Ebook Cancer epidemiology and prevention (4/E): Part 1

 i

Schottenfeld and Fraumeni
Cancer Epidemiology and Prevention


ii


 iii

Schottenfeld and Fraumeni
Cancer Epidemiology and
Prevention
Fourth Edition
Lead Editor

MICHAEL J. THUN, MD, MS
Epidemiology and Surveillance Research (Retired)
American Cancer Society
Atlanta, Georgia

Co-​Editors

MARTHA S. LINET, MD, MPH
Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

JAMES R. CERHAN, MD, PHD

Department of Health Sciences Research
Mayo Clinic
Rochester, Minnesota

1

CHRISTOPHER HAIMAN, SCD
Department of Preventive Medicine
Keck School of Medicine, University
of Southern California
Los Angeles, California

DAVID SCHOTTENFELD, MD, MSC
Department of Epidemiology (Retired)
University of Michigan School of Public Health
Ann Arbor, Michigan

Project Manager

ANNELIE M. LANDGREN, MPH, PMP


iv

1
Oxford University Press is a department of the University of Oxford. It furthers
the University’s objective of excellence in research, scholarship, and education
by publishing worldwide. Oxford is a registered trade mark of Oxford University
Press in the UK and certain other countries.
Published in the United States of America by Oxford University Press

198 Madison Avenue, New York, NY 10016, United States of America.
© Oxford University Press 2018
Third Edition published 2006
Second edition published 1996
First edition published 1982
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by license, or under terms agreed with the appropriate reproduction
rights organization. Inquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above.
You must not circulate this work in any other form
and you must impose this same condition on any acquirer.
Library of Congress Cataloging-in-Publication Data
Names: Thun, Michael J., editor. | Linet, Martha S., editor. |
Cerhan, James R., editor. | Haiman, Christopher, editor. | Schottenfeld, David, editor.
Title: Schottenfeld and Fraumeni Cancer Epidemiology and Prevention / lead editor,
Michael J. Thun ; co-editors, Martha S. Linet, James R. Cerhan,
Christopher Haiman, David Schottenfeld ; project manager, Annelie M. Landgren.
Other titles: Cancer epidemiology and prevention
Description: Fourth edition. | New York, NY : Oxford University Press, [2018] |
Preceded by Cancer epidemiology and prevention / edited by David Schottenfeld,
Joseph F. Fraumeni Jr. 3rd ed. 2006. | Includes bibliographical references and index.
Identifiers: LCCN 2017038170 | ISBN 9780190238667 (hardcover : alk. paper)
Subjects: | MESH: Neoplasms—epidemiology | Neoplasms—prevention & control
Classification: LCC RA645.C3 | NLM QZ 220.1 |
DDC 614.5/999—dc23
LC record available at />9 8 7 6 5 4 3 2 1
Printed by Sheridan Books, Inc., United States of America

 v

Contents

Acknowledgments 
Contributors 
Preface 

ix
xi
xix

1.Introduction 
Michael J. Thun, Martha S. Linet, James R. Cerhan, Christopher A. Haiman,
and David Schottenfeld

1

I  BASIC CONCEPTS
2. Biology of Neoplasia 
Michael Dean and Karobi Moitra

9

3. Morphological and Molecular Classification of Human Cancer 
Mark E. Sherman, Melissa A. Troester, Katherine A. Hoadley,
and William F. Anderson

19

4. Genomic Landscape of Cancer: Insights for Epidemiologists 
Christopher J. Maher and Elaine R. Mardis

43

5. Genetic Epidemiology of Cancer 
Kathryn L. Penney, Kyriaki Michailidou, Deanna Alexis Carere, Chenan Zhang,
Brandon Pierce, Sara Lindström, and Peter Kraft

53

6. Application of Biomarkers in Cancer Epidemiology 
Roel Vermeulen, Douglas A. Bell, Dean P. Jones, Montserrat Garcia-​Closas,
Avrum Spira, Teresa W. Wang, Martyn T. Smith, Qing Lan, and Nathaniel Rothman

77

7. Causal Inference in Cancer Epidemiology 
Steven N. Goodman and Jonathan M. Samet

97

II  THE MAGNITUDE OF CANCER
8. Patterns of Cancer Incidence, Mortality, and Survival 
Ahmedin Jemal, D. Maxwell Parkin, and Freddie Bray

107

9. Socioeconomic Disparities in Cancer Incidence and Mortality 
Candyce Kroenke and Ichiro Kawachi

141

10. The Economic Burden of Cancer in the United States 
K. Robin Yabroff, Gery P. Guy Jr., Matthew P. Banegas, and Donatus U. Ekwueme

169


vi

vi

Contents

III  THE CAUSES OF CANCER
11. Tobacco 
Michael J. Thun and Neal D. Freedman

185

12. Alcohol and Cancer Risk 
Susan M. Gapstur and Philip John Brooks

213

13. Ionizing Radiation 
Amy Berrington de González, André Bouville, Preetha Rajaraman,

and Mary Schubauer-​Berigan

227

14. Ultraviolet Radiation 
Adèle C. Green and David C. Whiteman

249

15. Electromagnetic Fields 
Maria Feychting and Joachim Schüz

259

16. Occupational Cancer 
Kyle Steenland, Shelia Hoar Zahm, and A. Blair

275

17. Air Pollution 
Jonathan M. Samet and Aaron J. Cohen

291

18. Water Contaminants 
Kenneth P. Cantor, Craig M. Steinmaus, Mary H. Ward,
and Laura E. Beane Freeman

305

19. Diet and Nutrition 
Marjorie L. McCullough and Walter C. Willett

329

20. Obesity and Body Composition 
NaNa Keum, Mingyang Song, Edward L. Giovannucci, and A. Heather Eliassen

351

21. Physical Activity, Sedentary Behaviors, and Risk of Cancer 
Steven C. Moore, Charles E. Matthews, Sarah Keadle, Alpa V. Patel,
and I-​Min Lee

377

22. Hormones and Cancer 
Robert N. Hoover, Amanda Black, and Rebecca Troisi

395

23. Pharmaceutical Drugs Other Than Hormones 
Marie C. Bradley, Michael A. O’Rorke, Janine A. Cooper, Søren Friis,
and Laurel A. Habel

411

24. Infectious Agents 
Silvia Franceschi, Hashem B. El-​Serag, David Forman, Robert Newton,
and Martyn Plummer

433

25. Immunologic Factors 
Eric A. Engels and Allan Hildesheim

461

IV  CANCERS BY TISSUE OF ORIGIN
26. Nasopharyngeal Cancer 
Ellen T. Chang and Allan Hildesheim

489

27. Cancer of the Larynx 
Andrew F. Olshan and Mia Hashibe

505

28. Lung Cancer 
Michael J. Thun, S. Jane Henley, and William D. Travis

519

29. Oral Cavity, Oropharynx, Lip, and Salivary Glands 
Mia Hashibe, Erich M. Sturgis, Jacques Ferlay, and Deborah M. Winn

543

30. Esophageal Cancer 

William J. Blot and Robert E. Tarone

579


 vi

vii

Contents

31. Stomach Cancer 
Catherine de Martel and Julie Parsonnet

593

32. Cancer of the Pancreas 
Samuel O. Antwi, Rick J. Jansen, and Gloria M. Petersen

611

33. Liver Cancer 
W. Thomas London, Jessica L. Petrick, and Katherine A. McGlynn

635

34. Biliary Tract Cancer 
Jill Koshiol, Catterina Ferreccio, Susan S. Devesa, Juan Carlos Roa,
and Joseph F. Fraumeni, Jr.

