Energy Balance and Cancer
Series Editor: Nathan A. Berger, Case Western Reserve University, Cleveland, OH, USA
For further volumes: http://www.springer.com/series/8282
Nathan A. Berger Editor
Cancer and Energy Balance, Epidemiology and Overview
13
Editor Nathan A. Berger School of Medicine Center for Science, Health & Society Case Western Reserve University 10900 Euclid Avenue Cleveland OH 44106 USA
[email protected]
ISBN 978-1-4419-5514-2 e-ISBN 978-1-4419-5515-9 DOI 10.1007/978-1-4419-5515-9 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2010923799 © 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 (www.springer.com)
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
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-05-010 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;341:1097–105
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: 1625–38 3. Calle EE, Kaaks R (2004). Overweight, obesity and cancer: epidemiological evidence and proposed mechanisms. Nature Rev Cancer, 4:579–591
Contents
Preface: Associations and Challenges . . . . . . . . . . . . . . . . . . .
v
Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
xi
1 Obesity and Cancer Epidemiology . . . . . . . . . . . . . . . . . . Rachel Ballard-Barbash, David Berrigan, Nancy Potischman, and Emily Dowling
1
2 Obesity and Cancer: Epidemiology in Racial/Ethnic Minorities . Colleen Doyle
45
3 Obesity and Cancer in Asia . . . . . . . . . . . . . . . . . . . . . . Wanghong Xu and Charles E. Matthews
65
4 Genetic Epidemiology of Obesity and Cancer . . . . . . . . . . . . Courtney Gray-McGuire, Indra Adrianto, Thuan Nguyen, and Chee Paul Lin
87
5 Obesity and Cancer: Overview of Mechanisms . . . . . . . . . . . Nora L. Nock and Nathan A. Berger
129
6 Caloric Restriction and Cancer . . . . . . . . . . . . . . . . . . . Fei Xue and Karin B. Michels
181
7 Physical Activity and Cancer . . . . . . . . . . . . . . . . . . . . . Leslie Bernstein, Yani Lu, and Katherine D. Henderson
201
8 Energy Balance, Cancer Prognosis, and Survivorship . . . . . . . Melinda L. Irwin
219
9 Behavior, Energy Balance, and Cancer: An Overview . . . . . . . Donna Spruijt-Metz, Selena T. Nguyen-Rodriguez, and Jaimie N. Davis
233
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 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
267
299 ix
Contributors
Indra Adrianto Arthritis and Immunology Research Program, Oklahoma Medical Research Foundation, Oklahoma City, OK, USA,
[email protected] Rachel Ballard-Barbash Applied Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute, Bethesda, MD, USA,
[email protected] Nathan A. Berger Center for Science, Health and Society, Case Western Reserve University, Cleveland, OH, USA,
[email protected] Leslie Bernstein Division of Cancer Etiology, Department of Population Sciences, Beckman Research Institute, City of Hope Comprehensive Cancer Center, Duarte, CA, USA,
[email protected] David Berrigan Division of Cancer Control and Population Sciences, , National Cancer Institute, Bethesda MD 20892, USA,
[email protected] Jaimie N. Davis Keck School of Medicine, Institute for Health Promotion and Disease Prevention Research, University of Southern California, Los Angeles, CA, USA,
[email protected] Emily Dowling Applied Research Program, Division of Cancer Control and Population Sciences National Cancer Institute, Bethesda, MD, USA,
[email protected] Colleen Doyle Nutrition and Physical Activity, American Cancer Society, Oklahoma City, OK 73123–1538, USA,
[email protected] Genevieve Dunton Department of Preventive Medicine, Institute for Health Promotion and Disease Prevention Research, University of Southern California, Los Angeles, CA, USA,
[email protected] Courtney Gray-McGuire Arthritis and Immunology Research Program, Oklahoma Medical Research Foundation, Oklahoma City, OK, USA,
[email protected]
xi
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,
[email protected] Lan Huang Physiology & Biophysics and Cell Biology, University of California Irvine, Irvine, CA, USA,
[email protected] Melinda L. Irwin Epidemiology and Public Health, Yale School of Public Health, New Haven, CT, USA,
[email protected] Chee Paul Lin Arthritis and Immunology Research Program, Oklahoma Medical Research Foundation, Oklahoma City, OK, USA,
[email protected] 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,
[email protected] Charles E Matthews Division of Cancer Epidemiology and Genetics, National Cancer Institute, Rockville, MD 20892–7344, USA,
[email protected] 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,
[email protected] Karin B. Michels Epidemiology, Harvard Medical School, Boston, MA, USA,
[email protected] Thuan Nguyen Arthritis and Immunology Research Program, Oklahoma Medical Research Foundation, Oklahoma City, OK, USA,
[email protected] Selena T. Nguyen-Rodriguez Keck School of Medicine, Institute for Health Promotion and Disease Prevention Research, University of Southern California, Los Angeles, CA, USA,
[email protected] Nora L. Nock Epidemiology and Biostatistics, Case Western Reserve University, Cleveland, OH, USA,
[email protected] Nancy Potischman Applied Research Program, Division of Cancer Control and Population Sciences National Cancer Institute, Bethesda, MD, USA,
[email protected] Donna Spruijt-Metz Keck School of Medicine, Institute for Health Promotion and Disease Prevention Research, University of Southern California, Los Angeles, CA, USA,
[email protected] Wanghong Xu Department of Epidemiology, School of Public Health, Fudan University, Shanghai, China,
[email protected] Fei Xue i3 Drug Safety, Ingenix, United Health Groups, Waltham, MA, USA,
[email protected]
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 e-mail:
[email protected]
N.A. Berger (ed.), Cancer and Energy Balance, Epidemiology and Overview, Energy Balance and Cancer 2, DOI 10.1007/978-1-4419-5515-9_1, C Springer Science+Business Media, LLC 2010
1
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].
Legend
>45.0
4000.02 km
Estimated % of population with BMI> = 30, Age-standardised to WHO World population.
8.0–14.9 15.0–29.9 30.0–44.9
<8.0
Prevalance of obesity, males, aged 30+, 2005
4000.02 km
4000.02 km
15.0–29.9
0
Legend 8.0-14.9 30.0–44.9
>45.0
4000.02 km
Estimated % of population with BMI> = 30, Age-standardised to WHO World population.
15.0-29.9
<8.0
Prevalance of obesity, males, aged 30+, 2015
0
Legend 8.0–14.9 >45.0 30.0–44.9 Estimated % of population with BMI> = 30, Age-standardised to WHO World population.
<8.0
Prevalance of obesity, females, aged 30+, 2015
Obesity and Cancer Epidemiology
Fig. 1.1 Global Trends in Obesity by Sex, Adults 30 + 2005 vs. 2015 (projected)
0
0
Legend
8.0–14.9 >45.0 15.0-29.9 30.0–44.9 Estimated % of population with BMI> = 30, Age-standardised to WHO World population.
<8.0
Prevalance of obesity, females, aged 30+, 2005
1 3
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.
1.1.1 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,
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. 1.1.2 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 18.50–24.99, overweight as BMI of 25.00 or higher. This latter overweight category is further subdivided into four categories: preobese 25.00–29.99, obese class I 30.00–34.99, obese class II 35.00–39.99, and obese class III ≥ 40.00 [167]. Recent meta-analyses summarized in this review have often estimated changes in risk by 5-unit (i.e., kg/m2 ) 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
6
R. Ballard-Barbash et al.
estimates by much smaller increments of BMI suggest that risk may vary across the range of BMI, depending on the chronic diseases and the populations examined.
1.1.3 Requirement for Large Sample Sizes to Assess Subgroup Variation in Association The association of BMI and other measures of body size varies by cancer site and often may vary by specific subtypes of cancer within one cancer site. In addition, research suggests that BMI and cancer associations may vary by a number of patient and tumor characteristics. As more factors are identified as potentially influencing body size and cancer associations, delineation of the underlying mechanisms requires further subgroup analysis, thereby increasing sample size requirements and the need for multi-institutional collaborative studies and meta-analyses.
2 Methods To review the epidemiologic literature on the association of obesity and cancer risk, we conducted a search of MEDLINE and PUBMED for all publications on weight, BMI, anthropometric factors, specific cancers, and cancer subtypes in human populations. We supplemented this search with a manual search of major relevant journals and also identified and reviewed major review papers or books. The literature search included all publications from 2002 to August 2008 and selected recent publications up to January 2009. This search updates previous literature reviews conducted on obesity and cancer, which have included all related publications until February 2003. Studies included in this review focus on some aspect of anthropometric risk factors in relation to cancer risk for the following cancers: breast, ovarian, endometrial, prostate, thyroid, renal cell, colorectal, esophageal, pancreatic, hepatocellular, gallbladder, lung, head and neck, and lymphoid and hematological. Figures were selected from metaanalyses to highlight emerging evidence on the association with obesity in cancer sites such as hepatocellular cancer, pancreatic cancer, non-Hodgkin’s lymphoma, and leukemia. Major associations are summarized in Table 1.1.
3 Breast Cancer 3.1 Incidence and Risk Factors Breast cancer is the most common cancer (182,460 cases in 2008) and the second most common cause of cancer deaths among women in the United States (40,480 cases in 2008) [7]. Factors associated with increased risk of breast cancer include older age, family history, race, exogenous estrogens such as hormone replacement
1
Obesity and Cancer Epidemiology
7
Table 1.1 BMI and central adiposity and cancer risk Decreases risk Exposure
Increases risk Cancer site
Convincing
Probable
BMI, adiposity
Breast (premenopausal)
Exposure
Cancer site
BMI, adiposity
Endometrium Breast (postmenopausal) Colorectum Kidney Esophagus Pancreas Colorectal Gallbladder
Central adiposity BMI, adiposity Central adiposity
Limited or suggestive
Adult weight gain BMI, adiposity
Underweight
Pancreas Breast (postmenopausal) Endometrium Thyroid (women) Breast (postmenopausal) Liver Non-Hodgkin’s lymphoma Leukemia Multiple myeloma Prostate (aggressive disease) Ovarian Lung Head and neck
Adapted and updated from Ref. [172]
therapy (HRT), reproductive factors (including age at first birth, parity, and lactation), inherited BRCA1 or BRCA2 mutations and other genetic factors, and lifestyle factors, including obesity, physical inactivity, alcohol intake, and some aspects of diet.
3.2 Overall Statement The association between overweight/obesity and breast cancer incidence has been an area of intense interest, beginning with studies in the 1970s that demonstrated that heavier women were at increased risk of postmenopausal breast cancer
8
R. Ballard-Barbash et al.
[64, 172]. Extensive evidence from cohort, case–control, and meta-analyses indicates a modest increase in risk of postmenopausal breast cancer in overweight or obese women, a two-fold to three-fold increase in risk from adult weight gain, and inconsistent changes in risk from central adiposity [64, 172]. Meta-analyses have found a 30% increase in postmenopausal breast cancer risk with BMIs of 28 or greater [163] and a 10% increase per 5 kg/m2 increase in BMI [130]. However, BMI-associated risk estimates are much higher (two-fold to three-fold) for overweight and obese women who have never used HRT. In contrast, among women who are current users of HRT, BMI is generally not associated with an increase in breast cancer risk. BMI-associated risk estimates also are much higher for developing hormone receptor-positive as compared to hormone receptor-negative tumors. For premenopausal women, meta-analyses of cohort studies have demonstrated an inverse association between BMI and premenopausal breast cancer, with a 50% reduction in risk among obese women [163] and an 8% reduction in risk per 5 kg/m2 increase in BMI [130].
3.3 Selected Issues 3.3.1 Life Cycle and Other Characteristics of Women May Be Relevant In addition to the effect on risk of factors such as menopausal status and age at diagnosis, hormone receptor status of the breast cancer, and exposure to exogenous estrogens, observed associations also vary by time period during the life cycle when weight or BMI are measured. A pooled analysis of 32 studies with 22,058 cases indicates a modest increase in risk for each standard deviation unit (0.5 kg) increase in birth weight (Relative Risk [RR]=1.06, 95% CI 1.02–1.09) [150]. Another metaanalysis reported a RR=1.24 (95% confidence interval [CI] 1.04–1.48) for a birth weight of 4,000 g or more compared with normal birth weights [114]. The association of weight-related measures during young adult life and breast cancer risk has been extensively explored, is based on self-report of weight and height from decades earlier, and generally finds a 10–30% decrease in risk of pre- and postmenopausal breast cancer among women reporting heavier weight or BMI during the teens to early twenties. 3.3.2 Evidence on Confounding Factors Is Limited The majority of studies on adult BMI and breast cancer risk have been adjusted for major breast cancer risk factors, including reproductive factors. Few studies have examined in detail the effect of confounding or interactions with diet and physical activity. Breast density has emerged as an important breast cancer risk factor and is known to be decreased among obese women, but few studies have examined either combined effects or interactions between breast density and BMI and breast cancer risk. A seminal study in Canada found that both BMI and mammographic breast density were independent risk factors for breast cancer and that the
1
Obesity and Cancer Epidemiology
9
strong inverse correlation between them may result in an underestimation of the adverse effects of either of these exposures on breast cancer risk if confounding is not addressed [17]. Unfortunately, most studies of either BMI or breast density have not collected measures of the other biological variables and so few studies can control for the potential interaction between BMI and breast density on breast cancer risk.
3.3.3 Physical Activity Is Protective Basic, clinical, and epidemiologic research has identified an inverse association between physical activity and breast cancer incidence with an overall reduction of about 20% observed among the most active compared to the least active women [84]. Most of the epidemiologic studies of physical activity and breast cancer incidence have adjusted for the confounding effect of obesity and have examined physical activity associations stratified by quantiles of BMI. In contrast, recent studies of BMI generally adjust for the potential confounding effect of physical activity and have generally reported that physical activity does not alter observed associations between BMI and breast cancer. A limited number of studies have reported on whether interactions between BMI and physical activity and breast cancer have been observed and most have not found evidence of statistically significant interaction [25, 35, 151, 96]. However, several of these studies have observed larger increases in risk among the most sedentary and overweight/obese women in both premenopausal [96] and postmenopausal women [149].
3.3.4 Family History Data Are Lacking Data are very limited on variation in BMI-related risk for postmenopausal breast cancer by family history, with some studies suggesting greater increases in risk with obesity among women with a positive family history [64]. Other studies find no differences in obesity-related risk by family history.
3.3.5 Limited Data Suggest Some Differences by Race and Ethnicity The majority of research on the association of obesity with breast cancer incidence has been based on white populations from Europe, Canada, and the United States. A limited number of studies in other racial or ethnic groups suggest some differences in observed associations among Hispanic and African-American women, with less increase in risk observed for overweight and obesity. In contrast, observed associations in large studies among Asian populations in the United States, Japan, and China have been similar to those among white women. Larger studies are needed in major racial/ethnic populations to better define how obesity is associated with breast cancer [172].
10
R. Ballard-Barbash et al.
3.3.6 Genetic Factors May Play a Role Recent studies have reported that the association of BMI with breast cancer may vary by the genetic expression patterns of luminal A and B (ER+), HER2 overexpression, basal-like, and triple-negative subtypes of tumors. Initial findings suggest that overweight and obesity may be associated with increased risk of luminal and triple-negative but not of non-luminal tumor types [120].
3.4 Conclusion Extensive evidence indicates that overweight and obesity and adult weight gain may increase risk of postmenopausal breast cancer while it decreases risk of premenopausal breast cancer. It is likely that these differential effects occur because of the differing effects of overweight and obesity on endogenous estrogen production in postmenopausal as compared to premenopausal women. The increased risk observed in postmenopausal women in association with overweight and obesity is much higher among women who have not used HRT, is higher for hormone receptorpositive tumors, and may be underestimated among women with increased breast density. Limited data suggest that obesity-related risk may vary among racial ethnic groups, by family history, and by some genetic tumor characteristics. A few recent studies have not found that the observed association between BMI and breast cancer is altered by physical activity.
4 Ovarian Cancer 4.1 Incidence and Risk Factors Ovarian cancer is the eighth most common cancer in the United States (21,650 cases in 2008) and a leading cause of cancer deaths (15,520 in 2008) among women because of the difficulty in early diagnosis [7]. Factors associated with increased risk for ovarian cancer include increasing age, use of estrogen-alone HRT, family history of breast or ovarian cancer, inherited BRCA1 or BRCA2 mutations, and inherited nonpolyposis colon cancer [7]. Reduced risks are related to higher parity, long-term oral contraceptive usage, history of tubal ligation, and hysterectomy [7]. Given that only a small percentage of cases are related to family history, most ovarian cancers are sporadic, suggesting that some of the disease is related to environmental or lifestyle factors.
4.2 Overall Statement Many studies have reviewed evidence on the association of obesity and ovarian cancer and results suggest a weak influence on disease risk. An early review [125] and a
1
Obesity and Cancer Epidemiology
11
more recent meta-analysis of 16 studies [111] showed similar increased risk of ovarian cancer with highest compared to lowest BMI (RR=∼1.3) and slightly lower risks observed for overweight (RR=1.16, 95% CI 1.10–1.32). Although another meta-analysis of 13 prospective studies reported a weak influence of a 5 kg/m2 increase in BMI on ovarian cancer incidence [130], results from the largest prospective studies are inconsistent. Two large studies each with a million women found no association of adult overweight or obesity with ovarian cancer [129, 37] but one study reported an increased risk associated with obesity during adolescence and young adulthood (RR=1.56, 95% CI 1.04–2.32) [37]. Thus, weak effects of adult body mass are observed overall, with some data suggesting that the relevant time period may be early in life.
4.3 Selected Issues 4.3.1 Influence of BMI Is Potentially Restricted to Early Life and Premenopausal Disease Results evaluating risk of disease associated with early body size are inconsistent, partly due to definitions of early BMI and type of disease. A meta-analysis of five case–control and four cohort studies [111] showed overweight and obesity in early adulthood (age 17–20 years) was associated with increased risk of ovarian cancer (RR=1.22, 95% CI 1.02–1.45), while a pooled analysis of six cohort studies with information on BMI in early adulthood (age 18–20 years) found no relation among this younger age group [147]. Adult obesity, however, was related to premenopausal but not postmenopausal ovarian cancer [147], similar to another evaluation of early adult body mass that suggested a possible restriction of risk to premenopausal disease [40]. Studies are inconsistent on the association of obesity with ovarian cancer subtypes [40, 100, 111, 112]. The lack of association or weak association of adult BMI with all ovarian cancers and restricted associations of early BMI and premenopausal disease [37, 100, 111] suggest some relation to endocrinologic factors during these time periods.
4.3.2 Evidence Exists for Higher Waist Circumference but Not Weight Gain Epidemiologic studies have consistently shown that high waist-to-hip ratio has been related to increased risk of ovarian cancer [32, 9, 57]. However, few studies have evaluated the relation of weight change to this cancer. A cohort study in Austria showed a strong relation of high weight gain and incident ovarian cancer [128], as did a case–control study [52]. Other studies, however, showed no association of adult weight gain with risk of ovarian cancer [40, 112]. The influence of starting BMI and weight gain needs to be addressed as evidenced by a study reporting higher risk for low weight gain from age 30 to diagnosis, yet cases were heavier at age 30 than controls [32].
12
R. Ballard-Barbash et al.
4.4 Conclusion A large number of studies indicate a weak association of adult BMI with ovarian cancers but many studies suggest that early life weight may influence risk to a greater extent. Most studies did not evaluate risk by menopausal status, yet there is now evidence that obesity may be more related to premenopausal disease. Further work remains for evaluation of lifelong weight history, relation of obesity to biochemical risk factors, and to ovarian cancer subtypes or tumor characteristics.
5 Endometrial Cancer 5.1 Incidence and Risk Factors Endometrial cancer is the fourth most common cancer among women (40,100 cases in 2008), the most common cancer of the female reproductive system, and the eighth most common cause of cancer deaths (7,470 in 2008) in women in the United States [7]. Known risk factors include obesity, use of estrogen-alone HRT, ovarian dysfunction, diabetes, infertility, nulliparity, and tamoxifen use [156].
5.2 Overall Statement Extensive evidence from case–control and cohort studies indicates a consistent, marked two-fold to four-fold increase in risk of endometrial cancer from overweight and obesity. A recent meta-analysis found a 60% increase in risk of endometrial cancer per 5 kg/m2 increase in BMI [130], with a doubling to tripling of risk observed among obese women. Obesity may account for about 40% of the worldwide variation in cumulative rates of endometrial cancer [64]. Therefore, a substantial portion of the international variation in the incidence of endometrial cancer may be explained by differences in the prevalence of obesity.
5.3 Selected Issues 5.3.1 Evidence Is Consistent The strong positive associations between obesity and endometrial cancer risk in case–control and cohort studies have remained high even after adjustment for several factors that may potentially influence both obesity and risk, such as parity, HRT, and smoking [64, 172]. These increased risks have been observed among a number of different racial and ethnic groups in the United States and internationally. The influence of weight gain during adulthood and central adiposity have been examined less extensively but also show strong positive associations with endometrial cancer in most studies [64, 172].
1
Obesity and Cancer Epidemiology
13
5.3.2 Physical Activity May Be Protective Recent studies have examined whether the association of BMI with endometrial cancer is altered by physical activity. Most studies have found that the elevated risk associated with obesity remains after adjustment for physical activity. Some studies on the association of physical activity with endometrial cancer risk have found that the protective effect of physical activity was most evident or limited to women with higher BMIs [84, 117].
5.3.3 Effect of Obesity Varies by HRT Use A limited number of studies have examined whether the BMI-related risk estimates vary by history of HRT use. Comparable to findings for breast cancer, risk estimates for BMI and endometrial cancer are higher (in the order of two-fold to four-fold) among women who have never used HRT compared to former or current HRT users [45, 13].
5.3.4 The Role of Diabetes Is Being Explored Evolving evidence suggests that diabetes, hyperinsulinemia, and long-term insulin therapy in patients with Type 1 diabetes may increase risk of endometrial cancer [43, 44]. A recent study found that the combination of obesity, physical inactivity, and diabetes was particularly detrimental. Women with those characteristics had a nearly 10-fold increase in risk compared to normal weight, physically active, nondiabetic women [43, 44]. In a related area, the association of dietary factors that may stimulate insulin production, including carbohydrate intake, glycemic index, and glycemic load, are being explored. Evidence in this area is somewhat inconsistent, as some evidence suggests that a high carbohydrate intake or high glycemic load may increase risk, particularly among overweight women with low physical activity [75].
5.4 Conclusion Evidence indicates that overweight and obesity increase endometrial cancer risk by two-fold to four-fold, among the strongest relative risks observed for obesity-related cancers. As with breast cancer, obesity-related risk is higher among women who have never used HRT compared to current or former users. The combination of obesity, physical inactivity, and diabetes appears to markedly increase risk. Increases in worldwide rates of obesity and diabetes may contribute to increasing rates of endometrial cancer worldwide in the future.
14
R. Ballard-Barbash et al.
6 Prostate Cancer 6.1 Incidence and Risk Factors Prostate cancer is among the most common cancers in men (186,320 cases in 2008) in the United States [7], although the incidence of this cancer overall and in different racial and ethnic groups is sensitive to the prevalence of prostate-specific antigen (PSA) testing [123]. To date, it has proved difficult to conclusively identify risk factors, except age, for any form of prostate cancer. Intensive efforts have focused on searching for links between prostate cancer, endogenous hormones [69], growth factors [135], and measures of energy balance, notably BMI [10, 48]. The WCRF notes the potential for physical activity to influence prostate cancer risk but does not report evidence for an association between obesity and prostate cancer [172]. Incidence of prostate cancer is higher in African-American men, although it is not yet clear whether this is because of biologically based differences in risk, behavioral differences, or differences in health-care access [18]. All three factors could play some role in this well-documented pattern.
6.2 Overall Statement Epidemiologic studies of associations between BMI and prostate cancer incidence dating back more than 30 years provide little evidence of a strong association between BMI and prostate cancer [8, 124, 21, 48]. The most recent meta-analysis of the association between body size and prostate cancer risk [94] reported a RR of 1.05 (95% CI 1.01–1.08) per 5 kg/m2 increase overall and 1.12 (95% CI 1.01–1.23) for late-stage disease per 5 kg/m2 increase. Height was also associated with risk, but waist circumference and waist-to-hip ratios were not. This analysis was based on results from 31 cohort and 25 case–control studies representing approximately 56,000 and 16,000 cases, respectively. A number of studies also have examined the association of anthropometric characteristics in younger people and prostate cancer risk [134]. There appears to be little or no association between youth and young adult BMI and future prostate cancer risk, but the studies are few in number and largely rely on self-reported height and weight. Moreover, most of the subjects were normal weight when young. Therefore, it is not yet known whether the current increase in overweight and obesity in children or young adults will influence prostate cancer incidence and mortality in the future.
6.3 Selected Issues 6.3.1 Contribution of Disease Stage at Diagnosis and Detection Is Unclear Past reviews and several recent studies point toward stronger associations between BMI and prostate cancer identified at an aggressive stage at diagnosis in contrast
1
Obesity and Cancer Epidemiology
15
to earlier research that did not examine associations by stage at diagnosis [50, 94, 42, 136, 174]. Such differences could be associated with changing ratios of positive and negative risk factors over the long developmental period associated with prostate hyperplasia. Obese men could be more likely to be diagnosed with latestage or aggressive prostate cancer because obesity reduces the probability of early detection by digital rectal exams, can lead to testosterone-mediated decreases in PSA levels, or result in hemodilution on PSA levels. Apparent reductions in PSA levels because of these latter two factors could delay follow-up and result in detection at later stages of tumor development [42]. Overall, associations between obesity and prostate cancer mortality also could arise because of later diagnoses, effects of comorbidity, or genuine associations between rate of progression and obesity. Studies to date do not seem to support the notion that poorer treatment outcomes in obese patients explain the association between obesity and prostate cancer mortality, but better data resources from health services research are needed to conclusively address this issue [154]. More effective integration of biological research and research on performance of diagnostic tests also are required to tease apart the relative contributions of diagnostic and biological processes to associations between BMI and prostate cancer incidence and mortality. A notable weakness of many studies is the failure to differentiate among stages of disease. Furthermore, among men diagnosed with early-stage prostate cancer, many more men die of other causes despite the presence of small prostate cancers than die of the cancer itself [39, 41]. Thus, it is critically important to understand risk factors for disease progression at different stages.
6.3.2 Serum Hormones and Growth Factors May Play a Role Because of notable differences in prostate cancer incidence in African-American and non-Hispanic whites in the United States, international variation in incidence [98], and extensive results from animal and cell culture studies [121], it is widely believed that prostate cancer incidence, progression, and mortality are influenced by endogenous hormones and growth factors. How these associations are linked to obesity remains unclear. The best studied growth factor is insulin-like growth factor-1 (IGF-1) [135].
6.3.3 Other Factors Also May Be Important Family history and genetic variation may account for significant variation in prostate cancer risk [177]. It is not yet known how obesity might interact with genetic predisposition to influence risk. More characterization of genotypes is obviously needed, but these studies will be most profitable if they are carried out in diverse populations and include careful phenotyping.
16
R. Ballard-Barbash et al.
6.4 Conclusion Obesity may be a risk factor for aggressive prostate cancer, but it has proved difficult to separate the influence of obesity on tumorigenesis versus its influence on detection. More work remains to be done concerning interactions between race/ethnicity and energy balance phenotypes on prostate cancer risk and prognosis.
7 Thyroid Cancer 7.1 Incidence and Risk Factors Incidence rates of thyroid cancer doubled between 1973 and 2002 [33] and continue to rise, with an estimated 37,340 cases in 2008 [7]. Thyroid cancer is much more common in women than in men, with an estimated 28,410 cases in women and 8,930 cases in men in 2008 [7]. Davies and Welch [33] argue convincingly that this increase is largely due to advances in imaging and diagnosis. Ionizing radiation, female sex, iodine deficiency, and obesity increase risk of this cancer. A few recent studies suggest that smoking and physical activity are protective. Despite recent advances in understanding the molecular processes altered in thyroid carcinogenesis [176], links between risk factors and these processes as well as a more complete catalog of risk factors have yet to be established.
7.2 Overall Statement Consistent evidence indicates small positive associations between BMI and thyroid cancer risk, particularly in women [172]. Meta-analysis of results from five cohort studies in Iceland, Korea, Austria, Norway, and Sweden including 1,212 cases overall resulted in risk ratios of 1.33 (95% CI = 1.04–1.70) for women and 1.14 (95% CI = 1.06–1.23) for men for a 5 kg/m2 increase in BMI. The studies included in this meta-analysis adjusted for diverse factors, including smoking, physical activity, alcohol consumption, and others, but different cohorts included adjustment for different sets of potential confounders. A cohort study in Korea [109] obtained similar risks of obesity in the entire population as compared to the subset of the population that never smoked. Pooled analysis of results from 12 case–control studies involving 2,573 cases (2,056 females and 417 males) reported elevated risk for height in both males and females, but BMI and thyroid cancer were associated only in females. Further analysis of the same pooled data of case–control studies of thyroid cancer reports a protective effect of ever or current smoking, even after adjustment for body mass [95]. One study also reports a protective effect of physical activity in women [140].
1
Obesity and Cancer Epidemiology
17
7.3 Selected Issues 7.3.1 Potential Role of Growth Hormone and IGF Axis Is Being Examined The observation that case–control studies indicate that height is a more consistent risk factor for thyroid cancer than weight supports an argument that the growth hormone and IGF axis may play a role in thyroid carcinogenesis [55]. No epidemiologic studies of growth hormone or IGF and thyroid cancer were identified, although a number of papers have discussed the role of growth factors and IGF in thyroid tumor growth and development, with local production of IGF increasing in thyroid cancer [30]. More recently, considerable attention has been paid to the PI3K/AKT signaling pathway in the differentiation of thyroid cancer precursor cells and this pathway is influenced by energy balance [103].
7.4 Conclusion Review of the literature concerning thyroid cancer and obesity highlights problems associated with identifying risk factors in relatively rare cancers. For example, we have little epidemiologic evidence concerning risk factors in different racial and ethnic groups, yet incidence rates from US population studies differ significantly (e.g., 4.9/100,000 in white men and 2.7/100,000 in African-American men) (http://seer.cancer.gov/statfacts/html/thyro.html). Standardization and expansion of anthropometric measures in diverse international cohorts could allow further productive analysis of pooled data.
8 Renal Cell Carcinoma 8.1 Incidence and Risk Factors Cancer of the kidney is among the 10 most common cancers in the United States (54,390 cases in 2008: 33,130 in men and 21,260 in women) [7]. Evidence linking kidney cancer and body fatness has become convincing since such an association was deemed probable in 1997 [172]. In addition to obesity, male sex, smoking, high blood pressure, and diverse occupational exposures are known risk factors [89, 103]. Kidney cancer comes in a number of types, but about 85–90% of cases in adults are renal cell carcinoma (RCC) [116]. Interestingly, in the United States, the incidence of RCC and renal pelvis cancer increased steadily from 1975 to 2005, rising from about 10 cases per 100,000 to almost 18 cases [132]. Improved detection methods and obesity and smoking trends in women are considered to account for some of the increase, but careful population modeling has yet to address the extent to which these risk factors can explain all of it [27, 28]. Trends in Europe have been mixed, with increased incidence in the United Kingdom, but decreased or
18
R. Ballard-Barbash et al.
stable incidence in other Northern European countries (notably in Sweden, where incidence decreased 18% between 1992 and 2002) [86, 87].
8.2 Overall Statement Obesity has been consistently identified as a risk factor for RCC in 17 cohort and more than 20 case–control studies, most of which are population based [102, 89, 172]. Risks attributable to elevated BMI versus smoking were similar in at least one study [12]. The increase in risk associated with BMI may be greater in women than men [10]. However, because the prevalence of multiple risk factors differs between men and women and sample sizes in RCC studies limit the potential to fully address confounding factors, it is difficult to fully interpret the differences in risk estimates for obesity between men and women.
8.3 Selected Issues 8.3.1 Study Results Are Heterogeneous Work to date has focused on quantifying age and sex-specific differences in the degree of risk associated with BMI or increases in BMI [10, 19]. Few consistent patterns have emerged beyond increased risk with elevated BMI. Results from a large and recent study (1,366 cases) suggest that BMI at older ages and weight gain in young and middle-aged adults are most strongly associated with risk [1]. Height and weight in this study were based on self-report and required recall over many years. The EPIC cohort study [122] (287 incident cases) resulted in estimated RR of 2.25 (95% CI 1.14–4.44) comparing highest and lowest BMI quintiles for women, but BMI was not associated with RCC in men. Different results are reported by Brennan et al., but their study involved the use of hospital-based controls [19]. 8.3.2 Little Data Available on Other Measures of Adiposity Few studies examine anthropomorphic measures beyond height, weight, and BMI. Even the largest cohort study to date [16], with 6,500 cases, could only evaluate height and weight. Adams and colleagues report that waist-to-hip ratio was positively associated with risk in women (after adjustment for height and other risk factors) and height with risk in both men and women (after adjustment for BMI and other risk factors) [1]. Pischon et al. argue that obesity is a risk factor in women regardless of fat distribution, whereas in men, fat distribution remains associated with RCC after adjustment for BMI [122]. The biological basis of such a difference is unknown. More complete assessment of overall distribution of body fat and studies in different race/ethnic groups are required to better characterize obesity and renal cancer associations.
1
Obesity and Cancer Epidemiology
19
8.3.3 Genetic Factors and Occupational Exposures Should Be Explored Increased RCC risk has been reported in dozens of industries, and a variety of exposures have been explored to account for these observations [103]. Arsenic, lead, dyes, polyaromatic hydrocarbons, and various solvents are among the candidate agents. It may be interesting to examine interactions between these exposures and other risk factors, such as obesity, hypertension, and smoking. However, few cohorts contain sufficient data about occupational, behavioral, and biological variables to examine all factors. Extensive studies have been carried out concerning the genetics of the hypoxia-inducible factor (HIF) pathway, which underlies the increased risk observed in Von-Hippel Lindau (VHL) syndrome. Tumors in sporadic cases of kidney cancer also exhibit VHL mutations in about 50% of the cases [103]. Given the association of RCC with occupational exposures, genetic variation in metabolismrelated pathways could mediate risk. A number of recent studies have found such associations, but most are based on few cases (<250) and therefore do not allow analysis of interactions with exposure or other risk factors [103]. 8.3.4 Biological Mechanisms Potentially Linking Obesity and RCC Lipid peroxidation has been suggested as a unifying mechanism linking obesity, smoking, and hypertension to RCC [46]. Levels of lipid peroxidation are increased in individuals at increased risk, and thus it could be an intermediary step linking many risk factors to DNA damage by peroxidation products. This hypothesis does seem to account for common features of diverse risk factors [51]. However, this hypothesis is based mostly on observational and indirect evidence, so it remains to be seen whether oxidative stress and its effects on lipid peroxidation is the dominant pathway linking obesity and RCC.
8.4 Conclusion Substantial evidence indicates that obesity is a risk factor for RCC.
9 Colorectal Cancers 9.1 Incidence and Risk Factors Cancers of the colon and rectum are the third most common cancers worldwide [172], with 108,070 and 40,740 new cases, respectively, in 2008 in the United States [7]. Rates of colon cancer increase with development and urbanization, and it is more common in men than women. Age, smoking, history of adenomatous polyps or inflammatory bowel disease, type 2 diabetes, and high intakes of red meat, processed meat, and alcohol are related to increased risk of colorectal cancers. Physical activity is associated with reduced risk, with a stronger influence on colon than rectal cancer.
20
R. Ballard-Barbash et al.
High intakes of vegetables and fruits [7], calcium, milk, dietary fiber, and garlic have been associated with reduced risks [172].
9.2 Overall Statement 9.2.1 BMI, Fat Distribution, and Colorectal Cancers Consistent and significant associations of body weight and adiposity with colorectal cancers have been reported, with differences in the associations by sex and menopausal status [77–81, 101, 130]. Two meta-analyses [77–81, 130] with approximately 23,000 male and 22,000 female cases demonstrated increased risk of colon cancer with a 5 kg/m2 increase in BMI that was stronger in men (RR=1.30, RR=1.24, respectively) than in women (RR=1.12, RR=1.09, respectively). Similar results with 70,000 cases were observed in a meta-analysis of obese compared with normal weight individuals (RR=1.46 men, RR=1.15 women) [101]. Estimates were significant within and across sex [77–81, 101, 130]. A meta-analysis of prospective studies for a 2007 World Cancer Research Fund [172] review found a linear association, with a summary risk estimate for men and women combined of 1.03 (1.02–1.04) per 1 kg/m2 increase [172]. The report concluded that abundant, consistent, and convincing epidemiologic evidence demonstrates a dose–response of greater body fatness and risk of colorectal cancer. Data from a limited number of studies on body fat distribution suggest increased colon cancer risk related to central adiposity [47]; this risk appears to be similar for men and women [31]. Four meta-analyses [77–81, 31, 101, 172] demonstrated stronger relationships with measures of fat distribution than with overall BMI. Comparing the highest versus lowest quartiles of waist-to-hip ratio showed relative risks of 1.91 (95% CI 1.46–2.49) for men and 1.49 (95% CI 1.23–1.81) for women [31]. In a meta-analysis with continuous measures, colon cancer risk increased with increasing waist circumference per 10 cm increment (RR=1.33 and 1.16 in men and women, respectively) and waist-to-hip ratio (0.1 unit increase: RR=1.43 and 1.20 in men and women, respectively) [77–81]. The WCRF review concluded that ample, consistent, and convincing evidence demonstrates a dose–response relation of colorectal cancer risk and abdominal obesity. For central adiposity, evidence suggests a stronger and possibly independent influence of body fat distribution compared to BMI [31]. 9.2.2 BMI and Rectal Cancer Meta-analyses of BMI and rectal cancer risk suggested weaker relationships than with colon cancer [130, 77–81, 172]. The influence of increasing BMI (5 kg/m2 ) was stronger in men (RR=1.09, 95% CI 1.06–1.12) than in women (RR=1.02, 95% CI 1.00–1.05), with significance of the difference in risk estimates between the sexes (p= 0.003) [130]. Similar results were observed in other meta-analyses [77–81, 31]. The few studies that have been published on fat distribution have indicated
1
Obesity and Cancer Epidemiology
21
an increased risk of rectal cancer with increasing waist circumference (per 10 cm increment) in both men and women and non-significantly increased risks for waistto-hip ratio [77–81].
9.3 Selected Topics 9.3.1 Various Factors Affect the Association of Weight and Colorectal Cancer Risk in Women Not only is the association of adiposity with colorectal cancer weaker in women than in men but the sex difference diminishes as women age [47]. Being premenopausal is associated with lower risk than being postmenopausal. In noting a lower risk of colon cancer among postmenopausal women on HRT compared to those not on therapy, many observational studies have suggested a protective role for estrogens. However, recent evaluation of the Women’s Health Initiative intervention trial and an analysis of a representative sample in the United Kingdom do not find a protective effect for either estrogen-alone or estrogen plus progestins in postmenopausal women [159, 160]. Inconsistent associations also have been observed for risks related to higher BMIs and specific subsites for colon cancer [65]. Thus, in women, obesity is a risk factor, and the premenopausal hormonal environment may be protective, and the effect of weight may differ by subsite within the colon. 9.3.2 Differences in Methodologic Approaches Do Not Alter Findings The large number of studies of body mass and colon cancer permits evaluation of several important methodologic issues. First, physical activity is a strong risk factor and a likely confounder to many of the colon cancer studies. Meta-analyses that evaluated the influence of controlling for physical activity concluded that results were similar for all studies compared with those that adjusted for this physical activity [77–81, 101]. Second, investigators who have evaluated the quality of weight and height measures used note that general results of meta-analyses were the same even though they found some bias in self-reported measures among women [130, 77–81]. Third, evaluation of publication bias suggests that small negative studies may be missing from some meta-analyses of colon cancer among women. Including such studies reduced the risk estimates only slightly and did not influence the significance [77–81]. Two other analyses reported publication bias [31, 101], the latter reporting that a 40% excess risk was reduced to 20% for obese compared with normal weight individuals. Further, no one large study was found to be driving the observed relative risks [130]. 9.3.3 Early Life Anthropometry May Be Important Most of the studies focus on adult BMI and not on weight change or the influence of early life body weight. Birth size [146, 105] and growth rates in childhood may be
22
R. Ballard-Barbash et al.
important but have not received much attention. Early adult and middle-aged adult obesity have been linked to increased colon cancer, suggesting both early and later influences on colon carcinogenesis [83]. A review of 21 cohort and 16 case–control studies led to the conclusion that convincing evidence indicates that greater attained height is associated with increased risk of colorectal cancer [172]. Because greater height is an indicator of genetic, environmental, hormonal, and nutritional factors affecting growth in adolescence, these data suggest that early life body size and growth influences later risk and thus should be evaluated in more detail.
9.4 Conclusion Substantial evidence indicates that high BMI and central adiposity are associated with colon cancer, with stronger effects among men than women and for colon than rectal cancer. The associations appear to be linear down to low levels of BMI and waist circumference. Evaluation of weight changes throughout life and other measures of anthropometry are needed. Given the suspicion of differences in risk factors by sex, investigators should address these issues in their analyses.
10 Esophageal Cancer 10.1 Incidence and Risk Factors Cancer of the esophagus occurs most commonly in the squamous cells that line the entire esophagus and as adenocarcinomas arising in glandular tissue near the stomach. There were 16,470 cases in 2008, with most (12,970) occurring in men [7]. In the United States, rates of squamous cell carcinoma are declining and those of adenocarcinoma are increasing in whites and African-Americans [34, 162]. Male sex, African-American race, Barrett’s esophagus, excessive alcohol consumption, tobacco use, and obesity are among the major risk factors for squamous cell cancer, adenocarcinoma, or both.
10.2 Overall Statement Studies to date have been interpreted as suggesting positive associations between BMI and adenocarcinomas and weakly positive or inverse associations between BMI and squamous cell carcinoma [38, 144, 88, 172]. For example, Engeland and colleagues report RRs of 1.8 (95% CI 1.48–2.19) and 2.58 (95% CI 1.81–3.68) in overweight and obese men, respectively, for esophageal adenocarcinoma, but risks of 0.72 (95% CI 0.63–0.82) and 0.68 (95% CI 0.50–0.93) for squamous cell carcinoma. A similar pattern has been reported in a recent comprehensive review of mortality studies [155]. Poor overall nutrition also is believed to be a risk factor
1
Obesity and Cancer Epidemiology
23
for squamous cell and gastric cardia cancer, and this may account for the inverse associations seen with BMI in undernourished populations [161]. Kubo and colleagues report on a meta-analysis of 2 cohort and 12 case–control studies involving 5,000 incident cases of adenocarcinoma of the esophagus and gastric cardia [71]. Elevated BMI was associated with increased risk of esophageal adenocarcinoma (OR 2.1 (1.7–2.4)), but results were more heterogeneous for gastric cardia cancer, with elevated risk associated with higher BMI in studies from Europe and the United States but not China. Individual studies have reported even larger risks. For example, Samnic and colleagues found a RR of 2.7 (1.3–5.6) for esophageal adenocarcinoma in obese compared to normal weight men in a large Swedish cohort.
10.3 Selected Issues 10.3.1 Role of Gastroesophageal Reflux Disease and Barrett’s Esophagus Remains Unclear in Explaining BMI Associations Increased risk associated with BMI for esophageal adenocarcinoma has been attributed to the role of obesity in increasing gastroesophageal reflux disease (GERD) [71, 155]. However, this association has not been conclusively established. At present, risk associated with BMI is perceived to be the same in cases with and without symptomatic GERD [73]. Pera and colleagues reviewed this controversy [119]. Most cases of adenocarcinoma of the esophagus in white men arise from Barrett’s esophagus. However, it has proved difficult to tease apart the roles of GERD and Barrett’s esophagus in esophageal cancer because many cases of Barrett’s esophagus remain undiagnosed in subjects with and without GERD. Thus, there appears to be more to the causal pathway connecting obesity and esophageal cancer than GERD alone and as mentioned, two studies have shown that BMI is a risk factor for adenocarcinoma independent of GERD [26, 73].
10.3.2 Gaps in the Epidemiology Persist A full understanding of the association between obesity and esophageal cancer is hampered by limited data on the different anthropometric measures. In the United States, overall incidence is lower in Hispanics (5.75/100,000) and whites (8.17) than in African-Americans (11.5), with much higher rates of squamous cell carcinoma in African-Americans and adenocarcinoma predominating in whites [175]. Varying distributions of risk factors, such as alcohol and tobacco use, are likely to account for these differences, but studies of risk factors related to energy balance in multiple racial and ethnic groups could be informative. Similarly, contrasts of incidence by acculturation status could help account for some trends in incidence in Hispanic populations. Such rates are predicted to rise because of increasing tobacco and alcohol use in immigrants.
24
R. Ballard-Barbash et al.
10.4 Conclusion The literature concerning esophageal cancer is somewhat inconsistent in part because studies have not routinely examined subtypes of esophageal cancer as separate entities. Nevertheless, considerable evidence indicates that obesity is a risk factor for esophageal adenocarcinoma and that GERD plays a role in this association.
11 Pancreatic Cancer 11.1 Incidence and Risk Factors Although pancreatic cancer occurs less frequently than other cancers (37,680 cases in 2008), it has a high mortality rate (34,290 deaths in 2008) and ranks fourth in cancer mortality in the United States in both men and women [36, 7, 132]. Smoking has been consistently reported to increase risk of this cancer, and diabetes is suggested as a strong causal link, but few other modifiable risk factors have been identified [76].
11.2 Overall Statement Results from cohort and case–control studies and meta-analyses suggest a slight positive association and a dose–response relationship between obesity and pancreatic cancer risk. Early studies of pancreatic cancer and obesity were inconclusive, but they lacked power to explore differences in risk across a range of BMI, used proxy respondents, and did not control for smoking [118]. Thus, potential associations between obesity and pancreatic cancer were limited. Recent studies have shown an excess of pancreatic cancer among obese populations and also suggest a dose–response relationship or increasing risk with increasing BMI [92]. Many prospective cohort studies have reported an increased risk of pancreatic cancer for obese individuals, with a RR of 1.2–3.0 compared to the risk for normal weight individuals [48]. Two recent meta-analyses reported increased risk of pancreatic cancer with obesity, suggesting a 19% increased risk for obese people compared to normal weight people [15] and a 12% increased risk per 5 kg/m2 increase in BMI [76] (Fig. 1.2).
11.3 Selected Issues 11.3.1 Findings About Sex Differences in Incidence Are Inconsistent In most studies, the association between obesity and pancreatic cancer incidence appears to be weaker for women compared to men. However, findings are inconsistent [93].
1
Obesity and Cancer Epidemiology
25
A Studies by sex
Relative Risk (95% CI)
Men Gapstur et al., 2000 Michaud et al., 2001 Stolzenberg-Solomon et al., 2002 Calle et al., 2003 Kuriyama et al., 2005 Patel et al., 2005 Batty et al., 2005 Oh et al., 2005 Rapp et al., 2005 Larsson et al., 2005 Lukanova et al., 2006 Nöthlings et al., 2006 Samanic et al., 2006 Summary estimate
1.85 (1.26-2.71) 1.28 (0.98-1.66) 0.87 (0.69-1.10) 1.16 (1.10-1.24) 0.99 (0.41-2.41) 1.55 (1.21-1.99) 1.06 (0.74-1.52) 0.92 (0.76-1.10) 1.53 (1.07-2.18) 1.34 (0.94-1.90) 0.68 (0.30-1.54) 1.21 (1.07-1.35) 1.02 (0.91-1.14) 1.16 (1.05-1.28)
Women Gapstur et al., 2000 Michaud et al., 2001 Calle et al., 2003 Kuriyama et al., 2005 Patel et al., 2005 Sinner et al., 2005 Rapp et al., 2005 Larsson et al., 2005 Lukanova et al., 2006 Nöthlings et al., 2006 Summary estimate
0.98 (0.58-1.65) 1.16 (0.98-1.37) 1.17 (1.10-1.25) 1.14 (0.60-2.16) 1.32 (0.99-1.74) 1.06 (0.88-1.27) 1.13 (0.82-1.56) 1.22 (0.87-1.70) 1.05 (0.69-1.58) 0.94 (0.83-1.05) 1.10 (1.02-1.19) 0.5
B
0.7
1
1.2
1.5
2
2.5
Relative Risk
Fig. 1.2 (a) Relative risk estimates of pancreatic cancer per 5 kg/m2 increase in body mass index in prospective studies, stratified by sex. Source: Larsson SC, Orsini N, Wolk A. Body mass index and pancreatic cancer risk: A meta-analysis of prospective studies. Int J Cancer. Vol. 129, No. 9, 2007, 1993–8. Reprinted with permission of John Wiley & Sons, Inc. (b) Relative risks of liver cancer associated with overweight and obesity. Relative risk estimates are for overweight and obese persons compared with normal weight persons. Source: Reprinted by permission from Macmillan Publishers Ltd: British Journal of Cancer; Larsson SC and Wolk A. Overweight, obesity and risk of liver cancer: a meta-analysis of cohort studies. Vol. 97, No. 7, 1005–8, 2007. (c) Relative risks of non-Hodgkin’s lymphoma associated with obesity, by histologic subtypes. SLL, small lymphocytic lymphoma; CLL, B-cell chronic lymphocytic leukemia. Source: Larsson SC and Wolk A. Obesity and risk of non-Hodgkin’s lymphoma: A meta-analysis. Int J Cancer. Vol. 121, No. 7, 2007, 1564–70. Reprinted with permission of John Wiley & Sons, Inc. (d) Association of obesity
26
R. Ballard-Barbash et al.
C
Relative risk (95% CI)
Diffuse large B-cell lymphoma Cerhan et al., 2002 Skibola et al., 2004 Pan et al., 2005 Chang et al., 2005 Willett et al., 2005 Cerhan et al., 2005 Summary estimate
1.20 (0.70-1.90) 1.80 (1.30-2.60) 1.35 (0.99-1.83) 1.10 (0.90-1.50) 1.90 (1.30-2.80) 1.35 (1.03-1.76) 1.40 (1.18-1.66)
Follicular lymphoma Cerhan et al., 2002 Skibola et al., 2004 Pan et al., 2005 Chang et al., 2005 Willett et al., 2005 Cerhan et al., 2005 Summary estimate
1.30 (0.70-2.60) 1.60 (1.10-2.50) 1.41 (0.97-2.03) 0.80 (0.50-1.10) 1.20 (0.80-2.00) 0.71 (0.52-0.99) 1.10 (0.82-1.47)
SLL/CLL Cerhan et al., 2002 (SLL) Cerhan et al., 2002 (CLL) Pan et al., 2005 (SLL) Chang et al., 2005 (SLL/CLL) Summary estimate
0.60 (0.20-1.50) 1.30 (0.60-2.60) 1.27 (0.66-2.44) 0.90 (0.70-1.20) 0.95 (0.76-1.20)
0.4
0.6
0.8
Tests for heterogeneity: Diffuse large B-cell lymphoma: p = 0.14, I 2= 39.8% Follicular lymphoma: p = 0.01, I 2 = 66.0% 2 SLL/CLL: p = 0.49, I = 0%
1
1.2
1.5
2
3
Relative risk
D Chronic lymphocytic leukemia Ross et al., 2004 Samanic et al., 2004 (White men) Samanic et al., 2004 (Black men) Engeland et al., 2006 (Men) Engeland et al., 2006 (Women) Overall
Relative risk No. of (95% CI) cases 1.10 (0.60–2.10) 84 1.30 (1.13–1.49)3140 1.72 (1.24–2.39)526 1.14 (0.92–1.40)1660 1.15 (0.97–1.35)1137 1.25 (1.11–1.41)6547
Acute lymphocytic leukemia Samanic et al., 2004 (White men) Samanic et al., 2004 (Black men) Engeland et al., 2006 (Men) Engeland et al., 2006 (Women) Fernberg et al., 2007 Overall
1.33 (0.82–2.15)263 0.69 (0.09–5.01)38 2.77 (1.49–5.12)119 1.49 (0.63–3.51)121 1.46 (0.43–4.98)47 1.65 (1.16–2.35)588
Acute myeloid leukemia Ross et al. 2004 Samanic et al., 2004 (White men) Samanic et al., 2004 (Black men) Engeland et al., 2006 (Men) Engeland et al., 2006 (Women) Fernberg et al., 2007 Overall
2.40 (1.30–4.50)72 1.59 (1.33–1.90)1607 2.64 (1.80–3.85)287 1.12 (0.89–1.42)1374 1.17 (1.00–1.36)1240 1.30 (0.77–2.17)224 1.52 (1.19–1.95)4804
Chronic myeloid leukemia Samanic et al., 2004 (White men) Samanic et al., 2004 (Black men) Engeland et al., 2006 (Men) Engeland et al., 2006 (Women) Fernberg et al., 2007 Overall
1.15 (0.92–1.45)1263 1.32 (0.77–2.27)253 1.65 (1.18–2.31)494 1.19 (0.91–1.57)419 1.35 (0.64–2.84)101 1.26 (1.09–1.46)2530
0.6
0.8
1 1.2 Relative risk
1.5
2
3
4.5
Fig. 1.2 (continued) with incidence of leukemia subtypes: chronic lymphocytic leukemia, acute lymphocytic leukemia, acute myeloid leukemia, chronic myeloid leukemia. Source: Larsson SC and Wolk A. Overweight and obesity and incidence of leukemia: A meta-analysis of cohort studies. Int J Cancer. Vol. 122, No. 6, 2008, 1418–1421. Reprinted with permission of John Wiley & Sons, Inc
1
Obesity and Cancer Epidemiology
27
11.3.2 Central Adiposity May Play a Role Few studies have examined the relationship of obesity and pancreatic cancer using measures of abdominal obesity, such as waist circumference or waist-to-hip ratio. Some studies have found an increased risk of pancreatic cancer with waist circumference or waist-to-hip ratio but not with increasing BMI [14, 93]. Additionally, men and women who reported more central weight gain compared to peripheral weight gain also were at an increased risk, with men at greater risk compared to women [118]. Central adiposity is related to glucose intolerance and is a risk factor for diabetes. This link therefore may be the mechanism by which central adiposity increases risk for pancreatic cancer [14]. The fact that men tend to gain weight more centrally than do women, combined with the findings of a stronger association of obesity with cancer in men compared to women, may suggest a role for central adiposity in pancreatic cancer incidence [118, 93].
11.4 Conclusion Although recent studies were adjusted for smoking status and proxy reporting for weight, the possibility of confounding from smoking and diabetes cannot be excluded. Future studies should examine the link between obesity and pancreatic cancer in never smokers and those without a history of diabetes. They also should use measures of fat distribution, including waist circumference and waist-to-hip measures, in men and women.
12 Hepatocellular Cancer 12.1 Incidence and Risk Factors Hepatocellular cancer (HCC) occurs relatively rarely in the United States (21,370 cases in 2008) [7], but is the third most common cause of death worldwide [115]. However, incidence and mortality rates of HCC have been increasing rapidly in the United States since the mid-1980s [131]. Hepatitis C and hepatitis B infection and alcoholic liver disease are known risk factors. However, only hepatitis C infection has contributed to the increase in the United States in recent decades, suggesting remaining unidentified risk factors [77–81]. Diabetes and aflatoxins also are known to be risk factors [22].
12.2 Overall Statement A limited literature has examined the association of increased risk of HCC with obesity [172]. A meta-analysis reported a 17% increased risk of HCC incidence among 5,037 overweight individuals and 89% increased risk among 6,042 individuals who
28
R. Ballard-Barbash et al.
were obese compared to individuals at a normal weight, with a higher risk in men compared to women [77–81]. Although it is not yet clear how obesity affects HCC incidence in relation to other risk factors, recent studies have suggested that obesity is independently related to liver cancer in patients with liver cirrhosis [126] and hepatitis C patients [110]. Furthermore, in the United States, obesity has become a major cause of non-alcoholic steatohepatitis (NASH) among adults and children. As NASH may progress to chronic active hepatitis, which is a common precursor to the development of HCC, it is anticipated that obesity may become a major cause of HCC in the United States.
12.3 Conclusion Overall, a small and limited literature has reported an increased risk of HCC with obesity and a suggestion of a dose–response relationship [172]. As many cohort studies do not adjust or stratify by alcohol use or hepatitis B or C status, current studies are limited by potential confounding issues. Future studies should aim to adjust or stratify by these risk factors.
13 Gallbladder Cancer 13.1 Incidence and Risk Factors Gallbladder cancer occurs relatively rarely (9,500 cases in 2008), shows geographic variation, and is more common in women then in men [127, 7]. The strongest known risk factor is gallstones. Because obesity increases the risk of gallstones, it is thought that obesity may be related to gallbladder cancer [77–81].
13.2 Overall Statement A few studies report an increased risk of gallbladder cancer with obesity and suggest a dose–response relationship. However, most studies have had small sample sizes [24]. A recent meta-analysis of 3,288 cases reported an increased risk of 15% for overweight individuals and 66% for obese individuals compared to normal weight individuals [77–81]. The WCRF/AICR report also documented a 23% increased risk per 5 kg/m2 in cohort data and a 19% increased risk per 5 kg/m2 in case–control data [172]. The association between obesity and gallbladder cancer was stronger in women compared to men [77–81]. This may be because women are at higher risk of gallstone disease compared to men across all ages [148].
1
Obesity and Cancer Epidemiology
29
13.3 Conclusion Overall, a small but consistent literature reports an increased risk of gallbladder cancer with obesity. Obesity is a probable cause of gallbladder cancer directly as well as indirectly through the formation of gallstones [172].
14 Lung Cancer 14.1 Incidence and Risk Factors Lung cancer is the second most common cause of cancer for both men and women (114,690 and 100,330 cases in 2008, respectively) and is the most common cause of cancer deaths in men and women (90,810 and 71,030 deaths, respectively) [7]. Tobacco use remains the major contributor to cancer incidence and death in the United States. Recent declines in lung cancer incidence and death among men and leveling off of the increase in these rates in women have been attributed to declines in tobacco use associated with widespread implementation of tobacco control and improvements in tobacco cessation services and treatment. Other less common causes of lung cancer include asbestos, arsenic in drinking water, and (in smokers only) pharmacological doses of beta-carotene. Fruits and foods containing carotenoids have been classified as probable factors in decreasing risk of lung cancer [172]. Other dietary components continue to be examined relative to lung cancer risk. Limited evidence suggests that physical activity may be associated with a reduced risk for lung cancer [84].
14.2 Overall Statement Twenty-one cohort studies, 24 case–control studies, and 1 ecologic study have examined the association between body fatness (largely using BMI) and lung cancer risk [172]. Nearly all of these studies have found an inverse association between BMI and lung cancer, which becomes less strong after adjustment for potential confounding by smoking status. In addition, studies have examined the association of BMI and lung cancer in subgroups, such as current and non-smokers, and found little association between BMI and lung cancer among never or former smokers. A systematic review of studies of BMI and lung cancer found that only five prospective cohort studies reported separately by smoking status. This review found a summary risk ratio of 0.76 (95% CI 0.67–0.85) in smokers with no association (RR=0.91; 95% CI 0.76–1.10) in non-smokers. This meta-analysis found that study-specific inverse associations became stronger as the proportion of smokers in the study increased. Overall, this evidence suggests that the inverse association between BMI and lung cancer is due to confounding from tobacco use. It is likely that risk factors for lung cancer may be different among tobacco users as compared to non-tobacco
30
R. Ballard-Barbash et al.
users. However, as the majority of lung cancer is caused by tobacco use, little research has examined risk factors in detail among non-tobacco users. There is no known mechanism by which greater body fatness might protect against lung cancer. The only study that has examined the association of body fat distribution by waist circumference and lung cancer risk found evidence of a differential association between waist circumference and histologic subtypes of lung cancer [113]. In that study among middle-aged Iowa women, risk for small cell and squamous cell lung cancer was increased by three-fold among women in the highest quintile of waist circumference. In contrast, no association was observed between waist circumference and adenocarcinoma of the lung.
14.3 Conclusion The inverse association between BMI and lung cancer is considered to occur because tobacco use results in lower body weights and acts as a confounding factor in these studies. There is no known mechanism by which a higher BMI would be protective against lung cancer. BMI is not associated with lung cancer among non-smokers.
15 Head and Neck Cancers 15.1 Incidence and Risk Factors Cancers of the oral cavity and pharynx are more common in men than in women and are ranked as the ninth most common cancer in men (25,310 cases in 2008), with 5,210 deaths occurring in 2008 [7]. Tobacco and alcohol use account for more than 90% of cancers of the head and neck in developing countries [62, 63].
15.2 Overall Statement A limited number of case–control studies have examined the association between weight or BMI and risk of head and neck cancers; no cohort studies or metaanalyses have been published on this topic. Several case–control studies have found an inverse association between weight and/or BMI and cancer of the oral cavity and pharynx [10]. Similar to findings for lung cancer, several studies found either no association or weaker associations between BMI and oral cancer among never smokers compared to current smokers. These results suggest that BMI is not causally related to risk of head and neck cancers. However, similar to lung cancer, the causal factors contributing to head and neck cancer among never smokers may be different than those among current or former smokers.
1
Obesity and Cancer Epidemiology
31
15.3 Conclusion The inverse association between BMI and head and neck cancer is considered to occur because tobacco use results in lower body weights and acts as a confounding factor in these studies. There is no known mechanism by which a higher BMI would be protective of head and neck cancer.
16 Lymphoid and Hematological Cancers Cancers of the lymphoid and hematological systems are primarily lymphomas (Hodgkin’s or non-Hodgkin’s), leukemias, and multiple myelomas. Although these cancers affect similar cancer sites, they have different mechanisms and associations with obesity, physical activity, and diet [172].
16.1 Incidence and Risk Factors Non-Hodgkin’s Lymphoma (NHL) is a heterogeneous group of cancers that occur in the immune system and has become the fifth most common cancer among men and women in the United States (66,120 cases in 2008) [7]. Most NHL cases have no known risk factors. Known risk factors, which account for a small proportion of NHL cases, include severe immunosuppression, auto-immune diseases, and HIV infection [97, 169]. Multiple myeloma (19,920 cases in 2008) [7] has no established risk factors other than male sex, increasing age, African-American race, and family history [6]. Like NHL, leukemia (44,270 cases in 2008) [7] has different subgroups, including acute lymphoblastic leukemia (ALL), chronic lymphocytic leukemia (CLL), acute myeloid leukemia (AML), and chronic myeloid leukemia (CML) [61]. Known risk factors for leukemia include smoking, infection with T-cell leukemia virus, radiation, and benzene exposure [172].
16.2 Overall Statement The risks of some lymphoid and hematological cancers may be related to obesity, but the evidence base is small and may differ by patient subgroup. A recent meta-analysis of 21,720 cases reported a 7% increased risk of NHL for overweight compared to normal weight individuals and a 20% increased risk in those who are obese compared to those of normal weight [77–81]. Although scant and inconsistent evidence exists about obesity and other lymphoma subtypes, it has been reported that obesity is associated with diffuse large B-cell lymphoma and follicular lymphoma [153]. A similar meta-analysis of 13,120 cases of multiple myeloma found an increased risk of 12–43% for those who were overweight and a 27–82% increased risk for those who were obese compared to normal weight adults [77–81].
32
R. Ballard-Barbash et al.
In leukemia, a meta-analysis suggested a 14% increased risk in overweight individuals and a 39% increase for obese individuals [82]. Risk appears to vary by subgroups of leukemia [139, 82]. A limited number of studies have examined early influences and the incidence of these cancers. However, some studies have examined weight gain in survivors of childhood cancers [108].
16.3 Conclusion Overall, obesity may be associated with risk of several types of lymphoid and hematological cancers. Future studies should analyze subgroups of lymphoma and leukemia patients separately unless it is biologically plausible to group them in the same category.
17 Cancer Sites with Insufficient Evidence for Conclusions Limited epidemiologic evidence exists on weight, BMI, or anthropometric measures and incidence of several cancers, including malignant melanoma, testicular, bladder cancers, and other types of lymphomas. Because of the limited ability to draw conclusions, these cancer sites are not summarized in this chapter. A summary of evidence for these cancer sites can be located in Food, Nutrition, Physical Activity and the Prevention of Cancer: A Global Perspective, a 2007 report from the American Institute for Cancer Research [172].
18 Common Areas of Challenge and Opportunity A number of gaps and areas requiring further exploration are common for research related to obesity and cancer. These challenges include the need for more comprehensive anthropometric measures to more completely assess the effects of overall and central adiposity and visceral fat; changes in association of these measures over the life cycle with various cancers; the need for more integrative studies that examine the combined effects of diet, physical activity, and body size and adiposity; the limited data among a number of major racial and ethnic groups other than whites and some Asian subgroups; the limited data on whether obesity associations vary by newer approaches for characterizing biologic and genetic tumor subtypes; and the many potential mechanisms by which obesity may influence risk for different cancers. These issues have been addressed in a number of previous reviews [10, 172]. In this review we comment on two emerging areas and related areas of research: use of animal models to explore obesity and cancer over the life cycle and the influence of in utero and early childhood exposures on both body size and subsequent adult cancer risk. Finally, we comment briefly on the potential to evaluate the influence
1
Obesity and Cancer Epidemiology
33
of weight loss on cancer risk from existing research and the categories of major mechanisms that may underlie obesity and cancer associations.
18.1 Animal Models May Help Guide Selection of Obesity Phenotypes for Epidemiologic Studies Inadequate characterization of body composition is a frequent problem in fully understanding associations between obesity and cancer risk based on epidemiologic studies. Animal models allow rapid exploration of the role of subcutaneous versus visceral fat, fatty liver, and other elements of the obesity phenotype that may require substantially greater time and costs in human studies [58]. More complete characterization of body composition within animal models might provide useful guidance in selecting from a range of possibilities for phenotyping in new studies. For example, rodent models of obesity, physical activity, and carcinogenesis can include detailed phenotyping in experimental studies [137, 59]. Furthermore, transgenic models such as the “fatless” mouse can help determine the contribution of adipose tissue to disease risk [107]. Another powerful tool for exploring effects of obesity on carcinogenesis involves chromosome substitution strains [56]. Substitution strains resistant to diet-induced obesity have already been identified, and such lines may offer powerful tools to explore traits that are normally correlated in wild-type populations, such as obesity and intake or aspects of regional fat deposition [20]. Great caution must be used in direct guidance from animal models because of major differences between humans and some of the main animal models of obesity and carcinogenesis, particularly mice [91, 165]. Nevertheless, strong parallels between human and mouse responses to obesity are possible [106]. In addition, animal models can be used to examine whether the influence of specific combinations of diet and physical activity interventions has differential effects on weight and metabolic changes by genetic factors that influence risk of obesity and cancer and whether those effects vary across the life cycle. Given the short time from exposure to event in most animal models, these issues can be examined over a much shorter time frame than in human studies with many more combinations of interventions tested than might be feasible to test in humans over a similar time frame.
18.2 In Utero and Early Childhood Exposures Influence Body Size and Subsequent Adult Cancer Risk Evidence for an influence of diet and other exposures early in life on obesity and cancers is scattered through the literature. The strongest evidence is from animal literature, where energy restriction modulates lower cancer incidence and longer latency and is related to lower body weight [166, 70, 60]. Primate studies also support an association of limited energy intake and healthier and longer lives [141, 157].
34
R. Ballard-Barbash et al.
In humans, the most compelling data come from a natural experiment of the Dutch famine in the Netherlands during the winter of 1945. Offspring that experienced severely limited calories in utero in the first trimester were heavier and longer at birth than those not exposed and as adults have higher BMIs, a more atherogenic lipid profile, and other serologic abnormalities. Those exposed to famine in the third trimester were lighter, shorter, and had smaller heads at birth than those unexposed, while as adults they demonstrate reduced glucose tolerance and other abnormalities [138]. Large studies have observed associations of higher birth size and risk of a variety of cancers [5] and the major determinant of offspring birth weight is maternal pre-pregnancy weight. Thus, factors in the in utero environment related to higher maternal BMI influence the developing fetus in a manner that results in higher birth weights and predisposes offspring to later cancer risk. The association between birth weight and later body size is not strong, however. Childhood obesity is related to lower risk for breast [130] and prostate [49] cancers in Western populations, although the mechanisms for these associations remain poorly understood. It may be important to evaluate several of these growth and body size parameters simultaneously, as adolescent growth and body size are related to breast cancer [4], in addition to diet during reproductive years. Thus, evidence from a variety of fields suggests that some exposures in utero, childhood, adolescence, and early adulthood have an impact on adult body size and some cancers. This time period should be investigated further, especially in terms of potential interactions with other exposures and possible mechanisms.
18.3 Existing Evidence of the Influence of Weight Loss on Cancer Risk Research from basic, animal, and clinical interventions suggests that reduction of fat mass through dietary, physical activity, and weight loss interventions leads to beneficial changes in a number of biological markers of mechanisms that are either postulated or known to reduce cancer risk for several cancers. Most of this research has focused on breast and colon cancer [99]. Interpreting the limited evidence on weight loss and cancer risk from observational epidemiologic research poses challenges in part because it is difficult from observational studies to clearly identify whether weight loss was intentional or due to underlying disease. In addition, the limited number of people who are able to achieve sustained weight loss in the current obesogenic environment of the United States and in many countries worldwide may further limit the ability to quantify the influence of weight loss on cancer risk. Further, a randomized control prevention trial of weight loss with cancer as an endpoint is not feasible for several reasons. Strong randomized clinical trial evidence has documented the benefit of weight loss for diabetes, hypertension, and cardiovascular disease. These endpoints are more common than cancer-specific endpoints and occur over a shorter time frame than it is anticipated cancer reductions would occur. However, given the evidence on the beneficial effects of weight loss on intermediate
1
Obesity and Cancer Epidemiology
35
endpoints and other chronic diseases that are common among people at risk for cancer, avoiding weight gain and attempting to lose weight for overweight and obese people are recommended [170]. As cohorts of bariatric surgery patients expand it may be possible to learn more about the effect of weight loss in this group of obese patients on subsequent cancer incidence and mortality. Research on cancer-related outcomes following bariatric surgery is emerging as mentioned in the introduction.
18.4 Other Mechanisms Also Appear to Be Influential Although another chapter in this volume (Chapter 5) is dedicated to detailing the mechanisms relating obesity to various cancers, a brief overview consistent with the epidemiologic review is presented here. The reproductive epidemiologic risk factors suggest a hormonal etiology for breast, ovarian, and endometrial cancers. A high BMI at various points in the life cycle also may be related to a steroid hormonal profile contributing to the development of these cancers. Postmenopausal obesity is related to higher exposures to bioavailable estrogens synthesized in the adipose tissue. The hormonal environment from obesity in adolescence and early adulthood is consistent with exposure to high levels of estrogens, androgens, and insulin and low levels of progesterone that affect many tissues, including the breast and ovaries [133, 67, 66]. The increased risk associated with high waist-to-hip ratio for a variety of cancers is consistent with hypothesized mechanisms of higher insulin and androgen exposures [23]. Many of these endocrinologic factors related to obesity can influence the balance of proliferation and apoptosis [23]. The association of height with breast, colorectal, and prostate cancers suggests growth factors during adolescence influence other tissues, perhaps making them susceptible to environmental influences later in life. Adult insulin-like growth factors have been related to colorectal, premenopausal breast, and prostate cancers with implications for environmental factors influencing the process. The higher levels of insulin-like growth factors in men compared with women throughout life may explain some of the excess risk for colorectal cancer among men [130]. Other obesity-related factors implicated in the etiology of several of the cancers are growth factors, adipokines including adiponectin and leptin, cytokines, and other inflammatory factors, altered immune response, and oxidative stress [23, 82, 130]. In general, many of these potential mechanistic risk factors have not been investigated together with obesity in relation to risk of cancer. Research on energy balance and many cancers is being examined in the NCI Cohort Consortium and includes efforts to examine potential genetic influences of these associations.
19 Summary and Conclusion Evidence is considered as 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
36
R. Ballard-Barbash et al.
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. Studies that have examined these combinations of factors have found much greater increases in risk among people who have these adverse health profiles. The continued global epidemics of obesity and diabetes mellitus are likely to contribute to global increases in a number of obesity-related cancers. Acknowledgments Special thanks to Anne Rogers for editing the manuscript and Penny RandallLevy for preparing the bibliography.
References 1. Adams KF, Leitzmann MF, Albanes D et al. (2008). Body size and renal cell cancer incidence in a large US cohort study. Am J Epidemiol, 168(3):268–277. 2. Adams TD, GRess RE, Smith SC et al. (2007). Long-term mortality after gastric bypass surgery. N Engl J Med, 357(8):753–761. 3. Adams TD, Stroug AM, Gress RE, et al. (2009). Cancer incidence and mortality after gastric bypass surgery. Obesity, 17(4):796–802. 4. Ahlgren M, Melbye M, Wohlfahrt J, Sorensen TI (2006). Growth patterns and the risk of breast cancer in women. Int J Gynecol Cancer, 16 Suppl 2:569–575. 5. Ahlgren M, Wohlfahrt J, Olsen LW, Sorensen TI, Melbye M (2007). Birth weight and risk of cancer. Cancer, 110(2):412–419. 6. Alexander DD, Mink PJ, Adami HO et al. (2007). Multiple myeloma: a review of the epidemiologic literature. Int J Cancer, 120 Suppl 12:40–61. 7. American Cancer Society (2008). Cancer facts & figures 2008. American Cancer Society, Atlanta. 8. Amling CL (2005). Relationship between obesity and prostate cancer. Curr Opin Urol, 15(3):167–171. 9. Anderson JP, Ross JA, Folsom AR (2004). Anthropometric variables, physical activity, and incidence of ovarian cancer: The Iowa Women’s Health Study. Cancer, 100(7):1515–1521. 10. Ballard-Barbash R, Friedenreich C, Slattery M, Thune I (2006). Obesity and body composition. In: Schottenfeld D, Fraumeni JF, Jr. (eds.), Cancer epidemiology and prevention, 3rd edn, pp. 422–448. Oxford University Press, New York. 11. Baumgartner RN (1997). Body composition and anthropometry. In: Gary PJ, Owen G, Eldridge TO (eds.), The New Mexico Aging Process Study: 1980–1997., pp. 88–145. University of New Mexico, New Mexico. 12. Benichou J, Chow WH, McLaughlin JK, Mandel JS, Fraumeni JF, Jr. (1998). Population attributable risk of renal cell cancer in Minnesota. Am J Epidemiol, 148(5):424–430. 13. Beral V, Bull D, Reeves G (2005). Endometrial cancer and hormone-replacement therapy in the Million Women Study. Lancet, 365(9470):1543–1551. 14. Berrington de Gonzalez A, Spencer EA, Bueno-de-Mesquita HB et al. (2006). Anthropometry, physical activity, and the risk of pancreatic cancer in the European prospective investigation into cancer and nutrition. Cancer Epidemiol Biomarkers Prev, 15(5):879–885. 15. Berrington de Gonzalez A, Sweetland S, Spencer E (2003). A meta-analysis of obesity and the risk of pancreatic cancer. Br J Cancer, 89(3):519–523.
1
Obesity and Cancer Epidemiology
37
16. Bjorge T, Tretli S, Engeland A (2004). Relation of height and body mass index to renal cell carcinoma in two million Norwegian men and women. Am J Epidemiol, 160(12): 1168–1176. 17. Boyd NF, Martin LJ, Sun L et al. (2006). Body size, mammographic density, and breast cancer risk. Cancer Epidemiol Biomarkers Prev, 15(11):2086–2092. 18. Brawley OW, Jani AB, Master V (2007). Prostate cancer and race. Curr Probl Cancer, 31(3):211–225. 19. Brennan P, van der Hel O, Moore LE et al. (2008). Tobacco smoking, body mass index, hypertension, and kidney cancer risk in central and eastern Europe. Br J Cancer, 99(11):1912–1915. 20. Buchner DA, Burrage LC, Hill AE et al. (2008). Resistance to diet-induced obesity in mice with a single substituted chromosome. Physiol Genomics, 35(1):116–122. 21. Buschemeyer WC, III, Freedland SJ (2007). Obesity and prostate cancer: epidemiology and clinical implications. Eur Urol, 52(2):331–343. 22. Caldwell SH, Crespo DM, Kang HS, Al-Osaimi AM (2004). Obesity and hepatocellular carcinoma. Gastroenterology, 127(5 Suppl 1):S97–103. 23. Calle EE, Kaaks R (2004). Overweight, obesity and cancer: epidemiological evidence and proposed mechanisms. Nat Rev Cancer, 4(8):579–591. 24. Calle EE, Thun MJ (2004). Obesity and cancer. Oncogene, 23(38):6365–6378. 25. Chang SC, Ziegler RG, Dunn B et al. (2006). Association of energy intake and energy balance with postmenopausal breast cancer in the prostate, lung, colorectal, and ovarian cancer screening trial. Cancer Epidemiol Biomarkers Prev, 15(2):334–341. 26. Chow WH, Blot WJ, Vaughan TL et al. (1998). Body mass index and risk of adenocarcinomas of the esophagus and gastric cardia. J Natl Cancer Inst, 90(2):150–155. 27. Chow WH, Devesa SS, Warren JL, Fraumeni JF, Jr. (1999). Rising incidence of renal cell cancer in the United States. JAMA, 281(17):1628–1631. 28. Chow WH, Gridley G, Fraumeni JF, Jr., Jarvholm B (2000). Obesity, hypertension, and the risk of kidney cancer in men. N Engl J Med, 343(18):1305–1311. 29. Christou NV, Lieberman M, Sampalis F, Sampalis JS (2008). Bariatric surgery reduces cancer risk in morbidly obese patients. Surv Obes Relat Dis, 4(6):691–5. 30. Ciampolillo A, DeTullio C, Giorgino F (2005). The IGF-I/IGF-I receptor pathway: Implications in the pathophysiology of thyroid cancer. Curr Med Chem, 12(24): 2881–2891. 31. Dai Z, Xu YC, Niu L (2007). Obesity and colorectal cancer risk: a meta-analysis of cohort studies. World J Gastroenterol, 13(31):4199–4206. 32. Dal Maso L, Franceschi S, Negri E et al. (2002). Body size indices at different ages and epithelial ovarian cancer risk. Eur J Cancer, 38(13):1769–1774. 33. Davies L, Welch HG (2006). Increasing incidence of thyroid cancer in the United States, 1973–2002. JAMA, 295(18):2164–2167. 34. Devesa SS, Blot WJ, Fraumeni JF, Jr. (1998). Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer, 83(10):2049–2053. 35. Dirx MJ, Voorrips LE, Goldbohm RA, van den Brandt PA (2001). Baseline recreational physical activity, history of sports participation, and postmenopausal breast carcinoma risk in the Netherlands Cohort Study. Cancer, 92(6):1638–1649. 36. Edwards BK, Brown ML, Wingo PA et al. (2005). Annual report to the nation on the status of cancer, 1975–2002, featuring population-based trends in cancer treatment. J Natl Cancer Inst, 97(19):1407–1427. 37. Engeland A, Tretli S, Bjorge T (2003). Height, body mass index, and ovarian cancer: a follow-up of 1.1 million Norwegian women. J Natl Cancer Inst, 95(16):1244–1248. 38. Engeland A, Tretli S, Bjorge T (2004). Height and body mass index in relation to esophageal cancer; 23-year follow-up of two million Norwegian men and women. Cancer Causes Control, 15(8):837–843.
38
R. Ballard-Barbash et al.
39. Etzioni R, Penson DF, Legler JM et al. (2002). Overdiagnosis due to prostate-specific antigen screening: lessons from U.S. prostate cancer incidence trends. J Natl Cancer Inst 94(13):981–990. 40. Fairfield KM, Willett WC, Rosner BA, Manson JE, Speizer FE, Hankinson SE (2002). Obesity, weight gain, and ovarian cancer. Obstet Gynecol, 100(2):288–296. 41. Frankel S, Smith GD, Donovan J, Neal D (2003). Screening for prostate cancer. Lancet, 361(9363):1122–1128. 42. Freedland SJ, Platz EA (2007). Obesity and prostate cancer: making sense out of apparently conflicting data. Epidemiol Rev, 29:88–97. 43. Friberg E, Mantzoros CS, Wolk A (2007). Diabetes and risk of endometrial cancer: a population-based prospective cohort study. Cancer Epidemiol Biomarkers Prev, 16(2): 276–280. 44. Friberg E, Orsini N, Mantzoros CS, Wolk A (2007). Diabetes mellitus and risk of endometrial cancer: a meta-analysis. Diabetologia, 50(7):1365–1374. 45. Friedenreich C, Cust A, Lahmann PH et al. (2007). Anthropometric factors and risk of endometrial cancer: the European prospective investigation into cancer and nutrition. Cancer Causes Control, 18(4):399–413. 46. Gago-Dominguez M, Castelao JE (2006). Lipid peroxidation and renal cell carcinoma: further supportive evidence and new mechanistic insights. Free Radic Biol Med, 40(4): 721–733. 47. Giovannucci E (2003). Diet, body weight, and colorectal cancer: a summary of the epidemiologic evidence. J Womens Health (Larchmt ), 12(2):173–182. 48. Giovannucci E, Michaud D (2007). The role of obesity and related metabolic disturbances in cancers of the colon, prostate, and pancreas. Gastroenterology, 132(6):2208–2225. 49. Giovannucci E, Rimm EB, Stampfer MJ, Colditz GA, Willett WC (1997). Height, body weight, and risk of prostate cancer. Cancer Epidemiol Biomarkers Prev, 6(8):557–563. 50. Gong Z, Neuhouser ML, Goodman PJ et al. (2006). Obesity, diabetes, and risk of prostate cancer: results from the prostate cancer prevention trial. Cancer Epidemiol Biomarkers Prev, 15(10):1977–1983. 51. Greenland S, Gago-Dominguez M, Castelao JE (2004). The value of risk-factor (“blackbox”) epidemiology. Epidemiology, 15(5):529–535. 52. Greer JB, Modugno F, Ness RB, Allen GO (2006). Anthropometry and the risk of epithelial ovarian cancer. Cancer, 106(10):2247–2257. 53. Gunnell D, Berney L, Holland P et al. (2000). How accurately are height, weight and leg length reported by the elderly, and how closely are they related to measurements recorded in childhood? Int J Epidemiol, 29(3):456–464. 54. Gunnell D, Okasha M, Smith GD, Oliver SE, Sandhu J, Holly JM (2001). Height, leg length, and cancer risk: a systematic review. Epidemiol Rev, 23(2):313–342. 55. Hard GC (1998). Recent developments in the investigation of thyroid regulation and thyroid carcinogenesis. Environ Health Perspect, 106(8):427–436. 56. Hill AE, Lander ES, Nadeau JH (2006). Chromosome substitution strains: a new way to study genetically complex traits. Methods Mol Med, 128:153–172. 57. Hoyo C, Berchuck A, Halabi S et al. (2005). Anthropometric measurements and epithelial ovarian cancer risk in African-American and White women. Cancer Causes Control, 16(8):955–963. 58. Huffman DM, Johnson MS, Watts A, Elgavish A, Eltoum IA, Nagy TR (2007). Cancer progression in the transgenic adenocarcinoma of mouse prostate mouse is related to energy balance, body mass, and body composition, but not food intake. Cancer Res, 67(1): 417–424. 59. Huffman DM, Moellering DR, Grizzle WE, Stockard CR, Johnson MS, Nagy TR (2008). Effect of exercise and calorie restriction on biomarkers of aging in mice. Am J Physiol Regul Integr Comp Physiol, 294(5):R1618–R1627.
1
Obesity and Cancer Epidemiology
39
60. Hursting SD, Lavigne JA, Berrigan D, Perkins SN, Barrett JC (2003). Calorie restriction, aging, and cancer prevention: mechanisms of action and applicability to humans. Annu Rev Med, 54:131–152. 61. Ilhan G, Karakus S, Andic N (2006). Risk factors and primary prevention of acute leukemia. Asian Pac J Cancer Prev, 7(4):515–517. 62. International Agency for Research on Cancer Working Group (1986). International Agency for Research on Cancer: tobacco smoking. IARC monographs on the evaluation of the carcinogenic risks to humans, vol. 38. IARC Press, Lyon, France. 63. International Agency for Research on Cancer Working Group (1988). International Agency for Research on Cancer: alcohol drinking. IARC monographs on the evaluation of the carcinogenic risks to humans, vol. 44. IARC Press, Lyon, France. 64. International Agency for Research on Cancer Working Group (2002). IARC Working Group on the evaluation of cancer-preventive strategies. IARC handbooks of cancer prevention, vol. 6, p. 315. Weight control and physical activity. IARC Press, Lyon, France. 65. Jacobs ET, Ahnen DJ, Ashbeck EL et al. (2009). Association between body mass index and colorectal neoplasia at follow-up colonoscopy: a pooling study. Am J Epidemiol, 169(6):657–666. 66. Jasik CB, Lustig RH (2008). Adolescent obesity and puberty: the “perfect storm”. Ann N Y Acad Sci, 1135:265–279. 67. Key TJ, Allen NE, Verkasalo PK, Banks E (2001). Energy balance and cancer: the role of sex hormones. Proc Nutr Soc, 60(1):81–89. 68. Kipping RR, Jago R, Lawlor DA (2008). Obesity in children. Part 1: Epidemiology, measurement, risk factors, and screening. BMJ, 337:a1824. 69. Kristal AR, Arnold KB, Schenk JM et al. (2008). Dietary patterns, supplement use, and the risk of symptomatic benign prostatic hyperplasia: results from the prostate cancer prevention trial. Am J Epidemiol, 167(8):925–934. 70. Kritchevsky D (2002). Caloric restriction and experimental carcinogenesis. Hybrid Hybridomics, 21(2):147–151. 71. Kubo A, Corley DA (2006). Body mass index and adenocarcinomas of the esophagus or gastric cardia: a systematic review and meta-analysis. Cancer Epidemiol Biomarkers Prev, 15(5):872–878. 72. Kuczmarski MF, Kuczmarski RJ, Najjar M (2001). Effects of age on validity of self-reported height, weight, and body mass index: findings from the Third National Health and Nutrition Examination Survey, 1988–1994. J Am Diet Assoc, 101(1):28–34. 73. Lagergren J, Bergstrom R, Lindgren A, Nyren O (1999). Symptomatic gastroesophageal reflux as a risk factor for esophageal adenocarcinoma. N Engl J Med, 340(11): 825–831. 74. Lahmann PH, Lissner L, Gullberg B, Olsson H, Berglund G (2003). A prospective study of adiposity and postmenopausal breast cancer risk: the Malmo Diet and Cancer Study. Int J Cancer, 103(2):246–252. 75. Larsson SC, Friberg E, Wolk A (2007). Carbohydrate intake, glycemic index and glycemic load in relation to risk of endometrial cancer: A prospective study of Swedish women. Int J Cancer, 120(5):1103–1107. 76. Larsson SC, Orsini N, Wolk A (2007). Body mass index and pancreatic cancer risk: A metaanalysis of prospective studies. Int J Cancer, 120(9):1993–1998. 77. Larsson SC, Wolk A (2007). Body mass index and risk of multiple myeloma: a metaanalysis. Int J Cancer, 121(11):2512–2516. 78. Larsson SC, Wolk A (2007). Obesity and colon and rectal cancer risk: a meta-analysis of prospective studies. Am J Clin Nutr, 86(3):556–565. 79. Larsson SC, Wolk A (2007). Obesity and risk of non-Hodgkin’s lymphoma: a meta-analysis. Int J Cancer, 121(7):1564–1570. 80. Larsson SC, Wolk A (2007). Obesity and the risk of gallbladder cancer: a meta-analysis. Br J Cancer, 96(9):1457–1461.
40
R. Ballard-Barbash et al.
81. Larsson SC, Wolk A (2007). Overweight, obesity and risk of liver cancer: a meta-analysis of cohort studies. Br J Cancer, 97(7):1005–1008. 82. Larsson SC, Wolk A (2008). Overweight and obesity and incidence of leukemia: a metaanalysis of cohort studies. Int J Cancer, 122(6):1418–1421. 83. Le Marchand L, Wilkens LR, Mi MP (1992). Obesity in youth and middle age and risk of colorectal cancer in men. Cancer Causes Control, 3(4):349–354. 84. Lee IM, Oguma Y (2006). Physical activity. In: Schottenfeld D, Fraumeni JF, Jr. (eds.), Cancer epidemiology and prevention, 3rd edn, pp. 449–467. Oxford University Press, New York. 85. Lee SY, Gallagher D (2008). Assessment methods in human body composition. Curr Opin Clin Nutr Metab Care, 11(5):566–572. 86. Levi F, Ferlay J, Galeone C et al. (2008). The changing pattern of kidney cancer incidence and mortality in Europe. BJU Int, 101(8):949–958. 87. Levi F, Lucchini F, Negri E, La VC (2004). Declining mortality from kidney cancer in Europe. Ann Oncol, 15(7):1130–1135. 88. Lindblad M, Rodriguez LA, Lagergren J (2005). Body mass, tobacco and alcohol and risk of esophageal, gastric cardia, and gastric non-cardia adenocarcinoma among men and women in a nested case-control study. Cancer Causes Control, 16(3):285–294. 89. Lipworth L, Tarone RE, McLaughlin JK (2006). The epidemiology of renal cell carcinoma. J Urol, 176(6 Pt 1):2353–2358. 90. Lohman TG, Roche AF, Martorell R (1988). Anthropometric standardization reference manual. Human Kinetics Books, Champaign, IL. 91. Lowell BB, Flier JS (1997). Brown adipose tissue, beta 3-adrenergic receptors, and obesity. Annu Rev Med, 48:307–316. 92. Lowenfels AB, Maisonneuve P (2002). Epidemiologic and etiologic factors of pancreatic cancer. Hematol Oncol Clin North Am, 16(1):1–16. 93. Luo J, Margolis KL, Adami HO, LaCroix A, Ye W (2008). Obesity and risk of pancreatic cancer among postmenopausal women: the Women’s Health Initiative (United States). Br J Cancer, 99(3):527–531. 94. MacInnis RJ, English DR (2006). Body size and composition and prostate cancer risk: systematic review and meta-regression analysis. Cancer Causes Control, 17(8): 989–1003. 95. Mack WJ, Preston-Martin S, Dal ML et al. (2003). A pooled analysis of case-control studies of thyroid cancer: cigarette smoking and consumption of alcohol, coffee, and tea. Cancer Causes Control, 14(8):773–785. 96. Malin A, Matthews CE, Shu XO et al. (2005). Energy balance and breast cancer risk. Cancer Epidemiol Biomarkers Prev, 14(6):1496–1501. 97. Maskarinec G, Erber E, Gill J, Cozen W, Kolonel LN (2008). Overweight and obesity at different times in life as risk factors for non-Hodgkin’s lymphoma: the multiethnic cohort. Cancer Epidemiol Biomarkers Prev, 17(1):196–203. 98. Mathers CD, Shibuya K, Boschi-Pinto C, Lopez AD, Murray CJ (2002). Global and regional estimates of cancer mortality and incidence by site: I. Application of regional cancer survival model to estimate cancer mortality distribution by site. BMC Cancer, 2:36. 99. McTiernan A (2006). Cancer prevention and management through exercise and weight control, p. 608. CRC Press, Boca Raton, FL. 100. Modesitt SC, van Nagell J, Jr. (2005). The impact of obesity on the incidence and treatment of gynecologic cancers: a review. Obstet Gynecol Surv, 60(10):683–692. 101. Moghaddam AA, Woodward M, Huxley R (2007). Obesity and risk of colorectal cancer: a meta-analysis of 31 studies with 70,000 events. Cancer Epidemiol Biomarkers Prev, 16(12):2533–2547. 102. Moore LE, Wilson RT, Campleman SL (2005). Lifestyle factors, exposures, genetic susceptibility, and renal cell cancer risk: a review. Cancer Invest, 23(3):240–255.
1
Obesity and Cancer Epidemiology
41
103. Moore T, Beltran L, Carbajal S et al. (2008). Dietary energy balance modulates signaling through the Akt/mammalian target of rapamycin pathways in multiple epithelial tissues. Cancer Prev Res (Phila Pa), 1(1):65–76. 104. Must A, Dietz WH, Willett WC (1993). Remote recall of childhood height, weight and body build by elderly subjects. Am J Epidemiol, 138(1):56–64. 105. Nilsen TI, Romundstad PR, Troisi R, Potischman N, Vatten LJ (2005) Birth size and colorectal cancer risk: a prospective population based study. Gut, 54(12):1728–1732. 106. Nunez NP, Carpenter CL, Perkins SN et al. (2007). Extreme obesity reduces bone mineral density: complementary evidence from mice and women. Obesity (Silver Spring), 15(8):1980–1987. 107. Nunez NP, Oh WJ, Rozenberg J et al. (2006). Accelerated tumor formation in a fatless mouse with type 2 diabetes and inflammation. Cancer Res, 66(10):5469–5476. 108. Oeffinger KC (2008). Are survivors of acute lymphoblastic leukemia (ALL) at increased risk of cardiovascular disease? Pediatr Blood Cancer, 50(2 Suppl):462–467. 109. Oh SW, Yoon YS, Shin SA (2005). Effects of excess weight on cancer incidences depending on cancer sites and histologic findings among men: Korea National Health Insurance Corporation Study. J Clin Oncol, 23(21):4742–4754. 110. Ohki T, Tateishi R, Sato T et al. (2008). Obesity is an independent risk factor for hepatocellular carcinoma development in chronic hepatitis C patients. Clin Gastroenterol Hepatol, 6(4):459–464. 111. Olsen CM, Green AC, Whiteman DC, Sadeghi S, Kolahdooz F, Webb PM (2007). Obesity and the risk of epithelial ovarian cancer: a systematic review and meta-analysis. Eur J Cancer, 43(4):690–709. 112. Olsen CM, Nagle CM, Whiteman DC, Purdie DM, Green AC, Webb PM (2008). Body size and risk of epithelial ovarian and related cancers: a population-based case-control study. Int J Cancer, 123(2):450–456. 113. Olson JE, Yang P, Schmitz K, Vierkant RA, Cerhan JR, Sellers TA (2002). Differential association of body mass index and fat distribution with three major histologic types of lung cancer: evidence from a cohort of older women. Am J Epidemiol, 156(7):606–615. 114. Park SK, Kang D, McGlynn KA et al. (2008). Intrauterine environments and breast cancer risk: meta-analysis and systematic review. Breast Cancer Res, 10(1):R8. 115. Parkin DM, Bray F, Ferlay J, Pisani P (2005). Global cancer statistics, 2002. CA Cancer J Clin, 55(2):74–108. 116. Pascual D, Borque A (2008). Epidemiology of kidney cancer. Adv Urol, :782381. 117. Patel AV, Feigelson HS, Talbot JT et al. (2008). The role of body weight in the relationship between physical activity and endometrial cancer: results from a large cohort of US women. Int J Cancer, 123(8):1877–1882. 118. Patel AV, Rodriguez C, Bernstein L, Chao A, Thun MJ, Calle EE (2005). Obesity, recreational physical activity, and risk of pancreatic cancer in a large U.S. Cohort. Cancer Epidemiol Biomarkers Prev, 14(2):459–466. 119. Pera M, Manterola C, Vidal O, Grande L (2005). Epidemiology of esophageal adenocarcinoma. J Surg Oncol, 92(3):151–159. 120. Phipps AI, Malone KE, Porter PL, Daling JR, Li CI (2008). Body size and risk of luminal, HER2-overexpressing, and triple-negative breast cancer in postmenopausal women. Cancer Epidemiol Biomarkers Prev, 17(8):2078–2086. 121. Pienta KJ, bate-Shen C, Agus DB et al. (2008). The current state of preclinical prostate cancer animal models. Prostate, 68(6):629–639. 122. Pischon T, Lahmann PH, Boeing H et al. (2006). Body size and risk of renal cell carcinoma in the European Prospective Investigation into Cancer and Nutrition (EPIC). Int J Cancer, 118(3):728–738. 123. Potosky AL, Miller BA, Albertsen PC, Kramer BS (1995). The role of increasing detection in the rising incidence of prostate cancer. JAMA, 273(7):548–552. 124. Presti JC, Jr. (2005). Obesity and prostate cancer. Curr Opin Urol, 15(1):13–16.
42
R. Ballard-Barbash et al.
125. Purdie DM, Green AC (2001). Epidemiology of endometrial cancer. Best Pract Res Clin Obstet Gynaecol, 15(3):341–354. 126. Qian Y, Fan JG (2005). Obesity, fatty liver and liver cancer. Hepatobiliary Pancreat Dis Int, 4(2):173–177. 127. Randi G, Franceschi S, La VC (2006). Gallbladder cancer worldwide: geographical distribution and risk factors. Int J Cancer, 118(7):1591–1602. 128. Rapp K, Klenk J, Ulmer H et al. (2008). Weight change and cancer risk in a cohort of more than 65,000 adults in Austria. Ann Oncol, 19(4):641–648. 129. Reeves GK, Pirie K, Beral V, Green J, Spencer E, Bull D (2007). Cancer incidence and mortality in relation to body mass index in the Million Women Study: cohort study. BMJ, 335(7630):1134. 130. Renehan AG, Tyson M, Egger M, Heller RF, Zwahlen M (2008). Body-mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studies. Lancet, 371(9612):569–578. 131. Ries LAG, Harkins D, Krapcho M, Mariotto A, Miller BA, Feuer EJ, Clegg L, Eisner MP, Horner MJ, Howlader N, Hayat M, Hankey BF, Edwards BK, (eds.), (2006). SEER cancer statistics review, 1975–2003, Bethesda, MD National Cancer Institute. http://seer.cancer.gov/csr/1975_2003/. 132. Ries LAG, Melbert D, Krapcho M, Stinchcomb DG, Howlader N, Horner MJ, Mariotto A, Miller BA, Feuer EJ, Altekruse SF, Lewis DR, Clegg L, Eisner MP, Reichman M, Edwards BK (2008). SEER Cancer Statistics Review, 1975–2005 http://seer.cancer.gov/csr/1975_2005/. Accessed: 23 Apr 2008. 133. Risch HA (1998). Hormonal etiology of epithelial ovarian cancer, with a hypothesis concerning the role of androgens and progesterone. J Natl Cancer Inst, 90(23):1774–1786. 134. Robinson WR, Poole C, Godley PA (2008). Systematic review of prostate cancer‘s association with body size in childhood and young adulthood. Cancer Causes Control, 19(8):793–803. 135. Roddam AW, Allen NE, Appleby P et al. (2008). Insulin-like growth factors, their binding proteins, and prostate cancer risk: analysis of individual patient data from 12 prospective studies. Ann Intern Med, 149(7):461–68. 136. Rodriguez C, Freedland SJ, Deka A et al. (2007). Body mass index, weight change, and risk of prostate cancer in the Cancer Prevention Study II Nutrition Cohort. Cancer Epidemiol Biomarkers Prev, 16(1):63–69. 137. Rogers CJ, Berrigan D, Zaharoff DA et al. (2008). Energy restriction and exercise differentially enhance components of systemic and mucosal immunity in mice. J Nutr, 138(1):115–122. 138. Roseboom TJ, van der Meulen JH, Ravelli AC, Osmond C, Barker DJ, Bleker OP (2001). Effects of prenatal exposure to the Dutch famine on adult disease in later life: an overview. Mol Cell Endocrinol, 185(1–2):93–98. 139. Ross JA, Parker E, Blair CK, Cerhan JR, Folsom AR (2004). Body mass index and risk of leukemia in older women. Cancer Epidemiol Biomarkers Prev, 13(11 Pt 1):1810–1813. 140. Rossing MA, Remler R, Voigt LF, Wicklund KG, Daling JR (2001). Recreational physical activity and risk of papillary thyroid cancer (United States). Cancer Causes Control, 12(10):881–885. 141. Roth GS, Ingram DK, Lane MA (2001). Caloric restriction in primates and relevance to humans. Ann N Y Acad Sci, 928:305–315. 142. Rowland ML (1990). Self-reported weight and height. Am J Clin Nutr, 52(6): 1125–1133. 143. Ruder EH, Dorgan JF, Kranz S, Kris-Etherton PM, Hartman TJ (2008). Examining breast cancer growth and lifestyle risk factors: early life, childhood, and adolescence. Clin Breast Cancer, 8(4):334–342. 144. Ryan AM, Rowley SP, Fitzgerald AP, Ravi N, Reynolds JV (2006). Adenocarcinoma of the oesophagus and gastric cardia: male preponderance in association with obesity. Eur J Cancer, 42(8):1151–1158.
1
Obesity and Cancer Epidemiology
43
145. Samaras TT, Elrick H, Storms LH (2003). Birthweight, rapid growth, cancer, and longevity: a review. J Natl Med Assoc, 95(12):1170–1183. 146. Sandhu MS, Luben R, Day NE, Khaw KT (2002). Self-reported birth weight and subsequent risk of colorectal cancer. Cancer Epidemiol Biomarkers Prev, 11(9):935–938. 147. Schouten LJ, Rivera C, Hunter DJ et al. (2008). Height, body mass index, and ovarian cancer: a pooled analysis of 12 cohort studies. Cancer Epidemiol Biomarkers Prev, 17(4):902–912. 148. Shaffer EA (2006). Gallstone disease: Epidemiology of gallbladder stone disease. Best Pract Res Clin Gastroenterol, 20(6):981–996. 149. Shin A, Matthews CE, Shu XO et al. (2009). Joint effects of body size, energy intake, and physical activity on breast cancer risk. Breast Cancer Res Treat, 113(1):153–161. 150. Silva Idos S, De Stavola B, McCormack V (2008). Birth size and breast cancer risk: reanalysis of individual participant data from 32 studies. PLoS Med, 5(9):e193. 151. Silvera SA, Jain M, Howe GR, Miller AB, Rohan TE (2006). Energy balance and breast cancer risk: a prospective cohort study. Breast Cancer Res Treat, 97(1):97–106. 152. Sjostrom L, Narbro K, Sjostrom D, et al. (2007). Effects of bariatric surgery on mortality in Swedish obese subjects. New Eng J Med, 357(8): 741–752. 153. Skibola CF (2007). Obesity, diet and risk of non-Hodgkin lymphoma. Cancer Epidemiol Biomarkers Prev, 16(3):392–395. 154. Skolarus TA, Wolin KY, Grubb RL, III (2007). The effect of body mass index on PSA levels and the development, screening and treatment of prostate cancer. Nat Clin Pract Urol, 4(11):605–614. 155. Smith M, Zhou M, Whitlock G et al. (2008). Esophageal cancer and body mass index: results from a prospective study of 220,000 men in China and a meta-analysis of published studies. Int J Cancer, 122(7):1604–1610. 156. Sorosky JI (2008). Endometrial cancer. Obstet Gynecol, 111(2 Pt 1):436–447. 157. Spindler SR (2005). Rapid and reversible induction of the longevity, anticancer and genomic effects of caloric restriction. Mech Ageing Dev, 126(9):960–966. 158. Stevens J, Keil JE, Waid LR, Gazes PC (1990). Accuracy of current, 4-year, and 28-year self-reported body weight in an elderly population. Am J Epidemiol, 132(6): 1156–1163. 159. Tannen RL, Weiner MG, Xie D, Barnhart K (2007a). A simulation using data from a primary care practice database closely replicated the women‘s health initiative trial. J Clin Epidemiol, 60(7):686–695. 160. Tannen RL, Weiner MG, Xie D, Barnhart K (2007b). Estrogen affects post-menopausal women differently than estrogen plus progestin replacement therapy. Hum Reprod, 22(6):1769–1777. 161. Tran GD, Sun XD, Abnet CC et al. (2005). Prospective study of risk factors for esophageal and gastric cancers in the Linxian general population trial cohort in China. Int J Cancer, 113(3):456–463. 162. Trivers KF, Sabatino SA, Stewart SL (2008). Trends in esophageal cancer incidence by histology, United States, 1998–2003. Int J Cancer, 123(6):1422–1428. 163. van den Brandt PA, Spiegelman D, Yaun SS et al. (2000). Pooled analysis of prospective cohort studies on height, weight, and breast cancer risk. Am J Epidemiol, 152(6):514–527. 164. Velie EM, Nechuta S, Osuch JR (2005). Lifetime reproductive and anthropometric risk factors for breast cancer in postmenopausal women. Breast Dis, 24:17–35. 165. Virtue S, Vidal-Puig A (2008) It’s not how fat you are, it‘s what you do with it that counts. PLoS Biol, 6(9):e237. 166. Weindruch R, Walford RL (1988). The retardation of aging and disease by dietary restriction, p. 436. Charles C. Thomas, Springfield, IL. 167. WHO Consultation on Obesity (2000). Obesity: preventing and managing the global epidemic. World Health Organization, Report of a WHO consultation, WHO technical reports series, no. 894. Geneva, Switzerland.
44
R. Ballard-Barbash et al.
168. WHO Expert Committee (1995). Physical status: the use and interpretation of anthropometry. World Health Organization, Report of a WHO Expert Committee. WHO Technical Reports Series, No. 854. Geneva, Switzerland. 169. Willett EV, Morton LM, Hartge P et al. (2008). Non-Hodgkin lymphoma and obesity: a pooled analysis from the InterLymph Consortium. Int J Cancer, 122(9):2062–2070. 170. Wolin KY, Colditz GA (2008). Can weight loss prevent cancer? Br J Cancer, 99(7):995–999. 171. Woodrow G (2009). Body composition analysis techniques in the aged adult: indications and limitations. Curr Opin Clin Nutr Metab Care, 12(1):8–14. 172. World Cancer Research Fund, American Institute for Cancer Research (2007). Food, nutrition, physical activity, and the prevention of cancer: a global perspective. American Institute for Cancer Research, Washington, DC. 173. World Health Organization (2008). WHO global infobase: data for saving lives http://www.who.int/infobase/comparestart.aspx. Accessed: 10 Feb 2009. 174. Wright ME, Chang SC, Schatzkin A et al. (2007). Prospective study of adiposity and weight change in relation to prostate cancer incidence and mortality. Cancer, 109(4):675–684. 175. Wu X, Chen VW, Ruiz B, Andrews P, Su LJ, Correa P (2006). Incidence of esophageal and gastric carcinomas among American Asians/Pacific Islanders, whites, and blacks: subsite and histology differences. Cancer, 106(3):683–692. 176. Xing M (2008). Recent advances in molecular biology of thyroid cancer and their clinical implications. Otolaryngol Clin North Am, 41(6):1135–1146, ix. 177. Zheng SL, Sun J, Wiklund F et al. (2008). Cumulative association of five genetic variants with prostate cancer. N Engl J Med, 358(9):910–919. 178. Zuccolo L, Harris R, Gunnell D et al. (2008). Height and prostate cancer risk: a large nested case-control study (ProtecT) and meta-analysis. Cancer Epidemiol Biomarkers Prev, 17(9):2325–2336.
Chapter 2
Obesity and Cancer: Epidemiology in Racial/Ethnic Minorities Colleen Doyle
Abstract Overweight and obesity continue to be major public health concerns in the United States and increasingly, throughout the world [1]. Obesity is the nation’s fastest rising public health problem and has become the second leading cause of preventable death in the United States, second only to tobacco use [1]. Obesity rates among US adults increased by more than 75% between 1991 and 2006, and rates doubled in children and tripled in teens over the past 20 years. While obesity rates have increased dramatically among most of the population, particular racial, ethnic, and socioeconomically disadvantaged groups have experienced disproportionate increases in the prevalence of overweight and obesity over this time [2]. This chapter will explore the differences in these trends, discuss implications for cancer prevention and control, examine contributing factors and review potential strategies for positively influencing overweight and obesity trends among all population groups.
