1994;28:47-54. 389. Laramore GE. The use of neutrons in cancer therapy: a historical perspective through the modern era. Semin Oncol 1997;24: 672-685. 390. Lindsley KL, Cho P, Stelzer KJ, et al. Fast neutrons in prostatic adenocarcinomas: worldwide clinical experience. Recent Results Cancer Res 1998;150:125-136. 391. Chuba PJ, Maughan R, Forman JD. Three dimensional conformal neutron radiotherapy for prostate cancer. Strahlenther Onkol 1999;175:79-81.

392. Slater JD, Rossi CJ Jr, Yonemoto LT, et al. Conformal proton therapy for early-stage prostate cancer. Urology 1999;53: 978-984. 393. Kabalin JN, Hodge KK, McNeal JE, et al. Identification of ­residual cancer in the prostate following radiation therapy: role of transrectal ultrasound guided biopsy and prostate specific antigen. J Urol 1989;142:326-331. 394. Babaian RJ, Kohima M, Saitoh M, et al. Detection of residual prostate cancer after external radiotherapy. Role of prostate specific antigen and transrectal ultrasonography. Cancer 1995;75:2153-2158. 395. Cox JD, Stoffel TJ. The significance of needle biopsy after irradiation for the stage C adenocarcinoma of the prostate. Cancer 1977;40:156-160. 396. Herr HW, Whitmore WF Jr. Significance of prostatic biopsies after radiation therapy for carcinoma of the prostate. Prostate 1982;3:339-350. 397. Cox JD, Kline RW. The lack of prognostic significance of biopsies after radiotherapy for prostatic cancer. Semin Urol 1983;4: 237-242. 398. Crook JM, Perry GA, Robertson S, et al. Routine prostate biopsies following radiotherapy for prostate cancer: results for 226 patients. Urology 1995;45:624-632. 399. Crook J, Malone S, Perry G, et al. Postradiotherapy prostate ­biopsies: what do they really mean? Results for 498 patients. Int J Radiat Oncol Biol Phys 2000;48:355-367. 400. Sartor CI, Strawderman MH, Lin XH, et al. Rate of PSA rise predicts metastatic versus local recurrence after definitive radiotherapy. Int J Radiat Oncol Biol Phys 1997;38:941-947. 401. Zagars GK, Pollack A. Kinetics of serum prostate-specific antigen after external beam radiation for clinically localized prostate cancer. Radiother Oncol 1997;44:213-221. 402. Lee WR, Hanks GE, Hanlon A. Increasing prostate-specific antigen profile following definitive radiation therapy for localized prostate cancer: clinical observations. J Clin Oncol 1997;15: 230-238. 403. Amling CL, Lerner SE, Martin SK, et al. Deoxyribonucleic acid ploidy and serum prostate specific antigen predict outcome following salvage prostatectomy for radiation refractory prostate cancer. J Urol 1999;161:857-862. 404. Cheng L, Lloyd RV, Weaver AL, et al. The cell cycle inhibitors p21WAF1 and p27KIP1 are associated with survival in patients treated by salvage prostatectomy after radiation therapy. Clin Cancer Res 2000;6:1896-1899. 405. Rogers E, Ohori M, Kassabian VS, et al. Salvage radical prostatectomy: outcome measured by serum prostate specific antigen levels. J Urol 1995;153:104-110. 406. Garzotto M, Wajsman Z. Androgen deprivation with salvage surgery for radiorecurrent prostate cancer: results at 5-year followup. J Urol 1998;159:950-954. 407. Gheiler EL, Tefilli MV, Tiguert R, et al. Predictors for maximal outcome in patients undergoing salvage surgery for radio­recurrent prostate cancer. Urology 1998;51:789-795. 408. Vaidya A, Soloway MS. Salvage radical prostatectomy for radiorecurrent prostate cancer: morbidity revisited. J Urol 2000;164:1998-2001. 409. Pisters LL, English SF, Scott SM, et al. Salvage prostatectomy with continent catheterizable urinary reconstruction: a novel approach to recurrent prostate cancer after radiation therapy. J Urol 2000;163:1771-1774. 410. Grossfeld GD, Sitter DM, Flanders SC, et al. Use of second treatment following definitive local therapy for local prostate cancer: data from the CaPSURE database. J Urol 1998;160:1398-1404. 411. de la Taille A, Hayek O, Benson MC, et al. Salvage cryotherapy for recurrent prostate cancer after radiation therapy: the Columbia experience. Urology 2000;55:79-84. 412. Perrotte P, Litwin MS, McGuire EJ, et al. Quality of life after salvage cryotherapy: the impact of treatment parameters. J Urol 1999;162:398-402. 413. Pisters LL, von Eschenbach AC, Scott SM, et al. The efficacy and complications of salvage cryotherapy of the prostate. J Urol 1997;157:921-925.

28  The Prostate 414. Lee F, Bahn DK, McHugh TA, et al. Cryosurgery of prostate cancer. Use of adjuvant hormonal therapy and temperature monitoring—a one-year follow-up. Anticancer Res 1997;17:1511-1515. 415. Miller RJ Jr, Cohen JK, Shuman B, et al. Percutaneous, transperineal cryosurgery of the prostate as salvage therapy for postradiation recurrence of adenocarcinoma. Cancer 1996;77: 1510-1514. 416. Beyer DC. Permanent brachytherapy as salvage treatment for recurrent prostate cancer. Urology 1999;54:880-883. 417. Grado GL, Collins JM, Kriegshauser JS, et al. Salvage brachytherapy for localized prostate cancer after radiotherapy failure. Urology 1999;53:2-10. 418. Reiner WG, Walsh PC. An anatomical approach to the surgical management of the dorsal vein and Santorini’s plexus during radical retropubic surgery. J Urol 1979;121:198-200. 419. Walsh PC. Radical prostatectomy for localized prostate cancer provides durable cancer control with excellent quality of life: a structured debate. J Urol 2000;163:1802-1807. 420. Walsh PC, Marschke P, Ricker D, et al. Patient-reported urinary continence and sexual function after anatomic radical prostatectomy. Urology 2000;55:58-61. 421. Catalona WJ, Basler JW. Return of erections and urinary incontinence following nerve sparing radical retropubic prostatectomy. J Urol 1993;150:905-907. 422. Benoit RM, Naslund MJ, Cohen JK. Complications after radical retropubic prostatectomy in the Medicare population. Urology 2000;56:116-120. 423. Fowler FJ, Roman A, Barry MJ, et al. Patient-reported complications and follow-up treatment after radical prostatectomy, the national Medicare experience: 1988-1990. Urology 1993;42:622-629. 424. Jonler M, Messing EM, Rhodes PR, et al. Sequelae of radical prostatectomy. Br J Urol 1994;74:352-358. 425. Talcott JA, Rieker P, Propert KJ, et al. Patient-reported impotence and incontinence after nerve-sparing radical prostatectomy. J Natl Cancer Inst 1997;89:1117-1123. 426. Kim ED, Scardino PT, Hampel O, et al. Interposition of sural nerve restores function of cavernous nerves resected during radical prostatectomy. J Urol 1999;l161:188-192. 427. Jonler M, Ritter MA, Brinkmann R, et al. Sequelae of definitive radiation therapy for prostate cancer localized to the pelvis. Urology 1994;44:876-882. 428. Crook J, Esche B, Futter N. Effect of pelvic radiotherapy for prostate cancer on bowel, bladder, and sexual function: the ­patient’s perspective. Urology 1996;47:387-394. 429. Litwin MS, Pasta DJ, Yu J, et al. Urinary function and bother after radical prostatectomy or radiation for prostate cancer: a longitudinal, multivariate quality of life analysis from the cancer of the prostate strategic urologic research endeavor. J Urol 2000;164:1973-1977. 430. Iwakiri J, Grandbois K, Wehner N, et al. An analysis of urinary prostate specific antigen before and after radical prostatectomy: evidence for secretion of prostate specific antigen by the periurethral glands. J Urol 1993;149:783. 431. Trapasso JG, deKernion JB, Smith RB, et al. The incidence and significance of detectable levels of serum prostate specific antigen after radical prostatectomy. J Urol 1994;152:1821-1825. 432. Amling CL, Blute ML, Bergstralh EJ, et al. Long-term hazard of progression after radical prostatectomy for clinically localized prostate cancer: continued risk of biochemical failure after 5 years. J Urol 2000;164:101-105. 433. Vallancien G, Bochereau G, Wetzel O, et al. Influence of preoperative positive seminal vesicle biopsy on the staging of prostatic cancer. J Urol 1994;152:1152-1156. 434. Narayan P, Gajendran V, Taylor SP, et al. The role of transrectal ultrasound-guided biopsy-based staging, preoperative serum prostate-specific antigen, and biopsy Gleason score in prediction of final pathologic diagnosis in prostate cancer. Urology 1995;46:205-212. 435. Hanlon AL, Hanks GE. Failure pattern implications following external beam irradiation of prostate cancer: long-term follow-up and indications of cure. Cancer J Sci Am 2000; 2:S193-S197.

727

436. Hanks GE, Hanlon AL, Pinover WH, et al. Evidence for cure of young men with prostate cancer treated by external beam radiation. Oncology 2001;15:563-567. 437. Van Poppel H, De Ridder D, Elgamal AA, et al. Neoadjuvant hormonal therapy before radical prostatectomy decreases the number of positive surgical margins in stage T2 prostate cancer: interim results of a prospective randomized trial. J Urol 1995;154:429-434. 438. Soloway MS, Sharifi R, McLeod D, et al. Randomized prospective study—radical prostatectomy alone vs radical prostatectomy preceded by androgen ablation in cT2b prostate cancer—initial results. J Urol 1996;155(Suppl):555A. 439. Goldenberg SL, Klotz LH, Srigley J, et al. Randomized, prospective, controlled study comparing radical prostatectomy alone and neoadjuvant androgen withdrawal in the treatment of localized prostate cancer. Canadian Urologic Oncology Group. J Urol 1996;156:873-877. 440. Labrie F, Cusan L, Gomez JL, et al. Neoadjuvant hormonal therapy: the Canadian experience. Urology 1997;49(Suppl 3A):56-64. 441. Fair WR, Cookson MS, Stroumbakis N, et al. The indications, rationale, and results of neoadjuvant androgen deprivation in the treatment of prostatic cancer: Memorial Sloan-Kettering Cancer Center results. Urology 1997;49(Suppl 3A):46-55. 442. Witjes WP, Schulman CC, Frans MJ. Preliminary results of a prospective randomized study comparing radical prostatectomy versus radical prostatectomy associated with neoadjuvant hormonal combination therapy in T2-3 N0 M0 prostatic carcinoma. Urology 1997;49(Suppl 3A):65-69. 443. Aus G, Abrahamsson PA, Ahlgren G, et al. Hormonal treatment before radical prostatectomy: a 3-year follow-up. J Urol 1998;159:2013-2017. 444. Pollack A, Zagars GK, Rosen I. Androgen ablation in addition to radiotherapy for prostate cancer: is there true benefit? Semin Radiat Oncol 1998;8:95-106. 445. Pollack A. Does radiotherapy add anything to androgen ablation? In Taylor J, ed: American Society of Clinical Oncology Spring Educational Book. Baltimore: Lippincott Williams & Wilkins, 2000, pp 320-326. 446. Klotz LH, Goldenberg SL, Jewett M, et al. CUOG randomized trial of neoadjuvant androgen ablation before radical prostatectomy: 36-month post-treatment PSA results. Canadian Urologic Oncology Group. Urology 1999;53:757-763. 447. Baert L, Goethuys HJ, De Ridder DJ, et al. Neoadjuvant treatment before radical prostatectomy decreases the number of positive margins in cT2-cT3 but has no impact on PSA progression. J Urol 1998;159(Suppl):61. 448. Meyer F, Moore L, Bairati I, et al. Neoadjuvant hormonal therapy before radical prostatectomy and risk of prostate specific antigen failure. J Urol 1999;162:2024-2028. 449. Gleave ME, Goldenberg SL, Jones EC, et al. Biochemical and pathological effects of 8 months of neoadjuvant androgen withdrawal therapy before radical prostatectomy in patients with clinically confined prostate cancer. J Urol 1996;155:213-219. 450. Gleave ME, La Bianca SE, Goldenberg SL, et al. Long-term neoadjuvant hormone therapy prior to radical prostatectomy: evaluation of risk for biochemical recurrence at 5-year followup. Urology 2000;56:289-294. 451. van der Kwast TH, Tetu B, Candas B, et al. Prolonged neoadjuvant combined androgen blockade leads to a further reduction of prostatic tumor volume: three versus six months of endocrine therapy. Urology 1999;53:523-529. 452. Montironi R, Diamanti L, Santinelli A, et al. Effect of total ­androgen ablation on pathologic stage and resection limit status of prostate cancer. Initial results of the Italian PROSIT study. Pathol Res Pract 1999;195:201-208. 453. McCarthy JF, Catalona WJ, Hudson MA, et al. Effect of radiation therapy on detectable serum prostate specific antigen levels following radical prostatectomy: early versus delayed treatment. J Urol 1994;1515:1575-1578. 454. Coetzee LJ, Hars V, Paulson DF. Postoperative prostate-specific antigen as a prognostic indicator in patients with margin­positive prostate cancer undergoing adjuvant radiotherapy after radical prostatectomy. Urology 1996;47:232-235.

728

PART 8  Male Genital Tract

455. Garg MK, Tekyi-Mensah S, Bolton S, et al. Impact of postprostatectomy prostate-specific antigen nadir on outcomes following salvage radiotherapy. Urology 1998;51:998-1002. 456. Schild SE. Radiation therapy after prostatectomy: now or later? Semin Radiat Oncol 1998;8:132-139. 457. Anscher MS. Adjuvant therapy for pathologic stage C prostate cancer: a casualty of the PSA revolution? Int J Radiat Oncol Biol Phys 1996;34:745-747. 458. Tefilli MV, Gheiler EL, Tiguert R, et al. Prognostic indicators in patients with seminal vesicle involvement following radical prostatectomy for clinically localized prostate cancer. J Urol 1998;160:802-806. 459. Ohori M, Wheeler TM, Kattan MW, et al. Prognostic significance of positive surgical margins in radical prostatectomy specimens. J Urol 1995;154:1818-1824. 460. Epstein JI, Partin AW, Sauvageot J, et al. Prediction of progression following radical prostatectomy. A multivariate analysis of 721 men with long-term follow-up. Am J Surg Pathol 1996;20: 286-292. 461. Lowe BA, Lieberman SF. Disease recurrence and progression in untreated pathologic stage T3 prostate cancer: selecting the ­patient for adjuvant therapy. J Urol 1997;158:1452-1456. 462. Bauer JJ, Connelly RR, Seterheinn IA, et al. Biostatistical modeling using traditional preoperative and pathological prognostic variables in the selection of men at high risk for disease recurrence after radical prostatectomy for prostate cancer. J Urol 1998;159:929-933. 463. Gilliland FD, Hoffman RM, Hamilton A, et al. Predicting extra­ capsular extension of prostate cancer in men treated with radical prostatectomy: results from the population based prostate cancer outcomes study. J Urol 1999;162:1341-1345. 464. Cheng L, Slezak J, Bergstralh EJ, et al. Preoperative prediction of surgical margin status in patients with prostate cancer treated by radical prostatectomy. J Clin Oncol 2000;18:2862-2868. 465. Blute ML, Bergstralh EJ, Partin AW, et al. Validation of Partin tables for predicting pathological stage of clinically localized prostate cancer. J Urol 2000;164:1591-1595. 466. Pollack A, Smith LG. Adjuvant external beam radiotherapy. In D’Amico AV, Carroll P, Kantoff PW, eds: Prostate Cancer. Philadelphia: Lippincott, Williams & Wilkins, 2000, pp 456-463. 467. Epstein JI. Incidence and significance of positive margins in radical prostatectomy specimens. Urol Clin North Am 1996;23: 651-663. 468. Kupelian PA, Katcher J, Levin HS, et al. Stage T1-2 prostate cancer: a multivariate analysis of factors affecting biochemical and clinical failures after radical prostatectomy. Int J Radiat Oncol Biol Phys 1997;37:1043-1052. 469. Zietman AL, Edelstein RA, Coen JJ, et al. Radical prostatectomy for adenocarcinoma of the prostate: the influence of preoperative and pathologic findings on biochemical disease-free outcome. Urology 1994;43:828-833. 470. Watson RB, Civantos F, Soloway MS. Positive surgical margins with radical prostatectomy: detailed pathological analysis and prognosis. Urology 1996;48:80-90. 471. Grossfeld GD, Tigrani VS, Nudell D, et al. Management of a positive surgical margin after radical postatectomy: decision analysis. J Urol 2000;164:93-100. 472. Gibbons RP, Cole BS, Richardson RG, et al. Adjuvant radiotherapy following radical prostatectomy: results and complications. J Urol 1986;135:65-68. 473. Meier R, Mark R, St Royal L, et al. Postoperative radiation therapy after radical prostatectomy for prostate carcinoma. Cancer 1992;70:1960-1966. 474. Cheng WS, Frydenberg M, Bergstralh EJ, et al. Radical prostatectomy for pathologic stage C prostate cancer: influence of pathologic variables and adjuvant treatment on disease outcome. Urology 1993;42:283. 475. Anscher MS, Robertson CN, Prosnita LR. Adjuvant radiotherapy for pathologic stage T3/4 adenocarcinoma of the prostate: ten-year update. Int J Radiat Oncol Biol Phys 1995;33:37-43. 476. Syndikus I, Pickles T, Kostashuk E, et al. Postoperative radiotherapy for stage pT3 carcinoma of the prostate: improved local control. J Urol 1996;155:1983-1986.

477. Leibovich BC, Engen DE, Patterson DE, et al. Benefit of adjuvant radiation therapy for localized prostate cancer with a positive surgical margin. J Urol 2000;163:1178-1182. 478. Forman JD, Meetze K, Pontes E, et al. Therapeutic irradiation for patients with an elevated post-prostatectomy prostate specific antigen level. J Urol 1997;158:1436-1439. 479. Forman JD, Velasco J. Therapeutic radiation in patients with a rising post-prostatectomy PSA level. Oncology 1998;12:33-39. 480. Zietman AL, Coen JJ, Shipley WU, et al. Adjuvant irradiation ­after radical prostatectomy for adenocarcinoma of prostate: analysis of freedom from PSA failure. Urology 1993;42:292-299. 481. Schild SE, Buskirk SJ, Wong WW, et al. The use of radiotherapy for patients with isolated elevation of serum prostate ­ specific antigen following radical prostatectomy. J Urol 1996;156: 1725-1729. 482. Morris MM, Dallow KC, Zietman AL, et al. Adjuvant and salvage irradiation following radical prostatectomy for prostate cancer. Int J Radiat Oncol Biol Phys 1997;38:731-736. 483. Valicenti RK, Gomella LG, Ismail M, et al. Pathologic seminal vesicle invasion after radical prostatectomy for patients with prostate carcinoma: effect of early adjuvant radiation therapy on biochemical control. Cancer 1998;82:1909-1914. 484. Valicenti RK, Gomella LG, Ismail M, et al. The efficacy of early adjuvant radiation therapy for pT3N0 prostate cancer: a matched-pair analysis. Int J Radiat Oncol Biol Phys 1999; 45:53-58. 485. Vicini FA, Ziaja EL, Kestin LL, et al. Treatment outcome with adjuvant and salvage irradiation after radical prostatectomy for prostate cancer. Urology 1999;54:111-117. 486. Petrovich Z, Lieskovsky G, Langholz B, et al. Comparison of outcomes of radical prostatectomy with and without adjuvant pelvic irradiation in patients with pathologic stage C (T3N0) adenocarcinoma of the prostate. Am J Clin Oncol 1999;22: 323-331. 487. Nudell DM, Grossfeld GD, Weinberg VK, et al. Radiotherapy after radical prostatectomy: treatment outcomes and failure patterns. Urology 1999;54:1049-1057. 488. Valicenti RK, Gomella LG, Ismail M, et al. Durable efficacy of early postoperative radiation therapy for high-risk pT3N0 prostate cancer: the importance of radiation dose. Urology 1998;52:1034-1040. 489. Anscher MS, Prosnitz LR. Multivariate analysis of factors predicting local relapse after radical prostatectomy—possible indications for postoperative radiotherapy. Int J Radiat Oncol Biol Phys 1991;21:941-947. 490. Thompson IM Jr, Tangen CM, Paradelo J, et al. Adjuvant ­radiotherapy for pathologically advanced prostate cancer: a randomized clinical trial. JAMA 2006;296:2329-2335. 491. Bolla M, van Poppel H, Collette L, et al. Postoperative radiotherapy after radical prostatectomy: a randomised controlled trial (EORTC trial 22911). Lancet 2005;366:572-578. 492. Van der Kwast TH, Bolla M, Van Poppel H, et al. Identification of patients with prostate cancer who benefit from immediate postoperative radiotherapy: EORTC 22911. J Clin Oncol 2007;25:4178-4186. 493. Swanson GP, Hussey MA, Tangen CM, et al. Predominant treatment failure in postprostectomy patient is local: analysis of patterns of treatment failure in SWOG 8794. J Clin Oncol 2007;25:2225-2229. 494. Amling CL, Bergstralh EJ, Blute ML, et al. Defining prostate specific antigen progression after radical prostatectomy: what is the most appropriate cut point? J Urol 2001;165:1146-1151. 495. Stephenson AJ, Kattan MW, Eastham JA, et al. Defining biochemical recurrence of prostate cancer after radical prostat­ ectomy: a proposal for a standardized definition. J Clin Oncol 2006;24:3973-3978. 496. Wu JJ, King SC, Montana GS, et al. The efficacy of postprostatectomy radiotherapy in patients with an isolated elevation of serum prostate-specific antigen. Int J Radiat Oncol Biol Phys 1995;32:317-323. 497. Zelefsky MJ, Aschkenasy E, Kelsen S, et al. Tolerance and early ­outcome results of postprostatectomy three-dimensional conformal radiotherapy. Int J Radiat Oncol Biol Phys 1997;139:327-333.

28  The Prostate 498. Rogers R, Grossfeld GD, Roach M 3rd, et al. Radiation therapy for the management of biopsy proved local recurrence after radical prostatectomy. J Urol 1998;160:1748-1753. 499. Anscher MS, Clough R, Dodge R. Radiotherapy for a rising prostate-specific antigen after radical prostatectomy: the first 10 years. Int J Radiat Oncol Biol Phys 2000;48:369-375. 500. Pisansky TM, Kozelsky TF, Myers RP, et al. Radiotherapy for isolated serum prostate specific antigen elevation after prostatectomy for prostate cancer. J Urol 2000;163:845-850. 501. Stephenson AJ, Shariat SF, Zelefsky MJ, et al. Salvage radiotherapy for recurrent prostate cancer after radical prostatectomy. JAMA 2004;291:1325-1332. 502. Lee AK, D’Amico AV. Utility of prostate-specific antigen kinetics in addition to clinical factors in the selection of patients for salvage local therapy. J Clin Oncol 2005;23:8192-8197. 503. Nguyen PL, D’Amico AV, Lee AK, Suh WW. Patient selection, cancer control, and complications after salvage local therapy for postradiation prostate-specific antigen failure: a systematic ­review of the literature. Cancer 2007;110:1417-1428. 504. Stephenson AJ, Scardino PT, Kattan MW, et al. Predicting the outcome of salvage radiation therapy for recurrent prostate cancer after radical prostatectomy. J Clin Oncol 2007;25: 2035-2041.

729

505. Lee F, Bahn DK, Badalament RA, et al. Cryosurgery for prostate cancer: improved glandular ablation by use of 6 to 8 cryoprobes. Urology 1999;54:135-140. 506. Zisman A, Leibovici D, Siegel YI, et al. Prostate cryoablation without an insertion kit using direct transperineal placement of ultrathin freezing probes. Tech Urol 2000;6:34-36. 507. De La Taille A, Benson MC, Bagiella E, et al. Cryoablation for clinically localized prostate cancer using an argon-based system: complication rates and biochemical recurrence. BJU Int 2000;85: 281-286. 508. Long JP, Fallick ML, LaRock DR, et al. Preliminary outcomes following cryosurgical ablation of the prostate in patients with clinically localized prostate carcinoma. J Urol 1998;159: 477-484. 509. Gould RS. Total cryosurgery of the prostate versus standard ­cryosurgery versus radical prostatectomy: comparison of early results and the role of transurethral resection in cryosurgery. J Urol 1999;62:1653-1657. 510. Koppie TM, Shinohara K, Grossfeld GD, et al. The efficacy of cryosurgical ablation of prostate cancer:the University of ­CaliforniaSan Francisco experience. J Urol 1999;162:427-432.

The Uterine Cervix 29 Patricia J. Eifel EFFECTS OF RADIATION ON THE NORMAL CERVIX EPIDEMIOLOGY OF CERVICAL CANCER ANATOMY AND PATTERNS OF TUMOR PROGRESSION PATHOLOGY Cervical Intraepithelial Neoplasia Microinvasive Carcinoma Squamous Cell Carcinoma Adenocarcinoma Undifferentiated Neuroendocrine Carcinoma FIGO STAGING OF CERVICAL CANCER CLINICAL EVALUATION Pelvic Examination Additional Clinical Examinations Radiographic Evaluation SURGICAL LYMPH NODE EVALUATION PROGNOSTIC FACTORS Tumor Size and Local Extent Lymph Node Involvement Lymphovascular Space Invasion Histologic Type and Tumor Grade Other Tumor Factors Patient-Related Factors

TREATMENT OPTIONS Stage IA1 Disease Stage IA2 Disease Stage IB1 or Small Stage IIA Disease Stage IB2 or Minimal Stage IIA Disease Stage II to III Disease Stage IVA Disease Regional Disease Locally Recurrent Disease Inappropriate Simple Hysterectomy for Invasive Cancer Metastatic Disease: Stage IVB or Recurrent Disease RADIATION THERAPY External Beam Radiation Therapy Brachytherapy Treatment Duration CHEMORADIATION THERAPY Concurrent Chemotherapy and Radiation Therapy Neoadjuvant Chemotherapy Followed by Radiation Therapy COMPLICATIONS OF RADIATION THERAPY Acute Side Effects Late Complications

Radiation therapy plays a central role in the management of invasive carcinoma of the uterine cervix. Because the uterus and vagina tolerate high doses of radiation and form an ideal receptacle for radioactive sources, intracavitary brachytherapy can be exploited to achieve high rates of central disease control. The low frequency of extraregional disease spread at the time of initial diagnosis means that many patients with cervical carcinoma can be cured with a combination of regional external beam radiation therapy (EBRT) and brachytherapy. Randomized studies demonstrate that radiation therapy can be even more effective when chemotherapy is given concurrently. However, this multidisciplinary treatment requires skill and cooperation. Careful technique is required to minimize the risk of complications in normal structures that are close to the uterus, such as the bladder and rectum. In the United States and other Western countries, effective screening has made invasive cervical cancer a relatively uncommon problem and has contributed to dramatic reductions in cervical cancer mortality rates. Implementation of vaccines against the causative agent human papillomavirus (HPV) should further reduce the incidence and public health impact of cervical cancer, which remains the leading cause of cancer death for women in many ­developing

c­ ountries. As the incidence of invasive cervical cancer declines in various communities, clinicians are increasingly challenged to maintain their skills in the specialized techniques that are used to treat this disease.

Effects of Radiation on the Normal Cervix Reproductive function usually is not maintained after irradiation of the uterus and cervix. If the ovaries have been transposed and the dose of pelvic radiation is moderate (20 to 40 Gy), spontaneous pregnancy is possible, but few cases have been reported in the literature.1,2 In women irradiated during adulthood, these pregnancies may result in normal term infants. However, in women who underwent irradiation of the uterus and cervix before menarche, pregnancies often fail in the second trimester, possibly because of incomplete development of the uterine and cervical smooth muscle.3 High-dose EBRT or intracavitary radiation therapy (ICRT) causes an early inflammatory reaction with superficial ulceration.4 During the first 1 to 3 years after radiation therapy, gradual scarring of the cervix and upper vagina

733

734

PART 9  Female Genital Tract

occur with obliteration of the cervical canal. Although the cervix is obstructed, the uterus rarely accumulates enough fluid to cause symptoms. However, if patients are exposed to cyclical hormones, hematometra may occur. The cervix can usually be treated to very high doses (up to 150 Gy) without symptomatic effects (aside from loss of reproductive capacity). Minor clinically evident necrosis of the cervix occurs in 5% to 10% of patients, usually within the first 6 months after treatment.5 This usually heals after 3 to 6 months with conservative local management.

Epidemiology of Cervical Cancer Worldwide, cervical cancer is the second leading cause of cancer death in women (after breast cancer), with approximately 500,000 new cases diagnosed each year. However, international incidences vary widely, reflecting in part differences in cultural attitudes toward sexual promiscuity and the availability of programs to screen for and treat preinvasive disease. Overall age-adjusted incidence rates in ­developing countries are approximately twice those in developed countries, with particularly high rates in sub­Saharan Africa, South America, Central America, and Southeast Asia. Rates are particularly low in China, North Africa, and the Middle East. The American Cancer Society estimates that 11,070 new cases of cervical cancer were diagnosed and 3870 women died of cervical cancer in the United States in 2008.6 This means that deaths from cervical cancer represented less than 2% of cancer deaths of U.S. women in 2008. In part because of the widespread use of cytologic screening, incidence rates in the United States have declined steadily since the 1930s.7 Between 1992 and 1996, rates continued to decrease by an average of about 2.1% per year. However, the incidence among black women (11.2 of 100,000) was significantly higher than the incidence among white women (7.3 of 100,000). The incidence among Hispanic and Vietnamese American women, many of whom were recent immigrants from countries where the incidence of cervical cancer is high, was also high; from 1995 through 1999, the incidence of cervical cancer among Hispanic American women was 2.1 times that of non-Hispanic white women.8 Although the incidence of squamous cell carcinoma is decreasing in the United States, the incidence of adenocarcinoma seems to have increased, particularly among women younger than 40 years.9,10 Epidemiologic and molecular studies convincingly demonstrate that cervical cancer is a sexually transmitted disease. Increased risk correlates with young age at first ­intercourse, high number of sexual partners, a history of sexually transmitted disease, and exposure to sexually promiscuous male partners.11 Compelling evidence that genitally transmitted infection with certain strains of HPV plays a causative role in the development of cervical cancer helps to explain the traditional risk factors for this disease. Studies have identified HPV DNA in 95% to 100% of cervical carcinomas.12 Research suggests that genomic integration of HPV DNA causes persistent transcription of the viral E6 and E7 genes, disrupting cell cycle control mechanisms through functional inactivation of the tumor suppressor genes TP53 and RB1, respectively.13,14 The oncogenicity of various HPV strains correlates with the ability of their E6

or E7 oncoproteins to bind and inactivate tumor suppressor genes. Strains that are most often associated with cervical cancer in North America include HPV types 16, 18, and, less often, 45, 31, and 33; at least 10 other rare subtypes have also been identified.15 Types 6 and 11 are associated with benign venereal warts but rarely with cancer. The compelling evidence that HPV plays a causative role in the development of cervical cancer quickly led to attempts to develop a preventive vaccine. In 2002, Koutsky and colleagues16 published a landmark trial demonstrating that vaccination of women with HPV-16 L1 virus-like particles provided 100% protection against persistent HPV16 infection and HPV-16–related cervical intraepithelial neoplasia. Subsequent studies have confirmed the efficacy of HPV vaccines against HPV-16, HPV-18, and other subtypes.17 Ongoing studies are exploring the durability of protection, the optimal age for vaccination, and the possibility of a therapeutic vaccine.17 Although persistent HPV infection is a necessary factor in the development of most (probably all) cervical cancers, epidemiologic evidence suggests that the cause of cervical cancer is multifactorial. Only a small fraction of women infected with oncogenic strains of HPV develop cancer. Additional genetic alterations, immunologic conditions, or other unknown factors may contribute to the development of a malignant phenotype. Smoking has repeatedly been found to be an independent risk factor for invasive and preinvasive disease,18-20 and potent carcinogens have been found in the cervical mucus of smokers,21 suggesting that smoking may play a role. The incidence of cervical cancer also may be increased in women who are undergoing immunosuppressive therapy.22 In 1993, cervical cancer was added to the list of acquired immunodeficiency syndrome (AIDS)–defining illnesses by the U.S. Centers for Disease Control and Prevention. However, the relationship between the human immunodeficiency virus (HIV) and cervical cancer is complex and can be obscured by common risk factors for the two diseases.23 The risk of persistent HPV infection and the prevalence of preinvasive cervical lesions seems to be increased in HIV-infected women.24,25 Although analyses of the association between AIDS and invasive cervical cancer incidence have yielded conflicting results,23,26,27 several studies have suggested that invasive cervical cancer manifests at a more advanced stage and behaves more aggressively in immunosuppressed patients.28,29 Maiman and colleagues29 reported a recurrence rate of 88% among 28 HIV-positive women treated for cervical cancer. Cancer was the immediate cause of death for 95% of these women, underscoring the importance of careful screening and effective treatment of localized disease in this population.29 A correlation between prolonged oral contraceptive use and cervical adenocarcinoma has been suggested, but a causative role for oral contraceptives has not been proven.30,31

Anatomy and Patterns of Tumor Progression The uterus is a flattened, muscular, thick-walled, hollow structure that lies in the true pelvis between the bladder and the rectosigmoid (Fig. 29-1). The uterus is divided by

29  The Uterine Cervix Suspensory ligament of ovary

735

Right ureter Ovary

Uterine tube External iliac vessels Fundus of uterus Vesicouterine recess Bladder Urethra Vagina

Ligament of ovary Rectouterine fold Rectouterine recess Posterior part of fornix Cervix uteri Rectal ampulla Anal canal

FIGURE 29-1.  Median sagittal section of the female pelvis. (From Gray’s Anatomy. Philadelphia: WB Saunders, 1980.)

the internal os into two regions—the corpus, or body, and the cervix. The uterus is most commonly anteverted (i.e., with its long axis approximately perpendicular to the axis of the vagina) and anteflexed, although it may be axial, retroverted, or retroflexed. The normal cervix is about 2.5 cm long and projects into the anterior wall of the vagina, which divides the cervix into the upper (supravaginal) cervix and the lower (vaginal) cervix, also known as the portio vaginalis cervicis. The external os is visible as a depression on the vaginal cervix. Most cervical carcinomas arise at the junction between the primarily columnar epithelium of the endocervix and the squamous epithelium of the ectocervix. Tumors that penetrate the basement membrane to invade the muscular stroma may remain clinically confined to the cervix until they are quite large, forming an exophytic, cauliflowershaped mass that protrudes into the vagina or causing an expanded, barrel-shaped cervix. However, they also may infiltrate beyond the cervix into adjacent structures. Exterior to the smooth muscle of the uterine body and cervix is a layer of visceral pelvic fascia called the parametrium (where the fascia surrounds the corpus) or paracervix (where the fascia surrounds the cervix). The uterine fundus is covered by the peritoneum, which extends posteriorly to cover the upper cervix and vaginal fornix and then on to the anterior rectum to form the rectouterine pouch (pouch of Douglas) or cul-de-sac. Although large cervical tumors may erupt into the cul-de-sac, direct invasion of the rectum is rare. Anteriorly, the peritoneum reflects from the fundus onto the posterior bladder at the level of the internal cervical os. The supravaginal cervix is below the peritoneal reflection and is separated from the bladder by only a relatively thin layer of paracervical fascia and cellular connective tissue. This layer may be traversed by deeply invasive cancer, although less than 5% of patients with cervical cancer present with cystoscopic evidence of bladder mucosal involvement. Although some clinical texts mistakenly refer to the junction between the cervical stroma and surrounding connective tissue as “serosa,” the cervix lacks a true serosal layer except in its posterior supravaginal portion, where the cervix is covered by the peritoneum. The uterus is connected to surrounding structures by a number of peritoneal folds and ligaments that can serve as conduits for direct extension of cancer from the cervix.

The uterus is supported by the cardinal (transverse), uterosacral, and pubocervical ligaments, which are attached ­peripherally to the lateral pelvic wall, sacrum, and pubis, respectively. These ligaments are formed of dense connective tissue that is continuous with the fibrous tissue of the paracervix. The broad ligaments are formed from peritoneal folds that extend from the sides of the uterus to the lateral pelvic wall. With the uterus, the broad ligaments divide the pelvis into anterior and posterior compartments containing the bladder and rectum, respectively. The uterine ­arteries pass between the two layers of the broad ligaments at their inferior borders, and the fallopian tubes are located at the upper, free borders of the broad ligaments. The broad ligaments also enclose the ovarian ligaments and proximal portions of the round ligaments. The cardinal ligaments consist of bands of dense fibroconnective tissue at the base of the broad ligaments; they connect the fascia of the lateral cervix and upper vagina with the fibrous tissue that surrounds the lateral pelvic vessels. The uterosacral ligaments lie within rectouterine peritoneal folds that extend posteriorly from the supravaginal cervix around either side of the rectum. Tumors that penetrate the paracervical fascia can involve the tissues of the broad ligaments or rectouterine spaces by direct extension; tumors also can gain access to these regions by lymphatic extension, particularly to nodes adjacent to the uterine vessels. The cervical mucosa and stroma are richly supplied with lymphatics that form extensive anastomoses with those of the lower uterine segment.32 These lymphatics penetrate the cervical stroma roughly perpendicularly and then spread out in a branching network of flattened lacunae that lie at the junction between the cervix and the surrounding connective tissue (or the peritoneum). The cervical lymphatics eventually merge to form collecting trunks, the most important of which exit laterally from the uterine isthmus in three groups (Fig. 29-2). Upper branches follow the uterine artery and terminate in the uppermost hypogastric nodes. Middle branches drain to deeper hypogastric (obturator) nodes, and the lowest channels course posteriorly to the inferior and superior gluteal, common iliac, and presacral nodes. Other posterior lymphatic channels arising from the posterior lip of the cervix drain to superior rectal nodes or pass through the retrorectal space to subaortic nodes that lie at the level of the sacral promontory. Some

736

PART 9  Female Genital Tract

Para-aortic nodes L5 Common iliac nodes Hypogastric nodes External iliac nodes

Obturator nodes

FIGURE 29-2.  Lymphatic drainage of the uterus and cervix. (From Gray’s Anatomy. Philadelphia: WB Saunders, 1980.)

anterior channels also parallel the course of the superior vesical arteries to terminate in the internal iliac nodes. In summary, the primary regional drainage of the cervix includes most of the lymph nodes of the pelvis. Tumors that invade the posterior vaginal wall may involve even the more inferior perirectal nodes, and tumors that invade the lower vagina may involve the inguinal nodes. However, unlike uterine malignancies, which may drain directly to the upper para-aortic lymph nodes (through the tubo-ovarian vessels), cervical cancer usually involves the pelvic lymph nodes first. The most common sites of distant metastasis from cervical cancer are the lungs and the secondary (e.g., para-­aortic) or other lymph nodes. Less commonly, cervical ­cancer may spread to liver, bone, and other sites.

Pathology Cervical Intraepithelial Neoplasia It was first suggested more than 100 years ago that invasive cervical cancer arose from preexisting intraepithelial lesions, subsequently called carcinomas in situ. In the 1940s, Papanicolaou and Traut33 reported a method of screening for these lesions using cytologic samples that could be readily collected from the cervix and vagina. The dramatic reduction in the incidence of invasive cervical cancer that came with successful screening for and treatment of premalignant cervical lesions supported the hypothesis that most invasive malignancies begin with a long preinvasive phase. The fact that the mean age of women with intraepithelial lesions is about 15 years younger than the mean age of women with invasive cancer also suggests that the progression to invasive disease is slow.34,35 Longitudinal

studies of patients followed for untreated preinvasive lesions confirm this but also suggest that less than 15% of high-grade preinvasive lesions progress to invasive carcinoma and that approximately one half undergo spontaneous regression.36 Early systems classed atypical lesions as dysplasias of various degrees (classes II to III), carcinoma in situ (class IV), or cancer (class V) (Table 29-1).20,37 In the 1960s, this method was replaced by the cervical intraepithelial neoplasia system, proposed by Richart and Barron,37 that more clearly described the spectrum of preinvasive squamous lesions. Although useful, cervical intraepithelial neoplasia grading was subjective and did not make the potentially important distinction between HPV-related koilocytotic atypia and other types of inflammatory change. In 1988, after a National Institutes of Health consensus conference, the Bethesda System was proposed to reduce ambiguities and improve the clinical relevance of cytologic classification of cervical lesions; with a few subsequent modifications,20 the Bethesda System is now used by most U.S. laboratories. According to this system, pathology reports must include a statement on the adequacy of the specimen; a general categorization to assist with triage (e.g., “within normal limits,” “infection or reactive changes,” “epithelial abnormality”); and a descriptive diagnosis. Squamous epithelial abnormalities are divided into four general categories: atypical squamous cells of undetermined significance (ASCUS), low-grade squamous intraepithelial lesions, high-grade squamous intraepithelial lesions (HSILs), and carcinomas. The ASCUS category, introduced to acknowledge limitations in the ability to identify some lesions as reactive or neoplastic with certainty, has become the most common abnormal Papanicolaou (Pap) test result.38

29  The Uterine Cervix

737

TABLE 29-1. Comparison of Systems Used to Classify Cervical Cytology Bethesda System

Dysplasia/CIN System

Papanicolaou System

Within normal limits

Normal

Class I

Infection (specify organism)

Inflammatory atypia (­organism)

Class II

Reactive and reparative changes Squamous cell abnormalities Atypical squamous cells of undetermined significance

Squamous atypia Class IIR HPV atypia

Low-grade squamous intraepithelial lesion

High-grade squamous intraepithelial lesion

Invasive squamous carcinoma

Mild dysplasia

CIN 1

Moderate ­dysplasia

CIN 2

Class III

Severe ­dysplasia CIN 3 Carcinoma in situ

Class IV

Invasive squamous carcinoma

Class V

CIN, cervical intraepithelial neoplasia; HPV, human papillomavirus. Data from Schiffman MH. Recent progress in defining the epidemiology of human papillomavirus infection and cervical neoplasia. J Natl Cancer Inst 1992;84:394-398; Richart RM, Barron BA. A follow-up study of patients with cervical dysplasia. Am J Obstet Gynecol 1969;105:386-393.

Although most ASCUS lesions are benign, patients with this finding should have a repeat Pap smear within 6 months because about 10% to 15% of cases of ASCUS are associated with underlying HSIL.39 However, concern about possible false-negative diagnoses has led many clinicians to perform colposcopy on all patients with a diagnosis of ASCUS, significantly increasing the cost of screening and the chance of overtreatment. This practice has led to the search for better methods of triage. Although HPV testing is not recommended for primary screening, studies have suggested that HPV testing of ASCUS smears followed by colposcopy in patients with HPV-positive lesions could be a highly accurate and cost-effective means of detecting underlying HSILs despite equivocal smears.40 Adenocarcinoma in situ (AIS) is diagnosed when malignant cells have replaced the normal endocervical glandular epithelium but the normal branching architecture is maintained and the cells do not invade the cervical stroma.41 High-risk HPV subtypes are often detected in AIS and invasive adenocarcinomas, and HSILs are often found adjacent to areas of AIS. Because AIS is commonly multifocal, residual AIS may remain even after a cone biopsy with negative margins.42

Microinvasive Carcinoma When it occurs, early stromal invasion is usually found close to the transformation zone adjacent to areas of squamous cervical intraepithelial neoplasia. Because the diagnosis of microinvasion is based on a quantitative assessment of the size and depth of the invasive lesion, diagnosis requires a cone biopsy specimen that encompasses the entire lesion and cervical transformation zone. The depth of invasion should be measured from the base of the epithelium to the deepest point of invasion. Although the International Federation of Gynecology and Obstetrics (FIGO) first included a category for ­minimally

invasive lesions in 1962, the definition was left nonspecific until 1995, when the current definitions of stage IA1 disease (invasion ≤ 3.0 mm deep) and stage IA2 disease (invasion > 3.0 mm and ≤ 5.0 mm deep) were adopted. In the interim, the Society of Gynecologic ­ Oncologists adopted a different standard that defined microinvasive lesions as those that invaded less than 3 mm into the cervical stroma without evidence of lymphovascular space invasion (LVSI). Although many clinicians have come to use the FIGO definitions, lingering use of the Society of Gynecologic Oncologists’ definition can still lead to some confusion (see “FIGO Staging of Cervical Cancer”). Because early invasive adenocarcinomas can arise from deep endocervical glandular epithelium, the depth of invasion can be difficult to determine. For this reason, such lesions are usually classified as FIGO stage IB.

Squamous Cell Carcinoma More than 75% of cases of cervical cancer are squamous cell carcinomas. Large-cell nonkeratinizing carcinomas are most common and are characterized by nests of neoplastic cells that may stain for keratin but lack the keratin pearls characteristic of large-cell keratinizing tumors. About 5% of cervical squamous cell carcinomas are characterized by relatively small cells that demonstrate minimal squamous differentiation, scant cytoplasm, small nucleoli, and abundant mitoses. These tumors have traditionally been called small cell squamous cell carcinomas, but this term should probably be avoided because it is easily confused with undifferentiated small cell carcinoma, which is associated with a much poorer prognosis (discussed in the next section). Rare variants of squamous cell carcinoma include verrucous carcinomas, well-differentiated papillomatous lesions that tend to be locally invasive but rarely metastasize,43 and papillary squamous cell carcinomas,44 which form papillae of atypical epithelium with frequent invasion at the base of the tumors.

738

PART 9  Female Genital Tract

Adenocarcinoma Microscopically, adenocarcinomas of the cervix demonstrate a variety of patterns and degrees of histologic differentiation. Mucinous (endocervical) carcinomas are most common, although other müllerian carcinomas (i.e., endometrioid, clear cell, and serous adenocarcinomas) also can originate in the cervix. Mixtures of these cell types are possible. Adenosquamous carcinomas contain malignant glandular and squamous elements. Reports conflict concerning the relative prognoses for adenocarcinomas and adenosquamous carcinomas of similar grade. Although histologic grade does not clearly influence ­outcome for patients with squamous cell carcinomas, it correlates strongly with outcome for patients with ­adenocarcinoma.45,46 Many rare subtypes of adenocarcinoma have been reported in the cervix, including signet-ring, glassy cell, adenoid cystic, adenoid basal, and villoglandular carcinomas. Minimal-deviation adenocarcinoma (adenoma malignum) is another rare type of carcinoma with an extremely welldifferentiated histologic appearance that can be mistakenly interpreted as benign on the basis of small biopsy specimens. Although early reports suggest that these tumors are associated with a poor prognosis and frequent local recurrence, later studies suggest that early lesions can be cured with local treatment.47

Undifferentiated Neuroendocrine Carcinoma Undifferentiated small cell or neuroendocrine carcinoma is a rare subtype of cervical cancer, but it is important ­because of its aggressive behavior. A very poorly ­differentiated ­carcinoma with neuroendocrine features, this type of tumor is histologically similar to anaplastic small cell carcinoma of the lung.48 Another rare entity, large cell neuroendocrine carcinoma, also has a poor prognosis.49 Neuroendocrine carcinomas tend to behave very aggressively, with frequent widespread metastases to many sites, including bone, liver, skin, and other organs. Brain metastases can occur when disease is advanced, but they are usually preceded by lung metastases.50 Efforts to treat these tumors by using approaches typically used for small cell carcinomas of the lung have had mixed results.50 Undifferentiated small cell carcinomas should not be confused with small cell squamous cell carcinomas, which have a better prognosis.

FIGO Staging of Cervical Cancer The FIGO clinical staging system51,52 (Box 29-1) is the internationally accepted method used to classify patients with cervical cancer. The four major stage groupings, ­designed to

BOX 29-1.  American Joint Committee on Cancer and International Federation of Gynecology and Obstetrics Staging Systems for Cancer of the Uterine Cervix AJCC TNM Categories

FIGO Stages

Description

TX T0 Tis T1 T1a

— — 0 I IA

T1a1 T1a2

IA1 IA2

T1b T1b1 T1b2 T2 T2a T2b T3

IB IB1 IB2 II IIA IIB III

T3a T3b T4

IIIA IIIB IVA

Primary tumor cannot be assessed No evidence of primary tumor Carcinoma in situ Cervical carcinoma confined to uterus (extension to corpus should be ­disregarded) Invasive carcinoma diagnosed only by microscopy. All macroscopically visible lesions—even with superficial invasion—are T1b/IB. Stromal invasion with a maximum depth of 5.0 mm measured from the base of the epithelium and a horizontal spread of 7.0 mm or less. Vascular space invasion, venous or ­lymphatic, does not affect classification Measured stromal invasion 3.0 mm or less in depth and 7.0 mm or less in ­horizontal spread Measured stromal invasion more than 3.0 mm and not more than 5.0 mm with a horizontal spread 7.0 mm or less Clinically visible lesion confined to the cervix or microscopic lesion greater than T1a/IA2 Clinically visible lesion 4.0 cm or less in greatest dimension Clinically visible lesion more than 4.0 cm in greatest dimension Cervical carcinoma invades beyond uterus but not to pelvic wall or to lower third of vagina Tumor without parametrial invasion Tumor with parametrial invasion Tumor extends to the pelvic wall and/or involves lower third of vagina, and/or causes hydronephrosis or nonfunctioning kidney Tumor involves lower third of the vagina, no extension to pelvic wall Tumor extends to pelvic wall and/or causes hydronephrosis or nonfunctioning kidney Tumor invades mucosa of the bladder or rectum, and/or extends beyond true pelvis (bullous edema is not sufficient to classify a tumor as T4)

Regional Lymph Nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Regional lymph node metastasis Distant Metastasis (M) MX Distant metastasis cannot be assessed M0 No distant metastasis M1 IVB Distant metastasis

29  The Uterine Cervix

739

BOX 29-1.  American Joint Committee on Cancer and International Federation of Gynecology and Obstetrics Staging Systems for Cancer of the Uterine Cervix—cont’d Stage Grouping Stage 0 Stage I Stage IA Stage IA1 Stage IA2 Stage IB Stage IB1 Stage IB2 Stage II Stage IIA Stage IIB Stage III Stage IIIA Stage IIIB

Stage IVA Stage IVB

Tis T1 T1a T1a1 T1a2 T1b T1b1 T1b2 T2 T2a T2b T3 T3a T1 T2 T3a T3b T4 Any T

N0 N0 N0 N0 N0 N0 N0 N0 N0 N0 N0 N0 N0 N1 N1 N1 Any N Any N Any N

M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M1

Notes about the Staging System* Stage 0 cases are those with full-thickness involvement of the epithelium with atypical cells but with no signs of invasion into the stroma. Cases of stage 0 disease should not be included in any therapeutic statistics for invasive carcinoma. It is usually impossible to estimate clinically whether a cancer of the cervix has extended to the corpus. Extension to the corpus should therefore be disregarded. A growth fixed to the pelvic wall by a short and indurated but not nodular parametrium should be assigned stage IIB. It is impossible at clinical examination to decide whether a smooth and indurated parametrium is truly cancerous or only inflammatory, and such cases should be classified as stage III only if the parametrium is nodular to the pelvic wall or if the growth itself extends to the pelvic wall. The presence of hydronephrosis or nonfunctioning kidney caused by stenosis of the ureter by cancer permits a case to be classified as stage III even if the case should be classified as stage I or stage II according to the other findings. The presence of bullous edema should not permit a case to be classified as stage IV. Ridges and furrows into the bladder wall should be interpreted as signs of submucous involvement of the bladder if they remain fixed to the growth at palposcopy (examination from the vagina or the rectum during cystoscopy). A finding of malignant cells in cytologic washings from the ­urinary bladder requires further examination and biopsy from the wall of the bladder. Rules for Clinical Staging The staging should be based on careful clinical examination and should be performed before any definitive therapy. Ideally, the examination should be performed under anesthesia by an experienced examiner. The clinical stage must under no circumstances be changed on the basis of subsequent findings. When doubt exists about the assignment of stage, the case must be classified as the earlier stage. The following examination methods are permitted for staging purposes: palpation, inspection, colposcopy, endocervical curettage, hysteroscopy, cystoscopy, proctoscopy, intravenous urography, and x-ray examination of the lungs and skeleton. Suspected bladder or rectal involvement should be confirmed by biopsy and histologic evidence. Findings on examinations such as lymphangiography, arteriography, venography, and laparoscopy are valuable for planning therapy, but because these studies are not generally available and because interpretation of the results varies, the findings of such studies should not be the basis for changing the clinical staging. Infrequently, hysterectomy is performed in the presence of unsuspected extensive invasive cervical carcinoma. These cases cannot be clinically staged or included in therapeutic statistics, but they should be reported separately. Only strict observance of the rules for clinical staging allows meaningful comparison of results between clinics and modes of therapy. *Notes are from the International Federation of Gynecology and Obstetrics. Annual Report on the Results of Treatment in Gynecological ­Cancer. Radiumhemmet, Stockholm: FIGO, 1994. From Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 259-265.

reflect the extent of local disease and the feasibility of successful surgical resection, are similar to those set forth in the 1938 staging system of the League of ­Nations Health Organization.53 Since then, FIGO has made many modifications that have altered the compositions of stage groups or defined various substages. Preinvasive lesions were included in stage I until 1950, when they were placed in a new (stage 0) category. Hydronephrosis was first included

as a criterion for labeling a tumor as stage III in 1976. Many efforts have been made to define subgroups of patients with more favorable “occult” or microinvasive disease. These descriptions were ambiguous until 1994, when the current definitions of microinvasion (stages IA1 and IA2) were added. All of these changes have created opportunities for stage migration, confounding comparisons between different ­experiences.

740

PART 9  Female Genital Tract

The staging system has been criticized for not including information about important predictors of outcome such as tumor size and lymph node status. In a significant improvement, FIGO first divided the stage IB category according to cervical diameter in 1994 (see Box 29-1), although substantial size heterogeneity still exists within the IB1 and IB2 groups and other stage groups. Radiographic and surgical assessments of lymph nodes have never been incorporated into the staging system for two important reasons. First, the expense of these studies is often prohibitive in the medically underserved countries where most of the world’s cases of cervical cancer occur. Second, the lack of a sensitive, widely accepted method of evaluating the pelvic and aortic lymph nodes has made it difficult to design a staging system that would document regional metastasis in a consistent fashion. The subjectivity of pelvic examination, the primary method used to define stage groups, creates hidden biases. Despite admonitions in FIGO’s staging notes to discount equivocal findings, clinicians tend to overstage tumors. For example, in a large surgical series, approximately two thirds of the cases labeled as stage IIB had no evidence of parametrial involvement in the hysterectomy specimens (Fig. 29-3).54 Comparisons of clinical stage with the findings from carefully sectioned hysterectomy specimens demonstrate substantial differences between institutions in the tendency to overstage or understage tumors.54 This is another important source of stage migration that influences the reporting of results. FIGO stage does correlate roughly with outcome, and despite the system’s flaws, it is the most consistent language currently used to compare results between institutions. For this reason, the staging system should always be applied strictly according to the rules defined by FIGO (see Box 29-1). When studies that are not used in staging are incorporated into treatment planning or patient selection, they should be carefully described in the reports of clinical trials. Only consistent staging and detailed description of methods will improve clinicians’ ability to determine the generalizability of studies to their own patients. The American Joint Committee on Cancer (AJCC) and FIGO suggest a tumor-node-metastasis (TNM) pathologic staging system that may be used to report surgical findings for patients treated with radical hysterectomy and lymph node dissection. This system cannot be applied to patients treated with initial irradiation. Although FIGO and the AJCC suggest a method for applying the TNM categories to stage groups, this approach causes confusion when patients with clinically staged disease are mixed with others whose disease has been classified according to pathologic stage. This inconsistent assignment of stage groupings becomes a particular problem in large epidemiologic studies that use tumor registries or multi-institutional databases to compare treatments according to stage.

Clinical Evaluation Every patient with cervical cancer should be carefully interviewed to obtain a history of the present illness and any other medical problems. Previous surgical procedures, pelvic infections, medical illnesses, habits (including smoking history), and other conditions that may influence the patient’s tolerance of treatment and risk of complications

MUNICH

350 300

Clinical stage Pathologic stage

250 200 150 100 50 0 IB

IIA

IIB

GRAZ 300 250 200 150 100 50 0 IB

IIA

IIB

FIGURE 29-3.  International Federation of Gynecology and Obstetrics (FIGO) clinical and pathologic stages for patients whose radical hysterectomy specimens were reviewed in two regional centers. Similar examination methods were used at the two centers. In Munich, findings on clinical examination predicted parametrial involvement in only about 50% of cases that were judged to be stage IIB on the basis of pathologic findings. In Graz, more than one half of the cases diagnosed clinically as FIGO stage IIB had no histologic evidence of parametrial involvement. (From Burghardt E, Hofmann HMH, Ebner F, et al. Results of surgical treatment of 1028 cervical cancers studied with volumetry. Cancer 1992;70:648-655.)

should be documented. The patient’s diet and medications and any herbal or complementary health products she is using should be documented because they can interact with treatment to worsen side effects and compromise the therapeutic benefit. An early assessment of the patient’s social circumstances is particularly important because women with cervical cancer often have distressed finances or young dependents and limited family support. Cultural, educational, and language differences between the patient and her physician can interfere with effective communication. Early recognition of any problems and appropriate intervention can greatly facilitate a patient’s ability to comply with the complicated treatment regimens used to treat cervical cancer.

Pelvic Examination Each patient should have a thorough physical examination, including a detailed pelvic examination. The findings should be documented in a tumor diagram. Pelvic examination should include careful inspection of the external

29  The Uterine Cervix

genitalia, vagina, and cervix. The capacity of the vagina should be assessed digitally before a speculum is inserted, and the speculum should be advanced carefully to minimize trauma to friable tumor. If a histologic diagnosis has not been made, a biopsy should be performed. If the lesion is not visible to the naked eye, colposcopy may be helpful. All areas of vaginal mucosa should be systematically visualized as the speculum is withdrawn. The pelvic examination should include a thorough digital vaginal examination, during which any vaginal abnormalities should be documented and an initial assessment of the size and morphology of the cervical tumor made. This should be done with as little trauma to the tumor as possible. With one or two fingers in the vagina in contact with the cervix and the other hand on the suprapubic abdomen, the physician should assess the uterus for its size, shape, mobility, and position in the pelvis. The adnexa should be carefully examined for masses or tenderness. A rectovaginal examination should then be performed to determine whether disease is breaking through the paracervix into the broad ligaments or uterosacral ligaments. The mobility of the uterus and any nodularity should be documented. The size of the cervix is evaluated by comparing its diameter with the width of the pelvis (usually about 12 cm at the level of the cervix). If significant bleeding follows the pelvic examination, a speculum should be reinserted immediately, and Monsel’s solution (ferric subsulfate solution) or vaginal packing should be applied to minimize continued blood loss.

Additional Clinical Examinations Cystoscopy and proctoscopy should be performed if symptoms, clinical examination findings, or magnetic resonance imaging (MRI) scan results suggest possible organ involvement. Suspected bladder or rectal involvement identified on MRI or computed tomography (CT) must be confirmed histologically before the FIGO stage can be increased (see Box 29-1).51 If MRI cannot be used, cystoscopy and proctoscopy may be indicated if bulky central disease or extensive vaginal involvement is present or if the patient has symptoms suggesting possible bladder or rectal ­abnormalities. All patients with cervical cancer should have a complete blood cell count and serum electrolyte, blood urea nitrogen, and creatinine levels measured at diagnosis. Additional studies may be required, depending on the patient’s symptoms and any concurrent illnesses.

Radiographic Evaluation Intravenous pyelography and radiography of the chest and bones are the only radiographic examinations that can be used to determine FIGO stage (see Box 29-1). However, unless the patient has very early disease (stage IA or IB1), additional radiographic studies are usually required to obtain important information about possible regional ­involvement. Clinicians often obtain a contrast-enhanced CT scan of the abdomen and pelvis for patients with stage IB2 or more advanced disease.55 Lymph nodes larger than 1.0 to 1.5 cm in diameter should raise suspicion of tumor involvement and should be biopsied. Unfortunately, microscopic metastases are not readily detected with

741

FIGURE 29-4.  Sagittal magnetic resonance imaging (MRI) scan of a large squamous cell carcinoma of the cervix. On clinical examination, this tumor was estimated to be an exophytic lesion, FIGO stage IIA, about 8 cm in diameter. MRI showed an 8 × 7 × 7.5 cm tumor with extensive invasion of the vagina (arrows), endocervix, and lower uterine segment with bulky pelvic lymphadenopathy.

CT, and the inflammation commonly associated with advanced disease may cause enlargement of nodes that do not contain metastases. The sensitivity of MRI in the detection of regional metastases is similar to that of CT. However, MRI provides more detailed images of the cervix and paracervical tissues (Fig. 29-4). Sagittal views are particularly helpful for determining the morphology of the tumor and the position of the uterus. MRI also yields a more objective assessment of tumor diameter than does pelvic examination. However, in medically underserved communities, particularly in developing nations, the cost of MRI may outweigh the benefit. For practical reasons, CT or MRI assessment of the ureters and renal pelvis is often substituted for traditional intravenous pyelography to determine whether a patient has tumor-related ­hydronephrosis. Studies have suggested that 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) is more sensitive and specific than standard radiographic studies for the ­detection of lymph node involvement by cervical cancer.56-58 Grigsby and colleagues59 reported a strong correlation between abnormal FDG uptake after therapy and tumor recurrence. However, inconsistent reimbursement for FDGPET continues to be a barrier to its widespread use for patients with cervical cancer in the United States, and its cost prevents its widespread international use for medically underserved patients.

742

PART 9  Female Genital Tract

Surgical Lymph Node Evaluation Although surgical evaluation is the most sensitive method of evaluating whether regional lymph nodes contain metastases, it is invasive and expensive, and it delays treatment of the primary lesion. Transperitoneal lymphadenectomy, the surgical staging approach most often used in the past, was associated with a high incidence of severe, even fatal, complications in patients treated with radiation therapy. The risk of severe complications seems to be much less when lymphadenectomy is performed using a retroperitoneal approach.60 Laparoscopic lymph node dissection is associated with a shorter postoperative recovery time and probably less late radiationrelated morbidity than open transperitoneal staging.61 The risk of occult para-aortic metastases is greatest for patients with grossly involved pelvic nodes; these patients may be the best candidates for surgical exploration if they have no medical contraindication to the procedure. Several small studies suggest that they also may benefit from removal of bulky lymph node metastases, which can be difficult to control with radiation therapy alone.62,63 Despite the sensitivity of lymph node dissection, positive lymph nodes can be missed during the procedure.64 When the results seem inconsistent with pretreatment imaging, repeat imaging is indicated to verify that abnormal nodes were removed. Several investigators are exploring the role of intraoperative lymphatic mapping for patients with cervical cancer and other gynecologic malignancies.65

Prognostic Factors Tumor Size and Local Extent Large tumor size is one of the most important predictors of local recurrence and death from cervical cancer. This is true for patients treated with hysterectomy54,66 (Fig. 29-5) and 1.0 ≤1.0 cm (n = 32)

0.9

1.1–2.0 cm (n = 49)

0.7 2.1–3.0 cm (n = 64)

0.6 0.5 >4.0 cm (n = 24)

0.4

3.1–4.0 cm (n = 30)

0.3 0.2 0.1

P = .008

0.0

Percent DSS

Proportion surviving

0.8

for patients treated with radiation therapy67-69 (Fig. 29-6). FIGO stage IB disease is now divided into two prognostic groups according to tumor size (see Box 29-1); other stage categories act in part as surrogates for tumor size. On clinical examination, clinicians may report the size of the visible lesion or the widest diameter of the cervix or palpable cervical lesion. Although the diameter of tumors that infiltrate and are fixed to the pelvis may be difficult to estimate from clinical examination, tumors that are fixed to both pelvic walls tend to be larger and have a poorer prognosis than those with unilateral fixation.70,71 Before the 1995 revision of the FIGO staging system, surgeons rarely reported clinical diameter; however, various investigators reported measurements from histologic sections of tumor thickness, depth of invasion, diameter, or volume.54,66,72 Retrospective studies have suggested that tomographic imaging methods, particularly MRI, yield more accurate measurements of tumor size than does clinical examination.73-75 However, results of tomographic imaging cannot be used to assign stage, and guidelines for consistent reporting of these results are needed. For patients treated with surgery, histologic evidence of extracervical spread is associated with a poorer prognosis. Parametrial extension is associated with a higher rate of lymph node involvement, local recurrence, and death from cancer.76,77 Paracervical extension occurs with about equal frequency in the right, left, or anterior direction; Landoni and colleagues78 found evidence of anterior spread through the vesicocervical septa or into the vesicocervical ligaments in 22 of 59 women with paracervical disease. This has important implications for postoperative treatment planning. Although clinical evidence of vaginal involvement (stage IIA) or parametrial involvement (stage IIB) is associated with a poorer prognosis after radiation therapy, some investigators have found that the predictive power of stage diminishes or is lost when comparisons are corrected for differences in clinical tumor diameter.79,80 Uterine body involvement is associated with an increased rate of distant metastases in patients treated with radiation therapy or surgery.81,82

100 90 80 70 60 50 40 30 20 10 0

Normal (150) Enlarged, <5 cm (301) 5–7.9 cm (154) ?8 cm (12)

0 0

1

2

3

4

5

6

7

8

9

10

Years

FIGURE 29-5.  Relationship between tumor diameter and survival for patients treated with radical hysterectomy for cervical cancer. (From Alvarez RD, Potter ME, Soong SJ, et al. Rationale for using pathologic tumor dimensions and nodal status to subclassify surgically treated stage IB cervical cancer patients. Gynecol Oncol 1991;43:108-112.)

1

2

3

4

5

6

7

8

9

10

Time (years)

FIGURE 29-6.  Relationship between tumor diameter and disease-specific survival (DSS) for 1526 patients treated with radiation therapy for FIGO stage IB squamous carcinoma of the cervix. (Modified from Eifel PJ, Morris M, Wharton JT, et al. The influence of tumor size and morphology on the outcome of patients with FIGO stage IB squamous cell carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys 1994;29:9-16.)

29  The Uterine Cervix

743

Lymphovascular Space Invasion

Lymph node involvement is another important independent predictor of outcome. After radical hysterectomy, reported 5-year survival rates are usually about 35% to 40% lower when the pelvic lymph nodes are involved.83,84 However, some studies suggest that postoperative chemoradiation therapy improves these results.85 Several oncologists have reported that survival decreases with increasing size of the largest involved nodes86,87 and with increasing number of nodes involved.84,88 Survival also decreases with increasing level of regional involvement and with the extent of central disease in the cervix. Overall, the survival rates for patients with positive para-aortic nodes are about one half of those for patients who have similar stages of disease without para-aortic lymph node involvement (Fig. 29-7).89-101 In patients who have para-aortic node metastases from relatively small tumors, cure rates are 40% to 50% with extended-field radiation therapy.93

LVSI correlates with an increased risk of recurrence. This reflects in part the strong correlation between LVSI and lymph node involvement. However, several studies of large series of patients treated with radical hysterectomy have demonstrated LVSI to be an independent predictor of prognosis.84,102-104 Although LVSI is rarely quantified in these studies, Roman and colleagues105 did report a correlation between the percentage of histopathologic sections containing LVSI and the incidence of lymph node metastases. A strong correlation also exists between LVSI and outcome for patients with adenocarcinoma of the cervix.45,106

Percent patients with FIGO III/IV

Lymph Node Involvement

100 90 80 70 60 50 40 30 20 10 0 0

10

20

30

40

50

60

Percent surviving

FIGURE 29-7.  Survival rates of patients treated with radiation for biopsy-confirmed para-aortic node metastases from cervical cancer. Each point represents one series of patients; the area is proportional to the number of patients (15 to 98) in the series. Blue and red circles represent series with survival rates calculated at 5 and 3 years, respectively.

Histologic Type and Tumor Grade A number of systems have been developed to subclassify squamous cell carcinomas on the basis of histologic type or tumor grade. Although many of these systems correlate with outcome, no single method has been confirmed to independently predict prognosis when other risk factors such as tumor size, depth of invasion, and lymph node status are considered.84,102,103,107 Investigators who have compared the outcome of patients with adenocarcinomas or squamous cell carcinomas have reached different conclusions about the relative prognoses of the two histologic types of cervical cancer. No significant difference was found in several retrospective series of patients treated with radiation.108-111 However, comparative studies have been confounded by the lack of a consistent definition of stage IA disease (particularly for adenocarcinomas) and by variations in treatment selection for patients with tumors of different histologic type. In a review of 1767 patients with stage IB disease (229 with adenocarcinoma), Eifel and colleagues112 found that patients with adenocarcinoma were at significantly higher risk for recurrence and death from disease (Table 29-2).112 This finding was independent of age, tumor size, or tumor morphology. The rate of distant metastasis for patients with bulky (≥4 cm) adenocarcinomas was almost twice that for patients with squamous cell carcinomas (37% versus 21%, P < .01). In contrast, no significant difference in the rate of

TABLE 29-2. Risk of Pelvic or Distant Relapse at 5 Years for Patients Treated with Radiation Alone for FIGO Stage IB Carcinoma of the Cervix Actuarial Risk of Relapse

Tumor Size

Histologic ­Subtype

Number of ­Patients

Pelvic*

Distant†

Any Site

<4 cm

SCC

706

2.6%

7.4%

9.6%

P < .01

P = .01

P = .03

≥4 cm

AC

113

6.3%

15.2%

16.9%

SCC

797

13.1%

21.2%

26.4%

P = .16

P < .01

P < .01

16.7%

36.8%

43.6%

AC *Progressive

106

or recurrent pelvic disease detected at any time after initial treatment. recurrence detected at any time after initial treatment. AC, adenocarcinoma, FIGO, International Federation of Gynecology and Obstetrics; SCC, squamous cell carcinoma. From Eifel PJ, Burke TW, Morris M, et al. Adenocarcinoma as an independent risk factor for disease recurrence in patients with stage IB ­cervical carcinoma. Gynecol Oncol 1995;59:38-44. †Extrapelvic

744

PART 9  Female Genital Tract

pelvic recurrence was found between patients with adenocarcinoma and those with other histologic subtypes. Several investigators have reported high recurrence rates after radical hysterectomy for adenocarcinomas that enlarge the cervix, are high grade, or are associated with LVSI.45,113,114 In one study, Eifel and colleagues45 ­reported that among patients with adenocarcinomas 3 to 4 cm in diameter, the rates of pelvic recurrence were 45% after treatment with radical hysterectomy and 11% after treatment with radiation therapy. This was true despite a greater prevalence of adverse prognostic features in the group treated with radiation. However, other investigators have suggested that adenocarcinomas are more effectively treated with hysterectomy or combined treatment than with radiation alone.115 In a subset analysis of a randomized Gynecologic Oncology Group (GOG) study, Rotman and others114 reported a high recurrence rate (11 of 25 patients) after treatment with radical hysterectomy alone for adenocarcinoma or adenosquamous carcinomas of the cervix; in contrast, only 3 of 31 patients who received postoperative radiation therapy experienced ­recurrent disease. All of these comparisons are compromised by the relatively small number of patients treated surgically for cervical ­adenocarcinomas. Although the prognostic significance of histologic grade has been disputed for squamous cell carcinomas, a clear correlation exists between the degree of differentiation and clinical behavior for adenocarcinomas.45,46,116,117 Anaplastic small cell carcinomas of the cervix have a distinctly poorer prognosis than squamous cell carcinomas, with a tendency to progress rapidly and metastasize widely.50,72,118

Other Tumor Factors Several investigators have reported a correlation between the serum concentration of squamous cell carcinoma antigen and the extent of squamous carcinoma of the cervix, although they disagree about the independent predictive power of the test.119-121 In one study of 284 patients treated with radical hysterectomy for squamous cell carcinoma, 65% of the patients with serum squamous cell carcinoma antigen levels above 8 μg/L had lymph node metastases, compared with only 14% of the patients with antigen levels below 8 μg/L.122 Other factors have been studied for their prognostic significance in cervical cancer. A strong inflammatory response in the cervical stroma tends to predict a good outcome,123 whereas increased tumor vascularity has been associated with a relatively poor prognosis.123,124 The reported incidence of tumor cells in peritoneal washings varies but is less than 5% in most series.125,126 Takeshima and colleagues126 reported a higher rate among patients with adenocarcinomas (11.4%) than among those with squamous cell carcinomas (1.7%). The finding of ­tumor cells in peritoneal washings is more common in ­patients with advanced-stage disease or lymph node metastases; however, the value of this finding as an independent predictor of prognosis or pattern of disease progression is uncertain. Some investigators have reported a correlation between HPV subtype and prognosis.127 Several investigators have reported higher recurrence rates among patients with histologically negative lymph nodes when a polymerase chain reaction assay of the lymph nodes was strongly positive for

HPV DNA.128,129 Others have correlated poor prognosis with the presence of HPV mRNA129 or HPV-related proteins130 in the peripheral blood of women with cervical cancer.

Patient-Related Factors The relationship between patient age and outcome is uncertain, with different investigators reporting young age to be a favorable prognostic factor,131,132 an adverse prognostic factor,133,134 or not predictive of outcome.135-137 Several factors may contribute to these differences. Although most investigators try to correct for potential confounding biases, they often do so incompletely. Studies of patients treated with surgery or radiation therapy may be biased because age is often used to select treatment. For example, patients with small tumors may be more likely to be treated with radiation if they are elderly than if they are young. In some series, older patients might have received less aggressive treatment. It is also possible that cultural or regional differences affect the influence of age on treatment selection or compliance. Many investigators have reported correlations between a low hemoglobin level before or during treatment and poor prognosis.138-141 These comparisons are striking but are often confounded by the prevalence of other adverse tumor features in anemic patients. Although many investigators have speculated that the poor prognosis of anemic patients is caused in part by hypoxia-induced radiation resistance, only one small randomized study in 1978, conducted at the Princess Margaret Hospital, has tested this hypothesis.138 Through the use of transfusions, patients’ hemoglobin levels were maintained at a level of at least 12 g/dL in one group of the study, whereas hemoglobin levels were allowed to drop to as little as 10 g/dL in the other group. The 25 patients in the control group of the study who had hemoglobin levels less than 12 g/dL had a significantly higher local recurrence rate than the patients in the experimental group who required transfusions to maintain a high hemoglobin level. Larger studies are still needed to confirm this finding. Unfortunately, other treatment protocols that have been designed to overcome the theoretical radiobiologic consequences of intratumoral hypoxia (through the use of hypoxic cell sensitizers, hyperbaric oxygen breathing, or neutron therapy) have not improved the outcome of patients with locally advanced disease.142-148 A prospective randomized study conducted by the GOG to investigate the use of exogenous erythropoietin to increase hemoglobin levels was aborted after an unexpectedly high rate of thromboembolic complications was identified in early patients randomized to the erythropoietin group; similar problems occurred in studies of tumors at other anatomic sites. Several investigators have linked low intratumoral oxygen tension levels with poor survival.148-150 However, those studies have also demonstrated that hypoxic tumors tend to have other adverse features that may explain their greater tendency to recur after treatment. Although high platelet count is thought by some to be associated with poor prognosis, large studies suggest that thrombocytosis may be a marker for more advanced disease but is not an independent predictor of poor prognosis.151 Several studies have demonstrated that in the ­ United States, women of color with invasive cervical ­ cancer

29  The Uterine Cervix

­ resent with higher-stage disease152-154 and lower p ­hemoglobin ­ levels155,156 than white women. Differences in ­ socioeconomic status may influence minority ­ patients’ access to care and ability to comply with treatment ­recommendations.156 In a study of 1304 women treated with radiation for cervical cancer, Kucera and colleagues157 reported that smokers with cervical cancer had a poorer 5-year survival rate than nonsmokers. This difference was statistically significant for patients with stage III disease (5-year survival 20.3% versus 33.9%, P < .01). However, the investigators did not analyze the possible influence of smoking-related deaths from causes other than cervical cancer.

Treatment Options Radiation therapy, surgery, and chemotherapy all play important roles in the curative management of cervical cancer. In recommending treatment for an individual patient, clinicians must consider the characteristics and extent of the cancer, possible side effects of treatment, and the patient’s personal needs. Positive findings from randomized trials have led to the routine use of concurrent chemoradiation therapy for appropriately selected patients with locoregionally advanced disease. The addition of chemotherapy to initial treatment regimens has improved outcome in some cases but has also increased the complexity of treatment, which is important because patients with cervical cancer often have severe socioeconomic limitations and little or no experience with health care systems. To achieve the best results, careful counseling and social services support must be integrated into the treatment plan from its inception.

Stage IA1 Disease Because the risk of extracervical involvement in patients with stage IA1 disease is low (<1%), the goal of treatment is to remove or sterilize disease in the cervix. All ­patients should have a cone biopsy because this is necessary to make the diagnosis of microinvasion and to verify that the paracervical tissues and lymph nodes do not require ­treatment. The most conventional treatment for stage IA1 cancer is a type I simple extrafascial hysterectomy. If the patient is young, the ovaries should be left in place. Stage IA disease also can be treated effectively with radiation therapy. ICRT alone is sufficient; it controls disease for more than 10 years in 95% to 100% of patients.158,159 Patients who want to maintain fertility may be treated with a generous cone biopsy alone if the tumor invades less than 3 mm without LVSI and if the margin of the cone biopsy is negative.160,161 These patients must be followed closely, but the risk of disease progression seems to be low.

Stage IA2 Disease Small lesions that invade 3 to 5 mm have an average risk of lymph node metastasis of about 5%. This risk is usually considered high enough to justify treatment of the paracervical tissues and pelvic lymph nodes.76 Surgical treatment of stage IA2 disease usually includes a type II radical hysterectomy and pelvic lymph node dissection. However, patients desiring to maintain fertility may be treated instead with a radical vaginal trachelectomy and laparoscopic lymph node dissection.162-164 This ­technique,

745

pioneered in France, has been shown to be safe and has a low rate of recurrence for carefully selected patients with stage IA or small IB1 tumors. Stage IA2 tumors also can be treated effectively with radiation therapy. Brachytherapy alone may be used if prolonged treatment is contraindicated, but the risk of pelvic lymph node metastasis, although low, is enough to justify treatment of the pelvic lymph nodes in most cases.

Stage IB1 or Small Stage IIA Disease Stage IB1 and small (<3 cm) stage IIA tumors are associated with a significant risk of microscopic paracervical extension or lymph node metastasis. In a prospective GOG study of patients treated with radical hysterectomy for stage I disease, lymph node metastases were found in 42 (16%) of 261 patients whose tumors were estimated clinically to have a maximum diameter less than 3 cm.76 Landoni and colleagues115 found positive lymph nodes in 28 (25%) of 114 patients treated with radical hysterectomy for stage IB-IIA tumors that measured 4 cm or less on initial clinical examination. This risk is clearly sufficient to warrant treatment of the paracervical ligaments and pelvic lymph nodes. However, because the probability of extrapelvic spread is low, effective locoregional treatment is curative for most patients.

Radical Radiation Therapy versus Radical Surgery Radical radiation therapy (external pelvic irradiation plus brachytherapy) and surgery (radical hysterectomy plus bilateral pelvic lymphadenectomy) are ­ acceptable ­ options for most patients. Conventional wisdom suggests that these two alternatives produce approximately equivalent outcomes, with overall 5-year survival rates of ­approximately 80% reported for each treatment. However, direct comparisons of retrospective reviews may not be ­appropriate because younger patients and those with more ­ favorable clinical characteristics tend to be selected for treatment with radical hysterectomy. Only one prospective randomized trial has directly compared radiation therapy for patients with stage IB or IIA disease.115 Patients were randomly assigned to receive treatment with type III radical hysterectomy or radiation therapy. Patients in the surgical group received postoperative radiation therapy if the tumor involved the parametria, surgical margins, deep stroma, or lymph nodes (Table 29-3).115 No significant differences were found in the rates of relapse or survival between the two treatment groups, but the overall rate of grade 2 to 3 complications was greater for patients treated with hysterectomy.

Role of Postoperative Radiation Therapy The role of postoperative radiation therapy for patients with stage IB or small IIA disease treated with radical surgery has been a subject of controversy. However, two randomized studies have added valuable information. Most practitioners agree that patients who have more than one positive lymph node, parametrial involvement, or positive margins require postoperative radiation therapy, because they have significantly lower pelvic recurrence rates with postoperative treatment. However, selection bias has made it difficult to evaluate the influence of postoperative radiation on survival for these patients, and the question has never been evaluated in prospective trials.

746

PART 9  Female Genital Tract

TABLE 29-3. Results of a Randomized Trial Comparing Radical Hysterectomy with Radiation Therapy for Stage IB or IIA Carcinoma of the Uterine Cervix Surgery

Radiation Therapy

Variable

Tumor ≤ 4 cm

Tumor > 4 cm

Tumor ≤ 4 cm

Tumor > 4 cm

Number of patients

115

55

113

54

—

—

Postoperative RT given

62 (54%)

46 (84%)

Positive findings on LAG

12 (10%)

12 (22%)

Surgically node positive

28 (24%)

17 (31%)

8 (7%)

8 (15%)

22 (19%)

19 (35%)

7 (6%)

Clinical stage IIA Surgical stage > pT2a Positive surgical margins

9 (8%)

13 (24%)

—

—

14 (12%)

9 (17%)

—

—

12 (22%)

—

—

5-Year survival rate

87%

70%

90%

72%

Pelvic relapses

11 (10%)

11 (20%)

12 (11%)

16 (30%)

Distant relapses

12 (10%)

8 (15%)

Grade 2-3 side effects

9 (8%)

48 (28%)

7 (13%) 19 (11%)

LAG, lymphangiography; RT, radiation therapy. From Landoni F, Maneo A, Colombo A, et al. Randomised study of radical surgery versus radiotherapy for stage Ib-IIa cervical cancer. ­Lancet 1997;350:535-540.

Findings of a prospective randomized trial (GOG 92), updated in 2006 by Rotman and colleagues,114 suggest that postoperative pelvic radiation therapy may benefit patients with intermediate-risk features after radical hysterectomy. In that trial, women whose hysterectomy specimens demonstrated various combinations of deep stromal invasion (invasion of more than one third to two thirds of the stroma), LVSI, or large tumor diameter (≥4 to 5 cm) were randomly assigned to receive postoperative radiation therapy or no further therapy. The authors of the study reported a 46% reduction in the risk of recurrence (P = .007) and significant improvements in rates of recurrence or death (hazard ratio = 0.58) when postoperative pelvic radiation was given (Table 29-4).114,165 Although a 30% reduction in the risk of death was observed among patients who received radiation therapy, this difference was not statistically significant (P = .07). In 2000, Peters and colleagues85 reported results of a Southwest Oncology Group (SWOG) trial in which ­patients with more than one positive lymph node, parametrial involvement, or positive surgical margins were randomly assigned to receive postoperative radiation therapy or postoperative radiation therapy with concurrent and adjuvant cisplatin and fluorouracil. This study demonstrated a highly significant improvement in overall survival for the patients who received chemotherapy in addition to radiation therapy (hazard ratio = 1.96, P = .007). The role of concurrent chemotherapy has not been evaluated for patients with high-risk local disease features but negative lymph nodes. However, of the 23 recurrences among patients treated with postoperative radiation therapy as part of the GOG 92 trial, 19 (83%) were confined to the pelvis, raising questions about the possible benefit of concurrent chemotherapy in selected intermediate-risk cases. In its 1996 consensus statement, a National Institutes of Health panel concluded that “to minimize morbidity, primary therapy should avoid the routine use of both

TABLE 29-4.   Recurrences by Treatment Regimen in a GOG Randomized Trial in Selected Patients with Stage IB, Node-Negative Carcinoma of the Cervix Treated with Initial Radical Hysterectomy Site of ­Recurrence

Radiation Therapy (n = 137)

No Further Therapy (n = 140)

No recurrence

113 (82.5%)

97 (69.3%)

Recurrence

24 (17.5%)

43 (30.7%)

Local only

19 (13.9%)

29 (20.7%)

Vagina

2

8

Pelvis

16

19

1

2

Vagina and pelvis Distant*

4 (2.9%)

12 (8.6%)

Unknown

1 (0.7%)

2 (1.4%)

*Includes

some patients who experienced failure at local and distant

sites. GOG, Gynecologic Oncology Group. From Rotman M, Sedlis A, Piedmonte M, et al. A phase III randomized trial of postoperative pelvic irradiation in stage IB cervical carcinoma with poor prognostic features: follow-up of a gynecologic oncology group study. Int J Radiat Oncol Biol Phys 2006;65:169176.

radical surgery and radiation therapy.” If a patient’s initial workup suggests a high likelihood of parametrial invasion, deep stromal invasion, or lymph node metastases, strong consideration should be given to treating the patient with radiation therapy alone or, in selected cases, with chemoradiation therapy. Patients who have an enlarged cervix with

29  The Uterine Cervix

circumferential tumor involvement or extensive LVSI fall into this category.

Stage IB2 or Minimal Stage IIA Disease Stage IB2 tumors appear, at least on clinical examination, to be confined to the cervix and are therefore technically resectable. Stage IIA tumors with minimal vaginal involvement may also be resectable. However, bulky stage IB2 and stage IIA tumors are usually deeply invasive and, when treated with initial surgery, often are found to be associated with microscopic paracervical disease or lymph node metastases. The ideal management of stage IB2 and minimal stage IIA tumors is a subject of considerable controversy, and the approach to these tumors varies widely between centers. Although some investigators continue to advocate initial surgical management for bulky stage IB and even some IIA tumors, most patients who are treated with initial surgery for these tumors require additional treatment with radiation or chemoradiation therapy. In the randomized trial reported by Landoni and colleagues,115 84% of surgically treated patients required postoperative radiation therapy, which contributed to a higher complication rate in that group. Even though the overall survival rates were similar for patients treated with initial surgery and those treated with radiation, the patients in the radiation therapy group received more protracted, lower-dose treatments than those recommended by most experts in the field. Some studies (discussed later) also suggest that chemoradiation therapy rather than radiation alone (the control group of the Landoni study) should be considered standard treatment for many patients with stage IB or IIA tumors. To improve the resectability of stage IB2 tumors, some investigators have advocated use of neoadjuvant (preoperative) chemotherapy. Only one prospective trial of this approach has been published, although several others are nearing completion. In the published study, reported by Sardi and colleagues,166 123 patients with bulky tumors (≥4 cm) were randomly assigned to treatment with radical hysterectomy with or without neoadjuvant chemotherapy. All patients received postoperative radiation therapy. The patients who received neoadjuvant chemotherapy were more likely to have had resectable tumors and had a better survival rate than were patients who had immediate exploration. However, no study has yet reported a comparison between this trimodality treatment and radiation alone or chemoradiation therapy. In 1996, the GOG initiated a trial (GOG 141) comparing radical hysterectomy with or without neoadjuvant chemotherapy with cisplatin and vincristine. The study was closed prematurely in 2001, and its results have yet to be published. Although primary radiation therapy is effective for many patients with stage IB2 disease, at least 8% to 10% of patients with bulky endocervical tumors treated with radiation therapy experience central disease recurrence.67,69,167 This led clinicians to investigate the role of adjuvant extrafascial hysterectomy after radiation therapy. Early retrospective studies from The University of Texas M.  D. Anderson Cancer Center168 suggested that pelvic disease control was improved when adjuvant surgery followed radiation therapy. These results led many clinicians to adopt a combined approach as standard. However, subsequent analyses of the M.  D. Anderson Cancer Center experience

747

suggested that the better survival rates observed among patients who received combined treatment reflected in part clinicians’ tendencies to select for this approach patients with smaller, more responsive tumors and those without regional metastases.80 In 1991, Mendenhall and colleagues169 reported the outcome of patients treated at the University of Florida before or after clinicians began to add adjuvant hysterectomy to the treatment of bulky (>6 cm) endocervical tumors. These investigators found no difference between the two treatment policies in pelvic disease control or survival rates. A prospective, randomized GOG study completed in 1990 (GOG 71)170 compared treatment with radiation alone or radiation plus extrafascial hysterectomy for 282 patients with stage IB tumors that were 4 cm or larger. Findings from that trial, published in 2004, indicated no significant difference in survival; the two treatment groups had virtually overlapping survival curves. However, the power of the trial to detect a difference between the two treatments was probably weak; the study included a large number of patients with small tumors, for whom the margin for improvement with adjuvant surgery in addition to radiation therapy is small. In a review of 1526 patients treated with radiation therapy alone for squamous carcinoma of the cervix, Eifel and colleagues67 reported central recurrence rates of only 1% for tumors smaller than 5 cm and 3% for exophytic tumors measuring 5 to 7.9 cm. The addition of concurrent chemotherapy with radiation therapy further reduces the risk of central recurrence ­after primary treatment with radiation, suggesting that the cost and side effects of adjuvant hysterectomy are rarely ­warranted. In 1999 and 2000, investigators reported five prospective trials demonstrating improved survival rates when patients treated with radiation for locoregionally advanced cervical cancer also were given concurrent cisplatin chemotherapy (Table 29-5).85,171-179 Patients with bulky stage IB disease were included in two of these trials. In one trial (GOG 123),172 patients were treated with radiation or radiation plus weekly cisplatin before brachytherapy and adjuvant hysterectomy (this trial was designed before the results of GOG 71 were known to be negative). Patients who received concurrent chemotherapy were more likely to have evidence of complete response in the hysterectomy specimen (52% versus 41%, P = .04) and had significantly better progression-free survival (P < .001) and overall survival (P = .008). The second trial, Radiation Therapy Oncology Group (RTOG) trial 90-01 (discussed in the next section),171,180 included 117 patients with bulky stage IB or IIA disease (30% of those entered in the trial). These findings strongly support the use of concurrent chemoradiation therapy as primary treatment for most patients with bulky (≥5 cm in diameter) stage IB tumors.

Stage II to III Disease Although a few surgeons have advocated ultraradical surgical procedures for patients with stage IIB to IIIB disease or for patients with more than minimal vaginal involvement, radiation therapy is usually considered to be the local treatment of choice in such cases. Remarkably, even very large tumors (even those 7 cm or more in diameter) have a chance of being cured with radiation therapy alone, particularly if there is no evidence of extensive regional

748

PART 9  Female Genital Tract

TABLE 29-5. Prospective Randomized Trials that Investigated the Role of Concurrent Radiation Therapy and Chemotherapy in Patients with Locoregionally Advanced Cervical Cancer Study and Reference al173

Rose et (GOG-120)

Eligibility FIGO IIB-IVA

Number of Patients 526

Chemotherapy in Control Arm

Relative Risk of ­ ecurrence R (90% CI)

P Value

Cisplatin 40 (up to six cycles)

g/m2

HU 3 (2×/wk)

0.57 (0.42-0.78)

<.001

Cisplatin 50 mg/m2 5-FU 4 g/m2/96 hr HU 2 g/m2 (2 ×/wk) (two cycles)

HU 3 g/m2 (2×/wk)

0.55 (0.40-0.75)

<.001

Chemotherapy in Investigational Arm mg/m2/wk

Eifel et al171 (RTOG 90-01)

FIGO IB-IIA (≥5 cm), IIB-IVA or pelvic nodes involved

403

Cisplatin 75 mg/m2 5-FU 4 g/m2/96 hr (three cycles)

None*

0.51 (0.36-0.66)

<.001

Keys et al172 (GOG-123)

FIGO IB (≥4 cm)

369

Cisplatin 40 mg/m2/wk (up to six cycles)

None‡

0.51 (0.34-0.75)

.001

Whitney et al174 (GOG-85)

FIGO IIB-IVA

368

Cisplatin 50 mg/m2 5-FU 4 g/m2/96 hr (two cycles)

HU 3 g/m2 (2×/wk)

0.79 (0.62-0.99)

.03

Peters et al85 (SWOG 87-97)

FIGO I-IIA after radical hysterectomy with nodes, margins, or parametrium positive

268

Cisplatin 70 mg/m2 5-FU 4 g/m2/96 hr (two cycles)

None

0.50 (0.29-0.84)

.01

Pearcey et al175 (NCI Canada)

FIGO IB-IIA (≥5 cm), IIB-IVA or pelvic nodes involved

259

Cisplatin 40 mg/m2/wk (up to six cycles)

None

0.91 (0.62-1.35)‡

.43

Wong et al179

FIGO IB-IIA (>4 cm), IIB-III

220

Epirubicin 60 mg/m2 then 90 mg/m2 every 4 wk for five more cycles§

None

≈ 0.65

.02

Thomas et al176

FIGO IB-IIA 234 (≥5 cm), IIB-IVA

5-FU 4 g/m2/96 hr (two cycles)

None¶

Lorvidhaya et al178

FIGO IIB-IVA

926

Mitomycin 10 mg/m2 and oral 5-FU 300 mg/m2/d × 14 days (2 cycles); ± adjuvant 5-FU

None or adjuvant 5-FU only

Not stated

.0001

Lanciano et al177 FIGO IIB-IVA (GOG-165)

316

5-FU 225 mg/m2/d for 5 days per week (protracted venous infusion)

Cisplatin 40 mg/m2/wk (up to six cycles)

1.29 (0.93-1.8)

Not stated

NS

*Patients in the control arm had prophylactic para-aortic irradiation. † All patients had extrafascial hysterectomy after radiation therapy. ‡Survival. §Chemotherapy was begun on day 1 and continued every 4 weeks during and after radiation therapy. ¶Patients were also randomly assigned to receive standard or hyperfractionated radiation therapy in a four-arm trial. CI, confidence interval; FIGO, International Federation of Gynecology and Obstetrics; 5-FU, fluorouracil; GOG, Gynecology Oncology Group; HU, hydroxyurea; NS, not significant; RTOG, Radiation Therapy Oncology Group.

29  The Uterine Cervix

749

TABLE 29-6.  Results after Low-Dose-Rate Brachytherapy and External Beam Radiation Therapy Study and Reference Coia et al,

1990182

Barillot et al, 1997181

Perez et al,

199269

M.  D. Anderson Cancer Center, 19801994 (unpublished study)

Lowrey et al,

1992167

FIGO Stage

Number of Patients

Pelvic Control (%)

5-Year Survival (%)

IB

168

88

74

II

243

73

56

III

114

49

33

I

482

94

89

IIA

223

94

77.5

IIB

543

87.5

67

IIIB

470

70.5

43

Tumors ≥ 6 cm (%)

11*

IB

384

94

90

13.5 (≥5 cm)

IIA

128

88

81

39.1 (≥5 cm)

IIB

353

83

77

III

293

64

59

IB1

524

98

86

IB2

482

81

67

22.8

IIA

149

84

67

20.5

IIB

211

81

54

58.8

III

328

70

47

IB

130

81

22

IIA

164

74

5

IIB

112

64

43

*For

the remaining patients with stage I or II disease, tumor diameter was less than 6 cm for 65% and unknown for 24%. FIGO, International Federation of Gynecology and Obstetrics.

metastases. Although local control of large tumors continues to be a significant problem, cervical cancer is probably the only epithelial cancer for which control of large tumors with radiation therapy alone is possible. These results reflect the effectiveness of brachytherapy in controlling central disease. Before the use of concurrent chemoradiotherapy, 5-year survival rates for patients with stage IIB disease were usually reported to be between 50% and 75% (Table 29-6).69,167,181,182 The broad range probably reflects differences in staging, patient selection, and treatment technique between reporting institutions. For ­ patients with stage IIIB disease, reported survival rates after treatment with radiation alone ranged between 30% and 50%.68,71,181-185 During the past 30 years, many investigators have attempted to improve on these results by combining radiation therapy with sequential or concurrent chemotherapy. Although neoadjuvant chemotherapy has produced some impressive tumor responses, randomized trials have repeatedly failed to demonstrate improvements in survival.186-192 In two trials,189,191 survival rates were actually poorer when neoadjuvant chemotherapy was added to radiation therapy. In contrast, trials comparing concurrent chemoradiation therapy with radiation therapy alone have demonstrated consistent improvements in survival with chemoradiation therapy, particularly when cisplatin is included in the chemotherapy regimen. RTOG 90-01171,180 and two GOG

t­ rials (GOG 85 and GOG 120)172,173 evaluated the use of cisplatin-based chemoradiation therapy for patients with stage IIB to IVA disease (RTOG 90-01 also included some patients with bulky stage IB or IIA disease). All three trials demonstrated significant improvements in local disease control, distant metastases, and survival (see Table 29-5). After publication of these results and an associated National Institutes of Health alert, many clinicians consider concurrent cisplatin-based chemoradiation therapy to be a standard treatment for patients with locally advanced cervical cancer. Although the preliminary findings of RTOG 90-01, first published in 1999, demonstrated a highly significant overall survival benefit from chemotherapy among patients with stages IB to IIB disease (79% versus 55% at 5 years, P < .0001), no significant difference was found in survival among patients with stages III to IVA disease; at the time, follow-up was incomplete, and the total number of patients with more advanced disease (117) was small, contributing to large confidence intervals on the results. Nevertheless, an update in 2004171 did show a significant difference in disease free-survival rates (54% versus 37%, P =. 05) and a trend toward improved overall survival (59% versus 45%, P = .07). Although concurrent chemotherapy should be considered for patients with stage IIIB disease, some patients cannot be treated with cisplatin because of compromised renal function from tumor-related hydronephrosis. In most cases, the ureters should be decompressed with stents or

PART 9  Female Genital Tract

nephrostomies to optimize renal function before a final determination is made about the feasibility of chemotherapy. If renal compromise prohibits the use of cisplatin, other drugs such as carboplatin or epirubicin may be considered, although the evidence supporting the use of these drugs is less certain (see “Chemoradiation Therapy”).

Stage IVA Disease Less than 5% of cases of invasive cervical cancer are truly stage IVA (i.e., with biopsy-proven bladder or rectal mucosal involvement). Most such cases are classified as IVA because of bladder involvement; rectal involvement is rarely present at initial diagnosis. The management of stage IVA disease is particularly challenging because the tumors usually are massive, fixed to pelvic structures, and regionally metastatic. Brachytherapy is often difficult or impossible to perform because vesicovaginal fistulas develop before diagnosis or during treatment (as the tumor responds to the radiation). However, at least 10% to 20% of stage IVA tumors are curable with radiation therapy alone. If the patient has not developed a fistula or if the fistula is small and the urinary stream has been diverted, brachytherapy should be attempted; if brachytherapy cannot be performed, central disease can sometimes be controlled with EBRT delivered by conformal techniques such as intensity-modulated radiation therapy (IMRT). A small proportion of patients may have disease that can be cured without brachytherapy with EBRT followed by anterior or total pelvic exenteration. Only patients who have relatively small, mobile central tumors are likely to be candidates for radical surgical treatment. Although too few patients have been treated with chemoradiation therapy to permit evaluation of the influence of combined treatment, concurrent chemotherapy is usually recommended if the patient’s renal function and performance status permit.

Regional Disease Most patients treated with radical hysterectomy who are found to have lymph node metastases in the lymphadenectomy specimens require postoperative radiation therapy. On the basis of the findings of Peters and colleagues,85 concurrent chemotherapy is also recommended if the patient has no medical contraindications. It is sometimes suggested that patients who have only one microscopic focus of regional disease may not require adjuvant treatment. However, because of the small number of women who have this finding and the inconsistent application of postoperative radiation therapy in most of the published series, it is difficult to determine whether adjuvant treatment can be safely omitted.84,103 Patients who are treated with primary radiation therapy for more than microinvasive disease also require treatment of the pelvic lymph nodes. Because bulky lymph node metastases can be difficult to control with EBRT alone, some investigators have suggested preirradiation resection of nodes that measure more than 1.5 to 2 cm in diameter. Several investigators have reported high local control rates for patients treated with excision of bulky nodes and radiation; however, the numbers of patients in those series were small, and this form of combined regional treatment requires additional study.62,63 Delivering more than 60 Gy to lateral pelvic structures without

100 Overall survival (%)

750

75

67% 55%

55%

44%

50 25 0

No. at risk 167 141 177 170 154 132

0

1

2

101 117

86 106

77 97

66 84

56 74

45 59

33 43

25 24

3

4

5

6

7

8

9

10

Years on study

Irradiation treatment

No. dead/total

Pelvic only Pelvic plus para-aortic

85/167 67/170 P = .02

FIGURE 29-8.  Overall survival of patients treated with pelvic radiation therapy or with extended fields (including the pelvis and para-aortic nodes) for FIGO stages IB to IIB cervical cancer in a Radiation Therapy Oncology Group prospective, randomized trial. (From Rotman M, Pajak M, Choi K, et al. Prophylactic extended-field irradiation of para-aortic lymph nodes in stages IIB and bulky IB and IIA cervical carcinomas. JAMA 1995;274:387393.)

exceeding normal tissue tolerance limits is difficult when using conventional EBRT techniques; moreover, this dose is likely to be insufficient to control bulky lymph nodes, particularly those that are larger than 2 to 3 cm in diameter. However, modern conformal or IMRT techniques may provide a means of delivering a higher dose to well-defined volumes containing bulky regional disease. Patients who have para-aortic lymph node involvement can be treated effectively with extended-field radiation. Five-year survival rates range between 25% and 50% (see Fig. 29-7). Because microscopic para-aortic disease may not be detected by routine clinical studies, some investigators have advocated prophylactic irradiation of the aortic nodes for patients at increased risk for metastases to this region. The value of prophylactic extended-field irradiation was tested in two randomized trials. In RTOG 79-2,193 patients with stage IB2 to IIB disease were randomly assigned to receive brachytherapy plus EBRT to the pelvis or to the pelvis and para-aortic nodes (Fig. 29-8). The absolute 5-year survival rate was significantly better for those who had para-aortic treatment (67% versus 55%, P = .02), but no significant difference was found in the disease-free survival rates. The investigators’ analysis of recurrence patterns in the two groups of patients was difficult to interpret because recurrences that occurred immediately above the pelvic field were not analyzed separately and might have been coded as distant or pelvic failures. The generalizability of this early study has been questioned because the initial status of the para-aortic ­region was not evaluated systematically. A concurrent European trial194 had a similar randomization scheme but included patients with more advanced disease (e.g., pelvic lymph node metastases or bulky stage IIB or IIIB disease). Patients were required to have negative para­aortic nodes by lymphangiography or surgical staging. No

751

29  The Uterine Cervix

TABLE 29-7. Results of Radiation Therapy for Pelvic Recurrence after Radical Hysterectomy for Cervical Cancer Study and Reference Parsons208

Year

Number of Patients

Treatment

Pelvic Control (%)

Survival (%)

1974

31

EB ± IC

23

19

al212

1990

28

EB

—

30

al209

1982

24

EB ± IC

38

22

Jobsen et al211

1989

18

EB ± IC

61

44

Thomas et al213

1987

17

RT + CT

47

47

1998

50

RT ± Brachy ± CT

62

33

Deutsch and Potter et

Hogan et

Ijaz et

al210

No pelvic wall disease

16

69

Pelvic node disease or fixation

34

18

Brachy, brachytherapy; CT, chemotherapy; EB, external beam radiation therapy; IC, intracavitary radiation therapy.

s­ ignificant ­ difference was found in the 4-year diseasefree survival rates between the two treatment groups in this study, but the investigators did not compare overall survival rates. The rate of para-aortic node recurrence was significantly higher for patients who did not have that area treated. In several phase II studies, concurrent chemotherapy has been given with extended-field irradiation. Although side effects are greater when treatment fields are enlarged, combined therapy may be tolerable if careful consideration is given to the chemotherapy regimen, volume of tissue irradiated, and other factors that may increase the risk of serious toxicity. The RTOG combined two cycles of cisplatin (75 mg/m2) and fluorouracil (4 g/m2 over 96 hours) with concurrent, twice-daily (1 Gy/fraction) extended-field irradiation for 29 patients with para-­aortic node metastases.195 The investigators reported severe acute and late gastrointestinal toxicity with this regimen and an overall survival rate that was not clearly different from that expected with radiation therapy alone. A GOG combination of once-daily extended-field irradiation (1.5 Gy/day to the aortic nodes) with cisplatin (50 mg/m2) and fluorouracil was better tolerated196 but resulted in more protracted treatment and used the most ­conservative cisplatin dose of the randomized trials. Malfetano and Keys197 reported results of a combination of extended-field irradiation and weekly cisplatin (1 mg/kg, not to ­ exceed 60 mg) for patients with negative para­aortic nodes. The dose used in their regimen was similar to the 50 mg/m2 used by the GOG with pelvic radiation therapy and seemed to be well tolerated. Although it is tempting to conclude from the impressive results achieved with combinations of pelvic irradiation and chemotherapy that this treatment is appropriate for patients with regional disease, patients should be informed that the cost-benefit ratio of concurrent chemotherapy and extended-field irradiation has not been formally tested. New imaging techniques such as PET and safer methods of surgical evaluation of lymph nodes (e.g., retroperitoneal or laparoscopic lymphadenectomy) may increase the precision with which the extent of regional disease can be ­estimated, thereby decreasing the margin for improvement to be gained by the prophylactic treatment of large volumes. No evidence has been found, however, that ­ giving

chemotherapy concurrently with radiation can prevent recurrence in unirradiated nodal sites (the rate of aortic recurrence in patients treated with pelvic chemoradiation therapy in RTOG 90-01 was 9%), and such recurrences can be very difficult to salvage.

Locally Recurrent Disease Some cases of disease that recurs centrally after initial treatment with radiation therapy can be cured with radical surgical resection. Only patients with tumors that are mobile and do not involve pelvic lymph nodes are considered candidates for surgery; those with a disease-free interval of less than 3 years tend to have a poor prognosis but occasionally experience cure.198-200 Total pelvic exenteration is usually required. In rare cases, small recurrences may be treated effectively with radical hysterectomy or anterior exenteration; however, the risk of complications is high when radical hysterectomy follows full-dose radiation therapy.201 The 5-year survival rate for patients who undergo pelvic exenteration for isolated local recurrence of cervical cancer is usually between 25% and 50%.199,202-204 Although several groups have used intraoperative radiation to treat recurrent pelvic wall disease, patients with this type of disease after radical radiation therapy have a very poor prognosis.205-207 Isolated pelvic recurrences after initial treatment with radical hysterectomy can sometimes be cured with radiation therapy (Table 29-7).208-213 Isolated central recurrences without pelvic wall fixation or regional metastasis can be cured in 60% to 70% of cases by using carefully tailored combinations of EBRT and brachytherapy.210 The prognosis is much poorer when the pelvic wall is involved; in that case, usually only 10% to 20% of patients survive 5 years after radiation therapy. Intraoperative EBRT or interstitial therapy may contribute to local control in some patients.205-207 Salvage rates may be lower for patients with locally recurrent adenocarcinomas.210,214 Radiation treatment techniques are similar to those used to treat primary vaginal cancer. Insufficient data are available to establish the benefit of concurrent administration of chemotherapy for patients with locally recurrent cervical cancer, although combined-modality therapy is often given based on extrapolation from its positive results for other types of cancer at advanced stages.213

752

PART 9  Female Genital Tract

Inappropriate Simple Hysterectomy for Invasive Cancer Treatment for cervical cancer that invades more than 3 mm should include the paracervical tissues and pelvic lymph nodes (see “Stage IA2 Disease”). Although cytologic screening should always be done before a hysterectomy, deeply invasive cancer is sometimes discovered unexpectedly in a type I hysterectomy specimen.215 In that situation, the hysterectomy is referred to as an inappropriate simple hysterectomy or cut-through hysterectomy; the latter term may be somewhat misleading because it refers to cases in which various disease extents may or may not involve the margins of the hysterectomy specimen. Patients treated with an inappropriate simple hysterectomy have been classified in five groups according to the amount of disease and presentation: (1) microinvasive cancer, (2) tumor confined to the cervix with negative surgical margins, (3) positive surgical margins but no gross residual tumor, (4) gross residual tumor by clinical examination documented by biopsy, and (5) patients referred for treatment more than 6 months after hysterectomy (usually for recurrent disease).216 If the hysterectomy specimen demonstrates more than 3 to 4 mm of invasion, simple hysterectomy is inadequate; however, with postoperative radiation therapy, local control can usually be achieved, and excellent survival rates can be expected in most cases.217-219 Roman and colleagues219 reported survival rates of 79% and 59% after radiation therapy for patients in groups 2 and 3, respectively, in the classification in the previous ­paragraph. However, the survival rate for patients with gross residual or recurrent disease (groups 4 and 5) was only 41%. Patients in groups 4 and 5 were usually treated with EBRT alone to a dose of 65 Gy. Interstitial therapy under laparoscopic guidance can be used to achieve somewhat higher doses if a fairly well-defined lesion is found. The value of chemoradiation therapy has not been proven in the setting of inappropriate simple hysterectomy. However, combined treatment is probably reasonable for patients with gross disease because of the relatively high local recurrence rates after radiation therapy alone.

Metastatic Disease: Stage IVB or Recurrent Disease Stage IVB disease is rare. This is fortunate because this advanced stage of disease is usually incurable, and 5-year survival rates are less than 10%.220 Possible rare exceptions include patients who have inguinal metastases from lower vaginal involvement (technically stage IVB disease, although the inguinal nodes are part of the primary nodal drainage from the distal vagina). Patients with solitary lung lesions also may be cured in some cases—possibly because they have two primary tumors that cannot be distinguished histologically. Most other patients who have stage IVB cervical cancer are treated with palliative intent. The most active single chemotherapy agent in patients with cervical carcinoma is cisplatin, which produces response rates of 30% to 50% in previously untreated patients. However, the median duration of response tends to be short (4 to 6 months), and complete responses are rare.221,222 Response rates tend to be lower (10% to 20%) for patients who receive cisplatin for recurrences after pelvic irradiation. Cisplatin analogues (e.g., carboplatin and

iproplatin) tend to produce somewhat lower response rates but have been tested less comprehensively. Other chemotherapy agents that can produce responses in a minority (<15%) of patients with advanced or recurrent cervical carcinoma include doxorubicin, bleomycin, mitomycin C, and fluorouracil. Many cisplatin-based combinations have been tested for advanced or recurrent cervical carcinoma. High overall response rates have been reported for previously untreated patients, but response rates are poor for tumors that recur within previously irradiated fields. The GOG223 compared cisplatin alone with cisplatin plus ifosfamide for patients with advanced or recurrent cervical cancer. Although the cisplatin-ifosfamide combination produced a higher response rate than cisplatin alone (33% versus 19%, P = .02), the cisplatin-ifosfamide combination also produced significantly more leukopenia, peripheral neuropathy, renal toxicity, and encephalopathy without improving the overall median survival time. In a subsequent study, Long and colleagues224,225 reported results of GOG 179, a randomized trial comparing three chemotherapy regimens: methotrexate, vinblastine, doxorubicin, and cisplatin (MVAC); topotecan and cisplatin (TC); and cisplatin alone. In 2001, the MVAC arm of the trial was closed after a 94% rate of grade 3 or 4 neutropenia and four deaths from neutropenic sepsis. Patients continued to be accrued to the remaining two treatment groups until the study was closed in 2002. A comparison between treatment with cisplatin only (146 patients) and treatment with TC (147 patients) was published in 2005.224 Rates of grade 3 or 4 neutropenia were much higher in the TC group than in the cisplatin-only group (70% versus 1%), but patients in the TC group had longer median survival (9.4 months versus 6.5 months, P = .02) and longer progression-free survival (4.6 months versus 2.9 months, P = .01) than those treated with cisplatin only. Radiation therapy can provide effective pain relief for localized metastases in bone, brain, lymph nodes, or other sites. Hypofractionated pelvic radiation therapy also can relieve pain and reduce bleeding for patients who present with stage IVB disease.226-228 Effective control of bleeding is of added importance if the patient’s treatment plan includes cytotoxic chemotherapy.

Radiation Therapy The goal of radiation therapy for cervical cancer is to deliver adequate doses over an acceptable period without exceeding normal tissue tolerance limits. The required dose varies according to the tumor burden in the cervix, paracervical sites, and regional lymph nodes. Factors that influence the tolerable dose of radiation include the patient’s vaginal and uterine anatomy, the degree of tumor-related tissue destruction and infection, and patient characteristics (e.g., body habitus, comorbid illnesses, smoking habits). The therapeutic ratio is optimized through careful integration of EBRT and brachytherapy, with the dose and intensity of treatment tailored according to estimates of the disease extent and the risk of complications.

External Beam Radiation Therapy Microscopic disease in regional lymph nodes can usually be controlled with an external beam radiation dose of 40 to 50 Gy. However, this dose may not be sufficient to control

29  The Uterine Cervix

753

FIGURE 29-9.  Example of anteroposterior and lateral fields that may be used to treat a patient with cervical cancer. Shading indicates the location of the common, external, and internal iliac lymph nodes and the approximate contour of this patient’s 6- to 7-cm endocervical tumor and upper vagina. Lighter shading on the lateral view indicates the contour of the bladder and distal rectum taken from a planning CT scan. When lateral fields are used, blocks must be designed to cover the regional lymph nodes and tumor, taking into account the variable position of the uterus and cervix in the pelvis.

some areas of heavy microscopic involvement. In particular, the sites of close or positive margins or extracapsular lymph node extension probably require a somewhat higher dose (e.g., 50 to 60 Gy). The risk of metastases within clinically uninvolved paracervical tissues in patients with locally advanced disease probably varies. If the uterus has not been removed, the combined dose from EBRT and ICRT addresses this risk. Adequate treatment of the paracervical tissues is more difficult after a hysterectomy has been performed because most of the paracervical tissues are no longer within the range of effective intracavitary therapy. Patients with stage IB to IVA cervical carcinoma require irradiation of the external iliac, internal iliac, common iliac, and presacral nodes. If the posterior vaginal wall is extensively involved, the perirectal nodes also must be treated. If cancer is present in the distal vagina, the inguinal nodes must be irradiated. EBRT fields must include the cervix (or operative bed) and paracervical tissues, including the broad ligaments and uterosacral ligaments. Treatment fields must be designed with the understanding that the central pelvic tissues are not static; the positions of the uterus, cervix, and upper vagina can change by 3 cm or more, depending on the filling of the bladder and rectum.229

Integration with Brachytherapy In addition to helping control regional metastasis, EBRT plays an important role in reducing the volume of central disease before intracavitary brachytherapy. Reduction of central disease volume facilitates insertion of brachytherapy applicators and permits delivery of a higher dose to gross disease by reducing the distance between the tumor periphery and intracavitary sources. Although patients are commonly treated with 40 to 45 Gy to the pelvis before brachytherapy, they should be examined regularly during treatment to determine the optimal point at which to begin brachytherapy. Because EBRT tends to cause some contraction of the upper vagina (limiting the size and packing of vaginal applicators), patients who have a narrow vagina should be considered for intracavitary therapy as soon

as the cervix is 4 cm or less in diameter. However, when brachytherapy treatments are scheduled in the middle of EBRT, the procedures should be scheduled carefully to avoid more than a 2- or 3-day interruption in treatment of the regional nodes.

Field Arrangements for Treatment of the Whole Pelvis The most common field arrangement for treatment of the whole pelvis consists of four fields (anteroposterior [AP], posteroanterior [PA], and right and left lateral). When radiation oncologists had only cobalt 60 (60Co) or 4- to 6-MV beams available for EBRT, four fields were needed to avoid delivering an unacceptably high dose to superficial tissues. More penetrating megavoltage beams (≥15 MV) are widely available; with these beams, two-field techniques produce relatively homogeneous dose distributions, even when the patient’s anteroposterior separation is 25 cm or more. However, most clinicians still use a four-field technique (Fig. 29-9) to reduce the dose to the anterior small bowel or rectum.55 Although in some cases—particularly after hysterectomy—this four-field approach can significantly reduce the dose of radiation to the small bowel, the use of two opposing anterior and posterior fields may be preferred when the patient has locoregionally advanced disease. If the tumor is bulky and lateral fields are used, relatively little of the pelvis can be spared treatment after the tumor, paracervix, and regional nodes are included with a margin (Fig. 29-10). Shielded lateral fields can result in undertreatment if the tumor size or organ motion is underestimated; shielded lateral fields also irradiate additional bone marrow that may be needed if the patient is to undergo chemoradiation therapy. Pelvic fields with an upper border at the L4 to L5 interspace are sufficient to cover primary-echelon lymph nodes with a margin. Somewhat smaller fields may be used for patients with small tumors or for patients receiving postoperative radiation therapy if a pelvic lymph node ­dissection showed no evidence of regional metastasis. However, patients with involvement of the common iliac or ­

754

PART 9  Female Genital Tract

A

B

FIGURE 29-10.  Sagittal MRI demonstrates a large, expanding endocervical cancer. A, Although this tumor was still clinically confined to the cervix and was mobile on pelvic examination, it almost filled the pelvis. In this case, little protection of normal tissues would be achieved with lateral pelvic fields. B, After 45 Gy of external beam irradiation and concurrent chemotherapy, there was an excellent partial response of the tumor.

para-aortic lymph nodes require treatment with larger fields. In general, the upper border of the radiation field should be at least 4 to 5 cm (or two vertebral bodies) above known disease. However, it is important to recognize the existence of a gradient of risk from the primary-echelon nodes to the upper para-aortic nodes. Placement of the upper field border is often based on a qualitative assessment of the risk of microscopic disease compared with the added risk of bowel complications from extended fields.

have been presented about whether IMRT can be used safely as an alternative to brachytherapy for the treatment of intact cervical cancer. The very high dose and dose gradients that can be achieved with intracavitary brachytherapy cannot be reproduced accurately with EBRT. Differences in the volumes of the bladder, rectum, and tumor that occur during the course of radiation therapy can produce substantial and unpredictable variation in the positions of the target and adjacent critical structures.229,235

Intensity-Modulated Radiation Therapy

Radiation Dose

Considerable interest has been expressed in the use of IMRT to treat the whole pelvis, particularly after hysterectomy for cervical or endometrial cancer (Fig. 29-11).230-233 Comparative analyses of radiation dose distributions indicate IMRT may allow the integral dose to bowel to be reduced relative to treatment with more standard fourfield conformal techniques.230,233 In preliminary analyses, Mundt and colleagues234 reported that patients treated with IMRT experienced less acute gastrointestinal toxicity and that IMRT plans could be formulated that reduce the bone marrow dose relative to four-field pelvic radiation therapy234; however, treatments that are designed to spare bone marrow probably spare the bowel less effectively. Although IMRT has promise as a means of reducing side effects from radiation therapy delivered after hysterectomy, it is also costly, time-consuming, and requires a comprehensive understanding of radiographic anatomy. Care must always be taken to account for intratreatment variations in the positions of the vaginal and paravaginal tissues. Some investigators have suggested that IMRT could be used to treat the primary tumor site in patients with locally advanced cervical cancer. There is little doubt that IMRT, when carefully applied, can safely deliver higher-than-usual doses to regions of gross disease, particularly in nodal sites that lie adjacent to critical structures. However, no data

Patients who have or are at risk for no more than microscopic regional disease can usually be treated effectively with 45 to 50 Gy of EBRT to the pelvic lymph nodes. If gross lymphadenopathy or extracapsular extension is discovered after surgical excision, small volumes may be boosted to a total dose of 56 to 66 Gy (including the contribution from intracavitary treatments) with conventional external beam techniques. Bulky nodes can be treated to doses of 65 Gy or more by using carefully planned conformal radiation therapy techniques that severely constrain the volume of bowel that is exposed to 60 Gy or more.

Field Arrangements for Treatment of the Para­aortic Region in Combination with the Pelvis A variety of techniques can be used to treat the para-­aortic region in combination with the pelvis. Although the entire volume can be encompassed in a single large pair of opposing AP and PA fields, this technique usually exposes a large amount of small bowel to the same dose of radiation as the lymph nodes. Alternatively, the entire volume can be treated with four fields (AP, PA, and two ­ lateral fields), or the pelvis can be treated with AP-PA fields and a different technique can be used for the para-aortic region. When the entire volume is treated in continuity, field-in-field techniques may be required to achieve a

29  The Uterine Cervix

A

755

B

FIGURE 29-11.  Radiation isodose distributions for intensity-modulated radiation therapy of the vagina, paravaginal tissues, and pelvic lymph nodes in a patient treated after hysterectomy for cervical cancer. The two scans show an axial view through the external iliac nodes (A) and a midline sagittal view through the vagina, presacral, and common iliac lymph nodes (B). (From Eifel P. Intensitymodulated radiation therapy for carcinomas of the uterine cervix and endometrium. In Bortfeld T, Schmidt-Ullrich R, De Neve, et al [eds]: Image-Guided IMRT. Berlin: Springer, 2006.)

homogeneous dose distribution. If the central axis of the radiation field is placed at the L4 to L5 or L5 to S1 interspace, the upper and lower half of the volume can be treated using different field arrangements without complex matching problems between the fields (Fig. 29-12). In some cases, IMRT may provide greater tissue-sparing than conventional techniques,236 although the technical concerns described in the previous section continue to pertain to treatment of extended fields.

Field Arrangement for Boost to Parametrium or Positive Pelvic Nodes In the case of lateral involvement of the paracervical tissues or positive pelvic nodes, small anterior and posterior opposing fields can be used to boost the parametrium. The goal is to deliver a total dose of 60 to 65 Gy to areas of gross disease or extracapsular extension. The match between the intracavitary system and external fields is complex. In general, boost fields should cover the implants and extend about 0.5 cm lateral to the colpostats (usually at about the 0.5 Gy/hr line). Some clinicians place a midline block during 50% or more of pelvic EBRT, compensating by giving more treatment with ICRT. This technique results in a lower total dose of radiation to portions of the bowel and bladder but may undertreat uterosacral disease.237 When more than one half of the pelvic treatment is delivered using a midline block, care should be taken not to overtreat the ureters; if they are included in EBRT fields and are exposed to a higher dose from intracavitary treatment, the risk of ureteral stricture may be increased.238

Treatment of Bulky Lymph Node Disease In selected cases in which bulky lymph node disease cannot be controlled with 60 Gy of radiation, conformal therapy or IMRT can be used to boost a small volume of tissue to a total dose (including the brachytherapy contribution) of up to 70 Gy. However, tightly conforming fields typically are not used to treat the cervix because of the substantial organ motion that can occur and because brachytherapy usually can be used to deliver even higher doses to the central target volume.

Brachytherapy The ability to control large cervical tumors with radiation therapy results in large part from the successful integration of EBRT and intracavitary brachytherapy. Fortunately, the uterus and vagina are ideal receptacles for intracavitary applicators for several reasons. Sources can be placed in the center of the tumor; sources can usually be positioned above and below the tumor (in the fundus and vagina), adding to the lateral dose penetration and improving coverage of the tumor periphery; the distensible vagina can be packed to displace rectum and bladder from intracavitary sources; and the uterus and vagina are relatively resistant to radiation damage. Traditionally, ICRT has been delivered at a low dose rate (LDR) using cesium 137 (137Cs), radium 226 (226Ra), or occasionally, 60Co sources. Although some clinicians have shifted from using this approach to using remote afterloading high-dose-rate (HDR) systems, LDR systems continue to have several advantages. Foremost is the radiobiologic advantage achieved when radiation is delivered at a rate of

756

PART 9  Female Genital Tract

48 40 Gy

45 Gy

FIGURE 29-12.  Extended fields can be treated by using a splitbeam technique that permits the use of different field arrangements above and below the central axis. In this case, very little small bowel was in the pelvis, which was treated using an anteroposterior-posteroanterior (AP-PA) technique that minimized the amount of bone marrow in the field. The para-aortic nodes were treated using AP, PA, and two lateral fields. The match was moved by 0.5 cm twice during treatment (not included in the plan shown).

less than 0.6 to 0.8 Gy/hr. At these rates, the recovery of sublethal injury in late-reacting normal tissues is similar to that achieved with EBRT delivered in fractions of 1.8 to 2 Gy. The use of pulsed-dose-rate (PDR) brachytherapy has also gained interest because it combines some of the advantages of remote-afterloading stepping-source technology with the radiobiologic advantages of LDR brachytherapy. These topics are discussed further in the “Pulsed-DoseRate Brachytherapy” section.

Low-Dose-Rate Therapy for Intact Cervical Tumors Isotopes for use in gynecologic LDR intracavitary therapy are usually encapsulated in 5- to 30-mg radium equivalent (RaEq) sources that are 17 to 22 mm long and 3 to 5 mm in diameter. Remote afterloading devices usually contain spherical cesium pellets of approximately 5 mg RaEq each. Sources are positioned in the uterus and vagina by using one of a variety of applicator systems. In early systems, applicators were preloaded with radiation sources before insertion into patients. Modern applicators are positioned in the uterus before the radioactive sources are inserted manually or remotely. Almost all applicator types include a hollow tube (tandem) that is inserted in the uterus. Intrauterine tandems may vary somewhat in their length, curvature, and caliber, but the most important variations between

a­ pplicator systems relate to the arrangement of sources in the vagina. Variations of the Fletcher-Suit-Delclos applicator system are the most common in use in the United States (Fig. 29-13). With this system, vaginal sources are positioned in the center of two cylindrical colpostats; the sources are roughly perpendicular to the axis of the intrauterine line of sources and are 1 to 1.5 cm from the surface of the cervix depending on the diameter of the colpostats (Fig. 29-14A). The anisotropic distribution of dose surrounding the linear sources and shielding in the inferomedial portion of the colpostats reduce the dose of radiation given to the bladder and rectum from the vaginal sources. If the vagina is too narrow to accommodate colpostats or if the lower vagina is involved with cancer, sources are positioned in the vaginal portion of the tandem and are displaced from the vagina with Lucite cylinders (see Fig. 29-14B). Other applicator systems arrange vaginal sources parallel to the tandem (e.g., Henschke applicators) or place the sources perpendicular to the tandem but closer to the cervix (e.g., mini-ovoids,239 French mould techniques240). Arrangement of Vaginal Sources, Packing of Vaginal Applicators, and Positioning of Tandem. A detailed discussion of intracavitary brachytherapy technique is beyond the scope of this chapter, but the most crucial points are outlined here. Sources are positioned to maximize the ratio between the dose to the cervix and paracervical tissues and the dose to the bladder and rectum. Sources in the endocervix deliver a very high dose to central disease; sources in the uterus and vagina expand the dose distribution to cover the periphery of the cervix, increase the dose to paracervical tissues, and treat the lower uterine segment and vagina. The arrangement and position of vaginal sources, vaginal packing, and the position of the tandem can be adjusted in several ways to minimize rectal and bladder exposure. The tandem is usually positioned in the midsagittal plane of the pelvis about two thirds of the way between the sacrum and pubis. If eccentric involvement of one side of the parametrium is present, the dose to that side can be increased by permitting some lateral deviation of the tandem. The colpostats should be up against the cervix; caudad displacement should be minimized because it reduces the dose to the tumor without decreasing the dose to normal tissues. The colpostats should be centered on the cervix. In most cases, centering them in this manner causes the colpostats to be approximately bisected by the ovoids. Excessive posterior displacement of the ovoids results in an underdosage of the anterior cervical lip and can lead to severe overdosage of rectum draped over the protruding posterior vagina (Fig. 29-15). The vaginal applicators should be carefully packed in place, displacing the vagina (and bladder and rectum) away from the sources. It is particularly important to avoid an anterior dip in the packing caudad to the colpostats; failure to adequately pack this area also exposes the rectum to an unnecessarily high dose. In most cases, the tandem should be placed up against the fundus of the uterus or high enough to permit loading of 6 to 8 cm of the intrauterine tandem. If the vagina is sufficiently ample, caps should be placed on the Fletcher-Suit-Delclos colpostats. They facilitate positioning and permit the use of somewhat greater activity in the vaginal applicators, improving the dose to large tumors and exposing more of the upper vaginal surface to a high radiation dose.

29  The Uterine Cervix

A

757

C

B

E

D

F

G

FIGURE 29-13.  The Fletcher-Suit-Delclos applicator system. A, Mini-colpostats. B, Small (2-cm) colpostats with various caps used to increase their diameter and length. C, Colpostats with various angles. D, Tandem and dummy markers. E, Metallic flanges used to set the length of the intrauterine tandem. F, Four tandems with various curvatures. G, Lucite cylinders with a central hole to accommodate the central tandem (used for patients who have a narrow vagina or extensive vaginal disease).

Even the most experienced practitioners occasionally perforate the uterus when they are sounding and dilating a cancerous cervix. The most common site of perforation is the posterior endocervix or lower uterine segment. Although this problem is sometimes appreciated when the sound penetrates beyond 8 cm without reaching a fundus, the findings can be more subtle if the tip of the probe passes into paracervical tissues or rests against the sacrum or other pelvic tissues. Uterine perforation can lead to disastrous secondary complications and underdosage of tumor. Whenever uterine perforation is a possible risk, the position of the tandem should be verified with CT or sonography. If a perforated uterus is subsequently probed successfully, the tandem can be left in place under antibiotic coverage.241 Like any operative technique, optimal intracavitary placement requires skill and experience regardless of whether LDR or HDR equipment is used. Because ­locally

advanced cervical cancer is an uncommon disease in most U.S. communities, radiation oncologists in the United States are not uniformly skilled in brachytherapy procedures. Some studies have shown that patients with cervical cancer who are treated at small facilities are more likely to not receive brachytherapy and to have protracted treatment courses.55 Because brachytherapy is a critical part of cervical cancer treatment, practitioners who treat very few cases of intact cervical cancer should consider referring patients to centers with greater experience. Selection of Sources and Calculation of Delivered Dose. Intracavitary treatments for carcinoma of the ­cervix are usually reported in terms of the doses to specific pelvic reference points (points A and B) that were initially defined as part of the Paris and Manchester treatment methods.242 In 1985, the International Commission on Radiation Units and Measurements (ICRU)243 described

758

PART 9  Female Genital Tract

75

50

30

75

20

A

50 30 20

B

FIGURE 29-14.  Anteroposterior (A) and lateral (B) radiographs of a Fletcher-Suit-Delclos tandem and small (2-cm) colpostats. The tandem bisects the colpostats on the lateral view and is in a roughly central position between the sacrum and bladder. The colpostats are within 0.5 cm of the marker seeds inserted in the anterior and posterior lips of the cervix. The vaginal packing (visible because of a radiopaque thread in the center of the packing) displaces the vagina away from the vaginal sources. The tandem was loaded with 35 mg RaEq of cesium, and the colpostats each contained 10 mg RaEq of cesium. Contours of the 20, 30, 50, and 75 cGy/hr isodose lines are shown superimposed on the radiographs. The dose rate at point A was approximately 50 cGy/hr.

the classic ­Manchester definition of point A as “a point 2 cm lateral to the central canal of the uterus and 2 cm up from the mucous membrane of the lateral fornix in the axis of the uterus” (Fig. 29-16). Point B was defined as “being in the transverse axis through points A, 5 cm from the midline.” Despite this and other attempts to standardize their definitions, points A and B continue to be calculated in a variety of ways, contributing to discrepancies between reporting institutions. Because point A is in a region of the pelvis where the dose of radiation from intracavitary sources is changing rapidly, small inaccuracies in the placement of point A can produce significant differences in the calculated result. In its report, the ICRU suggested an alternative method of reporting that included an estimate of the volume treated to 60 Gy and the reference air kerma. This approach never gained wide acceptance, and most clinicians continue to report their treatments in terms of the dose to point A. Although the dose to point A can be used as an approximate estimate of dose intensity, the dose to point A must be considered in the context of many other factors in designing treatments. At M.  D. Anderson Cancer Center, the quality of the intracavitary placement is the first consideration.244 If the applicator position or vaginal packing fails to meet acceptable standards, the applicator is adjusted and repacked. When the position has been optimized, the uterine tandem and vaginal applicators are loaded with relatively standard source combinations that deliver radiation to adjacent tissues at an acceptable rate. Typically, 30

to 40 mg RaEq is distributed in a tandem 6 to 8 cm long using three 10- to 15-mg RaEq sources (spacers may be required). If the tandem is well positioned, its tip may be loaded with 15 mg RaEq to improve the dose to the endocervix and lateral tissues; if the endocervix is expanded, a 15-mg RaEq source also may be used in the mid position to improve the distribution of dose to that region. The distal source should not protrude more than a few millimeters from the external cervical os when colpostats are being used; the use of a protruding source to supplement the dose between widely separated colpostats seems to increase the risk of vaginal fistulas.245 The loading of vaginal colpostats is usually chosen to give a dose rate of 1 Gy/hr or less at the vaginal surface. Lower dose rates are preferred when unshielded colpostats (i.e., mini-ovoids) or vaginal cylinders are used. Although most late treatment complications (e.g., anterior rectal ulcers, posterior bladder ulcers, fistulas) are probably the consequence of a high dose delivered to a relatively small volume of tissue, other complications (e.g., reduced bladder capacity, severe pelvic fibrosis) can result from a moderately high dose to a large volume. These types of complications are seen when more than 50 to 60 Gy is delivered to a large portion of the pelvis. Calculations of total reference air kerma or total mgRaEq hours are used as surrogates for the integral dose to the pelvis from ICRT. The total number of mg-RaEq hours of intracavitary treatment is generally limited to no more than 6000 to 6500 after 40 to 45 Gy to the whole pelvis

29  The Uterine Cervix 2 cm

B

A

759

3 cm

A

B

2 cm

FIGURE 29-15.  In this case, posterior displacement of the vaginal colpostats would have allowed the rectum to drape over the vaginal sources (arrow), producing a high rectal dose cephalad to the vaginal fornix. This problem was identified in the operating room, and the system was repositioned, improving the relationship between the tandem and colpostats.

(or 7500 to 8000 after 20 Gy to the whole pelvis). With typical loadings (usually equal to 65 mg RaEq), these limits are not exceeded in two 48-hour applications. However, when a capacious vagina or long tandem accommodates more than 65 to 70 mg RaEq, it is usually necessary to reduce the time the implants are left in place or their total activity. Although prescribing intracavitary treatments only to point A is dangerous, implants that adhere to the guidelines described usually result in a radiation dose rate of about 40 to 45 cGy/hr to point A. Steep dose gradients, organ motion, and anatomic changes that occur between implant placements make it very difficult to calculate accurate maximum or total organ doses from intracavitary implants. Most clinicians try to calculate doses to normal tissue reference points; these are selected in a wide variety of ways,246-249 including the method recommended by the ICRU in 1985 (Fig. 29-17).243 Increasing interest is being devoted to the use of image-based three-dimensional treatment planning for intracavitary and interstitial brachytherapy. Among the many potential advantages of image-based planning are its ability to provide a better sense of the true doses delivered to critical structures and possibly a more solid basis for comparisons between institutions. Studies of CT-based treatment planning have demonstrated that depending on the geometry of intracavitary systems, standard normal tissue reference points may be wildly inaccurate.248,250,251 However, no reliable data are available for comparing volumetric ­ calculations of critical structure doses with late

FIGURE 29-16.  In the classic Manchester system, point A is defined as a point 2 cm lateral to the central canal of the uterus and 2 cm up from the mucous membrane of the lateral fornix. In clinical practice, dose calculations are usually made from radiographs, and point A is taken 2 cm up from the flange of the uterine tandem and 2 cm lateral from the central canal.243 Alternatively, some clinicians use radiopaque markers in the cervix or a point midway between the surface of the vaginal colpostats to determine the starting point.

Balloon 7 cm3

Bladder reference point Intrauterine sources

Intravaginal sources

Vaginal posterior wall Rectal reference point

0.5 cm

FIGURE 29-17.  Rectal and bladder reference points according to the International Commission on Radiation Units and Measurements Report 38. (From International Commission on Radiation Units and Measurements. Dose and Volume Specification for Reporting Intracavitary Therapy in Gynecology. Bethesda: International Commission on Radiation Units and Measurements, 1985.)

760

PART 9  Female Genital Tract

c­ omplications, and ­ clinicians should be cautious about changing their practice without having this information. Nevertheless, image-based planning for intracavitary treatment is a burgeoning area of research that is expected to lead to improved treatment guidelines in the future. In the meantime, in the interest of achieving some consistency in nomenclature, the gynecology group of the Groupe ­Européen de Curiethérapie-European Society for Therapeutic Radiation and Oncology (GYN GEC-ESTRO) published a set of recommendations and target volume definitions that have been adopted by many investigators studying the use of three-dimensional image-based treatment planning for cervical brachytherapy.252

High-Dose-Rate Therapy for Intact Cervical Tumors In the past, radiation safety requirements prohibited handling of very high activity sources, necessitating the delivery of brachytherapy at relatively low dose rates. It eventually became apparent that LDR treatment had definite radiobiologic advantages, and the ability to administer ICRT at dose rates of 0.4 to 0.6 Gy/hr (to point A) was thought to be a major contributor to the success of cervical cancer treatment. Efforts to shorten the duration of ICRT by increasing the dose rate by as little as twofold (to 0.8 to 1.0 Gy/hr) have led to increased complication rates even after compensatory dose reductions.253 In the 1980s and 1990s, the advent of computer­controlled remote-afterloading technology made it possible to accurately deliver brachytherapy treatments at low dose rates or at very high dose rates (>1 Gy/min) without exposing personnel to ionizing radiation. The possibility that treatment could be delivered entirely within the radiation therapy department without inpatient hospitalization led vendors and investigators to develop methods and protocols for the delivery of HDR intracavitary therapy for patients with cervical cancer. Although many clinicians now consider HDR brachytherapy to be within the standard of care, it has never been rigorously compared with LDR brachytherapy. Consensus continues to be limited regarding which of the many techniques and fractionation schemes being used to treat patients is best. Three groups have tried to perform randomized trials comparing LDR and HDR brachytherapy.254-257 Although these investigators concluded that the two approaches were similar with respect to the effectiveness and toxicity of therapy, problems with the methods of treatment allocation, relatively small numbers of patients, and other flaws in the designs of these trials make it impossible to generalize the investigators’ conclusions about treatment efficacy.258,259 In the most recent comparison, a study of 237 patients treated in Thailand between 1995 and 2001,257 outcomes were similar for patients treated with HDR brachytherapy and for those treated with continuous brachytherapy exposures with 137Cs. However, the continuous treatment referred to as “LDR” in that study was delivered at a rate of 150 to 180 cGy/hr at point A. These dose rates, which are three to four times higher than those used by most investigators of LDR intracavitary cervical cancer treatment, would not be expected to have the same radiobiologic advantages over HDR as the more usual treatments given at 40 to 50 cGy/hr. Retrospective experiences demonstrate that combinations of EBRT and HDR brachytherapy can be curative

in a substantial proportion of cases.183,185,260-263 However, differences in the selection of patients, staging of tumors, and reporting of results make it difficult to compare experiences. Although these experiences are encouraging, they are insufficient to permit detection of moderate differences in outcome after LDR versus after HDR, particularly for subsets of patients with high-risk disease. Most of the published experiences with HDR brachytherapy are from Asia, Germany, and England.183,254-257, 260,261,264 Although the use of HDR brachytherapy in the United States has increased during the past 2 decades, few systematic studies of this technique have been conducted, and the experiences that have been reported include few patients. The University of Wisconsin experience is one of the largest.185 Investigators compared results for 191 patients treated with LDR brachytherapy between 1977 and 1988 and 173 patients treated with HDR brachytherapy after 1988. Although the survival rates with the HDR and LDR approaches were similar for patients with small tumors, the investigators reported a significantly lower survival rate for patients with stage IIIB disease who were treated with HDR brachytherapy (Table 29-8).183,185,261,265,266 Kapp and colleagues183 also reported encouraging survival rates for patients with small tumors, but the survival rates for patients with larger tumors were relatively low. Arai and colleagues267 reported high survival rates for patients with all stages of disease, but the total doses they recommended for HDR or LDR therapy were lower than conventional doses by 30% or more, making it difficult to reconcile their results with other experiences. Although the radiobiologic disadvantages of HDR intracavitary therapy can undoubtedly be overcome if the treatment is divided into a sufficiently large number of small fractions, doing so may compromise the purported advantages of the technique in terms of patient and physician convenience and cost. The radiobiologic disadvantage of HDR is probably overcome when adjacent critical structures can be displaced well away from the high-dose-perfraction portion of the intracavitary dose distribution, but this situation is not always possible. Although it has often been assumed that the risk of acute complications is lower with HDR brachytherapy, detailed studies suggest that this may not be true.268,269 HDR brachytherapy is gaining increasing acceptance in the community, but its widespread use to treat patients with intact cervical cancer continues to arouse controversy.

Pulsed-Dose-Rate Brachytherapy In PDR brachytherapy, a relatively new alternative to traditional methods of brachytherapy, treatment is delivered in intermittent pulses by using a stepping source of iridium 192 and remote-afterloading equipment very similar to that used for HDR brachytherapy. Unlike HDR units, which contain a radioactive source of approximately 10 Ci, PDR units typically contain a source of only 0.5 to 1 Ci. When treatment is delivered in hourly pulses of less than 50 to 70 cGy, the radiobiologic effect seems to be very similar to that of LDR.270 Although PDR treatments for cervical cancer are given in the hospital, they may be more comfortable for patients, because they have access to nursing and visitation during the intervals between pulses. Shielding requirements are similar to those for LDR. Other advantages of PDR brachytherapy over remote-­afterloading cesium units

761

29  The Uterine Cervix

TABLE 29-8.  Selected High-Dose-Rate Brachytherapy Regimens and Reported Treatment Outcome EBRT Dose (Gy) Study and Reference Kapp et al183

Petereit et al185

Utley et al265

Nakano et al261

Year 1998

1999

1984

2005

Tumor ­Description

Whole Pelvis

Parametrium

HDR Brachytherapy Schedule Dose per ­Fraction

Number of ­Fractions

Frequency Timing

Number of 5-Year ­Patients DSS (%)

4.7-5.3*

4

2 ×/wk

After EBRT

29

96

<3 cm

44-50.4

3-6 cm

30-60

After EBRT

97

66

>6 cm

>60

After EBRT

55

28

IB nonbulky

0

60

9.1-9.9

5

Weekly

52

85

IB bulky, IIA

20-30†

30-40

7.2-8.2†

5

Weekly

39

—

IIB

51†

9

3.7-4.9†

5

Weekly

50

69

III

60†

—

3.7-4.3†

5

Weekly

50

33

I

20

30

5

8-10

2 ×/wk

Concurrent

29

89

II

20-30

50

5

8-10

2 ×/wk

Concurrent

50

58

IIIB

20-50

50

5

8-10

2 ×/wk

Concurrent

43

33

IB‡

0

45

5.8

5

Weekly

Concurrent

147

94

II (small)

0

50

5.8

5

Weekly

Concurrent

256

80

II (large)

20

30

5.75

4

Weekly

Concurrent

—

—

III (small)

20-30

50

5.75

4

Weekly

Concurrent

515

66

III (large)

20-30

50

5-6

3-4

Weekly

Concurrent

—

—

*Schedules

evolved during the years covered in this study. These numbers represent the most recent schedules. schedules were changed in the final year of the study. Patients now receive 45 to 50 Gy to the whole pelvis and 5 fractions of HDR brachytherapy of 5.5 to 6 Gy each.185 ‡The study authors described tumors as smaller than 3 cm in 60%; only one stage IB tumor was larger than 5 cm. DSS, disease-specific survival; EBRT, external beam radiation therapy. †These

are improved optimization tools, the ability to use the same applicators for HDR and LDR, and the ability to use the units for interstitial applications and for cervical intracavitary brachytherapy. Although PDR brachytherapy has been used widely in Europe for more than 10 years,271 it only recently became available for use in the United States after changes were implemented in the regulatory requirements governing use of the necessary high-activity sources.

Interstitial Therapy for Intact Cervical Tumors Some radiation oncologists advocate the use of interstitial brachytherapy as an alternative to ICRT for selected patients with cervical cancer. Although criteria differ, patients with stage III disease or with poor vaginal anatomy are most often considered candidates for interstitial brachytherapy. To place interstitial implants for the treatment of patients with cervical cancer, most practitioners use a peri­neal template to guide evenly spaced needles into the cervix and paracervical tissues. An obturator in the vagina helps to determine the direction of needle insertion (Fig. 29-18). Supporters of interstitial therapy describe as advantages of this method the relatively homogeneous dose, the ease of inserting implants in patients whose uteri are difficult

to probe, and the ability to place sources directly into the parametrium. Even though interstitial templates have been used to treat cervical cancer for many years, the ­reported series have been small, and patient follow-up time has rarely been sufficient to permit calculation of long-term survival rates.272-274 In an early series, Syed and colleagues274 reported a projected 5-year survival rate of 53% for 26 patients with stage IIIB disease. However, survival results from two subsequent reports were disappointing. In a review of the combined experiences of Stanford University Medical Center and the Joint Center for Radiation Therapy,275 Hughes-Davies and colleagues reported 3-year disease-free survival rates of 36% for patients with stage IIB disease and only 18% for patients with stage IIIB disease; the corresponding local control rates were 22% and 44%, and in patients with local control, the rate of complications requiring surgical intervention was high. A report of the University of California-Irvine experience described 5-year survival rates of 21% and 29%, respectively, for patients with stage IIB and IIIB disease, also with high complication rates.276 These survival rates do not seem any better than those achieved with combinations of EBRT and ICRT (55% to 75% for IIB disease and 40% to 50% for IIIB).

762

PART 9  Female Genital Tract OFFSET

0.0

20

40

Isodoses 200 120 100 90 80 60 40 20

60 80

120

90

POINT DOSES X Y VALUE 0.0 0.0 140 0.8 0.4 126 0.8 0.3 127 0.8 0.1 132

100 100 200 8090

40

60

20

A

B

FIGURE 29-18.  A, Anterior view of a perineal implant placed for treatment of a proximal stage IIB carcinoma of the cervical stump using a perineal template and 10 needles. The screw just inferior to the opacified Foley balloon marks the superior end of the vaginal obturator. B, Dose distribution of the implant seen in A at the geometric center of the sources.

100

Percent of dose at surface

Some investigators have tried to improve the accuracy of needle placement by using laparotomy or sonographic guidance.277,278 However, no definitive evidence has been found to indicate that interstitial therapy improves outcome for patients with advanced intact cervical lesions. Interstitial brachytherapy may be useful for treating recurrent cervical cancer or, in combination with intracavitary brachytherapy, for lesions involving the distal vagina.278-280 The techniques are similar to those used to treat primary vaginal cancers.281 Laparoscopic guidance can be particularly helpful for accurate placement of needles in patients who have had a previous hysterectomy, as described in the following paragraphs.

2 cm 3 cm 4 cm

80

60

40

20

Vaginal Brachytherapy after Hysterectomy ICRT may be used to treat the vagina in patients with cervical cancer who have had a hysterectomy. Usually, vaginal brachytherapy is used to boost the dose to the vaginal apex in women who receive pelvic radiation therapy because of high-risk features in their hysterectomy specimen. The best argument for boosting this region to a higher dose than other tissues potentially at risk for microscopic disease is that the vaginal incision may harbor cells implanted during the surgical procedure. Patients who have disease at the vaginal mucosal margin are particularly good candidates for treatment with vaginal brachytherapy. However, it is important to recognize that most vaginal applicators have very shallow penetration (Fig. 29-19). Deep stromal margins (particularly lateral or anterior paracervical margins) are likely to lie beyond the reach of vaginal intracavitary therapy. A variety of applicators can be used to treat the vaginal cuff. Two vaginal ovoids can be used, although care must be taken to ensure that the vaginal incision is not packed away from the surface of the ovoids.282 At M.  D. Anderson, a domed vaginal cylinder is used to deliver LDR irradiation; the applicator is loaded with a single point source of cesium in the center of the hemispherical dome. Depending on the risk of recurrence in the vaginal cuff, a total dose between 30 Gy (prescribed at the vaginal surface) in

0 0

2

4

6

8

10

Distance from surface (mm)

FIGURE 29-19.  Dose of radiation delivered at various distances from the surface of a Delclos domed vaginal cylinder for cylinders of three different diameters. The rapid fall-off of dose must be considered in planning treatment, particularly when vaginal intracavitary treatment is used to treat gross or recurrent disease.

48 hours and 50 Gy in 72 hours may be given after 45 Gy of pelvic irradiation. The vaginal cuff can be treated by using HDR applicators with similar specifications. Many different schedules have been used. At M.  D. Anderson, a relatively conservative fractionation scheme of 3 fractions of 5 to 6 Gy prescribed at the vaginal surface is favored. When a positive vaginal margin requires delivery of a higher dose, we prefer to use LDR brachytherapy. It should always be remembered that the rectum is approximately 0.5 cm from the surface of the vagina. We prefer not to exceed a dose of 3 Gy/fraction at this depth out of concern that higher doses per fraction could produce an increased risk of rectal injury. However,

29  The Uterine Cervix

Probability of pelvic control (%)

100 80 60 40 20 b 0 30

40

50

60

70

80

Treatment time (days)

FIGURE 29-20.  Pelvic control as a function of treatment time for 279 patients with stage III or IVA carcinoma of the cervix treated with curative intent at Princess Margaret Hospital. (From Fyles A, Keane TJ, Barton M, et al. The effect of treatment duration in the local control of cervix cancer. Radiother Oncol 1992;25:273-279.)

many of the fractionation schemes recommended at other centers do deliver larger doses per fraction to the rectum.

Treatment Duration Several studies of patients with cervical cancer treated with irradiation have demonstrated correlations between longer treatment duration and decreased pelvic disease control rates.68,283,284 Although patients with high-risk lesions may have longer treatment courses,285 the results of these studies and radiobiologic principles strongly suggest that protraction of treatment beyond 7 to 8 weeks reduces the probability of pelvic disease control (Fig. 29-20).286 Results from the Patterns of Care studies287 and GOG trials288 suggest that treatment in many cases may require 10 weeks or more to complete. However, the average duration of treatment at facilities with large reported experiences (e.g., M.  D. Anderson and Washington University in St. Louis) tends to be shorter, with few patients requiring more than 8 to 9 weeks to complete the treatment.285 To minimize breaks in treatment, intracavitary therapy should begin no more than 1 to 7 days after the completion of whole-pelvis irradiation. Treatments to boost disease in the lateral parametrium or lymph nodes can be given between intracavitary treatments. If holidays or poor compliance extend a course of treatment, patients can be given two treatments, 6 to 8 hours apart, on 1 or 2 days.

Chemoradiation Therapy Concurrent Chemotherapy and Radiation Therapy In 1999 and 2000, the results of five prospective randomized trials demonstrated significant improvements in overall survival, disease-free survival, and local control rates when cisplatin-containing chemotherapy was administered concurrently with radiation therapy for patients with locoregionally advanced cervical cancer (see Table 29-5).85,172,180,173,174 These results stimulated an alert from the National Institutes of Health signaling a change

763

in the standard of care for many patients with locally advanced cervical cancer.289 In three of these randomized trials,180,173,174 the local treatment was radiation therapy alone, whereas the other two trials85,172 evaluated the role of chemotherapy in patients who received radiation before or after hysterectomy. However, despite differences in the eligibility criteria, the magnitude of the reduction in risk of recurrence was remarkably similar across all the studies. For five of the six treatment arms of the studies that included concurrent administration of cisplatin and radiation therapy, the risk of recurrence was reduced by 40% to 60% compared with the risk in the control group (radiation alone or radiation plus hydroxyurea). In these five trials, two schedules of chemotherapy emerged as superior. The first was a combination of cisplatin (75 mg/m2 every 3 weeks) with fluorouracil. One treatment group in the RTOG 90-01 trial involved receiving two or three cycles of this chemotherapy combined with pelvic radiation; this combination reduced the risk of recurrence by 52% compared with a control treatment of extended-field radiation therapy alone.180,290 The SWOG 87-97 trial85 of adjuvant treatment after hysterectomy for patients with high-risk disease compared pelvic radiation therapy with or without a similar combination of cisplatin and fluorouracil. The planned chemotherapy included two courses during pelvic radiation and two courses after the radiation; patients who received chemotherapy had a 50% reduction in the risk of recurrence. An earlier GOG trial (GOG-85174) compared radiation therapy and concurrent hydroxyurea with a combination of radiation and cisplatin and fluorouracil. The more conservative regimen of two cycles of cisplatin (50 mg/m2) and fluorouracil used in that trial was well tolerated but was associated with a less drastic, although still significant, risk reduction (21%). The second chemotherapy schedule was cisplatin at 40 mg/m2 weekly for up to 6 weeks during EBRT. This combination reduced the risk of recurrence by 40% to 50% compared with controls in the GOG-120 and GOG-123 trials.172,173 Weekly cisplatin was less toxic than a combination of cisplatin, fluorouracil, and hydroxyurea that achieved similar reductions in the risk of recurrence for patients with locally advanced disease.173 For these reasons, weekly cisplatin has been adopted as standard by the GOG and by many practicing clinicians. However, a later trial from the National Cancer Institute of Canada using the same regimen failed to demonstrate any significant improvement in outcome. Pelvic radiation alone was compared with a combination of weekly cisplatin and pelvic radiation (see Table 29-5).291 The eligibility criteria were similar to those of RTOG 90-01. No significant difference in overall survival was found between the two treatment groups (hazard ratio = 0.91; 95% confidence interval, 0.62 to 1.35). The Canadian trial was smaller than the earlier ones, resulting in relatively broad confidence intervals. The overall duration of radiation therapy for patients in the Canadian trial was 49 days, shorter than that of the GOG studies. The authors of the study argued that this more compact radiation therapy schedule may obviate chemotherapy. However, the 1-year pelvic disease control rates and overall survival rates reported in their initial abstract175 and in a subsequent publication291 are similar to those reported for the control group of RTOG 90-01, suggesting that the margin for improvement may have been similar.

764

PART 9  Female Genital Tract

The average total doses of cisplatin delivered with the weekly cisplatin regimens in GOG-120 and GOG-123 and the cisplatin plus fluorouracil regimens used in RTOG 90-01 and SWOG 87-97 were similar—approximately 200 mg/m2. The value of less intensive doses of cisplatin is unknown. The lower total dose of cisplatin in GOG-85 led to the smallest reduction in risk among the positive trials.174 A small trial of weekly cisplatin at a dose of 25 mg/m2 did not demonstrate an improvement over radiation therapy alone.292 The value of administering fluorouracil with radiation therapy has been well demonstrated in studies of gastrointestinal malignancies. Fluorouracil also has produced ­encouraging results in some smaller studies of patients with cervical cancer.293,176 Thomas and colleagues176 reported promising results from using a chronic continuous infusion of fluorouracil with radiation for patients with cervical ­cancer. However, a significant improvement was demonstrated in only one subset of the patients in that ­ trial— ­patients who had stage I or II disease and who received a single daily fraction rather than hyperfractionated radiation. In 2005, Lanciano and coworkers177 reported results of the GOG-165 trial comparing weekly cisplatin with a chronic continuous infusion of fluorouracil; that trial found no advantage for fluorouracil over weekly cisplatin. However, no trial has compared a combination of cisplatin with fluorouracil with cisplatin alone. RTOG 90-01171,180 was the only trial that encouraged the use of chemotherapy during ICRT. This is an interesting approach, particularly with LDR radiation, because a high dose of radiation is given over a short time (typically 20 to 25 Gy to point A over 2 days). Although the treatment seemed to be well tolerated, the benefits and risks of giving chemotherapy during brachytherapy have not been studied systematically. HDR ICRT was not permitted in any of the five trials previously discussed. Although HDR radiation is permitted as an alternative treatment in several prospective randomized trials, the best way of integrating HDR therapy with EBRT and chemotherapy and the risks of administering chemotherapy concurrently with large fractions of HDR ICRT are still incompletely understood. Although North American trials have focused on cisplatin-containing chemoradiation regimens, international trials have evaluated other regimens.178,179 In one such trial, Wong and colleagues179 randomly assigned 220 patients to receive radiation therapy only or radiation therapy with epirubicin administered every 4 weeks for five courses. Chemotherapy was begun on the first day of radiation therapy but was continued after radiation therapy until the five courses were completed. This group reported significantly better rates of disease-free survival (P = .03) and overall survival (P = .04) from the use of chemotherapy. In 2002, Lorvidhaya and colleagues178 published the results of a large Southeast Asian trial in which 926 patients with locally advanced cervical cancer were randomly assigned to receive one of four treatments: radiation therapy alone, radiation therapy with concurrent mitomycin and oral fluorouracil, radiation therapy with adjuvant fluorouracil, or radiation therapy with concurrent and adjuvant chemotherapy. Patients in the two treatment groups that included concurrent chemotherapy had higher diseasefree survival rates and lower rates of local recurrence than ­patients in other two groups. No evidence was found that

adjuvant fluorouracil improved the survival of patients treated initially with radiation alone or with radiation and concurrent chemotherapy. Preliminary results of a smaller randomized trial conducted in Venezuela294 suggested that mitomycin might be beneficial. Although the investigators did not find an increase in late complications after concurrent chemoradiation therapy with mitomycin C, an earlier study reported a correlation between use of mitomycin C and late bowel complications.213 All of the randomized trials of chemoradiation therapy excluded patients who had positive para-aortic nodes. However, because locoregional disease control can be ­ difficult to achieve in such patients, several investigators have evaluated the use of extended radiation fields with concurrent chemotherapy. Grigsby and colleagues295 ­reported ­unacceptable toxicity from a combination of hyperfractionated extended-field radiation and concurrent cisplatin and fluorouracil. The side effects seem to be greater when concurrent chemotherapy is included with comprehensive pelvic and para-aortic radiation, although combinations of extended-field radiation and weekly ­ cisplatin are usually tolerable.197

Neoadjuvant Chemotherapy Followed by Radiation Therapy Although cisplatin-containing chemotherapy produces some impressive responses in patients with cervical cancer, none of the seven randomized trials that have compared radiation therapy with or without neoadjuvant chemotherapy has demonstrated improved survival rates with the neoadjuvant chemotherapy approach (Table 29-9).173,187-192,296 To permit surgical treatment of patients with bulky lesions who would otherwise be poor candidates for radical hysterectomy, several trials have been initiated to study the role of induction chemotherapy before surgical exploration.166,297 In one of these trials, Sardi and colleagues166 found that stage IB tumors treated with chemotherapy were more often resectable and that the survival rate of patients who had received neoadjuvant chemotherapy was better than those who had not (81% versus 66%, P < .05). Most of the patients in that study also received postoperative radiation therapy. Benedetti-Panici and others297 reported similar outcomes for patients treated with radiation therapy alone or with neoadjuvant chemotherapy followed by radical hysterectomy; however, the radiation doses used in that trial (conducted before 1999) were low (median, 72 Gy to point A including LDR brachytherapy), the ­radiation treatments tended to be protracted, and concurrent chemotherapy was not given with the radiation.297 ­Ultimately, this approach may need to be compared with optimized chemoradiation therapy to determine the most effective, least toxic treatment.

Complications of Radiation Therapy Acute Side Effects External Beam Radiation Therapy Most patients experience at least some diarrhea during EBRT; symptoms may be more severe when concurrent chemotherapy (particularly fluorouracil) is given. Diarrhea

765

29  The Uterine Cervix

TABLE 29-9. Results of Prospective, Randomized Trials That Compared Neoadjuvant Chemotherapy Followed by Radiation Therapy with Radiation Therapy Alone for Patients with Locally Advanced Cervical Cancer Survival Rate (%)

Study and Reference

Year

Number of Patients

Stages

Drugs

CT + RT

RT Alone

P Value

End Point

Kumar et al187

1994

184

IIB-IVA

BIP

38

43

.5

OS at 32 mo

Tattersall et al192

1992

71

IIB-IVA

PVB

44*

40*

NS

OS at 48 mo

MtxCVP

47*

50*

NS

OS at 48 mo

69*

Chauvergne et Tattersall et

al186

al191

Sundfør et al190 Leborgne et

al188

Souhami et al189

1990

107

IIIB

1995

260

IIB-IVA

EpP

50*

.02

OS at 36 mo

1996

94

IIIB-IVA

PF

34*

37*

.9

OS at 48 mo

1997

96

IB2-IVA

BOP

38

45

.4

DFS at 60 mo

1991

107

IIIB

BOMP

23

39

.02

OS at 60 mo

*Percentages

were estimated from survival curves. B, bleomycin; C, chlorambucil; CT, chemotherapy; DFS, disease-free survival; Ep, epirubicin; F, fluorouracil; I, ifosfamide; M, mitomycin C; Mtx, methotrexate; NS, not significant; O, vincristine; OS, overall survival; P, cisplatin; RT, radiation therapy; V, vinblastine. From Eifel PJ. Chemoradiation for carcinoma of the cervix: advances and opportunities. Radiat Res 2000;154:229-236.

Intracavitary Radiation Therapy Fatal or life-threatening complications of ICRT are rare. Among 4042 patients who received 7662 LDR intracavitary treatments at M.  D. A   nderson between 1960 and 1992, there were 11 (0.3%) documented or suspected thromboembolic events resulting in four deaths (0.1%).268 All four of the patients who died had extensive pelvic wall disease that might have obstructed flow in pelvic vessels. Today, patients receive low-dose heparin during ICRT to reduce the risk of thromboembolism. However, because the number of events was so small before the institution of routine heparin prophylaxis, no significant risk reduction has been detectable. Less serious complications of ICRT included uterine perforation and vaginal laceration, which occurred in 2.8% and 0.3%, respectively, of patients in the M.  D. Anderson series. Both of these complications occurred more often in women who were more than 60 years old. Of the patients

10 Percent with major complications

is usually well controlled with oral medications and dietary modifications. Patients with severe diarrhea should have stool cultures done to rule out infection with Clostridium difficile. Less often, patients may experience bladder or urethral irritation. These symptoms can be treated with phenazopyridine hydrochloride (Pyridium) or antispasmodics after urinalysis and urine culture have ruled out a urinary tract complication. Patients with fever or flank pain should be evaluated for possible pyelonephritis. This is most likely to occur in patients with large tumors that have compromised urinary flow. When radiation fields cover the vulva to treat extensive vaginal involvement, severe skin erythema and even moist desquamation can develop. Treating the patient in an open-leg position can usually drastically reduce radiationinduced skin reactions. Patients who have unexpectedly severe skin reactions often have an underlying infection with Candida organisms, and treatment with oral or local antifungal agents can greatly improve symptoms even while radiation treatments continue.

9 Urinary tract

8 7 6 5

Rectal

4 3 2 1 0 0

5

10

15

20

25

Time (years)

FIGURE 29-21.  Rates of major rectal and urinary tract complications in 1784 patients treated with radiation therapy for FIGO stage IB carcinoma of the uterine cervix. (From Eifel PJ, Levenback C, Wharton JT, et al. Time course and incidence of late complications in patients treated with radiation therapy for FIGO stage IB carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys 1995;32:1289-1300.)

in that series, 14% had an elevated temperature, usually of uncertain origin.

Late Complications Most late complications of radiation therapy involve the rectum, bladder, or small bowel. Although most serious gastrointestinal complications occur within the first 3 years, serious side effects also may occur several decades after treatment (Fig. 29-21). The average time to onset of major urinary tract complications tends to be somewhat longer than that of intestinal complications. The most common serious late complications involve bleeding from the bladder or rectum. In a series of 1784 patients treated for stage IB disease at M.  D. Anderson,5 the actuarial risks of hematuria or hematochezia severe enough to require transfusion

766

PART 9  Female Genital Tract

were 2.6% and 0.7%, respectively, at 5 years. The overall risk of developing a gastrointestinal or urinary tract fistula was 1.7% at 5 years, with an increased risk for patients who underwent adjuvant hysterectomy or pretreatment transperitoneal lymphadenectomy. The risk of small bowel obstruction correlates strongly with several patient characteristics and treatment factors. Several studies have demonstrated threefold to fourfold increases in risk for patients who undergo transperitoneal lymphadenectomy before radiation therapy.5,60,298,299 Patients given high doses (>50 Gy) of EBRT delivered to the central pelvis have a greater risk of all types of complications, including small bowel obstruction.71,298 The risk of small bowel obstruction is also greater for patients who are very thin, have a history of pelvic infection, or smoke. A study from M.  D. Anderson revealed a sixfold increase in the risk of small bowel complications for patients who smoked more than one pack of cigarettes per day (12% versus 2% for nonsmokers).298 Mild to moderate apical vaginal ulceration or necrosis occurs in 5% to 10% of patients treated with primary radiation therapy for cervical cancer.5 This complication usually occurs within 12 months after the completion of radiation therapy, rarely progresses to fistula formation, and usually heals within 1 to 6 months with appropriate local care, including use of vaginal estrogen cream or 50% hydrogen peroxide douches (depending on severity). The true incidence of vaginal shortening and its influence on sexual dysfunction is poorly documented. Severe shortening is more common among postmenopausal patients, patients who are less sexually active, and patients with more advanced disease at presentation.

REFERENCES    1. Horning SJ, Hoppe RT, Kaplan HS, et al. Female reproductive potential after treatment for Hodgkin’s disease. N Engl J Med 1981;304:1377-1382.    2. Morice P, Thiam-Ba R, Castaigne D, et al. Fertility results after ovarian transposition for pelvic malignancies treated by external irradiation or brachytherapy. Hum Reprod 1998;13:660-663.    3. Mitchell MF, Gershenson DM, Soeters RP, et al. The long-term effects of radiation therapy on patients with ovarian dysgerminoma. Cancer 1991;67:1084-1090.    4. Fajardo LF. Pathology of Radiation Injury. New York: Masson, 1982.    5. Eifel PJ, Levenback C, Wharton JT, et al. Time course and incidence of late complications in patients treated with radiation therapy for FIGO stage IB carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys 1995;32:1289-1300.    6. American Cancer Society. Cancer Facts & Figures, 2008. Atlanta: American Cancer Society, 2008.    7. Wingo PA, Tong T, Bolden S. Cancer statistics, 1995. CA Cancer J Clin 1995;45:8-30.    8. American Cancer Society. Cancer Facts & Figures for Hispanics/ Latinos 2003-2005. Atlanta: American Cancer Society, 2006.    9. Vizcaino AP, Moreno V, Bosch FX, et al. International trends in the incidence of cervical cancer. I. Adenocarcinoma and adeno­ squamous cell carcinomas. Int J Cancer 1998;75:536-545.   10. Zheng T, Holford TR, Ma Z, et al. The continuing increase in adenocarcinoma of the uterine cervix: a birth cohort phenomenon. Int J Epidemiol 1996;25:252-258.   11. Bosch FX, Castellsague X, Munoz N, et al. Male sexual behavior and human papillomavirus DNA: key risk factors for cervical cancer in Spain. J Natl Cancer Inst 1996;88:1060-1067.   12. Bosch FX, de Sanjose S. Human papillomavirus and cervical cancer—burden and assessment of causality. J Natl Cancer Inst Monogr 2003;31:3-13.

  13. Kessler I. Epidemiological aspects of uterine cervix cancer. In Lurain JR, Sciarra J, eds: Gynecology and Obstetrics. Philadelphia: JB Lippincott, 1990, p 1.   14. Scheffner M, Werness BA, Huibregtse JM, et al. The E6 oncoprotein encoded by human papillomavirus types 16 and 18 promotes the degradation of p53. Cell 1990;63:1129-1136.   15. Munoz N, Bosch FX, de Sanjose S, et al. Epidemiologic classification of human papillomavirus types associated with cervical cancer. N Engl J Med 2003;348:518-527.   16. Koutsky LA, Ault KA, Wheeler CM, et al. A controlled trial of a human papillomavirus type 16 vaccine. N Engl J Med 2002;347:1645-1651.   17. Roden R, Wu TC. How will HPV vaccines affect cervical cancer? Nat Rev Cancer 2006;6:753-763.   18. Ho GY, Kadish AS, Burk RD, et al. HPV 16 and cigarette smoking as risk factors for high-grade cervical intra-epithelial neoplasia. Int J Cancer 1998;78:281-285.   19. Roteli-Martins CM, Panetta K, Alves VA, et al. Cigarette smoking and high-risk HPV DNA as predisposing factors for high-grade cervical intraepithelial neoplasia (CIN) in young Brazilian women. Acta Obstet Gynecol Scand 1998;77: 678-682.   20. Schiffman MH. Recent progress in defining the epidemiology of human papillomavirus infection and cervical neoplasia. J Natl Cancer Inst 1992;84:394-398.   21. Prokopczyk B, Cox JE, Hoffmann D, et al. Identification of tobacco-specific carcinogen in the cervical mucus of smokers and nonsmokers. J Natl Cancer Inst 1997;89:868-873.   22. Ozsaran AA, Ates T, Dikmen Y, et al. Evaluation of the risk of cervical intraepithelial neoplasia and human papilloma virus infection in renal transplant patients receiving immunosuppressive therapy. Eur J Gynaecol Oncol 1999;20:127-130.   23. Chirenje ZM. HIV and cancer of the cervix. Best Pract Res Clin Obstet Gynaecol 2005;19:269-276.   24. Ellerbrock TV, Chiasson MA, Bush TJ, et al. Incidence of cervical squamous intraepithelial lesions in HIV-infected women. JAMA 2000;283:1031-1037.   25. Palefsky JM, Minkoff H, Kalish LA, et al. Cervicovaginal human papillomavirus infection in human immunodeficiency virus-1 (HIV)-positive and high-risk HIV-negative women. J Natl Cancer Inst 1999;91:226-236.   26. Beral V, Newton R. Overview of the epidemiology of immunodeficiency-associated cancers. J Natl Cancer Inst Monogr 1998;23:1-6.   27. Massad LS, Seaberg EC, Watts DH, et al. Low incidence of invasive cervical cancer among HIV-infected US women in a prevention program. AIDS 2004;18:109-113.   28. Fruchter RG, Maiman M, Arrastia CD, et al. Is HIV infection a risk factor for advanced cervical cancer? J Acquir Immune Defic Syndr Hum Retrovirol 1998;18:241-245.   29. Maiman M, Fruchter RG, Clark M, et al. Cervical cancer as an AIDS-defining illness. Obstet Gynecol 1997;89:76-80.   30. Stubblefield PG. Oral contraceptives and neoplasia. J Reprod Med 1984;29:524-529.   31. Thomas DB, Ray RM. Oral contraceptives and invasive adenocarcinomas and adenosquamous carcinomas of the uterine ­cervix. The World Health Organization Collaborative Study of Neoplasia and Steroid Contraceptives. Am J Epidemiol 1996;144: 281-289.   32. Plentl AA, Friedman EA. Lymphatics of the cervix uteri.In: Lymphatic System of the Female Genitalia: The Morphologic Basis of Oncologic Diagnosis and Therapy. Philadelphia: WB Saunders, 1971, pp 75-115.   33. Papanicolaou G, Traut RF. The Diagnosis of Uterine Cancer by the Vaginal Smear. New York: Commonwealth Fund, 1943.   34. Kivlahan C, Ingram E. Papanicolaou smears without endocervical cells. Are they inadequate? Acta Cytol 1986;30:258-260.   35. Walton RJ. Editorial: the task force on cervical cancer screening programs. Can Med Assoc J 1976;114:981.   36. Syrjanen KJ. Spontaneous evolution of intraepithelial lesions ­according to the grade and type of the implicated human papillomavirus (HPV). Eur J Obstet Gynecol Reprod Biol 1996;65: 45-53.

29  The Uterine Cervix   37. Richart RM, Barron BA. A follow-up study of patients with cervical dysplasia. Am J Obstet Gynecol 1969;105:386-393.   38. Manos MM, Kinney WK, Hurley LB, et al. Identifying women with cervical neoplasia: using human papillomavirus DNA testing for equivocal Papanicolaou results. JAMA 1999;281: 1605-1610.   39. Kinney WK, Alvarez RD, Reid GC, et al. Value of adjuvant whole-pelvis irradiation after Wertheim hysterectomy for ­earlystage squamous carcinoma of the cervix with pelvic nodal metastasis: a matched-control study. Gynecol Oncol 1989;34: 258-262.   40. Cox JT. Human papillomavirus testing in primary cervical screening and abnormal Papanicolaou management. Obstet Gynecol Surv 2006;61:S15-25.   41. Hopkins MP, Roberts JA, Schmidt RW. Cervical adenocarcinoma in situ. Obstet Gynecol 1988;71:842-844.   42. Azodi M, Chambers SK, Rutherford TJ, et al. Adenocarcinoma in situ of the cervix: management and outcome. Gynecol Oncol 1999;73:348-353.   43. Crowther ME, Lowe DG, Shepherd JH. Verrucous carcinoma of the female genital tract: a review. Obstet Gynecol Surv 1998;43:263-280.   44. Randall ME, Anderson WA, Mills SE, et al. Papillary squamous cell carcinoma of the uterine cervix: a clinicopathologic study of nine cases. Int J Gynecol Pathol 1986;5:1-10.   45. Eifel PJ, Burke TW, Delclos L, et al. Early stage I adenocarcinoma of the uterine cervix: treatment results in patients with tumors = 4 cm in diameter. Gynecol Oncol 1991;41:199-205.   46. Raju K, Kjorstad KE, Abeler V. Prognostic factors in the treatment of stage IB adenocarcinoma of the cervix. Int J Gynaecol Obstet 1991;1:69-74.   47. Hirai Y, Takeshima N, Haga A, et al. A clinicocytopathologic study of adenoma malignum of the uterine cervix. Gynecol Oncol 1998;70:219-223.   48. Albores-Saavedra J, Gersell D, Gilks CB, et al. Terminology of endocrine tumors of the uterine cervix: results of a workshop sponsored by the College of American Pathologists and the National Cancer Institute. Arch Pathol Lab Med 1997;121:34-39.   49. Sato Y, Shimamoto T, Amada S, et al. Large cell neuroendocrine carcinoma of the uterine cervix: a clinicopathological study of six cases. Int J Gynecol Pathol 2003;22:226-230.   50. Viswanathan AN, Deavers MT, Jhingran A, et al. Small cell neuroendocrine carcinoma of the cervix: outcome and patterns of recurrence. Gynecol Oncol 2004;93:27-33.   51. Greene FL, Page DL, Fleming ID, et al (eds). Cervix uteri. In: AJCC Cancer Staging Manual. 6th ed, New York: Springer, 2002, pp 259-265.   52. International Federation of Gynecology and Obstetrics. Staging announcement. FIGO staging of gynecologic cancers; cervical and vulva. Int J Gynecol Cancer 1995;5:319.   53. League of Nations Health Organization. Inquiry into the results of radiotherapy in cancer of the uterus. Atlas illustrating the division of cancer of the uterine cervix into four stages according to the anatomo-clinical extent of the growth. Stockholm: Kungl Boktryckeriet PA Norstedt & Söner, 1938.   54. Burghardt E, Hofmann HMH, Ebner F, et al. Results of surgical treatment of 1028 cervical cancers studied with volumetry. ­Cancer 1992;70:648-655.   55. Eifel PJ, Moughan J, Erickson B, et al. Patterns of radiotherapy practice for patients with squamous carcinoma of the uterine cervix. A Patterns of Care study. Int J Radiat Oncol Biol Phys 2004;60:1144-1153.   56. Rose PG, Adler LP, Rodriguez M, et al. Positron emission tomography for evaluating para-aortic nodal metastasis in locally advanced cervical cancer before surgical staging: a surgicopathologic study. J Clin Oncol 1999;17:41-45.   57. Grigsby PW, Siegel BA, Dehdashti F. Lymph node staging by positron emission tomography in patients with carcinoma of the cervix. J Clin Oncol 2001;19:3745-3749.   58. Cohn DE, Dehdashti F, Gibb RK, et al. Prospective evaluation of positron emission tomography for the detection of groin node metastases from vulvar cancer. Gynecol Oncol 2002;85:179-184.

767

  59. Grigsby PW, Siegel BA, Dehdashti F, et al. Posttherapy [18F] fluorodeoxyglucose positron emission tomography in carcinoma of the cervix: response and outcome. J Clin Oncol 2004;22: 2167-2171.   60. Weiser EB, Bundy BN, Hoskins WJ, et al. Extraperitoneal versus transperitoneal selective paraaortic lymphadenectomy in the pretreatment surgical staging of advanced cervical carcinoma (a Gynecologic Oncology Group study). Gynecol Oncol 1981;33:283-289.   61. Possover M, Krause N, Plaul K, et al. Laparoscopic para-aortic and pelvic lymphadenectomy: experience with 150 patients and review of the literature. Gynecol Oncol 1998;71:19-28.   62. Hacker NF, Wain GV, Nicklin JL. Resection of bulky positive lymph nodes in patients with cervical carcinoma. Int J Gynecol Cancer 1995;5:250-256.   63. Potish RA, Downey GO, Adcock LL, et al. The role of surgical debulking in cancer of the uterine cervix. Int J Radiat Oncol Biol Phys 1989;17:979-984.   64. Coleman RL, Burke TW, Morris M, et al. Intra-operative radiographs to confirm the adequacy of lymph node resection in patients with suspicious lymphangiograms. Gynecol Oncol 1994;51:362-367.   65. Di Stefano AB, Acquaviva G, Garozzo G, et al. Lymph node mapping and sentinel node detection in patients with cervical carcinoma: a 2-year experience. Gynecol Oncol 2005;99:671-679.   66. Alvarez RD, Potter ME, Soong SJ, et al. Rationale for using pathologic tumor dimensions and nodal status to subclassify surgically treated stage IB cervical cancer patients. Gynecol Oncol 1991;43:108-112.   67. Eifel PJ, Morris M, Wharton JT, et al. The influence of tumor size and morphology on the outcome of patients with FIGO stage IB squamous cell carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys 1994;29:9-16.   68. Fyles AW, Pintilie M, Kirkbride P, et al. Prognostic factors in patients with cervix cancer treated by radiation therapy: results of a multiple regression analysis. Radiother Oncol 1995;35:107-117.   69. Perez CA, Grigsby PW, Nene SM, et al. Effect of tumor size on the prognosis of carcinoma of the uterine cervix treated with irradiation alone. Cancer 1992;69:2796-2806.   70. Horiot JC, Pigneux J, Pourquier H, et al. Radiotherapy alone in carcinoma of the intact uterine cervix according to G.H. Fletcher guidelines: a French cooperative study of 1383 cases. Int J Radiat Oncol Biol Phys 1988;14:605-611.   71. Logsdon MD, Eifel PJ. FIGO stage IIIB squamous cell carcinoma of the uterine cervix: an analysis of prognostic factors emphasizing the balance between external beam and intracavitary radiation therapy. Int J Radiat Oncol Biol Phys 1999;43:763-775.   72. Sevin BU, Lu Y, Bloch DA, et al. Surgically defined prognostic parameters in patients with early cervical carcinoma. A multivariate survival tree analysis. Cancer 1996;78:1438-1446.   73. Hawnaur JM, Johnson RJ, Buckley CH, et al. Staging, volume estimation and assessment of nodal status in carcinoma of the cervix: comparison of magnetic resonance imaging with surgical findings. Clin Radiol 1994;49:443-452.   74. Mayr NA, Yuh WT, Zheng J, et al. Tumor size evaluated by pelvic examination compared with 3-D quantitative analysis in the prediction of outcome for cervical cancer. Int J Radiat Oncol Biol Phys 1997;39:395-404.   75. Toita T, Kakinohana Y, Shinzato S, et al. Tumor diameter/volume and pelvic node status assessed by magnetic resonance imaging (MRI) for uterine cervical cancer treated with irradiation. Int J Radiat Oncol Biol Phys 1999;43:777-782.   76. Delgado G, Bundy BN, Fowler WC, et al. A prospective surgical pathological study of stage I squamous carcinoma of the cervix: a Gynecologic Oncology Group study. Gynecol Oncol 1989;36:314-320.   77. Zreik TG, Chambers JT, Chambers SK. Parametrial involvement, regardless of nodal status: a poor prognostic factor for cervical cancer. Obstet Gynecol 1996;87:741-746.   78. Landoni F, Bocciolone L, Perego P, et al. Cancer of the cervix, FIGO stages IB and IIA: patterns of local growth and paracervical extension. Int J Gynecol Cancer 1995;5:329-334.

768

PART 9  Female Genital Tract

  79. Stehman F, Perez C, Kurman R, et al. Uterine cervix. In Hoskins W, Perez C, Young R (eds): Principles and Practice of Gynecologic Oncology. Philadelphia: Lippincott, 2000, pp 591-662.   80. Thoms WW, Eifel PJ, Smith TL, et al. Bulky endocervical carcinomas: a 23-year experience. Int J Radiat Oncol Biol Phys 1992;23:491-499.   81. Noguchi H, Shiozawa I, Kitahara T, et al. Uterine body invasion of carcinoma of the uterine cervix as seen from surgical specimens. Gynecol Oncol 1988;30:173-182.   82. Perez CA, Camel HM, Askin F, et al. Endometrial extension of carcinoma of the uterine cervix: a prognostic factor that may modify staging. Cancer 1981;48:170-180.   83. Alvarez RD, Gelder MS, Gore H, et al. Radical hysterectomy in the treatment of patients with bulky early stage carcinoma of the cervix uteri. Surg Gynecol Obstet 1993;176:539-542.   84. Delgado G, Bundy B, Zaino R, et al. Prospective surgical-pathological study of disease-free interval in patients with stage IB squamous cell carcinoma of the cervix: a Gynecologic Oncology Group study. Gynecol Oncol 1990;38:352-357.   85. Peters WA III, Liu PY, Barrett RJ Jr, et al. Concurrent chemo­ therapy and pelvic radiation therapy compared with pelvic radiation therapy alone as adjuvant therapy after radical surgery in high-risk early-stage cancer of the cervix. J Clin Oncol 2000;18:1606-1613.   86. Dawtlatly R, Lavie O, Cross PA, et al. Prognostic factors in surgically-treated stage IB-IIB squamous cell carcinoma of the cervix with positive lymph nodes. Int J Gynecol Cancer 1998;8:46-47.   87. Inoue T, Chihara T, Morita K. The prognostic significance of the size of the largest nodes in metastatic carcinoma from the uterine cervix. Gynecol Oncol 1984;19:187-193.   88. Inoue T, Morita K. The prognostic significance of number of positive nodes in cervical carcinoma stages IB, IIA, and IIB. Cancer 1990;65:1923-1927.   89. Ballon SC, Berman ML, Lagasse LD, et al. Survival after extraperitoneal pelvic and paraaortic lymphadenectomy and radiation therapy in cervical carcinoma. Obstet Gynecol 1981;57:90-95.   90. Berman ML, Keys H, Creasman W, et al. Survival and patterns of recurrence in cervical cancer metastatic to periaortic lymph nodes (a Gynecologic Oncology Group study). Gynecol Oncol 1984;19:8-16.   91. Brookland RK, Rubin S, Danoff BF. Extended field irradiation in the treatment of patients with cervical carcinoma involving biopsy proven para-aortic nodes. Int J Radiat Oncol Biol Phys 1984;10:1875-1879.   92. Buchsbaum H. Extrapelvic lymph node metastases in cervical carcinoma. Am J Obstet Gynecol 1979;133:814-824.   93. Cunningham M, Dunton C, Corn B, et al. Extended-field radiation therapy in early-stage cervical carcinoma: survival and complications. Gynecol Oncol 1991;43:51-54.   94. Kim PY, Monk BJ, Chabra S, et al. Cervical cancer with paraaortic metastases: significance of residual paraaortic disease after surgical staging. Gynecol Oncol 1998;69:243-247.   95. Komaki R, Mattingly RF, Hoffman RG, et al. Irradiation of paraaortic lymph node metastases from carcinoma of the cervix or endometrium: preliminary results. Radiology 1983;147:245-248.   96. Nori D, Valentine E, Hilaris BS. The role of paraaortic node irradiation in the treatment of cancer of the cervix. Int J Radiat Oncol Biol Phys 1985;11:1469-1473.   97. Piver MS, Barlow JJ, Krishnamsetty R. Five-year survival (with no evidence of disease) in patients with biopsy-confirmed aortic node metastases from cervical carcinoma. Am J Obstet Gynecol 1981;139:575-578.   98. Podczaski E, Stryker JA, Kaminski P, et al. Extended-field radiation therapy for carcinoma of the cervix. Cancer 1990;66: 251-258.   99. Rubin SC, Brookland R, Mikuta JJ, et al. Para-aortic nodal metastases in early cervical carcinoma: long-term survival following extended-field radiotherapy. Gynecol Oncol 1984;18:213-217. 100. Tewfik HH, Buchsbaum HJ, Latourette HB, et al. Para-aortic lymph node irradiation in carcinoma of the cervix after exploratory laparotomy and biopsy-proven positive aortic nodes. Int J Radiat Oncol Biol Phys 1982;8:13-18.

101. Wharton JT, Jones HWI, Day T, et al. Preirradiation celiotomy and extended field irradiation for invasive carcinoma of the cervix. Obstet Gynecol 1977;49:333-338. 102. Comerci B, Bolger BS, Flannelly G, et al. Prognostic factors in surgically treated stage IB-IIB carcinoma of the cervix with negative lymph nodes. Int J Gynecol Cancer 1998;8:23-26. 103. Kamura T, Tsukamoto N, Tsuruchi N, et al. Multivariate analysis of the histopathologic prognostic factors of cervical cancer in patients undergoing radical hysterectomy. Cancer 1992;69:181-186. 104. Kristensen GB, Abeler VM, Risberg B, et al. Tumor size, depth of invasion, and grading of the invasive tumor front are the main prognostic factors in early squamous cell cervical carcinoma. Gynecol Oncol 1999;74:245-251. 105. Roman LD, Felix JC, Muderspach LI, et al. Influence of quantity of lymph-vascular space invasion on the risk of nodal metastases in women with early-stage squamous cancer of the cervix. Gynecol Oncol 1998;68:220-225. 106. Saigo PE, Cain JM, Kim WS, et al. Prognostic factors in adenocarcinoma of the uterine cervix. Cancer 1986;57:1584-1593. 107. Pickel H, Haas J, Lahousen M. Prognostic factors in cervical cancer. Eur J Obstet Gynecol 1997;71:209-213. 108. Grigsby PW, Perez CA, Kuske RR, et al. Adenocarcinoma of the uterine cervix: lack of evidence for a poor prognosis. Radiother Oncol 1988;12:289-296. 109. Kilgore LC, Soong SJ, Gore H, et al. Analysis of prognostic features in adenocarcinoma of the cervix. Gynecol Oncol 1988;31:137-148. 110. Kjörstad KE, Bond B. Stage IB adenocarcinoma of the cervix: metastatic potential and patterns of dissemination. Am J Obstet Gynecol 1984;150. 297–279. 111. Waldenstrom AC, Horvath G. Survival of patients with adenocarcinoma of the uterine cervix in western Sweden. Int J Gynecol Cancer 1999;9:18-23. 112. Eifel PJ, Burke TW, Morris M, et al. Adenocarcinoma as an independent risk factor for disease recurrence in patients with stage IB cervical carcinoma. Gynecol Oncol 1995;59:38-44. 113. Kleine W, Rau K, Schwoeorer D, et al. Prognosis of adenocarcinoma of the cervix uteri: a comparative study. Gynecol Oncol 1989;35:145-149. 114. Rotman M, Sedlis A, Piedmonte MR, et al. A phase III randomized trial of postoperative pelvic irradiation in stage IB cervical carcinoma with poor prognostic features: follow-up of a Gynecologic Oncology Group study. Int J Radiat Oncol Biol Phys 2006;65:169-176. 115. Landoni F, Maneo A, Colombo A, et al. Randomised study of radical surgery versus radiotherapy for stage Ib-IIa cervical cancer. Lancet 1997;350:535-540. 116. Berek JS, Hacker NS, Fu YS, et al. Adenocarcinoma of the uterine cervix: histologic variables associated with lymph node metastasis and survival. Obstet Gynecol 1985;65:46-52. 117. Matthews CM, Burke TW, Tornos C, et al. Stage I cervical adenocarcinoma: prognostic evaluation of surgically treated patients. Gynecol Oncol 1993;49:19-23. 118. Wang KL, Yang YC, Wang TY, et al. Neuroendocrine carcinoma of the uterine cervix: a clinicopathologic retrospective study of 31 cases with prognostic implications. J Chemother 2006;18:209-216. 119. Maiman M. The clinical application of serum squamous cell carcinoma antigen level monitoring in invasive cervical carcinoma. Gynecol Oncol 2002;84:4-6. 120. Micke O, Bruns F, Schafer U, et al. The impact of squamous cell carcinoma (SCC) antigen in patients with advanced cancer of uterine cervix treated with (chemo-)radiotherapy. Anticancer Res 2005;25:1663-1666. 121. Hong JH, Tsai CS, Lai CH, et al. Risk stratification of patients with advanced squamous cell carcinoma of cervix treated by radiotherapy alone. Int J Radiat Oncol Biol Phys 2005;63:492499. 122. Lin H, ChangChien CC, Huang EY, et al. The role of pretreatment squamous cell carcinoma antigen in predicting nodal metastasis in early stage cervical cancer. Acta Obstet Gynecol Scand 2000;79:140-144.

29  The Uterine Cervix 123. Kainz C, Gitsch G, Tempfer C, et al. Vascular space invasion and inflammatory stromal reaction as prognostic factors in patients with surgically treated cervical cancer stage IB to IIB. Anticancer Res 1994;14:2245-2248. 124. Höckel M, Schlenger K, Mitze M, et al. Tumor vascularity—a novel prognostic factor in advanced cancer of the uterine cervix. Proc Soc Gynecol Oncol 1995:48. 125. Estape R, Angioli R, Wagman F, et al. Significance of intraperitoneal cytology in patients undergoing radical hysterectomy. Gynecol Oncol 1998;68:169-171. 126. Takeshima N, Katase K, Hirai Y, et al. Prognostic value of peritoneal cytology in patients with carcinoma of the uterine cervix. Gynecol Oncol 1997;64:136-140. 127. Lombard I, Vincent-Salomon A, Validire P, et al. Human papillomavirus genotype as a major determinant of the course of cervical cancer. J Clin Oncol 1998;16:2613-2619. 128. Ikenberg H, Wiegering I, Pfisterer J, et al. Human papillomavirus DNA in tumor-free regional lymph nodes: a potential prognostic marker in cervical cancer. Cancer J Sci Am 1996;2:28-34. 129. Kobayashi Y, Yoshinouchi M, Tianqi G, et al. Presence of human papilloma virus DNA in pelvic lymph nodes can predict unexpected recurrence of cervical cancer in patients with histologically negative lymph nodes. Clin Cancer Res 1998;4:979-983. 130. Park JS, Park DC, Kim CJ, et al. HPV-16-related proteins as the serologic markers in cervical neoplasia. Gynecol Oncol 1998;69:47-55. 131. Alvarez RD, Soong SJ, Kinney WK, et al. Identification of prognostic factors and risk groups in patients found to have nodal metastasis at the time of radical hysterectomy for early-stage squamous carcinoma of the cervix. Gynecol Oncol 1989;35: 130-135. 132. Sigurdsson K, Hrafnkelsson J, Geirsson G, et al. Screening as a prognostic factor in cervical cancer: analysis of survival and prognostic factors based on Icelandic population data, 1964-1988. Gynecol Oncol 1991;43:64-70. 133. Dattoli MJ, Gretz HF III, Beller U, et al. Analysis of multiple prognostic factors in patients with stage IB cervical cancer: age as a major determinant. Int J Radiat Oncol Biol Phys 1989;17: 41-47. 134. Rutledge FN, Mitchell MR, Munsell M, et al. Youth as a prognostic factor in carcinoma of the cervix: a matched analysis. Gynecol Oncol 1992;44:123-130. 135. Hopkins MP, Morley GW. Stage IB squamous cell cancer of the cervix: clinicopathologic features related to survival. Am J Obstet Gynecol 1991;164:1520-1527. 136. Mitchell PA, Waggoner S, Rotmensch J, et al. Cervical cancer in the elderly treated with radiation therapy. Gynecol Oncol 1998;71:291-298. 137. Spanos WJ, King A, Keeney E, et al. Age as a prognostic factor in carcinoma of the cervix. Gynecol Oncol 1989;35:66-68. 138. Bush R. The significance of anemia in clinical radiation therapy. Int J Radiat Oncol Biol Phys 1986;12:2047-2050. 139. Girinski T, Pejovic-Lenfant M, Bourhis J, et al. Prognostic value of hemoglobin concentrations and blood transfusions in advanced carcinoma of the cervix treated by radiation therapy: results of a retrospective study of 386 patients. Int J Radiat Oncol Biol Phys 1989;16:37-42. 140. Kapp DS, Fischer D, Gutierrez E, et al. Pretreatment prognostic factors in carcinoma of the uterine cervix: a multivariable analysis of the effect of age, stage, histology and blood counts on survival. Int J Radiat Oncol Biol Phys 1983;9:445-455. 141. Thomas G. The effect of hemoglobin level on radiotherapy outcomes: the Canadian experience. Semin Oncol 2001;28: 60-65. 142. Brady LW, Plenk HP, Hanley JA, et al. Hyperbaric oxygen therapy for carcinoma of the cervix stages IIB, IIIA, IIIB and IVA: results of a randomized study by the Radiation Therapy Oncology Group. Int J Radiat Oncol Biol Phys 1981;7:990-998. 143. Dische S, Anderson PJ, Sealy R, et al. Carcinoma of the cervix—anaemia, radiotherapy and hyperbaric oxygen. Br J Radiol 1983;56:251-255.

769

144. Dische S, Chassagne D, Hope-Stone HF, et al. A trial of Ro 03-8799 (pimonidazole) in carcinoma of the uterine cervix: an interim report from the Medical Research Council Working Party on Advanced Carcinoma of the Cervix. Radiother Oncol 1993;26:93-103. 145. Grigsby PW, Winter K, Wasserman TH, et al. Irradiation with or without misonidazole for patients with stages IIIB and IVA carcinoma of the cervix: final results of RTOG 80-05. Radiation Therapy Oncology Group. Int J Radiat Oncol Biol Phys 1999;44:513-517. 146. Maor MH, Gillespie BW, Peters LJ, et al. Neutron therapy in cervical cancer: results of a phase III RTOG study. Int J Radiat Oncol Biol Phys 1988;14:885-891. 147. Overgaard J, Bentzen SM, Kolstad P, et al. Misonidazole combined with radiotherapy in the treatment of carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys 1989;16: 1069-1072. 148. Sundfør K, Lyng H, Rofstad EK. Tumour hypoxia and vascular density as predictors of metastasis in squamous cell carcinoma of the uterine cervix. Br J Cancer 1998;78:822-827. 149. Fyles AW, Milosevic M, Wong R, et al. Oxygenation predicts radiation response and survival in patients with cervix cancer [published erratum appears in Radiother Oncol 1999;50:371]. Radiother Oncol 1998;48:149-156. 150. Höckel M, Vorndran B, Schlenger K, et al. Tumor oxygenation: a new predictive parameter in locally advanced cancer of the uterine cervix. Gynecol Oncol 1993;51:141-149. 151. Hernandez E, Heller PB, Whitney C, et al. Thrombocytosis in surgically treated stage IB squamous cell cervical carcinoma (a Gynecologic Oncology Group study). Gynecol Oncol 1994;55:328-332. 152. Chen F, Trapido EJ, Davis K. Differences in stage at presentation of breast and gynecologic cancers among whites, blacks, and Hispanics. Cancer 1994;73:2838-2842. 153. Mandelblatt J, Andrews H, Kerner J, et al. Determinants of late stage diagnosis of breast and cervical cancer: the impact of age, race, social class, and hospital type. Am J Public Health 1991;81:646-649. 154. Shelton D, Paturzo D, Flannery J, et al. Race, stage of disease, and survival with cervical cancer. Ethn Dis 1992;2:47-54. 155. Mundt AJ, Connell PP, Campbell T, et al. Race and clinical outcome in patients with carcinoma of the uterine cervix treated with radiation therapy. Gynecol Oncol 1998;71:151-158. 156. Katz A, Eifel PJ, Moughan J, et al. Socioeconomic characteristics of patients with squamous cell carcinoma of the uterine cervix treated with radiotherapy in the 1992 to 1994 Patterns of Care study. Int J Radiat Oncol Biol Phys 2000;47:443-450. 157. Kucera H, Enzelberger H, Eppel W, et al. The influence of nicotine abuse and diabetes mellitus on the results of primary irradiation in the treatment of carcinoma of the cervix. Cancer 1987;60:1-4. 158. Grigsby P, Kuske R, Perez C, et al. Medically inoperable stage I adenocarcinoma of the endometrium treated with radiotherapy alone. Int J Radiat Oncol Biol Phys 1987;13:483-488. 159. Hamberger AD, Fletcher GH, Wharton JT. Results of treatment of early stage I carcinoma of the uterine cervix with intracavitary radium alone. Cancer 1978;41:980-985. 160. Burghardt E, Girardi F, Lahousen M, et al. Microinvasive carcinoma of the uterine cervix (International Federation of Gynecology and Obstetrics stage IA). Cancer 1991;67:1037-1045. 161. Morris M, Mitchell MF, Silva EG, et al. Cervical conization as definitive therapy for early invasive squamous carcinoma of the cervix. Gynecol Oncol 1993;51:193-196. 162. Dargent D. Radical abdominal trachelectomy and pelvic lymphadenectomy with uterine conservation and subsequent pregnancy in the treatment of early invasive cervical cancer. Am J Obstet Gynecol 2002;187:1728. author reply 1729. 163. Plante M, Renaud MC, Francois H, 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;94:614-623.

770

PART 9  Female Genital Tract

164. Burnett AF, Roman LD, O’Meara AT, et al. Radical vaginal trachelectomy and pelvic lymphadenectomy for preservation of fertility in early cervical carcinoma. Gynecol Oncol 2003;88:419-423. 165. Sedlis A, Bundy BN, Rotman MZ, et al. A randomized trial of pelvic radiation therapy versus no further therapy in selected patients with stage IB carcinoma of the cervix after radical hysterectomy and pelvic lymphadenectomy: a Gynecologic Oncology Group Study. Gynecol Oncol 1999;73:177-183. 166. Sardi JE, Giaroli A, Sananes C, et al. Long-term follow-up of the first randomized trial using neoadjuvant chemotherapy in stage IB squamous carcinoma of the cervix: the final results. Gynecol Oncol 1997;67:61-69. 167. Lowrey GC, Mendenhall WM, Million RR. Stage IB or IIA-B carcinoma of the intact uterine cervix treated with irradiation: a multivariate analysis. Int J Radiat Oncol Biol Phys 1992;24:205-210. 168. Durrance FY, Fletcher GH, Rutledge FN. Analysis of central recurrent disease in stages I and II squamous cell carcinomas of the cervix on intact uterus. AJR Am J Roentgenol 1969;106: 831-838. 169. Mendenhall WM, McCarty PJ, Morgan LS, et al. Stage IB-IIA-B carcinoma of the intact uterine cervix greater than or equal to 6 cm in diameter: is adjuvant extrafascial hysterectomy beneficial? Int J Radiat Oncol Biol Phys 1991;21:899-904. 170. Keys HM, Roberts JA, Brunetto VL, et al. A phase III trial of surgery with or without adjunctive external pelvic radiation therapy in intermediate risk endometrial adenocarcinoma: a Gynecologic Oncology Group study. Gynecol Oncol 2004;92:744-751. 171. Eifel PJ, Winter K, Morris M, et al. Pelvic irradiation with concurrent chemotherapy versus pelvic and para-aortic irradiation for high-risk cervical cancer: an update of Radiation Therapy Oncology Group trial (RTOG) 90-01. J Clin Oncol 2004;22: 872-880. 172. Keys HM, Bundy BN, Stehman FB, et al. Cisplatin, radiation, and adjuvant hysterectomy for bulky stage IB cervical carcinoma. N Engl J Med 1999;340:1154-1161. 173. Rose PG, Bundy BN, Watkins J, et al. Concurrent cisplatin-based chemotherapy and radiotherapy for locally advanced cervical cancer. N Engl J Med 1999;340:1144-1153. 174. Whitney CW, Sause W, Bundy BN, et al. A randomized comparison of fluorouracil plus cisplatin versus hydroxyurea as an adjunct to radiation therapy in stages IIB-IVA carcinoma of the cervix with negative para-aortic lymph nodes: a Gynecologic Oncology Group and Southwest Oncology Group study. J Clin Oncol 1999;17:1339-1348. 175. Pearcey R, Brundage M, Drouin P, et al. A clinical trial comparing concurrent cisplatin and radiation therapy versus radiation alone for locally advanced squamous cell carcinoma of the cervix carried out by the National Cancer Institute of Canada Clinical Trials Group [abstract]. Proc Am Soc Clin Oncol 2000;19:378a. 176. Thomas G, Dembo A, Ackerman I, et al. A randomized trial of standard versus partially hyperfractionated radiation with or without concurrent 5-fluorouracil in locally advanced cervical cancer. Gynecol Oncol 1998;69:137-145. 177. Lanciano R, Calkins A, Bundy BN, et al. Randomized comparison of weekly cisplatin or protracted venous infusion of ­fluorouracil in combination with pelvic radiation in advanced cervix cancer: a Gynecologic Oncology Group study. J Clin ­Oncol 2005;23:8289-8295. 178. Lorvidhaya V, Chitapanarux I, Sangruchi S, et al. Concurrent mitomycin C, 5-fluorouracil, and radiotherapy in the treatment of locally advanced carcinoma of the cervix: a randomized trial. Int J Radiat Oncol Biol Phys 2003;55:1226-1232. 179. Wong LC, Ngan HY, Cheung AN, et al. Chemoradiation and adjuvant chemotherapy in cervical cancer. J Clin Oncol 1999;17:2055-2060. 180. Morris M, Eifel PJ, Lu J, et al. Pelvic radiation with concurrent chemotherapy compared with pelvic and paraaortic radiation for high-risk cervical cancer. N Engl J Med 1999;340:1137-1143. 181. Barillot I, Horiot JC, Pigneux J, et al. Carcinoma of the intact uterine cervix treated with radiotherapy alone: a French cooperative study: update and multivariate analysis of prognostic factors. Int J Radiat Oncol Biol Phys 1997;38:969-978.

182. Coia L, Won M, Lanciano R, et al. The Patterns of Care Outcome Study for cancer of the uterine cervix. Results of the second national practice survey. Cancer 1990;66:2451-2456. 183. Kapp KS, Stuecklschweiger GF, Kapp DS, et al. Prognostic factors in patients with carcinoma of the uterine cervix treated with external beam irradiation and IR-192 high-dose-rate brachytherapy. Int J Radiat Oncol Biol Phys 1998;42:531-540. 184. Perez CA, Camel HM, Kuske RR, et al. Radiation therapy alone in the treatment of carcinoma of the uterine cervix: a 20-year experience. Gynecol Oncol 1986;23:127-140. 185. Petereit DG, Sarkaria JN, Potter DM, et al. High-dose-rate versus low-dose-rate brachytherapy in the treatment of cervical cancer: analysis of tumor recurrence—the University of Wisconsin experience. Int J Radiat Oncol Biol Phys 1999;45:12671274. 186. Chauvergne J, Rohart J, Héron JF, et al. Essai randomisé de chimiothérapie initiale dans 151 carcinomes du col utérin localement étendus (T2b-N1, T3b, M0). Bull Cancer 1990;77: 1007-1024. 187. Kumar L, Kaushal R, Nandy M, et al. Chemotherapy followed by radiotherapy versus radiotherapy alone in locally advanced cervical cancer: a randomized study. Gynecol Oncol 1994;54: 307-315. 188. Leborgne F, Leborgne JH, Doldán R, et al. Induction chemotherapy and radiotherapy of advanced cancer of the cervix: a pilot study and phase III randomized trial. Int J Radiat Oncol Biol Phys 1997;37:343-350. 189. Souhami L, Gil R, Allan S, et al. A randomized trial of chemotherapy followed by pelvic radiation therapy in stage IIIB carcinoma of the cervix. Int J Radiat Oncol Biol Phys 1991;29: 970-997. 190. Sundfør K, Trope CG, Hogberg T, et al. Radiotherapy and neoadjuvant chemotherapy for cervical carcinoma. A randomized multicenter study of sequential cisplatin and 5-fluorouracil and radiotherapy in advanced cervical carcinoma stage 3B and 4A. Cancer 1996;77:2371-2378. 191. Tattersall MHN, Lorvidhaya V, Vootiprux V, et al. Randomized trial of epirubicin and cisplatin chemotherapy followed by pelvic radiation in locally advanced cervical cancer. Am J Clin Oncol 1995;13:444-451. 192. Tattersall MHN, Ramirez C, Coppleson M. A randomized trial comparing platinum-based chemotherapy followed by radiotherapy vs. radiotherapy alone in patients with locally advanced cervical cancer. Int J Gynecol Cancer 1992;2:244-251. 193. Rotman M, Pajak M, Choi K, et al. Prophylactic extended-field irradiation of para-aortic lymph nodes in stages IIB and bulky IB and IIA cervical carcinomas. Ten-year treatment results of RTOG 79-20. JAMA 1995;274:387-393. 194. Haie C, Pejovic MH, Gerbaulet A, et al. Is prophylactic para-aortic irradiation worthwhile in the treatment of advanced cervical carcinoma? Results of a controlled clinical trial of the EORTC radiotherapy group. Radiother Oncol 1988;11:101-112. 195. Grigsby PW, Lu JD, Mutch DG, et al. Twice-daily fractionation of external irradiation with brachytherapy and chemotherapy in carcinoma of the cervix with positive para-aortic lymph nodes: phase II study of the Radiation Therapy Oncology Group 92-10. Int J Radiat Oncol Biol Phys 1998;41:817-822. 196. Varia MA, Bundy BN, Deppe G, et al. Cervical carcinoma metastatic to para-aortic nodes: extended field radiation therapy with concomitant 5-fluorouracil and cisplatin chemotherapy: a Gynecologic Oncology Group study. Int J Radiat Oncol Biol Phys 1998;42:1015-1023. 197. Malfetano JH, Keys H. Aggressive multimodality treatment for cervical cancer with paraaortic lymph node metastases. Gynecol Oncol 1991;42:44-47. 198. Eifel PJ, Jhingran A, Brown J, et al. Time course and outcome of central recurrence after radiation therapy for carcinoma of the cervix. Int J Gynecol Cancer 2006;16:1106-1111. 199. Shingleton HM, Soong SJ, Gelder MS, et al. Clinical and histopathologic factors predicting recurrence and survival after pelvic exenteration for cancer of the cervix. Obstet Gynecol 1989;73:1027-1034.

29  The Uterine Cervix 200. Hong JH, Tsai CS, Lai CH, et al. Recurrent squamous cell carcinoma of cervix after definitive radiotherapy. Int J Radiat Oncol Biol Phys 2004;60:249-257. 201. Coleman RL, Keeney ED, Freedman RS, et al. Radical hysterectomy for recurrent carcinoma of the uterine cervix after radiotherapy. Gynecol Oncol 1994;55:29-35. 202. Berek JS, Hacker NF, Lagasse LD. Rectosigmoid colectomy and reanastomosis to facilitate resection of primary and recurrent gynecologic cancer. Obstet Gynecol 1984;64:715-720. 203. Hatch KD, Shingleton H, Soong S, et al. Anterior pelvic exenteration. Gynecol Oncol 1988;31:205-216. 204. Stanhope CR, Webb MJ, Podratz KC. Pelvic exenteration for recurrent cervical cancer. Clin Obstet Gynecol 1990;33:897-909. 205. Höckel M, Baußmann E, Mitze M, et al. Are pelvic side-wall recurrences of cervical cancer biologically different from central relapses? Cancer 1994;74:648-655. 206. Mahé MA, Gérard JP, Dubois JB, et al. Intraoperative radiation therapy in recurrent carcinoma of the uterine cervix: report of the French Intraoperative Group on 70 patients. Int J Radiat ­Oncol Biol Phys 1996;34:21-26. 207. Stelzer KJ, Koh WJ, Greer BE, et al. The use of intraoperative radiation therapy in radical salvage for recurrent cervical cancer: outcome and morbidity. Am J Obstet Gynecol 1995;172:1881-1888. 208. Deutsch M, Parsons J. Radiotherapy for carcinoma of the cervix recurrent after surgery. Cancer 1974;34:2051-2055. 209. Hogan WM, Littman P, Griner L, et al. Results of radiation therapy given after radical hysterectomy. Cancer 1982;49:1278-1285. 210. Ijaz T, Eifel PJ, Burke T, et al. Radiation therapy of pelvic recurrence after radical hysterectomy for cervical carcinoma. Gynecol Oncol 1998;70:241-246. 211. Jobsen JJ, Leer JWH, Cleton FJ, et al. Treatment of locoregional recurrence of carcinoma of the cervix by radiotherapy after primary surgery. Gynecol Oncol 1989;33:368-371. 212. Potter ME, Alvarez RD, Gay FL, et al. Optimal therapy for pelvic recurrence after radical hysterectomy for early-stage cervical cancer. Gynecol Oncol 1990;37:74-77. 213. Thomas GM, Dembo AJ, Black B, et al. Concurrent radiation and chemotherapy for carcinoma of the cervix recurrent after radical surgery. Gynecol Oncol 1987;27:254-260. 214. Wang CJ, Lai CH, Huang HJ, et al. Recurrent cervical carcinoma after primary radical surgery. Am J Obstet Gynecol 1999;181:518-524. 215. Roman L, Morris M, Eifel P, et al. Reasons for inappropriate simple hysterectomy in the presence of invasive cancer of the cervix. Obstet Gynecol 1992;79:485-489. 216. Andras EJ, Fletcher G, Rutledge F. Radiotherapy of carcinoma of the cervix following simple hysterectomy. Am J Obstet Gynecol 1973;115:647-655. 217. Choi DH, Huh SJ, Nam KH. Radiation therapy results for patients undergoing inappropriate surgery in the presence of invasive cervical carcinoma. Gynecol Oncol 1997;65:506-511. 218. Hopkins MP, Peters WA, Anderson W, et al. Invasive cervical cancer treated initially by standard hysterectomy. Gynecol Oncol 1990;36:7-12. 219. Roman LD, Morris M, Mitchell MF, et al. Prognostic factors for patients undergoing simple hysterectomy in the presence of invasive cancer of the cervix. Gynecol Oncol 1993;50:179-184. 220. Benedet J, Odicino F, Maisonneuve P, et al. Carcinoma of the cervix uteri. J Epidemiol Biostat 1998;3:5-34. 221. Alberts DS, Garcia D, Mason-Liddil N. Cisplatin in advanced cancer of the cervix: an update. Semin Oncol 1991;18:11-24. 222. Park RC, Thigpen JT. Chemotherapy in advanced and recurrent cervical cancer. A review. Cancer 1993;71:1446-1450. 223. Omura GA, Blessing J, Vaccarello L, et al. A randomized trial of cisplatin versus cisplatin + mitolactol versus cisplatin + ifosfamide in advanced squamous cell carcinoma of the cervix by the Gynecologic Oncology Group [abstract]. Gynecol Oncol 1996;60:120. 224. Long HJ III, Bundy BN, Grendys EC Jr, et al. Randomized phase III trial of cisplatin with or without topotecan in carcinoma of the uterine cervix: a Gynecologic Oncology Group Study. J Clin Oncol 2005;23:4626-4633.

771

225. Long HJ III, Monk BJ, Huang HQ, et al. Clinical results and quality of life analysis for the MVAC combination (methotrexate, vinblastine, doxorubicin, and cisplatin) in carcinoma of the uterine cervix: a Gynecologic Oncology Group study. Gynecol Oncol 2006;100:537-543. 226. Boulware RJ, Caderao JB, Delclos L, et al. Whole pelvis megavoltage irradiation with single doses of 1000 rad to palliate advanced gynecologic cancers. Int J Radiat Oncol Biol Phys 1979;5:333-338. 227. Spanos WJ, Perez CA, Marcus S, et al. Effect of rest interval on tumor and normal tissue response—a report of phase III study of accelerated split course palliative radiation for advanced pelvic malignancies (RTOG-8502). Int J Radiat Oncol Biol Phys 1993;25:399-403. 228. Spanos WJ, Wasserman T, Meoz R, et al. Palliation of advanced pelvic malignant disease with large fraction pelvic radiation and misonidazole: final report of RTOG phase I/II study. Int J Radiat Oncol Biol Phys 1987;13:1479-1482. 229. Beadle B, Jhingran A, Salehpour M, et al. Tumor regression and organ motion during the course of chemoradiation for cervical cancer. Int J Radiat Biol Phys 2006;66:S44. 230. Ahamad A, D’Souza W, Salehpour M, et al. Intensity-modulated radiation therapy after hysterectomy: comparison with conventional treatment and sensitivity of the normal-tissue-sparing effect to margin size. Int J Radiat Oncol Biol Phys 2005;62:1117-1124. 231. Brixey CJ, Roeske JC, Lujan AE, et al. Impact of intensity-modulated radiotherapy on acute hematologic toxicity in women with gynecologic malignancies. Int J Radiat Oncol Biol Phys 2002;54:1388-1396. 232. Mundt AJ, Mell LK, Roeske JC. Preliminary analysis of chronic gastrointestinal toxicity in gynecology patients treated with intensity-modulated whole pelvic radiation therapy. Int J Radiat Oncol Biol Phys 2003;56:1354-1360. 233. Mundt AJ, Roeske JC, Lujan AE. Intensity-modulated radiation therapy in gynecologic malignancies. Med Dosim 2002;27: 131-136. 234. Lujan AE, Mundt AJ, Yamada SD, et al. Intensity-modulated radiotherapy as a means of reducing dose to bone marrow in gynecologic patients receiving whole pelvic radiotherapy. Int J Radiat Oncol Biol Phys 2003;57:516-521. 235. Huh SJ, Park W, Han Y. Interfractional variation in position of the uterus during radical radiotherapy for cervical cancer. Radio­ ther Oncol 2004;71:73-79. 236. Salama JK, Mundt AJ, Roeske J, et al. Preliminary outcome and toxicity report of extended-field, intensity-modulated radiation therapy for gynecologic malignancies. Int J Radiat Oncol Biol Phys 2006;65:1170-1176. 237. Chao C, Williamson JF, Grigsby PW, et al. Uterosacral space involvement in locally advanced carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys 1998;40:397-403. 238. McIntyre JF, Eifel PJ, Levenback C, et al. Ureteral stricture as a late complication of radiotherapy for stage IB carcinoma of the uterine cervix. Cancer 1995;75:836-843. 239. Delclos L, Fletcher GH, Moore EB, et al. Minicolpostats, dome cylinders, other additions and improvements of the Fletcher-Suit afterloadable system: indications and limitations of their use. Int J Radiat Oncol Biol Phys 1980;6:1195-1206. 240. Gerbaulet AL, Kunkler IH, Kerr GR, et al. Combined radiotherapy and surgery: local control and complications in early carcinoma of the uterine cervix—the Villejuif experience, 1975-1984. Radiother Oncol 1992;23:66-73. 241. Jhingran A, Burke TW, Eifel PJ. Definitive radiotherapy for patients with isolated vaginal recurrence of endometrial carcinoma after hysterectomy. Int J Radiat Oncol Biol Phys 2003;56: 1366-1372. 242. Schwarz G. An evaluation of the Manchester system of treatment of carcinoma of the cervix. Am J Roentgenol Radium Ther Nucl Med 1969;105:579-585. 243. International Commission on Radiation Units and ­Measurements. Dose and Volume Specification for Reporting Intracavitary Therapy in Gynecology. Bethesda: International Commission on Radiation Units and Measurements, 1985.

772

PART 9  Female Genital Tract

244. Katz A, Eifel PJ. Quantification and correlation of intracavitary brachytherapy parameters in patients with carcinoma of the cervix. Int J Radiat Oncol Biol Phys 2000;48:1417-1425. 245. Hamberger AD, Unal A, Gershenson DM, et al. Analysis of the severe complications of irradiation of carcinoma of the cervix: whole pelvis irradiation and intracavitary radium. Int J Radiat Oncol Biol Phys 1983;9:367-371. 246. Perez CA, Grigsby PW, Lockett MA, et al. Radiation therapy morbidity in carcinoma of the uterine cervix: dosimetric and clinical correlation. Int J Radiat Oncol Biol Phys 1999;44: 855-866. 247. Pourquier H, Delard R, Achille E, et al. A quantified approach to the analysis and prevention of urinary complications in radiotherapeutic treatment of cancer of the cervix. Int J Radiat Oncol Biol Phys 1987;13:1025-1033. 248. Schoeppel SL, LaVigne ML, Martel MK, et al. Three-dimensional treatment planning of intracavitary gynecologic implants: analysis of ten cases and implications for dose specification. Int J Radiat Oncol Biol Phys 1994;28:277-283. 249. Stuecklschweiger GF, Arian-Schad KS, Poier E, et al. Bladder and rectal dose of gynecologic high-dose-rate implants: comparison of orthogonal radiographic measurements with in vivo and CT- assisted measurements. Radiology 1991;181:889-894. 250. Pelloski CE, Palmer M, Chronowski GM, et al. Comparison ­between CT-based volumetric calculations and ICRU referencepoint estimates of radiation doses delivered to bladder and rectum during intracavitary radiotherapy for cervical cancer. Int J Radiat Oncol Biol Phys 2005;62:131-137. 251. Fellner C, Potter R, Knocke TH, et al. Comparison of radiography- and computed tomography-based treatment planning in cervix cancer in brachytherapy with specific attention to some quality assurance aspects. Radiother Oncol 2001;58:53-62. 252. Haie-Meder C, Potter R, Van Limbergen E, et al. Recommendations from Gynaecological (GYN) GEC-ESTRO Working Group (I): concepts and terms in 3D image based 3D treatment planning in cervix cancer brachytherapy with emphasis on MRI assessment of GTV and CTV. Radiother Oncol 2005;74: 235-245. 253. Haie-Meder C, Kramar A, Lambin P, et al. Analysis of complications in a prospective randomized trial comparing two brachytherapy low dose rates in cervical carcinoma. Int J Radiat Oncol Biol Phys 1994;29:1195-1197. 254. Patel FD, Sharma SC, Neigi PS, et al. Low dose rate vs. high dose rate brachytherapy in the treatment of carcinoma of the uterine cervix: a clinical trial. Int J Radiat Oncol Biol Phys 1993;28: 335-341. 255. Shigematsu Y, Nishiyama K, Masaki N, et al. Treatment of carcinoma of the uterine cervix by remotely controlled afterloading intracavitary radiotherapy with high-dose rate: a comparative study with a low-dose rate system. Int J Radiat Oncol Biol Phys 1983;9:351-356. 256. Teshima T, Inoue T, Ikeda H, et al. High-dose rate and low-dose rate intracavitary therapy for carcinoma of the uterine cervix. Final results of Osaka University Hospital. Cancer 1993;72:24092414. 257. Lertsanguansinchai P, Lertbutsayanukul C, Shotelersuk K, et al. Phase III randomized trial comparing LDR and HDR brachytherapy in treatment of cervical carcinoma. Int J Radiat Oncol Biol Phys 2004;59:1424-1431. 258. Eifel PJ. High dose-rate brachytherapy for carcinoma of the cervix: high tech or high risk? Int J Radiat Oncol Biol Phys 1992;24:383-386. 259. Scalliet P, Gerbaulet A, Dubray B. HDR versus LDR gynecological brachytherapy revisited. Radiother Oncol 1993;28:118126. 260. Akine Y, Arimoto H, Ogino T, et al. High-dose-rate intracavitary irradiation in the treatment of carcinoma of the uterine cervix: early experience with 84 patients. Int J Radiat Oncol Biol Phys 1988;14:893-898. 261. Nakano T, Kato S, Ohno T, et al. Long-term results of high-dose rate intracavitary brachytherapy for squamous cell carcinoma of the uterine cervix. Cancer 2005;103:92-101.

262. Fu KK, Phillips TL. High-dose-rate versus low-dose-rate intracavitary brachytherapy for carcinoma of the cervix. Int J Radiat Oncol Biol Phys 1990;19:791-796. 263. Souhami L, Corns R, Duclos M, et al. Long-term results of highdose rate brachytherapy in cervix cancer using a small number of fractions. Gynecol Oncol 2005;97:508-513. 264. Chen M, Lin F, Hong C, et al. High-dose-rate afterloading technique in the radiation treatment of uterine cervical cancer: 399 cases and 9 years experience. Int J Radiat Oncol Biol Phys 1991;20:915-919. 265. Utley J, von Essen C, Horn R, et al. High-dose-rate afterloading brachytherapy in carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys 1984;10:2259-2263. 266. Petereit DG, Pearcey R. Literature analysis of high dose rate brachytherapy fractionation schedules in the treatment of cervical cancer: is there an optimal fractionation schedule? Int J Radiat Oncol Biol Phys 1999;43:359-366. 267. Arai T, Nakano T, Morita S, et al. High dose rate remote afterloading intracavitary radiation therapy for cancer of the uterine cervix. Cancer 1992;69:175-180. 268. Jhingran A, Eifel PJ. Perioperative and postoperative complications of intracavitary radiation for FIGO stage I-III carcinoma of the cervix. Int J Radiat Oncol Biol Phys 2000;46:1177-1183. 269. Petereit DG, Sarkaria JN, Chappell RJ. Perioperative morbidity and mortality of high-dose-rate gynecologic brachytherapy. Int J Radiat Oncol Biol Phys 1998;42:1025-1031. 270. Bachtiary B, Dewitt A, Pintilie M, et al. Comparison of late toxicity between continuous low-dose-rate and pulsed-dose-rate brachytherapy in cervical cancer patients. Int J Radiat Oncol Biol Phys 2005;63:1077-1082. 271. Haie-Meder C, Peiffert D. [Innovation in gynaecological brachytherapy: new technologies, pulse dose-rate brachytherapy, image, definition of new volumes of interest and their impact on dosimetry: application in a clinical research programme “STIC”]. Cancer Radiother 2006;10:402-409. 272. Aristizabal SA, Woolfitt B, Valencia A, et al. Interstitial parametrial implants in carcinoma of the cervix stage II-B. Int J Radiat Oncol Biol Phys 1987;13:445-450. 273. Martinez A, Edmundson GK, Cox RS, et al. Combination of external beam irradiation and multiple-site perineal applicator (MUPIT) for treatment of locally advanced or recurrent prostatic, anorectal, and gynecologic malignancies. Int J Radiat Oncol Biol Phys 1985;11:391-398. 274. Syed AMN, Puthwala AA, Neblett D, et al. Transperineal interstitial-intracavitary “Syed-Neblett” applicator in the treatment of carcinoma of the uterine cervix. Endocuriether Hyperther Oncol 1986;2:1-13. 275. Hughes-Davies L, Silver B, Kapp D. Parametrial interstitial brachytherapy for advanced or recurrent pelvic malignancy: the Harvard/Stanford experience. Gynecol Oncol 1995;58:24-27. 276. Monk BJ, Tewari K, Burger RA, et al. A comparison of intracavitary versus interstitial irradiation in the treatment of cervical cancer. Gynecol Oncol 1997;67:241-247. 277. Choi JC, Ingenito AC, Nanda RK, et al. Potential decreased morbidity of interstitial brachytherapy for gynecologic malignancies using laparoscopy: a pilot study. Gynecol Oncol 1999;73:210215. 278. Erickson B, Gillin MT. Interstitial implantation of gynecologic malignancies. J Surg Oncol 1997;66:285-295. 279. Gupta AK, Vicini FA, Frazier AJ, et al. Iridium-192 transperineal interstitial brachytherapy for locally advanced or recurrent gynecological malignancies. Int J Radiat Oncol Biol Phys 1999;43:1055-1060. 280. Greven KM. Interstitial radiation for recurrent cervix or endometrial cancer in the suburethral region. Int J Radiat Oncol Biol Phys 1998;41:831-834. 281. Frank SJ, Jhingran A, Levenback C, et al. Definitive treatment of vaginal cancer with radiation therapy. Int J Radiat Oncol Biol Phys 2003;57:S194. 282. Kim RY, Pareek P, Duan J, et al. Postoperative intravaginal brachytherapy for endometrial cancer; dosimetric analysis of vaginal colpostats and cylinder applicators. Brachytherapy 2002;1:138-144.

29  The Uterine Cervix 283. Petereit DG, Sarkaria JN, Chappell R, et al. The adverse effect of treatment prolongation in cervical carcinoma. Int J Radiat Oncol Biol Phys 1995;32:1301-1307. 284. Girinski T, Rey A, Roche B, et al. Overall treatment time in advanced cervical carcinomas: a critical parameter in treatment outcome. Int J Radiat Oncol Biol Phys 1993;27:1051-1056. 285. Eifel P, Thames H. Has the influence of treatment duration on local control of carcinoma of the cervix been defined? Int J ­Radiat Oncol Biol Phys 1995;32:1527-1529. 286. Fyles A, Keane TJ, Barton M, et al. The effect of treatment duration in the local control of cervix cancer. Radiother Oncol 1992;25:273-279. 287. Eifel PJ, Moughan J, Owen JB, et al. Patterns of radiotherapy practice for patients with squamous carcinoma of the uterine cervix. A Patterns of Care study. Int J Radiat Oncol Biol Phys 1999;43:351-358. 288. Rose P. Locally advanced cervical carcinoma: the role of chemoradiation. Semin Oncol 1994;21:47-53. 289. McNeil C. New standard of care for cervical cancer sets stage for next questions [news]. J Natl Cancer Inst 1999;91:500a-501a. 290. Eifel PJ, Winter K, Morris M, et al. Pelvic radiation with concurrent chemotherapy versus pelvic and para-aortic radiation for high-risk cervical cancer: an update of RTOG 90-01 [abstract]. Int J Radiat Oncol Biol Phys 2002;54;(Suppl 1):1. 291. Pearcey R, Brundage M, Drouin P, et al. Phase III trial comparing radical radiotherapy with and without cisplatin chemotherapy in patients with advanced squamous cell cancer of the cervix. J Clin Oncol 2002;20:966-972. 292. Wong L, Choo Y, Choy D, et al. Long-term follow-up of potentiation of radiotherapy by cis-platinum in advanced cervical cancer. Gynecol Oncol 1989;35:159-163.

