Energy Balance and Cancer

Transcription

1

2 Energy Balance and Cancer Series Editor: Nathan A. Berger, Case Western Reserve University, Cleveland, OH, USA For further volumes:

3 Nathan A. Berger Editor Cancer and Energy Balance, Epidemiology and Overview 123

4 Editor Nathan A. Berger School of Medicine Center for Science, Health & Society Case Western Reserve University Euclid Avenue Cleveland OH USA ISBN e-isbn DOI / Springer New York Dordrecht Heidelberg London Library of Congress Control Number: Springer Science+Business Media, LLC 2010 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. While the advice and information in this book are believed to be true and accurate at the date of going to press, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Printed on acid-free paper Springer is part of Springer Science+Business Media (

5 Preface: Associations and Challenges In a series of landmark articles published in the New England Journal of Medicine [1, 2] and Nature Reviews [3] between 1999 and 2004, Dr. Eugenia Jeanne Calle alerted the entire scientific and medical community to the epidemiologic evidence providing definitive support for the association between body mass, all cause mortality, and cancer mortality. Based on results from a prospectively studied cohort of more than 900,000 US adults in the American Cancer Society, Cancer Prevention II Study, Jeanne identified the association of increased body mass index with death rate for all cancers combined, as well as for specific malignancies in both men and women. The ACS study showed that elevated body mass was associated with higher death rates from cancers of the esophagus, colon and rectum, liver, gallbladder, pancreas, kidney, non-hodgkin s lymphoma, and multiple myeloma. Trends were identified also for association of elevated BMI with deaths from prostate and stomach cancer in men and postmenopausal breast, uterus, cervix, and ovarian cancer in women. She put forth the alarming statistics that 14% of cancers in men and 20% of cancers in women were associated with obesity, all of which were of even greater concern due to the rising prevalence of overweight and obesity in the United States and on a worldwide basis. These studies clearly stressed the importance of obesity control to prevent the relative burdens of obesity-related morbidity and mortality. In subsequent years, Jeanne conducted important studies defining specific tumor types associated with overweight and obesity, investigated the contribution of specific nutrients to this problem, initiated studies to examine the interaction of adiposity with hormones and hormone-dependent malignancies, conducted studies on the impact of weight gain as opposed to established obesity on carcinogenesis and instituted studies on the importance of physical activity on adiposity and cancer. Jeanne earned her Ph.D. in Preventive Medicine from the Ohio State University. She subsequently worked at Oak Ridge National Laboratories, The Centers for Disease Control and The American Cancer Society where she ultimately became Vice President of Epidemiology. In her 2003 article she advocated that maintaining a BMI less than 25.0 might prevent 90,000 cancer deaths per year in the United States. Before her tragic death on February 17, 2009, Jeanne had become a leading spokesperson for obesity control and increased physical activity to prevent cancer incidence and mortality. v

6 vi Preface: Associations and Challenges On July 8, 2004, the National Cancer Institute responded to the alarming association of obesity and cancer by issuing RFA-CA to establish the Transdisciplinary Research on Energetics and Cancer (TREC) Centers in nutrition, energetics, energy balance, physical activity, and cancer. The primary mission of the TREC Centers was to foster collaboration among transdisciplinary teams of scientists with the goal of accelerating progress toward reducing cancer incidence, morbidity and mortality associated with obesity, low levels of physical activity and poor diet. This initiative was spearheaded at NCI by Linda Nebeling, Ph.D., M.P.H., RD, Chief, Health Promotion Research Branch, NCI, and her associates Robert Croyle, Ph.D., Director, Division of Cancer Control and Population Sciences, Rachel Ballard-Barbash, M.D, M.P.H. Associate Director, Applied Research Program and John Milner, Ph.D., Chief, Nutritional Science Research Group, Division of Cancer Prevention, NCI. After a competitive grant review process, TREC Centers were established at five institutions, Case Western Reserve University, University of Southern California, University of Minnesota, and Fred Hutchinson Cancer Research Center, along with a coordinating center at the Fred Hutchinson Cancer Research Center. Independent and collaborative research and training activities at the TREC Centers coupled with a series of interactive and national meetings have significantly accelerated development of progress focused on this critical area of cancer prevention and control. From the beginning, it was clear that the problems of energy balance and cancers would not be solved by individual scientists or even by interdisciplinary or multidisciplinary teams, but that a new transdisciplinary approach might lead to significant progress. This book series was stimulated by the need to encourage communication among investigators representing the multiple disciplines engaged in addressing this problem. Many of the authors in this introductory volume are TREC investigators or collaborators and, although they wrote their chapters from the perspective of their individual disciplines, they are very much aware of the need for transdisciplinary communication and research efforts and are already engaged in leading these initiatives. In her keynote address to a meeting entitled Energy Balance and Cancer: Mechanisms and Mediators sponsored by the American Association for Cancer Research, NCI, and TREC on October 24 26, 2008, in Lansdowne, Virginia, Jeanne Calle reviewed the relation between obesity and cancer, provided evidence for a growing list of obesity-related cancers, and stressed the need for accelerated research to identify the mechanisms involved. Just as she labored to prove the link between body mass and cancer, it remains for us to understand and break that link. Cleveland, OH, USA Nathan A. Berger References 1. Calle EE, Thun MJ, Petrelli HM, Rodriguez C, Heath CW Jr. (1999). Body mass and mortality in a prospective cohort of U.S. adults. N Engl J Med;341:

7 Preface: Associations and Challenges vii 2. Calle EE, Rodrigueez C, Walker-Thurmond KA, Thun MJ (2003). Oveweight, obesity and mortality from cancer in a prospectively studied cohort of U.S. Adults. N Engl J Med;348: Calle EE, Kaaks R (2004). Overweight, obesity and cancer: epidemiological evidence and proposed mechanisms. Nature Rev Cancer, 4:

8 Contents Preface: Associations and Challenges... v Contributors... xi 1 Obesity and Cancer Epidemiology... 1 Rachel Ballard-Barbash, David Berrigan, Nancy Potischman, and Emily Dowling 2 Obesity and Cancer: Epidemiology in Racial/Ethnic Minorities. 45 Colleen Doyle 3 Obesity and Cancer in Asia Wanghong Xu and Charles E. Matthews 4 Genetic Epidemiology of Obesity and Cancer Courtney Gray-McGuire, Indra Adrianto, Thuan Nguyen, and Chee Paul Lin 5 Obesity and Cancer: Overview of Mechanisms Nora L. Nock and Nathan A. Berger 6 Caloric Restriction and Cancer Fei Xue and Karin B. Michels 7 Physical Activity and Cancer Leslie Bernstein, Yani Lu, and Katherine D. Henderson 8 Energy Balance, Cancer Prognosis, and Survivorship Melinda L. Irwin 9 Behavior, Energy Balance, and Cancer: An Overview Donna Spruijt-Metz, Selena T. Nguyen-Rodriguez, and Jaimie N. Davis 10 Geographic and Contextual Effects on Energy Balance-Related Behaviors and Cancer David Berrigan, Robin McKinnon, Genevieve Dunton, Lan Huang, and Rachel Ballard-Barbash Subject Index ix

9 Contributors Indra Adrianto Arthritis and Immunology Research Program, Oklahoma Medical Research Foundation, Oklahoma City, OK, USA, Rachel Ballard-Barbash Applied Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute, Bethesda, MD, USA, Nathan A. Berger Center for Science, Health and Society, Case Western Reserve University, Cleveland, OH, USA, Leslie Bernstein Division of Cancer Etiology, Department of Population Sciences, Beckman Research Institute, City of Hope Comprehensive Cancer Center, Duarte, CA, USA, David Berrigan Division of Cancer Control and Population Sciences,, National Cancer Institute, Bethesda MD 20892, USA, Jaimie N. Davis Keck School of Medicine, Institute for Health Promotion and Disease Prevention Research, University of Southern California, Los Angeles, CA, USA, Emily Dowling Applied Research Program, Division of Cancer Control and Population Sciences National Cancer Institute, Bethesda, MD, USA, Colleen Doyle Nutrition and Physical Activity, American Cancer Society, Oklahoma City, OK , USA, Genevieve Dunton Department of Preventive Medicine, Institute for Health Promotion and Disease Prevention Research, University of Southern California, Los Angeles, CA, USA, Courtney Gray-McGuire Arthritis and Immunology Research Program, Oklahoma Medical Research Foundation, Oklahoma City, OK, USA, xi