1 The Burden of Overweight and Obesity in Racial/Ethnic Minority Populations Obesity rates in the United States have increased dramatically since the early 1980s, although recent data indicate no significant change in obesity prevalence between 2003–2004 and 2005–2006 for men or women [3]. Despite this positive finding, data from national surveys consistently indicate that there are large disparities in the prevalence of overweight and obesity among women, children, and adolescents in the United States. Non-Hispanic black women and children, Mexican-American women and children, Native Americans, and Pacific Islanders are all disproportionally affected. C. Doyle (B) Nutrition and Physical Activity, American Cancer Society, Oklahoma City, Oklahoma 73123-1538, USA e-mail:
[email protected]
N.A. Berger (ed.), Cancer and Energy Balance, Epidemiology and Overview, Energy Balance and Cancer 2, DOI 10.1007/978-1-4419-5515-9_2, C Springer Science+Business Media, LLC 2010
45
46
C. Doyle
1.1 Adults In adults, NHANES data consistently demonstrate trends of higher overweight and obesity prevalence for non-Hispanic blacks and Mexican Americans compared with non-Hispanic whites. The most current data from 2005 to 2006 show that nonHispanic blacks had the highest prevalence of overweight, followed closely by Mexican Americans. Large disparities exist in obesity prevalence by race/ethnicity among women; among men, however, the prevalence of obesity did not differ significantly by race/ethnic group (See Fig. 2.1). Non-Hispanic black and Mexican-American women were more likely to be obese than white women. Approximately 53% of non-Hispanic black women and 51% of Mexican-American women 40–59 years of age were obese, compared with about 39% of non-Hispanic white women of the same age. Among women 60 years and older, 61% of non-Hispanic black women were obese compared with 32% of non-Hispanic white women and 37% of Mexican-American women [2].
Fig. 2.1 Prevalence of obesity, by age, race/ethnicity, and sex, adults aged 20 years and older: United States, 2005–2006. (Source: CDC/NCHS, National Health and Nutrition Examination Survey) (1) Significantly different from the non-Hispanic white population (2) Significantly different from the non-Hispanic white and Mexican-American population. Note: Obesity is defined as body mass index>30
While NHANES provides data on non-Hispanic blacks and Mexican Americans, the survey does not include sufficient numbers of people from other minority backgrounds. Other data sources, do, however, indicate obesity prevalence is also higher across adult age ranges for American Indians and Alaska Natives, other Hispanic populations, Native Hawaiians, and Pacific Islanders when compared with non-Hispanic whites [4–5]. Studies also have found that the longer racial/ethnic minority immigrants are in the United States; the prevalence of obesity increases and approaches rates seen among US-born citizens [6–8]. Analysis of data over the past three decades reveals that the prevalence of overweight has increased at an average annual rate of approximately 0.3–0.9 percentage
2
Obesity and Cancer: Epidemiology in Racial/Ethnic Minorities
47
points across different racial/ethnic population groups. Assuming a similar increase in trends, it is estimated that by 2015, 75% of US adults are likely to be overweight or obese. Given the existing disparities in prevalence, some population groups will be more seriously affected: it is projected that among men, 78% of Caucasians, 66% of African Americans, and 82% of Hispanics will be overweight and among women, 69% of Caucasians, 87% of African Americans and 80% of Hispanics will be overweight [9].
1.2 Children and Adolescents The prevalence of obesity has tripled since 1980 among children 6–11 years of age and adolescents 12–17 years of age, and as in adults, racial/ethnic disparities in obesity prevalence are also seen in children and adolescents [10–12]. A recent study showed that these disparities may begin as early as by 4 years of age. (See Fig 2.2). Because it is estimated that about half of youngsters who are overweight as children will remain overweight in adulthood and that 70% of those who are overweight by adolescence will remain overweight as adults [13], it is critical that targeted efforts be made to establish positive, lifelong eating and exercise habits during childhood.
Fig. 2.2 Body mass index for age at or above the 95th percentile by race/ethnicity in 1999–2006. [92]
Although the overall prevalence of childhood obesity continued to increase during the first half of this decade (17% in 2006 vs. 14% in 2000), the differences by race/ethnicity appear to be diminishing, in part due to increases in obesity in non-Hispanic white children. Over time, non-Hispanic black children have experienced the steepest increase in overweight, as compared to Mexican American and non-Hispanic white children [14]. While there was no significant change in prevalence rates between 2003–2004 and 2005–2006, further data tracking will be needed to determine if rates have indeed reached a plateau. The most recent NHANES data showed that for boys ages 6–11 years of age, the prevalence rate of obesity (defined as ≥95% of the 2000 BMI-for-age growth charts)
48
C. Doyle
was highest for Mexican American boys (27%), followed by non-Hispanic black boys (18.6%) and non-Hispanic white boys (15.5%). Among girls, non-Hispanic black girls had the highest prevalence (24%), followed by Mexican American girls (19.7%) and non-Hispanic white girls (14.4%) [10]. For adolescent boys, the rate of obesity was higher among Mexican American boys (22.1%) than among non-Hispanic white boys (17.3%) and non-Hispanic black boys (18.5%) [10]. Data from NHANES III (1988–1994) through NHANES 2003– 2006 showed that the largest increases in the prevalence of obesity occurred among non-Hispanic black boys (7.8%) and Mexican American boys (8.0%) compared to non-Hispanic white boys (5.7%). Among non-Hispanic white boys, the prevalence of obesity increased from 11.6 to 17.3%. Among non-Hispanic black boys, the prevalence of obesity increased from 10.7 to 18.5%. Among Mexican American boys, the prevalence of obesity increased from 14.1 to 22.1%. Non-Hispanic black adolescent girls had the highest prevalence of obesity (27.7%) compared to that of non-Hispanic white (14.5%) and Mexican American (19.9%) girls [10]. Data from NHANES III (1988–1994) through NHANES 2003– 2006 showed that non-Hispanic black girls experienced the largest increase in the prevalence of obesity (14.5%) compared to non-Hispanic white girls (7.1%) and Mexican American (10.7%) girls. Among non-Hispanic white girls, the prevalence of obesity increased from 7.4 to 14.5%. Among non-Hispanic black girls, the prevalence of obesity increased from 13.2 to 27.7%. Among Mexican American girls, the prevalence of obesity increased from 9.2 to 19.9%. See Fig. 2.3 These rates may
Fig. 2.3 Obesity∗ , adolescents 12–19 years, by gender and race/ethnicity†, US, 1976–2006 ∗ BMI at or above the sex- and age-specific 95th percentile BMI cutoff points from the 2000 sexspecific BMI-for-age CDC Growth Charts. †Persons of Mexican origins may be of any race. Data estimates for white (non-Hispanic) and African American (non-Hispanic) races for 1999–2002 may not be strictly comparable with estimates for earlier years because of changes in Standards for Federal data on Race and Ethnicity. The differences in overweight estimates for current and earlier standards for these race categories do not exceed 0.5 percentage points. ‡Data for Mexican Americans are for 1982–1984
2
Obesity and Cancer: Epidemiology in Racial/Ethnic Minorities
49
be related in part to differences in stages of pubertal maturation. Girls who mature early tend to have higher a BMI during the teenage years than girls who mature later [15], and this relationship appears to be strongest in non-Hispanic black girls [16]. On average, these girls undergo pubertal maturation earlier than non-Hispanic white girls, which may account for some of the racial differences seen in adolescent obesity. Based on current trends, it is projected that by 2015, for children ages 6–11 years, the prevalence of overweight will be 23%. Among boys, 22% of nonHispanic whites, 24% of non-Hispanic blacks and 33% of Mexican Americans will be overweight. Among girls, 19% of non-Hispanic whites, 31% of nonHispanic blacks, 22% of Mexican Americans will be overweight. For adolescents aged 12–19 years, the prevalence of overweight will be 24%. Among males, 23% of non-Hispanic whites, 25% of non-Hispanic African Americans, and 28% of Mexican Americans will be overweight. Among females, 19% of non-Hispanic whites, 32% of non-Hispanic blacks, and 22% of Mexican Americans will be overweight [9].
1.3 Socioeconomic Disparities Population-based surveys also indicate a higher prevalence of obesity in populations with lower socioeconomic status (SES) [17–18]. Because of the association between race/ethnicity and socioeconomic status, some have hypothesized that differences in obesity among different racial/ethnic minorities might be easily explained by individual SES. Recent studies, however, have indicated that the racial/ethnic differences in obesity cannot be explained by SES alone; that two commonly used markers for SES – education and income – do not reflect SES level equally across racial/ethnic groups; and that the relationship between race/ethnicity, SES, gender, and obesity is quite complex [9]. For example, Wang and colleagues report that overall, those with less than a high school education tend to have higher rates of obesity than those with more education, although non-Hispanic black women are an exception and had the lowest prevalence of obesity when compared to those with more education. SES differences in children and adolescents appear to be equally complex and inconsistent across ethnicity, age, and gender [19–21]. Recent NHANES data, for example, shows an inverse association of obesity prevalence with SES in non-Hispanic white girls, while higher SES was associated with higher obesity rates in non-Hispanic black girls. A number of recent studies have also attempted to determine the extent to which neighborhood location may be related to obesity rates. In a study conducted by Drewnowski and colleagues, the authors concluded that neighborhood property values in the Seattle area predict local obesity rates better than education or income level [22]. For each additional $100,000 in the median price of homes, obesity rates in a given ZIP code dropped by 2%, and obesity rates reached 30% in the most deprived areas but were only around 5% in the most affluent ZIP codes.
50
C. Doyle
1.4 Geographic and Urban–Rural Differences Geographic variation in obesity has been reported by state and by degree of urbanization. Obesity rates remain highest in Southern states, according to the 2007 Behavioral Risk Factor Surveillance System (BRFSS) survey, nine of the top ten most obese states were in the South. In addition, all ten states with the highest rates of diabetes and hypertension, nine of the ten states with the highest rates of physical inactivity, and eight of the ten states with the highest rates of poverty are in the South. Northeastern and Western states have the lowest obesity rates [23]. The highest prevalence of obesity was seen in Mississippi, West Virginia, and Alabama, while the lowest prevalence was seen in Colorado and Hawaii [24]. Results of the National Health Interview Survey show that rural populations, when compared to urban and suburban populations, have a higher prevalence of obesity [25– 27]. Differences in obesity rates among these population groups may also reflect socioeconomic differences, with rural areas tending to experience higher levels of poverty [25].
2 Implications for Cancer Incidence and Mortality Because of the impact that excess weight has on cancer risk, current and projected trends in overweight and obesity among all population groups threaten to jeopardize progress made in cancer incidence and mortality since the early 1990s. In the United States, overweight and obesity contribute to 14–20% of all cancer-related mortality [28]. Overweight and obesity are clearly associated with increased risk for developing many cancers, including cancers of the breast in postmenopausal women, colon, endometrium, adenocarcinoma of the esophagus, and kidney. Evidence is highly suggestive that obesity also increases risk for cancers of the liver, prostate, stomach, pancreas, gallbladder, thyroid, ovary, and cervix, and for multiple myeloma, Hodgkin’s lymphoma, and aggressive prostate cancer. Increasing evidence also suggests that overweight and obesity increases the risk of breast cancer recurrence and decreases survival, and evidence is accumulating regarding other sites, as well [29–32]. Related diet and physical activity behaviors may also play a role in recurrence. Consumption of the typical “Western” diet among colorectal cancer survivors has been associated with a 3.5 times increased risk of recurrence [33]. The Women’s Intervention Nutrition Study (WINS) demonstrated reduced risk of breast cancer recurrence among intervention subjects following a low-fat diet as compared to controls (9.8% vs 12.4%) [34]. In the Nurses Health Study, higher levels of post-treatment physical activity were associated with a 26–40% reduction in the risk of breast cancer recurrence, breast cancer-specific mortality, and all-cause mortality [35]. Thus overweight and obesity are such strong risk factors for so many different types of cancers, and given disparate rates of overweight and obesity among
2
Obesity and Cancer: Epidemiology in Racial/Ethnic Minorities
51
racial/ethnic minority populations, it is likely that excess weight plays a role, at least in part, in the disparities in cancer incidence seen among these population groups.
2.1 Racial and Ethnic Differences in Cancer Incidence in the United States Compared with non-Hispanic whites, African Americans experience higher incidence rates for cancers of the colorectum, lung, prostate, liver, kidney, and cervix, and Hispanics, Asians, and Pacific Islanders experience higher incidence of cancers of the liver, stomach, and cervix [36]. The causes of these inequalities are complex and are predominantly thought to reflect social and economic disparities as opposed to biologic differences associated with race/ethnicity. These include inequalities in work, wealth, income, education, housing and overall standard of living, barriers to high-quality health care, and racial discrimination. Environmental issues that affect access to and availability of healthy, affordable foods and opportunities for safe, enjoyable physical activity, thus contributing to disparities in overweight and obesity, are part of the complexity of factors influencing disparities in cancer incidence. Studies that look specifically at obesity and risk of cancer in minority populations are limited. There is some evidence that, among African American women, the risk of breast cancer associated with obesity may be absent or less than that of other population groups [37–39]. However, a recent report showed that African American women who have a high BMI are more likely to have an advanced stage of breast cancer at diagnosis [40]. Another report showed that obese Hispanic women were twice as likely to develop breast cancer as non-obese Hispanics, but the researchers did not detect a difference in risk for obese Hispanic women before and after menopause [41]. Some have hypothesized that women with higher BMI’s may be less likely to undergo recommended cancer screening tests, such as Pap tests and mammography, because of embarrassment and/or discomfort associated with these tests. While a number of studies have shown that obese white women were significantly less likely to undergo cervical or breast cancer screening, in part because of feelings of embarrassment and/or discomfort associated with these screenings, BMI was not associated with mammography utilization in African American women, nor Pap testing among African American or Hispanic women, although overweight and obese Hispanic women were more likely than normal weight Hispanic women to cite cost concerns as a reason for not undergoing screening [42].
2.2 The Bottom Line Current trends in obesity have already negatively impacted cancer rates in the United States. Increasing trends in adenocarcinoma of the esophagus and kidney cancer
52
C. Doyle
have been attributed to, in part, the increasing rates of obesity [43–45]. And while the incidence of both colorectal cancer and postmenopausal breast cancer continues to decline, according to the American Cancer Society, it is likely that the declines in both would have started earlier and would have been steeper had it not been for the increasing prevalence of obesity. Efforts to curtail and ultimately reverse the obesity epidemic among all population groups are likely to have a considerable impact on reducing incidence rates of the many cancers that are impacted by excess weight.
3 Determinants of Obesity The determinants of obesity in the United States are complex, numerous, and involve a combination of social, economic, and other environmental effects, as well as individual behavior effects. When considering why trends in overweight and obesity have increased so dramatically overtime, it is necessary to consider the individual factors that play a role, but the broader environment in which individual food and physical activity choices occur must be taken into account. Indeed, based on the rapidity with which overweight and obesity trends have increased among all population groups, it is likely that social, economic, and other environmental factors – rather than individual behavior factors – have been key drivers of the accelerating trends over the past three decades.
3.1 Behavioral Determinants of Obesity A variety of individual factors have been associated with overweight and obesity. These include, but are not limited to, fruit and vegetable consumption, fast-food intake, soft drink consumption, television time, breastfeeding, and physical activity levels. Racial/ethnic differences in some of these factors may contribute in part to disparities seen in obesity rates among different population groups.
3.2 Fruit and Vegetable Consumption Emerging evidence suggests that increasing fruit and vegetable consumption may be associated with lower rates of obesity. While more research is needed to determine the exact relationship between produce consumption and weight, there are many benefits to consuming a diet high in fruits and vegetables. Unfortunately, consumption levels among both adults and youth, including by race/ethnicity, have been essentially flat for years. Among adults, 2007 BRFSS data indicate that while overall consumption of fruit and vegetable is low, there is little difference in fruit and vegetable consumption by race/ethnicity: 24.1% non-Hispanic whites, 23.1% of non-Hispanic blacks,
2
Obesity and Cancer: Epidemiology in Racial/Ethnic Minorities
53
and 24.7% of Mexican Americans report eating five or more servings of fruit and vegetable a day [24]. Similar to adults, overall consumption of fruits and vegetables remains low among youth. Overall, only 21.4% of youth consume fruits and vegetables five or more times a day. Only 18.8% of non-Hispanic white youth, 24.9% of non-Hispanic black youth, and 24% of Mexican American youth report eating the recommended number of fruit and vegetable servings [46].
3.3 Fast-Food Intake Fast-food consumption is associated with consumption of more calories, more saturated fat, fewer fruits and vegetables, and less milk [47–49]. Racial/ethnic differences in total calorie and fat consumption have been linked in part to high levels of fast-food consumption [48, 50]. Data indicates that fast food currently makes up nearly three-quarters of total restaurant visits [51] and that approximately one-fifth of restaurant meals were purchased from a car (e.g., drive-through or curbside) in 2005, up from 14% in 1998 [52]. Despite the addition of healthier items such as salads to restaurant menus, the top five most popular foods ordered in restaurants in 2005, for consumption on-site or take out, were for men – hamburgers, french fries, pizza, breakfast sandwiches, and side salads; for women – french fries, hamburgers, pizza, side salads, and chicken sandwiches; for students ages 18–24 – french fries, hamburgers, pizza, Mexican foods, and chicken sandwiches; and for children under age 6 – french fries, chicken nuggets, pizza, hamburgers, and ice cream [52].
3.4 Physical Activity Similar to fruit and vegetable consumption, the majority of adults do not get the minimum recommended amount of moderate physical activity. Overall, only 48.9% report meeting recommendations, and only 50.9% of non-Hispanic white adults, 41.3% of non-Hispanic black adults, and 45.1% of Mexican Americans report meeting the recommended 30 min of moderate activity five or more days per week [24]. Data on youth are even more dismal: overall, only 34.7% of youth meet the recommended 60 min minimum on five or more days per week. Thirty seven percent of non-Hispanic white youth, 31.1% of non-Hispanic black youth, and 30.2% of Mexican American youth report meeting the minimum recommendation [46]. Additional issues related to body image and perception of weight status may play a role in individual behaviors regarding diet and physical activity factors. For example, studies have shown that compared to women of other race/ethnicity, African American women are more likely to accept a larger ideal body image [53–55], and also appear to be less likely than non-Hispanic white or Mexican American women
54
C. Doyle
to report they are trying to lose weight [56]. In addition, overweight and obese nonHispanic blacks, compared to their non-Hispanic whites, were disproportionately more likely to categorize themselves as being “about the right” weight [57]. While these differences exist, the extent to which these factors may influence eating and physical activity behaviors is not currently known.
3.5 Television Time Increasing time spent watching television has been associated with excess weight among youth, in part because of its sedentary nature, but also due to increased snacking that occurs while watching television and the amount of exposure to food- and beverage-related advertisements that are viewed. Among youth, it has been reported that non-Hispanic black youth and Mexican American youth spend significantly more time watching TV than do non-Hispanic white youth. In 2007, 27.2% of nonHispanic white youth, 62.7% of non-Hispanic black youth, and 43% of Hispanic youth report watching three or more hours of television per day [46]. Media use differs, as well, by socioeconomic status: low-income youth spend more time watching TV than higher income children. In addition, the lower the education of the parent, the more likely it is that a child will have a television in the bedroom and that the family will watch television during meals, both of which are associated with higher caloric intake [58].
3.6 Breastfeeding Breastfeeding is associated with a reduced risk of obesity in children and is recommended by CDC as a “promising approach” to prevent obesity. Breast feeding rates have been increasing for all population groups, although non-Hispanic black women are less likely to breast feed immediately post-birth and to still be breastfeeding at six months, compared to Mexican American and non-Hispanic white women [59].
3.7 Environmental Influences on Obesity American society has become what has been termed “obesogenic,” characterized by environments that promote increased food intake; less healthy, energy-dense foods; and physical inactivity. While many Americans would like to adopt a healthy lifestyle, substantial barriers exist that make it difficult for many to follow nutrition and physical activity recommendations. Indeed, current trends toward increasing portion sizes, both at home and while eating out [13, 60–62]; increased consumption of high-calorie convenience foods, sugar-sweetened beverages and meals outside the home; and declining levels of physical activity are all factors that have contributed to the obesity epidemic [29, 63–64]. In addition, more time spent working outside
2
Obesity and Cancer: Epidemiology in Racial/Ethnic Minorities
55
of the home and more households with multiple wage earners reduce the amount of time available for meal preparation, resulting in increased consumption of meals outside the home, which tend to be higher in calories and less nutritious than foods prepared at home [65]. Portion sizes, especially of high-energy dense foods, have increased over time, and the availability of these is extensively marketed by restaurants, supermarkets, and food and beverage companies [13, 60, 62]. Reductions in leisure time, increased dependence on automobiles for transportation, and increased availability of electronic media all contribute to reduced physical activity [63, 64]. Increasing evidence also indicates that the built environment has the potential to impact obesity and physical activity levels [66, 67]: limited access to sidewalks, parks, and recreation facilities is associated with greater risk of obesity [68], while neighborhoods that are designed to facilitate walking and safe physical recreation tend to have lower obesity rates [66]. While these environmental issues impact the entire population, recent studies have demonstrated that racial/ethnic minority and low income groups may be especially impacted by these issues. Less availability of and access to affordable healthy foods; the marketing and availability of energy –dense foods and beverages; safety concerns that may limit opportunities for physical activity; and other factors may help explain, in part, why obesity does not affect all population groups equally.
3.8 The Food Environment Numerous studies have reported on differences in the accessibility of supermarkets in neighborhoods made up of predominantly racial/ethnic minority and/or low-SES residents. Access to supermarkets increases access to healthy foods and has been associated with more healthful diets, greater consumption of fruits and vegetables, and lower rates of obesity [69, 70]. In one study, communities with a greater proportion of ethnic minority residents were found to have approximately 30% fewer supermarkets that carry high-quality fresh fruits and vegetables and affordable healthy foods such as whole grains, low-fat dairy products, and meats. In another, African American and Hispanic neighborhoods had fewer chain supermarkets compared with white and non-Hispanic neighborhoods by about 50 and 70%, respectively [71]. Yet another study found that the poorest neighborhoods in Detroit with a high percentage of African American residents were further away from the closest supermarket than neighborhoods that were not as poor and with a lower percentage of African American residents [72]. Limited access to supermarkets frequently results in residents shopping for food at local convenience stores, where healthy food options tend to be of lesser quality and more expensive [73]. For example, one study reported that although low-fat milk was available in the majority of the smaller grocery stores in areas whose residents were predominantly Hispanics and those of low-SES status, some stores charged more for low-fat milk than for regular milk [74]. Evidence also suggests that higher prices for healthier foods have an effect on children’s weight: a recent study of elementary school children concluded
56
C. Doyle
that lower gains in BMI between kindergarten and third grade were seen among children living in areas with lower prices of fruits and vegetables. These effects were larger for low-SES children, as well as Asian and Hispanic children [75]. Coupled with limited access to supermarkets is potentially increased access to fast-food restaurants in neighborhoods with predominantly black and low-SES residents. Block and colleagues [50] found that neighborhoods where 80% of the residents were African American had 2.4 fast-food restaurants per square mile compared to neighborhoods where 80% of the residents were non-Hispanic white that had only 1.5 fast-food restaurants per square mile. The implication of this is that there were six more fast-food restaurants in an average-sized shopping area for the predominantly African American versus predominantly white neighborhoods. Another study conducted in California examining the availability of types of restaurants found that compared with restaurants in more affluent areas with fewer African Americans, restaurants in less affluent neighborhoods with more African American residents were more likely to be fast food and/or fast casual, and less likely to offer healthier options [76]. The availability of fast-food restaurants is an important consideration within discussions regarding the obesity epidemic, as higher consumption of fast food is associated with higher caloric intake, higher saturated fat intake, lower consumption of fruits and vegetables, and possibly obesity. Another important food-related factor impacting socioeconomically disadvantaged communities is the relative costs of low-calorie versus high-calorie foods. Calorie for calorie, refined grains, added sugars, and fats are relatively inexpensive, while more nutrient-dense foods such as fruits, vegetables, and whole grains tend to cost more [77], and the price disparity between the low-nutrient, high-calorie foods, and healthier food options continues to grow. While fats and sweets cost only 30% more than 20 years ago, the cost of fresh produce has increased more than 100%. More recent studies in Seattle supermarkets showed that foods with the lowest energy density (mostly fresh vegetables and fruit) increased in price by almost 20% over 2 years, whereas the price of energy-dense foods high in sugar and fat remained constant [78]. Therefore, even in neighborhoods where supermarkets are available, low-income residents may purchase a relatively higher calorie diet of less expensive, higher calorie foods, and indeed, studies have suggested that lower cost foods make up a greater proportion of the diet of lower income individuals [79]. In US Department of Agriculture (USDA) studies, female recipients of food assistance had more energy-dense diets, consumed fewer vegetables and fruit, and were more likely to be obese. In addition to increased access to fast food and less expensive, energy dense foods, racial/ethnic minorities may be exposed to more advertisements for lownutrient foods, due to both targeted marketing as well as higher rates of television viewing. Tirodkar and colleagues found that more food commercials are aired during black prime time than general prime time (4.78 per 30 min program vs. 2.89 per 30 min program on general prime time). The researchers also found that 30% of the food commercials featured candy and 13% featured soda, significantly more than on general prime time [80].
2
Obesity and Cancer: Epidemiology in Racial/Ethnic Minorities
57
Because of their heavy media use, ethnic minority and low-income youth are exposed to a great deal of food advertising at home, and research has found that such advertising can affect children’s food preferences even after a brief exposure [81]. Findings related to time spent watching television are important to consider within the context of addressing obesity trends: many more studies have confirmed that television-watching is associated with obesity, in part because of its sedentary nature but also because of the advertisements that are viewed and related snacking while watching [82].
3.9 The Physical Activity Environment In addition to disparities in the access to and availability of healthy foods among racial/ethnic minorities and low-SES populations, a few studies suggest that disparities also exist in the built environment, which is likely to contribute to differences in physical activity among population groups, particularly among low-income populations. Access to parks, gyms, and other opportunities for physical activity – such as the availability of sidewalks and the close proximity of residential areas to stores, jobs, schools, and recreation centers – have been shown to contribute to more physically active lifestyles [83–84]. However, Powell and colleagues studied 409 communities and found significantly fewer sports areas, parks, greenways, and bike paths in high-poverty areas when compared to areas with lower poverty rates [85]. Additionally, even when these facilities are available, cost factors, distance from exercise facilities, and transportation availability may still affect access among lowincome populations [86–87]. Heavy traffic, lack of street lighting, unleashed dogs, high crime rates, and lack of sidewalks and traffic calming measures are other factors that may present barriers to physical activity, particularly in low-income areas. A recent systematic review of the built environment and health behaviors among African Americans found that associations between the built environment and physical activity among African Americans were inconsistent [88]. In some but not all studies, light traffic and the presence of sidewalks were significantly associated with higher levels of physical activity among those living in both metropolitan and nonmetropolitan areas. Similarly, safety from crime was associated with higher activity levels among urban participants in some but not all studies. A variety of studies have been conducted to examine the built environments’ impact on physical activity among youth [89]. In low-income urban communities, the built environment appears to have a larger impact on children’s physical activity than that of adults. Many adults in these communities must rely on public transportation to get to and from work, for shopping, and for other activities, and must spend time being physically active to access such transportation. For safety reasons, however, parents may restrict their children’s outdoor activities, particularly when no adult is home. Indeed, a recent study suggests that social factors – such as increased social contact and the availability of a network of parents who know each other and can watch out for each other’s children – may have more of an impact on
58
C. Doyle
children’s activity levels than factors associated with the physical environment [90]. In low-income communities, it is also possible that family work schedules, discretionary time, money, and transportation challenges may make it difficult for parents and caregivers to transport children to sports and other recreational activities.
4 Improving Public Health and Reducing Disparities in Obesity – What Will It Take? Addressing the extraordinary increase in obesity among all population groups – and especially among those of racial/ethnic minority groups and of low SES – will require a broad range of strategies that include policy and environmental changes that improve the access of and availability of healthy, affordable, high-quality foods and opportunities for safe, enjoyable physical activity, as well as strategies that empower individuals with the knowledge and skills they need to make healthy food and physical activity choices. While many factors have been identified as contributing to the obesity trends that the United States has experienced over the last three decades, effectively addressing racial/ethnic and socioeconomic disparities in obesity will require in-depth knowledge of how these factors affect these populations disproportionately; how the interplay of social, economic, and cultural considerations specific to particular minority groups further impact these factors; and ultimately, as with any population group, identifying those combination of strategies that will be most effective in removing barriers to making healthy food and activity choices. Much more research, however, is needed to help guide action to reduce obesity levels and address disparities among minority populations. From identifying the impact that different home environments have on obesity rates, to understanding the impact that food advertising and marketing have on particular racial/ethnic minority groups, to quantifying neighborhood characteristics which will be most successful in facilitating physically active lifestyles among both youth and adults, many questions remain to be answered. In the meantime, however, lessons learned from other public health movements can provide guidance to public health and other professionals working to improve the nutritional and physical activity environments of racial/ethnic communities. Indeed, the tobacco epidemic exemplifies the importance of policy and environmental changes in positively influencing health behaviors. Adult per-capita cigarette consumption increased steeply from 1910 until 1964, when the first US Surgeon General Report publicized the health hazards of smoking. However, efforts that focused primarily on public education produced only a gradual decrease in cigarette consumption from 1964 through the early 1980s [91]. It was the subsequent introduction of community-wide policy and environmental change approaches that produced much larger reductions in cigarette smoking among children and adults, beginning in the mid-1980s. These included restrictions on cigarette advertising, increases in the price of tobacco products through taxation, laws preventing exposure to secondhand smoke in public places, and restrictions on the access of children
2
Obesity and Cancer: Epidemiology in Racial/Ethnic Minorities
59
to tobacco products. Only recently have communities begun to strategically consider policy and environmental approaches that might promote improved nutrition and physical activity at the population level, and it is likely that among all population groups, it is these types of changes that will be instrumental in reducing obesity rates. Government agencies, industries, non-profit organizations, schools, communities, the media, public health professionals and other professionals, and individuals all need to play an active role in the growing national, state, and local efforts to combat the obesity epidemic. Health care professionals and community leaders, in particular, have new opportunities to provide leadership in promoting policy and environmental changes in their communities that will help reduce obesity levels and the racial, ethnic, and socio-economic disparities in obesity seen today.
References 1. American Institute for Cancer Research (2007). Food, Nutrition and the Prevention of Cancer: a Global Perspective, World Cancer Research Fund/American Institute for Cancer Research. Washington, DC 2. Centers for Disease Control and Prevention (2007). National Health and Nutrition Examination Survey Public Use Data Files, 2005–2006 2007. Atlanta, GA: National Center for Health Statistics 3. Ogden CL, Carroll MD, McDowell MA, Flegal KM (2007). Obesity Among Adults in the United States – NO Statistically Significant Change Since 2003–2004. NCHS data brief no 1. Hyattsville, MD: National Center for Health Statistics 4. Mullis RM, Blair SN, Aronne LJ et al. (2004). Prevention Conference VII: Obesity, a worldwide epidemic related to heart disease and stroke: Group IV: prevention/treatment. Circulation. 110(18):e484–88 5. Adams PF, Schoenborn CA (2006). Health behaviors of adults: United States, 2002–2004. Vital Health Stat, (230):1–140 6. Goel MS, McCarthy EP, Phillips RS, Wee CC (2004). Obesity among US immigrant subgroups by duration of residence. Jama, 292(23):2860–867 7. Kaplan MS, Huguet N, Newsom JT, McFarland BH (2004). The association between length of residence and obesity among Hispanic immigrants. Am J Prev Med, 27(4):323–26 8. Himmelgreen DA, Perez-Escamilla R, Martinez D et al. (2004). The longer you stay, the bigger you get: length of time and language use in the U.S. are associated with obesity in Puerto Rican women. Am J Phys Anthropol, 125(1):90–96 9. Wang Y, Beydoun MA (2007). The obesity epidemic in the United States–gender, age, socioeconomic, racial/ethnic, and geographic characteristics: a systematic review and metaregression analysis. Epidemiol Rev, 29:6–28 10. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM (2006). Prevalence of overweight and obesity in the United States, 1999–2004. Jama, 295(13):1549–1555 11. Baskin ML, Ard J, Franklin F, Allison DB (2005). Prevalence of obesity in the United States. Obes Rev, 6(1):5–7 12. Hedley AA, Ogden CL, Johnson CL, Carroll MD, Curtin LR, Flegal KM (2004). Prevalence of overweight and obesity among US children, adolescents, and adults, 1999–2002. Jama, 291(23):2847–850 13. US Department of Health and Human Services: The Surgeon General’s call to action to prevent and decrease overweight and obesity. Washington, DC US Department of Health and Human Services, Public Health Service, Office of the Surgeon General
60
C. Doyle
14. Freedman DS, Khan LK, Serdula MK, Ogden CL, Dietz WH (2006). Racial and ethnic differences in secular trends for childhood BMI, weight, and height. Obesity (Silver Spring), 14(2):301–08 15. Biro FM, McMahon RP, Striegel-Moore R et al. (2001). Impact of timing of pubertal maturation on growth in black and white female adolescents: The National Heart, Lung, and Blood Institute Growth and Health Study. J Pediatr, 138(5):636–43 16. Adair LS, Gordon-Larsen P (2001). Maturational timing and overweight prevalence in US adolescent girls. Am J Public Health, 91(4):642–44 17. Zhang Q, Wang Y (2004). Trends in the association between obesity and socioeconomic status in U.S. adults: 1971–2000. Obes Res, 12(10):1622–632 18. Robert SA, Reither EN (2004). A multilevel analysis of race, community disadvantage, and body mass index among adults in the US. Soc Sci Med, 59(12):2421–434 19. Gordon-Larsen P, Adair LS, Popkin BM. The relationship of ethnicity, socioeconomic factors, and overweight in US adolescents. Obes Res, 11(1):121–29 20. Miech RA, Kumanyika SK, Stettler N, Link BG, Phelan JC, Chang VW (2006). Trends in the association of poverty with overweight among US adolescents, 1971–2004. Jama, 295(20):2385–393 21. Wang Y, Zhang Q (2006). Are American children and adolescents of low socioeconomic status at increased risk of obesity? Changes in the association between overweight and family income between 1971 and 2002. Am J Clin Nutr, 84(4):707–16 22. Drewnowski A, Rehm CD, Solet D (2007). Disparities in obesity rates: analysis by ZIP code area. Soc Sci Med, 65(12):2458–463 23. Levi J, Vinter S, St. Laurent R, Segal L (2008). F as in Fat: How Obesity Policies are Failing in America 2008, Trust for America’s Health, Robert Wood Johnson Foundation, Washington, DC 24. Centers for Disease Control and Prevention (CDC) (2008). Behavioral Risk Factor Surveillance System Survey Data 2007. Department of Health and Human Services, Centers for Disease Control and Prevention Atlanta, Georgia: U.S. ; 2008. http://apps.nccd.cdc.gov/brfss/index.asp. Accessed August 2008 25. Ramsey PW, Glenn LL (2002). Obesity and health status in rural, urban, and suburban southern women. South Med J, 95(7):666–71 26. Patterson PD, Moore CG, Probst JC, Shinogle JA (2004). Obesity and physical inactivity in rural America. J Rural Health, 20(2):151–59 27. Jackson JE, Doescher MP, Jerant AF, Hart LG (2005). A national study of obesity prevalence and trends by type of rural county. J Rural Health, 21(2):140–48. 28. Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ (2003). Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med, 348(17):1625–638 29. Vainio H, Bianchini F (2002). Weight Control and Physical Activity. Vol 6. Lyon, France: International Agency for Research on Cancer 30. Caan BJ, Kwan ML, Hartzell G et al. (2008). Pre-diagnosis body mass index, post-diagnosis weight change, and prognosis among women with early stage breast cancer. Cancer Causes Control, 19(10):1319–328 31. Dignam JJ, Polite BN, Yothers G et al. (2006). Body mass index and outcomes in patients who receive adjuvant chemotherapy for colon cancer. J Natl Cancer Inst, 98(22): 1647–654 32. Freedland SJ, Grubb KA, Yiu SK et al. (2005). Obesity and risk of biochemical progression following radical prostatectomy at a tertiary care referral center. J Urol, 174(3):919–22 33. Meyerhardt JA, Niedzwiecki D, Hollis D et al. (2007). Association of dietary patterns with cancer recurrence and survival in patients with stage III colon cancer. Jama, 298(7):754–64 34. Chlebowski RT, Blackburn GL, Thomson CA et al. (2006). Dietary fat reduction and breast cancer outcome: interim efficacy results from the Women’s Intervention Nutrition Study. J Natl Cancer Inst, 98(24):1767–776
2
Obesity and Cancer: Epidemiology in Racial/Ethnic Minorities
61
35. Holmes MD, Chen WY, Feskanich D, Kroenke CH, Colditz GA (2005). Physical activity and survival after breast cancer diagnosis. Jama, 293(20):2479–86 36. Sedjo RL, Byers T, Barrera E, Jr. et al. (2007). A midpoint assessment of the American Cancer Society challenge goal to decrease cancer incidence by 25% between 1992 and 2015. CA Cancer J Clin, 57(6):326–40 37. McTiernan A (2000). Associations between energy balance and body mass index and risk of breast carcinoma in women from diverse racial and ethnic backgrounds in the U.S. Cancer, 88(5 Suppl):1248–55. 38. Mayberry RM, Stoddard-Wright C (1992). Breast cancer risk factors among black women and white women: similarities and differences. Am J Epidemiol, 136(12):1445–56 39. Adams-Campbell LL KK, Dunston G et al. (1996). The relationship of body mass index to reproductive factors in pre– and postmenopausal African American women with and without breast cancer. Obesity Res, 4(5):451–56 40. Cui Y, Whiteman MK, Langenberg P et al. (2002). Can obesity explain the racial difference in stage of breast cancer at diagnosis between black and white women? J Womens Health Gend Based Med, 11(6):527–36 41. Wenten M, Gilliland FD, Baumgartner K, Samet JM (2002). Associations of weight, weight change, and body mass with breast cancer risk in Hispanic and non-Hispanic white women. Ann Epidemiol, 12(6):435–34 42. Wee CC, McCarthy EP, Davis RB, Phillips RS (2004). Obesity and breast cancer screening. J Gen Intern Med, 19(4):324–31 43. Vizcaino AP, Moreno V, Lambert R, Parkin DM (2002). Time trends incidence of both major histologic types of esophageal carcinomas in selected countries, 1973–1995. Int J Cancer, 99(6):860–68. 44. Hollingsworth JM, Miller DC, Daignault S, Hollenbeck BK (2006). Rising incidence of small renal masses: a need to reassess treatment effect. J Natl Cancer Inst, 98(18):1331–34 45. Brown LM, Devesa SS (2002). Epidemiologic trends in esophageal and gastric cancer in the United States. Surg Oncol Clin N Am, 11(2):235–56 46. Eaton DK, Kann L, Kinchen S et al. (2008). Youth risk behavior surveillance–United States, 2007. MMWR Surveill Summ, 57(4):1–131 47. Taveras EM, Berkey CS, Rifas-Shiman SL et al. (2005). Association of consumption of fried food away from home with body mass index and diet quality in older children and adolescents. Pediatrics, 116(4):e518–24 48. Schmidt M, Affenito SG, Striegel-Moore R et al. (2005). Fast-food intake and diet quality in black and white girls: the National Heart, Lung, and Blood Institute Growth and Health Study. Arch Pediatr Adolesc Med, 159(7):626–31 49. Bowman SA, Vinyard BT (2004). Fast food consumption of U.S. adults: impact on energy and nutrient intakes and overweight status. J Am Coll Nutr, 23(2):163–68. 50. Block JP, Scribner RA, DeSalvo KB (2004). Fast food, race/ethnicity, and income: a geographic analysis. Am J Prev Med, 27(3):211–17 51. Stewart H, Blisard N, Bhuyan S, Nayga R (2004). The Demand for Food Away from Home: Full-Service or Fast Food? Agricultural Economic Report #829. USDA ,Washington, DC 52. NPD Group (2005). The NPD Group, 20th Annual Eating Patterns in America Study. Port Washington, NYNPD Group 53. Sanchez-Johnsen LA, Fitzgibbon ML, Martinovich Z, Stolley MR, Dyer AR, Van Horn L (2004). Ethnic Differences in correlates of obesity between Latin-American and black Women. Obes Res, 12(4):652–60 54. Katz ML, Gordon-Larsen P, Bentley ME, Kelsey K, Shields K, Ammerman A (2004). “Does skinny mean healthy?” Perceived ideal, current, and healthy body sizes among African-American girls and their female caregivers. Ethn Dis Autumn,14(4): 533–41 55. Gipson GW, Reese S, Vieweg WV et al. (2005). Body image and attitude toward obesity in an historically black university. J Natl Med Assoc, 97(2):225–36
62
C. Doyle
56. Bish CL, Blanck HM, Serdula MK, Marcus M, Kohl HW3rd, Khan LK (2005). Diet and physical activity behaviors among Americans trying to lose weight: 2000 Behavioral Risk Factor Surveillance System. Obes Res, 13(3):596–607 57. Yancey A, Kumanyika S (2007). Bridging the Gap: Understanding the Structure of Social Inequities in Childhood Obesity. Am J Prev Med, 33(4, Supplement 1):S172–S174 58. The Henry J. Kaiser Family Foundation (2004) The Role of Media in Childhood Obesity. Menlo Park, CA 59. National Center for Health Statistics (2004). Health, United States, 2004 With Chartbook on Trends in the Health of Americans Hyattsville, MD 60. Smiciklas-Wright H, Mitchell DC, Mickle SJ, Goldman JD, Cook A (2003). Foods commonly eaten in the United States, 1989–1991 and 1994–1996: are portion sizes changing? J Am Diet Assoc, 103(1):41–47 61. Young LR, Nestle M (2002). The contribution of expanding portion sizes to the US obesity epidemic. Am J Public Health, 92(2):246–49 62. Nestle M (2003). Increasing portion sizes in American diets: more calories, more obesity. J Am Diet Assoc, 103(1):39–40 63. Dong L, Block G, Mandel S (2004). Activities Contributing to Total Energy Expenditure in the United States: Results from the NHAPS Study. Int J Behav Nutr Phys Act, 1(1):4 64. Kruger J, Galuska DA, Serdula MK, Kohl HW, 3rd (2005). Physical activity profiles of U.S. adults trying to lose weight: NHIS 1998. Med Sci Sports Exerc, 37(3):364–68 65. Paeratakul S, Ferdinand DP, Champagne CM, Ryan DH, Bray GA (2003). Fast-food consumption among US adults and children: dietary and nutrient intake profile. J Am Diet Assoc, 103(10):1332–38 66. Saelens BE, Sallis JF, Black JB, Chen D (2003). Neighborhood-based differences in physical activity: an environment scale evaluation. Am J Public Health, 93(9):1552–58 67. Frank LD, Andresen MA, Schmid TL (2004). Obesity relationships with community design, physical activity, and time spent in cars. Am J Prev Med, 27(2):87–96 68. Giles-Corti B, Macintyre S, Clarkson JP, Pikora T, Donovan RJ (2003). Environmental and lifestyle factors associated with overweight and obesity in Perth, Australia. Am J Health Promot, 18(1):93–102 69. Morland K, Diez Roux AV, Wing S (2006). Supermarkets, other food stores, and obesity: the atherosclerosis risk in communities study. Am J Prev Med, 30(4):333–39 70. Morland K, Wing S, Diez Roux A, Poole C (2002). Neighborhood characteristics associated with the location of food stores and food service places. Am J Prev Med, 22(1):23–29 71. Powell LM, Slater S, Mirtcheva D, Bao Y, Chaloupka FJ (2007). Food store availability and neighborhood characteristics in the United States. Prev Med, 44(3):189–195 72. Zenk SN, Schulz AJ, Israel BA, James SA, Bao S, Wilson ML (2005). Neighborhood racial composition, neighborhood poverty, and the spatial accessibility of supermarkets in metropolitan Detroit. Am J Public Health, 95(4):660–67 73. Horowitz CR, Colson KA, Hebert PL, Lancaster K (2004). Barriers to buying healthy foods for people with diabetes: evidence of environmental disparities. Am J Public Health, 94(9):1549–54 74. Wechsler H, Basch CE, Zybert P, Lantigua R, Shea S (1995). The availability of low-fat milk in an inner-city Latino community: implications for nutrition education. Am J Public Health, 85(12):1690–92 75. Sturm R, Datar A (2005). Body mass index in elementary school children, metropolitan area food prices and food outlet density. Public Health, 119(12):1059–68 76. Lewis LB, Sloane DC, Nascimento LM et al. (2005). African Americans’ access to healthy food options in South Los Angeles restaurants. Am J Public Health, 95(4):668–73 77. Andrieu E, Darmon N, Drewnowski A (2006). Low-cost diets: more energy, fewer nutrients. Eur J Clin Nutr, 60(3):434–36 78. Monsivais P, Drewnowski A (2007). The rising cost of low-energy-density foods. J Am Diet Assoc, 107(12):2071–76
2
Obesity and Cancer: Epidemiology in Racial/Ethnic Minorities
63
79. Darmon N, Drewnowski A (2008). Does social class predict diet quality? Am J Clin Nutr, 87(5):1107–17 80. Tirodkar MA, Jain A (2003). Food messages on African American television shows. Am J Public Health, 93(3):439–41 81. Borzekowski DL, Robinson TN (2001). The 30-second effect: an experiment revealing the impact of television commercials on food preferences of preschoolers. J Am Diet Assoc, 101(1):42–46 82. A Kaiser Family Foundation Report (2007). Food for Thought: Television Food Advertising to Children in the United States. Menlo Park, CA 83. Duncan MJ, Spence JC, Mummery WK (2005). Perceived environment and physical activity: a meta-analysis of selected environmental characteristics. Int J Behav Nutr Phys Act, 2:11 84. King WC, Belle SH, Brach JS, Simkin-Silverman LR, Soska T, Kriska AM (2005). Objective measures of neighborhood environment and physical activity in older women. Am J Prev Med, 28(5):461–69 85. Powell L, S S, FJ C (2004). The relationship between community physical activity settings and race, ethnicity and socioeconomic status. Evidence-Based Prev Med, 1:135–144 86. Gordon-Larsen P, Nelson MC, Page P, Popkin BM (2006). Inequality in the built environment underlies key health disparities in physical activity and obesity. Pediatrics, 117(2):417–24 87. Estabrooks PA, Lee RE, Gyurcsik NC (2003). Resources for physical activity participation: does availability and accessibility differ by neighborhood socioeconomic status? Ann Behav Med, 25(2):100–104 88. Casagrande SS, Whitt-Glover MC, Lancaster KJ, Odoms-Young AM, Gary TL (2009). Built environment and health behaviors among African Americans: a systematic review. Am J Prev Med, 36(2):174–81 89. Sallis JF, Glanz K (2004). The role of built environments in physical activity, eating, and obesity in childhood. Future Child, 16(1):89–108 90. Franzini L, Elliott MN, Cuccaro P et al. (2009). Influences of physical and social neighborhood environments on children’s physical activity and obesity. Am J Public Health, 99(2):271–78 91. Smoking and Health (1990). A National Status Report. A Report to Congress Center for Chronic Disease Prevention and Health Promotion. Office on Smoking and Health United States. Public Health Service. Office of the Surgeon General 1990–2002 (February 1990) 92. Ogden CL, Carroll MD, Flegal KM (2008). High body mass index for age among US children and adolescents, 2003–2006. JAMA, 299(20):2401–405
Chapter 3
Obesity and Cancer in Asia Wanghong Xu and Charles E. Matthews
Abstract Obesity has emerged as a major public health challenge for both developed and developing countries. Like more developed countries where obesity represents an increasingly important risk factor for many cancers [1, 2], numerous studies conducted in Asian countries suggest that long-term positive energy balance that leads to elevated adiposity also contributes to the development of many chronic diseases, including the burden associated with a number of different cancer types [3–12]. While the biological mechanisms linking elevated adiposity to cancer do not appear to be markedly different in these populations, there are several unique features of the obesity–cancer association in Asia that are important to understand in relation to cancer prevention and control. First, given the population size and age structure of the larger countries in Asia, the impact of these countries on the worldwide cancer burden is now substantial and in future years will only increase. Second, there are differences in the body size and composition of Asian as compared to Caucasian adults that translate to important differences in association between obesity and cancer, relative to findings from studies conducted among Caucasian adults in the West. In this chapter we will focus on several of these features through review of the epidemiologic evidence available from selected countries in East (e.g., China, Japan, Korea, and Singapore) and Southern Asia (e.g., India, Pakistan).
1 Changes in Cancer Epidemiology in Selected Asian Countries Cancer is a growing burden on global health and Asian countries contribute substantially to the overall burden. In 2007 it was estimated that there were 12.3 million new cancer cases worldwide and Asian countries accounted for nearly half (45.5%) of these cases [13]. Cancer is one of the leading causes of death in Japan [14] and C.E. Matthews (B) Division of Cancer Epidemiology and Genetics, National Cancer Institute, Rockville, MD 20892–7344, USA e-mail:
[email protected] N.A. Berger (ed.), Cancer and Energy Balance, Epidemiology and Overview, Energy Balance and Cancer 2, DOI 10.1007/978-1-4419-5515-9_3, C Springer Science+Business Media, LLC 2010
65
66
W. Xu and C.E. Matthews
Korea [15], and in China it is the second leading cause of death, inferior only to cardiovascular disease [16], and cancer is the fourth leading cause of death in rural India [17]. The population size and age structure of the larger Asian countries are important contributors to the cancer burden in the region. In 2006, Asia accounted for about 59% of the total 650 million people in the world, and China and India alone accounted for one-third of the global population. Accordingly, these two countries contribute the largest number of incident cancer cases to the worldwide cancer burden. While total population size is an important factor, reduced infant and childhood mortality, increasing life expectancy, and decreased fertility rates have resulted in a dramatic increase in both the number and the proportion of older individuals in many Asian countries. Between 2000 and 2050, the percentage of the Asian population over the age of 60 years is expected to more than double. In 2000, Japan, China, India, and Indonesia accounted for 72% of the elderly in the region and for 42% of the total number of older adults in the world. By 2050 it is estimated that people aged 60 years or older will account for 42% of the Japanese population, and 30, 22, and 21% of the populations of China, Indonesia, and India, respectively [18]. Therefore, even if Asian countries achieve success in lowering incidence of specific cancers through effective prevention, early detection, and active treatment, the actual number of cancer cases would still be expected to increase because of aging in these populations. Indeed, the global cancer burden is expected to grow to 27 million new cancer cases and 17.5 million cancer deaths by the year 2050, simply due to the growth and aging of the population [13]. At the same time, changes in age-specific and age-standardized cancer incidence rates are also occurring, mainly due to alterations in exposures to environmental risk factors, changing lifestyles and socioeconomic status, and the effects of early detection and treatment. Data from 13 cancer registries in Asia describe the changes in age-standardized rates in this region over the past three decades. While several major cancers linked to chronic infectious conditions such as liver, stomach, and cervical cancer are in decline, cancers related to obesity, for instance breast cancer, colorectal cancer, and endometrial cancer, are increasing rapidly [19–21]. Recent changes observed between 1970 and 2005 in Shanghai, the largest city in China, provide a useful description of these patterns. Among men, the decline in total cancer in this period was mainly attributable to decreasing rates for cancers of the esophagus, stomach, and liver [19]. In the early 1970s these three cancers accounted for 51% of all cases diagnosed in men, but since that time the proportion of cancers attributed to these sites decreased to 35% in 1993–1994 [19] and to 28% by 2004–2005 [22, 23]. Among women in Shanghai, the most marked change over this period was the 96% decline in the rates for invasive cervical cancer (from 26.7 to 1.1/100,000 women) [19, 23]. Also among women, substantial decreases were observed for esophageal, stomach, and liver cancers. In contrast the incidence for cancers of the colon, breast, corpus uteri, gallbladder, brain, and kidney increased substantially over the same time period [19]. Figure 3.1 describes age-specific increases in the incidence of proximal and distal colon cancer among Chinese women from Shanghai between 1972 and 1994 [24]. Rates were
3
Obesity and Cancer in Asia
67
Fig. 3.1 Incident colon cancer in women by site, year, and age group: Shanghai (1972–1994). Adapted from B.T. Ji et al. CEBP 7; 661, 1998
consistently increased within the period for women over the age of 35 years for tumors in the proximal and distal colon. While mortality from squamous cell esophageal cancer, a cancer associated with malnutrition, is generally in decline, particularly in certain countries such as China [25, 26] and Singapore [27], mortality from esophageal adenocarcinoma, which has been linked to obesity [28], appears to be rising [29]. A similar shift in gastric cancers has also been observed for non-cardia malignancies and cardia adenocarcinomas in some Asian countries [30]. While economic development in Asia has contributed to the many positive changes of cancer incidence in Asia, variation in development between and within countries has also contributed to considerable variation in the relations between obesity and cancer in these areas. That is, problems of both under- and overnutrition coexist in Asian populations and factors that influence cancer incidence (e.g., age structure and exposure to risk factors), detection and treatment (availability and use of medical practices), and even cancer registration (completeness of reporting) can be influenced by economic factors. Given the wide variation in these factors in many countries, regional differences in the types or burden of cancer should be expected, and it should be recognized that the health agencies in these countries must combat the twin challenges of too little and too much development. In the second half of the 20th century, most of the countries in Asia experienced rapid economic development, but there was considerable variation in the
68
W. Xu and C.E. Matthews
timing of the transitions. Japan experienced accelerated economic growth following World War II (e.g., from 1950 to 1970), while South Korea experienced a period of rapid growth following the end of the Korean War in the mid-1950s. China is currently experiencing a period of extreme economic development. Between 1990 and 2005 the gross domestic product per capita in China increased about four times (i.e., 1,625–6,012 yuan per capita) [18]. Economic development has generally improved the living conditions and food safety in the Chinese population, particularly among city dwellers. Widespread use of refrigeration, increased availability of fresh fruits and vegetables, and reduced salt- and nitrate-preserved food in China may account, at least in part, for the downward trend in cancers of the mouth, esophagus, stomach, and possibly liver. For example, improvements in food storage and transportation may have reduced exposure to the hepatic carcinogen aflatoxin [31], and hepatitis B vaccination is likely to have contributed to reduced mortality from liver cancer [32]. India has also undergone rapid social and economic transitions during the past decades [33, 34] and it has been suggested that noncommunicable diseases are an increasing contributor to premature deaths in Indian adults, particularly in the urban areas. As economic development in Asian countries has yielded reductions in the incidence of cancers linked to chronic infectious conditions and malnutrition, it is also beginning to lead to increases in cancers related to obesity. Along with economic development, urbanization has proceeded rapidly, resulting in a marked transition toward more Western dietary patterns and sedentary lifestyles. An increase in the prevalence of overweight and obesity has been one of the results [35]. However, unlike the more gradual transition that occurred in the United States and most European countries, the lifestyle transition in many Asian countries has been more rapid and appears to be contributing to increased incidence of many obesity-related cancers in many Asian countries [36].
2 Nutrition Transitions and Positive Energy Balance 2.1 Rebalancing the Equation: Changes in Intake and Expenditure Changes in dietary patterns and physical activity behaviors that typically follow economic development, and the subsequent rise of obesity and other chronic diseases, have been termed the “nutrition transition.” Dietary patterns typically change in composition and caloric density, and modernization of industry and increased personal wealth result in an increase in the prevalence of labor-saving devices and use of motorized transportation, both of which tend to result in reduced physical activity levels. Such transitions reflect the remarkable and continuing evolution of the ability of humans to optimize their capacity to obtain high levels of food energy on a consistent basis, but with only minimal physical effort [37]. China and India present vivid contemporary examples of the rapid changes that can occur during nutrition transitions. In China, between 1954 and 1979,
3
Obesity and Cancer in Asia
69
consumption of animal foods, mainly pork, poultry, eggs, and seafood, increased by 50–160% [38], and since 1979 this trend has continued [23]. Not surprisingly, the caloric density of the Chinese diet has increased because a greater proportion of dietary calories are consumed from fat, particularly among urban residents [39]. Similarly, in India supply of animal products has increased over the past several decades, although with wide differences between the socioeconomic strata and between urban and rural populations [33]. Data from Indian National Family Health Survey in 1998–1999 suggested that overnutrition was associated with overweight and obesity in middle- and upper class women, particularly those living in urban areas [40]. Unfortunately, as in many developing countries, in India this relationship also coexists with undernutrition whereby nearly 40% of rural women, the majority of whom are poor, are at an increased risk for being underweight. Two countries that modernized earlier than China and India – Japan and South Korea – have experienced a unique nutrition transition since the 1970s. While average consumption of fruits, dairy products, eggs, and meat has increased, the high consumption of vegetables, soy products, and fish has been maintained in Japan over the past 40 years [41]. Major dietary changes in South Korea include a large increase in the consumption of animal food products and a fall in total cereal intake. Uniquely, the amount and rate of increase in fat intake have remained low in South Korea [42]. Therefore, Japan and South Korea have a relatively low prevalence of obesity compared with other Asian countries with similar or much lower incomes. In terms of physical activity, nutrition transitions are characteristically associated with reductions in overall physical activity energy expenditure. The increased prevalence of labor- and time-saving devices at home and at work and access to motorized transportation all conspire to reduce daily requirements for physical activity. While many of these changes can increase the availability of leisure time, modernization and economic development also typically increase access to electronic media (i.e., television, DVDs, computers), which can quickly consume newly realized leisure time in sedentary – rather than physically active – behaviors. While high-quality data on overall physical activity levels over the last 50 years are not available in Asian countries, recent surveys shed some light on the magnitude of changes in activity levels that can occur with economic development and urbanization. In China, it has been estimated that the average weekly physical activity among adults fell by 32% between 1991 and 2006. Declines in occupational activity were the most pronounced, but reductions in household chores and walking and cycling for transportation also were substantial. For example, over this 15-year period, time spent in domestic chores was reduced from 15.0 to 5.8 h/wk (–61%) in men and from 28.0 to 11.9 hs/wk (–58%) in women. It is not known whether the time saved from household chores was re-allocated to other sedentary behaviors, but it was clear that these time savings were not re-allocated into leisure-time physical activity because the prevalence of exercise remained fairly consistent and low over time (<15%) [43]. Evidence from India, from a study that employed urban vs. rural comparisons as an indicator of the nutrition transition, suggested that physical activity is substantially lower among urban dwellers. Overall physical activity levels, including that from occupation, transportation, and leisure activities, were five to seven times lower
70
W. Xu and C.E. Matthews
among men and women in urban centers as compared to their peers in rural areas. These differences were driven by a lower level of occupational- and transportationrelated activity among adults in urban centers. Interestingly, women in rural areas were the most active and least sedentary group in this study [44].
2.2 Increasing Adiposity in Selected Countries Given the striking changes in the two behavioral parameters that influence the energy balance equation in developing Asian countries, it is not surprising that the prevalence of overweight and obesity is increasing in the segments of these countries that do not face nutritional scarcity [45]. In China, between 1982 and 1992, the prevalence of overweight and obesity in young adults (BMI > 25) increased from 9.7 to 14.9% in urban areas and from 6.2 to 8.4% in rural areas. The prevalence of overweight for school boys and school girls jumped from 3.4 and 2.8% in 1985 to 7.2 and 8.7% in 1995, respectively [46]. Data from the China Health and Nutrition Survey collected between 1989 and 2000 showed that the mean BMI increased from 21.3 to 22.4 kg/m2 for Chinese men and from 21.8 to 22.4 kg/m2 for Chinese women. During this period, the proportion of overweight (BMI ≥ 25) and obesity overweight (BMI ≥ 30) among men increased 21.9% (from 10.1 to 32.0%) in urban areas and increased 10.8% (from 4.7 to 15.5%) in rural areas in China. In Shanghai, the largest city in China, 35.2% of middle-aged and elderly women were overweight (BM ≥ 25) and 5.1% were obese (BMI ≥ 30) in the years 1997 to 2004 [47]. Among middle-aged men (40–74 years) from Shanghai during a similar time period, the prevalence of overweight and obesity was 30.4 and 2.5%, respectively [48]. In India, the increase of overweight and obesity in female adults was 5.0 percentage points between 1989 and 1994 [46]. The prevalence of obesity by using a BMI cutoff of 25.0 was 4.0 and 4.1% for rural men and women and 5.8 and 6.1% for those in urban areas, respectively [49]. However, the prevalence varied widely with socioeconomic status. In urban India, higher prevalence rates (32.2% among males, 50% among females) were observed in the upper strata than in the middle classes (16.2% males, 30.3% females), followed by the prevalence of 7.0% in males and 27.8% in females for the lower socioeconomic groups, and the lowest (1.0% males, 4.0% females) for poor urban areas [50]. The prevalence of overweight in Malaysia also reached 28.7% for males and 26.0% for females in 1990, higher than that in Japan at the same period [46]. Even in Japan and Korea, two developed countries with a relatively low prevalence of obesity compared with other Asian countries, the prevalence of overweight is also increasing. In Japan, the prevalence of overweight (BMI: 25–29.9 kg/m2 ) and obesity (BMI ≥ 30 kg/m2 ) in men increased from 14.5 and 0.8% in the time period 1976–1980 to 20.5 and 2.0% during 1991–1995, although the increasing trend of BMI during the period was observed only in elderly women (≥60 years) [51]. In 1995, the prevalence of overweight reached 24.5% for Japanese men and 21.4% for
3
Obesity and Cancer in Asia
71
women [46]. In Korea, the incidence of overweight and obesity increased to 29.6 and 4.0% in 2001 from 24.3 and 1.7% in 1998 among the male population and to 25.9 and 3.4% from 23.5 and 3.0% among females during the same period [52, 53].
3 Body Composition in Asian Adults In population-based studies simple anthropometric measures, such as body mass index (BMI, weight (kg) divided by height (m) squared) and circumference measures, are commonly used to estimate the level and distribution of body fat and evaluate health risks associated with either low or high levels of these exposures. Fat distribution, i.e., central versus peripheral fat distribution, is usually measured by waist and hip circumferences using waist circumference to estimate abdominal obesity and the waist-to-hip ratio (WHR) to estimate variation in upper vs. lower body obesity. Waist circumference is often a preferred measure of abdominal obesity compared to WHR [54]. Waist circumference is unrelated to height, correlates closely with BMI and WHR, and is a useful index of intra-abdominal fat mass. In some studies it has been reported to be a better indicator of intra-abdominal fat content than WHR [55–57]. There are a number of important differences in the body size and composition of Asian as compared to Caucasian adults. Differences in body build and body composition among Asian adults result in a different relationship between BMI and body fat relative to Caucasians, and fat distribution appears to differ as well [58, 59]. For example, in a recent study that compared the body composition of European (Caucasian), Maori, Pacific Island, and Asian adults across a wide age range, Asian Indians were observed to have more fat mass overall and more abdominal fat, but less lean mass (e.g., skeletal muscle and bone mass) than the other ethnic groups in this study [60]. Ethnic differences in the BMI–percentage body fat relationship have also been observed between Japanese and Australian men. Over the BMI range of 16–33 kg/m2 , Japanese men were observed to have a greater percentage body fat at any given BMI value than similarly aged Australian men. Accordingly, at an equivalent body fat percentage, the BMI levels of the Japanese men were about 1.5 kg/m2 lower than those of the Australians [61]. Similar relationships between BMI and body fat percentage have been reported for Chinese adults, but it should be noted that there is some variation between and within Asian countries on these relationships. Nevertheless, overall Asian populations tend to have a higher body fat percentage at a given BMI compared to Caucasians. Thus, the relationship between body fat and BMI among Asians can result in BMI levels that are up to 3–4 kg/m2 lower than Caucasian adults at the same body fat percentage [59]. Accordingly, the level of BMI that the World Health Organization (WHO) recommends for screening for overweight and obesity in Caucasians, typically defined as 25.0–29.9 kg/m2 and ≥30 kg/m2 , respectively, may not be applicable for Asian adults. Based on review of the morbidity data in Asians, the WHO Western Pacific Regional Office (WPRO), led by the International Association for the Study of
72
W. Xu and C.E. Matthews
Obesity and the International Obesity Task Force, proposed a new definition for obesity for use in the Asia-Pacific region. For Asian populations, overweight was defined as BMI 23.0–24.9 kg/m2 and obesity was defined as BMI ≥ 25 kg/m2 . Following this publication, a number of studies assessed whether the alternative cutoff points for Asians were useful for risk stratification. Findings in two cohort studies [12, 62] did not support the use of a lower BMI cutoff point for defining risk associated with overweight in Koreans for mortality and cancer outcomes. In contrast, results from an analysis of National Health Interview Survey data, a cohort study of 36,386 middle-aged Chinese in Taiwan, reported that significant mortality risks started at BMI ≥ 25.0 kg/m2 , supporting the use of BMI ≥ 25.0 kg/m2 as a new cutoff point for obesity and BMI = 23.0–24.9 kg/m2 for overweight [63]. Under the support of International Life Sciences Institute Focal Point, the Working Group on Obesity in China organized a meta-analysis on the relation between BMI and waist circumference on several metabolic risk factors and related chronic diseases (e.g., high blood glucose, lipoprotein disorders, and diabetes mellitus). Results supported the idea that the prevalence of these outcomes was elevated at relatively low BMI levels (e.g., 24 kg/m2 ) [64]. Also in this report, waist circumference beyond 85 cm for men and beyond 80 cm for women was associated with increased risk and thus was recommended as useful cutoff points for central obesity. The WHO has suggested that a waist circumference of 90 cm for men and 80 cm for women be used as interim lower values for Asians [65]. Corresponding waist circumference values for Caucasian adults are 94 cm in men and 80 cm in women [54]. In summary, Asian adults typically have higher levels of body fat for a given BMI, and consistent with the idea that elevated adiposity is associated with poor health, risk for obesity-related chronic diseases is often increased at lower BMI levels in Asian populations as compared to studies conducted among Caucasians. Thus, care should be taken when interpreting and comparing results from epidemiologic studies conducted in Asia to those from European countries. Additionally, the different thresholds indicative of overweight and obesity obviously will have an impact on prevalence estimates and therefore also would be expected to influence population-attributable risk estimates for these exposures.
4 Differences in the Association Between Adiposity and Cancer in Asia 4.1 Obesity and Cancer Risk The landmark study of Calle et al. provided considerable evidence that many different cancer types are associated with elevated adiposity, and many studies from Asia lend further support to these findings [2]. Among the developing countries in Asia, cancers with established links to obesity, such as postmenopausal breast, colon, and endometrial cancer, represent emerging problems for cancer prevention and control [1]. In contrast, for a number of cancer sites, such as esophageal, stomach,
3
Obesity and Cancer in Asia
73
and liver of which the burden is trending downward as a consequence of economic development and public health efforts, the rise of overweight and obesity may mitigate some gains for these cancers. While this is one unique dynamic for developing Asian countries, other differences in the association between obesity and cancer are associated with the different relationship between body fat and body mass index and a more pronounced influence of central adiposity on risk for Asian as compared to Caucasian adults. We will examine endometrial cancer and breast cancer to highlight these differences. In addition, we will highlight potential reasons for heterogeneity between and within the countries of Asia and briefly review selected literature for other cancers. As noted previously, compared to Caucasian adults, Asians tend to have smaller body frames and higher levels of overall body and abdominal fat at a given BMI level. Therefore, one of the primary differences among Asian populations in relation to the association between adiposity and cancer is the BMI level at which individuals may be at increased risk. Risk for endometrial cancer has consistently been found to be positively associated with elevated adiposity, and here we use it as an example to illustrate the impact of the different body fat–BMI relationships in Asian women at risk for this disease. In a large case–control study of endometrial cancer among Chinese women from urban Shanghai, we found that risk was increased among women with BMI levels of 21.5–23.7 kg/m2 (OR = 1.4 [95% CI = 1.0–1.9]) compared to women with BMI levels of less than 21.4 kg/m2 [57]. In Caucasian women this BMI level would be considered to be in the middle of the normal weight range (BMI 18.5–24.9 kg/m2 ) and would be considered to be low risk. Other studies from China have reported that BMI levels in this range were positively associated with endometrial cancer regardless of menopausal status [3, 66], but one Japanese study failed to observe an association with overall adiposity [67]. The accumulation of adipose in the abdominal region also appears to be more pronounced among Asian adults and therefore may be a particularly important indicator of risk for obesity-related cancers. Here, we maintain our focus on endometrial cancer to describe the influence of central adiposity and then we examine some of the interesting variability in the associations that have been observed between adiposity and premenopausal breast cancer in Asian women that may relate to the importance of central adiposity among Asian women. Earlier studies have reported that fat deposits on the trunk confer additional risk for endometrial cancer in Chinese women [3], and in our case–control study described earlier, Chinese women with greater waist circumferences were at increased risk for endometrial cancer, even if their BMI was low [57]. Figure 3.2 presents the joint associations between BMI and waist:hip ratio (WHR), and in this analysis low BMI and low WHR levels (quartiles) formed a common referent group. As expected, women with the highest levels for both of these indicators of adiposity were at highest risk (OR = 5.4). Interestingly, the joint classification of a lower BMI (<21.4 kg/m2 ) with a WHR above the median (>0.814) was associated with about a twofold increased risk, while women with the lowest WHR also appeared to be at increased risk if their BMI was above the median (>23.7 kg/m2 ; Fig. 3.2). The results suggest that both greater central adiposity and
74
W. Xu and C.E. Matthews
5.4
Odds Ratio*
6
4.6
3.8
3.5
5
2.4
3.0
4
2.2
2.0 3
1.9
2
1.1 2.1
1
1.0
1.1
1.6
0
< 21.4
2.7
2.2
21.5-23.7
23.8-26.2
Body mass index (kg/m 2)
> 26.2
> 0.855 0.815-0.855 0.783-0.814 < 0.782
Waist:Hip Ratio
Fig. 3.2 Association between endometrial cancer and body mass index and waist:hip ratio, Shanghai, China 1997–2001. Adapted from Xu et al. American Journal of Epidemiology 161: 939, 2005 ∗ Adjusted for age, education, year of menstruation, and number of pregnancies
elevated BMI levels are important risk factors for endometrial cancer and support the recommendation to include waist circumference measurements as a practical tool to assess abdominal fat, in conjunction with measures of overall adiposity as indicated by BMI in both clinical and research settings [57]. Breast cancer offers a second example, in the form of more heterogeneous study results among premenopausal women, of a potentially interesting difference in the association between adiposity and cancer among Asian women that may relate to the importance of central adiposity as a risk factor for certain cancers. In both Caucasian and Asian populations, the association between elevated adiposity and postmenopausal breast cancer risk has been consistently observed. For example, several studies have shown a significantly greater risk of breast cancer with higher level of BMI among postmenopausal Asian women [5, 7]. However, in contrast to numerous studies among Caucasian women – which indicate either a reduced risk for heavier women or no increase in risk – several studies among Asian women, but not all [68–70], suggest that elevated adiposity may be associated with increased risk for premenopausal breast cancer [71–73]. Gilani et al. observed a positive association between BMI and breast cancer in young Pakistani women (< 45 years). Women with a BMI ≥ 30 compared to women with values < 25 were at more than a five-fold greater risk, but the risk for overweight women was not statistically significant [71]. Similarly among urban and rural women in South India, Matthew et al. observed that premenopausal women with a BMI above 25 kg/m2 were at a 30–50% greater risk than their counterparts with BMIs below this level [72]. Chow et al. also reported an increased risk of 30–50% among premenopausal Chinese women with higher BMIs, although the results were not statistically significant [73].
3
Obesity and Cancer in Asia
75
The heterogeneity among these reports may be an artifact of the case–control study design which must often employ measures of body weight obtained after diagnosis and treatment have been completed. It is possible that weight gain associated with treatment [74] could bias risk estimates from these studies, although similar weight gain has not been observed in Asian women [75]. Alternatively, it may be that among premenopausal Asian women, the accumulation of abdominal body fat may play a more important role in breast cancer than overall obesity. In a prospective case–control study involving 1,086 Chinese women in Singapore, central obesity as indicated by women with a larger WHR was associated with highest risk for breast cancer, with OR being 9.18 (95% confidence interval, 4.8–17.5) comparing the last and first quintile, whereas BMI did not significantly predict risk for breast cancer [68]. In a prospective cohort study of 11,889 women conducted in Taiwan, central adiposity reflected by hip circumference was a significant predictor of breast cancer [76]. At least two studies in Asian women have reported that upper body fat accumulation as measured by WHR was positively associated with breast cancer in premenopausal women [69, 77], which is consistent with a recent meta-analysis of available case–control and cohort studies [78]. Future studies in Asian women, preferably from prospective cohorts, are needed to further clarify the potential for overall adiposity (BMI) and fat distribution to be positively associated with premenopausal breast cancer. While there are some differences in the association between obesity and cancer in Asian populations, there are also a number of similarities with Caucasian populations. We next briefly review the evidence linking obesity to colorectal, prostate, ovarian, pancreatic, gallbladder, esophageal, and gastric cancers in Asian adults. In terms of colorectal disease, several studies have reported an increased risk of colon cancer for Asian men with high BMIs [4, 10–12, 79], with an exception of a prospective study in Japanese men for which no positive association was observed for obesity and excessive weight gain with colon cancer death [80]. Among women, the associations observed in Asian studies have been inconsistent. The relationship between BMI and colorectal cancer risk in Asian women has been found to be stronger [7, 80], weaker [10, 81], or absent [11, 12], and as anticipated the associations varied with age and menopausal status [7, 10]. Higher BMI levels were associated with increased risk of colon cancer among premenopausal Chinese women, but elevated BMI levels appeared to reduce risk among postmenopausal women [10]. Abdominal obesity is also an important risk factor in colon cancer. Although few studies have yet directly compared waist circumference to colon cancer risk among Asians [82], several studies have linked abdominal obesity with an increased risk of colorectal adenoma, a precursor of colorectal cancer [83–85]. Evidence is more limited for the relationship between obesity and prostate or ovarian cancer in Asian populations. One case–control study conducted in China did not observe a significant association of prostate cancer with height, usual adult weight, or pre-adult and usual adult BMI. However, there was a positive association between abdominal adiposity and clinical prostate cancer [86]. Results suggest that abdominal fat should be carefully evaluated in future prostate cancer studies among Asian men. Similar to the situation in Western populations, it is unclear
76
W. Xu and C.E. Matthews
whether obesity influences ovarian cancer risk in Asian women. Some studies report an increased risk among obese women [87, 88], whereas others have found no association [89]. Conflicting findings also exist regarding the association between obesity and risk of pancreatic cancer in Asian populations. For example, a case–control study of pancreatic cancer in Shanghai, China, reported a positive relationship between BMI and this disease [90]. Similarly, a large prospective cohort study of Japanese men and women observed that BMI at baseline was associated with a non-significant increase in the risk of death from pancreatic cancer, but only in women. Among men, obesity at age 20 years, defined as a BMI of 30 or more, was associated with a 3.5fold increase in risk for pancreatic cancer death compared with men with a normal BMI [91]. In contrast, two Korean [9, 92] and one Japanese [7] cohort studies found no association. In a pooled analysis of 30 cohort studies including 22 from Asian countries, central obesity, as opposed to overall adiposity as indicated by BMI, was observed to be positively associated with pancreatic cancer. The age-adjusted hazard ratio for pancreatic cancer death was 1.76 (95% CI: 1.15–2.69) for a 2 cm increase in waist circumference [93]. Gallbladder cancer also has been found to be associated with obesity in Asians. Both overall and abdominal obesity, including obesity in early adulthood, have been associated with an increased risk of gallbladder cancer among Chinese adults [94]. In Japanese women, higher BMI levels were also significantly associated with increased risk [7]. Similarly, among Korean men, obesity (BMI ≥ 30 kg/m2 ) was associated with increased risk for this outcome [12]; however, another Korean study found no association [9]. As we noted previously, rates of esophageal, gastric, and liver cancer are trending downward in many Asian countries as food safety has improved and public health efforts to minimize other infectious exposures have been implemented. Nevertheless, the emergence of obesity as a possible risk factor for cancers at these sites has the potential to mitigate some gains in preventing these cancers. In the United States and other Western countries, overweight and obese adults are about two times more likely to develop esophageal adenocarcinoma and cancer of the gastric cardia than adults that maintain a healthy body weight [95, 96]. In contrast, in recent studies conducted in Asia, inverse associations between obesity and risk for squamous cell carcinoma of the esophagus have been observed [9, 97–99], associations between esophageal adenocarcinoma are absent, and associations with gastric cardia adenocarcinoma are not consistent [100, 101]. For example, a study conducted in Shanghai, China [100], reported a positive association between obesity and cancer of the gastric cardia, while Zhang et al. reported an inverse association between BMI and gastric cardia adenocarcinoma in a hospital-based study in Beijing, China [101]. Clearly, additional studies are needed to clarify these findings. Other cancers linked to obesity among Asians include kidney cancer [81, 102], liver cancer [12], papillary cancer in the thyroid, small-cell cancer in the lung, nonHodgkin’s lymphoma, melanoma [9], and leukemia [81]. However, results from these studies are far from conclusive and further studies are needed to evaluate these associations in Asian populations.
3
Obesity and Cancer in Asia
77
It should be noted that, similar to the findings in Western populations [103, 104], associations between obesity and cancer may differ among subsets of the population due to modifying effects of other risk factors [10, 57, 105]. Obesity itself can also function as a modifier of cancer risk, and this has been observed in Western [106, 107] and Asian populations [108, 109]. Given that distributions of risk factors for cancer can differ between and within populations, the risk of cancer associated with overweight or obesity might also differ because of modifications by these other factors. This may explain some of the variability in association observed in Asian studies. Furthermore, the changes in characteristics of populations have been suggested to have an influence on relative risk estimates for cancer with regards to obesity due to effect measure modification [110]. Therefore, changes in diet and lifestyles in Asia may not only result in increased prevalence of obesity in populations but these changes have the potential to also modify the relation between obesity and cancer risk.
4.2 Obesity and Cancer Survivorship Obesity also may be an indicator of prognosis among Asian cancer survivors, although the overall evidence is limited at present and elevated adiposity has been associated with both longer and shorter survival following a cancer diagnosis. Consistent with the result from a randomized trial of International Breast Cancer Study Group [111], a follow-up study of 1,455 breast cancer survivors aged 25–64 years in Shanghai, China, found that being overweight at cancer diagnosis or soon afterward, as measured by BMI, was associated with poorer overall survival and disease-free survival [112]. Interestingly, neither WHR nor waist circumference was independently associated with survival in this study [112]. A study among Japanese breast cancer survivors reported that obesity was associated with significantly shorter survival. The estimated survival probability for women with BMI of 20 at the time of operation was about 12% higher than that with BMI of 24 over 10 or more years of follow-up [113]. In a Korean study of survival following colorectal cancer, BMI and visceral adiposity were found to have no influence on overall colorectal cancer survival but overweight was observed to be inversely associated cumulative disease-free survival, and increased visceral adiposity was a significant predictor of poor disease-free survival in patients with resectable colorectal cancer [114]. As noted previously, the direction and magnitude of association between BMI and cancer survival among the studies conducted in Asia depend on the type or severity, or stage, of the cancer. In a Korean study including 14,578 men with cancer, higher BMI levels were associated with longer survival in head and neck (HR, 0.54; 95% CI, 0.39 to 0.74) and esophageal (HR, 0.44; 95% CI, 0.28–0.68) cancers, but not with other cancers [115]. In another study, mean survival was observed to be longer for individuals with stage 2 gastric cancers and lower BMI levels (1667 vs. 1322 days, P = 0.02), but survival was longer for cases with stage 3a disease
78
W. Xu and C.E. Matthews
and a higher BMI level (1431 vs. 943 days, P = 0.01) [116]. Favorable associations between high BMI on the prognosis of Japanese patients with renal cell carcinoma also have been reported [102]. It may be that there is a survival advantage to having greater energy reserves in the face of a diagnosis with later stage cancer.
5 Possible Biological Mechanisms The biologic mechanisms through which obesity influences the natural history of cancers are not completely understood in either Caucasian or Asian populations, but there are few reasons to anticipate major differences in these mechanisms between the two populations. A detailed review of the mechanisms linking obesity to cancer is beyond the scope of this chapter and detailed descriptions of these mechanisms may be found in other chapters of this book. In this section we briefly review selected mechanistic studies that may link obesity to cancer in Asian adults. One of the mechanisms that has been postulated to explain the obesity–cancer association emphasizes the role of systemic inflammation that can be associated with elevated adiposity [117, 118]. Several studies conducted among Asian adults lend support to this hypothesis. For example, in healthy Japanese men, those with abdominal obesity (waist circumference ≥ 85 cm) were observed to have higher levels of serum high sensitivity C-reactive protein and interleukin-6 and lower levels of adiponectin than men without abdominal obesity [119]. Evidence is also available for positive associations between inflammatory biomarkers and risk for cancers of the breast [120], endometrium [105], colon [121], and stomach [122]. Alterations in sex hormones (e.g., estrogen, progesterone, and androgens) associated with obesity may also explain the increased risk for hormone-sensitive cancers, such as breast, endometrial, and prostate cancers. Among postmenopausal women with elevated BMI levels, estrogen levels can be 50–100% higher than in lean women [123] and elevated estrogen levels are believed to account for the increased risk of postmenopausal breast cancer [124, 125]. For endometrial cancer, lifetime exposure to hormones and high levels of estrogen in obese women may be contributing factors [126]. Sex hormone-binding globulin, the major carrier protein for certain sex hormones in the plasma, also may play a role in the altered risk for these cancers in obese women [124–126]. The influence of sex hormone levels associated with elevated adiposity may differ for certain cancers. For example, while obesity has been consistently associated with increased risk of colon cancer in men [4, 9–12, 80], studies among women are more mixed, perhaps because of the protective effect of estrogens on colorectal cancer. In a study conducted in China, an elevated risk was observed in Chinese men, but among women, menopausal status was found to be a strong effect modifier of BMI–colon cancer association [10]. Results suggest that elevated estrogen levels associated with elevated adiposity in postmenopausal women may confer some protection against colon cancer among Chinese women. Studies of these relationships in the Chinese population may be
3
Obesity and Cancer in Asia
79
particularly insightful because historically use of postmenopausal hormones has been quite low (<5%) [47], and therefore the influence of the biology of endogenous, as opposed to exogenous, hormonal exposures on risk for colon cancer risk may be more readily evaluated. Hyperinsulinemia associated with insulin resistance also is believed to play an important role in the development of certain cancers. Insulin is known to stimulate cell proliferation and thus insulin may serve as a simple growth factor in the process of cancer development. Insulin resistance has been associated with cancers of the colon and rectum [83], breast [127–129], endometrium [105], pancreas [130], and prostate [131] in Asian men. In Chinese women, BMI and WHR have been significantly positively correlated with C-peptide levels, an indicator of insulin secretion [129], and elevated C-peptide levels have been associated with increased risk for breast cancer [132]. Hyperinsulinemia is thought to have a permissive influence on the bioavailability of sex hormones and insulin-like growth factors through its negative influence on sex hormone-binding globulins (SHBG) and certain IGF-binding proteins [126, 133]. The mechanisms by which obesity increases risk of adenocarcinoma of the esophagus, cancer of the gastric cardia, and liver cancer are also not well understood. It has been proposed that obesity influences risk for adenocarcinoma of the esophagus by increasing risk for gastroesophageal reflux disease, which can cause tissue damage, Barrett’s esophagus, and ultimately invasive disease [134, 135]. However, one of the few studies conducted in Asia did not find a direct relationship between gastroesophageal reflux disease and the occurrence of gastric cardia cancer [136]. In contrast, recent reports among Caucasians have highlighted a potential role for abdominal obesity [135], as well as alternate hormonal mechanisms that may be associated with these outcomes [137, 138]. Hepatocellular carcinoma accounts for a large proportion of all liver cancers, and liver damage from infection is thought to be an early event in the development of these cancers. Increasingly, non-alcoholic fatty liver disease, which is linked to obesity, also appears to contribute to liver damage, cirrhosis, and to invasive cancers [139, 140]. It may be that metabolic abnormalities associated with obesity may increase risk for these cancers.
6 Summary and Conclusions In summary, the consequences of obesity relevant to cancer risk are manifest among Asians whose body size and composition, and cancer incidence, are considerably different from the West. The rapid shift in diet and lifestyles due to economic development has increased the prevalence of obesity in Asian populations and may ultimately influence the type and distribution of cancer in this region. The increased prevalence of overweight and obesity and the marked upward trends in incidence of several obesity-linked cancers in Asian countries indicate that obesity prevention strategies are needed to prevent cancer even in Asian populations that are at the lower end of the worldwide BMI distribution.
80
W. Xu and C.E. Matthews
References 1. International Agency for Research on Cancer. (2002). Weight Control and Physical Activity. Lyon: International Agency for Research on Cancer. 2. Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ (2003). Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med, 348:1625–1638. 3. Shu XO, Brinton LA, Zheng W, Swanson CA, Hatch MC, Gao YT et al. (1992). Relation of obesity and body fat distribution to endometrial cancer in Shanghai, China. Cancer Res, 52:3865–3870. 4. Kono S, Handa K, Hayabuchi H, Kiyohara C, Inoue H, Marugame T et al. (1999). Obesity, weight gain and risk of colon adenomas in Japanese men. Jpn J Cancer Res, 90: 805–811. 5. Tung HT, Tsukuma H, Tanaka H, Kinoshita N, Koyama Y, Ajiki W et al. (1999). Risk factors for breast cancer in Japan, with special attention to anthropometric measurements and reproductive history. Jpn J Clin Onco, 29:137–146. 6. Shu XO, Jin F, Dai Q, Shi JR, Potter JD, Brinton LA et al. (2001). Association of body size and fat distribution with risk of breast cancer among Chinese women. Int J Cancer, 94:449–455. 7. Kuriyama S, Tsubono Y, Hozawa A, Shimazu T, Suzuki Y, Koizumi Y et al. (2005). Obesity and risk of cancer in Japan. Int J Cancer, 113:148–157. 8. Xu WH, Matthews CE, Xiang YB, Zheng W, Ruan ZX, Cheng JR et al. (2005). Effect of adiposity and fat distribution on endometrial cancer risk in Shanghai women. Am J Epidemiol, 161:939–47. 9. Oh SW, Yoon YS, Shin SA (2005). Effects of excess weight on cancer incidences depending on cancer sites and histologic findings among men: Korea National Health Insurance Corporation Study. J Clin Oncol, 23:4742–4754. 10. Hou L, Ji BT, Blair A, Dai Q, Gao YT, Potter JD et al. (2006). Body mass index and colon cancer risk in Chinese people: Menopause as an effect modifier. European J Cancer, 42: 84–90. 11. Isomura K, Kono S, Moore MA, Toyomura K, Nagano J, Mizoue T et al. (2006). Physical activity and colorectal cancer: the Fukuoka colorectal cancer study. Cancer Sci, 97: 1099–1104. 12. Jee SH, Yun JE, Park EJ, Cho ER, Park IS, Sull JW et al. (2008). Body mass index and cancer risk in Korean men and women. Int J Cancer, 123:1892–1896. 13. Garcia M, Jemal A, Ward K. D, Cerhan JR, Hara K, Siegmund B, Thun MJ (2007). Global Cancer Facts & Figures 2007. Atlanta, GA, American Cancer Society. 14. Pham TM, Fujino Y, Ide R, Tokui N, Kubo T, Mizoue T et al. (2009). Years of life lost due to cancer in a cohort study in Japan. Eur J Public Health, 19:189–192. 15. Kim SG, Hahm MI, Choi KS, Seung NY, Shin HR, Park EC (2008). The economic burden of cancer in Korea in 2002. Eur J Cancer Care, 17(2):136–44. 16. Wang L, Kong L, Wu F, Bai Y, Burton R (2005). Preventing chronic diseases in China. Lancet, 366:1821–1824. 17. Joshi R, Cardona M, Iyengar S, Sukumar A, Raju CR, Raju KR et al. (2006). Chronic diseases now a leading cause of death in rural India – mortality data from the Andhra Pradesh rural health initiative. IntJEpidemiol, 35:1522–1529. 18. Statistics Division, United Nations Economic and Social Commission for Asia and the Pacific UNESCAP (2006). Asia and the Pacific in figures 2006. 19. Jin F, Devesa SS, Chow WH, Zheng W, Ji BT, Fraumeni JF, Jr. et al. (1999). Cancer incidence trends in urban shanghai, 1972–1994: an update. Int J Cancer 83:435–440. 20. Katanoda K, Yako-Suketomo H (2009). Comparison of time trends in stomach cancer incidence (1973–2002) in Asia, from Cancer Incidence in Five Continents, Vols IV–IX. Jpn J Clin Oncol, 39:71–72.
3
Obesity and Cancer in Asia
81
21. Yako-Suketomo H, Marugame T (2009). Comparison of time trends in colon, rectum and anus cancer incidence (1973–2002) in Asia, from ‘Cancer Incidence in Five Continents, Vols IV–IX’. Jpn J Clin Oncol, 39:196–198. 22. Shanghai Municipal Center for Disease Control & Prevention (2006). Shanghai Cancer Report. 23. Shanghai Municipal Center for Disease Control & Prevention (2007). Shanghai Cancer Report. 24. Ji BT, Devesa SS, Chow WH, Jin F, Gao YT (1998). Colorectal cancer incidence trends by subsite in urban Shanghai, 1972–1994. Cancer Epidemiol Biomarkers Prev, 7:661–666. 25. Ke L (2002). Mortality and incidence trends from esophagus cancer in selected geographic areas of China circa 1970–90. Int J Cancer, 102:271–274. 26. Yang L, Parkin DM, Li L, Chen Y (2003). Time trends in cancer mortality in China: 1987–1999. Int J Cancer, 106:771–783. 27. Fernandes ML, Seow A, Chan YH, Ho KY. (2006). Opposing trends in incidence of esophageal squamous cell carcinoma and adenocarcinoma in a multi-ethnic Asian country. Am J Gastroenterol, 101:1430–1436. 28. Chow WH, Blot WJ, Vaughan TL, Risch HA, Gammon MD, Stanford JL et al. (1998). Body mass index and risk of adenocarcinomas of the esophagus and gastric cardia. J Natl Cancer Inst, 90:150–155. 29. Zhang H, Chen SH, Li YM (2004). Epidemiological investigation of esophageal carcinoma. World J Gastroenterol, 10:1834–1835. 30. Zhou Y, Zhang Z, Zhang Z, Wu J, Ren D, Yan X et al. (2008). A rising trend of gastric cardia cancer in Gansu Province of China. Cancer Lett, 269:18–25. 31. London WT, Mcglynn KA (1996). Liver Cancer. In Schottenfeld D, Fraumeni JR (eds.), Cancer Epidemiology and Prevention, pp. 772–793. New York: Oxford University Press. 32. Lee CL, Hsieh KS, Ko YC (2003). Trends in the incidence of hepatocellular carcinoma in boys and girls in Taiwan after large-scale hepatitis B vaccination. Cancer Epidemiol Biomarkers Prev, 12:57–59. 33. Shetty PS (2002). Nutrition transition in India. Public Health Nutr, 5(1A):175–182. 34. Popkin BM, Horton S, Kim S, Mahal A, Shuigao J, Popkin BM et al. (2001). Trends in diet, nutritional status, and diet-related noncommunicable diseases in China and India: the economic costs of the nutrition transition. [Review] [58 refs]. Nutr Rev, 59:379–390. 35. Popkin BM (2001). The nutrition transition and obesity in the developing world. J Nutr, 131:871S–3S. 36. Ferlay J, Bray F, Pisani P, Parkin DM (2004). GLOBOCAN 2002: Cancer incidence, mortality and prevalence worldwide. IARC Cancer Base No.5 Lyon. IARC Press. 37. Bellisari A (2008). Evolutionary origins of obesity. Obes Rev, 9:165–180. 38. Yu H, Harris RE, Gao YT, Gao R, Wynder EL (1991). Comparative epidemiology of cancers of the colon, rectum, prostate and breast in Shanghai, China versus the United States. Int J Epidemiol, 20:76–81. 39. Popkin BM, Du S (2003). Dynamics of the nutrition transition toward the animal foods sector in China and its implications: a worried perspective. J Nutr, 133:3898S–906S. 40. Griffiths PL, Bentley ME (2001). The nutrition transition is underway in India. J Nutr, 131:2692–2700. 41. Ministry of Health Labour and Welfare (2004). The National Nutrition Survey in Japan, 2002 (in Japanese). Tokyo: Daiichi-Shuppan. 42. Kim S, Moon S, Popkin BM (2000). The nutrition transition in South Korea. Am J Clin Nutr, 71:44–53. 43. Ng SW, Norton EC, Popkin BM (2009). Why have physical activity levels declined among Chinese adults? Findings from the 1991–2006 China health and nutrition surveys. Soc Sci Med, 68:1305–1314. 44. Yadav K, Krishnan A (2008). Changing patterns of diet, physical activity and obesity among urban, rural and slum populations in north India. Obes Rev, 9:400–408.
82
W. Xu and C.E. Matthews
45. World Health Organization (2003). Global Strategy on Diet, Physical Activity and Health. 46. Ke-You G, Da-Wei F (2001). The magnitude and trends of under- and over-nutrition in Asian countries. Biomed Environ Sci, 14:53–60. 47. Zheng W, Chow WH, Yang G, Jin F, Rothman N., Blair A. et al. (2005). The Shanghai Women’s Health Study: Rationale, study design, and baseline characteristics (2005). Am J Epidemiol, 162:1123–1131. 48. Lee SA, Xu WH, Zheng W, Li H, Yang G, Xiang YB et al. (2007). Physical activity patterns and their correlates among Chinese men in Shanghai. Med Sci Sports Exerc, 39:1700–1707. 49. Department of Women and Child Development (1998). India Nutrition Profile. New Dehli: Government of India, Ministry of Human Resource Development. 50. Gopalan C (1998). Obesity in the urban middle class. Nutr Found India Bull, 19:1–4. 51. Yoshiike N, Seino F, Tajima S, Arai Y, Kawano M, Furuhata T et al. (2002). Twenty-year changes in the prevalence of overweight in Japanese adults: the National Nutrition Survey 1976–95. Obes Rev, 3:183–190. 52. Ministry of Health and Welfare (1999). National Health and Nutrition Survey 1998 Korea. 53. Ministry of Health and Welfare (2002). National Health and Nutrition Survey 2001 Korea. 54. World Health Organization (1998). Obesity: Preventing and Managing the Global Epidemic. Geneva, Switzerland: WHO. 55. van der KK, Leenen R, Seidell JC, Deurenberg P, Droop A, Bakker CJ (1993). Waist-hip ratio is a poor predictor of changes in visceral fat. Am J Clin Nutr, 57:327–333. 56. Pouliot MC, Despres JP, Lemieux S, Moorjani S, Bouchard C, Tremblay A et al. (1994). Waist circumference and abdominal sagittal diameter: best simple anthropometric indexes of abdominal visceral adipose tissue accumulation and related cardiovascular risk in men and women. Am J Cardiol, 73:460–468. 57. Xu WH, Matthews CE, Xiang YB, Zheng W, Ruan ZX, Cheng JR et al. (2005). Effect of adiposity and fat distribution on endometrial cancer risk in Shanghai women. Am J Epidemiol, 161:939–947. 58. Deurenberg P, Deurenberg YM, Wang J, Lin FP, Schmidt G (1999). The impact of body build on the relationship between body mass index and percent body fat. Int J Obes Relat Metab Disord, 23:537–542. 59. Deurenberg P, urenberg-Yap M, Guricci S. (2002). Asians are different from Caucasians and from each other in their body mass index/body fat per cent relationship. Obes Rev, 3:141–146. 60. Rush EC, Freitas I, Plank LD (2009). Body size, body composition and fat distribution: comparative analysis of European, Maori, Pacific Island and Asian Indian adults. Br J Nutr, 1–10. 61. Kagawa M, Kerr D, Uchida H, Binns CW (2006). Differences in the relationship between BMI and percentage body fat between Japanese and Australian-Caucasian young men. Br J Nutr, 95:1002–1007. 62. Oh SW, Shin SA, Yun YH, Yoo T, Huh BY (2004). Cut-off point of BMI and obesity-related comorbidities and mortality in middle-aged Koreans. Obes Res, 12:2031–2040. 63. Wen CP, vid Cheng TY, Tsai SP, Chan HT, Hsu HL, Hsu CC et al. (2009). Are Asians at greater mortality risks for being overweight than Caucasians? Redefining obesity for Asians. Public Health Nutr, 12:497–506. 64. Zhou BF (2002). Predictive values of body mass index and waist circumference for risk factors of certain related diseases in Chinese adults – study on optimal cut-off points of body mass index and waist circumference in Chinese adults. Biomed Environ Sci, 15:83–96. 65. World Health Organization (2000). International Association for the Study of Obesity & International Obesity Task Force. The Asia-Pacific Perspective: Redefining Obesity and Its Treatment. Health Communications. 66. Inoue M, Okayama A, Fujita M, Enomoto T, Tanizawa O, Ueshima HA (1994). Casecontrol study on risk factors for uterine endometrial cancer in Japan. Jpn J Cancer Res, 85: 346–350.
3
Obesity and Cancer in Asia
83
67. Iemura A, Douchi T, Yamamoto S, Yoshimitsu N, Nagata Y (2000). Body fat distribution as a risk factor of endometrial cancer. J Obstet Gynaecol Res, 26:421–425 68. Ng EH, Gao F, Ji CY, Ho GH, Soo KC (1997). Risk factors for breast carcinoma in Singaporean Chinese women: the role of central obesity. Cancer, 80:725–731. 69. Shu XO, Jin F, Dai Q, Shi JR, Potter JD, Brinton LA et al. (2001). Association of body size and fat distribution with risk of breast cancer among Chinese women. Int J Cancer, 94:449–455. 70. Hirose K, Tajima K, Hamajima N, Takezaki T, Inoue M, Kuroishi T et al. (1999). Effect of body size on breast-cancer risk among Japanese women. Int J Cancer, 80:349–355. 71. Gilani GM, Kamal S (2004). Risk factors for breast cancer in Pakistani women aged less than 45 years. Ann Hum Biol, 31:398–407 72. Mathew A, Gajalakshmi V, Rajan B, Kanimozhi V, Brennan P, Mathew BS et al. (2008). Anthropometric factors and breast cancer risk among urban and rural women in South India: a multicentric case-control study. Br J Cancer, 99:207–213. 73. Chow LW, Lui KL, Chan JC, Chan TC, Ho PK, Lee WY et al. (2005). Association between body mass index and risk of formation of breast cancer in Chinese women. Asian J Surg, 28:179–184. 74. Demark-Wahnefried W, Peterson BL, Winer EP, Marks L, Aziz N, Marcom PK et al. (2001) Changes in Weight, Body Composition, and Factors Influencing Energy Balance Among Premenopausal Breast Cancer Patients Receiving Adjuvant Chemotherapy. J Clin Oncol, 19:2381–2389. 75. Han HS, Lee KW, Kim JH, Kim SW, Kim IA, Oh DY et al. (2009). Weight changes after adjuvant treatment in Korean women with early breast cancer. Breast Cancer Res Treat,114(1):147–153. 76. Wu MH, Chou YC, Yu JC, Yu CP, Wu CC, Chu CM et al. (2006). Hormonal and body-size factors in relation to breast cancer risk: a prospective study of 11,889 women in a lowincidence area. Ann Epidemiol, 16:223–229. 77. Sonnenschein E, Toniolo P, Terry MB, Bruning PF, Kato I, Koenig KL et al. (1999). Body fat distribution and obesity in pre- and postmenopausal breast cancer. Int J Epidemiol, 28: 1026–1031. 78. Harvie M, Hooper L, Howell AH (2003). Central obesity and breast cancer risk: a systematic review. Obes Rev, 4(3):157–173. 79. Oh SW, Yoon YS, Shin SA (2005). Effects of Excess Weight on Cancer Incidences Depending on Cancer Sites and Histologic Findings Among Men: Korea National Health Insurance Corporation Study. J Clin Oncol, 23:4742–4754. 80. Tamakoshi K, Wakai K, Kojima M, Watanabe Y, Hayakawa N, Toyoshima H et al. (2004). A prospective study of body size and colon cancer mortality in Japan: The JACC Study. International Journal of Obesity & Related Metabolic Disorders: J Intl Assoc Study Obes, 28(4):551–558. 81. Song YM, Sung J, Ha M (2008). Obesity and risk of cancer in postmenopausal Korean women. J Clin Oncol, 26:3395–3402. 82. Oh TH, Byeon JS, Myung SJ, Yang SK, Choi KS, Chung JW et al. (2008). Visceral obesity as a risk factor for colorectal neoplasm. J Gastroenterol Hepatol 23:411–417. 83. Otake S, Takeda H, Suzuki Y, Fukui T, Watanabe S, Ishihama K et al. (2005). Association of Visceral Fat Accumulation and Plasma Adiponectin with Colorectal Adenoma: Evidence for Participation of Insulin Resistance. Clin Cancer Res, 11:3642–3646. 84. Kim JH, Lim YJ, Kim YH, Sung IK, Shim SG, Oh SO et al. (2007). Is metabolic syndrome a risk factor for colorectal adenoma? Cancer Epidemiol Biomarkers Prev, 16:1543–1546. 85. Kim Y, Kim Y, Lee S (2009). An association between colonic adenoma and abdominal obesity: a cross-sectional study. BMC Gastroenterol, 9:4. 86. Hsing AW, Deng J, Sesterhenn IA, Mostofi FK, Stanczyk FZ, Benichou J et al. (2000). Body size and prostate cancer: a population-based case-control study in China. Cancer Epidemiol Biomarkers Prev, 9:1335–1341.