773

293. Chaney AW, Eifel PJ, Logsdon MD, et al. Mature results of a pilot study of pelvic radiotherapy with concurrent continuous infusion intra-arterial 5-FU for stage IIIB-IVA squamous cell carcinoma of the cervix. Int J Radiat Oncol Biol Phys 1999;45: 113-118. 294. Roberts KB, Urdaneta N, Vera R, et al. Interim results of a randomized trial of mitomycin C as an adjunct to radical radiotherapy in the treatment of locally advanced squamous-cell ­carcinoma of the cervix. Int J Cancer 2000;90:206-223. 295. Grigsby PW, Heydon K, Mutch DG, et al. Long-term follow-up of RTOG 92-10: cervical cancer with positive para-aortic lymph nodes. Int J Radiat Oncol Biol Phys 2001;51:982-987. 296. Eifel PJ. Chemoradiation for carcinoma of the cervix: advances and opportunities. Radiat Res 2000;154:229-236. 297. Benedetti-Panici P, Greggi S, Colombo A, et al. Neoadjuvant chemotherapy and radical surgery versus exclusive radiotherapy in locally advanced squamous cell cervical cancer: results from the Italian multicenter randomized study. J Clin Oncol 2002;20:179-188. 298. Eifel PJ, Jhingran A, Bodurka DC, et al. Correlation of smoking history and other patient characteristics with major complications of pelvic radiation therapy for cervical cancer. J Clin Oncol 2002;20:3651-3657. 299. Lanciano RM, Martz D, Montana GS, et al. Influence of age, prior abdominal surgery, fraction size, and dose on complications after radiation therapy for squamous cell cancer of the uterine cervix. A Patterns of Care Study. Cancer 1992;69:2124-2130.

30 The Endometrium Anuja Jhingran and Patricia J. Eifel ANATOMY OF THE UTERUS EFFECTS OF RADIATION ON THE NORMAL UTERUS ROUTES OF ENDOMETRIAL CANCER SPREAD PATHOLOGIC CLASSIFICATION Endometrioid Adenocarcinoma Mucinous Carcinoma Papillary Serous Carcinoma Clear Cell Carcinoma Other Subtypes PRETREATMENT EVALUATION STAGING PROGNOSTIC FACTORS Age Histologic Grade Histologic Subtype Depth of Myometrial Invasion

Carcinoma of the endometrium is the most common malignancy of the female genital tract in the United States. For 2008, the estimated number of new cases was 40,100, and the number of deaths was 7470.1 However, because endometrial carcinoma characteristically manifests as early-stage disease, usually with symptoms of bleeding in postmenopausal women, most cases are curable. The peak incidence occurs in the 50- to 70-year-old age group, with 75% of all cases occurring in postmenopausal women. Of the known risk factors for endometrial carcinoma, the one that most increases the risk is chronic estrogen exposure, whether because of use of oral unopposed estrogen or conditions that result in chronic increased or prolonged endogenous estrogen exposure (e.g., estrogen-secreting tumors, low number of births of viable offspring, early menarche, late menopause, and morbid obesity).2,3 Increased risk also has been associated with hypertension, diabetes mellitus, and longterm tamoxifen use among women breast cancer.2 Obesity and comorbid illnesses often complicate treatment. Another risk factor for endometrial cancer is hereditary nonpolyposis colorectal cancer (HNPCC), also called Lynch syndrome, which is an autosomal dominant disorder affecting 5% to 10% of women with colorectal cancer.4 In addition to being at increased risk for developing colorectal cancer, the hallmark cancer of the HNPCC syndrome, women from HNPCC families are also at significantly increased risk for gynecologic malignancies. Endometrial cancer is the second most common malignancy in Lynch syndrome, with an estimated lifetime risk between 40% and 70%5; the corresponding risk of developing ovarian cancer is 12%.6 The genetic basis for Lynch’s syndrome seems to be mutations in DNA mismatch repair genes, predominantly

774

Cervical Invasion Tumor Size Lymph Node Involvement Tumor Cells in Peritoneal Washings Adnexal Metastases and Simultaneous Ovarian Primary Tumors TREATMENT Surgery Adjuvant Radiation Therapy Radiation Therapy Alone for Medically or Surgically Inoperable Disease Treatment of Recurrent Disease Chemotherapy for Advanced, Recurrent, or Metastatic Endometrial Carcinoma UTERINE SARCOMA SUMMARY

MLH1 and MSH2 and, to a lesser extent, MSH6; mutations in MSH6 tend to be more strongly associated with endometrial cancer.6 These mutations and the associated microsatellite instability can be detected by immunohistochemical staining and DNA amplification tests of tissue samples. Screening for Lynch’s syndrome should be considered for young women presenting with endometrial cancer or for women with a family history of colorectal cancer.

Anatomy of the Uterus The uterus is a hollow, muscular organ situated in the midline of the pelvis, extending at right angles from the vagina, and lying between the bladder and the rectum. The uterus is divided into the corpus, which constitutes the upper two thirds of the organ, and the cervix, which makes up the lower third. The uterine cavity is lined by a mucous membrane, the endometrium, which is composed of columnar cells forming many tubular glands that extend into the stroma and muscular layers. The muscular layer of the uterus, the myometrium, is made up of smooth muscle fibers, arranged in no definite pattern, that interlace randomly. The uterus is covered by an outer serous coat, the peritoneum, that forms the broad ligaments laterally.

Effects of Radiation on the Normal Uterus After irradiation of the female pelvis, cyclical uterine bleeding usually ceases. Irradiation of the pelvis with doses as low as 10 to 20 Gy induces ovarian failure, which is an important contributor to long-term atrophy of the ­endometrium.

30  The Endometrium

Histologically, irradiation of the pelvis leads to acute necrosis of the tubular glands of the endometrium that is later replaced by atrophy of the stroma and glands.7 When intracavitary radiation therapy (ICRT) is used to treat endometrial carcinoma, subacute ulcers of the luminal wall may be seen if the uterus is removed 6 to 8 weeks after therapy. These ulcers are usually focal and probably correspond to areas where radiation sources were in contact with the uterine wall. In time, these lesions are replaced by dense fibrous tissue. Usual doses (40 to 50 Gy) of external beam radiation therapy (EBRT) usually do not cause significant fibrosis of the myometrium. Despite the changes in the endometrium after radiation therapy, exposure to cyclical hormones after radiation therapy often induces uterine bleeding, even after combined EBRT and ICRT. This suggests that many women retain at least some functional endometrium even after high-dose radiation therapy. If the cervical os is obstructed, this bleeding can accumulate as a hematometra. Primary carcinomas of the endometrium occasionally develop in women who have undergone definitive irradiation for cervical cancer; this also suggests that some functional endometrium remains after high-dose radiation therapy. Successful intrauterine pregnancy is probably not possible after combined pelvic EBRT and ICRT. However, successful pregnancies have been reported after 25 to 50 Gy of EBRT.8 Premenarchal treatment of the uterus ­probably arrests its normal development; pregnancies that occur in women who have undergone such ­treatment often end in spontaneous miscarriage during the second ­trimester.9

Routes of Endometrial Cancer Spread Endometrial carcinomas usually originate in the endometrial epithelium but may then invade locally to involve the myometrium or uterine cervix or metastasize locally to the ovaries, parametrium, or vagina. The uterus has a rich and complex lymphatic network composed of four lymphatic trunks that emerge from the lateral borders of the corpus (Fig. 30-1).10 The more superior lymphatic trunks pass through the broad ligament, paralleling the fallopian tube, and drain to the external iliac nodes near the ovary; from there, the lymphatic trunks drain into the para-aortic lymph nodes. Lymphatic vessels from the inferior portion of the corpus pass through the broad ligament to the common iliac nodes. Subserosal lymphatics near the junction of the fallopian tube and the uterine body form anastomoses with lymphatics of the fallopian tube to drain directly to the para-aortic lymph nodes. Lymphatics also pass from the corpus by way of the ovarian pedicle to form anastomoses with lymphatics of the fallopian tube and empty into the femoral lymph nodes. Uterine lymphatics interconnect with vessels that drain from the paracervical and paravaginal tissues. Invasion of the lymphatic network of the myometrium in the subserosa with subsequent involvement of the paracervical and paravaginal lymphatics by malignant tumors arising in the uterine corpus is accepted as the cause of recurrence of cancer in the vaginal vault after hysterectomy.

775

Para-aortic nodes Common iliac nodes Hypogastric nodes Interiliac nodes External iliac nodes Inguinal nodes Femoral nodes

FIGURE 30-1.  Lymph drainage and lymph node groups of the uterine corpus. Notice the difference in drainage between the lower and upper uterine segments. (From Dede JA, Plentl AA, Moore JG. Recurrent endometrial carcinoma. Surg Gynecol ­Obstet 1968;126:533-542.)

Tumors that penetrate the uterine serosa may ­ directly invade adjacent tissues, such as the bladder, colon, or ­adnexa, or exfoliate into the abdominal cavity to form implant metastases. Small tumor fragments also may gain access to the peritoneal cavity by traversing the fallopian tubes. Hematogenous dissemination of endometrial carcinoma is uncommon. Sites of distant spread include lung, liver, bone, and brain.

Pathologic Classification The International Society of Gynecologic Pathologists has proposed a pathologic classification for endometrial carcinoma (Box 30-1).11 Aspects of these various cellular subclassifications are discussed briefly in the following paragraphs.

Endometrioid Adenocarcinoma Endometrial adenocarcinomas are the most common form of carcinoma of the endometrium and account for about 75% to 80% of all cases. These tumors vary from a well-­differentiated form in which glandular architecture ­preserved is in more than 90% of the tumor (grade 1) to an undifferentiated form in which glandular architecture is preserved in less than 50% of the tumor (grade 3). The histopathologic grade of endometrial carcinoma is so important prognostically that grade is part of the staging classification (see “Staging”). However, the assignment of the histopathologic grade, especially the distinction between grades 1 and 2, varies among pathologists. In a study by Taylor and colleagues,12 two gynecologic ­pathologists independently reviewed 85 hysterectomy specimens from patients with endometrial carcinoma and had a difference of opinion of at least 1 grade in 26% of the specimens. In view of the uncertain clinical significance of grade 2 pathologic findings, Taylor and ­colleagues12 proposed a two-tiered grading system to improve the reproducibility and the prognostic significance of the tumor grade.

776

PART 9  Female Genital Tract

BOX 30-1.  International Society of Gynecologic Pathologists’ Pathologic Classification of Endometrial Carcinoma Endometrial adenocarcinoma Papillary villoglandular Secretory Ciliated cell Adenocarcinoma with squamous differentiation Mucinous carcinoma Papillary serous carcinoma Clear cell carcinoma Squamous carcinoma Undifferentiated carcinoma Mixed types Miscellaneous carcinoma Metastatic carcinoma From Zaino RJ. Pathologic indicators of prognosis in endometrial adenocarcinoma. Selected aspects emphasizing the Gynecologic Oncology Group experience. Pathol Annu 1995;30:1-28.

Secretory carcinomas account for 2% of all endometrial carcinomas and have a very good outcome. Ciliated cell carcinoma is also uncommon and is often associated with prior estrogen use. The prognosis associated with both of these subtypes is quite good. About 50% to 60% of cases of endometrial adenocarcinoma show some squamous differentiation. For a tumor to be labeled “adenocarcinoma with squamous differentiation,” at least 10% of the tumor must show squamous differentiation. Adenocarcinoma with squamous differentiation represents a continuum of lesions, ranging from adenoacanthoma, which is exceptionally well differentiated and has a benign-looking squamous component, to adenosquamous carcinoma, which is poorly differentiated and has a malignant squamous component.

Mucinous Carcinoma Mucinous carcinomas, defined as tumors in which at least 50% of the cells are mucinous, can occur in the endometrium and are similar to those that arise from the endocervix. Mucinous carcinomas constitute about 9% of all endometrial carcinomas; the prognosis associated with mucinous carcinomas is similar to that of endometrial ­adenocarcinomas.

Papillary Serous Carcinoma First described by Hendrickson and colleagues in 1982,13 papillary serous carcinoma of the endometrium is morphologically identical to high-grade serous carcinoma of the ovary and represents 10% of all cases of stage I endometrial cancer. Numerous studies have confirmed that papillary serous carcinoma behaves very aggressively and is often associated with deep myometrial invasion and lymphovascular space invasion, a marked tendency for intraperitoneal dissemination, and a poor survival rate compared with other subtypes. Papillary serous carcinomas should be distinguished from other papillary lesions (e.g., villoglandular adenocarcinomas) that are not serous and that have low-grade cytologic features and a relatively good prognosis.

Clear Cell Carcinoma Clear cell carcinoma represents less than 4% of all cases of endometrial carcinoma and commonly occurs in older black women. Clear cell carcinoma often occurs with papillary serous carcinoma of the uterus, which is associated with a poor prognosis. However, the prognosis associated with pure clear cell carcinoma is less ominous.14 It is important to distinguish between clear cell carcinoma and grade 1 carcinoma with secretory vacuolization (“secretory carcinoma”), which is associated with an excellent prognosis.

Other Subtypes Sarcomas arising in the uterus are rare. Pure leiomyosarcomas account for 20% to 30% of all uterine sarcomas; the mean patient age is somewhat younger than for other types of sarcoma. Endometrial stromal sarcomas, which account for 15% to 20% of uterine sarcomas, include benign stromal nodules, low-grade tumors, and high-grade stromal sarcoma. Low-grade tumors (e.g., endolymphatic stromal myosis) tend to recur locally and at distant sites years after initial surgery. High-grade stromal sarcomas tend to recur within the first 2 years after treatment, most commonly in the pelvis. Carcinosarcoma (i.e., müllerian mixed tumor) is the most common sarcoma of the uterus, accounting for more than 50% of all uterine sarcomas. Tumors contain malignant epithelial and sarcomatous elements. Patients with this disease usually are older than 60 years. The most common heterologous type is rhabdomyosarcoma, which contains benign epithelial elements and malignant stromal features.

Pretreatment Evaluation The possibility of endometrial carcinoma should be considered for postmenopausal women with any vaginal ­bleeding, perimenopausal women with heavy or prolonged vaginal bleeding, and premenopausal women with abnormal bleeding patterns who are obese or oligo-ovulatory. ­Although a formal dilatation and curettage was ­traditionally the standard technique for diagnosis, outpatient endometrial biopsy has replaced dilatation and curettage in most situations.15,16 When cervical involvement is suspected, a cervical biopsy should be performed; fractional curettage usually is not very accurate in this situation.17 Because endometrial carcinomas are usually surgically staged, the focus of the pretreatment evaluation should be on determining whether the disease can be resected and on determining the patient’s operative risk. In addition to a biopsy, the pretreatment evaluation should include a careful history and physical examination, laboratory studies, and chest radiography. Further studies are probably warranted only for patients with abnormal findings on pelvic examination. A serum cancer antigen (CA) 125 assay should be considered for women with high-risk histologic subtypes, such as papillary serous carcinoma, because the CA 125 level may be predictive of occult extrauterine disease and may be useful as a tumor marker.18,19 Computed tomography (CT) is recommended for all patients with high-grade tumors or clinical stage II or more advanced disease, because CT may demonstrate nodal enlargement or extrauterine spread. Contrast-enhanced

30  The Endometrium

­ agnetic resonance imaging (MRI) is not particularly m helpful in detecting nodal or peritoneal involvement but has had an accuracy of 82% in demonstrating the depth of myometrial invasion20,21 and an accuracy of 90% in demonstrating cervical involvement.22 In a meta-analysis, ­Kinkel and colleagues23 found that for assessing ­ myometrial invasion, contrast-enhanced MRI was significantly ­ better than nonenhanced MRI or sonography and may have been better than CT. These investigators recommended contrast-enhanced MRI as the “one-stop” examination with the highest efficacy.23 Because many women with endometrial cancer are ­elderly and have comorbid medical conditions such as obesity, diabetes, and hypertension, the pretreatment ­evaluation should be individualized on the basis of findings from the medical history and general physical examination.

Staging In 1971, the International Federation of Gynecology and Obstetrics (FIGO) adopted a clinical staging system for carcinomas of the uterine corpus, which at that time were often treated with radiation therapy before hysterectomy. The clinical staging system stratified patients with ­ early disease on the basis of a fractional biopsy specimen from the endocervix and the endometrium, the depth of the uterine cavity, and findings on physical examination. However, in the mid-1980s, these techniques for assessment of disease volume and spread were found to yield erroneous findings in as many as one third of patients compared with histopathologic findings at the time of laparotomy.24 At the same time, a report by Creasman and colleagues25 showed that 22% of patients with clinical stage I disease were found to have disease outside the uterus when a surgical staging procedure was ­performed. Surgeons started doing surgical staging, which led the FIGO in 1989 to design a surgical staging scheme for ­endometrial cancer.26 This surgical staging system has since been updated; the most recent update was published in 2002 (Box 30-2).27 The FIGO surgical staging system includes important prognostic information about the depth of myometrial infiltration and resolves some problems with the earlier FIGO staging systems, providing more detailed descriptions of tumor grades and for the first time emphasizing the prognostic importance of cytologic grade, particularly for patients with papillary serous adenocarcinomas. However, the FIGO surgical staging system is still problematic for several reasons. First, some tumors classified as stage I or II in the clinical staging system are classified as stage III in the surgical staging system, making it impossible to translate from the surgical to the clinical staging system; consequently, current results cannot be related to those of any studies that include patients treated before 1988. Second, the clinical staging system must still be used to categorize disease in patients who are treated with radiation therapy alone or with preoperative EBRT. Third, although surgical staging is mandated, FIGO does not define the extent of the procedure, which varies widely. Fourth, patients with tumor cells in peritoneal washings are classified as having stage III disease even though patients for whom peritoneal fluid is the sole site of extrauterine disease have an ­excellent prognosis.

777

Prognostic Factors In the absence of good prospective randomized studies of adjuvant therapy for endometrial carcinoma, decisions about the need for adjuvant treatment after surgical staging must be based on a thorough understanding of the natural history of the disease and risk factors for recurrence. In particular, the histologic grade and subtype of the carcinoma, the extent of disease in the uterus, and the presence and distribution of extrauterine disease can be used to predict the risk of tumor recurrence in the vagina, pelvis, and abdomen and at distant sites. The pathologic stage incorporates many of these risk factors and therefore is prognostic for survival and local recurrence.

Age Younger patients, especially those younger than 60 years, have a significantly better prognosis than older patients.28-33 Currently, however, no specific treatment recommendations are based on age.

Histologic Grade Tumor grade is one of the most sensitive indicators of prognosis. Grade correlates directly with the depth of myometrial invasion (Table 30-1)25 and the frequency of lymph node involvement.25,34-36 In most reports, the 5-year survival rate for patients with grade 1 tumors is at least 90%; for patients with grade 3 tumors, the 5-year survival rate is 65% to 70%.34,36 Lewis and colleagues37 reported that the rate of nodal involvement was 5.5% for patients with stage I, grade 1 tumors and 26% for patients with stage I, grade 3 tumors. Several investigators have reported that the incidence of deep myometrial invasion of grade 3 tumors is three to four times the incidence of grade 1 tumors.

Histologic Subtype Several investigators have reported that adenosquamous carcinoma is more malignant than adenocarcinoma or adenoacanthoma, probably because of the malignant squamous elements.38-40 Patients with adenosquamous carcinoma are best treated as though they have high-grade carcinoma of the uterine corpus. Papillary serous carcinoma of the endometrium should be treated as an aggressive high-grade tumor because it behaves more like its ovarian counterpart than like a typical endometrioid adenocarcinoma. A 2000 study by Cirisano and colleagues41 found that patients with papillary serous or clear cell carcinoma with myometrial invasion less than 2 mm had an estimated 5-year survival rate of 56%, compared with 93% for patients with endometrial adenocarcinoma, and that surgically staged patients had the same outcome as clinically staged patients. Clear cell carcinoma associated with papillary serous carcinoma should be treated as papillary serous carcinoma; however, pure clear cell carcinoma may have a better outcome and can be treated as endometrial adenocarcinoma.14

Depth of Myometrial Invasion The depth of myometrial invasion is one of the most important predictors of lymph node involvement and prognosis (Table 30-2).25,31,32,42 The depth of myometrial invasion is more strongly associated with lymphatic and extrauterine spread, treatment failure, and relapse than is

778

PART 9  Female Genital Tract

BOX 30-2.  American Joint Committee on Cancer and International Federation of Gynecology and Obstetrics Staging Systems for Cancer of the Corpus Uteri TNM Categories

FIGO Stage

TX T0 Tis T1 T1a T1b T1c T2 T2a

— — 0 I IA IB IC II IIA

T2b T3 T3a

IIB III IIIA

T3b T4

IIIB IVA

Regional Lymph Nodes (N) NX N0 N1 IIIC Distant Metastasis (M) MX M0 M1 IVB

Stage Grouping Stage 0 Stage I Stage IA Stage IB Stage IC Stage II Stage IIA Stage IIB Stage III Stage IIIA Stage IIIB Stage IIIC Stage IVA Stage IVB

Tis T1 T1a T1b T1c T2 T2a T2b T3 T3a T3b T1 T2 T3 T4 Any T

Primary tumor cannot be assessed No evidence of primary tumor Carcinoma in situ Tumor confined to corpus uteri Tumor limited to endometrium Tumor invades less than one half of the myometrium Tumor invades one half or more of the myometrium Tumor invades cervix but does not extend beyond uterus Tumor limited to the glandular epithelium of the endocervix. There is no evidence of connective tissue stromal invasion Invasion of the stromal connective tissue of the cervix Local and/or regional spread as defined below Tumor involves serosa and/or adnexa (direct extension or metastasis) and/or cancer cells in ascites or peritoneal washings Vaginal involvement (direct extension or metastasis) Tumor invades bladder mucosa and/or bowel mucosa (bullous edema is not sufficient to classify a tumor as T4) Regional lymph node(s) cannot be assessed No regional lymph node metastasis Regional lymph node metastasis to pelvic and/or para-aortic nodes Distant metastasis cannot be assessed No distant metastasis Distant metastasis (includes metastasis to abdominal lymph nodes other than para-aortic, and/or inguinal lymph nodes; excludes metastasis to vagina, pelvic serosa, or adnexa) N0 N0 N0 N0 N0 N0 N0 N0 N0 N0 N0 N1 N1 N1 Any N Any N

M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M1

From Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 267-273.

histologic grade.43 Regardless of grade, tumors that have not invaded the myometrium rarely involve lymph nodes and are usually curable with hysterectomy alone (although high-grade tumors occasionally recur in the vaginal apex). Uterine papillary serous carcinoma may be an exception; this tumor has been reported to behave aggressively in the absence of demonstrable myometrial invasion.44

Cervical Invasion Tumors that involve the cervix are classified by FIGO as stage II. Patients with tumors that involve only endocervical glands (FIGO stage IIA) have a relatively good

prognosis, probably similar to that of patients with tumors of comparable grade and depth that are confined to the fundus. Tumors that involve the endocervical stroma are associated with a poorer prognosis, with a higher likelihood of extrauterine involvement and disease recurrence. Presumably, involvement of the cervical stroma gives the tumor access to cervical regional routes of spread, placing pelvic nodes at relatively greater risk. In a small series of surgically staged patients, Fanning and colleagues45 reported recurrence in 5 of 8 patients with stage IIB disease and none of 12 patients with stage IIA disease.

30  The Endometrium

779

TABLE 30-1.  Correlation between Tumor Grade and Depth of Myometrial Invasion Percent of Tumors Depth of Myometrial Invasion

Grade 1

Grade 2

Grade 3

Endometrium

24%

11%

7%

Inner third

53%

45%

35%

Middle third

12%

24%

16%

Outer third

10%

20%

42%

Modified from Creasman W, Morrow C, Bundy B, et al. Surgical pathologic spread patterns of endometrial cancer. A Gynecologic Oncology Group Study. Cancer 1987;60:2035-2041.

TABLE 30-2.  Correlations of Depth of Myometrial Invasion, Tumor Grade, and Pelvic Node Involvement Number (%) of Patients with Pelvic Node Involvement Depth of Myometrial Invasion

Grade 1 (n = 180)

Grade 2 (n = 288)

Grade 3 (n = 153)

Endometrium (n = 86)

0 (0)

1 (3)

0 (0)

Inner third (n = 281)

3 (3)

7 (5)

5 (9)

Middle third (n = 115)

0 (0)

6 (9)

1 (4)

Outer third (n = 139)

2 (11)

11 (19)

22 (34)

Modified from Creasman W, Morrow C, Bundy B, et al. Surgical pathologic spread patterns of endometrial cancer. A Gynecologic Oncology Group Study. Cancer 1987; 60:2035-2041.

TABLE 30-3. Correlations of Depth of Myometrial Invasion, Tumor Grade, and Para-aortic Node Involvement Number (%) of Patients with Para-aortic Node Involvement Depth of Myometrial Invasion

Grade 1 (n = 180)

Grade 2 (n = 288)

Grade 3 (n = 153)

Endometrium (n = 86)

0 (0)

1 (3)

0 (0)

Inner third (n = 281)

1 (1)

5 (4)

2 (4)

Middle third (n = 115)

1 (5)

0 (0)

0 (0)

Outer third (n = 139)

1 (6)

8 (14)

15 (23)

Modified from Creasman W, Morrow C, Bundy B, et al. Surgical pathologic spread patterns of endometrial cancer. A Gynecologic Oncology Group Study. Cancer 1987; 60:2035-2041.

Tumor Size Schink and colleagues46 reported a relationship between the size of endometrial primary tumors and rates of lymph node metastasis and survival. The predictive value of tumor size seemed to be independent of the grade or depth of invasion of the endometrial cancer. Patients with tumors smaller than 2 cm were at a 4% risk for nodal ­metastasis compared with a 15% risk for those with tumors larger than 2 cm. Similarly, Mariani and colleagues47 found no nodal metastases in patients with a primary tumor diameter of 2 cm or less and a rate of 7% among those with tumors larger than 2 cm. That study demonstrated a significant difference in recurrence based on tumor size; however, in multivariate analysis, size was not an independently significant predictor of recurrence.47 In another series of 367 patients reported by Shah and colleagues,48 patients with tumors 2 cm in diameter or ­ smaller had a nodal metastasis rate of 6.3%, compared with a rate of 26.3% for patients with tumors larger than 2 cm (P = .0005).48

In cases in which the indications for adjuvant treatment are in a gray zone, tumor size is one factor that may influence the decision about whether such treatment is warranted.

Lymph Node Involvement With the advent of systematic surgical staging, more detailed information is becoming available about the incidence and prognostic importance of various patterns of extrauterine spread. The largest prospective study of surgical staging was conducted by the Gynecologic Oncology Group (GOG).25,49 This study has provided some of the best data available about the relationships between hysterectomy findings, nodal involvement, and outcome (Tables 30-1 to 30-3).25 Overall, lymph node metastases were reported in 70 (11%) of 621 evaluable patients.25 Approximately one half of the patients with documented nodal involvement had positive para-aortic nodes, 32% of those with pelvic node metastases also had positive para-aortic nodes, and 12 of

780

PART 9  Female Genital Tract

cells in peritoneal washings is quite low, and this finding would most likely not be an independent predictor for relapse.

100

Percent recurrence-free

90 80

Adnexal Metastases and Simultaneous Ovarian Primary Tumors

70 60 50 40 Node status Recurrence free Failures Total Both negative 784 104 680 Pos. pelvic/ 63 18 45 Neg. aortic Positive aortics 48 28 20

30 20 10 0 0

12

24

36

48

60

Months on study

FIGURE 30-2.  Recurrence-free interval in 784 patients with negative pelvic and para-aortic nodes, 63 patients with positive pelvic but negative para-aortic nodes, and 48 patients with positive para-aortic nodes. (From Morrow C, Bundy B, Kurman R, et al. Relationship between surgical-pathological risk factors and outcome in clinical stage I and II carcinoma of the endometrium: a Gynecologic Oncology Group Study. Gynecol Oncol 1991;40:55-65.)

70 patients with regional metastases had only para-aortic node involvement. A strong correlation exists between lymph node involvement, particularly para-aortic node involvement, and a worse prognosis (Fig. 30-2).42 However, disease that has metastasized to regional nodes seems to be curable in many cases. The GOG reported a 70% survival rate at 5 years among patients with histologic documentation of pelvic node involvement and no para-aortic node involvement. Although treatment details were not given, it was stated that most patients underwent pelvic irradiation. The survival rate for patients with positive para-aortic nodes in that study was 40%, which is comparable to survival rates in other reports of patients treated with extended-field radiation therapy.49-52 However, the 5-year disease-free survival rate drops to 25% in the presence of nodal involvement plus peritoneal, serosal, adnexal, or vaginal metatasis.51

Tumor Cells in Peritoneal Washings Although the presence of tumor cells in peritoneal washings is taken into account in the current FIGO staging system for endometrial carcinoma, the significance of this finding is uncertain. Some investigators have reported a poorer prognosis for patients with tumor cells in peritoneal washings, but these studies have not consistently identified patients with papillary serous carcinomas.42,53 Other investigators54,55 have concluded that abnormal peritoneal cytology is not an independent risk factor for recurrence. In a multivariate analysis by Milosevic and colleagues,54 the conclusion was that although the presence of tumor cells in peritoneal washings was strongly associated with disease recurrence (odds ratio = 4.7; 95% confidence interval [CI], 3.5 to 6.3), this finding largely reflected the association of malignant cytology with other adverse prognostic factors, which dominate the clinical presentation. In the absence of any other adverse features, the true incidence of tumor

Although endometrial carcinomas can metastasize to the ovaries, some patients who present with simultaneous disease in the endometrium and ovary actually have multiple primary tumors. Patients who have simultaneous low-grade endometrioid neoplasms with minimal myometrial invasion and early ovarian disease tend to have an excellent prognosis, with survival rates of 80% to 100%, suggesting that these lesions do not represent metastatic ovarian cancer or metastatic endometrial cancer.56,57 However, 5% of patients with stage I or occult stage II disease do have adnexal involvement, and this is associated with a fourfold increase in the risk of lymph node involvement and a worse prognosis.

Treatment Surgery The mainstay of treatment for adenocarcinoma of the endometrium is surgical removal of the uterus. Total abdominal hysterectomy with bilateral salpingo-oophorectomy has been widely regarded as preferable for most patients, although several less invasive surgical procedures are also available. Total laparoscopic hysterectomy and staging in particular are increasingly being used for the surgical management of endometrial cancer. Retrospective series have shown that patients who had total laparoscopic hysterectomy had considerably lower perioperative morbidity rates (17%) than did those who had the abdominal procedure (43%), and they required less postoperative analgesic medication.58-60 Length of hospital stay and return to normal function were shorter for the laparoscopic group than for the open hysterectomy group.61 The laparoscopic and open procedures seem to be comparable in terms of cost-­effectiveness,62,63 with one study reporting that laparoscopic hysterectomy was associated with higher ­perioperative costs but that abdominal hysterectomy was associated with higher complication rates and longer hospital stays.64 Disease-free and overall survival rates seem to be similar for patients undergoing total laparoscopic hysterectomy or total abdominal hysterectomy.58,60,65,66 The GOG has completed a phase III trial comparing laparoscopy with laparotomy for surgical staging in 2616 women with endometrial cancer. Preliminary analysis indicated that laparoscopy was associated with fewer grade 2 or higher complications (according to the Common Terminology Criteria for Adverse Events) (P < .0001), less use of antibiotics (P < .0001), and similar rates of transfusion. Operative time was longer in the laparoscopic group (203 versus 136 minutes, P < .0001), but hospital stay was shorter (3 versus 4 days, P < .0001). Numbers of positive pelvic nodes were similar in both groups, and only 25% of the laparoscopic cases were converted to open (laparotomy) cases.67 Other preliminary analyses showed that patients who underwent the laparoscopic procedure had superior quality of life at 6 weeks after surgery compared with those who underwent laparotomy, but those benefits

30  The Endometrium

were not sustained at 6 months after surgery.68 Information on disease recurrence and survival from this trial is not yet available. Findings from another randomized trial being conducted in Australia are expected to provide additional information on this topic.69 Options for evaluation of the regional lymph nodes range from palpation to selective nodal sampling to complete lymphadenectomy. Palpation of the lymph nodes has been shown to be inaccurate for determining lymph node status because the size of nodal metastases is often below the threshold for clinical detection. In a study of 76 patients with primary endometrial cancer of various histologic types, Girardi and colleagues70 found that 37% of lymph node metastases were less than 2 mm in diameter and almost 50% were less than 1 cm in diameter. These findings are consistent with those of Creasman and colleagues,25 who found that less than 10% of patients with nodal metastases had grossly positive nodes. Controversy exists in the gynecologic oncology community regarding the proper extent and the therapeutic value of lymphadenectomy. Chuang and colleagues71 ­demonstrated that for diagnostic purposes, a limited ­selective lymphadenectomy in which about 10 nodes were ­ removed accurately reflected the true nodal status. ­However, the trend in gynecologic oncology is increasingly toward performing a complete lymphadenectomy in which 40 to 50 pelvic lymph nodes are retrieved. Some investigators believe that complete lymphadenectomy improves survival,72-74 whereas others have concluded that lymphadenectomy provides prognostic information only.29,71,75,76 A large retrospective study published in 2006 by Chan and colleagues77 showed that complete lymphadenectomy could be beneficial in endometrial cancer. In that study, records from 12,333 women with endometrial cancer in the Surveillance, Epidemiology, and End Results database were evaluated for lymph node assessment and information on recurrence and survival. In that analysis, more extensive lymph node resection was associated with improved 5-year disease-specific survival for women with intermediate- or high-risk disease; rates were 75.3% for those who had 1 node resected, 81.5% for those with 2 to 5 nodes, 84.1% for those with 6 to 10 nodes, 85.3% for those with 11 to 20 nodes, and 86.6% for those with more than 20 nodes resected (P < .001). A benefit from more extensive node resection was also seen among patients with stage IIIC to IV disease, but no benefit could be demonstrated for those with low-risk disease. In multivariate analyses, a more extensive node resection remained a significant prognostic factor for improved survival among women with ­intermediate- or high-risk disease after adjusting for other factors including age, year of diagnosis, disease stage, tumor grade, use of adjuvant radiation therapy, and the presence of positive nodes (P < .001). No information was available on complications, time to recurrence, subsequent surgical and medical therapies, or chemotherapy given during the initial treatment.77 The retrospective studies described here all have flaws that could affect outcome. They were not randomized, they lacked standard surgical management protocols or standard pathologic evaluation of surgical specimens, or they ­included uncontrolled factors such as selection bias, tumor factors, and comorbid conditions. The U.K. Medical Research Council is conducting a randomized trial in an attempt

781

to address some of these shortcomings. This trial, called A Study in the Treatment of Endometrial Cancer (ASTEC), was designed to evaluate the therapeutic effects of lymphadenectomy and subsequent adjuvant irradiation for clinical stage I or stage II disease. Patients were randomly assigned to undergo simple hysterectomy plus bilateral salpingooophorectomy with or without dissection of the iliac and obturator nodes. Routine para-aortic node dissection was not required. Preliminary results presented at the Society of Gynecologic Oncologists meeting in March 200678 indicated no significant difference in survival between the two groups at 3 years (89% for those with node dissection and 88% for those without). Node dissection was found to have no benefit after age, performance status, depth of myometrial penetration, or cell type was controlled for.78 Complete lymphadenectomy is associated with higher rates of acute and late complications than are seen with less thorough methods of surgical lymph node evaluation.42,73 The increased risk of complications is justified if unsuspected disease is detected that would lead to the use of potentially curative pelvic radiation therapy; however, the risk of complications increases drastically if pelvic radiation therapy is added after lymphadenectomy.79,80 Most gynecologic oncologists still recommend pelvic radiation therapy for patients with stage IC, grade 3 disease with negative pelvic lymph nodes.81 If pelvic radiation therapy is recommended on the basis of known high-risk uterine features, a selective lymphadenectomy in which the paraaortic lymph nodes are also sampled leads to a reduction in the risk of radiation therapy complications and allows the most appropriate radiation fields to be determined. Pelvic lymphadenectomy may indirectly improve the therapeutic ratio by allowing identification of the patients most likely to benefit from adjuvant pelvic irradiation. The available evidence, including the results from the ASTEC trial, suggests that treatment of endometrial cancer should be designed to maximize cure rates and minimize complications.

Adjuvant Radiation Therapy Despite decades of study, the role of adjuvant radiation therapy (preoperative or postoperative) in the treatment of endometrial cancer has been difficult to define. Historically, patients with early-stage endometrial cancer were treated with radiation therapy delivered using one of a variety of techniques, followed by extrafascial hysterectomy. Supplemental postoperative radiation therapy was added in selected cases on the basis of the pathologic risk factors (e.g., deep myometrial extension, cervical involvement, positive nodes). This was the standard treatment for early-stage endometrial cancer in the United States and other countries by the 1950s, although a few clinicians questioned the advantages of selective postoperative irradiation.82,83 Later clinicians became more aware of the value of operative findings as a guide to the selection of adjuvant treatment. Because preoperative radiation therapy has never been proven to be more effective than tailored postoperative radiation therapy, the use of preoperative irradiation has declined in favor of initial surgical treatment.84-86 Radiation therapy has been shown to reduce the risk of local recurrence, but its effect on survival is unclear. As a result, investigators continue to disagree about the role of adjuvant radiation therapy in the management of endometrial carcinoma.

782

PART 9  Female Genital Tract

Several factors have made it difficult to study the ­value of adjuvant irradiation. Because most newly diagnosed endometrial carcinomas are confined to the uterus, treatment failures after surgery alone are expected in only about 15% of patients. The margin of improvement that can be achieved with radiation therapy is small, and very large groups of patients or highly selected populations are needed to demonstrate the effects of adjuvant irradiation. The influence of physicians’ biases on the selection of ­patients for adjuvant treatment has limited the value of retrospective studies, and retrospective studies are difficult to compare because of changes over time in the classification systems for staging and grading endometrial carcinoma. Few randomized trials of adjuvant irradiation have been conducted with patients with endometrial carcinoma, and most of those trials have included mainly patients with low- to intermediate-risk disease, for whom survival differences may not be as apparent. For these reasons, the decision of whether to administer adjuvant irradiation is based primarily on clinicians’ understanding of the risk factors for recurrence and the natural history of the disease, their interpretation of data, and their estimates of the risk of complications in each individual case.

Preoperative Radiation Therapy The primary goal of preoperative ICRT is to prevent vaginal recurrence. In early series, overall vaginal recurrence rates after surgery alone were as high as 15% to 25%.87,88 A few investigators have reported much lower vaginal recurrence rates after surgery alone and have suggested that the wide range of reported local recurrence rates may reflect variations in surgical technique that influence the risk of tumor cell implantation.89,90 Most of the evidence suggests that for patients with high-risk disease, preoperative irradiation reduces the incidence of vaginal recurrence from 10% to 15% to less than 5%.90-98 In theory, patients with gross tumor involvement of the cervix stand to receive the greatest benefit from preoperative irradiation. For those patients, the preoperative intracavitary technique delivers a greater dose to the paracervical tissues than is possible with postoperative vaginal irradiation. Intrauterine applicators are usually loaded with 35 to 40 mg radium equivalent (RaEq) of cesium and left in place for approximately 72 hours for a total exposure of 3500 to 4000 mg RaEq-hours. Vaginal applicators are loaded to ­deliver 60 to 70 Gy to the surface of the apical vagina over the same period. After this dose of radiation, hysterectomy can be safely performed within 2 to 3 days.91,92 Combined preoperative EBRT and ICRT may be indicated for patients with nonserous tumors that extensively infiltrate the cervix. In such cases, the hysterectomy should be delayed until 4 to 6 weeks after irradiation. Alternatively, some surgeons recommend radical hysterectomy for patients with bulky stage II disease, but the morbidity associated with the more radical surgery is greater, and postoperative radiation therapy may still be necessary.

Postoperative Radiation Therapy for Intermediate-Risk Disease Pelvic Irradiation. Postoperative pelvic irradiation clearly reduces the risk of pelvic recurrence, but no study has yet determined whether it improves the survival rates of patients with ­intermediate- or high-risk disease. In 1980,

Aalders and colleagues32 published a report on a randomized prospective trial of primary surgery and vaginal brachytherapy with or without additional EBRT for patients with endometrial cancer that was clinically confined to the uterus. Comprehensive surgical staging was not done, but patients with “proven metastases at surgery” were excluded. A significant reduction in the rate of vaginal and pelvic recurrences was observed in the group that received adjuvant radiation therapy (1.9% versus 6.9%), but they also experienced more distant recurrences; as a result, the overall 5-year survival rate was not improved by the use of EBRT. Within the subset of patients with deeply invasive grade 3 disease, patients treated with radiation therapy seemed to have a somewhat better survival rate, but the number of patients in this subset was small, and the results were not statistically significant. A second prospective randomized trial of adjuvant radiation therapy for patients with endometrial carcinoma was conducted by the Post Operative Radiation Therapy in Endometrial Carcinoma (PORTEC) group in the Netherlands and reported in April 2000 by Creutzberg and colleagues.99 In that trial, 715 patients with intermediate-risk endometrial carcinoma were randomly assigned after surgery to receive EBRT with 46 Gy or no radiation therapy. Nodes considered “suspicious” were sampled during surgery. Disease considered “intermediate risk” included grade 1 tumors with more than 50% depth of invasion, grade 2 tumors with any degree of invasion, and grade 3 tumors with less than 50% invasion. (Grade 3 tumors with outerhalf penetration were not included in this study.) A centralized pathology review was performed after the study was concluded. Many of the patients in this study were at relatively low risk for extrauterine involvement. Approximately 60% had tumor penetration of more than 50%, but 60% had grade 1 tumors. Overall survival rates at 5 years were equivalent in the radiation and no-radiation groups (81% versus 85%); however, 5-year locoregional recurrence rates were 4% for the irradiated group and 14% for the control group (Fig. 30-3).99 Of the recurrences, 73% were restricted to the vagina. The lack of effect of radiation on survival was accounted for by the low risk of recurrence among the patients in the study and by the high salvage rates for those with vaginal recurrence. The rates of serious complications were 3% in the irradiated group and less than 1% in the surgery-only group. A third prospective randomized trial, GOG-99,79 was different from the other two trials in that patients underwent complete surgical staging with pelvic and para-aortic lymph node dissection instead of just sampling of abnormal nodes. Patients found to have intermediate-risk surgical stage I disease were then randomly assigned to receive no further therapy or pelvic irradiation with 50.4 Gy. Most patients had relatively low-risk disease features; 80% had grade 1 or 2 disease, and 60% had stage IB disease. Pelvic irradiation was found to reduce the risk of recurrence from 12% to 3% but did not affect overall survival (92% versus 86% for observation only).79 Further analysis, however, led to the definition of a high-intermediate risk group comprising 132 patients (33% of the study population) with ­several combinations of four independent risk factors (i.e., older age, grade 2 or 3 tumor, outer third invasion, and lymphovascular space invasion). The cumulative 2-year ­recurrence rates for this high-intermediate risk group were 26% for

30  The Endometrium 20 No radiotherapy Radiotherapy

Cumulative (%)

15

10

5

0 0

12

24

36

48

60

Time after randomization (months)

FIGURE 30-3.  Probability of locoregional (vaginal or pelvic) relapse in patients with intermediate-risk endometrial carcinoma assigned to postoperative radiation therapy or no further treatment. (From Creutzberg CL, van Putten WL, Koper PC, et al. Surgery and postoperative radiotherapy versus surgery alone for patients with stage-1 endometrial carcinoma: multicentre randomised trial. PORTEC Study Group. Post Operative Radiation Therapy in Endometrial Carcinoma. Lancet 2000;355: 1404-1411.)

those receiving no radiation therapy and 6% for those receiving radiation.79 The complication rate associated with the pelvic irradiation in the GOG-99 study was much higher than that in the PORTEC study, probably because of the extent of surgical staging. The rate of severe (grade 3 or 4) gastrointestinal toxicity with radiation were 8% in the GOG study but only 3% in the PORTEC study. In all three of these randomized studies, pelvic radiation therapy clearly reduced the risk of local recurrence. However too few patients with high-risk disease were included and too few disease-related deaths occurred to demonstrate or rule out moderate differences in survival between patient subgroups. In the study by Creutzberg and colleagues,99,100 grade 3 disease and deep myometrial invasion doubled the relative risk of locoregional relapse, but grade 3 disease was the strongest predictor of endometrial carcinoma–related death. However, only 74 of the 714 evaluable patients in the study had grade 3 disease—too few to permit any survival difference to be ruled out in this high-risk population. Several studies have documented survival rates of 80% to 90% for patients with high-risk disease (e.g., high grade or deeply invasive) treated with postoperative radiation therapy29,75,76,101; these findings also provide indirect evidence of the treatment’s efficacy for intermediate-risk disease. Carey and colleagues29 reported a 5-year relapse-free survival rate of 81% for 157 patients with clinical stage I disease and high-risk features (deep myometrial invasion, grade 3, adenosquamous histology, or cervical involvement). The local recurrence rate was 4%, compared with 29% in a separate group of 28 patients who refused adjuvant treatment. The 70% projected 5-year survival rate reported by the GOG for 63 patients with positive pelvic nodes and negative para-aortic nodes may provide ­indirect evidence

783

of a survival benefit of pelvic irradiation for ­patients with intermediate- or high-risk disease.42 Naumann and colleagues81 surveyed members of the Society of Gynecologic Oncologists to determine whether the findings from the GOG-99 trial influenced their practice. Although 27% of respondents stated that they were less likely to recommend adjuvant radiation therapy as a result of the study, more than 77% reported that they still recommended adjuvant irradiation for patients with “highrisk” disease (e.g., stage IC or stage IB, grade 3 disease). What can be learned from these three randomized studies? For patients with low-risk endometrial cancer (i.e., stage IA or IB, grade 1 or 2), no further adjuvant therapy is needed. For patients with intermediate-risk disease (i.e., stage IB, grade 3; IC, grade 1 or 2; or stage II, grade 1 or 2), omitting radiation leaves those patients at significant risk for vaginal and pelvic relapse but does not affect overall survival rates. Whether vaginal brachytherapy alone may be sufficient to prevent vaginal relapse in patients with ­intermediate-risk disease is discussed further in the following section. For patients with high-intermediate risk disease (defined in terms of the four factors identified in GOG-99: older age, grade 2 or 3, outer third invasion, and lymphovascular space invasion), pelvic irradiation continues to be the most effective adjuvant treatment for ensuring pelvic control and perhaps providing a survival advantage. Vaginal Brachytherapy. In selected patients with ­endometrial carcinoma, vaginal ICRT may prevent vaginal recurrence. The magnitude of benefit from adjuvant ICRT is difficult to define, however, because criteria for adding external pelvic irradiation to ICRT are different among physicians and because reported vaginal recurrence rates after hysterectomy alone range from 2% to 20%. Because patients with inner- or middle-third myometrial invasion and grade 1 or 2 disease have a 3% risk of pelvic and para-aortic nodal involvement,25 they may benefit from vaginal brachytherapy without EBRT, regardless of whether staging lymphadenectomy is done (Table 30-4).42,72,73,76,102-105 In a few retrospective studies, vaginal brachytherapy alone has produced excellent results for patients with uterus-confined disease with poor prognostic indicators, such as high grade and deep myometrial penetration, but negative lymph nodes (Table 30-5).42,72,103,105-112 However, the only prospective trial that compared vaginal brachytherapy with vaginal brachytherapy plus pelvic radiation therapy for such patients suggested a benefit from the addition of EBRT.32 All of the trials listed in Tables 30-4 and 30-5 were retrospective, with small numbers of patients and very short ­follow-up times. Alektiar and colleagues103 observed that EBRT has a much longer track record with many more patients treated than does ICRT and that the factors that predict pelvic or vaginal recurrence remain unknown. In their study of hysterectomy followed by high-dose-rate vaginal ­brachytherapy for stage IB, grade 3 or stage IIB disease, they found the use of “comprehensive” disease staging (which included pelvic washings and pelvic or para-aortic lymph node sampling), tumor grade, and depth of invasion to be predictors of pelvic control.103 Preliminary results from the PORTEC-2 trial, designed to compare vaginal brachytherapy with EBRT for high­intermediate risk endometrial cancer, were presented at

784

PART 9  Female Genital Tract

TABLE 30-4. Incidence of Pelvic Recurrence after Vaginal Radiation Therapy Alone in Patients with Favorable Prognostic Indicators Study and Reference

Number of Patients

Risk Factors

Median Follow-up (yr)

Recurrence Rate (%)

Kucera et al76

376

Middle-third MMI, G1 Inner-third MMI, G1-3

Not stated Not stated

0.8 0.8

Eltabbakh et al102

303

Stage IB, G1-2

8.1

0

307

Stage IB, G1-2

2

2.6

Inner-third MMI, G1-3

Not stated

0

No Lymphadenectomy

Alektiar et

al103*

Lymphadenectomy Morrow et al42

49

Orr et al72

126

Stage IB, G1-2 Stage IC, G1

3.25

0

Mohan et al73

131

Stage IB, G1

8

0

Jolly et al104†

50

Stage IA, G1-2 Stage IB, G1-3 Stage IC, G1-2

3.2

4

Solhjem et al105

33

Stage IB, G1-2

2

0

*Forty

(13%) of 307 patients had comprehensive staging. lymphadenopathy only. G, histologic grade; MMI, myometrial invasion.

†Selective

TABLE 30-5. Incidence of Pelvic Recurrence after Vaginal Brachytherapy Alone after Staging Lymphadenectomy for Patients with Unfavorable Prognostic Indicators Study and Reference Morrow et al42 Fanning et

al106

Orr et al72 MacLeod et al107 Rittenberg et Ng et

al108

al109

Horowitz et

al110

al111

Risk Factors

Median Follow-up (yr)

Local Recurrence Rate (%)*

14

Inner-third MMI or G3

>3

0

18

Stage IC, G3, stage II

2.8

0

60

Stage IC or G3

Not stated

0

34

Stage IC or G3, stage II

6.9

3

12

Stage II

3

0

15

Stage II

3

0

133

Stage IC or G3 or stage II

5.5

4.5

Number of Patients

38

Stage IB, G3 or stage IC

2.5

0

Ng et al112

77

Stage IB, G3 or stage IC

3

10

Alektiar et al103

78

Stage IB, G3 or stage IC or II

3

8

Solhjem et al105

65

Stage IC or G3

2

0

Chadha et

G, histologic grade; MMI, myometrial invasion. *Local recurrence includes all recurrences in the pelvis or anywhere in the vagina but not distant metastasis, including para-aortic nodes

the American Society of Clinical Oncology meeting in May 2008.113 This trial, building on the previous PORTEC trial, randomly assigned patients with uterine-confined endometrial cancer to receive adjuvant EBRT (46 Gy in 23 fractions) or adjuvant vaginal brachytherapy (21 Gy in 3 high-dose-rate fractions or 30 Gy delivered by a low-doserate source to 0.5 cm depth). Eligible patients were women older than 60 years with stage IC, grade 1 disease; stage IC, grade 2 disease; or stage IB, grade 3 disease. Women of any age with stage IIA, grade 1 disease; stage IIA, grade 2 disease;

or stage IIA, grade 3 (<50% myometrial invasion) disease were also included. These criteria were designed to focus on the higher-risk spectrum of patients from PORTEC-1 by excluding all patients with stage IB, grade 2 disease and many patients who were younger than 60 years. As was true of the first PORTEC trial,99,100 PORTEC-2 did not include patients with stage IC, grade 3 disease. At 3 years of follow-up, the actuarial vaginal failure rates in the PORTEC-2 trial were 2.0% for those given EBRT and 0.9% for those given vaginal brachytherapy (P = .97).

30  The Endometrium

The vaginal brachytherapy group was more likely to experience pelvic relapse (3.6% versus 0.7% for the EBRT group, P = .03), but the absolute difference was small and was not associated with a difference in overall survival. The authors of that report concluded that vaginal brachytherapy was effective for preventing vaginal relapses, produced minimal treatment-related side effects, and was associated with better quality of life than EBRT. They further concluded that vaginal brachytherapy should be considered the treatment of choice for patients with uterine-confined endometrial cancer (excluding those with stage IC, grade 3 disease) deemed at sufficient risk to warrant adjuvant therapy.113 The lack of difference in survival in PORTEC-2 raises the question of whether adjuvant therapy is needed at all for patients with intermediate-risk disease. In a retrospective review, Obermair and colleagues114 compared the effects of surgery alone with those of surgery plus vaginal cuff irradiation for node-negative, intermediate-risk endometrial cancer. Disease in all 757 patients had been surgically staged. At a mean follow-up time of 38 months, the investigators’ analysis led them to conclude that postoperative vaginal brachytherapy did not reduce the risk of disease recurrence, including vaginal relapse, compared with surgery alone, even when cases were stratified by disease substage and grade.114 Another trial presented at the American Society of Clinical Oncology meeting in May 2007115 also questioned the benefit of radiation therapy (in this case, pelvic radiation therapy) for intermediate-risk disease. The United Kingdom Medical Research Council’s ASTEC trial and the National Cancer Institute of Canada’s EN.5 trial randomly assigned 905 patients with stage IA-IIA, grade 3 disease; stage IC, grade 1-2 disease; or stage IIA, grade 1-2 disease to undergo pelvic radiation therapy or observation after surgery. Most of the patients in these trials did not undergo full surgical staging, and brachytherapy was allowed at the discretion of the investigator. The 5-year overall survival rates were identical in both treatment groups (84%), as were the 5-year diseasespecific survival rates (89%).115 The cumulative incidence of vaginal recurrence favored pelvic radiation therapy (4% versus 7%), but vaginal recurrences represented only 42 of 123 recurrences. The authors of this analysis ­concluded that pelvic radiation therapy had no effect on overall survival or disease-specific survival rates but was associated with greater toxicity.115 These results have yet to be ­published in full.

Postoperative Radiation Therapy for High-Risk ­Disease: Stage IC, Grade 3 or Stage IIB to IIIC In 2004, Creutzberg and colleagues100 reported their findings for 104 patients with stage IC, grade 3 tumors who had registered but were not included in the first PORTEC trial, comparing their survival with that of the patients included in that trial. Grade 3 tumors were found to have a much higher metastasis rate than grade 1 or 2 tumors; multivariate analysis showed that histologic grade 3 was the most important adverse prognostic factor for relapse and death from endometrial cancer. These findings raise the question of whether radiation alone is enough for grade 3 tumors or whether the better strategy may be a combination of radiation and chemotherapy. Studies conducted in attempts to answer this question are described in the ­following paragraphs.

785

Radiation versus Chemoradiation Therapy. Two small phase II studies have shown the feasibility of adding concurrent chemotherapy to pelvic radiation followed by three or four courses of consolidation chemotherapy.116,117 Local control rates with this strategy were excellent in both of these studies, and the PORTEC-3 trial is currently ­accruing patients with high-risk factors to participate in a randomized trial comparing pelvic radiation alone with pelvic radiation plus concurrent chemotherapy followed by more chemotherapy. Early results of a study done by the Nordic Society of Gynecologic Oncology and the European Organisation for Research and Treatment of Cancer were presented at the American Society of Clinical Oncology meeting in May 2007.118 The NSGO-EC-9501/EORTC-55991 trial randomly assigned 382 patients between 1996 and 2007 to receive adjuvant radiation therapy (n = 196) or radiation therapy plus chemotherapy (n = 186). Patients with surgical stage I, II, IIIA (positive peritoneal fluid cytology only), or IIIC (positive pelvic lymph nodes only) disease were eligible if they were considered to be at sufficiently high risk for micrometastatic disease to qualify for adjuvant therapy. Patients with serous, clear cell, or anaplastic carcinomas were eligible regardless of risk factors. Patients with para-aortic metastases were not eligible. Lymph node exploration at surgical staging was optional. All patients ­underwent at least a total abdominal hysterectomy with bilateral salpingo-oophorectomy. Pelvic radiation therapy, with or without vaginal brachytherapy, was given to a dose of 44 Gy. Chemotherapy was given before or after the radiation. Before August 2004, the chemotherapy regimen consisted of four courses of cisplatin and doxorubicin or epirubicin; after that time, several chemotherapy regimens were allowed (e.g., cisplatin and doxorubicin or epirubicin; paclitaxel, epirubicin, and carboplatin; and paclitaxel and carboplatin). The median follow-up time was 4.3 years. The hazard ratio for progression-free survival in the study was 0.62 (95% CI, 0.40 to 0.97; P = .03), which translated to an estimated difference in 5-year progression-free survival rate of 7% (from 72% for those given radiation to 79% for those ­given chemoradiation therapy).118 The corresponding findings for overall survival were similar; the hazard ratio was 0.65 (95% CI, 0.40 to 1.06; P = .08), for an estimated difference in 5-year overall survival rate of 8% (74% for radiation versus 82% for chemoradiation ­ therapy). The investigators concluded that even though 27% of the patients assigned to the chemoradiation therapy group ­received little or none of the prescribed chemotherapy, chemoradiation therapy was still better than radiation alone for patients with early endometrial cancer at high risk for micrometastasis.118 Radiation versus Chemotherapy. Two other large studies compared radiation therapy with chemotherapy for high-risk endometrial carcinoma. In a study reported from Italy,119 the 345 patients had histologically confirmed endometrioid, adenoacanthoma, or adenosquamous carcinoma of FIGO stage IC, grade 3; stage IIA-B, grade 3 with deep myometrial invasion (50% or more); or stage III. ­ Radiation therapy consisted of whole-pelvis radiation or extended-field radiation for patients with positive para-­aortic nodes to a total dose of 45 to 50 Gy. The chemotherapy consisted of five cycles of cyclophosphamide,

786

PART 9  Female Genital Tract

doxorubicin, and cisplatin. Sixty-four percent of the patients had stage III disease. At a median follow-up time of 95.5 months, no differences were found in overall survival or progression-free survival between the two treatment groups. Radiation therapy seemed to delay local relapses and chemotherapy seemed to delay metastases, but these apparent effects were not statistically significant. The authors of this report proposed that improvements in therapy for stage III disease may require combinations of radiation therapy and chemotherapy.119 The second study, from the Japanese Gynecological Oncology Group,120 randomly assigned 385 patients with intermediate- to high-risk endometrial cancer to receive pelvic radiation therapy or chemotherapy. The radiation was given with anterior and posterior fields to a total dose of 45 to 50 Gy. The chemotherapy consisted of three or more cycles of cyclophosphamide, doxorubicin, and cisplatin. In terms of stage distribution, about 60% of patients had stage IC disease, and only 12% had stage III disease; in terms of histologic grade, 55% had grade 1 tumors, 30% had grade 2 tumors, and 15% had grade 3 tumors. No differences were found in progression-free survival rates (83.5% for the ­radiation group versus 81.8% for the chemotherapy group) or overall survival rates (85.3% for radiation versus 86.7% for chemotherapy).120 The rate of intrapelvic failure was 7% in both groups, and toxicities also were similar in both groups. The authors of this report concluded that adjuvant chemotherapy might be a useful alternative to radiation therapy for intermediate-risk endometrial carcinoma.120 The Italian119 and Japanese120 studies had some shortcomings that need to be considered when evaluating their findings. The patient populations in both trials were quite heterogenous, and both had large numbers of patients with low-risk disease. Neither study was designed to show noninferiority, which means that their findings should be considered inconclusive.121 The chemotherapy regimen used in both trials is no longer considered the standard treatment for endometrial carcinoma, and the doses were ­lower than those used in other studies. Results of the Italian trial119 indicated that radiation conferred an advantage in local control, but this effect was not seen in the ­Japanese trial.120 Their finding of improved local control with radiation therapy and improved distant metastasis with ­chemotherapy led the Italian group to hypothesize that combination therapy might be the best treatment for patients with high-risk endometrial carcinoma.119 It is hoped that the results of the ongoing PORTEC-3 trial will help to answer some of the questions raised by these studies. However, the one fundamental question that still remains—whether radiation therapy is important in the treatment of endometrial carcinoma or whether surgery followed by chemotherapy alone is more effective— can be answered only by a large, randomized study with sufficient numbers of patients.

Postoperative Therapy for Stage III Disease Stage III tumors constitute a broad and heterogeneous group with respect to patterns of failure and treatment outcomes. The FIGO staging classification defines three stage III subsets: stage IIIA (involvement of serosa or adnexa or positive cytologic findings); stage IIIB (vaginal metastases), and stage IIIC (pelvic or para-aortic node ­ involvement). The implications of finding and treating extrauterine nodal

disease cannot be considered in isolation but rather should be considered in the context of the presence or absence of other sites of disease. Positive cytologic findings on their own do not affect survival or prognosis (see “Tumor Cells in Peritoneal Washings”). However, if such findings are combined with lymphovascular space invasion, nonendometrioid histology, or adnexal or uterine serosal involvement, prognosis and survival rates decline and the incidence of distant metastasis rises.122 Surgical staging reveals that approximately one half of patients with positive nodes will have only positive nodes, whereas the other half will have positive nodes plus disease at other sites. Mariani and colleagues,51 in a study of 46 patients with stage IIIC disease, found clear evidence of two distinct subsets in 5-year recurrence-free survival rates. For the 22 patients with only nodal involvement, the rate was 68%, but the rate for the 24 patients with positive nodes plus peritoneal, serosal, adnexal, or vaginal metastasis was only 25%. Patients with only positive nodes had a very low risk of distant metastasis, with only one patient having lung metastasis; however, the risk of distant metastasis was much higher in the other group, with nine patients having disease outside the nodal areas.51 McMeekin and colleagues,123 reporting results for 47 patients with positive nodes, found the 3-year survival rate to be 93% for patients with only positive nodes and 39% for patients with positive nodes plus positive cytology or adnexal involvement. Both research groups concluded that a separate subcategory of stage IIIC disease should be established based on adnexal or serosal involvement and positive cytologic results.123 Radiation. Controversy and questions remain concerning the best adjuvant treatment for stage III disease. Treatment of positive pelvic nodes with pelvic radiation therapy has led to excellent local control and survival rates in several studies.124-126 In one of those studies, Nelson and colleagues124 found a 5-year disease-free survival rate of 81% for patients with only positive pelvic nodes treated with pelvic radiation. Two patients did experience para-aortic recurrence, however, suggesting that extended-field therapy may be useful for at least some patients with positive pelvic nodes. Mundt and colleagues127 reported 4 patients who experienced para-aortic failure among 20 who were treated with pelvic radiation alone, and that research group also proposed that extended-field therapy might be helpful for patients with para-aortic node–negative disease. Tailored extended-field radiation therapy can be curative in 30% to 50% of cases in which positive para-aortic nodes have been completely resected (Table 30-6).42,49,50,52,128-131 Radiation versus Chemotherapy. The GOG published the results of a randomized study (GOG-122) ­comparing whole-abdomen radiation therapy with chemotherapy for patients with stage III or IV endometrial cancer who had up to 2-cm residual intra-abdominal disease after surgical staging.132 That study showed that chemotherapy significantly improved disease-free survival at 2 years (58% versus 46% for radiation, P < .01). However, the difference in disease-free survival rates at 5 years was much smaller, although still favoring the chemotherapy group (42% versus 38%). Pattern-of-failure analyses of the two treatment groups demonstrated that chemotherapy reduced the rates of distant failure from 18% to 10%, but rates of pelvic or abdominal failure were no different between the two ­treatment groups. More grade 3 or 4 adverse

30  The Endometrium

787

48

37%

Rose et al50

17

53%

with stage III disease are clearly at higher risk for distant metastasis, specifically patients with nodal involvement plus other factors such as lymphovascular space invasion, positive peritoneal cytology, and adnexal or serosal involvement. In the GOG-122 study, 50% of the patients in the chemotherapy group experienced disease relapse.132 The best treatment for stage III endometrial cancer remains unclear, and as Maggi and colleagues indicated,119 trials evaluating combinations of chemotherapy and radiation therapy are eagerly awaited. Until then, the choice of treatment for patients with stage III endometrial cancer must be made on an individual basis.

Corn et al52

50

46%

Postoperative Whole-Abdomen Radiation The poor prognosis of uterine papillary serous carcinomas and their tendency to behave like ovarian carcinomas have led radiation oncologists to explore the value of treating these lesions with whole-abdomen radiation therapy. Unfortunately, published reports of this approach for uterine papillary serous carcinoma are still largely anecdotal,135-138 probably because this is a rare disease that has been well recognized for only 15 to 20 years and because it tends to affect elderly women who may not be ideal candidates for aggressive treatment. Smith and colleagues138 reported a series of patients with papillary serous carcinoma treated with whole­abdomen radiation therapy. Three-year disease-free and overall survival rates for patients with stage I or II disease were both 87%, but among patients with stage III or IV disease, the disease-free survival rate was only 32% and the overall survival rate 61%.129 Subsequently, Martinez and colleagues139 published 10-year outcomes for 55 women with stage I to III adenocarcinoma, papillary serous carcinoma, or clear cell carcinoma treated with whole-abdomen radiation therapy. The overall survival rates for the entire group were 58% at 5 years and 43% at 10 years; the corresponding cause-specific survival rates were 77% and 70%. Multivariate analysis identified advancing age—but not tumor histology—as having prognostic significance.139 In 2006, the GOG reported results of a phase II trial140 evaluating whole-abdomen radiation for clinical stage I or II papillary serous or clear cell carcinoma. The 5-year ­progression-free survival rates were 38.1% for those with papillary serous carcinoma and 53.9% for those with clear cell carcinoma. More than one half of the treatment failures were within the radiation field. The conclusion from that study was that future trials would be necessary to evaluate other therapies for these rare subtypes of endometrial cancer, such as systemic chemotherapy, radiosensitizing chemotherapy, or sequential radiation and chemotherapy.140 Some investigators have recommended whole-­abdomen radiation for patients with FIGO stage III nonserous ­tumors, but studies of this approach in this setting have been retrospective and have included few patients. Smith and colleagues138 reported 3-year disease-free survival and overall survival rates of 79% and 89%, respectively, after whole-abdomen radiation therapy for patients with surgical stage III or IV endometrial adenocarcinomas. ­Martinez and colleagues139 reported 5-year survival rates for 43 ­patients with stage I or II endometrioid carcinoma and 89 patients with stage III disease, all of whom were treated with whole-abdomen radiation. At a median ­ follow-up time of 6.4 years, the 5-year cause-specific ­ survival rate

TABLE 30-6. Survival Rates for Patients Treated with Extended-Field Radiation for Endometrial Carcinoma Metastatic to Para-aortic Lymph Nodes Study and Reference Komaki et al128 Potish49 Morrow et

al42

al129

Number of Patients

5-Year Survival Rate

7

60%

15

40%

11

27%

Blythe et

al130

13

54%

Onda et

al131

20

75%*

Hicks et

*Patients

also received three cycles of cisplatin-containing chemotherapy before radiation therapy.

events were experienced in the chemotherapy group than in the ­radiation group, as well as more grade 5 events (eight ­treatment-related deaths in the chemotherapy group and five in the radiation therapy group).132 Considerable controversy surrounds the GOG-122 study. The heterogeneity of the study population was not taken into account; approximately 25% of the patients had aggressive histologic subtypes, including papillary serous carcinoma and clear cell carcinoma (which have a higher propensity for distant metastasis), and 14% of the patients had stage IV disease. The potential for the prescribed radiation dose to control abdominopelvic disease was compromised by the dose used and the amount of residual disease present: the dose of 45 Gy to the pelvis would be ­expected to control 2-cm residual disease in only 30% of cases, and the dose of 30 Gy to the upper abdomen would only ­control about 10% of such cases. Chemotherapy has some activity against metastatic endometrial cancer. However, adjuvant chemotherapy alone is associated with fairly high pelvic failure rates. In one study, Mundt and colleagues133 reported a 3-year pelvic failure rate of 47%, with 31% of patients experiencing pelvic relapse as the first and only site of failure. A retrospective review from The University of Texas M. D. Anderson Cancer Center134 of 53 patients with stage IIIC disease (35 given radiation therapy and 18 chemotherapy only) indicated that radiation produced better 5-year overall survival rates (75% versus 42% for chemotherapy) and pelvic relapse–free survival rates (97% versus 60% for ­chemotherapy). Many clinicians do not recommend any adjuvant treatment for patients with grade 1 or grade 2 disease when abnormal peritoneal cytologic findings are the only evidence of extrauterine disease. Moreover, the prognosis for patients with isolated adnexal involvement and even positive lymph nodes (including para-aortic nodes) is favorable for those treated with radiation therapy. Presumably, improvements in radiation therapy techniques that allow higher doses to be given without toxicity may further improve ­ local control. Adding chemotherapy to radiation therapy may improve survival still further.131 However, some ­ patients

788

PART 9  Female Genital Tract

A

B

FIGURE 30-4.  Typical anterior field (A) and lateral field (B) for carcinoma of the endometrium. The rectum is outlined in yellow, and the vagina is outlined on a blue simulation from computed tomography.

was 83% for those with stage I or II disease and 73% for those with stage III disease.139 In the largest study of its kind, the GOG conducted a phase II study of wholeabdomen radiation therapy for 180 patients with stage III to IV endometrial carcinoma, 103 with high-risk histology (papillary serous carcinoma or clear cell carcinoma), and 77 with “typical” endometrial carcinoma of other histologic subtypes.141 The recurrence-free rate at 3 years was 11% for the patients with typical endometrial cancer; all patients with postoperative gross residual disease (up to 2 cm) died of their disease. These findings indicated that whole-abdomen radiation therapy, as delivered in this study, offered curative therapy for only a minority of patients with stage III or stage IV endometrial cancer, regardless of histologic subtype, and that new techniques and therapies are needed for endometrial cancer at these stages.141 The results of the GOG-122 trial confirmed these findings.132

Postoperative Radiation Therapy Techniques Pelvic Irradiation. Pelvic irradiation can be delivered by using a four-field technique (Fig. 30-4) or anterior and posterior opposed fields with 15- to 20-MV photons. Four fields are usually preferred for patients who have ­undergone an extensive lymphadenectomy or for those with wound problems. Serious complications are observed in 5% to 15% of all patients who undergo postoperative pelvic irradiation; however, the risk of complications may be greater for ­patients who have had a staging lymphadenectomy.80,142,143 The most significant side effects observed are small-bowel obstruction, fistula formation, proctitis, chronic diarrhea, vaginal stenosis, and insufficiency fractures of bone. The lateral fields should usually receive less weighting than the anterior and posterior fields (e.g., 50:50:40:40, respectively, for the anteroposterior, posteroanterior, right lateral, and left lateral fields) to reduce the lateral applied dose, and thus computerized planning should be used.

Pelvic treatment fields typically extend from the L4 or L5 lumbar vertebra to the middle or lower pubic ramus; a ­radiopaque marker is used to verify the presence of at least a 4- to 5-cm margin on the vaginal cuff (see Fig. 30-4). Prone ­ positioning, bladder distention, and use of a false tabletop have been used to reduce the volume of the small bowel within the pelvis. The total treatment dose is usually 40 to 50 Gy at 1.8 to 2.0 Gy/fraction. Extracapsular nodal extension or close margins in patients with cervical disease may warrant an external beam boost. Possible indications for vaginal vault irradiation are discussed under “Vaginal Brachytherapy.” Intensity-Modulated Radiation Therapy. The use of intensity-modulated radiation therapy (IMRT) in the postoperative setting can reduce the volume of small bowel receiving the prescription dose and reduce the overall dose to the bowel and rectum.144-146 A study conducted at the University of Chicago showed that use of whole-pelvis IMRT led to significant reductions in acute147 and chronic148 gastrointestinal toxicities compared with standard treatment. However, only one study from the University of Pittsburg149 has reported clinical outcomes after adjuvant IMRT for endometrial cancer. At a median follow-up time of 20 months for the 47 patients given IMRT, the local control rate was 100%, and the overall survival rate was 90%; the 3-year actuarial rate of grade 2 or higher toxicity was 3.3%.149 A phase II trial of IMRT as adjuvant treatment for patients with endometrial cancer being conducted by the Radiation Therapy Oncology Group is expected to provide more information on the efficacy of this approach. Extended-Field Radiation Therapy. Postoperative extended-field radiation therapy is indicated for patients with nonserous endometrial cancer with the only evidence of extrapelvic disease being in the para-aortic nodes. Fiveyear survival rates of 30% to 50% have been reported with this treatment (see Table 30-6).42,49,50,52,128-131 The upper border of the field should be at the T11 or T12 vertebra to

30  The Endometrium

ensure coverage of perirenal nodes, which are at risk for direct spread through the tubo-ovarian vessels. Treatment should be delivered with photons of at least 15 MV whenever possible. Four fields may be used, but CT planning should be done to verify that the dose to the kidneys, spinal cord, and para-aortic nodes is below that known to cause complications. The entire field probably cannot be treated with more than 45 to 50.4 Gy at 1.8 Gy/fraction; however, small boost fields may be used to increase the dose to residual gross disease or sites of extracapsular spread to 55 to 66 Gy, depending on the volume. Whole-Abdomen Irradiation.  Whole-abdomen radia­ tion therapy is usually given using anteroposterior-posteroanterior fields. The photon energy should be based on the thickness of the subcutaneous abdominal tissue. The upper border of the field should be designed under fluoroscopy to give a 1-cm margin on the upper excursion of the diaphragm under quiet respiration. CT is helpful for treatment planning and is used to verify the position of the kidneys and the borders of the peritoneal cavity. CT studies have shown that the pelvic peritoneum sometimes extends lateral to the anterior iliac crest, and use of standard anatomic landmarks therefore can result in underdosage of portions of the peritoneum. Doses of 25 to 30 Gy at 1.5 Gy/fraction are usually tolerated if the patient is treated appropriately with antiemetics and blood counts are followed closely. Posterior two to five halfvalue layer blocks are most often used to restrict the dose to the kidneys to less than 18 Gy, and liver blocks are most often used to shield the right lobe of the liver after about 25 Gy. Most investigators recommend a boost to the pelvis to bring the total dose to 45 to 50 Gy. Some investigators also boost the dose to the para-aortic nodes and the medial edges of the diaphragm to approximately 42 Gy. Vaginal Brachytherapy.  In the delivery of vaginal ICRT, opinions differ regarding the ideal dose, dose rate, target volume (i.e., vaginal length), and type of intracavitary applicators. It is thought that treating the entire vaginal length is unnecessary and may result in excess vaginal dysfunction or rectal morbidity. Most clinicians therefore treat only the upper third to half of the vagina. Colpostats or cylinders can be used to treat the vaginal mucosa, and radiation can be administered at low or high dose rates. If EBRT is not given, the typical dose delivered with lowdose-rate ICRT is 60 to 70 Gy to the vaginal surface over a period of about 72 hours. Some clinicians give additional ICRT to the vaginal cuff after pelvic radiation, although others have failed to document improved outcomes with this approach.30,150 A typical dose for prophylactic ICRT of the cuff after pelvic radiation is 30 Gy in 48 hours (at low dose rate) or 15 to 18 Gy in 3 fractions (at high dose rate) prescribed to the vaginal surface. High-dose-rate ICRT has some obvious advantages over low-dose-rate ICRT for prophylactic irradiation of the vaginal cuff. Placement of the applicator is simple, requires no sedation, and can be done as an outpatient procedure. The standard geometry of the applicators permits rapid calculation of source dwell times. The use of high-doserate ICRT is also less controversial in this setting because only a modest dose of radiation is needed to prevent tumor recurrence in the vagina. However, for most applicator designs, the dose and fraction size at 0.5 cm from the vaginal surface is approximately equivalent to the maximum rectal dose. Dose specification for low- or high-dose-rate delivery

789

should ­always include a calculation of the maximum vaginal surface dose. Sorbe and Smeds,151 in a prospective study comparing four fractionation schemes, found that the rate of rectal complications exceeded 80% in patients ­ treated with 4 fractions of 9 Gy prescribed at 10 mm. Several fractionation schemes have been recommended, including 6 fractions of 6 Gy152 or 2 fractions of 16.2 Gy154 to the vaginal surface and 3 fractions of 7Gy153 to the surface of the vaginal cylinder. A survey by the American Brachytherapy Society found that the most common fractionation scheme in the United States for vaginal brachytherapy without EBRT, used by 42% of the respondents, was 7 Gy in 3 fractions prescribed to 0.5 cm.155 All of these investigations have resulted in vaginal recurrence rates of less than 1.5% among patients with relatively favorable-risk tumors.

Radiation Therapy Alone for Medically or Surgically Inoperable Disease Even though hysterectomy is the primary treatment for early-stage endometrial cancer, patients with a very high risk of complications after surgery can be treated with radiation therapy alone. Disease-specific survival rates of 75% to 85% and local recurrence rates of 10% to 20% have been reported for patients with clinical stage I or II endometrial carcinoma treated with radiation therapy alone. Prognosis for these patients correlates with clinical stage and histologic grade.156-162 Brachytherapy alone yields excellent ­local control of most grade 1 tumors. Patients with large uterine cavities, cervical involvement, or grade 3 tumors should be considered for a course of EBRT before brachytherapy to treat regional nodes at risk. When brachytherapy is administered for endometrial carcinoma, it is important to cover the uterine fundus adequately. When the fundus is small, a single intrauterine ­tandem with increased activity at the tip should broaden the radiation dose distribution in the fundus. However, in patients with larger cavities, Heyman packing or Simon capsules163-166 can be placed in the uterus to achieve a better distribution to the fundus. Alternatively, a satisfactory dose distribution can sometimes be obtained by placing two tandems in the right and left cornua of the uterus and loading the tandems to achieve the desired distribution (Fig. 30-5). MRI may be helpful in planning treatment. With low-dose-rate radiation, uterine applicators are commonly loaded with 45 to 50 mg RaEq and left in place for two 72-hour or three 48-hour treatments (or for two 48-hour treatments after 40 to 45 Gy is delivered to the whole ­pelvis by EBRT). With this approach, reported control rates are high, and reported complication rates are low.158 The vagina does not require as high a dose as it does for cervical cancer unless the cervical stroma is involved. For a clinical stage I tumor, source loadings that yield a vaginal surface dose of 60 to 80 Gy should be sufficient to prevent vaginal recurrence in most patients. Because most patients with inoperable endometrial carcinoma have multiple medical problems, several groups have tested high-dose-rate ICRT for these patients.167-170 In one series, Kucera and colleagues167 used this approach to treat 228 patients with stage I endometrial carcinoma who had medical contraindications to surgery. The overall survival rates at 5 and 10 years were 59.7% and 30.2%, respectively, and the disease-specific survival rates at 5 and 10 years were 85.4% and 75.1%, respectively. ­Nguyen

790

PART 9  Female Genital Tract

FIGURE 30-5.  Radiograph shows the placement of two tandems in the uterus for the first intracavitary treatment of a patient with inoperable endometrial carcinoma. The loading of the tandem on the right is 10, 5, 10 mg RaEq of cesium, and the loading of the tandem on the left is 10, 10, 5, 5 mg RaEq of cesium. The implants were left in place for 72 hours.

and Petereit169 reported good uterine disease control rates (88%) but a high rate of perioperative morbidity and two perioperative deaths among 36 patients treated with highdose-rate brachytherapy for clinical stage I disease.169,171 Deep vein thrombosis, a possible complication of prolonged bed rest during low-dose-rate applications, is also a risk if patients receiving high-dose-rate therapy must remain in a lithotomy position.169,171 Petereit and colleagues172 suggested that perioperative morbidity could be reduced by using a tandem and a cylinder instead of a tandem and a pair of ovoids to reduce the time patients are in the lithotomy position; by using prophylaxis against deep vein thrombosis; and by using abdominal sonography to visualize the probe in the uterus.172 Whatever method of brachytherapy is used, patients with inoperable endometrial cancer should be treated only by a dedicated and experienced team in a well-controlled environment with adequate monitoring and emergency medical support.

Treatment of Recurrent Disease The most common site of vaginal recurrence of endometrial carcinoma is the region of the hysterectomy scar in the apical vagina. Recurrences also may occur under the urethra in the distal anterior vagina. Recurrences at other sites are uncommon. The mechanism of vaginal recurrence of endometrial carcinoma is not well understood, but at least two possible routes of spread are known: implantation into disturbed sites, principally in the vaginal apex, and spread

through the lymphovascular system. The first mechanism is ­ probably most consistent with the relatively high cure rates after ­ aggressive treatment of isolated vaginal recurrences and may explain the wide variation in vaginal recurrence rates reported by different institutions. Several case reports173,174 of episiotomy scar recurrences in women with cervical cancer (adenocarcinoma or squamous cell carcinoma) who delivered babies vaginally suggests that implantation is a potential mechanism of gynecologic tumor recurrence. For patients with isolated pelvic recurrence whose initial treatment was hysterectomy alone, 5-year local control rates of 40% to 80% and 5-year survival rates of 30% to 80% can be achieved with radical radiation therapy (Table 30-7).95,175-186 Survival correlates with the size, grade, extent, and location of the recurrent tumor; initial myometrial involvement; and the time to recurrence. Greven and Olds175 found that the size of the recurrence was the most important prognostic factor for local control. In their study, local control was achieved in 6 of 7 patients with tumor bulk less than 2 cm but in only 2 of 10 patients with tumor bulk of 2 cm or more.175 If the size and extent of recurrent disease are prognostic for outcome, this would suggest a need for careful follow-up of patients treated with surgery alone. In a more recent analysis, Jhingran and colleagues186 found that the type of treatment (i.e., EBRT alone versus EBRT plus brachytherapy) was a significant predictor of local control in a multivariate analysis, with patients ­given the combination treatment faring better than patients ­given only EBRT. In the treatment of recurrent disease, a combination of EBRT and brachytherapy is usually indicated; brachytherapy should be tailored to the size, extent, and initial response of the lesion. As is true for primary vaginal cancer, ICRT should not be used to treat endometrial tumors more than 5 mm thick. A variety of interstitial techniques may be indicated, depending on the clinical situation. For apical lesions, laparotomy or laparoscopic guidance may be necessary. Combined EBRT and interstitial radiation therapy is usually indicated for distal (usually suburethral) recurrences. The dose is important. Curran and colleagues176 reported that patients who received at least 60 Gy had significantly improved survival and pelvic control rates, and in a univariate analysis, Jhingran and colleagues186 found that patients who received 80 Gy or more had better local control rates.

Chemotherapy for Advanced, Recurrent, or Metastatic Endometrial Carcinoma Advanced, recurrent, or metastatic endometrial carcinomas are rare and rarely curable. In a review and meta-analysis that included information from the Cochrane Gynaecological Cancer Collaborative Review Group’s Trial Register and the Cochrane Central Register of Controlled Trials,187 the investigators concluded that the most active drugs in terms of tumor response were cisplatin, carboplatin, anthracyclines, and taxanes. The response rates varied among the trials, with the worst rates for the most heavily pretreated patients.188,189 In several trials, combination chemotherapy was more effective than single-agent chemotherapy, but for the most part, the improved response rates did not translate into improved progression-free or overall survival.190-192

30  The Endometrium

791

TABLE 30-7.  Survival Rates after Radiation Therapy for Pelvic Recurrences of Endometrial Carcinoma Study and Reference Aalders et Sears et

al185*

al177*

Curran et

al176* Olds175*

Time of Observation or Median Follow-up (yr)

Survival Rate (%)

3-19

14

7.5

44

5

31

3-10

33

Ingersol184*

5

48

Kuten et al183

5

18

1.25

35†

5

44

5

33

Greven and

Hart et

al178

Hoekstra et al181 Reddy et

al95 al179

5

43

Wylie et al182

5

53

Mandell et al180*

4

82

al186*

5

43

Ackerman et

Jhingran et *Vaginal †At

recurrences only. 5 years.

In the only trial that showed any such improvement, that ­ improvement came at the cost of higher toxicity.193 That trial compared doxorubicin, cisplatin, and paclitaxel with doxorubicin and cisplatin. In an attempt to reduce the toxicity of such treatments and clarify whether triplet ­ combinations that include paclitaxel are better than double combinations that include paclitaxel, the GOG is conducting a phase III trial to compare carboplatin and paclitaxel with doxorubicin, cisplatin, and paclitaxel.193 The ­Cochrane review suggested that more intense chemotherapy extended the median progression-free ­survival time (and possibly overall survival time) by about 1 month.187 However, given the toxicity of such treatment, it is questionable whether this gain translates into improved ­symptom control, quality of life, progression-free survival, or overall survival compared with other measures such as best supportive care.187

Uterine Sarcoma Surgery is the primary treatment for all sarcomas arising from the uterus. Total abdominal hysterectomy with bilateral salpingo-oophorectomy is the treatment of choice, although some investigators advocate a more radical surgical approach for low-grade endometrial stromal sarcomas.194 Lymph node dissection and sampling add information about prognosis but have not improved survival rates. Local failures are common, and most recurrences result from hematogenous dissemination of tumor. The role of radiation therapy in the management of carcinosarcoma is controversial. Most retrospective reviews of patients with carcinosarcoma have demonstrated a lower rate of pelvic tumor recurrence in patients who received postoperative pelvic irradiation; however, in most cases, this has not led to an improvement in survival.195-206 Potential biases from the small numbers of patients in most studies have made it impossible to determine the efficacy of treatment. The GOG retrospectively evaluated its

experience in treating uterine cancer in two studies.197,198 In both studies, the rate of pelvic recurrence was reduced with radiation therapy; however, more cases of distant metastasis occurred in the irradiated group, and overall survival was therefore the same in both groups. In another retrospective study by Callister and colleagues,201 pelvic radiation therapy significantly reduced the risk of pelvic recurrence and might have delayed the appearance of distant metastases after surgery; however, receipt of radiation therapy did not change overall survival. Only one retrospective study has shown improved survival from radiation therapy for uterine sarcoma.207 That study, a review of 2677 women with all types of uterine sarcoma, including leiomyosarcoma, carcinosarcoma, high-grade endometrial stromal sarcoma, adenosarcoma, and sarcoma not otherwise specified in the Surveillance, Epidemiology, and End Results database, found that adjuvant radiation therapy improved survival of women with stage II to stage IV disease. The problem with these studies, however, is that the tumor stage, grade, and depth of invasion and rates of lymph node involvement were not compared for patients receiving different treatments, making it impossible to determine the efficacy of radiation therapy. A phase III randomized study conducted by the GOG evaluated whole-abdomen radiation therapy versus combination chemotherapy with ifosfamide-mesna plus cisplatin for optimally debulked stage I to IV carcinosarcoma of the uterus.208 Preliminary results showed that adjuvant chemotherapy reduced the recurrence rate and prolonged overall survival relative to whole-abdomen radiation; however, the risk of recurrence within 5 years in the chemotherapy group was still 49%. The authors of this report concluded that the high rates of failure and death in both treatment groups indicate that the need for new adjuvant therapies remains. Few studies have separately reported control rates for patients given radiation therapy for leiomyosarcoma, and the numbers of patients are too small to allow treatment

792

PART 9  Female Genital Tract

G1, no invasion

No XRT

G1–2, < 50%, G3 no invasion

V

G2, no invasion

?V

G2, ≥ 50%, G3, < 50%

V±P

G2–3, ≥ 50%

P±V

FIGURE 30-6.  Possible schema for postoperative ­ treatment based on the grade and depth of invasion for patients with stage I endometrial carcinoma. Other prognostic factors, ­including tumor size and lymph node status, and risk factors for major complications also should be considered in the selection of treatment. G, histologic grade; P, pelvic radiation therapy;  V, vaginal vault radiation therapy; XRT, radiation therapy.

efficacy to be evaluated.197,199,200 Patients with close surgical margins may benefit from postoperative radiation. Because lymph node metastasis from leiomyosarcoma is uncommon, treatment of the operative bed (usually the lower pelvis) may be sufficient in most cases. Several investigators reporting small series of patients with endometrial stromal tumors have commented on the role of radiation therapy, with some finding it effective209-211 and others not.212 However, the reports of radiation ­therapy for these tumors are little more than anecdotal, and further study is needed to adequately define the role of radiation therapy in this setting.

Summary Adenocarcinoma of the endometrium occurs predominantly in postmenopausal women, for whom vaginal bleeding is a notable symptom. For this reason, the diagnosis is often made relatively early in the course of the disease, and disease control rates are good. Survival rates of 85% to 90% can be expected for patients with clinical stage I disease. Survival rates for patients with stage II or III disease are significantly lower, approximately 65% and 35%, respectively. In every stage of disease, poorly differentiated (grade 3) and deeply invasive tumors have a less favorable outlook. Regional lymph node metastases occur in a higher proportion of patients than previously suspected and are much more common in patients with poorly differentiated or advanced lesions involving the cervix or extending beyond the uterus. Overwhelming evidence supports the role of radiation for the treatment of endometrial carcinoma in terms of local control; however, little evidence is available to suggest that it improves survival. Radiation therapy does reduce the frequency of pelvic recurrences and may improve survival rates among patients with high-risk disease characteristics. A possible approach to selecting EBRT or ICRT for patients in various risk groups is illustrated in Figure 30-6. Patients with grade 1 or 2 disease with no myometrial invasion are unlikely to benefit from adjuvant radiation therapy. Patients with stage IA, grade 3 disease have a low incidence of lymph node involvement, and vaginal brachytherapy alone is probably adequate to ensure optimal local control and minimal sequelae. Patients with stage IB, grade 1 or 2 tumors probably do not need any form of radiation therapy, but patients with stage IB, grade 3 tumors probably

can benefit from vaginal brachytherapy alone. The patients most likely to benefit from pelvic radiation include those with high-grade disease or deep muscle invasion. This is true regardless of whether lymph node sampling has been performed. If patients have had extensive surgical staging or have other risk factors for serious bowel sequelae after pelvic radiation therapy, consideration should be given to treatment with vaginal applicators alone. Adding chemotherapy to individually tailored radiation therapy may improve survival for patients with endometrial carcinoma with high-risk features by improving local control and reducing distant metastasis while minimizing damage to normal tissues.

REFERENCES    1. American Cancer Society. Cancer Facts & Figures, 2008. ­Atlanta: American Cancer Society, 2008.    2. Parazzini F, LaVecchia C. Epidemiology of adenocarcinoma of the cervix. Gynecol Oncol 1990;39:40-46.    3. Antunes CM, Stolley PD, Rosenshein NB, et al. Endometrial cancer and estrogen use: report of a large case-control study. N Engl J Med 1979;300:9-13.    4. Rustgi AK. Hereditary gastrointestinal polyposis and nonpolyposis syndromes. N Engl J Med 1994;331:1694-1702.    5. Gruber SB. New developments in Lynch syndrome (hereditary nonpolyposis colorectal cancer) mismatch repair gene testing. Gastroenterology 2006;130:577-587.    6. Aarnio M, Sankila R, Pukkala E, et al. Cancer risk in mutation carriers of DNA-mismatch-repair genes. Int J Cancer 1999;81: 214-218.    7. Fajardo LF, Berthrong M, Anderson RE. Female reproductive ­organs. In Radiation Pathology. New York: Oxford University Press, 2001, pp 289-299.    8. Morice P, Thiam-Ba R, Castaigne D, et al. Fertility results after ovarian transposition for pelvic malignancies treated by external irradiation or brachytherapy. Hum Reprod 1998;13:660-663.    9. Mitchell MF, Gershenson DM, Soeters RP, et al. The long-term effects of radiation therapy on patients with ovarian dysgerminoma. Cancer 1991;67:1084-1090.   10. Dede JA, Plentl AA, Moore JG. Recurrent endometrial carcinoma. Surg Gynecol Obstet 1968;126:533-542.   11. Zaino RJ. Pathologic indicators of prognosis in endometrial ­adenocarcinoma. Selected aspects emphasizing the Gynecologic Oncology Group experience. Pathol Annu 1995;30:1-28.   12. Taylor RR, Zeller J, Lieberman RW, et al. An analysis of two versus three grades for endometrial carcinoma. Gynecol Oncol 1999;74:3-6.   13. Hendrickson M, Ross M, Eifel P, et al. Uterine papillary serous carcinoma. A highly malignant form of endometrial adenocarcinoma. Am J Surg Pathol 1982;6:93-108.   14. 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;14:30-38.   15. Greenwood SM, Wright DJ. Evaluation of the office endometrial biopsy in the detection of endometrial carcinoma and atypical hyperplasia. Cancer 1979;43:1474-1478.   16. Koss LG, Schreiber K, Moussouris H, et al. Endometrial carcinoma and its precursors: detection and screening. Clin Obstet Gynecol 1982;25:49-61.   17. Mannel RS, Berman ML, Walker JL, et al. Management of ­endometrial cancer with suspected cervical involvement. Obstet Gynecol 1990;75:1016-1022.   18. Rose PG, Sommers RM, Reale FR, et al. Serial serum CA-125 measurements for evaluation of recurrence in patients with ­endometrial carcinoma. Obstet Gynecol 1994;84:12-16.   19. Patsner B, Mann WJ. Use of serum CA-125 in monitoring ­patients with uterine sarcoma. A preliminary report. Cancer 1988;62:1355-1358.

30  The Endometrium   20. Frei KA, Kinkel K, Bonel HM, et al. Prediction of deep myometrial invasion in patients with endometrial cancer: clinical utility of contrast-enhanced MR imaging—a meta-analysis and Bayes­ ian analysis. Radiology 2000;216:444-449.   21. Hricak H, Stern JL, Fisher MR, et al. Endometrial carcinoma staging by MR imaging. Radiology 1987;162:297-305.   22. Seki H, Takano T, Sakai K. Value of dynamic MR imaging in ­assessing endometrial carcinoma involvement of the cervix. AJR Am J Roentgenol 2000;175:171-176.   23. Kinkel K, Kaji Y, Yu KK, et al. Radiologic staging in patients with endometrial cancer: a meta-analysis. Radiology 1999;212: 711-718.   24. Cowles TA, Magrina JF, Masterson BJ, et al. Comparison of ­ clinical and surgical staging in patients with endometrial ­carcinoma. Obstet Gynecol 1985;66:413-416.   25. Creasman W, Morrow C, Bundy B, et al. Surgical pathologic spread patterns of endometrial cancer. A Gynecologic Oncology Group Study. Cancer 1987;60:2035-2041.   26. International Federation of Gynecology and Obstetrics. Corpus cancer staging. Int J Gynaecol Obstet 1989;28:190.   27. Greene FL, Page DL, Fleming ID, et al (eds): AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 267–273.   28. Greven KM, Corn BW. Endometrial cancer. Curr Probl Cancer 1997;21:65-127.   29. Carey MS, O’Connell GJ, Johanson CR, et al. Good outcome ­associated with a standardized treatment protocol using selective postoperative radiation in patients with clinical stage I adenocarcinoma of the endometrium. Gynecol Oncol 1995;57:138-144.   30. Irwin C, Levin W, Fyles A, et al. The role of adjuvant radiotherapy in carcinoma of the endometrium—results in 550 patients with pathologic stage I disease. Gynecol Oncol 1998;70:247-254.   31. Jolly S, Vargas CE, Kumar T, et al. The impact of age on longterm outcome in patients with endometrial cancer treated with postoperative radiation. Gynecol Oncol 2006;103:87-93.   32. Aalders J, Abeler V, Kolstad P, et al. Postoperative external ­irradiation and prognostic parameters in stage I endometrial carcinoma. Obstet Gynecol 1980;56:419-427.   33. Rose PG. Endometrial carcinoma [published erratum appears in N Engl J Med 1997;336:1335]. N Engl J Med 1996;335: 640-649.   34. Wharton JT, Mikuta JJ, Mettlin C, et al. Risk factors and current management in carcinoma of the endometrium. Surg Gynecol Obstet 1986;162:515-520.   35. Sutton GP, Geisler HE, Stehman FB, et al. Features associated with survival and disease-free survival in early endometrial cancer. Am J Obstet Gynecol 1989;160:1385-1391.   36. Chambers SK, Kapp DS, Peschel RE, et al. Prognostic factors and sites of failure in FIGO stage I, grade 3 endometrial carcinoma. Gynecol Oncol 1987;27:180-188.   37. Lewis BV, Stallworthy JA, Cowdell R. Adenocarcinoma of the body of the uterus. J Obstet Gynaecol Br Commonw 1970;77:343-348.   38. Silverberg SG, Bolin MG, DeGiorgi LS. Adenoacanthoma and mixed adenosquamous carcinoma of the endometrium. A clinicopathologic study. Cancer 1972;30:1307-1314.   39. Ng AB, Reagan JW, Storaasli JP, et al. Mixed adenosquamous carcinoma of the endometrium. Am J Clin Pathol 1973;59:765-781.   40. Zaino RJ, Kurman R, Herbold D, et al. The significance of squamous differentiation in endometrial carcinoma. A Gynecologic Oncology Group Study. Cancer 1991;68:2293-2302.   41. Cirisano FD Jr, Robboy SJ, Dodge RK, 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.   42. Morrow C, Bundy B, Kurman R, et al. Relationship between surgical-pathological risk factors and outcome in clinical stage I and II carcinoma of the endometrium: a Gynecologic Oncology Group Study. Gynecol Oncol 1991;40:55-65.   43. Zaino RJ, Kurman RJ, Diana KL, et al. Pathologic models to predict outcome for women with endometrial adenocarcinoma: the importance of the distinction between surgical stage and clinical stage—a Gynecologic Oncology Group study [published erratum appears in Cancer 1997;79:422]. Cancer 1996;77:1115-1121.

793

  44. Silva EG, Jenkins R. Serous carcinoma in endometrial polyps. Mod Pathol 1990;3:120-128.   45. Fanning J, Alvarez PM, Tsukada Y, et al. Prognostic significance of the extent of cervical involvement by endometrial cancer. Gynecol Oncol 1991;40:46-47.   46. Schink JC, Rademaker AW, Miller DS, et al. Tumor size in ­endometrial cancer. Cancer 1991;67:2791-2794.   47. Mariani A, Webb MK, Keeney GL, et al. Low-risk corpus cancer: is lymphadenectomy or radiotherapy necessary? Am J Obstet Gynecol 2000;182:1506-1516.   48. Shah C, Johnson EB, Everett E, et al. Does size matter? Tumor size and morphology as predictors of nodal status and recurrence in endometrial cancer. Gynecol Oncol 2005;99:564-570.   49. Potish RA. Radiation therapy of periaortic node metastases in cancer of the uterine cervix and endometrium. Radiology 1987;165:567-570.   50. Rose P, Cha S, Tak W, et al. Radiation therapy for surgically proven para-aortic node metastasis in endometrial carcinoma. Int J Radiat Oncol Biol Phys 1992;24:229-233.   51. Mariani A, Webb MJ, Keeney GL, et al. Stage IIIC endometrial corpus cancer includes distinct subgroups. Gynecol Oncol 2002;87:112-117.   52. Corn BR, Lanciano K, Greven D, et al. Endometrial cancer with para-aortic adenopathy: patterns of failure and opportunities for cure. Int J Radiat Oncol Biol Phys 1992;24:223-227.   53. Turner DA, Gershenson DM, Atkinson N, et al. The prognostic significance of peritoneal cytology for stage I endometrial cancer. Obstet Gynecol 1989;74:775-780.   54. Milosevic MF, Dembo AJ, Thomas GM. The clinical significance of malignant peritoneal cytology in stage I endometrial carcinoma. Int J Gynecol Cancer 1992;2:225-235.   55. Grimshaw RN, Tupper C, Fraser RC, et al. Prognostic value of peritoneal cytology in endometrial carcinoma. Gynecol Oncol 1990;36:97-100.   56. Eifel P, Hendrickson M, Ross J, et al. Simultaneous presentation of carcinoma involving the ovary and the uterine corpus. Cancer 1982;50:163-170.   57. Pearl ML, Johnston CM, Frank TS, et al. Synchronous dual ­primary ovarian and endometrial carcinomas. Int J Gynaecol Obstet 1993;43:305-312.   58. Nascimento MC, Kelley A, Martitsch C, et al. Postoperative ­analgesic requirements—total laparoscopic hysterectomy ­versus vaginal hysterectomy. Aust NZ J Obstet Gynaecol 2005;45:140143.   59. O’Hanla KA, Huang GS, Lopez L, et al. Total laparoscopic hysterectomy for oncological indications with outcomes stratified by age. Gynecol Oncol 2004;95:196-203.   60. Obermair A, Manolitsas TP, Leung Y, et al. Total laparoscopic hysterectomy with total abdominal hysterectomy for obese women with endometrial cancer. Int J Gynecol Cancer 2005;15: 319-324.   61. Lenihan JP Jr, Kovanda C, Cammarano C. Comparison of ­laparoscopic-assisted vaginal hysterectomy with traditional hysterectomy for cost-effectiveness to employers. Am J Obstet Gynecol 2004;190:1714-1720.   62. Scribner DR Jr, Mannel RS, Walker JL, et al. Cost analysis of laparoscopy versus laparotomy for early endometrial cancer. Gynecol Oncol 1999;75:460-463.   63. Spirtos NM, Schlaerth JB, Gross GM, et al. Cost and quality of life analyses of surgery for early endometrial cancer: laparotomy versus laparoscopy. Am J Obstet Gynecol 1996;174:1795-1799.   64. Sculpher M, Manca A, Abbott J, et al. Cost effectiveness analysis of laparoscopic hysterectomy compared with standard hysterectomy: results from a randomized trial. BMJ 2004;328:134.   65. Obermair A, Manolitsas TP, Leung Y, et al. Total laparoscopic hysterectomy for endometrial cancer: patterns of recurrence and survival. Gynecol Oncol 2004;92:789-793.   66. Eltabbakh GH. Analysis of survival after laparoscopy in women with endometrial carcinoma. Cancer 2002;95:1894-901.   67. Walker JL, Piedmonte M, Spirtos NM, et al. Surgical staging of uterine cancer: randomized phase III trial of laparoscopy vs laparotomy—A gynecologic Oncology Group (GOG) Study: preliminary results [abstract]. J Clin Oncol 2006;24(18S):5010.

794

PART 9  Female Genital Tract

  68. Kornblith A, Walker J, Huang H, et al. Quality of life of patients in a randomized clinical trial of laparoscopy versus open laparotomy for the surgical resection and staging of uterine cancer: a Gynecologic Oncology Group study [abstract]. Gynecol Oncol 2006;101(Suppl 1):S47.   69. Janda M, Gebski V, Forder P, et al. Total laparoscopic versus open surgery for stage 1 endoetrial cancer: the LACE randomized controlled trial. Am J Obstet Gynecol 2005;193:1344-1352.   70. Girardi F, Petru E, Heydarfadai M, et al. Pelvic lymphadenectomy in the surgical treatment of endometrial cancer. Gynecol Oncol 1993;49:177-180.   71. Chuang L, Burke TW, Tornos C, et al. Staging laparotomy for endometrial carcinoma: assessment of retroperitoneal lymph nodes. Gynecol Oncol 1995;58:189-193.   72. Orr JW Jr, Holimon JL, Orr PF. Stage I corpus cancer: is teletherapy necessary? Am J Obstet Gynecol 1997;176:777-788.   73. Mohan DS, Samuels MA, Selim MA, et al. Long-term outcomes of therapeutic pelvic lymphadenectomy for stage I endometrial adenocarcinoma. Gynecol Oncol 1998;70:165-171.   74. Kilgore LC, Partridge EE, Alvarez RD, et al. Adenocarcinoma of the endometrium: survival comparisons of patients with and without pelvic node sampling. Gynecol Oncol 1995;56:29-33.   75. Piver MS, Hempling RE. A prospective trial of postoperative vaginal radium/cesium for grade 1-2 less than 50% myometrial invasion and pelvic radiation therapy for grade 3 or deep myometrial invasion in surgical stage I endometrial adenocarcinoma. Cancer 1990;66:1133-1138.   76. Kucera H, Vavra N, Weghaupt K. Benefit of external irradiation in pathologic stage I endometrial carcinoma: a prospective clinical trial of 605 patients who received postoperative vaginal irradiation and additional pelvic irradiation in the presence of unfavorable prognostic factors. Gynecol Oncol 1990;38: 99-104.   77. Chan JK, Cheung MK, Hug WK, et al. Therapeutic role of lymph node resection in endometrioid corpus cancer: a study of 12,333 patients. Cancer 2006;107:1823-1830.   78. Kitchener HC, Redman CW, Swart AMC, et al. ASTEC (surgery component): a study in the treatment of endometrial ­ cancer: a randomized trial of lymphadenectomy in the treatment of ­endometrial cancer [abstract]. Gynecol Oncol 2006;101(Suppl 1):S45.   79. Keys HM, Roberts JA, Brunetto VL, et al. A phase III trial of surgery with or without adjunctive external pelvic radiation therapy in intermediate risk endometrial adenocarcinoma: a Gynecologic Oncology Group study. Gynecol Oncol 2004;92:744-751.   80. Lewandowski G, Torrisi J, Potkul RK, et al. Hysterectomy with extended surgical staging and radiotherapy versus hysterectomy alone and radiotherapy in stage I endometrial cancer: a comparison of complication rates. Gynecol Oncol 1990;36:401-404.   81. Naumann RW, Higgins RV, Hall JB. The use of adjuvant radiation therapy by members of the Society of Gynecologic Oncologists. Gynecol Oncol 1999;75:4-9.   82. Javert C, Douglas R. Treatment of endometrial adenocarcinoma. Am J Roentgenol 1956;75:580.   83. Keller D, Kempson FL, Levine G, et al. Management of the ­patient with early endometrial carcinoma. Cancer 1974;33:1108-1116.   84. Maggino T, Romagnolo C, Landoni F, et al. An analysis of ­approaches to the management of endometrial cancer in North America: a CTF study. Gynecol Oncol 1998;68:274-279.   85. Partridge EE, Shingleton HM, Menck HR. The National Cancer Data Base report on endometrial cancer. J Surg Oncol 1996;61:111-123.   86. Society of Gynecologic Oncologists. Practice guidelines: uterine corpus—sarcomas. Society of Gynecologic Oncologists Clinical Practice Guidelines. Oncology (Williston Park) 1998;12: 284-286.   87. Jones HW. Treatment of adenocarcinoma of the endometrium. Obstet Gynecol Surv 1975;30:147-169.   88. Rutledge FN, Tan SK, Fletcher GH. Vaginal metastases from adenocarcinoma of the corpus uteri. Am J Obstet Gynecol 1958;75:167-174.   89. Lewis GC Jr, Mortel R, Slack NH. Adenocarcinoma of the body of the uterus. J Obstet Gynaecol Br Commw 1977;77:342-348.

  90. Eifel PJ, Ross J, Hendrickson M, et al. Adenocarcinoma of the endometrium. Analysis of 256 cases with disease limited to the uterine corpus: treatment options. Cancer 1983;52: 1026-1031.   91. Underwood PB, Lutz MH, Kreutner A, et al. Carcinoma of the endometrium: radiation followed immediately by operation. Am J Obstet Gynecol 1977;128:86-95.   92. Ohlsen JD, Johnson GH, Stewart JR, et al. Combined therapy for endometrial carcinoma: preoperative intracavitary irradiation followed promptly by hysterectomy. Cancer 1977;39: 659-664.   93. Bean HA, Bryant AJS, Carmichael JA, et al. Carcinoma of the endometrium in Saskatchewan: 1966 to 1971. Gynecol Oncol 1978;6:503-514.   94. Piver MS, Yazigi R, Blumenson L, et al. A prospective trial comparing hysterectomy, hysterectomy plus vaginal radium, and uterine radium plus hysterectomy in stage I endometrial carcinoma. Obstet Gynecol 1979;54:85-89.   95. Reddy S, Lee MS, Hendrickson FR. Pattern of recurrences in endometrial carcinoma and their management. Radiology 1979;133:737-740.   96. Sause WT, Fuller DB, Smith WG, et al. Analysis of preoperative intracavitary cesium application vs postoperative external beam radiation in stage I endometrial carcinoma. Int J Radiat Oncol Biol Phys 1990;18:1013-1019.   97. Grigsby PW, Perez CA, Kuten A, et al. Clinical stage I endometrial cancer: results of adjuvant irradiation and patterns of failure. Int J Radiat Oncol Biol Phys 1991;21:379-385.   98. Calais G, Vitu L, Descamps P, et al. Preoperative or postoperative brachytherapy for patients with endometrial carcinoma stage I and II. Int J Radiat Oncol Biol Phys 1990;19:523-527.   99. Creutzberg CL, van Putten WL, Koper PC, et al. Surgery and postoperative radiotherapy versus surgery alone for patients with stage-1 endometrial carcinoma: multicentre randomised trial. PORTEC Study Group. Post Operative Radiation Therapy in Endometrial Carcinoma. Lancet 2000;355:1404-1411. 100. Creutzberg CL, van Putten WL, Wárlám-Rodenhuis CC, et al. Outcome of high-risk stage IC, grade 3, compared with stage I endometrial carcinoma patients: the Postoperative ­Radiation Therapy in Endometrial Carcinoma trial. J Clin Oncol 2004;22:1234-1241. 101. Marchetti DL, Caglar H, Driscoll DL, et al. Pelvic radiation in stage I endometrial adenocarcinoma with high-risk attributes. Gynecol Oncol 1990;37:51-54. 102. Eltabbakh GH, Piver MS, Hempling RE, et al. Excellent longterm survival and absence of vaginal recurrences in 332 patients with low-risk stage I endometrial adenocarcinoma treated with hysterectomy and vaginal brachytherapy without formal staging lymph node sampling: report of a prospective trial. Int J Radiat Oncol Biol Phys 1997;38:373-380. 103. Alektiar KM, Venkatraman E, Chi DS, et al. Intravaginal brachytherapy alone for intermediate-risk endometrial cancer. Int J ­Radiat Oncol Biol Phys 2005;62:111-117. 104. Jolly S, Vargas C, Kumar T, et al. Vaginal brachytherapy alone: an alternative to adjuvant whole pelvis radiation for early stage endometrial cancer. Gynecol Oncol 2005;97:887-892. 105. Solhjem MC, Petersen IA, Haddock MG. Vaginal brachytherapy alone is sufficient adjuvant treatment of surgical stage I endometrial cancer. Int J Radiat Oncol Biol Phys 2005;62:1379-1384. 106. Fanning J, Nanavati PJ, Hilgers RD. Surgical staging and high dose rate brachytherapy for endometrial cancer: limiting ­external radiotherapy to node-positive tumors. Obstet Gynecol 1996;87:1041-1044. 107. MacLeod C, Fowler A, Duval P, et al. High-dose-rate brachytherapy alone post-hysterectomy for endometrial cancer. Int J Radiat Oncol Biol Phys 1998;42:1033-1039. 108. Rittenberg PVC, Lotocki RJ, Heywood MS, et al. Stage II ­endometrial carcinoma: limiting post-operative radiotherapy to the vaginal vault in node-negative tumors. Gynecol Oncol 2005;98:434-438. 109. Ng TY, Nicklin JL, Perrin LC, et al. Postoperative vaginal vault brachytherapy for node-negative stage II (occult) endometrial carcinoma. Gynecol Oncol 2001;81:193-195.

30  The Endometrium 110. Horowitz NS, Peters WA, Smith MR, et al. Adjuvant high dose rate vaginal brachytherapy as treatment of stage I and II endometrial carcinoma. Obstet Gyecol 2002;99:235-240. 111. Chadha M, Nanavati PJ, Liu P, et al. Patterns of failure in ­endometrial carcinoma stage IB grade 3 and IC patients treated with postoperative vaginal vault brachytherapy. Gynecol Oncol 1999;75:103-107. 112. Ng TY, Perrin LC, Nicklin JL, et al. Local recurrence in highrisk node-negative stage I endometrial carcinoma treated with postoperative vaginal vault brachytherapy. Gynecol Oncol 2000;79:490-494. 113. Nout RA, Putter J, Jurgenliemk-Schulz IM, et al. Vaginal brachytherapy versus external beam pelvic radiotherapy for high-risk intermediate risk endometrial cancer: results of the randomized PORTEC-2 trial [abstract]. J Clin Oncol 2008;26(Suppl): 1010S. 114. Obermair A, Cheuk R, Pak SC, et al. Disease-free survival after vaginal vault brachytherapy versus observation for patients with node-negative intermediate-risk endometrial adenocarcinoma. Gynecol Oncol 2008;110:280-285. 115. Orton J. Blake P on behalf of ASTEC/EN.5 collaborators. ­Adjuvant external beam radiotherapy (EBRT) in the treatment of endometrial cancer: results of the randomized MRC ASTEC and NCIC CTG EN.5 Trial [abstract 5504]. J Clin Oncol 2007;25(Suppl):275s. 116. Greven K, Winter K, Underhill K, et al. Final analysis of RTOG 9708: adjuvant postoperative irradiation combined with ­cisplatin/paclitaxel chemotherapy following surgery for patients with high-risk endometrial cancer. Gynecol Oncol 2006;103: 155-1559. 117. Mangili G, De Marzi P, Beatrice S, et al. Paclitaxel and concomitant radiotherapy in high-risk endometrial cancer patients: preliminary findings. BMC Cancer 2006;6:198. 118. Hogberg T, Rosenberg P, Kristensen G, et al. A randomized phase-III study of adjuvant treatment with radiation (RT) ± chemotherapy (CT) in early-stage high-risk endometrial cancer (NSGO-EC-9501/EORTC 55991) [abstract 5503]. J Clin Oncol 2007;25(Suppl):274s. 119. Maggi R, Lissoni A, Spina F, et al. Adjuvant chemotherapy vs radiotherapy in high-risk endometrial carcinoma: results of a randomized trial. Br J Cancer 2006;95:266-271. 120. Susumu N, Sagae S, Udagawa Y, et al. Randomized phase III trial of pelvic radiotherapy versus cisplatin-based combined chemotherapy in patients with intermediate- and high-risk endometrial cancer. A Japanese Gynecologic Oncology Group Study. Gynecol Oncol 2008;108:226-233. 121. Hogberg T. Adjuvant chemotherapy in endometrial carcinoma: overview of randomized trials. Clin Oncol (R Coll Radiol) 2008;20:463-469. 122. Mariani A, Webb MJ, Keeney GL, et al. Assessment of prognostic factors in stage IIIA endometrial cancer. Gynecol Oncol 2002;86:38-44. 123. McMeekin DS, Lashbrook D, Gold M, et al. Analysis of FIGO stage IIIc endometrial cancer patients. Gynecol Oncol 2001;81: 273-278. 124. Nelson G, Randall M, Sutton G, et al. FIGO stage IIIc endometrial carcinoma with metastases confined to pelvic lymph nodes: analysis of treatment outcomes, prognostic variables, and failure patterns following adjuvant radiation therapy. Gynecol Oncol 1999;75:211-214. 125. McMeekin DS, Lashbrook D, Gold M, et al. Nodal distribution and its significance in FIGO IIIc endometrial cancer. Gynecol Oncol 2001;82:375-379. 126. Greven KM, Lanciano RM, Corn B, et al. Pathologic stage III endometrial carcinoma. Prognostic factors and patterns of recurrence. Cancer 1993;71:3697-702. 127. Mundt AJ, Murphy KT, Rotmensch J, et al. Surgery and postoperative radiation therapy in FIGO stage IIIc endometrial carcinoma. Int J Radiat Oncol Biol Phys 2001;50:1154-1160. 128. Komaki R, Mattingly RF, Hoffman RG, et al. Irradiation of paraaortic lymph node metastases from carcinoma of the ­ cervix or endometrium. Preliminary results. Radiology 1983;147: 245-248.

795

129. Hicks ML, Piver MS, Puretz JL, et al. Survival in patients with paraaortic lymph node metastases from endometrial adenocarcinoma clinically limited to the uterus. Int J Radiat Oncol Biol Phys 1993;26:607-611. 130. Blythe JG, Edwards E, Heimbecker P. Paraaortic lymph node biopsy: a twenty-year study. Am J Obstet Gynecol 1997;176: 1157-1162. 131. Onda T, Yoshikawa H, Mizutani K, et al. Treatment of node­positive endometrial cancer with complete node dissection, chemotherapy and radiation therapy. Br J Cancer 1997;75:1836-1841. 132. Randall ME, Filiaci VL, Muss H, et al. Randomized phase III trial of whole-abdominal irradiation versus doxorubicin and cisplatin chemotherapy in advanced endometrial carcinoma: a Gynecologic Oncology Group study. J Clin Oncol 2006;24(18S): 36-44. 133. Mundt AJ, McBride R, Rotmensch J, et al. Significant pelvic recurrence in high-risk pathologic stage I-IV endometrial carcinoma patients after adjuvant chemotherapy alone: implications for adjuvant radiation therapy. Int J Radiat Oncol Biol Phys 2001;50:1145-1153. 134. Eifel PJ, Klopp A, Ramondetta LE, et al. Node-positive adenocarcinoma of the endometrium: outcome and patterns of recurrence [abstract 1065]. Int J Radiat Oncol Biol Phys 2006;66:S166. 135. Potish RA. Abdominal radiotherapy for cancer of the uterine cervix and endometrium. Int J Radiat Oncol Biol Phys 1989;16:1453-1458. 136. Mallipeddi P, Kapp DS, Teng NNH. Long-term survival with adjuvant whole abdominopelvic irradiation for uterine papillary serous carcinoma. Cancer 1993;71:3076-3081. 137. Grice J, Ek M, Greer B, et al. Uterine papillary serous ­carcinoma: evaluation of long-term survival in surgically staged patients. Gynecol Oncol 1998;69:69-73. 138. Smith RS, Kapp DS, Chen Q, et al. Treatment of high-risk uterine cancer with whole abdominopelvic radiation therapy. Int J Radiat Oncol Biol Phys 2000;48:767-778. 139. Martinez AA, Weiner S, Podratz K, et al. Improved outcome at 10 years for serous-papillary/clear cell or high-risk endometrial cancer patients treated by adjuvant high-dose whole abdominopelvic irradiation. Gynecol Oncol 2003;90:537-546. 140. Sutton G, Axelrod JH, Bundy BN, 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 2006;100: 349-354. 141. Sutton G, Axelrod JH, Bundy BN, 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;97:755-763. 142. Greven KM, Lanciano RM, Herbert SH, et al. Analysis of complications in patients with endometrial carcinoma ­ receiving ­adjuvant irradiation. Int J Radiat Oncol Biol Phys 1991;21: 919-923. 143. Corn BW, Lanciano RM, Greven KM, et al. Impact of improved irradiation technique, age and lymph node sampling on the ­severe complication rate of surgically staged endometrial cancer patients: a multivariate analysis. J Clin Oncol 1994;12:510-515. 144. Roeske JC, Lujan A, Romensch J, et al. Intensity-modulated whole pelvic radiation therapy in patients with gynecologic ­malignancies. Int J Radiat Oncol Biol Phys 2000;48:1613-1621. 145. Heron DE, Gerszten K, Selvaraj RN, et al. Conventional 3D conformal versus intensity-modulated radiotherapy for the ­adjuvant treatment of gynecologic malignancies: a comparative dosimetric study of dose-volume histograms. Gynecol Oncol 2003;91:39-45. 146. Ahamad A, D’Souza W, Salehpour M, et al. Intensity-modulated radiation therapy after hysterectomy: comparison with conventional treatment and sensitivity of the normal-tissue-­sparing effect to margin size. Int J Radiat Oncol Biol Phys 2005;62: 1117-1124. 147. Mundt AJ, Roeske JC, Lujan AE, et al. Initial clinical experience with intensity-modulated whole-pelvis radiation therapy in women with gynecologic malignancies. Gynecol Oncol 2001;82:456-463.

796

PART 9  Female Genital Tract

148. Mundt AJ, Mell LK, Roeske JC. Preliminary analysis of ­chronic gastrointestinal toxicity in gynecology patients treated with ­intensity-modulated whole pelvic radiation therapy. Int J Radiat Oncol Biol Phys 2003;56:1354-1360. 149. Beriwal S, Jain SK, Heron DE, et al. Clinical outcome with adjuvant treatment of endometrial carcinoma using ­ intensity­modulated radiation therapy. Gynecol Oncol 2006;102:195-199. 150. Greven KM, D’Agostino RB Jr, Lanciano RM, et al. Is there a role for a brachytherapy vaginal cuff boost in the adjuvant management of patients with uterine-confined endometrial cancer? Int J Radiat Oncol Biol Phys 1998;42:101-104. 151. Sorbe BG, Smeds AC. Postoperative vaginal irradiation with high dose rate after loading technique in endometrial carcinoma stage I. Int J Radiat Oncol Biol Phys 1990;18:305-314. 152. Thomas H, Pickerine GL, Dunn P, et al. Treating the vaginal vault in carcinoma of the endometrium using the Buchler afterloading system. Br J Radiol 1991;64:1044-1048. 153. Weiss E, Hirnle P, Arnold-Bofinger H, et al. Adjuvant vaginal high-dose-rate afterloading alone in endometrial carcinoma: patterns of relapse and side effects following low-dose therapy. Gynecol Oncol 1998;71:72-76. 154. Noyes WR, Bastin K, Edwards SA, et al. Postoperative vaginal cuff irradiation using high dose rate remote afterloading: a phase II clinical protocol. Int J Radiat Oncol Biol Phys 1995;32: 1439-1443. 155. Small W, Erickson B, Kwakwa F. American Brachytherapy Society survey regarding practice patterns of postoperative ­irradiation for endometrial cancer: current status of vaginal brachytherapy. Int J Radiat Oncol Biol Phys 2005;63: 1502-1507. 156. Grigsby P, Kuske R, Perez C, et al. Medically inoperable stage I adenocarcinoma of the endometrium treated with radiotherapy alone. Int J Radiat Oncol Biol Phys 1987;13:483-488. 157. Jones D, Stout R. Results of intracavitary radium treatment for adenocarcinoma of the body of the uterus. Clin Radiol 1986;37:169-171. 158. Kupelian PA, Eifel PJ, Tornos C, et al. Treatment of endometrial carcinoma with radiation therapy alone. Int J Radiat Oncol Biol Phys 1993;27:817-824. 159. Landgren R, Fletcher G, Delclos L, et al. Irradiation of endometrial cancer in patients with medical contraindication to surgery or with unresectable lesions. Am J Roentgenol 1976;126: 148-154. 160. Lehoczky O, Bosze P, Ungar L, et al. Stage I endometrial carcinoma: treatment of nonoperable patients with intracavitary radiation therapy alone. Gynecol Oncol 1991;43:211-216. 161. Rose PG, Baker S, Kern M, et al. Primary radiation therapy for endometrial carcinoma: a case controlled study. Int J Radiat ­Oncol Biol Phys 1993;27:585-590. 162. Rouanet P, Dubois JB, Gely S, et al. Exclusive radiation therapy in endometrial carcinoma. Int J Radiat Oncol Biol Phys 1993;26:223-228. 163. Heyman J. The so-called Stockholm method and the results of treatment of uterine cancer at the Radiumhemmet. Acta ­Radiol 1935;22:129. 164. Luk KH, Castro JR, Meyler TS. 137Cs afterloading intrauterine packing for stage I and stage II adenocarcinoma of the ­endometrium. Radiology 1978;129:229-231. 165. Chao CK, Grigsby PW, Perez CA, et al. Brachytherapy-related complications for medically inoperable stage I endometrial carcinoma. Int J Radiat Oncol Biol Phys 1995;31:37-42. 166. Grigsby PW. The role of radiotherapy in endometrial carcinoma. Ann Acad Med Singapore 1996;25:437-440. 167. Kucera H, Knocke TH, Kucera E, et al. Treatment of endometrial carcinoma with high-dose-rate brachytherapy alone in medically inoperable stage I patients. Acta Obstet Gynecol Scand 1998;77:1008-1012. 168. Nguyen C, Souhami L, Roman TN, et al. High-dose-rate brachytherapy as the primary treatment of medically inoperable stage I-II endometrial carcinoma. Gynecol Oncol 1995;59:370-375. 169. Nguyen TV, Petereit DG. High-dose-rate brachytherapy for medically inoperable stage I endometrial cancer. Gynecol Oncol 1998;71:196-203.

170. Knocke TH, Kucera H, Weidinger B, et al. Primary treatment of endometrial carcinoma with high-dose-rate brachytherapy: results of 12 years of experience with 280 patients. Int J Radiat Oncol Biol Phys 1997;37:359-365. 171. Petereit DG, Pearcey R. Literature analysis of high dose rate brachytherapy fractionation schedules in the treatment of cervical cancer: is there an optimal fractionation schedule? Int J Radiat Oncol Biol Phys 1999;43:359-366. 172. Petereit DG, Sarkaria JN, Chappell RJ. Perioperative morbidity and mortality of high-dose-rate gynecologic brachytherapy. Int J Radiat Oncol Biol Phys 1998;42:1025-1031. 173. Cliby WA, Dodson MK, Podratz KC. Cervical cancer ­complicated by pregnancy: episiotomy site recurrences following vaginal ­delivery. Obstet Gynecol 1994;84:179-182. 174. Copeland LJ, Saul PB, Sniege N. Cervical adenocarcinoma: tumor implantation in the episiotomy sites of two patients. Gynecol Oncol 1987;28:230-235. 175. Greven K, Olds W. Isolated vaginal recurrences of endometrial adenocarcinoma and their management. Cancer 1987;60: 419-421. 176. Curran WJ, Whittington R, Peters AJ, et al. Vaginal recurrences of endometrial carcinoma: the prognostic value of staging by a primary vaginal carcinoma system. Int J Radiat Oncol Biol Phys 1988;15:803-808. 177. Sears JD, Greven KM, Hoen HM, et al. Prognostic factors and treatment outcome for patients with locally recurrent endometrial cancer. Cancer 1994;74:1303-1308. 178. Hart KB, Han I, Shamsa F, et al. Radiation therapy for endometrial cancer in patients treated for postoperative recurrence. Int J Radiat Oncol Biol Phys 1998;41:7-11. 179. Ackerman I, Malone S, Thomas G, et al. Endometrial ­carcinoma— relative effectiveness of adjuvant irradiation vs therapy reserved for relapse. Gynecol Oncol 1996;60:177-183. 180. Mandell LR, Nori D, Hilaris B. Recurrent stage I endometrial carcinoma: results of treatment and prognostic factors. Int J ­Radiat Oncol Biol Phys 1985;11:1103-1109. 181. Hoekstra CJ, Koper PC, van Putten WL. Recurrent endometrial adenocarcinoma after surgery alone: prognostic factors and treatment. Radiother Oncol 1993;27:164-166. 182. Wylie J, Irwin C, Pintilie M, et al. Results of radical radiotherapy for recurrent endometrial cancer. Gynecol Oncol 2000;77:66-72. 183. Kuten A, Grigsby PW, Perez CA, et al. Results of radiotherapy in recurrent endometrial carcinoma: a retrospective analysis of 51 patients. Int J Radiat Oncol Biol Phys 1989;17:29-34. 184. Ingersol FM. Vaginal recurrences of carcinoma of the corpus— management and prevention. Am J Surg 1971;121:473-477. 185. Aalders JG, Abeler V, Kolstad P. Recurrent adenocarcinoma of the endometrium: a clinical and histopathological study of 379 patients. Gynecol Oncol 1984;17:85-103. 186. Jhingran A, Burke TW, Eifel PJ. Definitive radiotherapy for patients with isolated vaginal recurrence of endometrial carcinoma after hysterectomy. Int J Radiat Oncol Biol Phys 2003;56:1366-1372. 187. Humber C, Tierney J, Symonds P, et al. Chemotherapy for advanced, recurrent or metastatic endometrial carcinoma. ­Cochrane Database Syst Rev 2005(3):CD003915. 188. Pawinski A, Tumolo S, Hoesel G, et al. Cyclophosphamide or ifosfamide in patients with advanced and/or recurrent endometrial carcinoma: a randomized phase II study of the EORTC Gynecological Cancer Cooperative Group. Eur J Obstet Gynecol Reprod Biol 1999;86:179-183. 189. Edmonson JH, Krook JE, Hilton JF, et al. Randomized phase II studies of cisplatin and a combination of ­cyclophosphamide­doxorubicin-cisplatin (CAP) in patients with progestin-­refractory advanced endometrial carcinoma. Gynecol Oncol 1987;28: 20-24. 190. Aapro MS, Van Wijk FH, Bolis G, et al. Doxorubicin versus doxorubicin and cisplatin in endometrial carcinoma: definitive results of a randomized study (55872) by the EORTC-GCCG. Ann Oncol 2003;14:441-448. 191. Long HJ, Nelimark RA, Cha SS. Comparison of methotrexate, vinblastine, doxorubicin and cisplatin (MVAC) vs doxorubicin and cisplatin (AC) in advanced endometrial carcinoma [abstract 797]. Proc Am Soc Clin Oncol 1995;14:282.

30  The Endometrium 192. Thigpen JT, Brady ME, Homesley H, et al. Phase III trial of doxorubicin with or without cisplatin in advanced endometrial carcinoma: a Gynecologic Oncology Group Study. J Clin Oncol 2004;22:3902-3908. 193. Fleming GF, Brunetto VL, Cella D, et al. Phase III trial of doxorubicin plus cisplatin with or without paclitaxel plus filgrastim in advanced endometrial carcinoma: a Gynecologic Oncology Group Study. J Clin Oncol 2004;22:2159-2166. 194. DiSaia PJ, Castro JR, Rutledge FN. Mixed mesodermal sarcoma of the uterus. Am J Roentgenol Radium Ther Nucl Med 1973;117:632-636. 195. Echt G, Jepson J, Steel J, et al. Treatment of uterine sarcomas. Cancer 1990;66:35-39. 196. George M, Pejovic MH, Kramar A. Uterine sarcomas: ­prognostic factors and treatment modalities—study on 209 patients. ­Gynecol Oncol 1986;24:58-67. 197. 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;12:2127-2130. 198. Larson B, Silfverswärd C, Nilsson B, et al. Mixed Müllerian tumours of the uterus—prognostic factors: a clinical and histopathologic study of 147 cases. Radiother Oncol 1990;17: 123-132. 199. Major FJ, Blessing JA, Silverberg SG, et al. Prognostic factors in early-stage uterine sarcoma. Cancer 1993;71:1702-1709. 200. Moskovic E, Macsweeney E, Law M, et al. Survival, patterns of spread and prognostic factors in uterine sarcoma: a study of 76 patients. Br J Radiol 1993;66:1009-1015. 201. Callister M, Ramondetta LM, Jhingran A, et al. Malignant mixed Mullerian tumors of the uterus: analysis of patterns of failure, prognostic factors, and treatment outcome. Int J Radiat Oncol Biol Phys 2004;58:786-789. 202. Sorbe B. Radiotherapy and/or chemotherapy as adjuvant treatment of uterine sarcomas. Gynecol Oncol 1985;20:281-289.

797

203. Wheelock JB, Krebs HB, Schneider V, et al. Uterine sarcoma: analysis of prognostic variables in 71 cases. Am J Obstet Gynecol 1985;151:1016-1022. 204. Gerszten K, Faul C, Kounelis S, et al. The impact of adjuvant radiotherapy on carcinosarcoma of the uterus. Gynecol Oncol 1998;68:8-13. 205. Ferrer F, Sabater S, Farrús B, et al. Impact of radiotherapy on local control and survival in uterine sarcomas: a retrospective study from the Grup Oncologic Català-Occità. Int J Radiat ­Oncol Biol Phys 1999;44:47-52. 206. Chauveinc L, Deniaud E, Plancher C, et al. Uterine sarcomas: the Curie Institut experience. Prognosis factors and adjuvant treatments. Gynecol Oncol 1999;72:232-237. 207. Brooks SE, Zhan M, Cote T, et al. Surveillance, Epidemiology and End Results analysis of 2677 cases of uterine sarcoma 19891999. Gynecol Oncol 2004;93:204-208. 208. Wolfson AH, Brady RS, Mannel RS, et al. A Gynecologic ­Oncology Group randomized trial of whole abdominal irradiation (WAI) vs cisplatin-ifosfamide + mesna (CIM) in optimally staged I-IV carcinosarcoma (CS) of the uterus [abstract 5001]. J Clin Oncol 2006;24(18S)-256. 209. Rose PG, Boutselis JG, Sachs L. Adjuvant therapy for stage I uterine sarcoma. Am J Obstet Gynecol 1987;156:660-662. 210. Larson B, Silfverswärd C, Nilsson B, et al. Endometrial stromal sarcoma of the uterus. A clinical and histopathological study. The Radiumhemmet series 1936-1981. Eur J Obstet Gynecol Reprod Biol 1990;35:239-249. 211. Berchuck A, Rubin SC, Hoskins WJ, et al. Treatment of endometrial stromal tumors. Gynecol Oncol 1990;36:60-65. 212. DeFusco PA, Gaffey TA, Malkasian GD, et al. Endometrial stromal sarcoma: review of Mayo Clinic experience, 1945-1980. Gynecol Oncol 1989;35:8-14.

31 The Vulva and Vagina Patricia J. Eifel EFFECTS OF RADIATION ON THE NORMAL VULVA AND VAGINA Effects on the Vulva Effects on the Vagina EPIDEMIOLOGY OF VULVAR AND VAGINAL CANCER NATURAL HISTORY AND PATTERN OF SPREAD Vulvar Cancer Vaginal Cancer PATHOLOGY DIAGNOSIS, CLINICAL EVALUATION, AND STAGING PROGNOSTIC FACTORS Vulvar Cancer Vaginal Cancer

Effects of Radiation on the Normal Vulva and Vagina Effects on the Vulva The external vulva and perineal skin respond to radiation therapy like other skin responds, although complicating fungal or bacterial infections are more common at this ­location. Acutely, a brisk erythema may develop with epila­ tion of the pubic hair; as the dose of radiation is increased, patchy or confluent desquamation may occur. The acute skin reaction usually heals within 2 to 3 weeks after the completion of radiation therapy. Hyperpigmentation and epilation may resolve more slowly, and permanent perineal epilation can occur after high doses (>40 to 50 Gy). Over time, the vulvar skin may become atrophic and telangiectasia may develop, particu­ larly if the vulva has received more than 50 to 60 Gy. Di­ minished estrogen levels may increase the severity of vulvar atrophy. The effects of vulvar radiation therapy on sexual function and satisfaction have not been comprehensively studied.

Effects on the Vagina The effects of pelvic radiation therapy on vaginal function are related to the dose of radiation, the amount of vaginal mucosa treated to a high cumulative dose, the extent of tissue destruction by tumor, and the care given after ra­ diation (including estrogen support). Patient factors such as age, menopausal status, and sexual activity also play a role. The vagina is often said to tolerate very high doses of radiation, but these comments usually refer to the rela­ tively low incidence of major side effects involving adjacent

798

TREATMENT OF VULVAR CANCER Treatment of Preinvasive Disease Surgical Treatment of Invasive Carcinoma Postoperative Radiation Therapy Prophylactic Inguinal Irradiation Initial Radiation Therapy for Locoregionally Advanced Disease Radiation Therapy Techniques Management of Acute Radiation-Induced Side Effects Treatment of Metastatic Vulvar Cancer TREATMENT OF VAGINAL CANCER Treatment of Preinvasive Disease Treatment of Invasive Disease Role of Chemotherapy

organs (bladder, urethra, and rectum) after vaginal intra­ cavitary radiation therapy (ICRT). The effect of radiation on these structures is discussed later in this chapter and in Chapter 29. The influence of radiation therapy on vaginal function and specifically on sexual function is much less well docu­ mented. Preliminary studies suggest that women who have been treated for cervical or endometrial carcinoma tend to have gradual shortening of the vagina during the first few years after treatment.1 Decreased estrogen levels ­resulting from radiation effects on ovarian function also may contrib­ ute to vaginal changes. These changes are associated with a tendency for treated women to be less sexually active and to report a decrease in sexual satisfaction.2 ­However, the specific causes for this decreased sexual activity and ­satisfaction are not well established. Typically, definitive treatment of carcinoma of the ­cervix results in a maximum total dose of 100 to 140 Gy to the apical vagina. Acute effects of this treatment are edema and congestion of the vaginal mucosa.3 Superficial erosion may appear in the first few months. Over several years, the up­ per vagina and cervix may become scarred and agglutinated, obscuring the cervical os. These changes may progress over many years and probably accelerate if estrogen support is withdrawn or if the patient is sexually inactive. In some ­cases, severe foreshortening of the vagina occurs, compromis­ ing satisfying sexual intercourse. Severe narrowing and fore­ shortening of the vagina occur more commonly in women who are treated when they are postmenopausal.4 This ­finding may reflect less consistent use of estrogen replace­ ment therapy, less frequent sexual intercourse, and poorer compliance with recommended use of vaginal dilators.5

31  The Vulva and Vagina

Epidemiology of Vulvar and Vaginal Cancer Invasive carcinomas of the vulva and vagina are rare neoplasms, affecting only about 4% and 2% of women with gynecologic tumors, respectively.6,7 These types of cancer tend to affect elderly women; the median age of women diagnosed with vulvar or vaginal cancer is 65 to 70 years, and the incidence peaks in women older than 70 years.6-8 Evidence implicating human papillomavirus (HPV) in the pathogenesis of cervical cancer has led investigators to look for HPV deoxyribonucleic acid (DNA) in vulvar and vaginal neoplasms. Ikenberg and colleagues9 found HPV DNA in 10 of 18 cases of vaginal cancer. In a large ­population-based study, the prevalence of HPV infection in the vagina of women after hysterectomy was similar to the prevalence of HPV infection in the cervix of women who still had their uterus; in view of the relative rarity of primary vaginal cancer, the investigators concluded that the cervical transformation zone may not be required for HPV infection but may be important to the transforma­ tion process.10 It has been suggested that the squamous metaplasia that occurs during healing of mucosal abrasions might encourage HPV infections that lead to neoplastic transformation under appropriate conditions.11 Invasive vaginal carcinoma has also been associated with chronic ir­ ritant vaginitis, particularly the type caused by chronic use of a vaginal pessary.12 Although 80% to 90% of vulvar intraepithelial ­neoplasia (VIN) lesions contain HPV, only 30% to 50% of invasive ­ vulvar carcinomas are associated with evi­ dence of HPV ­infection.13,14 Women with HPV-­positive tumors tend to be relatively young and have risk ­factors similar to those ­ associated with cervical cancer.13-15 Based on the prevalence of HPV in these lesions, it has been estimated that HPV ­ vaccination could pre­ vent about one half of the ­ vulvar carcinomas in young women and about two thirds of the intraepithelial lesions in the lower genital tract. Some investigators have found a higher rate of TP53 ­ mutations in women with HPV-­negative tumors, suggesting that inactivation of TP53 is important in ­ patients with HPV-positive or HPV-­negative tumors.16,17 In the early 1970s, an increase in the rate of a rare form of clear cell adenocarcinoma of the vagina and cer­ vix was found to be associated with maternal ingestion of diethylstilbestrol (DES) during pregnancy.18 The peak number of DES-associated adenocarcinomas occurred in 1975; the incidence then declined as public awareness led to a decrease in the use of DES to prevent miscarriage.19 The peak risk period for development of adenocarcinoma of the vagina and cervix in women exposed to DES in utero was between the ages of 15 and 22 years. Newly reported cases have become rare in the cohort of exposed women, most of whom are now more than 40 years old. No clear evidence exists that DES-exposed women are at risk for lower genital tract malignancies other than clear cell carcinoma,20 although some studies have indicated that women who were exposed to DES in utero do have an increased risk of developing breast cancer after age 40 years.21

799

Natural History and Pattern of Spread Vulvar Cancer About two thirds of vulvar squamous cell carcinomas in­ volve the labia majora or minora.22,23 A smaller percentage arise in the clitoris or gynecologic perineum. About 10% are multifocal or too extensive to permit identification of a site of origin. From the site of origin, carcinomas of the vulva can extend to invade the vagina, urethra, or anus; ad­ vanced vulvar carcinoma can invade adjacent pelvic bones, particularly the pubis. The vulva is richly supplied with lymphatic vessels that often cross the midline. As a result, the risk of regional spread is significant for any vulvar carcinoma that has invaded to a depth of more than 1 mm.22,24 The primary site of lym­ phatic drainage is to the superficial inguinal nodes. Although carcinomas of the clitoris and Bartholin’s gland occasion­ ally spread directly to the deep femoral and even obturator lymph nodes, these nodal groups are rarely involved without evidence of superficial inguinal metastases.25,26 Despite the extensive anastomosis of lymphatics in the region, metas­ tasis of vulvar carcinoma to contralateral lymph nodes is ­uncommon in patients with well-lateralized T1 lesions.27 The most common site of hematogenous metastasis from vulvar cancer is the lung. Advanced cancer can also spread to bone, liver, and other sites.

Vaginal Cancer Vaginal cancer arises more often in the upper third of the vagina but can also originate in the middle or lower third; a fairly even distribution of lesions is evident, arising on the anterior, posterior, and lateral walls.28-32 About 60% of primary apical vaginal cancer is diagnosed in women who have undergone hysterectomy. This is probably a result of the International Federation of Gynecology and Obstetrics (FIGO) convention that classifies any tumor involving the cervix and vagina as a cervical cancer. Vaginal cancer spreads laterally to involve the paravagi­ nal tissues and may become fixed to pelvic wall structures. Although vaginal cancer can invade the adjacent urethra, bladder, or rectum, less than 10% of cases involve the mu­ cosa of these structures at presentation.6,30,31,33 The vagina, like the vulva, is richly supplied with mu­ cosal and submucosal lymphatics.26 Nearly all of the pelvic lymph nodes can serve as primary sites of lymphatic drain­ age from vaginal tumors; the specific nodes at risk depend on the primary tumor location. The drainage of apical vagi­ nal tumors is similar to that of cervical cancer, with initial regional spread primarily to the obturator and hypogastric nodes. The lymphatics of the posterior wall form anasto­ moses with those of the anterior rectal wall, draining to the superior and inferior gluteal nodes. Tumors involving the lower third of the vagina drain to the pelvic nodes or by way of the vulvar lymphatics to the inguinofemoral lymph nodes. Although few data have been collected concerning the risk of regional spread of vaginal cancer, available stud­ ies suggest that at least 25% of patients with stage II vaginal cancer have pelvic node metastases.34,35 The most common site of hematogenous metastasis from vaginal cancer is the lung. Advanced cancer can also spread to bone, liver, and other sites.

800

PART 9  Female Genital Tract

Pathology VIN and vaginal intraepithelial neoplasia (VAIN) share many of the clinical and histologic features of cervical intraepithelial neoplasia, although VIN and VAIN tend to occur in older women. Histologically, these lesions are characterized by disruption of the normal epithelial archi­ tecture with various degrees of cytoplasmic and nuclear maturation.36,37 About 4% of patients treated for VIN ­subsequently develop an invasive cancer.38 Paget’s disease of the vulva is a rare intraepithelial lesion of the epidermis and skin adnexa that typically manifests in postmenopausal white women as a pruritic, red, sharply demarcated lesion on the labia majora. Histologically, it is characterized by large, pale, mucopolysaccharide-rich cells that probably arise from primitive epithelial progenitor cells.39 About 5% to 10% of newly diagnosed Paget’s ­lesions are associated with underlying adenocarcinoma arising in the vulva or a distant site.40 More than 90% of invasive vulvar tumors and 80% to 90% of primary vaginal malignancies are squamous cell carcinomas.6-8 Invasive vulvar carcinomas are most com­ monly well-differentiated keratinizing squamous carcino­ mas; HPV-positive tumors often have a wartlike or basaloid growth pattern. Histologically, invasive vaginal carcinomas are similar to squamous tumors at other sites. They are usu­ ally well to moderately differentiated and may or may not exhibit prominent keratin formation. Verrucous carcinoma is a rare variant of squamous cell carcinoma that manifests as a warty, fungating mass in the vulva or vagina.41,42 This well-differentiated lesion rarely metastasizes but can be highly infiltrative and locally destructive. Although they are rare, primary small-cell anaplastic car­ cinomas can arise in the vulva or vagina. These tumors re­ semble neuroendocrine, small-cell anaplastic carcinomas of the lung or cervix, may be admixed with areas of squamous carcinoma or adenocarcinoma, and tend to behave aggres­ sively, with frequent recurrence and distant ­metastasis.43-45 When adenocarcinoma occurs in the vagina, metasta­ sis or local extension from other sites must be ruled out. ­Except for clear cell carcinomas associated with in utero DES exposure, most primary adenocarcinomas of the ­vagina occur in postmenopausal women. Adenocarcinomas account for less than 5% to 10% of vaginal cancer cases and can demonstrate a variety of müllerian histologic pat­ terns. They can arise in foci of adenosis or endometriosis, sometimes in or around Bartholin’s gland.46,47 The progno­ sis for these rare, non-DES–related vaginal adenocarcino­ mas seems to be much poorer than that for squamous cell ­carcinoma of the vagina.46 Malignant melanomas of the vulva and vagina ­represent less than 5% of primary neoplasms in these sites.48,49 A poorer prognosis is associated with deep ­ invasion, ­ulceration, clinical amelanosis, and older age.50 Chung and colleagues51 and Tasseron and colleagues52 proposed a modification of the Clark classification ­system (a meth­ od of classifying malignant cutaneous melanomas on the basis of level of invasion) that is ­often used to categorize ­vulvar melanomas.51,52 Primary vaginal melanomas ­usually ­originate in the lower third of the vagina and tend to have a poorer prognosis than vulvar melanomas; for vaginal ­melanomas, 5-year ­survival rates are only 15% to 20% de­ spite aggressive treatment.53,54

About 1% to 2% of vulvovaginal malignancies are soft tissue sarcomas. Leiomyosarcomas are most common, fol­ lowed by a variety of other histologic subtypes. Risk factors for local recurrence and metastasis are similar to those re­ ported for primary sarcomas arising at other sites and in­ clude large size, close surgical margins, and high grade.53,54 Embryonal rhabdomyosarcoma (sarcoma botryoides) oc­ curs in these sites, particularly in children younger than 6 years, and is usually treated with chemotherapy combined with surgery or radiation therapy.55 Rhabdomyosarcoma is discussed further in Chapter 37.

Diagnosis, Clinical Evaluation, and Staging Carcinomas involving the vulva and vagina are classified ac­ cording to FIGO guidelines. Cases are classified as vaginal carcinomas only when “the primary site of the growth is in the vagina.”6 Any tumor that has extended from the vulva to involve the vagina should be classified as a primary vulvar cancer, and any tumor that involves the area of the external cervical os should be classified as a cervical carcinoma. Patients with invasive vulvar carcinoma usually pres­ ent with pruritus or a mass; patients with VIN, however, may have no symptoms at diagnosis.56,57 Advanced vulvar lesions can be extremely painful. Patients with primary vaginal carcinoma usually present with abnormal vaginal bleeding but may also have pain (including dyspareunia), discharge, or a mass. All patients with invasive disease should be evaluated with a careful physical examination including a detailed pelvic examination, chest radiography, a complete blood cell count, and a biochemical profile. In some cases, col­ poscopy may be needed to rule out a concurrent cervi­ cal cancer. Cystoscopy should be performed in patients with tumors near the bladder or urethra, and proctoscopy should be performed in patients with tumors near the anus or rectum. Computed tomography (CT) or magnetic reso­ nance imaging (MRI) scans should be obtained to evaluate deep inguinal and pelvic lymph nodes, localize the kidneys, and rule out hydronephrosis. MRI also can be helpful in delineating the distribution of disease in patients with lo­ cally advanced vaginal cancer, although the results cannot be used to alter the clinical stage assigned to the lesion. The FIGO staging system for carcinoma of the vulva was changed from a clinical to a surgical staging system in 1988 and depends on “thorough histopathologic evaluation of the operative specimen (vulva and lymph nodes).” The staging system was revised again in 1994 to create a sepa­ rate stage IA for tumors that invade no more than 1 mm (Box 31-1).7,58 Many locally advanced vulvar lesions are be­ ing treated with initial radiation or chemoradiation therapy; consequently, the absence of a complete surgical specimen may interfere with accurate staging, particularly of the in­ guinal lymph nodes. FIGO has yet to address this problem. The FIGO rules for clinical staging of carcinoma of the va­ gina (Table 31-1)59 are the same as those for clinical ­staging of cervical cancer. The categories are also ­similar, ­although distinguishing stage I from stage II ­disease tends to be quite subjective because it is often impossible to ­accurately assess the presence or absence of subvaginal involvement by clini­ cal examination alone. Perez and ­colleagues31 recommend

31  The Vulva and Vagina

801

BOX 31-1.  American Joint Committee on Cancer and International Federation of Gynecology and Obstetrics Staging Systems for Cancer of the Vulva TNM Categories TX T0 Tis T1 T1a T1b

FIGO Stage — — 0 I IA IB

T2

II

T3 T4

III IVA

Primary tumor cannot be assessed No evidence of primary tumor Carcinoma in situ (preinvasive carcinoma) Tumor confined to the vulva or vulva and perineum, 2 cm or less in greatest dimension Tumor confined to the vulva or vulva and perineum, 2 cm or less in greatest dimension, and with stromal invasion no greater than 1 mm Tumor confined to the vulva or vulva and perineum, 2 cm or less in greatest dimension, and with stromal invasion greater than 1 mm Tumor confined to the vulva or vulva and perineum, more than 2 cm in greatest ­dimension Tumor of any size with contiguous spread to the lower urethra and/or vagina or anus Tumor invades any of the following: upper urethra, bladder mucosa, rectal mucosa, or is fixed to the pubic bone

Regional Lymph Nodes (N) NX N0 N1 N2

III IVA

Regional lymph nodes cannot be assessed No regional lymph node metastasis Unilateral regional lymph node metastasis Bilateral regional lymph node metastasis

Distant Metastasis (M) MX M0 M1 Stage Grouping Stage 0 Stage I Stage IA Stage IB Stage II Stage III

Stage IVA

Stage IVB

IVB

Distant metastasis cannot be assessed No distant metastasis Distant metastasis (including pelvic lymph node metastasis)

Tis T1 T1a T1b T2 T1 T2 T3 T3 T1 T2 T3 T4 Any T

N0 N0 N0 N0 N0 N1 N1 N0 N1 N2 N2 N2 Any N Any N

M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M1

*The depth of invasion is defined as the measurement of the tumor from the epithelial-stromal junction of the adjacent most superficial dermal papilla to the deepest point of invasion. † Every effort should be made to determine the site and laterality of lymph node metastases. However, if “regional lymph node metastasis, NOS” is the final diagnosis, the patient should be staged as N1. From Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 243-249.

a modification of the FIGO system that distinguishes be­ tween tumors that infiltrate the parametrium (stage IIB) and tumors with paravaginal submucosal extension only (stage IIA). However, this system fails to clearly define what is meant by parametrium, ­ particularly for women who have no uterus. The American Joint Committee on Cancer (AJCC) sug­ gests a tumor-node-metastasis (TNM) staging system that allows the stage assignment to be changed according to the presence of regional lymph node involvement. This system has not gained acceptance because patients rarely have ­surgical treatment for vaginal cancer, because the stage groupings are inconsistent with accepted FIGO staging cri­ teria, and because no accepted guidelines are available for determination of regional involvement.

Prognostic Factors Vulvar Cancer For patients with vulvar carcinoma, outcomes and the risk of regional metastasis correlate strongly with tumor size, depth of invasion, tumor thickness, and the pres­ ence or absence of lymphovascular space invasion.60-64 More than 75% of patients with lymphovascular space invasion have positive inguinal nodes.24,62,63 However, in a multivariate analysis of 389 patients, Raspagliesi and colleagues64 found lymphovascular space invasion and lymph node involvement to be independent predictors of outcome. Various conclusions have been drawn about the prognostic ­ importance of tumor grade and patient age.24,62,63,65-67

802

PART 9  Female Genital Tract 100

TABLE 31-1.  F  IGO Clinical Staging System for Carcinoma of the Vagina Carcinoma in situ, intraepithelial carcinoma

Stage I

The carcinoma is limited to the vaginal wall.

Stage II

The carcinoma has involved the subvaginal tissues but has not extended onto the pelvic wall.

Stage III

The carcinoma has extended onto the pelvic wall.

Stage IV

The carcinoma has extended beyond the true pelvis or has clinically involved the mucosa of the bladder or rectum. Bullous edema as such does not permit a case to be allotted to stage IV.

Stage IVA

Spread of the growth to adjacent organs and/or direct extension beyond the true pelvis

Stage IVB

Spread to distant organs

FIGO, International Federation of Gynecology and Obstetrics. From Shepherd J, Sideri M, Benedet J, et al. Carcinoma of the ­vagina. J Epidemiol Biostat 1998;3:103-109.

The presence or absence of inguinal lymph node i­nvolvement has traditionally been considered one of the strongest predictors of prognosis.60,64,65,68,69 In a 1991 ­review of 586 patients enrolled in two Gynecologic Oncology Group (GOG) trials,68 the investigators reported 5-year sur­ vival rates of 91% for patients with negative ­inguinal lymph nodes and 75%, 36%, and 24%, respectively, for patients with one or two, three or four, or five or six positive nodes. However, more than one half of the patients with positive nodes did not receive postoperative radiation therapy, which is currently considered standard treatment. For patients who did undergo radiation therapy in a GOG randomized trial published by Homesley and colleagues,70 no clear differ­ ence in outcome was found between patients with one and ­patients with two or more positive lymph nodes (Fig. 31-1). Several investigators64,71,72 have ­ reported correlations ­between the presence of extracapsular nodal extension and recurrence even for patients who have ­received postopera­ tive radiation. It is not known whether the development of extranodal extension can be overcome by intensifying the dose of radiation delivered to the regional operative bed. The very poor 5-year survival rates reported for ­patients with positive pelvic lymph nodes (23% in the GOG trial)70 led FIGO to categorize such disease as stage IV. However, these patients were usually treated with surgery alone; be­ cause patients treated with radiation therapy rarely ­undergo pelvic lymph node dissection, the curability of disease that in­ cludes pelvic node metastases is not well understood. Howev­ er, Raspagliesi and colleagues64 ­reported a 5-year survival rate of 30% for patients who had inguinal and pelvic lymph node involvement; all of the patients with lymph node involve­ ment in that study received ­postoperative ­radiation therapy. Heaps and colleagues61 reported a correlation between the tumor-free margin in vulvar excision specimens and the incidence of local failure. They saw no local recur­ rences in 91 cases in which the uninvolved margin was 8 mm or larger in the fixed specimen (equivalent to about

75 Percent survival

Stage 0

50

25

0 0

12

24

36

Months from protocol entry Radiation therapy/1 Pelvic node dissection/1 Radiation therapy/2+ Pelvic node dissection/2+ Censored observation

FIGURE 31-1.  Results of a Gynecologic Oncology Group r­ andomized trial enrolling patients who underwent radical vulvectomy and inguinal node dissection and were found to have inguinal lymph node metastases. Patients were randomly assigned intraoperatively to additional pelvic lymph node dissection (after frozen section diagnosis of inguinal metastases) or postoperative radiation therapy to the pelvis and inguinal nodes. Survival was related to treatment and the number of positive inguinal nodes. (Modified from Homesley HD, Bundy BN, Sedlis A, et al. Radiation therapy versus pelvic node resection for carcinoma of the vulva with positive groin nodes. Obstet Gynecol 1986;68:733.)

10 mm in fresh tissue). In contrast, 18 (50%) of 36 ­patients who had margins smaller than 8 mm experienced local ­recurrence.

Vaginal Cancer Outcome for patients with vaginal cancer correlates strong­ ly with tumor stage.29-31,73 Several investigators have also reported a correlation between outcome and tumor size (measured in centimeters or as the percentage of vaginal wall involved).28,30,33 Infiltrating, necrotic lesions tend to have a poorer prognosis than exophytic tumors.31 Tumors that involve the entire vagina tend to have a poor prog­ nosis; however, attempts to correlate the site of vaginal ­tumors with outcome have yielded inconsistent results.29-31,74 Investigators also disagree about the influence of histo­ logic grade and type (e.g., squamous cell carcinoma ­versus ­adenocarcinoma) on outcome.29,30,46,73

Treatment of Vulvar Cancer Treatment of Preinvasive Disease Small VIN lesions are treated with local excision. More ­extensive VIN can be treated with carbon dioxide laser va­ porization. Careful inspection and biopsy sampling to rule out invasive disease should be done before laser treatment. Extensive disease may require wide local excision or excision of the superficial skin of the vulva (skinning ­ vulvectomy).

31  The Vulva and Vagina

Although the morbidity of excision may be greater than that of laser vaporization, the ability to examine the surgical specimen is a relative advantage of surgical excision. VIN is often a multicentric disease, and local recurrences are com­ mon even after extensive surgical resection.75

Surgical Treatment of Invasive Carcinoma Minimally invasive vulvar tumors that infiltrate less than 1 mm from the most superficial basement membrane have a very low risk of regional metastasis (<1%) and can be treated with wide local excision only.22,76 In the latest modification of its staging system, FIGO subclassified these early lesions in a new stage IA category.7 Any tumor that invades more than 1 mm from the most superficial basement membrane requires treatment aimed at the local tumor and the regional nodes. Surgical manage­ ment of vulvar cancer has evolved away from the highly morbid radical en bloc resections that were traditional 25 years ago and toward increasingly individualized, ­tissuesparing procedures. In the early 1980s, Hacker and col­ leagues77 demonstrated that vulvar resection and inguinal lymphadenectomies could be performed through separate incisions without compromising cure rates and with much less morbidity. Most surgeons have accepted wide local excision ­combined with inguinal node dissection as the ­treatment of choice for early invasive vulvar lesions.76,78 The ­excision should extend down to the inferior fascia of the urogeni­ tal diaphragm, and an effort should be made to obtain a ­ tumor-free margin of at least 1 to 2 cm ­ unless doing so would compromise major organ function.76 ­Although wide local excision of very large T2 lesions may ­ require near-total vulvectomy, surgery for smaller ­ tumors can be more limited without increasing the risk of ­ local ­recurrence.76,78-81 This is particularly advantageous for patients with ­ lesions on the lateral or posterior vulva ­because the clitoris can be preserved, reducing the psy­ chosexual ­morbidity of treatment. When lesions are close to the ­ urethra, the distal half of the urethra can usually be ­resected without loss of ­continence.76 Although ultra­ radical surgery combining vulvectomy, exenteration, and lymphadenectomy was once recommended for patients with locally advanced disease involving the urethra or anus, most of these cases now are managed with organsparing treatments that ­include combinations of radiation therapy, surgery, and chemotherapy. The inguinal nodes must receive some form of treat­ ment if more than 1 mm of stromal invasion is detected in the vulvar specimen. Even 2 to 3 mm of invasion is ­associated with a 5% to 15% risk of inguinal metastasis, and more than 25% of patients with 3 to 5 mm of inva­ sion with have lymph node metastasis.76 Prevention of regional recurrence is critical to the curative manage­ ment of vulvar cancer because inguinal recurrence is rarely ­curable.76 The standard treatment for the inguinal nodes is an ­inguinofemoral lymphadenectomy that removes the superficial and deep inguinofemoral nodes. The contra­ lateral groin need not be dissected if the primary lesion is a well-­lateralized T1 ­ tumor. However, if the primary ­lesion ­approaches the ­midline or involves the anterior la­ bia ­minora, bilateral ­dissections should be done.76 At one time, pelvic node ­resection was performed for all patients with invasive vulvar ­cancer. Eventually this procedure was

803

l­imited to patients with positive inguinal nodes; with the current standard practice of administering radiation ther­ apy to all patients with ­regional disease, the morbidity of pelvic node dissection is rarely justified.76,82 Unfortunately, serious acute and subacute complica­ tions of radical inguinal lymphadenectomy are common despite the use of separate groin incisions.83-86 They in­ clude wound complications, chronic lymphedema, and thrombotic complications; the perioperative mortality rate is 2% to 5%.70,79-81 To reduce the risk of perioperative mor­ bidity and mortality, investigators have studied the use of less radical inguinal dissections. In the 1990s, several groups reported the use of so-called superficial inguinal node dis­ sections. The morbidity of this limited procedure was less than that of a radical inguinal lymphadenectomy; however, follow-up revealed groin recurrence rates of 10% to 15%, even for patients with histologically negative nodes and early primary lesions.87-89 Subsequent investigators com­ mented on the inconsistent nomenclature used to describe groin dissection practices and speculated that groin failures were occurring in primary echelon lymph nodes that had been missed in the dissection.89-91 Several groups have explored the use of lymphatic map­ ping and sentinel lymph node biopsy of the inguinal nodes.92 Although this technique may ultimately improve the ­accuracy of more limited regional dissection, data are insuf­ ficient to justify the routine use of this technique to tailor regional treatment. The role of radiation therapy in the man­ agement of regional disease is discussed in the next section. Five-year disease-specific survival rates of ­approximately 98% and 85% can be expected for patients treated surgical­ ly for stage I (T1 N0  M0) and stage II (T2 N0  M0) disease, respectively.68,76

Postoperative Radiation Therapy Twenty years ago, vulvar cancer was rarely treated with ­radiation. Today, radiation therapy is known to improve the survival rates of patients with regional metastases and permit organ preservation in patients with locally advanced disease. Prospective trials have confirmed the importance of radia­ tion therapy in the management of locoregionally advanced disease, but they have also demonstrated that special skill and excellent cooperation are necessary for effective multi­ modality care of patients with locoregional metastases. Postoperative radiation therapy is usually indicated in the following situations: more than one positive inguinal node, one grossly positive inguinal node or extracapsular extension of nodal disease, small tumors in patients with medical problems that prohibit radical inguinofemoral lymph node dissection, or a vulvar resection margin of less than 8 to 10 mm. In 1986, Homesley and colleagues70 published the ­results of a randomized prospective study that evaluated the role of postoperative radiation therapy for patients with inguinal lymph node metastases. Patients found to have in­ guinal metastases during radical vulvectomy and inguinal lymphadenectomy were randomly assigned to undergo additional pelvic lymph node dissection (as part of the same operative procedure) or postoperative radiation ther­ apy. Radiation was delivered by means of ­anteroposterior­posteroanterior fields that included the pelvic and inguinal nodes but not the vulva. The trial was closed after 114 eli­ gible patients had been entered because an interim analysis

804

PART 9  Female Genital Tract

revealed a significant survival advantage for the ­ radiation therapy group (P = .03). Subset analysis suggested that patients with clinically positive or multiple histological­ ly positive groin nodes derived the greatest benefit (see Fig. 31-1). Only 3 (5%) groin recurrences were evident in the 59 patients treated with radiation compared with 13 (24%) in the 55 patients treated with surgery alone (P = .02). Although the investigators detected no differ­ ence in the rates of pelvic recurrence, competing risks may have compromised this comparison. This study changed the standard of care for carcinoma of the vulva and remains one of the most important trials dealing with this rare dis­ ease. An update of the preliminary analysis first published in 1986 confirmed the benefit of radiation therapy in this setting.93 In a retrospective series that included 34 patients with inguinal lymph node metastases from vulvar carcinoma, Katz and colleagues88 reported an inguinal disease con­ trol rate of 90% at 5 years after lymph node dissection and postoperative radiation. Groin recurrence correlated strongly with delayed postoperative treatment (P = .03); of four patients with recurrences who received postoperative ­radiation, three were patients whose radiation therapy was begun more than 50 days after surgery. Whenever possible, postoperative radiation should begin within 4 to 6 weeks after surgery. Referrals for postoperative radiation therapy for treat­ ment of local disease occur most often when the surgeon is unable to obtain an adequate tumor-free margin with­ out compromising anal or urethral function. Patients with deeply invasive lesions, particularly those with lymphovas­ cular space invasion, may also have inadequate margins and may benefit from postoperative radiation. Although local recurrences are often controlled with ad­ ditional surgery, Faul and colleagues94 reported an overall 5-year survival rate of only 40% for 47 patients who had an isolated vulvar recurrence after definitive therapy. In an earlier paper, the investigators had demonstrated a local re­ currence rate of only 16% when patients who had close or positive surgical margins were treated with radiation and 58% if no radiation therapy was given.95 The reduction in the rate of recurrence was significant regardless of wheth­ er patients had only close (≤8 mm) margins (P = .04) or frankly positive margins (P = .05).

r­ adiation therapy alone, only the inguinal nodes were treated, whereas in patients found to have positive pelvic nodes at lymphadenectomy, the inguinal and pelvic nodes were treated. Second, patients were eligible if their groins were considered clinically negative for disease. However, this ­decision was based on clinical examination only, which might have missed large inguinal metastases that would now be detected on routine tomographic scans. Third, ­patients who received radiation therapy alone were treated by means of anterior fields with the dose prescribed to a depth of 3 cm; most patients were treated with a combina­ tion of 4- to 6- MV photons and 9- to 12-MeV electrons. An analysis of the five postradiation recurrences in the GOG study demonstrated that the inguinal nodes prob­ ably received less than the prescribed dose in all of those patients, in several cases by 30% or more.96 A review of femoral anatomy in 50 other cases demonstrated that the mean depth to the femoral vessels was 6.1 cm (range, 2.0 to 8.5 cm).96 The trial by Homesley and coworkers70 demonstrated that microscopic regional disease can be controlled with radiation therapy, and several retrospective studies sug­ gest that radiation therapy can be effective in controlling ­microscopic disease in patients with clinically negative groin nodes.84,88,97 Concurrent chemoradiation therapy may be even more effective, although it should be ­ considered ­investigational in this setting.98 The GOG experience clearly demonstrates the ­importance of careful technique that ensures delivery of an ­ adequate dose to the nodes at risk.86,96 Katz and ­colleagues88 also emphasized the importance of careful technique. They reported three recurrences in 29 patients treated with radiation therapy alone for clinically nega­ tive inguinal nodes; two of the three recurrences appeared lateral to radiation fields in the only two patients whose fields were less than 17 cm wide. Tomographic imaging should always be done to rule out lymphadenopathy that may ­ require a combined-modality approach, and CT simulation should be used to plan radiation treatment. In most cases in which the inguinal lymph nodes are at risk, treatment should include at least the distal pelvic nodes (Fig. 31-2).

Prophylactic Inguinal Radiation

Anecdotal reports of dramatic responses of very advanced vulvar cancer to preoperative radiation therapy99-103 first suggested that radiation therapy might have a role in the initial management of these lesions. In several cases, no ­residual disease was found in the vulvectomy specimen ­after modest doses of radiation (40 to 55 Gy). However, interest in the use of radiation therapy as a primary treat­ ment modality for advanced vulvar cancer remained luke­ warm until studies of gastrointestinal (particularly anal) and head and neck malignancies suggested that concur­ rent chemotherapy and radiation therapy could be very effective ­ treatment for carcinomas at other sites. Results of ­prospective trials of chemoradiation therapy for cervical cancer have fueled interest in this approach.104-118 Most studies of chemoradiation therapy for locally ­advanced or recurrent vulvar cancer have used combina­ tions of cisplatin, fluorouracil, and mitomycin C with radiation therapy (Table 31-2).104-116,118 Although the

The benefit of postoperative radiation therapy document­ ed in the trial by Homesley and colleagues70 suggested that radiation might also be used to sterilize potential ­microscopic disease in clinically negative lymph nodes and avoid the morbidity of inguinal lymphadenectomy. The GOG ­attempted to test this hypothesis in a trial in which patients with clinically negative nodes were randomly assigned to receive inguinal node radiation or radical lymphadenectomy (followed by inguinopelvic radiation for patients with positive nodes) after resection of the pri­ mary tumor.86 An interim analysis, performed after entry of only 58 ­patients, demonstrated a significantly higher rate of inguinal ­recurrence and death in the irradiated group. The investigators concluded that lymphadenectomy was the superior treatment and closed the trial. The study by Homesley and colleagues70 has been criti­ cized for several reasons. First, in patients treated with

Initial Radiation Therapy for Locoregionally Advanced Disease

31  The Vulva and Vagina

A

805

B

FIGURE 31-2.  Transverse (A) and coronal (B) views of the distribution of radiation dose in a patient treated for vulvar carcinoma.  A wide 6-MV anterior field encompassed the vulva and the inguinal, external iliac, and internal iliac lymph nodes. A narrower 18-MV posterior field excluded the femoral neck and part of the femoral head. The dose to the lateral inguinal lymph nodes was supplemented using electron fields matched to the lateral borders of the posterior field. A central block in the superior portion of the field reduced the amount of small bowel and rectum in the treatment fields. A modified frog-leg position reduced the skin dose in this patient, who had a posterior lesion.

TABLE 31-2.  C  oncurrent Chemoradiation Therapy for Patients with Locally Advanced or Recurrent Carcinoma of the Vulva

Study and Reference

Number of Patients

Chemotherapy Drugs

RT Dose (Gy)

Patients with Recurrent or Persistent Local Disease after RT ± Surgery (%)

Follow-up (mo)

Moore et al113

73

5-FU + CDDP

47.6

15 (21%)

22-72

14

5-FU + CDDP

45-50

4 (29%)

7-81

Cunningham et

al105

al110

58

5-FU + Mito

54

13 (22%)

4-48

Lupi et al112

31

5-FU + Mito

54

7 (23%)

22-73

Wahlen et al118

Landoni et

19

5-FU + Mito

45-50

1 (5%)

3-70

Eifel et

al106

12

5-FU + CDDP

40-50

5 (42%)

17-30

Koh et

al109

20

5-FU ± CDDP or Mito

30-54

9 (45%)

1-75

25

5-FU ± CDDP

47-72

6 (24%)

4-52

42

Bleomycin

45

39 (93%)

7-60

12

5-FU + CDDP

44-54

0

7-60

Russell et al114 Scheistroen and Berek et

al104 al116

Trope115

24

5-FU ± Mito

44-60

10 (42%)

5-43

al107

4

5-FU + Mito

25-70

2 (50%)

20-29

Levin et al111

6

5-FU + Mito

18-60

0

1-25

Bleomycin

15-40

11 (73%)*

4

Thomas et Evans et

Iversen108

15

*Most

patients had unresectable, stage IV lesions. CDDP, cisplatin; 5-FU, fluorouracil; Mito, mitomycin C; RT, radiation therapy.

studies have been small, most investigators have been im­ pressed by relatively high complete response rates among patients with very advanced disease. No randomized trials have been done to compare radiation alone with radiation plus chemotherapy for patients with vulvar carcinoma, al­ though the GOG is studying this question in patients with regionally metastatic vulvar cancer. Although these results are encouraging, it is important to remember that the median age of patients with vul­ var cancer is 65 to 70 years. The chemoradiation therapy schedules that have been used successfully in studies of carcinomas that primarily affect younger patients may not be tolerated as well by older women. Although it is probably reasonable to consider chemoradiation therapy

for relatively robust patients with high-risk disease, the need for concurrent chemotherapy has not been rigorously proven in patients with vulvar cancer. Aggressive multimo­ dality regimens should be used with particular caution for patients with concurrent medical problems or diminished performance status.

Radiation Therapy Techniques Patients for whom regional treatment is indicated usually are treated, at least initially, through opposing anterior and posterior photon fields. Using a narrower posterior pho­ ton field and supplementing the lateral inguinal nodes with electrons (see Fig. 31-2) can minimize the dose to the femoral heads. In some cases, positioning the patient with

806

PART 9  Female Genital Tract

legs separated or in a full frog-leg position can ­reduce the reaction of the skin on the upper medial thighs. However, the anterior vulvar skin usually receives less than the target dose when the patient is in this position unless a bolus dose is applied during treatment. Thermoluminescence dosime­ try may be helpful in selected cases. The frog-leg position is difficult to reproduce on a daily basis and requires the use of immobilization devices. Port films should be obtained periodically with markers placed on the medial borders of electron fields to verify the accuracy of the ­relationship ­between these supplementary fields and the lateral ­borders of the posterior photon field. Intensity-­modulated ­radiation therapy (IMRT) may be useful in s­elected cases, ­particularly those in which the inguinal nodes are too deep to be reached with high-energy electrons. However, great care must be taken to accurately identify the target vol­ ume, particularly in the regions of the vulva, perineum, and vagina. Fiducial skin markers should be used to assist in identifying areas of involvement that are not clearly identi­ fiable on tomographic images. An appositional electron beam can be used to boost the dose to the vulva. The electron energy is usually selected by examining the patient while she is in the treatment posi­ tion. A bolus may be required if low-energy electrons are used. Although interstitial implants are occasionally used to treat deep tissues at risk for recurrence, the more homo­ geneous dose delivered to the skin surface with an electron beam usually makes this a safer approach for superficial tumors.

Management of Acute Radiation-Induced Side Effects Radical radiation therapy requires delivery of a relatively high dose of radiation to sensitive skin and mucosa. Most patients develop confluent moist desquamation by the fourth or fifth week of treatment. Despite this ­reaction, it is rarely necessary or helpful to interrupt treatment. At the beginning of treatment, patients should be warned to expect about 3 weeks of significant discomfort and should be assured that the pain usually resolves within 2 to 3 weeks after the completion of treatment. They should be instructed carefully regarding local care and should be encouraged to wear loose-fitting garments (e.g., boxer shorts, sweat pants) and avoid trauma to the skin. Sitz baths with lukewarm water or aluminum ­acetate ­ solution (Domeboro’s solution) help to keep the area clean and, combined with analgesics, to reduce ­discomfort. ­Patients who develop brisk erythema before the 15th treatment usually have a superinfection, typi­ cally caused by Candida species. Treatment of the infection should be started immediately, usually with oral agents, which tend to be better tolerated and more practical than ­topical agents.

Treatment of Metastatic Vulvar Cancer Objective responses to chemotherapy in patients with metastatic vulvar cancer have been reported but are ­uncommon; even results of phase II studies of cisplatin have been discouraging.119-121 The ability to deliver aggres­ sive chemotherapy is often compromised by patients’ poor performance status. Palliative radiation therapy and sup­ portive care to control pain and local infection can relieve the discomfort of progressive local disease.

Treatment of Vaginal Cancer Treatment of Preinvasive Disease Carcinoma in situ of the vagina (VAIN 3) can be treated with local excision if the lesion is small; larger lesions are ­often treated with laser ablation after careful colposcopic examination and biopsy sampling to rule out invasion. Rarely, extensive lesions are treated with total vaginec­ tomy and vaginal reconstruction. ICRT is also an effec­ tive treatment for extensive disease, and recurrence rates after treatment of the vaginal surface to a dose of 60 to 80 Gy (given at a low dose rate) are approximately 10% to 15%.28,30,31 Perez and colleagues31 emphasized the im­ portance of treating the entire vaginal surface to prevent marginal recurrences. Several investigators have reported using fractionated high-dose-rate intracavitary treatment for VAIN 3. However, ­because the entire vagina must be treated, conservative fractionation schedules are recom­ mended to minimize the risk of adhesive vaginitis and ­rectal complications.122,123

Treatment of Invasive Disease Radiation therapy is the treatment of choice for most ­patients with invasive vaginal cancer. The proximity of the bladder and rectum often preclude organ-sparing surgical resection of even relatively small lesions. However, early tumors of the upper posterior vagina can sometimes be treated with radical hysterectomy and partial vaginectomy or, if the patient has no uterus, radical upper vaginecto­ my.33 Radical surgery (usually pelvic exenteration) is also indicated for most patients who have had prior radiation therapy. For selected patients with stage IVA disease, pelvic exenteration with vaginal reconstruction may be the treat­ ment of choice, particularly if a rectovaginal or vesicovagi­ nal fistula is present.34,124,125 Although radiation therapy can be quite effective for invasive vaginal cancer, optimal management of this rare disease requires a detailed understanding of the relevant anatomy and the natural history of vaginal cancer. Radia­ tion oncologists must be skilled in the selection and use of a variety of external beam radiation therapy (EBRT) and brachytherapy techniques. Some of the factors that must be considered before treatment is selected are whether the tumor is located in the proximal, middle, or distal vagina; whether the tumor is on the anterior, posterior, or lateral wall; the thickness of the lesion; the size of the lesion; wheth­ er the patient has had a hysterectomy; whether the lesion has been partially or grossly resected and the location of any suspect margins; whether the lesion is fixed to the pelvic wall or has invaded the bladder, urethra, or rectum; the pres­ ence of regional metastases; and the response to initial EBRT. As a result, treatment tends to be highly individualized. Several investigators have suggested that small, very ­superficial tumors can be treated with ICRT alone.28,31 This approach may be appropriate in certain highly s­elected cases. However, in a recent review of the experience at The University of Texas M.  D. Anderson Cancer ­Center, Frank and others29 questioned the wisdom of this approach. Of the nine patients in the study who were treated with brachytherapy alone for very early squamous cell carcino­ ma, three experienced recurrence in the pelvis, for a pelvic disease control rate of only 67%. The investigators consid­ ered this to be an unacceptable result and recommended

31  The Vulva and Vagina

that part of the treatment for invasive cancer be delivered by EBRT. However, it may not be necessary to treat the whole pelvis in patients with superficial tumors; Frank and colleagues found no pelvic recurrences in 11 patients who received local EBRT to the lower pelvis only combined with brachytherapy.29 The location and thickness of the lesion are important to consider in planning brachytherapy. For patients who have had a hysterectomy, an examination under anesthe­ sia, MRI scanning, or vaginal sonography may be needed to determine the thickness of apical lesions. For lesions of the middle or lower third of the vagina, the preferred treatment is a combination of intracavitary and interstitial brachytherapy with a single-plane implant to increase the total tumor dose to 80 to 85 Gy. The use of interstitial therapy, even for small superficial lesions, usually reduces the total vaginal dose and the ­volume of rectum treated to a high dose. If the lesion is palpable and particularly if it is more than 2 to 3 mm thick, the pelvic nodes should be treated ­because the risk of regional disease is significant and the likelihood of successful salvage therapy after recurrence is small. For tumors confined to the proximal half of the ­ vagina, ­external fields should include the true pelvis and the tumor (marked with radiopaque seeds) plus a distal margin of at least 4 cm. It is usually best to obtain two sets of treatment simulation scans, one with the patient’s ­bladder full and the other empty, to determine the ­maximum ­excursion of the vagina. My colleagues and I have found that the position of

807

the vaginal apex can shift by as much as 4 cm with changes in bladder and rectal filling. For this reason, an integrated target volume that encompasses the full excursion of the vaginal and paravaginal tissues should be used for planning. Treatment can be delivered by means of anteroposteriorposteroanterior fields or four fields. If the posterior vagi­ nal wall is involved, the entire sacrum should be included to cover the perirectal nodes. If the tumor approaches or ­involves the distal third of the vagina, the medial ingui­ nal nodes should be included in the field. More generous ­inguinal coverage may be indicated if inguinal metastases are present or if the lesion is deeply invasive and is centered in the distal vagina. If the introitus is included in the field, the morbidity of treatment can be markedly reduced by ­placing the patient in a frog-leg position. Brachytherapy should be tailored to the volume and distribution of the tumor and its response to EBRT. Apical tumors that are less than 3 to 4 mm thick after EBRT may be treated with a vaginal cylinder. For more distal tumors, a combination of intracavitary and interstitial brachytherapy, which reduces the total vaginal dose and the volume of rectum treated to a high dose, may be more appropriate. When the uterus is intact, tumors high in the posterior ­fornix can often be treated with a tandem and ovoids. Interstitial implants can be inserted freehand (Fig. 31-3) or with an interstitial template. Freehand placement of implants permits better control of the position of needles but requires considerable experience. Vaginal implants also can be positioned by using a perineal template. The parallel

B

40 cGy/hr 50 60 70 70 60 50 40

A

C

FIGURE 31-3.  Interstitial implant in a patient with squamous carcinoma involving the vagina and perineal vulva. The implant was used to boost the radiation dose to a 1-cm area of residual thickening after an excellent response to concurrent chemotherapy and external beam radiation. Because of its proximity to the anus and rectum, the tumor was not believed to be amenable to surgical resection. A, Five needles were inserted transperineally using a freehand technique. B, A 3-cm-diameter cylinder and a small amount of packing were placed in the vagina to displace uninvolved vagina. C, Dose distribution in a transverse plane through the center of the implant.

808

PART 9  Female Genital Tract

needle positions acquired with templates produce a very homogeneous dose, but the template interferes somewhat with the brachytherapist’s ability to appreciate the posi­ tion of the needles relative to the tumor and other struc­ tures. In the case of tumors involving the vaginal apex in patients who have had a hysterectomy, implants can often be effectively placed by using a template with laparoscopic or laparotomy guidance. Although brachytherapy is an important tool in the treatment of vaginal cancer, it should be used only if the entire tumor can be encompassed with an adequate mar­ gin.29,31,126,127 Even very small tumors will probably recur if a portion of the tumor receives less than 60 to 70 Gy. Apical tumors that have indistinct margins, deeply invasive tumors that involve the rectovaginal septum, and tumors that are close to the iliac vessels are particularly difficult to cover with brachytherapy. In some cases, tumors such as these may be better treated with EBRT alone using carefully designed shrinking fields.28,29,31,126,127 Chyle and colleagues28 reported three vaginal recurrences (11%) in 28 patients with stage II disease treated with EBRT alone, compared with 12 recurrences (21%) in 58 patients treated with combined EBRT and brachytherapy. Perez and col­ leagues31 reported local recurrence rates of 34% to 44% for patients with stage II disease treated in most cases with a combination of EBRT and brachytherapy. Three-dimensional conformal radiation therapy or IMRT techniques can improve the ability to deliver high doses to the tumor with EBRT alone. My colleagues and I have found IMRT to be particularly helpful for treating tumors that extensively involve the periurethral space, ­infiltrate the entire vagina, or involve the rectovaginal septum. However, great care must be taken to consider the ­ influence of internal organ motion and the complex ­regional lymphatic drainage of the vagina in designing the target volumes. Efforts to correlate outcome with the total dose of radiation delivered to the tumor have yielded mixed ­results.29,31,123,127 This may be because the tumor’s size, ­extent, and initial response to radiation often influence the dose of radiation and the ability to deliver brachytherapy. When good brachytherapy coverage of the tumor can be ­accomplished, an effort should be made to treat the tu­ mor to a dose of 75 to 85 Gy. When brachytherapy is not possible, patients may be treated with EBRT alone using shrinking pelvic fields to deliver a tumor dose of 60 to 66 Gy. With the use of IMRT, careful immobilization, at­ tention to organ motion, and setup verification, the gross target volume can sometimes be taken to an effective total dose of 70 Gy. If the gross target is treated concurrently with regional lymph nodes, the range in dose per fraction of radiation is an important factor, and the volume and ­effective dose delivered to critical structures must be care­ fully considered in the design of IMRT plans in this region. The brachytherapy and EBRT techniques used to treat these rare tumors are highly specialized; radiation oncolo­ gists who do not feel comfortable with these treatments should consider referring patients to a specialist. Treatment should be completed in less than 6 to 7 weeks.126 Overall, 5-year disease-specific survival rates for ­patients with invasive vaginal cancer treated with definitive radia­ tion range from 75% to 95% for stage I disease, 50% to 80% for stage II disease, 30% to 60% for stage III disease,

and 15% to 50% for stage IVA disease.28-31,126,127 Primary adenocarcinomas of the vagina, although rare, tend to have a poorer prognosis than squamous cell carcinomas. Frank and colleagues46 reported an overall 5-year survival rate of only 34% for 26 patients treated with radiation for ­ nonDES–related adenocarcinomas of the vagina. Probably because of the intimate association of the ­vagina with bladder and rectum, complication rates tend to be somewhat higher for vaginal cancer than for cervical cancer treated with definitive radiation and tend to corre­ late with the stage of disease.28-31,128 Frank and colleagues29 reported major complication rates of 4%, 9%, and 21% for patients treated definitively with radiation for stage I, II, or III-IVA disease, respectively; these results are similar to those reported in smaller series. Vaginal morbidity has not been systematically reported but is probably influenced by the initial tumor extent, secondary infection, the age and menopausal status of the patient, and the dose, site and volume of tissue irradiated.

Role of Chemotherapy Reports of chemoradiation therapy for patients with ­vaginal cancer are anecdotal129; it is unlikely that prospec­ tive ­ randomized trials comparing radiation alone with chemoradiation therapy will be feasible for patients with this rare form of cancer. However, because vaginal cancer is similar to cervical cancer in its epidemiology and ­natural history, it is probably reasonable to extrapolate from the randomized trials of treatment for cervical cancer130-133 about the use of concurrent cisplatin-based chemothera­ py for ­ selected patients with high-risk disease. ­ Cisplatin­containing ­ chemotherapy is sometimes used to treat ­patients with metastatic or recurrent vaginal cancer, but reported response rates to chemotherapy are poor.119,120

REFERENCES    1. Bruner DW, Lanciano R, Keegan M, et al. Vaginal stenosis and sexual function following intracavitary radiation for the treat­ ment of cervical and endometrial carcinoma. Int J Radiat Oncol Biol Phys 1993;27:825-830.    2. Frumovitz M, Sun CC, Schover LR, et al. Quality of life and sexual functioning in cervical cancer survivors. J Clin Oncol 2005;23:7428-7436.    3. Fajardo LF. Pathology of Radiation Injury. New York: Masson, 1982.    4. Eifel PJ, Levenback C, Wharton JT, et al. Time course and ­incidence of late complications in patients treated with radiation therapy for FIGO stage IB carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys 1995;32:1289-1300.    5. Robinson JW, Faris PD, Scott CB. Psychoeducational group ­increases vaginal dilation for younger women and reduces sexual fears for women of all ages with gynecological carcinoma treated with radiotherapy. Int J Radiat Oncol Biol Phys 1999;44:497-506.    6. Beller U, Benedet J, Creasman W, et al. Carcinoma of the vagina. Int J Gynaecol Obstet 2006;95(Suppl 1):S29-42.    7. Beller U, Quinn M, Benedet J, et al. Carcinoma of the vulva. Int J Gynaecol Obstet 2006;95(Suppl 1):S7-S27.    8. Creasman WT, Phillips JL, Menck HR. The National Cancer Data Base report on cancer of the vagina. Cancer 1998;83:10331040.    9. Ikenberg H, Runge M, Göppinger A, et al. Human papilloma­ virus DNA in invasive carcinoma of the vagina. Obstet Gynecol 1990;76:432-438.   10. Castle PE, Schiffman M, Bratti MC, et al. A population-based study of vaginal human papillomavirus infection in hysterect­ omized women. J Infect Dis 2004;190:458-467.

31  The Vulva and Vagina   11. Hoffman MS, DeCesare SL, Roberts WS, et al. Upper vaginec­ tomy for in situ and occult, superficially invasive carcinoma of the vagina. Am J Obstet Gynecol 1992;166:30-33.   12. Merino MJ. Vaginal cancer: the role of infectious and environ­ mental factors. Am J Obstet Gynecol 1991;165:1255-1262.   13. Crum CP, McLachlin CM, Tate JE, et al. Pathobiology of ­vulvar squamous neoplasia. Curr Opin Obstet Gynecol 1997;9: 63-69.   14. Hampl M, Sarajuuri H, Wentzensen N, et al. Effect of human papillomavirus vaccines on vulvar, vaginal, and anal intraepi­ thelial lesions and vulvar cancer. Obstet Gynecol 2006;108: 1361-1368.   15. Hørding U, Junge J, Daugaard S, et al. Vulvar squamous cell ­carcinoma and papillomaviruses: indications for two different etiologies. Gynecol Oncol 1994;52:241-246.   16. Hording U, Daugaard S, Junge J, et al. Human papilloma­ viruses and multifocal genital neoplasia. Int J Gynecol Pathol 1996;15:230-234.   17. Lee YY, Wilczanski SP, Chumakov A, et al. Carcinoma of the vulva: HPV and p53 mutations. Oncogene 1994;9: 1655-1659.   18. Herbst AL, Ulfelder H, Poskanzer DC. Adenocarcinoma of the vagina: an association of maternal stilbestrol therapy with tumor appearing in young women. N Engl J Med 1971;284:878-881.   19. Herbst AL, Anderson S, Hubby MM, et al. Risk factors for the ­ development of diethylstilbestrol associated clear cell ­adenocarcinoma: a case-control study. Am J Obstet Gynecol 1986;154:814-822.   20. Hatch EE, Palmer JR, Titus-Ernstoff L, et al. Cancer risk in ­women exposed to diethylstilbestrol in utero. JAMA 1998;280: 630-634.   21. Palmer JR, Wise LA, Hatch EE, et al. Prenatal diethylstilbestrol exposure and risk of breast cancer. Cancer Epidemiol Biomark­ ers Prev 2006;15:1509-1514.   22. Andreasson B, Nyboe J. Predictive factors with reference to low-risk of metastases in squamous cell carcinoma in the vulvar ­region. Gynecol Oncol 1985;21:196-206.   23. Plentl AA, Friedman EA. Clinical significance of the vulvar ­lymphatics. In: Lymphatic System of Female Genitalia. Philadel­ phia: WB Saunders, 1971, p 27.   24. Binder SW, Huang I, Fu YS, et al. Risk factors for the develop­ ment of lymph node metastasis in vulvar squamous carcinoma. Gynecol Oncol 1990;37:9-16.   25. DiSaia PJ, Creasman WT, Rich WM. An alternative approach to early cancer of the vulva. Am J Obstet Gynecol 1979;133: 825-832.   26. Plentl AA, Friedman EA. Lymphatics of the vagina. In: ­Lymphatic System of Female Genitalia. Philadelphia: WB Saunders, 1971, p 51-56.   27. Berek JS, Hacker NF. Practical Gynecologic Oncology. ­Baltimore: Williams & Wilkins, 1994.   28. Chyle V, Zagars GK, Wheeler JA, et al. Definitive radiotherapy for carcinoma of the vagina: outcome and prognostic factors. Int J Radiat Oncol Biol Phys 1996;35:891-905.   29. Frank SJ, Jhingran A, Levenback C, et al. Definitive radiation therapy for squamous cell carcinoma of the vagina. Int J Radiat Oncol Biol Phys 2005;62:138-147.   30. Kirkbride P, Fyles A, Rawlings GA, et al. Carcinoma of the ­vagina—experience at the Princess Margaret Hospital (1974-1989). Gynecol Oncol 1995;56:435-443.   31. Perez CA, Grigsby PW, Garipagaoglu M, et al. Factors ­affecting long-term outcome of radiation in carcinoma of the vagina. Int J Radiat Oncol Biol Phys 1999;44:37-45.   32. Plentl AA, Friedman EA. Clinical significance of the vaginal lym­ phatics. In: Lymphatic System of Female Genitalia. Philadelphia: WB Saunders, 1971, p 57-74.   33. Stock RG, Chen ASJ, Seski J. A 30-year experience in the man­ agement of primary carcinoma of the vagina: analysis of prognos­ tic factors and treatment modalities. Gynecol Oncol 1995;56: 45-52.   34. Al-Kurdi M, Monaghan JM. Thirty-two years experience in man­ agement of primary tumours of the vagina. Br J Obstet Gynecol 1981;88:1145-1150.

809

  35. Davis KP, Stanhope CR, Garton GR, et al. Invasive vagi­ nal carcinoma: analysis of early-stage disease. Gynecol Oncol 1991;42:131-136.   36. Crum CP. Carcinoma of the vulva: epidemiology and pathogen­ esis. Obstet Gynecol 1992;79:448-454.   37. Wilkinson EJ. The 1989 presidential address. International ­Society for the Study of Vulvar Disease. J Reprod Med 1990;35: 981-991.   38. Buscema J, Woodruff JD. Progressive histobiologic alterations in the development of vulvar cancer. Am J Obstet Gynecol 1980;138:146-150.   39. Koss LG. The Papanicolaou test for cervical cancer detection: a triumph and a tragedy. JAMA 1989;261:737-743.   40. Fanning J, Lambert HC, Hale TM, et al. Paget’s disease of the vulva: prevalence of associated vulvar adenocarcinoma, ­invasive Paget’s disease, and recurrence after surgical excision. Am J ­Obstet Gynecol 1999;180:24-27.   41. Isaacs JH. Verrucous carcinoma of the female genital tract. Gynecol Oncol 1976;4:259-269.   42. Japaze H, Van Dinh T, Woodruff JD. Verrucous carcinoma of the vulva: study of 24 cases. Obstet Gynecol 1982;60:462-466.   43. Gil-Moreno A, Garcia-Jimenez A, Gonzalez-Bosquet J, et al. Merkel cell carcinoma of the vulva. Gynecol Oncol 1997;64: 526-532.   44. Joseph RE, Enghardt MH, Doering DL, et al. Small cell neuroen­ docrine carcinoma of the vagina. Cancer 1992;70:784-789.   45. Miliauskas JR, Leong AS. Small cell (neuroendocrine) carcinoma of the vagina. Histopathology 1992;21:371-374.   46. Frank S, Deavers MT, Jhingran A, et al. Primary adenocarcinoma of the vagina not associated with diethylstilbestrol (DES) expo­ sure. Gynecol Oncol 2007;105:470-474.   47. Yaghsezian H, Palazzo JP, Finkel GC, et al. Primary vaginal ­adenocarcinoma of the intestinal type associated with adenosis. Gynecol Oncol 1992;45:62-65.   48. Raber G, Mempel V, Jackisch C, et al. Malignant melanoma of the vulva. Report of 89 patients. Cancer 1996;78:2353-2358.   49. 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;171:12251230.   50. Ragnarsson-Olding BK, Nilsson BR, Kanter-Lewensohn LR, et al. Malignant melanoma of the vulva in a nationwide, 25-year study of 219 Swedish females: predictors of survival. Cancer 1999;86:1285-1293.   51. Chung AF, Woodruff JM, Lewis JL Jr. Malignant melanoma of the vulva: a report of 44 cases. Obstet Gynecol 1975;45: 638-646.   52. Tasseron EWK, van der Esch EP, Hart AAM, et al. A clinico­ pathological study of 30 melanomas of the vulva. Gynecol On­ col 1992;46:170-175.   53. 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: 1177-1184.   54. Ragnarsson-Olding B, Johansson H, Rutqvist LE, et al. Malignant melanoma of the vagina. Cancer 1993;71:1893-1897.   55. Andrassy RJ, Wiener ES, Raney RB, et al. Progress in the surgi­ cal management of vaginal rhabdomyosarcoma: a 25-year review from the Intergroup Rhabdomyosarcoma Study Group. J Pediatr Surg 1999;34:731-734.   56. Bernstein SG, Kovacs BR, Townsend DE, et al. Vulvar carcinoma in situ. Obstet Gynecol 1983;61:304-307.   57. Friedrich EG Jr, Wilkinson EJ, Fu YS. Carcinoma in situ of the vulva: a continuing challenge. Am J Obstet Gynecol 1980;136: 830-843.   58. Benedet JL, Bender H, Jones H III, et al. FIGO staging classifica­ tions and clinical practice guidelines in the management of gyne­ cologic cancers. FIGO Committee on Gynecologic Oncology. Int J Gynaecol Obstet 2000;70:209-262.   59. Shepherd J, Sideri M, Benedet J, et al. Carcinoma of the vagina. J Epidemiol Biostat 1998;3:103-109.   60. Boyce J, Fruchter RG, Kasambilides E, et al. Prognostic factors in carcinoma of the vulva. Gynecol Oncol 1985;20:364-377.

810

PART 9  Female Genital Tract

  61. Heaps JM, Fu YS, Montz FJ, et al. Surgical-pathologic variables predictive of local recurrence in squamous cell carcinoma of the vulva. Gynecol Oncol 1990;38:309-314.   62. Homesley HD, Bundy BN, Sedlis A, et al. Prognostic factors for groin node metastasis in squamous cell carcinoma of the vulva (a Gynecologic Oncology Group study). Gynecol Oncol 1993;49:279-283.   63. Husseinzadeh N, Wesseler T, Schneider D, et al. Prognostic factors and the significance of cytologic grading in invasive ­squamous cell carcinoma of the vulva: a clinicopathologic study. Gynecol Oncol 1990;36:192-199.   64. Raspagliesi F, Hanozet F, Ditto A, et al. Clinical and pathologi­ cal prognostic factors in squamous cell carcinoma of the vulva. Gynecol Oncol 2006;102:333-337.   65. Kürzl R, Messerer D. Prognostic factors in squamous cell car­ cinoma of the vulva: a multivariate analysis. Gynecol Oncol 1989;32:143-150.   66. Pinto AP, Signorello LB, Crum CP, et al. Squamous cell carci­ noma of the vulva in Brazil: prognostic importance of host and viral variables. Gynecol Oncol 1999;74:61-67.   67. Sedlis A, Homesley H, Bundy BN, et al. Positive groin lymph nodes in superficial squamous cell vulvar cancer. A Gyneco­ logic Oncology Group study. Am J Obstet Gynecol 1987;156: 1159-1164.   68. Homesley HD, Bundy BN, Sedlis A, 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;164:997-1003.   69. Smyczek-Gargya B, Volz B, Geppert M, et al. A multivari­ ate analysis of clinical and morphological prognostic factors in squamous cell carcinoma of the vulva. Gynecol Obstet Invest 1997;43:261-267.   70. Homesley HD, Bundy BN, Sedlis A, et al. Radiation therapy ­versus pelvic node resection for carcinoma of the vulva with positive groin nodes. Obstet Gynecol 1986;68:733-740.   71. Origoni M, Sideri M, Garsia S, et al. Prognostic value of path­ ological patterns of lymph node positivity in squamous cell ­carcinoma of the vulva stage III and IVA FIGO. Gynecol Oncol 1992;45:313-316.   72. van der Velden J, Lindert ACM, Lammes FB, et al. Extracapsular growth of lymph node metastases in squamous cell carcinoma of the vulva. The impact on recurrence and survival. Cancer 1995;75:2885-2890.   73. Kucera H, Vavra N. Radiation management of primary carcino­ ma of the vagina: clinical and histopathological variables associ­ ated with survival. Gynecol Oncol 1991;40:12-16.   74. Vavra N, Seifert M, Kucera H, et al. Radiotherapy of primary vag­ inal carcinoma and effects of histological and clinical factors on the prognosis [in German]. Strahlenther Onkol 1991;167:1-6.   75. Kuppers V, Stiller M, Somville T, et al. Risk factors for recurrent VIN. Role of multifocality and grade of disease. J Reprod Med 1997;42:140-144.   76. Hacker NF. Vulvar cancer. In Berek JS, Hacker NF (eds): Prac­ tical Gynecologic Oncology, 3rd ed. Philadelphia: Lippincott ­Williams & Wilkins, 2005, pp 543-584.   77. Hacker NF, Leuchter RS, Berek JS, et al. Radical vulvectomy and bilateral inguinal lymphadenectomy through separate groin incisions. Obstet Gynecol 1981;58:574-579.   78. Burke TW, Stringer CA, Gershenson DM, et al. Radical wide excision and selective inguinal node dissection for squamous cell carcinoma of the vulva. Gynecol Oncol 1990;38:328-332.   79. Farias-Eisner R, Cirisano FD, Grouse D, et al. Conservative and individualized surgery for early squamous carcinoma of the ­vulva: the treatment of choice for stage I and II (T1-2 N0-1 M0) disease. Gynecol Oncol 1994;53:55-58.   80. Hacker NF, Berek JS, Lagasse LD, et al. Individualization of treatment for stage I squamous cell vulvar carcinoma. Obstet Gynecol 1984;63:155-162.   81. Magrina JF, Gonzalez-Bosquet J, Weaver AL, et al. Primary ­squamous cell cancer of the vulva: radical versus modified ­radical vulvar surgery. Gynecol Oncol 1998;71:116-121.

  82. Hacker NF, Berek JS, Lagasse LD, et al. Management of regional lymph nodes and their prognostic influence in vulvar cancer. ­Obstet Gynecol 1983;61:408-412.   83. Morley GW. Infiltrative carcinoma of the vulva: results of surgi­ cal treatment. Am J Obstet Gynecol 1976;124:874-888.   84. Petereit DG, Mehta MP, Buchler DA, et al. A retrospective review of nodal treatment for vulvar cancer. Am J Clin Oncol 1993;16:38-42.   85. Podratz KC, Symmonds RE, Taylor WF. Carcinoma of the vulva: analysis of treatment failures. Am J Obstet Gynecol 1982;143:340-351.   86. Stehman FB, Bundy BN, Thomas G, et al. Groin dissection ver­ sus groin radiation in carcinoma of the vulva: a ­ Gynecologic ­Oncology Group study. Int J Radiat Oncol Biol Phys 1992;24:389-396.   87. Burke TW, Levenback C, Coleman RL, et al. Surgical therapy of T1 and T2 vulvar carcinoma: further experience with radical wide excision and selective inguinal lymphadenectomy. Gynecol Oncol 1995;57:215-220.   88. Katz A, Eifel PJ, Jhingran A, et al. The role of radiation ther­ apy in preventing regional recurrences of invasive squamous cell ­ carcinoma of the vulva. Int J Radiat Oncol Biol Phys 2003;57:409-418.   89. Stehman FB, Bundy BN, Dvoretsky PM, et al. Early stage I car­ cinoma of the vulva treated with ipsilateral superficial inguinal lymphadenectomy and modified radical hemivulvectomy: a prospective study of the Gynecologic Oncology Group. Obstet Gynecol 1992;79:490-497.   90. Levenback C, Morris M, Burke TW, et al. Groin dissection prac­ tices among gynecologic oncologists treating early vulvar cancer. Gynecol Oncol 1996;62:73-77.   91. Micheletti L, Preti M, Zola P, et al. A proposed glossary of ter­ minology related to the surgical treatment of vulvar carcinoma. Cancer 1998;83:1369-1375.   92. Oonk MH, Hollema H, de Hullu JA, et al. Prediction of lymph node metastases in vulvar cancer: a review. Int J Gynecol Cancer 2006;16:963-971.   93. Kunos C, Simpkins F, Gibbons H, et al. Radiation therapy ver­ sus pelvic node resection for carcinoma of the vulva with posi­ tive groin nodes: an update of a Gynecologic Oncology study [abstract 5]. Gynecol Oncol 2008;108:S4.   94. Faul C, Mirmow D, Gerszten K, et al. Isolated local recurrence in carcinoma of the vulva: prognosis and implications for treat­ ment. Int J Gynecol Cancer 1998;8:409-414.   95. Faul CM, Mirmow D, Huang Q, et al. Adjuvant radiation for vulvar carcinoma: improved local control. Int J Radiat Oncol Biol Phys 1997;38:381-389.   96. Koh WJ, Chiu M, Stelzer KJ, et al. Femoral vessel depth and the implications for groin node radiation. Int J Radiat Oncol Biol Phys 1993;27:969-974.   97. Henderson RH, Parsons JT, Morgan L, et al. Elective ilioinguinal lymph node radiation. Int J Radiat Oncol Biol Phys 1984;10:811819.   98. Leiserowitz GS, Russell AH, Kinney WK, et al. Prophylactic chemoradiation of inguinofemoral lymph nodes in patients with locally extensive vulvar cancer. Gynecol Oncol 1997;66:509514.   99. Acosta AA, Given FT, Frazier AB, et al. Preoperative radiation therapy in the management of squamous cell carcinoma of the vulva: preliminary report. Am J Obstet Gynecol 1978;132: 198-206. 100. Boronow RC. Combined therapy as an alternative to exentera­ tion for locally advanced vulvo-vaginal cancer: rationale and ­results. Cancer 1982;49:1085-1091. 101. Fairey RN, MacKay PA, Benedet JL, et al. Radiation treatment of carcinoma of the vulva, 1950-1980. Am J Obstet Gynecol 1985;151:591-597. 102. Hacker NF, Berek JS, Juillard GJF, et al. Preoperative radiation therapy for locally advanced vulvar cancer. Cancer 1984;54: 2056-2061. 103. Jafari K, Magalotti M. Radiation therapy in carcinoma of the vulva. Cancer 1981;47:686-691.

31  The Vulva and Vagina 104. Berek JS, Heaps JM, Fu YS, et al. Concurrent cisplatin and 5-fluorouracil chemotherapy and radiation therapy for ­advanced-stage squamous carcinoma of the vulva. Gynecol ­Oncol 1991;42:197-201. 105. Cunningham MJ, Goyer RP, Gibbons SK, et al. Primary radia­ tion, cisplatin, and 5-fluorouracil for advanced squamous carci­ noma of the vulva. Gynecol Oncol 1997;66:258-261. 106. Eifel PJ, Morris M, Burke TW, et al. Prolonged continuous ­infusion cisplatinum and 5-fluorouracil with radiation for locally advanced or recurrent carcinoma of the vulva. Gynecol Oncol 1995;59:51-56. 107. Evans LS, Kersh CR, Constable WC, et al. Concomitant 5-fluorouracil, mitomycin-C, and radiotherapy for advanced ­gynecologic malignancies. Int J Radiat Oncol Biol Phys 1988;15:901-906. 108. Iversen T. Radiation and bleomycin in the treatment of inop­ erable vulval carcinoma. Acta Obstet Gynecol Scand 1982;61: 195-197. 109. Koh WJ, Wallace HJ, Greer BE, et al. Combined radiotherapy and chemotherapy in the management of local-regionally ­advanced vulvar cancer. Int J Radiat Oncol Biol Phys 1993;26:809-816. 110. Landoni F, Maneo A, Zanetta G, et al. Concurrent preopera­ tive chemotherapy with 5-fluorouracil and mitomycin C and radiotherapy (FUMIR) followed by limited surgery in locally advanced and recurrent vulvar carcinoma. Gynecol Oncol 1996;61:321-327. 111. Levin W, Goldberg G, Altaras M, et al. The use of concomitant chemotherapy and radiotherapy prior to surgery in advanced stage carcinoma of the vulva. Gynecol Oncol 1986;25:20-25. 112. Lupi G, Raspagliesi F, Zucali R, et al. Combined preopera­ tive chemoradiotherapy followed by radical surgery in locally advanced vulvar carcinoma. A pilot study. Cancer 1996;77: 1472-1478. 113. Moore DH, Thomas GM, Montana GS, et al. Preoperative chemoradiation for advanced vulvar cancer: a phase II study of the Gynecologic Oncology Group. Int J Radiat Oncol Biol Phys 1998;42:79-85. 114. Russell AH, Mesic JB, Scudder SA, et al. Synchronous radiation and cytotoxic chemotherapy for locally advanced or recurrent squamous cancer of the vulva. Gynecol Oncol 1992;47:14-20. 115. Scheistroen M, Trope C. Combined bleomycin and radiation in preoperative treatment of advanced squamous cell carcinoma of the vulva. Acta Oncol 1992;32:657. 116. Thomas G, Dembo A, DePetrillo A, et al. Concurrent radia­ tion and chemotherapy in vulvar carcinoma. Gynecol Oncol 1989;34:263-267. 117. van Doorn HC, Ansink A, Verhaar-Langereis M, et al. Neoad­ juvant chemoradiation for advanced primary vulvar cancer. ­Cochrane Database Syst Rev 2006;(3):CD003752. 118. Wahlen SA, Slater JD, Wagner RJ, et al. Concurrent radiation therapy and chemotherapy in the treatment of primary ­squamous cell carcinoma of the vulva. Cancer 1995;75:2289-2294. 119. Slayton R, Blessing J, Beecham J, et al. Phase II trial of etopo­ side in the management of advanced or recurrent squamous cell ­carcinoma of the vulva and carcinoma of the vagina. Cancer Treatment Reports 1987;71:869-870.

811

120. Thigpen JT, Blessing JA, Homesley H, 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;23:358-363. 121. Wagenaar HC, Colombo N, Vergote I, et al. Bleomycin, meth­ otrexate, and CCNU in locally advanced or recurrent, inop­ erable, squamous-cell carcinoma of the vulva: an EORTC ­Gynaecological Cancer Cooperative Group Study. European Organisation for Research and Treatment of Cancer. Gynecol Oncol 2001;81:348-354. 122. MacLeod C, Fowler A, Dalrymple C, et al. High-dose-rate brachytherapy in the management of high-grade intraepithelial neoplasia of the vagina. Gynecol Oncol 1997;65:74-77. 123. Ogino I, Kitamura T, Okajima H, et al. High-dose-rate int­ racavitary brachytherapy in the management of cervical and vaginal intraepithelial neoplasia. Int J Radiat Oncol Biol Phys 1998;40:881-887. 124. Benson C, Soisson AP, Carlson J, et al. Neovaginal reconstruc­ tion with a rectus abdominis myocutaneous flap. Obstet Gyne­ col 1993;81:871-875. 125. Berek JS, Hacker NF, Lagasse LD. Rectosigmoid colectomy and reanastomosis to facilitate resection of primary and recurrent gynecologic cancer. Obstet Gynecol 1984;64:715-720. 126. Lee WR, Marcus RB, Sombeck MD, et al. Radiotherapy alone for carcinoma of the vagina: the importance of overall treatment time. Int J Radiat Oncol Biol Phys 1994;29:983-988. 127. Spirtos NM, Doshi BP, Kapp DS, et al. Radiation therapy for primary squamous cell carcinoma of the vagina: Stanford Uni­ versity experience. Gynecol Oncol 1989;35:20-29. 128. Eddy GL, Jenrette JM, Creasman WT. Effect of radiotherapeutic technique on local control in primary vaginal carcinoma. Int J Gynecol Cancer 1993;3:399-404. 129. Dalrymple JL, Russell AH, Lee SW, et al. Chemoradiation for primary invasive squamous carcinoma of the vagina. Int J Gyne­ col Cancer 2004;14:110-117. 130. Eifel PJ, Winter K, Morris M, et al. Pelvic radiation with con­ current chemotherapy versus pelvic and para-aortic ­ radiation for high-risk cervical cancer: an update of Radiation Therapy ­Oncology Group trial (RTOG) 90-01. J Clin Oncol 2004;22: 872-880. 131. Rose PG, Bundy BN. Chemoradiation for locally advanced cervi­ cal cancer: does it help? J Clin Oncol 2002;20:891-893. 132. Pearcey R, Brundage M, Drouin P, et al. A clinical trial compar­ ing concurrent cisplatin and radiation therapy versus radiation alone for locally advanced squamous cell carcinoma of the cervix carried out by the National Cancer Institute of Canada Clinical Trials Group [abstract]. Proc Am Soc Clin Oncol 2000;19:378. 133. Whitney CW, Sause W, Bundy BN, et al. A randomized com­ parison of fluorouracil plus cisplatin versus hydroxyurea as an adjunct to radiation therapy in stages IIB–IVA carcinoma of the cervix with negative para-aortic lymph nodes: a Gynecologic Oncology Group and Southwest Oncology Group study. J Clin Oncol 1999;17:1339-1348.

32 The Ovary Anuja Jhingran and Gillian m. Thomas Prognostic Factors Clinical Evaluation Treatment Management of Borderline Epithelial Ovarian Tumors GRANULOSA CELL TUMORS DYSGERMINOMA

ANATOMY AND PHYSIOLOGY RESPONSE OF THE NORMAL OVARY TO IRRADIATION CARCINOMA OF THE OVARY Epidemiology Pathology Patterns of Spread Surgical Staging

Anatomy and Physiology The ovaries are almond-shaped glands located close to the lateral pelvic walls and suspended by the mesovarium of the broad ligament of the uterus. In prepubertal females, the surface of the ovary is covered by a smooth layer of ovarian surface epithelium that gives the surface a dull, grayish appearance, contrasting with the shiny surface of the adjacent peritoneal mesovarium with which it is continuous. After puberty, the surface becomes progressively scarred and distorted because of the repeated rupture of ovarian follicles and discharge of oocytes that are part of ovulation. Maturation of ovarian follicles results in the production of female hormones and ova. At puberty, the ovaries contain about 300,000 primordial follicles. These primitive follicles, 20 μm in diameter, consist of an oocyte arrested in prophase of the first meiotic division, surrounded by a single layer of specialized stromal cells, the granulosa cells (Fig. 32-1). During each menstrual cycle, under the influence of follicle-­stimulating hormone and luteinizing hormone, several hundred follicles develop.1 In each of these follicles, the granulosa cells and surrounding theca cells multiply, leading to formation of the ­graafian follicle, which reaches 2 cm in ­diameter before ovulation. The granulosa and theca cells are responsible for the cyclic production of estradiol and progesterone.2 ­During each menstrual cycle, the oocyte in the dominant follicle is stimulated to complete meiotic division and ­ mature into the ovum, which is shed at ovulation. The unsuccessful follicles and their oocytes degenerate and are reabsorbed. When the ovaries become depleted of primordial follicles, as occurs naturally at menopause or as a result of radiation therapy or alkylating chemotherapy, ­ ovulation and production of estradiol and progesterone cease. ­Because proliferation of the granulosa and theca cells depends on the presence of oocytes, estradiol production in females ­ depends on oogenesis to a far greater degree than ­testosterone production in males depends on spermatogenesis (see Chapter 27). The oocytes in primitive follicles are more sensitive to radiation damage than are the ova in mature follicles; the oocytes in graafian follicles may also be more radiosensitive than ova.

812

Response of the Normal Ovary to Irradiation The radiation dose necessary to induce ovarian failure depends on the age at exposure.3 Doses in excess of 5 Gy (5 to 10 Gy) are required to produce permanent cessation of menses in more than 95% of women younger than 40 years; a dose of 3.75 Gy produces amenorrhea in ­almost 100% of women older than 40 years.4 Doses of 1.25 to 5 Gy produced permanent amenorrhea in 14 (58%) of 24 women younger than 40 years, whereas doses of 1.25 to 3.75 Gy produced permanent amenorrhea in 18 (78%) of 23 women older than 40 years.4 Cole1 reported having used radiation-induced ovarian ablation for the treatment of breast cancer in 235 women. Usually, a single ­treatment of

Granulosa cells Primordial follicle Oocyte

Nonspecialized stroma

Theca cells

Ovum

Granulosa cells Basement membrane

Graafian (preovulatory) follicle Surface epithelium (with coelomic or germinal epithelium)

FIGURE 32-1.  Schematic diagram of a section of the ovary shows the functional structures and the tissues from which neoplasms arise.

32  The Ovary

4.5 Gy was administered to the pelvis. Permanent amenorrhea was induced in 67% (39 of 58) of the women younger than 40 years, 90% (70 of 78) of the women between 40 and 44 years old, 96% (71 of 74) of the women between 45 and 49 years old, and 100% (25 of 25) of the women older than 49 years. Stillman and colleagues2 reported the effects on ovarian function of abdominal irradiation for childhood malignancies. Most of the 182 girls in that study had been prepubertal at diagnosis, but all were more than 14 years old at the time of the study and had been followed for a minimum of 6 years. Primary or secondary amenorrhea and elevated pituitary gonadotropin levels occurred in 68% (17 of 25) of the girls who had both ovaries included in the radiation field (estimated ovarian doses of 12 to 50 Gy); in 14% (5 of 35) of the girls who had at least one ovary at the edge of the radiation field (estimated ovarian doses of 0.9 to 10 Gy); and in none (0 of 34) of the girls who had at least one ovary outside the treatment field (estimated ovarian doses of 0.05 to 1.5 Gy) or those (0 of 88) who had not received abdominal irradiation.2 These findings indicate that ovarian function can be ­ablated with low doses of radiation. However, in young women, oogenesis is not quite as sensitive to internal scatter radiation as is spermatogenesis in men, and ovarian function can often be preserved by shielding the ovaries from the primary radiation beam or, as is done in the treatment of Hodgkin disease and some cases of cervical carcinoma, by oophoropexy to move the ovaries away from the radiation field. When nonsurgical ablation of ovarian estrogen production is therapeutically desirable, as in the management of breast cancer, it can be effected in almost 100% of women with a dose of 10 to 15 Gy given in 4 or 5 fractions. An opposing pair of fields (15 cm wide × 10 cm high) is used, with the inferior border 1 cm below the top of the pubic symphysis. Cancer therapy in young women (i.e., surgery, radiation therapy, chemotherapy, or any combination of these) may result in unavoidable permanent ablation of ovarian function. Provided the tumor is not hormonally responsive, hormone replacement therapy is often of benefit in reversing menopausal symptoms and reducing the degree of ­osteoporosis and cardiovascular disease that may ensue.

Carcinoma of the Ovary A wide variety of tumors can affect the ovary. A working classification of ovarian neoplasms is presented in Box 32-1. This chapter emphasizes malignant tumors that can be treated successfully with radiation therapy or a combination of radiation therapy and other treatment.

Epidemiology In the United States, ovarian cancer is the leading cause of cancer death from gynecologic malignancies and is the fifth most common cause of cancer death in women.5 The approximate number of new cases of ovarian carcinoma in the United States in 2008 was 21,650, accounting for 3% of all cancers in women; the estimated number of deaths from ovarian cancer that year, 15,520, accounted for 6% of all cancer-related deaths in women.5 The median age at diagnosis is 63 years.6

813

Epidemiologic studies have identified several factors important in the carcinogenesis of ovarian cancer. Numerous studies have identified parity, multiple births, history of breast-feeding, and use of oral contraceptives as being protective against the development of ovarian cancer.7-9 The relationships between each of these factors and ovarian cancer incidence support the hypothesis that the development of ovarian cancer is related to “incessant ovulation”10,11 or prolonged exposure to estrogens. It is postulated that ovarian cancer may develop from an aberrant repair process of the surface epithelium during each ovulation cycle; if this is the case, factors that result in a decreased number of ovulatory cycles may reduce the risk of ovarian cancer. An individual patient’s risk of developing ovarian cancer also depends on familial and genetic factors. The lifetime risk for a woman in the general population is 1.6%, but this risk rises to approximately 5% if a single other first-degree family member is affected and to 7% if two relatives are ­affected.12 Familial ovarian cancer has been linked to at least three distinct genotypes. At least 10% of all epithelial ovarian cancers are thought to be hereditary, with mutations in the BRCA genes accounting for approximately 90%

BOX 32-1.  Working Classification of Ovarian Neoplasms* Epithelial Tumors (85% to 95%) Types 1. Serous (tubal type, accounts for approximately 50% of epithelial carcinomas) 2. Mucinous (cervical type, 10%) 3. Endometrioid (endometrial type, 20%) 4. Clear cell (4%) 5. Unclassified/undifferentiated (15%) Differentiation 1. Benign: cystadenoma 2. Borderline malignancy/low malignant potential (applies mainly to serous and mucinous types) 3. Malignant: cystadenocarcinoma, grades 1, 2, and 3 Stromal Tumors Specialized Stroma: Sex Cord Stromal Tumors (4% to 8%) 1. Granulosa cell–theca cell (estrogen-producing, about 3% to 6% of malignant ovarian tumors) 2. Sertoli-Leydig cell (virilizing, rare) Nonspecialized Stroma: Mixed Mesodermal Tumor, ­Lymphoma, Leiomyosarcoma, and others Germ Cell Tumors (2% to 4%) Dysgerminoma (1% to 2%) Embryonic Differentiation 1. Benign cystic teratoma 2. Malignant teratoma (AFP, β-HCG negative) Extraembryonic Differentiation 1. Endodermal sinus tumor/embryonal carcinoma (AFP positive) 2. Choriocarcinoma (rare, β-HCG positive) Secondary (Metastatic) Carcinomas *Percentages are the approximate frequency of primary ­malignancies in adults. AFP, alpha-fetoprotein; β-HCG, β-human chorionic gonadotropin.

814

PART 9  Female Genital Tract

of those cases and most of the remaining 10% attributable to hereditary nonpolyposis colorectal cancer (also known as Lynch syndrome).13,14 For women with a mutation in BRCA1, the risk of developing ovarian carcinoma by the age of 70 ranges from 35% to 60%.15,16 That risk is lower for women with a mutation in BRCA2, but it is still 10% to 27%. The median age for developing ovarian cancer is 54 years for BRCA1 mutation carriers and 62 years for BRCA2 mutation carriers.17 The risk of developing ovarian cancer is ­approximately 12% for women with Lynch syndrome.18

Pathology Epithelial tumors account for 90% of all ovarian malignancies. Two thirds of epithelial ovarian tumors, which span a range from borderline malignancies through markedly undifferentiated tumors, are frankly malignant. The borderline malignancies have cytologic features of malignancy and can metastasize, although they are characterized by an absence of destructive stromal invasion. The most common type of epithelial ovarian cancer arises from the ovarian surface epithelium. As the epithelium becomes malignant, it exhibits a variety of müllerian-type differentiations. The types of epithelial tumors are listed in Box 32-1 in order of decreasing frequency. In cystadenocarcinomas, the ­degree of histologic differentiation (grade) is an important predictor of the tumor’s clinical behavior, metastatic ­ potential, and lethality; this is particularly true of serous tumors. The endometrioid variant of ovarian carcinoma is associated with endometriosis in about 20% to 30% of cases, with a separate endometrioid uterine cancer (often stage I and low grade) simultaneously present in 15% of cases. Tumors with clear cell histology also may be associated with endometriosis and are more resistant to chemotherapy than are their more common serous counterparts.

Patterns of Spread Epithelial ovarian tumors arise from the surface epithelium and eventually penetrate the ovarian capsule. After the ovarian capsule is penetrated, malignant cells seem to exfoliate into the peritoneal cavity, where they follow the normal circulation of the peritoneal fluid and may implant and grow on any surface in the peritoneal cavity. Tumor cells can also obstruct lymphatics, resulting in ascites. Growth of ovarian cancer in the peritoneal cavity may cause matted loops of bowel, resulting in adhesions and mechanical obstruction. Lymphatic spread occurs most commonly to the paraaortic and pelvic nodes. The incidence of nodal metastases seems to be related to the stage of disease. Most patients with advanced ovarian cancer have involvement of pelvic and para-aortic nodes.19 Only about 15% of patients have extra-abdominal disease at the time of first relapse; in most patients, recurrent or persistent disease is intra-abdominal.

Surgical Staging Staging of ovarian cancer is based on findings at surgical exploration and pathologic examination of the surgical specimens. The 2002 revisions of the American Joint Committee on Cancer (AJCC) and International Federation of Gynecology and Obstetrics (FIGO) staging systems are shown in Box 32-2.20 Because postoperative therapy is often based on the disease stage and other prognostic ­factors, meticulous surgical staging is recommended, including (when possible) a vertical abdominal incision, ­ multiple

cytologic washings, intact tumor removal, complete ­abdominal ­exploration, removal of the ovaries, uterus and tubes, omentectomy, lymph node sampling, and random peritoneal biopsies. When surgical staging is incomplete, the extent of disease can be significantly underestimated: in a study of systematic restaging in 100 patients believed to have stage I or II disease after initial surgical staging, 31 patients (31%) were found to have more advanced disease than was initially diagnosed, and 23 (77%) of those 31 ­patients were found to have stage III disease.21 In an era when most patients with ovarian cancer are treated with combination chemotherapy regardless of the stage or extent of disease, it could be argued that the detection of occult or microscopic residual disease would not change treatment recommendations for individual patients. However, careful surgical staging is an essential part of maximal removal of tumor, an important aspect of therapy and prognosis.

Prognostic Factors The most important independent prognostic factors in ovarian cancer are tumor stage, volume of residual disease, tumor grade, histologic subtype, patient age, and the presence or absence of malignant ascites.22 Tumor stage remains the strongest prognostic factor. The 5-year survival rate for patients with stage IA disease and grade 1 or 2 histology is more than 90% after surgery alone, and many investigators include patients with stage IB, grade 1 or 2 disease in this good-prognosis group.23 However, the relapse rate for patients with stage IC or stage II disease who do not undergo postoperative therapy is 30% to 40%, and the 5-year survival rate for patients with disease at these stages is 80%, regardless of whether they receive adjuvant therapy.23 Most patients present with stage III or stage IV disease, for which the survival rates are considerably lower. After surgery, the 5-year survival rate for patients with stage III or stage IVA disease that is optimally debulked (i.e., any sites of gross residual disease are 1 cm or less in diameter) is approximately 20% to 30%, and this rate drops to less than 10% if the disease is not optimally debulked.24,25 A multivariate analysis of patients treated at the ­University of Toronto has confirmed that in addition to ­tumor stage and amount of residual tumor remaining after surgery, tumor grade and patient age are important independent prognostic factors, whereas histopathologic subtype is less important and may be correlated with other prognostic factors. A multifactorial classification system for patients with small-volume or no macroscopic residual disease has been developed and validated in sequential patient cohorts treated at the University of Toronto (Figs. 32-2 and 32-3).26-30 In that population, a combination of stage, ­residuum, and tumor grade clearly stratified patients into subgroups with widely disparate outcomes after treatment with radiation therapy. This type of classification has since been validated by other investigators.31-33 As for other possible prognostic factors, preoperative cancer antigen (CA) 125 levels usually reflect the volume of disease and do not seem to have an independent effect on survival. However, postoperative CA 125 levels during and after completion of first-line chemotherapy do seem to have some prognostic value.34,35 Several investigators have demonstrated that normalization of serum CA 125 levels after three cycles of chemotherapy is associated with more

32  The Ovary

815

BOX 32-2.  American Joint Committee on Cancer and International Federation of Gynecology and Obstetrics Staging Systems for Cancer of the Ovary AJCC

FIGO

Primary Tumor (T) TX — T0 — T1 I T1a IA T1b

IB

T1c

IC

T2 T2a

II

IIA

T2b

IIB

T2c T3

IIC III

T3a T3b T3c

IIIA IIIB IIIC

M1

IV

Primary tumor cannot be assessed No evidence of primary tumor Tumor limited to 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 ruptured, tumor on ovarian surface, malignant cells in ascites or peritoneal washings Tumor involves one or both ovaries with pelvic extension and/or implants Extension and/or implants on uterus and/or tube(s). No malignant cells in ascites or peritoneal washings Extension to and/or implants on other pelvic tissues. No malignant cells in ascites or peritoneal washings Pelvic extension and/or implants (T2a or T2b) with malignant cells in ascites or peritoneal washings Tumor involves one or both ovaries with microscopically confirmed peritoneal metastasis outside the pelvis† Microscopic peritoneal metastasis beyond pelvis (no macroscopic tumor) 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)�

Regional Lymph Nodes (N) Regional lymph node(s) cannot be assessed NX N0 No regional lymph node metastasis N1 IIIC Regional lymph node metastasis Distant Metastasis (M) MX M0 M1 IV

Distant metastasis cannot be assessed No distant metastasis Distant metastasis (excludes peritoneal metastasis)

Stage Grouping Stage I T1 Stage IA T1a Stage IB T1b Stage IC T1c Stage II T2 Stage IIA T2a Stage IIB T2b Stage IIC T2c Stage III T3 Stage IIIA T3a Stage IIIB T3b Stage IIIC T3c Any T Stage IV Any T

N0 N0 N0 N0 N0 N0 N0 N0 N0 N0 N0 N0 N1 Any N

M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M1

From Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 275-283. *The presence of nonmalignant ascites is not classified. The presence of ascites does not affect staging unless malignant cells are present. †Liver capsule metastases are T3/stage III; liver parenchymal metastasis, M1/stage IV. Pleural effusion must have positive cytology for M1/ stage IV.

favorable outcome, as is the achievement of a CA 125 nadir level of 10 U/mL or less on completion of treatment.34,35

Clinical Evaluation The presenting symptoms of women with ovarian cancer are often vague and indistinct. The most common symptoms are abdominal discomfort, abdominal pain, and ­abdominal distention. Gastrointestinal symptoms are relatively common and may include a change in bowel habits,

nausea or dyspepsia, or early satiety. Unfortunately, 75% to 85% of patients have disease that is disseminated throughout the peritoneal cavity at presentation. Because reliable procedures for the early detection of ovarian cancer have yet to be found,36-38 early-stage disease is usually diagnosed after detection of an asymptomatic adnexal mass on routine pelvic examination. However, most adnexal masses, particularly in premenopausal ­women, are not ­malignant.

