Article:Abdominal Obesity and Lung Cancer Risk: Systematic Review and Meta-Analysis of Prospective Studies. (5188465)

From ScienceSource
Jump to: navigation, search

This page is the ScienceSource HTML version of the scholarly article described at Its title is Abdominal Obesity and Lung Cancer Risk: Systematic Review and Meta-Analysis of Prospective Studies. and the publication date was 2016-12-15. The initial author is Khemayanto Hidayat.

Fuller metadata can be found in the Wikidata link, which lists all authors, and may have detailed items for some or all of them. There is further information on the article in the footer below. This page is a reference version, and is protected against editing.

Converted JATS paper:

Journal Information

Title: Nutrients

Abdominal Obesity and Lung Cancer Risk: Systematic Review and Meta-Analysis of Prospective Studies

  • Khemayanto Hidayat
  • Xuan Du
  • Guochong Chen
  • Minhua Shi
  • Bimin Shi

1Department of Endocrinology and Metabolism, the First Affiliated Hospital of Soochow University, Suzhou 215006, China; (K.H.); (X.D.)

2Department of Nutrition and Food Hygiene, School of Public Health, Soochow University, Suzhou 215123, China;

3Department of Respiratory Medicine, the Second Affiliated Hospital of Soochow University, Suzhou 215000, China;

Publication date (epub): 12/2016

Publication date (collection): 12/2016


Several meta-analyses of observational studies have been performed to examine the association between general obesity, as measured by body mass index (BMI), and lung cancer. These meta-analyses suggest an inverse relation between high BMI and this cancer. In contrast to general obesity, abdominal obesity appears to play a role in the development of lung cancer. However, the association between abdominal obesity (as measured by waist circumference (WC) (BMI adjusted) and waist to hip ratio (WHR)) and lung cancer is not fully understood due to sparse available evidence regarding this association. PubMed and Web of Science databases were searched for studies assessing the association between abdominal obesity and lung cancer up to October 2016. The summary relative risks (RRs) with 95% confidence intervals (CIs) were calculated with a random-effects model. Six prospective cohort studies with 5827 lung cancer cases among 831,535 participants were included in our meta-analysis. Each 10 cm increase in WC and 0.1 unit increase in WHR were associated with 10% (RR 1.10; 95% CI 1.04, 1.17; I2 = 27.7%, p-heterogeneity = 0.198) and 5% (RR 1.05; 95% CI 1.00, 1.11; I2 = 25.2%, p-heterogeneity = 0.211) greater risks of lung cancer, respectively. According to smoking status, greater WHR was only positively associated with lung cancer among former smokers (RR 1.11; 95% CI 1.00, 1.23). In contrast, greater WC was associated with increased lung cancer risk among never smokers (RR 1.11; 95% CI 1.00, 1.23), former smokers (RR 1.12; 95% CI 1.03, 1.22) and current smokers (RR 1.16; 95% CI 1.08, 1.25). The summary RRs for highest versus lowest categories of WC and WHR were 1.32 (95% CI 1.13, 1.54; I2 = 18.2%, p-heterogeneity = 0.281) and 1.10 (95% CI 1.00, 1.23; I2 = 24.2%, p-heterogeneity = 0.211), respectively. In summary, abdominal obesity may play an important role in the development of lung cancer.


1. Introduction

Lung cancer is the most common cancer in the world with 1.8 million new cases diagnosed in 2012, accounting for 12.9% of total cancer incidences. In terms of mortality, 1.6 million deaths were caused by lung cancer in 2012, accounting for 19.4% of total cancer deaths [[1]]. There is emerging evidence that general obesity and/or abdominal obesity is associated with increased risk of certain types of cancers, including postmenopausal breast cancer, colorectal, endometrial, esophagus, kidney, pancreatic, thyroid, and gallbladder cancers [[2],[3],[4],[5],[6],[7],[8],[9],[10],[11]]. However, evidence is less convincing and somewhat controversial for lung cancer. Several meta-analyses of observational studies have been performed to examine the association between general obesity, as measured by body mass index (BMI), and lung cancer. These meta-analyses suggest an inverse relation between high BMI and this cancer [[12],[13],[14]]. Cigarette smoking has been directly associated with increased lung cancer risk; thus, an inverse relationship between BMI and lung cancer may be a reflection of incomplete adjustment for the effects of cigarette smoking, as smoking habits may affect both body weight and body composition [[15],[16],[17]]. In contrast to general obesity, body fat distribution—particularly abdominal obesity—appears to play a role in the development of lung cancer [[18],[19],[20],[21]]. Abdominal obesity is reflected by a higher waist to hip ratio (WHR) and a higher waist circumference (WC) relative to others with similar BMI [[21]]. Findings from several prospective cohort studies have found a positive association between lung cancer incidence and WHR and/or WC after adjustment for BMI [[18],[19],[20],[21]]. While unadjusted WC reflects general obesity, only adjusted WC reflects abdominal obesity [[21]]. Furthermore, another study has also found a positive association between higher WC and lung cancer mortality [[22]]. Thus far, the association between abdominal obesity and lung cancer is not fully understood due to sparse available evidence regarding this association [[18],[19],[20],[21],[23],[24]]. Given these considerations, we conducted the present meta-analysis of prospective studies with the following objectives: (1) provide insight into and robust evidence concerning the association between abdominal obesity and lung cancer by using published prospective data; and (2) investigate and quantify the potential dose-response relationship between abdominal adiposity measures and risk of lung cancer that has not been investigated before.