84
W. Xu and C.E. Matthews
87. Zhang M, Xie X, Holman CD (2005). Body weight and body mass index and ovarian cancer risk: a case-control study in China. Gynecol Oncol, 98:228–234. 88. Niwa Y, Yatsuya H, Tamakoshi K, Nishio K, Kondo T, Lin Y et al. (2005). Relationship between body mass index and the risk of ovarian cancer in the Japanese population: findings from the Japanese Collaborate Cohort (JACC) study. J Obstet Gynaecol Res 31:452–458. 89. Mori M, Nishida T, Sugiyama T, Komai K, Yakushiji M, Fukuda K et al. (1998). Anthropometric and other risk factors for ovarian cancer in a case-control study. Jpn J Cancer Res, 89:246–253. 90. Ji BT, Hatch MC, Chow WH, McLaughlin JK, Dai Q, Howe GR et al. (1996). Anthropometric and reproductive factors and the risk of pancreatic cancer: a case-control study in Shanghai, China. Int J Cancer, 66:432–437 91. Lin Y, Kikuchi S, Tamakoshi A, Yagyu K, Obata Y, Inaba Y et al. (2007). Obesity, physical activity and the risk of pancreatic cancer in a large Japanese cohort. Int J Cancer, 120: 2665–2671 92. Yun JE, Jo I, Park J, Kim MT, Ryu HG, Odongua N et al. (2006). Cigarette smoking, elevated fasting serum glucose, and risk of pancreatic cancer in Korean men. Int J Cancer, 119: 208–212 93. Ansary-Moghaddam A, Huxley R, Barzi F, Lawes C, Ohkubo T, Fang X et al. (2006). The effect of modifiable risk factors on pancreatic cancer mortality in populations of the AsiaPacific region. Cancer Epidemiol Biomarkers Prev, 15:2435–2440 94. Hsing AW, Sakoda LC, Rashid A, Chen J, Shen MC, Han TQ et al. (2008). Body size and the risk of biliary tract cancer: a population-based study in China. Br J Cancer, 99:811–815 95. Brown LM, Swanson CA, Gridley G, Swanson GM, Schoenberg JB, Greenberg RS et al. (1995). Adenocarcinoma of the esophagus: role of obesity and diet. J Natl Cancer Inst, 87:104–109 96. Vaughan TL, Davis S, Kristal A, Thomas DB (1995). Obesity, alcohol, and tobacco as risk factors for cancers of the esophagus and gastric cardia: adenocarcinoma versus squamous cell carcinoma. Cancer Epidemiol Biomarkers Prev, 4:85–92 97. Guo W, Blot WJ, Li JY, Taylor PR, Liu BQ, Wang W et al. (1994). A nested case-control study of oesophageal and stomach cancers in the Linxian nutrition intervention trial. Int J Epidemiol, 23:444–450 98. Tran GD, Sun XD, Abnet CC, Fan JH, Dawsey SM, Dong ZW et al. (2005). Prospective study of risk factors for esophageal and gastric cancers in the Linxian general population trial cohort in China. Int J Cancer, 113:456–463 99. Smith M, Zhou M, Whitlock G, Yang G, Offer A, Hui G et al. (2008). Esophageal cancer and body mass index: results from a prospective study of 220,000 men in China and a metaanalysis of published studies. Int J Cancer, 122:1604–1610 100. Ji BT, Chow WH, Yang G, McLaughlin JK, Gao RN, Zheng W et al. (1997) Body mass index and the risk of cancers of the gastric cardia and distal stomach in Shanghai, China. Cancer Epidemiol Biomarkers Prev, 6:481–485 101. Zhang J, Su XQ, Wu XJ, Liu YH, Wang H, Zong XN et al. (2003). Effect of body mass index on adenocarcinoma of gastric cardia. World J Gastroenterol, 9:2658–2661 102. Awakura Y, Nakamura E, Ito N, Yamasaki T, Kamba T, Kamoto T et al. (2007). Influence of body mass index on prognosis of Japanese patients with renal cell carcinoma. Urology, 70:50–54. 103. Jonsson F, Wolk A, Pedersen NL, Lichtenstein P, Terry P, Ahlbom A et al. (2003). Obesity and hormone-dependent tumors: cohort and co-twin control studies based on the Swedish Twin Registry. Int J Cancer, 106:594–599 104. Guilera M, Connelly-Frost A, Keku TO, Martin CF, Galanko J, Sandler RS (2005). Does physical activity modify the association between body mass index and colorectal adenomas? Nutr Cancer, 51:140–145 105. Wen W, Cai Q, Xiang YB, Xu WH, Ruan ZX, Cheng J et al. (2008). The modifying effect of C-reactive protein gene polymorphisms on the association between central obesity and endometrial cancer risk. Cancer, 112:2409–2416
3
Obesity and Cancer in Asia
85
106. Sherman B, Wallace R, Bean J (1983). Estrogen use and breast cancer. Interaction with body mass. Cancer, 51(8):1527–1531. 107. Kim S, Baron JA, Mott LA, Burke CA, Church TR, Keown-Eyssen GE et al. (2006). Aspirin may be more effective in preventing colorectal adenomas in patients with higher BMI (United States). Cancer Causes Control, 17:1299–1304. 108. Dai Q, Franke AA, Yu H, Shu XO, Jin F, Hebert JR et al. (2003). Urinary phytoestrogen excretion and breast cancer risk: evaluating potential effect modifiers endogenous estrogens and anthropometrics. Cancer Epidemiol Biomarkers Prev, 12:497–502. 109. Dai Q, Gao Y-T, Shu XO, Yang G, Milne G, Cai Q et al. (2009). Oxidative stress, Obesity, and Breast Cancer Risk: Results from the Shanghai Women’s Health Study (SWHS). J Clin Oncol, (in press). 110. Slattery ML, Murtaugh MA, Quesenberry C, Caan BJ, Edwards S, Sweeney C (2007). Changing population characteristics, effect-measure modification, and cancer risk factor identification. Epidemiol Perspect Innov, 4:10. 111. Berclaz G, Li S, Price KN, Coates AS, Castiglione-Gertsch M, Rudenstam CM et al. (2004). Body mass index as a prognostic feature in operable breast cancer: the International Breast Cancer Study Group experience. Ann Oncol, 15(6):875–884. 112. Tao MH, Shu XO, Ruan ZX, Gao YT, Zheng W, Tao MH et al. (2006). Association of overweight with breast cancer survival. Am J Epidemiol 163:101–7. 113. Kyogoku S, Hirohata T, Takeshita S, Nomura Y, Shigematsu T, Horie A (1990). Survival of breast-cancer patients and body size indicators. Intl J Cancer, 46(5):824–831. 114. Moon HG, Ju YT, Jeong CY, Jung EJ, Lee YJ, Hong SC et al. (2008). Visceral obesity may affect oncologic outcome in patients with colorectal cancer. Ann Surg Oncol, 15(7): 1918–1922. 115. Park SM, Lim MK, Shin SA, Yun YH (2006). Impact of prediagnosis smoking, alcohol, obesity, and insulin resistance on survival in male cancer patients: National health insurance corporation study. J Clin Oncol, 24:5017–5024. 116. Moriwaki Y, Kunisaki C, Kobayashi S, Harada H, Imai S, Kasaoka C (2003). Does body mass index (BMI) influence morbidity and long-term survival in gastric cancer patients after gastrectomy? Hepatogastroenterology, 50:284–8. 117. Gunter MJ, Leitzmann MF (2006). Obesity and colorectal cancer: epidemiology, mechanisms and candidate genes. J Nutr Biochem, 17:145–156 118. Algul H, Treiber M, Lesina M, Schmid RM (2007). Mechanisms of disease: chronic inflammation and cancer in the pancreas – a potential role for pancreatic stellate cells? Nat Clin Pract Gastroenterol Hepatol, 4:454–462. 119. Nishida M, Moriyama T, Sugita Y, Yamauchi-Takihara K (2007). Abdominal obesity exhibits distinct effect on inflammatory and anti-inflammatory proteins in apparently healthy Japanese men. Cardiovasc Diabetol, 6:27. 120. Liu J, Zhai X, Jin G, Hu Z, Wang S, Wang X et al. (2006). Functional variants in the promoter of interleukin-1beta are associated with an increased risk of breast cancer: a case-control analysis in a Chinese population. Int J Cancer, 118:2554–8. 121. Chiu HM, Lin JT, Chen TH, Lee YC, Chiu YH, Liang JT et al. (2008). Elevation of C-reactive protein level is associated with synchronous and advanced colorectal neoplasm in men. Am J Gastroenterol, 103:2317–2235 122. Li CQ, Pignatelli B, Ohshima H (2000). Coexpression of interleukin-8 and inducible nitric oxide synthase in gastric mucosa infected with cagA+ Helicobacter pylori. Dig Dis Sci, 45:55–62. 123. Hankinson SE, Willett WC, Manson JE, Hunter DJ, Colditz GA, Stampfer MJ et al. (1995). Alcohol, height, and adiposity in relation to estrogen and prolactin levels in postmenopausal women. J Natl Cancer Inst, 87:1297–1302. 124. Nagata C, Shimizu H, Takami R, Hayashi M, Takeda N, Yasuda K (2000). Relations of insulin resistance and serum concentrations of estradiol and sex hormone-binding globulin to potential breast cancer risk factors. Jpn J Cancer Res, 91:948–953
86
W. Xu and C.E. Matthews
125. Key TJ, Appleby PN, Reeves GK, Roddam A, Dorgan JF, Longcope C et al. (2003). Body mass index, serum sex hormones, and breast cancer risk in postmenopausal women. J Natl Cancer Inst, 95:1218–1226. 126. Kaaks R, Lukanova A, Kurzer MS. (2002). Obesity, endogenous hormones, and endometrial cancer risk: a synthetic review. Cancer Epidemiol Biomarkers Prev, 11:1531–1543. 127. Suga K, Imai K, Eguchi H, Hayashi S, Higashi Y, Nakachi K (2001). Molecular significance of excess body weight in postmenopausal breast cancer patients, in relation to expression of insulin-like growth factor I receptor and insulin-like growth factor II genes. Jpn J Cancer Res, 92:127–134. 128. Yu H, Jin F, Shu XO, Li BD, Dai Q, Cheng JR et al. (2002). Insulin-like growth factors and breast cancer risk in Chinese women. Cancer Epidemiol Biomarkers Prev, 11:705–712. 129. Fair AM, Dai Q, Shu XO, Matthews CE, Yu H, Jin F et al. (2007). Energy balance, insulin resistance biomarkers, and breast cancer risk. Cancer Detect Prev, 31:214–9. 130. Berrington de GA, Yun JE, Lee SY, Klein AP, Jee SH (2008). Pancreatic cancer and factors associated with the insulin resistance syndrome in the Korean cancer prevention study. Cancer Epidemiol Biomarkers Prev, 17:359–364. 131. Hsing AW, Gao YT, Chua S Jr, Deng J, Stanczyk FZ (2003). Insulin resistance and prostate cancer risk. J Natl Cancer Inst, 95:67–71. 132. Yang G, Lu G, Jin F, Dai Q, Best R, Shu XO et al. (2001). Population-based, case-control study of blood C-peptide level and breast cancer risk. Cancer Epidemiol, Biomarkers Prev, 10:1207–1211. 133. Calle EE,Kaaks R (2004). Overweight, obesity and cancer: epidemiological evidence and proposed mechanisms. Nat Rev Cancer 4:579–591. 134. Mayne ST,Navarro SA (2002). Diet, obesity and reflux in the etiology of adenocarcinomas of the esophagus and gastric cardia in humans. J Nutr 132:3467S–3470S. 135. Corley DA, Kubo A, Levin TR, Block G, Habel L, Zhao W et al. (2007). Abdominal obesity and body mass index as risk factors for Barrett’s esophagus. Gastroenterology, 133:34–41. 136. Okabayashi T, Gotoda T, Kondo H, Inui T, Ono H, Saito D et al. (2000). Early carcinoma of the gastric cardia in Japan: is it different from that in the West? Cancer 89:2555–9. 137. Kendall BJ, Macdonald GA, Hayward NK, Prins JB, Brown I, Walker N et al. (2008). Leptin and the risk of Barrett’s oesophagus. Gut, 57:448–454. 138. Whiteman DC, Sadeghi S, Pandeya N, Smithers BM, Gotley DC, Bain CJ et al. (2008). Combined effects of obesity, acid reflux and smoking on the risk of adenocarcinomas of the oesophagus. Gut, 57:173–180. 139. Clark JM (2006). The epidemiology of nonalcoholic fatty liver disease in adults. J Clin Gastroenterol, 40:S5–10. 140. El-Serag HB,Rudolph KL (2007). Hepatocellular carcinoma: epidemiology and molecular carcinogenesis. Gastroenterology, 132:2557–2576.
Chapter 4
Genetic Epidemiology of Obesity and Cancer Courtney Gray-McGuire, Indra Adrianto, Thuan Nguyen, and Chee Paul Lin
Abstract To understand the genetic epidemiology of obesity and cancer means to consider the genetics of not only these two complex diseases but also the many risk factors associated with them and the molecular mechanisms likely to contribute to their underlying etiology. It is therefore the purpose of this chapter to highlight the following: genetic mapping studies that have been done to isolate genes related to obesity, genetic studies aimed to identify genes for obesity risk factors, and, finally, a discussion of the genes shared across or among the studies of obesity and/or its risk factors and the various cancers. This is by no means an exhaustive discussion as these are each complex traits with complex molecular mechanisms. It is, however, intended to be an overview of the genes likely to be involved in the cascade of pathways linking obesity and cancer. It is well known that obesity continues to be a leading public health concern not only in the United States but now worldwide. From 1980 to 2002, the prevalence of obesity doubled in adults aged over 20 and overweight prevalence tripled in children and adolescents aged 6–9 years. The estimates of overweight and obesity in 2003– 2004 showed that 17.1% of US children and adolescents were overweight and 32.2% of adults were obese [143]. The increasing rates of obesity among children are especially alarming and suggest continuing increases in the rates of obesity-related cancers. Since being overweight in childhood predicts obesity in adulthood [208], obesity occurring in childhood may operate as a ‘cumulative’ exposure that influences cancer risk in later life. Obese children may be subject to potentially long periods of hormonal exposures that operate during adolescence to influence propensity to neoplastic diseases. In females, childhood obesity may also affect the levels of sex hormone at
C. Gray-McGuire (B) Arthritis and Immunology Research Program, Oklahoma Medical Research Foundation, Oklahoma City, OK 73104-5097, USA e-mail:
[email protected]
N.A. Berger (ed.), Cancer and Energy Balance, Epidemiology and Overview, Energy Balance and Cancer 2, DOI 10.1007/978-1-4419-5515-9_4, C Springer Science+Business Media, LLC 2010
87
88
C. Gray-McGuire et al.
adolescence. A prospective study of 55-year follow-up on 508 adolescents at 13–18 years from Harvard Growth Study showed that overweight adolescent boys had a nine-fold increased risk of colorectal cancer mortality [161]. Associations between adolescent weight and cancer persisted even after adjusting for adult BMI. A 50-year follow-up of >2,000 British children showed an overall 9% increase in cancer incidence per standard deviation increase in BMI, with effects three times larger for smoking-related cancers. There is also evidence from an Israeli case–control study which indicated that being in the upper quartile of BMI at age 18 years was associated with a 42% increase for ovarian cancer [24]. However, the mechanisms underlying the association between childhood obesity and cancer are not well understood, particularly in conjunction with genetic data. In one case– control study of 40 obese and 40 non-obese prepubertal children, the obese group showed higher levels of IGF-1, insulin, and lower sex-binding hormone than that of the non-obese group [182]. This suggests that high levels of growth factors and altered sex hormone profiles are present in obese children and the exposures to an adverse metabolic milieu may begin early in life.
1 Genes for Obesity This public health concern has led to an increase in the number of studies aimed to isolate genes associated with obesity, but implications of the importance of heredity in obesity began several decades ago. Twin studies [127] indicated an important role of genetic factors, supported by several family studies, even in somewhat isolated populations. In a study of a Hutterite group, Paganini-Hill et al. [146] found evidence for a major gene in the determination of ‘bulk factor.’ In a study in the Danish Adoption Register, Stunkard et al. [185] found a strong relation between weight classes, thin, median weight, overweight, or obese, and the body mass index of the biologic parents but no significant correlation between the weight class of adoptees and the body mass index of their adoptive parents. Another twin study subjected 12 pairs of identical male twins to overfeeding by 1,000 kcal per day, 6 days a week, for a period of 100 days. The variance between pairs in response to overfeeding was about three times greater than that within pairs. With respect to the changes in regional fat distribution and amount of abdominal visceral fat, the differences were particularly striking, with six times as much variance between pairs compared to within pairs [14]. These results supported involvement of genetic factors in both storage of energy as either fat or lean tissue and the various determinants of resting expenditure of energy. Several family studies have also been conducted which support a genetic contribution to obesity. A study of nuclear families with school-aged children in northern Italy that included 67 families with children classified as obese and 112 families with non-obese children and their parents and sibs suggested a dominant major gene with a weak effect [225]. Recessive inheritance was implicated in a study of a relative fat pattern index (RFPI), i.e., the ratio of subscapular skinfold thickness to the sum of subscapular and suprailiac skinfold thicknesses which included 774
4
Genetic Epidemiology of Obesity and Cancer
89
adults from 59 pedigrees ascertained through cases of cardiovascular disease [75]. This same study suggested that the variance in RFPI is 42.3% due to the major locus, 9.5% due to polygenic inheritance, and 48.2% due to random environmental effects. Moll et al. [130] in a study investigating the role of genetic and environmental factors in determining variability in ponderosity (body weight relative to height) also suggested a single recessive locus with major effect. Ponderosity was measured by body mass index (BMI; kg/m2 ) in the mothers, fathers, and sibs of 284 school children in Muscatine, Iowa, and the major locus accounted for almost 35% of the adjusted variation in BMI. Polygenic loci accounted for an additional 42% of the variation. Approximately 23% of the adjusted variation was not explained by genetic factors. Several family-based linkage studies have been conducted, some of which have focused on special populations or specific genetic effects. For example, a genomewide linkage scan was conducted in the somewhat isolated and homogeneous population of Pima Indians (which also has a very high incidence of obesity). This study found single-marker linkages to percentage body fat using sib pair analysis for quantitative traits. From these analyses, the best evidence of genes influencing body fat came from markers at chromosome 11q21-q22 and 3p24.2-p22 [142]. Other studies have aimed to identify the interplay between the many regions identified as a part of linkage and association studies. For example, Dong et al. [50] evaluated potential epistatic interactions using independent obese-affected sibling pairs in 244 families. Both the affected sib pair-specific IBD-sharing probability and the family-specific NPL score revealed that there were strong positive correlations between the effects on chromosome 10q (88–97 cM) and 20q (65–83 cM). To detect potentially imprinted, obesity-related genetic loci, these same investigators [49] performed genome-wide parent-of-origin linkage analyses. They studied a European-American sample of 1,297 individuals from 260 families and then replicated in two smaller, independent samples. For discrete trait analysis, they found evidence for a maternal effect in 10p12 across the three samples, with both multipoint LOD scores over 4.0 in the pooled sample. For quantitative trait analysis, they found a maternal effect in region 12q24 (multipoint LOD of 4.01 for BMI and 3.69 for waist circumference) and a paternal effect (multipoint LOD of 3.72 for BMI) in region 13q32 in Caucasians. Few of the genetic effects suggested for obesity have been replicated and are generally accepted. One of those is the FTO locus on chromosome 16q12.2. Frayling et al. [55] identified an association between obesity and a common variant in this gene as a part of a genome-wide association study of 1,924 type II diabetes patients and 2,938 controls from the United Kingdom for close to 500,000 SNPs. SNPs in the FTO region were confirmed as a part of the same study using an additional 3,757 type II diabetics and 5,346 controls. The diabetes risk alleles at FTO were strongly associated with increased BMI, and further analysis showed that the association of the FTO SNPs with type II diabetes was mediated through an increased risk for BMI. Frayling et al. [55] analyzing an additional 19,424 white European adults confirmed association of the A allele of rs9939609 with increased BMI from seven population-based studies and 10,172 white European children from two
90
C. Gray-McGuire et al.
population-based studies. In all adult population-based studies, the risk for higher BMI was additive, such that those homozygous for the A allele had a higher BMI than those heterozygous for the A allele and the low-risk T allele. The attributable risk for rs9939609 was approximately 20% for BMI >30 and approximately 13% for being overweight (BMI more than 25 kg/m2 ). Study of at-risk children showed that rs9939609 was associated with increased BMI and obesity by the age of 7 years. Further sequence analysis of 47 patients with a BMI of more than 40 kg/m2 did not reveal any obvious functional variants in the FTO coding region, minimal splice sites, or 3 UTR. In a study of almost 3,000 affected individuals and over 5,000 controls, Dina et al. [47] identified two potentially functional SNPs in intron 1 of the FTO gene that were consistently strongly associated with early-onset and severe obesity. The at-risk haplotype yielded a proportion of attributable risk of 22% for common obesity. An additional study of 2,726 children found that the A allele of the FTO variant rs9939609 was associated with increased weight and increased fat mass but not lean mass [25]. The authors concluded that the FTO variant that confers a predisposition to obesity may play a role in the control of food intake and food choice, perhaps involving a hyperphagic phenotype or a preference for energy-rich foods. Other GWAS have also highlighted the role of FTO but have identified unique loci as well. Loos et al. [116], for example, performed a meta-analysis of data from four European population-based studies and three case–control studies, involving a total of 16,876 individuals of European descent, and confirmed the association between FTO and BMI as well as found a significant association between rs17782313, near the melanocortin-4 receptor (MC4R), and BMI in adults and children. Likewise, Thorleifsson et al. [189] conducted a GWAS for both weight and body mass index (BMI) in a sample of over 30,000 Caucasian and 1,100 African Americans and combined the results with previously published results from the Diabetes Genetics Initiative (DGI) on 3,024 Scandinavians. They too confirmed previously identified variants close to or in the FTO, MC4R, BDNF, and SH2B1 genes, in addition to variants at seven loci not previously connected with obesity and not specific to any given pathway. Additional genes, including APOE and TGF-beta-1, have been associated with the obesity phenotypes of fat mass, percentage fat mass, and lean mass and ENPP1 and PCSK1 with childhood and adult obesity and increased risk of glucose intolerance and type II diabetes. Long et al. [113] analyzed several SNPs of each gene in 1,873 subjects from 405 white families to test for linkage or association with BMI, fat mass, percentage fat mass, or lean mass. A significant linkage disequilibrium was observed between pairs of SNPs within each gene. Within-family association was observed in the APOE gene for percentage fat mass and fat mass and was found between lean mass and the TGFB1 gene. Finally, Meyre et al. [128] analyzed genome-wide association data from 1,380 Europeans with early-onset and morbid adult obesity and 1,416 age-matched normal weight controls and confirmed association at one of three SNPs in a risk haplotype in ENPP1, with replication in an additional 14,186 European individuals.
4
Genetic Epidemiology of Obesity and Cancer
91
2 Genetically Influenced Environmental Risk Factors As mentioned, there are many factors that contribute to obesity, each with their own genetic susceptibility. While genes specific to fat mass and weight have been identified, it is most likely the interplay between the genes predisposing to the risk factor and the genes for obesity that lead to the many complications associated with this complex disease. Among these environmental risk factors are diet, exercise, sleep, and other mechanisms such as viruses and thermoregulation. Some argue that although genetic susceptibility may explain up to 40% of the obesity phenotype, technological, lifestyle, and cultural changes over the past 50 years are the most likely causes of the recent obesity epidemic. This section addresses both the environmental risk factors for obesity and the molecular mechanisms with which they are most likely associated.
2.1 Food Intake and Appetite Regulation Diet and eating habits play an important role in controlling the weight of a person [163, 159]. Individuals with improper balance of caloric intake over the short or long term may develop a high risk of obesity. Excessive intake of energy-dense foods and sugary drinks combined with decreased physical expenditure are involved in the growing obesity epidemic. Dietary intakes, particular nutrients, and foods are all associated with both obesity and several types of cancer. Specifically, people at risk have a very high fat content in their diets, low fiber, and eat fewer fruits and vegetables. In addition, there is often a lack of exercise in their lifestyle. The World Health Organization (WHO) also suggests that up to one-third of the cancer burden could be reduced by implementing cancer prevention strategies, which include making simple changes in diet and lifestyle [220]. 2.1.1 Genes for Food Intake and Appetite Regulation Interestingly, there have been several studies implicating genes as contributors of food intake and appetite regulation. Previous studies have shown that the body regulates energy intake and consumption through hormonal signals from the adipose stores to the hypothalamus, which sends the neuronal signals to the brain to reduce appetite and increase energy expenditure. Thus the central nervous system plays a critical role in the regulation of blood glucose. A study by Schwartz et al. emphasized that there are two key components of hypothalamic neurons involved in body weight regulation: (a) the anorexigenic melanocortin neurons which express proopiomelancortin (POMC) and reduce appetite while elevating energy expenditure and (b) the orexigenic neurons that express neuropeptide Y (NPY) which have the opposite effects on the central energy metabolism [173]. Further studies by Lee et al. showed that imbalances from either POMC or NPY neuronal genes can lead to an inappropriate increase in
92
C. Gray-McGuire et al.
appetite and reduction in energy expenditure [105]. Studies led by Cone et al. also showed that leptin acts directly on two distinct classes of neurons: one class that co-expresses anorexigenic peptide POMC and CART (cocaine and amphetaminerelated transcript) which reduce food intake, while the other co-expresses the orexigenic peptides NPY and AgRP (agouti-related protein) which increases food intake. These neuronal peptides are regulated by leptin which sends a signal directly to the hypothalamus. These studies have also shown that MC3 and MC4 receptors, expressed in the brain and CNS, play a central role in the control of body weight [35]. In addition to the genes in the central nervous system, genes in the endocrine system have also shown to be a critical mediator in energy balance and body weight regulation. The laboratory of J.M. Friedman cloned the obese gene (ob) in mice and demonstrated that leptin-deficient mice (ob/ob) given leptin are capable of sustained weight loss. Similar studies from other groups found that leptin also plays a major role in energy balance and intake in that it belongs to a family of proteins secreted by adipocytes [192] that signals certain neuronal factors in the hypothalamus to reduce appetite because of fat stores and increase energy expense. Leptin belong to a family of adipokines, which includes resistin, adiponectin, FIAF (fasting-induced adipose factor), visfatin, vaspin, and a few unexpected candidates including nerve growth factors. The diabetes gene was cloned and identified as the leptin receptor shortly after the cloning of the leptin gene [188, 28, 31]. The leptin receptor (ObRb) was shown to be an important mediator of many different signaling cascades, such as STAT3, IFG-1, PI3K, SOC3, PTP1B (protein tyrosine phosphatase 1B). The importance of the central nervous system and likely genes controlling eating behavior was highlighted in a study by Willer et al., which performed a metaanalysis of 15 genome-wide association studies for BMI (n > 32,000) and followed up top signals in 14 additional cohorts (n > 59,000). They strongly confirmed FTO and MC4R, known obesity genes, but also identified six additional loci: TMEM18, KCTD15, GNPDA2, SH2B1, MTCH2, and NEGR1, all genes which are known to be highly expressed or to act in the central nervous system, emphasizing, as in rare monogenic forms of obesity, the role of the central nervous system in predisposition to obesity [210].
2.2 Physical Activity Physical activity is one of the most effective ways to both prevent and control obesity. With industrialization, urbanization, and mechanization, there are increasing numbers of people adapted to sedentary ways of life. It suggests that people who are not spending adequate time in maintaining a minimal level of physical activity to balance their caloric intake and caloric expenditure may obtain a positive energy balance that gives rise to obesity and increases the risk for cancer. However, the efficiency with which individuals expend calories, accumulate lean or fat mass is also
4
Genetic Epidemiology of Obesity and Cancer
93
highly influenced by genes. We present here a description of the risk associated with lack of exercise and also review both animal and human studies implicating genes in the regulation of calorie expenditure and efficiency of exercises, including fitness phenotypes. The prevalence of a sedentary lifestyle in youth is particular alarming, the most recent Youth Risk Behavior Survey (2007) indicated that 35% of high school students watched TV and 25% played video game/computer games or spent more than 3 h per average school day on the computer for something unrelated to school work. The study also reported 65% of the students did not meet recommended levels of physical activity and 46% did not attend physical education classes [193]. These behaviors may potentially be the risk factor to obesity in youth. A longitudinal study was conducted to assess the relationship between sedentary behavior and overweight in adolescent girls. The results showed an inverse association between sedentary behavior and light activity over the time period from 6th to 8th grade. Over the 2-year time period, adolescent girls shift their time from light activity to more sedentary activity. However, the changes in sedentary and light activity were not linked to the changes in BMI [63]. A longer follow-up may be required to observe the relationship between sedentary activity and BMI. The sedentary behavior and physical activity are not mutually exclusive; the association exists between sedentary behavior, physical activity, and obesity. In theory, young adults may be highly active and highly sedentary. For example, youth could be involved in both moderate to high physical activity and sedentary activity over the course of the day. The sedentary activity is particularly evident on weekends. Likewise, youth could also fit into other subgroups such as those who are highly active and low sedentary, low active and low sedentary, or low active and highly sedentary. A cross-sectional study showed that low-active, high-sedentary boys were 1.6 times more likely to be overweight than high-active, low-sedentary boys. The low-active, high-sedentary girls are two-fold more likely to be overweight than high-active, low-sedentary girls. However, high-active, high-sedentary girls were 1.91 times more likely to be overweight than were high-active, low-sedentary girls. These results indicate that sedentary behavior may moderate the relationship between physical activity and overweight [216]. Of course, as mentioned previously, we know that molecular mechanisms underlying the efficiency with which an individual responds to physical activity likely play a role in this relationship as well. Therefore, we present here a discussion of the role of genes in human physical performance and health-related fitness. The physical performance phenotypes include cardiorespiratory endurance and muscle strength whereas the health-related fitness phenotypes are grouped into four categories: hemodynamics; anthropometry and body composition; insulin and glucose metabolism; and blood lipid, lipoprotein, and hemostatic factors [16]. Animal and human studies using case–control and other designs are described. The mechanism of the effect of exercise on inflammatory processes has been well characterized and calorie restriction, physical activity alone, or in combination are known to reduce certain inflammatory markers. In general, acute inflammation is a process that is beneficial to the host by providing protection from invading
94
C. Gray-McGuire et al.
pathogens and initiating wound healing, whereas chronic systemic inflammation, which typically accompanies obesity, is a low-grade pervasive form of inflammation that damages the endothelial linings of arteries and a variety of tissues and organs that causes disruptive effects to the nervous, endocrine, and other systems. Several cross-sectional studies showed an association between physical inactivity and systemic inflammation. The cancer risk is lowered by the effect of physical activity in reduction of chronic inflammation, as exercise-induced increases in systemic IL-6 may result in reduced pro-inflammatory mediators and elevated anti-inflammatory factors [85]. 2.2.1 Genes for Physical Activity Animal studies can be used to investigate the genetic analysis of complex phenotypes. The main advantage of animal studies on the genetics of performance and fitness phenotypes over human studies is the control on several factors including heterogeneity, breeding, variation, physiological studies, therapies, and hypothesis testing. A study by Ways et al. [206] applied the quantitative trait locus (QTL) mapping to identify genes contributing to maximal exercise endurance in mice. They found three QTLs linked with exercise endurance in an F2 rat model derived from Copenhagen (COP) and DA inbred progenitors. They found a significant linkage of exercise endurance with D16Rat17 (LOD of score of 4.0), a suggestive linkage with D16Rat55 (LOD score of 2.9) on chromosome 16, and a suggestive linkage with D3Rat56 (LOD score of 2.2) on chromosome 3. Lightfoot et al. [111] conducted a QTL study on F2 mice derived from an intercross of two inbred strains with high maximal exercise endurance (Balb/cJ) or low maximal exercise endurance (DBA/2 J) run on a treadmill to test their exercise endurance. They identified a significant QTL on chromosome X for all mice and a suggestive QTL in the female mice on chromosome 8 from selective mapping. From fine mapping, their findings were confirmed. A significant QTL was found at 57.9 cM (DXMIT31, LOD score of 2.26) on chromosome X. Suggestive QTLs were found for DXMit121 (LOD score of 2.13), DXMIT5 (LOD score of 2.13) and DXMIT236 (LOD score of 2.10) on chromosome X in all mice, and D8Mit359 (LOD score of 1.19) on chromosome 8 in the female mice. A recent study by MacArthur et al. [124] revealed that alpha actinin-3 deficiency alters skeletal muscle metabolism and increases endurance performance in a knockout mouse model. Homozygous knockout (Actn3–/– ) mice that do not have alpha actinin-3 protein were compared to wild-type (Actn3+/+ ) littermate controls. They found that Actn3–/– mice have more intense staining muscle sections than Actn3+/+ mice. In the intrinsic exercise capacity test, where mice were placed on a motorized treadmill at increasing speeds until they were exhausted, Actn3–/– mice ran on average 33% further than Actn3+/+ mice. They then suggested that the ACTN3 577X allele also influences muscle metabolism in humans. The role of the angiotensin-converting enzyme (ACE) insertion/deletion (I/D) polymorphism on endurance performance has been extensively investigated using
4
Genetic Epidemiology of Obesity and Cancer
95
case–control studies [7, 8, 34, 59, 82, 120, 132, 137, 139, 170, 195]. The ACE is involved in the degradation of vasodilator bradykinin and converts angiotensin I to the vasoconstrictor angiotensin II that influences blood pressure [7, 132]. Recent case–control studies have identified a higher frequency of the I allele in the ACE locus in British high-altitude mountaineers [132], Australian rowers [59], British Olympic-standard distance runners [137], and South African triathlon athletes [34] compared to controls. An increased frequency of the homozygous II ACE genotype has been reported in Italian road cyclists, track and field endurance athletes, and cross-country skiers [170]; Russian swimmers, skiers, triathlon athletes, and track and field athletes [139]; Spanish cyclists, long-distance runners, and handball players (P = 0.0009) [7]; and Turkish university athletes [195]. On the other hand, an excess of the deletion (D) allele in the ACE locus has been found among elite short distance swimmers [137, 218] and sprinters [137] that require more power than endurance. Additional studies examining aerobic phenotypes implicate several other genes. A case–control study by Wolfarth et al. [215] revealed significant differences in allele and genotype frequencies for the alpha-2A-adrenoceptor gene (ADRA2A) between elite endurance athletes and sedentary controls. Another study by Wolfarth et al. [214] found a significant difference in the genotype distribution between athletes and controls, indicating a significant association between the Arg16Gly single nucleotide polymorphism in the beta2-adrenergic receptor (ADRB2) gene and endurance performance. Other case–control studies found genes that associated with endurance performance including the adenosine monophosphate deaminase 1 (AMPD1) [165], the peroxisome proliferator-activated receptor-gamma coactivator 1 alpha (PPARGC1A) [120], the bradykinin beta 2 receptor (BDKRB2) [169], and peroxisome proliferator-activated receptor alpha (PPARalpha) [5]. Hagberg et al. [68] found that, in postmenopausal women, the ACE insertion/insertion (II) genotype group had a significant higher maximal O2 consumption (VO2max ) than the ACE deletion/deletion (DD) genotype group after assessing the effect of physical activities levels. They also reported that the ACE II genotype group had a significantly higher VO2max than the ACE insertion/deletion (ID) genotype group. The ACE II genotype group also had a higher maximal arterial–venous O2 difference (a-vDO2 ) compared to D-allele carriers. Similar findings have also been reported by Hagberg et al. [69], Zhao et al. [224], and Kasikcioglu et al. [94]. Abraham et al. [3] studied 57 patients with congestive heart failure and found that the ACE II genotype group had a significantly higher maximum exercise time on the treadmill and VO2peak than the ACE ID and DD genotype groups. A study on 33 patients with chronic obstructive pulmonary disease (COPD) by Kanazawa et al. [93] revealed that the mean pulmonary arterial pressure (mPAP) after exercise was significantly higher in patients with the DD genotype than in those with the II genotype. Pulmonary vascular resistance (PVR) after exercise was also significantly higher in patients with the DD genotype than in those with the II genotype or ID genotype. Furthermore, DO2 after exercise was significantly lower in patients with the DD genotype than in those with the II genotype or ID genotype. No significant difference was found in PVR or DO2 between placebo or nifedipine administered
96
C. Gray-McGuire et al.
controls or all genotype groups. However, they found that nifedipine administration significantly reduced mPAP after exercise in all groups. Another study in 232 patients with heart failure [203] examined polymorphisms of ADRB2 in relation to exercise capacity. They reported that patients with the Ile164 polymorphism had a lower VO2peak than patients with Thr164. Liggett et al. [110] suggested that patients with Ile164 progress more quickly to death or transplantation than those with Thr164. McCole et al. [126] revealed an association between ADRB2 and a-vDO2 in postmenopausal women during treadmill exercise. They found that the Gln/Gln homozygote women had higher a-vDO2 than the Glu/Glu carriers during submaximal and maximal exercise. The same study indicated that the Gln carriers had higher VO2max than the Glu carriers. Moore et al. [133] investigated 63 non-obese postmenopausal Caucasian women and found that VO2max was lower in ADRB2 Glu27Glu than in ADRB2 Glu27Gln and Gln27Gln genotype women. Hence, they suggested that the Gln27Gln and Gln27Glu genotypes might associate with elite endurance performance in older women. Wagoner et al. [204] studied polymorphisms of ADRB1 in 263 patients with congestive heart failure in relation to exercise capacity and found that patients homozygous for Gly389 had significantly lower VO2peak compared with those with Arg389. Sandilands et al. [168] also confirmed the Wagoner study although the differences between 389R and 389G homozygotes in the Sandilands study were somewhat larger than that reported by Wagoner et al. [204]. Defoor et al. [41] conducted the CAREGENE (cardiac rehabilitation and genetics of exercise performance) study on 935 patients with coronary artery disease (CAD). They found that patients with the Gly49Gly genotype of ADRB1 had significantly higher covariate-adjusted aerobic power at baseline than those with Ser49Ser and Ser49Gly. Lopez-Alarcon et al. [117] examined the association of a genetic polymorphism of the insulin-like growth factor, IGF-I, on body composition, exercise performance, and exercise economy in 114 premenopausal women. After adjusting for African admixture, they found that IGF-I was negatively associated with lean body mass and lean leg mass, but not with leg strength, and IGF-I carriers had a longer time on the treadmill after adjusting for AFADM. A negative relationship was found between oxygen uptake during cycling. Hersh et al. [79] studied genetic associations for COPD-related phenotypes, including measures of exercise capacity, pulmonary function, and respiratory symptoms in 304 COPD patients. They found an association of single nucleotide polymorphisms (SNPs) in the microsomal epoxide hydrolase (EPHX1) and in the latent transforming growth factor-beta binding protein-4 (LTBP4) maximal output on cardiopulmonary exercise testing (Wmax ). Hautala et al. [76] examined peroxisome proliferator-activated receptor-delta (PPARdelta) gene polymorphisms in relation to cardiorespiratory fitness and plasma lipid responses to endurance training. They found a lower training response in maximal power output (Wmax ) in the exon 4 +15 C/C homozygotes compared with the heterozygotes and the T/T homozygotes in black subjects, and a similar finding was reported in white subjects.
4
Genetic Epidemiology of Obesity and Cancer
97
They also found that the exon 4 +15 C/C homozygotes had a smaller traininginduced increase in maximal oxygen consumption (P = 0.028) compared with the C/T and T/T genotypes in black subjects. The HERITAGE family study [162] investigated quantitative trait loci for maximal exercise capacity phenotypes and their responses to a standardized 20-week endurance training program in sedentary black and white subjects. They found promising linkages in the sedentary state for VO2max on chromosome 11p15 in Caucasians and for Wmax on chromosome 10q23 in Caucasians, and on chromosome 1p31 in African Americans for VO2max and chromosome 5q23 in Caucasians for Wmax for their responsiveness to training. They also reported suggestive evidence of linkage on 13q33 and 18q12 for baseline Wmax in whites and on 1p31, 7q32, and 7q36 for baseline VO2max in blacks. The suggestive evidence of linkage of VO2max training response was reported on 16q22 and 20q13.1 in blacks and on 4q27, 7q34, and 13q12 in whites. Genes for anaerobic activity have also been identified, implicating the interplay of both aerobic and anaerobic components of physical performance. Ahmetov et al. [5] found an increasing linear trend of the C allele with increasing anaerobic components of physical performance by studying the intron 7 G/C polymorphism in the peroxisome proliferator-activated receptor alpha (PPARA) gene in Russian athletes. They also reported that GG genotype frequencies in enduranceoriented and power-oriented athletes were significantly different compared to controls. Furthermore, the muscle fiber analysis revealed that the CC genotype carriers had significantly lower percentages of slow-twitch fibers compared to the GG genotype carriers. An examination of the alpha actinin-3 (ACTN3) gene locus and its nonsense R577X in African athletes by Yang et al. [221] did not find any significant genotype frequency differences between Nigerian sprinters and controls [16]. A study by Oh [144] on the distribution of the I/D polymorphism in the ACE gene in elite Korean athletes also did not reveal a significant difference between athletes and controls. Moran et al. [135] investigated 40 m sprint performance in 992 adolescent Greeks and found that there was a significant association (P = 0.003) between the ACTN3 R577X polymorphism and the sprint time, with the 577R allele contributing the faster times. Vincent et al. [201] examined the ACTN3 R577X polymorphism in relation to isometric and isokinetic knee extension strength in 90 young males and found significantly higher relative dynamic quadriceps torques at 300◦ /s in RR carriers compared with XX carriers (P = 0.04). This same group studied [135] the associations of the ACE polymorphisms with physical, physiological, and skill parameters in 1,027 teenage Greeks. They reported a strong association (P < 0.001) between the ACE I/D (insertion/deletion) polymorphism and both handgrip strength and vertical jump in females where homozygotes for the I-allele exhibiting higher performance-related phenotype scores. However, they did not find a significant association in the males for either performance measure. A study by van Rossum et al. [200] revealed that carriers of the ER22/23EK polymorphism in the glucocorticoid receptor gene were taller, had more lean body mass, greater thigh circumference, and more muscle strength in arms and legs compared to
98
C. Gray-McGuire et al.
noncarriers in males at 36 years of age. In females at the age of 36 years, waist and hip circumferences were smaller in ER22/23EK carriers, but there was no difference in body mass index compared to noncarriers. A linkage study by Huygens et al. [86] was performed in 329 young male sibs from 146 families using a polymorphic marker in RB1 (D13S153 on 13q14.2) for trunk strength. They reported evidence for linkage between locus D13S153 at 13q14.2 and several measurements of trunk flexion with LOD scores between 1.62 and 2.78, but did not find evidence for linkage with trunk extension. The next study by Huygens et al. [87] examined the potential role of the myostatin (GDF8) pathway in relation to muscle strength and estimated muscle cross-sectional area in 329 young male sibs with a candidate gene approach. They observed linkage patterns between knee extension and flexion peak torque with markers D2S118 (GDF8), D6S1051 (CDKN1A), and D11S4138 (MYOD1), and a maximum LOD score of 2.63 (P = 0.0002) was reported with D2S118.
2.3 Sleep Several recent studies have linked lack of sleep with obesity and other chronic health problems, yet millions of people do not get enough sleep and many suffer from sleep deprivation. Insufficient sleep makes an individual more vulnerable to obesity. The causes of sleep deprivation vary; the problems can directly or indirectly be linked to abnormalities in physiological systems such as the brain and nervous system, cardiovascular system, metabolic functions, and immune system. Furthermore unhealthy conditions, disorders, and diseases also contribute to sleeplessness. The percentage of adults getting <7 h sleep has risen from 15 to 39% with the median sleep time decreased from 8 to 7 h between 1959 and 2000 [138]. Much of the reduction in sleep time reflects changes in behavior including late night activities such as TV viewing, Internet usage (reported by 43% of adults), or need of late hour work (reported by 45% of adults). There is also increasing number of preadolescents and adolescents with sleep deprivation. More than 33% of teens were reported getting less than the recommended 9 h of sleep with objective evidence that they suffer rising daytime sleepiness [24]. It is believed that the social/school pressure along with biological changes in circadian rhythm and sleep architecture occurring in the developmental stage contribute to insufficient sleep in adolescents. Early school start times conflict with a delayed circadian phase that make occurrence of sleep deprivation most likely. In addition, the likelihood of insufficient sleep may be higher in certain ethnic groups. Data from the Cleveland Clinic indicate that African-American children are almost five times more likely to have delayed bedtimes compared to that of European Americans. African American children are the group most likely to have insufficient sleep [182]. Sleep deprivation has clear adverse effects on mood, neurocognitive function, and quality of life [48, 74, 83, 153, 178]. Three recent large epidemiological studies (Nurse Health Study, Massachusetts Male Aging Study, and Sleep Heart Health
4
Genetic Epidemiology of Obesity and Cancer
99
Study) [109, 213] indicate a 10–60% increased risk of developing glucose intolerance or DM2 in adults reporting <6 or 7 h of sleep per night with risk as high as 250% for those getting <5 h of sleep [64]. There is also an observed association between approximately 30% increased odds of coronary heart disease [10] and mortality [149, 211]. Other studies suggest that insufficient sleep increases the risk of obesity. In a study of 2,000 individuals aged >15 years from Spain, the result showed that obesity was significantly associated with decreased sleep duration of <6 h (odds ratio 1.67). With each additional hour of sleep time, the risk of obesity were estimated to be 24% lower for each individual [202]. The mechanistic bases for obesity and metabolic dysfunction occurring secondary to insufficient sleep are uncertain, but they are likely to include effects associated with sympathetic nervous system overactivity as well as hypothalamic pituitary–adrenal axis dysfunction. The latter effects are possibly associated with selective reduction of delta (slow wave) sleep. Experimental sleep deprivation of healthy young adults causes elevations of evening cortisol levels, reduced glucose tolerance, and an altered profile of growth hormone secretion [11, 106, 181, 198]. The impairment in glucose tolerance is thought to be secondary to enhanced sympathetic nervous system activation with associated secretion of insulin antagonists and beta-adrenergic stimulation of visceral fat and increased lipolysis. Experimental sleep restriction also affects the corticotrophin and somatotropin hypothalamic–pituitary axes with effects on eating, energy balance, and metabolism. The sleep deprivation is common in patients with sleep apnea. It is characterized by numerous involuntary breathing pauses during sleep which cause an obstruction to the sleep pattern. There are two types of sleep apnea: the first type is central sleep apnea that occurs when the brain fails to send the appropriate signals to the muscles to initiate breathing. Central sleep apnea is less common than obstructive sleep apnea. The second type is obstructive sleep apnea – occurs when air cannot flow into or out of the person’s nose or mouth, although efforts to breathe continue [102]. Sleep apnea affects 1–2% children and 4–15% of middle-aged adults [11, 157]. In particular, young African Americans have a four to six times higher risk of sleep apnea than European Americans [184]. Associations between sleep apnea, obesity, and metabolic syndrome are likely to reflect both causal and non-causal relationships. The data suggest that obese children like adults are approximately at four times higher risk to suffer from sleep apnea than non-obese individuals [160]. Obesity may precipitate or exacerbate sleep apnea by influence that are related to abdominal mass loading with effects on breathing pattern and responsiveness to respiratory drive or via upper airway at fat deposition compromising airway patency. Conversely, obesity may be exacerbated by the effects of sleep apnea due in part to reduced physical expenditure and associated mood and energy changes or possible apnea-induced dysregulation of leptin [149]. 2.3.1 Genes for Sleep Disturbances The demand for discovering genetic factors in sleep pathology has gained substantial attention in recent years. Early studies suggested that sleep disorders are
100
C. Gray-McGuire et al.
inherited [99]. Studies in twin pairs indicated that sleep architecture and some sleep disorders are inheritable through families. For example, monozygotic (MZ) twins showed similarities in sleep latency, duration of sleep cycles, and appearance of rapid eye movement (REM) sleep [207]. Furthermore, resemblance in alpha rhythms during waking was observed within MZ but not dizygotic (DZ) twins [78]. For sleep disorders, insomnia is typical of familial occurrence. A recent study discovered vulnerabilities in healthy probands who have family members diagnosed with insomnia [129]. Higher REM density was observed for both affected and non-affected family members. Although insomnia can be a manifestation prior to the onset of affective disorders that are also of genetic hereditary, other disordered sleep may also be influenced by genetic factors. Thus sleep regulation or dysregulation is likely to be associated with genetic control [99]. Several sleep disorders are heritable including restless leg syndrome (RLS), narcolepsy, sleep apnea syndrome, circadian sleep disorders, and Kleine–Levin syndrome (KLS). RLS is a sleep disorder with a strong genetic component. The syndrome is characterized by occurrence of an urge to move the lower limbs with unpleasant sensation at rest in the evening or at night. It was reported that 60–90% of patients have a positive family history [179, 212]. Additional investigation on MZ twins revealed 83% to be concordant for RLS. It supports the importance of a genetic contribution to the disease [145]. Narcolepsy, a human sleep disorder, causes excessive daytime sleepiness, cataplexy, hypnagogic hallucinations, and sleep paralysis. The disease occurs at irregular intervals and its pathophysiological association is to both environmental and genetic susceptibility factors interacting with each other. Familial cases for the disease are rare and a risk of 2% for narcolepsy development is observed in first-degree relatives. Up to one-third of MZ twins are concordant for narcolepsy. This suggests that non-genetic factors play a role in the etiology of the disorder [187]. Sleep apnea syndrome – obstructive sleep apnea is characterized by recurrent episodes of upper airway obstruction, snoring, and sleepiness. It is linked with other epidemics such as obesity, altered craniofacial morphology, upper airway tissue enlargement, and cardiovascular morbidity. Moreover, obstructive sleep apnea is also an independent risk factor for hypertension, myocardial infarction, and insulin resistance [99]. Numerous studies have found that a familial aggregation of some related conditions is involved in the pathogenesis of sleep apnea. For instance, an extensive study indicated that patients with specific craniofacial features have a familial aggregate [66]. The other two disorders, circadian sleep disorders and KLS, are very rare; only a few familial cases have been studied. Obesity is one of the most influential risk factors for sleep disorders especially sleep apnea. Studies of genetics of sleep apnea together with other related cardiovascular diseases (CVD), hypertension (HTN), and diabetes phenotypes provide considerable understanding of genetic and pathophysiological basis of inter-related attributes. In fact, there may be shared genetic pathways for sleep apnea and obesity-related phenotypes. Many of the important molecular pathways in obesity may also affect other risk factors for sleep apnea. Increasing evidence demonstrates that the overlapping pathophysiological pathways influence body weight,
4
Genetic Epidemiology of Obesity and Cancer
101
sleep regulation, circadian rhythm, and ventilation [27, 71, 81, 101, 140, 141, 197]. Sleep, respiratory, and metabolic functions are integrated through cholinergic, serotonergic, and orexinergic pathways located in hypothalamus and in the brain stem raphe nucleus. Examples include serotonin neurotransmission which affects weight through influences on appetite and thermoregulation and ventilatory control, sleep– wake regulation, and upper airway muscle tone. Other candidate genes for obesity, such as adenosine deaminase and melanocortin-3 receptor, in a variety of tissues are also important in breathing control [72]. Genes that affect sleep–wake and circadian rhythm may also influence obesity and metabolic syndrome such as deficiency in clock transcription factor, orexin, and leptin [71, 103]. Clock transcription factor is a key element of the molecular circadian clock within pacemaker neurons of the hypothalamic suprachiasmatic nucleus. It plays a role in regulating fat and glucose metabolism in adipose tissue, muscle, and liver. Orexins are a pair of neuropeptides that are expressed in specific neurons in the lateral hypothalamic area. It has a mediating role in the control of appetite and body weight. Lacking of orexin signaling is linked to the narcolepsy. Leptin, an adipocyte-derived cytokine produced in the white adipose tissue, which regulates body weight via appetite and thermogenesis, also influences lung growth, respiratory control, and sleep architecture and may also influence energy balance through additional mechanisms. Discussion of the obesity-related genes associated with sleep and circadian rhythm in mammalian species including humans is presented in the following studies. Animals with clock transcription factor deficiency develop a remarkably abnormal diurnal feeding rhythm, hormonal, and clinical changes consistent with the metabolic syndrome. As circadian clock is affected, the disruption of circadian rhythms may lead to obesity and metabolic disorders. An experiment was conducted on clock mutant and wild-type mice in the light–dark cycle controlled environment; the result revealed a significant increase in activity during the light phase and a change in the temporal pattern of total activity during dark phase. Particularly, wild-type mice showed two distinctive peaks of activity occurring after light off and before light on, and these peaks were attenuated in clock mutant mice [194]. The mutant mice also exhibited abnormal diurnal rhythmicity in food intake where only 53% of food intake occurred during the dark phase when compared to 75% in wildtype mice. The similar rhythm was seen in energy expenditure. Furthermore, clock mutant mice showed elevated energy intake and body weight on regular and high-fat diets. In mouse models, disruptive sleep–wake patterns and reduced physical expenditure are observed for orexin-deficient mice. They showed a phenotype similar to human narcolepsy including behavioral arrests, premature entry into REM sleep, and irregular sleep–wake behavior [71]. These may lead to the development of obesity. The physiologic mechanisms through which orexin deficiency reduces locomotor activity are demonstrated by a study design where wheel running activity and its relationship to sleep–wake behavior was examined between wild-type and orexin-knockout mice. The result suggests that orexin-deficient mice ran 42% less than wild-type mice. The running bouts were shorter for orexin-deficient mice and
102
C. Gray-McGuire et al.
they often experienced cataplexy or quick transition into sleep after running in comparison to wild-type mice [99]. Individuals with depletion in orexin develop symptoms of narcolepsy. Narcoleptic patients also have increased prevalence of obesity. In a structured assessment of symptoms for narcoleptic patients, the results suggested that narcolepsy patients are more obese in comparison to general populations. They showed elevated BMI and approximately 18% of the narcoleptic patients were above the 97th BMI percentile [39]. Further study also showed that the obesity and overweight occurred more often among narcoleptic patients with prevalence of nearly three-folds of healthy individuals [101]. Another sleep disorder associated to orexin genes is obstructive sleep apnea syndrome (OSAS). Comparison of plasma levels of orexin-A was conducted between 13 healthy controls, 37 OSAS patients, and 14 patients treated with positive airway pressure (CPAP). All OSAS patients and patients treated with CPAP were reported having lower orexin-A plasma levels than healthy individuals. It suggests that orexin-A plasma levels are abnormally low in patients with OSAS and do not change with patients receiving treatment [21]. Animals with leptin deficiency are hyperphagic and obese, they also have increased sleep fragmentation and impairments in sleep–wake regulations. A study was conducted to examine the effect of leptin deficiency on sleep architecture and homeostasis using the ob/ob mouse, a genetic model of leptin deficiency resulting from a spontaneous mutation in the gene (ob) encoding leptin [103]. Elevated sleep fragmentation was found in ob/ob mice compared to age-matched control group including increased number of arousals from sleep, frequent stage shifts, increased sleep bout numbers, and reduced sleep bout durations. Furthermore, significant increased amount of 24 h non-rapid eye movement (NREM) sleep time was observed for ob/ob mice. In comparison to wild-type mice, ob/ob mice had overall lower body temperature and locomotor activity counts. There is an attenuated diurnal rhythm of sleep–wake stages, NREM delta power, and locomotor activity for ob/ob mice. After sleep deprivation, ob/ob mice also exhibited smaller amounts of NREM and REM recovery sleep [52]. Studies of leptin have intensified with the discovery of the anti-obesity hormone, leptin, in the human body [125]. The leptin levels increase exponentially with fat mass. Higher circulating leptin levels are observed (i.e., 300% higher) in most obese individuals than lean individuals, and cerebrospinal fluid concentrations are only 30% higher in obese subjects. The implication of this is that leptin cerebrospinal fluid ratio is four to five times higher in lean than obese individuals, which suggests that limitations in the active receptor-mediated transportation of the hormone across the blood–brain barrier may contribute to leptin resistance and elevated leptin levels in obese subjects. Obesity is a leptin-resistant state in most circumstances due to possible receptor or post-receptor defects [125]. Another study also reported very low serum leptin levels in two obese children homozygous for a mutation in the gene for leptin who had a normal weight at birth but were obese at 3–4 months of age. This further suggests that leptin also critically influences energy balance in prepubertal individuals [15].
4
Genetic Epidemiology of Obesity and Cancer
103
While these three genes are widely discussed, there is a possibility that other genes may also influence the pathophysiology of major sleep disturbances [186]. Genetic control of sympathovagal balance is important in blood pressure and ventilatory control, thus genes that impact autonomic control may influence the expression of both hypertension and sleep apnea. Other genes that influence both traits through different mechanisms include endothelin-1 (ET-1), angiotensin-converting enzyme (ACE) gene, and hypoxia-inducible factor 1 (HIF-1). Regulation of appetite and food intake is influenced by circadian variation. Irregular circadian pattern has implications for obesity. For instance, the timed seasonal development of obesity in animals may be induced by biological clock [180]. An examination of association of obesity with gene variants and haplotypes of the linkage disequilibrium of the clock transcription factor showed four out of six SNPs (i.e., rs1554483, rs6843722, rs6850524, rs4864548) were significantly associated with overweight or obesity. Paired haplotypes including rs1554483 and rs4864548 had an effect on overweight or obesity. In addition, there were also significant differences in BMIs across genotypic groups. This result suggests the clock polymorphism and related haplotypes play an influential role in susceptibility to obesity. ET-1 is a potent and long-acting vasoconstrictor with mitogenic properties. The condition of hypoxia triggers the production of ET-1. Animals with intermittent hypoxia often had increased ET-1 and blood pressure [92]. Blood pressure decreased with an ET-1 blocker treatment. Subjects with untreated sleep apnea were accompanied by a significant increase in both blood pressure and ET-1, the symptoms were alleviated after treatment, whereas ET-1 levels did not change for healthy subjects under treatment. This suggests the effects of ET-1 may be implicated in the elevated blood pressure in individuals with sleep apnea. Angiotensin-converting enzyme (ACE) influences the metabolism of angiotensin II, and inactivation of bradykinins and tachykinins which are potent bronchial constrictors and mediators of inflammation asthma and ACE, is heavily expressed in lungs [99]. The polymorphisms in ACE gene vary among individuals, with different levels of ventilatory responses to hypoxia or high altitude known to be associated with severity of sleep apnea [78]. Individuals with mild to moderate degrees of sleep apnea are more likely to develop hypertension [129]. Hypoxia-inducible factor 1 (HIF-1) is a regulator of oxygen homeostasis. It plays an important signaling role in hypoxia-mediated blood pressure responses. HIF-1 heterozygous mice have completed blunted hypertensive responses to intermittent hypoxemia compared to wild-type mice [71]. Therefore, the variability of susceptibility of hypertension among subjects with sleep apnea may be related to variation in HIF-1 activity. Further, sleep disturbances that alter the expression of hypertension and obesity phenotypes may be gene-by-environment effects. Examples include uncoupling protein and peroxisome proliferator-activated receptor alpha (PPARγ). Uncoupling proteins (UCPs) include UCP1 which has a thermogenic role in the brown adipose tissue, UCP2 is ubiquitous, UCP3 is expressed in skeletal muscle, and others such as UCP4 and UCP5 are expressed in the brain [118]. Uncoupling proteins
104
C. Gray-McGuire et al.
mediate the mitochondrial proton leak which accounts for a significant component of resting energy expenditure. Polymorphism in UCP genes may have a magnifying impact on these genes from sleep disorders. As UCP1 is only expressed in brown adipose tissue, resting energy expenditure is not affected solely by UCP1. Expression of both UCP2 and UCP3 was increased in mice with lack of sleep. In particular, UCP2 may be important for changes in energy expenditure related to sleep loss. Since a high magnitude of elevated UCP2 expression was observed in the liver and skeletal muscle of long-term sleep-deprived mice compared to controls and skeletal muscle is the largest tissue in the body, UCP2 is likely to affect the overall resting energy expenditure and may thus contribute its increase after sleep deprivation [32]. PPARγ is a member of the nuclear hormone receptor family which influences insulin sensitivity, glucose metabolism, and blood triglyceride levels [37]. Hypoxia inhibits the production of PPARγ, and polymorphisms in PPARγ may variably influence diabetes susceptibility in individuals with different backgrounds of hypoxic exposure as would occur in sleep apnea. A cross-sectional study on the relationship of PPARγ (Pro12Ala) with obstructive sleep apnea (OSA) in Asian Indians showed that the frequency of the ‘Ala12’ allele in PPARγ (Pro12Ala) was significantly higher in obese individuals with OSA when compared to obese individuals without OSA. The result suggests that there is an association between PPARγ gene polymorphisms and OSA, independent of obesity. This gene may serve as a marker for the development of OSA in obese subjects [176]. Occurrence of hypertension accompanied by sleep apnea is due to the effects of hypoxia and arousal on the stimulation of sympathetic nervous system activity, leading to a cascade of inflammatory and hormonal responses. However, there is variable individual susceptibility to sleep apnea-related hypertension. Compelling data show an association of OSA with hypertension especially in obese individuals. OSA does in fact contribute to the obesity-related hypertension. A study (Wisconsin Sleep Cohort Study) reported an independent dose–response relation between sleepdisordered breathing at baseline and the development of hypertension 4 years later. The likelihoods of incident hypertension for sleep apnea individuals at follow-up were 1.42, 2.03, and 2.80 times more than at baseline [150]. The effective treatment of OSA with continuous positive airway pressure results in a decrease in both daytime and nighttime blood pressure. This further supports the causal relationship between OSA and chronic hypertension [54, 12, 151]. The stresses imposed by obesity may interact with genetic loci to exacerbate the development of sleep apnea. The effect of genetic polymorphisms is especially apparent in obese subjects where the fat deposition in upper airway precipitates the expression of elevated laxity of connective tissue. Obesity of the centripetal type where adipose tissue is distributed predominately in the regions of the abdominal viscera, upper body, and neck is significantly more associated with OSA than the peripheral pattern of obesity where the distribution of fat is found in the hip and thighs. Neck circumference is believed to be a strong predictor of sleep apnea [40]. Studies have shown that significant weight loss of approximately 10% is associated with varying degrees of improvement in sleep apnea. OSA patients with 17%
4
Genetic Epidemiology of Obesity and Cancer
105
decrease in body weight had significant reduction in airway collapsibility (greater than 50% decrement in disordered breathing). The evidence suggests that adipose tissue within the lateral pharyngeal walls, tongue, uvula, and under the mandible decreases with weight loss and it contributes to the overall increase in volume of the upper airway. Therefore, obesity may affect the tissue surrounding the airway and thereby alter the airway biomechanical properties that give rise to sleep apnea [102].
2.4 Viral Infection, Thermoregulation, and Other Genetically Influenced Risk Factors In recent years viral infections have been recognized as possible contributors to obesity. Viruses that cause an increase of adiposity in animals are canine distemper virus, Rous-associated virus, Borna virus, an avian adenovirus, SMAM-1, and human virus adenovirus 36 (Ad-36). Among those, Ad-36 and SMAM-1 are linked with human obesity. In particular, the common human virus Ad-36 increases the replication, differentiation, lipid accumulation, and insulin sensitivity in fat cells and reduces those cells’ leptin secretion and expression [199]. In animal models, it was found that animals inoculated with Ad-36 developed a syndrome of increased adipose tissue and paradoxically low levels of serum cholesterol and triglycerides [44, 45]. In non-human primate studies, occurrence of AD-36 antibodies in 15 male rhesus monkeys was observed and a significant longitudinal association of positive antibody status with weight gain and lowered plasma cholesterol. Furthermore, there was a three-fold body weight gain, a greater fat gain, and lower serum cholesterol for three male marmosets inoculated with Ad-36 compared to three uninfected controls in a randomized controlled experiment. These experiments demonstrated that Ad-36 has an adiposity promoting effect on non-human primate species [46]. The mechanistic basis of Ad-36 in human obesity is unknown. Due to ethical reasons, the infection could not be performed on human subjects. However, two observational studies were conducted to determine the prevalence of Ad-36 antibodies in obese and non-obese humans and examine the association of Ad-36 antibodies with body mass index (BMI) and serum lipids. The result of the first study indicated that there is a significant association of obesity and positive Ad-36 antibody status independent of age, gender, and collection site. The prevalence of Ad-36 antibodies is almost three-fold higher in obese versus non-obese subjects. Lower serum cholesterol and triglycerides were observed in individuals with positive Ad-36 antibody compared to individuals with negative Ad-36 antibody. The second study demonstrated that for twin pairs, antibody-positive twins had higher BMI and percentage body fat compared to antibody-negative twins. No apparent association of Ad-2, Ad-31, or Ad-37 antibodies with BMI or serum lipids was present. Hence there is an association of Ad-36 with the increase in body weight and lower serum levels in humans [9].
106
C. Gray-McGuire et al.
Another risk factor such as thermoregulation may also be associated with obesity. Particular attention was placed on reduction in variability in ambient temperature. Thermoregulation is defined as the ability to maintain body temperature in a safe range generally between 97 and 99 F. Internal and external factors affect the body temperature of an individual. Internal factors include physical activity, peripheral circulation, amount of subcutaneous fat, metabolic rate, amount and type of food and fluids ingested. External factors include environmental temperature, humidity, air movement, and amount and type of clothing [61]. With the technological advancement, heaters/air conditioners are readily available. Individuals are now living more in the thermoneutral zone (TNZ) than they did 30 years ago. For instance, the average internal UK home temperature increased from 13 to 18◦ C between 1970 and 2000. The US thermal standard for winter comfort increased from 18 to 24.6◦ C from 1923 to 1986. The percentage of US home with central air conditioning increased from 23 to 47% between 1978 and 1997. A two-fold increase in the percentage of homes with air conditioning is observed in the southern United States with some of the highest obesity rates. The thermoneutral zone, by definition, is the range of ambient temperature in which energy expenditure is not required for homeothermy. Energy expenditure increases with ambient temperature above or below TNZ, the energy stored decreases. This effect was evidenced in short-term controlled human experiments and decreases in adiposity were observed in controlled animal experiments [96]. Increased exposure to TNZ triggers reduction in body’s ability to regulate its temperature which promotes adiposity [118]. Some other interesting plausible contributors to obesity include but not limited to pharmaceutical iatrogenesis, changes in distribution of ethnicity and age, increase of Gravida age, and decreased smoking [96]. However, only selective factors with the strongest evidences were discussed in the previous section. Environmental factors do certainly have a profound effect on obesity. Considerable evidence indicates associations between obesity and cancers. In 2002, about 41,000 new cases of cancer in the United States were estimated to be due to obesity [155]. This suggests that about 3.2% of all new cancers are linked to obesity. While the mechanisms underlying the obesity–carcinogenesis relationship are not fully characterized, sufficient evidence exists to support recommendations that individuals maintain reasonable weight and healthy lifestyles for multiple health benefits including decreasing their risk of cancer.
2.5 Stomach Flora The two predominant populations of microbiota in both the mouse and human gut are members of the bacterial groups known as the Firmicutes and the Bacteroidetes. Ley et al. [107] found that genetically obese mice (ob/ob) had 50% fewer Bacteroidetes and correspondingly more Firmicutes than their lean wild-type sibs. They also showed that the relative proportion of Bacteroidetes was decreased in
4
Genetic Epidemiology of Obesity and Cancer
107
obese people in comparison with lean people, and that this proportion increased with weight loss on two types of low-calorie diet. They concluded that obesity has a microbial component, which might have potential therapeutic implications. Turnbaugh et al. [196] demonstrated through metagenomic and biochemical analyses that these changes affect the metabolic potential of the mouse gut microbiota. Their results indicated that the obese microbiome has an increased capacity to harvest energy from the diet. Furthermore, this trait was transmissible: colonization of germ-free mice with an ‘obese microbiota’ resulted in a significantly greater increase in total body fat compared to colonization with a ‘lean microbiota.’ They concluded that their results identified the gut microbiota as an additional contributing factor to the pathophysiology of obesity.
3 Current Models of Shared Genetic Pathways Between Cancer and Obesity Cancer or malignant neoplasm is a genetic term for the rapid creation of abnormal cells, which grow beyond their usual boundaries. It has been the second leading cause of death in the United States since 1999 exceeded only by heart disease. Globally, the World Health Organization (WHO) estimates that cancer kills roughly 7.6 million people annually [220]. Many forms of cancer occur at various places in the body. The origins of cancer are complex and their exact causes are still unknown; however, scientists have found that the development of cancer is influenced by several acquired risk factors that include environmental exposures and genetic contributions. It has been estimated that 15–20% of all cancer deaths in the United States can be attributed to being overweight and obese [22]. The genetic complexity of cancer shares the common metabolic pathways for several common traits with obesity. Thus many genetic studies have attempted to determine the linkage between this complex human disease and the common traits with obesity. There has been increasing evidence of obesity and metabolic dysfunction portending a substantial increase in cancer morbidity and mortality. It is therefore important to identify the sources of genetic variation for these comorbidities with cancer to elucidate the important pathophysiological pathways and pave the way for targeted intervention and prevention. Obesity is associated with a wide range of metabolic and hormonal effects, reflecting an active role of adipose tissue in mediating cytokine, growth factor, and sex steroid levels [166, 90, 98]. Many of these cytokines play a key role in the pathogenesis of diabetes, autoimmune diseases, and many types of cancer. These cytokines and growth factors interactively modify the actions and secretion of insulin, which, in turn, influences the metabolism of fatty acids and expression of growth factors and inflammatory cytokines which includes leptin, interleukin (IL) IL-6, resistin, tumor necrosis factor (TNF)α, vascular endothelial growth factor (VEGF) (a potent stimulus of angiogenesis [190], and adiponectin [158, 64]). Several hormones and growth factors serve as intermediate and long-term
108
C. Gray-McGuire et al.
communicators of nutritional state throughout the biosystem and have been implicated in both energy balance and carcinogenesis. The increasing rate of obesity among children is especially alarming and suggests continuing increases in the rates of obesity-related cancers for many years to come [85]. Increasing animal and human data suggest that obesity has roles in immune regulation, neovascularization, apoptosis, cellular proliferation, and estrogen biosynthesis and can influence the development or expression of various cancers, including colon, breast, endometrium, renal cell, esophagus, and prostate [166, 90, 98]. Since being overweight in childhood predicts obesity in adulthood [209], obesity occurring in childhood may operate as a ‘cumulative’ exposure that influences cancer risk in later life. The Harvard Growth Study reported 55-year follow-up data on 508 adolescents 13–18 years of age, showing that overweight adolescent boys (but not girls) had nine times increased risk of colorectal cancer mortality [136]. Associations between adolescent weight and cancer persisted even after adjusting for adult BMI. A 50-year follow-up of >2,000 British children showed an overall 9% increase in cancer incidence per one unit increase in BMI standard deviation, with effects three times larger for smoking-related cancers [88]. An Israeli case– control study reported that being in the upper quartile of BMI at 18 years of age was associated with a 42% increase for ovarian cancer [119]. Even less research has examined the association between biomediators in childhood and cancer risk, particularly in conjunction with genetic data. One case–control study of 40 obese and 40 non-obese prepubertal children showed increased IGF-1 and insulin and lower sex-binding hormone in the obese group [58], suggesting that high levels of growth factors and altered sex hormone profiles are present in obese children and that exposures to an adverse metabolic milieu may begin early in life. The two potential mechanisms of energy balance which links obesity to cancer is mediated through hormones, growth factors, and inflammatory pathways. The hormones and growth factors that serve as intermediate and long-term communicators of nutritional state throughout the biosystem include insulin growth factor (IGF-I), insulin, adiponectin and leptin, sex steroid hormones, as well as several factors associated with inflammation and oxidative stress [84]. This section will review common genetic pathways that link cancer to obesity.
3.1 Networks of Insulin Growth Factors in Cancer and Obesity Link IGF-I has been implicated in transformation, cell migration, and propensity for metastasis in vivo colorectal cancer [154]. It has been associated with colon cancer risk in men [123] and endometrial cancer [122]. Insulin-like growth factor (IGF) bioavailability is increased in obesity, perhaps partly mediated by high insulin levels. Insulin increases hepatic production of IGF-1 and reduces insulin-like growth factor-binding protein (IGFBP-1), together resulting in an increased bioavailability of IGF-1. Animal studies suggest that high levels of insulin and IGF-1 and
4
Genetic Epidemiology of Obesity and Cancer
109
low IGFBP-1 promote tumorigenesis via effects on apoptosis and cell proliferation [60, 98, 62]. Human work indicates that high insulin and/or IGF-1 levels increase propensity for colonic polyps and cancers of the breast (pre-menopausal), pancreas, colon, kidney, and endometrium, with associations that persist after adjustment for BMI [91, 121, 123, 172, 70, 104, 122, 183, 191, 26, 147]. Cancer mortality also has been associated with high insulin and/or IGF-1 levels [62]. Understanding the association between growth factors and cancer risk, however, requires consideration of differences in biomediator availability (e.g., IGF-1-binding patterns), which vary markedly by sex and ethnicity [43]. Given the potential role of IGF-1 levels in the development of neoplastic disease and the variable prognosis of cancers across ethnic groups, it is important to identify factors that modify IGF-1 expression across population subgroups. Insulin-like growth factors (IGFs) play a fundamental role in somatic growth, cellular differentiation, metabolism, and survival. It is said that the IGF-1-like signaling pathway postpones or attenuates cancer processes. Longo et al. reported that the ability of GH (growth hormone) and IGF-1 to lower antioxidant defenses in hepatocytes suggested that IGF-1 can promote cellular damage and disease in mammals [114]. Thirty years ago, Silberberg showed that Pit-1 dwarf mice, which are deficient in plasma GH and IGF-1 [177], had less osteoarthritis than wild-type mice. Since then, high levels of IGF-1 have been associated with an increase risk of several human diseases including breast, lung, colorectum, and prostate cancer [1]. IGF-1 appears to also promote cancer in mice as tumors in Pit-1 or Prop-1 dwarf mice are either reduced or delayed. The dwarf mice with elevated GH and IGF-I exhibited severe kidney lesions and a shorter life span [36, 177]. Another network of pathways that regulate the cellular growth and metabolism is the PI3K/Akt pathway, which induces downstream targets of the IGF-I receptor and insulin receptor [23]. Activation of the PI3K/Akt pathway affects cell growth, proliferation, survival, and metabolism. This pathway is most commonly altered in human epithelial tumors [85]. Activation of the receptor tyrosine kinases (RTKs) and/or Ras oncogene stimulates PI3K to produce the lipid second messenger, phosphatidylinositol 3,4,5-triphosphate (PIP3). PIP3 recruits and anchors Akt to the cell membrane where it can be further phosphorylated and activated [73, 77, 217]. The Akt signaling cascade has been shown to be associated with an elevation in the mTOR signaling pathway. The mTOR (mammalian target of rapamycin) is nutrient pathway that regulates cellular growth. mTOR is a highly conserved serine/threonine protein kinase which is activated by Akt and inhibited by an opposing signal from AMP-activated kinase (AMPK). mTOR dictates translational control of new proteins in response to both growth factor signals and nutrient availability through phosphorylation of its downstream mediators, S6K and 4EBP-1. The activation of mTOR results in cell growth, cell proliferation, and a resistance to apoptosis [85]. Another key regulator of cellular growth is the AMP-activated protein kinase (AMPK) pathway. AMPK acts as a sensory mechanism for cells to process signals of energy consumption. Storage of AMP/ATP ratio would activate AMPK, which inhibits energy-consuming processes and activates energy-producing processes to
110
C. Gray-McGuire et al.
restore energy homeostasis [205]. Energy balance is highly influenced by the Atk and the AMPK pathways by activation of mTOR. This energy balance functions to reduce cellular energy processes and protects against stress-induced apoptosis. There has been evidence of a variety of cancers that have overactivated mTOR signaling [171]. Multiple tumor suppressor genes such as LKBI, PTEN, and TSC2/1 have been identified to be negative regulators of mTOR signaling, which makes AMPK a possible tumor suppressor [205]. PTEN is a phosphatase and tensin homolog deleted on chromosome TEN that interacts with the PI3K signaling cascade. Its relevance has been shown in cancer from three independent groups who characterized PTEN being highly mutated and lost in several cancers [108].
3.2 Role of Cytokines and Hormones in Cancer and Obesity Link Leptin is a peptide hormone secreted from adipocytes that is involved with appetite control and energy metabolism through its hypothalamic influence [85]. In the nonobese state, rising leptin levels result in decreased appetite and increased energy metabolism through a series of neuroendocrine changes. The obese state is associated with high circulating levels of leptin [115, 131, 219, 223], suggesting that the obese may develop leptin resistance. The leptin pathway is involved in multiple signaling cascades that regulate the downstream signals of cancer genes to be turned on or off. Adiponectin is a 28-kDa peptide hormone produced exclusively by adipocytes and intimately involved in the regulation of insulin sensitivity and carbohydrate and lipid metabolism. The link between adiponectin and cancer risk is not well characterized, although there is a report that adiponectin infusion inhibits endothelial proliferation and inhibits transplanted fibrosarcoma growth. It has been shown that leptin and adiponectin interact antagonistically to influence carcinogenesis, although this interaction is not well established [85]. Other factors like sex steroid hormones which include estrogens, androgens, and progesterone are likely to have a role in the relationship between energy balance and certain types of cancer as does aromatase, also called estrogen synthetase, a cytochrome P450 of aromatic C18 estrogens from C19 androgens. However, details of the role of these agents in the relationship between obesity and cancer are beyond the scope of this chapter.
4 Cancer-Specific Shared Genetic Effects Since there are multiple pathways that link obesity and cancer, there are many overlaps among the specific cancers. This next section will highlight some of the genetic studies that may link these pathways.
4
Genetic Epidemiology of Obesity and Cancer
111
4.1 Stomach Gastric cancer is the second leading cause of cancer death worldwide [148]. Gastrointestinal malignancies may be associated with obesity. Mechanisms that are associated with obesity include a particular metabolic state characterized by hyperinsulinemia or insulin resistance along with elevated serum leptin. Leptin is derived from adipocytes and appears to play a role in the regulation of ghrelin, a peptide derived from the stomach and small intestine that stimulates appetite and weight gain. Other studies have shown androgens and estrogens playing a role in gastric cancer etiology. Freedman et al. investigated the association of gastric cancer with single nucleotide polymorphisms (SNPs) in six genes (COMPT, CYP1B1, CYP17A1, CYP19A1, HSD17B1, and SHBG) involved in estrogen and androgen synthesis and metabolism, with 58 haplotype-tagging SNPs genotyped in 295 gastric cancer cases and 415 controls from a population-based study in Poland. The study found that polymorphisms in COMPT, involved in estrogen inactivation, and SHBG, a modulator of hormone bioavailability, are associated with gastric cancer. However, these studies will need to be confirmed with additional studies due to possible selection bias [57].
4.2 Endometrial Endometrial cancer affects more than 40,000 women a year in the United States [4]. Prolonged exposure to estrogens unopposed by progesterone plays an important role in the etiology of endometrial cancer [22, 89]. While the association of this disease with high body mass index and sex steroid hormones is well known, there are many questions about etiology that have not been resolved. Little is known about the genetic basis for risk associated with hormones or obesity, other common genetic factors associated with risk, or gene–environment interactions. Studies have shown that high endogeneous levels of estrogens are related to increased risk of endometrial cancer [6, 121, 156, 222]. There are many candidate genes that are a plausible risk variant for endometrial cancer such as MYH; however, there are two known candidate genes, CYP19A1 [174] and CYP17A1 [13], associated with environmental risk factors such as physical activity and reduced risk of endometrial cancer, especially in woman who are older and overweight. Aromatase, encoded by CYP19A1, converts androstenedione to estrone and testosterone to estradiol. After menopause, the primary source of estrogens is via peripheral conversion of androgrens in adipose tissue catalyzed by aromatase. Given its key role in estrogen biosynthesis, it is possible that polymorphisms in CYP19A1 that alter estrogen production could be involved in endometrial carcinogenesis. Factors influencing circulating estrogen levels, insulin-mediated pathways, or energy balance through obesity-related mechanisms, such as physical activity, have been proposed as potential risk factors for endometrial cancer [174]. A prospective study using the American Cancer Society Cancer Prevention Study II Nutrition Cohort examined measures of physical activity in relation to
112
C. Gray-McGuire et al.
endometrial cancer risk using information obtained at baseline in 1992. From 1992 to 2003, 466 incident endometrial cancers were identified among 42,672 postmenopausal women with intact uteri who were cancer-free at enrollment. The inverse relationship was seen only among overweight or obese women (trend P = 0.003) and not in normal weight women (trend P = 0.51). In summary, light and moderate physical activity including daily life activities were associated with lower endometrial cancer risk in this study, especially among women who are overweight or obese [174].
4.3 Breast, Premenopausal Breast cancer is the most common cancer in women, with nearly 180,000 women diagnosed with breast cancer in the United States annually [4]. There are several plausible mechanisms linking obesity to breast cancer risk. These mechanisms have evolved from a focus on estrogen excess to the combined effects of estrogen and progesterone and, most recently, attempts to understand the factors defining the bioavailability and effects of estrogen and androgens and their metabolites on specific end organs. Increases in obesity have been associated with increases in androgens, triglycerides, and insulin and decreases in SHBG [38]. These hormonal changes increase the bioavailability of estradiol and its metabolites and may also directly promote tumor growth [17, 18, 53, 67, 95, 100, 156]. The bioavailability of estradiol is dependent on the degree and strength of binding to several protein carriers. SHBG is the predominant carrier of estradiol and the percentage of free estradiol is inversely related to the level of SHBG [65]. Increases in free fatty acids, such as triglycerides, have been reported to increase the level of free estradiol by displacing estradiol from SHBG [65]. Therefore, both decreases in SHBG and increases in triglycerides may result in increases in free estradiol. Key and Pike first hypothesized that the effect of adiposity on the estrogen bioavailability was modulated by menopausal changes in estrogen and progesterone production and as a result explained the contradictory findings for premenopausal and postmenopausal breast cancer [97]. Before menopause, ovarian production of estrogen overwhelms changes in estrogen metabolism related to the overall level of adiposity. As a result, estradiol in ovulatory cycle does not differ significantly in obese compared with lean women. However, estradiol levels are reduced in anovulatory cycles that are more frequently in obese than lean premenopausal women. In addition, obese premenopausal women have been found to have markedly reduced progesterone levels, both due to anovulations and decreased production during the luteal phase of the menstrual cycle. After menopause, the lower risk associated with premenopausal obesity diminishes over time. In addition, in postmenopausal women, the overall level of adiposity results in increased estrogenic activity due to an increase in estrogen production from the aromatization of higher levels of androgens in adipose tissue [100], decreased estrogen binding [20] due to decreases in SHBG [18, 95, 134], and increases in triglycerides [17]. Furthermore, insulin and insulin-like growth factors have been found to promote cancer cell growth, and their production
4
Genetic Epidemiology of Obesity and Cancer
113
can be increased by estrogen [33]. However, although five studies have shown that higher IGF-I levels are associated with a higher risk for breast cancer [19, 42, 70, 152, 191], only one of these studies found this association to be statistically significant [19]. In addition, published studies, thus far, have not found an association between IGF-I and postmenopausal breast cancer. IGF-1 has been shown to be positively associated with muscle mass, may be positively related to increased premenopausal breast cancer risk among tall and lean women. Leptin is characterized as a growth factor for breast cancer. Thus far, there has only been one study showing the relationship among premenopausal women finding nonsignificantly lower level of leptin in breast cancer cases compared to controls. Breast cancer results from complex interactions among genetic, hormonal, and environmental factors. Thus, knowledge of genetic risk factors that contribute to breast cancer is crucial to design both preventative and therapeutic strategies and to identify at-risk individuals, in order to reduce the incidence of and death from this disease.
4.4 Breast, Postmenopausal Obese women have an increased risk for postmenopausal breast cancer. The association of the overweight and obesity with breast cancer risk has been well studied. However, the physiological mechanism by which obesity contributes to breast tumorigenesis is not understood. In addition to the well-characterized BCRA1/BCRA2 genes that are associated with breast cancer, the HCCR-1 oncogene which contributes to breast tumorigenesis as a negative regulator of p53 and detection of HCCR-1 serological levels was useful for the diagnosis of breast cancer. In this study, we found that the HCCR-1 level is elevated in breast cancer tissues and cell lines compared to normal breast tissues. We identified apolipoprotein E (ApoE) interacting with HCCR-1. Our data show that HCCR-1 inhibits the anti-proliferative effect of ApoE, which was mediated by diminishing ApoE secretion in breast cancer cells. Finally, HCCR-1 induced the severe obesity in transgenic mice. Those obese mice showed severe hyperlipidemia. In conclusion, our results suggest that HCCR-1 might play a role in the breast tumorigenesis while the overexpression of HCCR-1 induces the obesity probably by inhibiting the cholesterol-lowering effect of ApoE. Therefore, HCCR-1 seems to provide the molecular link between the obesity and the breast cancer risk. There is convincing evidence supporting a probable preventive role for physical activity in postmenopausal breast cancer. The mechanisms by which long-term physical activity affects risk, however, remain unclear. The aims of this review were to propose a biological model whereby long-term physical activity lowers postmenopausal breast cancer risk and to highlight gaps in the epidemiologic literature. To address the second aim, we summarized epidemiologic literature on 10 proposed biomarkers, namely body mass index. Estrogens, androgens, sex hormone-binding globulin, leptin, adiponectin, markers of insulin resistance, tumor necrosis factoralpha, interleukin-6, and C-reactive protein, in relation to postmenopausal breast
114
C. Gray-McGuire et al.
cancer risk and physical activity, respectively. Associations were deemed ‘convincing,’ ‘probable,’ ‘possible,’ or ‘hypothesized’ using set criteria. Our proposed biological model illustrated the co-occurrence of overweight/obesity, insulin resistance, and chronic inflammation influencing cancer risk through inter-related mechanisms. The most convincing epidemiologic evidence supported associations between postmenopausal breast cancer risk and BMI, estrogens, and androgens, respectively. In relation to physical activity, associations were most convincing for BMI, estrogen, insulin resistance, and C-reactive protein. Only BMI and estrone were convincingly (or probably) associated with both postmenopausal breast cancer risk and physical activity. There is a need for prospective cohort studies relating the proposed biomarkers to cancer risk and for long-term exercise randomized controlled trials comparing biomarker changes over time, specifically in postmenopausal women. Future etiologic studies should consider interactions among biomarkers, whereas exercise trials should explore exercise effects independently of weight loss, different exercise prescriptions, and effects on central adiposity.
4.5 Colorectal Colorectal cancer is the second most common cause of death and illness in developed countries. Previous reviews have suggested that obesity may be associated with 30–60% greater risk of colorectal cancer, but little consideration was given to the possible effect of publication bias on the reported association. Random-effects meta-analyses were done, involving 70,000 cases of incident colorectal cancer from 31 studies, of which 23 were cohort studies and 8 were case–control studies. Dietary and lifestyle factors play a large role in colorectal cancer (CRC) etiology and may account for a majority of CRC cases. In particular, high intakes of fats, red/processed meats, and low intake of dietary fiber, fruits and vegetables, accompanied with physical inactivity and overweight are thought to be associated with increased risk of colon and rectal cancers. Such regimens have major metabolic consequences, including the development of insulin resistance and hyperinsulinemia. However, insulin is the key to regulation of energy metabolism and may directly influence CRC risk via mitogenic/anitiapoptotic effects through the insulin receptor or by increasing energy provision to colon and rectal cancer cells with growth-promoting consequences. Indirectly, insulin may also increase the bioactivity of insulin-like growth factor I (IGF-I) by reducing the levels of two IGF-binding proteins (IGFBP), IGFBP-I and IGFBP-2, although total IGF-I concentrations in blood are primarily associated with increased CRC risk. Thus, chronically elevated fasting and postprandial insulin levels may be closely associated with increased colon and rectal cancer risk. Indeed, recent reviews suggest that prevention of weight gain and a greater level of physical activity, both of which inhibit loss of insulin sensitivity and tend to lower blood insulin levels, may help to reduce the risk of colon cancer, although evidence for an association with rectal cancer is currently insufficient. Evidence for the association of blood insulin levels with CRC risk comes from both experimental and epidemiologic studies, showing a higher
4
Genetic Epidemiology of Obesity and Cancer
115
CRC risk in type II diabetics and in individuals presenting risk factors are thought to be related to chronic hyperinsulinemia. However, only a few prospective studies have assessed baseline levels of circulating insulin in association with CRC risk. In a small cohort of elderly subjects, Schoen et al. showed an almost two-fold increased CRC risk with higher nonfasting insulin levels 2 h after a glucose challenge, while in a Swedish population, baseline insulin levels have been associated with a nonstatistically significant positive risk of colon and rectal cancers. Some other prospective studies have utilized C-peptide, which is considered as a valid marker of pancreatic insulin secretion because it has a longer half-life than insulin itself. A study based on a cohort of New York women showed that higher C-peptide levels were significantly associated with increased CRC risk, which was even stronger when analyses were limited to the colon. Results from the North American cohorts have also shown a positive association between C-peptide and CRC risk, which was statistically significant in men (effects similar in the colon and rectum) but not in women (effect appeared stronger in the colon, but was not statistically significant). Similarly, a Norwegian study has shown a nonstatistically significant positive cancer risk association in the colon and no association in the rectum.
4.6 Esophagus The incidence of esophageal adenocarcinoma (EADC) is rapidly increasing in Western countries and obesity is thought to be a major risk factor. Previous research on the association between BMI and EADC and gastric cardia adenocarcinoma has relied almost exclusively on case–control studies because the low incidence rates have precluded accruing sufficient case numbers in most cohorts. To date there has been three documented prospective studies that attempted to examine the association between BMI and EADC and gastric cardia. One of the three studies did not properly control for important potential confounders [51] such as cigarette smoking, and the other two had incomplete information on confounders [112, 167]. A study by Abnet et al. prospectively examined the association between BMI and EADC, gastric cardia adenocarcinoma, and gastric noncardia adenocarcinoma in a cohort of approximately 500,000 people in the United States (NIH-AARP Diet and Health Study Cohort). The study results showed that compared to people with a BMI of 18.5–25 kg/m2 , a BMI ≥ 35 was associated with significantly increased risk of EADC, HR (95% CI) = 2.27 (1.44–3.59), and gastric cardia adenocarcinoma 2.46 (1.60–3.80), but not gastric noncardia adenocarcinoma 0.84 (0.50–1.42) [2].
4.7 Prostate Prostate cancer is the leading cancer diagnosed among men in the United States. There has been evidence that androgen plays an important role in prostate cancer growth, proliferation, and progression.
116
C. Gray-McGuire et al.
A study by Huggins and Hodges was the first study to show that marked reductions in serum testosterone levels by castration or high-dose estrogen therapy resulted in the regression of metastatic prostate cancer. This work and many others lead to the belief that higher testosterone levels caused enhanced growth of prostate cancer and resulted in the use of androgen ablation therapy for the management of prostate cancer [175]. In men, testosterone is synthesized primarily in the testes and, to some extent, in the adrenal glands. In the circulation, about 45% of the total testosterone binds to sex hormone-binding globulin (SHBG), about 50% binds loosely to albumin, and <4% is unbound free testosterone [29, 30, 83]. The genetic susceptibility at several loci on chromosome 8q24 that is approximately 300 kb from the MYC gene were identified in two independent studies as plausible risk loci for prostate cancer. Although it is uncertain how much these risk variants influence the risk of prostate cancer, these findings provided evidence of how androgen biosynthesis and metabolism may affect the MYC gene which may be mediating communication between cascades of pathways and the androgen receptor.
References 1. Abbott B, Abolins M, Abramov V, Acharya BS, Adams DL, Adams M, Ahn S, Akimov V, Alves GA, Amos N and others (2000). Measurement of the W–>taunu production cross section in pp collisions at square root s=1.8 TeV. Phys Rev Lett, 84(25):5710–5. 2. Abnet CC, Freedman ND, Hollenbeck AR, Fraumeni JF, Jr., Leitzmann M, Schatzkin A (2008). A prospective study of BMI and risk of oesophageal and gastric adenocarcinoma. Eur J Cancer, 44(3):465–71. 3. Abraham MR, Olson LJ, Joyner MJ, Turner ST, Beck KC, Johnson BD (2002). Angiotensinconverting enzyme genotype modulates pulmonary function and exercise capacity in treated patients with congestive stable heart failure. Circulation, 106(14):1794–9. 4. ACS (2008). Cancer Facts and Figures 2008. American Cancer Society. 5. Ahmetov, II, Mozhayskaya IA, Flavell DM, Astratenkova IV, Komkova AI, Lyubaeva EV, Tarakin PP, Shenkman BS, Vdovina AB, Netreba AI and others (2006). PPARalpha gene variation and physical performance in Russian athletes. Eur J Appl Physiol, 97(1):103–8. 6. Allen NE, Key TJ, Dossus L et al. (2008). Endogenous sex hormones and endometrail cancer risk in women in the European Prospective Investigation into cancer and Nutrion (EPIC). Endocr Relat Cancer, 15:485–497. 7. Alvarez R, Terrados N, Ortolano R, Iglesias-Cubero G, Reguero JR, Batalla A, Cortina A, Fernandez-Garcia B, Rodriguez C, Braga S and others (2000). Genetic variation in the reninangiotensin system and athletic performance. Eur J Appl Physiol, 82(1–2):117–20. 8. Amir O, Amir R, Yamin C, Attias E, Eynon N, Sagiv M, Meckel Y (2007). The ACE deletion allele is associated with Israeli elite endurance athletes. Exp Physiol, 92(5):881–6. 9. Atkinson RL, Dhurandhar NV, Allison DB, Bowen RL, Israel BA, Albu JB, Augustus AS (2005). Human adenovirus-36 is associated with increased body weight and paradoxical reduction of serum lipids. Int J Obes (Lond), 29(3):281–6. 10. Ayas NT, White DP, Al-Delaimy WK, Manson JE, Stampfer MJ, Speizer FE, Patel S, Hu FB (2003). A prospective study of self-reported sleep duration and incident diabetes in women. Diabetes Care, 26(2):380–4. 11. Barker DJ, Winter PD, Osmond C, Phillips DI, Sultan HY (1995). Weight gain in infancy and cancer of the ovary. Lancet, 345(8957):1087–1088. 12. Becker HF, Jerrentrup A, Ploch T, Grote L, Penzel T, Sullivan CE, Peter JH (2003). Effect of nasal continuous positive airway pressure treatment on blood pressure in patients with obstructive sleep apnea. Circulation, 107(1):68–73.
4
Genetic Epidemiology of Obesity and Cancer
117
13. Berstein LM IE, Kovalevskij AJ, et al. (2004). CYP17 and CYP19 genetic polymorphisms in endometrail cancer: association with intratumoral aromatase activity. Cancer Lett, 207: 191–196. 14. Bouchard C, Perusse L, Leblanc C (1990). Using MZ twins in experimental research to test for the presence of a genotype-environment interaction effect. Acta Genet Med Gemellol (Roma), 39(1):85–9. 15. Bradbury J (1997). Leptin deficiency linked to human obesity at last. Lancet, 349 (9069):1889. 16. Bray MS, Hagberg JM, Perusse L, Rankinen T, Roth SM, Wolfarth B, Bouchard C (2009). The human gene map for performance and health-related fitness phenotypes: the 2006–2007 update. Med Sci Sports Exerc, 41(1):35–73. 17. Bruning PF, Bonfrèr JM, Hart AA, van Noord PA, van der Hoeven H, Collette HJ, Battermann JJ, Jong-Bakker M, Nooijen WJ, de Waard F (1992a). Body Measurements, estrogen availability and the risk of human breast cancer, A case-control study. Int J Cancer, 51:14–19. 18. Bruning PF, Bonfrèr JM, van Noord PA, Hart AA, Jong-Bakker M, Nooijen WJ (1992b). Insulin resistance and breast-cancer risk. Int J Cancer, 52:511–516. 19. Bruning PF, Van Doorn J, Bonfrer JM, van Noord PA, Korse CM, Linders TC, Hart AA (1995). Insulin-like growth-factor-binding protein 3 is decreased in early-stage operable premenopausal breast cancer. Int J Cancer, 62:266–270. 20. Brunning PF (1987). Endogenous estrogens and breast cancer a possible relationship between body fat distribution and estrogen availability. J Steroid Biochem, 27:487–492. 21. Busquets X, Barbe F, Barcelo A, de la Pena M, Sigritz N, Mayoralas LR, Ladaria A, Agusti A (2004). Decreased plasma levels of orexin-A in sleep apnea. Respiration, 71(6):575–9. 22. Calle EE, Kaaks R (2004). Overweight, obesity and cancer: epidemiological evidence and proposed mechanisms. Nat Rev Cancer, 4(1):579–591. 23. Carling D (2004). The AMP-activated protein kinase cascade-a unifiying system for energy control. TRENDS Biochem Sci, 29(1):18–24. 24. Carskadon MA, Wolfson AR, Acebo C, Tzischinsky O, Seifer R (1998). Adolescent sleep patterns, circadian timing, and sleepiness at a transition to early school days. Sleep, 21(8):871–81. 25. Cecil JE, Tavendale R, Watt P, Hetherington MM, Palmer CN (2008). An obesity-associated FTO gene variant and increased energy intake in children. N Engl J Med, 359(24): 2558–66. 26. Chan JM, Stampfer MJ, Giovannucci E, Gann PH, Ma J, Wilkinson P, Hennekens CH, Pollak M (1998). Plasma insulin-like growth factor-I and prostate cancer risk: a prospective study. Science, 279(5350):563–566. 27. Chemelli RM, Willie JT, Sinton CM, Elmquist JK, Scammell T, Lee C, Richardson JA, Williams SC, Xiong Y, Kisanuki Y and others (1999). Narcolepsy in orexin knockout mice: molecular genetics of sleep regulation. Cell, 98(4):437–51. 28. Chen H, Charlat O, Tartaglia LA, Woofe EA, Weng X, Ellis SJ, Lakey ND, et al. (1996). Evidence that the diabetes gene encodes the leptin receptor: identification of a mutation in the leptin receptor gene in db/db mice. Cell, 84:491–495. 29. Chokkalingam AP, Stanczyk FZ, Reichardt JKV, Hsing AW (2007). Molecular epidemiology of prostate cancer: hormone-related genetic loci. Front Biosci, 12:3436–3446. 30. Chu LW, Reichardt JK, Hsing AW (2008). Androgens and the molecular epidemiology of prostate cancer. Curr Opin Endocrinol Diabetes Obes, 15:261–270. 31. Chua SC Jr., Chung WK, Wu-Peng XS, Zhang Y, Liu SM, Tartaglia LA, Leibel RL (1996). Phenotypes of mouse diabetes and rat fatty due to mutations in the OB (leptin) receptor. Science, 271:994–996. 32. Cirelli C, Gutierrez CM, Tononi G (2004). Extensive and divergent effects of sleep and wakefulness on brain gene expression. Neuron, 41(1):35–43. 33. Clayton SJ, May FEB, Westley BR (1997). Insulin-like growth factors control the regulation of oestrogen and progesterone receptor expression by oestrogens. Mol Cell Endocrinol, 128:57–68.
118
C. Gray-McGuire et al.
34. Collins M, Xenophontos SL, Cariolou MA, Mokone GG, Hudson DE, Anastasiades L, Noakes TD (2004). The ACE gene and endurance performance during the South African Ironman Triathlons. Med Sci Sports Exerc, 36(8):1314–20. 35. Cone RD (2005). Anatomy and regulation of the central melanocortin system. Nat Neurosci, 8(5):571–578. 36. Coschigano KT, Clemmons D, Bellush LL, Kopchick JJ (2000). Assessment of growth parameters and life span of GHR/BP gene-disrupted mice. Endocrinology, 141(7): 2608–2613. 37. Coulston AM, Boushey CJ (2008). Nutrition in the Prevention and Treatment of Disease: Academic Press, San Diego, CA. 38. Coyle YM (2009). Lifestyle, Genes, and Cancer. Methods of Molecular Biology, Cancer Epidemiol, 472:25–56. 39. Dahmen N, Bierbrauer J, Kasten M (2001). Increased prevalence of obesity in narcoleptic patients and relatives. Eur Arch Psychiatry Clin Neurosci, 251(2):85–9. 40. Davies RJ, Ali NJ, Stradling JR (1992). Neck circumference and other clinical features in the diagnosis of the obstructive sleep apnoea syndrome. Thorax, 47(2):101–5. 41. Defoor J, Martens K, Zielinska D, Matthijs G, Van Nerum H, Schepers D, Fagard R, Vanhees L (2006). The CAREGENE study: polymorphisms of the beta1-adrenoceptor gene and aerobic power in coronary artery disease. Eur Heart J, 27(7):808–16. 42. Del Giuduce ME, Fantus IG, Ezzat S, McKeown-Eyssen G, Page D, Goodwin PJ (1998). Insulin and related factors in premenopausal breast cancer risk. Breast Cancer Res Treat, 47:111–120. 43. DeLellis K, Rinaldi S, Kaaks RJ, Kolonel LN, Henderson B, Le ML (2004). Dietary and lifestyle correlates of plasma insulin-like growth factor-I (IGF-I) and IGF binding protein-3 (IGFBP-3): the multiethnic cohort. Cancer Epidemiol Biomarkers Prev, 13(9):1444–1451. 44. Dhurandhar NV, Israel BA, Kolesar JM, Mayhew GF, Cook ME, Atkinson RL (2000). Increased adiposity in animals due to a human virus. Int J Obes Relat Metab Disord, 24(8):989–96. 45. Dhurandhar NV, Kulkarni PR, Ajinkya SM, Sherikar AA, Atkinson RL (1997). Association of adenovirus infection with human obesity. Obes Res, 5(5):464–9. 46. Dhurandhar NV, Whigham LD, Abbott DH, Schultz-Darken NJ, Israel BA, Bradley SM, Kemnitz JW, Allison DB, Atkinson RL (2002). Human adenovirus Ad-36 promotes weight gain in male rhesus and marmoset monkeys. J Nutr, 132(10):3155–60. 47. Dina C, Meyre D, Gallina S, Durand E, Korner A, Jacobson P, Carlsson LM, Kiess W, Vatin V, Lecoeur C and others (2007). Variation in FTO contributes to childhood obesity and severe adult obesity. Nat Genet, 39(6):724–6. 48. Dinges DF, Pack F, Williams K, Gillen KA, Powell JW, Ott GE, Aptowicz C, Pack AI (1997). Cumulative sleepiness, mood disturbance, and psychomotor vigilance performance decrements during a week of sleep restricted to 4–5 hours per night. Sleep, 20(4):267–77. 49. Dong C, Li WD, Li D, Price RA (2005). Interaction between obesity-susceptibility loci in chromosome regions 2p25-p24 and 13q13-q21. Eur J Hum Genet, 13(1):102–108. 50. Dong C, Wang S, Li WD, Li D, Zhao H, Price RA (2003). Interacting genetic loci on chromosomes 20 and 10 influence extreme human obesity. Am J Hum Genet, 72(1):115–124. 51. Engeland A, Tretli S, Bjorge T (2004). Height and body mass index in relation to esophageal cancer; 23-year follow-up of two million Norwegian men and women. Cancer Causes Control, 15(8):837–43. 52. Espana RA, McCormack SL, Mochizuki T, Scammell TE (2007). Running promotes wakefulness and increases cataplexy in orexin knockout mice. Sleep, 30(11):1417–25. 53. Evans DJ, Hoffmann RG, Kalkhoff RK, Kissebah AH (1984). Relationship of body fat topography to insulin sensitivity and metabolic profiles in premenopausal women. Metabolism, 33:68–75. 54. Faccenda JF, Mackay TW, Boon NA, Douglas NJ (2001). Randomized placebo-controlled trial of continuous positive airway pressure on blood pressure in the sleep apnea-hypopnea syndrome. Am J Respir Crit Care Med, 163(2):344–8.
4
Genetic Epidemiology of Obesity and Cancer
119
55. Frayling TM (2007). Genome-wide association studies provide new insights into type 2 diabetes aetiology. Nat Rev Genet, 8(9):657–62. 56. Frayling TM, Timpson NJ, Weedon MN, Zeggini E, Freathy RM, Lindgren CM, Perry JR, Elliott KS, Lango H, Rayner NW and others (2007). A common variant in the FTO gene is associated with body mass index and predisposes to childhood and adult obesity. Science, 316(5826):889–94. 57. Freedman ND, Ahn J, Hou L, Lissowska J, Zatonski W, Yeager M, Chanock SJ, Chow WH, Abnet CC (2008). Polymorphisms in estrogen- and androgen- metabolizing genes and the risk of gastric cancer. Carcinogenesis, 30(1):71–77. 58. Gascon F, Valle M, Martos R, Zafra M, Morales R, Castano MA (2004). Childhood obesity and hormonal abnormalities associated with cancer risk. Eur J Cancer Prev, 13(3):193–197. 59. Gayagay G, Yu B, Hambly B, Boston T, Hahn A, Celermajer DS, Trent RJ (1998). Elite endurance athletes and the ACE I allele – the role of genes in athletic performance. Hum Genet, 103(1):48–50. 60. Giovannucci E, Pollak MN, Platz EA, Willett WC, Stampfer MJ, Majeed N, Colditz GA, Speizer FE, Hankinson SE (2000). A prospective study of plasma insulin-like growth factor1 and binding protein-3 and risk of colorectal neoplasia in women. Cancer Epidemiol Biomarkers Prev, 9(4):345–349. 61. Gloria W (2003). Basic Geriatric Nursing: Elsevier Health Sciences, St. Louis, Mo. 62. Goodwin PJ, Ennis M, Pritchard KI, Trudeau ME, Koo J, Madarnas Y, Hartwick W, Hoffman B, Hood N (2002). Fasting insulin and outcome in early-stage breast cancer: results of a prospective cohort study. J Clin Oncol, 20(1):42–51. 63. Gortmaker SL, Must A, Sobol AM, Peterson K, Colditz GA, Dietz WH (1996). Television viewing as a cause of increasing obesity among children in the United States, 1986–1990. Arch Pediatr Adolesc Med, 150(4):356–62. 64. Gottlieb DJ, Punjabi NM, Newman AB, Resnick HE, Redline S, Baldwin CM, Nieto FJ (2005). Association of sleep time with diabetes mellitus and impaired glucose tolerance. Arch Intern Med, 165(8):863–7. 65. Group. TEHaBCC (2002). Endogenous sex hormones and breast cancer in postmenopausal women, reanalysis of nine prospective studies. J Natl Cancer Inst, 94:606–616. 66. Guilleminault C, Partinen M, Hollman K, Powell N, Stoohs R (1995). Familial aggregates in obstructive sleep apnea syndrome. Chest, 107(6):1545–51. 67. Haffner SM, Katz MS, Stern MP, Dunn JF (1989). Relationship of sex hormone binding globulin to overall adiposity and body fat distribution in a biethnic population. Int J Obes, 13:1–9. 68. Hagberg JM, Ferrell RE, McCole SD, Wilund KR, Moore GE (1998). VO2 max is associated with ACE genotype in postmenopausal women. J Appl Physiol, 85(5):1842–6. 69. Hagberg JM, McCole SD, Brown MD, Ferrell RE, Wilund KR, Huberty A, Douglass LW, Moore GE (2002). ACE insertion/deletion polymorphism and submaximal exercise hemodynamics in postmenopausal women. J Appl Physiol, 92(3):1083–8. 70. Hankinson SE, Willett WC, Colditz GA, Hunter DJ, Michaud DS, Deroo B, Rosner B, Speizer FE, Pollak M (1998). Circulating concentrations of insulin-like growth factor-I and risk of breast cancer. Lancet, 351(9113):1393–1396. 71. Hara J, Yanagisawa M, Sakurai T (2005a). Difference in obesity phenotype between orexin-knockout mice and orexin neuron-deficient mice with same genetic background and environmental conditions. Neurosci Lett, 380(3):239–42. 72. Hara J, Yanagisawa M, Sakurai T (2005b). Difference in obesity phenotype between orexin-knockout mice and orexin neuron-deficient mice with same genetic background and environmental conditions. Neurosci Lett, 380(3):239–242. 73. Hardie DG, Carling D, Carlson M (1998). The AMP-activated/SNFI protein kinase subfamily: metabolic sensors of the eukaryotic cell? Ann Rev Biochem, 67:821–855. 74. Hart RP, Buchsbaum DG, Wade JB, Hamer RM, Kwentus JA (1987). Effect of sleep deprivation on first-year residents’ response times, memory, and mood. J Med Educ, 62(11):940–2.
120
C. Gray-McGuire et al.
75. Hasstedt SJ, Ramirez ME, Kuida H, Williams RR (1989). Recessive inheritance of a relative fat pattern. Am J Hum Genet, 45(6):917–25. 76. Hautala AJ, Leon AS, Skinner JS, Rao DC, Bouchard C, Rankinen T (2007). Peroxisome proliferator-activated receptor-delta polymorphisms are associated with physical performance and plasma lipids: the HERITAGE Family Study. Am J Physiol Heart Circ Physiol, 292(5):H2498–505. 77. Hawley SA, Boudeau J, Reid JL, Mustard KJ, Udd L, Makela TP, Alessi DR, Hardie DG (2003). Complexes between the LKBI tumor suppressor, STRAD alpha/beta and MO25 alpha/beta are upstream kinases in the AMP-activated protein kinase cascade. J Biol, 2(4):28. 78. Heath AC, Kendler KS, Eaves LJ, Martin NG (1990). Evidence for genetic influences on sleep disturbance and sleep pattern in twins. Sleep, 13(4):318–35. 79. Hersh CP, Demeo DL, Lazarus R, Celedon JC, Raby BA, Benditt JO, Criner G, Make B, Martinez FJ, Scanlon PD and others (2006). Genetic association analysis of functional impairment in chronic obstructive pulmonary disease. Am J Respir Crit Care Med, 173(9):977–84. 80. Horne JA, Anderson NR, Wilkinson RT (1983). Effects of sleep deprivation on signal detection measures of vigilance: implications for sleep function. Sleep, 6(4):347–58. 81. Horvath TL, Gao XB (2005). Input organization and plasticity of hypocretin neurons: possible clues to obesity’s association with insomnia. Cell Metab, 1(4):279–286. 82. Hruskovicova H, Dzurenkova D, Selingerova M, Bohus B, Timkanicova B, Kovacs L (2006). The angiotensin converting enzyme I/D polymorphism in long distance runners. J Sports Med Phys Fitness, 46(3):509–13. 83. Hsing AW, Chokkalingam AP (2006). Prostate Cancer Epidemiology. Front Biosci, 11: 1388–1413. 84. Hursting SD, Lashinger LM, Colbert LH, Rogers CJ, Wheatley KW, Nunez NP, Mahabir S, Barrett JC, Forman MR, Perkins SN (2007). Energy balance and carcinogenesis: underlying pathways and targets for intervention. Curr Cancer Drug Targets, 7(5):484–91. 85. Hursting SD, Lashinger LM, Wheatley KW, Rogers CJ, Colbert LH, Nunez NP, Perkins SN (2008). Reducing the weight of cancer: mechanistic targets for breaking the obesitycarcinogenesis link. Best Pract Res Clin Endocrinol Metab, 22(4):659–69. 86. Huygens W, Thomis MA, Peeters MW, Aerssens J, Janssen R, Vlietinck RF, Beunen G (2004a). Linkage of myostatin pathway genes with knee strength in humans. Physiol Genomics, 17(3):264–70. 87. Huygens W, Thomis MA, Peeters MW, Aerssens J, Janssen RG, Vlietinck RF, Beunen G (2004b). A quantitative trait locus on 13q14.2 for trunk strength. Twin Res, 7(6):603–6. 88. Jeffreys M, Smith GD, Martin RM, Frankel S, Gunnell D (2004). Childhood body mass index and later cancer risk: a 50-year follow-up of the Boyd Orr study. Int J Cancer, 112(2):348–351. 89. Kaaks R, Lukanova A, Sommersberg B (2000a). Plasma androgens, IGF-1, body size, and prostate cancer risk: a synthetic review. Prostate Cancer ProstaticDis, 3(3):157–172. 90. Kaaks R, Lundin E, Rinaldi S, Manjer J, Biessy C, Soderberg S, Lenner P, Janzon L, Riboli E, Berglund G and others (2002). Prospective study of IGF-I, IGF-binding proteins, and breast cancer risk, in northern and southern Sweden. Cancer Causes Control, 13(4): 307–316. 91. Kaaks R, Toniolo P, Akhmedkhanov A, Lukanova A, Biessy C, Dechaud H, Rinaldi S, Zeleniuch-Jacquotte A, Shore RE, Riboli E (2000b). Serum C-peptide, insulin-like growth factor (IGF)-I, IGF-binding proteins, and colorectal cancer risk in women. J Natl Cancer Inst 92(19):1592–1600. 92. Kanagy NL, Walker BR, Nelin LD (2001). Role of endothelin in intermittent hypoxiainduced hypertension. Hypertension, 37(2 Part 2):511–5. 93. Kanazawa H, Tateishi Y, Yoshikawa J (2004). Acute effects of nifedipine administration in pulmonary haemodynamics and oxygen delivery during exercise in patients with chronic obstructive pulmonary disease: implication of the angiotensin-converting enzyme gene polymorphisms. Clin Physiol Funct Imaging, 24(4):224–8.