816

PART 9  Female Genital Tract STAGE

100 90 80 70 60 50

AGE

(79%)

I (135) P < 0.0005 II (224) (57%)

40

(52%) < 50 (265)

(39%) Clear (35) (35%) Serous (397)

(31%) > 50 (553)

(21%) 20

(61%) Mucinous (93) (51%) Endometrioid (163)

P <.0005

P <.0005

30

Percent actuarial survival

PATHOLOGY

III (350) P <.0005

IV 2.6% at 5 years (109)

10

(14%) (121) Undifferentiated 0

Well differentiated (196) P <.0005

5

6

GRADE (73%) 1 (196) P <.0005 (39%) 2 (244) P <.0005

Moderate to poor differentiation (612)

20

4

Small (142)

(27%)

30

3

(68%) 0/? (312) P <.0005 (40%)

(73%)

40

2

RESIDUUM

GRADE

100 90 80 70 60 50

1

P <.0005

(20%) 3 (331)

(11%) Large (364)

10 0

1

2

3

4

5

6

0

1

2

3

4

5

6

0

1

2

3

4

5

6

Years after diagnosis

FIGURE 32-2.  Survival curves for 818 patients with epithelial carcinoma of the ovary treated at the Princess Margaret Hospital between 1971 and 1978 are classified according to disease stage, age at diagnosis, histologic subtype, degree of differentiation, amount of residual disease (residuum), and grade. The percentage in parentheses above each curve is the 5-year survival rate; the number in parentheses after each curve is the number of patients in the corresponding group.

Stage

Residuum

I

0

II

0

II

< 2 cm

III

0

III

< 2 cm

Grade 1

Grade 2

Grade 3

Low risk

Intermediate risk

High risk

FIGURE 32-3.  Prognostic subgroupings according to tumor stage, residuum, and grade for patients with stages I to III ovarian cancer with little or no residual disease. Abdominopelvic irradiation is recommended as the sole postoperative treatment for patients in the intermediate-risk group. (Modified from Carey MS, Dembo AJ, Fyles AW, et al. Testing the validity of a prognostic classification in patients with surgically optimal ovarian carcinoma: a 15-year review. Int J Gynecol Cancer 1993;3:24-35.)

Abdominal ultrasonography, computed tomography, and magnetic resonance imaging are often used to evaluate early and advanced disease. Although imaging may aid in ruling out the possibility of hepatic metastases, the extent of intraperitoneal disease is best evaluated at surgery.

The most useful serum marker for ovarian cancer is CA 125. Between 80% and 85% of patients with epithelial ovarian cancer have elevated CA 125 levels (>35 U/mL).39 More than 85% of patients with the serous subtype of ­epithelial ovarian cancer have increased CA 125 levels, but

32  The Ovary

elevations are less common among those with mucinous tumors. For postmenopausal women with asymptomatic pelvic masses, a CA 125 level of greater than 65 U/mL has a sensitivity of 97% and a specificity of 78% in the ­diagnosis of ovarian cancer.40 However, in premenopausal women, nonmalignant conditions, including endometriosis, fibroids, and pelvic inflammatory disease, can produce ­ elevations of serum CA 125, lowering the specificity of the test. The finding of an elevated CA 125 level in patients older than 50 years suggests epithelial ovarian cancer and should prompt surgical exploration.

Treatment In the past, treatment for ovarian cancer was selected on the basis of tumor stage and clinical-pathologic features. Postsurgical adjuvant therapy included radiation therapy of the pelvis or whole abdomen, intraperitoneal colloid therapy, chemotherapy, administration of biological response modifiers, or combinations of these treatments. Most ­patients with ovarian cancer in the United States currently are treated with primary cytoreductive surgery and, when adjuvant therapy is indicated, combination chemotherapy. However, radiation therapy remains useful in the treatment of certain patient subgroups, as is discussed throughout the “Treatment” section of this chapter.

Surgery Except in certain subsets of patients with stage I disease, the standard surgical approach to the management of epithelial ovarian cancer is a staging laparotomy with a bilateral salpingo-oophorectomy, a hysterectomy, and surgical cytoreduction of intraperitoneal tumor masses. At many centers, routine para-aortic and pelvic lymph node dissection is performed if this procedure, along with optimal debulking, can render the patient free of residual macroscopic disease. The volume of tumor remaining at the completion of the surgery and the number of residual masses are significant prognostic factors for outcome (see “Prognostic Factors”). Most surgeons accept that attempts should be made to minimize the tumor residuum at the time of surgery. It is recognized that the amount of residual disease should be treated as a continuum for prognostic purposes, and most investigators define the optimal residuum as the largest diameter of the largest residual lesion being less than 1 cm. Hoskins and colleagues41 studied the frequency with which optimal primary cytoreduction could be achieved. On average, optimal cytoreduction could be achieved in only about one third of patients.41 The ultimate value of primary cytoreduction must be interpreted cautiously ­because even reduction to 1 g of residual tumor leaves ­approximately 109 tumor cells. The most effective chemotherapy available usually reduces the number of residual cells by about three logs. The number of residual cells is increased by any tumor proliferation that occurs during chemotherapy and by any resistance that develops as the result of mutations as the tumor ages. Cure is unlikely even if all gross tumor is removed. The better outcome for patients who undergo optimal cytoreductive surgery was previously attributed to the therapeutic effect of surgery, but it could be that patients in whom optimal cytoreductive surgery is possible have biologically less aggressive or more responsive disease. However, because cure is rarely

817

achieved for patients who have a residuum greater than 1 cm, it is usually worthwhile to attempt optimal cytoreduction before treatment is begun, in the hope that a smaller tumor burden can increase the chance of cure. Among patients with optimal cytoreduction, long-term survival is better for those having no macroscopic residual disease than it is for those having small-volume macroscopic residual disease.

Adjuvant Therapy for Early-Stage Disease Only 10% to 15% of patients with epithelial ovarian cancer are diagnosed with stage I or II disease after careful staging laparotomy. These patients have a much better prognosis than do patients with advanced disease, and they constitute approximately one third of the patients with epithelial ovarian cancer who are eventually cured. Several methodologic factors have hampered a precise understanding of the possible contribution of adjuvant therapy to the treatment of patients with stage I or II disease: the relative infrequency of the presentation, underestimates of disease extent during the time when laparotomies were performed less carefully, and lack of understanding of prognostic factors for early-stage disease. Because the prognosis for patients with stage I or II disease is relatively good, conducting phase III randomized, controlled trials of postoperative adjuvant therapy in this subgroup is hampered by the anticipated low event rate and the small numbers of patients. No controlled study has established a curative advantage for any form of postoperative therapy. Decisions about postoperative therapy for patients with stage I or II disease therefore are based not on established benefit but rather on the predicted risk of relapse for an individual ­patient, as determined by pathologic predictive factors and the physician’s practice. Patients with stage I disease with grade 1 (or sometimes grade 2) differentiation have a 5-year survival rate in excess of 90%.23,28 Findings from several studies23,42 support the current practice of advising that no postoperative adjuvant therapy be given to this subgroup. For patients with stage I, grade 2 or 3 disease, the risk of relapse may be as high as 30%. The results of two randomized trials from Europe indicated that platinum-based adjuvant chemotherapy could improve overall and recurrence-free survival rates at 5 years for patients with high-risk early-stage ovarian carcinoma.43 This same group of investigators, in a prospective, unblinded, randomized phase III trial to test the efficacy of adjuvant chemotherapy for patients with early-stage ovarian cancer, found that the benefit of adjuvant chemotherapy seemed to be limited to patients who had undergone less than optimal surgical staging—in other words, for patients at higher risk for unappreciated residual disease.44 However, most physicians favor treating high-risk, earlystage disease with adjuvant chemotherapy. The optimal number of cycles of therapy is still under debate. A study by the Gynecologic Oncology Group (GOG) to compare the ­ effectiveness of three cycles versus six cycles of carboplatin and paclitaxel45 unfortunately was underpowered to answer that question. Currently, six cycles is considered standard if adjuvant chemotherapy is to be given. External Beam Radiation Therapy. Clinical trials have been undertaken to evaluate the impact of postoperative radiation therapy on the survival of patients with earlystage epithelial ovarian cancer. In a GOG study,46 patients

818

PART 9  Female Genital Tract

with stage I disease were randomly assigned after surgery to undergo observation, pelvic external beam radiation therapy (EBRT), or chemotherapy with melphalan. The results of this study were interpreted as inconclusive because surgical staging was not required and significant numbers of patients were eliminated from the analysis.46 A study in Toronto compared postoperative pelvic irradiation with observation for patients with stage I or II or asymptomatic stage III disease47 and failed to demonstrate the superiority of either approach; in that study, as was true for the GOG study, comprehensive staging was not mandated.47 Pelvic irradiation did reduce the number of pelvic relapses, but because of an increased rate of recurrence in the upper abdomen in this subgroup, it did not improve survival.47 No study has rigorously examined the role of abdominopelvic irradiation in stage I disease. A retrospective comparison of pelvic radiation therapy and no treatment48 showed a statistically insignificant reduction in relapse risk with radiation only for patients with grade 2 or 3 disease and for patients with densely adherent tumors—a subset of patients in whom disease may be more correctly classified and treated as stage II. Intraperitoneal Radiocolloid Therapy. Because ovarian cancer tends to spread through the coelom, the intraperitoneal instillation of radiocolloids, which theoretically deliver high doses of radiation to the peritoneal surfaces, is intuitively attractive for the treatment of early-stage disease. Colloids labeled with radioactive isotopes of phosphorus (32P) or gold (198Au) have been used. 198Au is no longer available for use in North America because it poses a greater radiation hazard to people who come in contact with treated patients. No randomized trials have examined the benefits of 32P versus observation only for early-stage disease; however, several randomized trials have compared 32P with various chemotherapeutic agents (Table 32-1).49 When 32P was compared with single-agent cisplatin in a study conducted at the Norwegian Radium Hospital,50 5-year disease-free survival rates for patients with stage I disease were 86% for patients treated with 32P and 83%

for patients treated with cisplatin. A small subgroup of patients with adhesions too extensive to permit treatment with 32P were treated with whole-abdomen irradiation, and the 5-year disease-free survival rate for that subgroup was 95%. Because the incidence of late bowel obstructions requiring surgery was higher in the group treated with 32P (4%) than in the cisplatin group, the investigators concluded that cisplatin was the treatment of choice despite the fact that no benefit was found with respect to median survival time, time to relapse, or proportion experiencing disease relapse. A similar study performed in Italy42 compared 32P with single-agent cisplatin for patients with stage IC ovarian cancer. The disease-free survival interval was longer in the cisplatin group; however, no difference was evident in 5-year overall survival rates (81% for cisplatin versus 79% for 32P). The GOG conducted a trial to compare 32P with three cycles of cisplatin and cyclophosphamide for patients with unfavorable early-stage disease. No differences in survival were observed, but the investigators concluded that chemotherapy was the preferred alternative because 32P was associated with a higher incidence of bowel toxicity.51 A therapeutic value for 32P has not been established. Because of the variable and unpredictable radiation dose distribution to peritoneal surfaces and the lack of penetration of this pure beta-emitter into tissues, it is unclear that 32P would provide adequate radiation therapy to the entire peritoneal cavity or adequately penetrate the retroperitoneal nodes. If radiation therapy is to be used in ovarian cancer, it is better delivered with EBRT techniques, which provide a uniform and predictable radiation dose to the peritoneal cavity and adequately irradiate the entire ­abdominal contents. General Recommendations. The optimal postoperative treatment for high-risk early-stage epithelial ovarian cancer remains to be established. Appropriate treatment options at this time include pelvic irradiation, platinumbased chemotherapy, and paclitaxel-based chemotherapy.

TABLE 32-1. Five-Year Disease-Free and Overall Survival Rates in Prospective, Randomized Trials Comparing 32P Radiocolloid Therapy with Various Chemotherapy Agents in Patients with Early-Stage Ovarian Cancer Series

Criteria for High Risk

Number of ­Patients

5-Year Disease-Free Survival (%)

5-Year Overall Survival (%)

National Cancer ­Institute of Canada

IA, IB, G3; IC-III, OpD

257

Pelvic RT + 32P vs. pelvic RT + ­melphalan vs. whole-abdomen RT

—

66

— —

61 62

Gynecologic Oncology Group

IA, IB, G3; IC-II OpD

141

32P

vs. melphalan

80 80

78 81

Norwegian Radium Hospital

I-III OpD

340

Cisplatin vs. 32P

75 81

— —

Italian Interregional Cooperative Group of Gynaecologic Oncology

IC

161

Cisplatin vs. 32P

95 65

81 79

Study Design

OpD, optimally debulked; 32P, radioactive isotope of phosphorus; RT, radiation therapy. Modified from Cardenes H, Randall ME. Integrating radiation therapy in the curative management of ovarian cancer: current issues and future directions. Semin Radiat Oncol 2000;10:61-70.

32  The Ovary

Because no form of adjuvant therapy has been shown to provide a survival advantage compared with observation, observation alone can also be considered; however, many physicians believe that the available evidence supports six cycles of chemotherapy for patients with high-risk earlystage disease.

Postoperative Whole-Abdomen and Pelvis Irradiation Identification of several chemotherapeutic agents with significant activity against advanced ovarian cancer has led to extremely limited use of radiation as definitive postoperative therapy for epithelial ovarian cancer. Nevertheless, the evidence is incontrovertible for the efficacy of radiation therapy for certain subsets of patients. Extensive experience with the platinum compounds, discovered in the late 1970s, and with paclitaxel, introduced in the 1990s, has demonstrated that these drugs can prolong survival times. However, it is unclear whether the drugs produce significant improvements in survival rates. Most patients treated with platinum- or paclitaxel-based chemotherapy still experience disease relapse that leads to death. Response rates to radiation therapy, even in ­chemotherapyresistant ovarian cancer, are very high.52-55 Although most physicians avoid recommending radiation therapy because of concern that it may impair bone marrow reserve (a crucial issue for patients being treated with hematotoxic chemotherapy agents), given the efficacy of radiation therapy, it seems premature to discard this modality from the definitive treatment armamentarium for ovarian cancer. The discussion that follows reviews the evidence supporting the contribution of radiation therapy to the cure of ovarian cancer and suggests specific patient subgroups in whom definitive postoperative adjuvant irradiation of the whole abdomen plus pelvis (hereafter referred to as abdominopelvic irradiation) may have efficacy equivalent to that of chemotherapy.56,57 Evidence of Efficacy. In ovarian cancer, the dominant route of dissemination is throughout the abdominal cavity, and tumors tend to remain confined to the abdominal cavity for extended periods; at first relapse, regardless of the type of therapy received, tumor is confined to the abdominal cavity in approximately 85% of patients.58,59 In radiation therapy for ovarian cancer, techniques that ­encompass

819

the entire peritoneal cavity rather than just the pelvis or lower abdomen are likely to have the greatest ­ curative ­potential. Several studies have compared abdominopelvic irradiation with pelvic radiation therapy alone or pelvic radiation therapy combined with single-agent alkylating chemotherapy.57,60 These studies have demonstrated a superior outcome with abdominopelvic irradiation for patients with minimal residual disease after primary surgery. Substantial evidence is available from five trials to ­support the curative potential of abdominopelvic irradiation in patients with advanced ovarian cancer (Table 32-2).27,31,57,61-63 Many retrospective analyses of outcomes in patients treated with abdominopelvic irradiation include patients in whom no residual disease was apparent. In these series, some patients were undoubtedly cured by surgery alone, and the magnitude of additional benefit conferred by the radiation therapy is not clear. However, all patients included in the trials summarized in Table 32-2 had known macroscopic residual disease at the time of irradiation. The cure rates were determined by the stage at presentation and the volume of residual disease after surgery. For the four studies that included follow-up of more than 10 years (which should permit accurate estimates of the true cure rate because 90% of recurrences develop within 5 years of treatment), the outcomes were similar: Between 38% and 62% of patients with residual disease smaller than 2 cm were cured after abdominopelvic irradiation. Among ­patients with stage II disease, the cure rate was higher for patients in whom the pelvis was treated to a higher dose than was the upper abdomen. For patients with residual disease in excess of 2 cm, the probability of cure ranged from 0% to 14%. Although these studies provide strong evidence that abdominopelvic irradiation can cure small-volume residual ovarian cancer, certain questions have been raised about interpretation of the results. Because aggressive cytoreductive surgery and comprehensive staging were not used in many of these studies, the presence and volume of residual disease might have been underestimated or overesti­mated. However, the finding of similar long-term disease-free survival rates for patients with known small-volume macroscopic residual disease after surgery constitutes strong ­evidence that radiation therapy can be curative for selected patients with ovarian cancer.

TABLE 32-2.  Outcomes after Abdominopelvic Irradiation for Patients with Residual Disease after Surgery Volume of Residual Disease before Irradiation Study and Reference Dembo,

198527

Martinez et al, Fuller et al,

198562

198757

Study End Point

<2 cm

>2 cm

10-year relapse-free survival rate (%)

38

6

15-year failure-free survival rate (%)

50

14

10-year relapse-free survival rate (%)

62

0

Wesier et al, 198863

10-year overall survival rate (%)

42

10

Goldberg and Peschel, 198831

Surviving fraction at 6 years

0.41

—*

*No patient had residual disease measuring more than 2 cm. Modified from Ozols RF, Rubin SC, Thomas G, et al. Epithelial ovarian cancer. In Hoskins WJ, Perez CA, Young RC (eds): Principles and Practice of Gynecologic Oncology, 2nd ed. Philadelphia: Lippincott-Raven, 1997, pp 919-986.

820

PART 9  Female Genital Tract

Selection of Patients. During the past 4 decades, through studies begun at the Princess Margaret Hospital and The University of Texas M. D. Anderson Cancer Center, patients with all stages and extents of ovarian cancer have been treated with abdominopelvic irradiation.47,64,65 Using data from these studies, Dembo and colleagues28 performed a detailed multivariate analysis of pathologic prognostic factors and defined the patient and tumor factors that predicted favorable outcome after the administration of this treatment. The investigators classified patients with small-volume or no residual ovarian cancer in three groups defined in terms of the risk of recurrence after surgery alone: low, intermediate, and high risk (Fig. 32-4; see also Fig. 32-3).28 Classification of patients in these risk groups has been validated by three other groups of investigators in New Haven,29 Denmark,32 and Germany.33 The low-risk group is composed of patients with stage I, grade 1 ovarian cancer without evidence of dense adherence or ascites. Multifactorial analysis of prognostic factors determined that these patients have such a good survival rate after surgery alone (96% ± 2%) that no postoperative therapy is warranted.28 In patients with stage II or III disease, the amount and site of residual disease and the tumor grade are strong determinants of the success of abdominopelvic irradiation. Given the restrictions on the dose that can be safely delivered to the upper abdomen and pelvis, abdominopelvic irradiation should be used only for patients with no macroscopic disease in the upper abdomen and with small-volume macroscopic (0 to 2 cm) residual disease in the pelvis. The intermediate-risk group consists of patients with stage I disease with grade 2 or 3 lesions or dense adherence or ascites; those with stage II disease, except grade 3 lesions with residual tumor after surgery; and those with stage III, grade 1 disease with no macroscopic residual tumor or with macroscopic residual tumor measuring less than 2 cm and located in the pelvis. The patients in the ­intermediaterisk category constitute about one third of the total ­patient population with ovarian cancer. It is in this intermediate-risk group that abdominopelvic irradiation is most

Stage

Residuum

a­ ppropriate as the sole postoperative treatment method. With the use of this approach, more than 67% of intermediate-risk patients were alive and disease free 10 years after ­treatment, with minimal late morbidity.66,67 Patients in the high-risk group (see Fig. 32-3) have a 10-year recurrence-free rate of approximately 20% after treatment with abdominopelvic irradiation alone after surgery.67 Although patients in this high-risk group usually are treated with cisplatin- or paclitaxel-based chemotherapy, one study has suggested that among patients with highrisk disease who have undergone optimal cytoreduction, cisplatin-based combination chemotherapy followed by abdominopelvic irradiation could significantly improve median survival times and relapse-free survival rates over those that have been achieved among historical controls treated with radiation therapy alone.68 A study from Australia reported in 199769 examined the outcome of 58 patients fulfilling Dembo’s definition of intermediate risk.28 These patients were treated with abdominopelvic irradiation after cytoreductive surgery, and their reported relapse-free and overall survival rates—52% and 55% at 5 years—were similar to those reported by Dembo and colleagues. The consistency of these reports supports the conclusion of a significant curative benefit from abdominopelvic irradiation for patients defined as having intermediate-risk disease according to this prognostic classification system. Radiation therapy should therefore be considered a rational choice of therapy for a large proportion of patients with optimally cytoreduced ovarian cancer. Treatment Technique. The irradiated volume and the total dose are critical determinants of the therapeutic benefit of abdominopelvic irradiation. Treatment must include the entire peritoneal cavity. Originally, coverage of the peritoneal cavity was accomplished by using a “moving strip” technique developed at M. D. Anderson Cancer Center.70,71 The open-field technique is currently preferred because it is simpler, takes less time (6 versus 10 to 13 weeks), and is associated with less morbidity.62,70,72 Two randomized studies compared the moving strip and

Grade 1

Grade 2

Grade 3

I

0

96 ± 2% (n = 80)

78 ± 5 (71)

62 ± 8 (39)

II

0

91 ± 4 (45)

73 ± 7 (46)

52 ± 7 (47)

II

< 2 cm

No relapses (5)

78 ± 14 (9)

21 ± 11 (14)

III

0

63 ± 14 (15)

26 ± 14 (12)

29 ± 11 (20)

III

< 2 cm

88 ± 12 (8)

45 ± 11 (20)

39 ± 10 (27)

FIGURE 32-4.  Five-year relapse-free survival rates (± standard deviation) for patients with ovarian cancer according to tumor stage, residuum, and grade. The numbers in parentheses indicate the number of patients in each cell. Low-risk subgroups are indicated by gray-blue boxes; intermediate-risk subgroups are indicated by blue boxes; and high-risk subgroups are indicated by yellow boxes. (Fourteen patients without specific grade assignments were excluded.) (Modified from Carey MS, Dembo AJ, Fyles AW, et al. Testing the validity of a prognostic classification in patients with surgically optimal ovarian carcinoma: a 15-year review. Int J Gynecol Cancer 1993;3:24-35.)

32  The Ovary

open-field techniques and showed them to be equally effective with respect to tumor control.72,73 Parallel opposing portals with a beam energy sufficient to ensure dosage variation of no more than 5% should be used. Simulator and portal films are required to ensure that the margins are outside the peritoneal reflection laterally and that the treatment portal is above the level of the diaphragm on expiration. The field margins should be outside the iliac wings and below the level of the obturator foramina to ensure adequate coverage of the pelvic floor. Preliminary results from a randomized study by the ­National Cancer Institute of Canada suggest that survival is significantly reduced when the peritoneal surfaces are inadequately encompassed by the radiation field. An analysis of several series in which radiation therapy was administered to patients with stage I to III ovarian cancer also showed the prognostic importance of the adequacy of peritoneal coverage.74 Tumor doses limited to 22 to 30 Gy result in acceptable rates of side effects. Treatment usually is delivered in daily fractions of 1 to 1.2 Gy, with both fields treated each day. Appropriate renal shielding posteriorly should be used to limit the renal dose to 18 to 20 Gy. A boost dose usually is given to the pelvis in 1.8- to 2-Gy fractions to bring the total pelvic dose to 45 to 50 Gy. However, it may be best to limit this practice to patients with residual disease confined to the pelvis; in patients with residual macroscopic disease at other sites, boost doses to the sites of distant metastases may be most appropriate. Some investigators have advocated routinely delivering a boost dose to the para-aortic lymph nodes and the medial domes of the diaphragm,65 although this technique has not been used routinely at the University of Toronto. One randomized study75 compared two doses of abdominopelvic irradiation for patients with intermediate-risk ovarian cancer: 22.5 Gy in 22 fractions and 27.5 Gy in 27 fractions. No significant difference in overall survival rate was found between the two groups (83% in the low-dose group and 72% in the high-dose group). Similarly, no difference was found in patterns of relapse, hematologic complications, or late complications between the two groups. At least within this dose range, no advantage to using the higher dose of radiation is evident. Within this dose range, hepatic shielding is not required. Opponents of the use of abdominopelvic irradiation for ovarian cancer have argued that the doses that can be used safely are ineffective in sterilizing residual disease. ­Although Fletcher’s seminal work suggested that 50 Gy is required to produce a 90% probability of control of microscopic residual disease,76 later radiobiologic modeling77 based on data from the clinical literature about multiple tumor sites has demonstrated that the dose-response curve for control of subclinical metastases of ovarian cancer that are present after surgery is different from that previously accepted. The model suggests that although microscopic disease was previously assumed to imply a tumor burden of 108 or 109 residual clonogenic cells in all patients, in all probability, the residual tumor burden is heterogeneous, ranging from a single cell to 108 or 109 cells, with many patients having residual tumor burdens near the small end of the range. One implication of this model is that in many patients with lower residual tumor burdens, disease control and reduction in the risk of recurrence can be achieved

821

with radiation doses lower than 45 to 50 Gy. Available data regarding abdominopelvic irradiation for control of subclinical ovarian cancer indicate that doses of 22 to 25 Gy in 1-Gy fractions can reduce the rate of metastasis by 38% to 40%.57,78,79 The model of heterogeneity of residual tumor cell burden in ovarian cancer helps to explain the efficacy and limitations of conventional radiation therapy for ovarian cancer. It explains why the radiation doses used, although lower than the 50-Gy dose once believed necessary for disease control, have led to cure in a substantial proportion of patients. Because tolerance limits the dose that can be delivered to the upper abdomen compared with the pelvis, only patients with microscopic upper abdominal disease and a small macroscopic pelvic residuum may be cured with abdominopelvic irradiation. Side Effects. Abdominopelvic irradiation is associated with acute and chronic side effects. The acute effects usually resolve within 2 to 3 weeks after completion of treatment. Patients commonly report fatigue, which increases during the course of radiation therapy. Gastrointestinal side effects are also common; most patients experience some degree of nausea or anorexia. Fortunately, nausea is often well relieved by antinausea agents such as ondansetron. Approximately 75% of patients experience diarrhea with or without abdominal cramping. Acute hematologic toxicity is modest. In a randomized trial from the University of Toronto75 in which patients were assigned to receive one of two abdominal doses of radiation—22.5 Gy in 22 fractions or 27.5 Gy in 27 fractions—five patients in the lowdose group and seven in the high-dose group had white blood cell counts of less than 2 × 109/L at the completion of ­radiation therapy, and three patients in each group had grade 3 thrombocytopenia (Table 32-3).75 A review of almost 600 patients treated with abdominopelvic irradiation80 showed that 10% could not complete treatment, with the most commonly cited reason being ­myelosuppression. Radiation therapy should be ­interrupted for thrombocytopenia (platelet count <50,000/μL [50 × 109/L]) or neutropenia (neutrophil count <1000/μL [1 × 109/L]). The late effects of abdominopelvic irradiation include asymptomatic basal pneumonitis or fibrosis, which is ­detectable in 15% to 20% of patients on radiography. This effect results from the necessity of including 1 to 2 cm of the inferior lung field in the radiation portal to ensure adequate coverage of the hemidiaphragm. Eleven of 125 patients in the University of Toronto randomized study of two doses of radiation had radiographic evidence of pneumonitis, although only 2 patients, both in the high-dose group, had associated symptoms.75 With upper abdominal doses of 22 to 25 Gy in 22 to 25 fractions, about 50% of ­patients will develop transient elevation of the alkaline phosphatase level a few months after radiation therapy. A rising alkaline phosphatase level in this setting does not usually signify recurrence in the liver.27 Late gastrointestinal complications are the complications of greatest concern to most investigators. The frequency and severity of late gastrointestinal side effects depend on the total dose of radiation, the dose per fraction, and the extent and number of previous operations.81,82 The risk of late complications seems to be increased if lymph node sampling was performed as part of the initial operation.83

822

PART 9  Female Genital Tract

TABLE 32-3.   Side Effects of Abdominopelvic Irradiation in the University of Toronto Randomized Trial of LowDose (22.5 Gy) versus High-Dose (27.5 Gy) Treatment Low-Dose Group

High-Dose Group

White blood cell count, median (range)

2.7 (1.1-4.1)

2.6 (1.4-5.2)

Platelet count, median (range)

134 (46-224)

121 (43-268)

13

10

2

0

32

29

0

1

7

4

Grade 1

2

0

Grade 3

1

1

Side Effect Acute Effects*

Late Effects† Bowel Grade 1 or 2 Grade 3 Liver Grade 1 or 2 Grade 3 Lung Grade 1 or 2 Bladder

*Reported as the count × 109/L. †Reported as the number of patients. Modified from Fyles AW, Thomas GM, Pintilie M, et al. A randomized study of two doses of abdominopelvic radiation therapy for patients with optimally debulked stage I, II, and III ovarian cancer. Int J Radiat Oncol Biol Phys 1998;41:543-549.

The risk of serious bowel complications can be minimized if the technical principles outlined in the section on “Treatment Technique” earlier in this chapter are followed. About 10% to 15% of patients report some diarrhea or persistent bloating related to particular dietary intolerance, but frank malabsorption is rare. Obstruction requiring surgical correction is much less common than is ­ usually ­appreciated. In four studies including a total of almost 1100 patients, 5.6% of the patients required bowel surgery for late radiation-related complications (the rates for the individual studies ranged from 1.4% to 14%).27,31,84,85 In these collected series, four patients (<0.5%) died as a result of radiation-induced bowel damage. In an update of these results,86 investigators presented 25-year findings on outcome after sequential abdominal radiation therapy and melphalan chemotherapy. The actuarial rate of gastrointestinal toxicity was 14% and remained unchanged from 2 years onward. The likelihood of small bowel complications was related to the number of prior abdominal ­operations, a thin physique, and hypertension.86 The use of intensity-modulated radiation therapy may help to reduce toxicity and may allow some dose escalation, especially around the diaphragmatic area.87,88

No cases have been reported of second malignancy as a result of abdominopelvic irradiation. This is in distinct contrast to the findings of Travis and colleagues89 regarding the risk of leukemia among patients with ovarian cancer treated with platinum-based chemotherapy. The relative risk of leukemia among 28,971 patients who received platinumbased chemotherapy was 4.0 (95% confidence interval, 1.4 to 11.4). The relative risks for treatment with carboplatin and cisplatin were 6.5 and 3.3, respectively.

Whole-Abdomen and Pelvis Irradiation versus Chemotherapy The relative effectiveness of abdominopelvic irradiation and combination platinum-based chemotherapy as adjuvant treatment for patients with small-volume or no macroscopic residual ovarian cancer after surgery is often ­debated. A review of the current literature does not resolve which of these two treatments is preferable. The GOG in the United States and the National Cancer Institute of Canada have conducted studies to compare abdominopelvic irradiation and platinum-based chemotherapy. Both studies failed to accrue sufficient numbers of patients to permit valid comparisons of the two treatments, probably because of physician biases and the disparate ­nature of the treatments. Numerous nonrandomized studies of radiation therapy or chemotherapy for ovarian cancer have been reported, but for many reasons, valid comparisons between these studies are not possible. The radiation therapy series tend to be older and tend not to have included rigorous staging or aggressive cytoreductive surgery, in contrast to the practice in the later chemotherapy studies. In the radiation therapy studies, the extent of disease might have been underestimated, and the efficacy of irradiation therefore might have been underestimated. However, patients in the radiation therapy series who had minimal macroscopic ­residual disease after surgery that did not include extensive cytoreduction might have had a more favorable prognosis than patients in the chemotherapy series who had minimal macroscopic disease after more aggressive debulking surgery. Further complicating comparisons of the radiation therapy and chemotherapy series is the introduction of newer diagnostic techniques and more accurate staging methods, which could have changed the stage distribution of ovarian cancer. Long-term overall survival rates and progression-free survival rates have been reported for patients with advanced ovarian cancer treated with cisplatin-based chemotherapy. In the largest reported series, the disease-free survival rate at 4 years was 35%.90 Presumably, 10% to 15% more patients would experience relapse with longer ­follow-up. Overall survival rates reported for women given abdominopelvic irradiation for optimal stage III disease are approximately 30% to 35%.28 These findings seem to ­indicate that radiation therapy and cisplatin-based chemotherapy produce similar outcomes in patients with ovarian cancer, but given the limitations outlined in the preceding paragraph, this conclusion is problematic. Radiation therapy is effective in the management of ovarian cancer, but when radiation therapy is used as the primary postoperative therapy, optimal results require strict adherence to the criteria for selecting patients and the technical aspects of radiation delivery. The similarity

32  The Ovary

823

TABLE 32-4.   Results of Consolidation or Salvage Radiation Therapy for Patients with No or Only Microscopic Residual Disease after Surgery and Chemotherapy Number Surviving/Number at Risk Study and Reference

No Residual Disease (Consolidation)

Microscopic (<5 mm) Residual Disease (Salvage)

Chiara et al, 199198

6/9

5/11

Greiner et al,

198499

14/15

5/9*

Reddy et al,

1989100

—

4/5

Kuten et al,

1988101

5/5

12/18

Solomon et al, 1988102

—

4/12

Kong et al, 198897

—

3/5

4/4

3/4

14/23

—

1/4

10/15

Rosen et al, Haie et al,

1986103

1989104

Kersh et al,

1988105

Goldhirsch et al,

1988106

19/24

—

Menczer et al, 1986107

9/10†

—

Morgan et al, 198896

—

8/8*

*A 1-cm residuum. †Result interpreted as negative by authors. From Thomas GM. Is there a role for consolidation or salvage radiotherapy after chemotherapy in advanced epithelial ovarian cancer? Gynecol Oncol 1993;51:97-103.

of the long-term survival rates for patients treated with platinum-containing chemotherapy and patients treated with radiation therapy should stimulate renewed interest in the use of radiation therapy in ovarian cancer. Many investigators were quick to discard radiation therapy from the treatment armamentarium after cisplatin’s activity was recognized. However, because the long-term results with cisplatin have been less than satisfactory, radiation therapy deserves continued study. No precedent is evident from any type of solid tumor for discarding from the treatment armamentarium a single agent with the level of activity seen with radiation therapy in the treatment of ovarian cancer. No reports have been published of comparisons between abdominopelvic irradiation and the newer chemotherapy agents that have been shown to be active against ovarian cancer. Prospective randomized trials have demonstrated that cisplatin plus paclitaxel improves the survival of ­patients with advanced ovarian cancer compared with cisplatin plus cyclophosphamide, the previous standard.91,92 The chemotherapy regimen of choice for all patients with ovarian cancer is a combination of paclitaxel with cisplatin or carboplatin. Further explorations of the role of radiation therapy as primary postoperative adjuvant therapy for ­ epithelial ovarian cancer—radiation therapy as a single agent or ­radiation therapy as part of concurrent or sequential combined-­modality therapy—are justified, although such studies currently have low priority. The issues being explored with ­regard to optimizing chemotherapy include the dose intensity of the platinum compounds, prolongation of the ­number of treatment courses, and newer multidrug combinations containing other drugs with identified ­ activity against ovarian cancer (e.g., gemcitabine, topotecan,

docetaxel). Comparisons of intraperitoneal with intravenous therapy are being conducted.93

Consolidation or Salvage Radiation Therapy after Chemotherapy A strong theoretical basis is evident for the use of sequential multimodality therapy for ovarian cancer. If no cross-­resistance occurs among surgery, chemotherapy, and ­radiation therapy and if these modalities can be combined with acceptable toxicity without diminution of their individual efficacies, sequential multimodality therapy offers a means of increasing the total cytoreductive capacity of treatment. In 1978, Fuks and colleagues94 were among the first to propose the use of sequential multimodality therapy. Their regimen consisted of cisplatin-based chemotherapy followed by cytoreductive surgery and then abdominopelvic irradiation for patients with advanced disease.94 Since then, numerous reports of sequential multimodality therapy have appeared. In 1993, Thomas95 published a review of 28 studies that examined the role of radiation therapy for consolidation of results after surgery and chemotherapy or as salvage treatment for residual disease in patients with ovarian ­cancer (Table 32-4).95-107 The studies included a total of 713 ­patients. The review did not provide definitive conclusions about the value of this treatment because most of the studies were single-arm trials without appropriate controls. The balance of the evidence, however, was against a significant curative benefit for radiation therapy used as salvage or consolidation treatment (see Table 32-4).95-107 The use of only radiation therapy for consolidation purposes has been tested in five randomized trials (Table 32-5).86,108-112 Findings from two trials supported its use

824

PART 9  Female Genital Tract

over chemotherapy or observation, findings from two ­other trials found no difference between radiation therapy or chemotherapy, and the fifth study was closed early ­ after seeming to favor chemotherapy over radiation therapy. The variety in definitions of consolidation, the variations in chemotherapeutic regimens used, and the differences in radiation doses and volumes make comparisons difficult, but these findings suggest that the radiation technique and the extent of residual tumor are important factors. Patients with no evidence of disease during a second-look operation fare the best, followed by those with microscopic disease identified during at second-look operation and then by those with macroscopic residuum of less than 2 cm.113 The failure of this review to detect a benefit from salvage or consolidation radiation therapy may indicate that sequential therapy is not beneficial. However, methodologic reasons may account for the findings; most of the studies were observational, without appropriate controls, and patient selection was often inappropriate, with inclusion of patients with large-volume residual disease after chemotherapy, patients who had undergone many previous operations, patients who had received protracted chemotherapy and had demonstrated chemoresistance, patients with poor bone marrow reserve, and patients with extensive adhesions. Selection of Patients. In studies that reported the amount of residual disease, survival clearly depended on the tumor residuum before radiation therapy. For 473 patients in the 28 series, sufficient information was available to permit calculation of the amount of residual disease. Among these patients, survival rates were 76% (86 of 113) for ­patients with no residuum, 49% (77 of 158) for ­patients with microscopic residual disease or residuum less than 5 mm, and 17% (34 of 202) for patients with macroscopic residual disease.95 The survival rate for patients with macroscopic disease is probably no different from that achieved with primary surgery and chemotherapy. This finding suggests that salvage radiation therapy is inappropriate for most patients with larger-than-microscopic residual disease after chemotherapy.

It is unclear whether the better survival rates among ­ atients with microscopic or no residual disease are p ­attributable to the abdominopelvic irradiation or to the inherently better prognosis of those patients. The observed outcomes are compatible with the variability of already known prognostic factors in ovarian cancer, including ­patient age, ­ tumor grade, and previous response to chemotherapy; therefore, an unequivocal treatment benefit for radiation therapy cannot be established. However, the suggestion of better results among patients with microscopic or no residual disease (see Table 32-4) implies that at least some possible benefit may accrue to such patients and that future randomized controlled studies should examine the value of consolidation radiation therapy for such patients. Although most reported studies of consolidation radiation therapy have included too few patients to allow meaningful analyses of possible prognostic and treatment factors that contribute to the observed outcomes, the findings of Schray and colleagues114 are valuable because of the method of analysis and presentation. In a multivariate analysis of prognostic factors for disease-free survival, these investigators established that the following factors are significant: tumor grade (70% survival for grades I and II versus 10% for grades III and IV) and initial residuum before chemotherapy (50% survival for ≤2 cm versus 14% for >2 cm). Schray and colleagues114 also examined the prognostic impact of the amount of residual disease at the time of radiation therapy. Three of five patients with microscopic ­ disease remained disease-free after radiation therapy, whereas only 5 of 21 patients with macroscopic disease that had been debulked to microscopic amounts before consolidation radiation therapy remained disease free. Schray and colleagues114 examined the prognostic ­impact of the number of favorable factors: amount of residual disease before chemotherapy, amount of residual disease before radiation therapy, and tumor grade. Rates of freedom from disease after radiation therapy were 0% (0 of 17) for patients with none of the favorable factors, 29% (4 of 14) for patients with one favorable factor, 53% (8 of 15)

TABLE 32-5.    Survival in Prospective Randomized Trials of Consolidation Therapy for Ovarian Cancer Type of Treatment Study and Reference

Number of Patients

Extent of Residual Disease

Radiation Therapy

Chemotherapy

Control

P Value

19%*

0.006

35%*

0.032

Pickel et al, 1999108

64

CCR

45%*

Sorbe et al, 2000109

98

PCR

56%*

36%*

117

PCR or <2 cm after 2nd look

18%*

15%*

0.920

Lawton et al, 1990111

50

<2 cm after 2nd look

24%†

50%†

0.110

Bruzzone et al, 1990112

41

PCR or <2 cm after 2nd look

38%‡

64%‡

0.080

Lambert et al,

1993110

*Five-year recurrence-free survival rate. †Three-year actuarial survival rate. ‡Three-year progression-free survival rate. From Dusenbery KE, Bellairs EE, Potish RA, et al. Twenty-five year outcome of sequential abdominal radiotherapy and melphalan: implications for future management of epithelial carcinoma of the ovary. Gynecol Oncol 2005;96:307-313. CCR, complete clinical remission; PCR, pathologic complete remission.

32  The Ovary

for patients with three favorable factors, and 67% (4 of 6) for patients with four favorable factors.114 Appropriate selection of patients for studies of consolidation or salvage radiation therapy should be guided by this information. Radiation therapy is likely to improve cure rates among patients with no macroscopic disease after chemotherapy. However, it is doubtful whether radiation therapy can be beneficial for patients requiring further resection of chemoresistant disease before consideration of the use of radiation therapy.27,28,115,116 Because radiation therapy is not curative for patients with larger volumes of residual disease, it is unlikely ­(because of many biologic factors that may be operative, including accelerated clonogen proliferation) that salvage radiation therapy can be curative for patients with largevolume disease who have already undergone extensive surgery (including secondary debulking and chemotherapy, to which resistance may have already been demonstrated). Altered Fractionation Schedules. Altered fractionation schedules are being investigated as a potential means of increasing the biologically effective dose of consolidation radiation therapy and reducing its toxicity. The protracted course of antineoplastic therapy involved in ­ sequential multimodality treatment may allow accelerated clonogen proliferation and contribute to treatment failure. This theoretical concern, along with the desire to reduce the toxicity of therapy, is the basis for explorations of hyperfractionated regimens. Morgan and colleagues96 treated 15 patients with stage III disease and known residual disease after cisplatin-based chemotherapy with 30.4 Gy given in 0.8-Gy fractions twice daily plus a boost of 15 to 19 Gy to the pelvis. All ­patients completed the planned treatment course, although two required a break because of thrombocytopenia. No ­episodes of small bowel obstruction occurred. Kong and colleagues97 performed a similar study of sequential chemotherapy and radiation therapy with a radiation dose of 30 Gy in twice-daily 1-Gy fractions. In that study, only 1 of 23 patients so treated did not complete the planned ­therapy. Six patients with gross residual disease were also given a limited-field boost of 15 Gy in 15 fractions after the 30-Gy whole-abdomen dose, but none experienced control of disease, and five required surgery for small bowel obstruction attributed to radiation damage. Among the 17 patients who did not receive the boost dose, 5 ­developed bowel obstructions, but all were attributed to tumor recurrence and not ­irradiation. Randall and colleagues117 published the results of a GOG phase II trial of three cycles of cisplatin plus cyclophosphamide followed by hyperfractionated abdominopelvic irradiation (0.8 Gy per fraction twice daily to a total dose of 30.4 Gy) for patients with optimal stage III disease. This regimen was “well tolerated” and thought to be a “promising management option in selected patients.” Complications. Many studies of abdominopelvic irradiation after chemotherapy have reported complication rates higher than those reported for radiation therapy alone, which may be the result of the use of high radiation doses, protracted chemotherapy, and multiple abdominal operations.118 Because the benefit of abdominopelvic irradiation after chemotherapy is equivocal and probably confined to a small subset of patients, the risk of complications is a major consideration. This risk can be minimized by restricting the

825

radiation dose for patients who have undergone ­ secondlook operations. No reports are available on the use of whole-abdomen irradiation after taxane-based chemotherapy. Concern regarding the use of radiation therapy after taxanes includes issues about renal function and possible chemotherapy or radiation recall phenomena. General Recommendations. In summary, use of radiation therapy as consolidative treatment probably should be confined to patients with negative or microscopic disease at second-look laparotomy, prechemotherapy tumor bulk minimally larger than that of patients with optimally cytoreduced disease, no more than six courses of chemotherapy consisting of effective agents that are the least toxic to the bone marrow, and radiation therapy delivered with a dose and technique ensuring acceptable complication rates or designed to explore the possible theoretical benefits of ­hyperfractionated irradiation at somewhat higher than conventionally accepted doses.

Palliation Ovarian cancer that recurs or persists after first-line chemotherapy is incurable. Recurrent or persistent disease is often symptomatic, with generalized or localized abdominal symptoms from intraperitoneal disease. Usually, second-, third-, or fourth-line chemotherapy is used in attempts to prolong life and palliate symptoms. Few specific findings are available to permit evaluation of the quality of life or degree of symptom palliation achieved with this approach. Reported response rates range from 10% to 43%, and a ­variety of side effects have been reported.52,85,119-127 Radiation therapy as a palliative modality in ovarian cancer is often neglected but may be very useful if the sole or dominant site of symptomatic disease can be safely encompassed within a radiation field. In these situations (e.g., a fixed pelvic mass eroding the vaginal mucosa and causing bleeding, pain, or bowel or bladder dysfunction in the absence of obvious symptomatic peritoneal disease), local irradiation often produces tumor regression or relief from symptoms. Similarly, radiation therapy can be used to treat localized masses elsewhere in the abdomen, such as the retroperitoneal nodes. Palliative irradiation may also be useful for extra-abdominal disease, particularly supraclavicular or inguinal node masses and bone or brain metastases. Radiation therapy may also be beneficial for the acute relief of painful hepatomegaly resulting from capsular distention. Limited data have been published concerning the palliative efficacy of radiation therapy in these settings.52,53,128 Investigators at M. D. Anderson treated patients with largefraction therapy (1 to 3 fractions of 10 Gy, with fractions delivered 4 weeks apart) and reported overall response rates of 55% among patients treated for pain and 71% among patients treated for bleeding.128 However, after 6 of the 42 patients so treated developed severe radiation-induced bowel injury, the procedure was revised such that only 2 fractions were used, and no further complications have been reported. Another report of palliative radiation therapy for recurrent ovarian cancer after initial ­management with cisplatin-containing regimens documented that radiation therapy produced a mean symptom-free interval of 8.5 months in a group of patients with a median survival of 19.5 months.53 Corn and colleagues52 reported the use of fractionated palliative irradiation of 47 anatomic sites for 33 patients

826

PART 9  Female Genital Tract

with symptomatic recurrence of ovarian cancer after frontline cisplatin-based chemotherapy. Palliative responses occurred in 79% of the patients, and complete palliative responses occurred in 51%. The median duration of palliation was 4 months; in 90% of cases, palliation lasted until death. Ninety percent of patients treated for vaginal bleeding and 85% of patients treated for rectal bleeding experienced control of the bleeding; 83% of patients treated for pain experienced pain relief. A range of radiation doses was used, but it is unclear whether patients with better performance status received the higher doses. Palliation was more likely among patients with a higher Karnofsky performance status score and among patients who received a ­biologically effective dose of irradiation of at least 44 Gy, which was calculated as the total dose {1+ [fractional dose ÷ (α/β)]}, in which α/β was set at 10.52 Too few patients were in this study to permit a multivariate analysis of factors predictive of the observed palliative responses, and recommendations regarding the optimal palliative dose cannot be made from the findings of this study. The fractionation schemes used did not allow the construction of a full doseresponse curve. However, the study did show that durable palliation can be achieved by using local radiation therapy for ovarian cancer that recurs after cisplatin-based chemotherapy, particularly for patients with good performance status.

Future Directions in the Treatment of Ovarian Cancer Studies are required to explore the integration of radiation therapy with newer cytotoxic agents for the treatment of ovarian cancer, with particular attention paid to the use of concurrent chemotherapy and radiation therapy. Studies of other tumors sensitive to radiation therapy—­including carcinomas of the anal canal, esophagus, rectum, head and neck, vulva, and cervix129-133—show that the use of concurrent chemotherapy with radiation therapy can improve local control and survival. Despite the limitations on the radiation dose imposed by the large treatment volume and the need for inclusion of the upper abdomen, concurrent chemotherapy with radiation therapy for epithelial ovarian cancer needs to be explored. In vitro evidence has confirmed that low doses of radiation therapy can sensitize cisplatin-resistant ovarian cancer cell lines to cisplatin.134 These findings suggest that radiation therapy may have a role in the clinical treatment of platinum-resistant ovarian cancer. A phase I study of 11 patients with stage III or IV ovarian cancer confirmed the feasibility of abdominopelvic ­irradiation to a dose of 20 Gy with intraperitoneal cisplatin after debulking surgery.135 Because hematologic and gastrointestinal toxic effects were significant, the use of colony-stimulating growth factors was recommended for future prospective trials.135 The use of abdominopelvic ­irradiation in conjunction with low-dose weekly cisplatin (15 mg/m2) has been explored and found to be feasible for patients with advanced endometrial cancer.136 In a similar study137 of chemotherapy-refractory ovarian cancer, ­patients ­received 30 Gy of abdominopelvic irradiation in 30 fractions with 5 mg/m2 of weekly cisplatin. The projected 5- and 10-year survival rates from that study were 44% and 35%, respectively. Despite the current focus on exploration of multiple combinations of newer

c­ hemotherapeutic agents, studies to explore the integration of radiation therapy with these newer agents are indispensable. The taxane compounds, use of which is becoming standard for all stages of ovarian cancer, are ­effective cytotoxic agents and are in vitro radiation sensitizers.138 Concurrent paclitaxel with abdominopelvic ­irradiation needs to be explored. Another option deserving of exploration is the use of limited-field radiation therapy for carefully selected ­patients. Four studies have evaluated this approach for ­patients with recurrent disease.54,139-141 In one of those studies,139 the survival rates at 2 years after recurrence were 57% for 21 patients with no liver or extra-abdominal metastases at the start of the radiation therapy but 0% for 7 patients with liver or extra-abdominal metastases. Four of the 28 patients were still alive without evidence of disease at more than 5 years after treatment. In another study of 35 patients with recurrent ovarian cancer treated with local radiation therapy,140 the local recurrence–free survival rate at 5 years after the radiation was 66%, and the overall and disease-free survival rates at that time were both 34%. The corresponding rates were 89%, 72%, and 50% for patients who had had optimal resection before radiation therapy and 42%, 22%, and 19% for patients with gross residual ­disease or unresectable tumors. The current practice at M. D. Anderson is to treat patients with local recurrence in nodal or pelvic areas with local radiation therapy. Although the results are still preliminary, they show promise that this approach is another avenue to explore in the treatment of recurrent or primary ovarian cancer.

Management of Borderline Epithelial Ovarian Tumors Within the subgroup of ovarian tumors known as borderline tumors, natural history and histologic features are better understood for the serous subtype than for the less common subtypes. Most patients with stage I borderline ovarian malignancies are treated with total hysterectomy and bilateral ­oophorectomy without postoperative adjuvant therapy. Five-year survival rates with this approach range from 90% to 100%. When patients wish to preserve fertility, conservative therapy consisting of just a unilateral oophorectomy can be used. Although recurrence rates are higher after conservative surgery than after total hysterectomy and ­bilateral oophorectomy (15% versus 5%), the higher recurrence rates are not associated with measurable differences in long-term survival, possibly because of the long natural history of ovarian cancer or because bilaterality does not confer a worse prognosis. Only one prospective randomized trial of adjuvant therapy for stage I borderline ovarian cancer has been done.142 After complete surgery, patients were randomly assigned to undergo observation, pelvic irradiation, or oral melphalan. Only one patient experienced recurrent disease. This study and the excellent overall results of published nonrandomized studies show that adjuvant therapy is not necessary for stage I borderline tumors. Approximately 20% of borderline epithelial ovarian tumors manifest as stage III or IV disease. Metastatic borderline tumors are lethal in one third to one half of cases, although the clinical course is often protracted over many years.143 The significance of peritoneal implants in

32  The Ovary

advanced disease is uncertain. The suggestion has been made that invasive implants carry a worse prognosis than noninvasive implants.144,145 Noninvasive implants may be not true metastases but rather autochthonous extraovarian neoplasia, treatment of which is different from treatment of invasive disease. Kaern and colleagues146 used cellular deoxyribonucleic acid (DNA) content to identify which patients with borderline ovarian tumors had an unfavorable prognosis. Among patients older than 60 years with advanced mucinous or serous tumors, the 15-year survival rate was 75% for those with diploid tumors and 20% for those with aneuploid tumors. Other investigators subsequently confirmed the prognostic significance of aneuploidy in borderline tumors of the ovary.147,148 Although patients with aneuploid tumors are recognized as being at high risk for relapse, no evidence is available to show that they benefit from any type of adjuvant therapy. Various postoperative adjuvant therapies have been evaluated, including single-agent alkylating drugs and platinum-based chemotherapy, but their value has not been established.149 Patients have also been treated with EBRT, intraperitoneal radioactive colloid therapy, and combinedmodality therapies. Second-look laparotomies and clinical observations have demonstrated that these tumors can respond to chemotherapy and radiation therapy.150 However, it is impossible to determine whether those therapies prolonged disease-free survival or overall survival. Although 5-year survival rates for patients with advanced-stage borderline tumors range from 64% to 96%,149 the mortality rates from disease climb to 30% to 50% over time.143 The rarity of metastatic borderline tumors and their long natural history make randomized studies of therapy extremely difficult.

Granulosa Cell Tumors Granulosa cell tumors are diagnosed as unilateral nonmetastatic tumors in more than 80% of cases.151,152 Granulosa cell tumors spread to contiguous structures and extend through the peritoneum. Metastasis to lymph nodes, liver, or beyond the abdomen at diagnosis is rare. Malignant behavior occurs in 25% to 50% of pure granulosa cell and mixed granulosa cell–theca cell tumors. ­Because granulosa cell tumors often produce estrogen, many patients have endometrial hyperplasia, and about 10% have coexisting endometrial carcinoma. Metastases and bilaterality are good predictors of disease relapse and probably of dense adherence. Evidence in the literature is conflicting about whether relapse of nonmetastatic granulosa cell tumors can be predicted by cyst size greater than 15 cm, age older than 40 years, high ­mitotic rate, atypia, and diffuse pattern.153-156 Rupture, capsular penetration, and capillary space invasion seem to be nonprognostic in the absence of tumor spread. Granulosa cell tumors have a propensity for late recurrence after primary therapy, with more than one half of deaths occurring more than 5 years after diagnosis. Less than 10% of patients with surgical stage I disease experience recurrence, whereas 30% or more of patients with more advanced stage disease experience recurrence. More than 70% of patients who develop recurrent disease will die of granulosa cell tumor. Stenwig and colleagues152

827

r­ eported a mean interval to recurrence of 8.9 years (range, 1 to 22 years). The effect of adjuvant therapy on stage I disease cannot be evaluated, perhaps because the disease is rare and has a long natural history. Published reports often span decades of experience during which radiation therapy was applied sporadically and usually with small-volume techniques. In nonrandomized studies of stage I disease, relapse rates were relatively low regardless of whether postoperative radiation therapy was used or not.152,157 Patients with more extensive disease usually have been treated with surgical resection and postoperative radiation therapy and chemotherapy. Although some patients ­treated with this approach have remained disease-free, others have experienced recurrent disease, often with a protracted clinical course, which makes it impossible to determine the relative benefits of surgery and radiation therapy.157 It is reasonable to offer abdominopelvic irradiation (as in epithelial ovarian cancer; see “Postoperative Whole-­Abdomen and Pelvis Irradiation”) to patients with ­bilateral, metastatic, high-risk nonmetastatic, or recurrent ­granulosa cell tumors provided that the residual tumor bulk is not large. The published experience with chemotherapy is relatively sparse. Certain multidrug regimens may be more ­effective than single alkylating agents or the combinations of dactinomycin, fluorouracil, and cyclophosphamide; doxorubicin and cisplatin; or doxorubicin, cisplatin, and cyclophosphamide.158-162 In a GOG study in which vincristine, dactinomycin, and cyclophosphamide were used for ­ patients with advanced or recurrent granulosa cell ­ tumors, only a 20% partial response rate was observed.163 The highest activity has been reported for the combination of cisplatin, vinblastine, and bleomycin.164-166 Other investigators are using combinations of bleomycin, etoposide, and cisplatin or of paclitaxel and a platinum drug.

Dysgerminoma Dysgerminoma is different from the epithelial ovarian malignancies in many respects. It mainly occurs in much younger patients (85% of patients are younger than 29 years at diagnosis); its principal mode of dissemination is nodal rather than transperitoneal; and it is extremely sensitive to radiation therapy, with radiosensitivity resembling that of the equivalent male tumor, seminoma of the testis.167,168 Disease is confined to a single ovary at the time of diagnosis in about 75% of cases. Because dysgerminoma has been shown to be curable with chemotherapy and radiation therapy,168 management of this disease has changed appreciably over the past 4 decades. A laparotomy procedure similar to that required for epithelial ovarian tumors is recommended for patients with dysgerminoma, with careful palpation and possible biopsy of retroperitoneal lymph nodes where the ovarian vessels originate from the renal artery and vein. Cytologic examination of peritoneal fluid and omental biopsy are also recommended. Three large published series have examined the results of conservative surgery without postoperative radiation therapy for 145 patients with stage I dysgerminoma.167,169,170 Conservative surgery consisted of unilateral oophorectomy only, with or without ipsilateral salpingectomy. The other

828

PART 9  Female Genital Tract

ovary and tube and the uterus were not removed. Surgical staging was apparently not performed routinely in these series, but the 10-year survival rate for the patients was 91%—­similar to survival rates achieved by using more ­aggressive initial and complete surgery followed by radiation therapy. ­Approximately 25% of patients with stage I disease experienced recurrence after surgery, but disease in most of these patients was successfully controlled with radiation therapy, chemotherapy, or further surgery, bringing the 10-year survival rate to the reported 91%. As the efficacy of chemotherapy against dysgerminoma becomes recognized, investigators are advising unilateral adnexectomy even in the case of minor involvement of the opposite ovary or nodal or peritoneal metastases. One ­series showed that the use of cisplatin, vinblastine, and bleomycin chemotherapy in these situations resulted in eradication of disease in the contralateral ovary and apparent cure, with preservation of child-bearing capacity.171 Baseline investigations after surgery should include chest radiography, bipedal lymphangiography, abdominal computed tomography, and measurement of the levels of the serum tumor markers alpha-fetoprotein and β-human chorionic gonadotropin. If a conservative surgical approach was used, close postoperative monitoring for at least 2 years with chest radiography, abdominal computed tomography, and tumor marker evaluations is desirable. The objective of close monitoring is to detect recurrences before they become large, permitting chemotherapy or radiation therapy to be used with curative intent. Published findings are conflicting regarding the prognostic importance of primary tumor size in stage IA disease. Some have suggested that the risk of recurrence after conservative surgery may be related to the size of the primary tumor, but the evidence is insufficient to be persuasive against the use of conservative approaches, even for ­patients with large primary tumors. It is likely that adopting careful surveillance after conservative surgery will ­allow any relapse to be detected early and treated effectively with salvage therapy.169,172 The management of stage IB and higher-stage dysgerminoma is controversial. For patients who do not wish to ­maintain fertility and for patients with disease in the abdominopelvic cavity, radiation therapy is clearly ­indicated and usually curative. Because of the exquisite ­radiosensitivity of this tumor, it is unnecessary to ­exceed doses of 25 Gy in 20 to 25 fractions to the whole ­abdomen, with possible boosts of 10 Gy to sites of bulky ­ residual disease. Prophylactic mediastinal irradiation is not advocated for stage III disease because extra-abdominal nodal relapse is rare without uncontrolled abdominal disease, and mediastinal radiation therapy may reduce marrow reserves and prevent the effective use of salvage chemotherapy.173 For patients with stage II or III disease who wish to maintain fertility and for patients with supradiaphragmatic or visceral disease, combination chemotherapy is the treatment of choice.174 Accumulated data from three series show resumption of menses after unilateral ­oophorectomy and chemotherapy with the combination of cisplatin, ­vinblastine, and bleomycin in 19 of 20 patients, with two subsequent successful pregnancies.175 The optimal types of drugs and duration of therapy have not been determined, although the two most commonly used regimens are ­ vincristine, doxorubicin, and cyclophosphamide and

c­ isplatin, vinblastine, and bleomycin. The latter combination has been used successfully as salvage therapy for disease that has relapsed after doxorubicin and cyclophosphamide, and it may be the more effective of the two combinations. Some findings suggest that the combination of bleomycin, etoposide, and cisplatin cures most cases of recurrent or metastatic dysgerminoma regardless of its bulk.176,177 This combination is less toxic than cisplatin, vinblastine, and bleomycin. However, it is important that methods be developed for stratifying dysgerminomas on the basis of risk of recurrence so that chemotherapy regimens can be more effectively tailored to the risk level. Like testicular seminomas, dysgerminomas may regress slowly and incompletely after chemotherapy but usually do not represent residual disease unless they are enlarging. Consolidation radiation therapy after chemotherapy is not recommended unless a clear sign of progressive disease is evident after chemotherapy. For the rare patient for whom chemotherapy fails to control the disease and the disease is no longer considered curable, radiation therapy may provide excellent palliation.

REFERENCES    1. Cole MP. Suppression of ovarian function in primary breast cancer. In Forrest APM, Kunkler PB (eds): Prognostic Factors in Breast Cancer. Edinburgh: E & S Livingston, ��������������� 1968.    2. Stillman RJ, Schinfeld JS, Schiff I, et al. Ovarian failure in longtesrm survivors of childhood malignancy. Am J Obstet Gynecol 1981;139:62-66.    3. Fajardo LF, Berthrong M, Anderson RE. Female reproductive ­organs. In Radiation Pathology. New York: Oxford University Press, 2001, pp 289-300.    4. Peck WS, McGreer JT, Kretzschmar NR, et al. Castration of the female by irradiation: the results in 334 patients. Radiology 1940;34:176-186.    5. Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2008. CA Cancer J Clin 2008;58:71-96.    6. Edwards B, Howe HL, Ries LA, et al. Annual report to the ­nation on the status of cancer, 1973-1999, featuring implications of age and aging on the U.S. cancer burden. Cancer 2002;94: 2766-2792.    7. Greene MH, Clark JW, Blayney DW. The epidemiology of ovarian cancer. Semin Oncol 1984;11:209-226.    8. Cramer DW, Hutchison GB, Welch WR, et al. Factors affecting the association of oral contraceptives and ovarian cancer. N Engl J Med 1982;307:1047-1051.    9. Hartge P, Schiffman MH, Hoover R, et al. A case-control study of epithelial ovarian cancer. Am J Obstet Gynecol 1989; 161:10-16.   10. Casagrande JT, Louie EW, Pike MC, et al. “Incessant ovulation” and ovarian cancer. Lancet 1979;2:170-173.   11. Fathalla MF. Incessant ovulation—a factor in ovarian neoplasia? Lancet 1971;2:163.   12. Schildkraut JM, Thompson WD. Familial ovarian cancer: a population-based case-control study. Am J Epidemiol 1988;128: 456-466.   13. Boyd J. Specific keynote: hereditary ovarian cancer: what we know. Gynecol Oncol 2003;88:S8-S10.   14. Risch HA, McLaughlin J, Cole D, et al. Prevalence and pentrance of germline BRCA1 and BRCA 2 mutations in population series of 649 women with ovarian cancer. Am J Hum Genet 2001;68:700-710.   15. Antoniou A, Pharoah PD, Narod S, et al. Average risks of breast and ovarian cancer associated with BRCA1 or BRCA2 mutations detected in series unselected for family history: a combined analysis of 22 studies. Am J Hum Genet 2003;72: 1117-1130.

32  The Ovary   16. King MC, Marks JH, Mandell JB. New York Breast Cancer Study Group. Breast and ovarian cancer risks due to inherited mutations in BRCA1 and BRCA2. Science 2003;302:643-646.   17. Boyd J, Sonoda Y, Federici MG, et al. Clinicopathologic features of BRCA-linked and sporadic ovarian cancer. JAMA 2000;283:2260-2265.   18. Aarnio M, Sankila R, Pakkula E, et al. Cancer risk in mutation carriers of DNA-mismatch repair genes. Int J Cancer 1999;81: 214-218.   19. Bergman F. Carcinoma of the ovary: a clinicopathological study of 86 autopsied cases with special reference to mode of spread. Acta Obstet Gynecol Scand 1966;45:211-231.   20. Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 275-283.   21. Young RC, Decker DG, Wharton JT, et al. Staging laparotomy in early ovarian cancer. JAMA 1983;250:3072-3076.   22. Omura GA, Brady MF, Homesley HD, et al. Long-term ­followup and prognostic factor and analysis in advanced ovarian ­carcinoma: the Gynecologic Oncology Group experience. J Clin Oncol 1991;9:1138-1150.   23. Young RC, Walton LA, Ellenberg SS, et al. Adjuvant therapy in stage I and stage II epithelial ovarian cancer: results of two prospective randomized trials. N Engl J Med 1990;322:1021-1027.   24. Ozols RF, Bundy BN, Greer BE, et al. Phase III trial of carboplatin and paclitaxel compared with cisplatin and paclitaxel in patients with optimally resected stage III ovarian cancer: a Gynecology Oncology Group study. J Clin Oncol 2003;21:3194-3200.   25. McGuire WP, Hoskins WJ, Brady MF, et al. Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in ­patients with stage III and stage IV ovarian cancer. N Engl J Med 1996;334:1-6.   26. Hainsworth JD, Grosh WW, Burnett LS, et al. Advanced ovarian cancer: long term results of treatment with intensive ­cisplatin-based chemotherapy of brief duration. Ann Intern Med 1988;108:165-170.   27. Dembo AJ. Abdominopelvic radiotherapy in ovarian cancer: a 10-year experience. Cancer 1985;55:2285-2290.   28. Carey MS, Dembo AJ, Fyles AW, et al. Testing the validity of a prognostic classification in patients with surgically optimal ovarian carcinoma: a 15-year review. Int J Gynecol Cancer 1993; 3:24-35.   29. Gershenson DM, Copeland LJ, Wharton JT, et al. Prognosis of surgically determined complete responders in advanced ovarian cancer. Cancer 1985;55:1129-1135.   30. Greco FA, Hande KR, Jones HW, et al. Advanced ovarian cancer: long-term follow up after brief intensive chemotherapy ­[abstract]. Proc Am Soc Clin Oncol 1984:166.   31. Goldberg N, Peschel RE. Postoperative abdominopelvic radiation therapy for ovarian cancer. Int J Radiat Oncol Biol Phys 1988;14:425-429.   32. Sell A, Bertelsen K, Andersen JE, et al. Randomized study of whole abdomen irradiation versus pelvic irradiation plus ­cyclophosphamide in treatment of early ovarian cancer. Gynecol ­Oncol 1990;37:367-373.   33. Lindner H, Willich H, Atzinger A. Primary adjuvant whole ­abdominal irradiation in ovarian carcinoma. Int J Radiat Oncol Biol Phys 1990;19:1203-1206.   34. Meyer T, Rustin G. Role of tumour markers in monitoring epithelial ovarian cancer. Br J Cancer 2000;82:1535-1538.   35. Crawford SM, Peace J. Does the nadir CA125 concentration predict a long-term outcome after chemotherapy for carcinoma of the ovary? Ann Oncol 2005;16:47-50.   36. Badguell D, Bast RC Jr. Early detection of ovarian cancer. Dis Markers 2007;23:397-410.   37. Bast RC Jr, Brewer M, Zou C, et al. Prevention and early detection of ovarian cancer: Mission impossible? Recent Results Cancer Res 2007;174: 91-100.   38. Petricoin EF III, Ardekani AM, Hitt BA, et al. Use of proteanic patterns in serum to identify ovarian cancer. Lancet 2002;359:572-577.   39. Niloff JM, Bast RC Jr, Schaetzl EM, et al. Predictive value of CA 125 antigen levels in second-look procedures for ovarian cancer. Am J Obstet Gynecol 1985;151:981-986.

829

  40. Malkasian GD Jr, Knapp RC, Lavin PT, et al. Preoperative evaluation of serum CA-125 levels in premenopausal and postmenopausal patients with pelvic masses: discrimination of ­benign from malignant disease. Am J Obstet Gynecol 1988;159: 341-346.   41. Hoskins WJ, Bundy BN, Thigpen JT, et al. The influence of ­cytoreductive surgery on recurrence free interval and survival in small-volume stage III epithelial ovarian cancer: a Gynecologic Oncology Group study. Gynecol Oncol 1992;47:159-166.   42. Bolis G, Colombo N, Pecorelli S, et al. Adjuvant treatment for early epithelial ovarian cancer: results of two randomised clinical trials comparing cisplatin to no further treatment or chromic phosphate (32P). Ann Oncol 1995;6:887-893.   43. Trimbos JB, Parmar M, Vergote I, et al. International collaborative ovarian neoplasm trial 1 and adjuvant chemotherapy in ovarian neoplasm trial: two parallel randomized phase III trials of adjuvant chemotherapy in patients with early-stage ovarian carcinoma. J Natl Cancer Inst 2003;95:105-112.   44. Trimbos JB, Vergote I, Bolis G, et al. Impact of adjuvant chemotherapy and surgical staging in early-stage ovarian carcinoma: European Organisation for Research and Treatment of CancerAdjuvant Chemotherapy in Ovarian Neoplasm trial. J Natl Cancer Inst 2003;95:113-125.   45. Bell J, Brady MF, Young RC, et al. Randomized phase III trial of three versus six cycles of carboplatin and paclitaxel in early stage ovarian epithelial carcinoma: a Gynecology Oncology Group study. Gynecol Oncol 2006;102:432-439.   46. Hreschyshyn MM, Park RC, Blessing JA, et al. The role of ­adjuvant therapy in stage I ovarian cancer. Am J Obstet Gynecol 1980;138:139-145.   47. Dembo AJ, Bush RS, Beale FA, et al. The Princess Margaret Hospital study of ovarian cancer: stage I, II and asymptomatic III presentations. Cancer Treat Rep 1979;63:249-254.   48. Dembo AJ, Davy M, Stenwig AE, et al. Prognostic factors in patients with stage I epithelial ovarian cancer. Obstet Gynecol 1990;75:263-273.   49. Cardenes H, Randall ME. Integrating radiation therapy in the curative management of ovarian cancer: current issues and ­future directions. Semin Radiat Oncol 2000;10:61-70.   50. Vergote IB, Vergote-De Vos LN, Abeler VM, et al. Randomized trial comparing cisplatin with radioactive phosphorus or wholeabdomen irradiation as adjuvant treatment of ovarian cancer. Cancer 1992;69:741-749.   51. Young RC, Brady MF, Nieberg RK, et al. Adjuvant treatment for early ovarian cancer: randomized clinical trial of adjuvant treatment for early ovarian cancer: a randomized phase III trial of intraperitoneal 32P or intravenous cyclophosphamide and cisplatin—a Gynecologic Oncology Group study J Clin Oncol 2003;21:4350-4355.   52. Corn BW, Lanciano RM, Boente M, et al. Recurrent ovarian cancer: effective radiotherapeutic palliation after chemotherapy failure. Cancer 1994;74:2979-2983.   53. May LF, Belinson JL, Roland TA. Palliative benefit of ­radiation therapy in advanced ovarian cancer. Gynecol Oncol 1990;37: 408-411.   54. Davidson SA, Rubin SC, Mychalczak B, et al. Limited-field ­radiotherapy as salvage treatment of localized persistent or ­recurrent epithelial ovarian cancer. Gynecol Oncol 1993;51: 349-354.   55. Suzuki S, Fujiwara K, Yamauchi H, et al. Efficacy of radiation therapy for chemoresistant epithelial ovarian cancer [abstract P191]. Presented at the 6th Biennial Meeting of the IGCS. Fukuoka, Japan: Blackwell Science, 1997, p 81.   56. Dubois JB, Joyeux H, Solassol CL, et al. Les tumeurs epitheliales de l’ovaire: resultats therapeutiques a propos de 165 stades II et III. Gynecol Obstet Biol Reprod 1979;14:627-632.   57. Fuller DB, Sause WT, Plenk HP, et al. Analysis of postoperative radiation therapy in stage I through III epithelial ovarian carcinoma. J Clin Oncol 1987;5:897-905.   58. Piver MS, Barlow JJ, Lele SB. Incidence of subclinical metastasis in stage I and II ovarian carcinomas. Obstet Gynecol 1978;52:100-104.

830

PART 9  Female Genital Tract

  59. Rosenoff SH, Young RC, Anderson T, et al. Peritoneoscopy: a valuable staging tool in ovarian carcinoma. Ann Intern Med 1975;83:37-41.   60. Dembo A, Bush R, Beale F, et al. Ovarian carcinoma: improved survival following abdominopelvic irradiation in patients with a completed pelvic operation. Am J Obstet Gynecol 1979;134: 793-800.   61. Ozols RF, Rubin SC, Thomas G, et al. Epithelial ovarian cancer. In Hoskins WJ, Perez CA, Young RC (eds): Principles and Practice of Gynecologic Oncology, 2nd ed. Philadelphia: LippincottRaven, 1997, pp 919-986.   62. Martinez A, Schray MF, Howes AE, et al. Postoperative radiation therapy for epithelial ovarian cancer: the curative role based on a 24-year experience. J Clin Oncol 1985;3:901-911.   63. Weiser EB, Burke TW, Heller PB, et al. Determinants of survival of patients with epithelial ovarian carcinoma following whole abdominal irradiation (WAR). Gynecol Oncol 1988;30: 201-208.   64. Bush RS, Allt WEC, Beale FA, et al. Treatment of epithelial carcinoma of the ovary: operation, irradiation and chemotherapy. Am J Obstet Gynecol 1977;127:692-704.   65. Delclos L, Smith JP. Ovarian cancer, with special regard to types of radiotherapy. Natl Cancer Inst Monogr 1975;42:129-135.   66. Klaassen D, Shelley W, Starreveld A, et al. Early stage ovarian cancer: a randomized clinical trial comparing whole abdominal radiotherapy, melphalan, and intraperitoneal chromic phosphate: a National Cancer Institute of Canada Clinical Trials Group ­report. J Clin Oncol 1988;6:1254-1263.   67. Redman JR, Petroni GR, Saigo PE, et al. Prognostic factors in advanced ovarian cancer. J Clin Oncol 1986;4:515-523.   68. Ledermann JA, Dembo AJ, Sturgeon JFG, et al. Outcome of patients with unfavorable optimally cytoreduced ovarian cancer treated with chemotherapy and whole abdominal radiation. Gynecol Oncol 1991;41:30-35.   69. Hruby G, Bull CA, Langlands AO, et al. WART revisited: the treatment of epithelial ovarian cancer by whole abdominal ­radiotherapy. Australas Radiol 1997;41:276-280.   70. Dembo AJ, Van Dyk J, Japp B, et al. Whole abdominal irradiation by a moving-strip technique for patients with ovarian cancer. Int J Radiat Oncol Biol Phys 1979;5:1933-1942.   71. Delclos L, Braun EJ, Herrera JR, et al. Whole abdominal irradiation by 60Co moving-strip technique. Radiology 1963;81: 632-641.   72. Fazekas JT, Maier JG. Irradiation of ovarian carcinomas: a prospective comparison of the open-field and moving-strip techniques. AJR Am J Roentgenol 1974;120:118-123.   73. Dembo AJ, Bush RS, Beale FA, et al. A randomised clinical trial of moving strip versus open field whole abdominal irradiation in patients with invasive epithelial cancer of ovary [abstract]. Int J Radiat Oncol Biol Phys 1983;9(Suppl 1):97.   74. Lanciano RM, Randall M. Update on the role of radiotherapy in ovarian cancer. Semin Oncol 1991;18:233-247.   75. Fyles AW, Thomas GM, Pintilie M, et al. A randomized study of two doses of abdominopelvic radiation therapy for patients with optimally debulked stage I, II, and III ovarian cancer. Int J Radiat Oncol Biol Phys 1998;41:543-549.   76. Fletcher GH. Clinical dose-response curves of human malignant epithelial tumours. Br J Radiol 1973;46:1-12.   77. Marks LB. A standard dose of radiation for “microscopic disease” is not appropriate. Cancer 1990;66:2498-2502.   78. Dembo AJ. Radiotherapeutic management of ovarian cancer. Semin Oncol 1984;11:238-250.   79. Withers HR, Peters LJ, Taylor JMG. Dose-response relationship for radiation therapy of subclinical disease. Int J Radiat Oncol Biol Phys 1995;31:353-359.   80. Fyles AW, Dembo AJ, Bush RS, et al. Analysis of complications in patients treated with abdomino-pelvic radiation therapy for ovarian carcinoma. Int J Radiat Oncol Biol Phys 1992;22: 847-851.   81. Perez CA, Lorba A, Zivnuska F, et al. 60Co moving strip technique in the management of carcinoma of the ovary: analysis of tumor control and morbidity. Int J Radiat Oncol Biol Phys 1978;4:379-388.

  82. Schray MF, Martinez A, Howes AE. Toxicity of open field whole abdominal irradiation as primary post-operative treatment in gynecological malignancy. Int J Radiat Oncol Biol Phys 1988;16:397-403.   83. van Bunnigen B, Bouma J, Kooijman C, et al. Total abdominal irradiation in stage I and II carcinoma of the ovary. Radiother Oncol 1988;11:305-310.   84. Dietl J, Marzusch K. Ovarian surface epithelium and human ovarian cancer. Gynecol Obstet Invest 1993;35:129-135.   85. Einzig AI, Wiernik PH, Sasloff J, et al. Phase II study and longterm follow up of patients treated with taxol for advanced ovarian adenocarcinoma. J Clin Oncol 1992;10:1748-1753.   86. Dusenbery KE, Bellairs EE, Potish RA, et al. Twenty-five year outcome of sequential abdominal radiotherapy and melphalan: implications for future management of epithelial carcinoma of the ovary. Gynecol Oncol 2005;96:307-313.   87. Hong L, Alektiar K, Chui C, et al. IMRT of large fields: wholeabdomen irradiation. Int J Radiat Oncol Biol Phys 2002;54: 278-289.   88. Duthoy W, De Gersem W, Vegoete K, et al. Whole abdominopelvic radiotherapy (WAPRT) using intensity modulated arc therapy (IMAT): first clinical experience. Int J Radiat Oncol Biol Phys 2003;57:1019-1032.   89. Travis LB, Holowaty EJ, Bergfeldt K, et al. Risk of leukemia after platinum-based chemotherapy for ovarian cancer. N Engl J Med 1999;340:351-357.   90. Omura GA, Bundy BN, Berek JS, et al. Randomized trial of ­cyclophosphamide plus cisplatin with or without doxorubicin in ovarian carcinoma: a Gynecologic Oncology Group study. J Clin Oncol 1989;7:457-465.   91. McGuire WP, Hoskins WJ, Brady MF, et al. A phase III trial comparing cisplatin/cytoxan (PC) and cisplatin/taxol (PI) in ­advanced ovarian cancer (AOC) [abstract 808]. Proc Am Soc Clin Oncol 1993;12:255.   92. McGuire WP, Hoskins WJ, Brady MF, et al. Taxol and cisplatin (TP) improves outcome in advanced ovarian cancer (AOC) as compared to cytoxan and cisplatin (CP) [abstract]. Proc Am Soc Clin Oncol 1995;14:771.   93. Jaaback K, Johnson N. Intraperitoneal chemotherapy for the initial management of primary epithelial ovarian cancer. Cochrane Database Syst Rev 2006;(1): CD005340.   94. Fuks Z, Rizel S, Anteby SO, et al. Current concepts in cancer: ovary—treatment for stages III and IV. The multimodal approach to the treatment of stage III ovarian carcinoma. Int J Radiat Oncol Biol Phys 1982;8:903-908.   95. Thomas GM. Is there a role for consolidation or salvage radiotherapy after chemotherapy in advanced epithelial ovarian cancer? Gynecol Oncol 1993;51:97-103.   96. Morgan L, Chafe W, Mendenhall W, et al. Hyperfractionation of whole-abdomen radiation therapy: salvage treatment of persistent ovarian carcinoma following chemotherapy. Gynecol Oncol 1988;31:122-134.   97. Kong JS, Peters LJ, Wharton JT, et al. Hyperfractionated ­splitcourse whole abdominal radiotherapy for ovarian carcinoma: ­tolerance and toxicity. Int J Radiat Oncol Biol Phys 1988;14: 737-743.   98. Chiara S, Orsatti M, Franzone P, et al. Abdominopelvic radiotherapy following surgery and chemotherapy in advanced ovarian cancer. Clin Oncol (R Coll Radiol) 1991;3:340-344.   99. Greiner R, Goldhirsch A, Davis BW, et al. Whole-abdomen radiation in patients with advanced ovarian carcinoma after surgery, chemotherapy and second-look laparotomy. J Cancer Res Clin Oncol 1984;107:94-98. 100. Reddy S, Hartsell W, Graham J, et al. Whole abdomen radiation therapy in ovarian carcinoma: its role as a salvage therapeutic modality. Gynecol Oncol 1989;35:307-313. 101. Kuten A, Stein M, Steiner M, et al. Whole abdominal irradiation following chemotherapy in advanced ovarian carcinoma. Int J Radiat Oncol Biol Phys 1988;14:273-279. 102. Solomon HJ, Atkinson KH, Coppleson JVM, et al. Ovarian carcinoma: abdominopelvic irradiation following reexploration. Gynecol Oncol 1988;31:396-401.

32  The Ovary 103. Rosen EM, Goldberg ID, Rose C, et al. Sequential multi-agent chemotherapy and whole abdominal irradiation for stage III ovarian carcinoma. Radiother Oncol 1986;7:223-232. 104. Haie C, Pejovic-Lenfant MH, George M, et al. Whole ­abdominal irradiation following chemotherapy in patients with minimal ­residual disease after second look surgery in ovarian carcinoma. Int J Radiat Oncol Biol Phys 1989;17:15-19. 105. Kersh CR, Randall ME, Constable WC, et al. Whole abdominal radiotherapy following cytoreductive surgery and chemotherapy in ovarian carcinoma. Gynecol Oncol 1988;31:113-120. 106. Goldhirsch A, Greiner R, Dreher E, et al. Treatment of advanced ovarian cancer with surgery, chemotherapy and consolidation of response by whole-abdominal radiotherapy. Cancer 1988;62: 40-47. 107. Menczer J, Modan M, Brenner J, et al. Abdominopelvic irradiation for stage II-IV ovarian carcinoma patients with limited or no residual disease at second-look laparotomy after completion of ­ cisplatinum-based chemotherapy. Gynecol Oncol 1986;24:149-154. 108. Pickel H, Lagousen M, Petru E, et al. Consolidation radiotherapy after carboplatin-based chemotherapy in radically operated advanced ovarian cancer. Gynecol Oncol 1999;72:215-219. 109. Sorbe B for the Swedish-Norwegian Ovarian Cancer Study Group. Consolidation treatment of advanced (FIGO stage III) ovarian carcinoma in complete surgical remission after induction chemotherapy: a randomized, controlled, clinical trial comparing whole abdominal radiotherapy, chemotherapy, and no further treatment. Int J Gynecol Cancer 2003;13:278-286. 110. Lambert HE, Rustin GJ, Gregory WM, et al. A randomized clinical trial comparing single-agent carboplatinum followed by ­radiotherapy for advanced ovarian cancer. North Thames Ovary Group Study. J Clin Oncol 1993;11:440-448. 111. Lawton F, Luesley D, Blackledge BG, et al. A randomized trial comparing whole abdominal radiotherapy with chemotherapy following cisplatinum cytoreduction in epithelial ovarian cancer. West Midlands Ovarian Cancer Group Trial II. Clin Oncol 1990;2:4-9. 112. Bruzzone M, Repetto L, Chiara S, et al. Chemotherapy versus radiotherapy in the management of ovarian cancer patients with pathological complete response or minimal residual disease at second look. Gynecol Oncol 1990;38:392-395. 113. Goldberg H, Stein ME, Steiner M, et al. Consolidation ­radiation therapy following cytoreductive surgery, chemotherapy and ­second-look laparotomy for epithelial ovarian carcinoma: longterm follow-up. Tumori 2001;87:248-251. 114. Schray MF, Martinez A, Howes AE, et al. Advanced epithelial ovarian cancer: salvage whole abdominal irradiation for patients with recurrent or persistent disease after combination chemotherapy. J Clin Oncol 1988;6:1433-1439. 115. Dembo AJ, Bush RS. Current concepts in cancer: ovary— ­treatment of stages III and IV. Choice of postoperative therapy based on prognostic factors. Int J Radiat Oncol Biol Phys 1982;8: 893-897. 116. Dembo AJ, Bush RS, Brown TC. Clinicopathological correlates in ovarian cancer. Bull Cancer 1982;69:292-298. 117. Randall ME, Barrett RJ, Spirtos NM, et al. Chemotherapy, early surgical reassessment, and hyperfractionated abdominal ­radiotherapy in stage III ovarian cancer: results of a Gynecologic Oncology Group study. Int J Radiat Oncol Biol Phys 1996;34: 139-147. 118. Whelan TJ, Dembo AJ, Bush RS, et al. Complications of whole abdominal and pelvic radiotherapy following ­ chemotherapy for advanced ovarian cancer. Int J Radiat Oncol Biol Phys 1992;22:853-858. 119. Hatch KD, Beecham JB, Blessing JA, et al. Responsiveness of ­patients with advanced ovarian carcinoma to tamoxifen. A Gynecologic Oncology Group study of second-line therapy in 105 patients. Cancer 1991;68:269-271. 120. Markman M, Hakes T, Reichman B, et al. Ifosfamide and ­mesna in previously treated advanced epithelial ovarian cancer: activity in platinum-resistant disease. J Clin Oncol 1992;10: 243-248.

831

121. Markman M, Rothman R, Hakes T, et al. Second-line platinum therapy in patients with ovarian cancer previously treated with cisplatin. J Clin Oncol 1991;9:389-393. 122. McGuire WP, Rowinsky EK, Rosenshein NB, et al. Taxol: a unique antineoplastic agent with significant activity in ­advanced ovarian epithelial neoplasms. Ann Intern Med 1989;111:273-279. 123. Moore DH, Fowler WC, Jones CP, et al. Hexamethylmelamine chemotherapy for persistent or recurrent epithelial ovarian cancer. Am J Obstet Gynecol 1991;165:573-576. 124. Reed E, Jacob J, Ozols RF, et al. 5-Fluorouracil (5-FU) and ­leucovorin in platinum-refractory advanced stage ovarian carcinoma. Gynecol Oncol 1992;46:326-329. 125. Rosen GF, Lurain JR, Newton M. Hexamethylmelamine in ovarian cancer after failure of cisplatin based multiple-agent chemotherapy. Gynecol Oncol 1987;27:173-179. 126. Thigpen JT, Vance RB, Khansur T. Second-line chemotherapy for recurrent carcinoma of the ovary. Cancer 1993;71: 1559-1564. 127. Vergote I, Himmelmann A, Frankendal B, et al. Hexamethylmelamine as a second-line therapy in platinum-resistant ovarian cancer. Gynecol Oncol 1992;47:282-286. 128. Adelson MD, Wharton JT, Delclos L, et al. Palliative radio­ therapy for ovarian cancer. Int J Radiat Oncol Biol Phys 1987; 13:17-21. 129. Rose PG, Bundy BN, Watkins EB, et al. Concurrent ­ cisplatinbased chemoradiation in locally advanced cervical cancer. N Engl J Med 1999;340:1144-1153. 130. Keys HM, Bundy BN, Stehman FB, et al. Cisplatin, radiation and adjuvant hysterectomy compared with radiation and adjuvant hysterectomy for bulky stage IB cervical carcinoma. N Engl J Med 1999;340:1154-1161. 131. Whitney CW, Sause W, Bundy BN, et al. A randomized comparison of fluorouracil plus cisplatin versus hydroxyurea as an adjunct to radiation therapy in stages IIB-IVA carcinoma of the cervix with negative para-aortic lymph nodes. A Gynecologic Oncology Group and Southwest Oncology Group study. J Clin Oncol 1999;17:1339-1348. 132. Peters WA III, Liu PY, Barrett RJ, et al. Cisplatin and 5­fluorouracil plus radiation therapy are superior to radiation therapy as ­adjunctive in high-risk early-stage carcinoma of the cervix after radical hysterectomy and pelvic lymphadenectomy: report of a phase III intergroup study [abstract]. In Program of the 30th Annual Conference, Society of Gynecologic ­ Oncologists. San Francisco: Academic Press, 1999, p 443. 133. Morris M, Eifel PJ, Lu J, et al. Pelvic radiation with concurrent chemotherapy versus pelvic and para-aortic radiation for highrisk cervical cancer. A randomized Radiation Therapy Oncology Group clinical trial. N Engl J Med 1999;340:1137-1143. 134. Silver DF, Wheeless CR, Dubin NH. Radiotherapy as a ­cisplatinsensitizer in a resistant ovarian carcinoma cell line. Cancer 1996;77:1850-1853. 135. King LA, Downey GO, Potish RA, et al. Concomitant whole­abdominal radiation and intraperitoneal chemotherapy in ­advanced ovarian carcinoma: a pilot study. Cancer 1991;67: 2867-2871. 136. Reisinger SA, Asbury R, Liao SY, et al. A phase I study of weekly cisplatin and whole abdominal radiation for the treatment of stage III and IV endometrial carcinoma: a Gynecologic Oncology Group pilot study. Gynecol Oncol 1996;63:299-303. 137. Dowlatshahi M, Miller M, Semrad N, et al. Use of concomitant low dose cisplatinum and whole abdominal radiation (WARP) as salvage therapy for persistent ovarian carcinoma after positive second look procedure: a pilot study (abstract). Cancer J Sci Ann 1999;5:123. 138. Steren A, Sevin B, Perras J, et al. Taxol as a radiation sensitizer: a flow cytometric study. Gynecol Oncol 1993;50:89-93. 139. Firat S, Erickson B. Selective irradiation for the treatment of recurrent ovarian carcinoma involving the vagina or rectum. Gynecol Oncol 2001;80:213-220. 140. Albuguerue KV, Singla R, Potkul RK, et al. Impact of tumor volume-directed involved field radiation therapy integrated in the management of recurrent ovarian cancer. Gynecol Oncol 2005;96:701-704.

832

PART 9  Female Genital Tract

141. Fujiwara K, Suzuke S, Yoden E, et al. Local radiation therapy for localized relapse or refractory ovarian cancer patients with or without symptoms after chemotherapy. Int J Gynecol Cancer 2002;12:250-256. 142. Creasman WT, Park R, Norris H, et al. Stage I borderline ovarian tumors. Obstet Gynecol 1982;59:93-96. 143. Colgan TJ, Norris HJ. Ovarian epithelial tumors of low malignant potential: a review. Int J Radiat Oncol Biol Phys 1982;8: 219-226. 144. Bell DA, Weinstock MA, Scully RE. Peritoneal implants of ovarian serous borderline tumors. Cancer 1988;62:2212-2222. 145. McCaughey WTE, Kirk ME, Lester W, et al. Peritoneal epithelial lesions associated with proliferative serous tumors of ovary. Histopathology 1984;8:195-208. 146. Kaern J, Trope C, Kjorstad KE, et al. Cellular DNA content as a new prognostic tool in patients with borderline tumors of the ovary. Gynecol Oncol 1990;38:452-457. 147. Drescher CW, Flint A, Hopkins M, et al. Prognostic significance of DNA content and nuclear morphology in borderline ovarian tumors. Gynecol Oncol 1993;36:242-246. 148. Padberg BC, Arps H, Franke U, et al. DNA cytophotometry and prognosis in ovarian tumors of borderline malignancy. Cancer 1992;69:2510-2514. 149. Chambers JT. Borderline ovarian tumors: a review of treatment. Yale J Biol Med 1989;62:351-356. 150. Gershenson DM, Silva EG. Serous ovarian tumors of low malignant potential with peritoneal implants. Cancer 1990;65: 578-585. 151. Norris HJ, Taylor HB. Prognosis of granulosa-theca tumors of the ovary. Cancer 1968;21:255-263. 152. Stenwig JT, Hazekamp JT, Beecham JB. Granulosa cell tumors of the ovary: a clinicopathological study of 118 cases with long term follow-up. Gynecol Oncol 1979;136-152. 153. Bompas E, Freyer G, Vitrey D, et al. Granulosa cell tumour: review of the literature [in French]. Bull Cancer 2000;87: 709-714. 154. Fujimoto T, Sakuragi N, Okuyama K, et al. Histopathological prognostic factors of adult granulosa cell tumors of the ovary. Acta Obstet Gynecol Scand 2001;80:1069-1074. 155. Juric G, Zarkovic N, Nola M, et al. The value of cell proliferation and angiogenesis in the prognostic assessment of ovarian granulosa cell tumors. Tumori 2001;87:47-53. 156. Miller BE, Barron BA, Wan JY, et al. Prognostic factors in adult granulosa cell tumor of the ovary. Cancer 1997;79:1951-1955. 157. Kalavathi N. Granulosa cell tumour: hormonal aspects and ­radiosensitivity. Clin Radiol 1971;22:524-527. 158. Schwartz PE, Smith JP. Treatment of ovarian stromal tumors. Am J Obstet Gynecol 1976;125:402-411. 159. Jacobs AJ, Deppe G, Cohen CJ. Combination chemotherapy of ovarian granulosa cell tumor with cisplatinum and doxorubicin. Gynecol Oncol 1982;14:294-297.

160. Camlibel F, Caputo TA. Chemotherapy of granulosa cell tumors. Am J Obstet Gynecol 1983;145:763-765. 161. 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. 162. Pectasides D, Alevizakos N, Athanassiou AE. Cisplatin­containing regimen in advanced or recurrent granulosa cell tumours of the ovary. Ann Oncol 1992;3:316-318. 163. Slayton RE. Management of germ cell and stromal tumors of the ovary. Semin Oncol 1984;11:299-313. 164. Colombo N, Sessa C, Landoni F, et al. Cisplatin, vinblastine, and bleomycin combination chemotherapy in metastatic granulosa cell tumor of the ovary. Obstet Gynecol 1986;67:265-268. 165. Chiara S, Merlini L, Campora E, et al. Cisplatin-based chemotherapy in recurrent or high risk ovarian granulosa cell tumor patients. Eur J Gynaecol Oncol 1993;14:314-317. 166. Zambetti M, Escobedo A, Pilotti S, et al. Cisplatin/vinblastine/ bleomycin combination chemotherapy in advanced or recurrent granulosa cell tumors of the ovary. Gynecol Oncol 1990;36: 317-320. 167. Asadourian LA, Taylor HB. Dysgerminoma: an analysis of 105 cases. Obstet Gynecol 1969;33:370-379. 168. De Palo G, Pilotti S, Kenda R, et al. Natural history of dysgerminoma. Am J Obstet Gynecol 1982;143:799-807. 169. Gordon A, Lipton D, Woodruff JD. Dysgerminoma: a review of 158 cases from the Emil Novak Ovarian Tumor Registry. Obstet Gynecol 1981;58:497-504. 170. Malkasian GD Jr, Symmonds RE. Treatment of the ­ unilateral ­encapsulated ovarian dysgerminoma. Obstet Gynecol 1964;90: 379-382. 171. Bianchi UA, Sartori E, Favalli G, et al. New trends in the ­treatment of ovarian dysgerminoma [abstract]. Gynecol Oncol 1986;23:246. 172. Krepart G, Smith JP, Rutledge F, et al. The treatment of dysgerminoma of the ovary. Cancer 1978;41:986-990. 173. Thomas GM, Dembo AJ, Hacker HF, et al. Current concepts of therapy for dysgerminoma of the ovary. Obstet Gynecol 1987;70:268-275. 174. Einhorn LH, Williams SD. Chemotherapy of disseminated seminoma. Cancer Clinical Trials 1980;3:307-313. 175. Taylor MH, Depetrillo AD, Turner AR. Vinblastine, bleomycin and cisplatin in malignant germ cell tumors of the ovary. Cancer 1985;56:1341-1349. 176. Brewer M, Gershenson DM, Herzog CE, et al. Outcome and reproductive function after chemotherapy for ovarian dysgerminoma. J Clin Oncol 1999;17:2631-2632. 177. Williams SD, Kauderer J, Burnett AF, et al. Adjuvant therapy of completed resected dysgerminoma with carboplatin and etoposide: a trial of the Gynecologic Oncology Group. Gynecol Oncol 2004;95:496-499.

The Brain and Spinal Cord 33 shiao Y. Woo TOLERANCE OF THE CENTRAL NERVOUS SYSTEM TO IRRADIATION Postirradiation Cerebral Necrosis Postirradiation Myelopathy Other Late Effects CENTRAL NERVOUS SYSTEM NEOPLASMS TUMORS COMMON IN CHILDREN Medulloblastoma Astrocytoma Brainstem Gliomas Ependymomas Pineal Region and Germ Cell Tumors

Craniopharyngioma Primitive Neuroectodermal Tumors Tumors in Children Younger than 3 Years TUMORS COMMON IN ADULTS Malignant Gliomas: Anaplastic Astrocytoma and Glioblastoma Multiforme Oligodendroglioma Meningioma Primary Malignant Lymphoma of the Central Nervous System Carcinoma Metastatic to the Brain SPINAL CORD TUMORS

Tolerance of the Central Nervous System to Irradiation The central nervous system (CNS) is a relatively sensitive, often dose-limiting tissue in curative radiation therapy for CNS and adjacent neoplasms.1 Radiation injury has been documented on the basis of clinical signs, imaging findings, and histopathologic correlations.2-6 CNS effects are more pronounced in children and seem to be inversely related to age, with the youngest children having the greatest CNS sensitivity. Of further concern is that numerous chemotherapy agents are associated with unique or enhanced ­sequelae when given in conjunction with radiation therapy.7 Studies of laboratory models of cerebral and spinal cord changes have helped to clarify the pathophysiology of ­ radiation-induced neural damage and to associate the extent of damage with time, dose, delivery, and volume.8-10   For example, Caveness11 reported a detailed investigation of the clinical, histologic, and physiologic findings in ­primates after irradiation comparable with that used clinically. In that study, a single fraction of 20 Gy was uniformly fatal by week 26; this dose produced increased intracranial pressure and ventriculomegaly with scattered necrotic foci most dramatically affecting the brainstem. A single fraction of 15 Gy produced discrete microfoci of white matter necrosis by 26 weeks, with coalescing areas of white matter necrosis evident by 56 weeks. Pathologic findings were less pronounced in animals killed 78 weeks after irradiation, suggesting that some recovery had taken place. No clinical or histologic abnormalities were observed after wholebrain irradiation with a single fraction of 10 Gy or with  40 Gy delivered in 20 fractions over 4 weeks. In experiments more relevant to clinical radiation therapy, delivering 60 Gy to the whole brain in fractions of 2 Gy over a period of 6 weeks caused papilledema and ­anorexia in pubertal monkeys, but no overt clinical findings were apparent in adult primates. Histologically, the adult brains showed discrete microfoci of white matter necrosis

with calcifications by 26 weeks; healing of the degenerative foci along with increased microcalcifications was seen by 52 weeks. The radiation-induced changes were quantitatively greater in younger animals. After 80 Gy delivered in 40 fractions, confluent areas of necrosis appeared in the cerebral hemispheres of younger animals by 32 weeks, progressive necrosis was observed at 52 weeks, and coalescent necrotic regions and atrophy were seen by 78 weeks. Caveness’ findings at lower dose levels are similar to the mineralizing microangiopathy seen in children with acute leukemia treated with cranial irradiation (20 to 24 Gy) and methotrexate.12 Small, coalescent foci of necrosis ­ related to microvascular changes may account for the clinically ­silent lacunar lesions evident in magnetic resonance imaging (MRI) studies of long-term survivors of childhood brain tumors.13 Postirradiation CNS effects are typically classified temporally as acute, subacute, and late reactions. Fractionated radiation rarely produces clinically apparent acute changes, even when the total dose approaches 70 to 80 Gy.14 Anecdotal reports of transient exaggeration of neurologic signs have been related theoretically to perilesional or intralesional edema that develops at the beginning of ­irradiation.15-17 Young and colleagues18 found abrupt neurologic deterioration in 50% of patients who were given large (7.5- to 10Gy) fractions for the treatment of symptomatic cerebral metastases; the investigators attributed this effect to the radiation therapy. Although no similar phenomena were reported in studies by the Radiation Therapy Oncology Group (RTOG) in which 6-Gy fractions were administered, this finding might have been related to the more consistent use of corticosteroids by the RTOG.16 Subacute CNS effects are rather common. Jones19 ­observed transient, self-limited paresthesias with neck flexion (Lhermitte’s sign) that lasted for 1 to 2 months after spinal cord irradiation. These paresthesias, which ­occur in 15% to 25% of patients after mantle irradiation for Hodgkin disease, are thought to result from transient

835

836

PART 10  Central Nervous System

demyelination. Comparable transitory neurologic changes, called the somnolent syndrome, occur after cranial irradiation such as that given for acute leukemia; mild encephalopathy or focal neurologic changes can appear after the treatment of intracranial tumors.20,21 The cranial syndrome is thought to stem from a transient demyelinating event and is the result of the effects of radiation on the replicating oligodendrocytes.8 MRI can detect specific dose-related changes in gray and white matter within several weeks of conventionally fractionated irradiation.22 The potential for self-limited neurologic deterioration, most often occurring during the second month after irradiation, is important to bear in mind because clinical and radiologic changes can sometimes mimic tumor progression.23,24 Late reactions of the normal brain are the dose-limiting toxicities for radiation treatment. Clinical manifestations of focal neurologic deficits, encephalopathy, or neuropsychological dysfunction accompany various morphologic findings, such as atrophy, calcifications, diffuse white matter degeneration, and focal necrosis. Permanent or progressive myelopathy can result from spinal cord irradiation. Late changes in hypothalamic-pituitary function manifest as specific radiation-induced endocrine deficits.

Postirradiation Cerebral Necrosis Postirradiation cerebral necrosis, the most direct effect of CNS irradiation, is thought to result from direct effects of radiation on the replicating glial cell compartments and the capillary endothelial cells.8,9 The theoretical relationship between primary glial cells (i.e., type II astrocytes and oligodendrocytes) and vascular endothelial components is shown in Figure 33-1.8,25-29 Clinical signs of postirradiation cerebral necrosis are seen 6 to 36 months after

Vascular endothelium

Astrocyte I

Astrocyte II

Oligodendrocyte

Myelinated nerve fiber Blood-brain barrier (BBB) Internodal

Paranodal 1–2 Transient edema ? (high dose) months Progressive BBB breakdown

3–6 months

?

Early demyelination

Postirradiation Myelopathy

Progressive demyelination

Capillaries venules

White matter necrosis

Telangiectasia infarcts

Late delayed effects

therapy.15,16,30 Cerebral changes often mimic residual or recurrent tumor, with mass effect, enhancement, or cyst formation evident on computed tomography (CT) scans; these changes occur at or near the site of the primary tumor (i.e., the high-dose volume for intracranial neoplasms).31 Cerebral necrosis unrelated to intraparenchymal neoplasms has been described in reviews by Kramer and colleagues32 and Sheline and coworkers.16 Necrosis has been observed after therapy for extracranial tumors (predominantly those of the skin and paranasal sinuses) and for noninvasive pituitary tumors or craniopharyngiomas. Postirradiation necrosis is a dose-related phenomenon with a relative threshold radiation dose of 50 to 55 Gy if treatment is given in a conventional fractionation schedule (1.8- to 2-Gy fractions given once daily).16,30,33,34 Marks and colleagues30 reported the occurrence of necrosis in 5% of 139 patients with intracranial tumors treated with conventional fractionation to doses higher than 45 Gy; 6 of the 7 patients who had necrosis were given total doses in excess of 63 Gy. The dose-time relationship is critical in determining CNS tolerance, with fraction size the dominant factor among patients receiving doses of 60 to 63 Gy. Necrosis in patients who receive less than 60 Gy usually has been ­associated with fractions of greater than 2.5 Gy.16 Aristizabal and colleagues33 documented necrosis after treatment for pituitary tumors in 3% of 106 patients given less than 2.2 Gy per day (total dose of about 50 Gy) compared with 15% of 13 patients given doses of greater than 2.2 Gy daily. The influence of fraction size was made apparent in the isoeffect analyses of brain necrosis reported by Sheline and colleagues.16 Sheline and coworkers16 contend that the brain can tolerate a total radiation dose of 50 to 54 Gy given in fractions of 2 Gy once daily. At that dose, cerebral necrosis is estimated to occur in 0.04% to 0.4% of patients. The likelihood of necrosis is higher among patients with primary brain tumors, although the incidence is still estimated to be less than 5% among patients receiving a dose of 55 Gy at 1.8 to 2 Gy per fraction.34 Tolerance is affected by age; ­ experimental and clinical data indicate that children younger than 2 years can tolerate approximately 10% to 20% less radiation to the brain than those older than 2 years.11,35,36

Glial atrophy

FIGURE 33-1.  The direct effects of radiation on glial cells (astrocyte II and oligodendrocyte) and the indirect effects resulting from vascular changes are shown in a scheme developed by van der Kogel. (From van der Kogel AJ. Central nervous system radiation injury in small animal models. In Gutin PH, Leibel SA, Sheline GE [eds]: Radiation Injury to the Nervous System. New York: Raven Press, 1991, pp 91-112.)

Postirradiation myelopathy is the spinal cord equivalent of cerebral necrosis. The myelopathy manifests as ­sensory and motor signs referable to a single cord level, signs that are often clinically indicative of hemisection of the cord (Brown-Séquard syndrome).37,38 Symptoms occur as early as 6 months after treatment (median, 20 months).39 The pathophysiology of spinal cord myelopathy is similar to that of cerebral necrosis, and there is continued debate concerning the relative importance of vascular endothelial changes (particularly in the spinal cord with end-arterial blood supply to portions of the thoracic cord) and direct effects of the radiation on glial elements.28,29,40 The time-dose relationship and tolerance of the spinal cord to irradiation have been analyzed in the context of ­extraspinal tumors. The thoracic spinal cord seems to be the most sensitive portion; few instances of lumbar myelopathy have been reported. Phillips and Buschke41 suggested that the thoracic spine could tolerate a dose as high as 

33  The Brain and Spinal Cord

837

50 Gy given in fractions of 2 Gy once daily. Abbatucci and colleagues42 observed cervical cord myelopathy after doses of 55 Gy or more. The incidence of myelopathy rises rapidly with increasing doses, approaching 50% at 65 Gy.29 Many clinical trials of large-fraction irradiation for lung cancer have confirmed that the frequency of reactions ­ increases as the fraction size exceeds 2.5 Gy.43-46 Fraction size also seems to correlate inversely with the latent ­interval before the development of myelopathy42; however, no experimental evidence exists to suggest improved tolerance to fractions less than 2 Gy.47 The Continuous Hyperfractionated, Accelerated Radiation Therapy (CHART) regimen for lung cancer was originally based on a presumed cord tolerance of a low dose per fraction, but that regimen initially was associated with an unacceptable risk of myelopathy.48,49 Tolerance decreases as the volume or length of cord irradiated increases.41,42

energy transfer (LET) radiation for pituitary or parasellar tumors.63 The optic chiasm is often considered a dose-limiting structure when irradiation for suprasellar or adjacent temporal lobe lesions is being planned, and special attention is given to limiting the risk of dose-related visual complications.64 Optic nerve or chiasm injury has been documented in almost 20% of patients treated with fraction sizes of 2.5 Gy or greater (to total doses of 45 to 50 Gy), compared with 0 of 27 patients and 0 of 500 patients treated with fraction sizes of 1.7 to 2 Gy to the same level.16,65 Late injury to the brachial plexus observed after the treatment of breast cancer similarly seems to occur largely after highdose fractions.66

Other Late Effects

Primary tumors of the CNS represent 2% of all cases of cancer in the United States. Approximately 20,500 new cases are diagnosed annually.67 The incidence of these ­tumors exceeds that of Hodgkin disease in adults, and they account for 20% of all cases of childhood cancer. The natural history of tumors of the brain and spinal cord is relatively unusual because aggressive malignant ­lesions rarely disseminate beyond the neuraxis. Categorizing CNS tumors as benign or malignant is often difficult because histologically low-grade or benign neoplasms are sometimes associated with a poorer prognosis than certain histologically malignant tumors. It is instead the potential for local invasiveness or metastasis along the cerebrospinal fluid (CSF) pathways that determines the “malignancy” of CNS tumors. The challenge for neurosurgeons and radiation oncologists is to remove or devitalize neoplastic tissue in the functionally vital, anatomically confined nervous system.

A unique late effect of cranial irradiation combined with ­chemotherapy, called leukoencephalopathy, has been ­observed in children with acute lymphoblastic leukemia, adults with small cell carcinoma of the lung, and patients with primary CNS tumors.12,50,51 Leukoencephalopathy is a profoundly demyelinating, necrotizing reaction that usually occurs 4 to 12 months after combined treatment with methotrexate and radiation.52 Dementia and ­ dysarthria may progress to seizures, ataxia, focal long-tract signs, or death. Although most of these symptoms regress after ­systemic, intrathecal, or intraventricular methotrexate ­therapy is discontinued, patients are left with permanent neurologic deficits. The occurrence of ­leukoencephalopathy in ­patients treated with systemic and intrathecal methotrexate without radiation and a dose-effect relationship that is more apparent for methotrexate than for radiation suggest that the clinical leukoencephalopathy is more ­significantly related to the methotrexate than to the radiation.7 However, white matter changes interpreted as leukoencephalopathy (and so coded in the NCI Common Terminology Criteria for Adverse Events) must be distinguished from the clinical syndrome previously described. Decline in intellectual function has been related to cranial irradiation of adults and particularly of children.53-57 Neurocognitive changes in children correlate significantly with cranial irradiation, young age at diagnosis and ther­ apy, and functional level at the time the irradiation was begun.53-55,58,59 The occurrence of dose-related neuroco­ gnitive alterations suggests a role for continued, if cautious, reduction in the full cranial radiation dose for patients with medulloblastoma, for example, who exhibit favorable presenting signs.60 Changes in white matter volume obtained by MRI segmentation constitute an objective variable that is closely tied to intellectual function and that may permit more detailed studies of the effects of radiation therapy and chemotherapy in the current generation of clinical trials.61 Intellectual impairment in adults has occurred primarily ­after intensive chemotherapy and radiation for the preven­ tion of metastasis from small cell carcinoma of the lung and for treatment of primary CNS lymphomas.62 Peripheral nervous system tissue is relatively resistant to irradiation. Acute and subacute symptoms or signs ­ occur infrequently; transient cranial nerve neuropathies are ­observed only with high-dose, single-fraction, high–linear

Central Nervous System Neoplasms

Tumors Common in Children Brain tumors in children account for one in five cases of childhood cancer. The incidence of primary CNS tumors, particularly the astrocytic lesions, increased in children during the first half of the 1990s; since then, the incidence has leveled off, perhaps because of the added diagnostic sensitivity of MRI.68 Pediatric brain tumors arise most often in the posterior fossa. Posterior fossa tumors often produce symptoms and signs of increased intracranial pressure because the growing tumor obstructs the flow of CSF through the fourth ventricle. Ataxia often accompanies pressure-related headaches and morning vomiting. Brainstem gliomas cause cranial nerve palsies, ataxia, and long-tract signs, including lateralizing weakness and sensory deficits. Supratentorial tumors occur in the cerebral ­hemispheres and the central or deep-seated areas of the thalamus, ­hypothalamus, suprasellar area, and pineal–third ­ventricular region. Symptoms are local and include lateralizing neurologic deficits and seizures, which are associated with hemispheric and thalamic tumors; visual and endocrine changes; and specific oculomotor findings, which are associated with tumors of the pineal region. Primary tumors of the spinal cord account for only 5% of tumors in children. Metastasis to the CNS is uncommon

838

PART 10  Central Nervous System

in childhood cancer. Contiguous extradural ­ compression within the spinal canal or at the base of the skull (e.g., in neuroblastoma, rhabdomyosarcoma, primary bone ­tumors) is more common than hematogenous metastasis to the brain. Common primary intracranial tumors diagnosed in children are listed in Table 33-1.25,26 The histologic classification of pediatric brain tumors has been relatively standardized by the World Health Organization (WHO).69 The most common tumors are the astrocytic lesions (i.e., astrocytomas, malignant gliomas, and brainstem gliomas). Medulloblastoma, which accounts for 20% of all childhood CNS tumors, is part of the broader category of supratentorial primitive neuroectodermal tumors (PNETs) and other embryonal CNS tumors (i.e., cerebral neuroblastoma, ­ ependymoblastoma, and pineoblastomas).69-71 Biologic findings increasingly define the nature of CNS tumors in children, providing some associations with tumor aggressiveness and outcome. Medulloblastoma, for example, is often associated with gene deletions in chromosomal region p17; expression of the EGFR gene (formerly designated ERBB1), which encodes the epidermal growth factor receptor, and NTRK3 (formerly called TRKC), which encodes the neurotrophic tyrosine kinase receptor type 3, are negatively and positively correlated with outcome, respectively.72,73 Several other molecular markers, such as cyclin-dependent kinase 6, MYC, apurinic or apyrimidinic endonuclease activity, and members of the WNT pathway, also have prognostic significance.74-77 Several investigators have proposed combining information on gene expression patterns and clinical stage for stratifying risk groups in ­medulloblastoma.78,79

TABLE 33-1.   Relative Incidence of Brain Tumors in Children Tumor Type

Incidence (%)

Infratentorial

55

Medulloblastoma

25

Astrocytoma (cerebellum)

15

Brainstem glioma

10

Ependymoma (fourth ventricular region)

6

Supratentorial

45

Astrocytoma

23

Malignant glioma (anaplastic astrocytoma, glioblastoma multiforme)

6

Embryonal (primitive neuroectodermal) tumors

4

Ependymoma

3

Craniopharyngioma

6

Pineal region tumors

4

Oligodendroglioma

2

Modified from Duffner PK, Cohen ME, Myers MH, et al. Survival of children with brain tumors: SEER program, 1973-1980. Neurology 1986;36:597-601; Childhood Brain Tumor Consortium. A study of childhood brain tumors based on surgical biopsies from 10 North American institutions: sample description. J Neurooncol 1988;6:9-23.

Medulloblastoma Medulloblastoma represents 20% of childhood brain ­tumors. Common throughout the pediatric age range, the median age at diagnosis is 6 years, and 25% of tumors occur in children younger than 3 years. Approximately 10% to 20% of medulloblastomas occur in adults, predominantly in the third to fifth decades. The classic presenting triad of posterior fossa tumors such as medulloblastomas consists of headaches and vomiting, which are signs of increased intracranial pressure and ataxia. Medulloblastoma is by definition a posterior fossa ­tumor, most often arising in the cerebellar vermis, within the cerebellar hemispheres (typically in adolescents and frequently as the desmoplastic variant) or along the cerebellopontine peduncle. The tumor grows to fill the fifth ventricle and often attaches to the brainstem. Medulloblastoma is the characteristic “seeding” CNS tumor and is associated with subarachnoid dissemination through the CSF; metastasis may be apparent from cytologic findings or from images of radiographic deposits within the intracranial subarachnoid space (i.e., along the convexities or within the basal cisterns, sometimes limited to the region of the pituitary stalk) or within the spinal subarachnoid space. Subarachnoid disease is apparent at diagnosis in 20% to 25% of affected children.13,80 Extraneural disease, usually involving the bone or bone marrow, is an uncommon finding at ­diagnosis or disease recurrence.81 The embryonal CNS tumors represent a group of histologically related neoplasms that occur predominantly in children (Box 33-1).69 The WHO classification defines the embryonal tumors as round, blue, small cell tumors that tend to seed the neuraxis and respond to radiation and chemotherapy, although the prognosis varies widely. Less common tumors in this category include the supratentorial PNETs; the rare, primitive medulloepithelioma; and atypical or rhabdoid tumor (see “Tumors in Children Younger than 3 Years”). Pineoblastoma is classified separately (see “Pineal Region and Germ Cell Tumors”). The term posterior fossa PNET is synonymous with medulloblastoma.69

BOX 33-1.  Embryonal Central Nervous System Tumors* Medulloblastoma Central neuroblastoma Ependymoblastoma Primitive neuroectodermal tumor (PNET)† Additional Tumors with a Distinct Histology and Different Genetic Pathway of Evolution Medulloepithelioma Atypical teratoid or rhabdoid tumor Pineal Parenchymal Tumor Classified Clinically as an Embryonal Tumor Pineoblastoma *In the definition proposed by Kleihues and Cavenee,69 ­embryonal central nervous system tumors are “histologically identifiable ­entities [that] all develop on the background of an undifferentiated ­round-cell tumor but show a variety of divergent patterns of ­differentiation.” †PNET refers to supratentorial tumors; tumors that occur in the ­posterior fossa are known as medulloblastomas.

33  The Brain and Spinal Cord

Medulloblastoma is thought to originate from the primitive external granular layer of the developing ­cerebellum. The tumor is characterized by ­ undifferentiated small, round, blue cells, often including ­neuroblastic rosettes and with various degrees of neuronal and glial ­differentiation. The desmoplastic tumors are ­classically identified by their prominent nodularity and dense ­areas of reticular fibers.69 ­Other histopathologic ­subsets ­include medulloblastoma with extensive nodularity or ­advanced neuronal ­differentiation and large cell or anaplastic ­medulloblastoma.73

Surgery and Staging Surgery is the initial therapy, with the goal of maximal ­resection for the sake of therapy and reestablishing CSF flow. Intraoperative ventriculostomy has replaced the routine initial placement of a ventriculoperitoneal shunt; with current surgical approaches, less than 25% of children ­require postoperative conversion of the temporary ventriculostomy to a permanent ventriculoperitoneal shunt. Gross total resection is limited only by substantial invasion of the tumor into the brainstem or infiltration into the peduncle. Gross total or near-total resection (i.e., more than 90% of the tumor removed, with less than 1.5 cm2 of residual disease) is achieved in almost 90% of children older than  3 years in the United States.82 The operative mortality rate should be less than 1% or 2% for children treated with current approaches, and the postoperative morbidity is typically acceptable, with a small proportion of children left with permanent new cranial nerve deficits or marked ataxia.82,83 The posterior fossa syndrome, marked by severe truncal ataxia, difficulties with speech and swallowing stemming from bulbar effects, and sometimes mutism, seems to be associated with extensive dissection along the lower brainstem; symptoms usually abate considerably over several months after the surgery, although late residual deficits of various degrees of severity have been seen.84 Postoperative neuraxis staging is standard in defining risk groups and subsequent therapy for medulloblastoma and the other embryonal tumors. Immediate postoperative cranial MRI or CT (preferably done within 1 to 3 days of surgery) defines the postoperative residual effect, which is a more reliable prognostic factor than the operative coding of the extent of resection.82 Neuraxis staging begins with a contrast-enhanced spinal MRI (preferably more than 10 days after surgery to eliminate the confusion of findings with postoperative debris) and subsequent lumbar CSF cytologic studies; routine bone scanning and bone marrow assessments are standard only in investigational settings.13,80 Staging is based on the extent of residual tumor (average risk is defined as the presence of ≤1.5 cm2 of residual tumor) and the presence or absence of subarachnoid or extraneural metastasis (defined according to a modification of the Chang system85) (Box 33-2). Most important from the standpoint of planning therapy is determining whether isolated CSF cytologic evidence of metastatic disease (M1 disease), imaging evidence of intracranial disease outside the posterior fossa (M2 disease), or imaging evidence of spinal metastasis (M3 disease) is present; extranodal disease is classified as M4.85 Standard or  average-risk disease is defined as that which has been totally removed or has residual tumor volume of less than 1.5 cm2 with M0 disease; almost 60% to 65% of children enrolled in U.S. studies at the turn of the century were so classified.

839

The presence of only local residual disease is uncommon in children with high-risk disease; more often, these children have neuraxis dissemination.

Radiation Therapy Postoperative management consists of radiation therapy and chemotherapy.86 Infants and young children (i.e., those younger than 3 years) are treated in accordance with ­developing investigational approaches (see “Tumors in Children Younger than 3 Years”). The standard sequence of therapy for children 3 years old or older is to begin radiation therapy within 28 to 31 days of surgery, followed by chemotherapy.87 Although studies of preirradiation chemotherapy have added significantly to our knowledge of chemosensitivity, randomized trials have shown no benefit from administering chemotherapy before the radiation, and in some instances, results have suggested that this sequence is ultimately detrimental to disease control.88-91 Radiation therapy for medulloblastoma is technically demanding and biologically unforgiving. It requires the systematic inclusion of the entire subarachnoid space (i.e., craniospinal irradiation [CSI]) followed by a boost to the posterior fossa. One standard technique involving immobilization of patients in the prone position is illustrated in Figure 33-2. Lateral craniocervical fields encompass the ­intracranial and upper cervical spine regions, with particular attention to including the subfrontal and infratemporal regions (at the cribriform plate in the midline subfrontal area and along the infratemporal regions) (Fig. 33-3). The conventional dose to the neuraxis in patients ­given surgery and radiation is 36 Gy (delivered in 1.8-Gy

BOX 33-2.  Chang Staging System for Medulloblastoma Primary Tumor (T)* T1 Tumor less than 3 cm in diameter T2 Tumor at least 3 cm in diameter T3a Tumor larger than 3 cm in diameter with extension into the aqueduct of Sylvius and/ or into foramen of Luschka T3b Tumor larger than 3 cm in diameter with unequivocal extension into the brainstem T4 Tumor larger than 3 cm in diameter with extension up past the aqueduct of Sylvius and/or down past the foramen magnum (i.e., beyond the posterior fossa) Distant Metastasis (M) M0 No evidence of gross subarachnoid or hematogenous metastasis M1 Microscopic tumor cells found in cerebrospinal fluid M2 Gross nodule seedings demonstrating in the cerebellar, cerebral subarachnoid space, or in the third or lateral ventricles M3 Gross nodular seeding in spinal subarachnoid space M4 Metastasis outside the cerebrospinal axis *The T stage is no longer thought to be prognostically related to outcome. Modified from Chang CH, Housepian EM, Herbert C Jr. An ­operative staging system and a megavoltage radiotherapeutic technic for ­cerebellar medulloblastoma. Radiology 1969;93:1351-1359.

840

PART 10  Central Nervous System

Posterior spine

A

B

FIGURE 33-2.  A, Classic craniospinal irradiation configurations, with a divergent posterior spinal field matching opposed lateral craniocervical fields. The collimator rotation enabling the craniocervical fields to be matched to the divergence of the posterior spinal field is typically modified at least every 5 fractions. Field markings for medulloblastoma are determined by the cribriform plate at the orbital level (see Fig. 32-3). B, The parallel inferior border from the craniocervical fields is achieved by couch rotation, calculated to achieve a three-dimensional match in the depicted plane.

FIGURE 33-3.  A conventional fluoroscopic image used for simulating the lateral craniocervical fields used in craniospinal irradiation for medulloblastoma (see Fig. 33-2). The long arrow indicates the margins along the cribriform plate. The  radiopaque dot and the small arrows indicate the anterior margins of the lateral orbital rims. Divergence rather than a collimator angle is needed when only the subarachnoid space is to be ­targeted.

f­ ractions once daily). The progression-free survival rate for patients with localized disease after maximal resection and CSI (36 Gy) is approximately 65% at 5 to 10 years (Table 33-2).36,92-96 A randomized study of patients with averagerisk medulloblastoma conducted by the Pediatric Oncology Group (POG) and Children’s Cancer Group (CCG) in the late 1980s established the superiority of a total dose of

36 Gy compared with 23.4 Gy for CSI and the benchmark outcome of more than 90% tumor removal with less than 1.5 cm2 of residual disease.92 Later findings suggest that the combination of reduced-dose CSI (23.4 Gy) and chemotherapy produces disease control rates comparable with or superior to those for conventional-dose CSI alone.97 Large-scale prospective studies are documenting the efficacy of what is now considered the standard treatment for localized medulloblastoma, which is radiation (CSI dose of 23.4 Gy) followed by cisplatin-containing chemotherapy. Reports of outcome from nonrandomized, single-arm trials, further supported by neuropsychological findings from the POG-CCG study previously mentioned, suggest that cognitive function in children with average-risk disease after therapy is superior among those given CSI to 23.4 Gy instead of 36 Gy, especially in children 3 to 8.5 years old.60 The current Children’s Oncology Group (COG) trial is testing, in children between 3 and 7 years old with averagerisk disease, whether the CSI dose can be further reduced from 23.4 Gy to 18 Gy. Children with metastatic disease at presentation (i.e., M1 to M3 disease) require a dose of 36 to 39 Gy to the neuraxis. The use of proton therapy for CSI has a theoretical advantage over photon therapy because of the lack of exit dose from proton therapy. Miralbell and colleagues estimated that the risk of second neoplasms could be reduced with proton therapy relative to photon therapy.98 Irradiation includes a boost to the posterior fossa ­region, classically defined as the entire infratentorial ­volume. The use of opposed lateral posterior fossa fields has been standard, but radiation oncologists are increasingly using a three-dimensional conformal radiation therapy approach99 to spare the auditory apparatus and, when possible, the pituitary-hypothalamic region (Fig. 33-4). Proton therapy for pediatric medulloblastoma is also being considered as a way of maximizing target dose coverage and sparing normal tissues (Fig. 33-5). The cumulative dose delivered to the posterior fossa is typically 54 to 55.8 Gy. A few institutional summaries of patterns of treatment failure for patients with medulloblastoma and descriptions of preliminary experience with target volumes confined

33  The Brain and Spinal Cord

841

TABLE 33-2.    Survival Rates after Craniospinal Irradiation for Medulloblastoma Study and Reference Bloom et al, Berry et al,

196936

198193 198694

Period

Number of Patients

Survival Rate at 5 Years (%)

Survival Rate at 10 Years (%)

1950-1964

71

40

30

1958-1978

122

56

43 42

1977-1985

53

64

199095

1970-1981

53

53

45

Jenkin et al, 199096

1977-1987

72

71

63

Thomas et al, 200092

1978-1991

42 (CSI = 36 Gy)

67*

67*

46 (CSI = 23 Gy)

52*

52*

Hughes et al, Bloom et al,

*Event-free

survival rates at 5 and 8 years. CSI, craniospinal irradiation.

A FIGURE 33-4.  A, Treatment plans for a three-dimensional conformal technique demonstrating the ideal dose distribution for a sixfield, non-coplanar approach targeting the entire posterior fossa. Anatomically, this approach includes the tentorium as it extends superiorly in the posterior clinoid and then inferiorly to define the anterior and superior margins of the posterior fossa. The inferior and posterior margins are defined by the calvarium and the lower margin of foramen magnum.  (Continued)

to the operative or tumor bed highlight investigational ­approaches that are being studied prospectively in a COG study.100,101

Chemotherapeutic Agents Several categories of chemotherapeutic agents have shown efficacy in medulloblastoma: alkylating agents (e.g., ­cyclophosphamide, lomustine), platinating agents (e.g., cisplatin, carboplatin), topoisomerase II inhibitors (e.g., etoposide), and the vinca alkaloids (e.g., vincristine). The most common adjuvant treatment for standard- or average-risk disease was described by Packer and colleagues87 and consists of concurrent vincristine during irradiation and postirradiation cycles of lomustine, cisplatin, and vincristine. The addition of concurrent carboplatin or etoposide and high-dose, peripheral stem cell–supporting chemotherapy regimens has been studied in two separate protocols for the treatment of high-risk disease.102 The first ­ report of 

a multi-institutional phase II trial of this approach showed 5-year event-free survival rates of 83% for children with average-risk disease and 70% for children with high-risk disease.103 Although posterior fossa recurrence previously was ­regarded as the dominant pattern of failure, later studies have indicated that more patients exhibit neuraxis or ­combined local and disseminated failure. Approaches to the management of disease recurrence are not standardized but often include chemotherapy, further surgery for localized disease, and the judicious use of additional irradiation. Secondary or salvage therapies are rarely successful.

Astrocytoma Astrocytomas are the most common glial tumors in children, developing in the cerebellum, brainstem, cerebral hemispheres, diencephalon (i.e., thalamus, hypothalamus, and third ventricular region, including the optic chiasm),

842

PART 10  Central Nervous System

B FIGURE 33-4.—cont’d  B, Three-dimensional conformal dose distribution demonstrates coverage of the primary tumor site, which in this case is the operative bed after surgical resection of a medulloblastoma. A multifield, non-coplanar, 6-MeV photon approach uses a gross tumor volume that outlines the operative bed, expanded by 2 cm (as modified by anatomic boundaries) to identify the clinical target volume. A 5-mm planning target volume geometrically surrounds the clinical target volume.

or spinal cord.104-106 In adults, low-grade tumors occur ­primarily in the cerebral hemispheres and thalamus; however, malignant astrocytomas far exceed the number of histologically benign lesions beyond the age of 20 years. The WHO grades astrocytomas as grade I (juvenile ­ pilocytic astrocytoma [JPA]) or grade II (fibrillary ­astrocytoma or astrocytoma not otherwise specified); highgrade ­lesions are deemed grade III (anaplastic [malignant] astrocytoma) or grade IV (glioblastoma multiforme).69 Grossly, the ­ tumors are solid with discrete or infiltrating margins or cystic. Pilocytic astrocytomas are biologically nonaggressive ­lesions that are often cystic, occurring ­predominantly in the cerebellum and diencephalon in ­children (Fig. 33-6). The natural history of astrocytomas depends on the age of the patient and the site of origin and histologic classification of the tumor. Pilocytic tumors rarely progress to a more malignant histology; however, fibrillary astrocytomas progress to a more malignant histology in 30% to 50% of cases.107-109 The proliferative index, assessed by staining for Ki-67 with the murine monoclonal antibody MIB1, has been reported to influence prognosis.110

Cerebellar Astrocytomas Cerebellar astrocytomas involve the vermis or hemispheres in 75% of patients; in 25% of children, tumors manifest as transitional neoplasms that involve the cerebellum and the cerebellar peduncles or adjacent brainstem.111-114 Cerebellar astrocytomas are classically cystic lesions; the textbook appearance consisting of a large cyst associated with a ­mural nodule occurs in about one fourth of cases. Grossly, most tumors are mixed cystic and solid lesions or apparently solid tumors with small cystic components. Histologically, 75% of astrocytomas in this location are JPAs.111,112,114,115 Debate continues about whether high-grade gliomas occur

in the cerebellum in children; if they do, they are quite uncommon.116 Therapy for cerebellar astrocytomas is largely surgical. Reports from institutions indicate that gross total resection is done in 65% to 90% of cases, although this rate has risen to approximately 85% to 90% in recent series.100,112,114,115 Resection is more often incomplete in patients with tumors involving the cerebellar peduncles or the brainstem.112-115 The operative mortality rate should be less than 4%, unlike the rate for patients with medulloblastomas and ependymomas that may invade the adjacent brainstem. Morbidity after radical resection of cerebellar astrocytomas usually is limited. Recurrence of tumor after radioimaging confirmation of complete surgical removal is uncommon; in the most ­recent series, the recurrence rate is less than 10%.112,114,117,118 Prognosis most closely parallels the extent of resection; multivariate analyses have also identified solid (rather than cystic) lesions, transitional lesions (versus those confined to the cerebellum), and lesions showing a fibrillary or diffuse histology (versus JPA) as potentially significant prognostic factors.112,114,115,119 A direct correlation between the volume of residual tumor and outcome has also been reported.112 Indications for radiation therapy are limited in patients with cerebellar astrocytomas. Although radiation therapy has been suggested to be effective for incompletely ­resected solid lesions, enthusiasm is generally limited for postoperative irradiation for any cerebellar astrocytomas.117,120 A small subset of tumors, often those that primarily involve the cerebellar peduncle, will recur after initial surgery; our approach to the treatment of such tumors is to use irradiation only after tumor progression after a second attempt at resection. MRI (T2-weighted or fluid attenuation inversion recovery [FLAIR] sequences) and MRI/CT fusion are

33  The Brain and Spinal Cord

A

D

B

E

C

F

843

FIGURE 33-5.  Medulloblastoma treatment plans. Posterior fossa axial slices are shown for three-dimensional conformal radiation therapy (3D CRT) (A), intensity-modulated radiation therapy (IMRT) (B), and protons (C). Isodose lines are designated for 3D CRT and IMRT: blue, 30.6 Gy; red, 15 Gy; yellow, 10 Gy; purple, 5 Gy. Isodose lines are designated for protons: red, 30.6 Gy; green,  15 Gy. For craniospinal irradiation, axial slices are shown for 3D CRT electrons (D), 3D CRT photons (E), and protons (F). Isodose lines are designated for 3D CRT and IMRT: blue, 23.4 Gy; red, 20 Gy; yellow, 10 Gy; purple, 5 Gy. Isodose lines are designated for protons: aqua/green, 23.4 Gy; light blue, 20 Gy; darker blue, 10 Gy; indigo, 5 Gy.  (Continued)

useful for target delineation, and a margin of 0.5 to 1 cm seems adequate. The total dose is 50 to 54 Gy in conventional fractionation.

Supratentorial Astrocytomas Supratentorial astrocytomas include diencephalic or ­central lesions and lesions of the cerebral hemispheres. In children, total resection can be achieved for up to 90% of ­ astrocytomas in the cerebral hemisphere.104,108,121,122 ­Diffuse cerebral lesions, including those labeled ­gliomatosis cerebri, can arise in several cerebral lobes with an often lowgrade histologic appearance (i.e., fibrillary, WHO grade II) but clinically malignant potential. Similarly, ­thalamic ­tumors represent a mix of circumscribed JPAs and ­diffuse, sometimes bithalamic fibrillary tumors; up to 40% of ­thalamic

gliomas are histologically malignant (i.e., WHO grade III to IV, anaplastic astrocytoma or ­ glioblastoma).122-124 ­  Biopsy alone has been the standard surgery for most thalamic ­ astrocytomas because knowledge of the tumor histology is important in guiding therapy.95,125,126 However, Bernstein and colleagues127 reported that major resections were performed in up to 33% of patients with circumscribed thalamic tumors, and the incidence of major operative morbidity was only 7%. Further experience confirmed that limited or aggressive resection might each have a role in the treatment of low-grade thalamic gliomas.128 The role of radiation therapy in childhood astrocytomas is evolving. It is thought to depend on tumor histology and site, age of the patient, and presence or absence of clinical signs. Several reviews129 have shown that radiation therapy

844

PART 10  Central Nervous System

G

H

I FIGURE 33-5.—cont’d  Sagittal slices are shown for 3D CRT electrons (G), 3D CRT photons (H), and protons (I). Isodose lines are designated for 3D CRT and IMRT: green, 54 Gy; orange, 35 Gy; blue, 20 Gy; red, 15 Gy; yellow, 10 Gy; purple, 5 Gy. Isodose lines are designated for protons: orange, 54 Gy; green, 35 Gy; light blue, 20 Gy; aqua/green, 15 Gy; violet, 10 Gy; dark purple, 5 Gy. (From Lee CT, Bilton SD, Famiglietti RM, et al. Treatment planning with protons for pediatric retinoblastoma, medulloblastoma, and pelvic sarcoma: how do protons compare with other conformal techniques? Int J Radiat Oncol Biol Phys 2005;63:362-372.)

is effective in terms of tumor response, prolonged progression-free interval, and long-term disease control. A joint CCG-POG prospective trial of postoperative irradiation for incompletely resected astrocytomas was abandoned in the early 1990s because the proportion of children in whom gross total or near-total resection could be achieved had increased substantially and because the wisdom of performing irradiation before symptomatic or radioimaging-documented progression of disease precluded uniform ­acceptance of irradiation as a randomized option. Preliminary experience in that trial suggested that observation might be appropriate after substantial resection because disease seems to progress in only a minority of such cases  3 to 5 years after surgery126 (J. Wisoff, personal communication, 2001). For some patients with central low-grade gliomas (e.g., tectal plate or midbrain; less often, hypothalamic or thalamic), cautious observation can identify indolent, stable tumors that may not require immediate intervention if the child is neurologically stable.106,130 However, this  approach implies a commitment to intervene with surgery or radiation, or both, when clinical or radiographic  evidence of tumor progression appears. The systematic use of postoperative irradiation has also been questioned in adults.125,131-133 Adults more often have sizable peripheral cerebral hemispheric tumors. Outcome at 5 and 10 years seems to be improved for patients given radiation therapy for incompletely (as opposed to

completely) resected tumors.131-134 A trial by the European Organisation for Research and Treatment of Cancer to evaluate postoperative observation versus irradiation for WHO grade I and II hemispheric lesions in adults showed no convincing advantage for early irradiation or for lowdose (45 Gy) versus high-dose (60 Gy) therapy.135 The effect of radiation therapy is well documented for central supratentorial lesions. Neurologic improvement has been observed in 80% of patients after irradiation for diencephalic tumors. Objective reduction in the size of ­supratentorial astrocytomas has been confirmed in ­approximately one half of the patients studied by prospective CT assessment. Long-term survival of patients with biopsy-proven diencephalic lesions usually has required irradiation to therapeutic dose levels.125 Limited-volume, fractionated irradiation has been reported to control most such lesions, with toxicity apparently being more limited in the pediatric age group.136

Optic Pathway Glioma Tumors of the optic nerves and chiasm are low-grade ­astrocytomas that occur predominantly in young children, often younger than 3 to 5 years. In 25% to 30% of cases, the tumors are associated with neurofibromatosis.137 Tumors of the optic nerve often behave as hamartomatous lesions, although up to 50% may extend proximally toward the chiasm. Chiasmatic gliomas manifest with visual and ­often

33  The Brain and Spinal Cord

h.t.

A

h.t.

845

of ­ patients receiving radiation therapy, and ­ neuroimaging shows a ­ reduction in the size of the lesion in these ­patients.145,146 Survival rates after irradiation for chiasmatic gliomas ­reportedly range between 75% and 100% in large series of patients followed for 5 to 10 years.138,143-145 Controversy regarding management appropriately surrounds the often-benign clinical course of the disease in relation to the potential for treatment-related morbidity with surgery or irradiation. Exophytic JPAs arising in the chiasmatic or hypothalamic region are amenable to  incomplete surgical resection147; the potential advantage of partial resection lies in delaying more definitive irradiation. Particularly among younger children, a balance between disease progression and the potential risk of adverse endocrine and intellectual effects from irradiation has favored a conservative approach.138,143 Packer and colleagues148-150 described their experience with chemotherapy, ­particularly the carboplatin and vincristine regimen. In most young children, previously progressive or symptomatic disease was stabilized with chemotherapy; objective responses were documented in up to 50% of cases, and the median time to disease progression was in excess of 3 years. Initial management with relatively toxicity-free chemotherapy approaches is logical for children with progressive symptomatic chiasmatic or hypothalamic tumors, certainly for those younger than 3 to 5 years and possibly for all children before puberty. The frequency and durability of disease control are not clear for children for whom irradiation is begun when their disease progresses during or after chemotherapy. Outside the investigational setting, the treatment of choice remains primary irradiation for children between 5 and 10 years old because long-term disease control can be achieved in up to 90% of these cases.138,143-145

Radiation Therapy Technique

B FIGURE 33-6.  Midline sagittal T1-weighted MRI scans depict a hypothalamic–optic chiasmatic juvenile pilocytic astrocytoma that filled much of the sella and suprasellar region at ­diagnosis (A) and at 2 years after completion of radiation therapy (54 Gy delivered in 30 fractions in a multifield, three-dimensional conformal approach) (B). The involution is typical of juvenile pilocytic astrocytomas at this site, including the fairly long interval to reduction. h.t., hypothalamic tumor.

with endocrine signs; it is often difficult to distinguish chiasmatic from hypothalamic tumors. In collected reports of 500 cases, 25% of patients with chiasmatic involvement died of tumor-related events.138,139 In an often-quoted ­series of 36 children, Hoyt and Baghdassarian140 indicated that chiasmatic tumors were indolent and nonthreatening. However, follow-up at a median of 20 years documented a 25% tumor-related mortality rate in a population weighted toward the apparently more benign tumor associated with neurofibromatosis.141,142 The efficacy of radiation therapy for tumors of the ­optic chiasm with or without hypothalamic involvement has been documented in numerous series.138,143,144 Visual acuity or visual fields are improved in approximately 25%

Radiation therapy for low-grade astrocytomas uses local target volumes to encompass the typically discrete primary tumor quite narrowly.136,151,152 Uncommonly, low-grade  tumors diffusely infiltrate one or more cerebral lobes, requiring more wide-field or total cranial irradiation (Fig. 33-7).  Low-grade, often pilocytic, astrocytomas are associated with multifocal disease or subarachnoid involvement at ­diagnosis (or, less commonly, at recurrence) in 15% to 20% of patients with childhood astrocytomas.153-155 Multifocal or metastatic disease is rarely confined to the spinal subarachnoid region; multiple disease sites are almost always ­apparent on cranial MRI studies, obviating any requirement for spinal imaging or staging in patients with this tumor. Specific histologic subtypes of tumors (i.e., pleomorphic xanthoastrocytoma and meningocerebral astrocytoma) can directly infiltrate the contiguous meninges, but this seems to be associated with an indolent, benign course after surgery alone.156-158 Establishing the target volume requires detailed MRI. A 1-cm margin beyond the abnormalities shown on T1- or T2-weighted images seems adequate on the basis of early follow-up data from patients for whom three-­dimensionally planned and delivered radiation techniques were used.136,151,152,159 A phase II trial by the COG is using MRI/ CT fusion for target delineation, a 0.5-cm margin for clinical target volume, and a 0.3- to 0.5-cm margin for ­ planning ­target volume. The planning target volume is treated by using a conformal technique to 54 Gy in 30 fractions.

846

PART 10  Central Nervous System

FIGURE 33-7.  MRI scan demonstrates the diffuse, multilobar involvement of a biopsy-proven fibrillary astrocytoma. Although predominantly in the left temporal lobe, the lesion also extends throughout the left parietal, right parietal, and mesencephalic regions. Objective improvement in response to irradiation is ­often transitory. This type of diffuse cerebral tumor, sometimes called gliomatosis cerebri, is a supratentorial lesion analogous to the more typical diffusely infiltrating pontine glioma.

A dose-response relationship for astrocytoma has not been well defined. In most series, a radiation dose of ­approximately 50 Gy delivered in 1.8-Gy, once-daily fractions has proved adequate.129,138 However, in statistical analyses of rather sizable series of patients, Albright125 and Laws108 and their associates observed an apparent benefit conferred by doses above 40 Gy. Justifying dose levels above 54 to 55 Gy in children is difficult in view of the increase in risk at radiation levels approaching cerebral tolerance. For adults with supratentorial astrocytomas, Shaw and colleagues,134 having shown a dose-response relationship indicating improved survival after at least 53 Gy, recommend levels of 55 to 60 Gy. However, two randomized studies of adults with low-grade gliomas testing 50.4 Gy versus 64.8 Gy or 45 Gy versus 59.6 Gy showed no significant difference in outcome between the lower doses and the higher doses.135,160

Prognosis Laws and colleagues108 reviewed the Mayo Clinic experience with 461 patients of all ages with supratentorial astrocytomas treated between 1915 and 1975. The most important correlates of 10- and 15-year survival were age (86% long-term survival for children and adolescents younger than 20 years versus 33% for those 20 years old or older) and the degree of tumor involvement, reflected by the neurologic deficit and the extent of resection  (37% survival rate for patients after biopsy or limited removal compared with 55% for those after subtotal or ­total resection). In Shaw’s review134 of the same Mayo experience with adult patients with supratentorial ­astrocytomas,

a role for radiation therapy was clearly indicated for most patients with astrocytomas, with 5- and 10-year survival rates of 68% and 39%, respectively, after at least 53 Gy. Patients treated with less than 53 Gy or no irradiation had a 5-year survival rate of 47% (both groups) and 10-year survival rates of 27% and 11%, respectively. The long-term potential for radiation therapy to control disease specifically in children with hemispheric astro­ cytomas has been documented in reviews by Bloom95 and Jenkin96 and their colleagues, who independently reported that the survival rate reached a plateau between 5 and 15 years after postoperative irradiation. The benefit of adding irradiation after incomplete resection for children with ­supratentorial hemispheric tumors was made apparent in the Pittsburgh series reported by Pollack and colleagues.161 Site-specific survival rates suggest that patients with hypothalamic and chiasmatic astrocytomas enjoy longer survival times than do patients with most other types of supratentorial astrocytomas. Bloom and colleagues95 cited 5-year survival rates of 60% to 70% for patients with thalamic or hypothalamic tumors after primary irradiation. Albright and colleagues125 reported a mean survival duration of 5.1 years after radiation therapy for patients with diencephalic gliomas compared with a mean survival time of less than 1 year for patients who did not undergo irradiation. The virtual lack of tumor progression beyond 5 years after radiation therapy further suggests a role for radiation therapy as a curative modality in the primary management of hypothalamic or chiasmatic tumors.145,162

Brainstem Gliomas Tumors of the brainstem most often involve the pons (70%), followed by the medulla (15% to 20%) and the midbrain (10% to 15%).163,164 Brainstem gliomas are characterized as diffusely infiltrating or focal, depending on the pattern of involvement; focal tumors are well demarcated, confined to one anatomic segment of the brainstem, and occupy less than one half of the brainstem.165 Most focal tumors are JPAs histologically; they occur more often in the midbrain or medulla but can arise in the pons.163,166168 Pontine gliomas are largely diffuse (Fig. 33-8).164,165 Dorsally exophytic lesions are relatively benign tumors that usually arise at the pontomedullary junction or from the medulla; histologically, they are JPAs.166-169 In the past, brainstem tumors have been treated conservatively, with histologic confirmation limited to focal or cystic lesions for which knowledge of the histology or drainage might facilitate treatment.163,164 Findings from biopsied cases indicate that malignant glioma exists in 40% of patients at diagnosis164; the remaining tumors are largely fibrillary astrocytomas.163,170 At autopsy, however, malignant gliomas are found in more than 80% of cases.164,170 Radiation therapy is primarily used for diffusely infiltrating brainstem gliomas. Substantial neurologic improvement occurs in 60% to 70% of patients with diffusely  infiltrating pontine gliomas, although long-term survival is limited to 10% to 20% of patients with these gliomas.164 Dorsally exophytic tumors are usually identified during surgery for a tumor that otherwise fills much of the fourth ventricle. The lesions in more than one half of the patients remain indolent after incomplete resection; the 40% or more of tumors that do progress are most often controlled with radiation (Fig. 33-9).167,168

33  The Brain and Spinal Cord

A

847

B

FIGURE 33-8.  A, Sagittal MRI scan depicts a diffusely infiltrating brainstem glioma involving much of the pons and extending toward the midbrain and the medulla. B, A typical two-dimensional treatment approach is shown. Use of three-dimensional conformal techniques also may be advantageous.

Radiation Therapy Technique Brainstem gliomas tend to infiltrate longitudinally, usually extending beyond the pons to involve the midbrain or medulla and extending peripherally into the cerebellum or medial temporal lobes (see Fig. 33-8). Irradiation is limited to the brainstem and immediately adjacent neural tracts. The infiltrative nature of these tumors can be better appreciated on T2-weighted MRI studies. Clinical target volumes are defined with 2-cm margins along the tracts of infiltration (i.e., into the midbrain or medulla, through the peduncles, and into the adjacent cerebellum).171 Given the poor overall outcome for patients with pontine lesions, many centers continue to use two-dimensional, opposed lateral fields in this setting, providing a margin around the known tumor extent. The standard dose for brainstem gliomas is 54 to 56 Gy, given in a conventional fractionation scheme. In a series of trials, hyperfractionated doses given to total doses of 66, 70 to 72, and 76 to 78 Gy indicated no real benefit from the higher doses.172-174 Most long-term survivors have been adults or children with focal brainstem lesions often ­located outside the pons (i.e., in the midbrain or medulla).172 The decade-long experience with hyperfractionated therapy for these tumors has encouraged neurosurgeons, radiation oncologists, and pediatric oncologists to enroll patients in multi-institutional protocols. In a randomized trial conducted by the POG, conventional fractionation to 54 Gy was documented to be as effective as 70 Gy given in twice-daily 1.2-Gy fractions.175 This trial also documented the apparent negative effect of irradiation in combination with concurrent cisplatin in this setting.175,176

Prognosis The 5-year overall survival rates are 10% to 20% for  children with brainstem gliomas and 20% to 30% for adults.172-174 Median survival after irradiation is only 8 months for patients with classic intrinsic tumors in the pons. Hoffman and colleagues166,167 initially identified as highly favorable a subset of children in whom exophytic tumors extended into the fourth ventricles. However, it is

FIGURE 33-9.  Sagittal MRI scan shows a dorsally exophytic brainstem glioma. This discrete, briskly enhancing tumor fills much of the fourth ventricle, attaching along the posterior ­aspect of the brainstem at the pontomedullary junction. ­Invasion may not be apparent on imaging, and at surgery, ­lesions often are found to blend imperceptibly with the underlying brainstem.

now understood that these tumors are usually JPAs168; survival after surgery, with or without radiation, approaches 90% in this selected group.166-168 Intrinsic pontine tumors, constituting most of the classically identified brainstem gliomas, are usually associated with multiple cranial nerve palsies, diffusely hypodense lesions on CT or MRI studies, and a high-grade histologic appearance indicative of an infiltrating, aggressive neoplasm. Long-term survival rates

848

PART 10  Central Nervous System

for the latter group have consistently been less than 15%. Less common intrinsic tumors that are focal and confined to a segment of the pons or midbrain are associated with slightly better long-term survival rates.165 Chemotherapy regimens have little effect on brainstem gliomas. Adjuvant therapy did not affect outcome in the only published randomized trial.177

Ependymomas arise from the ependymal cells lining the ventricular system and the central spinal canal. The tumor characteristically includes ependymal rosettes and perivascular pseudorosettes. Ependymomas are differentiated or low-grade lesions; variants include cellular, clear cell, tanycytic, and papillary types.69 In one series, 15% to 30% of ependymomas were classified as anaplastic tumors (i.e., WHO grade III), marked by increased cellularity and  mitotic rate. The tumors usually are well demarcated but can show parenchymal invasion.69,115 Although ependymomas were regarded by leading neuropathologists in the 1980s as being of little clinical significance, an association has since been recognized between an anaplastic histology and a less favorable outcome.186-192 Ependymoblastomas are rare embryonal tumors and are discussed with the ­other supratentorial PNETs. Surgery is the initial intervention for most intracranial ependymomas. Even extensive tumors of the fourth ventricle can often be grossly resected; total resection of tumors extending into or arising in the cerebellopontine angle has been achieved.180,190 The advantage of near-total or total resection is apparent, with considerable improvement in disease control in children in whom complete resection was achieved.189-191 The price of this control is a significant incidence of new postoperative neurologic deficits, including the posterior fossa syndrome and major new bulbar signs (e.g., palsies of cranial nerves IX and X resulting in difficulty with speech and airway protection), sometimes requiring postoperative tracheostomy or the placement of a feeding tube.88 Institutional experience shows that gross total resection can be accomplished in 50% to 60% of ­patients with infratentorial ependymoma; however, a total resection rate as high as 90% has been reported.190 Postoperative irradiation is essentially routine for ­patients with ependymomas, almost regardless of age, because adjuvant radiation has been shown to increase disease control rates in several major series.190,192 However, a single experience suggests that disease could be controlled in a certain subgroup of children with supratentorial ­ ependymomas who are treated with total resection alone; in this ongoing

Ependymomas Intracranial ependymomas occur predominantly in young children, but they are also seen in older children and adults. Ependymomas also occur as primary spinal cord neoplasms, but this presentation is more common in adults. Posterior fossa ependymomas typically manifest with the same symptom triad as other posterior fossa tumors in children—headache, vomiting, and ataxia. Associated symptoms include torticollis and lower cranial nerve dysfunction. The median age of children with these tumors is 4 years; however, almost 30% of infant brain tumors (i.e., those in children younger than 3 years) are ependymomas.178,179 ­Infratentorial lesions in younger children may occur more often in girls than in boys. Supratentorial lesions more ­often manifest with seizures or focal neurologic abnormalities and are most common in adolescents and adults. Tumors of the posterior fossa occur predominantly in children; these tumors show a predilection for the fourth ventricle in young children and often involve or originate within the region of the foramen of Luschka (Fig.  33-10).179,180 Direct extension through the foramen magnum into the upper cervical spinal canal is a characteristic pattern of growth. Ependymomas disseminate into the subarachnoid space in approximately 12% of children.95,181-184 However, this manifestation is more often seen at diagnosis or as a primary site of recurrence in very young children and may be associated with an anaplastic histology; in older children, this pattern is more common after local recurrence than earlier in the disease.183,184 Supratentorial tumors often arise in regions adjacent to the ventricular system, and growth occurs largely through intraparenchymal extension.185

A

B

FIGURE 33-10.  A typical ependymoma involving the fourth ventricle. A, Axial CT scan reveals a lesion that fills much of the fourth ventricle and extends through the right foramen of Luschka. B, Sagittal MRI scan shows the tumor extending caudally along the medulla and below the foramen magnum into the upper cervical spinal canal.

33  The Brain and Spinal Cord

COG prospective trial, only a small proportion of patients with such ependymomas is being treated with observation only.193 For all lesions of the fourth ventricle or cerebellopontine angle, high rates of tumor control have been achieved only for patients treated with local postoperative radiation and radical resection.189-191,194 Long-term disease control was accomplished in approximately 70% of such children, compared with 20% to 30% of those treated with incomplete resection and similar radiation. Ependymomas respond to a variety of chemotherapeutic agents, including cyclophosphamide and vincristine, ­cisplatin and carboplatin, and etoposide. A randomized trial of adjuvant lomustine, vincristine, and prednisone ­conducted by the CCG showed no benefit from this ­chemotherapy.195 The initial experience in infants with ependymomas indicated an inverse relationship between the duration of ­ chemotherapy (equivalent to the time of initiating radiation therapy) and disease control, encouraging current institutional and cooperative group trials to limit the use of chemotherapy for this tumor; only for the youngest children (<18 months old) has a brief (3 to 4 month) course of chemotherapy been used before radiation (or second resection and radiation if measurable disease is  apparent at the completion of chemotherapy).194 For ­infants, ­ interest in chemotherapy dose escalation continues because of positive preliminary results from the second POG study of infant brain tumors, results that suggested that dose-intensified chemotherapy produced better disease control rates than did standard chemotherapy.196 In older children, the usefulness of limited-duration chemoresponsiveness in facilitating second operative resections before radiation is being examined in a COG trial. Most clinical experience with ependymomas is based on treatment volumes for infratentorial lesions that  include the entire posterior fossa and often the upper cervical spinal canal to one to two vertebral levels below the caudal tumor extent. Because recurrent disease is almost always localized to the initial tumor bed and often the site of attachment and because minimal residual disease is left adjacent to the brainstem or peduncle, current prospective investigations are targeting only the local tumor bed (with margins as tight as 1 cm) using three-dimensional conformal or intensity-modulated radiation therapy (Fig. 33-11).159 However, findings during an as yet insufficient follow-up period indicate that the limited treatment volumes should be used with caution outside a controlled setting. For supratentorial lesions, a 1- to 2-cm margin around the initial tumor extent is indicated, with potential modifications when the normal brain is reconfigured by tumor resection. Treatment volumes for differentiated and anaplastic ependymomas usually are not different.197,198 Previous interest in CSI for anaplastic infratentorial lesions has not continued into contemporary protocols or treatment standards, except in cases in which neuraxis involvement is documented at diagnosis. Classically, radiation doses of 45 to 54 Gy have been used.182,189,192 However, doses of 54 to 55.8 Gy have been reported to achieve superior results and are indicated even after gross total resection.199,200 Despite apparently encouraging results from a small POG trial of hyperfractionated radiation (70 Gy given twice daily in 1.2-Gy fractions) for patients with posterior fossa ependymomas, no substantiated role has been established for altered fractionation. The

849

potential value of dose escalation is being explored in trials testing cumulative doses approaching 60 Gy.159

Pineal Region and Germ Cell Tumors A unique group of tumors occur in the pineal region: germ cell tumors, pineal parenchymal tumors, and glial and ­ ependymal tumors (Table 33-3).69,116,201 Intracranial germ cell tumors occur predominantly in the pineal region, ­although they can also arise in the anterior third ventricular (suprasellar) region and, less commonly, in thalamic or deep cerebral hemispheric areas. This unique group of tumors manifests with clinically and histologically linked syndromes. Pineal region ­germinomas occur predominantly in adolescent boys and are ­associated with multifocal or metastatic disease in the ­ anterior third ventricular region in 10% to 40% of cases.202-206   Suprasellar (anterior third ventricular) germ cell tumors occur more broadly throughout the pediatric age group and about equally in girls and boys; these tumors are often germinomas but also include other germ cell types. ­Pineal parenchymal tumors in children are overwhelmingly ­malignant pineoblastomas, with a considerable proportion occurring in infants and children younger than 2 to 3 years as highly aggressive, widely metastasizing, small, round, blue cell tumors. Pineocytomas are decidedly less common in children occurring more often in adults.69,207-209 Regardless of histology, pineal region tumors typically manifest with Parinaud’s syndrome, consisting of near-light dissociation (i.e., pupillary responses are diminished to

FIGURE 33-11.  The treatment plan shows the isodose volumes for a three-dimensional conformal approach, using multiple non-coplanar 6-MeV photon fields to encompass the tumor within the lower aspect of the posterior fossa and the uppermost spinal canal. The target volume in contemporary protocols more often addresses the tumor bed or surgical bed than the full posterior fossa.

850

PART 10  Central Nervous System

TABLE 33-3.   Histologic Classification and Relative Incidence of Pineal Region Tumors Histologic Type*

Relative Frequency†

Tumors of Germ Cell Origin Germinoma

60% to 80%

Teratoma (differentiated, immature, malignant)

65%

Endodermal sinus tumor

18%

Embryonal carcinoma

7%

Choriocarcinoma

5%

Tumors of Pineal Parenchymal Origin

14%

Pineoblastoma Pineocytoma Tumors of Glial Origin

15%

Astrocytoma Malignant glioma Ganglioglioma (ganglioneuroma) Ependymoma *Modified from Burger PC, Scheithauer BW, Vogel FS (eds). Surgical Pathology of the Nervous System and its Coverings, 3rd ed. New York: Churchill Livingstone, 1991. †Modified from Burger PC, Scheithauer BW, Vogel FS (eds). Surgical Pathology of the Nervous System and its Coverings, 3rd ed. New York: Churchill Livingstone, 1991; Kleihues P, Cavenee WK (eds). ­Tumours of the Nervous System. Lyon, France: IARC Press, 2000.

light but are maintained with attempted accommodation), lack of upward gaze, and incomplete ability to accommodate. Obstructive hydrocephalus often occurs when the ­tumors are near the sylvian aqueduct. Suprasellar lesions can also manifest as hydrocephalus when the growing ­tumor ­obstructs the foramen of Monro; diabetes insipidus and visual field deficits are often apparent. Although sometimes debated in the past, it is important to histologically establish the diagnosis of pineal region tumors. Suprasellar lesions have usually been biopsied because of their relative accessibility and the broad differential diagnosis that includes astrocytomas of the hypothalamic or chiasmatic region and craniopharyngioma. Germ cell tumors can occasionally be definitively diagnosed on the basis of chemical markers. Any significant elevation of the alpha-fetoprotein level in the serum or CSF is diagnostic of a malignant germ cell histiotype (typically embryonal carcinoma, yolk sac tumor, or teratocarcinoma), and marked elevation of β-human chorionic gonadotrophin level (βHCG; typically more than 100 to 1000 mIU/mL) is diagnostic of a choriocarcinoma or malignant mixed germ cell tumor (with choriocarcinomatous elements).210,211 Moderate elevation of the β-HCG level (50 to 100 mIU/mL) is compatible with a germinoma, but cytologic or histologic evidence is required to confirm the diagnosis.212 After a diagnosis is established, neuraxis staging (e.g., spinal MRI, lumbar CSF cytology) is mandatory for any of the germ cell tumors and pineoblastoma.

An open craniotomy is usually required to biopsy suprasellar tumors.213 Biopsy has been possible during third ventriculoscopy, often at the time of ventriculoperitoneal shunt insertion or third ventriculostomy. Pineal region ­tumors can be approached stereotactically, but their challenging location adjacent to the major draining intracranial veins has led many neurosurgeons to prefer an open approach. Tumor resection is feasible in the pineal region, although done in only a few instances because of the high rate of potential complications. Although insertion of a ventriculoperitoneal shunt is often necessary, the notable incidence of shunt metastasis (with tumor arising throughout the peritoneal cavity),212,213 although low in absolute terms, is relatively more common for this group of tumors than for any other intracranial lesions. Performance of a third ventriculostomy, rather than placement of a ventriculoperitoneal shunt, is encouraged for accomplishing internal decompression.

Radiation Therapy Technique Radiation therapy is highly curative in patients with ­germinomas in the pineal region, anterior third ventricle, or multiple midline locations.205,211,214-220 Considerable debate, however, surrounds the appropriate radiation volume for these tumors.206 Given the recognized incidence of intraventricular and subarachnoid dissemination, craniospinal (or less often, full cranial) irradiation has been recommended for pineal region and suprasellar germinomas.205,214,221 The outcome in many series indicates that disease is controlled in up to 95% of patients treated with full neuraxis radiation as a component of their therapy (Table 33-4).203,206,215,218-225 Local irradiation, variably defined as encompassing the primary tumor site with a margin or the entire third ventricular region, has been ­associated with long-term disease-free survival rates of more than 80% (see Table 33-4).203,206,215,218-225 The incidence of neuraxis dissemination varies from less than 10% to 35%.206,214,219,221,226 No consensus has been reached regarding the appropriate radiation volume because cure rates are relatively high for local and full-neuraxis irradiation, and the toxic effects associated with the relatively low doses necessary for this disease are exquisitely age ­related.218,220 Evidence is increasing that 45 Gy delivered to the primary tumor site is adequate to ensure virtually uniform control of pineal region and suprasellar germinomas. ­Wider-field components (craniospinal or, as done at some ­institutions, full cranial irradiation) can be limited to doses of 25 Gy.216,218,219 CSI doses as low as 18 to 19.8 Gy in children with no detectable leptomeningeal metastasis from germinomas at diagnosis have been reported to be ­effective in preventing leptomeningeal recurrence.227 Considerable interest has been expressed in chemotherapy for intracranial germinomas. Although a high rate of response has been documented, durable control of disease has been achieved in only about one half of the patients treated with chemotherapy only, including some of the more rigorous treatment regimens that carry a 5% to 10% rate of toxicity-related death.222,224,225,228 CSI is often a successful salvage treatment for germinomas that recur after such therapy.218 Early studies conducted by Allen and colleagues223,229 ­established that disease control was excellent after ­combined

33  The Brain and Spinal Cord

851

TABLE 33-4.    Therapy and Outcome in Intracranial Germinomas Radiation Therapy Therapy

CSI

Local (Gy)

RT only

+

Chemotherapy and reduced RT

Chemotherapy only

Chemotherapy

5-Year NED Rate (%)

Study and Reference

50-55

85

Shibamoto et al, 1988221

+

45-55

86

Rich et al, 1985203

+

45-59

97

Hardenbergh et al, 1997219*

+

45-50

91

Bamberg et al, 1999222

CrI, VtI

50-54

91

Wolden et al, 1995206

0

40-50

87

Calaminus et al, 1994220

0

44

90

Haddock et al, 1997215

+

50

100

Merchant et al, 2000218

0

30

Cyclophosphamide

91

Allen et al, 1987223

0

40

VP-16/CBCDA; VP-16/ ifosfamide

96†

Bouffet et al, 1999224

0

30

(CBCDA or CDDP); vinblastine; bleomycin

92

Calaminus et al, 1994220

0

0

CDDP, VP-16, bleomycin

42

Balmaceda et al, 1996225

*Includes cases from Rich et al, 1985203 (CSI or CrI, dose of 15 to 44 Gy). †Event-free survival at 3 years. CBCDA, carboplatin; CDDP, cisplatin; CrI, full cranial irradiation; CSI, craniospinal irradiation (dose levels 24 to 30+ Gy for localized tumors); NED, no evidence of disease; RT, radiation therapy; VP-16, etoposide; VtI, full ventricular volume irradiation (typically to doses of 20 to 36 Gy), +, used; 0, not used.

chemotherapy (with cyclophosphamide or carboplatin) followed by limited-volume, reduced-dose radiation. Strategies incorporating moderately aggressive ­ chemotherapy and local radiation (to 30 to 36 Gy for patients showing a good partial or complete response, as shown on imaging studies and by normalization of any elevated markers) or limited-dose CSI (for those with disease dissemination) result in disease control rates exceeding 90%, with limited toxic effects.220,222,224,228 Whether such an approach offers any advantage over radiation therapy alone is unclear; prospective trials of this approach are being conducted in Europe and by the COG. It is difficult to improve on the disease control rates produced by primary irradiation, although prepubertal patients and those who require CSI before completion of skeletal growth may benefit from a combined-modality approach. In patients with malignant germ cell tumors (i.e., embryonal or yolk sac carcinoma, choriocarcinoma), treatment with radiation therapy alone has rarely controlled disease in more than 25% to 40% of cases.213,220 Combined radiation and chemotherapy seems to be beneficial in such cases, but the chemotherapy must be relatively aggressive, and the radiation dose to the craniospinal axis is usually 24 to 36 Gy, followed by a boost dose to the tumor bed of 9 to 25 Gy, depending on the response to prior chemotherapy.230,231

Prognosis A fairly favorable outcome has been observed for patients with pineoblastomas treated with radiation therapy (including CSI) and chemotherapy (including alkylating

agents or nitrosoureas in conjunction with vincristine).232 Disease has been controlled in up to 60% of patients, compared with the 0% to 10% survival rates for infants treated with prolonged chemotherapy with or without ­radiation.178

Craniopharyngioma Craniopharyngiomas are histologically benign tumors derived from squamous cell rests in the region of the pituitary stalk. More than one half of these tumors occur in patients younger than 18 years, and craniopharyngiomas represent 3% to 9% of the intracranial tumors of children. The tumor arises as an extra-axial lesion, typically a densely calcified, solid tumor in the suprasellar region with large cystic components (Fig. 33-12). These histologically classic adamantine tumors occur in children and adults.233 Less common are the noncalcified, squamous papillary type, which usually occur in adults.69,115 Total surgical resection is usually curative in patients with craniopharyngiomas.233-240 The decision about ­whether to use surgery or radiation therapy depends on the site of origin and growth of the tumor, which often damages the hypothalamus or optic chiasm and sometimes damages the adjacent major vessels. Although encapsulated, the lesion usually adheres to these vital structures, even though it does not infiltrate to the level of the tuber cinereum.234-236   Modern microsurgical techniques allow total resection in up to 75% of cases, with operative mortality rates of 1% to 3%.236,241-243 The incidence of long-term functional deficits may be higher among patients who undergo radical resection than among those treated with radiation.237,244-247

852

PART 10  Central Nervous System

Prognosis

FIGURE 33-12.  Midline sagittal non-enhanced MRI scan demonstrates a craniopharyngioma involving the upper aspect of the sella, with a solid component extending into the suprasellar region along the pituitary-hypothalamic stalk. The cystic component is visible superiorly.

Kramer and colleagues248,249 established the effectiveness of high-dose radiation after limited cyst aspiration, initially reporting 5-year survival for 9 of 10 patients and with later follow-up showing 13- to 15-year survival for 6 of 6 children.250 Subsequent series showed similarly ­impressive longterm survival data.239,251-254 Bloom and colleagues95 updated Kramer’s initial findings, reporting a 10-year survival rate of 85% for the children and a 10-year relapse-free survival rate of 80%.95 Disease control is even better in young adults (92% at 20 years for patients 16 to 39 years old at the time of diagnosis).254 These latter figures are equal to the best results observed in primary surgical series when corrected to account for cases with incomplete resection.235,243

Radiation Therapy Technique Radiation therapy for craniopharyngioma is designed to narrowly encompass the solid and cystic components of a well-delineated, central neoplasm. Much of the published information involves treatment using coronal arcs of 180 to 220 degrees; three-dimensional conformal or fractionated stereotactic radiation therapy is being used to improve dose conformity.239,251-254 Craniopharyngiomas show a radiation dose-­response ­relationship. Bloom and colleagues95,181 reported a superior outcome in response to 50 to 55 Gy compared with the response to less than 50 Gy. Doses of 50 to 54 Gy at 1.8 Gy per fraction are recommended for children. It is not clear that higher doses are necessary or desirable for adults. Doses in excess of 60 Gy are associated with significantly greater toxicity, especially in the visual system, with little improvement in survival.95,181,255 Single-fraction radiosurgery has been used more broadly in Europe than in the United States.255,256 Radiosurgery may be appropriate for children with small foci of residual disease, particularly disease that is limited to intrasellar deposits or foci of disease remote from the hypothalamus and chiasm.255,257

Excellent long-term survival rates have been attained for patients who undergo total resection or planned limited surgery and radiation. The proportion of patients in recent series who underwent successful resection approximates 75%.233-241,243,258 However, even the most vociferous advocates of surgery acknowledge the existence of residual calcifications or apparently viable, radiographically obvious tumor in up to 50% of postoperative CT studies of patients who have undergone gross total resection.234-236 The rate of disease recurrence after complete resection ranges from 10% to 30%.233-235,241-243,259 Despite the benign nature of craniopharyngioma, tumor progresses in 70% to 90% of patients after incomplete resection and without radiation at a median of 2 to 3 years.254,259,260 Primary radiation after cyst decompression or partial resection has been associated with progression-free survival rates of 80% to 95% at 5 to 20 years.95,181,238,  251-255,258,259,261 Survival rates of 90% to 100% are common among children and adults. The relative toxicity of primary surgery and radiation for craniopharyngioma are becoming increasingly apparent. Diabetes insipidus occurs in 75% to 100% of patients after complete surgical resection.233,241,246,247 Endocrine dysfunction involving other pituitary-hypothalamic hormones has occurred in 40% to 80% of patients after surgery.246,247 Hypothalamic obesity is apparent in more than 50% of long-term survivors; it is less common among those who undergo limited surgery and radiation.246,247 Undesirable personality changes are associated with resections that damage the hypothalamus. An increased endocrine deficit has been identified in patients whose primary treatment was surgery compared with those who underwent limited resection and radiation.246 Up to 25% of children show significant decreases in vision after resection241,243; reports of visual complications from radiation therapy have been anecdotal.255 Overall performance, intellectual function, and memory seem to be equal or better in the population treated by ­conservative surgery and radiation than in those treated ­surgically.237,245,253,255

Primitive Neuroectodermal Tumors In 1973, Hart and Earle262 described the cerebral PNET as a malignant embryonal tumor that affects children and young adults. The tumor accounts for approximately 3% of supratentorial neoplasms. PNETs are large, often cystic hemispheric tumors characterized histologically by a ­highly cellular, undifferentiated infiltrate that includes vascular endothelial hyperplasia and necrosis.69,262,263 In classic ­descriptions, focal areas of glial or neuronal differentiation constitute less than 5% to 10% of the tumor. The clinically similar cerebral neuroblastoma is distinguished from the Hart/Earle PNET by its degree of neuronal differentiation.264-266 The specific cerebral PNET encountered in a patient, however, must be distinguished from the broad concept of PNETs advanced by Rorke and colleagues, as discussed earlier in this chapter.69,70,162,267 Cerebral PNETs infiltrate widely despite their circumscribed gross appearance. The tumor can also arise ­multifocally. About one third of cases involve ventricular or spinal subarachnoid dissemination.261 The more

Tumors in Children Younger than 3 Years Between 15% and 20% of pediatric brain tumors occur in ­infants and children younger than 3 years. The most common tumors in this age group are astrocytomas (primarily optic chiasmatic or hypothalamic tumors; cerebellar astrocytomas are uncommon), embryonal tumors (medulloblastoma, PNET, and pineoblastoma), ependymoma, and ­ malignant gliomas.272-274 Tumors occurring almost uniquely in this age group include atypical teratoid or rhabdoid tumors and choroid plexus neoplasms (papillomas and carcinomas).275-279 Desmoplastic infantile astrocytoma and ganglioglioma are uncommon, and when they do occur, they are typically large tumors that are biologically benign.280 The operative mortality is relatively high in this age group because of tumor presentations (often sizable, midline or multilobed supratentorial lesions) and the relative lack of elasticity of the still-myelinating brain. The infant brain is particularly vulnerable to the effects of radiation therapy, particularly with regard to neurocognitive development.281 Survival rates for patients in this age group with ependymoma or medulloblastoma are significantly lower than those for other age groups because the tumor in these very young patients is often advanced at presentation and because tumoricidal radiation doses cannot be used.95,177,181,271,282-284 The use of prolonged chemotherapy after surgery to delay or obviate the use of radiation therapy for very young children has been studied for almost 20 years (Fig. 33-13).178,285-291 Trials of treatment for medulloblastoma in infants conducted by the POG, CCG, and International ­Society of Paediatric Oncology (SIOP) have ­documented that chemotherapy is effective in only a minority of ­patients; durable disease control has been achieved in 20% to 30% of patients with chemotherapy alone and in only 35% of those treated with response-defined consolidative radiation.178,288-292 Salvage radiation has achieved a surprising degree of secondary control after progression ­ during chemotherapy, but the dose required to overcome chemotherapy resistance produces late changes in ­ intelligence quotient that make this an ethically unacceptable alternative.291,293 A trial from Germany in which ­ postoperative

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

(51 ± 10%) (21 ± 8%)

0

A Intelligence quotient

­ ifferentiated cerebral neuroblastoma has been associated d with CSF metastasis in an equal proportion of patients at autopsy, although isolated involvement of the subarachnoid space is uncommon.261,264,266,268 Studies conducted by the CCG have documented the efficacy of combining chemotherapy (the tested regimen included lomustine, prednisone, and vincristine) with CSI for children with cerebral PNETs who are more than 3 or 4 years old.232,263,267,269-271 Long-term survival ­approached 60% in this group. For infants and young children, ­ therapeutic approaches are comparable with those used for ­ other embryonal CNS lesions in this age group (see “Tumors in Children Younger than 3 Years”). Despite anecdotal reports of disease control after cranial irradiation, CSI is considered standard therapy for supratentorial PNETs, including cerebral neuroblastoma and pineoblastoma.232,268 The doses and techniques used for supratentorial PNETs are similar to those described for medulloblastoma, with typically a 2-cm margin planned around the tumor bed to define the clinical target volume.

Probability

33  The Brain and Spinal Cord

4

Survival

Progression-free survival

6

8

10

1

Years postdiagnosis 120 90 60 30 0

B

2

853

4

8

Time from diagnosis in years

FIGURE 33-13.  A, Survival and progression-free survival for  29 infants and children younger than 4 years with medulloblastoma. During the study period (1984-1995), prolonged chemotherapy was used in the hope of delaying or ­obviating the ­substantial neurotoxicity associated with irradiation in young children. The stable progression-free survival rate ­ after 18 months (21%)  reflects the use of irradiation as ­adjuvant therapy (for those with resected or stable disease during 12 to 24 months of chemotherapy); the 51% overall survival rate at 5 years reflects a high salvage rate obtained from the use of craniospinal irradiation for persistent or progressive disease. B, Among the 19 patients who survived for more than 24 months, the high survival rate came at the cost of a steady decline in full-scale intelligence quotient scores, from a pretreatment level ­ approximating 90 to a level just above 60 at 7 years. (From ­Walter AW, Mulhern RK, Gajjar A, et al. Survival and neurodevelopmental outcome of young children with medulloblastoma at St. Jude Children’s Research Hospital. J Clin Oncol 1999;17:3720-3728.)

i­ntravenous and intraventricular chemotherapy without radiation ­ therapy was used for infants with medulloblastoma showed 5-year progression-free survival rates of 82%, 50%, and 33% for infants who had complete resection,  residual tumor, and macroscopic metastases, respectively.294 Because these results seem to be better than those in any previous report, they will need confirmation. For infants with ependymoma, earlier intervention with ­radiation after prolonged chemotherapy may help control disease; however, this strategy too seems to work in only a minority of cases.194 Chemotherapy alone has been an unsuccessful postoperative intervention.295 Reports of survival of patients with pineoblastoma and atypical teratoid or rhabdoid tumors treated with this approach have been ­anecdotal.178,194,232,296 The current generation of infant brain tumor trials foc­ uses on the judicious use of irradiation. Based on ­preliminary data for children with recurrent ­ medulloblastoma, the frontline approach in the Pediatric Brain Tumor ­Consortium and COG trials is to introduce early, limited-volume, threedimensional conformal radiation therapy to the posterior fossa and primary tumor site alone for patients with M0 medulloblastoma and to the local tumor bed only for those

854

PART 10  Central Nervous System

with supratentorial PNETs.292 For patients with atypical teratoid or rhabdoid tumors, targeted radiation volumes are used after just 2 months of chemotherapy for children 12 months of age and older.296 For those with ependymomas, the trend is toward immediate postoperative radiation, at least for patients with tumors that can be grossly resected.159

Tumors Common in Adults Malignant Gliomas: Anaplastic Astrocytoma and Glioblastoma Multiforme The dominant primary intracranial tumors that occur in adults in terms of frequency and mortality are the ­malignant gliomas. These tumors occur in the cerebral hemispheres as sizable, rapidly growing lesions with a characteristic ringlike, enhancing appearance on CT or MRI, with central necrosis, infiltrating margins, and surrounding low-density changes (Fig. 33-14). Malignant gliomas occur in all age groups but predominate in the fifth and sixth decades, and they account for 35% to 45% of all adult brain tumors. Histologically, malignant gliomas are heterogeneous neoplasms composed of fibrillary astrocytes; gemistocytes; large, bizarre glial cells; and small anaplastic cells.69,115 In Kernohan and Sayre’s classic categorization297 of astrocytic neoplasms, which defined these tumors as grade III and grade IV astrocytomas, malignant gliomas were thought to represent progressively dedifferentiated astrocytic tumors marked by a certain degree of pleomorphism, hyperchromaticism, and mitosis. In contemporary neuropathology, malignant gliomas are classified as anaplastic astrocytoma or the more malignant glioblastoma multiforme, with the latter identified by the presence of tumor necrosis.298 ­Anaplastic astrocytomas account for 10% to 15% of malignant gliomas; more than 85% are glioblastomas.299

A

Malignant gliomas typically infiltrate the adjacent normal brain. Detailed comparisons of imaging and histologic findings have established the anatomy of malignant gliomas. Histologically, the central low-density region seen on neuroimaging studies is a necrotic and hypocellular area. The surrounding ring of enhancing tissue is composed of densely cellular tumor. The hypodense area beyond the enhancing tissue seen on CT scans is hypocellular and edematous but infiltrated by small anaplastic T cells.300 In 50% of cases, scattered tumor cells immediately surrounding the hypodense volume are evident histologically.301 The often sizable area of increased signal on T2-weighted MRI similarly contains infiltrates of viable T cells.302 Tumor cells characteristically extend along neural tracts and across the corpus callosum.301 Distant subarachnoid or subependymal extension occurs in 5% to 9% of patients with malignant gliomas.303,304 Surgery for malignant gliomas is usually limited to subtotal resection. Macroscopically, complete removal is achieved in only 10% to 20% of cases.305-308 The extent of residual tumor visualized on postresection CT scans seems to correlate with outcome, with increased time to progression and survival being associated with smaller residual tumor volumes among adults and children.297,309 Radiation therapy has been the most effective treatment for these aggressive lesions, but it usually can only delay disease progression or recurrence. Early clinical trials conducted by the Brain Tumor Study Group confirmed the clinical efficacy of radiation. Median survival increased from 14 weeks after surgery to 36 weeks for ­patients given postoperative radiation.306 The survival rate at 1 year was 24% for patients also treated with ­radiation therapy, compared with only 3% for patients who had surgery only.306 One fourth to one half of patients who receive radiation therapy show clinical improvement. Up to 60% of patients are able to return to work and achieve full function.310-312

B

FIGURE 33-14.  MRI scans show a glioblastoma multiforme arising in the left thalamic region and extending just beyond midline.  A, Gadolinium-enhanced T1-weighted MR image shows a ring-enhancing periphery with a dark, necrotic center. B, T2-weighted MR image shows surrounding edematous changes.

33  The Brain and Spinal Cord

Despite documented improvement in time to progression and survival, the response to radiation illustrated by imaging studies is complex and rarely indicative of rapid or significant disease reduction. Essentially all glioblastomas and 65% to 80% of anaplastic astrocytomas recur within 2 to 5 years and result in rapid death, causing malignant gliomas to live up to their reputation of being one of the most aggressive and lethal tumors.310 Although the anatomy of malignant gliomas suggests that disease extension or metastasis is the source of recurrence, recurrent disease almost always arises from within the central or enhancing portion of the tumor.313-317 The central tumor is relatively radioresistant, perhaps because of an inherent lack of radiosensitivity or a hypoxic, relatively radioinsensitive necrotic core.316,317 This pattern of intralesional recurrence has stimulated several lines of clinical investigation to overcome hypoxic cell resistance; considerable interest has also been expressed in exploiting the time-dose relationship to identify an altered fractionation regimen of value in treating this tumor.299,312 Several trials have tested modifications in radiation delivery, including physical modalities other than photons and local boost techniques incorporating interstitial brachytherapy or stereotactic radiosurgery.318-329 Early trials of radiosensitizers focused on the nitroimidazoles misonidazole and etanidazole.330 Later studies have investigated carbogen and nicotinamide. Despite some improvement realized from use of these agents with suboptimal radiation, clinically meaningful benefit was not achieved.331,332 Long-standing interest in the halogenated pyrimidines bromodeoxyuridine and iododeoxyuridine has resulted in several trials of their use with conventional and altered fractionation, but only a marginal benefit has been suggested by the results.333-336 The use of radiation with high-LET particles has been examined in clinical trials as a possible way of overcoming the apparent oxygen-related radioresistance of these tumors. However, the use of fast–neutron beam radiation has produced no significant improvement over conventional  photon radiation.318,322,337 The rationale for using proton radiation stems more from the ability to restrict the volume of tissue irradiated than from its being less affected by oxygenation; trials of doses as high as 90 Gy suggest a potential benefit in terms of central tumor control, but no early demonstration of improvement in outcome has been seen.338 Stereotactic interstitial implantation of radioactive sources has been done at several European centers since the 1950s. Over the past 2 decades, prospective studies of temporary stereotactic brain implants using high-activity iodine 125 (125I) have evaluated the toxicity and efficacy of volume-limited, high-dose radiation delivery.319-321 Among selected patients with limited-volume recurrent malignant gliomas, geometrically planned implantations of radiation sources delivering an average of 60 Gy to the enhancing lesion over 6 days resulted in survival durations of approximately 1 year for those with glioblastomas and 1.5 years for those with anaplastic astrocytomas.319-321,326,339 ­Results of the Brain Tumor Cooperative Group’s prospective randomized trial suggested that interstitial boost radiation could be beneficial in appropriately selected cases.340 Patients with malignant gliomas who are potentially suitable for local types of therapy (e.g., brachytherapy, radiosurgery) are those with a relatively favorable prognosis,

855

constituting 20% to 40% of cases. The criteria for implantation include Karnofsky performance status scores above 70, unifocal supratentorial lesions less than 6 cm in the greatest diameter, and no involvement of the midline, corpus callosum, or subependyma.319,321 A retrospective review of unselected patients treated with standard external beam radiation therapy to the brain (EBRT) revealed that outcome was significantly better among ­ implant-eligible patients with glioblastoma (median survival of 14 months) than among those who were ineligible (median survival of less than 6 months),341 indicating that any perceived improvement in outcome in patients attributed to the brachytherapy may be artificial. However, median survival duration and 2- and 3-year survival rates for patients with glioblas­ toma treated with 125I brachytherapy seem to be superior in single-institution experiences and in comparison with the previously described findings.319,321,341 Comparable data for anaplastic astrocytomas are less ­convincing.321,341 Symptomatic local necrosis occurs in 40% to 50% of patients with interstitial implants, and this often ­requires surgical intervention.320,321,326,327,332 Ultimate ­survival has been reported to be superior in patients who  require a ­second operation, which presumably reflects the ­combination-treatment effect.320 Stereotactic radiosurgery has been used to achieve a highdose boost. Eligibility criteria are selective, typically including single lesions less than 4 to 5 cm in diameter without subependymal extension and more than 5 mm away from the optic chiasm or brainstem. In the initial U.S. ­ series, which included a high proportion of patients who had undergone limited surgical resection, the median survival time was at least equal to that among patients with implants: 26 months for those with glioblastoma and more than  36 months for those with anaplastic astrocytoma.342,343 In a multi-institutional trial of 189 patients, the median survival time was 86 weeks for those with glioblastoma who met all criteria for gamma knife intervention, compared with 40 weeks for those with more extensive tumors or a less favorable performance status score.323,328,341 The incidence of symptomatic necrosis may be less than that for patients with interstitial implants.271 Several trials in the United States, Europe, and Japan have assessed modifications in the radiation time-dose ­rel ationship.299,312,338,344-347 Despite an early suggestion that ­hyperfractionated delivery to higher doses might be beneficial, subsequent studies conducted by the RTOG have shown no substantial benefit from such therapy. A large randomized dose-escalating trial conducted by the RTOG tested hyperfractionated doses of 64.8 to 81.6 Gy, given in 1.2-Gy fractions twice daily with an interfractional interval of 4 to 8 hours. The results indicated that the ideal dose response occurred at 72 Gy, although the superiority of this schedule over 60 Gy delivered in conventionally fractionated radiation therapy was inapparent.344 Patients who received doses of 75 to 81.6 Gy fared less well.344 Without documentation of any reduction in neurotoxicity from hyperfractionated delivery, few objective data indicate that the therapeutic ratio and effect of hyperfractionation are greater than those for conventional fractionation.344,348,349 Alternative time-dose regimens have included accelerated hyperfractionation (1.5- or 1.6-Gy fractions given twice daily), accelerated fractionation (1.9- or 2.0-Gy fractions given three times daily), and hypofractionation ­ (fractions

PART 10  Central Nervous System

of 2.5 to 6 Gy).332,334-347 Although the hypofractionation regimens may expedite palliation in patients with advanced or unfavorable conditions, these regimens have provided no overall benefit. Chemotherapy has been investigated extensively for the treatment of glioblastomas. As sole adjuvants to ­ surgery, ­nitrosoureas have been ineffective.306 When combined with radiation, nitrosoureas or combinations of agents (e.g., carmustine-hydroxyurea plus procarbazine-­teniposide, ­carmustine-hydroxyurea, vincristine-­procarbazine) have been associated with only marginally improved survival times for patients enrolled in prospective studies.306,311,350 Despite one report of encouraging early results with penciclovir,351 a randomized trial conducted by the Medical ­Research Council in the United Kingdom of more than  670 patients showed no survival benefit for patients given penciclovir in addition to standard radiation therapy.352 Two randomized trials testing concurrent temozolomide plus  radiation therapy with radiation therapy alone ­demonstrated a significant survival benefit for the combination.353-354 In a trial by the European Organisation for ­ Research and Treatment of Cancer, the survival benefit from temozolomide was more apparent when the MGMT gene was  epigenetically silenced.355 The current RTOG trial uses MGMT status in the tumor as one of the criteria for stratification. Many biological agents are available for the treatment of malignant gliomas, including compounds with specific biological targets (e.g., epidermal growth factor, platelet­derived growth factor) and antiangiogenesis agents.356 One ongoing trial is using a vaccine against a truncated form of the epidermal growth factor receptor (EGFRvIII) for patients whose glioblastomas have undergone gross total resection and demonstrate overexpression of EGFRvIII. Studies of local chemotherapy, specifically carmustine-impregnated wafers (Gliadel), have yielded ­inconclusive results in patients with malignant gliomas, producing an apparent response but being of uncertain value as an adjuvant therapy.357

Radiation Therapy Technique Radiation therapy remains the most effective single agent in prolonging survival in patients with malignant gliomas. It should consist of standard EBRT with wide local coverage of the neoplasm.310,314 Anatomic studies of tumor extent, imaging analyses of patterns of recurrence, and clinical trials have shown that the appropriate initial target volume should extend 2 to 3 cm beyond the low-density periphery shown on CT scans or 2 cm beyond the abnormal signal shown on T2-weighted MRI studies.301,302,314,315,350 Margins of less than 2 cm are theoretically inadequate because of the histologic nature of the tumor301 and the way in which the initial sites of disease progress.315 The use of limited high-dose treatment volumes delivered by three-dimensional conformal or proton radiation may be associated with a relative increase in the peripheral failure rate.338,358 Central or intralesional disease progression is less common in patients with interstitial implants. Studies of patterns of failure in series of patients treated with brachytherapy have shown that disease recurs primarily in the tumor margin, often within 2 cm of the high-dose implant; the incidence of distant intracranial recurrence is also relatively high.316,317 Dose-response data were established early in clinical trials involving patients with malignant gliomas. Walker

70 25th percentile

60 Survival in weeks

856

50 Median

0.004

40

0.110

30

0.174 75th percentile

20 10 4500

5000

5500

6000

Rads

FIGURE 33-15.  Dose-response analysis for survival of adults with malignant gliomas. (Data from Walker MD, Green SB, Byar DP, et al. Randomized comparisons of radiotherapy and nitrosoureas for the treatment of malignant glioma after surgery. N Engl J Med 1980;303:1323-1329.)

and colleagues306 showed that median survival time improved from 28 weeks for patients given 50 Gy in 25 to  28 fractions to 42 weeks for patients given 60 Gy in the first Brain Tumor Study Group studies (Fig. 33-15).306 Three-dimensional conformal and proton-based regimens have made it possible to escalate doses to 70 to 90 Gy; however, although central tumor control may be improved, no increase in survival has been found.338,358 As summarized earlier, none of the altered fractionation regimens has been shown to improve outcome over that achieved with conventionally fractionated radiation to 60 to 65 Gy (in 1.8- to 2-Gy fractions given once daily).345 No data exist to suggest that these recommendations for radiation therapy for adults be adjusted for children; however, the requirement for high-dose radiation to reasonably large volumes limits the use of this modality in young children with malignant gliomas.

Prognosis Reports of survival beyond 2 years for adults with glioblastoma remain anecdotal.359 However, approximately 15% to 25% of patients with anaplastic astrocytoma survive 5 years, with gradually decreasing survival beyond that time documented in most series. Levin and colleagues351 reported that adjuvant penciclovir chemotherapy is associated with a 35% survival rate at 3 to 5 years for patients with anaplastic lesions; however, these and similar results seen in other small series were not substantiated in a Medical Research Council trial.314,352,360 Other factors influencing prognosis include age (survival is progressively better among younger patients, with the best among 18 to 39 year olds, the next best among 40 to 59 year olds, and the worst among those 60 years old or older) and initial performance status.305,325 Among children with malignant gliomas, the overall progression-free survival rates at 3 to 5 years range from 15% to 20%; for the small subset with completely resected lesions, corresponding survival rates of up to 75% have been  reported.309,361,362

33  The Brain and Spinal Cord

Oligodendroglioma Oligodendroglioma is a relatively uncommon tumor that occurs in all age groups. The tumor arises most often in the frontal lobes and occasionally involves the infratentorial or spinal regions. The most common symptom is seizures. The relatively nonaggressive nature of this tumor is indicated by the median survival time, which is more than 5 years from the onset of symptoms and diagnosis.363 The co-­deletion of the short arm of chromosome 1 (1p) and the long arm of chromosome 19 (19q) has been associated with less aggressive behavior of the tumor and better response to therapy.364,365 Primary surgical resection is the initial therapy for lowgrade oligodendrogliomas.366,367 The role of postoperative radiation has been poorly defined. In a small, retrospective study, Sheline and colleagues368 found an improvement in 5-year survival rates for patients who underwent irradiation compared with those who did not (85% versus 55%), but the survival rate at 10 years was not statistically improved. Chin and colleagues369 reported a progression-free survival rate of 100% at 5 years for 24 patients completing postoperative radiation. Lindegaard and colleagues370 observed a benefit from radiation only among those who had undergone incomplete surgical resection. In a study of a similar group of patients treated during roughly the same period, Reedy and colleagues371 found no survival advantage from radiation therapy. In the absence of a prospective trial assessing the efficacy of radiation therapy, it seems justified and prudent to use postoperative radiation for subtotally resected lesions. Local fields to doses of 50 to 54 Gy approximate the treatment recommended for other low-grade tumors.106,372 Although 5-year survival rates in excess of 65% to 90% have been observed in major clinical series, the ultimate survival rate beyond 15 or 20 years was only 30% in two large reviews from an earlier era that examined the pathologic behavior of tumors from more than  500 patients.363,373 Late tumor progression and death take place more than 5 to 10 years after diagnosis. Histologic grading is of increasing importance because a role for penciclovir chemotherapy in the treatment of anaplastic oligodendrogliomas has been established.374 Longterm survival is higher for patients younger than 40 or  50 years.373

Meningioma Meningiomas are common CNS tumors, most often occurring as intracranial tumors along the convexities or in parasagittal regions. The usually benign meningioma constitutes 15% of adult CNS tumors, with a female-to-male ratio of 2.5 to 1. The natural history of this tumor was elegantly described in Cushing and Eisenhardt’s 1938 monograph,375 in which they identified the broad dural attachment (or local infiltration), growth along or into the venous sinuses, and osseous invasion or overlying reactive bone formation that typify these lesions. Angioblastic meningiomas have a less favorable outcome than the other so-called benign tumors.376 Less than 10% of meningiomas are malignant, and these lesions seem to be associated with an unfavorable outcome.376-378 Surgery is the primary treatment for most peripherally located intracranial meningiomas. Local tumor recurrence

857

after complete resection occurs in 10% to 20% of cases, related in part to the extent of dural and sinus excision.376-379 Symptomatic tumor recurrence or progression takes place in approximately 50% of patients with subtotally ­excised lesions, most often in patients with meningiomas of the sphenoid ridge and parasellar areas, which represent ­almost 30% of all meningiomas (Fig. 33-16).376,379-381 The efficacy of radiation therapy for incompletely resected meningiomas has been well documented. Wara and colleagues382 reported local tumor recurrence in 74% of patients after incomplete resection and in 29% of patients given postoperative radiation. Progression-free survival rates beyond  5 years average 70% to 90% for patients undergoing incomplete resection and radiation; disease control rates in excess of 70% to 80% at 8 to 10 years are associated with survival rates exceeding 90%.383-385 Similar outcomes have been observed among patients who receive radiosurgery alone or as an adjunct to incomplete surgery.386-389 Neurologic improvement has occurred in 44% of patients after radiation for unresected primary lesions.376 Serial CT studies have confirmed the regression of neoplasms in response to radiation therapy (see Fig. 33-16).381 In sites such as the optic nerve sheath, operative intervention may be associated with a significant functional deficit. Primary irradiation can improve tumor-related defects in visual acuity and visual fields in these patients.390

Radiation Therapy Technique Because meningiomas are locally infiltrating tumors that produce local tissue changes by means of direct tumor extension or have subclinical, limited multicentricity,391 local irradiation should include a reasonable margin despite the circumscribed appearance of these tumors on CT scans. Large tumors of the sphenoid ridge may extend beyond the cranial vault into the orbit or the facial or cervical regions. Optimal radiologic assessment of tumor extent and the use of wide inferior treatment margins are important in these cases.381 Three-dimensional conformal techniques, intensity-modulated radiation therapy, and proton therapy are suitable for the treatment of meningiomas. Carella and colleagues383 were unable to identify a doseresponse relationship for meningiomas between 50 and 75 Gy. However, excellent local tumor control has been achieved at doses approaching CNS tolerance, and the recommended treatment is a total dose of 50 to 55 Gy given in 25 to 30 fractions.376,381,384 The use of radiosurgery as the primary treatment for selected lesions has produced equally positive results; at many centers, radiosurgery is the treatment of first choice for limited-volume lesions that are difficult to approach surgically or are residual or recurrent after initial resection.386,388,389,392,393

Prognosis The benign nature of meningiomas is confirmed by the limited number of recurrences seen in patients after total surgical resection. After incomplete resection, however, the median time to clinically apparent tumor progression is 4 years after surgery. The median intervals to tumor progression in patients given radiation therapy exceed 5 to  10 years.381,383,384 Initial postoperative irradiation for incompletely ­resected or biopsied meningiomas has achieved excellent local ­tumor control, as described in the previous ­paragraphs.

858

PART 10  Central Nervous System

FIGURE 33-16.  CT scans of an orbital sphenoidal meningioma demonstrate residual disease after surgery (top row) and 2 years after irradiation (bottom row). (From Petty AM, Kun LE, Meyer GA. Radiation therapy for incompletely resected meningiomas. J Neurosurg 1985;62:502-507.)

Indications for initial or delayed radiation therapy remain controversial, although increasing information suggests that incompletely resected lesions along the cavernous sinus and those with malignant histologic features may warrant immediate postoperative irradiation.377,385,389,392 No apparent decrease in tumor control has been observed in most series of patients who undergo repeat surgery and radiation therapy for disease that recurs after initial incomplete resection.376,380,384 A more rapid rate of tumor regrowth after subtotal excision has been indicated in children, raising some concern about the advisability of delaying radiotherapeutic intervention in this age group.394,395 Identifying the histologic subtype of meningioma69,378 is important because the local recurrence rate is higher for malignant (anaplastic) meningiomas and because postoperative radiation therapy can effectively control these tumors. For example, Dziuk and colleagues377 and others378,380 have reported 15% to 80% differences in 5-year progressionfree survival rates related to radiation. A long-term disease control rate of only 25% at 5 years has been observed for patients who receive single-fraction radiosurgery, in most cases for recurrent malignant meningiomas.393

Primary Malignant Lymphoma of the Central Nervous System Primary malignant lymphoma of the CNS was previously a rare neoplasm, occurring spontaneously in conjunction with immunosuppression. However, the frequency with which it occurs has increased greatly in conjunction with

the acquired immunodeficiency syndrome (AIDS) epidemic.396-398 The tumor occurs primarily in the supratentorial region, most often as an isodense, diffusely enhancing lesion involving the basal ganglia or thalamus, the periventricular white matter, or the corpus callosum.399 Up to 15% to 40% of tumors are multifocal intracranial lesions.400,401 The CSF contains tumor cells in 10% to 20% of cases; involvement of the ocular vitreous at diagnosis or metachronously occurs in 10% to 25% of cases.400,402 The lesion is usually classified as an immunoblastic or lymphoblastic diffuse, large cell, malignant lymphoma, most often of a B-cell immunophenotype. Systematic evaluation is necessary to distinguish primary lymphoma of the brain from secondary CNS manifestations of malignant lymphoma, although intraparenchymal deposits occur more often with primary tumors than with secondary tumors. Surgery is limited, often to biopsy alone, but radiation therapy produces a high rate of objective response. Neurologic improvement can be achieved in up to 70% of patients, but median survival has been only 7 to 8 months.403 Long-term survival after radiation alone has been limited.398,400,401,404,405 Chemotherapy has shown increasing efficacy in patients with CNS malignant lymphoma, with rates of objective response often exceeding 50% to 70% in those treated with regimens including high-dose methotrexate or, less often, with conventional systemic chemotherapy regimens (e.g., cyclophosphamide, doxorubicin, vincristine, and prednisone).406-410 The combination

33  The Brain and Spinal Cord

of ­preirradiation chemotherapy and cranial irradiation resulted in median survival times as long as 60 months and 4-year progression-free and overall survival rates of approximately 50% in a moderate-size series from Memorial Sloan-Kettering Cancer Center.408,411 One drawback to the use of combined methotrexate-radiation regimens has been their considerable neurotoxicity, which sometimes produces frank dementia, particularly in patients older than 60 years.407,408,412 Given the size and often the multiplicity of most primary CNS lymphomas, the minimal target volume is often full cranial irradiation.401-404 Isolated subarachnoid spinal recurrence after cranial radiation occurs in 4% to 25% of patients.106,404 The inability to control the primary tumor within the cranium in a high proportion of patients and the need to coordinate radiation with chemotherapy have led clinicians away from considering CSI. This raises a valid argument for avoiding spinal irradiation in deference to the potential benefits conferred by chemotherapy. A radiation dose-response relationship has been ­suggested in the treatment of primary malignant lymphoma of the CNS. Murray and colleagues401 reviewed 198 cases reported in the literature and found a statistically ­significant improvement in the 5-year survival rate for  patients given doses of more than 50 Gy (42% versus 13%) (Fig. 33-17).401 However, an analysis at Memorial SloanKettering Cancer Center did not find additional benefit from doses higher than 45 Gy. Recurrence of tumor at the primary site even in patients treated with doses of 50 to 55 Gy led to a major RTOG trial, the results of which failed to demonstrate a significant added benefit from high-dose cranial irradiation.403 In small series of patients, however, the addition of radiation therapy seems to have prolonged disease control and improved survival relative to the outcome for patients given chemotherapy only.398,408,411

Prognosis Overall survival in patients with primary malignant lymphoma seems to have increased among those treated with combined, sequential chemoradiotherapy; median survival times have increased several-fold to approximately 4 to  5 years. However, most patients do die of the disease, sometimes as long as 5 years after diagnosis.408,411

Carcinoma Metastatic to the Brain Intracranial metastasis, primarily in the cerebrum, occurs in 10% to 15% of patients with cancer.413 Of the types of cancer that metastasize preferentially to the brain, cancer of the lung and breast predominate, and gastrointestinal and renal carcinomas are less common. Brain metastasis often occurs at multiple sites. In a classic autopsy series, solitary metastases were found in approximately 40% of cases, two to three lesions in 25% of cases, and four or more foci in 35%.414 Treatment of brain metastases is palliative; long-term survival can be achieved for up to 20% of patients with favorable presenting characteristics, such as age younger than 60 to 65 years, favorable performance status (typically, a Karnofsky performance status score of at least 70), control of the primary tumor, and absent or controlled extracranial metastasis.415-417 Corticosteroids alone can transiently reduce symptoms in 60% of patients.347 However, cranial

1.00 0.9 0.8 0.7 Fraction survival

Radiation Therapy Technique

859

0.6 0.5 ³50 Gy

0.4 0.3

P<.05

0.2 <50 Gy

0.1

0

4

8 12 16 20 24 28 32 36 40 44 48 52 56 60 Survival (months)

FIGURE 33-17.  Dose-response analysis for survival of patients with primary cerebral malignant lymphoma after irradiation. (From Murray K, Kun L, Cox J. Primary malignant lymphoma of the central nervous system. J Neurosurg 1986;65:600-607.)

irradiation achieves more prolonged palliation, with objective improvement in neurologic function in 50% to 70% of patients treated in this manner.319 Earlier time-dose studies showed the relative equivalence of the following treatment regimens: 30 to 36 Gy given in 10 to 12 fractions; 20 Gy given in 5 fractions; and 40 Gy given in 15 fractions.417a The current standard therapy is 30 Gy given in 10 fractions; however, randomized trials to assess the benefit of surgery or radiosurgery in selected settings favor 37.5 Gy given in 15 fractions.418,419 For patients with a solitary metastasis, the addition of surgical resection or radiosurgery has significantly improved survival time (from a median of 15 weeks to a median of  40 weeks in a randomized trial reported by Patchell and colleagues420) and the rate of local control at metastatic sites (from less than 50% for patients treated with cranial irradiation alone to 80% for patients undergoing ­resection). This increase in survival also has been associated with improved functional outcome.420 In several series, the ­outcome from radiosurgery has been similar to that from operative ­removal; in all of these series, the patients with favorable factors (e.g., age, performance status, absence of disease at other sites) are the ones who benefit the most from the addition of ­local therapy.388,418,419,421,422 Surgery and radiosurgery seem to achieve equivalent results in most reports.419,421 Detailed analyses have shown that the addition of full cranial irradiation improves survival and ­local tumor control and may be associated with long-term survival of a small but meaningful number of patients undergoing ­ aggressive local therapy.423 One randomized trial ­comparing ­radiosurgery alone

860

PART 10  Central Nervous System

with the combination of ­radiosurgery and whole-brain irradiation showed significantly fewer intracranial recurrences with the combination therapy, although survival times were similar in both groups.424 Results for patients with malignant melanoma that has metastasized to the brain have generally paralleled those seen for patients with other carcinomas; however, ­local therapy may be particularly advantageous for ­patients with few metastatic sites and favorable performance ­status.421,425 The treatment selected for patients with brain metastasis depends on the absence or presence of the clinical ­features consistently associated with improved outcome (i.e., age younger than 60 to 65 years, Karnofsky ­performance status score of 70 or higher, and the lack of apparent or symptomatic disease at the primary site or other extracranial locations). However, there does seem to be a definite advantage to surgery or radiosurgery for patients with only one metastatic focus; a similar advantage seems to exist for those with only two or three metastatic foci.388,418,419,421,422,426 In all of these patients and in those with more advanced cranial or extraneural disease, the addition of full cranial irradiation is an important and worthwhile palliative intervention.415,418,423

Spinal Cord Tumors Primary tumors of the spinal cord represent 10% to 15% of CNS neoplasms. Extramedullary tumors, including schwannomas and meningiomas, account for almost one half of the lesions in the spinal canal. Primary intramedullary neoplasms are most often gliomas or vascular tumors. Intramedullary gliomas include ependymomas (60%) and astrocytomas (30%). Ependymomas occur more ­often in the lumbar region, affecting the conus and cauda ­equina. Spinal ependymomas usually are low-grade neoplasms; those occurring in the cauda equina usually are well­differentiated myxopapillary tumors.427 The ependymomas tend to develop as discrete lesions, facilitating subtotal or gross total resection. Gross total resection led to systemic disease control in recent series with a 5-year follow-up. Aggressive surgery, however, is associated with postoperative neurologic deterioration in 15% of patients.428 Response to radiation is well documented, with 75% of patients showing improvement in neurologic status and progression-free survival rates of 70% to 100% at 10 years.429-431 A local tumor control rate of 80% with no identifiable long-term neurologic toxicity despite doses in excess of 45 Gy indicates that radiation has a role in the treatment of patients whose spinal tumors are incompletely resected.431 Patients with tumors of the cauda equina region and conus have particularly high survival rates after radiation or surgical resection.432,433 Spinal astrocytomas are evenly distributed along the length of the spinal cord. These lesions often contain cystic components and most often are low-grade fibrillary tumors. Astrocytomas are more infiltrating than ependymomas, limiting surgery to cyst decompression and partial resection.432 Epstein and colleagues,434-437 however, described operative techniques that can be used to completely resect many low-grade astrocytomas in children and adults. Aggressive operative intervention is not ­advantageous for high-grade spinal astrocytomas.429,438,439   Long-term control in patients with spinal astrocytomas treated with radiation is less predictable than that in

­ atients with ependymomas despite frequent neurologic p improvement.429,431 The disease-free survival rate at 5 to  10 years ranges from 25% to 60% among patients with spinal astrocytomas.429,438,439 However, the histologic grade affects outcome, in that patients with low-grade astrocytomas enjoy a short-term (5-year) survival rate of as high as 89%, compared with 0% among patients with higher-grade tumors (e.g., glioblastoma).438 Spinal cord tolerance is an important consideration when reviewing results of patients with intramedullary ­ tumors treated with radiation.440 Doses of approximately 50 Gy are recommended for most spinal cord tumors; however, higher doses and the suggestion of a dose-response effect at or above 50 Gy have been reported for localized spinal cord lesions.441,442 Extradural metastasis is the most common oncologic diagnosis for patients with spinal cord disease. An estimated 5% of patients who die of cancer have clinical signs of epidural metastases.443 Extradural involvement usually results from the contiguous extension of a vertebral metastasis, most often in patients with primary cancer of the breast, lung, or prostate. Less common are vertebral metastases from malignant lymphomas and myeloma or carcinomas of the kidney and gastrointestinal tract. Early evaluation with spinal MRI has been advocated for patients with pain and radiographic evidence of vertebral metastasis without objective neurologic signs.443,444 The treatment of epidural metastases is controversial, largely with respect to the use of surgery before radiation. Conventional procedures have included decompressive laminectomy, at least for patients with rapidly evolving signs or complete block shown by myelography; patients for whom the diagnosis is uncertain or the diagnosis of malignancy is not established require surgical intervention.339 Retrospective and prospective studies comparing laminectomy and postoperative radiation with radiation alone confirm an apparent advantage to surgical intervention only for patients with nonambulatory symptoms; however, there has been little difference in outcome for most patients.443,445-447 Experience with vertebral body resection in selected cases indicates improved neurologic outcome after a more aggressive operative approach.448-450 Primary irradiation is often the treatment of choice for certain histologic types of tumors known to respond relatively rapidly and for patients with residual functional deficits. Irradiation for spinal cord compression is considered a radiotherapeutic emergency. Prompt initiation of dexamethasone and high-dose (2- to 4-Gy) fractions are indicated. After two or three treatments, the dose per fraction and total dose depend on the type of malignancy and clinical status of the patient. Carcinomatous meningitis or diffuse leptomeningeal infiltration is a relatively uncommon late manifestation of systemic cancer. Primary tumors are usually carcinomas of the breast or lung; small cell lung cancer in particular is often associated with intracranial metastasis. Symptoms and signs referable to the spine predominate in 25% of patients; signs related to spinal root infiltration are often apparent.451 Treatment consists of cranial irradiation or CSI (depending on the histologic findings and coordinated use of chemotherapy), with or without intrathecal or systemic chemotherapy. Responses to treatment are common, but long-term survival results are largely anecdotal.

33  The Brain and Spinal Cord

REFERENCES    1. Schultheiss TE , Kun L E , Ang K K , et al. Radiation response of the central nervous system. Int J Radiat Oncol Biol Phys 1995;31:1093-1112.    2. Fajardo LF, Berthrong M, Anderson RE. Central nervous system and spinal cord. In Radiation Pathology. New York: Oxford University Press, 2001, pp 351-361.    3. Burger PC, Mahaley MS Jr, Dudka L, et al. The morphologic effects of radiation administered therapeutically for intracranial gliomas: a postmortem study of 25 cases. Cancer 1979;44:  1256-1272.    4. Constine LS, Konski A, Ekholm S, et al. Adverse effects of brain irradiation correlated with MR and CT imaging. Int J Radiat  Oncol Biol Phys 1988;15:319-330.    5. Valk PE, Dillon WP. Radiation injury of the brain. AJNR Am J Neuroradiol 1991;12:45-62.    6. Gutin PH. Treatment of radiation necrosis of the brain. In Gutin PH, Leibel SA, Sheline GE (eds): Radiation Injury to the Nervous System. New York: Raven Press, 1991, pp 271-282.    7. DeAngelis LM, Shapiro WR. Drug/radiation interactions and central nervous system injury. In Gutin PH, Leibel SA, Sheline GE (eds): Radiation Injury to the Nervous System. New York: Raven Press, 1991, pp 361-382.    8. Van der Kogel AJ. Central nervous system radiation injury in small animal models. In Gutin PH, Leibel SA, Sheline GE (eds): Radiation Injury to the Nervous System. New York: Raven Press, 1991, pp 91-112.    9. Fike JR, Gobbe GT. Central nervous system radiation injury in large animal models. In Gutin PH, Leibel SA, Sheline GE (eds): Radiation Injury to the Nervous System. New York: Raven Press, 1991, pp 113-136.   10. Ostertag CB. Experimental central nervous system injury from implanted isotopes. In Gutin PH, Leibel SA, Sheline GE (eds): Radiation Injury to the Nervous System. New York: Raven Press, 1991, pp 183-190.   11. Caveness WF. Experimental observations: delayed necrosis in normal monkey brain. In Gilbert HA, Kagen AR (eds): Radiation Damage to the Nervous System. New York: Raven Press, 1980, pp 1-38.   12. Price RA, Jamieson PA. The central nervous system in childhood leukemia. Subacute leukoencephalopathy. Cancer 1975;35 (Pt 2):306-318.   13. Fouladi M, Langston J, Mulhern R, et al. Silent lacunar ­lesions detected by magnetic resonance imaging of children with brain tumors: a late sequela of therapy. J Clin Oncol 2000;18:824-831.   14. Salazar OM, Rubin P, Feldstein ML, et al. High dose radiation therapy in the treatment of malignant gliomas: final report. Int J Radiat Oncol Biol Phys 1979;5:1733-1740.   15. Kramer S, Lee KF. Complications of radiation therapy: the central nervous system. Semin Roentgenol 1974;9:72-83.   16. Sheline GE, Wara WM, Smith V. Therapeutic irradiation and brain injury. Int J Radiat Oncol Biol Phys 1980;6:1215-1228.   17. Sheline GE. Irradiation injury of the human brain: a review of clinical experience. In Gilbert HA, Kagen AR (eds): Radiation Damage to the Nervous System. New York: Raven Press, 1980, pp 39-58.   18. Young DF, Posner JB, Chu F, et al. Rapid-course radiation therapy of cerebral metastases: results and complications. Cancer 1974;34:1069-1076.   19. Jones A. Transient radiation myelopathy (with reference to Lhermitte’s sign of electrical paresthesia). Br J Radiol 1964;37:  727-744.   20. Freeman JE, Johnston PGB, Voke JM. Somnolence after prophylactic cranial irradiation in children with acute lymphoblastic leukaemia. Br Med J 1973;4:523-525.   21. Boldrey E, Sheline G. Delayed transitory clinical manifestations after radiation treatment of intracranial tumors. Acta Radiol Ther Phys Biol 1966;5:5-10.   22. Steen RG, Koury M, Granja CI, et al. Effect of ionizing radiation on the human brain: white matter and gray matter T1 in pediatric brain tumor patients treated with conformal radiation therapy. Int J Radiat Oncol Biol Phys 2001;49:79-91.

861

  23. Hoffman WF, Levin VA, Wilson CB. Evaluation of malignant glioma patients during the postirradiation period. J Neurosurg 1979;30:624-628.   24. Longee DC, Fiedman HS, Djang NW, et al. Transient late magnetic resonance imaging changes suggesting progression of brain stem glioma: implications for entry criteria for phase II trials. Neurosurgery 1988;23:248-253.   25. Duffner PK, Cohen ME, Myers MH, et al. Survival of children with brain tumors: SEER program, 1973-1980. Neurology 1986;36:597-601.   26. Childhood Brain Tumor Consortium. A study of childhood brain tumors based on surgical biopsies from 10 North American ­institutions: sample description. J Neurooncol 1988;6:9-23.   27. Groothuis DR, Wright DC, Ostertag CB. The effect of 125I ­interstitial radiotherapy on blood-brain barrier function in ­normal canine brain. J Neurosurg 1987;67:895-902.   28. Schultheiss TE, Stephens LC. The pathogenesis of radiation myelopathy: widening the circle. Int J Radiat Oncol Biol Phys 1992;23:1089-1091.   29. Schultheiss TE, Stephens LC. Invited review: permanent radiation myelopathy. Br J Radiol 1992;65:737-753.   30. Marks JE, Baglan RJ, Prassad SC, et al. Cerebral radionecrosis: incidence and risk in relation to dose, time, fractionation and volume. Int J Radiat Oncol Biol Phys 1981;7:242-252.   31. Mikhael MA. Dosimetric considerations in the diagnosis of ­radiation necrosis of the brain. In Gilbert HA, Kagen AR (eds): Radiation Damage to the Nervous System. New York: Raven Press, 1980, pp 59-92.   32. Kramer S, Southard ME, Mansfield CM. Radiation effect and tolerance of the central nervous system. Front Radiat Ther Oncol 1972;6:332-345.   33. Aristizabal S, Caldwell WL, Avila J. The relationship of timedose fractionation factors to complications in the treatment of pituitary tumors by irradiation. Int J Radiat Oncol Biol Phys 1977;2:667-673.   34. Leibel SA, Sheline GE. Tolerance of the brain and spinal cord to conventional irradiation. In Gutin PH, Leibel SA, Sheline GE (eds): Radiation Injury to the Nervous System. New York: Raven Press, 1991, pp 239-256.   35. Clemente CD, Yamazaki JN, Bennett LR, et al. Brain radiation in newborn rats and differential effects of increased age. Neurology 1960;10:669-675.   36. Bloom HJG, Wallace ENK, Henk JM. The treatment and prognosis of medulloblastoma in children. AJR Am J Roentgenol 1969;105:43-62.   37. Boden G. Radiation myelitis of the brain stem. J Fac Radiol (Great Britain) 1950;2:79-94.   38. Pallis CA, Louis S, Morgan RL. Radiation myelopathy. Brain 1961;84:460-479.   39. Schultheiss TE, Higgins EM, el-Mahdi AM. The latent period in clinical radiation myelopathy. Int J Radiat Oncol Biol Phys 1984;10:1109-1115.   40. Myers R, Rogers MA, Hornsey S. A reappraisal of the role of glial and vascular elements in the development of white matter necrosis in irradiated rat spinal cord. Br J Cancer 1986;7:  221-223.   41. Phillips TL, Buschke F. Radiation tolerance of the thoracic spinal cord. Am J Roentgenol Rad Ther Nucl Med 1969;105:659-664.   42. Abbatucci JS, Delozier T, Quint R, et al. Radiation myelopathy of the cervical spinal cord: time, dose and volume factors. Int J Radiat Oncol Biol Phys 1978;4:239-248.   43. Hatievoll R, Host H, Kaalhus O. Myelopathy following radiotherapy of bronchial carcinoma with large single fractions: a retrospective study. Int J Radiat Oncol Biol Phys 1983;9:41-44.   44. Wara WM, Phillips TL, Sheline GE, et al. Radiation tolerance of the spinal cord. Cancer 1975;35:1558-1562.   45. Cohen L, Creditor M. An isoeffect table for radiation tolerance of the human spinal cord. Int J Radiat Oncol Biol Phys 1981;7:  961-966.   46. Van der Kogel AJ. Radiation tolerance of the rat spinal cord: time-dose relationships. Radiology 1977;122:505-509.

862

PART 10  Central Nervous System

  47. Ang KK, van der Kogel AJ, van der Schueren E. Lack of ­evidence for increased tolerance of rat spinal cord with decreasing ­fraction doses below 2 Gy. Int J Radiat Oncol Biol Phys 1985;11:105-110.   48. Dische S, Saunders MI. Continuous, hyperfractionated, accelerated radiotherapy (CHART): an interim report upon late morbidity. Radiother Oncol 1989;16:67-74.   49. Dische S. Accelerated treatment and radiation myelitis. Radiother Oncol 1991;20:1-2.   50. Norrell H, Wilson CB, Slagel DE, et al. Leukoencephalopathy following the administration of methotrexate into the cerebrospinal fluid in the treatment of primary brain tumors. Cancer 1974;33:923-932.   51. Lee JS, Umsawasdi T, Lee Y-Y, et al. Neurotoxicity in long-term survivors of small cell lung cancer. Int J Radiat Oncol Biol Phys 1986;12:313-321.   52. Bleyer WA. Neurologic sequelae of methotrexate and ionizing radiation: a new classification. Cancer Treatment Reports 1981;65:89-98.   53. Mulhern RK, Ochs J, Kun LE. Changes in intellect associated with cranial radiation therapy. In Gutin PH, Leibel SA, Sheline GE (eds): Radiation Injury to the Nervous System. New York: Raven Press, 1991, pp 325-340.   54. Mulhern RK, Kovnar E, Langston J, et al. Long-term survivors of leukemia treated in infancy: factors associated with neuropsychologic status. J Clin Oncol 1992;10:1095-1102.   55. Jannoun L, Bloom HJG. Long-term psychological effects in children treated for intracranial tumors. Int J Radiat Oncol Biol Phys 1990;18:747-753.   56. Hochberg FH, Slotnick B. Neuropsychologic impairment in ­astrocytoma survivors. Neurology 1980;30:172-177.   57. Maire J, Coudin B, Guerin J, et al. Neuropsychologic ­impairment in adults with brain tumors. Am J Clin Oncol 1987;10:  156-162.   58. Kun LE, Mulhern RK, Crisco JJ. Quality of life in children treated  for brain tumors: intellectual, emotional, and academic function. J Neurosurg 1983;58:1-6.   59. Radcliffe J, Packer RJ, Atkins TE, et al. Three- and four-year cognitive outcome in children with noncortical brain tumors treated with whole-brain radiotherapy. Ann Neurol 1992;32:551-554.   60. Mulhern RK, Kepner JL, Thomas PRM, et al. Neuropsychological functioning of survivors of childhood medulloblastoma randomized to receive conventional (3,600 cGy/20) or reduced (2,340/13) dose craniospinal irradiation: a Pediatric Oncology Group study. J Clin Oncol 1998;16:1723-1728.   61. Mulhern RK, Reddick WE, Palmer SL, et al. Neurocognitive deficits in medulloblastoma survivors and white matter loss. Ann Neurol 1999;46:834-841.   62. Craig JB, Jackson DV, Moody D, et al. Prospective evaluation of changes in computed cranial tomography in patients with small cell lung carcinoma treated with chemotherapy and prophylactic cranial irradiation. J Clin Oncol 1984;2:1151-1156.   63. Urie MM, Fullerton B, Tatsuzaki H, et al. A dose response analysis of injury to cranial nerves and/or nuclei following proton beam radiation therapy. Int J Radiat Oncol Biol Phys 1992;23:27-29.   64. Parsons JT, Bova FJ, Fitzgerald CR, et al. Radiation optic neuropathy after megavoltage external-beam irradiation: analysis of time-dose factors. Int J Radiat Oncol Biol Phys 1994;30:  755-763.   65. Harris JR, Levene MB. Visual complications following irradiation for pituitary adenomas and craniopharyngiomas. Radiology 1976;120:167-171.   66. Svensson H, Westling P, Larsson LG. Radiation-induced lesions of the bronchial plexus correlated to the dose-time-fraction schedule. Acta Radiol 1975;14:228-238.   67. Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2007. CA Cancer J Clin 2007;57:43-66.   68. Linabery AM, Ross JA. Trends in cancer incidence among children in the U.S. (1992-2004). Cancer 2008;112:416-432.   69. Kleihues P, Cavenee WK (eds). Tumours of the Nervous System. Lyon, France: IARC Press, 2000.   70. Rorke L. The cerebellar medulloblastoma and its relationship to primitive neuroectodermal tumors. J Neuropathol Exp Neurol 1983;42:1-15.

  71. Hart MN, Earle KM. Primitive neuroectodermal tumors of the brain in children. Cancer 1973;32:890-898.   72. Grotzer MA, Janss AJ, Fung JA, et al. TrkC expression ­predicts good clinical outcome in primitive neuroectodermal brain ­tumors. J Clin Oncol 2000;18:1027-1035.   73. McLendon RE, Friedman HS, Fuchs HE, et al. Diagnostic markers in paediatric medulloblastoma: a Pediatric Oncology Group study. Histopathology 1999;34:154-162.   74. Mendrzyk F, Radlwimmer B, Joos S, et al. Genomic and ­protein expression profiling identifies CDK6 as novel independent prognostic marker in medulloblastoma. J Clin Oncol 2005;34:  8853-8862.   75. Aldosari N, Bigner SH, Burger PC, et al. MYCC and MYCN oncogene amplification in medulloblastoma. A fluorescence in situ hybridization study on paraffin sections from the Children’s Oncology Group. Arch Pathol Lab Med 2002;126:540-544.   76. Bobola MS, Finn LS, Ellenbogen RG, et al. Apurinic/apyrimidinic  endonuclease activity is associated with response to radiation and chemotherapy in medulloblastoma and primitive neuroectodermal tumors. Clin Cancer Res 2005;11:7405-7414.   77. Clifford SC, Lusher ME, Lindsey JC, et al. Wnt/Wingless pathway activation and chromosome 6 loss characterize a distinct molecular sub-group of medulloblastomas associated with a ­favorable prognosis. Cell Cycle 2006;5:2666-2670.   78. Lamont JM, McManamy CS, Pearson AD, et al. Combined histopathological and molecular cytogenetic stratification of medulloblastoma patients. Clin Cancer Res 2004;10:5482-5493.   79. Gajjar A, Hernan R, Kocak M, et al. Clinical, histopathologic and molecular markers of prognosis: toward a new disease risk stratification system for medulloblastoma. J Clin Oncol 2004;22:  971-974.   80. Gajjar A, Fouladi M, Walter AW, et al. Comparison of lumbar and shunt cerebrospinal fluid specimens for cytologic ­detection of leptomeningeal disease in pediatric patients with brain ­tumors. J Clin Oncol 1999;17:1825-1828.   81. Kleinman GM, Hockberg FH, Richardson EP Jr. Systemic ­metastases from medulloblastoma: report of two cases and ­review of the literature. Cancer 1981;48:2296-2309.   82. Albright AL, Wisoff JH, Zeltzer PM, et al. Effects of medulloblastoma resections on outcome in children: a report from the Children’s Cancer Group. Neurosurgery 1996;38:265-271.   83. Cochrane DD, Gustavsson B, Poskitt KP, et al. The surgical and natural morbidity of aggressive resection for posterior fossa ­tumors in childhood. Pediatr Neurosurg 1994;20:19-29.   84. Aguiar PH, Plese JP, Ciquini O, et al. Transient mutism following a posterior fossa approach to cerebellar tumors in children: a critical review of the literature. Childs Nerv Syst 1995;11:  306-310.   85. Chang CH, Housepian EM, Herbert C Jr. An operative staging system and a megavoltage radiotherapeutic technic for cerebellar medulloblastoma. Radiology 1969;93:1351-1359.   86. Kun LE, Constine LS. Medulloblastoma: caution regarding new treatment approaches. Int J Radiat Oncol Biol Phys 1991;20:  897-899.   87. Packer RJ, Sutton LN, Elterman R, et al. Outcome for children with medulloblastoma treated with radiation and cisplatin, CCNU, and vincristine chemotherapy. J Neurosurg 1994;81:  690-698.   88. Heideman RL, Kovnar EH, Kellie SJ, et al. Preirradiation chemotherapy with carboplatin and etoposide in newly ­ diagnosed embryonal pediatric CNS tumors. J Clin Oncol 1995;13:  2247-2254.   89. Kovnar EH, Kellie SJ, Horowitz ME, et al. Pre-irradiation cisplatin and etoposide in the treatment of high-risk medulloblastoma and other malignant embryonal tumors of the central nervous system: a phase II study. J Clin Oncol 1990;8:330-336.   90. Bailey CC, Gnekow A, Wellek S, 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;25:166-178.