10 xii Contributors Katherine D. Henderson Division of Cancer Etiology, Department of Population Sciences Beckman Research Institute and City of Hope National Medical Center, Duarte, CA, USA, Lan Huang Physiology & Biophysics and Cell Biology, University of California Irvine, Irvine, CA, USA, Melinda L. Irwin Epidemiology and Public Health, Yale School of Public Health, New Haven, CT, USA, Chee Paul Lin Arthritis and Immunology Research Program, Oklahoma Medical Research Foundation, Oklahoma City, OK, USA, Yani Lu Department of Population Sciences, Beckman Research Institute, City of Hope Comprehensive Cancer Center, Duarte, CA, USA; Department of Preventive Medicine, University of Southern California Keck School of Medicine, Los Angeles, CA, USA, Charles E Matthews Division of Cancer Epidemiology and Genetics, National Cancer Institute, Rockville, MD , USA, charles.matthews2@nih.gov Robin McKinnon Risk Factor Monitoring and Methods Branch, Applied Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute, Bethesda MD 20892, USA, robin.mckinnon@nih.gov Karin B. Michels Epidemiology, Harvard Medical School, Boston, MA, USA, kmichels@rics.bwh.harvard.edu Thuan Nguyen Arthritis and Immunology Research Program, Oklahoma Medical Research Foundation, Oklahoma City, OK, USA, nguyent@omrf.org Selena T. Nguyen-Rodriguez Keck School of Medicine, Institute for Health Promotion and Disease Prevention Research, University of Southern California, Los Angeles, CA, USA, selenang@usc.edu Nora L. Nock Epidemiology and Biostatistics, Case Western Reserve University, Cleveland, OH, USA, nln@case.edu Nancy Potischman Applied Research Program, Division of Cancer Control and Population Sciences National Cancer Institute, Bethesda, MD, USA, nancy.potischman@nih.gov Donna Spruijt-Metz Keck School of Medicine, Institute for Health Promotion and Disease Prevention Research, University of Southern California, Los Angeles, CA, USA, dmetz@usc.edu Wanghong Xu Department of Epidemiology, School of Public Health, Fudan University, Shanghai, China, wanghong.xu@fudan.edu.cn Fei Xue i3 Drug Safety, Ingenix, United Health Groups, Waltham, MA, USA, fei.xue@i3drugsafety.com

11 Chapter 1 Obesity and Cancer Epidemiology Rachel Ballard-Barbash, David Berrigan, Nancy Potischman, and Emily Dowling Abstract Evidence has expanded extensively in the past two decades on the association between body mass index (BMI) and other measures of body composition and weight gain with many cancers. Evidence is convincing for obesity as a risk factor for cancers of the esophagus, pancreas, colon and rectum, postmenopausal breast, endometrium, kidney, and thyroid and as probable for cancer of the gallbladder. Although not yet definitive, research is expanding rapidly for a number of other rare cancers and suggests associations for obesity and cancers of the ovary and liver and for several types of lymphoid and hematological malignancies. Associations between obesity and lung and head and neck cancers are confounded by tobacco use. An important shift in research has been the effort to examine the combined effect of overweight/obesity, physical inactivity, and poor diet. Generally, studies that have examined these combinations of factors have found much greater increases in risk among people who have these adverse health profiles. A number of mechanisms are being explored related to obesity and cancer, including changes in sex hormones, insulin-related growth factors, inflammation, immune function, and other growth factors. Data on racial/ethnic groups other than non-hispanic whites and Asians are limited for most cancers, but suggest there may be some differences in BMI and cancer associations in some subgroups. The continued global epidemics of obesity and diabetes mellitus are likely to contribute to global increases in a number of obesity-related cancers. R. Ballard-Barbash (B) Applied Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute, Bethesda, MD, USA barbashr@mail.nih.gov N.A. Berger (ed.), Cancer and Energy Balance, Epidemiology and Overview, Energy Balance and Cancer 2, DOI / _1, C Springer Science+Business Media, LLC