661

35. Small Intestine Cancer 
Jennifer L. Beebe-​Dimmer, Fawn D. Vigneau, and David Schottenfeld

671

36. Cancers of the Colon and Rectum 
Kana Wu, NaNa Keum, Reiko Nishihara, and Edward L.Giovannucci

681

37. Anal Cancer 
Andrew E. Grulich, Fengyi Jin, and I. Mary Poynten

707

38. Leukemias 
Martha S. Linet, Lindsay M. Morton, Susan S. Devesa, and Graça M. Dores

715

39. Hodgkin Lymphoma 
Henrik Hjalgrim, Ellen T. Chang, and Sally L. Glaser

745

40. The Non-​Hodgkin Lymphomas 
James R. Cerhan, Claire M. Vajdic, and John J. Spinelli

767

41. Multiple Myeloma 
Mark P. Purdue, Jonathan N. Hofmann, Elizabeth E. Brown, and Celine M. Vachon

797

42. Bone Cancers 
Lisa Mirabello, Rochelle E. Curtis, and Sharon A. Savage

815

43. Soft Tissue Sarcoma 
Marianne Berwick and Charles Wiggins

829

44. Thyroid Cancer 
Cari M. Kitahara, Arthur B. Schneider, and Alina V. Brenner

839

45. Breast Cancer 
Louise A. Brinton, Mia M. Gaudet, and Gretchen L. Gierach

861

46. Ovarian Cancer 
Shelley S. Tworoger, Amy L. Shafrir, and Susan E. Hankinson

889

47. Endometrial Cancer 
Linda S. Cook, Angela L. W. Meisner, and Noel S. Weiss

909

48. Cervical Cancer 
Rolando Herrero and Raul Murillo

925

49. Vulvar and Vaginal Cancers 
Margaret M. Madeleine and Lisa G. Johnson

947

50. Choriocarcinoma 
Julie R. Palmer

953

51. Renal Cancer 
Wong-​Ho Chow, Ghislaine Scelo, and Robert E. Tarone

961

52. Bladder Cancer 
Debra T. Silverman, Stella Koutros, Jonine D. Figueroa, Ludmila Prokunina-​Olsson,
and Nathaniel Rothman

977

53. Prostate Cancer 
Catherine M. Tangen, Marian L. Neuhouser, and Janet L. Stanford

997


vi

viii

Contents

54. Testicular Cancer 
Katherine A. McGlynn, Ewa Rajpert-​De Meyts, and Andreas Stang

1019

55. Penile Cancer 
Morten Frisch

1029

56. Nervous System 
E. Susan Amirian, Quinn T. Ostrom, Yanhong Liu, Jill Barnholtz-​Sloan,
and Melissa L. Bondy

1039

57. Melanoma 
Bruce K. Armstrong, Claire M. Vajdic, and Anne E. Cust

1061

58. Keratinocyte Cancers 
Anala Gossai, Dorothea T. Barton, Judy R. Rees, Heather H. Nelson,
and Margaret R. Karagas

1089

59. Childhood Cancers 
Eve Roman, Tracy Lightfoot, Susan Picton, and Sally Kinsey

1119

60. Multiple Primary Cancers 
Lindsay M. Morton, Sharon A. Savage, and Smita Bhatia

1155

V  CANCER PREVENTION AND CONTROL
61. Framework for Understanding Cancer Prevention 
Michael J. Thun, Christopher P. Wild, and Graham Colditz

1193

62. Primary Prevention of Cancer 

1205

62.1. Tobacco Control 
Jeffrey Drope, Clifford E. Douglas, and Brian D. Carter

1207

62.2.Prevention of Obesity and Physical Inactivity 
Ambika Satija and Frank B. Hu

1211

62.3.Prevention of Infection-​Related Cancers 
Marc Bulterys, Julia Brotherton, and Ding-​Shinn Chen

1217

62.4.Protection from Ultraviolet Radiation 
Robyn M. Lucas, Rachel E. Neale, Peter Gies, and Terry Slevin

1221

62.5. Preventive Therapy 
Jack Cuzick

1229

62.6. Regulation 
Jonathan M. Samet and Lynn Goldman

1239

63. Cancer Screening 
Jennifer M. Croswell, Russell P. Harris, and Barnett S. Kramer

1255

Index1271


 ix

Acknowledgments

We are indebted to the more than 190 chapter authors who generously contributed their time,
labor, and expertise to produce this comprehensively updated fourth edition. The multi-​authored
text reflects the increasingly interdisciplinary and collaborative nature of the field; it provides a
resource for researchers seeking to harness the unprecedented advances in genetic and molecular research into large-​scale population studies of cancer etiology, and ultimately into effective
preventive interventions. We owe special thanks to Ms. Annelie Landgren, whose energy, enthusiasm, and organizational expertise as project manager have been invaluable in bringing this
text to completion. We also thank Dr. Stephen Chanock for his early and unfailing encouragement and for supporting the critical infrastructure necessary for such a collaborative enterprise.
This book would not have been possible without the generous forbearance of our spouses and
families. Finally, Michael Thun thanks Dr. Lynne Moody for her insights as a sounding board
throughout this process.

ix


x

 xi

Contributors

E. Susan Amirian, PhD

Jennifer L. Beebe-​Dimmer, PhD, MPH

Dan L Duncan Cancer Center
Baylor College of Medicine
Houston, Texas

Wayne State University School of Medicine
Karmanos Cancer Institute
Detroit, Michigan

William F. Anderson, MD, MPH

Douglas A. Bell, PhD

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Environmental Epigenomics, Immunity, Inflammation and
Disease Laboratory
National Institute of Environmental Health Sciences
Research Triangle Park, North Carolina

Samuel O. Antwi, PhD

Department of Health Sciences Research
Mayo Clinic College of Medicine
Rochester, Minnesota

Bruce K. Armstrong, MD, PhD*
School of Public Health
The University of Sydney
Sydney, New South Wales, Australia

Matthew P. Banegas, PhD, MPH
Center for Health Research
Kaiser Permanente
Portland, Oregon

Jill Barnholtz-​Sloan, PhD
Case Comprehensive Cancer Center
Case Western Reserve University School of Medicine
Cleveland, Ohio

Dorothea T. Barton, MD
Department of Surgery
Dartmouth-​Hitchcock Medical Center
Lebanon, New Hampshire

Laura E. Beane Freeman, PhD
Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Amy Berrington de González, DPhil

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Marianne Berwick, PhD
Department of Internal Medicine
University of New Mexico
Albuquerque, New Mexico

Smita Bhatia, MD, MPH
Institute of Cancer Outcomes and Survivorship
University of Alabama at Birmingham, School of Medicine
Birmingham, Alabama

Amanda Black, PhD, MPH
Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

A. Blair, PhD
Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

William J. Blot, PhD
Vanderbilt-​Ingram Cancer Center
Nashville, Tennessee

* Retired

xi


xi

xii

Contributors

Melissa L. Bondy, PhD

James R. Cerhan, MD, PhD, (Editor)

Department of Medicine, Section of Epidemiology and Population Sciences
Baylor College of Medicine
Houston, Texas

Department of Health Sciences Research
Mayo Clinic
Rochester, Minnesota

AndrÉ Bouville, PhD*

Ellen T. Chang, ScD

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Center for Health Sciences Exponent Inc.