33  The Brain and Spinal Cord   91. Hartsell WF, Gajjar A, Heideman RL, et al. Patterns of failure in children with medulloblastoma: effects of pre-irradiation chemotherapy. Int J Radiat Oncol Biol Phys 1997;39:15-24.   92. Thomas PR, Deutsch M, Kepner JL, et al. Low-stage medulloblastoma: final analysis of trial comparing standard-dose with reduced-dose neuraxis irradiation. J Clin Oncol 2000;18:  3004-3011.   93. Berry MP, Jenkin RDT, Keen CW, et al. Radiation treatment for medulloblastoma: a 21-year review. J Neurosurg 1981;55:  43-51.   94. Hughes EN, Winston K, Cassady JR, et al. Medulloblastoma: ­results of surgery and radical radiation therapy at the JCRT 1968-1984. Int J Radiat Oncol Biol Phys 1986;12:179.   95. Bloom HJG, Glees J, Bell J. The treatment and long-term prognosis of children with intracranial tumors: a study of 610 cases, 1950-1981. Int J Radiat Oncol Biol Phys 1990;18:723-745.   96. Jenkin D, Goddard K, Armstrong D, et al. Posterior fossa medulloblastoma in childhood: treatment results and a proposal for a new staging system. Int J Radiat Oncol Biol Phys 1990;19:  265-274.   97. Packer RJ, Goldwein J, Nicholson HS, et al. Treatment of children with medulloblastomas with reduced-dose craniospinal radiation therapy and adjuvant chemotherapy: a Children’s Cancer Group study. J Clin Oncol 1999;17:2127-2136.   98. Miralbell R, Lomax A, Cella L, et al. Potential reduction of the incidence of radiation-induced second cancers by using proton beams in the treatment of pediatric tumors. Int J Radiat Oncol Biol Phys 2002;54:824-829.   99. Fukunaga-Johnson N, Sandler HM, Marsh R, et al. The use of 3D conformal radiotherapy (3D CRT) to spare the cochlea in patients with medulloblastoma. Int J Radiat Oncol Biol Phys 1998;41:77-82. 100. Merchant TE, Happersett L, Finlay J, et al. Preliminary results of conformal radiation therapy for medulloblastoma. Neurooncology 1999;1:177-187. 101. Fukunaga-Johnson N, Lee JH, Sandler HM, et al. Patterns of failure following treatment for medulloblastoma: is it necessary to treat the entire posterior fossa? Int J Radiat Oncol Biol Phys 1998;42:143-146. 102. Strother D, Ashley D, Kellie SJ, et al. Feasibility of four consecutive high-dose chemotherapy cycles with stem-cell ­ rescue for ­patients with newly diagnosed medulloblastoma or ­supratentorial primitive neuroectodermal tumor after ­ craniospinal radiotherapy: results of a collaborative study. J Clin Oncol 2001;19:  2696-2704. 103. Gajjar A, Chintagumpala M, Ashley D, et al. Risk-adapted craniospinal radiotherapy followed by high-dose chemotherapy and stem-cell rescue in children with newly diagnosed medulloblastoma (St Jude Medulloblastoma-96): long-term results from a prospective, multicentre trial. Lancet Oncol 2006;7:  813-820. 104. Gajjar A, Sanford RA, Heideman R, et al. Low-grade astrocytoma: a decade of experience at St. Jude Children’s Research Hospital. J Clin Oncol 1997;15:2792-2799. 105. Leibel SA, Sheline GE, Wara WM, et al. The role of radiation therapy in the treatment of astrocytomas. Cancer 1975;35:  1551-1557. 106. Morantz RA. Radiation therapy in the treatment of cerebral ­astrocytoma. Neurosurgery 1987;20:975-982. 107. Wallner KE, Gonzales MF, Edwards MSB, et al. Treatment ­results of juvenile pilocytic astrocytoma. J Neurosurg 1988;69:  171-176. 108. Laws ER, Taylor WF, Clifton MB, et al. Neurosurgical management of low-grade astrocytoma of the cerebral hemispheres.  J Neurosurg 1984;61:665-673. 109. Krieger MD, Gonzalez-Gomez I, Levy ML, et al. Recurrence patterns and anaplastic change in a long-term study of pilocytic astrocytomas. Pediatr Neurosurg 1997;27:1-11. 110. Roessler K, Bertalanffy A, Jezan H, et al. Proliferative activity as measured by MIB-1 labeling index and long-term outcome of cerebellar juvenile pilocytic astrocytomas. J Neurooncol 2002;58:141-146.

863

111. Ilgren EB, Stiller CA. Cerebellar astrocytomas. Clinical characteristics and prognostic indices. J Neurooncol 1987;4:  293-308. 112. Smoots DW, Geyer JR, Lieberman DM, et al. Predicting disease progression in childhood cerebellar astrocytoma. Childs Nerv Syst 1998;14:636-648. 113. Tomita T, Chou P, Reyes-Mugica M. IV ventricle astrocytomas in childhood: clinicopathological features in 21 cases. Childs Nerv Syst 1998;14:537-546. 114. Pencalet P, Maixner W, Sainte-Rose C, et al. Benign cerebellar astrocytomas in children. J Neurosurg 1999;90:265-273. 115. Sgouros S, Fineron PW, Hockley AD. Cerebellar astrocytoma of childhood: long-term follow-up. Childs Nerv Syst 1995;11:  89-96. 116. Burger PC, Scheithauer BW, Vogel FS (eds). Surgical Pathology of the Nervous System and its Coverings, 3rd ed. New York: Churchill Livingstone, 1991. 117. Garcia DM, Marks JE, Latifi HR, et al. Childhood cerebellar ­astrocytomas: is there a role for postoperative irradiation? Int J Radiat Oncol Biol Phys 1990;18:815-818. 118. Schneider JH, Raffel C, McComb JG. Benign cerebellar astrocytomas of childhood. Neurosurgery 1992;30:58-63. 119. Ilgren EB, Stiller CA. Cerebellar astrocytomas: therapeutic management. Acta Neurochir 1986;81:11-26. 120. Larson DA, Wara WM, Edwards MSB. Management of childhood cerebellar astrocytoma. Int J Radiat Oncol Biol Phys 1980;18:971-973. 121. Hirsch J-F, Rose CS, Pierre-Kahn A, et al. Benign astrocytic and oligodendrocytic tumors of the cerebral hemispheres in children. J Neurosurg 1989;70:568-572. 122. Hoffman HJ, Soloniuk DS, Humphreys RP, et al. Management and outcome of low-grade astrocytomas of the midline in children: a retrospective review. Neurosurgery 1993;33:964-971. 123. Reardon D, Gajjar A, Walter AW, et al. Bithalamic ­involvement predicts poor outcome among children with thalamic glial ­tumors. Pediatr Neurosurg 1998;29:29-35. 124. Krouwer HGJ, Prados MD. Infiltrative astrocytomas of the thalamus. J Neurosurg 1995;82:548-557. 125. Albright AL, Price RA, Guthkelch AN. Diencephalic gliomas of children: a clinicopathologic study. Cancer 1985;55:2789-2793. 126. Cairncross JG, Laperriere NJ. Low-grade glioma: to treat or not to treat? Arch Neurol 1989;46:1238-1239. 127. Bernstein M, Hoffman HJ, Halliday WC, et al. Thalamic ­tumors in children: long-term follow-up and treatment guidelines.  J Neurosurg 1984;61:649-656. 128. Kelly PJ. Stereotactic biopsy and resection of thalamic astrocytomas. Neurosurgery 1989;25:185-194. 129. Leibel SA, Sheline GE. Review article: radiation therapy for neoplasms of the brain. J Neurosurg 1987;66:1-22. 130. Boydston WR, Sanford RA, Muhlbauer MS, et al. Gliomas of the tectum and periaqueductal region of the mesencephalon. Pediatr Neurosurg 1991;17:234-238. 131. Medbery CA, Straus KL, Steinberg SM, et al. Low-grade astrocytomas: treatment results and prognostic variables. Int J Radiat Oncol Biol Phys 1988;15:837-841. 132. Whitton AC, Bloom HJG. Low-grade glioma of the cerebral hemispheres in adults: a retrospective analysis of 88 cases. Int J Radiat Oncol Biol Phys 1990;18:783-786. 133. Leighton C, Fisher B, Bauman G, et al. Supratentorial low-grade glioma in adults: an analysis of prognostic factors and timing of radiation. J Clin Oncol 1997;15:1294-1301. 134. Shaw EG, Daumas-Duport C, Scheithauer BW, et al. Radiation therapy in the management of low-grade supratentorial astrocytomas. J Neurosurg 1989;70:853-861. 135. Karim AB, Maat B, Hatlevoll R, et al. A randomized trial of dose-response in radiation therapy of low-grade cerebral glioma: EORTC study 22844. Int J Radiat Oncol Biol Phys 1996;36:  549-556. 136. Dunbar SF, Tarbell NJ, Kooy HM, et al. Stereotactic radiotherapy  for pediatric and adult brain tumors: preliminary report. Int J Radiat Oncol Biol Phys 1994;30:531-539.

864

PART 10  Central Nervous System

137. Listernick R, Louis DN, Packer RJ, et al. Optic pathway gliomas in children with neurofibromatosis 1: consensus statement from the NF1 Optic Pathway Gliomas Task Force. Ann Neurol 1997;41:143-149. 138. Horwich A, Bloom HJG. Optic gliomas: radiation therapy and prognosis. Int J Radiat Oncol Biol Phys 1985;11:1067-1079. 139. Alvord EC, Lofton S. Gliomas of the optic nerve or chiasm: outcome by patient’s age, tumor site, and treatment. J Neurosurg 1988;68:85-98. 140. Hoyt WF, Baghdassarian SA. Optic glioma of childhood: natural history and rationale for conservative management. Br J Ophthalmol 1969;53:793-798. 141. Imes RK, Hoyt WF. Childhood chiasmal gliomas: update on the fate of patients in the 1969 San Francisco study. Br J Ophthalmol 1986;70:179-182. 142. Wong JYC, Wara W, Sheline GE. Optic gliomas: a reanalysis of the University of California, San Francisco, experience. Cancer 1987;60:1847-1855. 143. Jenkin D, Angyalfi S, Becker L, et al. Optic glioma in children: surveillance, resection, or irradiation? Int Radiat Oncol Biol Phys 1993;25:215-225. 144. Bataini JP, Delanian S, Ponvert D. Chiasmal gliomas: results of irradiation management in 57 patients and review of literature. Int J Radiat Oncol Biol Phys 1991;21:615-623. 145. Tao ML, Barnes PD, Billett AL, et al. Childhood optic chiasm gliomas: radiographic response following radiotherapy and longterm clinical outcome. Int J Radiat Oncol Biol Phys 1997;39:  579-587. 146. Fletcher WA, Imes RK, Hoyt WF. Chiasmal gliomas: appearance and long-term changes demonstrated by computerized tomography. J Neurosurg 1986;65:154-159. 147. Wisoff JH, Abbott R, Epstein F. Surgical management of exophytic chiasmatic-hypothalamic tumors of childhood. J Neurosurg 1990;73:661-667. 148. Packer RJ, Lange B, Ater J, et al. Carboplatin and vincristine for recurrent and newly diagnosed low-grade gliomas of childhood. J Clin Oncol 1993;11:850-856. 149. Packer RJ, Ater J, Allen J, et al. Carboplatin and vincristine chemotherapy for children with newly diagnosed progressive low-grade gliomas. J Neurosurg 1997;86:747-754. 150. Mahoney DH, Cohen ME, Friedman HS, et al. Carboplatin is effective therapy for young children with progressive optic pathway tumors: a Pediatric Oncology Group phase II study. Neurooncology 2000;2:213-220. 151. Somaza SC, Kondziolka D, Lunsford LD, et al. Early outcomes after stereotactic radiosurgery for growing pilocytic astrocytomas in children. Pediatr Neurosurg 1996;25:109-115. 152. Grabb PA, Lunsford LD, Albright AL, et al. Stereotactic ­radiosurgery for glial neoplasms of childhood. Neurosurgery 1996;38:696-702. 153. Pollack IF, Hurtt M, Pang D, et al. Dissemination of low grade astrocytomas in children. Cancer 1994;73:2869-2878. 154. Prados M, Mamelak AN. Metastasizing low grade gliomas in children. Redefining an old disease. Cancer 1994;73:2671-2673. 155. Gajjar A, Bhargava R, Jenkins JJ, et al. Low-grade astrocytoma with neuraxis dissemination at diagnosis. J Neurosurg 1995;83:67-71. 156. 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;44:1839-1852. 157. Taratuto AL, Monges J, Lylyk P, et al. Superficial cerebral astrocytoma attached to dura. Cancer 1984;54:2505-2512. 158. Fouladi M, Jenkins J, Burger P, et al. Pleomorphic xanthoastrocytoma (PXA): favorable outcome following complete surgical resection. Neuro Oncol 2001;3:184-192. 159. Merchant TE, Zhu Y, Thompson SJ, et al. Preliminary ­results from a phase II trial of conformal radiation therapy for pediatric ­ patients with localized low-grade astrocytomas and ­ ependymoma. Int J Radiat Oncol Biol Phys 2002;52:  325-332.

160. Shaw E, Arusell R, Scheithauer B, et al. Prospective randomized trial of low- versus high-dose radiation therapy in adults with supratentorial low-grade glioma: initial report of a North Central Cancer Treatment Group/Radiation Therapy Oncology Group/Eastern Cooperative Oncology Group Study. J Clin Oncol 2002;20:2367-2376. 161. Pollack IF, Claassen D, Al-Shboul Q, et al. Low grade gliomas of the cerebral hemispheres in children: an analysis of 71 cases.  J Neurosurg 1995;82:536-547. 162. McLaurin RL, Breneman J, Aron B. Hypothalamic gliomas: review of 18 cases. Concepts Pediatr Neurosurg 1987;7:19. 163. Albright AL, Price RA, Guthkelch AN. Brainstem gliomas of children: a clinicopathologic study. Cancer 1983;52:2313-2319. 164. Albright AL, Guthkelch AN, Packer RJ, et al. Prognostic factors in pediatric brainstem gliomas. J Neurosurg 1986;65:751-755. 165. Barkovich AJ, Krischer J, Kun LE, et al. Brain stem gliomas: a classification system based on magnetic resonance imaging. Pediatr Neurosurg 1990;16:73-83. 166. Hoffman HJ, Becker I, Craven MA. A clinically and pathologically distinct group of benign brainstem gliomas. Neurosurgery 1980;7:243-248. 167. Stroink AR, Hoffman HJ, Hendrick EB, et al. Diagnosis and management of pediatric brainstem gliomas. J Neurosurg 1986;65:745-750. 168. Khatib ZA, Heideman RL, Kovnar EH, et al. Predominance of pilocytic histology in dorsally exophytic brain stem tumors. Pediatr Neurosurg 1994;20:2-10. 169. Fisher PG, Breiter SN, Carson BS, et al. A clinicopathologic reappraisal of brain stem tumor classification. Identification of pilocytic astrocytoma and fibrillary astrocytoma as distinct entities. Cancer 2000;89:1569-1576. 170. Mantravadi RVP, Phatak R, Bellur S, et al. Brainstem gliomas: an autopsy study of 25 cases. Cancer 1982;49:1294-1296. 171. Schulz-Ertner D, Debus J, Lohr F, et al. Fractionated stereotactic conformal radiation therapy of brain stem gliomas: outcome and prognostic factors. Radiother Oncol 2000;57:215-223. 172. Shrieve DC, Wara WM, Edwards MSB, et al. Hyperfractionated radiation therapy for gliomas of the brainstem in children and in adults. Int J Radiat Oncol Biol Phys 1992;24:599-610. 173. Freeman CR, Krischer JP, Sanford RA, et al. Final results of a study of escalating doses of hyperfractionated radiotherapy in brain stem tumors in children: a Pediatric Oncology Group study. Int J Radiat Oncol Biol Phys 1993;27:197-206. 174. Freeman CR, Bourgouin PM, Sanford RA, et al. Long term survivors of childhood brain stem gliomas treated with hyperfractionated radiotherapy, clinical characteristics and treatment related toxicities. Cancer 1996;77:555-562. 175. Mandell LR, Kadota R, Freeman C, et al. There is no role for hyperfractionated radiotherapy in the management of children with newly diagnosed diffuse intrinsic brainstem tumors: results of a Pediatric Oncology Group phase III trial comparing conventional vs. hyperfractionated radiotherapy. Int J Radiat Oncol Biol Phys 1999;43:959-964. 176. Freeman CR, Kepner J, Kun LE, et al. A detrimental effect of a combined chemotherapy-radiotherapy approach in children with diffuse intrinsic brain stem gliomas? Int J Radiat Oncol Biol Phys 2000;47:561-564. 177. Tait DM, Thornton-Jones H, Bloom HJG, et al. Adjuvant chemotherapy for medulloblastoma: the first multicenter control trial of the International Society of Paediatric Oncology (SIOP I). Eur J Cancer Clin Oncol 1990;26:464-469. 178. Duffner PK, Horowitz ME, Krischer JP, et al. Postoperative chemotherapy and delayed radiation in children less than three years of age with malignant brain tumors. N Engl J Med 1993;328:1725-1731. 179. Ilgren EB, Stiller CA, Hughes JT, et al. Ependymomas: a clinical and pathologic study, part 1: biologic features. Clin Neuropathol 1984;3:113-121. 180. Sanford RA, Kun LE, Heideman RL, et al. Cerebellar pontine angle ependymoma in infants. Pediatr Neurosurg 1997;27:  84-91.

33  The Brain and Spinal Cord 181. Bloom HJG. Intracranial tumors: response and resistance to therapeutic endeavors. 1970-1980. Int J Radiat Oncol Biol Phys 1982;8:1083-1113. 182. Vanuystel LJ, Bessell EM, Ashley SE, et al. Intracranial ependymoma: long-term results of a policy of surgery and radiotherapy. Int J Radiat Oncol Biol Phys 1982;23:313-319. 183. Goldwein JW, Glauser TA, Packet RJ, et al. Recurrent intracranial ependymomas in children: survival, patterns of failure, and prognostic factors. Cancer 1990;66:557-563. 184. Rezai AR, Woo HH, Lee M, et al. Disseminated ependymomas of the central nervous system. J Neurosurg 1996;85:  618-624. 185. Centeno RS, Lee AA, Winter J, et al. Supratentorial ependymomas. Neuroimaging and clinicopathological correlation. J Neurosurg 1986;64:209-215. 186. Rawlings CE III, Giangaspero F, Burger PC, et al. Ependymomas: a clinicopathologic study. Surg Neurol 1988;29:271-281. 187. Rorke LB. Relationship of morphology of ependymoma in children to prognosis. Prog Exp Tumor Res 1987;30:170-174. 188. Ross GW, Rubinstein LJ. Lack of histopathological correlation of malignant ependymomas with postoperative survival. J Neurosurg 1989;70:31-36. 189. Rousseau P, Habrand JL, Sarrazin D, et al. Treatment of intracranial ependymomas of children: review of a 15-year experience. Int J Radiat Oncol Biol Phys 1994;28:381-386. 190. Pollack IF, Gerszten PC, Martinez AJ, et al. Intracranial ependymomas of childhood: long-term outcome and prognostic factors. Neurosurgery 1995;37:655-667. 191. Healey EA, Braners PD, Kupsky WJ, et al. The prognostic significance of postoperative residual tumor in ependymoma. Neurosurgery 1991;28:666-672. 192. Nazar GB, Hoffman HJ, Becker LE, et al. Infratentorial ependymomas in childhood: prognostic factors and treatment. J Neurosurg 1990;72:408-417. 193. Hukin J, Epstein F, Lefton D, et al. Treatment of intracranial ependymoma by surgery alone. Pediatr Neurosurg 1998;29:  40-45. 194. Duffner PK, Krischer JP, Sanford RA, et al. Prognostic factors in infants and very young children with intracranial ependymomas. Pediatr Neurosurg 1998;28:215-222. 195. Lefkowitz I, Evans A, Sposto R, et al. Adjuvant chemotherapy of childhood posterior fossa (PF) ependymoma: craniospinal irradiation with or without CCNU, vincristine (VCR) and prednisone (P) [abstract]. Proc Am Soc Clin Oncol 1989;8:87a. 196. Strother D, Kepner J, Aronin P, et al. Dose-intensive (DI) chemotherapy (CT) prolongs event-free survival (EFS) for very young children with ependymoma (EP). Results of Pediatric Oncology Group (POG) Study 9233 [abstract]. Proc Am Soc Clin Oncol 2000;19. 585a, 2302. 197. Goldwein JW, Corn BW, Finlay JL, et al. Is craniospinal irradiation required to cure children with malignant (anaplastic) intracranial ependymomas? Cancer 1991;67:2766-2771. 198. Merchant TE, Haida T, Wang M-H., et al. Anaplastic ependymoma: treatment of pediatric patients with or without craniospinal radiation therapy. J Neurosurg 1997;86:943-949. 199. Kovnar EH, Curran W, Tomita T, et al. Hyperfractionated ­irradiation for childhood ependymoma: improved local ­control in subtotally resected tumors [abstract]. Paper presented at the Eighth International Symposium on Pediatric Neuro-Oncology, May 6-9, 1998, Rome, Italy. 200. Aggarwal R, Yeung D, Muhlbauer M, et al. Efficacy and feasibility of stereotactic radiosurgery in the primary management of unfavorable pediatric ependymoma. Radiother Oncol 1997;43:  269-273. 201. Herrick MK. Pathology of pineal tumors. In Neuwelt EA (ed): Diagnosis and Treatment of Pineal Region Tumors. Baltimore: Williams & Wilkins, 1984, pp 31-60. 202. Glenn OA, Barkovich AJ. Intracranial germ cell tumors: a comprehensive review of proposed embryonic deprivation. Pediatr Neurosurg 1996;24:242-251. 203. Rich TA, Cassady JR, Strand RD, et al. Radiation therapy for pineal and suprasellar germ cell tumors. Cancer 1985;55:  932-940.

865

204. Ho DM, Lieu H-C. Primary intracranial germ cell tumor. Pathologic study of 51 patients. Cancer 1992;70:1577-1584. 205. Dernaley DP, A’Hern RP, Whittaker S, et al. Pineal and CNS germ cell tumors: Royal Marsden Hospital experience 19621987. Int J Radiat Oncol Biol Phys 1990;18:773-781. 206. Wolden SL, Wara WM, Larson DA, et al. Radiation therapy for primary intracranial germ-cell tumors. Int J Radiat Oncol Biol Phys 1995;32:943-949. 207. Borit A, Blackwood W, Mair WGP. The separation of pineocytoma from pineoblastoma. Cancer 1980;45:1408-1418. 208. DiSclafani A, Hudgins RJ, Edwards MSB, et al. Pineocytomas. Cancer 1989;63:302-304. 209. Schild SE, Scheithauer BW, Schomberg PJ, et al. Pineal parenchymal tumors. Clinical, pathologic, and therapeutic aspects. Cancer 1993;72:870-880. 210. Schild SE, Haddock MG, Scheithauer BW, et al. Nongerminomatous germ cell tumors of the brain. Int J Radiat Oncol Biol Phys 1996;36:557-563. 211. Shibamoto Y, Takahashi M, Sasai K. Prognosis of intracranial germinoma with syncytiotrophoblastic giant cells treated by radiation therapy. Int J Radiat Oncol Biol Phys 1997;37:505-510. 212. Jennings MT, Gelman R, Hochberg F. Intracranial ­ germ-cell ­tumors: natural history and pathogenesis. J Neurosurg 1985;63:  155-167. 213. Jenkin D, Berry M, Chan H, et al. Pineal region germinomas in childhood: treatment considerations. Int J Radiat Oncol Biol Phys 1990;18:541-545. 214. Sung D, Harisiadis L, Chang CG. Midline pineal tumors and suprasellar germinomas: highly curable by irradiation. Radiology 1978;128:745-751. 215. Haddock MG, Schild SE, Scheithauer BE, et al. Radiation therapy for histologically confirmed primary central nervous system germinoma. Int J Radiat Oncol Biol Phys 1997;38:915-923. 216. Shibamoto Y, Takahashi M, Abe M. Reduction of the radiation dose for intracranial germinoma: a prospective study. Br J Cancer 1994;70:984-989. 217. Shirato H, Nishio M, Sawamura Y, et al. Analysis of long-term treatment of intracranial germinoma. Int J Radiat Oncol Biol Phys 1997;37:511-515. 218. Merchant TE, Sherwood SH, Mulhern RK, et al. CNS germinoma: disease control and long-term functional outcome for  12 children treated with craniospinal irradiation. Int J Radiat Oncol Biol Phys 2000;46:1171-1176. 219. Hardenbergh PH, Golden J, Billet A, et al. Intracranial germinoma: the case for lower dose radiation therapy. Int J Radiat Oncol Biol Phys 1997;39:419-426. 220. Calaminus G, Bamberg M, Baranzelli MC, et al. Intracranial germ cell tumors: a comprehensive update of the European data. Neuropediatrics 1994;25:26-32. 221. Shibamoto Y, Abe M, Yamashita J, et al. Treatment results of intracranial germinoma as a function of the irradiation volume. Int J Radiat Oncol Biol Phys 1988;15:285-290. 222. Bamberg M, Kortmann R-D, Calaminus G, et al. Radiation therapy for intracranial germinoma: results of the German Cooperative Prospective Trials MAKEI 83/86/89. J Clin Oncol 1999;17:2585-2592. 223. Allen JC, Kim JH, Packer RJ. Neoadjuvant chemotherapy for newly diagnosed germ cell tumors of the central nervous system. J Neurosurg 1987;67:65-70. 224. Bouffet E, Baranzelli MC, Patte C, et al. Combined treatment modality for intracranial germinomas: results of a multicentre SFOP experience. Br J Cancer 1999;79:1199-1204. 225. Balmaceda C, Heller G, Rosenblum M, et al. Chemotherapy without irradiation. A novel approach for newly diagnosed CNS germ cell tumors: results of an international cooperative trial.  J Clin Oncol 1996;14:2908-2915. 226. Linstadt D, Wara WM, Edwards MSB, et al. Radiotherapy of primary intracranial germinomas: the case against routine craniospinal irradiation. Int J Radiat Oncol Biol Phys 1988;15:291-297. 227. Maity A, Shu HK, Janss A, et al. Craniospinal radiation in the treatment of biopsy-proven intracranial germinomas: twentyfive years’ experience in a single center. Int J Radiat Oncol Biol Phys 2004;58:1165-1170.

866

PART 10  Central Nervous System

228. Buckner JC, Peethambaram PP, Smithson WA, et al. Phase II trial of primary chemotherapy followed by reduced-dose radiation for CNS germ cell tumors. J Clin Oncol 1999;17:933-940. 229. Allen JC, DaRosso RC, Donahue B, et al. A phase II trial of preirradiation carboplatin in newly diagnosed germinoma of the central nervous system. Cancer 1994;74:940-944. 230. Robertson PL, DaRosso RC, Allen JC. Improved prognosis of intracranial non-germinoma germ cell tumors with multimodality therapy. J Neurooncol 1997;32:71-80. 231. Aoyama H, Shirato H, Yoshida H, et al. Retrospective multi­institutional study of radiotherapy for intracranial non-germinomatous germ cell tumors. Radiother Oncol 1998;49:55-59. 232. Jakacki RI, Zeltzer PM, Boyett JM, 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;13:1377-1383. 233. Adamson TE, Wiestler OD, Kleihues P, et al. Correlation of clinical and pathological features in surgically treated craniopharyngiomas. J Neurosurg 1990;73:12-17. 234. Hoffman HJ, De Silva M, Humphreys RP, et al. Aggressive surgical management of craniopharyngiomas in children. J Neurosurg 1992;76:47-52. 235. Hoffman HJ. Surgical management of craniopharyngioma. Pediatr Neurosurg 1994;21:44-49. 236. Yarsargil MG, Curcic M, Kis S, et al. Total removal of craniopharyngiomas: approaches and long-term results in 144 patients.  J Neurosurg 1990;73:3-11. 237. Sanford RA. Craniopharyngioma: results of survey of the American Society of Pediatric Neurosurgery. Pediatr Neurosurg 1994;21:39-43. 238. Scott RM, Hetelekidis S, Barnes PD, et al. Surgery, radiation, and combination therapy in the treatment of childhood craniopharyngioma—a 20-year experience. Pediatr Neurosurg 1994;21:  75-81. 239. Wen B-C., Hussey DH, Staples J, et al. A comparison of the roles of surgery and radiation therapy in the management of craniopharyngiomas. Int J Radiat Oncol Biol Phys 1989;16:17-24. 240. Sweet WH. History of surgery for craniopharyngiomas. Pediatr Neurosurg 1994;21:28-38. 241. Tomita T, McLone DG. Radical resections of childhood craniopharyngiomas. Pediatr Neurosurg 1993;19:6-14. 242. Yasargil MC, Curcic M, Kis S, et al. Total removal of craniopharyngiomas: approaches and long-term results in 144 patients.  J Neurosurg 1990;73:3-11. 243. Tomita T, McLone DG. Radical resections of childhood craniopharyngiomas. Pediatr Neurosurg 1993;19:6-14. 244. Fischer EG, Welch K, Shillito J Jr, et al. Craniopharyngiomas in children: long-term effects of conservative surgical ­ procedures combined with radiation therapy. J Neurosurg 1990;73:  534-540. 245. Fischer EG, Welch K, Shilito J Jr, et al. Craniopharyngiomas in children. Long-term effects of conservative surgical procedures combined with radiation therapy. J Neurosurg 1990;73:  534-540. 246. Sklar CA. Craniopharyngioma: endocrine sequelae of treatment. Pediatr Neurosurg 1994;21:120-123. 247. Curtis J, Daneman D, Hoffman HJ, et al. The endocrine outcome after surgical removal of craniopharyngiomas. Pediatr Neurosurg 1994;21:24-27. 248. Kramer S, Southard M, Mansfield CM. Radiotherapy in the management of craniopharyngiomas. AJR Am J Roentgenol 1968;103:44-52. 249. Kramer S, McKissock W, Concannon JP. Craniopharyngiomas: treatment by combined surgery and radiation therapy. J Neurosurg 1961;18:217-226. 250. Kramer S, Southard M, Mansfield CM. Radiotherapy in the management of craniopharyngiomas: further experience and late results. Am J Roentgenol Radium Ther Nucl Med 1968;103:  44-52. 251. Regine WF, Kramer S. Pediatric craniopharyngiomas: long-term results of combined treatment with surgery and radiation. Int J Radiat Oncol Biol Phys 1992;24:611-617.

252. Brada M, Thomas DGT. Craniopharyngioma revisited. Int J ­Radiat Oncol Biol Phys 1993;27:471-475. 253. Hetelididis S, Barnes PD, Tao ML, et al. Twenty year experience in childhood craniopharyngioma. Int J Radiat Oncol Biol Phys 1993;27:189-195. 254. Rajan B, Ashley S, Gorman C, et al. Craniopharyngioma—long term results following limited surgery and radiotherapy. Radiother Oncol 1993;26:1-10. 255. Flickinger JC, Lunsford LD, Singer J, et al. Megavoltage ­external beam irradiation of craniopharyngiomas: analysis of tumor control and morbidity. Int J Radiat Oncol Biol Phys 1990;19:  117-122. 256. Lunsford LD, Pollack BE, Kondziolka DS, et al. Stereotactic options in the management of craniopharyngioma. Pediatr Neurosurg 1994;21:90-97. 257. Backlund EO, Axelsson B, Bergstrand CG, et al. Treatment of craniopharyngiomas—the stereotactic approach in a ten to twenty-three years’ perspective. Surgical, radiological and ophthalmological aspects. Acta Neurochir 1989;99(Pt 1):11-19. 258. Baskin DS, Wilson CB. Surgical management of craniopharyngiomas: a review of 74 cases. J Neurosurg 1986;65:22-27. 259. Merchant TE, Kiehna EN, Sanford RA, et al. Craniopharyngioma: the St. Jude Children’s Research Hospital experience. Int J Radiat Oncol Biol Phys 2002;53:533-542. 260. Hoogenhout J, Otten BJ, Kazem I, et al. Surgery and radiation therapy in the management of craniopharyngiomas. Int J Radiat Oncol Biol Phys 1984;10:2293-2297. 261. Sung DI, Chang CH, Harisiadis L, et al. Treatment results in craniopharyngiomas. Cancer 1981;47:847-852. 262. Hart MN, Earle KM. Primitive neuroectodermal tumors of the brain in children. Cancer 1973;32:890-897. 263. Kosnik EJ, Boesel CP, Bau J, et al. Primitive neuroectodermal tumors of the central nervous system in children. J Neurosurg 1978;48:741-746. 264. Bennett JP, Rubinstein LJ. The biological behavior of primary cerebral neuroblastoma: a reappraisal of the clinical course in a series of 70 cases. Ann Neurol 1984;16:21-27. 265. Cruz-Sanchez FF, Rossi ML, Hughes JT, et al. Differentiation in embryonal neuroepithelial tumors of the central nervous system. Cancer 1991;67:965-976. 266. Berger MS, Edwards MSB, Wara WM, et al. Primary cerebral neuroblastoma: long-term follow-up review and therapeutic guidelines. J Neurosurg 1983;59:418-423. 267. Gaffney CC, Sloane JP, Bradley NJ, et al. Primitive neuroectodermal tumors of the cerebrum: pathology and treatment.  J Neurooncol 1985;3:23-33. 268. Berger MS, Edwards MSB, LaMasters D, et al. Pediatric brainstem tumors: radiographic, pathological, and clinical correlations. Neurosurgery 1983;12:298-302. 269. Horten BC, Rubinstein LJ. Primary cerebral neuroblastoma: a clinicopathological study of 35 cases. Brain 1976;99:735-756. 270. Ashwal S, Hinshaw DB Jr, Bedros A. CNS primitive neuroectodermal tumors of childhood. Med Pediatr Oncol 1984;12:  180-188. 271. Zeltzer PM, Boyett JM, Finlaye JL, 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 study. J Clin Oncol 1999;17:832-845. 272. Stiller CA, Bunch KJ. Brain and spinal tumors in children aged under two years: incidence and survival in Britain, 1971-1985. Br J Cancer 1992;66:S50-S53. 273. Cohen BH, Packer RJ, Siegel KR, et al. Brain tumors in children under 2 years: treatment, survival and long-term prognosis. Pediatr Neurosurg 1993;19:171-179. 274. Brown K, Mapstone TB, Oakes WJ. A modern analysis of intracranial tumors of infancy. Pediatr Neurosurg 1997;26:25-32. 275. Gilles FH, Sobel EL, Tavare CJ, et al. Age-related changes in ­diagnoses, histological features, and survival in children with brain tumors: 1930-1979. Neurosurgery 1995;37:1056-1068.

33  The Brain and Spinal Cord 276. Rorke LB, Packer RJ, Biegel JA, et al. Central nervous system atypical teratoid/rhabdoid tumors of infancy and childhood: definition of an entity. J Neurosurg 1996;85:56-65. 277. Burger PC, Yu I-T, Tihan T, et al. Atypical teratoid/rhabdoid tumor of the central nervous system: a highly malignant tumor of infancy and childhood frequently mistaken for medulloblastoma. A Pediatric Oncology Group study. Am J Surg Pathol 1998;22:1083-1092. 278. Pancelet P, Sainte-Rose C, Lellough-Tubiana A, et al. Papillomas and carcinomas of the choroid plexus in children. J Neurosurg 1998;88:521-528. 279. Weiss E, Behring B, Behnke J, et al. Treatment of primary malig­ nant rhabdoid tumor of the brain: report of three cases and review of the literature. Int J Radiat Oncol Biol Phys 1998;41:  1013-1019. 280. Taratuto AL, Monges J, Lylyk P, et al. Superficial cerebral astrocytoma attached to dura: report of six cases in infants. Cancer 1984;54:2505-2512. 281. Mulhern RK, Hancock J, Fairclough D, et al. Neuropsychological status of children treated for brain tumors: a critical review and integrative analysis. Med Pediatr Oncol 1992;20:181-191. 282. Evans AE, Jenkin RDT, Sposto R, 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;72:572-582. 283. Deutsch M. Radiotherapy for primary brain tumors in very young children. Cancer 1982;50:2785-2789. 284. Jenkin D, Greenberg M, Hoffman H, et al. Brain tumors in children: long-term survival after radiation therapy. Int J Radiat ­Oncol Biol Phys 1995;31:445-451. 285. Allen JC, Helson L, Jereb B. Preradiation chemotherapy for just this minute diagnosed childhood brain tumors: a modified phase II  trial. Cancer 1983;52:2001-2006. 286. Baram TZ, van Eys J, Dowell RE, et al. Survival and neurologic outcome in infants with medulloblastoma treated with surgery and MOPP chemotherapy: a preliminary report. Cancer 1987;60:173-177. 287. Horowitz ME, Mulhern RK, Kun LE, et al. Brain tumors in the very young child: postoperative chemotherapy in combined­modality treatment. Cancer 1988;61:428-434. 288. Duffner PK, Horowitz ME, Krischer JP, et al. The treatment of malignant brain tumors in infants and very young children: an update of the Pediatric Oncology Group experience. Neurooncology 1999;1:152-161. 289. Geyer JR, Zeltzer PM, Boyett JM, et al. Survival of infants with primitive neuroectodermal tumors or malignant ependymomas of the CNS treated with eight drugs in 1 day: a report from the Children’s Cancer Group. J Clin Oncol 1994;12:1607-1615. 290. Kuhl J, Beck J, Bode U, et al. Delayed radiation therapy (RT) after postoperative chemotherapy (PCH) in children less than 3 years of age with medulloblastoma. Results of the trial HITSKK’87, and preliminary results of the pilot trial HIT-SKK’92 [abstract]. Med Pediatr Oncol 1995;25:250. 291. Walter AW, Mulhern RK, Gajjar A, et al. Survival and neurodevelopmental outcome of young children with medulloblastoma at St. Jude Children’s Research Hospital. J Clin Oncol 1999;17:3720-3728. 292. Dupuis-Girod S, Hartmann O, Benhamou E, 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. 293. Gajjar A, Mulhern RK, Heideman RL, et al. Medulloblastoma in very young children: outcome of definitive craniospinal irradiation following incomplete response to chemotherapy. J Clin Oncol 1994;12:1212-1216. 294. Rutkowski S, Bode U, Deinlein F, et al. Treatment of early childhood medulloblastoma by postoperative chemotherapy alone.  N Engl J Med 2005;352:978-986. 295. Grill J, Le Deley M-C., Gambarelli D, et al. Postoperative chemotherapy without irradiation for ependymoma in children under 5 years of age: a multicenter trial of the French Society of Pediatric Oncology. J Clin Oncol 2001;19:1288-1296.

867

296. Hilden JM, Watterson J, Longee DC, et al. Central nervous system atypical teratoid tumor/rhabdoid tumor: response to intensive therapy and review of the literature. J Neurooncol 1998;40:265-275. 297. Kernohan JW, Sayre GP. Tumors of the Central Nervous System. Washington, DC: American Registry of Pathology, Armed Forces Institute of Pathology, 1952. 298. Nelson JS, Taukada Y, Schoenfeld D, et al. Necrosis as a prognostic criterion in malignant supratentorial, astrocytic gliomas. Cancer 1983;52:550-554. 299. Werner-Wasik M, Scott CB, Nelson DF, et al. Final report of a phase I/II trial of hyperfractionated and accelerated hyperfractionated radiation therapy with carmustine for adults with supratentorial malignant gliomas. Radiation Therapy Oncology Group study 83-02. Cancer 1996;77:1535-1543. 300. Burger PC, Heinz ER, Shibata T, et al. Topographic anatomy and CT correlations in the untreated glioblastoma multiforme.  J Neurosurg 1988;68:698-704. 301. Halperin EC, Bentel G, Heinz ER, et al. Radiation therapy treatment planning in supratentorial glioblastoma multiforme: an analysis based on postmortem topographic anatomy with CT correlations. Int J Radiat Oncol Biol Phys 1989;17:1347-1350. 302. Kelly PJ, Daumas-Duport C, Kispert DB, et al. Imaging-based stereotaxic serial biopsies in untreated intracranial glial neoplasms. J Neurosurg 1987;66:865-874. 303. Choucair AK, Levin VA, Gutin PH, et al. Development of multiple lesions during radiation therapy and chemotherapy in patients with gliomas. J Neurosurg 1986;65:654-658. 304. Kopelson G, Linggood R. Infratentorial glioblastoma: the role of neuraxis irradiation. Int J Radiat Oncol Biol Phys 1982;8:  999-1003. 305. Nelson DF, Nelson JS, Davis DR, et al. Survival and prognosis of patients with astrocytoma with atypical or anaplastic features.  J Neurooncol 1985;3:99-103. 306. Walker MD, Green SB, Byar DP, et al. Randomized comparisons of radiotherapy and nitrosoureas for the treatment of malignant glioma after surgery. N Engl J Med 1980;303:1323-1329. 307. Prados MD, Gutin PH, Phillips TL, et al. Highly anaplastic ­astrocytoma: a review of 357 patients treated between 1977 and 1989. Int J Radiat Oncol Biol Phys 1992;23:3-8. 308. Wood JR, Green SB, Shapiro WR. The prognostic importance of tumor size in malignant gliomas: a CT scan study by the Brain Tumor Cooperative Group. J Clin Oncol 1988;6:338-343. 309. Wisoff JH, Boyett JM, Berger MS, et al. Current neurosurgical management and the impact of the extent of resection in the treatment of malignant gliomas of childhood: a report of the Children’s Cancer Group Trial no. CCG-945. J Neurosurg 1998;89:52-59. 310. Garden AS, Maor MH, Yung WKA, et al. Outcome and patterns of failure following limited-volume irradiation for malignant astrocytomas. Radiother Oncol 1991;20:99-110. 311. Fine HA, Dear KB, Loeffler JS, et al. Meta-analysis of radiation therapy with and without adjuvant chemotherapy for malignant gliomas in adults. Cancer 1993;71:2585-2597. 312. Murray KJ, Nelson DF, Scott C, et al. Quality-adjusted survival analysis of malignant glioma. Patients treated with twice-daily radiation (RT) and carmustine: a report of Radiation Therapy Oncology Group study 83-02. Int J Radiat Oncol Biol Phys 1995;31:453-459. 313. Sneed PK, Gutin PH, Larson DA, et al. Patterns of recurrence of glioblastoma multiforme after external irradiation followed by implant boost. Int J Radiat Oncol Biol Phys 1994;29:  719-727. 314. Hochberg FH, Pruitt A. Assumptions in the radiotherapy of glioblastoma. Neurology 1980;30:907-911. 315. Wallner KE, Galicich JH, Krol G, et al. Patterns of failure following treatment for glioblastoma multiforme and anaplastic astrocytoma. Int J Radiat Oncol Biol Phys 1989;16:1405-1409. 316. Loeffler JS, Alexander E III, Hochberg FH, et al. Clinical patterns of failure following stereotactic interstitial irradiation for malignant gliomas. Int J Radiat Oncol Biol Phys 1990;19:  1455-1462.

868

PART 10  Central Nervous System

317. Agbi CB, Bernstein M, Laperriere N, et al. Patterns of recurrence of malignant astrocytoma following stereotactic interstitial brachytherapy with 125I implants. Int J Radiat Oncol Biol Phys 1992;23:321-326. 318. Laramore GE, Diener-West M, Griffin TW, et al. Randomized neutron dose searching study for malignant gliomas of the brain: results of an RTOG study. Int J Radiat Oncol Biol Phys 1988;14:1093-1102. 319. Loeffler JS, Alexander E III, Wen PY, et al. Results of stereotactic brachytherapy used in the initial management of patients with glioblastoma. J Natl Cancer Inst 1990;82:1918-1921. 320. Scharfen CO, Sneed PK, Wara WM, et al. High activity iodine125 interstitial implant for gliomas. Int J Radiat Oncol Biol Phys 1992;24:583-591. 321. Prados MD, Gutin PH, Phillips TL, et al. Interstitial brachytherapy for newly diagnosed patients with malignant gliomas: the UCSF experience. Int J Radiat Oncol Biol Phys 1992;24:593-597. 322. Stelzer KJ, Laramore GE, Griffin TW, et al. Fast neutron radiotherapy: the University of Washington experience. Acta Oncol 1994;33. 275-280. 323. Larson DA, Gutin PH, McDermott M, et al. Gamma knife for glioma: selection factors and survival. Int J Radiat Oncol Biol Phys 1996;36:1045-1053. 324. Pickles T, Goodman GB, Rheaume DE, et al. Pion radiation for high grade astrocytoma: results of a randomized study. Int J Radiat Oncol Biol Phys 1997;37:491-497. 325. Sneed PK, Prados MD, McDermott MW, et al. Large effect of age on survival of patients with glioblastoma treated with radiotherapy and brachytherapy boost. Neurosurgery 1995;36:898-904. 326. Wen PY, Alexander E III, Black PM, et al. Long-term results of stereotactic brachytherapy used in the initial treatment of patients with glioblastomas. Cancer 1994;73:3029-3036. 327. Shrieve DC, Alexander E III, Wen PY, et al. Comparison of stereotactic radiosurgery and brachytherapy in the treatment of recurrent glioblastoma multiforme. Neurosurgery 1995;36:275-284. 328. Mehta MP, Masciopinto J, Rozental J, et al. Stereotactic radiosurgery for glioblastoma multiforme: report of a prospective study evaluating prognostic factors and analyzing long-term survival advantage. Int J Radiat Oncol Biol Phys 1994;30:541-549. 329. Buatti JM, Friedman WA, Bova FJ, et al. Linac radiosurgery for high grade gliomas: the University of Florida experience. Int J Radiat Oncol Biol Phys 1995;32:205-210. 330. Nelson DF, Diener-West M, Weinstein AS, et al. A randomized comparison of misonidazole sensitized radiotherapy plus BCNU and radiotherapy plus BCNU for treatment of malignant glioma after surgery: final report of an RTOG study. Int J Radiat Oncol Biol Phys 1986;12:1793-1800. 331. Urtusan R, Band P, Chapman JD, et al. Radiation and high-dose metronidazole in supratentorial glioblastomas. N Engl J Med 1976;294:1364-1367. 332. Miralbell R, Mornex F, Greiner R, et al. Accelerated radiotherapy, carbogen, and nicotinamide in glioblastoma multiforme: ­report of European Organization for Research and Treatment of Cancer Trial 22933. J Clin Oncol 1999;17:3143-3149. 333. Hegarty TJ, Thornton AF, Diaz RF, et al. Intraarterial bromodeoxyuridine radiosensitization of malignant gliomas. Int J Radiat Oncol Biol Phys 1990;19:421-428. 334. Sullivan FJ, Herscher LL, Cook JA, et al. National Cancer Institute (phase II) study of high-grade glioma treated with accelerated hyperfractionated radiation and iododeoxyuridine: results in anaplastic astrocytoma. Int J Radiat Oncol Biol Phys 1994;30:583-590. 335. Groves MD, Maor MH, Meyers C, et al. A phase II trial of high-dose bromodeoxyuridine with accelerated fractionation radiotherapy followed by procarbazine, lomustine, and vincristine for glioblastoma multiforme. Int J Radiat Oncol Biol Phys 1999;45:127-135. 336. Prados MD, Scott C, Sandler H, et al. A phase 3 randomized study of radiotherapy plus procarbazine, CCNU, vincristine (PCV) with or without BrdU for the treatment of anaplastic astrocytoma: a preliminary report of the Radiation Therapy Oncology Group study 9404. Int J Radiat Oncol Biol Phys 1999;45:1109-1115.

337. Saroja KR, Mansell J, Hendrickson FR, et al. Failure of accelerated neutron therapy to control high grade astrocytomas. Int J Radiat Oncol Biol Phys 1989;17:1295-1297. 338. Fitzek MM, Thornton AF, Rabinov JD, et al. Accelerated proton/ photon irradiation to 90 cobalt gray equivalent for glioblastoma multiforme: results of a phase II prospective trial. J Neurosurg 1999;91:251-260. 339. Leibel SA, Gutin PH, Wara WM, et al. Survival and quality of life after interstitial implantation of removable high-activity iodine 125 sources for treatment of patients with recurrent malignant gliomas. Int J Radiat Oncol Biol Phys 1989;17:1129-1139. 340. Selker RG, Shapiro WR, Green S, et al. A randomized trial of interstitial radiotherapy (IRT) boost for the treatment of newly diagnosed malignant glioma (glioblastoma multiforme, anaplastic astrocytoma, anaplastic oligodendroglioma, malignant mixed glioma). Brain Tumor Cooperative Group study 87-01 [abstract]. Paper presented at the 45th Meeting of the Congress of Neurological Surgeons. San Francisco: CA, October 14-18, 1995. 341. Florell RC, MacDonald DR, Irish W, et al. Selection bias, survival, and brachytherapy for glioma. J Neurosurg 1992;76:  179-183. 342. Loeffler JSI, Alexander E III, Shea WM, et al. Radiosurgery as part of the initial management of patients with malignant gliomas. J Clin Oncol 1992;10:1379-1385. 343. Sankaria JN, Mehta MP, Loeffler JS, et al. Radiosurgery in the initial management of malignant gliomas: survival comparison with the Radiation Therapy Oncology Group recursive partitioning analysis. Int J Radiat Oncol Biol Phys 1995;32:931-942. 344. Nelson DF, Curran WJ Jr, Scott C, et al. Hyperfractionated  radiation therapy and BIS-chlorethyl nitrosourea in the treatment of malignant glioma: possible advantage observed at  72 Gy in 1.2 Gy bid fractions: report of the Radiation Therapy Oncology Group protocol 8302. Int J Radiat Oncol Biol Phys 1993;25:193-207. 345. Fallia C, Olmi P. Hyperfractionated and accelerated radiation therapy in central nervous system tumors (malignant gliomas, pediatric tumors, and brain metastases). Radiother Oncol 1997;43:235-246. 346. Brada M, Sharpe G, Rajan B, et al. Modifying radical radiotherapy in high grade gliomas: shortening the treatment through acceleration. Int J Radiat Oncol Biol Phys 1999;43:287-292. 347. Levin VA, Maor MH, Thall PF, et al. Phase II study of accelerated fractionation radiation therapy with carboplatin followed by vincristine chemotherapy for the treatment of glioblastoma multiforme. Int J Radiat Oncol Biol Phys 1995;33:357-364. 348. Halperin EC. Multiple-fraction-per-day external beam radiotherapy for adults with supratentorial malignant gliomas. J Neurooncol 1992;14:255-262. 349. Fulton DS, Urtasun RC, Scott-Brown I, et al. Increasing radiation dose intensity using hyperfractionation on patients with malignant glioma: final report of a prospective phase I-II dose response study. J Neurooncol 1992;14:63-72. 350. Shapiro WR, Green SB, Burger PC, et al. Randomized trial of three chemotherapy regimens and two radiotherapy regimens in postoperative treatment of malignant glioma: Brain Tumor ­Cooperative Group Trial no. 8001. J Neurosurg 1989;71:1-9. 351. Levin VA, Silver P, Hannigan J, et al. Superiority of post­radiotherapy adjuvant chemotherapy with CCNU, procarbazine, and vincristine (PCV) over BCNU for anaplastic gliomas: NCOG 6G61 final report. Int J Radiat Oncol Biol Phys 1990;18:  321-324. 352. Medical Research Council Brain Tumour Working Party. Randomized trial of procarbazine, lomustine, and vincristine in the adjuvant treatment of high-grade astrocytoma: a Medical ­Research Council Trial. J Clin Oncol 2001;19:509-518. 353. Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma.  N Engl J Med 2005;352:987-996. 354. Athanassiou H, Synodinou M, Maragoudakis E, et al. Randomized phase II study of temozolomide and radiotherapy compared with radiotherapy alone in newly diagnosed glioblastoma multiforme. J Clin Oncol 2005;23:2372-2377.

33  The Brain and Spinal Cord 355. Hegi ME, Diserens AC, Gorlia T, et al. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med 2005;352:997-1003. 356. del Rowe J, Scott C, Werner-Wasik M, et al. Single-arm, openlabel phase II study of intravenously administered tirapazamine and radiation therapy for glioblastoma multiforme. J Clin Oncol 2000;18:1254-1259. 357. Valtonen S, Timonen U, Toivanan P, et al. Interstitial chemotherapy with carmustine-loaded polymers for high-grade gliomas: a randomized double-blind study. Neurosurgery 1997;41:44-49. 358. Nakagawa K, Aoki Y, Fujimaki T, et al. High-dose conformal radiotherapy influenced the pattern of failure but did not improve survival in glioblastoma multiforme. Int J Radiat Oncol Biol Phys 1998;40:1141-1149. 359. Sheline GE. Radiotherapy for high grade gliomas. Int J Radiat Oncol Biol Phys 1990;18:793-803. 360. Yung WKA, Janus TJ, Maor M, et al. Adjuvant chemotherapy with carmustine and cisplatin for patients with malignant gliomas. J Neurooncol 1992;12:131-135. 361. Heideman RL, Kuttesch J, Gajjar AJ, et al. Supratentorial malignant gliomas in childhood: a single institution perspective. Cancer 1997;80:497-504. 362. Finlay JL, Boyett JM, Yates AJ, et al. Randomized phase III trial in childhood high-grade astrocytoma comparing vincristine, lomustine, and prednisone with the eight-drugs-in-1 regimen.  J Clin Oncol 1995;13:112-123. 363. Mork SJ, Lindegaard K-F., Halvorsen TB, et al. Oligodendro­ glioma: incidence and biological behavior in a defined population. J Neurosurg 1985;63:881-889. 364. Jenkins RB, Blair H, Ballman KV, et al. A t(1;19)(q10;p10) mediates the combined deletions of 1p and 19q and predicts a better prognosis of patients with oligodendroglioma. Cancer Res 2006;66:9852-9861. 365. Brandes AA, Tosoni A, Cavallo G, et al. Correlations between O6-methylguanine DNA methyltransferase promoter methylation status, 1p and 19q deletions, and response to temozolomide in anaplastic and recurrent oligodendroglioma: a prospective GICNO study. J Clin Oncol 2006;24:4746-4753. 366. Shaw EG, Scheithauer BW, O’Fallon JR, et al. Oligodendrogliomas: the Mayo Clinic experience. J Neurosurg 1992;76:  428-434. 367. Shaw EG, Scheithauer BW, O’Fallon JR. Management of ­supratentorial low-grade gliomas. Oncology 1993;7:97-107. 368. Sheline GE, Boldney E, Karisburg P, et al. Therapeutic considerations in tumors affecting the central nervous system: oligodendrogliomas. Radiology 1964;82:44-49. 369. Chin HW, Hazel JJ, Kim TH, et al. Oligodendrogliomas. I. A clinical study of cerebral oligodendrogliomas. Cancer 1980;45:  1458-1466. 370. Lindegaard K-F., Mork SJ, Eide GE, et al. Statistical analysis of clinicopathological features, radiotherapy, and survival in 270 cases of oligodendroglioma. J Neurosurg 1987;67:224-230. 371. Reedy PD, Bay JW, Hahn JF. Role of radiation therapy in the treatment of cerebral oligodendroglioma: an analysis of 57 cases and a literature review. Neurosurgery 1983;67:224-230. 372. Allison RR, Schulsinger A, Vongtama V, et al. Radiation and chemotherapy improve outcome in oligodendroglioma. Int J ­Radiat Oncol Biol Phys 1997;37:399-403. 373. Ludwig CL, Smith MT, Godfrey AD, et al. A clinicopathological study of 323 patients with oligodendrogliomas. Ann Neurol 1986;19:15-21. 374. Cairncross JG, MacDonald DR. Successful chemotherapy for ­recurrent malignant oligodendroglioma. Ann Neurol 1988;23:  460-464. 375. Cushing H, Eisenhardt L. Meningiomas: Their Classification, ­Regional Behavior, Life History, and Surgical End Results. Springfield, IL: Charles C Thomas, 1938. 376. Glaholm J, Bloom HJG, Crow JH. The role of radiotherapy in the management of intracranial meningiomas: the Royal Marsden Hospital experience with 186 patients. Int J Radiat Oncol Biol Phys 1990;18:755-761.

869

377. Dziuk T, Woo S, Butler B, et al. Malignant meningioma: an ­indication for initial aggressive surgery and adjuvant radiotherapy.  J Neurooncol 1998;37:177-188. 378. Jaaskelainen J, Haltia M, Servo A. Atypical and anaplastic meningiomas: radiology, surgery, radiotherapy, and outcome. Surg Neurol 1986;25:233-242. 379. Mirimanoff RO, Rosoretz DE, Linggood RM, et al. Meningioma: analysis of recurrence and progression following neurosurgical resection. J Neurosurg 1985;62:18-24. 380. Goldsmith BJ, Wara WM, Wilson CB, et al. Postoperative ­irradiation for subtotally resected meningiomas. A retrospective analysis of 140 patients treated from 1967 to 1990. J Neurosurg 1994;80:195-201. 381. Petty AM, Kun LE, Meyer GA. Radiation therapy for incompletely resected meningiomas. J Neurosurg 1985;62:502-507. 382. Wara WM, Sheline GE, Newman H, et al. Radiation therapy of meningiomas. AJR Am J Roentgenol 1975;123:453-458. 383. Carelia RJ, Ransohoff J, Newall J. Role of radiation therapy in the management of meningioma. Neurosurgery 1982;10:332339. 384. 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;2:1139-1143. 385. Maguire PD, Clough R, Friedman AH, et al. Fractionated external-beam radiation therapy for meningiomas of the cavernous sinus. Int J Radiat Oncol Biol Phys 1999;44:75-79. 386. Lunsford L. Contemporary management of meningiomas: radiation therapy as an adjuvant and radiosurgery as an alternative to surgical removal? J Neurosurg 1994;80:187-190. 387. Kondziolka D, Flickinger JC, Perez B. Judicious resection and/ or radiosurgery for parasagittal meningiomas: outcomes from a multicenter review. Neurosurgery 1998;43:405-413. 388. Kondziolka D, Levy EI, Niranjan A, et al. Long-term outcomes after meningioma radiosurgery: physician and patient perspectives. J Neurosurg 1999;91:44-50. 389. Shafron DH, Friedman WA, Buatti JM, et al. Linac radiosurgery for benign meningiomas. Int J Radiat Oncol Biol Phys 1999;43:321-327. 390. Smith JL, Vuksanovic MM, Yates BM, et al. Radiation therapy for primary optic nerve meningiomas. J Clin Neuroophthalmol 1981;1:85-89. 391. Borovich B, Doron Y, Braun J, et al. Recurrence of intracranial meningiomas: the role played by regional multicentricity. Clinical and radiological aspects. J Neurosurg 1986;65(Pt 2):  168-171. 392. Roche P-H, Regis J, Dufour H, et al. Gamma knife radiosurgery in the management of cavernous sinus meningiomas. J Neurosurg 2000;93:68-73. 393. Ojemann SG, Sneed PK, Larson DA, et al. Radiosurgery for malignant meningioma: results in 22 patients. J Neurosurg 2000;93:62-67. 394. Deen HC Jr, Scheithauer BW, Ebersold MJ. Clinical and pathologic study of meningiomas of the first two decades of life.  J Neurosurg 1982;56:317-322. 395. Drake JM, Hendrick EB, Becker LE, et al. Intracranial meningiomas in children. Pediatr Neurosci 1986;12:134-139. 396. Snow RB, Lavyne MH. Intracranial space-occupying lesions in AIDS patients. Neurosurgery 1985;16:148-153. 397. So YT, Beckstead JH, Davis RL. Primary central nervous system lymphoma in AIDS: a clinical and pathologic study. Ann Neurol 1986;20: 566-S72. 398. Reni M, Ferreri AJ, Garancini MP, et al. Therapeutic management of primary central nervous system lymphoma in immunocompetent patients: results of a critical review of the literature. Ann Oncol 1997;8:227-234. 399. Jack CR, Reese DF, Scheithauer BW. Radiographic findings in 32 cases of primary CNS lymphoma. AJR Am J Roentgenol 1986;146:271-276. 400. Hochberg FH, Miller DC. Primary central nervous system lymphoma. J Neurosurg 1988;68:835-853. 401. Murray K, Kun L, Cox J. Primary malignant lymphoma of the central nervous system. J Neurosurg 1986;65:600-607.

870

PART 10  Central Nervous System

402. Socie G, Piprot-Chauffat C, Schlienger M, et al. Primary lymphoma of the central nervous system: an unresolved therapeutic problem. Cancer 1990;65:322-326. 403. Nelson DF, Martz KL, Bonner H, 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 8315. Int J Radiat Oncol Biol Phys 1992;23:9-17. 404. Loeffler JS, Ervin TJ, Mauch P, et al. Primary lymphomas of the central nervous system: patterns of failure and factors that influence survival. J Clin Oncol 1985;3:490-494. 405. Laperriere NJ, Cerezo L, Milosevic MF, et al. Primary lymphoma of brain: results of management of a modern cohort with radiation therapy. Radiother Oncol 1997;43:247-252. 406. Glass J, Gruber ML, Cher L, et al. Preirradiation methotrexate chemotherapy of primary central nervous system lymphoma: long-term outcome. J Neurosurg 1994;81:188-195. 407. Abrey LE, DeAngelis LM, Yahalom J. Long-term survival in primary CNS lymphoma. J Clin Oncol 1998;16:859-863. 408. Abrey LE, Yahalom J, DeAngelis LM. Treatment for primary CNS lymphoma: the next step. J Clin Oncol 2000;18:  3144-3150. 409. Schultz C, Scott C, Sherman W, 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;14:556-564. 410. O’Neill BP, Wang C-H., O’Fallon JR, et al. Primary central nervous system non-Hodgkin’s lymphoma (PCNSL): Survival advantages with combined initial therapy? A final report of the North Central Cancer Treatment Group study 86-72-52. Int J Radiat Oncol Biol Phys 1999;43:559-563. 411. Ferreri AJM, Reni M, Villa E. Therapeutic management of primary central nervous system lymphoma: lessons from prospective trials. Ann Oncol 2000;11:927-937. 412. DeAngelis LM, Seiferheld W, Schold SC, et al. Combination chemotherapy and radiotherapy for primary central nervous system lymphoma: Radiation Therapy Oncology Group Study 93-10. J Clin Oncol 2002;20:4643-4648. 413. Posner JB. Management of central nervous system metastases. Semin Oncol 1977;4:81-91. 414. Pickren JW, Lopez G,Tsukada Y, et al. Brain metastases: an autopsy  study. Cancer Treatment Symposia 1983;2:295-313. 415. Gaspar L, Scott C, Rotman M, et al. Recursive partitioning analysis (RPA) of prognostic factors in three Radiation Therapy Oncology Group brain metastases trials. Int J Radiat Oncol Biol Phys 1997;37:745-751. 416. Swift PS, Phillips T, Martz K, et al. CT characteristics of patients with brain metastases treated in Radiation Therapy ­ Oncology Group study 79-16. Int J Radiat Oncol Biol Phys 1993;25:  209-214. 417. Smalley SR, Laws ER Jr, O’Fallon JR, et al. Resection for solitary brain metastasis: role of adjuvant radiation and prognostic variables in 229 patients. J Neurosurg 1992;77:531-540. 417a. Borgelt G, Gelber R, Kramer S, et al. The palliation of brain metastases: final results of the first two studies by the Radiation theraphy Oncology Group. Int J Radiat Oncol Biol Phys. 1980;6:11-9. 418. Shaw EG. Radiotherapeutic management of multiple brain metastases: “3000 in 10” whole brain radiation is no longer a “no brainer.” Int J Radiat Oncol Biol Phys 1999;45:253-254. 419. Auchter RM, Lamond JP, Alexander E III, et al. A multi­institutional outcome and prognostic factor analysis of radiosurgery for resectable single brain metastasis. Int J Radiat Oncol Biol Phys 1996;35:27-35. 420. Patchell RA, Tibbs PA, Walsh JW, et al. A randomized trial of surgery in the treatment of single metastases to the brain. N Engl J Med 1990;322:494-500. 421. Alexander E III, Moriarty TM, Davis RB, et al. Stereotactic radiosurgery for the definitive, noninvasive treatment of brain metastases. J Natl Cancer Inst 1995;87:34-40.

422. Flickinger JC, Kondziolka D, Lunsford LD, et al. A multi­institutional experience with stereotactic radiosurgery for ­resectable single brain metastasis. Int J Radiat Oncol Biol Phys 1994;28:797-802. 423. Smalley SR, Laws ER Jr, O’Fallon JR, et al. Resection for solitary brain metastasis: role of adjuvant radiation and prognostic variables in 229 patients. J Neurosurg 1992;77:531-540. 424. Aoyama H, Shirato H, Tago M, et al. Stereotactic radiosurgery plus whole-brain radiation therapy vs stereotactic radiosurgery alone for treatment of brain metastases: a randomized controlled trial. JAMA 2006;295:2483-2491. 425. Choi KN, Withers HR, Rotman M. Intracranial metastases from melanoma: clinical features and treatment by accelerated fractionation. Cancer 1985;56:1-9. 426. Weltman E, Salvajoli JV, Brandt RA, et al. Radiosurgery for brain metastases: a score index for predicting prognosis. Int J Radiat Oncol Biol Phys 2000;46:1155-1161. 427. Mork SJ, Loken AC. Ependymoma: a follow-up study of 101 cases. Cancer 1977;40:907-915. 428. McCormick PC, Torres R, Post KD, et al. Intramedullary ependymoma of the spinal cord. J Neurosurg 1990;72:523-532. 429. Reimer R, Onofrio BM. Astrocytomas of the spinal cord in children and adolescents. J Neurosurg 1985;63:669-675. 430. Kopelson G, Linggood RM, Leinman GM, et al. Management of intramedullary spinal cord tumors. Radiology 1980;135:  473-479. 431. Chun HC, Schmidt-Ullrich RK, Wolfson A, et al. External beam radiotherapy for primary spinal cord tumors. J Neurooncol 1990;9:211-217. 432. Guidetti B, Mercuri S, Vagnozzi R. Long-term results of the surgical treatment of 129 intramedullary spinal gliomas. J Neurosurg 1981;54:323-330. 433. Cooper PR, Epstein F. Radical resection of intramedullary spinal cord tumors in adults: recent experience in 29 patients. J Neurosurg 1985;63:492-499. 434. Epstein J. Spinal cord astrocytomas of childhood. Prog Exp ­Tumor Res 1987;30:135-153. 435. Epstein FJ, Farmer J-P., Freed D. Adult intramedullary astrocytomas of the spinal cord. J Neurosurg 1992;77:355-359. 436. Epstein F, McCleary EL. Intrinsic brain-stem tumors of childhood: surgical indications. J Neurosurg 1986;64:11-15. 437. Epstein F. Intraaxial tumors of the cervicomedullary junction in children. Concepts Pediatr Neurosurg 1987;7:117. 438. Kopelson G, Linggood RM, Kleinman GM, et al. Management of intramedullary spinal cord tumors. Radiology 1980;135:  473-479. 439. O’Sullivan C, Jenkin RD, Doherty MA, et al. Spinal cord tumors in children: long-term results of combined surgical and radiation treatment. J Neurosurg 1994;81:507-512. 440. Kopelson G. Radiation tolerance of the spinal cord previously damaged by tumor and operation: long term neurological ­improvement and time-dose-volume relationships after irradiation of intraspinal gliomas. Int J Radiat Oncol Biol Phys 1982;8:  925-929. 441. Shaw EG, Evans RG, Scheithauer BW, et al. Radiotherapeutic management of adult intraspinal ependymomas. Int J Radiat Oncol Biol Phys 1986;12:323-327. 442. Garcia DM. Primary spinal cord tumors treated with surgery and postoperative irradiation. Int J Radiat Oncol Biol Phys 1985;11:1933-1939. 443. Bryne TN. Spinal cord compression from epidural metastases.  N Engl J Med 1992;327:614-619. 444. Rodichok LD, Harper GR, Ruckdeschel JC, et al. Early diagnosis of spinal epidural metastases. Am J Med 1981;70:1181-1188. 445. Young RF, Post EM, King GA. Treatment of spinal epidural ­metastases. J Neurosurg 1980;53:741-748. 446. Sorensen PS, Borgesen SE, Rohde K, et al. Metastatic epidural spinal cord compression: results of treatment and survival. Cancer 1990;65:1502-1508. 447. Latini P, Maranzano E, Ricci S, et al. Role of radiotherapy in metastatic spinal cord compression: preliminary results from a prospective trial. Radiother Oncol 1989;15:227-233.

33  The Brain and Spinal Cord 448. Sundaresan N, Galicich JH, Lane JM, et al. Treatment of neoplastic epidural cord compression by vertebral body resection and stabilization. J Neurosurg 1985;63:676-684. 449. Siegal T, Seigal T. Surgical decompression of anterior and posterior malignant epidural tumors compressing the spinal cord: a prospective study. Neurosurgery 1985;17:424-432.

871

450. Herbert SH, Solin LJ, Rate WR, et al. The effect of palliative radiation therapy on epidural compression due to metastatic ­malignant melanoma. Cancer 1991;67:2472-2476. 451. Olsen ME, Chernik NL, Posner JB. Infiltration of the leptomeninges by systemic cancer. Arch Neurol 1974;30:122-137.

Leukemias and Lymphomas 34 Bouthaina Dabaja, Chul S. Ha, James D. Cox RESPONSE OF NORMAL HEMATOPOIETIC TISSUES TO IRRADIATION Bone Marrow Changes in the Peripheral Blood after Total-Body Irradiation Lymph Nodes Spleen LEUKEMOGENIC EFFECTS OF IONIZING RADIATION TREATMENT OF LEUKEMIA HODGKIN LYMPHOMA Pathology and Clinical Presentation Epidemiology and Etiology Staging Evaluation Treatment of Early-Stage Disease: Selected Stage I and II

Treatment of Intermediate-Stage Disease: Unfavorable Stages I and II and Less-Advanced Stage III Treatment of Advanced Disease: Stage IIB, III, or IV Subdiaphragmatic Disease Lymphocyte-Predominant Hodgkin Lymphoma Radiation Therapy Techniques NON-HODGKIN LYMPHOMA Epidemiology and Etiology Pathology, Clinical Presentation, and Staging Subtypes of Non-Hodgkin Lymphoma

Response of Normal Hematopoietic Tissues to Irradiation The hematopoietic stem cells are among the most exquisitely radiosensitive cells in the body. Total-body irradiation (TBI) produces profound changes in the bone marrow, changes that are reflected in the peripheral blood at various intervals after irradiation. Similar doses delivered to whole blood in vitro produce few immediate or delayed effects.1 The changes in the peripheral blood reflect the effects of radiation on the blood-forming organs, not on the circulating cells themselves. Lymphopenia has been demonstrated after extracorporeal irradiation of the peripheral blood,2 but doses several orders of magnitude higher than those that cause bone marrow effects are required to have any influence on the other elements of the peripheral blood. The following discussion is confined to the clinically important changes produced by doses and types of ionizing radiation that may be encountered in the clinical setting.

Bone Marrow A voluminous literature is available concerning the effects of whole-body irradiation on laboratory animals. The single TBI dose that is lethal in 30 days for 50% of a group of laboratory animals, the LD50/30, is clearly defined for most mammals. For obvious reasons, little information is available for humans, and most of that comes from accidental exposures in which the dose to the bone marrow has been estimated with a greater or lesser degree of certainty. Tubiana and colleagues3-6 provided the most definitive data from their experiences with using TBI for immunosuppression purposes during the first renal transplantation

attempts in Europe between 1959 and 1962. Forty-one patients received total-body doses ranging from 1 to 6 Gy in 1 or 2 fractions. Patients were maintained in a controlled environment until leukocyte levels returned to normal. The investigators observed that the initial decrease in the peripheral blood counts was more rapid and the nadir was reached sooner with higher doses. They concluded that the LD50 in humans is higher than 4.5 Gy. The larger experience with single-dose or fractionated irradiation followed by bone marrow transplantation7,8 is somewhat more difficult to interpret because most ­patients treated with lethal doses followed by bone marrow reconstitution had primary disease processes that affected the bone marrow.9,10 Bloom11 carefully studied the changes in rabbit bone marrow after TBI. Mitosis in the marrow ceased 30 minutes after a dose of 3.5 Gy. Numbers of nucleated red cells had begun to decrease, and more mature forms predomin­ ated. The megakaryocytes remained normal in number and appearance at first but disappeared in 2 days. The rather abrupt and nearly simultaneous regeneration of platelets, reticulocytes, and granulocytes suggested a rapid proliferation of a multipotent stem cell with accelerated differ­ entiation during the fourth week after TBI. Local irradiation of limited amounts of bone marrow is associated with recovery of the marrow that is faster than recovery after TBI. Knospe and colleagues12 gave single doses of 20, 40, 60, or 100 Gy to the hind limbs of rats and examined the marrow at intervals up to 1 year. Regeneration of the bone marrow began 7 to 14 days after irradiation regardless of the dose, but the degree of cellularity at any time was dose-dependent. During the second and third months after irradiation, sinusoidal disruption and disappearance were observed in all irradiated marrow specimens.

875

876

PART 11  Leukemias and Lymphomas

This was associated with the maximum decrease in bone marrow cellularity. After 6 months, marrow regeneration was evident only in the animals that received 20 Gy or less, and this recovery followed regeneration of sinusoidal structures. This observation suggests that hematopoietic recovery depends on regeneration of the microcirculation of the bone marrow. Fishburn and colleagues13 demonstrated that marrow in juvenile rats could repopulate with higher levels of cellularity and after higher radiation doses than could the marrow from adult rodents. In humans, locally irradiated marrow shows a marked decrease in cellularity after being exposed to 4 Gy over a 3-day period.14 After a dose of 10 Gy given over 8 days, complete disappearance of normoblasts and early granulocytes occurs, along with a significant reduction in marrow cellularity. Some mature granulocytes persist after a dose of 20 Gy given over 16 days, but megakaryocytes are absent. No suppression of the unirradiated bone marrow is evident. Morardet and colleagues15,16 found that regeneration of the marrow depended on the absorbed dose and on the volume of bone marrow irradiated. With lesser volumes ­irradiated (20% to 30% of the bone marrow), recovery was less complete than when the volume of irradiated marrow was larger (60% or more). They suggested that this difference in regeneration was the result of increased stimulation of hematopoiesis caused by the inability of the unirradiated marrow to achieve sufficient production. Knowledge of the distribution of active marrow is useful in anticipating the magnitude of the effects of irradiating various sites in clinical practice. The best available information in this regard is the work of Ellis (Table 34-1).17 Approximately 40% of the active bone marrow in the adult is in the pelvis, and an additional 25% is located in the thoracic and lumbar vertebrae. This distribution is significant in the extensive treatment of patients with Hodgkin lymphoma (also called Hodgkin disease) or follicular lymphoma. In children, a relatively greater amount of active bone marrow is present in the long bones (45% at age 5 years), but this rapidly decreases in adolescence. After age 20, negligible amounts of active bone marrow are present in the long bones.17,18 High local doses such as those administered routinely in clinical radiation therapy may produce sufficient bone marrow injury that repopulation never occurs. Sykes and colleagues19,20 found failure of repopulation in the sternal bone marrow at doses of approximately 30 Gy, but doses up to 25 Gy consistently produced regeneration within 1 month. Total doses in excess of 40 Gy with common fractionation presumably produce bone marrow ablation. The failure of recovery is undoubtedly related to disruption of the microcirculation of the marrow. When the bone marrow is not repopulated, the predominant picture is one of fatty replacement. Despite the inability of the bone marrow to repopulate, it can support the growth of metastatic cancer.19,20 Studies of bone marrow regeneration after total nodal irradiation for Hodgkin lymphoma indicate that total doses of 40 Gy or more produce prolonged suppression of marrow that can continue for at least 1 year.21,22 For a group of 27 patients studied with bone marrow scanning agents, Rubin and colleagues21,22 reported that 50% showed evidence of partial or complete regeneration at 1 year, and 80% showed regeneration at 2 or 3 years after irradiation.

TABLE 34-1.   Distribution of Active Bone Marrow in the Adult Site

Total Red Marrow (%)

Head

13.1

Upper limb girdle

8.3

Sternum

2.3

Ribs

7.9

Vertebrae, cervical

3.4

Vertebrae, thoracic

14.1

Vertebrae, lumbar

10.9

Sacrum

13.9

Lower limb girdle

26.1

From Ellis RE. The distribution of active bone marrow in the adult. Phys Med Biol 1961;5:255-258.

They described redistribution of bone marrow activity in one half of the patients studied. Sacks and colleagues23 used indium 111 (111In) to evaluate the bone marrow of 48 patients (15 to 69 years old) given radiation therapy for a variety of malignant diseases (three fourths of which were Hodgkin neoplasms). They found that local marrow ­regeneration was influenced by the total dose, the age of the patient, and the total amount of bone marrow irradiated. Regeneration was seen in 25 of 31 patients who underwent treatment of bone marrow on both sides of the diaphragm but was seen in only 1 of 11 patients who underwent treatment of bone marrow on only one side of the diaphragm. These findings are in keeping with that of greater stimulation of marrow regeneration with larger treatment volumes described by Morardet and colleagues.15,16

Changes in the Peripheral Blood after Total-Body Irradiation Normally, a delicate balance exists among the life span of the circulating blood cells, the rate of replacement of these cells, and the body’s demand for them. Irradiation of hematopoietic tissues upsets this balance by reducing or ­interrupting the supply of blood cells. Leukopenia, granulocytopenia, anemia, and thrombocytopenia develop at rates that depend on the severity of the damage and the normal life span of the circulating cells. The postirradiation fluctuations of leukocyte counts and the percentages of granulocytes, erythrocytes, and platelets are shown graphically in Figure 38-1. The extreme sensitivity of lymphopoietic tissue manifests early as a rapid lymphocytopenia. The degree to which the short life span of the lymphocyte and its unique characteristic of undergoing interphase death explain this change in peripheral lymphocyte counts is unclear. Granulocyte counts change more slowly. They are little altered by a TBI dose of 0.5 Gy, but they change to a greater degree and more rapidly as the TBI dose increases (Fig. 38-2).

Platelets Megakaryocytes, from which platelets are produced, are the least radiosensitive of the myeloid elements. Platelets have a circulating life of 8 to 11 days. Radiation damage to

34  Leukemias and Lymphomas

A

877

B

C FIGURE 34-1.  Response of mesenteric lymph node of the rabbit to irradiation. A, Normal nodule with active germinal center (×245). B, Seventeen hours after a single dose of 800 roentgens (×245). C, Twenty-four hours after a single dose of 800 roentgens (×245). (Photographs courtesy of the United States Atomic Energy Commission; from De Bruyn PPH. Lymph node and intestinal tissue. In Bloom W [ed]: Histopathology of Irradiation. New York: McGraw-Hill, 1948.)

megakaryocytes is reflected in the circulating blood in 2 to 3 days. The nadir of thrombocytopenia is reached approximately 2 weeks after a single large dose of TBI (see Fig. 38-1), and recovery begins at about 3 weeks. Unlike their precursors, circulating platelets are resistant to high doses of radiation (75 Gy).24

Red Blood Cells

FIGURE 34-2.  Para-aortic lymph node 3 years after a dose of 35 Gy given in 4 weeks for metastatic seminoma. The nodal anatomy appears well preserved. The capsular and medullar ­sinusoids contain lipid droplets from a recent lymphangiogram. An embryonal carcinoma was discovered in the remaining testicle (×160).

Circulating red blood cells, like other circulating formed elements, are resistant to direct damage by irradiation. The normal red blood cell has an average circulating life span of 120 days. After TBI, the red blood cell count decreases more slowly and returns to normal earlier than do the other formed elements. In clinical practice, anemia rarely results from a course of radiation therapy. When fractionated low-dose TBI is given over a period of many months or years, an entirely different picture is seen in the peripheral blood. The highly radiosensitive erythropoietic stem cells are not allowed to recover, and anemia develops in addition to leukopenia. The exact dose levels necessary to produce these changes in humans are not known. Total doses above 20 Gy have been achieved with fractionated TBI for chronic lymphocytic leukemia.25

878

PART 11  Leukemias and Lymphomas

Lymph Nodes A vasculoconnective tissue framework serves as the supporting structure for lymphocytes and lymphoblasts within the lymph nodes. Lymph enters the nodes through sinuses and passes between lymphocytes and macrophages, some of which are phagocytic. Each of the cellular elements (e.g., lymphocytes, macrophages, endothelial cells) exhibits a different radiosensitivity. The cells that constitute the lymphopoietic tissues vary considerably in their sensitivity to ionizing radiation. The response of the lymphoid tissue in the rabbit is representative of this variation in radiation sensitivity.26 Within one-half hour after the delivery of an LD50/30 dose to rabbits (a total-body dose of approximately 8 Gy), nuclear debris appears in the lymph node and rapidly increases to a maximum at 8 hours. Macrophages are not noticeably ­altered, but they become active phagocytically as the debris appears. Most of the debris is phagocytized in 20 hours (Fig. 34-1).26 Within 24 hours, the fibrovascular framework and macrophages dominate the picture; hemorrhagic areas and perivascular edema can be seen. Approximately 5 days after the single-dose TBI, lymphocytes begin to reappear in the lymph node. Plasma cells are seen about 9 days after irradiation but become less numerous several weeks later. Three weeks after the LD50/30 dose, lymphopoiesis is observed in cortical areas, and by 4 months, the lymph node looks normal histologically. With smaller doses, destruction is less complete, and the return to normal is more rapid. For example, after 0.5 Gy, the lymph node appears normal in 27 hours. Engeset27 described the effects of local irradiation with 30 Gy given to the popliteal lymph nodes of rats. The same processes were evident as in the rabbit, but lymphocytic repopulation was evident at 24 hours; by 48 hours, the nuclear debris had disappeared. Although recovery was nearly complete 1 week after irradiation, chronic changes appeared after 2 weeks. A gradual increase in connective tissue and progressive decreases in germinal centers and cortical lymphocytes were observed. One year after irradiation, the normal lymph node architecture was almost completely disrupted. In addition to changing lymphopoietic function, moderately high radiation doses affect the reticuloendothelial elements. Doses of 20 to 30 Gy diminish their functions, and doses of 65 Gy abolish them,28 although even this high a dose does not lead to obstruction to the flow of lymph through the nodes. In terms of functional changes in reticuloendothelial cells, several investigators27,29-32 have evaluated the ability of locally irradiated lymph nodes to remove particulate material (e.g., tumor cells, erythrocytes, charcoal particles, colloidal gold) infused into the afferent lymphatics. Most of those studies suggest that the barrier function of nodes is reduced by irradiation. However, the experiments involved the use of a few large fractions and evaluated immediate effects. Correlation of these changes with those expected in the clinical setting is lacking. Transient suppression of lymph node function undoubtedly occurs in the clinical setting, but with the usual clinical fractionation and moderate or even high total doses, lymph nodes seem to recover rather completely (Fig. 34-2). Lymphatic vessels are resistant to doses of radiation that are sufficient to produce cutaneous and subcutaneous necrosis in the hind limb of dogs.33 Nineteen months after

conventional x-irradiation (36 Gy in 3 fractions in 14 days), Sherman and O’Brien33 demonstrated normal lymphatics in the irradiated zone of animals that had developed extensive soft tissue necroses. The resistance of lymphatic vessels to irradiation has been confirmed by Engeset27 and by Lenzi and Bassani.34 Although irradiation has been implicated as a direct cause of lymphatic obstruction, evidence of obliteration of lymphatic vessels is rarely seen in clinical situations except when the lymph node was known to be involved by cancer or when surgical intervention had interrupted normal lymphatic pathways. In such circumstances, lymphedema may develop as a result of fibrosis or enhancement of postoperative scarring.

Spleen The response of the spleen to ionizing radiation is very similar to that of the lymph nodes. The white pulp is similar in structure to lymph nodes, and the red pulp contains the remaining formed elements of whole blood. In humans, lymphopoiesis normally occurs in the white pulp of the spleen. Granulopoiesis occurs in the red pulp of the spleen in rabbits, rats, and mice (and in humans in certain abnormal conditions), thereby providing the opportunity to study the relative radiosensitivity of lymphopoiesis and granulopoiesis side by side. Murray35 described the morphologic changes in the spleen after a TBI dose of approximately 8 Gy in the rabbit. As previously described for the bone marrow, mitosis ceases within 30 minutes of this LD50/30 dose. By 3 hours, the white pulp retains only a few lymphocytes, and by 8 hours, the reticular cells show nuclear abnormalities, but they do not die. During this same period, destruction becomes evident in all elements of the red pulp except reticular cells and the cells of the circulating blood. Lymphopoiesis is reestablished in about 9 days, erythroblasts and myeloblasts reappear by 14 days, and myelopoiesis becomes hyperactive in about 3 weeks. As a result of the rapid cell loss, the spleen shrinks to one half of its normal weight within 24 hours. The false impression of reticuloendothelial hyperplasia is the result of the less radiosensitive cells being forced into a smaller volume. After a single dose of 8 Gy, lymphopoiesis is always seen before myelopoiesis, even though the lymphocyte count in the peripheral blood is slower to return to normal than the granulocyte count. Splenic atrophy can occur after irradiation with relatively high total doses (40 Gy),36 and the consequences are similar to those seen after splenectomy. However, experience with hundreds of ­patients followed for 10 or more years after treatment with total doses of 30 Gy suggests that functional consequences at this dose level are absent or rare.

Leukemogenic Effects of Ionizing Radiation The principles of mutagenesis and carcinogenesis, including leukemogenesis, can be found in Chapter 1. Susceptibility to the development of leukemia among mammalian species varies widely. Humans and mice are relatively susceptible, whereas rats, rabbits, and guinea pigs are relatively resistant. TBI has been demonstrated as a cause of leukemia in ­humans. The Radiation Effects Research Foundation,

34  Leukemias and Lymphomas

f­ ormerly the Atomic Bomb Casualty Commission, has carefully followed the residents of Hiroshima who were exposed to a single acute dose of photons and neutrons from a fission bomb. The relative risk of leukemia and several types of cancer (Fig. 34-3)37 was higher with the greater doses associated with proximity to the hypocenter. Acute granulocytic leukemia was seen within 18 months of exposure. The peak incidence of acute leukemia was seen 5 years after the blast and approached normal levels approximately 15 years later for those who were younger than 15 years at the time of the exposure. The risk for ­older individuals remained above normal more than 25 years ­after exposure.37 An increased risk of leukemia among the pioneers of radiology was suspected as early as 1911.38 March39 found an increased risk of death from leukemia among early radiologists compared with physicians who were not radiologists, and this risk persisted among physicians who died in the 1950s.40 With more complete understanding of this risk and appropriate rules for radiation protection, the increase in leukemia among radiologists has been eliminated entirely.41 Court-Brown and Abbatt42 described an increased frequency of death from leukemia in a group of patients given irradiation to the spine for ankylosing spondylitis between 1940 and 1954. The work of Court-Brown and Doll43 has suggested a linear dose-response relationship in radiation leukemogenesis, although the findings of the Radiation ­Effects Research Foundation44 have led investigators to suspect a linear-quadratic dose-response relationship. Levine and Bloomfield45 provided a comprehensive review of the extensive literature concerning the risk of acute leukemia and myelodysplastic syndromes resulting from treatment for Hodgkin lymphoma. They concluded that the use of radiation therapy alone, even with large fields, was associated with few or no cases of secondary acute leukemias: “The overwhelming majority of the cases resulted in the setting of chemotherapy use, almost always involving an alkylating agent.”45

Malignant tumors

879

Coltman and Dixon46 found an increased risk of acute leukemia after treatment of Hodgkin lymphoma that was related to the age at which the patients were treated. Those who were older than 40 years (153 patients) at the time of treatment had a risk of nearly 20%, compared with lower risks among patients 20 to 39 years old (376 patients) and an even lower risk for patients who were younger than 20 years. Intravenous radioactive phosphorus (32P) was thought to be a cause of leukemic evolution after its administration for polycythemia vera; however, a prospective trial of phlebotomy versus intravenous 32P administration revealed no difference in the frequency of leukemia after treatment.47

Treatment of Leukemia The natural history of the different types of acute leukemia before effective therapy was developed justified the designations of acute or chronic at that time. Therapeutic successes over the past 40 years have profoundly altered the natural history of some of these diseases, most notably acute leukemia in children.48,49 More than one half of children with acute leukemia, properly treated, are longterm survivors, free of all evidence of disease and no longer on therapy.50 A lesser degree of progress has occurred for adults with acute leukemia.51 Chronic leukemias have proved far more resistant to treatment, but recent results are encouraging. Radiation therapy has a limited role in the care of patients with leukemia. Its use is confined to palliative treatment. The most common indications for irradiation are urgent ones, including irradiation of the base of the skull for cranial nerve abnormalities due to infiltration of the leptomeninges or involvement of the cauda equina with pain and weakness in the lower extremities. The nature of these leukemic infiltrates is such that imaging studies may be normal. In the presence of manifest leukemia and neurologic abnormalities, irradiation on clinical grounds alone is often palliative. Larger tumor infiltrations may be seen in patients with lymphocytic or myelocytic leukemias. Biopsies of the former may be interpreted as lymphomas, and those of the latter may be called granulocytic sarcomas (formerly called chloromas). Low total doses such as 20 to 25 Gy are usually sufficient for rapid palliation.52

Risk

Hodgkin Lymphoma Acute leukemia

0

5

10

15

20

25

Years after exposure

FIGURE 34-3.  Schematic model of the risk of leukemia and malignant tumors in survivors of the Hiroshima atomic bomb. (Modified from United Nations Scientific Committee on the ­Effects of Atomic Radiation [UNSCEAR]. Genetic and somatic ­effects of ionizing radiation. 1986 Report to the General Assembly. New York: United Nations, 1986.)

Hodgkin lymphoma, also known as Hodgkin disease, is an unusual form of cancer in that the neoplastic cells (i.e., Reed-Sternberg cells) account for only a very small portion of the disease process; most of that process results from the wide range of inflammatory cells and fibrous bands that accompany the Reed-Sternberg cells. Hodgkin lymphoma arises mainly within the lymph nodes, and evidence is accumulating to suggest that most of the Reed-Sternberg cells are of B-cell origin.53 Considered uniformly fatal 50 years ago, Hodgkin lymphoma is now considered one of the most curable types of cancer.54 Areas of active research include understanding the pathogenesis of Hodgkin lymphoma at the molecular level and minimizing side effects from treatment while maintaining high cure rates. The remainder of this section on Hodgkin lymphoma consists of an overview of the disease, with emphasis on its management.

880

PART 11  Leukemias and Lymphomas

Pathology and Clinical Presentation According to the Revised European-American Classification of Lymphoid Neoplasms (REAL), published in 1994, Hodgkin disease can be classified as classic Hodgkin disease or a (nodular) lymphocyte-predominant form. Histologic subtypes of classic Hodgkin disease proposed at that time were nodular sclerosis, mixed cellularity, lymphocyte depletion, and lymphocyte-rich; the lymphocyte-rich category was considered provisional at that time.55 The later World Health Organization (WHO) classification scheme56 essentially incorporates the REAL classification; in the WHO scheme, the lymphocyte-rich form is considered a category of classic Hodgkin disease, and the terms Hodgkin disease and Hodgkin lymphoma are considered interchangeable.56 Classic Hodgkin disease can be distinguished from the lymphocyte-predominant form on the basis of morphology and immunophenotype. Morphologically, classic Hodgkin disease is characterized by the presence of Reed-Sternberg cells or lacunar cells on a background of lymphocytes, histiocytes, eosinophils, and plasma cells. The lymphocytepredominant form of Hodgkin disease is characterized by the presence of abundant lymphocytic and histiocytic (L&H or “popcorn”) cells. Immunophenotypically, the Reed-Sternberg cells in classic Hodgkin disease express the CD15 and CD30 cell surface antigens but not CD45; in the lymphocyte-predominant form, the L&H cells do not express CD15 or CD30 but do express CD45. CD20 is also expressed in the lymphocyte-predominant form but is not common in classic Hodgkin disease.55 Classic and lymphocyte-predominant forms of Hodgkin disease can be distinguished by their clinical presentations. Classic Hodgkin disease has a bimodal age distribution in which the first peak occurs around age 20 and the second past age 50; in contrast, the age at appearance of the lymphocyte-predominant form is not bimodal but rather is between the mid-20s and mid-30s. Classic Hodgkin disease tends to manifest with centrally located lesions at stage II or III and in about 40% of cases with constitutional or B symptoms (see “Staging Evaluation”). The lymphocyte-predominant form tends to manifest with peripheral ­lesions and at stage I or II, and less than 20% of patients have B symptoms at presentation.57

Epidemiology and Etiology The expected number of new cases of Hodgkin lymphoma in the United States in 2008 was 8220 (4400 male and 3820 female patients), and the number of deaths was estimated to be 1350 (700 male and 650 female patients).58 The most common histologic subtype of Hodgkin disease in the United States and Canada is nodular sclerosis (70%), followed by mixed cellularity (25%), lymphocyte predominant (5%), lymphocyte depleted (<1%), and lymphocyte rich (incidence not yet defined); the variations among countries and over time are substantial.57 Risk factors associated with Hodgkin disease are family history, higher socioeconomic and educational status, and Epstein-Barr virus infection. Monozygotic twins have a 99-fold increased risk of Hodgkin disease relative to dizygotic twins, suggesting a genetic basis for the disease.59 Evidence of links between Hodgkin disease and EpsteinBarr virus infection includes the appearance of elevated immunoglobulin G and A (IgG and IgA) antibody titers

at least 3 years before the diagnosis of Hodgkin disease.60 The detection of the viral genome in the Reed-Sternberg cells by in situ hybridization also raises the speculation that Epstein-Barr virus may be involved in the development of Hodgkin disease.61

Staging Evaluation Staging evaluations for Hodgkin disease should include complete history and physical examination; complete blood cell count; assessment of electrolytes, erythrocyte sedimentation rate (ESR), and β2-microglobulin levels; liver function tests; bilateral bone marrow biopsy and aspiration; chest radiography; computed tomography (CT) scanning of the neck, chest, abdomen, and pelvis; gallium scanning; and in some cases, lymphangiography. The most widely accepted staging system for Hodgkin disease remains the Ann Arbor system62 as modified at the Cotswolds meeting,63 although definitions of lymph node regions and interpretations of extralymphatic sites or ­organs have varied. Some generally accepted definitions of typically involved lymph node regions are the cervical, axillary, mediastinal, para-aortic, splenic, mesenteric, and iliac-inguinal-femoral regions; some investigators report involvement of the infraclavicular, hilar, and inguinofemoral areas separately.64 Less commonly involved regions are Waldeyer’s ring, the epitrochlear or brachial nodes, and the popliteal nodes. The Ann Arbor/American Joint Committee on Cancer system is summarized in Box 34-1.65 Each stage grouping also includes A and B categories, with B denoting the presence of constitutional symptoms such as unexplained temperatures above 38°C (100.4°F) for 3 or more consecutive days, drenching night sweats, or unexplained weight loss of more than 10% during the 6 months before diagnosis. The A designation denotes the absence of these symptoms.65

Treatment of Early-Stage Disease: Selected Stage I and II Radiation Therapy In the early 1920s, Rene Gilbert, a Swiss radiotherapist, recognized the importance of treating contiguous lymph node–bearing regions that are not clinically involved when radiation is to be used as the only treatment for Hodgkin disease.66 Despite this observation, the administration of low doses of radiation to clinically involved sites remained the standard of care over the next 4 decades. Consequently, Hodgkin disease was considered incurable. Although Peters reported in 1950 that early-stage Hodgkin disease could be cured with radiation therapy,67 physicians remained ­skeptical. In 1963, Easson and Russell’s landmark article, “The Cure of Hodgkin’s Disease,” reemphasized that radiation therapy could be curative if administered properly.68 In the 1960s and 1970s, Henry Kaplan at Stanford University helped to refine radiation therapy techniques, and with Saul ­Rosenberg, he helped popularize the use of radiation therapy for early-stage disease.69 This work led to the first decline in the age-adjusted mortality rate for Hodgkin disease after 1970.70 Since the 1970s, many patients with ­early-stage Hodgkin disease have been treated with subtotal nodal irradiation (also known as extended-field radiation therapy, which includes the treatment of the mantle and

34  Leukemias and Lymphomas

BOX 34-1.  Ann Arbor/American Joint Committee on Cancer Staging System for Hodgkin Disease

*The number of lymph node regions involved may be indicated by a subscript (e.g., II3). †The location of stage IV disease is identified further by specifying the site according to the notations listed for stage III disease. From Greene FL, Page DL, Fleming ID, et al (eds). AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 393-406.

para-aortic regions and the spleen), and radiation therapy remains the single most effective therapeutic agent. An extensive study of radiation therapy for clinical stage I or II supradiaphragmatic Hodgkin disease done at the Princess Margaret Hospital in Toronto involved 250 patients treated between 1978 and 1986 with a median follow-up of 6.3 years. In that study, the actuarial overall survival rate was 83%, the cause-specific survival rate was 90%, and the relapse-free survival (RFS) rate was 72% at 8 years (Fig. 34-4).71 The local control rate was 95%, with only two true in-field failures. The criteria for treatment with radiation only were evolving throughout the period of this retrospective analysis; nevertheless, the factors predicting favorable cause-specific survival were age less than 50 years and an ESR of less than 40, and factors predicting favorable RFS were age less than 50, ESR less than 40, and the lymphocyte-predominant or nodular-sclerosis histologic subtype.71 Some of the best-known prognostic factors for Hodgkin disease were derived from a series of prospective trials carried out by the European Organisation for Research and Treatment of Cancer (EORTC). This group classified stages I and II Hodgkin disease as one of three prognostic categories: very favorable, favorable, and unfavorable. Very favorable disease is that which meets all six of the following criteria: (1) clinical stage IA, occurring in (2) females

248

100

232

172

80 Percent

Stage Grouping I Involvement of a single lymph node region (I) or localized involvement of a single extralymphatic organ or site in the absence of any lymph node involvement (IE) (rare in Hodgkin lymphoma) II Involvement of two or more lymph node regions on the same side of the diaphragm (II) or localized involvement of a single associated extralymphatic organ or site in association with regional lymph node involvement with or without involvement of other lymph node regions on the same side of the diaphragm (IIE)* III Involvement of lymph node regions on both sides of the diaphragm (III), which also may be ­accompanied by extralymphatic ­extension in ­association with adjacent lymph node ­involvement (IIIE) or by involvement of the spleen (IIIS) or both (IIIE+S) IV Diffuse or disseminated involvement of one or more extralymphatic organs, with or without associated lymph node involvement, or isolated ­extralymphatic organ involvement in the absence of adjacent regional lymph node involvement, but in conjunction with disease in site(s), or any ­involvement of the liver or bone marrow, or ­nodular involvement of the lung(s)†

881

OS (250–64) RFR

93

60 40 20 0 0

1

2

3

4

5

6

7

8

Time (years)

FIGURE 34-4.  The actuarial overall survival (OS) and relapsefree survival (RFR) rates for 250 patients with stage I or II ­supradiaphragmatic Hodgkin disease treated with radiation at the Princess Margaret Hospital in Toronto, Canada, between 1978 and 1986. (From Gospodarowicz MK, Sutcliffe SB, Clark RM, et al. Analysis of supradiaphragmatic clinical stage I and II Hodgkin’s disease treated with radiation alone. Int J Radiat ­Oncol Biol Phys 1992;22:859-865.)

(3) younger than 40 years with (4) an ESR of less than 50; (5) lymphocyte-predominant or nodular-sclerosis histology; and (6) a mediastinum-to-thoracic ratio (i.e., ratio of the horizontal diameter of the mediastinal tumor to the thoracic diameter measured at the T5-T6 interspace) of less than 35%. The presence of any of the following factors connotes unfavorable disease: age 50 years or older, B symptoms, an ESR of 50 or more, disease at more than three sites, and a mediastinum-to-thoracic ratio of more than 35%. Favorable disease is connoted by clinical stage I or II disease that does not meet the very favorable or unfavorable criteria. Results from the H7 trials of the EORTC indicated that treatment of 165 patients with favorable disease with subtotal nodal irradiation produced an RFS rate of 81% and an overall survival rate of 96% at 6 years.72 Despite the high cure rates obtained with radiation therapy alone, some proportion of patients who are cured of Hodgkin disease will experience some long-term complications resulting from the radiation therapy (see “Radiation Toxicity” later in this chapter and see Chapter 1). Although pulmonary and cardiac complications have been reduced greatly through improvements in techniques and better understanding of how to minimize long-term side effects, the risk of developing a second malignancy persists, a risk that poses particular problems for young patients who are otherwise cured of the disease.73-75 Ongoing attempts to modify treatment for early-stage Hodgkin disease to reduce the adverse effects of radiation or chemotherapy without sacrificing high cure rates are ­described in the following paragraphs.

Radiation Volume Attempts to reduce the volume of radiation delivered to patients with early-stage Hodgkin disease include the recent EORTC H7VF trial, in which 40 patients with very favorable disease were treated with mantle radiation therapy only.72 At 6 years, the overall survival rate for this group was 96%, and the RFS rate was 73%. However, most ­relapses appeared in the nodal extension area, which

882

PART 11  Leukemias and Lymphomas

would have been covered by irradiation of the para-aortic and spleen areas. The RFS rate of 73% was thought to be too low, and mantle radiation therapy alone has not been used in any subsequent EORTC trial.76 The H5 trial of the EORTC involved 237 patients with favorable disease (i.e., clinical stage I or clinical stage II without mediastinal disease, age 40 or younger, ESR 70 or less, and lymphocyte-predominant or nodular-sclerosis histology).77 The 198 patients who had no evidence of disease on laparotomy were randomly assigned to undergo mantle field or mantle plus para-aortic field radiation therapy. No difference was found between the groups in the 6-year RFS rate (74% for mantle only versus 72% for mantle plus paraaortic) or overall survival rate (96% versus 89%). The H7 trial also compared radiation with radiation plus chemotherapy with epirubicin, bleomycin, vinblastine, and prednisone (EBVP regimen). Patients with favorable disease were randomly assigned to receive subtotal nodal ­irradiation (n = 165) or six cycles of EBVP followed by ­involved-field radiation therapy (n = 168).72,76 The RFS rate at 6 years was 81% for the radiation-only group and 92% for the chemotherapy plus radiation group (P = .004). The corresponding overall survival rates were 96% and 98% (P = .156), numbers that reflect a good salvage rate for those who experienced relapse after subtotal nodal irradiation. The H9F trial assigned patients with very favorable or favorable disease to undergo six cycles of EBVP followed by (1) no radiation therapy, (2) 20 Gy of involved-field radiation therapy, or (3) 36 Gy of involved-field radiation therapy.78 Preliminary results from that trial have been published in abstract form79; a full report awaits further analyses and a longer follow-up period.

Radiation Dose In an attempt to evaluate whether radiation dose could be reduced for early-stage Hodgkin disease, Vijayakumar and Myrianthopoulos80 reviewed the literature and found that the likelihood of in-field recurrence after the treatment of subclinical disease with radiation therapy alone to 27 Gy was only 5%. Seydel and colleagues81 and Mendenhall and coworkers82 have also suggested that 20 to 30 Gy is adequate for treating microscopic disease when the only treatment is to be radiation. For gross disease, Kaplan83 found a 4.4% recurrence rate after delivery of 35 to 40 Gy; an apparent linear dose­response relationship was described in this 1966 retrospective review, which included patients with non-Hodgkin lymphoma,84 a group in which the recurrence rate after radiation therapy was unusually high,85 and a group in which some patients were given 2.75 Gy daily to a single anterior or posterior field 4 days per week.86 After excluding these patients, Fletcher and Shukovsky87 observed a sigmoid dose-response relationship for Hodgkin disease, findings that were subsequently corroborated by several other investigators. In that analysis, administration of 35 Gy ­resulted in combined in-field and marginal recurrence rates of 5% or less. Vijayakumar and Myrianthopoulos80 reviewed megavoltage data from the treatment of 4117 at-risk sites for Hodgkin disease and suggested that achieving an infield control rate of 98% would require 36.9 Gy for disease smaller than 6 cm and 37.4 Gy for disease larger than 6 cm. Brincker and Bentzen88 later reported that fitting a dose-response curve to the data compiled by Vijayakumar

and Myrianthopoulos80 revealed a plateau in local control at doses greater than 32.5 Gy. In contrast, the Patterns of Care Outcome Study for Hodgkin disease89 and studies by Thar and colleagues,90 Schewe and associates,91 and Sears and coworkers92 found no benefit from doses higher than 30 Gy. Part of the discrepancy in these findings may have resulted from failure to distinguish marginal from in-field recurrences.93 Most of these analyses also used prescribed dose rather than the minimum dose actually delivered to the tumor. The German Hodgkin’s Lymphoma Study Group, in its HD4 trial, compared relapse-free and overall survival rates after 40 Gy given as extended-field irradiation (i.e., subtotal nodal irradiation) or 30 Gy given as extended-field irradiation followed by a 10-Gy boost to involved disease sites.94 Patients had pathologically proven stage IA, IIA, or IIB disease.94 The 5-year rates of freedom from treatment failure were 82% for those given the boost dose and 70% for those given 40 Gy without the boost (P = .026). The 5-year overall actuarial survival rates were 97% for the boost group and 93% for the no-boost group (P = .067). No true in-field failures were found within the extendedfield volume in either group, and the authors of the study concluded that 30 Gy was sufficient to control subclinical disease.

Radiation Toxicity Late complications of mantle irradiation include lung, heart, and thyroid dysfunction; second malignancies; growth abnormalities; and Lhermitte’s sign. However, in at least one study,95 total radiation doses of less than 30.6 Gy, given in 17 fractions, produced a low incidence of late complications. Boivin and colleagues96 suggested that mediastinal ­irradiation modestly increases the age-adjusted risk of death from myocardial infarction. At an average follow-up time of 9.5 years, Hancock and colleagues97 reported that mediastinal irradiation to a total dose of less than 30 Gy given in daily doses of 1.5 to 2.75 Gy did not increase the risk of cardiac death but that mediastinal doses of more than 30 Gy increased the relative risk of cardiac death to 3.5. Subcarinal blocking after 30 to 35 Gy reduces the relative risk for cardiac diseases such as congestive heart failure and pancarditis from 5.3 to 1.4 but does not significantly reduce the risk of acute myocardial infarction because the proximal coronary arteries still receive the full dose of ­radiation. The increased risk of second malignancies from previous irradiation has been well documented.74,98 The main longterm hazard associated with radiation therapy is the risk of forming solid tumors. The incidence of breast cancer is increased among women who were younger than 30 years at the time of mantle irradiation; the younger the patient and the greater the radiation dose, the higher the relative risk.99-101 The cumulative risk of developing a solid tumor such as osteosarcoma or carcinoma of the breast, thyroid, lung, stomach or colon is approximately 0.3% to 0.5% per year102,103 and increases with time, particularly 10 years or more after the irradiation. Irradiated patients also have an increased risk of developing non-Hodgkin lymphoma102 but not acute nonlymphocytic leukemia, unlike those who undergo chemotherapy with alkylating agents.98 Children who undergo radiation therapy can experience growth ­abnormalities.

34  Leukemias and Lymphomas

883

Staging Laparotomy

Radiation versus Chemotherapy

Other evaluations of prognostic factors for early-stage Hodgkin disease have led to investigations of the value of histologic findings obtained from laparotomy. The EORTC H6F trial, conducted from 1982 to 1988, included 262 patients with favorable clinical stage I or II disease who were randomly assigned to undergo staging laparotomy (with splenectomy) or no laparotomy.104 Patients who did not have laparotomy (i.e., those who underwent clinical staging) were treated with subtotal nodal irradiation. Patients assigned to the laparotomy group who showed no evidence of disease on laparotomy were treated with mantle radiation therapy (if the histology was lymphocyte predominant or nodular sclerosis) or with mantle and para-aortic radiation therapy (if the histology was mixed cellularity or lymphocyte depleted). If the laparotomy findings were positive, the patients were treated with chemotherapy plus radiation therapy as follows: six cycles of mechlorethamine, vincristine, procarbazine, and prednisone (MOPP regimen) or doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD regimen), with mantle radiation therapy given between the third and fourth cycles. At a median followup of 64 months, the 6-year overall survival rate was 93% for the no-laparotomy group and 89% for the laparotomy group (P = .24); the corresponding 6-year freedom from progression rates were 78% and 83% (P = .27). All three treatment-related deaths were in the laparotomy group. Even though the treatment outcomes were equivalent, the recommendation from this study was to abandon laparotomy because of the mortality associated with laparotomy and splenectomy. The University of Texas M. D. Anderson Cancer Center reported results from a retrospective analysis of 145 patients with pathologically confirmed stage I or II Hodgkin disease who were treated with mantle radiation therapy alone.105 Patients were treated between 1967 and 1991, and the median follow-up period was 12.2 years. Four factors (male sex, age of 40 years or older, B symptoms, and disease at least three sites) were found to be adverse prognostic factors for progression-free survival. Of the 145 patients studied, 54 had no adverse prognostic factors, 70 had one, and 21 had two adverse prognostic factors. Lower numbers of adverse prognostic factors were associated with better overall survival (P = .006) and progression-free survival (P < .0001). Progression-free survival rates were 96% in the absence of any adverse prognostic factors and 57% to 63% in the presence of one or two of these factors. As a general rule, prognostic factors obtained from retrospective analyses should be verified by prospective trials, as is being done at the Harvard Joint Center for Radiation Therapy.106 The eligibility criteria for this trial of mantle radiation therapy alone are having pathologic stage IA or IIA, nonbulky upper mediastinal disease (tumor not more than one third of the thoracic diameter) of lymphocytepredominant or nodular-sclerosis histology (pathologic stage IA disease can be of mixed cellularity). In this trial, laparotomy is not required if the patient presents with stage IA lymphocyte-predominant Hodgkin disease. Preliminary findings from this trial suggest that in selected patients with early-stage disease, mantle irradiation alone can produce disease control rates comparable to those after extended-field irradiation.107

In a study conducted in Italy, Biti and colleagues108 evaluated outcome for 89 patients with pathologic stage I or IIA Hodgkin disease who were randomly assigned to undergo subtotal nodal irradiation or six cycles of MOPP. At a median follow-up of 8 years, the overall survival rate was higher after radiation than after chemotherapy (93% versus 56%, P < .001). Rates of freedom from progression were similar for the two groups (76% versus 64%). Moreover, azoospermia was documented in 100% of the men in the chemotherapy group but 0% of those in the radiation group; amenorrhea was documented in 50% of the women in the chemotherapy group and in 10% of those in the radiation group. The investigators contended that radiation therapy alone should remain the treatment of choice for early-stage Hodgkin disease with favorable prognostic factors until more effective and less toxic chemotherapy ­becomes available. The U.S. National Cancer Institute published a report on 106 patients with pathologically confirmed stage IA, IB, IIA, IIB, or IIIA1 disease who were randomly assigned to receive subtotal or total nodal irradiation or at least six cycles of MOPP (or two additional cycles after a complete response).109 At a median follow-up time of 7.5 years, the projected 10-year disease-free survival rate was 60% for those treated with radiation and 86% for those treated with MOPP (P = .009). The projected 10-year overall survival rate was 76% for the radiation therapy and 92% for the MOPP group (P = .051). However, these intergroup differences in disease-free and overall survival rates disappeared when patients with bulky mediastinal disease and stage IIIA1 disease were excluded from the analysis. The conclusion was that MOPP was as effective as radiation in the treatment of selected patients with early-stage Hodgkin disease. Survival rates and complications over the long term remain to be seen. The question of which chemotherapy agents are most effective in the treatment of early-stage Hodgkin disease remains unanswered. Ongoing trials to address this question include the EORTC H9F trial, in which six cycles of EBVP are followed by randomization among no radiation therapy, 20 Gy, or 36 Gy.78 The HD10 trial, conducted by the German Hodgkin’s Lymphoma Study Group, involves two or four cycles of ABVD followed by randomization ­between 20 and 30 Gy of involved-field radiation therapy.110 ­Another trial of the Stanford V regimen (i.e., doxorubicin, vinblastine, mechlorethamine, vincristine, bleomycin, etoposide, and prednisone with granulocyte colony-­stimulating factor) with involved-field radiation111 suggested that this approach was quite effective for locally extensive or advanced Hodgkin disease, but that further tests were needed to see if the use of Stanford V, with or without radiation therapy, represented a therapeutic advantage over ABVD.

Chemotherapy Toxicity Although current trials are focusing on reducing radiation dose and volume in an attempt to reduce radiation-induced toxicity, chemotherapy also has toxic effects. These effects (including those not yet known because of the relatively short follow-up from chemotherapy trials) should be kept in mind when chemotherapy is ­recommended for patients with favorable stage I or II Hodgkin disease.

884

PART 11  Leukemias and Lymphomas

Of interest in this regard are the findings from two large studies comparing radiation therapy and chemotherapy for the treatment of Hodgkin disease. The first of these studies was a review of 1013 patients with Hodgkin disease treated at M. D. Anderson Cancer Center between 1966 and 1987.112 The median follow-up period was more than 5 years (except for patients who were treated with cyclophosphamide, vincristine, procarbazine, and prednisolone alternated with doxorubicin, bleomycin, dacarbazine, lomustine, and prednisone, for whom it was 3 years). At 20 years, the overall and disease-specific survival rates for the entire population were 60% and 76%. The relative hazard ratios for developing acute myelogenous leukemia, nonHodgkin lymphoma, or solid tumors were higher for those treated with chemotherapy alone than for those treated with radiation therapy with or without chemotherapy. This phenomenon was thought to result from the fact that patients who received chemotherapy as part of the combinedmodality treatment received less chemotherapy than did patients treated with chemotherapy only. The other study, conducted at the Harvard Joint Center for Radiation Therapy, retrospectively reviewed 794 patients with pathologically confirmed stages IA and IIIB Hodgkin disease treated between 1969 and 1988.73 In that study, 489 patients were treated with radiation therapy only; 158 were treated with radiation therapy and chemotherapy; and the other 147 patients had been treated initially with radiation, experienced relapse, and were then treated with chemotherapy. The 20-year actuarial survival rate was 73%. However, the absolute excess mortality risk from second malignancy and causes other than Hodgkin disease was higher for those treated with combined-modality therapy than for those treated with radiation only. Although the death rate from Hodgkin disease declined over time, the risk of death from other causes such as second malignancy and cardiac disease continued to rise. It was projected that the mortality from the second malignancy would surpass that from Hodgkin disease by 15 to 20 years after treatment. The HD10 trial of the German Hodgkin’s Lymphoma Study Group is testing the effectiveness of two or four cycles of ABVD, which has proved effective in advanced disease given with lower-dose, lower-volume radiation therapy (20 or 30 Gy to involved fields) for patients with favorable stage I or II Hodgkin disease.110 The hope is that this approach will be effective with fewer of the side effects associated with other chemotherapy or radiation therapy regimens. In summary, subtotal nodal irradiation remains the standard treatment for good-prognosis, early-stage Hodgkin disease; it has the longest track record. Carefully selected patients with pathologically staged disease may respond well to mantle irradiation alone. However, the morbidity and mortality associated with laparotomy and splenectomy must be kept in mind. Results from chemotherapy trials, with or without involved-field radiation therapy, are promising, but long-term follow-up is needed, especially with regard to second malignancies and salvage rates.

Treatment of Intermediate-Stage Disease: Unfavorable Stages I and II and Less-Advanced Stage III Intermediate-stage disease seems to respond to combination chemotherapy and radiation therapy, although the optimal chemotherapy agents, number of cycles, field sizes,

and radiation dose have yet to be identified. Studies that address these and related questions are summarized in this section. In its H7U trial, the EORTC described the outcome for 389 patients with unfavorable disease (i.e., patients 50 years old or older or with an ESR of 50 or more, B symptoms, disease at more than three sites, or a mediastinum-to-­thoracic ratio of more than 35%) who were randomized to receive six cycles of EBVP followed by involved-field radiation therapy or six cycles of alternating MOPP and ABV followed by involved-field radiation therapy.72 At 6 years, the RFS rates were 94% for the MOPP-ABV group and 79% for the EBVP group, and corresponding overall survival rates were 89% and 82%. Given these rather disappointing findings, EBVP has not been used in subsequent trials involving patients with unfavorable disease. Instead, the subsequent H8U trial randomly assigned patients with unfavorable disease into one of three groups. The first group was given six alternating cycles of MOPP-ABV followed by involved-field radiation therapy to 36 Gy, the second was given four cycles of MOPP-ABV followed by involved-field radiation therapy to 36 Gy, and the third was given four cycles of MOPP-ABV followed by subtotal nodal irradiation to 36 Gy.78 The findings from H8U have been published only in abstract form.113 In the current H9U trial, patients with unfavorable disease are randomly assigned to receive six cycles of ABVD, four cycles of ABVD, or four cycles of bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone (BEACOPP regimen); all groups receive involved-field radiation therapy to 30 Gy after the chemotherapy.78,113 Preliminary findings are available from an Italian study of 114 patients randomly assigned to undergo four cycles of ABVD followed by involved-field radiation therapy to 36 Gy or subtotal nodal irradiation to 30.6 Gy with a boost to the involved sites.114 Eligible patients could have clinical stage I with bulky disease or stage IB, IIA, or IIAE disease. At 12 years, the complete response rates were 97% for those given involved-field irradiation and 100% for those given subtotal nodal irradiation. Corresponding freedom from progression rates were 94% (95% confidence interval [CI], 88% to 100%) and 93% (95% CI, 83% to 100%), and overall survival rates were 94% (95% CI, 89% to 100%) and 96% (95% CI, 91% to 100%). Given the lack of statistical difference in these rates between the two groups, involvedfield irradiation after four cycles of AVBD was thought to be effective and safe for early Hodgkin disease with favorable or unfavorable features. However, the authors of the study reported that the statistical power of this study was insufficient to establish the superiority of involved-field or subtotal nodal irradiation in this context.114 The German Hodgkin’s Lymphoma Study Group HD8 trial115 also compared the outcome for patients at intermediate risk who received two cycles of cyclophosphamide, vincristine, procarbazine, and prednisone (COPP regimen) with two cycles of ABVD followed by subtotal nodal irradiation (30 Gy followed by a 10-Gy boost) or involvedfield radiation therapy (30 Gy followed by a 10-Gy boost). Eligible patients had stage I or II disease with at least one of the following risk factors: bulky mediastinal or extranodal disease, massive splenic involvement, an ESR of 50 or more (or an ESR of 30 or more in combination with B symptoms), or lymph node involvement in at least three sites, or stage IIIA disease in the absence of any of these risk factors.

34  Leukemias and Lymphomas

A total of 1204 patients were randomized. At 5 years, no difference was apparent between the groups in RFS rates (85.8% for subtotal nodal irradiation versus 84.2% for involved-field irradiation) or overall survival rates (90.8% for subtotal nodal irradiation versus 92.4% for involved-field irradiation). The authors of the study concluded that involved-field irradiation is sufficient after this chemotherapy regimen.115 However, they also observed that patients given subtotal nodal irradiation experienced more acute toxicity and a tendency for more severe late effects.115 The subsequent German Hodgkin’s Lymphoma Study Group HD9 trial involved a comparison of involved-field radiation therapy given after one of three chemotherapy regimens. The first group was given four cycles of COPP and four cycles of ABVD followed by involved-field radiation therapy of 30 Gy to initially bulky disease sites (≥5 cm) and 40 Gy to the area of residual disease; the second group was given eight cycles of baseline BEACOPP followed by the same radiation therapy; and the third group was given escalated-dose BEACOPP with granulocyte colony-­stimulating factor support followed by the same radiation therapy.116,117 Eligible patients were 16 to 65 years old, with stage IIB disease with bulky mediastinal disease, extranodal involvement, or massive splenic disease; stage IIIA disease with any of the previously described risk factors, an ESR of more than 50, or at least three areas of lymph node involvement; had stage IIIB disease; or stage IV disease. This study accrued 1282 patients between 1993 and 1998. At the time of the final data analysis in August 2001, the 5-year freedom from treatment failure rates were 69% for the COPP-ABVD group, 76% for the BEACOPP group, and 87% for the escalated-dose BEACOPP group; corresponding survival rates were 83%, 88%, and 91% (P = .002 for COPP-ABVD versus escalated-dose BEACOPP).117 Freedom from treatment failure was significantly better (P < .001) in the escalated-dose BEACOPP group than in either of the other two groups in this analysis.117 The COPP-ABVD treatment was dropped from the trial after the first interim analysis, which was completed in September 1996. An alternative chemotherapy plus radiation therapy protocol being tested for bulky or advanced Hodgkin disease is the Stanford V program.111,118 The treatment protocol involves 12 weeks of intensive chemotherapy followed by 36 Gy of involved-field radiation therapy to sites of bulky (≥5 cm) disease and to macroscopic splenic disease. For the 121 patients accrued, the median age was 27 years, and the median follow-up was 4.5 years. The 5-year survival rate for the entire group was 96%, and the failure-free survival rate was 89%. The failure-free survival rate for the 37 patients with stage I or II disease was 100%, and that for the 46 patients with stage III disease and the 38 patients with stage IV disease was 85%. Encouraging results from this trial led to initiation of trial 2496 by the Eastern Cooperative Oncology Group (ECOG) and Southwest Oncology Group (SWOG), in which the Stanford V program will be compared with ABVD chemotherapy.118

Treatment of Advanced Disease: Stage IIB, III, or IV Treatment with chemotherapy or combined chemotherapy and radiation has been less successful in controlling Hodgkin disease that manifests at advanced stages than it

885

is for early-stage disease. Multiagent chemotherapy seems more effective than single-agent chemotherapy or radiation alone, but the most effective and least toxic of these regimens has yet to be identified. Findings regarding the efficacy of treatments for advanced disease are summarized in the remainder of this section. One of the earliest large-scale randomized studies comparing the standard MOPP combination with the non– cross-resistant ABVD combination for Hodgkin disease was performed at the Instituto Nazionale Tumori in ­Milan, where the ABVD combination was initially conceived. ­Patients with stage IIB, IIIA, or IIIB disease (verified by laparotomy or laparoscopy) were randomly assigned to receive three cycles of MOPP followed by radiation therapy followed by three more cycles of MOPP or to receive three cycles of ABVD followed by radiation therapy followed by three more cycles of ABVD.119 Subtotal or total nodal irradiation was used according to whether the para-aortic nodes were involved. Patients were accrued to the study between 1974 and 1982. A total of 232 patients were randomized, and the groups were stratified by tumor histology, laparotomy versus laparoscopy, and disease stage. At 7 years, the ABVD treatment had produced superior rates of freedom from progression (81% versus 63%, P < .002), RFS (88% versus 77%, P = .06), and overall survival (77% versus 68%, P = .03). Another trial addressing the question of which chemotherapy regimen is the most effective in advanced Hodgkin disease was a Cancer and Leukemia Group B (CALGB) study120 in which patients with stage IIIA2, IIIB, IVA, or IVB disease (including those in first relapse after previous radiation therapy) were randomized to undergo six to eight cycles of MOPP or ABVD or 12 cycles of alternating MOPP and ABVD. No radiation therapy was given in this study. Laparotomy was performed only if findings on lymphangiography and abdominal CT scanning were equivocal or negative. At 5 years, the failure-free survival rates were 50% for the MOPP-only group, 61% for the ABVD-only group, and 65% for the alternating MOPP and ABVD group; corresponding overall survival rates were 66%, 73%, and 75%. Both the ABVD and the MOPP-ABVD treatments produced higher failure-free survival rates than did MOPP alone (P = .02), but no differences were found among groups in overall survival rate. The investigators concluded that ABVD alone was as effective as MOPP-ABVD and that both were superior to MOPP alone; moreover, ABVD alone was less myelotoxic than the MOPP regimens. Another modification of the MOPP-ABVD combination for advanced disease was tested in an Intergroup trial in which patients were randomly assigned to undergo ­ sequential MOPP-ABVD chemotherapy or a hybrid MOPP-ABV regimen.121 Eligibility criteria were the same as those for the CALGB trial described previously (i.e., stage IIIA2, IIIB, IVA, or IVB disease or first relapse after radiation therapy). Patients were stratified by age, previous radiation therapy, bulky disease, and performance status. Those assigned to the sequential group were treated with six cycles of MOPP; if complete response was achieved, they were given three cycles of ABVD. If the first six cycles of MOPP produced a partial response, two more cycles of MOPP were given. If complete response or stable disease was attained at that point, patients were then given three cycles of ABVD. Patients assigned to the hybrid regimen

886

PART 11  Leukemias and Lymphomas

group were given six cycles of the hybrid chemotherapy. If those six cycles produced a complete response, patients were given two more cycles of hybrid chemotherapy; if those six cycles produced a partial response, hybrid chemotherapy was repeated until two cycles beyond complete response (to a total of 12 cycles). No radiation therapy was given as a part of this trial. Patients were accrued between January 1987 and July 1989. A total of 691 patients were randomized and analyzed at a median follow-up of 7.3 years. The 8-year failure-free survival rate was 64% for the hybrid chemotherapy group and 54% for the sequential chemotherapy group (P = .01). The corresponding 8-year overall survival rates were 79% and 71% (P = .02). Although fewer incidents of acute leukemia and myelodysplasia were documented in the hybrid group than in the sequential group (P = .01), those receiving the hybrid chemotherapy developed more grade 4 or 5 leukopenia (P < .001), neutropenia (P < .001), and pulmonary toxicity (P = .004) than those in the sequential group. Another large-scale comparison of ABVD with a MOPPABV hybrid regimen for advanced Hodgkin disease was the CALGB 8952 trial, which involved the SWOG, the ECOG, and the National Cancer Institute of Canada.122 A total of 856 patients with a stage III or IV or recurrent disease after radiation therapy alone were accrued and stratified by age, stage, and previous radiation therapy. Chemotherapy was given for up to 8 to 10 cycles for those who responded. The complete response rate was 76% for the ABVD group and 80% for the MOPP-ABV hybrid group (P = .4); at 5 years, the corresponding failure-free survival rates were 63% and 66% (P = .3), and the overall survival rates were 82% and 81% (P = .8). However, the study was closed in November 1995 because of excessive numbers of treatment-related deaths and second malignancies among patients in the hybrid group (15 deaths and 18 second malignancies in the hybrid group versus 9 deaths [P = .13] and 28 second malignancies [P = .006] in the ABVD group).

Radiation Therapy after Complete Response to Chemotherapy A SWOG study, published in 1994, explored whether the addition of low-dose radiation after chemotherapy could prevent relapse and improve survival among patients with stage III or IV Hodgkin disease.123 Patients with clinical or pathologic stage III or IV disease were given six cycles of mechlorethamine, vincristine, prednisone, bleomycin, doxorubicin, and procarbazine (MOP-BAP regimen). ­Patients who achieved a complete response were ­randomly assigned to undergo low-dose, involved-field ­ radiation therapy or no further treatment. Radiation doses (all delivered only to initially involved sites) were 20 Gy (in 1.5-Gy fractions) to nodes, 15 Gy (in 1.5-Gy fractions) to liver, 15 Gy (in 1.25-Gy fractions) to spleen, and 10 Gy (in 1-Gy fractions) to lungs. Between 1978 and 1988, 530 eligible patients underwent six cycles of MOP-BAP, and 322 achieved a complete response; among these patients, 278 agreed to participate in the randomization, and 135 were assigned to receive radiation and 143 to receive no further treatment. However, only 104 of the 135 patients in the radiation therapy group actually underwent irradiation, and 130 of the 143 in the no-treatment group actually received no further treatment. Intent-to-treat analysis of these 278 patients showed the 5-year remission-duration estimate to

be 79% for the radiation therapy group and 68% for the no-treatment group (P = .09). However, analysis of the 234 patients who actually received the randomized treatment showed the 5-year remission-duration rates to be 85% for the radiation therapy group and 67% for the no­treatment group (P = .002). Limiting the analysis to the 169 patients with nodular-sclerosis histology revealed 5-year remission-duration rates of 82% for the radiation therapy group and 60% for the no-treatment group (P = .002). Further limiting the analysis to the 74 patients with nodular-sclerosis histology and bulky disease made a still larger difference in 5-year remission-duration rates: 76% (radiation therapy group) and 46% (no-treatment group) (P = .006); analysis of the 95 patients with nodular-sclerosis histology and nonbulky disease revealed 5-year remission-duration rates of 88% (radiation therapy) and 68% (no treatment) (P = .06). Bulky disease was defined in this trial as any mass 6 cm in diameter or larger. Shortcomings of this study ­ included a change in study design after early problems with accrual and the fact that 16% of the patients did not receive the assigned treatment. Nevertheless, this trial demonstrated a significant benefit for low-dose, involvedfield radiation therapy, at least for some subsets of patients. The HD3 trial of the German Hodgkin’s Lymphoma Study Group also studied the effectiveness of adjuvant radiation or chemotherapy after a complete response to chemotherapy.124 Patients with clinical or pathologic stage IIIB or IV disease were treated with three cycles of COPP and ABVD chemotherapy. Those who achieved complete response were randomly assigned to undergo involved-field radiation therapy (20 Gy to initially involved areas) or one more cycle of COPP-ABVD. Between 1984 and 1988, 288 patients were accrued, and 171 achieved a complete response after three cycles of COPP-ABVD. Only 100 of these patients were randomized, 51 to the radiation-­therapy group and 49 to the additional-chemotherapy group. The two groups did not have different rates of freedom from treatment failure or overall survival. The relative risk of failure for 21 patients who refused further treatment after experiencing a complete response to three chemotherapy cycles was 3.67 compared with patients given consolidation treatment. Although the number of patients studied was small, these findings underscore the importance of consolidation treatment. Whether more than six cycles of chemotherapy can eliminate the need for consolidation therapy remains to be seen. A similar study was conducted by the Groupe d’Études des Lymphomes de l’Adulte (GELA).125 In their H89 trial, 418 patients with stage IIIB or IV Hodgkin lymphoma who had achieved a complete or partial response to six cycles of doxorubicin, bleomycin, vinblastine, procarbazine, and prednisone (ABVPP regimen) or a MOPP-ABV hybrid regimen were randomly assigned to undergo two additional cycles of chemotherapy or extended-field radiation therapy to 30 to 40 Gy. The radiation fields were subtotal nodal (i.e., mantle, para-aortic, and spleen). However, the inverted Y with spleen fields was used if the iliac or inguinal nodes were involved. Although consolidation treatment with radiation seemed to confer a slight advantage over consolidation with chemotherapy in terms of disease-free survival (P = .07), no difference was found in overall survival rate between those given radiation or chemotherapy consolidation. However, because the radiation fields were

34  Leukemias and Lymphomas

substantially larger than the involved fields in this study, the putative increase in toxicity might have negated any benefit achieved by radiation in terms of an improvement in disease-free survival.

Radiation Therapy after Partial Response to Chemotherapy The EORTC and the Groupe Pierre-et-Marie-Curie performed a randomized trial comparing involved-field radiation therapy and no radiation therapy for patients who achieved a complete response to six cycles of MOPPABV hybrid chemotherapy for stage III or IV Hodgkin disease.126-128 In that study of 739 enrolled patients, 420 (67%) had experienced a complete response and 247 (33%) a partial response to the induction chemotherapy. Patients who achieved complete responses were randomly assigned to undergo involved-field radiation or no radiation; the results from that aspect of the study, published in 2003,127 indicated no difference between the two groups in 5-year overall survival rates (85% for those who received radiation versus 91% for those who did not; P = .07) or event-free survival rates (79% for the radiation group versus 84% for the no-radiation group; P = .35). However, findings from those who achieved partial response to induction chemotherapy, who were then given involved-field radiation therapy (30 Gy, with a boost of 4 to 10 Gy as needed, in 1.5- to 2-Gy fractions), are encouraging. At a median follow-up time of 7.8 years, the 8-year progression-free and overall survival rates for the 227 partial responders were 76% and 84%, respectively, rates similar to those for patients who achieved complete response to the chemotherapy.128 These results strongly support the role of radiation therapy for patients who achieve a partial response to induction chemotherapy because the prognosis for those patients seems to be as good as the prognosis for those who achieved a complete response to induction ­chemotherapy.127,128

Predictors of Outcome The International Prognostic Factors Project sought to identify prognostic factors (in addition to Ann Arbor stage) for advanced Hodgkin lymphoma.129 Analyzed data were pooled from 5141 patients from 25 centers in Europe and North America. Eligible patients had to be between 15 and 65 years old with advanced Hodgkin disease that had been treated with at least four planned cycles of “state-of-theart” combination chemotherapy, with or without radiation therapy. The definition of advanced disease was left to the contributors. Of the 4695 patients analyzed, more than 75% had been treated with doxorubicin-based regimens, 20% had been treated with MOPP or a MOPP-like regimen, and 40% had undergone radiation therapy. Another requirement was that treatment must have begun before January 1, 1992, to ensure sufficiently long follow-up. The variables at diagnosis examined in relation to the end point (freedom from disease progression) were age; sex; tumor histology; Ann Arbor stage; presence of B symptoms; extent of mediastinal involvement; presence of inguinal node, lung, liver, and bone marrow involvement; hemoglobin level; serum albumin level; ESR; white blood cell and platelet counts; absolute and relative lymphocyte counts; and alkaline phosphatase, lactate dehydrogenase, and creatinine levels. Seven of these factors were identified in a Cox regression model as having independent ­adverse

887

significance: serum albumin level of less than 4 g/dL, ­hemoglobin level of less than 10.5 g/dL, male sex, age of 45 years or older, stage IV disease, white blood cell count of 15,000/mm3 or more, and lymphocyte count of less than 600/mm3 or less than 8% of the number of white blood cells. The prognostic score was defined as the number of these adverse variables present at diagnosis. That score predicted the rate of freedom from progression as follows: 0 or no factors (7% of the patients): 84% 1 factor (22% of the patients): 77% 2 factors (29% of the patients): 67% 3 factors (23% of the patients): 60% 4 factors (12% of the patients): 51% 5 or more factors (7% of the patients): 42% The rates of freedom from progression seemed to reach a plateau after 4 years, regardless of prognostic score. ­Unfortunately, this scoring technique could not identify a subgroup of patients at very high risk for progression. In summary, for the treatment of intermediate- and advanced-stage Hodgkin disease, consolidation radiation therapy probably improves at least RFS after a complete response to chemotherapy. Radiation therapy is recommended after a partial response in most situations. The balance between benefit and risk probably varies with the number of involved sites and the bulkiness of the disease.

Subdiaphragmatic Disease About 5% to 12% of the patients with stage I or II Hodgkin disease present with disease that appears only below the diaphragm. Compared with supradiaphragmatic disease, subdiaphragmatic Hodgkin disease is more likely to appear in men than in women and at older rather than younger ages, is less likely to be of nodular-sclerosis histology, and is more likely to be of lymphocyte-predominant histology. Given the rarity of this condition, no prospective randomized trials have been conducted, and most of the available information comes from retrospective reports from individual institutions. Reported adverse prognostic factors are central disease (versus peripheral disease),130,131 older age, B symptoms,132 and low albumin and hemoglobin levels.133 Laparotomy is no longer thought to be necessary given the ability to visualize abdominopelvic disease on CT scans, the morbidity associated with splenectomy, and satisfactory treatment outcome without laparotomy.134 Radiation therapy administered in inverted Y and spleen fields seems to be adequate when the disease is limited to the inguinofemoral areas. For more centrally located disease (i.e., in the para-aortic or iliac node regions), total nodal irradiation or combined-modality therapy is recommended because of a high incidence of recurrence above the diaphragm. Combined-modality therapy usually is recommended for patients with stage IIB disease; whether those with stage IB disease can also benefit from chemotherapy is less clear. Long-term follow-up of patients with subdiaphragmatic Hodgkin disease shows overall and RFS rates similar to those that would be expected in supradiaphragmatic Hodgkin disease.133

Lymphocyte-Predominant Hodgkin Lymphoma Whether lymphocyte-predominant Hodgkin disease behaves any differently from classic Hodgkin disease remains a matter of debate.

888

PART 11  Leukemias and Lymphomas

Survival and Relapse-Free Rates In 1988, a group from Stanford published a retrospective analysis of 73 patients with lymphocyte-predominant Hodgkin disease treated between 1963 and 1985.134 In this group, 41 patients had the diffuse subtype, and 32 had the nodular subtype. The median follow-up was 9.4 years (range, 0.3 to 23.5 years). Among those with the nodular subtype, RFS rates were 83% at 5 years and 55% at 10 years; the 10-year RFS rate for patients with the diffuse subtype was 92% (P = .017). The median time to relapse was 5.2 years for those with the nodular subtype and 3.7 years for those with the diffuse subtype. Overall survival rates were not different for the two subtypes. In 1999, investigators at M.  D. Anderson published a retrospective analysis of 70 patients with lymphocytepredominant Hodgkin disease treated between 1960 and 1992.135 Of those 70 patients, 58 had the nodular subtype, and 12 had the diffuse subtype; cases with any discernable nodularity were classified as nodular. The median age of the patients was 25 years. Median follow-up for living patients was 12.3 years (range, 3 to 37 years). The overall survival rates for the entire group were 97% at 5 years, 84% at 10 years, and 80% at 15 years; the corresponding RFS rates were 84%, 71%, and 71%. No differences in RFS or overall survival rates were found between patients with the nodular or diffuse subtypes. In 1999, the European Task Force on Lymphoma reported an analysis of 219 patients with lymphocyte­predominant Hodgkin disease pooled from 17 European and American centers.136 The median observation time was 6.8 years. Survival and failure-free survival rates were not better among these patients than were stage-­comparable results for classic Hodgkin disease from the German Hodgkin’s Lymphoma Study Group trials. Although the actual data were not presented in the Task Force study, no significant differences in patient characteristics or survival results reportedly were observed among the nodular, diffuse, or nodular plus diffuse cases of lymphocyte-predominant Hodgkin disease.136 Significantly, 27% of the patients with lymphocyte-predominant disease who experienced relapse had multiple relapses, which raises the question of how best to minimize the treatment for lymphocyte-predominant disease in light of good salvage rates and good survival rates after relapse. Moreover, diffuse lymphocyte-predominant Hodgkin disease seems to be disappearing as a diagnostic entity as more cases are recognized as being lymphocyterich classic Hodgkin disease.55

Secondary Malignancies Another finding from the European Task Force study was a 2.9% incidence of non-Hodgkin lymphoma among patients with lymphocyte-predominant Hodgkin disease. This finding raises the question of whether patients with this form of the disease are more susceptible to developing nonHodgkin lymphoma than are patients with classic Hodgkin disease.136 The answer is not straightforward. Analysis of the international database on Hodgkin disease shows that the risk of developing non-Hodgkin lymphoma after treatment for Hodgkin disease continues to increase over a 20-year period, from about 1% at 10 years to a 4% cumulative incidence at 20 years.75 Bodis and colleagues,137 in reviewing 325 cases of lymphocyte-predominant Hodgkin

disease, found a 2.5% chance of developing secondary nonHodgkin lymphoma. Because lymphocyte-predominant Hodgkin disease sometimes coexists with large cell lymphoma even at presentation, it may seem to have a higher chance of transforming into non-Hodgkin lymphoma.138 Evaluations of the incidence of non-Hodgkin lymphoma in the presence of lymphocyte-predominant Hodgkin disease should focus on actuarial cumulative incidence rather than crude incidence rates.

Response to Chemotherapy In a study conducted by the Harvard Joint Center for Radiation Therapy, 16 patients with lymphocyte-predominant Hodgkin disease were given chemotherapy as initial treatment or at first relapse.137 Eight of 12 patients treated with MOPP or MOPP-like regimens achieved a “durable” complete response, compared with two of six patients treated with ABVD and etoposide, vinblastine, and doxorubicin (EVA) regimens, raising the question of whether MOPP was more effective than ABVD-EVA for this disease. Findings from a later M. D. Anderson review suggested that patients treated with radiation therapy alone had better RFS rates than patients treated with chemotherapy alone or combined chemoradiation therapy.135 However, data from both studies should be interpreted cautiously because both were retrospective analyses and selection bias was present in terms of who was treated with chemotherapy. In summary, no conclusive evidence exists to suggest that the lymphocyte-predominant form of Hodgkin disease behaves in a different manner from the classic form. Because the salvage rates for the lymphocyte-predominant form may be better than those for the classic form, even after multiple recurrences, some clinicians may advocate less-intensive treatment for patients with lymphocyte­predominant Hodgkin disease.

Radiation Therapy Techniques Mantle Field Irradiation In mantle field irradiation, the bilateral cervical, supraclavicular, infraclavicular, axillary, mediastinal, and hilar lymphatics are the target volumes (Fig. 34-5A). Simulations are usually performed with the patient supine with arms ­akimbo. This position has the advantage of easy reproducibility, although keeping the hands raised rather than at the waist would allow a bit more of the lung parenchyma to be spared through superolateral movement of the axillary lymphatics.139 The patient’s neck is slightly ­hyperextended so that the mouth block can also block the occipital head without compromising the coverage of the cervical ­lymphatics. Equally weighted 6-MV opposed photon fields typically are used for mantle radiation. Lung blocks are shaped superolaterally for adequate coverage of the axillary and infraclavicular lymphatics. Medially, the lung blocks should allow margins adequate to cover the mediastinum and ­bilateral hila. The inferolateral border of the lung blocks is determined by the lateral-most point of the sixth rib. The inferior border of the mantle fields is placed immediately above the highest point of excursion of the left hemidiaphragm on expiration, unless this placement compromises tumor coverage in the lower mediastinum. If the inferior border is lower than this, field matching with the para­aortic and spleen fields may take place in the spleen, which

34  Leukemias and Lymphomas

A

889

B

FIGURE 34-5.  A, A typical initial mantle irradiation field for a patient with Hodgkin disease to be treated with radiation only. B, Dosimetry for para-aortic and spleen fields for the patient in A.

is undesirable. The left ventricle of the heart is shielded throughout the treatment unless this shielding would compromise tumor coverage. After 25.2 Gy, the inferior field border is raised superiorly to about two vertebral bodies below the carina to shield additional heart volume, unless this movement would compromise tumor coverage.140 An anterior larynx block is placed over the true vocal cords after 19.8 Gy unless it compromises tumor coverage.

Para-aortic and Spleen Field ������������� Irradiation Expanding the radiation field to include the para-aortic nodes and spleen (see Fig. 34-5B) usually involves the use of equally weighted 18-MV opposed photon fields. The superior border of these fields is matched with the inferior border of the mantle fields. The inferior border is usually at the L4-L5 interspace, which would allow coverage of the entire para-aortic lymphatics. The margin left around the spleen should be sufficient to allow splenic excursion with breathing.

Involved-Field Irradiation The term involved fields usually implies treatment of the tumor volumes that were present before chemotherapy, with some additional margins. The exact extent of the fields depends on the location of the initial disease, the extent of residual disease after chemotherapy, and the exposure to adjacent organs such as kidneys, heart, and salivary glands. The advantages of higher dose and larger target volume must be balanced against the additional morbidity associated with this technique.

Non-Hodgkin Lymphoma Epidemiology and Etiology In contrast to Hodgkin lymphoma, the incidence and mortality rates for non-Hodgkin lymphoma have increased over the past 30 years.65,141 The American Cancer Society estimated that 66,120 people were diagnosed with and 19,160 died of non-Hodgkin lymphoma in the United States in 2008,58 making it the fifth most common type of cancer (except for cutaneous carcinomas) in the United States.

Non-Hodgkin lymphoma is a heterogeneous group of diseases characterized by malignant lymphocytes of B cell, T cell, and rarely, natural killer (NK) cell origin. The cause remains unknown, although Helicobacter pylori infection, ­viral infections, chemical and physical agents, congenital and acquired immunodeficiency states, and autoimmune disorders may play roles in the pathogenesis of non­Hodgkin lymphoma.142 H. pylori, a gram-negative bacillus, has been linked with mucosa-associated lymphoid tissue (MALT) lymphomas of the stomach.143,144 Epstein-Barr ­virus, a deoxyribonucleic acid virus in the herpesvirus ­family, was recognized in lymphoblasts from an African patient with Burkitt’s lymphoma in 1964,145 and it is present in more than 95% of endemic variants of Burkett’s lymphoma.146 ­ Patients with acquired immunodeficiency syndrome (AIDS) are at particular risk for Epstein-Barr virus–related B-cell lymphomas. The incidence of lymphoma among patients with immunosuppression, whether occurring as a result of disease or induced by drugs or antibodies for the purpose of organ transplantation, has been increasing as the survival time for such patients is becoming longer.147,148 Recipients of mismatched T-cell–depleted marrow stem cell grafts who are infected with the EpsteinBarr virus are also at increased risk for lymphoproliferative disease.149 Several chromosomal abnormalities have been identified in various subtypes of lymphoma. The most common of these abnormalities is the t(14;18)(q32;21) translocation, which appears in approximately 85% of follicular center cell lymphomas. This translocation activates the BCL2 oncogene, and the resultant increase in the expression of BCL2 protein is thought to extend the lifespan of centrocytes by inhibiting their apoptosis.150 Burkitt’s lymphoma has been associated with the t(8;14) translocation and consequent rearrangement of the MYC oncogene or, less commonly, with t(2;8)(p13;q23) or t(8;22)(p13;q24) translocations.151 In anaplastic large cell lymphoma, the t(2;5)(p23;q35) translocation is identified in about 50% of systemic cases and is somewhat less common in primary cutaneous cases.152 The t(11;14)(q13;q32) translocation, which leads to overexpression of the cell cycle control

890

PART 11  Leukemias and Lymphomas

protein cyclin D1, has been found in 70% of cases of mantle cell lymphoma.153-155

Pathology, Clinical Presentation, and Staging Many classification schemes have been used for non­Hodgkin lymphoma worldwide. The sheer number and evolution of those systems during the 20th century has greatly complicated comparisons between studies conducted at different institutions. Briefly, in the 1950s, Rappaport and colleagues recognized the importance of growth pattern in some types of non-Hodgkin lymphoma and used pattern, in addition to cell size and shape, as the basis of a new classification.156,157 In the 1970s, the recognition that nonHodgkin lymphomas were tumors of the immune system prompted the creation of the immunologically based classification systems of Lukes and Collins158,159 and Lennert and associates (the Kiel system).160-162 In an attempt to unify terminology and improve the effectiveness of communications between pathologists and clinicians, the National Cancer Institute proposed their Working Formulation in 1982.163,164 Over the next 2 decades, however, the Kiel classification came to dominate clinical practice in Europe, and the Working Formulation became the main classification system used in North America. In an attempt to include newly recognized lymphoid disease entities that were not part of the Working Formulation164 and to provide a common classification system that could be used internationally,165 the International Lymphoma Study Group in 1994 published a revised version of the Working Formulation that included additional information on the molecular-genetic characteristics of the lymphoid cells. The resulting REAL ­classification system55 is essentially a list of distinctive clinicopathologic entities that can be recognized by available techniques; the list can be and has been modified easily as new information becomes available. A comparison of the Working Formulation and REAL schemes is presented in Table 34-2.55,166 The latest classification system, the WHO classification,56,167 was formally published in 2001 and represents an update of the 1994 REAL classification (Table 34-3). The clinical relevance of the REAL classification has been validated in several studies. The largest of these studies, the Non-Hodgkin’s Lymphoma Classification Project,168 was a retrospective analysis of 1378 cases of lymphoma from nine institutions in eight countries. The incidence of the most common subtypes of lymphoma in this study was as follows: large B-cell lymphoma (31%), follicular lymphoma (22%), marginal zone B-cell MALT lymphoma (8%), peripheral T-cell lymphoma (7%), small B-cell lymphocytic lymphoma (also known as chronic lymphocytic leukemia) (7%), mantle cell lymphoma (6%), anaplastic large T/null cell lymphoma (2%), and primary mediastinal large B-cell lymphoma (2%); tumors of uncommon histology constituted the remainder of the non-Hodgkin lymphomas.168 One such tumor, monocytoid B-cell lymphoma, is particularly uncommon. It occurs in older individuals, involves lymph nodes rather than extranodal tissues, and is similar morphologically and immunologically to MALT lymphoma.169,170 Differences in overall survival according to REAL subtype are illustrated in Figure 34-6.168 Aspects of presentation and treatment of the more common subtypes

and some of the less common subtypes are described later (see “Subtypes of Non-Hodgkin Lymphoma”). Cell surface marker analysis (i.e., immunophenotyping) can help in resolving cases that are difficult to diagnose on the basis of morphology.171 B-cell lineage is suggested by the presence of CD19, CD20, and CD22, and T-cell lineage is suggested by the expression of CD2, CD3, and CD7. Follicular lymphomas are characteristically CD5− and CD43−, MALT lymphomas are CD5− and CD10−, mantle cell lymphomas are CD5+ and CD23−, and angiocentric T-cell/NK cell lymphomas are CD56+. Although cell surface marker analysis constitutes one piece of the diagnostic puzzle, few cell surface markers are lineage-specific.

Presenting Sites of Disease Non-Hodgkin lymphomas can arise in many different anatomic sites, including the following (in order of decreasing frequency): Lymph nodes Gastrointestinal tract Stomach Small bowel Colon and rectum Appendix Esophagus Waldeyer’s ring Nasopharynx Palatine tonsil Base of tongue Orbit Brain Thyroid Paranasal sinuses Testis Oral cavity Bone Breast Uterus Bladder Skin Lung Salivary glands Diseases arising at these sites have different etiologic ­factors, distinct natural histories, and dissimilar responses to treatment.168

Staging Evaluation The pretreatment evaluation for non-Hodgkin lymphoma is similar to that for Hodgkin lymphoma and should include physical examination, with attention to node-bearing areas, including Waldeyer’s ring, and the size of liver and spleen; assessment of B symptoms (i.e., fever, night sweats, weight loss) and performance status; complete blood cell counts, with platelets and differentials; ESR; lactate dehydrogenase levels; comprehensive metabolic panel; hepatitis B testing; bone marrow aspirate and biopsy; CT scans of the head and neck, chest, abdomen, and pelvis with adequate contrast; and nuclear imaging. Nuclear imaging has evolved over the past several decades from gallium 67 scanning to positron emission tomography (PET) scanning. Small, single-institution studies suggest that PET scanning is more sensitive than gallium scanning for detecting hepatic, splenic, and bone

34  Leukemias and Lymphomas

891

TABLE 34-2.   Comparisons of the NCI Working Formulation164 and the REAL Classifiction55 Definitions of Non-Hodgkin Lymphoma REAL Classification NCI Working Formulation

B-Cell Neoplasms

T-Cell Neoplasms

Small lymphocytic, consistent with CLL

B-cell CLL/PLL/SLL Marginal zone/MALT Mantle cell

T-cell CLL/PLL LGL ATL/L (chronic and smoldering types)

Small lymphocytic, plasmacytoid

Lymphoplasmacytic-immunocytoma Marginal zone/MALT

Follicular, predominantly small cleaved cell

Follicle center, follicular, grade I Mantle cell Marginal zone/MALT

Low Grade

Follicular, mixed small cleaved and large cell Follicle center, follicular, grade II Marginal zone/MALT Intermediate Grade Follicular, large cell

Follicle center, follicular, grade III

Diffuse, small cleaved cell

Mantle cell Follicle center, diffuse small cell Marginal zone/MALT

Angiocentric

T-cell CLL/PLL LGL ATL/L Angioimmunoblastic

Diffuse, mixed small and large cell

Large B-cell lymphoma (rich in T cells) Follicle center, diffuse small cell Lymphoplasmacytoid Marginal zone/MALT Mantle cell

Peripheral T-cell unspecified ATL/L Angioimmunoblastic Angiocentric Intestinal T-cell lymphoma

Diffuse, large cell

Diffuse large B-cell lymphoma

Peripheral T-cell lymphoma ATL/L Angioimmunoblastic Angiocentric Intestinal T-cell

Large-cell immunoblastic

Diffuse large B-cell lymphoma

Peripheral T-cell unspecified Angioimmunoblastic Angiocentric Intestinal T-cell Angioblastic large cell

Lymphoblastic

Precursor B-lymphoblastic

Precursor T-cell lymphoblastic

High Grade

Small noncleaved cell Burkitt’s

Burkitt’s

Non-Burkitt

High-grade B-cell, Burkitt-like diffuse large B-cell

ATL/L, adult T-cell lymphoma/leukemia; CLL, chronic lymphocytic leukemia; LGL, large granular lymphocyte leukemia; MALT, mucosa­associated lymphoid tissue; NCI, National Cancer Institute; PLL, prolymphocytic leukemia; REAL, Revised European-American Classification of Lymphoid Neoplasms; SLL, small lymphocytic lymphoma. From Shipp MA, Mauch PM, Harris NL. Non-Hodgkin’s lymphomas. In DeVita VT Jr, Hellman S, Rosenberg SA (eds): Cancer: Principles and Practice of Oncology, 5th ed. Philadelphia: Lippincott-Raven, 1997, pp 2165-2220.

­ arrow involvement172,173 and may be particularly usem ful for ­ revealing sites of bone marrow involvement that are not detected in iliac crest biopsies.174,175 According to one meta-analysis, the sensitivity of PET for detecting bone marrow involvement in non-Hodgkin lymphoma depended greatly on histologic subtype, being better for tumors with more aggressive histology than for less aggressive subtypes,176 perhaps because indolent disease is

less likely to involve the marrow or involves only 10% to 20% of it.174,177 PET may be superior to CT for detecting different lymphoma subtypes at various anatomic sites. Some studies suggest that 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) is more accurate than CT for ­detecting nodal and extranodal disease, which presumably would have a considerable impact on the staging of

892

PART 11  Leukemias and Lymphomas

TABLE 34-3.  Differences in Terminology between the Revised European-American and World Health Organization Classifications of Lymphoid Neoplasms REAL Classification (1994)

WHO Classification (2001)

Precursor Cell Lymphoma

Precursor Cell Lymphoma

Lymphoblastic lymphoma, T cell

Lymphoblastic lymphoma, T-cell

Lymphoblastic lymphoma, B cell

Lymphoblastic lymphoma, B-cell

Peripheral (Mature) B-Cell Neoplasms B-chronic lymphocytic leukemia/small lymphocytic lymphoma/prolymphocytic leukemia

B-cell chronic lymphocytic leukemia/small lymphocytic ­lymphoma B-cell prolymphocytic leukemia

Lymphoplasmacytoid lymphoma

Lymphoplasmacytic lymphoma

Mantle cell lymphoma

Mantle cell lymphoma

Follicular center cell lymphoma

Follicular lymphoma

Marginal-zone B-cell lymphoma Extranodal

Extranodal marginal zone B-cell lymphoma of MALT type

Nodal (provisional entity)

Nodal marginal zone B-cell lymphoma

Splenic marginal zone B-cell lymphoma (provisional entity)

Splenic marginal zone B-cell lymphoma

Hairy cell leukemia

Hairy cell leukemia

Diffuse large B-cell lymphoma

Diffuse large B-cell lymphoma

Burkitt’s lymphoma

Burkitt’s lymphoma (including Burkitt-like lymphoma)

High-grade B-cell lymphoma, Burkitt-like (provisional entity) Plasmacytoma/plasma cell myeloma

Plasmacytoma/plasma cell myeloma

Peripheral (Mature) T and NK Cell Neoplasms

Peripheral (Mature) T and K Cell Neoplasms

T-cell chronic lymphocytic leukemia/prolymphocytic leukemia

T-cell prolymphocytic leukemia

Large granular lymphocytic leukemia (T or NK cell)

T-cell granular lymphocytic leukemia Aggressive NK-cell leukemia

Mycosis fungoides/Sézary syndrome

Mycosis fungoides/Sézary syndrome

Peripheral T-cell lymphoma, unspecified

Peripheral T-cell lymphoma, not otherwise characterized

Angioimmunoblastic T-cell lymphoma

Angioimmunoblastic T-cell lymphoma

Angiocentric lymphoma

Extranodal NK/T-cell lymphoma, nasal and nasal type

Intestinal T-cell lymphoma

Enteropathy-type T-cell lymphoma

Hepatosplenic γδ T-cell lymphoma (provisional entity)

Hepatosplenic γδ T-cell lymphoma

Subcutaneous panniculitis-like T-cell lymphoma (provisional entity)

Subcutaneous panniculitis-like T-cell lymphoma

Anaplastic large-cell lymphoma, T/null cell

Anaplastic large-cell lymphoma, primary systemic type Anaplastic large-cell lymphoma, primary cutaneous type

Adult T-cell lymphoma/leukemia

(HTLV1+)

Adult T-cell lymphoma/leukemia (HTLV1+)

HTLV+, positive for human T-cell leukemia virus; MALT, mucosa-associated lymphoid tissue; NK, natural killer; REAL, Revised European-American Classification of Lymphoid Neoplasms; WHO, World Health Organization.

non-Hodgkin lymphomas. FDG-PET has revealed disease that would change the stage for 10% to 40% of patients, but for only about one half of them would the restaging (which is more often upward than downward) lead to changes in the treatment strategy or outcome.178 The sensitivity and specificity of PET seem to be better for aggressive ­lymphomas than for indolent lymphomas,179 but reported values vary considerably. For example, the sensitivity of FDG-PET for detecting follicular lymphoma can

be as high as 98%, but only 67% of marginal zone lymphomas and 40% of peripheral T-cell lymphomas showed avid FDG uptake.177,179 Other evidence exists to suggest that PET can help to distinguish low-grade from high-grade or transformed cases.180 While findings from prospective trials of the use of PET for staging lymphoma (or for monitoring response to treatment) are awaited, the National Comprehensive Cancer Network181 recommends in its 2008 guidelines that PET be used in pretreatment disease staging.

893

34  Leukemias and Lymphomas 100

100

90

90 70

MZ MALT

60

FL

50 40 30

Overall survival (%)

Overall survival (%)

80

ALCL

80

70 60 40 30

20

20

10

10 0

0 0

1

2

3

A

4

5

6

7

8

0

100

90

90

80

80

70 60 Med LBC

50

DLBC

Burkitt’s HG, BL

30

0 4

5

6

7

8

Years

T-LB

30

0 3

6

7

8

40

10 2

5

50

10 1

4

60

20

0

3

70

20

C

2

Years

100

40

1

B

Years

Overall survival (%)

Overall survival (%)

MZ nodal LP SL

50

PTCL MC

0

1

2

3

D

4

5

6

7

8

Years

FIGURE 34-6.  Overall survival rates for subtypes of non-Hodgkin lymphomas. A, Subtypes for which overall survival rates are more than 70% are anaplastic large T/null-cell lymphoma (ALCL), marginal-zone B-cell lymphoma of mucosa-associated lymphoid tissue (MZ MALT), and follicular lymphoma (FL). B, Subtypes for which overall survival rates range from 70% to 50% are marginal-zone B-cell lymphoma of nodal type (MZ nodal), lymphoplasmacytoid lymphoma (LP), and small lymphocytic lymphoma (SL). C, Subtypes for which overall survival rates range from 49% to 30% are primary mediastinal large B-cell lymphoma (Med LBC), diffuse large B-cell lymphoma (DLBC), high-grade B-cell Burkitt-like lymphoma (HG, BL), and Burkitt’s lymphoma. D, Subtypes for which overall survival rates are less than 30% are precursor T-cell lymphoblastic lymphoma (T-LB), peripheral T-cell lymphoma (PTCL), and mantle cell lymphoma (MC). (From A clinical evaluation of the International Lymphoma Study Group classification of non-Hodgkin lymphoma. Blood 1997;89:3909-3918.)

Ann Arbor Staging System Although originally developed for Hodgkin disease, the American Joint Committee on Cancer has adopted the Ann Arbor staging system for non-Hodgkin lymphoma to avoid confusion (see Box 34-1). In the Ann Arbor staging system, the thymus, appendix, Waldeyer’s ring, and Peyer’s patches are considered lymphatic sites. However, because of the clinical behavior of lymphomas at these sites, most oncologists refer to the aforementioned sites as extranodal. The Ann Arbor staging system is of limited use in predicting the outcome of patients with non-Hodgkin lymphoma,182,183 primarily because it fails to address tumor burden.184 It is more useful when combined with the serum lactate dehydrogenase level, performance status, presence of extranodal sites of disease, and age of the patient. These five factors, which independently predict overall survival, make up the age-adjusted international prognostic index proposed by the International Non-Hodgkin’s Lymphoma Prognostic Factors Project (Table 34-4).185

TABLE 34-4.   Age-Adjusted International Prognostic Index for Non-Hodgkin Lymphoma Prognostic Factor

Adverse Feature*

Lactate dehydrogenase level

>Normal

Ann Arbor stage

III or IV

Zubrod performance status

≤2

Extranodal Age *Each

involvement†

>1 site >60 years

adverse feature is assigned one point. The sum of the points is the international prognostic index (IPI). A total IPI value ≤ 2 indicates a good prognosis and an IPI value > 2 indicates a poor prognosis. †Extranodal involvement is considered only for patients older than 60 years. Modified from A predictive model for aggressive non-Hodgkin’s lymphoma: the International Non-Hodgkin’s Lymphoma Prognostic Factors Project. N Engl J Med 1993;329:987-994.

894

PART 11  Leukemias and Lymphomas

TABLE 34-5.    Zubrod Performance Status Scale Points

Description

0

Normal activity, asymptomatic

1

Symptomatic but fully ambulatory

2

Symptomatic and in bed 1% to 49% of the day

3

Symptomatic and in bed 50% to 99% of the day

4

100% bedridden

5

Dead

Modified from Stanley KE. Prognostic factors for survival in patients with inoperable lung cancer. J Natl Cancer Inst 1980;65:25-32.

Subtypes of Non-Hodgkin Lymphoma The WHO classification (see Table 34-3)56 represents the most recent attempt to categorize specific subtypes of ­lymphoma according to their morphologic, immunophenotypic, genetic, and clinical features. The goal was to define disease entities that could be reliably recognized by pathologists and be clinically relevant as well. The Clinical ­Advisory ­Committee for the new WHO ­classification cautioned against grouping disease entities on the ­ basis of their clinical or biologic ­ behavior because it would ­hamper understanding of the specific features of some of the diseases.56 The remainder of this chapter describes the clinical ­ presentation, pathologic diagnosis, treatment, and outcome of some of the more common subtypes of non-Hodgkin lymphoma.168 The chapter concludes with brief descriptions of three special cases—primary central nervous system (CNS) lymphoma in immunocompetent ­patients, AIDS-related non-Hodgkin lymphoma, and ­solitary ­plasmacytomas.

Diffuse Large B-Cell Lymphoma Diffuse large B-cell lymphoma comprises a clinically, morphologically, and genetically heterogeneous group of tumors characterized by a diffuse proliferation of large neoplastic B lymphoid cells with very large nuclei (as large as or larger than the nuclei in normal macrophages). This type of lymphoma accounts for 30% to 40% of non-Hodgkin lymphomas in adults in Western countries. Morphologic variants include centroblastic, immunoblastic, T-cell/histiocyte-rich, and anaplastic. Immunophenotypically, diffuse large B-cell lymphoma expresses various pan-B markers such as CD19, CD20, CD22, and CD79a. Most of the ­anaplastic ­variants and some of the nonanaplastic variants express CD30. Proliferation rates, measured by immunohistochemical staining with the antibody Ki-67, are usually high, and tumors in which 90% or more of the cells are positive for Ki-67 have been linked with particularly poor outcome.186 The molecular heterogeneity of these tumors, reflected in differences in their gene expression patterns, may account for some of the variation in tumor proliferation rate, host ­response to therapy, and differentiation state of the tumor. One study involving DNA microarray analysis of diffuse large B-cell lymphomas identified one type that expressed genes characteristic of germinal center B cells and ­another that ­ expressed genes normally induced during in vitro activation of peripheral blood B cells.187

Prognosis. The prognosis for patients with aggressive lymphomas such as diffuse large B-cell lymphoma can be estimated from a combination of factors that collectively constitute the international prognostic index (see Table 34-4).185 (The Zubrod performance scale,188 which is part of the international prognostic index, is given in Table 34-5.) This index score has been shown to be predictive of overall survival rate and failure-free survival rate.185 ­Minor shortcomings of this prognostic index are its occasional lack of clarity in distinguishing stage I and II disease from stage III and IV disease and in defining what constitutes an extranodal site. Patients who have an elevated serum lactate dehydrogenase level and more than one extranodal site of disease are at high risk for recurrent disease in the CNS, which carries a poor prognosis.189 Patients with lymphoma involving the testes are also at increased risk for CNS involvement.190 Treatment of Localized Disease: Ann Arbor Stage I or II. Approximately 30% to 40% of diffuse large B-cell lymphomas are diagnosed as stage I or II disease, often referred to as localized disease. The most common treatment strategy for stage I or II disease has been anthracyclinebased chemotherapy followed by radiation. The addition of the anti-CD20 antibody rituximab to chemotherapy consisting of cyclophosphamide, hydroxydaunomycin, vincristine, and prednisone (CHOP regimen) became the standard of care after showing benefit in terms of time to progression and overall survival time compared with CHOP alone.191 Unresolved issues include the role of radiation as consolidation therapy and the optimal number of cycles of ­chemotherapy. Several randomized trials have been undertaken to address these issues. Trials by the SWOG192 and the ECOG193 demonstrated a benefit from involved-field radiation therapy after CHOP induction chemotherapy for patients with clinical stage I or stage II large B-cell lymphoma. In the SWOG trial,192 which compared three cycles of CHOP plus involved-field irradiation (40 to 55 Gy to the prechemotherapy treatment volume) with eight cycles of CHOP without irradiation, the corresponding 5-year diseasefree survival rates were 77% for chemoradiation therapy and 64% for chemotherapy only (P = .03). Higher doses of radiation (i.e., 45 to 55 Gy) were typically used if the lymphoma was initially larger than 10 cm or if the induction chemotherapy produced only a partial response. The chemoradiation protocol also seemed to improve overall survival rates over those produced by chemotherapy alone (82% and 72% at 5 years, P = .02), but an update of the SWOG trial published in 2001194 showed no differences in failure-free survival or overall survival rates between the two treatment groups, probably because of late relapses and lymphoma deaths more than 5 years after treatment among patients given the abbreviated chemotherapy followed by radiation therapy.194 In the ECOG trial,193 patients who achieved a complete response to eight cycles of CHOP were randomly assigned to observation or to radiation, whereas patients with a partial response all went on to receive radiation therapy. ­ Despite a slight imbalance in patient characteristics between groups (more patients in the combinedtherapy group had bulky disease), the administration of involved-field radiation therapy (30 Gy to the prechemotherapy tumor volume) was found to significantly improve

34  Leukemias and Lymphomas

disease-free ­survival rates (6-year rates of 73% versus 56%, P = .05). A 12% difference in overall survival rate seeming to favor radiation therapy was not statistically significant. When eight cycles of CHOP chemotherapy produced a partial response and radiation therapy to 40 Gy was administered, the 6-year disease-free and overall survival rates were both 63%. This important observation suggests that patients with a partial response to chemotherapy who go on to receive radiation therapy did almost as well as did those with a complete response to chemotherapy. These results are similar to those found by the EORTC for advanced Hodgkin disease.127 The GELA conducted a similar study comparing ­aggressive chemotherapy (i.e., dose-intensified doxorubicin, cyclophosphamide, vindesine, bleomycin, and prednisone [ACVBP regimen]) alone with abbreviated chemotherapy (i.e., three cycles of CHOP) followed by involved-field radiation therapy for stage I or II, mostly low-risk, aggressive lymphoma.195 All patients in this study were younger than 60 years. The 5-year event-free (82%) and overall survival (90%) rates were significantly better for the chemotherapy group than for the combined-modality group (74% and 81%, respectively). Although the addition of radiation therapy reduced relapses at the initial disease sites, it was not enough to overcome the excessive number of relapses in the abbreviated-chemotherapy group.195 Despite results indicating the inability of abbreviated chemotherapy plus radiation to prevent out-of-field relapses, the GELA conducted another trial (GELA LHN 93-4) comparing CHOP with CHOP plus involved-field radiation therapy to 40 Gy, this time for patients older than 60 years.196 (The use of ACVBP had been dropped by the time this trial was undertaken because of excessive toxicity.) At a median follow-up time of 7 years, no significant differences were evident in 5-year event-free survival rates (61% for chemotherapy alone versus 64% for chemoradiation therapy) or overall survival rates (72% versus 68%). The results for the chemotherapy-only group were similar to results for the group that received eight cycles of CHOP in the previously discussed SWOG trial. However, because relapses can be expected to appear beyond 5 years, ­another 5 to 10 years should be allowed to elapse before this approach is widely adopted. Until that time, conclusions that can be drawn from these trials are that (1) combined­modality therapy should be still considered the standard of care for early-stage diffuse large B-cell lymphoma; (2) abbreviated chemotherapy should not be used because of the pattern of relapse outside the radiation field; and (3) randomized trials comparing rituximab plus CHOP with or without radiation should be conducted to address the two most pressing unresolved issues—the benefit of rituximab and the optimal number of chemotherapy cycles. Treatment of Advanced Disease: Ann Arbor Stage III or IV. Whether radiation therapy has a role in the treatment of bulky stage III or IV large B-cell lymphoma ­remains a matter of debate. Some groups see no benefit,197-199 and others suggest that delivery of involved-field ­radiation to lymphomas measuring 4 cm or more before induction chemotherapy may improve disease-free and overall survival rates.200-203 However, most of these studies were done ­retrospectively; the only prospective randomized study conducted demonstrated that radiation therapy could ­benefit patients with bulky (>10 cm) stage IV diffuse large

895

B-cell lymphoma.204 In that study, patients who experienced a complete response to anthracycline-based chemotherapy were randomly assigned to receive or not receive involved-field radiation therapy (40 to 50 Gy) to initially bulky sites of disease. The addition of radiation therapy produced significant improvements in disease-free survival rates (5-year rates of 72% and 35%, P < .01) and overall survival rates (5-year rates of 81% and 55%, P < .01).204 Radiation Dose-Response Analysis. Before the 1980s, aggressive lymphomas were typically treated with radiation therapy alone. Between 1979 and 1987, four randomized studies205-208 demonstrated an improvement in 5-year disease-free survival from the addition of cyclophosphamide, vincristine, and prednisone chemotherapy to radiation therapy. Anthracyclines such as doxorubicin were ­subsequently found to be active against this disease, giving rise to the current use of CHOP209,210 followed by involved-field radiation therapy. Stryker and colleagues211 reported that the mean dose that produced local control for patients treated with radiation therapy alone was 48 Gy, compared with 37 Gy for patients treated with chemotherapy and radiation ­therapy. Mirza and colleagues212 have suggested that among patients who experience a complete response to chemotherapy, a total radiation dose of only 30 Gy is sufficient. Kamath and colleagues213 have also suggested that 30 Gy is an adequate dose for large B-cell lymphomas that are initially 6 cm or smaller and completely respond to chemotherapy. Similarly, Wilder and colleagues214 at M.  D. Anderson observed that 30 Gy is adequate for lymphomas smaller than 3.5 cm that completely respond to a median of three cycles of chemotherapy. In that study, 45 Gy was recommended for lymphomas that were larger than 10 cm before chemotherapy and completely responded to six cycles of chemotherapy.214 After analyzing a different group of patients, Fuller and colleagues,215 also at M.  D. Anderson, recommended doses of at least 40 Gy for lymphomas that are initially 10 cm or larger and completely respond to chemotherapy. Others have reported that doses of at least 40 Gy seem to be necessary for local control of large B-cell lymphomas that are larger than 6 cm at the start of chemotherapy213 or that respond only partially to chemotherapy.213,216,217 The dose-response relationship in these studies may reflect the presence of a greater residual tumor burden after treatment of bulky lymphomas with induction chemotherapy.218 Primary Mediastinal Lymphoma. Primary mediastinal lymphoma is a distinct subtype of diffuse large cell lymphoma originating from thymic cells. Accounting for about 2% of all non-Hodgkin lymphoma cases,219 primary mediastinal lymphoma is thought to be derived from mature germinal center or postgerminal center B cells.220 Disease is limited to the chest in approximately 80% of cases. It is a disease of young age, with a slight preponderance among females. Gene expression patterns resemble those of ­Hodgkin lymphoma.221 Believing primary mediastinal lymphoma to be more aggressive than other large cell lymphomas, the International Extranodal Lymphoma Study Group retrospectively evaluated a variety of dose-intense or high-dose chemotherapy regimens, among them methotrexate, doxorubicin, ­cyclophosphamide, vincristine, prednisone, and bleomycin (MACOP-B regimen), with or without radiation ­therapy.222

896

PART 11  Leukemias and Lymphomas

In at least one such study,223 relapse-free response rates were found to be as high as 93%. The question of whether radiation therapy after chemotherapy improves survival rates was raised by the GELA in 1996.219 Others, however, have argued that radiation improves outcome for patients treated with MACOP-B,223 a conclusion confirmed in a retrospective review at M.   D. Anderson showing that the addition of radiation led to better local control rates.224

Follicular Lymphoma Follicular lymphoma is the second most common lymphoma in the United States and Western Europe, constituting about 20% of all non-Hodgkin lymphomas. ­Follicular lymphoma is more common among whites than Asians or blacks,225 and the median age at diagnosis is 60 years.225,171 Morphologically, these lymphomas are composed of ­follicles containing a mixture of centrocytes (i.e., cleaved ­follicle center cells or small cells) and centroblasts (i.e., noncleaved follicle center cells or large cells) that ­proliferate within a network of nonmalignant follicular dendritic cells and T cells. The tumor cells gradually ­replace the normal nodal architecture in a pattern that can be entirely follicular or include some diffuse component. The proportion of centroblasts determines the tumor grade; in the WHO system, grade 1 disease has 0 to 5 centroblasts per high-powered microscopic field (HPF), grade 2 has 6 to 15 centroblasts/HPF, and grade 3 disease has more than 15 centroblasts/HPF. The tumor cells are derived from normal germinal center B cells and express B-cell markers such as CD19, CD20, CD22, and CD79; the germinal center B-cell markers BCL-6, CD38, and CD10; and membrane-bound immunoglobulin. Unlike normal germinal center B cells, virtually all follicular lymphoma cells strongly express BCL2, although their proliferation rate (according to Ki-67 staining) is lower than that of reactive follicles. Expression of CD5 and CD43 is usually absent, and that of CD23 varies. The genetic hallmark of follicular lymphoma, present in 80% to 90% of cases, is the t(14;18)(q32;q21) translocation, which leads to rearrangement and subsequent ­ activation of the antiapoptotic protein BCL2.226,227 Prognosis. The usefulness of the international prognostic index developed for non-Hodgkin lymphoma (see Table 34-4)185 for follicular lymphomas was controversial shortly after its introduction,228,229 and this index was ­subsequently modified to better predict prognosis in cases of follicular lymphoma. Like its predecessor, the Follicular Lymphoma International Prognostic Index230 (FLIPI) includes age (>60 years versus <60 years), Ann Arbor disease stage (III to IV versus I to II), number of involved nodal areas (>4 versus <4), and serum lactate dehydrogenase level (above normal versus normal or below); the fifth prognostic factor is hemoglobin level (<120 g/L versus > 120 g/L). The three risk groups defined in terms of the FLIPI are low risk (0 or 1 adverse factor), intermediate risk (2 factors), and poor risk (>3 adverse factors). This index was validated in a prospective trial by the German Low-Grade Lymphoma Group,231 who assessed 362 patients with advanced stage follicular lymphoma given rituximab and CHOP as firstline therapy. At 2 years, the event-free survival rate was 67% for the high-risk group, 92% for the low-risk group, and 90% for the intermediate-risk group (P < .001 for the three-way comparison).231

Molecular markers are being examined for their potential use as indicators of disease status and prognosis. For example, groups are studying whether molecular response, as assessed by polymerase chain reaction analysis, can be considered an early end point with which to predict long-term disease-free survival.232 Bertoni and colleagues233 reported that a molecular complete response to chemotherapy was associated with better overall survival. Similarly, LopezGuillermo and coworkers234 suggested that a molecular complete response to chemotherapy was associated with improved failure-free survival. This finding was corroborated by Apostolidis and colleagues,235 who reported that a molecular complete response to chemotherapy was associated with lower risks of relapse and death. Gribben and associates236 found that the reappearance of t(14;18)– ­positive circulating cells was strongly associated with recurrence of lymphoma after myeloablative therapy. Ha and colleagues237 and Cox238 have suggested that a molecular complete response is a favorable prognostic sign in patients treated with central lymphatic irradiation. Two factors influence the course of treatment in follicular lymphoma. The first is the considerable variability in the course of the disease (i.e., indolent, characterized by waxing and waning versus rapid growth and massive tumors that produce symptoms that require treatment), and the second is the tendency for histologic transformation to diffuse large B-cell lymphoma, which occurs at rates of 5% to 10% per year depending on the initial percentage of large cells in the specimen.239 Treatment of Localized Disease: Ann Arbor Stage I or II. Unlike aggressive lymphoma, follicular lymphoma is usually considered incurable with conventional treatment. The course of the disease, however, can be quite long. Besa and colleagues240 at M.   D. Anderson reported recurrences 16 years after involved-field radiation therapy. Although MacManus and Hoppe241 at Stanford observed no recurrences 14 to 30 years after radiation therapy to one side of the diaphragm, relapses occurred 22 years after total nodal irradiation. Paryani and colleagues242 and MacManus and Hoppe241 at Stanford have reported that total lymphoid irradiation for stage I or II follicular lymphomas does not significantly improve overall survival compared with involved-field radiation therapy (15-year rates of 45% for total nodal irradiation versus 43% for involved-field irradiation, P = .19). As a result, at many institutions, patients with stage I or II follicular lymphomas are treated with involved-field radiation therapy.243-247 Use of this approach to treat stage I or stage II disease has led to freedom from treatment failure rates of 41% to 49% and overall survival rates of 43% to 79%.241,244,248-250 M.  D. Anderson is one of the few institutions where central lymphatic irradiation is administered for clinical stage I to III follicular lymphomas.251 Central lymphatic irradiation differs from total lymphoid irradiation in that all of the mesenteric lymph nodes are included in the clinical target volume (Fig. 34-7). Consequently, wider abdominal and pelvic portals are irradiated than in total lymphoid irradiation. This aggressive treatment is advocated because follicular lymphomas are less likely to ­ recur with this approach (compare Fig. 34-8252-254 and Fig. 34-9241). Central lymphatic irradiation ­typically consists of opposed equally weighted mantle, whole

34  Leukemias and Lymphomas A

Gehan P-values 1 vs 2 0.0012

100 Probability (%)

897

80 60 1

40 20 2

2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Time (years)

Percent surviving relapse free

FIGURE 34-7.  Clinical target volumes in central lymphatic ­irradiation. Treatment is begun to the region with the greatest ­tumor burden. The superior-most region shown is irradiated only if Waldeyer’s ring or the upper cervical nodes are involved. The daily fraction size is 1.5 Gy for the whole abdomen and 1.8 Gy for the remaining regions.

De Los Santos et al. 1993 (n = 21) Jacobs et al. 1993 (n = 34) Paryani et al. 1994 (n = 61)

100 80 60 40 20 0 0

2

4

6

8

10

12

14

16

18

20

Years

FIGURE 34-8.  Freedom from relapse of follicular lymphoma treated with central lymphatic irradiation or total nodal irradiation. The green curve represents patients with stage I, II, or III disease, whereas the red and blue curves include only patients with stage III disease. The number of patients in each study is shown in parentheses. (Data from Paryani SB, Hoppe RT, Cox RS, et al. The role of radiation therapy in the management of stage III follicular lymphomas. J Clin Oncol 1984;2:841-848; Jacobs JP, Murray KJ, Schultz CJ, et al. Central lymphatic irradiation for stage III nodular malignant lymphoma: long-term results. J Clin Oncol 1993;11:233-238; De Los Santos JF, Mendenhall NP, Lynch JW. Is comprehensive lymphatic irradiation for low-grade nonHodgkin’s lymphoma curative therapy? Long-term experience at a single institution. Int J Radiat Oncol Biol Phys 1997;38:3-8.)

a­ bdominal, and whole pelvic fields (see Fig. 34-7) that are treated in three stages separated by a 4-week break. De Los Santos and colleagues253 at the University of Florida College of Medicine reported that treatment with central lymphatic irradiation produced 5-year, 10-year, and 15-year RFS rates of 75%, 58%, and 58% for patients with stage I to III follicular lymphomas (see Fig. 34-8). Adverse prognostic factors identified in multivariate analyses have

FIGURE 34-9.  Effect of radiation therapy field size on freedom from relapse in 177 patients with Ann Arbor stage I or II follicular lymphomas. Curve 1 (red) represents the results of radiation therapy to one side of the diaphragm (136 patients), and curve 2 (blue) represents the results of radiation therapy to both sides of the diaphragm (41 patients). (From MacManus MP, Hoppe RT. Is radiotherapy curative for stage I and II low grade follicular lymphoma? Results of a long-term follow-up study of patients treated at Stanford University. J Clin Oncol 1996;14:1282-1290.)

varied across studies and have included age older than 60 years, grade 3 histology, bulky disease, and extranodal presentation. Local irradiation can still be offered for follicular lymphoma, except for tumors of grade 3 histology, which are treated similarly to aggressive diffuse large B-cell ­lymphoma. The incurable nature of this disease has led to the ­exploration of alternative approaches for the treatment of follicular lymphoma. Advani and colleagues255 at Stanford University published a retrospective study of 43 patients with stage I or II grade 1 or 2 follicular lymphoma in whom treatment had been deferred for at least 3 months because of advanced patient age, the need for a large abdominal irradiation field, or the physician’s or patient’s choice. At a median follow-up time of 86 months, 27 of the original 43 patients had received no treatment; 4 of the 16 patients who had been treated had experienced transformation to a higher-grade lymphoma. In all, nine patients died, six of progressive disease. The 10-year overall survival rate was 85%. These investigators concluded that deferring therapy might be appropriate in selected cases. In a different approach, investigators at the Peter ­MacCallum Cancer Institute in Australia256 sought to ­determine whether adding chemotherapy to involved-field irradiation would improve disease control and ­ survival. In that study, 102 patients with stage I or II indolent ­lymphoma were treated with 10 cycles of cyclophosphamide, ­ vincristine, prednisone, and bleomycin, with ­doxorubicin added for patients with extranodal disease, high lactate dehydrogenase levels, bulky disease, or diffuse small ­lymphocytic histology. This chemotherapy was ­followed by ­ involved-field irradiation to a dose of 30 to 40 Gy. Among the 85 patients with disease of follicular ­histology, the 10-year time to treatment failure rate was 72%, and the overall survival rate was 80%. None of the other 17 patients (whose tumors were of diffuse small lymphocytic or mucosa-associated histology) experienced relapse. The 10-year survival rate for those who did ­experience relapse was 46%.256

898

PART 11  Leukemias and Lymphomas

Treatment of Ann Arbor Stage III Disease. Jacobs and colleagues252 observed a 15-year disease-free survival rate of 40% after the administration of central lymphatic irradiation or total lymphoid irradiation for stage III follicular lymphoma (see Fig. 34-8). Paryani and colleagues254 also reported a 15-year RFS rate of 40% for patients with stage III follicular lymphomas treated with total lymphoid irradiation (see Fig. 34-8). Central lymphatic irradiation can be offered for follicular lymphomas that have recurred after treatment with chemotherapy alone, provided that the bone marrow has never been involved.257 In such salvage central lymphatic irradiation, liver blocks are used from the start to reduce the risk of hepatitis.200 Mahe and colleagues258 used irradiation of the whole abdomen and pelvis (0.8 Gy twice daily to a total dose of 20.0 Gy over 3 weeks) to treat 22 patients with follicular lymphomas that had recurred subdiaphragmatically after chemotherapy. This approach resulted in 10-year diseasefree and overall survival rates of 37% and 45%, respectively. As a result, the investigators suggested that abdominopelvic irradiation may be a feasible alternative when bone marrow transplantation is not possible. Chemotherapy alone is commonly recommended for patients with stage III follicular lymphomas that are bulky or are causing symptoms or organ dysfunction.259,260 The addition of involved-field radiation therapy has not improved failure-free or overall survival for patients with stage III follicular lymphoma.261 The use of combined­modality therapy remains controversial for patients with stage I to III follicular lymphomas.259 Treatment of Ann Arbor Stage IV Disease. The optimal timing and type of treatment for stage IV follicular lymphomas are a matter of debate. Twenty-three percent of follicular lymphomas spontaneously regress, at least partially.262 Horning and Rosenberg263 at Stanford observed no overall survival benefit for immediate as opposed to delayed chemotherapy in patients with stage III or IV follicular lymphoma. Similarly, a prospective trial conducted by the National Cancer Institute found no overall survival benefit for immediate compared with delayed treatment.264 Consequently, most oncologists advocate a watch-and-wait philosophy for previously untreated, smaller than 10 cm, stage III or IV follicular lymphomas that are not causing symptoms or organ dysfunction.262,265 However, some oncologists recommend that such patients undergo immediate chemotherapy.266 Antisense oligonucleotides that inhibit BCL2 protein expression have been introduced267 that may sensitize t­umor cells to chemotherapy.268 Rituximab immunotherapy also constitutes a promising approach.269 Rituximab, a chimeric anti-CD20 monoclonal antibody,270 is the first monoclonal antibody to gain U.S. Food and Drug Administration approval for cancer therapy. Radioimmunotherapy may be helpful.271 Radiation Dose-Response Analysis. The optimal ­radiation dose for control of follicular lymphoma is a ­matter of debate. Arthur272 found that doses ranging from 20 to 30 Gy produced local control in 27 of 27 low-grade lymphomas. However, Sutcliffe and colleagues273 did not observe a dose-response relationship with doses ranging from 20 Gy to 40 Gy. Cox and colleagues274 found that doses of less than 25 Gy produced local control in

Helicobacter pylori positive

H. pylori negative

Antibiotics

Complete response

Incomplete response

Involved-field radiation therapy

Follow with endoscopic biopsies every 3–6 mo; deliver involved-field radiation therapy if gastric MALT recurs.

FIGURE 34-10.  Treatment algorithm for localized gastric ­mucosa-associated lymphoid tissue (MALT) lymphoma.

83 (90%) of 92 low-grade lymphomas, and doses ranging from 25 Gy to 46 Gy controlled 21 of 21 lymphomas. Kamath and colleagues213 have suggested that 30 Gy is sufficient for most low-grade lymphomas, except for small-volume disease in the orbit, for which 20 to 25 Gy may be sufficient.

Marginal Zone or Mucosa-Associated Lymphoid Tissue Lymphoma MALT lymphoma can manifest as extranodal, nodal, or splenic disease. The most common presentation, ­extranodal, represents an infiltration of extranodal zone B cells into epithelial tissues in the gastrointestinal tract, salivary glands, breast, thyroid, prostate, skin, orbit, conjunctiva, lung, kidney, brain, or soft tissues. MALT lymphomas are immunophenotypically positive for B-cell markers such as CD19, CD20, and CD22 and are negative for CD5, CD10, and CD23. Microbial agents that have been linked with the development of MALT lymphoma include H. pylori,143,275 Borrelia burgdorferi,276,277 Campylobacter jejuni,278 and Chlamydia psittaci.279 Autoimmune disorders, such as Hashimoto’s thyroiditis in thyroid MALT lymphoma280 and Sjögren’s syndrome in thymic281 or salivary-gland MALT lym­phoma, have also been implicated.282,283 Treatment of an H. pylori infection in gastric MALT lymphoma with antimicrobial agents and a potent acid-inhibiting drug or bismuth284 can produce complete remission in about almost two thirds of gastric MALT lymphomas.285-287 Consequently, an ­antibiotic-based approach can be offered to patients with localized gastric MALT lymphomas associated with an H. pylori infection. For patients with H. pylori–negative or antibiotic-resistant gastric MALT lymphomas, involved-field ­radiation therapy alone is recommended (Fig. 33-10).288-290 After a median follow-up of 30 months, Yahalom and colleagues290 ­reported a disease-free survival rate of 94% ± 5% for patients given a median dose of 30 Gy in 20 fractions over 4 weeks. Zinzani and colleagues291 reported results from 75 patients with previously untreated stage IE to IV nongastrointestinal MALT lymphomas who were given treatment ranging from local administration of interferon-α to surgery and postoperative chemoradiation therapy. The 5-year freedom from progression rate for the 47 patients with stage

34  Leukemias and Lymphomas

IE or IIE disease, many of whom underwent involved-field radiation therapy, was 71%. The National Comprehensive Cancer Network Practice Guidelines292 recommend that stages IE and IIE, nongastric and H. pylori–negative or antibiotic-­resistant gastric MALT lymphomas be treated with involved-field radiation therapy to 30 to 33 Gy. Rituximab, alone or in combination with chemotherapy, is used for ­advanced disease.293,294

Peripheral T-Cell Lymphoma The incidence of peripheral T-cell lymphomas is increasing particularly rapidly, and the cause is unknown.295 Patients with peripheral T-cell lymphomas tend to be younger than 35 years and to have widespread disease at presentation.296 A complete response to anthracycline-based chemo­therapy occurs in only about 45% of cases, and disease-free sur­vival typically is brief (see Fig. 34-6D).297 The prognosis for peripheral T-cell lymphoma is poorer than that for most other non-Hodgkin lymphomas (see Fig. 34-6D).168 Chemotherapy followed by involved-field radiation therapy to initially bulky sites of disease is recommended.

Extranodal Natural Killer Cell/T-Cell Lymphoma, Nasal and Nasal-Type Pathologically, nasal NK/T-cell lymphomas show a broad morphologic spectrum, frequent ­ angioinvasion and angio­ centricity with zone necrosis, and nearly ­ ubiquitous ­association with Epstein-Barr virus infection.298 Immunophenotypically, the tumor cells usually express CD2 and CD56. Clinically, this entity is characterized by its ­predominance in young males, its tendency to ­manifest as localized stage I or stage II disease, its resistance to ­conventional ­chemotherapy, and its sensitivity to radiation therapy.299 The typical clinical presentation is nasal obstruction and bleeding, and local invasion of the orbit, sinuses, nasopharynx, and oropharynx is common. Local radiation therapy emerged during the early 2000s as being the most important component of treatment; the 5-year progression-free sur­vival rate after local radiation therapy is about 61%, regardless of whether radiation is used alone or in combination with chemotherapy.300 A radiation dose-response analysis301 showed a sigmoid curve for response to radiation between 20 and 54 Gy, followed by a plateau for doses higher than 54 Gy; the current recommendation is that the target volume plus a generous margin should be treated to 54 Gy.301

899

pleomorphic or large cells and a very high (>90%) proliferation fraction.56 Treatment for Burkitt’s lymphoma consists largely of dose-intensive multiagent chemotherapy, with high-dose systemic methotrexate and cytosine arabinoside used to control CNS and systemic disease. The use of radiation for CNS prophylaxis has been abandoned by most, but not all,303 groups. For patients with stage IV disease, the recommended dose of craniospinal radiation is 24 Gy in 12 fractions over 2.5 weeks.302 This protocol can also be used to palliate symptoms in patients with a chemorefractory CNS recurrence. In general, 3-year overall survival rates for ­patients with Burkitt’s lymphoma range from about 49% to 74%.304,305

Mantle Cell Lymphoma Mantle cell lymphoma occurs as one of two morphologic subtypes. The classic or typical form comprises 90% of cases, and the blastoid variant constitutes the other 10%. Immunophenotypically, mantle cell lymphomas resemble mature B cells; their lack of expression of CD10 and CD23 helps to distinguish them from chronic lymphocytic leukemia. Cytogenetically, mantle cell lymphoma cells are characterized by the t(11;14)(q13;q32) translocation, which leads to overexpression of cyclin D1.306 Mantle cell lymphoma typically manifests as advanced disease that cannot be cured with conventional chemotherapy (see Fig. 34-6D).154,155,307 Patients at M.   D. Anderson who have mantle cell lymphoma and are younger than 60 years undergo hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone chemotherapy alternating with high-dose methotrexate and cytarabine chemotherapy, TBI, and bone marrow transplantation. For previously untreated patients with stage IV disease, this approach produced 3-year event-free and overall survival rates of 72% and 92%, rates that compare favorably with the 28% event-free and 56% overall-survival rates observed for historical controls treated with CHOP-based chemotherapy.308 Despite the paucity of randomized trials showing a benefit for transplantation, many oncologists ­advocate an aggressive approach in young patients because the ­ median survival duration after standard, CHOP-like chemotherapy is only 3 years.309

Burkitt’s Lymphoma

Precursor B-Cell and T-Cell Lymphoblastic ­Lymphomas

Burkitt’s lymphoma is an uncommon, highly aggressive tumor of B-cell origin that often arises in extranodal sites involving the abdomen (e.g., cecum, mesentery)199,302 and shows a high affinity for the CNS despite the use of chemotherapeutic or radiation prophylaxis. Three clinical and genetic subtypes were proposed in the WHO classification: endemic, sporadic, and immunodeficiency-related.56 Morphologically, most endemic cases and most sporadic cases show the classic pathologic starry-sky pattern; immunophenotypically, most cells express B-cell markers, although expression of CD10 is also common. All types of Burkitt’s lymphoma have a translocation involving MYC at band q24 on chromosome 8 to the Ig heavy chain region of chromosome 14 or to the Ig light chain loci on chromosome 22 or chromosome 2. A morphologic variant, Burkitt-like lymphoma, resembles Burkitt’s lymphoma but has more

Precursor lymphoblastic lymphomas of B-cell or T-cell origin are each recognized as distinct entities in the WHO classification. T-cell origin is by far the more common of the two, contributing 85% to 90% of cases. Patients with precursor T-cell lymphoblastic lymphomas tend to be younger than those with precursor B-cell lymphoblastic lymphomas and to more often present with bone marrow involvement.310 Lymphoblastic lymphomas involving the mediastinum can progress to a leukemic phase that is cytologically indistinguishable from acute lymphoblastic leukemia.311 Given the natural history of the disease, the treatment is similar to that for acute leukemia and consists of intensive chemotherapy312,313 and intrathecal chemotherapy for CNS prophylaxis. At some institutions, CNS prophylaxis also includes cranial irradiation to a dose of 24 Gy given in 12 fractions over 2.5 weeks.314,315

900

PART 11  Leukemias and Lymphomas

TABLE 34-6.   Mycosis Fungoides Cooperative Group Staging Guidelines for Cutaneous T-Cell Lymphoma Clinical Stage

T Category (Skin)*

Palpable Nodal Sites

I

T0-T1

0-1

II

T0-T1

2-8

T2

0-1

T2

2-8

T3

0-8

T4

0-8

III IV *Primary

tumor classification: T1, plaques, papules, or eczematous patches involving less than 10% of the skin; T2, plaques, papules, or eczematous patches involving at least 10% of the skin; T3, at least 3 tumors; and T4, generalized erythroderma or Sézary syndrome. From Lamberg SI, Green SB, Byar DP, et al. Clinical staging for cutaneous T-cell lymphoma. Ann Intern Med 1984;100:187-192.

The use of mediastinal irradiation is controversial, with some groups reserving this approach for patients who present with respiratory distress or those who achieve only a partial response to chemotherapy.311,316 Others317 deliver mediastinal irradiation to all adults with precursor T-cell lymphoblastic lymphomas that involve the mediastinum in an effort to improve local control.

Mycosis Fungoides Mycosis fungoides, also known as Sézary syndrome or cutaneous T-cell lymphoma, is a clinicopathologic diagnosis. Staging guidelines from the Mycosis Fungoides Cooperative Group are presented in Table 34-6.318 Pathologic features include an infiltrate of atypical T lymphocytes with cerebriform nuclei in the epidermis (i.e., epidermotropism).55 These cells may form aggregates known as Pautrier’s microabscesses.319,320 In most cases, malignant T cells express CD2, CD3, CD4, and CD5 but not CD7 or CD30. T-cell receptor genes are clonally rearranged, with γ region chain rearrangements.320 Because of the rarity of mycosis fungoides (only 1000 new cases appear each year in the United States) and the consequent paucity of randomized trials, treatment remains controversial.321 Treatments used for early-stage disease have included topical steroids, topical mechlorethamine, topical retinoids, phototherapy, and total-skin electronbeam radiation therapy. Total-skin electron-beam radiation therapy has been used to treat patients with mycosis fungoides since 1951.322 Acute side effects of this therapy include generalized erythema, dry desquamation, and soreness of the hands and feet. The treatment also causes reversible alopecia, loss of fingernails and toenails, decreased ability to sweat for 6 to 12 months, chronically dry skin, and an increased risk of second cutaneous cancer. The Ontario Cancer Foundation has suggested that total-skin electron-beam radiation therapy is potentially curative in approximately 29% of patients with T1 or T2 mycosis fungoides, provided that no palpable nodal sites are present.323 The Stanford group has suggested that total-skin electron-beam radiation therapy is potentially curative in patients with early-stage disease. Hoppe and

colleagues324 reported that for patients with plaques involving less than 25% of the skin surface (i.e., limited-plaque disease), total-skin electron-beam radiation ­therapy resulted in a 10-year disease-free survival rate of 45%. No recurrences were observed at 4 to 10 years after the ­radiation. Kim and colleagues325 reported that patients with plaques involving less than 10% of the skin treated with topical mechlorethamine have achieved life expectancies similar to those of age- and sex-matched control populations. Hamminga and colleagues326 have reported that topical mechlorethamine and total-skin electron-beam radiation therapy are equally effective for the treatment of patients who have plaques involving less than 10% of the skin. Two other groups327,328 have suggested that topical mechlorethamine is potentially curative in patients with early-stage mycosis fungoides. The complete response rate for patients with limited-plaque disease after topical mechlorethamine in ointment (approximately 55%) is not as high as that after electron-beam radiation therapy (approximately 95%).319 Nevertheless, these patients can be treated with total-skin electron-beam radiation therapy if disease progresses after topical mechlorethamine, and they achieve the same outcome as if they had initially undergone radiation therapy.329 One prospective randomized trial has addressed the use of aggressive or conservative treatment for mycosis fungoides. No benefit was found in terms of disease-free or overall survival from early, aggressive treatment consisting of cyclophosphamide, doxorubicin, etoposide, and ­vincristine chemotherapy and total-skin electron-beam radiation therapy compared with a more conservative approach beginning with topical mechlorethamine.330 The toxicity of ­ chemotherapy and total-skin electron-beam radiation therapy in that study was considerable. Localized radiation therapy331-333 (e.g., 30 Gy in 15 fractions over 3 weeks using a 2-cm margin for visible and palpable disease334) and psoralen activated with ultraviolet-A light (PUVA)335,336 are effective treatment options for patients with early-stage mycosis fungoides.320 Total-skin electron-beam radiation therapy is typically recommended for patients who present with thick plaques or have a recent history of rapid progression of cutaneous disease and are no longer responding to topical mechlorethamine or PUVA.320 Among patients who undergo total-skin electron-beam radiation therapy, overall survival is significantly better when a total dose of 30 Gy or more is delivered in daily 1.5- to 2.0-Gy fractions.324 Some groups326,337 have reported that the application of topical mechlorethamine after electron-beam radiation therapy prolongs disease-free survival. Zackheim338 and Chinn and colleagues337 recommend chemotherapy for the initial palliative management of cutaneous symptoms in patients with lymph node or visceral metastases, reserving total-skin electron-beam radiation therapy for disease that does not respond to chemotherapy. Progression-free survival time after high-dose chemotherapy and autologous stem cell or bone marrow rescue has typically been less than 4 months, although experience with this approach is limited.339-341 In summary, most physicians recommend topical mechlorethamine,325-328,337,342-344 localized radiation therapy,331-333 or PUVA335,336 for patients with mycosis ­ fungoides

34  Leukemias and Lymphomas

involving less than 10% of the skin. Combination chemotherapy is commonly recommended for patients with lymph node or visceral metastases.320,327,338,345 At M.   D. Anderson, total-skin electron-beam radiation therapy is administered by using a Stanford dual fixed-angle six-field technique346 modified to include a 1.2-cm Lucite scatter plate to reduce the electron energy and improve dose homogeneity.347 A total dose of 32 Gy is delivered in nine cycles over 8 weeks.347

CD30-Positive Cutaneous Lymphoproliferative ­Disorders: Anaplastic Large Cell Lymphoma and Lymphomatoid Papulomatosis CD30+ lymphoproliferative disorders are the second most common form of cutaneous T-cell lymphoma after mycosis fungoides, contributing about 25% of all cutaneous T-cell lymphomas. They tend to predominate in men, with a ratio of approximately 1.5 to 2.0:1. Clinically and biologically, cutaneous T-cell lymphomas are a heterogeneous group of non-Hodgkin lymphomas defined by clonal proliferation of skin-homing malignant T lymphocytes. Their common phenotypic hallmark is the CD30 antigen, which in lymphocytes is upregulated by Epstein-Barr ­virus, human herpesvirus, and human T-cell lymphotropic viruses 1 and 2. This upregulation has been implicated in the pathogenesis of CD30+ lymphoproliferative diseases.348 CD30+ lymphoproliferative disorders are also characterized by spontaneous regression of skin lesions and a generally low-grade malignant course with an excellent ­prognosis.349,350 Lymphomatoid papulomatosis typically manifests as chronic, recurrent, self-healing erythematous papules and nodules that may become hemorrhagic, necrotic, or ­ulcerative.351 In most cases, cutaneous anaplastic large cell lymphomas manifest as solitary or regional nodules or tumors that often show ulceration. Extracutaneous or regional lymph node involvement is seen in 10% of patients at ­presentation. No curative treatment is available for cutaneous CD30+ lymphoproliferative disorders. Historically, the most commonly reported treatment modalities for lymphomatoid papulomatosis are doxycycline, PUVA, UVB light therapy, low-dose methotrexate, interferon-α, topical steroid and bexarotene (Targretin) formulations, and low-dose radiation. However, because none of these treatments alters the natural course of disease, the short-term benefits should be weighed against potentially harmful side effects. For patients with few lesions, observation is recommended; for patients with more disseminated disease, low-dose methotrexate or UV light treatment could be effective in clearing the lesions.352,353 The preferred treatments for CD30+ cutaneous anaplastic large cell lymphoma are surgery or spot irradiation (34 to 44 Gy) for solitary or localized lesions, with systemic chemotherapy reserved for cases with a large ­tumor burden or extracutaneous involvement.354 However, relapse occurs in approximately 40% of patients despite treatment.350 Chemotherapy regimens include CHOP or CHOP-like combinations, interferon-α, or oral bexarotene. According to the Dutch Cutaneous Lymphoma Group, multiagent systemic chemotherapy was no more effective than single-agent therapy in terms of cure rate or ­preventing relapses.350

901

Primary Central Nervous System Lymphoma in ­Immunocompetent Patients Primary CNS lymphoma is an uncommon variant of extranodal non-Hodgkin lymphoma and accounts for 3% of all primary CNS tumors diagnosed each year in the United States. The mean age at diagnosis of primary CNS lymphoma in immunocompetent patients is 55 years.355 Most patients present with a focal neurologic deficit such as hemiparesis or dysphasia.356,357 Histopathologic confirmation, preferably by stereotactic biopsy, is essential,355 and the extent of disease should be established by gadoliniumenhanced magnetic resonance imaging or with contrast­enhanced CT.358 Most of the current knowledge of treatment for primary CNS lymphoma has come from nonrandomized trials of small numbers of patients or from retrospective reviews of multicenter experiences, and comparisons of the various treatment options have been difficult.359 However, it seems clear that whole-brain irradiation alone is insufficient for durable local control,360 and it can be associated with significant neurotoxicity, particularly among patients older than 60 years. Combination chemoradiation therapy involving highdose methotrexate followed by whole-brain irradiation is superior to whole-brain irradiation alone,360-362 although some groups have maintained that the addition of radiation to chemotherapy may not be necessary.360,363-365 The initial report on the use of high-dose methotrexate followed by cranial irradiation from Memorial Sloan­Kettering Cancer Center indicated a median cause-specific survival time of 42 months.363 An update of the Memorial Sloan-Kettering experience, however, indicated that the risk of treatment-related neurotoxicity after whole-brain irradiation was considerable—75% for patients older than 60 years and 26% for patients younger than 60 years.366 The authors of that report recommended deferring radiation therapy for older patients until relapse to minimize treatment-related toxicity.366 Because most of the older patients in that study who received chemotherapy alone died of disease progression and because no gain in survival was attained from the addition of radiation, the impact of treatment-related neurotoxicity was thought to be roughly equivalent to the rate of disease progression when radiation was omitted. A radiation dose-response relationship seems to exist for primary CNS lymphoma; in one study, reducing the radiation dose from 45 Gy to 30 Gy led to an increased rate of relapse and shorter overall survival time.367 However, a study conducted by the Radiation Therapy Oncology Group showed that adding a 20-Gy boost to an original whole-brain irradiation dose of 40 Gy did not have a positive influence on outcome.368 Several groups are testing other chemotherapy agents in combination with high-dose methotrexate without whole-brain irradiation to improve out­ come and reduce treatment-related neurotoxicity369-371; however, these regimens have also been associated with ­significant neurotoxicity, which raises the question of whether the radiation is solely responsible for the neurotoxicity. No formal randomized trials been conducted to address this question, and the trials that have been done have not necessarily included formal assessments of neurocognitive function before therapy or accounted for ­confounding

902

PART 11  Leukemias and Lymphomas

influences such as the relatively higher prevalence of cognitive disorders among older patients or the bias in giving radiation preferentially to patients who do not respond or show only partial response to chemotherapy. Moreover, some trials have reported acceptably low risks of neurotoxicity from high-dose methotrexate followed by wholebrain irradiation.372 In terms of radiation therapy techniques, the portion of the orbits posterior to the lenses of the eyes should be included in the planning target volume. If slit-lamp examination or vitrectomy reveals involvement of the vitreous humor, the anterior chambers of the eyes should also be included in the radiation therapy fields.373 Methotrexate should not be given after whole-brain irradiation because of the potential for supra-additive neurotoxicity.374 Because of the shortcomings of the available data on treatment approaches, many oncologists recommend high-dose methotrexate-based induction chemotherapy followed by whole-brain irradiation as the initial treatment for patients with good performance status. Disabled patients requiring special care and assistance can be treated with a palliative course of whole-brain irradiation.366

Acquired Immunodeficiency Syndrome–Related Non-Hodgkin Lymphoma No standard therapy exists for AIDS-related non-Hodgkin lymphoma. Although complete response rates to chemotherapy are initially high, many patients later die of opportunistic infections or other AIDS-related complications.375 Recent improvements in overall survival (relative to that for historical controls) since the introduction of highly ­active antiretroviral therapy376,377 seem to reflect preservation of immune function and lower rates of opportunistic infections rather than any increase in the response to ­chemotherapy.

for extramedullary plasmacytomas of the oral cavity, oropharynx, or nasopharynx. However, regional lymph nodes are not included in the clinical target volume for extramedullary plasmacytomas of the maxillary sinus or nasal ­cavity. Radiation therapy results in a median survival of 9.5 years.378 Solitary Bone Plasmacytoma. Solitary plasmacytomas in bone occur in approximately 2% to 5% of patients with myeloma. These rare lesions are typically lytic and most ­often located in a vertebra. The criteria for staging, the natural history after radiation treatment, and the potential curability of solitary bone plasmacytoma are all controversial, in large part because most of the available information has come from retrospective studies involving small numbers of patients and inconsistent treatment strategies. More than 50% of patients with a solitary bone plasmacytoma eventually develop multiple myeloma.384,386 The staging workup should include magnetic resonance imaging, which can be useful for detecting occult lesions that are otherwise undetectable with standard staging procedures such as bone surveys.387,388 The presence of myeloma protein in the serum before therapy is not associated with the risk of progression to multiple myeloma, but the persistence of myeloma protein after radiation indicates a higher risk of subsequently developing multiple myeloma.388,389 Radiation therapy can produce high local control rates, but as is true for extramedullary lesions, the recommended dose varies among investigators. M.   D. Anderson recommends 45 to 50 Gy in 25 fractions.388 Mendenhall and colleagues390 at the University of Florida found through a retrospective dose-response analysis that the minimum required dose was 40 Gy, whereas Tsang and colleagues391 at the Princess Margaret Hospital in Toronto found that a median dose of 35 Gy was sufficient, especially for tumors smaller than 5 cm.

Solitary Plasmacytomas Extramedullary Plasmacytoma. Extramedullary plasmacytoma is a rare type of cancer characterized by the ­presence of clonal CD38+ plasma cells outside the bone marrow.55 Extramedullary plasmacytomas tend to arise in the mucosa of the head and neck region. The results achieved with radiation therapy at M.   D. Anderson were updated in 1999 by Liebross and colleagues.378 Radiation therapy resulted in a 5-year local control rate of 95% and a 5-year freedom from progression rate of 56%, suggesting that ­approximately one third of cases diagnosed as extramedullary plasmacytoma may have occult malignant cells in the bone marrow outside the radiation therapy fields (i.e., undiagnosed multiple myeloma). Dose-response analyses have been hampered by small numbers of patients (often fewer than 25) and high local control rates (92% to 100%) in published reports.378-384 Current practice at M.   D. Anderson consists of the delivery of 45 Gy in 25 fractions over 5 weeks, with the acknowledgment that the optimal dose of radiation remains unclear. The clinical target volume, like the dose, is controversial. At some institutions, the regional lymph nodes are electively irradiated,383,384 whereas at other institutions, they are not.380,385 At M.   D. Anderson, the decision regarding elective nodal irradiation is based on the guidelines for a squamous cell carcinoma arising at the same site.378 For example, regional lymph nodes are electively irradiated

REFERENCES    1. Pritchard SL, Rogers PCJ. Rationale and recommendations for the irradiation of blood products. Crit Rev Oncol Hematol 1987;7:115-138.    2. Schiffer LM, Chanana AD, Cronkite EP, et al. Extracorporeal irradiation of the blood. Semin Hematol 1966;3:154-167.    3. Dealy DB Jr, Tubiana M. Hematological responses to inhomogeneous and homogeneous whole-body irradiation. Ann NY Acad Sci 1964;114:268-283.    4. Tubiana M, Frindel E, Croizat H, et al. Effects of radiations on bone marrow. Pathol Biol 1979;27:326-334.    5. Tubiana M, Lalanne CM, Surmont J. Total body irradiation for organ transplantation. Proc R Soc Med 1961;54:1143-1153.    6. Tubiana M, Lalanne CM. [Hematologic course of patients subjected to total body irradiation for organ transplantation] [in French]. Ann de Radiol 1963;6:561-580.    7. Shank B, Andreeff M, Li D. Cell survival kinetics peripheral blood and bone marrow during total body irradiation for marrow transplantation. Int J Radiat Oncol Biol Phys 1983;9: 1613-1623.    8. Bortin MM, Gale RP, Rimm AA. Allogeneic bone marrow transplantation for 144 patients with severe aplastic anemia. JAMA 1981;245:1132-1139.    9. Champlin RE, Gale RP. Role of bone marrow transplantation in the treatment of hematologic malignancies and solid tumors: critical review of syngeneic, autologous, and allogeneic transplants. Cancer Treat Rep 1984;68:145-161.   10. Santos GW. Bone marrow transplantation in leukemia-current status. Cancer 1984;54:2732-2740.

34  Leukemias and Lymphomas   11. Bloom W. Histopathology of Irradiation from External and Internal Sources. New York: McGraw-Hill, 1948.   12. Knospe WH, Blom J, Crosby WH, et al. Regeneration of locally irradiated bone marrow: I. Dose dependent, long-term changes in the rat, with particular emphasis on vascular and stromal reaction. Blood 1966;28:398-415.   13. Fishburn RI, Dobelbower RR, Patchefsky AS, et al. The effect of age on the long-term response of bone marrow to local fractionated irradiation. Cancer 1975;35:1685-1691.   14. Lehar TJ, Kiely JM, Pease GL, et al. Effect of local irradiation on human bone marrow. Am J Roentgenol 1966;96:183-190.   15. Morardet N, Parmentier C, Flamant R. Etude par le fer 59 des effete de la radiothérapie étendue des hématosarcomes sur l’erythropoiese [in French]. Biomedicine 1973;18:228-234.   16. Morardet N, Parmentier C, Hayat M, et al. Effects of ­radiotherapy on the bone marrow granulocytic stem cells of patients with ­malignant lymphomas, part II: long-term effects. Int J Radiat ­Oncol Biol Phys 1978;4:853-857.   17. Ellis RE. The distribution of active bone marrow in the adult. Phys Med Biol 1961;5:255-258.   18. Atkinson HR. Bone marrow distribution as a factor in estimating radiation to the blood-forming organs: a survey of present knowledge. J Coll Radiol Aust 1962;6:149-154.   19. Sykes MP, Chu FC, Saval H, et al. The effects of varying dosages of irradiation upon sternal marrow regeneration. Radiology 1964;83:1084-1088.   20. Sykes MP, Chu FC, Wilkerson WG. Local bone marrow changes secondary to therapeutic irradiation. Radiology 1960;75:919-924.   21. Rubin P, Constine L, Bennett JM. Hodgkin’s disease IIB or not to-be using irradiation alone or in combination with chemotherapy? That is the question! J Clin Oncol 1986;4:455-457.   22. Rubin P, Landman S, Mayer E, et al. Bone marrow regeneration and extension after extended field irradiation in Hodgkin’s disease. Cancer 1973;32:699-711.   23. Sacks EL, Goris ML, Glatstein E, et al. Bone marrow regeneration following large field radiation. Cancer 1978;42:1057-1065.   24. Greenberg ML, Chanana AD, Cronkite EP, et al. Extracorporeal irradiation of blood in man: radiation resistance of circulating platelets. Radiat Res 1968;35:147-154.   25. Del Regato JA. Total body irradiation in the treatment of chronic lymphogenous leukemia: Janeway Lecture, 1973. Am J Roentgenol Radium Ther Nucl Med 1974;120:504-520.   26. De Bruyn PPH. Lymph node and intestinal lymphatic tissue. In Bloom W (ed): Histopathology of Irradiation. New York: ­McGraw-Hill, 1948.   27. Engeset A. Irradiation of lymph nodes and vessels. Experiments in rats, with reference to cancer therapy. Acta Radiol Diagn (Stockh) 1964;229(Suppl):1.   28. Teneff S, Stoppani F. L’influenza delle irradiazioni sulle linfoghiandole e sulfa circolazione linfatica [in Italian]. Radiol Med 1935;22:768-787.   29. Dettman PN, King ER, Zimberg YH. Evaluation of lymph node function following irradiation or surgery. Am J Roentgenol 1966;96:711-718.   30. Fisher B, Fisher ER. Barrier functions of lymph node to ­tumor cells and erythrocytes. Normal nodes. Cancer 1967;20(Pt 1): 1907-1913.   31. Fisher B, Fisher ER. Barrier function of lymph node to tumor cells and erythrocytes. Effect of x-ray, inflammation, sensitization, and tumor growth. Cancer 1967;20(Pt 2):1914-1919.   32. O’Brien PH, Moss WT, Ujiki GT, et al. Effect of irradiation on tumor-infused lymph nodes. Radiology 1970;94:407-411.   33. Sherman JO, O’Brien PH. Effects of ionizing irradiation on normal lymphatic vessels and lymph nodes. Cancer 1967;20: 1851-1858.   34. Lenzi M, Bassani G. The effect of radiation on the lymph and on the lymph vessels. Radiology 1963;80:814-817.   35. Murray RG. The spleen. In Bloom W, (ed): Histopathology of ­Irradiation. New York: McGraw-Hill, 1948.   36. Dailey MO, Coleman CN, Kaplan HS. Radiation-induced splenic atrophy in patients with Hodgkin’s disease and nonHodgkin’s lymphomas. N Engl J Med 1980;302:215-217.

903

  37. United Nations Scientific Committee on the Effects of ­Atomic Radiation. Genetic and Somatic Effects of Ionizing Radiation: 1986 Report to the General Assembly. New York: United ­Nations, 1986.   38. Von Jagic N, Schwarz G, von Siebenrock L. Blutefunde bei Rontgenologen. Klin Wochenschr 1911;48:1220-1222.   39. March HC. Leukemia in radiologists in a 20 year period. Am J Med Sci 1950;220:282-286.   40. March HC. Leukemia in radiologists, 10 years later. Am J Med Sci 1961;242:137-149.   41. Logue JN, Barrick MK, Jessup TG. Mortality of radiologists and pathologists in the radiation registry of physicians. J Occup Med 1986;28:91-99.   42. Court-Brown WMC, Abbatt JD. The incidence of leukemia in ankylosing spondylitis treated with x-ray: a preliminary report. Lancet 1955;1:1283-1285.   43. Court-Brown WM, Doll R. Leukemia and aplastic anemia in patients irradiated for ankylosing spondylitis. Spec Rep Ser Med Res Counc (G B) 1957;295:1-135.   44. Wakabayashi T, Kato H, Ikeda T, et al. Studies of the mortality of A-bomb survivors, report 7: incidence of cancer in 1959-1978, based on the tumor register, Nagasaki. Radiat Res 1983;93(Pt 3):112-146.   45. Levine EG, Bloomfield CD. Leukemias and myelodysplastic syndromes secondary to drug, radiation, and environmental exposure. Semin Oncol 1992;19:47-84.   46. Coltman CA Jr, Dixon DO. Second malignancies complicating Hodgkin’s disease: a Southwest Oncology Group 10-year ­follow-up. Cancer Treat Rep 1982;66:1023-1033.   47. Berk PD, Goldberg JD, Silverstein MN, et al. Increased incidence of acute leukemia in polycythemia vera associated with chlorambucil therapy. N Engl J Med 1981;304:441-447.   48. Pinkel D. General Motors Cancer Research Foundation Prizes: Charles F. Kettering Prize: curing children of leukemia. Cancer 1987;59:1683-1691.   49. Pinkel D. The ninth annual David Karnofsky Lecture: treatment of acute lymphocytic leukemia. Cancer 1979;43:1128-1137.   50. Bleyer WA, Harland S, Coccio P, et al. The staging of childhood acute lymphoblastic leukemia: strategies of the Children’s Cancer Study Group and a 3-dimensional technic of multivariate analysis. Med Pediatr Oncol 1986;14:271-280.   51. Clarkson BD, Gee T, Mertelsmann R, et al. Current status of treatment of acute leukemia in adults: an overview of the ­Memorial experience and review of literature. Crit Rev Oncol Hematol 1986;4:221-248.   52. Ha CS, Chung WK, Koller CA, et al. Role of radiation therapy to the brain in leukemic patients with cranial nerve palsies in the absence of radiological findings. Leuk Lymphoma 1999;32: 497-503.   53. Cossman J, Annunziata CM, Barash S, et al. Reed-Sternberg cell genome expression supports a B-cell lineage. Blood 1999;94: 411-416.   54. Cox J, Ha C. Prevention of diagnostic and treatment-related ­sequelae in stage I and II Hodgkin’s disease. In Meyer J (ed): ­Frontiers of Radiation Therapy and Oncology. Basel: Karger, 1999, pp 85-97.   55. Harris NL, Jaffe ES, Stein H, et al. A Revised European­American classification of lymphoid neoplasms: a proposal from the International Lymphoma Study Group. Blood 1994;84: 1361-1392.   56. Harris NL, Jaffe ES, Diebold J, et al. The World Health Organization classification of neoplastic diseases of the haematopoietic and lymphoid tissues: report of the Clinical Advisory Committee Meeting, Airlie House, VA, November 1997. Hematol J 2000;1:53-66.   57. Harris NL. The many faces of Hodgkin’s disease around the world: what have we learned from its pathology? Ann Oncol 1998;9(Suppl 5):S45-S56.   58. American Cancer Society. Cancer Facts & Figures, 2008. Atlanta: American Cancer Society, 2008.   59. Mack TM, Cozen W, Shibata DK, et al. Concordance for Hodgkin’s disease in identical twins suggesting genetic susceptibility to the young-adult form of the disease. N Engl J Med 1995;332:413-418.

904

PART 11  Leukemias and Lymphomas

  60. Mueller N, Evans A, Harris NL, et al. Hodgkin’s disease and Epstein-Barr virus. Altered antibody pattern before diagnosis. N Engl J Med 1989;320:689-695.   61. Chapman AL, Rickinson AB. Epstein-Barr virus in Hodgkin’s disease. Ann Oncol 1998;9(Suppl 5):S5-S16.   62. Carbone P, Kaplan H, Musshoff K. Report of the committee on Hodgkin’s disease classification. Cancer Res 1971;31. 1869-1861.   63. Lister TA, Crowther D, Sutcliffe SB, et al. Report of a committee convened to discuss the evaluation and staging of patients with Hodgkin’s disease: Cotswolds meeting [published erratum appears in J Clin Oncol 1990;8:1602]. J Clin Oncol 1989;7: 1630-1636.   64. Kaplan HS, Rosenberg SA. The treatment of Hodgkin’s disease. Med Clin North Am 1966;50:1591-1610.   65. Greene FL, Page DL, Fleming ID, et al (eds). Lymphoid neoplasms. In AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 383-406.   66. Gilbert R. Radiotherapy in Hodgkin’s disease (malignant granulomatosis): anatomic and clinical foundations; governing principles; results. Am J Roentgenol Radium Ther Nucl Med 1939;41:198-241.   67. Peters MV. A study in survivals in Hodgkin’s disease treated ­radiologically. Am J Roentgenol 1950;63:299-311.   68. Easson E, Russell M. The cure of Hodgkin’s disease. Br Med J 1963;1:1704-1707.   69. Rosenberg SA, Kaplan HS, Hoppe RT, et al. The Stanford randomized trials of the treatment of Hodgkin’s disease 1967-1980. In Rosenberg SA, Kaplan HS (eds): Malignant Lymphomas: Etiology, Immunology, Pathology, Treatment. New York: Academic Press, 1982, pp 513-522.   70. Devessa S, Silverman D, Young J, et al. Cancer incidence and mortality trends among whites in the United States, 1947-1984. J Natl Cancer Inst 1987;79:701-770.   71. Gospodarowicz MK, Sutcliffe SB, Clark RM, et al. Analysis of supradiaphragmatic clinical stage I and II Hodgkin’s disease treated with radiation alone. Int J Radiat Oncol Biol Phys 1992;22: 859-865.   72. Noordijk EM, Carde P, Hagenbeek A, et al. Combination of radiotherapy and chemotherapy is advisable in all patients with clinical stage I-II Hodgkin’s disease. Six-year results of the EORTCGPMC controlled clinical trials H7-VF, H7-F and H7-U [abstract 177]. Int J Radiat Oncol Biol Phys 1997;39:173.   73. Mauch PM, Kalish LA, Marcus KC, et al. Long-term survival in Hodgkin’s disease: relative impact of mortality, second ­tumors, infection, and cardiovascular disease. Cancer J Sci Am 1995;1: 33-42.   74. Tucker MA, Coleman CN, Cox RS, et al. Risk of second cancers after treatment for Hodgkin’s disease. N Engl J Med 1988; 318:76-81.   75. Henry-Amar M. Second cancer after the treatment for Hodgkin’s disease: a report from the International Database on Hodgkin’s disease. Ann Oncol 1992;3:117-128.   76. Noordijk EM, Carde P, Mandard AM, et al. Preliminary results of the EORTC-GPMC controlled clinical trial H7 in early-stage Hodgkin’s disease. EORTC Lymphoma Cooperative Group and Groupe Pierre-et-Marie-Curie. Ann Oncol 1994;5:107-112.   77. Carde P, Burgers JM, Henry-Amar M, et al. Clinical stages I and II Hodgkin’s disease: a specifically tailored therapy according to prognostic factors. J Clin Oncol 1988;6:239-252.   78. Wolf J, Tesch H, Parsa-Parsi R, et al. Current clinical trials for the treatment of adult Hodgkin’s disease: common strategies and perspectives. Ann Oncol 1998;9:S79-S82.   79. Thomas J, Fermé C, Noordijk EM, et al. Six courses of EBVP followed by 36 Gy involved-field radiotherapy vs no irradiation in favourable supradiaphragmatic clinical stage I-II Hodgkin’s lymphoma: the EORTC-GELA strategy in 771 patients (H9-F trial-20982) [abstract E11a]. Eur J Haematol 2004;73 (Suppl 65):40.   80. Vijayakumar S, Myrianthopoulos LC. An updated dose-response analysis in Hodgkin’s disease. Radiother Oncol 1992;24:1-13.   81. Seydel HG, Bloedorn FG, Wizenberg MJ. Time-dose-volume ­relationship in Hodgkin’s disease. Radiology 1967;89:919-922.