4
Genetic Epidemiology of Obesity and Cancer
121
94. Kasikcioglu E, Kayserilioglu A, Ciloglu F, Akhan H, Oflaz H, Yildiz S, Peker I (2004). Angiotensin-converting enzyme gene polymorphism, left ventricular remodeling, and exercise capacity in strength-trained athletes. Heart Vessels, 19(6):287–93. 95. Kaye SA, Folsom AR, Soler JT, Prinease RJ, Potter JD (1991). Associations of body mass and fat distribution with sex hormone concentrations in post menopausal women. Int J Epidemiol, 20:151–156. 96. Keith SW, Redden DT, Katzmarzyk PT, Boggiano MM, Hanlon EC, Benca RM, Ruden D, Pietrobelli A, Barger JL, Fontaine KR and others (2006). Putative contributors to the secular increase in obesity: exploring the roads less traveled. Int J Obes (Lond), 30(11):1585–94. 97. Key TJ, Pike MC (1988). The dose-effect relationship between “unopposed” oestrogens and endometrial mitotic rate, its central role in explaining and predicting endometrial cancer risk. Br J Cancer, 57:205–212. 98. Khandwala HM, McCutcheon IE, Flyvbjerg A, Friend KE (2000). The effects of insulin-like growth factors on tumorigenesis and neoplastic growth. Endocr Rev, 21(3):215–244. 99. Kimura M, Winkelmann J (2007). Genetics of sleep and sleep disorders. Cell Mol Life Sci, 64(10):1216–26. 100. Kirschner MA, Samojlik E, Drejka M, Szmal E, Schneider G, Ertel N (1990). Androgenestrogen metabolism in women with upper body versus lower body obesity. J Clin Endocrinol Metab, 70:473–479. 101. Kok SW, Overeem S, Visscher TL, Lammers GJ, Seidell JC, Pijl H, Meinders AE (2003). Hypocretin deficiency in narcoleptic humans is associated with abdominal obesity. Obes Res, 11(9):1147–1154. 102. Kushida C (2007). Obstructive Sleep Apnea: Diagnosis and Treatment. Informa Health Care. 103. Laposky AD, Shelton J, Bass J, Dugovic C, Perrino N, Turek FW (2006). Altered sleep regulation in leptin-deficient mice. Am J Physiol Regul Integr Comp Physiol, 290(4): R894–R903. 104. Lawlor DA, Smith GD, Ebrahim S (2004). Hyperinsulinaemia and increased risk of breast cancer: findings from the British Women’s Heart and Health Study. Cancer Causes Control, 15(3):267–275. 105. Lee M, Korner J (2008). Review of physiology, clinical manifestations, and management of hypothalamic obesity in humans. Pituitary, 12(2):87–95. 106. Leproult R, Copinschi G, Buxton O, Van CE (1997). Sleep loss results in an elevation of cortisol levels the next evening. Sleep, 20(10):865–870. 107. Ley RE, Turnbaugh PJ, Klein S, Gordon JI (2006). Microbial ecology: human gut microbes associated with obesity. Nature, 444(7122):1022–3. 108. Li J, Yen C, Liaw D, Podsypanina K, Bose S, Wang SI, Puc J, Miliaresis C, Rodgers L, McCombie R and others (1997). PTEN, a putative protein tyrosine phosphatase gene mutated in human brain, breast, and prostate cancer. Science, 275(5308):1943–1947. 109. Li TY, Rana JS, Manson JE, Willett WC, Stampfer MJ, Colditz GA, Rexrode KM, Hu FB (2006). Obesity as compared with physical activity in predicting risk of coronary heart disease in women. Circulation, 113(4):499–506. 110. Liggett SB, Wagoner LE, Craft LL, Hornung RW, Hoit BD, McIntosh TC, Walsh RA (1998). The Ile164 beta2-adrenergic receptor polymorphism adversely affects the outcome of congestive heart failure. J Clin Invest, 102(8):1534–9. 111. Lightfoot JT, Turner MJ, Knab AK, Jedlicka AE, Oshimura T, Marzec J, Gladwell W, Leamy LJ, Kleeberger SR (2007). Quantitative trait loci associated with maximal exercise endurance in mice. J Appl Physiol, 103(1):105–10. 112. Lindblad M, Rodriguez LA, Lagergren J (2005). Body mass, tobacco and alcohol and risk of esophageal, gastric cardia, and gastric non-cardia adenocarcinoma among men and women in a nested case-control study. Cancer Causes Control, 16(3):285–294. 113. Long JR, Liu PY, Liu YJ, Lu Y, Xiong DH, Elze L, Recker RR, Deng HW (2003). APOE and TGF-beta1 genes are associated with obesity phenotypes. J Med Genet, 40(12):918–24. 114. Longo VD, Finch CE (2003). Evolutionary Medicine: From Dwarf Model Systems to Healthy Centenarians? Science, 299:1342–1346.
122
C. Gray-McGuire et al.
115. Lonnqvist F, Wennlund A, Arner P (1997). Relationship between circulating leptin and peripheral fat distribution in obese subjects. Int J Obesity, 21:255–260. 116. Loos RJ, Lindgren CM, Li S, Wheeler E, Zhao JH, Prokopenko I, Inouye M, Freathy RM, Attwood AP, Beckmann JS and others (2008). Common variants near MC4R are associated with fat mass, weight and risk of obesity. Nat Genet, 40(6):768–75. 117. Lopez-Alarcon M, Hunter GR, Gower BA, Fernandez JR (2007). IGF-I polymorphism is associated with lean mass, exercise economy, and exercise performance among premenopausal women. Arch Med Res, 38(1):56–63. 118. Lowell BB, Spiegelman BM (2000). Towards a molecular understanding of adaptive thermogenesis. Nature, 404(6778):652–60. 119. Lubin F, Chetrit A, Freedman LS, Alfandary E, Fishler Y, Nitzan H, Zultan A, Modan B (2003). Body mass index at age 18 years and during adult life and ovarian cancer risk. Am J Epidemiol, 157(2):113–120. 120. Lucia A, Gomez-Gallego F, Barroso I, Rabadan M, Bandres F, San Juan AF, Chicharro JL, Ekelund U, Brage S, Earnest CP and others (2005). PPARGC1A genotype (Gly482Ser) predicts exceptional endurance capacity in European men. J Appl Physiol, 99(1):344–8. 121. Lukanova A, Lundin E, Micheli A, Arslan A, Ferrari P, Rinaldi S, Krogh V, Lenner P, Shore RE, Biessy C and others (2004a). Circulating levels of sex steroid hormones and risk of endometrial cancer in postmenopausal women. Int J Cancer, 108(3):425–432. 122. Lukanova A, Zeleniuch-Jacquotte A, Lundin E, Micheli A, Arslan AA, Rinaldi S, Muti P, Lenner P, Koenig KL, Biessy C and others (2004b). Prediagnostic levels of C-peptide, IGF-I, IGFBP -1, -2 and -3 and risk of endometrial cancer. Int J Cancer, 108(2):262–268. 123. Ma J, Pollak M, Giovannucci E, Chan JM, Tao Y, Hennekens C, Stampfer MJ (2000). A prospective study of plasma levels of insulin-like growth factor I (IGF-I) and IGF-binding protein-3, and colorectal cancer risk among men. Growth Horm IGF Res, 10 Suppl A: S28–S29. 124. MacArthur DG, Seto JT, Raftery JM, Quinlan KG, Huttley GA, Hook JW, Lemckert FA, Kee AJ, Edwards MR, Berman Y and others (2007). Loss of ACTN3 gene function alters mouse muscle metabolism and shows evidence of positive selection in humans. Nat Genet, 39(10):1261–5. 125. Marik PE (2000). Leptin, obesity, and obstructive sleep apnea. Chest, 118(3):569–71. 126. McCole SD, Shuldiner AR, Brown MD, Moore GE, Ferrell RE, Wilund KR, Huberty A, Douglass LW, Hagberg JM (2004). Beta2- and beta3-adrenergic receptor polymorphisms and exercise hemodynamics in postmenopausal women. J Appl Physiol, 96(2):526–30. 127. Medlund P, Cederlof R, Floderus-Myrhed B, Friberg L, Sorensen S (1976). A new Swedish twin registry containing environmental and medical base line data from about 14,000 samesexed pairs born 1926–58. Acta Med Scand Suppl, 600:1–111. 128. Meyre D, Delplanque J, Chevre JC, Lecoeur C, Lobbens S, Gallina S, Durand E, Vatin V, Degraeve F, Proenca C and others (2009). Genome-wide association study for early-onset and morbid adult obesity identifies three new risk loci in European populations. Nat Genet, 41(2):157–9. 129. Modell S, Ising M, Holsboer F, Lauer CJ (2002). The Munich Vulnerability Study on Affective Disorders: stability of polysomnographic findings over time. Biol Psychiatry, 52(5):430–7. 130. Moll PP, Burns TL, Lauer RM (1991). The genetic and environmental sources of body mass index variability: the Muscatine Ponderosity Family Study. Am J Hum Genet, 49(6): 1243–55. 131. Montague CT, Prins JB, Sanders L et al. (1997). Depot- and sex-specific differences in human leptin mRNA expression: implications for the control of regional fat distribution. Diabetes, 46:342–347. 132. Montgomery HE, Marshall R, Hemingway H, Myerson S, Clarkson P, Dollery C, Hayward M, Holliman DE, Jubb M, World M and others (1998). Human gene for physical performance. Nature, 393(6682):221–2.
4
Genetic Epidemiology of Obesity and Cancer
123
133. Moore GE, Shuldiner AR, Zmuda JM, Ferrell RE, McCole SD, Hagberg JM (2001). Obesity gene variant and elite endurance performance. Metabolism, 50(12):1391–2. 134. Moore JW, Key TJ, Bulbrook RD, Clark GM, Allen DS, Wang DY, Pike MC (1987). Sex hormone binding globulin and risk factors for breast cancer in a population of normal women who had never used exogenous sex hormones. Br J Cancer, 56:661–666. 135. Moran CN, Yang N, Bailey ME, Tsiokanos A, Jamurtas A, MacArthur DG, North K, Pitsiladis YP, Wilson RH (2007). Association analysis of the ACTN3 R577X polymorphism and complex quantitative body composition and performance phenotypes in adolescent Greeks. Eur J Hum Genet, 15(1):88–93. 136. Must A, Jacques PF, Dallal GE, Bajema CJ, Dietz WH (1992). Long-term morbidity and mortality of overweight adolescents. A follow-up of the Harvard Growth Study of 1922 to 1935. N Engl J Med, 327(19):1350–5. 137. Myerson S, Hemingway H, Budget R, Martin J, Humphries S, Montgomery H (1999). Human angiotensin I-converting enzyme gene and endurance performance. J Appl Physiol, 87(4):1313–6. 138. National Sleep Foundation (2000). Omnibus Sleep in America Poll. National Sleep Foundation, Washington, DC. 139. Nazarov IB, Woods DR, Montgomery HE, Shneider OV, Kazakov VI, Tomilin NV, Rogozkin VA (2001). The angiotensin converting enzyme I/D polymorphism in Russian athletes. Eur J Hum Genet, 9(10):797–801. 140. Nishino S, Mignot E (1997). Pharmacological aspects of human and canine narcolepsy. Prog Neurobiol, 52(1):27–78. 141. Nishino S, Ripley B, Overeem S, Lammers GJ, Mignot E (2000). Hypocretin (orexin) deficiency in human narcolepsy. Lancet, 355(9197):39–40. 142. Norman RA, Thompson DB, Foroud T, Garvey WT, Bennett PH, Bogardus C, Ravussin E (1997). Genomewide search for genes influencing percent body fat in Pima Indians: suggestive linkage at chromosome 11q21-q22. Pima Diabetes Gene Group. Am J Hum Genet, 60(1):166–173. 143. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM (2006). Prevalence of overweight and obesity in the United States, 1999–2004. JAMA, 295(13):1549–55. 144. Oh SD (2007). The distribution of I/D polymorphism in the ACE gene among Korean male elite athletes. J Sports Med Phys Fitness, 47(2):250–4. 145. Ondo WG, Vuong KD, Wang Q (2000). Restless legs syndrome in monozygotic twins: clinical correlates. Neurology, 55(9):1404–6. 146. Paganini-Hill A, Martin AO, Spence MA (1981). The S-leut anthropometric traits: genetic analysis. Am J Phys Anthropol, 55(1):55–67. 147. Palmqvist R, Hallmans G, Rinaldi S, Biessy C, Stenling R, Riboli E, Kaaks R (2002). Plasma insulin-like growth factor 1, insulin-like growth factor binding protein 3, and risk of colorectal cancer: a prospective study in northern Sweden. Gut, 50(5):642–646. 148. Parkin DM, Bray F, Ferlay J, Pisani P (2005). Global cancer statistics, 2002. CA Cancer J Clin, 55:74–108. 149. Patel SR, Palmer LJ, Larkin EK, Jenny NS, White DP, Redline S (2004). Relationship between obstructive sleep apnea and diurnal leptin rhythms. Sleep, 27(2):235–9. 150. Peppard PE, Young T, Palta M, Skatrud J (2000). Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med, 342(19):1378–1384. 151. Pepperell JC, Ramdassingh-Dow S, Crosthwaite N, Mullins R, Jenkinson C, Stradling JR, Davies RJ (2002). Ambulatory blood pressure after therapeutic and subtherapeutic nasal continuous positive airway pressure for obstructive sleep apnoea: a randomised parallel trial. Lancet, 359(9302):204–10. 152. Peyrat JP, Bonneterre J, Hecquet B, Vennin P, Louchez MM, Fournier C, Lefebvre J, Demaille A (1993). Plasma insuin-like growth factor-I (IGF-I) concentrations in human breast cancer. Eur J Cancer, 29A:492–497. 153. Pirolli A, Smith C (1989). REM sleep deprivation in humans impairs learning a complex task. Sleep Research, 18:375.
124
C. Gray-McGuire et al.
154. Playford MP, Bicknell D, Bodmer WF, Macaulay VM (2000). Insulin-like growth factor 1 regulates the location, stability, and transcriptional activity of beta-catenin. Proc Natl Acad Sci USA, 97(22):12103–12108. 155. Polednak AP (2003). Trends in incidence rates for obesity-associated cancers in the US. Cancer Detect Prev, 27(6):415–21. 156. Potischman N, Hoover RN, Brinton LA, et al. (1996). Case-control study of endogenous steroid hormones and endometrial cancer. J Natl Cancer Inst, 88:1127–1135. 157. Potischman N, Troisi R (1999). In-utero and early life exposures in relation to risk of breast cancer. Cancer Causes Control, 10(6):561–573. 158. Rajala MW, Obici S, Scherer PE, Rossetti L (2003). Adipose-derived resistin and gut-derived resistin-like molecule-beta selectively impair insulin action on glucose production. J Clin Invest, 111(2):225–230. 159. Ravussin E, Lillioja S, Knowler WC, Christin L, Freymond D, Abbott WG, Boyce V, Howard BV, Bogardus C (1988). Reduced rate of energy expenditure as a risk factor for body-weight gain. N Engl J Med, 318(8):467–72. 160. Redline S, Tishler PV, Schluchter M, Aylor J, Clark K, Graham G (1999). Risk factors for sleep-disordered breathing in children. Associations with obesity, race, and respiratory problems. Am J Respir Crit Care Med, 159(5 Pt 1):1527–1532. 161. Reigle BS, Wonders K (2009). Breast cancer and the role of exercise in women. Methods Mol Biol, 472:169–89. 162. Rico-Sanz J, Rankinen T, Rice T, Leon AS, Skinner JS, Wilmore JH, Rao DC, Bouchard C (2004). Quantitative trait loci for maximal exercise capacity phenotypes and their responses to training in the HERITAGE Family Study. Physiol Genomics, 16(2):256–60. 163. Roberts SB, Young VR (1988). Energy costs of fat and protein deposition in the human infant. Am J Clin Nutr, 48(4):951–5. 164. Rose D, Gilhooly EM, Nixon D (2002). Adverse effects of obesity on breast cancer prognosis, and biological actions of leptin. Int J Oncol, 6:1285–1292. 165. Rubio JC, Martin MA, Rabadan M, Gomez-Gallego F, San Juan AF, Alonso JM, Chicharro JL, Perez M, Arenas J, Lucia A (2005). Frequency of the C34T mutation of the AMPD1 gene in world-class endurance athletes: does this mutation impair performance? J Appl Physiol, 98(6):2108–12. 166. Salgado R, Junius S, Benoy I, Van DP, Vermeulen P, Van ME, Huget P, Dirix LY (2003). Circulating interleukin-6 predicts survival in patients with metastatic breast cancer. Int J Cancer, 103(5):642–646. 167. Samanic C, Chow WH, Gridley G, Jarvholm B, Fraumeni JF Jr. (2006). Relation of body mass index to cancer risk in 362, 552 Swedish men. Cancer Causes Control, 17(7):901–909. 168. Sandilands AJ, Parameshwar J, Large S, Brown MJ, O’Shaughnessy KM (2005). Confirmation of a role for the 389R>G beta-1 adrenoceptor polymorphism on exercise capacity in heart failure. Heart, 91(12):1613–4. 169. Saunders CJ, Xenophontos SL, Cariolou MA, Anastassiades LC, Noakes TD, Collins M (2006). The bradykinin beta 2 receptor (BDKRB2) and endothelial nitric oxide synthase 3 (NOS3) genes and endurance performance during Ironman Triathlons. Hum Mol Genet, 15(6):979–87. 170. Scanavini D, Bernardi F, Castoldi E, Conconi F, Mazzoni G (2002). Increased frequency of the homozygous II ACE genotype in Italian Olympic endurance athletes. Eur J Hum Genet, 10(10):576–7. 171. Schmelzle T, Hall MN (2000). TOR, a central controller of cell growth. Cell, 103(2):253–62. 172. Schoen RE, Tangen CM, Kuller LH, Burke GL, Cushman M, Tracy RP, Dobs A, Savage PJ (1999). Increased blood glucose and insulin, body size, and incident colorectal cancer. J Natl Cancer Inst, 91(13):1147–1154. 173. Schwartz MW, Porte D Jr. (2005). Diabetes, obesity, and the brain. Science, 307(5708): 375–379.
4
Genetic Epidemiology of Obesity and Cancer
125
174. Setiawan VW, Doherty JA, Shu X, Akbari MR, Chen C, et al. (2009). Two Estrogen-Reated Variants in CYP19A1 and Endometrial Cancer Risk: A Pooled Analysis in the Epidemiology of Endometrail Cancer Consortium. Cancer Epidemiol Biomarkers Prev, 18(1):242–247. 175. Sharifi N, Gulley J, Dahut WL (2005). Androgen deprivation therapy for prostate cancer. JAMA, 294:238–244. 176. Shirai H, Oishi K, Kudo T, Shibata S, Ishida N (2007). PPARalpha is a potential therapeutic target of drugs to treat circadian rhythm sleep disorders. Biochem Biophys Res Commun, 357(3):679–82. 177. Silberberg R (1972). Articular aging and osteoarthrosis in dwarf mice. Pathol Microbiol, 38(6):417–430. 178. Smith C, Pirolli A (1989). Learning deficits following REM sleep deprivation of either the first two or last two REM periods of the night. Sleep Research, 18:377. 179. Smith JE, Tolson JM (2008). Recognition, diagnosis, and treatment of restless legs syndrome. J Am Acad Nurse Pract, 20(8):396–401. 180. Sookoian S, Gemma C, Gianotti TF, Burgueno A, Castano G, Pirola CJ (2008). Genetic variants of Clock transcription factor are associated with individual susceptibility to obesity. Am J Clin Nutr, 87(6):1606–15. 181. Spiegel K, Leproult R, Colecchia EF, L’Hermite-Baleriaux M, Nie Z, Copinschi G, Van CE (2000). Adaptation of the 24-h growth hormone profile to a state of sleep debt. Am J Physiol Regul Integr Comp Physiol, 279(3):R874–R883. 182. Spilsbury JC, Storfer-Isser A, Drotar D, Rosen CL, Kirchner LH, Benham H, Redline S (2004). Sleep behavior in an urban US sample of school-aged children. Arch Pediatr Adolesc Med, 158(10):988–94. 183. Stattin P, Lukanova A, Biessy C, Soderberg S, Palmqvist R, Kaaks R, Olsson T, Jellum E (2004). Obesity and colon cancer: does leptin provide a link? Int J Cancer, 109(1):149–152. 184. Stavola BL, Hardy R, Kuh D, Silva IS, Wadsworth M, Swerdlow AJ (2000). Birthweight, childhood growth and risk of breast cancer in a British cohort. Br J Cancer, 83(7):964–968. 185. Stunkard AJ, Foch TT, Hrubec Z (1986). A twin study of human obesity. Jama, 256(1):51–4. 186. Tafti M, Maret S, Dauvilliers Y (2005). Genes for normal sleep and sleep disorders. Ann Med, 37(8):580–9. 187. Taheri S, Mignot E (2002). The genetics of sleep disorders. Lancet Neurol, 1(4):242–50. 188. Tartaglia LA, Dembski M, Weng X, Deng N, Culpepper J, Devos R, Richards GJ, Campfield LA, Clark FT, Deeds J, Muir C, et al. (1995). Identification and expression cloning of a leptin receptor, OB-R. Cell, 83:1263–1271. 189. Thorleifsson G, Walters GB, Gudbjartsson DF, Steinthorsdottir V, Sulem P, Helgadottir A, Styrkarsdottir U, Gretarsdottir S, Thorlacius S, Jonsdottir I and others (2009). Genomewide association yields new sequence variants at seven loci that associate with measures of obesity. Nat Genet, 41(1):18–24. 190. Toi M, Inada K, Suzuki H, Tominaga T (1995). Tumor angiogenesis in breast cancer: its importance as a prognostic indicator and the association with vascular endothelial growth factor expression. Breast Cancer Res Treat, 36(2):193–204. 191. Toniolo P, Bruning PF, Akhmedkhanov A, Bonfrer JM, Koenig KL, Lukanova A, Shore RE, Zeleniuch-Jacquotte A (2000). Serum insulin-like growth factor-I and breast cancer. Int J Cancer, 88(5):828–832. 192. Trayhurn P, Bing C, Wood IS (2006). Adipose tissue and adipokines–energy regulation from the human perspective. J Nutr, 136(7 Suppl):1935S–1939S. 193. Treuth MS, Baggett CD, Pratt CA, Going SB, Elder JP, Charneco EY, Webber LS (2009). A Longitudinal Study of Sedentary Behavior and Overweight in Adolescent Girls. Obesity (Silver Spring). 17(5):1003–8. 194. Turek FW, Joshu C, Kohsaka A, Lin E, Ivanova G, McDearmon E, Laposky A, Losee-Olson S, Easton A, Jensen DR and others (2005). Obesity and metabolic syndrome in circadian Clock mutant mice. Science, 308(5724):1043–5.
126
C. Gray-McGuire et al.
195. Turgut G, Kaptanoglu B, Turgut S, Enli Y, Genc O (2004). Effects of chronic aluminum administration on blood and liver iron-related parameters in mice. Yonsei Med J, 45(1): 135–9. 196. Turnbaugh PJ, Ley RE, Mahowald MA, Magrini V, Mardis ER, Gordon JI (2006). An obesity-associated gut microbiome with increased capacity for energy harvest. Nature, 444(7122):1027–31. 197. Van Cauter E, Copinschi G (2000). Interrelationships between growth hormone and sleep. Growth Horm IGF Res, 10 Suppl B:S57–62. 198. Van CE, Plat L, Copinschi G (1998). Interrelations between sleep and the somatotropic axis. Sleep, 21(6):553–566. 199. van Ginneken V, Sitnyakowsky L, Jeffery JE (2009). Infectobesity: viral infections (especially with human adenovirus-36: Ad-36) may be a cause of obesity. Med Hypotheses, 72(4):383–8. 200. van Rossum EF, Voorhoeve PG, te Velde SJ, Koper JW, Delemarre-van de Waal HA, Kemper HC, Lamberts SW (2004). The ER22/23EK polymorphism in the glucocorticoid receptor gene is associated with a beneficial body composition and muscle strength in young adults. J Clin Endocrinol Metab, 89(8):4004–9. 201. Vincent B, De Bock K, Ramaekers M, Van den Eede E, Van Leemputte M, Hespel P, Thomis MA (2007). ACTN3 (R577X) genotype is associated with fiber type distribution. Physiol Genomics, 32(1):58–63. 202. Vioque J, Torres A, Quiles J (2000). Time spent watching television, sleep duration and obesity in adults living in Valencia, Spain. Int J Obes Relat Metab Disord, 24(12):1683–8. 203. Wagoner LE, Craft LL, Singh B, Suresh DP, Zengel PW, McGuire N, Abraham WT, Chenier TC, Dorn GW, 2nd, Liggett SB (2000). Polymorphisms of the beta(2)-adrenergic receptor determine exercise capacity in patients with heart failure. Circ Res, 86(8):834–40. 204. Wagoner LE, Craft LL, Zengel P, McGuire N, Rathz DA, Dorn GW, 2nd, Liggett SB (2002). Polymorphisms of the beta1-adrenergic receptor predict exercise capacity in heart failure. Am Heart J, 144(5):840–6. 205. Wang W, Guan KL (2009). AMPK and Cancer. Acta Physiol. 206. Ways JA, Cicila GT, Garrett MR, Koch LG (2002). A genome scan for Loci associated with aerobic running capacity in rats. Genomics, 80(1):13–20. 207. Webb WB, Campbell SS (1983). Relationships in sleep characteristics of identical and fraternal twins. Arch Gen Psychiatry, 40(10):1093–5. 208. Whitaker RC, Wright JA, Pepe MS, Seidel KD, Dietz WH (1997a). Predicting obesity in young adulthood from childhood and parental obesity. N Engl J Med, 337(13):869–73. 209. Whitaker RC, Wright JA, Pepe MS, Seidel KD, Dietz WH (1997b). Predicting obesity in young adulthood from childhood and parental obesity. N Engl J Med, 337(13):869–873. 210. Willer CJ, Speliotes EK, Loos RJF, Li S, Lindgren CM, Heid IM, et al. (2009). Six new loci associated with body mass index highlight a neuronal influence on body weight regulation. Nat. Genetics, 41(1):25–34. 211. Wingard DL, Berkman LF (1983). Mortality risk associated with sleeping patterns among adults. Sleep, 6(2):102–107. 212. Winkelmann J, Wetter TC, Collado-Seidel V, Gasser T, Dichgans M, Yassouridis A, Trenkwalder C (2000). Clinical characteristics and frequency of the hereditary restless legs syndrome in a population of 300 patients. Sleep, 23(5):597–602. 213. Wolf AM, Hunter DJ, Colditz GA, Manson JE, Stampfer MJ, Corsano KA, Rosner B, Kriska A, Willett WC (1994). Reproducibility and validity of a self-administered physical activity questionnaire. Int J Epidemiol, 23(5):991–9. 214. Wolfarth B, Rankinen T, Muhlbauer S, Scherr J, Boulay MR, Perusse L, Rauramaa R, Bouchard C (2007). Association between a beta2-adrenergic receptor polymorphism and elite endurance performance. Metabolism, 56(12):1649–51.
4
Genetic Epidemiology of Obesity and Cancer
127
215. Wolfarth B, Rivera MA, Oppert JM, Boulay MR, Dionne FT, Chagnon M, Gagnon J, Chagnon Y, Perusse L, Keul J and others (2000). A polymorphism in the alpha2aadrenoceptor gene and endurance athlete status. Med Sci Sports Exerc, 32(10):1709–12. 216. Wong SL, Leatherdale ST (2009). Association between sedentary behavior, physical activity, and obesity: inactivity among active kids. Prev Chronic Dis, 6(1):A26. 217. Woods A, Johnstone SR, Dickerson K, Leiper FC, Fryer LG, Neumann D, Schlattner U, Wallimann T, Carlson M, Carling D (2003). LKB1 is the upstream kinase in the AMPactivated protein kinase cascade. Curr Biol, 13(22):2004–2008. 218. Woods D, Hickman M, Jamshidi Y, Brull D, Vassiliou V, Jones A, Humphries S, Montgomery H (2001). Elite swimmers and the D allele of the ACE I/D polymorphism. Hum Genet, 108(3):230–2. 219. Woods SC, Seeley RJ, Porte D, Jr. et al. (1998). Signals that regulate food intake and energy homeostasis. Science, 280:1378–1383. 220. World Health Organization (2006). World Health Organization Database. 221. Yang N, MacArthur DG, Wolde B, Onywera VO, Boit MK, Lau SY, Wilson RH, Scott RA, Pitsiladis YP, North K (2007). The ACTN3 R577X polymorphism in East and West African athletes. Med Sci Sports Exerc, 39(11):1985–8. 222. Zeleniuch-Jacquotte A, Akhmedkhanov A, Kato I, et al. (2001). Postmenopausal endogenous oestrogens and risk of endometrial cancer: results of a prospective study. Br J Cancer, 84:975–981. 223. Zhang Y L (1998). Molecular physiology of leptin and its receptor. N Engl J Med, 14:17–35. 224. Zhao B, Moochhala SM, Tham S, Lu J, Chia M, Byrne C, Hu Q, Lee LK (2003). Relationship between angiotensin-converting enzyme ID polymorphism and VO(2max) of Chinese males. Life Sci, 73(20):2625–30. 225. Zonta LA, Jayakar SD, Bosisio M, Galante A, Pennetti V (1987). Genetic analysis of human obesity in an Italian sample. Hum Hered, 37(3):129–39.
Chapter 5
Obesity and Cancer: Overview of Mechanisms Nora L. Nock and Nathan A. Berger
Abstract This chapter provides an overview of the putative pathophysiological mechanisms explaining the positive association between obesity and various cancers. After presenting each of the major factors and pathways involved in obesity-related carcinogenesis and their potential synergisms from a physiological, epidemiological and mechanistic perspective, we conclude with a discussion of the therapeutic opportunities to reduce obesity and obesity-related cancers.
1 Background The rates of overweight (defined as a body mass index (BMI) ≥ 25.0 kg/m2 and ≤ 30.0 kg/m2 ) and obesity (defined as a BMI ≥ 30.0 kg/m2 ) are increasing in epidemic proportions in both developed and developing nations with over 1 billion overweight and 315 million obese adults currently estimated worldwide [1–3]. In the United States, over two-thirds of the adult population is overweight and approximately one-third is obese with a clear upward shift in the mean BMI occurring over the past 20 years [4] (see Chapter 1 for a detailed review of the epidemiology of obesity). Obesity is clearly a risk factor for certain cancers. Strong and consistent epidemiological associations have been observed between obesity and several malignancies including colon, postmenopausal breast, endometrial, esophageal adenocarcinoma and renal cell cancers [5–7]. In addition, there is evidence for an association between obesity and other malignancies including pancreatic, gallbladder, and hepatocellular cancers [7–9]. However, a positive association between obesity and cancer is not universal across all tissue types. For example, there is no evidence for a positive association between obesity and lung cancer; and, in fact, some studies N.L. Nock (B) Epidemiology and Biostatistics, Case Western Reserve University, Cleveland, OH, USA e-mail:
[email protected]
N.A. Berger (ed.), Cancer and Energy Balance, Epidemiology and Overview, Energy Balance and Cancer 2, DOI 10.1007/978-1-4419-5515-9_5, C Springer Science+Business Media, LLC 2010
129
130
N.L. Nock and N.A. Berger
have observed an inverse association between BMI and lung cancer risk, particularly among smokers [7]. Furthermore, the severity or stage of cancer at diagnosis may also play a role as observed in prostate cancer, whereby obesity has only been positively associated with aggressive or high-grade prostrate tumours [10]. Obesity is also a risk factor for increased cancer mortality. In the United States, approximately 20% of all cancer deaths in women and 14% in men are estimated to be attributed to being overweight or obese [8]. Obesity at diagnosis has also been associated with increased risk of recurrence and/or shorter survival time in endometrial [11–13], postmenopausal breast [13–15], colon [16], and aggressive prostate [10, 17] cancers. Although it is clear that energy restriction inhibits progression of many cancers in rodent models [18], the prognostic role of obesity after cancer diagnosis in humans remains an active area of research. While there is strong epidemiological evidence for an association between obesity and increased cancer risk at several organ sites, the mechanisms driving these associations are not well understood. This is due, in part, to the complex physiological, molecular, and biochemical perturbations involved in the manifestation of obesity and because many of these processes likely act cooperatively in carcinogenesis. Table 5.1 lists several of the putative mediators and mechanisms involved in the link between obesity and carcinogenesis. Figure 5.1 provides an illustration of the major actions and interactions of these processes, most of which appear to affect promotion and progression of cancer. For example, obesity modifies circulating concentrations of endogenous hormones such as insulin, insulin-like growth factors, and sex steroid hormones, which can modify cellular proliferation, apoptosis, and angiogenesis [19]. Obesity may also induce chronic low-grade inflammation resulting in an alteration of local and systemic levels of cytokines (e.g., TNF-α, IL-6, CRP) and adipokines (e.g., leptin, adiponectin), which may, in turn, affect mitosis, apoptosis, and angiogenesis [19]. Other candidate mechanisms, which could be involved in cancer initiation and/or promotion, include oxidative stress and dietary intake of carcinogens. Obesity leads to other disorders which could indirectly increase the risk of cancer. For example, obesity causes non-alcoholic fatty liver disease (NAFLD), which increases the risk of hepatocellular carcinoma [20]. Obesity has also been associated with gastroesophageal reflux disease (GERD) [21], which increases the risk of esophageal adenocarcinoma [20, 22]. Each of the factors listed in Table 5.1 may play an independent role in carcinogenesis; however, it is more likely that the association between obesity and cancer is best explained by the interplay between several, if not all, of these factors together with the modifying effects of behavioral, social, and genetic factors. Thus, the overall impact of obesity or disordered energy homeostasis on cancer development and growth is most likely the result of a multifactorial process that enables the bypass of cell cycle checkpoints leading to unrestricted cancer cell growth. This chapter will review the major factors putatively involved in obesity-related carcinogenesis and their potential synergisms from epidemiological, biochemical, and mechanistic viewpoints. Given the rising rates of obesity on a worldwide basis, disentangling the roles of the multiple putative mechanisms involved in obesityrelated carcinogenesis is becoming increasingly more important to ensure future
5
Obesity and Cancer: Overview of Mechanisms
131
Table 5.1 Obesity and cancer: Putative mechanistic factors and their role in carcinogenesis
Factor
Primary tissue of origin
Peptide hormones ↑ Insulin
Pancreas
↑ IGF-1
Liver
Sex steroid hormones ↑ Estrogen
Ovaries
– Secondary sex characteristics in females – Promotes growth in hormonedependent tissues
Testosterone
Testis
– Secondary sex characteristics in males
Adipose and gut hormones ↑ Leptin
↓ Adiponectin
Initiation (I)/ Promotion (P)
Normal function
Role in carcinogenesis
– Glucose uptake and utilization in muscle, liver, and adipose tissue – Blocks gluconeogenesis – Insulin-like effects – Promotes cell growth – Inhibits apoptosis
– Stimulates cell growth and proliferation – Anti-apoptotic
P
– Stimulates cell growth and proliferation – Pro-angiogenic
P
– Promotes cell proliferation – Anti-apoptotic and pro-angiogenic in hormone-dependent tissues (breast, uterus, ovary) – Mutagenic DNA adducts – Conversion to estrogen by aromatase
I and P
Adipocytes – Inhibits brain orexigenic pathways – Stimulates anorexigenic pathways Adipocytes – Insulin sensitization – Increased glucose uptake – Increased fatty acid oxidation
P
– Mitogenic, anti-apoptotic, pro-angiogenic
P
– Permissive effect on insulin resistance
P
132
N.L. Nock and N.A. Berger Table 5.1 (continued)
Factor Cytokines ↑ TNF-α
↑ IL-6 ↑ CRP ↑ MCP-1 Free radicals ↑ ROS
Primary tissue
Normal function
Macrophages Adipocytes
– Proinfammatory
Macrophages Smooth muscle Liver and adipocytes Macrophages Smooth muscle
– Proinflammatory – Anti-apoptotic – Proinflammatory – Proinflammatory
– Ubiquitous
Fatty acid metabolites ↑ Prostaglandins (PGE2 )
Angiogenesis ↑ VEGF
Ubiquitous
Xenobiotic carcinogenic agents ↑ PAH
Diet
↑ PhIP
Diet
Role in carcinogenesis
Initiation (I) versus promotion (P)
– Inflammation – Insulin resistance/IRS inhibition – Pro-angiogenic – Upregulates aromatase – Mitogenic – Pro-angiogenic
I and P
– Stimulates IL-6 – Stimulates TNF-α – Metastasis – Chemotaxis
P
– Signal transduction – Cellular senescence – Apoptosis – Autophagy
– DNA damage/mutagenesis – Promotes insulin resistance
I and P
– Regulates adipocyte and lipid metabolism – Stimulates growth
– Stimulates cell growth and proliferation – Anti-apoptotic, pro-angiogenic – Metastasis
P
Promotes normal angiogenesis
– Promotes tumor angiogenesis – Pro-angiogenic
P
– PAH–DNA adducts – Mutagenesis – PhIP–DNA adducts – Mutagenesis
I
P
P
I
5
Obesity and Cancer: Overview of Mechanisms
133
Fig. 5.1 Putative factors involved in obesity-related carcinogenesis. Adapted from Cowey and Hardy [268], with permission from the American Society for Investigative Pathology. Factors denoted in bold red text are core features of the Metabolic Syndrome. Factors denoted in bold blue text are additional features that may also be components of the Metabolic Syndrome. Abbreviations used: CRP, C-reactive protein; FFA, free fatty acids; IGF-1, insulin-like growth factor 1; IGFBP, insulin-like growth factor-binding protein; IL-6, interleukin-6; MAC, macrophage; MCP-1, monocyte chemoattractant protein 1; Mito, mitochondria; PAI-1, plasminogen activator inhibitor-1; ROS, reactive oxygen species; SHBG, steroid hormone-binding globulin; TG, triglycerides; TNF-α, tumor necrosis factor α; VEGF, vascular endothelial growth factor
research is directed toward designing the most cost-effective and efficient control and prevention strategies.
2 Insulin and Insulin-Like Growth Factor (IGF) Axis 2.1 Insulin Insulin circulates as a two-chain (α and β) peptide hormone composed of 51 amino acids with a molecular mass of 5.8 kDa. It is synthesized in pancreatic β-islet cells from the 91 amino acid proinsulin protein and its synthesis and release are stimulated in response to elevated blood glucose levels. Amino acids are proteolytically cleaved from the interior of the proinsulin molecule following which the terminal fragments (α and β) are linked by disulfide bonds before secretion
134
N.L. Nock and N.A. Berger LEP
EGF
IGF-1
INS
TNF-α
ADIPO
LEPR
EGFR
IGF-IR
INSR
TNFR
ADIPOR
IRS
GRB2
Ras
IKK
TRAF2
PI (4,5)P2 LKB1 PI3K Exercise PIP3
JAK2 Src Raf
AMP ATP
Glucose
AMPKK/ STK11
G6P
ROS AMPK
PDK1
SHP-2 PTPN11 MEK 1/2
PEPCK FOX01
Stat 3 AKT
Erk1/2/ MAPK
TSC2/1
mTOR
GSK3B 4EBp1
C-Myc
Amino Acids Autophagy
p70S6K
S6
eIF4E
ROS
eIF4B
HlF1α
Hypoxia
VEGF
Cyclin D1 Protein Translation Cell Proliferation
Cell Growth
Angiogenesis Cell Survival
Tumor Promotion / Progression
Fig. 5.2 Peptide growth factors, adipokines, and nutrients putatively involved in obesity-related carcinogenesis. Adapted from Moore et al. [317], with permission from American Association for Cancer Research. Obesity is usually associated with increased peptide growth factors including insulin (INS), insulin-like growth factor-1 (IGF-1), leptin (LEP), decreased adiponectin (ADIPO) as well as increased nutrients such as glucose and amino acids. INS and IGF-1 bind their respective transmembrane receptors, the insulin receptor (INSR) and the IGF-1 receptor (IGF-1R), which function as protein tyrosine kinases to phosphorylate and activate the insulin receptor substrate (IRS) that in turn activates phosphoinositide 3 kinase (PI3K) and the AKT system of proteins, originally named for a transforming retrovirus. Intermediary steps in this pathway include activation of phosphoinositide-dependent kinase (PDK1). PI3K converts the signaling molecule phosphoinositide 4,5-diphosphate (PI(4,5)P2) to phosphotidylinositol 3,4,5-triphosphate (PIP3). AKT activates the mammalian target of rapamycin system (mTOR) which serves as a central regulatory focus connecting energy metabolism to cell growth and proliferation. mTOR is blocked by a pathway initiated by an increased adenosine monophosphate (AMP)/adenosine triphosphate (ATP) ratio, which activates AMP kinase (AMPK) that in turn increases activity of the tuberous sclerosis complex (TSC 2/1) to block mTOR function. Increased glucose is metabolized to increase the ATP/AMP ratio, thereby eliminating stimulation of AMPK, removing the TSC 2/1 block to mTOR. Through several intermediate steps (not shown), increased amino acids and hypoxia each activates mTOR. Activated mTOR prevents activation of eukaryotic translation initiation factor 4E binding protein 1 (4EBP1) which associates with translation factor eIF4E to inhibit ribosomal subunit assembly and repress translation. Dissociation of 4EBP1 from the complex allows initiation of protein translation. Its release contributes to increased protein translation of multiple factors involved in cell growth and proliferation including cyclin D1 which is critical to cell cycle traverse and cell proliferation. mTOR also activates the 70-kDa ribosomal protein S6 kinase (p70S6K) which in turn phosphorylates the 40S ribosomal protein S6 (S6) and eukaryotic translation initiation factor 4B (eIF4B), both of which contribute to increased protein translation and cell growth. In addition, mTOR stimulates release of hypoxia-inducible factor 1α (HIFα) which results in increased activity
5
Obesity and Cancer: Overview of Mechanisms
135
Fig. 5.2 (continued) of vascular endothelial growth factor (VEGF) and angiogenesis which in turn facilitates increased gas and nutrient supply to tumor cell masses. Epidermal growth factor (EGF) functions by activating the transmembrane epidermal growth factor receptor (EGFR). EGFR, IGF-1R, and INSR are all protein tyrosine kinases that dimerize and phosphorylate their intracellular domains upon binding by their ligands. EGFR, IGF1R, and INSR are involved in cross talk at the intracellular level. Activation of the EGFR stimulates cell proliferation through the PI3K pathway, the Src kinase (Src) pathway, and the extracellular signal-regulated kinase (ERK1/2), mitogen-activated protein kinase (MAPK) pathway. The MAPK pathway includes activation of Ras, a membrane associated small molecular weight guanine nucleotide-binding protein, Raf, a serine/threonine protein kinase, MEK 1/2, MAPK kinase, Erk 1/2, and c-Myc, the cellular homologue of the myelocytomatosis virus oncogene to activate cyclin D1 leading to cell proliferation. Activation of the Src (Stat3) signal transduction and activation of transcription pathway contributes to increased activity of cyclin D1. Glycogen synthetase 3B phosphorylates cyclin D1 leading to its degradation, a process which is interfered by AKT. Cross talk interactions between insulin and EGF networks may amplify mitogenic signaling. In addition to activating cell proliferation through the PI3K pathway, insulin (INS) binding to the insulin receptor (INSR) phosphorylates IRS proteins, which are linked to the activation of Ras/ERK pathway through binding to the growth factor receptor-binding protein 2 (GRB2)-son of sevenless (SOS) complex. Leptin (LEP) binding to its receptor (LEPR), a transmembrane protein belonging to the cytokine receptor superfamily, induces Janus kinase (JAK) proteins (JAK2 and possibly JAK1) which are activated by transphosphorylation. The activated JAK proteins, in turn, phosphorylate tyrosine residues, providing a docking site for STAT proteins; the phosphorylated STAT molecules dimerize and translocate to the nucleus where they modulate gene transcription. JAK proteins then induce protein tyrosine phosphatase, non-receptor type 11 (SHP-2/PTPN11), which, in turn, activates the MAPK signaling pathway. Adiponectin (ADIPO) binding to its receptor (ADIPOR), which belongs to a seventransmembrane receptor family termed PAQR that has its N terminus facing the cytosol, has been shown to stimulate AMP-activated protein kinase (AMPK), activation through AMPK kinase, or serine/threonine protein kinase 11 (AMPKK/STK11). Upon activation, AMPK increases cellular energy levels by inhibiting anabolic energy-consuming pathways (e.g., fatty acid synthesis, protein synthesis) and stimulating catabolic energy-producing pathways (e.g., fatty acid oxidation, glucose transport). However, under hypoxic conditions, which may occur within tumors, activation of AMP-activated protein kinase (AMPK) leads to the phosphorylation and activation of the kinase activity of phosphofructokinase-2 (PFK-2, not shown) causing upregulation of anaerobic glycolysis (G6P, glucose 6-phosphatase) known as the ‘Pasteur effect’ whereby an increased rate of glycolysis occurs in response to hypoxia. Activation of AMPK, especially by metformin used to treat type 2 diabetes, has been shown to inhibit hepatic gluconeogenesis through AMPKdependent regulation of phosphoenol pyruvate carboxykinase (PEPCK). Tumor necrosis factor alpha (TNF-α) binding to its receptor (TNFR) induces TNF receptor-associated factor 2 (TRAF2), which recruits the multi-component protein kinase IkappaB kinase (IKK), and the IKK complex has been shown to attenuate insulin action via the phosphorylation of insulin receptor substrate 1 (IRS-1). Forkhead box O transcription factors (FOXO) play a significant role in regulating whole body energy metabolism. In particular, when nutrient and insulin levels are low, FOXO1 promotes expression of gluconeogenic enzymes; however, in the fed state under conditions of insulin resistance, the negative signaling afforded by FOXO1 is compromised leading to dysregulated glucose uptake. Reactions which stimulate or activate others are indicated by an −−>, whereas substrates that inhibit or interfere with subsequent pathways or activities are indicated by a − − −| .
as the α- and β-insulin dimer and the free 30 amino acids (C-peptide). Insulin facilitates glucose uptake and utilization in muscle, liver, and adipose tissue to provide energy, while concomitantly decreasing the amount of glucose produced by
136
N.L. Nock and N.A. Berger
the liver. Under normal conditions, insulin acts by directly binding to its receptor and through a myriad of signaling molecules (e.g., IRS-1, Akt, and to a lesser extent IRS-2), causing movement of glucose transporters (e.g., GLUT4 in skeletal muscle) to the cell membrane to bring glucose into the cell for subsequent use in energy metabolism pathways such as glycolysis and glycogen and fatty acid synthesis [9]. However, obesity causes a decreased sensitivity to the action of insulin (insulin resistance), leading to a compensatory increase in the production of insulin (hyperinsulinemia) [23]. Serum insulin levels fluctuate with the level of blood glucose and the state of fasting; therefore, surrogates of pancreatic insulin secretion, such as C-peptide, may be preferred in epidemiological studies because C-peptide is not removed by the liver, has a slower metabolic clearance, and lacks cross-reactivity with antibodies [24]. Since C-peptide is cleaved along with insulin from proinsulin and secreted into the circulation in equimolar amounts when insulin is required, it is used as a surrogate marker for insulin secretion. Higher circulating levels of C-peptide have been associated with obesity [25] and with an increased risk of endometrial [26], postmenopausal breast [27], colorectal [28], and pancreatic [29] cancers. However, a recent meta-analysis concluded that only colorectal and pancreatic cancer patients have increased pre-diagnostic blood levels of insulin and glucose [30], and effects in carcinogenesis may only be relevant in the postprandial state [29]. In terms of the potential mechanistic role(s) of insulin in carcinogenesis, chronic hyperinsulinemia may lead to increased cell proliferation (mitogenesis) and inhibition of programmed cell death (apoptosis). This may occur directly through the effects of insulin binding to its receptor or indirectly by decreasing the levels of insulin-like growth factor-binding proteins (IGFBP-1 and -2), resulting in higher levels of bioavailable IGF-1 [19, 31]. As shown in Fig. 5.2, insulin activity at the cellular level is initiated by binding to its receptor (INSR), one of several cell membrane receptor tyrosine kinases, which upon phosphorylation initiate a cascade of events through the insulin receptor substrate (IRS-1) and phosphoinositide 3-kinase (PI3K)/Akt pathway leading to activation of mammalian target of rapamycin (mTOR), which serves as a central regulator of cell metabolism through several downstream pathways. However, insulin may only be directly mitogenic at levels higher than those typically found in normal physiological conditions [32]. Interestingly, IRS-1-associated PI3K activity has been found to be markedly impaired in obese women and GLUT4 protein levels substantially decreased in obese compared to lean subjects [33]. As shown in Fig. 5.2, activated PI3K then converts phosphatidylinositol 4,5-bisphosphate (PI(4,5)P2) to phosphatidylinositol 3,4,5-triphosphate (PIP3), which recruits Akt and phosphoinositide-dependent kinase-1 (PDK1) to the plasma membrane and, then, PDK1 phosphorylates and activates Akt [34]. Aberrant Akt signaling is observed in many cancers and may induce upregulation of mTOR, which controls protein synthesis in response to growth factor signaling and nutrient availability [35]. Activation of the PI3K/AKT pathway is also associated with deactivation of glycogen synthetase kinase 3B (GSK3B), resulting in upregulation of cyclin D1 and increased cell proliferation (Fig. 5.2).
5
Obesity and Cancer: Overview of Mechanisms
137
Recent evidence suggests mTOR is also regulated by TNF-α and Wnt, both of which have critical roles in the development of many human neoplasias [35]. AMPactivated protein kinase (AMPK) responds to cellular AMP/ATP ratios and when AMP is high (or ATP is low) the tuberous sclerosis complex (TSC) is activated, which, in turn, inhibits mTOR signaling [36]. Insulin may also inhibit sex hormonebinding globulin (SHBG) production [37], leading to an increase in the amount of bioavailable steroid hormones (see Section 3 for further discussion).
2.2 Insulin-Like Growth Factor (IGF) The IGF system is complex involving multiple ligands (IGF-1, IGF-2), receptors (IGF-1R, IGF-2R), and binding proteins (IGFBP-1, 2, 3, 4, 5, 6), which are important for regulating normal tissue growth and regeneration. Although both IGF-1 and IGF-2 are involved in prenatal growth, only IGF-I appears to be relevant in postnatal growth [38]. IGF-1 may have a beneficial effect on glucose homeostasis, since IGF-1 can bind to insulin receptors, stimulating insulin-like actions and enhancing insulin sensitivity; however, IGF-1 has a low affinity for these insulin receptors and, therefore, may only act indirectly through suppression of growth hormone (GH), an insulin antagonist [39]. Circulating levels of IGF-1 are determined primarily by hepatic synthesis, which is regulated by nutrient status and GH [40] while IGF-1 levels in extrahepatic tissues are determined by a plethora of factors including GH, IGFBPs, and local levels of IGF-1 [41]. Approximately 80–90% of IGF-I is bound to IGFBP-3, the most abundant of the six IGFBP isoforms found in humans [42], thereby reducing its bioavailability. However, high circulating levels of insulin can decrease the production of IGFBP-1 and 2, resulting in a larger amount of bioavailable IGF-1 [19]. IGF-1 levels are typically measured in serum or plasma to estimate the amount of bioavailable IGF-1. Most, but not all, epidemiological studies suggest a nonlinear relationship between IGF-1 and body fatness, as measured by BMI and waist circumference. Lean subjects (BMI < 25 kg/m2 ) exhibit positive associations between IGF-I levels and body fatness while overweight and obese subjects (BMI > 25 kg/m2 ) show inverse associations between IGF-1 levels and body fatness [44]. The divergence observed may be due, in part, to differences in laboratory methods, the use of plasma versus serum, the presence of type II diabetes [45], decreased circulating levels of GH with increased levels of body fatness [44], or current use of hormone replacement therapy [46]. Associations between levels of IGFBP1 and IGFBP-3 and body fatness are also inconsistent across studies; however, most suggest an inverse association between IGF-binding proteins and body fatness and a positive association between the ratio of IGF-1 to IGFBPs (IGF-1:IGFBP-3; IGF-1:IGFBP-1) and body fatness [44, 46]. Associations between circulating levels of IGF-1 and IGF-binding proteins vary across different cancer types. A recent meta-analysis involving 21 studies (3,609 cases and 7,137 controls) showed that higher IGF-1 levels are associated with an
138
N.L. Nock and N.A. Berger
increased risk of premenopausal breast and prostate cancers [47]. Increased circulating concentrations of IGF-1 and acromegaly, a condition arising from sustained hypersecretion of GH and elevated IGF-1, have also been associated with increased risk of colonic neoplasia [48]. Although IGF-binding proteins are generally believed to be inversely associated with increased cancer risk, the studies, even within the same cancer type, are too divergent to draw any general conclusions [49]. The level of bioavailable IGF-1 is believed to be a critical factor in carcinogenesis because IGF-1 binding to its receptor (IGF-1R) induces a variety of actions, such as mitogenesis, anti-apoptosis, and pro-angiogenesis, which favor tumor growth. As shown in Fig. 5.2, the signaling cascades induced by IGF-1 binding to IGF-R1, a transmembrane tyrosine kinase receptor, are similar to those discussed for insulin and include PI3K and extracellular signal-regulated kinase (ERK) mitogen-activated protein kinase (MAPK) pathways [50]. In vitro studies have consistently shown that IGF-1 enhances the growth of cancer cells in various cancer cell lines [51] and IGF-R1 has been found to be overexpressed in many tumors [52]. IGF-1 may also act indirectly through interactions with the tumor suppressor gene, p53 [53, 54]. Furthermore, IGF-I may also promote migration of cancer cells through the activation of IGF-1R [38], which may be important in tumor metastases. On the other hand, the binding of IGF-1 to IGFBP-3 can invoke apoptosis through interactions with caspase-7 and caspase-8 [43]. Recent evidence suggests cross talk between IGF-1 and other factors potentially involved in obesity-related carcinogenesis including leptin, estrogen, and prolactin which may synergize the processes associated with neoplastic progression in human cancer cells, particularly breast cancer cells [55–57]. In particular, cross talk between IGF-1 and estrogen receptor (ER) signaling pathways (discussed further in Section 3) results in synergistic growth, whereby estrogen enhances IGF signaling by inducing expression of IGF-R1 and downstream signaling molecules (IRS-1, IRS-2), which, in turn, result in enhanced tyrosine phosphorylation of IRS-1 followed by enhanced MAPK activation [58].
3 Sex Steroid Hormones Steroid hormones are lipids characterized by a carbon skeleton with four fused rings and a variety of functional side groups. Corticosteroids are synthesized primarily by the adrenal glands and endogenous sex steroid hormones including estrogens (estrone (E1), estradiol (E2)), androgens (testosterone, androstenedione, dihydrotestosterone (DHT), dehydroepiandrosterone (DHEA)), and progestogens (progesterone), which are responsible for the development of secondary sex characteristics and are synthesized primarily in gonadal tissues. In females, estrogen and progesterone are made primarily in the ovary (and in the placenta during pregnancy). Estrogens promote the development of female sex characteristics as well as help regulate the menstrual cycle by thickening the endometrium. Progesterone converts the endometrium to its secretory stage (primarily to prepare
5
Obesity and Cancer: Overview of Mechanisms
139
the uterus for implantation) and, therefore, levels are relatively low during the pre-ovulatory phase of the menstrual cycle, rise after ovulation, and remain elevated during the luteal phase. Estrogen and progestin production by the ovaries are regulated by hypothalamic production of follicle-stimulating hormone (FSH) and luteinizing hormone (LH). In males, androgens function to support sperm production and development of testes, which are the predominant organ synthesizing testosterone. Higher levels of androgens may be responsible for the increased skeletal muscle mass observed in males [59] compared to females. Estrogens and androgens also play a role in the development and regulation of metabolic-related systems including the hypothalamus–pituitary axis (HPA), bone metabolism, and cardiovascular systems [60]. Although estrogens and progestogens are generally considered ‘female sex hormones’ and androgens considered ‘male sex hormones,’ both types are present in each gender at differing levels. Sex steroid hormones are ultimately synthesized from cholesterol and, notably, estrogen can be synthesized from testosterone via aromatase enzymes as shown in Fig. 5.1. This peripheral estrogen synthesis may be an important mechanism in postmenopausal women, particularly those who are obese, since excess central adiposity has been associated with increased levels of aromatase in postmenopausal but not premenopausal women [61]. Furthermore, obesity has been associated with higher levels of testosterone in postmenopausal women [62]. In contrast, in men, obesity as measured by BMI and waist circumference has been associated with lower levels of total and free testosterone [63, 64]. Levels of sex hormone-binding globulin (SHBG), which can bind and deactivate bioavailable estrogens and androgens (implications of which are discussed further below), are also affected by levels of adiposity with obese individuals (BMI > 30.0 kg/m2 ) having as much as 50% less SHBG than thin individuals (BMI < 22.0 kg/m2 ) [65, 66]. Weight loss has also been associated with lower levels of estrone and total and bioavailable estradiol in obese breast cancer survivors [67]. Not surprisingly, sex steroid hormones have been associated with cancers that are considered ‘hormone’ dependent including breast, uterine, and ovarian. A pooled analysis of nine prospective cohort studies showed that higher levels of circulating estrogens (estrone, estradiol) and androgens (testosterone, androstenedione, DHEA) and lower levels of SHBG were associated with an increased risk of postmenopausal breast cancer, and this risk may be driven predominantly by the parallel increase in circulating estradiol with increased BMI [62]. Similarly, higher levels of testosterone and androstenedione were associated with increased premenopausal breast cancer risk in the European Prospective Investigation into Cancer and Nutrition (EPIC) study but they found no association with SHBG [68]. Interestingly, higher levels of androgens (androstenedione, testosterone) have been associated with increased endometrial cancer risk in both pre- and postmenopausal women, but increased levels of estrogens (estradiol, estrone) only appear to increase postmenopausal endometrial cancer risk [69]. The relationship between sex steroid hormones and prostate cancer is more complex. Epidemiological studies of associations between circulating levels of
140
N.L. Nock and N.A. Berger
androgens and prostate cancer have been inconsistent and a recent pooled analysis of 18 prospective studies (3,886 cases and 6,438 healthy controls) showed no association between serum androgen levels and prostate cancer risk [70]. However, Hsing et al. [71] argue that circulating levels of testosterone do not adequately reflect androgen action in the prostate and that dihydrotestosterone (DHT) and the androgen receptor (AR) have key roles in prostate carcinogenesis. Furthermore, obese men appear to have a differential risk for prostate cancer depending on the aggressiveness of the disease – with a decreased risk observed in men with low-grade prostate cancer and an increased risk in men with high-grade disease [72], which may be driven by lower serum levels of testosterone and SHBG in obese compared to normal weight men as discussed above. Mechanistically, sex steroid hormones could have multiple functions in obesityrelated carcinogenesis. First and foremost, when estrogen binds to its receptor (ER), it activates nuclear and transcriptional processes, regulating intracellular signaling pathways such as MAPK, which may stimulate initiated cancer cells in the Go /G1 resting phase to progress through the G1 –S phase and complete cell division, leading to tumor progression. In addition, estrogens may also be involved in cancer initiation because their metabolism may produce multiple forms of DNA damage, which, if not repaired, could lead to DNA mutation(s). Specifically, aromatase (CYP19) metabolizes testosterone to estradiol (E2) (Fig. 5.1) and androstendione to estrone (E1), which can be further metabolized to E2 by CYP17-β HSD; thus, enhanced expression of aromatase from excess adipose tissue (obesity) could lead to higher levels of unbound estradiol. The E2 may then be further metabolized by CYP1A1 and CYP1B1 to 2-OH-E2 and 4-OH-E2, respectively, and, if these metabolites, particularly 4-OH-E2, are not methylated and detoxified by COMT, they may generate E2-quinones, which could directly bind to DNA (creating bulky DNA adducts). Otherwise, the E2-quinones could enter into a futile redox cycle, generating substantial quantities of reactive oxygen species (ROS) that could cause oxidative DNA base lesions (see Section 8; Table 5.1). Estrogens may also interact with the IGF system (see Section 2) and inhibit apoptosis, at least in the endometrium [19]. Furthermore, progesterone levels, which would normally diminish the proliferative actions of estrogens in the endometrium by stimulating the breakdown of estradiol and inducing the synthesis of IGFBP-1 [19, 69], become altered during menopause transitioning (perimenopause) and during chronic anovulation, which may occur as a result of polycystic ovary syndrome (PCOS) [69], yielding variable periods of ‘unopposed estrogen.’ The prevailing hypothesis of ‘unopposed estrogen’ is that unbound, biologically active estrogen leads to increased mitogenesis of endometrial tissue, thus increasing endometrial cancer risk [61]. Interestingly, a recent study using urinary biomarkers of estrogen and progesterone concluded that the total number of days of ‘unopposed estrogen’ exposure may be much greater than previously appreciated and that variation in the amount of time spent in the menopause transition may be an important risk factor for ovarian, breast, and endometrial cancers [73]. However, additional cohort studies evaluating estrogen and progesterone urinary biomarkers in obese women as
5
Obesity and Cancer: Overview of Mechanisms
141
they transition through menopause may lend additional insight to the risk associated with longer bouts of ‘unopposed estrogen.’ The mechanistic role for androgens in hormonal cancers is complex. Androgens may either stimulate or inhibit proliferation of normal epithelial and cancer mammary cells [74]. The underlying mechanisms driving these opposing actions are not well understood but may depend, in part, on the availability of estrogens. In the absence of estrogens, androgens stimulate breast cancer cell growth via binding to ER-α but, in the presence of estrogens, androgens act as anti-estrogens by inhibiting growth of breast cancer cells via the androgen receptor [74]. The ‘androgen hypothesis’ for ovarian cancer, which speculates that androgens stimulate epithelial cell proliferation, was initially proposed by Risch in 1998 [75]; however, direct evidence to support this hypothesis is lacking [76]. Furthermore, although androgen ablation has been the primary non-surgical treatment for prostate cancer, many prostate tumors eventually become refractory to this treatment, which could involve the differential proliferative, apoptotic, and angiogenic events following androgen activation of its receptor [77]. Other factors that may contribute to enhanced androgen receptor (AR) signaling in an androgen-depleted environment include mutations in the AR that allow activation by other steroid hormones such as progesterone and estrogens [78]. Nevertheless, the presence of elevated levels of both estrogens and androgens may increase the risk of breast, endometrial, and prostate cancers, since treating rodents with both estrogens and androgens appears to have a synergistic effect on uterine and mammary cancers in female mice [79] and prostatic hyperplasia in male rats [80]. There is also evidence for cross talk between sex steroid hormones and other putative mechanisms in obesity-related carcinogenesis. For example, cross talk between estrogen and IGF-1 appears to enhance the transcriptional activation of the ER leading to even higher levels of cellular proliferation than either acting alone, and the function of ER-α may be required to maintain IGF signaling [57]. Estrogen and insulin may act synergistically to promote cell cycle progression through the differential regulation of c-Myc and cyclin D1, which are downstream targets of Wnt signaling [81]. There is also evidence for cross talk between the androgen receptor and cytokines (e.g., IL-6) and growth factors (e.g., IGF-1R) [78], which may enhance AR activation of AKT and MAPK signaling pathways and exacerbate tumor progression.
4 Adipokines and Gut Hormones Polypeptide hormones derived from adipocytes are known as ‘adipokines’ and over 50 different adipokines have been identified [82]. The most abundant and wellstudied adipokines with a potential role in obesity-related carcinogenesis include leptin, adiponectin, resistin, and visfatin, which we discuss in further detail in this section. We also discuss ghrelin, which is not secreted by adipoytes but by the stomach mucosa, in this section (cytokines and chemokines are discussed in Section 5).
142
N.L. Nock and N.A. Berger
4.1 Leptin Leptin is a 16-kDa protein hormone that is secreted primarily by white adipocytes in a proportion nearly linear to the level of adipose tissue [83]. Leptin plays a key role in regulating energy balance by binding to its receptors, which are members of the cytokine family of transmembrane receptors, located in the hypothalamus. Although at least six leptin receptor isoforms exist primarily as a result of mRNA alternative splicing, only the long form (OB-Rb), which has an intracellular domain, appears to be involved in hypothalamic regulation of energy systems in humans [84]. Leptin acts by inhibiting the activity of orexigenic pathway neurons such as neuropeptide Y (NPY) and agouti-related peptide (AGRP) and, acts by increasing the activity of anorexigenic pathway neurons such as cocaine- and amphetamine-regulated transcript (CART) and proopiomelanocortin (POMC), leading to elevated levels of α-melanocyte-stimulating hormone (α-MSH) that, in turn, decrease appetite by binding to melanocortin receptors [85]. Thus, as adiposity increases, compensatory levels of leptin should rise to decrease dietary intake (via orexogenic pathways) and enhance energy expenditure (via anorexigenic pathways). However, leptin may be involved in more than just regulating energy systems as it appears to play a role in the development and maintenance of reproductive tissues [86] and in innate and adaptive immune responses [87]. Leptin has also been shown to be expressed in colorectal and mammary epithelial tissues [88]. Although leptin was not formally discovered until 1994 by Friedman and colleagues [89], the effects of a deficiency in leptin were observed four decades earlier in a colony of mice that developed obesity through a random mutation, which was designated by the symbol ‘ob’ [90]. Specifically, they observed that mice homozygous for this mutation (ob/ob) had a rapid increase in weight, which became noticeable at 4–6 weeks of age, and grew to about four times the weight of normal animals [90]. In humans, mutations in the leptin (LEP) and leptin receptor (LEPR) genes contribute to severe obesity and other rare metabolic disorders (see Chapter 4). Epidemiological studies have shown that obese individuals have higher circulating levels of leptin compared to normal weight individuals and may become insensitive (resistant) to the action of leptin [91], much like individuals with type 2 diabetes become resistant to insulin action. Furthermore, some research has shown that circulating leptin levels may be higher in women than in men [92], which may be attributed, in part, to higher levels of subcutaneous fat in females compared to males [93, 94]. The association between circulating levels of leptin and cancer is inconsistent across cancer types. A recent review indicated that only 3 out of 10 studies reported a positive association between increased leptin levels and increased breast cancer risk [95]. The potential relationship between leptin and endometrial cancer has not been well studied but there is some evidence that higher serum leptin levels are associated with increased risk of endometrial cancer and endometrial hyperplasia (EH) compared to levels found in women with a normal endometrium [96]. Higher circulating
5
Obesity and Cancer: Overview of Mechanisms
143
levels of leptin have also been associated with increased prostate cancer risk, particularly more aggressive prostate cancer [97], but not all studies have observed this effect [98]. Several studies have observed an association between higher leptin levels and increased risk of colon cancer [99, 100] and colorectal adenomas [101, 102]; however, these associations were only found to be statistically significant in men and not women [99, 101]. Why leptin may play a more important role in men compared to women in manifesting certain cancers is not well understood. Some researchers have suggested that heterogeneity in endogenous sex steroid hormones and certain LEPR genotypes may be partially responsible for the observed sex differences in the relationship between obesity and colorectal adenoma risk [101]. However, leptin concentrations appear to be higher in subcutaneous compared to visceral fat [106] and, as mentioned briefly above, women have a higher proportion of subcutaneous fat than men [93, 94]. Moreover, leptin has been shown to upregulate the transcription of aromatase [107] and estrogen may be involved in stimulating leptin secretion [108]. Thus, one might expect leptin to have a greater role in women, particularly obese postmenopausal women, and not men. Nevertheless, as indicated in Fig. 5.1, leptin has been shown to exert mitogenic, anti-apoptotic, and angiogenic effects in several cell lines [95, 109, 110, 111] and, therefore, may be involved in tumor growth and metastasis. Mechanistic studies have also shown that human colon cancer cell lines treated with leptin produce higher numbers of cells, an effect which is most likely mediated through the PI3K pathway [103], and that preneoplastic colon epithelial cells treated with leptin orchestrate VEGF-driven angiogenesis [104]. Higher serum leptin levels have also been associated with an increased risk of Barrett s esophagus, a precursor condition to esophageal adenocarcinoma, but this association was only statistically significant among men and not women [105]. Leptin’s role in breast cancer may be unique in that it may also depend on the specific leptin receptor isoform that leptin binds to, which may invoke a different signaling pathway. For example, ER-positive MCF-7 and T47D cell lines express substantial quantities of the long form of the leptin receptor (LRb) but ER-negative breast cancer cell lines express only the short form. This, in turn, may lead to differences in the cell cycle signaling pathway activated since LRb can activate both STAT and MAPK, whereas the short form of the leptin receptor primarily activates only MAPK (Fig. 5.2) [95]. Interestingly, studies conducted with A-Zip/F-1 ‘fatless’ mice, which have no white adipose tissue or detectable serum levels of leptin or adiponectin but display accelerated tumor formation, suggest that adipokines may not play an integral role in the enhancement of tumor development [112]. Furthermore, the MMTV-Wnt-1 transplantable breast cancer model showed a lower incidence of breast cancer and generated smaller tumors in ob/ob leptin-deficient mice compared to wild-type control mice (Reizes O, Hursting SD, 2009 Failure of breast cancer progression in leptin deficient mice, Personal Communication), which supports an important role for leptin in breast tumor development.
144
N.L. Nock and N.A. Berger
4.2 Adiponectin Adiponectin is a 30-kDa protein hormone produced by mature adipocytes, which is abundant in the serum at a concentration ranging between 1 to 15 μg/ml. Adiponectin has a complex assembly involving a low molecular weight (LMW) trimer, a middle molecular weight (MMW) hexamer, and a high molecular weight (HMW) multimer, with the LMW oligomers being the most predominant form found in the circulation and the HMW oligomers representing the most abundant intracellular form [113]. As shown in Fig. 5.2, adiponectin is involved in the regulation of a number of metabolic processes including glucose and fatty acid catabolism and may be an important insulin-sensitizing agent since adiponectin replacement experimentally is able to diminish insulin resistance [114]. Adiponectin (ADIPO) acts by binding to its receptors ADIPO-R1 and ADIPO-R2, which are most highly expressed in the skeletal muscle and liver, respectively [115]. This binding leads to the activation of AMP protein kinase (AMPK) pathway, subsequent inhibition of acetyl CoA carboxylase and an increase in fatty acid β-oxidation [116]. Increased glucose utilization and fatty acid oxidation in skeletal muscle together with inhibition of liver gluconeogenesis may result in increased insulin sensitivity [117]. Adiponectin levels correlate negatively with obesity. Specifically, lower levels of adiponectin have been associated with increased waist circumference, visceral fat, and obesity in both adults and children [116, 118]. Furthermore, weight loss appears to significantly increase circulating levels of adiponectin in obese women [119]. However, adiponectin levels were not associated with a change in weight over 6 years in a cohort of older (50–77 years of age at baseline), non-diabetic men and women [120]. Future studies would benefit from researchers clearly identifying which oligomeric form(s) of adiponectin are being measured, particularly since the HMW form may be a better marker of ectopic fat [113]. The epidemiological evidence for an inverse association between adiponectin levels and cancer risk has been fairly consistent across cancer types. For example, lower concentrations of adiponectin have been associated with increased risk of endometrial cancer in premenopausal and postmenopausal women [121] and with an increased risk of endometrial cancer and endometrial hyperplasia in obese women [122]. Lower adiponectin levels have been fairly consistently associated with an increased risk of breast cancer in postmenopausal but not premenopausal women [123, 124]. Although there are far fewer reports, lower adiponectin levels have been observed to increase colorectal cancer in men [125] and may have prognostic significance in non-metastatic disease [126]. The inverse association between adiponectin and colon cancer is further supported by mechanistic studies which demonstrate that adiponectin deficiency induced by azoxymethane enhances colorectal carcinogenesis and liver tumor formation in mice [127]. Lower adiponectin levels have been associated with prostate cancer and may be substantially lower in more aggressive or advanced forms of the disease [128, 129]. A protective role for adiponectin in carcinogenesis is not well developed but may involve upregulation of the peroxisome proliferator-activated receptor alpha
5
Obesity and Cancer: Overview of Mechanisms
145
(PPAR-α) pathway, which controls cellular proliferation and differentiation, including adipocyte differentiation [117]. Tumor inhibition may also involve inactivation of the MAPK pathway and/or anti-angiogenesis via caspase-mediated apoptosis of endothelial cells [130]. Adiponectin also appears to interact with inflammatoryrelated factors secreted by adipocytes including TNF-α and IL-6 (see Section 5). TNF-α and IL-6 can decrease adiponectin expression and adiponectin may be able to counteract the pro-inflammatory effects of TNF-α and IL-6 [131, 132]. Since adiponectin is an insulin-sensitizing agent, it may also act indirectly through insulin-related carcinogenic mechanisms (see Section 2). However, adiponectin’s putative role in carcinogenesis has recently been challenged by researchers who have observed accelerated tumor formation in A-Zip/F-1 ‘fatless’ mice despite the fact that these mice have no detectable levels of adiponectin (or leptin) [112].
4.3 Resistin Resistin is a 12.5-kDa polypeptide hormone produced by adipocytes in rodents and immunocompetent cells in humans [133]. The physiological role for resistin in humans remains controversial. Although initially thought to provide a link between inflammation and insulin resistance, resistin now appears to have a greater role in inflammatory-related processes rather than in insulin sensitivity [134]. In particular, resistin has been shown to increase transcriptional events leading to higher expression of pro-inflammatory cytokines including TNF-α and IL-6 [135]. Although resistin is not secreted by adipocytes in humans, serum levels of resistin have been shown to be elevated in obese compared to lean subjects [136]. In addition, resistin levels have been positively correlated with the homeostasis model assessment ratio (HOMA-R), a marker for insulin resistance as determined by fasting insulin and glucose levels, in some [137] but not all [136] studies. However, circulating resistin levels have not been associated with type II diabetes risk after adjustment for BMI [138]. Interestingly, resistin levels appear to differ by gender with significantly higher levels observed in women compared to men [137, 138]. The potential association between circulating resistin levels and various cancer types has not been well explored and studies published thus far have been in fairly small sample sizes. One study in Korean women reported that higher serum resistin levels were associated with breast cancer and that resistin levels increased with tumor grade [139]. Additional support for resistin’s potential role in breast cancer comes from a study in Chinese women whereby serum levels of resistin were found to be significantly higher in breast cancer patients versus healthy controls and in patients with lymph node metastasis (LNM) compared to those without LNM [140]. Higher resistin levels have been observed in colon cancer patients compared to healthy controls but levels were not significantly different between patients with colon cancer and those with colon adenomas [102]. Serum resistin levels did not differ significantly in prostate cancer patients compared to men with benign prostate hyperplasia [141].
146
N.L. Nock and N.A. Berger
The potential mechanisms for resistin in cancer are not well understood but may involve inflammatory (see Section 5) and angiogenic pathways. mRNA expression of resistin has been shown to be induced by TNF-α and IL-6, and insulin sensitizers with anti-inflammatory properties, including a synthetic PPARγ agonist (rosiglitazone), as well as aspirin, have been shown to suppress resistin expression in macrophages [135]. Furthermore, resistin induces human endothelial cell proliferation and migration and, upregulates expression of vascular endothelial growth factor receptors (VEGFR-1, VEGFR-2) and matrix metalloproteinases (MMP-1, MMP-2) at both the mRNA and protein levels [142], suggesting a potential role for resistin in tumor progression and metastasis.
4.4 Visfatin Visfatin occurs in both intracellular and extracellular forms with some serum visfatin appearing at 50 kDa but most is found at 100 kDa, suggesting that it primarily circulates in multimeric forms. Visfatin is a relatively new member of the adipokine family, which is believed to be secreted by visceral fat and to mimic the effects of insulin [143]. Visfatin was first identified as pre-B cell colony-enhancing factor (PBEF) based on its ability to work in association with IL-7 and as a stem cell factor to increase pre-B cell colony-forming activity [144, 145]. Visfatin has additional pro-inflammatory activities including stimulation of cytokines IL-6 and IL-8 in cultured amniotic cells. Visfatin is elevated in obese children, adolescents, and women with polycystic ovary syndrome and in patients with type 2 diabetes mellitus [144–146]. Increased visfatin has also been correlated with increased HDL, cholesterol, leptin, and IL-6 [144–146]. Serum visfatin is increased in patients with inflammatory bowel disease and sepsis. Visfatin has also been implicated in inducing vascular endothelial growth factor [144–146], and elevated levels have been observed in patients with colorectal cancer [147, 148]. Visfatin stimulates glucose uptake and, although initially thought to bind to and activate the insulin receptor, this has not been substantiated [149]. Nevertheless, visfatin is associated with phosphorylation of several components of the insulin signaling pathway including the insulin receptor, IRS-1, and IRS-2, as well as with glucose utilization in selected cell types [145]. Both visfatin and PBEF are identical to the circulating, extracellular form of the enzyme, nicotinamide phosporibosyltransferase (Nampt), which converts nicotinic acid to nicotinamide mononucleotide (NMN), the rate-limiting step for enzymatic synthesis of NAD, a critical cofactor for oxidation–reduction processes, DNA repair process, and many other signaling processes [139]. Therefore, it is probable that many of the observed functions for visfatin are related to a regulatory function determined by effects on NMN or NAD levels. For example, the Nampt-mediated synthesis of NAD content could serve to regulate glucose-stimulated insulin secretion which could contribute to its systemic effect on regulation of glucose levels. Interestingly, NMN biosynthesis and
5
Obesity and Cancer: Overview of Mechanisms
147
NAD levels have been implicated in regulation of the NAD-dependent deacetylase, SIRT1, which modulates circadian rhythm [150] and may contribute to sleep disturbances, an important determinant of obesity [144, 149].
4.5 Ghrelin Ghrelin, a 28-amino acid peptide which is acylated at the serine 3 position with an octanoyl group from the ghrelin gene, is predominantly secreted from the stomach mucosa. Ghrelin acts by binding to its type 1a growth hormone secretagogue receptor (GHSR1a) to release growth hormone and acts as an orexigenic signal from the gut to the brain to stimulate appetite by activating NPY and AGRP neurons in the arcuate nucleus of the hypothamalus [151]. During fasting, ghrelin levels rise and fall to a nadir within 1 h of eating [151]. In addition to the stomach and hypothalamus, ghrelin has been shown to be expressed in several other tissues including the small and large intestines, pancreas, and testes [151] and may be involved in functions regulating tissue growth and development. Non-acylated forms of the ghrelin gene product such as obestatin, a 23-amino acid peptide, appear to have opposite physiological actions (e.g., induce satiety/decrease food intake); however, the receptors for non-acylated forms and their mechanisms of action have not been well established [152] and are not discussed further here. In contrast to leptin, serum levels of ghrelin are generally inversely associated with BMI and body fat and may differ by gender. Serum levels of ghrelin appear to be higher in women compared to men, particularly during the follicular phase of menstruation [151]. Studies have also shown that fasting serum ghrelin concentrations are significantly lower in overweight and obese compared to normal weight (BMI < 25 kg/m2 ) adults [153]. This evidence, however, does not support overt dysregulation of ghrelin in obesity and, therefore, evaluating changes in fasting ghrelin levels with weight loss (or gain) may be more informative. In contrast to studies involving gastric banding, which have shown no change in ghrelin levels with weight loss (median weight loss of 45.7%) [154], ghrelin levels have been observed to decrease significantly after substantial weight loss (mean weight loss of 62.5%) following Roux-en-Y gastric bypass surgery [155]. This suggests drastic physical change in the gastrointestinal tract may be required to induce ghrelin-related appetite suppression. Reports evaluating serum levels of ghrelin in cancer patients compared to healthy controls are scant. One study reported that serum levels of ghrelin were not significantly different in prostate cancer patients compared to men with benign prostatic hyperplasia [156]. Plasma ghrelin levels were reported to be higher among breast and colon cancer patients with cachexia, which is a complex state characterized by the loss of muscle mass and adipose tissue together with anorexia, compared to noncachectic patients [157]. Putative mechanisms for ghrelin in cancer are lacking. However, ghrelin and its receptors have been observed to be upregulated in breast, endometrial, prostate
148
N.L. Nock and N.A. Berger
and pancreatic cancer tissue and/or cell lines, and these increases in ghrelin have been shown to increase cancer cell proliferation [151, 158]. Thus, ghrelin could play a role in tumor progression but additional studies are clearly needed to better understand its role in carcinogenesis.
5 Inflammatory Factors: Cytokines and Chemokines Cytokines are a group of diverse polypeptides secreted by multiple cell types, which are involved in a variety of cell signaling, particularly in inflammatory and immunological processes. The work of Hotamisligil and colleagues [159] conducted over 15 years ago, which first described elevated cytokine (tumor necrosis factor alpha (TNF-α)) levels in diet-induced obese rodents, has helped pave the way for a paradigm shift in the way adipose tissue is viewed. Instead of adipose tissue being thought of as just a site of energy storage, it is viewed as a biologically active endocrine organ. It is now widely accepted that obesity induces a state of chronic, low-grade inflammation resulting in the increase of several inflammatoryrelated molecules such as TNF-α, interleukin-6 (IL-6), C-reactive protein (CRP), and monocyte chemoattractant protein-1 (MCP-1). It is also widely accepted that chronic inflammation is associated with processes that contribute to the onset and/or progression of several types of cancer including colorectal [160], prostate [161], and esophageal [162]. Therefore, it is quite plausible that the chronic inflammation is a mediator of the relationship between obesity and carcinogenesis. In the sections that follow, we present the most well-studied cytokines and chemokines released by adipose tissue and discuss their putative role(s) in the carcinogenic process. However, it is important to note that secretion of a substance from adipose tissue does not necessarily imply the substance is synthesized by adipocytes. Recent evidence shows that obese individuals have an increased infiltration of macrophages into their adipose tissue and that these macrophages, which are mononuclear phagocytes whose primary function is to provide defense against invading foreign organisms, may be the primary source of the inflammatory cytokines [163]. However, it is not clear what actually causes the increased macrophage infiltration. Hypothesized processes responsible for macrophage invasion include altered adipocyte size and adipokine signaling, nutritional induction of metabolic endotoxemia or reduced angiogenesis, and local adipose cell hypoxia [163]. MCP-1 (discussed further below) has also been shown to contribute to macrophage infiltration into adipose tissue [164].
5.1 Tumor Necrosis Factor Alpha (TNF-α) The cytokine TNF-α is produced primarily as a 212-amino acid type II transmembrane protein, which forms a 51-kDa soluble homotrimeric cytokine (sTNF) and a 17-kDa protomer. TNF-α is produced mainly by macrophages and is secreted
5
Obesity and Cancer: Overview of Mechanisms
149
in response to lipopolysaccharide (a component of gram-negative bacteria) and interleukin-1 (IL-1). TNF-α has two receptors, TNF-R1 and TNF-R2. Although TNF-R1 is expressed in most tissues and can be fully activated by both the membrane-bound and soluble form, TNF-R2 is found only in immune system cells and can activate only the membrane-bound form. The binding of TNF-α to TNFR1 leads to the binding of TNF-RSF1A-associated death domain (TRADD) and the subsequent activation of nuclear factor kappa B (NF-κB), MAPK, or caspase-8 signaling cascades, which induce transcription of proteins involved in inflammatory response, cell proliferation, cell differentiation, and apoptosis (discussed further below). TNF-α is expressed in adipocytes but whether or not it is secreted independent of the presence of macrophages in human white adipose tissue has been debated [165]. TNF-α expression in white adipose tissue was first identified in obese rodents, where TNF-α expression was higher and shown to regulate insulin action in ob/ob mice [159]. Obese mice lacking TNF-α or its receptor were found to be protected against developing insulin resistance [166]. TNF-α appears to have a role in human adipocytes as a regulator of the synthesis and release of other cytokines (e.g., IL-6) as well as a regulator of apoptosis [165]. TNF-α may also alter adipocyte-related energy metabolism by inhibiting lipoprotein lipase, stimulating hormone-sensitive lipase, inducing uncoupling protein expression, and downregulating insulin-stimulated glucose uptake [167]. The extent to which TNF-α produced by human adipocytes is released into the circulation has also been a matter of debate; however, several studies have shown that TNF-α levels are higher in obese compared to normal weight subjects [85, 168]. Furthermore, TNF-α has been shown to decrease substantially with weight loss in obese subjects [169, 170]. Epidemiological studies examining associations between circulating levels of TNF-α and cancer outcomes are relatively scant, perhaps, because understanding the specific source of its production is much more informative than circulating levels which may arise due to infection, tissue damage, or various disease states. Nevertheless, the Health Aging and Body Composition cohort study of older adults (70–79 years) found that higher baseline plasma levels of TNF-α were more strongly associated with death from any cancer than incidence of any cancer at an average follow-up of 5.5 years but, site-specific analyses revealed no association between TNF-α levels with colon, breast, or prostate cancer [171]. Although no other studies examining TNF-α on colorectal cancer have been reported, higher levels of TNF-α have recently been associated with increased colorectal adenomas [172]. Moreover, significantly higher levels of serum concentrations of TNF-α have been observed in metastatic compared to localized prostate cancer [173], lending more evidence for the prognostic value of this marker. However, serum concentrations of TNF-α were not found to be significantly different in a small sample of overweight and obese breast cancer patients compared to BMI-matched controls [174]. Although TNF-α is toxic to tumor cells at high doses [175], physiological levels may be involved in several mechanisms linking obesity to carcinogenesis. First, exposure to TNF-α promotes insulin resistance by activating
150
N.L. Nock and N.A. Berger
pro-inflammatory pathways that modify glucose uptake in adipocytes (and myocytes) and by impairing insulin signaling at the level of the insulin receptor substrate (IRS) proteins mainly through the expression of protein tyrosine phosphatase (PTP)1B [176]. Furthermore, it has been hypothesized that the absence or inhibition of PTP1B in insulin target tissues could confer protection against insulin resistance induced by TNF-α [176]. As mentioned above, the binding of TNF-α to TNF-R1 induces a series of intracellular events that results in the activation of NF-κB, phosphatidylinositol 3-kinase (PI3K)/Akt, and c-Jun NH(2)terminal kinase (JNK), which can inhibit cell apoptosis [177]. However, JNK may induce pro- or anti-apoptotic functions, depending on the cell type, nature of the death stimulus, duration of its activation, and the activity of other signaling pathways [178]. In particular, peroxisome proliferator-activated receptors (PPAR) may antagonize the activities of NF-κB. In addition, TNF-α modulates the synthesis of other enzymes in the adipose milieu including IL-6 [179] and estrogen through the stimulation of aromatase [180], which suggests a role for TNF-α in obesity-related carcinogenesis, particularly among postmenopausal women.
5.2 Interleukin-6 (IL-6) Interleukin-6 (IL-6), which is also known as interferon-beta 2, is a 26-kDa protein with a myriad of functions. IL-6 is secreted by macrophages to stimulate B-cell immune response and by osteoblasts to stimulate osteoclast formation. IL-6 is also known as a ‘myokine,’ a cytokine produced from muscle, which increases in response to muscle contraction. In muscle and adipose tissue, IL-6 stimulates energy mobilization leading to increased body temperature. IL-6 binding to its receptor (IL-6R) activates a signaling cascade through Janus kinases (JAK) and signal transducers and activators of transcription (STAT) factors. Although IL-6 has receptors in the hypothalamus of mice, suggesting it has a direct role in the regulation of energy homeostasis, no IL-6 receptors in the human hypothalamus have been identified [181, 182]. A soluble form of the receptor (sIL-6R), however, has been identified in human serum, and IL-6 binding to these soluble receptors appears to yield neuronal interaction. Circulating levels of IL-6 have been shown to be elevated in obese compared to normal weight subjects [165]. Fasting plasma IL-6 concentrations have also been positively correlated with percent body fat [183]. Moreover, weight loss results in a significantly decreased levels of IL-6 [172, 173]. Circulating levels of IL-6 have been associated with several cancers but, as with TNF-α, the association seems to be stronger in predicting cancer mortality than cancer incidence. For example, although higher baseline plasma levels of IL-6 were associated with death from any cancer, they were not associated with incident colon, breast, or prostate cancer [171]. Furthermore, significantly higher levels of serum concentrations of IL-6 have been observed in metastatic compared to localized prostate [173] and breast [184] cancers; and, may have prognostic value, particularly in predicting breast cancer survival [185]. IL-6 levels have been shown to
5
Obesity and Cancer: Overview of Mechanisms
151
be higher in a small sample of overweight and obese breast cancer patients compared to BMI-matched controls [174]. Higher plasma levels of IL-6 have also been associated with the risk of colorectal adenomas [172]. IL-6 is known to stimulate cell growth and inhibit apoptosis [186]; however, IL-6 may have tumor-promoting or tumor-inhibitory effects depending on the presence of other modulating factors. For example, IL-6 expression induced by treatment with leptin has been reported to promote cell proliferation in an Apc Min/+ colon epithelial cell line [187] and, thus, may play more of a role in the progression of mutated cells. IL-6 expression is induced by hypoxia and subsequently able to upregulate vascular endothelial growth factor (VEGF) transcription, suggesting a potential role for IL-6 as an angiogenic factor that facilitates the production and the distribution of VEGF to metastatic sites [185]. Furthermore, because IL-6 promotes osteoclast formation and inhibits dendritic cell proliferation, it may play a role in metastatic growth [188].
5.3 C-Reactive Protein (CRP) C-reactive protein (CRP) is a 117.5-kDa protein released by hepatocytes and adipocytes. Because it is released in response to inflammation, high-sensitivity CRP (hs-CRP) often serves as a non-specific prognostic marker of systemic inflammation for cardiovascular disease with levels >3 mg/L suggesting elevated risk for myocardial infarction and stroke [189]. Higher levels of CRP have been associated with obesity in adults, irrespective of ethnicity [190]. Furthermore, CRP levels decreased significantly among obese women who lost, on average, 10% of their body weight over 12 months [119]. Similar associations showing decreased levels of CRP with 10% or more weight loss were also observed in an earlier study of obese premenopausal women [191]. Plasma CRP concentrations have been shown to be significantly higher in colorectal cancer patients in several prospective studies [192]; however, this association may be driven by those who developed colon and not rectal cancer [193]. Plasma concentrations of CRP were not associated with prostate cancer risk in two case– control studies [194, 195] or in four prospective studies [196]. CRP levels were not significantly associated with clinical characteristics in a cohort of breast cancer survivors at 31 months following diagnosis [197]. Authors of a review evaluating CRP in various cancers concluded that the limited number of prospective studies conducted provided no strong evidence for a causal role of CRP in cancer [196]. However, additional prospective studies are needed to determine if CRP has prognostic value in colon, prostate, and breast carcinogenesis. The role for CRP in obesity-related carcinogenesis is likely mediated through chronic inflammation (Fig. 5.1). CRP upregulates the expression of adhesion molecules and stimulates the release of IL-6 and TNF-α [198], which suggests CRP may activate mechanisms similar to those previously described for IL-6 and TNF-α.
152
N.L. Nock and N.A. Berger
5.4 Monocyte Chemotactic Protein-1 (MCP-1) Monocyte chemotactic protein-1 (MCP-1), also known as the chemokine C–C motif (CCL2), is a low molecular weight 13-kDa protein. Although its main function is chemotaxis to monocytic cells and recruitment and activation of leukocytes during inflammation, MCP-1 is also produced by smooth muscle cells, bone (osteoblasts, osteoclasts, bone marrow), and endothelial cells. MCP-1 and its receptor, chemokine receptor 2 (CCR2), have been shown to be upregulated in obese subjects (Fig. 5.1) [199]. Visceral adipose tissue, which tends to accumulate around the organs in the abdominal region, has been shown to have higher numbers of macrophages and to express more MCP-1 compared to subcutaneous fat tissue in obese individuals [200], which may contribute to the increased secretion of inflammatory factors from adipose tissue. Higher serum levels of MCP-1 have been associated with less advanced disease in breast cancer patients [201]. In contrast, higher levels of MCP-1 have been associated with advanced stage prostate cancer [202]. MCP-1 appears to be actively involved in the carcinogenic process as MCP-1 and its receptor, CCR2, are upregulated in prostate, breast, and other cancer cells [202]. MCP-1 may be involved in modifying obesity-related carcinogenesis via insulin resistance mechanisms as MCP-1 can lead to the inhibition of serine phosphorylation of IRS-1 and impaired uptake of glucose in human skeletal muscle [163] and possibly other tissues. Interestingly, knockout of either MCP1 in db/db mice [164] or the MCP1 receptor (CCR2) in C57BL/6 J mice fed a high-fat diet [203] attenuated obesity and afforded protection against insulin resistance. MCP-1 has also been shown to play a role in tumor-induced osteoclastogenesis and bone resorption in prostate cancer [202] and may, therefore, play a role in metastasis of other tissue tumors.
6 Angiogenic Factors 6.1 Vascular Endothelial Growth Factor (VEGF) Vascular endothelial growth factor (VEGF) represents a family of growth factors that regulate angiogenesis, which is the growth of blood vessels from pre-existing vasculature. Angiogenesis is critical for nutrient and gas supply to all tissues and, therefore, an important requirement of tumor growth and metastasis. VEGF-A is the most well-studied isoform, which acts as a vasodilator increasing microvascular permeability and stimulates monocyte migration. VEGF is produced in many tissues with high levels in the prostate gland [204]. VEGF-A can bind to receptors (primarily VEGFR-2) stimulating a tyrosine kinase signaling pathway that leads to angiogenesis. Under conditions of low oxygen or ‘hypoxia,’ hypoxia-inducible factor (HIF-1) stimulates the release of VEGF (Fig. 5.2).
5
Obesity and Cancer: Overview of Mechanisms
153
Circulating levels of VEGF have been observed to be higher with increasing BMI in obese and overweight compared to lean subjects [205]. Serum VEGF-A levels were also found to be significantly higher in morbidly obese patients undergoing bariatric surgery compared to lean individuals with levels being significantly lower after weight loss 9–12 months after surgery [206]. Although several smaller clinical studies have shown that VEGF levels are associated with more advanced prostate cancer, pre-diagnostic plasma VEGF levels were not associated with development of prostate cancer in a larger nested case– control study derived from the Physicians’ Health Study cohort [204]. Studies evaluating VEGF levels and breast cancer risk have also been a bit conflicting with some studies reporting that higher levels are associated with more advanced disease [207] and others finding no association [208]. Similar patterns of association with more advanced disease have also been observed in colorectal cancer; however, the association may depend on the use of serum versus plasma, with serum levels being a better predictor of angiogenic tumor activity and colorectal cancer survival [209]. The formation of new blood vessels is essential for tumor growth and, therefore, VEGF may play an intuitive role in the progression of cancer; however, its role in obesity-related carcinogenesis is, perhaps, less obvious. Leptin, which is secreted by adipocytes (see Section 3), has been shown to induce preneoplastic colon epithelial cells to VEGF-driven angiogenesis, which provides a specific mechanism for obesity-associated colon cancer [103]. Cross talk with other obesity-related factors has been reported, whereby dietary restriction in rats reduced autocrine/paracrine IGF-1 expression, which contributed to reduced VEGF expression and signaling and inhibition of prostate tumor angiogenesis [210].
7 Fatty Acid Metabolism: Eicosanoids and Prostaglandins Eicosanoids are signaling molecules synthesized through the oxygenation of 20 carbon omega-6 (ω-6) or omega-3 (ω-3) fatty acids that regulate inflammation and immunity. There are four families of eicosanoids (prostaglandins, prostacyclins, thromboxanes, leukotrienes), which are characterized by the attachment of differing functional groups at distinct locations [211]. Prostaglandin synthesis is initiated by the metabolism of arachidonic acid, an ω-6 fatty acid with 20 carbons and 4 double bonds in cis (ω-6 20:4), by cyclooxygenase enzymes (COX-1, COX-2), which generates the formation of the five-member prostane ring, involving carbons 8–12, from the arachidonic acid molecule. As discussed in more detail below, another pathway of eicosanoid metabolism, independent of COX enzymes, is through free radical catalysis of arachidonic acid resulting in isoprostanes. Lipid peroxidation is necessary for initiation of COX activity [212], which leads to the formation of the prostaglandin precursor, PGH2 . Then, PGH2 generates series-2 prostanoids, most notably, PGE2 . PGE2 is involved in the downregulation of lipolysis in adipocytes
154
N.L. Nock and N.A. Berger
and adipocyte differentiation. Moreover, PGE2 binding to its protein-coupled receptor, EP2, activates the PI3K/AKT pathway, while its binding to EP4 phosphorylates the epidermal growth factor receptor (EGFR) leading to cellular proliferation and anti-apoptosis [213]. PGE2 can be degraded by 15-prostaglandin dehydrogenase (15-PGDH) to downregulate these pathways [213]. Interestingly, knockout mice with a homozygous deletion of 15-PGDH have an increased susceptibility to colon cancer [214]. Higher PGE2 levels have been associated with higher BMI [215] but large-scale epidemiological studies evaluating PGE2 levels in obesity are lacking. Explants grown in tissue culture from a small study of obese humans have shown increased formation of PGE2 compared to adipocytes derived from normal individuals [216], lending some evidence, albeit indirect, that higher PGE2 levels may be secreted by adipocytes in obese subjects. PGE2 levels have been examined in tissue samples extracted from cancer patients. Higher PGE2 levels have been observed in colon tumors compared to normal colonic mucosa [217, 218]. Elevated PGE2 levels have also been found in lung cancer [219, 220]. However, PGE2 levels have not been consistently associated with prostate or breast cancers [221]. Mechanistically, PGE2 levels have been shown to increase the growth and motility of colon cancer cells, interfere with apoptosis, stimulate angiogenesis [213, 222], and stimulate production of IL-6 [223]. Furthermore, non-steroidal antiinflammatory drugs (NSAIDs), which act by inhibiting COX-1 and COX-2, have been shown to substantially lower the risk of colon cancer [213, 224]. However, the use of COX inhibitors has been limited by the relatively recent observation that their use leads to increased adverse cardiovascular events. Interestingly, PGE2 has been shown to stimulate estrogen biosynthesis and a strong linear association between aromatase and COX-1 and COX-2 expression has been observed in breast cancer specimens, which may occur via increases in intracellular cyclic AMP levels [225], suggesting multiple potential roles for PGE2 in breast cancer.
8 Reactive Oxygen Species and Oxidative Stress The term ‘oxidative stress’ is used to define disruption of redox homeostasis or, more specifically, an imbalance in the rate at which the intracellular content of free radicals increases relative to the rate at which the cell scavenges free radicals. Free radicals are generated from many endogenous processes as well as through the metabolism of exogenous agents and include reactive oxygen species (ROS) and reactive nitrogen oxide species (RNOS). In terms of ROS production, the mitochondrial electron transport chain is a major endogenous source of the superoxide anion (O2 –˙), which results from electron leakage between complex I and III [226]. As mentioned in Section 3, the superoxide anion can also be generated from the futile redox cycling of E2 quinones, which may be formed during the metabolism of endogenous estrogen. Similarly, many common
5
Obesity and Cancer: Overview of Mechanisms
155
lipophilic exogenous chemicals (e.g., polycyclic aromatic hydrocarbons (PAH), which are found in cigarette smoke, fuel exhaust, and grilled meats) can generate large quantities of the superoxide anion by futile redox cycling of their quinones [227]. The membrane-associated reduced nicotinamide adenine dinucleotide phosphate (NAD(P)H) oxidase, cytochrome c, and xanthine oxidase complexes can also generate ROS. Nitric oxide synthase (NOS) isozymes (mitochondrial (mtNOS), endothelial (eNOS), inducible (iNOS)) can generate nitric oxide (NO),˙ a RNOS [228]. Notably, the superoxide anion (O2 –˙) can react with nitric oxide (NO)˙ to form the highly reactive RNOS, peroxynitrite (ONOO– ) [229]. If free radicals are not neutralized, they have the capacity to attack proteins, lipids, and nucleic acids (DNA), which may modulate their structure and function and lead to dysregulated mRNA expression and DNA mutations. Therefore, biological systems have developed several lines of defense against ROS and RNOS. In the first line of defense, the superoxide anion and hydrogen peroxide can be scavenged non-enzymatically by vitamin C and E, respectively [230]. O2 –˙ can be metabolized by superoxide dismutase (SOD) to yield hydrogen peroxide (H2 O2 ) and O2 , and, then, the H2 O2 can be neutralized by catalase (CAT) to yield H2 O and O2 [231]. H2 O2 can also be neutralized to O2 through coupled reactions involving catalysis by glutathione peroxidase (GPX) in the presence of reduced glutathione (GSH) [232]. On the other hand, H2 O2 may be converted to the highly reactive hydroxyl radical, HO,˙ in the presence of iron (Fe(II) and Fe(III)) through the Fenton reaction or via the Haber–Weiss reaction [233]. Macrophages, which may invade adipose tissue in obese individuals, may also generate hydrogen peroxide (H2 O2 ). Because free radicals are generally short lived, direct measurement is not feasible and surrogate measures are used in an attempt to quantify oxidative stress levels. The markers fall into oxidized proteins, nucleic acids (DNA), and lipids categories; and, while some represent the final end product of damage, others represent interim products, which can result in further (secondary) effects as discussed below.
8.1 Biomarkers of Oxidative Protein Damage Oxidized proteins are often functionally inactive; however, cells are generally very successful in removing oxidized proteins through proteolysis [234]. Protein carbonyl levels have been used as a biomarker of protein oxidation but a variety of mechanisms can lead to protein carbonyls [234], which limits their usefulness as a marker of oxidative stress.
8.2 Biomarkers of Oxidative Lipid Damage The attack of free radicals on lipids results in lipid peroxidation. Polyunsaturated fatty acids (PUFAs) with two or more double bonds are subject to greater oxidation from free radicals than saturated and monounsaturated fatty acids because of
156
N.L. Nock and N.A. Berger
instability in the hydrogen atom adjacent to the double bond [235]. Oxidized lowdensity lipoprotein cholesterol (Ox-LDL) is a commonly used biomarker of lipid peroxidation, which has been reported to be positively correlated with abdominal visceral adipose tissue in males [236]. Higher circulating Ox-LDL levels have also been associated with increased waist circumference in both men and women independent of body mass index [237]. Furthermore, circulating Ox-LDL levels have been shown to decrease after weight loss in morbidly obese patients following laproscopic banding surgery [238]. Studies measuring Ox-LDL in cancer patients are limited. In one small study, Ox-LDL levels were observed to be higher among breast and ovarian cancer patients compared to healthy control subjects [239]. When comparing the highest quartile to the lowest quartile, serum Ox-LDL levels were found to be associated with increased colorectal cancer risk in a Japanese cohort [240]. Lipid peroxyl radicals can produce aldehydes such as malondialdehyde (MDA), which can cause DNA damage. Increased levels of malondialdehyde (MDA) have been reported in tumor tissue from colorectal cancer patients compared with normal mucosa from the same subjects [217], which point toward an initiating versus a promoting mechanism. Furthermore, oxidized LDL has been shown to downregulate base excision repair (BER) activity in extracts of mouse monocytes [241], which could potentially amplify oxidative DNA damage since BER is the primary repair mechanism for oxidative DNA damage.
8.3 Biomarkers of Oxidative Nucleic Acid (DNA) Damage By inducing hydroxylation of the C-8 position of 2 -deoxyguanosine, ROS can create DNA damage (Fig. 5.1) in the form of the oxidative DNA base lesion, 8oxo-deoxyguanosine (8-oxo-dG). Although 20 oxidative base lesions have been identified, 8-oxo-dG is the most abundant DNA lesion caused by ROS and it is highly mutagenic [242]. 8-Oxo-dG is a reliable marker of oxidative DNA damage in tissue or urine [242]. RNOS (NO ˙and ONOO–) have also been observed to attack DNA bases and induce single-strand breaks [243]. Although levels of 8-oxo-dG have not been well studied in obesity, urinary 8oxo-dG excretion has been observed to be positively correlated with BMI in subjects with obstructive sleep apnea who were either overweight or obese [244]. Another study reported 8-oxo-dG levels were associated with body fat and higher in muscle obtained from self-reported weight gainers ( >5 kg) compared with weight maintainers (≤ 4 kg) in a small sample of hernia patients [245]. Cancer cells generally have higher levels of ROS and oxidative DNA lesions compared to normal cells [246]. In particular, human colon and breast cancer cells have increased steady-state levels of ROS (O2 –˙; H2 O2 ) relative to their respective normal cells [247]. Increased levels of 8-oxo-dG have also been observed in various tumors [248]. 8-Oxo-dG levels were significantly reduced in postmenopausal breast cancer survivors with the highest quartile of plasma carotenoids compared to those
5
Obesity and Cancer: Overview of Mechanisms
157
with the lowest quartile [249]. An association between urinary 8-oxo-dG levels and breast cancer risk was observed in another study but only after cases that underwent radiation treatment were removed from the analysis [250]. The increased ROS levels in cancer cells may result from endogenous and exogenous sources discussed above or from increased metabolic activity, mitochondrial malfunction, or infiltration of the tumor by inflammatory cells [246]. Regardless of how they are generated, ROS can attack and damage DNA and, therefore, may play a role in cancer cell initiation if the DNA base lesions induced are not repaired properly by DNA repair systems. Although base excision repair (BER) is the primary mechanism for 8-oxo-dG repair, nucleotide excision repair (NER), which is a more complex process that typically repairs larger lesions, may also be involved in repairing oxidative DNA lesions. Furthermore, since the mitochondrial electron transport chain is a major source of ROS generation, the vulnerability of the mitochondrial DNA (mtDNA) to ROS-induced damage may amplify oxidative stress in cancer cells; however, studies supporting the level of ROS-induced mtDNA damage in obesity are lacking. In addition to direct mutagenic effects, ROS may also play a role in tumor progression. For example, ROS are a necessary component of the signal transduction mechanisms by which many growth factors and cytokines activate the PI3K pathway and elicit their cellular responses [251]. Although a low to moderate level of ROS can activate the PI3K signaling pathway and promote cell survival (Fig. 5.2), severe or chronic oxidative stress levels may inhibit the PI3K pathway leading to apoptosis [251]. ROS also regulate autophagy (Fig. 5.2), a self-digestion process that degrades intracellular structures in response to stress, through several distinct mechanisms involving catalase and the mitochondrial electron transport chain [252]. Interestingly, ROS-regulated autophagy can be blocked by mTOR under nutrient-rich conditions [246]. Increased oxidative stress may also result in dysregulation of mRNA expression in mice and tissue culture. Treatment of 3T3-L1 adipocytes with H2 O2 was found to downregulate mRNA expression of adiponectin and PPARγ and upregulate mRNA expression of plasminogen activator inhibitor-1 (PAI-1), IL-6 and MCP-1, suggesting that ROS can affect multiple factors, albeit indirectly, involved in obesity and the Metabolic Syndrome (discussed in Section 9). Anti-oxidants may also be particularly important in obesity-related carcinogenesis because obesity may create ‘malnubesity,’ a condition whereby substantially lower levels of micronutrients are available relative to the level of macronutrients, leading to inefficiencies in oxidative energy metabolism [253]. Decreased concentrations of plasma anti-oxidants such as vitamin E, a known ROS scavenger, have been correlated with obesity [254]. Moreover, decreased enzymatic activity of anti-oxidants such as SOD and CAT have been reported in obese compared to non-obese individuals [255] and in some cancer cells compared to non-cancer cells [246]. Futhermore, anti-oxidant deficiencies may alter DNA repair mechanisms which are highly dependent on micronutrients. Interestingly, co-treatment of mouse monocytes with oxidized LDL and anti-oxidants (vitamins C and E) prevented the downregulation of BER [241].