12 2 R. Ballard-Barbash et al. 1 Introduction Evidence has expanded extensively in the past two decades on the association between body mass index (BMI) and other measures of body composition and weight gain with many cancers. The most recent World Cancer Research Fund/American Institute for Cancer Research report identifies evidence that body fatness increases risk of cancers of the esophagus, pancreas, colon and rectum, postmenopausal breast, endometrium, and kidney and as probable for cancer of the gallbladder [172]. In addition, the associations of BMI, central adiposity, and weight gain are being examined relative to many other cancers resulting in emerging evidence of other associations. The growing evidence about the role of obesity on cancer risk and survival has coincided with the expanding global epidemic of obesity. The World Health Organization (WHO) estimated that in 2005, many countries had prevalence rates of obesity of more than 30%, with rates of obesity much higher in women compared to men. Furthermore, WHO estimates that by 2015, prevalence rates of obesity will have continued to increase, particularly in North and South America, reaching more than 45% in many countries on those continents, with rates in men equal to or surpassing rates in women in some countries (Fig. 1.1) [173]. The convergence of the evidence on the potential role of obesity on cancer risk and the rising global epidemic of obesity has led to focused attention on the influence of obesity on cancer risk and survival. This chapter will focus on the evidence of the association of obesity with cancer incidence from observational epidemiologic research, briefly summarizing evidence on the potential mechanisms that have been explored in human studies and noting promising future research directions. At present, there is no evidence from randomized controlled trials on the influence of weight loss on cancer risk. Data from the Swedish Obesity Study, a prospective, non-randomized trial of bariatric surgery involving over 4,000 obese subjects, found statistically significant lower overall mortality rates among subjects undergoing bariatric surgery than those who did not. Cancer deaths were lower among those undergoing bariatric surgery; however, the study was not powered to examine differences in cancer alone [152]. Several other studies have been published since 2007 using other cohorts of bariatric surgical patients and have reported statistically significant reductions in cancer mortality following bariatric surgery [2, 29]. Similar to results for the Swedish Obesity Study, deaths from causes not due to chronic diseases, such as accidents and suicide, appear to be higher among the bariatric surgery patients compared to obese patients not undergoing bariatric surgery. The reductions in cancer mortality are reported to be observed irrespective of cancer site, while reductions in cancer incidence appear to be limited to cancer that have been associated with obesity in prior research [3]. Investigators are beginning to explore possible mechanisms for these effects in small samples of patients following bariatric surgery. Evidence on obesity and cancer mortality and survival are addressed in another chapter of this book. Because of space limitations, the evidence on height and cancer risk is not included in this chapter but has been well summarized in recent reviews [54, 145, 164, 94, 143, 147, 178].

13 1 Obesity and Cancer Epidemiology 3 Prevalance of obesity, females, aged 30+, 2005 Prevalance of obesity, females, aged 30+, 2015 Legend Legend < >45.0 Estimated % of population with BMI>=30, Age-standardised to WHO World population. < >45.0 Estimated % of population with BMI>=30, Age-standardised to WHO World population km km Prevalance of obesity, males, aged 30+, 2005 Prevalance of obesity, males, aged 30+, 2015 Legend Legend < >45.0 Estimated % of population with BMI>=30, Age-standardised to WHO World population. < >45.0 Estimated % of population with BMI>=30, Age-standardised to WHO World population km km Fig. 1.1 Global Trends in Obesity by Sex, Adults vs (projected)