  82. Mendenhall NP, Rodrigue LL, Moore-Higgs GJ, et al. The optimal dose of radiation in Hodgkin’s disease: an analysis of clinical and treatment factors affecting in-field disease control. Int J Radiat Oncol Biol Phys 1999;44:551-561.   83. Kaplan HS. Evidence for a tumoricidal dose level in the radiotherapy of Hodgkin’s disease. Cancer Res 1966;26:1221-1224.   84. Jacobs ML, Marasso FJ. Lymphomas: relationship between dosage and recurrence. Radiology 1964;88:106-107.   85. Scott RM, Brizel HE. Time-dose relationships in Hodgkin’s disease. Radiology 1964;82:1043-1049.   86. Raubitschek A, Glatstein E. The never-ending controversies in Hodgkin’s disease. Int J Radiat Oncol Biol Phys 1989;17: 1115-1117.   87. Fletcher GH, Shukovsky LJ. The interplay of radiocurability and tolerance in the irradiation of human cancers. J Radiol Electrol 1975;56:383-400.   88. Brincker H, Bentzen SM. A re-analysis of available dose-­response and time-dose data in Hodgkin’s disease. Radiother ­ Oncol 1994;30:227-230.   89. Hanks GE, Kinzie JJ, White RL, et al. Patterns of Care Outcome Studies: results of the national practice in Hodgkin’s disease. Cancer 1983;51:569-573.   90. Thar TL, Million RR, Hausner RJ, et al. Hodgkin’s disease, stages I and II: relationship of recurrence to size of disease, radiation dose, and number of sites involved. Cancer 1979;43:1101-1105.   91. Schewe KL, Reavis J, Kun LE, et al. Total dose, fraction size, and tumor volume in the local control of Hodgkin’s disease. Int J Radiat Oncol Biol Phys 1988;15:25-28.   92. Sears JD, Greven KM, Ferree CR, et al. Definitive irradiation in the treatment of Hodgkin’s disease. Analysis of outcome, prognostic factors, and long-term complications. Cancer 1997;79:145-151.   93. Schewe KL, Kun LE, Cox JD. A step toward ending the controversies in Hodgkin’s disease. Int J Radiat Oncol Biol Phys 1989;17:1123-1124.   94. Duhmke E, Diehl V, Loeffler M, et al. Randomized trial with early-stage Hodgkin’s disease testing 30 Gy vs. 40 Gy ­extended field radiotherapy alone. Int J Radiat Oncol Biol Phys 1996;36:305-310.   95. Vlachaki MT, Ha CS, Hagemeister FB, et al. Stage I Hodgkin disease: radiation therapy and chemotherapy at the University of Texas M. D. Anderson Cancer Center, 1996-1997. Radiology 1998;208:739-747.   96. Boivin JF, Hutchinson GB, Lubin JH, et al. Coronary artery disease mortality in patients treated for Hodgkin’s disease. Cancer 1992;69:1241-1247.   97. Hancock SL, Tucker MA, Hoppe RT. Factors affecting late mortality from heart disease after treatment of Hodgkin’s disease. JAMA 1993;270:1949-1955.   98. Swerdlow A, Douglas A, Vaughan HG, et al. Risk of second primary cancers after Hodgkin’s disease by type of treatment: analysis of 2486 patients in the British National Lymphoma ­Investigation. BMJ 1992;304:1137-1143.   99. Hancock S, Tucker M, Hoppe R. Breast cancer after treatment of Hodgkin’s disease. J Natl Cancer Inst 1993;85:25-31. 100. Aisenberg AC, Finkelstein DM, Doppke KP, et al. High risk of breast carcinoma after irradiation of young women with Hodgkin’s disease. Cancer 1997;79:1203-1210. 101. Bhatia S, Robison LL, Oberlin O, et al. Breast cancer and other second neoplasms after childhood Hodgkin’s disease. N Engl J Med 1996;334:745-751. 102. Vlachaki MT, Ha CS, Hagemeister FB, et al. Long-term outcome of treatment for Ann Arbor stage 1 Hodgkin’s disease: patterns of failure, late toxicity and second malignancies. Int J Radiat ­Oncol Biol Phys 1997;39:609-616. 103. Nyandoto P, Muhonen T, Joensuu H. Second cancer among long-term survivors from Hodgkin’s disease. Int J Radiat Oncol Biol Phys 1998;42:373-378. 104. Carde P, Hagenbeek A, Hayat M, et al. Clinical staging versus laparotomy and combined modality with MOPP versus ABVD in early-stage Hodgkin’s disease: the H6 twin randomized trials from the European Organization for Research and Treatment of Cancer Lymphoma Cooperative Group. J Clin Oncol 1993;11:2258-2272.

34  Leukemias and Lymphomas 105. Liao Z, Ha CS, Vlachaki MT, et al. Mantle irradiation alone for pathologic stage I and II Hodgkin’s disease: long-term follow-up and patterns of failure. Int J Radiat Oncol Biol Phys 2001;50:971-977. 106. Jones E, Mauch P. Limited radiation therapy for selected patients with pathological stage IA and IIA Hodgkin’s disease. Semin Radiat Oncol 1996;6:162-171. 107. Backstrand KH, Ng AK, Takvorian RW, et al. Results of a prospective trial of mantle irradiation alone for selected patients with early-stage Hodgkin’s disease. J Clin Oncol 2001;19:736-741. 108. Biti GP, Cimino G, Cartoni C, et al. Extended-field radiotherapy is superior to MOPP chemotherapy for the treatment of pathologic stage I-IIA Hodgkin’s disease: eight-year update of an Italian prospective randomized study. J Clin Oncol 1992;10:378-382. 109. Longo DL, Glatstein E, Duffey PL, et al. Radiation therapy versus combination chemotherapy in the treatment of early-stage Hodgkin’s disease: seven-year results of a prospective randomized trial. J Clin Oncol 1991;9:906-917. 110. Jox A, Sieber M, Wolf J, et al. Hodgkin’s disease-new treatment strategies toward the cure of patients. Cancer Treat Rev 1999;25:169-176. 111. Horning SJ, Hoppe RT, Breslin S, et al. Stanford V and radiotherapy for locally extensive and advanced Hodgkin’s disease: mature results of a prospective clinical trial. J Clin Oncol 2002;20:630-637. 112. Rodriguez MA, Fuller LM, Zimmerman SO, et al. Hodgkin’s disease: study of treatment intensities and incidences of second malignancies. Ann Oncol 1993;4:125-131. 113. Fermé C, Eghbali H, Habenbeek A, et al. MOPP/ABV (M/A) hybrid and irradiation in unfavorable supradiaphragmatic clinical stages I-II HD: comparison of three treatment modalities, preliminary results of the EORTC/GELA H8-U randomized trial in 995 patients [abstract]. Blood 2000;96:576a. 114. Bonadonna G, Bonfante V, Viviani S, et al. ABVD plus subtotal nodal versus involved-field radiotherapy in early-stage Hodgkin’s disease: long-term results. J Clin Oncol 2004;22:2835-2841. 115. Engert A, Schiller P, Josting A, et al. Involved-field radiotherapy is equally effective and less toxic compared with extended-field radiotherapy after four cycles of chemotherapy in patients with early-stage unfavorable Hodgkin’s lymphoma: results of the HD8 trial of the German Hodgkin’s Lymphoma Study Group. J Clin Oncol 2003;21:3601-3608. 116. Diehl V, Franklin J, Hasenclever D, et al. BEACOPP, a new dose-escalated and accelerated regimen, is at least as effective as COPP/ABVD in patients with advanced-stage Hodgkin’s lymphoma: interim report from a trial of the German Hodgkin’s Lymphoma Study Group. J Clin Oncol 1998;16:3810-3821. 117. Diehl V, Franklin J, Pfreundschuh M, et al. Standard and increaseddose BEACOPP chemotherapy compared with COPP-ABVD for advanced Hodgkin’s disease. N Engl J Med 2003;348:2386-2395. 118. Horning SJ, Williams J, Bartlett NL, et al. Assessment of the Stanford V regimen and consolidative radiotherapy for bulky and advanced Hodgkin’s disease: Eastern Cooperative Oncology Group pilot study E1492. J Clin Oncol 2000;18:972-980. 119. Santoro A, Bonadonna G, Valagussa P, et al. Long-term results of combined chemotherapy-radiotherapy approach in Hodgkin’s disease: superiority of ABVD plus radiotherapy versus MOPP plus radiotherapy. J Clin Oncol 1987;5:27-37. 120. Canellos GP, Anderson JR, Propert KJ, et al. Chemotherapy of advanced Hodgkin’s disease with MOPP, ABVD, or MOPP alternating with ABVD. N Engl J Med 1992;327:1478-1484. 121. Glick JH, Young ML, Harrington D, et al. MOPP/ABV hybrid chemotherapy for advanced Hodgkin’s disease significantly ­improves failure-free and overall survival: the 8-year results of the Intergroup trial. J Clin Oncol 1998;16:19-26. 122. Duggan DB, Petroni GR, Johnson JL, et al. Randomized comparison of ABVD and MOPP/ABV hybrid for the treatment of advanced Hodgkin’s disease: report of an Intergroup trial. J Clin Oncol 2003;21:607-614. 123. Fabian CJ, Mansfield CM, Dahlberg S, et al. Low-dose involved field radiation after chemotherapy in advanced Hodgkin disease. A Southwest Oncology Group randomized study. Ann Intern Med 1994;120:903-912.

905

124. Diehl V, Loeffler M, Pfreundschuh M, et al. Further chemotherapy versus low-dose involved-field radiotherapy as consolidation of complete remission after six cycles of alternating chemotherapy in patients with advanced Hodgkin’s disease. German Hodgkin’s Lymphoma Study Group (GHSG). Ann Oncol 1995;6:901-910. 125. Ferme C, Sebban C, Hennequin C, et al. Comparison of chemotherapy to radiotherapy as consolidation of complete or good partial response after six cycles of chemotherapy for patients with advanced Hodgkin’s disease: results of the Groupe d’Études des Lymphomes de l’Adulte H89 trial. Blood 2000;95:2246-2252. 126. Raemaekers J, Burgers M, Henry-Amar M, et al. Patients with stage III/IV Hodgkin’s disease in partial remission after MOPP/ ABV chemotherapy have excellent prognosis after additional involved-field radiotherapy: interim results from the ongoing EORTC-LCG and GPMC phase III trial. The EORTC Lymphoma Cooperative Group and Groupe Pierre-et-Marie-Curie. Ann Oncol 1997;8:111-114. 127. Aleman BM, Raemaekers JM, Tirelli U, et al. Involved-field ­radiotherapy for advanced Hodgkin’s lymphoma. N Engl J Med 2003;348:2396-2406. 128. Aleman BM, Raemaekers JM, Tomisic R, et al. Involved-field radiotherapy for patients in partial remission after chemotherapy for advanced Hodgkin’s lymphoma. Int J Radiat Oncol Biol Phys 2007;67:19-30. 129. Hasenclever D, Diehl V, Armitage JO, et al. A prognostic score for advanced Hodgkin’s disease. International Prognostic Factors Project on Advanced Hodgkin’s Disease. N Engl J Med 1998;339:1506-1514. 130. Krikorian JG, Portlock CS, Mauch PM. Hodgkin’s disease presenting below the diaphragm: a review. J Clin Oncol 1986;4: 1551-1562. 131. Roos DE, O’Brien PC, Wright J, et al. Treatment of subdiaphragmatic Hodgkin’s disease: is radiotherapy alone appropriate only for inguino-femoral presentations?. Int J Radiat Oncol Biol Phys 1994;28:683-691. 132. Leibenhaut MH, Hoppe RT, Varghese A, et al. Subdiaphragmatic Hodgkin’s disease: laparotomy and treatment results in 49 patients. J Clin Oncol 1987;5:1050-1055. 133. Liao Z, Ha CS, Fuller LM, et al. Subdiaphragmatic stage I and II Hodgkin’s disease: long-term follow-up and prognostic factors. Int J Radiat Oncol Biol Phys 1998;41:1047-1056. 134. Regula D, Hoppe RT, Weiss L. Nodular and diffuse types of lymphocyte predominance Hodgkin’s disease. N Engl J Med 1988;318:214-219. 135. Ha CS, Kavadi V, Dimopoulos MA, et al. Hodgkin’s disease with lymphocyte predominance: long-term results based on current histopathologic criteria. Int J Radiat Oncol Biol Phys 1999;43:329-334. 136. Diehl V, Sextro M, Franklin J, 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;17:776-783. 137. Bodis S, Kraus M, Pinkus G, et al. Clinical presentation and ­outcome in lymphocyte-predominant Hodgkin’s disease. J Clin Oncol 1997;15:3060-3066. 138. Sundeen JT, Cossman J, Jaffe ES. Lymphocyte-predominant Hodgkin’s disease nodular subtype with coexistent large cell lymphoma-histological progression or composite malignancy? Am J Surg Pathol 1988;12:599-606. 139. Pergolizzi S, Settineri N, Gaeta M, et al. What is the best position of the arms in mantle field for Hodgkin’s disease? Int J Radiat Oncol Biol Phys 2000;46:119-122. 140. Plowman PN. Radiotherapy considerations in patients with Hodgkin’s disease who receive mediastinal radiotherapy and anthracycline-containing chemotherapy. Clin Oncol 1998;10: 384-391. 141. Rabkin CS, Devesa SS, Zahm SH, et al. Increasing incidence of non-Hodgkin’s lymphoma. Semin Hematol 1993;30:286-296. 142. Grulich AE, Vajdic CM. The epidemiology of non-Hodgkin lymphoma. Pathology 2005;37:409-419.

906

PART 11  Leukemias and Lymphomas

143. Wotherspoon AC, Ortiz-Hidalgo C, Falzon MR, et al. Helicobacter pylori–associated gastritis and primary B-cell gastric lymphoma. Lancet 1991;338:1175-1176. 144. Isaacson PG, Spencer J. Gastric lymphoma and Helicobacter ­pylori. Important Adv Oncol 1996:111-121. 145. Epstein MA, Achong BG, Barr YM. Virus particles in cultured lymphoblasts from Burkitt’s lymphoma. Lancet 1964;1:702-703. 146. Magrath I, Jain V, Bhatia K. Epstein-Barr virus and Burkitt’s lymphoma. Semin Cancer Biol 1992;3:285-295. 147. Penn I. Cancers complicating organ transplantation. N Engl J Med 1990;323:1767-1769. 148. Swinnen LJ, Costanzo-Nordin MR, Fisher SG, et al. Increased incidence of lymphoproliferative disorder after immunosuppression with the monoclonal antibody OKT3 in cardiac-transplant recipients. N Engl J Med 1990;323:1723-1728. 149. Skinner JC, Gilbert EF, Hong R, et al. B cell lymphoproliferative disorders following T-cell-depleted allogeneic bone marrow transplantation. Am J Pediatr Hematol Oncol 1988;10:112. 150. Lopez-Guillermo A, Cabanillas F, McDonnell TI, et al. Correlation of bcl-2 rearrangement with clinical characteristics and outcome in indolent follicular lymphoma. Blood 1999;93:3081-3087. 151. Neri A, Barriga F, Knowles DM, et al. Different regions of the immunoglobulin heavy-chain locus are involved in chromosomal translocations in distinct pathogenetic forms of Burkitt lymphoma. Proc Natl Acad Sci USA 1988;85:2748-2752. 152. DeCoteau JF, Butmarc JR, Kinney MC, et al. The t(2;5) chromosomal translocation is not a common feature of primary cutaneous CD30+ lymphoproliferative disorders: comparison with anaplastic large-cell lymphoma of nodal origin. Blood 1996;87:3437-3441. 153. Medeiros LJ, Van Krieken JH, Jaffe ES, et al. Association of bcl-1 rearrangements with lymphocytic lymphoma of intermediate differentiation. Blood 1990;76:2086-2090. 154. Garcia-Conde J, Cabanillas F. Mantle cell lymphoma: a lymphoproliferative disorder associated with aberrant function of the cell cycle. Leukemia 1996;10(Suppl 2):S78-S83. 155. de Boer CJ, van Krieken JH, Schuuring E, et al. Bcl-1/cyclin D1 in malignant lymphoma. Ann Oncol 1997;8(Suppl 2):109-117. 156. Rappaport H, Winter WI, Hicks EB. Follicular lymphoma: a reevaluation of its position in the scheme of malignant lymphoma, based on a survey of 253 cases. Cancer 1954;9:792. 157. Rappaport H. Tumors of the hematopoietic system. Atlas of Tumor Pathology, Section III, Fascicle 8. Washington, DC: Armed Forces Institute of Pathology, 1966. 158. Lukes RJ, Collins RD. Immunologic characterization of human malignant lymphomas. Cancer 1974;34(Suppl):1488-1503. 159. Lukes RJ, Collins RD. New approaches to the classification of the lymphomata. Br J Cancer Suppl 1975;2:1-28. 160. Lennert K, Mohri N, Stein H, et al. The histopathology of malignant lymphoma. Br J Haematol 1975;31(Suppl):193. 161. Lennert K. Malignant Lymphomas Other Than Hodgkin’s Disease. New York: Springer-Verlag, 1978. 162. Stansfeld A, Diebold J, Kapanci Y, et al. Updated Kiel classification of lymphomas. Lancet 1988;1:292-293. 163. Classification of non-Hodgkin’s lymphomas. Reproducibility of major classification systems. NCI non-Hodgkin’s Classification Project Writing Committee. Cancer 1985;55:91-95. 164. National Cancer Institute-sponsored study of classifications of non-Hodgkin’s lymphomas: summary and description of a working formulation for clinical usage. The Non-Hodgkin’s Lymphoma Pathologic Classification Project. Cancer 1982;49:2112-2135. 165. Fisher RI, Miller TP, Grogan TM. New REAL clinical entities. Cancer J Sci Am 1998;4(Suppl 2):S5-12. 166. Shipp MA, Mauch PM, Harris NL. Non-Hodgkin’s lymphomas. In DeVita VT Jr, Hellman S, Rosenberg SA (eds): Cancer: Principles and Practice of Oncology, 5th ed. Philadelphia: Lippincott-Raven, 1997, pp 2165-2220. 167. Pileri SA, Milani M, Fraternali-Orcioni G, et al. From the REAL classification to the upcoming WHO scheme: a step toward universal categorization of lymphoma entities? Ann Oncol 1998;9:607-612. 168. No authors listed. A clinical evaluation of the International Lymphoma Study Group classification of non-Hodgkin’s lymphoma. The Non-Hodgkin’s Lymphoma Classification Project. Blood 1997;89:3909-3918.

169. Sheibani K, Sohn CC, Burke JS, et al. Monocytoid B-cell lymphoma: a novel B-cell neoplasm. Am J Pathol 1986;124: 310-318. 170. Nizze H, Cogliatti SB, von Schilling C, et al. Monocytoid Bcell lymphoma: morphological variants and relationship to lowgrade B-cell lymphoma of the mucosa-associated lymphoid tissue. Histopathology 1991;18:403-414. 171. Armitage JO, Weisenburger DD. New approach to classifying non-Hodgkin’s lymphomas: clinical features of the major histologic subtypes. Non-Hodgkin’s Lymphoma Classification Project. J Clin Oncol 1998;16:2780-2795. 172. Bangerter M, Griesshammer M, Bergmann L. Progress in medical imaging of lymphoma and Hodgkin’s disease. Curr Opin ­Oncol 1999;11:339-342. 173. Kostakoglu L, Leonard JP, Kuji I, et al. Comparison of fluorine18 fluorodeoxyglucose positron emission tomography and Ga-67 scintigraphy in evaluation of lymphoma. Cancer 2002;94: 879-888. 174. Moog F, Bangerter M, Kotzerke J, et al. 18F-fluorodeoxyglucosepositron emission tomography as a new approach to detect lymphomatous bone marrow. J Clin Oncol 1998;16:603-609. 175. Carr R, Barrington SF, Madan B, et al. Detection of lymphoma in bone marrow by whole-body positron emission tomography. Blood 1998;91:3340-3346. 176. Pakos EE, Fotopoulos AD, Ioannidis JP. 18F-FDG PET for evaluation of bone marrow infiltration in staging of lymphoma: a meta-analysis. J Nucl Med 2005;46:958-963. 177. Elstrom R, Guan L, Baker G, et al. Utility of FDG-PET scanning in lymphoma by WHO classification. Blood 2003;101: 3875-3876. 178. Kostakoglu L, Leonard JP, Coleman M, et al. The role of FDGPET imaging in the management of lymphoma. Clin Adv Hematol Oncol 2004;2:115-121. 179. Tsukamoto N, Kojima M, Hasegawa M, et al. The usefulness of (18)F-fluorodeoxyglucose positron emission tomography ((18)F-FDG-PET) and a comparison of (18)F-FDG-pet with (67)gallium scintigraphy in the evaluation of lymphoma: relation to histologic subtypes based on the World Health Organization classification. Cancer 2007;110:652-659. 180. Rodriguez M, Rehn S, Ahlström H, et al. Predicting ­malignancy grade with PET in non-Hodgkin’s lymphoma. J Nucl Med 1995;36:1790-1796. 181. National Comprehensive Cancer Network. Clinical Practice Guidelines in Oncology, v.3.2008. Non-Hodgkin’s lymphomas. Available at http://www.nccn.org/ (accessed June 2008). 182. Rosenberg SA. Validity of the Ann Arbor staging classification for the non-Hodgkin’s lymphomas. Cancer Treat Rep 1977;61: 1023-1027. 183. Urba WJ, Duffey PL, Longo DL. Treatment of patients with aggressive lymphomas: an overview. J Natl Cancer Inst Monogr 1990;10:29-37. 184. Moore DF Jr, Cabanillas F. Overview of prognostic factors in non-Hodgkin’s lymphoma. Oncology 1998;12:17-24. 185. No authors listed. A predictive model for aggressive non­Hodgkin’s lymphoma: the International Non-Hodgkin’s Lymphoma Prognostic Factors Project. N Engl J Med 1993;329: 987-994. 186. Miller TP, Grogan TM, Dahlberg S, et al. Prognostic significance of the Ki-67-associated proliferative antigen in aggressive non-Hodgkin’s lymphomas: a prospective Southwest Oncology Group trial. Blood 1994;83:1460-1466. 187. Alizadeh AA, Eisen MB, Davis RE, et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 2000;403:503-511. 188. Stanley KE. Prognostic factors for survival in patients with inoperable lung cancer. J Natl Cancer Inst 1980;65:25-32. 189. van Besien K, Ha CS, Murphy S. Risk factors, treatment, and outcome of central nervous system recurrence in adults with intermediate-grade and immunoblastic lymphoma. Blood 1998;91:1178-1184. 190. Touroutoglou N, Dimopoulos MA, Younes A, et al. Testicular lymphoma: late relapses and poor outcome despite doxorubicin-based therapy. J Clin Oncol 1995;13:1361-1367.

34  Leukemias and Lymphomas 191. Coffier B, Lepage E, Briere J, 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: 235-242. 192. Miller TP, Dahlberg S, Cassady JR, et al. Chemotherapy alone compared with chemotherapy plus radiotherapy for localized intermediate- and high-grade non-Hodgkin’s lymphoma. N Engl J Med 1998;339:21-26. 193. Horning SJ, Weller E, Kim K, et al. Chemotherapy with or without radiotherapy in limited-stage diffuse aggressive non­Hodgkin’s lymphoma: Eastern Cooperative Oncology Group study 1484. J Clin Oncol 2004;22:3032-3038. 194. Miller T, Leblanc M, Spier C, et al. CHOP alone compared to CHOP plus radiotherapy for early stage aggressive non-­Hodgkin’s lymphomas: update of the Southwest Oncology Group (SWOG) randomized trial [abstract]. Blood 2001;98:724a. 195. Reyes F, Lepage E, Ganem G, et al. ACVBP versus CHOP plus radiotherapy for localized aggressive lymphoma. N Engl J Med 2005;352:1197-1205. 196. Bonnet C, Fillet G, Mounier N, et al. CHOP alone compared with CHOP plus radiotherapy for localized aggressive lymphoma in elderly patients: a study by the Groupe d’Etude des Lymphomes de l’Adulte. J Clin Oncol 2007;25:787-792. 197. Danieu L, Wong G, Koziner B, et al. Predictive model for prognosis in advanced diffuse histiocytic lymphoma. Cancer Res 1986;46:5372-5379. 198. Shipp MA, Klatt MM, Yeap B, et al. Patterns of relapse in large-cell lymphoma patients with bulk disease: implications for the use of adjuvant radiation therapy. J Clin Oncol 1989;7: 613-618. 199. Coiffier B. Treatment of aggressive non-Hodgkin’s lymphoma. Semin Oncol 1999;26:12-20. 200. Velasquez W, Fuller LM, Oh KK, et al. Combined modality therapy in stage III and stage IIIE diffuse large cell lymphomas. Cancer 1984;53:1478-1483. 201. Schlembach PJ, Wilder RB, Tucker SL, et al. Impact of involved field radiotherapy after CHOP-based chemotherapy on stage III-IV, intermediate grade and large-cell immunoblastic lymphomas. Int J Radiat Oncol Biol Phys 2000;48:1107-1110. 202. Crowther D. New approaches to the management of patients with non-Hodgkin’s lymphoma of high-grade pathology. First Gordon Hamilton-Fairley memorial lecture. Br J Cancer 1981;43:417-435. 203. Ferreri AJ, Dell’Oro S, Reni M, et al. Consolidation radiotherapy to bulky or semibulky lesions in the management of stage III-IV diffuse large B cell lymphomas. Oncology 2000;58:219-226. 204. Avilés A, Delgado S, Nambo MJ, et al. Adjuvant radiotherapy to sites of previous bulky disease in patients stage IV diffuse large cell lymphoma. Int J Radiat Oncol Biol Phys 1994;30: 799-803. 205. Landberg TG, Hakansson LG, Moller TR, et al. CVP-remission maintenance in stage I or II non-Hodgkin’s lymphomas. Preliminary results of a randomized study. Cancer 1979;44:831-838. 206. Monfardini S, Banfi A, Bonadonna G, et al. Improved five-year survival after combined radiotherapy-chemotherapy for stage I-II non-Hodgkin’s lymphoma. Int J Radiat Oncol Biol Phys 1980;6:125-134. 207. Nissen NI, Ersboll J, Hansen HS, et al. A randomized study of radiotherapy versus radiotherapy plus chemotherapy in stage III non-Hodgkin’s lymphomas. Cancer 1983;52:1-7. 208. Somers R, Burgers M, Quasim M, et al. EORTC trial non­Hodgkin’s lymphomas. Eur J Cancer Clin Oncol 1987;23: 283-293. 209. Fisher RI, Gaynor ER, Dahlberg S, et al. Comparison of a standard regimen (CHOP) with three intensive chemotherapy regimens for advanced non-Hodgkin’s lymphoma. N Engl J Med 1993;328:1002-1006. 210. Fisher RI. Cyclophosphamide, doxorubicin, vincristine, and prednisone versus intensive chemotherapy in non-Hodgkin’s lymphoma. Cancer Chemother Pharmacol 1997;40:S42-S46. 211. Stryker JA, Abt AB, Eyster MA, et al. Histopathologic subclassification of diffuse histiocytic lymphoma and response to therapy. Radiology 1982;142:501-506.

907

212. Mirza MR, Brincker H, Specht L. The integration of radiotherapy into the primary treatment of non-Hodgkin’s lymphoma. Crit Rev Oncol Hematol 1992;12:217-229. 213. Kamath SS, Marcus RB Jr, Lynch JW, et al. The impact of radiotherapy dose and other treatment-related and clinical factors on in-field control in stage I and II non-Hodgkin’s lymphoma. Int J Radiat Oncol Biol Phys 1999;44:563-568. 214. Wilder RB, Tucker SL, Ha CS, et al. Dose-response analysis for radiotherapy delivered to patients with intermediate grade and large-cell immunoblastic lymphomas that completely responded to induction, CHOP-based chemotherapy. Int J Radiat Oncol Biol Phys 2001;49:17-22. 215. Fuller LM, Krasin MJ, Velasquez WS, et al. Significance of tumor size and radiation dose to local control in stage I-III diffuse large cell lymphoma treated with CHOP-bleo and radiation. Int J Radiat Oncol Biol Phys 1995;31:3-11. 216. Glick JH, Kim KJ, Earle J, et al. An ECOG randomized phase III trial of CHOP vs. CHOP + radiotherapy (XRT) for intermediate grade early stage non-Hodgkin’s lymphoma (NHL) [abstract]. Proc Am Soc Clin Oncol 1995;14:391. 217. Aref A, Narayan S, Tekyi-Mensah S, et al. Value of radiation therapy in the management of chemoresistant intermediate grade nonHodgkin’s lymphoma. Radiat Oncol Investig 1999;7:186-191. 218. Fletcher GH. Lucy Wortham James Lecture. Subclinical disease. Cancer 1984;53:1274-1284. 219. Cazals-Hatem D, Lepage E, Brice P, et al. Primary mediastinal large B-cell lymphoma. A clinicopathologic study of 141 cases compared with 916 nonmediastinal large B-cell lymphomas, a GELA (“Groupe d’Etude des Lymphomes de l’Adulte”) study. Am J Surg Pathol 1996;20:877-888. 220. Pileri SA, Zinzani PL, Gaidano G, et al. Pathobiology of primary mediastinal B cell lymphoma. Leuk Lymphoma 2003;44:S21-S26. 221. Savage KJ, Monti S, Kutok JL, 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;102:3871-3879. 222. Zinzani PL, Martelli M, Bertini M, et al. for the International Extranodal Lymphoma Study Group. Induction chemotherapy strategies for primary mediastinal large B-cell lymphoma with sclerosis: a retrospective multinational study on 426 previously untreated patients. Haematologica 2002;87:1258-1264. 223. Zinzani PL, Martelli M, Magagnoli M, et al. Treatment and clinical management of primary mediastinal large B-cell lymphoma with sclerosis: MACOP-B regimen and mediastinal radiotherapy monitored by (67)gallium scan in 50 patients. Blood 1999;94:3289-3293. 224. Nguyen LN, Ha CS, Hess M, et al. The outcome of combinedmodality treatments for stage I and II primary large B-cell lymphoma of the mediastinum. Int J Radiat Oncol Biol Phys 2000;47:1281-1285. 225. Groves FD, Linet MS, Travis LB, et al. Cancer surveillance ­series: non-Hodgkin’s lymphoma incidence by histologic subtype in the United States from 1978 through 1995. J Natl Cancer Inst 2000;92:1240-1251. 226. Bloomfield CD, Arthur DC, Frizzera G, et al. Nonrandom chromosome abnormalities in lymphoma. Cancer Res 1983;43: 2975-2984. 227. Bende RJ, Smit LA, van Noesel CJ. Molecular pathways in follicular lymphoma. Leukemia 2007;21:18-29. 228. Aviles A. The International Index is not useful in the classification of low-grade lymphoma. J Clin Oncol 1994;12:2766-2768. 229. Lopez-Guillermo A, Montserrat E, Bosch F, et al. Applicability of the International Index for aggressive lymphomas to patients with low-grade lymphoma. J Clin Oncol 1994;12:1343-1348. 230. Solal-Celigny P, Roy P, Colombat P, et al. Follicular lymphoma international prognostic index. Blood 2004;104:1258-1265. 231. Buske C, Hoster E, Dreyling M, et al. The Follicular Lymphoma International Prognostic Index (FLIPI) separates high-risk from intermediate- or low-risk patients with advanced-stage follicular lymphoma treated front-line with rituximab and the combination of cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) with respect to treatment outcome. Blood 2006;108:1504-1508.

908

PART 11  Leukemias and Lymphomas

232. Cabanillas F. Can we cure indolent lymphomas? Clin Cancer Res 1997;3:2655-2659. 233. Bertoni F, Bosshard G, Roggero E, et al. Clinical relevance of bcl-2(MBR)/J(H) rearrangement detected by polymerase chain reaction in the peripheral blood of patients with follicular lymphoma. Br J Cancer 1997;76:36-39. 234. Lopez-Guillermo A, Cabanillas F, McLaughlin P, et al. Molecular response assessed by PCR is the most important factor predicting failure-free survival in indolent follicular lymphoma: update of the MDACC series. Ann Oncol 2000;11:137-140. 235. Apostolidis J, Gupta RK, Grenzelias D, et al. High-dose therapy with autologous bone marrow support as consolidation of remission in follicular lymphoma: long-term clinical and molecular follow-up. J Clin Oncol 2000;18:527-536. 236. Gribben JG, Neuberg D, Barber M, et al. Detection of residual lymphoma cells by polymerase chain reaction in peripheral blood is significantly less predictive for relapse than detection in bone marrow. Blood 1994;83:3800-3807. 237. Ha CS, Tucker SL, Lee M-S, et al. The significance of molecular response of follicular lymphoma to central lymphatic irradiation as measured by polymerase chain reaction for t(14;18)(q32;q21). Int J Radiat Oncol Biol Phys 2001;49:727-732. 238. Cox JD. Cure and follicular lymphoma. Int J Radiat Oncol Biol Phys 1997;38:1-2. 239. Longo DL. What’s the deal with follicular lymphomas? J Clin Oncol 1993;11:202-208. 240. Besa PC, McClaughlin PW, Cox JD, et al. Long-term assessment of patterns of treatment failure and survival in patients with stage I or II follicular lymphoma. Cancer 1995;75:23612367. 241. MacManus MP, Hoppe RT. Is radiotherapy curative for stage I and II low grade follicular lymphoma? Results of a long-term follow-up study of patients treated at Stanford University. J Clin Oncol 1996;14:1282-1290. 242. Paryani SB, Hoppe RT, Cox RS, et al. Analysis of non-Hodgkin’s lymphomas with nodular and favorable histologies, stages I and II. Cancer 1983;52:2300-2307. 243. Sutcliffe SB, Gospodarowicz MK, Bergsagel DE, et al. Prognostic groups for management of localized Hodgkin’s disease. J Clin Oncol 1985;3:393-401. 244. Pendlebury S, el Awadi M, Ashley S, et al. Radiotherapy results in early stage low grade nodal non-Hodgkin’s lymphoma. Radiother Oncol 1995;36:167-171. 245. MacManus MP, Hoppe RT. Overview of treatment of localized low-grade lymphomas. Hematol Oncol Clin North Am 1997;11:901-918. 246. Gospodarowicz M, Lippuner T, Pintilie M, et al. Stage I and II follicular lymphoma: long-term outcome and patterns of failure following treatment with involved field radiation therapy alone [abstract]. Int J Radiat Oncol Biol Phys 1999;45:217. 247. Tsang RW, Gospodarowicz M, Bezjak A, et al. Staging and management of localized non-Hodgkin’s lymphomas: variations among experts in radiation oncology [abstract]. Int J Radiat ­Oncol Biol Phys 1999;45:389. 248. Lawrence TS, Urba WJ, Steinberg SM, et al. Retrospective analysis of stage I and II indolent lymphomas at the National Cancer Institute. Int J Radiat Oncol Biol Phys 1988;14: 417-424. 249. Vaughan Hudson B, Vaughan Hudson G, MacLennan KA, et al. Clinical stage 1 non-Hodgkin’s lymphoma: long-term follow-up of patients treated by the British National Lymphoma Investigation with radiotherapy alone as initial therapy. Br J Cancer 1994;69:1088-1093. 250. Wilder RB, Jones D, Tucker SL, et al. Long-term results with radiotherapy for stage I-II follicular lymphomas. Int J Radiat Oncol Biol Phys 2001;51:1219-1227. 251. Ha CS, Kong JS, Tucker SL, et al. Central lymphatic irradiation for stage I-III follicular lymphoma: report from a singleinstitutional prospective study. Int J Radiat Oncol Biol Phys 2003;57:316-320. 252. Jacobs JP, Murray KJ, Schultz CJ, et al. Central lymphatic ­irradiation for stage III nodular malignant lymphoma: long-term results. J Clin Oncol 1993;11:233-238.

253. De Los Santos JF, Mendenhall NP, Lynch JW. Is comprehensive lymphatic irradiation for low-grade non-Hodgkin’s lymphoma curative therapy? Long-term experience at a single institution. Int J Radiat Oncol Biol Phys 1997;38:3-8. 254. Paryani SB, Hoppe RT, Cox RS, et al. The role of radiation therapy in the management of stage III follicular lymphomas. J Clin Oncol 1984;2:841-848. 255. Advani R, Rosenberg SA, Horning SJ. Stage I and II follicular non-Hodgkin’s lymphoma: long-term follow-up of no initial therapy. J Clin Oncol 2004;22:1454-1459. 256. Seymour JF, Pro B, Fuller LM, et al. Long-term follow-up of a prospective study of combined modality therapy for stage I-II indolent non-Hodgkin’s lymphoma. J Clin Oncol 2003;21: 2115-2122. 257. Ha CS, Tucker SL, Blanco AI, et al. Salvage central lymphatic irradiation in follicular lymphomas following failure of chemotherapy: a feasibility study. Int J Radiat Oncol Biol Phys 1999;45:1207-1212. 258. Mahe M, Bourdin S, Le Pourhiet-Le Mevel A, et al. Salvage ­extended-field irradiation in follicular non-Hodgkin’s lymphoma after failure of chemotherapy. Int J Radiat Oncol Biol Phys 2000;47:735-738. 259. Hiddemann W, Unterhalt M, Buske C, et al. Treatment of follicular follicle centre lymphomas: current status and future perspectives. J Intern Med Suppl 1997;740:55-62. 260. Peterson BA. Current treatment of follicular low-grade lymphomas. Semin Oncol 1999;26:2-11. 261. Ha CS, Kong JS, McLaughlin P, et al. Stage III follicular lymphoma: long-term follow-up and patterns of failure. Int J Radiat Oncol Biol Phys 2003;57:748-754. 262. Rosenberg SA. The low-grade non-Hodgkin’s lymphomas: challenges and opportunities. J Clin Oncol 1985;3:299-310. 263. Horning SJ, Rosenberg SA. The natural history of initially ­untreated low-grade non-Hodgkin’s lymphomas. N Engl J Med 1984;311:1471-1475. 264. Young RC, Longo DL, Glatstein E, et al. The treatment of indolent lymphomas: watchful waiting v. aggressive combined modality treatment. Semin Hematol 1988;25:11-16. 265. Portlock CS, Rosenberg SA. No initial therapy for stage III and IV non-Hodgkin’s lymphomas of favorable histologic types. Ann Intern Med 1979;90:10-13. 266. Cabanillas FF. Management of patients with low-grade follicular lymphoma. Am J Clin Oncol 1989;12:81-87. 267. Cotter FE. Antisense therapy of hematologic malignancies. Semin Hematol 1999;36:9-14. 268. Pich A, Rancourt C. A role for intracellular immunization in chemosensitization of tumor cells? Gene Ther 1999;6:1202-1209. 269. Bendandi M, Longo DL. Biologic therapy for lymphoma. Curr Opin Oncol 1999;11:343-350. 270. McLaughlin P, White CA, Grillo-Lopez AJ, et al. Clinical status and optimal use of rituximab for B-cell lymphomas. Oncology 1998;12:1763-1769. 271. Knox SJ, Meredith RF. Clinical radioimmunotherapy. Semin ­Radiat Oncol 2000;10:73-93. 272. Arthur K. Follicular lymphoma: a review of 58 cases. Clin Radiol 1968;19:347-350. 273. Sutcliffe SB, Gospodarowicz MK, Bush RS, et al. Role of radiation therapy in localized non-Hodgkin’s lymphoma. Radiother Oncol 1985;4:211-223. 274. Cox JD, Koehl RH, Turner WM, et al. Irradiation in the local control of malignant lymphoreticular tumors (non-Hodgkin’s malignant lymphoma). Radiology 1974;112:179-185. 275. Parsonnet J, Hansen S, Rodriguez L, et al. Helicobacter ­ pylori infection and gastric lymphoma. N Engl J Med 1994;330: 1267-1271. 276. Roggero E, Zucca E, Mainetti C, et al. Eradication of Borrelia burgdorferi infection in primary marginal zone B-cell lymphoma of the skin. Hum Pathol 2000;31:263-268. 277. Bogle MA, Riddle CC, Triana EM, et al. Primary cutaneous B-cell lymphoma. J Am Acad Dermatol 2005;53:479-484. 278. Lecuit M, Abachin E, Martin A, et al. Immunoproliferative small intestinal disease associated with Campylobacter jejuni. N Engl J Med 2004;350:239-248.

34  Leukemias and Lymphomas 279. Ferreri AJ, Guidoboni M, Ponzoni M, et al. Evidence for an association between Chlamydia psittaci and ocular adnexal lymphomas. J Natl Cancer Inst 2004;96:586-594. 280. Licci S, Boscaino A, De Palma M, et al. Concurrence of marginal zone B-cell lymphoma MALT-type and Langerhans cell histiocytosis in a thyroid gland with Hashimoto disease. Ann Hematol 2008;87:855-857. 281. Kinoshita N, Ashizawa K, Abe K, et al. Mucosa-associated lymphoid tissue lymphoma of the thymus associated with Sjögren’s syndrome: report of a case. Surg Today 2008;38:436-439. 282. Ekström Smedby K, Vajdic CM, Falster M, et al. Autoimmune disorders and risk of non-Hodgkin lymphoma subtypes: a pooled analysis within the InterLymph Consortium. Blood 2008;111:4029-4038. 283. Ambrosetti A, Zanotti R, Pattaro C, et al. Most cases of primary salivary mucosa-associated lymphoid tissue lymphoma are associated either with Sjoegren syndrome or hepatitis C virus infection. Br J Haematol 2004;126:43-49. 284. Unge P. Eradication therapy of Helicobacter pylori. A review. Report from a workshop organized by the Swedish and Norwegian Medical Products Agencies, September 1995. J Gastroenterol 1998;33:57-61. 285. Steinbach G, Ford R, Glober G, et al. Antibiotic treatment of gastric lymphoma of mucosa-associated lymphoid tissue. An uncontrolled trial. Ann Intern Med 1999;131:88-95. 286. Wotherspoon AC, Doglioni C, Diss TC, et al. Regression of primary low-grade B-cell gastric lymphoma of mucosa-associated lymphoid tissue type after eradication of Helicobacter pylori. Lancet 1993;342:575-577. 287. Du MQ, Isaacson PG. Gastric MALT lymphoma: from aetiology to treatment. Lancet Oncol 2002;3:97-104. 288. Taal BG, Burgers JM, van Heerde P, et al. The clinical spectrum and treatment of primary non-Hodgkin’s lymphoma of the stomach. Ann Oncol 1993;4:839-846. 289. Schechter NR, Portlock CS, Yahalom J. Treatment of mucosa­associated lymphoid tissue lymphoma of the stomach with ­radiation alone. J Clin Oncol 1998;16:1916-1921. 290. Yahalom J, Schecter NR, Gonzales M, et al. Effective treatment of MALT lymphoma of the stomach with radiation alone. Ann Oncol 1999;10(Suppl 3):135. 291. Zinzani PL, Magagnoli M, Galieni P, et al. Nongastrointestinal low-grade mucosa-associated lymphoid tissue lymphoma: analysis of 75 patients. J Clin Oncol 1999;17:1254. 292. Zelenetz AD, Hoppe RT. NCCN Non-Hodgkin’s Lymphoma Practice Guidelines Panel. NCCN: Non-Hodgkin’s Lymphoma. Cancer Causes Control 2001;8(Suppl 2):102-113. 293. Conconi A, Martinelli G, Thieblemont C, et al. Clinical activity of rituximab in extranodal marginal zone B-cell lymphoma of MALT type. Blood 2003;102:2741-2745. 294. Zucca E, Conconi A, Pedrinis E, et al. Nongastric marginal zone B-cell lymphoma of mucosa-associated lymphoid tissue. Blood 2003;101:2489-2495. 295. Morgan G, Vornanen M, Puitinen J, et al. Changing trends in the incidence of non-Hodgkin’s lymphoma in Europe Biomed Study Group. Ann Oncol 1997;8:49-54. 296. Melnyk A, Rodriguez A, Pugh WC, et al. Evaluation of the ­Revised European-American Lymphoma classification confirms the clinical relevance of immunophenotype in 560 ­ cases of ­ aggressive non-Hodgkin’s lymphoma. Blood 1997;89:4514-4520. 297. Lopez-Guillermo A, Cid J, Salar A, et al. Peripheral T-cell lymphomas: initial features, natural history, and prognostic factors in a series of 174 patients diagnosed according to the REAL classification. Ann Oncol 1998;9:849-855. 298. Harabuchi Y, Imai S, Wakashima J, et al. Nasal T-cell ­lymphoma causally associated with Epstein-Barr virus: clinicopathologic, phenotypic, and genotypic studies. Cancer 1996;77:2137-2149. 299. Cheung MMC, Chan JK, Lau WH, et al. Early stage nasal T/ NK-cell lymphoma: clinical outcome, prognostic factors, and the effect of treatment modality. Int J Radiat Oncol Biol Phys 2002;54:182-190. 300. Li YX, Yao B, Jin J, et al. Radiotherapy as primary treatment for stage IE and IIE nasal natural killer/T-cell lymphoma. J Clin Oncol 2006;24:181-189.

909

301. Koom WS, Chung EJ, Yang WI, et al. Angiocentric T-cell and NK/T-cell lymphomas: radiotherapeutic viewpoints. Int J Radiat Oncol Biol Phys 2004;59:1127-1137. 302. Ostronoff M, Soussain C, Zambon E, et al. Burkitt’s lymphoma in adults: a retrospective study of 46 cases. Nouv Rev Fr Hematol 1992;34:389-397. 303. Rizzieri DA, Johnson JL, Niedzwiecki D, 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;100:1438-1448. 304. Soussain C, Patte C, Ostronoff M, et al. Small noncleaved cell lymphoma and leukemia in adults. A retrospective study of 65 adults treated with the LMB pediatric protocols. Blood 1995;85:664-674. 305. Thomas DA, Cortes J, O’Brien S, et al. Hyper-CVAD program in Burkitt’s-type adult acute lymphoblastic leukemia. J Clin Oncol 1999;17:2461-2470. 306. Bertoni F, Zucca E, Cotter FE. Molecular basis of mantle cell lymphoma. Br J Haematol 2004;124:130-140. 307. Meusers P, Hense J, Brittinger G. Mantle cell lymphoma: diagnostic criteria, clinical aspects and therapeutic problems. Leukemia 1997;11:S60-S64. 308. Khouri IF, Romaguera J, Kantarjian H, et al. Hyper-CVAD and high-dose methotrexate/cytarabine followed by stem-cell transplantation: an active regimen for aggressive mantle-cell ­lymphoma. J Clin Oncol 1998;16:3803-3809. 309. Morrison VA, Peterson BA. High-dose therapy and transplantation in non-Hodgkin’s lymphoma. Semin Oncol 1999;26:84-98. 310. Soslow RA, Baergen RN, Warnke RA. B-lineage lymphoblastic lymphoma is a clinicopathologic entity distinct from other histologically similar aggressive lymphomas with blastic morphology. Cancer 1999;85:2648-2654. 311. Weinstein HJ, Vance ZB, Jaffe N, et al. Improved prognosis for patients with mediastinal lymphoblastic lymphoma. Blood 1979;53:687-694. 312. Hoelzer D, Gokbuget N, Digel W, et al. Outcome of adult patients with T-lymphoblastic lymphoma treated according to protocols for acute lymphoblastic leukemia. Blood 2002;99:4379-4385. 313. Thomas DA, O’Brien S, Cortes J, et al. Outcome with the hyper-CVAD regimens in lymphoblastic lymphoma. Blood 2004;104:1624-1630. 314. Coleman CN, Picozzi VJ Jr, Cox RS, et al. Treatment of lymphoblastic lymphoma in adults. J Clin Oncol 1986;4: 1628-1637. 315. Colgan JP, Andersen J, Habermann TM, et al. Long-term maintenance and central nervous system prophylaxis in lymphoblastic non-Hodgkin’s lymphoma. Leuk Lymphoma 1994;15:291-296. 316. Levitt LJ, Aisenberg AC, Harris NL, et al. Primary nonHodgkin’s lymphoma of the mediastinum. Cancer 1982;50: 2486-2492. 317. Levine AM, Forman SJ, Meyer PR, et al. Successful therapy of convoluted T-lymphoblastic lymphoma in the adult. Blood 1983;61:92-98. 318. Lamberg SI, Green SB, Byar DP, et al. Clinical staging for cutaneous T-cell lymphoma. Ann Intern Med 1984;100:187-192. 319. Diamandidou E, Cohen PR, Kurzrock R. Mycosis fungoides and Sezary syndrome. Blood 1996;88:2385-2409. 320. Kim YH, Hoppe RT. Mycosis fungoides and the Sezary syndrome. Semin Oncol 1999;26:276-289. 321. Duncan K, Heald P. Cutaneous T-cell lymphoma: centuries of controversy. Semin Cutan Med Surg 1998;17:133-140. 322. Lo TC, Salzman FA, Moschella SL, et al. Whole body surface electron irradiation in the treatment of mycosis fungoides. Radiology 1979;130:453-457. 323. Tadros AA, Tepperman BS, Hryniuk WM, et al. Total skin electron irradiation for mycosis fungoides: failure analysis and prognostic factors. Int J Radiat Oncol Biol Phys 1983;9:1279-1287. 324. Hoppe RT, Cox RS, Fuks Z, et al. Electron-beam therapy for mycosis fungoides: the Stanford University experience. Cancer Treat Rep 1979;63:691-700. 325. Kim YH, Jensen RA, Watanabe GL, et al. Clinical stage IA (limited patch and plaque) mycosis fungoides: a long-term outcome analysis. Arch Dermatol 1996;132:1309-1313.

910

PART 11  Leukemias and Lymphomas

326. Hamminga B, Noordijk EM, van Vloten WA. Treatment of mycosis fungoides: total-skin electron-beam irradiation vs topical mechlorethamine therapy. Arch Dermatol 1982;118:150-153. 327. Zachariae H, Thestrup-Pedersen K, Sogaard H. Topical nitrogen mustard in early mycosis fungoides. A 12-year experience. Acta Derm Venereol 1985;65:53-58. 328. Vonderheid EC, Tan ET, Kantor AF, et al. Long-term efficacy, curative potential, and carcinogenicity of topical mechlorethamine in cutaneous T cell lymphoma. J Am Acad Dermatol 1989;20:416-428. 329. Hoppe RT. The management of mycosis fungoides at ­Stanford: standard and innovative treatment programmes. Leukemia 1991;5:46-48. 330. Kaye FJ, Bunn PA, Steinberg SM, et al. A randomized trial comparing combination electron beam radiation and chemotherapy with topical therapy in the initial treatment of mycosis fungoides. N Engl J Med 1989;321:1784-1790. 331. Lo TCM, Salzman FA, Costey GE, et al. Megavolt electron ­irradiation for localized mycosis fungoides. Acta Radiol Oncol 1981;20:71-74. 332. Wilson LD, Kacinski BM, Jones GW. Local superficial radiotherapy in the management of minimal stage IA cutaneous T-cell lymphoma. Int J Radiat Oncol Biol Phys 1998;40:109-115. 333. Micaily B, Miyamoto C, Kantor G. Radiotherapy for unilesional mycosis fungoides. Int J Radiat Oncol Biol Phys 1998;42: 361-364. 334. Cotter GW, Baglan RJ, Wasserman TH, et al. Palliative radiation treatment of cutaneous mycosis fungoides-A dose response. Int J Radiat Oncol Biol Phys 1983;9:1477-1480. 335. Herrmann JJ, Roenigk HH Jr, Hurria A, et al. Treatment of mycosis fungoides with photochemotherapy (PUVA): long-term follow-up. Acad Dermatol 1995;33:234-242. 336. Watson A. Photochemotherapy for mycosis fungoides: current status. Australas J Dermatol 1997;38:9-11. 337. Chinn DM, Chow S, Kim YH, et al. Total skin electron beam therapy with or without adjuvant topical nitrogen mustard or nitrogen mustard alone as initial treatment of T2 and T3 mycosis fungoides. Int J Radiat Oncol Biol Phys 1999;43:951-958. 338. Zackheim HS. Treatment of cutaneous T-cell lymphoma. Semin Dermatol 1994;13:207-215. 339. Ferra C, Servitje O, Petriz L, et al. Autologous haematopoietic progenitor transplantation in advanced mycosis fungoides. Br J Dermatol 1999;140:1188-1189. 340. Marcus R, Burrows NP, Roberts SO, et al. Erythrodermic mycosis fungoides treated with total body irradiation and autologous bone marrow transplantation. Clin Exp Dermatol 1995;20:73-75. 341. Bigler RD, Crilly P, Micaily B, et al. Autologous bone marrow transplantation for advanced stage mycosis fungoides. Bone Marrow Transplant 1991;7:133-137. 342. Hoppe RT, Abel EA, Deneau DG, et al. Mycosis fungoides: management with topical nitrogen mustard. J Clin Oncol 1987;5:1796-1803. 343. Ramsay DL, Halperin PS, Zeleniuch-Jacquotte A. Topical mechlorethamine for early stage mycosis fungoides. J Am Acad Dermatol 1988;19:684-691. 344. Kemme DJ, Bunn PA Jr. State of the art therapy of mycosis fungoides and Sézary syndrome. Oncology (Williston Park) 1992;6:31-42; discussion 44, 47-48. 345. Wilson LD, Kacinski BM, Edelson RL, Cutaneous T-cell lymphomas. In DeVita VT Jr, Hellman S, Rosenberg SA (eds): Cancer: Principles and Practice of Oncology. 5th ed, Philadelphia: Lippincott-Raven, 1997, pp 2220-2232. 346. Hoppe RT, Fuks Z, Bagshaw MA. Radiation therapy in the management of cutaneous T-cell lymphoma. Cancer Treat Rep 1979;63:625-632. 347. Antolak JA, Cundiff JH, Ha CS. Utilization of thermoluminescent dosimetry in total skin electron beam radiotherapy of mycosis fungoides. Int J Radiat Oncol Biol Phys 1998;40: 101-108. 348. Querfeld C, Kuzel TM, Guitart J, et al. Primary cutaneous CD30+ lymphoproliferative disorders: new insights into biology and therapy. Oncology (Williston Park) 2007;21:689-696; discussion 699–700.

349. Paulli M, Berti E, Rosso R, et al. CD30/Ki-1-positive lymphoproliferative disorders of the skin-clinicopathologic correlation and statistical analysis of 86 cases: a multicentric study from the European Organization for Research and Treatment of Cancer ­Cutaneous Lymphoma Project Group. J Clin Oncol 1995;13:1343-1354. 350. Bekkenk MW, Geelen FA, van Voorst Vader PC, 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;95:3653-3661. 351. Macaulay WL. Lymphomatoid papulosis. A continuing self­healing eruption, clinically benign-histologically malignant. Arch Dermatol 1968;97:23-30. 352. Vonderheid EC, Sajjadian A, Kadin ME. Methotrexate is ­effective therapy for lymphomatoid papulosis and other primary cutaneous CD30-positive lymphoproliferative disorders. J Am Acad Dermatol 1996;34:470-481. 353. Hughes PS. Treatment of lymphomatoid papulosis with imiquimod 5% cream. J Am Acad Dermatol 2006;54:546-547. 354. Yu JB, Mc Niff JM, Lund MW, et al. Treatment of primary cutaneous CD3D+ anaplastic large-cell lymphoma with radiation therapy. Int J Radiat Oncol Biol Phys 2008;70:1542-1545. 355. Newton HB, Fine HA. NCCN clinical practice guidelines for non-immunosuppressed primary CNS lymphoma. Oncology 1999;13:153-160. 356. Fine HA, Mayer RJ. Primary central nervous system lymphoma. Ann Intern Med 1993;119:1093-1104. 357. Deangelis LM. Current management of primary central nervous system lymphoma. Oncology 1995;9:63-71. 358. Abrey L, Batchelor T, Ferreri A, et al. Report of an international workshop to standardize baseline evaluation and response criteria for primary CNS lymphoma. J Clin Oncol 2005;23:50345043. 359. Ferreri AJ, Abrey LE, Blay JY, et al. Summary statement on primary central nervous system lymphomas from the Eighth International Conference on Malignant Lymphoma, Lugano, ­Switzerland, June 12 to 15, 2002. J Clin Oncol 2003;21:24072414. 360. Ferreri AJ, Reni M, Pasini F, et al. A multicenter study of treatment of primary CNS lymphoma. Neurology 2002;58: 1513-1520. 361. DeAngelis LM, Seiferheld W, Schold SC, et al. Combination chemotherapy and radiotherapy for primary central nervous system lymphoma: Radiation Therapy Oncology Group Study 93-10. J Clin Oncol 2002;20:4643-4648. 362. O’Brien P, Roos D, Pratt G, et al. Combined-modality therapy for primary central nervous system lymphoma: long-term data from a Phase II multicenter study (Trans-Tasman Radiation Oncology Group). Int J Radiat Oncol Biol Phys 2006;64:408-413. 363. Abrey LE, DeAngelis LM, Yahalom J. Long-term survival in primary CNS lymphoma. J Clin Oncol 1998;16:859-863. 364. Blay JY, Conroy T, Chevreau C, 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;16:864-871. 365. Blay JY, Conroy T, Chevreau C, et al. Late neurological toxicity after the treatment of primary cerebral lymphoma: multivariate analysis on 208 patients [abstract]. Proc Am Soc Clin Oncol 1997;16:A86. 366. Gavrilovic IT, Hormigo A, Yahalom J, et al. Long-term followup of high-dose methotrexate-based therapy with and without whole brain irradiation for newly diagnosed primary CNS lymphoma. J Clin Oncol 2006;24:4570-4574. 367. Bessell EM, Lopez-Guillermo A, Villa S, 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;20:231-236. 368. Nelson DF, Martz KL, Bonner H, 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.

34  Leukemias and Lymphomas 369. Freilich RJ, Delattre JY, Monjour A, et al. Chemotherapy without radiation therapy as initial treatment for primary CNS lymphoma in older patients. Neurology 1996;46:435-439. 370. Sandor V, Stark-Vancs V, Pearson D, et al. Phase II trial of chemotherapy alone for primary CNS and intraocular lymphoma. J Clin Oncol 1998;16:3000-3006. 371. Pels H, Schmidt-Wolf IG, Glasmacher A, et al. Primary central nervous system lymphoma: results of a pilot and phase II study of systemic and intraventricular chemotherapy with deferred radiotherapy. J Clin Oncol 2003;21:4489-4495. 372. Glass J, Gruber ML, Cher L, et al. Preirradiation methotrexate chemotherapy of primary central nervous system lymphoma: long-term outcome. J Neurosurg 1994;81:188-195. 373. Nelson DF. Radiotherapy in the treatment of primary central nervous system lymphoma (PCNSL). J Neurooncol 1999;43:241-247. 374. Delattre JY, Shapiro WR, Posner JB. Acute effects of low-dose cranial irradiation on regional capillary permeability in experimental brain tumors. J Neurol Sci 1989;90:147-153. 375. DeMario MD, Liebowitz DN. Lymphomas in the immunocompromised patient. Semin Oncol 1998;25:492-502. 376. Antinori A, Cingolani A, Alba L, et al. Better response to chemotherapy and prolonged survival in AIDS-related lymphomas responding to highly active antiretroviral therapy. AIDS 2001;15:1483-1491. 377. Navarro JT, Ribera JM, Oriol A, et al. Influence of highly active anti-retroviral therapy on response to treatment and survival in patients with acquired immunodeficiency syndrome-related non-Hodgkin’s lymphoma treated with cyclophosphamide, hydroxydoxorubicin, vincristine and prednisone. Br J Haematol 2001;112:909-915. 378. Liebross RH, Ha CS, Cox JD, et al. Clinical course of solitary extramedullary plasmacytoma. Radiother Oncol 1999;52:245249. 379. Bush SE, Goffinet DR, Bagshaw MA. Extramedullary plasmacytoma of the head and neck. Radiology 1981;140:801-805.

911

380. Knowling MA, Harwood AR, Bergsagel DE. Comparison of extramedullary plasmacytomas with solitary and multiple plasma cell tumors of bone. J Clin Oncol 1983;1:255-262. 381. Wasserman TH. Diagnosis and management of plasmacytomas. Oncology 1987;1:37-41. 382. Brinch L, Hannisdal E, Abrahamsen AF, et al. Extramedullary plasmacytomas and solitary plasmacytoma of the head and neck. Eur J Haematol 1990;44:131-134. 383. Mayr NA, Wen BC, Hussey DH, et al. The role of radiation therapy in the treatment of solitary plasmacytomas. Radiother Oncol 1990;17:293-303. 384. Bolek TW, Marcus RB, Mendenhall NP. Solitary plasmacytoma of bone and soft tissue. Int J Radiat Oncol Biol Phys 1996;36: 329-333. 385. Susnerwala SS, Shanks JH, Banerjee SS, et al. Extramedullary plasmacytoma of the head and neck region: clinicopathological correlation in 25 cases. Br J Cancer 1997;75:921-927. 386. Bataille R, Sany J. Solitary myeloma: clinical and prognostic features of a review of 114 cases. Cancer 1981;48:845-851. 387. Moulopoulos LA, Dimopoulos MA, Weber D, et al. Magnetic resonance imaging in the staging of solitary plasmacytoma of bone. J Clin Oncol 1993;11:1311-1315. 388. Liebross RH, Ha CS, Cox JD, et al. Solitary bone plasmacytoma: outcome and prognostic factors following radiotherapy. Int J Radiat Oncol Biol Phys 1998;41:1063-1067. 389. Wilder RB, Ha CS, Cox JD, et al. Persistence of myeloma protein for more than one year after radiotherapy is an adverse prognostic factor in solitary plasmacytoma of bone. Cancer 2002;94: 1532-1537. 390. Mendenhall CM, Thar TL, Million RR. Solitary plasmacytoma of bone and soft tissue. Int J Radiat Oncol Biol Phys 1980;6: 1497-1501. 391. Tsang RW, Gospodarowicz MK, Pintilie M, et al. Solitary plasmacytoma treated with radiotherapy: impact of tumor size on outcome. Int J Radiat Oncol Biol Phys 2001;50:113-120.

The Bone 35 Valerae O. Lewis RESPONSE OF NORMAL BONE TO IRRADIATION Tolerance and Radiosensitivity Radiation Effects on Bone Growth EFFECTS OF IRRADIATION ON HETEROTOPIC OSSIFICATION AND SECONDARY MALIGNANCIES Prevention of Heterotopic Bone Formation Induction of Secondary Tumors

BONE IMAGING PRIMARY MALIGNANT TUMORS OF BONE Plasmacytoma Chondrosarcoma Primary Malignant Lymphoma of Bone Chordoma Malignant Fibrous Histiocytoma Hemangioendothelioma Giant Cell Tumor of Bone Eosinophilic Granuloma RADIATION THERAPY FOR METASTATIC BONE DISEASE

Bone tumors represent less than 0.5% of all primary ­tumors, and bone destruction is their sine qua non. This discussion of primary malignant tumors of bone stresses the role of radiation therapy and comments on the diagnostic imaging of bone tumors. Although bone tumors disseminate almost exclusively through the blood, radiation fields must be designed to spare soft tissue in the areas of lymphatic drainage to reduce the risk of late lymphedema. In combined-modality therapy, orthopedic surgeons, diagnostic radiologists, and radiation and medical oncologists must be involved in all treatment decisions from the outset.

Response of Normal Bone to Irradiation Tolerance and Radiosensitivity Bone is almost twice as dense as soft tissue, with an effective atomic number approaching 14. However, in the range of machine energies used, photon radiation is absorbed primarily by means of the Compton process, which is essentially independent of the atomic number of the attenuating tissues. Although bone absorption of particle radiation with high linear energy transfer is greater than that of photons, the difference is significant only when neutron irradiation is used for osteogenic and other skeletal sarcomas. It is important to distinguish between the relative radiosensitivity of growing cartilage and bone in a child and the relative radioresistance of mature bone and cartilage in the adult. Arrested growth can be seen in children, for whom the dose at which the risk of injury is 5% at 5 years (tolerance dose or TD5/5) is 10 Gy and dose at which the risk of injury is 50% at 5 years (TD50/5) is 30 Gy. The corresponding values for necrosis and fracture in the adult are a TD5/5 of 60 Gy and a TD50/5 of 100 Gy given with conventional fractionation. The threshold for radiographic changes in normal adult bone is about 50 Gy given over 5 weeks, with the changes manifested at least 2 years ­after

the irradiation, although in the dose ranges used clinically, these radiographically demonstrated changes in the trabecular pattern and cortical thickening are usually not significant. It is highly likely, however, that if concomitant or sequential multiple-agent chemotherapy is given with the radiation, these TD5/5 and TD50/5 guidelines should be modified downward.

Radiation Effects on Bone Growth The anatomy and physiology of the growth plate was well described in a 1995 report of a late effects consensus conference.1 Much of the early work was performed by ­ Kember and colleagues.2,3 To summarize, ossification centers develop within the cartilage of a fetus or growing child. Some bones (e.g., wrist, ankle, zygoma) develop from a single ossification center. Others (e.g., femur) develop from one point in the center and one or more others in the epiphyseal end. The growth plate is a column of cells that remains between the secondary epiphysis and metaphysis until after puberty, when it fuses. The growth plate consists of three layers: reserve cells, proliferative cells, and hypertrophic cells. As the cells mature, they progress down the column. When they reach a certain distance from the top of the growth plate, calcium is released, and ossification occurs. In rodents, most cells in the proliferative zone are dividing.2,4 In humans, the cell growth rate is fastest at birth, slower during childhood (with a cell cycle time of approximately 20 to 30 days at age 5 to 8 years), and then faster again during puberty.3 This proliferative zone is sensitive to radiation therapy and its fractionation.5,6 In mice, irradiated osteoblasts demonstrate decreased cell proliferation and a dose-dependent, sustained reduction in collagen production relative to control cells7; irradiated tibias show dose-dependent reductions in bone growth.8 The effects of hyperfractionation and accelerated hyperfractionation in these models are ­being investigated.5,6,9 In children, irradiation of the axial skeleton can cause height abnormalities (especially in children younger than

915

916

PART 12  Musculoskeletal System

6 years or children between 11 and 13 years old given more than 33 to 35 Gy),10-12 partial irradiation of vertebral bodies can cause scoliosis,13-15 and hip irradiation can cause slipped capital femoral epiphysis (especially in children younger than 4 years given more than 25 Gy)16 and avascular necrosis (especially in children given more than 30 Gy, steroids, or concomitant chemotherapy).1,17 Ultimately, stature depends on age at the time of treatment, total radiation dose, the volume treated,11 and the growth hormone supply (which may be reduced secondary to cranial irradiation).18 Kroll and colleagues19 reviwed the records of 236 patients who had been given radiation for malignant disease during childhood; some degree of bone deformity was seen in 40%, and it occurred more commonly when radiation therapy was administered before 2 years of age. When ­tumor control requires the irradiation of growing epiphyseal cartilage, the recommendations from the late effects consensus conference1 are to avoid dose gradients across vertebral bodies (especially laterally), to use relatively small fraction sizes (≤1.8 Gy), and to make sure that the patient and family are aware of the various risks of bone irradiation. Early identification and treatment of bone abnormalities are essential. For example, patients with spinal curvatures of more than 20 degrees may benefit from brace support. Patients with mild leg length discrepancies may benefit from shoe lifts; patients with severe leg length discrepancies may need amputation. Patients with severe epiphyseal widening or mild epiphyseal slippage may benefit from surgical pinning; patients with severe epiphyseal slippage (more than 60 degrees) may need osteotomy and osteoplasty.1 Amifostine may confer some protection from radiation effects.20-22 To determine the relative benefits of sparing longitudinal bone growth by fractionation alone compared with pretreatment with amifostine, Damron and colleagues21 randomized 24 weanling, 4-week-old, male Sprague-Dawley rats to receive fractionated radiation therapy alone (n = 12) or with amifostine pretreatment (n = 12). The distal femur and proximal tibia in the right leg of each animal were exposed to 17.5 Gy total in 3 or 5 fractions, with the contralateral leg used as a control. The use of amifostine for radioprotection when the rats received radiation in 5 fractions significantly reduced the femoral and overall percentage growth arrest and limb length ­ discrepancy compared with rats that received only the 5 fractions. Anticipated growth loss from a single 17.5-Gy radiation dose was reduced by 30.8% in the femur, 20.3% in the tibia, and 25.7% overall in the total limb. In another study, Damron and colleagues22 found a directly proportional relationship between amifostine dose and its protective effects on the growth plate. They exposed 36 weanling Sprague-Dawley rats to a single dose of 17.5 Gy to the right knee, with the contralateral limb used as a nonirradiated control. Six groups of six rats were each given amifostine intraperitoneally 20 minutes before radiation exposure at doses of 0, 50, 100, 150, 200, or 250 mg/kg. Six weeks after treatment, the rats were killed, and the lower limbs were disarticulated, skeletonized, imaged, and measured. Statistically significant dose-related differences were observed between amifostine dosage groups for mean right-sided growth, growth loss, and limb length discrepancy. The mean amount of right-sided growth recovered by

amifostine administration increased from 14% at 50 mg/kg to 57% at 250 mg/kg. Growth loss and limb discrepancy were significantly reduced in proportion to increasing amifostine doses.22 Cranial and total-body irradiation (TBI) in young patients can cause hormone deficiencies affecting stature and bone density.18,23-27 Causes of short height and osteopenia are probably multifactorial, but growth hormone deficiency is the most common endocrine deficiency observed after cranial irradiation.26,28 TBI alone causes height loss. Clement-de Boers and colleagues29 observed that among patients who in childhood had undergone bone marrow transplantation (BMT) after chemotherapy and TBI with 7.5 Gy or more, final height was compromised because of blunted growth in puberty, but patients who had not undergone TBI suffered no height loss. Thyroid function and adrenal function were normal in all patients, and no growth hormone deficiency was detected. When cranial irradiation is added to TBI, especially single-fraction TBI, the effects on height are even worse. Cohen and colleagues25 reviewed the cases of 181 patients who had had BMT before puberty and who had reached their final height. Previous cranial irradiation plus singledose TBI had the greatest negative effect on final height. Fractionation of TBI reduced this effect significantly, and conditioning with busulfan and cyclophosphamide seemed to eliminate it. Girls did better than boys, and the younger the person was at time of BMT, the greater the loss in height. Nevertheless, most patients (140 of 181) reached an adult height that was within the normal range of the general population. As Murray and colleagues30 reported, the clinical picture for adults with growth hormone deficiency resulting from primary hypothalamic-pituitary disease during childhood is similar to that described for longterm survivors of childhood cancer: increased fat mass and reductions in lean mass, strength, exercise tolerance, bone mineral density, and quality of life. Arikoski and colleagues26 found that for survivors of childhood acute lymphoblastic leukemia, being male and having a history of cranial irradiation were associated with low lumbar and femoral bone marrow densities. Plasma insulin-like growth factor type 1 (IGF1) and its growth hormone–dependent binding protein (IGFBP3) are of no diagnostic value for growth hormone deficiency after TBI. Brauner and colleagues31 interpreted the presence of normal or high IGF1 levels despite having had low growth hormone peaks in response to a stimulation test as suggesting that bone irradiation induced lesions that caused resistance to IGF1. Over time, the production of growth hormone may ­improve on its own (years after BMT), but early growth hormone replacement has shown benefit.23,24 Hovi and colleagues24 found that 3 years of growth hormone treatment, begun 2 to 6 years after BMT, yielded an increase in mean height standard-deviation score of 0.4, whereas untreated patients suffered a decrease in standard-deviation score of 0.8 during the corresponding time (P = .009). Other hormonal problems, such as gonadal failure, may contribute to growth impairment. The importance of initiating hormone replacement therapy promptly after treatment cannot be overemphasized.32 Clinically relevant effects of radiation therapy on the dental and craniofacial development of children include

35  The Bone

malocclusion, reduced mobility of the temporomandibular joint, growth retardation, and osteoradionecrosis.33 Malocclusion and reduced mobility of the temporomandibular joint can cause muscle pain and headaches,33 in part from treatment-induced inflammation and eventual fibrosis of the joint.33 Qualitative changes, such as regularly spaced lines in the enamel and dentine of the teeth, have been found to correspond with cycles of chemotherapy. Qualitative and quantitative changes, such as true dental hypoplasia, have been associated with TBI.34 Children treated with cranial irradiation before 5 years of age exhibit a reduced growth of the mandible; the mandible is four times more radiosensitive than the maxilla.35 Radiation-induced alterations of factors related to bone remodeling and wound healing may participate in the pathogenesis of osteoradionecrosis.36 Ashamalla and colleagues37 found that hyperbaric oxygen could be safely used to prevent and treat radiation-induced bone and soft tissue complications in children. In mice, radiation effects on bone growth can be significantly reduced by hyperbaric oxygen after 10 or 20 Gy but not after 30 Gy. At 30 Gy, basic fibroblast growth factor (bFGF, or FGF2) still significantly reduced the degree of bone shortening, but hyperbaric oxygen provided no added benefit to the FGF2 therapy.7 Rodent studies suggest that recombinant human bone morphogenic protein-2 in soluble collagen type 1 carrier delivered to an irradiated bone defect (3 mm, calvarial) can increase new bone formation (34% ± 14% of the defect area within 21 days, compared with 7% ± 2% for type 1 collagen carrier alone and 5% ± 2% if untreated; untreated, unirradiated bone defects show a spontaneous osseous regeneration of 90% ± 7% of the defect area within 21 days).38 Rapidly reproducing bone marrow cells are highly radiosensitive. The effects of radiation on regenerative potential vary with the volume irradiated. Management of the hematopoietic effects of radiation therapy includes surveillance, antibiotic therapy, blood product transfusion, and occasionally hematopoietic growth factors, BMT, or reinfusion of peripheral blood stem cells.39 Ding and colleagues40 studied the bone marrow protectant effects of acidic FGF (aFGF, or FGF1) and FGF2 given within a day of radiation therapy. The radioprotectant effect of these drugs in mice treated with a single dose of TBI was characterized by a dose-modification factor of 1.15. In fractionated systems, FGF2 seems to be more effective than FGF1 (respective dose-modification factors of 1.22 and 1.04). FGF1 given 24 and 4 hours before irradiation with an otherwise lethal dose of TBI improved the ­repopulation of hematopoietic progenitor cells of bone marrow; this effect was not associated with an increase in the S-phase fraction or detectable circulating levels of the cytokines interleukin-3, tumor necrosis factor alpha (TNF-α), or granulocyte-macrophage colony-stimulating factor, suggesting that other mechanisms of protection were responsible.40 A protective effect was observed even when FGF1 was given 24 hours after radiation therapy.41 FGF2, given 24 and 4 hours before the first dose of fractionated TBI, aided in the recovery of bone marrow hematopoietic cells and peripheral white blood cells, red blood cells, and ­platelets.40,42 In rodents, neither the rate of tumor growth or metastasis nor the radiosensitivity of tumors seems to be affected

917

by treatment with FGF1 or FGF2.40,43,44 When radiation therapy is used alone, prophylactic use of hematopoietic growth factors has not been justified.45 Using laser Doppler flow measurements of blood flow in rodent tibia, Okunieff and colleagues8,44,46 studied whether the angiogenic effect of FGF2 would counteract the chronic antiangiogenic effect of radiation on bone growth. They found that chronic radiation bone toxicity was associated with decreased perfusion, without elevation of circulating or soft tissue TNF-β or TNF-α.8 Radiation toxicity to rodent bone (30 Gy in a single fraction to the hind limb) is alleviated by bFGF therapy, even if that therapy is begun up to 5 weeks after the irradiation. Okunieff and colleagues46 postulated that the increased laser Doppler flow of the contralateral nonirradiated tibia was due to circulating angiogenesis factors, levels of which were elevated to compensate for the ­ radiationinduced antiangiogenesis.

Effects of Irradiation on Heterotopic Ossification and Secondary Malignancies Prevention of Heterotopic Bone Formation Heterotopic ossification (HO) is thought to occur when pluripotent mesenchymal cells inappropriately differentiate into osteoblastic stem cells in sites of surgery or extensive trauma.47-52 HO is particularly problematic when it occurs near joints, because as it advances, it can bridge the two sides of the joint, limiting motion and causing significant pain.53-55 The risk of HO exceeds 33% after total hip arthroplasty,56 and the risk is still higher (approximately 50%) after extensive dissection or extensive trauma to tissues. Risk factors for HO include being male and having a history of previous postoperative HO in the ipsilateral or contralateral hip and extensive soft tissue dissection.48,57-62 Other factors that can increase the risk of HO include prior trauma or surgery, hypertrophic osteoarthritis, diffuse idiopathic skeletal hyperostosis, ankylosing spondylitis, Paget’s disease, spinal cord injury, and closed head ­trauma.47,48,57,60,63-67 Classification systems for HO after total hip arthroplasty have usually been based on anteroposterior pelvic radiographs and have not included volumetric assessments of the ossification or consideration of its clinical or functional implications. A four-category classification system for HO of the hip described by Brooker and colleagues57 is the only such system to have been validated. According to this system, class I HO is defined as an island of bone within soft tissue; class II as bone spurs from the proximal femur or pelvis with at least 1 cm between opposing bone surfaces; class III as bone spurs from the proximal femur or pelvis with less than 1 cm between opposing bone surfaces; and class IV� ������������������������������������������������������ as ankylosis. class ���������������������������������������� IV� ������������������������������� HO, which involves significant restriction in the range of motion, is uncommon, occurring in only about 2% to 7% of patients with HO.57 Classification systems for HO after total knee arthroplasty are also based on anteroposterior or lateral radiographic measurements and do not consider volume or clinical or functional effects, but none of those grading systems has been independently validated. Because no nonsurgical means have been found to prevent the continued ­formation and

918

PART 12  Musculoskeletal System

maturation of existing HO, the best management is prevention.68-70 Treatment options for the prevention of HO include the use of bisphosphonates, nonsteroidal anti-inflammatory drugs, and external beam radiation therapy (EBRT).69-79 Although bisphosphonates can delay osteoid mineralization in some cases, clinical trials have not shown evidence that bisphosphonates decrease HO formation,80 and the delay in mineralization does not seem to significantly improve the range of motion of involved hips.81 Moreover, the effects of bisphosphonates are not sustained after their discontinuation,82,83 and their use has fallen into disfavor. Nonsteroidal anti-inflammatory drugs, particularly indomethacin, have been shown in several studies71-74 to reduce the formation of heterotopic bone; the recommended dosage is 75 mg/day for 1 to 6 weeks. However, the gastrointestinal and hematologic side effects can be serious, particularly for patients who are taking steroids or anticoagulants and patients with postoperative ileus or peptic ulcer disease.77,84 Side effects from nonsteroidal antiinflammatory agents prevent more than one third of such patients from completing a 2-week treatment course (starting on postoperative day 1).85,86 These drugs are also thought to interfere with the ingrowth of bone and to affect the stability of noncemented components.86,87 Use of EBRT for HO prophylaxis was first proposed by Coventry and Scanlon,68 who showed that a 20 Gy ­cumulative dose divided into 10 postoperative treatments prevented the formation of HO after total hip arthroplasty. Irradiation effectively inhibited the pluripotential mesenchymal cells from differentiating into osteoblasts. Other subsequent attempts to reduce the cumulative dose, number of treatments, and size of the radiation fields revealed that as long as the radiation was given before the fifth day after surgery, results from single-dose regimens of 8, 7, and 6 Gy were equivalent to those from higher cumulative doses.78,88-90 Moreover, use of single doses is more cost­effective, more convenient, and more tolerable for the patient. Significant acute or subacute adverse effects of a 7-Gy postoperative radiation dose are rare, as are second malignancies.70 For the treatment of advanced HO, reoperation may be necessary, followed by a single 7-Gy dose of radiation.70 The best time for postoperative radiation therapy is within the first 72 hours after surgery.52,63,68,69,78,88,91-97 Preoperative and postoperative radiation therapy for the prevention of HO have been compared, and although neither has been proven to be better than the other, postoperative treatment has been favored by some because of the belief that the rapidly dividing postoperative cells are more radiosensitive than their preoperative counterparts.94,98-100 ­ Pellegrini and Gregoritch98 reported that preoperative treatment eliminated the discomfort and morbidity associated with postoperative radiation after total hip arthroplasty. The field for postoperative radiation therapy need include only the soft tissue plane around the neck of the implant to be treated. For example, Lo and colleagues69 treated a 5 to 6 × 14 cm field with a 25-degree collimator angle without custom blocking. Patient selection precludes definitive evaluation of the success of radiation therapy for this purpose; however, modern series with a 7-Gy postoperative dose show failure rates of 5% to 16%.69,72,78,88,101 Lo70 regards successful radiation therapy as that which can reduce the incidence of HO within the radiation field within the first 6 months after surgery to 15% or less.

Induction of Secondary Tumors As increasing numbers of patients become long-term ­cancer survivors,102,103 the prevalence of secondary cancer unfortunately also increases.102,103 Although some forms of ­cancer are themselves associated with an increased likelihood of developing secondary cancer, some secondary ­tumors arise as a result of the treatment of the ­ primary cancer and have been called iatrogenic diseases of success.104 Some patients have a genetic predisposition for the development of secondary tumors105,106; for example, in one study of 1604 patients with retinoblastoma who ­survived at least 1 year after diagnosis, the incidence of subsequent cancer was ­elevated only among the 961 patients with hereditary retinoblastoma (190 cases of cancer versus the 6.3 expected in the general population; relative risk = 30; 95% ­confidence interval [CI], 26 to 47).105 The ­cumulative incidence (±standard error) of a second cancer at 50 years after diagnosis was 51.0% (±6.2%) for those with hereditary retinoblastoma and 5.0% (±3.0%) for those with ­nonhereditary retinoblastoma.105 Radiation therapy can further increase the risk of secondary malignancies. Radiation doses as small as 10 mGy received by a fetus in utero lead to an increase in the risk of childhood cancer, with an excess absolute risk coefficient at this level of exposure of approximately 6%/Gy.107 The most common bone tumor appearing after local field irradiation is osteosarcoma, the incidence of which seems to be dose-dependent (at least at high doses).105,108-112 The accumulation of skeletal doses from radium 226 (226Ra) and 225Ra by radium dial painters was shown to be an important predictor of risk of death from bone sarcoma; the risk is even greater in those exposed before skeletal growth is complete.108,109,113 In the retinoblastoma series discussed previously,105 114 of the 190 cancer diagnoses after irradiation were sarcomas, and the sarcomas were of diverse histologic types. For soft tissue sarcomas, the relative risk values showed a stepwise increase at all dose categories and were statistically significant at 10 to 29.9 Gy and at 30 to 59.9 Gy. A radiation risk for all sarcomas combined was evident at doses above 5 Gy, rising to a 10.7-fold increase at doses of 60 Gy or greater (P < .05).105 Tucker and colleagues106 reviewed the cases of 64 patients in whom bone cancer developed after childhood cancer and compared them with 209 matched controls who had survived cancer in childhood but who did not develop bone cancer. Patients who had had radiation therapy had 2.7 times the risk of those who had not (95% CI, 1.0 to 7.7), and a sharp dose-response gradient was evident, reaching a 40-fold higher risk after doses to the bone of more than 60 Gy. After adjustment for radiation therapy, treatment with alkylating agents was also linked to bone cancer (relative risk, 4.7; 95% CI, 1.0 to 22.3), with the risk increasing as cumulative drug exposure rose.106 In a review by Kuttesch and colleagues110 of 266 survivors of Ewing’s sarcoma, the median latency to the diagnosis of the second malignancy was 7.6 years (range, 3.5 to 25.7 years). The estimated cumulative incidence rate at 20 years for any second malignancy was 9.2% (standard deviation, 2.7%) and that for secondary sarcoma was 6.5% (standard deviation, 2.4%). The cumulative incidence rate of ­secondary sarcoma depended on the radiation dose (P = .002); no secondary

35  The Bone

sarcomas developed among patients who had received less than 48 Gy, whereas the absolute risk of secondary sarcoma was 130 cases per 10,000 person-years of observation among patients who had received 60 Gy or more.110 Hematologic tumors (e.g., leukemia) and solid secondary tumors (e.g., osteosarcoma) can develop after TBI.114-117 As Niethammer and Mayer115 reported, different types of secondary cancer occur at various times after BMT; ­lymphoproliferative disease can occur during the first year, ­leukemias and myelodysplastic syndromes after several years, and solid tumors even later. The incidence of ­second neoplasias after BMT seems to be higher in children than in adults.115 When corrected for competing risks, TBI ­increases the risk of cancer in rhesus monkeys by a factor of 8.116 The risk of leukemia after irradiation of the pelvis is relatively small.118-121 Of a cohort of 110,000 women with invasive cancer of the uterine corpus who survived at least 1 year after the initial cancer (assembled from nine ­population-based cancer registries), the leukemia risk associated with partial-body radiation therapy for uterine corpus cancer was small—about 14 excess cases per 10,000 women followed for 10 years.120 The quoted risk for acute myelogenous leukemia in patients with prostate cancer treated with radiation therapy is 0.1% at 10 years.121 The risk of leukemia after radioactive iodine treatment also is small. De Vathaire and colleagues122 studied 1771 patients treated for a thyroid cancer at two institutions. None of these patients had undergone EBRT, and 1497 had received iodine 131. After a mean follow-up time of 10 years, no cases of leukemia were observed, compared with 2.5 expected according to the coefficients derived from Japanese atomic bomb survivors (P =.1).122 Strontium 89 (89Sr) has been associated with cases of leukemia in patients with metastatic prostate cancer,123 but the exact risk is unknown and needs to be investigated.

Bone Imaging Accurate diagnosis of bone tumors requires a multimodality approach with clinical-radiologic-pathologic correlation.124,125 Appropriate diagnostic and staging evaluation should be performed before biopsy, as careful imaging and review of blood chemistry results may eliminate the need for a biopsy.124 Plain radiographs in two orthogonal views (anteroposterior and lateral) can help determine tumor morphology; patterns of bone destruction (i.e., lytic, blastic, mixed, moth-eaten, or permeated); lesion margins (i.e., from well-delineated sclerotic rim to ill-defined margin) and internal characteristics (i.e., nonmatrix-producing tumors, nonmineralized matrix-producing tumors, and mineralized matrix-producing tumors); type of host bone ­response (i.e., medullary, periosteal, or none); the location (i.e., ­femur, tibia, humerus, or others), site (i.e., metaphysis, diaphysis, or epiphysis), and position (i.e., central, eccentric, or periosteal) of the lesion in the skeletal system and in the individual bone; soft tissue involvement; and number of ­ lesions.124 Plain radiographs also can help identify ­radiation-induced bone changes such as osteoradionecrosis (with its characteristically ill-defined cortical destruction) and avascular necrosis. Computed tomography (CT) and magnetic resonance imaging (MRI) are useful for determining the extent of

919

bone tumors.126 CT is particularly useful for the assessment of cortical invasion, periosteal reaction, soft tissue or lesion matrix calcification or ossification, and fracture fragment positioning or configuration, especially in complex anatomic sites such as the spine, pelvis, and hindfoot.124,125 Because bony anatomy is well defined on CT, CT scans can provide useful information regarding the best site for biopsy, such as in areas of cortical thinning. MRI can visualize the intraosseous extent of the lesion and the presence and dimensions of soft tissue masses. MRI is also useful for assessing the response of primary bone tumors to systemic treatment and for follow-up after surgical treatment.127 Because bone scintigraphy visualizes local osteoblastic activity, it is useful for assessing the entire skeleton for the presence of metastases. However, not all lesions are “hot” (i.e., detectable on bone scans); one example is skeletal lesions from multiple myeloma, which should be evaluated by skeletal survey (whole-body radiography) rather than bone scanning. Bone scintigraphy is also useful for documenting the presence or extent of the osteoblastic response ­necessary for the concentration of therapeutic radionuclides such as 89Sr or samarium 153 (153Sm)–­lexidronam.125,128,129 Gallium scans are useful for evaluating primary and metastatic lesions of malignant fibrous histiocytoma when the tumor invades the skeletal system.130 Another technique that has become widely available in recent years is 18F-fluorodeoxyglucose positron emission tomography (FDG-PET), which has also assumed an important role in oncologic evaluations. FDG-PET can be quite effective for diagnosing osseous metastases.131-135 Fujimoto and others132 retrospectively compared the diagnostic accuracy of PET with that of conventional bone scintigraphy for the diagnosis of bone metastases in 95 consecutive patients with a variety of types of cancer. In that study, findings from “whole-body” PET scans (from face to upper thigh) were compared with those from standard whole-body scintigraphic scans obtained within a month of one another. The diagnostic accuracy was comparable for both techniques, but PET had the advantage of being able to visualize the response of bone lesions to treatment and the presence of soft tissue metastases. The limited field of view used for PET in this study precluded detection of solitary bone metastases in the skull or below the femoral diaphysis.

Primary Malignant Tumors of Bone Plasmacytoma Solitary plasmacytoma of bone is a rare presentation of plasma cell dyscrasia. Solitary plasmacytomas of bone constitute approximately 2% of all cases of myeloma and 70% of all plasmacytomas.136 In the past, solitary plasmacytomas of bone were considered to be an early manifestation of multiple myeloma137-139; however, diagnostic criteria have been established to define solitary plasmacytoma of the bone as a distinct entity. According to the strictest of these criteria, solitary plasmacytoma of bone can be considered a distinct entity if no other skeletal lesions are radiographically evident, if bone marrow is free of plasma cells, and if results from serum and urine protein electrophoresis are negative. Unfortunately, published diagnostic criteria vary

920

PART 12  Musculoskeletal System

considerably,137,140-142 including histologically confirmed solitary lesion on skeletal survey (with or without a soft tissue mass), bone marrow plasmacytosis level of less than 10% or of less than 5%, and serum and urine electrophoresis findings that are positive, negative, or become negative after treatment of the lesion.136,138,140,143-147 These different diagnostic criteria make comparisons among studies difficult, and as Liebross and colleagues147 have suggested, this variation may be a factor in the different time courses that have been reported for the lesions to progress to systemic disease. Like multiple myeloma, solitary plasmacytoma of bone is more common in men, with a reported male-to-female ratio of 3 to 1. Although multiple myeloma is most common in the sixth and seventh decade, solitary plasmacytoma of bone manifests at younger ages.136,137 The most common locations for solitary plasmacytomas of bone are in the long bones, pelvis, and spine. In a review of 46 patients with solitary plasmacytomas of bone, Frassica and colleagues140 found that 54% of these patients had lesions in the spine. Pain is the most common symptom, and other symptoms tend to reflect the location of the tumor. Lesions in the spine may manifest with neurologic compromise. Neurologic changes are usually accompanied by lumbar or thoracic pain of several months’ duration. Solitary plasmacytomas of bone are exquisitely radiosensitive and their response to radiation is rapid. In one series of 37 patients with plasmacytoma (27 with solitary plasmacytomas of bone), Bolek and colleagues139 reported that radiation therapy effectively achieved local control. In that study, only one patient with solitary plasmacytoma of bone developed local failure, but all cases had progressed to multiple myeloma within 15 years. Liebross and colleagues147 found that local megavoltage radiation (median dose, 50 Gy; range, 30 to 70 Gy) achieved effective local control in 96% of their patients with solitary plasmacytomas of bone. No late toxic effects were found. Of their patients, 29 (51%) developed multiple myeloma. No radiation dose-response relationship was found for local control or subsequent progression to multiple myeloma.147 In another study of 16 patients with solitary plasmacytosis, however, Mill and Griffith142 did find a dose-response relationship between radiation dose and progression of solitary plasmacytoma of bone to multiple myeloma; they found that patients given more than 50 Gy to their lesions had less likelihood of developing multiple myeloma. Although local control rates of solitary plasmacytomas are quite good after radiation therapy (up to 90%), the question arises of whether the use of local radiation is addressing the entire clinical picture—whether radiation is achieving satisfactory disease control if more than 50% of cases reportedly progress to multiple myeloma. Aviles and colleagues,148 in attempting to address this question, ­examined the role of adjuvant chemotherapy after radiation therapy for patients with solitary ­ plasmacytomas. After a median follow-up of 8.9 years, disease-free and overall survival rates were improved among the patients treated with combined therapy: 22 of the patients in the ­combined-treatment group remained alive and free of disease compared with only 13 patients given radiation only.148 However, the groups in that study were too small to draw definitive ­conclusions regarding the use of ­adjuvant chemotherapy.

Another small study by Mayr and colleagues149 produced results concordant with those of the Aviles group; none of five patients with solitary plasmacytomas of bone who received adjuvant chemotherapy experienced progression to multiple myeloma, compared with 9 of 12 patients (75%) who did not receive chemotherapy. However, in another series of 32 cases of solitary plasmacytomas of bone, Holland and colleagues138 reported that adjuvant chemotherapy did not affect the incidence of conversion but did seem to delay it. Collectively, these findings suggest that combined-modality therapy for solitary plasmacytoma of bone warrants further investigation. Radiation therapy provides an excellent treatment option for local control.

Chondrosarcoma Chondrosarcomas were first described in 1943 by Lichtenstein and Jaffe150 as sarcomas in which the tumor cells are associated with a cartilaginous matrix. Although chondrosarcomas are the second most common primary tumor of bone, they are nonetheless uncommon. According to the Mayo Clinic, less than 9.2% of the malignant tumors in their series were chondrosarcomas.137 Lesions are referred to as central, peripheral, primary, or secondary depending on where in the bone they originate or whether they originate from preexisting lesions. However, this classification bears no relation to overall prognosis. Central chondrosarcomas arise within the medullary canal, and peripheral chondrosarcomas originate from the cortex of the bone. Primary chondrosarcomas that arise de novo tend to be central and constitute approximately 75% of all chondrosarcomas.151 Secondary chondrosarcomas constitute approximately 25% of all lesions, tend to be peripheral, and arise from preexisting benign cartilage lesions. The most common preexisting lesions from which secondary chondrosarcomas arise are enchondromas, but chondrosarcomas also can arise from Paget’s disease, fibrous dysplasia, unicameral bone cysts, irradiated bone, chondromyxoid fibromas, chondroblastomas, and synovial chondromatosis. Malignant transformation of solitary enchondromas is thought to occur in less than 3% of cases. In multiple enchondromatosis (Ollier’s disease), Maffucci’s syndrome (endochondromatosis with hemangiomas), and hereditary multiple exostoses, the incidence of transformation is higher. Primarily a tumor of adulthood, chondrosarcomas are most common in the fourth to seventh decades. They tend to have a slight male predominance. Chondrosarcomas are rare in children,152-154 although secondary chondrosarcomas tend to occur at a considerably earlier age than primary lesions.155 Local swelling and pain are the most common presenting symptoms. Pain is considered to be an indication of growth of a lesion and often heralds malignant transformation of a preexisting lesion of the bone to a secondary chondrosarcoma. Progressive pain, pain at rest, and nocturnal pain are all harbingers of malignancy. Pathologic fracture occurs in 3% to 8% of patients with chondrosarcoma.156,157 Although chondrosarcomas can arise in any bone in the body, they occur most commonly in the pelvis, hip girdle, and shoulder girdle. In the long bones of the body, these lesions tend to be metaphyseal and extend into the diaphysis. Epiphyseal involvement is rare. Chondrosarcomas have a characteristic appearance on radiography, appearing as

35  The Bone

poorly marginated destructive lesions of bone with areas of randomly distributed calcifications. Areas of lucency surrounding the matrix calcification suggest malignancy. Endosteal scalloping is common, and cortical destruction is also an indication of malignancy. New periosteal bone is rarely seen. CT (which is quite sensitive for revealing subtle cortical disruption) and MRI can provide valuable information concerning the extent of the tumor and its relationship to the surrounding anatomy. The natural history of conventional chondrosarcomas is one of relatively slow growth and low metastatic potential. However, dedifferentiated lesions, high-grade lesions, and mesenchymal lesions are often characterized by aggressive local growth and early metastatic spread. Dahlin and ­Beabout158 in 1971 defined dedifferentiated chondrosarcomas as a distinct clinical entity characterized by low-grade cartilage lesions juxtaposed to high-grade nonchondroid sarcoma. These lesions account for 6% of all chondrosarcomas. Their behavior and treatment is based on their highgrade sarcoma component. The prognosis for patients with chondrosarcoma depends largely on the histologic grade and clinical characteristics of the disease, which vary according to the disease subtype. Tumor size and anatomic site influence survival rates because they often dictate treatment and adequacy of surgical resection. The definitive treatment for chondrosarcoma is wide surgical resection. After adequate surgical treatment, survival rates are 80% to 90% for patients with grade 1 disease, 50% to 80% for grade 2 disease, and 40% for grade 3 disease.151 Inadequate surgical margins are associated with a high likelihood of local recurrence, with reported rates as high as 75% to 92%.157,159,160 As is true for other types of sarcoma, the stage of the disease is directly related to the prognosis. Metastasis at the time of initial presentation is the most powerful determinant of long-term survival.161 Chondrosarcomas tend to metastasize to lung and rarely to soft tissues and bone. Because these lesions tend to metastasize late in the course of the disease, 5-year survival statistics may not be appropriate for determining the natural history and overall prognosis of metastatic disease. The best type of surgical treatment for chondrosarcoma remains controversial. Wide surgical resection necessitates reconstruction of the skeletal defect. The options for intercalary reconstruction include the use of allografts, vascularized autografts (i.e., vascularized fibula), or prostheses. If joint reconstruction is needed, the options are allograftprosthetic composites, arthrodesis, osteoarticular allografts, and endoprosthetic reconstruction. Given the morbidity of these procedures, the trend for treating low-grade sarcomas has been for conservative, intralesional surgery with adjuvant therapy to preserve joint anatomy and function. Several studies of patients with low-grade chondrosarcomas treated with intralesional excision and adjuvant therapy have demonstrated no compromise in clinical outcome.162-164 An evaluation of 23 patients with medullary grade 1 chondrosarcoma treated with intralesional reduction showed good oncologic results.165 Chondrosarcomas are considered to be relatively radioresistant,137,161 but the biologic basis for this resistance is not known. Moussavi-Harami and others166 contend that radiation resistance is related to the expression of CDKN2A (formerly designated p16ink4a), a major tumor

921

s­ uppressor protein that participates in cell cycle regulation. The research group showed that ectopic expression of CDKN2A in an immortalized human chondrosarcoma cell line that does not normally express CDKN2A significantly enhanced the radiation sensitivity of that cell line to lowdose gamma irradiation (1 to 5 Gy) in clonogenic assays. These results suggest that restoration of CDKN2A expression may enhance the radiosensitivity of chondrosarcomas lacking its expression. In another effort to increase the radiosensitivity of chondrosarcomas, Rhomberg and colleagues investigated the effects of radiation therapy in combination with the radiation-sensitizing agent razoxane.167 Razoxane was given at a dose of 125 mg twice daily, starting 5 days before irradiation (median tumor dose, 60 Gy). Among the 13 patients with chondrosarcoma tested in this way, 8 had unresectable or recurrent disease, 2 showed no change, 5 showed partial responses, and 1 showed a complete response. Although this study was limited by the lack of a control group, the results support the possible benefit from a prospective, randomized trial for further evaluating razoxane as a radiosensitizing agent in chondrosarcoma. Cranial base chondrosarcomas remain a formidable management challenge. Reports of successful treatment of unresectable chondrosarcomas in the cervical spine and base of the skull with proton beam and three-dimensional conformal treatment techniques suggest that local control can be achieved.168-172 In one series by Rosenberg and colleagues,172 chondrosarcoma located at the base of the skull was treated with high-dose postoperative fractionated precision radiation therapy at doses of 64.2 to 79.6 cobaltgray-equivalents; 5- and 10-year local control rates were 99% and 98%, ­respectively, and 5- to 10-year survival rates were 99%. Hug and ­ colleagues170 gave high-dose proton radiation therapy postoperatively to patients with residual tumor after surgical ­resection and observed an actuarial 5year survival of 100%, with toxic effects identified in only 7% of the patients. Munzenrider and Liebsch173 found proton therapy to be “ideal for treating skull base and cervical spine tumors, for the dose can be focused while sparing the brain, brain stem, cervical cord, and optic nerves.” Wide surgical resection remains the definitive treatment, but proton-beam and three-dimensional conformational radiation therapy may offer promising results when used in the adjuvant setting or for inaccessible lesions. Long-term (15to 20-year) ­follow-up studies are still needed. Another technique being evaluated for chondrosarcomas in the head and neck region, particularly at the base of the skull, is stereotactic radiosurgery. These tumors tend to be intimately associated with cranial nerves and major blood vessels, which often precludes complete surgical resection. The numbers of patients treated with stereotactic radiosurgery for chondrosarcomas of the skull base or upper cervical spine are too few to permit conclusions to be drawn about the feasibility of this procedure for primary treatment.174,175 However, stereotactic radiosurgery can be useful as an adjuvant treatment for residual disease or small recurrences.

Primary Malignant Lymphoma of Bone Lymphoma of bone is a rare entity that was first identified by Oberling176 in 1928. However, it was not until 1939, when Parker and Jackson177 reported on their series

922

PART 12  Musculoskeletal System

B

A

C

D

E

FIGURE 35-1.  A and B, Plain radiographs of the right knee of a 13-year-old boy with pain in that knee for 3 to 4 months reveal a poorly circumscribed lytic lesion in the medial femoral condyle crossing the growth plate. C, Axial T1-weighted MR image of the right knee shows a poorly circumscribed lesion in the medial condyle extending to the trochlea. The lesion itself shows a gross decrease in signal intensity. D, Proton-weighted MR image shows a poorly defined lesion with decreased signal intensity in the medial femoral condyle. E, Bone scan shows increased uptake about the right distal femur. A biopsy revealed malignant lymphoma of large B-cell type, which was consistent with a primary large B-cell lymphoma of bone.

of “reticulum cell sarcoma,” that lymphoma of the bone was recognized as a distinct clinical entity. Multiple terms, including reticulum cell sarcoma, malignant lymphoma of bone, primary skeletal lymphoma, and osteolymphoma, have been used to describe the disease. Primary lymphoma of bone, as it is most commonly referred to, constitutes approximately 0.2% of all bone tumors and 5% of all extranodal lymphomas.178,179 Primary lymphoma of bone was initially described as a malignant lymphoid infiltrate within bone, with or without cortical invasion or soft tissue extension but without concurrent involvement of regional lymph nodes or distant viscera.180,181 Some investigators have expanded the definition to permit lymph node involvement but have stipulated a 6-month time frame between the appearance of the primary focus and the emergence of nodal involvement.182,183 Primary lymphoma of bone occurs with a male-to­female ratio of 1.8 to 1 and can occur at any age.137,184-187 The sites most commonly affected are the long bones. Pain is the most common presenting symptom, and pathologic fracture has been observed in 20% to 30% of cases.188 ­Although radiography can help establish the diagnosis,

no pathognomonic radiographic features have been determined. Radiographically, these lesions tend to be permeative and located in the metaphyses or diaphyses of long bones (Fig. 35-1A and B). Phillips and colleagues189 evaluated characteristic radiographic findings and found that pathologic fracture, layered periosteal new bone, broken periosteal new bone, cortical breakthrough, soft tissue mass, and soft tissue swelling were the most helpful criteria in establishing the prognosis of the lesion.189 CT and MRI can assess the extent of the lesion and the presence of an associated soft tissue mass (see Fig. 35-1C and D). The importance of a complete staging workup cannot be overemphasized. Such a workup consists of radiography of the bone involved, a bone scan, and CT of the abdomen and pelvis. Before CT became widely available for abdominal and pelvic evaluation, patients with primary lymphoma of bone underwent staging laparotomies to exclude other possible primary tumor sites; many extraosseous lesions may have been missed in such patients. CT scanning of the abdomen and pelvis can facilitate a more precise assessment of extraosseous disease and has therefore become an integral part of the evaluation of primary lymphoma

35  The Bone

of bone. Bone scans are used to evaluate the skeleton for evidence of multifocal disease (see Fig. 35-1E). Many investigators have described cases in which lesions that were initially considered to represent primary bone lymphoma were subsequently found to represent systemic disease after a more careful staging analysis.190 Treatment of primary lymphoma of bone remains ­controversial. Surgery is used primarily as a ­ diagnostic aid or in the treatment of pathologic fractures. Some ­investigators have recommended radiation therapy for the ­treatment of localized disease191 and chemotherapy for patients with more extensive disease.187 They have proposed that radiation be used for Ann Arbor stage 1E disease and that ­combination therapy be used for higher-stage disease. Combined-modality therapy has been advocated for all stages of disease because of the high distant relapse rates observed after treatment with radiation therapy only.180,182,192,193 Combined-modality therapy in adults has now been shown to yield survival superior to that produced by radiation alone. With the use of combined-modality therapy, the prognosis for patients with primary lymphoma of bone is considered to be good. Bacci and colleagues194 found that 88% of patients treated with chemotherapy and radiation were continuously disease free after a median follow-up of 87 months. Ferreri and colleagues182 observed that patients with primary lymphoma of bone treated with chemotherapy and whole-bone radiation to a dose of 40 Gy, followed by a boost to a total dose of 45 Gy to areas of bulky disease, had a lower local relapse rate (88% to 90%) and a more favorable 5-year survival rate (40% to 63%) than did patients treated with radiation alone. The studies of Dubey and colleagues186 suggest that radiation doses of about 46 Gy provide optimal local control and, when combined with chemotherapy, improved long-term survival. Children also may present with solitary osseous involvement of lymphoma. In pediatric cases, it is not clear whether such a presentation represents primary lymphoma of bone or an early manifestation of systemic disease.191,195 Typically, children with disease considered to be at high risk of relapse have been treated with more aggressive therapy. Loeffler and colleagues196 reported an 8-year actuarial survival rate of 90% for children with primary lymphoma of bone who were treated with combination therapy. However, they also reported in the same study the development of two cases of secondary sarcoma in the radiation field among the 11 children in whom the original disease had been successfully treated. In an attempt to reduce the incidence of side effects and the risk of secondary malignancies while maintaining similar control of the primary disease, some investigators have investigated less intensive regimens. In a series of 31 patients, Suryanarayan and colleagues191 found that patients who were treated with only 9 weeks of chemotherapy and no radiation (two thirds of the total population) had a projected 5-year survival rate of 100% and an event-free survival rate of 95%. Although long-term follow-up is still required, these results led the investigators to conclude that primary lymphoma of bone in children can be treated successfully with shorter-duration chemotherapy regimens and without radiation. The reason for the differences in outcome observed between children and adults is unclear.

923

However, until prospective randomized clinical trials have been performed, combined-modality therapy should be considered the treatment of choice for adults with primary lymphoma of bone.

Chordoma Chordomas are rare malignant tumors that arise from the remnants of the notochord. Chordomas represent approximately 1% of all malignant bone tumors. They display a predilection for the ends of the spinal cord; approximately 40% to 60% of lesions are located in the sacrococcygeal region of the spine, 15% to 20% are located in the thoracic, lumbar, and cervical spine, and approximately 40% are located at the base of the skull. Chordomas are uncommon in patients younger than 30 years. They do, however, occur in a broad age range and are most common in the fifth to seventh decades. Chordomas usually affect men more often than women, with reported ratios of between 1.2 to 1 and 2 to 1. In the Mayo Clinic series, approximately 64% of 356 patients with chordomas were male,137 whereas Bjornsson and colleagues156 reported that 48% of their 40 patients were male. Chordomas are indolent lesions, and it is not uncommon for symptoms to be present for several years before diagnosis.156 Various symptoms are related to the location and size of the tumor. Headaches, increased intracranial pressure, or cranial nerve symptoms are often associated with cranial tumors. Pain, neurologic deficits, or symptoms similar to those resulting from a herniated disk can be produced by vertebral tumors. Pain, neurologic compromise, and bowel or bladder dysfunction can result from sacral or sacrococcygeal tumors. Radiographic features of chordomas depend on their location. Because these lesions originate from the remnants of the notochord, they appear at the midline. In sacral and sacrococcygeal tumors, bony destruction is central and irregular. A soft tissue mass is usually present anteriorly, and calcification within the soft tissue mass is common. CT and MRI are helpful because routine radiography often fails to adequately delineate sacral lesions. CT can be used to delineate the lesion, assess the extent of peripheral reactive sclerosis, and define the degree and extension of the calcifications.197 MRI can delineate soft tissue mass extension, longitudinal extension of tumors, and sacral root involvement.198,199 Chordomas that arise in the mobile spine generally arise within the vertebral bodies, but extension into the posterior elements is common. Vertebral destruction is associated with peripheral sclerosis, and reactive new bone is often observed. Multiple contiguous vertebral sites of involvement can be seen. Cranial lesions are almost always associated with radiographic changes. Destruction of the clivus or sella turcica and bony destruction in the sphenooccipital and hypophyseal regions are often readily apparent. MRI of these regions facilitates visualization of soft tissue extension. Although chordomas are slow-growing malignant tumors, the prognosis is guarded. Most often, death occurs from local extension of the tumor rather than distant metastasis. In a series of 40 patients with chordoma described by Bjornsson and colleagues,200 the long-term survival rate was 37% even though only 5% of patients developed lung metastases. Patient age and tumor location are important prognostic factors, with age less than 40 years and ­proximal

924

PART 12  Musculoskeletal System

vertebral lesions associated with improved long-term survival.201,202 Depending on the location of the lesion, treatment and prognosis vary. Wide or complete excision is the treatment of choice. However, wide excision is not feasible for all tumors. Kaiser and colleagues203 reported that the recurrence rate among tumors that had been entered during surgery was approximately three times higher than that for tumors that had not been compromised at the time of surgery. Radiation has been advocated as an adjuvant treatment when surgical margins are close or positive for residual disease.168,201,204-206 Conventional fractionated high-dose and proton-photon beam therapy have been used for postoperative treatment.204,207-209 Munzenrider and Liebsch173 suggested that charged-particle beams are ideal for treating skull base and cervical chordomas because such beams can deliver precise high-dose radiation to localized areas of disease while sparing surrounding normal tissues. These techniques have proved particularly useful for spinal and cranial lesions. al-Mefty and colleagues204 concluded that proton-photon therapy effectively controlled tumors after radical surgery, but the follow-up time in that study was short. Borba and colleagues207 and Habrand and coworkers210 have advocated proton-photon therapy for children, and Borba and colleagues reported a 68% 5-year actuarial survival rate and a 63% disease-free survival rate in their series of four pediatric patients with intracranial chordomas. Conventional therapy consisting of high-dose fractionated radiation therapy has produced comparable results. Romero and colleagues211 demonstrated that the prognosis for patients given a postoperative radiation dose of 48 Gy or more was better than that for patients given doses of less than 40 Gy. Their study suggested that high-dose postoperative radiation may increase the disease-free interval for patients with microscopic residual disease. Klekamp and Samii212 found that patients treated with postoperative radiation doses of 60 to 70 Gy had a significantly longer recurrence-free interval than did patients who had suboptimal excision and no radiation therapy. These results suggest that radiation therapy with ­protonphoton or conventional high-dose beams is a valuable adjuvant treatment option for patients with incompletely resected disease. However, when possible, wide surgical resection should remain the definitive treatment of choice. As emphasized by Catton and colleagues,205 radiation therapy may play an important role in controlling microscopic disease and affording palliative pain control, but it rarely cures gross residual disease.

Malignant Fibrous Histiocytoma Malignant fibrous histiocytoma of bone is most likely a clinicopathologic entity distinct from other sarcomas of bone and from its soft tissue counterpart.213,214 An aggressive tumor, malignant fibrous histiocytoma of bone is often associated with extensive destruction of cortical bone evident on radiography; periosteal reactions are irregular, and matrix calcifications are rarely present. Histologically, malignant fibrous histiocytoma is identical to that of soft tissue, being a poorly differentiated neoplasm, often with strongly pleomorphic and hyperchromatic nuclei, probably derived from primitive mesenchymal cells such as fibroblast precursor cells. Areas of fibrogenic differentiation (often in a storiform pattern), or areas of ­histiocytes

and anaplastic ­giant cells, are often present.214-217 Malignant fibrous histiocytoma can usually be classified as storiform-pleomorphic (63%) or histiocytic (19%) based on the predominance of one of the patterns.213 Malignant fibrous histiocytoma of bone can be difficult to distinguish histologically from fibroblastic osteosarcoma, but in such cases osteoid tumor formation favors the diagnosis of ­osteosarcoma. This high-grade bone tumor (manifesting at grade 2 in 1% of cases, grade 3 in 27%, and grade 4 in 72%) is seen with fairly equal frequency in men and women. It can occur at any age, but unlike osteosarcoma, it is more common in the fifth to seventh decades.213,218 The most common clinical symptom on presentation is pain, with or without swelling or a mass; other symptoms and signs depend on the location of the lesion and can include limping, deformity, joint stiffness, and proptosis.213,218 The most common sites of involvement are the metaphyseal regions of the distal femur, proximal tibia, and proximal femur. In a series of 81 patients described by Nishida and colleagues,213 the most common sites were around the knee joint (29%) and around the pelvis and pelvic girdle, including the proximal femur (22%). The most common site of metastasis is the lung (82%).218 Predisposing factors for malignant fibrous histiocytoma of bone include Paget’s disease, bone infarcts, malignant disorders of the hematopoietic system, and the prolonged intake of corticosteroids. Several cases have been described of malignant fibrous histiocytoma of bone developing near endoprostheses of the hip. Malignant fibrous histiocytoma of bone can occur in response to irradiation, but it more often arises spontaneously; in the Nishida series,213 13 of 81 patients had a history of previous irradiation for a variety of primary tumors. Radiographically, the tumor appears aggressive, with a permeative pattern of bone destruction, ill-defined lesion margins, and a wide zone of transition from normal to abnormal bone (Fig. 35-2).213 Among the 13 patients in the Nishida series213 who had radiographs available for review, 62% of the lesions were purely osteolytic, 38% were mixed osteolytic and sclerotic, 38% caused a periosteal reaction, 85% were associated with cortical destruction and an extraosseous soft tissue mass, 92% showed no matrix calcification, none were exclusively epiphyseal in location, and a few were associated with pathologic fracture.213 Most patients described in the literature have been treated surgically with wide local excision or ­amputation.213,215,218 The prognosis of malignant fibrous histiocytoma is best for tumors treated with chemotherapy and wide resection with negative margins. The 1-, 5-, and 10-year disease-free survival rates are 72.8%, 66.5%, and 60.9%, respectively, if adequate margins are taken, compared with 59.8%, 48.3%, and 26.4% if not.213 Without chemotherapy, the prognosis for patients with malignant fibrous histiocytoma of bone is poor, with 5-year survival rates of about 15%.219 Use of neoadjuvant chemotherapy has improved diseasefree ­ survival rates in some series to 67%,220 69%,221 and 59%.222 Radiation therapy has also been shown to improve the prognosis when tumors excised with inadequate margins (2-, 5-, and 10-year disease-free survival rates of 61.5%, 61.5%, and 41.0%, compared with 58.3%, 35.0%, and 11.7%, respectively), but the differences among these numbers are not statistically significant.213 The role of

35  The Bone

r­ adiation therapy for malignant fibrous histiocytoma of bone has been minimized of late, and the treatment of choice is now chemotherapy plus wide surgical resection.

Hemangioendothelioma Originally described in soft tissues in 1982 by Weiss and Enzinger,223 epithelioid hemangioendothelioma is defined as an intermediate-grade malignant vascular neoplasm, the clinical course of which falls between that of epithelioid hemangioma and that of angiosarcoma. These tumors can be locally aggressive with a tendency to recur locally but to metastasize only rarely.224,225 Epithelioid hemangioendotheliomas can occur at almost any age but are most common during the second or third decades of life and are almost twice as common in males as in females.226,227,223 Epithelioid hemangioendothelioma most often involves the calvarium, axial skeleton, and lower limbs. Long bones most often affected are the tibia, femur, and humerus,227 and lesions can be proximal or distal. Lesions can be metaphyseal, diaphyseal, or, less commonly, epiphyseal. Multifocal disease is present at diagnosis in more than one half of cases This type of lesion tends to involve bones of the same region, most commonly the foot, and rarely involves the viscera (lung, liver, or spleen). Because of the potential for multifocal bone or visceral involvement, patients

should be thoroughly evaluated with CT scanning of the chest and abdomen, bone scintigraphy, and a full-skeleton x-ray series. The classic radiographic appearance of epithelioid hemangioendothelioma is a lytic lesion that involves multiple bones about 40% of the time, most often in a single anatomic region or extremity.228 Cortical destruction and cortical expansion are uncommon. Other imaging findings include matrix mineralization and periosteal reaction. The imaging characteristics thought to be most indicative of epithelioid hemangioendothelioma from a radiological perspective are the multifocal nature of the condition and its tendency for local clustering.229 Differential diagnoses depend on the patient’s age and include Langerhans cell histiocytosis, giant cell tumors, fibrous dysplasia, and lymphoma in adolescents and metastatic carcinoma and myeloma in adults.230 The definitive diagnosis is often difficult to make on the basis of radiologic findings alone, and a final diagnosis requires histopathologic confirmation. Histologically, epithelioid hemangioendothelioma typically appears as solid cell nests or cords of epithelial-like cells and prominent endothelial-lined vascular spaces surrounded by proliferating tumor cells.224 Intravascular ­papillary-like projections with epithelial or histiocyte-like cells may also be present.230

B

A

C

925

D

E

FIGURE 35-2.  A and B, Anteroposterior and lateral radiographs of the distal femur and proximal tibia of a 64-year-old woman with recently developed pain and swelling indicate an ill-defined lytic lesion in the distal femur. C, Axial T1-weighted MR image through the thigh shows a large soft tissue mass about the femur that extends laterally. The mass and marrow are low-density features. D, Axial T2-weighted MR image shows a very large, heterogeneous soft tissue mass located medially that appears to be within the knee joint.  E, Additional gadolinium-enhanced images through the distal femur confirm the large size and heterogeneity of the soft tissue mass.

926

PART 12  Musculoskeletal System

F

G

H I FIGURE 35-2—cont’d  F, Sagittal T1-weighted MR image shows a lesion of low signal intensity extending from the distal femur to the femoral diaphysis. G, Sagittal T1-weighted MR image shows a very large soft tissue mass located laterally in addition to the intramedullary lesion extending from the distal femur to the femoral diaphysis. H and I, Photographs of the gross specimen without the surrounding soft tissue or muscle. The soft tissue mass seems to be growing off the lateral femoral cortex and condyle. The specimen was identified as a high-grade malignant fibrous histiocytoma with minimal necrosis.

Epithelioid hemangioendothelioma of bone has a variable presentation and clinical course, and unfortunately, its behavior cannot be predicted on the basis of the radiographic or histologic features alone. Although some studies have reported that multifocal lesions are associated with better prognosis, the numbers of cases in those studies have been few.226 Kleer and others228 postulated that visceral involvement is the most important criterion for predicting poor prognosis. Definitive treatment for epithelioid hemangioendotheliomas of bone is not well established. Most studies recommend wide surgical resection,227,228,231 although curettage may also be an option, especially for multifocal lesions. Radiofrequency ablation was used successfully in ­combination with surgical resection with the goal of limiting the extent of surgery and sparing the patient amputation.232 Radiation therapy can be used alone or in conjunction with surgery, particularly for unresectable

l­esions or metastatic disease. Radiation therapy should be considered for tumors appearing in weight-bearing bones, vertebral bodies, or in wide distribution in the skeleton or foot.224,233-236 With 30 to 40 Gy given over 4.5 to 5 weeks, at least 75% of these patients should experience complete disappearance of pain.237 The role of chemotherapy is not clear at this time. Treatment plans should be tailored individually in accordance with the location and extent of disease.

Giant Cell Tumor of Bone Giant cell tumors of bone have been described as the most challenging benign tumors of bone.161 They constitute approximately 5% of all primary bone tumors and 21% of all benign tumors of bone.137 These tumors have a slight female predominance, with a female-to-male ratio of 1.3 to 1.238,239 Most commonly appearing in the third decade of life, less than 5% of giant cell tumors occur in skeletally

35  The Bone

i­mmature patients.240,241 In a Mayo Clinic series, 84% of giant cell tumors occurred in patients older than 19 years.137 Most giant cell tumors are found in the epiphyses of long bones, but they may extend into the metaphyses as well. In several published series,242-247 only 1.2% of giant cell tumors were found to involve the metaphysis or diaphysis without epiphyseal involvement. Approximately 50% were located about the knee (i.e., distal femur and proximal tibia), with the proximal humerus and distal radius being the third and fourth most common sites. Pain and swelling are the most common presenting symptoms. The incidence of pathologic fracture can range from 11% to 37%.243,248,249 Soft tissue extension is common. Radiographically, these lesions are eccentrically located, are lytic, have narrow zones of transition, and tend to be juxta-articular. The lesions lack peripheral sclerosis and surrounding periosteal reaction. Giant cell tumors can be quite aggressive and are characterized by extensive local bony destruction, cortical breakthrough, and soft tissue expansion. Campanacci and colleagues243 proposed a grading system for giant cell tumors that was based on their radiographic appearance. The Campanacci system is similar to that proposed by Ennecking250 for benign bone tumors. No correlation has been found between these grading systems and prognosis. The natural history of giant cell tumors varies widely and can include local bony destruction, local metastasis, metastasis to the lung, metastasis to lymph nodes (rare), or malignant transformation. Approximately 3% of giant cell tumors eventually metastasize to the lung. These metastases appear as clusters of giant cell tumors located within the lung.251-255 Lung metastases appear an average of 3 to 5 years after the initial diagnosis of the primary lesion. The significance of these metastases is unclear; reports have been made of multiple unresectable lung lesions that remain stable and some that spontaneously regress.251,256-260 Long-term follow-up is important for patients with metastatic disease. Spontaneous malignant transformation of giant cell tumor is relatively uncommon. Dahlin137 defined malignant transformation as a sarcoma associated with a typical giant cell tumor at presentation or as a sarcoma arising at the site of a preexisting giant cell tumor. Malignant transformation can occur as long as 25 years after the initial presentation.261,262 Many investigators have hypothesized that ­malignant transformation of giant cell tumors may be caused by or at least associated with radiation ­therapy.137,161,263,264 Treatment of giant cell tumor has ranged from simple curettage to wide en bloc resection. Wide en bloc resection results in the lowest recurrence rate but also necessitates prosthetic reconstruction and is associated with considerable surgical and functional morbidity. However, simple ­curettage leads to minimal surgical and functional morbidity but has produced recurrence rates as high as 30% in some series.249,265-267 Attempts to treat giant cell tumors with simple curettage and high-speed burring of the cavity followed by reconstruction with insertion of polymethylmethacrylate or a bone graft and have resulted in slightly lower local recurrence rates (20% to 25%).268,269 Adjuvant therapy with phenol, argon laser, or cryotherapy to the curetted cavity, followed by packing with ­polymethylmethacrylate, has further reduced recurrence rates to 2% to 10%.270-276

927

Radiation therapy has been used as a primary or adjunct therapy for giant cell tumors. Although these lesions were previously considered to be radioresistant, more recent ­findings suggest that they may be radiosensitive.277 Radiation is generally not recommended in the orthopedic literature as a primary or adjunctive treatment for lesions of the axial skeleton264 because of concerns regarding malignant transformation. However, those concerns were based on studies done when giant cell tumors were treated with orthovoltage radiation. In those studies, patients were given several orthovoltage treatments, resulting in a high cumulative dose and high dose to the bone. Megavoltage radiation, at doses of 40 to 70 Gy, limits the dose to the bone while still preventing tumor progression. Because malignant transformation seems to depend on dose, lower rates of transformation have been observed when modern radiation therapeutic techniques are used. Megavoltage treatments have become the preferred radiation regimen.277-281 Radiation therapy has been proposed to treat patients who cannot undergo surgery for medical reasons, those with tumors in locations not amenable to operative treatment, or those in whom tumor relapse or subsequent surgery poses substantial for morbidity.279,282 Chakravarti and colleagues278 concluded from their study of 20 patients that giant cell tumors of bone can be effectively treated with megavoltage radiation, with clinical results comparable to those observed after curettage and adjuvant treatment. A series of 20 patients reported by Nair and Jyothirmayi283 also suggested that treatment with radiation at doses of 40 to 60 Gy, given in 15 to 30 fractions, also produced results comparable to those from surgical treatment, leading these investigators to conclude that radiation could achieve local control without major complications or risk of malignant transformation. A report from Caudell and colleagues from The University of Texas M.  D. Anderson Cancer Center284 retrospectively compared outcome after radiation therapy with or without surgical resection among 25 consecutive patients with pathologically confirmed giant cell tumor of bone treated during a 44-year time span (1956-2000). Of these 25 patients, 13 had been referred for treatment of a primary tumor, and 12 had locally recurrent disease after one or more other treatments; 14 had undergone radiation therapy for gross disease, and 11 had undergone radiation therapy after gross total resection. Actuarial rates of local control and distant metastasis-free survival at 5 years were 62% and 81%, respectively. Further analysis suggested that treatment for recurrent (as opposed to primary) disease correlated with lower disease-free survival rate (83% versus 33%, P < .06), lower distant metastasis-free survival rate (100% versus 64%, P < .08), and lower local control rate (83% versus 42%, P < .08) at 5 years. The conclusion from this study was that radiation therapy could be considered as an adjuvant to surgery or, for tumors that are unresectable or for which excision would result in substantial functional deficits, as an alternative to surgery. This analysis also underscored the importance of delivering radiation therapy early during treatment because of the lower likelihood of local control among patients who had undergone multiple prior therapies. In each of these studies, radiation therapy has been shown to be a valuable adjuvant treatment option for ­giant cell tumor. However, the likelihood of long-term disease control must be weighed against the risk of ­ sarcomatous

928

PART 12  Musculoskeletal System

transformation from radiation therapy. Even though no cases of malignant transformation were reported, the ­median follow-up times for the studies summarized here (9.9 years,278 48 months,283 and 8.8 years284) might have been too short for malignant transformation to become apparent. The median time for malignant transformation is thought to be approximately 10 years. In summary, operative treatment consisting of curettage and reconstruction with methylmethacrylate or bone grafting combined with adjuvant therapy remains the standard of care for giant cell tumor. However, radiation therapy may offer an alternative treatment option for inaccessible giant cell tumors, tumors of the axial skeleton, and tumors in individuals who are not candidates for surgery.

Eosinophilic Granuloma Eosinophilic granuloma is a solitary or multifocal osseous lesion that is part of a spectrum known as Langerhans cell histiocytosis or, in earlier days, histiocytosis X. Langerhans cell histiocytosis consists of a systemic fulminant form (Letterer-Siwe disease), a chronic disseminated form (HandSchüller-Christian disease), and a localized form limited to bone or lung (eosinophilic granuloma).285 Most patients (60% to 80%) with Langerhans cell histiocytosis have the localized form, but they can develop disseminated disease. If dissemination is going to occur, it most commonly does so during the first year after presentation.286 Mononuclear cells of the dendritic line, found in the epidermis but derived from bone marrow precursors, are the primary proliferative element. These cells, known as Langerhans cells, characteristically contain cytoplasmic organelles called Birbeck granules. Older lesions may appear dominantly fibrotic and be misdiagnosed as chronic osteomyelitis.285 The age incidence for single lesions peaks between 5 and 10 years, with most lesions occurring in patients younger than 20 years. Almost all series report predominance in boys, with a male-to-female ratio of 2 to 1. Almost any bone can be involved, and the appearance of the lesions on skeletal surveys depends on the bones involved. The distribution of lesions varies among case series, but the skull, mandible, ribs, vertebrae, and tubular long bones of the extremities usually are the bones most often involved. Radiographically, the bone lesions often have a characteristic punched-out appearance, with a periosteal reaction that resembles an onion’s skin (Fig. 35-3), although lesions from eosinophilic granuloma can have other manifestations. ­Spinal involvement is rare. Although vertebral plana, in which massive destruction of the entire vertebral body leads to vertebral body collapse, is considered a radiographic hallmark of spinal eosinophilic granuloma, it seems to be present in only about 40% of cases that involve the spine. To rule out multifocal osseous disease technetium99 the imaging workup should include skeletal survey. Because bone scanning fails to demonstrate the lesion in approximately one third of cases, the imaging workup should include a skeletal ­survey.287 Treatment modalities are based on the type of bone ­involvement (solitary or multiple lesions, with or without involvement of other organ systems) and the site and size of the tumor and have included curettage, bone-­grafting, chemotherapy, local or systemic corticosteroids, and radiation therapy.288 Aggressive surgical curettage is an

a­ cceptable method of treatment for many lesions, although total ­removal of the lesion is not always necessary.289 Even biopsy with or without curettage can initiate healing, and some lesions simply resolve spontaneously over months to years. Intralesional corticosteroid injection is a minimally invasive, effective, convenient, and safe method for controlling local disease.290 Systemic disease is often treated with chemotherapy, including vinblastine and etoposide.290-293 When localized lesions are symptomatic or occur in weightbearing bones or critical sites of the head and neck such as the orbit, mastoid, or mandible, radiation therapy is the treatment of choice.294 Eosinophilic granulomas are exquisitely radiosensitive, and radiation can be quite effective for persistent or ­recurrent painful lesions.288,295 Most radiation oncologists favor doses in the range of 6 to 8 Gy, given in 3 or 4 fractions and using fields that include any soft tissue extension. Some reports describe the use of larger radiation doses (e.g., 10 to 15 Gy) for adults and for patients whose disease includes a soft tissue component.294-297 In-field recurrences are rare after radiation therapy, unless the extent of soft tissue involvement is underestimated.294,295 In a study from the University of California at Los Angeles,295 35 (88%) of 40 bone lesions were controlled by radiation, but when soft tissues were involved, the rate of control dropped to 69% (11 of 16). All in-field recurrences of disease occurred in adults who had involvement of multiple bones and soft tissue extension.295 If a lesion thought to be eosinophilic granuloma fails to resolve after retreatment with an additional 10 to 15 Gy, the initial diagnosis should be reconsidered, especially if the lesion is solitary and in the femur, pelvis, or vertebral bodies. Given the availability of effective minimally invasive treatment options, use of radiation therapy for children should be restricted to those rare cases in which lesions are aggressive and unresectable.

Radiation Therapy for Metastatic Bone Disease Bone metastases can cause pain because of periosteal stretching, muscle spasm, mechanical damage, or nerve ­entrapment.298 The destructive effects of metastases do not result from direct damage to the bone by tumor cells; ­rather, the tumor cells secrete osteoclast-activating factors that recruit and activate osteoclasts and stimulate peritumoral osteoblastic activity.299 Radiation therapy for palliation of pain can be quite effective, yielding complete pain relief in approximately 50% of patients and partial pain relief in an additional 40%.300,301 Ironically, however, the use of radiation can increase the risk of pathologic fractures, as does cortical destruction by the bone metastases.302 Pathologic fractures are defined as fractures resulting from physiological loading. Predicting the likelihood of pathologic fracture is difficult and has been the subject of a great deal of investigation,303-305 as has the optimal treatment modality for patients at risk. The risk of pathologic fracture is assessed on the basis of the patient’s symptoms, including pain, and evidence of cortical destruction and lesion location and type (lytic or blastic) on plain radiographs. In 1989, Mirels306 described a scoring system for predicting the risk of pathologic fracture that is still in wide use today. In that system, up to 3 points

35  The Bone

A

C

929

B

D

FIGURE 35-3.  A, Anteroposterior radiograph of the right forearm of a 4-year-old boy with pain of several months’ duration and no history of significant trauma shows a lytic lesion within the midshaft of the radius with significant periosteal reaction about the lesion. A fracture may be present within this lesion. Significant periosteal reaction can be observed. B, Lateral radiograph shows a large lytic lesion within the distal radius. C, CT scan through the middle forearm shows significant periosteal reaction around the radius, with “onion skin” layering. D, CT scan through the diaphyseal portion of the radius shows a circumferential periosteal reaction, the absence of a radial cortex, and thinning of the periosteal reaction. The final diagnosis was eosinophilic granuloma.

are assigned to each of four variables: location of the lesion, pattern of destruction, amount of destruction, and amount of pain. According to this scoring system, patients with a score of 10 or more were prone to fracture, whereas those with a score of 7 or less could be treated conservatively with radiation alone. In general, purely lytic lesions that have destroyed more than 50% the cortex of a weightbearing bone carry the highest risk of fracture, and bones so affected are best treated with operative stabilization. Assessment of impending fracture and the choice of treatment option must be made on an individual basis, with several factors weighed before creating a treatment plan. The extent of disease (osseous and systemic), the histology of the tumor, and the patient’s expected survival time, functional status, and comorbid conditions should all be considered. Operative stabilization of an upper extremity is generally thought to be warranted if pathologic fracture is deemed impending and the patient is expected to survive for more than 3 months; operative stabilization

of a lower extremity is warranted if pathologic fracture is impending and the patient is expected to survive for more 1 month. A brief fractionated course of radiation therapy (e.g., 20 Gy given in 5 fractions) is routinely given after surgery.302,307 For patients in whom prophylactic operative fixation can be avoided, radiation therapy is an excellent palliative treatment option. Radiation therapy is thought to act on osseous metastases by causing necrosis and degeneration of the tumor cells, followed by collagen proliferation and the production of rich vascular fibrous stroma, within which intense osteoblast activity lays down new bone.308 Radiographically, the maximal radiographic response (healing of the lesion) is visible 3 months after irradiation of the ­lesion. Radiation therapy is an effective means of relieving pain from bone metastases, particularly for those with metastases from prostate or breast primary tumors.300 In general, the time to pain relief is on the order of weeks; 96% of

930

PART 12  Musculoskeletal System

patients who eventually experience minimal relief do so within the first 4 weeks of treatment, but almost 50% of those who achieve complete relief do so for the first time more than 4 weeks after treatment. The treatment effect lasts for several months (medians, 12 weeks for complete relief and 23 weeks for partial relief).300 The rate of pathologic fracture after radiation therapy is approximately 8%.300,309 Persistent or recurrent pain after therapy is an indication for prophylactic fixation.298,310 With regard to local radiation therapy, the benefit of fractionated compared with single-dose radiation therapy remains controversial, although most North American physicians still use fractionated schedules.311-322 The most common fractionation scheme in the United States is 30 Gy given in 10 fractions.323 In 1982, the Radiation Therapy Oncology Group (RTOG) reported that among more than 1000 patients (266 with solitary metastases and 750 with multiple osseous metastases), the low-dose, shortcourse schedules seemed to be as effective as high-dose, long-course schedules; however, in 1985, Peter Blitzer301 reanalyzed the data and found that evaluating the percentage of patients with “complete combined relief” (i.e., success meant that pain and use of narcotics fall to zero) revealed a trend favoring protracted programs (40.5 Gy in 15 2.7Gy fractions and 30 Gy in 10 3-Gy fractions). For patients with solitary metastases, complete combined relief occurred in 55% of patients (38 of 69) treated with 15 2.7-Gy fractions compared with only 37% (25 of 68) treated with 5 4-Gy fractions. For those with multiple metastases, complete combined relief occurred in 46% of patients (72 of 158) treated with 10 3-Gy fractions compared with only 36% (48 of 135) treated with 5 3-Gy fractions, 40% (59 of 147) treated with 5 4-Gy fractions, and 28% (39 of 137) treated with 5 5-Gy fractions. Multivariate analysis of the solitary and multiple metastasis groups together showed the number of fractions to be associated with complete combined relief (P = .0003) but not the dose per fraction (P > .10).301 The practice at M.  D. Anderson Cancer Center is to treat bone metastases with 30 Gy given in 10 3-Gy fractions. Other institutions support the use of single-dose or shortcourse local-field radiation therapy as an equally ­effective, more convenient option.314,317,318,324-328 In a ­ prospective randomized trial of 270 patients, Hoskin and colleagues318 demonstrated a dose-response relationship with singledose radiation therapy: at 4 weeks, 8 Gy provided a higher probability of pain relief than 4 Gy, although the ability of single-dose radiation therapy to provide pain relief in this study (27% to 35% complete relief and 58% to 88% partial relief) was less than that in the RTOG trials of fractionated treatment.314 Similarly, Uppelschoten and colleagues309 found that 6-Gy single-dose, local-field radiation provided partial pain relief in 88% of patients with painful bony metastases and complete relief in 39%. When the treatment was given to the vertebrae, in-field spinal cord compression developed in 9% of patients; when given to the pelvis or femur, in-field fractures developed in 8%.309 This experience is similar to the 8% risk of fracture after fractionated radiation therapy observed in the RTOG series.300 These findings require further confirmation, but they suggest that 6- to 8-Gy single-dose radiation therapy may prove to be a useful alternative to fractionated regimens such as 10 3-Gy fractions, although trends have been found ­favoring

the more protracted schedules in terms of pain relief and recalcification.300,309,314,318,327-329 To further investigate the relative benefit of single-fraction compared with multiple-fraction radiation therapy for the treatment of painful metastases, several prospective randomized controlled studies have been conducted. In a study by Kaasa and colleagues330 of 376 patients, treatment outcomes were similar between groups, with both groups experiencing similar pain relief within the first 4 months that was maintained throughout the follow-up period. No differences were found in fatigue or global quality of life, and survival times were similar in both groups. These findings were confirmed by the Dutch Bone Metastasis Study, which prospectively enrolled 320 patients who were randomly assigned to receive a single fraction of 8 Gy or 24 Gy in 6 fractions for painful bone metastases.331 All 320 patients survived more than 52 weeks, and the response duration and progression rates were similar between the two groups. In a systemic review and meta-analysis of randomized trials comparing single-fraction and multifraction radiation therapy for the palliation of metastatic bone pain, Sze and colleagues332 identified 11 trials involving 3435 patients. Meta-analyses revealed no significant difference between single-fraction and multifraction radiation therapy in reducing the overall pain response, but receipt of singlefraction therapy was associated with a higher percentage of pathologic fractures and a greater need for retreatment. ­Although single-fraction therapy may be as effective as multifraction therapy for relieving pain, the increased incidence of pathologic fracture and the need for retreatment may render it less economically feasible in terms of health care dollars and patient outcome. Studies of hemibody irradiation (HBI) confirm the superiority of fractionated radiation for durable palliation. For single-fraction HBI, the RTOG 7810 trial showed doses of 6 Gy for upper-body HBI and 8 Gy for lower-body HBI to be the most effective and least toxic.333 Zelefsky and colleagues334 at Memorial Sloan-Kettering Cancer Center compared fractionated HBI (25 to 30 Gy in 9 or 10 fractions) with 6-Gy (upper body HBI) or 8-Gy (lower body HBI) single doses. The need for retreatment was far less in the fractionated group (2 [13%] of 15) than in the single-dose group (10 [71%] of 14) (P = .001). Palliation also lasted longer in the fractionated group (8.5 versus 2.8 months).334 The major dose-limiting toxic effect of HBI is hematologic (i.e., thromboleukopenia). The major nonhematologic side effects are gastrointestinal, which can be alleviated by premedication with ondansetron or other antiemetic agents.335,336 An advantage of HBI is the speed of relief; most patients have significant relief of pain within 2 weeks of completing fractionated HBI and within 72 hours of completing single-dose HBI.333,334,337,338 Considering the relatively short life span of patients with widespread bony metastases, the duration of pain relief from HBI is excellent. In a series described by Algara and colleagues,338 the mean duration of response to single-fraction HBI (6 Gy for upper HBI and 8 Gy for lower HBI) was 101 days, or 70% of the expected life span. The RTOG also has investigated the use of single-dose HBI (6 Gy for upper and 8 Gy for lower body) after ­localfield fractionated radiation therapy (10 3-Gy fractions) to delay the development of new sites of disease (RTOG 8206; phase III; N = 499). Although toxicity rates were

35  The Bone

higher in the group given HBI than in those given fractionated radiation, toxicity was reversible and was balanced by a doubling in the time to the appearance of new disease (from 6.3 to 12.6 months).339 An alternative to HBI for the management of micrometastatic disease is the use of radioisotopes such as 89Sr (Metastron).340 89Sr, with a half-life of 50.6 days and maximum energy of 1.4 MeV, is a popular radiopharmaceutical for the treatment of bone metastases, with overall response rates as high as 80% for patients with metastases from prostate or breast cancer.341-345 Relief begins in 1 to 3 weeks, peaks at 6 weeks, and lasts for a mean of 12 months.346 Radioisotopes are particularly useful when patients (with adequate marrow reserve and no evidence of impending spinal cord compression) have bone scan evidence of metastatic bony disease on both sides of the diaphragm.346,347 89Sr is less toxic to the bone marrow than its predecessor isotope 32P, although multiple or sequential high doses should be used with caution because of myelotoxicity.341 The boneto-marrow uptake ratio for 89Sr is 10 to 1, and the retention time varies greatly (e.g., 11% to 88% at 3 months) depending on the degree of metastatic bony involvement.348-351 Up to 22% of patients are pain-free at 3 months.346 The optimum palliative dose is 40 µCi/kg (1.48 MBq/kg).344 With hematologic surveillance, repeated treatment is an option, and the efficacy is potentially the same as that achieved with the first dose.352 Low-dose concomitant platinum therapy has been used to further improve the pain control achieved with 89Sr alone.353,354 153Sm-lexidronam is another bone-seeking radiopharmaceutical licensed for use in the United States.355 The relative advantage of 153Sm-lexidronam is its short half-life (46.3 hours) and low energy (mean beta particle energies of 810 keV [20%], 710 keV [50%], and 640 keV [30%]; mean gamma photon energy of 103 keV).356,357 Pain relief from 153Sm-lexidronam is prompt, with significant pain improvement as early as the first week after drug administration.358 Pain relief occurs in most patients (61% to 90%) and lasts at least 3 or 4 months.358-365 Repeated administrations are possible with careful monitoring of white blood cell and platelet counts; with a standard dose of 1.0 mCi/kg, nadir reportedly occurs in 3 to 5 weeks and recovery is within 5 to 8 weeks of drug administration.362

REFERENCES    1. Eifel PJ, Donaldson SS, Thomas PR. Response of growing bone to irradiation: a proposed late effects scoring system. Int J Radiat Oncol Biol Phys 1995;31:1301-1307.    2. Kember N. Cell population kinetics of bone growth in long bones. Clin Orthop 1971;76:213-230.    3. Kember N, Sissons H. Quantitative histology of the human growth plate. J Bone Joint Surg Br 1976;58:426-435.    4. Walker K, Kember N. Cell kinetics of growth cartilage in the rat tibia. I. Measurements in young rats. Cell Tissue Kinetics 1972;5:401-408.    5. Eifel PJ. Decreased bone growth arrest in weanling rats with multiple radiation fractions per day. Int J Radiat Oncol Biol Phys 1988;15:141-145.    6. Alheit H, Baumann M, Thames HD, et al. Fractionation effect on radiation-induced growth retardation of tibia in rabbits and rats. Acta Oncol 1998;37:151-158.    7. Wang X, Ding I, Xie H, et al. Hyperbaric oxygen and basic ­fibroblast growth factor promote growth of irradiated bone. Int J Radiat Oncol Biol Phys 1998;40:189-196.

931

   8. Okunieff P, Wang X, Rubin P, et al. Radiation-induced changes in bone perfusion and angiogenesis. Int J Radiat Oncol Biol Phys 1998;42:885-889.    9. Wechsler-Jentzsch K, Huepfel H, Schmidt W, et al. Failure of hyperfractionated radiotherapy to reduce bone growth arrest in rats. Int J Radiat Oncol Biol Phys 1993;26:427-431.   10. Probert JC, Parker BR. The effects of radiation therapy on bone growth. Radiology 1975;114:155-162.   11. Willman K, Cox R, Donaldson S. Radiation induced height ­impairment in pediatric Hodgkin’s disease. Int J Radiat Oncol Biol Phys 1994;28:85-92.   12. Neuhauser EB, Wittenborg MH, Berman CZ. Irradiation ­effects of roentgen therapy on the growing spine. Radiology 1952;59:637-649.   13. Riseborough E, Grabias S, Burton R, et al. Skeletal alterations following irradiation for Wilms’ tumor. J Bone Joint Surg 1976;58:526-536.   14. Willich E, Kuttig H, Pfeil G, et al. Wirbelsaulenveranderungen nach bestrahlung wegen wilmstumor im kleinkindesalter. Retrospektive interdisziplinare langzeitstudie an 82 kindern. Strahlenther Onkol 1990;166:815-821.   15. Heaston D, Libshitz H, Chan R. Skeletal effects of megavoltage irradiation in survivors of Wilms’ tumor. AJR Am J Roentgenol 1979;133:389-395. 16. Silverman C, Thomas P, McAlister W, et al. Slipped capital femoral epiphysis in irradiated children: dose volume and age relationships. Int J Radiat Oncol Biol Phys 1981;7:1357-1363. 17. Libshitz H, Edeikin B. Radiotherapy changes of the pediatric hip. AJR Am J Roentgenol 1981;137:585-588. 18. Cohen A, van Lint MT, Uderzo C, et al. Growth in patients after allogeneic bone marrow transplant for hematological diseases in childhood. Bone Marrow Transplant 1995;15:343-348. 19. Kroll SS, Woo SY, Santin A, et al. Long-term effects of radiotherapy administered in childhood for the treatment of malignant diseases. Ann Surg Oncol 1994;1:473-479. 20. Tamurian RM, Damron TA, Spadaro JA. Sparing radiation­induced damage to the physis by radioprotectant drugs: laboratory analysis in a rat model. J Orthop Res 1999;17:286-292. 21. Damron TA, Spadaro JA, Tamurian RM, et al. Sparing of radiation-induced damage to the physis: fractionation alone compared to amifostine pretreatment. Int J Radiat Oncol Biol Phys 2000;47:1067-1071. 22. Damron TA, Spadaro JA, Margulies B, et al. Dose response of amifostine in protection of growth plate function from irradiation effects. Int J Cancer 2000;90:73-79. 23. Holm K, Nysom K, Rasmussen MH, et al. Growth, growth hormone and final height after BMT. Possible recovery of irradiation-induced growth hormone insufficiency. Bone Marrow Transplant 1996;18:163-170. 24. Hovi L, Saarinen-Pihkala UM, Vettenranta K, et al. Growth in children with poor-risk neuroblastoma after regimens with or without total body irradiation in preparation for autologous bone marrow transplantation. Bone Marrow Transplant 1999;24: 1131-1136. 25. Cohen A, Rovelli A, Bakker B, et al. Final height of patients who underwent bone marrow transplantation for hematological disorders during childhood: a study by the Working Party for Late Effects-EBMT. Blood 1999;93:4109-4115. 26. Arikoski P, Komulainen J, Voutilainen R, et al. Reduced bone mineral density in long-term survivors of childhood acute lymphoblastic leukemia. J Pediatr Hematol Oncol 1998;20:234240. 27. Fletcher BD, Crom DB, Krance RA, et al. Radiation-induced bone abnormalities after bone marrow transplantation for childhood leukemia. Radiology 1994;191:231-235. 28. Barr JD, Barr MS, Lemley TJ, et al. Percutaneous vertebroplasty for pain relief and spinal stabilization. Spine 2000;25:923-928. 29. Clement-De Boers A, Oostdijk W, Van Weel-Sipman MH, et al. Final height and hormonal function after bone marrow transplantation in children. J Pediatr 1996;129:544-550. 30. Murray RD, Brennan BM, Rahim A, et al. Survivors of childhood cancer: long-term endocrine and metabolic problems dwarf the growth disturbance. Acta Paediatr Suppl 1999;88:5-12.

932

PART 12  Musculoskeletal System

31. Brauner R, Adan L, Souberbielle JC, et al. Contribution of growth hormone deficiency to the growth failure that follows bone marrow transplantation. J Pediatr 1997;130:785-792. 32. Locatelli F, Giorgiani G, Pession A, et al. Late effects in children after bone marrow transplantation: a review. Haematologica 1993;78:319-328. 33. Dahllof G, Rozell B, Forsberg CM, et al. Histologic changes in dental morphology induced by high dose chemotherapy and ­total body irradiation. Oral Surg Oral Med Oral Pathol 1994;77: 56-60. 34. Dahllof G, Krekmanova L, Kopp S, et al. Craniomandibular dysfunction in children treated with total-body irradiation and bone marrow transplantation. Acta Odontol Scand 1994;52:99105. 35. Dahllof G. Craniofacial growth in children treated for malignant diseases. Acta Odontol Scand 1998;56:378-382. 36. Gal TJ, Munoz-Antonia T, Muro-Cacho CA, et al. Radiation ­effects on osteoblasts in vitro: a potential role in osteoradionecrosis. Arch Otolaryngol Head Neck Surg 2000;126:1124-1128. 37. Ashamalla HL, Thom SR, Goldwein JW. Hyperbaric oxygen therapy for the treatment of radiation-induced sequelae in children. The University of Pennsylvania experience. ­ Cancer 1996;77:2407-2412. 38. Wurzler KK, DeWeese TL, Sebald W, et al. Radiation-induced impairment of bone healing can be overcome by recombinant human bone morphogenetic protein-2. J Craniofac Surg 1998;9:131-137. 39. Coquard R. Late effects of ionizing radiations on the bone marrow. Cancer Radiother 1997;1:792-800. 40. Ding I, Wu T, Matsubara H, et al. Acidic fibroblast growth factor (FGF1) increases survival and haematopoietic recovery in total body irradiated C3H/HeNCr mice. Cytokine 1997;9:59-65. 41. Ding I, Huang K, Wang X, et al. Radioprotection of hematopoietic tissue by fibroblast growth factors in fractionated radiation experiments. Acta Oncol 1997;36:337-340. 42. Okunieff P, Mester M, Wang J, et al. In vivo radioprotective ­effects of angiogenic growth factors on the small bowel of C3H mice. Radiat Res 1998;150:204-211. 43. Ding I, Huang K, Snyder ML, et al. Tumor growth and tumor radiosensitivity in mice given myeloprotective doses of fibroblast growth factors. J Natl Cancer Inst 1996;88:1399-1404. 44. Okunieff P, Abraham EH, Moini M, et al. Basic fibroblast growth factor radioprotects bone marrow and not RIF1 tumor. Acta Oncol 1995;34:435-438. 45. Janssens P, Mitine C, Beauduin M, et al. Is there potential for granulocyte or granulocyte-macrophage colony stimulating factors in radiotherapy? Eur J Cancer 1994;30A:642-645. 46. Okunieff P, Wang X, Li M, et al. Chronic radiation bone toxicity is associated with decreased perfusion without elevation of circulating or soft tissue TGF beta or TNF alpha. Adv Exp Med Biol 1998;454:325-333. 47. Jowsey J, Coventry MB, Robins PR. Heterotopic ossification: theoretical consideration, possible etiologic factors, and a clinical review of total hip arthroplasty patients exhibiting this phenomenon, in the hip. Presented at the Fifth Open Scientific Meeting of the Hip Society, St. Louis, MO, 1977. 48. Ritter MA, Vaughan RB. Ectopic ossification after total hip ­arthroplasty: predisposing factors, frequency, and effect on ­results. J Bone Joint Surg Am 1977;59:345-351. 49. Nollen AJ, Slooff TJ. Para-articular ossification after total hip ­replacement. Acta Orthop Scand 1973;44:230-241. 50. Buring K. On the origin of cells in heterotopic bone formation. Clin Orthop 1975;110:293-301. 51. Chalmers J, Grary DH, Rush J. Observations on the induction of bone in soft tissue. J Bone Joint Surg Br 1975;57:36-45. 52. Parkinson JR, Evarts CM, Hubbard LF. Radiation therapy in the prevention of heterotopic ossification after total hip arthroplasty. The Hip: Proceedings of the Tenth Open Scientific Meeting of The Hip Society. St Louis, MO: CV Mosby; 1982, pp 211-227. 53. Garland DE. A clinical perspective on common forms of acquired heterotopic ossification. Clin Orthop Relat Res 1991;263: 13-29.

54. Stover SL, Niemann KM, Tulloss JR. Experience with surgical resection of heterotopic bone in spinal cord injury patients. Clin Orthop Relat Res 1991;263:71-77. 55. Fingeroth RJ, Ahmed AQ. Single dose 6 Gy prophylaxis for heterotopic ossification after total hip arthroplasty. Clin Orthop Relat Res 1995;317:131-140. 56. Nollen AJ, van Douveren FQ. Ectopic ossification in hip arthroplasty: a retrospective study of predisposing factors in 637 cases. Acta Orthop Scand 1993;64:185-187. 57. Brooker AF, Bowerman JW, Robinson RA. Ectopic ossification following total hip replacement: incidence and a method of classification. J Bone Joint Surg Am 1973;55:1629-1632. 58. DeLee J, Ferrari A, Charnley J. Ectopic bone formation following low friction arthroplasty of the hip. Clin Orthop 1976;121: 53-59. 59. Kjaergaard-Andersen P, Hougaard K, Linde F. Heterotopic bone formation after total hip arthroplasty in patients with primary or secondary coxarthrosis. Orthopedics 1990;13:1211-1217. 60. Ahrengart L, Lindgren U. Functional significance of heterotopic bone formation after total hip arthroplasty. J Arthroplasty 1989;4:125-131. 61. Lazansky MG. Complications revisited: the debit side of total hip replacement. Clin Orthop 1973;95:96-103. 62. Sodemann B, Persson PE, Nilsson OS. Nonsteroid anti­inflammatory drugs prevent the recurrence of heterotopic ossification after excision. Arch Orthop Trauma Surg 1990;109:5356. 63. Ayers DC, Evarts CM, Parkinson JR. The prevention of heterotopic ossification in high-risk patients by low-dose radiation therapy after total hip arthroplasty. J Bone Joint Surg Am 1986;68:1423-1430. 64. Evans EB, Smith JR. Bone and joint changes following burns: a roentgenographic study. Preliminary report. J Bone Joint Surg Am 1959;41:785-799. 65. Bisla RS, Ranawat CS, Inglis AE. Total hip replacement in ­patients with ankylosing spondylitis with involvement of the hip. J Bone Joint Surg Am 1976;58:233-238. 66. Ahrengart L, Lindgren U. Heterotopic bone after hip arthroplasty. Defining the patient at risk. Clin Orthop 1993;293: 153-159. 67. Han CD, Choi CH, Suh CO. Prevention of heterotopic bone formation after total hip arthroplasty using 600 rad in single dose in high risk patient. Yonsei Med J 1997;38:96-100. 68. Coventry MB, Scanlon PW. The use of radiation to discourage ectopic bone: a nine-year study in surgery about the hip. J Bone Joint Surg Am 1981;63:201-208. 69. Lo TC, Healy WL, Covall DJ. Heterotopic bone formation after hip surgery: prevention with single-dose postoperative hip irradiation. Radiology 1988;168:851-854. 70. Lo TC. Radiation therapy for heterotopic ossification. Semin ­Radiat Oncol 1999;9:163-170. 71. Knelles D, Barthel T, Karrer A. Prevention of heterotopic ossification after total hip replacement: a prospective, randomized study using acetylsalicylic acid, indomethacin, and fractional or single-dose irradiation. J Bone Joint Surg Am 1997;79:596-602. 72. Elmstedt E, Lindhilm TS, Nilsson OS. Effect of ibuprofen on heterotopic ossification after hip replacement. Acta Orthop Scand 1985;56:25-27. 73. Pagnani MJ, Pellicci PM, Salvati EA. Effect of aspirin on heterotopic ossification after total hip arthroplasty in men who have osteoarthritis. J Bone Joint Surg Am 1991;73:924-929. 74. Ritter MA, Gioe TJ. The effect of indomethacin on paraarticular ectopic ossification following total hip arthroplasty. Clin Orthop 1982;167:113-117. 75. Spry NA, Dally MJ, Benjamin B, et al. Heterotopic bone formation affecting the hip joint is preventable in high risk patients by post-operative radiation. Australas Radiol 1995;39:379-383. 76. Moore KD, Goss K, Anglen JO. Indomethacin versus radiation therapy for prophylaxis against heterotopic ossification in acetabular fractures: a randomised, prospective study. J Bone Joint Surg Br 1998;80:259-263. 77. Lewallen DG. Heterotopic ossification following total hip ­arthroplasty. Instr Course Lect 1995;44:287-292.

35  The Bone 78. Healy WL, Lo TC, DeSimone AA, et al. Single-dose irradiation for the prevention of heterotopic ossification after total hip ­arthroplasty. A comparison of doses of five hundred and fifty and seven hundred centigray. J Bone Joint Surg Am 1995;77: 590-595. 79. Heyd R, Strassmann G, Kirchner J, et al. Postoperative ­radiotherapy in the prevention of heterotopic ossification after endoprosthetic hip joint replacement. Strahlenther Onkol 1996;172: 543-552. 80. Haran M, Bhuta T, Lee B. Pharmacological interventions for treating acute heterotopic ossification. Cochrane Database Syst Rev 2004;(4):CD003321. 81. Thomas BJ, Amstutz HC. Results of the administration of ­diphosphonate for the prevention of heterotopic ossification after total hip arthroplasty. J Bone Joint Surg Am 1985;67:400403. 82. Bijvoet OL, Nollen AJ, Slooff TJ, Feith R. Effect of a diphosphonate on para-articular ossification after total hip replacement. Acta Orthop Scand 1974;45:926-934. 83. Finerman GA, Krengel WF Jr, Lowell JD, et al. Role of diphosphonate (EHDP) in the prevention of heterotopic ossification after total hip arthroplasty: a preliminary report. The Hip: Proceedings of the Fifth Open Scientific Meeting of the Hip Society. St Louis: CV Mosby, 1977, pp 222-34. 84. Kjaersgaard-Andersen P, Ritter MA. Prevention of formation of heterotopic bone after total hip arthroplasty. J Bone Joint Surg Am 1991;73:942-947. 85. Cella JP, Salvati EA, Sculco TP. Indomethacin for the ­prevention of heterotopic ossification following total hip arthroplasty: ­ effectiveness, contraindications, and adverse effects. J ­Arthroplasty 1988;3:229-234. 86. Trancik T, Mills W, Vinson N. The effect of indomethacin, ­aspirin, and ibuprofen on bone ingrowth into a porous-coated ­implant. Clin Orthop Relat Res 1989;249:113-121. 87. Maloney WJ, Krushell RJ, Jasty M. Incidence of heterotopic ­ossification after total hip replacement: effect of the type of fixation of the femoral component. J Bone Joint Surg Am 1991;73: 191-193. 88. Healy WL, Lo TC, Covall DJ. Single-dose radiation therapy for prevention of heterotopic ossification after total hip arthroplasty. J Arthroplasty 1990;5:369-375. 89. Pellegrini VD Jr, Konski AA, Gastel JA, et al. Prevention of heterotopic ossification with irradiation after total hip arthroplasty. Radiation therapy with a single dose of eight hundred centigray administered to a limited field. J Bone Joint Surg Am 1992;74:186-200. 90. Hedley AK, Mead LP, Hendren DH. The prevention of heterotopic bone formation following total hip arthroplasty using 600 rad in a single dose. J Arthroplasty 1989;4:319-325. 91. Brunner R, Morscher E, Hunig R. Para-articular ossification in total hip replacement: an indication for irradiation therapy. Arch Orthop Trauma Surg 1987;106:102-107. 92. MacLennan I, Keys HM, Evarts CM. Usefulness of postoperative hip irradiation in the prevention of heterotopic bone formation in a high risk group of patients. Int J Radiat Oncol Biol Phys 1984;10:49-53. 93. Kennedy WF, Gruen TA, Chessin H. Radiation therapy to prevent heterotopic ossification after cementless total hip arthroplasty. Clin Orthop 1991;262:185-191. 94. Seegenschmiedt MH, Keilholz L, Martus P, et al. Prevention of heterotopic ossification about the hip: final results of two randomized trials in 410 patients using either preoperative or postoperative radiation therapy. Int J Radiat Oncol Biol Phys 1997;39:161-171. 95. Evarts CM, Ayers DC, Puzas JE. Prevention of heterotopic bone formation in high-risk patients by postoperative irradiation, in the hip. Presented at the Fourteenth Open Scientific Meeting of the Hip Society, St. Louis, MO, 1987. 96. Childs HA III, Cole T, Falkenberg E, et al. A prospective evaluation of the timing of postoperative radiotherapy for preventing heterotopic ossification following traumatic acetabular fractures. Int J Radiat Oncol Biol Phys 2000;47:1347-1352.

933

97. Arnold M, Stas P, Kummermehr J, et al. Radiation-induced ­impairment of bone healing in the rat femur: effects of ­radiation dose, sequence and interval between surgery and irradiation. ­Radiother Oncol 1998;48:259-265. 98. Pellegrini VD Jr, Gregoritch SJ. Preoperative irradiation for prevention of heterotopic ossification following total hip arthroplasty. J Bone Joint Surg Am 1996;78:870-881. 99. Craven PL, Urist MR. Osteogenesis by radioisotope labelled cell populations in implants of bone matrix under the influence of ionizing radiation. Clin Orthop Relat Res 1971;76:231-233. 100. van Leeuwen WM, Deckers P, de Lange WJ. Preoperative ­irradiation for prophylaxis of ectopic ossification after hip ­arthroplasty. a randomized study in 62 hips. Acta Orthop Scand 1998;69:116-118. 101. Blount LH, Thomas BJ, Tran L. Postoperative irradiation for the prevention of heterotopic bone: analysis of different dose schedules and shielding considerations. Int J Radiat Oncol Biol Phys 1990;19:809-811. 102. Huvos AG, Woodard HQ, Cahan WG, et al. Postradiation osteogenic sarcoma of bone and soft tissues. A clinicopathologic study of 66 patients. Cancer 1985;55:1244-1255. 103. Robinson E, Neugut AI, Wylie P. Clinical aspects of post­irradiation sarcomas. J Natl Cancer Inst 1988;80:233-240. 104. Seeler R. Book review: Pediatric Oncology: A Treatise for the Clinician. JAMA 1983;251:1619-1620. 105. Wong FL, Boice JD Jr, Abramson DH, et al. Cancer incidence after retinoblastoma. Radiation dose and sarcoma risk. JAMA 1997;278:1262-1267. 106. Tucker MA, D’Angio GJ, Boice JD, et al. Bone sarcomas linked to radiotherapy and chemotherapy in children. N Engl J Med 1987;317:588-593. 107. Doll R, Wakeford R. Risk of childhood cancer from fetal irradiation. Br J Radiol 1997;70:130-139. 108. Carnes BA, Groer PG, Kotek TJ. Radium dial workers: issues concerning dose response and modeling. Radiat Res 1997;147:707714. 109. Gossner W. Pathology of radium-induced bone tumors: new aspects of histopathology and histogenesis. Radiat Res 1999;152:S12-S15. 110. Kuttesch JF Jr, Wexler LH, Marcus RB, et al. Second malignancies after Ewing’s sarcoma: radiation dose-dependency of secondary sarcomas. J Clin Oncol 1996;14:2818-2825. 111. Le Vu B, de Vathaire F, Shamsaldin A, et al. Radiation dose, chemotherapy and risk of osteosarcoma after solid tumours during childhood. Int J Cancer 1998;77:370-377. 112. Nekolla EA, Kreisheimer M, Kellerer AM, et al. Induction of ­malignant bone tumors in radium-224 patients: risk estimates based on the improved dosimetry. Radiat Res 2000;153:93103. 113. Fry SA. Studies of U.S. radium dial workers: an epidemiological classic. Radiat Res 1998;150:S21-S29. 114. Greenberger JS, Anderson J, Berry LA, et al. Effects of ­irradiation of CBA/CA mice on hematopoietic stem cells and stromal cells in long-term bone marrow cultures. Leukemia 1996;10: 514-527. 115. Niethammer D, Mayer E. Long-term survivors: an overview on late effects, sequelae and second neoplasias. Bone Marrow Transplant 1998;21:S61-S63. 116. Broerse JJ, Bartstra RW, van Bekkum DW, et al. The carcinogenic risk of high dose total body irradiation in non-human primates. Radiother Oncol 2000;54:247-253. 117. Kirova YM, Rafi H, Voisin MC, et al. Radiation-induced bone sarcoma following total body irradiation: role of additional ­radiation on localized areas. Bone Marrow Transplant 2000;25: 1011-1013. 118. Kleinerman RA, Littlefield LG, Tarone RE, et al. Chromosome aberrations in lymphocytes from women irradiated for benign and malignant gynecological disease. Radiat Res 1994;139: 40-46. 119. Kleinerman RA, Boice JD Jr, Storm HH, et al. Second primary cancer after treatment for cervical cancer. An international cancer registries study. Cancer 1995;76:442-452.

934

PART 12  Musculoskeletal System

120. Curtis RE, Boice JD Jr, Stovall M, et al. Relationship of leukemia risk to radiation dose following cancer of the uterine corpus. J Natl Cancer Inst 1994;86:1315-1324. 121. Gershkevitsh E, Rosenberg I, Dearnaley DP, et al. Bone ­marrow doses and leukaemia risk in radiotherapy of prostate cancer. ­Radiother Oncol 1999;53:189-197. 122. de Vathaire F, Schlumberger M, Delisle MJ, et al. Leukaemias and cancers following iodine-131 administration for thyroid ­cancer. Br J Cancer 1997;75:734-739. 123. Kossman SE, Weiss MA. Acute myelogenous leukemia after ­exposure to strontium-89 for the treatment of adenocarcinoma of the prostate. Cancer 2000;88:620-624. 124. Priolo F, Cerase A. The current role of radiography in the ­assessment of skeletal tumors and tumor-like lesions. Eur J ­Radiol 1998;27(Suppl 1):S77-S85. 125. Pomeranz SJ, Pretorius HT, Ramsingh PS. Bone scintigraphy and multimodality imaging in bone neoplasia: strategies for ­imaging in the new health care climate. Semin Nucl Med 1994;24: 188-207. 126. Panicek DM, Gatsonis C, Rosenthal DI, et al. CT and MR ­imaging in the local staging of primary malignant ­musculoskeletal ­neoplasms: report of the Radiology Diagnostic Oncology Group. Radiology 1997;202:237-246. 127. Balzarini L, Sicilia A, Ceglia E, et al. Magnetic resonance in primary bone tumors: a review of 10 years of activities. Radiol Med (Torino) 1996;91:344-347. 128. Thomann KH, Dul MW. Bone scintigraphy: a review of the procedure and its applications. Optom Vis Sci 1994;71:502-507. 129. Donohoe KJ. Selected topics in orthopedic nuclear medicine. Orthop Clin North Am 1998;29:85-101. 130. Lin WY, Kao CH, Hsu CY, et al. The role of Tc-99m MDP and Ga-67 imaging in the clinical evaluation of malignant fibrous histiocytoma. Clin Nucl Med 1994;19:996-1000. 131. Kato H, Miyazaki T, Nakajima M, et al. Comparison between whole-body positron emission tomography and bone scintigraphy in evaluating bony metastases of esophageal carcinomas. Anticancer Res 2005;25:4439-4444. 132. Fujimoto R, Higashi T, Nakamoto Y, et al. Diagnostic accuracy of bone metastases detection in cancer patients: comparison ­between bone scintigraphy and whole-body FDG-PET. Ann Nucl Med 2006;20:399-408. 133. Schirrmeister H, Guhlmann A, Elsner K, et al. ­ Sensitivity in ­detecting osseous lesions depends on anatomic ­ localization: ­planar bone scintigraphy versus 18F PET. J Nucl Med 1999;40: 1623-1629. 134. Schirrmeister H, Glatting G, Hetzel J, et al. Prospective evaluation of the clinical value of planar bone scans, SPECT, and (18)F-labeled NaF PET in newly diagnosed lung cancer. J Nucl Med 2001;42:1800-1804. 135. Even-Sapir E, Metser U, Mishani E, et al. The detection of bone metastases in patients with high-risk prostate cancer: 99mTcMDP planar bone scintigraphy, single- and multi-field-of-view SPECT, 18F-fluoride PET, and 18F-fluoride PET/CT. J Nucl Med 2006;47:287-297. 136. Hernandez JA, Land KJ, McKenna RW. Leukemias, ­ myeloma, and other lymphoreticular neoplasms. Cancer 1995;75: 381-394. 137. Unni KK. Dahlin’s Bone Tumors: General Aspects and Data on 11,087 Cases. 5th ed. New York: Lippincott-Raven, 1996, p 463. 138. Holland J, Trenkner DA, Wasserman TH, et al. Plasmacytoma. Treatment results and conversion to myeloma. Cancer 1992;69:1513-1517. 139. Bolek TW, Marcus RB, Mendenhall NP. Solitary plasmacytoma of bone and soft tissue. Int J Radiat Oncol Biol Phys 1996;36: 329-333. 140. Frassica DA, Frassica FJ, Schray MF, et al. Solitary plasmacytoma of bone: Mayo Clinic experience. Int J Radiat Oncol Biol Phys 1989;16:43-48. 141. Lewanski CR, Bates T, Bowen J, et al. Solitary bone plasmacytoma: management of isolated local relapse following radiotherapy. Clin Oncol 1999;11:348-351.

142. Mill WB, Griffith R. The role of radiation therapy in the management of plasma cell tumors. Cancer 1980;45:647-652. 143. Bullough PG. Cartilage-forming tumors and tumor-like conditions. In Bullough and Vigorita’s Orthopaedic Pathology. 3rd ed. London: Mosby-Wolfe, 1997, pp 357-365. 144. Chang MY, Shih LY, Dunn P, et al. Solitary plasmacytoma of bone. J Formos Med Assoc 1994;93:397-402. 145. Dimopoulos MA, Goldstein J, Fuller L, et al. Curability of solitary bone plasmacytoma. J Clin Oncol 1992;10:587-590. 146. Greenberg P, Parker RG, Fu YS, et al. The treatment of solitary plasmacytoma of bone and extramedullary plasmacytoma. Am J Clin Oncol 1987;10:199-204. 147. Liebross RH, Ha CS, Cox JD, et al. Solitary bone plasmacytoma: outcome and prognostic factors following radiotherapy. Int J Radiat Oncol Biol Phys 1998;41:1063-1067. 148. Aviles A, Huerta-Guzman J, Delgado S, et al. Improved outcome in solitary bone plasmacytomata with combined therapy. Hematol Oncol 1996;14:111-117. 149. Mayr NA, Wen BC, Hussey DH, et al. The role of radiation therapy in the treatment of solitary plasmacytomas. Radiother Oncol 1990;17:293-303. 150. Lichtenstein L, Jaffe HL. Chondrosarcomas of bone. Am J Pathol 1943;19:553-589. 151. Marcove RC, Mike V, Hutter RV, et al. Chondrosarcoma of the pelvis and upper end of the femur. An analysis of factors influencing survival time in one hundred and thirteen cases. J Bone Joint Surg 1972;54:561-572. 152. Aprin H, Riseborough EJ, Hall JE. Chondrosarcoma in children and adolescents. Clin Orthop 1982;166:226-232. 153. Young CL, Sim FH, Unni KK, et al. Case report 559. Chondrosarcoma of the proximal humeral epiphysis. Skeletal Radiol 1989;18:403-405. 154. Young CL, Sim FH, Unni KK, et al. Chondrosarcoma of bone in children. Cancer 1990;66:1641-1648. 155. Mirra JM, Gold R, Downs J, et al. A new histological approach to the differentiation of enchondroma from chondrosarcoma of bones: a clinicopathologic analysis of 51 cases. Clin Orthop 1985;201:214-237. 156. Bjornsson J, McLeod RA, Unni KK, et al. Primary chondrosarcoma of long bones and limb girdles. Cancer 1998;83: 2105-2119. 157. Pritchard DJ, Lunke RJ, Taylor WF, et al. Chondrosarcoma: a clinicopathologic and statistical analysis. Cancer 1980;45: 149-157. 158. Dahlin DC, Beabout JW. Dedifferentiation of low-grade chondrosarcomas. Cancer 1971;28:461-466. 159. Gitelis S, Bertoni F, Picci P, et al. Chondrosarcoma of bone. The experience at the Istituto Ortopedico Rizzoli. J Bone Joint Surg Am 1981;63:1248-1257. 160. Lee FY, Mankin HJ, Fondren G, et al. Chondrosarcoma of bone: an assessment of outcome. J Bone Joint Surg Am 1999;81: 326-338. 161. Simon MA, Springfield D (eds): Surgery for Bone and Soft­Tissue Sarcomas, vol 1. New York: Lippincott Williams & Wilkins, 1998, p 756. 162. Bauer HC, Brosjo O, Kreicbergs A, et al. Low risk of recurrence of enchondroma and low-grade chondrosarcoma in extremities. 80 patients followed for 2-25 years. Acta Orthop Scand 1995;66:283-288. 163. Marcove RC, Stovell PB, Huvos AG, et al. The use of ­cryosurgery in the treatment of low and medium grade chondrosarcoma. A preliminary report. Clin Orthop 1977;122:147-156. 164. Schreuder HW, Pruszczynski M, Veth RP, et al. Treatment of ­benign and low-grade malignant intramedullary chondroid tumours with curettage and cryosurgery. Eur J Surg Oncol 1998;24:120-126. 165. Etchebehere M, de Camargo OP, Croci AT, et al. Relationship between surgical procedure and outcome for patients with grade I chondrosarcomas. Clinics (Sao Paulo) 2005;60:121-126. 166. Moussavi-Harami F, Mollano A, Martin JA, et al. Intrinsic radiation resistance in human chondrosarcoma cells. Biochem Biophys Res Commun 2006;346:379-385.

35  The Bone 167. Rhomberg W, Eiter H, Bohler F, et al. Combined radiotherapy and razoxane in the treatment of chondrosarcomas and chordomas. Anticancer Res 2006;26:2407-2411. 168. Debus J, Haberer T, Schulz-Ertner D, et al. Carbon ion irradiation of skull base tumors at GSI. First clinical results and future perspectives. Strahlenther Onkol 2000;176:211-216. 169. Gripp S, Pape H, Schmitt G. Chondrosarcoma of the larynx: the role of radiotherapy revisited-a case report and review of the literature. Cancer 1998;82:108-115. 170. Hug EB, Loredo LN, Slater JD, et al. Proton radiation therapy for chordomas and chondrosarcomas of the skull base. J Neurosurg 1999;91:432-439. 171. Muthukumar N, Kondziolka D, Lunsford LD, et al. Stereotactic radiosurgery for chordoma and chondrosarcoma: further experiences. Int J Radiat Oncol Biol Phys 1998;41:387-392. 172. Rosenberg AE, Nielsen GP, Keel SB, et al. Chondrosarcoma of the base of the skull: a clinicopathologic study of 200 cases with emphasis on its distinction from chordoma. Am J Surg Pathol 1999;23:1370-1378. 173. Munzenrider JE, Liebsch NJ. Proton therapy for tumors of the skull base. Strahlenther Onkol 1999;175:57-63. 174. Gwak HS, Yoo HJ, Youn SM, et al. Hypofractionated stereotactic radiation therapy for skull base and upper cervical chordoma and chondrosarcoma: preliminary results. Stereotact Funct Neurosurg 2006;83:233-243. 175. Krishnan S, Foote RL, Brown PD, et al. Radiosurgery for cranial base chordomas and chondrosarcomas. Neurosurgery 2005;56:777-784. 176. Oberling C. Les reticulosarcomes at les reticuloendotheliosarcomes de la moelle osseuse sarcomes d’Ewing. Bull Assoc Fr Etud Cancer (Paris) 1928;17:259-296. 177. Parker JW, Jackson H. Primary reticulum cell sarcoma of bone. Surg Gynecol Obstet 1939;68:45-53. 178. Freeman C, Berg JW, Cutler SJ. Occurrence and prognosis of extranodal lymphomas. Cancer 1972;29:252-260. 179. Rudders RA, Ross ME, DeLellis RA. Primary extranodal lymphoma: response to treatment and factors influencing prognosis. Cancer 1978;42:406-416. 180. Boston HC, Dahlin DC, Ivins JC, et al. Malignant lymphoma (socalled reticulum cell sarcoma) of bone. Cancer 1974;34:11311137. 181. Pettit CK, Zukerberg LR, Gray MH, et al. Primary lymphoma of bone. Am J Surg Pathol 1990;14:329-334. 182. Ferreri AJ, Reni M, Ceresoli GL, et al. Therapeutic management with adriamycin-containing chemotherapy and radiotherapy of monostotic and polyostotic primary non-Hodgkin’s lymphoma of bone in adults. Cancer Invest 1998;16:554-561. 183. Shoji H, Miller TR. Primary reticulum cell sarcoma of bone. Significance of clinical features upon the prognosis. Cancer 1971;28:1234-1244. 184. Baar J, Burkes RL, Bell R, et al. Primary non-Hodgkin’s lymphoma of bone. Cancer 1994;73:1194-1199. 185. Chin HW, McGuire MH, Block M, et al. Primary lymphoma of bone. Nebr Med J 1990;75:303-306. 186. Dubey P, Ha CS, Besa PC, et al. Localized primary malignant lymphoma of bone. Int J Radiat Oncol Biol Phys 1997;37: 1087-1093. 187. Nishiyama N, Nakatani S, Inoue K, et al. Primary lymphoma of bone originating in a rib. Jpn J Thorac Cardiovasc Surg 2000;48:180-183. 188. Heyning FH, Hogendoorn PC, Kramer MH, et al. Primary nonHodgkin’s lymphoma of bone: a clinicopathological investigation of 60 cases. Leukemia 1999;13:2094-2098. 189. Phillips WC, Kattapuram SV, Doseretz DE, et al. Primary lymphoma of bone: relationship of radiographic appearance and prognosis. Radiology 1982;144:285-290. 190. Jones D, Kraus MD, Dorfman DM. Lymphoma presenting as a solitary bone lesion. Am J Clin Pathol 1999;111:171-178. 191. Suryanarayan K, Shuster JJ, Donaldson SS, et al. Treatment of localized primary non-Hodgkin’s lymphoma of bone in ­children: a Pediatric Oncology Group study. J Clin Oncol 1999;17: 456-459.

935

192. Wang CC, Fleischli DJ. Primary reticulum cell sarcoma of bone. With emphasis on radiation therapy. Cancer 1968;22:994-998. 193. Wang CC. Treatment of primary reticulum-cell sarcoma of bone by irradiation. N Engl J Med 1968;278:1331-1332. 194. Bacci G, Picci P, Bertoni F, et al. Primary non-Hodgkin’s lymphoma of bone: results in 15 patients treated by radiotherapy combined with systemic chemotherapy. Cancer Treat Rep 1982;66:1859-1862. 195. Link MP, Shuster JJ, Donaldson SS, et al. Treatment of children and young adults with early-stage non-Hodgkin’s lymphoma. N Engl J Med 1997;337:1259-1266. 196. Loeffler JS, Tarbell NJ, Kozakewich H, et al. Primary lymphoma of bone in children: analysis of treatment results with Adriamycin, prednisone, Oncovin (APO), and local radiation therapy. J Clin Oncol 1986;4:496-501. 197. Meyer JE, Lepke RA, Lindfors KK, et al. Chordomas: their CT appearance in the cervical, thoracic and lumbar spine. Radiology 1984;153:693-696. 198. Hudson TM, Galceran M. Radiology of sacrococcygeal chordoma. Difficulties in detecting soft tissue extension. Clin Orthop 1983;175:237-242. 199. Rosenthal DI, Scott JA, Mankin HJ, et al. Sacrococcygeal chordoma: magnetic resonance imaging and computed tomography. AJR Am J Roentgenol 1985;145:143-147. 200. Bjornsson J, Wold LE, Ebersold MJ, et al. Chordoma of the ­mobile spine. A clinicopathologic analysis of 40 patients. Cancer 1993;71:735-740. 201. Cheng EY, Ozerdemoglu RA, Transfeldt EE, et al. Lumbosacral chordoma. Prognostic factors and treatment. Spine 1999;24:1639-1645. 202. Mitchell A, Scheithauer BW, Unni KK, et al. Chordoma and chondroid neoplasms of the spheno-occiput. An immunohistochemical study of 41 cases with prognostic and nosologic implications. Cancer 1993;72:2943-2949. 203. Kaiser TE, Pritchard DJ, Unni KK. Clinicopathologic study of sacrococcygeal chordoma. Cancer 1984;53:2574-2578. 204. al-Mefty O, Borba LA. Skull base chordomas: a management challenge. J Neurosurg 1997;86:182-189. 205. Catton C, O’Sullivan B, Bell R, et al. Chordoma: long-term follow-up after radical photon irradiation. Radiother Oncol 1996;41:67-72. 206. Slater JM, Slater JD, Archambeau JO. Proton therapy for cranial base tumors. J Craniofac Surg 1995;6:24-26. 207. Borba LA, Al-Mefty O, Mrak RE, et al. Cranial chordomas in children and adolescents. J Neurosurg 1996;84:584-591. 208. Logroscino CA, Astolfi S, Sacchettoni G. Chordoma: longterm evaluation of 15 cases treated surgically. Chir Organi Mov 1998;83:87-103. 209. Schoenthaler R, Castro JR, Petti PL, et al. Charged particle ­irradiation of sacral chordomas. Int J Radiat Oncol Biol Phys 1993;26:291-298. 210. Habrand JL, Mammar H, Ferrand R, et al. Proton beam therapy (PT) in the management of CNS tumors in childhood. Strahlenther Onkol 1999;175:91-94. 211. Romero J, Cardenes H, la Torre A, et al. Chordoma: results of radiation therapy in eighteen patients. Radiother Oncol 1993;29:27-32. 212. Klekamp J, Samii M. Spinal chordomas-results of treatment over a 17-year period. Acta Neurochir (Wien) 1996;138:514-519. 213. Nishida J, Sim FH, Wenger DE, et al. Malignant fibrous histiocytoma of bone. A clinicopathologic study of 81 patients. Cancer 1997;79:482-493. 214. Dahlin DC, Unni KK, Matsuno T. Malignant (fibrous) histiocytoma of bone-fact or fancy? Cancer 1977;39:1508-1516. 215. Capanna R, Bertoni F, Bacchini P. Fibrous histiocytoma of bone. Cancer 1984;76:177-187. 216. Huvos AG, Heilweil M, Bretsky SS. The pathology of malignant fibrous histiocytoma of bone: a study of 130 patients. Am J Surg Pathol 1985;9:853-871. 217. Dahlin DC, Unni KK. Bone Tumors: General Aspects and Data on 8,542 Cases, 4th ed. Springfield, MA: Charles C Thomas, 1986, pp 357-365.

936

PART 12  Musculoskeletal System

218. Weiss SW, Enzinger FM. Malignant fibrous histiocytoma, an analysis of 200 cases. Cancer 1978;41:2250-2266. 219. Dunham WK, Wilborn WH. Malignant fibrous histiocytoma of bone. Report of two cases and review of the literature. J Bone Joint Surg Am 1979;61:939-942. 220. Picci P, Bacci G, Ferrari S, et al. Neoadjuvant chemotherapy in malignant fibrous histiocytoma of bone and in osteosarcoma ­located in the extremities: analogies and differences between the two tumors. Ann Oncol 1997;8:1107-1115. 221. Bielack SS, Schroeders A, Fuchs N, et al. Malignant fibrous histiocytoma of bone: a retrospective EMSOS study of 125 cases. European Musculo-Skeletal Oncology Society. Acta Orthop Scand 1999;70:353-360. 222. Bacci G, Ferrari S, Bertoni F, et al. Neoadjuvant chemotherapy for osseous malignant fibrous histiocytoma of the extremity: results in 18 cases and comparison with 112 contemporary osteosarcoma patients treated with the same chemotherapy regimen. J Chemother 1997;9:293-299. 223. Weiss SW, Enzinger FM. Epithelioid hemangioendothelioma: a vascular tumor often mistaken for a carcinoma. Cancer 1982;50:970-981. 224. Wold LE, Unni KK, Beabout JW, et al. Hemangioendothelial sarcoma of bone. Am J Surg Pathol 1982;6:59-70. 225. Campanacci M, Boriani S, Giunti A. Hemangioendothelioma of bone. Cancer 1980;46:804-814. 226. Tsuneyoshi M, Dorfman HD, Bauer TW. Epithelioid hemangioendothelioma of bone. A clinicopathologic, ultrastructural, and immunohistochemical study. Am J Surg Pathol 1986;10:754-764. 227. Ignacio EA, Palmer KM, Mathur SC, et al. Epithelioid ­hemangioendothelioma of the lower extremity. Radiographics 1999;19:531-537. 228. Kleer CG, Unni KK, McLeod RA. Epithelioid hemangioendothelioma of bone. Am J Surg Pathol 1996;20:1301-1311. 229. Adler B, Naheedy J, Yeager N, et al. Multifocal epithelioid hemangioendothelioma in a 16-year-old boy. Pediatr Radiol 2005;35:1014-1018. 230. Tillman RM, Choong PF, Beabout JW, et al. Epithelioid hemangioendothelioma of bone. Orthopedics 1997;20:177-180. 231. Gosheger G, Hardes J, Ozaki T, et al. The multicentric epithelioid hemangioendothelioma of bone: a case example and review of the literature. J Cancer Res Clin Oncol 2002;128:11-18. 232. Rosenthal DI, Treat ME, Mankin HJ, et al. Treatment of epithelioid hemangioendothelioma of bone using a novel combined approach. Skeletal Radiol 2001;30:219-222. 233. Larsson SE, Lorentzon R, Boquist L. Malignant hemangioendothelioma of bone. J Bone Joint Surg Am 1975;57:84-89. 234. Glenn JN, Reckling FW, Mantz FA. Malignant hemangioendothelioma in a lumbar vertebra: a rare tumor in an unusual location. J Bone Joint Surg Am 1974;56:1279-1282. 235. Garcia-Moral CA. Malignant hemangioendothelioma of bone. Review of world literature and report of two cases. Clin Orthop Relat Res 1972;82:70-79. 236. Welles L, Dorfman H, Valentine ES, et al. Low grade malignant hemangioendothelioma of bone: a disease potentially curable with radiotherapy. Med Pediatr Oncol 1994;23:144-148. 237. Faria SL, Schlupp WR, Chiminazzo H. Radiotherapy in the treatment of vertebral hemangiomas. Int J Radiat Oncol Biol Phys 1985;11:387-390. 238. Frassica FJ, Sanjay BK, Unni KK, et al. Benign giant cell tumor. Orthopedics 1993;16:1179-1183. 239. Schajowicz F, Granato DB, McDonald DJ, et al. Clinical and radiographic features of atypical giant cell tumors of bone. Br J Radiol 1991;64:877-889. 240. Kransdorf MJ, Sweet DE, Buetow PC, et al. Giant cell tumor in skeletally immature patients. Radiology 1992;184:233-237. 241. Picci P, Manfrini M, Zucchi V, et al. Giant cell tumor of bone in skeletally immature patients. J Bone Joint Surg 1983;65:486-490. 242. Bogumill GB, Schultz MA, Johnson LC. Giant cell tumor: a metaphyseal lesion. J Bone Joint Surg 1972;54:1558. 243. Campanacci M, Giunti A, Olmi R. Metaphyseal and diaphyseal localization of giant cell tumors. Chir Organi Mov 1975;62: 29-34.

244. Fain JS, Unni KK, Beabout JW, et al. Nonepiphyseal giant cell tumor of the long bones. Clinical, radiologic, and pathologic study. Cancer 1993;71:3514-3519. 245. Peison B, Feigenbaum J. Metaphyseal giant cell tumor in a girl of 14. Radiology 1976;118:145-146. 246. Sherman M, Fabricus R. Giant cell tumors in metaphysis of a child. J Bone Joint Surg Am 1961;43:1225-1229. 247. Wilkerson JA, Cracchiolo A. Giant cell tumor of the tibial diaphysis. J Bone Joint Surg Am 1969;51:1205-1209. 248. Hudson TM, Schiebler M, Springfield DS, et al. Radiology of giant cell tumors of bone: computed tomography, arthro-tomography, and scintigraphy. Skeletal Radiol 1984;11:85-95. 249. McDonald DJ, Sim FH, McLeod RA, et al. Giant cell tumor of bone. J Bone Joint Surg Am 1986;68:235-242. 250. Ennecking WF. Musculoskeletal Tumor Surgery, vol 1. New York: Churchill Livingstone, 1983. 251. Bertoni F, Present D, Sudanese A, et al. Giant cell tumor of bone with pulmonary metastases. Six case reports and a review of the literature. Clin Orthop Relat Res 1988;237:275-285. 252. Maloney WJ, Vaughan LM, Jones HH, et al. Benign metastasizing giant cell tumor of bone. Report of three cases and review of the literature. Clin Orthop 1989;243:208-215. 253. Szyfelbein WM, Schiller AL. Cytologic diagnosis of giant cell tumor of bone metastatic to lung. A case report. Acta Cytol 1979;23:460-464. 254. Connell D, Munk PL, Lee MJ, et al. Giant cell tumor of bone with selective metastases to mediastinal lymph nodes. Skeletal Radiol 1998;27:341-345. 255. Tubbs WS, Brown LR, Beabout JW, et al. Benign giant cell tumor of bone with pulmonary metastases: clinical findings and radiologic appearance of metastases in 13 cases. AJR Am J Roentgenol 1992;158:331-334. 256. Mirra JM, Ulich T, Magidson J, et al. A case of probable benign pulmonary “metastases” or implants arising from a giant cell ­tumor of bone. Clin Orthop Relat Res 1982;46:245-254. 257. Nojima T, Takeda N, Matsuno T, et al. Case report 869. ­Benign metastasizing giant cell tumor of bone. Skeletal Radiol 1994;23:583-585. 258. Rock MG, Pritchard DJ, Unni KK. Metastases from histologically benign giant cell tumor of bone. J Bone Joint Surg Am 1984;66:269-274. 259. Cheng JC, Johnston JO. Giant cell tumor of bone. Prognosis and treatment of pulmonary metastases. Clin Orthop Relat Res 1997;338:205-214. 260. Kay RM, Eckardt JJ, Seeger LL, et al. Pulmonary metastasis of ­benign giant cell tumor of bone. Six histologically confirmed cases, including one of spontaneous regression. Clin Orthop Relat Res 1994;302:219-230. 261. Gitelis S, Wang JW, Quast M, et al. Recurrence of a giant cell tumor with malignant transformation to a fibrosarcoma twentyfive years after primary treatment. A case report. J Bone Joint Surg Am 1989;71:757-761. 262. Mori Y, Tsuchiya H, Karita M, et al. Malignant transformation of a giant cell tumor 25 years after initial treatment. Clin Orthop Relat Res 2000;381:185-191. 263. Nascimento AG, Huvos AG, Marcove RC. Primary malignant giant cell tumor of bone: a study of eight cases and review of the literature. Cancer 1979;44:1393-1402. 264. Rock MG, Sim FH, Unni KK, et al. Secondary malignant giant cell tumor of bone. Clinicopathological assessment of nineteen patients. J Bone Joint Surg Am 1986;68:1073-1079. 265. Dahlin DC, Cupps RE, Johnson EW. Giant cell tumor: a study of 195 cases. Cancer 1970;25:1061-1070. 266. Goldenberg RR, Campbell CJ, Bonfiglio M. Giant cell tumor of bone. An analysis of two hundred and eighteen cases. J Bone Joint Surg Am 1970;52:619-664. 267. Kreicbergs A, Lonnqvist PA, Nilsson B. Curettage of benign lesions of bone. Factors related to recurrence. Int Orthop 1985;8:287-293. 268. Blackley HR, Wunder JS, Davis AM, et al. Treatment of giant cell tumors of long bones with curettage and bone-grafting. J Bone Joint Surg Am 1999;81:811-820.

35  The Bone 269. Leeson MC, Lippitt SB. Thermal aspects of the use of polymethylmethacrylate in large metaphyseal defects in bone. A clinical review and laboratory study. Clin Orthop Relat Res 1993;295:239245. 270. Gitelis S, Mallin BA, Piasecki P, et al. Intralesional excision compared with en bloc resection for giant cell tumors of bone. J Bone Joint Surg Am 1993;75:1648-1655. 271. Malawer MM, Bickels J, Meller I, et al. Cryosurgery in the treatment of giant cell tumor. A long-term followup study. Clin ­Orthop 1999;359:176-188. 272. Marcove RC, Sheth DS, Brien EW, et al. Conservative surgery for giant cell tumors of the sacrum. The role of cryosurgery as a supplement to curettage and partial excision. Cancer 1994;74:1253-1260. 273. O’Donnell RJ, Springfield DS, Motwani HK, et al. Recurrence of giant cell tumors of the long bones after curettage and packing with cement. J Bone Joint Surg Am 1994;76:1827-1833. 274. Persson BM, Wouters HW. Curettage and acrylic cementation in surgery of giant cell tumors of bone. Clin Orthop Relat Res 1976;120:125-133. 275. Pogrel MA. The management of lesions of the jaws with liquid nitrogen cryotherapy. J Calif Dent Assoc 1995;23:54-57. 276. Lewis VO, Wei A, Mendoza T, et al. Argon beam coagulation as an adjuvant for local control of giant cell tumor. Clin Orthop Relat Res 2007;454:192-197. 277. Bennett CJ Jr, Marcus RB Jr, Million RR, et al. Radiation therapy for giant cell tumor of bone. Int J Radiat Oncol Biol Phys 1993;26:299-304. 278. Chakravarti A, Spiro IJ, Hug EB, et al. Megavoltage radiation therapy for axial and inoperable giant cell tumor of bone. J Bone Joint Surg Am 1999;81:1566-1573. 279. Malone S, O’Sullivan B, Catton C, et al. Long-term follow-up of efficacy and safety of megavoltage radiotherapy in high-risk giant cell tumors of bone. Int J Radiat Oncol Biol Phys 1995;33: 689-694. 280. Seider MJ, Rich TA, Ayala AG, et al. Giant cell tumors of bone: treatment with radiation therapy. Radiology 1986;161:537-540. 281. Suit H, Spiro I. Radiation treatment of benign mesenchymal disease. Semin Radiat Oncol 1999;9:171-178. 282. Schwartz LH, Okunieff PG, Rosenberg A. Radiation therapy in the treatment of difficult giant cell tumors. Int J Radiat Oncol Biol Phys 1989;17:1085-1088. 283. Nair MK, Jyothirmayi R. Radiation therapy in the treatment of giant cell tumor of bone. Int J Radiat Oncol Biol Phys 1999;43:1065-1069. 284. Caudell JJ, Ballo MT, Zagars GK, et al. Radiotherapy in the management of giant cell tumor of bone. Int J Radiat Oncol Biol Phys 2003;57:158-165. 285. Stull MA, Kransdorf MJ, Devaney KO. Langerhans cell histiocytosis of bone. Radiographics 1992;12:801-823. 286. Schajowicz F, Slullitel J. Eosinophilic granuloma of bone and its relationship to Hand-Schüller-Christian and Letterer-Siwe ­syndromes. J Bone Joint Surg Br 1973;55:545-565. 287. Siddiqui AR, Tashjian JH, Lazarus K. Nuclear medicine studies and evaluation of skeletal lesions in children with histiocytosis X. Radiology 1981;140:287-289. 288. Sessa S, Sommelet D, Lascombes P, et al. Treatment of Langerhans-cell histiocytosis in children. Experience at the Children’s Hospital of Nancy. J Bone Joint Surg Am 1994;76:1513-1525. 289. Hartman KS. A review of 114 cases with oral involvement. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1980;49:38-54. 290. Egeler RM, Thompson RCJ, Voute PA, et al. Intralesional infiltration of corticosteroids in localized Langerhans cell histiocytosis. J Pediatr Orthop 1992;12:811-814. 291. Cline MJ. The histiocytic malignancies. In Haskell CM (ed): ­Cancer Treatment, 2nd ed. Philadelphia: WB Saunders, 1987, pp 843-850. 292. Raney RB Jr. Chemotherapy for children with aggressive ­fibromatosis and Langerhans’ cell histiocytosis. Clin Orthop 1991;262:58-63. 293. Ishii E, Matsuzaki A, Okamura J, et al. Treatment of Langerhans cell histiocytosis in children with etoposide. Am J Clin Oncol 1992;15:515-517.

937

294. Anonsen CK, Donaldson SS. Langerhans’ cell histiocytosis of the head and neck. Laryngoscope 1987;97:537-542. 295. Selch MT, Parker RG. Radiation therapy in the management of Langerhans cell histiocytosis. Med Pediatr Oncol 1990;18: 97-102. 296. Pereslegin IA, Ustinova VF, Podlyaschuk EL. Radiotherapy for eosinophilic granuloma of bone. Int J Radiat Oncol Biol Phys 1981;7:317-321. 297. Greenberger JS, Cassady JR, Jaffe N. Radiation therapy in ­patients with histiocytosis: management of diabetes insipidus and bone lesions. Int J Radiat Oncol Biol Phys 1979;5:17491755. 298. Mercadante S. Malignant bone pain: pathophysiology and treatment. Pain 1997;69:1-18. 299. Coleman RE. Clinical features of metastatic bone disease and risk of skeletal morbidity. Clin Cancer Res 2006;12:6243s6249s. 300. Tong D, Gillick L, Hendrickson FR. The palliation of symptomatic osseous metastases: final results of the RTOG. Cancer 1982;50:893-899. 301. Blitzer PH. Reanalysis of the RTOG study of palliation of symptomatic osseous metastasis. Cancer 1985;55:1468-1472. 302. Frassica DA. General principles of external beam radiation therapy for skeletal metastases. Clin Orthop Relat Res 2003;415(Suppl): S158-S164. 303. Hipp JA, Springfield DS, Hayes WC. Predicting pathologic fracture risk in the management of metastatic bone defects. Clin Orthop Relat Res 1995;312:120-135. 304. Snyder BD, Hauser-Kara DA, Hipp JA, et al. Predicting fracture through benign skeletal lesions with quantitative computed ­tomography. J Bone Joint Surg Am 2006;88:55-70. 305. Hong J, Cabe GD, Tedrow JR, et al. Failure of trabecular bone with simulated lytic defects can be predicted noninvasively by structural analysis. J Orthop Res 2004;22:479-486. 306. Mirels H. Metastatic disease in long bones. A proposed ­scoring system for diagnosing impending pathologic fractures. Clin ­Orthop Relat Res 1989;249:256-264. 307. Hoegler D. Radiotherapy for palliation of symptoms in incurable cancer. Curr Probl Cancer 1997;21:129-183. 308. Menendez LR, Healey JH, Eckardt JJ, et al (eds): Orthopaedic Knowledge Update: Musculoskeletal Tumors. Rosemont, IL: Musculoskeletal Tumor Society/American Academy of Orthopaedic Surgeons, 2002. 309. Uppelschoten JM, Wanders SL, de Jong JM. Single-dose radiotherapy (6 Gy): palliation in painful bone metastases. Radiother Oncol 1995;36:198-202. 310. Janjan NA. Radiation for bone metastases: conventional techniques and the role of systemic radiopharmaceuticals. Cancer 1997;80:1628-1645. 311. Rose CM, Kagan R. The final report of the export panel for the Radiation Oncology Bone Metastases Work Group of the American College of Radiology. Int J Radiat Oncol Biol Phys 1998;40:1117-1124. 312. Mandoliti G, Polico C, Capirci C, et al. Radiation therapy of bone metastases in the elderly: a multicentric survey of the Italian Geriatric Radiation Oncology Group. Rays 1997;22:57-60. 313. Rasmusson B, Vejborg I, Jensen AB, et al. Irradiation of bone metastases in breast cancer patients: a randomized study with 1 year follow-up. Radiother Oncol 1995;34:179-184. 314. Niewald M, Tkocz HJ, Abel U, et al. Rapid course radiation therapy vs. more standard treatment: a randomized trial for bone metastases. Int J Radiat Oncol Biol Phys 1996;36:10851089. 315. Booth M, Summers J, Williams MV. Audit reduces the reluctance to use single fractions for painful bone metastases. Clin Oncol (R Coll Radiol) 1993;5:15-18. 316. Chander SS, Sarin R. Single fraction radiotherapy for bone ­metastases: are all questions answered? Radiother Oncol 1999;52: 191-193. 317. Price P, Hoskin PJ, Easton D. Prospective randomized trial of single and multi-fraction radiotherapy schedules in the treatment of painful bony metastases. Radiother Oncol 1986;6: 247-255.

938

PART 12  Musculoskeletal System

318. Hoskin PJ, Price P, Easton D. A prospective randomized trial of 4 Gy or 8 Gy single doses in the treatment of metastatic bone pain. Radiother Oncol 1992;23:74-78. 319. Stevens G, Firth I. Patterns of fractionation for palliation of bone metastases. Australas Radiol 1995;39:31-35. 320. Roos DE. Continuing reluctance to use single fractions of ­radiotherapy for metastatic bone pain: an Australian and New Zealand practice survey and literature review. Radiother Oncol 2000;56:315-322. 321. Bremer M, Rades D, Blach M, et al. Effectiveness of hypofractionated radiotherapy in painful bone metastases. Two prospective studies with 1 × 4 Gy and 4 × 4 Gy. Strahlenther Onkol 1999;175:382-386. 322. Chow E, Danjoux C, Wong R, et al. Palliation of bone metastases: a survey of patterns of practice among Canadian radiation oncologists. Radiother Oncol 2000;56:305-314. 323. Coia LR, Hanks GE, Martz K. Practice patterns of palliative care for the United States 1984-1985. Int J Radiat Biol Phys 1988;14:1261-1269. 324. Madsen EL. Painful bone metastasis: efficacy of radiotherapy ­assessed by the patients. A randomized trial comparing 4 Gy × 6 versus 10 Gy × 2. Int J Radiat Oncol Biol Phys 1983;9:17751779. 325. Okawa T, Kita M, Goto M. Randomized prospective clinical study of small, large and twice-a-day fraction radiotherapy for painful bone metastases. Radiother Oncol 1988;13:99-104. 326. Gaze MN, Kelly CG, Kerr GR, et al. Pain relief and quality of life following radiotherapy for bone metastases: a randomised trial of two fractionation schedules. Radiother Oncol 1997;45: 109-116. 327. Arcangeli G, Giovinazzo G, Saracino B, et al. Radiation therapy in the management of symptomatic bone metastases: the effect of total dose and histology on pain relief and response duration. Int J Radiat Oncol Biol Phys 1998;42:1119-1126. 328. Nielsen OS, Bentzen SM, Sandberg E, et al. Randomized trial of single dose versus fractionated palliative radiotherapy of bone metastases. Radiother Oncol 1998;47:233-240. 329. Koswig S, Budach V. Remineralization and pain relief in bone metastases after different radiotherapy fractions (10 times 3 Gy vs. 1 times 8 Gy). A prospective study. Strahlenther Onkol 1999;175:500-508. 330. Kaasa S, Brenne E, Lund JA, et al. Prospective randomised multicenter trial on single fraction radiotherapy (8 Gy × 1) versus multiple fractions (3 Gy × 10) in the treatment of painful bone metastases. Radiother Oncol 2006;79:278-284. 331. van der Linden YM, Steenland E, van Houwelingen HC, et al. ­Patients with a favourable prognosis are equally palliated with single and multiple fraction radiotherapy: results on survival in the Dutch Bone Metastasis Study. Radiother Oncol 2006;78: 245-253. 332. Sze WM, Shelley MD, Held I, et al. Palliation of metastatic bone pain: single fraction versus multifraction radiotherapy—a systematic review of randomised trials. Clin Oncol (R Coll Radiol) 2003;15:345-352. 333. Salazar OM, Rubin P, Hendrickson FR. Single dose half body irradiation for palliation of multiple bone metastases from solid tumors. Cancer 1986;58:29-36. 334. Zelefsky MJ, Scher HI, Forman JD. Palliative hemiskeletal irradiation for widespread metastatic prostate cancer: a comparison of single dose and fractionated regimens. Int J Radiat Oncol Biol Phys 1989;17:1281-1285. 335. Scarantino CW, Caplan R, Rotman M, et al. A phase I/II study to evaluate the effect of fractionated hemibody irradiation in the treatment of osseous metastases-RTOG 88-22. Int J Radiat Oncol Biol Phys 1996;36:37-48. 336. Priestman TJ. Clinical studies with ondansetron in the control of radiation-induced emesis. Eur J Cancer 1989;25:S29-S33. 337. Hayashi S, Hoshi H, Iida T, et al. Multi-fractionated wide-field radiation therapy for palliation of multiple symptomatic bone metastases from solid tumors. Radiat Med 1999;17:411-416. 338. Algara M, Valls A, Ruiz V, et al. Half-body irradiation. Palliative efficacy and predictive factors of response in 78 procedures. Med Clin (Barc) 1994;103:85-88.

339. Poulter CA, Cosmatos D, Rubin P. A report of RTOG 8206: a phase III study of whether the addition of single dose hemibody irradiation is more effective than local field irradiation alone in the treatment of symptomatic osseous metastases. Int J Radiat Oncol Biol Phys 1992;23:207-214. 340. Quilty PM, Kirk D, Bolger JJ, et al. A comparison of the palliative effects of strontium-89 and external beam radiotherapy in metastatic prostate cancer. Radiother Oncol 1994;31:33-40. 341. Porter AT, McEwan AJB. Strontium-89 as an adjuvant to ­external beam radiation improves pain relief and delays disease progression in advanced prostate cancer: results of a randomized controlled trial. Semin Oncol 1993;20:38-43. 342. Porter AT, McEwan AJB, Powe JE. Results of a randomized phase-III trial to evaluate the efficacy of strontium-89 adjuvant to local field external beam irradiation in the management of endocrine resistant metastatic prostate cancer. Int J Radiat Oncol Biol Phys 1993;25:805-813. 343. Robinson RG, Preston DF, Spicer JA, et al. Radionuclide therapy of intractable bone pain: emphasis on strontium-89. Semin Nucl Med 1992;22:28-32. 344. Robinson RG, Spicer JA, Blake GM. Strontium-89: treatment results and kinetics in patients with painful metastatic prostate and breast cancer in bone. Radiographics 1989;9:271-281. 345. Lee BC, Shav-Tal Y, Peled A, et al. A hematopoietic organ­specific 49-kD nuclear antigen: predominance in immature normal and tumor granulocytes and detection in hematopoietic precursor cells. Blood 1996;87:2283-2291. 346. Baumrucker S. Palliation of painful bone metastases: ­ strontium-89. Am J Hosp Palliat Care 1998;15:113-115. 347. Brundage MD, Crook JM, Lukka H. Use of strontium-89 in ­ endocrine-refractory prostate cancer metastatic to bone. ­Provincial Genitourinary Cancer Disease Site Group. Cancer Prev Control 1998;2:79-87. 348. Blake GM, Gray JM, Zivanovic MA. Strontium-89 radionuclide therapy: a dosimetric study using impulse response function analysis. Br J Radiol 1987;60:685-692. 349. Blake GM, Zivanovic MA. Sr-89 therapy: strontium kinetics in disseminated carcinoma of the prostate. Eur J Nucl Med 1986;12:447-454. 350. Blake GM, Zivanovic MA, McEwan AJ. Sr-89 radionuclide therapy: dosimetry and hematologic toxicity in two patients with metastasizing prostatic carcinoma. Eur J Nucl Med 1987;13: 41-46. 351. Breen SL, Powe JE, Porter AT. Dose estimation in strontium89 radiotherapy of metastatic prostatic carcinoma. J Nucl Med 1992;33:1316-1323. 352. Pons F, Herranz R, Garcia A, et al. Strontium-89 for palliation of pain from bone metastases in patients with prostate and breast cancer. Eur J Nucl Med 1997;24:1210-1214. 353. Sciuto R, Festa A, Tofani A, et al. Platinum compounds as ­radiosensitizers in strontium-89 metabolic radiotherapy. Clin Ter 1998;149:43-47. 354. Sciuto R, Maini CL, Tofani A, et al. Radiosensitization with lowdose carboplatin enhances pain palliation in radioisotope therapy with strontium-89. Nucl Med Commun 1996;17:799-804. 355. McEwan AJ. Use of radionuclides for the palliation of bone metastases. Semin Radiat Oncol 2000;10:103-114. 356. Singh A, Holmes RA, Farhanghi M. Human pharmacokinetics of samarium-153 EDTMP in metastatic cancer. J Nucl Med 1989;30:1814-1818. 357. Heggie JCP. Radiation absorbed dose calculations for samarium153-EDTMP localized in bone. J Nucl Med 1991;32:840-844. 358. Serafini AN, Houston SJ, Resche I. Palliation of pain associated with metastatic bone cancer using samarium-153 lexidronam: a double-blind placebo-controlled clinical trial. J Clin Oncol 1998;16:1574-1581. 359. Eary JF, Collins C, Appelbaum FR. Sm-153 EDTMP treatment of hormone refractory prostate carcinoma [abstract 203]. J Nucl Med 1990;31:755. 360. Turner JH, Claringbold PG. A phase II study of treatment of painful multifocal skeletal metastases with single and repeated dose samarium-153 ethylenediaminetetramethylene phosphonate. Eur J Cancer 1991;27:1084-1086.

35  The Bone 361. Bayouth JE, Macey DJ, Kasi LP, et al. Dosimetry and toxicity of samarium-153 EDTMP administered for bone pain due to skeletal metastases. J Nucl Med 1994;35:63-69. 362. Collins C, Eary JF, Donaldson G, et al. Samarium-153-EDTMP in bone metastases of hormone refractory prostate carcinoma: a phase I/II trial. J Nucl Med 1993;34:1839-1844. 363. Serafini AN. Current status of systemic intravenous radiopharmaceuticals for the treatment of painful metastatic bone disease. Int J Radiat Oncol Biol Phys 1994;30:1187-1194.

939

364. Farhanghi M, Holmes RA, Volkert WA, et al. Samarium-153EDTMP: pharmacokinetic, toxicity and pain response using an escalating dose schedule in treatment of metastatic bone cancer. J Nucl Med 1992;33:1451-1458. 365. van Rensburg AJ, Alberts AS, Louw WK. Quantifying the radiation dosage to individual skeletal lesions treated with samarium153 EDTMP. J Nucl Med 1998;39:2110-2115.

36 The Soft Tissue Matthew T. Ballo and Gunar K. Zagars EPIDEMIOLOGY AND ETIOLOGY PATHOLOGY PRESENTATION, NATURAL HISTORY, AND PROGNOSIS Metastasis Local Recurrence STAGING Staging System Diagnostic and Staging Workup TREATMENT Surgery Adjuvant Radiation Therapy RADIATION THERAPY TECHNIQUES Volume Considerations Dose Considerations

By convention, the term soft tissue refers to all of the nonepithelial, nonreticuloendothelial extraskeletal tissues of the body, including the fibrous connective tissue, adipose tissue, muscle (skeletal and smooth), blood and lymph vessels, and peripheral nerve sheaths. Neoplasms arising within and consisting of soft tissue elements are called soft tissue tumors. Despite their histopathologic diversity, these tumors share many pathobiologic features, justifying their inclusion in one oncotaxonomic group. Soft tissue tumors can be broadly classified according to their clinical behavior as benign, borderline, or malignant. Benign tumors (e.g., lipoma, neurofibroma, leiomyoma) closely resemble their normal tissue of origin, produce little or no local invasion, have no metastatic propensity, and rarely recur after excision. Borderline lesions (e.g., desmoid tumor, dermatofibrosarcoma protuberans) are locally invasive and have a significant tendency to recur after excision, and some types occasionally metastasize. Malignant soft tissue tumors—soft tissue sarcomas (STSs)—are locally invasive and have a high propensity for hematogenous metastasis. Although many sarcomas have a benign histogenetic counterpart (e.g., liposarcoma-lipoma, ­ leiomyosarcomaleiomyoma, fibrosarcoma-fibroma), transformation of soft tissue tumors from benign to malignant forms is rare, and most sarcomas arise without known precursor lesions. Moreover, because the benign tumors are easily cured by resection, radiation therapy has little role in their management. This chapter deals only with sarcomas and selected intermediate soft tissue tumors.

Epidemiology and Etiology Approximately 9200 new cases and 3500 deaths are caused by STS in the United States each year. However, STS is rare and accounts for less than 1% of all new non–skin cancer

940

MANAGEMENT ISSUES Complications Radiation Therapy Alone Chemotherapy Recurrent Disease Special Anatomic Sites SELECTED TUMOR TYPES Malignant Fibrous Histiocytoma Synovial Sarcoma Liposarcoma Desmoid Tumor Dermatofibrosarcoma Protuberans Hemangiopericytoma

diagnoses.1 The median patient age at presentation varies from 40 to 60 years, although these tumors can occur at any age from infancy on.1-5 Males are slightly more often affected than females, but this sex difference is small.2 No racial predilection has been found.5 Etiologic factors for STS are poorly defined. Although patients sometimes report antecedent trauma to the site of their sarcoma (e.g., prior surgery, reaction to a foreign body6-8), it is thought that trauma serves to bring the patient’s attention to the lesion rather than having any causative role. Chronic lymphedema of the upper extremity as a sequela of breast cancer management is associated with the development of lymphangiosarcoma, a phenomenon known as the Stewart-Treves syndrome.9 Postirradiation sarcoma is a well-recognized potential complication of radiation therapy. The diagnosis of radiation-induced sarcoma requires that the sarcoma arise in a previously irradiated site after a latency period of at least a few years.10 With an incidence of 0.03% to 0.22% and a median latency period of 10 years, ­ radiation-induced sarcomas have been difficult to ­characterize.11 However, they usually manifest in an advanced stage, are high grade, metastasize early, respond poorly to therapy, and are fatal. Malignant fibrous ­histiocytoma (MFH) is the most commonly diagnosed radiation-induced sarcoma, followed by angiosarcoma.12,13 Postirradiation liposarcoma has not been reported.5 The etiologic significance of exposure to phenoxyacetic acid herbicides (e.g., Agent Orange) remains controversial, although most reports do support at least a weak association.14-16 In keeping with the concept that oncogenes are responsible for carcinogenesis, evidence is mounting for genetic etiologic factors in selected STSs. Several sarcomas occur in well-defined hereditary settings. Neurofibromatosis type 1 (von Recklinghausen’s disease), inherited as an autosomal dominant defect that consists of a mutation that causes a

36  The Soft Tissue

lack of function in the neurofibromin gene, is characterized by the development of multiple benign neurofibromas and by a high incidence of malignant peripheral nerve sheath tumors (i.e., malignant schwannoma, neurogenic sarcoma, or neurofibrosarcoma).17 Optic glioma, astrocytoma, and pheochromocytoma also occur in these patients. Neurofibromatosis type 2 (hereditary acoustic neuroma), an ­autosomal dominant disorder defined by a mutant merlin protein, is characterized by the development of acoustic neuroma and is associated with meningioma, glioma, and schwannoma.5 Hereditary retinoblastoma—also an autosomal dominant disease, with a mutation that causes a lack of function in the retinoblastoma gene—is associated with an increased risk of bone and STSs.18 Gardner’s syndrome, an autosomal dominant variant of hereditary polyposis coli, is associated with desmoid tumors and rarely with fibrosarcoma or ­liposarcoma.5 Among conditions heralding a hereditary predisposition to sarcoma is the Li-Fraumeni syndrome, which is inherited as an autosomal dominant trait with incomplete penetrance and is caused by a mutation that causes a lack of function in the TP53 gene. This disorder manifests with a high incidence of a variety of tumors, among them MFH, rhabdomyosarcoma, osteosarcoma, and breast cancer occurring in patients younger than is typical.19,20 Genetic factors are further implicated by the finding of characteristic chromosome translocations in some sarcomas. Consistent translocations have been described in myxoid liposarcoma (t[12;16]), clear cell sarcoma (t[12;22]), alveolar rhabdomyosarcoma (t[2;13]), synovial sarcoma (t[X;18]), and Ewing’s sarcoma (t[11;22]).21

Pathology The pathologic classification of STSs is based on the type of tissue comprising the tumor—that is, on its histologic appearance.5,22,23 For many STSs, this histologic classification also implies a specific histogenesis or cell type of origin, such as leiomyosarcoma from smooth muscle, liposarcoma from fat, angiosarcoma from lymph or blood vessels, or rhabdomyosarcoma from skeletal muscle. For some STSs (e.g., MFH, alveolar soft part sarcoma, synovial sarcoma), the histologic classification does not imply a well-defined cell of origin. About 25 major categories of STSs are recognized, and when subcategories are considered, about 60 types have been described.5,24 Complicating this taxonomy is the use of some 300 synonyms for these 60 types of STS.24 The more common types are MFH (20% to 30% of cases), ­liposarcoma (10% to 20%), synovial sarcoma (5% to 10%), rhabdomyosarcoma (5% to 10%), leiomyosarcoma (5% to 10%), malignant peripheral nerve sheath tumor (i.e., malignant schwannoma) (5% to 10%), and fibrosarcoma (5% to 10%).5,24-26 These tumors account for approximately 70% to 80% of all STSs. Accurate recognition of STS type requires expertise in the pathology of these tumors. Even then, the rate of agreement among pathologists on histologic type is only about 70%.27,28 In difficult cases, the use of immunocytochemical testing or electron microscopy can be helpful.29 Although no marker pattern is diagnostic of a particular tumor, the following markers can aid in diagnosis.5,21

941

Vimentin, a cytoskeletal protein normally present in mesenchymal tissue but absent in epithelia, occurs in most STSs and may occasionally occur in adenocarcinomas. Desmin, an integral part of the cytoskeleton of cardiac, skeletal, and smooth muscle, occurs in most rhabdomyosarcomas but occurs in only one half of leiomyosarcomas and is occasionally expressed in Ewing’s ­sarcoma, liposarcoma, and MFH. Cytokeratin, characteristic of epithelia, is a classic feature for the diagnosis of carcinoma but has been found in some epithelioid sarcomas, synovial sarcomas, and other STSs. S-100 protein, a brain-specific protein, is present in most melanomas and clear cell sarcomas but is also found in many malignant nerve sheath tumors and in some carcinomas. In addition to classifying an STS according to its type, the pathologist must assign a grade—an estimate of the degree of the tumor’s virulence based on its histologic appearance. Features typically assessed for grading are cellular differentiation, cellular pleomorphism, degree of cellularity, ­mitotic activity, necrosis, and invasiveness.30 Three-tier27,31-33 and four-tier34 grading schemes have been proposed. The American Joint Committee on Cancer (AJCC) staging system recommends a four-tier system (Box 36-1),35 but a three-tier system of low, intermediate, and high grade seems to be the most widely used. Unfortunately, the grading of STSs remains largely subjective and is extremely controversial. Different investigators have found different pathologic features to be significant as grading criteria. Some found degree of cellularity to be significant,31,36 and others not28,32; some found degree of differentiation to be significant,28 and others not31,36; pleomorphism was significant according to some,31,36 but not to others28; likewise, necrosis was found to be significant by some,28,31 but not by others.36 Most investigators have found mitotic rate to be a significant factor.28,31,36 In the face of this uncertainty, it may seem surprising that all of the various grading systems seem to predict the metastatic potential of STSs.28,31-34,36,37 An alternative to tumor grading on an individual basis is the use of histologic type as an indicator of grade.5 Histologic grading is based on multivariate analyses that include nonhistologic features such as tumor size and site (superficial versus deep) in the statistical model. These studies have shown that most of the classic histopathologic indicators of grade lose independent prognostic significance.22,26 For many years, we have used a purely histologic grading classification (Box 36-2), which correlates well with prognosis.22,38-41 Advances in understanding molecular markers suggest that objective measures of tumor virulence may emerge.42-44 Meanwhile, every effort should be made to grade each individual patient’s tumor because grade, however assigned, remains the strongest predictor of outcome.

Presentation, Natural History, and Prognosis Although STSs can arise in any part of the body, the most common site is an extremity, especially a lower extremity (Table 36-1).3,45-47 Most STSs arise within muscle, but

942

PART 12  Musculoskeletal System

BOX 36-1.  American Joint Committee on Cancer Staging System for Soft Tissue Sarcomas Primary Tumor (T) TX Primary tumor cannot be assessed T0 No evidence of primary tumor T1 Tumor 5 cm or less in greatest dimension T1a Superficial tumor T1b Deep tumor T2 Tumor more than 5 cm in greatest dimension T2a Superficial tumor T2b Deep tumor Regional Lymph Nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastases N1 Regional lymph node metastases Distant Metastases (M) MX Distant metastases cannot be assessed M0 No distant metastases M1 Distant metastases

Metastasis

Histopathologic Grade GX Grade cannot be assessed G1 Well differentiated G2 Moderately differentiated G3 Poorly differentiated G4 Undifferentiated Stage Grouping Stage I G1-2 G1-2 Stage II G3-4 Stage III G3-4 Stage IV Any G Any G

T1a-1b T2a-2b T1a-1b-2a T2b Any T Any T

N0 N0 N0 N0 N1 N0

As STSs grow, they typically respect the major intermuscular septa and remain confined to the compartment of origin. Extracompartmental extension may occur ­ after ­surgical manipulation but is typically found late because local extension occurs through perforating vascular ­channels.48 Most STSs grow along the direction of muscle bundles or along aponeurotic sheaths in the vicinity of their origin. In extremity STSs, this pattern of growth dictates generous longitudinal margins for surgery and radiation but permits relatively narrow transverse margins. Several histologic subtypes are characterized by unique local growth patterns. Desmoid tumor49 and epithelioid sarcoma50 lack the well-defined pseudocapsule described previously, making delineation and complete surgical resection difficult. Malignant peripheral nerve sheath tumors are difficult to resect completely because of their propensity for local extension along major nerve trunks. Angiosarcoma of the skin has a striking tendency for presentation with widespread dermal involvement that is often underappreciated clinically.51

M0 M0 M0 M0 M0 M1

From Greene FL, Page DL, Fleming ID, et al, eds. AJCC Cancer ­Staging Manual, 6th ed. New York: Springer, 2002, pp 193-200.

they also can arise in intermuscular planes, subcutaneous tissue, fat, nerves, veins, and other tissues. Although any type of STS can arise in any site, some types have a predilection for particular sites; for example, liposarcoma is the most common retroperitoneal STS, synovial sarcoma tends to arise near but not within joints, and myxoid liposarcoma favors the upper medial thigh. Usually, the patient notices a painless mass, although pain can occur if the lesion impinges on neurovascular structures. Typical are a 4-month delay between the onset of symptoms and presentation to a physician and a 1-month delay between presentation and diagnosis.3 The tumor’s growth rate depends on its histologic subtype and grade; local spread is directed along the pathways of least resistance. Growth begins radially, pushing surrounding tissues into a well-delineated compression zone. Beyond this area of compressed normal tissue lies a reactive zone ­characterized by edema, neovascularity, and inflammatory infiltrate. The combination of these two zones creates a pseudocapsule, which sometimes prompts inexperienced surgeons to perform a “shell-out” procedure.48 However, because the pseudocapsule invariably contains numerous points of microscopic tumor penetration,30 local failure is common if no additional treatment is given.

Lymph node metastases from STSs are uncommon, occurring in less than 10% of patients at any time during the disease.25,52 In a review of the experience at Massachusetts General Hospital, Mazeron and Suit25 reported that the incidence of STS metastases was 5.9% among 323 patients. This finding was consistent with their evaluation of pooled data from published reports on more than 2000 patients, in which the crude incidence of lymph node metastases was 3.9%. When combining the pooled data with their own institutional experience, these investigators found that histologic subtype, grade, and size were the most important determinants of lymph node metastasis. Lymph node involvement was highest for the clear cell (27.5%), epithelioid (20%), synovial-cell (13.7%), angiosarcoma (11.4%), and rhabdomyosarcoma (14.8%) subtypes. No lymph node metastases occurred among the 63 patients with grade 1 lesions, compared with incidences of 2% and 12% among the 118 and 142 patients with grade 2 and grade 3 lesions, respectively. Tumor diameter of less than 5 cm was associated with a 3% incidence of lymph node involvement, compared with 15% for lesions of 5 cm or larger. Hematogenous metastasis is a common and serious event in patients with STS. The lung is the most common site of metastasis, but lesions can appear in bone and, ­especially in the case of retroperitoneal STS, in the liver. Metastases are detected in 20% to 25% of patients at the time of initial presentation.3 New metastases develop within 5 years of local treatment in most patients,53-55 and the incidence of these metastases correlates closely with tumor grade and size. Each of these factor—grade and size—is a highly significant and independent determinant of metastatic relapse. The importance of grade in predicting metastases has been illustrated in several reports. Lindberg and colleagues,56 reporting on 300 patients, found metastasis rates of 5%, 28%, and 43% for grades 1, 2, and 3 tumors, respectively. Trojani and colleagues27 reviewed the experience of 155 patients and reported metastasis-free survival rates of 69%, 46%, and 16% at 5 years for those with grades 1, 2, and 3 lesions, respectively. Markhede and colleagues36 described similar findings using a four-tier system applied

36  The Soft Tissue

943

BOX 36-2.  Malignancy Grouping of Specific Types of Soft Tissue Sarcoma* Group 1 Locally recurring soft tissue neoplasms with no or very little metastatic potential Desmoid tumor Atypical lipomatous tumor Dermatofibrosarcoma protuberans

Synovial sarcoma Pleomorphic liposarcoma Cellular myxoid liposarcoma Dedifferentiated liposarcoma Angiosarcoma Leiomyosarcoma Malignant peripheral nerve sheath tumor (malignant ­schwannoma) Rhabdomyosarcoma (all types) Alveolar soft part sarcoma Extraskeletal osteosarcoma Extraskeletal Ewing’s sarcoma Malignant melanoma of soft parts (clear cell sarcoma) Epithelioid sarcoma

Group 2 Sarcomas of intermediate aggressiveness (but with metastatic capability) Myxoid liposarcoma Extraskeletal myxoid chondrosarcoma Myxoid malignant fibrous histiocytoma Hemangiopericytoma Group 3 Aggressive sarcomas Malignant fibrous histiocytoma

*Including locally recurring nonmetastasizing neoplasms. Modified from Evans HL. Classification and grading of soft-tissue sarcomas. Hematol Oncol Clin North Am 1995;9:653-656.

TABLE 36-1.    Anatomic Sites of 6882 Soft-Tissue Sarcomas Study and Reference

Lower Extremity

Upper Extremity

Head and Neck

Abbas et al45

81

42

24

al46

Trunk* 90

152

59

12

84

Torosian et al47

208

81

21

182

Lawrence et al3

2110

594

406

1440

681

225

84

306

Potter et

M. D. Anderson Cancer Center, 1960-2003 Total *Approximately

3232 (47%)

1001 (14%)

547 (8%)

2102 (31%)

50% were retroperitoneal.

to 97 patients; the metastasis rate was 0%, 30%, 34%, and 60% for grade 1, 2, 3, and 4 lesions, respectively. The influence of tumor size was illustrated in a report by Lindberg and colleagues,56 who found that tumors less than 5 cm in largest dimension had a metastasis rate of 13%, compared with a metastasis rate of 35% for tumors exceeding 5 cm. Although 5 cm is generally accepted as the boundary between “large” and “small” STS and is the basis of the AJCC T classification system,35 the relationship between size and metastatic propensity is more or less continuous. The continuity of this relationship is illustrated by our data on intermediate- and high-grade sarcomas showing that the actuarial 15-year rates of metastatic relapse were 11%, 25%, 40%, and 49% for tumors of less than 2 cm, 2 to 5 cm, 5 to 10 cm, and more than 10 cm, respectively.57 The independent effects of grade and tumor size on metastatic propensity are illustrated in our study of MFH.39 Pleomorphic MFH, a high-grade STS according to our system of classification (see Box 36-2),22 had 10-year metastasis rates of 23% and 51% for tumors of up to 5 cm and greater than 5 cm, respectively; myxoid MFH, an intermediate-grade STS, had 10-year metastasis rates of 4% and 22% for tumors of up to 5 cm and larger than 5 cm, respectively. Although grade was an important determinant of metastasis, tumor size was also important in each grade group.