2. Materials and Methods

2.1. Search Strategy

This meta-analysis was planned, conducted, and reported according to “Meta-Analysis of Observational Studies in Epidemiology (MOOSE) group” guidelines [[25]]. PubMed and Web of Science databases were searched for studies assessing the association between abdominal obesity and lung cancer up to October 2016. The following search terms were employed to retrieve the relevant literature in the databases: (adiposity OR body size OR anthropometric OR abdominal obesity OR central obesity OR obese OR abdominal adiposity OR obesity OR body composition OR body fat distribution OR body fat patterning OR retroperitoneal fat OR visceral fat OR abdominal fat OR intra-abdominal fat OR waist to hip ratio OR waist hip ratio OR waist circumference OR girth circumference OR abdominal adiposity measures OR adiposity measures) AND (lung cancer OR cancer of lung OR lung carcinoma OR carcinoma of lung OR adenocarcinoma of the lung OR lung adenocarcinoma OR small cell lung cancer OR small cell lung carcinoma OR non-small cell lung cancer OR non-small cell lung carcinoma OR squamous cell lung cancer OR squamous cell lung carcinoma OR neoplasm of lung OR lung neoplasm OR lung tumor OR tumor of lung OR lung tumour OR tumour of lung OR NSCLC OR SCLC) AND (cohort OR prospective OR follow-up OR follow up OR observational study). The search strategy had no language, publication date, or publication type restriction. In addition, the reference lists of retrieved full publications were reviewed to complement the search and to identify relevant studies that were missed during the electronic database search.

2.2. Study Selection

To be included in this meta-analysis, the studies had to meet the following inclusion criteria: (a) the study had a prospective design (including prospective cohort study, nested case-control study, and case-cohort study); (b) examined the association between measures of abdominal obesity (WC and/or WHR) and risk of lung cancer; and (c) relative risks (RRs) or hazard ratios (HRs) or odds ratios (ORs) with 95% confidence intervals (CIs) were available. Accordingly, retrospective studies, or studies on lung cancer mortality or recurrence were excluded. If multiple publications from the same study were identified, the publication containing the largest number of cases and most detailed information (i.e., reporting data for subgroup or dose-response analyses) was selected.

2.3. Data Extraction and Quality Assessment

Using a standardized data-collection form, the following data were abstracted from each study: the first author’s last name, publication year, country, study population, duration of follow-up, number of participants, number of cases, ascertainment of adiposity, measures of abdominal adiposity, most fully adjusted risk estimates with their corresponding 95% CIs for each category of abdominal adiposity measures, adjustment for anthropometric variables, and adjustment for potential confounding factors. If multiple RRs of the association were available, we extracted RRs with their corresponding 95% CIs from the models that reflected the maximum extent of adjustment for potentially confounding variables, and from the models that further adjusted for anthropometric variables (e.g., BMI, height). When studies provided risk estimates according to smoking status, we extracted all of them and used the data in subgroup analysis. The study quality was assessed using the 9-star Newcastle-Ottawa Scale (NOS) [[26]], in which each study was judged based on the selection of the study groups (representativeness, selection of non-exposed cohort, ascertainment of exposure, no disease at start of study), the comparability of the groups, and three for the quality of the outcome (assessment of outcome, length of follow-up and adequacy of follow-up). Studies with NOS values of six or greater were considered moderate to high-quality studies and those with a NOS value of less than six were regarded low-quality studies. Two investigators (K.H. and X.D.) participated in literature search, study selection, data extraction, and quality assessment independently. Any discrepancies regarding inclusion were solved through group discussion, with input from the senior investigator (B.M.S.).

