An Integrative Approach for Deciphering the Causal Associations of Physical Activity and Cancer Risk: The Role of Adiposity

An Integrative Approach for Deciphering the Causal Associations of Physical Activity and Cancer... Abstract Higher physical activity levels have been associated with about 10% to 25% reductions in up to 13 cancers. In isolation, these results are suggestive but not compelling enough to conclude causal associations, except for colon and breast cancer. However, knowledge on the relationships between obesity and cancer, between physical activity and overall and visceral adiposity, and between physical activity and obesity-related mediators of cancer risk can inform the epidemiology of physical activity and cancer. Excluding primarily smoking-related malignancies, for which residual confounding by smoking may occur, all 13 cancers associated with lower physical activity are also obesity-related. Moreover, the magnitude of the inverse association between physical activity and cancer type correlates highly with the association with body mass index (BMI) and cancer type (Spearman r = .79, two-sided P = .004). Physical activity lowers essentially all the obesity-related mediators of cancer, probably mediated largely through reductions in visceral adiposity. These findings strongly suggest that physical activity and adiposity are largely operating through similar carcinogenic mechanisms. That BMI has more robust associations than physical activity with cancer largely reflects that most populations studied have had great variation in BMI and little in physical activity. In populations with higher levels of physical activity and a lower range of BMI, physical activity may appear relatively more important. It may be useful to emphasize to clinicians and the public that physical activity, by acting on the same mechanisms, is likely to reduce risk of obesity-related cancers, even if the impact on lowering BMI is minimal. Determining the causal role of physical activity, a modifiable lifestyle factor, on cancer risk is of high interest. Typically, randomized controlled trials (RCTs) are considered to provide the strongest level of evidence, but they are not feasibly conducted for physical activity and cancer incidence. Hence, the examination of independent causal effects of physical activity has focused on strongly designed cohort studies, which can control for confounding factors. These have typically been conducted in highly economically developed countries (eg, in North America, Europe), and leisure time physical activity has been the predominant form of exercise studied. To date, lower levels of physical activity have generally been acknowledged to be associated with only cancers of the colon, breast, and possibly endometrium (1–3). The main limitations of these studies have been moderate associations (eg, 10%–20% reductions in risk) and the potential for residual confounding because leisure time physical activities tend to be correlated with healthy behaviors. In addition, studies have generally adjusted for body mass index (BMI), but whether effects of physical activity are truly independent from adiposity is unclear. The focus on leisure time physical activity may be practical because leisure activity is most feasibly modified in current populations. Nonetheless, in evaluating the evidence for the role of physical activity on cancer risk, a broader-based synthesis might be elucidative. First, a careful consideration of the relationship between physical activity and adiposity across the life course is important. In particular, how the types and ranges of activities in populations relate to adiposity level in the population is important to consider. Second, this knowledge between physical activity and adiposity may help explain epidemiologic patterns for both obesity and physical activity and cancer, including differences in cancer rates across populations and over time, as well as between individuals within populations. Third, the inter-relations between physical activity and body adiposity can be integrated into studies focused on mechanisms. Overall, this synthesis will evaluate the evidence on whether physical activity should be considered a factor for obesity-related cancers, acting on cancer through largely the same mechanisms operative for obesity and cancer. The Relationship Between Physical Activity and Adiposity Level Although it seems intuitive that physical activity influences adiposity, the precise role of physical activity on body fatness is not straightforward. We can consider two broad aspects of physical activity and adiposity, the relationship in terms of reduction of body weight among the overweight and obese vs the prevention of weight gain over the life course. Much research has focused on physical activity in terms of weight loss. In general, a reduction in energy intake is likely more important than physical activity for weight loss (4). Yet, the cumulative effect of physical activity for preventing weight gain over the life course, as well as in preventing regain once weight is lost, may be important. This section will consider 1) the relationship between physical activity and adiposity from a population-level perspective, particularly time trends; 2) this relationship studied between individuals within populations; 3) effects of physical activity on adiposity compartments, particularly on visceral adiposity, which may operate on a shorter term, and which are not reflected, or only poorly reflected, by BMI measures. Life Course Aspects: Across Populations Ng and Popkin used an array of longitudinal and cross-sectional data sets to assess energy expenditure in four domains of physical activity (occupation, domestic production, travel, and leisure) for adults from United States, the United Kingdom, Brazil, and China, going as far back as 1961 (5). Though the timing differed across the countries, there was a marked decrease of physical activity, primarily due to occupational activity, as well as transportation and household chores, and an increase in sedentary behaviors, including TV watching, computing, and Internet use, among others. The timing is related to changes in the economy and occupations (earlier in the United States and United Kingdom, later in China and Brazil). The decline in occupational physical activity and increase in sedentary time have corresponded in time to increasing rates of obesity across populations worldwide (6). Populations that maintain high rates of occupational and transport-related physical activity with minimal use of technologies that facilitate sedentary lifestyles have a relatively uniformly low body mass. For example, in groups that have refrained from adopting modern technology (eg, Amish), both adults and youth have much higher levels of physical activity and lower BMIs compared with their contemporaries (7,8). Changes in the availability and quality of food have undoubtedly contributed to the rising rates of obesity as well (9). It is beyond the scope of this review to quantify the relative contributions of changes in energy expenditure and intake to the rise in obesity, but it appears undeniable that decreases in physical activity are important contributors. Life Course Aspects: Among Individuals Within Populations Quantifying the influence of physical activity over the life course on adiposity at the individual level in a modern population is difficult. Assuming that occupational activity will not rise, leisure time physical activity is most relevant. Short-term intervention studies suggest that physical activity could help in preventing population weight gain. For example, a physical activity intervention over six to 12 months may result in less weight gain (eg, 1 kg) in the intervention group; assuming such an intervention is carried forward and adhered to for multiple decades, an important influence on cumulative body weight would likely emerge. Overall, the evidence strongly supports that physical activity can be an important factor in reducing weight gain over the life course, but how much so is difficult to estimate. Although the longer-term effects are not feasibly studied with interventions, observational studies support that some level of physical activity is useful for preventing or minimizing weight gain. For example, based on long-term observational studies, the American College of Sports Medicine concluded in 2009 that 150–250 minutes per week of moderate-intensity physical activity are effective in preventing weight gain (10). Sedentary lifestyle and daily life activities (eg, ≤1.5 metabolic equivalents energy expenditure) are also likely to influence cumulative weight gain (11), but further study for quantification of effects in populations is required. Shorter-Term Aspects of Physical Activity on Visceral Adiposity A consideration of measures of the specific adiposity used in epidemiologic studies of physical activity to determine independent effect of adiposity is important. BMI, which is typically used, is a reasonably good surrogate of overall adiposity, but it may not capture all of the adiposity-related effects of physical activity. Visceral fat is only a few percent of total bodyweight. A loss of half a kilogram of visceral fat would have major health benefits, but this would hardly impact the total BMI. Intervention studies of supervised aerobic exercise ranging from weeks to one year show sizable absolute reduction in visceral adiposity, with no or minimal loss in overall body weight (12–15). A meta-analysis showed that even in the absence of weight loss, exercise was associated with a 6.1% decrease in visceral adiposity (16). Further, weight loss by physical activity induced a larger loss in visceral fat than a comparable weight loss by diet (16). Some evidence also suggests that physical inactivity (eg, bed rest) preferentially leads to an increase in visceral adiposity (17). These changes in visceral adiposity may largely underlie effects of physical activity on obesity-related biomarkers. Physical activity normalizes metabolic biomarkers (eg, insulin, inflammatory markers, glucose, adiponectin, and insulin resistance) in proportion to reductions in visceral adiposity independent of weight loss or BMI) (18–23). As a typical example, a 12-week physical activity intervention in obese elderly had a minimal impact on BMI (reduced from 33.3 to 32.1 kg/m2) but reduced visceral adipose tissue by 22.4% and insulin resistance by 31.2%, and the change in visceral adipose reduction correlated highly with the change in insulin resistance (r = .66) (23). Thus, physical activity has an important influence on adiposity-related biomarkers not captured by BMI measures, even in obese individuals. How sedentary lifestyle and daily life activities influence visceral adiposity is less understood. Physical Activity, Excess Adiposity, and Cancer: An Epidemiologic Perspective The brief description of the relationship between physical activity and adiposity above help explain the epidemiology of physical activity and obesity and cancer across the spectrum of human populations. High levels of physical activity in pre-industrial populations help keep population-wide BMI levels in the low range. When the average level of physical activity in a population becomes very low, various contributors to the variation in BMI will then be unleashed; these include genetics, epigenetics, dietary factors, self-imposed behavioral constraints on overeating, smoking, and leisure time and transportation physical activities. As physical