Menlo Park, California

Marie C. Bradley, PhD, MScPH

Hepatitis Research Center
National Taiwan University Hospital
Taipei, Taiwan

Division of Cancer Control and Population Sciences
National Cancer Institute
Bethesda, Maryland

Freddie Bray, PhD

Ding-​Shinn Chen, MD

Wong-​Ho Chow, PhD

Section of Cancer Surveillance
International Agency for Research on Cancer
Lyon, France

Department of Epidemiology
The University of Texas
MD Anderson Cancer Center
Houston, Texas

Alina V. Brenner, MD, PhD

Aaron J. Cohen, MPH, DSc‡

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Health Effects Institute
Boston, Massachusetts

Louise A. Brinton, PhD

Division of Public Health Services
Washington University
St. Louis, Missouri

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Philip John Brooks, PhD
Laboratory of Neurogenetics
National Institute on Alcohol Abuse and Alcoholism, NIH
Bethesda, Maryland

Julia Brotherton, MD, PhD
National HPV Vaccination Program Register
Victorian Cytology Service
East Melbourne, Victoria, Australia

Elizabeth E. Brown, PhD, MPH
Department of Pathology

University of Alabama at Birmingham
Birmingham, Alabama

Marc Bulterys, MD, PhD
HIV/​Hepatitis Department
World Health Organization
Geneva, Switzerland

Kenneth P. Cantor, PhD, MPH*
Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Deanna Alexis Carere, ScD, CGC
Department of Pathology and Molecular Medicine
McMaster University
Hamilton, Ontario, Canada

Brian D. Carter, MPH
Epidemiology Research Program
American Cancer Society
Atlanta, Georgia
* Retired
‡
Consultant/​Contractor

Graham Colditz, MD, DrPH

Linda S. Cook, PhD
Department of Internal Medicine

University of New Mexico
Albuquerque, New Mexico

Janine A. Cooper, PhD
School of Pharmacy
Queen’s University Belfast
Belfast, Northern Ireland

Jennifer M. Croswell, MD, MPH
Patient-​Centered Outcomes Research Institute
Washington, DC

Rochelle E. Curtis, MA
Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Anne E. Cust, PhD
School of Public Health and Melanoma Institute Australia
The University of Sydney
Sydney, New South Wales, Australia

Jack Cuzick, PhD
Wolfson Institute of Preventive Medicine
Queen Mary University of London
London, United Kingdom

Catherine de Martel, MD, PhD
Infections and Cancer Epidemiology Group
International Agency for Research on Cancer

Lyon, France


 xi

Contributors

Michael Dean, PhD

David Forman, PhD

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

International Agency for Research on Cancer
Lyon, France

Susan S. Devesa, PhD*

International Agency for Research on Cancer
Lyon, France

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Graça M. Dores, MD§
Division of Cancer Epidemiology and Genetics
National Cancer Institute

Bethesda, Maryland

Clifford E. Douglas, JD
Center for Tobacco Control
American Cancer Society
Atlanta, Georgia

Jeffrey Drope, PhD
Economic & Health Policy Research
American Cancer Society
Atlanta, Georgia

Donatus U. Ekwueme, PhD, MS
National Center for Chronic Disease Prevention and Health Promotion
Centers for Disease Control and Prevention
Atlanta, Georgia

A. Heather Eliassen, ScD
Brigham & Women’s Hospital and Harvard Medical School
Harvard TH Chan School of Public Health
Boston, Massachusetts

Hashem B. El-​Serag, MD, MPH
Gastroenterology and Hepatology
Baylor College of Medicine
Houston, Texas

Eric A. Engels, MD
Division of Cancer Epidemiology and Genetics
National Cancer Institute

Bethesda, Maryland

Jacques Ferlay, MSc
Section of Cancer Surveillance
International Agency for Research on Cancer
Lyon, France

Catterina Ferreccio, MD, MPH
Division of Public Health and Family Medicine
School of Medicine, Pontificia Universidad Católica de Chile
Santiago, Chile

Maria Feychting, PhD
Institute of Environmental Medicine
Karolinska Institutet
Stockholm, Sweden

Jonine D. Figueroa, PhD
Usher Institute of Population Health Sciences and Informatics,
CRUK Edinburgh Centre
University of Edinburgh
Edinburg, United Kingdom
* Retired
§
Adjunct

Silvia Franceschi, MD

Joseph F. Fraumeni, Jr., MD
Division of Cancer Epidemiology and Genetics

National Cancer Institute
Bethesda, Maryland

Neal D. Freedman, PhD
Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Søren Friis, MD
Statistics and Pharmacoepidemiology
Danish Cancer Society Research Center
Copenhagen, Denmark

Morten Frisch, MD, PhD, DrSci(Med)
Department of Epidemiology Research
Statens Serum Institut
Copenhagen, Denmark

Susan M. Gapstur, PhD, MPH
Epidemiology Research Program
American Cancer Society
Atlanta, Georgia

Montserrat Garcia-​Closas, MD, DrPH
Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Mia M. Gaudet, PhD
Epidemiology Research Program

American Cancer Society
Atlanta, Georgia

Gretchen L. Gierach, PhD
Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Peter Gies, PhD
Australian Radiation Protection and Nuclear Safety Agency
Melbourne, Victoria, Australia

Edward L. Giovannucci, MD, ScD
Departments of Nutrition and Epidemiology
Harvard TH Chan School of Public Health
Boston, Massachusetts

Sally L. Glaser, PhD
Cancer Prevention Institute of California
Fremont, California

Lynn Goldman, MD, MS, MPH
Milken Institute School of Public Health
George Washington University
Washington, DC

xiii


vxi

xiv

Contributors

Steven N. Goodman, MD, PhD

Henrik Hjalgrim, MD, PhD, DrSci(med)

Department of Medicine, Clinical and Translational Research
Stanford University School of Medicine
Stanford, California

Department of Epidemiology Research
Statens Serum Institut
Copenhagen, Denmark

Anala Gossai, MPH, PhD

Katherine A. Hoadley, PhD

Geisel School of Medicine
Dartmouth College
Hanover, New Hampshire

Department of Genetics, Lineberger Comprehensive Cancer Center
University of North Carolina at Chapel Hill
Chapel Hill, North Carolina

Adèle C. Green, MD, PhD

Jonathan N. Hofmann, PhD

Population Health Division
QIMR Berghofer Medical Research Institute
Brisbane, Queensland, Australia

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Andrew E. Grulich, PhD

Robert N. Hoover, MD, ScD

Kirby Institute
The University of New South Wales
Sydney, New South Wales, Australia

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Gery P. Guy Jr, PhD, MPH

Frank B. Hu, MD, PhD

National Center for Chronic Disease Prevention and Health
Promotion
Centers for Disease Control and Prevention

Atlanta, Georgia

Departments of Nutrition and Epidemiology
Harvard TH Chan School of Public Health
Boston, Massachusetts

Laurel A. Habel, PhD, MPH

Department of Public Health
North Dakota State University
Fargo, North Dakota

Division of Research
Kaiser Permanente Northern California
Oakland, California

Christopher A. Haiman, ScD, (Editor)
Department of Preventive Medicine
Keck School of Medicine of University of Southern California
Los Angeles, California

Susan E. Hankinson, ScD
Department of Biostatistics and Epidemiology
University of Massachusetts
Amherst, Massachusetts

Russell P. Harris, MD, MPH§
Lineberger Comprehensive Cancer Center
University of North Carolina School of Medicine
Chapel Hill, North Carolina

Mia Hashibe, PhD
Department of Family and Preventive Medicine
Huntsman Cancer Institute, University of Utah School of Medicine
Salt Lake City, Utah

S. Jane Henley, MSPH
Division of Cancer Prevention and Control
US Centers for Disease Control and Prevention
Atlanta, Georgia

Rolando Herrero, MD, PhD
Prevention and Implementation Group
International Agency for Research on Cancer
Lyon, France

Allan Hildesheim, PhD
Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland
§

adjunct

Rick J. Jansen, MS, PhD

Ahmedin Jemal, DVM, PhD
Surveillance and Health Services Research Program
American Cancer Society
Atlanta, Georgia

Fengyi Jin, PhD
Kirby Institute
The University of New South Wales
Sydney, New South Wales, Australia

Lisa G. Johnson, PhD, MPH
Division of Public Health Sciences
Fred Hutchinson Cancer Research Center
Seattle, Washington

Dean P. Jones, PhD
Department of Medicine
Emory University
Atlanta, Georgia

Margaret R. Karagas, PhD
Department of Epidemiology
Geisel School of Medicine at Dartmouth
Hanover, New Hampshire

Ichiro Kawachi, MD, PhD
Department of Social and Behavioral Sciences
Harvard School of Public Health
Boston, Massachusetts