The only factor other than grade and size that ­regularly correlates with metastatic relapse is tumor depth, with ­lesions superficial to the deep fascia displaying a significantly lower metastatic propensity than similar lesions located deep in the fascia.44,58-60 This feature is incorporated in the current AJCC staging system (see Box 36-1).35 Factors such as patient age and sex have an uncertain effect on ­metastatic propensity.5,36,57,61,62 Likewise, the significance of tumor primary site as an independent determinant of outcome is uncertain. Although STSs arising in the head and neck or in the retroperitoneum are associated with a significantly poorer outcome than are STSs at other sites,63-65 this phenomenon is thought to reflect the anatomic constraints on radical resection of head and neck lesions and the large size of retroperitoneal tumors, rather than any intrinsic virulence of lesions arising in these sites. Because patient survival is largely determined by whether metastasis develops, the prognostic factors for survival are the same as those for metastatic relapse.

Local Recurrence Because of their infiltrative growth pattern, STSs have a marked tendency to recur locally, especially after conservative excision. Even after more radical treatment, local recurrence remains a potential problem. After definitive

944

PART 12  Musculoskeletal System

treatment, whether by radical surgery alone or by combined conservation surgery and radiation, local relapse occurs in 15% to 25% of cases. The major prognostic factors for local recurrence are resection margins positive for disease, a history of prior recurrence, older age and tumor location within the deep trunk or head and neck.57,60 Tumor grade and size have some influence on the likelihood of local recurrence, but this effect is small and inconsistently documented.36,46,57,59,60,66-70 The propensity for local recurrence of low- or intermediate-grade tumors (e.g., desmoids, dermatofibrosarcoma protuberans, myxoid MFH) is similar to that of high-grade lesions (e.g., pleomorphic MFH, synovial sarcoma). Certain histologic subtypes have been associated with poor local control (e.g., MFH, neurogenic sarcoma, epithelioid sarcoma).57 Retroperitoneal STSs are notorious for their recurrence rates, which exceed 80% and even 90% in some series because of the anatomic impossibility of achieving adequate resection margins.63,71,72 It is important to realize that the prognostic factors for local recurrence (prior local recurrence, disease in the resection margin) are different from those for metastatic relapse (tumor grade, size of primary lesion, and depth of invasion).

Staging Staging System The AJCC staging system (see Box 36-1) uses three known prognostic factors: tumor size, tumor grade, and tumor depth. This system applies to all STSs except for Kaposi’s sarcoma, dermatofibrosarcoma protuberans, and desmoid tumor. Sarcomas arising within the confines of the dura mater, including the brain, and sarcomas arising within parenchymatous organs or hollow viscera are also excluded. Depth is defined in relation to the superficial muscular fascia. Any tumor involving the superficial muscular fascia and retroperitoneal disease are considered deep. The validity of this clinicopathologic staging system and its ability to predict relapse (in the form of distant metastasis) and tumor-related mortality have been confirmed.60,73,74 The AJCC’s use of tumor grade in assigning tumor stage in STSs is almost unique among the group’s staging systems, highlighting the importance of grade as a determinant of outcome for patients who have this type of cancer.

Diagnostic and Staging Workup Virtually all new, enlarging, or symptomatic soft tissue masses should be regarded as potential sarcomas. A timely and correctly performed biopsy is an important early step in evaluation. Four biopsy techniques are available: incisional biopsy, excisional biopsy, core-needle biopsy, and fine-needle aspiration. Excisional biopsy—removal of the entire grossly evident lesion without a margin—invariably passes through the ­apparent plane of the pseudocapsule (i.e., shell-out procedure), often contaminates surrounding tissue, complicates subsequent surgical management, and should not be performed.75-77 Incisional biopsy—removal of a wedge of ­tumor—usually is the most appropriate method but should be performed by an oncologic surgeon. The incision should be located so that the tumor can be excised easily during the definitive surgical procedure; the incision also should be

placed directly over the tumor to avoid tunneling through tissue. The incision should be longitudinal for extremity lesions, and hemostasis must be meticulous.75,78 A poorly planned or executed biopsy procedure can significantly reduce the chances for local control and increase the number and complexity of surgical procedures required.79,80 The less-invasive biopsy procedures—core-needle ­biopsy and fine-needle aspiration—have been criticized because of the small amounts of tissue obtained, but in experienced hands, they have been shown to be successful. Core-needle biopsy, which retrieves a thin sliver of tissue, usually by using a Tru-Cut needle, allows the correct histopathologic type and grade to be assigned to approximately 90% of STSs.81 In our practice, core-needle biopsy is the preferred initial biopsy technique and can be guided by computed tomography (CT), if appropriate. This technique has the potential to eliminate inappropriate biopsy-incision placement and to decrease the incidence of wound-healing ­complications.82 Less certain is the role of fine-needle aspiration in the initial diagnostic workup because even experienced cytopathologists are often unable to assign a grade or histopathologic type to samples obtained in this way.5,36,83 However, fine-needle aspiration is useful in documenting relapse in established cases of STS. A thorough assessment of the AJCC staging features and of relevant patient factors is crucial to the formulation of an optimal treatment strategy for a patient with STS. Clinical findings are usually disappointingly nonspecific, and radiologic imaging studies are indispensible.84 The primary imaging studies for evaluating STS are CT and magnetic resonance imaging (MRI), supplemented by plain radiography, ultrasonography, and angiography when appropriate. The optimal modality for evaluating a primary tumor varies with the tumor type, its location, and the features of interest.84 MRI usually is the modality of choice for evaluating the extent of STSs because it can delineate tumor margins and neurovascular structures and can provide images in several planes. In direct comparisons, preoperative MRI has been found to be superior to CT for assessing tumor resectability85 and for distinguishing muscle from tumor.86,87 However, CT provides better visualization of fatty tumors and tumor calcification and permits a better assessment of tissue-bone interfaces than does MRI.88 For the radiation oncologist, CT has the advantage of ­providing “scout” images from which the tumor can be related to bony landmarks. Possible uses of positron emission tomography (PET)89 and radionuclide imaging with novel radionuclides90 are areas of active research. In a prospective evaluation of 30 suspected malignant masses, PET scans were obtained and used to categorize the tumors as benign or one of three malignant grades. All 12 of the high-grade lesions were correctly identified by the PET scan characteristics. The lowgrade lesions could not be distinguished from the benign lesions.89 Although research is in its early phase, PET scanning and radionuclide imaging may become increasingly important, particularly in evaluating the response of STSs to nonsurgical treatment.90,91 Because STSs have a marked propensity for metastasis to the lungs, all patients with intermediate- or high-grade tumors who have normal chest x-ray images should undergo CT scanning of the chest.92 Patients with ­retroperitoneal

36  The Soft Tissue

STS should be evaluated for hepatic metastases by CT. Because myxoid liposarcoma metastasizes preferentially to retroperitoneal soft tissues,40 patients who have this type of STS should undergo CT scanning of the abdomen. Patients with alveolar soft part sarcoma should undergo a radionuclide brain scan because of this sarcoma’s peculiar tendency to metastasize to the brain.93 The low incidence of regional nodal metastases suggests that specific evaluation of lymph nodes is unnecessary, although radionuclide sentinel node mapping94 may be appropriate for patients with epithelioid or clear cell sarcoma (i.e., melanoma of soft parts). In practice, the regional nodes are often ­included in the imaging studies of the primary tumor and further investigation is not needed.

Treatment Surgery Surgical resection remains the mainstay of treatment for most patients with localized STS. Except for select sarcomas, such as the round cell types (i.e., extraskeletal Ewing’s sarcoma, malignant peripheral neuroectodermal tumor, embryonal and alveolar rhabdomyosarcoma, and neuroblastoma), which can be successfully treated with a combination of radiation and chemical agents, most STSs cannot be reliably eradicated without resection of gross disease. Historical experience with limited excision has demonstrated a high incidence of residual microscopic and even gross tumor76 and an incidence of local recurrence as high as 90%.45,48,53,55,95,96 The infiltrative nature of these tumors often dictates the use of radical surgical procedures. In an attempt to rationalize surgery for STS, Enneking and colleagues48 presented a classification of surgical

945

­ rocedures. This system was developed for lesions of the exp tremities; true radical resection is rarely possible for lesions arising in the head and neck, trunk wall, mediastinum, or retroperitoneum. The Enneking classification has two surgical procedures—local excision or amputation—and four surgical margins—intralesional, marginal, wide, or radical. The resulting eight combinations of terms describe the type of surgical procedure possible and the relationship between the plane of dissection, the lesion, its pseudocapsule, and surrounding normal tissues (Table 36-2).48 An amputation, for example, can range from a radical procedure, in which the level of amputation passes proximal to the origin of the muscles containing the lesion, to a debulking procedure, in which the level of amputation actually passes through the lesion itself. A local procedure can be equally radical in that the entire compartment containing the lesion is removed, or it may be incisional, wherein only a small portion of the lesion is removed for diagnostic purposes. The terms wide or marginal reflect margins as outlined in Table 36-2. In practice, the terms amputation, radical compartmental resection (i.e., monobloc soft part resection),97 and wide local excision describe the majority of therapeutic surgical procedures performed for STS. The use of this terminology, however, is no substitute for pathologic assessment and designation of the final margin as positive or negative for disease. As indicated, unsatisfactory early experience with ­limited excision resulted in the adoption of radical ­resection—radical monobloc soft part resection, radical ­compartment resection, and amputation—as the standard surgical procedures for STS. Table 36-3 summarizes the local control rates achieved with radical surgery as the sole treatment.36,45,55,67,98,99 In general, local recurrence of

TABLE 36-2.    Four Surgical Margins and Related Procedures Surgical Margins and Related Procedures

Definition

Radical margin

An aggressive procedure that removes the entire anatomic compartment involved by tumor, including neurovascular structures.

Radical amputation

Amputation is proximal to the origin of the muscles adjacent to the ­lesion, usually above the joint proximal to the tumor.

Radical compartment resection

All compartmental muscles are removed from origin to insertion.

Wide margin

The lesion and surrounding reactive zone are removed, along with a cuff of normal tissue.

Wide amputation

The level of amputation is within normal tissue outside the reactive zone.

Wide local excision

A rim of normal tissue surrounds the plane of dissection, but no attempt is made to remove the entire muscle.

Myectomy

The involved muscle is removed only from origin to insertion.

Marginal margin

The plane of dissection passes through the pseudocapsular-reactive zone. This is usually associated with positive microscopic margins.

Marginal amputation

The level of amputation passes through the reactive zone.

Marginal excision (excisional biopsy or “shell-out”)

This method is appropriate for benign lesions only.

Intralesional margin

This diagnostic procedure consists of resection inside the pseudocapsule. It is invariably associated with positive microscopic margins.

Debulking amputation

The level of amputation passes through the lesion.

Incisional biopsy or intralesional debulking

This is the preferred diagnostic procedure.

Modified from Enneking WF, Spanier SS, Malawer MM. The effect of the anatomic setting on the results of surgical procedures for soft parts sarcoma of the thigh. Cancer 1981;47:1005-1022.

946

PART 12  Musculoskeletal System

TABLE 36-3.    Local Recurrence Rates after Radical Surgical Procedures Alone Study and Reference

Number of Patients

Number of Primary Amputations (%)

5-Year Local Recurrence Rate (%)

Cantin et al55

653

Not available

29

Abbas et al45

143

51 (36)

26

45

23 (51)

17

97

15 (15)

22

139 (47)

18

29 (54)

17

Simon et al98 Markhede et Shiu et

al36

al99

Berlin et

al67

297 54

­ isease occurs in 10% to 25% of patients, depending on the d surgical series. Amputation achieves local control in more than 90% of patients,36,67,68,100 whereas radical, nonamputative resection achieves local control in 75% to 85% of patients.36,48,60,67,68,100,101 Radical surgery alone can achieve high rates of local control. However, this salutary effect comes with significant morbidity and a rate of primary amputation of 50% in some series (see Table 36-3). Moreover, the expected survival time for patients treated with amputation has often been inferior to that for patients treated with less radical resection because patients who undergo amputation generally have large, high-grade lesions and a high incidence of micrometastatic disease.100 The high local control rate achieved by amputation has never been reflected in overall survival rates. The use of truly radical surgery is anatomically limited to extremity tumors and is rarely applicable to nonextremity sites. Recognizing these problems, many investigators have explored the combination of conservative surgery and ­radiation therapy. Other surgical oncologists, however, have sought to redefine the selection criteria for tumors that might be adequately managed using less radical surgery— myectomy or function-sparing wide local excision—as the sole treatment.53,83,101 The results of both approaches are considered in the following paragraphs.

Adjuvant Radiation Therapy Although radical surgery alone provides satisfactory local control, it does so at the expense of significant functional deficits, and it is rarely applicable to nonextremity tumors. By the 1960s, it was known that complete responses could be attained when radiation therapy was used alone for gross disease,102 dispelling the long-held belief that STSs were highly resistant to radiation. Clinical research then turned to combining limited (conservative or conservation) surgical excision with adjuvant radiation therapy. Suit and colleagues103 initially reported the successful use of this technique at The University of Texas M. D. Anderson Cancer Center in the early 1970s. Lindberg and colleagues56 updated that experience, reviewing 300 patients with STS treated with conservative surgical resection (excisional biopsy or wide local excision) and postoperative radiation. No patient was treated for gross disease, and radiation doses ranged from 60 to 75 Gy. The 5-year overall survival and local control rates of 61% and 78% were comparable to those attained by radical surgery alone but the combination of conservative surgery and postoperative radiation did not cause loss of limb

function. The ultimate limb preservation rate of 84% is consistent with the 83%,104 84%,61 87%,105 and 90%100 rates reported by colleagues applying similar conservation techniques. The only randomized trial comparing radical amputation with limb-sparing surgery plus radiation therapy was performed by the National Cancer Institute (NCI).106 In that trial, 43 adult patients with high-grade STS of an extremity were prospectively and randomly assigned, in a 2-to-1 fashion, to receive limb-sparing surgery plus radiation therapy (27 patients) or radical amputation (16 patients). Both randomization groups received doxorubicin, ­cyclophosphamide, and high-dose methotrexate chemotherapy. Patients in the limb-sparing group received 50 Gy to the entire anatomic area at risk, with a boost to the tumor bed of an additional 10 to 20 Gy. At a median followup time of 56 months, four local recurrences had occurred in the group given limb-sparing surgery plus radiation therapy, but no local recurrences of disease occurred in the amputation group (P = .06). Despite the slight difference in the incidence of local failure, no differences were found in ­disease-free or overall survival rates.

Conservation Surgery plus Radiation Therapy In an attempt to further define the role of radiation therapy, two prospective, randomized trials (Table 36-4) were undertaken to compare conservation surgery plus radiation therapy with conservation surgery alone.107-109 The first of these trials,107 from Memorial Sloan-Kettering Cancer Center, randomly assigned patients with extremity or superficial trunk STS to undergo conservation surgery alone (86 patients) or conservation surgery plus brachytherapy (78 patients). The brachytherapy was performed using catheters afterloaded with iridium 192 (192Ir) to deliver 42 to 45 Gy over 4 to 6 days. After a median follow-up time of 76 months, the 5-year actuarial local control rate was 82% in the conservation surgery plus brachytherapy group compared with 69% in the conservation surgery–only group (P = .04). The positive effect of adjuvant radiation therapy was limited to the 119 patients who had high-grade lesions and was not apparent in the 45 patients who had low-grade disease. Differences in the treatment effect could have resulted from small patient numbers,107 particularly among those expected to be at the highest risk of local recurrence (only 9 of 45 patients who had low-grade disease and positive margins were randomized), or radiobiologic reasons (cells in slow-growing low-grade tumors might not have

36  The Soft Tissue

947

TABLE 36-4.   Randomized Trials Comparing Conservation Surgery plus Radiation Therapy with Conservation Surgery Alone Study, Reference, and Tumor Grade

Number of Patients

Local Failure Rate (%)

P Value

Median Follow-up Time (yr)

RT

44

0 (0)

.003

9.6

63 Gy EBRT

No RT

47

9 (22)

RT

26

1 (4)

.016

9.9

63 Gy EBRT

No RT

24

8 (37)

RT

56

5 (11)

.0025

6.3

42-45 Gy brachytherapy

No RT

63

19 (34)

RT

22

8 (33)

.49

6.3

42-45 Gy brachytherapy

No RT

23

6 (24)

Treatment

Radiation Dose and Technique

Yang et al109 High-grade lesions Low-grade lesions Pisters et al107 High-grade lesions

Low-grade lesions

EBRT, external beam radiation therapy; RT, radiation therapy.

been given an opportunity to traverse radiosensitive phases of the cell cycle during the implantation period).108 The second trial109 from the NCI randomly assigned patients with extremity STS to undergo limb-sparing surgery alone (71 patients) or limb-sparing surgery plus external beam radiation therapy (EBRT) (70 patients). The radiation consisted of 45 Gy (1.8 Gy/fraction) to wide fields ­followed by an 18-Gy boost to the tumor bed (defined by surgical clips). At a median follow-up time of 115 months, 17 failures had occurred in the surgery-only group, ­compared with one failure in the group receiving surgery plus radiation (P = .0001). Unlike the Memorial SloanKettering Cancer Center experience, in this study, radiation therapy improved local control of the 91 high-grade tumors and the 50 low-grade tumors. This discrepancy suggests a radiobiologic difference in the sensitivity of STSs to brachytherapy and EBRT, but it also may reflect the fact that more of the low-grade tumors in the NCI series were at high risk of local recurrence (i.e., 21 of 50 patients had low-grade disease and close or positive margins). Numerous experiences in which conservation surgery has been followed by radiation therapy have been published, and overall, this approach has been found to result in local control for 80% to 90% of patients who have extremity or nonretroperitoneal truncal STS (Table 36-5).56,61,62,68,70,96,100,104,107,109-114 These results are comparable to those achieved by using radical surgery alone (see Table 36-3). The most consistently identified prognostic factors for local recurrence after conservation surgery and radiation therapy are resection margins positive for microscopic disease60,,68,69,112,115 and prior local recurrence.39,60,70,114 Retroperitoneal STSs are notorious for local recurrence because of anatomic constraints precluding complete resection and normal tissue tolerances limiting the permissible radiation doses.63,71,72,116,117 Similar constraints hamper local control in the head and neck region.118 Less consistently reported factors that may influence local control include radiation dose (Table 36-6)61,68,70,96,100,112,114,119

or field size61,100 and tumor histologic subtype.40,69,114 Factors that do not correlate with local control when using the combined-modality treatment approach are tumor size and grade and depth with respect to the superficial fascia (superficial versus deep).40,41,60,68,70,74,100,104,107 Traditionally, radiation has been applied postoperatively (see Table 36-5). This sequencing has the advantages of allowing the radiation therapy dose and field size to be tailored to fit the surgical findings (e.g., margin status, tumor histologic subtype and grade); eliminating delay in complete resection of the primary lesion; and possibly eliminating the need for radiation therapy altogether in select cases. However, preoperative radiation has its advantages, too. It requires smaller field sizes and lower radiation doses; it eliminates any delay between incomplete surgical resection and the start of radiation therapy; it treats a fully oxygenated tumor and limits the dissemination of viable tumor cells at the time of resection; and it increases resectability by eliminating peripheral microscopic tumor extension.

Preoperative Radiation The potential for smaller radiation volumes in the preoperative setting was confirmed by Nielson and colleagues,120 who reported on a series of 26 patients who had undergone preoperative radiation and independent repeat simulation after undergoing definitive surgery. The target volume in the postoperative simulation included the entire surgical cavity and drain sites rather than gross tumor only, as it had in the preoperative simulation. The size of the preoperative radiation field and the number of joints included were significantly smaller than in the postoperative setting. The potential effectiveness of a lower radiation dose makes preoperative radiation more attractive than postoperative radiation. Although never definitively proven, the notion that tumors in surgically undisturbed tissues respond better to radiation, perhaps by virtue of their greater oxygenation, is hallowed by long tradition, going back to the work of Fletcher in the 1960s.121 It is generally accepted that 50 Gy delivered preoperatively has the

948

PART 12  Musculoskeletal System

TABLE 36-5.    Local Control Rates after Postoperative Radiation Therapy Study and Reference Leibel et

al96

Potter et

al62

Abbatucci et

al110

al68

Number of Patients

Follow-up Time

Evaluation Period

Local Control Rate

29

2 yr (minimum)

Crude

90%

123

3.2 yr (median)

Crude

92%

113

Not stated

5 yr

86%

100

1.2 yr (median)

Crude

72%

Pao and Pilepich100

50

5.8 yr (median)

Crude

78%

Dinges et al70

93

Not stated

5 yr

82%

117

7.4 yr (median)

5 yr

81%

150

Not stated

5 yr

87%

67

2.6 yr (median)

5 yr

87%

Bell et

Keus et

al104

Suit and

Spiro111

al112

Fein et

al61

50

3.6 yr (median)

5 yr

76%

Pisters et al107

78

6.3 yr (median)

5 yr

82%

Cheng et al113

64

5.3 yr (mean)

5 yr

91%

70

9.6 yr (median)

10 yr

98%

743

12 yr (median)

5 yr

85%

Mundt et

Yang et

al109

Zagars and

Ballo114

TABLE 36-6.    Radiation Dose-Response Data for Microscopic Disease Study and Reference al96

Leibel et Bell et

al68*

Pao and

Pilepich100 al119†

Median Dose

Dose Range

Dose Response

—

50-75 Gy

NS

—

24-60 Gy

NS

60 Gy (mean)

45-69 Gy

NS

—

40-75 Gy

NS

Dinges et al70

59 Gy

45-72 Gy

>65 Gy

Fein et al112‡

60.4 Gy

39.6-71 Gy

>62.5 Gy

—

50-68 Gy

>60 Gy

62 Gy

40-75 Gy

≥64 Gy

Robinson et

Mundt et

al61

Zagars and

Ballo114

*Fraction

size varied from 2 to 8 Gy. size varied from 1.25 Gy twice daily to 4.4 Gy once daily. ‡Some patients received an interstitial boost of 10 to 20 Gy (median, 16 Gy). NS, no statistically significant association found.

†Fraction

same effect on microscopic tumor as 60 to 70 Gy delivered postoperatively. Several investigators have suggested that local control might be optimized through the selective use of preoperative or postoperative radiation therapy,38,111,122,123 ­although no comparative study has proved the superiority of one treatment sequence over another.113,124 We reviewed the experience of 271 patients given preoperative radiation therapy to a median dose of 50 Gy before excision or reexcision at M. D. Anderson Cancer Center and 246 patients given postoperative radiation therapy to a median dose of 60 Gy after definitive excision.124 When comparing the 10-year rates of local control, the treatment approach did have an effect (83% preoperative versus 72% postoperative on univariate analysis, P = .005); however, this effect was not significant on multivariate analysis and was explained by an imbalance in other prognostic factors. The complication rates among these patients were clearly associated

with the timing of the radiation therapy. In our experience, acute wound complications are more common among patients who undergo preoperative radiation therapy (34% preoperative versus 16% postoperative, P < .001), whereas late radiation-related complication rates depend on radiation dose when that dose is delivered postoperatively (18% for ≥ 60 Gy versus 9% for < 60 Gy, P = .04). These findings have been confirmed by a randomized trial comparing acute and chronic complication rates after preoperative or postoperative radiation therapy.125,126 Although the trial was not sufficiently powered to prove equivalence, no clinically significant differences were found in local control. Preoperative radiation, however, was ­ associated with more acute wound complications, as were lower extremity tumor site and tumor size exceeding 10 cm.125,127 From these findings, we conclude that the ­sequencing of radiation therapy should be based on the site of disease and the likelihood of long-term radiation-­related

36  The Soft Tissue

949

TABLE 36-7.    Local Control Rates after Preoperative Radiation Therapy Study and Reference Willet et Suit et

al128

al123

Dinges et

al70

Mundt et

al61

Cheng et al113 M. D. Anderson

Number of Patients

Follow-up Time

Evaluation Period

1.4 yr (mean)

Crude

100%

160

Not stated

5 yr

92%

9

Not stated

5 yr

89%

7

3.6 yr (median)

5 yr

88%

5.3 yr (mean)

5 yr

83%

5.9 yr (median)

6 yr

86%

27

48 374

complications. For most patients, preoperative radiation is preferable because of the lower radiation dose and ­smaller radiation treatment fields; preoperative radiation may ­result in temporary acute wound complications but will ultimately result in fewer long-term radiation-related complications. The experience of several investigators with preoperative radiation has shown local control rates to be on the order of 85% (Table 36-7).61,70,113,123,128 Whether the two approaches (preoperative versus postoperative) yield any real difference in local control remains unclear. However, resection margins containing tumor cells are particularly ominous for the patient treated with preoperative radiation. Tanabe and colleagues115 reviewed 95 consecutive patients with intermediate- or high-grade extremity STS who received preoperative radiation therapy (median dose, 50 Gy) and then limb-sparing surgery. The final microscopic margin was positive for disease in 24 patients and no additional radiation therapy was delivered. The 5-year rates of local control were 91% for ­marginnegative tumors and 62% for margin-positive tumors (P = .005). Multivariate analysis showed that positive surgical margins and intraoperative tumor violation independently predicted an increased risk of local failure. No satisfactory strategy has been developed for tumors found to have positive margins at surgery after preoperative irradiation. Potential approaches include the use of intraoperative boost radiation, a brachytherapy boost, and an EBRT boost. All of these approaches suffer from the theoretical disadvantage that they represent split-course ­irradiation. Eliminating radiation therapy for patients who clearly would not benefit from it is desirable, although identifying such patients is difficult. Tumor size has consistently been shown to be unrelated to the probability of local control,39-41,60,68-70,112,115 but interest has been renewed in managing lesions less than 5 cm in largest diameter with surgery alone. In a retrospective review of 174 patients with STS of the extremities and a median follow-up time of 48 months, Geer and colleagues101 reported that surgery alone yielded a local recurrence rate of only 11% among patients with small high-grade lesions and 7% among those with small low-grade lesions. The favorable overall 5-year survival rate of 94% for all patients and the lack of benefit from adjuvant radiation therapy or chemotherapy led the investigators to conclude that patients with small lesions should not be included in trials of adjuvant therapy. In contrast, another review of patients with small high-grade extremity lesions revealed 5-year local failure and overall survival rates of 18% and 83%, respectively.74 Because of

Local Control Rate

the long median follow-up time in the latter study (76 months), the investigators stressed the importance of longterm follow-up and reemphasized the importance of adjuvant therapy. It is still not clear which patients with T1 tumors can be treated effectively with conservation surgery alone; it would seem prudent to approach all such cases with combined surgery and radiation until findings from prospective randomized trials can be used to identify patients who can be spared the potential toxicity of ­combined-modality treatment.

Brachytherapy As an alternative to EBRT, brachytherapy has the advantage of delivering a high dose of radiation to a limited volume of tissue while accelerating the time in which the total dose can be delivered. Although the procedure is less expensive than EBRT129 and patients can usually return to “normal” activity in a relatively short period, clinical experience with this strategy is still rather limited (Table 36-8).107,130,131 After wide local excision and placement of metal clips to outline the areas at risk, afterloaded catheters are used to place single-plane implants of 192Ir. The target volume is defined as the surgical bed plus a margin of approximately 2 cm.108 The surgical scar and drain sites are not included. The catheters are loaded 6 or more days after surgery because the risk of wound complications increases if catheters are loaded within 5 days.132 The minimum dose is 42 to 45 Gy delivered over 4 to 6 days. The skin dose should be kept below 20 Gy.132a The relatively small target volume in brachytherapy remains a topic of controversy. It is at odds with the aim of EBRT, which stresses the importance of wide radiation margins. Some reports of the use of brachytherapy have substantiated this concern. Habrand and colleagues,130 from the Institut Gustave Roussy, reviewed the experience of 48 patients treated by iridium implantation with a 2- to 3-cm margin around the surgical bed. They reported local failure in 16 patients (33%); 14 of the failures (88%) occurred in the margins. Another concern with brachytherapy is its apparent ineffectiveness compared with EBRT for lowgrade tumors. Prospective randomized trials will be needed to delineate the relative places of brachytherapy and teletherapy as adjuvants to the surgical resection of STSs. The importance of local control and its relationship to overall survival remains controversial. In retrospective analyses, local recurrence has consistently been shown to be associated with an increased incidence of metastatic relapse and decreased rates of overall survival55,79,133-135; however, a causal relationship cannot be assumed. Local

950

PART 12  Musculoskeletal System

TABLE 36-8.    Surgery and Brachytherapy for Soft Tissue Sarcoma Study and Reference Habrand et

al130

Dose of Iridium 192 25-72 Gy over 2-19 (median, 60 Gy)

Number of Patients

Local Control Rate

48

67%

days*

Pisters et al107

42-45 Gy over 4-5 days

78

82%

Chaudhary et al131

29-50 Gy (median, 30 Gy)

33

75%

*Dose

range included four patients receiving external beam radiation therapy.

recurrence may simply represent aggressive disease, with distant metastases occurring at the time of initial presentation rather than at the time of local recurrence. The NCI and Memorial Sloan-Kettering Cancer Center randomized trials discussed previously showed statistically different local control rates without differences in overall survival. Although these trials suggested a lack of survival benefit through improved local control, their power was insufficient to prove no difference. Several retrospective studies have attempted to address this controversy further. Gustafson and colleagues136 reviewed the experience of 375 patients with STS of the extremities or trunk wall. Of the 128 patients who developed distant metastases, 63 had local failure and 65 did not. The timing of the metastases and the distribution of clinicopathologic factors were similar in the two groups, supporting the conclusion that local control was of minor importance as a determinant of metastatic failure. Similarly, Heslin and colleagues137 reviewed 168 patients with primary, high-grade, deep, and large (≥5 cm) extremity STS and found 21 local recurrences (13%). The time between the development of distant metastases and the date of initial presentation was no different for patients with and without local failure. Further elucidation of this problem awaits additional study, perhaps using local recurrence as a time-­dependent variable in a multivariate regression analysis.107,134,135 ­Although the influence of local recurrence on metastatic outcome remains uncertain, the importance of local control should not be underestimated. At the very least, local con­ t­rol secures the primary site, avoiding further ­ potentially debilitating local treatment; it also may have a salutary ­effect on metastatic dissemination.

Radiation Therapy Techniques Radiation therapy techniques for STSs depend on the location and size of the tumor and on the extent of prior surgery. The diversity in clinical situations encountered with these tumors requires a corresponding diversity in technical approaches, which can include the use of photons, electrons, compensating filters, wedge filters, three-dimensional treatment planning, intensity-modulated radiation therapy, and brachytherapy, all supplemented by a variety of positioning and immobilization devices. In the face of this technical diversity, the choice of an appropriate approach for an individual patient can be perplexing because no data demonstrate the superiority of one technique over another. Treatment planning is therefore based on general principles applicable to most clinical situations. Postoperative EBRT is considered first. Treatment should not begin before the surgical wound has healed completely and usually commences 4 to 6 weeks after the operation.

Volume Considerations The requisite postoperative radiation volume remains controversial, but it is universally accepted that the surgical bed plus a variable margin judged to be at risk of harboring occult tumor should constitute the treatment volume. Controversy surrounds the optimal width for this margin of normal tissue. Traditionally, margins of 5 to 7 cm for most tumors,56 or 5 to 10 cm for small lesions and 10 to 15 cm for large lesions,111,123,138 have been recommended. Few studies report the outcome according to radiation margin. Two reports found that margins exceeding 5 cm were superior to narrower ones, but that margins greater than 10 cm were not apparently beneficial.61,100 We typically use a 5- to 7-cm margin, as originally proposed by ­Lindberg.56 The radiation margin referred to is in the direction of the involved muscle compartment—the direction of potential spread of sarcoma—not in a direction perpendicular to it. For extremity sarcomas, this is the longitudinal direction, parallel to the limb. The radiation margin ­ perpendicular to the path of least resistance is usually 2 to 3 cm, and it cannot be larger without treating the whole circumference of the extremity. Drain holes and surgical scars are usually included in the radiation volume. For sarcomas at sites other than an extremity, these margin recommendations often cannot be achieved and lesser margins are used. The need for large radiation volumes, particularly the need to treat scars and drain holes, has been called into question by proponents of brachytherapy, who typically treat a 2- to 3-cm proximal-distal, 1- to 2-cm lateral, and 1.5- to 3-cm thick volume without irradiating the scar or drain holes.107,108 It is possible that the phenomenon of marginal recurrence has been overemphasized; in our experience, only about 20% of local recurrences are at the edge of or just beyond the radiation field.39-41 Nevertheless, until this issue is resolved, we continue to recommend the modest 5- to 7-cm longitudinal margin. Diagnostic imaging with MRI or CT before and after surgery is crucial to the delineation of an appropriate target volume. A radiation therapy planning CT scan is ­indispensable in most situations, and centers that cannot perform such scans should not be irradiating patients who have STS. General principles and some data103,139 dictate consideration of the following issues regarding radiation volume. The whole circumference of an extremity should never be treated to more than about 50 Gy, and a generous strip of unirradiated tissue should remain to allow lymphatic drainage. At least one half of the cross section of a weight-bearing bone should be spared, if possible, and the entire cavity of a major joint should be irradiated only if absolutely necessary. Likewise, attempts should be made to spare major tendons, such as the patellar and Achilles tendons. Organ

36  The Soft Tissue

A

951

B

FIGURE 36-1.  Radiation therapy fields used for treatment of thigh soft tissue sarcoma. A, A 53-year-old man underwent wide local excision in December 1980 for a 9 × 7 × 4 cm tumor within the proximal rectus femoris muscle. Pathologic examination confirmed malignant fibrous histiocytoma and resection margins negative for disease. Parallel-opposed, anteriorly weighted 6-MV photon fields were used to deliver 50 Gy over 5 weeks. An additional 10 Gy was given through an appositional electron beam over the scar and surgical bed. The patient was alive with no evidence of disease as of January 1998. B, A 75-year-old woman underwent incisional biopsy of a rapidly enlarging medial thigh mass in March 1984. Pathologic examination confirmed malignant fibrous histiocytoma. After seven cycles of cisplatin, vindesine, dacarbazine, and doxorubicin chemotherapy, preoperative radiation was delivered to a dose of 50 Gy over 5 weeks using opposed oblique fields of 6-MV photons. The leg was abducted, laterally rotated, and flexed during treatment.

tolerance must be respected when irradiating intrathoracic or retroperitoneal STSs. Lesions of the hand and foot can be irradiated with an acceptable outcome if attention is paid to technical details.140

Dose Considerations Having defined the radiation volume, attention turns to the selection of an appropriate dose. This too is controversial. As summarized in Table 36-6, no agreement has been reached on a dose-response relation within the range of 60 to 70 Gy, but most radiation oncologists recommend a total dose of at least 60 Gy. Our current practice, based on recent analysis,114 calls for a postoperative radiation dose of 60 Gy at 2 Gy/fraction for tumors with margins negative for microscopic disease and a dose of 68 Gy when the resection margins are positive. For patients with recurrent disease or primary disease in head, neck, or deep trunk ­locations, we recommend 64 to 66 Gy if allowed by local anatomic ­constraints. To minimize the risk of complications, a shrinking-field approach is used. The initial large volume is treated to 50 Gy, and the next 10 Gy is given in an area of “conedown” to the surgical bed (outlined by surgical clips or defined by the scar) plus a 1- to 2-cm margin. If a further boost is ­given, it is usually directed to the tumor bed itself. The defi­ nition of radiation volume for preoperative radiation proceeds along the same lines. We recommend the same 5- to 7-cm margin longitudinally and a 2- to 3-cm margin ­laterally.

The dose for preoperative radiation is 50 Gy at 2 Gy/fraction, and no field reductions are made routinely.111,114 After the volume and dose are decided on, detailed planning proceeds to delineate patient positioning, immobilization, and beam geometry. Most tumors arising in an extremity can be irradiated by using parallel opposed beams, with the extremity suitably positioned and with the use of custom blocking (Figs. 36-1 and 36-2). In general, extremity lesions are best treated with 4- to 6-MeV photon beams, supplemented by boost fields as appropriate. The use of high-energy beams may be appropriate, but care must be taken not to underdose superficial regions. In this regard, the judicious use of bolus material is useful for lesions of the hands and feet and for superficially located lesions. The upper extremity is often best treated in an abducted position so that it is away from the trunk, but such a setup can prevent a planning CT scan from being obtained with the patient in the treatment ­ position. Lesions of the upper inner thigh are best treated with the extremity in a frog-leg position so that the radiation is directed away from the contralateral thigh and the genitalia. Lesions of the buttock or posterior thigh are best approached with the patient in a prone position and with a wedge-pair field arrangement. Superficial trunk lesions can be treated using a pair of tangents or an appositional electron beam. Retroperitoneal tumors pose a significant therapeutic problem, and the utility of radiation is often debated.

952

PART 12  Musculoskeletal System

into the target site. Orthogonal radiographs provide threedimensional data for calculating the prescribed dose, which is specified at 1 cm from the plane of the implant. For adjuvant irradiation with brachytherapy alone, a dose of 45 Gy over 3 to 4 days is commonly prescribed.108,141 The catheters are not loaded with radionuclide until at least the sixth day after surgery to allow some wound healing and thereby reduce the risk of wound complications.132

T 5250 5000

4500

Management Issues Complications

A

cm

B FIGURE 36-2.  A 24-year-old woman underwent excisional biopsy at a referring institution in November 1999 for a slowly enlarging mass measuring 4 × 5 cm and located anterior to the left tibia. Pathologic examination confirmed epithelioid sarcoma. Physical and radiologic examination revealed gross residual disease measuring 1.5 × 2 cm. A, Preoperative radiation was delivered to a dose of 50 Gy over 5 weeks using left posterior and right anterior oblique fields of 60Co without wedges. B, The patient was positioned to avoid irradiation of the contralateral leg. Immobilization and a three-point laser line setup were used for daily reproducibility.

­ arallel-opposed fields, often oblique, commonly cover the P lesion while sparing adjacent organs. Occasionally, a more complex setup is necessary, and three-dimensional conformal techniques may be valuable. In any event, the radiation margin for these tumors tends to be smaller than that for extremity tumors. The deep location of these tumors dictates the use of high-energy photons (10 to 18 MV). Lesions of the head and neck are often amenable to three-dimensional or intensity-modulated therapy because their daily setup is highly reproducible and radiation volumes tend to be small. These volumes must be defined sharply to avoid critical structures. Similar high-precision techniques are indispensable for paravertebral tumors. The technique of brachytherapy involves the placement of a single-plane implant in the surgical bed.108,141-143 ­After excision of the tumor, the surgical bed is outlined with radiopaque clips, and brachytherapy afterloading parallel catheters, spaced 1 cm apart, are inserted percutaneously

A variety of complications can follow the use of combined surgery and radiation. Among the complications that may occur are fibrosis and induration, wound breakdown, ­decreased range of joint motion, edema, cutaneous telangiectasia, joint contracture, cutaneous ulceration, peripheral neuropathy, osteonecrosis, bone fracture, and various internal organ dysfunctions, depending on the treatment site. Attributing complications to radiation or to surgery is difficult because both modalities can contribute to the genesis of any particular sequela. In practice, complications developing months or years after treatment are ascribed to radiation. Significant late complications requiring surgical management or interfering with function occur in 8% to 15% of patients irradiated postoperatively.39-41,56,62,68,112 Most complications are evident by 5 years, but a small proportion occur beyond this time.39,140 Patients should be counseled that an irradiated area remains at risk for complications indefinitely and that activities that pose a significant risk of trauma to the treated site need to be avoided. Although it has been difficult to prove, the incidence of radiation complications seems to correlate with dose and volume irradiated and with anatomic site treated. Doses exceeding 60 to 65 Gy seem to increase complications,55,139 as do increasing radiation field sizes.139 In our experience with primary lower extremity STS, receipt of preoperative radiation therapy and large tumor size are associated with acute wound complications, whereas proximal lower extremity tumor site, prior surgical complication, and radiation doses of 60 Gy or more are associated with chronic radiation-related complications.144 Irradiating the whole circumference of an extremity to doses beyond 50 Gy is fraught with risks. One study reported that 6 of 18 patients in whom the whole circumference of an extremity was irradiated required amputation because of severe complications, including edema.103 It is also well documented that the distal lower extremity tolerates treatment less well than does the upper extremity. In our experience with patients treated for STS of the distal extremities, the overall incidence of complications was higher for those with tumors of the foot and ankle; 2 of 33 patients underwent amputation in one series we studied, whereas no patients with a tumor of the wrist or hand required amputation because of a complication.140 Bone fractures are rare after radiation therapy but seem to be related to the volume of bone irradiated and whether periosteal stripping was performed. In a group of 334 patients with primary STS in the groin, thigh, or knee, we observed a total of five fractures (1.5%), and all five occurred in patients receiving radiation to the entire circumference of the femur. The actuarial fracture rate at 10 years in patients having had the whole circumference of bone

36  The Soft Tissue

irradiated and having had the bone exposed at the time of surgery for a groin, thigh, or knee primary tumor was 11%.144 Given this low rate of fracture, routine prophylactic fixation is not warranted. We have found that instead of radiation causing fracture, an otherwise uncomplicated bone fracture resulting from trauma may heal slowly because of prior radiation therapy and may require surgical repair.144 Treatment planning should aim for the following features to minimize the risk of complications. A generous strip of tissue should be spared to avoid irradiating the whole circumference of an extremity; at least one half of the cross section of a weight-bearing bone should be spared; treatment of the whole of a major joint cavity should be avoided; treatment of major tendons should be avoided; and the tolerance of other organs should be considered when treating intrathoracic, abdominal, and pelvic STSs. Physical therapy to maintain joint mobility in the face of radiation-related muscle wasting and shortening may prevent or ameliorate functional deficits.139 Quality-of-life assessments have failed to reveal any significant differences between patients treated with amputation and those undergoing conservation surgery and radiation.145,146 This finding probably reflects humans’ remarkable ability to adjust to adversity; it is hardly plausible that life without a limb is just as good as life with one. Radiation delivered preoperatively carries a significant risk of wound complications after surgery. Bujko and colleagues147 reported on 202 patients treated with preoperative radiation and conservation surgery. Of these patients, 143 were given a postoperative boost by an interstitial implant or EBRT. The overall wound complication rate was 37%: wound dehiscence occurred in 24% of patients, infection in 6%, seroma in 3.5%, skin graft breakdown in 3%, and hematoma in 0.5%. Reoperation was required in 17% of cases, and six patients (3%) required amputation as a result of wound-healing problems. Multivariate analysis showed that tumor in a lower extremity, increasing patient age, and postoperative boost with an interstitial implant predicted an increased risk of wound morbidity. The investigators suggested that gentle handling of tissues during the operation, meticulous hemostasis, avoidance of closure under tension, and elimination of dead space through the use of drains and rotation flaps could decrease wound complication rates. Retrospective analyses113,124 and one randomized trial125 have documented that wound complications are more common after preoperative than postoperative radiation. In a randomized trial reported by O’Sullivan and colleagues,125 190 patients with extremity STS were randomly ­allocated to receive 50 Gy as preoperative radiation or 66 Gy as postoperative radiation. At a median follow-up time of 3.3 years, 35% of the patients in the preoperative group and 17% of patients in the postoperative group had experienced a wound complication within 120 days of surgery (P = .01). Six weeks after surgery, patients treated preoperatively had worse lower extremity function and higher pain scores because of the higher incidence of acute wound complications, but late radiation morbidity tended to be worse in the patients treated postoperatively (48.2% versus 31.5%, P = .07) because fibrosis, joint stiffness, and edema adversely affect long-term function.125,127 Although some retrospective evidence exists to suggest that the appropriate use of

953

TABLE 36-9.   Local Control of Gross Disease with Radiation Therapy Alone Study and Reference

Number of Patients

Dewer and Duncan149

25

20%

Slater et al150

72

29%

al151

112

45%

Kepka et

Local Control Rate

vascularized tissue transfer can reduce complication rates after preoperative radiation, this randomized trial and our own experiences have not confirmed this finding.125,148

Radiation Therapy Alone The role of radiation therapy in the management of STS has generally been that of controlling residual microscopic disease after conservation surgical resection. In a limited number of cases, however, radiation has been applied as primary therapy for gross disease, with local control reported in up to a third of patients (Table 36-9).149-151 Windeyer and colleagues102 reviewed the experience of patients with fibrosarcoma, including detailed data about their clinical responses to doses of radiation ranging from 60 to 85 Gy. Among the 22 patients given radiation therapy alone for unresectable gross primary tumors or surgical recurrences, the clinical complete response rate was 64%. An additional five patients (23%) had a partial response, and in three the disease remained unchanged. Among the 14 patients achieving a complete response, 6 developed subsequent recurrence. Slater and colleagues150 reviewed 72 patients treated with photons alone (57 patients) or photons combined with neutrons (15 patients). The overall 5-year local control rate was 29%, although control was obtained in 11 (42%) of the 27 patients treated with doses of 65 Gy or higher. The median duration of local control was 21 months for the group given doses higher than 65 Gy and 13 months for the group given lower doses. Similar results were reported by Kepka and colleagues in their update of the Massachusetts General Hospital experience.151,152 In that study of 112 patients with STS who underwent radiation therapy for gross disease, the 5-year actuarial local control rate was 45% and was associated with tumor size and radiation dose. The local control rates according to tumor size were 51% for tumor smaller than 5 cm, 45% for tumors 5 to 10 cm, and 9% for tumor larger than 10 cm. In terms of radiation dose, the local control rates were 22% when the dose was less than 63 Gy and 60% when the dose was 63 Gy or more. Further analysis revealed significantly more complications when doses higher than 68 Gy were delivered, prompting their recommendation of 68 Gy for the treatment of gross disease. Definitive radiation therapy may have a role for patients for whom the only alternative means of controlling the primary tumor would be radical amputation.

Chemotherapy Although the need for effective systemic therapy for patients with high-grade large STSs is evident, the role of adjuvant chemotherapy in the management of localized STS

954

PART 12  Musculoskeletal System

TABLE 36-10.    Trials Comparing Adjuvant Doxorubicin with Observation for Resectable Soft Tissue Sarcomas Study and Reference Intergroup Sarcoma Study Group155-159

N

Median Follow-up Time

Gynecologic Oncology Group164

68%

Observation

—

59%

62%

60 mg/m2 doxorubicin q4wk for 9 cycles

92%

62%

75%

Observation

92%

56%

70%

120 mg/m2 4’-epidoxorubicin for 5 cycles

94%*

50%*

69%*

Observation

83%*

37%*

50%*

90 mg/m2 doxorubicin for 5 cycles

96%

58%

84%

62

Observation

89%

54%

80%

75

mg/m2

60 doxorubicin q3wk for 8 cycles

—

59%

60%

Observation

—

47%

52%

77

53

57

81 *Rates

Overall Survival Rate

67%

40 mo

59 mo

51 University of California, Los Angeles163

Disease-Free Survival Rate

—

54 mo

77 Italian Cooperative Group (Bologna)161,162

mg/m2

Locoregional Control Rate

450 doxorubicin (cumulative dose)

80

88 Scandinavian Sarcoma Group160

Treatment

28 mo

—

Sites All

All

Extremity

Extremity

Uterine

at 4 years.

remains controversial.153,154 Numerous trials have compared doxorubicin chemotherapy alone with observation (Table 36-10)155-164 or have compared combination chemotherapy with observation (Table 36-11).160,161,165-172 The results of these trials are mixed, and the trials have been criticized for having small patient numbers, long accrual times, and high ineligibility rates. The patients ­included often had cancer of mixed histologic subtypes or grades and tumor locations. In an effort to overcome these limitations, Zalupski and colleagues173 performed a meta-analysis of published data and found that disease-free and overall survival rates at 10 years were improved by the use of doxorubicin-based chemotherapy. The analysis was subsequently repeated174 using the raw patient data that had been updated by the individual investigators of each trial. That meta-analysis included 1568 patients from 14 trials, with a median followup time of 9.4 years. Adjuvant chemotherapy was found to significantly improve local relapse-free survival rate (81% versus 75%, P = .016) and recurrence-free survival rate (55% versus 45%, P = .0001). However, the use of adjuvant chemotherapy conferred no improvement in overall survival (54% versus 50%, P = .12). The results of any meta-analysis must be interpreted with caution because they can only be as good as the results of the individual trials included.148 For example, in this large meta-analysis, tumor histologic subtype and grade were reviewed in only 59% and 25% of patients, respectively.174 The

­ uropean Organisation for Research and Treatment of ­Cancer E (EORTC)165 and the Scandinavian Sarcoma Group160 trials, two of the largest, had poor patient compliance with chemotherapy and very high rates of patient ineligibility. Because 58% of patients’ tumors were located in an extremity and because patients with recurrent disease were excluded, it is unclear whether the results can be applied to nonextremity sarcomas or to patients presenting with recurrent disease. The meta-analysis did confirm that local control could be improved through the use of systemic chemotherapy (27% reduction in the risk of local recurrence, absolute benefit 6%). The NCI trial168,169 and the EORTC trial165 had shown improvements in local control resulting from doxorubicin alone and combination chemotherapy, respectively (details of the combination therapy are given in Table 36-11). Within the EORTC trial, however, the benefit was limited to head, neck, and trunk lesions, whereas only extremity lesions were included in the NCI trial. These findings suggest that the combination of chemotherapy and radiation deserves investigation as a strategy to improve local control. The only completed trial to examine modern chemotherapy for STS in an appropriately high-risk cohort of patients was conducted by the Italian Cooperative Group (see Table 36-11).160,161 In that study, 104 patients who had localized extremity or limb girdle, high-grade (3 or 4), deeply seated, and primary tumors larger than 5 cm or recurrent lesions of any size were randomized to ­observation or five

36  The Soft Tissue

955

TABLE 36-11.   Trials Comparing Adjuvant Combination Chemotherapy with Observation for Resectable Soft Tissue Sarcomas

Study and Reference Italian Cooperative Group (Bologna)161,162

Median Follow-up Time

N 53

90 mo

51 EORTC165

145

80 mo

31

20 37 17

8 30 31

*P

16%

46%

56%*

63% 56%

CyVADIC for 9 cycles

97%

87%*

87%*

Observation

82%

54%

37%

VAC-VAdC

90%

55%*

65%

Observation

65%

35%

57%

AC-MTX

97%*

75%*

82%

Observation

86%

54%

60%

AC-MTX

—

77%

68%

Observation

—

49%

58%

29 mo

AC-MTX

—

50%

47%

Observation

—

86%

100%

64.3 mo

VADIC/VAdC

53%

88%

82%

Observation

52%

68%

82%

52.4 mo

120 mo 7.1 yr 35 mo

7 Mayo172

†

43%

14 NCI-Retroperitoneum171

—

66%*

69%

28 NCI-Head & Neck, Breast, and Trunk170

Observation

47%

Observation

23 NCI-Extremity168,169

—

Overall Survival Rate

83%*

28 M. D. Anderson167

Epirubicin, ifosfamide, mesna, G-CSF

Disease-Free Survival Rate

CyVADIC for 8 cycles

172 Fondation Bergonié (Bordeaux)166

Treatment

Locoregional Control Rate

Sites Extremity

All

All

Extremity

All

< .05.

†Reduction

in local recurrence was significant in head, neck, and trunk soft tissue sarcomas only. A, Adriamycin (doxorubicin); Ad, actinomycin D; Cy and C, cyclophosphamide; DIC, dacarbazine; EORTC, European Organization for the Research and Treatment of Cancer; G-CSF, granulocyte colony-stimulating factor; MTX, methotrexate; NCI, National Cancer Institute; V, vincristine.

cycles of epirubicin, ifosfamide, mesna, and granulocyte colony-stimulating factor. After a median follow-up time of 90 months and according to an intent-to-treat analysis, combination chemotherapy did not improve the overall survival rate (P = .07); however, when the seven ­patients in the chemotherapy group who had never received chemotherapy were excluded from the analysis, improvements in overall survival (P = .04) and perhaps in disease-free survival (P = .06) were seen.160,161 Given the uncertainties surrounding the effectiveness of adjuvant chemotherapy, patients with large high-grade sarcomas should be enrolled in established chemotherapy trials. No role has been definitively established for adjuvant chemotherapy in localized STS.

Recurrent Disease The management of relapsed STS is determined by the site of relapse. Isolated local recurrence is managed according to the principles outlined previously. If disease recurs after prior resection alone (without adjuvant radiation) and the lesion is amenable to wide local resection, this treatment

should be combined with adjuvant radiation in virtually all cases. The optimal management strategy for local recurrence in patients treated with prior surgery and radiation is unclear; additional wide resection, radical soft part resection, and amputation have all been used.39,56,138 Results from two early studies suggested that adjuvant brachytherapy might have a role in treating recurrent disease. These studies looked at a total of 66 patients who had isolated local recurrences after surgery and EBRT to doses of 45 to 70 Gy; the results indicated that re-resection plus interstitial irradiation to doses of 45 to 50 Gy achieved local control in 68%143 and 52%142 of patients, with an acceptable incidence of complications. We re-examined our experience with brachytherapy for such patients and found unacceptably high rates of long-term radiation-related complications compared with patients treated with surgery alone.175 The 5-year actuarial local control rate among all 62 previously irradiated patients with isolated STS local recurrence was 51%. Among the 37 patients that were re­ irradiated (33 with low-dose-rate brachytherapy implants), the local control rate was 58%, compared with 39% for

956

PART 12  Musculoskeletal System

25 patients who underwent wide local excision alone (P = .4). However, further analysis showed higher rates of long-term complications among the reirradiated group (80% versus 17% without irradiation, P < .001), suggesting that surgical re-excision alone should be the initial treatment for isolated local recurrences.175 Patients who develop metastatic relapse are generally given chemotherapy, but their long-term survival rate is poor (approximately 10% at 10 years after metastatic relapse39). Nevertheless, resection of metastases (metastasectomy) is appropriate in select situations, especially in patients who have isolated lung metastases, for whom a 3-year survival rate of 40% after resection has been reported.46

A

Special Anatomic Sites At many sites, including the retroperitoneum, head and neck, hand, and foot, anatomic constraints often preclude wide local excision, and the tolerance of normal structures often limits the delivery of appropriate doses of radiation. For STS arising in these locations, local control remains a formidable challenge.

T 100% 95%

Retroperitoneum Retroperitoneal STSs are distinguished by bulky presentation, fixation to vital structures, and local recurrence, even after complete surgical resection (Fig. 36-3). The ­predominant histologic subtype is liposarcoma; leiomyosarcoma is the second most common histologic subtype, and other subtypes have been described.63,171,176 At the time of surgical resection, only 55% of patients are found to have grossly resectable disease, and of these, 80% require the removal of adjacent organs; in about 25%, debulking resection is possible, and in approximately 20%, the tumor is deemed unresectable.63,71,171,176 The extent of surgical resection remains the most important predictor of long-term survival. Patients undergoing complete gross resection have a 10-year survival rate of approximately 45%, compared with an 8% survival rate for patients in whom only debulking is possible.63,176 Patients in whom partial resection is achieved have a 5-year survival rate of approximately 25%, compared with a 9% survival rate among those in whom no resection is done.63 It is unclear whether these results reflect a true benefit of resection or merely pathobiologic features that favor resectability. Even after complete gross resection, local recurrence is common, occurring in up to 90% of patients if they survive to 10 years.63 The use of adjuvant radiation therapy has met with limited success in improving local control. Fein and colleagues116 reviewed the experience of 21 patients treated with surgical resection and radiation. Only one patient had surgical margins that were microscopically negative for disease. The median radiation dose was 54 Gy. The 2-year rates of local control and overall ­survival were 72% and 69%, respectively. Local control seemed to be improved by doses greater than 55.2 Gy. Tepper and colleagues72 similarly reported a modest dose-related response, with improvements in local control seen for doses greater than 60 Gy. Local control may also be improved through the use of an intraoperative radiation therapy boost. In another randomized trial from the NCI,117 35 patients with resected retroperitoneal STS were randomized to undergo 20 Gy as an intraoperative boost in combination with 35 to 40 Gy postoperative EBRT (15

B

cm

FIGURE 36-3.  A 28-year-old woman with neurofibromatosis developed left flank pain in May 1998 and was found to have a large retroperitoneal mass displacing the kidney. Pathologic examination of a CT-guided fine-needle aspiration biopsy specimen confirmed high-grade neurogenic sarcoma. A, Diagnostic CT scan shows an 8 × 8 cm retroperitoneal mass with areas of enhancement and necrosis. B, The simulation and the treatment were done with the patient in the left lateral decubitus position to spare as much bowel as possible. Using lateral parallel-opposed 6-MV photon fields, a tumor dose of 50 Gy was delivered over 5 weeks before surgery, with concurrent doxorubicin chemotherapy. Complete surgical resection was followed by 15 Gy of intraoperative electron beam radiation therapy. The patient was alive with bone metastases in the sacrum at the most recent follow-up.

­ atients) or 50 to 55 Gy postoperative EBRT alone. After p a median follow-up time of 8 years, the overall survival rates were similar, but local control was better in those patients given the intraoperative boost (60% versus 20%, P < .05). However, the overall intra-abdominal relapse rates were similar in the two groups. The incidence of radiation-related peripheral neuropathy was higher in the intraoperative-boost group, but the incidence of radiationrelated enteritis was lower. Our experience with 83 patients with retroperitoneal sarcoma indicates that at a median follow-up time of 47 months, the actuarial overall disease-specific survival, distant metastasis-free survival, and local control rates were 44%, 67% and 40%, respectively.177 Of the 38 patients who died of disease, local disease progression was the sole site of recurrence in 16 and a component of progression in

36  The Soft Tissue

another 11, highlighting the need for improved local therapy. Five patients developed a chronic radiation-related complication (10% at 5 years), and all five had received postoperative EBRT. Efforts to improve survival through the use of adjuvant chemotherapy have been unsuccessful. In a small ­randomized trial reported by the NCI,171 the 2-year actuarial survival rate was worse for eight patients given doxorubicin-based chemotherapy than for seven patients who were only observed (47% versus 100%, P = .06). The initial treatment for all 15 patients was complete surgical resection. Our adjuvant treatment recommendations are based on disease presentation. For patients with biopsy-proven retroperitoneal STS, regardless of grade, we recommend 50.4 Gy preoperative radiation therapy followed by surgical resection. In the preoperative setting, tumor often displaces the bowel out of the field of radiation, allowing treatment to be delivered with little toxicity. If surgical margins are then found to be close or positive, we recommend intraoperative radiation therapy. When surgical resection has already been performed, we discourage the routine use of postoperative radiation unless the tumor is clearly away from dose-limiting structures. Although postoperative doses for patients with extremity tumors are 60 to 68 Gy, these doses for patients with retroperitoneal STS are associated with increased toxic effects.

Head and Neck Among STSs affecting the head and neck, the most commonly reported histologic subtype is MFH.64,118,178 Overall survival rates consistently decreases as tumor grade increases, whereas surgical margins positive for microscopic disease are prognostic for local recurrence.65,178,179 Although head and neck STSs are rare and manifest heterogeneously, local control rates after conservative resection and radiation therapy have consistently been 65% to 75%.65,118,179,180 Adjuvant radiation therapy has a significant role in the management of head and neck STSs in view of the difficulty in achieving resection margins that are free of disease. Adjuvant radiation is particularly important in the management of angiosarcoma arising in the skin of the scalp.181

957

A

B FIGURE 36-4.  A 24-year-old man presented in July 1983 with an 8-year history of a slowly enlarging lesion on the plantar ­aspect of the right foot. Physical examination revealed a 4.5 ×  3 cm mass. Pathologic examination of an excisional biopsy specimen confirmed extraosseous Ewing’s sarcoma. He received two courses of chemotherapy (cyclophosphamide, vincristine, doxorubicin, and dacarbazine) and was then referred for radiation therapy. A, The patient was positioned prone with his foot on a pillow. A single, appositional 12-MeV electron beam delivered 50 Gy over 5 weeks, with a small margin around the primary tumor. An additional 10 Gy was delivered using 9-MeV electrons. The dose was prescribed to the 90% isodose line, and the skin dose was maximized through the use of a wax bolus.  B, After delivery of 50 Gy, significant erythema and blistering were apparent. It resolved completely within 1 month. There was no evidence of disease as of January 1994.

Distal Extremities For STSs arising in the hand, wrist, foot, and ankle, anatomic constraints make it difficult to perform wide local excision and still preserve function. There is an unsubstantiated belief that these anatomic locations tolerate radiation therapy poorly.182 An early experience with surgical management alone for STS of the hand and foot resulted in local failure in only 7 of 50 patients (14%); however, radical or limited amputation was required in 74% to achieve this result.182 Five of the seven local failures occurred in the patients who underwent wide local excision alone. This high amputation rate contrasts with the 24% rate reported by Talbert and colleagues140 after conservative surgery and radiation therapy in 78 patients with wrist, hand, ankle, or foot STSs. Reasons for amputation included salvage after recurrent disease in 13 patients and complications in 6. At a median follow-up time of 7.9 years, the actuarial 5- and 10-year local control rates were 80% and 74%, respectively. The 69% overall survival rate at 10 years compared favorably with that obtained with radical surgery, but a normal

or fairly normal extremity was retained in more than two thirds of patients. Amputation is necessary to treat a small percentage of primary distal extremity STSs, but a conservative approach has been successful in most cases. When irradiating the hand or foot, the radiation oncologist must be aware of the unusual acute radiation reaction that can affect the palm or sole; in our experience, patients often develop an edematous, painful, and erythematous response by the fifth week (Fig. 36-4). Although uncomfortable and alarming, this reaction heals within approximately 1 month, and its development is probably related to the unusual anatomic tethering of skin to underlying tissue in these locations.

Selected Tumor Types The following paragraphs describe the more common variants of STS and those of special interest to radiation ­oncologists.

958

PART 12  Musculoskeletal System

but myxoid MFH has significantly less metastatic potential than the nonmyxoid variant.

Synovial Sarcoma

FIGURE 36-5.  Typical microscopic appearance of pleomorphic malignant fibrous histiocytoma (×100). It is often characterized by a storiform pattern and a mixture of bizarre-appearing, plump fibroblastic cells, giant cells containing multiple hyperchromatic irregular nuclei, and increased mitotic activity. (Courtesy A. K. Raymond, M. D. Anderson Cancer Center, Houston, TX)

Malignant Fibrous Histiocytoma MFH is the most commonly diagnosed STS. The term represents a pleomorphic high-grade STS characterized microscopically by plump spindle cells arranged in a cartwheel or storiform pattern around slitlike vessels (Fig. 36-5).5,183 Although numerous histologic subtypes of MFH have been described, only myxoid change, characteristic of myxoid MFH, seems to be of prognostic value. Myxoid MFH is categorized as intermediate grade; other variants (storiform-pleomorphic, giant-cell, and inflammatory) are considered to be high-grade disease.22 In our experience with patients with MFH treated with conservative surgery and radiation showed that among 462 such patients, the overall survival rates at 10, 15, and 20 years were 60%, 47%, and 39%, respectively.60 These rates are similar to rates reported by others.184,185 Determinants of metastatic relapse and overall survival were histologic variant (myxoid versus nonmyxoid) and tumor size. The 10-year overall survival and metastatic relapse rates were 79% and 12%, respectively, for myxoid tumors, compared with 55% and 44%, respectively, for the nonmyxoid tumors. The 10-year metastasis rates for large (>5 cm) and small myxoid tumors were 17% and 6%, respectively, compared with 40% and 23%, respectively, for nonmyxoid tumors.57 Adverse determinants of local control included margins microscopically positive for disease (39% of patients with positive margins experienced local failure versus 17% of others, P = .01), patient age (32% of patients over the age of 65 experienced local failure versus 16% of others, P < .001), and head and neck or deep trunk primary site (45% of patients with a head and neck or deep trunk primary experienced local failure versus 19% of others, P < .001). Tumor size (≤5 cm versus >5 cm) and histologic variant (myxoid versus nonmyxoid) were not associated with the likelihood of local control. It is important to distinguish the myxoid variant from its much more common nonmyxoid counterpart. Both subtypes are equally aggressive locally,

Synovial sarcoma is a high-grade STS characterized by a predilection for the para-articular regions of the extremities in adolescents and young adults. It often manifests as a painful, calcified mass in close association with tendon sheaths, bursae, and joint capsules, although it does not arise from synovial structures.5 Microscopically, distinct epithelial and spindle cell components can be seen in various proportions. This pattern creates a morphologic spectrum between two histologic variants, biphasic and monophasic synovial sarcoma. Further subclassification into biphasic, monophasic fibrous, monophasic epithelial, and poorly differentiated synovial sarcoma has been described,5 although the prognostic value of these distinctions has been ­questioned.22 In our experience with 150 patients with synovial sarcoma treated with conservative resection and radiation,186 the 5-, 10-, and 15-year overall survival rates were 77%, 58%, and 51%, respectively. The 10-year metastasis rate was 44%, with mortality resulting almost entirely from metastatic relapse. Multivariate analysis showed that small ­tumor size and young patient age (≤20 years) independently predicted favorable overall survival. As has been described by others,187 the survival probability correlated particularly closely with tumor size. The distinction of biphasic versus monophasic histology was not a significant determinant of outcome. Ten-year metastasis rates showed a clear relationship to tumor size (for lesions ≤2 cm, the 10-year metastasis rate was 0%; for tumors >2 but ≤5 cm in diameter, the rate was 36%; for those > 5 but ≤10 cm across, the rate was 51%; and for those > 10 cm in diameter, the rate was 93%; P < .0001). Adjuvant chemotherapy was not found to have a favorable effect on overall survival; however, the high metastatic propensity highlights the need for effective systemic therapy. Our 10-year local recurrence rate of 18% is comparable to that reported by others.188

Liposarcoma The pathobiologic characteristics of liposarcoma range from low-grade, nonmetastasizing lesions to highly aggressive ones with a propensity for local and metastatic recurrence. Behavior can be predicted by the microscopic appearance of the primary tumor and by estimating the degree of differentiation.189 Enzinger and Weiss5 suggested a five-tiered classification system; that system was subsequently modified by Evans190 to three histologic variants with distinct clinical behaviors: (a) the low-grade, well-differentiated liposarcoma (i.e., atypical lipomatous tumor); (b) the intermediate-grade myxoid liposarcoma; and (c) the high-grade pleomorphic liposarcoma. A fourth variant known as dedifferentiated liposarcoma represents the coexistence of well-differentiated liposarcoma with a poorly differentiated, nonlipogenic spindle cell element (usually MFH) within the same neoplasm. This variant usually occurs in the setting of multiply recurrent, well-differentiated liposarcoma and is included in the high-grade category because of its metastatic potential.190 Liposarcoma commonly occurs in the thigh and retroperitoneum, although the tumor can arise anywhere. Certain histologic variants seem to have a predilection for specific anatomic locations. Myxoid liposarcoma and ­pleomorphic

36  The Soft Tissue

sarcoma arise within the upper thigh, whereas well-differentiated and dedifferentiated liposarcoma arise within the retroperitoneum.40,190 Myxoid liposarcoma has an unusual proclivity for metastasis to the soft tissue (particularly the retroperitoneum) rather than to the lung.40,190,191 This histologic variant is also uniformly associated with the t(12;16) chromosome translocation.193 Liposarcoma is one of the few STSs that has not been reported as a radiationinduced sarcoma.5 A review of our experience with conservative surgery and radiation therapy for 207 patients with liposarcoma193 confirmed its wide spectrum of clinical behavior. The overall survival rates at 10, 15, and 20 years were 72%, 66%, and 56%, respectively, but survival was highly dependent on the histologic variant. The actuarial 10-year overall survival rates for the well-differentiated, myxoid, pleomorphic, and dedifferentiated variants were 93%, 77%, 49%, and 100% respectively; the rates of metastasis in these same groups were 0%, 24%, 33%, and 10% respectively. In multivariate analysis, factors predicting decreased disease-specific survival were pleomorphic histologic variant (versus myxoid) and tumor size larger than 5 cm. In contrast to other STSs, for which local control seems to be independent of histologic variant, the subcategory of liposarcoma type seems to predict the probability of local control.40,191 We found a local recurrence rate of 28% for patients who had ­pleomorphic liposarcomas, in contrast to 3% for those who had the myxoid variant.57

Desmoid Tumor Desmoid tumor, also known as aggressive fibromatosis, is a locally invasive tumor with no metastatic potential. Microscopically, these tumors are characterized by ­spindle-shaped cells in a collagenous matrix lacking the pleomorphic, atypical, or hyperchromatic nuclei of malignancy (Fig. 36-6).5 One of the major determinants of local control after surgical management is the status of the resection margins.194-197 In a review of 118 patients from M. D. Anderson,198 a local recurrence rate of 54% was attained by surgery alone when the microscopic margins contained cancer cells; the local recurrence rate dropped to 27% if the resection margins were negative for disease (P = .003). In a second group of 46 patients whose desmoid tumors were managed with gross total resection and radiation therapy, the prognostic significance of microscopically positive margins was lost. Postsurgical treatment recommendations are observation in cases of completely resected disease and radiation to 50 Gy in cases of microscopic residual disease. In certain situations in which recurrence may compromise functional and cosmetic outcome, postoperative radiation therapy is recommended regardless of the margin status. In a separate analysis of 23 patients199 treated for gross disease with radiation alone, the rate of local control was 69%, comparable to that reported by others.197-201 Our findings suggested a modest dose-related response for doses greater than 50 Gy. Beyond doses of 56 Gy, the incidence of complications increased significantly, prompting our final recommendation of 56 Gy for control of gross disease.199

Dermatofibrosarcoma Protuberans Dermatofibrosarcoma protuberans is an uncommon, lowgrade, cutaneous STS characterized by aggressive local behavior but little metastatic potential. Microscopically, the

959

A

B

C FIGURE 36-6.  A 38-year-old man presented in August 1995 with a right posterior shoulder mass that had enlarged slowly over 5 years. A, CT (shown) and MRI revealed a 7.5 × 7.5 × 3.5 cm mass arising from the musculature of the right upper back, with invasion of the supraspinatus muscle and subcutaneous fatty tissue. No definite bone abnormality was detected.  B, Grossly, the tumor abutted the scapula, and resection ­required near-total scapulectomy. The tumor had a firm consistency and a coarsely trabecular appearance when sectioned.  C, Pathologic examination revealed elongated, slender, spindleshaped cells of uniform appearance and abundant surrounding collagen consistent with a desmoid tumor (×200). Microscopically, no bone invasion was seen, but margins were close, and postoperative radiation therapy was recommended. Oblique, parallel-opposed 6-MV photon fields were used to deliver  50 Gy over 5 weeks. There was no evidence of disease as of September 1999. (C, Courtesy A.  K. Raymond, M. D. Anderson ­Cancer Center, Houston, TX)

960

PART 12  Musculoskeletal System

tumor can be recognized by slender fibroblasts arranged in a storiform pattern.5 Similar to the desmoid type, dermatofibrosarcoma protuberans tumors are locally infiltrative, with control determined primarily by the adequacy of the surgical resection. Wide local resection with at least a 3-cm margin, including the underlying fascia, has been the recommended treatment.202 Local recurrence rates of 10% to 20%, however, can still be expected.203-205 Mohs micrographic surgical excision has been associated with a less than 10% rate of recurrence,206 but it is labor intensive, ­results in significant tissue defects, and is difficult to perform on larger or recurrent lesions. More limited surgical resection combined with radiation therapy can be used when complete surgical resection of the lesion would be difficult. Ballo and colleagues206 reported only one local recurrence among 19 patients, 10 of whom had already experienced at least one prior recurrence and 6 of whom had positive microscopic margins. Suit and colleagues207 reported three local recurrences among 15 patients, 5 of whom had experienced prior relapse and 12 of whom had positive microscopic margins. Current recommendations are complete resection plus observation if margins are negative and radiation therapy with 50 to 60 Gy if the margins are close (<2 to 3 cm) or positive for disease. Adjuvant radiation therapy may be indicated when complete resection would entail significant cosmetic or functional deficits.

Hemangiopericytoma Hemangiopericytoma, otherwise known as solitary fibrous tumor, is an uncommon STS derived from vascular pericytes5 that is of interest to radiation oncologists by virtue of its radiosensitivity. These tumors most commonly arise in the pelvis, thigh, or head and neck.208-210 They are highly vascular, like glomus tumors. Histologic grading is difficult, but a high mitosis rate seems to correlate with metastatic potential. Highly mitotic tumors are associated with a 5-year survival rate of less than 50%.208,209 The radiosensitivity of hemangiopericytomas is indicated by the achievement of local control in approximately 50% of patients irradiated for gross tumor to doses of 50 to 65 Gy.211,212 These same reports show a local control rate exceeding 60% when radiation is given as an adjuvant to resection. Radiation is indicated for unresectable disease and as an adjuvant for high-grade tumors. In the adjuvant setting, the high vascularity of these tumors suggests that preoperative radiation may be more effective than postoperative radiation, but this theory has not been tested.

REFERENCES    1. American Cancer Society. Cancer Facts & Figures, 2007. Atlanta: American Cancer Society, 2007.    2. Pollock RE, Karnell LH, Menck HR, et al. The National Cancer Data Base report on soft tissue sarcomas. Cancer 1996;78: 2247-2257.    3. Lawrence W, Donegan WL, Natarajan N, et al. Adult soft tissue sarcomas: a pattern of care survey of the American College of Surgeons. Ann Surg 1987;205:349-359.    4. Rydholm A, Berg NO, Gullberg B, et al. Epidemiology of softtissue sarcoma in the locomotor system. Acta Path Microbiol Immunol Scand [A] 1984;92:363-374.    5. Enzinger FM, Weiss SW. Soft Tissue Tumors, 3rd ed. St Louis: Mosby, 1995.

   6. Inoshita T, Youngberg GA. Malignant fibrous histiocytoma arising in previous surgical sites: a report of two cases. Cancer 1984;53:176-183.    7. Jennings TA, Peterson L, Axiotis CA, et al. Angiosarcoma associated with foreign body material: a report of three cases. Cancer 1988;62:2436-2444.    8. Lindeman G, McKay MJ, Taubman KL, et al. Malignant fibrous histiocytoma developing in bone 44 years after shrapnel trauma. Cancer 1990;66:2229-2232.    9. Stewart FW, Treves N. Lymphangiosarcoma in postmastectomy lymphedema: a report of six cases in elephantiasis chirurgica. Cancer 1948;1:64-81.   10. Cahan WG, Woodard HQ, Higinbotham NL, et al. Sarcoma arising in irradiated bone: report of eleven cases. Cancer 1948;1: 3-29.   11. Robinson E, Neugut AI, Wylie P. Clinical aspects of postirradiation sarcomas. J Natl Cancer Inst 1988;80:233-240.   12. Laskin WB, Silverman TA, Enzinger FM. Postirradiation soft tissue sarcomas: an analysis of 53 cases. Cancer 1988;62:23302340.   13. Patel SG, See ACH, Williamson PA, et al. Radiation induced sarcoma of the head and neck. Head Neck 1999;21:346-354.   14. Hardell L, Eriksson M. The association between soft tissue sarcomas and exposure to phenoxyacetic acids: a new case-referent study. Cancer 1988;62:652-656.   15. Wingren G, Fredrikson M, Brage HN, et al. Soft tissue sarcoma and occupational exposure. Cancer 1990;66:806-811.   16. Dich J, Zahm SH, Hanberg A, et al. Pesticides and cancer. Cancer Causes Control 1997;8:420-443.   17. Hosoi K. Multiple neurofibromatosis (von Recklinghausen’s disease) with special reference to malignant transformation. Arch Surg 1931;22:258-281.   18. Wong FL, Boice JD, Abramson DH, et al. Cancer incidence after retinoblastoma: radiation dose and sarcoma risk. JAMA 1997;278:1262-1267.   19. Li FP, Fraumeni JF. Soft-tissue sarcomas, breast cancer, and other neoplasms: a familial syndrome? Ann Intern Med 1969;71:747-752.   20. Li FP, Fraumeni JF, Mulvihill JJ, et al. A cancer syndrome in twenty-four kindreds. Cancer Res 1988;48:5358-5362.   21. Fisher C. Current aspects of the pathology of soft tissue sarcomas. Semin Radiat Oncol 1999;9:315-327.   22. Evans HL. Classification and grading of soft-tissue sarcomas. Hematol Oncol Clin North Am 1995;9:653-656.   23. Recommendations for the reporting of soft tissue sarcomas. Mod Pathol 1998;11:1257-1261.   24. Hajdu SI. Pathology of Soft Tumors. Philadelphia: Lea & Febiger, 1979.   25. Mazeron J, Suit HD. Lymph nodes as sites of metastases from sarcomas of soft tissue. Cancer 1987;60:1800-1808.   26. Hashimoto H, Daimaru Y, Takeshita S, et al. Prognostic significance of histologic parameters of soft tissue sarcomas. Cancer 1992;70:2816-2822.   27. Trojani M, Contesso G, Coindre JM, et al. Soft-tissue sarcomas of adults: study of pathological prognostic variable and definition of a histopathological grading system. Int J Cancer 1984;33:37-42.   28. Coindre JM, Trojani M, Contesso G, et al. Reproducibility of a histopathologic grading system for adult soft tissue sarcoma. Cancer 1986;58:306-309.   29. Leyvraz S, Costa J. Issues in the pathology of sarcomas of the soft tissue and bone. Semin Oncol 1989;16:273-280.   30. Broders AC, Hargrave R, Meyerding HW. Pathological features of soft tissue fibrosarcoma with special reference to the grading of its malignancy. Surg Gynecol Obstet 1939;69:267-280.   31. Costa J, Wesley RA, Glatstein E, et al. The grading of soft tissue sarcomas: results of a clinicohistopathologic correlation in a series of 163 cases. Cancer 1984;53:530-541.   32. Van Unnik JAM, Coindre JM, Contesso C, et al. Grading of soft tissue sarcomas: experience of the EORTC Soft Tissue and Bone Sarcoma Group. Eur J Cancer 1993;29A:2089-2093.   33. Myhre-Jensen O, Kaae S, Hjollund Madsen E, et al. Histological grading in soft-tissue tumours. Acta Path Microbiol Immunol Scand [A] 1983;91:145-150.

36  The Soft Tissue   34. Angervall L, Kindblom L, Rydholm A, et al. The diagnosis and prognosis of soft tissue tumors. Semin Diagn Pathol 1986;3: 240-258.   35. Greene FL, Page DL, Fleming ID, et al (eds). soft tissue sarcoma. In AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002, pp 193-199.   36. Markhede G, Angervall L, Stener B. A multivariate analysis of the prognosis after surgical treatment of malignant soft-tissue tumors. Cancer 1982;49:1721-1733.   37. Guillou L, Coindre JM, Bonichon F, et al. Comparative study of the National Cancer Institute and French Federation of Cancer Centers sarcoma group grading systems in a population of 410 adult patients with soft tissue sarcoma. J Clin Oncol 1997;15:350-362.   38. Barkley HT, Martin RG, Romsdahl MM, et al. Treatment of soft tissue sarcomas by preoperative irradiation and conservative surgical resection. Int J Radiat Oncol Biol Phys 1988;14:693-699.   39. Zagars GK, Mullen JR, Pollack A. Malignant fibrous histiocytoma: outcome and prognostic factors following conservative surgery and radiotherapy. Int J Radiat Oncol Biol Phys 1996;34: 683-694.   40. Zagars GK, Goswitz MS, Pollack A. Liposarcoma: outcome and prognostic factors following conservation surgery and radiation therapy. Int J Radiat Oncol Biol Phys 1996;36:311-319.   41. Mullen JR, Zagars GK. Synovial sarcoma outcome following conservation surgery and radiotherapy. Radiother Oncol 1994;33:23-30.   42. Pollock RE. Molecular determinants of soft tissue sarcoma proliferation. Semin Surg Oncol 1994;10:315-322.   43. El-Naggar AK, Ayala AG, Abdul-Karim FW, et al. Synovial sarcoma: a DNA flow cytometry study. Cancer 1990;65: 2295-2300.   44. Pisters PWT, Pollock RE. Staging and prognostic factors in soft tissue sarcoma. Semin Radiat Oncol 1999;9:307-314.   45. Abbas JS, Holyoke ED, Moore R, et al. The surgical treatment and outcome of soft-tissue sarcoma. Arch Surg 1981;116:765-769.   46. Potter DA, Glenn J, Kinsella T, et al. Patterns of recurrence in patients with high-grade soft-tissue sarcomas. J Clin Oncol 1985;3:353-366.   47. Torosian MH, Friedrich C, Godbold J, et al. Soft-tissue sarcoma: initial characteristics and prognostic factors in patients with and without metastatic disease. Semin Surg Oncol 1988;4:13-19.   48. Enneking WF, Spanier SS, Malawer MM. The effect of the anatomic setting on the results of surgical procedures for soft parts sarcoma of the thigh. Cancer 1981;47:1005-1022.   49. Hunt RT, Morgan MC, Ackerman LV. Principles in the management of extra-abdominal desmoids. Cancer 1960;20:825-836.   50. Batsakis JG. Epithelioid sarcoma. Ann Otol Rhinol Laryngol 1989;98:659-660.   51. Rosai J, Sumner HW, Kostianovsky M, et al. Angiosarcoma of the skin: a clinicopathologic and fine structural study. Hum Pathol 1976;7:83-109.   52. Fong Y, Coit DG, Woodruff JM, 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;217:72-77.   53. Rydholm A, Rööser B. Surgical margins for soft-tissue sarcoma. J Bone Joint Surg Am 1987;69:1074-1078.   54. Dewer JA, Duncan W. A retrospective study of the role of radiotherapy in treatment of soft tissue sarcoma. Clin Radiol 1985;36:629-632.   55. Cantin J, McNeer GP, Chu FC, et al. The problem of local recurrence after treatment of soft tissue sarcoma. Ann Surg 1968;168:47-53.   56. Lindberg RD, Martin RG, Romsdahl MM, et al. Conservative surgery and postoperative radiotherapy in 300 adults with softtissue sarcomas. Cancer 1981;47:2391-2397.   57. Zagars GK, Ballo MT, Pisters PW, et al. Prognostic factors for ­patients with localized soft tissue sarcoma treated with conservation surgery and radiation therapy: an analysis of 1225 patients. Cancer 2003;97:2530-2543.

961

  58. Tsujimoto M, Aozasa D, Ueda K, et al. Multivariate analysis for histologic prognostic factors in soft tissue sarcomas. Cancer 1988;62:994-998.   59. Collin C, Godbold J, Hajdu S, et al. Localized extremity soft tissue sarcomas: an analysis of factors affecting survival. J Clin Oncol 1987;5:601-612.   60. Pisters PWT, Leung DHY, Woodruff J, et al. Analysis of prognostic factors in 1041 patients with localized soft tissue sarcomas of the extremities. J Clin Oncol 1996;14:1679-1689.   61. Mundt AJ, Awan A, Sibley GS, et al. Conservative surgery and adjuvant radiation therapy in the management of adult soft ­tissue sarcoma of the extremities: clinical and radiobiological ­results. Int J Radiat Oncol Biol Phys 1995;32:977-985.   62. Potter DA, Kinsella T, Glatstein E, et al. High-grade soft tissue sarcomas of the extremities. Cancer 1986;58:190-205.   63. Storm FK, Eilber FR, Mirra J, et al. Retroperitoneal sarcomas: a reappraisal of treatment. J Surg Oncol 1981;17:1-7.   64. Willers H, Hug EB, Spiro IJ, et al. Adult soft tissue sarcomas of the head and neck treated by radiation and surgery or radiation alone: patterns of failure and prognostic factors. Int J Radiat ­Oncol Biol Phys 1995;33:585-593.   65. Tran LM, Mark R, Meier R, et al. Sarcomas of the head and neck. Cancer 1992;70:169-177.   66. Singer S, Corson JM, Demetri GD, et al. Prognostic factors ­predictive of survival for truncal and retroperitoneal soft-tissue sarcoma. Ann Surg 1995;221:185-195.   67. Berlin O, Stener B, Angervall L, et al. Surgery for soft tissue sarcoma in the extremities. Acta Orthop Scand 1990;61: 475-486.   68. Bell RS, O’Sullivan B, Liu FF, et al. The surgical margin in softtissue sarcoma. J Bone Joint Surg Am 1989;71:370-375.   69. Herbert SH, Corn BW, Solin LJ, et al. Limb-preserving treatment for soft tissue sarcomas of the extremities: the significance of surgical margins. Cancer 1993;72:1230-1238.   70. Dinges S, Budach V, Budach W, et al. Local recurrences of soft tissue sarcomas in adults: a retrospective analysis of prognostic factors in 102 cases after surgery and radiation therapy. Eur J Cancer 1994;30A:1636-1642.   71. Karakousis CP, Velez AF, Emrich LJ. Management of retroperitoneal sarcomas and patient survival. Am J Surg 1985;150: 376-380.   72. Tepper JE, Suit HD, Wood WC, et al. Radiation therapy of retroperitoneal soft tissue sarcomas. Int J Radiat Oncol Biol Phys 1984;10:825-830.   73. Gaynor JJ, Tan CC, Casper ES, et al. Refinement of clinicopathologic staging for localized soft tissue sarcoma of the extremity: a study of 423 adults. J Clin Oncol 1992;10:1317-1329.   74. Fleming JB, Berman RB, Cheng S, et al. Long-term outcome of patients with American Joint Committee on Cancer stage IIB extremity soft tissue sarcomas. J Clin Oncol 1999;17:27722780.   75. Rosenberg SA, Glatstein EJ. Perspectives on the role of surgery and radiation therapy in treatment of soft tissue sarcomas of the extremities. Semin Oncol 1981;8:190-200.   76. Guiliano AE, Eilber FR. The rationale for planned reoperation after unplanned total excision of soft-tissue sarcomas. J Clin ­Oncol 1985;3:1344-1348.   77. Karakousis CP, Driscoll DL. Treatment and local control of primary extremity soft tissue sarcomas. J Surg Oncol 1999;71: 155-161.   78. Romsdahl MM. Open biopsy of sarcomas. In Ryan JR, Baker LO, (eds): Recent Concepts in Sarcoma Treatment. Dordrecht: Kluwer Academic, 1987, pp 69-75.   79. Gustafson P, Dreinhöfer KE, Rydholm A. Soft tissue sarcoma should be treated at a tumor center: a comparison of quality of surgery in 375 patients. Acta Orthop Scand 1994;65:47-50.   80. Rööser B, Rydholm A, Alvegård T. Centralization of soft tissue sarcoma: status in Sweden in 1982. Acta Orthop Scand 1987;58:641-644.   81. Ball ABS, Fisher C, Pittam M, et al. Diagnosis of soft tissue ­tumours by Tru-cut biopsy. Br J Surg 1990;77:756-758.

962

PART 12  Musculoskeletal System

  82. Ayala AG, Ro JY, Fanning CV, et al. Core needle biopsy and fine needle aspiration in the diagnosis of bone and soft-tissue lesions. Hematol Oncol Clin North Am 1995;9:633-651.   83. Rydholm A, Gustafson P, Rööser B, et al. Limb-sparing surgery without radiotherapy based on anatomic location of soft tissue sarcoma. J Clin Oncol 1991;9:1757-1765.   84. Manaster BJ. Musculoskeletal oncologic imaging. Int J Radiat Oncol Biol Phys 1991;21:1643-1651.   85. Bland KI, McCoy DM, Kinard RE, et al. Application of magnetic resonance imaging and computer tomography as an adjunct to the surgical management of soft tissue sarcomas. Ann Surg 1987;205:473-481.   86. Chang AE, Matory YL, Dwyer AJ, et al. Magnetic resonance imaging versus computed tomography in the evaluation of soft tissue tumors of the extremities. Ann Surg 1987;205:340-348.   87. Petasnick JP, Turner DA, Charter JR, et al. Soft-tissue masses of the locomotor system: comparison of MR imaging with CT. Radiology 1986;160:125-133.   88. Kransdorf MJ, Moser RP, Jelinek J, et al. Soft tissue masses: diagnosis using MR imaging. AJR Am J Roentgenol 1989;153:541-547.   89. Lucas JD, O’Doherty MJ, Cronin BF, et al. Prospective evaluation of soft tissue masses and sarcomas using fluorodeoxyglucose positron emission tomography. Br J Surg 1999;86:550-556.   90. Nishizawa K, Okunieff P, Elmaleh D, et al. Blood flow of human soft tissue sarcomas measured by thallium-201 scanning: prediction of tumor response to radiation. Int J Radiat Oncol Biol Phys 1991;20:593-597.   91. Ramanna L, Waxman A, Binney G, et al. Thallium-201 scintigraphy in bone sarcoma: comparison with gallium-67 and technetium-MDP in the evaluation of chemotherapeutic response. J Nucl Med 1990;31:567-572.   92. Pass HI, Dwyer A, Makuch R, et al. Detection of pulmonary metastases in patients with osteogenic and soft tissue sarcomas: the superiority of CT scans compared with conventional tomograms using dynamic analysis. J Clin Oncol 1985;3:1261-1265.   93. Lieberman PH, Brennan MF, Kimmel M, et al. Alveolar softpart sarcoma: a clinicopathologic study of half a century. Cancer 1989;63:63-70.   94. Morton DL, Ollila DW. Critical review of the sentinel node ­hypothesis. Surgery 1999;126:815-819.   95. Gerner RE, Moore GE, Pickren JW. Soft tissue sarcomas. Ann Surg 1975;181:803-808.   96. Leibel SA, Tranbaugh RF, Wara WM, et al. Soft tissue sarcomas of the extremities: survival and patterns of failure with conservative surgery and postoperative irradiation compared to surgery alone. Cancer 1982;50:1076-1083.   97. Bowden L, Booher RJ. The principles and techniques of resection of soft parts for sarcoma. Surgery 1958;44:963-977.   98. Simon MA, Enneking WF. The management of soft-tissue sarcomas of the extremities. J Bone Joint Surg Am 1976;58:317327.   99. Shiu MH, Castro EB, Hajdu SI, et al. Surgical treatment of 297 soft tissue sarcomas of the lower extremity. Ann Surg 1975;182:597-602. 100. Pao WJ, Pilepich MV. Postoperative radiotherapy in the treatment of extremity soft tissue sarcomas. Int J Radiat Oncol Biol Phys 1990;19:907-911. 101. Geer RJ, Woodruff J, Casper ES, et al. Management of small soft-tissue sarcoma of the extremity in adults. Arch Surg 1992;127:1285-1289. 102. Windeyer B, Dische S, Mansfield CM. The place of radiotherapy in the management of fibrosarcoma of the soft tissues. Clin Radiol 1966;17:32-40. 103. Suit HD, Russell WO, Martin RG. Management of patients with sarcoma of soft tissue in an extremity. Cancer 1973;31: 1247–1255. 104. Keus RB, Rutgers EJT, Ho GH, et al. Limb-sparing therapy of extremity soft tissue sarcomas: treatment outcome and longterm functional results. Eur J Cancer 1994;30A:1459-1463. 105. Brandt TA, Parsons JT, Marcus RB, et al. Preoperative irradiation for soft tissue sarcomas of the trunk and extremities in adults. Int J Radiat Oncol Biol Phys 1990;19:899-906.

106. Rosenberg SA, Tepper J, Glatstein E, et al. The treatment of soft-tissue sarcomas of the extremities: prospective randomized evaluations of (1) limb-sparing surgery plus radiation therapy compared with amputation and (2) the role of adjuvant chemotherapy. Ann Surg 1982;196:305-315. 107. Pisters PWT, Harrison LB, Leung DHY, et al. Long-term results of a prospective randomized trial of adjuvant brachytherapy in soft tissue sarcoma. J Clin Oncol 1996;14:859-868. 108. Harrison LB, Franzese F, Gaynor JJ, et al. Long-term results of a prospective randomized trial of adjuvant brachytherapy in the management of completely resected soft tissue sarcomas of the extremity and superficial trunk. Int J Radiat Oncol Biol Phys 1993;27:259-265. 109. Yang JC, Chang AE, Baker AR, 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. 110. Abbatucci JS, Boulier N, De Ranieri J, et al. Local control and survival in soft tissue sarcomas of the limbs, trunk walls and head and neck: a study of 113 cases. Int J Radiat Oncol Biol Phys 1986;12:579-586. 111. Suit HD, Spiro IJ. The role of radiation in patients with soft tissue sarcomas. Cancer Control 1994;1:592-598. 112. Fein DA, Lee WR, Lanciano RM, et al. Management of extremity soft tissue sarcomas with limb-sparing surgery and postoperative irradiation: do total dose, overall treatment time, and the surgery-radiotherapy interval impact on local control? Int J Radiat Oncol Biol Phys 1995;32:969-976. 113. Cheng EY, Dusenbery KE, Winters MR, et al. Soft tissue sarcomas: preoperative versus postoperative radiotherapy. J Surg Oncol 1996;61:90-99. 114. Zagars GK, Ballo MT. Significance of dose in postoperative ­radiotherapy for soft tissue sarcoma. Int J Radiat Oncol Biol Phys 2003;56:473-481. 115. Tanabe KK, Pollock RE, Ellis LM, et al. Influence of surgical margins on outcome in patients with preoperatively irradiated extremity soft tissue sarcomas. Cancer 1994;73:1652-1659. 116. Fein DA, Corn BW, Lanciano RM, et al. Management of retroperitoneal sarcomas: does dose escalation impact on locoregional control? Int J Radiat Oncol Biol Phys 1995;31:129-134. 117. Sindelar WF, Kinsella TJ, Chen PW, et al. Intraoperative radiotherapy in retroperitoneal sarcomas: final results of a prospective, randomized, clinical trial. Arch Surg 1993;128:402-410. 118. Krause DH, Dubner S, Harrison LB, et al. Prognostic factors for recurrence and survival in head and neck soft tissue sarcomas. Cancer 1994;74:697-702. 119. Robinson M, Barr L, Fisher C, et al. Treatment of soft tissue sarcomas with surgery and radiotherapy. Radiother Oncol 1990;18:221-233. 120. Nielsen OS, Cummings B, O’Sullivan B, et al. Preoperative and postoperative irradiation of soft tissue sarcoma: effect of radiation field size. Int J Radiat Oncol Biol Phys 1991;21: 1595-1599. 121. Fletcher GH. The evolution of the basic concepts underlying the practice of radiotherapy from 1949 to 1977. Radiology 1978;127:3-19. 122. Atkinson L, Garvan JM, Newton NC. Behavior and management of soft connective tissue sarcomas. Cancer 1963;16:1552-1562. 123. Suit HD, Mankin HJ, Wood WC, et al. Preoperative, intraoperative, and postoperative radiation in the treatment of primary soft tissue sarcoma. Cancer 1985;55:2659-2667. 124. Zagars GK, Ballo MT, Pisters PW, 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;56:482-488. 125. O’Sullivan B, Davis AM, Turcotte R, et al. Preoperative versus postoperative radiotherapy in soft-tissue sarcoma of the limbs: a randomized trial. Lancet 2002;359:2235-2241. 126. Davis AM, O’Sullivan B, Turcotte R, et al. Late radiation morbidity following randomization to preoperative versus postoperative radiotherapy in extremity soft tissue sarcoma. Radiother Oncol 2005;75:48-53.

36  The Soft Tissue 127. Davis AM, O’Sullivan B, Turcotte R, et al. Function and health status outcomes in a randomized trial comparing preoperative and postoperative radiotherapy in extremity soft tissue sarcoma. J Clin Oncol 2002;20:4472-4477. 128. Willett CG, Schiller AL, Suit HD, et al. The histologic response of soft tissue sarcoma to radiation therapy. Cancer 1987;60:15001504. 129. Janjan NA, Yasko AW, Reece GP, et al. Comparison of ­charges related to radiotherapy for soft-tissue sarcomas treated by ­preoperative external-beam irradiation versus interstitial implantation. Ann Surg Oncol 1994;1:415-422. 130. Habrand JL, Gerbaulet A, Pejovic MH, et al. Twenty years experience of interstitial iridium brachytherapy in the management of soft tissue sarcomas. Int J Radiat Oncol Biol Phys 1991;20: 405-411. 131. Chaudhary AJ, Laskar S, Badhwar R. Interstitial brachytherapy in soft tissue sarcomas: the Tata Memorial Hospital experience. Strahlenther Onkol 1998;174:522-528. 132. Ormsby MV, Hilaris BS, Nori D, et al. Wound complications of adjuvant radiation therapy in patients with soft-tissue sarcomas. Ann Surg 1989;210:93-99. 132a. Ballo MT, Zagars GK. Radiation therapy for soft tissue sar­coma. Surg Oncol Clin N Am 2003;12:449-467. 133. Lewis JJ, Leung D, Heslin M, et al. Association of local recurrence with subsequent survival in extremity soft tissue sarcoma. J Clin Oncol 1997;15:646-652. 134. Emrich LJ, Ruka W, Driscoll DL, et al. The effect of local recurrence on survival time in adult high-grade soft tissue sarcomas. J Clin Epidemiol 1989;42:105-110. 135. Stotter AT, A’Hern RP, Fisher C, et al. The influence of local recurrence of extremity soft tissue sarcoma on metastasis and survival. Cancer 1990;65:1119-1129. 136. Gustafson P, Rööser B, Rydholm A. Is local recurrence of ­minor importance for metastases in soft tissue sarcoma? Cancer 1991;67:2083-2086. 137. Heslin MJ, Woodruff J, Brennan MF. Prognostic significance of a positive microscopic margin in high-risk extremity soft tissue sarcoma: implications for management. J Clin Oncol 1996;14: 473-478. 138. Suit HD, Mankin HJ, Wood WC, et al. Treatment of the patient with stage M0 soft tissue sarcoma. J Clin Oncol 1988;6: 854-862. 139. Stinson SF, DeLaney TF, Greenberg J, et al. Acute and long-term effects on limb function of combined modality limb sparing therapy for extremity soft tissue sarcomas. Int J Radiat Oncol Biol Phys 1991;21:1493-1499. 140. Talbert ML, Zagars GK, Sherman NE, et al. Conservative surgery and radiation therapy for soft tissue sarcomas of the wrist, hand, ankle, and foot. Cancer 1990;66:2482-2491. 141. Shiu MH, Hilaris BS, Harrison LB, et al. Brachytherapy and function-saving resection of soft tissue sarcoma arising in the limb. Int J Radiat Oncol Biol Phys 1991;21:1485-1492. 142. Pearlstone DB, Janjan NA, Feig BW, et al. Re-resection with brachytherapy for locally recurrent soft tissue sarcoma arising in a previously radiated field. Cancer J Sci Am 1999;5:25-33. 143. Nori D, Shupak K, Shiu MH, et al. Role of brachytherapy in recurrent extremity sarcoma in patients treated with prior surgery and irradiation. Int J Radiat Oncol Biol Phys 1991;20: 1229-1233. 144. Cannon CP, Ballo MT, Zagars GK, et al. Complications of combined modality treatment of primary lower extremity soft-tissue sarcomas. Cancer 2006;107:2455-2461. 145. Sugarbaker PH, Barofsky I, Rosenberg SA, et al. Quality of life assessment of patients in extremity sarcoma clinical trials. Surgery 1982;91:17-23. 146. Weddington WW, Segraves KB, Simon MA. Psychological outcome of extremity sarcoma survivors undergoing amputation or limb salvage. J Clin Oncol 1985;3:1393-1399. 147. Bujko K, Suit HD, Springfield DS, et al. Wound healing after preoperative radiation for sarcoma of soft tissues. Surg Gynecol Obstet 1993;176:124-134.

963

148. Bell RS, Mahoney J, O’Sullivan B, et al. Wound healing complications in soft tissue sarcoma management: comparison of three treatment protocols. J Surg Oncol 1991;46:190-197. 149. Dewer JA, Duncan W. A retrospective study of the role of ­radiotherapy in the treatment of soft tissue sarcoma. Clin Radiol 1985;36:629-632. 150. Slater JD, McNeese MD, Peters LJ. Radiation therapy for unresectable soft tissue sarcomas. Int J Radiat Oncol Biol Phys 1986;12:1729-1734. 151. Kepka L, DeLaney TF, Suit HD, et al. Results of radiation ­therapy for unresectable soft-tissue sarcomas. Int J Radiat Oncol Biol Phys 2005;63:852-859. 152. Tepper JE, Suit HD. Radiation therapy alone for sarcoma of soft tissue. Cancer 1985;56:475-479. 153. Verweij J, Seynaeve C. The reason for confining the use of adjuvant chemotherapy in soft tissue sarcoma to the investigational setting. Semin Radiat Oncol 1999;9:352-359. 154. Benjamin RS. Evidence for using adjuvant chemotherapy as standard treatment of soft tissue sarcoma. Semin Radiat Oncol 1999;9:349-351. 155. Antman K, Ryan L, Borden E, et al. Pooled results from three randomized adjuvant studies of doxorubicin versus observation in soft tissue sarcoma: 10 year results and review of the literature. In Salmon S (ed): Adjuvant Therapy of Cancer VI. Philadelphia: WB Saunders, 1990, pp 529-543. 156. Baker LH. Adjuvant therapy for soft tissue sarcomas. In Ryan JR, Baker LO (eds): Recent Concepts in Sarcoma Treatment. Dordrecht: Kluwer Academic, 1988, pp 131-136. 157. Wilson RE, Wood WC, Lerner HL, et al. Doxorubicin chemotherapy in the treatment of soft-tissue sarcoma. Arch Surg 1986;121:1354-1359. 158. Lerner HJ, Amato DA, Savlov ED, et al. Eastern Cooperative Oncology Group: a comparison of adjuvant doxorubicin and ­observation for patients with localized soft tissue sarcoma. J Clin Oncol 1987;5:613-617. 159. Antman K, Suit H, Amato D, et al. Preliminary results of a randomized trial of adjuvant doxorubicin for sarcomas: lack of apparent difference between treatment groups. J Clin Oncol 1984;2:601-608. 160. Alvegård TA, Sigurdsson H, Mouridsen H, et al. Adjuvant chemotherapy with doxorubicin in high-grade soft tissue sarcoma: a randomized trial of the Scandinavian Sarcoma Group. J Clin Oncol 1989;7:1504-1513. 161. Frustaci S, Gherlinzoni F, De Paoli A, 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;19:1238-1247. 162. Frustaci S, De Paoli A, Bidoli E, et al. Ifosfamide in the adjuvant therapy of soft tissue sarcomas. Oncology 2003;65(Suppl 2): 80-84. 163. Eilber FR, Giuliano AE, Huth JF, et al. A randomized prospective trial using postoperative adjuvant chemotherapy (Adriamycin) in high-grade extremity soft-tissue sarcomas. Am J Clin ­Oncol 1988;11:39-45. 164. Omura GA, Blessing JA, Major F, et al. A randomized clinical trial of adjuvant Adriamycin in uterine sarcomas: a Gynecologic Oncology Group study. J Clin Oncol 1985;3:1240-1245. 165. Bramwell V, Rouesse J, Steward W, et al. Adjuvant CYVADIC chemotherapy for adult soft tissue sarcoma—reduced local ­recurrence but no improvement in survival: a study of the European Organization for Research and Treatment of Cancer Soft Tissue and Bone Group. J Clin Oncol 1994;12:1137-1149. 166. Ravaud A, Bui NB, Coindre J, et al. Adjuvant chemotherapy with CyVADIC in high risk soft tissue sarcoma: a randomized prospective trial. In Salmon S (ed): Adjuvant Therapy of Cancer VI. Philadelphia: WB Saunders, 1990, pp 556-566. 167. Benjamin RS, Terjanian TO, Fenoglio CJ, et al. The importance of combination chemotherapy for adjuvant treatment of highrisk patients with soft-tissue sarcomas of the extremity. In Sal­ mon S (ed): Adjuvant Therapy of Cancer V. Orlando, FL: Grune & Stratton, 1987, pp 735-744.

964

PART 12  Musculoskeletal System

168. Rosenberg SA, Tepper J, Glatstein E, et al. Prospective randomized evaluation of adjuvant chemotherapy in adults with soft tissue sarcomas of the extremities. Cancer 1983;52:424-434. 169. Chang AE, Kinsella T, Glatstein E, et al. Adjuvant chemotherapy for patients with high-grade soft-tissue sarcomas of the extremity. J Clin Oncol 1988;6:1491-1500. 170. Glenn J, Kinsella T, Glatstein E, 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;55:1206-1214. 171. Glenn J, Sindelar WF, Kinsella T, et al. Results of multimodality therapy of resectable soft-tissue sarcomas of the retroperitoneum. Surgery 1985;97:316-325. 172. Edmonson JH, Fleming TR, Ivins JC, et al. Randomized study of systemic chemotherapy following complete excision of nonosseous sarcomas. J Clin Oncol 1984;2:1390-1396. 173. Zalupski MM, Ryan JR, Hussein ME, et al. Defining the role of adjuvant chemotherapy for patients with soft tissue sarcoma of the extremities. In Salmon SE (ed): Adjuvant Therapy of Cancer VII. Philadelphia: JB Lippincott, 1993, pp 385-392. 174. Tierney JF, Stewart LA, Parmar MKB. Adjuvant chemotherapy for localised resectable soft-tissue sarcoma of adults: meta­analysis of individual data. Lancet 1997;350:1647-1654. 175. Torres MA, Ballo MT, Butler CE, et al. Management of locally recurrent soft-tissue sarcoma after prior surgery and radiation therapy. Int J Radiat Oncol Biol Phys 2007;67:1124-1129. 176. Jaques DP, Coit DG, Hajdu SI, et al. Management of primary and recurrent soft-tissue sarcoma of the retroperitoneum. Ann Surg 1990;212:51-59. 177. Ballo MT, Zagars GK, Pollock RE, et al. Retroperitoneal soft tissue sarcoma: an analysis of radiation and surgical treatment. Int J Radiat Oncol Biol Phys 2007;67:158-163. 178. Le QT, Fu KK, Kroll S, et al. Prognostic factors in adult soft­tissue sarcomas of the head and neck. Int J Radiat Oncol Biol Phys 1997;37:975-984. 179. Weber RS, Benjamin RS, Peters LJ, et al. Soft tissue sarcomas of the head and neck in adolescents and adults. Am J Surg 1986;152:386-392. 180. McKenna WG, Barnes MM, Kinsella TJ, et al. Combined modality treatment of adult soft tissue sarcomas of the head and neck. Int J Radiat Oncol Biol Phys 1987;13:1127-1133. 181. Morrison WH, Byers RM, Garden AS, et al. Cutaneous angiosarcoma of the head and neck. Cancer 1995;76:319-327. 182. Owens JC, Shiu MH, Smith R, et al. Soft tissue sarcomas of the hand and foot. Cancer 1985;55:2010-2018. 183. Fletcher CDM. Pleomorphic malignant fibrous histiocytoma: fact or fiction? A critical reappraisal based on 159 tumors diagnosed as pleomorphic sarcoma. Am J Surg Pathol 1992;16:213228. 184. Rööser B, Willén H, Gustafson P, et al. Malignant fibrous histiocytoma of soft tissue: a population-based epidemiologic and prognostic study of 137 patients. Cancer 1991;67:499-505. 185. Fagundes HM, Lai PP, Dehner LP, et al. Postoperative radiotherapy for malignant fibrous histiocytoma. Int J Radiat Oncol Biol Phys 1992;23:615-619. 186. Guadagnolo BA, Zagars GK, Ballo MT, et al. Long-term outcomes for synovial sarcoma treated with conservation surgery and radiotherapy. Int J Radiat Oncol Biol Phys 2007;69:11731180. 187. Wright PH, Sim FH, Soule EH, et al. Synovial sarcoma. J Bone Joint Surg Am 1982;64:112-122. 188. Rööser B, Willén H, Hugoson A, et al. Prognostic factors in synovial sarcoma. Cancer 1989;63:2182-2185. 189. Chang HR, Hajdu SI, Collin C, et al. The prognostic value of histologic subtypes in primary extremity liposarcoma. Cancer 1989;64:1514-1520. 190. Evans HL. Liposarcoma: a study of 55 cases with a reassessment of its classification. Am J Surg Pathol 1979;3:507-523. 191. Cheng EY, Springfield DS, Mankin HJ. Frequent incidence of extrapulmonary sites of initial metastasis in patients with lipo­ sarcoma. Cancer 1995;75:1120-1127.

192. Knight JC, Renwick PJ, Cin PD, et al. Translocation t(12;16) (q13;p11) in myxoid liposarcoma and round cell lipo­sarcoma: molecular and cytogenetic analysis. Cancer Res 1995;55: 24-27. 193. Guadagnolo BA, Zagars GK, Ballo MT, et al. Excellent local control rates and distinctive patterns of failure in myxoid liposarcoma treated with conservation surgery and radiotherapy. Int J Radiat Oncol Biol Phys 2008;70:760-765. 194. Pritchard DJ, Nascimento AG, Petersen IA. Local control of extra-abdominal desmoid tumors. J Bone Joint Surg Am 1996;78:848-854. 195. McKinnon JG, Neifeld JP, Kay S, et al. Management of desmoid tumors. Surg Gynecol Obstet 1989;169:104-106. 196. Goy BW, Lee SP, Eilber F, et al. The role of adjuvant radiotherapy in the treatment of resectable desmoid tumors. Int J Radiat Oncol Biol Phys 1997;39:659-665. 197. Spear MA, Jennings LC, Mankin HJ, et al. Individualizing management of aggressive fibromatosis. Int J Radiat Oncol Biol Phys 1998;40:637-645. 198. Ballo MT, Zagars GK, Pollack A, et al. Desmoid tumor: prognostic factors and outcome after surgery, radiation therapy, or combined surgery and radiation therapy. J Clin Oncol 1999;17: 158-167. 199. Ballo MT, Zagars GK, Pollack A. Radiation therapy in the management of desmoid tumors. Int J Radiat Oncol Biol Phys 1998;42:1007-1014. 200. Schmitt G, Mills EED, Levin V, et al. Radiotherapy of aggressive fibromatosis. Eur J Cancer 1992;28A:832-835. 201. Kamath SS, Parsons JT, Marcus RB, et al. Radiotherapy for local control of aggressive fibromatosis. Int J Radiat Oncol Biol Phys 1996;36:325-328. 202. Bendix-Hansen K, Myhre-Jensen O, Kaae S. Dermatofibrosarcoma protuberans: a clinico-pathological study of nineteen cases and a review of world literature. Scand J Plast Reconstr Surg 1983;17:247-252. 203. McPeak C, Cruz T, Nicastri A. Dermatofibrosarcoma protuberans: an analysis of 86 cases—five with metastases. Ann Surg 1967;166:803-816. 204. Roses DF, Valensi Q, LaTrenta G, et al. Surgical treatment of dermatofibrosarcoma protuberans. Surg Gynecol Obstet 1986;162:449-452. 205. Gloster HM, Harris KR, Roenigk RK. A comparison between Mohs micrographic surgery and wide surgical excision for the treatment of dermatofibrosarcoma protuberans. J Am Acad Dermatol 1996;35:82-87. 206. Ballo MT, Zagars GK, Pisters P, et al. The role of radiation therapy in the management of dermatofibrosarcoma protuberans. Int J Radiat Oncol Biol Phys 1998;40:823-827. 207. Suit H, Mankin HJ, Efrid J, et al. Radiation in management of patients with dermatofibrosarcoma protuberans. J Clin Oncol 1996;14:2365-2369. 208. McMaster MJ, Soule EH, Ivins JC. Hemangiopericytoma: a clinicopathologic study and long-term followup of 60 patients. Cancer 1975;36:2232-2244. 209. Enzinger FM, Smith BH. Hemangiopericytoma: an analysis of 106 cases. Hum Pathol 1976;7:61-82. 210. Smullens SN, Scotti DJ, Osterholm JL, et al. Preoperative embolization of retroperitoneal hemangiopericytomas as an aid in their removal. Cancer 1982;50:1870-1875. 211. Mira JG, Chu FC, Fortner JG. The role of radiotherapy in the management of malignant hemangiopericytoma. Cancer 1977; 39:1254-1259. 212. Jha N, McNeese M, Barkley HT, et al. Does radiotherapy have a role in hemangiopericytoma management? Report of 14 new cases and a review of the literature. Int J Radiat Oncol Biol Phys 1987;13:1399-1402.

Childhood Cancer 37 Shiao Y. Woo NORMAL TISSUE EFFECTS ACUTE LEUKEMIAS Clinical and Biologic Features Meningeal Leukemia Testicular Leukemia Prognosis WILMS’ TUMOR Radiation Therapy Prognosis NEUROBLASTOMA Staging

Surgery, Observation, Chemotherapy, and Radiation Therapy RHABDOMYOSARCOMA Radiation Therapy Prognosis RETINOBLASTOMA Radiation Therapy Prognosis LIVER NEOPLASMS

Approximately 9000 new cases of childhood cancer occur annually in the United States. Cancer is the second most common cause of death in children younger than  18 years. The most common pediatric malignant diseases are listed in Table 37-1. Two categories of malignancies account for one half of all cases of childhood cancer: acute leukemias and central nervous system (CNS) tumors. Many of the other common neoplasms are broadly classified as small round cell tumors, reflecting the basic histologic features of neuroblastoma, malignant lymphoma, Hodgkin disease, rhabdomyosarcoma, Ewing’s sarcoma, and retinoblastoma.1 Pediatric cancer is relatively sensitive to radiation ­therapy and chemotherapy. Combinations of surgery and radiation therapy with multiagent chemotherapy since the 1960s have resulted in sharply increased survival rates for children with most types of cancer. Controlled clinical ­trials have sought to define the roles of complementary treatment modalities and newer therapeutic interventions while seeking to optimize disease control and minimize late sequelae. Given the risks of radiation therapy, particularly for young children, it is essential to clearly identify those for whom radiation therapy is truly important for locoregional disease control and survival. The advent of more sophisticated, image-guided radiation techniques has been particularly advantageous in pediatric oncology, and in several disease settings, these techniques have increased the role for radiation therapy in childhood cancer treatment.

Normal Tissue Effects Radiation therapy in children requires a delicate balance between efficacy and radiation-related late toxic effects. Acute radiation effects are similar to those occurring in adults; however, the heightened toxicity of combined chemoradiation therapy must be borne in mind because multiagent chemotherapy is often used in sequence or concurrently with radiation.

It is incumbent on the radiation oncology community to recognize the functional consequences of childhood ­cancer and its therapies. Of 142 children followed for 6 to 21 years after successful cancer management during the 1950s and 1960s, Li and Stone2 described major physical defects in 52%, but virtually all survivors were living “fully active lives.” Later series similarly indicate physical abnormalities in most long-term survivors, 40% of whom show functional deficits after childhood cancer therapy.3 Data from long-term survivors of Wilms’ tumor who were treated before they were 1 year old confirm the frequency of significant, largely musculoskeletal changes but the relative lack of life-threatening or disabling problems.2,4,5 Growth disturbances are unique effects of cancer ­therapy in growing children, and they have been observed in approximately 40% of long-term survivors.2,4,6 ­Scoliosis, for example, had been identified as a consequence of inhomogeneous vertebral irradiation.7 Vertebral growth ­changes after current radiation techniques are identifiable although clinically insignificant at dose levels below 35 Gy.4,8 Reduction in vertebral height after radiation therapy at dose levels of at least 24 Gy is most apparent in children treated before they are 6 years  old or during the adolescent growth spurt.8 Changes in soft tissue development often exceed changes in developing bone after modern radiation therapy. Muscular hypoplasia, for example, is common, ­resulting in functional or aesthetic alterations depending on the region irradiated (Fig. 37-1).9 Radiation changes in long bones include reduction in bone length resulting from treatment of epiphyseal growth centers.6 The apparent threshold for significant changes in bone length is age-related; changes can be anticipated ­after doses as low as 10 to 15 Gy in very young children  (<3 years) and more than 25 Gy in children older than 3 to 5 years. Growth deficits are proportional to the amount of potential growth related to the irradiated epiphysis at the time of treatment. In the head and neck region, doses as low as 18 Gy are associated with observable defects in skull or facial bone development.10 Orbital growth is altered after surgery plus radiation for retinoblastoma.10

967

968

PART 13  Childhood Cancer 160

TABLE 37-1.    Cancer in Children

Cancer Type Acute leukemia

Percentage Occurring in Children

Peak Incidence by Age

31.0

2-8 yr

Acute lymphoblastic leukemia

21.0

2 yr, then stable

Other

10.0

2-18 yr

19.0

Varies with ­histology

Central nervous system Malignant lymphoma

8.0

Varies with type

Hodgkin disease

6.0

Increases with age

Neuroblastoma

8.0

0-3 yr

Wilms’ tumor

6.0

<5 yr

Soft tissue sarcomas

5.0

<5 yr; >10-12 yr

Bone sarcomas

5.0

Increases with age >10 yr

Retinoblastoma

2.5

0-2 yr

Germ cell tumors

2.5

<1 yr; >8-10 yr

Liver tumors

2.5

4 yr

Alterations in growth also occur as a result of radiationrelated endocrine dysfunction. Growth hormone ­deficiency has been identified in more than 50% of children who have undergone irradiation of the pituitary-­hypothalamic region.11,12 A dose effect is apparent, with growth hormone deficits after doses above 25 to 30 Gy in children treated for primary brain tumors and sarcomas of the head and neck region.12-14 After 24 Gy of preventive cranial irradiation for acute lymphoblastic leukemia (ALL), subtle growth hormone deficits become apparent as the child undergoes the pubertal growth spurt.13,15 In girls, even low radiation doses (18 to 24 Gy for ALL) have been associated with early onset of puberty.16 Hypothyroidism also occurs relatively often after cranial irradiation; when the thyroid region is included within the radiation volume, clinical or chemical evidence of decreased thyroid function is apparent in 10% to 25% of children after doses of 18 to 26 Gy (e.g., for ALL or Hodgkin disease) and in more than 50% of children after doses exceeding 25 to 30 Gy (e.g., for medulloblastoma, sarcoma, or Hodgkin disease).17-20 Complicating matters are recent findings indicating that up to 50% of children with brain tumors have preexisting hormonal deficits before undergoing irradiation.21 Other radiation-related somatic changes include late pulmonary deficits. These are caused primarily by ­diminished thoracic and lung parenchymal volumes after ­ low-dose ­irradiation (15 Gy) in young children with Wilms’ ­tumor and after carmustine as a component of multimodality therapy in children with brain tumors.22,23 The identification of CNS changes after cranial ­irradiation in conjunction with intrathecal or intravenous chemotherapy in ALL has led to relatively limited use of preventive cranial irradiation in this setting. A subacute syndrome of somnolence, often associated with nausea, irritability, or

140

120

100

80

60

FIGURE 37-1.  Muscular hypoplasia after mantle irradiation during childhood.

low-grade fever, has been documented in approximately 50% of children 4 to 8 weeks after cranial irradiation to 18 to 24 Gy.24,25 Cerebral atrophy and mineralizing microangiopathy are identifiable on computed tomography (CT) or magnetic resonance imaging (MRI) studies after treatment with cranial irradiation or chemotherapy.26-29 Leukoencephalopathy is a more profound, sometimes fatal entity histologically identified as focal ­ coagulative necrosis of the white matter.30 The clinical syndrome includes lethargy, seizures, perceptual changes, and cerebellar dysfunction. The imaging diagnosis is based on characteristic diffuse white matter changes (Fig. 37-2). Although leukoencephalopathy is independently related to cranial irradiation and to methotrexate (predominately systemic but also intrathecal), its incidence is highest among patients treated with high-dose or prolonged intravenous methotrexate after cranial irradiation.31,32 Neuropsychological alterations have been well documented in survivors of ALL. Eiser33 initially described significant deterioration of the intelligence quotient (IQ) in children treated with cranial irradiation. Several subsequent studies34-37 have confirmed IQ declines related to cranial irradiation and to methotrexate administration.36 Treatment of children younger than 3 to 4 years generally has been associated with a greater risk of ­ neuropsychological deficits.38,39 Reproductive and genetic effects of abdominal irradiation include an increased risk of perinatal mortality and low-birth-weight infants. No excess congenital malformations or early malignancies in first-generation offspring have been described.40-42

37  Childhood Cancer

A

969

B

FIGURE 37-2.  Transverse section (A) and coronal section (B) T2-weighted MRI scans reveal extensive white matter changes (arrow) characteristic of leukoencephalopathy in an 11-year-old child given high-dose methotrexate without irradiation for acute lympho­ blastic leukemia. The imaging findings were associated with the onset of generalized seizures and lassitude, all of which improved over the ensuing 6 to 8 months.

Perhaps the single most important issue affecting the risk-benefit ratio for radiation therapy in children is the frequency of secondary tumors. After orthovoltage irradiation, Li and colleagues43 reported a 10-year actuarial incidence of treatment-related second neoplasms approaching 14%. Although related to alkylating agents and radiation therapy, 90% of the tumors in Li’s initial study occurred within the radiation fields. The frequency of secondary tumors after megavoltage irradiation approximates 4% to 7% at 20 years.44,45 Secondary cancer is more common in girls than in boys, reflecting an incidence of radiation-related breast cancer of as much as 20% after the second decade after irradiation for pediatric Hodgkin disease.44,46 Some findings suggest that certain disease entities, including Ewing’s sarcoma, are associated with a higher ­incidence of secondary cancer, especially sarcomas.47,48 An increased incidence of spontaneous and treatment-related neoplasms is apparent in children with genetically determined disorders such as neurofibromatosis, xeroderma pigmentosum, and Gorlin’s or basal cell nevus syndrome.49-54 The highest risk of secondary cancer is in children with the genetic form of retinoblastoma. In this group of young children with known familial disease or spontaneous bilateral disease, the chance of secondary tumors, largely but not entirely radiation-related bone sarcomas, has been 25% to 40% by 20 years after treatment and up to 50% by 50 years after irradiation (compared with only 5% in patients with similarly treated but spontaneous unilateral retinoblas­ toma).44,55 Fully one third of second malignant neoplasms after radiation therapy identified in the Late Effects Study Group44 occurred in patients with genetic disorders.56-60 In considering the added benefit of irradiation in disease control, oncologists must always consider the potential for functional limitations and even death that follow therapeutic interventions in this often vulnerable age group.44 As the role for technically sophisticated radiation therapy is

reexamined and often expanded, it is important to neither discount the importance of locoregional control, often best achieved with irradiation, nor ignore the late consequences of irradiation in young and developing children.

Acute Leukemias Clinical and Biologic Features Acute leukemias are the most common types of cancer in children. ALL in particular represents 80% of pediatric leukemias and about 20% of all cases of childhood cancer. The most common presenting symptoms and signs relate to bone marrow replacement and include fever and ­neutropenia, petechiae or hemorrhage associated with thrombocytopenia, fatigue related to anemia, and bone and joint pain. The incidence of ALL peaks between the ages of 2 and 8 years. ALL is a systemic disease. Clinical features that predict a relatively poor prognosis include age younger than 1 year or older than 9 years, white blood cell count higher than 50,000, and T-cell or mature B-cell immunophenotype (versus the more favorable B-cell progenitor ­immunophenotype).61,62 Immunologic studies identify a B-cell lineage in about 85% of ALL cases; 60% have surface B-cell differentiation antigens (early pre-B cell, associated with a relatively favorable prognosis), 20% have intracellular immunoglobulin heavy chains (pre-B cell), and 1% have actual surface immunoglobulin (B cell). T-cell characteristics are present in 15% of ALL cases, classically associated with age older than 10 years, high white blood cell count, and mediastinal lymphadenopathy.63 Leukocyte immunophenotype is determined by using monoclonal antibodies to lineage-specific  clusters-of-differentiation (CD) molecular ­ markers.61,62 Therapy and outcome are defined by the immunophenotype, with modern risk-adapted regimens producing

970

PART 13  Childhood Cancer

e­ xcellent outcomes in patients with B-lineage or T-lineage ALL.61,62,64-66 Cytogenetic features affect the prognosis of ALL. ­Hyperdiploidy and the presence of the TEL-AML1 ­fusion gene correlate strongly with a favorable outcome among children 1 to 9 years old; in contrast, hypodiploidy, the presence of the BCR-ABL fusion gene or the Philadelphia chromosome in ALL identifies a high-risk patient ­cohort.61,62,67,68 The poor prognosis for children with ALL who are younger than 1 year old is associated with the relatively specific MLL gene rearrangement characteristic of the disease in this age group.69-71 In addition to the clinical and biologic features associated with outcome in ALL, early response to therapy is highly predictive of disease control. Conversely, the presence of residual disease (whether evident clinically or only by molecular analysis [defined as minimal residual disease]) early during induction therapy or at the completion of the induction regimen is associated with a high risk of subsequent relapse.68,72,73 Acute myelogenous and myelomonocytic leukemias represent about 15% of childhood leukemias. Treatment results show improvement in outcome in these leukemias, although not to the impressive level now associated with ALL.74,75

Meningeal Leukemia Preventive Therapy for Occult Disease The concept of prolonged multiagent chemotherapy for curative treatment of ALL was introduced in the mid1960s.76,77 The basic strategy of induction therapy followed by consolidation chemotherapy produced durable ­hematologic remissions while highlighting the problem of meningeal (CNS) leukemia. In early trials in which no specific CNS therapy was used, the incidence of initial meningeal failure was as high as 70%.78 A prospective randomized ­ trial at St. Jude Children’s Research Hospital79 confirmed the utility of preventive craniospinal irradiation (CSI) (24 Gy in 15 to 16 fractions) early during hematologic remission, reducing the incidence of CNS relapse from 67% to 4%. Subsequent studies showed that cranial irradiation combined with intrathecal methotrexate was as effective as CSI and was associated with less hematologic suppression.76,77 Sequential studies by the Pediatric Oncology Group, Children’s Cancer Group, and European researchers, in addition to several large single-institution trials in the United States, have shown that systemic methotrexate with ­repeated intrathecal instillations of methotrexate alone or in combination with cytarabine and hydrocortisone results in a CNS failure rate of less than 5% in children with B-lineage ALL or T-cell disease plus a white blood cell count of less than 50,000 to 100,000.61,62,66,80-87 The current treatment regimens for ALL incorporate induction, intensification (or consolidation; this phase often uses moderate- to high-dose systemic methotrexate), and continuation therapy for a total of 2.5 years after remission. In conjunction with repeated intrathecal chemotherapy, the more intensive methotrexate-based regimens have improved the control of hematologic and non-CNS  extramedullary (e.g., testicular) disease.85,86,88 The higherrisk patients do have a greater incidence of CNS relapse

when treated without cranial irradiation, and the BerlinFrankfurt-Munster group has continued to recommend somewhat less intensive methotrexate regimens than are typical in the United States, in combination with 12 Gy of  cranial irradiation.84,88 Unclear at this time is where the balance between avoiding radiation-associated sequelae and avoiding methotrexate-related neurologic effects (seizures and white matter alterations occurred at an alarming rate in several U.S. studies) will ultimately set the threshold for cases requiring preventive cranial irradiation.89 Series in Italy and the Netherlands and an ongoing study at St. Jude Children’s Research Hospital have sought to obviate cranial irradiation for all ALL cases without CNS involvement at diagnosis.84,87,90 Compared with the ­ Italian and Dutch chemotherapy-only studies with similar systemic and intrathecal components, the Berlin-­  Frankfurt-Munster trials for children with very-high-risk disease (i.e., T-cell disease manifesting with a blood cell count of more than 100,000) have shown significantly superior results with the addition of 12 Gy of preventive cranial irradiation. The event-free survival rates in the  Berlin-Frankfurt-Munster trials were twice the rates in Dutch and German chemotherapy-only studies, and the incidence of CNS failure was one fifth that seen without cranial irradiation.82-85,87,90

Technique for Cranial Irradiation Preventive cranial irradiation in the United States is currently restricted to children or adolescents with T-cell ALL and a white blood cell count of 50,000 to 100,000 or more. The technique used must ensure adequate ­coverage of the entire intracranial subarachnoid space. Treatment demands careful attention to margins at the base of the skull, particularly the region of the cribriform plate and the temporal fossa (Fig. 37-3). The target volume includes the posterior retina and orbital apex, requiring specific techniques to block the anterior half of the eye and the lens. A lower margin at the bottom of the second cervical vertebra has become standard, ensuring clearance at the base of the skull (see Fig. 37-3). The radiation field around the calvarium should be sufficient to ensure full dose delivery at the subarachnoid space, typically 5 to 10 mm ­beyond the skull. The total dose for the small cohort requiring preventive cranial irradiation is 18 Gy (most often given as once-daily 1.5-Gy fractions); this dose was recently reduced to 12 Gy in selected studies in Europe and the United States.84,85

Treatment of Overt Disease Meningeal leukemia is documented at diagnosis in approximately 5% of ALL cases.91-94 CNS disease is defined by the presence of leukemic cells and at least five white blood cells per milliliter in the cerebrospinal fluid (CSF). CNS-2 disease, with blasts but fewer than five white blood cells in the CSF, has been identified as a potentially negative prognostic factor and has consistently been managed with somewhat augmented intrathecal chemotherapy.83,87,90,92,95 Most clinical trials incorporate cranial irradiation (12 or  18 Gy) for children with CNS involvement at diagnosis, with ongoing studies further testing the role of chemotherapy alone in this setting.61,82,88 CNS relapse after initial remission requires careful integration of radiation therapy and chemotherapy. Early studies

37  Childhood Cancer

971

the gantry to establish parallel, nondivergent beams at the level of the orbital rim; the collimators are also angled to parallel the cephalad divergence of the entering posterior spinal field. Correction for caudal divergence of the entering lateral craniocervical fields can be achieved by angling the treatment pedestal; alternatively, this divergence can be corrected to meet in the midplane by using a calculated 5- to 10-mm gap at the lateral cervical region to match the projection of the light fields at the midplane posteriorly. It is good practice to change the junction by 1 to 2 cm at least once during CSI with the spinal doses recommended for ALL. Immobilization and reproducibility can be improved by using a prone position with a custom-made device to ensure appropriate patient positioning.

c

t

FIGURE 37-3.  Cranial irradiation field for preventive therapy for a patient with acute lymphoblastic leukemia. The field includes the cribriform plate (c) of the subfrontal region and the temporal fossa (t) at the base of the skull.

using CSI for overt CNS relapse were marked by frequent second CNS relapses.79 Subsequent studies confirmed consistent CNS control after clearance of malignant cells from the CSF with intrathecal chemotherapy and prompt initiation of CSI.96-99 Reports of successful management of CNS relapse without adequate CSI are limited.100,101 The most recent Pediatric Oncology Group and European reports indicate a role for intensive chemotherapy lasting 6 to  12 months followed by CSI (cranial irradiation to 24 to 25 Gy and spinal irradiation to 15 Gy).99,102,103 For patients who sustain late CNS relapse (i.e., relapse more than 18 months after initial remission), a secondary event-free survival rate of 83% at 4 years has been reported.102 An ongoing trial by the Children’s Oncology Group will study whether irradiation in patients with late CNS relapse can be limited to only cranial therapy (18 Gy). For children with early CNS relapse, an approach of systemic chemotherapy with CSI results in a 4-year event-free survival rate of only 46%, indicating a continuing requirement for CSI and perhaps a need for bone marrow transplantation in that setting.102

Technique for Craniospinal Irradiation The target volume for CSI includes the same intracranial volume described above for cranial irradiation plus the ­spinal subarachnoid space. The lower margin must extend below S2 to encompass the caudal limit of the thecal sac. Lateral craniocervical fields adjoining a posterior spinal field are most often used. Minimizing inhomogeneity at the field junction requires correcting for divergence (see Fig. 33-2 in Chapter 33). The lateral craniocervical fields are angled at

Testicular Leukemia With current chemotherapy regimens, the occurrence of testicular relapse of ALL has become anecdotal.62,90 In instances of testicular involvement at relapse, bilateral ­testicular irradiation to 20 to 26 Gy has been uniformly successful in achieving local disease control, with the ­ultimate outcome dependent on the timing of testicular ­failure.104-106 Electron-beam irradiation is simplest and ­appropriate in this setting.

Prognosis Studies in which current therapies are used report 70% to 80% overall rates of continuous complete remission for children older than 2 years with ALL.62,65,66,84,85,87,88,103,107   Even for children with high-risk disease factors, treatment with current regimens can produce long-term disease ­control and survival at rates of 70% or more.62,65,66,90 The ­major challenge remaining in ALL is in achieving a leukemia-free survival rate of more than 50% among infants (i.e., those younger than 2 years) and in the limited cohort of patients with specific biologically unfavorable presentations (e.g., Philadelphia chromosome–positive ALL), for whom bone marrow transplantation has become the standard early in remission.61,68

Wilms’ Tumor Wilms’ tumor is a malignant renal tumor occurring predominantly in young children. The tumor is believed to arise from embryonic nephrogenic rests. Almost 70% of Wilms’ tumors occur in children younger than 4 years. A palpable or visible abdominal mass is usually the only presenting sign. Abdominal pain or hematuria occurs in less than one fourth of patients. Wilms’ tumor has been associated with genitourinary anomalies (e.g., cryptorchidism, hypospadias) and somatic malformations (most commonly hemihypertrophy or aniridia). Approximately 10% of children show abnormalities in the WT1 gene (11p13 deletion) associated with the WAGR syndrome (Wilms’ tumor, aniridia, genitourinary abnormalities, and mental retardation).108 The Beckwith-Wiedemann ­syndrome includes ­ exophthalmos, macroglossia, and gigantism often in association with Wilms’ tumor.109 Multivariate analysis showed that changes in the expression of genes such as HEY2, TRIM22, and TOP2A were associated with metastasis or mortality from Wilms’ tumors.110 Confirmation will need to come from large prospective trials.

972

PART 13  Childhood Cancer

It is important to distinguish between Wilms’ tumor and mesoblastic nephroma, a mesenchymal hamartoma occurring in infancy. Fully 25% of mesoblastic nephromas are diagnosed in newborns.111,112 Mesoblastic nephroma has a benign course after resection. Wilms’ tumors comprise epithelial components, including primitive tubules and glomeruli, mixed with areas of mesenchymal tumor, including fibrous, smooth, and skeletal muscle elements. Nephrogenic rests are present in 40% of cases. A single focus of intralobar rests is usually associated with Wilms’ tumor in the WAGR syndrome. The finding of several foci of intralobar rests (or nephroblastomatosis) in unilateral Wilms’ tumor is associated with metachronous disease in the remaining kidney.108,112-114 Perilobar nephrogenic rests are more common than intralobar rests in the general pediatric population and are multifocal and often bilateral. The National Wilms’ Tumor Study (NWTS) has identified striking correlations between histology and prognosis. In 85% of cases, tumors contain only differentiated ­epithelial or mesenchymal elements. These tumors have a favorable prognosis. Histologic findings associated with an unfavorable prognosis are identified in 12% to 15% of ­patients. Anaplastic Wilms’ tumor is seen in 5% of cases; it is rare in infants, but accounts for 13% of tumors in children older than 5 years.113,115-117 Lymph node metastasis and hematogenous dissemination are more common in anaplastic tumors.113,115,116 Clear cell sarcoma of the kidney is technically distinct from Wilms’ tumor but shares a common clinical presentation and age distribution. The clear cell tumors metastasize preferentially to bone and brain.113,118 Rhabdoid tumors are highly aggressive lesions occurring primarily in the ­  kidney (but also in other visceral or CNS sites) in very young children.113,119 Rhabdoid tumors metastasize widely

to liver, lungs, and bone. Brain tumors arise independently in infants with renal rhabdoid tumors, which are associated with a chromosome 22 deletion and the SMARCB1 gene (formerly designated INI1).120-122 Rhabdoid tumors are truly independent of Wilms’ tumors. Prognostic factors identified in the report of Garcia and colleagues123 in 1963 led to the first suggestions for staging of Wilms’ tumor. Cassady and colleagues124 proposed a tumor-node-metastasis (TNM) system built on Garcia’s data, correlating prognosis with tumor size, capsular invasion, lymph node involvement, and distant metastasis. The NWTS initially used groups I to IV (and V for bilateral disease) based on tumor extension and resectability. Analysis of the first two trials conducted by the NWTS (NWTS-1  and NWTS-2) indicated the importance of lymph node metastasis.113,116 The current NWTS staging system  (Table 37-2) is based on histologic evidence of resection margins, infiltrative characteristics of the lesion, and the presence of lymph node metastasis.125,126 With nephrectomy and postoperative abdominal irradiation, Gross and Neuhauser127 reported a disease-free survival rate of 47% among 38 patients with Wilms’ ­tumor by 1950. Farber128 later described a 70% survival rate with the addition of dactinomycin. However, subsequent ­studies including the NWTS-1 trial found that disease-free survival with adjuvant dactinomycin only slightly exceeded  the 50% rate established with surgery and irradiation alone.114,124 The improved survival reported by Farber128 largely reflected the effectiveness of dactinomycin and pulmonary irradiation in salvage therapy for nearly 50% of patients with lung metastases. Prospective trials have proved the efficacy of surgery and combination chemotherapy with vincristine and dactinomycin, or vincristine alone in early-stage disease, for ­patients with differentiated Wilms’ tumor.116,125,129,130

TABLE 37-2.    National Wilms’ Tumor Study Staging System for Wilms’ Tumor Stage

Description

I

Tumor limited to kidney and completely resected Renal capsule intact or infiltrated but not penetrated; renal sinus invaded <2 mm; vessels of renal sinus not involved; tumor not ruptured or biopsied (fine-needle aspiration allowed); and no evidence of tumor at or beyond the margins of resection

II

Tumor extends beyond kidney but completely resected Renal sinus invaded >2 mm; vessels of renal sinus or outside the renal parenchyma involved; and/or tumor biopsied (except fine-needle aspiration) or spillage confined to the flank and does not involve peritoneal surface No evidence of tumor at or beyond the margins of resection

III

Residual nonhematogenous tumor confined to the abdomen Abdominopelvic lymph nodes involved; tumor penetrates peritoneal surface; tumor implants on peritoneal surface; gross or microscopic residual tumor after surgery; tumor incompletely resected because of local infiltration into vital structures; and/or tumor spillage not confined to flank

IV

Hematogenous metastases (e.g., lung, liver, bone, brain); and/or lymph node metastases outside the ­abdominopelvic region; “CT only” lung metastases do not mandate stage IV

V

Bilateral renal involvement at diagnosis Each side should be staged according to above criteria

CT, computed tomography. Adapted from Neville HL, Ritchey ML. Wilms’ tumor. Overview of National Wilms’ Tumor study Group results. Urol Clin North Am 2000; 27:435-442.

37  Childhood Cancer

Radiation Therapy Indications for Abdominal Irradiation Routine postoperative irradiation has been effective in virtually eliminating abdominal recurrences of Wilms’ ­tumor.32,124,131 With increasingly effective chemotherapy, however, abdominal irradiation can be omitted in some groups.132,133 The indications for abdominal irradiation relate directly to disease extent and inversely to the intensity of chemotherapy. Results of the NWTS-1 and NWTS-2 ­trials for ­ patients with early-stage disease (intrarenal stage I)  confirmed this relationship, indicating that postoperative ­ abdominal irradiation could be avoided when using ­effective chemotherapy for what is now considered stage I ­  differentiated Wilms’ tumor.125,135 With more intensive chemotherapy, anaplastic stage I tumors also do not require postoperative irradiation. Stage I clear cell sarcomas are treated with local irradiation after chemotherapy.133 For stage II disease (i.e., microscopic residual disease without lymph node involvement), the third NWTS trial (NWTS-3) included a comparison of 20 Gy of tumor bed irradiation plus chemotherapy versus chemotherapy alone (vincristine and dactinomycin, with or without doxorubicin) and found a 2-year relapse-free survival rate of 88% with chemotherapy for favorable-histology tumors and 92% with the addition of radiation therapy. Abdominal failure occurred in only 3% of cases without irradiation; protocols now obviate irradiation for differentiated stage II disease while requiring it for tumors with anaplastic ­changes or unfavorable histologic types.117,132,133,135 Breslow and colleagues116,136 described tumor factors that were associated with abdominal failure in the early NWTS trials. Statistically significant factors include abdominal spread (defined as contiguous disease extension into the peritoneum or other abdominal organs, usually associated with macroscopic residual abdominal tumor), operative rupture (locally into the tumor bed or diffusely through the peritoneum), lymph node metastasis, and unfavorable histologic type. In the NWTS-2 trial, abdominal tumor recurrences occurred in 20% of cases with unfavorable histologic types.134 All patients with stage III disease (i.e., macroscopic residual abdominal disease, lymph node metastasis, or diffuse peritoneal tumor spillage) require postoperative irradiation.125,134 However, for stage III disease exclusive of diffuse peritoneal spillage, no evidence exists to suggest that whole-abdomen irradiation is necessary. Rather, irradiation of the local tumor bed and para-aortic lymph nodes is indicated for these patients. No obvious radiation therapy dose-response relationship for Wilms’ tumor has been apparent. The NWTS-1 trial showed an equal frequency of abdominal failure after doses of 18 to 20 Gy, 20 to 24 Gy, and 24 to 40 Gy.132-134 In the NWTS-3 trial, patients with stage III disease were randomly assigned to receive 10.8 or 20 Gy after surgery. An investigator option to boost the dose to areas of gross residual disease to 20 Gy in the 10.8-Gy group was used in only 2% of cases. With 10.8 Gy and three-drug chemotherapy, abdominal failure occurred in only 4% of cases; with two drugs, the 10.8-Gy dose was inadequate, leading to abdominal recurrence in 11% of cases.132 Similar single-institution results have established a local dose of

973

10 to 12 Gy as the standard for stage III disease with favorable histology.137,138 Most investigators use a 10-Gy boost dose to sites of overt disease, identified arbitrarily in the NWTS schema as residual disease measuring at least 3 cm in diameter. (One of the more common indications for a boost dose is residual disease in the inferior vena cava.139) For anaplastic Wilms’ tumor and rhabdoid tumor in particular, radiation doses of 20 to 30 Gy were used in the NWTS-2, -3, and -4 trials, but the current protocol (NTWS-5) does not prescribe different doses on the basis of histology, although the rationale for this is not readily apparent. Preoperative irradiation was tested in the earlier European International Society of Pediatric Oncology (SIOP) trials.6 However, preoperative irradiation has not been used extensively in the United States because of concerns regarding the certainty of diagnosis and the potential loss of information regarding tumor stage and histology.133,135

Technique for Abdominal Irradiation Treatment fields for local irradiation encompass the preoperative tumor bed and adjacent para-aortic lymph nodes. In practice, treatment fields extend from the diaphragm to a level below the primary tumor as typically defined by CT. The lateral aspect of the treatment field tangentially includes the abdominal wall; the medial border extends beyond the contralateral vertebral margin to include the paraaortic lymph node chain. Opposed anterior and posterior fields are used. The dose levels are defined in the previous paragraphs. For patients requiring whole-abdomen irradiation, the treatment volume must include the entire peritoneal cavity. The upper margins include the diaphragm bilaterally; the inferior margin is usually defined at the midobturator level, blocking the acetabular and femoral head regions. The dose to the remaining (contralateral) kidney should be minimized; when dose levels in excess of 10.8 Gy are indicated, the dose to any substantial portion of the remaining kidney should not exceed 10 to 12 Gy. When treating bilateral Wilms’ tumor, maximal reduction in the remaining functioning renal substance should be confirmed by a dose-volume histogram. For patients with bilateral tumors or metastatic disease, the need for abdominal irradiation is defined by the local extent of disease in each affected kidney.

Pulmonary Irradiation The ability to achieve durable disease-free survival among patients with pulmonary metastasis from Wilms’ tumor followed the introduction of combined chemotherapy (dactinomycin) and lung irradiation.124,128,140 Ultimate disease control after metastatic failure in the lungs was seen in approximately 50% of patients in earlier series.141,142 The findings indicated the necessity to treat the entire lung ­volume rather than limited fields.143,144 The likelihood of disease control after metastasis is inversely related to the intensity of the initial chemotherapy regimen. This can be seen most easily in terms of ­secondary disease-free survival (i.e., control of posttreatment recurrences) in the first three NWTS studies: 51% in the NWTS-1, 40% in the NWTS-2, and 34% in the NWTS3.141 At the same time, overall survival consistently improved over the three NWTS studies, indicating that the

974

PART 13  Childhood Cancer

proportion of ­patients developing metastasis progressively diminished but that the ability to salvage treatment failures also ­diminished. Approximately 12% of patients with Wilms’ tumor have pulmonary metastasis at diagnosis. Treatment including pulmonary irradiation and chemotherapy has resulted in disease control in 75% of stage IV (lung metastasis) cases with favorable histology.141,144,145 However, a subset analysis of the NWTS-3 trial showed no apparent benefit from the addition of lung irradiation for those with lung metastases evident only on CT (i.e., those for whom plain x-ray films of the chest were normal); the relapse-free survival rate was 85% with or without thoracic irradiation.146 The United Kingdom Children’s Cancer Study Group First Wilms’ Tumor Study, in contrast, showed a less favorable 65% disease control rate for patients with metastatic disease who did not receive thoracic irradiation, suggesting that thoracic irradiation does have a role.130 The radiation dose commonly used is 12 Gy in 8 fractions. Tumor foci that persist at 2 weeks after pulmonary irradiation may be excised or treated to an additional 7.5 Gy in 5 fractions.

Hepatic Irradiation Documented nonresectable hepatic metastases are irradiated with up to 30.6 Gy in fractions of 1.5 to 1.8 Gy as long as less than 75% of the liver receives that dose. If the entire liver is involved, a dose of 19.8 Gy in 11 fractions to the whole liver is commonly used. The dose to the remaining kidney must be limited to 14.4 Gy in the process.

Prognosis In recent series, disease-free survival rates for all stages combined approach 90%. For those with tumors of favorable histology, the NWTS-3 trial showed 4-year survival rates of 96% for those with stage I disease, 92% for stage II disease, and 86% for stage III disease.125 The relapse-free survival rate for patients with nonmetastatic favorablehistology Wilms’ tumor was 88%.125,126 Age older than 4 years, large primary tumor size, and lymph node metastasis are prognostically negative factors.126 Recent information on anaplastic Wilms’ tumor shows an improvement in survival for patients with anaplastic tumors; children with tumors that have only focal anaplastic changes have better survival rates than children whose tumors have diffuse anaplasia.117,118,147 The use of doxorubicin-containing chemotherapy and irradiation for treatment of clear cell sarcoma produce results that approach those for Wilms’ tumors of favorable histology.125,147,148

cells) and ganglioneuroma (composed entirely of mature cellular elements, some of which are presumed to represent spontaneous maturation from earlier occult ­neuroblastoma). The median age at diagnosis of neuroblastoma is 21 months; 30% of neuroblastoma cases are diagnosed in infants younger than 1 year and 80% in children younger than 4 years. The primary tumor site is most often in the abdomen, including the adrenal medulla, celiac axis, or paravertebral ganglia. Posterior mediastinal tumors account for 15% of cases, and 10% arise in the cervical sympathetic chain. Neuroblastoma typically spreads to bone marrow, liver, and bone; 60% of children older than 1 year have identifiable hematogenous metastasis at diagnosis. Local disease extension occurs most notably in paravertebral locations, with tumor infiltrating through the neural foramina and spreading within the contiguous spinal canal as so-called “dumbbell” tumors. Lymph node metastasis has been ­reported in 33% to 48% of cases.152,153 A special pattern of metastatic neuroblastomal (stage 4S) associated with a high frequency of spontaneous cellular maturation or regression has been described.151,154 Infants with stage 4S disease have small primary tumors and metastatic involvement confined to the liver, skin, or bone marrow, and they have a good prognosis with observation only or with very limited therapy.155-157 The clinicopathologic prognostic scheme defined by Shimada includes patient age, the mitosis-karyorrhexis index, proportion of differentiated neuroblasts, and stromal pattern.158 Favorable-prognosis cases in this system are ­defined as follows: stroma-rich without a nodular pattern in patients of all ages; stroma-poor, differentiated pattern, and a mitosis-karyorrhexis index of less than 100 in patients 1.5 to 5 years old; or stroma-poor pattern and a mitosis-  karyorrhexis index of greater than 200 in patients younger than 1.5 years. Unfavorable-prognosis cases have converse defining characteristics.158 Among the biologic variables predictive of therapeutic response and outcome in neuroblastoma, the most important is MYCN gene amplification, present in 25% of ­patients at diagnosis (5% to 10% of those with locoregional disease and 30% to 40% with disseminated disease).159-161 An allelic loss at 1p36 leading to a loss of heterozygosity at chromosome 1 correlates with MYCN amplification.162 Other prognostic biologic characteristics include BCL2 gene expression, a negative indicator correlated with unfavorable histology and MYCN amplification; TRK nerve growth factor expression, a positive indicator correlated with very young age and limited stage; and P-glycoprotein expression, a positive indicator.163-165

Neuroblastoma

Staging

Neuroblastoma is a primitive neuronal tumor of sympathetic nervous tissue. These classic “small, round, blue cell” tumors are derived from primitive neural crest cells. ­Described as enigmatic, neuroblastoma has been associated with spontaneous regression or maturation, or both; it also has a particular age-dependent prognosis, with persistently high mortality rates among children older than 1 year.149-151   A spectrum of cellular maturation occurs among tumors described as ganglioneuroblastoma (with features of immature, undifferentiated neuroblasts and mature ­ganglion

Several staging systems have been used for neuroblastoma. Of the systems most commonly used in the past, the EvansD’Angio and Pediatric Oncology Group systems are based, respectively, on anatomic extent and on extent, degree of resection, and nodal involvement (Table 37-3).166-168 The International Neuroblastoma Staging System (INSS) was revised to better define the midline (i.e., INSS stage 3 was redefined to require disease extending across the width of the vertebral column). The use of INSS stage 4S was also limited to infants younger than 1 year.169 The latter

37  Childhood Cancer

975

TABLE 37-3.    Staging Systems for Neuroblastomas Evans, D’Angio, and Randolph166

Pediatric Oncology Group168

International Neuroblastoma Staging System167

Stage I

Stage A

Stage 1

Tumor confined to the organ or ­structure of origin

Complete gross resection of primary tumor, with or without microscopic residual disease Intracavitary lymph nodes that are not ­adhered to or ­removed in continuity with primary tumor and are histologically free of tumor (nodes adhered to or within tumor resection may be positive for tumor without upstaging to stage C) If primary tumor in abdomen or pelvis, liver histologically free of tumor

Localized tumor confined to the area of origin; complete gross excision, with or without microscopic residual disease­ Identifiable ipsilateral and contralateral lymph nodes negative microscopically

Stage II

Stage B

Stage 2A

Tumor extending in continuity beyond the origin or structure of origin but not crossing the midline Regional lymph nodes on the ipsilateral side may be involved

Gross unresected primary tumor Nodes and liver same as stage A

Unilateral tumor with incomplete gross excision Identifiable ipsilateral and contralateral lymph nodes negative microscopically Stage 2B Unilateral tumor with complete or ­incomplete gross excision Positive ipsilateral regional lymph nodes Identifiable contralateral lymph nodes negative microscopically

Stage III

Stage C

Stage 3

Tumor extending in continuity beyond the midline Regional lymph nodes may be involved bilaterally

Complete or incomplete resection of primary tumor Intracavitary nodes that are not adhered to primary tumor and are histologically positive for tumor Liver as in stage A

Tumor infiltrating across the midline with or without regional lymph node ­involvement or unilateral tumor with contralateral regional lymph node involvement or midline tumor with bilateral regional lymph node involvement

Stage IV

Stage D

Stage 4

Remote disease involving the skeleton, bone marrow, soft tissue, distant lymph node groups, and other sites (see stage IV-S)

Any dissemination of disease beyond intracavitary nodes (i.e., extracavitary nodes, liver, skin, bone marrow, or bone)

Dissemination of tumor to distant lymph nodes, bone, bone marrow, liver, and/or other organs (except as defined in stage 4S)

Stage IV-S

Stage 4S

Patients who would otherwise be stage I or stage II but who have remote disease confined to liver, skin, or bone marrow without radiographic evidence of bone metastases on complete skeletal survey

Localized primary tumor (as defined for stage1, 2A, or 2B), with dissemination limited to skin, liver, and/or bone ­marrow (limited to infants <1 year old)

s­ pecial category refers to cases with local stage 1 or 2 primary ­lesions and metastases limited to the skin, liver, and bone marrow (specifically excluding those with imaging evidence of osseous disease).151 The complex interactions of clinical factors (e.g., stage, age), biology (e.g., MYCN amplification; for infants, deoxyribonucleic acid [DNA] ploidy), and histology define low-, intermediate-, and high-risk presentations for risk-related protocol therapy in trials conducted by the Children’s Oncology Group.170,171 Localized neuroblastoma has a very favorable prognosis after surgery, with or without ­chemotherapy.160,171-176

Surgery, Observation, Chemotherapy, and Radiation Therapy Most cases of localized neuroblastoma are managed by surgery alone, with disease-free survival rates of 85% to 90% and overall survival rates approaching 95%.160,171,172,174-176 No benefit has been proven for adjuvant irradiation or chemotherapy.174,176 Almost identical disease control rates can be achieved for patients with extracapsular but node-negative presentations (INSS stage 2A) after surgery (used as primary  treatment or as adjuvant treatment after chemotherapy for sizable primary lesions) by using relatively ­ nonintensive

976

PART 13  Childhood Cancer

chemotherapy.177 Investigators from Memorial-Sloan Kettering Cancer Center reported similar disease-free and overall survival rates after using observation and limited additional surgery, essentially without cytotoxic chemotherapy, in a cohort of moderate size.174 Chemotherapeutic regimens used have included cyclophosphamide, doxorubicin, cisplatin, and etoposide or teniposide. The incidence of local failure is less than 10% after adjuvant ­chemotherapy.177 The areas of controversy with regard to use of radiation therapy focus on locally advanced or regional stage disease. INSS stage 2B includes patients with unilateral primary tumors and ipsilateral nodal metastasis. The impact of nodal involvement has been somewhat unclear; several early reports have indicated an adverse effect.157,178 Whether more consistent use of regional irradiation would eliminate any effect of nodal extension was speculated on in series ­reported more than 20 years ago.153 In a Pediatric ­Oncology Group prospective randomized trial of postoperative chemotherapy with or without irradiation (24 to 30 Gy in 16 to 20 fractions) for children older than 1 year with stage C neuroblastoma, children with local tumor extent that was comparable to INSS stage 3 or Evans-D’Angio stage III who received radiation therapy had superior rates of complete response (76% with irradiation versus 56% with chemotherapy only), event-free survival (59% versus 32%), and overall survival (73% versus 41%) than did children with comparably staged disease who were given chemotherapy only. The mediastinal and supraclavicular nodal regions received doses of 24 and 18 Gy, respectively.179   A subsequent Pediatric Oncology Group trial with stage C cases used dose-intensive chemotherapy and secondlook surgery, reserving radiation for those with residual ­locoregional disease after the second-look surgery.180 The overall results were said to be comparable to those of the earlier study group that received radiation but showed a poor outcome (<10% event-free survival) for those with MYCN amplification.180 Experience with treating residual, nonmetastatic neuroblastoma in children older than 1 year with observation and surgery or with nonmyeloablative chemotherapy has been interpreted to indicate that limited indications for irradiation exist in children with less than metastatic disease at diagnosis.172-176,181 The use of intraoperative irradiation has also been explored in selected settings, often for regional consolidation therapy in children who initially had INSS stage 4 disease and responded to chemotherapy.182,183 For children older than 1 or 2 years presenting with INSS stage 4 disease, long-term disease control rates range from 20% to 40%.114,170,184,185 The degree of progress over the past 2 decades is unclear. The use of high-dose, myeloablative chemotherapy (with autologous marrow or peripheral stem cell reconstitution), sometimes combined with ­totalbody irradiation, has resulted in survival rates approximating 40% in some series. Concurrent Children’s Cancer Group trials found that intensive chemotherapy produced better 4-year event-free survival rates than did conventional chemotherapy (40% versus 19%).170,184,185 A benefit from primary tumor resection for patients with INSS stage 4 disease has been supported in subsequent analyses.186,187 A benefit from local or regional irradiation in stage 4  disease has never been proven, although patterns of failure analyses have shown locoregional disease recurrence in up

to one third of children with stage 4 disease.181,188 The use of 18 to 21 Gy (typically 1.5 Gy per fraction, delivered once or twice daily) to locoregional sites of residual disease before high-dose chemotherapy and total-body irradiation has been studied. Institutional and cooperative group experiences suggest that this regimen is tolerable and effective.181,188-191 For patients with evidence of residual disease on images obtained before planned ablative chemotherapy, regional irradiation seems to be beneficial. Total-body irradiation regimens typically involve 12 Gy given in fractions of 2 Gy once or twice daily or 1.2 Gy twice daily.189,191 For infants with stage INSS 2B or 3 disease, less aggressive (i.e., not myeloablative) chemotherapy has resulted in disease control rates of more than 90%.192,193 The DNA index has been an important prognostic variable, with inferior outcome among infants with diploid patterns. Those with stage 4 disease have survival rates of 75% to 80% after modern chemotherapy.179,194 It is uncommon that irradiation be required in this age group. Infants presenting with INSS stage 4S disease represent a small group with a relatively favorable prognosis.151 Survival rates approach 90% with limited intervention such as limited-intensity chemotherapy. A significant minority of patients present with urgent respiratory compromise resulting from massive hepatomegaly; some series indicate that up to two thirds of cases requiring any form of therapy will need hepatic irradiation.156 Typical hepatic irradiation regimens deliver a total of 4.5 Gy, given in three consecutive daily 1.5-Gy fractions to the entire liver.156 Neuroblastoma often manifests in paravertebral locations, both intrathoracic and intra-abdominal. The so-called dumbbell pattern, with direct extension through the neural foramina into the epidural compartment of the spinal canal, is relatively common. Current therapy for paravertebral neuroblastoma is most often chemotherapy alone, with occasional indications for laminectomy or irradiation at presentation.160,195 For patients presenting with progressive abdominal disease or disseminated osseous disease, palliative irradiation is highly effective. Local doses of 21 to 30 Gy given in conventional or accelerated fractionation schedules often produce immediate relief of bone pain or improvement in bulky retroperitoneal disease. Radiolabeled meta-iodobenzylguanidine has been shown to be effective against diffuse, symptomatic bone metastases.157,196,197

Rhabdomyosarcoma Rhabdomyosarcoma is a primitive tumor of skeletal muscle origin accounting for more than 70% of childhood soft tissue sarcomas. The most common sites of origin are the head and neck (40%), the genitourinary tract (20% to 25%), the extremities (15% to 20%), and the trunk (15% to 20%). Children younger than 6 years more commonly have tumors of the head and neck and genitourinary tract  (especially vaginal, bladder, and prostatic tumors) ­compared with adolescents, who more often have tumors of the ­extremities, trunk (thorax, retroperitoneum, or perineum), and paratesticular regions. Ultrastructural studies of rhabdomyosarcoma often ­reveal the striated muscle origin. In studies conducted by the U.S. Intergroup Rhabdomyosarcoma Study (IRS) Group,198 embryonal rhabdomyosarcoma has been the most common histologic subtype; alveolar ­ rhabdomyosarcoma has been

37  Childhood Cancer

the second most common subtype; and undifferentiated sarcomas and peripheral neuroepitheliomas (also called peripheral primitive neuroectodermal tumors or extraosseous Ewing’s sarcoma) have accounted for the remaining tumors. The histologic subtypes are associated with various clinical and biologic findings. Embryonal rhabdomyosarcoma, which accounts for 55% of cases, occurs primarily in children younger than 8 years and at head and neck or genitourinary sites; the tumor is associated with loss of heterozygosity at chromosome 11 (11p15). Alveolar rhabdomyosarcoma, which accounts for 20% of cases, is more common in adolescents, originating in extremity or truncal sites; a specific t(2;13)(q35;q14) translocation marks this entity. Botryoid rhabdomyosarcoma, which accounts for 5% of cases, is an exophytic growth, marked histologically by a subepithelial cambium layer of dense cellularity; the tumor occurs predominantly in infants as bladder or vaginal lesions and less commonly in young children within the nasopharynx. Other uncommon subtypes of rhabdomyosarcoma include undifferentiated and pleomorphic sarcomas. Extraosseous Ewing’s sarcoma (or peripheral primitive neuroectodermal tumor) is considered a separate entity, initially identified by the IRS Group but now managed and reported with ­osseous Ewing’s tumors.199-205 Uncommon lesions related to rhabdomyosarcoma include ectomesenchymomas (i.e., rhabdomyosarcoma with neuronal differentiation) and Triton tumors (i.e., malignant peripheral nerve sheath tumors with rhabdomyosarcoma components).201-203,205 The natural history of rhabdomyosarcoma is of local tissue infiltration, regional lymph node involvement, and ­ hematogenous dissemination. The tumor notoriously spreads along soft tissue planes. Primary tumors in the ­nasopharynx (and nasal cavity), paranasal sinuses, middle ear, and infratemporal or pterygoid fossa are grouped as parameningeal lesions; they share a tendency to involve the adjacent base of the skull by direct bone erosion or infiltration along cranial nerves (Fig. 37-4).206 Rhabdomyosarcoma is typically an invasive tumor, manifesting as locoregional disease in 83% of cases and as overt

A

977

metastases in 17%.198,207,208 Lymph node metastasis has been documented in 15% of children at diagnosis; primary tumors were in the prostate in 40%, paratesticular region in 25%, extremities in 12%, head or neck (excluding orbit, where no node-positive cases were reported in IRS-II) in 7%, and trunk in 5%.207,208 The initial IRS studies (IRS-I, -II, and -III) used postoperative disease extent to define stage groups  (Table 37-4).208-210 The postoperative staging of children with ­ rhabdomyosarcoma in these studies indicated that 15% had localized, completely resected tumors (group I) and 18% had metastatic disease (group IV).211,212 One half of all tumors could not be completely resected (group III), and 15% to 20% of patients had microscopic local or regional residual disease after grossly complete resection (group II).211,212 Although these results confirmed the prognostic value of the grouping system, individual tumor factors (i.e., primary tumor site, size, invasiveness, and lymph node metastasis) more accurately predicted outcome.210,213 The current preoperative IRS staging system is therefore based on primary tumor site, tumor extent (local invasiveness and size), and the presence or absence of regional lymph node metastasis or distant metastasis (Table 37-5).208,209 The system is easily simplified by classifying nonmetastatic tumors in favorable sites as stage I, localized tumors up to 5 cm in diameter in other sites as stage II, size larger than 5 cm or the presence of lymph node metastasis as stage III, and hematogenous dissemination as stage IV. The IRS-IV study used stage to determine chemotherapy regimen and group to determine radiation regimen.

Radiation Therapy Central Role in Treatment Therapy for rhabdomyosarcoma includes surgery and chemotherapy for essentially all presentations; irradiation is indicated for microscopic (group II) or overt (group III) locoregional disease and sites of metastatic involvement (group IV). Earlier studies of surgery alone documented limited disease control after radical surgery at selected

B

FIGURE 37-4.  Primary parameningeal rhabdomyosarcoma of the nasopharynx. A, Direct CT Scan with contrast outlines a sizable, complex tumor that fills the nasopharynx and extends through the base of the skull to involve the region of the right cavernous sinus. B, CT Scan at a bone window setting shows the extent of destruction of the right sphenoid.

978

PART 13  Childhood Cancer

TABLE 37-4.   Surgical-Pathologic Grouping System Used in Intergroup Rhabdomyosarcoma Study Group Trials Group

Criteria

I

Localized disease, completely resected: A. Confined to the muscle or organ of origin B. Infiltration outside organ or muscle of origin; regional lymph nodes not involved

II

Compromised or regional resection of three types, including the following: A. Grossly resected tumors with microscopic residual B. Regional disease, completely resected, in which lymph nodes may be involved and/or extension of tumor into an adjacent organ may be present C. Regional disease with involved lymph nodes, macroscopically resected but with ­evidence of microscopic residual disease

III

Incomplete resection or biopsy with macroscopic residual disease

IV

Distant metastases present at diagnosis

From Lawrence W Jr, Anderson JR, Gehan EA, et al. Pretreatment TNM staging of childhood rhabdomyosarcoma: a report of the Intergroup Rhabdomyosarcoma Study Group. Cancer 1997;80:1165-1170.

TABLE 37-5.    Preoperative Staging System Used by the Intergroup Rhabdomyosarcoma Study Group Stage

Sites

Invasiveness*

Size

Lymph Nodes†

Distant Metastases‡

I

Orbit, head and neck (except ­parameningeal)

T1 or T2

a (<5 cm) or b (≥5 cm)

N0 or N1 or N2

M0

T1 or T2

a (<5 cm)

N0 or NX

M0

T1 or T2

a (<5 cm)

N1

M0

b (≥5 cm)

N0 or N1 or NX

a (<5 cm) or b (≥5 cm)

N0 or N1

Genitourinary tract (except bladder or prostate) II

Bladder or prostate Extremity Head and neck parameningeal Other (e.g., trunk, retroperitoneum)

III

Bladder or prostate Extremity Head and neck parameningeal Other (e.g., trunk, retroperitoneum)

IV

All

T1 or T2

M1

*T1,

confined to anatomic site of origin; T2, extension and/or fixation to the surrounding tissue. regional lymph nodes not clinically involved; N1, regional lymph nodes clinically involved by neoplasm; NX, clinical status of regional lymph nodes unknown (especially with sites that preclude lymph node evaluation). ‡M0, no distant metastasis; M1, metastasis present. Modified from Lawrence W Jr, Anderson JR, Gehan EA, et al. Pretreatment TNM staging of childhood rhabdomyosarcoma: a report of the Intergroup Rhabdomyosarcoma Study Group. Cancer 1997;80:1165-1170.

†N0,

sites—a 45% disease-free survival rate for patients with orbital tumors and up to 70% disease-free survival rates for patients with limited bladder lesions outside the trigone.214,215 Therapy in North America almost uniformly follows the guidelines of the IRS protocols. Rhabdomyosarcoma was initially thought to respond poorly to irradiation. Reports by Cassady and colleagues,216 Edland,217 and Nelson218 in the 1960s established the efficacy of radiation therapy, describing local tumor control after wide-field irradiation to dose levels of 50 to 60 Gy. For orbital tumors, primary irradiation achieved local ­ tumor

control in 90% of cases, although survival was limited to 66% because of hematogenous metastases.219 The addition of chemotherapy has improved local tumor control and survival. Jereb and colleagues220 reported a 2-year disease-free survival rate of 82% for patients with localized disease treated with chemotherapy and 47% for patients treated without chemotherapy. Combining chemotherapy and irradiation with less radical surgery has proved feasible in many cases. Initial chemotherapy followed by surgery and response-directed irradiation has proved effective for a significant proportion

37  Childhood Cancer

979

TABLE 37-6.   Protocol-Required Total Radiation Dose for Group III Rhabdomyosarcoma Patients and Corresponding Local Failure Rates Patient Age <6 yr ≥6 yr

Tumor Size

Dose

Local Failure Rate*

<5 cm

40-45 Gy

16%

≥5 cm

45-50 Gy

23%

<5 cm

45-50 Gy

14%

≥5 cm

50-55 Gy

26%

*Special pelvic sites excluded. From Wharam MD, Hanfelt JJ, Tefft MC, 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.

of lesions that are not easily and completely resectable at presentation.221-225 For children with localized rhabdomyosarcoma (group I), the addition of radiation therapy did not improve overall disease control in the initial IRS studies.212,226 The failurefree survival rate with late follow-up of the IRS-I through IRS-III trials was only 79% in a 1999 update from Wolden and colleagues.227 Most group I lesions are relatively ­limited in size, are often located in the extremities, are unusually small bladder lesions, or are located in nonparameningeal sites in the head or neck. Findings regarding local control without irradiation indicated higher rates of local failure for tumors larger than 5 cm, tumors with alveolar or undifferentiated histology, and tumors at sites other than the genitourinary tract. In group I cases with alveolar or undifferentiated histology in the IRS-I and -II trials, the addition of radiation therapy significantly enhanced local tumor control (73% versus 44%) and overall survival (82% versus 52%).227 For embryonal tumors, a trend toward improved failure-free survival was not statistically significant, and no difference in overall survival was apparent.227 In later trials, a reduced radiation dose (36 to 41.4 Gy) has been recommended for children with disease of unfavorable histology or for those with group I disease (i.e., localized, resected disease). For patients with group II disease (i.e., microscopic residual disease only), dose levels of 36 to 40 Gy produce durable local control in more than 90% of cases.198,228-230 For those with group III tumors (i.e., up to two thirds of patients with locoregional presentations), dose levels of 45 Gy (selectively used in children younger than 6 years with tumors smaller than 5 cm) to 50 Gy (in ­conventional fractions, typically 1.8 Gy, occasionally reduced to 1.5 Gy, once daily) produce an 80% likelihood of local tumor control (Table 37-6).231 The rate of locoregional failure is 15% for tumors originating within the head and neck or genitourinary region and no larger than 10 cm. For lesions originating in the chest, pelvis, trunk, or extremities or for primary lesions larger than 10 cm at other, more favorable sites, the rate of locoregional failure is significantly higher, at 35%.231 The importance of radiation therapy is apparent from the high rate of local recurrence among the most favorable orbital tumors when the primary treatment is chemotherapy. The SIOP experience indicates a local failure rate of 45% without irradiation compared with 10% with irradiation in the IRS experience.231-233 A comparative analysis of parameningeal rhabdomyosarcoma trials from the IRS,

SIOP, and German and Italian cooperative studies indicated significantly superior 5-year event-free survival and overall survival rates in the IRS experience (71% eventfree survival and 74% overall survival), in which radiation therapy (median dose, 53 Gy) was used systematically, compared with the SIOP experience (36% event-free survival and 55% overall survival), in which irradiation was limited to children older than 5 years with signs of intracranial extension, or the German (47% and 47%, respectively) or Italian (39% and 39%, respectively) experiences, in which radiation therapy was delayed for 3 months or limited to a median dose of 32 to 46 Gy with no central quality ­control.232 Orbital rhabdomyosarcoma is approached with biopsy and irradiation as group III disease; a minority of patients (group II) undergo initial resection, usually short of exenteration, and irradiation. The target volume typically includes the entire orbit, although more limited radiation therapy may be appropriate for anterior lesions. Disease control and survival rates exceed 90% with ­ approaches that include irradiation as local therapy.231,232,234 More than 80% of patients develop cataracts after radiation therapy for orbital rhabdomyosarcoma, and reduced vision has been documented in 70% of cases.235 Surgery for symptomatic sequelae of primary radiation therapy or disease recurrence is required in approximately 15% of cases with long-term follow-up.235 Xerophthalmia occurs less often than might be expected, but posttherapy orbital hypoplasia has been documented in most surviving children.235 Parameningeal rhabdomyosarcoma often demonstrates overt extension into the base of the skull, manifesting as cranial nerve palsy, base-of-skull erosion on CT or MRI, or direct intracranial extension.61,236,237 The initial IRS experience demonstrating a 35% rate of meningeal (largely extradural rather than subarachnoid) extension has subsequently been reduced to rates of less than 15% or 20% for local or marginal failures with wide local radiation volumes (including margins of 2 cm above the skull base for signs of cranial nerve or bone erosion, and wider margins around sites of overt intracranial extension on neuroimaging).61,231,237,238 Tumors at other head and neck sites have a very favorable outcome with primary irradiation or irradiation combined with judicious surgery.231,239 For bladder and prostate rhabdomyosarcoma, therapy in earlier IRS studies involved a change from primary surgery to primary chemotherapy and selected, delayed irradiation, resulting in disease control with bladder preservation in

980

PART 13  Childhood Cancer

only 25% of children in the IRS-I and -II trials.222-224,240 Subsequent experience with early use of local irradiation (usually about 45 Gy at week 20) in the IRS-III trial has achieved 80% disease-free survival and 60% preservation of bladder control.240 Use of partial cystectomy for ­lesions away from the bladder trigone, often in conjunction with radiation therapy, has been particularly successful.222 ­Combined-modality therapy has produced favorable functional outcomes.241 Paratesticular rhabdomyosarcoma has been among the most favorable presentations, although in recent IRS trials, the use of less aggressive chemotherapy when ­assessment of para-aortic nodes has been done only with imaging ­resulted in a higher rate of failure.61,201,242 The less common vaginal rhabdomyosarcoma has been successfully managed with brachytherapy.243 Rhabdomyosarcoma originating in the extremities and trunk (including thorax, abdomen or retroperitoneum, pelvis, and perineum) is common in adolescents and often is of alveolar histology. Treatment usually includes induction chemotherapy followed by primary irradiation or delayed resection, the latter best followed by dose-adjusted irradiation depending on the extent of ­resection223,228,230,231,244,245

Technique Radiation therapy can be initiated at diagnosis or after induction chemotherapy. No evidence exists to suggest that delaying radiation therapy for up to 4 months is deleterious except for tumors at parameningeal sites. For patients with intracranial extension of parameningeal tumors, ­radiation therapy should be started at the very beginning of ­combined-modality therapy. The radiation technique used should ensure broad coverage of the primary tumor, with 2-cm margins beyond the preoperative, prechemotherapy tumor volume in most instances. For tumors in areas where abdominal or pelvic viscera present problems with radiation tolerance, margins may be limited to 1 cm. For tumors at parameningeal sites, it is important to identify 2-cm dosimetric margins in all three dimensions. The advantage of three-dimensional conformal techniques is just beginning to be apparent.246 Treatment of regional lymph nodes is indicated ­primarily for patients with proven nodal involvement; however, the high incidence of lymph node metastasis in extremity ­tumors of alveolar or undifferentiated histology warrants consideration of adjacent lymph node irradiation in this setting.231,247 Proton therapy has been shown to be ­feasible for the treatment of rhabdomyosarcoma and has been shown to have dosimetric advantages over photon therapy for sparing normal tissues.248,249 The established dose level of 40 Gy for microscopic residual disease (IRS group II) provides local tumor control in 95% to 98% of cases.250,251 However, limited data exist to suggest that doses less than 40 Gy may be adequate for microscopic residual disease after initial surgery.229 ­Current protocols used by the Children’s Oncology Group involve delivering 36 Gy to microscopic disease at the primary ­tumor site and 41.4 Gy to nodal microscopic residual (N1) disease. For macroscopic residual disease, the IRS reports indicate that 50.4 Gy produces excellent disease control in most settings; however, review data substantiate a higher

local failure rate for tumors that are large or of unfavorable histologic types, findings that argue for consideration of higher dose levels in such settings.231 Hyperfractionated irradiation to dose levels of 59.4 Gy (0.11 Gy twice daily) has produced outcomes rather similar to those with conventionally fractionated irradiation, although the final results of the IRS-V randomized trial testing hyperfractionated delivery are not yet available.230,252 Flamant and colleagues243 reported disease control and excellent functional preservation in 15 of 17 patients with vaginal rhabdomyosarcoma treated with an individualized technique of intracavitary irradiation.

Prognosis The IRS trials have resulted in continual improvements in outcome, with 5-year survival rates for patients with ­rhabdomyosarcoma increasing from 56% (IRS-I) to 63% (IRS-II) to 72% (IRS-III).198,231 Outcome depends on ­tumor site and histology, extent of local disease, and lymph node involvement.

Retinoblastoma Retinoblastoma is a malignant embryonal neoplasm arising in the retina during fetal life or early childhood. The tumor is unilateral in two thirds of patients and bilateral in one third of patients. More than 90% of these tumors occur in children younger than 4 years; the median age is 8 months for patients with bilateral retinoblastoma and 26 months for those with unilateral disease. Diagnosis in children typically is made on the basis of leukokoria (the white eye reflex, in which the retinal tumor reflects light), strabismus, or glaucoma. Ten percent of patients have a family history of retinoblastoma. Retinoblastoma is the classic genetically inherited pediatric cancer; genetic studies of the disease have confirmed Knudson’s two-hit hypothesis of cancer development and have provided the evidence for the discovery of cancer suppressor genes.253 Retinoblastoma results from the loss or mutation of both copies of the RB1 gene, ­located at 13q14.55,254,255 Retinoblastoma arises from the inner nuclear layer of the retina. Histologically, the tumor comprises small, uniform neuroectodermal cells with various degrees of photoreceptor differentiation. Tumor cells grow through the retinal layers, often infiltrating into the surrounding choroid or the vitreous. Lesions with pronounced late choroidal involvement may extend extrasclerally along the emissary vessels. Tumor cells gain access to the optic nerve by direct infiltration at the optic disk or by extension along the central ­vessels. Lesions that involve the optic nerve can be associated with diffuse subarachnoid involvement, an uncommon late manifestation of the disease. Even with negative surgical margins, however, tumors that infiltrate through the lamina cribrosa (i.e., into the optic nerve proper) have been associated with an increased risk of orbital recurrence or metastatic disease.256 Staging for retinoblastoma has classically been based on the Reese-Ellsworth system (Box 37-1).257 Stage groupings are defined by the number, size, and location of intraocular ­tumors and indicate the suitability of a tumor for less-than­exenterative treatment (i.e., radiation therapy, cryotherapy, or ­photocoagulation).

37  Childhood Cancer

BOX 37-1.  Reese-Ellsworth Staging System for Retinoblastoma Group I (very favorable) a. Solitary tumor, less than 4 dd,* at or behind the ­equator b. Multiple tumors, none over 4 dd, all at or behind the equator Group II (favorable) a. Solitary lesion, 4-10 dd, at or behind the equator b. Multiple tumors, 4-10 dd, behind the equator Group III (doubtful) a. Any lesion anterior to the equator b. Solitary lesions larger than 10 dd behind the equator Group IV (unfavorable) a. Multiple tumors, some larger than 10 dd b. Any lesion extending anteriorly to the ora serrata Group V (very unfavorable) a. Massive tumors involving over one half of the retina b. Vitreous seeding

Contact lens assembly Scale (mm)

Rod

981

Linear Beam accelerator 3-cm

Collimator Adjustable Applicator scatterer

FIGURE 37-5.  Diagram of a treatment approach for retinoblastoma involving opposed lateral fields, closely calibrated in this example to identify the posterior margin of the optic lens, with a fixed correlation between the central ray of the photon beam (defining the anterior coronal plane) and the anterior position of the eye (affixed to the head of the accelerator by a suction contact). Current approaches include opposed lateral beams such as those shown and three-dimensional conformal configurations. (Technique described in Schipper J, Tan KE, van Peperzeel HA. Treatment of retinoblastoma by precision megavoltage radiation therapy. Radiother Oncol 1985;3:117-132.)

*One disk diameter (dd) = 1.6 mm.

From Reese AB. Tumors of the Eye, 3rd ed. Hagerstown, MD: Harper & Row, 1976.

Radiation Therapy Advanced Ocular Disease Treatment for retinoblastoma has been determined largely by the presence or absence of advanced ocular disease at diagnosis. Enucleation alone is appropriate for many unilateral tumors, which are often sufficiently advanced to obviate consideration of eye-preserving local therapies. For children older than 2 or 3 years in whom the genetic form of disease can reasonably be excluded if the presentation is unilateral, focal therapy alone can be considered (more often photocoagulation or cryotherapy, but also brachytherapy or, in some institutions, focal external beam radiation therapy [EBRT]).258 The use of standard, full ocular radiation techniques for sporadic cases of unilateral disease has been documented; the combination of chemotherapy and more focal irradiation in this setting has been used ­infrequently.259 In cases of bilateral retinoblastoma, at least one eye has group V disease in more than 90% of patients. The principle of therapy is to preserve the eye if there is a reasonable expectation that functional vision may result. In practice, most children with bilateral disease begin treatment with chemotherapy, followed by local measures including focal surgical approaches (e.g., cryotherapy for anterior lesions or photocoagulation for more posterior areas) and irradiation focal brachytherapy plaques or EBRT, depending on the extent of disease after chemotherapy).260-263 EBRT has been used with considerable success for moderate to advanced intraocular disease; findings over the past 4 decades have documented disease control in 70% to 80% of cases with group I to III disease. Cassady and colleagues264 reported a 73% disease control rate, with good to fair vision, among 120 patients treated with EBRT between 1954 and 1963; orthovoltage and megavoltage (22.5-MV) x-ray techniques were used during this period. In the 1980s, Schipper reported from Utrecht that

tumor was controlled and vision preserved in 33 (94%) of 35 patients with group I to III disease by using an anatomically precise lateral linear accelerator technique with ocular fixation and a narrowly defined target volume  (Fig. 37-5).265 For more advanced ocular disease, Schipper and colleagues266 described ocular preservation in 11 (79%) of 14 patients with group IV disease but 0 of 5 patients with group V disease. The Utrecht irradiation technique has also achieved disease control in a limited number of cases involving vitreous seeding and otherwise limited tumor extent.266,267 The lower integral dose of proton beam therapy relative to photon therapy may reduce the risks of second malignancy and cosmetic and functional sequelae.268,269 The use of chemotherapy has resulted in primary ­tumor control in only a very limited subset of cases but has achieved substantial reductions in the extent of ocular disease, facilitating apparently successful use of local treatment modalities in most cases.260-263 Objective responses to alkylating agents (especially cyclophosphamide, ifosfamide, and cisplatin), etoposide, vincristine, and doxorubicin have been documented.261,270 One goal of initial ­chemotherapy approaches has been to limit the use of EBRT because of the relatively high rate of secondary cancer in children with the genetic form of disease (bilateral or multifocal disease) and late orbital deformities after irradiation in this very young population.10,45,184,261-263 More focal types of ­therapy, such as focal surgical or radiotherapeutic interventions, have shown excellent rates of disease control in ocular disease without vitreous seeding.10,184,259,261-263 For advanced cases with massive intraocular or vitreous disease, initial chemotherapy followed by EBRT for those who respond to the chemotherapy has allowed ocular preservation in some cases that would otherwise have required enucleation.260-263,271 Close ophthalmologic observation during initial chemotherapy is critical to maximize local “salvage” approaches. Local irradiation techniques have included cobalt 60 (60Co) applicators and the use of iridium 192 (192Ir) or iodine 125 (125I). The high rate of recurrence after the use

982

PART 13  Childhood Cancer

of only focal radiation therapy had limited the enthusiasm for brachytherapy for cases with bilateral presentations, but renewed interest in this modality has been shown in conjunction with chemotherapy.258 When irradiation is used as the primary therapy, cryotherapy or photocoagulation may be needed for local sites of apparent disease persistence or as adjuvant therapy for tumors extending anteriorly within the globe.266 Visual acuity after irradiation depends on tumor size and location. In 22 eyes with tumor control, Egbert and colleagues272 reported visual acuity of better than 20/40 in 5, up to 20/100 in 5, 20/200 to hand movements in 9, and no vision in 3 eyes. Poor visual results correlate primarily with tumor involvement of the macular region.258,272 ­Deterioration in visual status and the need for enucleation can occasionally result from late effects of therapy rather than tumor recurrence; late effects include vitreous ­ hemorrhage, retinopathy, and ischemic optic nerve ­neuropathy.258,272 Orbital irradiation is indicated postoperatively for documented scleral extension of tumor, extraocular disease, or tumor infiltration of the optic nerve beyond the lamina cribrosa.273 Cranial irradiation or CSI has been used for patients with microscopic extension of retinoblastoma to the cut end of the optic nerve or CNS involvement at diagnosis. Long-term disease control in this setting has been rare.57,270

Technique The anterior field margin for lateral photon irradiation is placed just behind the lens; in practice, this is 7 to 9 mm posterior to the anterior surface of the eye.265 With threedimensional conformal radiation ­ approaches, multifield, non-coplanar configurations may offer an ­ advantage in full retinal disease control, with relative sparing of the optic lens and, in most circumstances, with avoiding irradiation to the contralateral eye when appropriate.256 Use of a single anterior electron field or multiple oblique electron fields may achieve uniform retinal irradiation, with rates of later cataractogenesis and ocular effects similar to those produced by the three-­dimensional  approaches.256,274 Sedation is usually necessary to ensure the accuracy of daily positioning and immobilization. Schipper265 has described an exquisite radiation therapy technique in which a globe-fixing contact lens mechanically attached to the linear accelerator gantry is used to ensure ocular immobilization and proper field placement on a daily basis. No conclusive radiation dose-response data have been published for retinoblastoma. The report of Cassady and colleagues264 of the early Columbia experience showed equal tumor control rates with doses of 35 to 40 Gy and doses of more than 40 Gy delivered in hypofractionated regimens (3.33- to 4-Gy fractions delivered three times weekly). Schipper and colleagues266 also used 3 fractions per week, delivering 45 Gy in 15 treatments. The rates of tumor control and visual preservation seem to approach a balance at 40 to 50 Gy, suggesting that a fractionated regimen of 2 Gy, given four to five times weekly to a total dose of 40 to 45 Gy, is effective. Whether dose levels can be successfully reduced in the setting of adjuvant or, more often, salvage irradiation after initial response to or progression after chemotherapy remains uncertain.260,261

Prognosis The survival rate after treatment of retinoblastoma is excellent, with more than 90% survival rates documented for all stages.259-262,266,275,276 Even among patients with advanced (group V) disease, survival rates of 70% to 90% have been reported in long-term series from North America and  Europe.57,277 Advanced, often extraocular extension of disease at diagnosis limits results in developing countries to one half of that level.278 No evidence exists to suggest that conservative primary therapy has increased the risk of subsequent hematogenous metastasis or death. Overall survival rates for patients with advanced bilateral disease are virtually identical after ­bilateral enucleation (92%) and initial bilateral irradiation with delayed enucleation for disease recurrence (88%).277 Death from retinoblastoma in patients with group I to IV disease has been rare.260-262

Liver Neoplasms Primary tumors of the liver are rare in children. Hepatoblastoma is the most common malignant liver neoplasm in children up to age 5 years, after which hepatocellular carcinoma becomes the most common. Hepatoblastoma is a tumor type that occurs as part of the Beckwith-­Wiedemann syndrome.279 Hepatoblastoma is often a unifocal tumor. Histologically, it is identified as an epithelial tumor (with fetal or embryonal components) or a mixed lesion (also including mesenchymal components).280 Hepatocellular carcinoma occurs in children older than 5 years; as in adults, it is associated with hepatitis B surface antigen.281 Hepatoblastoma and hepatocellular carcinoma manifest as abdominal masses, usually associated with ­elevated serum alpha-fetoprotein levels. Hepatic tumors have been managed primarily by surgical resection, with major series showing 50% to 65% of such tumors to be potentially resectable at diagnosis.282-284   Among 48 cases reported by Giacomantonio and colleagues,283 16 patients underwent gross surgical resection; 11 of these 16 patients survived without recurrence: 7 of 9 after surgery alone and 4 of 7 after preoperative chemotherapy and tumor removal. Operative mortality rates are approximately 2% to 10%.283-285 Preoperative irradiation has been described anecdotally; studies published in the late 1980s and early 1990s282,286,287 have emphasized tumor regression with chemotherapy (e.g., dactinomycin, cyclophosphamide, vincristine, and fluorouracil; and more recently, doxorubicin and cisplatin) before attempted resection. Among patients with residual disease after surgery, responses to radiation therapy or chemotherapy, or both, have been documented, and a suggestion in the series by Habrand and colleagues288 was made that radiation doses of 24 to 45 Gy can successfully control limited postoperative ­residual disease.

REFERENCES    1. Donaldson SS. Patterns of failure in childhood solid tumors. Cancer Treat Symp 1983;2:267-283.    2. Li FP, Stone R. Survivors of cancer in childhood. Ann Intern Med 1976;84:551-553.

37  Childhood Cancer    3. Garre ML, Gandus S, Cesana B, et al. Health status of long-term survivors after cancer in childhood. Results of an uni-institutional study in Italy. Am J Pediatr Hematol Oncol 1994;16:143-152.    4. Thomas PR, Griffith KD, Fineberg BB, et al. Late effects of treatment for Wilms’ tumor. Int J Radiat Oncol Biol Phys 1983;9:  651-657.    5. Taylor RE. Morbidity from abdominal radiotherapy in the First United Kingdom Children’s Cancer Study Group Wilms’ ­Tumour Study. United Kingdom Children’s Cancer Study Group. Clin Oncol (R Coll Radiol) 1997;9:381-384.    6. Jereb B, Burgers JM, Tournade MF, et al. Radiotherapy in the SIOP (International Society of Pediatric Oncology) nephroblastoma studies: a review. Med Pediatr Oncol 1994;22:221-227.    7. Riseborough EJ, Grabias SL, Burton RI, et al. Skeletal alterations following irradiation for Wilms’ tumor: with particular reference to scoliosis and kyphosis. J Bone Joint Surg Am 1976;58:  526-536.    8. Probert JC, Parker BR. The effects of radiation therapy on bone growth. Radiology 1975;114:155-162.    9. Oliver JH, Gluck G, Gledhill RB, et al. Musculoskeletal deformities following treatment of Wilms’ tumour. Can Med Assoc J 1978;119:459-464.   10. Kaste SC, Chen G, Fontanesi J, et al. Orbital development in long-term survivors of retinoblastoma. J Clin Oncol 1997;15:  1183-1189.   11. Shalet SM, Beardwell CG, Morris-Jones PH, et al. Pituitary function after treatment of intracranial tumours in children. Lancet 1975;2:104-107.   12. Shalet SM, Beardwell CG, Aarons BM, et al. Growth impairment in children treated for brain tumours. Arch Dis Child 1978;53:491-494.   13. Swift PG, Kearney PJ, Dalton RG, et al. Growth and hormonal status of children treated for acute lymphoblastic leukaemia. Arch Dis Child 1978;53:890-894.   14. Lustig RH, Rose SR, Burghen GA, et al. Hypothalamic obesity caused by cranial insult in children: altered glucose and insulin dynamics and reversal by a somatostatin agonist. J Pediatr 1999;135:162-168.   15. Shalet SM, Beardwell CG, Twomey JA, et al. Endocrine function following the treatment of acute leukemia in childhood.  J Pediatr 1977;90:920-923.   16. Kaleita TA, Reaman GH, MacLean WE, et al. Neurodevelopmental outcome of infants with acute lymphoblastic leukemia: a Children’s Cancer Group report. Cancer 1999;85:1859-1865.   17. Samaan NA, Vieto R, Schultz PN, et al. Hypothalamic, pituitary and thyroid dysfunction after radiotherapy to the head and neck. Int J Radiat Oncol Biol Phys 1982;8:1857-1867.   18. Robison LL, Nesbit ME, Sather HN. Thyroid abnormalities in long-term survivors of childhood acute lymphoblastic leukemia [abstract]. Pediatr Res 1985;19:266.   19. Oberfield SE, Allen JC, Pollack J, et al. Long-term endocrine ­sequelae after treatment of medulloblastoma: prospective study of growth and thyroid function. J Pediatr 1986;108:219-223.   20. Constine LS, Donaldson SS, McDougall IR, et al. Thyroid dysfunction after radiotherapy in children with Hodgkin’s disease. Cancer 1984;53:878-883.   21. Merchant TE, Kushner BH, Sheldon JM, et al. Effect of lowdose radiation therapy when combined with surgical resection for Ewing sarcoma. Med Pediatr Oncol 1999;33:65-70.   22. Wohl ME, Griscom NT, Traggis DG, et al. Effects of therapeutic irradiation delivered in early childhood upon subsequent lung function. Pediatrics 1975;55:507-516.   23. Jakacki RI, Schramm CM, Donahue BR, et al. Restrictive lung disease following treatment for malignant brain tumors: a ­potential late effect of craniospinal irradiation. J Clin Oncol 1995;13:1478-1485.   24. Freeman JE, Johnston PG, Voke JM. Somnolence after prophylactic cranial irradiation in children with acute lymphoblastic leukaemia. Br Med J 1973;4:523-525.   25. Bleyer WA. Neurologic sequelae of methotrexate and ionizing radiation: a new classification. Cancer Treat Rep 1981;65:8998.

983

  26. Price RA, Birdwell DA. The central nervous system in childhood leukemia. Mineralizing microangiopathy and dystrophic calcification. Cancer 1978;42(Pt 3):717-728.   27. Peylan-Ramu N, Poplack DG, Pizzo PA, et al. Abnormal CT scans of the brain in asymptomatic children with acute lymphocytic leukemia after prophylactic treatment of the central nervous system with radiation and intrathecal chemotherapy.  N Engl J Med 1978;298:815-818.   28. Riccardi R, Brouwers P, Di Chiro G, et al. Abnormal computed tomography brain scans in children with acute lymphoblastic leukemia: serial long-term follow-up. J Clin Oncol 1985;3:  12-18.   29. Ochs JJ, Parvey LS, Whitaker JN, et al. Serial cranial ­computedtomography scans in children with leukemia given two different forms of central nervous system therapy. J Clin Oncol 1983;1:793-798.   30. Price RA, Jamieson PA. The central nervous system in childhood leukemia. Subacute leukoencephalopathy. Cancer 1975;  35(Pt 2):306-318.   31. Aur RJ, Simone JV, Verzosa MS, et al. Childhood acute lymphocytic leukemia: study VIII. Cancer 1978;42:2123-2134.   32. Bleyer WA, Griffin TW. White matter necrosis, mineralizing microangiopathy, and intellectual abilities in survivors of childhood leukemia: associations with CNS irradiation and metho­ trexate therapy. In Gilbert HA, Kagan AR (eds): Radiation ­Damage to the Nervous System. New York: Raven Press, 1980, pp 155-174.   33. Eiser C. Intellectual abilities among survivors of childhood leukaemia as a function of CNS irradiation. Arch Dis Child 1978;53:391-395.   34. Meadows AT, Gordon J, Massari DJ, et al. Declines in IQ scores and cognitive dysfunctions in children with acute lymphocytic leukaemia treated with cranial irradiation. Lancet 1981;2:  1015-1018.   35. Rowland JH, Glidewell OJ, Sibley RF, et al. Effects of different forms of central nervous system prophylaxis on neuropsychologic function in childhood leukemia. J Clin Oncol 1984;2:  1327-1335.   36. Williams JM, Davis KS. Central nervous system prophylactic treatment for childhood leukemia: neuropsychological outcome studies. Cancer Treat Rev 1986;13:113-127.   37. Whitt JK, Wells RJ, Lauria MM, et al. Cranial radiation in childhood acute lymphocytic leukemia. Neuropsychologic sequelae. Am J Dis Child 1984;138:730-736.   38. Jannoun L. Are cognitive and educational development affected by age at which prophylactic therapy is given in acute lympho­ blastic leukaemia? Arch Dis Child 1983;58:953-958.   39. Walter AW, Hancock ML, Pui CH, et al. Secondary brain ­tumors in children treated for acute lymphoblastic leukemia at St Jude Children’s Research Hospital. J Clin Oncol 1998;16:  3761-3767.   40. Li FP, Gimbrere K, Gelber RD. Adverse pregnancy outcome ­after radiotherapy for childhood Wilms’ tumors [abstract]. Proc Am Soc Clin Oncol 1986;5:202.   41. Li FP, Fine W, Jaffe N, et al. Offspring of patients treated for cancer in childhood. J Natl Cancer Inst 1979;62:1193-1197.   42. Green DM, Fine WE, Li FP. Offspring of patients treated for unilateral Wilms’ tumor in childhood. Cancer 1982;49:2285-2288.   43. Li FP, Cassady JR, Jaffe N. Risk of second tumors in survivors of childhood cancer. Cancer 1975;35:1230-1235.   44. Hudson MM, Jones D, Boyett J, et al. Late mortality of longterm survivors of childhood cancer. J Clin Oncol 1997;15:  2205-2213.   45. Wong FL, Boice JD Jr, Abramson DH, et al. Cancer incidence after retinoblastoma. Radiation dose and sarcoma risk. JAMA 1997;278:1262-1267.   46. Beaty O III, Hudson MM, Greenwald C, et al. Subsequent ­malignancies in children and adolescents after treatment for Hodgkin’s disease. J Clin Oncol 1995;13:603-609.   47. Tucker MA, D’Angio GJ, Boice JD Jr, et al. Bone sarcomas linked to radiotherapy and chemotherapy in children. N Engl J Med 1987;317:588-593.

984

PART 13  Childhood Cancer

  48. Kuttesch JF Jr, Wexler LH, Marcus RB, et al. Second malignancies after Ewing’s sarcoma: radiation dose-dependency of secondary sarcomas. J Clin Oncol 1996;14:2818-2825.   49. Garber JE, Goldstein AM, Kantor AF, et al. Follow-up study of twenty-four families with Li-Fraumeni syndrome. Cancer Res 1991;51:6094-6097.   50. Riccardi VM. Von Recklinghausen neurofibromatosis. N Engl J Med 1981;305:1617-1627.   51. Listernick R, Charrow J, Greenwald M, et al. Natural history of optic pathway tumors in children with neurofibromatosis type 1: a longitudinal study. J Pediatr 1994;125:63-66.   52. Matsui I, Tanimura M, Kobayashi N, et al. Neurofibromatosis type 1 and childhood cancer. Cancer 1993;72:2746-2754.   53. Kraemer KH, Lee MM, Scotto J. Xeroderma pigmentosum. ­Cutaneous, ocular, and neurologic abnormalities in 830 published cases. Arch Dermatol 1987;123:241-250.   54. Swift M, Morrell D, Massey RB, et al. Incidence of cancer in 161 families affected by ataxia-telangiectasia. N Engl J Med 1991;325:1831-1836.   55. Eng C, Li FP, Abramson DH, et al. Mortality from second tumors among long-term survivors of retinoblastoma. J Natl Cancer Inst 1993;85:1121-1128.   56. Abramson DH, Ellsworth RM, Zimmerman LE. Nonocular cancer in retinoblastoma survivors. Transact Am Acad Ophthalmol Otolaryngol 1976;81:454-457.   57. Rubin CM, Robison LL, Cameron JD, et al. Intraocular retinoblastoma group V: an analysis of prognostic factors. J Clin Oncol 1985;3:680-685.   58. Lueder GT, Judisch F, O’Gorman TW. Second nonocular ­tumors in survivors of heritable retinoblastoma. Arch Ophthalmol 1986;104:372-373.   59. Abramson DH, Ellsworth RM, Kitchin FD, et al. Second nonocular tumors in retinoblastoma survivors. Are they radiationinduced? Ophthalmol 1984;91:1351-1355.   60. Sagerman RH, Cassady JR, Tretter P, et al. Radiation induced neoplasia following external beam therapy for children with retinoblastoma. Am J Roentgenol Radium Ther Nucl Med 1969;105:529-535.   61. Raney RB Jr, Tefft M, Newton WA, et al. Improved prognosis with intensive treatment of children with cranial soft tissue ­sarcomas arising in nonorbital parameningeal sites. A report from the Intergroup Rhabdomyosarcoma Study. Cancer 1987;59:  147-155.   62. Pui CH. Childhood leukemias. N Engl J Med 1995;332:  1618-1630.   63. Crist WM, Grossi CE, Pullen DJ, et al. Immunologic ­ markers in childhood acute lymphocytic leukemia. Semin Oncol 1985;12:105-121.   64. Pui CH, Behm FG, Crist WM. Clinical and biologic relevance of immunologic marker studies in childhood acute lymphoblastic leukemia. Blood 1993;82:343-362.   65. Reiter A, Schrappe M, Ludwig WD, et al. Chemotherapy in 998 unselected childhood acute lymphoblastic leukemia patients. Results and conclusions of the multicenter trial ALL-BFM 86. Blood 1994;84:3122-3133.   66. Schorin MA, Blattner S, Gelber RD, et al. Treatment of childhood acute lymphoblastic leukemia: results of Dana-Farber Cancer Institute/Children’s Hospital Acute Lymphoblastic Leukemia Consortium Protocol 85-01. J Clin Oncol 1994;12:740-747.   67. Rubnitz JE, Downing JR, Pui CH, et al. TEL gene rearrangement in acute lymphoblastic leukemia: a new genetic marker with prognostic significance. J Clin Oncol 1997;15:1150-1157.   68. Ribeiro RC, Broniscer A, Rivera GK, et al. Philadelphia ­chromosome-positive acute lymphoblastic leukemia in children: durable responses to chemotherapy associated with low initial white blood cell counts. Leukemia 1997;11:1493-1496.   69. Behm FG, Raimondi SC, Frestedt JL, et al. Rearrangement of the MLL gene confers a poor prognosis in childhood acute lymphoblastic leukemia, regardless of presenting age. Blood 1996;87:2870-2877.   70. Pui CH, Kane JR, Crist WM. Biology and treatment of infant leukemias. Leukemia 1995;9:762-769.

  71. Reaman G, Zeltzer P, Bleyer WA, et al. Acute lymphoblastic leukemia in infants less than one year of age: a cumulative ­experience of the Children’s Cancer Study Group. J Clin Oncol 1985;3:1513-1521.   72. Gajjar A, Ribeiro R, Hancock ML, et al. Persistence of circulating blasts after 1 week of multiagent chemotherapy confers a poor prognosis in childhood acute lymphoblastic leukemia. Blood 1995;86:1292-1295.   73. Cave H, van der Werff ten Bosch J, Suciu S, et al. Clinical significance of minimal residual disease in childhood acute lymphoblastic leukemia. European Organization for Research and ­ Treatment of Cancer—Childhood Leukemia Cooperative Group. N Engl J Med 1998;339:591-598.   74. Wells RJ, Woods WG, Buckley JD, et al. Treatment of newly diagnosed children and adolescents with acute myeloid leukemia: a Children’s Cancer Group study. J Clin Oncol 1994;12:  2367-2377.   75. Creutzig U, Ritter J, Riehm H, et al. Improved treatment ­results in childhood acute myelogenous leukemia: a report of the German cooperative study AML-BFM-78. Blood 1985;65:  298-304.   76. Aur RJ, Simone JV, Hustu HO, et al. Cessation of therapy during complete remission of childhood acute lymphocytic leukemia.  N Engl J Med 1974;291:1230-1234.   77. Aur RJ, Hustu HO, Verzosa MS, et al. Comparison of two methods of preventing central nervous system leukemia. Blood 1973;42:349-357.   78. Pinkel D. Patterns of failure in acute lymphocytic leukemia. Cancer Treat Symp 1983;2:259-266.   79. Hustu HO, Aur RJ. Extramedullary leukaemia. Clin Haematol 1978;7:313-337.   80. Bleyer WA, Poplack DG. Prophylaxis and treatment of leukemia in the central nervous system and other sanctuaries. Semin ­Oncol 1985;12:131-148.   81. Sullivan MP, Chen T, Dyment PG, et al. Equivalence of intrathecal chemotherapy and radiotherapy as central nervous system prophylaxis in children with acute lymphatic leukemia: a Pediatric Oncology Group study. Blood 1982;60:948-958.   82. Laver JH, Barredo JC, Amylon M, et al. Effects of cranial radiation in children with high risk T cell acute lymphoblastic leukemia: a Pediatric Oncology Group report. Leukemia 2000;14:369-373.   83. Pui CH, Mahmoud HH, Rivera GK, et al. Early intensification of intrathecal chemotherapy virtually eliminates central nervous system relapse in children with acute lymphoblastic leukemia. Blood 1998;92:411-415.   84. Conter V, Schrappe M, Arico M, et al. Role of cranial radiotherapy for childhood T-cell acute lymphoblastic leukemia with high WBC count and good response to prednisone. Associazione Italiana Ematologia Oncologia Pediatrica and the Berlin-­FrankfurtMunster groups. J Clin Oncol 1997;15:2786-2791.   85. Nachman J, Sather HN, Cherlow JM, et al. Response of children with high-risk acute lymphoblastic leukemia treated with and without cranial irradiation: a report from the Children’s Cancer Group. J Clin Oncol 1998;16:920-930.   86. Freeman AI, Boyett JM, Glicksman AS, et al. Intermediate-dose methotrexate versus cranial irradiation in childhood acute lymphoblastic leukemia: a ten-year follow-up. Med Pediatr Oncol 1997;28:98-107.   87. Kamps WA, Bokkerink JP, Hahlen K, et al. Intensive treatment of children with acute lymphoblastic leukemia according to ALLBFM-86 without cranial radiotherapy: results of Dutch Childhood Leukemia Study Group Protocol ALL-7 (1988-1991). Blood 1999;94:1226-1236.   88. Schrappe M, Reiter A, Henze G, et al. Prevention of CNS ­recurrence in childhood ALL: results with reduced radiotherapy combined with CNS-directed chemotherapy in four consecutive ALL-BFM trials. Klin Padiatr 1998;210:192-199.   89. Mahoney DH Jr., Shuster JJ, Nitschke R, et al. Acute neurotoxicity in children with B-precursor acute lymphoid leukemia: an ­association with intermediate-dose intravenous methotrexate and intrathecal triple therapy—a Pediatric Oncology Group study.  J Clin Oncol 1998;16:1712-1722.

37  Childhood Cancer   90. Pui CH, Boyett JM, Rivera GK, et al. Long-term results of Total Therapy studies 11, 12 and 13A for childhood acute lymphoblastic leukemia at St Jude Children’s Research Hospital. Leukemia 2000;14:2286-2294.   91. Rivera GK, Mauer AM. Controversies in the management of childhood acute lymphoblastic leukemia: treatment intensification, CNS leukemia, and prognostic factors. Semin Hematol 1987;24:12-26.   92. Mahmoud HH, Rivera GK, Hancock ML, et al. Low leukocyte counts with blast cells in cerebrospinal fluid of children with newly diagnosed acute lymphoblastic leukemia. N Engl J Med 1993;329:314-319.   93. Cherlow JM, Sather H, Steinherz P, et al. Craniospinal irradiation for acute lymphoblastic leukemia with central nervous system disease at diagnosis: a report from the Children’s Cancer Group. Int J Radiat Oncol Biol Phys 1996;36:19-27.   94. Kun LE. CNS disease at diagnosis: a continuing challenge in childhood lymphoblastic leukemia. Int J Radiat Oncol Biol Phys 1996;36:257-259.   95. Gilchrist GS, Tubergen DG, Sather HN, et al. Low numbers of CSF blasts at diagnosis do not predict for the development of CNS leukemia in children with intermediate-risk acute lymphoblastic leukemia: a Children’s Cancer Group report. J Clin Oncol 1994;12:2594-2600.   96. Kun LE, Camitta BM, Mulhern RK, et al. Treatment of meningeal relapse in childhood acute lymphoblastic leukemia. I. ­Results of craniospinal irradiation. J Clin Oncol 1984;2:359-364.   97. Kumar P, Mulhern RK, Regine WF, et al. A prospective neurocognitive evaluation of children treated with additional chemotherapy and craniospinal irradiation following isolated central nervous system relapse in acute lymphoblastic leukemia. Int J Radiat Oncol Biol Phys 1995;31:561-566.   98. Ribeiro RC, Rivera GK, Hudson M, et al. An intensive re-treatment protocol for children with an isolated CNS relapse of acute lymphoblastic leukemia. J Clin Oncol 1995;13:333-338.   99. Winick NJ, Smith SD, Shuster J, et al. Treatment of CNS relapse in children with acute lymphoblastic leukemia: a Pediatric ­Oncology Group study. J Clin Oncol 1993;11:271-278. 100. Buhrer C, Hartmann R, Fengler R, et al. Importance of effective central nervous system therapy in isolated bone marrow relapse of childhood acute lymphoblastic leukemia. BFM (Berlin-Frankfurt-Munster) Relapse Study Group. Blood 1994;83:34683472. 101. Land VJ, Thomas PR, Boyett JM, et al. Comparison of maintenance treatment regimens for first central nervous system ­relapse in children with acute lymphocytic leukemia. A Pediatric Oncology Group study. Cancer 1985;56:81-87. 102. Ritchey AK, Pollock BH, Lauer SJ, et al. Improved survival of children with isolated CNS relapse of acute lymphoblastic leukemia: a Pediatric Oncology Group study. J Clin Oncol 1999;17:3745-3752. 103. van den Berg H, Odink AE, Behrendt H. Delayed craniospinal irradiation for a first isolated central nervous relapse of acute lymphoblastic leukemia: report on 14 cases. Med Pediatr Oncol 2000;34:402-406. 104. Wofford MM, Smith SD, Shuster JJ, et al. Treatment of occult or late overt testicular relapse in children with acute lymphoblastic leukemia: a Pediatric Oncology Group study. J Clin Oncol 1992;10:624-630. 105. Finklestein JZ, Miller DR, Feusner J, et al. Treatment of overt isolated testicular relapse in children on therapy for acute lymphoblastic leukemia. A report from the Children’s Cancer Group. Cancer 1994;73:219-223. 106. Buchanan GR, Boyett JM, Pollock BH, et al. Improved treatment results in boys with overt testicular relapse during or shortly after initial therapy for acute lymphoblastic leukemia. A Pediatric Oncology Group study. Cancer 1991;68:48-55. 107. Mahoney DH Jr., Shuster JJ, Nitschke R, et al. Intensification with intermediate-dose intravenous methotrexate is effective therapy for children with lower-risk B-precursor acute lympho­ blastic leukemia: a Pediatric Oncology Group study. J Clin Oncol 2000;18:1285-1294.

985

108. Huff V, Miwa H, Haber DA, et al. Evidence for WT1 as a Wilms’ tumor (WT) gene: intragenic germinal deletion in bilateral WT. Am J Hum Genet 1991;48:997-1003. 109. Breslow NE, Beckwith JB. Epidemiological features of Wilms’ tumor: results of the National Wilms’ Tumor Study. J Natl Cancer Inst 1982;68:429-436. 110. Wittmann S, Wunder C, Zirn B, et al. New prognostic markers revealed by evaluation of genes correlated with clinical parameters in Wilms’ tumors. Genes Chromosomes Cancer 2008;47:386395. 111. Howell CG, Otherson HB, Kiviat NE, et al. Therapy and outcome in 51 children with mesoblastic nephroma: a report of the National Wilms’ Tumor Study. J Pediatr Surg 1982;7:  826-831. 112. Beckwith JB, Kiviat NB, Bonadio JF. Nephrogenic rests, nephroblastomatosis, and the pathogenesis of Wilms’ tumor. Pediatr Pathol 1990;10:1-36. 113. Beckwith JB, Palmer NF. Histopathology and prognosis of Wilms’ tumors: results from the First National Wilms’ Tumor Study. Cancer 1978;41:1937-1948. 114. Bowman LC, Hancock ML, Santana VM, et al. Impact of intensified therapy on clinical outcome in infants and children with neuroblastoma: the St. Jude Children’s Research Hospital experience, 1962 to 1988. J Clin Oncol 1991;9:1599-1608. 115. Bonadio JF, Storer B, Norkool P. Anaplastic Wilms’ tumor: clinical and pathologic studies. J Clin Oncol 1985;3:513-520. 116. Breslow N, Churchill G, Beckwith JB, et al. Prognosis for Wilms’ tumor patients with nonmetastatic disease at diagnosis—­results of the Second National Wilms’ Tumor Study. J Clin Oncol 1985;3:521-531. 117. Green DM, Beckwith JB, Breslow NE, et al. Treatment of children with stages II to IV anaplastic Wilms’ tumor: a report from the National Wilms’ Tumor Study Group. J Clin Oncol 1994;12:2126-2131. 118. Beckwith JB, Larson E. Case 7. Clear cell sarcoma of kidney. Pediatric Pathology 1989;9:211-218. 119. Weeks DA, Beckwith JB, Mierau GW, et al. Rhabdoid tumor of kidney. A report of 111 cases from the National Wilms’ Tumor Study Pathology Center. Am J Surg Pathol 1989;13:439-458. 120. Bonnin JM, Rubinstein LJ, Palmer NF, et al. The association of embryonal tumors originating in the kidney and in the brain. A report of seven cases. Cancer 1984;54:2137-2146. 121. Rorke LB, Packer R, Biegel J. Central nervous system atypical teratoid/rhabdoid tumors of infancy and childhood. J Neurooncol 1995;24:21-28. 122. Biegel JA, Rorke LB, Packer RJ, et al. Monosomy 22 in rhabdoid or atypical tumors of the brain. J Neurosurg 1990;73:710-714. 123. Garcia M, Douglass C, Schlosser JV. Classification and prognosis in Wilms’ tumor. Radiology 1963;80:574-580. 124. Cassady JR, Tefft M, Filler RM, et al. Considerations in the radiation therapy of Wilms’ tumor. Cancer 1973;32:598-608. 125. D’Angio GJ, Breslow N, Beckwith JB, et al. Treatment of Wilms’ tumor. Results of the Third National Wilms’ Tumor Study. Cancer 1989;64:349-360. 126. Breslow N, Sharples K, Beckwith JB, et al. Prognostic factors in nonmetastatic, favorable histology Wilms’ tumor. Results of the Third National Wilms’ Tumor Study. Cancer 1991;68:23452353. 127. Gross RE, Neuhauser EBD. Treatment of mixed tumors of the kidney in childhood. Pediatrics 1950;6:843. 128. Farber S. Chemotherapy in the treatment of leukemia and Wilms’ tumor. JAMA 1966;198:826-836. 129. Tournade MF, Com-Nougue C, de Kraker J, et al. Optimal ­duration of preoperative therapy in unilateral and nonmetastatic Wilms’ tumor in children older than 6 months: results of the Ninth International Society of Pediatric Oncology Wilms’ Tumor Trial and Study. J Clin Oncol 2001;19:488-500. 130. Pritchard J, Imeson J, Barnes J, et al. Results of the United Kingdom Children’s Cancer Study Group First Wilms’ Tumor Study. J Clin Oncol 1995;13:124-133. 131. D’Angio GJ, Breslow NE. Treatment of Wilms’ tumor: results of the National Wilms’ Tumor Study. Cancer 1989;64:349-360.

986

PART 13  Childhood Cancer

132. Thomas PR, Tefft M, Compaan PJ, et al. Results of two radiation therapy randomizations in the Third National Wilms’ Tumor Study. Cancer 1991;68:1703-1707. 133. Thomas PR. Wilms’ tumor: changing role of radiation therapy. Semin Radiat Oncol 1997;7:204-211. 134. Thomas PR, Tefft M, Farewell VT, et al. Abdominal relapses in irradiated Second National Wilms’ Tumor Study patients. J Clin Oncol 1984;2:1098-1101. 135. Burgers JM, Tournade MF, Bey P, et al. Abdominal recurrences in Wilms’ tumours: a report from the SIOP Wilms’ Tumour Trials and Studies. Radiother Oncol 1986;5:175-182. 136. Breslow NE, Palmer NF, Hill LR, et al. Wilms’ tumor: prognostic factors for patients without metastases at diagnosis: results of the National Wilms’ Tumor Study. Cancer 1978;41:1577-1589. 137. Tobin RL, Fontanesi J, Kun LE, et al. Wilms’ tumor: reduceddose radiotherapy in advanced-stage Wilms’ tumor with favorable histology. Int J Radiat Oncol Biol Phys 1990;19:867-871. 138. Flentje M, Weirich A, Graf N, et al. Abdominal irradiation in unilateral nephroblastoma and its impact on local control and survival. Int J Radiat Oncol Biol Phys 1998;40:163-169. 139. Ritchey ML, Kelalis PP, Haase GM, et al. Preoperative ­therapy for intracaval and atrial extension of Wilms’ tumor. Cancer 1993;71:4104-4110. 140. Green DM, Jaffe N. The role of chemotherapy in the treatment of Wilms’ tumor. Cancer 1979;44:52-57. 141. Breslow NE, Churchill G, Nesmith B, et al. Clinicopathologic features and prognosis for Wilms’ tumor patients with metastases at diagnosis. Cancer 1986;58:2501-2511. 142. Green DM, Breslow NE, Evans I, et al. Treatment of children with stage IV favorable histology Wilms’ tumor: a report from the National Wilms’ Tumor Study Group. Med Pediatr Oncol 1996;26:147-152. 143. Monson KJ, Brand WN, Boggs JD. Results of small-field irradiation of apparent solitary metastasis from Wilms’ tumor. Radiology 1972;104:157-160. 144. Macklis RM, Oltikar A, Sallan SE. Wilms’ tumor patients with pulmonary metastases. Int J Radiat Oncol Biol Phys 1991;21:1187-1193. 145. de Kraker J, Lemerle J, Voute PA, et al. Wilms’ tumor with pulmonary metastases at diagnosis: the significance of primary chemotherapy. International Society of Pediatric Oncology Nephroblastoma Trial and Study Committee. J Clin Oncol 1990;8:1187-1190. 146. Green DM, Fernbach DJ, Norkool P, et al. The treatment of Wilms’ tumor patients with pulmonary metastases detected only with computed tomography: a report from the National Wilms’ Tumor Study. J Clin Oncol 1991;9:1776-1781. 147. Corey SJ, Andersen JW, Vawter GF, et al. Improved survival for children with anaplastic Wilms’ tumors. Cancer 1991;68:  970-974. 148. Green DM, Breslow NE, Beckwith JB, et al. Treatment of children with clear-cell sarcoma of the kidney: a report from the ­National Wilms’ Tumor Study Group. J Clin Oncol 1994;12:2132-2137. 149. Brodeur GM, Azar C, Brother M, et al. Neuroblastoma. ­Effect of genetic factors on prognosis and treatment. Cancer 1992;70:1685-1694. 150. De Bernardi B, Pianca C, Boni L, et al. Disseminated neuroblas­ toma (stage IV and IV-S) in the first year of life. Outcome ­related to age and stage. Italian Cooperative Group on Neuroblastoma. Cancer 1992;70:1625-1633. 151. van Noesel MM, Hahlen K, Hakvoort-Cammel FG, et al. Neuroblastoma 4S: a heterogeneous disease with variable risk factors and treatment strategies. Cancer 1997;80:834-843. 152. Hayes FA, Green A, Hustu HO, et al. Surgicopathologic staging of neuroblastoma: prognostic significance of regional lymph node metastases. J Pediatr 1983;102:59-62. 153. Rosen EM, Cassady JR, Kretschmar C, et al. Influence of localregional lymph node metastases on prognosis in neuroblastoma. Med Pediatr Oncol 1984;12:260-263. 154. D’Angio GJ, Evans AE, Koop CE. Special pattern of widespread neuroblastoma with a favourable prognosis. Lancet 1971;1:10461049.

155. Spunt SL, Poquette CA, Hurt YS, et al. Prognostic factors for children and adolescents with surgically resected nonrhabdomyosarcoma soft tissue sarcoma: an analysis of 121 patients treated at St. Jude Children’s Research Hospital. J Clin Oncol 1999;17:3697-3705. 156. Nickerson HJ, Matthay KK, Seeger RC, et al. Favorable biology and outcome of stage IV-S neuroblastoma with supportive care or minimal therapy: a Children’s Cancer Group study. J Clin Oncol 2000;18:477-486. 157. Matthay KK. Stage 4S neuroblastoma: what makes it special?   J Clin Oncol 1998;16:2003-2006. 158. Shimada H, Chatten J, Newton WA Jr., et al. Histopathologic prognostic factors in neuroblastic tumors: definition of subtypes of ganglioneuroblastoma and an age-linked classification of neuroblastomas. J Natl Cancer Inst 1984;73:405-416. 159. Brodeur GM, Hayes FA, Green AA, et al. Consistent N-myc copy number in simultaneous or consecutive neuroblastoma samples from sixty individual patients. Cancer Res 1987;47:  4248-4253. 160. Rubie H, Hartmann O, Michon J, et al. N-Myc gene amplification is a major prognostic factor in localized neuroblastoma: ­results of the French NBL 90 study. Neuroblastoma Study Group of the Société Française d’Oncologie Pediatrique. J Clin Oncol 1997;15:1171-1182. 161. Perez CA, Matthay KK, Atkinson JB, et al. Biologic variables in the outcome of stages I and II neuroblastoma treated with surgery as primary therapy: a Children’s Cancer Group study. J Clin Oncol 2000;18:18-26. 162. Caron H, van Sluis P, van Hoeve M, et al. Allelic loss of chromosome 1p36 in neuroblastoma is of preferential maternal origin and correlates with N-myc amplification. Nat Genet 1993;4:187-190. 163. Castle VP, Heidelberger KP, Bromberg J, et al. Expression of the apoptosis-suppressing protein Bcl-2 in neuroblastoma is associated with unfavorable histology and N-myc amplification. Am J Pathol 1993;143:1543-1550. 164. Nakagawara A, Arima-Nakagawara M, Scavarda NJ, et al. Association between high levels of expression of the TRK gene and favorable outcome in human neuroblastoma. N Engl J Med 1993;328:847-854. 165. Chan HS, Haddad G, Thorner PS, et al. P-glycoprotein expression as a predictor of the outcome of therapy for neuroblastoma. N Engl J Med 1991;325:1608-1614. 166. Evans AE, D’Angio GJ, Randolph J. A proposed staging for children with neuroblastoma. Children’s Cancer Study Group A. Cancer 1971;27:374-378. 167. Brodeur GM, Maris JM. Neuroblastoma. In Pizzo PA, Poplack DG (eds): Principles and Practice of Pediatric Oncology, 5th ed. Philadelphia: Lippincott Williams & Wilkins, 2000, pp 933-970. 168. Brodeur GM, Seeger RC, Barrett A, et al. International criteria for diagnosis, staging, and response to treatment in patients with neuroblastoma. J Clin Oncol 1988;6:1874-1881. 169. Brodeur GM, Pritchard J, Berthold F, et al. Revisions of the ­international criteria for neuroblastoma diagnosis, staging, and response to treatment. J Clin Oncol 1993;11:1466-1477. 170. Matthay KK. Intensification of therapy using hematopoietic stem-cell support for high-risk neuroblastoma. Pediatr Transplant 1999;3(Suppl 1):72-77. 171. Nitschke R, Smith EI, Shochat S, et al. Localized neuroblastoma treated by surgery: a Pediatric Oncology Group Study. J Clin Oncol 1988;6:1271-1279. 172. Matthay KK, Sather HN, Seeger RC, et al. Excellent outcome of stage II neuroblastoma is independent of residual disease and radiation therapy. J Clin Oncol 1989;7:236-244. 173. Rubie H, Michon J, Plantaz D, et al. Unresectable localized neuroblastoma: improved survival after primary chemotherapy including carboplatin-etoposide. Neuroblastoma Study Group of the Société Française d’Oncologie Pediatrique (SFOP). Br J Cancer 1998;77:2310-2317. 174. Kushner BH, Cheung NK, LaQuaglia MP, et al. Survival from locally invasive or widespread neuroblastoma without cytotoxic therapy. J Clin Oncol 1996;14:373-381.

37  Childhood Cancer 175. Garaventa A, De Bernardi B, Pianca C, et al. Localized but ­unresectable neuroblastoma: treatment and outcome of 145 cases. Italian Cooperative Group for Neuroblastoma. J Clin Oncol 1993;11:1770-1779. 176. De Bernardi B, Conte M, Mancini A, et al. Localized resectable neuroblastoma: results of the second study of the Italian Cooperative Group for Neuroblastoma. J Clin Oncol 1995;13:884-893. 177. Nitschke R, Smith EI, Altshuler G, et al. Postoperative treatment of nonmetastatic visible residual neuroblastoma: a Pediatric ­Oncology Group study. J Clin Oncol 1991;9:1181-1188. 178. Ninane J, Pritchard J, Morris Jones PH, et al. Stage II neuroblastoma. Adverse prognostic significance of lymph node involvement. Arch Dis Child 1982;57:438-442. 179. Castleberry RP, Kun LE, Shuster JJ, et al. Radiotherapy improves the outlook for patients older than 1 year with Pediatric ­Oncology Group stage C neuroblastoma. J Clin Oncol 1991;9:789-795. 180. Strother D. Treatment of Pediatric Oncology Group stage C neuroblastoma: a preliminary POG report [abstract]. Proc Am Soc Clin Oncol 1993;12:1422. 181. Wolden SL, Gollamudi SV, Kushner BH, et al. Local control with multimodality therapy for stage 4 neuroblastoma. Int J ­Radiat Oncol Biol Phys 2000;46:969-974. 182. Haas-Kogan DA, Fisch BM, Wara WM, et al. Intraoperative radiation therapy for high-risk pediatric neuroblastoma. Int J Radiat Oncol Biol Phys 2000;47:985-992. 183. Leavey PJ, Odom LF, Poole M, et al. Intra-operative radiation therapy in pediatric neuroblastoma. Med Pediatr Oncol 1997;28:424-428. 184. Matthay KK, Villablanca JG, Seeger RC, et al. Treatment of high-risk neuroblastoma with intensive chemotherapy, radiotherapy, autologous bone marrow transplantation, and 13-cisretinoic acid. Children’s Cancer Group. N Engl J Med 1999;341:  1165-1173. 185. Frappaz D, Michon J, Coze C, et al. LMCE3 treatment strategy: results in 99 consecutively diagnosed stage 4 neuroblastomas in children older than 1 year at diagnosis. J Clin Oncol 2000;18:468-476. 186. La Quaglia MP, Kushner BH, Heller G, et al. Stage 4 neuroblastoma diagnosed at more than 1 year of age: gross total resection and clinical outcome. J Pediatr Surg 1994;29:1162-1165. 187. Ikeda H, August CS, Goldwein JW, et al. Sites of relapse in ­patients with neuroblastoma following bone marrow transplantation in relation to preparatory “debulking” treatments.  J ­Pediatr Surg 1992;27:1438-1441. 188. Matthay KK, Atkinson JB, Stram DO, et al. Patterns of relapse after autologous purged bone marrow transplantation for neuroblastoma: a Children’s Cancer Group pilot study. J Clin Oncol 1993;11:2226-2233. 189. McCowage GB, Vowels MR, Shaw PJ, et al. Autologous bone marrow transplantation for advanced neuroblastoma using teniposide, doxorubicin, melphalan, cisplatin, and total-body irradiation. J Clin Oncol 1995;13:2789-2795. 190. Lindsley K. Wilms’ tumor and neuroblastoma. In Leibel S, Phillips T (eds): Textbook of Radiation Oncology. Philadelphia: WB Saunders, 1998, pp 994-1012. 191. Stram DO, Matthay KK, O’Leary M, et al. Consolidation chemoradiotherapy and autologous bone marrow transplantation versus continued chemotherapy for metastatic neuroblastoma: a report of two concurrent Children’s Cancer Group studies.  J Clin Oncol 1996;14:2417-2426. 192. Look AT, Hayes FA, Mitschke R, et al. Cellular DNA content as a predictor of response to chemotherapy in infants with unresectable neuroblastoma. N Engl J Med 1984;311:231-235. 193. Look AT, Hayes FA, Shuster JJ, et al. Clinical relevance of ­tumor cell ploidy and N-myc gene amplification in childhood neuroblastoma: a Pediatric Oncology Group study. J Clin Oncol 1991;9:581-591. 194. Katzenstein HM, Bowman LC, Brodeur GM, et al. Prognostic significance of age, MYCN oncogene amplification, tumor cell ploidy, and histology in 110 infants with stage D(S) neuroblastoma: the Pediatric Oncology Group experience—a Pediatric Oncology Group study. J Clin Oncol 1998;16:2007-2017.