activity has generally diminished in many populations, it has become a less important factor in accounting for variation in BMI within a population. How the economic transition in countries generates comparisons for physical activity and cancer from three epidemiologic perspectives is illustrated in Figure 1: A) cancer rates in highly active populations vs those in sedentary populations (“ecologic” comparisons); B) physically active vs less active individuals within sedentary populations; C) physically active vs less active individuals within highly active populations. Figure 1. View largeDownload slide Conceptual model of transitioning from a highly physically active traditional population to a sedentary modern population and the relationship to population range in body mass index (BMI, kg/m2). A) Ecologic comparison; high physical activity level–low adiposity are strongly intercorrelated, so it is difficult to separate their independent effects. B) Within a highly active population, body mass index (BMI) range is limited and physical activity may appear as a dominant factor. C) With low mandatory activities in modern sedentary population, a wide range of BMIs emerge, for which leisure time physical activity is a moderate determinant among many (eg, diet, genetics); BMI may appear as a dominant factor. BMI = body mass index. Figure 1. View largeDownload slide Conceptual model of transitioning from a highly physically active traditional population to a sedentary modern population and the relationship to population range in body mass index (BMI, kg/m2). A) Ecologic comparison; high physical activity level–low adiposity are strongly intercorrelated, so it is difficult to separate their independent effects. B) Within a highly active population, body mass index (BMI) range is limited and physical activity may appear as a dominant factor. C) With low mandatory activities in modern sedentary population, a wide range of BMIs emerge, for which leisure time physical activity is a moderate determinant among many (eg, diet, genetics); BMI may appear as a dominant factor. BMI = body mass index. Cancer Rates in Highly Active Populations vs Those in Sedentary Active Populations Based primarily on associations in cohort study populations, obesity is generally considered to be related to increased risk of at least 13 cancers (24). Many of the 13 cancers now associated with obesity have been relatively rare or much less prevalent in pre-industrial populations and are generally more prevalent in economically developed countries. Exceptions are liver and stomach cancer, which, although related to obesity, have additional strong infectious causes. Differences in physical activity and correspondingly in adiposity may contribute to if not largely account for variability in obesity-related cancer rates between populations. Although ecologic studies are less able to account for confounding factors and are downplayed, they represent the most extreme contrast of “highly physically active/low adiposity” vs “sedentary/high adiposity.” In this ecologic type of comparison, high physical activity level and low adiposity are strongly intercorrelated, so it is difficult, or perhaps fruitless, to attempt to separate their independent causal effects. For example, in the extreme theoretical scenario that everyone is highly physically active and hence everyone has a low BMI, there would be no cancers attributable to obesity. In the opposite extreme of a completely sedentary population, the obesity rate would be high, and its variation would be completely due to factors other than physical activity. Thus, compared with the highly physically active population, all of the apparent “obesity-related” cancers are due to lack of physical activity, but from the perspective of those in the nonactive population, physical activity appears to be an irrelevant factor. Physically Active vs Less Active Individuals Within Sedentary Populations In sedentary populations, physical activity, mostly leisure time activity, will be only a moderate determinant of BMI. Other determinants, including diet, eating habits, and genetics will contribute to wide variability of BMI (from 22 to 35 kg/m2, for example). In this context, BMI is a relatively robust measure of adiposity, and substantial variation exists between individuals. Additionally, BMI tracks well over time in an individual, so even a single measure provides a robust assessment of long-term exposure, though repeated measures may strengthen estimates (25). Indeed, all these features of even a relatively simple measure such as BMI have helped establish obesity as a causal risk factor for at least 13 cancers. Independently of BMI, abdominal adiposity is associated with an increased risk of cancer (24,26–28). The argument that these associations with obesity are causal is supported by robust dose-response associations in cohort studies, numerous plausible mechanisms, and recent Mendelian randomization studies of adiposity-related traits and cancer risk (27,29–34). For physical activity and cancer, the evidence for causality has been less robust than for obesity, with the exceptions of cancers of the colon, breast, and possibly endometrium (1–3). Interestingly, some evidence suggests that sedentary behaviors may be associated with higher risk of colon and endometrial cancer (35). The above synthesis of the inter-relation of physical activity and adiposity raises the question of should physical activity also be associated with obesity-related cancers. To shed light on this question, Table 1 shows a comparison of results from a recent analysis that pooled data from 12 prospective US and European cohorts with self-reported leisure time physical activity. This is the largest relevant study to date, with 1.44 million participants, providing sufficient statistical power to examine multiple cancer types (36). The authors reported statistically significant or suggestive (P ≤ .1) inverse associations with 17 cancers (Table 1). A striking pattern is apparent. Among the cancers associated with lower physical activity levels, 13 are also established obesity-related cancers (details in table 1). Moreover, an increase in cancer risk associated with BMI (from meta-analyses of cohort studies) correlates strongly with the reduction in cancer risk associated with physical activity (Pearson r = .75, two-sided P = .008; Spearman r = .79, two-sided P = .004) (see Figure 2 for details). This high concordance between obesity-related and physical activity–related cancers strongly suggests that physical activity and adiposity are largely operating through similar carcinogenic mechanisms. If physical activity and adiposity were acting on cancer through entirely different mechanisms, we would not expect this striking pattern. Table 1. Cancers inversely associated with leisure time physical activity from Moore et al. (36) and whether they are associated with obesity and tobacco Cancer type Obesity-related Tobacco-related* Esophageal adenocarcinoma† Yes No Gallbladder† Yes No Liver† Yes No Lung No Yes Kidney† Yes No Small intestine (47–49) Yes No Gastric cardia† Yes No Endometrial† Yes No Esophageal squamous No Yes Myeloid leukemia (50) Yes No Myeloma (51) Yes No Colon† Yes No Head and neck No Yes Rectum† Yes No Bladder No Yes Breast† Yes No Non-Hodgkin lymphoma (52) Yes No Cancer type Obesity-related Tobacco-related* Esophageal adenocarcinoma† Yes No Gallbladder† Yes No Liver† Yes No Lung No Yes Kidney† Yes No Small intestine (47–49) Yes No Gastric cardia† Yes No Endometrial† Yes No Esophageal squamous No Yes Myeloid leukemia (50) Yes No Myeloma (51) Yes No Colon† Yes No Head and neck No Yes Rectum† Yes No Bladder No Yes Breast† Yes No Non-Hodgkin lymphoma (52) Yes No * Those cancers for which tobacco accounts for at least 40% of the total (53; modifed from [54–56]). † Acknowledged as an established association with obesity by the World Cancer Research Fund/American Institute of Cancer Research. Table 1. Cancers inversely associated with leisure time physical activity from Moore et al. (36) and whether they are associated with obesity and tobacco Cancer type Obesity-related Tobacco-related* Esophageal adenocarcinoma† Yes No Gallbladder† Yes No Liver† Yes No Lung No Yes Kidney† Yes No Small intestine (47–49) Yes No Gastric cardia† Yes No Endometrial† Yes No Esophageal squamous No Yes Myeloid leukemia (50) Yes No Myeloma (51) Yes No Colon† Yes No Head and neck No Yes Rectum† Yes No Bladder No Yes Breast† Yes No Non-Hodgkin lymphoma (52) Yes No Cancer type Obesity-related Tobacco-related* Esophageal adenocarcinoma† Yes No Gallbladder† Yes No Liver† Yes No Lung No Yes Kidney† Yes No Small intestine (47–49) Yes No Gastric cardia† Yes No Endometrial† Yes No Esophageal squamous No Yes Myeloid leukemia (50) Yes No Myeloma (51) Yes No Colon† Yes No Head and neck No Yes Rectum† Yes No Bladder No Yes Breast† Yes No Non-Hodgkin lymphoma (52) Yes No * Those cancers for which tobacco accounts for at least 40% of the total (53; modifed from [54–56]). † Acknowledged as an established association with obesity by the World Cancer Research Fund/American Institute of Cancer Research. Figure 2. View largeDownload slide The correlation between percent (%) risk reduction due to physical activity from Moore et al. (36) and the % risk increase due to body mass index (BMI) in 11 cancer sites from studies with available data of BMI from meta-analyses of prospective cohort studies (24). All meta-analyses are from the World Cancer Research Fund/American Institute of Cancer Research, except for lymphoma (52), multiple myeloma (51), and myeloid leukemia (50). Figure 2. View largeDownload slide The correlation between percent (%) risk reduction due to physical activity from Moore et al. (36) and the % risk increase due to body mass index (BMI) in 11 cancer sites from studies with available data of BMI from meta-analyses of prospective cohort studies (24). All meta-analyses are from the World Cancer Research Fund/American Institute of Cancer Research, except for lymphoma (52), multiple myeloma (51), and myeloid leukemia (50). Of note, the remaining four cancers associated with lower physical activity (lung, esophagus [squamous], bladder, and head and neck) but not with obesity (36) have smoking as an important primary cause (table 1). For these malignancies, residual confounding from smoking is possible, as those who suffer the greatest toxic effects of smoking (especially respiratory) may be less able to exercise. In fact, these four cancers were not associated with physical activity among never-smokers, an analysis less prone to smoking-related confounding. It is also plausible that physical activity could offset some specific carcinogenic effects related primarily to smoking, such as reduced immunity. Physically Active vs Less Physically Active Individuals Within Highly Active Populations Minimal data exist on physical activity and cancer risk in populations with high levels of physical activity. Interestingly, a case–control study of colon cancer conducted in Shanghai, China, between 1990 and 1993 may represent a more traditional population transitioning to a sedentary modern one (37). Around this time, colon cancer rates, historically extremely low, were increasing in China. The amount of physical activity was much higher than in modern populations, and average BMI was low, with very few individuals in the overweight/obese range. Among those in the high normal range in BMI, a sevenfold risk gradient was observed for low versus high physical activity. This