Sarah Keadle, PhD, MPH
Kinesiology Department
California Polytechnic State University
San Luis Obispo, California

 xv

Contributors

NaNa Keum, ScD

Yanhong Liu, PhD

Department of Nutrition
Harvard TH Chan School of Public Health
Boston, Massachusetts

Department of Medicine, Section of Epidemiology
and Population Sciences
Baylor College of Medicine
Houston, Texas

Sally Kinsey, MD, FRCP
Department of Pediatric Hematology
Leeds Teaching Hospitals NHS Trust
Leeds, United Kingdom

W. Thomas London, MD*

Cari M. Kitahara, PhD

Robyn M. Lucas, MBChB, MPH&TM, PhD

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Radiation Health Services Branch
The Australian National University
Canberra, The Australian Capital Territory, Australia

Jill Koshiol, PhD

Margaret M. Madeleine, PhD

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Division of Public Health Sciences
Fred Hutchinson Cancer Research Center
Seattle, Washington

Stella Koutros, PhD

Christopher J. Maher, PhD

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

McDonnell Genome Institute
Washington University School of Medicine

St. Louis, Missouri

Peter Kraft, PhD

Elaine R. Mardis, PhD

Departments of Epidemiology and Biostatistics
Harvard TH Chan School of Public Health
Boston, Massachusetts

McDonnell Genome Institute
Washington University School of Medicine
St. Louis, Missouri

Barnett S. Kramer, MD, MPH

Charles E. Matthews, PhD

Division of Cancer Prevention
National Cancer Institute
Bethesda, Maryland

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Candyce Kroenke, ScD, MPH

Marjorie L. McCullough, ScD, RD

Division of Research
Kaiser Permanente Northern California
Oakland, California

Epidemiology Research Program
American Cancer Society
Atlanta, Georgia

Qing Lan, MD, MPH

Katherine A. McGlynn, PhD

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

I-​Min Lee, MBBS, ScD

Angela L. W. Meisner, MPH

Brigham and Women’s Hospital
Harvard Medical School
Boston, Massachusetts

New Mexico Tumor Registry
University of New Mexico

Albuquerque, New Mexico

Tracy Lightfoot, PhD

Kyriaki Michailidou, PhD

Department of Health Sciences
University of York
York, United Kingdom

Department of Electron Microscopy/​Molecular Pathology
The Cyprus Institute of Neurology and Genetics
Nicosia, Cyprus

Sara Lindström, PhD

Lisa Mirabello, PhD

Department of Epidemiology
University of Washington
Seattle, Washington

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Martha S. Linet, MD, MPH, (Editor)

Karobi Moitra, PhD

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Trinity Washington University
Washington, DC

* Retired

Fox Chase Cancer Center
Philadelphia, Pennsylvania

xv


xvi

xvi

Contributors

Steven C. Moore, PhD

Alpa V. Patel, PhD

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Epidemiology Research Program

American Cancer Society
Atlanta, Georgia

Lindsay M. Morton, PhD

Kathryn L. Penney, ScD

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

ScD Department of Medicine
Brigham and Women’s Hospital /​Harvard Medical School
Boston, Massachusetts

Raul Murillo, MD

Gloria M. Petersen, PhD

Prevention and Implementation Group
International Agency for Research on Cancer
Lyon, France

Department of Health Sciences Research
Mayo Clinic College of Medicine
Rochester, Minnesota

Rachel E. Neale, PhD

Jessica L. Petrick, PhD

Population Health Division
QIMR Berghofer Medical Research Institute
Brisbane, Queensland, Australia

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Heather H. Nelson, MPH, PhD

Susan Picton, BMBS, FRCPCH

Division of Epidemiology, Masonic Cancer Center
University of Minnesota, Twin Cities
Minneapolis, Minnesota

Department of Pediatric Oncology
Leeds Teaching Hospitals NHS Trust
Leeds, United Kingdom

Marian L. Neuhouser, PhD

Brandon Pierce, PhD

Division of Public Health Sciences
Fred Hutchinson Cancer Research Center
Seattle, Washington

Departments of Public Health Sciences and Human Genetics

University of Chicago
Chicago, Illinois

Robert Newton, PhD

Martyn Plummer, PhD

MRC/​UVRI Uganda Research Unit
Entebbe, Uganda

International Agency for Research on Cancer
Lyon, France

Reiko Nishihara, PhD

I. Mary Poynten, PhD Kirby Institute

Department of Pathology
Brigham and Women’s Hospital
Boston, Massachusetts

The University of New South Wales
Sydney, New South Wales, Australia

Michael A. O’Rorke, PhD

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

School of Medicine Dentistry and Biomedical Sciences
Queen’s University Belfast
Belfast, Northern Ireland

Andrew F. Olshan, PhD
Department of Epidemiology
Gillings School of Global Public Health
University of North Carolina
Chapel Hill, North Carolina

Quinn T. Ostrom, MA, MPH
Case Comprehensive Cancer Center
Case Western Reserve University School of Medicine
Cleveland, Ohio

Julie R. Palmer, ScD
Slone Epidemiology Center at Boston University
Boston, Massachusetts

D. Maxwell Parkin, MD, DSc

Ludmila Prokunina-​Olsson, PhD

Mark P. Purdue, PhD
Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Preetha Rajaraman, PhD
Division of Cancer Epidemiology and Genetics

National Cancer Institute
Bethesda, Maryland

Ewa Rajpert-​De Meyts, MD, PhD
Department of Growth & Reproduction
Copenhagen University Hospital Rigshospitalet
Copenhagen, Denmark

Judy R. Rees, BM, BCh, PhD

Nuffield Department of Public Health University of Oxford
Oxford, United Kingdom

Department of Epidemiology
Geisel School of Medicine at Dartmouth
Hanover, New Hampshire

Julie Parsonnet, MD

Juan Carlos Roa, MD, Msc

Department of Medicine
Stanford University School of Medicine
Stanford, California

Department of Pathology
School of Medicine, Pontificia Universidad Católica de Chile
Santiago, Chile

 xvi

Contributors

Eve Roman, PhD

Martyn T. Smith, PhD

Department of Health Sciences
University of York
York, United Kingdom

School of Public Health
University of California at Berkeley
Berkeley, California

Nathaniel Rothman, MD, MPH

Mingyang Song, MD, ScD

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Clinical and Translational Epidemiology Unit and Division
of Gastroenterology
Massachusetts General Hospital and Harvard Medical School
Boston, Massachusetts

Jonathan M. Samet, MD

Department of Preventive Medicine
Keck School of Medicine of University of Southern California
Los Angeles, California

John J. Spinelli, PhD Cancer Control Research

Ambika Satija, ScD

Avrum Spira, MD, MSc

Department of Nutrition
Harvard TH Chan School of Public Health
Boston, Massachusetts

Division of Computational Biomedicine
Boston University School of Medicine
Boston, Massachusetts

Sharon A. Savage, MD

Janet L. Stanford, PhD

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Division of Public Health Sciences
Fred Hutchinson Cancer Research Center
Seattle, Washington

Ghislaine Scelo, PhD

Andreas Stang, MD, MPH

Genetic Epidemiology Group
International Agency for Research on Cancer
Lyon, France

Institute of Medical Informatics, Biometry and Epidemiology
University Hospital Essen
Essen, Germany

Arthur B. Schneider, MD, PhD*

Kyle Steenland, PhD

Section of Endocrinology, Diabetes and Metabolism
University of Illinois at Chicago College of Medicine
Chicago, Illinois

Rollins School of Public Health Emory University
Atlanta, Georgia

David Schottenfeld, MD, MSc (Editor)*

School of Public Health
University of California Berkeley
Berkeley, California

Department of Epidemiology

University of Michigan School of Public Health
Ann Arbor, Michigan

Mary Schubauer-​Berigan, PhD
Division of Surveillance Hazard Evaluation and Field Studies
Centers for Disease Control and Prevention
Atlanta, Georgia