2.4. Statistical Analysis

RR was chosen as the common measure of association across this study, and HR was directly considered as RR. A DerSimonian and Laird random-effects model [[27]] was used to calculate the summary risk estimates. The degree of heterogeneity in the relationship between measures of abdominal obesity and lung cancer across studies was assessed using Q and I2 statistics. For the Q statistic, p < 0.1 was considered statistically significant; and for the I2 statistic, the following conventional cut-off points were used: <25% (low heterogeneity), 25%–50% (moderate heterogeneity) and >75% (severe heterogeneity). Both Begg’s rank correlation test and Egger’s linear regression test were performed to investigate potential publication bias [[28]]. If evidence of publication bias was observed, the trim and fill method was applied to correct the bias [[29]].

We only performed subgroup analysis according to smoking status (never smokers, former smokers and current smokers) due to the limited number of studies included in this meta-analysis. In addition, to investigate the impacts of individual studies on the overall results, we also performed a sensitivity analysis by omitting one study in each turn while pooling results from the remainder. We performed a linear dose-response analysis examining the association between measures of abdominal obesity and lung cancer risk according to the method proposed by Greenland and Longnecker [[30]] and Orsini et al. [[31]]. This method requires the number of cases and person-years and the risk estimates with their variance estimates for at least three quantitative abdominal adiposity measures categories. For the studies that did not provide the number of cases and/or person-years in each abdominal obesity measure category, we estimated these data from total number of cases and person-years. For each study, the median or mean level of abdominal obesity measures for each category was assigned to each corresponding risk estimate. When the median or mean abdominal obesity measures per category were not provided, we considered the midpoint of the upper and lower boundaries in each category as a new reference. If the highest or lowest category was open-ended, we assumed the width of the interval to be the same as in the closest category. Forest plots of the linear dose-response meta-analysis were presented for each 10 cm increase in WC and for 0.1 unit increase in WHR. All statistical analyses were performed using STATA software, version 11.0 (STATA Corp., College Station, TX, USA). All p-values were two-sided, and the level of significance was at <0.05, unless explicitly stated.

3. Results

3.1. Study Characteristics

A flow chart of study selection, including reasons for exclusion, is presented in Figure 1. During the initial search, we briefly identified 1414 articles from PubMed and Web of Science databases; most were excluded because they were retrospective studies or because the exposure or outcome was not relevant to our analysis, leaving 20 potentially eligible articles for full-text review. After careful review, 14 articles were further excluded; among excluded articles, two articles were conducted in the same populations [[15],[32]], one article examined central obesity and lung cancer mortality [[22]], and the remaining 11 studies were excluded because the risk estimate for the association of interest was not available. Finally, six prospective cohort studies [[18],[19],[20],[21],[23],[24]] were included into our final analysis. The characteristics of the included studies are summarized and listed in Table 1. These studies were published between 2002 and 2016. All of the included studies had a prospective cohort design. A total of 5827 lung cancer cases were diagnosed among 831,535 participants. One prospective cohort study was conducted in China [[24]], one in European countries [[21]], and the remaining four in the USA [[18],[19],[20],[23]]. Regarding the sex of the participants, four [[18],[19],[23],[24]] studies evaluated only women, and the remaining two [[20],[21]] included both sexes. The length of follow-up ranged from seven to 15.1 years. Individual studies adjusted for a wide range of potential confounding factors, such as age, physical activity, and smoking. The details of quality assessment according to the nine-star NOS are presented in the online Supplementary Materials Table S1. All studies were given scores of ≥7.

3.2. WC and Lung Cancer

Five prospective cohort studies [[18],[19],[20],[21],[23]] were eligible for the analysis of WC and risk of lung cancer. All studies concerning this association were further adjusted for BMI. The summary RR for a 10 cm increase in WC was 1.10 (95% CI 1.04, 1.17) with moderate heterogeneity (I2 = 27.7%, p = 0.198) (Figure 2A). No evidence of publication bias was observed across studies (Begg, p = 0.404; Egger, p = 0.842). The summary RRs according to smoking status were 1.11 (95% CI 1.00, 1.23) for never smokers, 1.12 (95% CI 1.03, 1.22) for former smokers, and 1.16 (95% CI 1.08, 1.25) for current smokers (Figure 2B). Sensitivity analysis investigating the influence of a single study on the overall risk estimate by omitting one study at each turn yielded a range of RRs from 1.09 (95% CI 1.01, 1.18) to 1.13 (95% CI 1.08, 1.19). Furthermore, the summary RR for the highest versus lowest categories of WC was 1.32 (95% CI 1.13, 1.54) with low heterogeneity (I2 = 18.2%, p = 0.281) (Online Supplementary Materials Figure S1).