987

195. De Bernardi B, Pianca C, Pistamiglio P, et al. Neuroblastoma with symptomatic spinal cord compression at diagnosis: treatment and results with 76 cases. J Clin Oncol 2001;19:  183-190. 196. Sisson JC, Shapoiro B, Hutchinson RJ, et al. Survival of patients with neuroblastoma treated with 125I MIBG. Am J Clin Oncol 1996;19:144-148. 197. Matthay KK, DeSantes K, Kasegawa B, et al. Phase I dose ­escalation of 131I-metaiodobenzylguanidine with autologous bone marrow support in refractory neuroblastoma. J Clin Oncol 1998;16:229-236. 198. Crist W, Gehan EA, Ragab AH, et al. The Third Intergroup Rhabdomyosarcoma Study. J Clin Oncol 1995;13:610-630. 199. Scrable H, Cavenee W, Ghavimi F, et al. A model for embryonal rhabdomyosarcoma tumorigenesis that involves genome ­imprinting. Proc Natl Acad Sci U S A 1989;86:7480-7484. 200. Douglass EC, Shapiro DN, Valentine M, et al. Alveolar rhabdomyosarcoma with the t(2;13): cytogenetic findings and clinicopathologic correlations. Med Pediatr Oncol 1993;21:83-87. 201. Wexler LH, Helman LJ. Rhabdomyosarcoma and the undifferentiated sarcomas. In Pizzo PA, Poplack DG (eds): Principles and Practice of Pediatric Oncology.. Philadelphia: LippincottRaven, 1997, pp 799-829. 202. Asmar L, Gehan EA, Newton WA, et al. Agreement among and within groups of pathologists in the classification of rhabdomyosarcoma and related childhood sarcomas. Report of an international study of four pathology classifications. Cancer 1994;74:2579-2588. 203. Newton WA Jr, Soule EH, Hamoudi AB, et al. Histopathology of childhood sarcomas, Intergroup Rhabdomyosarcoma Studies I and II: clinicopathologic correlation. J Clin Oncol 1988;6:6775. 204. Tsokos M, Webber BL, Parham DM, et al. Rhabdomyosarcoma. A new classification scheme related to prognosis. Arch Pathol Lab Med 1992;116:847-855. 205. Arndt CA, Crist WM. Common musculoskeletal tumors of childhood and adolescence. N Engl J Med 1999;341:342-352. 206. Raney RB, Meza J, Anderson JR, et al. Treatment of children and adolescents with localized parameningeal sarcoma: experience of the Intergroup Rhabdomyosarcoma Study Group protocols IRS-II through -IV, 1978-1997. Med Pediatr Oncol 2002;38:  22-32. 207. Raney RB Jr, Tefft M, Maurer HM, et al. Disease patterns and survival rate in children with metastatic soft-tissue sarcoma. A report from the Intergroup Rhabdomyosarcoma Study (IRS)-I. Cancer 1988;62:1257-1266. 208. Lawrence W Jr, Anderson JR, Gehan EA, et al. Pretreatment TNM staging of childhood rhabdomyosarcoma: a report of the Intergroup Rhabdomyosarcoma Study Group. Cancer 1997;80:1165-1170. 209. Lawrence W Jr, Hays DM, Heyn R, et al. Lymphatic metastases with childhood rhabdomyosarcoma. A report from the Intergroup Rhabdomyosarcoma Study. Cancer 1987;60:910-915. 210. Pedrick TJ, Donaldson SS, Cox RS. Rhabdomyosarcoma: the Stanford experience using a TNM staging system. J Clin Oncol 1986;4:370-378. 211. Crist WM, Garnsey L, Beltangady MS, et al. Prognosis in children with rhabdomyosarcoma: a report of the Intergroup Rhabdomyosarcoma Studies I and II. Intergroup Rhabdomyosarcoma Committee. J Clin Oncol 1990;8:443-452. 212. Maurer HM, Gehan EA, Beltangady M, et al. The Intergroup Rhabdomyosarcoma Study-II. Cancer 1993;71:1904-1922. 213. Rodary C, Gehan EA, Flamant F, et al. Prognostic factors in 951 nonmetastatic rhabdomyosarcoma in children: a report from the International Rhabdomyosarcoma Workshop. Med Pediatr ­Oncol 1991;19:89-95. 214. Jones IS, Reese AB, Kraut J. Orbital rhabdomyosarcoma. An analysis of 62 cases. Am J Ophthalmol 1966;61:721-736. 215. Tefft M, Jaffe N. Proceedings: sarcoma of the bladder and prostate in children. Rationale for the role of radiation therapy based on a review of the literature and a report of fourteen additional patients. Cancer 1973;32:1161-1177.

988

PART 13  Childhood Cancer

216. Cassady JR, Sagerman RH, Tretter P, et al. Radiation therapy for rhabdomyosarcoma. Radiology 1968;91:116-120. 217. Edland RW. Embryonal rhabdomyosarcoma. AJR Am J Roentgenol Radium Ther Nucl Med 1965;93:671. 218. Nelson AJ III. Embryonal rhabdomyosarcoma. Report of ­twentyfour cases and study of the effectiveness of radiation therapy upon the primary tumor. Cancer 1968;22:64-68. 219. Sagerman RH, Tretter P, Ellsworth RM. The treatment of orbital rhabdomyosarcoma of children with primary radiation therapy. Am J Roentgenol Radium Ther Nucl Med 1972;114:31-34. 220. Jereb B, Cham W, Lattin P, et al. Local control of embryonal rhabdomyosarcoma in children by radiation therapy when combined with concomitant chemotherapy. Int J Radiat Oncol Biol Phys 1976;1:217-225. 221. Hays DM, Lawrence W Jr, Crist WM, et al. Partial cystectomy in the management of rhabdomyosarcoma of the bladder: a report from the Intergroup Rhabdomyosarcoma Study. J Pediatr Surg 1990;25:719-723. 222. Hays DM, Raney RB, Wharam MD, et al. Children with vesical rhabdomyosarcoma (RMS) treated by partial cystectomy with neoadjuvant or adjuvant chemotherapy, with or without radiotherapy. A report from the Intergroup Rhabdomyosarcoma Study (IRS) Committee. J Pediatr Hematol Oncol 1995;17:  46-52. 223. Raney RB Jr, Gehan EA, Hays DM, et al. Primary chemo therapy with or without radiation therapy and/or surgery for children with localized sarcoma of the bladder, prostate, ­vagina, uterus, and cervix. A comparison of the results in Intergroup Rhabdomyosarcoma Studies I and II. Cancer 1990;66:  2072-2081. 224. Martelli H, Oberlin O, Rey A, et al. Conservative treatment for girls with nonmetastatic rhabdomyosarcoma of the genital tract: a report from the Study Committee of the International Society of Pediatric Oncology. J Clin Oncol 1999;17:2117-2122. 225. Breneman JC, Wiener ES. Issues in the local control of rhabdomyosarcoma. Med Pediatr Oncol 2000;35:104-109. 226. Maurer HM, Beltangady M, Gehan EA, et al. The Intergroup Rhabdomyosarcoma Study-I. A final report. Cancer 1988;61:  209-220. 227. Wolden SL, Anderson JR, Crist WM, et al. Indications for ­radiotherapy and chemotherapy after complete resection in ­rhabdomyosarcoma: a report from the Intergroup Rhabdomyosarcoma Studies I to III. J Clin Oncol 1999;17:3468-3475. 228. Regine WF, Fontanesi J, Kumar P, et al. Local tumor control in rhabdomyosarcoma following low-dose irradiation: comparison of group II and select group III patients. Int J Radiat Oncol Biol Phys 1995;31:485-491. 229. Mandell L, Ghavimi F, Peretz T, et al. Radiocurability of microscopic disease in childhood rhabdomyosarcoma with radiation doses less than 4,000 cGy. J Clin Oncol 1990;8:1536-1542. 230. Regine WF, Fontanesi J, Kumar P, et al. A phase II trial evaluating selective use of altered radiation dose and fractionation in ­ patients with unresectable rhabdomyosarcoma. Int J Radiat ­Oncol Biol Phys 1995;31:799-805. 231. Wharam MD, Hanfelt JJ, Tefft MC, 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. 232. Oberlin O, Rey A, Anderson J, et al. Treatment of orbital rhabdomyosarcoma: survival and late effects of treatment—results of an international workshop. J Clin Oncol 2001;19:197-204. 233. Kodet R, Newton WA Jr, Hamoudi AB, et al. Orbital rhabdomyosarcomas and related tumors in childhood: relationship of morphology to prognosis—an Intergroup Rhabdomyosarcoma Study. Med Pediatr Oncol 1997;29:51-60. 234. Fiorillo A, Migliorati R, Vassallo P, et al. Radiation late effects in children treated for orbital rhabdomyosarcoma. Radiother ­Oncol 1999;53:143-148. 235. Raney RB, Anderson JR, Kollath J, et al. Late effects of therapy in 94 patients with localized rhabdomyosarcoma of the orbit: report from the Intergroup Rhabdomyosarcoma Study (IRS)-III, 1984-1991. Med Pediatr Oncol 2000;34:413-420.

236. Tarbell NJ. Extent of bone erosion predicts survival in non-­orbital rhabdomyosarcoma of the head and neck in children ­[abstract]. Proc Am Soc Clin Oncol 1987;6:222a. 237. Mandell LR, Massey V, Ghavimi F. The influence of extensive bone erosion on local control in non-orbital rhabdomyosarcoma of the head and neck. Int J Radiat Oncol Biol Phys 1989;17:  649-653. 238. Tefft M, Fernandez C, Donaldson M, et al. Incidence of meningeal involvement by rhabdomyosarcoma of the head and neck in children: a report of the Intergroup Rhabdomyosarcoma Study (IRS). Cancer 1978;42:253-258. 239. Raney RB, Asmar L, Vassilopoulou-Sellin R, et al. Late complications of therapy in 213 children with localized, nonorbital softtissue sarcoma of the head and neck: a descriptive report from the Intergroup Rhabdomyosarcoma Studies (IRS)-II and-III. IRS Group of the Children’s Cancer Group and the Pediatric Oncology Group. Med Pediatr Oncol 1999;33:362-371. 240. Heyn R, Newton WA, Raney RB, et al. Preservation of the bladder in patients with rhabdomyosarcoma. J Clin Oncol 1997;15:69-75. 241. Raney RB Jr, Heyn R, Hays DM, et al. Sequelae of treatment in 109 patients followed for 5 to 15 years after diagnosis of sarcoma of the bladder and prostate. A report from the Intergroup Rhabdomyosarcoma Study Committee. Cancer 1993;71:2387-  2394. 242. Wiener ES, Lawrence W, Hays D, et al. Retroperitoneal node biopsy in paratesticular rhabdomyosarcoma. J Pediatr Surg 1994;29:171-177. 243. Flamant F, Gerbaulet A, Nihoul-Fekete C, et al. Long-term ­sequelae of conservative treatment by surgery, brachytherapy, and chemotherapy for vulval and vaginal rhabdomyosarcoma in children. J Clin Oncol 1990;8:1847-1853. 244. Heyn R, Beltangady M, Hays D, et al. Results of intensive therapy in children with localized alveolar extremity rhabdomyos­ arcoma: a report from the Intergroup Rhabdomyosarcoma Study. J Clin Oncol 1989;7:200-207. 245. Ghavimi F, Mandell LR, Heller G, et al. Prognosis in childhood rhabdomyosarcoma of the extremity. Cancer 1989;64:22332237. 246. Merchant TE. Conformal therapy for pediatric sarcomas. Semin Radiat Oncol 1997;7:236-245. 247. Donaldson SS. The value of adjuvant chemotherapy in the management of sarcomas in children. Cancer 1985;55:2184-2197. 248. York T, Schneider R, Friedmann A, et al. Proton radiotherapy for orbital rhabdomyosarcoma: clinical outcome and a dosimetric comparison with photons. Int J Radiat Oncol Biol Phys 2005;63:1161-1168. 249. Timmermann B, Schuck A, Niggli F, et al. Spot-scanning proton therapy for malignant soft tissue tumors in childhood: first experience at the Paul Scherrer Institute. Int J Radiat Oncol Biol Phys 2007;67:497-504. 250. Tefft M, Wharam M, Gehan E. Local and regional control by radiation of rhabdomyosarcoma in IRS-II [abstract 83]. Int J ­Radiat Oncol Biol Phys 1988;(Suppl 1):159. 251. Kun L. Treatment factors affecting local control in childhood rhabdomyosarcoma [abstract]. Proc Am Soc Clin Oncol 1986;5:207a. 252. Donaldson SS, Asmar L, Breneman J, et al. Hyperfractionated radiation in children with rhabdomyosarcoma—results of an Intergroup Rhabdomyosarcoma Pilot Study. Int J Radiat Oncol Biol Phys 1995;32:903-911. 253. Knudson AG Jr. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci USA 1971;68:820-823. 254. Gallie BL. Retinoblastoma gene mutations in human cancer.  N Engl J Med 1994;330:786-787. 255. Moll AC, Imhof SM, Bouter LM, et al. Second primary tumors in patients with retinoblastoma. A review of the literature. Ophthalmic Genet 1997;18:27-34. 256. Steenbakkers RJ, Altschuler MD, D’Angio GJ, et al. Optimized lens-sparing treatment of retinoblastoma with electron beams. Int J Radiat Oncol Biol Phys 1997;39:589-594.

37  Childhood Cancer 257. Reese AB. Tumors of the Eye, 3rd ed. Hagerstown, MD: Harper &  Row, 1976. 258. Hernandez JC, Brady LW, Shields JA, et al. External beam radiation for retinoblastoma: results, patterns of failure, and a proposal for treatment guidelines. Int J Radiat Oncol Biol Phys 1996;35:125-132. 259. Pradhan DG, Sandridge AL, Mullaney P, et al. Radiation therapy for retinoblastoma: a retrospective review of 120 patients. Int J Radiat Oncol Biol Phys 1997;39:3-13. 260. Shields CL, Shields JA, Needle M, et al. Combined chemoreduction and adjuvant treatment for intraocular retinoblastoma. Ophthalmology 1997;104:2101-2111. 261. Friedman DL, Himelstein B, Shields CL, et al. Chemoreduction and local ophthalmic therapy for intraocular retinoblastoma.  J Clin Oncol 2000;18:12-17. 262. Murphree AL, Villablanca JG, Deegan WF III, et al. Chemotherapy plus local treatment in the management of intraocular retinoblastoma. Arch Ophthalmol 1996;114:1348-1356. 263. Gallie BL, Budning A, DeBoer G, et al. Chemotherapy with ­focal therapy can cure intraocular retinoblastoma without radiotherapy. Arch Ophthalmol 1996;114:1321-1328. 264. Cassady JR, Sagerman RH, Tretter P, et al. Radiation therapy in retinoblastoma. An analysis of 230 cases. Radiology 1969;93:405409. 265. Schipper J. An accurate and simple method for megavolt age radiation therapy of retinoblastoma. Radiother Oncol 1983;1:  31-41. 266. Schipper J, Tan KE, van Peperzeel HA. Treatment of retinoblastoma by precision megavoltage radiation therapy. Radiother ­Oncol 1985;3:117-132. 267. Freeman CR, Esseltine DL, Whitehead VM, et al. Retinoblastoma: the case for radiotherapy and for adjuvant chemotherapy. Cancer 1980;46:1913-1918. 268. Krengli M, Hug EB, Adams JA, et al. Proton radiation therapy for retinoblastoma: comparison of various intraocular tumor ­locations and beam arrangements. Int J Radiat Oncol Biol Phys. 2005;61:583-593. 269. Lee CT, Bilton SD, Famiglietti RM, et al. Treatment planning with protons for pediatric retinoblastoma, medulloblastoma, and pelvic sarcoma: how do protons compare with other conformal techniques? Int J Radiat Oncol Biol Phys 2005;63:362372. 270. Pratt CB, Crom DB, Howarth C. The use of chemotherapy for extraocular retinoblastoma. Med Pediatr Oncol 1985;13:  330-333. 271. Madreperla SA, Hungerford JL, Doughty D, et al. Treatment of retinoblastoma vitreous base seeding. Ophthalmology 1998;105:120-124. 272. Egbert PR, Donaldson SS, Moazed K, et al. Visual results and ocular complications following radiotherapy for retinoblastoma. Arch Ophthalmol 1978;96:1826-1830.

989

273. Khelfaoui F, Validire P, Auperin A, et al. Histopathologic risk factors in retinoblastoma: a retrospective study of 172 patients treated in a single institution. Cancer 1996;77:1206-1213. 274. Merchant TE, Gould CJ, Hilton NE, et al. Ocular preservation after 36-Gy external beam radiation therapy for retinoblastoma. J Pediatr Hematol Oncol 2002;24:246-249. 275. Howarth C, Meyer D, Hustu HO, et al. Stage-related combined modality treatment of retinoblastoma. Results of a prospective study. Cancer 1980;45:851-858. 276. Czyzewski EA, Goldman S, Mundt AJ, et al. Radiation therapy for consolidation of metastatic or recurrent sarcomas in children treated with intensive chemotherapy and stem cell rescue. A feasibility study. Int J Radiat Oncol Biol Phys 1999;44:569-577. 277. Abramson DH, Ellsworth RM, Tretter P, et al. Simultaneous bilateral radiation for advanced bilateral retinoblastoma. Arch Ophthalmol 1981;99:1763-1766. 278. Gaitan-Yanguas M. Retinoblastoma: analysis of 235 cases. Int J Radiat Oncol Biol Phys 1978;4:359-365. 279. Sotelo-Avila C, Gonzalez-Crussi F, Fowler JW. Complete and incomplete forms of Beckwith-Wiedemann syndrome: their ­oncogenic potential. J Pediatr 1980;96:47-50. 280. Lack EE, Neave C, Vawter GF. Hepatoblastoma. A clinical and pathologic study of 54 cases. Am J Surg Pathol 1982;6:693705. 281. Ohaki Y, Misugi K, Sasaki Y, et al. Hepatitis B surface antigen positive hepatocellular carcinoma in children. Report of a case and review of the literature. Cancer 1983;51:822-828. 282. Gauthier F, Valayer J, Thai BL, et al. Hepatoblastoma and hepatocarcinoma in children: analysis of a series of 29 cases. J Pediatr Surg 1986;21:424-429. 283. Giacomantonio M, Ein SH, Mancer K. Thirty years of experience with pediatric primary malignant liver tumors. J Pediatr Surg 1984;19:523. 284. Price JB Jr, Schullinger JN, Santulli TV. Major hepatic resections for neoplasia in children. Arch Surg 1982;117:1139-1141. 285. Tagge EP, Tagge DU, Reyes J, et al. Resection, including transplantation, for hepatoblastoma and hepatocellular carcinoma: impact on survival. J Pediatr Surg 1992;27:292-296. 286. Weinblatt ME, Siegel SE, Siegel MM, et al. Preoperative chemotherapy for unresectable primary hepatic malignancies in children. Cancer 1982;50:1061-1064. 287. Ortega JA, Krailo MD, Haas JE, et al. Effective treatment of unresectable or metastatic hepatoblastoma with cisplatin and continuous infusion doxorubicin chemotherapy: a report from the Children’s Cancer Study Group. J Clin Oncol 1991;9:21672176. 288. Habrand JL, Nehme D, Kalifa C, et al. Is there a place for radiation therapy in the management of hepatoblastomas and hepatocellular carcinomas in children? Int J Radiat Oncol Biol Phys 1992;23:525-531.

Radiation Therapy for Bone Marrow 38 or Stem Cell Transplantation Selim Y. Firat and Colleen A. Lawton HISTORY OF TOTAL-BODY IRRADIATION FOR MARROW OR STEM CELL TRANSPLANTATION RESPONSE OF NORMAL BONE MARROW TO TOTAL-BODY IRRADIATION Response of Peripheral Blood Effects on the Spleen TYPES OF TRANSPLANTS Autologous Transplants Syngeneic Transplants Allogeneic Transplants USES OF MARROW OR STEM CELL TRANSPLANTATION ROLES OF TOTAL-BODY IRRADIATION IN MARROW OR STEM CELL TRANSPLANTATION AND DELIVERY TECHNIQUES

Dose Rate Total Dose and Fractionation Selective Organ Shielding OTHER TYPES OF IRRADIATION USED WITH MARROW OR STEM CELL TRANSPLANTATION Irradiation of the Spleen Irradiation of the Central Nervous System Boost Doses for Local Disease Nonmyeloablative Conditioning Regimens SEQUELAE OF TOTAL-BODY IRRADIATION AND MARROW OR STEM CELL TRANSPLANTATION Acute Toxic Effects Long-Term Effects CONCLUSIONS

History of Total-Body Irradiation for Marrow or Stem Cell Transplantation Total-body irradiation (TBI) has played a pivotal role in bone marrow transplantation since the first successful transplants were performed in mice in the 1950s.1 Supralethal doses of TBI followed by the infusion of compatible bone marrow allowed cells to maintain viability. After this preliminary work, similar studies were performed with canines2 and primates.3 In this early transplantation era, TBI of animals was shown to have a profound effect on the immune system, but the administration of appropriate types and quantities of bone marrow could lead to repair of the damage produced by the radiation. This preliminary animal research quickly led to the exploration of bone marrow transplantation in humans. Bone marrow transplantation in humans began in the late 1950s, when Thomas and colleagues4 performed bone marrow infusion on a handful of patients with fatal diseases, mostly leukemias. Thomas’s early work provided several important insights into bone marrow transplantation. That work involved the first successful infusions of bone marrow into humans without significant clinical problems such as pulmonary emboli. Subsequently, Thomas’s group showed that it was the radiation rather than the chemotherapy (as it was known in the late 1950s) that produced sufficient immunosuppression for the infused marrow to be accepted. They proved that the doses of TBI necessary to adequately suppress the patient’s immune system to allow successful engraftment were well above the generally accepted median lethal dose for humans. With this early work, autologous (between host and self) and syngeneic (between identical siblings) bone marrow

transplantations were well under way. Unfortunately, allogeneic transplantation (transplantation between two persons who are not identical twins) was significantly hindered in its initial development because of problems with nonengraftment.5 After this problem was overcome, as a result of better understanding of the histocompatibility antigens, the next hurdle was the serious and potentially lethal sequelae of transplantation now recognized as graft-versus-host disease (GVHD), in which the donor graft attacks the host’s organs. A suggestion of this process had been seen in rodent models,6 but a real understanding of it in the human system did not occur until 1975, when Thomas and colleagues7 described the disease and developed a grading system for it. With the advent of drugs such as methotrexate and cyclosporine, physicians were able to control this lethal complication,8 allowing allogeneic bone marrow transplantation to flourish. Most transplants today are allogeneic and are given ­after a conditioning regimen of chemotherapy (usually cyclophosphamide) to destroy the normal bone marrow followed by external beam radiation therapy (EBRT) for the entire body. This combination effectively kills tumor cells and suppresses the host immune system at lower doses than would be necessary if either modality was used alone. TBI remains an important part of the transplantation strategy.

Response of Normal Bone Marrow to Total-Body Irradiation It is readily apparent that TBI, alone or in conjunction with bone marrow transplantation, affects virtually every organ system, depending on the dose and fractionation. This

993

994

PART 14  Special Considerations

15

2

10

1

5

0

0

0

10

20

30

40

10

3 2

5

1 0

0

50

Days after irradiation Platelets Leucocytes

Erythrocytes Granulocytes

FIGURE 38-1.  Changes in the relative amounts of peripheral blood elements in humans after total-body irradiation with 4.3 Gy in 2 fractions. Day 0 is the first day of irradiation. (From Tubiana M, Frindel E, Croizat H, et al. Effects of radiations on bone marrow. Pathol Biol (Paris) 1979;27:326-334.)

to localized breast carcinoma, sternal biopsies showed little or no marrow regeneration in 54 cases (96.5%).

Response of Peripheral Blood The work of Tubiana and colleagues10 showed that TBI has a profound and dose-dependent effect on peripheral blood counts. Figure 38-1 shows that significant doses of TBI, such as those used in association with marrow or stem cell transplantation, produce leukopenia, granulocytopenia, anemia, and thrombocytopenia. As discussed in the following paragraphs, the magnitude of the respective cytopenias depends on the dose of the radiation, on the inherent radiosensitivity of the circulating cells and their precursors, and on the normal life span of the circulating cells.

Leukocytes Being the most radiosensitive of the peripheral blood cells, lymphocyte counts drop drastically soon after TBI.13 Lymphoid precursors are also highly radiosensitive; the time required for recovery of recirculating lymphocytes depends on the radiation dose, and the process requires weeks or even months in rodents. Granulocytes and their precursors are less radiosensitive, and therefore granulocyte counts fall 1 to 2 days after TBI (see Fig. 38-1).10 Significant granulocyte regeneration usually does not occur until approximately 20 to 25 days after TBI.

Platelets Circulating platelets are relatively radioresistant.14 ­Given their normal circulating life span of 8 to 11 days, the drop in peripheral platelet counts is delayed relative to that of the other myeloid cells. The platelet precursor cells, the megakaryocytes, are the least radiosensitive of the myeloid elements13; measurable damage reflected in the peripheral blood counts appears approximately  2 to 3 days after TBI. The nadir lasts approximately 2 to  3 weeks. 1000 Granulocytes (%)

3

4

Granulocytes (´ 103/mm3)

20

Erythrocytes (´ 106/mm3)

4

Leucocytes (´ 103/mm3)

Platelets (´ 105/mm3)

s­ ection focuses on the effects of irradiation on normal bone marrow and other hematopoietic tissues. Additional details are given in Chapters 1 and 34. The effects of TBI on experimental animals have been well documented; the LD50/30 (i.e., the single dose that is lethal at 30 days to 50% of the experimental animals tested) is well defined for most mammals. However, little information is available regarding TBI in humans except in cases of accidental exposure, for which retrospective ­estimations of dose make the findings less reliable.9 Tubiana and colleagues10 have provided some unique information regarding the effects of TBI on humans. They used TBI for immunosuppression in 41 patients undergoing kidney transplantation. After irradiation with 1 to 6 Gy given in 1 or 2 fractions, patients were maintained in “aseptic” rooms and given a controlled diet until their leukocyte levels returned to normal. Despite these precautions, two patients who received 6 Gy died within a month after the transplantation, with aplasia resulting from the irradiation being the most likely cause of death. Seven of 17 patients given between 3.5 and 4.5 Gy in one or two sessions died within 40 days after transplantation. Four of those deaths most likely were the result of aplasia or infection enhanced by the aplasia resulting from the irradiation. Figure 38-1 shows the changes in amounts of peripheral blood elements in patients from this series. The investigators observed that the initial decrease in peripheral blood counts was more rapid and the nadir was reached sooner after higher radiation doses, as Figure 38-2 illustrates for granulocyte counts. They concluded that the median lethal dose in humans must be greater than the previous estimate of 4.5 Gy, a conclusion substantiated by observations of the victims of the Chernobyl nuclear reactor accident. Even when large amounts of bone marrow are exposed to high doses of radiation, marrow recovery can occur provided that a significant portion of the bone marrow is shielded.11 However, Sykes and colleagues12 showed that 30 Gy given at 10 Gy per week seems to be the dose beyond which bone marrow regeneration does not occur; after 56 women were given radiation doses of 30 Gy or more

100 10 1 0 0

5

10

15

20

25

30

35

40

Days after irradiation 0.5 Gy 1 Gy 2.5 Gy

4 Gy 6 Gy

FIGURE 38-2.  Comparison of changes in circulating granulocytes in five patients given total-body irradiation in single doses ranging from 0.5 to 6 Gy. Levels are expressed as percent of preirradiation granulocyte counts. (From Tubiana M, Frindel E, Croizat H, et al. Effects of radiations on bone marrow. Pathol Biol (Paris) 1979;27:326-334.)

995

38  Radiation Therapy for Bone Marrow or Stem Cell Transplantation

Erythrocytes

Allogeneic Transplants

Red blood cells have the longest circulating life span of the peripheral blood cells (120 days), and in their mature, circulating form are resistant to direct damage from irradiation. After TBI in preparation for bone marrow transplantation, the erythrocyte counts drop the slowest and recover the fastest of all the circulating cell types (see Fig. 38-1). In this regard, the TBI given for marrow or stem cell transplantation is unlike the fractionated low doses of TBI given over periods of months to years for chronic lymphocytic leukemia, from which erythropoietic stem cells do not completely recover.15

Allogeneic marrow or stem cell transplantation takes place between two nonidentical individuals of the same species (e.g., two humans) and can be subdivided into two broad categories: related and unrelated. Related allogeneic bone marrow transplantation occurs between two genetically related individuals. These transplants are further categorized according to the degree of human leukocyte antigen (HLA) matching. A completely matched donor and recipient transplant occurs between HLA-identical siblings, whereas a mismatched related transplant occurs between relatives who match at only three, four, or five of the six major HLA loci. The donor can be a parent, aunt, cousin, or other relatives having the appropriate HLA type. Unrelated allogeneic marrow or stem cell transplantation takes place between two genetically unrelated ­people. Success with this type of transplant requires that the ­donor and recipient share most, if not all, of the six major  HLA loci.

Effects on the Spleen The white pulp of the spleen is structurally similar to lymph nodes, and the response of the spleen to ionizing radiation is similar to that of the lymph nodes (see Chapter 34). In humans, lymphocyte maturation takes place largely in the white pulp of the spleen and in the lymph nodes. Extensive experience with splenic irradiation in patients with early-stage Hodgkin disease has revealed no significant clinical effect of the radiation. No reports have suggested a clinically significant change in the spleen after local irradiation for Hodgkin disease or after TBI in conjunction with marrow or stem cell transplantation.

Types of Transplants Cells to be used for transplantation can be obtained in one of two ways. In the traditional bone marrow harvest, the patient is anesthetized, and multiple marrow samples are collected by needle aspiration from the iliac crest or occasionally from the sternum. In the more common procedure, donors are given cytokines, with or without ­chemotherapy, to stimulate the production and release of stem cells from the marrow into the peripheral blood (i.e., stem cell mobilization). Stem cells are then removed from the peripheral blood by apheresis, and the remaining blood is returned to the donor. The relationship of the donor to the recipient defines the type of marrow or stem cell transplant: autologous, syngeneic, or allogeneic. Each type is described briefly in the following paragraphs.

Uses of Marrow or Stem Cell Transplantation The use of marrow or stem cell transplantation has been increasing over time (Fig. 38-3) as success rates continue to improve, especially for leukemias and lymphomas. Most marrow or stem cell transplantations have been performed in patients with leukemia.16 Initially, transplantation was considered only for acute leukemias in refractory phases or after multiple relapses,17 but currently, many patients with high-risk acute leukemias are referred for transplantation during their first remission. Patients include children  with Philadelphia chromosome–positive acute ­lymphocytic leukemia and adults with acute leukemia with high-risk ­features.18,19 Patients with chronic leukemia did not routinely undergo transplantation in the early years because the natural history of the disease allowed many patients to live for years before succumbing to the disease. However, as bone marrow transplantation became safer, it was considered and used successfully for patients with chronic leukemias

In autologous transplants, the donor is also the recipient. In this procedure, stem cells are harvested from the patient and frozen, after which the patient undergoes a conditioning regimen involving ablative chemotherapy with or without irradiation. After the conditioning regimen has been completed, the frozen stem cells are thawed and reinfused. If the stem cells are thought to be contaminated by malignant cells, they may be purged before being reinfused. This process involves exposing the isolated stem cells to agents such as monoclonal antibodies in an attempt to remove any neoplastic cells.

Syngeneic Transplants Syngeneic transplantation takes place between genetically identical twins. Some of the first successful transplants in humans were syngeneic. This type of transplant is rarely used because few patients have genetically identical twins.

Number of transplants

Autologous Transplants

40,000 36,000 32,000 28,000 24,000 20,000 16,000 12,000 8,000 4,000 0 1970

Autologous

Allogeneic

1975

1980

1985

1990

1995

Year

FIGURE 38-3.  Numbers of patients undergoing allogeneic and autologous blood and marrow transplantation worldwide, 1970 to 1998. (Courtesy of the International Bone Marrow Transplant Registry, Milwaukee, WI.)

996

PART 14  Special Considerations

(especially chronic granulocytic leukemia).20-22 Currently, adult patients with chronic myelogenous leukemia ­routinely ­undergo transplantation while they are in early chronicphase disease if an appropriate donor can be found.22-24 Lymphomas make up the other major disease category for which transplantation is performed. Hodgkin and nonHodgkin lymphomas have been successfully treated with autologous, syngeneic, or allogeneic bone marrow transplantation.25,26 Most patients with lymphomas are treated with standard chemotherapy or radiation, or both, and they are considered for marrow or stem cell transplantation only if the disease proves refractory to these therapies. Experimentally, peripheral blood stem cell infusions may be used to enhance marrow recovery after intensive chemotherapy. Some reports suggest that autologous marrow or stem cell transplantation can be appropriate front-line therapy for patients with certain aggressive lymphomas.27-29 Bone marrow transplantation is used for ­noncancerous but otherwise fatal conditions, such as severe aplastic anemia and some congenital immunodeficiency disorders,18,30- 33 and for some solid tumors. The most successful transplantations for treating solid tumors have been done for children with neuroblastoma34 or Ewing’s sarcoma.35-37 Transplantations for solid tumors in adults, especially autologous transplants for breast cancer, have been studied extensively but remain controversial.38-43

Roles of Total-Body Irradiation in Marrow or Stem Cell Transplantation and Delivery Techniques TBI serves two different but important functions in preparation for marrow or stem cell transplantation. The first of these functions is to suppress the immune system, and TBI was first used in humans expressly for this purpose.4 The second function is to kill tumor cells. For cancer patients, this role can be as important as immunosuppression. Techniques for the delivery of TBI have evolved in North America since Heublein44 published the first description of a dedicated unit in 1932. In that unit, four beds were placed in a single room with an x-ray machine at one end, and four patients could be treated simultaneously. The exposure rate varied from 1.26 roentgen (r)/hr for patients in the closer beds to 0.68 r/hr for those in the more distant beds. This technique was quite progressive for its time, but modern techniques are very different. Kim and colleagues45 and subsequently Van Dyk46 described the various techniques for delivery of TBI in use at several major centers across North America during the 1980s. From these reports, it is clear that several different techniques can successfully deliver uniform TBI (i.e.,  ± 10% dose homogeneity) through the use of single or opposing cobalt sources or single or opposing low-energy liner accelerators. Several centers also use high-energy linear accelerators, but the use of such equipment requires careful attention to superficial doses.47-49 The position of the patient during treatment also varies; although no one position is inherently better than another, dose uniformity is easier to obtain when the patient is treated while lying down or standing. Opposed lateral techniques can be

A

B

FIGURE 38-4.  The stand for total-body irradiation was ­designed by Mick-Radio Nuclear (New York, NY) and is used at the Medical College of Wisconsin. Notice the partial transmission blocks.

c­ ombined with anteroposterior-posteroanterior techniques to improve dose homogeneity and to make selective organ shielding easier.44,50-52 Treatment stands for TBI can be purchased and customized to fit a transplantation center’s needs. Use of such stands (Fig. 38-4) allows safe, effective, and uniform dose delivery in combination with effective partial organ shielding. When the appropriate energy range of x-rays is determined and dose uniformity is obtained, the next two issues to be addressed are dose rate and fractionation.

Dose Rate Many reports suggest that lower dose rates produce the largest therapeutic gains, because the repair capacity of bone marrow stem cells and malignant hematopoietic cells is more limited than that of other cells in mammals.50,53-56 Clinical data in support of the use of low dose rates can be found in an analysis of the International Bone Marrow Transplant Registry, which described a statistically significant difference in the incidence of interstitial pneumonitis between the use of very low dose rates (up to 0.04 Gy/min) and higher ones. In a later retrospective study,58 a dose rate of 0.15 Gy/min was associated with higher rates of interstitial pneumonitis than was 0.075 Gy/min. The higher dose rate was an independently significant predictor of the development of interstitial pneumonitis and was found in univariate analyses to be associated with poorer survival.58 These findings led some transplant centers to continue to use very low dose rates to deliver TBI. However, other evidence suggests that the therapeutic gain from the use of lower dose rates is limited and that the use of higher dose rates (e.g., 0.2 to 0.3 Gy/min) is safe.49,59,60 The most common dose rates in use today range from 0.05 to 0.2 Gy/min.

Total Dose and Fractionation The value of fractionation in TBI is better understood than in the past. When TBI was initially performed, the usual dose was 8 to 10 Gy, given as a single fraction at low dose

38  Radiation Therapy for Bone Marrow or Stem Cell Transplantation

rates (up to 5 Gy/min).60a Single-fraction TBI given at ­higher doses (≥10 Gy) has been associated with increased risk of death unrelated to relapse, as shown in an analysis of 6691 patients in the International Bone Marrow Transplant Registry.61 Single-fraction TBI is also more likely to produce severe chronic renal failure and veno-occlusive disease than are fractionated schemes.62,63 The consensus is that fractionating the radiation allows higher total doses to be given safely and improves long-term survival.64,65 Shank and colleagues66 introduced the concept of hyperfractionated radiation as a way of reducing the incidence of pneumonitis and allowing even higher total doses to be delivered. Doses above 12 Gy seemed to be necessary for transplants from siblings who were less than fully HLAmatched.67,68 Clift,29,70 Thomas,22 and their colleagues have tested fractionated doses of up to 15.75 Gy given in once-daily fractions, without shielding, over 7 days. Their randomized studies have shown better disease control with 15.75 Gy for patients with acute or chronic myelogenous leukemia, but the toxicity associated with this dose was unacceptably high, resulting in a decrement in overall survival compared with the more conventional dose of 12 Gy given in 6 fractions over 6 days.22,69,70 It seems that the dose limit for TBI has been reached, even with fractionation. The most common fractionation schedule used in the United States is 12 Gy in 6 fractions, given once daily over 6 days or twice daily over 3 days.

997

A

Selective Organ Shielding Selective organ shielding has been used by several centers across the United States in an effort to reduce the toxicity associated with marrow or stem cell transplantation, especially toxicity to the lung, liver, and kidneys.49,51,66 One technique for selective shielding of the lungs is to use thin lead shields (Fig. 38-5A), with the goal of attenuating the dose only enough to account for the difference in density between the lung and soft tissues. Another option is to use thicker lung blocks (see Fig. 38-5B), with the goal of reducing the dose to the lung by a substantial amount (e.g., 50% or 60% of the total dose).49 The thicker lung blocks are used most often at centers that use doses higher than the standard 12 Gy in 6 fractions, because ­pulmonary ­transplantrelated mortality increases with higher TBI doses if no shielding is used. One study71 indicated an 11% incidence of interstitial pneumonitis from a 12-Gy TBI regimen given in 6 fractions with no lung shielding, but shielding reduced that rate to 2.3%. Another study72 also showed that lung shielding conferred a significant survival advantage relative to no shielding by reducing the incidence of pulmonary transplant-related mortality in patients receiving a total TBI dose of 13.6 Gy. Selective shielding designed specifically for the liver and kidneys has been developed at the Medical College of Wisconsin (see Fig. 38-5B). In our experience, rates of late renal dysfunction have been reduced from 26% to 6% with the use of customized renal shielding that reduces the dose to the kidneys from 14 Gy to 12 Gy.49,51 The differences in kidney position when patients are upright or when they are supine must be considered when designing partial transmission blocks for kidneys.73 Customized shielding of lung and kidneys has also been used to test dose escalation of TBI in preparation for autologous peripheral blood stem cell transplantation for

B FIGURE 38-5.  A, Thin lead shields are used to correct for density differences between the lung and soft tissue. B, Lung, liver, and kidney blocks are used in total-body irradiation at the Medical College of Wisconsin for significant shielding of those organs. (A, Courtesy Nancy Tarbell, MD, Department of Radiation Oncology, Harvard Medical School, Boston, MA.)

­ atients with chemotherapy-refractory lymphoma.74 In p that study, use of kidney shields (that reduced the renal dose by 25%) and lung shields (that reduced the pulmonary dose by 50%) allowed TBI doses to be escalated from 16 Gy to 20 Gy (given in 2-Gy fractions twice daily).74 A later study showed that using a computed ­tomography– guided intensity-modulated radiation therapy technique (helical tomotherapy) could reduce the median organ dose by a factor of 1.7 to 7.5 compared with conventional TBI during targeted total marrow irradiation or total marrow and lymphoid irradiation.75 Although this technique seems promising for allowing dose escalation, further studies and clinical experience are necessary to see how it compares with other conventional TBI techniques.

Other Types of Irradiation Used with Marrow or Stem Cell Transplantation Localized irradiation can be used in conjunction with ­marrow or stem cell transplantation, with or without TBI. Specific uses of localized irradiation have varied widely, 

998

PART 14  Special Considerations

but the most common focus on three goals: treating tumor in the spleen, treating tumor in the central nervous system, and boosting the radiation dose to areas of local disease.

transplantation, especially for lymphomas in which ­residual bulky disease is present. When TBI is not used, the boost doses used are those appropriate to the type of tumor.

Irradiation of the Spleen

Nonmyeloablative Conditioning Regimens

Before the development of bone marrow transplantation for chronic myelogenous leukemia in the late 1970s and early 1980s, splenectomy was often used in an attempt to reduce tumor burden and prevent relapse and blastic transformation.76,77 Since then, several investigators have observed that splenectomy, with its associated morbidity, can be avoided by giving low doses of radiation to the spleen in conjunction with TBI and bone marrow transplantation.52,76-80 Results from some randomized trials79 have suggested a disease-free survival advantage from the use of splenic irradiation for some patients with chronic myelogenous leukemia. The usual boost doses to the spleen, given in conjunction with TBI for marrow or stem cell transplantation, vary considerably, from a high of 10 Gy in one fraction79 to a low of 2.5 Gy in five fractions.80

Irradiation of the Central Nervous System Central nervous system irradiation as part of the treatment of acute leukemia is not a new concept. Children with high-risk acute lymphocytic leukemia routinely have been treated with cranial and sometimes with spinal irradiation as prophylaxis against the development of disease in the central nervous system or as treatment of known central nervous system involvement.81 This therapy is not without sequelae, but when it is intended for therapeutic purposes, there is often little choice. Although the role of cranial boost doses in addition to TBI has not been extensively studied, one report showed that the risk of central nervous system recurrence was low for patients presenting with only hematologic disease regardless of whether a cranial boost was used with TBI or not.82 This finding probably reflects the effectiveness of TBI for central nervous system prophylaxis in patients with uninvolved cerebrospinal fluid. When the central nervous system is irradiated in conjunction with marrow or stem cell transplantation, the ­doses vary widely, depending mostly on whether TBI is part of the transplantation strategy. If the central nervous system is irradiated in conjunction with the usual dose of TBI (12 Gy in 6 fractions), the boost doses are usually about 9 to 12 Gy, given in daily 1.5- to 2-Gy fractions. Doses for central nervous system irradiation for marrow or stem cell transplantation without TBI can be much higher (24 to  30 Gy, given in a fractionated course).

Boost Doses for Local Disease In conjunction with marrow or stem cell transplantation, boosting the radiation dose to areas of local disease with EBRT has been used for hematologic malignancies such as leukemias and lymphomas and for solid tumors such as neuroblastomas in children. Boost doses also have been used for women with breast cancer who are undergoing autologous marrow or stem cell transplantation. The doses involved vary widely and depend on the type of tumor, the type of transplant, and whether TBI will be used with transplantation. In patients with leukemias or lymphomas who are undergoing allogeneic bone marrow transplantation, boost doses of approximately 10 to 20 Gy (fractionated) are used before the TBI.18,26 Higher doses can be used after

Because conventional myeloablative preparative regimens for hematopoietic stem cell transplantation are often associated with significant acute toxicity and mortality, use of such regimens is usually restricted to younger, medically fit patients who have no major comorbid conditions. To reduce the incidence and severity of toxicities such as infections and those that involve the hematologic, gastrointestinal, hepatic, pulmonary, and metabolic systems, nonmyeloablative regimens have been developed with the goal of allowing older patients to undergo allogeneic stem cell transplantation.83 Nonmyeloablative regimens often involve TBI in a single 2-Gy fraction, with no shielding, in combination with postgrafting immunosuppression; the goal is to permit stable allogeneic engraftment, with the expectation that a graft-versus-tumor effect will develop and address any remaining disease.84 Although nonmyeloablative regimens have been shown to be less toxic than conventional myeloablative regimens, they also carry a greater likelihood of tumor recurrence.83,85,86

Sequelae of Total-Body Irradiation and Marrow or Stem Cell Transplantation Sequelae caused solely by TBI are somewhat difficult to identify, because that irradiation is given in conjunction with conditioning chemotherapy—which is also toxic— and the transplantation procedure. The acute and late sequelae of TBI given in the context of marrow or stem cell transplantation are discussed briefly with the understanding that many of these sequelae have multiple causes.

Acute Toxic Effects The major life-threatening acute toxicity of TBI at the ­doses used for marrow or stem cell transplantation is to the hematopoietic system. The effects of the usual TBI dose on peripheral blood counts are so drastic that death would result if stem cell support (or transplantation) were not performed.11 However, because marrow or stem cell transplantation is an integral part of the procedure, the hematopoietic toxicity of TBI is not a problem in this context. TBI also has gastrointestinal toxic effects, most commonly nausea and vomiting that begin within the first hour after irradiation begins and last up to 48 hours after the final dose. Antiemetics such as ondansetron and granisetron can reduce nausea and vomiting considerably but cannot eliminate them completely.80 Diarrhea is common after TBI, and it is aggravated by the antibiotics and chemotherapy that are part of the transplantation process. The parotid glands are acutely sensitive to radiation (see Chapter 8), and parotitis is experienced by most patients who undergo TBI. Parotitis manifests as acute swelling of the parotid glands with associated jaw or facial pain. Parotitis can be treated effectively with steroids, which are ­commonly used during transplantation. Left untreated, parotitis subsides in 2 to 3 days.

38  Radiation Therapy for Bone Marrow or Stem Cell Transplantation

Skin erythema or tanning is common after TBI at the usual 12-Gy dose level. This effect is transient, and desquamation, whether dry or moist, is rare. Oral mucositis also is common after TBI, but it is intimately associated with the conditioning chemotherapy. Conditioning chemotherapy, particularly with cyclophosphamide,7,87 is known to cause mucositis, but the TBI increases its severity. Most patients experience alopecia within 7 to 14 days after receiving TBI88; although chemotherapy contributes to hair loss, irradiation alone can cause it as well. Alopecia is usually complete but reversible, and significant regrowth of hair usually occurs within 3 to 6 months after the end of the treatment. Significant damage to vital organs can occur as an acute effect of TBI. Damage to specific organs and long-term effects are discussed in the following sections.

Long-Term Effects Hepatotoxicity

Veno-occlusive disease of the liver is a significant and often life-threatening complication of marrow or stem cell transplantation; it can occur after conditioning with TBI or chemotherapy, or both.89-91 This disease, which was first described as a clinicopathologic entity in 1977, manifests as hepatic dysfunction that usually develops within 3 weeks after marrow transplantation.91 Characteristic features include ascites and hepatomegaly associated with abdominal pain and weight gain. Histologic changes include concentric subintimal thickening and luminal narrowing in the terminal hepatic venules or small sublobular veins. Venoocclusive disease of the liver is fatal in approximately one third of affected individuals.91 The development of veno-occlusive disease of the liver depends on many factors, the most notable of which are the presence of underlying liver disease, the patient’s age, and the conditioning regimen used.91 Use of busulfan and use of cytoreductive therapy regimens that include cyclophosphamide in combination with TBI to a total dose of more than 12 Gy are risk factors for veno-occlusive disease.92,93 Veno-occlusive disease does not seem to be related to GVHD. Findings from one study63 suggest that use of single-fraction TBI was associated with increased mortality from veno-occlusive disease of the liver and that the TBI dose was related to the incidence of veno­occlusive disease. A randomized trial comparing TBI to 12 Gy (given at 2 Gy/day) with TBI to 15.75 Gy (given at 2.25 Gy/day) showed an increased incidence of venoocclusive disease and an increase in mortality related to veno-occlusive disease among the group given the higher dose.70 For TBI at doses that exceed 12 Gy in 6 fractions, the use of partial organ shielding should be considered to avoid liver damage.

Pulmonary Toxicity Interstitial pneumonia or pneumonitis is a common complication of TBI.94 Interstitial pneumonitis has been documented in up to 50% of all patients who have undergone transplantation, and approximately one half of those cases are fatal.95 Although pulmonary pathogens have been recovered in many cases, the pneumonitis seems to be of idiopathic origin in 30% to 40% of patients experiencing this effect after stem cell transplantation.96 Interstitial pneumonitis usually manifests as fever, nonproductive cough,

999

and some shortness of breath within 90 days after TBI. Histopathologic findings include lung edema, fibrosis, variable cellular infiltrates, and alveolar exudates. Studies have shown that the risk of developing interstitial pneumonitis after marrow or stem cell transplantation depends on patient-, disease-, and treatment-related factors.85,86,97 In one analysis of 224 patients who developed pulmonary infiltrates after an allogeneic stem cell transplantation,85 bronchoalveolar lavage, lung biopsy, or autopsy identified infectious organisms in about one half of cases. Among the 81 patients who fulfilled the criteria of idiopathic pneumonia syndrome, the following risk factors were identified: a diagnosis of acute leukemia or myelodysplastic syndrome; patient age older than 40 years; the presence of grade III through IV GVHD; conventional (as opposed to nonmyeloablative) conditioning regimen; high-dose (>12 Gy) TBI; and use of methotrexate.85 Other studies have also shown that baseline pulmonary functions such as 1-second forced expiratory volume (FEV1), patient age, smoking history, and type of conditioning regimen used (especially myeloablative regimens that include busulfan or TBI) can influence the development of interstitial pneumonitis.71,72,86,97 In addition to the risk factors previously cited, risk factors related to TBI are important in development of interstitial pneumonitis. Unquestionably, the radiation dose delivered to the lungs during TBI participates in the development of pneumonitis, especially idiopathic pneumonitis. The radiation dose, dose rate, and fractionation schedule all seem to be related to the incidence of pneumonitis.57,98 In one classic study, Keane and colleagues94 compared the incidence of interstitial pneumonitis at centers where TBI was delivered as a single 10-Gy dose and those where the dose was lower (4 to 9.5 Gy). They found that the incidence of pneumonitis of infectious or idiopathic origin was much lower when lower doses were used. A later study showed an increased incidence of lethal pulmonary complications when mean lung doses were higher than 9.4 Gy with fractionated TBI.99 A randomized trial by Clift and colleagues comparing two fractionated regimens (15.75 Gy and  12 Gy) for chronic myelogenous leukemia70 showed that the risk of fatal cytomegalovirus pneumonia was higher in the group given the higher dose of TBI. Interstitial pneumonitis can be fatal despite aggressive supportive therapy, and every effort should be made to reduce the total lung dose to minimize the risk of interstitial pneumonitis.71,72,85 A study by Sampath and colleagues71 indicated that the incidence of interstitial pneumonitis after 12-Gy TBI given in 6 daily fractions was 11% but that this rate could be substantially reduced to less than 5% with the use of 50% shielding.71 Similar results were found in another study in which the use of lung shielding led to lower pulmonary transplant–related mortality and improved overall survival.72 The use of low dose rates also seems to reduce the incidence of interstitial pneumonitis.98 In one such study, Carruthers and Wallington58 showed that the incidence of interstitial pneumonitis was higher after TBI given at 0.15 Gy/min than that after TBI given at 0.075 Gy/min. Similarly, Ozsahin and others97 showed that dose rates higher than 0.09 Gy/min were an independent risk factor for interstitial pneumonitis. In addition to total dose and dose rate, fractionation seems to affect the incidence of interstitial pneumonitis. 

1000

PART 14  Special Considerations

Shank and colleagues66 showed that fractionating the ­irradiation significantly reduced the incidence of interstitial pneumonitis from 70% for a single dose to 33% for a regimen in which 13.2 Gy was given in 11 fractions over 4 days. This beneficial effect of fractionation on the incidence of interstitial pneumonitis has also been shown elsewhere.64,100 In an effort to clarify the effect of bone marrow transplantation that includes TBI on lung function, investigators at some centers have used pulmonary function tests at selected intervals after transplantation.101-103 Results from one study indicate that the incidence of pneumonitis was associated with the conditioning regimen used and with preexisting lung damage.102 Results from a large study of pulmonary function after conditioning that included TBI at the Medical College of Wisconsin104 indicate that patients who survived for 100 days after a bone marrow transplant showed an initial decline in pulmonary function but subsequently recovered unless factors such as infection or significant GVHD were present. Radiation dose to the lung also affected pulmonary function, specifically forced vital capacity.104 TBI has a role in the development of interstitial pneumonitis after bone marrow transplantation, but it is only one of several contributory factors. For patients who are to be treated with TBI at doses that exceed the usual  12 Gy given in 6 fractions, the use of selective partial lung shielding should be seriously considered to minimize the incidence of pneumonitis.104,105

Cardiotoxicity Cardiac damage has been reported to follow marrow or stem cell transplantation, but cardiac effects generally are related more to the chemotherapy than to the irradiation.106-109 Cyclophosphamide in particular is a significant cardiotoxin, especially in high doses, and some centers recommend that TBI be used so that the dose of cyclophosphamide can be reduced.107 Cardiac toxicity can be expressed as cardiomyopathy or pericarditis, or both. The mortality rate among affected patients is approximately 10%.107,109

Renal Toxicity Unlike cardiotoxicity, renal toxicity associated with bone marrow transplantation seems to be strongly related to the TBI rather than the chemotherapy.110-113 The clinical presentation of the renal damage is consistent with radiation nephritis and includes increased serum creatinine levels, decreased glomerular filtration rates, anemia, and hypertension, all occurring approximately 6 to 20 months after bone marrow transplantation.111 Even the histologic appearance of the renal injury resembles radiation damage. However, because the total doses of radiation are rather low relative to the amount of renal damage produced, the cause of the damage probably depends on other factors in addition to the radiation. In a European study of 75 patients, Mirabell and colleagues114 found that the incidence of renal injury associated with bone marrow transplantation was related to the total radiation dose and to the dose per fraction. Similar findings were shown in a study by Delgado and others,62 who showed that in addition to age and use of fludarabine, TBI fractionation had a significant effect on the incidence of severe chronic renal failure and endstage renal disease. Because the incidence of bone marrow

transplant–related nephropathy is strongly associated with the radiation dose received by the kidneys, selective renal shielding during TBI should be considered to reduce the incidence of such nephropathy.51,115,116

Cataracts Cataracts can develop in patients who undergo marrow or stem cell transplantation. The incidence of cataract development in patients conditioned with chemotherapy only is 15% to 20%.117 However, the use of single-dose TBI in conjunction with bone marrow transplantation results in a much higher incidence of cataract formation (­approximately 80% at 5 years after transplantation).118 Fortunately, this risk can be greatly reduced (to the levels associated with chemotherapy only) by using fractionated radiation.119-122 The incidence and severity of cataract formation are also related to factors such as the type of transplant, the presence of GVHD, patient age, and use of steroids.123 If cataracts do develop and impair the patient’s vision, they can be surgically corrected.

Effects on Gonadal Function and Fertility TBI can lead to hypergonadotropic hypogonadism, which leads to increased levels of luteinizing hormone and ­folliclestimulating hormone.124-127 The radiosensitivity of the testicular germinal epithelium (see Chapter 27) means that spermatogenesis is usually reduced or absent after TBI, and radiation-induced infertility is a problem after undergoing marrow or stem cell transplantation.124 Ninety-five percent of male patients have azoospermia after TBI. Leydig cells (which secrete testosterone) are less sensitive to radiation than Sertoli cells, but up to 33% of male patients require testosterone replacement after TBI.124,125 Hypergonadotropic hypogonadism is likely in female patients after TBI.126 More than 95% of postmenarchal females become amenorrheic after bone marrow transplantation, but a few resume menstruation. The potential for recovery depends on age at transplantation, with younger patients having a better chance of gonadal recovery.125,126 Approximately 80% of girls who are premenarchal at the time of transplantation ultimately achieve menarche after the procedure; postmenarchal girls who are younger than 18 years are very likely to recover sufficient ovarian function to resume menstration.128,129 Patients should be ­informed that loss of fertility is likely but that cases of childbearing and fathering of children have been documented after transplantation.130-132

Effects on Growth and Development TBI delivered to prepubertal patients can produce problems with growth and development.101,128,133-136 Data from the University of California at San Francisco135 showed that among prepubertal patients with leukemia who are given TBI in preparation for transplantation, 80% to 90% had diminished growth, 17% had delayed or arrested puberty, and 100% had minor abnormalities in the intelligence quotient (IQ) at 1 year, with recovery within approximately 3 years. The final height achieved after bone marrow transplantation during childhood is affected by genetic, nutritional, hormonal, and psychological factors and by the type of anticancer therapy given.124,137 However, the two main reasons for decreased standard deviation scores from those of the general population reflect the effects of ­ radiation

38  Radiation Therapy for Bone Marrow or Stem Cell Transplantation

on bone epiphyses and on the hypothalamus and pituitary glands, especially if a cranial boost is used.137-139 Younger age at the time of transplantation and being male influence final height.139 TBI is thought to lead to a loss of height of approximately 1 standard deviation score from the expected (genetic) final height. Busulfan and cyclophosphamide can also affect final height, albeit to a lesser extent than TBI; however, at least one study has shown that growth retardation was equivalent among children given a conditioning regimen consisting of only chemotherapy (with busulfan and cyclophosphamide) and those given a conditioning regimen of cyclophosphamide and TBI.139,140 In another study, children with acute myeloid leukemia whose conditioning regimen consisted of busulfan and cyclophosphamide also showed abnormal gonadotropin levels, with the abnormality being more prominent in girls than in boys.141

Thyroid Dysfunction TBI, whether given as a single or fractionated dose, can cause thyroid dysfunction, which itself can affect growth and development in children. Usually, the hypothyroidism is mild and subclinical, and it may resolve spontaneously.124,127,142-144 However, the incidence of hypothyroidism does increase over time after irradiation, and the hypothyroidism can occur late; consequently, thyroid function should be followed up for several years after TBI. In a study evaluating the late effects of bone marrow transplantation for children with stage IV neuroblastoma, 60% needed thyroid replacement for thyroid dysfunction, which was most often of peripheral origin.145

Secondary Malignancies As the survival rates of patients undergoing marrow or stem cell transplantation improve, long-term complications, including secondary malignancies, are increasingly being recognized. Secondary malignancies are not uncommon after marrow or stem cell transplantation.146-150 Witherspoon and colleagues151 reported that the age-adjusted incidence of secondary malignancies after transplantation is 6.7 times that of primary cancer in the general population. Secondary malignancies can be caused by TBI, by chemotherapy, and by additional localized radiation given in preparation for transplantation.152 Three types of secondary malignancies appear after bone marrow transplantation: posttransplantation lympho­ proliferative disorders, hematologic malignancies such as transplant-related myelodysplastic syndrome, and solid tumors. In an analysis of 3372 patients, Baker and ­colleagues149 reported a cumulative incidence for the development of any posttransplantation malignancy of 6.9% at 20 years, with a 3.8% incidence of solid tumors. Although the incidence of myelodysplastic syndrome, acute myelogenous leukemia, and lymphoproliferative disorders remained stable at about 1.4% at 10 years after transplantation, the incidence of solid tumors did not reach a plateau and was 3.8% at 20 years after transplantation. B-cell lymphoproliferative disorders can occur under conditions of prolonged immunosuppression other than those associated with transplantation, but they are becoming increasingly common among bone marrow transplant recipients because of the use of mismatched transplants and T-cell depletion techniques.153,154 These B-cell ­ disorders

1001

seem to be related to the Epstein-Barr virus; when the virus infects B lymphocytes in immunocompromised patients, the lymphocytes can proliferate uncontrollably, with extensive infiltration and organ damage (possibly fatal), or they can convert to an aggressive, usually fatal form of lymphoma.126 Fortunately, the risk of developing lymphoproliferative disorders is low (0.6%), and the use of TBI in the conditioning regimen does not seem to be a causative factor.154 Although secondary leukemias can appear after bone marrow transplantation, their incidence is low compared with those of lymphomas and solid tumors.151 An association has been observed between conditioning regimens that contain TBI and the occurrence of secondary solid tumors in humans and dogs.151,155 In a series reported by Witherspoon and colleagues,151 13 of the 35 cases of secondary cancer that appeared after bone marrow transplantation were solid tumors. In a larger analysis of 19,229 patients, the risk of new solid tumors in transplant recipients was 8.3 times higher than expected at 10 years after transplantation. Multivariate analyses showed that a single fraction of 10 Gy or more or a fractionated total dose of  13 Gy or more was associated with a higher risk of solid tumor malignancies.148 Although marrow or stem cell transplantation does carry the risk of secondary malignancies, patients undergoing this procedure usually have lethal diseases for which transplantation offers the possibility of cure. In this context, the risk of second malignancies, although high, may be acceptable.

Neurocognitive and Psychological Effects Because hematopoietic cell transplantation involves several potentially neurotoxic agents, including TBI, cranial irradiation, busulfan or other myeloablative chemotherapeutic agents, and drugs used for prophylaxis or treatment of GVHD, concerns have been raised regarding long-term complications in neurocognitive functioning as the longterm survival rates after such procedures continue to increase.156 Although prophylactic cranial irradiation for the treatment of leukemia usually has been associated with long-term declines in full-scale IQ scores in children, other evidence has shown no difference between children given no cranial irradiation, those given 18 Gy, and even those given 24 Gy.81,157,158 Because cranial irradiation doses are lower when TBI is used, the effects of such irradiation would be expected to be less evident and even more difficult to demonstrate in clinical studies. A cohort of 102 children who survived transplantation showed no difference in cognitive or academic functioning among whose conditioning regimen included TBI and those who did not undergo TBI.156 However, age was an important factor in this study; children younger than 3 years were more vulnerable to neurocognitive effects and showed significant declines in IQ after marrow transplantation, regardless of whether TBI was used.156 Another study by Kramer and colleagues159 showed significant declines in IQ between baseline and 1 year; however, neither the conditioning regimen nor the radiation dose used influenced this outcome. No further declines in IQ were observed at 3-year followup evaluations.159 Psychological effects are an important aspect of the late sequelae of TBI and the transplantation process. Given the gravity of the diseases being treated and the intensity of the treatment, some psychological morbidity would be

1002

PART 14  Special Considerations

expected. A prospective quantitative psychosocial evaluation of 31 adults who underwent bone marrow transplantation was conducted by Leigh and colleagues in the United Kingdom.160 In that study, the incidence of psychosocial morbidity (measured with standardized psychological measurement tools) was 54%; moreover, morbidity was still present at 6 to 9 months after the transplantation, but the type of transplantation was unrelated to the occurrence of morbidity. Patients with chronic myelogenous leukemia fared worse in terms of psychosocial morbidity after transplantation than did patients with other diseases, probably because they had no experience with intense hematologic therapy before undergoing the transplantation.160 With regard to pediatric patients, a retrospective review of 73 children who had undergone allogeneic transplants showed that about 75% reported having few psychosocial problems after transplantation.136 Cancer patients who received adjuvant psychosocial therapy after diagnosis in another study161 showed less psychological morbidity than patients who did not. It is possible that psychological intervention would be helpful to at least some patients undergoing transplantation, and such intervention should be considered.161

Conclusions Because the morbidity associated with the use of TBI in marrow or stem cell transplantation is fairly significant, it is reasonable to question the use of TBI. According to Shank,162 TBI for cancer treatment has several theoretical advantages over chemotherapy with regard to cytoreduction: sanctuary sites such as the testes are not spared, the dose is homogeneous throughout the body and independent of blood supply, radiation has no known cross-­resistance with other agents, the effective dose is not affected by ­detoxification or excretion, and the dose distribution can be adjusted by using partial tissue blocks for radiosensitive normal tissues (e.g., lung) or by boosting the dose to radioresistant areas. With these advantages in mind, it makes sense to use TBI for immunosuppression in circumstances in which the alternative (busulfan and cyclophosphamide) is not as effective, such as for transplantation of T-cell–depleted material. Outcomes of bone marrow transplantation with or without TBI for acute leukemia have been studied by French and Japanese investigators. In the study from France, Blaise and colleagues163 showed a statistically significant improvement in disease-free survival for patients with acute myeloid leukemia who underwent TBI. In the study from Japan,164 81 patients underwent TBI and 42 did not, and a survival advantage favored those who underwent TBI. A meta-analysis comparing busulfan plus cyclophosphamide with TBI as conditioning regimens indicated that the chemotherapy-based conditioning regimen produced a higher incidence of veno-occlusive disease of the liver and suggested that overall survival and disease-free survival rates were higher among those given TBI (Fig. 38-6).165 The aforementioned evidence supports the continued use of TBI in preparation for marrow or stem cell transplantation in selected patients. Further investigation of dose rates and fractionation schedules is needed to minimize the toxic effects of TBI while maximizing its effectiveness.

Total body-irradiation superior

Busulfan superior Blaise Survival Blume Clift Ringden Devergie All studies Blaise Blume Clift Ringden Devergie

Disease-free survival

All studies 1

0.1

A

10

Odds ratio Busulfan superior

Total body-irradiation superior

Veno-occlusive disease

Blaise Blume Ringden Devergie All studies Interstitial pneumonitis

Blaise Ringden Devergie All studies 0.01

B

0.1

1

10

100

Odds ratio

FIGURE 38-6.  Results from a meta-analysis of survival and complications associated with the use of regimens that include busulfan plus cyclophosphamide versus regimens that include total-body irradiation as conditioning for bone marrow transplantation. A, Survival and disease-free survival are not statistically different, but a trend favoring the use of total-body irradiation is evident. B, Patients conditioned for transplantation with busulfan plus cyclophosphamide have a higher rate of veno-occlusive disease (and perhaps interstitial pneumonitis) than do patients conditioned with regimens that included ­total-body irradiation. (From Hartman AR, Williams SF, Dillon JJ. Survival, disease-free survival and adverse effects of conditioning for ­ allogeneic bone marrow transplantation with busulfan/­ cyclophosphamide vs total body irradiation: a meta-analysis. Bone Marrow Transplant 1998;22:439-443.)

REFERENCES    1. Odell TF, Tausche FG, Lindsley DL, et al. Homotransplantation of functional erythropoietic elements in rat following total-body irradiation. Ann N Y Acad Sci 1957;64:811-825.    2. Alpen EL, Baum SJ. Acute radiation protection of dogs by bone marrow auto transfusions. Presented at a meeting of the Radiation Research Society, May 13-15, 1957, Rochester, NY.    3. Crouch BG, Overman RR. Whole-body radiation protection in primates. Fed Proc 1957;16:27.    4. Thomas ED, Lochte HL Jr, Lu WC, et al. Intravenous infusion of bone marrow in patients receiving radiation and chemotherapy. N Engl J Med 1957;257:491-496.

38  Radiation Therapy for Bone Marrow or Stem Cell Transplantation    5. Bortin MM. A compendium of reported human bone marrow transplants. Transplantation 1970;9:571-587.    6. Trentin JJ. Mortality and skin transplantability in x-irradiated mice receiving isologous, homologous or heterologous bone marrow. Proc Soc Exp Biol Med 1956;92:688-693.    7. Thomas ED, Storb R, Clift RA, et al. Bone-marrow transplantation (second of two parts). N Engl J Med 1975;292:895-902.    8. Storb R, Deeg HJ, Thomas ED, et al. Marrow transplantation for chronic myelocytic leukemia: a controlled trial of cyclosporine versus methotrexate for prophylaxis of graft-versus-host disease. Blood 1985;66:698-702.    9. Champlin R. Symposium: treatment for victims of nuclear ­accidents: the role of bone marrow transplantation. Radiat Res 1988;113:205-210.   10. Tubiana M, Frindel E, Croizat H, et al. Effects of radiations on bone marrow. Pathol Biol (Paris) 1979;27:326-334.   11. Rubin P, Landman S, Mayer E, et al. Bone marrow regeneration and extension after extended field irradiation in Hodgkin’s disease. Cancer 1973;32:699-711.   12. Sykes HP, Chu FCH, Wilkerson WG. Local bone marrow ­changes secondary to therapeutic irradiation. Radiology 1960;75:  919-929.   13. Fajardo LF, Berthrong M, Anderson RE. Radiation and host ­defense mechanisms. In Radiation Pathology. New York: Oxford University Press, 2001, pp 87-96.   14. Greenberg ML, Chanana AD, Cronkite EP, et al. Extracorporeal irradiation of blood in man: radiation resistance of circulating platelets. Radiat Res 1968;35:147-154.   15. Del Regato JA. Total body irradiation in the treatment of chronic lymphogenous leukemia: Janeway Lecture, 1973. Am J Roentgenol Radium Ther Nucl Med 1974;120:504-520.   16. Horowitz MM, Loberiza FR, Bredeson CN, et al. Transplant registries: guiding clinical decisions and improving outcomes. ­Oncology (Williston Park) 2001;15:649-659.   17. Thomas ED, Buckner CD, Banaji M, et al. One hundred ­patients with acute leukemia treated by chemotherapy, total body irradiation, and allogeneic marrow transplantation. Blood 1977;49:511-533.   18. Ash RC, Casper JT, Chitambar CR, et al. Successful allogeneic transplantation of T-cell-depleted bone marrow from closely HLA-matched unrelated donors. N Engl J Med 1990;322:  485-494.   19. Tallman MS, Kopecky KJ, Amos D, et al. Analysis of prognostic factors for the outcome of marrow transplantation or further chemotherapy for patients with acute nonlymphocytic leukemia in first remission. J Clin Oncol 1989;17:326-337.   20. Doney K, Buckner D, Sales GE, et al. Treatment of chronic granulocytic leukemia by chemotherapy, total body irradiation and allogeneic bone marrow transplantation. Exp Hematol 1978;6:738-747.   21. Fefer A, Cheever MA, Thomas ED, et al. Disappearance of Ph1positive cells in four patients with chronic granulocytic leukemia after chemotherapy, irradiation and marrow transplantation from an identical twin. N Engl J Med 1979;300:333-337.   22. Thomas ED. Total body irradiation regimens for marrow grafting. Int J Radiat Oncol Biol Phys 1990;19:1285-1288.   23. Gratwohl A, Hermans J, Niederwieser D, et al. Bone marrow transplantation for chronic myeloid leukemia: long-term ­results. Chronic Leukemia Working Party of the European Group for Bone Marrow Transplantation. Bone Marrow Transplant 1993;12:509-516.   24. Jones RJ. Biology and treatment of chronic myeloid leukemia. Curr Opin Oncol 1997;9:3-7.   25. Carella AM, Congiu AM, Gaozza E, et al. High-dose chemotherapy with autologous bone marrow transplantation in 50 advanced resistant Hodgkin’s disease patients: an Italian study group report. J Clin Oncol 1988;6:1411-1416.   26. Lundberg JH, Hansen RM, Chitambar CR, et al. Allogeneic bone marrow transplantation for relapsed and refractory lymphoma using genotypically HLA-identical and alternative donors. J Clin Oncol 1991;9:1848-1859.

1003

  27. Gulati SC, Shank B, Black P, et al. Autologous bone marrow transplantation for patients with poor-prognosis lymphoma.  J Clin Oncol 1988;6:1303-1313.   28. Freedman AS, Takvorian E, Neuberg D, et al. Autologous bone marrow transplantation in poor-prognosis intermediate-grade and high-grade B-cell non-Hodgkin’s lymphoma in first remission: a pilot study. J Clin Oncol 1993;11:931-936.   29. Gianni AM, Bregni M, Siena S, et al. High-dose chemotherapy and autologous bone marrow transplantation compared with MACOP-B in aggressive B-cell lymphoma. N Engl J Med 1997;336:1290-1297.   30. Good RA. Bone marrow transplantation symposium: bone marrow transplantation for immunodeficiency disease. Am J Med Sci 1987;294:68-74.   31. Ramsay NKC, Kim T, Nesbit ME, et al. Total lymphoid irradiation and cyclophosphamide as preparation for bone marrow transplantation in severe aplastic anemia. Blood 1980;55:  344-346.   32. Bacigalupo A, Oneto R, Bruno B, et al. Current results of bone marrow transplantation in patients with acquired severe aplastic anemia. Report of the European Group for Blood and Marrow Transplantation. Acta Haematol 2000;103:19-25.   33. Margolis DA, Casper JT. Alternative-donor hematopoietic stemcell transplantation for severe aplastic anemia. Semin Hematol 2000;37:43-55.   34. Halperin EC, Constine LS, Tarbell NJ, et al (eds): Pediatric Radiation Oncology, 3rd ed. Philadelphia: Lippincott Williams & Wilkins, 1999, pp 163-202.   35. Marcus RB, Graham-Pole JR, Springfield DS, et al. High-risk ­Ewing’s sarcoma: end-intensification using autologous bone marrow transplantation. Int J Radiat Oncol Biol Phys 1988;15:53-59.   36. Ladenstein R, Laset C, Pinkerton R, et al. Impact of megatherapy in children with high-risk Ewing’s tumors in complete remission: a report from EBMT Solid Tumour registry. Bone Marrow Transplant 1995;15:697-705.   37. Ladenstein R, Gadner H, Hartmann O, et al. The European ­experience with megadose therapy and autologous bone marrow transplantation in solid tumors with poor prognosis (Ewing sarcoma, germ cell tumors and brain tumors). Wien Med Wochenschr 1995;145:55-57.   38. Bonadonna G, Gianni AM. High-dose chemotherapy and autologous bone marrow transplant for adjuvant treatment of poorrisk breast cancer. Oncol J Club 1990;2:3-11.   39. Peters WP, Shpall EJ, Jones RB, et al. High-dose combination alkylating agents with bone marrow support as initial treatment for metastatic breast cancer. J Clin Oncol 1988;6:1368-1376.   40. Shea T, Bitran J, Abboud C, et al. Multi-center study with multiple cycles of PBSC supported high-dose therapy for stage IV breast cancer [abstract 230]. Proc Am Soc Clin Oncol 2000;19:59a.   41. Klein JL, Dansey R, Karanes C, et al. High-dose chemotherapy (STAMP 1) and peripheral blood progenitor cell transplantation following Adriamycin induction chemotherapy for high-risk primary and locally advanced breast cancer [abstract 234]. Proc Am Soc Clin Oncol 2000;19:60a.   42. Sundaram S, Lancaster D, Silva J, et al. Oral capecitabine is an active and well tolerated primary palliative treatment in ­patients with metastatic breast cancer who progress after high dose chemotherapy and autologous stem cell support [abstract 223]. Proc Am Soc Clin Oncol 2000;19:58a.   43. Martinelli G, Cinieri S, Ferrucci P, et al. Multiple high-dose chemotherapy with docetaxel, epirubicin and cyclophosphamide for high risk breast cancer: clinical results of a phase II study [abstract 218]. Proc Am Soc Clin Oncol 2000;19:56a.   44. Heublein AC. A preliminary report on continuous irradiation of the entire body. Radiology 1932;18:1051-1062.   45. Kim TH, Khan FM, Galvin JM. A report of the work party: comparison of total body irradiation techniques for bone marrow transplantation. Int J Radiat Oncol Biol Phys 1980;6:779-784.   46. Van Dyk J. Magna field irradiation: physical considerations. Int J Radiat Oncol Biol Phys 1983;9:915.

1004

PART 14  Special Considerations

  47. Findley DO, Skov DD, Blume KG. Total body irradiation with a 10 MV linear accelerator in conjunction with bone marrow transplantation. Int J Radiat Oncol Biol Phys 1980;6:695-702.   48. Glasgow GP, Mill WB, Phillips GL II, et al. Comparative 60CO total body irradiation (220 cm SAD) and 25-meV total body irradiation (370 cm SAD) dosimetry. Int J Radiat Oncol Biol Phys 1980;6:1243-1250.   49. Lawton CA, Barber-Derus SW, Murray KJ, et al. Technical modifications in hyperfractionated total body irradiation for T-­lymphocyte-depleted bone marrow transplant. Int J Radiat ­Oncol Biol Phys 1989;17:319-322.   50. Kolb HJ, Rieder I, Bodenberger U, et al. Dose rate and dose fractionation studies in total body irradiation of dogs. Pathol Biol (Paris) 1979;27:370-372.   51. Lawton CA, Barber-Derus SW, Murray KJ, et al. Influence of renal shielding on the incidence of late renal dysfunction associated with bone marrow transplantation in adult patients. Int J Radiat Oncol Biol Phys 1992;23:681-686.   52. Lapidot T, Singer TS, Salomon O, et al. Booster irradiation to the spleen following total body irradiation: a new immunosuppressive approach for allogeneic bone marrow transplantation.  J Immunol 1988;141:2619-2624.   53. Krebs JS, Jones DCL. The LD50 and the survival of bone-marrow colony forming cells in mice: effect of rate of exposure to ionizing radiation. Radiat Res 1972;5:374-380.   54. Lichter AS, Tracy D, Lam WC, et al. Total body irradiation in bone marrow transplantation: the influence of fractionation and delay of marrow infusion. Int J Radiat Oncol Biol Phys 1980;6:301-309.   55. Peters LJ, Withers HR, Cundiff JH, et al. Radiobiological considerations in the use of total body irradiation for bone-marrow transplantation. Radiology 1979;131:243-247.   56. Travis EL, Peters LJ, McNeill J, et al. Effect of dose-rate on ­total body irradiation: lethality and pathologic findings. Radiother Oncol 1985;4:341-351.   57. Weiner RS, Bortin MM, Gale RP, et al. Interstitial pneumonitis after bone marrow transplantation. Ann Intern Med 1986;  104:168-175.   58. Carruthers SA, Wallington MM. Total body irradiation and pneumonitis risk: a review of outcomes. Br J Cancer 2004;90:2080-2084.   59. Kim TH, McGlave PB, Ramsay N, et al. Comparison of two total body irradiation regimens in allogeneic bone marrow transplantation for acute non-lymphoblastic leukemia in first remission. Int J Radiat Oncol Biol Phys 1990;19:889-897.   60. Song CW, Kim TH, Khan FM, et al. Radiobiological basis of ­total body irradiation with different dose rate and fractionation: ­repair capacity of hemopoietic cells. Int J Radiat Oncol Biol Phys 1981;7:1695-1701. 60a. Thomas ED, Clift RA, Hersman J, et al. Marrow transplantation for acute nonlymphoblastic leukemia in first remission using fractionated or single-dose irradiation. Int J Radiat Oncol Biol Phys 1982;8:817-821.   61. Socie G, Stone JV, Wingard JR, et al. Long-term survival and late deaths after allogeneic bone marrow transplantation.  N Engl J Med 1999;341:14-21.   62. Delgado J, Cooper N, Thompson K, et al. The importance of age, fludarabine, and total body irradiation in the incidence and severity of chronic renal failure after allogeneic hematopoietic cell transplantation. Biol Blood Marrow Transplant 2006;12:75-83.   63. Girisnky T, Benhamou E, Bourhis JH, et al. Prospective randomized comparison of single-dose versus hyperfractionated ­total-body irradiation in patients with hematologic malignancies.  J Clin Oncol 2000;18:891-986.   64. Cosset JM, Baume D, Pico JL, et al. Single dose versus hyperfractionated total body irradiation before allogeneic bone marrow transplantation: a nonrandomized comparative study of 54 patients at the Institut Gustave-Roussy. Radiother Oncol 1989;15:151-160.   65. Deeg HJ, Storb R, Longton G, et al. Single dose or fractionated total body irradiation and autologous marrow transplantation in dogs: effects of exposure rate, fraction size, and fractionation ­interval on acute and delayed toxicity. Int J Radiat Oncol Biol Phys 1988;15:647-653.

  66. Shank B, Hopfan S, Kim JH, et al. Hyperfractionated total body  irradiation for bone marrow transplantation. I. Early results in leuk­ emia patients. Int J Radiat Oncol Biol Phys 1981;7:1109-1115.   67. Soderling CCB, Song CW, Blazar BR, et al. A correlation ­between conditioning and engraftment in recipients of MHC­mismatched T-cell-depleted murine bone marrow transplants. J Immunol 1985;135:941-946.   68. Trigg ME, Billing R, Sondel PM, et al. Clinical trial depleting T lymphocytes from donor marrow for matched and mismatched allogeneic bone marrow transplants. Cancer Treat Rep 1985;69:377-386.   69. Clift RA, Buckner CD, Appelbaum FR, et al. Allogeneic marrow transplantation in patients with acute myeloid leukemia in first remission: a randomized trial of two irradiation regimens. Blood 1990;76:1867-1871.   70. Clift RA, Buckner CD, Appelbaum FR, et al. Allogeneic marrow transplantation in patients with chronic myeloid leukemia in the chronic phase: a randomized trial of two irradiation regimens. Blood 1991;177:1660-1665.   71. Sampath S, Schultheiss TE, Wong J. Dose response and factors related to interstitial pneumonitis after bone marrow transplant. Int J Radiat Oncol Biol Phys 2005;63:876-884.   72. Sarvani BN, Montero A, Wu C. Prediction and prevention of transplant-related mortality from pulmonary causes after total body irradiation and allogeneic stem cell transplantation. Biol Blood Marrow Transplant 2005;11:223-230.   73. Reiff JE, Werner-Wasik M, Valicenti RK, et al. Changes in the size and location of kidneys from the supine to standing positions and the implications for block placement during total body irradiation. Int J Radiat Oncol Biol Phys 1999;45:447-449.   74. McAfee SL, Powell SN, Colby C, et al. Dose-escalated total body irradiation and autologous stem cell transplantation for refractory hematologic malignancy. Int J Radiat Oncol Biol Phys 2002;53:151-156.   75. Wong JYC, Liu A, Schultheiss T, et al. Targeted total marrow irradiation using three-dimensional image-guided ­ tomographic intensity-modulated radiation therapy: an alternative to standard total body irradiation. Biol Blood Marrow Transplant 2006;12:306-315.   76. Clift RA, Buckner CD, Thomas FD, et al. Treatment of chronic granulocytic leukaemia in chronic phase by allogeneic marrow transplantation. Lancet 1982;2:621-623.   77. Goldman JM, Baughan AS, McCarthy DM, et al. Marrow transplantation for patients in the chronic phase of chronic granulocytic leukaemia. Lancet 1982;2:623-625.   78. Gratwohl A, Hermans J, Biezen AV, et al. No advantage for ­patients who receive splenic irradiation before bone marrow transplantation for chronic myeloid leukemia. Bone Marrow Transplant 1992;10:147-152.   79. Gratwohl A, Hermans J, Biezen A, et al. Splenic irradiation ­before bone marrow transplantation for chronic myeloid leukemia. Chronic Leukemia Working Party of the European Group for Blood and Marrow Transplantation (EBMT). Br J Haematol 1996;95:494-500.   80. Jabro G, Koc Y, Boyle T, et al. Role of splenic irradiation in patients with chronic myeloid leukemia undergoing allogeneic bone marrow transplantation. Biol Blood Marrow Transpl 1999;5:173-179.   81. Halberg FE, Kramer JH, Moore IM, et al. Prophylactic cranial irradiation dose effects on late cognitive function in children treated for acute lymphoblastic leukemia. Int J Radiat Oncol Biol Phys 1992;22:13-16.   82. Alexander BM, Wechsler D, Braun TM, et al. Utility of cranial boost in addition to total body irradiation in the treatment of high risk acute lymphoblastic leukemia. Int J Radiat Oncol Biol Phys 2005;63:1191-1196.   83. Diaconescu R, Flowers CR, Storer B, et al. Morbidity and ­mortality with nonmyeloablative compared with myeloablative conditioning before hematopoietic cell transplantation from HLA-matched related donors. Blood 2004;104:1550-1558.   84. McSweeney PA, Niederwieser D, Shizuru JA, et al. Hematopoietic cell transplantation in older patients with hematologic ­malignancies: replacing high-dose cytotoxic therapy with graftversus-tumor effects. Blood 2001;97:3390-3400.

38  Radiation Therapy for Bone Marrow or Stem Cell Transplantation   85. Fukuda T, Hackman RC, Guthrie KA, et al. Risks and outcomes of idiopathic pneumonia syndrome after nonmyeloablative and conventional conditioning regimens for allogeneic hematopoietic stem cell transplantation. Blood 2003;102:2777-2785.   86. Chien JW, Maris MB, Sandmaier BM, et al. Comparison of lung function after myeloablative and 2 Gy of total body irradiationbased regimens for hematopoietic stem cell transplantation. Biol Blood Marrow Transplant 2005;11:288-296.   87. Thomas ED, Storb R, Clift RA, et al. Bone-marrow transplantation. N Engl J Med 1975;292:832-843.   88. Thomas ED, Sanders JE, Flournoy N, et al. Marrow transplantation for patients with acute lymphoblastic leukemia in remission. Blood 1979;54:468-476.   89. Ganem G, Saint-Marc Girardin MF, Kuentz M, et al. Venocclusive disease of the liver after allogeneic bone marrow transplantation in man. Int J Radiat Oncol Biol Phys 1988;14:879-884.   90. Jones RJ, Lee KSK, Beschorner WE, et al. Venocclusive disease of the liver following bone marrow transplantation. Transplantation 1987;44:778-783.   91. Shulman HM, McDonald GB, Matthews D, et al. An analysis of hepatic venoocclusive disease and centrilobular hepatic degeneration following bone marrow transplantation. Gastroenterology 1980;79:1178-1191.   92. McDonald GB, Hinds MS, Fisher LD, et al. Veno-occlusive disease of the liver and multiorgan failure after bone marrow transplantation: a cohort study of 355 patients. Ann Intern Med 1993;118:255-267.   93. Vogelsang GB, Dalal J. Hepatic venoocclusive disease in blood and bone marrow transplantation in children: incidence, risk factors and outcome. J Pediatr Hematol Oncol 2002;24:706-709.   94. Keane TJ, Van Dyk J, Rider WD. Idiopathic interstitial pneumo­ nia following bone marrow transplantation: the relationship with total body irradiation. Int J Radiat Oncol Biol Phys 1981;7:  1365-1370.   95. Gluckman E, Devergie A, Dutreix A, et al. Total body irradiation in bone marrow transplantation. Pathol Biol (Paris) 1979;27:  349-352.   96. Neiman PE, Reeves W, Ray G, et al. A prospective analysis of interstitial pneumonia and opportunistic viral infection among recipients of allogeneic bone marrow grafts. J Infect Dis 1977;  136:754-767.   97. Ozsahin M, Belkacemi Y, Pene F, et al. Interstitial pneumonitis following autologous bone-marrow transplantation conditioned with cyclophosphamide and total-body irradiation. Int J Radiat Oncol Biol Phys 1996;34:71-77.   98. Bortin MM, Kay HEM, Gale RP, et al. Factors associated with interstitial pneumonitis after bone marrow transplantation for acute leukaemia. Lancet 1982;1:437-439.   99. Volpe AD, Ferreri AJM, Annaloro C, et al. Lethal pulmonary complications significantly correlate with individually assessed mean lung dose in patients with hematologic malignancies treated with total body irradiation. Int J Radiat Oncol Biol Phys 2002;52:483-488. 100. Latini P, Aristei C, Aversa F, et al. Lung damage following bone marrow transplantation after hyperfractionated total body irradiation. Radiother Oncol 1991;22:127-132. 101. Deeg HJ. Acute and delayed toxicities of total body irradiation. Int J Radiat Oncol Biol Phys 1983;9:1933-1939. 102. Springmeyer SC, Silvestri R, Flournoy N, et al. Pulmonary function of marrow transplant patients. Exp Hematol 1984;12:  805-810. 103. Tait RC, Burnett AK, Robertson AG, et al. Subclinical pulmonary function defects following autologous and allogeneic bone marrow transplantation: relationship to total body irradiation and graft-versus-host disease. Int J Radiat Oncol Biol Phys 1991;20:1219-1227. 104. Gore EM, Lawton CA, Ash RC, et al. Pulmonary function changes in long-term survivors of bone marrow transplantation. Int J Radiat Oncol Biol Phys 1996;36:67-75. 105. Labar B, Bogdanic V, Nemet D, et al. Total body irradiation with or without lung shielding for allogeneic bone marrow transplantation. Bone Marrow Transplant 1992;9:343-347.

1005

106. Braverman AC, Antin JH, Plappert MT, et al. Cyclophosphamide cardiotoxicity in bone marrow transplantation: a prospective evaluation of new dosing regimens. J Clin Oncol 1991;9:  1215-1223. 107. Cazin B, Gorin NC, Laporte JP, et al. Cardiac complications after bone marrow transplantation: a report on a series of 63 consecutive transplantations. Cancer 1986;57:2061-2069. 108. Lele SS, Durrant STS, Atherton JJ, et al. Demonstration of late cardiotoxicity following bone marrow transplantation by assessment of exercise diastolic filling characteristics. Bone Marrow Transplant 1996;17:1113-1118. 109. Hertenstein B, Stefanic M, Schmeiser T, et al. Cardiac toxicity of bone marrow transplantation: predictive value of cardiologic evaluation before transplant. J Clin Oncol 1994;12:998-1004. 110. Juckett M, Perry EH, Daniels BS, et al. Hemolytic uremic syndrome following bone marrow transplantation. Bone Marrow Transplant 1991;7:405-409. 111. Lawton CA, Cohen EP, Barber-Derus SW, et al. Late renal dysfunction in adult survivors of bone marrow transplantation. Cancer 1991;67:2795-2800. 112. Tarbell NJ, Guinan EC, Niemeyer C, et al. Late onset of renal dysfunction in survivors of bone marrow transplantation. Int J Radiat Oncol Biol Phys 1988;15:99-104. 113. Van Why SK, Friedman AL, Wei LJ, et al. Renal insufficiency after bone marrow transplantation in children. Bone Marrow Transplant 1991;7:383-388. 114. Mirabell R, Bieri S, Mermillod B, et al. Renal toxicity after ­allogenic bone marrow transplantation: the combined effects of total-body irradiation and graft-versus-host disease. J Clin Oncol 1996;14:579-585. 115. Lawton CA, Cohen EP, Murray KJ, et al. Long-term results of selective renal shielding in patients undergoing total body irradiation in preparation for bone marrow transplantation. Bone Marrow Transplant 1997;20:1069-1074. 116. Casper J, Camitta B, Truitt R, et al. Unrelated bone marrow ­donor transplants for children with leukemia or myelodysplasia. Blood 1995;85:2354-2363. 117. Bray LC, Carey PJ, Proctor SJ, et al. Ocular complications of bone marrow transplantation. Br J Ophthalmol 1991;75:611-614. 118. Gordon KB, Char DH, Sagerman R. Late effects of radiation on the eye and ocular adnexa. Int J Radiat Biol Oncol Phys 1995;31:1123-1139. 119. Deeg HJ, Flournoy N, Sullivan KM, et al. Cataracts after total body irradiation and marrow transplantation: a sparing effect of dose fractionation. Int J Radiat Oncol Biol Phys 1984;10:  957-964. 120. Benyunes MC, Sullivan KM, et al. Cataracts after bone marrow transplantation: long/term follow-up of adults treated with fractionated total body irradiation. Int J Radiat Oncol Biol Phys 1995;32:661-670. 121. Belkacemi Y, Labopin M, Vernant JP, et al. Cataracts after total body irradiation and bone marrow transplantation in patients with acute leukemia in complete remission: a study of the European Group for Blood and Marrow Transplantation. Int J Radiat Oncol Biol Phys 1998;41:659-668. 122. Zierhut D, Lohr F, Schraube P, et al. Cataract incidence after ­total body irradiation. Int J Radiat Oncol Biol Phys 2000;46:  131-135. 123. Van Kempen-Harteveld M, Struikmans H, Kal HB, et al. ­Cata­ract-free interval and severity of cataract after total body irradiation and bone marrow transplantation: influence of treatment parameters. Int J Radiat Oncol Biol Phys 2000;48:  807-815. 124. Socie G, Salooja N, Cohen A, et al. Nonmalignant late effects after allogeneic stem cell transplantation. Blood 2003;101:  3373-3385. 125. Faraci M, Barra S, Cohen A, et al. Very late nonfatal ­consequences of fractionated TBI in children undergoing bone marrow transplant. Int J Radiat Oncol Biol Phys 2005;63:1568-1575. 126. Mertens AC, Ramsay NKC, Kouris S, et al. Patterns of gonadal dysfunction following bone marrow transplantation. Bone Marrow Transplant 1998;23:345-350.

1006

PART 14  Special Considerations

127. Kauppila M, Koskinen P, Irjala K, et al. Long-term effects of ­allogeneic bone marrow transplantation on pituitary, gonad, thyroid and adrenal function in adults. Bone Marrow Transplant 1998;22:331-337. 128. Spinelli S, Chiodi S, Bacigalupo A, et al. Ovarian recovery after total body irradiation and allogenic bone marrow transplantation: long-term follow-up of 79 females. Bone Marrow Transplant 1994;14:373-380. 129. Mertens AC, Ramsay NKC, Kouris S, et al. Patterns of gonadal dysfunction following bone marrow transplantation. Bone Marrow Transplant 1998;22:345-350. 130. Giri N, Vowels MR, Barr AL, et al. Successful pregnancy after total body irradiation and bone marrow transplantation for acute leukaemia. Bone Marrow Transplant 1992;10:93-95. 131. Gulati SC, Van Poznak C. Pregnancy after bone marrow transplantation. J Clin Oncol 1998;16:1978-1985. 132. Wang WS, Tzeng CH, Ksieh RK, et al. Successful pregnancy following very high-dose total body irradiation (1575 cGy) and bone marrow transplantation in a woman with acute myeloid leukemia. Bone Marrow Transplant 1998;21:415-417. 133. Sanders J. Effects of cyclophosphamide and total body irradiation on ovarian and testicular function [abstract]. Exp Hematol 1982;10(Suppl 11):49. 134. Sanders JE, Pritchard S, Mahoney P, et al. Growth and development following marrow transplantation for leukemia. Blood 1986;68:1129-1135. 135. Chou RH, Wong GB, Kramer JH, et al. Toxicities of total-body irradiation for pediatric bone marrow transplantation. Int J ­Radiat Oncol Biol Phys 1996;34:843-851. 136. Matthes-Martin S, Lamche M, Ladenstein R, et al. Organ toxicity and quality of life after allogeneic bone marrow transplantation in pediatric patients: a single centre retrospective analysis. Bone Marrow Transplant 1999;23:1049-1053. 137. Couto-Silva AC, Trivin C, Esperou H, et al. Final height and ­gonad function after total body irradiation during childhood. Bone Marrow Transplant 2006;38:427-432. 138. Cohen A, Duell T, Socie G, et al. Nutritional status and growth after bone marrow transplantation (BMT) during childhood: EBMT late-effects working party retrospective data. Bone Marrow Transplant 1999;23:1043-1047. 139. Cohen A, Rovelli A, Bakker B, et al. Final height of patients who underwent bone marrow transplantation for hematological disorders during childhood: a study by the working party for lateeffects-EBMT. Blood 1999;93:4109-4115. 140. Wingard JR, Plotnick LP, Freemer CS, et al. Growth in children after bone marrow transplantation: busulfan plus cyclophosphamide versus cyclophosphamide plus total body irradiation. Blood 1992;79:1068-1073. 141. Afify Z, Shaw P, Clavano-Harding A, et al. Growth and endocrine function in children with acute myeloid leukemia after bone marrow transplantation using busulfan/cyclophosphamide. Bone Marrow Transplant 2000;25:1087-1092. 142. Boulad F, Bromley M, Black P, et al. Thyroid dysfunction following bone marrow transplantation using hyperfractionated radiation. Bone Marrow Transplant 1995;15:71. 143. Al-Fiar FZ, Colwill R, Lipton JH, et al. Abnormal thyroid ­stimulating hormone (TSH) levels in adults following allogeneic bone marrow transplants. Bone Marrow Transplant 1997;19:  1019-1022. 144. Thomas O, Mahé M, Campion L, et al. Long-term complications of total body irradiation in adults. Int J Radiat Oncol Biol Phys 2001;49:125-131. 145. Flandin I, Hartmann O, Michon J, et al. Impact of TBI on late effects in children treated by megatherapy for stage IV neuroblastoma. A study of the French Society of Pediatric Oncology. Int J Radiat Oncol Biol Phys 2006;64:1424-1431. 146. Brown JR, Yeckes H, Friedberg JW, et al. Increasing incidence of late second malignancies after conditioning with cyclophosphamide and total-body irradiation and autologous bone marrow transplantation for non-Hodgkin’s lymphoma. J Clin Oncol 2005;23:2208-2214.

147. Hasegawa W, Pond GR, Rifkind JT, et al. Long-term follow-up of secondary malignancies in adults after allogeneic bone marrow transplantation. Bone Marrow Transplant 2005;35:51-55. 148. Curtis RE, Rowlings PA, Deeg HJ, et al. Solid cancers after bone marrow transplantation. N Engl J Med 1997;336:897-904. 149. Baker KS, Defor TE, Burns LJ, et al. New malignancies after blood or marrow stem-cell transplantation in children and adults: incidence and risk factors. J Clin Oncol 2003;21:13521358. 150. Kolb HJ, Socie G, Duell T, et al. Malignant neoplasms in longterm survivors of bone marrow transplantation. Ann Intern Med 1999;131:738-744. 151. Witherspoon RP, Fisher LD, Schoch G, et al. Secondary cancers after bone marrow transplantation for leukemia or aplastic anemia. N Engl J Med 1989;321:784-789. 152. Kirova Y, Rafi H, Voisin M-C., et al. Radiation-induced bone ­sarcoma following total body irradiation: role of additional ­radiation on localized areas. Bone Marrow Transplant 2000;25:  1011-1013. 153. Shapiro RS, McClain K, Frizzera G, et al. Epstein-Barr virusassociated B cell lymphoproliferative disorders following bone marrow transplantation. Blood 1988;71:1234-1243. 154. Zutter MM, Martin PJ, Sale GE, et al. Epstein-Barr virus lymphoproliferation after bone marrow transplantation. Blood 1988;72:520-529. 155. Deeg HJ, Storb R, Prentice R, et al. Increased cancer risk in ­canine radiation chimeras. Blood 1980;55:233-239. 156. Phipps S, Dunavant Maggi, Srivastava DK, et al. Cognitive and academic functioning in survivors of pediatric bone marrow transplantation. J Clin Oncol 2000;18:1004-1011. 157. Mulhern RK, Fairclough D, Ochs J. A prospective comparison of neuropsychologic performance of children surviving leukemia who received 18-Gy, 24-Gy, or no cranial irradiation. J Clin Oncol 1991;9:1348-1356. 158. Langer T, Martus P, Ottensmeier H, et al. CNS late-effects after all therapy in childhood. Part III: neuropsychological performance in long-term survivors of childhood ALL: impairments of concentration, attention and memory. Med Pediatr Oncol 2002;38:320-328. 159. Kramer JH, Crittenden MR, DeSantes K, et al. Cognitive and adaptive behavior 1 and 3 years following bone marrow transplantation. Bone Marrow Transplant 1997;19:607-613. 160. Leigh S, Wilson KCM, Burns R, et al. Psychosocial morbidity in bone marrow transplant recipients: a prospective study. Bone Marrow Transplant 1995;16:635-640. 161. Greer S, Moorey S, Baruch JD, et al. Adjuvant psychological therapy for patients with cancer: a prospective randomized trial. BMJ 1992;304:675-679. 162. Shank B. Total body irradiation for marrow or stem-cell transplantation. Cancer Invest 1998;16:397-404. 163. Blaise D, Maraninchi D, Archimbaud E, et al. Allogeneic bone marrow transplantation for acute myeloid leukemia in first ­remission: a randomized trial of a busulfan-Cytoxan versus ­Cytoxan-total body irradiation as preparative regimen: a report from the Group d’Etudes de la Greefe de Moelle Osseuse. Blood 1992;79:2578-2582. 164. Inoue T, Ikeda H, Yamazaki H, et al. Role of total body irradiation as based on the comparison of preparation regimens for ­allogeneic bone marrow transplantation for acute leukemia in first complete remission. Strahlenther Onkol 1993;169:250-255. 165. Hartman AR, Williams SF, Dillon JJ. Survival, disease-free survival and adverse effects of conditioning for allogeneic bone marrow transplantation with busulfan/cyclophosphamide vs total body irradiation: a meta-analysis. Bone Marrow Transplant 1998;22:439-443.

ADDITIONAL READING Thomas ED. Bone marrow transplantation: past experiences and future prospects. Semin Oncol 1992;19:3-6.

Palliative Care 39 Nora A. Janjan, Marc E. Delclos, and Christopher H. Crane SCOPE AND SOCIOECONOMIC COSTS OF PALLIATIVE CARE Debilitation and Financial Burden End-of-Life Care Pain Treatment Prostate Cancer PROGNOSTIC FACTORS AFFECTING PALLIATIVE CARE Metastasis Quality of Life Estimating Survival Duration PALLIATIVE VERSUS CURATIVE CARE: PHILOSOPHICAL AND ECONOMIC DIFFERENCES Treatment Modalities Evaluating Outcomes of Palliative Therapy SYMPTOM RELIEF IN LOCALLY ADVANCED OR METASTATIC DISEASE Pain Categories Pharmacologic Principles Clinical Considerations

BONE METASTASES Symptom Management Morbidity Radiation Tolerance of the Spinal Cord Treating Disseminated Bone Metastases BRAIN METASTASES LUNG TUMORS OR METASTASES HEAD AND NECK TUMORS PELVIC DISEASE SKIN AND SUBCUTANEOUS TISSUE INVOLVEMENT DIAGNOSTIC AND THERAPEUTIC RECOMMENDATIONS

Between 1990 and 2008, 23.7 million cases of cancer were diagnosed, and about 10 million people died of the disease. Approximately one half of the patients diagnosed with cancer develop metastatic disease. More than 70% of all cancer patients develop symptoms from the primary ­tumor or metastatic disease.1-4 The total number of cancer deaths per year has been rising steadily since 2000.5 In the United States, cancer is the second most common cause of death, accounting for 23% of all deaths in 1998, and is the leading cause of death among women between the ages of 40 and 79 years. In 2008, an estimated 1,437,180 new cases of cancer were diagnosed, and 565,650 patients died of the disease, which is more than 1500 people each day. Cancer of the lung, colon or rectum, pancreas, and prostate and non-Hodgkin lymphoma constitute more than 50% of all cancer deaths. Palliation represents a large component of cancer treatment and includes the use of therapeutic and supportive care measures. Unlike other aspects of cancer therapy, ­tumor control and survival are not the only end points of success in palliative care. Quality of life is recognized as an end point secondary in importance only to survival.6,7 The goal of palliative care is to effectively and efficiently relieve symptoms and to maintain maximum quality of life for the duration of the patient’s life.2-4,8-13 Effective palliation of cancer-related pain can enhance quality of life by improving functional and emotional well-being. The effectiveness of palliative therapy is measured in terms of the percentage of patients who continue to experience persistent or recurrent symptoms and whether those symptoms can

be relieved by using other therapeutic modalities.12,14,15 Like other aspects of cancer care, effective palliative care ­requires a multidisciplinary approach.16,17

Scope and Socioeconomic Costs of Palliative Care Debilitation and Financial Burden The debility that results from cancer and cancer treatment is an important socioeconomic issue. Among more than 9700 community-based Medicare beneficiaries surveyed in a study by Stafford and Cyr,18 the 1647 people who had cancer reported having poorer health, more limitations on the activities of daily living, and greater use of health care resources than those who did not have cancer. ­Cancer patients reported difficulty walking (38%), getting out of a chair (21%), completing heavy housework (34%), and shopping (17%). Poorer health was observed more often among patients with lung, breast, prostate, or colon ­cancer. Having lung, bladder, or prostate cancer predicted increased health care use, and lung cancer was most often associated with limitation in activities of daily living. Even though the decrease in functional capacity among cancer patients is not as prolonged as that among patients with other chronic diseases (e.g., arthritis, stroke, emphysema), the annual health care costs for cancer patients are higher than the health care costs for those with other chronic diseases.19 The mean annual Medicare reimbursement for lung cancer was more than twice that for colon,

1007

1008

PART 14  Special Considerations

breast, or prostate cancer. Assuming that not all of the people in the Medicare review who had cancer had active disease at the time of the analysis, the proportion of cancer patients who are functionally impaired and using more health care resources would be even greater if only those with active disease were included in the analysis. Approximately 300,000 patients in the United States receive palliative radiation therapy each year at a total annual cost of $900 million. For comparison, about 100,000 patients receive curative radiation therapy at a total annual cost of $1.1 billion.20 The National Institutes of Health ­estimates that the overall cost of cancer in 2007 was $219.2 billion, including $89 billion in direct medical costs, $18.2 billion resulting from lost productivity due to illness, and $112 billion resulting from lost productivity due to death.5 Because health care costs are a key ­economic and political issue, these analyses are increasingly important. The goal for the medical community is to develop palliative care strategies that cost-effectively relieve suffering, increase personal independence, and prevent complications that may be caused by the disease and its treatment. Terminal illnesses impose substantial economic and other burdens on patients and those who care for them. In a study by Emanuel and colleagues,21 988 terminally ill patients and 893 caregivers from six randomly selected cities across the United States were interviewed, and transportation, nursing care, personal care, and financial needs were evaluated. The leading causes of terminal illness were cancer (51.8%), heart disease (18%), and chronic obstructive pulmonary disease (10.9%). The mean age of the participants was 66.5 years (range, 22 to 109 years), and 59.4% were at least 65 years old. Among these patients, 50.2% experienced substantial pain, 70.9% had shortness of breath after walking one block or less, 35.5% had urinary or fecal incontinence, 16.8% had symptoms of depression, and 17.5% were bedridden for more than one half of ­every day. In the 6 months before the study, 66.5% had been hospitalized, 22.3% of whom in intensive care units, and 36.8% had undergone a surgical procedure. Overall, 35% of patients had substantial health care needs that consumed more than 10% of their household income; 16% of families had to take out a loan, spend their savings, or work an additional job to cover medical costs. Patients with substantial health care needs were more likely to consider euthanasia. Administering care significantly affected the lives of more than 35% of caregivers.

End-of-Life Care Of the families included in the study by Emanuel and colleagues,21 only 28% felt burdened by the illness if the ­attending physician was involved in administering ­terminal care, compared with 42% of families whose physicians were not involved in this level of care (P = .005). However, physicians are rarely trained in how to administer end-oflife care. On average, medical textbooks devote only 2% of their total page count to end-of-life care.22 In a review of the 50 top-selling textbooks from multiple medical specialties, only 24% of textbooks—and only 38% of oncology textbooks—had helpful information on end-of-life care. Difficulties in providing adequate end-of-life care were reflected in findings from another report of physical symptoms among 350 hospice inpatients with cancer.23 The mean number of symptoms per patient increased from 6

on admission to 10 during the course of hospice care.23 The mean number of symptoms correlated directly with the patients’ performance status; patients with seven symptoms had performance status scores of 10 to 20, those with six symptoms had scores of 30 to 50, and those with four symptoms had scores of 60 or higher (a higher score indicates better functioning). Among patients with performance scores of 10 to 20, 94% had anorexia, 47% had edema, and 50% had dyspnea. In contrast, these symptoms occurred in only 29%, 16%, and 10%, respectively, of ­patients with a performance status score of 60 or more. In this study, pain was reportedly treated in 65% of patients at admission and in 88% of patients at the time of death.23

Pain Treatment Barriers to effective pain treatment include lack of ­assessment, patients’ reluctance to report pain and to use analgesics, and physicians’ reluctance to prescribe ­opioids.1,16,17,24,25 In a study of hospital inpatients referred to a pain service,16 the mean pain score on a visual analog scale (on which 10 is the worst pain imaginable) was 5.2 at presentation; 56% of the patients in this study had pain scores of 5 or higher, and 20% had pain scores of 8 or ­higher. All patients had previously used opioids, and 41 were receiving opioids at the time of consultation. Within 24 hours of consulting the pain service, the mean pain score had dropped to 2.7 (P < .05).16 Similar findings were observed among 108 patients ­referred to an outpatient multidisciplinary bone metastasis clinic.17 The median age of that population (27% of whom were working full or part time) was 55 years, and 69% of the patients were younger than 65 years. The time since diagnosis of the primary tumor ranged from 2 weeks to 23 years (median, 22 months), and metastases had been diagnosed in 30% of patients within the past 6 months. Pain was the presenting symptom in 74% of patients at diagnosis. The patients’ average pain was rated as moderate to severe by 79% of patients and severe by 21%; the patients’ worst pain was rated as severe by 78% and intolerable by 22%. Only 45% of patients experienced good relief from prescribed analgesics, and 23% indicated that the prescribed analgesics were ineffective.17 Extended life spans and improvements in screening techniques have greatly increased the number of diagnosed cases of the four most common types of cancer: lung, breast, prostate, and colorectal. All of these tumors have high rates of metastatic spread to bone and to visceral structures, which can cause pain.5,26,27 Musculoskeletal pain can occasionally also reflect an undiagnosed malignancy. In a study of 491 patients with new or recurring experiences of bone pain who had no known underlying malignancy, 4% of the entire group and 9% of those who were 50 years old or older had definite evidence of metastasis.28,29 The pain was diffuse in 57% of patients, was localized to the back in 52%, to the hips or pelvis in 16%, and to the extremities in 15%. Malignancies subsequently were diagnosed in the lung in 32% of patients and in the prostate in 16% of patients.

Prostate Cancer Prostate cancer is the second leading cause of cancer death among men,5,30,31 and pain develops during the course of the disease in 75% of cases.32 Radiographically detectable bone metastases develop in 20% of patients with stage A2

39  Palliative Care

(T1b) or B (T2) disease, 40% of patients with stage C (T3) disease, and 62% of patients with stage D1 (T4) disease presentations.33-37 In the 1970s, only 50% of prostate tumors were confined to the prostate gland at diagnosis, and metastases were present in 30% of cases at diagnosis. In a populationbased study of prostate cancer in which routine screening was not performed and only palliative intervention was provided, associations were seen among patients with ­regard to disease-specific death rate, stage of disease, and tumor differentiation.29,33 Of 719 new cases, 45% were diagnosed incidentally, only 31% had organ-confined disease, and 35% of patients had bone metastases at diagnosis. Symptoms at diagnosis included prostatism (42%), bone pain (12%), and fatigue (9%). The median age at diagnosis was 75 years. Age at diagnosis, however, did not influence disease-specific survival rates. Prostate cancer was the cause of death in 62% of these patients, and the disease-specific ­survival rates at 1, 5, and 10 years were 80%, 38%, and 17%, respectively. Even today, when patients who present with symptoms are referred to a urologist, only 70% have disease that is confined to the prostate gland; among patients undergoing prostatectomy, only 50% are confirmed to have disease that is confined to the prostate gland. Although routine screening with digital rectal examination and prostate-specific antigen measurements can detect more than 90% of ­cases, only 70% are confirmed on pathologic examination as being confined to the gland. The risk of subsequent development of distant ­metastasis is significantly lower when the primary tumor is controlled by treatment than when it is not. For example, survival rates after an isolated recurrence of prostate cancer are influenced by the initial stage of the disease and the disease-free interval between initial treatment and recurrence.34,36,37 In more than 70% of cases of locally recurrent prostate cancer, radiation can control symptoms such as hematuria, urinary outflow obstruction, ureteral obstruction, and lower extremity edema.36 With or without local recurrence, the survival rate is compromised by the presence of distant metastases. The survival rates at 5 years and 10 years after pelvic recurrence of prostate cancer are 50% and 22%, respectively. When distant metastases are present, the survival rate at 5 years is 20% and that at 10 years is less than 5%.34,37

Prognostic Factors Affecting Palliative Care Metastasis Prognosis in many types of cancer is influenced by the overall metastatic burden and the number and location of sites involved by disease. The site of the primary disease and the presence of a solitary metastatic site are predictive of more prolonged survival.38-42 The presence of bone metastasis predicts progression of disease to other sites, and it is the most common cause of cancer-related pain.5,26,27,33,35-37,43 The median duration of survival is a critical factor in evaluating the effects of palliative therapy and in determining the appropriate type of palliative therapy. When metastases are found in the lung, liver, or central nervous system in addition to the bone, the prognosis is especially poor.

1009

After bone metastasis is diagnosed, median survival duration is 12 months for patients with breast cancer, 6 months for patients with prostate cancer, and 3 months for patients with lung cancer. For patients with breast cancer, the median survival duration is 48 months if metastases are confined to the skeletal system, but it decreases to only 9 months if visceral metastasis is also present.26,27 For patients with prostate cancer, the presence and distribution of bone ­metastases have prognostic significance. Survival times are significantly longer when the metastases are restricted to the pelvis and lumbar spine and longer among patients in whom salvage hormonal therapy induces a response.34,35,43 However, any metastatic involvement outside the pelvis and lumbar spine shortens survival time, regardless of the effect of salvage hormonal therapy. In one study, a bone scan index was formulated on the basis of proportions of tumor involvement in individual bones, and that index was then examined in relation to known prognostic factors and survival in patients with androgenindependent prostate cancer.44 In multivariate ­proportional hazards analyses, only bone scan index, age, hemoglobin level, and lactate dehydrogenase level were ­ associated with survival. Survival durations were 18.3 months for patients with a bone scan index of less than 1.4%, 15.5 months for those with an index of 1.4% to 5.1%, and 8.1 months for those with an index of more than 5.1%.

Quality of Life Quality-of-life measurements can enhance the prognostic information derived from the Karnofsky performance status score and extent of disease and aid in predicting ­survival. Physical symptoms such as pain, dry mouth, constipation, change in taste, lack of appetite and energy, bloating, nausea, vomiting, weight loss, and drowsiness or dizziness portend a poor prognosis.38 In one study of 208 patients with terminal cancer (median survival duration, 15 weeks), shorter survival time was independently associated with the following factors: primary tumor site (lung cancer ­ versus breast or gastrointestinal cancer), the presence of liver metastases or comorbid conditions, weight loss of more than 8 kg in the previous 6 months, and clinical estimation by the treating physician of a survival duration of less than 2 months.45 Laboratory assessments, including serum albumin levels of less than 35 g/L, lymphocyte counts of less than 1 × 109/L, and lactate dehydrogenase levels of more than 618 U/L also were associated with a poor prognosis. After these independent factors were accounted for in the analysis, other factors such as performance status, symptoms other than nausea and vomiting, tumor burden, and socioeconomic characteristics did not independently influence the prognosis.45

Estimating Survival Duration It is important to consider issues regarding the estimation of prognosis when developing a palliative care plan. Some of these issues can affect analgesic management. Many physicians prescribe opiate analgesics only after patients are thought to have less than 6 months to live or after palliative radiation therapy fails to induce a response.1,25,46 This practice results in substantial undertreatment of cancer-related pain. First, it is difficult for physicians to make an accurate prognosis. In a review of five studies ­involving a total of 468 patients considered to have less than 6 months

1010

PART 14  Special Considerations

to live, physicians accurately predicted survival duration for only about one half of cases (range, 22% to 70% of ­cases). In these studies, the actual median survival duration was 3.5 weeks (range, 2 to 5 weeks), compared with the median survival time of 6 weeks (range, 4.5 to 8 weeks) estimated by the physicians.47 Second, patients should not have to suffer with symptoms just because their physicians believe that they may live for more than 6 months. Some of the uncertainties involved in making a prognosis were overcome in the Study to Understand ­Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT).48 This study first retrospectively evaluated the prognoses of 4301 patients and developed a model with which to predict survival duration based on the outcome of that evaluation. The model was prospectively tested with 4028 new patients.49 The model then was applied by physicians in 1757 cases to estimate the likelihood that a ­patient would survive for 2 months or for 6 months. Among the cases analyzed were 406 patients with advanced colon cancer, 764 patients with lung cancer, and 594 patients with multiple organ system failure with malignancy. The physicians’ estimates fell within 7.5% of those generated by the model; model estimates were always within 3% of the actual survival duration, and physicians’ estimates were within 9% of the actual survival duration. Although the physicians’ estimates of prognosis were generally accurate, the patients themselves tended to overestimate their survival duration. However, after patients acknowledged that they had at least a 10% chance of dying within 6 months, they were more likely to choose palliative care over therapeutic care. Despite this finding and despite efforts to educate physicians and patients about palliative care options and about end-of-life issues, many patients in that study continued to receive ineffective or inappropriate cancer treatments.47

Palliative versus Curative Care: Philosophical and Economic Differences Most of the cost of cancer care is incurred in the final months of the patient’s life. For several reasons, only a small percentage of patients with extensive disease receive innovative therapies as part of a clinical trial. With regard to administration of palliative therapy, cancer is treated as a chronic disease, and cancer symptoms are controlled in two ways—by direct treatment of the symptoms (e.g., giving the patient opioids for pain) and by treating the cause of the symptoms, the cancer. The goal of the latter method is to shrink the tumor as much as possible and thereby ­reduce symptoms. In this context, cancer treatment does not always produce a sufficient improvement in survival duration; it serves only to maintain comfort. Physicians must communicate to patients and their families the outcomes that can reasonably be achieved through the use of palliative care and begin discussions of end-of life issues. There is a growing recognition that physicians must be better trained in end-of-life issues, including hospice care, so that they can discuss these issues knowledgeably and compassionately.

Keeping the goals of palliative care in mind, palliative therapy should target symptomatic areas or areas at which morbidity caused by disease progression is likely to occur. Aggressive therapy should be reserved for patients with an extended survival prognosis, to maintain functional integrity (e.g., surgical stabilization of a pathologic fracture), or for those in a clinical trial. Therapies that cause ­morbidity and provide little to no palliative benefit should not be ­administered. The role of systemic agents (e.g., chemotherapy drugs, hormones, bisphosphonates, and radiopharmaceuticals, given alone or in combination with localized irradiation) in palliative care has not been clearly defined. Interest has increased in palliative chemotherapy with agents such as gemcitabine that provide little or no increase in survival duration but do relieve cancer-related symptoms.50,51 Many of the other agents, when used to treat bone metastases, do not provide a level of pain relief comparable with localized irradiation. Although palliation constitutes a large part of radiation therapy practice, radiation is generally underused in ­palliative care. At The University of Texas M. D. Anderson Cancer Center, about 25% of all radiation treatments are administered with palliative intent, and this has been the pattern of practice for more than 40 years.52 Consistent with practices in the United States, approximately 25% of patients in Sweden to whom radiation therapy was administered received that radiation with palliative intent; however, only 10% of all Swedish cancer patients were given any radiation.53 Because symptoms of metastases develop in more than one half of patients who have prostate, breast, or lung cancer, palliative radiation therapy is being greatly underused.

Treatment Modalities Many types of clinical problems require palliative care, and identifying the most appropriate multidisciplinary ­approach to treatment can be challenging. Among patients with bone metastases, for example, several different clinical problems, such as pathologic fracture and spinal cord compression, can occur and require different multidisciplinary interventions. Among the issues that need to be addressed are how therapies can be combined to achieve additive or synergistic effects and how to treat visceral and bone metastases. In regard to combining therapies for palliative care, as is true for curative therapies, studies are needed to evaluate the effectiveness of the combination of radiation (external beam radiation therapy [EBRT] and radiopharmaceuticals) with chemotherapy, hormonal therapy, and bisphosphonates. Cost-benefit considerations also are ­important. A Trans-Canada study found that the combination of localized irradiation and the radioisotope strontium 89 (89Sr) was more cost-effective than radiation alone.54-56 Specific to radiation therapy, studies of the costs and benefits of single-fraction radiation, radiopharmaceuticals, and systemic therapies need to be pursued. Although ­abbreviated radiation courses may result in cost savings, the patients’ responses to these modalities must be critically analyzed. The analyses must account for the type of primary tumor, location and extent of metastases, and prognosis. Ineffective palliative therapy given by any means is especially costly because unrelieved pain and disability can result in the continued need for analgesics, extended

39  Palliative Care

f­ unctional limitation, and the need for systemic therapies and surgery.57,58

Bisphosphonates Much controversy exists surrounding the cost-benefit ratio associated with using bisphosphonates to treat bone ­metastases. One study showed that the cost of pamidronate exceeded the cost savings realized from preventing the adverse events.59 The projected net cost of preventing an adverse skeletal-related event was $3940 with chemotherapy and $9390 with hormonal therapy.59 The cost per unit benefit was $108,200 with chemotherapy and $305,300 with hormonal therapy per quality-adjusted life year (QALY). A cost-analysis study conducted in Canada showed similar findings60: over a 12-month period, the total cost in the pamidronate group was 44% higher than in the placebo group. Without surgical intervention, pamidronate increased costs by $31,600/QALY. Surgical intervention, which was required by 6% of the study group in addition to the pamidronate, increased the total cost of palliative care by $13,300/QALY. Based on the number of breast cancer patients who would meet the clinical indications for receiving pamidronate, the added cost for bisphosphonates would exceed $10 million (Canadian currency)/year in Ontario. Studies are needed to show how the use of bisphosphonates in combination with other therapeutic modalities can be optimized to improve the overall cost profile of this family of drugs.

Radiation Radiation therapy is highly cost-effective when compared with analgesic therapy. The cost of radiation therapy was evaluated in 66 patients who received palliative radiation totaling 30 Gy in 10 fractions (72%) or 20 Gy in 5 fractions (28%).61 Pain scores decreased about 4 points (0 to 10 scale) after radiation therapy, from a median level of 5.58 points before treatment to 1.55 points after treatment.61 Incident-related pain with movement decreased 5 points, from 7.32 points before treatment to 1.94 points after treatment. The estimated cost of radiation therapy ranged from $1200 to $2500/patient. With durable pain relief for 9 months, the estimated cost of analgesics over the same period would have totaled $9000 to $36,000. In this study, a brief course of radiation therapy proved to significantly reduce pain and was cost-effective compared with the continued need for analgesics. This is a particularly important consideration for the treatment of hospice patients.

Chemotherapy versus Bisphosphonates Despite cost-benefit concerns, bisphosphonates have been recommended for breast cancer patients who have lytic bone metastases and who are undergoing chemotherapy or hormonal therapy. In a report of more than 700 breast cancer patients randomly assigned to receive chemotherapy alone or with bisphosphonates,62,63 bone metastases were first diagnosed 18 months before study entry, and in about 70% of patients, bone was the only site of metastatic disease. This study demonstrated that the development of skeletal complications was delayed by the bisphosphonates. With chemotherapy alone, skeletal complications occurred at 7.0 months, whereas with systemic chemotherapy plus bisphosphonates skeletal ­ complications

1011

occurred at 12 months. With chemotherapy alone, the proportion of patients experiencing skeletal complications was 56%, whereas with systemic chemotherapy plus bisphosphonates, the proportion was 43%.62,63 With hormonal therapy alone, skeletal complications occurred at 7.0 months, compared with 10 months with hormonal therapy plus bisphosphonates.64 The proportions of ­patients experiencing skeletal complications were reduced by 67% with hormonal therapy alone and by 56% with the combination therapy. In both studies, the occurrence of nonvertebral pathologic fractures, surgery or radiation therapy, bone pain, and hypercalcemia were all substantially reduced with bisphosphonates. However, no survival advantage was observed in the bisphosphonate group; the median survival duration was 37 months. With extended follow-up, 78% of patients in the bisphosphonate and placebo groups died because of progression of breast cancer before completing 24 months of treatment, with a median survival duration of about 18 months.65 In these trials, bisphosphonates were used to increase ­response to systemic therapy, because these drugs delayed the development of skeletal complications and decreased the number of patients referred for palliative radiation therapy or, in some cases, delayed the need for palliative radiation therapy. Radiation therapy in these studies was given only if symptomatic progressive disease appeared while the bisphosphonates and systemic therapy were ­ being administered. Although statistically significant improvements were ­observed, less than one half of the original cohort of breast ­cancer patients completed the first year of the trial and less than one third continued the randomized therapy for 2 years.62-68 Reasons for discontinuing therapy included the occurrence of adverse events, poor response to therapy, or refusal by nearly one half of the patients to continue treatment.65 Almost one half of these patients developed a skeletal event while being given bisphosphonates. Although a benefit from bisphonate use was shown, research is needed to determine the optimal schedule of administration for bisphosphonates and how these drugs can best be combined with other modalities in consideration of the cost-benefit ­issues.69

Remaining Research Questions The need for further research on bisphosphonates is best demonstrated by the intriguing results of a study of bisphosphonates with adjuvant systemic therapy for breast cancer. This prospective randomized trial involved more than 300 patients who had immunohistochemical evidence of at least one tumor cell in a bone marrow aspirate sample.70 More than 80% of patients had T1 or T2 disease, almost 50% had node-negative disease, and more than 70% had estrogen receptor-positive tumors; 81% of patients were given adjuvant systemic chemotherapy, hormonal therapy, or a combination. Substantial decreases in the occurrence of distant metastases (13% versus 29%), bone metastases (8% versus 17%), and visceral metastases (8% versus 19%) were observed among patients who received adjuvant bisphosphonate therapy, resulting in a substantial increase in survival rate (96% versus 85%). The effects of bisphosphonates on apoptosis, adhesion molecules, and proteases are thought to affect the ­development of visceral metastases. When combined with ­cytotoxic agents that suppress cell proliferation, however,

1012

PART 14  Special Considerations

bisphosphonates interfere with tumor cell adhesion and ­invasion.70 These characteristics raise many questions. It is not known how long bisphosphonates should be infused or whether another type of administration, such as pulsed therapy, would be more effective. Another consideration is whether the schedule of administration should be different for adjuvant or palliative treatment. It is not known whether other bone-targeting therapies, such as radiopharmaceuticals, or more cost-effective approaches would improve response over that produced by an adjuvant or palliative combination of chemotherapy plus hormonal therapy.

Evaluating Outcomes of Palliative Therapy The findings of the studies previously discussed demonstrate two key points regarding evaluation of the outcomes of palliative therapy. First, palliative therapy for symptomatic metastases often does not provide a significant survival benefit, even if the same approach improves survival when given in the adjuvant setting. Physicians must not confuse what can be accomplished with adjuvant therapy with what can be accomplished with palliative therapy. The goal of palliative therapy, whether administered in an oncologic or hospice practice, is the prompt and cost-effective relief of symptoms with little treatment-related morbidity. In most cases, radiation therapy can be given efficiently, with few side effects, and at a relatively low cost compared with systemic therapies. Second, almost one third of patients develop distant metastases despite receiving adjuvant systemic chemotherapy plus hormonal therapy. In one study of patients with T1 breast cancer, the frequencies of first distant metastases were 1% to 2% per month for those with node-negative disease, 2% to 4% per month for those with one to three positive nodes, and 3% to 6% per month for those with more than three positive nodes.71 From this finding and from the assumption that adjuvant therapy reduces the mortality rate by 30%, patients with T1 disease would potentially have a higher rate of distant metastases if they underwent neoadjuvant chemotherapy (1% to 4% for each month that surgery is delayed).71 Novel approaches are needed to increase the effectiveness of adjuvant therapies to prevent disease ­recurrence. Clearly defined approaches are needed to treat advanced and metastatic cancer as a chronic disease.

Symptom Relief in Locally Advanced or Metastatic Disease Most patients with cancer have symptoms, and those symptoms must be controlled regardless of whether treatment is administered with curative or palliative intent. Radiation oncologists need to be knowledgeable about the management of cancer-related pain, because controlling pain ­facilitates the ability to create radiation-field simulations and administer palliative therapy. The Joint Commission on Accreditation of Healthcare Organizations issued a new set of standards for the assessment and management of pain that went into effect on January 1, 2001. These standards state that all patients have the right to appropriate assessment and management of pain and that patients should be taught that pain management is a part of treatment.

The results of a survey of Radiation Therapy Oncology Group (RTOG) physicians showed that 83% believed that most cancer patients with pain were undermedicated.46 In their practice, 40% of the physicians surveyed ­reported that pain management was of poor or fair quality, and 23% said they would not use opioid analgesics until the patient’s estimated survival duration was 6 months or less. Seventy-seven percent of physicians cited poor pain ­assessment as a barrier to adequate pain management, 60% cited the reluctance of patients to report pain, 72% cited the reluctance of patients to take analgesics, and 40% cited the reluctance of the physicians to prescribe opioids. For 72% of the RTOG physicians, the prescription of opioid analgesics was prompted by the failure to respond to palliative radiation therapy.46 These results proved to be not substantially different from the responses to the same survey among Eastern Cooperative Oncology Group physicians 7 years earlier,25 before publication of the Agency for Health Care Policy and Research’s Cancer Pain Management Guidelines.2,13 One of the barriers to effective pain management as ­observed by physicians is a lack of knowledge about assessing symptoms and determining treatment.1,25 Patients must be encouraged to report symptoms such as pain. Objective signs of acute pain, such as tachycardia, sweating, and ­facial grimacing, often are absent in chronic pain syndromes. The patients should be reassured that their pain can be controlled and that pain management is an integral part of cancer therapy. Patients who become fatigued because of pain often cannot tolerate the side effects of antineoplastic therapy. A clear history that identifies factors that could be related to the pain and a careful physical and neurologic examination should be performed to help direct diagnostic studies. Because more than 70% of the pain that is experienced by cancer patients is related to the disease, diagnostic evaluations should be performed frequently.

Pain Categories Cancer pain can be divided into two broad categories: ­somatic or visceral pain and neuropathic pain. These categories of pain have different causes, symptoms, and ­responses to analgesics and palliative radiation. Somatic or visceral pain arises from direct stimulation of afferent nerves caused by tumor infiltration of skin, soft tissue, or viscera.2,72,73 Somatic pain is often described as well-localized, dull or aching discomfort. Visceral pain tends to be poorly localized and is often referred to a dermatomic area remote from the source of pain. Common causes of somatic pain include bone metastases; liver metastases that result in capsular distention and biliary, bowel, or ureteral obstruction; and distention of normal skin or soft tissue structures from an expanding tumor mass (Box 39-1).74 Somatic or visceral pain usually can be controlled with palliative antineoplastic treatment and conventional analgesics. Neuropathic pain develops after injury to or chronic compression of nerves. Common causes of neuropathic pain include tumor invasion of a nerve plexus, spinal nerve root compression, herpes zoster infection, or surgical ­interruption of intercostal nerves (by thoracotomy or mastectomy). Neuropathic pain is described as sharp, burning, or shooting and often is associated with paresthesias and dysesthesias. Response to conventional analgesics is usually poor, but antidepressant or anticonvulsant ­medications

39  Palliative Care

1013

BOX 39-1.  Specific Pain Syndromes in Patients with Cancer Pain Syndromes Associated with Direct Tumor ­Infiltration Tumor infiltration of bone Base of skull syndromes, with metastases to Jugular foramen Clivus Sphenoid sinus Vertebral body syndromes, with metastases to C2 C7, T1 L1 Sacral syndrome Tumor infiltration of nerve Peripheral neuropathy Plexopathy Brachial Lumbar Sacral Nerve root Leptomeningeal metastases Spinal cord Epidural spinal cord compression Intramedullary metastasis Pain Syndromes Associated with Cancer Therapy Postoperative syndromes Postthoracotomy syndrome: intercostal nerve injury

Postmastectomy syndrome: intercostal brachial radial nerve injury Postradical neck dissection syndrome: cervical plexus injury Phantom limb syndrome Postchemotherapy syndromes Peripheral neuropathy Aseptic necrosis of the femoral head Steroid pseudorheumatism Posttherapeutic neuralgia Postirradiation syndromes Radiation fibrosis of brachial and lumbar plexus Radiation myelopathy Radiation-induced second primary tumors Radiation necrosis of bone Pain Syndromes Not Associated with Cancer or Cancer Therapy Cervical and lumbar osteoarthritis Thoracic and abdominal aneurysms Diabetic neuropathy Myofascial pain syndromes Migraine and tension headaches

From Foley KM, Arbelt E. Management of cancer pain. In DeVita VT, Hellman S, Rosenberg SA (eds): Cancer: Principles and Practice of ­Oncology, 3rd ed. Philadelphia: JB Lippincott, 1989, pp 2:2064-2087.

may be helpful. Radiation therapy is one of the most effective—and often the only—therapeutic option for relieving pain caused by nerve compression or infiltration by a malignant tumor. Examples of neuropathic pain syndromes commonly encountered in radiation therapy include superior sulcus (Pancoast) tumor, otalgia from head and neck cancer, brachial plexopathy in breast cancer, and epidural metastases. Among patients predicted to have an extended survival time, neuropathic pain often can be best relieved by a high-dose, extended course of radiation therapy in an attempt to secure durable pain relief.

Pharmacologic Principles Tolerance and Addiction Knowledge concerning appropriate use of analgesics is ­essential for any physician caring for cancer patients. The choice of an analgesic drug must be individualized. Too often, this choice is influenced by unfounded fears of the patient and physician about tolerance, addiction, and respiratory depression rather than by the level of pain ­experienced.2,3,24,75-78 Tolerance is defined as the need to increase the dose of a drug to achieve the same analgesic effect. Tolerance is a normal physiologic response to ­opioids and is characterized by a decrease in the duration of the analgesic effect. The dose may be increased, occasionally to enormous levels, because there is no upper limit on the prescribed dose for strong opioids such as morphine. If tolerance develops, an alternative opioid may be used; adjuvant analgesics, such nonsteroidal anti-inflammatory drugs, may be given; or these interventions may be combined. In contrast to tolerance, physical dependence is a normal physiologic response that occurs in response to the

a­ dministration of steroids, some antihypertensive agents, and analgesics. Physical dependence becomes obvious when opioids are discontinued, and it is manifested through withdrawal symptoms, such as diaphoresis, nausea, and vomiting. Analgesics must be discontinued in the same way as steroids are, by slow tapering of the dose. ­Withdrawal symptoms can be prevented or treated by administering a dose of analgesic of about 25% of the previous daily dose. The onset of withdrawal symptoms is related to the half-life of the drug being used; symptoms of withdrawal may begin 6 to 12 hours after morphine is discontinued, whereas these symptoms may be delayed for several days if methadone is prescribed. Psychological dependence (i.e., addiction) is an ­abnormal behavioral condition characterized by an overwhelming involvement in the acquisition and use of a drug for reasons other than pain relief.2,72,73 Addiction is rare among patients who receive opioids for control of cancer pain. Tolerance and physical dependence are predictable pharmacologic effects of chronic narcotic administration, and they are distinct from addiction. Patients and their families should be counseled about the rarity of addiction when opioids are prescribed for legitimate medical indications. Opioid pseudoaddiction describes behavior that mimics psychological dependence that occurs as a consequence of inadequate treatment of cancer pain. Pseudoaddiction may result from inadequate analgesic dosing schedules (“as needed” prescriptions) during periods of continuous pain, the use of dosing intervals that are greater than the ­duration of action of the prescribed analgesic, or the use of prescribed analgesics that are too weak to address the level of pain.76 Unlike true addiction, pseudoaddictive behavior resolves after the pain is adequately managed.

1014

PART 14  Special Considerations

Fear of addiction limits the use of narcotic analgesics in clinical practice. The unsubstantiated concerns of the past regarding physician prescription records and regulatory ­issues present an additional barrier to effective pain management. Federal and state laws affirm the essential ­medical value of controlled substances for cancer patients with intractable pain. Patients and physicians should be reassured that the rates of addiction and abuse have not increased even though the medical use of opioid analgesics has increased since the publication of pain guidelines. From 1990 to 1996, the medical use of morphine increased by 59%, fentanyl by 1168%, oxycodone by 23%, and hydromorphone by 19%. During the same time, the use of meperidine decreased by 35%.79 Although the incidence of drug abuse increased 6.6%, the proportion of abuse cases related to opioids decreased from 5.1% to 3.8%. Reports of abuse of meperidine decreased by 39%, abuse of oxycodone by 29%, abuse of fentanyl by 59%, and abuse of hydromorphone by 15%; abuse associated with morphine increased by 3%. The trend of increasing medical use of opioid analgesics to treat pain therefore does not seem to contribute to the abuse of opioid analgesics.

Approaches for Prescribing Analgesics The World Health Organization has advocated the use of a three-step approach in prescribing analgesics, and this approach is discussed at length in Chapters 24 and 25. The approach is indexed to the patient’s self-report of the intensity of the pain as mild, moderate, or severe.80 The schedule of administration and the type of analgesic are then based on the frequency and severity of the pain. If the pain is constant, a long-acting analgesic should be prescribed. Long-acting analgesics can be administered orally every 8 or every 12 hours, depending on the pain cycle. Transdermal fentanyl, which is administered as a patch, is an alternative means of continuous analgesic administration; patches can be replaced every 48 to 72 hours. A short-acting opioid also should be prescribed as supplemental treatment for breakthrough pain until the next dose of a long-acting analgesic is given. Breakthrough pain is an incident-related symptom event; for example, it can be brought on by walking or undergoing a diagnostic procedure or by an inadequate dose or schedule of administration of the long-acting analgesic.73,81 Short-acting analgesics also can be used alone for intermittent pain. Moderate to severe pain can be treated with opioid analgesics around the clock, used alone or in combination with a nonsteroidal anti-inflammatory agent (Table 39-1).3 In accordance with the course of the disease and response to treatment, analgesic needs may increase or decrease for any individual patient. Patients and families should be ­ given information about how to titrate analgesics so that the patient’s pain can be controlled around the clock and the patient can enjoy normal activities and sleep patterns. Analgesics that are complexed with acetaminophen or aspirin should be avoided because of possible renal and ­hepatic toxic effects. Nonsteroidal anti-inflammatory agents also have serious potential side effects, including gastrointestinal bleeding, renal toxicity, and inhibition of platelets. Although these agents can be effective as co-analgesics, the possibility of toxic reactions persists. Analgesics should be given orally whenever possible, because intramuscular

injections can cause discomfort, infection, or bruising at the injection site as well as irregular drug absorption. ­Analgesics should be administered around the clock rather than as needed to avoid cycles of pain. Higher doses of analgesics are needed to treat pain as opposed to preventing pain. Analgesic side effects, such as constipation, should be anticipated and treated. More than 90% of cancer-related pain can be controlled. A variety of alternative long- and short-acting opioid analgesics and routes of administration have become available. When oral administration is not possible, rectal suppositories can be used as an alternative or as a breakthrough analgesic.81 A list of oral and rectal opioid analgesics and their characteristics is provided in Table 39-2.81 Giving opioids in oral form also has the advantage of their being ­directly absorbed into the bloodstream from the oral ­ mucosa, thereby avoiding the first-pass effect (in which drug metabolism begins as the drug passes through the gut ­mucosa and liver). If the oral mucosa is intact but the ­patient cannot swallow, morphine can be delivered sublingually, at an absorption rate of about 22%. Another ­product, a lozenge that contains fentanyl, provides pain relief within minutes of administration and lasts for 1 to 2 hours. Other means, such as a nerve block or an epidural catheter that administers opioids directly to the spine, should be considered if the pain or analgesic side effects are poorly controlled.82-84 A common example is a celiac axis block for pain resulting from pancreatic cancer.

Clinical Considerations Choice of Palliative Modalities Curative therapy attempts to render the patient free of disease. This approach can be applied in the treatment of primary or metastatic disease. The prognosis for a patient who is rendered disease-free after resection of a single ­liver metastasis from rectal cancer is better than that for a patient with extensive, inoperable lung cancer who does not have metastatic disease. Unlike curative therapy, in which a high total dose of radiation is administered to eradicate a tumor, palliation radiation therapy usually involves use of a hypofractionated schedule. Typical hypofractionated radiation schedules for palliative therapy are 2.5-Gy fractions given over a 3-week period for a total dose of 35 Gy; 20 Gy given over a 1-week period; and a single dose of 4 to 8 Gy. In the United States, the most common schedule involves giving 30 Gy in 10 fractions over 2 weeks.12,85-93 The dose fractionation schedule used depends on prognosis, volume, and irradiation target site. Treatment with palliative intent is intended to control the symptoms of disease when the disease cannot be eradicated. Effective antineoplastic therapy provides the greatest palliative benefit fo locally advanced or metastatic cancer in any organ. Radiation can be used alone or in combination with other antineoplastic therapies. The symptoms most commonly relieved by radiation therapy are pain, bleeding, and obstruction. The palliative interventions recommended depend on the patient’s clinical status, burden of disease, and the site of symptoms. For locally advanced or metastatic disease, these factors are indexed to the relative effectiveness, durability, and morbidity of each palliative intervention. Prognosis represents the single most ­important factor in choosing a palliative therapy.

39  Palliative Care

1015

TABLE 39-1.  Classes of Analgesic Agents Used to Treat Cancer Pain Drug Classes and Names

Usual Doses for Adults

Group I: Nonnarcotic Analgesics (Nonsteroidal Anti-inflammatory Drugs) Commonly Used Drugs Acetylsalicylic acid (aspirin)

500-1000 mg q4-6h PO

Acetaminophen

500-1000 mg q4-6h PO

Indomethacin

25-50 mg tid PO

Naprosyn

500 mg; then 250 mg q6-8h PO

Ibuprofen (Motrin, others)

200-400 mg q4-6h PO

Ketoprofen (Orudis)

25-50 mg q4-6h PO

Ketorolac (Toradol)

30-60 mg

Group II: Narcotic Agonists and Antagonists Agonists Morphine and congeners Morphine*

15-30 mg q3-4h PO

Hydromorphone

4-8 mg q3-4h PO

Codeine

30-60 mg q3-4h PO

Oxycodone*

15-30 mg q3-4h PO

Levorphanol

2-3 mg q3-4h PO

Meperidine and congeners Meperidine (Demerol)

Not recommended

Methadone and congeners Propoxyphene (Darvon)

65-130 mg q4-6h

Methadone (Dolophine)

5-10 mg q4-8h

Transdermal fentanyl (Duragesic)

25-100 μg/h q3d

Partial Agonists and Agonist-Antagonists Pentazocine

None is recommended for chronic dosing in cancer patients.

Buprenorphine Nalbuphine Butorphanol Group III: Adjuvant Analgesic Drugs Anticonvulsants Phenytoin (Dilantin)

100-300 mg tid PO

Carbamazepine

200-600 mg qid PO (Tegretol)

Phenothiazines Methotrimeprazine (Levoprome)

15 mg q4-6h IM

Tricyclic Antidepressants Amitriptyline

10-150 mg/day PO

Desipramine

25-150 mg/day PO

Nortriptyline

25-150 mg/day PO

Steroids Dexamethasone

4 mg qid PO

Antihistamines Hydroxyzine (Vistaril)

25 mg q4-6h PO (Continued)

1016

PART 14  Special Considerations

TABLE 39-1.  Classes of Analgesic Agents Used to Treat Cancer Pain—Cont’d Drug Classes and Names

Usual Doses for Adults

Stimulants Amphetamine

2.5-10 mg PO

Methylphenidate (Ritalin)

5-10 mg PO

Caffeine

100-200 mg/day

Mexiletine

150 mg bid/tid PO

*Morphine/oxycodone can be given as an immediate release formulation, or as a sustained release preparation. Modified from Jacox A, Carr DB, Payne R. New clinical practice guidelines for the management of pain in patients with cancer. N Engl J Med 1994;330:651-655.

TABLE 39-2.  Dosages and Characteristics of Rectal and Oral Opioids Opioid

Dosage Form

Time to Peak Level

Duration of Action

Morphine

Oral tablets

1.1 hr

<6 hr

Morphine

Controlled release

5.4 hr

8-12 hr

Morphine

Rectal solution

5 hr

6 hr

Morphine

Rectal suppositories

1.1 hr

<6 hr

Hydromorphone

Rectal suppositories

1 hr

4-6 hr

From Lipman AG. New and alternative noninvasive opioid dosage forms and routes of administration. In Berger A, Portenoy RK, Weissman DE (eds): Principles and Practice of Supportive Oncology, Update. Philadelphia: Lippincott Williams & Wilkins, 2000, pp 1-8.

Several clinical, prognostic, and therapeutic factors must be considered in developing a treatment regimen for a course of palliative radiation therapy. Clinical status is accounted for in the treatment setup and in the number of treatments prescribed. The radiation dose-fractionation schedule and technique also should take into account the integration of other therapies. Many radiation therapy options exist for tumors that cause localized symptoms. Brachytherapy and conformal and intensity-modulated radiation therapy can be used to better localize radiation dose and reduce side effects, especially in a previously ­irradiated area.53,58,86,93-95

Reirradiation Issues regarding reirradiation are especially important in palliative therapy. Experimental data suggest that acutely responding tissues recover from radiation injury in a few months and can tolerate another full course of radiation therapy. However, the variability in recovery from radiation among late-reacting tissues is considerable.96 The heart, bladder, and kidney do not seem to recover from any late radiation effects, but the skin, mucosa, lung, and spinal cord do recover from subclinical radiation injury. This recovery depends on the organ irradiated, the size of the initial dose of radiation, and the interval between doses of radiation. Clinical correlates exist, and limited toxic reactions have been reported with reirradiation.97

with palliative intent depends more on prognosis than on the histology of the primary tumor. Common sites palliated by radiation alone or in combination with other treatments include tumor involvement of the lung, pelvis, skin and subcutaneous tissues, brain, and bone. Symptoms resulting from cancer in these sites include pain, bleeding, visceral obstruction, and lymphovascular obstruction.

Normal Tissue Tolerance The limited radiation tolerance of normal tissues, such as the spinal cord, that are adjacent to a bone metastasis makes it impossible to administer a radiation dose that is large enough to eradicate a measurable volume of tumor. Palliative radiation should produce sufficient tumor regression from critical structures to relieve symptoms for the duration of the patient’s life. Symptoms that recur after palliative radiation most commonly result from localized regrowth of tumor. Control of cancer-related pain with the use of analgesics is imperative to allow comfort during and while awaiting response to therapeutic interventions. Pain represents a sensitive measure of disease activity. Close ­follow-up should be performed after any palliative therapeutic intervention to ensure control of cancer and treatment-related pain and to initiate diagnostic studies to identify progressive or recurrent disease.

Symptomatic Site Location

Bone Metastases

In choosing palliative care, the location of the symptomatic site is more important than the state or type of tumor at that site. In general, radiation portals are used to palliate bronchial obstruction, whether caused by a primary lung cancer, recurrent breast cancer, or metastatic melanoma. The number of radiation fractions prescribed for treatment

Although metastasis can occur at any site, in more than one half of cases, palliative therapy is prescribed to relieve symptoms of bone metastases.1-4 In 1993, about 800,000 patients in the United States were estimated to have metastatic disease, and metastases to bone were the most common.92 Metastatic involvement of a solitary bony site

39  Palliative Care

or of only the bones is uncommon, occurring in less than 10% of patients. Among hospitalized patients, severe pain was experienced by more than 50% of patients with bone metastases.4 One of the most important goals in the treatment of bone metastases is to relieve suffering and return the patient to independent function.52,93 The location of the metastasis, however, can influence the degree of pain relief that can be achieved by palliative interventions. Metastases to weight-bearing bones and bones involved in ambulation and activities of daily living are less likely to respond completely to palliative interventions. Pain relief can be achieved with radiation therapy in 73% of patients with spine metastases, 88% with limb lesions, 67% with pelvic metastases, and 75% with metastases to other parts of the skeleton.14 Bone metastases from prostate or breast cancer involve the axial skeleton in more than 80% of cases because of the predilection of these tumors to involve the red marrow. Metastatic invasion of the bone cortex rarely happens without red marrow involvement.98-109 For this reason, the spine, pelvis, and ribs usually are involved before metastases become evident in the skull, femora, humeri, scapulae, and sternum. The mechanisms involved in the metastatic spread of cancer to the bone are complex. The predilection of specific types of tumors to metastasize to bone is not well understood. Bone is unique because it is continuously remodeled by local and systemic growth factors. The process involved in bone remodeling includes an increase in osteoclast activity followed by the appearance and proliferation of osteoblast precursors, which lay down new bone as they mature.­102-109 Tumors can release osteoclastic stimulatory factors ­ locally or systemically. These factors include the parathyroid hormone–related protein, transforming growth ­factor α, transforming growth factor β, prostaglandins, and ­ cytokines.110 Tumors also can indirectly stimulate bone loss through the activation of immune mechanisms that release cytokines, including tumor necrosis factor and interleukin-1. Metastases to bone result from a synergistic relationship between the cancer cell and the bone that causes increased bone resorption by means of the release of mediators from the cancer cells or leukocytes. Factors released during bone resorption promote growth of the cancer through activation of proliferation-associated oncogenes. The relationship between bone invasion and bone pain is unknown. Pain may result from the stimulation of nerve endings by the release of factors such as prostaglandins, bradykinin, histamine, or substance P in areas of bone involvement.107 A mass effect from stretching of the periosteum, pathologic fracture, or direct tumor growth into adjacent nerves and tissues may explain the appearance of pain from larger metastatic deposits. Multidisciplinary evaluation of patients with metastatic disease to the bone allows comprehensive management of the associated symptoms, determination of the risk of pathologic fracture, and coordination of the administration of a wide range of available antineoplastic therapies.12,108,109 Bone scans are the most sensitive and specific method of detecting bone metastases, but magnetic resonance imaging (MRI) is the best available technique for evaluating neoplastic invasion of the bone marrow, vertebrae, central nervous system, and peripheral nerves.43,98-100,110 Metastases to bone and other areas are rarely undetected when radiographic diagnosis is pursued. When radiographic

1017

confirmation of malignancy is equivocal, bone biopsy should be considered.101-103

Symptom Management Bone metastases can be treated with localized or systemic therapies or a combination of both.93,86,102,107,111 Because localized radiation treats only a localized symptomatic site of disease, often it is used in coordination with systemic therapies such as chemotherapy, hormonal therapy, and bisphosphonates. Radiopharmaceuticals are another systemic option for treating diffuse symptomatic bone ­metastases. Because the radiation is deposited directly at the involved area in the bone, radiopharmaceuticals such as 89Sr also can be used to treat bone metastases when symptoms recur in a previously irradiated site.54-56,112-123 Radiopharmaceuticals also can be used as an adjuvant to localized EBRT and can reduce the development of other symptomatic sites of disease.

Radiation Schedules: The Radiation Therapy ­Oncology Group Trial Experience with palliative radiation for bone metastases comes mainly from a prospective clinical trial conducted by the RTOG to compare a variety of treatment schedules.124,125 To account for prognosis, patients were stratified on the basis of whether they had solitary or multiple sites of bone metastasis. Initial analysis of the results led to the conclusion that low-dose, short-course treatment schedules were as effective as high-dose, protracted treatment programs.124 For patients with solitary bone metastases, no difference in pain relief was reported after a dose of 20 Gy given in 4-Gy fractions than after a dose of 40.5 Gy given in 2.7-Gy fractions (Table 39-3).124 Pain relapsed in 57% of patients at a median of 15 weeks after completion of therapy at each dose level. For patients with multiple bone metastases, the following dose schedules were compared: 30 Gy at 3 Gy/fraction, 15 Gy at 3 Gy/fraction, 20 Gy at 4 Gy/ fraction, and 25 Gy at 5 Gy/fraction. No difference was identified in the rates of pain relief for these treatment schedules. Partial pain relief was achieved in 83% of the ­patients, and complete relief was achieved in 53% of ­patients. More than 50% of these patients developed recurrent pain; the fracture rate was 8%; and the median ­duration of pain control was 12 weeks for all radiation schedules used to treat multiple bone metastases. Prognostic factors for ­response included the initial pain score and site of the primary tumor. Reanalysis of Results. A subsequent reanalysis of the results from this RTOG study included use of a different definition of complete pain relief (absence of pain and cessation of the use of narcotics).125 When this definition was used, pain relief was found to be related to the number of fractions and the total dose of radiation.125 Complete pain relief was achieved in 55% of patients with solitary bone metastases who received 40.5 Gy at 2.7 Gy/fraction and in 37% of patients who received a total dose of 20 Gy at 4 Gy/fraction (Table 39-4).125 A similar relationship was observed on reanalysis for patients with multiple bone ­metastases. Complete pain relief was achieved in 46% of patients who received 30 Gy at 3 Gy/fraction and in 28% of patients treated to 25 Gy in 5-Gy fractions. In most ­cases, the interval to response was 4 weeks for complete and minimal relief of symptoms.

1018

PART 14  Special Considerations

TABLE 39-3.  Percentage of Patients Responding to Radiation Patient Group and Total Dose (Gy)

Number of Weeks after Radiation Therapy Dose/Fraction (Gy)

Tumor Dose at 2 Gy/Fraction

<2

2-4

4-12

12-20

Solitary Metastases 40.5

2.7

42.9

7%

29%

53%

77%

20.0

4

23.3

16%

50%

66%

82%

30.0

3

32.5

19%

48%

73%

84%

15.0

3

16.2

34%

70%

84%

93%

20.0

4

23.3

28%

53%

75%

88%

25.0

5

31.25

22%

41%

72%

80%

Multiple Metastases

Data from Tong D, Gillick L, Hendrickson FR. The palliation of symptomatic osseous metastases: final results of the study by the Radiation Therapy Oncology Group. Cancer 1982;50:893-899.

TABLE 39-4.  D  ose Response Evaluation from Reanalysis of the Radiation Therapy Oncology Group Bone Metastases Protocol Patient Group and Dose/Fraction (Gy)

Total Dose (Gy)

Tumor Dose at 2 Gy/Fraction

Complete Response Rate*

P Value <0.0003

Solitary Bone Metastases 40.5

2.7

42.9

55%

20.0

4.0

23.3

37%

Multiple Bone Metastases 30

3.0

32.5

46%

15.0

3.0

16.2

36%

20.0

4.0

23.3

40%

25.0

5.0

31.25

28%

<0.0003

*Excludes

the use of analgesics and accounts for retreatment. Data from Blitzer PH. Reanalysis of the RTOG study of the palliation of symptomatic osseous metastasis. Cancer 1985;55:1468-1472.

Three important issues can be identified from this experience. First, the results of the reanalysis demonstrated the importance of defining response to therapy. Second, the revised definitions of response showed that the total radiation dose did influence the degree to which pain was relieved (see Table 39-4); higher radiation doses resulted in higher rates of pain relief. These results also correspond well to those of another study of patients with breast, prostate, or lung cancer in which patients treated with total doses of 40 Gy or more had higher rates of complete pain relief (63% to 71%) than did those treated with less than 40 Gy (44% to 62%).14 Third, the amount of time needed to experience pain relief after radiation therapy was fairly long (see Table 39-3); only one half of the patients who eventually responded showed relief of symptoms at 2 to 4 weeks after radiation therapy,124,125 underscoring the need for continued analgesic support after completion of radiation therapy. Accomplishment of maximum relief consistently took 12 to 20 weeks after radiation therapy, a period that may reflect the time needed for ­reossification.

Radiographic evidence of recalcification is observed in about one fourth of such cases, and in 70%, recalcification is evident within 6 months of completing radiation and other palliative therapies.106-109 Determining the time to response and defining the parameters of response are critical to an evaluation of outcome. The total dose of radiation also may be a significant factor for complete pain relief. Similar issues were observed when pretreatment pain characteristics were used to determine the efficacy of palliative radiation.39,72,73,126 Among patients who had a life expectancy of only a few months, less than 25% had a complete response to palliative radiation. A complete response was defined as becoming pain-free within 28 days after the start of treatment, with “stable” use of analgesics. A partial response was achieved in 35% of patients, and 42% had no response to palliative radiation. Pretreatment evidence of neuropathic pain was the only significant prognostic or clinical variable that predicted failure to respond to palliative radiation, and severe neurogenic symptoms were ­present in 33% of patients treated.39,72,73,126

39  Palliative Care

Other Schedules A variety of radiation therapy schedules have been evaluated in palliative therapy for bone metastases. One trial compared 30 Gy given in 10 fractions with 15 Gy given in 3 fractions at 2 fractions/week to 217 patients with breast cancer.127 No difference was observed in pain scores, analgesic use, treatment side effects, or survival. At 12 months of follow-up, about 70% of patients had little or no limitation on activity.127 Consistent with other reports, the survival rates were 45% at 3 months, 33% at 6 months, and only 20% at 1 year.

Response to Treatment Response to treatment may be the most important prognostic factor in bone metastases. A prospective randomized trial compared 30 Gy given in 2-Gy fractions with 20 Gy given in 4-Gy fractions to 100 patients with a ­ median ­follow-up time of 12 months.128 The primary tumor was located in the breast in 43% of patients, lung in 24%, and prostate in 14%. The pelvis (31%), vertebrae (30%), and ribs (20%) were the most common sites treated. These characteristics were equally distributed between the study groups. The biologically effective doses were 28 Gy10 (Gy10 ­indicates an alpha/beta ratio of 10) and 36 Gy10 for the two treatment groups. No significant differences were observed between the groups in frequency of pain relief, duration of pain relief, mobility, or pathologic fractures.128 The mean interval from the completion of radiation to pain relief was approximately 10 days for both groups. The mean duration of pain relief was 245 days in each group, and it lasted for up to 84% of the remaining survival time. Importantly, responders had a highly significant (P < .0001) improvement in the duration of relief. In addition to the radiation schedule, other factors that influenced overall survival rate were primary tumor site (P < .0001), performance status (P < .0001), age (P < .03), and biologically effective dose (P < .04). The importance of the relationship between pain control and survival was observed unexpectedly in a trial comparing best medical management of pain with placement of an intrathecal catheter for pain control in patients with locally advanced and recurrent cancer. Most of the patients in this trial, preliminary results of which were presented at the 2001 meeting of the American Society of Clinical Oncology, had lung or ovarian cancer. Improvements in pain control were associated with a significant survival benefit among patients with the intrathecal catheters. It is possible that survival may be linked to improved mobility afforded by pain control. It is not known whether pain, through its effects on sleep and other functions, also affects cytokine release and other immunologic variables that may ultimately affect survival. This is an area of active research. Whether a dose-response relationship exists in the treatment of bone metastases is unknown. The published results do not control for histology, type (blastic or lytic), or location of the metastases or type of pain (neuropathic, mechanical, or somatic), and many of the older reports do not include validated pain surveys or account for use of analgesics. The definition and duration of response is variable.85,129,130 It is often unclear how many patients who had been assigned to a high-dose group completed therapy

1019

and how many who had been assigned to a low-dose group received a high dose of radiation as a result of retreatment.85 Three reports suggest the existence of a dose-response relationship in the treatment of bone ­metastases.125,131,132 The RTOG study124,125 demonstrated that the level of pain correlated with prognosis for patients with multiple bone metastases. This survival difference may be an important observation, because unrelieved pain and the resultant sequelae of immobility can contribute to mortality and morbidity.18,19,34,35,37,38,40,126 Larger data sets that include prognostic factors in multivariate analysis are required to determine the effect of pain and the response to therapy on quality of life and survival. Further research may determine that the radiation dose schedules should be specific to the primary tumor, the site of metastases, and the type of pain if response is to be maximized. Regardless of the radiation therapy schedule used, retreatment should be considered at the first sign of symptom progression to achieve the maximum and most durable level of palliation. Pain management should be ­optimized in light of reports of its relationship to survival. Future research should target level of pain relief and the cost-effectiveness of treatment.

Single-Fraction versus Multiple-Fraction Radiation Therapy Although equal rates of response have been observed for single and for several radiation fractions, future analyses should evaluate survival and durability of response. Criteria need to be established to determine which clinical presentation is best treated with single-fraction therapy compared with multiple-fraction therapy.130,133 Most importantly, these analyses should assess the efficacy and costs of treating symptoms that recur in an irradiated field. Single-fraction radiation is used extensively in Europe and is the subject of an RTOG trial. Because it is convenient, cost-effective, and can be repeated, single-fraction radiation is of special interest in the palliative-care setting. In the United States, however, the efficacy of this treatment continues to raise concern. The cost and convenience benefits from a shorter course of radiation may be lost through the continued need for analgesics. A cost-effectiveness comparison was conducted among 1171 patients who were randomly assigned to receive ­radiation at 8 Gy in 1 fraction or 24 Gy in 6 fractions.134 The two treatment groups were equivalent in terms of palliation, including use of analgesics; however, 25% of the 8-Gy group required retreatment compared with 7% of the 24-Gy multifraction group. A cost analysis showed that treatment in the 8-Gy group cost $1734 (European currency) and $2305 (European currency) for those in the multifraction group, representing a 25% cost difference. However, when the costs of retreatment were included in the analysis, the cost difference was reduced to only 8%, although an economic advantage continued to exist because of the savings in radiation therapy capacity. Many issues must be evaluated before single-fraction radiation can be advocated for the treatment of all bone metastases; close follow-up is needed to determine which patients treated in this way will require retreatment.130,133 Discussions regarding radiation dose fractionation must further be considered in the context of multidisciplinary treatment.

1020

PART 14  Special Considerations

A single large radiation fraction is reportedly as effective in relieving pain as is radiation delivered in several smaller doses. Radiobiologically, a single 8-Gy fraction produces the same side effects in late-reacting tissues as those produced by 18 Gy given in nine 2-Gy fractions.135 The radiobiologically equivalent dose to the tumor of a single 8-Gy fraction would be 12 Gy if 2-Gy fractions were used. The most common dose fractionation schedule used for palliative radiation in the United States is 30 Gy given in 10 fractions. Radiobiologically, this is equivalent to 36 Gy at 2 Gy/fraction for late-reacting tissues and 32.5 Gy to the tumor. When single-dose and multiple-dose schedules from several studies were compared, no difference was ­reported in how quickly symptoms resolved or the duration of pain relief.12,89-92,133,136-142 In each case, symptom relief ­lasted 3 months in 70% of patients, 6 months in 37%, and 12 months in 20%. As was true in the RTOG study, about 69% of patients responded at 4 weeks, and response rates reached a plateau in 80% of patients at 8 weeks.124,125,136-142 Complete response rates after a single 8-Gy fraction ­totaled 15% at 2 weeks, 23% at 4 weeks, 28% at 8 weeks, and 39% at 12 weeks after irradiation. The Bone Pain Trial Working Party138 conducted a trial among 765 patients with skeletal metastases, comparing a single 8-Gy fraction to 20 Gy given in 5 fractions or 30 Gy given in 10 fractions. Pain severity and analgesic requirements were recorded before treatment, 2 weeks after treatment, and at 1, 2, 3, 4, 5, 6, 8, 10, and 12 months of follow-up. Pain relief was the primary end point. In all three groups, pain management with analgesics before radiation therapy had been inadequate. Consistent with other findings, 70% of patients in this study had moderate to severe pain when they presented for radiation therapy despite the fact that 70% were taking opioid analgesics. No difference was identified in the time to first improvement in pain, time to complete pain relief, time to pain recurrence, risk of spinal cord compression or pathologic fracture, ­nausea or vomiting, and analgesic requirements between the ­singledose and multiple-dose radiation schedules. Overall, 57% of patients experienced a complete response to radiation. At 6 and 12 months of follow-up, however, about 55% of patients still required opioid analgesics. Retreatment was twice as common in the single-dose 8-Gy group (23% versus 10%), although this difference did not influence overall pain relief. The investigators thought that physicians were more likely to administer retreatment after a single 8-Gy course of radiation therapy than after a multiple-fraction course of therapy. Reirradiation for persistent or recurrent pain is often precluded when higher radiation doses are administered. Because the radiobiologic dose from a single dose is ­relatively low, reirradiation usually is possible.15,136-142 Although reirradiation was necessary for 25% of the patients in the Bone Pain Trial who received a single 8-Gy radiation fraction, all of those patients reportedly responded to the second dose of radiation. When a single 4-Gy fraction was compared with a single 8-Gy fraction, the rate of response was slightly lower and fewer acute radiation-induced reactions were seen in the lower-dose group, but a greater proportion of patients required reirradiation.141 When reirradiation was accounted for, the overall rate of response was equivalent for the 8-Gy and 4-Gy groups.

Reirradiation with a single 4-Gy fraction also yielded a 74% response rate, including a 31% complete response rate.136 Pain relief was achieved in 46% of patients who did not initially respond to a single 4- or 8-Gy fraction. Patients who had a previous complete response to therapy were more likely to respond to reirradiation compared with the group that previously experienced a partial response (85% versus 67%). A shorter radiation therapy schedule, such as the use of single fractions, is advantageous for patients with poor prognostic factors. First, it is easier for patients with poor performance status to complete therapy. Second, response rates are equal for single- and multiple-fraction therapy at 3 months because the median survival duration is less than 6 months in patients with poor prognostic factors.12,87,92,139-143 The option of retreatment after a single fraction of radiation also may be an advantage among patients with good prognostic factors in terms of periodically reducing tumor burden and controlling symptoms in noncritical anatomic sites. Higher radiation doses that provide more durable pain relief are considered to be warranted for patients with good prognostic factors who require treatment over the spine and other critical sites.14,86,93,144-150

Survival Duration The projected survival duration is the critical issue in determining the optimal radiation dose and schedule for palliative radiation. In one study,88 only 12 of 245 patients were alive at the time of analysis, with approximately 50% alive at 6 months, 25% at 1 year, 8% at 2 years, and 3% at 3 years after palliative radiation. For the patients with breast cancer in that study, the corresponding survival rates after palliative radiation were 60% at 6 months, 44% at 1 year, 20% at 2 years, and 7% at 3 years; for the patients with prostate cancer, the survival rates were 60% at 6 months, 24% at 1 year, and 0% at 2 years.88 In the RTOG trial, the median survival duration for patients with solitary bone metastases was 36 weeks; for those with multiple bone metastases, it was 24 weeks.124,125 It may not matter that the response to radiation therapy is optimized through a more protracted course if a doseresponse relationship exists or through retreatment.61,85 Consideration must be given to patient mobility and access, convenience, and cost-effectiveness. A durable and significant response from radiation, not the schedule of administration, is the key issue in palliative radiation.

Morbidity Pathologic fracture and spinal cord compression are significant consequences of bone metastasis. Pain that persists or recurs after palliative radiation should be evaluated to exclude as its cause progression of disease, extension of disease outside the radiation portal (resulting in referred pain), and bone fracture. Reduced cortical strength can cause compression, stress, or microfractures. Plain radiographs have a 91% concordance rate in detecting after-treatment disease progression and fractures, compared with the 57% specificity rate of bone scans performed after radiation therapy.108

Pathologic Fractures Pathologic fractures occur in 8% to 30% of patients with bone metastases.151-156 Proximal long bones are involved more commonly than distal bones; consequently, 50% of pathologic fractures occur in the femur, and 15% occur in

39  Palliative Care

the humerus (Fig. 39-1). The femoral neck and head are the most common locations for pathologic fracture because of the propensity for metastases to involve proximal bones and because of the stress of weight placed on this part of the femur. More than 80% of pathologic fractures occur in patients with breast (50%), kidney, lung, or thyroid cancer. Approximately 10% to 30% of pathologic fractures that occur in long bones where a metastatic lesion is present require surgical intervention. Clinically, the outcome for patients with a pathologic fracture resulting from a bone metastasis after surgical repair is comparable to that of patients sustaining a traumatic fracture.152-154 Prognosis is generally poor if hypercalcemia is present and if parenteral narcotics are required to control pain from other sites of bone metastases. In such cases, the decision for surgical intervention should be based on the severity of the symptoms associated with the fracture.152 As shown in Figure 39-2, postoperative radiation is often given after surgical fixation of a pathologic fracture to reduce the risk of progressive disease in the bone that could lead to instability of the internal fixation.157 The choice of treatment for pathologic fracture or impending fracture depends on the bone involved and the patient’s clinical status. Indications for surgical intervention of pathologic fracture or impending fracture include an expected survival duration of more than 6 weeks, the ability to accomplish internal stability of the fracture site, and the absence of coexistent conditions that would preclude ­early mobilization (e.g., metastasis involving weightbearing bones and lytic lesions more than 2 or 3 cm in

A

1021

size or ­metastases that destroy more than 50% of the cortex).152-156 It is unclear whether osteolytic metastases are more likely to lead to fracture than are osteoblastic lesions because osteoblastic lesions, by definition, have an osteolytic component that facilitates the formation of new bone.

Spinal Cord Compression The vertebral column is the site of metastatic tumors in 40% of patients who die of cancer. Approximately 70% of vertebral metastases involve the thoracic spine, 20% ­involve the lumbosacral region, and 10% involve the cervical spine. The average time from the original diagnosis of cancer to the development of metastatic spinal disease is 30 months. The development of spinal cord compression is associated with a poor overall prognosis, and the median survival duration among these patients is 7 months, with a 36% probability of a 1-year survival. After the diagnosis of epidural spinal cord compression, the median survival ­duration is 5 months. After epidural spinal cord compression is diagnosed, the mean survival duration is 14 months for patients with breast cancer, 12 months for those with prostate cancer, 6 months for those with malignant melanoma, and 3 months for those with lung cancer.5,144,151 Symptoms.  Pain is the initial symptom in ­approximately 90% of ­patients with spinal cord compression. Paraparesis or paraplegia occurs in more than 60% of cases, sensory loss in 70% to 80%, and bladder or bowel disturbances in 14% to 77%.88,144-150 The extent of the epidural mass influences prognosis, ­because a complete spinal block results in ­greater residual neurologic impairment than a partial block.

B

FIGURE 39-1.  A, Plain radiograph shows an extensive lytic lesion in the proximal humerus. B, Prophylactic internal fixation was performed to prevent a pathologic fracture. The patient, who experienced pain primarily in the hip, would have been put on crutches to reduce stress on the involved femur. A bone scan and radiographs obtained to exclude other sites of metastatic involvement identified this lesion in the humerus. The humerus would certainly have fractured if all of the patient’s weight had been displaced to the upper extremities with crutches.

FIGURE 39-2.  Typical radiation portal after fixation of a ­pathologic fracture of the femur. Radiation is given to treat ­residual disease around the internal fixation device, pubis, and acetabulum.

1022

PART 14  Special Considerations

Weakness can signal the rapid progression of symptoms, and 30% of patients with weakness become ­ paraplegic within 1 week. Rapid development of weakness, defined as weakness occurring less than 2 months after the appearance of spinal cord compression, most commonly occurs in patients with lung cancer, whereas in those with breast or prostate cancer, weakness can progress more slowly. Neurologic deficits can develop within a few hours in up to 20% of patients with spinal cord compression.144150,158,159 The severity of weakness at presentation is the most important predictor of recovery of function; 90% of patients who are ambulatory at presentation will be ambulatory after treatment. Only 13% of patients who are paraplegic will regain function, particularly if the paraplegia had been present for more than 24 hours before the initiation of therapy. More than 30% of patients who develop spinal cord compression are alive 1 year later, and 50% of them will remain ambulatory with appropriate therapy. Pain may be present for days or months before neurologic dysfunction appears. Unlike degenerative joint disease, which primarily occurs in the low cervical and low lumbar regions, pain caused by epidural spinal cord compression can occur anywhere in the spinal axis and is aggravated by recumbency. The most common site for epidural metastases is the thorax, followed by the lumbar region and the cervical region. Lung and breast cancer tends to metastasize to the thoracic spine, but colon and pelvic tumors have a predilection for the lumbosacral region.157 Back pain in any cancer patient, especially those with known metastatic involvement of the vertebral bodies,

should raise the suspicion of spinal cord compression. The risk of spinal cord compression exceeds 60% among ­patients with back pain and plain-film evidence of vertebral collapse from metastatic cancer.144-150,157-165 Epidural spinal cord disease has been documented in 17% of asymptomatic patients who have an abnormal bone scan but normal plain films; 47% of asymptomatic patients with vertebral metastases identified on bone scans and plain films also have associated epidural disease.158 Symptomatic patients with a normal vertebral contour and osteoblastic changes seen on a plain film and bone scan also should be evaluated for spinal cord compression (Fig. 39-3). Radiographic Determination. Radiographic determination of the involved spinal levels is essential for radiation treatment planning. Clinical assessment of the location of epidural spinal cord compression is incorrect in 33% of ­cases.144-146 Plain-film radiographs show involvement of more than one spinal level in about one third of patients. On these films, destruction of the pedicles commonly indicates spine metastases. In contrast, computed tomography (CT) scans show that the initial anatomic location of metastases is in the posterior portion of the vertebral body and that destruction of the pedicles occurs only in combination with involvement of the vertebral body (Fig. 39-4).162 ­

A

B FIGURE 39-3.  This bone scan demonstrates multifocal metastatic disease. Involvement of weight-bearing areas such as the pelvis and lower lumbar area significantly affects mobility.

FIGURE 39-4.  A, Sagittal MR image of the thoracic spine shows ­ involvement of the posterior aspect of the vertebral body, resulting in partial spinal cord compression. B, Axial CT scan shows direct impingement on the spinal cord.

39  Palliative Care

Osteoblastic bony expansion, common in prostate and breast cancers, can result in spinal cord compromise and vertebral compression fractures.161 If the results of MRI, tomographic studies, and surgical findings are included, more than 85% of patients are found to have evidence of multiple sites of vertebral involvement.144-150,158,162,166,167 Multiple sites of epidural involvement also are common; 10% to 38% of epidural metastases involve multiple noncontiguous spinal levels.157 Findings from 337 patients evaluated between 1985 and 1993 were reviewed to assess the benefit of complete spinal imaging.166 When the entire spine was evaluated radiographically, 32% of patients had multiple sites of ­epidural involvement; when imaging was limited to the symptomatic site, multiple areas of epidural tumor spread were detected in only 18% of cases. Although failure to image the cervical spine would have missed only 1% of epidural tumor deposits, this percentage increased to 21% if the thoracic or lumbar region was not fully imaged. The presence of multiple epidural tumor deposits was an independent prognostic factor for poorer survival. Secondary epidural tumor deposits were included in 93% of cases within the radiation portals. Treatment. Treatment for spinal cord compression includes emergent corticosteroids, radiation therapy, neurosurgical intervention, or a combination thereof. Radiation therapy is the treatment of choice in most cases (Fig. 39-5).95,157,168 Functional outcome depends on the level of symptoms at the time radiation is administered. The outcome of 166 patients who presented to a cancer center with spinal cord compression confirmed on MRI was reviewed in terms of performance and neurologic status.168 More than one half of the patients had prostate or breast cancer, and 13% had lung cancer. The thoracic spine was most commonly involved, alone (57%) or in combination with lesions in the lumbar region (11%). Multiple levels of epidural involvement were documented in 28% of ­patients. Symptoms included back pain in 122, limb ­ weakness

A

B

FIGURE 39-5.  A, Typical radiation portal for the treatment of multifocal disease involvement of the vertebral bodies and epidural region. B, Blastic lesions can be seen in the pedicles, ­especially at L2.

1023

in 128, sensory loss in 90, and sphincter disturbance in 53 patients. More than 50% of the patients had moderate to major neurologic disturbances. Radiation therapy was given to 92% of patients. Median survival duration from confirmation of spinal cord compression was 82 days (range, 1 to 349 days). Performance status improved in 16% of patients, and neurologic status improved in only 20%. In 55% to 60% of patients, performance and neurologic status remained the same, and 25% were worse after treatment.168 Thirty-two percent of patients died in the hospital. Most of the remaining 68% returned home; only 11% were sent to another hospital, 4% to a rehabilitation center, and 5% to a hospice. ­Pretreatment performance and neurologic status proved to be highly predictive of outcome. Because disability and pain continue even after treatment, ongoing medical intervention is needed to assist in home care efforts. Importance of Early Treatment.  Spinal cord compression resulting from metastatic tumor can be prevented or effectively treated when diagnosed early.95,144-150,157,158,162,168 The most common radiation therapy schedule for spinal cord compression and carcinomatous plexopathy is 30 Gy in 10 fractions; pain relief can be accomplished in 73% of patients treated in this ­manner.158,159 Although more prolonged courses of radiation therapy have been given for spinal cord compression, a short course was evaluated in a patient group with a poor overall prognosis.169 Among the 53 patients so treated, 11 had lung cancer, 11 had prostate cancer, 9 had gastrointestinal ­cancer, 5 had renal cell cancer, and 3 had breast cancer. Pain was present in 96% of patients, and motor weakness was present in 84%; 41% were unable to walk, and 12% were paraplegic. Radiation therapy consisted of a single 8-Gy fraction; a second 8-Gy fraction was given after 1 week to patients whose tumors responded or stabilized.169 The median field size was 136 cm2, and all patients treated to the upper abdomen were premedicated. Pain relief was accomplished in 67% of patients, and motor function improved in 63%. Early diagnosis and therapy were very important in predicting response to radiation therapy. Among patients who were ambulatory and had urinary continence before treatment, the ability to ambulate was preserved in 91% of patients, and the ability to maintain continence was preserved in 98%. However, fewer of the patients who were nonambulatory or incontinent experienced improvement in these functions (38% and 44%, respectively).169 The ­median survival duration was 5 months; 30% survived for 1 year. Survival was significantly better among ambulatory than among nonambulatory patients. For patients with spinal cord compression and poor prognostic factors, a short course of radiation seemed to provide efficient and effective palliation. In the treatment of epidural spinal cord compression, a statistically significant improvement in functional outcome has been reported for laminectomy plus radiation relative to either modality used alone for certain clinical presentations. Laminectomy has been recommended for prompt reduction of tumor volume in an attempt to provide rapid relief from compression injury of the spinal cord and stabilize the spinal axis.95 The rate of tumor regression after radiation therapy is too slow in such cases to effect recovery of lost neurologic function.

1024

PART 14  Special Considerations

Several studies have indicated that the use of radiation to treat a partial spinal cord block resulted in 64% of patients regaining ambulation, 33% achieving normalization of sphincter tone, and 72% becoming pain-free, with a median survival duration of 9 months.147-150 With a complete spinal cord block, only 27% of patients show improvement in motor function after radiation therapy alone, and 42% continue to have pain. Surgical intervention therefore should be considered for patients presenting with a complete spinal cord block. In paraparetic patients who undergo laminectomy plus radiation, 82% regain the ability to walk, 68% have improved sphincter function, and 88% have relief of pain. Laminectomy for vertebral and epidural metastases is ­indicated for tumor progression in a previously irradiated area, for cases in which the spine is stabilized, for paraplegic patients with limited disease and good probability of ­survival, and for establishing a diagnosis.144-150 Adjuvant radiation therapy is often given to treat microscopic residual disease after neurosurgical intervention among patients who have not previously undergone irradiation. Surgical restoration of the vertebral alignment may be required because of neurologic compromise and pain caused by progressive vertebral collapse. Vertebral collapse may occur as a result of cancer or vertebral instability after cancer therapy (Fig. 39-6). Appropriate diagnostic studies and interventions should be pursued, because neurologic compromise and pain from vertebral instability can be as devastating as that from epidural spinal cord metastases.149,163 Paravertebral masses are most commonly associated with lung cancer and are rare in prostate cancer.164,165 Approximately 20% of patients with epidural spinal cord ­ compression have an associated paravertebral mass. ­Surgical resection combined with radiation therapy has been suggested to improve functional outcome when a

FIGURE 39-6.  Sagittal MR image shows a normal spinal cord but recurrent tumor extending from the vertebral body. Specialized treatment planning is needed to ensure that adequate radiation doses are delivered to the involved region.

paravertebral mass is associated with spinal cord compression because radiation alone is less effective for large tumor burdens.

Radiation Tolerance of the Spinal Cord The potential for developing radiation myelitis from total radiation doses that exceed 40 Gy at 2 Gy/fraction represents the limiting factor in the treatment of large tumor burdens near or involving the spinal canal. Moreover, the length of the spinal cord that needs to be irradiated substantially affects its radiation tolerance.170-172 ­Histopathologic changes observed after experimental fractionated irradiation of the spinal cord include white matter necrosis, ­massive hemorrhage, and segmental parenchymal atrophy; these changes are consistently associated with abnormal neurologic signs.172 Other pathologic responses involve focal fiber loss and white matter vacuolation. Other experimental findings have shown that the time course and the extent of long-term recovery from radiation depend on the specific type and age of the tissue.170,172 Radiation tolerance of the spinal cord, like that of other tissues, is based on the dose per fraction, total dose, and volume of tissue treated. The most important of these factors is the isodose per fraction. Clinical and experimental experience have failed to demonstrate any difference in radiosensitivity in different segments of the spinal cord.170,173 The risk of radiation myelitis in the cervicothoracic spine is less than 5% when a 60-Gy dose is given in 1.72-Gy fractions or when 50 Gy is given in daily 2-Gy fractions. Among patients who have received chemotherapy or have had spinal injury, the total dose to the spinal cord usually is limited to 40 Gy given in 2-Gy fractions to minimize the risk of irreversible radiation injury to the spinal cord. In addition to isodose, the total dose is extremely important in defining the radiation tolerance of the spinal cord. A steep curve based on total radiation dose predicts the risk of developing radiation myelopathy; a small increase in total radiation dose can result in a large increase in the risk of radiation myelopathy.170,172 However, estimates from a linear-quadratic model173 suggest that no progressive myelopathies would be produced from doses of 60 Gy in 2-Gy fractions to the spinal cord. Concerns have been raised that the conservative limit of radiation dose to the spinal cord has compromised tumor control. With conventional fractionation of 1.8 to 2 Gy, the incidence of myelitis at 5 years was 0.2% after 45 Gy and 1% to 5% after 60 Gy.173 These determinations from a linear-quadratic model have correlated well with clinical findings. Retreatment of a previously irradiated segment of the spinal cord places the patient at high risk for radiationinduced myelopathy, because other neurologic pathways cannot compensate for an injury to a specific level of the spinal cord. Within 6 months, repair of spinal tissue is about 30%. If 50 Gy was administered initially, this percentage of repair would represent 85% of the tolerance dose to the spinal cord.96,172 Allowing 1 to 2 years between radiation doses for repair, the tolerance dose to the spinal cord is determined to be about 135% of that with reirradiation. A total dose of 75 to 81 Gy, given in two radiation courses with conventional fractionation spaced 1 to 2 years apart, should not result in substantial injury. The use of conformal or intensity-modulated radiation therapy techniques in combination with spinal cord recovery provides another

39  Palliative Care

1025

option for reirradiation.95 If the patient cannot be treated surgically, reirradiation may be the only viable option for maintaining neurologic integrity for the remainder of the patient’s life (see Fig. 39-6). Given the limited overall prognosis, the risk of long-term radiation injury is small. The radiation tolerance of the spinal cord can be compromised by prior injury. Vasogenic edema of the spinal cord and nerve roots can be caused by compression injury, along with venous hemorrhage, loss of myelin, and ische­ mia. Synthesis of prostaglandin E2, which increases in the presence of vasogenic edema, can be inhibited by steroids or nonsteroidal antiinflammatory agents.157,170,172 Radiation can cause two forms of injury: white matter damage and vasculopathy. White matter damage is associated with diffuse demyelination and swollen axons that can be focally necrotic and have associated glial reaction. Vascular damage has been shown experimentally to be age dependent, and it can result in vascular necrosis.

Types of Injury Six major types of injury have been shown experimentally to result from irradiation of the spinal column; five occur in the spinal cord and one in the dorsal root ganglia. The most severe spinal lesions, all of which are caused by vascular damage and result in neurologic dysfunction, include white matter necrosis, hemorrhage, and segmental parenchymal atrophy. The two less severe forms of spinal lesion include focal fiber loss and scattered white matter vacuolation caused by damage to glial cells, axons, the vasculature, or a combination of these. Less severe sequelae occur after lower total doses of radiation and are less likely to result in neurologic dysfunction. In dorsal root ganglia, radiation damage causes the formation of intracytoplasmic vacuoles and loss of neurons and satellite cells that can affect sensory function. These injuries are distinct from the demyelination of the posterior columns associated with the selflimiting Lhermitte’s syndrome.172 Meningeal thickening and fibrosis can be observed after radiation, but the clinical significance of these conditions is unknown. Ependymal and nerve root injuries from radiation are rare. Experimental findings that high doses of steroids are more effective than lower doses in reversing edema and improving neurologic function are consistent with findings from clinical trials, including a well-designed randomized trial in which radiation therapy was given with high-dose corticosteroids or with placebo.157 The group that received corticosteroids was more likely to retain or regain ambulation than was the group that received placebo. Pain relief also was more rapid and complete with high-dose steroids (initial bolus of 100 mg followed by 4 mg of dexamethasone four times daily for the duration of radiation therapy) among patients suspected of having spinal cord ­compression. Persistent or recurrent pain after radiation therapy for vertebral metastases should be investigated to exclude the possibility of progressive disease inside or outside the radiation portal or mechanical spinal instability caused by a vertebral compression fracture (Fig. 39-7). Radiation-induced neural usually injury does not result in pain. Compression fractures that occur shortly after completion of radiation therapy are related more to bone insufficiency with tumor cell kill than to the effects of radiation. Radiation-related changes in the bone marrow that appear on MRI scans­

FIGURE 39-7.  Compression fraction of the 12th thoracic vertebral body in a patient who had previously experienced a pain-free interval after palliative radiation. Vertebral weakness and rapid tumor regression resulted in the compression fracture, which caused recurrent back pain because of spinal instability.

a­ fter radiation therapy can be distinguished from those characteristic of progressive disease.98-100,162,166,167,174,175 Bone mineral density evaluations are being used in trials to determine the response of lytic bone lesions to therapeutic radiation.176 Clinical and radiographic evidence suggests that leptomeningeal carcinomatosis also should be considered in the diagnostic evaluation of persistent or recurrent pain. Leptomeningeal carcinomatosis occurs more often than expected. For example, only one half of the cases of leptomeningeal carcinomatosis occurring with breast cancer will be diagnosed before the patients die.144,150,158,177,178 At least three cerebrospinal fluid samples are necessary to exclude the possibility of leptomeningeal disease, because the initial cytologic analysis fails to show tumor cells in 10% to 40% of cases.177 MRI can identify leptomeningeal disease among patients with normal findings on cerebrospinal fluid cytology, and MRI is sensitive and specific in locating regions of nodular leptomeningeal involvement. Except for nodular leptomeningeal involvement, for which localized radiation therapy may be of some benefit as an adjuvant treatment, intrathecal chemotherapy is the treatment of choice.178

Treating Disseminated Bone Metastases Other therapeutic approaches, including wide-field radiation therapy, systemic radionuclides, and bisphosphonates, have been used to treat disseminated bone metastases. These approaches are useful in augmenting the therapeutic effect of localized radiation and in preventing the

1026

PART 14  Special Considerations

­ rogression of asymptomatic bone lesions. Although it is p not usually a significant consideration in localized radiation, adequate bone marrow reserve is required for widefield radiation therapy and systemic radionuclides. Bone marrow scans can be performed to determine the volume of functioning marrow and to assess the feasibility of using these techniques.98,102,113,114,117

Wide-Field (Hemibody) Radiation Therapy Hemibody irradiation, with 6- to 10-Gy doses given in a single fraction to the upper, middle, or lower body, has been used to treat diffuse bone metastases. Response rates are consistently reported to be more than 70%, and 20% or more of patients experience complete relief of pain. Among patients with prostate cancer, the overall response rate is 80% (30% having complete relief of pain); the mean duration of pain relief is 15 weeks, and the mean survival duration is 25 weeks.178-181 The overall response rate for all types of primary tumors is 80%. About one half of the patients who receive hemibody irradiation experience relief of pain within 48 hours of treatment. More than one half of the patients so treated do not require further palliative irradiation for recurrent bone pain over the remainder of their lives. An RTOG study demonstrated that hemibody irradiation reduced the time to disease progression and decreased the need for subsequent palliative radiation therapy of bone metastases at 1 year of follow-up when compared with local-field irradiation alone.180 These results are consistent with the experiences of others using hemibody irradiation.179-181 Median survival duration after hemibody irradiation was substantially better among patients who ­ presented with a good performance status score; ­approximately 90% of patients experiencing complete response and 70% of those showing partial response had a good to excellent performance status before radiation ­therapy. Prior systemic therapy does not influence response to wide-field radiation therapy. In one study, patients given hemibody irradiation for symptomatic bone metastases refractory to chemotherapy and hormonal therapy had a complete response rate of 70% and partial response rate of 24%.181 Symptoms were palliated in 88% of cases when a previously treated area was reirradiated. Significant toxicity was observed in less than 10% of patients, whereas 50% experienced stabilization of disease at 1 year. With current strategies, the acute toxic effects associated with hemibody irradiation are manageable. Premedication prevents nausea, and partial shielding minimizes the dose to the lung and the risk of radiation pneumonitis. Acute hematologic ­depression is limited. Because of the potential for toxic effects to visceral structures and the difficulties in treatment setup, hemibody irradiation is not routinely used to palliate multifocal bone metastases. Hemibody irradiation may have permanent effects on bone marrow reserve, which can be problematic if subsequent chemotherapy is needed. For these reasons, radiopharmaceuticals, which cause no systemic toxic effects other than affecting blood counts, have gained popularity over hemibody irradiation for treating multifocal bone ­metastases. However, radiopharmaceuticals are most useful for blastic bone lesions, and hemibody irradiation may be an important palliative option for patients with diffuse ­lytic bone metastases that are refractory to other therapies.

Radiopharmaceuticals The most common radiopharmaceuticals used in the treatment of bone metastases are 89Sr and samarium 153 (153Sm). Many reports show that radiopharmaceuticals provide effective palliation of pain lasting up to 6 months in 60% to 80% of patients with breast or prostate cancer.54-56,112,117-123,182,183 Improvements in functional status and quality of life have been observed, and about 20% of ­patients have complete resolution of pain. Strontium 89. Pain control has been reported to be superior among patients with disseminated prostate cancer treated with 89Sr and localized radiation compared with localized radiation alone.54-56 89Sr combines with the calcium component of hydroxyapatite in osteoblastic lesions. Because the activity of 89Sr is limited to bone, its use is ­contraindicated when epidural disease is associated with vertebral metastases. Myelotoxicity, expressed as a 25% decline in initial platelet and white blood cell counts, is usually transient and is the only substantial toxic effect ­associated with 89Sr.54,114,117 Several clinical trials have shown 89Sr to be effective and easily administered in an outpatient setting. Because the radiation dose absorbed by the bone marrow after a dose of 89Sr is 2 to 50 times less than the dose delivered to osteoblastic lesions, the radiation dose that can be delivered to metastatic bone lesions with 89Sr can range from 3 Gy to more than 300 Gy. Clinical response to 89Sr, which is comparable to response to wide-field radiation therapy, has been documented in subjective and objective terms. Subjective response, defined as symptomatic improvement in a validated survey, has been reported by more than 80% of patients with prostate cancer. Evidence of objective response, documented as reductions in alkaline and acid phosphatase levels, also were associated with a decrease in uptake by metastatic lesions on sequential bone scans.54-56,112,117-123,183 Prior therapies for prostate cancer, including localized radiation therapy and systemic chemotherapy or hormone therapy, do not influence the toxicity of or the clinical response to 89Sr. As an adjuvant to localized EBRT for treatment of metastatic prostate cancer, 89Sr can improve pain relief and delay progression of disease. At 3 months’ follow-up, almost twice as many patients treated with 89Sr were reported to be painfree as patients treated with only localized EBRT.54-56 Analgesics were no longer required by 17% of patients treated with 89Sr, whereas only 2% of the patients treated with localized EBRT alone were able to discontinue analgesics. Quality-of-life assessments demonstrated increased physical activity along with improved pain relief after 89Sr was administered in conjunction with localized EBRT. A costbenefit analysis has suggested that the administration of 89Sr results in reductions in the cost of hospitalization for tertiary care.54-56 Samarium 153 and Other Radiopharmaceuticals. Several other radiopharmaceuticals available for treatment of disseminated bone metastases include 153Sm, gallium nitrate, phosphorus 32 (32P), and rhenium 186 (186Re).54-56,112,117-123,182,183 The therapeutic mechanism of action of radiopharmaceuticals reflects their physical and biologic half-lives in the bone lesion, their mean energy, and the delivered dose. Table 39-5182 summarizes some of the physical characteristics and clinical data for various

39  Palliative Care

1027

TABLE 39-5.   Characteristics of Radiopharmaceuticals Used to Treat Bone Metastases

Half-life (days)

Typical ­Response Rate (%)

Typical Time to Response (days)

Typical Duration of Response (mo)

Extent of Myelosuppression

↑ Energy beta

14

78

14

1.5-11.3

3+

Strontium 89

↓ Energy beta

50

70-80

7-20

≤6

1+

Samarium 153

Beta/gamma

 2

65

7-21

3-4

3+

Rhenium 186

Beta/gamma

 4

80

7-21

2

1+

Tin 117m

Beta/gamma

14

—

—

—

?

Iodine 131

↓ Energy beta

 8

72

—

≤2

0 to +1

Radioisotope

Radiation Type

Phosphorus 32

↑, increased; ↓, decreased. From Friedland J. Local and systemic radiation for palliation of metastatic disease. Urol Clin North Am 1999;26:391-402.

r­ adionuclides. 32P and 89Sr emit pure beta rays (i.e., little penetration in tissue), whereas 186Re and 153Sm emit beta rays and relatively high-energy gamma-ray photons (103 to 159 KeV) that penetrate tissue for some distance. Because 153Sm has a gamma-ray component, the distribution of the radiation dose can be imaged directly. Scans generated after injection of 153Sm are comparable with diagnostic scans obtained with technetium 99m; both show that mean skeletal uptake is more than 50% of the dose.121 Nonskeletal sites receive negligible radiation doses, and unabsorbed radiation is cleared within 6 to 8 hours of administration.120 In a double-blind, placebocontrolled clinical trial, 153Sm was shown to be effective in palliating painful bone metastases in patients with breast cancer.116 Pain relief occurred within 1 week and was maintained for at least 16 weeks after administration. Approximately 65% of patients responded within the first 4 weeks, and 43% had relief of pain of at least 16 weeks’ duration. No substantial toxic effects to the bone marrow have been observed. Recommended doses range from 1.0 to 1.5 mCi/kg. Multiple administrations are possible in about one third of cases.123 The mechanism of action of 186Re, which concentrates selectively in bone, is similar to that of technetium 99m diphosphonate, which is used in diagnostic bone scans. This characteristic allows direct imaging of the deposition of 186Re in bone metastases. The metastatic lesion receives tens of gray of radiation after administration of 186Re, whereas the radiation dose to the marrow is limited to 0.75 Gy.112,183,184 Thrombocytopenia seems to be the dose-limiting toxicity. As is true for 89Sr, decreases in prostate-specific antigen levels in men with prostate cancer have been observed after the administration of 186Re.184 32P has been used for more than 30 years to treat bone metastases; 77% of patients with prostate cancer experience substantial pain relief.112 The rates and durations of response observed with 32P are similar to those observed with wide-field radiation and 89Sr. However, the main disadvantage of 32P is severe hematologic toxicity, which develops in about 30% of patients. Marrow suppression resulting from radioisotopes is caused by penetrating gamma-radiation or the long half-life of the radioisotope. For example, 89Sr produces transient cytopenia because it has a fairly long half-life

(51 days), even though it emits a beta particle of low penetrance and low energy (1.46 MeV) (see Table 39-5).182 Hematologic toxicity is more pronounced in patients who begin ­treatment with platelet counts of less than 6 × 104, white blood cell counts of less than 2.5 × 103, or more than 30% involvement of the red marrow–bearing bone.114,117 ­Compromise of red marrow–bearing bone can be a consequence of tumor or prior radiation or chemotherapy. Comparative Trials. Comparative clinical trials are necessary to determine whether any difference exists in the onset of response and the pattern of response to 153Sm, 89Sr, and other radiopharmaceuticals. Clinical experience with 186Re and other radiopharmaceuticals is still limited, and these agents are in various stages of clinical investigation.54-56,112,117-123,182-184 Several strategies have been proposed to enhance the therapeutic effectiveness of these and other agents, including further dose-intensity studies, adjunctive administration of bisphosphonates, and concurrent administration with chemotherapy.118 Although bone marrow toxicity is limited with the doses of radiopharmaceuticals currently administered (alone or in combination with other agents), future dose-intensity studies may require temporary hematologic support with colony­stimulating factors. Sequential radiographs and bone scans after hormonal and radiopharmaceutical therapy for breast or prostate cancer demonstrate an osteoblastic response that ­ reflects remodeling of the bone in osteolytic osseous ­metastases.108,109 Approximately one third of patients have ­evidence of increased tracer uptake (flare) on bone scans ­obtained 8 to 16 weeks after treatment. Of these patients, 72% will respond to the treatment, but only 36% of those with little or no flare will respond to treatment. The advantages of radiopharmaceuticals include ease of administration, low toxicity, and potential for reuse. Future clinical studies will examine the use of radiopharmaceuticals earlier in the course of metastatic disease and in combination with other therapies. For example, 89Sr has been combined with several other chemotherapeutic agents in the treatment of prostate cancer, and early reports suggest synergistic effects in analgesia and cytotoxicity.183 Other areas in which research is needed are dose and schedule of administration and selection of the radiopharmaceutical that can provide the optimal response based on tumor type and other factors.

1028

PART 14  Special Considerations

Brain Metastases Radiation is used to relieve the symptoms of headache, seizure, nausea, vomiting, and neurologic dysfunction ­associated with brain metastases. Surgery, alone or in combination with radiation, is often performed for a solitary brain metastasis if the patient’s performance status is good and if the cancer burden is otherwise limited. Radiation therapy usually is given over 2 to 3 weeks in daily ­fractions of 2.5 to 3 Gy. Total radiation doses range from 25 Gy ­(given after resection) to 30 Gy (given for unresectable disease that carries a poor prognosis) (Fig. 39-8).185 Reirradiation of brain tumors and brain metastases is common. The role of radiosurgery was evaluated in an RTOG study of 156 cases of recurrent brain tumor and brain metastasis.186 Patients with recurrent brain tumors constituted 36% of the group, and these patients had had a median prior radiation dose of 60 Gy. The remaining 64% of patients had recurrent brain metastases, and their median prior radiation dose was 30 Gy. The maximum tolerated dose for tumors 2 cm or smaller was 24 Gy; that for 2 to 3-cm tumors was 21 Gy; and that for 3 to 4-cm tumors was 18 Gy.186 Tumor size was associated with a significant increase in the risk of grade 3 or higher neurotoxicity: tumors between 2 and 4 cm in diameter were 7 to 16 times more likely to be associated with grade 3 or higher neurotoxicity. The incidence of radiation-induced necrosis in this study was 5% at 6 months, 8% at 12 months, 9% at 19 months, and 11% at 24 months. In another study,97 patients with recurrent brain metastases treated with a linear accelerator instead of a gamma knife had 2.8 times the risk of progression within the radiosurgical target volume. This finding was not related to patient group, because 61% of patients with recurrent brain tumors, compared with 30% of patients with brain metastases, were treated with the gamma knife.97 EBRT has been used to re-treat patients with persistent symptoms caused by brain metastases who are not candidates for radiosurgery. If symptoms recur within 3 to 4 months of administration of radiation, reirradiation is unlikely to be effective. However, additional therapy may benefit patients who develop recurrent neurologic symptoms 6 to 9 months later and who have a Karnofsky ­performance status score of more than 50. With steroid administration and reirradiation, response rates of 60% have

A

been reported. A total dose of 25 to 30 Gy is prescribed in 2- to 2.5-Gy fractions.97 The risk of late toxicity depends on the total cumulative dose, volume of tumor reirradiated, and interval between the first and second course of radiation therapy.96 A 14% risk of necrosis exists when a dose of 86 Gy or more is given in 2-Gy fractions. Reirradiation with EBRT for recurrent gliomas was investigated in 24 patients who had previously received a median radiation dose of 60 Gy.97 With reirradiation, the total cumulative dose was limited to 92 Gy; reirradiation was given in 1.5-Gy fractions in combination with oral lomustine (130 mg/m2 every 6 weeks). The response rate was 57%, and two patients survived more than 5 years. Median time to progression and overall survival duration were 8 and 14 months, respectively. No substantial neurologic toxicity was reported from use of this approach.97 Brachytherapy has been used to treat recurrent brain ­tumors. This technique has been applied in selected ­patient groups: those with good neurologic function and whose tumors are well-defined and located within 5 cm of midline.97 The primary complication has been necrosis, which necessitated reoperation in 49% of patients. Median ­survival duration among patients with glioblastoma multiforme was 54 weeks.97 Prognosis also has been linked to scores on the MiniMental Status Examination (MMSE) before and after palliative radiation for brain metastases.187 In an RTOG study, 445 patients with brain metastases were given 1.6 Gy twice daily to a total dose of 54.4 Gy or a radiation therapy regimen totaling 30 Gy given in 10 fractions to the entire brain. The average score before radiation was 26.5 (range, 11 to 30); 16% of the patients had scores of less than 23 before radiation, indicating possible dementia.187 Median survival during treatment was 4.2 months, and 37% of patients were alive at 6 months. On multivariate analysis, the pretreatment MMSE score and Karnofsky performance score were predictive of outcome. Survival rates of 81% at 6 months and 66% at 1 year were observed for patients whose MMSE scores before and after irradiation were higher than 23. Death from brain metastases was associated with decline in MMSE score and site of the primary tumor (breast versus lung). Overall, a dose of 30 Gy administered over 2 weeks proved to be an effective treatment for brain metastases, and no substantial advantage was observed with a more prolonged course of therapy.

B

FIGURE 39-8.  A, A typical radiation portal used to treat brain metastases. B, A simulation film obtained to encompass the base of the skull.

39  Palliative Care

Lung Tumors or Metastases Involvement of the lung and mediastinum, whether by a locally advanced primary tumor or metastases from elsewhere, can result in a variety of symptoms, some of them emergent, and it often requires palliative intervention.188 Pain can result from tumor invasion of the ribs and nerve roots of the chest wall (Fig. 39-9). Vertebral involvement can be associated with spinal cord compression. Obstructive pneumonitis and hemoptysis can result from ­bronchial obstruction. Mediastinal infiltration can cause superior vena cava syndrome. All of these clinical presentations can be palliated with EBRT that encompasses disease evident on diagnostic images and that treats pain referred along involved nerve roots. A schedule of 30 Gy delivered in 10 fractions over 2 to 3 weeks is typically prescribed for treatment of previously unirradiated sites. Using such a palliative radiation schedule for 65 patients with non–small cell lung cancer resulted in decreased hemoptysis in 79% of patients, decreased arm or shoulder pain in 56%, decreased chest wall pain in 53%, decreased cough in 49%, and decreased dyspnea in 39%.189 However, reductions in fatigue (experienced by 22% of patients) and loss of appetite (experienced by 11%) were minimal. Decreases in dyspnea strongly correlated with ­radiographic evidence of response.189 Although palliative radiation is effective, most patients have residual symptoms that require medical management. If the tumor site has been previously irradiated, radiation therapy techniques that exclude critical anatomic structures such as the spinal cord are used. A review of more than 1500 cases of lung cancer treated from 1982 to 1997 revealed that 23 patients underwent reirradiation for relief of symptoms.190 Median duration of follow-up was 3.2 months (range, 0 to 17.5 months). Consistent with other published reports, hemoptysis decreased or resolved in all 6 patients with that symptom, 4 of 5 patients had reduced pain, 9 of 15 had reduced coughing, and 11 of 15 had decreased dyspnea.190 The risk of radiation pneumonitis associated with such treatments can be decreased by using narrow margins around the radiation portal.

1029

Other approaches, such as brachytherapy, can be used when the symptomatic site is well localized and accessible. In these cases, large doses of radiation can be ­delivered within a few minutes by a high-dose-rate brachytherapy unit.94,181 Brachytherapy can be used as a boost to EBRT or to re-treat a previously irradiated area. Precise details of techniques and radiation doses vary, but most schedules involve giving 6 to 10 Gy every 1 to 2 weeks for up to 3 fractions or a single fraction of 10 to 20 Gy. Overall symptom response rates range between 70% and 90%, and palliation is durable in 60% of patients.181 Relief of specific symptoms included hemoptysis in 54% of patients, cough in 94%, dyspnea in 86%, and pulmonary collapse in 51%.94 Patients with recurrent symptoms can be treated with another round of brachytherapy or with other methods. The incidence of radiation bronchitis and stenosis among such patients was 17%, and fatal hemoptysis occurred in 5.5%. Tumors of the right main stem bronchus and the right upper lobe bronchus seem to be more susceptible to development of fatal hemorrhage than left-sided lesions. Symptoms of dyspnea should be addressed medically while a response to radiation therapy is awaited. Oxygen therapy, maintenance of adequate hemoglobin levels, and opioids are effective treatments for dyspnea; respiratory depression will not occur if the patients have been previously treated with opioids. Although benzodiazepines often are used, these agents have been found to be ineffective in four of five randomized trials.191 Brachytherapy also has been used in the treatment of esophageal cancer. Rates of relief of dysphagia in several reports range between 70% and 85%.94 A combination of a short course of EBRT (30 Gy in 10 fractions) plus brachytherapy as a boost relieves dysphagia for 3 to 6 months. Relative contraindications include a tumor length of 10 cm or more, extension to the gastroesophageal junction or ­cardia, skip lesions, extensive extraesophageal spread, macroscopic regional adenopathy, tracheoesophageal fistula, cervical esophageal involvement, or stenosis that cannot be bypassed.94

Head and Neck Tumors

FIGURE 39-9.  Axial CT scan shows tumor masses invading the chest wall.

Palliation of locally advanced head and neck malignancies often requires a more protracted radiation schedule because of the sensitivity of the structures, such as the arytenoids and cervical spinal cord, in the radiation field. List and colleagues192 evaluated quality-of-life variables before and during chemoradiation therapy for locally advanced head and neck cancer and at 3-month intervals ­after treatment. Severe treatment-related toxic effects led to declines in all quality-of-life and functional domains. By 12 months, marked improvement was observed, although up to one third of patients continued to experience difficulty swallowing, hoarseness, and mouth or throat pain, all of which were present at levels comparable to those before treatment. Substantial increases in dry mouth (58% versus 17%), inability to distinguish tastes (32% versus 8%), and the need for a soft-food diet (82% versus 42%) were documented after chemoradiation therapy. The pretreatment quality-of-life level was the best predictor of quality of life, regardless of residual side effects or functional impairment.

1030

PART 14  Special Considerations

Aggressive management of treatment-related side e­ ffects can improve tolerance of head and neck irradiation. In a study by Janjan and colleagues,193 symptoms of head and neck irradiation were effectively controlled by using a specific pain management regimen that included opioid analgesics. With this regimen, the incidence of pain lasting more than 12 hours/day decreased by 35%, and the incidence of mean weight loss of more than 5 kg decreased by 41%. Moderate to severe pain during the first week of radiation therapy was experienced by 15% of patients in this study. Significant improvements in symptom control were observed from the use of opioid analgesics over the use of traditional codeine elixirs to control symptoms. Pain intensity decreased by 11% by the third week of radiation therapy, by 24% during the fifth week, and by 26% during the seventh week. At the completion of radiation therapy, pain intensity at its worst was mild in 43% of patients, moderate in 23% of patients, and severe in 6%. The frequency of the pain was occasional in 46% of patients, most of the day in 17%, and always in 6%. Late radiation effects are of particular concern in reirradiation of recurrent head and neck cancer. In an assessment of the total biologic effect of the retreatment of recurrent nasopharyngeal cancer,194 the severity of damage during the initial course of radiation was the major determinant of complications associated with retreatment. Repair of normal tissues was evident because the tolerated doses were higher than expected for a single course of radiation.194 The complication-free survival rate at 5 years was lower in the retreatment group (48%) than in a group initially treated with a single course of radiation (81%). A 20% risk of complications at 5 years would occur with a total biologically equivalent dose of 143 Gy3 with retreatment or 111 Gy3 among patients treated with a single course.

Pelvic Disease Locally advanced or metastatic disease in the pelvis often manifests as hemorrhage and visceral, lymphovascular, or nerve root obstruction. Treatment may require emergent radiation therapy or surgery. Hemorrhage is commonly associated with tumors involving the rectum and genitourinary tracts. As is true for tumors in the lung, radiation is an effective means of stopping active bleeding. Colorectal cancer is often diagnosed among patients with unexplained bleeding. Colorectal tumor involvement also can produce obstruction, often requiring the placement of a stent to maintain the integrity of the visceral lumen while the radiation is being given.188,195 Occasionally, a diverting colostomy is required to bypass intestinal obstruction or the formation of a fistula. If these procedures are not performed, intestinal colic can be palliated quickly with opioids, but anticholinergics such as scopolamine, atropine, and loperamide decrease peristalsis in the smooth muscle of the intestinal tract.195,196 Intestinal obstruction also can produce nausea and vomiting, which can be palliated with nasogastric decompression and pharmacologic agents. Octreotide, an analogue of somatostatin, reduces gastrointestinal secretions, motility, and splanchnic blood flow; has a proabsorptive effect on water and ions; and may inhibit the secretion of vasoactive intestinal peptides. These actions help to disrupt the cycle of distention–intestinal secretion–peristalsis and bleeding.196,197

Because colorectal tumors tend to be locally advanced, radiation therapy is given preoperatively with conventional fractionation if little or no metastatic disease is evident. This regimen is intended to stop bleeding, render the tumor operable, and provide a chance for cure. For extensive metastatic disease, 30 to 35 Gy of radiation can be given over 2 to 3 weeks to palliate symptoms of ­bleeding and obstruction. In two studies, 35 Gy in 14 fractions, 30 Gy in 10 fractions, and 30 Gy in 6 fractions were given twice weekly for 3 weeks for this purpose.198,199 A diverting colostomy was required in 16% of patients before radiation therapy. No substantial treatment-related toxic effects were observed. Regardless of whether radiation was given in conventional or hypofractionated schedules or with or without infusional fluorouracil, symptoms from the primary tumor resolved in 94% of cases, and the endoscopic complete response rate was 36%. The colostomy-free survival rate was 87% among patients who underwent palliative chemoradiation ­ therapy and 51% among patients who received preoperative chemoradiation therapy. When surgery was performed, the 2-year survival rate was 46%, compared with 11% after palliative chemoradiation therapy. Control of symptomatic pelvic disease was not substantially different between the two treatments (81% for palliative chemoradiation ­therapy and 91% for preoperative chemoradiation therapy).198 ­Predictors of control of pelvic disease included the severity of pelvic pain at presentation, a biologically equivalent radiation dose of less than 35 Gy at 2 Gy/fraction, and poor tumor differentiation. Poor tumor differentiation and selection for palliative chemoradiation therapy predicted worse overall survival. Tumors involving the cervix can hemorrhage and require emergent radiation therapy or other gynecology structures. Such therapy can be delivered by means of a cone applied directly to the cervix, by opposed anterior and posterior pelivic fields, or by brachytherapy. Radiation is typically given in single large (e.g., 10 Gy) fractions, repeated as needed every 3 to 4 weeks.201 Bladder cancer or tumors that secondarily invade the bladder can result in significant bleeding that can be palliated by EBRT. Urinary obstruction is common in locally advanced pelvic cancer, especially tumors of the prostate and cervix. Occasionally, placement of a urinary stent or a urostomy or nephrostomy is required until radiation­induced tumor regression is sufficient to reestablish the integrity of the urinary tract.188 As is true of tumors in the bowel and gynecologic tracts, the formation of vesical fistulas from the tumor itself or from tumor regression remains a concern. In one study,200 palliative radiation administered as 35 Gy in 10 fractions or 21 Gy in 3 fractions was evaluated in 272 patients with bladder cancer. At a follow-up of at least 3 months, 71% of those who received 35 Gy and 64% of those who received 21 Gy achieved symptomatic improvement.200 No difference in the radiation schedules was identified in terms of efficacy, toxicity, or survival. The overall survival rates were approximately 77% at 3 months, 35% at 12 months, and 19% at 24 months. The pelvic lymph nodes and major blood vessels can become obstructed by tumor. This condition occurs most often when tumors arise in pelvic structures but also can occur with pelvic metastases from breast and other types

39  Palliative Care

of cancer.188 Lymphovascular obstruction results in painful edema that is refractory to diuretic and other therapies. Severe obstruction can lead to fluid and electrolyte imbalances. Pelvic irradiation can relieve lymphovascular obstruction by causing tumor regression. Pelvic tumors can invade the sacral plexus and result in intractable pain. These tumors proceed along nerve roots and can invade the sacrum. Pain caused by visceral obstruction or lymphovascular obstruction, or both, often responds more rapidly to palliative radiation than does the neuropathic pain arising from sacral plexus involvement. Other radiation therapy approaches such as brachytherapy are of limited effectiveness when the cancer persists or recurs after EBRT; pain management techniques are often required to control pain associated with sacral plexus ­involvement. Reirradiation of the pelvis for symptomatic recurrence has been well tolerated, although data are limited. Care must be taken to exclude the intestines from the radiation portals given the risk of late small-bowel toxicity. Although data on treatment volume are limited, the cumulative radiation dose, reirradiation dose, and time interval between courses have not been related to late toxicity.96

Skin And Subcutaneous Tissue Involvement Tumors can cause ulceration of the skin and subcutaneous tissues that is often painful and distressing because of constant drainage and because they create the potential for sepsis in immunocompromised patients. Localized radiation can be used to destroy tumor and allow re-­epithelialization of the skin. Typically, electron-beam radiation therapy is used because it can better target the skin and subcutaneous tissues while avoiding irradiation of underlying normal structures. Although 10 radiation treatments are usually given, the course of irradiation can be abbreviated further and can be administered over 1 to 5 days. Occasionally, skin and subcutaneous lesions are treated with brachytherapy, in which the radioactive sources are placed in a mold that sits on top of the tumor, and treatment is delivered over a few minutes (for high-dose-rate therapy) or over a few days (for low-dose-rate therapy).

Diagnostic and Therapeutic Recommendations Pain that persists after palliative radiation should be evaluated to exclude progression of disease in the treated area; extension of disease outside the radiation portal, which can cause referred pain; and bone fracture. Reduced cortical strength can result in compression, stress, or microfractures. Plain radiographs often clearly delineate lytic changes and fractures after radiation therapy. Radiation therapy remains an important modality in palliative care. Several clinical, prognostic, and therapeutic factors must be considered in choosing the optimal regimen. Adequate management of cancer-related pain is important during and after palliative radiation. For locally advanced and metastatic cancer, efficient and effective palliative treatment is imperative to relieve symptoms, improve function, and minimize disease-related morbidity.

1031

More specific prognostic criteria are needed to identify the appropriate treatment regimen for bone metastases. A staging system for metastatic disease affecting the ­skeleton should take into consideration performance ­status, type and extent of bone and visceral involvement, time to disease progression, and primary tumor site, and it should account for prior failed therapies. Because only the bone is treated, EBRT or radiopharmaceuticals continue to be considered localized therapies for systemic disease. The response to and outcomes after radiation therapy must be specifically defined relative to the irradiated and ­unirradiated sites of disease, and other antineoplastic and supportive care therapies should be administered as needed. Radiation therapy is an important means of treating localized symptoms related to tumor involvement because it provides a wide range of therapeutic options. Radiobiologic principles, the radiation tolerance of adjacent normal tissues, and the patient’s clinical condition influence the selection of the radiation technique, dose, and fraction size. However, the optimal dose and treatment schedule are not fully defined. Further study is necessary to integrate validated pain scores, use of analgesics, prognostic factors, and radiobiologic principles to better define the most efficient and efficacious treatment schedule according to the clinical presentation. Palliative radiation should be integrated into multidisciplinary therapeutic approaches because of the need to treat associated symptoms and other underlying medical problems. Antineoplastic therapy can induce tumor regression, relieve cancer-related symptoms, and maintain functional integrity. Control of cancer-related pain through the use of analgesics is imperative to provide comfort during radiation therapy and while awaiting response to that therapy. Pain represents a sensitive measure of disease activity. Close follow-up should be performed to ensure control of cancer and treatment-related pain and to identify progressive or recurrent disease.

REFERENCES    1. Cleeland CS, Gonin R, Hatfield AK, et al. Pain and its treatment in outpatients with metastatic cancer. N Engl J Med 1994;330:592-596.    2. Jacox AK, Carr DB, Payne R, (eds). Management of Cancer Pain. Clinical Practice Guideline No. 9. AHCPR publication 94-0592. Rockville, MD: Agency for Health Care Policy and Research, 1994.    3. Jacox A, Carr DB, Payne R. New clinical practice guidelines for the management of pain in patients with cancer. N Engl J Med 1994;330:651-655.    4. Brescia FJ, Portenoy RK, Ryan M, et al. Pain, opioid use, and survival in hospitalized patients with advanced cancer. J Clin Oncol 1992;10:149-155.    5. American Cancer Society. Cancer Facts & Figures, 2008. Atlanta: American Cancer Society, 2008.    6. American Society of Clinical Oncology. Outcomes of cancer treatment for technology assessment and cancer treatment guidelines. J Clin Oncol 1996;14:671-679.    7. Cassel EJ. The nature of suffering and the goals of medicine. N Engl J Med 1982;306:639-645.    8. Porzsolt F. Goals of palliative cancer therapy: scope of the problem. Cancer Treat Rev 1993;19(Suppl A):3-14.    9. Rubens RD. Approaches to palliation and its evaluation. Cancer Treat Rev 1993;19(Suppl A):67-71.   10. Porzsolt F, Tannock I. Goals of palliative cancer therapy. J Clin Oncol 1993;11:378-381.

1032

PART 14  Special Considerations

  11. Cella DF, Tulsky DS. Quality of life in cancer: definition, purpose, and method of measurement. Cancer Invest 1993;11:327336.   12. Janjan NA. Radiation for bone metastases-conventional techniques and the role of systemic radiopharmaceuticals. Cancer 1997;80:1628-1645.   13. Liu L, Meers K, Capurso A, et al. The impact of radiation therapy on quality of life in patients with cancer. Cancer Practice 1998;6:237-242.   14. Arcangeli G, Micheli A, Arcangeli G, et al. The responsiveness of bone metastases to radiotherapy: the effect of site, histology and radiation dose on pain relief. Radiother Oncol 1989;14:95-101.   15. Mithal NP, Needham PR, Hoskin PJ. Retreatment with radiotherapy for painful bone metastases. Int J Radiat Oncol Biol Phys 1994;29:1011-1014.   16. Manfredi PL, Chandler S, Pigazzi A, et al. Outcomes of cancer pain consultations. Cancer 2000;89:920-924.   17. Janjan NA, Payne R, Gillis T, et al. Presenting symptoms in patients referred to a multidisciplinary clinic for bone metastases. J Pain Symptom Manage 1998;16:171-178.   18. Stafford RS, Cyr PL. The impact of cancer on the physical function of the elderly and their utilization of health care. Cancer 1997;80:1973-1980.   19. Cohen HJ. Cancer and the functional status of the elderly. Cancer 1997;80:1883-1886.   20. Hanks GE. The crisis in health care costs in the United States: some implications for radiation oncology. Int J Radiat Oncol Biol Phys 1992;23:203-206.   21. Emanuel EJ, Fairclough DL, Slutsman J, et al. Understanding economic and other burdens of terminal illness: the experience of patients and their caregivers. Ann Intern Med 2000;132:451459.   22. Rabow MW, Hardie GE, Fair JM, et al. End of life care content in 50 textbooks from multiple specialties. JAMA 2000;283:771778.   23. Morita T, Tsunoda J, Inoue S, et al. Contributing factors to physical symptoms in terminally ill cancer patients. J Pain Symptom Manage 1999;18:338-346.   24. Pargeon KL, Hailey BJ. Barriers to effective cancer pain management: a review of the literature. J Pain Symptom Manage 1999;18:358-368.   25. Von Roenn JH, Cleeland CS, Gonin R, et al. Physicians attitudes and practice in cancer pain management: a survey from the Eastern Cooperative Oncology Group. Ann Intern Med 1993;119:121-126.   26. Sherry MM, Greco FA, Johnson DH, et al. Breast cancer with skeletal metastases at initial diagnosis-distinctive clinical characteristics and favorable prognosis. Cancer 1986;58:178-182.   27. Sherry MM, Greco FA, Johnson DH, et al. Metastatic breast cancer confined to the skeletal system. Am J Med 1986;81:381386.   28. Jacobson AF. Musculoskeletal pain as an indicator of occult ­malignancy. Arch Intern Med 1997;157:105-109.   29. Vuorinen E. Pain as an early symptom in cancer. Clin J Pain 1993;9:272-278.   30. Pienta KJ, Esper PS. Risk factors for prostate cancer. Ann Intern Med 1993;118:793-833.   31. Abbas F, Scardino PT. The natural history of clinical prostate carcinoma. Cancer 1997;80:827-833.   32. Chisholm GD, Rana A, Howard GCW. Management options for painful carcinoma of the prostate. Semin Oncol 1993;20(Suppl 2): 34-37.   33. Borre M, Nerstrom B, Overgaard J. The natural history of prostate carcinoma based on a Danish population treated with no intent to cure. Cancer 1997;80:917-928.   34. Soloway MS, Hardeman SW, Hickey D, et al. Stratification of patients with metastatic prostate cancer based on extent of disease on initial bone scan. Cancer 1988;61:195-202.   35. Yamashita K, Denno K, Ueda T, et al. Prognostic significance of bone metastases in patients with metastatic prostate cancer. Cancer 1993;71:1297-1302.   36. Perez CA, Cosmatos D, Garcia DM, et al. Irradiation in relapsing carcinoma of the prostate. Cancer 1993;71:1110-1122.

  37. Lai PP, Perez CA, Lockett MA. Prognostic significance of pelvic recurrence and distant metastases in prostate carcinoma following definitive radiotherapy. Int J Radiat Oncol Biol Phys 1992;24:423-430.   38. Chang VT, Thaler HT, Polyak TA, et al. Quality of life and ­survival–the role of multidimensional symptom assessment. Cancer 1998;83:173-179.   39. Grabowski CM, Unger JA, Potish RA. Factors predictive of completion of treatment and survival after palliative radiation therapy. Radiology 1992;184:329-332.   40. Reuben DB, Mor V, Hiris J. Clinical symptoms and length of survival in patients with terminal cancer. Arch Int Med 1988;148:1586-1591.   41. Fielding LP, Henson DE. Multiple prognostic factors and outcome analysis in patients with cancer—communication from the American Joint Committee on Cancer. Cancer 1993;71:24262429.   42. Portenoy RK, Miransky J, Thaler HT, et al. Pain in ambulatory patients with lung or colon cancer. Cancer 1992;70:1616-1624.   43. Knudson G, Grinis G, Lopez-Majano V, et al. Bone scan as a stratification variable in advanced prostate cancer. Cancer 1991;68:316-320.   44. Sabbatini P, Larson SM, Kremer A, et al. Prognostic significance of extent of disease in bone in patients with androgen-independent prostate cancer. J Clin Oncol 1999;17:948-957.   45. Vigano A, Bruera E, Jhangri GS, et al. Clinical survival predictors in patients with advanced cancer. Arch Intern Med 2000;160:861-868.   46. Cleeland CS, Janjan NA, Scott CB, et al. Cancer pain management by radiotherapists: a survey of Radiation Therapy Oncology Group physicians. Int J Radiat Oncol Biol Phys 2000;47:203-208.   47. Lamont EB, Christakis NA. Some elements of prognosis in terminal cancer. Oncology 1999;13:1165-1170.   48. SUPPORT: Study to understand prognoses and preferences for outcomes and risks of treatments. Study design. J Clin Epidemiol 1990;43:1S-123S.   49. Zhong Z, Lynn J. Review of the Lamont/Christakis article [Some elements of prognosis in terminal cancer]. Oncology 1999;13:1172-1173.   50. Archer VR, Billingham LJ, Cullen MH. Palliative chemotherapy: no longer a contradiction in terms. Oncologist 1999;4:470-477.   51. Oh WK, Kantoff PW. Treatment of locally advanced prostate cancer: is chemotherapy the next step? J Clin Oncol 1999;17:36643675.   52. Janjan NA. An emerging respect for palliative care in radiation oncology. J Palliat Med 1998;1:83-88.   53. SBU: The Swedish Council on Technology Assessment in Health Care. Acta Oncol 1997;35(Suppl 6):89-97.   54. Porter AT, Ben-Josef E. Strontium 89 in the treatment of bony metastases. In DeVita VT, Hellman S, Rosenberg SA (eds): Important Advances in Oncology. Philadelphia: JB Lippincott, 1995, pp 87-94.   55. Porter AT, McEwan AJB, Powe JE, et al. Results of a randomized phase III trial to evaluate the efficacy of strontium 89 adjuvant to local field external beam irradiation in the management of endocrine resistant metastatic prostate cancer. Int J Radiat Oncol Biol Phys 1993;25:805-813.   56. Porter AT, McEwan AJB. Strontium 89 as an adjuvant to external beam radiation improves pain relief and delays in disease progression in advanced prostate cancer: results of a randomized controlled trial. Semin Oncol 1993;20:38-43.   57. Dale RG, Jones B. Radiobiologically based assessments of the net costs of fractionated radiotherapy. Int J Radiat Oncol Biol Phys 1996;36:739-746.   58. Janjan NA. Radiotherapeutic approaches to cancer pain management. Highlights Oncol Pract 1997;14:103-113.   59. Hillner BE, Weeks JC, Desch CE, et al. Pamidronate in prevention of bone complications in metastatic breast cancer: a cost­effectiveness analysis. J Clin Oncol 2000;18:72-79.   60. Dranitsaris G, Hsu T. Cost utility analysis of prophylactic pamidronate for the prevention of skeletal related events in patients with advanced breast cancer. Support Care Cancer 1999;7:271279.