158
N.L. Nock and N.A. Berger
8.4 Biomarkers of Systemic Oxidative Stress: F2-Isoprostanes Isoprostanes are prostaglandin-like bioactive compounds that are synthesized independent of cyclooxygenases (COX) through free-radical catalysis of arachidonic acid in a multi-step process involving the removal of a labile hydrogen, addition of oxygen, formation of peroxy radicals, and resulting in endoperoxide intermediates designated as either 5-, 8-, 12-, or 15-series isomers based on the carbon atom to which the side chain hydroxyl group is positioned [235]. Isoprostanes are cleaved from phospholipids by phospholipases and, then, circulate in plasma with subsequent excretion in urine. Due to its chemical stability, the urinary metabolite 8-iso-prostaglandin F2α (8-iso-PGF2α) has been identified as a reliable index of in vivo oxidative stress and ensuing lipid peroxidation [256]. Increased urinary levels of 8-iso-PGF2α have been observed in obese compared to non-obese women, particularly in those with android (defined as a BMI > 28 kg/m2 and a waist-to-hip ratio (WHR) ≥ 0.86) compared to gynoid (defined as a BMI > 28 kg/m2 and a WHR < 0.86) obesity [257]. Obese men were found to have significantly higher plasma concentrations of 8-iso-PGF2α than non-obese men [258]. Increased urinary isoprostanes have also been associated with increased visceral adipose tissue [259] and a 5 kg/m2 increase in BMI has been associated with a 9.9% increase in urinary isoprostane levels [260]. Studies examining 8-iso-PGF2α levels in cancer patients are limited. Increased isoprostanes have been associated with increased breast cancer using immunoassay [250] and mass spectrometry [261] techniques. This association was even greater among women with higher BMI values [261]. Interestingly, 8-iso-PGF2α levels were found to be significantly decreased in breast cancer patients who lost weight after participation in a 12-month low-fat (15% of energy from fat) dietary intervention program [262]. However, urinary 8-iso-PGF2α levels were not found to be significantly different in a small study of prostate cancer cases compared to controls [263]. Although mechanistic studies are lacking, 8-iso-PGF2α levels have been observed to be significantly higher in estradiol-induced mammary tumors than levels in control mammary tissue from age-matched rats leading to proliferative changes in the breast tissue after only 7 days, with the first palpable tumor appearing 128 days after estradiol exposure, suggesting that the oxidative changes occurred prior to tumor development [264].
9 Metabolic Syndrome The Metabolic Syndrome, also known as the insulin resistance syndrome and syndrome ‘X,’ is characterized by multiple core traits including insulin resistance, impaired glucose tolerance, obesity (central body), hypertension, and dyslipidemias (Fig. 5.1). Recent research suggests that the Metabolic Syndrome may also consist of additional factors including a prothrombotic state characterized by high PAI-1
5
Obesity and Cancer: Overview of Mechanisms
159
and a pro-inflammatory state characterized by elevated C-reactive protein [265, 266]. Disturbed sleep has also been identified as a component of the Metabolic Syndrome, prompting the term syndrome ‘Z’ [267]. The Metabolic Syndrome has been associated with diseases such as diabetes, cardiovascular disease, and non-alcoholic fatty liver disease as well as certain cancers [268, 269]. Patients with the Metabolic Syndrome have a higher risk of colon adenomas and carcinomas [270] and pancreatic cancer [271]. Furthermore, women with type 2 diabetes have a 16% higher risk of developing breast cancer compared to nondiabetics, and this risk is greatest in postmenopausal women [268]. Although there is limited direct evidence for the role of the Metabolic Syndrome in breast cancer, the relation is supported indirectly by the multiple and consistent associations observed between individual components of the syndrome (insulin resistance, dyslipidemia, abdominal obesity) and increased postmenopausal breast cancer risk [269, 271]. Studies in prostate cancer have produced equivocal findings, perhaps, due to the opposing associations observed between BMI and less advanced (decreased risk) compared to more advanced (increased risk) prostate cancer [269]. The role of the Metabolic Syndrome in cancer clearly depends on a multitude of genetic and environmental factors. When mice with the Apc Min/+ mutation are fed a high-fat diet promoting obesity and the Metabolic Syndrome, they show more rapid progression of intestinal polyps than mice without the Metabolic Syndrome (Doerner S, Nadeau JH, 2009, Diet dependent growth of intestinal polyps in Apc Min/+ mice, Unpublished Work). The same mice without the Apc Min/+ mutation, that are susceptible to diet-induced obesity and Metabolic Syndrome, go on to develop NAFLD and then hepatocellular carcinoma (HCC) at even later ages [272]. The importance of understanding the role of the Metabolic Syndrome in cancer stems from the fact that this syndrome involves several factors previously discussed that have been shown to independently affect tumor growth (see Sections 2, 3, 4, and 5). However, these factors need to be considered in aggregate, since interruption of any one component, such as blocking the cellular growth stimulated by insulin or IGF, is unlikely to completely block cancer promotion by other factors involved in the Metabolic Syndrome. On the other hand, it may be possible to simultaneously eliminate multiple individual risk factors for cancer by ablating the Metabolic Syndrome through weight loss induced by diet/caloric restriction and exercise and, perhaps, where necessary, bariatric surgery. Interestingly, recent studies have shown that bariatric surgery for obesity not only improves multiple metabolic abnormalities but also reduces the incidence of cancer and cancer-related deaths [273–275] see Section 11.
10 Dietary Intake of Carcinogens and DNA Damage Although the concept has not been well explored, it is plausible that excess dietary intake of carcinogens may occur commensurate with obesity and may be compounded by ‘malnubesity’ (decreased micronutrient relative to macronutrient level)
160
N.L. Nock and N.A. Berger
leading to increased risk of carcinogenesis. For example, increased dietary intake of pan-fried and grilled fish, chicken, and red meats can increase levels of exposure to carcinogenic xenobiotic agents such as heterocyclic amines (HCAs) and polycyclic aromatic hydrocarbons (PAHs). Furthermore, the lipophilic nature of these compounds may lead to bioaccumulation in body fat, which may be later released during periods of stress or weight loss, which may increase risk in ‘yo-yo’ dieters. In this section, we discuss the most mutagenic and commonly studied dietary carcinogens with a potential role in obesity-related carcinogenesis.
10.1 Heterocyclic Amines (HCAs) The most well-studied and abundant HCAs are 2-amino-3,8-dimethylimidazo[4,5f]quinoxaline (MeIQx) and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), with PhIP being the most abundant. Bioactivation of HCAs including PhIP to carcinogenic species is initiated by N-oxidation of the compound, which is catalyzed by cytochrome P4501A2 and subsequent acetylation or sulfation of the N-hydroxy–PhIP is catalyzed by N-acetyltransferases (NAT) or sulfotransferases (SULT), which generate N-acetoxy- or N-sulfonyloxy–PhIP electrophilic compounds that bind covalently to DNA to form PhIP adducts [276]. PhIP–DNA adducts and other bulky adducts formed from HCAs can lead to mutations if they are not repaired by NER processes. Investigations between HCA levels and body composition, particularly body fat, are limited. A single study in a Swedish cohort examined HCA and BMI levels and found that increased dietary intake of HCA levels was associated with increased BMI, where obese women and men had a significantly higher risk of being in the highest HCA quintile compared to average weight women and men [277]. Many studies have used meat intake as a surrogate measure of HCA intake level. A meta-analysis involving 15 cohort studies (7,367 cases) found that an increase of 120 g/day of red meat was associated with a 28% increased risk of colorectal cancer [278]. Interestingly, a recent study observed a 47% increased risk of colorectal adenomas when comparing the highest to the lowest quartile of dietary intake of PhIP, but no statistically significant association was observed for MeIQx [279]. A study conducted in the Agricultural Health Study cohort recently reported that intake of well or very well done meat was associated with a 1.26-fold increased risk of incident prostate cancer and a 1.97-fold increased risk of advanced disease when the highest tertile was compared with the lowest; however, they found no statistically significant association with estimated levels of individual HCAs including PhIP and MeIQx [280]. In addition, PhIP–DNA adducts in men undergoing radical prostatectomy for prostate cancer treatment who had increased grilled red meat consumption had higher PhIP-DNA adduct levels in prostate tumor but not adjacent non-tumor prostate cells [276]. In a recent review between dietary factors and breast cancer risk, one cohort study observed a positive association between HCA levels and breast cancer [281].
5
Obesity and Cancer: Overview of Mechanisms
161
PhIP has been shown to increase mutation frequency and tumor incidence in animal models [280]. PhIP-DNA adducts have been found in prostate and other organ cells [276]. Moreover, the mutations induced by these DNA adducts can occur in genes controlling key homeostatic mechanisms such as tumor suppression (e.g., p53) or tumor promotion (e.g., KRAS), potentially leading to accelerated tumor growth.
10.2 Polycyclic Aromatic Hydrocarbons (PAHs) Polycyclic aromatic hydrocarbons (PAHs) are large aromatic planar compounds that comprise a class of over 200 chemicals with three or more benzene rings. The most notable PAH is benzo(a)pyrene (b(a)p), which is classified as a known human carcinogen. PAH are metabolized by the cytochrome P450 enzyme system into reactive electrophiles, which may be detoxified and made water soluble through conjugation with glutathione or glucuronic acid; however, if these reactive electrophilic PAH species are not detoxified they may bind to DNA and form PAH–DNA adducts [282]. If these bulky adducts are not repaired by NER systems, they may cause mutations in the DNA. Human exposure to PAHs can come from many sources including cigarette smoking, fuel combustion/diesel exhaust, and our diet through the consumption primarily of cooked meats, but appreciable levels have also been found in grains and leafy vegetables [283]. Similar to HCAs, dietary intake of PAHs can be estimated using food questionnaires together with a database that has quantified levels of PAHs, most notable b(a)p, in various foods [284]. Studies examining PAH levels and body composition are lacking. However, one study found that BMI was inversely associated with the presence of detectable PAH (b(a)p)–DNA adducts [285]. As mentioned above, a study conducted in the Agricultural Health Study cohort found that intake of well or very well done meat was associated with a 1.26-fold increased risk of incident prostate cancer and a 1.97-fold increased risk of advanced disease when the highest tertile was compared with the lowest. However, they found no statistically significant association with estimated levels of individual PAHs including b(a)p [280]. PAH–DNA adducts have been associated with prostate cancer cells and this association differed by race as well as variation in genes involved in PAH metabolism, conjugation, and repair [286]. In a pooled analysis (873 cases; 941 controls), detectable compared to non-detectable PAH–DNA adducts in peripheral mononuclear cells were found to be associated with increased breast cancer risk but no dose–response with quantile levels was observed [287]. A recent study observed that higher intake of b(a)p was associated with increased risk of colorectal adenomas [288]. PAH can induce PAH–DNA adducts that can lead to mutations, primarily G to T transversions, if they are not properly repaired by NER systems [282]. B[a]p
162
N.L. Nock and N.A. Berger
has been shown to preferentially form adducts with DNA bases in the p53 tumor suppresser gene which are frequently mutated in cancer [282] and, therefore, may exacerbate cancer progression.
11 Therapeutic Opportunities: Diet, Exercise, and Pharmaceuticals Given that obesity is a causal risk factor for certain cancers, diet and exercise are potential modifiable factors that could prevent carcinogenesis. However, there is little in the way of randomized control trial evidence that weight loss can prevent cancer incidence or mortality other than that found in bariatric studies where weight reduction has been associated with decreased cancer mortality [270, 271].
11.1 Diet In animal models, restricting calories by 10–40% results in decreased cell proliferation and increased apoptosis [51]. As discussed within each of the previous sections, weight loss may modify many of the factors involved in obesity-related mechanisms. However, caloric restriction alone does not appear to be feasible for sustaining long-term weight loss in obese humans and strategies involving caloric restriction and physical activity (as well as possibly other psychosocial aspects) seem much more promising.
11.2 Exercise Exercise may alter several factors involved in the various mechanisms discussed in the preceding sections including insulin–IGF axis and AMPK pathways. Increased insulin action afforded by exercise is due to increased movement of glucose transporters (e.g., GLUT4 in skeletal muscle) to cell membranes. Other processes may involve the upregulation of enzymes responsible for the phosphorylation, storage, and oxidation of glucose as well as a greater capillary density and possibly conversion of fast-twitch glycolytic fibers (Type IIb) to fast-twitch oxidative fibers (Type IIa) [289], depending on the modality of training. Strength training in particular has been shown to reduce fasting insulin, fasting glucose, and IGF-I levels; however, no consistent change in IGFBP1 or IGFBP3 has been observed [290]. Exercise also activates AMPK, a fuel-sensing enzyme, in skeletal muscle and possibly adipose tissue and other organs in humans, which, in turn, stimulates energy-generating processes (e.g., glucose uptake and fatty acid oxidation) and decreases energy-consuming processes (e.g., protein and lipid synthesis) (Fig. 5.2) [291].
5
Obesity and Cancer: Overview of Mechanisms
163
The relationship between exercise and estrogen and androgen levels is not clear and may be affected by many factors including age, gender, body composition, exercise prescription, and pre-trial training status. However, most studies seem to show that exercise increases SHBG levels, which would decrease the availability of biologically active estrogens and androgens [66, 292]. In terms of cytokines and other factors secreted by adipocytes, aerobic exercise training studies have demonstrated reductions in hs-CRP concentrations ranging from 16% to 41%, an effect that may be independent of baseline levels of CRP, body composition, or weight loss [293]. Furthermore, IL-6, the first ‘myokine’ which is a cytokine produced and released by contracting skeletal muscle fibers, increases up to 100-fold in the circulation during physical exercise [294]. Regular exercise also induces anti-inflammatory effects by suppressing TNFα production, which together with changes in IL-6 may allow exercise to serve as a means to control the low-grade systemic inflammation [295, 296] associated with obesity. An acute bout of exercise has been shown to decrease plasma VEGF [297]; however, much additional research is needed to understand how obesity-related angiogenic factors are modified in humans as a result of chronic exercise. A transgenic mouse model containing the Apc Min/+ mutation and a lifelong enhanced exercise phenotype showed delayed development of tumors and prolonged survival compared to mice without the enhanced exercise phenotype. This may be attributed, in part, to lower levels of insulin, leptin, and IL-6 (Berger NA, Hanson R, Nadeau JH, 2009, Effect of PEPCK_Cmus transgenic mice on enhanced exercise and colon polyps in Apc Min/+ mice, Unpublished Work). Regular physical activity has been shown to decrease plasma levels of OxLDL [298]. Decreased oxidative damage to lipids as well as to DNA may occur as a result of the upregulation of GSH and SOD observed with chronic exercise [299]. Chronic exercise also increases the number and activity of natural killer (NK) and lymphokine-activated killer cells [300], which would enhance innate immunity and potentially reduce ensuing inflammation-mediated oxidative stress damage. Physical activity has been shown to decrease gut transit time, possibly through increased vagal tone and subsequent increased peristalsis [301, 302]. Decreased transit time would decrease contact time of colon mucosa with carcinogens as well as decrease potential exposure of carcinogens to other cells from reuptake through the portal vein. High levels of physical activity have been associated with decreased PGE2 levels and increased prostaglandin F2α levels, which may enhance gut motility [215, 303].
11.3 Pharmaceuticals Pharmacologic investigations targeting the obesity–cancer link focus on either promoting weight loss or disrupting the cell growth/tumor promotion process.
164
N.L. Nock and N.A. Berger
Investigational drugs targeting obesity by catabolic (anorexigenic) pathways include leptin, agonists of melanocortin receptor-4, 5-HT and dopamine, beta-3 adrenergic receptor agonists, adiponectin derivatives, and glucagon-like peptide-1 and those targeting anabolic (orexigenic) pathways consist of the ghrelin receptor, neuropeptide Y receptor and melanin-concentrating hormone-1 antagonists, peroxisome proliferator-activated receptor-gamma and -beta/delta agonists, and cannabinoid-1 receptor antagonists [304]. Sibutramine is a serotonin and noradrenaline reuptake inhibitor, which has been studied in a large number of randomized, double-blind, placebo-controlled trials lasting up to 24 months or more. Sibutramine has been found to induce weight loss in a dose–response manner (e.g., weight losses of 1.2, 6.1 and 8.8% at 10, 20, and 30 mg doses, respectively); however, this drug has been shown to cause many adverse effects including increasing diastolic blood pressure and, thus, is not utilized in patients with hypertension [305]. Orlistat, which is a lipase inhibitor that works in the gastrointestinal tract to reduce the body’s absorption of fat, has been well studied and shown to decrease weight as well as LDL cholesterol, systolic and diastolic blood pressure, and fasting insulin and glucose; however, it reduces the absorption of fat-soluble vitamins such as A, D, E, and K, and beta carotene [305]. In terms of pharmaceuticals that aim to interrupt the cell growth/tumorpromoting processes, PPAR agonists appear effective in treating insulin resistance and in treating various types of cancers by inhibiting proliferation through arrest at the G1 phase of the cell cycle; however, their usefulness is limited since they increase the risk of myocardial infarction and death from cardiovascular diseases [306]. Antibodies to the IGF-1 receptor have been used to interfere with the mitogenic effects of IGF-1 in tissue culture [307] and against a variety of tumors in clinical trials [308, 309]. Metformin, used as an ‘insulin sensitizer’ for treatment of type 2 diabetes mellitus, functions as an activator of AMP kinase, which inhibits activity of the mTOR pathway, thereby, reducing tumor growth rates [308, 310]. Metformin has also been shown to reduce the incidence of cancer in patients with diabetes and is currently in clinical trials for cancer prevention and treatment [308]. Furthermore, in women receiving neoadjuvant chemotherapy for early stage breast cancer, the concomitant administration of metformin for treatment of type II diabetes mellitus was associated with a three-fold increase in pathologic response rate; however, metformin administration was not associated with prolongation of relapse-free survival [311, 312]. Although the use of metformin by these women during neoadjuvant chemotherapy produced some degree of weight loss, the anti-cancer effects of metformin are likely not driven by the modest, short-term weight loss. Other agents in various stages of preclinical and clinical evaluation and use targeting the PI3K–mTOR pathway include LY294002 and everolimus [313]. Everolimus (RAD001) is currently in clinical use for treatment of renal cell carcinoma, which has been associated with obesity [314]. Newer agents with dual PI3K/mTOR inhibitor activity which are now in preclinical development may be more effective and include PI-103 and NVP-BEZ235. PI-103 has been shown to have anti-tumor effects against chordomas [315] and NVP-BEZ235 has been shown to have anti-tumor effects against trastuzumab-resistant breast cancer [316].
5
Obesity and Cancer: Overview of Mechanisms
165
Acknowledgments Support for this work was derived in part from NIH Grants K07CA129162, to Nora Nock and U54 CA116867 and P30 CA043703 to Nathan Berger.
References 1. Caterson ID, Gill TP (2002). Obesity: epidemiology and possible prevention. Best Prac Res Clin Endocrinol Metab, 16:595–610. 2. Ogden CL, Carroll MD, Curtin LR et al. (2008). Prevalence of overweight and obesity in the United States, 1999–2004. JAMA, 295:1549–55. 3. World Health Organization (2003). World Cancer Report. IARC Nonserial Publication, Geneva, Switzerland. 4. Ogden CL, Yanvoski SZ, Carroll MD et al. (2007). The epidemiology of obesity. Gastroenterology, 132:2087–102. 5. World Health Organization (2000). Obesity: Preventing and Managing the Global Epidemic. Report of a WHO Consultation. Rep. 894, World Health Organization, Geneva. 6. World Cancer Research Fund and American Institute for Cancer Research (2007). Food, Nutrition, Physical Activity, and the Prevention of Cancer: a Global Perspective. AICR, Washington, DC. 7. Renehan AG, Tyson M, Egger M et al. (2008). Body-mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studies. Lancet. 371: 569–78. 8. Calle EE, Rodriquez C, Walker-Thurmond K et al. (2003). Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med, 348:1625–38. 9. Li D, Morris JS, Liu J et al. (2009). Body mass index and risk, age of onset, and survival in patients with pancreatic cancer. JAMA, 301:2553–62. 10. Freedland SJ, Platz EA (2007). Obesity and prostate cancer: making sense out of apparently conflicting data. Epidemiol Rev, 29:88–97. 11. Chia VM, Newcomb PA, Trentham-Dietz A et al. (2007). Obesity, diabetes, and other factors in relation to survival after endometrial cancer diagnosis. Int J Gynecol Cancer, 17: 441–6. 12. Lane G (2008). Obesity and gynaecological cancer. Menopause Int, 14:33–7. 13. Reeves GK, Pirie K, Beral V et al. (2007). Cancer incidence and mortality in relation to body mass index in the Million Women Study: cohort study. BMJ, 335:1134. 14. Caan BJ, Kwan ML, Hartzell G et al. (2008). Pre-diagnosis body mass index, post-diagnosis weight change, and prognosis among women with early stage breast cancer. Cancer Causes Control, 19:1319–28. 15. Carmichael AR (2006). Obesity and prognosis of breast cancer. Obesity Rev, 7:333–40. 16. Dignam JJ, Polite BN, Yothers G et al. (2006). Body mass index and outcomes in patients who receive adjuvant chemotherapy for colon cancer. J Natl Cancer Inst, 98:1647–54. 17. Kristal AR, Gong Z (2007). Obesity and prostate cancer mortality. Future Oncol, 3:557–67. 18. Dirx MJ, Zeegers MP, Dagnelie PC et al. (2003). Energy restriction and the risk of spontaneous mammary tumors in mice: a meta-analysis Int J Cancer, 106:766–70. 19. Calle EE, Kaaks R (2004). Overweight, obesity and cancer: epidemiological evidence and proposed mechanisms. Nat Rev Cancer, 4:579–91. 20. Clark JM (2006). The epidemiology of nonalcoholic fatty liver disease in adults. J Clin Gastroenterol, 40:S5–S10. 21. Anand G, Katz PO (2008). Gastroesophageal reflux disease and obesity. Rev Gastroenterol Disord, 8:233–9. 22. Nordenstedt H, Lagergren J (2008). Environmental factors in the etiology of gastroesophageal reflux disease. Expert Rev Gastroenterol Hepatol, 2:93–103.
166
N.L. Nock and N.A. Berger
23. Rabinowitz D, Zierler KL (1962). Forearm metabolism in obesity and its response to intra-arterial insulin. Characterization of insulin resistance and evidence for adaptive hyperinsulinism. J Clin Invest, 41:2173–81. 24. Bonser AM, Garcia-Webb P (1984). C-peptide measurement: methods and clinical utility. Crit Rev Clin Lab Sci, 19:297–352. 25. Allen NE, Appleby PN, Kaaks R et al. (2003). Lifestyle determinants of serum insulin-like growth-factor-I (IGF-I), C-peptide and hormone binding protein levels in British women. Cancer Causes Control, 14:65–74. 26. Lukanova A, Zeleniuch-Jacquotte A, Lundin E et al. (2004). Prediagnostic levels of C-peptide, IGF-I, IGFBP -1, -2 and -3 and risk of endometrial cancer. Int J Cancer, 108:262–8. 27. Schairer C, Hill D, Sturgeon SR et al. (2004). Serum concentrations of IGF-I, IGFBP-3 and c-peptide and risk of hyperplasia and cancer of the breast in postmenopausal women. Int J Cancer, 108:773–9. 28. Ma J, Giovannucci E, Pollak M et al. (2004). A prospective study of plasma C-peptide and colorectal cancer risk in men. J Natl Cancer Inst, 96:546–53. 29. Michaud DS, Wolpin B, Giovannucci E et al. (2007). Prediagnostic plasma C-peptide and pancreatic cancer risk in men and women. Cancer Epidemiol Biomarkers Prev, 16:2101–9. 30. Pisani P (2008). Hyper-insulinaemia and cancer, meta-analyses of epidemiological studies. Arch Physiol Biochem, 114:63–70. 31. Carling D (2004). The AMP-activated protein kinase cascade–a unifying system for energy control. Trends Biochem Sci, 29:18–24. 32. Renehan AG, Frystyk J, Flyvbjerg A (2006). Obesity and cancer risk: the role of the insulinIGF axis. Trends Endocrinol Metab, 17:328–36. 33. Björnholm M, Al-Khalili L, Dicker A et al. (2002). Insulin signal transduction and glucose transport in human adipocytes: effects of obesity and low calorie diet. Diabetologia, 45:1128–35. 34. Vanhaesebroeck B, Alessi DR (2000). The PI3K-PDK1 connection: more than just a road to PKB. Biochem J, 346:561–76. 35. Rosner M, Hanneder M, Siegel N et al. (2008). The mTOR pathway and its role in human genetic diseases. Mutat Res, 659:284–92. 36. Wang W, Guan KL (2009). AMP-activated protein kinase and cancer. Acta Physiol (Oxf), 196:55–63. 37. Singh A, Hamilton-Fairley D, Koistinen R et al. (1990). Effect of insulin-like growth factortype I (IGF-I) and insulin on the secretion of sex hormone binding globulin and IGF-I binding protein (IBP-I) by human hepatoma cells. J Endocrinol, 124:3. 38. Marelli MM, Moretti RM, Procacci P et al. (2006). Insulin-like growth factor-I promotes migration in human androgen-independent prostate cancer cells via the alphavbeta3 integrin and PI3-K/Akt signaling. Int J Oncol, 28:723–30. 39. Clemmons DR (2006). Involvement of insulin-like growth factor-I in the control of glucose homeostasis. Curr Opin Pharmacol, 6:620–5. 40. Le Roith D, Scavo L, Butler A (2001). What is the role of circulating IGF-I? Trends Endocrinol Metab, 12:48–52. 41. Le Roith D, Bondy C, Yakar S et al. (2001). The somatomedin hypothesis: 2001. Endocr Rev, 22:53–74. 42. Allen NE, Roddam AW, Allen DS, Fentiman IS et al. (2005). A prospective study of serum insulin-like growth factor-I (IGF-I), IGF-II, IGF-binding protein-3 and breast cancer risk Br J Cancer. 92:1283–7. 43. Kim HS, Ingermann AR, Tsubaki J et al. (2004). Insulin-like growth factor binding protein 3 induces caspase-dependent apoptosis through a death receptor-mediated pathway in MCF-7 human breast cancer cells. Cancer Res. 64:2229–37. 44. Sandhu MS, Gibson JM, Heald AH et al. (2004). Association between insulin-like growth factor-I: insulin-like growth factor-binding protein-1 ratio and metabolic and anthropometric factors in men and women. Cancer Epidemiol Biomarkers Prev, 13:166–70.
5
Obesity and Cancer: Overview of Mechanisms
167
45. Frystyk J, Skjaerbaek C, Vestbo E et al. (1999). Circulating levels of free insulin-like growth factors in obese subjects: the impact of type 2 diabetes. Diabetes Metab Res Rev, 15:314–22. 46. Holmes MD, Pollak MN, Hankinson SE (2002). Lifestyle correlates of plasma insulin-like growth factor I and insulin-like growth factor binding protein 3 concentrations. Cancer Epidemiol Biomarkers Prev, 11:862–7. 47. Renehan AG, Zwahlen M, Minder C et al. (2004). Insulin-like growth factor (IGF)-I, IGF binding protein-3, and cancer risk: systematic review and meta-regression analysis. Lancet, 363:1346–53. 48. Giovannucci E (2001). Insulin, insulin-like growth factors and colon cancer: a review of the evidence. J Nutr, 131:3109S–20S. 49. Renehan AG, Harvie M, Howell A (2006). Insulin-like growth factor (IGF)-I, IGF binding protein-3, and breast cancer risk: eight years on. Endocr Relat Cancer, 13:273–8. 50. Nakae J, Kido Y, Accili D (2001). Distinct and overlapping functions of insulin and IGF-I receptors. Endocr Rev, 22:818–35. 51. Hursting SD, Lavigne JA, Berrigan D et al. (2003). Calorie restriction, aging, and cancer prevention: mechanisms of action and applicability to humans. Annu Rev Med, 54:131–52. 52. LeRoith D, Roberts CT Jr. (2003). The insulin-like growth factor system and cancer. Cancer Lett, 195:127–37. 53. Buckbinder L, Talbott R, Velasco-Miguel S et al. (1995). Induction of the growth inhibitor IGF-binding protein 3 by p53. Nature, 377:646–9. 54. Takahashi K, Suzuki K (1993). Association of insulin-like growth-factor-I-induced DNA synthesis with phosphorylation and nuclear exclusion of p53 in human breast cancer MCF-7 cells. Int J Cancer, 55:453–8. 55. Karver KC, Schuler LA (2008). Prolactin does not require insulin-like growth factor intermediates but synergizes with insulin-like growth factor I in human breast cancer cells. Mol Cancer Res, 6:634–43. 56. Saxena NK, Taliaferro-Smith L, Knight BB et al. (2008). Bidirectional crosstalk between leptin and insulin-like growth factor-I signaling promotes invasion and migration of breast cancer cells via transactivation of epidermal growth factor receptor. Cancer Res, 68:9712–22. 57. Yee D, Lee AV (2000). Crosstalk between the insulin-like growth factors and estrogens in breast cancer. Mammary Gland Biol Neoplasia, 5:107–15. 58. Lee AV, Jackson JG, Gooch JL et al. (1999). Enhancement of insulin-like growth factor signaling in human breast cancer: estrogen regulation of insulin receptor substrate-1 expression in vitro and in vivo. Mol Endocrinol, 13:787–96. 59. Sinha-Hikim I, Taylor WE, Gonzalez-Cadavid NF et al. (2004). Androgen Receptor in Human Skeletal Muscle and Cultured Muscle Satellite Cells: Up-Regulation by Androgen Treatment. J Clin Endocrinol Metab, 89:5245–55. 60. Ray R, Novotny NM, Crisostomo PR et al. (2008). Sex steroids and stem cell function. Mol Med, 14:493–501. 61. Key TJ, Allen NE, Verkasalo PK et al. (2001). Energy balance and cancer: the role of sex hormones. Proc Nutr Soc, 60:81–9. 62. Key TJ, Appleby PN, Reeves GK et al. (2003). Body mass index, serum sex hormones, and breast cancer risk in postmenopausal women. J Natl Cancer Inst, 95:1218–26. 63. Derby CA, Zilber S, Brambilla D et al. (2006). Body mass index, waist circumference and waist to hip ratio and change in sex steroid hormones: the Massachusetts Male Ageing Study. Clin Endocrinol (Oxf), 65:125–31. 64. Pan RS, Pu SJ (2002). Impact of obesity on hypogonadism in the andropause. Int J Androl, 25:195–201. 65. Franks S, Kiddy DS, Hamilton-Fairley D et al. (1991). The role of nutrition and insulin in the regulation of sex hormone binding globulin. J Steroid Biochem Mol Biol, 39:835–8. 66. McTiernan A, Rajan KB, Tworoger SS et al. (2003). Adiposity and sex hormones in postmenopausal breast cancer survivors. J Clin Oncol, 21:1961–6.
168
N.L. Nock and N.A. Berger
67. Jen KL, Djuric Z, DiLaura NM et al. (2004). Improvement of metabolism among obese breast cancer survivors in differing weight loss regimens. Obes Res, 12:306–12. 68. Kaaks R, Berrino F, Key T et al. (2005). Serum sex steroids in premenopausal women and breast cancer risk within the European Prospective Investigation into Cancer and Nutrition (EPIC). J Natl Cancer Inst, 97:755–65. 69. Kaaks R, Lukanova A, Kurzer MS (2002). Obesity, endogenous hormones, and endometrial cancer risk: a synthetic review. Cancer Epidemiol Biomarkers Prev, 11:1531–43. 70. Endogenous Hormones and Prostate Cancer Collaborative Group, Roddam AW, Allen NE et al. (2008). Endogenous sex hormones and prostate cancer: a collaborative analysis of 18 prospective studies. J Natl Cancer Inst, 100:170–83. 71. Hsing AW, Chu LW, Stanczyk FZ (2008). Androgen and prostate cancer: is the hypothesis dead? Cancer Epidemiol Biomarkers Prev, 17:2525–30. 72. Hsing AW, Sakoda LC, Chua SC Jr. (2007). Obesity, metabolic syndrome, and prostate cancer. Am J Clin Nutr, 86:S857. 73. O’Connor KA, Ferrell RJ, Brindle E et al. (2009). Total and unopposed estrogen exposure across stages of the transition to menopause. Cancer Epidemiol Biomarkers Prev, 18:828–36. 74. Liao DJ, Dickson RB (2002). Roles of androgens in the development, growth, and carcinogenesis of the mammary gland. J Steroid Biochem Mol Biol, 80:175–89. 75. Risch HA (1998). Hormonal etiology of epithelial ovarian cancer, with a hypothesis concerning the role of androgens and progesterone. J Natl Cancer Inst, 90:1774–86. 76. Olsen CM, Green AC, Nagle CM et al. (2008). Epithelial ovarian cancer: testing the ‘androgens hypothesis’. Endocr Relat Cancer, 15:1061–8. 77. Debes JD, Tindall DJ (2002). The role of androgens and the androgen receptor in prostate cancer. Cancer Lett, 187:1–7. 78. Mellado B, Codony J, Ribal MJ et al. (2009). Molecular biology of androgen-independent prostate cancer: the role of the androgen receptor pathway. Clin Transl Oncol, 11:5–10. 79. Zhang J, Sun Y, Liu Y et al. (2004). Synergistic effects of androgen and estrogen on the mouse uterus and mammary gland. Oncol Rep, 12:709–16. 80. Ho SM, Leav I, Merk FB et al. (1995). Induction of atypical hyperplasia, apoptosis, and type II estrogen-binding sites in the ventral prostates of Noble rats treated with testosterone and pharmacologic doses of estradiol-17 beta. Lab Invest, 73:356–65. 81. Mawson A, Lai A, Carroll JS et al. (2005). Estrogen and insulin/IGF-1 cooperatively stimulate cell cycle progression in MCF-7 breast cancer cells through differential regulation of c-My and cyclin D1. Mol Cell Endocrinol, 229:161–73. 82. MacDougald OA, Burant CF (2007). The rapidly expanding family of adipokines. Cell Metab, 6:159–61. 83. Considine RV, Sinha MK, Heiman ML et al. (1996). Serum immunoreactive-leptin concentrations in normal-weight and obese humans. N Engl J Med, 334:292–5. 84. Guerra B, Santana A, Fuentes T et al. (2007). Leptin receptors in human skeletal muscle. J Appl Physiol, 102:1786–92. 85. Trayhurn P, Bing C (2006). Appetite and energy balance signals from adipocytes. Philos Trans R Soc Lond B Biol Sci, 361:1237–49. 86. Blüher S, Mantzoros CS (2007). Leptin in reproduction. Curr Opin Endocrinol Diabetes Obes, 14:458–64. 87. Maya-Monteiro CM, Bozza PT (2008). Leptin and mTOR: partners in metabolism and inflammation. Cell Cycle, 7:1713–7. 88. Baratta M (2002). Leptin – from a signal of adiposity to a hormonal mediator in peripheral tissues. Med Sci Monit, 8:RA282–RA292. 89. Zhang Y, Proenca R, Maffei M et al. (1994). Positional cloning of the mouse obese gene and its human homologue. Nature, 372:425–32. 90. Ingalls AM, Dickie MM, Snell GD (1950). Obese, a new mutation in the house mouse. J Hered, 41:317–8.
5
Obesity and Cancer: Overview of Mechanisms
169
91. Considine RV, Sinha MK, Heiman ML et al. (1996). Serum immunoreactive-leptin concentrations in normal-weight and obese humans. N Engl J Med, 334:292–5. 92. Schäffler A, Schölmerich J, Buechler C (2007). Mechanisms of disease: adipokines and breast cancer – endocrine and paracrine mechanisms that connect adiposity and breast cancer. Nat Clin Pract Endocrinol Metab, 3:345–54. 93. Magni P, Liuzzi A, Ruscica M et al. (2005). Free and bound plasma leptin in normal weight and obese men and women: relationship with body composition, resting energy expenditure, insulin-sensitivity, lipid profile and macronutrient preference. Clin Endocrinol (Oxf), 62:189–96. 94. Smith SR, Lovejoy JC, Greenway F et al. (2001). Contributions of total body fat, abdominal subcutaneous adipose tissue compartments, and visceral adipose tissue to the metabolic complications of obesity. Metabolism, 50:425–35. 95. Vona-Davis L, Rose DP (2007). Adipokines as endocrine, paracrine, and autocrine factors in breast cancer risk and progression. Endocr Relat Cancer, 14:189–206. 96. Cymbaluk A, Chudecka-Glaz A, Rzepka-Górska I (2008). Leptin levels in serum depending on Body Mass Index in patients with endometrial hyperplasia and cancer. Eur J Obstet Gynecol Reprod Biol, 136:74–7. 97. Mistry T, Digby JE, Desai KM et al. (2007). Obesity and prostate cancer: a role for adipokines. Eur Urol, 52:46–53. 98. Stattin P, Kaaks R, Johansson R et al. (2003). Plasma leptin is not associated with prostate cancer risk. Cancer Epidemiol Biomarkers Prev, 12:474–5. 99. Stattin P, Palmqvist R, Söderberg S et al. (2003). Plasma leptin and colorectal cancer risk: a prospective study in Northern Sweden. Oncol Rep, 10:2015–21. 100. Stattin P, Lukanova A, Biessy C et al. (2004). Obesity and colon cancer: does leptin provide a link? Int J Cancer, 109:149–52. 101. Chia VM, Newcomb PA, Lampe JW et al. (2007). Leptin concentrations, leptin receptor polymorphisms, and colorectal adenoma risk. Cancer Epidemiol Biomarkers Prev, 16:2697–703. 102. Kumor A, Daniel P, Pietruczuk M et al. (2009). Serum leptin, adiponectin, and resistin concentration in colorectal adenoma and carcinoma (CC) patients. Int J Colorectal Dis, 24:275–81. 103. Hoda MR, Keely SJ, Bertelsen LS et al. (2007). Leptin acts as a mitogenic and antiapoptotic factor for colonic cancer cells. Br J Surg, 94:346–54. 104. Birmingham JM, Busik JV, Hansen-Smith FM et al. (2009). Novel mechanism for obesityinduced colon cancer progression. Carcinogenesis, 30:690–7. 105. Kendall BJ, Macdonald GA, Hayward NK et al. (2008). Leptin and the risk of Barrett’s oesophagus. Gut, 57:448–54. 106. Orel M, Lichnovská R, Gwozdziewiczová S et al. (2004). Gender differences in tumor necrosis factor alpha and leptin secretion from subcutaneous and visceral fat tissue. Physiol Res, 53:501–5. 107. Catalona S, Marsico S, Giordano C et al. (2003). Leptin enhances via AP-1 expression of aromatase in MCF-7 cell line. J Biol Chem, 278:28668–76. 108. Machinal-Quélin F, Dieudonné MN, Leneveu MC et al. (2002). Direct in vitro effects of androgens and estrogens on ob gene expression and leptin secretion in human adipose tissue. Endocrine, 18:179–84. 109. Hu X, Juneja SC, Maihle NJ et al. (2002). Leptin-a growth factor in normal and malignant breast cancer cells and for normal mammary gland development. J Natl Cancer Inst, 94:1704–11. 110. Hardwick JC, Van Den Brink GR, Offerhaus GJ et al. (2001). Leptin is a growth factor for colonic epithelial cells. Gastroenterology, 121:79–90. 111. Sharma D, Saxena NK, Vertino PM et al. (2006). Leptin promotes the proliferative response and invasiveness in human endometrial cancer cells by activating multiple signaltransduction pathways. Endocr Relat Cancer, 13:629–40.
170
N.L. Nock and N.A. Berger
112. Hursting SD, Nunez NP, Varticovski L et al. (2007). The obesity-cancer link: lessons learned from a fatless mouse. Cancer Res, 67:2391–3. 113. Kantartzis K, Staiger H, Machann J et al. (2009). Adiponectin oligomers and ectopic fat in liver and skeletal muscle in humans. Obesity (Silver Spring), 17:390–2. 114. Haluzík M, Parízková J, Haluzík MM (2004). Adiponectin and its role in the obesity-induced insulin resistance and related complications. Physiol Res, 53:123–9. 115. Yamauchi T, Karmon J, Ito Y et al. (2003). Cloning of adiponectin receptors that mediate antidiabetic metabolic effects. Nature, 423:762–9. 116. Gil-Campos M, Cañete RR, Gil A (2004). Adiponectin, the missing link in insulin resistance and obesity. Clin Nutr, 23:963–74. 117. Yamauchi T, Kamon J, Minokoshi Y et al. (2002). Adiponectin stimulates glucose utilization and fatty-acid oxidation by activating AMP-protein kinase. Nat Med, 8:1288–95. 118. Steffes MW, Gross MD, Schreiner PJ et al. (2004). Serum adiponectin in young adults – interactions with central adiposity, circulating levels of glucose and insulin resistance: the CARDIA study. Ann Epidemiol, 14:492–8. 119. Coppola A, Marfella R, Coppola L et al. (2009). Effect of weight loss on coronary circulation and adiponectin levels in obese women. Int J Cardiol, 134(3):414–6. 120. Wedick NM, Snijder MB, Dekker JM et al. (2009). Prospective investigation of metabolic characteristics in relation to weight gain in older adults: the Hoorn study. Obesity (Silver Spring), 17(8):1609–14. 121. Cust AE, Kaaks R, Friedenreich C et al. (2007). Plasma adiponectin levels and endometrial cancer risk in pre- and postmenopausal women. J Clin Endocrinol Metab, 92:255–63. 122. Rzepka-Górska I, Bedner R, Cymbaluk-Ploska A et al. (2008). Serum adiponectin in relation to endometrial cancer and endometrial hyperplasia with atypia in obese women. Eur J Gynaecol Oncol, 29:594–7. 123. Tian YF, Chu CH, Wu MH et al. (2007). Anthropometric measures, plasma adiponectin, and breast cancer risk. Endocr Relat Cancer, 14:669–77. 124. Mantzoros C, Petridou E, Dessypris N et al. (2004). Adiponectin and breast cancer risk. Clin Endocrinol Metab, 89:1102–7. 125. Wei EK, Giovannucci E, Fuchs CS et al. (2005). Low plasma adiponectin levels and risk of colorectal cancer in men: a prospective study. J Natl Cancer Inst, 97:1688–94. 126. Ferroni P, Palmirotta R, Spila A et al. (2007). Prognostic significance of adiponectin levels in non-metastatic colorectal cancer. Anticancer Res, 27:483–9. 127. Nishihara T, Baba M, Matsuda M et al. (2008). Adiponectin deficiency enhances colorectal carcinogenesis and liver tumor formation induced by azoxymethane in mice. World J Gastroenterol, 14:6473–80. 128. Sher DJ, Oh WK, Jacobus S et al. (2008). Relationship between serum adiponectin and prostate cancer grade. Prostate, 68:1592–8. 129. Goktas S, Yilmaz MI, Caglar K et al. (2005). Prostate cancer and adiponectin. Urology. 65:1168–72. 130. Bråkenhielm E, Veitonmäki N, Cao R et al. (2004). Adiponectin-induced antiangiogenesis and antitumor activity involve caspase-mediated endothelial cell apoptosis. Proc Natl Acad Sci U S A, 101:2476–81. 131. Kappes A, Loffler G (2000). Influences of ionomycin, dibutyryl-cycloAMP and tumor necrosis factor-alpha on intracellular amount and secretion of apM1 in differentiating primary human adipocytes. Hormone Metabolic Res, 32:548–54. 132. Fasshauer M, Kralisch S, Klier M et al. (2003). Adiponectin gene expression and secretion is inhibited by interleukin-6 in 3T3-L1 adipocytes. Biochem Biophys Res Comm, 301:1045–50. 133. Steppan CM, Brown EJ, Wright CM et al. (2001). A family of tissue-specific resistin-like molecules. Proc Natl Acad Sci U S A, 98:502–6. 134. Haluzik M, Haluzikova D. 2006. The role of resistin in obesity-induced insulin resistance. Curr Opin Investig Drugs, 7:306–11.
5
Obesity and Cancer: Overview of Mechanisms
171
135. Silswal N, Singh AK, Aruna B et al. (2005). Human resistin stimulates the pro-inflammatory cytokines TNF-alpha and IL-12 in macrophages by NF-kappaB-dependent pathway. Biochem Biophys Res Commun, 334:1092–101. 136. Degawa-Yamauchi MBJE, Juliar BE, Watson W et al. (2003). Serum resistin (FIZZ3) protein is increased in obese humans. J Clin Endocrinol Metab, 88:5452–5. 137. Silha JV, Krsek M, Skrha JV et al. (2003). Plasma resistin, adiponectin and leptin levels in lean and obese subjects: correlations with insulin resistance. Eur J Endocrinol, 149:331–5. 138. Chen BH, Song Y, Ding EL et al. (2009). Circulating levels of resistin and risk of type 2 diabetes in men and women: results from two prospective cohorts. Diabetes Care, 32: 329–34. 139. Kang JH, Yu BY, Youn DS (2007). Relationship of serum adiponectin and resistin levels with breast cancer risk. J Korean Med Sci, 22:117–21. 140. Hou WK, Xu YX, Yu T et al. (2007). Adipocytokines and breast cancer risk. Chin Med J (Engl), 120:1592–6. 141. Housa D, Vernerova Z, Heracek J et al. (2008). Serum resistin levels in benign prostate hyperplasia and non-metastatic prostate cancer: possible role in cancer progression. Neoplasma, 55:442–6. 142. Mu H, Ohashi R, Yan S et al. (2006). Adipokine resistin promotes in vitro angiogenesis of human endothelial cells. Cardiovasc Res, 70:146–57. 143. Fukuhara AM, Matsuda M, Nishizawa K et al. (2005). Visfatin: a protein secreted by visceral fat that mimics the effects of insulin. Science, 307:426–30. 144. Garten A, Petzold S, Korner A et al. (2009). Nampt: linking NAD biology, metabolism and cancer. Trends Endocrinol. Metab, 20:130–8. 145. Kowalska I, Straczkowski M, Nikolajuk A et al. (2007). Serum visfatin in relation to insulin resistance and markers of hyperandrogenism in lean and obese women with polycystic ovary syndrome. Hum. Reprod, 22:1824–9. 146. Tilg H, Moschen AR (2008). Role of adiponectin and PBEF/visfatin as regulators of inflammation: involvement in obesity-associated diseases. Clin Sci (Lond), 114:275–88. 147. Hufton SE, Moerkerk PT, Brandwijk R et al. (1999). A profile of differentially expressed genes in primary colorectal cancer using suppression subtractive hybridization FEBS Lett, 463:77–82. 148. Van B, Jr., Moerkerk PT, Gerbers AJ et al. (2002). Target validation for genomics using peptide-specific phage antibodies: a study of five gene products overexpressed in colorectal cancer. Int, J, Cancer, 101:118–27. 149. Ramsey KM, Yoshino J, Brace CS et al. (2009). Circadian clock feedback cycle through NAMPT-mediated NAD+ biosynthesis, Science, 324:651–4. 150. Nakahata Y, Kaluzova M, Grimaldi B et al. (2008). The NAD+-dependent deacetylase SIRT1 modulates CLOCK-mediated chromatin remodeling and circadian control, Cell, 134:329–40. 151. Katergari SA, Milousis A, Pagonopoulou O et al. (2008). Ghrelin in pathological conditions. Endocr J, 55:439–53. 152. Soares JB, Leite-Moreira AF (2008). Ghrelin, des-acyl ghrelin and obestatin: three pieces of the same puzzle. Peptides, 29:1255–70. 153. Beasley JM, Ange BA, Anderson CA et al. (2009). Characteristics associated with fasting appetite hormones (obestatin, ghrelin, and leptin). Obesity (Silver Spring), 17: 349–54. 154. Shak JR, Roper J, Perez-Perez GI et al. (2009). The effect of laparoscopic gastric banding surgery on plasma levels of appetite-control, insulinotropic, and digestive hormones. Obes Surg, 18:1089–96. 155. Roth CL, Reinehr T, Schernthaner GH et al. (2009). Ghrelin and obestatin levels in severely obese women before and after weight loss after Roux-en-Y gastric bypass surgery. Obes Surg, 19:29–35.
172
N.L. Nock and N.A. Berger
156. Mungan NA, Eminferzane S, Mungan AG et al. (2008). Diagnostic value of serum ghrelin levels in prostate cancer. Urol Int, 80:245–8. 157. Wolf I, Sadetzki S, Kanety H et al. (2006). Adiponectin, ghrelin, and leptin in cancer cachexia in breast and colon cancer patients. Cancer, 106:966–73. 158. Lanfranco F, Baldi M, Cassoni P et al. (2008). Ghrelin and prostate cancer. Vitam Horm, 77:301–24. 159. Hotamisligil GS, Shargill NS, Spiegelman BM (1993). Adipose expression of tumor necrosis factor-alpha: direct role in obesity-linked insulin resistance. Science, 259:87–91. 160. Rhodes JM, Campbell BJ (2002). Inflammation and colorectal cancer: IBD-associated and sporadic cancer compared. Trends Mol Med, 8:10–6. 161. Nelson WG, De Marzo AM, DeWeese TL et al. (2004). The role of inflammation in the pathogenesis of prostate cancer. J Urol, 172:S6–S11. 162. McCann J (1999). Esophageal cancers: changing character, increasing incidence. J Natl Cancer Inst, 91:497–8. 163. Heilbronn LK, Campbell LV (2008). Adipose tissue macrophages, low grade inflammation and insulin resistance in human obesity. Curr Pharm Des, 14:1225–30. 164. Kanda H, Tateya S, Tamori Y et al. (2006). MCP-1 contributes to macrophage infiltration into adipose tissue, insulin resistance, and hepatic steatosis in obesity. J Clin Invest, 116:1494–505. 165. Trayhurn P, Wood IS (2004). Adipokines: inflammation and the pleiotropic role of white adipose tissue. Br J Nutr, 92:347–55. 166. Uysal KT, Wiesbrock SM, Marino MW et al. (1997). Protection from obesity-induced insulin resistance in mice lacking TNF-alpha function. Nature, 389:610–4. 167. Coppak SW (2001). Pro-inflammatory cytokines and adipose tissue. Proc Nutr Soc, 60: 349–56. 168. Bahceci M, Gokalp D, Bahceci S et al. (2007). The correlation between adiposity and adiponectin, tumor necrosis factor alpha, interleukin-6 and high sensitivity C-reactive protein levels. Is adipocyte size associated with inflammation in adults? J Endocrinol Invest, 30:210–4. 169. Hotamisligil GS, Arner P, Caro JF et al. (1995). Increased adipose expression of tumor necrosis factor- in human obesity and insulin resistance. J Clin Invest, 95:2409–15. 170. Kern PA, Saghizadeh M, Ong JM et al. (1995). The expression of tumor necrosis factor in human adipose tissue. Regulation by obesity, weight loss, and relationship to lipoprotein lipase. J Clin Invest, 95:2111–9. 171. Il’yasova D, Colbert LH, Harris TB et al. (2005). Circulating levels of inflammatory markers and cancer risk in the health aging and body composition cohort. Cancer Epidemiol Biomarkers Prev, 14:2413–8. 172. Kim S, Keku TO, Martin C et al. (2008). Circulating levels of inflammatory cytokines and risk of colorectal adenomas. Cancer Res, 68:323–8. 173. Michalaki V, Syrigos K, Charles P et al. (2004). Serum levels of IL-6 and TNF-alpha correlate with clinicopathological features and patient survival in patients with prostate cancer. Br J Cancer, 90:2312–6. 174. Gonullu G, Ersoy C, Evrensel T et al. (2005). Relation between insulin resistance and serum concentrations of IL-6 and TNF-alpha in overweight or obese women with early stage breast cancer. Cytokine, 31:264–9. 175. Carswell EA, Old LJ, Kassel RL et al. (1975). An endotoxin-induced serum factor that causes necrosis of tumors. Proc Natl Acad Sci U S A, 72:3666–70. 176. Nieto-Vazquez I, Fernández-Veledo S, Krämer DK et al. (2008). Insulin resistance associated to obesity: the link TNF-alpha. Arch Physiol Biochem, 114:183–94. 177. Osawa Y, Nagaki M, Banno Y et al. 2002. Tumor necrosis factor alpha-induced interleukin-8 production via NF-kappaB and phosphatidylinositol 3-kinase/Akt pathways inhibits cell apoptosis in human hepatocytes. Infect Immun, 70:6294–301. 178. Liu J, Lin A (2005). Role of JNK activation in apoptosis: a double-edged sword. Cell Res, 15:36–42.
5
Obesity and Cancer: Overview of Mechanisms
173
179. do Nascimento CO, Hunter L, Trayhurn P (2004). Regulation of haptoglobin gene expression in 3T3-L1 adipocytes by cytokines, catecholamines, and PPARgamma. Biochem Biophys Res Commun, 313:702–8. 180. Purohit A, Newman SP, Reed MJ (2002). The role of cytokines in regulating estrogen synthesis: implications for the etiology of breast cancer. Breast Cancer Res, 4:65–9. 181. Jansson JO, Wallenius K, Wernstedt I et al. (2003). On the site and mechanism of action of the anti-obesity effects of interleukin-6. Growth Horm IGF Res, 13:S28–S32. 182. Wallenius K, Wallenius V, Sunter D et al. (2002). Intracerebroventricular interleukin-6 treatment decreases body fat in rats. Biochem Biophys Res Commun, 293:560–5. 183. Vozarova B, Weyer C, Hanson K et al. (2001). Circulating interleukin-6 in relation to adiposity, insulin action, and insulin secretion. Obes Res, 9:414–7. 184. Salgado R, Junius S, Benoy I et al. (2003). Circulating interleukin-6 predicts survival in patients with metastatic breast cancer. Int J Cancer, 103:642–6. 185. Bachelot T, Ray-Coquard I, Menetrier-Caux C et al. (2003). Prognostic value of serum levels of interleukin 6 and of serum and plasma levels of vascular endothelial growth factor in hormone-refractory metastatic breast cancer patients. Br J Cancer, 88:1721–6. 186. Lotem J, Sachs L (2001). Different mechanisms for suppression of apoptosis by cytokines and calcium mobilizing compounds. Proc Natl Acad Sci U S A, 95:4601–6. 187. Fenton JI, Hursting SD, Perkins SN et al. (2006). Interleukin-6 production induced by leptin treatment promotes cell proliferation in an Apc (Min/+) colon epithelial cell line. Carcinogenesis, 27:1507–15. 188. Grano M, Mori G, Minielli V et al. (2000). Breast cancer cell line MDA-231 stimulates osteoclastogenesis and bone resorption in human osteoclasts. Biochem Biophys Res Commun, 270:1097–100. 189. Ndumele CE, Pradhan AD, Ridker PM (2006). Interrelationships between inflammation, C-reactive protein, and insulin resistance. J Cardiometab Syndr, 1:190–6. 190. Wee CC, Mukamal KJ, Huang A et al. (2008). Obesity and C-reactive protein levels among white, black, and hispanic US adults. Obesity (Silver. Spring), 16:875–80. 191. Marfella R, Esposito K, Siniscalchi M et al. (2004). Effect of weight loss on cardiac synchronization and proinflammatory cytokines in premenopausal obese women. Diabetes Care, 27:47–52. 192. Tsilidis KK, Branchini C, Guallar E et al. (2008). C-reactive protein and colorectal cancer risk: a systematic review of prospective studies. Int J Cancer, 123:1133–40. 193. Erlinger TP, Platz EA, Rifai N et al. (2004). C-reactive protein and the risk of incident colorectal cancer. JAMA, 291:585–90. 194. Platz EA, De Marzo AM, Erlinger TP et al. (2004). No association between pre-diagnostic plasma C-reactive protein concentration and subsequent prostate cancer. Prostate, 59: 393–400. 195. Stark JR, Li H, Kraft P et al. (2009) Circulating prediagnostic interleukin-6 and C-reactive protein and prostate cancer incidence and mortality. Int J Cancer, 124(11):2683–9. 196. Heikkilä K, Ebrahim S, Lawlor DA (2007). A systematic review of the association between circulating concentrations of C reactive protein and cancer. J Epidemiol Community Health, 61:824–33. 197. Pierce BL, Neuhouser ML, Wener MH et al. (2009). Correlates of circulating C-reactive protein and serum amyloid A concentrations in breast cancer survivors. Breast Cancer Res Treat, 114:155–67. 198. Black S, Kushner I, Samols D (2004). C-reactive protein. J Biol Chem, 279:48487–90. 199. Huber J, Kiefer FW, Zeyda M et al. (2008). CC chemokine and CC chemokine receptor profiles in visceral and subcutaneous adipose tissue are altered in human obesity. J Clin Endocrinol Metab, 93:3215–21. 200. Harman-Boehm I, Blüher M, Redel H et al. (2007). Macrophage infiltration into omental versus subcutaneous fat across different populations: effect of regional adiposity and the comorbidities of obesity. J Clin Endocrinol Metab, 92:2240–7.
174
N.L. Nock and N.A. Berger
201. Dehqanzada ZA, Storrer CE, Hueman MT et al. (2006). Correlations between serum monocyte chemotactic protein-1 levels, clinical prognostic factors, and HER-2/neu vaccine-related immunity in breast cancer patients. Clin Cancer Res, 12:478–86. 202. Lu Y, Cai Z, Xiao G et al. (2007). Monocyte chemotactic protein-1 mediates prostate cancerinduced bone resorption. Cancer Res, 67:3646–53. 203. Weisberg SP, Hunter D, Huber R et al. (2006). CCR2 modulates inflammatory and metabolic effects of high-fat feeding. J Clin Invest, 116:115–24. 204. Li H, Kantoff PW, Ma J et al. (2005). Prediagnostic plasma vascular endothelial growth factor levels and risk of prostate cancer. Cancer Epidemiol Biomarkers Prev, 14:1557–61. 205. Silha JV, Krsek M, Sucharda P et al. (2005). Angiogenic factors are elevated in overweight and obese individuals. Int J Obesity (Lond), 29:1308–14. 206. García de la Torre N, Rubio MA, Bordiú E et al. (2008). Effects of weight loss after bariatric surgery for morbid obesity on vascular endothelial growth factor-A, adipocytokines, and insulin. J Clin Endocrinol Metab, 93:4276–81. 207. Balasubramanian SP, Cox A, Cross SS et al. (2007). Influence of VEGF-A gene variation and protein levels in breast cancer susceptibility and severity. Int J Cancer, 121:1009–16. 208. Reeves KW, Ness RB, Stone RA et al. (2009). Vascular endothelial growth factor and breast cancer risk. Cancer Causes Control, 20:375–86. 209. Werther K, Christensen IJ, Nielsen HJ et al. (2002). Prognostic impact of matched preoperative plasma and serum VEGF in patients with primary colorectal carcinoma. Br J Cancer, 86:417–23. 210. Powolny AA, Wang S, Carlton PS et al. (2008). Interrelationships between dietary restriction, the IGF-I axis, and expression of vascular endothelial growth factor by prostate adenocarcinoma in rats. Mol Carcinog, 47:458–65. 211. Milne GL, Yin H, Morrow JD (2008). Human biochemistry of the isoprostane pathway J Biol Chem, 283:15533–7. 212. Taketo MM (1998). Cyclooxygenase-2 inhibitors in tumorigenesis. J Natl Cancer Inst, 90:1529–36. 213. Markowitz SD (2007). Aspirin and colon cancer – targeting prevention? N Engl J Med, 356:2195–8. 214. Myung SJ, Rerko RM, Yan M et al. (2006). 15-Hydroxyprostaglandin dehydrogenase is an in vivo suppressor of colon tumorigenesis. Proc Natl Acad Sci U S A, 103:12098–102. 215. Martínez ME, Heddens D, Earnest DL et al. (1999). Physical activity, body mass index, and prostaglandin E2 levels in rectal mucosa. J Natl Cancer Inst, 91:950–3. 216. Fain JN, Kanu A, Bahouth SW et al. (2002). Comparison of PGE2, prostacyclin and leptin release by human adipocytes versus explants of adipose tissue in primary culture. Prostaglandins Leukot Essent Fatty Acids, 67:467–73. 217. Hendrickse CW, Kelly RW, Radley S et al. (1994). Lipid peroxidation and prostaglandins in colorectal cancer. Br J Surg, 81:1219–23. 218. Nicholson ML, Neoptolemos JP, Clayton HA et al. (1991). Increased cell membrane arachidonic acid in experimental colorectal tumors. Gut, 32:418. 219. Hughes D, Otani T, Yang P et al. (2008). NAD+-dependent 15-hydroxyprostaglandin dehydrogenase regulates levels of bioactive lipids in non-small cell lung cancer. Cancer Prev Res (Phila Pa), 1:241–9. 220. LeFever A, Funahashi A (1990). Elevated prostaglandin E2 levels in bronchoalveolar lavage fluid of patients with bronchogenic carcinoma. Chest, 98:1397–402. 221. Larré S, Tran N, Fan C et al. (2008). PGE2 and LTB4 tissue levels in benign and cancerous prostates. Prostaglandins Other Lipid Mediat, 87:14–9. 222. Sheng H, Shao J, Washington MK et al. (2001). Prostaglandin E2 increases growth and motility of colorectal carcinoma cells. J Biol Chem, 276:18075–81. 223. Hinson RM, Williams JA, Shacter E (1996). Elevated interleukin 6 is induced by prostaglandin E2 in a murine model of inflammation: possible role of cyclooxygenase-2. Proc Natl Acad Sci U S A, 93:4885–90.
5
Obesity and Cancer: Overview of Mechanisms
175
224. Kawamori T, Rao CV, Seibert K et al. (1998). Chemopreventive activity of celecoxib, a specific cyclooxygenase-2 inhibitor, against colon carcinogenesis. Cancer Res, 58: 409–12. 225. Brueggemeier RW, Richards JA, Petrel TA (2003). Aromatase and cyclooxygenases: enzymes in breast cancer. J Steroid Biochem Mol Biol, 86:501–7. 226. Le Bras M, Clément MV, Pervaiz S et al. (2005). Reactive oxygen species and the mitochondrial signaling pathway of cell death. Histol Histopathol, 20:205–19. 227. Flowers L, Bleczinski WF, Burczynski ME et al. (1996). Disposition and biological activity of benzo[a]pyrene-7,8-dione. A genotoxic metabolite generated by dihydrodiol dehydrogenase. Biochemistry, 35:13664–72. 228. Ghafourifar P, Cadenas E (2005). Mitochondrial nitric oxide synthase. Trends Pharmacol Sci, 26:190–5. 229. Szabó C, Ischiropoulos H, Radi R (2007). Peroxynitrite: biochemistry, pathophysiology and development of therapeutics. Nat Rev Drug Discov, 6:662–80. 230. Valko M, Izakovic M, Mazur M et al. (2004). Role of oxygen radicals in DNA damage and cancer incidence. Mol Cell Biochem, 266:37–56. 231. Faraci FM, Didion SP (2004). Vascular protection: superoxide dismutase isoforms in the vessel wall. Arterioscler Thromb Vasc Biol, 24:1367–73. 232. Brigelius-Flohé R (2006). Glutathione peroxidases and redox-regulated transcription factors. Biol Chem, 387:1329–35. 233. Kehrer JP (2000). The Haber-Weiss reaction and mechanisms of toxicity. Toxicology, 149:43–50. 234. Dean RT, Fu S, Stocker R et al. (1997). Biochemistry and pathology of radical-mediated protein oxidation. Biochem J, 324:1–18. 235. Basu S (2008). F2-Isoprostanes in human health and diseases: From molecular mechanisms to clinical implications. Antioxid Redox Signal, 10:1405–34. 236. Couillard C, Ruel G, Archer WR et al. (2005). Circulating levels of oxidative stress markers and endothelial adhesion molecules in men with abdominal obesity. J Clin Endocrinol Metab, 90:6454–9. 237. Weinbrenner T, Schröder H, Escurriol V et al. (2006). Circulating oxidized LDL is associated with increased waist circumference independent of body mass index in men and women. Am J Clin Nutr, 83:30–5. 238. Uzun H, Zengin K, Taskin M et al. (2004). Changes in leptin, plasminogen activator factor and oxidative stress in morbidly obese patients following open and laparoscopic Swedish adjustable gastric banding. Obes Surg, 14:659–65. 239. Delimaris I, Faviou E, Antonakos G et al. (2007). Oxidized LDL, serum oxidizability and serum lipid levels in patients with breast or ovarian cancer. Clin Biochem, 40: 1129–34. 240. Suzuki K, Ito Y, Wakai K et al. (2004). Serum oxidized low-density lipoprotein levels and risk of colorectal cancer: a case-control study nested in the Japan Collaborative Cohort Study. Cancer Epidemiol Biomarkers Prev, 13:1781–7. 241. Chen KH, Srivastava DK, Singhal RK et al. (2000). Modulation of base excision repair by low density lipoprotein, oxidized low density lipoprotein and antioxidants in mouse monocytes. Carcinogenesis, 21:1017–22. 242. Haghdoost S, Czene S, Näslund I et al. (2005). Extracellular 8-oxo-dG as a sensitive parameter for oxidative stress in vivo and in vitro. Free Rad Res, 39:153–62. 243. Tsuzuki T, Nakatsu Y, Nakabeppu Y (2007). Significance of error-avoiding mechanisms for oxidative DNA damage in carcinogenesis. Cancer Sci, 98:465–70. 244. Yamauchi M, Nakano H, Maekawa J et al. (2005). Oxidative stress in obstructive sleep apnea. Chest, 127:1674–9. 245. de la Maza MP, Olivares D, Hirsch S et al. (2006). Weight increase and overweight are associated with DNA oxidative damage in skeletal muscle. Clin Nutr, 25:968–76.
176
N.L. Nock and N.A. Berger
246. Azad MB, Chen Y, Gibson SB (2009). Regulation of autophagy by reactive oxygen species (ROS): implications for cancer progression and treatment. Antioxid Redox Signal, 11:777–90. 247. Aykin-Burns N, Amhad IM, Zhu Y et al. (2009). Increased levels of superoxide and H2 O2 mediate the differential susceptibility of cancer cells versus normal cells to glucose deprivation. Biochem J, 418:29–37. 248. Valko M, Rhodes CJ, Moncol J et al. (2006). Free radicals, metals and antioxidants in oxidative stress-induced cancer. Chem Biol Interact, 160:1–40. 249. Thomson CA, Giuliano AR, Shaw JW et al. (2007). Diet and biomarkers of oxidative damage in women previously treated for breast cancer. Nutr Cancer, 51:146–54. 250. Rossner P Jr, Gammon MD, Terry MB et al. (2006). Relationship between urinary 15F2t-isoprostane and 8-oxodeoxyguanosine levels and breast cancer risk. Cancer Epidemiol Biomarkers Prev, 15:639–44. 251. Leslie NR (2006). The redox regulation of PI 3-kinase-dependent signaling. Antioxid Redox Signal, 8:1765–74. 252. Chen Y, Azad MB, Gibson SB (2009). Superoxide is the major reactive oxygen species regulating autophagy. Cell Death. Differ, 16:1040–52. 253. McGill AT (2008). Malnutritive obesity (‘malnubesity’): is it driven by human brain evolution? Metab Syndr Relat Disord, 6:241–6. 254. Reitman A, Friedrich I, Ben-Amotz A et al. (2002). Low plasma antioxidants and normal plasma B vitamins and homocysteine in patients with severe obesity. Isr Med Assoc J, 4:590–3. 255. Zhu YG, Zhang SM, Wang JY et al. (2006). Overweight and obesity-induced oxidative stress in children. Biomed Environ Sci, 19:353–9. 256. Lawson JA, Rokach J, FitzGerald GA (1999). Isoprostanes: formation, analysis and use as indices of lipid peroxidation in vivo. J Biol Chem, 274:24444. 257. Davi G, Guagnano MT, Ciabattoni G et al. (2002). Platelet activation in obese women: role of inflammation and oxidant stress. JAMA, 288:2008–14. 258. Urakawa H, Katsuki A, Sumida Y et al. (2003). Oxidative stress is associated with adiposity and insulin resistance in men. J Clin Endocrinol Metab, 88:4673–6. 259. Pou KM, Massaro JM, Hoffmann U et al. (2007). Visceral and subcutaneous adipose tissue volumes are cross-sectionally related to markers of inflammation and oxidative stress: the Framingham Heart Study. Circulation, 116:1234–41. 260. Keaney JF, Jr., Larson MG, Vasan RS et al. (2003). Obesity and systemic oxidative stress: clinical correlates of oxidative stress in the Framingham Study Arterioscler. Thromb Vasc Biol, 23:434–9. 261. Dai Q, Gao YT, Shu XO et al. (2009). Oxidative stress, obesity, and breast cancer risk: results from the Shanghai Women’s Health Study. J Clin Oncol, 27:2482–8. 262. Chen G, Heilbrun LK, Venkatramanamoorthy R et al. (2004). Effects of low-fat and/or high-fruit-and-vegetable diets on plasma levels of 8-isoprostane-F2alpha in the Nutrition and Breast Health study. Nutr Cancer, 50:155–60. 263. Camphausen K, Ménard C, Sproull M et al. (2004). Isoprostane levels in the urine of patients with prostate cancer receiving radiotherapy are not elevated. Int J Radiat Oncol Biol Phys, 58:1536–9. 264. Mense SM, Remotti F, Bhan A et al. (2008). Estrogen-induced breast cancer: alterations in breast morphology and oxidative stress as a function of estrogen exposure. Toxicol Appl Pharm, 232:78–85. 265. Fantuzzi G (2005). Adipose tissue, adipokines, and inflammation J Allergy Clin Immunol, 115:911–9. 266. Shoelson SE, Lee J, Goldfine AB (2006). Inflammation and insulin resistance. J Clin Invest, 116:1793–801.
5
Obesity and Cancer: Overview of Mechanisms
177
267. Nock NL, Li L, Larkin EK et al. (2009). Empirical evidence for “syndrome Z”: a hierarchical 5-factor model of the metabolic syndrome incorporating sleep disturbance measures. Sleep, 32:615–22. 268. Cowey S, Hardy RW (2006). The metabolic syndrome: A high-risk state for cancer? Am J Pathol, 169:1505–22. 269. Pothiwala P, Jain SK, Yaturu S (2009). Metabolic syndrome and cancer. Metab Syndr Relat Disord, 7:279–88. 270. Giovannucci E (2007). Metabolic syndrome, hyperinsulinemia, and colon cancer: a review. Am J Clin Nutr, 86:s836–s842. 271. Xue F, Michels KB (2007). Diabetes, metabolic syndrome, and breast cancer: a review of the current evidence. Am J Clin Nutr 86:s823–s835. 272. Hill-Baskin AE, Markiewski MM, Buchner DA et al. (2009). Diet-induced hepatocellular carcinoma in genetically predisposed mice. Hum Mol Genet, 18: 2975–88. 273. Sjostrom L, Narbro K, Sjostrom CD et al. (2007). Effects of bariatric surgery on mortality in Swedish obese subjects. N Engl J Med, 357:741–52. 274. Sjostrom L, Gummesson A, Sjostrom CD et al. (2009). Effects of bariatric surgery on cancer incidence in obese patients in Sweden (Swedish Obese Subjects Study): a prospective, controlled intervention trial. Lancet Oncol, 10:653–62. 275. Adams TD, Hunt SC (2009). Cancer and obesity: effect of bariatric surgery. World J Surg, 33(10):2028–33. 276. Tang D, Liu JJ, Rundle A et al. (2007). Grilled meat consumption and PhIP-DNA adducts in prostate carcinogenesis. Cancer Epidemiol Biomarkers Prev, 16:803–8. 277. Ericson U, Wirfält E, Mattisson I et al. (2007). Dietary intake of heterocyclic amines in relation to socio-economic, lifestyle and other dietary factors: estimates in a Swedish population. Public Health Nutr, 10:616–27. 278. Larsson SC, Wolk A (2006). Meat consumption and risk of colorectal cancer: a metaanalysis of prospective studies. Int J Cancer, 119:2657–64. 279. Rohrmann S, Hermann S, Linseisen J (2009). Heterocyclic aromatic amine intake increases colorectal adenoma risk: findings from a prospective European cohort study. Am J Clin Nutr, 89:1418–24. 280. Koutros S, Cross AJ, Sandler DP et al. (2009). Meat and meat mutagens and risk of prostate cancer in the Agricultural Health Study. Cancer Epidemiol Biomarkers Prev, 17: 80–7. 281. Michels KB, Mohllajee AP, Roset-Bahmanyar E et al. (2007). Diet and breast cancer: a review of the prospective observational studies. Cancer, 109:2712–49. 282. Rybicki BA, Nock NL, Savera AT et al. (2006). Polycyclic aromatic hydrocarbon-DNA adduct formation in prostate carcinogenesis. Cancer Lett, 239:157–67. 283. Phillips DH (1999). Polycyclic aromatic hydrocarbons in the diet. Mutat Res, 443:139–47. 284. Sinha R, Cross A, Curtin J et al. (2005). Development of a food frequency questionnaire module and databases for compounds in cooked and processed meats. Mol Nutr Food Res, 49:648–55. 285. Rundle A, Madsen A, Orjuela M et al. (2005). The association between benzo[a]pyreneDNA adducts and body mass index, calorie intake and physical activity. Biomarkers, 12:123–32. 286. Nock NL, Tang D, Rundle A et al. (2007). Associations between smoking, polymorphisms in polycyclic aromatic hydrocarbon (PAH) metabolism and conjugation genes and PAH-DNA adducts in prostate tumors differ by race. Cancer Epidemiol Biomarkers Prev, 16:1236–45. 287. Gammon MD, Sagiv SK, Eng SM et al. (2004). Polycyclic aromatic hydrocarbon-DNA adducts and breast cancer: a pooled analysis. Arch Environ Health, 59:640–9. 288. Sinha R, Peters U, Cross AJ et al. (2005). Meat, meat cooking methods and preservation, and risk for colorectal adenoma. Cancer Res, 65:8034–41.
178
N.L. Nock and N.A. Berger
289. Ivy JL (1997). Role of exercise training in the prevention and treatment of insulin resistance and non-insulin-dependent diabetes mellitus. Sports Med, 24:321–36. 290. Schmitz KH, Ahmed RL, Yee D (2002). Effects of a 9-month strength training intervention on insulin, insulin-like growth factor (IGF)-I, IGF-binding protein (IGFBP)-1, and IGFBP-3 in 30–50-year-old women. Cancer Epidemiol Biomarkers Prev, 11:1597–604. 291. Richter EA, Ruderman NB (2009). AMPK and the biochemistry of exercise: implications for human health and disease. Biochem J, 418:261–71. 292. Bell RJ, Davison SL, Papalia MA et al. (2007). Endogenous androgen levels and cardiovascular risk profile in women across the adult life span. Menopause, 14:630–8. 293. Plaisance EP, Grandjean PW (2006). Physical activity and high-sensitivity C-reactive protein. Sports Med, 36:443–58. 294. Pedersen BK, Febbraio MA (2008). Muscle as an endocrine organ: focus on muscle-derived interleukin-6. Physiol Rev, 88:1379. 295. Mathur N, Pedersen BK (2008). Exercise as a mean to control low-grade systemic inflammation. Mediators Inflamm, 2008:109502. 296. Petersen AM, Pedersen BK (2006). The role of IL-6 in mediating the anti-inflammatory effects of exercise. J Physiol Pharmacol, 57:43–51. 297. Gu JW, Gadonski G, Wang J et al. (2004). Exercise increases endostatin in circulation of healthy volunteers. BMC Physiol, 4:2. 298. Elosua R, Molina L, Fito M et al. (2003). Response of oxidative stress biomarkers to a 16week aerobic physical activity program, and to acute physical activity, in healthy young men and women. Atherosclerosis, 167:327–34. 299. Wang JS, Lee T, Chow SE (2008). Role of exercise intensities in oxidized low-density lipoprotein-mediated redox status of monocyte in men. J Appl Physiol, 101:740–4. 300. Woods JA, Davis JM, Smith JA et al. (1999). Exercise and cellular innate immune function. Med Sci Sports Exerc, 31:66. 301. Cordain L, Latin RW, Behnke JJ (1986). The effects of aeorobic running program in bowel transit time. J Sports Med Phys Fitness, 26:101–4. 302. Holdstock DJ, Misiewicz JJ, Smith T et al. (1970). Propulsion (mass movements) in the human colon and its relationship to meals and somatic activity. Gut, 11:91–9. 303. Qualtrough D, Kaidi A, Chell S et al. (2007). Prostaglandin F(2alpha) stimulates motility and invasion in colorectal tumor cells. Int J Cancer, 121:734–40. 304. Mancini MC, Halpern A (2006). Investigational therapies in the treatment of obesity. Expert Opin Investig Drugs, 15:897–915. 305. Jakicic JM, Clark K, Coleman E et al. (2001). American College of Sports Medicine position stand. Appropriate intervention strategies for weight loss and prevention of weight regain for adults. Med Sci Sports Exerc, 33:2145–56. 306. Nissen SE, Wolski K (2007). Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes. N Engl J Med, 356:2457–71. 307. Sachdev D, Li SL, Hartell JS et al. (2003). A chimeric humanized single-chain antibody against the type I insulin-like growth factor (IGF) receptor renders breast cancer cells refractory to the mitogenic effects of IGF-I. Cancer Res, 63:627–35. 308. Pollak M (2008). Insulin and insulin-like growth factor signalling in neoplasia. Nat Rev Cancer, 8:915–28. 309. Ryan PD, Goss PE (2008). The emerging role of the insulin-like growth factor pathway as a therapeutic target in cancer. Oncologist, 13:16–24. 310. Zakikhani M, Dowling R, Fantus IG et al. (2006). Metformin is an AMP kinase-dependent growth inhibitor for breast cancer cells. Cancer Res, 66:1(0269)–73. 311. Goodwin PJ, Ligibel JA, Stambolic V (2009). Metformin in breast cancer: time for action. J Clin Oncol, 27:3271–3. 312. Jiralerspong S, Palla SL, Giordano SH et al. (2009). Metformin and pathologic complete responses to neoadjuvant chemotherapy in diabetic patients with breast cancer. J Clin Oncol, 27:3297–302.
5
Obesity and Cancer: Overview of Mechanisms
179
313. Maira SM, Stauffer F, Schnell C et al. (2009). PI3K inhibitors for cancer treatment: where do we stand? Biochem Soc Trans, 37:265–72. 314. Motzer RJ, Escudier B, Oudard S et al. (2008). Efficacy of everolimus in advanced renal cell carcinoma: a double-blind, randomised, placebo-controlled phase III trial. Lancet, 372:449–56. 315. Schwab J, Antonescu C, Boland P et al. (2009). Combination of PI3K/mTOR inhibition demonstrates efficacy in human chordoma. Anticancer Res, 29:1867–71. 316. Serra V, Markman B, Scaltriti M et al. (2008). NVP-BEZ235, a dual PI3K/mTOR inhibitor, prevents PI3K signaling and inhibits the growth of cancer cells with activating PI3K mutations. Cancer Res, 68:8022–30. 317. Moore T, Beltran L, Carbajal S et al. (2009). Dietary energy balance modulates signaling through the Akt/mammalian target of rapamycin pathways in multiple epithelial tissues. Cancer Prev Res (Phila Pa), 1:65–76.
Chapter 6
Caloric Restriction and Cancer Fei Xue and Karin B. Michels
Abstract In various animal models, caloric restriction is the most effective and reproducible intervention to extend life span and to reduce risks of aging-related chronic diseases, particularly cancer. Findings from human studies based on ecologic comparisons, the Norwegian famine during World War II, and patients with anorexia nervosa suggest that caloric restriction reduces cancer risk, especially the risk of breast cancer. In contrast, transient and abrupt caloric restriction with malnutrition followed by compensatory overnutrition may counter any protection conferred. Several earlier hypotheses for the biological mechanisms underlying the association between caloric restriction and longer life span and decreased cancer risk such as retarded growth and development, reduced metabolism rate, endocrinological changes, and decreased accumulation of oxidative damage were refuted by laboratory results. More recent findings suggest a hormesis hypothesis proposing that caloric restriction conveys a low-intensity biological stress on organisms, which may elicit an adaptive response of enhanced maintenance and repair. The identification of a new class of caloric restriction mimetic molecules that target the SIR2 family of longevity-promoting enzymes may provide a novel intervention for the prevention and treatment of cancer and other aging-related chronic diseases. Epigenetic mechanisms may also play a role.
1 Caloric Restriction Studies in Animals In 1935, McCay and colleagues published data suggesting that restricted diets after weaning may increase the mean and maximum life spans in laboratory rats [59]. In this study of 106 white rats, 34 rats were allowed all the feed desired and grew K.B. Michels (B) Obstetrics and Gynecology Epidemiology Center, Department of Obstetrics, Gynecology and Reproductive Biology, Brigham and Women’s Hospital, Harvard Medical School, 221 Longwood Avenue, Boston, MA 02115, USA; and Department of Epidemiology, Harvard School of Public Health, Boston, MA, USA e-mail:
[email protected]
N.A. Berger (ed.), Cancer and Energy Balance, Epidemiology and Overview, Energy Balance and Cancer 2, DOI 10.1007/978-1-4419-5515-9_6, C Springer Science+Business Media, LLC 2010
181
182
F. Xue and K.B. Michels
normally, 36 were restricted in food intake to limit weight gain to 10 g every 2–3 months from the time of weaning, and 36 were allowed sufficient feed to permit normal growth for 2 weeks after weaning and then were restricted in food intake. Relative to the group on an ad libitum diet (483 ± 59 days), the two groups with restricted feeding were more likely to attain extreme ages, and the male rats on a restricted diet had a longer mean life span (820 ± 113 days) [59]. During the following 70 years, McCay’s findings have been confirmed in various animal species, ranging from non-mammalian species such as fish and flies to rodents and mice [97, 93, 62, 64, 65]. Evidence from additional studies suggested that the increased survival among rodents by food restriction was due to caloric restriction rather than restriction of specific nutrients (e.g., protein or fat or minerals) [63]. This extension of the life span of rodents was found to be attained along with retarded age-related pathology and diseases such as cancer, and it was proposed that caloric restriction increases survival by delaying these processes rather than altering aging and senescence [62]. While some studies suggest that caloric restriction preferentially affects physiological aging processes [61, 98, 52], a large body of laboratory studies indicate that caloric restriction can prevent pathology of chronic diseases, including cancer. The first observations on the effect of caloric restriction on cancer occurrence were made in 1909 by Moreschi and in 1914 by Rous; both studies demonstrated that animals on a lower calorie diet showed reduced growth of transplanted tumors than ad libitum-fed controls. Subsequently, Tannenbaum et al. showed that underfeeding inhibited the development of spontaneous or induced tumors in several different mouse strains [85]. Since the earlier studies of caloric restriction involved underfeeding of the same diet as that for controls, it seemed likely that deprivation of essential nutrients rather than caloric restriction was related to cancer growth. To address this question, Tannenbaum formulated a diet of known energy content and studied levels of energy present in various diets in relation to cancer risk [86] and found that at the same level of energy intake, a diet high in fat was related to higher tumor incidence [88]. This finding was later confirmed by Boutwell et al. [8]. In addition, Tannenbaum showed that caloric restriction was more effective in reducing cancer risk during the promotion rather than the initiation stage of tumorigenesis and caloric restriction during the initiation stage was ineffective if followed by an ad libitum diet [87]. Subsequently, data from decades of research among rodents consistently suggested that caloric restriction lowers the incidence of a variety of spontaneous as well as induced or transplanted tumors and that this effect is proportional to the extent of caloric restriction [49, 43, 74, 51]. A meta-analysis of 14 animal experiments reported a 55% reduction of the incidence of spontaneous mammary tumors associated with caloric restriction ranging from 23 to 40% in mice [18]. Kritchevsky et al. and Klurfeld et al. found that 40% caloric restriction effectively inhibited the growth of induced tumors among animals [49, 43]. In addition to the established tumor-inhibiting effect of 40% caloric restriction, it was found that caloric restriction of 10, 20, and 30% in 7,12-Dimethylbenz(a)anthracene (DMBA)-treated female Sprague-Dawley rats was related to a reduced tumor incidence or tumor multiplicity [44]. Though energy restriction in experimental animal studies can be achieved in
6
Caloric Restriction and Cancer
183
several ways and sources of energy may include various nutrients such as fat, carbohydrates, and protein, the effect of caloric restriction on cancer risk was found to be independent of dietary components [2, 74, 44, 45, 51]. Klurfeld et al. reported a significantly reduced tumor incidence among DMBA-treated female rats fed a diet containing 20 or 26.7% fat under conditions of 25% energy restriction compared to DMBA-treated rats fed with diets containing 5, 15, or 20% fat ad libitum [45]. Compared with consistent results from short-lived mammals such as rodents, less data are available for non-human primates. Preliminary results suggest that caloric restriction may prevent or delay the onset and decrease the mortality due to several aging-related diseases, including cancer, among non-human primates (rhesus and squirrel monkeys) [79, 54]. In a large non-human primate study at the National Institute on Aging and another similar study from the University of Wisconsin, rhesus and squirrel monkeys with caloric restriction had substantially lower number and types of neoplasia than controls [54].
2 Caloric Restriction Studies in Humans 2.1 Ecologic Studies In an ecologic study, Armstrong and Doll assessed the correlation between total per capita caloric intake and cancer incidence and mortality based on concurrent food and cancer data from 33 countries [5]. The site and sex-specific correlation coefficient suggested a linear and statistically significant relation for several cancer sites: it was 0.57 for breast cancer, 0.66 for colon cancer, 0.56 for rectal cancer, 0.65 for endometrial cancer, and 0.64 for kidney cancer among women; among men the coefficient was 0.60 for colon cancer, 0.75 for rectal cancer, 0.55 for kidney cancer, and 0.56 for cancer of the nervous system [5]. Another similar study compared per capita food consumption with cancer mortality among 19 countries and reported a correlation coefficient between total calorie intake and cancer of 0.77 for breast cancer, 0.60 for prostate cancer, and 0.74 and 0.71 for male and female colon cancer, respectively [47]. Several cancer sites are more prevalent in developed countries than in developing countries (breast cancer, colon cancer, prostate cancer, pancreatic cancer, endometrial cancer, ovarian cancer, bladder cancer, kidney cancer, and leukemia) [1]; however, distributions are changing with the increase in obesity worldwide. Cancer incidence increases among migrants from low-incidence countries to high-incidence countries, suggesting the importance of environmental and lifestyle factors in cancer etiology [1].
2.2 Anorexia Nervosa Studies A human starvation study was conducted in the 1940s investigating the mechanism of famine edema, in which famine edema was produced experimentally in normal
184
F. Xue and K.B. Michels
men who lost a quarter of their body weight while subsisting for 6 months on a European type of semi-starvation diet [40]. Though extracellular water to cellular tissue was roughly doubled among these study subjects, there were no signs of renal or cardiac failure, suggesting famine edema is not simply a result of hypoproteinemia or of renal or cardiac failure [40]. Caloric restriction as an intervention in humans is generally considered infeasible and unethical. Voluntary starvation found in patients who suffer from anorexia nervosa provides an opportunity to examine the effect of low caloric intake on cancer incidence. Anorexia nervosa is an eating disorder that occurs primarily during adolescence and early adulthood and is characterized by very low caloric intake, low intake of all micro- and macro nutrients, low body mass index, and amenorrhea. The relation of anorexia nervosa to subsequent cancer incidence has been assessed in two studies [69, 70] (Table 6.1). In a retrospective cohort of hospitalized anorexia nervosa patients identified from the population-based Danish Psychiatric Case Register, Mellemkjaer et al. reported a slight but not statistically significant reduction in the overall cancer incidence among women (standardized incidence ratio [SIR] = 0.80; 95% CI 0.52–1.18) as compared with that of the general population [69]. In a retrospective cohort study conducted in Sweden, women hospitalized for anorexia nervosa prior to age 40 identified from the Swedish inpatient registry had a 53% lower incidence of breast cancer later in life (SIR = 0.47; 95% CI 0.19–0.97) compared to the general population [70]. The reduced breast cancer incidence was more pronounced among parous women with anorexia nervosa (SIR=0.24, 95% CI 0.03–0.87) than among nulliparous women (SIR=0.77, 95% CI 0.25–1.79). In this study, overall cancer risk was not significantly reduced among women who had suffered from anorexia nervosa; however, parous women who had experienced anorexia tended to have a lower risk (SIR=0.66, 95% CI 0.38–1.07) [70]. While anorexia nervosa is generally associated with an extended period of very low caloric intake early in life, some affected patients also exercise excessively. Therefore, strenuous physical activity has to be considered as alternative or partial explanation for the inverse association between caloric intake and cancer incidence.
2.3 Famine Studies Observations made following famines allow a unique opportunity to investigate the influence of caloric restriction on cancer incidence. Individuals who experience time periods of famine generally have a substantially reduced caloric intake but may also experience malnutrition as they lack relevant amounts of micro- and macronutrients. In Norway, average caloric intake during World War II dropped from 3,475 kcal daily in 1939 to a minimum of 2,700 kcal in late 1944 and early 1945 [91, 90, 73, 77] (Table 6.2). Findings from studies based on the Norwegian famine data consistently suggested that caloric restriction during the war was associated with a reduced risk of breast cancer later in life, though the exposure to caloric restriction was not assessed at individual levels in these studies. In particular, the incidence of breast cancer was found to be lower than expected among women who experienced
Mellemkjaer, 2001, Denmark
Study (author, year, country)
Retrospective cohort
Design
Cancer of all sites
Outcome 25 cases of all cancer, 2,151 women and 186 men with anorexia nervosa (population)
No. of cases and controls or population Anorexia nervosa identified in the Psychiatric Case Registry and the National Registry of Patients
Exposure Observed number of cases in anorexia nervosa patients divided by the expected number of cases in general population
Exposure categories
Table 6.1 Epidemiologic studies on anorexia nervosa and risk of cancer
SIR: All malignant neoplasms: 0.8 (0.52–1.2) Cancer sites: Buccal cavity: 3.2 (0.1–17.4) Digestive organs: 1.3 (0.3–3.9) Lung: 2.2 (0.5–6.4) Breast: 0.8 (0.3–1.7) Female genital organs: 0.3 (0.0–1.2) Urinary system: 1.6 (0.0–8.7) Skin: 0.6 (0.2–1.5) Brain and nervous system: 0.6 (0.0–3.2) Thyroid gland: 1.8 (0.0–10.0) Non-Hodgkin’s lymphoma: 3.0 (0.4–10.8)
OR/HR/SIR (95%) and P for trend/interaction
Standardized by age, gender, and calendar time
Adjustment for covariates
6 Caloric Restriction and Cancer 185
Design
Retrospective cohort
Study (author, year, country)
Michels and Ekbom, 2004
Cancer of all sites and breast cancer
Outcome 52 cases of all cancer and 7 cases of breast cancer, 7,303 women with anorexia nervosa (population)
No. of cases and controls or population Hospitalization and treated for anorexia nervosa identified in the Swedish Inpatient Registry
Exposure
Table 6.1 (continued)
Observed number of cases in anorexia nervosa patients divided by the expected number of cases in general population
Exposure categories SIR: All cancer: Overall: 0.92 (0.69–1.21) Parous: 0.66 (0.38–1.07) Nulliparous: 1.12 (0.78–1.55) Breast cancer: 0.47 (0.19–1.97) Parous: 0.24 (0.03–0.87) Nulliparous: 0.77 (0.25–1.79)
OR/HR/SIR (95%) and P for trend/interaction
Standardized by age and calendar time
Adjustment for covariates
186 F. Xue and K.B. Michels
6
Caloric Restriction and Cancer
187
puberty during the war [90]. When residential history in areas with and without food production was used as the proxy to severity of exposure to caloric restriction, a lower incidence of breast cancer was found among women who resided in the non-food-producing compared to those in food-producing areas [77]. Furthermore, when height was assessed as a proxy of early nutritional status, height was more strongly associated with breast cancer incidence among women who were born or who went through peri-pubertal growth during the famine years [91, 73]. These results suggest the greater nutritional diversity in early life during famine years, as reflected by variations in height, would be of particular importance for women in their later risk of breast cancer. During the end of World War II, the densely populated Western parts of the Netherlands were cut off from food supply and daily rations per capita dropped from about 1,500 kcal to about 700 kcal between September 1944 and May 1945 with about 500 kcal at the height of the famine [21]. In the initial Dutch Famine Study “place of residence” during the famine was assessed as a proxy for individual exposure level; residents of the Western area who were exposed to famine had a higher risk of breast cancer than those who resided in unexposed areas [17]. In a subsequent study, in which the individual level of exposure to famine was assessed using a severity score derived from a questionnaire inquiring about hunger, cold, and weight loss, women severely exposed to famine had a higher risk of breast cancer (hazard ratio [HR]=1.48, 95% confidence interval [CI] 1.09–2.01) relative to unexposed women; the association was stronger for women exposed at ages 2–9 compared to women exposed at older ages (≥10 years) [21]. Overall cancer incidence among the Dutch famine population was 25% higher among women who experienced severe famine (95% CI 1–55) than the unexposed, but this increase was largely driven by the increase in breast cancer incidence and the association was attenuated after breast cancer was excluded (HR=1.12, 95% CI 0.87–1.43) [24]. The involuntary starvation of individuals during the famine and the voluntary starvation among individuals suffering from anorexia nervosa may have different physiologic effects. Food deprivation during the famine was relatively brief and spanned less than 1 year. It started and ended abruptly and was followed by availability of sufficient food. It is likely that the individuals exposed to famine compensated for the starvation by a feasting period. Indeed, individuals exposed to the Dutch famine had a higher body mass index (BMI) 25 years later than individuals who were not exposed, and the BMI was higher the more extreme the starvation had been: the average body mass index of subjects with severe exposure to famine (mean=26.1 kg/m2 ) was found to be higher than those with moderate exposure (mean=25.8 kg/m2 ) or no exposure (mean=25.7 kg/m2 ) [21]. Accordingly, exposure to famine was associated with elevated postmenopausal serum levels of insulin-like growth factor (IGF)-I [22, 23]. In contrast, women suffering from anorexia nervosa usually fast for extended periods of time, often years, and their deprivation is not following by overeating, but their caloric intake may approach normal ranges at best. This difference in behavior patterns may affect hormone levels differentially and explain the opposite association of exposure to the Dutch famine and to anorexia nervosa with breast cancer incidence.
Dirx, 1999, The Netherlands
Study (author, year, country)
Cohort
Design
Breast cancer
Outcome 1,009 cases, 62,573 (population)
No. of cases and controls or population Residence in the hunger winter (1944–1945) and war years (1940–1944) and fathers’ employment status (1932–1940) as indicators of caloric restriction
Exposure 1932–1940: Father had a job Father had no job 1940–1944: Rural area in 1942 City in 1942 1944–1945: Non-West Western rural area Western city
Exposure categories
Table 6.2 Epidemiologic studies on famine and risk of cancer
1.1 (0.9–1.4)
1.0 1.3 (1.0–1.7)
1.0 (0.8–1.2)
1.0
1.0 1.0 (0.7–1.3)
OR/HR/SIR (95%) and P for trend/interaction
Age, age at menopause, parity, age at first birth, maternal breast cancer, breast cancer in sister(s), benign breast cancer disease, alcohol use, energy consumption, education, age at menarche, height
Adjustment for covariates
188 F. Xue and K.B. Michels
Nilsen and Vatten, 2001, Norway
Study (author, year, country)
Cohort
Design
Breast cancer
Outcome 215 cases, 25,204 (population)
No. of cases and controls or population Differences in achieved adult height as a reflection of differences in childhood nutrition
Exposure
Table 6.2 (continued)
Tertiles (cutoff=162 cm and 167 cm) of height in birth cohorts 1925–1929: T1 T2 T3 1930–1934: T1 T2 T3 1935–1940: T1 T2 T3 1940–1945: T1 T2 T3 ≥1946: T1 T2 T3
Exposure categories
1.0 1.1 (0.5–2.4) 1.0 (0.5–2.3)
1.0 1.1 (0.5–2.7) 2.5 (1.2–5.5)
1.0 0.8 (0.3–1.9) 1.3 (0.5–3.1)
1.0 0.9 (0.5–2.0) 0.9 (0.4–2.3)
1.0 0.9 (0.5–1.6) 0.5 (0.2–1.1)
OR/HR/SIR (95%) and P for trend/interaction
Adjusted for age at study entry. Further adjustment for BMI, smoking, and physical activity did not change the results
Adjustment for covariates
6 Caloric Restriction and Cancer 189
Robsahm and Tretli, 2002, Norway
Study (author, year, country)
Cohort
Design
Breast cancer
Outcome 7,311 cases, 597,906 (population)
No. of cases and controls or population Residence in areas with and without food production according to the main income sources for the municipality in 1968
Exposure
Table 6.2 (continued) OR/HR/SIR (95%) and P for trend/interaction 1.0 1.17 (1.1–1.24)
Exposure categories Food No food
Age, age at first childbirth, education, occupational physical activity, birth cohort
Adjustment for covariates
190 F. Xue and K.B. Michels
Nested case– cohort
Nested case– cohort
Elias, 2005, The Netherlands
Design
Elias, 2004, The Netherlands
Study (author, year, country)
Cancer of all sites
Breast cancer
Outcome
718 cases of cancer of all sites, including 459 cases exclusive of breast cancer 2,352 (sub-cohort) randomly selected from 15,396 women
585 cases, 2,352 (sub-cohort) randomly selected from 15,396 women
No. of cases and controls or population OR/HR/SIR (95%) and P for trend/interaction 1.0 1.13 (0.92–1.38) 1.48 (1.09–2.01) 0.016
1.0 1.10 (0.96–1.27) 1.25 (1.01–1.55) 0.03 1.0 1.07 (0.91–1.26) 1.12 (0.87–1.43) 0.318
Exposure categories Not exposed Moderately exposed Severely exposed P for trend
All cancer sites: Unexposed Moderately exposed Severely exposed P for trend All cancer sites, exclusive of breast cancer: Unexposed Moderately exposed Severely exposed
Exposure Famine score assessing personal experience with hunger, cold, and weight loss during famine
Famine score assessing personal experience with hunger, cold, and weight loss during famine
Table 6.2 (continued)
Age, age squared, body mass index, height, socioeconomic status, age at menarche, parity, age at birth of first child, and family history of breast cancer (first-degree relative) Age, age squared, body mass index, height, socioeconomic status, cigarette smoking habits
Adjustment for covariates
6 Caloric Restriction and Cancer 191
192
F. Xue and K.B. Michels
3 Potential Mechanisms A number of biologic and metabolic responses to caloric restriction may influence cancer risk. Reduced cell proliferation and DNA synthesis and enhanced apoptosis and DNA repair limit the number of preneoplastic lesions. Caloric restriction also alters levels of several hormones and growth factors, including estrogen, insulin, prolactin, leptin, IGF-I, and insulin - like growth factor binding protein (IGFBP)-3, which may affect hormone-related cancers [51, 2, 94, 39, 34]. Despite the consistency of findings on extended life span and reduced risk of cancer and other diseases among animals exposed to caloric restriction, the underlying mechanisms have remained largely unknown. Early on, caloric restriction was thought to extend life span by retarding the growth and development of animals [59]. However, observations that caloric restriction also extended the life span of adults suggest that the effect of caloric restriction was not mediated by limiting growth and development [48]. Sacher later proposed that caloric restriction reduces metabolic rate and may thereby extend survival among rodents [81]. However, this hypothesis was contradicted by a series of subsequent laboratory tests [67, 68]. In two groups of 6-month-old male rats fed ad libitum or maintained on a life-prolonging foodrestriction regimen for 4.5 months, the metabolic rate per kilogram lean body mass measured by O2 consumption was found to be the same for both groups [67]. In another study examining changes in metabolic rate immediately after restriction of food, both metabolism rate at normal daily living conditions and basal metabolic rate decreased after feeding was restricted to 60%, but this decrease was transient, so that within a few weeks metabolic rate of restricted rats was the same as that of rats fed ad libitum [68]. Precise measurements of metabolic rate are more difficult to achieve in humans compared to controlled animal studies. Although individuals have been observed to have physiologic or pathologic changes under conditions of extended starvation or anorexia nervosa, relevant studies were not designed to assess metabolic rate; onset and continuity of metabolic rate change during caloric restriction were not as carefully monitored. Endocrinologic changes were also considered as possible mechanisms for the reduced caner risk observed. Caloric restriction was found to alter levels of a number of hormones and growth factors, including increased glucocorticoids and IGFBP-3 levels and decreased levels of IGF-1, insulin, prolactin, estrogen, and leptin, all of which may reduce the risk of cancer [74, 2, 51, 94, 39]. Although the endocrine system plays an important role in cancer etiology, evidence on the effect of caloric restriction on cancer risk through the endocrinologic changes remains conflicting [82]. For instance, though IGF-1 and insulin levels were found to be lower in both animals with caloric restriction and in dwarf mice, Bartke et al. have reported that the life span of the dwarf mice can be further increased by caloric restriction [6], suggesting caloric restriction can further increase life span at least partially through mechanisms other than reduced IGF-1 level. Caloric restriction may also reduce cancer risk by decreasing the accumulation of oxidative damage [99, 100] due to reduced production of mitochondrial free radicals [25], enhanced protective
6
Caloric Restriction and Cancer
193
mechanisms, including altering levels of antioxidant enzymes [96, 30], detoxication enzymes [4], plasma membrane antioxidant system [16, 35], and DNA repair capacity [29, 10, 84]. Nonetheless, the oxidative damage attenuation hypothesis remains elusive because findings from several other studies have suggested that oxidative stress does not alter the aging process in animals [66]. Hormesis refers to the phenomenon in which a usually detrimental environmental agent such as radiation or chemical substance provides beneficial effects when administered at low intensities or concentrations [28, 66]. The beneficial effects include increased longevity, retardation of senescent deterioration, retardation of age-associated diseases, and enhanced coping with intense stressors [66]. Caloric restriction may be one such low-intensity stressors as it induces daily elevation of circadian peak plasma-free corticosterone levels throughout the life span in both rats and mice [80, 31]. The activation of the intracellular cell-autonomous signaling pathways in response to biological stress and low nutrition would protect cells and tissues and regulate glucose, fat, and protein metabolism [82]. Increased expression of SIR2, a conserved longevity factor involved in the response to caloric restriction, was found to lengthen life span by acting on biological processes that promote survival under conditions of scarcity in yeast [32]. The homologs of SIR2 were found in a wide array of organisms in addition to yeast, ranging from bacteria [89, 78] to mammals, including humans [26, 27]. At the cellular level, SIRT1, a homolog of the SIR2 longevity factor, was found to be induced in tissues of rats with caloric restriction [13] and has been suggested to be involved in the epigenetic regulation of gene expression in cancer cells [41]. In many cancers, SIRT1 localizes to the promoters of several aberrantly silenced tumor suppressor genes whose DNA is hypermethylated [38], and the inhibition of SIRT1 increases H4-K16 and H3-K9 acetylation at endogenous promoters and suffices to induce gene re-expression in breast and colon cancer cells [75]. In addition, SIRT1 was also found to deacetylate non-histone proteins, including various transcription factors involved in growth regulation [56], DNA repair [9, 72], and apoptosis [58, 92], in the fundamental progression of cancer. The acetylation of H3K56 is increased in multiple types of cancer and SIRT1 and SIRT2 deactylate H3K56ac [15a].
4 Discussion Through nearly a century’s research, caloric restriction has been shown to reduce the incidence of spontaneous and induced tumors and increase the life span in animal models. This effect was independent of dietary composition and the period of life when caloric restriction occurred. In contrast to the abundant and consistent evidence from animal models, only limited data directly assessing the impact of caloric restriction are available from human studies, largely due to the difficulties of systematic follow-up of starving populations. It is likely that sustained rather than transient caloric restriction reduces cancer risk and expands life span. While reduction of caloric intake up to 60% in animals has been associated with a reduced risk of cancer [2], caloric restriction
194
F. Xue and K.B. Michels
among rats followed by ad libitum feeding was found to lead to rapid weight gain with concomitant appearance of tumors [50]. In another study, an 8-day period of re-feeding following chronic dietary energy restriction reversed the anti-cancer effects of caloric restriction in rats [101]. This situation is similar to the shortterm caloric restriction of the Dutch famine, where subjects may have compensated for their nutritional deprivation by overeating after the end of the food embargo. Unlike individuals in the Norway famine study who experienced a mild caloric restriction or patients with anorexia nervosa who voluntarily reduced their calorie intake, victims who experienced the Dutch famine are more likely to have subsequent compensatory eating behavior and gain extra weight, which may compromise the anti-cancer effect of caloric restriction, since higher body mass index increases cancer risk. Recently, mimetic molecules of caloric restriction have been proposed as a possible preventive intervention [82]. One example of such mimetic molecules is 2-deoxyglucose (2DG), which is a synthetic glucose analog that inhibits the glycolytic enzyme phosphohexose isomerase [82]. Although 2DG has effects similar to caloric restriction, including suppressed tumor growth, reduced insulin, and increased glucocorticoids in rodents [36], the chronic use of it may enhance the risk of congestive heart failure [36]. Another family of mimetic molecules is biguanides, including metformin, buformin, and phenformin, which may inhibit the incidence of cancer by stimulating AMP-dependent kinase, modulating appetite, glucose, and insulin metabolism [11]; however, these compounds may increase the risk of lactic acidosis [60]. A new class of promising caloric restriction mimetics with fewer side effects are molecules that target the SIR2 family of longevity-promoting enzymes (sirtuins). Howitz et al. identified 18 small molecules from plants, including resveratrol, butein, and piceatannol, that were shown to potentiate human SIRT1 activity in vitro and in vivo [33]. Among them, resveratrol, a polyphenol that is found in many plants species, had the strongest stimulatory activity [82]. Resveratrol is found in grapes [55] and other plants, including peanuts and Itadori tea, a Japanese traditional herbal remedy, which may be the richest known source of resveratrol [76]. Many epidemiologic and clinical studies on red wine, grapes, and grape juice have suggested that resveratrol may aid in disease prevention [12]. The level of total resveratrol in red seedless table grape skin was found to be 2780 μg/g [76]. Red wine contains an average of 1.9 ± 1.7 mg/l trans-resveratrol (8.2 ± 7.5 μM), ranging from non-detectable levels to 14.3 mg/l (62.7 μM) trans-resveratrol depending on grape type [83]. By mimicking caloric restriction, resveratrol extended the life span of a variety of species ranging from yeast to Caenorhabditis elegans and Drosophila melanogaster [95]. In in vitro studies, resveratrol was found to protect mammalian cells in vitro and in vivo from oxidative damage at a concentration of 10–50 μM [53] or 10–100 μM [3], from gamma radiation at a dosage of 100 μM [33], and from Bax-mediated apoptosis [14]. Furthermore, in preliminary in vitro experiments and animal studies, resveratrol was found to act as an antioxidant and anti-mutagenic agent and selectively suppresses the transcriptional activation of cytochrome P-450 1A1 and
6
Caloric Restriction and Cancer
195
inhibits the formation of carcinogen-induced preneoplastic lesions and was effective in treating esophageal cancer, breast cancer, and liver cancer, with a wide range of median effective dose (3.7–85 μM) according to various anti-inflammatory pathways [37, 7]. Resveratrol was also found to be effective in treating several other diseases, including oral herpes [19, 20], chronic obstructive pulmonary disease [15], hyperlipidemia [71], and ischemic events [42], through its antioxidant and anti-inflammatory effects in some animals and in vitro studies. The other two sirtuinactivating compounds, including quercetin and butein, were also effective against several diseases related to aging, including ischemic heart disease, cerebrovascular disease, cancer, asthma, and diabetes [46, 57]. Future studies may elucidate the long-term efficacy and potential side effects of these caloric restriction mimicking molecules and examine possible synergistic effects with lifestyle factors, e.g., diet and physical activity.