3.3. WHR and Lung Cancer

Six prospective cohort studies [[18],[19],[20],[21],[23],[24]] were eligible for the analysis of WHR and risk of lung cancer. The summary RR for a 0.1 unit increase in WHR was 1.05 (95% CI 1.00, 1.11) with moderate heterogeneity (I2 = 25.2%, p = 0.211) (Figure 3A). No evidence of publication bias was observed across studies (Begg, p = 0.851; Egger, p = 0.962). The summary RRs according to smoking status were 1.07 (95% CI 0.94, 1.21) for never smokers, 1.11 (95% CI 1.00, 1.23) for former smokers, and 0.99 (95% CI 0.94, 1.04) for current smokers (Figure 3B). In a sensitivity analyses in which we omitted one study at a time, the overall association became slightly attenuated (RR 1.05, 95% CI 0.98, 1.12) by excluding the study by Dewi et al. [[21]]. In addition, the summary RR for the highest versus lowest categories of WC was 1.10 (95% CI 1.00, 1.23) with low heterogeneity (I2 = 24.2%, p = 0.211) (Online Supplementary Materials Figure S2).

4. Discussion

To our knowledge, the present meta-analysis is the first quantitative review of prospective studies concerning the association between abdominal obesity and lung cancer risk. Our dose-response analysis revealed that each 10 cm increase in WC and 0.1 unit increase in WHR were associated with 10% and 5% greater risks of lung cancer, respectively. In contrast, observational studies have reported a consistent inverse relationship between general obesity, as measured by BMI, and lung cancer [[12],[13],[14]]. Thus, our findings suggest that an excess of abdominal adipose tissue, but not overall body fatness, may be a better predictor of lung cancer. Although limited, our findings may have significant public health and clinical implications because WC (BMI adjusted) and to a lesser extent WHR may represent a better predictor for lung cancer than BMI. In that regard, obtaining a WC and WHR measurement instead of BMI in individuals already at increased risk for lung cancer (e.g., in current smokers and former smokers) may provide essential information that might not be delivered by BMI because smokers tend to have greater abdominal adiposity than nonsmokers do with similar BMI [[33],[34],[35]].

The exact mechanisms for the association between abdominal obesity and lung cancer remain poorly understood. One speculative biological mechanism for the contrary associations of abdominal obesity and general obesity to lung cancer may involve complex biologic pathways, such as hyperinsulinemia, decreased levels of sex hormone binding globulin (SBHG), and increased levels of unbound androgens and estrogens. All of these biologic pathways are more strongly related to abdominal fatness than to body fatness [[15],[18],[20],[36],[37],[38]]. Moreover, several in vitro studies have shown that small-cell lung cancer and non-small-cell lung cancer respond to insulin-like growth factors I (IGF-I) [[39],[40]] and that lung cancer cells contain receptors for steroid hormones, including estrogens and androgens [[41],[42],[43]]. Nevertheless, the biologic mechanism underlying the association between abdominal obesity and lung cancer warrants further research.

Furthermore, there are several reasonable explanations for the contrary relationships of abdominal obesity and general obesity to lung cancer in regard to the residual confounding by smoking. The interrelations between general and abdominal obesity measures and smoking are complex and are subject to change over time. Smoking is a well-established risk factor for lung cancer [[44]], and is also inversely associated with body weight [[45],[46]]. Current smokers are associated with both lower BMI [[35]] and increased risk of lung cancer. By comparison, current smokers are associated with more visceral adipose tissue accumulation [[22]], and it is known that smoking cessation is associated with increased body weight and WC [[18],[22]]. If this is the case, limiting the investigation to participants who never smoked may help to overcome the issue of residual confounding by smoking in regard to both general obesity and abdominal obesity.