finding suggests when the level of physical activity is high, it may be a powerful determinant of colon cancer risk, even if few are in the overweight/obese range (BMI > 25 kg/m2) in that population. Of note, in Asian populations with low BMI, visceral adiposity is independently associated with higher risk of colorectal neoplasia (26). Thus, the lower risk of colon cancer in the Shanghai case–control study in highly physically active participants could possibly reflect low visceral fat. Physical Activity, Excess Adiposity, and Cancer: A Mechanistic Perspective The mechanisms underlying the association between adiposity and cancer risk are not fully established, but the main mechanisms generally discussed include sex steroids (eg, breast, endometrial cancers), metabolic hormones (eg, gastrointestinal, reproductive cancers) and inflammation (eg, immune-related and other cancers) (24). Insulin and insulin-like growth factor 1 (IGF1) are the major hormones that activate the PI3K-Akt and ras-MAPK signaling pathways, which stimulate mitosis and inhibit apoptosis in sensitive tissues (38). Estrogen is an established causal factor for breast and endometrial cancer (39), and insulin (or C-peptide) levels are robustly associated with colorectal, pancreatic, and endometrial cancer, and IGF1 with colorectal and prostate cancer. Additional obesity-related hormones, including leptin, adiponectin, IGF-binding proteins, and sex hormone binding globulin may also be involved. Higher levels of many of these hormones have been associated with a higher risk of specific cancers in prospective studies (24). Not surprisingly, obesity-related biomarkers (eg, estrogen, metabolic hormones) are lower in countries with high physical activity and low BMI (40). These hormones, as well as likely unidentified ones, are biologically inter-related. These biomarkers can be utilized to examine the roles of adiposity and physical activity cancer risk from a mechanistic basis. From a practical perspective, these biomarkers do not necessarily have to be the precise isolated causal factors, but rather, they act as surrogates of causal mediators. In general, BMI has a stronger association with cancer-related biomarkers than physical activity, at least in cross-sectional studies conducted in sedentary, high-BMI populations. In some studies, including intervention studies, the effect of physical activity on biomarkers including insulin, estrogens, and inflammatory markers appeared to be partially or even largely due to effects on adiposity, especially visceral adiposity (41–45). It is apparent that physical activity affects essentially all of the putative obesity-related biomarkers for cancer, but with weaker associations than BMI has with these biomarkers, a similar pattern as with physical activity obesity with cancer risk. Moreover, adjusting for BMI generally attenuated the associations between physical activity and cancer, though inverse associations remained statistically significant for most of the cancers (36). Similarly, the association between physical activity and biomarkers generally weakens when adjusted for BMI, but typically remains. A potential reason, as described above, is that physical activity is related to visceral adiposity, which influences biomarkers, independently of BMI. Summary and Public Health and Research Implications The synthesis integrated herein was based on our understanding of the close relationship between physical activity and adiposity, their effects on putative biochemical mediators, and the range of physical activity and BMI across the human experience and may help explain the following observations: 1) in comparisons across countries, the traditional populations with high physical activity (especially occupational) and low BMI generally have lower risks of the 13 “obesity-related” cancers than modern sedentary populations; 2) these 13 cancers have been associated with obesity (primarily BMI) at the individual level within populations, and more weakly with physical activity; 3) the increase in cancer risk associated with BMI correlates strongly with the reduction in cancer risk associated with physical activity; 4) in modern sedentary populations, BMI, which has a wide range and is measured relatively well, is more strongly associated with cancer risk than is leisure time physical activity; 5) though data are limited, in traditional populations, the association between the physical activity and cancer may be stronger than that for BMI and cancer, because physical activity is high whereas BMI is low and has a narrow range; 6) the relationships between BMI and physical activity and putative cancer-related biomarkers are similar to those of the relationship between BMI and physical activity and cancer risk; and 7) physical activity reduces visceral fat independently of effects on weight loss or BMI, and visceral fat predicts cancer-related biomarkers and cancer risk independently of BMI. If these conclusions are valid, physical activity may help prevent many more cancers than is currently appreciated. Only two or three cancers are considered associated with physical activity, but there may be up to 13, with about 10% to 25% reductions in relative risk for the most active compared with the least active individuals. Although the magnitudes in reduction appear relatively modest, several factors are important to note. First, this number may be an underestimate due to measurement error, which attenuates associations, and to adjustment for BMI; such adjustment may be considered “overcontrol” if the effects or physical activity are largely adiposity-related. Second, the level of physical activity in most populations is very low, at least from the perspective of human history before 50 years ago. Third, most studies are based on one measurement and may not incorporate the potential benefits of physical activity in minimizing weight gain over the life course. It may be useful to emphasize to clinicians and the public that physical activity, by acting on the same obesity-related mechanisms, is likely to reduce risk of obesity-related cancers even if its impact on lowering BMI is minimal. Given the difficulty in achieving lasting weight loss, this advice may be important. From a research perspective, rather than considering physical activity and adiposity as two separate concepts, a more integrated approach may be useful. For example, short-term physical activity interventions may inform BMI-independent effects but may not necessarily be relevant for considering life-long cumulative effects of physical activity on weight maintenance. Potential adiposity-independent mechanisms of physical activity should be studied, but concluding that an effect is truly adiposity independent may require more sensitive measures of adiposity, such as visceral adiposity tissue. The influence of different types (eg, intensities) of physical activities on levels obesity-related biomarkers (eg, sex hormones, metabolic hormones, inflammation) may be informative for cancer risk; these studies need to be complemented with a better understanding of how these factors influence cancer risk. A better understanding of how different components of adiposity (eg, subcutaneous, ectopic, visceral) influence cancer risk will likely ultimately aid in understanding the role of physical activity and cancer (46). Studies in populations with higher levels of physical activity should be conducted when feasible. The issue of measurement error in assessing physical activity is critically important. Most of the evidence for an association of physical activity 1) on measures of adiposity, 2) in relation to biomarkers (from both observational and interventional studies), and 3) on cancer risk have focused on moderate and vigorous physical activities, mostly leisure time. In part, this may reflect that these activities are generally better and have been more often assessed by conventional methods, and although they incur measurement error, the measurement may be adequate to detect these associations. Although questionnaires may do a reasonable job of assessing planned moderate and vigorous physical activities, they may not do as well in measuring sedentary behaviors, daily life activities, and long bouts of physical inactivity. Although not as well studied as moderate/vigorous activities, some evidence suggests that sedentary behaviors may be associated with higher risk of some cancers associated with obesity and physical inactivity (colon, endometrial cancer) but not with other associated cancers (35). More study is required to determine how measurement error in the current measures affects these findings. Of note, sedentary time has increased across populations over time, corresponding to the decrease in occupational physical activity, but whether physiologically important variation in sedentary time can be captured well enough to robustly predict biomarkers, adiposity, and cancer risk remains to be established. More precise measures of these may be helpful. Although further research is required to strengthen evidence for or clarify some specific issues, the overall coherence of these observations strongly suggests that physical activity and adiposity largely operate through similar carcinogenic mechanisms. Notes Affiliations of author: Department of Nutrition and Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA; Channing Division of Network Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, MA. The author has no disclosures to report. References 1 Rezende LFM , Sá TH , Markozannes G , et al. . Physical activity and cancer: An umbrella review of the literature including 22 major anatomical sites and 770 000 cancer cases . Br J Sports Med. Nov 16, 2017: bjsports-2017-098391. doi: 10.1136/bjsports-2017-098391. [Epub ahead of print]. 2 International Agency for Research on Cancer (IARC) . Weight control and physical activity. In: Vainio H, Bianchini F, eds. IARC Handbook of Cancer Prevention . Lyon : IARC Press ; 2002 :83–167. 3 World Cancer Research Fund/American Institute for Cancer Research CUP . Food, nutrition, physical activity and the prevention of colorectal cancer. 2017 . https://www.wcrf.org/int/continuous-update-project/continuous-update-project-findings-reports. Accessed March 27, 2018. 4 Greenway F. Physiological adaptations to weight loss and factors favouring weight regain . Int J Obes (Lond) . 2015 ; 39 8 : 1188 – 1196 . Google Scholar Crossref Search ADS PubMed 5 Ng SW , Popkin BM. Time use and physical activity: A shift away from movement across the globe . Obes Rev. 2012 ; 13 8 : 659 – 680 . Google Scholar Crossref Search ADS PubMed 6 NCD Risk Factor Collaboration (NCD-RisC) . Trends in adult body-mass index in 200 countries from 1975 to 2014: A pooled analysis of 1698 population-based measurement studies with 19-2 million participants . Lancet . 2016 ; 387 10026 : 1377 – 1396 . Crossref Search ADS PubMed 7 Bassett DR , Schneider PL , Huntington GE. Physical activity in an old order Amish community . Med Sci Sports Exerc. 2004 ; 36 1 : 79 – 85 . Google Scholar Crossref Search ADS PubMed 8 Bassett DR Jr , Tremblay MS , Esliger DW , et al. . Physical activity and body mass index of children in an old order Amish community . Med Sci Sports Exerc. 2007 ; 39 3 : 410 – 415 . Google Scholar Crossref Search ADS PubMed 9 Brownson RC , Boehmer TK , Luke DA. Declining rates of physical activity in the United States: What are the contributors? Annu Rev Public Health. 