References 1. Adami HO, Hunter D, Trichopoulos D. (2002). Textbook of Cancer Epidemiology New York: Oxford University Press, Inc. 2. Albanes D (1987). Total calories, body weight, and tumor incidence in mice. Cancer Res, 47:1987–1992. 3. Araki T, Sasaki Y, Milbrandt J (2004). Increased nuclear NAD biosynthesis and SIRT1 activation prevent axonal degeneration. Science, 305:1010–1013. 4. Armeni T, Pieri C, Marra M et al. (1998). Studies on the life prolonging effects of food restriction: glutathione levels and glyoxylase enzymes in rat liver. Mech Ageing Dev, 101:101–110. 5. Armstrong B, Doll R (1975). Environmental factors and cancer incidence and mortality in different countries, with special reference to dietary practices. Int J Cancer, 15:617–631. 6. Bartke A, Wright JC, Mattison JA, et al. (2001). Extending the lifespan of long-lived mice. Nature, 414:412. 7. Bhat KP, Pezzuto JM (2002). Cancer chemopreventive activity of resveratrol. Ann N Y Acad Sci, 957:210–229. 8. Boutwell RK, Brush MK, Rusch HP (1949). The stimulatory effect of dietary fat on carcinogenesis. Cancer Res, 741–746. 9. Brunet A, Sweeney LB, Sturgill JF, et al. (2004). Stress-dependent regulation of FOXO transcription factors by the SIRT1 deacetylase. Science, 303:2011–2015. 10. Cabelof DC, Yanamadala S, Raffoul JJ, et al. (2003). Caloric restriction promotes genomic stability by induction of base excision repair and reversal of its age-related decline. DNA Repair (Amst), 2:295–307. 11. Carling D (2004). The AMP-activated protein kinase cascade – a unifying system for energy control. Trends Biochem Sci, 29:18–24. 12. Cassidy A., Hanley B, Lamuela-Raventos RM (2000). Isoflavones, lignans and stilbenes – Origins, metabolism and potential importance to human health. J Sci Food Agric, 80: 1044–1062. 13. Cohen HY, Miller C, Bitterman KJ, et al. (2004a). Calorie restriction promotes mammalian cell survival by inducing the SIRT1 deacetylase. Science, 305:390–392. 14. Cohen HY, Lavu S, Bitterman KJ, et al. (2004b). Acetylation of the C terminus of Ku70 by CBP and PCAF controls Bax mediated apoptosis. Mol Cell, 13:627–638. 15. Culpitt SV, Rogers DF, Fenwick PS, et al. (2003). Inhibition by red wine extract, resveratrol, of cytokine release by alveolar macrophages in COPD. Thorax, 58:942–6.
196
F. Xue and K.B. Michels
15a. Das C, Lucia MS, Hansen KC, et al. (2009). CBP/p300-mediated acetylation of histone H3 on lysine 56. Nature, 459:113–117. 16. De Cabo R, Cabello R, Rios M, et al. (2004). Calorie restriction attenuates age-related alterations in the plasma membrane antioxidant system in rat liver. Exp Gerontol, 39:297–304. 17. Dirx MJ, van den Brandt PA, et al. (1999). Diet in adolescence and the risk of breast cancer: results of the Netherlands Cohort Study. Cancer Causes Control, 10:189–99. 18. Dirx MJ, Zeegers MP, Dagnelie PC, et al. (2003). Energy restriction and the risk of spontaneous mammary tumors in mice: a meta-analysis. Int J Cancer, 106:766–770. 19. Docherty JJ, Fu MM, Stiffler BS, et al. (1999). Resveratrol inhibition of herpes simplex virus replication. Antivir Res, 43:145–155. 20. Docherty JJ, Smith JS, Fu MM, et al. (2004). Effect of topically applied resveratrol on cutaneous herpes simplex virus infections in hairless mice. Antivir Res, 61:19–26. 21. Elias SG, Peeters PH, Grobbee DE, et al. (2004a). Breast cancer risk after caloric restriction during the 1944–1945 Dutch famine. J Natl Cancer Inst, 96:539–546. 22. Elias SG, Onland-Moret NC, Peeters PH, et al. (2004b). Urinary endogenous sex hormone levels in postmenopausal women after caloric restriction in young adulthood. Br J Cancer, 90:115–117. 23. Elias SG, Keinan-Boker L, Peeters PH, et al. (2004c). Long term consequences of the 1944–1945 Dutch famine on the insulin-like growth factor axis. Int J Cancer, 108:628–630. 24. Elias SG, Peeters PH, Grobbee DE, et al. (2005). The 1944–1945 Dutch famine and subsequent overall cancer incidence. Cancer Epidemiol Biomarkers Prev, 14:1981–1985. 25. Feuers RJ, Weindruch R, Hart RW (1993). Caloric restriction, aging, and antioxidant enzymes. Mutat Res, 295:191–200. 26. Frye RA (1999). Characterization of five human cDNAs with homology to the yeast SIR2 gene: Sir2-like proteins (sirtuins) metabolize NAD and may have protein ADPribosyltransferase activity. Biochem Biophys Res Commun, 260:273–279. 27. Frye RA (2000). Phylogenetic classification of prokaryotic and eukaryotic Sir2-like proteins. Biochem Biophys Res Commun, 273:793–798. 28. Furst A (1987). Hormetic effects in pharmacology: pharmacological inversions as prototypes for hormesis. Health Phys, 52:527–530. 29. Guo Z, Heydari A, Richardson A (1998). Nucleotide excision repair of actively transcribed versus nontranscribed DNA in rat hepatocytes: effect of age and dietary restriction. Exp Cell Res, 245:228–238. 30. Guo ZM, Yang H, Hamilton ML, et al. (2001). Effects of age and food restriction on oxidative DNA damage and antioxidant enzyme activities in the mouse aorta. Mech Ageing Dev, 2001;122:1771–1786. 31. Han ES, Levin N, Bengani N, et al. (1995). Hyperadrenocorticism and food restrictioninduced life extension in the rat: evidence for divergent regulation of pituitary proopiomelanocortin RNA and adrenocorticotropic hormone biosynthesis. J Gerontol A Biol Sci Med Sci, 50:B288–294. 32. Hekimi S, Guarente L (2003). Genetics and the specificity of the aging process. Science, 299:1351–1354. 33. Howitz KT, Bitterman KJ, Cohen HY, et al. (2003). Small molecule activators of sirtuins extend Saccharomyces cerevisiae lifespan. Nature, 425:191–196. 34. Hursting SD, Lavigne JA, Berrigan D, et al. (2003). Calorie restriction, aging, and cancer prevention: mechanisms of action and applicability to humans .Annu Rev Med, 54:131–152. 35. Hyun DH, Emerson SS, Jo DG, et al. (2006). Calorie restriction up-regulates the plasma membrane redox system in brain cells and suppresses oxidative stress during aging. Proc Natl Acad Sci U S A, 103:19908–19912. 36. Ingram DK, Anson RM, de Cabo R, et al. (2004). Development of calorie restriction mimetics as a prolongevity strategy. Ann N Y Acad Sci, 1019:412–423. 37. Jang M, Cai L, Udeani GO, Slowing KV et al. (1997). Cancer chemopreventive activity of resveratrol, a natural product derived from grapes. Science, 275:218–220.
6
Caloric Restriction and Cancer
197
38. Jones PA, Baylin SB (2002). The fundamental role of epigenetic events in cancer. Nat Rev Genet, 3:415–428. 39. Kaaks R, Lukanova A (2001). Energy balance and cancer: the role of insulin and insulin-like growth factor-I. Proc Nutr Soc, 60:91–106. 40. Keys A, Taylor HL, Mickelsen O, et al. (1946). Famine Edema and the Mechanism of Its Formation. Science, 103:669–670. 41. Kim EJ, Um SJ (2008). SIRT1: roles in aging and cancer. BMB Rep, 41:751–6. 42. Kiziltepe U, Turan NN, Han U, et al. (2004). Resveratrol, a red wine polyphenol, protects spinal cord from ischemia-reperfusion injury. J Vasc Surg, 40:138–145. 43. Klurfeld DM, Weber MM, Kritchevsky D (1987). Inhibition of chemically-induced mammary and colon tumor promotion by caloric restriction in rats fed increased dietary fat. Cancer Res, 47:2759–2762. 44. Klurfeld DM, Welch CB, Davis MJ, et al. (1989a). Determination of degree of energy restriction necessary to reduce DMBA-induced mammary tumorigenesis in rats during the promotion phase. J Nutr, 119:286–291. 45. Klurfeld DM, Welch CB, Lloyd LM, et al. (1989b). Inhibition of DMBA-induced mammary tumorigenesis in rats fed high fat diets. Int J Cancer, 43:922–925. 46. Knekt P, Kumpulainen J, Järvinen R, et al. (2002). Flavonoid intake and risk of chronic diseases. Am J Clin Nutr, 76:560–568. 47. Kodama M, Kodama T (1990). Interrelation between Western type cancers and non-Western type cancers as regards their risk variations in time and space. II. Nutrition and cancer risk. AntiCancer Res, 10:1043–1049. 48. Kristal, BS, Yu, BP (1994). Aging and its modulation by dietary restriction. in: Yu, B. P., (ed.), Modulation of Aging Processes by Dietary Restriction, pp.1–36. CRC Press, Boca Raton, FL,. 49. Kritchevsky D, Weber MM, Klurfeld DM (1984). Dietary fat versus caloric content in initiation and promotion of 7,12-dimethylbenz(a)anthracene-induced mammary tumorigenesis in rats. Cancer Res, 44:3174–3177. 50. Kritchevsky D, Welch CB, Klurfeld DM (1989). Response of mammary tumors to caloric restriction for different time periods during the promotion phase. Nutr Cancer, 12: 259–269. 51. Kritchevsky D (1997). Caloric restriction and experimental mammary carcinogenesis. Breast Cancer Res Treat, 46:161–7. 52. Kubo C, Johnson BC, Gajjar A, et al. (1987). Crucial dietary factors in maximizing life span and longevity in autoimmune-prone mice. J Nutr, 117:1129–1135. 53. Kutuk O, Adli M, Poli G, et al. (2004). Resveratrol protects against 4-HNE induced oxidative stress and apoptosis in Swiss 3T3 fibroblasts. Biofactors, 20:1–10. 54. Lane MA, Black A, Handy A, et al. (2001). Caloric restriction in primates. Ann N Y Acad Sci, 928:287–295. 55. Langcake P, Pryce RJ (1976). The production of resveratrol by Vitis vinifera and other members of the Vitaceae as a response to infection and injury. Physiol. Plant Pathol, 9:77–86. 56. Langley E, Pearson M, Faretta M, et al. (2002). Human SIR2 deacetylates p.53 and antagonizes PML/p53-induced cellular senescence. EMBO J, 21:2383–2396. 57. Lim SS, Jung SH, Ji J, et al. (2001). Synthesis of flavonoids and their effects on aldose reductase and sorbitol accumulation in streptozotocin-induced diabetic rat tissues. J Pharm Pharmacol, 53(5):653–668. 58. Luo J, Nikolaev AY, Imai S, et al. (2001). Negative control of p53 by Sir2alpha promotes cell survival under stress. Cell, 107:137–148. 59. McCay CM, Crowell MF, Maynard LA (1935). The effect of retarded growth upon the length of lifespan and upon the ultimate body size. J Nutr, 10:63–79. 60. McCarty MF (2004). Chronic activation of AMP-activated kinase as a strategy for slowing aging. Med Hypotheses, 63:334–339. 61. Masoro EJ (1985). Nutrition and aging – a current assessment. J Nutr, 115:842–8.
198
F. Xue and K.B. Michels
62. Masoro EJ (1988). Food restriction in rodents: an evaluation of its role in the study of aging. J Gerontol, 43:B59–64. 63. Masoro EJ (1990). Assessment of nutritional components in prolongation of life and health by diet. Proc Soc Exp Biol Med, 193:31–4. 64. Masoro EJ, Shimokawa I, Yu BP (1991). Retardation of the aging processes in rats by food restriction. Ann N Y Acad Sci, 621:337–352. 65. Masoro, EJ (2001). Dietary restriction: An experimental approach to the study of the biology of aging. In: Masoro, EJ. and Austad, SN., (eds.), Handbook of the Biology of Aging, 5th edn, pp. 396–420. Academic Press, San Diego, CA. 66. Masoro EJ (2005). Overview of caloric restriction and ageing. Mech Ageing Dev, 126:913–922. 67. McCarter R, Masoro EJ, Yu BP (1985). Does food restriction retard aging by reducing the metabolic rate? Am J Physiol, 248:E488–490. 68. McCarter RJ, McGee JR (1989). Transient reduction of metabolic rate by food restriction. Am J Physiol, 257:E175–9. 69. Mellemkjaer L, Emborg C, Gridley G et al. (2001). Anorexia nervosa and cancer risk. Cancer Causes Control, 12:173–177. 70. Michels KB, Ekbom A (2004). Caloric restriction and incidence of breast cancer. JAMA, 291:1226–1230. 71. Miura D, Miura Y, Yagasaki K (2003). Hypolipidemic action of dietary resveratrol, a phytoalexin in grapes and red wine, in hepatoma-bearing rats. Life Sci, 73:1393–400. 72. Motta MC, Divecha N, Lemieux M, et al. (2004). Mammalian SIRT1 represses forkhead transcription factors. Cell, 116:551–563. 73. Nilsen TI, Vatten LJ (2001). Adult height and risk of breast cancer: a possible effect of early nutrition. Br J Cancer, 85:959–961. 74. Pariza MW (1987). Dietary fat, calorie restriction, ad libitum feeding, and cancer risk. Nutr Rev, 45:1–7. 75. Pruitt K, Zinn RL, Ohm JE, et al. (2006). Inhibition of SIRT1 reactivates silenced cancer genes without loss of promoter DNA hypermethylation. PLoS Genet, 2:e40. 76. Ragab AS, Van Fleet J, Jankowski B, et al. (2006). Detection and quantitation of resveratrol in tomato fruit (Lycopersicon esculentum Mill.). J Agric Food Chem, 54:7175–9. 77. Robsahm TE, Tretli S (2002). Breast cancer incidence in food- vs non-food producing areas in Norway: possible beneficial effects of World War II. Br J Cancer, 86:362–6. 78. Rogina B, Helfand SL (2004). Sir2 mediates longevity in the fly through a pathway related to calorie restriction. Proc Natl Acad Sci U S A, 101:15998–16003. 79. Roth GS, Ingram DK, Lane MA (1999). Calorie restriction in primates: will it work and how will we know? J Am Geriatr Soc, 47:896–903. 80. Sabatino F, Masoro EJ, McMahan CA, et al. (1991). Assessment of the role of the glucocorticoid system in aging processes and in the action of food restriction. J Gerontol, 46:B171–179. 81. Sacher G.A (1977). Life table modification and life prolongation. In Finch, C.E. and Hayflick, L. (eds.), Handbook of the biology of aging, Van Nostrand Reinhold, pp. 582–638. New York. 82. Sinclair DA (2005). Toward a unified theory of caloric restriction and longevity regulation. Mech Ageing Dev, 126:987–1002. 83. Stervbo U, Vang O, Bonnesen C (2007). A review of the content of the putative chemopreventive phytoalexin resveratrol in red wine. Food Chem, 101:449–457. 84. Stuart JA, Karahalil B, Hogue BA, et al. (2004). Mitochondrial and nuclear DNA base excision repair are affected differently by caloric restriction. FASEB J, 18:595–597. 85. Tannenbaum A (1940). The initiation and growth of tumors. Introduction. Effects of undernutrition. Am J Cancer, 38:335–350. 86. Tannenbaum A (1942). The genesis and growth of tumors. II. Effects of caloric restriction per se. Cancer Res, 2:460–467.
6
Caloric Restriction and Cancer
199
87. Tannenbaum A (1944). The dependence of the genesis of induced skin tumors on the caloric intake during different stage of carcinogenesis. Cancer Res, 4:463–477. 88. Tannenbaum A (1945). The dependence of tumor formation on the composition of the calorie-restricted diet as well as on the degree of restriction. Cancer Res, 5:616–625. 89. Tissenbaum HA, Guarente L (2001). Increased dosage of a sir-2 gene extends lifespan in Caenorhabditis elegans. Nature, 410:227–230. 90. Tretli S, Gaard M (1996). Lifestyle changes during adolescence and risk of breast cancer: an ecologic study of the effect of World War II in Norway. Cancer Causes Control, 7:507–512. 91. Vatten LJ, Kvinnsland S (1990). Body height and risk of breast cancer. A prospective study of 23,831 Norwegian women. Br J Cancer, 61:881–885. 92. Vaziri H, Dessain SK, Ng Eaton E, et al. (2001). hSIR2(SIRT1) functions as an NADdependent p53 deacetylase. Cell, 107:149–159. 93. Weindruch R, Walford RL, Fligiel S, et al. (1986). The retardation of aging in mice by dietary restriction: longevity, cancer, immunity and lifetime energy intake. J Nutr, 116:641–654. 94. Weindruch R, Keenan KP, Carney JM, et al. (2001). Caloric restriction mimetics: metabolic interventions. J Gerontol A Biol Sci Med Sci, 56 Spec No, 1:20–33. 95. Wood JG, Rogina B, Lavu S, et al. (2004). Sirtuin activators mimic caloric restriction and delay ageing in metazoans. Nature, 430:686–689. 96. Xia E, Rao G, Van Remmen H, et al. (1995). Activities of antioxidant enzymes in various tissues of male Fischer 344 rats are altered by food restriction. J Nutr, 125:195–201. 97. Yu BP, Masoro EJ, Murata I, et al. (1982). Life span study of SPF Fischer 344 male rats fed ad libitum or restricted diets: longevity, growth, lean body mass and disease. J Gerontol, 37:130–141. 98. Yu BP, Masoro EJ, McMahan CA (1985). Nutritional influences on aging of Fischer 344 rats: I. Physical, metabolic, and longevity characteristics. J Gerontol, 40:657–670. 99. Yu BP (1996). Aging and oxidative stress: modulation by dietary restriction. Free Radic Biol Med, 21:651–668. 100. Zainal TA, Oberley TD, Allison DB, et al. (2000). Caloric restriction of rhesus monkeys lowers oxidative damage in skeletal muscle. FASEB J, 14:1825–1836. 101. Zhu Z, Jiang W, McGinley J, et al. (2005). Effects of dietary energy repletion and IGF-1 infusion on the inhibition of mammary carcinogenesis by dietary energy restriction. Mol Carcinog, 42:170–176.
Chapter 7
Physical Activity and Cancer Leslie Bernstein, Yani Lu, and Katherine D. Henderson
Abstract This chapter provides a synopsis of conclusions from existing epidemiologic literature on the association between physical activity and cancer risk and considers potential biological mechanisms underlying observed associations. Understanding the relationships between physical activity and risk of cancer will offer clues to the etiologic underpinnings of cancer development that should have important public health implications. Case–control studies and cohort studies, which provide most of the epidemiological evidence on physical activity and cancer risk, usually rely on widely varying self-reported measures of activity. Ideally, activity levels are represented by type (recreational or occupational), duration, frequency, and intensity of activity. The overall health benefits of participating in regular physical activity are widely documented and include reductions in risk of cardiovascular disease, diabetes, osteoporosis, obesity, depression, fatigue, and reduced overall mortality rates. The evidence linking physical activity and cancer risk is quite strong for breast cancer and colon cancer. Evidence that physical activity influences endometrial cancer is increasing. Results are still not confirmed or are conflicting for cancer at other sites. Although it is clear that public health recommendations for appropriate changes in activity levels are needed, we have no exact physical activity prescriptions to give the public. Many questions remain to be answered: What are the ages when physical activity provides its greatest benefit? What types of activity will provide the greatest protection against cancer? What activity patterns (intensity, frequency and/or duration of activity) are optimal? Knowledge about the mechanisms involved in the relationship between physical activity and each cancer type will be important in understanding the etiology of these cancers and in formulating public health recommendations.
L. Bernstein (B) Division of Cancer Etiology, City of Hope National Medical Center, Duarte, CA, USA e-mail:
[email protected]
N.A. Berger (ed.), Cancer and Energy Balance, Epidemiology and Overview, Energy Balance and Cancer 2, DOI 10.1007/978-1-4419-5515-9_7, C Springer Science+Business Media, LLC 2010
201
202
L. Berstein et al.
1 Introduction This chapter provides an overview of the epidemiologic studies on the association between physical activity and risk of cancer at several organ sites. The relationship between energy balance, prognosis, and survivorship is addressed in Chapter 8. Because the literature regarding the epidemiology of physical activity and cancer risk is extensive, this is not a comprehensive review, but rather a synopsis of conclusions available from comprehensive reports [1–3] and meta-analyses, highlighting key results from several important research studies. Potential biological mechanisms explaining observed associations are considered, and areas for future research are discussed. Few established risk factors for cancer are modifiable. Physical activity, particularly participation in recreational forms of exercise activity, offers one potential lifestyle modification that may impact risk for cancer at several organ sites. Understanding the relationships between physical activity and risk of these cancers will offer clues to the etiologic underpinnings of cancer development and will have important public health implications. An emphasis on physical activity began in 1956 when President Eisenhower established the President’s Council on Youth Fitness, a national campaign emphasizing physical fitness objectives in physical education. This was expanded during the Kennedy administration in 1960 to include adult fitness, military fitness, and community recreation. Over the years a series of guidelines on the optimal levels of physical activity to be performed by children, adolescents, and adults has been published by several organizations. The most recent recommendation of the American College of Sports Medicine indicates that between 150 and 300 minutes of moderate intensity physical activity is needed to prevent weight gain and promote modest weight loss and health [4]. Interest in whether physical activity impacts cancer risk began in the early 1980s. One of the earliest studies used cancer registry data to determine whether a man’s occupational physical activity level as reported at diagnosis influenced his colon cancer risk [5]. Although the measure of occupational activity used was crude and not clearly temporally relevant, this study showed that men working in sedentary occupations had greater colon cancer risk than men whose jobs required substantial physical activity. In the ensuing years, physical activity has been studied in far greater detail in relation to many types of cancer by investigators throughout the world as it offers a workable approach for risk modification and ultimately, one hopes, cancer prevention [1, 6]. Physical activity is defined as any movement of the body that results in energy expenditure [7]. In this review, we consider studies of recreational physical activity (exercise activity) or occupational activity (including household activity) and cancer risk. In assessing how different types of activity influence risk, it is important to consider not only duration and frequency of activity, but also the intensity of activity that is being performed. For example, occupational activity may occur over a longer period of time than recreational activity, but occupational activity might require less
7
Physical Activity and Cancer
203
energy expenditure per hour than bouts of vigorous or moderate recreational physical activity [8]. The distinction between recreational and occupational activity is also important because occupational physical activity, with increasing mechanization and technological advances, has decreased, particularly in developed areas of the world, leading to an overall decrease in energy expenditure through physical activity. Case–control studies and cohort studies provide most of the epidemiological evidence on physical activity and cancer risk. In these studies, activity levels are usually self-reported. These measures vary widely. They may be extremely comprehensive, obtaining detailed lifelong diaries of physical activity that represent lifetime histories of activity, they may be based on questionnaire items about activities at defined ages or time points in life, or they may only include questions on current or recent usual activity. Ideally activity levels are represented by objective measures of type (recreational or occupational), duration, frequency, and intensity of activity. While some studies have measured overall activity, others have asked respondents to classify their own activity as vigorous or moderate and report their extent of participation in each. Some studies have assessed sedentary time, representing hours of sitting per day, in addition to time spent in exercise or other physical activities [9, 10]. More comprehensive assessments take a lifetime or long-term history of individual activities using carefully designed in-person interviews which make use of memory probes by creating lifetime calendars of key life events [11]. One challenge regarding collection of information on physical activities in epidemiologic studies is the lack of a “gold standard” measure for the complex behavior of physical activity. Even in the best-designed studies (i.e., those using in-person interviews or those collecting information prior to disease onset), random errors in measurement of physical activity may lead to non-differential misclassification of exposure with respect to outcome status and thus an underestimation of the association between physical activity and cancer risk. Meta-analyses use a statistical approach to combine effect estimates from a series of studies addressing a research question to provide one summary effect estimate and its confidence interval. When interpreting results from meta-analyses, one must consider the myriad of methods for collecting physical activity data across studies. Meta-analyses often lack standardization with respect to the comparison group and generally present a comparison of the “most” active to the “least” active in a study. These categories will differ by study even if the study data collection methods are similar, unless a standard set of definitions is used for “most” and “least” active. In these studies, physical activity may represent a mixture of current activity, recent activity, past activity at a certain age, activity over a particular age range, or lifetime activity; and activity may be defined as sports activities, occupational activities, or other activities such as gardening. Many epidemiologic studies of physical activity convert information into metabolic equivalents (METs) of energy expenditure using an estimate of energy expenditure per hour for each activity [12] to summarize intensity, frequency, and duration across all physical activities. Such measures are not commonly applied in meta-analyses. Although meta-analyses provide some insight into the potential for physical activity to influence cancer risk, the summary
204
L. Berstein et al.
relative risk estimates they provide are limited in detail and are somewhat difficult to interpret or use in making public health recommendations. The overall health benefits of participating in regular physical activity are widely documented and include reductions in risk of cardiovascular disease, diabetes, osteoporosis, obesity, depression and fatigue, and reduced overall mortality rates [13]. In 2001, the International Agency for Research on Cancer convened a committee to assess the relationship between cancer and two modifiable cancer risk factors, inactivity and obesity, and to determine the strength of the evidence by cancer site [1]. Evidence that a physically active lifestyle lowers colon cancer risk was considered to be the strongest and was classified as “convincing.” Evidence for breast cancer was also strong; a lower breast cancer risk, in the range of 20–40%, was reported for the most physically active versus the least physically active women. This review also considered the available evidence for other cancer sites and for these cancers, concluding that the current level of evidence was either limited or conflicting. This chapter focuses on the association between physical activity and risk of breast cancer, colon cancer, endometrial cancer, prostate cancer, lung cancer, and ovarian cancer and provides a short summary of available evidence for other, less well-studied cancer sites.
2 Breast Cancer Results from case–control studies [1, 14–18] and cohort studies [1, 19–24] have shown that invasive breast cancer risk is reduced by 13–50% among physically active women. One of the earliest studies, a case–control study of women 40 years or younger, showed a dramatic reduction in risk (approximately 50%) among women who averaged about 4 hours of activity per week during their reproductive years [14]. Similarly, among postmenopausal women, those with higher amounts of recreational physical activity during their lifetimes have lower breast cancer risk [15, 16]. These results have been confirmed in more than 50 studies and have been observed in different demographic subgroups of the population such as Asian American women [17] and African American women [18]. A study that focused solely on in situ breast cancer showed similar reductions in risk [25]. The Women’s Health Initiative Observational Study and the American Cancer Society Cancer Prevention Study II (CPS II) Nutrition Cohort have reported 22% and 29% reduced risk of breast cancer, respectively [19, 20]. In the Nurses’ Health Study cohort, risk of breast cancer was 18% lower among women engaging in moderate or vigorous activity for at least 7 hours a week compared to women that exercised less than 1 hour per week as adults [26]. For the California Teachers Study cohort, Dallal and colleagues reported that invasive breast cancer risk was inversely associated with long-term vigorous activity as was risk of in situ breast cancer [22]. Participation in activities like aerobics, running, distance swimming, and cycling on hills was accumulated from high school through age 54 years in this study. The protective effect of long-term vigorous activity on risk of invasive breast cancer was more
7
Physical Activity and Cancer
205
pronounced in younger women (< 55 years). Both vigorous and moderate longterm recreational activities were associated with reduced risk of estrogen receptor (ER)-negative but not ER-positive invasive breast cancer. The California Teachers Study results for receptor status were confirmed by the National Institutes of Health (NIH)-American Association of Retired Persons (AARP) Diet and Health Study which used information on frequency of moderate-to-vigorous activity each week at the time women were recruited into the study [27]. Despite results from these two large studies, receptor status-specific results have not been consistent across all studies. One of the earliest reports, based on two case–control studies from Los Angeles, showed greater reductions for ER-negative, progesterone receptor (PR)-negative breast cancer compared to ER-positive, PR-positive breast cancer, but reductions in risk for the two subtypes were not statistically significantly different [28]. A recent case–control study from Germany found that the breast cancer– physical activity association was restricted to ER-positive, PR-positive breast cancer [29], whereas, in the Breast Cancer Detection Demonstration Project Follow-Up Study, which showed an overall inverse association among postmenopausal women participating in vigorous activity, results did not differ by hormone receptor status [24]. Not all studies have found that recreational physical activity impacts breast cancer risk. In the European Prospective Investigation into Cancer and Nutrition (EPIC) study [23], increased physical activity in the form of household activity (highest vs. lowest quartile) was associated with a reduction in breast cancer risk among postmenopausal and premenopausal women, but neither recreational activity nor occupational activity was significantly associated with breast cancer risk. A meta-analysis of 19 cohort and 29 case–control studies published prior to publication of the EPIC, California Teachers, and NIH-AARP studies has provided strong evidence for an inverse association between physical activity and postmenopausal breast cancer [30]. For premenopausal breast cancer the evidence was considered weaker. The authors reported that evidence for a dose–response relationship, on the order of a 6% decrease in risk for each additional hour of physical activity per week, was observed in approximately half of the “higher quality” studies that reported a decreased risk. The totality of results for breast cancer has begun to clarify the protective role of physical activity, suggesting that long-term vigorous activity is required for the effect to be apparent. The issue of whether physical activity affects risk of ER-negative breast cancer is clearly important because current chemopreventive approaches are not effective against ER-negative breast cancer.
3 Colon Cancer The epidemiological literature suggests that increased physical activity is protective for colon cancer [1, 31]. In early studies, this effect was observed more consistently in men than in women. In cohort studies, a greater benefit among men is still observed, although results for case–control studies indicate similar exercise benefits
206
L. Berstein et al.
for colon cancer among men and women [31]. Risk reductions from a meta-analysis comparing individuals in the highest physical activity category to the lowest category found a 31% reduction in colon cancer risk for case–control studies and a 17% reduction in risk for cohort studies [31]. Among men, reductions in colon cancer risk have been associated with both occupational and leisure-time physical activity [1]. One factor that might account for weaker results of studies of physical activity on colon cancer risk among older women is use of hormone therapy, which has been shown to reduce colon cancer risk [32]. This possibility has been investigated in the California Teachers Study. Mai and colleagues [33] reported that combined lifetime moderate and vigorous leisure-time physical activity was modestly associated with decreased risk of colon cancer. Women who exercised at least 4 hours per week during their reproductive years had a 25% lower risk of colon cancer relative to women who exercised no more than 30 minutes per week. Importantly, among postmenopausal women, those who had never used hormone therapy experienced a substantial 46% decrease in colon cancer risk if they averaged at least 4 hours of exercise per week, whereas those who had used hormone therapy experienced no benefit from exercise, but retained a benefit from having used hormone therapy that was comparable to 4 hours of activity per week. It has been argued that distal and proximal colon cancers have distinct etiologies [34]. Several studies of physical activity and colon cancer have examined risk by anatomic subsite, but results are inconsistent with regard to whether the association is stronger for distal tumors or for proximal tumors [35–39]. In contrast to colon cancer, nearly all studies have failed to show a relationship between physical activity and rectal cancer risk [1]. However, the NIH-AARP Diet and Health Study shows a modest reduction in rectal cancer risk for men but not for women after 6.9 years of follow-up [39]. Sedentary behavior itself may be associated with colon cancer risk; in the NIHAARP Diet and Health Study cohort, men who spent at least 9 waking hours per day watching television had 56% greater risk of colon cancer than men who spent less than 3 hours per day in this sedentary behavior [39]. The questions of what type and what amount of exercise to recommend as a public health intervention are still ill defined for colon cancer. Based on the NIH-AARP Diet and Health Study results, sedentary behavior and physical activity appear to be independent predictors of colon cancer risk [39], suggesting that reduction of sitting time and increasing moderate-to-vigorous physical activity are both important. Few studies have evaluated different types of activities; several have looked at whether walking alone is sufficient to lower risk [38, 40], but results are mixed.
4 Endometrial Cancer The evidence showing that regular physical activity lowers endometrial cancer risk is accumulating, but is not currently considered convincing. Results supporting this association [1, 10, 41–44] are not as definitive as those for obesity and colon cancer
7
Physical Activity and Cancer
207
[1, 2] or for physical activity and breast cancer (see Section 2). Studies to date have suggested that risk of endometrial cancer is decreased 20–40% in women who are in the highest category of physical activity [1, 41], with a median relative risk of 0.73 overall and 0.70 among studies that have adjusted for body mass index in their statistical analyses. The lack of change in relative risk estimates, upon adjustment for body mass index, suggests that physical activity is an independent risk factor for endometrial cancer. Although physical activity is associated with decreased risk of endometrial cancer in both normal weight and obese women, two recent studies have suggested that this association is more pronounced for obese women [10, 44]. Three recent studies have also reported an increased risk of endometrial cancer in sedentary women [10, 45, 46]. Two meta-analyses of the physical activity–endometrial cancer association identify some inconsistency in dose–response relationships, documenting the importance of differences in activity type (total, occupational, household, recreational) and intensity [41, 42]. Little evidence exists on how long-term or lifetime physical activity and activity patterns in different life periods influence endometrial cancer risk; some studies have suggested that recent activity and long-term activity may be more important than activity in the distant past [42].
5 Prostate Cancer More than 20 studies have assessed the potential association between physical activity and prostate cancer [1]. Regardless of the varied methods, population bases, and sample sizes used in these studies, the majority have suggested a modest reduction in risk with increasing level of physical activity [1]. Exercise also appears to lower risk for benign prostatic hyperplasia [47, 48]. In a review of the literature, Friedenreich and Orenstein concluded that prostate cancer risk is reduced 10–30% when comparing the most active men to the least active men; further, they suggested that it may be high levels of physical activity earlier in life that are most relevant to this disease [49]. Physical activity during adolescence and overall lifetime vigorous activity were modestly related to lower prostate cancer risk in a population-based case–control study of advanced prostate cancer (stage T2 or greater) conducted in Canada [50]. Cohort studies, which have essentially observed no association between adult levels of physical activity and overall prostate cancer risk, have shown reduced risk for advanced disease [51–53] and for fatal prostate cancer [51, 53] with high levels of physical activity, suggesting that regular vigorous activity may impact progression of prostate cancer. It is important to consider several caveats regarding these results including small numbers of exposed cases [51] and the potential for confounding by prostatespecific antigen (PSA) testing [52, 53]. One would expect that PSA testing should increase the incidence of prostate cancer overall, but decrease the incidence of tumors of advanced stage and high Gleason score. In a study based on the American
208
L. Berstein et al.
Cancer Society CPS II cohort, Patel and colleagues [52] reported that a history of PSA testing was more common among physically active men than among inactive men. No association between physical activity and risk of aggressive prostate cancer was reported in another cohort study [54]; however, investigators conducting this study reported a statistically significant interaction between physical activity and body mass index on prostate cancer risk, with the protective effect of physical activity limited to normal weight men. This result suggests that overall energy balance may play a role in prostate cancer etiology.
6 Lung Cancer The existence of a relationship between physical activity and lung cancer risk is controversial. Physical activity may be protective for lung cancer, yet this effect is not considered well established [1]. A meta-analysis of nine studies published between 1989 and 2003 reported a 13% decreased risk for lung cancer associated with moderate recreational physical activity and a 30% decreased risk associated with vigorous activity [55]. The impact appeared to be slightly stronger among women than among men. Although these effects have been observed among smokers, and after considering measures of pack-years smoked, it is still possible that controls for confounding factors by smoking status and smoking intensity was incomplete in these studies, and that the lower risk of lung cancer reflects unmeasured differences in smoking habits. If there is a causal relationship between physical activity and lung cancer, we would expect to see the protective effect of physical activity among never smokers, yet studies to date have not shown such an association. In the NIH-AARP Diet and Health Study [56], a cohort study with 6,745 lung carcinoma patients identified during an average of 7.2 years of follow-up, physical activity at the time of study initiation was inversely associated with lung cancer risk among former and current smokers, but was not associated with risk of lung cancer overall or with risk of any individual histologic subtype among never smokers. Further, among current or past smokers, increasing physical activity was associated with decreasing risk of adenocarcinomas, but not of other cell types. These results are consistent with those from two other studies presenting data stratified by smoking status [57, 58], which appeared after the meta-analysis publication. Thus the protective effect of physical activity observed among smokers may be due to residual confounding by cigarette smoking.
7 Ovarian Cancer The literature on risk of ovarian cancer in relation to physical activity has been inconclusive. More than 18 studies have assessed the impact of physical activity on
7
Physical Activity and Cancer
209
risk [1, 9, 59, 60]. Existing studies have shown reductions in risk of ovarian cancer with increasing activity ranging from 10 to 67% or no statistically significant effect. In a meta-analysis of 12 studies, ovarian cancer risk was 21% lower in case–control studies and 19% lower in cohort studies among women who exercised the most compared to those with the least amount of exercise, suggesting a modest inverse association [60]. The results among postmenopausal women from the American Cancer Society CPS II cohort showed no evidence that activity was associated with ovarian cancer risk; but in this study, a history of sedentary behavior (sitting ≥ 6 hours a day versus < 3 hours a day) was associated with more than a 50% increase in risk [9]. A similar positive association between sedentary hours and ovarian cancer risk has been reported in one case–control study [61]. Three cohort studies published after the meta-analysis, one in which most women were premenopausal [62] and two in which most women were postmenopausal [63, 64], have reported no association between physical activity and ovarian cancer risk.
8 Other Cancers Limited data exist on the relationship between physical activity and risk of other types of cancer; but, given the growing literature for cancers of the breast, colon, endometrium, ovary, prostate, and lung, physical activity is now being evaluated in relation to several other cancers.
8.1 Gastric Cancer The risk of gastric cancers occurring over 18 years of follow-up in a populationbased cohort of residents in Nord-Trondelag County in Norway was examined in relation to physical activity [65]. When participants were classified according to their overall activity, taking frequency, intensity, and duration of activity into account, risk for gastric cancer declined with increasing level of activity; this reduction in risk was most apparent for non-cardia gastric cancers where individuals with moderate and high levels of activity had one-half the risk of those with no activity. Similarly, in the NIH-AARP Diet and Health Study cohort, risk of distal gastric adenocarcinomas was more than 30% lower among cohort members who reported participating in any level of physical activity at cohort initiation [66]. One interpretation of these results is that protection results from minimal activity; another is that the results are due to residual confounding by a healthy lifestyle among active individuals. It is also possible that some subclinical disease process results in inactivity among those at high risk for distal gastric cancer and results in an apparent marked reduction in risk for active individuals. Although the NIH-AARP Diet and Health Study showed an impact of physical activity on risk of adenocarcinomas of the distal stomach and esophagus (see Section 8.2), no impact on risk of gastric cardia adenocarcinomas was observed [66]. A modest, but not statistically significant,
210
L. Berstein et al.
decreased risk of both distal gastric cancer and gastric cardia cancer was observed in a case–control study that assessed lifetime occupational activity in relation to gastric cancer [67].
8.2 Esophageal Cancer One case–control study and one cohort study have examined risk of adenocarcinoma of the esophagus in relation to physical activity [66, 67]. In the case–control study, the activity measure was based on lifetime occupational activity with occupations classified as involving vigorous, moderate or sedentary activity [67]. Lifetime occupational activity was modestly related to risk of adenocarcinoma of the esophagus. A more pertinent measure, average annual level of occupational activity before age 65 years, was more strongly related to risk. Study participants in the highest occupational activity group experienced a roughly 40% reduction in risk of esophageal adenocarcinoma compared with those in the lowest occupational activity category. Results from the NIH-AARP Diet and Health Study are consistent with those from the case–control study and showed that participants who engaged in physical activity at least five times per week at cohort entry had a 25% lower risk of esophageal adenocarcinoma [66]. No beneficial effect was noted for squamous cell esophageal cancer [66].
8.3 Renal Cell Cancer At least six cohort studies have explored the relationship between renal cell cancer and physical activity, in part because of the known deleterious effects of high body mass index and hypertension on the risk of renal cell cancer [1, 68]. Despite this rationale for a potential association, no association between physical activity and renal cell cancer has been established [1]. In a large Canadian case–control study, physical activity was not associated with renal cell cancer among men or women [69]. In contrast, in the Hawaii and Los Angeles Multiethnic Cohort, physical activity was associated with renal cell cancer risk among women, but no association was found for men [68]. As pointed out by the authors, given the marked impact of obesity on risk in women, this could be an artifact of inadequate control of the confounding effects of adiposity.
8.4 Pancreatic Cancer Pancreatic cancer is generally diagnosed at an advanced stage. Therefore, case– control studies are likely to exclude patients who die soon after diagnosis or are too ill to participate. Prospective studies require long duration of follow-up to accrue sufficient numbers of pancreatic cancer cases to examine risk factors and rule out
7
Physical Activity and Cancer
211
changes in behavior due to early symptoms in the first years of follow-up. Despite this, at least 12 prospective studies have examined risk of pancreatic cancer in relation to physical activity (summarized in [70]). Given the mixed results from these studies, current evidence does not support an association between physical activity and risk of pancreatic cancer.
9 Mechanisms Mechanisms by which physical activity can influence cancer risk, including those discussed in Chapter 5 of the current volume, may vary by cancer site. In brief, changes in estrogen and progesterone exposure over a woman’s lifetime are the main biological mechanisms considered in linking physical activity to risk of breast cancer. For example, evidence has shown that exercise influences exposure to estrogen and progesterone during a woman’s reproductive years by lowering body fat in youth [71] which can delay the age when first menses occurs [72]. Physical activity can also disrupt menstrual cyclicity which manifests across a continuum ranging from luteal phase defects to anovulation to secondary amenorrhea [73]. This may reflect an imbalance between energy expenditure and energy intake. In adolescence, even moderate levels of physical activity increase the likelihood of experiencing anovulatory menstrual cycles [74]. Having fewer ovulatory menstrual cycles is likely to reduce a woman’s cumulative lifetime exposure to endogenous ovarian hormones (estradiol and progesterone) [75]. Physical activity can reduce weight and weight gain in adulthood, both of which are associated with greater risk of breast cancer after menopause. In the postmenopausal woman, adipose tissue is the primary source of endogenous estrogen via aromatization of androstenedione to estrone [76]; thus, heavier postmenopausal women have higher levels of circulating estrogen than women with less adipose tissue, and these higher levels of estrogen likely increase breast cancer risk. Physical activity can also have a direct impact on circulating estrogen levels among postmenopausal women [77]. Another mechanism by which exercise might influence breast cancer risk is through regulation of melatonin levels; Knight and colleagues [78] have shown that exercise increases melatonin levels in women, which would be expected to lower breast cancer risk. Exposure to unopposed estrogen is the major cause of endometrial cancer [79]. The likely pathway by which physical activity influences endometrial cancer risk is by altering endogenous hormone profiles. Physical activity may counter the proliferative effects of estrogen either directly or by influencing body weight or weight gain. By lowering body fat after menopause, when ovarian production of estradiol and progesterone ceases, women have little or no exposure to androstenedione-derived estrogen. Studies have linked elevated insulin levels and diabetes to endometrial cancer risk [79]. Because physical activity also influences insulin sensitivity, this may also
212
L. Berstein et al.
explain a potential association between physical activity and endometrial cancer risk. The impact of physical activity on circulating hormone profiles may also influence ovarian cancer risk and prostate cancer risk. The mechanisms cited above for breast cancer and endometrial cancer may affect ovarian cancer risk; however, risk may also involve changes in gonadotropin exposure. For prostate cancer, proposed mechanisms include paradoxical effects of testosterone on low-grade versus more advanced prostate cancer and alterations in insulin and circulating insulin-like growth factor (IGF)-I profiles [80]. It has been suggested that physical activity reduces colon cancer risk by stimulating stool transit through the colon, thereby decreasing the exposure of the colonic mucosa to carcinogens in the stool [81]. Physical activity may alter the control of cellular proliferation and apoptosis. A year-long randomized controlled clinical trial using a home-based exercise intervention confirmed this for men, showing that moderate to vigorous intensity exercise (6 hours/week) increased the levels of the pro-apoptotic protein, BAX, at the base of colonic crypts; but the same was not true for women [82]. In fact, results for women were in the opposite direction [82]. Colon mucosal prostaglandin concentrations did differ between the exercise and control arms of this clinical trial [83]. The insulin and IGF pathways have been proposed to mediate associations with colon cancer risk [84]. High insulin levels may promote cell proliferation and tumor growth in the colon [84] and may suppress expression of IGF-binding proteins 1 and 2 leading to increased bioavailable IGF-I levels [85], which have been associated with higher risk of colon cancer [86]. Inflammation is a pathway that may mediate the relationship between physical activity and a number of cancers, and in particular colon cancer and breast cancer. Chronic colon conditions, such as inflammatory bowel disease, are associated with increased incidence of colon cancer [84]. Further, use of aspirin and other non-steroidal anti-inflammatory drugs reduce colon cancer risk [84]. Recent evidence suggests that physical activity may reduce inflammation [87]. Muscle-derived interleukin-6 (IL-6) is the major cytokine that is affected by acute vigorous and moderate exercise, increasing exponentially in response to exercise. Although IL-6 levels decline following exercise, other anti-inflammatory cytokines induced by IL6 increase. IL-6 has inhibitory effects on proinflammatory cytokine proteins [88]. This evidence of the anti-inflammatory impact of physical activity is supported by studies showing that an individual’s level of physical activity (light, moderate, or vigorous) is inversely associated with level of circulating C-reactive protein, a marker of inflammation [89–91].
10 Summary In summary, the evidence regarding physical activity and cancer risk is quite strong for breast cancer and colon cancer. Evidence that physical activity influences endometrial cancer risk is increasing. Results are still mixed for ovarian cancer and
7
Physical Activity and Cancer
213
those for prostate cancer may be influenced by PSA testing. The association with lung cancer requires careful assessment of the possibility that smoking patterns may account for differences in activity levels. Although it is clear that recommendations for appropriate changes in activity levels are important public health messages, we still have no exact physical activity prescriptions to give the public. Many questions remain to be answered: What are the ages when physical activity provides its greatest benefit? What types of activity will provide the greatest protection against cancer? What activity patterns (intensity, frequency and/or duration of activity) are optimal? What is the role of the built environment in affecting physical activity levels? Knowledge about the mechanisms involved in the relationship between physical activity and each cancer type will be important in understanding the etiology of these cancers and formulating public health recommendations.
References 1. Vainio H, Bianchini F (eds.) (2000). IARC handbooks of cancer prevention. Weight control and physical activity, Vol. 6. IARC Press, Lyon, France 2. World Cancer Research fund / American institute for Cancer Research (2007). Food, Nutrition, Physical Activity, and the Prevention of Cancer: a Global Perspective. American Institute for Cancer Research, Washington, DC 3. Physical Activity Guidelines Advisory Committee (2008). Physical Activity Guidelines Advisory Committee Report. U.S. Department of Health and Human Services, Washington, DC 4. Donnelly JE, Blair SN, Jakicic JM, Manore MM, Rankin JW, Smith BK (2009). Appropriate physical activity intervention strategies for weight loss and prevention of weight regain for adults. Med Sci Sports Exerc, 41:459–71. 5. Garabrant DH, Peters JM, Mack TM, Bernstein L (1984). Job activity and colon cancer risk. Am J Epidemiol, 119:1005–1014. 6. Friedenreich C (2001). Physical activity and cancer prevention. From observational to intervention research. Cancer Epidemiol Biomarkers Prev, 10:287–301. 7. http://www.cdc.gov/nccdphp/dnpa/physical/terms/index.htm 8. Kohl HW, LaPorte RE, Blair SN (1988). Physical activity and cancer. An epidemiological perspective. Sports Med, 6:222–237. 9. Patel AV, Rodriguez C, Pavluck AL, Thun MJ, Calle EE (2006). Recreational physical activity and sedentary behavior in relation to ovarian cancer risk in a large cohort of US women. Am J Epidemiol, 163:709–716. 10. Gierach GL, Chang SC, Brinton LA, et al. (2009). Physical activity, sedentary behavior, and endometrial cancer risk in the NIH-AARP diet and health study. Int J Cancer, 124:2139–2147. 11. Friedenreich CM, Courneya KS, Bryant HE (1998). The lifetime total physical activity questionnaire: development and reliability. Med Sci Sports Exerc, 30:266–274. 12. Ainsworth BE, Haskell WL, Whitt MC, et al. (2000). Compendium of physical activities: an update of activity codes and MET intensities. Med Sci Sports Exerc, 32:S498–S504. 13. National Center for Chronic Disease Prevention and Health Promotion (1996). Physical Activity and Health: A Report of the Surgeon General. National Center for Chronic Disease Prevention and Health Promotion, Atlanta, GA 14. Bernstein L, Henderson BE, Hanisch R, Sullivan-Halley J, Ross RK (1994). Physical exercise and reduced risk of breast cancer in young women. J Natl Cancer Inst, 86:1403–1408. 15. Carpenter CL, Ross RK, Paganini-Hill A, Bernstein L (1999). Lifetime exercise activity and breast cancer risk among post-menopausal women. Br J Cancer, 80:1852–1858.
214
L. Berstein et al.
16. Carpenter C, Ross RK, Paganini-Hill A, Bernstein L (2003). Effect of family history, obesity, and exercise on breast cancer risk among postmenopausal women. Int J Cancer, 106:96–102. 17. Yang D, Bernstein L, Wu AH (2003). Physical activity and breast cancer risk among AsianAmerican women in Los Angeles. Cancer, 97:2565–2575. 18. Bernstein L, Patel AV, Ursin G, et al. (2005). Lifetime recreational exercise activity and breast cancer risk among black and white women. J Natl Cancer Inst, 97:1671–1679. 19. McTiernan A, Kooperberg C, White E, et al. (2003). Recreational physical activity and the risk of breast cancer in postmenopausal women: the women’s health initiative cohort study. JAMA, 290:1331–1336. 20. Patel AV, Calle EE, Bernstein L, Wu AH, Thun MJ (2003). Recreational physical activity and risk of postmenopausal breast cancer in a large cohort of US women. Canc Causes Contr, 14:519–529. 21. Dirx MJ, Voorrips LE, Goldbohm RA, van den Brandt PA (2001). Baseline recreational physical activity, history of sports participation, and postmenopausal breast carcinoma risk in the Netherlands cohort study. Cancer, 92(6):1638–1649. 22. Dallal CM, Sullivan-Halley J, Ross RK, et al. (2007). Long-term recreational physical activity and risk of invasive and in situ breast cancer. Arch Int Med, 167:408–415. 23. Lahmann P, Friedenreich C, Schuit A, et al. (2007) Physical activity and breast cancer risk: the European prospective investigation into cancer and nutrition. Cancer Epidemiol Biomarkers Prev, 16:36–42. 24. Leitzman MF, Moore SC, Peters TM, et al. (2008). Prospective study of physical activity and risk of postmenopausal breast cancer. Breast Cancer Res, 10:R92 (http://breast-cancerresearch.com/content/10/5/R92) 25. Patel AV, Press MF, Meeske K, Calle EE, Bernstein L (2003). Lifetime recreational physical activity and risk of breast carcinoma in situ. Cancer, 98:2161–2169. 26. Rockhill B, Willett WC, Hunter DJ, Manson JE, Hankinson SE, Colditz GA (1999). A prospective study of recreational physical activity and breast cancer risk. Arch Int Med, 159:2290–2296. 27. Peters TM, Schatzkin A, Gierach GL, et al. (2009). Physical activity and postmenopausal breast cancer risk in the NIH-AARP diet and health study. Cancer Epidemiol Biomarkers Prev, 18:289–296. 28. Enger SM, Ross RK, Paganini-Hill A, Carpenter CL, Bernstein L (2000). Body size, physical activity, and breast cancer hormone receptor status: results from two case–control studies. Cancer Epidemiol Biomarkers Prev, 9:681–687. 29. Schmidt ME, Steindorf K, Mutschelknauss E, et al. (2008). Physical activity and postmenopausal breast cancer: effect modification by breast cancer subtypes and effective periods in life. Cancer Epidemiol Biomarkers Prev, 17:3402–3410. 30. Monninkhof EM, Elias SG, Vlems FA, et al. (2007). Physical activity and breast cancer: a systematic review. Epidemiol, 18:137–157. 31. Wolin KY, Yan Y, Colditz GA, Lee IM (2009). Physical activity and colon cancer prevention: a meta-analysis. Br J Cancer, 100:611–616. 32. Chlebowski RT, Wactawski-Wende J, Ritenbaugh C, et al. (2004). Estrogen plus progestin and colorectal cancer in postmenopausal women. N Engl J Med, 350:991–1004. 33. Mai PL, Sullivan-Halley J, Ursin G, et al. (2007). Physical activity and colon cancer risk among women in the California Teachers Study. Cancer Epidemiol Biomarkers Prev, 16: 517–525. 34. Iacopetta B (2002). Are there two sides to colorectal cancer? Int J Cancer, 101:403–408. 35. Friedenreich C, Norat T, Steindorf K, et al. (2006). Physical activity and risk of colon and rectal cancers: the European prospective investigation into cancer and nutrition. Cancer Epidemiol Biomarkers Prev, 15:2398–2407. 36. Moradi T, Gridley G, Bjork J, et al. (2008). Occupational physical activity and risk for cancer of the colon and rectum in Sweden among men and women by anatomic subsite. Eur J Cancer Prev, 17:201–208.
7
Physical Activity and Cancer
215
37. Nilsen TI, Romundstad PR, Petersen H, Gunnell D, Vatten LJ (2008). Recreational physical activity and cancer risk in subsites of the colon (the Nord-Trondelag Health Study). Cancer Epidemiol Biomarkers Prev, 17:183–188. 38. Wolin KY, Lee IM, Colditz GA, Glynn RJ, Fuchs C, Giovannucci E (2007). Leisure-time physical activity patterns and risk of colon cancer in women. Int J Cancer, 121:2776–2781. 39. Howard RA, Freedman DM, Park Y, Hollenbeck A, Schatzkin A, Leitzman MF (2008). Physical activity, sedentary behavior, and the risk of colon and rectal cancer in the NIH-AARP diet and health study. Cancer Causes Control, 19:939–953. 40. Chao A, Connell CJ, Jacobs EJ, et al. (2004). Amount, type, and timing of recreational physical activity in relation to colon and rectal cancer in older adults: the Cancer Prevention Study II Nutrition Cohort. Cancer Epidemiol Biomarkers Prev, 13:2187–2195. 41. Voskuil DW, Monninkhof EM, Elias SG, Vlems FA, van Leeuwen FE, the Task Force Physical Activity and Cancer. (2007). Physical activity and endometrial cancer risk, a systematic review of current evidence. Cancer Epidemiol Biomarkers Prev, 639–648. 42. Cust A, Armstrong B, Friedenreich C, Slimani N, Bauman A. (2007). Physical activity and endometrial cancer risk: a review of the current evidence, biologic mechanisms and the quality of physical activity assessment methods. Cancer Causes Control, 18:243–258. 43. Friedenreich C, Cust A, Lahmann PH, et al. (2007). Physical activity and risk of endometrial cancer: the European prospective investigation into cancer and nutrition. Int J Cancer, 121:347–355. 44. Patel AV, Feigelson HS, Talbot JT, et al. (2008). The role of body weight in the relationship between physical activity and endometrial cancer: results from a large cohort of US women. Int J Cancer, 123:1877–1882. 45. Friberg E, Mantzoros CS, Wolk A (2006). Physical activity and risk of endometrial cancer: a population-based prospective cohort study. Cancer Epidemiol Biomarkers Prev, 15: 2136–2140. 46. Matthews CE, Xu WH, Zheng W, et al. (2005). Physical activity and risk of endometrial cancer: a report from the Shanghai endometrial cancer study. Cancer Epidemiol Biomarkers Prev, 14:779–785. 47. Platz EA, Kawachi I, Rimm EB, et al. (1998). Physical activity and benign prostatic hyperplasia. Arch Int Med, 158:2349–2356. 48. Lagiou A, Samoli E, Georgila C, et al. (2008). Occupational physical activity in relation with prostate cancer and benign prostatic hyperplasia. Eur J Cancer Prev, 17:336–339. 49. Friedenreich CM, Orenstein MR (2002). Physical activity and cancer prevention: etiologic evidence and biological mechanisms. J Nutr, 132:3456S–3464S. 50. Friedenreich CM, McGregor SE, Courneya KS, Angyalfi SJ, Elliott FG (2004). Case–control study of lifetime total physical activity and prostate cancer risk. Am J Epidemiol, 159: 740–749. 51. Giovannucci EL, Liu Y, Leitzman MF, Stampfer MJ, Willett WC (2005). A prospective study of physical activity and incident and fatal prostate cancer. Arch Int Med, 165:1005–1010. 52. Patel AV, Rodriguez C, Jacobs EJ, Solomon L, Thun MJ, Calle EE (2005). Recreational physical activity and risk of prostate cancer in a large cohort of US men. Cancer Epidemiol Biomarkers Prev, 14:275–279. 53. Nilsen TI, Romundstad PR, Vatten LJ (2006). Recreational physical activity and risk of prostate cancer: a prospective population-based study in Norway (the HUNT study). Int J Cancer, 119:2943–2947. 54. Littman AJ, Kristal AR, White E (2006). Recreational physical activity and prostate cancer risk (United States). Cancer Causes Control, 17:831–841. 55. Tardon A, Lee WJ, Delgado-Rodriguez M, et al. (2005). Leisure-time physical activity and lung cancer: a meta-analysis. Cancer Causes Control, 16:389–397. 56. Leitzmann MF, Koebnick C, Abnet CC, et al. (2009). Prospective study of physical activity and lung cancer by histologic type in current, former, and never smokers. Am J Epidemiol, 169:542–553.
216
L. Berstein et al.
57. Kubik A, Zatloukal P, Tomasek L, et al. (2007). Interactions between smoking and other exposures associated with lung cancer risk in women: diet and physical activity. Neoplasma, 54:83–88. 58. Sinner P, Folsom AR, Harnack L, Eberly LE, Schmitz KH. The association of physical activity with lung cancer incidence in a cohort of older women: the Iowa Women’s Health Study. Cancer Epidemiol Biomarkers Prev, 15:2359–2363. 59. Lukanova A, Kaaks R (2005). Endogenous hormones and ovarian cancer: Epidemiology and current hypotheses. Cancer Epidemiol Biomarkers Prev, 14:98–107. 60. Olsen CM, Bain CJ, Jordan SJ, et al. Recreational physical activity and epithelial ovarian cancer: a case–control study, systematic review, and meta-analysis. Cancer Epidemiol Biomarkers Prev, 16:2321–2330. 61. Zhang M, Xie X, Lee AH, Binns CW (2004). Sedentary behaviours and epithelial ovarian cancer risk. Cancer Causes Control, 15:83–89. 62. Weiderpass E, Margolis KL, Sandin S, et al. (2006). Prospective study of physical activity in different periods of life and the risk of ovarian cancer. Int J Cancer, 118:3153–3160. 63. Lahmann PH, Friedenreich C, Schulz M, et al. (2009). Physical activity and ovarian cancer risk: the European prospective investigation into cancer and nutrition. Cancer Epidemiol Biomarkers Prev, 18:351–354. 64. Leitzmann F, Koebnick C, Moore SC, et al. (2008). Prospective study of physical activity and the risk of ovarian cancer. Cancer Causes Control, Dec 31. [Epub ahead of print] 65. Sjödahl K, Jia C, Vatten L, Nilsen T, Hveem K, Lagergren J (2008). Body mass and physical activity and risk of gastric cancer in a population-based cohort study in Norway. 17:135–140. 66. Leitzman MF, koebnick C, Freedman ND, et al. (2009). Physical activity and esophageal and gastric carcinoma in a large prospective study. Am J Prev Med, 36:112–119. 67. Vigen C, Bernstein L, Wu AH. (2006). Occupational physical activity and risk of adenocarcinomas of the esophagus and stomach. Int J Cancer, 118:1004–1009. 68. Setiawan VW, Stram DO, Nomura AMY, Kolonel LN, Henderson BE (2007). Risk factors for renal Cell cancer; The Multiethnic Cohort. Am J Epidemiol, 166:932–940. 69. Pan SY, DesMeules M, Morrison H, Wen SW, Canadian Cancer Registries Epidemiology Research Group (2006). Obesity, high energy intake, lack of physical activity and the risk of kidney cancer. Cancer Epidemiol Biomarkers Prev, 15:2453–2460. 70. Nöthlings U, Wilkens LR, Murphy SP, Hankin JH, Henderson BE, Kolonel LN (2007). Body mass index and physical activity as risk factors for pancreatic cancer: the Multiethnic Cohort Study. Cancer Causes Control, 18:165–175. 71. Goran MI (2000). Energy metabolism and obesity. Med Clinics No Am, 84:347–362. 72. Frisch RE, McArthur JW (1974). Menstrual cycles: fatness as a determinant of minimum weight for height necessary for their maintenance or onset. Science, 185:949–951. 73. Williams NI (2003). Lessons from experimental disruptions of the menstrual cycle in humans and monkeys. Med Sci Sports Exerc, 35:1564–1572. 74. Bernstein L, Ross RK, Lobo RA, Hanisch R, Krailo MD, Henderson BE (1987). The effects of moderate physical activity on menstrual cycle patterns in adolescence: implications for breast cancer prevention. Br J Cancer, 55:681–685. 75. Bernstein L (2002). Epidemiology of endocrine-related risk factors for breast cancer. J Mammary Gland Biol Neopl, 7:3–15. 76. MacDonald PC, Edman CD, Hemsell DL, Porter JC, Siiteri PK (1978). Effect of obesity on conversion of plasma androstenedione to estrone in postmenopausal women with and without endometrial cancer. Am J Obstet Gynecol, 130:448–455. 77. Cauley JA, Gutai JP, Kuller LH, LeDonne D, Powell JG (1989). The Epidemiology of serum sex hormones in postmenopausal women. Am J Epidemiol 129:1120–1131. 78. Knight JA, Thompson S, Raboud JM, Hoffman BR (2005). Light and exercise and melatonin production in women. Am J Epidemiol, 162:1114–1122. 79. Kaaks R, Lukanova A, Kurzer MS (2002). Obesity, endogenous hormones, and endometrial cancer risk: a synthetic review. Cancer Epidemiol Biomarkers Prev, 11:1531–1543.
7
Physical Activity and Cancer
217
80. Rodriguez C, Freedland SJ, Deka A, et al. (2007). Body mass index, weight change, and risk of prostate cancer in the cancer prevention study II nutrition cohort. Cancer Epidemiol Biomarkers Prev, 16:63–69. 81. Hardman AE (2001). Physical activity and cancer risk. Proc Nutr Soc, 60:107–113. 82. Campbell KL, McTiernan A, Li SS, et al. (2007). Effect of a 12-month exercise intervention on the apoptotic regulating proteins Bax and Bcl-2 in colon crypts: a randomized controlled trial. Cancer Epidemiol Biomarkers Prev, 16:1767–1774. 83. Abrahamson PE, King IB, Ulrich CM, et al. (2007). No effect of exercise on colon mucosal prostaglanding concentrations: A 12-month randomized controlled trial. Cancer Epidemiol Biomarkers Prev, 16:2351–2356. 84. Gunter MJ, Leitzman MF (2006). Obesity and colorectal cancer: Epidemiology, mechanisms and candidate genes. J Nutr Biochem, 17:145–156. 85. Calle EE, Kaaks R (2004). Overweight, obesity and cancer: Epidemiological evidence and proposed mechanisms. Nature Rev Cancer, 4:579–591. 86. Ma J, Pollak MN, Giovannucci E, et al. (1999). Prospective study of colorectal cancer risk in men and plasma levels of insulin-like growth factor (IGF)-I and IGF-binding protein-3. J Natl Cancer Inst, 91:620–625. 87. Petersen AMW, Pedersen BK (2006). The role of IL-6 in mediating the anti-inflammatory effects of exercise. J Physiol Pharmacol, 57 (Suppl 10):43–51. 88. Pedersen BK (2006). The anti-inflammatory effect of exercise: its role in diabetes and cardiovascular disease control. Essays Biochem, 42:105–117. 89. Mattusch F, Dufaux B, Heine O, Mertens I, Rost R (2000). Reduction of the plasma concentration of C-reactive protein following nine months of endurance training. Int J Sports Med, 21:21–24. 90. Ford ES (2002). Does exercise reduce inflammation? Physical activity and C-reactive protein among US adults. Epidemiol, 13:561–568. 91. Abramson JL, Vaccarino V (2002). Relationship between physical activity and inflammation among apparently healthy middle-aged and older US adults. Arch Int Med, 162:1286–1292.
Chapter 8
Energy Balance, Cancer Prognosis, and Survivorship Melinda L. Irwin
Abstract An increasing number of men and women are being diagnosed with cancer and many cancer survivors are seeking lifestyle-based approaches to improve survival. Physical activity and diet are modifiable behaviors with a multitude of health benefits, and an increasing number of publications have shown a strong relationship between physical activity, diet, weight, and cancer survival. The purpose of this chapter is to review (1) the evidence supporting the effect of weight, physical activity, and diet on cancer prognosis and survivorship; (2) mechanisms mediating the observed associations between energy balance and improved prognosis; and (3) approaches to favorably changing physical activity, diet, and weight to improve cancer prognosis and survivorship. Since a majority of cancer survivors are overweight, and not participating in recommended levels of physical activity or eating a prudent diet, improving these behaviors has the potential to benefit a large number of cancer survivors.
1 Introduction There has been significant research directed toward improving survival and quality of life in men and women diagnosed with cancer. One of the most common questions cancer survivors ask is “What can I do to improve my survival?” The impact of energy balance, or specifically weight, physical activity and diet, may have on cancer survival is of growing interest to scientists, clinicians, professional organizations, and survivors. Physical activity and diet are modifiable behaviors with a multitude of health benefits, and an increasing number of publications have shown a strong relationship between physical activity, diet, weight, and cancer survival [1–8].
M.L. Irwin (B) Epidemiology and Public Health, Yale School of Public Health, New Haven, CT, USA e-mail:
[email protected]
N.A. Berger (ed.), Cancer and Energy Balance, Epidemiology and Overview, Energy Balance and Cancer 2, DOI 10.1007/978-1-4419-5515-9_8, C Springer Science+Business Media, LLC 2010
219
220
M.L. Irwin
The purpose of this chapter is to review (1) the evidence examining the impact of weight, physical activity, and diet on cancer prognosis and survivorship; (2) mechanisms mediating the observed associations between energy balance and improved prognosis; and (3) approaches to favorably changing physical activity, diet, and weight to improve cancer prognosis and survivorship. Since a majority of cancer survivors are overweight, and not participating in recommended levels of physical activity or eating a prudent diet [9–11], improving these behaviors has the potential to benefit a large number of cancer survivors.
2 Obesity at Initial Diagnosis and Cancer Survival Many observational studies have looked at the relationship between weight at diagnosis and cancer outcomes, and the vast majority of these have demonstrated an increased risk of cancer recurrence and death in men and women who are overweight or obese at the time of cancer diagnosis [5–8]. In a study conducted by the American Cancer Society, obesity in adult men and women was associated with increased mortality from cancers of the colon, breast, endometrium, kidney (renal cell), esophagus (adenocarcinoma), gastric cardia, pancreas, prostate, gallbladder, and liver [8]. Estimates from this study suggest 14% of all cancer deaths in men and 20% of all cancer deaths in women from a range of cancer types are attributable to overweight and obesity. Furthermore, there was a 52 and 88% increase in the risk of all cancer death for men and women, respectively, who were severely obese (BMI ≥ 40 kg/m2 ) compared with men and women who were normal body weight (BMI < 25 kg/m2 ). Recently, two studies observed associations between obesity and increased risk for recurrence and death from colon and prostate cancer. Dignam and colleagues investigated the association between BMI at diagnosis and risk of recurrence, second primary cancer, and mortality in 4288 colon cancer patients [12]. A BMI greater than 35.0 kg/m2 at diagnosis was associated with a 38 and 49% increased risk of recurrence or death, respectively, as compared to a BMI less than 25 kg/m2 . Wright and colleagues examined BMI in relation to prostate cancer mortality in 287,760 men in the NIH-AARP Diet and Health Study [13]. A significant two-fold elevation in prostate cancer mortality was observed in men with BMI levels greater than 35 kg/m2 as compared with men with BMI levels less than 25 kg/m2 . Given that breast cancer is the most frequently diagnosed invasive cancer among women and that rates of obesity are increasing among women and breast cancer survivors, there has been significant research directed toward the relationship between obesity and breast cancer prognosis. A meta-analysis of these observational studies demonstrated a hazard ratio for breast cancer recurrence at 5 years of 1.78 (95% CI 1.5–2.11) and for breast cancer death at 10 years of 1.36 (95% CI 1.19–1.55) for women in higher BMI categories compared with women at lower BMI categories [14]. The findings that obesity is associated with cancer mortality are apparent even after adjustment for stage at diagnosis and adequacy of treatment.
8
Energy Balance, Cancer Prognosis, and Survivorship
221
Associations between obesity and breast cancer prognosis in younger women appear to be even stronger; Daling and colleagues [15] reported that women younger than 45 years of age who had invasive breast cancer and a BMI > 25 kg/m2 were 2.5 times as likely to die of their disease within 5 years of diagnosis compared with women with breast cancer and a BMI < 21 kg/m2 .
3 Weight Gain and Cancer Survival Epidemiological studies have also shown that weight gain after a breast cancer diagnosis is associated with an increased risk for recurrence and death compared with maintaining normal weight after diagnosis [16]. This is especially worrisome given the fact that, especially among women treated for breast cancer, a majority of them gain a significant amount of weight in the year following breast cancer diagnosis, and return to pre-diagnosis weight is rare [9]. Analyses from the Nurses’ Health Study showed that weight gain after diagnosis (∼5–10 lbs) was related to approximately 50% higher rates of breast cancer recurrence and death [16]. The findings were especially apparent in women who never smoked, among women with earlier stage disease or those who were normal weight before diagnosis. While these findings are intriguing, not all studies have observed an association between obesity or weight gain and poor survival. Caan and colleagues did not observe an association between post-diagnosis weight gain and breast cancer recurrence risk in the first 5–7 years post-diagnosis [17]. Meyerhardt and colleagues, using data from the Cancer and Leukemia Group B (CALGB) 89803 study of 1,053 patients who had stage III colon cancer, demonstrated no association between BMI or weight change and survival in colon cancer patients [18]. It is unknown if chemotherapy dose specifications may account for the differences between these studies and the studies showing an increased risk of death with higher BMI and weight gain. Thus, obesity is either associated with poor prognosis or may be associated with receiving inadequate chemotherapy doses. However, Buist and colleagues examined the association between BMI and receipt of appropriate primary tumor therapy and adjuvant therapy in 897 women diagnosed with breast cancer. They found that receipt of appropriate primary therapy and adjuvant therapy was not associated with BMI in women treated for breast cancer, implying that the majority of studies that have shown an association between obesity and poor prognosis may in fact be true [19]. One final concern recently raised by Daniell and colleagues is that being obese, compared with being normal weight, prior to and at cancer diagnosis is associated with earlier tumor metastasis, or more rapid growth of node metastases before diagnosis, as well as differences in hormone receptor status [20]. Thus, these genetic differences in tumors among obese patients may have already influenced the growth of metastic tissue before their initial diagnosis. Therefore, weight loss after diagnosis may not influence prognosis because of the already established genetic alterations. However, without a methodologically strong weight loss trial conducted in overweight and obese cancer survivors, we are unable to definitively
222
M.L. Irwin
know whether weight loss impacts survival or not. Regardless, obesity and weight gain still have adverse effects on risk of other new cancers and overall survival. Specifically, there is evidence that cancer survivors die of non-cancer causes at a higher rate than persons in the general population (deaths being primarily from cardiovascular disease and diabetes) [21]. Therefore, surviving cancer requires not only treating the primary cancer, but also avoiding second cancers for which patients are at increased risk. To improve overall survival, it is critically important for cancer survivors to prevent obesity. One of the primary methods for preventing or treating obesity and weight gain is by eating a healthy diet and increasing physical activity levels. Physical activity in particular has also been presented as a therapeutic strategy to address both the psychological and the physical concerns faced by cancer survivors. In summary, given the strong observational evidence suggesting that being overweight and obese is associated with poor prognosis in breast, colon, and prostate cancers, there is a need to develop clinical trials testing the effect of intentional weight loss upon cancer recurrence and mortality. If, e.g., 10% weight loss is associated with an improved disease-free survival rate, then perhaps more behavior/ lifestyle change programs will be available to cancer survivors. However, even if a benefit of weight loss on disease-free survival is not observed, there are additional benefits of weight loss including reduced therapy-related complications, improved quality of life, and reduced risk of death from other causes. Oncologists are, therefore, encouraged to counsel patients on maintaining a healthy weight via increased physical activity and eating a prudent diet high in fruits and vegetables, whole grains, and poultry.
4 Physical Activity and Cancer Survival Numerous observational studies have recently been published demonstrating that participation in moderate-intensity recreational physical activity after diagnosis is associated with improved survival in women who develop breast cancer [1–4]. These studies have demonstrated an approximate 50% reduction in the risk of total deaths and risk of breast cancer deaths in women who are physically active after breast cancer diagnosis compared with women reporting no recreational physical activity. These studies also showed that the decreased risk of death associated with physical activity was observed in pre- and post-menopausal women, overweight and normal weight women, and women with stage I–III disease. While any amount of recreational physical activity performed after diagnosis has been associated with a decreased risk of death, the maximal benefit occurred in women who performed the equivalent of brisk walking 3 hrs per week. The type of physical activity assessed in these studies was sports/recreational physical activity; however, one study showed similar yet slightly attenuated associations for any moderate-intensity physical activity (e.g., heavy household work, gardening, occupational activities) [2]. While most of the studies included samples of breast cancer survivors that were primarily non-Hispanic white, one study showed a similar association in African American and Hispanic women [2]. Furthermore, there is
8
Energy Balance, Cancer Prognosis, and Survivorship
223
little reason to believe that the biological mechanisms by which physical activity could improve survival would differ by race/ethnicity. Given that women who are more physically active after diagnosis may have been similarly active before diagnosis, these studies cannot exclude the possibility that physically active individuals who develop breast cancer acquire tumors that are biologically less aggressive. Therefore, being physically active prior to diagnosis may have been associated with an earlier disease stage. Two studies, assessing physical activity in the year prior to diagnosis, observed nonsignificant reduced risks of breast cancer death with higher levels of pre-diagnosis physical activity [2, 3]. However, one study examined change in physical activity from before to after breast cancer diagnosis; and observed increased risk of death associated with decreasing physical activity [2]. Furthermore, compared with women who remained physically inactive both before and after diagnosis, increasing physical activity after diagnosis was associated with a reduced risk of death. These finding emphasizes the importance of maintaining or increasing physical activity levels after a diagnosis of breast cancer to gain the maximum benefits of physical activity on survival. Two large observational studies have also shown that participation in 3 hr/wk of moderate-intensity recreational physical activity after a diagnosis of colon cancer is associated with a 50–63% reduction in the risk of total death and 39–59% reduction in the risk of colon cancer death [22, 23]. The inverse relations between postdiagnosis physical activity and colon cancer mortality remained largely unchanged across strata of sex, BMI, age, disease stage, or year since diagnosis. In summary, these observational findings of post-diagnosis physical activity and improved survival suggest that exercise may confer additional improvements in breast and colon cancer survival beyond surgery, radiation, and chemotherapy. However, despite this growing body of observational evidence suggesting a strong link between physical activity and cancer survival, there is still the potential for confounding by unknown or poorly characterized variables. For example, physical activity may be a marker of overall health behaviors including adherence to adjuvant treatments. Thus, randomized controlled trials testing the effects of physical activity on cancer survival and/or surrogate/biological markers mediating the association between physical activity and survival are necessary and would provide critical information for cancer survivors about whether and how much lifestyle change can affect their prognosis. While a trial of physical activity on cancer survival has yet to be done, a small number of randomized trials of exercise on surrogate/biological markers of survival have been published. Refer to the Biological Mechanisms section 7 below for a review of these studies.
5 Nutrition and Cancer Survival With the growing population of cancer survivors, it is now recognized that evidence regarding nutrition, cancer prognosis, and survivorship is needed. At present, there are hundreds of studies that have investigated nutritional factors in the etiology of various cancers, but fewer that have evaluated nutrition in relation to survival. Several new cohort studies of cancer survivors that include post-diagnosis dietary
224
M.L. Irwin
assessments are now underway as reviewed by Kushi and colleagues [24]. A recent publication by Kwan and colleagues from the Life After Cancer Epidemiology Study examined dietary patterns and breast cancer survival. They observed a prudent dietary pattern (i.e., high intake of fruits, vegetables, whole grains, and poultry) was associated with a 43% reduced risk of overall death (p = 0.02). In contrast, a Western diet (high intakes of red and processed meats and refined grains) was related to a 53% increased risk of overall death (p = 0.05) [25]. Two important large-scale multi-site studies evaluated dietary interventions in breast cancer survivors. The Women’s Intervention Nutrition Study (WINS) was a randomized trial of a dietary intervention designed to reduce fat intake in 2437 women with resected, early-stage breast cancer [26]. The low-fat diet group consumed an estimated 33 g total fat/day in comparison with 51 g total fat/day in the usual care group. An interim analysis based on 5 years of follow-up reported that the intervention group had a lower risk of relapse events (HR = 0.76, 95% CI 0.60–0.98). Of particular interest, when the results were stratified by the estrogen receptor (ER) status (positive vs. negative) of the women’s first breast cancer, a more substantial benefit was observed in women who had ER-negative breast cancer (RR: 0.58, 95% CI: 0.37, 0.91). Women with ER-positive breast cancer who were randomized to the low-fat diet also had fewer second breast cancers, but the result was not statistically significant (RR: 0.85, 95% CI: 0.63, 1.14). However, the updated results from the WINS study (presented at the San Antonio Breast Conference in December 2006, www.sabcs.org ) demonstrated a non-significant improvement in disease-free survival in the intervention group compared with usual care. The Women’s Healthy Eating and Living Study (WHEL) was a randomized clinical trial of high fruit and vegetable and low-fat diet versus usual diet in breast cancer survivors. This trial demonstrated that women randomized to a high fruit and vegetable diet increased consumption of fruits and vegetables, as supported by marked increases in plasma carotenoids, which are biomarkers of fruit and vegetable intake [27]. Despite this apparent adherence, and a 7.3-year follow-up period, rates of second breast cancer were similar in the two dietary arms of this trial (HR = 0.96, 95% CI 0.80–1.14) and mortality was lower, but not significantly (HR = 0.91, 95% CI 0.72–1.15). The women in the non-intervention arm of this study were consuming 3.8 servings of vegetables and 3.4 servings of fruit at baseline; therefore, this trial was evaluating incremental benefits in cancer survival for consuming very high intakes versus high intakes. The observational evidence commonly suggests that the incremental benefits are largest when intervening in populations with low intakes, so the results of this trial should be interpreted in their proper context (no additional benefit to consuming more than 5 servings a day). These two trials are by far the largest trials of nutritional interventions in cancer survivors; however, some smaller trials of dietary interventions as well as some larger randomized trials of nutrient supplements have been done as recently reviewed [28] by Davies et al. This review includes trials in cancer patients as well as patients with premalignant end points such as adenomatous polyps. The authors noted that all-cause mortality was nonsignificantly lower in trials of a “healthy diet”
8
Energy Balance, Cancer Prognosis, and Survivorship
225
(OR 0.90, 95% CI 0.46–1.77). Analyses of nutrient supplements, including antioxidant supplements, showed no clear evidence of benefit or harm in cancer survivors. However, it should be emphasized to patients that high-dose nutrient supplements are not recommended for generalized cancer prevention, and some studies of cancer survivors have observed adverse effects from nutrient supplements. As an example, Bairati and colleagues observed a higher rate of second primary cancers and poorer cancer-free survival in head and neck cancer patients randomized to receive alphatocopherol as compared with placebo [29]. The meta-analysis, published in 2006, concluded that there is an urgent need to better understand the effects of diet and nutrient supplements on cancer outcomes. In regards to supplements, vitamin D and its potential association with prognosis is receiving a lot of attention. Recent research suggests breast cancer patients might fare worse if they suffer from vitamin D deficiency. These results were presented at the 2008 annual meeting of the American Society of Clinical Oncology by Goodwin and colleagues [30]. They measured vitamin 25-OH D levels in the blood of 512 newly diagnosed breast cancer patients and followed them for 12 years. Compared with women with adequate vitamin D, women with deficient levels had significantly less disease-free survival (hazard ratio [HR] 1.94, p = 0.02) and overall survival (HR 1.73, p = 0.02). The prognostic significance of vitamin D levels was independent of patient age or weight, tumor stage, or tumor grade. In summary, because cancer survivors are often at heightened risk for non-cancer chronic diseases such as coronary heart disease, and frequently live a long time and may die of other diseases, it is obviously prudent to emphasize dietary patterns associated with lowered all-cause mortality. Future trials of low calorie diets with physical activity are necessary to better understand the impact of weight loss, diet, and physical activity on disease-free survival.
6 Impact of Both Nutrition and Physical Activity on Cancer Survival A recent observational analysis of the non-intervention arm of the WHEL study suggested that the combination of fruit and vegetable intake with physical activity was beneficial [4]. Specifically, breast cancer survivors who consumed 5+ daily servings of fruits and vegetables and who exercised an amount equivalent to walking 30 min a day, 6 days per week, had a significant survival advantage. This was not seen in women who engaged in only one behavior (diet or physical activity); rather, the combined influence of diet and physical activity was associated with risk reduction. Notably, only a minority of breast cancer survivors engaged in both health-promoting behaviors, suggesting obvious opportunities for survivorship interventions. Given the lack of randomized trials of both diet and exercise or weight loss on survival in men and women diagnosed with cancer, a couple small-scale trials of diet and exercise on surrogate markers of survival (serum hormones and body composition) have recently been initiated. Mefferd and colleagues recently published results
226
M.L. Irwin
from a cognitive behavioral therapy intervention for weight loss through exercise and diet modification on risk factors for recurrence of breast cancer [31]. Eighty-five overweight breast cancer survivors were randomly assigned to a once weekly, 16-week intervention, or wait-list control group. Significant differences in weight, BMI, and percent body fat were observed between intervention and control groups (p < 0.05). Their results indicate that 16 weeks of a cognitive behavioral therapy program for weight management may reduce obesity in overweight breast cancer survivors.
7 Mechanisms Linking Energy Balance to Cancer Survival The beneficial effects of weight loss, healthy eating, and physical activity on improved prognosis after a cancer diagnosis may be mediated through beneficial changes in metabolic (insulin) and sex hormones (androgens and estrogens), growth factors (insulin-like growth factor (IGF)-I and IGFBP-3), adipokines (leptin, adiponectin), and/or inflammation (C-reactive protein) [32]. Specifically, there is increasing evidence that high insulin levels strongly increase the risk of breast and colon cancer recurrence and death. Three recent studies have observed an approximate three-fold increased risk of all-cause mortality among women with high insulin levels, measured approximately 2 years after breast cancer diagnosis, relative to women with low insulin levels [33]. The strong association between fasting insulin levels and death has led a number of oncologists and scientists to consider the targeting of insulin as a therapeutic modality in breast cancer, particularly because insulin can be modified by lifestyle and pharmacologic interventions. Similar to insulin, IGF-I has potent mitogenic and antiapoptotic properties in normal and malignant epithelial cells, whereas IGFBP-3 can either stimulate or suppress cellular proliferation by restricting IGF-I’s availability and biological activity. For insulin, some mitogenic effects may be mediated by interaction with IGF-I receptors, as hyperinsulinemia promotes the synthesis and activity of IGF-I. Although the data are not consistent, high levels of IGF-I and low levels of IGFBP-3 have been associated with an increased risk of breast cancer and adverse prognostic factors; however, a study by Goodwin and colleagues found high levels, rather than low levels, of IGFBP-3 predicted distant recurrence of breast cancer in postmenopausal women [34]. Most recently, two studies have demonstrated decreased serum insulin levels in breast cancer survivors in response to recommended amounts of physical activity [35, 36], and one of the studies also observed beneficial changes in IGFs [36]. Thus, physical activity may improve prognosis via favorable changes in insulin and IGFs. Obesity, a high insulin level and altered IGF levels are also associated with a less favorable sex hormone profile [5, 8, 38, 40]. Sex steroid hormones have powerful mitogenic and proliferative influences and are strongly associated with the development and control of breast cancer. A number of clinical trials show that estrogen ablation increases survival following a diagnosis of breast cancer [37]. Changes in sex hormones are perhaps the most consistently cited potential mechanism for
8
Energy Balance, Cancer Prognosis, and Survivorship
227
the association between physical activity and improved breast cancer survival. The primary mechanism of reduced calorie diets and physical activity influencing sex hormones in postmenopausal women is via decreased body fat, a substrate for estrogen and testosterone production, which results in less tissue capable of aromatization of the adrenal androgens to estrogens. Observational and clinical trials have shown favorable effects of healthy eating, physical activity, and weight loss on sex hormone concentrations [38–40]. Preliminary in vitro and epidemiologic data have also suggested a link between adipokines (e.g., leptin and adiponectin), inflammatory markers (e.g., C-reactive protein), and poor cancer prognosis [41]. Adipokines exhibit strong associations with body mass index, abdominal fat mass, and hyperinsulinemia. In addition, several adipokines including leptin promote angiogenesis, which is essential for cancer development and progression and can stimulate estrogen biosynthesis by the induction of aromatase activity. C-reactive protein (CRP) is a well-known systemic marker for inflammation that is produced by the liver, and is only present during episodes of chronic inflammation, and has been shown to be associated with worse prognosis. While these hormones may be related to cancer survival, very few studies have been published examining the impact of weight loss, diet, or exercise on adipokines and inflammation in cancer survivors. Future studies need to examine the effect of energy balance on adipokines and inflammation. In summary, with numerous publications showing statistically and clinically significant associations between energy balance and poor prognosis, it becomes increasingly important to identify modifiable factors that improve survival and can be used as surrogate markers of survival. Although much further research is needed, evidence to date suggests that weight loss, healthy eating, and physical activity favorably modify surrogate markers of cancer survival.
8 Energy Balance and Psycho-Social/Quality of Life Benefits Because of the continually improving survival rates, resulting in a large population of over 11 million cancer survivors in the United States alone, the psychological well-being and physical functioning of survivors is important from a public health standpoint. Furthermore, treatment advances, new chemotherapeutic agents, hormone therapies, and biologic therapy have, at least in part, altered the psychological impact of a diagnosis of cancer. As a result of the cancer diagnosis, surgery, and adjuvant treatments, some cancer survivors experience fatigue, depression, anxiety, reduced overall quality of life, (QOL) and weight gain. Recent systematic reviews and meta-analyses have reported clear benefits of physical activity for improved quality of life, reduced fatigue, and cardiovascular fitness among cancer survivors [42]. In one of the largest studies to date, Courneya and colleagues examined the effects of aerobic exercise alone, resistance exercise alone, or usual care on quality of life in 242 breast cancer survivors initiating chemotherapy [43]. There were significant favorable effects of both aerobic and resistance exercise on multiple outcomes including self-esteem, fitness, and body
228
M.L. Irwin
composition, as well as increased chemotherapy completion rates compared with usual care. Furthermore, no significant adverse events were reported; lymphedema did not increase or was not exacerbated by aerobic or resistance exercise. Courneya and colleagues also completed a similar trial of aerobic exercise vs. usual care in breast cancer survivors who had completed adjuvant treatment and observed similar favorable effects of exercise on overall quality of life [44]. Overall, these, and other, studies have demonstrated that exercise is safe in cancer survivors and produces beneficial effects on quality of life and cancer-related symptoms with no adverse side effects. Most recently, Segal and colleagues examined the effect of 24 weeks of aerobic or resistance exercise compared with usual care on fatigue and quality of life in 121 prostate cancer patients initiating radiotherapy with or without androgen deprivation therapy [45]. Both resistance and aerobic exercise mitigated fatigue in prostate cancer survivors, and resistance training generated longer-term improvements and additional benefits for QOL. Lastly, Demark-Wahnefried and colleagues are currently conducting a trial that tests whether a home-based multi-behavior intervention focused on exercise and including a low saturated fat, plant-based diet, would improve physical functioning in 641 older, long-term (> 5 year post-diagnosis) survivors of breast, prostate, or colorectal cancer [46]. In summary, given that many existing cancer therapies are costly and have significant side effects that can result in long-term morbidity and even mortality in cancer patients, non-pharmacologic methods to lower cancer recurrence and death, especially those that are also associated with improvements in quality of life, depression, and fatigue, may offer an attractive addition to the currently available treatment options. Additionally, men and women who have survived cancer have an increased risk for developing cardiovascular disease [21], thus an intervention that might have a positive impact on this outcome would be beneficial for cancer survivors.
9 Approaches to Improving Weight, Nutrition, and Physical Activity Behaviors in Cancer Survivors Physical activity and nutrition counseling have not traditionally been a part of the cancer treatment/survivorship program. This is despite the fact that weight gain is common after a cancer diagnosis. Further, a large proportion of cancer survivors do not perform regular physical activity, and many cancer survivors decrease their physical activity levels after diagnosis. Suboptimal diets have also been observed in men and women diagnosed with cancer. Nutritional needs of cancer survivors vary, depending upon their stage in the cancer continuum. Many things can impact cancer survivors’ nutritional status (i.e., reduced appetite and alterations in taste due to chemotherapy), and dietary strategies are available to help cancer patients through the treatment and recovery process. Given, physical activity and nutrition programs carry tremendous potential to affect length and quality of survival in a positive manner and prevent or control morbidity associated with cancer or its treatment, the motivation to maintain a positive behavior change may be higher among cancer
8
Energy Balance, Cancer Prognosis, and Survivorship
229
survivors than men and women without cancer. Survey studies have shown that cancer survivors want to receive information about lifestyle and cancer survivorship [47]. Survey studies have shown that oncologists agree that exercise, diet, and weight loss counseling is beneficial for cancer survivors during and after treatment [48]. Thus, oncologists appear to have a favorable attitude toward recommending exercise, healthy eating, and weight loss to cancer survivors, yet several barriers, such as not being aware of the benefits of these behaviors or referral opportunities, may prevent them from providing counseling. Most recently, working together, the American Cancer Society and the American College of Sports Medicine developed a certification called “the certified cancer exercise trainer” for personal trainers, physical therapists, nurse practitioners, or other health professionals to become certified in counseling and training cancer survivors in how to exercise safely and at recommended levels. These “Certified Cancer Exercise Trainers” are knowledgeable about the potential physical limitations associated with surgery and treatment and have the skills and abilities to help cancer survivors overcome some of the recent and late effects of surgery and treatment (go to www.acsm.org or www.cancer.org for more information on physical activity and diet counseling referral opportunities). In summary, oncologists should discuss with their patients the benefits of physical activity, healthy eating, and weight loss after a diagnosis of cancer, and also reassure them that exercise is safe and associated with improved overall quality of life. While large-scale trials of dietary-induced weight loss and physical activity on survival have yet to be conducted, cancer survivors should seek out the growing number of opportunities that exist toward being physically activity and eating healthy, and oncologists should also become aware of the benefits of these healthy behaviors after a cancer diagnosis, as well as existing referral networks.
10 Conclusions An increasing number of men and women are being diagnosed with cancer, and many cancer survivors are seeking lifestyle-based approaches to improve survival. There are, clearly, many questions to be answered concerning who would benefit from lifestyle change, when these behavior changes would be most beneficial, and what type of diet and exercise program would be most valuable. Future research must be done both to establish the efficacy and effectiveness of weight control, nutrition, and physical activity to lower cancer recurrence and death and to understand the biologic mechanisms through which these methods impact cancer development and malignant potential. However, until these studies are conducted, it may well be a benefit for men and women with cancer to maintain a healthy weight, eat a prudent diet of fruits and vegetables, whole grains and poultry and to exercise 30 min per day. Since a majority of cancer survivors are not currently participating in recommended levels of physical activity or eating a healthy diet, these targeted therapies have the potential to benefit a large number of cancer survivors. Thus, oncologists and primary-care physicians should be encouraged to counsel cancer survivors proactively about nutrition, exercise, and weight control.
230
M.L. Irwin
Future research might also benefit from specifically targeting those survivors who are experiencing psychosocial impairment or reduced quality of life. Despite overall improvements in the health and well-being of cancer survivors, quality of life remains a major concern for certain subgroups of survivors, including young survivors, survivors with a lower level of education, survivors who are diagnosed with later-stage cancer and those who undergo chemotherapy, hormone therapy, or extensive and debilitating treatment regimens. Improved understanding in these research areas will pave the way for physical activity and dietary interventions and programs to become a routine component of cancer treatment and recovery and will hopefully provide the necessary evidence to convince policy makers to include weight loss, exercise, and nutrition counseling in cancer management, and encourage second party payers to reimburse cancer survivors for receipt of their counseling.
References 1. Holmes MD, Chen WY, Feskanich D, Kroenke CH, Colditz GA (2005). Physical activity and survival after breast cancer diagnosis. JAMA, 293(20):2479–2486. 2. Irwin ML (2008). Influence of pre- and post-diagnosis physical activity on survival in breast cancer survivors: the health, eating, activity, and lifestyle (HEAL) study. J Clinic Oncol, 26(24):1–7. 3. Holick CN, Newcomb PA, Trentham-Dietz A, et al. (2008). Physical activity and survival after diagnosis of invasive breast cancer. Cancer Epidemiol Biomarkers Prev, 17(2):379–386. 4. Pierce JP, Stefanick ML, Flatt SW, et al. (2007). Greater survival after breast cancer in physically active women with high vegetable-fruit intake regardless of obesity. J Clinic Oncol, 25(17):2345–2351. 5. Doyle C, Kushi LH, Byers T, et al. (2006). Nutrition and physical activity during and after cancer treatment: an American cancer society guide for informed choices. CA Cancer J Clin, 56(6):323–353. 6. Daling JR, Malone KE, Doody DR, Johnson LG, Gralow JR, Porter PL (2001). Relation of body mass index to tumor markers and survival among young women with invasive ductal breast carcinoma. Cancer, 92(4):720–729. 7. Chlebowski RT, Aiello E, McTiernan A (2002). Weight loss in breast cancer patient management. J Clin Oncol, 20(4):1128–1143. 8. Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ (2003). Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med, 348(17):1625–1638. 9. Irwin ML, McTiernan A, Baumgartner RN, et al. (2005). Changes in body fat and weight after a breast cancer diagnosis: influence of demographic, prognostic, and lifestyle factors. J Clin Oncol, 23(4):774–782. 10. Irwin ML, Crumley D, McTiernan A, et al. (2003). Physical activity levels before and after a diagnosis of breast carcinoma: the health, eating, activity, and lifestyle (HEAL) study. Cancer, 97(7):1746–1757. 11. Blanchard CM, Courneya KS, Stein K (2008). Cancer survivors’ adherence to lifestyle behavior recommendations and associations with health-related quality of life: results from the American Cancer Society’s SCS-II. J Clin Oncol, 26(13): 2198–2204. 12. Dignam JJ, Polite BN, Yothers G, et al. (2006). Body mass index and outcomes in patients who receive adjuvant chemotherapy for colon cancer. J Natl Cancer Inst, 98(22):1647–1654. 13. Wright ME, Chang SC, Schatzkin A, et al. (2007). Prospective study of adiposity and weight change in relation to prostate cancer incidence and mortality. Cancer, 109(4):675–684.
8
Energy Balance, Cancer Prognosis, and Survivorship
231
14. Goodwin, P (1995). In: Bitzer J, Stauber M (eds.), Psychosomatic Obstetrics and Gynecology (pp. 491–496). Monduzzi Editore, International Proceedings Division, Bologna, Italy. 15. Daling JR, Malone KE, Doody DR, Johnson LG, Gralow JR, Porter PL (2001). Relation of body mass index to tumor markers and survival among young women with invasive ductal breast carcinoma. Cancer, 92(4):720–729. 16. Kroenke CH, Chen WY, Rosner B, Holmes MD (2005). Weight, weight gain, and survival after breast cancer diagnosis. J Clin Oncol,23(7):1370–1378. 17. Caan BJ, Emond JA, Natarajan L, et al. (2006). Post-diagnosis weight gain and breast cancer recurrence in women with early stage breast cancer. Breast Cancer Res Treat, 99(1):47–57. 18. Meyerhardt JA, Niedzwiecki D, Hollis D, et al. (2008). Impact of body mass index and weight change after treatment on cancer recurrence and survival in patients with stage III colon cancer: Findings from CALGB 89803. J Clin Oncol, 26: 4109–4115. 19. Buist D, Ichikawa L, Prout M, et al. (2007). Receipt of appropriate primary breast cancer therapy and adjuvant therapy are not associated with obesity in older women with access to health care. J Clin Oncol, 25:3428–3436. 20. Daniell HW (2009). Weight loss after breast cancer diagnosis may not improve prognosis. J Clin Oncol, 27:829–830. 21. Carver JR, Shapiro CL, Ng A, et al. (2007). American society of clinical oncology clinical evidence review on the ongoing care of adult cancer survivors: cardiac and pulmonary late effects. J Clin Oncol, 25(25):3991–4008. 22. Meyerhardt JA, Giovannucci EL, Holmes MD, et al. (2006). Physical activity and survival after colorectal cancer diagnosis. J Clin Oncol, 24(22):3527–3534. 23. Meyerhardt JA, Heseltine D, Niedzwiecki D, et al. (2006). Impact of physical activity on cancer recurrence and survival in patients with stage III colon cancer: findings from CALGB 89803. J Clin Oncol, 24(22):3535–3541. 24. Kushi LH, Kwan ML, Lee MM, Ambrosone CB (2007). Lifestyle factors and survival in women with breast cancer. J Nutr, 137(1 Suppl):236S–242S. 25. Kwan ML, Weltzien E, Kushi LH, et al. (2008). Dietary patterns and breast cancer recurrence and survival among women with early-stage breast cancer. J Clin Oncol, 27:1–8. 26. Chlebowski RT, Blackburn GL, Thomson CA, et al. (2006). Dietary fat reduction and breast cancer outcome: interim efficacy results from the women’s intervention nutrition study. J Natl Cancer Inst, 98(24):1767–1776. 27. Pierce JP, Natarajan L, Caan BJ, et al. (2007). Influence of a diet very high in vegetables, fruit, and fiber and low in fat on prognosis following treatment for breast cancer: the women’s healthy eating and living (WHEL) randomized trial.[see comment]. JAMA, 298(3):289–298. 28. Davies AA, Davey Smith G, Harbord R, et al. (2006). Nutritional interventions and outcome in patients with cancer or preinvasive lesions: systematic review. J Natl Cancer Inst, 98(14):961–973. 29. Bairati I, Meyer F, Gelinas M, et al. (2005). A randomized trial of antioxidant vitamins to prevent second primary cancers in head and neck cancer patients. J Natl Cancer Inst, 97(7):481–488. 30. Goodwin P. Vitamin D and breast cancer prognosis. J Clin Oncol, 26: abstr 511. 31. Mefferd K, Nichols JF, Pakiz B, Rock CL (2007). A cognitive behavioral therapy intervention to promote weight loss improves body composition and blood lipid profiles among overweight breast cancer survivors. Breast Cancer Res Treat, 104(2):145–152. 32. Hursting SD, Lashinger LM, Colbert LH, et al. (2007). Energy balance and carcinogenesis: underlying pathways and targets for intervention. Curr Cancer Drug Targets, 7(5):484–491. 33. Goodwin PJ, Ennis M, Pritchard KI, et al. (2002). Fasting insulin and outcome in early-stage breast cancer: results of a prospective cohort study. J Clin Oncol, 20(1):42–51. 34. Goodwin PJ, Ennis M, Pritchard KI, et al. (2002). Insulin-like growth factor binding proteins 1 and 3 and breast cancer outcomes. Breast Cancer Res Treat, 74(1):65–76. 35. Ligibel JA, Campbell N, Partridge A, et al. (2008). Impact of a mixed strength and endurance exercise intervention on insulin levels in breast cancer survivors. J Clin Oncol, 26(6): 907–912.
232
M.L. Irwin
36. Irwin ML, Varma K, Alvarez-Reeves, et al. (2009). Randomized controlled exercise trial on insulin and IGFs in breast cancer survivors: the Yale exercise and survivorship study. Cancer Epidemiol Biomarker Prev, 18(1): 306–313. 37. Howell A, Cuzick J, Baum M, et al. (2005). Results of the ATAC (Arimidex, Tamoxifen, Alone or in Combination) trial after completion of 5 years’ adjuvant treatment for breast cancer. Lancet, 365(9453):60–62. 38. McTiernan A, Tworoger SS, Ulrich CM, et al. (2004). Effect of exercise on serum estrogens in postmenopausal women: a 12-month randomized clinical trial. Cancer Res, 64(8):2923–2928. 39. World Cancer Research Fund, American Institution for Cancer Research (2007). Food, nutrition, physical activity, and the prevention of cancer: a global perspective. The Second Expert Report. AICR, Washington, DC. 40. McTiernan A, Rajan K, Tworoger S, et al. (2003). Adiposity and sex hormones in postmenopausal breast cancer survivors. J Clin Oncol, 21:1961–1966. 41. Pierce BL, Neuhouser ML, Wener MH, et al. (2009). Correlates of circulating C-reactive protein and serum amyloid A concentrations in breast cancer survivors. Breast Cancer Res Treat, 114:155–167. 42. Bicego D, Brown K, Ruddick M, et al. (2008). Effects of exercise on quality of life in women living with breast cancer: A systematic review. Breast J, Dec 12. 43. Courneya KS, Segal RJ, Mackey JR, et al. (2007). Effects of aerobic and resistance exercise in breast cancer patients receiving adjuvant chemotherapy: a multicenter randomized controlled trial. J Clin Oncol, 25(28):4396–4404. 44. Courneya KS, Mackey JR, Bell GJ, Jones LW, Field CJ, Fairey AS (2003). Randomized controlled trial of exercise training in postmenopausal breast cancer survivors: cardiopulmonary and quality of life outcomes. J Clin Oncol, 21(9):1660–1668. 45. Segal RJ, Reid RD, Courneya KS, et al. (2009). Randomized controlled trial of resistance or aerobic exercise in men receiving radiation therapy for prostate cancer. J Clin Oncol, 27: 344–351. 46. Snyder DC, Morey MC, Sloane R, et al. (2008). Reach out to ENhancE Wellness in Older Cancer Survivors (RENEW): design, methods and recruitment challenges of a home-based exercise and diet intervention to improve physical function among long-term survivors of breast, prostte, and colorectal cancer. Psychooncology, 18(4):429–439. 47. Jones LW, Courneya KS (2002). Exercise counseling and programming preferences of cancer survivors. Cancer Pract, 10(4):208–215. 48. Jones LW, Courneya KS, Peddle C, Mackey JR (2005). Oncologists’ opinions towards recommending exercise to patients with cancer: a Canadian national survey. Support Care Cancer, 13(11):929–937.
Chapter 9
Behavior, Energy Balance, and Cancer: An Overview Donna Spruijt-Metz, Selena T. Nguyen-Rodriguez, and Jaimie N. Davis
1 Overview: Physical Activity, Diet, and Sleep Impact on Both Obesity and Cancer There are arguably only three lifestyle behaviors that are imperative for survival. We can, for instance, survive without working, reading, listening to music, writing, traveling, or driving a car. However, humans (and animals) have to eat, move, and sleep in order to maintain life. All three of these behaviors have been linked to obesity as well as cancer. Thus, not only do obesity and cancer share some common mechanisms, consequences, and reciprocal influences, as shown in previous chapters, they also appear to be influenced by the same lifestyle behaviors. More than 66% of adults, 18 years or older, are overweight or obese (defined as a body mass index (BMI) of 25 or over) [1]. More than 1 million Americans were diagnosed with cancer in 2008, and over 35% of this cancer is preventable through exercise, diet, and obesity reduction [2–8]. It stands to reason that if people know that healthy diet, physically active lifestyles, and sufficient sleep are linked to lower weight and decreased cancer risk, they would attempt to change these behaviors. In a recent nation-wide study, 47% of adults aged ≥20 years said they tried to lose weight during the preceding 12 months [9]. Consumers spend more than $35 billion a year on weight loss products and programs to change diet and physical activity [10]. However, physical activity, diet, and sleep are difficult to change and are strongly influenced by a host of determinants including habit, stress, and the social and built environment. Therefore, although these behavioral domains seem to be under volitional control, they have proven notoriously difficult to change, even for brief periods of time, and long-term change has proven elusive. Each of these behavioral domains encompasses a broad range of specific behaviors, some of which have much stronger empirically documented relationships with obesity and cancer
D. Spruijt-Metz (B) Keck School of Medicine, Institute for Health Promotion and Disease Prevention Research, University of Southern California, Los Angeles, CA, USA e-mail:
[email protected]
N.A. Berger (ed.), Cancer and Energy Balance, Epidemiology and Overview, Energy Balance and Cancer 2, DOI 10.1007/978-1-4419-5515-9_9, C Springer Science+Business Media, LLC 2010
233
234
D. Supruijt-Metz et al.
than others. Current knowledge on relationships between specific behaviors should inform choices of intervention targets. Therefore, this chapter is designed to provide up-to-date information on the relationships of specific behaviors with obesity and cancer in order to inform the design of future interventions to prevent obesity and some forms of cancer. In this chapter we will (1) review the basic tenets of interventions to change lifestyle behaviors, (2) review the relationships of specific components of these three key behavioral domains with obesity and cancer risk, (3) discuss challenges involved in interventions to change these behaviors, and (4) examine the effects of behavior changes on obesity and cancer outcomes. Each section provides examples of studies that focus on specific types of physical activity, diet, or sleep behaviors that have been associated with any form of cancer risk, and is not exhaustive, but meant to illustrate the current state of the science in each field.
2 Intervening to Change Lifestyle Behaviors Figure 9.1 shows a flowchart of choices that need to be made in the course of intervention development. Each of the top boxes will generate a cascade of choices as shown in the corresponding lower boxes and combine to inform the study design.
Population(s) •Ethnicity •Gender •Socioeconomic status •Age •Health status •Cognitive ability
Behavior(s) • Specific behaviors • # of behaviors • Outcome Measures
Setting(s) • Workplace • Schools • Home •Community Settings • Public space • Laboratory • Combination
Theory(s) • Appropriate for behaviors • Appropriate for target population • Measure Theoretical Constructs • Intervene on Theoretical Constructs
Level(s) (Intervention unit) • Individual • Family • Group • Environment • Policy Makers
Modality(s) • Phone • In-person • Text messages • Web • Mass Media • Pharmaceutical
Outcomes • Did behavior change? • How much–clinical vs. statistical significance • Correlated Changes?
• • • •
Intensity Number of contacts Length of contacts Length of Intervention Participant burden (travel time, measurement burden, homework, etc)
Evaluation • Did intervention work as planned? • How much did it cost? • How to improve in future?
Delivery • Research Team • Professional Trainer • Nutritionist • Medical Professional • etc.