A dose-response meta-analysis by Duan et al. [[14]] did not find inverse association of high BMI and lung cancer when restricting analysis to non-smokers. Thus far, only a few prospective studies have examined the association between abdominal obesity and risk of lung cancer among never smokers [[18],[19],[20],[21]]. Olson et al. [[18]], Kabat et al. [[19]], and Dewi et al. [[21]] found WC was positively associated with lung cancer among smokers, but not among never smokers. All of these studies found that WC was positively associated with risk of lung cancer in smokers, but only after adjustment for BMI. Furthermore, in the largest prospective study of never smoking lung cancer to date, Lam et al. [[20]] also observed a significant positive association between WC, conditional on BMI, and lung cancer risk; this finding suggests that the observed positive associations from previous studies were not merely due to residual confounding by smoking. Nevertheless, data on an association between abdominal obesity and risk of non-smoking lung cancer remains inconclusive. Therefore, further large prospective studies conducted on never smokers are needed to eliminate residual confounding by cigarette smoking.

Our stratified analysis according to smoking status revealed that higher WC was associated with 11% increased risk of lung cancer in never smokers, whereas no association was observed for WHR. As expected, greater WC was also positively associated with lung cancer among former smokers and current smokers. Regarding WHR, we observed a positive association among former smokers, but not among current smokers. These findings were somewhat surprising and challenging to explain since both WC (BMI adjusted) and WHR are measures of abdominal obesity. There are several explanations for the discrepancies between measures of abdominal obesity. First, WC (BMI adjusted) may be a better predictor of abdominal obesity than WHR [[47],[48]]. Second, the null association of lung cancer and WHR among never smokers and current smokers may be because WHR are difficult to interpret since participants could have low abdominal fat and abundant gluteal fat or vice versa; although not for all studies, higher fat deposition in hip region (gluteofemoral fat) has been shown to be inversely associated with lung cancer [[20]]. Thus, this inverse association may weaken the overall positive association. Third, the observed positive association between greater WHR and lung cancer among former smokers is reasonable, since smoking cessation is associated with increased WC [[22]] and further corroborated the association. Finally, given a lower number of studies on WHR that further stratified for smoking status, as compared to studies on WC, the observed null association could be due to lack of statistical power to detect the true effect. Nevertheless, both WC (BMI adjusted) and WHR are both crude measures of abdominal obesity and cannot distinguish between subcutaneous fat and visceral fat. Therefore, future studies with advanced imaging techniques (i.e., magnetic resonance imaging, computed tomography) are warranted to confirm these findings.

As mentioned earlier, the findings from several prospective cohort studies have found a positive association between lung cancer incidence and WC, but only after further adjustment for BMI [[18],[19],[20],[21]]. In contrast to unadjusted WC, which is related with overall body fat, adjusted WC is associated with having more abdominal and visceral adipose tissue than others do with similar BMI and height [[21],[49]]. However, because keeping BMI and height constant reflects keeping total body mass constant, an increase in abdominal adipose tissue must be accompanied by a decrease in other parts of the body, for example lean mass reduction or gluteofemoral fat [[21],[50]]. Furthermore, lean mass reduction may be associated with preclinical lung cancer, which could also explain the positive associations from these studies [[21]].

Finally, it is worth taking into consideration that abdominal obesity and cancer share many risk factors, such as smoking, physical inactivity, and poor diet [[51],[52]]. For example, findings from our stratified analysis revealed that higher WC has the strongest effect in current smokers than other groups. It is possible that higher WC among smokers is a consequence of greater levels of smoking or other poor lifestyle habits (e.g., lack of physical activity, unhealthy diet) that may contribute to lung carcinogenesis. Given these considerations, the observed association between abdominal obesity and lung cancer may be partly due to similar confounding factors shared by both conditions. Further clarification for the issue of whether abdominal obesity itself is associated with increased risk of lung cancer, rather than a proxy for another cancer risk factor, is needed.

Strengths and Limitations

This meta-analysis has several strengths, including incorporated evidence and relevant studies to the date. Because results from individual studies often had insufficient statistical power, the enlarged sample size from this present meta-analysis may enhance the power to detect a significant association and provide more precise estimates of the effects. All of the included studies have long follow-up durations and have a prospective nature that thereby reduced the likelihood of potential biases (e.g., recall and selection biases). Furthermore, given the considerably heterogeneous categories of abdominal obesity measures among included studies, a dose-response meta-analysis is necessary since it provides better precision of risk estimates than merely conducting high versus low analysis.