Joachim Schüz, PhD
Section of Environment and Radiation
International Agency for Research on Cancer
Lyon, France

Amy L. Shafrir, ScD
Division of Adolescent and Young Adult Medicine
Boston Children’s Hospital
Boston, Massachusetts

Mark E. Sherman, MD
Health Sciences Research
Mayo Clinic College of Medicine
Jacksonville, Florida

Debra T. Silverman, ScD, ScM
Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Terry Slevin, MPH
Cancer Council Western Australia

Perth, Western Australia, Australia
* Retired

British Columbia Cancer Agency
Vancouver, British Columbia, Canada

Craig M. Steinmaus, MD, MPH

Erich M. Sturgis, MD, MPH
Department of Head and Neck Surgery
The University of Texas MD Anderson Cancer Center
Houston, Texas

Catherine M. Tangen, DrPH
Division of Public Health Sciences
Fred Hutchinson Cancer Research Center
Seattle, Washington

Robert E. Tarone, PhD*
International Epidemiology Institute
Rockville, Maryland

Michael J. Thun, MD, MS (Editor)*
Epidemiology and Surveillance Research
American Cancer Society
Atlanta, Georgia

William D. Travis, MD
Department of Pathology
Memorial Sloan Kettering Cancer Center

New York, New York

Melissa A. Troester, PhD
Department of Epidemiology
Lineberger Comprehensive Cancer Center
University of North Carolina at Chapel Hill
Chapel Hill, North Carolina

xvii


xvii

xviii

Contributors

Rebecca Troisi, ScD, MA

David C. Whiteman, MD, PhD

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Population Health Division
QIMR Berghofer Medical Research Institute
Brisbane, Queensland, Australia

Shelley S. Tworoger, PhD

Charles Wiggins, PhD, MPH

Harvard Medical School and the Brigham and Women’s Hospital
Harvard TH Chan School of Public Health
Boston, Massachusetts

Department of Internal Medicine
University of New Mexico
Albuquerque, New Mexico

Celine M. Vachon, PhD

Christopher P. Wild, PhD

Department of Health Sciences Research
Mayo Clinic
Rochester, Minnesota

Director’s Office
International Agency for Research on Cancer
Lyon, France

Claire M. Vajdic, PhD

Walter C. Willett, MD, DrPH

Centre for Big Data Research in Health
University of New South Wales
Sydney, New South Wales, Australia

Department of Nutrition
Harvard TH Chan School of Public Health
Boston, Massachusetts

Roel Vermeulen, PhD

Deborah M. Winn, PhD

Institute for Risk Assessment Sciences
Utrecht University
Utrecht, The Netherlands

Division of Cancer Control and Population Sciences
National Cancer Institute
Bethesda, Maryland

Fawn D. Vigneau, JD, MPH

Kana Wu, MD, PhD

Wayne State University School of Medicine
Karmanos Cancer Institute
Detroit, Michigan

Department of Nutrition
Harvard TH Chan School of Public Health
Boston, Massachusetts

Teresa W. Wang, PhD

K. Robin Yabroff, PhD

Division of Computational Biomedicine
Boston University School of Medicine
Boston, Massachusetts

Division of Cancer Control and Population Sciences
National Cancer Institute
Bethesda, Maryland

Mary H. Ward, PhD

Shelia Hoar Zahm, ScD‡

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Division of Cancer Epidemiology and Genetics
National Cancer Institute
Bethesda, Maryland

Noel S. Weiss, MD, DrPH

Chenan Zhang, PhD

Department of Epidemiology
University of Washington
Seattle, Washington

Department of Epidemiology and Biostatistics
University of California, San Francisco
San Francisco, California

‡

Consultant/​Contractor


x i

Preface

The Schottenfeld and Fraumeni text on Cancer Epidemiology and Prevention has served as the
premier reference text for population research on the causes and prevention of cancers since the
publication of the first edition in 1982 (Schottenfeld and Fraumeni, 1982). It is written for colleagues pursuing careers in research in cancer epidemiology and, more broadly, in preventive
oncology. The founding editors, Dr. David Schottenfeld, now emeritus professor of epidemiology at the University of Michigan, and Dr. Joseph Fraumeni, recently retired as the director of the
Division of Cancer Epidemiology and Genetics at the National Cancer Institute (NCI), updated
their landmark text in 1996 and 2006 (Schottenfeld and Fraumeni, 1996, 2006).
The current edition again provides a comprehensive update of research advances in cancer
epidemiology, prevention, and related fields in the past 10–​15 years, and honors the founding
editors in the title. The new editorial team is led by Dr. Michael Thun (editor-​in-​chief), formerly
with the American Cancer Society, and includes four senior co-​editors: Drs. Martha Linet from
NCI, James Cerhan from the Mayo Clinic, Christopher Haiman from the University of Southern
California, and David Schottenfeld. We are also deeply indebted to the internationally recognized experts who authored the 63 chapters. Without their generous effort and commitment, this
updated synthesis would not be possible.

xix

x


 1

1Introduction
MICHAEL J. THUN, MARTHA S. LINET, JAMES R. CERHAN,
CHRISTOPHER A. HAIMAN, AND DAVID SCHOTTENFELD

I

n this introduction, we provide an overview of the text and highlight
cross-​cutting developments and new opportunities that are transforming our understanding of the causes and prevention of cancer. As
in previous editions, the text is grouped into five major parts: “Basic
Concepts,” “The Magnitude of Cancer,” “The Causes of Cancer,”
“Cancers by Tissue of Origin,” and “Cancer Prevention and Control.”
Part I first describes research advances in understanding “the biology of neoplasia,” including the progressive disruption of genetic and
epigenetic controls that regulate cell growth, division, and survival
(Chapter 2). Advances in high-​throughput technologies have greatly
expanded the ability to identify germline and somatic mutations and
to relate these to etiology, prognosis, and treatment. Tumor classification is also changing for certain cancers, as data on the molecular
features and lineage of the neoplastic cells is combined with information on the primary anatomic location and the morphologic, histopathologic and clinical characteristics of the tumor (Chapter 3). The
“landscape” of genomic and epigenomic alterations in tumor tissue
has been cataloged for multiple human cancers (Chapter 4), revealing
both the singularity of individual cancer genomes and the commonality of genetic alterations that drive cancer in different tissues. Chapter
5 describes advances in research on inherited genomic variants that
affect cancer risk. Genome-​wide association studies (GWAS) have
identified more than 700 germline loci associated with increased or
decreased risk for various types of cancer, although the risk estimates

for almost all are small to modest. Innovations in genomics and other
“OMIC” technologies are identifying biomarkers that reflect internal
exposures, biological processes, and intermediate outcomes in large
population studies (Chapter 6). While research in many of these areas
is still in its infancy, mechanistic and molecular insights are extending
the traditional criteria for inferring causation in epidemiologic studies
of cancer (Chapter 7).
Part 2 of the book discusses the global public health impact of cancer and its relationship to demographic trends, changing risk factors,
socioeconomic disparities, and economic development. It considers
the direct and indirect costs of cancer in the United States to illustrate
the economic burden in a high income country. Parts 3–​5 of the book
discuss the growing list of exposures known to affect cancer risk, the
epidemiology of over 30 types of cancer by tissue of origin, and the
encouraging progress in cancer prevention and control. Major developments in these areas are discussed below, beginning with those that
affect the public health impact of cancer.