2005 ; 26 : 421 – 443 . Google Scholar Crossref Search ADS PubMed 10 Donnelly JE , Blair SN , Jakicic JM , et al. . American College of Sports Medicine position stand. Appropriate physical activity intervention strategies for weight loss and prevention of weight regain for adults . Med Sci Sports Exerc. 2009 ; 41 2 : 459 – 471 . Google Scholar Crossref Search ADS PubMed 11 Qi Q , Li Y , Chomistek AK , et al. . Television watching, leisure time physical activity, and the genetic predisposition in relation to body mass index in women and men . Circulation. 2012 ; 126 15 : 1821 – 1827 . Google Scholar Crossref Search ADS PubMed 12 Lee S , Kuk J , Davidson L , et al. . Exercise without weight loss is an effective strategy for obesity reduction in obese individuals with and without Type 2 diabetes . J Appl Physiol (1985) . 2005 ; 99 3 : 1220 – 1225 . Google Scholar Crossref Search ADS PubMed 13 Johnson N , Sachinwalla T , Walton D , et al. . Aerobic exercise training reduces hepatic and viseral lipids in obese individuals without weight loss . Hepatology. 2009 ; 50 4 : 1105 – 1112 . Google Scholar Crossref Search ADS PubMed 14 Gepner Y , Shelef I , Schwarzfuchs D , et al. . Effect of distinct lifestyle interventions on mobilization of fat storage pools: The CENTRAL Magnetic Resonance Imaging randomized controlled trial . Circulation . 2018 ; 137 11 : 1143 – 1157 . Google Scholar Crossref Search ADS PubMed 15 Friedenreich C , Woolcott C , McTiernan A , et al. . Adiposity changes after a 1-year aerobic exercise intervention among postmenopausal women: A randomized controlled trial . Int J Obes (Lond) . 2011 ; 35 3 : 427 – 435 . Google Scholar Crossref Search ADS PubMed 16 Verheggen RJ , Maessen MF , Green DJ , et al. . A systematic review and meta-analysis on the effects of exercise training versus hypocaloric diet: Distinct effects on body weight and visceral adipose tissue . Obes Rev. 2016 ; 17 8 : 664 – 690 . Google Scholar Crossref Search ADS PubMed 17 Belavy D , Mohlig M , Pfeiffer A , et al. . Preferential deposition of visceral adipose tissue occurs due to physical inactivity . Int J Obes (Lond) . 2014 ; 38 11 : 1478 – 1480 . Google Scholar Crossref Search ADS PubMed 18 Lesser IA , Guenette JA , Hoogbruin A , et al. . Association between exercise-induced change in body composition and change in cardiometabolic risk factors in postmenopausal South Asian women . Appl Physiol Nutr Metab. 2016 ; 41 9 : 931 – 937 . Google Scholar Crossref Search ADS PubMed 19 Friedman WJ , Dreyfus CF , McEwen B , et al. . Presynaptic transmitters and depolarizing influences regulate development of the substantia nigra in culture . J Neurosci. 1988 ; 8 10 : 3616 – 3623 . Google Scholar Crossref Search ADS PubMed 20 Lee S , Deldin AR , White D , et al. . Aerobic exercise but not resistance exercise reduces intrahepatic lipid content and visceral fat and improves insulin sensitivity in obese adolescent girls: A randomized controlled trial . Am J Physiol Endocrinol Metab. 2013 ; 305 10 : E1222 – E1229 . Google Scholar Crossref Search ADS PubMed 21 Borel AL , Nazare JA , Smith J , et al. . Visceral and not subcutaneous abdominal adiposity reduction drives the benefits of a 1-year lifestyle modification program . Obesity (Silver Spring). 2012 ; 20 6 : 1223 – 1233 . Google Scholar Crossref Search ADS PubMed 22 Kim K , Valentine RJ , Shin Y , et al. . Associations of visceral adiposity and exercise participation with C-reactive protein, insulin resistance, and endothelial dysfunction in Korean healthy adults . Metabolism. 2008 ; 57 9 : 1181 – 1189 . Google Scholar Crossref Search ADS PubMed 23 O'Leary VB , Marchetti CM , Krishnan RK , et al. . Exercise-induced reversal of insulin resistance in obese elderly is associated with reduced visceral fat . J Appl Physiol (1985) . 2006 ; 100 5 : 1584 – 1589 . Google Scholar Crossref Search ADS PubMed 24 Keum N , Song M , Giovannucci E , et al. . Obesity and body composition. In: Thun M , Linet M , Cerhan J , et al. ., eds. Schottenfeld and Fraumeni Cancer Epidemiology and Prevention . 4th ed . New York : Oxford University Press ; 2018 . 25 Thygesen LC , Gronbaek M , Johansen C , et al. . Prospective weight change and colon cancer risk in male US health professionals . Int J Cancer. 2008 ; 123 5 : 1160 – 1165 . Google Scholar Crossref Search ADS PubMed 26 Keum N , Lee D , Kim R , et al. . Visceral adiposity and colorectal adenomas: Dose-response meta-analysis of observational studies . Ann Oncol. 2015 ; 26 6 : 1101 – 1109 . Google Scholar Crossref Search ADS PubMed 27 Jarvis D , Mitchell J , Law P , et al. . Mendelian randomisation analysis strongly implicates adiposity with risk of developing colorectal cancer . Br J Cancer. 2016 ; 115 2 : 266 – 272 . Google Scholar Crossref Search ADS PubMed 28 Zhang C , Rexrode K , Dam Rv , et al. . Abdominal obesity and the risk of all-cause, cardiovascular, and cancer mortality: Sixteen years of follow-up in US women . Circulation. 2008 ; 117 13 : 1658 – 1667 . Google Scholar Crossref Search ADS PubMed 29 Tsilidis KK , Dai JY , Peters U. Editorial: Mendelian randomization analysis identifies body mass index and fasting insulin as potential causal riak factors for pancreatic cancer risk . J Natl Cancer Inst. 2017 ; 109 9 :djx043. 30 Mao Y , Yan C , Lu Q , et al. . Genetically predicted high body mass index is associated with increased gastric cancer risk . Eur J Hum Genet. 2017 ; 25 9 : 1061 – 1066 . Google Scholar Crossref Search ADS PubMed 31 Painter J , O'Mara T , Marquart L , et al. . Genetic risk score mendelian randomization shows that obesity measured as body mass index, but not waist:hip ratio, is causal for endometrial cancer . Cancer Epidemiol Biomarkers Prev. 2016 ; 25 11 : 1503 – 1510 . Google Scholar Crossref Search ADS PubMed 32 Gao C , Patel C , Michailidou K , et al. . Mendelian randomization study of adiposity-related traits and risk of breast, ovarian, prostate, lung and colorectal cancer . Int J Epidemiol. 2016 ; 45 3 : 896 – 908 . Google Scholar Crossref Search ADS PubMed 33 Dixon S , Nagle C , Thrift A , et al. . Adult body mass index and risk of ovarian cancer by subtype: A Mendelian randomization study . Int J Epidemiol. 2016 ; 45 3 : 884 – 895 . Google Scholar Crossref Search ADS PubMed 34 Thrift AP , Shaheen NJ , Gammon MD , et al. . Obesity and risk of esophageal adenocarcinoma and Barrett's esophagus: A Mendelian randomization study . J Natl Cancer Inst. 2014 ; 106 11 :dju252. 35 Schmid D , Leitzmann MF. Television viewing and time spent sedentary in relation to cancer risk: A meta-analysis . J Natl Cancer Inst. 2014 ; 106 7 :dju098. 36 Moore S , Lee I , Weiderpass E , et al. . Association of leisure time physical activity with risk of 26 types of cancer in 1.44 million adults . JAMA Intern Med. 2016 ; 176 6 : 816 – 825 . Google Scholar Crossref Search ADS PubMed 37 Hou L , Ji B , Blair A , et al. . Commuting physical activity and risk of colon cancer in Shanghai, China . Am J Epidemiol. 2004 ; 160 9 : 860 – 867 . Google Scholar Crossref Search ADS PubMed 38 Klement RJ , Fink MK. Dietary and pharmacological modification of the insulin/IGF-1 system: Exploiting the full repertoire against cancer . Oncogenesis. 2016 ; 5 : e193 . Google Scholar Crossref Search ADS PubMed 39 Brown SB , Hankinson SE. Endogenous estrogens and the risk of breast, endometrial, and ovarian cancers . Steroids. 2015 ; 99 ( Pt A ): 8 – 10 . Google Scholar Crossref Search ADS PubMed 40 Wang DY , Key TJ , Pike MC , et al. . Serum hormone levels in Bristish and rural Chinese females . Breast Cancer Res Treat. 1991 ; 18(Suppl 1) : S41 – S45 . Google Scholar Crossref Search ADS PubMed 41 Monteiro-Junior RS , Maciel-Pinheiro PdT , Portugal EdMM , et al. . Effect of exercise on inflammatory profile of older persons: Systematic review and meta-analyses . J Phys Act Health. 2018 ; 15 1 : 64 – 71 . Google Scholar Crossref Search ADS PubMed 42 Fedewa M , Hathaway E , Ward-Ritacco C. Effect of exercise training on C reactive protein: A systematic review and meta-analysis of randomised and non-randomised controlled trials . Br J Sports Med. 2017 ; 51 8 : 670 – 676 . Google Scholar Crossref Search ADS PubMed 43 Kang D , Lee J , Suh S , et al. . Effects of exercise on insulin, IGF axis, adipocytokines, and inflammatory markers in breast cancer survivors: A systematic review and meta-analysis . Cancer Epidemiol Biomarkers Prev. 2017 ; 26 3 : 355 – 365 . Google Scholar Crossref Search ADS PubMed 44 Ennour-Idrissi K , Maunsell E , Diorio C. Effect of physical activity on sex hormones in women: A systematic review and meta-analysis of randomized controlled trials . Breast Cancer Res. 2015 ; 17 1 : 139 . Google Scholar Crossref Search ADS PubMed 45 Cronin O , Keohane D , Molloy M , et al. . The effect of exercise interventions on inflammatory biomarkers in healthy, physically inactive subjects: A systematic review . QJM . 2017 ; 110 10 : 629 – 637 . Google Scholar PubMed 46 Liesenfeld DB , Grapov D , Fahrmann JF , et al. . Metabolomics and transcriptomics identify pathway differences between visceral and subcutaneous adipose tissue in colorectal cancer patients: The ColoCare study . Am J Clin Nutr. 2015 ; 102 2 : 433 – 443 . Google Scholar Crossref Search ADS PubMed 47 Lu Y , Cross A , Murphy N , et al. . Comparison of abdominal adiposity and overall obesity in relation to riak of small intestinal cancer in a European prospective cohort . Cancer Causes Control. 2016 ; 27 7 : 919 – 927 . Google Scholar Crossref Search ADS PubMed 48 Boffetta P , Hazelton W , Chen Y , et al. . Body mass, tobacco smoking, alcohol drinking and risk of cancer of the small intestine—a pooled analysis of over 500,000 subjects in the Asia cohort consortium . Ann Oncol. 2012 ; 23 7 : 1894 – 1998 . Google Scholar Crossref Search ADS PubMed 49 Samanic C , Gridley G , Chow W , et al. . Obesity and cancer risk among white and black United States veterans . Cancer Causes Control. 2004 ; 15 1 : 35 – 43 . Google Scholar Crossref Search ADS PubMed 50 Larsson S , Wolk A. Overweight and obesity and incidence of leukemia: A meta-analysis of cohort studies . Int J Cancer. 2008 ; 122 6 : 1418 – 1421 . Google Scholar Crossref Search ADS PubMed 51 Wallin A , Larsson S. Body mass index and risk of multiple myeloma: A meta-analysis of prospective studies . Eur J Cancer. 2011 ; 47 11 : 1606 – 1615 . Google Scholar Crossref Search ADS PubMed 52 Larsson S , Wolk A. Body mass index and risk of non-Hodgkin's and Hodgkin's lymphoma: A meta-analysis of prospective studies . Eur J Cancer. 2011 ; 47 16 : 2422 – 2430 . Google Scholar Crossref Search ADS PubMed 53 Thun M , Tobacco Freedman N. In: Thun M , Linet M , Cerhan J , et al. ., eds. Schottenfeld and Fraumeni Cancer Epidemiology and Prevention . 4th ed . New York : Oxford University Press ; 2018 . 54 US Department of Health and Human Services. The health consequences of smoking: 50 years of progress. In: A Report of the Surgeon General. Rockville, MD: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2014. 55 Carter BD , Abnet CC , Feskanich D , et al. . Smoking and mortality - beyond established causes . N Engl J Med. 2015 ; 372 7 : 631 – 640 . Google Scholar Crossref Search ADS PubMed 56 Siegel RL , Jacobs EJ , Newton CC , et al. . Deaths due to cigarette smoking for 12 smoking-related cancers in the United States . JAMA Intern Med. 2015 ; 175 9 : 1574 – 1576 . Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JNCI: Journal of the National Cancer Institute Oxford University Press

An Integrative Approach for Deciphering the Causal Associations of Physical Activity and Cancer Risk: The Role of Adiposity

Loading next page...