There are several limitations in the present meta-analysis that should be acknowledged. First, almost all of the included studies were conducted in women. Thus, our findings might not apply to men. Unfortunately, we are unable to perform subgroup analysis according to sex of the participants in order to clarify whether the differences in body fat distribution between men and women may influence the overall positive association between abdominal obesity and risk of lung cancer. Further studies focusing on the influence of sex differences on this association are needed to address this issue. Second, although individual studies have considered a wide range of potential confounders in their analyses, the potential impacts of residual/unknown confounding factors on our findings cannot be completely excluded; for example, confounding from smoking may have contributed to observed positive association of abdominal obesity and lung cancer. Third, our analysis was limited by the number of included studies, which limits our ability to perform subgroup analyses (e.g., according to lung cancer subtypes, sex). Fourth, the possible errors in WC and WHR measurements obtained by self-report and self-measurement have led to overestimation or underestimation of the true association between abdominal obesity and lung cancer. Fifth, the possible dose-response meta-analysis measurement error should also be acknowledged, as it requires assumptions such as extrapolating the width of the open-ended lowest and highest boundaries from the closest category and assigning the midpoint of each category of abdominal obesity measures to corresponding relative risk. Sixth, because the findings of the current meta-analysis were mainly based on data from studies conducted in Western populations, additional research in other populations is warranted to generalize the findings. Finally, given all of these limitations, the results from our meta-analysis should always be treated with caution.

5. Conclusions

In summary, abdominal obesity may play an important role in the development of lung cancer. Further large prospective studies conducted in never smokers and both sexes are needed to extend and confirm our findings.



This study was supported by grants from the National Natural Science Foundation of China (No. 81272610) and from the Suzhou Social Development Fund (No. SYSD2013080 and SYS201455).