39  Palliative Care   61. Macklis RM, Cornelli H, Lasher J. Brief courses of palliative radiotherapy for metastatic bone pain. Am J Clin Oncol 1998;21:617-622.   62. Hortobagyi GN, Theriault RL, Porter L, et al. Efficacy of pamidronate in reducing skeletal complications with breast cancer and lytic bone metastases. N Engl J Med 1996;335:1785-1791.   63. Hortobagyi GN, Theriault RL, Lipton A, et al. Long-term prevention of skeletal complications of metastatic breast cancer with pamidronate. J Clin Oncol 1998;16:2038-2044.   64. Theriault RL, Lipton A, Hortobagyi GN, et al. Pamidronate reduces skeletal morbidity in women with advanced breast cancer and lytic bone lesions: a randomized, placebo-controlled trial— Protocol 18 Aredia Breast Cancer Study Group. J Clin Oncol 1999;17:846-854.   65. Lipton A, Theriault RL, Hortobagyi GN, et al. Pamidronate prevents skeletal complications and is effective palliative treatment in women with breast carcinoma and osteolytic bone metastases. Cancer 2000;88:1082-1090.   66. Body JJ, Bartl R, Burckhardt P, et al. Current use of bisphosphonates in oncology. J Clin Oncol 1998;16:3890-3899.   67. Bloomfield DJ. Should bisphosphonates be part of the standard therapy of patients with multiple myeloma or bone metastases from other cancers? An evidence-based review. J Clin Oncol 1998;16:1218-1225.   68. Lipton A. Bisphosphonates and breast cancer. Cancer 1997;80: 1668-1673.   69. Hillner BE. The role of bisphosphonates in metastatic breast ­cancer. Semin Radiat Oncol 2000;10:250-253.   70. Diel IJ, Solomayer EF, Costa SD, et al. Reduction in new metastases in breast cancer with adjuvant clodronate treatment. N Engl J Med 1998;339:357-363.   71. Thames HD, Buchholz TA, Smith CD. Frequency of first metastatic events in breast cancer: implications for sequencing of systemic and local-regional treatment. J Clin Oncol 1999;17:26492658.   72. Kelly JB, Payne R. Pain syndromes in the cancer patient. Neurologic Clinics 1991;9:937-953.   73. Portenoy RK. Cancer pain management. Semin Oncol 1993;20:19-35.   74. Foley KM, Arbelt E. Management of cancer pain. In DeVita VT, Hellman S, Rosenberg SA (eds): Cancer: Principles and Practice of Oncology, 3rd ed, vol 2.Philadelphia: JB Lippincott, 1989, pp 2064-2087.   75. Paice JA, Toy C, Shott S. Barriers to cancer pain relief: fear of tolerance and addiction. J Pain Symptom Manage 1998;16:1-9.   76. Weissman DE. Doctors, opioids, and the law: the effect of controlled substances regulations on cancer pain management. Semin Oncol 1993;20(Suppl 1):53-58.   77. Hill CS. The barriers to adequate pain management with opioid analgesics. Semin Oncol 1993;20(Suppl 1):1-5.   78. Lebovits AH, Florence I, Bathina R, et al. Pain knowledge and attitudes of healthcare providers: practice characteristic differences. Clin J Pain 1997;13:237-243.   79. Joranson DE, Ryan KM, Gilson AM, et al. Trends in medical use and abuse of opioid analgesics. JAMA 2000;283:1710-1714.   80. World Health Organization. Cancer Pain Relief. Geneva: World Health Organization, 1986.   81. Lipman AG. New and alternative noninvasive opioid ­ dosage forms and routes of administration. In Berger A, Portenoy RK, Weissman DE (eds): Principles and Practice of Supportive ­Oncology, Update. Philadelphia: Lippincott Williams & Wilkins, 2000, pp 1-8.   82. Hassenbusch SJ, Portenoy RK. Current practices in intraspinal therapy—a survey of clinical trends and decision making. J Pain Symptom Manage 2000;20:S4-S11.   83. Bennett G, Serafini M, Burchiel K, et al. Evidence-based review of the literature on intrathecal delivery of pain medication. J Pain Symptom Manage 2000;20:S12-S36.   84. Bennett G, Burchiel K, Buchser E, et al. Clinical guidelines for intraspinal infusion: report of an expert panel. J Pain Symptom Manage 2000;20:S37-S43.

1033

  85. Ratanatharathorn V, Powers WE, Moss WT, et al. Bone ­metastases: review and critical analysis of random allocation trials of local field treatment. Int J Radiation Oncol Biol Phys 1999;44:1-18.   86. Awan AM, Weichselbaum RR. Palliative radiotherapy. Hematol Oncol Clin North Am 1990;4:1169-1181.   87. Bates T, Yarnold JR, Blitzer P, et al. Bone metastases consensus statement. Int J Radiat Oncol Biol Phys 1992;23:215-216.   88. Bates T. A review of local radiotherapy in the treatment of bone metastases and cord compression. Int J Radiat Oncol Biol Phys 1992;23:217-221.   89. Niewald M, Tkocz HJ, Abel U, et al. Rapid course radiation therapy vs. more standard treatment: a randomized trial for bone metastases. Int J Radiat Oncol Biol Phys 1996;36:1085-1089.   90. Gaze MN, Kelly CG, Kerr GR, et al. Pain relief and quality of life following radiotherapy for bone metastases: a randomised trial of two fractionation schedules. Radiother Oncol 1997;45:109116.   91. Ben-Josef E, Shamsa F, Williams AO, et al. Radiotherapeutic management of osseous metastases: a survey of current patterns of care. Int J Radiat Oncol Biol Phys 1998;40:915-921.   92. Rose CM, Kagan AR. The final report of the expert panel for the Radiation Oncology Bone Metastasis Work Group of the American College of Radiology. Int J Radiat Oncol Biol Phys 1998;40:1117-1124.   93. Powers WE, Ratanatharathorn V. Palliation of bone metastases. In Perez CA, Brady LW (eds): Principles and Practice of Radiation Oncology, 3rd ed. Philadelphia: Lippincott Raven, 1998, pp 2199-2219.   94. Shasha D, Harrison LB. The role of brachytherapy for palliation. Semin Radiat Oncol 2000;10:221-239.   95. Bilsky MH, Lis E, Raizer J, et al. The diagnosis and treatment of metastatic spinal tumor. Oncologist 1999;4:459-469.   96. Nieder C, Milas L, Ang KK. Tissue tolerance to reirradiation. Semin Radiat Oncol 2000;10:200-209.   97. Morris DE. Clinical experience with retreatment for palliation. Semin Radiat Oncol 2000;10:210-221.   98. Steiner RM, Mitchell DG, Rao VM, et al. Magnetic resonance imaging of diffuse bone marrow disease. Radiol Clin North Am 1993;31:383-409.   99. Algra PR, Bloem JL, Tissing H, et al. Detection of vertebral ­metastases: comparison between MR imaging and bone scintigraphy. Radiographics 1991;11:219-232. 100. Le Bihan DJ. Differentiation of benign versus pathologic compression fractures with diffusion-weighted MR imaging: a closer step toward the “holy grail” of tissue characterization? Radiology 1998;207:305-307. 101. Albright JA, Gillespie TE. Butaud TR. Treatment of bone metastases. Semin Oncol 1980;7:418-433. 102. Nielsen OS, Munro AJ, Tannock IF. Bone metastases: pathophysiology and management policy. J Clin Oncol 1991;9:509524. 103. Hendrix RW, Rogers LF, Davis TM Jr. Cortical bone metastases. Radiology 1991;181:409-413. 104. Garrett IR. Bone destruction in cancer. Semin Oncol 1993;20 (Suppl 2):4-9. 105. Orr FW, Kostenuik P, Sanchez-Sweatman OH, et al. Mechanisms involved in the metastasis of cancer to bone. Breast Cancer Res Treat 1993;25:151-163. 106. Ford HT, Yarnold JR. Radiation therapy—pain relief and recalcification. In Stoll BA, Parbhoo S (eds): Bone Metastases: Monitoring and Treatment. New York: Raven, 1983, pp 343-354. 107. Mercadante S. Malignant bone pain: pathophysiology and treatment. Pain 1997;69:1-18. 108. Hortobagyi GN, Libshitz HI, Seabold JE. Osseous metastases of breast cancer-clinical, biochemical, radiographic, and scintigraphic evaluation of response to therapy. Cancer 1984;53:577582. 109. Vogel CL, Schoenfelder J, Shemano I, et al. Worsening bone scan in the evaluation of antitumor response during hormonal therapy of breast cancer. J Clin Oncol 1995;13:1123-1128. 110. Pluijm G, Lowik C, Papapoulos S. Tumour progression and angiogenesis in bone metastasis from breast cancer: new approaches to an old problem. Cancer Treat Rev 2000;26:11-27.

1034

PART 14  Special Considerations

111. Conte PF, Latreille J, Mauriac L, et al. Delay in progression of bone metastases in breast cancer patients treated with intravenous pamidronate: results from a multinational randomized controlled trial. J Clin Oncol 1996;14:2552-2559. 112. Holmes RA. Radiopharmaceuticals in clinical trials. Semin ­Oncol 1993;20:22-26. 113. Robinson RG, Preston DF, Baxter KG, et al. Clinical experience with strontium 89 in prostatic and breast cancer patients. Semin Oncol 1993;20:44-48. 114. Robinson RG, Preston DF, Schiefelbein M, et al. Strontium 89 therapy for the palliation of pain due to osseous metastases. JAMA 1995;274:420-424. 115. Krishnamurthy GT, Swailem FM, Srivastava SC, et al. Tin117m(4+)DTPA: pharmacokinetics and imaging characteristics in patients with metastatic bone pain. J Nucl Med 1997;38:230237. 116. Serafini AN, Houston SJ, Resche I, et al. Palliation of pain associated with metastatic bone cancer using samarium-153 lexidronam: a double-blind placebo-controlled clinical trial. J Clin Oncol 1998;16:1574-1581. 117. Rogers CL, Speiser BL, Ram PC, et al. Efficacy and toxicity of intravenous strontium-89 for symptomatic osseous metastases. J Brachytherapy International 1998;14:133-142. 118. Sciuto R, Maini CL, Tofani A, et al. Radiosensitization with low-dose carboplatin enhances pain palliation in radioisotope ­therapy with strontium-89. Nucl Med Commun 1996;17:799804. 119. Bolger JJ, Dearnaley DP, Kirk D, et al. Strontium-89 [Metastron] versus external beam radiotherapy in patient with painful bone metastases secondary to prostatic cancer: preliminary report of a multicenter trial. Semin Oncol 1993;20:32-33. 120. Eary JF, Collins C, Stabin M, et al. Samarium 153-EDTMP biodistribution and dosimetry estimation. J Nucl Med 1993;34:10311036. 121. Bayouth JE, Macey DJ, Kasi LP, et al. Dosimetry and toxicity of samarium-153-EDTMP administered for bone pain due to skeletal metastases. J Nucl Med 1994;35:63-69. 122. McEwan AJ, Amyotte GA, McGowan DG, et al. A retrospective analysis of the cost-effectiveness of treatment with Metastron in patients with prostate cancer metastatic to bone. Eur Urol 1994;26:26-31. 123. Alberts AS, Smit BJ, Louw WKA, et al. Dose response relationship and multiple dose efficacy and toxicity of samarium153-EDTMP in metastatic cancer to bone. Radiother Oncol 1997;43:175-179. 124. Tong D, Gillick L, Hendrickson FR. The palliation of symptomatic osseous metastases: final results of the study by the Radiation Therapy Oncology Group. Cancer 1982;50:893-899. 125. Blitzer PH. Reanalysis of the RTOG study of the palliation of symptomatic osseous metastasis. Cancer 1985;55:1468-1472. 126. Rutten EHJM, Crul BJP, van der Toorn PPG, et al. Pain characteristics help to predict the analgesic efficacy of radiotherapy for the treatment of cancer pain. Pain 1997;69:131-135. 127. Rasmusson B, Vejborg I, Jensen AB, et al. Irradiation of bone metastases in breast cancer patients: a randomized study with 1 year follow-up. Radiother Oncol 1995;34:179-184. 128. Niewald M, Tkocz HJ, Abel U, et al. Rapid course radiation therapy vs. more standard treatment: a randomized trial for bone metastases. Int J Radiat Oncol Biol Phys 1996;36:1085-1089. 129. Dawson R, Currow D, Stevens G, et al. Radiotherapy for bone metastases: a critical appraisal of outcome measures. J Pain Symptom Manage 1999;17:208-218. 130. Chandler SS, Sarin R. Single fraction radiotherapy for bone metastases: are all questions answered? Radiother Oncol 1999;52:191-193. 131. Gillick LS, Goldberg S. Technical Report No 185R: Final Analysis of RTOG Protocol No. 74-02. Boston: Department of Biostatistics, Sidney Farber Cancer Institute, 1981. 132. Jeremic B, Shibamoto Y, Adimovic L, et al. A randomized trial of three single-dose radiation therapy regimens in the treatment of metastatic bone pain. Int J Radiat Oncol Biol Phys 1998;42: 161-167.

133. Nielsen OS. Palliative radiotherapy of bone metastases: there is now evidence for the use of single fractions. Radiother Oncol 1999;52:95-96. 134. Steenland E, Leer JW, van Houwelingen H, et al. The effect of a single fraction compared to multiple fractions on painful bone metastases: a global analysis of the Dutch Bone Metastasis Study. Radiother Oncol 1999;52:101-109. 135. Barton M. Tables of equivalent dose in 2 Gy fractions: a simple application of the linear quadratic formula. Int J Radiat Oncol Biol Phys 1995;31:371-378. 136. Jeremic B, Shibamoto Y, Igrutinovic I. Single 4 Gy re-irradiation for painful bone metastasis following single fraction ­radiotherapy. Radiother Oncol 1999;52:123-127. 137. Nielsen OS, Bentzen SM, Sandberg E, et al. Randomized trial of single dose versus fractionated palliative radiotherapy of bone metastases. Radiother Oncol 1998;47:233-240. 138. Bone Pain Trial Working Party. 8 Gy single fraction radiotherapy for the treatment of metastatic skeletal pain: randomised comparison with a multifraction schedule over 12 months of patient follow-up. Radiother Oncol 1999;52:111-121. 139. Barak F, Werner A, Walach N, et al. The palliative efficacy of a single high dose of radiation in treatment of symptomatic ­osseous metastases. Int J Radiat Oncol Biol Phys 1987;13:12331235. 140. Cole DJ. A randomized trial of a single treatment versus conventional fractionation in the palliative radiotherapy of painful bone metastases. Clin Oncol 1989;1:59-62. 141. Hoskin PJ, Price P, Easton D, et al. A prospective randomised trial of 4 Gy or 8 Gy single doses in the treatment of metastatic bone pain. Radiother Oncol 1992;23:74-78. 142. Price P, Hoskin PJ, Easton D, et al. Prospective randomised trial of single and multifraction radiotherapy schedules in the treatment of painful bone metastases. Radiother Oncol 1986;6: 247-255. 143. Madsen EL. Painful bone metastasis: efficacy of radiotherapy assessed by the patients: a randomized trial comparing 4 Gy × 6 versus 10 Gy × 2. Int J Radiat Oncol Biol Phys 1983;9: 1775-1779. 144. Boogerd W, van der Sande JJ. Diagnosis and treatment of spinal cord compression in malignant disease. Cancer Treat Rev 1993;19:129-150. 145. Byrne TN. Spinal cord compression from epidural metastases. New Engl J Med 1992;327:614-619. 146. Grant R, Papadopoulos SM, Greenberg HS. Metastatic epidural spinal cord compression. Neurol Clin 1991;9:825-841. 147. Maranzano E, Latini P, Checcaglini F, et al. Radiation therapy in metastatic spinal cord compression—a prospective analysis of 105 consecutive patients. Cancer 1991;67:1311-1317. 148. Janjan NA. Radiotherapeutic management of spinal metastases. J Pain Symptom Manage 1996;1:47-56. 149. Loblaw DA, Laperriere NJ. Emergency treatment of malignant extradural spinal cord compression: an evidence-based guideline. J Clin Oncol 1998;16:1613-1624. 150. Boogerd W. Central nervous system metastasis in breast cancer. Radiother Oncol 1996;40:5-22. 151. Paterson AHG. Bone metastases in breast cancer, prostate cancer and myeloma. Bone 1987;8(Suppl 1):17-22. 152. Bunting RW, Boublik M, Blevins FT, et al. Functional outcome of pathologic fracture secondary to malignant disease in a rehabilitation hospital. Cancer 1992;69:98-102. 153. Townsend PW, Smalley SR, Cozad SC, et al. Role of postoperative radiation therapy after stabilization of fractures caused by metastatic disease. Int J Radiat Oncol Biol Phys 1995;31: 43-49. 154. Heisterberg L, Johansen TS. Treatment of pathologic fractures. Acta Orthop Scand 1979;50:787-790. 155. Fidler M. Incidence of fracture through metastases in long bones. Acta Orthop Scand 1981;52:623-627. 156. Oda MAS, Schurman DJ. Monitoring of pathological fracture. In Stoll BA, Parbhoo S (eds): Bone Metastases: Monitoring and Treatment. New York: Raven, 1983, pp 271-288. 157. Quinn JA, De Angelis LM. Neurologic emergencies in the cancer patient. Semin Oncol 2000;27:311-321.

39  Palliative Care 158. Boogerd W, van der Sande JJ, Kroger R. Early diagnosis and treatment of spinal metastases in breast cancer: a prospective study. J Neurol Neurosurg Psychiatry 1992;55:1188-1193. 159. Turner S, Marosszeky B, Timms I, et al. Malignant spinal cord compression: a prospective evaluation. Int J Radiat Oncol Biol Phys 1993;26:141-146. 160. Peteet J, Tay V, Cohen G, MacIntyre J. Pain characteristics and treatment in an outpatient cancer population. Cancer 1985;57:1259-1265. 161. Wada E, Yamamoto T, Furuno M, et al. Spinal cord compression secondary to osteoblastic metastasis. Spine 1993;18:1380-1381. 162. Algra PR, Heimans JJ, Valk J, et al. Do metastases in vertebrae begin in the body or the pedicles?—imaging study in 45 patients. AJR Am J Roentgenol 1992;158:1275-1279. 163. Landmann C, Hunig R, Gratzi O. The role of laminectomy in the combined treatment of metastatic spinal cord compression. Int J Radiat Oncol Biol Phys 1992;24:627-631. 164. Kim RY, Smith JW, Spencer SA, et al. Malignant epidural spinal cord compression associated with a paravertebral mass: its radiotherapeutic outcome on radiosensitivity. Int J Radiat Oncol Biol Phys 1993;27:1079-1083. 165. Bach F, Agerlin N, Sorensen JB, et al. Metastatic spinal cord compression secondary to lung cancer. J Clin Oncol 1992;10:17811787. 166. Schiff D, O’Neill BP, Wang CH, et al. Neuroimaging and treatment implications of patients with multiple epidural spinal ­metastases. Cancer 1998;83:1593-1601. 167. Saip P, Tenekeci N, Aydiner A, et al. Response evaluation of bone metastases in breast cancer: value of magnetic resonance imaging. Cancer Invest 1999;17:575-580. 168. Cowap J, Hardy JR, A’Hern R. Outcome of malignant spinal cord compression at a cancer center: implications for palliative care services. J Pain Symptom Manage 2000;19:257-264. 169. Maranzano E, Latini P, Perrucci E, et al. Short-course radiotherapy [8 Gy × 2] in metastatic spinal cord compression: an effective and feasible treatment. Int J Radiat Oncol Biol Phys 1997;38:1037-1044. 170. Jeremic B, Djuric L, Mijatovic L. Incidence of radiation myelitis of the cervical spinal cord at doses of 5500 cGy or greater. Cancer 1991;68:2138-2141. 171. Wen PY, Blanchard KL, Block CC, et al. Development of Lhermitte’s sign after bone marrow transplantation. Cancer 1992;69:2262-2266. 172. Powers BE, Thames HD, Gillette SM, et al. Volume effects in the irradiated canine spinal cord: do they exist when the probability of injury is low? Radiother Oncol 1998;46:297-306. 173. Fowler JF, Bentzen SM, Bond SJ, et al. Clinical radiation doses for spinal cord: the 1998 international questionnaire. Radiother Oncol 2000;55:295-300. 174. Sugimura H, Kisanuki A, Tamura S, et al. Magnetic resonance imaging of bone marrow changes after irradiation. Invest Radiol 1994;29:35-41. 175. Yankelevitz DF, Henschke C, Knapp PH, et al. Effect of radiation therapy on thoracic and lumbar bone marrow: evaluation with MR imaging. Am J Roentgenol 1991;157:87-92. 176. Shapiro CL, Keating J, Angell JE, et al. Monitoring therapeutic response in skeletal metastases using dual-energy x-ray absorptiometry: a prospective feasibility study in breast cancer patients. Cancer Invest 1999;17:566-574. 177. Bach F, Bjerregaard B, Soletormos G, et al. Diagnostic value of cerebrospinal fluid cytology in comparison with tumor marker activity in central nervous system metastases secondary to breast cancer. Cancer 1993;72:2376-2382. 178. Russi EG, Pergolizzi S, Gaeta M, et al. Palliative radiotherapy in lumbosacral carcinomatous neuropathy. Radiother Oncol 1993;26:172-173. 179. Salazar OM, Rubin P, Hendrickson FR, et al. Single-dose halfbody irradiation for palliation of multiple bone metastases from solid tumors—final Radiation Therapy Oncology Group Report. Cancer 1986;58:29-36.

1035

180. Poulter CA, Cosmatos D, Rubin P, et al. A report of RTOG 8206: A phase III study of whether the addition of single dose hemibody irradiation to standard fractionated local field irradiation is more effective than local field irradiation alone in the treatment of symptomatic osseous metastases. Int J Radiat Oncol Biol Phys 1992;23:207-214. 181. Barton R, Kirkbride P. Special techniques in palliative radiation oncology. J Palliat Med 2000;3:75-83. 182. Friedland J. Local and systemic radiation for palliation of metastatic disease. Urol Clin North Am 1999;26:391-402. 183. Silberstein EB. Systemic radiopharmaceutical therapy of painful osteoblastic metastases. Semin Radiat Oncol 2000;10:240-249. 184. Maxon HR, Deutsch EA, Thomas SR, et al. Re-186(Sn) HEDP for treatment of multiple metastatic foci in bone: human biodistribution and dosimetric studies. Radiology 1988;166:501-507. 185. Kagan AR. Palliation of brain and spinal cord metastases. In ­Perez CA, Brady LW (eds): Principles and Practice of Radiation Oncology, 3rd ed. Philadelphia: Lippincott Raven, 1998, pp 2187-2198. 186. Shaw E, Scott C, Souhami L, et al. Single dose radiosurgical treatment of recurrent previously irradiated primary brain tumors and brain metastases: final report of RTOG Protocol 9005. Int J Radiat Oncol Biol Phys 2000;47:291-298. 187. Murray KJ, Scott C, Zachariah B, et al. Importance of the minimental status examination in the treatment of patients with brain metastases: a report from the Radiation Therapy Oncology Group Protocol 91-04. Int J Radiat Oncol Biol Phys 2000;48:59-64. 188. Kagan AR. Palliation of visceeral recurrences and metastases. In Perez CA, Brady LW (eds): Principles and Practice of Radiation Oncology, 3rd ed. Philadelphia: Lippincott-Raven, 1998, pp 2219-2226. 189. Langenduk JA, ten Velde GPM, Aaronson NK, et al. Quality of life after palliative radiotherapy in non-small cell lung cancer: a prospective study. Int J Radiat Oncol Biol Phys 2000;47:149-155. 190. Gressen EL, Werner-Waski M, Cohn J, et al. Thoracic reirradiation for symptomatic relief after prior radiotherapeutic management for lung cancer. Am J Clin Oncol 2000;23:160-163. 191. Ripamonti C. Management of dyspnea in advanced cancer ­patients. Support Care Cancer 1999;7:233-243. 192. List MA, Siston A, Haraf D, et al. Quality of life and performance in advanced head and neck cancer patients on concomitant chemoradiotherapy: a prospective examination. J Clin Oncol 1999;17:1020-1028. 193. Janjan NA, Weissman DE, Pahule A. Improved pain management during radiation therapy for head and neck carcinoma. Int J Radiat Oncol Biol Phys 1992;23:647-652. 194. Lee AWM, Foo W, Law SCK, et al. Total biological effect on late reactive tissues following reirradiation for recurrent nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2000;46:865-872. 195. Mancini I, Bruera E. Constipation in advanced cancer patients. Support Care Cancer 1998;6:356-364. 196. Rousseau P. Management of malignant bowel obstruction in ­advanced cancer: a brief review. J Palliative Med 1998;1:65-72. 197. Gagnon B, Mancini I, Pereira J, et al. Palliative management of bleeding events in advanced cancer patients. J Palliative Care 1998;14:50-54. 198. Crane CH, Janjan NA, Abbruzzese JL, et al. Effective pelvic symptom control using initial chemoradiation without colostomy in metastatic rectal cancer. Int J Radiat Oncol Biol Phys 2001;49:107-116. 199. Janjan NA, Breslin T, Lenzi R, et al. Avoidance of colostomy placement in advanced colorectal cancer with twice weekly ­ hypofractionated radiation plus continuous infusion 5-fluorouracil. J Pain Symptom Manage 2000;20:266-272. 200. Duchesne GM, Bolger JJ, Griffiths GO, et al. A randomized trial of hypofractionated schedules of palliative radiotherapy in the management of bladder carcinoma: results of medical research council trial BA09. Int J Radiat Oncol Biol Phys 2000;47:379-388. 201. Onsrud M, Hagen B, Strickert T, 10-Gy single-fraction pelvic irradiation for palliation and life prolongation in patients with cancer of the cervix and corpus uteri. Gynecol Oncol 2001;82:167-171.

40 Proton Therapy James D. Cox SETTING THE STAGE FOR PROTON THERAPY: DEVELOPMENT AND EVOLUTION OF CONFORMAL RADIATION THERAPY Three-Dimensional Conformal Radiation Therapy versus Two-Dimensional Radiation Therapy Intensity-Modulated Radiation Therapy versus Three-Dimensional Conformal Radiation Therapy PROTON BEAM THERAPY Biologic Effects Dosimetric Advantages

Protons were discovered in 1919 by Ernest Rutherford.1 Unlike x-rays, which were being used for therapeutic purposes within a few years of their discovery by Roentgen, the potential for therapy with protons was first suggested by Robert Wilson nearly 3 decades after their discovery.2 Cornelius Tobias began treating patients with protons at the Lawrence Berkeley National Laboratory in 1954, but after 30 patients had been so treated, the focus at that facility shifted to helium ions (alpha particles),3 which were easier to produce with the accelerator there.4 In 1961, Kjellberg and colleagues treated the first patient at the Harvard Cyclotron Laboratory,5 where physicists, engineers, computer scientists, dosimetrists, and physicians made many important advances in proton therapy over the next 40 years. When the Harvard cyclotron facility was closed in 2002, 9116 patients had undergone proton therapy there. By the mid-1960s, many research facilities devoted to nuclear physics around the world had begun to allow some portion of the time available with their accelerators, mostly cyclotrons, to be used for therapy with protons, helium, and heavy ions, primarily carbon (12C). Because the ­ primary activity of these facilities was fundamental ­research in nuclear physics, the availability of the beams for therapeutic purposes was limited. Very few patients could be treated during the brief periods when the beam was available, and uncommon diseases were selected for proton beam therapy. Most of the more than 40,000 ­patients who received proton beam therapy were treated for uveal ­ melanoma, an uncommon ocular tumor occurring almost entirely in whites. Nevertheless, this experience clearly ­demonstrated the safety and efficacy of proton beam ­ therapy. The ­ limited uses of proton beam therapy that were necessitated by the beam energies, field sizes, and beam availability in the ­research facilities arguably retarded the clinical ­development of that therapy. In 1990, the first hospital-based proton facility was opened at Loma Linda University Medical Center. More than 10,000 patients have been treated at the Loma Linda

1036

CLINICAL EXPERIENCE WITH PROTON THERAPY Ocular Melanoma Central Nervous System Tumors in Adults Skull Base and Cervical Spine Tumors Cancer in Children Head and Neck Tumors Non–Small Cell Lung Cancer Esophageal Cancer Prostate Cancer PROTONS IN COMBINED-MODALITY THERAPIES CHALLENGES FOR PROTON THERAPY

facility, most of whom have been men with adenocarcinoma of the prostate. In 2001, Massachusetts General Hospital opened the second hospital-based facility, the Northeast Proton Therapy Center, drawing on the large experience gained at the Harvard Cyclotron Laboratory.6 The main components of hospital-based proton therapy systems are an accelerator (cyclotron or synchrotron) to generate the proton beam, a transport system to carry the beam from the accelerator to the patient, and a delivery system to control and shape the beam to the desired threedimensional specifications and deliver it to the tumor site in the patient. Proton treatment systems have been approved by the U.S. Food and Drug Administration. Considering the evolution of radiation oncology practice over the past 2 decades is valuable for putting proton beam therapy into context. The lessons learned, especially in the development of the evidence base for new technologies, are germane to the current state of proton beam therapy.

Setting the Stage for Proton Therapy: Development and Evolution of Conformal Radiation Therapy Advances in diagnostic imaging in the 1980s provided the basis for tailoring radiation therapy to the specific ­anatomy of individual patients rather than to anatomic atlases. Computed tomography and other tomographic scanners enabled three-dimensional display of tumors relative to surrounding normal anatomy. The development of ­accurate threedimensional radiation dose computations and ­ multileaf collimators in linear accelerators permitted a beam’s eye view of tumors and the conformal delivery of radiation to those tumors. Commercial treatment planning systems facilitated the widespread use of three-dimensional conformal radiation therapy (CRT) by the mid-1990s. Computer simulations of dose distributions demonstrated that

40  Proton Therapy

these systems could deliver higher total doses to the gross tumor volume7 than was possible with two-dimensional treatment planning, and normal tissues could be avoided or at least exposed to much lower doses. The rapid adoption of three-dimensional CRT was based entirely on ­computer-generated treatment plans that showed reductions in the volume of normal tissue irradiated with this method compared with two-dimensional planning and delivery systems. Findings from formal prospective clinical trials were reported a few years after three-dimensional CRT was ­already in wide use.8,9 Some of those findings are described in the following sections. Delivery of small beams of x-rays with different intensities (i.e., intensity-modulated radiation therapy [IMRT]) permitted further shaping of the high-dose volume. As physicists developed ways of optimizing the different intensities and through the use of dynamic multileaf collimators, IMRT came to be fully realized. In contrast to the rather rapid adoption of three-dimensional CRT, IMRT was appreciated and introduced more slowly.10 Physicians, physicists, and dosimetrists had to devote much more time and effort to plan patient treatments with three-dimensional CRT and especially with IMRT,11,12 but the benefit of reduced toxicity was and is a worthy goal.13 Such precision in the delivery of radiation requires careful target delineation, treatment planning, and quality assurance. Moreover, the risk of missing tumors that may move during treatment (e.g., with respiration) or between daily doses requires that imaging be conducted at each treatment session. The term image-guided radiation therapy can refer to imaging at each session and to imaging conducted in the course of planning and delivering three-dimensional CRT and IMRT. Findings from formal trials comparing two-dimensional radiation therapy with three-dimensional conformal therapy are reviewed in the next two sections.

Three-Dimensional Conformal Radiation Therapy versus Two-Dimensional Radiation Therapy In an early investigation of three-dimensional CRT for prostate cancer, Pollack and colleagues9 reported the results of a phase III trial testing dose escalation with two-­dimensional or three-dimensional CRT for men with adenocarcinoma of the prostate. Clinical and biochemical evidence of disease progression were used as end points. This study was not a pure comparison of two-dimensional and three-­dimensional CRT because both groups of patients had been treated initially with a traditional two-­dimensional, four-field box technique to the pelvis to a total dose of 46 Gy in 23 fractions. From that point onward, the standard-dose group ­received another 24 Gy (total dose of 70 Gy) with the box configuration with reduced fields, and the dose-­escalation group received a six-field, three-dimensional CRT boost to 78 Gy; all doses were given at 2 Gy per fraction. The higher total dose resulted in a statistically significant increase in freedom from biochemical disease progression (defined as a drop followed by a stable level of circulating prostatespecific antigen [PSA]). A second report of this study14 confirmed the higher rate of progression-free survival, but the effects on bowel and bladder were not decreased in the dose-escalation group, undoubtedly because both groups had initially been treated with the four-field technique to the whole pelvis. These ­ investigators further identified a

1037

critical volume of rectum as the determinant of toxicity in that randomized trial.15 In another randomized trial, Dearnaley and collegues8 compared three-dimensional CRT with two-dimensional radiation for carcinoma of the prostate. This was a more definitive comparison that focused on reducing the side ­effects of radiation; tumor control was expected to be similar because the total doses were the same in both treatment groups. The results showed that three-dimensional CRT produced significantly fewer acute and late effects on bowel and bladder. However, the total dose given in that trial was only 64 Gy, which is well below the total dose required for control of prostate carcinoma.16,17 In contrast, findings from the Radiation Therapy Oncology Group (RTOG) protocol 94-0618,19 indicated that a total dose of 79.2 Gy at 1.8 Gy per fraction was well tolerated and produced grade 3 bowel and bladder toxicity rates of approximately 3%, rates well below the 12% to 17% reported in earlier RTOG studies of two-dimensional treatments. The only other study to unequivocally demonstrate improved outcome with three-dimensional CRT over twodimensional treatment was reported in 2006 by Fang and colleagues.20 In that study, patients with stage I non–small cell lung cancer who could not undergo surgery (usually because of cardiac or pulmonary disease) were treated with radiation therapy alone. Local tumor control and survival were better among those who received three­dimensional CRT than in a historical group treated with two-­dimensional techniques. Distant metastasis occurred with equal frequency in both groups. Another study that spanned the transition from twodimensional to three-dimensional therapy was a phase I trial that permitted comparison of three-dimensional CRT with two-dimensional techniques for preventing acute toxicity when given in combination with gemcitabine for non–small cell lung carcinoma.21 Gemcitabine was given weekly during the radiation therapy (63 Gy in 35 fractions over 7 weeks). The dose-limiting toxicity—grade 3 or higher esophageal reactions—was much less common in the three-dimensional group, and higher doses of gemcitabine were possible with the three-dimensional CRT. The volume of esophagus that received 60 Gy or more was only 11% in the three-dimensional group as compared with 71% in the two-dimensional group.21

Intensity-Modulated Radiation Therapy versus Three-Dimensional Conformal Radiation Therapy Although comparative dosimetry favors IMRT over threedimensional CRT, relatively few studies have shown improvement in clinical outcomes from the use of IMRT. Available data for carcinoma of the prostate, head and neck, and breast treated with IMRT have been summarized by Bucci and colleagues.22 In a long-term study in which carcinoma of the prostate was treated to total doses up to 82 Gy with IMRT, Zelefsky and others23 reported that grade 3 or greater toxicity was experienced by less than 1% of these patients. Yom and colleagues24 also reported a highly significant reduction in the occurrence of treatment-related pneumonitis with IMRT compared with three-dimensional CRT for non–small cell lung cancer, but the follow-up times for that study were relatively short: 8 months for the IMRT group and 9 months for the three-dimensional CRT group.

1038

PART 14  Special Considerations

Proton Beam Therapy The observed improvement in tumor control and diminished toxicity from three-dimensional CRT or IMRT compared with two-dimensional techniques set the scene for the consideration of proton beam therapy in two ways. First, the advantages of three-dimensional CRT indicated by comparative dosimetry led to adoption of three-­dimensional CRT by the radiation oncology community well before clinical trials or retrospective studies had demonstrated clinical benefit. Second, the same measure of comparative dosimetry has consistently shown an ­advantage for proton beam therapy over three-dimensional techniques in terms of avoidance of normal tissues.

Biologic Effects More than a century of experience with x-rays and gamma rays provides the basis for understanding and predicting the biologic effects of protons. In terms of relative biologic effectiveness (RBE, a concept discussed further in Chapter 1), protons have almost the same RBE value as photons. Summarizing available data from numerous experiments with protons, Paganetti25 concluded that the RBE of protons is approximately 1.1. The RBE for carbon ions is approximately 3, similar to the RBE for neutrons. Higher RBE values may be advantageous in terms of tumor control, but they are disadvantageous in terms of damaging normal tissues.

Dosimetric Advantages The primary advantage of proton beam therapy results from its dose distribution. Unlike photons, for which the maximum dose always occurs at or near the point of entry into biologic systems (e.g., the skin), protons deposit their energy abruptly as the proton velocity decreases. Depth dose distributions for protons are characterized by a Bragg peak that occurs at a depth that depends on the energy of the incident proton beam (Fig. 40-1). The clinical ramifications of this are profound. Tissues near the surface of entry do not receive the maximum dose, as is the case with photons, and there is no exposure to radiation beyond the sharp fall of the Bragg peak to zero dose. Because the Bragg peak can be spread out to ­encompass

the clinical target volume by means of range modulator wheels (Fig. 40-2) and the high-dose volume can be shaped with brass apertures (in the x and y dimensions) (Fig. 40-3A) and with plastic (see Fig. 40-3B) (in the z dimension, the dimension of the spread-out Bragg peak), it is possible to deliver the desired dose to the tumor and to avoid almost all normal tissues, especially those distal to the tumor.

Clinical Experience with Proton Therapy Several diseases have been treated with proton beam therapy for many years at various physics research facilities. Outcome data from the treatment of these diseases are mature, and the demonstration of the value of proton beam therapy for their treatment is widely appreciated. These experiences are described briefly in the following sections.

Ocular Melanoma Malignant tumors of the eye and orbit are uncommon. They can usually be cured by enucleation or orbital exenteration, with the loss of function and esthetics. Radiation therapy with the intent of preserving the eye was first attempted in 1929.26 Choroidal (uveal) melanoma is the most common primary intraocular malignancy, and it can be lethal. Eighty percent of uveal melanomas occur in the posterior choroid; the remainder involve the ciliary body or the iris.27 The most significant predictor of metastasis in ocular melanomas is the size of the lesion at diagnosis.28 Enucleation, long the mainstay of treatment for ocular malignancies, seems to offer no survival advantage over procedures that conserve the globe. The two most common approaches to controlling melanoma while preserving the eye are the use of radioactive plaques, most often incorporating iodine 125 (125I), and charged particle therapy. A Collaborative Ocular Melanoma Study of 1300 patients showed that 125I plaque therapy resulted in 5-year29 and 10-year28 survival rates equivalent to enucleation. The only trial comparing 125I with charged particle therapy, conducted by Char and colleagues,30 involved a total dose of 70 cobalt

PRISTINE BRAGG PEAK, 250 MeV 120.0

Percent depth dose

100.0 80.0 60.0 40.0 20.0 0.0 0

50

100

150

200

250

300

350

Depth (mm)

FIGURE 40-1.  Distribution of dose deposition from 250-MeV protons. The red line indicates the pristine Bragg peak; the blue line, the spread-out Bragg peak.

40  Proton Therapy

Gray equivalents (CGE) delivered over 7 to 11 days continuously with brachytherapy (98 patients) or in 5 fractions with helium ions (86 patients). The local recurrence rate in the brachytherapy group was 13% compared with 0% in the helium ion group. Enucleations were required more often after brachytherapy than after helium ion treatment.

1039

Proton therapy for ocular tumors was studied extensively at the Harvard Cyclotron Laboratory by investigators from the Massachusetts General Hospital and the ­Massachusetts Eye and Ear Infirmary, with more than 3000 uveal melanomas treated by the end of 2002.6 In that experience, the 5-year actuarial local control rate was 96%, and the

SPREAD-OUT BRAGG PEAK FROM PRISTINE BRAGG PEAKS 120.0 100.0

PDD

80.0 60.0 40.0 20.0 0.0 0

50

100

150

200

250

300

350

Depth (mm)

FIGURE 40-2.  Multiple Bragg peaks contribute to the spread-out Bragg peak. Multiple Bragg peaks are generated by range modulator wheels, as shown in the two photographs. PDD, percent depth dose.

A

B

FIGURE 40-3.  A, Brass aperture to be inserted in the proton beam to shape the dose in the x and y dimensions. B, Plastic compensator to be inserted in the proton beam to absorb protons and thereby shape the dose in the z dimension, which is the direction of the spread-out Bragg peak.

1040

PART 14  Special Considerations

­ robability of eye retention at 5 years was 90%. At 15 years, p the local control rate was 95%, with eye retention in 84% of patients.31 Investigators from the Paul Scherrer Institut in Switzerland reported eye-conservation rates after proton therapy of 89% at 5 years, 86% at 10 years, and 83% at 15 years among the 2645 patients so treated.32

Central Nervous System Tumors in Adults This section describes tumors of the central nervous system (CNS) in adults; CNS tumors in children are discussed separately in the section “Cancer in Children.” Malignant gliomas are best controlled with maximal tumor resection followed by postoperative radiation. However, this sequence is rarely curative because of the invasive nature of these tumors and their inherent resistance to radiation; virtually all malignant gliomas recur within 5 years. Investigators at the Harvard Cyclotron Laboratory escalated total doses to such tumors to 90 CGE, 50% higher than the total dose that is commonly accepted as tolerable with photons alone. Although the normal brain could be spared better with protons than photons, even these high doses did not result in control of the disease because of disease recurrences, some at considerable distances from the margins of the irradiated volume.33 Temozolomide has been shown to extend survival in patients with malignant gliomas,34 but no data have been published regarding the combination of proton radiation and temozolomide. Meningiomas are relatively common tumors of the CNS. They are usually benign and can be resected with ­excellent results. Radiation therapy is indicated only for those ­tumors that cannot be completely removed or those that recur ­after resection. Investigators at the Harvard ­ Cyclotron ­Laboratory treated 46 patients with such ­ meningiomas by using a combination of photons and protons to a median dose to the gross tumor volume of 59 CGE. Five-year and 10-year recurrence-free rates were 100% and 88%, ­respectively. Eight of the 46 patients developed severe late ­effects of treatment, including ophthalmologic toxicity in 4 patients and neurologic effects in 4 patients. All patients with late ophthalmologic effects had received a total dose in excess of 54 CGE, which is now considered the maximum tolerable dose for the optic nerves and chiasm.35 Proton therapy has been used to treat pituitary adenomas.36 Kjellberg treated almost 3000 patients at the Harvard Cyclotron Laboratory between 1961 and 1993, most of whom had pituitary adenomas, arteriovenous malformations, or vestibular schwannomas.5 Harsh and colleagues37 reported a study of stereotactic irradiation of vestibular schwannomas (acoustic neuromas) conducted between 1992 and 1998. In that study, 68 patients received a single dose of 12 CGE to the 70% isodose line (maximum dose, 17 CGE) for lesions that showed progression on sequential imaging or were causing clinical deterioration. Actuarial estimates of freedom from tumor progression were 94% at 2 years and 84% at 5 years; the average period of observation with imaging was 34 months. In another study, Bush and colleagues38 used fractionated proton radiation to treat 30 patients with acoustic neuromas between 1991 and 1999; patients with useful hearing received a total dose of 54 CGE in 30 fractions, and those without useful hearing received 60 CGE in 30 to 33 fractions. Magnetic resonance imaging revealed no cases of tumor growth and 11 cases

of tumor regression after treatment. All 13 patients with useful hearing maintained it. No injuries to cranial nerves V or VII occurred. Sparing all normal tissues in treatment of CNS lesions is critically important. Most tumors require total doses of 54 CGE to 60 CGE, which is very near the tissue tolerance of the normal brain, and even doses lower than these can affect cognitive function. When radiation therapy is indicated for CNS tumors, proton beam radiation is without compare.

Skull Base and Cervical Spine Tumors Tumors that arise in the bones or soft tissues at the base of the skull are extraordinarily difficult to treat because of the many critical structures within or adjacent to them. The tumors are poorly responsive to radiation therapy or chemotherapy. Surgical resection is the mainstay of treatment whenever possible. Procedures undertaken by teams of neurosurgeons and head and neck surgeons increase the probability of gross macroscopic removal. However, the inability to adhere to the oncologic principle of removal without transecting the tumor ensures a high likelihood of recurrence under the best of circumstances. Chondrosarcomas have long been recognized as controllable with radiation therapy, although their regression is slow and incomplete. Munzenrider and Liebsch39 reported the results from 229 patients with skull base chondrosarcoma treated with proton therapy at Massachusetts General Hospital between 1975 and 1998. Actuarial local control rates were 98% at 5 years and 94% at 10 years. Doses in that study ranged from 66 to 83 CGE (median, 68 CGE), given in fractions of 1.8 GY or 1.92 CGE. A dose-response relationship was found such that the risk of recurrence was greater if the total dose was less than 67 CGE39; a dosevolume effect was also observed in which high target doses to the pituitary (�70 CGE) or hypothalamus (�50 CGE) were predictive of endocrinopathies, particularly of the anterior pituitary.40 Results of proton therapy for chondrosarcomas of the base of the skull at Loma Linda University Medical Center between 1992 and 1998, reported by Hug and colleagues,41 showed that local control was achieved in 23 of 25 patients at a mean follow-up time of 33 months. In that study, 1.8-CGE fractions were used, and total doses ranged from 64.8 CGE to 79.2 CGE (mean, 70.7 CGE). Noel and colleagues from the Centre de Protonthérapie d’Orsay42 described achieving local control in 9 of 11 patients observed for a mean interval of 30.5 months. The median total dose in that study was 67 CGE (range, 60 to 70 CGE) to the gross tumor volume. Chondrosarcomas of the cervical spine have proved more difficult to control than those of the skull base; Munzenrider and Liebsch39 reported a local control rate of 54% at 5 years and at 10 years for 17 patients. Among skull base tumors, chordomas are much more difficult to control than chondrosarcomas. Munzenrider and Liebsch39 reported actuarial local control rates of 73% at 5 years and 54% at 10 years for 169 patients. The outlook for females with chordomas is poorer than that for males; local control rates at 5 and 10 years for males were 81% and 65%, whereas those for females were 65% and 42%, respectively. Chordomas involving the cervical spine are also difficult to control, with local control rates in one study of 85 patients of only 69% at 5 years and 48% at

40  Proton Therapy

10 years, but no differences were found in control rates between males and females. Weber and colleagues43 at the Paul Scherrer Institut, reporting the results of proton therapy with spot-scanning for skull base tumors, found control rates at 3 years of 87.5% for chordoma and 100% for chondrosarcoma. Irradiation of skull base and cervical spine tumors carries a significant risk of adverse effects, even with proton therapy. In the Massachusetts General Hospital experience, injury to the brainstem and temporal lobes was observed in 8% of patients at 5 years and 13% at 10 years. Optic neuropathy was observed in 4%, hearing loss was found in almost one half of the patients evaluated, and cranial nerve injury was identified in 15 of 27 patients studied.39 With regard to endocrinopathies, 10-year actuarial rates were 84% for hyperprolactinemia, 63% for hypothyroidism, 36% for hypogonadism, and 28% for ­hypoadrenalism.40

Cancer in Children Proton beam therapy shows striking benefit for treating malignant tumors in children. Children usually have growing bone and muscle, and exposure to ionizing radiation is far more damaging for them than for adults. Genetic susceptibility to cancer in terms of secondary tumors appearing long after initial successful treatment is also a matter of concern for children. Total-body irradiation resulting from leakage from linear accelerators may further contribute to the risk of secondary cancer.44

3D CRT, electrons

1041

St. Clair and colleagues from Massachusetts General Hospital45 described significant decreases in doses to critical structures, which are expected to translate into decreased incidence in long-term toxicity, from the use of proton therapy. This group compared three-dimensional CRT, IMRT, and proton therapy for craniospinal irradiation that could be used for medulloblastoma and showed considerable sparing of normal tissues. Structures anterior to the vertebral bodies in particular were almost completely avoided with posterior proton beams, and the dose to the cochlea could be reduced to less than 5%. Lee and others46 performed similar comparisons and corroborated these findings (Fig. 40-4). They also showed considerable sparing of normal tissues during irradiation for retinoblastoma (Fig. 40-5)46 and pelvic sarcoma (Fig. 40-6).46 These studies showed that IMRT was superior to three-dimensional CRT, but proton beam therapy was the best of the three approaches. Clinical experience supports the advantages of proton beam therapy predicted by these dosimetric comparisons. Investigators at the Loma Linda University ­Medical Center reported reduced acute toxicity among three children treated with craniospinal proton beam therapy for medulloblastoma.47 Others from the same institution used ­proton beam therapy for 27 children with low-grade ­astrocytomas that had recurred after partial resection or that were considered unresectable.48 At a mean follow-up time of 3.3 years (range, 0 to 6.8 years), brainstem tumors were controlled in 60% of patients, hemispheric tumors were controlled in 71%, and central (diencephalic) tumors were controlled in

3D CRT, photons

Proton therapy

FIGURE 40-4.  Dosimetric comparisons of three techniques for craniospinal irradiation for medulloblastoma in a child. Sagittal slices are shown for plans generated for three-dimensional conformal radiation therapy (3D CRT) with electrons, 3D CRT with photons, or proton therapy. Isodose lines for the electron and photon therapy plans are as follows: green, 54 Gy; orange, 35 Gy; blue, 20 Gy; red, 15 Gy; yellow, 10 Gy; and purple, 5 Gy. Isodose lines for the proton therapy plan are as follows: orange, 54 Gy; green, 35 Gy; light blue, 20 Gy; aqua/green, 15 Gy; violet, 10 Gy; dark purple, 5 Gy. (From Lee CT, Bilton SD, Famiglietti RM, et al. Treatment planning with protons for pediatric retinoblastoma, medulloblastoma, and pelvic sarcoma: how do protons compare with other conformal techniques? Int J Radiat Oncol Biol Phys 2005;63:362-372.)

1042

PART 14  Special Considerations

A

B

C

D

E

F

FIGURE 40-5.  Dosimetric comparisons of three techniques for treatment of retinoblastoma in a child. Axial slices are shown for plans generated for three-dimensional conformal radiation therapy (3D CRT) with electrons, single appositional field (A); intensitymodulated radiation therapy (B); proton therapy (C); 3D CRT with a lateral beam (D); 3D CRT with anterior and lateral beam and lens block (E); and 3D CRT with anterior and lateral beam without a lens block (F). Isodose lines for the 3D CRT and intensity-modulated radiation therapy plans (A, B, D, E, F) are as follows: purple, 36 Gy; red, 20 Gy; orange, 10 Gy; yellow, 5 Gy. Isodose lines for the proton therapy plan (C) are as follows: red, 36 Gy; green, 20 Gy; blue, 10 Gy; and dark blue, 5 Gy. (From Lee CT, Bilton SD, Famiglietti RM, et al. Treatment planning with protons for pediatric retinoblastoma, medulloblastoma, and pelvic sarcoma: how do protons compare with other conformal techniques? Int J Radiat Oncol Biol Phys 2005;63:362-372.)

87%. Additional findings are awaited from further clinical investigations being conducted at the Loma Linda, Massachusetts General Hospital, Orsay, and Paul Scherrer facilities. The U.S. National Cancer Institute has also approved the use of proton beam therapy for children being enrolled in prospective trials conducted by the Children’s Oncology Group.

Head and Neck Tumors The greatest opportunities for benefit from proton therapy for tumors arising in the head and neck are for tumors at the base of the skull. Treatment considerations for tumors in these locations are similar to those found for

c­ hondrosarcomas and chordomas—critical neural pathways lie close to tumors that require high total doses for control.

Carcinoma of the Paranasal Sinuses Primary malignant neoplasms of the paranasal sinuses are uncommon in the United States. Local recurrence remains the major cause of death for patients with carcinoma of the paranasal sinuses.49 Extensive surgical resection, although possibly improving local control, may require removal of the eye and produce adverse esthetic consequences. Postoperative radiation improves outcome over the results of surgery or radiation therapy alone.50 Nevertheless, the morbidity

40  Proton Therapy

A A

L

1043

normal tissues, particularly the optic chiasm and frontal lobes. Conventional therapies for paranasal sinus tumors involve suboptimal but tolerated levels of radiation delivered before or after surgery to complement often compromised surgical resections. A trial of hyperfractionated, accelerated proton beam radiation therapy is being conducted at Massachusetts ­General Hospital and Loma Linda University Medical Center in an attempt to spare patients the morbidity of ­extensive surgical resections, which are often inadequate, and the complications of the aggressive irradiation ­ necessary to cure these tumors. The research question posed in this phase I/II study is whether increasing the dose to 76 CGE for gross residual disease and to 68.4 CGE for high-risk microscopic disease will improve local control and survival with acceptable morbidity. Radiation is delivered in accelerated fractionation, incorporating computed tomography–based, three-dimensional planned conformal proton/photon radiation therapy and a daily stereotactic patient setup, as described elsewhere for glioblastoma multiforme.33

Carcinoma of the Nasopharynx

B

P

The juxtaposition of the nasopharynx to the brainstem and optic apparatus justifies consideration of proton therapy for carcinoma of the nasopharynx, an often-curable disease. Protons are of particular advantage for T3 and T4 nasopharyngeal carcinomas that have extended to the base of the skull or posteriorly to surround the spinal cord and brainstem, both of which are dose-limiting structures. Curable doses can be delivered with proton therapy that cannot be delivered with any other radiation modality with acceptable toxicity. However, irradiation of regional lymph nodes is a challenge for proton therapy, and the excellent results obtained with IMRT51 raise the question of the relative values of IMRT and proton therapy for this purpose.

Carcinoma of the Oropharynx C FIGURE 40-6.  Dosimetric comparisons of three treatment techniques for a pelvic sarcoma in a child. Axial slices are shown for plans generated for three-dimensional conformal radiation therapy (3D CRT) (A); intensity-modulated radiation therapy (B); and proton therapy (C). Isodose lines for the 3D CRT and intensity-modulated radiation therapy plans (A and B) are as follows: red, 40 Gy; aqua, 30 Gy; purple, 15 Gy; and yellow, 5 Gy. Isodose lines for the proton therapy plan (C) are as follows: orange/red, 40 Gy; green, 30 Gy; light blue, 15 Gy; and dark blue, 5 Gy. (From Lee CT, Bilton SD, Famiglietti RM, et al. Treatment planning with protons for pediatric retinoblastoma, medulloblastoma, and pelvic sarcoma: how do protons compare with other conformal techniques? Int J Radiat Oncol Biol Phys 2005;63:362-372.)

associated with surgery and radiation for ­ malignancies of the paranasal sinuses remains formidable. Although technical improvements have been made in microvascular reconstructive procedures and combined neurosurgical and otolaryngologic skull base resections, preservation of vision with complete surgical ablation continues to be difficult when disease is advanced. Delivery of tumoricidal doses of conventional radiation therapy is constrained by nearby

As is true for nasopharyngeal tumors, nodal irradiation is an important part of treating oropharyngeal tumors. The challenge for proton therapy is to adequately encompass regional lymph nodes that are involved or at high risk from tumors that arise in the tonsillar fossa and base of the tongue. IMRT can effectively treat primary tumors and lymph nodes at the same time; whether proton therapy can also do so has yet to be seen.

Non–Small Cell Lung Cancer Protons interact with normal lung tissue in a manner similar to x-rays (see Chapter 19). The advantage of proton therapy for lung cancer, as for other cancers, is the ability to focus it more precisely to the site of the tumor and to avoid surrounding normal tissues; this can be extremely valuable in the chest. Investigators at the Loma Linda University Medical Center used proton beam therapy (without chemotherapy) to treat clinical stage I, II, or IIIA non–small cell carcinoma of the lung in patients who were not candidates for surgical resection.52 In that study, the local control rate at 2 years was 87%, and no patient developed pneumonitis requiring hospitalization. Shioyama and colleagues53 reported findings for 51 patients who underwent proton beam therapy at Tsukuba University. In that study, the local control rates were 89% for the 9 patients with stage IA disease and 39% for the 19 patients with stage IB tumors.

1044

PART 14  Special Considerations

Only one patient had severe radiation pneumonitis, and no patient experienced life-threatening toxicity. Concurrent chemotherapy and radiation therapy, considered the standard of care in the United States for stage III tumors, was not used for any of the patients in this study. Chang and colleagues at The University of Texas M.  D. Anderson Cancer Center reported detailed comparisons of normal tissue doses in various types of radiation therapy for stage I and stage III non–small cell lung cancer.54 In this comparison of three-dimensional CRT, IMRT, and proton beam therapy, proton therapy clearly resulted in reduced doses to normal tissues, even when the tumor doses were higher than those that can safely be given with the other techniques (Fig. 40-7).54

Esophageal Cancer Proton therapy can be used to treat carcinoma of the esophagus to higher fractional and total doses without exceeding the tolerance of the spinal cord and with irradiation of a reduced volume of normal lung.55,56 Concurrent chemotherapy and radiation therapy, the standard treatment for inoperable esophageal cancer, is associated with considerable toxicity. Adding surgical resection after chemotherapy and radiation therapy carries a substantial risk (about 20%) of postoperative pneumonia that is directly related to the volume of lung that was irradiated.57 Proton therapy can reduce the volume of lung and heart irradiated and thereby

reduce the risk of toxicity. Formal trials of proton therapy versus x-ray therapy with chemotherapy for this disease have not been done.

Prostate Cancer Radiation therapy for carcinoma of the prostate has improved greatly since the advent of three-dimensional CRT in the early to mid-1990s. Dose escalation has proved feasible and effective with three-dimensional CRT compared with two-dimensional delivery of x-rays. IMRT permitted even higher doses, and preliminary results suggest that the rates of rectal and bladder morbidity are very low.58 Proton therapy for prostate cancer has been investigated since the early 1980s. Investigators at Massachusetts General Hospital completed a phase III trial comparing 67.2 Gy of photons with 75.6 CGE delivered with a conformal perineal proton boost (necessitated by the energy limitations of the proton beam at the Harvard Cyclotron ­Laboratory).59 From 1982 through 1992, 202 patients with T3-T4 prostate cancer received 50.4 Gy with four-field photons. Patients then received 25.2 CGE with conformal protons or a 16.8-Gy photon boost. No differences were found in overall survival, total recurrence-free survival, or ­overall local recurrence-free survival in the two groups. At 7 years of follow-up, the local recurrence-free survival rate for men with poorly differentiated (Gleason score 9 or 10) tumors was 85% in the proton-boost group and 37% in the

Isodose displays comparison in stage III NSCLC Photon IMRT

Proton

FIGURE 40-7.  Dosimetric comparison of intensity-modulated radiation therapy (IMRT) and proton radiation for the treatment of stage III non–small cell lung cancer (T2 N3 M0). Contralateral hilar lymph node involvement was treated to 60 Gy. (From Chang JY, Zhang X, Wang X, et al. Significant reduction of normal tissue dose by proton radiotherapy compared with three-dimensional conformal or intensity-modulated radiation therapy in stage I or stage III non–small-cell lung cancer. Int J Radiat Oncol Biol Phys 2006;65:1087-1096.)

40  Proton Therapy

photon-boost group. Rates of grade 1 and 2 rectal bleeding were higher in the proton group (32% versus 12% in the photon group), as were the rates of urethral stricture (19% in the proton group versus 8% in the photon group). Dose escalation to 75.6 CGE by conformal proton boost led to increased late sequelae of radiation but did not improve overall, disease-specific, or total recurrence-free survival rates in any subgroup. However, the proton boost did improve local recurrence-free survival among patients with poorly differentiated tumors. The results of the Loma Linda University Medical Center study suggested that higher total doses could be advantageous for early-stage (T1-T2b) prostate carcinoma.60 In that study, 319 men were treated with 74 to 75 CGE protons or combined proton and photon therapy. The overall 5-year biochemical disease–free survival rate was 88%, similar to that after radical prostatectomy, and no severe treatment-related morbidity was observed. These results were used as the basis for initiating the Proton Radiation Oncology Group protocol 95-09, a randomized phase III trial conducted at Massachusetts Gen­eral Hospital and Loma Linda University Medical Center to test conformal photons with a proton boost for early-stage prostate cancer.61 Patients were randomly assigned to two different proton-boost dose levels, with patients receiving a total dose of 79.2 Gy or 70.2 Gy. Patients deemed to be at low risk for local failure (i.e., PSA level of <4 ng/mL) and those at high risk for distant disease (i.e., PSA level >15 ng/mL, with or without positive lymph nodes) were excluded. This trial was designed to detect a 20% increase in freedom from local failure and biochemical relapse at 5 years. This trial was closed with 390 patients entered. A statistically significant reduction in local failure rate, with a corresponding decrease in the 5-year biochemical failure rate, was found with the higher proton dose. The rates of failure among the highproton-dose group were approximately one half of those of the group given the lower total dose.61 The suggestion has been made that IMRT can achieve high tumor control rates in men with prostate cancer, with a similarly low risk of side effects as that achieved with proton therapy.58 However, because IMRT requires the delivery of low doses of radiation to very large volumes of normal tissues in the pelvis, IMRT must be assumed to increase the risk of a second malignant tumor through exposure of pelvic organs to unnecessary radiation and the necessity for more monitor units (i.e., more time “turned on”) for any dose delivered to the prostate, which exposes the entire body to additional radiation. Brenner and colleagues62 showed that the risk of second malignant tumors in the pelvis was increased among patients receiving radiation therapy for cancer of the prostate relative to those undergoing prostatectomy. The large volume of pelvic tissues exposed with IMRT relative to proton therapy suggests an advantage in favor of proton therapy for this specific late effect of treatment.

Protons in Combined-Modality Therapies Combining chemotherapy with radiation therapy has become the standard treatment approach for many common types of cancer. Carcinomas of the esophagus, lung,

1045

head and neck, cervix, and rectum, for example, are usually treated with concurrent chemotherapy and radiation therapy. Logically, the ability of proton therapy to avoid normal structures that can give rise to acute toxicity can be expected to diminish the adverse effects of concurrent therapy. Preliminary results from the Proton Therapy Center in Houston suggest that esophageal reactions from concurrent chemotherapy and proton therapy are less common and milder than those from chemotherapy and three-dimensional conformal x-rays. Proton therapy may facilitate better tolerance of standard chemotherapy regimens and may allow still higher doses of chemotherapy, as has been demonstrated with three-dimensional CRT.21 Moreover, avoidance of normal tissues may permit intensification of the radiation therapy by higher total doses or larger doses per fraction, or both. Dose escalation of chemotherapy and proton therapy may translate into improved survival. Studies seeking to confirm these potential advantages include prospective randomized phase II trials comparing IMRT with proton therapy, both with concurrent chemotherapy, for locally advanced non–small cell lung cancer.

Challenges for Proton Therapy A companion article63 to a review of clinical experiences with particle therapy64 for the disease sites reviewed here emphasized the lack of large randomized comparative trials that include long-term observations. The absence of such data is not surprising, because until the past few years, the world has had only one hospital-based proton facility, and most proton treatments have been administered in physics research facilities with limited availability of beam time, energies, and field sizes. Nonetheless, the comparative dose distributions for protons versus photons are at least as compelling as those that drove the development of three-dimensional CRT and IMRT. The potential benefits from combining proton therapy with concurrent chemotherapy for common diseases were not discussed in either article. A major problem for conducting randomized trials of proton therapy versus photon therapy is the tension between a central requirement for randomized clinical trials (equipoise between treatment groups) and ethical issues ­regarding fully informed consent and the willingness of patients to be enrolled.65 The potential for proton therapy to improve treatments by avoiding normal tissues is just beginning to be appreciated. Dosimetric comparisons have shown the value of proton therapy for a wide variety of malignant diseases, including carcinoma of the lung and esophagus; tumors of the paranasal sinuses and nasopharynx; carcinoma of the rectum; and virtually any tumor close to critical and highly sensitive normal structures. The value of three­dimensional CRT over two-dimensional treatment was first demonstrated by comparisons of dose distributions; that value was then demonstrated in clinical practice. The value of proton therapy over three-dimensional CRT, clearly shown by comparing the quality of dose distribution, should translate into better outcomes, as reflected by consistent reductions of the probability of adverse effects on normal tissues and increased control of tumors that are relatively large or located around critical normal tissues. Worldwide ­ experiences with three-dimensional CRT and

1046

PART 14  Special Considerations

IMRT ­ provide compelling arguments for the advantages of proton therapy, especially scanning proton beams that ­permit intensity-modulated proton therapy.

REFERENCES   1. Del Regato JA. Earnest Rutherford. In Radiological Physicists. New York: American Institute of Physics, 1985, pp 37-49.   2. Wilson RR. Radiological uses of fast protons. Radiology 1946; 47:487-491.   3. Sisterson J. World wide charged particle patient totals. Particles [newsletter] 2005;36:1-11. Available at http://ptcog.web.psi. ch/Archive/ptles36.pdf (accessed April 2009).   4. Halperin EC. Particle therapy and the treatment of cancer. Lancet Oncol 2006;7:676-685.   5. Wilson R. A Brief History of the Harvard University Cyclotrons. Cambridge, MA: Harvard University Press, 2004.   6. Levin WP, Kooy H, Loeffler JS, et al. Proton beam therapy. Br J Cancer 2005;93:849-854.   7. International Commission on Radiation Units and Measurements. Prescribing, Recording, and Reporting Photon Beam Therapy (ICRU Report 50). Bethesda, MD: International Commission on Radiation Units and Measurements, 1993.   8. Dearnaley DP, Khoo VS, Norman AR, et al. Comparison of radiation side-effects of conformal and conventional radiotherapy in prostate cancer: a randomised trial. Lancet 1999;353:267-272.   9. Pollack A, Zagars GK, Smith LG, et al. Preliminary results of a randomized radiotherapy dose-escalation study comparing 70 Gy with 78 Gy for prostate cancer. J Clin Oncol 2000;18:3904-3911. 10. Mell LK, Roeske JC, Mundt AJ. A survey of intensity-modulated radiation therapy use in the United States. Cancer 2003;98: 204-211. 11. Bentzen SM. Radiation therapy: intensity modulated, image guided, biologically optimized and evidence based. Radiother Oncol 2005;77:227-230. 12. Miles EA, Clark CH, Urbano MT, et al. The impact of introducing intensity modulated radiotherapy into routine clinical practice. Radiother Oncol 2005;77:241-246. 13. Lin A, Kim HM, Terrell JE, et al. Quality of life after parotidsparing IMRT for head-and-neck cancer: a prospective longitudinal study. Int J Radiat Oncol Biol Phys 2003;57:61-70. 14. Pollack A, Zagars GK, Starkschall G, et al. Prostate cancer radiation dose response: results of the M.D. Anderson phase III randomized trial. Int J Radiat Oncol Biol Phys 2002;53:1097-1105. 15. Storey MR, Pollack A, Zagars G, et al. Complications from ­radiotherapy dose escalation in prostate cancer: preliminary results of a randomized trial. Int J Radiat Oncol Biol Phys 2000;48: 635-642. 16. Hanks GE, Hanlon AL, Pinover WH, et al. Survival advantage for prostate cancer patients treated with high-dose three-­dimensional conformal radiotherapy. Cancer J Sci Am 1999;5:152-158. 17. Peeters ST, Heemsbergen WD, Koper PC, et al. Dose-response in radiotherapy for localized prostate cancer: results of the Dutch multicenter randomized phase III trial comparing 68 Gy of radiotherapy with 78 Gy. J Clin Oncol 2006;24:1990-1996. 18. Michalski JM, Winter K, Purdy JA, et al. Toxicity after three­dimensional radiotherapy for prostate cancer on RTOG 9406 dose level V. Int J Radiat Oncol Biol Phys 2005;62:706-713. 19. Michalski JM, Winter K, Purdy JA, et al. Preliminary evaluation of low-grade toxicity with conformal radiation therapy for prostate cancer on RTOG 9406 dose levels I and II. Int J Radiat ­Oncol Biol Phys 2003;56:192-198. 20. Fang LC, Komaki R, Allen P, et al. Comparison of outcomes for patients with medically inoperable stage I non-small cell lung cancer treated with two-dimensional versus three-dimensional radiotherapy. Int J Radiat Oncol Biol Phys 2006;66:108-116. 21. Fossella FV, Zinner RG, Komaki R, et al. Gemcitabine (G) with concurrent chest radiation (XRT) followed by consolidation chemotherapy with gemcitabine plus cisplatin (CDDP): a phase I trial for patients with stage III non-small cell lung cancer (NSCLC) [abstract]. J Clin Oncol 2001;20:312a.

22. Bucci MK, Bevan A, Roach M 3rd. Advances in radiation ­therapy: conventional to 3D, to IMRT, to 4D, and beyond. CA Cancer J Clin 2005;55:117-134. 23. Zelefsky MJ, Chan H, Hunt M, et al. Long-term outcome of high dose intensity modulated radiation therapy for patients with clinically localized prostate cancer. J Urol 2006;176: 1415-1419. 24. Yom SS, Liao Z, Liu HH, et al. Initial evaluation of treatmentrelated pneumonitis in advanced-stage non-small-cell lung cancer patients treated with concurrent chemotherapy and ­intensity-modulated radiotherapy. Int J Radiat Oncol Biol Phys 2007;68:94-102. 25. Paganetti H. Significance and implementation of RBE variations in proton beam therapy. Technol Cancer Res Treat 2003;2: 413-426. 26. Moore RF. Choroidal sarcoma treated by intra-ocular insertion of radon seeds. Br J Ophthalmol 1930;14:145-152. 27. Char D. Current treatments and trials in uveal melanoma. ­Oncology 1989;3:113-124. 28. The Collaborative Ocular Melanoma Study (COMS) randomized trial of pre-enucleation radiation of large choroidal melanoma: IV. Ten-year mortality findings and prognostic factors. COMS Report No. 24. Am J Ophthalmol 2004;138:936-951. 29. Diener-West M, Earle JD, Fine SL, et al. The COMS randomized trial of iodine 125 brachytherapy for choroidal melanoma, III: initial mortality findings. COMS Report No. 18. Arch Ophthalmol 2001;119:969-982. 30. Char DH, Quivey JM, Castro JR, et al. Helium ions versus iodine 125 brachytherapy in the management of uveal melanoma. A prospective, randomized, dynamically balanced trial. Ophthalmology 1993;100:1547-1554. 31. Gragoudas E, Li W, Goitein M, et al. Evidence-based estimates of outcome in patients irradiated for intraocular melanoma. Arch Ophthalmol 2002;120:1665-1671. 32. Egger E, Zografos L, Schalenbourg A, et al. Eye retention after proton beam radiotherapy for uveal melanoma. Int J Radiat ­Oncol Biol Phys 2003;55:867-880. 33. Fitzek MM, Thornton AF, Rabinov JD, et al. Accelerated fractionated proton/photon irradiation to 90 cobalt gray equivalent for glioblastoma multiforme: results of a phase II prospective trial. J Neurosurg 1999;91:251-260. 34. Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005;352:987-996. 35. Wenkel E, Thornton AF, Finkelstein D, et al. Benign ­meningioma: partially resected, biopsied, and recurrent intracranial tumors treated with combined proton and photon radiotherapy. Int J Radiat Oncol Biol Phys 2000;48:1363-1370. 36. Vicente A, Estrada J, de la Cuerda C, et al. Results of external pituitary irradiation after unsuccessful transsphenoidal surgery in Cushing’s disease. Acta Endocrinol (Copenh) 1991;125: 470-474. 37. Harsh GR, Thornton AF, Chapman PH, et al. Proton beam stereotactic radiosurgery of vestibular schwannomas. Int J Radiat ­Oncol Biol Phys 2002;54:35-44. 38. Bush DA, McAllister CJ, Loredo LN, et al. Fractionated proton beam radiotherapy for acoustic neuroma. Neurosurgery 2002;50:270-273; discussion; 273-275. 39. Munzenrider JE, Liebsch NJ. Proton therapy for tumors of the skull base. Strahlenther Onkol 1999;175(Suppl 2):57-63. 40. Pai HH, Thornton A, Katznelson L, et al. Hypothalamic/pituitary function following high-dose conformal radiotherapy to the base of skull: demonstration of a dose-effect relationship using dose-volume histogram analysis. Int J Radiat Oncol Biol Phys 2001;49:1079-1092. 41. Hug EB, Loredo LN, Slater JD, et al. Proton radiation therapy for chordomas and chondrosarcomas of the skull base. J Neurosurg 1999;91:432-439. 42. Noel G, Habrand JL, Mammar H, et al. Combination of photon and proton radiation therapy for chordomas and chondrosarcomas of the skull base: the Centre de Protontherapie D’Orsay ­experience. Int J Radiat Oncol Biol Phys 2001;51:392-398.

40  Proton Therapy 43. Weber DC, Rutz HP, Pedroni ES, et al. Results of spot-scanning proton radiation therapy for chordoma and chondrosarcoma of the skull base: the Paul Scherrer Institut experience. Int J Radiat Oncol Biol Phys 2005;63:401-409. 44. Hall EJ. Intensity-modulated radiation therapy, protons, and the risk of second cancers. Int J Radiat Oncol Biol Phys 2006;65:1-7. 45. St Clair WH, Adams JA, Bues M, et al. Advantage of protons compared to conventional X-ray or IMRT in the treatment of a pediatric patient with medulloblastoma. Int J Radiat Oncol Biol Phys 2004;58:727-734. 46. Lee CT, Bilton SD, Famiglietti RM, et al. Treatment planning with protons for pediatric retinoblastoma, medulloblastoma, and pelvic sarcoma: how do protons compare with other conformal techniques? Int J Radiat Oncol Biol Phys 2005;63:362-372. 47. Yuh GE, Loredo LN, Yonemoto LT, et al. Reducing toxicity from craniospinal irradiation: using proton beams to treat medulloblastoma in young children. Cancer J 2004;10:386-390. 48. Hug EB, Muenter MW, Archambeau JO, et al. Conformal proton radiation therapy for pediatric low-grade astrocytomas. Strahlenther Onkol 2002;178:10-17. 49. Beale FA, Garrett PG. Cancer of the paranasal sinuses with particular reference to maxillary sinus cancer. J Otolaryngol 1983;12:377-382. 50. Allen MW, Schwartz DL, Rana V, et al. Long-term radiotherapy outcomes for nasal cavity and septal cancers. Int J Radiat Oncol Biol Phys 2008;71:401-406. 51. Lee N, Xia P, Quivey JM, 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. 52. Bush DA, Slater JD, Bonnet R, et al. Proton-beam radiotherapy for early-stage lung cancer. Chest 1999;116:1313-1319. 53. Shioyama Y, Tokuuye K, Okumura T, et al. Clinical evaluation of proton radiotherapy for non–small-cell lung cancer. Int J Radiat Oncol Biol Phys 2003;56:7-13. 54. Chang JY, Zhang X, Wang X, et al. Significant reduction of normal tissue dose by proton radiotherapy compared with three­dimensional conformal or intensity-modulated radiation therapy in stage I or stage III non–small-cell lung cancer. Int J Radiat Oncol Biol Phys 2006;65:1087-1096.

1047

55. Koyama S, Tsujii H, Yokota H, et al. Proton beam therapy for ­patients with esophageal carcinoma. Jpn J Clin Oncol 1994; 24:144-153. 56. Sugahara S, Tokuuye K, Okumura T, et al. Clinical results of proton beam therapy for cancer of the esophagus. Int J Radiat Oncol Biol Phys 2005;61:76-84. 57. Lee HK, Vaporciyan AA, Cox JD, et al. Postoperative pulmonary complications after preoperative chemoradiation for esophageal carcinoma: correlation with pulmonary dose-volume histogram parameters. Int J Radiat Oncol Biol Phys 2003;57:1317-1322. 58. Zelefsky MJ, Fuks Z, Hunt M, et al. High-dose intensity modulated radiation therapy for prostate cancer: early toxicity and biochemical outcome in 772 patients. Int J Radiat Oncol Biol Phys 2002;53:1111-1116. 59. Shipley WU, Verhey LJ, Munzenrider JE, et al. Advanced prostate cancer: the results of a randomized comparative trial of high dose irradiation boosting with conformal protons compared with conventional dose irradiation using photons alone. Int J Radiat Oncol Biol Phys 1995;32:3-12. 60. Slater JD, Rossi CJ Jr, Yonemoto LT, et al. Conformal proton therapy for early-stage prostate cancer. Urology 1999;53: 978-984. 61. Zietman AL, DeSilvio ML, Slater JD, et al. Comparison of ­conventional-dose vs high-dose conformal radiation therapy in clinically localized adenocarcinoma of the prostate: a randomized controlled trial. JAMA 2005;294:1233-1239. 62. Brenner DJ, Curtis RE, Hall EJ, et al. Second malignancies in prostate carcinoma patients after radiotherapy compared with surgery. Cancer 2000;88:398-406. 63. Brada M, Pijls-Johannesma M, De Ruysscher D. Proton ­therapy in clinical practice: current clinical evidence. J Clin Oncol 2007;25:965-970. 64. Schulz-Ertner D, Tsujii H. Particle radiation therapy using proton and heavier ion beams. J Clin Oncol 2007;25:953-964. 65. Goitein M, Cox JD. Should randomized clinical trials be required for proton radiotherapy? J Clin Oncol 2008;26:175-176.

Index

A

Abdominopelvic radiation. See Ovarian cancer, radiation therapy in Absolute growth delay, 104–105, 105f ABT-510, 124 ABT-526, 124 Acetaminophen, 1015t Acetylsalicylic acid, 1015t Acral lentiginous melanoma, 144–145 Acute lymphoblastic leukemia, 969–971 cranial radiation therapy in, 970–971, 971f craniospinal radiation therapy in, 971 meningeal, 970–971 prognosis for, 971 testicular, 971 Ad-Delta24-RGD, 131 Ad-p53 (adenoviral-mediated TP53), 130 Addiction, 1013–1014 Adenoma parotid gland, 173, 173f, 176–177, 177t pituitary gland, 1040 α-Adrenergic receptors, radiation effects on, 416–417 β-Adrenergic receptors, radiation effects on, 416–417 Age breast cancer and, 357, 357t breast-conservation therapy and, 370 cervical cancer and, 744 endometrial cancer and, 777 Alcohol use breast cancer and, 357t, 358 colorectal cancer and, 561, 563 Alopecia, total-body irradiation and, 999 Alpha-fetoprotein in germ cell tumor, 850 in testicular cancer, 664 Alpha particles, 4 energy track of, 33 Alveolus (alveoli), 424. See also Lung(s) type I cells of, 424 type II cells of, 424 Amenorrhea, radiation-induced, 812–813 Amifostine, 32–33 with chemoradiation therapy, 33 growth plate protection with, 916 in radiation pneumonitis, 432–433, 432f Amitriptyline, 1015t Amphetamine, 1015t Ampulla of Vater carcinoma, radiation therapy in, 548 Amyloid, in medullary thyroid cancer, 336, 336f Anal cancer, 606 anal margin, 616 clinical presentation of, 609–610 epidemiology of, 607–608 metastatic, 609–610 molecular features of, 610

Anal cancer (Continued) pathology of, 608 prognosis for, 610 radiation therapy and chemotherapy in, 113–114, 610–615, 612t cisplatin in, 613–615 5-fluorouracil in, 612–615, 614–615t HIV infection and, 615 mitomycin C in, 612–613, 614–615t vs. radiation therapy, 611–612 toxicity of, 619–620, 620t radiation therapy in, 609t age-related toxicity of, 618 dose-time consideration in, 618 intensity-modulated, 618 interstitial iridium-192 in, 619 vs. radiation therapy and chemotherapy, 611–612 techniques of, 616–619, 616–617f toxicity of, 618–620, 620t recurrence of, 610t risk factors for, 607b staging of, 607f, 610 surgery in, 609t, 620 Anal sphincter, preservation of, in rectal cancer treatment, 575, 586–589, 587t Analgesia, 1013–1014 prescription for, 1014, 1015–1016t Anaplastic large cell lymphoma, 901 Anastrozole, in early-stage breast cancer, 379 Androgen, prostate cancer and, 679 Androgen ablation, in prostate cancer, 684, 701–704, 701–702t, 709–710, 709t Angiogenesis, 122–125 inhibition of. See Antiangiogenic therapy radiation-induced, 123 Angiography in auricular cancer, 324 in chemodectoma, 328, 329f Angiostatin, radiation therapy and, 125–126 Animal assays, 16–17 dilution, 16 growth delay, 16 lung colony, 16 tumor-control dose, 16 in vivo/in vitro, 16–17 Anorexia, radiation-related, 821 Antiangiogenic therapy, 110, 123–125 chemotherapy with, 125–126 in mouse model, 125–126 radiation therapy with, 125–126 Anticonvulsants, 1015t Antidepressants, 1015t 131I-Antiferritin antibody, in hepatocellular carcinoma, 545 Antihistamines, 1015t Antisense oligonucleotides, 131 Antivascular therapy, 123

Anus anatomy of, 606–607, 607f cancer of. See Anal cancer radiation effects on, 607 Aorta, radiation-induced growth arrest in, 414 Apocrine carcinoma, 146 treatment of, 152 Aromatase inhibitors, in early-stage breast cancer, 378–379 Arteriosclerosis, radiation-induced, 413 Artery (arteries), radiation-induced injury to arteriosclerosis and, 413 atherosclerosis and, 413 growth arrest and, 414 hypertension and, 413 increased permeability and, 412, 412f mechanisms of, 413–414 vessel occlusion and, 412–413, 413f Aryepiglottic folds, 282, 283–284f tumors of, 285, 286f Arytenoid cartilage, 282–283, 283f Aspirin, 1015t Astrocytoma adult, 854–856 prognosis for, 856 radiation therapy in, 844, 856, 856f spinal cord, 860 pediatric, 841–846, 845f cerebellar, 842–843 grade of, 842 natural history of, 842 prognosis for, 846 radiation therapy in, 845–846, 846f spinal cord, 860 supratentorial, 843–844 Atherosclerosis, radiation-induced, 413–414 Atrial fibrillation, radiation-induced, 480 Atrioventricular block, radiation-induced, 417 Auricle, 320, 321f cancer of, 323. See also Ear(s), cancer of radiation effects on, 320–321 Autoradiography, 18, 18f Axillary lymph nodes, 354, 355f, 376–378 adenocarcinoma with unknown primary of, 387–388 evaluation of, 387–388 radiation therapy in, 387–388, 387–388t radiation therapy for, 376–378 sentinel node surgery for, 377–378 skip metastases of, 354 surgical treatment of, 376

B

B-mode Acquisition and Target (BAT) ­system, in prostate cancer, 690, 691f Bacillus Calmette-Guérin, in bladder cancer, 633–634

Pages followed by f indicate figures; t, tables; b, boxes.

1049

1050

Index

Basal cell cancer. See Skin cancer, basal cell Basal cell nevus syndrome, 142 BAX, 631 BCL2, 9, 631, 889–890 Beckwith-Wiedemann syndrome, 971 Benign mixed (pleomorphic) adenoma, of parotid gland, 173, 173f, 176–177, 177t Beta particle plaques in ocular melanoma, 80–81, 81f in pterygium, 313 Bevacizumab, 124 in colorectal cancer, 565 in pancreatic cancer, 126, 529 Bile duct cancer. See Biliary tract cancer; Cholangiocarcinoma Biliary tract, anatomy of, 534–535 Biliary tract cancer, 545–555 prognosis for, 555 radiation therapy in acute effects of, 536 adjuvant, 546–549 brachytherapy for, 551–552 charged-particle, 550–551 intraoperative, 552 neoadjuvant, 549–550 palliative, 550–553 techniques of, 553–555, 553–554f surgery in, 545–546 Biological response modifiers, 38 Biopsy breast, 362 pineal region tumor, 850 prostate gland, 680–683, 706 soft tissue tumor, 944 Bisphosphonates, in palliative care, 1011–1012 Bladder anatomy of, 627 cancer of. See Bladder cancer capacity of, 627 lymphatic drainage of, 627 motion of, radiation therapy and, 641, 641f radiation effects on, 627–629, 644 early, 628 late, 628, 629f management of, 629 patient factors and, 628–629 Bladder cancer, 629–645 chemotherapy in, 633–634, 638t adjuvant, 637 neoadjuvant, 636–637 palliative, 639–640 radiation therapy with, 114, 636t, 637 clinical presentation of, 631–632 epidemiology of, 629–630 genetic factors in, 631 imaging of, 631, 632f lymph node metastases in, 632, 634 with muscularis propria invasion, 631 without muscularis propria invasion, 630–631 palliative treatment in, 639–640, 639b chemotherapy for, 639–640 radiation therapy for, 639, 1030 papillary, 629f, 630–631, 630b pathology of, 630–631, 630b radiation therapy and chemotherapy in, 114, 636t, 637 radiation therapy in, 640–644, 642f accelerated, 644 complications of, 644 dose for, 642–643 fractionation for, 642–643

Bladder cancer (Continued) hyperfractionated, 643–644 hypofractionated, 644 intensity-modulated, 643, 643f palliative, 639, 1030 patient selection for, 640 recurrence after, 644–645 volume definition for, 640–641, 641f recurrence of, 644–645 risk factors for, 629–630 sessile, 631 in situ, 630, 636 staging of, 632, 633b superficial, 630 T2-T4 chemotherapy in adjuvant, 637 neoadjuvant, 636–637 radiation therapy with, 636t, 637 lymphadenectomy in, 634 radiation therapy in, 635–636, 635t bladder motion and, 641, 641f chemotherapy after, 637 chemotherapy before, 636–637, 638t chemotherapy with, 636t, 637 hypoxia and, 635–636 interstitial, 638–639 nodal, 641 postoperative, 638 preoperative, 637–638 in situ cancer after, 636 techniques of, 641–644 surgery in, 634–635 radiation after, 638 radiation before, 637–638, 638t salvage, 645 Ta, T1, Tis chemotherapy in, 633–634 immunotherapy in, 633–634 radiation therapy in, 634 surgery in, 632–633 Bladder wall, 627 Blood cells, radiation effects on, 876–877 Blood vessels, 411–414. See also Artery (­arteries); Capillary (capillaries) anatomy of, 411 large, radiation effects on, 413–414 small, radiation effects on, 411–413, 412–413f Bone, 915 chondrosarcoma of, 920–921 chordoma of, 923–924 eosinophilic granuloma of, 928 giant cell tumor of, 926–928 growth of, radiation effects on, 915–917 hemangioendothelioma of, 925–926 heterotopic, 917 classification of, 917–918 prevention of, 917–919 risk of, 917 imaging of, 919 lymphoma of, 921–923, 922f malignant fibrous histiocytoma of, 924–925, 925f metastatic disease of, 1016–1027 disseminated, 1025–1027 hemibody radiation therapy in, 1026 evaluation of, 1017 fracture with, 1020–1021, 1021f morbidity of, 1020–1024 radiation therapy in, 928–931

Bone (Continued) comparative trials of, 1027 hemibody, 930–931, 1026 outcomes of, 1017–1019, 1018t 32P in, 805, 1027t 186Re in, 1027, 1027t schedules in, 1017–1018 single- vs. multiple-fraction, 930 single-fraction vs. multiple-fraction, 1019–1020 153Sm in, 931, 1026–1027, 1027t 89Sr in, 931, 1017, 1026, 1027t survival duration and, 1020 vertebral, 1021–1024, 1022–1025f plasmacytoma of, 919–920 primary tumors of, 919–928 radiation effects on, 915–917, 967 secondary cancer of, 918–919 Bone marrow, radiation effects on, 875–876, 876t, 993–995, 994f Bone marrow (stem cell) transplantation, 993 allogeneic, 995 autologous, 995 boost radiation dose in, 998 CNS irradiation in, 998 historical perspective on, 993 indications for, 995–996, 995f localized irradiation in, 997–998 in noncancerous disorders, 996 nonmyeloablative conditioning regimens in, 998 in solid tumors, 996 splenic irradiation in, 998 syngeneic, 995 total-body irradiation in, 996–997, 996f acute toxic effects of, 998–999 alopecia with, 999 cardiac effects of, 1000 cataracts with, 1000 cutaneous effects of, 999 dose rate for, 996 fractionation for, 996–997 gastrointestinal effects of, 998 gonadal effects of, 1000 growth effects of, 1000–1001 hepatic effects of, 999 long-term effects of, 999–1002 meta-analysis of, 1002, 1002f neurocognitive effects of, 1001–1002 organ shielding for, 997, 997f parotid effects of, 998 psychological effects of, 1001–1002 pulmonary effects of, 999–1000 renal effects of, 1000 second malignancy after, 1001 sequelae of, 998–1002 thyroid effects of, 1001 total dose for, 996–997 types of, 995 Bone scan in metastatic disease, 919 in prostate cancer, 683 Brachytherapy, 76–81 in biliary tract cancer, 551–552 in brain tumor, 78–79, 79f in breast cancer, 79–80, 80f in cervical cancer, 76–78, 77–78f, 753, 755–763, 757–759f, 761t, 762f in cholangiocarcinoma, 551–552 in coronary artery stenosis, 419–420 in early-stage breast cancer, 374–376, 374f in endometrial cancer, 783–785, 784t, 789 in esophageal cancer, 475, 1029

Index Brachytherapy (Continued) in gallbladder cancer, 551–552 historical perspective on, 3–4 liquid source, 78–79, 79f MammoSite, 79–80, 80f, 374, 374f, 376 mandibular osteonecrosis with, 276 in nasal vestibule cancer, 189, 197, 198–199f in nasopharyngeal cancer, 215–216 in ocular melanoma, 80–81, 81f in oral cavity cancer, 265–266, 266–267f in prostate cancer, 691–694, 692–694f, 707 See also Prostate cancer, ­brachytherapy in in rectal cancer, 583 in soft palate cancer, 238 in soft tissue sarcoma, 949–950, 950t, 952 in tongue base cancer, 235 in tonsillar region cancer, 237 in uveal melanoma, 315 in vaginal cancer, 806–808, 807f Brain leukoencephalopathy of, 837 necrosis of, 836, 836f radiation effects on, 835–837, 836f Brain tumors. See also specific tumors in adults, 854–860 in children, 837–854, 838t, 838b in infants, 853–854, 853f liquid source brachytherapy for, 78–79, 79f metastatic, 859–860 in breast cancer, 391 in lung cancer, 434, 435t, 444–445, 448–449, 448t radiation therapy in, 1028, 1028f stereotactic radiosurgery/radiation therapy in, 63, 63f Brainstem glioma, 846–848, 847f prognosis for, 847–848 radiation therapy in, 847 Branchiomeric paraganglia, 327 BRCA1 breast cancer and, 359–361, 360t breast-conservation therapy and, 371 ovarian cancer and, 813–814 BRCA2 breast cancer and, 359–361 breast-conservation therapy and, 371 ovarian cancer and, 813–814 Breast. See also Breast cancer anatomy of, 353–354, 354f axillary extension of, 353–354 biopsy of, 362 ducts of, 354 examination of, 361–362, 361f lobules of, 354 lymphatics of, 354–355, 354f radiation effects on, 355–356 in children, 356 removal of. See also Breast cancer, ­mastectomy in prophylactic, 361 Breast cancer, 353 advanced breast-conservation therapy in, 386–387 palliative radiation therapy in, 391 age and, 357, 357t antiangiogenic agents in, 125 biopsy in, 362 body weight and, 358 bone metastases in, 391 brachytherapy for, 79–80, 80f brain metastases in, 391

Breast cancer (Continued) BRCA1 in, 359–361, 360t BRCA2 in, 359–361 contralateral breast in, 355 diagnosis of, 361–362, 361f early, 92 dietary factors and, 357t, 358 ductal in situ, 359, 362–367 breast-conservation therapy in, 364–367 EORTC 10853 trial of, 365t, 366 vs. mastectomy, 364 NSABP B-06 trial of, 365 NSABP B-17 trial of, 365–366, 365t prospective trials of, 366–367 radiation therapy in, 365, 365t SweDCIS trial of, 365t, 366 UK ANZ trial of, 365t, 366 Van Nuys data on, 366 clinical presentation of, 362 comedo necrosis in, 362–364, 365f estrogen receptor–positive, 367 high-grade, 362–364, 365f hormonal therapy in, 367 mammography in, 362, 365f pathology of, 362–364 tamoxifen therapy in, 367 treatment of, 364–367 well-differentiated, 362–364 early-stage, 367–380 accelerated partial irradiation for, 373–376 axillary treatment in, 376–378 radiation therapy in, 376–378 sentinel lymph node surgery in, 377–378 surgical, 376 breast-conservation therapy for, 376–368 age and, 369–370 brachytherapy in, 374–376, 374f Christie Hospital trial of, 374–375 conformal external beam therapy in, 374–376, 375f contraindications to, 370 Hungarian National Institute of Oncology trial of, 375 MammoSite brachytherapy in, 374, 374f, 376 vs. mastectomy, 354, 367–368 outcomes in, 370–373 extensive intraductal component and, 371–372 family history and, 371 genetic factors and, 371 histology and, 372–373 hormone receptor status and, 372–373 multicentric disease and, 372 multifocal disease and, 372 negative (surgical) margin in, 371–372, positive (surgical) margin in, 371–372 race and, 370–271 radiation therapy in, 368–369 modified radical mastectomy for, 367–368, 372 radiation therapy for, 368–369, 369f, 374–375f accelerated partial irradiation, 373–376 cardiac effects of, 418 chemotherapy with, 379

1051

Breast cancer (Continued) hormonal therapy with, 379–380 techniques of, 374 trastuzumab and, 380 tumor bed boost, 373 sequence of treatments for, 379–380 systemic therapy for, 373, 378–379 biologic, 379–380 chemotherapy in, 378, 380 hormonal, 368–369, 378–380 vs. radiation therapy, 368–369 environmental factors and, 358 epidemiology of, 356–359 estrogen replacement therapy and, 357 examination of, 361–362, 361f exercise and, 358 family history of, 357t, 358 fibrocystic disease and, 359 genetic factors in, 357t, 359–360 HER2/NEU in, 360 hormonal factors and, 357–358, 357t incidence of, 356, 357t inflammatory, 389–390 breast-conservation therapy in, 389 mastectomy in, 389 neoadjuvant chemotherapy in, 389 radiation therapy in, 389–390 lactation and, 358–359 lobular in situ, 362 lymphangiosarcoma and, 940 lymphatic metastases in, 354, 354f axillary, 354, 355f, 376–378 radiation therapy for, 376–378 sentinel node surgery for, 377–378 skip metastasis, 354 surgical treatment of, 376 internal mammary, 355, 355f postmastectomy radiation for, 384–385, 385–386t retrograde flow and, 355 supraclavicular, 354–355 radiation therapy for, 395–396, 396f, 398f in males, 357 mastectomy in, 367–368 chemotherapy after British Columbia Trial of, 381–382 Danish Breast Cancer Cooperative Group trials of, 382 M.D. Anderson studies of, 382–383, 382t chemotherapy before, 383–386 prophylactic, 361 radiation therapy after, 380–386 boost dose in, 396, 398f British Columbia Trial of, 381–382 cardiac effects of, 418 chest wall, 393–395, 395f clinical trials of, 381–383 Danish Breast Cancer Cooperative Group trials of, 382 dose in, 396–397 dosimetry in, 397–398 internal mammary chain, 384–385, 385–386t, 394, 396 patient selection for, 381–383 supraclavicular fossa, 395, 396f target volume in, 385–386 recurrence after, 383, 383–384t radiation therapy in, 390–391 metastasis in, 92 See also Breast cancer, lymphatic metastases in

1052

Index

Breast cancer (Continued) bone, 391 brain, 391 palliative radiation therapy in, 391 spinal, 391 microcalcifications in, 362, 365f molecular classification of, 360 mortality rates in, 356–357 neoadjuvant chemotherapy in, 383–386 palliative radiation therapy in, 391 parity and, 357t pathogenesis of, 359–361 BRCA1 in, 359–360, 360t BRCA2 in, 359–360 TP53 in, 359 pathologic risk factors and, 359 peau d’orange appearance in, 361–362, 361f prevention of, 360–361 experimental, 358 prophylactic surgery in, 361 raloxifene in, 361 tamoxifen in, 360–361 quadrant of, 354 radiation-induced ovarian ablation in, 812–813 radiation therapy in, 392–398 axilla, 395–396, 396f boost fields in, 396, 397–398f cardiac morbidity with, 99, 394–395 chest wall, 393–395, 394–395f complications of, 355–356, 370 connective tissue disease and, 370 cosmetic complications of, 356 CT simulation for, 393 dose prescriptions for, 396–397 dosimetry for, 397–398 excision cavity, 396 fields for, 392–396, 392–393f high-tangent technique for, 393 infraclavicular, 396, 397f intact breast, 392–393, 392–393f internal mammary lymph nodes, 396, 397f lactation after, 356 rod and chain device for, 393 second cancer after, 356 supraclavicular fossa, 395–396, 396f, 398f raloxifene in, 361 reproductive factors and, 357–358, 357t risk factors for, 357–359, 357t individual determination of, 359 salivary gland tumors and, 172 soft tissue metastases in, 391 spinal metastases in, 391 staging of, 362, 363b tamoxifen in, 360–361, 774 tamoxifen therapy in, 367 TP53 in, 359 Breast-conservation therapy. See Breast ­cancer, ductal in situ, breast-conservation therapy in; Breast cancer, early-stage, breast-conservation therapy in Broad ligaments, 735 Bronchoscopy, in esophageal cancer, 468 Brown’s sign, 328 Buccal mucosa cancer, 262t, 263, 272–273. See also Oral cavity cancer Buprenorphine, 1015t Burkitt’s lymphoma, 893f, 899 Butorphanol, 1015t

C

CA125, in ovarian cancer, 814–817 CA19-9 in cholangiocarcinoma, 539–540 in pancreatic cancer, 518 Caffeine, 1015t Calcitonin, in medullary thyroid cancer, 342–343 Cancer. See also specific cancers biology of, 8–10 genetics of, 10 multistep nature of, 9 second. See Second malignancy Candidiasis, oral cavity, radiation-related, 275 Cantlie’s line, 534 Capecitabine bevacizumab with, 124 in pancreatic cancer, 529 in rectal cancer, 580, 589 Capillary (capillaries), radiation injury to decreased permeability and, 412–413, 413f fibrosis and, 413 increased permeability and, 412 telangiectasia and, 413 vessel dilation and, 412 Carbamazepine, 1015t Carbogen, radiosensitizing effects of, 31 Carbon ion radiation therapy, 35 in hepatocellular carcinoma, 543 Carcinoembryonic antigen in colorectal cancer, 569–571 in medullary thyroid cancer, 342 in pancreatic cancer, 518 Carcinogenesis, 8–9 radiation-related, 39–40. See also Second malignancy absolute vs. relative risk models for, 40, 41f latent period in, 40 Carcinoid tumor lung, 433 small intestine, 509 Cardinal ligaments, 735 Caretaker genes, 9 Carotid artery, aberrant, vs. chemodectoma, 328, 329f Carotid artery disease, radiation-induced, 414 Cataracts, 311, 1000 CD30-positive cutaneous lymphoproliferative disorders, 901 CDKN2A, 631 in esophageal cancer, 483t in nasopharyngeal carcinoma, 210 CDKN2B, in esophageal cancer, 483t Celecoxib, in non–small cell lung cancer, 127–128, 442–443, 443f Cell(s) accelerated proliferation of, 109, 109–110f treatment-stimulated regeneration of, 108–110, 109f Cell cycle, 9–10, 17–19, 17f chemotherapy effects on, 107–108 flash labeling for, 18, 18f percent-labeled mitoses curve for, 18, 18f radiation effects on, 107–108 S phase of, 18, 18f, 108 taxane effect on, 107–108, 107f Cell death, 39 Cell loss fraction, 19

Cell survival curve, 13, 13f for normal tissue, 13–15, 14f Central nervous system(CNS). See Brain; Spinal cord Cerebral necrosis, postirradiation, 836, 836f Cervical cancer, 733 adenocarcinoma, 738, 743–744, 743t adenocarcinoma in situ, 737 age and, 744 as AIDS-defining disease, 734 anaplastic, 744 Bethesda System in, 736, 737t brachytherapy in, 755–763 complications of, 765 high-dose-rate, 760, 761t after hysterectomy, 762–763, 762f interstitial, 761–762, 762f low-dose-rate, 749t, 756–760, 757–759f applicator systems for, 756, 757–758f colpostats positioning for, 756–757, 759f complications of, 765 dose calculation for, 757–760, 759f tandem positioning for, 756–757, 759f uterine perforation with, 757 pulsed-dose-rate, 76–78, 77–78f, 760–761 chemotherapy in preoperative, 747 radiation therapy after, 764, 765t radiation therapy with, 751, 763–764 in stage IB2 or IIA disease, 747, 748t in stage II to III disease, 749–750 in stage IVB or recurrent disease, 752 cigarette smoking and, 745 clinical evaluation of, 740–741 cystoscopy in, 741 early diagnosis of, 92 epidemiology of, 734 human papillomavirus in, 734, 744 hysterectomy in, 744–745 after radiation therapy, 747 radiation therapy after, 750–751, 750f in stage IB1 or IIA disease, 745–747, 746t in stage IB2 or IIA disease, 747 recurrence after, 744 vaginal brachytherapy after, 762–763, 762f imaging in, 741, 741f inflammatory response in, 744 laboratory tests in, 741 lymph nodes in, 735–736, 736f, 743 para-aortic, 742–743, 743f postoperative radiation therapy for, 750–751, 750f surgical evaluation of, 742 metastatic, 736, 743–744, 743t, 752 microinvasive, 737 neuroendocrine carcinoma, 738 oral contraceptives and, 734 pathology of, 736–738, 737t, 743–744 pelvic examination in, 740–741 proctoscopy in, 741 prognosis for, 742–745 histology and, 743–744, 743t human papillomavirus subtype and, 744 local extent and, 742 lymph node involvement and, 743, 743f lymphovascular space invasion and, 743 patient-related factors and, 744–745

Index Cervical cancer (Continued) peritoneal tumor cells and, 744 serum squamous cell carcinoma antigen and, 744 tumor size and, 742, 742f radiation therapy in, 752–763, 1030. See also Cervical cancer, brachytherapy in chemotherapy before, 764, 765t chemotherapy with, 113, 751, 763–764 in stage IB2 or IIA disease, 747, 748t in stage II to III disease, 749–750 complications of, 764–766 acute, 764–765, 798 late, 765–766, 765f, 798 reproductive, 733–734 cystitis with, 629f duration of, 763, 763f external beam, 752–755 brachytherapy with, 753 bulky lymph node disease and, 755 complications of, 764–765 dose for, 754 field arrangements for, 753–754, 753–754f para-aortic region in, 754–755, 756f parametrial boost in, 755 positive pelvic nodes and, 755 hypoxia and, 744 hysterectomy after, in stage IB2 or IIA disease, 747 in inappropriate-simple-hysterectomy disease, 752 intensity-modulated, 754, 755f in locally recurrent disease, 751, 751t postoperative, 750–751, 750f in stage IB1 or IIA disease, 745–747 in stage IB2 or IIA disease, 747 second malignancy after, 41–42 in stage IB1 or IIA disease, 745, 746t in stage II to III disease, 747–750, 749t in stage IVA disease, 750 in stage IVB or recurrent disease, 752 rectovaginal examination in, 741 recurrence of, 743f, 744, 751–752 size of, 742, 742f spread of, 734–736, 742 squamous cell, 737, 743–744, 743t staging of, 738–740, 738b, 740f total pelvic exenteration in, 751 treatment of, 745–752 hypoxia and, 744 inappropriate hysterectomy in, 752 in locally recurrent disease, 751, 751t in minimal stage IIA disease, 747, 748t race and, 744–745 in recurrent disease, 752 in regional disease, 750–751, 750f in stage IA1 disease, 745 in stage IA2 disease, 745 in stage IB1 disease, 745–747, 746t in stage IB2 disease, 747, 748t in stage II to III disease, 747–750, 749t in stage IIA disease, 745–747, 746t, 748t in stage IV B disease, 752 in stage IVA disease, 750 vaginal trachelectomy in, 745 vascularity in, 744 Cervical intraepithelial neoplasia, 736–737, 737t Cervix, 733 anatomy of, 734–736, 735f atypical squamous cells of undetermined significance of, 736–737

Cervix (Continued) endometrial cancer invasion of, 778 lymphatics of, 735–736, 736f radiation effects on, 733–734 Cetuximab, 37–38 in colorectal cancer, 122, 564–565 in head and neck squamous cell ­carcinoma, 121–122, 166, 166–167t in non–small cell lung cancer, 443 in pancreatic cancer, 530 Checkpoint genes, 10 Chemical exposures bladder cancer and, 630 skin cancer and, 142 Chemodectoma, 326–331 clinical presentation of, 328 evaluation of, 328, 329f pathology of, 328 staging of, 328–329 treatment of, 329–330, 330t radiation therapy in, 330–331, 330t surgery in, 329, 330t Chemokine and bone marrow–homing receptor, in esophageal cancer, 483t Chemotherapy, 36–37. See also Radiation therapy and chemotherapy and at specific cancers adjuvant, 111–112 in advanced breast cancer, 386–387 antiangiogenic effects of, 110 antiangiogenic therapy with, 125–126 in bladder cancer, 633–634, 636–637, 638t, 639–640 in cerebral primitive neuroectodermal tumors, 853 in cervical cancer, 747, 752 in CNS lymphoma, 858–859 in colorectal cancer, 597 concurrent, 111–113, 112t in diffuse large B-cell lymphoma, 1026–1027 in dysgerminoma, 828 in early-stage breast cancer, 373, 378–380 in endometrial cancer, 785–787, 790–791 in ependymoma, 849 in esophageal cancer, 470–471, 471f esophagitis and, 431–432 in follicular lymphoma, 898 in gastric cancer, 500–504 in glioma, 856 in Hodgkin lymphoma, 883–884, 886–887 induction (neoadjuvant), 110–112 in inflammatory breast cancer, 389 in intracranial germinoma, 850–851 in laryngeal cancer, 291–292, 291t in laryngopharyngeal cancer, 293–294 in lymphocyte-predominant Hodgkin lymphoma, 888 in malignant fibrous histiocytoma, 924–925 in maxillary sinus cancer, 196–197 in medulloblastoma, 839, 841 in melanoma, 151 in mycosis fungoides, 900 in nasopharyngeal cancer, 216–219, 217–218t in non–small cell lung cancer, 440–442, 441–442f, 444t in orbital lymphoma, 316 organ-specific toxicity of, 114t in osseous lymphoma, 923 ototoxicity of, 322

1053

Chemotherapy (Continued) in ovarian cancer, 817, 818t, 819, 822–823, 826 in palliative care, 1011 in plasmacytoma, 920 postmastectomy, 381–383, 382t premastectomy, 383–386 with proton therapy, 1045 radiation pneumonitis and, 430, 430f in retinoblastoma, 981 in seminoma, 667–668 in soft tissue sarcoma, 953–955, 954–955t in thyroid cancer, 341–342 in vaginal cancer, 808 Chernobyl nuclear accident, 334 Child-Pugh score, 539 Children, 967, 968t astrocytoma in, 841–846, 845f cerebellar, 842–843 grade of, 842 natural history of, 842 prognosis for, 846 radiation therapy in, 845–846, 846f supratentorial, 843–844 brain tumors in, 837–854, 838t, 838b. See also specific tumors brainstem glioma in, 846–848, 847f prognosis for, 847–848 radiation therapy in, 847 craniopharyngioma in, 851–852, 852f prognosis for, 852 radiation therapy in, 852 ependymoma in, 848–849, 848–849f hepatic tumors in, 982 leukemia in, 969–971 biologic features of, 969–970 clinical features of, 969–970 meningeal, 970–971, 971f testicular, 971 medulloblastoma in, 838–841 chemotherapy in, 841 proton therapy in, 1041, 1041f radiation therapy in, 839–841, 840–841f, 841t, 851t staging of, 839, 839b surgery in, 839 nasopharyngeal cancer in, 220 neuroblastoma in, 974–976 staging of, 974–975, 975t treatment of, 975–976 pelvic sarcoma in, proton therapy in, 1041, 1043f pineal region tumor in, 849–851, 849f, 850t biopsy in, 850 radiation therapy in, 850–851 primitive neuroectodermal tumor in, 838, 852–853 proton therapy in, 1041–1042, 1041–1043f radiation-related tissue effects in, 915–917, 967–969 retinoblastoma in, 980–982, 981b prognosis for, 982 proton therapy in, 1041, 1042f radiation therapy in, 981–982, 981f rhabdomyosarcoma in, 976–980, 977f, 978t prognosis for, 980 radiation therapy for, 977–980, 979t second tumors in, 969 Wilms’ tumor in, 971–974, 972t

1054

Index

Children (Continued) abdominal radiation therapy in, 973 hepatic radiation therapy in, 974 prognosis for, 974 pulmonary radiation therapy in, 973–974 Cholangiocarcinoma distal radiation therapy in, 547–548 surgery in, 546–547 evaluation of, 539–540 perihilar radiation therapy in, 546–547 surgery in, 546 transplantation in, 552–553 radiation therapy in, 553–555, 553–554f adjuvant, 546–549 brachytherapy for, 551–552 charged-particle, 550–551 intraoperative, 552 late effects of, 536 neoadjuvant, 549–550 palliative, 550–553 preoperative, 552–553 proton, 550–551 respiratory movement and, 553–554 risk factors for, 539 staging of, 539–540 surgery in, 545–546 transplantation in, 552–553 Cholangitis, vs. cholangiocarcinoma, 539–540 Chondronecrosis, of ear, 321 Chondrosarcoma, 920–921 classification of, 920 cranial base, 921 dedifferentiated, 921 proton therapy for, 1040 radioresistance of, 921 surgery for, 921 Chordoma, 923–924 proton therapy for, 1040–1041 Choriocarcinoma, 660. See also Testicular cancer Chromosome(s) aberrations in, 11–12, 12f acentric fragment of, 12, 12f dicentric fragment of, 12, 12f fluorescence in situ hybridization of, 11–12, 12f ring, 12, 12f translocation of, 11–12, 12f Cigarette smoking anal cancer and, 608 bladder cancer and, 630 cervical cancer and, 745 oral cavity cancer and, 252, 259 pancreatic cancer and, 517 Cirrhosis Child-Pugh score for, 539 hepatocellular carcinoma and, 538 Cisplatin. See also Chemotherapy in anal cancer, 613–615 ototoxicity of, 322 Clear cell sarcoma, 972 Cloacogenic cancer, 608. See also Anal cancer Cochlea, 320, 321f Codeine, 1015t Cognition, radiation-related effects on, 837, 968, 1001–1002 Colon anatomy of, 560 cancer of. See Colorectal cancer; Rectal cancer

Colon (Continued) familial adenomatous polyposis of, 562 polyps of, 561 removal of, 566–567 radiation effects on, 560–561, 765–766, 765f, 821–822, 822t acute, 560–561 late, 561 Colonoscopy, 567–568 after colorectal cancer resection, 569–570 guidelines for, 568–569 Colorectal cancer, 561–571 anatomic distribution of, 561 antiangiogenic factors in, 565 bevacizumab in, 565 cetuximab in, 122 cyclooxygenase 2 in, 564 diet and, 563 early diagnosis of, 92 epidemiology of, 561–564, 565f epidermal growth factor receptor in, 564–565 etiology of, 561–564 familial adenomatous polyposis and, 562 genetic factors in, 562, 566 hereditary nonpolyposis, 562, 566 hormone replacement therapy and, 563 incidence of, 565f inflammatory bowel disease and, 563–564 molecular targets in, 564–565 nonsteroidal anti-inflammatory drugs and, 562–563 proximal, 596–597 adjuvant chemotherapy in, 597 adjuvant radiation therapy in, 596–597 race and, 561 radiation therapy in adjuvant, 596–597 intraoperative, 594–595 palliative, 1030 preoperative, 1030 risk factors for, 561–564, 563b risk-reduction strategies in, 566–567 screening for, 565–571 barriers to, 566 guidelines for, 568–569 staging of, 571–573, 572b endorectal sonography in, 571–573, 573f imaging modalities in, 571 Complete response, 16 Compton effect, 4, 5f Computed tomography (CT) in auricular cancer, 323–324 in bladder cancer, 640–641, 641f in bone cancer, 919 in bone lymphoma, 922–923 in cervical cancer, 741 in chemodectoma, 328, 329f in chordoma, 923 in colorectal cancer, 571 after colorectal cancer resection, 570–571 cone-beam, 54, 55f diaphragmatic motion during, 54, 54f in endometrial cancer, 776–777 in esophageal cancer, 466–467, 468–469f, 485, 485f in gastric cancer, 497 helical, in treatment planning, 53–54, 54f in hypopharyngeal cancer, 288, 288f in-room, 73–75, 75f in laryngeal cancer, 288 in lung cancer, 51–52, 51–52f, 74–75, 76f, 436

Computed tomography (CT) (Continued) in oral cavity cancer, 254–257, 254t, 255–257f in pancreatic cancer, 517 of parotid gland, 170f in parotid gland adenoma, 173f in parotid gland cancer, 172–173f in prostate cancer, 74, 75f, 683, 690, 690f in radiation-induced injury, 414–415 in radiation pneumonitis, 429f, 430 on rails, 64f, 74, 75f in respiratory motion measurement, 68–69 in soft tissue sarcoma, 944–945 of submandibular gland, 171f in testicular cancer, 662–664 in tongue base cancer, 230f in tonsillar region cancer, 230f in treatment planning, 53–56, 54–55f Conduction, cardiac, radiation effects on, 417 Conformal radiation therapy, 1036–1037 vs. intensity-modulated radiation therapy, 1037 vs. two-dimensional therapy, 1037 Conjunctiva, melanoma of, 315 Convolution algorithm, dose calculation, 57–58 Cornea radiation effects on, 311 Corniculate cartilage, 282–283 Coronary arteries radiation effects on, 414, 417–418 stenosis of, radiation prevention of, 419–420 Corticosteroids in orbital pseudotumor, 313 in radiation pneumonitis, 432 Cowden’s syndrome, 334 CpG oligodeoxynucleotides, 131–132, 132f Cranial radiation therapy in leukemia, 970–971, 971f, 998, 1001 in marrow/stem cell transplantation, 998 in non–small cell lung cancer, 444–445 in small cell lung cancer, 448–449, 448t Craniopharyngioma, 851–852, 852f prognosis for, 852 radiation therapy in, 852 Cranium, radiation effects on, 968, 969f Creatinine clearance, renal irradiation and, 536 Cricoid cartilage, 282–283 Crohn’s disease anal cancer and, 608 colorectal cancer and, 563–564 Cryotherapy, in prostate cancer, 707, 716 Cryptorchidism, testicular cancer and, 658–659 CT. See Computed tomography (CT) Cuneiform cartilage, 282–283 Cutaneous lymphoproliferative disorders, CD30-positive, 901 Cyclin(s), 128–129 in oral cavity cancer, 259 Cyclin D1, in esophageal cancer, 483t Cyclin-dependent kinases, 17, 17f, 128–129 function of, 128 inhibitors of, 128–129 Cyclooxygenase(s), 126–128, 127f expression of, 126–127 inhibitors of, 127–128, 128f tumor expression of, 126–127

Index Cyclooxygenase2 in colorectal cancer, 564 in esophageal cancer, 483t Cystectomy, 633–634 partial, in bladder cancer, 635 radiation after, 638 radiation before, 637–638, 638t salvage, 636, 645 Cystitis, radiation-related, 628, 629f Cystoscopy, in cervical cancer, 741 Cytokeratin, in soft tissue tumor, 941 Cytotoxic viruses, 38

D

Death-regulating gene, 8 Deformable image registration, 75–76, 77f Deoxyribonucleic acid double-strand breaks in, 11, 11f, 105 radiochemical lesions in, 10–11, 10f, 105 single-strand breaks in, 11, 11f structure of, 11, 11f Dermatofibrosarcoma protuberans, 959–960 Dermis, 141 Desipramine, 1015t Desmin, in soft tissue tumor, 941 Desmoid tumor, 959, 959f Dexamethasone, 1015t Diabetes insipidus, after craniopharyngioma treatment, 852 Diarrhea, radiation-related in cervical cancer, 764–765 in ovarian cancer, 821 in rectal cancer, 591–592 Diet breast cancer and, 357t, 358 colorectal cancer and, 561, 563 gastric cancer and, 496 pancreatic cancer and, 517 prostate cancer and, 679 Diethylstilbestrol, vaginal cancer and, 799 Diffuse large B-cell lymphoma, 894–896 advanced, 895 chemotherapy in for advanced disease, 895 for localized disease, 894–895 localized, 894–895 mediastinal, 895–896 prognosis for, 893f, 894 radiation therapy in for advanced disease, 895 dose-response in, 895 for localized disease, 894–895 Dilution assays, 16 Docetaxel, radiation therapy with, 107–108, 107–108f Dose. See Radiation dose Dose-response curve, 94, 94f for clonogenic cell survival, 13–14, 14f, 94, 94f for functional end points, 14, 14f linear energy transfer and, 34, 34f Doubling time, 19 Doxorubicin, cardiotoxicity of, 417 Drugs bladder cancer and, 630 chemotherapeutic. See Chemotherapy; Radiation therapy and chemotherapy Dry eye, 311 Ductal Carcinoma in situ (DCIS). See Breast cancer, ductal in situ

Dysgerminoma, 827–828 chemotherapy in, 828 oophorectomy in, 827–828 radiation therapy in, 828 Dysphagia, in esophageal cancer, 478–482

E

E-cadherin, in esophageal cancer, 483t Ear(s) anatomy of, 320, 321f cancer of, 323–326 clinical presentation of, 323 evaluation of, 323–324 in external auditory canal, 323 in middle ear, 323 in outer ear, 323 pathology of, 323 prognosis for, 324, 324–325t spread of, 323 staging of, 324–325, 324t treatment of, 325–326 radiation therapy in, 325–326, 327–328f results of, 324–325t, 326 surgery in, 324–326, 325f, 325t in T1 disease, 325, 325f in T2 disease, 325–326 in T3 disease, 326 in T4 disease, 326 chondronecrosis of, 321 inner, 320, 321f radiation effects on, 322, 322f middle, 320, 321f radiation effects on, 321–322 outer, 320, 321f radiation effects on, 320–321 radiation effects on, 320–322 Eccrine carcinoma, 146 treatment of, 152 Electrocardiography in pericarditis, 415–416 in radiation-induced injury, 414–415 Electron, 4 Electron beam therapy, in orbital ­lymphoma, 316, 316f Embryonal carcinoma, 660. See also ­Testicular cancer Embryonal CNS tumors, 838–841, 838b Embryonal rhabdomyosarcoma, 800, 977 End-of-life care, 1008. See Palliative care Endodermal sinus tumor, 660 Endometrial cancer, 774. See also Uterus adenocarcinoma, 775–776 cervical invasion of, 778 chemotherapy in, 785–787, 790–791 clear cell carcinoma, 776 epidemiology of, 774 evaluation of, 776–777 hereditary nonpolyposis colorectal cancer and, 774 hysterectomy in, 780–781 imaging in, 776–777 lymph nodes in, 775, 775f, 779–780, 779t, 780f lymphadenectomy in, 781 metastatic, 775, 790–791 lymphatic, 779–780, 779t, 780f mucinous, 776 myometrial invasion of, 777–778, 779t ovarian cancer and, 780

1055

Endometrial cancer (Continued) papillary serous, 776 pathology of, 775–777, 776b peritoneal washing in, 780 prognosis for, 777–780, 792 age and, 777 cervical invasion and, 778 histology and, 777 lymph node involvement and, 779–780, 779t, 780f myometrial invasion and, 777–778, 779t peritoneal tumor cells and, 780 tumor size and, 779 radiation therapy in, 781–789, 792, 792f adjuvant, 781–789 complications of, 775 cystitis with, 628 for inoperable disease, 789–790, 790f postoperative, 782–785, 783f vs. chemoradiation therapy, 785 vs. chemotherapy, 785–787 extended-field, 788–789 for high-risk disease, 785–786 intensity-modulated, 788 pelvic, 782–783, 788, 788f for stage III disease, 786–787, 787t vaginal, 783–785, 784t, 789 whole-abdomen, 787–789 preoperative, 782 in recurrent disease, 790, 791t vaginal brachytherapy for, 783–785, 784t, 789 sarcoma, 776 spread of, 775, 775f staging of, 777 surgery in, 780–781 treatment of, 780–791 tumor size in, 779 Endorectal sonography, in rectal cancer, 571–573, 573f Endoscopic sonography, in esophageal ­cancer, 466, 468f, 481 Endoscopy, in pancreatic cancer, 517–518 Endostatin, 124 postradiation levels of, 123–124 Eosinophilic granuloma, 928, 929f Ependymoma pediatric, 848–849, 848–849f spinal cord, 860 Epicardium, 414 Epidermal growth factor receptor, 118–122 cellular radiation sensitivity and, 119–120, 120f in colorectal cancer, 564–565 in epithelial neoplasms, 119 in esophageal cancer, 482, 483t function of, 118–119 in head and neck squamous cell ­carcinoma, 119–120, 121f, 166 inhibitors of, 120–122 mutations in, 122 nuclear, 119 oropharyngeal carcinoma and, 228 radiation-induced activation of, 119 structure of, 118–119 Epidermis, 141 radiation-related necrosis of, 142 Epididymis, 653, 654f Epididymitis, vs. testicular cancer, 660 Epiglottic cartilage, 282–283 Epiglottis, 282, 283f

1056

Index

Epstein-Barr virus infection gastric cancer and, 497 nasopharyngeal cancer and, 210 ERBB2, in breast cancer, 360 Erectile dysfunction, prostate cancer ­treatment and, 707 Erlotinib, in pancreatic cancer, 529–530 Erythema, radiation-induced, 411–412 Erythrocytes, radiation effects on, 994f, 995 Erythroplakia, 224, 259 Erythropoietin, in cervical cancer, 744 Esophageal cancer, 463–464 adenocarcinoma, 463–464 celiac adenopathy in, 481, 481f clinical evolution of, 464–465, 467f dysphagia in, 478–482 epidemiology of, 463–464 etiology of, 464 evaluation of, 465–468 bronchoscopy in, 468 computed tomography in, 466–467, 468–469f endoscopic sonography in, 466, 468f, 481 esophagography in, 466, 468f molecular markers in, 482, 483t positron emission tomography in, 467–468, 469f, 482 incidence of, 464 metastatic, 464–465, 468t lymphatic, 464, 467f, 481, 481f mortality from, 464–465 palliation in, 478–482 prognosis for, 480–482, 480t, 481f proton therapy for, 484, 484f, 1044–1045 radiation therapy in, 471–478 adverse effects of, 478–480 brachytherapy for, 475, 1029 cardiac effects of, 418, 480 chemotherapy and surgery with, 477–478 chemotherapy with, 113, 472–477, 473–474f, 473t, 476–477f adverse effects of, 462 pathologic response to, 481–482, 482f three-step strategies in, 474–475, 474t dose for, 486, 486f esophagitis with, 479–480 intensity-modulated, 484 nodal, 486 pathologic response to, 481–482 postoperative, 471 preoperative, 472–474 proton therapy for, 484, 484f, 1044–1045 pulmonary effects of, 479, 479f target volume for, 485–486, 485–486f techniques of, 482–486 three-step strategies in, 474–475, 474t squamous cell, 463–464 staging of, 468–469, 470b surgery in, 469–471, 470t, 471f chemotherapy before, 470–471, 471f complications of, 479 radiation therapy after, 471 radiation therapy and chemotherapy before, 472–475, 473–474f, 473–474t, 477–478

Esophageal cancer (Continued) adverse effects of, 478–480, 479f pathologic response to, 481–482, 482f pulmonary effects of, 479, 479f radiation therapy before, 472 Esophagectomy, 469–471, 470t chemotherapy before, 470–471, 471f pulmonary complications of, 479 radiation therapy after, 471 radiation therapy and chemotherapy before, 472–475, 477–478 pathologic response to, 481–482, 482f radiation therapy before, 472 Esophagitis, radiation-induced, 431–432, 479–480 Esophagography, in esophageal cancer, 466, 468f Esophagus, 461 anatomy of, 461–462, 462–465f, 467f blood supply to, 461–462, 465f cancer of. See Esophageal cancer cervical, 461, 464f cross-sectional anatomy of, 461, 464f intrathoracic, 461 layers of, 461 lower, 461, 464f lymphatic system of, 462, 467f midthoracic, 461, 464f radiation effects on, 462–463, 479–480 acute, 462 late, 463 venous system of, 461–462, 465f wall of, 461 Esthesioneuroblastoma, 185 clinical features of, 187 incidence of, 187 staging of, 187 treatment of, 192–194, 193–194t Estrogen ovarian cancer and, 813 prostate cancer and, 679 Estrogen replacement therapy, breast cancer and, 357 Etanidazole, radiosensitizing effects of, 31 Ethmoid air cell cancer clinical manifestations of, 186–187 natural history of, 185 radiation therapy in, 191–192, 198–200, 201f dose for, 198–199 intensity-modulated, 200, 202f setup for, 199–200 target volume for, 198–199 side effects of, 204t staging of, 187, 188b surgery in, 191–192 treatment of, 189–194 Ethmoid sinuses, 184 Eustachian tube, 207, 320, 321f Exercise, breast cancer and, 358 External auditory canal, 320, 321f cancer of, 323. See also Ear(s), cancer of Eye(s) anatomy of, 309, 309f melanoma of, 313–315, 314b radiation effects on, 309–312 shield for, 316, 316f Eyelashes, radiation effects on, 310 Eyelids carcinoma of, 147, 149t radiation effects on, 310 sebaceous carcinoma of, 146

F

Facial nerve paralysis, in parotid gland tumor, 172 Familial adenomatous polyposis, 562, 566 screening for, 569 Farnesyltransferase, inhibitors of, 129 FAS/FAS ligands, in esophageal cancer, 483t Fecal occult blood test, 567–569 Fentanyl, transdermal, 1015t Fertility, male, stem cell transplantation for, 658 Fetus, radiation effects on, 918 Fibroblast growth factor, radioprotective effects of, 917 Fibrosis abdominopelvic, 821 capillary, 413 pulmonary, 425, 425f Fibrous histiocytoma, malignant, 924–925, 925f, 958, 958f Flavopiridol, 128–129, 129f Floor of mouth cancer, 264, 267f, 273. See also Oral cavity cancer Flow cytometry, in prostate cancer, 686 Fludarabine, radioenhancing effects of, 106–107, 107f Fluorescence in situ hybridization, 11–12, 12f Fluoride, dental application of, 277 5-Fluorouracil. See also Chemotherapy in anal cancer, 612–615, 614–615t gene delivery of, 130 in pancreatic cancer, 526–528, 527t in rectal cancer, 579–580, 588–589 Follicular lymphoma, 896–898 chemotherapy in for stage III disease, 898 for stage IV disease, 898 prognosis for, 893f, 896 radiation therapy in dose-response analysis in, 898 for stage I or II disease, 896–897, 897f for stage III disease, 898 rituximab immunotherapy in, 898 Fractionation, 20–26 accelerated, 24 accelerated repopulation and, 25–26, 27f ARCON protocol for, 25 CHART trial of, 24 early-responding tissue and, 22t, 22f historical perspective on, 3 hyperfractionation for, 24 large-dose, 23 late-responding tissue and, 22, 22f, 22t multiple, 23–25 nominal standard dose for, 23 potentially lethal damage repair and, 20–21 reassortment and, 21 relative biologic effectiveness and, 34, 34f reoxygenation and, 21, 29, 30f repair and, 20–21 repopulation and, 21–22 RTOG trial 9003 of, 24–25 Strandqvist plot for, 23 sublethal damage repair and, 20, 21–22f treatment time and, 25, 26f tumor control and, 23 Fracture pathologic, 928–929, 1020–1021, 1021f vertebral, 1025, 1025f Free radicals, in DNA injury, 10–11, 10f

Index Freund’s adjuvant, 131 Functional imaging, 87 Fungal infection, oral cavity, ­ radiation-­related, 275

G

Gallbladder cancer, 545–555 radiation therapy in, 548–549 brachytherapy for, 551–552 neoadjuvant, 549–550 palliative, 550–553 recurrence of, 548–549 risk factors for, 539 staging of, 540b surgery in, 545–546, 548 Gamma index, 61–62 Gamma rays, 4 Gardner’s syndrome, 334, 941 Gastric cancer chemotherapy in, 500–504 historical perspective on, 500 induction, 502 intraperitoneal, 506 palliative, 504 postoperative, 500–502 preoperative, 502–503 supportive care with, 504 in unresectable disease, 503–504 clinical presentation of, 497 diet in, 496 diffuse type of, 496 environmental factors in, 496 epidemiology of, 496 etiology of, 496–497 evaluation of, 497 host-related factors in, 496 infectious factors in, 497 intestinal type of, 496 mortality in, 500 pathology of, 498 peptic ulcer disease and, 496 prognosis for, 498 radiation therapy in, 504–505, 505–506f adverse effects of, 493–494 conformal, 506 extended-field, 506–507 fractionation in, 505 hepatic-directed, 506–507 neutron therapy and, 506 palliative, 504 peritoneal-directed, 506–507 postoperative, 500–502 preoperative, 502–503 radiosensitizers with, 505–506 supportive care with, 504 in unresectable disease, 503–504 screening for, 497 spread of, 497, 498f staging of, 498 supportive care in, 504 surgery in, 498–500 chemotherapy after, 500–502 chemotherapy before, 502–503 unresectable, 503–504 Gastrointestinal tract. See Colon; Small intestine; Stomach Gated beam delivery, 69, 69f Gatekeeper genes, 9 Gefitinib, in rectal cancer, 580 Gemcitabine in pancreatic cancer, 527t, 528 radioenhancing effects of, 106–107

Gene therapy, 38–39, 129–131 gene silencing, 15–17 immunomodulatory, 38 oncolytic viruses for, 38, 130–131 radiation-activated, 38–39 radiosensitizing drug delivery with, 130 replacement, 130 suicide gene, 38 suppressive, 131 tumor suppressor gene, 38 Germ cell(s) primordial, 653–654 radiation effects on, 39, 655–656 Germ cell tumor extragonadal, 671 gonadal. See Dysgerminoma; Testicular cancer intracranial, 849–851, 851t Giant cell tumor, 926–928 malignant transformation of, 927 metastases from, 927 radiation therapy in, 927–928 surgery in, 927 Gingival cancer, 264, 273. See also Oral cavity cancer Gleason score, in prostate cancer, 680–685, 685f, 710–711 GliaSite Radiation Therapy System, 78–79, 79f Glioblastoma multiforme, 854–856, 854f antiangiogenic agents in, 125 prognosis for, 856 radiation therapy in, 856, 856f Glioma, 854–856, 854f. See also ­Astrocytoma brainstem, 846–848, 847f prognosis for, 847–848 radiation therapy in, 847 chemotherapy in, 856 MGMT in, 856 optic, 844–845 prognosis for, 856 radiation therapy in, 854–855, 1040 high-LET particles for, 855 interstitial implants for, 855 radiosensitizers for, 855 technique of, 856, 856f time-dose regimens for, 855–856 spinal cord, 860 surgery in, 854 Gliomatosis cerebri, 843, 846f Glomus body, 327–328 Glomus tumor. See Chemodectoma Glottic cancer airway obstruction with, 284, 285f clinical manifestations of, 287 radiation therapy in, 295–297, 296–298f complications of, 305–306 dose and fractionation for, 300–301 dose distribution for, 296 fields for, 295–297, 296–297f intensity-modulated, 297–298f, 301 results of, 302–304, 303–304t in T1 and T2 tumors, 292, 302 in T3 tumors, 292, 302–304, 304t in T4 tumors, 292, 304 three-field technique for, 296 spread of, 284, 285f staging of, 289–290, 289b surgery in, 292 T1 and T2, 292, 302

1057

Glottic cancer (Continued) T3, 292, 302–304 T4, 292, 304 treatment of, 292. See also Glottic cancer, radiation therapy in surgery in, 292 Glottis. See also Glottic cancer anatomy of, 282, 283–284f Glut1, in esophageal cancer, 483t Gonad(s) radioiodine therapy effect on, 346 total-body irradiation effects on, 1000 tumors of. See Ovarian cancer; Testicular cancer Gonadoblastoma, 671 Gonocytes, 653–654 Granuloma, eosinophilic, 928, 929f Granulosa cell tumors, 827 Graves’ ophthalmopathy, 312–313 radiation therapy in, 312–313 Growth, radiation effects on, 967–968, 1000–1001 Growth delay assay, 16 Growth fraction, 19 Growth hormone, radiation-related ­deficiency of, 916, 968 Growth plate, 915 radiation effects on, 915–916

H

Hair follicles, radiation effects on, 142 Hard palate cancer, 264, 273 Head and neck cancer, 161. See also Head and neck squamous cell cancer and at specific sites molecular biomarkers in, 165–166 radiation therapy in altered fractionation regimens in, 161–162, 162t, 167 high-precision, 162–163 Head and neck squamous cell cancer biomarkers for, 166 epidermal growth factor receptor in, 119–120, 121f locally advanced, radiation therapy and chemotherapy in, 163–165, 164–165t, 1029–1030 radiation therapy in altered fractionation regimens in, 161–162, 162t cetuximab with, 121–122, 166, 166–167t chemotherapy with, 113, 163–165, 164–165t intensity-modulated, 162 Hearing loss, radiation-related, 215, 322, 322f Heart anatomy of, 414 doxorubicin effects on, 417 radiation-induced injury to, 414–420 assessment of, 414–415 breast cancer treatment and, 99, 418 esophageal cancer treatment and, 418, 480 exposed volume and, 415 follow-up for, 420 fraction size and, 415 Hodgkin lymphoma treatment and, 418–419, 882 postmastectomy, 394–395 prevention of, 420

1058

Index

Heart (Continued) seminoma treatment and, 419 tolerance dose in, 415, 420 total dose and, 415, 415f total-body irradiation effects on, 1000 Heart valves, radiation-induced injury to, 417 Heavy ions, 5 Heavy metal exposures, bladder cancer and, 630 Helical tomotherapy, 61 Helicobacter pylori infection gastric cancer and, 497 MALT lymphoma and, 898, 898f Hemangioendothelioma, 925–926 Hemangiopericytoma, 960 Hematuria, radiation-induced, 629 Hemoglobin, in cervical cancer treatment, 744 Hepatitis, radiation, 495–496, 517 Hepatitis B, hepatic radiation therapy and, 536 Hepatitis C, hepatocellular carcinoma and, 538 Hepatobiliary cancer, 538–539. See also Cholangiocarcinoma; Gallbladder ­cancer; Hepatocellular ­carcinoma epidemiology of, 538–539 evaluation of, 539–540 radiation therapy in acute effects of, 535–536 chemotherapy and, 538 late effects of, 536 normal-tissue complication probability model and, 536–538, 537–538f risk factors for, 539 staging of, 539–540, 540–541b surgery in, 539, 539f Hepatoblastoma, 982 Hepatocellular carcinoma, 540–545 in children, 982 epidemiology of, 538–539 evaluation of, 539–540 liver-directed therapy in, 540–542 monoclonal antibody therapy in, 545 mortality in, 538–539 percutaneous ethanol injection in, 540–541 radiation therapy in, 542–545 acute effects of, 535–536 chemotherapy with, 544 conformal, 542–543, 543f heavy ion therapy in, 543–544 intratumoral, 544–545 proton therapy in, 543–544 respiratory motion during, 542 transarterial chemoembolization with, 542–543 viral hepatitis and, 536 radiofrequency ablation in, 540–541 radionuclide therapy in, 544–545 staging of, 539–540 surgery in, 540 transarterial chemoembolization in, 541–542 radiation therapy with, 542–543 transplantation in, 540 vascular endothelial growth factor in, 544 HER2/NEU in breast cancer, 360 in esophageal cancer, 482, 483t Hereditary nonpolyposis colorectal cancer, 562, 566 endometrial cancer and, 774 screening for, 569

Herpes simplex virus infection, oral cavity, radiation-related, 275 Heterotopic ossification, 917 classification of, 917–918 prevention of, 917–919 risk of, 917 Hoarseness in hypopharyngeal cancer, 288 in laryngeal cancer, 287–288 Hodgkin lymphoma, 879–889 chemotherapy in vs. radiation therapy, 883 radiation therapy after, 886–887 toxicity of, 883–884 clinical presentation of, 880 epidemiology of, 880 etiology of, 880 lymphocyte-predominant, 887–888 chemotherapy in, 888 prognosis for, 888 second malignancy and, 888 pathology of, 880 prognosis for, 881, 881f radiation therapy in cardiac effects of, 417–419, 882 after chemotherapy, 886–887 vs. chemotherapy, 883 dose in, 882 involved-field, 889 mantle field, 888–889 para-aortic field in, 889, 889f second malignancy after, 42, 333, 882 splenic field in, 889, 889f stage I and II disease, 880–881, 881f technique in, 888–889 toxicity of, 882 volume in, 881–882 risk factors for, 880 staging of, 880, 881b laparotomy for, 883 subdiaphragmatic, 887 treatment of. See also Hodgkin lymphoma, radiation therapy in in advanced disease, 885–887 in less-advanced stage III, 884–885 in stage I and II disease, 880–884, 881f in unfavorable stage I and II, 884–885 Hormonal therapy. See also Androgen ­ablation in ductal in situ breast cancer, 367 in early-stage breast cancer, 378–380 Hormone replacement therapy breast cancer and, 357 colorectal cancer and, 563 Hospice care. See Palliative care Hot potato voice, 287 Human chorionic gonadotropin (hCG) in germ cell tumor, 850 in testicular cancer, 664 Human immunodeficiency virus (HIV) infection anal cancer treatment and, 615 cervical cancer and, 734 non-Hodgkin lymphoma and, 902 Human papillomavirus (HPV) infection anal cancer and, 607–608 cervical cancer and, 734, 744 oral cavity cancer and, 259–260 oropharyngeal carcinoma and, 227–228 vaccine against, 734 vaginal cancer and, 799 vulvar cancer and, 799

Human ribosomal protein, in esophageal cancer, 483t Hürthle cell tumor, 335–336. See also ­Thyroid cancer Hydromorphone, 1015–1016t Hydroxyzine, 1015t Hypercalcitoninemia, in medullary thyroid cancer, 342–343 Hyperfractionation, 24 Hypertension, radiation-induced, 413 Hyperthermia, 35–36 Hypogonadism, total-body irradiation and, 1000 Hypopharyngeal cancer clinical manifestations of, 288 evaluation of, 288 pathology of, 287 radiation therapy in complications of, 305–306 dose and fractionation for, 302 fields for, 299–300, 299–300f intensity-modulated, 299–300, 300f, 302 results of, 305 spread of, 285–286 lymphatic, 287 staging of, 288–290, 289b treatment of, 293–294 in advanced disease, 291–292, 291t chemotherapy in, 293–294 in early-stage disease, 290–291 surgery in, 293 Hypopharynx, anatomy of, 283–284, 284f Hypopituitarism, radiation-related, 215 Hypothyroidism, radiation-related, 215, 333, 968 Hypoxia, tumor, 29–31, 29–30f carbogen breathing effects on, 31 chemical radiosensitizer effects on, 30–31 chemotherapy effects on, 108 development of, 108 hyperbaric oxygen effects on, 30 mitomycin effects on, 108 nicotinamide effects on, 31 taxane effects on, 108, 109f tirapazamine effects on, 31–32, 31f, 108 tumor progression and, 29–30 Hysterectomy in cervical cancer, 744–747, 746t, 750–751, 750f in endometrial cancer, 780–781 in ovarian cancer. See Ovarian cancer, surgery in

I

Ibuprofen, 1015t Image-guided adaptive radiation therapy, 75–76, 77f Image-guided radiation therapy, 50, 70–76, 70–71f, 86 on-board kilovoltage x-ray imaging for, 71–72, 72f sonographic, 72–73, 73–74f three-dimensional, 73–75, 75–76f Immobilization, in treatment planning, 56–57, 57f, 57t Immune system, radiation effects on, 19–20 Immunomodulatory gene therapy, 38 Immunotherapy, 131–132 in bladder cancer, 633–634 in follicular lymphoma, 898 mechanisms of, 131 molecular, 131–132, 132f

Index In-room imaging, 70–76, 70–71f, 86 image-guided adaptive radiation therapy and, 75–76, 77f on-board kilovoltage x-ray imaging for, 71–72, 72f sonographic, 72–73, 73–74f three-dimensional, 73–75, 75–76f In vivo/in vitro assays, 16–17 Indomethacin, 1015t Infant. See also Children brain tumors in, 853–854, 853f Infection Epstein-Barr virus gastric cancer and, 497 nasopharyngeal cancer and, 210 Helicobacter pylori gastric cancer and, 497 MALT lymphoma and, 898, 898f human immunodeficiency virus anal cancer treatment and, 615 cervical cancer and, 734 non-Hodgkin lymphoma and, 902 human papillomavirus anal cancer and, 607–608 cervical cancer and, 734, 744 oral cavity cancer and, 259–260 oropharyngeal carcinoma and, 227–228 vaccine against, 734 vaginal cancer and, 799 vulvar cancer and, 799 oral cavity, radiation-related, 275 urinary tract, bladder cancer and, 630 vulvar, radiation-related, 806 Inflammatory bowel disease anal cancer and, 608 colorectal cancer and, 563–564 Inflammatory breast cancer, 389–390 breast-conservation therapy in, 389 mastectomy in, 389 neoadjuvant chemotherapy in, 389 radiation therapy in, 389–390 Intelligence quotient, radiation effects on, 968, 1001 Intensity-modulated radiation therapy (IMRT), 50, 1037 in anal cancer, 618 in bladder cancer, 643, 643f in cervical cancer, 754, 755f in endometrial cancer, 788 in esophageal cancer, 484 in ethmoid air cell cancer, 200, 202f field-in-field planning in, 59, 60f fixed-beam attenuation in, 59–61 in glottic cancer, 297–298f, 301 in head and neck squamous cell ­carcinoma, 162 helical tomotherapy in, 61 in hypopharyngeal cancer, 299–300, 300f, 302 inverse planning in, 58–59 in maxillary sinus cancer, 198t, 201, 203–204f in nasal cavity cancer, 200, 202f in nasopharyngeal cancer, 162–163, 213, 215, 216f in oral cavity cancer, 270 in oropharyngeal cancer, 162–163 in oropharyngeal wall cancer, 241 in pancreatic cancer, 520–521 portal verification for, 70–71, 71f in prostate cancer, 688–690, 688f

quality assurance of, 61–62, 61f in salivary gland tumors, 171–172, 179, 179–180f in sinonasal cancer, 198t, 205 in soft palate cancer, 241, 246f in submandibular gland cancer, 179, 180f in supraglottic cancer, 298, 299f, 301 vs. three-dimensional conformal radiation therapy, 1037 in thyroid cancer, 345, 345f in tongue base cancer, 232–233, 232f, 241, 245f in tonsillar region cancer, 233, 241, 241–242f in vaginal cancer, 808 Interferon-alpha, in melanoma, 151 Internal mammary lymph nodes, 355, 355f postmastectomy radiation of, 384–385, 385–386t, 394 Interstitial radiation therapy, 98. See also Brachytherapy in anal cancer, 619 in bladder cancer, 638–639 in cervical cancer, 761–762, 762f in glioma, 855 Intestine. See Colon; Small intestine Intraoperative radiation therapy, 98 in cholangiocarcinoma, 552 in colorectal cancer, 594–595 in rectal cancer, 594–595 Intraoral cone therapy, in oral cavity cancer, 266 Intraperitoneal radiocolloid therapy, after ovarian cancer, 818, 818t Inverse planning, 58–59 cost function (objective function) in, 58 dose-volume constraint in, 58 equivalent uniform dose in, 58 Iodine-131– Lipiodol injection therapy, in hepatocellular carcinoma, 545 Iodine-125 plaques, in ocular melanoma, 80, 1038–1039 Iodine-125 seeds, in prostate cancer, 691–694, 692–693f Ionization, 4 Ionizing radiation, 4 Iridium-192 therapy in anal cancer, 619 in prostate cancer, 693–694 Irinotecan, in gastric cancer, 505–506 Isobologram analysis, 103–104, 104f Isotretinoin, in leukoplakia, 265

J

JAK/STAT pathway, 119

K

Keratitis, radiation-induced, 311 Ketoprofen, 1015t Ketorolac, 1015t Ki-67 in esophageal cancer, 483t in prostate cancer, 685–686 Kidney(s) clear cell sarcoma of, 972 radiation effects on, 536 rhabdoid tumor of, 972 total-body irradiation effects on, 1000 Wilms’ tumor of. See Wilms’ tumor

L

1059

L-778123, 129 Labyrinth, 320, 321f Lacrimal gland, radiation effects on, 311 Lactation breast cancer and, 358–359 after breast cancer radiation therapy, 356 Large intestine. See Colon Laryngeal cancer, 282. See also ­Hypopharyngeal cancer clinical manifestations of, 287–288 evaluation of, 288, 288f glottic airway obstruction with, 284, 285f clinical manifestations of, 287 radiation therapy in, 295–297, 296–298f dose and fractionation for, 300–301 results of, 302–304, 303–304t spread of, 284, 285f T1 and T2, 292 T3, 292 T4, 292 treatment of, 292 pathology of, 287 radiation therapy in, 294–302 beam energies in, 294 complications of, 305–306 dose and fractionation in, 300–302 in early-stage disease, 291 fields for, 295–300 in glottic cancer, 295–297, 296–298f, 300–304, 304t results of, 302–305, 303–304t setup for, 295, 295f simulation for, 294–295, 294f in supraglottic cancer, 297–298, 298–299f, 301–302, 304–305 in situ, 287 spread of, 284–287, 285–286f distant, 287 lymphatic, 286–287 staging of, 288–290, 289b subglottic clinical manifestations of, 287–288 spread of, 285 treatment of, 293 supraglottic clinical manifestations of, 287 early-stage, 292–293 radiation therapy in, 297–298, 298–299f dose and fractionation for, 301–302 results of, 304–305 spread of, 285, 286f T3, 293 T4, 293 treatment of, 292–293 treatment of, 290–294. See also Laryngeal cancer, radiation therapy in in advanced disease, 291–292, 291t chemotherapy in, 291–292, 291t cost of, 291 in early-stage disease, 290–291 in glottic cancer, 292 patient preferences for, 292 radiation therapy and chemotherapy in, 114 in subglottic cancer, 293 in supraglottic cancer, 292–293 surgery in, 290–291, 291t Laryngoscopy, in laryngeal cancer, 288

1060

Index

Larynx anatomy of, 282–283, 283–284f cancer of. See Laryngeal cancer cartilages of, 282–283 regions of, 282 Law of Bergonié and Tribondeau, 3 Left ventricular ejection fraction, radiation effects on, 417–418 Lens, radiation effects on, 311 Lentigo maligna melanoma, 144, 149 Lethal dose, 19 Letrozole, in early-stage breast cancer, 379 Leukemia bone marrow transplantation for, 995–996. See also Bone marrow transplantation metastatic prostate cancer treatment and, 919 pediatric, 969–971 biologic features of, 969–970 clinical features of, 969–970 cranial radiation therapy in, 968, 998 meningeal, 970–971, 971f neuropsychological alterations in, 968 testicular, 971 radiation-related, 878–879, 879f, 919 radioiodine-related, 346, 919 treatment of, 879 Leukocytes, radiation effects on, 994, 994f Leukoencephalopathy, 837, 968, 969f Leukoplakia, 224, 225f, 259, 265 Levorphanol, 1015t Leydig cell, radiation effects on, 657–658 Leydig cell tumor, 671 Lhermitte’s sign, 835–836 Li-Fraumeni syndrome, 359, 941 Linear energy transfer, 34 dose-response curve and, 34, 34f oxygen enhancement ratio and, 34f, 35 relative biologic effectiveness and, 34, 34f Linitis plastica, 498 Lip cancer, 262t, 263, 264f, 272. See also Oral cavity cancer 131I-Lipiodol injection therapy, in ­hepatocellular carcinoma, 545 Liposarcoma, 958–959 Liquid source brachytherapy, in brain tumor, 78–79, 79f Liver anatomy of, 534–535 blood supply of, 534 cancer of. See Hepatobiliary cancer; Hepatocellular carcinoma functional anatomy of, 534 functional testing of, 535 histology of, 534–535 lobes of, 534 prophylactic radiation to, in pancreatic cancer, 530 radiation effects on, 506–507, 517, 535–538 acute, 535–536 chemotherapy and, 538 late, 536 normal-tissue complication probability model for, 536–538, 537–538f total-body irradiation and, 999 transplantation of in cholangiocarcinoma, 552–553 in hepatocellular carcinoma, 540 Local-regional tumor control, 92

Lumpectomy. See Breast cancer, ductal in situ, breast-conservation therapy in; Breast cancer, early-stage, ­breast­conservation therapy in Lung(s) esophagectomy-related disorders of, 479 functional sections of, 424 lymphatic drainage of, 434, 434f radiation injury to, 424–433, 968 clinical manifestations of, 429 in esophageal cancer treatment, 479, 479f exudative phase of, 425, 425f fibrotic phase of, 425, 425f functional effects of, 430–432, 431f latent phase of, 425 molecular mechanisms of, 425–427, 426f radiologic changes in, 429–430, 429f total-body irradiation and, 999–1000 treatment of, 432–433, 432f, 433t volume irradiated and, 427–429, 427–428f Lung cancer, 433–450 adenocarcinoma, 433, 434t metastatic, 434 radiography in, 434–435, 435t apical sulcus, 449–450, 449f carcinoid, 433 classification of, 433–434 computed tomography in, 51–52, 51–52f, 74–75, 76f, 436 evaluation of, 435–436 incidence of, 433 inoperable, 450–451 Karnofsky score in, 450, 451t palliative radiation therapy in, 450–451 prognosis for, 450, 451t radiation therapy in, 438–445, 439t large cell, 433, 434t metastatic, 434 radiography in, 434–435, 435t lymphatic drainage in, 434, 434f, 435t, 438f metastatic brain, 434, 435t, 444–445, 448–449, 448t lymphatic, 434, 435t palliative radiation therapy in, 450–451, 1029, 1029f mortality from, 450, 451t non–small cell mortality in, 438–440, 439t prophylactic cranial irradiation in, 444–445 proton therapy in, 1043–1044, 1044f radiation therapy in, 440f Ad-p53 gene transfer with, 130 celecoxib with, 127–128 chemotherapy and, 440–442, 441–442f computed tomography in, 440, 440f cranial, 444–445 for inoperable disease, 438–445 postoperative, 436–438 preoperative, 436 targeted molecular therapy and, 442–444, 443–445f, 444t techniques of, 440, 440–441f targeted molecular therapy in, 442–444, 443–445f, 444t Pancoast’s syndrome in, 449 paraneoplastic phenomena in, 435 positron emission tomography in, 436

Lung cancer (Continued) prognosis for, 450, 451t radiation therapy in, 433–450 dose-volume constraints in, 432–433, 433t indications for, 436 in inoperable non–small cell lung ­carcinoma, 438–445, 439t irradiated lung volume in, 427–429, 427–428f palliative, 450–451, 1029 postoperative, 98, 436–438 pretreatment pulmonary function and, 430–431, 431f radiography of, 434–435, 435t small cell, 433, 434t metastatic, 434 prophylactic cranial irradiation in, 448–449, 448t radiation therapy in, 445–449 chemotherapy with, 113, 445–446, 446f cranial, 448–449, 448t fractionation of, 447–448, 447f timing of, 446–447, 447f radiography in, 434–435, 435t spread of, 434, 435t squamous cell, 433–434, 434t radiography in, 434–435, 435t staging of, 436, 437b, 438f, 439b stereotactic radiation therapy in, 65–67 delivery of, 67 fractionation for, 65 image guidance system in, 67 immobilization for, 65–66, 66f planning for, 66–67, 67t superior vena cava obstruction in, 450 surgery in, 433 radiation therapy after, 436–438 symptoms and signs of, 434 Lung colony assays, 16 Luteinizing hormone, 657–658 Lymph nodes. See also Metastases, lymphatic and at specific cancers axillary, 354, 355f, 376–378 adenocarcinoma with unknown ­primary of, 387–388 evaluation of, 387–388 radiation therapy in, 387–388, 387–388t Lymphangiography, in testicular cancer, 662 Lymphangiosarcoma, 940 Lymphoblastic lymphoma, precursor T-cell/ B-cell, 899–900 Lymphoma bone, 921–923, 922f bone marrow transplantation for, 996. See also Bone marrow transplantation Burkitt’s, 893f, 899 CNS, 858–859 in immunocompetent patients, 901–902 prognosis for, 859 radiation therapy in, 859, 859f diffuse large B-cell. See Diffuse large B-cell lymphoma follicular. See Follicular lymphoma Hodgkin. See Hodgkin lymphoma lymphoblastic, 899–900 MALT, 893f, 898–899, 898f mantle cell, 893f, 899 natural killer cell, 899 non-Hodgkin, 889–902

Index Lymphoma (Continued) AIDS-related, 902 classification of, 890, 891–892t clinical presentation of, 890–893 epidemiology of, 889–890 etiology of, 889–890 pathology of, 890–893 prognosis for, 893, 893–894t sites of, 890 staging of, 890–893 subtypes of, 892t, 894–902. See also specific lymphomas orbital, 315–316, 316f peripheral T-cell, 893f, 899 testicular, 671–672 thyroid gland, 344 Lymphomatoid papulomatosis, 901 Lynch’s syndrome, 774

M

Magnetic resonance imaging (MRI) in auricular cancer, 323–324 in bladder cancer, 631, 632f, 640 in bone cancer, 919 in bone lymphoma, 922f in brainstem glioma, 847, 847f in breast cancer, 360 in cervical cancer, 741, 741f in chemodectoma, 328, 329f in chordoma, 923 in colorectal cancer, 571 after colorectal cancer resection, 570 in endometrial cancer, 776–777 in malignant fibrous histiocytoma, 925f in maxillary sinus tumor, 186f in nasopharyngeal cancer, 209–210f, 212f in oral cavity cancer, 254–257, 254t in pancreatic cancer, 517 in prostate cancer, 681 in soft tissue sarcoma, 944 in tongue base cancer, 230f Malignant fibrous histiocytoma osseous, 924–925, 925f soft tissue, 958, 958f MALT lymphoma, 893f, 898–899, 898f Mammography, in ductal in situ breast cancer, 362, 365f MammoSite system, 79–80, 80f, 374, 374f, 376 Mantle cell lymphoma, 893f, 899 Margin internal, 52 setup, 52 Mastectomy, 367–368. See also Breast cancer, mastectomy in vs. breast-conservation therapy, 364, 367–368 prophylactic, 361 Maxillary sinus, 184, 184f Maxillary sinus cancer chemotherapy in, 196–197 clinical manifestations of, 186f, 187–189 natural history of, 185 physical examination in, 187 radiation therapy in, 200–201 complications of, 197t intensity-modulated, 198t, 201, 203–204f outcomes of, 194, 194t postoperative, 194–196, 194–196t bilateral, 201 dose for, 200, 203f external beam setup for, 200–201

Maxillary sinus cancer (Continued) ipsilateral, 201 patient care for, 201–203 side effects of, 204t target volume for, 200, 203f primary, 196, 200, 204f side effects of, 204t staging of, 188–189, 188b surgery in, 194, 194t outcomes of, 194, 194t Mechlorethamine, in mycosis fungoides, 900 Mediastinum, thymoma of, 451 Medulloblastoma infant, 853, 853f pediatric, 838–841 chemotherapy in, 839, 841 genetic factors in, 838 pathology of, 839 proton therapy in, 1041, 1041f radiation therapy in, 839–841, 840–841f, 841t, 851t staging of, 839, 839b surgery for, 839 Melanoma conjunctival, 315 cutaneous, 144–145 acral lentiginous, 144–145 amelanotic, 144 biology of, 145 Breslow classification of, 146 Clark classification of, 147 clinical presentation of, 144–145 epidemiology of, 141 evaluation of, 146–147 lentigo maligna, 144 nodular, 144 pagetoid, 144 pathology of, 144–145 spread of, 145, 145f staging of, 147, 148b sunlight exposure and, 142 superficial spreading, 144 treatment of, 149–151 chemotherapy in, 151 elective lymph node dissection in, 149–150 interferon-alpha in, 151 radiation therapy in, 150–151, 153, 153f sentinel node identification in, 150 surgery in, 149 systemic adjuvant therapy in, 151 therapeutic lymph node dissection in, 150 vaccine against, 151 ocular, 313 125I plaques for, 80, 1038–1039 proton therapy for, 1039–1040 106Ru/106Rh plaques for, 80–81, 81f, 1038–1039 uveal, 313–315, 314b brachytherapy in, 315 enucleation in, 314–315 particle beam therapy in, 315 vaginal, 800 vulvar, 800 Meningeal leukemia, 970–971 Meningioma, 857–858 anaplastic, 858 prognosis for, 857–858 radiation therapy in, 857, 858f, 1040 Menopause, breast cancer and, 357 Meperidine, 1015t

1061

Merkel cell carcinoma, 141, 145–146 treatment of, 151–153 Mesorectum, 560 resection of, 574–575 Metastases. See also at specific cancers in anal cancer, 609–610 in anaplastic thyroid cancer, 343 in basal cell skin cancer, 143–144 bone. See Bone, metastatic disease of brain, 859–860 in breast cancer, 391 in lung cancer, 434, 435t, 444–445, 448–449, 448t radiation therapy in, 1028, 1028f in cervical cancer, 736, 743–744, 743t, 752 in endometrial cancer, 775, 790–791 in esophageal cancer, 464–465, 468t in giant cell tumor, 927 hematogenous in oral cavity cancer, 259 in soft tissue sarcoma, 942 in testicular cancer, 661–662 in hypopharyngeal cancer, 287 local-regional treatment and, 92–93 lymphatic in bladder cancer, 632, 634 in breast cancer, 354–355, 355f. See also Breast cancer, lymphatic metastases in in differentiated thyroid cancer, 342 in endometrial cancer, 779–780, 779t, 780f in esophageal cancer, 464, 467f, 481, 481f in hypopharyngeal cancer, 287 in laryngeal cancer, 286–287 in lung cancer, 434, 435t in nasopharyngeal cancer, 208, 210f, 212f in oral cavity cancer, 257–258, 257–258f, 258t, 264 in oropharyngeal cancer, 226–227, 231–232, 238–239, 241–244, 247 in ovarian cancer, 814 in prostate cancer, 683, 704–706, 705f, 705t, 1008–1009 in soft tissue sarcoma, 942 in testicular cancer, 660–661, 660–661f in medullary thyroid cancer, 343 orbital, 317 in oropharyngeal cancer, 239 in parotid gland cancer, 172, 173f prognosis and, 1009 radiation-associated, 123 in salivary gland tumors, 175, 179–181 in soft tissue sarcoma, 942–943 spinal. See Spinal tumors, metastatic in thyroid cancer, 341 time of, 92 in uveal melanoma, 314 in vulvar cancer, 806 Methadone, 1015t Methotrimeprazine, 1015t Mexiletine, 1015t MGMT, 856 Misonidazole, radiosensitizing effects of, 30, 108 Mitomycin C in anal cancer, 612–613, 614–615t in hypoxic cell killing, 108 Mitotic index, 18 Mixed germ cell tumor, 660. See also ­Testicular cancer

1062

Index

Molecular therapeutics. See Radiation therapy and molecular therapeutics Monte Carlo model, dose calculation, 58 Morphine, 1015–1016t Motion, 67–70 bladder, radiation therapy and, 641f interfractional, 67, 69–70 intrafractional, 67–69, 68–69f respiratory, 67–69 in cholangiocarcinoma radiation therapy, 553–554 during computed tomography, 54, 54f in hepatocellular carcinoma radiation therapy, 542 measurement of, 68–69, 68f mitigation of, 69 Mouth. See Oral cavity MRI. See Magnetic resonance imaging (MRI) Mucosa-associated lymphoid tissue (MALT) lymphoma, 893f, 898–899, 898f Mucositis confluent, 250–251 patchy, 250 radiation-related, 275–276, 346 Multileaf collimation, 59 Multiple endocrine neoplasia, type2, 334, 342 Muscle, radiation effects on, 967, 968f Mutation, germline, 10 Mutator phenotype, 10 Myasthenia gravis, thymoma and, 453 MYC, in non-Hodgkin lymphoma, 889–890 Mycosis fungoides, 900–901, 900t Myelopathy, postirradiation, 836–837 Myocardial infarction, radiation therapy and, 417–419, 882 Myocardium, 414 radiation-induced injury to, 416–417

N

Nalbuphine, 1015t Naprosyn, 1015t Nasal cavity, anatomy of, 183–184 Nasal cavity cancer, 183. See also Nasal ­vestibule cancer; Sinonasal cancer clinical manifestations of, 186–187 natural history of, 185 radiation therapy in, 189–194, 200–201f dose for, 198–199 external beam setup for, 199–200 intensity-modulated, 200, 202f outcomes of, 189–191, 190–192t relapse after, 190t target volume for, 198–199 side effects of, 204t staging of, 187, 188b surgery in, 189–194 outcomes of, 189–191, 190–192t Nasal-type T-cell lymphoma, 899 Nasal vestibule, 183 Nasal vestibule cancer clinical manifestations of, 185–186 natural history of, 185 radiation therapy in, 189, 190t brachytherapy for, 189, 197, 198–199f dose for, 197 external beam setup for, 197–198, 199f side effects of, 204t target volume for, 197 surgery in, 189 Nasolacrimal duct, radiation-related stenosis of, 311

Nasopharyngeal cancer, 207 anterior extension of, 207, 209f chemotherapy in, 211–212, 216–219 adjuvant, 217–218 neoadjuvant, 216–217, 217t radiation therapy with, 218–219, 218t in children, 220 environmental factors in, 210 epidemiology of, 210 Epstein-Barr virus and, 210 genetic factors in, 210 histopathology of, 210 inferior extension of, 208, 209f lateral extension of, 208, 209f nodal spread in, 208, 210f, 212f posterior extension of, 208, 209f proton therapy in, 1043 radiation therapy in, 212–216 brachytherapy for, 215–216 chemotherapy with, 218–219, 218t complications of, 215 cutaneous sequelae of, 215 dose for, 213 endocrine sequelae of, 215 fractionation for, 213 intensity-modulated, 162–163, 213, 215, 216f in local recurrence, 219–220, 219t neurologic sequelae of, 215 oral sequelae of, 215 planning for, 212–213, 212f results of, 214–215, 214t recurrent, 219–220, 219t treatment of, 219–220, 219t salvage therapy in, 219–220, 219t spread of, 207–208, 209–210f, 212f staging of, 208–210, 211b stereotactic radiosurgery in, 215–216 superior extension of, 207–208, 209f Nasopharynx anatomy of, 207, 207–208f carcinoma of. See Nasopharyngeal cancer lymphoepithelioma of, 210 lymphoma of, 210 Natural killer cell lymphoma, 899 Nausea, 20, 821 Necrosis, cerebral, 836, 836f Neuroblastoma, 974–976 staging of, 974–975, 975t treatment of, 975–976 Neuroendocrine cancer cervical, 738 prostate, 680 Neurologic examination, in chemodectoma, 328 Neutron(s), 4–5, 35 energy track of, 33 Neutron therapy, 35 in gastric cancer, 506 in prostate cancer, 706 in salivary gland tumors, 179–181 Nicotinamide, radiosensitizing effects of, 31 Nimorazole, radiosensitizing effects of, 31 Nominal standard dose, for fractionation, 23 Non-Hodgkin lymphoma, 889–902 anatomic sites, 890 classification systems, 890–893 NCI Working Formulation, 891t REAL classification, 891t, 892t WHO classification, 892t epidemiology and etiology, 889–890 International Prognostic Index, 893t

Non-Hodgkin lymphoma (Continued) orbital, 315, 316, 316f staging evaluation for, 890–893 Ann Arbor staging system, 893 subtypes of, 894–902, 891t, 892t Burkitt’s lymphoma, 899 diffuse large B-cell lymphoma, 894–896 follicular lymphoma, 896–898 mantle cell lymphoma, 899 mucosa-associated lymphoid tissue lymphoma (MALT), 898–899 plasmacytomas, solitary, 902 precursor lymphoblastic lymphomas, 899–900 primary CNS disease, 901–902 survival according to, 893f T-cell lymphomas, 899 mycosis fungoides (Sézary ­syndrome), 900–901 anaplastic large cell lymphoma, 901 lymphomatoid papulomatosis, 901 Nonchromaffin paraganglioma. See ­Chemodectoma Nonsteroidal anti-inflammatory drugs, 127–128, 1015t colorectal cancer and, 562–563 Normal tissue cell survival curve for, 13–15, 14f radiation effects on, 15. See also at specific organs and tissues Normalized growth delay, 104–105, 105f Nortriptyline, 1015t Nutrition, in oral cavity cancer, 276

O

Obesity breast cancer and, 358 after craniopharyngioma treatment, 852 Occam, William of, 8 Occult blood test, 567–569 Ohngren’s line, 184, 184f Olfactory region, 183–184 Olfactory region cancer clinical manifestations of, 187 natural history of, 185 side effects of, 204t treatment of, 192–194, 193–194t Oligodendroglioma, 857 On-board kilovoltage x-ray imaging, 71–72, 72f Oncogenes, 8 functions of, 9 Oncolytic viruses, for gene therapy, 130–131 ONYX-15, 130–131 Oophorectomy in dysgerminoma, 827–828 in ovarian cancer. See Ovarian cancer, surgery in prophylactic, 361, 371 Opioids, 1015–1016t addiction to, 1013–1014 prescription for, 1014, 1015–1016t tolerance to, 1013–1014 Optic chiasm glioma of, 844–845 radiation effects on, 312, 837 Optic nerve glioma of, 844–845 radiation effects on, 312, 837

Index Oral cavity, 250 anatomy of, 252–253, 252–253f dysplasia of, 259 erythroplakia of, 259 leukoplakia of, 259, 265 premalignant changes in, 259 radiation effects on, 250–251, 251t Oral cavity cancer, 250 buccal mucosa, 262t, 263, 272–273 chemoprevention of, 264–265 cigarette smoking and, 252, 259 clinical presentation of, 253, 253t coincident second malignancy and, 257 cyclins in, 259 death from, 251t epidemiology of, 250, 251t etiology of, 251–252 evaluation of, 253–257, 253t floor of mouth, 264, 267f, 273 gingival, 264, 273 hard palate, 264, 273 hematogenous metastases in, 259 host factors in, 252 imaging in, 254–257, 254t, 255f lip, 262t, 263, 264f, 272 lymphatic metastases in, 257–258, 257–258f, 258t evaluation of, 255–257, 256–258f pathology of, 259–260, 261f physical examination in, 254 radiation therapy in, 261, 265–270, 270t altered-fractionation regimens for, 267–268 brachytherapy for, 265–266, 266–267f care after, 277 complications of, 251t, 274–277, 274b, 275t dental decay with, 276–277, 277f external beam, 266–270, 268–269f infection with, 275 intensity-modulated, 270 intraoral cone therapy for, 266 local control and, 271 mucositis with, 275–276 nodal, 258, 258t nutritional problems with, 276 osteonecrosis with, 276 pain with, 275–276 portals for, 268–270, 268–269f postoperative, 268 regression patterns and, 271, 271–272t results of, 270–274, 271–272t soft tissue necrosis with, 276 xerostomia with, 274–275 recurrent, 264 risk factors for, 252 staging of, 259, 260b tongue, 262t, 264, 266–267f, 273–274 TP53 in, 259 treatment of, 260–265. See also Oral ­cavity cancer, radiation therapy in in buccal mucosa lesions, 263, 272–273 care after, 277 in floor of mouth lesions, 264, 273 in gingival lesions, 264, 273 in hard palate lesions, 264, 273 in high-risk resected disease, 263 in lip lesions, 263, 264f, 272 nodal, 264 principles of, 260–265, 262f, 262t in recurrent lesions, 264 in retromolar trigone lesions, 264, 273

Oral cavity cancer (Continued) selection of, 261–263 surgical, 261 in T1 and T2 lesions, 263 in T3 and T4 lesions, 263 in tongue lesions, 264, 273–274 viruses in, 259–260 Oral contraceptives, cervical cancer and, 734 Orbit. See also specific structures anatomy of, 309, 310f benign disease of, 312–313 Graves’ ophthalmopathy of, 312–313 lymphoma of, 315–316, 316f malignant disease of, 313–317. See also Melanoma, ocular; Melanoma, uveal metastasis to, 317 pseudotumor of, 313 radiation effects on, 309–312 Orchiectomy in gonadal stromal tumor, 671 in nonseminomatous testicular tumor, 670–671 in seminomatous testicular tumor, 668 Orchioblastoma, 660 Organ-at-risk volume, 52 Organ of Corti, 320, 321f Oropharyngeal cancer, 224 detection of, 228 distant metastatic disease in, 239 epidemiology of, 227–228 epidermal growth factor receptor in, 228 etiology of, 227–228 evaluation of, 228–231, 230f lymphatic metastases in, 226–227 treatment of, 231–232, 238–239, 241–244, 247 oropharyngeal wall, 227, 228f intensity-modulated radiation therapy in, 241 standard fractionated radiation therapy in, 233, 238 pathology of, 224 prevention of, 228 proton therapy for, 1043 recurrent, 239–240 soft palate, 227, 227f brachytherapy in, 238 chemotherapy in, 233 external beam radiation therapy in, 233, 237–238, 237t intensity-modulated radiation therapy in, 241, 246f surgery in, 233 staging of, 228–231, 229b tongue base, 225–226, 226f, 230f brachytherapy in, 235 external beam radiation therapy in, 235–236, 235t intensity-modulated radiation therapy in, 232–233, 232f, 241, 245f radiation therapy and surgery in, 236 surgery in, 232–233, 236 tonsillar region, 226, 227f, 227t, 230f brachytherapy in, 237 chemotherapy in, 233 external beam radiation therapy in, 233, 236–237, 236t, 243f intensity-modulated radiation therapy in, 233, 241, 241–242f treatment of, 231–233 in advanced infiltrative disease, 231 biological agents in, 247–248

1063

Oropharyngeal cancer (Continued) in distant metastatic disease, 239 in early-stage disease, 231 intensity-modulated radiation therapy in, 162–163, 233–234, 240–246 dose-fractionation schedules for, 244–246 oropharyngeal wall, 241 setup for, 240 soft palate, 241, 246f tongue base, 232–233, 232f, 241, 245f tonsillar region, 233, 241, 241–242f in intermediate-stage disease, 231 in nodal disease, 231–232, 238–239, 241–244, 247 palliative, 240 postoperative radiation therapy in, 247, 248f principles of, 231 proton therapy in, 1043 in recurrent disease, 239–240 results of, 233–240, 235–237t site-specific, 232–233 standard-field radiation therapy in, 234–235, 247 Oropharyngeal wall cancer, 227, 228f intensity-modulated radiation therapy in, 241 standard fractionated radiation therapy in, 233, 238 Oropharynx anatomy of, 224, 227t cancer of. See Oropharyngeal cancer Osteoradionecrosis auricular ossicles, 322 mandibular, 276 prevention of, 917 temporal bone, 322 Osteosarcoma, radiation-induced, 918–919 Otitis media, serous, radiation-related, 321–322 Ovarian cancer, 813–827 aneuploidy in, 827 borderline, 826–827 BRCA genes and, 813–814 CA125 in, 814–817 chemotherapy in, 817, 818t, 819, 822 vs. abdominopelvic radiation therapy, 822–823 palliative, 825 radiation therapy after, 823–825, 823t radiation therapy with, 826 second malignancy and, 822 classification of, 813, 813b endometrial cancer and, 780 epidemiology of, 813–814 estrogen exposure and, 813 evaluation of, 815–817 lymph nodes in, 814 pathology of, 814 prognosis for, 814–815, 816f radiation therapy in consolidation, 823–825, 823–824t altered fractionation schedules for, 825 complications of, 825 patient selection for, 824–825 results of, 823–824, 824t intraperitoneal radiocolloids for, 818, 818t limited-field, 826 palliative, 825–826

1064

Index

Ovarian cancer (Continued) postoperative, 819–822 vs. chemotherapy, 822–823 dose for, 821 efficacy of, 819, 819t, 820f fibrosis after, 821 gastrointestinal side effects of, 821–822, 822t hematologic side effects of, 821, 822t patient selection for, 816f, 820, 820f pneumonitis after, 821 side effects of, 821–822, 822t technique of, 820–821 salvage, 823–825, 823t altered fractionation schedules for, 825 complications of, 825 patient selection for, 824–825 risk factors for, 813–814 spread of, 814 staging of, 814, 815b surgery in, 817 chemotherapy after, 817, 818t, 819, 822–823, 823t intraperitoneal radiocolloid therapy after, 818, 818t radiation therapy after, 817–822, 818–819t, 820f Ovary (ovaries) anatomy of, 812, 812f chemotherapy effects on, 813 dysgerminoma of, 813b, 827–828 epithelial cancer of. See Ovarian cancer granulosa cell tumors of, 813b, 827 primordial follicles of, 812 prophylactic removal of, 361, 371 radiation effects on, 812–813 shielding for, 813 Oxaliplatin in gastric cancer, 505–506 in rectal cancer, 580, 589–590 Oxycodone, 1015t Oxygen hyperbaric in osteoradionecrosis, 917 radiosensitizing effects of, 30 tumor, 27–30, 28f. See also Hypoxia, ­tumor; Reoxygenation, tumor Oxygen enhancement ratio, 34f, 35

P

Paclitaxel in gastric cancer, 505–506 in pancreatic cancer, 527t, 528–529 radiation therapy with, 107–108, 107f tumor oxygen effects of, 108, 109f Paget’s disease, vulvar, 800 Pain, 1012–1013, 1013b back. See Spinal tumors, metastatic management of, 1008, 1012–1016. See also Palliative radiation therapy addiction and, 1013–1014 prescription for, 1014, 1015–1016t tolerance and, 1013–1014 neuropathic, 1012–1013 oral cavity, radiation-related, 275–276 somatic, 1012 visceral, 1012 Pair production, 4, 5f Palatal cancer. See Hard palate cancer; Soft palate cancer Palladium-103 seeds, in prostate cancer, 693

Palliative care, 1007 bisphosphonates in, 1011 vs. chemotherapy, 1011 research on, 1011–1012 chemotherapy in, 1011 vs. curative care, 1010–1012 metastasis and, 1009 outcomes of, 1012 pain management in, 1008, 1012–1016, 1013b. See also Palliative radiation therapy addiction and, 1013–1014 prescription for, 1014, 1015–1016t tolerance and, 1013–1014 in prostate cancer, 1008–1009 quality of life and, 1009 radiation therapy in, 1011. See also ­Palliative radiation therapy socioeconomic costs of, 1007–1009 survival duration and, 1009–1010 treatment modalities in, 1010–1012 Palliative radiation therapy, 1011, 1014–1016 in biliary tract cancer, 550–553 in bladder cancer, 639 in bone metastases, 1016–1027, 1018t in brain metastases, 1028, 1028f in breast cancer, 391 in cholangiocarcinoma, 550–553 in cutaneous and subcutaneous disease, 1031 in disseminated bone metastases, 1025–1027, 1027t in gallbladder cancer, 550–553 in gastric cancer, 504 in head and neck tumors, 1029–1030 in lung cancer, 450–451 normal tissue tolerance and, 1016 in ovarian cancer, 825–826 in pelvic disease, 1030–1031 in pulmonary metastases, 1029 recommendations for, 1031 in rectal cancer, 595–596, 595f reirradiation and, 1016 site location for, 1016 in skin cancer, 152 socioeconomic costs of, 1008 in spinal cord compression, 1021–1024, 1023–1024f in thyroid cancer, 346 Pancoast’s syndrome, 449 Pancreas anatomy of, 516 cancer of. See Pancreatic cancer radiation effects on, 516–517 acute, 516 late, 516–517 Pancreatic cancer bevacizumab in, 126, 529 cetuximab in, 530 chemotherapy in, 521–525, 522t capecitabine-based, 529 5-fluorouracil-based, 526–528, 527t gemcitabine-based, 527t, 528 paclitaxel-based, 527t, 528–529 preoperative, 523–525 clinical manifestations of, 517–518 diet and, 517 epidemiology of, 517 erlotinib in, 529–530 etiology of, 517 evaluation of, 517–518, 525 locally advanced

Pancreatic cancer (Continued) radiation therapy and chemotherapy in, 526–529 capecitabine-based, 529 5-fluorouracil-based, 526–528, 527t gemcitabine-based, 527t, 528 paclitaxel-based, 527t, 528–529 radiation therapy in, 519–520, 520f resectability of, 525–526 marginally resectable, 525 molecular therapy in, 529–530 nonoperative biliary decompression in, 525 pancreaticoduodenectomy in, 518 prognosis for, 518–519 radiation therapy in, 519–521 adjuvant, 519, 520f chemotherapy with, 521–529, 524b, 527t dose escalation in, 521 dose in, 516–517, 521 hepatic, 530 intensity-modulated, 520–521 for locally advanced disease, 519–520, 520f preoperative, 523–525 stereotactic, 521 recurrent, 519 resectability of, 525–526 risk factors for, 517 staging of, 518, 518b surgery in, 518–519 vascular extension of, 525 Pancreaticoduodenectomy, 518 Paracervix, 735 Paraganglia, 327–328 chemodectoma of. See Chemodectoma Parametrium, 735 Paranasal sinuses anatomy of, 184, 184f cancer of. See Sinonasal cancer Paraneoplastic syndromes, in lung cancer, 435 Parathyroid cancer, of thyroid gland, 344 Paresthesias, radiation-induced, 835–836 Parinaud’s syndrome, 849–850 Parotid duct, 170, 170f Parotid glands acinic cell carcinoma of, 174, 175t adenocarcinoma of, 174, 175t adenoid cystic carcinoma of, 172f, 174, 175t anatomy of, 170, 170f benign mixed (pleomorphic) adenoma of, 173, 173f treatment of, 176–177, 177t carcinoma ex pleomorphic adenoma of, 174 malignant mixed tumor of, 174, 175t mucoepidermoid carcinoma of, 173f, 174 poorly differentiated carcinoma of, 173f squamous cell carcinoma of, 174–175, 175t total-body irradiation effects on, 998 tumors of, 175t benign, 173, 173f clinical presentation of, 172, 173f epidemiology of, 172 malignant, 174–175 metastasis with, 172, 173f neural spread of, 178–179 radiation therapy in intensity-modulated, 179f postoperative, 175–177, 176t primary, 177 technique of, 177, 178f

Index Parotid glands (Continued) staging of, 173, 174b surgery for, 175 radiation therapy after, 175–177, 176t Warthin’s tumor of, 173 Parotitis, total-body irradiation and, 998 Partial response, 16 Particle beam therapy. See Neutron therapy; Proton therapy Pelvic examination, in cervical cancer, 740–741 Pelvic spaces, 560 Pentazocine, 1015t Peptic ulcer disease gastric cancer and, 496 radiation therapy in, 493–495 Percent-labeled mitoses curve, 18, 18f Percutaneous ethanol injection, in ­hepatocellular carcinoma, 540–542 Pericardial effusion, radiation-induced, 415–416, 480 Pericarditis acute, radiation-induced, 414–416 chronic, radiation-induced, 416 Pericardium, 414 radiation-induced injury to, 415–416 Peripheral nervous system, radiation effects on, 837 Peripheral T-cell lymphoma, 893f, 899 Periplakin, in esophageal cancer, 483t Peritoneum, radiation injury to, 506–507 PET. See Positron emission tomography (PET) Pharyngeal walls, 283–284 Phenothiazines, 1015t Phenytoin, 1015t Phosphatidylinositol-3-kinase, 119 Phosphorus-32 therapy in bone metastases, 1027, 1027t in ovarian cancer, 818, 818t Photoelectric effect, 4, 5f Photon, energy track of, 33, 33f Photon beam radiation therapy. See also Intensity-modulated radiation therapy (IMRT) delivery of, 59–62 fixed-beam attenuation in, 59–61 helical tomotherapy in, 61 multileaf collimation in, 59 quality assurance in, 61–62, 61–62f planning for, 51–59 dose-calculation algorithms in, 57–58 field-in-field planning in, 59, 60f imaging in, 53–56, 53–56f immobilization techniques in, 56–57, 57f, 57t inverse planning in, 58–59 target volume delineation in, 51–53, 51–52f Pi-mesons, 5 Pineal region tumor, 849–851, 849f, 850t biopsy in, 850 radiation therapy in, 850–851 Pineoblastoma, 849–851, 850t Pineocytoma, 849, 850t Piriform pyriform sinuses, 283, 284f tumors of. See Hypopharyngeal cancer Pituitary adenoma, proton therapy in, 1040 Plaque brachytherapy in ocular melanoma, 80–81, 81f, 1038–1039 in pterygium, 313

Plasmacytoma, 902, 919–920 chemotherapy in, 920 extramedullary, 902 radiation therapy in, 920 solitary, 902 Platelets, radiation effects on, 876–877, 994, 994f Platelet-derived growth factor, 124–125 Plummer-Vinson syndrome, 464 Pneumonia, interstitial, total-body ­irradiation and, 999–1000 Pneumonitis, radiation, 425, 425f, 821 clinical manifestations of, 429, 429f computed tomography in, 429f, 430 concurrent chemotherapy and, 430, 430f functional effects of, 430–432, 431f irradiated lung volume and, 427–429, 427–428f radiography in, 429–430, 429f radioiodine and, 346 total-body irradiation and, 999–1000 transforming growth factor β in, 426–427 treatment of, 432–433, 432f Poikiloderma, 142 Polypectomy, 566–567 Pons, glioma of, 846–848, 847f Population kinetics, of tumor, 19 Positron emission tomography (PET) in bone cancer, 919 in cervical cancer, 741 after colorectal cancer resection, 570–571 in esophageal cancer, 467–468, 469f, 482, 485, 485f in lung cancer, 436 in non-Hodgkin lymphoma, 890–892 in oral cavity cancer, 254–257 in soft tissue sarcoma, 944 in treatment planning, 54–55, 56f Postoperative radiation therapy, 97. See also at specific cancers benefits of, 97 dose for, 95, 95f limitations of, 97 Potentially lethal damage repair, 20–21 inhibition of, 105–107 Precursor B-cell lymphoblastic lymphoma, 899–900 Precursor T-cell lymphoblastic lymphoma, 899–900 Pregnancy breast cancer and, 357–358, 357t breast cancer treatment and, 370 after pelvic radiation, 775 Preoperative radiation therapy, 96–97. See also at specific cancers benefits of, 96 dose for, 95, 95f limitations of, 97 Presacral space, 560 Primitive neuroectodermal tumors, 838, 852–853 Primordial germ cell, 653–654 Proctocolectomy, prophylactic, 566 Proctoscopy, in cervical cancer, 741 Proliferative potential, of tumor cell, 27–28 Propoxyphene, 1015t ProstaScint scan, 683 Prostate cancer, 678–680 androgen ablation in, 684, 701–704, 701–702t EORTC trial of, 702t, 703

1065

Prostate cancer (Continued) M.D. Anderson diethylstilbestrol trial of, 701–702, 702t Medical Research Council trial of, 702t, 703–704 in node-positive disease, 704–706 prostatectomy and, 709–710, 709t RTOG 85-31 trial of, 702–703, 702t RTOG 86-10 trial of, 702, 702t RTOG 92-02 trial of, 702t, 703 short-course, 702t, 704 Umea University trial of, 702t, 703 on autopsy, 679 B-mode Acquisition and Target system in, 690, 691f biopsy in, 680–683, 706 bone scan in, 683 brachytherapy in, 691–694 external beam radiation therapy with, 693, 697 historical perspective on, 691 planning for, 691–693, 692f postimplantation evaluation for, 693, 693–694f salvage, 707 in T1-T2a disease, 694–697, 696t temporary iridium-192 implants for, 693–694 urinary complications of, 697 computed tomography in, 683, 690, 690f cryotherapy in, 716 detection of, 679–680 DNA content characteristics in, 685–686 epidemiology of, 678–679 etiology of, 679 failed treatment of, 683–684 family history in, 686 flow cytometry in, 686 Gleason score in, 680–685, 685f, 710–711 grading of, 680 growth rate in, 685–686 histology of, 680 imaging in, 683 in-room computed tomography in, 74, 75f Ki-67 in, 685–686 lymph node dissection in, 683 lymph node–positive, 683 treatment of, 704–706, 705f, 705t magnetic resonance imaging in, 681 metastatic leukemia and, 919 lymphatic, 683, 704–706, 705f, 705t pain and, 1008–1009 molecular markers in, 686 mortality rates with, 678–679 neuroendocrine, 680 palpability of, 680, 681t prognosis for, 680, 683–686 ProstaScint scan in, 683 prostate-specific antigen in. See Prostate-specific antigen proton therapy for, 1044–1045 radiation therapy in, 686–691, 697–706 androgen ablation with, 684, 701–704, 701–702t biochemical failure after, 683–684 biopsy after, 706 bladder volume in, 687, 690, 690–691f brachytherapy for, 691–694, 692–693f, 697, 707 colorectal cancer and, 563 complications of, 677–678

1066

Index

Prostate cancer (Continued) conformal, 686–688, 688f cryotherapy after, 707 dose-escalation in, 698–700, 698t, 699f dose-response evaluation in, 698–700, 698t, 699f, 700t in favorable (T1-T2a) disease, 694–697, 694f, 695t intensity-modulated, 688–690, 688f nodal, 701, 704–706, 705f, 705t particle beam therapy for, 706 prostate apex localization for, 691 prostate-specific antigen nadir after, 684 prostatectomy after, 707 rectal reaction with, 700, 700f, 700t rectal volume changes and, 690, 690–691f rectum volume in, 687 results of, 694–707 in favorable disease, 694–697, 694f, 695–696t in intermediate-to-high-risk disease, 697–706, 698t, 699f, 701–702t salvage procedures after, 706–707 second malignancy after, 41, 41f three-dimensional planning for, 686–688, 687f rectal examination in, 680–681 recurrence biochemical evidence of, 683–684. See also prostate-specific antigen. seminal vesicle involvement in, 713 staging of, 680–683, 681t, 682b controversies in, 681 surgery in. See Prostatectomy testosterone in, 679, 686 transrectal ultrasonography in, 680–681 transurethral resection in, 685 ultrasonographic localization of, 72–73, 73–74f zonal distribution of, 676–677 Prostate gland, 676 anatomy of, 676–677, 677f biopsy of, 680–683, 706 bladder cancer spread to, 632 cancer of. See Prostate cancer excision of. See Prostatectomy permanent radioactive seeds, implantation in. See Prostate cancer, brachytherapy in radiation effects on, 677–678, 678t zones of, 676 Prostate-specific antigen, 679–680 age-related levels of, 679 bounce phenomenon of, 684 free-to-total ratio of, 679–680 nadir level of, 684 pelvic radiation effects on, 677 pretreatment, 684–685, 685f prognostic significance of, 681, 714, 716 after radiation therapy, 683–684 Prostatectomy, 707–716 adjuvant radiation therapy after, 710–716, 712t randomized trials of, 713–714, 713t vs. salvage radiation therapy, 716, 716t androgen ablation with, 709–710, 709t biochemical end points after, 707–708 biochemical failure after, 683–684 clinical end points after, 707–708 erectile function after, 707 for favorable disease, 708, 708t high-risk pathology after, 710–713, 710–711t

Prostatectomy (Continued) for intermediate-risk-to-high-risk disease, 708–710, 709t prostate-specific antigen after, 683–684. See also Prostate-specific antigen salvage, 707 salvage radiation therapy after, 710–716, 714t, 715f, 716t vs. adjuvant radiation therapy, 716, 716t in T1-T2a disease, 694–696, 695t urinary incontinence after, 707 Proton(s), 4 passively scattered, 82–83, 83f production of, 81–82, 82f Proton-photon therapy, in chordoma, 924 Proton therapy, 35, 81–87, 1038 accelerator for, 81–82, 82f beam shape in, 83–84, 83–84f beam width in, 82–83, 82–83f biologic effects of, 1038 in cervical spine tumors, 1040–1041 chemotherapy with, 1045 in children, 1041–1042, 1041–1043f in cholangiocarcinoma, 550–551 in chondrosarcoma, 1040 in chordoma, 1040–1041 compensator in, 83–84, 84f components of, 81–84, 82–83f, 1036 cyclotron for, 81–82 depth dose curve in, 81, 81f discrete spot scanning in, 85–86 dose distribution in, 83–84, 83–84f dosimetric advantages of, 1038, 1038–1039f in esophageal cancer, 484, 484f, 1044 evaluation of, 1045–1046 in glioma, 1040 in head and neck tumors, 1042–1043 in hepatocellular carcinoma, 543–544 historical perspective on, 1036 in medulloblastoma, 840–841, 843f, 1041, 1041f in meningioma, 1040 in nasopharyngeal cancer, 1043 in non–small cell lung cancer, 440, 1043–1044, 1044f in ocular melanoma, 1038–1040 in oropharyngeal cancer, 1043 in paranasal sinus cancer, 1042–1043 in pelvic sarcoma, 1041, 1043f in pituitary adenoma, 1040 planning for, 84–85 in prostate cancer, 706, 1044–1045 radiobiologic effectiveness of, 86 range in, 83 range modulator wheel in, 82–83, 83f in retinoblastoma, 1041, 1042f in skull base tumors, 1040–1041 spread-out Bragg peak in, 82–83, 83f synchrotron for, 81–82, 82f in uveal melanoma, 315 Pseudoaddiction, 1013 Pseudosarcoma, laryngeal, 287 Pseudotumor, orbital, 313 Pterygium, 313 PTK787 (vatalanib), bevacizumab with, 124 Pubocervical ligament, 735 Pulmonary fibrosis clinical manifestations of, 429 transforming growth factor β in, 426–427 Pulsed-dose-rate brachytherapy, in cervical cancer, 76–78, 77–78f, 760–761 Pyelonephritis, radiation-related, in cervical cancer, 765

Q

Quality of life, 1009

R

R115777 (tipifarnib), 129 Race, breast-conservation therapy and, 370–371 Radiation. See also Radiation therapy and at specific types of therapy and specific cancers carcinogenesis with, 39–40. See also ­Second malignancy absolute vs. relative risk models for, 40, 41f latent period in, 40 electromagnetic, 4. See also Gamma rays; X-rays historical perspective on, 3–4 particulate, 4–5 therapy-related production of, 5–6, 6t types of, 4–5 Radiation dose, 6–7 cell survival curve and, 13, 13f depth-dose curves for, 6f maximum tumor dose for, 7 measurement of, 7, 7f microdosimetry for, 7, 8f minimum tumor dose for, 7 tumor dose for, 7 Radiation hepatitis, 495–496, 517 Radiation-induced liver disease, 535–536 normal-tissue complication probability model in, 536–538, 537–538f Radiation pneumonitis. See Pneumonitis, radiation Radiation therapy. See also at specific types and specific cancers biologic basis of, 10–42, 10f cell cycle and, 17–19, 17–18f cell survival curves with, 13, 13f chemistry of, 10–11, 10f chemotherapy and. See Radiation therapy and chemotherapy chromosomal aberrations with, 11–12, 12f DNA target of, 10–11, 10–11f energy deposition in, 7, 8f, 33–35, 33f fractionation for, 20–26, 21–22f, 26–27f. See also Fractionation hereditary effects of, 39 historical perspective on, 3–4 molecular therapeutics and. See Radiation therapy and molecular therapeutics normal tissue effects of, 13–15, 14f oxygen effects in, 28–33, 28–30f. See also Hypoxia, tumor planning for, 51–59 quality of, 33, 33–34f radiation production for, 5–6, 6t selection of, 7–8 surgery and. See Radiation therapy and surgery systemic effects of, 19–20 tumor effects of, 15–17, 16f tumor kinetics and, 19 tumor sensitivity to, 26–28 tumor volume and, 26 Radiation therapy and chemotherapy, 36–37, 37f, 102 additive effects of, 103–105, 104f assessment of, 104–105, 104–105f, 106t mechanisms of, 105–110

Index Radiation therapy and chemotherapy (Continued) adjuvant chemotherapy in, 111–112 amifostine with, 33 in ampulla of Vater carcinoma, 548 in anal cancer, 113–114, 610–615, 612t angiogenesis inhibition in, 110 in apical sulcus lung cancer, 449–450, 449f in bladder cancer, 114, 636t, 637 cell cycle effects of, 107–108, 107–108f cell regeneration and, 108–110, 109–110f cellular repair inhibition in, 105–107, 106t, 107f in cervical cancer, 113, 747, 748t, 749–751, 763–764 in cholangiocarcinoma, 547–548 clinical applications of, 111 concurrent chemotherapy in, 111–113, 112t DNA damage in, 105 in endometrial cancer, 785 in esophageal cancer, 472–478, 476–477f in gallbladder cancer, 549 in gastric cancer, 500–504, 503f in head and neck squamous cell ­carcinoma, 163–165 in hepatocellular carcinoma, 544 in hypopharyngeal cancer, 293–294, 305 hypoxic cell reduction in, 108, 109f induction chemotherapy in, 111–112 in laryngeal cancer, 114 in nasopharyngeal cancer, 218–219, 218t in non–small cell lung cancer, 440–442, 441–442f, 444t in oral cavity cancer, 267–268 organ conservation with, 113–114 in ovarian cancer, 826 in pancreatic cancer, 521–529, 522t, 524b radioprotectants in, 103 in rectal cancer, 114, 577–580, 577–578t, 588–591 sequencing in, 111 in small cell lung cancer, 113, 445–449, 446f spatial cooperation in, 103 surgery and, 113 survival curves for, 104, 104f therapeutic ratio in, 102–103, 103f tissue toxicity of, 103, 114–115, 114t toxicity in, 103 tumor response enhancement in, 103 in vulvar cancer, 804–805, 805t Radiation therapy and molecular ­therapeutics, 118 angiogenesis in, 122–125 antiangiogenic therapy in, 123–125 clinical studies of, 126 preclinical findings in, 125–126 antivascular therapy in, 123 COX-2 inhibitors in, 127–128, 128f cyclin-dependent kinase inhibitors in, 128–129, 129f cyclin-dependent kinases in, 128–129 cyclins in, 128–129 cyclooxygenases in, 126–128, 127f epidermal growth factor receptor in, 118–122 gene therapy in, 129–131 immunotherapy in, 131–132 in non–small cell lung cancer, 442–444, 443–445f, 444t in pancreatic cancer, 529–530 RAS signaling pathway in, 129

Radiation therapy and surgery, 92. See also Postoperative radiation; Preoperative radiation and at specific cancers clinical applications of, 98 complications of, 98–99 local-regional tumor control and, 92 principles of, 99 salvage application of, 93 subclinical disease and, 93–96, 93–95f time intervals in, 97–98 topographic relationships in, 96 Radiodermatitis, 142, 153–154 Radiofrequency ablation in epithelioid hemangioendothelioma, 926 in hepatocellular carcinoma, 540–542 Radiography in bone lymphoma, 922, 922f in cervical cancer, 741 in chordoma, 923 in eosinophilic granuloma, 929f in epithelioid hemangioendothelioma, 925 in giant cell tumor, 927 in lung cancer, 434–435, 435t in malignant fibrous histiocytoma, 925f in radiation pneumonitis, 429–430, 429f in spinal metastases, 1022–1023, 1022–1023f Radioiodine ablation secondary malignancy and, 346 in thyroid cancer, 338–339, 341, 346 Radionuclide imaging in bone cancer, 919 in radiation-induced cardiac injury, 414–415, 418 of thyroid gland, 337 Radionuclide injection therapy, in ­hepatocellular carcinoma, 544–545 Radioprotection, 32–33 Radioresistance epidermal growth factor receptor and, 119, 120f hypoxia and. See Hypoxia, tumor Radioresponsiveness, 15–16 Radiosensitivity, 15–17, 16f, 26–28. See also Hypoxia, tumor oxygen and, 28–30, 28–30f Radiosensitizers, 30–31 gene delivery of, 130 Raloxifene, in breast cancer prevention, 361 RAS, 129 RAS-RAF-MAPK pathway, 119, 129 Razoxane, in chondrosarcoma treatment, 921 RB1, 8–10 in bladder cancer, 631 Rectal cancer. See also Colorectal cancer endorectal sonography in, 571–573, 573f radiation exposure history and, 563 radiation therapy and chemotherapy in, 113–114, 577–580, 577–578t, 588–591 adjuvant, 577–578 biological agents with, 580 capecitabine for, 580, 589 diarrhea with, 591–592 5-fluorouracil for, 579–580, 588–589 historical perspective on, 579 late effects of, 592 local excision and, 590–591 neoadjuvant, 578–579

1067

Rectal cancer (Continued) oxaliplatin for, 580, 589–590 repeat, 592–594 side effects of, 591–592 three-step bowel management program with, 591 radiation therapy and surgery in, 98 radiation therapy in, 580–589 belly board for, 581, 581f brachytherapy for, 583 cystitis with, 628 endocavitary, 583 inferior border for, 582 inguinal region, 582 intraoperative, 594–595 outcomes of, 585t palliative, 595–596, 595f postoperative, 583–585 preoperative, 583–585 radiation-surgery interval and, 588 recurrence after, 582–583 small bowel exclusion for, 581, 581f superior border for, 582, 582f toxicity of, 585–586 recurrence of, 582–583, 592–596 intraoperative radiation in, 594–595 palliative radiation therapy in, 595–596, 595f radiation therapy in, 592–594 surgery in, 592 sphincter preservation in, 586–589, 587t staging of, 571–573, 572b surgery in, 573–577 complications of, 577 imaging after, 570–571 local excision in, 575–576 in locally advanced disease, 576 mesorectal resection in, 574–575 outcomes of, 576–577, 578t, 585t radiation-surgery interval and, 588 radiation therapy after, 581, 583–585 radiation therapy before, 581, 583–589, 587t radiation therapy during, 594–595 salvage, 592 sphincter preservation in, 575, 586–589 surveillance after, 569–571 Rectum anatomy of, 573, 574f cancer of. See Rectal cancer examination of, in prostate cancer, 680–681 radiation effects on, 560–561 acute, 560–561 late, 561 Red blood cells, radiation effects on, 877 Relative biologic effectiveness fractionation and, 34, 34f linear energy transfer and, 34, 34f radiation quality and, 35 Reoxygenation, tumor, 29, 30f. See also Hypoxia, tumor vascular changes in, 413 Respiratory motion, 67–69 in cholangiocarcinoma radiation therapy, 553–554 during computed tomography, 54, 54f in hepatocellular carcinoma radiation therapy, 542 measurement of, 68–69, 68f mitigation of, 69 RET, 334, 337, 342 Rete testis, 653

1068

Index

Retina, radiation effects on, 311–312 Retinoblastoma, 980–982, 981b chemotherapy in, 981 prognosis for, 982 proton therapy in, 1041, 1042f radiation therapy in, 981–982, 981f secondary tumors and, 918 soft tissue sarcoma and, 941 staging of, 980, 981b 13-cis-Retinoic acid, in leukoplakia, 265 Retinopathy, radiation-related, 311–312 Retromolar trigone cancer, 264, 273. See also Oral cavity cancer Rhabdoid tumor, renal, 972 Rhabdomyosarcoma, 976–980, 977f, 978t bladder, 979–980 embryonal, 800, 977 extremity, 980 orbital, 979 parameningeal, 979 paratesticular, 980 prognosis for, 980 prostate, 979–980 radiation therapy for, 977–980, 979t staging of, 977, 978t Rhenium-186 therapy, in bone metastases, 1027, 1027t Ritalin, 1015t Rituximab, in follicular lymphoma, 898 RNA interference, 131 Roentgen, Wilhelm Conrad, 3 Rosenmüller’s fossa, 207, 207f Ruthenium-106/rhodium-106 plaques, in ocular melanoma, 80–81, 81f, 1038–1039

S

S-100 protein, in soft tissue tumor, 941 Salivary gland(s), 169 anatomy of, 169–171, 169–170f major, 169 tumors of. See Salivary gland tumors minor, 169 tumors of, 172–173, 178, 179f radiation effects on, 171–172, 250, 251t, 275–276 saliva secretion by, 169 postradiation, 171–172 tubuloalveolar structure of, 169, 169f Salivary gland tumors, 260 benign, 173, 176–177, 177t clinical presentation of, 172–173, 172–173f distribution of, 169 epidemiology of, 172 inoperable, 179–181 locally advanced, 179–181 malignant, 174–175 metastasis with, 175, 179–181 neural spread of, 178–179 outcomes of, 175–176, 176t pathogenesis of, 169 pathology of, 173–175, 175t radiation therapy in, 177 fast neutron, 179–181 for inoperable disease, 179–181 intensity-modulated, 171–172, 179, 179–180f for neural spread, 178–179, 179–180f for parotid gland tumors, 177, 178f postoperative, 175–177, 176t for submandibular gland tumors, 177–178, 178f

Salivary gland tumors (Continued) techniques of, 177–181, 178–180f threshold dose and, 171 staging of, 173, 174b surgery in, 175 Samarium-153 therapy, in bone metastases, 919, 934 Sarcoma experimental, CpG oligodeoxynucleotides in, 131–132, 132f pelvic, 1041, 1043f proton therapy in, 1041, 1043f radiation-induced, 940 renal, 972 small intestine, 509 soft tissue, 940. See also specific soft tissue sarcomas chemotherapy in, 953–955, 954–955t distal extremities, 957, 957f head and neck, 957 histology of, 941, 943b metastases from, 942–943 pathology of, 941 presentation of, 941–944, 943t radiation therapy in, 950–952 bone fracture after, 952–953 complications of, 952–953 dose considerations in, 951–952 physical therapy after, 953 postoperative, 946–947, 947–948t preoperative, 947–949, 948–949t primary, 953, 953t in recurrent disease, 955–956 volume considerations in, 950–951, 951–952f wound complications after, 953 recurrence of, 943–944, 955–956 retroperitoneal, 944, 951–952, 956–957, 956f staging of, 941, 942b, 944–945 surgery in, 945–946, 945–946t brachytherapy after, 949–950, 952 radiation therapy after, 946–947, 947–948t radiation therapy before, 947–949, 948–949t synovial, 958 uterine, 791–792 vaginal, 800 vulvar, 800 Schistosomiasis, bladder cancer and, 630 Sclera, radiation effects on, 311 Scoliosis, radiation-related, 967 Screening colorectal cancer, 565–571 familial adenomatous polyposis, 569 gastric cancer, 497 hereditary nonpolyposis colorectal cancer, 569 skin cancer, 142–143 Sebaceous carcinoma, 146 treatment of, 152 Second malignancy, 41–42, 41f, 918–919 breast cancer therapy and, 356 cervical cancer therapy and, 41–42 in children, 969 Hodgkin lymphoma therapy and, 42, 333, 882, 888 oral cavity cancer and, 257 ovarian cancer therapy and, 822 prostate cancer therapy and, 41, 41f radioiodine ablation–related, 346

Second malignancy (Continued) retinoblastoma therapy and, 918 total-body irradiation and, 1001 Semaxanib (SU5416), 124 Seminiferous tubules, 653, 654f Seminoma, 659–660, 659f anaplastic, 659 classic, 659 evaluation of, 662–664 extragonadal, 671 spermatocytic, 659–660 spread of, 660–662, 660–661f staging of, 662, 663b treatment of, 664–668, 665t chemotherapy in, 667–668 radiation therapy in, 665–666 in bulky abdominal disease, 667, 668t, 670 cardiac effects of, 419 nodal, 666–667, 667t portals for, 666, 670 for postchemotherapy residual disease, 668 recommendations for, 668–670, 669f in stage I disease, 665–666, 669 in stage IIA disease, 666, 669 in stage IIB disease, 666, 669–670 in stage IID disease, 666, 670 in stage III disease, 670 in stage IIIC disease, 669–670 in stage IV disease, 670 tumor markers in, 664 Sentinel node in breast cancer, 377–378 in melanoma, 150 Sertoli cell, radiation effects on, 657 Sertoli cell tumor, 671 Sexual function, radiation therapy and, 798 Sézary syndrome, 900–901, 900t Sigmoidoscopy, 567–569 Signal transduction, 9 Single photon emission computed ­tomogra­phy(spect), in treatment planning, 55, 56f Sinonasal cancer, 183 chemotherapy in, 205 detection of, 184–185 epidemiology of, 184 ethmoid air cell clinical manifestations of, 186–187 natural history of, 185 side effects of, 204t treatment of, 189–194, 198–200, 201–202f etiology of, 184 intensity-modulated radiation therapy in, 198t, 205 maxillary sinus clinical manifestations of, 186f, 187–189 natural history of, 185 side effects of, 204t staging of, 188b treatment of, 194–197, 194–198t, 200–201, 203–204f nasal cavity clinical manifestations of, 186–187 natural history of, 185 side effects of, 204t staging of, 188b treatment of, 189–194, 190–192t, 198–200, 200–201f

Index Sinonasal cancer (Continued) nasal vestibule clinical manifestations of, 185–186 natural history of, 185 side effects of, 204t treatment of, 189, 190t, 197–198, 198–199f natural history of, 185 olfactory region clinical manifestations of, 187 natural history of, 185 side effects of, 204t treatment of, 192–194, 193–194t prevention of, 184–185 proton beam therapy in, 205, 1042–1043 surgery in, 203–205 Skin, 141 anatomy of, 141 cancer of. See Melanoma; Skin cancer dry desquamation of, 250–251 moist desquamation of, 250–251 radiation effects on, 141–142, 153–154, 411–412 acute, 142 chronic, 142 total-body irradiation and, 999 Skin cancer. See also Melanoma apocrine, 146 treatment of, 152 basal cell, 143–144 basosquamous (metatypical, keratotic), 143 biology of, 143–144 clinical presentation of, 143 epidemiology of, 141 infiltrative, 143 metastatic, 143–144 morphea-like (sclerosing), 143 nodal disease in, 147–149 nodulo-ulcerative (rodent ulcer), 143 pathology of, 143 perineural invasion in, 143 radiation therapy in, 152–154 spread of, 143–144 staging of, 146, 147b superficial, 143 terebrant, 143 treatment of, 147–149 eccrine, 146 treatment of, 152 epidemiology of, 142 etiology of, 142 evaluation of, 146–147 Merkel cell (trabecular, neuroendocrine), 141, 145–146 treatment of, 151–153 perineural invasion in, 143–144 prevention of, 142–143 screening for, 142–143 sebaceous, 146 treatment of, 152 squamous cell, 144 biology of, 144 clinical presentation of, 144 epidemiology of, 141 nodal disease in, 147–149 pathology of, 144 radiation therapy in, 152–154 spread of, 144 staging of, 146, 147b treatment of, 147–149 treatment of, 147–149, 149t palliative, 152, 1031

Skin cancer (Continued) radiation therapy in, 152–154 patient care during, 153–154 Small intestine, 507 anatomy of, 507 histology of, 507 radiation effects on acute, 507 late, 507 treatment-related exclusion of, 581, 581f Small intestine cancer adenocarcinoma, 508–509 carcinoid, 509 epidemiology of, 507 etiology of, 507 radiation therapy in, 508–509 techniques of, 509 sarcoma, 509 staging of, 507–508, 508b Socioeconomic costs, cancer-related, 1007–1009 Soft palate cancer, 227, 227f brachytherapy in, 238 chemotherapy in, 233 external beam radiation therapy in, 233, 237–238, 237t intensity-modulated radiation therapy in, 241, 246f surgery in, 233 Soft tissue tumors, 940. See also specific tumors anatomic sites of, 941–942, 943t benign, 940 biopsy in, 944 borderline, 940 chemotherapy for, 953–955, 954–955t classification of, 940 clinical presentation of, 941–944 definition of, 940 epidemiology of, 940–941 etiology of, 940–941 growth of, 942 histology of, 941, 943b imaging in, 944–945 local recurrence of, 943–944 malignant, 940, 943b metastases from, 942–943 natural history of, 941–944 pathology of, 941 prognosis for, 941–944 radiation therapy in, 950–952 complications of, 952–953 in distal extremities, 957, 957f dose in, 951–952, 951–952f in head and neck disease, 957 primary, 953, 953t in recurrent disease, 955–956 in retroperitoneal disease, 956–957, 956f volume considerations in, 950–951 recurrence of, 955–956 staging of, 941, 942b, 944–945 surgery in, 945–946, 945–946t brachytherapy after, 949–950, 950t radiation therapy after, 946–947, 947–948t radiation therapy before, 947–949, 949t Solitary fibrous tumor (hemangiopericytoma), 960 Somnolent syndrome, 835–836, 968 Sorafenib (Nexavan), 125 Spence, tail of, 353–354 Sperm count, radiation-related decline in, 656 Spermatocytes, 655 radiation effects on, 655

1069

Spermatogenesis, 654–658, 655f postirradiation recovery of, 656 radiation effects on, 655–657 Spermatogonia, 654–655 experimental transplantation of, 658 radiation effects on, 655 Spermatozoa, 655 radiation-induced mutation in, 657 Spinal cord compression of, 1021–1024. See also Spinal tumors, metastatic radiation effects on, 835–837, 1025 radiation tolerance of, 1024–1025 tumors of, 860 in children, 837–838 Spinal tumors, metastatic, 860, 1021–1024 in breast cancer, 391 radiation therapy in, 1023, 1023–1024f fracture after, 1025, 1025f spinal cord injury with, 1025 radiography of, 1022–1023, 1022–1023f stereotactic radiation therapy in, 63–65, 64–65f symptoms of, 1021–1022, 1022f Spleen irradiation of, 998 bone marrow transplantation and, 998 radiation effects on, 878, 995 Squamous cell carcinoma antigen, 744 Stem cell transplantation, 993 allogeneic, 995 autologous, 995 boost irradiation in, 998 historical perspective on, 993 indications for, 995–996 localized irradiation in, 997–998 nonmyeloablative conditioning regimens in, 998 syngeneic, 995 total-body irradiation in, 996–997, 996f acute toxic effects of, 998–999 alopecia with, 999 cardiac effects of, 1000 cataracts after, 1000 cutaneous effects of, 999 dose rate for, 996 fractionation for, 996–997 gastrointestinal effects of, 998 gonadal effects of, 1000 growth effects of, 1000–1001 hepatic effects of, 999 neurocognitive effects of, 1001–1002 organ shielding for, 997, 997f parotid effects of, 998 psychological effects of, 1001–1002 pulmonary effects of, 999–1000 renal effects of, 1000 second malignancy and, 1001 sequelae of, 998–1002 acute, 998–999 long-term, 999–1002 thyroid effects of, 1001 total dose for, 996–997 Stereotactic radiation therapy, 62–67 in pancreatic cancer, 521 in spinal tumors, 63–65, 64–65f in thoracic tumors, 65–67, 66f, 67t Stereotactic radiosurgery in chondrosarcoma, 921 in cranial tumors, 63, 63f in glioma, 855 in nasopharyngeal cancer, 215–216

1070

Index

Stewart-Treves syndrome, 940 Stimulants, 1015t Stomach, 493–496 anatomy of, 493, 494f cancer of. See Gastric cancer histology of, 493, 494f radiation effects on, 493–495 acute, 493–495, 495f late, 495–496 Strandqvist plot, 23 Strontium-89, in bone metastases, 925, 931, 1026, 1027t SU5416 (semaxanib), 124 SU11248 (sunitinib), 125 Subclinical disease, 93–96 clonogen density in, 95–96 definition of, 93 vs. microscopic, 93 radiation biology of, 94–96, 94–95f, 95t surgical removal of, 93–94, 93f Subcutaneous tissue, 141 Subglottic cancer clinical manifestations of, 287–288 spread of, 285 staging of, 289–290, 289b treatment of, 293 Sublethal damage repair fractionation and, 20, 21–22f inhibition of, 105–107 Sublingual glands, 170–171. See also Salivary gland(s) Submandibular glands. See also Salivary gland(s) anatomy of, 170–171, 171f tumors of, 172, 175t intensity-modulated radiation therapy in, 179, 180f staging of, 173, 174b surgery for, 175 radiation therapy after, 175–178, 176t, 178f Suicide gene therapy, 38 Sunitinib (SU11248), 125 Sunlight exposure, skin cancer and, 142 Superior vena cava, obstruction of, in lung cancer, 450 Supraclavicular lymph nodes, 354–355 radiation therapy for, 395–396, 398f Supraglottic cancer clinical manifestations of, 287 early-stage, 292–293 radiation therapy in complications of, 305–306 dose and fractionation for, 301–302 in early-stage disease, 293 fields for, 297–298, 298f intensity-modulated, 298, 299f, 301 postoperative, 293 results of, 304–305 spread of, 285, 286f staging of, 289–290, 289b T3, 293 T4, 293 treatment of, 292–293 in early-stage disease, 292–293 neck dissection in, 293 surgery in, 292–293 in T3 disease, 293 in T4 disease, 293 Supraglottic voice, 287 Supraglottis, 282

Surgery. See also Radiation therapy and surgery and at specific cancers subclinical disease and, 93–94, 93f Synovial sarcoma, 958

T

T bar and vacuum bag system, 57, 57f, 57t T-cell lymphoma nasal, 899 peripheral, 893f, 899 Tail of Spence, 353–354 Tamoxifen in breast cancer prevention, 360–361 in ductal in situ breast cancer, 367 in early-stage breast cancer, 368–369, 373, 378–380 endometrial cancer and, 774 Target volume, 51–53 clinical, 51–52, 52f respiratory motion and, 68 internal, 52 respiratory motion and, 68–69 planning, 52, 52f gross tumor volume and, 52–53 respiratory motion and, 69 Targeted therapy, 37–38. See also Cetuximab; Rituximab; Trastuzumab Taste buds, radiation effects on, 250 Taxanes radiation therapy with, 107–108, 107–108f tumor oxygen effects of, 108, 109f TEL-AML1, 970 Telecurietherapy, 4 Telomerase, in esophageal cancer, 483t Temporal bone chemodectoma of, 326–331 clinical presentation of, 328 evaluation of, 328 pathology of, 328 staging of, 328–329 treatment of, 329–330, 330t radiation therapy in, 330–331, 330t surgery in, 329, 330t radiation-related necrosis of, 322 Teratoma, 660. See also Testicular cancer Terminal illness, 1008. See also Palliative care Test phantom, for quality assurance, 61–62, 61–62f Testicular cancer, 658–664 alpha-fetoprotein in, 664 classification of, 659–660, 659f computed tomography in, 662–664 cryptorchidism and, 658–659 differential diagnosis of, 660 evaluation of, 662–664 geographic variation in, 658 hematogenous metastasis in, 661–662 human chorionic gonadotropin in, 664 imaging in, 662–664 incidence of, 658–659 local extension of, 660 lymphangiography in, 662 lymphatic metastasis in, 660–661, 660–661f lymphoma, 671–672 nonseminomatous, 659–660, 659f treatment of, 670–671 in stage I disease, 670 in stage II disease, 670–671

Testicular cancer (Continued) pathology of, 659 seminomatous, 659–660, 659f anaplastic, 659 classic, 659 spermatocytic, 659–660 treatment of, 664–668, 665t chemotherapy in, 667–668 radiation therapy in, 665–666 in bulky abdominal disease, 667, 668t cardiac effects of, 419 nodal, 666–667, 667–668t portals for, 666, 670 for postchemotherapy residual disease, 668 recommendations for, 668–670, 669f in stage I disease, 665–666, 669 in stage IIA disease, 666, 669 in stage IIB disease, 666, 669–670 in stage IIC disease, 666, 670 in stage III disease, 670 in stage IIIC disease, 669–670 in stage IV disease, 670 spread of, 660–662, 660f staging of, 662, 663b stromal, 671 treatment of, 664–671 tumor markers in, 664 Testicular shield, 657, 657f Testis (testes) anatomy of, 653, 654f cancer of. See Testicular cancer development of, 653–654 radiation effects on, 653 dose and, 656–657 endocrine, 657–658 genetic, 657 paracrine, 657–658 shield prevention of, 657, 657f somatic, 658 spermatogenic, 655–657, 655f Testosterone, prostate cancer and, 679, 686 Therapeutic ratio, in radiation therapy and chemotherapy, 102–103, 103f Thermoplastic mask, 56–57, 57f, 57t Thorium oxide (Thorotrast), maxillary sinus carcinoma and, 184 Three-step bowel management program in anal cancer, 619 in rectal cancer, 591 Thromobocytosis, in cervical cancer, 744 Thymectomy, 451 Thymic carcinoma, 452 Thymoma, 451–452 myasthenia gravis and, 453 prognosis for, 453 radiation therapy in, 453, 453t recurrence of, 452–453, 452t resection of, 452–453, 452t staging of, 452 Thymus, 451–453 Thyrohyoid membrane, 283 Thyroid cancer, 333 anaplastic, 336, 336f, 343–344 clinical manifestations of, 343 evaluation of, 343 management of, 343–344 prognosis for, 343 classification of, 335–336 differentiated, 337–342, 339f, 339b

Index Thyroid cancer (Continued) adjuvant radiation therapy in, 340–341, 340t chemotherapy in, 341–342 clinical manifestations of, 337 evaluation of, 337, 339b, 341 follow-up for, 339–340 metastatic, 341 nodal recurrence of, 341 prognosis for, 337 radioiodine ablation in, 338–339, 341 recurrence of, 341 staging of, 337, 338b thyroglobulin-positive, scan-negative, 341 thyroidectomy in, 337–340, 339f epidemiology of, 334 etiology of, 334–335 follicular, 335, 336f genetic factors in, 334 iodine insufficiency and, 334–335 Hürthle cell, 335–336 incidence of, 334 insular, 336 medullary, 336, 336f, 342–343 clinical manifestations of, 342 evaluation of, 342 familial, 334 management of, 342 hypercalcitoninemia and, 342–343 in metastatic disease, 343 prognosis for, 342 papillary, 335, 335f diffuse sclerosing variant of, 335 genetic factors in, 334 ground-glass appearance of, 335, 335f Orphan Annie appearance of, 335, 335f tall-cell variant, 335 radiation-related, 334 radiation therapy in, 344–346 adjuvant, 340–341, 340t complications of, 346 external beam, 341, 345–346 in anaplastic cancer, 343–344 complications of, 346 in medullary cancer, 342 to thyroid bed alone, 345 to thyroid bed and lymph nodes, 345, 345f gonadal failure and, 346 intensity-modulated, 345, 345f leukemia and, 346 palliative, 346 pericardial effusion after, 416 radioiodine, 338–339, 341, 344–346 in recurrent disease, 341 secondary, 333 Thyroid cartilage, 282–283 Thyroid gland cancer of. See Thyroid cancer fine needle aspiration of, 337 lymphoma of, 344 nodules of, 337. See also Thyroid cancer parathyroid carcinoma of, 344 radiation effects on, 333 squamous cell carcinoma of, 344 total-body irradiation effects on, 1001 Thyroidectomy in anaplastic cancer, 343 in differentiated cancer, 337–340, 339f in medullary cancer, 342

Tipifarnib (R115777), 129 Tirapazamine, 31–32, 31f, 108 TNF, vector-based tumor delivery of, 130 Tolerance, analgesic, 1013–1014 Tongue cancer oral, 262t, 266–267f treatment of, 264, 273–274 oropharyngeal, 225–226, 226f, 230f brachytherapy in, 235 external beam radiation therapy in, 235–236, 235t intensity-modulated radiation therapy in, 232–233, 232f, 241, 245f radiation therapy and surgery in, 236 surgery in, 232–233, 236 Tonsillar region cancer, 226, 227f, 227t, 230f brachytherapy in, 237 chemotherapy in, 233 external beam radiation therapy in, 233, 236–237, 236t, 243f intensity-modulated radiation therapy in, 233, 241, 241–242f Tooth (teeth) fluoride for, 277 preradiation evaluation of, 276 radiation-related effects on, 276–277, 277f, 916–917 Torus tubarius, 207, 207f Total-body irradiation, 996–997, 996f acute toxic effects of, 998–999 alopecia with, 999 cardiac effects of, 1000 cataracts with, 1000 cutaneous effects of, 999 dose rate for, 996 erythrocyte response to, 995 fractionation for, 996–997 gastrointestinal effects of, 998 gonadal effects of, 1000 growth effects of, 1000–1001 hepatic effects of, 999 historical perspective on, 993 hormone deficiencies with, 916 leukocyte response to, 994 long-term effects of, 999–1002 meta-analysis of, 1002, 1002f neurocognitive effects of, 1001–1002 normal bone marrow response to, 993–995, 994f organ shielding for, 997, 997f parotid effects of, 998 peripheral blood response to, 994–995, 994f platelet response to, 994 psychological effects of, 1001–1002 pulmonary effects of, 999–1000 renal effects of, 1000 second malignancy after, 919, 1001 sequelae of, 998–1002 splenic effects of, 995 systemic effects of, 19–20 thyroid effects of, 1001 total dose for, 996–997 TP53, 8–10, 130 in anal cancer, 610 in bladder cancer, 631 in breast cancer, 359 in esophageal cancer, 483t in Li-Fraumeni syndrome, 941 in oral cavity cancer, 259 Trachea, radiation effects on, 424–433. See also at Lung(s) Trachelectomy, in cervical cancer, 745

1071

Transarterial embolization, in hepatocellular carcinoma, 541–542 Transforming growth factor α, in esophageal cancer, 483t Transforming growth factor β, in radiationinduced lung injury, 426–427, 426f Transplantation. See Bone marrow transplantation; Stem cell transplantation Trastuzumab in early-stage breast cancer, 379–380 in inflammatory breast cancer, 389 Tricyclic antidepressants, 1015t Tumor bed effect, 123–124 Tumor cell kill, 14–15 Tumor-control dose (TCD50) assay, 16, 104–105 Tumor growth delay assay, 104–105, 104f Tumor necrosis factor-α, radiation-induced expression of, 130 Tumor suppressor gene, 8–9 functions of, 9 Tumor suppressor gene therapy, 38 Tumor volume, 26, 27t gross, 51, 51–52f planning target volume and, 52–53 Two-compartment model, 123 Tympanic cavity, 320, 321f Tympanic membrane, 320, 321f in chemodectoma, 328 TYMS, 564 Tyrosine kinase inhibitors, in non–small cell lung cancer, 443, 444f

U

Ulcer, cutaneous, radiation-related, 142 Ulcerative colitis anal cancer and, 608 colorectal cancer and, 563–564 Ultrasonography in prostate cancer, 72–73, 73f, 690, 691f of thyroid gland, 337 transrectal, in prostate cancer, 680–681 Urinary tract infection, bladder cancer and, 630 Uterine cervix, 733 anatomy of, 734–736, 735f atypical squamous cells of undetermined significance of, 736–737 cancer of. See Cervical cancer endometrial cancer invasion of, 778 epidemiology of, 734 lymphatics of, 735–736, 736f radiation effects on, 733–734 Uterine sarcoma, 791–792 Uterosacral ligament, 735 Uterus anatomy of, 734–736, 735f, 774 cancer of. See Endometrial cancer radiation effects on, 774–775 sarcoma of, 791–792 Uvea, melanoma of, 313–315, 314b brachytherapy in, 315 enucleation in, 314–315 particle beam therapy in, 315

V

Vaccine human papillomavirus, 734 melanoma, 151

1072

Index

Vagina. See also Vaginal cancer melanoma of, 800 radiation effects on, 766, 798 sarcoma of, 800 Vaginal cancer, 799 chemotherapy in, 808 diethylstilbestrol and, 799 epidemiology of, 799 evaluation of, 800–801 human papillomavirus and, 799 lymph nodes in, 799 natural history of, 799 pathology of, 800 prognosis for, 802 radiation therapy in, 806–808, 807f brachytherapy for, 806–808, 807f chemotherapy with, 808 complications of, 808 intensity-modulated, 808 results of, 808 spread of, 799 staging of, 800–801, 802t treatment of, 806–808 chemotherapy in, 808 in invasive disease, 806–808, 807f in preinvasive disease, 806 Vaginal intraepithelial neoplasia, 800, 806 Vandetanib (ZD6474), 125–126 in non–small cell lung cancer, 443–444, 444–445f Vascular endothelial growth factor, 124 in colorectal cancer, 565 in esophageal cancer, 483t in hepatocellular carcinoma, 544 inhibitors of, 124–125 side effects of, 126 Vasectomy, prostate cancer and, 679 Vatalanib (PTK787), bevacizumab with, 124 Vertebrae metastases to See Spinal tumors, ­metastatic radiation effects on, 967 Vimentin, in soft tissue tumor, 941 Virus cytotoxic, 38 Epstein-Barr, 210, 497

Virus (Continued) herpes simplex, 275 HIV, 615, 734, 902 HPV. See Human papillomavirus (HPV) infection oncolytic, 130–131 Vision decrease in, after craniopharyngioma treatment, 852 radiation-related impairment of, 311–312, 982 Vocal cords false, 282, 283–284f true, 282, 283–284f. See also Glottis Vomiting, 20 Vulva. See also Vulvar cancer melanoma of, 800 Paget’s disease of, 800 radiation effects on, 765, 798 sarcoma of, 800 Vulvar cancer, 799 chemotherapy in, 804–805, 805t epidemiology of, 799 evaluation of, 800–801 human papillomavirus and, 799 inguinal lymphadenectomy in, 803 lymph nodes in, 799 metastatic, 806 natural history of, 799 pathology of, 800 prognosis for, 801–802, 802f radiation therapy in, 805–806 chemotherapy with, 804–805, 805t infection with, 806 inguinal, 804, 805f postoperative, 803–804 preoperative, 804–805 prophylactic, 804 side effects of, 806 spread of, 799 staging of, 800–801, 801b surgery in, 803 treatment of, 802–806 in invasive disease, 803 in locoregionally advanced disease, 804–805, 805t

Vulvar cancer (Continued) in metastatic disease, 806 postoperative radiation therapy in, 803–804 in preinvasive disease, 802–803 prophylactic inguinal radiation therapy in, 804, 805f Vulvar intraepithelial neoplasia, 800, 802–803

W

WAGR syndrome, 971 Warthin’s tumor, 173 Weight, body, breast cancer and, 358 William of Occam, 8 Wilms’ tumor, 971–974, 972t abdominal radiation therapy in, 973 hepatic radiation therapy in, 974 prognosis for, 974 pulmonary radiation therapy in, 973–974

X

X-rays, 4 energy track of, 33, 33f Xerostomia, radiation-related, 171, 215, 274–275 management of, 172, 277 prevention of, 172

Y

Yolk sac tumor, 660 Yttrium-90 microsphere therapy, in ­hepatocellular carcinoma, 544–545

Z

ZD6126, 123 ZD6474 (vandetanib), 125–126 in non–small cell lung cancer, 443–444, 444–445f Zubrod performance status score, 894t