 
/lp/ou_press/an-integrative-approach-for-deciphering-the-causal-associations-of-StuU0urg4x
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com
ISSN
0027-8874
eISSN
1460-2105
D.O.I.
10.1093/jnci/djy091
Publisher site
See Article on Publisher Site

Abstract

Abstract Higher physical activity levels have been associated with about 10% to 25% reductions in up to 13 cancers. In isolation, these results are suggestive but not compelling enough to conclude causal associations, except for colon and breast cancer. However, knowledge on the relationships between obesity and cancer, between physical activity and overall and visceral adiposity, and between physical activity and obesity-related mediators of cancer risk can inform the epidemiology of physical activity and cancer. Excluding primarily smoking-related malignancies, for which residual confounding by smoking may occur, all 13 cancers associated with lower physical activity are also obesity-related. Moreover, the magnitude of the inverse association between physical activity and cancer type correlates highly with the association with body mass index (BMI) and cancer type (Spearman r = .79, two-sided P = .004). Physical activity lowers essentially all the obesity-related mediators of cancer, probably mediated largely through reductions in visceral adiposity. These findings strongly suggest that physical activity and adiposity are largely operating through similar carcinogenic mechanisms. That BMI has more robust associations than physical activity with cancer largely reflects that most populations studied have had great variation in BMI and little in physical activity. In populations with higher levels of physical activity and a lower range of BMI, physical activity may appear relatively more important. It may be useful to emphasize to clinicians and the public that physical activity, by acting on the same mechanisms, is likely to reduce risk of obesity-related cancers, even if the impact on lowering BMI is minimal. Determining the causal role of physical activity, a modifiable lifestyle factor, on cancer risk is of high interest. Typically, randomized controlled trials (RCTs) are considered to provide the strongest level of evidence, but they are not feasibly conducted for physical activity and cancer incidence. Hence, the examination of independent causal effects of physical activity has focused on strongly designed cohort studies, which can control for confounding factors. These have typically been conducted in highly economically developed countries (eg, in North America, Europe), and leisure time physical activity has been the predominant form of exercise studied. To date, lower levels of physical activity have generally been acknowledged to be associated with only cancers of the colon, breast, and possibly endometrium (1–3). The main limitations of these studies have been moderate associations (eg, 10%–20% reductions in risk) and the potential for residual confounding because leisure time physical activities tend to be correlated with healthy behaviors. In addition, studies have generally adjusted for body mass index (BMI), but whether effects of physical activity are truly independent from adiposity is unclear. The focus on leisure time physical activity may be practical because leisure activity is most feasibly modified in current populations. Nonetheless, in evaluating the evidence for the role of physical activity on cancer risk, a broader-based synthesis might be elucidative. First, a careful consideration of the relationship between physical activity and adiposity across the life course is important. In particular, how the types and ranges of activities in populations relate to adiposity level in the population is important to consider. Second, this knowledge between physical activity and adiposity may help explain epidemiologic patterns for both obesity and physical activity and cancer, including differences in cancer rates across populations and over time, as well as between individuals within populations. Third, the inter-relations between physical activity and body adiposity can be integrated into studies focused on mechanisms. Overall, this synthesis will evaluate the evidence on whether physical activity should be considered a factor for obesity-related cancers, acting on cancer through largely the same mechanisms operative for obesity and cancer. The Relationship Between Physical Activity and Adiposity Level Although it seems intuitive that physical activity influences adiposity, the precise role of physical activity on body fatness is not straightforward. We can consider two broad aspects of physical activity and adiposity, the relationship in terms of reduction of body weight among the overweight and obese vs the prevention of weight gain over the life course. Much research has focused on physical activity in terms of weight loss. In general, a reduction in energy intake is likely more important than physical activity for weight loss (4). Yet, the cumulative effect of physical activity for preventing weight gain over the life course, as well as in preventing regain once weight is lost, may be important. This section will consider 1) the relationship between physical activity and adiposity from a population-level perspective, particularly time trends; 2) this relationship studied between individuals within populations; 3) effects of physical activity on adiposity compartments, particularly on visceral adiposity, which may operate on a shorter term, and which are not reflected, or only poorly reflected, by BMI measures. Life Course Aspects: Across Populations Ng and Popkin used an array of longitudinal and cross-sectional data sets to assess energy expenditure in four domains of physical activity (occupation, domestic production, travel, and leisure) for adults from United States, the United Kingdom, Brazil, and China, going as far back as 1961 (5). Though the timing differed across the countries, there was a marked decrease of physical activity, primarily due to occupational activity, as well as transportation and household chores, and an increase in sedentary behaviors, including TV watching, computing, and Internet use, among others. The timing is related to changes in the economy and occupations (earlier in the United States and United Kingdom, later in China and Brazil). The decline in occupational physical activity and increase in sedentary time have corresponded in time to increasing rates of obesity across populations worldwide (6). Populations that maintain high rates of occupational and transport-related physical activity with minimal use of technologies that facilitate sedentary lifestyles have a relatively uniformly low body mass. For example, in groups that have refrained from adopting modern technology (eg, Amish), both adults and youth have much higher levels of physical activity and lower BMIs compared with their contemporaries (7,8). Changes in the availability and quality of food have undoubtedly contributed to the rising rates of obesity as well (9). It is beyond the scope of this review to quantify the relative contributions of changes in energy expenditure and intake to the rise in obesity, but it appears undeniable that decreases in physical activity are important contributors. Life Course Aspects: Among Individuals Within Populations Quantifying the influence of physical activity over the life course on adiposity at the individual level in a modern population is difficult. Assuming that occupational activity will not rise, leisure time physical activity is most relevant. Short-term intervention studies suggest that physical activity could help in preventing population weight gain. For example, a physical activity intervention over six to 12 months may result in less weight gain (eg, 1 kg) in the intervention group; assuming such an intervention is carried forward and adhered to for multiple decades, an important influence on cumulative body weight would likely emerge. Overall, the evidence strongly supports that physical activity can be an important factor in reducing weight gain over the life course, but how much so is difficult to estimate. Although the longer-term effects are not feasibly studied with interventions, observational studies support that some level of physical activity is useful for preventing or minimizing weight gain. For example, based on long-term observational studies, the American College of Sports Medicine concluded in 2009 that 150–250 minutes per week of moderate-intensity physical activity are effective in preventing weight gain (10). Sedentary lifestyle and daily life activities (eg, ≤1.5 metabolic equivalents energy expenditure) are also likely to influence cumulative weight gain (11), but further study for quantification of effects in populations is required. Shorter-Term Aspects of Physical Activity on Visceral Adiposity A consideration of measures of the specific adiposity used in epidemiologic studies of physical activity to determine independent effect of adiposity is important. BMI, which is typically used, is a reasonably good surrogate of overall adiposity, but it may not capture all of the adiposity-related effects of physical activity. Visceral fat is only a few percent of total bodyweight. A loss of half a kilogram of visceral fat would have major health benefits, but this would hardly impact the total BMI. Intervention studies of supervised aerobic exercise ranging from weeks to one year show sizable absolute reduction in visceral adiposity, with no or minimal loss in overall body weight (12–15). A meta-analysis showed that even in the absence of weight loss, exercise was associated with a 6.1% decrease in visceral adiposity (16). Further, weight loss by physical activity induced a larger loss in visceral fat than a comparable weight loss by diet (16). Some evidence also suggests that physical inactivity (eg, bed rest) preferentially leads to an increase in visceral adiposity (17). These changes in visceral adiposity may largely underlie effects of physical activity on obesity-related biomarkers. Physical activity normalizes metabolic biomarkers (eg, insulin, inflammatory markers, glucose, adiponectin, and insulin resistance) in proportion to reductions in visceral adiposity independent of weight loss or BMI) (18–23). As a typical example, a 12-week physical activity intervention in obese elderly had a minimal impact on BMI (reduced from 33.3 to 32.1 kg/m2) but reduced visceral adipose tissue by 22.4% and insulin resistance by 31.2%, and the change in visceral adipose reduction correlated highly with the change in insulin resistance (r = .66) (23). Thus, physical activity has an important influence on adiposity-related biomarkers not captured by BMI measures, even in obese individuals. How sedentary lifestyle and daily life activities influence visceral adiposity is less understood. Physical Activity, Excess Adiposity, and Cancer: An Epidemiologic Perspective The brief description of the relationship between physical activity and adiposity above help explain the epidemiology of physical activity and obesity and cancer across the spectrum of human populations. High levels of physical activity in pre-industrial populations help keep population-wide BMI levels in the low range. When the average level of physical activity in a population becomes very low, various contributors to the variation in BMI will then be unleashed; these include genetics, epigenetics, dietary factors, self-imposed behavioral constraints on overeating, smoking, and leisure time