Study Design • Number of measurement points • Time to follow-up • Crossover, Control/Comparison Groups, Matching, other design considerations
Fig. 9.1 Intervention decision node flowchart – choices to be made in the development of lifestyle interventions
9
Behavior, Energy Balance, and Cancer: An Overview
235
The combination of elements will dictate, among other things, (1) the success of the intervention at changing the targeted behaviors, (2) the ability to detect behavior change, and (3) the ability to explain behavior change. Interventions based on relevant behavioral theories that incorporate possible mediators and moderators of behavioral change into the study design, outcome measures, and statistical analyses are more informative and arguably more effective [11–13]. Moderators are variables not targeted for intervention and in most cases not expected to change, but which could interact with other factors to influence the outcomes of the intervention. In Fig. 9.1, the first box represents frequent moderators of intervention effects. For instance, several interventions to change diet and physical activity have been more successful in one gender than another [14], in which case gender moderates the effect of the intervention. Mediators are constructs that are hypothesized to fall in the causal pathway between intervention components and behavior. For instance, increases in self-efficacy to increase fruit and vegetable intake could serve as a mediator between improved dietary knowledge and increases in fruit and vegetable intake. Theories of behavior comprise sets of moderators and mediators and can be population specific. For instance, a behavioral theory that posits rational decision making as the main mediator in dietary behaviors might not be suitable for development of an intervention targeting adolescent populations [13]. A review of theories of health-related behaviors is beyond the scope of this chapter. See, for instance, Sallis et al. [15] and Spruijt-Metz and Saelens [13] for reviews of theories and correlates of physical activity in youth, Wendel-Vos et al. [16] and Trost et al. [17] for correlates of physical activity in adults, Shaikh et al. [18], for correlates of dietary intake in adults, and Blanchette et al. [19] and McClain et al. [20] for correlates of dietary intake in youth. The existing literature on psychosocial factors and sleep tends to focus on the psychosocial effects of sleep disturbance (i.e., psychosocial factors are the outcome rather than the predictor). For example, children who experience more sleep problems have more depressive symptoms [21] and their mothers experience more negative mood and parenting stress [22]. In a study of older adults, sleep disturbances were found to be related to depression [23]. The study of psychosocial correlates affecting sleep behavior is a new area of research, and to the best of our knowledge, an overview of these correlates has not yet been published. Aspects of the built environment can both mediate and moderate physical activity, diet, and sleep behaviors. There are many aspects of the built environment that might influence energy balance-related behaviors, such as proximity and availability and costs of healthy food choices [24], proximity of fast-food restaurants, parks, and safe places to play [25], noise levels, and connectivity [26]. These have yet to be fully enumerated and consistently incorporated into behavioral theories [27]. An important correlate of the built environment that also influences energy balance-related behaviors is socioeconomic status. Lower socioeconomic status has been linked to more “obesogenic” environments and an increased risk of obesity in both adult and child populations [28]. Chapter 10 of this book reviews the evidence for interrelationships between aspects of the built environment and energy balance-related behaviors.
236
D. Supruijt-Metz et al.
3 Physical Activity, Sedentary Behavior, and Cancer Risk Chapter 7 in this volume documents the epidemiological evidence linking physical activity to cancer risk and prevention. Physical activity seems to exert preventive effects at several stages of carcinogenesis, including tumor initiation and progression. There is convincing evidence for preventive effects of physical activity on breast, colon, endometrial, prostate, and lung cancer [3]. The biological mechanisms that underlie the protective effects of physical activity are not clear [2], although some have been characterized to certain extent. These include reduction in body fat stores, activity-related changes in sex-hormone levels, improved immune function, improved insulin dynamics, activity-induced changes in insulin-like growth factors, reductions in inflammation, and enhanced DNA repair systems [2, 29].
3.1 Domains, Levels, and Measures of Physical Activity One of the difficulties in understanding the mechanisms by which physical activity influences cancer is that physical activity is not one unified behavior. Physical activity is defined as body movement that is produced by the contraction of muscle that increases energy expenditure above the baseline level [30]. All domains of physical activity are included in this definition, including leisure-time physical activity, occupational physical activity, transportation physical activity, household chores, and any other activity other than complete body stillness [12, 31]. There are literally hundreds of different physical activity domains [32, 33]. A recent review of 47 physical activity interventions found 99 distinct physical activity outcomes in the 47 studies reviewed [12]. Different physical activity behaviors, or domains, can have different physiological and biological effects. Walking, leisure activities, work-related activities, organized sports such as basketball, swimming, strength training, and aerobic exercise – these and other activities may affect various biological mechanisms and thus affect cancer in different ways. Furthermore, many of these domains have some overlap. Brisk walking, for instance, is subsumed under the rubric of moderate physical activity, but may not have the same physiological effect or ease of execution as other domains that would also fit under that same rubric. Sedentary behavior, once conceptualized as the absence of physical activity, is now seen as a distinct realm of behaviors such as time spent watching TV, in motorized transportation, or sitting at a desk [34]. Time spent in sedentary behavior has been linked to obesity [35] and cancer risk [36] independently of other domains of physical activity. Within each domain of physical activity there are at least four levels of activity: (1) Intensity, which is the ratio of working metabolic rate to resting metabolic rate (metabolic equivalent tasks or METs), where one MET, or metabolic equivalent, represents the metabolic rate of an individual at rest and is estimated at 3.5 ml of oxygen consumed per kilogram of body mass per minute or 1 kcal/kg/h [32]. METs are frequently used to classify physical activity into light (<3 METs), moderate (3–6 METs), and vigorous (>6 METs) activity [37]. Sometimes MET-hours are used to
9
Behavior, Energy Balance, and Cancer: An Overview
237
quantify activity levels. MET-hours are METs per hour of each activity multiplied by hours per week of each activity; (2) Energy expenditure, which is the amount of kilocalories (kcals) or kilocalories per kilogram of body weight expended in any activity; (3) Duration – time spent in a particular activity; and (4) Frequency with which a particular activity is undertaken – how often in a day, week, month, or year. Thus, the effect of any activity on cancer and obesity might be influenced by properties of the activity itself (for instance, is it weight-bearing, aerobic, which muscles does it use?), the intensity at which it is practiced, the energy expended, the time spent doing that activity, and/or the frequency with which that activity is undertaken. Domains and levels of physical activity in which an individual participates in the course of a day, a week, or some other period of time combine to reveal patterns of physical activity. The protective effects of any specific activity will be related to the level at which the activity is undertaken, including duration and intensity, and in other cases, frequency and energy expenditure. Finally, each domain and level of physical activity can be measured in many different ways, including questionnaires and diaries, objective measures such as pedometers and accelerometers, and observation [31]. The choice of measures depends upon the research design and outcome of interest, study budget, and population to be studied. Many studies use a combination of these modes of measurement to assess the same behavioral outcome. For instance, if total physical activity is a target, this might be measured objectively (by accelerometry) and subjectively (by questionnaire). Each of these measures may be more or less reliable and valid, and may or may not show strong agreement with each other. The mode of measurement may influence research findings. Therefore, to examine the effects of physical activity on cancer or obesity, the behavioral domain, level, and mode of measurement needs to be taken into consideration.
3.2 Physical Activity Domains and Cancer Risk Chapter 7 of this volume gives a complete epidemiological overview of the relationships between any type of physical activity and specific cancers. This section, on the other hand, provides examples of specific types of physical activity that have been associated with any form of cancer risk. To illustrate physical activity type– cancer relationships here, examples of some significant relationships are reviewed. Although many types of physical activity have been related to reduced breast cancer risk, we only highlight a few here. Walking: Walking is a common and safe physical activity behavior that has been related to a myriad of health benefits in people of all ages. Walking to school is associated with lower BMI in youth [38]. In a case–control study in adults in Shanghai in 931 colon cancer patients and 1,552 randomly selected controls examined lifetime commuting to work and cancer risk. Walking for more than 30 minutes per day reduced colon cancer risk by 29% in men and 43% in women [39]. In a prospective study of 74,171 women aged 50–79 [40], women who participated in
238
D. Supruijt-Metz et al.
1.25–2.5 hours of brisk walking per week had an 18% decreased incidence of breast cancer compared with less active women. The protective domain of physical activity was walking; the intensity was brisk (3.8 METs or moderate physical activity) and the duration was between 1.25 and 2.5 hours. All of these parameters may be important to achieve effective prevention. Bicycling: In the case–control study by Hou et al. mentioned above [39], bicycling for transportation purposes was also associated with colon cancer. Risk was reduced consistently with an increasing average time spent bicycling for both men and women, with risk reduction of nearly 59% for those who rode a bicycle for more than 2 hours per day. Decreased risk of breast cancer with increasing cycling activity levels has also been reported in premenopausal and postmenopausal women [41]. Leisure time or recreational physical activity: Leisure time or recreational physical activity is a composite that can encompass sports, biking, walking, household chores, gardening, and leisure-time sports, depending upon the physical activity measure used. McTiernan et al. [42] found that >7 hours of moderate to vigorous recreational physical activity per week was protective against breast cancer incidence (RR = 0.79, 95% CI, 0.63–0.99) in a sample of 74,171 US postmenopausal women aged 50–79. Recreational physical activity was a composite of time spent in activities such as (but not limited to) aerobics, aerobic dancing, jogging, tennis, swimming laps, biking outdoors, using an exercise machine, calisthenics, easy swimming, and popular, folk or slow dancing, bowling, and golf. Another study in 685 colon cancer cases and 2,434 control subjects found significantly lower risks of colon cancer with as little as 10 minutes of leisure-time physical activity 2–6 times per week [43]. Occupational physical activity: A large European cohort study examined the relative risk of breast cancer between women whose jobs involved walking, lifting, or heavy manual labor and women whose jobs were sedentary. The study found a 52% reduction in breast cancer risk for women who reported doing heavy manual labor [44]. It should be noted that these are potentially confounded by health at baseline, which was not controlled. Women in better health might be more capable of performing heavy manual labor, therefore it is unclear if heavy manual labor or superior health actually reduced cancer risk. Household physical activity: Household physical activity, like recreational and occupational physical activity, is defined variously and measures are not uniform or necessarily comparable. Household physical activity is, however, for many women a substantial percentage of their overall physical activity. One study found that household activity was associated with a significantly reduced risk of breast cancer in postmenopausal (HR, 0.81; 95% confidence interval, 0.70–0.93, highest versus the lowest quartile; p = 0.001) and premenopausal (HR, 0.71; 95% confidence interval, 0.55–0.90, highest versus lowest quartile; p = 0.003) women [45]. MET-hours, moderate, and vigorous physical activity: Almost any activity can be used to calculate MET-hours (METs per hour of each activity multiplied by hours per week of each activity) to understand time spent in moderate to vigorous physical activity, depending on the measurement instruments used. In a
9
Behavior, Energy Balance, and Cancer: An Overview
239
large cohort of 416,277 men and women across 10 European countries, Steindorf et al. [46] found that women who participated in vigorous non-occupational physical activity for between 13.5 and 33.5 MET-hours a week had a significantly reduced risk for lung cancer (RR = 0.6, 95% CI, 0.40–0.89), as did men who participated in at least 18 MET-hours a week of sports (RR = 0.71, 95% CI, 0.50–0.98). Energy expenditure: In the Harvard Alumni Health Study of 17,148 men, energy expenditure (assessed by kilojoules per week (4.2 kJ=1 kcal) spent in physical activity from blocks walked, flights climbed, and participation in sports or recreational activities [47]) was associated with reduced risk in colon cancer [48]. Risk ratios (95% CI) for colon cancer, with inactive set at 1.00, was 0.52 (0.28–0.94) for moderately active participants and 0.50 (0.27–0.93) for highly active participants. Aerobic physical activity: The 2008 Physical Activity Guidelines synthesized the current data for the protective effects of aerobic physical activity, finding that those who participated in aerobic physical activity for approximately 3–4 hours per week at moderate or greater levels of intensity had on average a 30% reduction in colon cancer risk and a 20–40% lower risk of breast cancer, compared with those who were sedentary [49]. Strength training: Although strength training has not been directly linked to cancer prevention, it has been shown to affect important cancer biomarkers including body fat, waist circumference, fasting insulin, fasting glucose, insulin-like growth factor I (IGF-I), and several IGF-binding proteins [50]. Sedentary behaviors: Sedentary behavior is no longer considered merely a lack of activity, but has been defined as a class of behaviors, such as television viewing. One study examined the association between physical activity, sedentary behavior, and ovarian cancer risk in the American Cancer Society Cancer Prevention Study II Nutrition Cohort, a prospective study of cancer incidence and mortality [51]. Baseline information was collected in 1992. From 1992 to 2001, 314 ovarian cancer cases were found among 59,695 postmenopausal women who were cancer free at enrollment. In this large cohort, there were no associations between past physical activity or recreational physical activity at baseline and risk of ovarian cancer. However, a prolonged duration of sedentary behavior (for 6 vs. <3 hours per day) was associated with an increased risk of ovarian cancer (hazard rate ratio = 1.55, ptrend = 0.01). Clearly, most of these behavioral categories might be further parsed – perhaps into clusters of types of physical activity with similar physiological effects and outcomes – in order to understand that health benefits of particular classes of physical activity.
4 Diet, Nutrition, and Cancer Risk Nutrition has been widely studied as a leading behavioral factor in the prevention and treatment of various types of cancers, including breast, ovarian, colon, and prostate cancers. It has been well documented that positive energy intake, or
240
D. Supruijt-Metz et al.
over-nutrition, is linked to increased cancer risks. As shown in previous chapters, excessive energy intake can lead to increased cancer risk through a variety of proposed pathways, including increased obesity, which are associated with increased IGF-1, insulin, insulin resistance, and adipokines, all of which can lead to inflammation and sex steroid metabolism.
4.1 Domains, Levels, and Measures of Dietary Intake Like physical activity, dietary intake is not one behavior, but a vast array of behaviors that can be characterized by domains – which can be categorized as micronutrients or macronutrients, and levels, i.e., the amount of food ingested. Domains and levels of an individual’s dietary intake over the course of a day, a week, or some other period of time combine to reveal dietary patterns. Dietary intake is notoriously difficult to measure, and as with physical activity, can be measured in various ways. There are some objective measures that have been used to assess dietary intake. For instance, there are several biomarkers that have been used to assess specific nutrients, for instance, intake of protein by urinary nitrogen. Other objective measures include direct observation, often in combination with measures of plate waste. However, most studies use some form of self-report to assess diet. Self-report measures of dietary intake include recall by interview or dietary history by interview, dietary intake diaries, food frequency questionnaires, dietary screeners, and for children parental proxy recalls [52, 53], and more recently using web-based interfaces, hand-held and mobile technologies [54]. Some of these measures, such as the dietary recall, are meant for capture total energy intake as well as a complete inventory of macronutrients and micronutrients and individual foods and beverages, capturing both level and domain. These types of measures are very intensive for both researchers and participants and can be expensive. Other measures, such as food frequency questionnaires and dietary screeners, are more suited for capturing consumption frequency of certain types of foods rather than total energy or nutrient intake, and are often meant to reflect dietary intake over a period of time, such as a month. As is the case for physical activity, the choice of measures depends upon the research design, outcome of interest, study budget, and population to be studied. Each of these measures may be more or less reliable and valid, and may or may not show strong agreement with each other. The mode of measurement may influence research findings.
4.2 Macronutrients, Micronutrients, and Cancer Risk Much attention has been paid to the influences of macronutrients on cancer risk and protection. Because there is not a separate chapter in this book on the epidemiology of nutrition and cancer risk, we will review several nutrients and foods that are
9
Behavior, Energy Balance, and Cancer: An Overview
241
linked to either an increase or decrease risk of cancer. Specifically, we review the evidence for dietary fat, protein intake, soy products, carbohydrates, fiber and whole grain foods, dietary sugar, glycemic index, glycemic load and added sugars. Dietary fat: Dietary fat is the most commonly studied macronutrient in relation to various types of cancer. There is much debate over whether dietary fat is associated with increased cancer risk. Recently this topic has become a source of public confusion and controversy. In animal models, higher dietary fat intake was linked to increased cancer risk [55]. However, the data for humans are less clear. Numerous studies have shown that total dietary fat is associated with an increased cancer risk [56–58] while others show no association [59, 60]. In a meta-analysis of 12 international case–control studies [58], there was a positive association between dietary fat intake and breast cancer, while in an analysis of seven Western cohort studies [60] no such association was found. The relationship between the types of dietary fat and cancer is also a topic of much debate. Some studies have shown that there is no association between type of dietary fat, i.e., saturated (SFA), monounsaturated (MUFA), and polyunsaturated (PUFA) fat intake and cancer risk [61, 62], while others show that SFAs [63] and MUFAs [64] are positively associated and PUFAs [65] are inversely related to cancer risk. There are some data to suggest that PUFAs high in ω–6 fatty acids (or popularly referred to as omega-6) have a strong direct association with breast cancer [66], while ω–3 fatty acids (or omega-3), which are high in fish oils, are protective against breast cancer [67]. Trans fatty acids have also been associated with an increased breast cancer risk [68]. Protein intake: The majority of the studies that have shown a relationship between protein intake and cancer risk have assessed the different foods that contain this macronutrient, because there are a wide range of high-fat and low-fat protein sources. One study found that women who consumed the highest quartile of red meat had a 91% increased risk for breast cancer [69]. In this same study, fish intake and dairy intake, specifically low-fat dairy, were associated with a 28 and 34% reduction in breast cancer. Another study found that women consuming fish five or more times a week was associated with a 25% reduction in breast cancer when compared to women who consumed fish fewer than three times a month [70]. However, as mentioned above, this relationship between protein sources and cancer risk may be more related to the fat content. Soy products: In the last two decades, there has been a great deal of interest in the role that soy products play in cancer risk, specifically breast cancer. In the year 2000, more than 25% of individuals reported eating soy products at least once per week [71]. Also, soy products are increasingly being used as food additives and meat substitutes in the US market. Soybeans contain phytoestrogens, i.e., isoflavones genistin and daidzein, which may exert a biologic effect that potentially reduces breast cancer. However, there is also some evidence to suggest that soy or isoflavones can increase breast cancer risk [72, 73]. One study showed that in postmenopausal Chinese women, soy protein intake increased breast cancer risk, whereas it was inversely associated when pre- and postmenopausal women were combined [72]. In a meta-analysis of 18 epidemiologic studies published from
242
D. Supruijt-Metz et al.
1978 through 2004, high soy intake was modestly associated with a 14% reduction in breast cancer risk, and this association was stronger for premenopausal women [74]. Carbohydrates – fruit and vegetable intake: The relationship between carbohydrate intake and cancer risk has been widely studied, usually assessing carbohydrate food groups. It has been well documented that fruit and vegetable consumption is associated with a lower risk of various types of cancers. One study showed that approximately one additional serving of raw vegetables per day was associated with a 26% reduction in colon cancer, 16% reduction in rectal cancer, and a 15% reduction in breast cancer [75]. Total intake of cooked vegetables was associated with a substantial decrease in colon (28%) and rectum (26%) cancers, but only a modest reduction in breast cancer (4%) [75]. As for fruit intake, only apples, pears, and kiwi were associated with risk reductions of at least 5% for all cancers [75]. A case–control study in Germany found that intake of total and raw vegetables were associated with a 62 and 51% decrease in breast cancer, respectively [76]. Another study showed that there was no association between total vegetable intake and breast cancer risk, but dark green vegetables and yellow-orange vegetables were inversely associated [77]. This study also showed that total fruit consumption, regardless of type, was associated with a decreased risk of breast cancer. However, these significant associations may be more related to the increased fiber intake and subsequent decreased consumption of calorically dense foods. Fiber and whole grain foods: There is substantial evidence that dietary fiber and whole grain foods are related to decreased risk of cancer. In a meta-analyses of 13 case–control studies, Howe et al. [78] showed that adults with a high intake of fiber had a 50% lower risk of colorectal cancer relative to people with a low intake of fiber. Similarly, total fiber intake and especially cereal fiber were associated with a 50% reduced risk of breast cancer in postmenopausal women receiving hormone therapy [79]. In the European Prospective Investigation into Cancer and Nutrition (EPIC) study, people in the highest fifth of dietary fiber intake had a 29% lower risk of colorectal cancer compared to the lowest fifth of fiber intake [80]. In 492,321 adults from the National Institute of Health – AARP Diet and Health Study, total dietary fiber, fiber from grains and whole grain foods were inversely associated with small intestinal cancer incidence [81]. In the study by Adzersen et al. [76], intake of whole grain products was associated with a 57% reduction in breast cancer risk. Although many of these studies controlled for energy intake, the findings could be, in part, related to the decreased energy intake. Regardless, the evidence is strong and suggests that dietary fiber is protective against various types of cancers. Sugar: Recently, there has been a great deal of interest in assessing the effects of dietary sugar on cancer risk. In the World Cancer Research Fund review [82], where 12 case–control studies and one cohort study were assessed, 9 of these studies reported that diets high in refined sucrose or sucrose-containing foods were associated with increased risk of colorectal cancer. In a population-based case–control study consisting of 4,403 men and women, added sugar was associated with a 30%
9
Behavior, Energy Balance, and Cancer: An Overview
243
increase in colon cancer and the sucrose to dietary fiber ratio was associated with over a two-fold increase in colon cancer risk [83]. The risk was even greater for people who were also sedentary and had a higher body mass index. In a case–control study of 2,019 women, consumption of sweets, particularly sodas and desserts, 9.8 more times per week compared with <2.8 times per week was associated with a 35% increase in breast cancer [84], after adjusting for energy intake. Glycemic index (GT) and glycemic load (GL): Carbohydrates are often classified in terms of GI and GL to better access their impact on glucose metabolism. GI refers to the quantitative assessment of foods based on postprandial blood glucose response as compared to a reference food (white bread or glucose) [85]. GL is calculated by multiplying the GI by the amount of carbohydrate consumed [86]. Glycemic index (GI) and glycemic load (GL) have been recently studied for their roles in reducing or promoting cancer risk. In the Women’s Health Study, out of 180 subjects who were diagnosed with pancreatic cancer, sugar intake was not associated with overall pancreatic cancer risk; however, a high glycemic load intake and fructose intake was associated [87]. This association was strengthened in those women with a high body mass index and low physical activity levels. In the National Breast Screening Study [88], consumption of high GI foods was associated with an increased risk of breast cancer among postmenopausal women and this association was also strengthened among women with a high BMI and low physical activity levels. Added sugar intake, Sugar-sweetened beverage intake has increased steadily as documented by both food supply data and nation-wide food consumption survey data in the past couple of decades [89–91]. Increased total sugar and added sugar intake has been linked to increased insulin resistance and hyperinsulinemia in both adult [92–94] and child populations [95, 96]. Given that it is well documented that hyperinsulinemia and insulin resistance increases risk for various types of cancers [97], we would expect that excessive sugar intake would play an even more active role in the promotion of various types of cancers in the future. Other micronutrients: Briefly, there are also several micronutrients that may have protective effects against various types of cancer [77]. Specifically, vitamin C, ß-carotene, vitamin D, calcium, zinc, selenium, and antioxidants have all been shown to be protective against cancer risk. However, this chapter focuses on the macronutrients, which make a more substantial contribution to energy balance.
5 Sleep and Cancer Risk Research on sleep, energy balance, and cancer risk has received less attention than physical activity and diet, and the importance of sleep is just beginning to be recognized. Thus, the literature linking sleep patterns to cancer risk is in its infancy. The following sections highlight the growing work in this area.
244
D. Supruijt-Metz et al.
5.1 Domains, Levels, and Measures of Sleep Sleep, like physical activity and diet, is a multifaceted behavior. The domains of sleep most relevant to obesity and cancer fall into three primary categories: (1) sleep duration (e.g., average time, short sleep, and long sleep), which also might be construed as quantity of sleep; (2) sleep quality (e.g., continuity, fragmentation, architecture, and disorders); and (3) sleep phase (e.g., circadian rhythm and night shifts). Sleep parameters are studied by a variety of objective and subjective methods, depending upon the research questions and hypotheses, available expertise, and available resources. Polysomnography (PSG) is the gold standard for sleep measurement [98]. It monitors and records multiple channels of physiological data during sleep. PSG allows for detailed study of numerous sleep parameters, including sleep duration, sleep continuity, sleep stages, arousals, breathing, oxygen saturation, body position, heart rate, and leg movements. The sensors required for this multichannel recording need to be applied by trained technicians and the resulting data are most reliably scored by specialized polysomnologists. Wrist actigraphy provides information on sleep patterns and circadian patterns, and since it uses a single sensor that can be self-applied, is suitable for use over multiple nights, thus improving its reliability and allowing night to night variability in sleep parameters to be recorded. The most common form of sleep assessment is self-report via sleep diaries (e.g., [99]) and questionnaires. Areas that may be characterized with self-report pertain to sleep habits [100, 101], sleep quality [102], daytime sleepiness [103], and symptoms of sleep disorders, such as snoring or restless legs. As alluded to above, when choosing measures, the best fit will depend on study hypotheses, available expertise and equipment, study population, targeted behaviors, budget, and feasibility.
5.2 Sleep and Cancer Risk Sleep duration: Recommended sleep duration is specific to age, thus reference to sleep duration is not absolute and study findings refer to developmentally appropriate durations. Outcomes associated with sleep duration tend to show a U-shape distribution, where both shorter and longer sleep durations are associated with adverse health outcomes [104]. Sleep duration has received periodic attention as it relates to overall mortality. Large epidemiological studies, such as the Cancer Prevention Study I and II, have shown that sleeping more than 8 hours as well as less than 7 hours per night was associated with increased all-cause mortality [105, 106]. Exploration of covariates showed that most of the increased risk of mortality due to short sleep could be explained by co-morbid conditions. However, the authors were unable to explain why longer sleep was associated with increased risk of mortality. Tamakoshi and Ohno [107] also found that both short and long sleep were associated with increased likelihood of morbidity, including cancer. Interestingly, when psychological conditions were controlled for in men, the increased risk due to short sleep disappeared, indicating a J-shaped curve. It was speculated that perhaps long
9
Behavior, Energy Balance, and Cancer: An Overview
245
sleep may actually be the result of medical conditions which together lead to mortality. However, other studies have shown that sleep duration is related to increased risk of mortality beyond co-morbidities [108–110]. Thus, it is unclear whether or not sleep duration is a direct cause of excessive mortality risk. As such, it is not possible to state that changing sleep duration may decrease overall mortality risk. The biological bases for these observations, and the extent to which variations in sleep duration directly impact health, or rather, represent a marker for other underlying risk factors, is an area of active research. One such burgeoning area of research is that short sleep is a novel risk factor for obesity. Research indicates a link between short sleep and increased obesity [105, 111, 112]. In an examination of environmental risk factors for obesity, short sleep time was a stronger predictor of obesity in children than snacking and watching television [113]. The risk for obesity is nearly three times higher for children who sleep less than 8 hours per night [114], and short sleep has led to same year increases in BMI in adolescent females [115]. Gupta et al. [116] reported that obese adolescents reported less sleep than their normal weight counterparts, calculating an 80% increased risk of obesity for each hour of lost sleep. A study based on parent reports of child sleep duration found that odds of obesity increased 41% for each hour less of sleep [117]. This relationship is further supported by longitudinal studies that have found that sleep duration at 3–4 years old predicted obesity at 7 [118] and 9.5 years old [119]. A prospective study using parents report also found that short sleep in third grade was related to overweight in sixth grade; the authors also found an association of sleep duration in sixth grade to overweight in sixth grade [120]. Sleep thus clearly impacts obesity and several other cancer risk factors. Plausible pathways from short sleep to obesity include the effects of short sleep on diet and physical activity behaviors. Sleep deprivation can lead to increased hunger for high-energy dense foods [121, 122]. Epidemiological data have shown that irregular meal patterns, snacking, overseasoning, and inadequate fruit and vegetable intake were related to insufficient sleep [123]. Short sleep has also led to increased ghrelin and decreased leptin [99, 111, 124]. While ghrelin stimulates appetite, leptin signals satiety; the implications for obesity are clear: short sleep can lead to increased energy intake which can lead to overweight/obesity if not counteracted by equal increase in energy expenditure. However, another study showed that improved sleep quality was related to increases in ghrelin and decreases in leptin in postmenopausal women [125]. Further exploration of these relationships is necessary. Information about the effects of sleep deprivation on human exercise and physical activity is quite limited. One study found that sleep disturbance time was related to decreased levels of physical activity in adolescents [116]. Results from AddHealth data show that healthy sleep habits were related to more physical activity, more fruit and vegetable intake, and less fast food intake [126]. Increased sleep has also been found to be related to physical activity and fitness [127, 128]. Littman and colleagues [125] unexpectedly found that of participants in a physical activity intervention, those that slept less at follow-up (vs. baseline durations) lost more weight than those who did not change sleep duration. Although sleep patterns may have
246
D. Supruijt-Metz et al.
more indirect effects on obesity, there appears to be mechanisms at work aside from energy intake and expenditure. Chaput and colleagues [129] found that both short sleep and long sleep present a risk for future weight gain, even after adjusting for dietary intake and physical activity. Despite growing research on the effects of sleep duration on cancer and obesity risk, further study is necessary to further delineate pathways of influence and increased risk. For example, exploration of sleep parameters other than sleep duration may provide further insight into the etiology of these health outcomes. Sleep quality: It is necessary to identify potential mechanisms through which sleep duration may be associated with negative health outcomes, such as obesity and cancer. A study of elderly participants in Rotterdman revealed that sleep fragmentation may be a mechanism through which short sleep is related to obesity [130]. Although they found a U-shaped association between sleep duration and BMI, this relationship became non-significant when sleep fragmentation was included in the model. Recent research has also explored the possibility that sleep disruption affecting circadian rhythms may be associated with increased risk for cancer. This stems from studies focusing on occupations that are susceptible to conditions that would disrupt normal human circadian function. Endocrine function and hormonal profiles are influenced by light exposure and sleep disruption, therefore this may be the avenue through which these factors may impact the etiology of hormone-related cancers [131]. Another hypothesized pathway per circadian disruption has to do with the light– dark cycle and melatonin release that can be affected by occupations that are susceptible to conditions that would disrupt normal human circadian function. Melatonin is normally released at night; however, light exposure may reduce melatonin levels [132]. Biological evidence of the oncostatic properties of melatonin provides support for this mechanism [132, 133]. The effects of sleep disturbances on immune function provide another plausible avenue of impact. Studies have found that poor sleep can alter normal functioning of the immune system’s natural killer cell activity and production of cytokines [134, 135]. These potential mechanisms are supported by a growing body of research showing that those with occupations that expose workers to sleep and/or circadian disruption are at increased risk for cancer. Studies have found an increased risk of breast cancer for women who engaged in shift work, such as caterers in Denmark [124], women who worked the graveyard shift [136], and nurses working on rotating shifts [137]. Results from the Nurse’s Health Study also show increased risk for colorectal cancer in females working rotating night shifts [138]. Pilots and flight attendants have also been studied, as they are susceptible to sleep disruption due to travel across time zones. A study of flight attendants in California revealed a 60% increased risk of breast cancer for those assigned to international flights [139]. Several studies have also found that airline pilots are at increased risk for prostate cancer [140, 141]. These findings provide support for the mechanisms mentioned above and point to sleep as a potential modifiable risk factor of obesity and cancer. It should also be noted that the domains of sleep (duration and quality) and their impact on health are not mutually exclusive. Obesity and cancer risk may increase for those who
9
Behavior, Energy Balance, and Cancer: An Overview
247
experience short sleep, which results from sleep fragmentation, which leads to circadian disruption. Additional research is needed to parse out the individual effects of these parameters as well as to explore the interacting and/or additive effects of these factors.
6 Interventions to Change Physical Activity and Sedentary Behavior: Overview There have been several recent reviews of physical activity interventions in adults and children. One review examined 45 physical activity interventions in healthy populations that included at least 75 participants and at least a 3-month followup [12]. Thirty of these studies comprised one intervention, 11 comprised two, 4 comprised three, and 2 comprised four interventions. A total of 99 outcomes were measured, with some overlapping but none entirely identical. Interventions occurred in 9 different settings with some occurring in multiple settings. Three-fourths of the studies had some form of in-person contact, while a quarter of the studies relied on contact by mail or telephone alone. Length of interventions varied from less than 2 weeks to over 3 years. Interventions were scored on a scale from 1 to 4 for intensity, based on type of contact, form of contact, number of contacts, and length of intervention. Of the 45 studies reviewed, 45% demonstrated a positive effect on physical activity, and only 4 had an effect size larger that 0.5 at follow-up, 0.5 representing a medium effect size. Only half of the studies were based on any behavioral theory. Whether or not an intervention was theory based had no clear effect on attaining significant intervention effects. It appeared that more intensive studies were more likely to be effective, but less likely to be theory based. These findings are commensurate with other reviews of preventive interventions to increase physical activity in children and other populations. Approximately 50% of interventions had some significant positive effect on health outcome [14].
6.1 Interventions to Change Physical Activity and Sedentary Behavior: Outcomes In a recent review of 43 physical activity weight loss interventions in adults, 21 evaluated a walking intervention, 10 evaluated cycle ergometry (exercise bicycle), 8 evaluated jogging, 8 evaluated weight training, 5 evaluated commercial aerobics, 5 evaluated treadmill exercise, 2 evaluated stair stepping, and 1 evaluated each of dancing, ball games, calisthenics, rowing, and aqua jogging, respectively. No trials evaluated swimming or water aerobics as weight loss interventions. In youth, frequent targets for intervention differ from those in adults, and include such modalities as organized physical activity at school and reductions in sedentary behavior [14]. Many interventions intervene on multiple types of physical activity. Below, we
248
D. Supruijt-Metz et al.
will provide an overview of effects of some of the more frequently targeted behaviors. Where possible we will highlight interventions that focus solely on one type of physical activity. Studies are included here only if they measured outcomes that have previously been related to energy balance and cancer (BMI, % body fat, insulin homeostasis, oxidative stress, etc). Walking: Because walking is relatively safe, and easily incorporated into daily life, it is one of the most frequently chosen behavioral targets, often in combination with dietary or other behavioral targets. In one study, 24 obese non-insulindependent diabetes mellitus (NIDDM) patients were divided into two groups for a 6–8 week training program: 10 were managed by diet alone and 14 were placed in a diet plus walking group. The diet plus walking group was instructed to walk at least 10,000 steps/day on a flat field as monitored by pedometer, while the diet only group was told to maintain a normal daily routine. A glucose clamp procedure was performed before and after the program. Body weight decreased significantly in both groups during the study (p < 0.01). The diet plus walking group lost more weight than the diet only group (7.8 ± 0.8 vs. 4.2 ± 0.5 kg, P < 0.01). Insulin dynamics (clearance rate and glucose infusion rate) improved significantly in the diet plus walking group only, and there was a significant effect of exercise (time × exercise, P = 0.0005) for improvements in insulin homeostasis [142]. Bicycling: Many studies have included cycling as part of an intervention. One small study is of interest because it was in healthy, untrained subjects and measured fat oxidation, body composition by underwater weighing, and skeletal muscle mRNA expression by muscle biopsy. Six male participants, aged 43 ± 2.0 years, BMI 22.7 ± 1.0, took part in a 12-week training program, consisting of cycling on an ergometer at a low intensity (40% of predetermined VO2 max). The average number of sessions was three per week. Average exercise duration was 2 hours a week. At completion of the program, body weight tended to be lower (p=0.07) but there was no effect on % body fat. However, this minimal amount of physical activity increased fat oxidation and led to marked changes in the expression of genes encoding for key enzymes in fat metabolism [143]. Leisure time or recreational physical activity: Increasing recreational (or lifestyle) physical activity has been shown to be effective in changing body composition, insulin dynamics, and a host of other outcomes, usually in combination with dietary interventions [144]. In one study, 173 sedentary overweight postmenopausal women (ages 50–75 years) [144] were randomly assigned to an intervention consisting of exercise facility and home-based, moderate-intensity exercise (n=87) or a stretching control group (n=86). At 12 months, women in the exercise group participated in moderate-intensity sports/recreational activity for a mean (SD) of 3.5 (1.2) day/week for 176 (91) minutes/week. Walking was the most frequently reported activity. Exercisers showed statistically significant differences from controls in baseline to 12-month changes in body weight (–1.4 kg; 95% confidence interval [CI], –2.5 to –0.3 kg), total body fat (–1.0%; 95% CI, –1.6 to –0.4%), and subcutaneous abdominal fat (–28.8 g/cm2 ; 95% CI, –47.5 to –10.0 g/cm2 ). A significant reduction in body fat was observed with increasing duration of exercise.
9
Behavior, Energy Balance, and Cancer: An Overview
249
Aerobic physical activity: Aerobics have been shown to be effective in combination with dietary interventions in improving body composition as well as other cancer biomarkers [145]. In one school-based intervention that focused on increasing aerobic physical activity without a companion dietary intervention, kindergarten children (n = 292, mean age 4.5 ± 0.4 years, 58% boys) were randomized by class (n=10) into an exercise group or a control group (five classes in each) [146]. Specialists delivered a physical activity intervention that included 15 minutes of walking plus 20 minutes of aerobic exercise three times a week for 29.6 weeks. Treatment and control groups also received the usual school physical education provision. Evaluation at 29.6 weeks showed a reduction of the prevalence of obesity in the intervention pre-school children that nearly reached statistical significance (P=0.057). The study showed that intervention girls had a lower likelihood of having an increased BMI slope than control girls (odds ratio 0.32; 95%CI, 0.18–0.56). There was no effect for boys (odds ratio 1.08; 95% CI, 0.62–1.89). Strength training: Exercise training (or strength training) is known to improve insulin resistance in adults. A recent study showed that strength training could also improve insulin dynamics in youth. In a 16-week resistance-training exercise intervention, 22 overweight Latino males aged 15.1 ± 0.5 were randomly assigned to either a twice-per-week resistance-training group [11] or a non-exercising control group [11]. Strength was assessed by one-repetition maximum, body composition was quantified by dual-energy x-ray absorptiometry, and insulin sensitivity was determined by the frequently sampled intravenous glucose tolerance test with minimal modeling. The resistance-training group showed significant increases in upper and lower body strength as compared to the control group. The resistance-training group significantly increased insulin sensitivity compared with the control group (p<0.05), and this increase remained significant after adjustment for changes in total fat mass and total lean tissue mass (p<0.05). Compared with baseline values, insulin sensitivity increased 45.1±7.3% in the resistance-training group versus –0.9±12.9% in controls (p<0.01). Other physical activity forms: Tai Chi and other eastern physical activity modalities such as yoga have been studied as intervention modalities. In one study, 15 healthy adults, mean age 52 ±1.1 years participated in a Tai Chi class approximately 2 times a week for 12 months, and were compared to 17 health controls, mean age 50 ±1.2 years of age. After 12 months, Tai Chi participants had improved levels of HDL, decreased DNA damage, and decreased oxidative damage [147]. Sedentary behaviors: Sedentary behavior has an independent effect on obesity and cancer biomarkers, and may be easier to change, at least in youth [148]. Planet Health was a 2-year school-based interdisciplinary intervention in 1,295 ethnically diverse grade 6 and 7 public school students [149]. Curriculum, incorporated into classroom curriculum, focused on decreasing television viewing and increasing moderate and vigorous physical activity. The intervention reduced television hours among both girls and boys, but had no impact on physical activity. Only girls reduced BMI significantly. Reductions in television viewing predicted obesity change and mediated the intervention effect. Each hour of reduction in television
250
D. Supruijt-Metz et al.
viewing predicted reduced obesity prevalence (odds ratio, 0.85; 95% confidence interval, 0.75–0.97; p = 0.02) (girls only).
7 Interventions to Change Dietary Behaviors: Overview There are few reviews of dietary interventions in adults [150], and most of the studies reviewed include a dietary and physical activity component. The outcome measures have generally been BMI or body weight between intervention and control groups. The majority of dietary interventions range in length from 6 months to 1 year. For adults, a review of 9 large RCT studies showed that only 1 of them focused only on diet [150]. This study was the Women’s Health Initiative with 50,000 postmenopausal women and showed that women who were randomized to a low-fat dietary intervention had significant reductions in BMI at 1 and 7.5 year follow-ups compared to a control group [151]. There have also been several recent reviews on dietary interventions in child populations [152, 153]; however, most of these studies also include a physical activity or lifestyle component. In a review of 21 international school-based interventions aimed at preventing obesity in children [153], 9 of them targeted nutritional behaviors and the general focus was on increasing fruit and vegetable consumption and reducing fat intake. These studies ranged in length from 3 months to 1 year. Although all of these 9 studies showed some improvement in dietary intake, only 1 of the 9 studies showed that a dietary intervention resulted in BMI reduction. Below we illustrate specific dietary behaviors that have been intervened upon to change any form of cancer risk.
7.1 Interventions to Change Dietary Behaviors: Outcomes Fast food: Greater distribution of fast-food restaurants is associated with a greater prevalence of overweight/obesity [154]. One study showed that over 30% of children and adolescents reported consuming fast food on a typical day [155]. Children who ate more fast food compared with those who did not consumed more total energy (187 kcal), more total fat (9 g), more total carbohydrate (24 g), more added sugar (26 g), and more sugar-sweetened beverages (228 g), less fiber (–1 g), less milk (–65 g), and fewer fruits and non-starchy vegetables (–45 g) [155]. Decreasing fast-food consumption is often a component in dietary interventions. However, the effect of decreasing fast-food consumption on weight loss has not yet been isolated separately. Several studies have shown that a dietary intervention that included a fast-food reduction component resulted in modest weight loss [156–158]. Portion sizes: Increased portion sizes have been linked to obesity in children and adults [159, 160]. One study showed that men given 175 g bags of potato chips tripled the amount of chips they ate compared with mean given 25 g bags, taking in an extra 311 kcal [161]. In a randomized control trial, obese adults who received
9
Behavior, Energy Balance, and Cancer: An Overview
251
a 6-month intervention focused on decreasing portion sizes, lost more weight, and decreased type 2 diabetes risk factors compared to the control group [162]. Meal timing: Some studies have shown that infrequent eating frequency called “gorging” vs. frequent eating called “nibbling or snacking” has been linked to excessive energy intake and obesity [163–165]. Eating frequent meals throughout the day suppresses hunger and allows an individual to feel satiated after a smaller meal. One adult study showed that four or more eating episodes compared to three or fewer was associated with a 45% lower risk of obesity [166]. It has been documented that skipping meals often leads to overeating at subsequent meals [167]. Speechly et al. [163] showed that subjects who were fed one-third of their daily energy needs in one meal versus those fed this amount in five meals over 5 hours resulted in ad libitum intake after the 5 hours that was nearly one-third greater in energy intake. Breakfast: National data show that 20% of adults and 30% of children regularly skip breakfast [168, 169]. Recently, numerous studies, in both adults and children, have shown that skipping breakfast is associated with increased obesity. One report showed that skipping breakfast was associated with a 450% increased risk for obesity [166]. Several studies report that individuals who do not eat breakfast have a greater overall daily energy intake [170, 171] and consume more high-fat foods later in the day [172]. Two randomized clinical trials with obese adult women, found that eating breakfast for 12 weeks resulted in significant weight loss [167, 172]. The type of breakfast consumed appears to be important. High fiber, low-GI breakfast meals have been shown to be more satiating and reduce hunger compared to the typical high sugar or high GI breakfast foods [173, 174]. One study showed that obese children who consumed a high-GI breakfast had a 53% higher voluntary food intake for the rest of the day compared to those who consumed a medium- or low-GI breakfast [173]. Consuming a high-GI breakfast has also been shown to increase postprandial blood glucose levels and insulin response [173, 174], which could, in part, explain the drive for overeating at subsequent meals. Therefore, eating breakfast, one that is high in fiber and low in sugar, is helpful in increasing satiation and hunger control, normalizing glucose and insulin response and promoting weight loss. Sugar-sweetened beverages: Several studies have demonstrated higher intakes of sugar-sweetened beverages are associated with increased obesity in youth [175– 178] and adults [179]. On average, a 12-oz serving of soda provides 150 kcal and 40–50 g sugar in the form of high-fructose corn syrup, which is equivalent to about 10 teaspoons of sugar. So if individuals reduced their intake by as little as 1 soda/day, without increasing intake from other sources, it could lead to a 15 lb weight loss in 1 year [180]. In addition, sugar and sugar-sweetened beverage intake has been associated with increased insulin resistance and type 2 diabetes risk in adult [179] and child populations [95, 96]. In a randomized intervention trial with Latino overweight adolescents, those participants who decreased added sugar by the equivalent of 1 can of soda had a 33% reduction in insulin secretion or type 2 diabetes risk [181]. Given that hyperinsulinemia and increased insulin resistance are linked to various types of cancer, reductions in added sugar, specifically sugar-sweetened beverages, could also have a profound effect on decreasing risk of cancer. More research in this area is definitely warranted.
252
D. Supruijt-Metz et al.
Eating in front of the TV: Adult and childhood studies have linked excessive television viewing to obesity. One-quarter of adults who reported watching more than 3 hours of TV a day were classified as obese [35]. National data report that over 25% of children watch TV ≥ 4 hours a day, even on school days [182]. Children who reported ≥ 4 hours of TV per day were 40% more likely to be overweight compared to children who watched ≤ 1 hour of TV per day [182]. In a clinical trial, children who received a 6-month intervention to reduce television viewing had lower body mass index and body fat compared to control subjects [183]. Not only has increased TV time been linked to decreased physical activity levels, as shown above, but it is also linked to unhealthy eating habits [184, 185]. Similarly, when the televisions were on during meals, children consume more red meat, pizza, snack foods, and soda and fewer fruits and vegetables [186, 187]. Caloric restriction: There is substantial and consistent evidence that caloric restriction diets are the most potent dietary intervention to protect against a variety of cancers in animals [188–190]. Evidence that caloric restriction protects against human cancer is scarce and inconsistent. Some studies show that caloric restriction is associated with decreased cancer risk, while others have not. One study found that caloric restriction was associated with a 57% reduction in rectal cell proliferation [191]. Whereas, in a large Dutch study, famine and severe caloric restriction were not associated with a reduction in breast cancer risk [192]. Decreasing dietary fat: There are even fewer randomized control trials that have assessed whether dietary interventions focused on specific macronutrients significantly reduce cancer risk and report mixed results. The two most notable dietary interventions are the Women’s Health Initiative (WHI) and the Women’s Intervention Nutrition Study (WINS) which focused on decreasing fat intake in order to decrease breast cancer risk. However, these two well-known studies showed mixed results. In the Women’s Health Initiative WHI study, women (n=19541) were randomly assigned to a dietary modification intervention group, which goals were to reduce fat intake and increase fruits and vegetable intake or to a control group (n=29294). After 8 years, women who decreased their fat intake did not have a statistically significant reduction in breast cancer risk. However, this study had a fairly minimal goal of fat reduction, i.e., total fat intake of 20% of energy and a fairly minimal maintenance behavioral modification program, i.e., quarterly maintenance sessions from year 2 to 8. In contrast, the interim results from the Women’s Intervention Nutrition Study (WINS) showed that women who were randomly assigned to a low-fat dietary intervention (<15% of energy from fat) (n=975) had a 24% reduction in breast cancer compared to the control group (n=1,462) at the 5-year follow-up. In addition to the reductions in fat intake, women in the WINS dietary intervention also had significant reductions in energy intake and weight and increases in fiber intake, which all could have played a role in reducing cancer risk. Decreasing dietary fat and increasing fiber: One randomized trial showed that women following a 2-year dietary intervention focused on reduction of dietary fat and increased consumption of complex carbohydrates, primarily through increased dietary fiber intake, had reduced breast cancer risk compared to the control group, and this significant reduction remained after accounting for weight loss [193].
9
Behavior, Energy Balance, and Cancer: An Overview
253
Increasing fruit and vegetable intake: A majority of the nutrition interventions in children have focused on increasing fruits and vegetable consumption. A 2-year controlled community based intervention in 5–12-year-old children found that an intervention that targeted increased fruit and vegetable intake and a reduction in sweetened drinks, resulted in a significant increase in fruit intake per day and a small, yet significant, decrease in BMI and waist circumstances [194]. In contrast, a randomized control trial in 10 elementary schools found that a 1-year lifestyle intervention that focused primarily on increasing fruit and vegetable intake resulted in significant increases in fruit and vegetable intake but no reduction in weight or body mass index was found [195]. Another notable family-based study conducted by Epstein et al., showed that parents exposed to a 1-year dietary intervention focused on increasing fruit and vegetable intake had a greater loss of more weight compared to a 1-year dietary intervention focused on decreasing high-fat/high-sugar intake. However, there was no weight reduction in children in either dietary intervention [196]. Although the increasing fruit and vegetable intake approach has been widely used in children, the results are mixed and result in no or small reductions in weight. The Mediterranean diet: The Mediterranean diet is usually rich in vegetables and low in red meat, a moderate amount of fat and a relatively high proportion of monounsaturated fat. A randomized control trial found that women who received a traditional Mediterranean diet had a 40% reduction in circulating hormones, which are known breast cancer risk factors. The Mediterranean dietary intervention in this study focused on the following: (1) reductions in refined sugar, saturated fat, and total fat intake; (2) increased consumption of MUFA and PUFAs; (3) increased intake of fruits and vegetables; and (4) increased intake of foods rich in phytoestrogens [197]. In a recent 2-year trial, 322 moderately obese subjects (mean age, 52 years; mean body mass index 31; male sex, 86%) were randomized to one of three diets: low-fat, restricted-calorie; Mediterranean, restricted-calorie; or low-carbohydrate, non-restricted-calorie. The Mediterranean diet group consumed the largest amounts of dietary fiber and had the highest ratio of monounsaturated to saturated fat. The low-carbohydrate group consumed the smallest amount of carbohydrates and the largest amounts of fat, protein, and cholesterol and had the highest percentage of participants with detectable urinary ketones. Among the 272 participants who completed the intervention, the mean weight losses were 3.3 kg for the low-fat group, 4.6 kg for the Mediterranean-diet group, and 5.5 kg for the low-carbohydrate group. The relative reduction in the ratio of total cholesterol to high-density lipoprotein cholesterol was 20% in the low-carbohydrate group and 12% in the lowfat group. Among the 36 subjects with diabetes, changes in fasting plasma glucose and insulin levels were more favorable among those assigned to the Mediterranean diet than among those assigned to the low-fat diet (P<0.001 for the interaction among diabetes and Mediterranean diet and time with respect to fasting glucose levels). The authors concluded that the Mediterranean and low-carbohydrate diets may be effective alternatives to low-fat diets. The more favorable effects on lipids (with the low-carbohydrate diet) and on glycemic control (with the Mediterranean diet) suggest that personal preferences and metabolic considerations might inform
254
D. Supruijt-Metz et al.
individualized tailoring of dietary interventions [198]. These results suggest that dietary interventions focused on multiple nutrient intakes, not just reductions in dietary fat, may be more beneficial at decreasing cancer risk. Increased soy intake: As mentioned above, soy-containing foods are abundant in phytoestrogens and research suggests that these compounds exhibit chemoprotectant activity against a number of human cancers, including colon, breast, and prostate cancer. Several dietary interventions have shown a direct association between modest consumption of soy products and a reduction in circulating steroid hormone levels. One study showed that in premenopausal women, daily consumption of a 36-oz portion of soy milk, containing 154 mg isoflavones, for the duration of a single menstrual cycle resulted in significant decreases in circulating hormones and breast cancer risk [199]. Similarly, a 3-month study with daily ingestion of just 40 mg of isoflavones resulted in significant reductions in circulating hormones and increased menstrual cycle length [200]. Conversely, a year-long dietary intervention trial where premenopausal women consumed 100 mg/day of isoflavones had no effect on menstrual cycle length or circulating hormone levels [201].
8 Interventions to Change Sleep: Overview Based on the available literature, inclusion of sleep components in cancer and obesity prevention efforts seems warranted, as findings support the potential for changes in sleep patterns to affect mechanisms to these morbidities. Despite these findings, to date, we know of only one pilot sleep intervention study with the larger aim of reduction of risk for obesity and cancer. Data reported at a scientific conference showed success in increasing sleep regularity and that the intervention was acceptable to parents [117]. There have been no published articles of interventions that focus on changing sleep patterns in order to reduce cancer and obesity risk. Although it is likely implausible to see effects of sleep intervention on cancer risk, it is quite possible to assess the impact of improving sleep on obesity risk. Research has shown that sleep patterns can be changed. This provides support for targeting sleep as a modifiable risk factor to target in disease prevention efforts. Evidence comes primarily from literature focusing on reducing insomnia, but also from interventions to improve sleep parameters in cancer patients and/or survivors.
8.1 Interventions to Change Sleep: Outcomes Insomnia: A review of the literature reports that cognitive behavioral therapy (CBT) has shown to be quite effective more than 70% of the time in reducing both difficulties falling asleep and staying asleep [202]. The primary objectives of CBT are to change poor sleep habits, regulate sleep–wake patterns, change maladaptive beliefs/attitudes about sleep, to help participants develop effective coping strategies, and to prevent relapse. There are three components to CBT: (1) behavioral, (2) cognitive, and (3) educational. This involves behavioral strategies such as relaxation which is used to decrease cognitive and physiological arousal through methods
9
Behavior, Energy Balance, and Cancer: An Overview
255
such as muscle relaxation and meditation. Cognitive strategies include changing unrealistic sleep expectations. Educational components focus on teaching proper sleep hygiene, such as avoiding caffeine after dinner time [202]. In a meta-analysis of 59 interventions in over 2000 patients, CBT has proven to reduce sleep-onset latency and night-time awakenings [203]. It was also concluded that stimulus control was the most effective standalone method. Importantly, participants have also reported improved sleep experience and control over sleep which reduces distress over insomnia [202]. Since this distress often leads to insomnia, its reduction may lead to more sustained outcomes. Sleep quality, onset, and efficiency: Similar strategies as those used in CBT have been effective in improving other sleep parameters. An 8-week program that utilized stress reduction and mindfulness led to improved sleep quality and reduced sleep disturbances in cancer patients [204]. Davidson et al. [205] found that stimulus control and relaxation techniques were successful in reducing number of night-time awakenings, time awake after sleep onset, sleep efficiency, and sleep quality in cancer survivors. Another study of breast cancer patients undergoing chemotherapy found that development of an individual sleep program targeting sleep hygiene, relaxation, stimulus control, and sleep restriction were feasible methods of sleep intervention [206]. These reported data serve a two-fold purpose: (1) to show that sleep parameters are indeed modifiable and (2) to provide specific evidence-based strategies to be incorporated into interventions to improve sleep.
9 Conclusions In the realm of physical activity, not all studies find significant relationships between types of physical activity and cancer risk, and each relationship is subject to considerations of level and dose that are only sometimes discussed. Questions that remain to be answered for each type of behavior include how vigorous (METs, energy expenditure), for how long (bout), and how often (per day, week, etc.) must the activity be performed in order to reap the protective benefits? Definitive answers to these questions remain to be answered for most physical activity types. A major limitation in current research is that many studies do not differentiate between types of physical activity, such as aerobic or resistance exercise, or to their subtypes such as jogging versus walking. Rather, activities are often combined into measures of MET-hours per week, or to measures of frequency, or total duration of activity per week. However, evidence is quite strong that different forms of physical activity have different metabolic and physiological effects [145, 207]. Several current overviews are beginning to include data on specific physical activity forms and cancer, such as Miles [3], the IARC Handbook [208], and the 2008 Physical Activity Guidelines [49]. In the realm of nutrition, although the relationship between nutrition and cancer has been studied for decades, there still appears to be many unanswered questions.
256
D. Supruijt-Metz et al.
The literature suggests that caloric restriction and increased intake of certain nutrients and foods, such as some polyunsaturated fatty acids, complex carbohydrates, fish, dairy, fruits and vegetables, dietary fiber, and soy, are protective against a variety of cancers. Other nutrients and foods, such as total and saturated fat, red meat, refined grains and added sugar, may actually promote cancer risk. There is substantial evidence that weight loss can reduce cancer risk. In this chapter, we highlighted several strategies that have been successful at promoting and maintaining weight loss, including decreasing fast-food consumption, decreasing portion sizes, increasing meal frequency, consuming breakfast regularly, decreasing sugar-sweetened beverage intake, and reducing TV time, especially eating in front of the TV. The literature showing that dietary interventions reduce cancer risk is still fairly sparse and inconclusive. It remains unclear whether a low-fat diet can reduce cancer risk. A low-fat approach in combination with increased dietary fiber and decreased sugar may result in more substantial weight loss and cancer risk reduction. In the realm of sleep, there is much to be learned about the links with obesity and cancer. Findings thus far highlight the need for continuing exploration of these associations. Clearly, sleep duration has a link to diet, physical activity, and obesity, which all have associations with cancer risk. Further, studies of circadian disruption have provided plausible physiological mechanisms by which sleep patterns may be related to development of cancer. Although research has shown that sleep is indeed a novel modifiable risk factor, we have yet to show empirical evidence that improving sleep can reduce risk of cancer and obesity. Considering that diet and physical activity interventions have not been unequivocally successful at reducing weight, body fat, or BMI, incorporating sleep and sleep hygiene into these intervention efforts might be useful. Overall, interventions to change behavior related to obesity, insulin sensitivity, inflammation, other cancer biomarkers, and other factors associated with cancer risk have been moderately successful. There is much to be learned about which specific behaviors, levels, and doses will lead to clinically relevant responses, and how best to help people make healthy behavioral choices. However, lifestyle change remains the key to obesity prevention and cancer risk reduction.
References 1. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM (2006). Prevalence of overweight and obesity in the United States, 1999–2004. JAMA, 295:1549–1555 2. Rogers CJ, Colbert LH, Greiner JW, Perkins SN, Hursting SD (2008). Physical activity and cancer prevention : pathways and targets for intervention. Sports Med, 38:271–296 3. Miles L (2007). Physical activity and the prevention of cancer: a review of recent findings. Nutr Bull, 32:250–282 4. Clavel-Chapelon F, Niravong M, Joseph RR (1997). Diet and breast cancer: review of the Epidemiologic literature. Cancer Detect Prevent, 21:426–440 5. La Vecchia C (2001). Diet and human cancer: a review. Eur J Cancer Prev, 10:177–181 6. Michels KB, Mohllajee AP, Roset-Bahmanyar E, Beehler GP, Moysich KB (2007). Diet and breast cancer: a review of the prospective observational studies. Cancer, 109: 2712–2749
9
Behavior, Energy Balance, and Cancer: An Overview
257
7. Ryan-Harshman M, Aldoori W (2007). Diet and colorectal cancer: review of the evidence. Can Fam Physician, 53:1913–1920 8. Tsugane S, Sasazuki S (2007). Diet and the risk of gastric cancer: review of Epidemiological evidence. Gastric Cancer, 10:75–83 9. Centers for Disease Control and Prevention (2008). Mortality and morbidity weekly report, 57 10. Cleland RL, Gross WC, Koss LD, Daynard M, Muoio KM (2002). Weight-Loss Advertising: An Analysis of Current Trends. Federal Trade Commission, Rockville, MD 11. Lewis BA, Marcus BH, Pate RR, Dunn AL (2002). Psychosocial mediators of physical activity behavior among adults and children. Am J Prev Med, 23:26–35 12. Holtzman J, Schmitz K, Babes G, et al. (2004). Effectiveness of behavioral interventions to modify physical activity behaviors in general populations and cancer patients and survivors. In: Quality AfHRa, ed. Summary, Evidence Report/Technology Assessment No. 102. Rockville, MD: Prepared by the University of Minnesota Evidence-based Practice Center, under contract no. 290-02-0009 13. Spruijt-Metz D, Saelens B (2005). Behavioral aspects of physical activity in childhood and adolescence. In: Goran MI, Southern M, (eds.), Handbook of Pediatric Obesity: Etiology, Pathophysiology and Prevention, pp. 227–250. Taylor & Francis Books/CRC Press, Boca Raton, FL 14. Summerbell CD, Waters E, Edmunds LD, Kelly S, Brown T, Campbell KJ (2005). Interventions for preventing obesity in children [update of Cochrane Database Syst Rev. PMID: 12076426]. Cochrane Database Syst Rev: CD001871. 15. Sallis JF, Prochaska JJ, Taylor WC (2000). A review of correlates of physical activity of children and adolescents. Med Sci Sports Exerc, 32:963–975 16. Wendel-Vos W, Droomers M, Kremers S, Brug J, van Lenthe F (2007). Potential environmental determinants of physical activity in adults: a systematic review obesity reviews, Early online articles 17. Trost SG, Owen N, Bauman AE, Sallis JF, Brown W (2002). Correlates of adults’ participation in physical activity: review and update. Med Sci Sports Exerc, 34:1996 18. Shaikh AR, Yaroch AL, Nebeling L, Yeh MC, Resnicow K (2008). Psychosocial predictors of fruit and vegetable consumption in adults a review of the literature. Am J Prev Med, 34:535–543 19. Blanchette L, Brug J (2005). Determinants of fruit and vegetable consumption among 6-12year-old children and effective interventions to increase consumption. J Human Nutr Diet, 18:431 20. McClain A, Chappuis C, Nguyen-Rodriguez ST, Yaroch AL, Spruijit-Metz D (2009). Psychosocial correlates of eating behavior in children and adolescents: A review Int J Behav Nuts Phys Act, 6(1):54 21. Gregory AM, Rijsdijk FV, Dahl RE, McGuffin P, Eley TC (2006). Associations between sleep problems, anxiety, and depression in twins at 8 years of age. Pediatrics, 118:1124 22. Meltzer LJ, Mindell JA (2007). Relationship between child sleep disturbances and maternal sleep, mood, and parenting stress: a pilot study. J Fam Psychol, 21:67–73 23. Newman AB, Enright PL, Manolio TA, Haponik EF, Wahl PW (1997). Sleep disturbance, psychosocial correlates, and cardiovascular disease in 5201 older adults: the cardiovascular health study. J Am Geriat Soc, 45:1–7 24. Lewis LB, Sloane DC, Nascimento LM, et al. (2005). African Americans’ access to healthy food options in South Los Angeles restaurants. Am J Public Health, 95:668–673 25. Cohen DA, McKenzie TL, Sehgal A, Williamson S, Golinelli D, Lurle N (2007). Contribution of public parks to physical activity. Am J Public Health, 97:509–514 26. Frank LD, Schmid TL, Sallis JF, Chapman J, Saelens BE (2005). Linking objectively measured physical activity with objectively measured urban form: findings from SMARTRAQ. Am J PrevMed, 28:117–125
258
D. Supruijt-Metz et al.
27. Dunton GF, Kaplan J, Wolch J, Jerrett M, Reynolds KD (2009). Physical environmental correlates of childhood obesity: a systematic review. Obes Rev, 10(4):393–402 28. Papas MA, Alberg AJ, Ewing R, Helzlsouer KJ, Gary TL, Klassen AC (2007). The built environment and obesity. Epidemiol Rev, 29:129–243 29. Fair AM, Montgomery K (2009). Energy balance, physical activity, and cancer risk. Method Molecul Biol, 472:57–88 30. USDHHS (1996). Physical Activity and Health: A Report of the Surgeon General. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Atlanta, GA 31. Spruijt-Metz D, Berrigan D, Kelly LA, et al (2009). Measures of physical activity and exercise. In: Allison DB, Baskin ML, (eds.), Handbook Of Assessment Methods For Eating Behaviors And Weight Related Problems: Measures, Theory And Research. Sage PUblications, Inc., Thousand Oaks, CA pp. 187–254 32. Ainsworth BE, Haskell WL, Leon AS, et al. (1993). Compendium of physical activities: classification of energy costs of human physical activities. Med Sci Sports & Exerc, 25: 71–80 33. Ainsworth BE, Haskell WL, Whitt MC, et al. (2000). Compendium of physical activities: an update of activity codes and MET intensities. Med Sci Sports Exerc, 32:S498–S504 34. Spanier PA, Marshall SJ, Faulkner GE (2006). Tackling the obesity pandemic: a call for sedentary behaviour research. Can J Public Health, 97:255–257 35. Shields M, Tremblay MS (2008). Sedentary behaviour and obesity. Health Rep 19:19–30 36. Gierach GL, Chang SC, Brinton LA, et al. (2008). Physical activity, sedentary behavior, and endometrial cancer risk in the NIH-AARP Diet and Health Study. Int J Cancer 37. Pate RR, Pratt M, Blair SN, et al. (1995). Physical activity and public health. A recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. JAMA, 273:402–407 38. Evenson KR, Huston SL, McMillen BJ, Bors P, Ward DS (2003). Statewide prevalence and correlates of walking and bicycling to school. Arch Pediatr Adolesc Med, 157:887–892 39. Hou L, Ji BT, Blair A, Dai Q, Gao YT, Chow WH (2004). Commuting physical activity and risk of colon cancer in Shanghai, China. Am J Epidemiol, 160:860–867 40. McTiernan A, Kooperberg C, White E, et al. (2003). Recreational physical activity and the risk of breast cancer in postmenopausal women: the women’s health initiative cohort study [see comment]. JAMA, 290:1331–1336 41. Steindorf K, Schmidt M, Kropp S, Chang-Claude J (2003). Case–control study of physical activity and breast cancer risk among premenopausal women in Germany. Am J Epidemiol, 157:121–130 42. McTiernan A, Kooperberg C, White E, et al. (2003). Recreational physical activity and the risk of breast cancer in postmenopausal women. JAMA, 290:1331–1336 43. Zhang Y, Cantor K, Dosemeci M, Lynch C, Zhu Y, Zheng T (2006). CME occupational and leisure-time physical activity and risk of colon cancer by subsite. J Occup Environ Med, 48:236 44. Thune I, Brenn T, Lund E, Gaard M (1997). Physical activity and the risk of breast cancer. New Engl J Med, 336:1269–1275 45. Lahmann PH, Friedenreich C, Schuit AJ, et al. (2007). Physical activity and breast cancer risk: the European prospective investigation into cancer and nutrition. Cancer Epidemiol Biomark Prev, 16:36–42 46. Steindorf K, Friedenreich C, Linseisen J, et al. (2006). Physical activity and lung cancer risk in the European prospective investigation into cancer and nutrition cohort. Int J Cancer, 119:2389–2397 47. Sesso HD, Paffenbarger RS, Jr., Lee IM (2000). Physical activity and coronary heart disease in men: the Harvard alumni health study. Circulation, 102:975–980 48. Paffenbarger Jr R, Lee I, Wing A (1992). The influence of physical activity on the incidence of site-specific cancers in college alumni. Adv Exp Med Biol, 322:7–15
9
Behavior, Energy Balance, and Cancer: An Overview
259
49. Physical Activity Guidelines Advisory Committee (2008). Physical Activity Guidelines Advisory Committee Report. Department of Health and Human Services, Washington, DC 50. Schmitz KH, Ahmed RL, Yee D (2002). Effects of a 9-Month Strength Training Intervention on Insulin, Insulin-like Growth Factor (IGF)-I, IGF-binding Protein (IGFBP)-1, and IGFBP-3 in 30-50-Year-Old Women. Cancer Epidemiology, Biomarkers and Prevention, 11:1597–1604 51. Patel AV, Rodriguez C, Pavluck AL, Thun MJ, Calle EE (2006). Recreational physical activity and sedentary behavior in relation to ovarian cancer risk in a large cohort of US women. Am J Epidemiol, 163:709–716 52. Livingstone MBE., Robson PJ (2000). Measurement of dietary intake in children. Proceedings of the Nutrition Society, 59:279–293 53. Thompson FE, Subar A (2001). Dietary assessment methodology. In: Coulston AM, Boushey C, (eds.), Nutrition in the Prevention and Treatment of Disease, pp. 3–39. Boston Academic Press/Elsevier, Amsterdam 54. Lee G, Tsai C, Griswold W, Raab F, Patrick K (2006). PmEB: A Mobile Phone Application For Monitoring Caloric Balance, pp. 1013–1018. ACM, New York 55. Prentice RL, Sheppard L (1990). Dietary fat and cancer: consistency of the Epidemiologic data, and disease prevention that may follow from a practical reduction in fat consumption. Cancer Causes Control, 1:81–97, discussion 99–109 56. Schulz M, Hoffmann K, Weikert C, Nothlings U, Schulze MB, Boeing H (2008). Identification of a dietary pattern characterized by high-fat food choices associated with increased risk of breast cancer: the European prospective investigation into cancer and nutrition (EPIC)-Potsdam study. Br J Nutr, 100:942–946 57. Wang J, John EM, Horn-Ross PL, Ingles SA (2008). Dietary fat, cooking fat, and breast cancer risk in a multiethnic population. Nutr Cancer, 60:492–504 58. Howe GR, Hirohata T, Hislop TG, et al. (1990). Dietary factors and risk of breast cancer: combined analysis of 12 case–control studies. J Natl Cancer Inst, 82:561–569 59. Crowe FL, Key TJ, Appleby PN, et al. (2008). Dietary fat intake and risk of prostate cancer in the European prospective investigation into cancer and nutrition. Am J Clin Nutr 87: 1405–1413 60. Hunter DJ, Spiegelman D, Adami HO, et al. (1996). Cohort studies of fat intake and the risk of breast cancer–a pooled analysis. N Engl J Med, 334:356–361 61. Holmes MD, Hunter DJ, Colditz GA, et al. (1999). Association of dietary intake of fat and fatty acids with risk of breast cancer. JAMA, 281:914–920 62. Smith-Warner SA, Spiegelman D, Adami HO, et al. (2001). Types of dietary fat and breast cancer: a pooled analysis of cohort studies. Int J Cancer, 92:767–774 63. Decarli A, Favero A, La Vecchia C, et al. (1997). Macronutrients, energy intake, and breast cancer risk: implications from different models. Epidemiology, 8:425–428 64. Velie E, Kulldorff M, Schairer C, Block G, Albanes D, Schatzkin A (2000). Dietary fat, fat subtypes, and breast cancer in postmenopausal women: a prospective cohort study. J Natl Cancer Inst, 92:833–839 65. De Stefani E, Deneo-Pellegrini H, Mendilaharsu M, Ronco A (1998). Essential fatty acids and breast cancer: a case–control study in Uruguay. Int J Cancer, 76:491–494 66. Hursting SD, Thornquist M, Henderson MM (1990). Types of dietary fat and the incidence of cancer at five sites. Prev Med, 19:242–253 67. Caygill CP, Charlett A, Hill MJ (1996). Fat, fish, fish oil and cancer. Br J Cancer, 74:159–164 68. Kohlmeier L, Simonsen N, Margolin B, Thamm M (1995). Stores of trans fatty acids and breast cancer risk. Am J Clin Nutr, 61:896–902 69. Shannon J, Cook LS, Stanford JL (2003). Dietary intake and risk of postmenopausal breast cancer (United States). Cancer Causes Control, 14:19–27
260
D. Supruijt-Metz et al.
70. Hirose K, Takezaki T, Hamajima N, Miura S, Tajima K (2003). Dietary factors protective against breast cancer in Japanese premenopausal and postmenopausal women. Int J Cancer, 107:276–282 71. Dale D, Corbin CB, Dale KS (2000). Restricting opportunities to be active during school time: do children compensate by increasing physical activity levels after school? Res Quart Exerc Sport, 71:240–248 72. Lee HP, Gourley L, Duffy SW, Esteve J, Lee J, Day NE (1992). Risk factors for breast cancer by age and menopausal status: a case–control study in Singapore. Cancer Causes Cont, 3:313–322 73. Yuan JM, Wang QS, Ross RK, Henderson BE, Yu MC (1995). Diet and breast cancer in Shanghai and Tianjin, China. Br J Cancer, 71:1353–1358 74. Trock BJ, Hilakivi-Clarke L, Clarke R (2006). Meta-analysis of soy intake and breast cancer risk. J Natl Cancer Inst, 98:459–471 75. Franceschi S, Parpinel M, La Vecchia C, Favero A, Talamini R, Negri E (1998). Role of different types of vegetables and fruit in the prevention of cancer of the colon, rectum, and breast. Epidemiology, 9:338–341 76. Adzersen KH, Jess P, Freivogel KW, Gerhard I, Bastert G (2003). Raw and cooked vegetables, fruits, selected micronutrients, and breast cancer risk: a case–control study in Germany. Nutr Cancer, 46:131–137 77. Malin AS, Qi D, Shu XO, et al. (2003). Intake of fruits, vegetables and selected micronutrients in relation to the risk of breast cancer. Int J Cancer, 105:413–418 78. Howe GR, Benito E, Castelleto R, et al. (1992). Dietary intake of fiber and decreased risk of cancers of the colon and rectum: evidence from the combined analysis of 13 case–control studies. J Natl Cancer Inst, 84:1887–1896 79. Suzuki R, Rylander-Rudqvist T, Ye W, Saji S, Adlercreutz H, Wolk A (2008). Dietary fiber intake and risk of postmenopausal breast cancer defined by estrogen and progesterone receptor status–a prospective cohort study among Swedish women. Int J Cancer, 122:403–412 80. Bingham SA, Norat T, Moskal A, et al. (2005). Is the association with fiber from foods in colorectal cancer confounded by folate intake? Cancer Epidemiol Biomark Prev, 14: 1552–1556 81. Schatzkin A, Park Y, Leitzmann MF, Hollenbeck AR, Cross AJ (2008). Prospective study of dietary fiber, whole grain foods, and small intestinal cancer. Gastroenterology, 135: 1163–1167 82. World Cancer Research Fund (1997). Food, nutrition, and the prevention of cancer: a global perspective. American Institute for Cancer Research, Washington, DC 83. Slattery ML, Benson J, Berry TD, et al. (1997). Dietary sugar and colon cancer. Cancer Epidemiol Biomarkers Prev, 6:677–685 84. Potischman N, Coates RJ, Swanson CA, et al. (2002). Increased risk of early-stage breast cancer related to consumption of sweet foods among women less than age 45 in the United States. Cancer Causes Cont, 13:937–946 85. Wolever TM, Jenkins DJ, Jenkins AL, Josse RG (1991). The glycemic index: methodology and clinical implications. Am J Clin Nutr, 54:846–854 86. Foster-Powell K, Holt SHA, Brand-Miller JC (2002). International table of glycemic index and glycemic load values. Am J Clin Nutr, 76:5–56 87. Michaud DS, Liu S, Giovannucci E, Willett WC, Colditz GA, Fuchs CS (2002). Dietary sugar, glycemic load, and pancreatic cancer risk in a prospective study. J Natl Cancer Inst, 94:1293–1300 88. Navarro Silvera SA, Jain M, Howe GR, Miller AB, Rohan TE (2005). Dietary carbohydrate and breast cancer risk: a prospective study of roles of overall glycemic index and glycemic load. Int J Cancer, 114:653–658 89. Kantor LS (1998). A dietary assessment of the US food supply. Comparing per capital food consumption with food guide pyramid service recommendations. Food and
9
Behavior, Energy Balance, and Cancer: An Overview
90.
91. 92.
93. 94.
95.
96. 97. 98. 99. 100.
101. 102. 103. 104. 105. 106. 107. 108.
109.
110.
261
Rural Economics Division, Economics Research Service, U.S. Department of Agriculture, Agricultural Economic Report no. 772 1998: Government Printing Office Washington, DC Johnson RK, Frary C (2001). Choose beverages and foods to moderate your intake of sugars: the 2000 dietary guidelines for Americans–what’s all the fuss about? J Nutr, 131: 2766S–2771S Krebs-Smith SM (2001). Choose beverages and foods to moderate your intake of sugars: measurement requires quantification. J Nutr, 131:527S–535S McKeown NM, Meigs JB, Lui S, Saltzman E, Wilson PWF, Jacques PF (2004). Carbohydrate nutrition, insulin resistance, and the prevalence of the metabolic syndrome in the Framingham offspring cohort. Diabet Care, 27:538–546 Pereira MA, Jacobs DR, Jr., Pins JJ, et al. (2002). Effect of whole grains on insulin sensitivity in overweight hyperinsulinemic adults. Am J Clin Nutr, 75:848–855 Ludwig DS, Pereira MA, Kroenke CH, Van Horn L, Slattery ML, Jacobs DR (1999). Dietary fiber, weight gain, and cardiovascular disease risk factors in young adults. JAMA, 282: 1539–1546 Davis JN, Alexander KE, Ventura EE, et al. (2007). Associations of dietary sugar and glycemic index with adiposity and insulin dynamics in overweight Latino youth. Am J Clin Nutr, 86:1331–1338 Davis J, Ventura E, Weigensberg M, et al. (2005). The relation of sugar intake to beta-Cell function in overweight Latino children. Am J Clin Nutr, 82:1004–1010 Calle EE, Kaaks R. (2004). Overweight, obesity and cancer: Epidemiological evidence and proposed mechanisms. Nat Rev Cancer, 4:579–591 AARC-APT (1995). AARC-APT clinical practice guideline: Polysomnography. Respir Care, 40:1336–1343 Spilsbury JC, Storfer-Isser A, Drotar D, et al. (2004). Sleep behavior in an urban US sample of school-aged children. Archives of Pediatrics & Adolescent Medicine 158:988–994 LeBourgeois MK, Giannotti F, Cortesi F, Wolfson AR, Harsh J. The relationship between reported sleep quality and sleep hygiene in Italian and American adolescents. Pediatrics 2005;115:257–265 Wolfson AR, Carskadon MA, Acebo C, et al. (2003). Evidence for the validity of a sleep habits survey for adolescents. Sleep 26:213–216 Carpenter JS, Andrykowski MA (1998). Psychometric evaluation of the pittsburgh Sleep quality index. J Psychos Res, 45:5–13 Johns MW (1992). Reliability and factor analysis of the Epworth Sleepiness scale. Sleep, 15:376–381 Knutson KL, Turek FW (2006). The U-shaped association between Sleep and health: the 2 peaks do not mean the same thing. Sleep, 29:878–879 Kripke DF, Garfinkel L, Wingard DL, Klauber MR, Marler MR (2002). Mortality associated with Sleep duration and insomnia [see comment]. Arch GenPsychiatry, 59:131–136 Kripke DF, Simons RN, Garfinkel L, Hammond EC (1979). Short and long Sleep and Sleeping pills. Is increased mortality associated? Arch Gen Psychiatry, 36:103–116 Tamakoshi A, Ohno Y, Group JS (2004). Self-reported Sleep duration as a predictor of allcause mortality: results from the JACC study, Japan [see comment]. Sleep, 27:51–54 Shankar A, Koh WP, Yuan JM, Lee HP, Yu MC (2008). Sleep duration and coronary heart disease mortality among Chinese adults in Singapore: a population-based cohort study. Am J Epidemiol, 168:1367–1373 Mallon L, Broman JE, Hetta J (2002). Sleep complaints predict coronary artery disease mortality in males: a 12-year follow-up study of a middle-aged Swedish population. J Intern Med, 251:207–216 Dew MA, Hoch CC, Buysse DJ, et al. (2003). Healthy older adults’ Sleep predicts all-cause mortality at 4 to 19 Years of Follow-Up. Am Psychosomatic Soc, :63–73
262
D. Supruijt-Metz et al.
111. Taheri S, Lin L, Austin D, Young T, Mignot E (2004). Short Sleep duration is associated with reduced Leptin, Elevated Ghrelin, and increased body mass index. PLOS MEDICINE, 1:210 112. Vioque J, Torres A, Quiles J (2000). Time spent watching television, Sleep duration and obesity in adults living in Valencia, Spain. Int J Obesity, 24:1683–1688 113. Locard E, Mamelle N, Billette A, Miginiac M, Munoz F, Rey S (1992). Risk factors of obesity in a five year old population. Parental versus environmental factors. Int J Obes Relat Metab Disord, 16:721–729 114. Sekine M, Yamagami T, Handa K, et al. (2002). A dose-response relationship between short Sleeping hours and childhood obesity: results of the Toyama birth cohort study. Child Care Health Develop, 28:163–170 115. Berkey CS, Rockett HRH, Colditz GA (2008). Weight gain in older adolescent females: the internet, Sleep, coffee, and alcohol. J Pediatr, 153(5):635–639 116. Gupta NK, Mueller WH, Chan W, Meininger JC (2002). Is obesity associated with poor Sleep quality in adolescents? Am J Hum Biol, 14:762–768 117. Ievers-Landis CE, Heinberg L, Donovan LM, et al. (2008). Feasibility of a Sleep intervention for adolescents who are obese. Transdisciplinary research on energetics and cancer centers’ scientific meeting, Bethesda, MD 118. Reilly JJ, Armstrong J, Dorosty AR, et al. (2005). Early life risk factors for obesity in childhood: cohort study. Br Med J, 330:1357 119. Agras WS, Hammer LD, McNicholas F, Kraemer HC (2004). Risk factors for childhood overweight: a prospective study from birth to 9.5 years. J Pediatr, 145:20–25 120. Lumeng JC, Somashekar D, Appugliese D, Kaciroti N, Corwyn RF, Bradley RH (2007). Shorter Sleep duration is associated with increased risk for Being overweight at ages 9 to 12 years. Pediatrics, 120:1020–1029 121. Spiegel K, Leproult R, L’Hermite-Baleriaux M, Copinschi G, Penev PD, Van Cauter E (2004). Leptin levels are dependent on Sleep duration: relationships with sympathovagal balance, Carbohydrate regulation, cortisol, and thyrotropin. Endocrine Soc, 5762–5771 122. Spiegel K, Leproult R, Tasali E, Penev P, Van Cauter E (2003). Sleep curtailment results in decreased leptin levels and increased hunger and appetite. Sleep, 26:A174 123. Imaki M, Hatanaka Y, Ogawa Y, Yoshida Y, Tanada S (2002). An Epidemiological study on relationship between the hours of Sleep and life style factors in Japanese factory workers. J Physiol Anthropol Appl Human Sci, 21:115–120 124. Spiegel K, Tasali E, Penev P, Cauter EV (2004). Brief communication: Sleep curtailment in healthy young men is associated with decreased leptin levels, elevated ghrelin levels, and increased hunger and appetite. Annals Intern Med, 141:846–850 125. Littman AJ, Vitiello MV, Foster-Schubert K, et al. (2007). Sleep, ghrelin, leptin and changes in body weight during a 1-year moderate-intensity physical activity intervention. Int J Obesity, 31:466–475 126. Harris KM, King RB, Gordon-Larsen P (2006). Healthy habits among adolescents: Sleep, exercise, diet, and body image. In: Moore KA, Lippman LH (eds.), What Do Children Need to Flourish? pp. 111–132. Springer, US 127. Montgomery I, Trinder J, Paxton SJ (1982). Energy expenditure and total Sleep time: effect of physical exercise. Sleep, 5:159–168 128. Youngstedt SD, O’Connor PJ, Dishman RK (1997). The effects of acute exercise on Sleep: a quantitative synthesis. Sleep, 20:203–214 129. Chaput JP, Després JP, Bouchard C, Tremblay A (2008). The association between Sleep duration and weight gain in adults: a 6-year prospective study from the Quebec family study. Sleep, 31:517–523 130. van den Berg JF, Neven AK, Tulen JHM, et al. (2008). Actigraphic Sleep duration and fragmentation are related to obesity in the elderly: the Rotterdam study. Int J Obes, 32(7):1083–1090
9
Behavior, Energy Balance, and Cancer: An Overview
263
131. Davis S, Mirick DK (2006). Circadian disruption, shift work and the risk of cancer: a summary of the evidence and studies in Seattle. Cancer Causes Cont, 17:539–545 132. Hansen J (2001). Increased breast cancer risk among women who work predominantly at night. Epidemiology, 12:74 133. Stevens RG, Rea MS (2001). Light in the built environment: potential role of circadian disruption in endocrine disruption and breast cancer. Cancer Causes Cont, 12:279–287 134. Irwin M (2001). Neuroimmunology of disordered Sleep in depression and alcoholism. Neuropsychopharmacology, 25:45–49 135. Vgontzas AN, Chrousos GP (2002). Sleep, the hypothalamic–pituitary–adrenal axis, and cytokines: multiple interactions and disturbances in Sleep disorders. Endocrin Metabol Clin North Am, 31:15–36 136. Davis S, Mirick DK, Stevens RG (2001). Night Shift Work, Light at Night, and Risk of Breast Cancer, pp. 1557–1562. Oxford University Press, Oxford 137. Schernhammer ES, Laden F, Speizer FE, et al. (2001). Rotating Night Shifts and Risk of Breast Cancer in Women Participating in the Nurses’ Health Study, pp. 1563–1568. Oxford University Press, Oxford 138. Schernhammer ES, Laden F, Speizer FE, et al. (2003). Night-Shift Work and Risk of Colorectal Cancer in the Nurses’ Health Study, pp. 825–828. Oxford University Press, Oxford 139. Reynolds P, Cone J, Layefsky M, Goldberg DE, Hurley S (2002). Cancer incidence in California flight attendants (United States). Cancer Causes Cont, 13:317–324 140. Pukkala E, Aspholm R, Auvinen A, et al. (2002). Incidence of cancer among Nordic airline pilots over five decades: occupational cohort study. Br Med J, 325:567 141. Rafnsson V, Hrafnkelsson J, Tulinius H (2000). Incidence of cancer among commercial airline pilots. Occup Environ Med, 57:175–179 142. Yamanouchi K, Shinozaki T, Chikada K, et al. (1995). Daily walking combined with diet therapy is a useful means for obese NIDDM patients not only to reduce body weight but also to improve insulin sensitivity. Diabet Care, 18:775 143. Schrauwen P, van Aggel-Leijssen DPC, Hul G, et al. (2002). The effect of a 3-month low-intensity endurance training program on fat oxidation and Acetyl-CoA Carboxylase-2 expression. Diabetes, 51:2220–2226 144. Irwin ML, Yasui Y, Ulrich CM, et al. (2003). Effect of exercise on total and intra-abdominal body fat in postmenopausal women: a randomized controlled trial. JAMA, 289:323–330 145. Davis JN, Tung A, Chak S, et al (2009). Aerobic and strength training reduces adiposity in overweight Latina adolescents. Med Sci Sports Exerc, 41(7):1494–1503 146. Mo-suwan L, Pongprapai S, Junjana C, Puetpaiboon A (1998). Effects of a controlled trial of a school-based exercise program on the obesity indexes of preschool children. Am J Clin Nutr, 68:1006–1011 147. Goon J, Aini A, Musalmah M, Anum M, Nazaimoon W, Ngah W (2009). Effect of Tai Chi exercise on DNA damage, antioxidant enzymes and oxidative stress in middle aged adults. J Phys Act Health, 6:43–54 148. Spruijt-Metz D, Nguyen-Michel ST, Goran MI, Chou CP, Huang TT (2008). Reducing sedentary behavior in minority girls via a theory-based, tailored classroom media intervention. Int J Pediatr Obes, 3:240–248 149. Gortmaker SL, Peterson K, Wiecha J, et al. (1999). Reducing obesity via a school-based interdisciplinary intervention among youth: planet health. Arch Pediat Adol Med, 153: 409–418 150. Lemmens VE, Oenema A, Klepp KI, Henriksen HB, Brug J (2008). A systematic review of the evidence regarding efficacy of obesity prevention interventions among adults. Obes Rev, 9:446–455 151. Howard BV, Van Horn L, Hsia J, et al. (2006). Low-fat dietary pattern and risk of cardiovascular disease: the women’s health initiative randomized controlled dietary modification trial. JAMA, 295:655–666
264
D. Supruijt-Metz et al.
152. Kamath C, Vickers K, Ehrlich A, et al. (2008). Behavioral interventions to prevent childhood obesity: a systematic review and metaanalyses of randomized trials. J Clin Endocrinol Metab, 93:4606 153. Sharma M (2006). School-based interventions for childhood and adolescent obesity. Obesity Rev, 7:261–269 154. Li F, Harmer PA, Cardinal BJ, et al. (2008). Built environment, adiposity, and physical activity in adults aged 50–75. Am J Prev Med, 35:38–46 155. Bowman SA, Gortmaker SL, Ebbeling CB, Pereira MA, Ludwig DS (2004). Effects of fastfood consumption on energy intake and diet quality among children in a national household survey. Pediatrics, 113:112–118 156. Klohe-Lehman DM, Freeland-Graves J, Clarke KK, et al. (2007). Low-income, overweight and obese mothers as agents of change to improve food choices, fat habits, and physical activity in their 1-to-3-year-old children. J Am Coll Nutr, 26:196–208 157. Davis JN, Ventura EE, Alexander KA, et al. (2007). Development and testing of a culturally tailored nutrition education program for reducing sugar and increasing fiber intake in overweight Latina adolescents. Int J Ped Obes, 2:22–30 158. Summerbell C, Ashton V, Campbell K, Edmunds L, Kelly S, Waters E (2003). Interventions for preventing obesity in children. Cochrane Database Syst Rev, 3:CD001872 159. Fisher JO, Kral TV (2008). Super-size me: Portion size effects on young children’s eating. Physiol Behav, 94:39–47 160. Cohen D, Farley TA (2008). Eating as an automatic behavior. Prev Chronic Dis, 5:A23 161. Rolls BJ, Roe LS, Kral TV, Meengs JS, Wall DE (2004). Increasing the portion size of a packaged snack increases energy intake in men and women. Appetite, 42:63–69 162. Pedersen SD, Kang J, Kline GA (2007). Portion control plate for weight loss in obese patients with type 2 Diabetes mellitus: a controlled clinical trial. Arch Intern Med, 167:1277–1283 163. Speechly DP, Rogers GG, Buffenstein R (1999). Acute Appetite reduction associated with an increased frequency of eating in obese males. Int J Obes Relat Metab Disord, 23:1151–1159 164. Kirk TR (2000). Role of dietary carbohydrate and frequent eating in body-weight control. Proc Nutr Soc, 59:349–358 165. Bellisle F, McDevitt R, Prentice AM (1997). Meal frequency and energy balance. Br J Nutr, 77 Suppl 1:S57–S70 166. Ma Y, Bertone ER, Stanek EJ, 3rd, et al. (2003). Association between eating patterns and obesity in a free-living US adult population. Am J Epidemiol, 158:85–92 167. Schlundt DG, Sbrocco T, Bell C (1989). Identification of high-risk situations in a behavioral weight loss program: application of the relapse prevention model. Int J Obes, 13:223–134 168. Cho S, Dietrich M, Brown CJ, Clark CA, Block G (2003). The effect of breakfast type on total daily energy intake and body mass index: results from the Third National Health and Nutrition Examination Survey (NHANES III). J Am Coll Nutr, 22:296–302 169. Siega-Riz AM, Popkin BM, Carson T (1998). Trends in breakfast consumption for children in the United States from 1965–1991. Am J Clin Nutr, 67:748S–756S 170. Stanton JL, Jr., Keast DR (1989). Serum cholesterol, fat intake, and breakfast consumption in the United States adult population. J Am Coll Nutr, 8:567–572 171. Morgan KJ, Zabik ME, Stampley GL (1986). The role of breakfast in diet adequacy of the U.S. adult population. J Am Coll Nutr, 5:551–563 172. Keim NL, Van Loan MD, Horn WF, Barbieri TF, Mayclin PL (1997). Weight loss is greater with consumption of large morning meals and fat-free mass is preserved with large evening meals in women on a controlled weight reduction regimen. J Nutr, 127:75–82 173. Ludwig DS, Majzoub JA, Al-Zahrani A, Dallal GE, Blanco I, Roberts SB (1999). High glycemic index foods, overeating, and obesity. Pediatrics, 103:E26 174. Warren JM, Henry CJ, Simonite V (2003). Low glycemic index breakfasts and reduced food intake in preadolescent children. Pediatrics, 112:e414 175. Phillips SM, Bandini LG, Naumova EN, et al. (2004). Energy-dense snack food intake in adolescence: longitudinal relationship to weight and fatness. Obes Res, 12:461–472
9
Behavior, Energy Balance, and Cancer: An Overview
265
176. Ludwig DS, Peterson KE, Gortmaker SL (2001). Relation between consumption of sugarsweetened drinks and childhood obesity: a prospective, observational analysis. Lancet, 357:505–508 177. Berkey CS, Rockett HR, Field AE, Gillman MW, Colditz GA (2004). Sugar-added beverages and adolescent weight change. Obes Res, 12:778–788 178. Troiano RP, Briefel RR, Carroll MD, Bialostosky K (2000). Energy and fat intakes of children and adolescents in the united states: data from the national health and nutrition examination surveys. Am J Clin Nutr, 72:1343S–1353S 179. Schulze MB, Manson JE, Ludwig DS, et al. (2004). Sugar-sweetened beverages, weight gain, and incidence of type 2 Diabetes in young and middle-aged women. JAMA, 292: 927–934 180. Apovian CM (2004). Sugar-sweetened soft drinks, obesity, and type 2 Diabetes. JAMA, 292:978–979 181. Ventura EE, Davis JN, Byrd-Williams C, et al (2009). Reduction in risk factors for type 2 Diabetes in response to a low-sugar, high-fiber dietary intervention in overweight Latino youth. Arch Pediatr Adolesc Med, 163(4):320–327 182. Eisenmann JC, Bartee RT, Wang MQ (2002). Physical activity, TV viewing, and weight in U.S. youth: 1999 youth risk behavior survey. Obes Res, 10:379–385 183. Robinson TN (1999). Reducing children’s television viewing to prevent obesity: a randomized controlled trial. JAMA, 282:1561–1567 184. French SA, Story M, Neumark-Sztainer D, Fulkerson JA, Hannan P (2001). Fast food restaurant use among adolescents: associations with nutrient intake, food choices and behavioral and psychosocial variables. Int J Obes Relat Metab Disord, 25:1823–1833 185. Blass EM, Anderson DR, Kirkorian HL, Pempek TA, Price I, Koleini MF (2006). On the road to obesity: Television viewing increases intake of high-density foods. Physiol Behav, 88:597–604 186. Coon KA, Goldberg J, Rogers BL, Tucker KL (2001). Relationships between use of television during meals and children’s food consumption patterns. Pediatrics, 107:E7 187. Rehm CD, Matte TD, Van Wye G, Young C, Frieden TR (2008). Demographic and behavioral factors associated with daily sugar-sweetened soda consumption in New York City adults. J Urban Health, 85:375–385 188. Ross MH, Bras G (1971). Lasting influence of early caloric restriction on prevalence of neoplasms in the rat. J Natl Cancer Inst, 47:1095–1113 189. Boissonneault GA, Elson CE, Pariza MW (1986). Net energy effects of dietary fat on chemically induced mammary carcinogenesis in F344 rats. J Natl Cancer Inst, 76:335–338 190. Reddy BS, Wang CX, Maruyama H (1987). Effect of restricted caloric intake on azoxymethane-induced colon tumor incidence in male F344 rats. Cancer Res, 47:1226–1228 191. Steinbach G, Heymsfield S, Olansen NE, Tighe A, Holt PR (1994). Effect of caloric restriction on colonic proliferation in obese persons: implications for colon cancer prevention. Cancer Res, 54:1194–1197 192. Elias SG, Peeters PH, Grobbee DE, van Noord PA (2005). The 1944–1945 Dutch famine and subsequent overall cancer incidence. Cancer Epidemiol Biomarkers Prev, 14:1981–1985 193. Boyd NF, Greenberg C, Lockwood G, et al. (1997). Effects at two years of a low-fat, high-carbohydrate diet on radiologic features of the breast: results from a randomized trial. Canadian diet and breast cancer prevention study group. J Natl Cancer Inst, 89:488–496 194. Taylor RW, McAuley KA, Barbezat W, Strong A, Williams SM, Mann JI (2007). APPLE Project: 2-y findings of a community-based obesity prevention program in primary school age children. Am J Clin Nutr, 86:735–742 195. Sahota P, Rudolf MC, Dixey R, Hill AJ, Barth JH, Cade J (2001). Randomised controlled trial of primary school based intervention to reduce risk factors for obesity. Bmj, 323: 1029–1032 196. Epstein LH, Gordy CC, Raynor HA, Beddome M, Kilanowski CK, Paluch R (2001). Increasing fruit and vegetable intake and decreasing fat and sugar intake in families at risk for childhood obesity. Obes Res, 9:171–178
266
D. Supruijt-Metz et al.
197. Carruba G, Granata O, Pala V, et al. (2006). A traditional mediterranean diet decreases endogenous estrogens in healthy postmenopausal women. Nutr Cancer, 56:253–259 198. Shai I, Schwarzfuchs D, Henkin Y, et al. (2008). Weight loss with a low-carbohydrate, Mediterranean, or low-fat diet. N Engl J Med, 359:229–241 199. Lu LJ, Anderson KE, Grady JJ, Kohen F, Nagamani M (2000). Decreased ovarian hormones during a soya diet: implications for breast cancer prevention. Cancer Res, 60:4112–4121 200. Kumar NB, Cantor A, Allen K, Riccardi D, Cox CE (2002). The specific role of isoflavones on estrogen metabolism in premenopausal women. Cancer, 94:1166–1174 201. Maskarinec G, Williams AE, Inouye JS, Stanczyk FZ, Franke AA (2002). A randomized isoflavone intervention among premenopausal women. Cancer Epidemiol Biomarkers Prev, 11:195–201 202. Morin CM (2004). Cognitive-behavioral approaches to the treatment of insomnia. J Clin Psychiatry, 65:33–40 203. Morin CM, Culbert JP, Schwartz SM (1994). Nonpharmacological Interventions for Insomnia. Am J Psychiat, 1:1173 204. Carlson LE, Garland SN (2005). Impact of Mindfulness-Based Stress Reduction (MBSR) on Sleep, mood, stress and fatigue symptoms in cancer outpatients. Int J Behav Med, 12: 278–285 205. Davison KK, Birch LL (2001). Childhood overweight: a contextual model and recommendations for future research. Obesity Rev, 2:159–171 206. Berger AM, VonEssen S, Khun BR, et al. (2002). Feasibilty of a Sleep intervention during adjuvant breast cancer chemotherapy. Oncology Nursing Forum, Online. 29:1431–1441 207. Shaibi GQ, Cruz ML, Ball GD, et al. (2006). Effects of resistance training on insulin sensitivity in overweight Latino adolescent males. Med Sci Sports Exer, 38:1208–1215 208. IARC W (2002). IARC handbooks of cancer prevention. Volume 6: Weight control and physical activity. IARC Press, Lyon
Chapter 10
Geographic and Contextual Effects on Energy Balance-Related Behaviors and Cancer David Berrigan, Robin McKinnon, Genevieve Dunton, Lan Huang, and Rachel Ballard-Barbash
Abstract This chapter concerns the analysis of geographic and contextual effects on health behaviors related to energy balance, such as diet, weight, and physical activity. We adopt a broad definition of the environment to include not just ambient exposures to toxins, meteorological conditions, or other traditional foci of environmental sciences but also the built, economic, social, and policy aspects of the environment. The chapter includes a selective review of research in this area concerning urban sprawl and obesity, built environment and physical activity, social context and energy balance, and the food environment. We then focus on ongoing challenges to advances in the analysis of contextual effects, including discussion of confounding self-selection as a barrier to causal inference, variable selection and spatial scan statistics, and the general problem of incorporating maps into research on health behavior variables. Experimental and longitudinal designs have not been applied extensively in this field and are required to examine causal associations and to quantify which types of interventions are likely to have the largest effect in specific situations.
1 Introduction Recent years have seen an intense interest in environmental, contextual, and geographic variation in health behaviors and in disease, including cancer [137, 151, 101, 163]. Interest in spatial patterns of disease and disease risk factors began with Snow’s analysis of cholera, the emergence of occupational health research, and nineteenth century mapping of infectious disease prevalence in different countries [163, 58]. Because of this historical element and because of the major influence of
D. Berrigan (B) Division of Cancer Control and Population Sciences, National Institutes of Health, National Cancer Institute, 6130 Executive Blvd, Bethesda, MD 20892, USA e-mail:
[email protected]
N.A. Berger (ed.), Cancer and Energy Balance, Epidemiology and Overview, Energy Balance and Cancer 2, DOI 10.1007/978-1-4419-5515-9_10, C Springer Science+Business Media, LLC 2010
267
268
D. Berrigan et al.
spatial characteristics on infectious disease, population dynamics and environmental toxins, spatial epidemiology, toxicology, and population biology have focused on infectious disease and environmental hazards [1, 45, 7]. However, more recently, this focus has expanded to include significant interest in geographic variation in chronic disease, and diverse risk factors related to such diseases as cancer, diabetes, and heart disease. Growth in this research area can be traced to at least three factors: (1) rapid growth in the use and sophistication of geographic tools, such as geographic information systems (GIS), global positioning systems (GPS), and the statistical approaches needed to exploit data generated by such tools; (2) advances in the fields of transportation and urban planning that have focused attention on the role of the built environment as a determinant of transportation mode choice; and (3) the failure of individually based approaches to lead to marked change at the population level in adverse health behaviors involving diet, weight, and physical activity. The study of spatial variation in risk factors and disease encompasses multiple disciplines, including geography, epidemiology, statistics, ecology, transportation and planning, economics, and sociology. This makes the topic both rich and potentially confusing, as the different disciplines have developed diverse vocabularies and conceptual frameworks to analyze and understand spatial variation in disease and disease risk factors. In this chapter we emphasize the application of research in these disciplines to understanding health behaviors related to cancer and energy balance. This area has seen particularly rapid growth because of epidemic increases in the prevalence of obesity in the United States and many other countries. Increases in investments in this area of research by public and private organizations as well as by municipalities seeking sources of information to inform planning decisions also have fostered growth in this area. The underlying logic of such research is predicated on an ecological model of health behaviors [106, 138] and a population approach to public health through prevention [136]. In the socioecological model, environmental and individual factors interact to influence health behaviors and in turn influence risk of disease (Fig. 10.1). Such a model can be conceptualized in linear fashion (Fig. 10.1a), which highlights the key point that environmental variables can have a strong influence on causal pathways reaching all the way to the molecular level. A major research goal in animal and human ecophysiology is to characterize such causal pathways. An alternative conceptualization of the ecological model (Fig. 10.1b) focuses on the diversity of the environment, highlighting that environmental variables include domains such as the home, school, family, and workplace as well as policy and economic variables [138, 151, 63]. The second diagram (Fig. 10.1b) has been widely used to motivate multilevel analyses of factors influencing health behavior and efforts to develop and improve measures of diverse aspects of the environment. Much past research on obesity and most contemporary research on cancer have focused on the individual level aspects of human disease with a strong concentration on biology and individual level demographic and clinical characteristics. Here “environmental” factors are defined to include not just exposures to toxins in the environment, weather, and the natural landscape, but also the built and
10
Geographic and Contextual Effects
269
A. Linear Model
Environment
Behavior Physiological Status
Carcinogenesis
B. Nested Model
Fig. 10.1 A simple model motivating transdisciplinary studies of spatial covariation in environments and behavior. (a). Linear model emphasizing presence of causal links between adjacent levels. (b). Model highlighting hierarchical and nested characteristics of social and environmental factors influencing individual behavioral and biological characteristics (Derived and Revised from diverse sources in the literature, e.g., [138]
retail environments and social variables such as neighborhood cohesion, crime, and personal contacts. Identifying such contextual factors suggests avenues for population level interventions that can improve health behaviors [136, 63, 4, 60, 78]. A broad definition of “environment” also acts to encourage collaboration and learning among research communities investigating environmental exposures to toxins, such as arsenic or particulate air pollution, that have direct effects on
270
D. Berrigan et al.
health and those investigating exposures such as the distribution of retail outlets or street connectivity that are hypothesized to influence behavior and therefore disease risk. The conceptual model outlined in Fig. 10.1 does not emphasize genetic variation, or genotype by environment interactions, which influence behavioral or physiological responses to the environment. Substantial evidence indicates that genetic variation is important for the origin and maintenance of obesity [118, 15] by environment interaction. As genotyping continues to become increasingly economical and comprehensive and as additional genetic markers related to environmental sensitivity are identified, the role of genotypes in the causal pathways among different levels of causation presented in Fig. 10.1 can be examined [3, 50, 134, 83, 14]. Indeed, it is well known that diverse genetic markers in humans have considerable geographic variation, although the appropriate geographic scale at which this variation manifests itself is not yet fully described [116]. Finally, genotype by environment interaction in the presence of spatial variation in genes influencing disease or behavior could result in confounding. This issue is beyond the scope of this chapter but is a major historical focus of quantitative and evolutionary genetics [145] and is a major challenge for the interpretation of spatial variation in disease. This chapter touches on diverse topics from a variety of disciplines. In the first section we review recent literature concerning factors that influence geographic and contextual variables, including urban sprawl and obesity, built environment and physical activity, food environments and disparities in food access, and behavior-setting theory and energy balance. In the second section of the chapter, we illustrate some of the major challenges to research in this area. We focus on four topics: confounding as it applies to spatial patterns, neighborhood selection as a barrier to causal inference in cross-sectional studies of the association between environment and behavior, the potential contribution of tools for detecting spatial clusters of behaviors, and the major challenge of incorporating maps into the analysis of geographic variation in energy balancerelated variables and cancer. In this final discussion, we briefly consider limitations to the use of spatial methodologies for exploring risk factor and disease associations when risk factors are measured poorly at the spatial level and the diseases being examined are often multi-factorial in origin without large relative risks identified for any single risk factor. At present, the most conclusive research using spatial methodologies to explore risk factors and cancer associations has examined the association of spatial variation in risk factors that are well measured and are known to be the predominant contributor of a specific cancer. The best examples of such research are spatial variation in tobacco use and lung cancer [123] and human papilloma virus (HPV) exposure and cervical cancer [38]. Better measurement of environment and behavior at the population level is needed to determine the ultimate value of a geographic approach to diseases with multiple and difficult to measure candidate risk factors. This effort seems worthwhile. Finally, we end the chapter with a discussion of some overarching problems and prospects in the analysis of geographic variation in behavior and disease, notably
10
Geographic and Contextual Effects
271
involving protection of confidential data, and the potential of contextual variables to contribute to clinical trials of behavioral interventions aimed at energy balance behaviors.
2 Geographic and Contextual Effects on Energy Balance Variables The analysis of geographic and contextual effects on diet, weight, and physical activity (here abbreviated as “energy balance”) has been used to identify environments that facilitate the adoption of recommended health behaviors by individuals and has expanded markedly [56, 9, 69, 55]. Some of this growth in research output can be attributed to the collaboration across disciplines that is common to many areas of behavioral research in the past decade. In the case of physical activity and the built environment, many of these issues were aptly summarized at a conference held in 2001 (Amer J Prev Med Vol. 23, Issue 2, 2002). This conference arose from substantial efforts by the organizers to build mutual understanding of diverse vocabularies, research goals, and analytical tools among transportation researchers, urban planners, epidemiologists, and other researchers. The conference highlighted many themes present in the contextual analysis of behavior, including need for better measures of the environment, creative theory building, and greater attention to experimental design and statistical analysis of complex data sets. These and related issues are still at the forefront of research on physical activity and the built environment [8, 13, 60].
2.1 Urban Sprawl and Obesity The first studies reporting a direct relationship between the built environment and obesity were carried out in the early 2000s [49, 28]. In these studies, the built environment was conceptualized as “sprawl” and the underlying hypothesis proposed that an automobile-oriented development pattern promotes sedentary behavior, positive energy balance, and, therefore, obesity. Ewing and colleagues [49] used county-level data concerning US adults to analyze associations between urban sprawl and obesity. Estimates of obesity were obtained from the Behavioral Risk Factor Surveillance System and a county sprawl index was developed to summarize the built environment. The sprawl index is a composite of six variables related to residential density and street accessibility combined through principal components analysis [49]. The six variables included gross population density, percent of population living at less than 1,500 people/square mile, percent of population living at more than 12,500 per square mile, county population divided by the amount of urban land, average block size, and percent of blocks smaller than approximately 500 feet per side. Thus, this and other indices of urban sprawl are strongly influenced by population density, but are not measured by density alone [98]. Related indices have now been used in diverse studies [119]. Adults living in sprawling
272
D. Berrigan et al.
counties had higher body mass indices (BMIs) and were more likely to be obese (BMI > 30 kg/m2 ) than were their counterparts living in compact counties. This association was still present after controlling for age, education, fruit and vegetable consumption, and other sociodemographic and behavioral covariates. Papas and colleagues [119] reviewed 18 cross-sectional studies linking built environment and energy balance variables. Four of these studies examined associations between obesity and sprawl and identified weak but statistically significant associations between sprawl and obesity. The above studies were cross-sectional and included only adults. Research is limited on built environment and youth and results have been mixed [19, 156, 144, 160, 161, 42]. Ewing and colleagues [46] investigated associations between sprawl and obesity in a combined cross-sectional and longitudinal study of US adolescents. In their cross-sectional analyses, adolescents living in sprawling counties were more likely to be overweight or at risk of overweight than those living in compact counties. Likewise, young adults living in sprawling counties were more likely to be obese. However, the longitudinal analyses showed no statistically significant association between sprawl and weight gain among youth. The differences between these two analyses could be because of inaccurate measurements of exposures, time lags in the effects of sprawl on energy balance, the examination of children at different times in adolescence when trajectories of weight gain may differ greatly, or insufficient elapsed time to observe significant change. Sampling design, study duration, and magnitude of the environmental effects also could influence the relative power of cross-sectional versus longitudinal analyses to identify statistically significant environmental effects. The association between sprawl and behavior could be clarified by further theoretical and empirical work. Such studies require better measures of the environment and more powerful study designs. For example, models that delineate categories of behavior and the specific settings in which they can occur, along with appropriate constraints involving time use and economic factors, might prove more successful at quantifying the possible contribution of different policy or environmental factors and help to identify worthy candidates for further study using more rigorous (and potentially more costly) experimental designs.