MAJOR NEW DEVELOPMENTS
Global Trends in Cancer Risk and Burden
Part II, “The Magnitude of Cancer,” provides a global public health
perspective on cancer. The human and economic costs of cancer are
increasing worldwide (http://​globocan.iarc.fr). The World Health
Organization (WHO) estimates that 14 million new cases and 8.2 million deaths from cancer occurred in 2012. This burden is projected to
increase to 24 million cases and 13 million deaths annually by 2035
(Ferlay et al., 2013). Chapter 8 decribes the disproportionate increase
in the cancer burden in low-​and middle-​income countries (LMICs),
which can least afford additional health-​related, social and financial

costs. In 2012, these countries accounted for over half (57%) of all
incident cancers; this is projected to increase to nearly two-​thirds
(65%) by 2035. Much of the increase will result from the growth and

aging of populations, since LMICs currently comprise about 80% of
the world’s population, and large numbers of young adults are now
surviving to older ages, when cancer becomes more common. In
addition to the effect of demographic changes, cancer incidence and
mortality rates are increasing in LMICs because of the widespread
adoption of Western patterns of diet, physical inactivity, excess body
fat, delayed reproduction, and tobacco smoking, especially of manufactured cigarettes. As countries advance economically, the incidence
rates of cancers traditionally associated with Westernization (e.g.,
breast, colorectum, lung, and prostate) increase more rapidly than the
decrease in cancers caused partly or wholly by infectious agents (e.g.,
stomach, liver, uterine cervix). Survival after a diagnosis of cancer is
also lower in LMICs than in high-​resource countries, because of later
stage at diagnosis, a higher proportion of tumors diagnosed clinically
rather than incidentally, and limited access to standard and state-​of-​
the-​art treatment protocols.
In economically developed countries, the incidence rates of most
cancers are either stabilizing at a high level or decreasing, depending on the temporal trends of underlying risk factors and utilization of
cancer screening. Despite the decreasing rates, the disease burden, or
number of cancer cases and deaths, continues to increase. The increasing burden results from the aging and growth of populations, and the
decline in competing causes of death from circulatory and infectious
diseases. Mortality rates are decreasing more rapidly than incidence
rates for many cancer sites due to a combination of prevention, early
detection, and improvements in treatment.
Part III, “The Causes of Cancer,” discusses 15 broad categories
of exposure that affect cancer risk. These include exposures that are
typically considered “environmental” by the public (chemical carcinogens, ionizing radiation, occupational exposures, pollutants in air and
drinking water), as well as exposures that are less widely recognized as
carcinogenic (infectious agents, metabolic factors, body composition,
reproductive and other hormones, pharmaceutical drugs, and immunological conditions). All of these exposures are “environmental” in
the sense that they are acquired after conception rather than inherited.

Some are genotoxic and damage the structure of DNA or alter DNA
repair; others modify gene expression, induce oxidative stress and/​
or chronic inflammation, suppress host immunity, immortalize cells,
modulate receptors, and/​or alter cell proliferation, cell death, or nutrient supply (Smith et al., 2016).
Although some exposures are conventionally perceived as “lifestyle choices”, they are by no means entirely voluntary. For example,
behavioral risk factors such as tobacco smoking, energy imbalance, and physical inactivity are strongly influenced by factors in
the social, economic, and cultural environment, beginning in early
childhood. Physiologic addiction is a major driver of tobacco use at
all ages.
Part IV of the book describes “Cancers by Tissue of Origin” for
33 anatomic sites, multiple primary tumors, and cancers in children.
Rapid advances in discovering the molecular events that drive certain
forms of cancer are transforming clinical diagnoses and treatment, and
affecting tumor classification. This will influence future endpoints in
etiologic studies and population-​based cancer surveillance.

1


2

2Introduction
Part V, “Cancer Prevention and Control,” discusses the impact of
interventions that effectively reduce carcinogenic exposures or disrupt
the multistage progression of tumors. It focuses on interventions that
demonstrably reduce cancer risk in the general population, rather than
in special circumstances or high-​risk subgroups. Examples of these
are discussed in Chapters 61–​63. In all cases, the design and implementation of preventive measures require translational research to
ensure safety, optimize feasibility and impact, and critically evaluate
all stages of the process.

Cancer Prevention and Control
A growing number of population-​level preventive interventions are
proving to be highly effective, as confirmed by the decreases in incidence as well as mortality rates from certain cancers (Chapters 61–​63).
Tobacco control has reduced the age-​standardized incidence rate of
lung cancer by up to 40% among men in high-​and middle-​income
countries. Increased screening for colorectal cancer and removal of
precursor lesions is credited for the 30% decrease in the incidence
rates at this site in the United States. Universal neonatal vaccination
against hepatitis B virus (HBV) has markedly decreased the prevalence of chronic HBV infection and liver cancer at younger ages in
high-​risk areas of East Asia and will yield maximal benefits against
cancer in the future. The development of safe and effective vaccines
against human papillomavirus (HPV) and less expensive and less onerous screening tests for cervical cancer have greatly expanded opportunities to prevent HPV-​related cancers among women in many LMICs.
Increased funding is becoming available for application research and
cancer preventive services in LMICs. Cancer prevention presents both
opportunities and challenges, as discussed in Part V of the text. The
best practices developed for tobacco control provide an encouraging model of how health-​related policies can address the behavioral
causes of cancer. However, these must be tailored to fit the particular
social, economic, and other considerations that affect the exposure
(Chapter 61).

Advances in Genomics and other OMICs
Technological advances in high-​
throughput genotyping/​
sequencing and gene expression arrays have transformed research on both
inherited (germline) susceptibility variants and the largely acquired
(somatic) mutations in tumor tissue. Epidemiologic studies of cancer
genetics have focused mainly on germline variants associated with
cancer risk and etiology, whereas clinical and basic researchers have
characterized the landscape of somatic alterations in tumor cells that

drive the development and progression of cancer.

Germline Susceptibility Variants

The tools to identify inherited genetic susceptibility variants have
advanced enormously since publication of the previous edition of
this text in 2006. At that time, studies involved either high-​risk families or the evaluation of a small number of pre-​specified “candidate
genes” in case-control studies of sporadic cancers in the general population. The candidate gene approach was largely unsuccessful in
identifying robust associations for several reasons, including small
sample size, limited statistical power, failure to account for multiple testing (generating negative and false positive results, respectively), and limited biologic knowledge to inform the selection of
candidate genes. Following the completion of the Human Genome
Project in 2003, genome-​wide maps of single nucleotide polymorphisms (SNPs) became available. Advances in high-​
throughput
genotyping technology, combined with knowledge about the structure of genetic linkage disequilibrium, created opportunities to conduct exploratory (hypotheis-​free or “agnostic”) surveys across the
entire genome. Over the past decade, GWAS have robustly identified more than 700 common (i.e., minor allele frequency >5%) susceptibility loci associated with cancer risk, as discussed for specific
sites in Part IV, “Cancers by Tissue of Origin.” Because GWAS test
millions of alleles across the genome, they require stringent criteria

(“genome-​wide significance”), large sample size, and replication
in more than one study to exclude chance associations. Most of the
associations identified through GWAS are modest (per allele ORs:
1.5–​2.0) or weak (ORs <1.5), but in aggregate these loci can distinguish a wide range of risk in the population, thus providing opportunities for targeted screening and prevention. While our current
knowledge regarding germline risk comes from studies in populations of European ancestry, the identification of population-​specific
risk loci highlights the importance of conducting GWAS in diverse
racial and ethnic populations.
Statistical modeling suggests that, for many cancers, additional
variants remain to be identified, yet the search for variants with smaller
effect sizes, as well as less common variants, drives the need for
even larger studies. With the recent development of next-​generation
sequence technology, it is now practical to sequence whole exomes

(coding regions plus regulatory regions) and whole genomes in
population-​and family-​based studies in the search for heritability not
identified through common variation in GWAS.
An important limitation of GWAS is biological interpretation,
as the vast majority of risk variants revealed through GWAS are in
non-​coding genetic sequences. Functional analyses are underway to
address this issue. The process is time-​consuming, however, since
it incorporates new bioinformatics tools and a comparison of gene
expression in tumor and normal tissue to localize the functional SNP
and ultimately the affected gene.
There has as yet been little progress in identifying interactions
between inherited germline loci identified through GWAS and acquired
“environmental” risk factors. Candidate gene studies have documented gene–​environment interactions between tobacco smoking and
the slow NAT-​2 acetylation phenotype for bladder cancer (Chapter 52)
and between alcohol consumption and slow ADH1B metabolizers for
esophageal cancer (Chapter  30). However, much larger GWAS with
more precise measures of exposure and risk will be needed to assess
other, subtler gene–​environment interactions.