  1. J. FerlayI. SoerjomataramR. DikshitS. EserC. MathersM. RebeloD.M. ParkinD. FormanF. BrayCancer incidence and mortality worldwide: Sources, methods and major patterns in GLOBOCAN 2012Int. J. Cancer2015136E359E38610.1002/ijc.2921025220842
  2. G.C. ChenS.J. ChenR. ZhangK. HidayatJ.B. QinY.S. ZhangL.Q. QinCentral obesity and risks of pre- and postmenopausal breast cancer: A dose-response meta-analysis of prospective studiesObes. Rev.2016171167117710.1111/obr.1244327432212
  3. Y. MaY. YangF. WangP. ZhangC. ShiY. ZouH. QinObesity and risk of colorectal cancer: A systematic review of prospective studiesPLoS ONE20138e5391610.1371/journal.pone.005391623349764
  4. D. AuneD.A. Navarro RosenblattD.S. ChanS. VingelieneL. AbarA.R. VieiraD.C. GreenwoodE.V. BanderaT. NoratAnthropometric factors and endometrial cancer risk: A systematic review and dose-response meta-analysis of prospective studiesAnn. Oncol.2015261635164810.1093/annonc/mdv14225791635
  5. F. WangY. XuBody mass index and risk of renal cell cancer: A dose-response meta-analysis of published cohort studiesInt. J. Cancer20141351673168610.1002/ijc.2881324615287
  6. X. XiaW. ChenJ. LiX. ChenR. RuiC. LiuY. SunL. LiuJ. GongP. YuanBody mass index and risk of breast cancer: A nonlinear dose-response meta-analysis of prospective studiesSci. Rep.20144748010.1038/srep0748025504309
  7. J.M. GenkingerD. SpiegelmanK.E. AndersonL. BernsteinP.A. Van Den BrandtE.E. CalleD.R. EnglishA.R. FolsomJ.L. FreudenheimC.S. FuchsA pooled analysis of 14 cohort studies of anthropometric factors and pancreatic cancer riskInt. J. Cancer20111291708171710.1002/ijc.2579421105029
  8. S. HongQ. CaiD. ChenW. ZhuW. HuangZ. LiAbdominal obesity and the risk of colorectal adenoma: A meta-analysis of observational studiesEur. J. Cancer Prev.20122152353110.1097/CEJ.0b013e328351c77522343656
  9. F. TuratiI. TramacereC. La VecchiaE. NegriA meta-analysis of body mass index and esophageal and gastric cardia adenocarcinomaAnn. Oncol.20132460961710.1093/annonc/mds24422898040
  10. D. SchmidC. RicciG. BehrensM.F. LeitzmannAdiposity and risk of thyroid cancer: A systematic review and meta-analysisObes. Rev.2015161042105410.1111/obr.1232126365757
  11. Z.M. LiZ.X. WuB. HanY.Q. MaoH.L. ChenS.F. HanJ.L. XiaL.S. WangThe association between BMI and gallbladder cancer risk: A meta-analysisOncotarget20167436694367910.18632/oncotarget.966427248320
  12. A.G. RenehanM. TysonM. EggerR.F. HellerM. ZwahlenBody-mass index and incidence of cancer: A systematic review and meta-analysis of prospective observational studiesLancet200837156957810.1016/S0140-6736(08)60269-X18280327
  13. Y. YangJ. DongK. SunL. ZhaoF. ZhaoL. WangY. JiaoObesity and incidence of lung cancer: A meta-analysisInt. J. Cancer20131321162116910.1002/ijc.2771922777722
  14. P. DuanC. HuC. QuanX. YiW. ZhouM. YuanT. YuA. KouroumaK. YangBody mass index and risk of lung cancer: Systematic review and dose-response meta-analysisSci. Rep.2015191693810.1038/srep1693826582414
  15. C.R. DrinkardT.A. SellersJ.D. PotterW. ZhengR.M. BostickC.L. NelsonA.R. FolsomAssociation of body mass index and body fat distribution with risk of lung cancer in older womenAm. J. Epidemiol.19951426006077653468
  16. J. Audrain-McGovernN. BenowitzCigarette smoking, nicotine, and body weightClin. Pharmacol. Ther.20119016416810.1038/clpt.2011.10521633341
  17. A. ChioleroD. FaehF. PaccaudJ. CornuzConsequences of smoking for body weight, body fat distribution, and insulin resistanceAm. J. Clin. Nutr.20088780180918400700
  18. J.E. OlsonP. YangK. SchmitzR.A. VierkantJ.R. CerhanT.A. SellersDifferential association of body mass index and fat distribution with three major histologic types of lung cancer: Evidence from a cohort of older womenAm. J. Epidemiol.200215660661510.1093/aje/kwf08412244029
  19. G.C. KabatM. KimJ.R. HuntR.T. ChlebowskiT.E. RohanBody mass index and waist circumference in relation to lung cancer risk in the Women’s Health InitiativeAm. J. Epidemiol.200816815816910.1093/aje/kwn10918483121
  20. T.K. LamS.C. MooreL.A. BrintonL. SmithA.R. HollenbeckG.L. GierachN.D. FreedmanAnthropometric measures and physical activity and the risk of lung cancer in never-smokers: A prospective cohort studyPLoS ONE20138e7067210.1371/journal.pone.007067223940620
  21. N.U. DewiH.C. BoshuizenM. JohanssonP. VineisE. KampmanA. SteffenA. TjønnelandJ. HalkjærK. OvervadG. SeveriAnthropometry and the Risk of Lung Cancer in EPICAm. J. Epidemiol.201618412913910.1093/aje/kwv29827370791
  22. M.F. LeitzmannS.C. MooreA. KosterT.B. HarrisY. ParkA. HollenbeckA. SchatzkinWaist circumference as compared with body-mass index in predicting mortality from specific causesPLoS ONE20116e1858210.1371/journal.pone.001858221541313