14 4 R. Ballard-Barbash et al. 1.1 Methodologic and Measurement Complexities in Obesity and Cancer Epidemiology This review of the existing observational epidemiologic evidence on the influences of energy balance and body size on cancer highlights three notable limitations: validity of exposure assessment, difficulty of comparisons across studies, and insufficient sample sizes to explore fully the many factors that relate to body size and fat mass and that could potentially modify observed associations. These numerous factors include age, race or ethnic background, related health behaviors of diet and physical activity, sleep, alcohol intake, tobacco use, medications, and comorbid conditions Validity of Exposure Assessment Among many health exposures related to energy balance that have been examined relative to cancer outcomes, anthropometric measures are among the most valid and reliable, particularly if done by direct measurement rather than by self-report. Weight and height are the most standardized measurements and least subject to variability. A few recent studies have directly measured various anthropometric indices from study participants, thereby reducing the possibility of random and systematic error. However, most studies of cancer etiology have relied on self-reported weight and height. Studies on self-reported height suggest a reasonable degree of accuracy, with a bias in over-reporting height that is somewhat greater in men compared to women and that increases with age [142] or is limited to people older than age 60 years [72], presumably due to age-related loss in height. Misreporting of selfreported weight is more common, with significant misreporting at the extremes of weight, suggesting that heavier people under-report and lighter people over-report their weight [142, 158, 104]. Therefore, studies of chronic diseases that rely on self-reported height and weight, used to derive BMI, are thought to underestimate the risk associated with these measures [53]. Nonetheless, compared to estimates of correlation between reported versus measured assessment of other exposures, such as dietary intake or physical activity, correlation coefficients between recalled and measured weight suggest that weight is recalled with a reasonable degree of accuracy, often with correlations over 0.95 for recent recall. Some decrement in recall has been documented for more distant recall [158]. In addition to measurement of total adiposity as estimated by BMI, research since the late 1980s also has explored the association of distribution of adiposity with disease endpoints, including cancer. Distribution of body fat is generally estimated by measurement of skinfolds or circumferences at the trunk, waist, hip, and extremities. Measurement of skinfolds and circumferences is more complex and hence less reliable than the measurement of weight and height [90, 168]. However, reliability is improved with standardization in measurement technique. It is not feasible to obtain these anthropometric measures from self-report. Measurement of waist circumference by trained personnel continues to be the most convenient, commonly used,

15 1 Obesity and Cancer Epidemiology 5 and simple measure of abdominal or central adiposity for epidemiologic research [167, 168]. Some studies continue to use ratios of waist to hip (WHR) or waist to thigh (WTR); however, use of waist alone is considered the most relevant measure of central adiposity that is feasible for large population-based studies of cancer etiology [172]. Percent body fat can be estimated from bioelectric impedance but rarely has been reported in studies of cancer etiology [74]. Use of bioelectric impedance for estimating body composition has expanded in large epidemiologic studies because it is portable, inexpensive, easy to use, and highly precise. Issues related to use of different methods for estimating body composition for chronic disease epidemiology are well summarized [11, 68, 85, 171]. Gold standard technologies for precisely measuring total and individual compartments of body fat, such as visceral or intramuscular fat, include dual X-ray absorptiometry (DXA) or magnetic resonance (MRI). These technologies are in common use in many research designs, have improved over time, and some machines used with these methods occupy less space than earlier versions. However, these methods remain expensive, not easily portable, and therefore, not practical for use in large population studies of cancer etiology. Finally, many cancer cohorts have only one self-report measure of weight and height at the beginning of the study, generally in mid-adult life. A number of studies that have obtained reports of weight and height at several ages during adolescent and adult life have demonstrated that weight change during adult life has larger and more consistent associations with subsequent cancer than does the single BMI estimate from weight and height obtained at the study baseline. Therefore, the absence of repeated measures of weight during adult life may be a limitation for a number of studies Complexity in Cross-Study Comparisons Comparison across studies is difficult because most studies have examined risk by quantile distributions (most commonly tertile or quartile) for BMI or other anthropometric measures used. As the distribution of BMI varies across populations, these quantile groups are not comparable across studies. To facilitate comparison of the risk of BMI across many chronic diseases, investigators have begun to examine risk by standard WHO BMI categories: underweight as BMI of less than 18.50, normal weight as BMI of , overweight as BMI of or higher. This latter overweight category is further subdivided into four categories: preobese , obese class I , obese class II , and obese class III [167]. Recent meta-analyses summarized in this review have often estimated changes in risk by 5-unit (i.e., kg/m 2 ) increments in BMI. This approach allows comparison across cancers for which increases in risk may vary only modestly across a wide range of BMI. However, a 5-unit increment in BMI translates to a very large difference in weight. For example, an increase in BMI from 25 to 30 in a person 5-ft tall is equivalent to 25.5 lbs and in a person 6-ft tall is equivalent to 36.8 lbs. Larger studies and meta-analyses that have the potential to explore risk