Somatic Genomic Alterations

Most of the somatic genomic alterations, including mutations,
indels, copy number alterations, and chromosomal rearrangements,
that drive neoplastic progression in tumor tissue are acquired rather
than inherited. As mentioned, the Human Cancer Genome Project
and other international laboratory and clinical collaborations have
characterized so-​called driver mutations (i.e., those which confer
growth advantage to a mutated cell line) for multiple types of human
cancer (Chapter 4). These mutations represent the events involved in
the multistage development of particular forms of cancer (Armitage

and Doll, 2004; Hornsby et al., 2007; Wu et al., 2016). It is noteworthy that discoveries in somatic mutations over the past three decades
provide strong support for the theory of multistage carcinogenesis
that was proposed by Armitage and Doll, 10 years before elucidation
of the structure of DNA, and over 30 years before the identification
of the first proto-​oncogenes and tumor suppression genes (Armitage
and Doll, 1954). Sequencing studies have also implicated epigenetic
modification as a major source of alterations in cancer (Chapters 2
and 4).
Variable combinations of genetic and epigenetic abnormalities
account for the phenotypic heterogeneity within and among cancers.
Molecular characterization of tumors is increasingly used to predict
prognosis and to guide the use of targeted therapies for individual cancer patients. These markers are only beginning to be evaluated and
integrated into large-​scale epidemiologic studies, yet they are already
changing the taxonomy of some types of cancers and are likely to profoundly affect future etiologic studies (Chapter  3). There has been
some progress in efforts to link specific classes of somatic mutations, such as the mutational signatures of ultraviolet (UV) radiation,
tobacco smoke, and oncogenic viruses, to established carcinogenic
exposures (Chapters  2 and 4). These molecular signatures may, in
the future, identify the causal exposure(s) for cancer in individuals as
well as populations. Hopefully this goal will motivate interdisciplinary
collaborations between epidemiologists, cancer prevention scientists,
geneticists, cancer biologists, and clinicians.


 3

Introduction

Other OMICs

While genomic research is the poster child for the value of agnostic, comprehensive explorations of germline variants associated with

cancer, other areas of OMIC research are moving toward this goal
(Chapter  6). The development of technologies to screen many thousands of analytes related to gene expression (e.g., RNAseq), epigenetics (methylation; ChIP-​Seq), metabolomics, and the microbiome
will open new opportunities to identify the connections between exposures and the biologic effects that mediate carcinogenesis. Various
OMIC technologies are at different stages of development. One of the
more advanced efforts along these lines is the identification of hormone metabolites that influence breast cancer risk, illustrating the
potential of these new technologies (Chapter 22).

OUTCOMES AND EXPOSURES
Changing Taxonomy of Cancer
Accurate and reproducible classification of neoplastic diseases is
essential for advances in diagnosis and treatment, for quantifying
geographic, temporal, and demographic variations in incidence, and
for identifying etiologic relationships and mechanisms. Tumor classification has historically been based on the primary anatomic site and
morphology for most solid tumors and on histologic characteristics for
leukemias. Classification systems have evolved to incorporate information on morphology, genetics, cell lineage, developmental characteristics, and an array of molecular, clinical, and etiologic factors
(ICD-​O-​3, 2013) (Fritz et al., 2000).
While the refinement of cancer endpoints based on molecular or
other characteristics will potentially increase the ability of etiologic
studies to detect associations with specific tumor subtypes, it also
poses serious challenges. Very large studies will be needed for both
discovery and replication. Even well-​established tumor markers are
not measured uniformly in all patients. Newer classification systems
based on molecular features have generally been evaluated in only a
few hundred sporadic cancers, with little consideration of patient or
population characteristics. Clonal heterogeneity within tumors and
changes in tumor pathology during treatment further complicate classification (Norum et al., 2014). While the use of automated algorithms
and computer-​based image analysis is increasing among pathologists,
these methods and the assessment of reproducibility and validity may
not be reported to clinicians, epidemiologists, and others using the data.
Population-​based cancer registries are already challenged by efforts to

keep abreast of changing tumor classifications, especially when the
new criteria are not uniformly applied in a standardized manner.

Exposures and Exposure Measurement
More than 100 different agents and exposures are now designated as
causally related to cancer in humans (Group I) by the International
Agency for Research on Cancer (IARC). Variations in the prevalence
and intensity of these exposures account for the striking geographic
and temporal variations in the occurrence of many types of cancer.
While many exposures such as tobacco, alcohol, and numerous industrial chemicals have long been classified as human carcinogens, exposure patterns change, new agents are introduced, the ability to measure
exposures or outcomes progresses, and the quality, quantity, and/​or
specificity of evidence improves. For example, the contining global
increase in obesity, metabolic syndrome, and type II diabetes, combined with improvements in laboratory assays to measure hormones
in large population studies, has created new opportunities to study
metabolic and hormonal effects on cancer. Thus, the discussion of
“Hormones and Cancer” (Chapter 22) has been expanded to consider
endogenous as well as exogenous exposures and peptide hormones
(insulin, insulin-​like growth factors, growth hormone, leptin, adiponectin, resistin, ghrelin, etc.) in addition to steroidal sex hormones.
Several agents recently classified as Group  1 human carcinogens
affect massive numbers of people. These include outdoor air pollution

3

(Chapter 17), the combustion of coal as household fuel (Chapters 16,
17, and 28), diesel engine exhaust (Chapter  17), the consumption
of red and processed meat (Chapter 19), and to a lesser extent, UV-​
emitting tanning beds (Chapter 14). The search for other modifiable
causes of cancer continues. New associations have been reported with
shift work, sedentary behavior (as distinct from physical inactivity),
computed tomography scans during childhood, sun-​sensitizing pharmaceutical drugs, and others. While the studies may be methodologically strong, the evidence for causality is not yet considered definitive.

There is a continuing need to monitor both the immediate and long-​
term effects of more recently implemented medical technologies
and newly developed drugs, products such as cellular phones and e-​
cigarettes, nuclear accidents, exposures occurring in war zones, and
other exposures potentially related to cancer.
Technological advancements in biomarker studies will allow more
comprehensive examination of an individual’s metabolome, microbiome, genome, epigenome, and exposome. The use of specific biomarkers to assess internal exposures and identify children, adolescents, and
other subsets of individuals who may be particularly susceptible to the
factor being investigated (for example, dietary exposure, pesticide acting as a hormonal disruptor, or medication) could increase our ability
to detect complex exposure–​disease relationships.

ESTIMATES OF ATTRIBUTABLE FRACTION
Epidemiologists have long debated the fraction of cancer cases or
deaths that could be avoided by preventive interventions (Chapter 61).
Estimates of the percent of cancer deaths that could theoretically be
avoided, if the exposures were eliminated, range from 50% to 80%,
although the potential for primary prevention differs for incidence and
mortality, by geographic region, gender, and attained age (Whiteman
and Wilson, 2016). About half of the deaths that could be avoided in
principle relate to 11 potentially avoidable risk factors, including the
behavioral risk factors discussed above. The attributable fraction estimates are predicated on the idea that cancer risk is largely acquired,
rather than inherited. Inherited factors do contribute to the variation in
risk among individuals, but they cannot account for the large temporal changes in risk within countries, or the differences in risk among
migrants who move from one country to another.
Chapter 19, “Diet and Nutrition,” provides the first estimates of the
fraction of all cancers attributable to diet, in combination with or separate from overweight and physical inactivity. The authors estimated
the total as about 20% overall, which is weaker than previously estimated. The proportion contributed by dietary composition independent of adiposity is less clear because some dietary factors have yet to
be identified or established with sufficient certainty, and associations
are likely underestimated because of measurement error or misspecification of temporal relationships. The authors estimate an etiologic
contribution of 5%–​12% for dietary composition alone, but suggest

that this could be appreciably higher when considering nutrient and
genetic interactions.