and transportation physical activities. As physical activity has generally diminished in many populations, it has become a less important factor in accounting for variation in BMI within a population. How the economic transition in countries generates comparisons for physical activity and cancer from three epidemiologic perspectives is illustrated in Figure 1: A) cancer rates in highly active populations vs those in sedentary populations (“ecologic” comparisons); B) physically active vs less active individuals within sedentary populations; C) physically active vs less active individuals within highly active populations. Figure 1. View largeDownload slide Conceptual model of transitioning from a highly physically active traditional population to a sedentary modern population and the relationship to population range in body mass index (BMI, kg/m2). A) Ecologic comparison; high physical activity level–low adiposity are strongly intercorrelated, so it is difficult to separate their independent effects. B) Within a highly active population, body mass index (BMI) range is limited and physical activity may appear as a dominant factor. C) With low mandatory activities in modern sedentary population, a wide range of BMIs emerge, for which leisure time physical activity is a moderate determinant among many (eg, diet, genetics); BMI may appear as a dominant factor. BMI = body mass index. Figure 1. View largeDownload slide Conceptual model of transitioning from a highly physically active traditional population to a sedentary modern population and the relationship to population range in body mass index (BMI, kg/m2). A) Ecologic comparison; high physical activity level–low adiposity are strongly intercorrelated, so it is difficult to separate their independent effects. B) Within a highly active population, body mass index (BMI) range is limited and physical activity may appear as a dominant factor. C) With low mandatory activities in modern sedentary population, a wide range of BMIs emerge, for which leisure time physical activity is a moderate determinant among many (eg, diet, genetics); BMI may appear as a dominant factor. BMI = body mass index. Cancer Rates in Highly Active Populations vs Those in Sedentary Active Populations Based primarily on associations in cohort study populations, obesity is generally considered to be related to increased risk of at least 13 cancers (24). Many of the 13 cancers now associated with obesity have been relatively rare or much less prevalent in pre-industrial populations and are generally more prevalent in economically developed countries. Exceptions are liver and stomach cancer, which, although related to obesity, have additional strong infectious causes. Differences in physical activity and correspondingly in adiposity may contribute to if not largely account for variability in obesity-related cancer rates between populations. Although ecologic studies are less able to account for confounding factors and are downplayed, they represent the most extreme contrast of “highly physically active/low adiposity” vs “sedentary/high adiposity.” In this ecologic type of comparison, high physical activity level and low adiposity are strongly intercorrelated, so it is difficult, or perhaps fruitless, to attempt to separate their independent causal effects. For example, in the extreme theoretical scenario that everyone is highly physically active and hence everyone has a low BMI, there would be no cancers attributable to obesity. In the opposite extreme of a completely sedentary population, the obesity rate would be high, and its variation would be completely due to factors other than physical activity. Thus, compared with the highly physically active population, all of the apparent “obesity-related” cancers are due to lack of physical activity, but from the perspective of those in the nonactive population, physical activity appears to be an irrelevant factor. Physically Active vs Less Active Individuals Within Sedentary Populations In sedentary populations, physical activity, mostly leisure time activity, will be only a moderate determinant of BMI. Other determinants, including diet, eating habits, and genetics will contribute to wide variability of BMI (from 22 to 35 kg/m2, for example). In this context, BMI is a relatively robust measure of adiposity, and substantial variation exists between individuals. Additionally, BMI tracks well over time in an individual, so even a single measure provides a robust assessment of long-term exposure, though repeated measures may strengthen estimates (25). Indeed, all these features of even a relatively simple measure such as BMI have helped establish obesity as a causal risk factor for at least 13 cancers. Independently of BMI, abdominal adiposity is associated with an increased risk of cancer (24,26–28). The argument that these associations with obesity are causal is supported by robust dose-response associations in cohort studies, numerous plausible mechanisms, and recent Mendelian randomization studies of adiposity-related traits and cancer risk (27,29–34). For physical activity and cancer, the evidence for causality has been less robust than for obesity, with the exceptions of cancers of the colon, breast, and possibly endometrium (1–3). Interestingly, some evidence suggests that sedentary behaviors may be associated with higher risk of colon and endometrial cancer (35). The above synthesis of the inter-relation of physical activity and adiposity raises the question of should physical activity also be associated with obesity-related cancers. To shed light on this question, Table 1 shows a comparison of results from a recent analysis that pooled data from 12 prospective US and European cohorts with self-reported leisure time physical activity. This is the largest relevant study to date, with 1.44 million participants, providing sufficient statistical power to examine multiple cancer types (36). The authors reported statistically significant or suggestive (P ≤ .1) inverse associations with 17 cancers (Table 1). A striking pattern is apparent. Among the cancers associated with lower physical activity levels, 13 are also established obesity-related cancers (details in table 1). Moreover, an increase in cancer risk associated with BMI (from meta-analyses of cohort studies) correlates strongly with the reduction in cancer risk associated with physical activity (Pearson r = .75, two-sided P = .008; Spearman r = .79, two-sided P = .004) (see Figure 2 for details). This high concordance between obesity-related and physical activity–related cancers strongly suggests that physical activity and adiposity are largely operating through similar carcinogenic mechanisms. If physical activity and adiposity were acting on cancer through entirely different mechanisms, we would not expect this striking pattern. Table 1. Cancers inversely associated with leisure time physical activity from Moore et al. (36) and whether they are associated with obesity and tobacco Cancer type Obesity-related Tobacco-related* Esophageal adenocarcinoma† Yes No Gallbladder† Yes No Liver† Yes No Lung No Yes Kidney† Yes No Small intestine (47–49) Yes No Gastric cardia† Yes No Endometrial† Yes No Esophageal squamous No Yes Myeloid leukemia (50) Yes No Myeloma (51) Yes No Colon† Yes No Head and neck No Yes Rectum† Yes No Bladder No Yes Breast† Yes No Non-Hodgkin lymphoma (52) Yes No Cancer type Obesity-related Tobacco-related* Esophageal adenocarcinoma† Yes No Gallbladder† Yes No Liver† Yes No Lung No Yes Kidney† Yes No Small intestine (47–49) Yes No Gastric cardia† Yes No Endometrial† Yes No Esophageal squamous No Yes Myeloid leukemia (50) Yes No Myeloma (51) Yes No Colon† Yes No Head and neck No Yes Rectum† Yes No Bladder No Yes Breast† Yes No Non-Hodgkin lymphoma (52) Yes No * Those cancers for which tobacco accounts for at least 40% of the total (53; modifed from [54–56]). † Acknowledged as an established association with obesity by the World Cancer Research Fund/American Institute of Cancer Research. Table 1. Cancers inversely associated with leisure time physical activity from Moore et al. (36) and whether they are associated with obesity and tobacco Cancer type Obesity-related Tobacco-related* Esophageal adenocarcinoma† Yes No Gallbladder† Yes No Liver† Yes No Lung No Yes Kidney† Yes No Small intestine (47–49) Yes No Gastric cardia† Yes No Endometrial† Yes No Esophageal squamous No Yes Myeloid leukemia (50) Yes No Myeloma (51) Yes No Colon† Yes No Head and neck No Yes Rectum† Yes No Bladder No Yes Breast† Yes No Non-Hodgkin lymphoma (52) Yes No Cancer type Obesity-related Tobacco-related* Esophageal adenocarcinoma† Yes No Gallbladder† Yes No Liver† Yes No Lung No Yes Kidney† Yes No Small intestine (47–49) Yes No Gastric cardia† Yes No Endometrial† Yes No Esophageal squamous No Yes Myeloid leukemia (50) Yes No Myeloma (51) Yes No Colon† Yes No Head and neck No Yes Rectum† Yes No Bladder No Yes Breast† Yes No Non-Hodgkin lymphoma (52) Yes No * Those cancers for which tobacco accounts for at least 40% of the total (53; modifed from [54–56]). † Acknowledged as an established association with obesity by the World Cancer Research Fund/American Institute of Cancer Research. Figure 2. View largeDownload slide The correlation between percent (%) risk reduction due to physical activity from Moore et al. (36) and the % risk increase due to body mass index (BMI) in 11 cancer sites from studies with available data of BMI from meta-analyses of prospective cohort studies (24). All meta-analyses are from the World Cancer Research Fund/American Institute of Cancer Research, except for lymphoma (52), multiple myeloma (51), and myeloid leukemia (50). Figure 2. View largeDownload slide The correlation between percent (%) risk reduction due to physical activity from Moore et al. (36) and the % risk increase due to body mass index (BMI) in 11 cancer sites from studies with available data of BMI from meta-analyses of prospective cohort studies (24). All meta-analyses are from the World Cancer Research Fund/American Institute of Cancer Research, except for lymphoma (52), multiple myeloma (51), and myeloid leukemia (50). Of note, the remaining four cancers associated with lower physical activity (lung, esophagus [squamous], bladder, and head and neck) but not with obesity (36) have smoking as an important primary cause (table 1). For these malignancies, residual confounding from smoking is possible, as those who suffer the greatest toxic effects of smoking (especially respiratory) may be less able to exercise. In fact, these four cancers were not associated with physical activity among never-smokers, an analysis less prone to smoking-related confounding. It is also plausible that physical activity could offset some specific carcinogenic effects related primarily to smoking, such as reduced immunity. Physically Active vs Less Physically Active Individuals Within Highly Active Populations Minimal data exist on physical activity and cancer risk in populations with high levels of physical activity. Interestingly, a case–control study of colon cancer conducted in Shanghai, China, between 1990 and 1993 may represent a more traditional population transitioning to a sedentary modern one (37). Around this time, colon cancer rates, historically extremely low, were increasing in China. The amount of physical activity was much higher than in modern populations, and average BMI was low, with very few individuals in the overweight/obese range. Among those in the high normal range in BMI, a sevenfold risk