2.2 Built Environment and Physical Activity Many more studies have examined associations between the built environment and physical activity than have examined associations between the built environment and obesity [68, 47, 53, 4, 167]. These studies address diverse forms of physical activity, but a major focus has been on active transportation (mostly walking and bicycling). This focus arose in part because of the logical connection between urban form and walking and in part because of the influence of the transportation/planning literature [47, 69]. This work has explored the effects of population density, site design, and building characteristics on transportation behavior. Many of these studies report that increased density and mixed-use development
10
Geographic and Contextual Effects
273
are associated with more walking and bicycling [54, 121, 122]. However, considerable debate continues over the consistency and interpretation of such associations [55, 89, 12]. Much of this debate has centered on whether or not causal inferences can be made from the analysis of cross-sectional associations. Such studies also have been extended to examine walking and bicycling for multiple purposes [53, 4, 17]. In its simplest form, this area of research attempts to quantify the relative contribution of different environmental measures, such as street connectivity, sidewalks, or safe crosswalks, with the prevalence of walking. For example, a number of researchers have suggested that measures of street connectivity (e.g., block length and pattern of the street network) could be associated with walking [53]. There is a consensus that density, diversity, and design interact to influence walking. However, because the influence of built environment is much smaller than the influence of demographic variables such as age, sex, employment status, and family circumstances, and potentially because of remaining challenges in the measurement of environmental variables, there does not seem to be a consensus concerning the relative importance of such factors or the optimal interventions that could be selected to increase walking [29, 53, 69, 151, 60]. Nevertheless, many communities are already acting on this evidence (http://www.smartgrowthamerica.org) despite the fact that strong evidence of causal associations between the built environment and increased physical activity is lacking from the peer-reviewed literature [79]. Critics of this approach have suggested that changing the urban landscape will take too long or cost too much to have significant impact. However, changes to the built environment that improve walking behavior range from modest to expensive and can occur on multiple time scales. For example, block length may be fixed in many areas, but traffic calming measures, such as narrowing sections of streets or installing speed bumps, might be feasible in the short term at modest cost, while longer range changes, such as lower residential density, might allow walking or bike paths that reduce effective block length for non-motorized transport, may be supported as part of urban planning. More nimble funding for evaluation projects could help improve the evidence base linking community action, zoning regulations, planning activity, and physical activity [8]. In addition to research using transportation data resources, public health investigators are exploring such issues with existing national health surveys. For example, we report increased odds of walking one or more miles for respondents living in residences built before 1946 based on data from an existing nationally representative survey (NHANES III) of US adults [9]. This study includes many features of recent work on the built environment and physical activity. We describe this study to highlight the gaps in current health surveys that limited early research on the environmental determinants of physical activity. Four problems seem particularly noteworthy. First, the study is cross-sectional in design. Thus, self-selection or confounding could account for the association between environment and behavior [96]. Second, the measure of the built environment (time period when the resident’s home was built) is a proxy measure of the environmental variables that determine a latent construct “walkability” [85]. Third,
274
D. Berrigan et al.
the study considers only environmental features around the home; it does not include the entire walking environment such as work location, school, or other venues where pedestrian activity might occur [61]. Fourth, the estimate of walking is based on self-report and includes only a subset of all walking trips [10, 91, 158]. Since 2002, many analyses of associations between environment and walking in multiple domains have used more comprehensive measures of the environment [166, 99] and more elaborate instruments to measure walking, including objective measurement by accelerometer [57]. However, longitudinal study designs remain rare. A review of this field [167], identified 47 studies of the association between environmental variables and physical activity from 2000 to 2004. Most of the studies occurred in the United States or Australia and most of the studies examined walking as the outcome. Overall, few associations were strongly supported except for a consistent relationship between social support and diverse measures of physical activity. The review illustrates the lack of consistency in extant studies of measures of both environment and physical activity, with few studies using the same measures of environment or behavior. However, this field is evolving rapidly. Searching the ISI Web of Knowledge for “built environment” and “physical activity” results in 272 citations, 230 of them published after 2004. This reflects the emergence of “built environment” as shorthand for environmental influences on physical activity, but it also reflects growing interest and funding for this kind of research as concern about the obesity epidemic has grown. Despite the use of improved and more comprehensive measures of the built environment and, most recently, objective measurement of physical activity, cross-sectional studies continue to result in small, albeit statistically significant, associations between the built environment and physical activity. As it is likely that the effect size of any single environmental factor is small, it is critical that comprehensive theories of how built, social, and economic environments influence physical activity be used to quantify the contribution of these factors to identify potential targets for interventions at multiple levels including the home, workplace, community, or policy level.
2.3 Food Environments and Food Deserts Research has grown on the potential effects of the built food environment on dietary behavior, although it is not as developed as in the field of built environment and physical activity. The Dietary Guidelines for Americans, updated every 5 years, stress the importance of (1) consuming sufficient quantities of fruit, vegetables, whole grains, and reduced-fat or non-fat milk, and (2) limiting intake of added sugars, saturated and trans fats, refined grains and salt [US Department of Health and Human Services 2005]. In order to meet dietary guidelines, people need access to recommended foods. However, it appears that some communities may be “food deserts.” In other words, they have reduced access to nutritious and affordable food. First used in 1996 by the United Kingdom’s Nutrition Task Force’s Low Income Project, the term is used to describe “areas of relative exclusion where people experience physical and economic barriers to accessing healthy food” (p. 138) [129, 35].
10
Geographic and Contextual Effects
275
Food deserts may pose a challenge for public health policymakers, as it would seem particularly difficult to alter dietary behavior in an unsupportive food environment. Food deserts have been studied from the perspective of a number of disciplines and fields, including public health, nutrition, epidemiology, medicine, retail studies, economics, agricultural economics, planning and design, and geography. Researchers using quantitative research methods have used several measures to attempt to identify food deserts, and have used different tools based on the question under investigation. Measures used by researchers include (1) GIS to assess and display geospatial differences in food stores, (2) market baskets to evaluate price differences between geographic areas, and (3) other instruments such as checklists, inventories, and interview/questionnaires to assess food availability and quality within food outlets [105, 33]. Few national-level studies exist, and those that do focus on issues such as access to supermarkets within a geographic area [127]. Studies that assess disparities in healthy food access and affordability with greater granularity (e.g., availability of “indicator foods,” such as low-fat milk and fresh fruit and vegetables, and quality of food items), typically assess stores within defined geographic areas, such as within cities or counties [11, 73–75, 82, 147, 165]. A number of studies support the existence of food deserts in the United States [31, 130, 41, 147, 74, 11, 169, 73, 75, 82, 110, 112, 170]. Research on food deserts in other developed countries is limited, and provides conflicting evidence of their existence [2, 33, 59, 65, 148]. Despite the fact that the concept of food deserts first gained popularity in the United Kingdom, more recent research suggests that food deserts may not exist there [37]. In the United States, research suggests that community access to food stores differs by store type, and that the differences in store type may be responsible for the geographical variation in food prices, variety, and quality. Supermarkets—in comparison to grocery stores—appear to offer greater variety of food choices [11] at lower cost [31, 34, 113], in addition to providing higher quality, and a greater variety of produce [113, 11, 170]. Several studies found that low-socioeconomic status (SES) and/or non-white communities had more grocery stores and fewer supermarkets relative to communities with a high percentage of white residents [31, 127, 74, 11, 169, 111, 82, 170, 128]. This difference in access to food store types may have important consequences for dietary behavior, if food selection is influenced by the type of retail outlet present. Studies that are restricted to comparing supermarkets only or grocery stores only showed little variation in price between high- or low-SES communities [71, 74, 11, 73]. Most of the research in this field has assessed the relationship between the food environment and demographic variables of different geographical areas, such as race/ethnicity and SES. Fewer studies have examined whether availability of nutritious food choices in food stores and individual dietary intake are related. However, Fisher and Strogatz found a positive association between the availability of low-fat milk in stores and individual consumption [52]. Other researchers have found links between food store availability/access and intake of nutritious foods in general [97], or fruit and vegetables specifically [135]. Behavior-setting theory could help guide
276
D. Berrigan et al.
studies of the multilevel effects of food availability in retail outlets and specific dietary choices. Establishing a causal connection between the food environment and dietary behavior is particularly challenging, as most of the studies on food deserts are cross-sectional in design. However, two longitudinal studies have used the implementations of supermarkets in low-SES neighborhoods to evaluate changes in intake over time. Results from these UK-based projects provide conflicting evidence on food availability and its effect on diet. One study from a development in Glasgow, Scotland, found that people switched to shopping at the newly implemented supermarket, but did not change their fruit and vegetable intake [35, 36]. Another study in Leeds found that introduction of a new supermarket in a low-SES community was associated with an increase in residents’ fruit and vegetable intake [167]. However, self-reports of diet contain measurement error. The brief dietary assessment instrument (two questions regarding fruit and vegetable intake) in the Scottish study is imprecise and may be inadequate to capture any dietary changes, should they have occurred. The method used in the Leeds study (7-day food diary pre- and post- intervention) likely provides more precise estimates of intake, and thus change, but also may suffer from under-reporting intake overall [155, 131]. Evidence clearly suggests that disparities exist in the geographic accessibility and affordability of healthy food, even in a prosperous and developed country such as the United States. Further research on spatial and other variation in food resources could provide useful guidance for urban planners, public health agencies, and legislators in their efforts to influence planning and policy for health. It is not yet known how much of an impact changes in the distribution of retail outlets or outlet type could improve diets in the US or other countries.
2.4 Behavior-Setting Theory and the Social Context of Energy Balance Studies of sprawl, the built environment and physical activity, and obesity largely measure the presence, availability of, and access and distance to healthful or unhealthful resources (e.g., parks, programs, fast food restaurants). Unfortunately, this research approach provides limited information about the extent to which individuals actually encounter or interact with these features or visit particular settings. Lack of free range for play [80], lack of knowledge about available resources [43], lack of time [140], safety concerns [27], and other barriers may prevent children and adults from using available facilities. For example, while a regularly active individual may have access to a particular recreational facility in his or her neighborhood (or the setting being observed), the frequency with which he or she uses that environmental resource or the intensity/duration of his or her activity in that setting is seldom measured in existing studies. An alternative approach to investigating contextual effects on physical activity and diet is to study individuals’ interactions with their immediate environments through their exposure to, and use, and experience of different settings. There is growing interest in context-specific behaviors
10
Geographic and Contextual Effects
277
and behavior-specific aspects of the environment relevant to physical activity and healthy eating [141]. Researchers have used in-person observational techniques to measure activity levels in specific environments such as on school playground [103], on multi-use trails and parks [108, 132], or during physical education classes [114]. Fewer studies, however, have taken a systematic human–environment interaction approach—that is, comparing different contexts in terms of the types of energy balance behaviors that occur within them or comparing different individuals in terms of the types of environments in which they engage in physical activity and healthy eating. The human–environment interaction approach to studying the problem of the environment and energy balance is derived from an early theory that proposed that human behavior is set by the social and physical environment. This theory is termed “behavior setting” [6]. Attributes and characteristics of one’s immediate situation, such as type of social company and physical location, are thought to shape the behavior taking place in that setting [159]. Salvy and colleagues adopted this perspective in their examination of children’s physical activity across different social contexts. They found that children had greater motivation for physical activity [142] and engaged in higher intensity activities when in the presence of others as compared to when they were alone [143]. Other research has compared children’s physical activity across physical settings, showing that activity intensity is higher outdoors than inside homes [5]. Research using time use survey methodology found that adults’ exercise bouts were shorter when exercising alone as compared to with family members or friends/acquaintances. Also, the average duration of exercise bouts was greater when the activity occurred outdoors as compared to when at home, work, or at gym/health club [42]. Behavior-setting theory provides a conceptual framework for further understanding these kinds of empirical observations through its emphasis on interactions between people and specific aspects of social and physical settings. Exposure to and use of different social and physical environments for physical activity and dietary behaviors can vary across population subgroups. In a study using electronic diaries among high school students, Dunton and colleagues found that boys were more likely than girls to report exercising and walking in outdoor locations. Also, activity contexts varied by age. The proportion of exercise bouts occurring with classmates, family members, and at school decreased from 9th to 12th grade [44]. Data from the 2003–2006 American Time Use Survey reveal demographic patterns in adults’ exercise contexts. More exercise bouts occurred alone and outdoors among adults aged 60 years and older as compared to younger age groups. College graduates were more likely to exercise at a gym/health club as compared to adults with lower levels of education [41]. Future work is needed to understand whether demographic disparities in the use of different environments for sports and exercise result from differential preferences or opportunities across these groups. Nonetheless, the human–environment interaction approach to understanding the effects of the environment on energy balance behaviors may help guide the development of targeted interventions and policies to increase physical activity and healthy eating in specific populations. Programs to promote healthy
278
D. Berrigan et al.
lifestyles can most likely benefit from information about how social features and physical settings influence the types of behavior taking place in those contexts. One of the limitations of current data within such time use surveys is the lack of objectively measured physical activity. Methods for objectively measuring physical activity have become less expensive and more feasible for use in large population level surveys. The addition of such measures in even sub-samples of people within these surveys could greatly enhance the accuracy of measurement of physical activity. Above we have focused on physical environments that potentially influence obesity, physical activity, and transportation choices. There is also considerable recent interest in the potential for social networks to influence these energy balance variables. Ideas, feelings, and behaviors can be spread through social networks, so aspects of energy balance involving eating and physical activity might also be influenced by such networks. Analysis of social networks in the Framingham Heart Study indicates that risk of obesity increased significantly among people with friends who became obese during a particular time period [30]. Such effects also applied to adult siblings and spouses [30], but were not seen in study participants living nearby who were not socially connected with the subjects embedded in a network exhibiting weight gain. Cohen-Cole and Fletcher [32] carried out a similar analysis using data from the Study of Adolescent Health Behaviors (ADD Health), including obesity (http://www.cpc.unc.edu/projects/addhealth). These authors were able to replicate the results of the Framingham study. However, they note that some of the residual variance in obesity might be attributed to other (unmeasured) contextual factors. The authors do not speculate further, but these features could be any shared environments used by members of a social network differentially compared to people not members of that network. One recent study of 562 adolescent girls does report that overweight adolescents were more likely to have overweight friends than were normal weight peers and that adjustment for classroom and school effects did not eliminate these associations [Valente and Spruijt Metz, In Press]. However, contextual factors beyond the classroom and school could still play a role if the association between such environments and obesity varied. This study highlights one of the key limitations in this field of research, which is the need for further data on both the social and the environmental context of eating behavior, physical activity, and weight trajectories. Time use and transportation research studies that have assessed factors influencing adoption or patterns of physical activity or examined associations of physical activity with disease outcomes have included measures of social or environmental context. However, the addition of even a limited set of these measures to research examining eating behavior, physical activity, and weight trajectories could greatly enhance the potential to explore the role of social context in energy balance-related behaviors. At present a limited set might include measures of the age, sex, race, and obesity status of people with whom subjects interact in different environmental contexts. The addition of such measures is likely to be more informative within studies that are examining human behavior within defined social contexts, such as schools or workplaces, where individual friends or family and their characteristics can be
10
Geographic and Contextual Effects
279
identified. In general, many health studies and larger surveys will be limited to very simple measures of social context. For example, many time use surveys include information concerning who was present during self-reported activities over the past 24 h. Dunton and colleagues used the American Time Use Survey to describe the social and environmental context of leisure time activity [41]. The survey includes questions such as “Who was with you/Who accompanied you?” with interviewer coded responses into 22 different categories for each activity in the previous 24 h. Such studies can provide novel information about interactions between contextual and demographic variables. Transportation researchers are also developing an interest in the extent to which social networks influence travel behavior, travel demand, and mode choice [26, 24, 25]. Some of this work has been facilitated by the fact that household transportation surveys may include information about with whom and where people travel. Transportation surveys are adding GPS elements and this could also increase the physical contextual data available. More work remains to integrate methods for analyzing social networks from other disciplines, develop new data resources, and develop new approaches to teasing apart the relative importance of social and other contextual effects. Collaborative work concerning networks by economists, sociologists, transportation researchers, psychologists, and epidemiologists could accelerate progress. This point is nicely illustrated by Cohen-Cole and Fletcher’s discussion [32] of different vocabularies used by sociologists and economists and by the observation that behavior-setting theory appears to address the problem of built environment and physical activity directly, yet is seldom addressed in the epidemiological literature on this topic. Across these disciplines very different vocabularies are used to discuss similar or related concepts (e.g., homophily vs. selection; social network effect vs. endogeneous social effect). Perhaps analyses of an overarching and controversial social and health problem such as the US obesity epidemic will provide a forum for work to integrate vocabularies and methods concerning social network influences on behavior and behavior change. Such integration is needed to quantify the relative contributions of social and environmental variables to energy balance. Behaviorsetting theory can help guide the design of such studies and the level of detail required to address different questions concerning the influence of social and environmental contexts on energy balance.
3 Major Challenges for the Analysis of Geographic and Contextual Effects on Energy Balance In this section we consider four major challenges to the analysis of contextual effects on energy balance from a public health and epidemiological perspective. These challenges include (1) confounding as it applies to spatial patterns, (2) neighborhood self-selection as a barrier to causal inference in cross-sectional studies of the association between environment and behavior,
280
D. Berrigan et al.
(3) The problem of variable selection for examining determinants of behavior and the potential contribution of tools for detecting geographically identified clusters of behavior, and (4) The major challenge of incorporating maps into the analysis of geographic variation in disease and the puzzle of how to interpret lack of congruence between spatial variation in known risk factors and associated diseases. Although not a specific challenge to analysis, a key limitation in this field is that many of the measures used for estimating risk factors in spatial studies are derived from very short questionnaires that are known to measure such exposures with significant error. If measures of exposure are poor, it is unlikely that addressing the challenge in analysis will improve the ability for spatial studies to identify associations should they exist.
3.1 Confounding Given the diverse challenges to inference about spatial variation in disease, it is possible to neglect the basic challenge of epidemiological inference, namely confounding. Confounding occurs when an estimate of an association is distorted by one or more additional variables that are associated with both the exposure of interest and the putative outcome [96, 126]. This could be the most serious problem for spatial analysis because spatial studies naturally include variables related to multiple levels from the individual to social, policy, and environmental factors. This data complexity reflects causal pathways linking geographic, contextual, and biological variables to behavior and disease. However, multi-level and multiple factors require careful attention to the issues of confounding and the development of theories that define causal pathways. We illustrate this issue with a very simple example. The United States exhibits substantial statewide variation in obesity rates (www.cdc.gov/nccdphp/dnpa/obesity/trend/maps/). Some of this variation could be due to differences in levels of physical activity. Much physical activity occurs outdoors [41] and hot, humid weather is known to be a significant barrier to physical activity. We analyzed the association between mean July temperature and percent overweight/obese in US adults at the state level. Remarkably, we found a high correlation between temperature and obesity with higher temperatures related to higher prevalence of obesity. This might suggest that climate may be a key contributor to differences in obesity in the United States and interventions aimed at the issue could focus on providing opportunities for indoor recreation. Alternatively, other confounding variables could account for the association between temperature and the prevalence of obesity. Inspection of the state-level maps of temperature and obesity naturally led us to consider race/ethnicity as a potential confounder. And indeed, inclusion of the percentage of the population who are African American as a covariate eliminates the association between temperature and obesity. This is because African Americans have elevated BMIs on average,
10
Geographic and Contextual Effects
281
and are more prevalent in the hotter parts of the southeastern US. This is a simple example, but it highlights the logic of confounding in a spatial context and the need for caution in interpreting even strong associations between energy balance-related variables and environmental characteristics. Additionally, it is important to note that this does not mean that outdoor environmental characteristics do not influence individual propensity for outdoor activity. Rather, this association does not provide a direct explanation for the observed state-level pattern of obesity prevalence.
3.2 Self-Selection and Interpretation of Cross-Sectional Associations Between Environment and Behavior Humans are not randomly assigned to habitats, and therefore, a major concern for the interpretation of cross-sectional studies of environment and behavior is the possibility that adults select homes in environments that facilitate behaviors they prefer. This is an example of problems with causal inference from cross-sectional data. In other words, environmental variables do not cause people to increase their walking. Instead, people who choose to walk may also choose to move to neighborhoods that are walking friendly. If this is true, changing the environment would improve conditions for walkers, but it might not cause non-walkers to walk more. This is an important issue for public health, urban and transportation planning, housing, education, and many other areas associated with improving communities [23, 21, 70]. The issue of the role of self-selection in residence and its association with health behaviors have been recently reviewed [109, 70]. Random assignment to different environments followed by measurement of health behaviors related to energy balance is not feasible at the population level. Therefore, causal inference concerning associations between the built environment and physical activity must rely on observational study designs. Mokhtarian and her colleagues describe statistical approaches to modeling the relative contributions of built environment, preferences and attitudes, including statistical control, instrumental variable modeling, sample selection designs, and propensity scoring. Such approaches, developed in various disciplines, including economics, epidemiology, travel behavior modeling, and social epidemiology, all offer improvements over simpler regression analyses. However, none of these approaches can overcome the inherent limitations of cross-sectional study designs. Currently, more than 30 studies incorporate all these analytical approaches. A review of this research concludes that both self-selection and built environment may influence various aspects of physical activity and transportation behavior, consistent with an emerging consensus that public health efforts should address both individual preferences and behaviors as well as the many environmental forces that shape them. As in epidemiological studies of association between cancer and energy balance, longitudinal studies of associations between environment and behaviors may reduce the risk of confounding. Such studies may occur in convenience samples
282
D. Berrigan et al.
of selected populations [67] or in populations associated with a specific environmental change—a natural experiment or a “before-and-after” study design [100]. Major approaches include comparisons of movers versus non-movers and the different environments they inhabit [90] or secular changes in neighborhood characteristics and attitudes such as construction of walking trails [18]. Although such study designs are more powerful than cross-sectional studies, they are also more expensive and require careful planning and design. Further, most of the extant studies are small and highly localized to specific communities [20]. Larger panel surveys and careful before-and-after studies of major alterations to the built environment are required to provide definitive evidence of causal associations between built environmental characteristics and physical activity or active transportation [62, 22].
3.3 Spatial Scan Statistics to Identify Geographic Clusters of Behavior Many studies have reported associations between diverse measures of urban form and active transportation or leisure activities such as walking and bicycling [79, 53, 4, 119, 167]. However, the amount of variation in active transportation in these studies has been relatively small, the results are inconsistent, and the evidence is based almost entirely on cross-sectional studies [4, 167]. These studies typically involve compilation of diverse demographic and contextual variables followed by regression models used to identify relevant environmental variables. Such approaches have provided significant insight into the relative contributions of density, diversity, design, and demography to levels of active transportation [53]. Nonetheless, it seems possible that such approaches are limited by the inability of simple models to capture complex interactions among the many contextual and environmental influences, particularly if these influences vary geographically. A complementary approach to the classical method might involve the identification of neighborhoods or clusters with elevated levels of active transportation. Once these clusters were identified, efforts could be made to compare and contrast the properties of both individuals in the clusters and the environmental features of each cluster. Such an approach might identify related factors that could be tested for causality in more rigorous designs. Of the more than 100 statistical tests proposed for overall clustering or cluster identification [93, 95], the spatial scan method [64, 163] is a particularly suitable tool for implementing this approach because of its analytical capacity to account for covariates. The likelihood-based spatial scan approach is also good at detecting localized clusters [92, 45, 153, 40, 149, 163, 76]. This method primarily has been used to detect clusters related to infectious disease and accidents [154, 66, 72, 115, 126] but it also has been applied to forestry, toxicology, psychology, and criminology data [81, 150, 102, 133].
10
Geographic and Contextual Effects
283 A: Clusters of high/low prevalence in LA without age-adjustment
Location of the subjects in LA
6 Low
2 High 7 Low 3 High
1 High 5 High 4 High
B: Clusters of high/low prevalence in LA with age-adjustment
4 Low
3 High 2 High
1 High
5 Low
Fig. 10.2 Clusters of walking in LA. Spatial scan statistics can identify spatial clusters of health behaviors based on geographically identified presence or absence of such behaviors. (a). Clusters identified in a crude analysis. (b). Clusters identified after age adjustment
Huang and colleagues have applied such a technique to variation in active transportation in Los Angeles and San Diego counties (Fig. 10.2) [77]. They identified clusters of active transportation in both counties and characterized neighborhoods by demographic and built environmental factors that might influence walking. As anticipated, they found differential associations with walking for different neighborhood characteristics. Specifically, clusters of walking occurred in areas with both high and low SES, although all high walking prevalence clusters were characterized by relatively high levels of street connectivity and relatively high population density. Age accounts for some, but not all of the clusters of active transportation identified in this analysis (compare Fig. 10.2a and 10.2b). What sets this approach apart is that not only is it possible to make inferences about pre-selected explanatory variables, but also specific neighborhoods can be identified and targeted for
284
D. Berrigan et al.
further formative or intervention research, and the potential role of covariates can be addressed directly. Use of the spatial scan approach, coupled with alternative study designs, such as ethnographic approaches, focus groups, direct observation, and other qualitative methods, could help generate a new conceptual framework or at the very least a new appreciation of the diversity of factors involved in the association between the built environment and physical activity. This method is best used for population-based studies in a large region where geographic variation of the behavior is of interest and where geographic identifiers are available. Choice of a maximum window size for the cluster scan will be a compromise between including a large enough population to provide sufficient statistical power to detect significant clusters and small enough to identify local neighborhoods with distinct characteristics. The spatial scan approach may have more limited application in regards to diet and nutrition. The scan approach requires categorizing respondents as cases or controls. Thus, the scan approach will be limited in its capacity to explore geographic variation in diet beyond categories such as adherent or non-adherent to certain recommendations or with and without a certain dietary pattern. Nevertheless it might be of considerable interest to determine whether clusters of certain dietary patterns could be identified. Similarly, it might be quite interesting to determine whether the spatial distribution of farmers markets was associated with elevated fruit and vegetable consumption. The spatial scan approach could be adopted to address this kind of question. Finally, the availability of large spatially referenced data sets with detailed dietary data needed to characterize dietary patterns is limited. Standardized regional and national surveys such as the National Health Interview Survey (NHIS) or Californial Health Interview Survey (CHIS) often measure diet with a limited set of screener questions, precluding more complete analysis of diet.
3.4 Geographic Analyses of Energy Balance-Related Measures and Cancer Up to this point we have focused on associations between environmental variables and health behaviors related to energy balance. Before concluding this chapter we briefly touch on approaches to studying the geographic distribution of cancer per se. It is worth noting that the US National Cancer Institute has been a leader in developing tools for the geospatial analysis of cancer incidence and mortality [124]. Diverse GIS applications and resources related to analysis of this spatial variation are available at http://gis.cancer.gov/ [Accessed March 2, 2009]. Furthermore, interest in global variation in cancer incidence and mortality is substantial [120, 16, 157, 84, 104]. Geographic analyses of disease, including cancer, have proved most informative when disease incidence and key exposures have been measured with accuracy and, in some cases, where one or a few risk factors are the dominant cause of disease.
10
Geographic and Contextual Effects
285
Diverse infectious diseases [163], cancer caused by exposure to infectious agent, such as cervical cancer [38], tobacco-related cancers [123], and rare cancers caused by exposure to specific environmental agents such as asbestos [152], are examples of model systems in which geographic analyses have done much to strengthen our understanding of causal associations between risk factors and diseases. However, the large-scale analysis of disease incidence and health behaviors related to energy balance remains a significant challenge for two reasons. First, the spatial and temporal scales over which cancer develops versus the scales over which environment influences health behaviors related to energy balance may be
Uterine Cancer Incidence
29 0
30 2
28 4
28 9
26 0
28 1
25 3
25 8
23 9
25 0
21 7
23 7
19 3
21 6
18 2
18 6
15 8
18 0
Data Not Available
Colon Cancer Incidence
Fig. 10.3 State-level variation in obesity and cancer incidence in the United States. (a). Uterine cancer (Women). (b). Colon cancer (Men and Women Combined). (c). Proportion with BMI ≥ 30.0. Cancer incidence from http://statecancerprofiles. cancer.gov/accessed March 2, 2009. BMI data from 2006 based on self report in the Behavioral Risk Factor Surveillance System (BRFSS) at http://www.cdc.gov/ NCCDPHP/DNPA/obesity accessed March 2, 2009
54 7
58 0
54 2
54 6
51 9
54 1
50 7
51 8
49 6
50 6
48 5
49 5
46 9
48 4
44 2
46 8
38 0
44 1
Proportion Obese 0 293
0 317
0 278
0 292
0 272
0 277
0 260
0 271
0 253
0 259
0 246
0 252
0 234
0 245
0 218
0 233
0 184
0 217
286
D. Berrigan et al.
quite different. For example, built environmental variables might have heterogeneous influences on walking behavior at the block group level over the course of a few years during which a subject lives in a certain neighborhood. In contrast, physical activity may only exert a salient influence on behavior over multiple decades. Furthermore, any given energy balance-related variable may have only a small influence on risk. Several chapters in this volume address the magnitude of these risks, and few if any are as large as those associated with tobacco or HPV. Thus geographic and contextual analyses may prove useful for studying causes of behaviors related to cancer risk, even if they do provide direct insight into causal associations between risk factors and disease or account for geographic variation in disease. Second, both variables related to energy balance, particularly those related to diet and physical activity, and built and food environments are currently measured with substantial error [107, 86, 51, 105]. Thus, despite significant geographic variation in obesity and in cancers known to be related to energy balance (Fig. 10.3), it seems unlikely that ecological regression or iterative analyses of spatial clusters will be able to tease apart complex causal pathways relating environments, energy balance, and cancer. This challenge is exacerbated by the fact that associations between individual aspects of diet, physical activity, and cancer are generally modest. Progress has been made in objective measurements of physical activity at the national level [158], development of new tools for objective measurements of diet (http://www.gei.nih.gov/), and methods for characterizing the built and food environments [105]. These developments may lead to greater success in the analysis of geographic variation in behavior, obesity, and cancer to develop new targets for prevention and new hypotheses concerning the causes of obesity.
4 Future Directions The built environment has been recognized as a powerful influence on health for several centuries. This recognition was further strengthened when the Industrial Revolution led to a concentration of adverse housing conditions and rendered many cities in Europe and the United States virtually uninhabitable. However, only recently have diverse research disciplines begun to focus on the potential role of the environment as an influence on health behaviors, such as diet and physical activity, and begun to characterize communities as healthy based on whether they facilitate or inhibit such behaviors. This research effort is of considerable importance to researchers focused on cancer prevention and control because substantial evidence already suggests that built environment and contextual effects can have significant influences on physical activity, diet, obesity, alcohol and tobacco use, and use of health services. Health researchers have much to contribute to this research endeavor because of our focus on rigorous measurement, our history of identifying biological markers of behavior, and our intense interest in demonstrating causal associations between environmental exposures and health or behavioral outcomes. In turn, health researchers have much to learn from urban and transportation planners, ecologists,
10
Geographic and Contextual Effects
287
sociologists, landscape architects, and researchers in other traditions who are also focused on the role of the built environment as a determinant of diverse individual behaviors and social patterns. Almost every aspect of research on geographic and contextual effects on energy balance variables is at the developmental phase. Thus, significant gaps exist in the development of tools to measure environments, the theoretical background needed to motivate such tools, and the availability of appropriate data resources. In addition, experimental studies are distinctly lacking, to some extent, an analytical framework for such analyses. This latter point may not be as serious a problem as the other three areas because many powerful statistical and modeling approaches have been developed for economic, transportation, and sociological research problems. Many of these have yet to be applied widely to analytical efforts aimed at understanding contextual effects on energy balance variables. Measurement, analysis, and theory development are among the major challenges for this research endeavor. McKinnon and colleagues have recently compiled results of a workshop in this and concluded that considerable work remains to develop and validate measures of the built and food environments [105]. Public health researchers, transportation modelers, geographers, spatial statisticians, economists, and sociologists, just to name a few, are developing and testing complex new analytical tools aimed at understanding associations between multi-level variables and teasing apart the relative importance of particular variables from a complex web of causation [109]. However, such researchers are scarce and to accelerate progress in the transdisciplinary science of contextual effects on energy balance and carcinogenesis, we need a cadre of researchers with the skills to understand and use tools from such diverse disciplines and with a clear awareness of their strengths, weaknesses, and assumptions. Transdisciplinary studies of contextual effects on energy balance suffer from proliferation of variables at multiple levels of organization and behavior resulting in a “combinatorial explosion” of potential analyses. McKinnon et al. and Oakes et al. describe two different visions to deal with such data complexity [105, 117]. Oakes argues that we can reduce confusion from data complexity in the first place by good conceptual models, rigorous study design, and clear thinking. Alternatively, it is possible that despite such efforts, clear solutions may elude us because food and physical activity research is inherently multilevel and complex [105]. The merits of these two perspectives are likely to be the topic of an intense debate in the coming years [105, 117]. In the near term, tools for managing data complexity are necessary. For example, ArcView GIS software catalyzed explosive growth in geographic analysis. We are waiting for the appropriate tool or set of tools that takes the integration of public health, cancer research, and geographic/contextual effects to the next level. Few longitudinal and before-and-after studies of the association between built environment and variables related to energy balance have been conducted. Such studies are needed to improve our estimates of the relative contributions of individual preferences, self-selection, and built environment to physical activity and diet as well as other behaviors. People move for many different reasons, and longitu-
288
D. Berrigan et al.
dinal studies that select people from professions (e.g., the military) or at stages in their work life (e.g., physicians assigned to residencies) where geographic relocation is likely to occur independently of self-selection related to the built environment, might be a particularly useful for determining the importance of built environment variables [139]. A strong strain of advocacy can be seen in some of the literature on urban design, walkability, and healthy communities. Care should be taken in interpreting the extant literature to avoid publication bias. Additionally, it is critically important to recognize that the same variable may have positive or negative effects on a health behavior depending on nearby residents’ stage in the life course or other factors. For example, a recent study indicates that children are more active when living in or near suburban cul-de-sacs than are children in traditional urban areas, but adults are more active in the traditional urban areas [67]. A major technical problem in this research area is that of data de-identification associated with use of geographic identifiers [162, 146]. This problem has emerged as much more important in health studies than it had been in the transportation and housing areas because of the sensitivity of personal information involving health status, health services use, and health behaviors. In the United States, results from national and regional health surveys are released for analysis but only when the data have been effectively de-identified. This often precludes the use of sufficiently finescale geographic information to allow meaningful inference. For example, it may be important to know a home location within a few tens of meters in order to calculate the average walking distance from the home to various amenities. Disclosure of a home address or a geographic location with a few tens of meters of a home address (or much more in a rural area) is equivalent to allowing easy identification of the respondent. Data containing fine-scale geographic information are available for analysis in protected data centers (http://www.cdc.gov/nchs/r&d/rdc.htm). However, the restrictions on analytical practice in such centers and the monetary and logistical expense of sending researchers to such sites have proved a strong barrier to the use of these resources. Many solutions to these problems have been proposed but no consensus that could lead toward additional analysis of existing data sets has been reached. Future health studies can avoid this problem through the use of the proper consent practices, but more work is needed to improve access and use of existing studies. Randomized controlled trials of interventions designed to change energy balance behaviors or measure the effects of behavior change generally result in significant fractions of the study population failing to achieve the goals of the intervention. For example, in the Diabetes Prevention Program (DPP) only 50% of the participants achieved the goal of weight loss of 7% at 24 weeks into the study and only 74% self-reported achieving the goal of at least 150 min of physical activity per week at 24 weeks on study [88]. Understanding why many people fail to comply to interventions or maintain compliance might help increase the power of studies testing the effects of behavioral interventions and also improve the interventions. It seems possible that the built environment could explain in part the failure to achieve physical activity targets by some participants. Brisk walking was a major way in which
10
Geographic and Contextual Effects
289
DPP participants could achieve their PA goals and substantial evidence suggests that the environment around the home could either encourage or discourage walking. At present, built environment variables are seldom explored as a potential influence on compliance to physical activity interventions. Another topic of interest in this area involves the potential role of the built environment in explaining or contributing to changes in energy balance behaviors over the life course. As in the example of cul-de-sacs and child versus adult activity levels, it seems clear that the same environments can have very different effects on people of different ages or at with different time use patterns. We know little about this issue and lack a research agenda aimed at integrating life course epidemiology, analyses of aging, and studies of the built environment in a way that integrates health and urban design.
5 Conclusions Cancer epidemiology and basic biomedical research have demonstrated strong causal associations between diverse aspects of energy balance and carcinogenesis. Public health interventions aimed at energy balance could reduce the burden of cancer and other diseases in the United States and worldwide. Research is needed on the interaction between individual and environmental determinants of behaviors such as diet and physical activity to identify promising new candidates for cancer prevention and control. Studies of the built environment are a significant element of this research agenda. Such studies require creative efforts designed to synthesize theoretical and empirical advances from multiple disciplines and the creation of trans-disciplinary teams with appropriate training [13] that invest the time and resources required to learn each other’s languages and integrate each other’s measurement tools and data resources. Acknowledgments Thanks to Sarah Locke, David Stinchcomb, Fran Thompson, and Donna Spruijt-Metz for many helpful comments on the manuscript. Special thanks to Anne Rogers for editing the entire manuscript, Penny Randall-Levy for preparing the bibliography, and Alyssa Grauman for advice on Fig. 10.1.
References 1. Anderson RM, May RM (1991). Infectious Diseases of Humans: Dynamics and Control, p. 757. Oxford University Press, Oxford. 2. Apparicio P, Cloutier M, Shearmur R (2007). The case of Montreal’s missing food deserts: evaluation of accessibility to food supermarkets. Int J Health Geogr, 6(4): 3. Arnold SJ (1987). Genetic correlation and the evolution of physiology. In: Feder ME, Bennett AF, Burggren WW, Huey RB (eds.), New Directions in Ecological Physiology, pp. 189–215. Cambridge University Press, Cambridge. 4. Badland HM, Schofield G (2005). The built environment and transport-related physical activity: what we do and do not know. J Phys Act Health, 2(4):433–442.
290
D. Berrigan et al.
5. Baranowski T, Thompson WO, DuRant RH, Baranowski J, Puhl J (1993). Observations on physical activity in physical locations: age, gender, ethnicity, and month effects. Res Q Exerc Sport, 64(2):127–133. 6. Barker RG (1968). Ecological psychology: concepts and methods for studying the environment of human behavior. Stanford University Press, Palo Alto, CA. 7. Beale L, Abellan JJ, Hodgson S, Jarup L (2008). Methodologic issues and approaches to spatial Epidemiology. Environ Health Perspect, 116(8):1105–1110. 8. Berrigan D, McKinnon RA (2008). Built environment and health. Prev Med, 47(3):239–240. 9. Berrigan D, Troiano RP (2002). The association between urban form and physical activity in U.S. adults. Am J Prev Med, 23(2 Suppl):74–79. 10. Berrigan D, Troiano RP, McNeel T, DiSogra C, Ballard-Barbash R (2006). Active transportation increases adherence to activity recommendations. Am J Prev Med, 31(3): 210–216. 11. Block D, Kouba J (2006). A Comparison of the availability and affordability of a market basket in two communities in the Chicago area. Public Health Nutr, 9(7):837–845. 12. Boarnet M, Crane R (2001). The influence of land use on travel behavior: specification and estimation strategies. Transp Res, A 35:823–845. 13. Botchwey ND, Hobson SE, Dannenberg AL et al. (2009). A model curriculum for a course on the built environment and public health: training for an interdisciplinary workforce. Am J Prev Med 36(2 Suppl):S63–S71. 14. Bouchard C (2008). Gene-environment interactions in the etiology of obesity: defining the fundamentals. Obesity (Silver Spring), 16 Suppl, 3:S5–S10. 15. Bouchard C, Agurs-Collins T (2008). Studying gene-behavior interactions: summary of recommendations. Obesity (Silver Spring), 16 Suppl 3:S95–S96. 16. Boyle P, Zaridze DG, Smans M (1985). Descriptive Epidemiology of colorectal cancer. Int J Cancer, 36(1):9–18. 17. Brownson RC, Hagood L, Lovegreen SL et al. (2005). A multilevel ecological approach to promoting walking in rural communities. Prev Med, 41(5–6):837–842. 18. Brownson RC, Housemann RA, Brown DR et al. (2000). Promoting physical activity in rural communities: walking trail access, use, and effects. Am J Prev Med, 18(3):235–241. 19. Burdette HL, Whitaker RC (2004). Neighborhood playgrounds, fast food restaurants, and crime: relationships to overweight in low-income preschool children. Prev Med, 38(1): 57–63. 20. Cao SX, Dhahbi JM, Mote PL, Spindler SR (2001). Genomic profiling of short- and longterm caloric restriction effects in the liver of aging mice. Proc Natl Acad Sci U S A, 98(19):10630–10635. 21. Cao X, Mokhtarian PL, Handy SL (2007). Do changes in neighborhood characteristics lead to changes in travel behavior? A structural equations modeling approach. Transportation, 34(5):535–556. 22. Cao X, Mokhtarian PL, Handy SL. (2009). Examining the impacts of residential selfselection on travel behavior: a focus on empirical findings. Transport Rev (in press). 23. Cao XY, Handy SL, Mokhtarian PL (2006). The influences of the built environment and residential self-selection on pedestrian behavior: evidence from Austin, TX. Transportation, 33(1):1–20. 24. Carrasco JA, Hogan B, Wellman B, Miller EJ (2008a). Agency in social activity interactions: the role of social networks in time and space. Tijdschr Econ Soc Geogr, 99(5): 562–583. 25. Carrasco JA, Hogan B, Wellman B, Miller EJ (2008b). Collecting social network data to study social activity-travel behavior: an egocentric approach. Environ Plann B Plann Des, 35(6):961–980. 26. Carrasco JA, Miller EJ, Wellman B (2008). How far and with whom do people socialize? Empirical evidence about distance between social network members. Transp Res Rec, 2076:114–122.
10
Geographic and Contextual Effects
291
27. Carver A, Timperio A, Crawford D (2008). Playing it safe: the influence of neighbourhood safety on children’s physical activity. A review. Health Place, 14(2):217–227. 28. Catlin TK, Simoes EJ, Brownson RC (2003). Environmental and policy factors associated with overweight among adults in Missouri. Am J Health Promot, 17(4): 249–258. 29. Cervero R, Duncan M (2003). Walking, bicycling, and urban landscapes: evidence from the San Francisco Bay Area. Am J Public Health, 93(9):1478–1483. 30. Christakis NA, Fowler JH (2007). The spread of obesity in a large social network over 32 years. N Engl J Med, 357(4):370–379. 31. Chung C, Myers SL (1999). Do the poor pay more for food? An analysis of grocery store availability and food price disparities. J Consum Aff, 33(2):276–296. 32. Cohen-Cole E, Fletcher JM (2008). Is obesity contagious? Social networks vs. environmental factors in the obesity epidemic. J Health Econ, 27(5):1382–1387. 33. Coveney J, O’Dwyer LA (2009). Effects of mobility and location on food access. Health Place, 15(1):45–55. 34. Crockett EG, Clancy KL, Bowering J (1992). Comparing the cost of a thrifty food plan market basket in three areas of New York State. J Nutr Educ, 24(1):71S–78S. 35. Cummins S, Findlay A, Petticrew M, Sparks L (2005). Healthy Cities: the impact of food retail led regeneration on food access, choice and retail structure. Built Environ, 31(4): 288–301. 36. Cummins S, Findlay A, Petticrew M, Sparks L (2008). Reducing inequalities in health and diet: the impact of food retail development. Environ Plan A, 40(2):402–422. 37. Cummins S, MacIntyre S (2006). Food environments and obesity––neighbourhood or Nation? Int J Epidemiol, 35(1):100–104. 38. de Sanjose S, Diaz M, Castellsague X et al. (2007). Worldwide prevalence and genotype distribution of cervical human papillomavirus DNA in women with normal cytology: a metaanalysis. Lancet Infect Dis, 7(7):453–459. 39. DHHS and USDA (2005). Dietary guidelines for Americans, 6th edn. U.S Government Printing Office, Washington, DC. 40. Duczmal L, Assuncao R (2004). A simulated annealing strategy for the detection of arbitrarily shaped spatial clusters. Comput Stat Data Anal, 45:269–286. 41. Dunton GF, Berrigan D, Ballard-Barbash R, Graubard BI, Atienza AA (2008). Social and physical environments of sports and exercise reported among adults in the American Time Use Survey. Prev Med, 47(5):519–524. 42. Dunton GF, Berrigan D, Graubard BI, Ballard-Barbash R, Atienza AA. (2009). Social and physical environmental influences on the intensity and duration of exercise in a U.S. time use survey. Med Sci Sports Exerc, 41(9):1698–1705. 43. Dunton GF, Jamner MS, Cooper DM (2003). Assessing the perceived environment among minimally active adolescent girls: validity and relations to physical activity outcomes. Am J Health Promot, 18(1):70–73. 44. Dunton GF, Whalen CK, Jamner LD, Floro JN (2007). Mapping the social and physical contexts of physical activity across adolescence using ecological momentary assessment. Ann Behav Med, 34(2):144–153. 45. Elliott P, Wakefield JC, Best NG, Briggs DJ (2000). Spatial Epidemiology: methods and applications. Oxford University Press, New York. 46. Ewing R, Brownson RC, Berrigan D (2006). Relationship between urban sprawl and weight of United States youth. Am J Prev Med, 31(6):464–474. 47. Ewing R, Cervero R (2001). Travel and the built environment: synthesis. Transp Res Rec, 1780:87–114. 48. Ewing R, Haliyur P, Page GW (1994). Getting around a traditional city, a suburban planned unit development, and everything in between. Transp Res Rec, 1466:53–62. 49. Ewing R, Schmid T, Killingsworth R, Zlot A, Raudenbush S (2003). Relationship between urban sprawl and physical activity, obesity, and morbidity. Am J Health Promot, 18(1): 47–57.
292
D. Berrigan et al.
50. Feder ME, Bennett AF, Burggren WW, Huey RB, (eds.), (1987). New directions in physiological ecology. Cambridge University Press, Cambridge. 51. Ferrari P, Friedenreich C, Matthews CE (2007). The role of measurement error in estimating levels of physical activity. Am J Epidemiol, 166(7):832–840. 52. Fisher BD, Strogatz DS (1999). Community measures of low-fat milk consumption: comparing store shelves with households. Am J Public Health, 89(2):235–237. 53. Frank L, Engelke P, Schmid T (2003). Health and community design: the impact of the built environment on physical activity, p. 250. Island Press, Washington, DC. 54. Frank L, Pivo G (1994). Impacts of mixed use and density on utilization of three modes of travel: single-occupant vehicle, transit, and walking. Transp Res Rec, 1466:44–52. 55. Frank LD, Andresen MA, Schmid TL (2004). Obesity relationships with community design, physical activity, and time spent in cars. Am J Prev Med, 27(2):87–96. 56. Frank LD, Engelke PO (2001). The built environment and human activity patterns: exploring the impacts of urban form on public health. J Plann Lit, 16(2):202–218. 57. Frank LD, Schmid TL, Sallis JF, Chapman J, Saelens BE (2005). Linking objectively measured physical activity with objectively measured urban form: findings from SMARTRAQ. Am J Prev Med, 28(2 Suppl 2):117–125. 58. Frerichs RR (2009). John Snow http://www.ph.ucla.edu/EPI/snow.html. Accessed: 12 Feb 2009. 59. Furey S, Farley H, Strugnell C (2002). An Investigation into the availability and economic accessibility of food items in rural and urban areas of Northern Ireland. Int J Consum Stud, 26(4):313–321. 60. Giles-Corti B, King AC (2009). Creating active environments across the life course: “thinking outside the square”. Br J Sports Med, 43(2):109–113. 61. Giles-Corti B, Timperio A, Bull F, Pikora T (2005). Understanding physical activity environmental correlates: increased specificity for ecological models. Exerc Sport Sci Rev, 33(4):175–181. 62. Glanz K, Sallis JF, Saelens BE, Frank LD (2007). Nutrition Environment Measures Survey in stores (NEMS-S): development and evaluation. Am J Prev Med, 32(4):282–289. 63. Glass TA, McAtee MJ (2006). Behavioral Science at the crossroads in public health: extending horizons, envisioning the future. Soc Sci Med, 62(7):1650–1671. 64. Glaz J, Naus J, Wallenstein S (2001). Scan Statistics, p. 370. Springer Verlag, New York. 65. Guy CM, Clarke G, Eyre H (2004). Food retail change and the growth of “Food deserts”: a case study of Cardiff. Int J Retail Distrib Manag, 32(2):72–88. 66. Han D, Rogerson PA, Nie J et al. (2004). Geographic clustering of residence in early life and subsequent risk of breast cancer (United States). Cancer Causes Control, 15(9):921–929. 67. Handy S, Cao X, Mokhtarian PL (2006). Self-selection in the relationship between the built environment and walking. J Am Plann Assoc, 72(1):55–74. 68. Handy SL (1996). Urban form and pedestrian choices: a study of Austin neighborhoods. Transp Res Rec, 1552:135–144. 69. Handy SL, Boarnet MG, Ewing R, Killingsworth RE (2002). How the built environment affects physical activity: views from urban planning. Am J Prev Med, 23(2 Suppl): 64–73. 70. Handy SL, Cao X, Mokhtarian PL (2008). The causal influence of neighborhood design on physical activity within the neighborhood: evidence from Northern California. Am J Health Promot, 22(5):350–358. 71. Hayes LR (2000). Are prices higher for the poor in New York city? J Consum Pol, 23: 127–152 72. Heffernan R, Mostashari F, Das D, Karpati A, Kulldorff M, Weiss D (2004). Syndromic surveillance in public health practice, New York City. Emerg Infect Dis, 10(5): 858–864. 73. Hendrickson D, Smith C, Eikenberry N (2006). Fruit and vegetable access in four lowincome food deserts communities in Minnesota. Agric Human Values, 23:371–383.
10
Geographic and Contextual Effects
293
74. Horowitz CR, Colson KA, Hebert PL, Lancaster K (2004). Barriers to buying healthy foods for people with diabetes: evidence of environmental disparities. Am J Public Health, 94(9):1549–1554. 75. Hosler AS, Varadarajulu D, Ronsani AE, Frederick BL, Fisher BD (2006). Low-fat milk and high-fiber bread availability in food stores in urban and rural communities. J Public Health Manag Pract, 12(6):556–562. 76. Huang L, Kulldorff M, Gregorio D (2007). A spatial scan statistic for survival data. Biometrics, 63(1):109–118. 77. Huang L, Stinchcomb D, Pickle L, Dill J, Berrigan D. (2009). Identifying clusters of active Transportation using spatial scan statistics. Am J Prev Med, 37(2):157–166. 78. Huang TT, Glass TA (2008). Transforming research strategies for understanding and preventing obesity. JAMA, 300(15):1811–1813. 79. Institute of Medicine (2005). Does the built environment influence physical activity? Examining the evidence – special report 282. National Academies Press, Washington, DC. 80. Irwin LG, Johnson JL, Henderson A, Dahinten VS, Hertzman C (2007). Examining how contexts shape young children’s perspectives of health. Child Care Health Dev, 33(4): 353–359. 81. Jefferis ES (1998). A multi-method exploration of crime hot spots: SaTScan results. National Institute of Justice, Crime Mapping Research Center, Washington, DC. 82. Jetter KM, Cassady DL (2006). The availability and cost of healthier food alternatives. Am J Prev Med, 30(1):38–44. 83. Khoury MJ, Davis R, Gwinn M, Lindegren ML, Yoon P (2005). Do we need genomic research for the prevention of common diseases with environmental causes? Am J Epidemiol, 161(9):799–805. 84. King ES, Rimer BK, Trock B, Balshem A, Engstrom P (1990). How valid are mammography self-reports? Public Health Rep, 80(11):1386–1388. 85. King WC, Belle SH, Brach JS, Simkin-Silverman LR, Soska T, Kriska AM (2005). Objective measures of neighborhood environment and physical activity in older women. Am J Prev Med, 28(5):461–469. 86. Kipnis V, Subar AF, Midthune D et al. (2003). Structure of dietary measurement error: results of the OPEN biomarker study. Am J Epidemiol, 158(1):14–21. 87. Kitamura R, Mokhtarian PL, Laidet L (1997). A micro-analysis of land use and travel in five neighborhoods in the San Francisco Bay Area. Transportation, 24:125–158. 88. Knowler WC, Barrett-Connor E, Fowler SE et al. (2002). Reduction in the incidence of type 2 Diabetes with lifestyle intervention or metformin. N Engl J Med, 346(6):393–403. 89. Krizek KJ (2000). A pre-test/post-test strategy for researching neighborhood-scale urban form and travel behavior. Transp Res Rec, 1722:48–55. 90. Krizek KJ (2003). Residential relocation and changes in urban travel: does neighborhoodscale urban form matter? J Am Plann Assoc, 69(3):265–281. 91. Kruger J, Ham SA, Berrigan D, Ballard-Barbash R (2008). Prevalence of Transportation and leisure walking among U.S. adults. Prev Med, 47(3):329–334. 92. Kulldorff M (1997). A spatial scan statistic. Commun Stat Theory Methods, 26(6): 1481–1496. 93. Kulldorff M (2006). SaTScan Software for spatial temporal space time statistics. http://www.satscan.org/reviewed 2/3/2010 94. Kulldorff M (2006). Tests of spatial randomness adjusted for an inhomogeneity: a general framework. J Am Stat Assoc, 101(475):1289–1305. 95. Kulldorff M, Song C, Gregorio D, Samociuk H, DeChello L (2006). Cancer map patterns: are they random or not? Am J Prev Med, 30(2 Suppl):S37–S49. 96. Langbein LI, Lichtman AJ (1978). Ecological inference. Sage Publications, Beverly Hills, CA.
294
D. Berrigan et al.
97. Laraia BA, Siega-Riz AM, Kaufman JS, Jones SJ (2004). Proximity of supermarkets is positively associated with diet quality index for pregnancy. Prev Med, 39(5):869–875. 98. Lopez R (2004). Urban sprawl and risk for being overweight or obese. Am J Public Health, 94(9):1574–1579. 99. Lovasi GS, Moudon AV, Pearson AL et al. (2008). Using built environment characteristics to predict walking for exercise. Int J Health Geogr, 7:10. 100. Macbeth AG (1999). Bicycle lanes in Toronto. ITE J, 69(4):38–46. 101. Macintyre S, Ellaway A, Cummins S (2002). Place effects on health: how can we conceptualise, operationalise and measure them? Soc Sci Med, 55(1):125–139. 102. Margai F, Henry N (2003). A community-based assessment of learning disabilities using environmental and contextual risk factors. Soc Sci Med, 56(5):1073–1085. 103. McKenzie TL, Marshall SJ, Sallis JF, Conway TL (2000) Leisure-time physical activity in school environments: an observational study using SOPLAY. Prev Med, 30(1):70–77. 104. McKeown-Eyssen G (1994). Epidemiology of colorectal cancer revisited: are serum triglycerides and/or plasma glucose associated with risk? Cancer Epidemiol Biomarkers Prev, 3(8):687–695. 105. McKinnon RA, Reedy J, Handy SL, Rodgers AB. (2009). Measuring the food and physical activity environments: shaping the research agenda. Am J Prev Med, 36(4 suppl):S81–S85. 106. McLeroy KR, Bibeau D, Steckler A, Glanz K (1988). An ecological perspective on health promotion programs. Health Educ Q, 15(4):351–377. 107. Michels KB (2001). A renaissance for measurement error. Int J Epidemiol, 30(3):421–422. 108. Milat AJ, Stubbs J, Engelhard S, Weston P, Giles-Corti B, Fitzgerald S (2002). Measuring physical activity in public open space––an electronic device versus direct observation. Aust N Z J Public Health, 26(1):50–51. 109. Mokhtarian PL, Cao X (2008). Examining the impacts of residential self-selection on travel behavior: a focus on methodologies. Transp Res, B 42(3):204–228. 110. Moore LV, Diez Roux AV (2006). Associations of neighborhood characteristics with the location and type of foodstores. Am J Public Health 96(2):325–331. 111. Morland K, Wing S, Diez Roux A, Poole C (2002). Neighborhood characteristics associated with the location of foodstores and food service places. Am J Prev Med, 22(1):23–29. 112. Morland K, Diez Roux A, Wing S (2006). Supermarkets, other foodstores, and obesity: the atheriosclerosis risk in communities study. Am J Prev Med, 30(4):333–339. 113. Morris PM, Leuhauser L, Campbell C (1992). Food security in rural America: a study of the availability and costs of food. J Nutr Educ, 24(1):52S–58S. 114. Nader PR (2003). Frequency and intensity of activity of third-grade children in physical education. Arch Pediatr Adolesc Med, 157(2):185–190. 115. Nkhoma ET, Hsu CE, Hunt VI, Harris AM (2004). Detecting spatiotemporal clusters of accidental poisoning mortality among Texas counties, U.S., 1980–2001. Int J Health Geogr, 3(1):25. 116. Novembre J, Johnson T, Bryc K et al. (2008). Genes mirror geography within Europe. Nature, 456(7218):98–101. 117. Oakes JM, Masse LC, Messer LC (2009). Work Group III: methodological issues in research on the food and physical activity environment: addressing data complexity. Am J Prev Med, 36(4 suppl):S177–S181. 118. O’Rahilly S, Farooqi IS (2006). Genetics of obesity. Philos Trans R Soc Lond B Biol Sci, 361(1471):1095–1105. 119. Papas MA, Alberg AJ, Ewing R, Helzlsouer KJ, Gary TL, Klassen AC (2007). The built environment and obesity. Epidemiol Rev, 29:129–143. 120. Parkin DM (2004). International variation. Oncogene, 23(38):6329–6340. 121. Parsons Brinckerhoff Quade & Douglas Inc. (1994). Building orientation. A supplement to the Pedestrian Environment. Vol. 4B. 1000 Friends of Oregon, Portland, OR. 122. Parsons Brinckerhoff Quade & Douglas Inc. (1993). The pedestrian environment. Vol. 4A. 1000 Friends of Oregon, Portland, OR.
10
Geographic and Contextual Effects
295
123. Pickle LW, Su Y (2002). Within-state geographic patterns of health insurance coverage and health risk factors in the United States. Am J Prev Med, 22(2):75–83. 124. Pickle LW, Szczur M, Lewis DR, Stinchcomb DG (2006). The crossroads of GIS and health information: a workshop on developing a research agenda to improve cancer control. Int J Health Geogr, 5:51. 125. Pollack LA, Gotway CA, Bates JH et al. (2006). Use of the spatial scan statistic to identify geographic variations in late stage colorectal cancer in California (United States). Cancer Causes Control, 17(4):449–457. 126. Porta M (2008). Dictionary of Epidemiology 5th edn, pp. 320. Oxford University Press, New York. 127. Powell LM, Slater S, Mirtcheva D, Bao Y, Chaloupka FJ (2007). Food store availability and neighborhood characteristics in the United States. Prev Med, 44(3):189–195. 128. Raja S, Ma C, Yadav P (2008). Beyond food deserts: measuring and mapping racial disparities in neighborhood food environments. J Plann Educ Res, 27:469–482. 129. Reisig VM, Hobbiss A (2000). Food deserts and how to tackle them: a study of one city’s approach. Health Educ J, 59:137–149. 130. Replogle M (1995). Integrating pedestrian and bicycle factors into regional Transportation planning models: summary of the state-of-the-art and suggested steps forward. Environmental Defense Fund, Washington, DC. 131. Resnicow K, Odom E, Wang T et al. (2000). Validation of three food frequency questionnaires and 24-hour recalls with serum carotenoid levels in a sample of African-American adults. Am J Epidemiol, 152(11):1072–1080. 132. Reynolds KD, Wolch J, Byrne J et al. (2007). Trail characteristics as correlates of urban trail use. Am J Health Promot, 21(4 Suppl):335–345. 133. Riitters KH, Coulston JW (2005). Hot spots of perforated forest in the eastern United States. Environ Manage, 35(4):483–492. 134. Roff DA (1997). Evolutionary quantitative genetics. Chapman and Hall, New York. 135. Rose D, Richards R (2004). Food store access and household fruit and vegetable use among participants in the US Food Stamp Program. Public Health Nutr, 7(8): 1081–1088. 136. Rose G (1992). The strategy of preventive medicine. Oxford University Press, New York. 137. Sallis JF, Bauman A, Pratt M (1998). Environmental and policy interventions to promote physical activity. Am J Prev Med, 15(4):379–397. 138. Sallis JF, Owen N (1996). Ecological models. In: Glanz K, Lewis FM, Rimer BK (eds.), Health behavior and health education: theory, research and practice, pp. 403–424. JosseyBass, San Francisco, CA. 139. Sallis JF, Story M, Lou D (2009). Study designs and analytic strategies for environmental and policy research on obesity, physical activity, and diet: recommendations from a meeting of experts. Am J Prev Med, 36(2 Suppl):S72–S77. 140. Salmon J, Salmon L, Crawford DA, Hume C, Timperio A (2007). Associations among individual, social, and environmental barriers and children’s walking or cycling to school. Am J Health Promot, 22(2):107–113. 141. Salmon J, Timperio A (2007). Prevalence, trends and environmental influences on child and youth physical activity. Med Sport Sci, 50:183–199. 142. Salvy SJ, Bowker JW, Roemmich JN et al. (2008a) Peer influence on children’s physical activity: an experience sampling study. J Pediatr Psychol, 33(1):39–49. 143. Salvy SJ, Roemmich JN, Bowker JC, Romero ND, Stadler PJ, Epstein LH (2009). Effect of peers and friends on youth physical activity and motivation to be physically active. J Pediatr Psychol, 34(2):217–255. 144. Sanchez A, Norman GJ, Sallis JF, Calfas KJ, Cella J, Patrick K (2007). Patterns and correlates of physical activity and nutrition behaviors in adolescents. Am J Prev Med, 32(2):124–130.
296
D. Berrigan et al.
145. Sgro CM, Hoffmann AA (2004). Genetic correlations, tradeoffs and environmental variation. Heredity, 93(3):241–248. 146. Sherman JE, Fetters TL (2007). Confidentiality concerns with mapping survey data in reproductive health research. Stud Fam Plann, 38(4):309–321. 147. Sloane DC, Diamant AL, Lewis LB, Yancey AK, Flynn G, Miller LM, McCarthy WJ, Guinyard JJ, Cousineau MR (2003). Improving the nutritional resource environment for healthy living through community-based participatory research. J Gen Intern Med, 18: 568–575. 148. Smoyer-Tomic KE, Spence JC, Amrhein C (2006). Food deserts in the prairies? Supermarket accessibility and neighborhood need in Edmonton, Canada. Prof Geogr, 58(3):307–326. 149. Song C, Kulldorff M (2003). Power evaluation of disease clustering tests. Int J Health Geogr, 2(1):9. 150. Sudakin DL, Horowitz Z, Giffin S (2002). Regional variation in the incidence of symptomatic pesticide exposures: applications of geographic information systems. J Toxicol Clin Toxicol, 40(6):767–773. 151. Swinburn B, Egger G, Raza F (1999). Dissecting obesogenic environments: the development and application of a framework for identifying and prioritizing environmental interventions for obesity. Prev Med, 29(6 Pt 1):563–570. 152. Tagnon I, Blot WJ, Stroube RB et al. (1980). Mesothelioma associated with the shipbuilding industry in coastal Virginia. Cancer Res, 40(11):3875–3879. 153. Tango T (2000). A test for spatial disease clustering adjusted for multiple testing. Stat Med, 19(2):191–204. 154. Thomas A, Carlin BP (2003). Late detection of breast and colorectal cancer in Minnesota counties: an application of spatial smoothing and clustering. Stat Med, 22(1):113–127. 155. Thompson FE, Subar AF (2008). Dietary assessment methodology. In: Coulston AM, Boushey CJ (eds.), Nutrition in the prevention and treatment of disease, 2nd edn. Academic Press, San Diego, CA. 156. Timperio A, Salmon J, Telford A, Crawford D (2005). Perceptions of local neighbourhood environments and their relationship to childhood overweight and obesity. Int J Obes (Lond), 29(2):170–175. 157. Trock B, Lanza E, Greenwald P (1990). Dietary fiber, vegetables, and colon cancer: critical review and meta-analyses of the Epidemiologic evidence. J Natl Cancer Inst, 82:650–661. 158. Troiano RP, Berrigan D, Dodd KW, Masse LC, Tilert T, McDowell M (2008). Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc, 40(1): 181–188. 159. Turner L, Mermelstein R, Flay B (2004) Individual and contextual influences on adolescent smoking. Ann N Y Acad Sci, 1021:175–197. 160. van der Horst K, Oenema A, Ferreira I et al. (2007). A systematic review of environmental correlates of obesity-related dietary behaviors in youth. Health Educ Res, 22(2): 203–226. 161. van der Horst K, Paw MJ, Twisk JW, Van Mechelen W (2007). A brief review on correlates of physical activity and sedentariness in youth. Med Sci Sports Exerc, 39(8): 1241–1250. 162. VanWey LK, Rindfuss RR, Gutmann MP, Entwisle B, Balk DL (2005). Confidentiality and spatially explicit data: concerns and challenges. Proc Natl Acad Sci U S A, 102(43): 15337–15342. 163. Waller LA, Gotway CA (2004). Applied spatial statistics for public health data. John Wiley & Sons, Hoboken, NJ. 164. Walter SD (2000). Disease mapping: a historical perspective. In: Elliott P, Wakefield J, Best N, Briggs DJ (eds.), Spatial Epidemiology: Methods and Applications, pp. 223–252. Oxford University Press, Oxford. 165. Wechsler H, Basch CE, Zybert P, Lantigua R, Shea S (1995). The availability of low-fat milk in an inner-city Latino Community: implications for nutrition education. Am J Public Health, 85(12):1690–1692.
10
Geographic and Contextual Effects
297
166. Wendel-Vos GC, Schuit AJ, de NR, Boshuizen HC, Saris WH, Kromhout D (2004). Factors of the physical environment associated with walking and bicycling. Med Sci Sports Exerc, 36(4):725–730. 167. Wendel-Vos W, Droomers M, Kremers S, Brug J, van Lenthe F (2007). Potential environmental determinants of physical activity in adults: a systematic review. Obes Rev, 8(5):425–440. 168. Wrigley N, Warm D, Margetts B (2003). Deprivation, diet, and food-retail access: findings from the Leeds “food deserts” study. Environ Plann, A 35(1):151–188. 169. Zenk SN, Schulz AJ, Israel BA, James SA, Bao S, Wilson ML (2005b). Neighborhood racial composition, neighborhood poverty, and the spatial accessibility of supermarkets in metropolitan Detroit. Am J Public Health, 95(4):660–667. 170. Zenk SN, Schulz AJ, Israel BA, James SA, Bao S, Wilson ML (2006). Fruit and vegetable access differs by community racial composition and socioeconomic position in Detroit, Michigan. Ethn Dis, 16(Winter):275–280.
Subject Index
Note: The letters ‘f’ and ‘t’ following the locators refer to figures and tables respectively.
A Acute lymphoblastic leukemia (ALL), 31 Acute myeloid leukemia (AML), 26f, 31 Adenosine monophosphate deaminase 1 (AMPD1), 95 Adipokines and gut hormones adiponectin, 144–145 ghrelin, 147–148 leptin, 141–142 resistin, 145–146 visfatin, 146–147 Adiponectin, 144–145 ADIPO-R1 and ADIPO-R2, receptors, 144 and cancer risk, 144 in carcinogenesis, 144–145 30-kDa protein hormone, 144 liver neoglucogenesis, 144 Aerobics commercial, 247 exercise, 163, 204, 227–228, 236, 238 training studies, 163 vs. usual care in breast cancer survivors, 228 genes for anaerobic activity, 97 and resistance exercise, 227 studies examining aerobic phenotypes, 95 swimming or water aerobics, 247 Agricultural Health Study cohort, 160–161 AICR, see American Institute of Cancer Research (AICR) report ALL, see Acute lymphoblastic leukemia (ALL) Alpha-2A-adrenoceptor gene (ADRA2A), 95 Alpha actinin-3 (ACTN3), 94, 97 American Cancer Society Cancer Prevention Study II (CPS II) Nutrition Cohort, 111–112, 204, 209, 239 American College of Sports Medicine, 202, 229
American Institute of Cancer Research (AICR) report, 2, 28 American Time Use Survey, 278 2-amino-3,8-dimethylimidazo[4,5-f] quinoxaline (MeIQx), 160 2-amino-1-methyl-6-phenylimidazo[4,5-b] pyridine (PhIP), 132t, 160–161 AML, see Acute myeloid leukemia (AML) AMP-activated protein kinase (AMPK) pathway, 109–110, 134f–135f, 137, 144, 162 Androgen ablation therapy, 116 Androgen hypothesis, 141 Angiotensin-converting enzyme (ACE), 94–95, 97, 103 Anorexia nervosa studies, 183–184 eating disorder, 184 and risk of cancer, 185t–186t voluntary starvation, 184 Apc Min/+ mutation, 159, 163 Auto-immune diseases, 31 B Banding surgery, laproscopic, 156 Bariatric surgery, 3, 34–35, 153, 159 Barrett’s esophagus, 22–23, 79 Base excision repair (BER), 156–157 Behavioral Risk Factor Surveillance System (BRFSS) survey, 50, 52, 271, 285f Behavior, energy balance, and cancer diet, nutrition, and cancer risk, 239–243 intervening to change lifestyle behaviors, 234–235 See also Lifestyle behaviors, intervening to change interventions to change dietary behaviors, 250–254 See also Dietary behaviors, interventions to change
N.A. Berger (ed.), Cancer and Energy Balance, Epidemiology and Overview, Energy Balance and Cancer 2, DOI 10.1007/978-1-4419-5515-9, C Springer Science+Business Media, LLC 2010
299
300 Behavior, energy balance, and cancer (cont.) interventions to change physical activity and sedentary behavior, 247–250 See also Physical activity and sedentary behavior interventions to change sleep, 254–255 See also Sleep and cancer risk physical activity/diet/sleep, impact on obesity and cancer, 233–234 overweight or obese, definition, 233 sedentary behavior/physical activity and cancer risk, 236–239 sleep and cancer risk, 243–247 Behaviors class of, 239 dietary behaviors, see Dietary behaviors, interventions to change eating, 92 energy balance-related, 235, 267–289 irregular sleep–wake, 101 lifestyle behaviors, see Lifestyle behaviors, intervening to change sedentary, see Physical activity and sedentary behavior See also Behavior, energy balance, and cancer Behavior-setting theory and social context of energy balance, 276–279 American Time Use Survey, 277–278 Cohen-Cole and Fletcher’s discussion, 279 human–environment interaction approach, 277 Study of Adolescent Health Behaviors (ADD Health), 278 time use and transportation research studies, 278 See also Geographic and contextual effects Beta2-adrenergic receptor (ADRB2) gene, 95–96 Biomarkers of oxidative lipid damage, 155–156 higher circulating/serum Ox-LDL levels, 156 lipid peroxidation, 155 lipid peroxyl radicals, 156 of oxidative nucleic acid (DNA) damage, 156–157 cancer cells, 156 ‘malnubesity,’ 157 8-oxo-dG, 156 ROS levels, role of, 157
Subject Index of oxidative protein damage, 155 protein carbonyls, 155 proteolysis, 155 of systemic oxidative stress: F2-isoprostanes, 158 8-iso-prostaglandin F2α (8-iso-PGF2α), 158 BMI, see Body mass index (BMI) Body mass index (BMI), 2, 25f, 46f–47f, 71, 73–74f, 88–90, 98, 105, 111, 113, 129, 156, 184, 187, 191t, 194, 207–208, 210, 227, 233, 243, 252–253 Bradykinin beta 2 receptor (BDKRB2), 95 Breast cancer aerobics exercise vs. usual care in breast cancer survivors, 228 evidences, 8 family history data, 9 genetic factors, role of, 10 incidence and risk factors, 7 leptin, role in, 143 life cycle and characteristics of women, 8 physical activity, 8 postmenopausal, 113–114 premenopausal, 112–113 prognosis and obesity in younger women, 221 race and ethnicity, 9 studies on, 204–205 Breast Cancer Detection Demonstration Project Follow-Up Study, 205 BRFSS, see Behavioral Risk Factor Surveillance System (BRFSS) survey Butein, 194–195 C Californial Health Interview Survey (CHIS), 284 Caloric restriction and cancer, 181–195 animal studies, 181–183 decrease mortality, 183 effect of caloric restriction on cancer, 182 lowers spontaneous/induced/ transplanted tumors, 182 McCay’s findings, 182 normal diet mice vs. restricted diet mice, 181–182 human studies, 183–191
Subject Index anorexia nervosa studies, 183–184 ecologic studies, 183 famine studies, 184–191 mimetic molecules, 194 2-deoxyglucose (2DG), 194 molecules from plants, 194 other molecules, 194 SIR2 family, 194 potential mechanisms, 192–193 beneficial effects, 193 endocrinologic changes, 192 hormesis, 193 hormones and growth factors, 192 IGF-1 and insulin levels, 192 intracellular cell-autonomous signaling pathways, 193 SIRT1, 193 Cancer and Leukemia Group B (CALGB) 89803, 221 Cancer and obesity in Asia, see Obesity and cancer in Asia epidemiology in racial/ethnic minorities, 45–59 See also Obesity and cancer epidemiology at initial diagnosis and cancer survival, 220–221 mechanisms, 129–164 risk and obesity, see Obesity and cancer risk Cancer and physical activity, see Physical activity and cancer Carcinogenesis breast, 151 colon, 21 colorectal, 144 endometrial, 111 obesity-related, 130, 133f–134f, 138, 141, 150–153, 157, 160 prostate, 140 thyroid, 16 Carcinogens/DNA damage, dietary intake of HCAs, 160–161 PAHs, 161–162 therapeutic opportunities diet, 162 exercise, 162–163 pharmaceuticals, 163–164 Cardiac rehabilitation and genetics of exercise performance (CAREGENE) study, 96 “Certified Cancer Exercise Trainers,” 229
301 Chemokine C–C motif (CCL2), see Monocyte chemotactic protein-1 (MCP-1) Childhood obesity, 34, 47, 87–88 China Health and Nutrition Survey, 70 Chronic exercise, 163 Chronic lymphocytic leukemia (CLL), 25f–26f, 31 Chronic myeloid leukemia (CML), 26f, 31 Chronic obstructive pulmonary disease (COPD), 95–96, 195 Circadian sleep disorders, 100 Class of behaviors, 239 CLL, see Chronic lymphocytic leukemia (CLL) Clock transcription factor deficiency, 101 CML, see Chronic myeloid leukemia (CML) Cocaine and amphetamine related transcript (CART), 92, 142 Colon cancer, 205–206 factors, weaker results, 206 NIH-AARP Diet and Health Study, 206 Colorectal cancers, 114–115 BMI and rectal cancer, 20 BMI/fat distribution/colon/colorectal cancers, 20–21 early life anthropometry, 21 incidence and risk factors, 19 methodologic approaches, 21 and weight risk in women, factors affecting, 21 C-reactive protein (CRP), 151, 227 higher levels, 151 117.5-kDa protein, 151 in obesity-related carcinogenesis, 151 plasma CRP concentrations, 151 Cytokines and chemokines CRP, 151 IL-6, 150–151 MCP-1, 152 TNF-α, 148–150 D Danish Psychiatric Case Register, 184 Deficiency(ies) adiponectin, 144 alpha actinin-3, 94 iodine, 16 leptin, 102, 142 orexin, 101 vitamin D, 225 Diabetes Genetics Initiative (DGI), 90
302 Diet ad libitum, 182 Chinese, 69 energy-dense, 56 healthy, 222, 224–225, 229, 233 higher calorie, 56 high-fat, 101, 152, 159 low-calorie, 107, 182 low-carbohydrate, 253 low-fat, 50, 224, 253, 256 Mediterranean, 253–254 plant-based, 228 semi-starvation, 184 suboptimal, 228 “Western,” 50, 68, 224 Dietary behaviors, interventions to change, 250–254 breakfast, 251 caloric restriction, 252 decreasing dietary fat, 252 and increasing fiber, 252 eating in front of TV, 252 fast food, 250 increased soy intake, 254 increasing fruit and vegetable intake, 253 meal timing “nibbling or snacking,” 251 Mediterranean diet, 253–254 portion sizes, 251 sugar-sweetened beverages, 250–251 Dietary patterns, 68, 224–225, 240, 284 Diet, nutrition, and cancer risk, 239–243 domains, levels, and measures of dietary intake, 240 macro/micro nutrients and cancer risk, 240–243 macronutrients, micronutrients, and cancer risk added sugar intake, 243 carbohydrates – fruit and vegetable intake, 242 dietary fat, 241 fiber and whole grain foods, 242 glycemic index (GT) and glycemic load (GL), 243 other micronutrients, 243 protein intake, 241 soy products, 241–242 sugar, 242–243 Dizygotic (DZ) twins, 100 Dual X-ray absorptiometry (DXA), 5 DXA, see Dual X-ray absorptiometry (DXA)
Subject Index E Eicosanoids, 153–154 20 carbon ω-6 or ω-3 fatty acids, 153 families, four, 153 metabolism pathway, 153 Endometrial cancer, 111–112, 206–207 candidate genes (CYP19A1/CYP17A1), risk, 111 evidence, 12 HRT, effect of obesity, 12 incidence and risk factors, 12 levels of estrogens, 111 association–physical activity, meta-analyses, 207 physical activity, 13 role of diabetes, 13 Energy balance and psycho-social/quality of life benefits, 227–228 aerobic and resistance exercise, 227 aerobic exercise vs. usual care in breast cancer survivors, 228 Energy balance/cancer prognosis/survivorship improving weight/nutrition/physical activity behaviors, 228–229 mechanisms linking energy balance, 226–227 C-reactive protein (CRP), 227 decreased serum insulin levels, 226 IGFBP-3, 226 IGF-I, 226 link between adipokines, 227 sex steroid hormones, 226–227 nutrition Life After Cancer Epidemiology, 224 and physical activity, 225–226 vitamin D and potential association with prognosis, 225 WHEL, 224 WINS, 224 obesity at initial diagnosis, 220–221 NIH-AARP Diet and Health Study, 220 obesity and breast cancer prognosis in younger women, 221 obesity and increased risk, studies on, 220 physical activity, 222–223 assessing activity in year prior to diagnosis, 223 moderate-intensity recreational activity after diagnosis, 223 psycho-social/quality of life benefits, 227–228 aerobic and resistance exercise, 227
Subject Index aerobic exercise vs. usual care in breast cancer survivors, 228 weight gain, 221–222 Cancer and Leukemia Group B (CALGB) 89803, 221 genetic differences in tumors among obese patients, 221 healthy diet and increasing physical activity, 222 Nurses’ Health Study, 221 Energy-consuming processes, 162 Environment definition, 269 and physical activity, 272–274 environmental measures, 273 ISI Web of Knowledge, 274 nationally representative survey (NHANES III) of US adults, 273 problems, 273–274 Environmental risk factors, obesity and cancer food intake/appetite regulation and genes, 91–92 adipocytes, 92 body weight regulation, hypothalmic neurons in, 91 diabetes gene, 92 genes controlling eating behavior, 92 leptin, role of, 92 regulation of blood glucose, CNS, 91 physical activity and genes, 92–98 ACE-insertion/deletion (I/D) polymorphism, role of, 94–95 ACE insertion/insertion (II) genotype group, 95 ACTN3 gene locus polymorphism, 97 acute inflammation, 93–94 ADRA2A, allele and genotype differences, 95 ADRB2 gene and polymorphism, 95–96 alpha actinin-3 deficiency, 94 anaerobic activity, genes for, 97 animal studies, advantage of, 94 chronic systemic inflammation, 94 COPD, genetic associations for, 96 ER22/23EK polymorphism, 97–98 genes associated with endurance performance, 95 HERITAGE family study, 97 human physical performance and health-related fitness, 93 IGF-I, polymorphism of, 96
303 inflammatory processes, effect of exercise on, 93 lack of exercise, 92–93 myostatin (GDF8) pathway, role of, 98 PVR, 95–96 QTL mapping, 94 sedentary behavior and physical activity, 93 Wagoner study, 96 Youth Risk Behavior Survey, 93 sleep and genes for sleep disturbances, 99–105 clock transcription factor deficiency, 101 condition of hypoxia, 103 effects of sleep deprivation, 98–99 familial inheritable diseases, 100 genetic control of sympathovagal balance, 103 genetics of sleep apnea, 100 HIF-1, role of, 103 higher REM density, 100 hypertension and obesity phenotypes, 103 insufficient sleep risk diseases, 99 irregular circadian pattern, 103 leptin deficiency, 102 obesity/metabolic syndrome/sleep apnea, associations between, 99 orexin-deficient mice, 101 OSA with hypertension, 104 PPARγ, 104 reduction in sleep time, 98 serotonin neurotransmission, 101 sleep apnea, 99 sleep architecture and sleep disorders, inheritable, 100 sleep–wake and circadian rhythm, genes affecting, 101 symptoms of narcolepsy, 102 UCPs, role of, 103–104 stomach flora, 106–107 Firmicutes and Bacteroidetes, 106 ‘lean microbiota,’ 107 ‘obese microbiota,’ 107 viral infection and thermoregulation, 105–106 AD-36 antibodies in male rhesus monkey, 105 increase of adiposity in animals, 105 thermoneutral zone (TNZ), 106 thermoregulation, 106
304 Esophageal cancer, 210 incidence and risk factors, 22 overall statement, 22 selected issues epidemiology persistence, 23 GERD and Barrett’s esophagus, role of, 23 Estrogen synthetase, 110 European prospective investigation into cancer and nutrition (EPIC) study, 18, 139, 205, 242 European type of semi-starvation diet, 184 Exercise aerobic, 163 chronic, 163 lifelong enhanced, 163 physical, 163 regular, 163 F Famine studies, 184–191 difference in behavior patterns, 187 Dutch Famine Study, 187 food deprivation, 187 involuntary starvation, 187 malnutrition, 184 Norwegian famine data, 184 peri-pubertal growth, 187 and risk of cancer, 188t–191t Fasting-induced adipose factor (FIAF), 92 Fast-twitch glycolytic fibers (Type IIb), 162 Fast-twitch oxidative fibers (Type IIa), 162 Fat–BMI relationships, 73 Fatty acid metabolism eicosanoids, 153–154 prostaglandins, 153–154 Fenton reaction, 155 Food environments and food deserts, 274–276 behavior-setting theory, 275 dietary guidelines for Americans, 274 quantitative research methods, 275 United Kingdom’s Nutrition Task Force’s Low Income Project, 274 See also Geographic and contextual effects Food, Nutrition, Physical Activity and the Prevention of Cancer: A Global Perspective, a 2007 report, 32 G Gallbladder cancer incidence and risk factors, 28 Gastric cancer, 111, 209–210 gastric cardia cancer, 79
Subject Index Gastroesophageal reflux disease (GERD), 23, 79, 130 Gastrointestinal malignancies, 111 Genetic epidemiology of obesity and cancer, 87–116 cancer-specific shared genetic effect breast, postmenopausal, 113–114 breast, premenopausal, 112–113 colorectal, 114–115 endometrial, 111–112 esophagus, 115 prostate, 115–116 stomach, 111 genes for obesity, 88–90 APOE and TGF-beta-1, 90 FTO, role of, 90 genome wide linkage scan, 89 recessive inheritance in RFPI study, 88 twin studies, 88 type II diabetes, 89 genetically influenced environmental risk factors food intake/appetite regulation and genes, 91–92 physical activity and genes, 92–98 sleep and genes for sleep disturbances, 99–105 stomach flora, 106–107 viral infection and thermoregulation, 105–106 genetic pathways networks of insulin growth factors, 108–110 role of cytokines and hormones, 110 Geographic and contextual effects behavior-setting theory, 276–279 American Time Use Survey, 278 2003–2006 American Time Use Survey, 277 Cohen-Cole and Fletcher’s discussion, 279 human–environment interaction approach, 277 Study of Adolescent Health Behaviors (ADD Health), 278 time use and transportation research studies, 278 built environment and physical activity, 272–274 environmental measures, 273
Subject Index ISI Web of Knowledge, 274 nationally representative survey (NHANES III) of US adults, 273 problems, 273–274 “environment,” definition, 269 food environments and food deserts, 274–276 behavior-setting theory, 275 dietary guidelines for Americans, 274 quantitative research methods, 275 United Kingdom’s Nutrition Task Force’s Low Income Project, 274 future directions, 286–289 Diabetes Prevention Program (DPP), 288 role of built environment, 288 studies of contextual effects on energy balance, 287 technical problem, 288 growth, factors, 268 major challenges for analysis clusters of walking in LA., 283f confounding, 280–281 energy balance-related measures and cancer, 284–286 self-selection and interpretation of cross-sectional associations, 281–282 spatial scan approach, 282 state-level variation in obesity and cancer incidence, 285f socioecological model, 268–270 linear/nested model, 269f urban sprawl and obesity, 271–272 association between sprawl and behavior, 272 sprawl index, 271 Geographic information systems (GIS), 268, 275, 284, 287 GERD, see Gastroesophageal reflux disease (GERD) Ghrelin, 147–148 28-amino acid peptide, 147 GHSR1a, 147 plasma ghrelin levels, 147 Roux-en-Y gastric bypass surgery, 147 serum levels, 147 Global positioning systems (GPS), 268, 279 Glucose clamp procedure, 248 “Gold standard” technologies, 5, 203, 244 “Gorging” vs. frequent eating (“nibbling or snacking”), 251
305 H Haber–Weiss reaction, 155 Harvard Growth Study, 88, 108 HCC, see Hepatocellular cancer (HCC) HCCR-1 oncogene, 113 Head and neck cancers incidence and risk factors, 30 Health Aging and Body Composition cohort study, 149 Hepatocellular cancer (HCC) incidence and risk factors, 27 HERITAGE family study, 97 Heterocyclic amines (HCAs), 160–161 and body composition/BMI levels, 160 intake level, 160 MeIQx and PhIP, 160 PhIP–DNA adducts, 160–161 HIF, see Hypoxia-inducible factor pathway (HIF) HIF-1, 103, 134f, 152 High-dose estrogen therapy, 116 Homeostasis model assessment ratio (HOMA-R), 145 Human starvation study, 183–184 Hypoproteinemia or renal/cardiac failure, 184 Hypoxia-inducible factor pathway (HIF), 19 HIF-1, 103, 134f, 152 I Immunosuppression, 31 Indian National Family Health Survey, 69 Inflammation, 212 chronic, 94, 114, 148, 151, 227 chronic low-grade, 130, 148 low-grade systemic, 163 systemic, 78, 94, 151, 163 Inflammatory related molecules, 148 Insomnia, 100, 254–255 Insulin, 133–137 α and β peptide hormone, 133–135 C-peptide, 136 energy metabolism pathways fatty acid synthesis, 136 glycogen, 136 glycolysis, 136 hyperinsulinemia, 136 obesity-related carcinogenesis peptide growth factors/ adipokines/nutrients, 134f–135f potential mechanistic role(s), 136–137 resistance, 136 serum insulin levels, 136
306 Insulin-like growth factor (IGF), 96, 109, 137–138 estrogen receptor (ER) signaling pathways, 138 IGFBP isoforms, 137 IGFBP-3, 137, 192, 226 IGF-1/IGF-2 receptors, 137, 226 PI3K/ERK/MAPK, 138 Insulin receptor substrate (IRS) proteins, 134f–135f, 136, 150 Insulin resistance syndrome, see Metabolic syndrome Interleukin-6 (IL-6), 150–151, 212 circulating levels in cancers/obese, 150 fasting plasma IL-6 concentrations, 150 higher plasma levels, 151 26-kDa protein, 150 ‘myokine,’ 150 serum concentrations, 150 stimulate cell growth and inhibit apoptosis, 151 International Agency for Research on Cancer, 204 International Association for the Study of Obesity, 71 International Breast Cancer Study Group, 77 International Life Sciences Institute Focal Point, the Working Group on Obesity in China, 72 International Obesity Task Force, 72 8-iso-prostaglandin F2α (8-iso-PGF2α), 158 in cancer patients, 158 in estradiol-induced mammary tumors, 158 urinary levels, 158 J Janus kinases (JAK), 134f–135f, 150 Japanese breast cancer survivors, 77 K Kidney cancer, 17–18, 51–52, 76, 183 Kleine–Levin syndrome (KLS), 100 Korean study of survival, 77 L Leptin, 141–142 and cancer, association between, 142–143 effects of deficiency, 142 epidemiological studies, 142 16-kDa protein hormone, 142 mechanistic studies, 143 orexigenic pathway neurons, 142 role in breast cancer, 143 Life After Cancer Epidemiology, 224
Subject Index Lifestyle behaviors, intervening to change, 234–235 combination of elements, 235 aspects of built environment, 235 mediators, 235 moderators, 235 “obesogenic” environments, 235 intervention decision node flowchart, 234–234f Lifetime occupational activity, 209–210 Low-grade systemic inflammation, 163 Lung cancer, 208 incidence and risk factors, 29 protective effect of physical activity, 208 Lymphoid and hematological cancers incidence and risk factors, 31 M Magnetic resonance (MRI), 5 Mammalian target of rapamycin (mTOR), 109–110, 134f, 136–137, 157, 164 Mammography, 51 Massachusetts Male Aging Study, 98–99 Mean pulmonary arterial pressure (mPAP), 95–96 Mechanisms linking energy balance to cancer survival, 226–227 C-reactive protein (CRP), 227 decreased serum insulin levels, 226 IGFBP-3, 226 IGF-I, 226 link between adipokines, 227 sex steroid hormones, 226–227 Mechanisms, physical activity and cancer, 211–212 exposure to unopposed estrogen, 211 impact on circulating hormone profiles, 212 inflammation, 212 insulin and IGF pathways, 212 interleukin-6 (IL-6), 212 See also Physical activity and cancer Mediterranean dietary intervention, 253 Melanocortin-4 receptor (MC4R), 90, 92 Metabolic equivalents (METs) of energy expenditure, 203, 236–238 Metabolic syndrome, 158–159 associated diseases, 159 in cancer stems, 159 disturbed sleep, syndrome ‘Z,’ 159 risk factors, 159 syndrome ‘X,’ 158 Monocyte chemotactic protein-1 (MCP-1), 152
Subject Index in carcinogenic process, 152 and chemokine receptor 2 (CCR2), 152 higher serum levels, 152 13-kDa protein, 152 Monounsaturated fatty acid (MUFA), 155–156, 241, 253 Monozygotic (MZ) twins, 100 MRI, see Magnetic resonance (MRI) Multiple myeloma, 6t, 31, 50 Myokine, 150, 163 N Narcolepsy, 100–102 NASH, see Non-alcoholic steatohepatitis (NASH) National Health Interview Survey (NHIS), 50, 72, 84, 284 National Institutes of Health-American Association of Retired Persons (NIH-AARP) Diet and Health Study, 115, 205–206, 208–210, 220 Neuropeptide Y (NPY), 91–92, 142, 147, 164 N –6 fatty acids (or) omega-6, 241 NHANES data, 46–49, 273 NHL, see Non-Hodgkin’s lymphoma (NHL) “Nibbling or snacking,” 251 NIDDM, see Noninsulin- dependent diabetes mellitus (NIDDM) Non-alcoholic steatohepatitis (NASH), 28 Non-Hodgkin’s lymphoma (NHL), 6, 25f, 31, 76, 185t Noninsulin- dependent diabetes mellitus (NIDDM), 248 Non-rapid eye movement (NREM) sleep time, 102 Nucleotide excision repair (NER), 157, 160–162 Nurses Health Study, 50, 204, 221 Nutrition and cancer survival, 223–225 Life After Cancer Epidemiology, 224 scarcity, nutritional, 70, 224 vitamin D and potential association with prognosis, 225 Women’s Healthy Eating and Living Study (WHEL), 224 Women’s Intervention Nutrition Study (WINS), 224 ‘transition,’ 68–71 O Obesity cancer mortality, 130 childhood, 34, 47, 87–88
307 definition, 129 disorders, 130 endogenous hormones, 130 induce chronic low-grade inflammation, 130 Obesity and cancer epidemiology, 1–36 BMI and central adiposity and cancer risk, 6t breast cancer, 6–10 cancer sites with insufficient evidence, 32 challenge and opportunity areas influential mechanisms, 35 obesity phenotypes selection, animal models for, 33 in utero and early childhood exposures, influences, 33–34 weight loss on cancer risk, evidences, 34–35 colorectal cancers, 19–22 endometrial cancer, 12–13 esophageal cancer, 22–24 gallbladder cancer, 28–29 global trends in obesity by sex, 2f head and neck cancers, 30–31 hepatocellular cancer, 27–28 lung cancer, 29–30 lymphoid and hematological cancers, 31–32 methodologic and measurement complexities cross-study comparisons, 5–6 subgroup variation in association, assessment, 5 validity of exposure assessment, 4–5 methods, 6 ovarian cancer, 10–12 pancreatic cancer, 24–27 prostate cancer, 14–16 renal cell carcinoma, 17–19 thyroid cancer, 16–17 Obesity and cancer, epidemiology in racial/ethnic minorities, 45–59 burden of overweight and obesity, 48f adults, 46–47 children and adolescents, 47–49 geographic and urban–rural differences, 50 socioeconomic disparities, 49 cancer incidence and mortality, implications bottom line, 51–52 cancer incidence in United States, 51 determinants of obesity
308 Obesity and cancer (cont.) behavioral determinants, 52 breastfeeding, 54 environmental influences, 54–55 fast-food intake, 53 food environment, 55–57 fruit and vegetable consumption, 52–53 physical activity, 53–54 physical activity environment, 57–58 television time, 54 public health improvement and reducing disparities, 58–59 Obesity and cancer in Asia, 65–79 adiposity and cancer, differences between risk, 72–77 survivorship, 77–78 biological mechanisms, 78–79 BMI–colon cancer association, 78 hyperinsulinemia, 79 sex hormones, alterations in, 78 systemic inflammation, role of, 78 body composition in adults, 71–72 anthropometric measures, 71 BMI–percentage body fat relationship, ethnic differences in, 71 body size and composition, differences in, 71 fat distribution, 71 obesity-related chronic diseases, 72 screening for overweight and obesity, 71 waist circumference, 71 cancer epidemiology, changes in, 65–68 age-specific and age-standardized cancer rates, 66 economic development, 67–68 esophageal adenocarcinoma, 67 incident colon cancer in women, 67f mortality rate, 65 population size and age structure, 66 squamous cell esophageal cancer, 67 under- and over nutrition risk, 67 nutrition transitions and positive energy balance increasing adiposity in selected countries, 70–71 intake and expenditure, changes in, 68–70 Obesity and cancer, mechanisms, 129–164 adipokines and gut hormones adiponectin, 144–145 ghrelin, 147–148 leptin, 141–142
Subject Index resistin, 145–146 visfatin, 146–147 angiogenic factors VEGF, 152–153 cytokines and chemokines CRP, 151 IL-6, 150–151 MCP-1, 152 TNF-α, 148–150 dietary intake of carcinogens and DNA damage diet/exercise/pharmaceuticals, 162–164 HCAs, 160–161 PAHs, 161–162 fatty acid metabolism eicosanoids and prostaglandins, 153–154 insulin and IGF axis IGF, 137–138 insulin, 133–137 metabolic syndrome, 158–159 obesity-related carcinogenesis, 133f putative factors and carcinogenesis, 131t–132t reactive oxygen species and oxidative stress, 155–158 See also Stress (oxidative) and reactive oxygen species sex steroid hormones, 138–141 Obesity and cancer risk, 72–77 abdominal obesity, 75 BMI and colorectal cancer, relationship risk, 75 and WHR associations risk, 73 breast cancer, 75 central adiposity for Asian, 73 colon cancer, 75 colorectal adenoma, 75 endometrial cancer, 73 BMI/WHR, association between, 74f fat–BMI relationships, 73 gallbladder cancer, 76 ovarian cancer, 76 overweight women, 74 pancreatic cancer, 76 postmenopausal breast cancer, 74 squamous cell carcinoma, 76 Obesity at initial diagnosis and cancer survival 220-221 NIH-AARP Diet and Health Study, 220
Subject Index obesity and breast cancer prognosis in younger women, 221 obesity and increased risk, studies on, 220 “Obesogenic,” 34, 54, 235 Obstructive sleep apnea syndrome (OSAS), 102 Occupational activity, 69, 202–203, 205, 209–210 Ovarian cancer, 208–209 early life and premenopausal disease, 11 incidence and risk factors, 10 waist circumference and weight gain, 11 Overeating, 187, 194, 251 Overweight in childhood, 107 definition, 129 Oxidative stress, 19, 35, 108, 130, 154–158, 163, 193, 248 See also Stress (oxidative) and reactive oxygen species Oxidized low density lipoprotein cholesterol (Ox-LDL), 156, 163 8-oxo-deoxyguanosine (8-oxo-dG), 156–157 P Pancreatic cancer, 210–211 central adiposity, role of, 27 incidence and risk factors, 24 RR estimation of increase in BMI, 25f–26f sex differences findings, 24–26 Pap tests, 51 Peroxisome proliferator-activated receptor (PPAR), 95–97, 103, 144–145, 150, 164 Pharmaceuticals, 163–164 anabolic (orexigenic) pathways, drugs, 164 catabolic (anorexigenic) pathways, drugs, 164 everolimus, 164 metformin, ‘insulin sensitizer,’ 164 neoadjuvant chemotherapy, 164 orlistat, 164 PI-103 and NVP-BEZ235, 164 sibutramine, 164 Phosphatidylinositol 3-kinase (PI3K), 17, 92, 109–110, 134f, 135f, 136, 138, 143, 150, 154, 157, 164 Physical activity and cancer breast cancer, studies on, 204–205 case–control studies and cohort studies, 203 colon cancer, 205–206 factors, weaker results, 206
309 NIH-AARP Diet and Health Study, 206 “convincing,” 204 definition, 202 endometrial cancer, 206–207 meta-analyses of the physical activity–endometrial cancer association, 207 lung cancer, 208 protective effect of physical activity, 208 mechanisms, 211–212 exposure to unopposed estrogen, 211 impact on circulating hormone profiles, 212 inflammation, 212 insulin and IGF pathways, 212 interleukin-6 (IL-6), 212 meta-analyses, 203 metabolic equivalents (METs) of energy expenditure, 203 other cancers esophageal cancer, 210 gastric cancer, 209–210 pancreatic cancer, 210–211 renal cell cancer, 210 ovarian cancer, 208–209 prostate cancer, 207–208 American Cancer Society CPS II cohort, 209 physical activity and risk of, 208 survival, 222–223 assessing in year prior to diagnosis, 223 moderate-intensity recreational activity after diagnosis, 223 Physical activity and sedentary behavior domains and cancer risk, 237–239 aerobic physical activity, 239 bicycling, 238 energy expenditure, 239 household physical activity, 238 leisure time or recreational physical activity, 238 MET-hours, moderate/vigorous physical activity, 238–239 occupational physical activity, 238 sedentary behaviors, 239 strength training, 239 walking, 237–238 interventions to change, 247–250 aerobic physical activity, 249 bicycling, 248 leisure time or recreational physical activity, 248–249
310 Physical activity and sedentary behavior (cont.) other forms, 249 sedentary behaviors, 249–250 strength training, 249 walking, 248 levels/measures, 236–237 duration, 237 energy expenditure, 237 frequency, 237 intensity, 236 modes of measurement, 237 Physicians’ Health Study cohort, 153 PI3K/Akt pathway, 109, 136, 154 Polycyclic aromatic hydrocarbons (PAHs), 161–162 Agricultural Health Study cohort, 161 benzo(a)pyrene (b(a)p), human carcinogen, 161 PAH–DNA adducts, 161–162 Polyunsaturated fatty acid (PUFA), 155–156, 241, 253, 256 Pre-B cell colony-enhancing factor (PBEF), 146 Proopiomelancortin (POMC), 91–92, 142 Prostaglandins, 153–154 elevated/mechanistic PGE2 levels, 154 higher PGE2 levels, 154 lipid peroxidation, 153–154 synthesis, 153 use of COX inhibitors, 154 Prostate cancer, 115–116, 207–208 American Cancer Society CPS II cohort, 209 disease stage at diagnosis and detection, 14–15 incidence and risk factors, 14 other factors, 15 physical activity and risk of, 208 serum hormones and growth factors, role of, 15 Prostate-specific antigen (PSA) testing, 14, 207–208, 213 Protein carbonyls, 155 Protein tyrosine phosphatase (PTP)1B, 92, 134f–135f, 150 Proteolysis, 155 PSA, see Prostate-specific antigen (PSA) testing Pulmonary vascular resistance (PVR), 95–96 Q Quantitative trait locus (QTL) mapping, 94 Quercetin, 195
Subject Index R Rapid eye movement (REM) sleep, 100–102 Ratios of waist to hip (WHR), 5, 71, 73–75, 77, 79, 158 Ratios of waist to thigh (WTR), 5 RCC, see Renal cell carcinoma (RCC) Recreational and occupational activity, distinction, 203 Regular exercise, 163 Relative fat pattern index (RFPI), 88–89 Relative risk (RR), 8, 13–14, 20–21, 25–26f, 77, 97, 203–204, 207, 224, 238–239, 270 Renal cell carcinoma (RCC) adiposity, data measures of, 18 biological mechanisms, 19 genetic factors and occupational exposures, 19 heterogeneous study results, 18 incidence and risk factors, 17 Resistin, 145–146 and cancer types, 145 HOMA-R, 145 12.5-kDa polypeptide hormone, 145 mRNA expression, 146 Restless leg syndrome (RLS), 100 Resveratrol antioxidant and antimutagenic agent, 194 effective in treating diseases, 195 in grapes and other plants, 194 protect mammalian cells, 194 Roux-en-Y gastric bypass surgery, 147 RR, see Relative risk (RR) S Saturated fatty acid (SFA), 241 Scarcity, nutritional, 70 Self-digestion process, 157 SES, see Socioeconomic status (SES) Sex hormone-binding globulin (SHBG), 78–79, 111–112, 116, 133f, 137, 139–140, 163 Sex hormones, alterations in estrogen/progesterone/androgens, 78 Sex steroid hormones, 138–141, 226–227 androgens, 139 endogenous, 138 EPIC study, 139 estrogens, 138–139 ‘female sex hormones,’ 139 hormonal cancers, 141 IGF system, 140 ‘male sex hormones’, 139
Subject Index in obesity related carcinogenesis, 140 ovarian cancer, 141 perimenopause, 140 progesterone and estrogens, 139 and prostate cancer, relationship between, 139–140 synthesized from cholesterol, 139 ‘unopposed estrogen,’ 140–141 Sex steroid metabolism, 240 SHBG, see Sex hormone-binding globulin (SHBG) Signal transducers and activators of transcription (STAT) factors, 134f–135f, 143, 150 Single nucleotide polymorphisms (SNPs), 89–90, 96, 103, 111 Sleep and cancer risk, 243–247 circadian disruption, 246–247 domains, levels, and measures of sleep, 244 effects of sleep deprivation, 245 interventions to change, outcomes, 254–255 CBT, components to, 255 insomnia, 254 sleep quality, onset, and efficiency, 255 short sleep, 245 sleep duration, 244–246 sleep quality, 246 Sleep apnea, 99 central, 99 obstructive, 99 syndrome, 100 Sleep disorders circadian sleep disorders, 100 Kleine–Levin syndrome (KLS), 100 narcolepsy, 100 restless leg syndrome (RLS), 100 sleep and cancer risk, 243–247 sleep apnea syndrome, 100 Sleep Heart Health Study, 98–99 Sleep–wake behavior, 101 Socioecological model, 268–270 linear/nested model, 269f Socioeconomic status (SES), 49, 54–58, 66, 70, 191t, 234f, 235, 275–276, 283 Spatial scan approach, 282, 284 Stomach cancer, 50–51, 66, 72–73, 78, 111 Stress (oxidative) and reactive oxygen species biomarkers, 155–158 See also Biomarkers free radicals, 154 futile redox cycling, 155 Haber–Weiss reaction, 155
311 lipophilic exogenous chemicals, 155 ROS production, 154 Swedish Obesity Study, 2 Syndrome Kleine–Levin, 100 metabolic, 99, 101, 133f, 157–159 obstructive sleep apnea, 102 polycystic ovary, 140, 146 restless leg, 100 sleep apnea, 100 Von-Hippel Lindau (VHL), 19 ‘X’, 158 ‘Z,’ 159 T Thyroid cancer growth hormone and IGF axis, potential role of, 17 incidence and risk factors, 16–17 Tumor necrosis factor alpha (TNF-α), 148–150 212-amino acid type II transmembrane protein, 148 and cancer outcomes, 149 expression in adipocytes, 149 promoting insulin resistance, 149–150 role in human adipocytes, 149 TNF-R1/TNF-R2, receptors, 149 Type 1a growth hormone secretagogue receptor (GHSR1a), 147 Type 1 diabetes, 13 U Uncoupling proteins (UCPs), 103–104 United Kingdom’s Nutrition Task Force’s Low Income Project, 274 Urban sprawl and obesity, 271–272 association between sprawl and behavior, 272 sprawl index, 271 See also Geographic and contextual effects USDA, see US Department of Agriculture (USDA) studies US Department of Agriculture (USDA) studies, 56 V Vascular endothelial growth factor (VEGF), 152–153 angiogenesis, 152 circulating levels and serum levels, 152–153 clinical studies, 153 low oxygen or ‘hypoxia,’ 152
312 Vascular endothelial growth (cont.) in obesity-associated colon cancer, 153 VEGF-A, 152 VHL, see Von-Hippel Lindau (VHL) syndrome Visfatin, 146–147 50 kDa and 100 kDa, 146 PBEF, 146 and PBEF, 146 serum, 146 Vitamin D deficiency, 225 and potential association with prognosis, 225 Von-Hippel Lindau (VHL) syndrome, 19 W WCRF, see World Cancer Research Fund (WCRF) Weight gain and cancer survival, 221–222 Cancer and Leukemia Group B (CALGB) 89803, 221 genetic differences in tumors among obese patients, 221 healthy diet and increasing physical activity, 222 Nurses’ Health Study, 221 WHEL, see Women’s Healthy Eating and Living Study (WHEL)
Subject Index WHO, see World Health Organization (WHO) WHO Western Pacific Regional Office (WPRO), 71 WHR, see Ratios of waist to hip (WHR) WINS, see Women’s Intervention Nutrition Study (WINS) Wisconsin Sleep Cohort Study, 104 Women’s Health Initiative intervention trial, 21 observational study, 204 Women’s Healthy Eating and Living Study (WHEL), 224–225 Women’s Intervention Nutrition Study (WINS), 50, 224, 252 World Cancer Research Fund (WCRF), 2, 14, 20, 28, 242 World Health Organization (WHO), 2, 5, 71–72, 91, 107 WPRO, see WHO Western Pacific Regional Office (WPRO) WTR, see Ratios of waist to thigh (WTR) Y Youth Fitness, national campaign, 202 Youth risk behavior survey, 93 ‘Yo-yo’ dieters, 160