ONGOING CHALLENGES
Tumor Diagnosis and Classification
In LMICs, the completeness and specificity of tumor diagnosis varies depending on economic resources and medical infrastructure.
Less than 10% of people in Africa and South America are covered
by population-​
based tumor registries (Chapter  8). In high-​
income
countries, tumor classification is more advanced because of earlier
application of diagnostic innovations and revised classification systems. Classification systems evolve with the introduction of new histochemical and molecular markers and advances in understanding tumor
biology. Even in high-​income countries, there is wide variability in
the proportion of tumors incompletely or inadequately characterized.
Molecular profiling at major cancer centers may include a range of
established tumor markers, exome or whole genome sequencing, copy


4

4Introduction
number, messenger and micro RNA sequencing, DNA methylation,
and proteomics analysis (Hoadley et al., 2014). While this information
may be useful clinically, it is not yet available for most cancer patients,
nor are the data routinely incorporated into cancer surveillance systems (Chapter 8). Even cancers that have undergone intensive multidisciplinary review to improve classification, such as hematopoietic
and lymphoproliferative malignancies, include subtypes characterized
as “not otherwise specified” or provisionally classified (Swerdlow
et al., 2016). Thus epidemiologic studies of cancer must collect and
archive tumor tissue in order to ensure a uniform, more complete, and
contemporary approach to molecular testing.

Over-​diagnosis
“Over-​diagnosis” refers to the incidental detection of small and/​or
indolent cancers that otherwise might not cause clinical problems
during the patient’s lifetime (Chapter  8). Extreme examples of this
have been the sudden increase in prostate cancer diagnoses following the introduction of PSA screening (Chapter 53), and the increase
in thyroid cancer diagnoses due to screening programs using ultrasound (Chapter 44). Overdiagnosis is most problematic if it leads to
unnecessary treatment and serious adverse effects. The introduction
of new screening tests can also distort temporal trends in incidence
and bias observational studies of the impact of screening (Chapter 63).
The likelihood of over-​diagnosis varies by cancer site and depends
on the baseline incidence and risk characteristics of a population.
An estimated 10–​30% of newly diagnosed breast cancers identified
with screening mammography may reflect over-​diagnosis (Bleyer &
Welch, 2012; Loeb et al., 2014; Vickers et al., 2014). In the absence of
molecular markers that reliably distinguish indolent from aggressive
tumors, clinicians must grapple with the potential for “over-​treatment”
(Chapter 3). Over-​diagnosis poses a greater clinical dilemma for early
stage cancers in internal organs than for premalignant lesions detected
by colorectal or cervical screening, because the treatment is more
invasive.

Exposure Measurement Issues
Regardless of the epidemiologic study design used, it is challenging
to characterize accurately many types of acquired exposures due to
a lack of comprehensive measurements during the relevant exposure
window. This presents a greater problem for some exposures than others. For example, as described in Chapters 19–​21, misclassification of
exposure is of great concern in characterizing patterns of nutrition and
physical activity, especially at earlier stages of life. It presents less of a
problem in studying cigarette smoking, menopausal hormonal therapy,

or exposure to microbial agents, since these exposures can be reasonably well defined qualitatively, and biomarkers exist to supplement
questionnaire data. Certain exposures are experienced in multiple
settings, including workplace, residential, recreational, medical, war
zone, and other settings. Chemical exposures often occur as mixtures
in air or water. Surrogate measures, such as job titles for occupational
exposures and administrative databases for residential exposures, do
not capture variations among individuals or over time.
The timing of exposure is an area of special interest and challenge.
Exposures that occur during a particular time window or a susceptible
stage of development may have adverse effects that are not evident
when exposure occurs later in life. A classic example of this involved
in utero exposure to high doses of the hormone di-​ethyl stilbesterol
(DES) in the daughters of mothers treated to prevent pregnancy complications, who subsequently developed vaginal carcinoma in adolescence (Chapter 49). For breast cancer, it is hypothesized that hormonal
exposures received in utero or immediately postnatally may affect
early stages in tumor development, or that breast tissue may be more
susceptible to certain exposures (e.g., ionizing radiation, cigarette
smoking, or alcohol consumption) during the period between menarche and first full-​term pregnancy, when cells are proliferating but not
fully differentiated (Chapter  45). Similar hypotheses have been proposed for nasopharyngeal cancer and Hodgkin lymphoma in relation

to early childhood infection with Epstein-​Barr virus (Chapters 26 and
39). Although these hypotheses are important, they are difficult to test
without biomarkers or experiments of nature that demarcate the timing
of exposure. Serial acquisition of biological samples over many years
of follow-​up would be informative, although this approach must be
tempered by feasibility and cost considerations.

FUTURE RESEARCH DIRECTIONS
While individual chapters outline future research directions for specific exposures or cancers, we highlight here selected cross-​cutting
issues that apply broadly to many cancers.

Team Science
One of the benefits of GWAS and pooled risk factor studies has been
the formation of multi-​institutional international consortia to share
biospecimens and primary data in order to maximize statistical power
for discovery and robust replication (Boffetta et al., 2007). These consortia have been instrumental in creating new models of funding, leadership, and authorship, and in sharing and harmonizing primary data
across studies. The formation of larger and more complex data sets has
stimulated innovations in informatics and the development and application of novel analytic methods. Further advances in high-​throughput
laboratory technology will continue to create new opportunities to
explore the complex biology of genomics, metabolomics, proteomics,
and so on, in large population studies. This will generate vast amounts
of data to be analyzed. It will also require insights from diverse disciplines to plan analyses and to interpret the results.
“Transdisciplinary research” signifies a level of collaboration in
which researchers from different scientific disciplines come together
to identify the most important research questions, design the optimal
approach, and collect, analyze, and publish the study results jointly
(Rosenfield, 1992). This level of team science transcends disciplinary
boundaries and benefits all parties. For example, the application of
haplotype analyses based on the structure of genetic linkage disequilibrium, initially proposed by geneticists, greatly accelerated exploratory analyses across the entire genome in large population studies.
Similarly, the involvement of epidemiologists in tumor genomics has
brought a population science perspective to this field, increasing attention to population sampling, sex, and racial/​ethnic differences, and
exposures such as smoking that can affect the mutational spectrum
of various cancers. There are many opportunities for collaborations
involving laboratory scientists, analytic chemists, epidemiologists,
and others to accelerate research on etiologic issues related to cancer. One example would be transdisciplinary research to understand
hormonal carcinogenesis at the molecular level in human populations
(Chapter 22).

Cancer Survivorship
The number of people surviving after a diagnosis of cancer has
increased rapidly in high-​income and many middle-​income countries

due to improvements in treatment and the effects of screening on both
early diagnosis and more complete ascertainment. In the United States,
the estimated number of people alive for at least 5 years after a diagnosis of cancer increased from 4 million in 1978 (~1.8% of the population) to 13.7 million in 2012 (~4% of the population), and is predicted
to approach 18 million by 2022 (de Moor et al., 2013; Harrop et al.,
2011). A substantial proportion of these are long-​term survivors. Forty
percent of individuals who survived for at least 5  years were alive
10 or more years after diagnosis, and 15% had survived 20 or more
years (Howlader et al., 2015). To address the rapidly increasing population of cancer survivors, the NCI Office of Cancer Survivorship
and a 2006 publication from the Institute of Medicine (Committee
on Cancer Survivorship, 2006) provided a framework for identifying
and addressing the unmet needs of this growing population. Disease
and treatment affect multiple health domains (e.g., medical, physical