gradient was observed for low versus high physical activity. This finding suggests when the level of physical activity is high, it may be a powerful determinant of colon cancer risk, even if few are in the overweight/obese range (BMI > 25 kg/m2) in that population. Of note, in Asian populations with low BMI, visceral adiposity is independently associated with higher risk of colorectal neoplasia (26). Thus, the lower risk of colon cancer in the Shanghai case–control study in highly physically active participants could possibly reflect low visceral fat. Physical Activity, Excess Adiposity, and Cancer: A Mechanistic Perspective The mechanisms underlying the association between adiposity and cancer risk are not fully established, but the main mechanisms generally discussed include sex steroids (eg, breast, endometrial cancers), metabolic hormones (eg, gastrointestinal, reproductive cancers) and inflammation (eg, immune-related and other cancers) (24). Insulin and insulin-like growth factor 1 (IGF1) are the major hormones that activate the PI3K-Akt and ras-MAPK signaling pathways, which stimulate mitosis and inhibit apoptosis in sensitive tissues (38). Estrogen is an established causal factor for breast and endometrial cancer (39), and insulin (or C-peptide) levels are robustly associated with colorectal, pancreatic, and endometrial cancer, and IGF1 with colorectal and prostate cancer. Additional obesity-related hormones, including leptin, adiponectin, IGF-binding proteins, and sex hormone binding globulin may also be involved. Higher levels of many of these hormones have been associated with a higher risk of specific cancers in prospective studies (24). Not surprisingly, obesity-related biomarkers (eg, estrogen, metabolic hormones) are lower in countries with high physical activity and low BMI (40). These hormones, as well as likely unidentified ones, are biologically inter-related. These biomarkers can be utilized to examine the roles of adiposity and physical activity cancer risk from a mechanistic basis. From a practical perspective, these biomarkers do not necessarily have to be the precise isolated causal factors, but rather, they act as surrogates of causal mediators. In general, BMI has a stronger association with cancer-related biomarkers than physical activity, at least in cross-sectional studies conducted in sedentary, high-BMI populations. In some studies, including intervention studies, the effect of physical activity on biomarkers including insulin, estrogens, and inflammatory markers appeared to be partially or even largely due to effects on adiposity, especially visceral adiposity (41–45). It is apparent that physical activity affects essentially all of the putative obesity-related biomarkers for cancer, but with weaker associations than BMI has with these biomarkers, a similar pattern as with physical activity obesity with cancer risk. Moreover, adjusting for BMI generally attenuated the associations between physical activity and cancer, though inverse associations remained statistically significant for most of the cancers (36). Similarly, the association between physical activity and biomarkers generally weakens when adjusted for BMI, but typically remains. A potential reason, as described above, is that physical activity is related to visceral adiposity, which influences biomarkers, independently of BMI. Summary and Public Health and Research Implications The synthesis integrated herein was based on our understanding of the close relationship between physical activity and adiposity, their effects on putative biochemical mediators, and the range of physical activity and BMI across the human experience and may help explain the following observations: 1) in comparisons across countries, the traditional populations with high physical activity (especially occupational) and low BMI generally have lower risks of the 13 “obesity-related” cancers than modern sedentary populations; 2) these 13 cancers have been associated with obesity (primarily BMI) at the individual level within populations, and more weakly with physical activity; 3) the increase in cancer risk associated with BMI correlates strongly with the reduction in cancer risk associated with physical activity; 4) in modern sedentary populations, BMI, which has a wide range and is measured relatively well, is more strongly associated with cancer risk than is leisure time physical activity; 5) though data are limited, in traditional populations, the association between the physical activity and cancer may be stronger than that for BMI and cancer, because physical activity is high whereas BMI is low and has a narrow range; 6) the relationships between BMI and physical activity and putative cancer-related biomarkers are similar to those of the relationship between BMI and physical activity and cancer risk; and 7) physical activity reduces visceral fat independently of effects on weight loss or BMI, and visceral fat predicts cancer-related biomarkers and cancer risk independently of BMI. If these conclusions are valid, physical activity may help prevent many more cancers than is currently appreciated. Only two or three cancers are considered associated with physical activity, but there may be up to 13, with about 10% to 25% reductions in relative risk for the most active compared with the least active individuals. Although the magnitudes in reduction appear relatively modest, several factors are important to note. First, this number may be an underestimate due to measurement error, which attenuates associations, and to adjustment for BMI; such adjustment may be considered “overcontrol” if the effects or physical activity are largely adiposity-related. Second, the level of physical activity in most populations is very low, at least from the perspective of human history before 50 years ago. Third, most studies are based on one measurement and may not incorporate the potential benefits of physical activity in minimizing weight gain over the life course. It may be useful to emphasize to clinicians and the public that physical activity, by acting on the same obesity-related mechanisms, is likely to reduce risk of obesity-related cancers even if its impact on lowering BMI is minimal. Given the difficulty in achieving lasting weight loss, this advice may be important. From a research perspective, rather than considering physical activity and adiposity as two separate concepts, a more integrated approach may be useful. For example, short-term physical activity interventions may inform BMI-independent effects but may not necessarily be relevant for considering life-long cumulative effects of physical activity on weight maintenance. Potential adiposity-independent mechanisms of physical activity should be studied, but concluding that an effect is truly adiposity independent may require more sensitive measures of adiposity, such as visceral adiposity tissue. The influence of different types (eg, intensities) of physical activities on levels obesity-related biomarkers (eg, sex hormones, metabolic hormones, inflammation) may be informative for cancer risk; these studies need to be complemented with a better understanding of how these factors influence cancer risk. A better understanding of how different components of adiposity (eg, subcutaneous, ectopic, visceral) influence cancer risk will likely ultimately aid in understanding the role of physical activity and cancer (46). Studies in populations with higher levels of physical activity should be conducted when feasible. The issue of measurement error in assessing physical activity is critically important. Most of the evidence for an association of physical activity 1) on measures of adiposity, 2) in relation to biomarkers (from both observational and interventional studies), and 3) on cancer risk have focused on moderate and vigorous physical activities, mostly leisure time. In part, this may reflect that these activities are generally better and have been more often assessed by conventional methods, and although they incur measurement error, the measurement may be adequate to detect these associations. Although questionnaires may do a reasonable job of assessing planned moderate and vigorous physical activities, they may not do as well in measuring sedentary behaviors, daily life activities, and long bouts of physical inactivity. Although not as well studied as moderate/vigorous activities, some evidence suggests that sedentary behaviors may be associated with higher risk of some cancers associated with obesity and physical inactivity (colon, endometrial cancer) but not with other associated cancers (35). More study is required to determine how measurement error in the current measures affects these findings. Of note, sedentary time has increased across populations over time, corresponding to the decrease in occupational physical activity, but whether physiologically important variation in sedentary time can be captured well enough to robustly predict biomarkers, adiposity, and cancer risk remains to be established. More precise measures of these may be helpful. Although further research is required to strengthen evidence for or clarify some specific issues, the overall coherence of these observations strongly suggests that physical activity and adiposity largely operate through similar carcinogenic mechanisms. Notes Affiliations of author: Department of Nutrition and Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA; Channing Division of Network Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, MA. The author has no disclosures to report. References 1 Rezende LFM , Sá TH , Markozannes G , et al. . Physical activity and cancer: An umbrella review of the literature including 22 major anatomical sites and 770 000 cancer cases . Br J Sports Med. Nov 16, 2017: bjsports-2017-098391. doi: 10.1136/bjsports-2017-098391. [Epub ahead of print]. 2 International Agency for Research on Cancer (IARC) . Weight control and physical activity. In: Vainio H, Bianchini F, eds. IARC Handbook of Cancer Prevention . Lyon : IARC Press ; 2002 :83–167. 3 World Cancer Research Fund/American Institute for Cancer Research CUP . Food, nutrition, physical activity and the prevention of colorectal cancer. 2017 . https://www.wcrf.org/int/continuous-update-project/continuous-update-project-findings-reports. Accessed March 27, 2018. 4 Greenway F. Physiological adaptations to weight loss and factors favouring weight regain . Int J Obes (Lond) . 2015 ; 39 8 : 1188 – 1196 . Google Scholar Crossref Search ADS PubMed 5 Ng SW , Popkin BM. Time use and physical activity: A shift away from movement across the globe . Obes Rev. 2012 ; 13 8 : 659 – 680 . Google Scholar Crossref Search ADS PubMed 6 NCD Risk Factor Collaboration (NCD-RisC) . Trends in adult body-mass index in 200 countries from 1975 to 2014: A pooled analysis of 1698 population-based measurement studies with 19-2 million participants . Lancet . 2016 ; 387 10026 : 1377 – 1396 . Crossref Search ADS PubMed 7 Bassett DR , Schneider PL , Huntington GE. Physical activity in an old order Amish community . Med Sci Sports Exerc. 2004 ; 36 1 : 79 – 85 . Google Scholar Crossref Search ADS PubMed 8 Bassett DR Jr , Tremblay MS , Esliger DW , et al. . Physical activity and body mass index of children in an old order Amish community . Med Sci Sports Exerc. 2007 ; 39 3 : 410 – 415 . Google Scholar Crossref Search ADS PubMed 9 Brownson RC , Boehmer TK , Luke DA. Declining rates of physical activity in the United States: What are the contributors? Annu Rev Public Health. 2005 ; 26 : 421 – 443 . Google Scholar Crossref Search ADS PubMed 10 Donnelly JE , Blair SN , Jakicic JM , et al. . American College of Sports Medicine position stand. Appropriate physical activity intervention strategies for weight loss and prevention of weight regain for adults . Med Sci Sports Exerc. 2009 ; 41 2 : 459 – 471 . Google Scholar Crossref Search ADS PubMed 11 Qi Q , Li Y , Chomistek AK , et al. . Television watching, leisure time physical activity, and the genetic predisposition in relation to body mass index in women and men . Circulation. 2012 ; 126 15 : 1821 – 1827 . Google Scholar Crossref Search ADS PubMed 12 Lee S , Kuk J , Davidson L , et al. . Exercise without weight loss is an effective strategy for obesity reduction in obese individuals with and without Type 2 diabetes . J Appl Physiol (1985) . 2005 ; 99 3 : 1220 – 1225 . Google Scholar Crossref Search ADS PubMed 13 Johnson N , Sachinwalla T , Walton D , et al. . Aerobic exercise training reduces hepatic and viseral lipids in obese individuals without weight loss . Hepatology. 2009 ; 50 4 : 1105 – 1112 . Google Scholar Crossref Search ADS PubMed 14 Gepner Y , Shelef I , Schwarzfuchs D , et al. . Effect of distinct lifestyle interventions on mobilization of fat storage pools: The CENTRAL Magnetic Resonance Imaging randomized controlled trial . Circulation . 2018 ; 137 11 : 1143 – 1157 . Google Scholar Crossref Search ADS PubMed 15 Friedenreich C , Woolcott C , McTiernan A , et al. . Adiposity changes after a 1-year aerobic exercise intervention among postmenopausal women: A randomized controlled trial . Int J Obes (Lond) . 2011 ; 35 3 : 427 – 435 . Google Scholar Crossref Search ADS PubMed 16 Verheggen RJ , Maessen MF , Green DJ , et al. . A systematic review and meta-analysis on the effects of exercise training versus hypocaloric diet: Distinct effects on body weight and visceral adipose tissue . Obes Rev. 2016 ; 17 8 : 664 – 690 . Google Scholar Crossref Search ADS PubMed 17 Belavy D , Mohlig M , Pfeiffer A , et al. . Preferential deposition of visceral adipose tissue occurs due to physical inactivity . Int J Obes (Lond) . 2014 ; 38 11 : 1478 – 1480 . Google Scholar Crossref Search ADS PubMed 18 Lesser IA , Guenette JA , Hoogbruin A , et al. . Association between exercise-induced change in body composition and change in cardiometabolic risk factors in postmenopausal South Asian women . Appl Physiol Nutr Metab. 2016 ; 41 9 : 931 – 937 . Google Scholar Crossref Search ADS PubMed 19 Friedman WJ , Dreyfus CF , McEwen B , et al. . Presynaptic transmitters and depolarizing influences regulate development of the substantia nigra in culture . J Neurosci. 1988 ; 8 10 : 3616 – 3623 . Google Scholar Crossref Search ADS PubMed 20 Lee S , Deldin AR , White D , et al. . Aerobic exercise but not resistance exercise reduces intrahepatic lipid content and visceral fat and improves insulin sensitivity in obese adolescent girls: A randomized controlled trial . Am J Physiol Endocrinol Metab. 2013 ; 305 10 : E1222 – E1229 . Google Scholar Crossref Search ADS PubMed 21 Borel AL , Nazare JA , Smith J , et al. . Visceral and not subcutaneous abdominal adiposity reduction drives the benefits of a 1-year lifestyle modification program . Obesity (Silver Spring). 2012 ; 20 6 : 1223 – 1233 . Google Scholar Crossref Search ADS PubMed 22 Kim K , Valentine RJ , Shin Y , et al. . Associations of visceral adiposity and exercise participation with C-reactive protein, insulin resistance, and endothelial dysfunction in Korean healthy adults . Metabolism. 2008 ; 57 9 : 1181 – 1189 . Google Scholar Crossref Search ADS PubMed 23 O'Leary VB , Marchetti CM , Krishnan RK , et al. . Exercise-induced reversal of insulin resistance in obese elderly is associated with reduced visceral fat . J Appl Physiol (1985) . 2006 ; 100 5 : 1584 – 1589 . Google Scholar Crossref Search ADS PubMed 24 Keum N , Song M , Giovannucci E , et al. . Obesity and body composition. In: Thun M , Linet M , Cerhan J , et al. ., eds. Schottenfeld and Fraumeni Cancer Epidemiology and Prevention . 4th ed . New York : Oxford University Press ; 2018 . 25 Thygesen LC , Gronbaek M , Johansen C , et al. . Prospective weight change and colon cancer risk in male US health professionals . Int J Cancer. 2008 ; 123 5 : 1160 – 1165 . Google Scholar Crossref Search ADS PubMed 26 Keum N , Lee D , Kim R , et al. . Visceral adiposity and colorectal adenomas: Dose-response meta-analysis of observational studies . Ann Oncol. 2015 ; 26 6 : 1101 – 1109 . Google Scholar Crossref Search ADS PubMed 27 Jarvis D , Mitchell J , Law P , et al. . Mendelian randomisation analysis strongly implicates adiposity with risk of developing colorectal cancer . Br J Cancer. 2016 ; 115 2 : 266 – 272 . Google Scholar Crossref Search ADS PubMed 28 Zhang C , Rexrode K , Dam Rv , et al. . Abdominal obesity and the risk of all-cause, cardiovascular, and cancer mortality: Sixteen years of follow-up in US women . Circulation. 2008 ; 117 13 : 1658 – 1667 . Google Scholar Crossref Search ADS PubMed 29 Tsilidis KK , Dai JY , Peters U. Editorial: Mendelian randomization analysis identifies body mass index and fasting insulin as potential causal riak factors for pancreatic cancer risk . J Natl Cancer Inst. 2017 ; 109 9 :djx043. 30 Mao Y , Yan C , Lu Q , et al. . Genetically predicted high body mass index is associated with increased gastric cancer risk . Eur J Hum Genet. 2017 ; 25 9 : 1061 – 1066 . Google Scholar Crossref Search ADS PubMed 31 Painter J , O'Mara T , Marquart L , et al. . Genetic risk score mendelian randomization shows that obesity measured as body mass index, but not waist:hip ratio, is causal for endometrial cancer . Cancer Epidemiol Biomarkers Prev. 2016 ; 25 11 : 1503 – 1510 . Google Scholar Crossref Search ADS PubMed 32 Gao C , Patel C , Michailidou K , et al. . Mendelian randomization study of adiposity-related traits and risk of breast, ovarian, prostate, lung and colorectal cancer . Int J Epidemiol. 2016 ; 45 3 : 896 – 908 . Google Scholar Crossref Search ADS PubMed 33 Dixon S , Nagle C , Thrift A , et al. . Adult body mass index and risk of ovarian cancer by subtype: A Mendelian randomization study . Int J Epidemiol. 2016 ; 45 3 : 884 – 895 . Google Scholar Crossref Search ADS PubMed 34 Thrift AP , Shaheen NJ , Gammon MD , et al. . Obesity and risk of esophageal adenocarcinoma and Barrett's esophagus: A Mendelian randomization study . J Natl Cancer Inst. 2014 ; 106 11 :dju252. 35 Schmid D , Leitzmann MF. Television viewing and time spent sedentary in relation to cancer risk: A meta-analysis . J Natl Cancer Inst. 2014 ; 106 7 :dju098. 36 Moore S , Lee I , Weiderpass E , et al. . Association of leisure time physical activity with risk of 26 types of cancer in 1.44 million adults . JAMA Intern Med. 2016 ; 176 6 : 816 – 825 . Google Scholar Crossref Search ADS PubMed 37 Hou L , Ji B , Blair A , et al. . Commuting physical activity and risk of colon cancer in Shanghai, China . Am J Epidemiol. 2004 ; 160 9 : 860 – 867 . Google Scholar Crossref Search ADS PubMed 38 Klement RJ , Fink MK. Dietary and pharmacological modification of the insulin/IGF-1 system: Exploiting the full repertoire against cancer . Oncogenesis. 2016 ; 5 : e193 . Google Scholar Crossref Search ADS PubMed 39 Brown SB , Hankinson SE. Endogenous estrogens and the risk of breast, endometrial, and ovarian cancers . Steroids. 2015 ; 99 ( Pt A ): 8 – 10 . Google Scholar Crossref Search ADS PubMed 40 Wang DY , Key TJ , Pike MC , et al. . Serum hormone levels in Bristish and rural Chinese females . Breast Cancer Res Treat. 1991 ; 18(Suppl 1) : S41 – S45 . Google Scholar Crossref Search ADS PubMed 41 Monteiro-Junior RS , Maciel-Pinheiro PdT , Portugal EdMM , et al. . Effect of exercise on inflammatory profile of older persons: Systematic review and meta-analyses . J Phys Act Health. 2018 ; 15 1 : 64 – 71 . Google Scholar Crossref Search ADS PubMed 42 Fedewa M , Hathaway E , Ward-Ritacco C. Effect of exercise training on C reactive protein: A systematic review and meta-analysis of randomised and non-randomised controlled trials . Br J Sports Med. 2017 ; 51 8 : 670 – 676 . Google Scholar Crossref Search ADS PubMed 43 Kang D , Lee J , Suh S , et al. . Effects of exercise on insulin, IGF axis, adipocytokines, and inflammatory markers in breast cancer survivors: A systematic review and meta-analysis . Cancer Epidemiol Biomarkers Prev. 2017 ; 26 3 : 355 – 365 . Google Scholar Crossref Search ADS PubMed 44 Ennour-Idrissi K , Maunsell E , Diorio C. Effect of physical activity on sex hormones in women: A systematic review and meta-analysis of randomized controlled trials . Breast Cancer Res. 2015 ; 17 1 : 139 . Google Scholar Crossref Search ADS PubMed 45 Cronin O , Keohane D , Molloy M , et al. . The effect of exercise interventions on inflammatory biomarkers in healthy, physically inactive subjects: A systematic review . QJM . 2017 ; 110 10 : 629 – 637 . Google Scholar PubMed 46 Liesenfeld DB , Grapov D , Fahrmann JF , et al. . Metabolomics and transcriptomics identify pathway differences between visceral and subcutaneous adipose tissue in colorectal cancer patients: The ColoCare study . Am J Clin Nutr. 2015 ; 102 2 : 433 – 443 . Google Scholar Crossref Search ADS PubMed 47 Lu Y , Cross A , Murphy N , et al. . Comparison of abdominal adiposity and overall obesity in relation to riak of small intestinal cancer in a European prospective cohort . Cancer Causes Control. 2016 ; 27 7 : 919 – 927 . Google Scholar Crossref Search ADS PubMed 48 Boffetta P , Hazelton W , Chen Y , et al. . Body mass, tobacco smoking, alcohol drinking and risk of cancer of the small intestine—a pooled analysis of over 500,000 subjects in the Asia cohort consortium . Ann Oncol. 2012 ; 23 7 : 1894 – 1998 . Google Scholar Crossref Search ADS PubMed 49 Samanic C , Gridley G , Chow W , et al. . Obesity and cancer risk among white and black United States veterans . Cancer Causes Control. 2004 ; 15 1 : 35 – 43 . Google Scholar Crossref Search ADS PubMed 50 Larsson S , Wolk A. Overweight and obesity and incidence of leukemia: A meta-analysis of cohort studies . Int J Cancer. 2008 ; 122 6 : 1418 – 1421 . Google Scholar Crossref Search ADS PubMed 51 Wallin A , Larsson S. Body mass index and risk of multiple myeloma: A meta-analysis of prospective studies . Eur J Cancer. 2011 ; 47 11 : 1606 – 1615 . Google Scholar Crossref Search ADS PubMed 52 Larsson S , Wolk A. Body mass index and risk of non-Hodgkin's and Hodgkin's lymphoma: A meta-analysis of prospective studies . Eur J Cancer. 2011 ; 47 16 : 2422 – 2430 . Google Scholar Crossref Search ADS PubMed 53 Thun M , Tobacco Freedman N. In: Thun M , Linet M , Cerhan J , et al. ., eds. Schottenfeld and Fraumeni Cancer Epidemiology and Prevention . 4th ed . New York : Oxford University Press ; 2018 . 54 US Department of Health and Human Services. The health consequences of smoking: 50 years of progress. In: A Report of the Surgeon General. Rockville, MD: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2014. 55 Carter BD , Abnet CC , Feskanich D , et al. . Smoking and mortality - beyond established causes . N Engl J Med. 2015 ; 372 7 : 631 – 640 . Google Scholar Crossref Search ADS PubMed 56 Siegel RL , Jacobs EJ , Newton CC , et al. . Deaths due to cigarette smoking for 12 smoking-related cancers in the United States . JAMA Intern Med. 2015 ; 175 9 : 1574 – 1576 . Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

JNCI: Journal of the National Cancer InstituteOxford University Press

Published: Sep 1, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off