  23. T.N. BetheaL. RosenbergM. CharlotG.T. O’ConnorL.L. Adams-CampbellJ.R. PalmerObesity in relation to lung cancer incidence in African American womenCancer Causes Control2013241695170310.1007/s10552-013-0245-623744044
  24. Y. LiuS. Warren AndersenW. WenY.T. GaoQ. LanN. RothmanB.T. JiG. YangY.B. XiangX.O. ShuW. ZhengProspective cohort study of general and central obesity, weight change trajectory and risk of major cancers among Chinese womenInt. J. Cancer20161391461147010.1002/ijc.3018727177094
  25. D.F. StroupJ.A. BerlinS.C. MortonI. OlkinG.D. WilliamsonD. RennieD. MoherB.J. BeckerT.A. SipeS.B. ThackerMeta-analysis of observational studies in epidemiology: A proposal for reporting. Meta-Analysis of Observational Studies in Epidemiology (MOOSE) groupJAMA20002832008201210.1001/jama.283.15.200810789670
  26. G.A. WellsB. SheaD. O’ConnellJ. PetersonV. WelchM. LososP. TugwellThe Newcastle-Ottawa Scale (NOS) for Assessing the Quality of Nonrandomized Studies in Meta-AnalysesAvailable online: on 10 March 2016)
  27. R. DerSimonianN. LairdMeta-analysis in clinical trialsControl Clin. Trials1986717718810.1016/0197-2456(86)90046-23802833
  28. M. EggerG. Davey SmithM. SchneiderC. MinderBias in meta-analysis detected by a simple, graphical testBMJ199731562963410.1136/bmj.315.7109.6299310563
  29. S. DuvalR. TweedieTrim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysisBiometrics20005645546310.1111/j.0006-341X.2000.00455.x10877304
  30. S. GreenlandM.P. LongneckerMethods for trend estimation from summarized dose-response data, with applications to meta-analysisAm. J. Epidemiol.1992135130113091626547
  31. N. OrsiniR. BelloccoS. GreenlandGeneralized least squares for trend estimation of summarized dose-response dataStata J.200664057
  32. A.R. FolsomL.H. KushiK.E. AndersonP.J. MinkJ.E. OlsonC.P. HongT.A. SellersD. LazovichR.J. PrineasAssociations of general and abdominal obesity with multiple health outcomes in older women: The Iowa Women’s Health StudyArch. Intern. Med.20001602117212810.1001/archinte.160.14.211710904454
  33. S.A. KayeA.R. FolsomR.J. PrineasJ.D. PotterS.M. GapsturThe association of body fat distribution with lifestyle and reproductive factors in a population study of postmenopausal womenInt. J. Obes.19901458359110.1016/0378-5122(91)90321-G2228393
  34. H. ShimokataD.C. MullerR. AndresStudies in the distribution of body fat. III. Effects of cigarette smokingJAMA19892611169117310.1001/jama.1989.034200800890372915440
  35. E. Barrett-ConnorK.T. KhawCigarette smoking and increased central adiposityAnn. Intern. Med.198911178378710.7326/0003-4819-111-10-7832817625
  36. J.T. SolerA.R. FolsomS.A. KayeR.J. PrineasAssociations of abdominal adiposity, fasting insulin, sex hormone binding globulin, and estrone with lipids and lipoproteins in post-menopausal womenAtherosclerosis198979212710.1016/0021-9150(89)90029-42803343
  37. V. Mohamed-AliJ.H. PinkneyS.W. CoppackAdipose tissue as an endocrine and paracrine organInt. J. Obes. Relat. Metab. Disord.1998221145115810.1038/sj.ijo.08007709877249
  38. B.L. WajchenbergSubcutaneous and visceral adipose tissue: Their relation to the metabolic syndromeEndocr. Rev.20002169773810.1210/edrv.21.6.041511133069
  39. V.M. MacauleyJ.D. TealeM.J. EverardG.P. JoshiI.E. SmithJ.L. MillarSomatomedin-C/insulin-like growth factor I is a potent mitogen for human small cell lung cancerBr. J. Cancer198857919310.1038/bjc.1988.162831929
  40. D.P. AnkrappD.R. BevanInsulin-like growth factor I and human fibroblast-derived insulin-like growth factor-I stimulate the proliferation of human lung carcinoma cells in vitroCancer Res.199353339934048391925
  41. C.W. BeattieN.W. HansenP.A. ThomasSteroid receptors in human lung cancerCancer Res.198545420644143875407
  42. P.T. CagleD.R. ModyM.R. SchwartzEstrogen and progesterone receptors in bronchiogenic carcinomaCancer Res.199050663266352208126
  43. D.C. Marquez-GarbanV. MahM. AlaviE.L. MareshH.W. ChenL. BagryanovaS. HorvathD. ChiaE. GaronL. GoodglickProgesterone and estrogen receptor expression and activity in human non-small cell lung cancerSteroids20117691092010.1016/j.steroids.2011.04.01521600232
  44. A.J. AlbergM.V. BrockJ.M. SametEpidemiology of lung cancer: Looking to the futureJ. Clin. Oncol.2005233175318510.1200/JCO.2005.10.46215886304
  45. D.R. Jacobs Jr.S. GottenborgSmoking and weight: The Minnesota Lipid Research ClinicAm. J. Public Health19817139139610.2105/AJPH.71.4.3917468879
  46. S.G. WannametheeA.G. ShaperM. WalkerWeight change, body weight and mortality: The impact of smoking and ill healthInt. J. Epidemiol.20013077778610.1093/ije/30.4.77711511602
  47. R.W. TaylorD. KeilE.J. GoldS.M. WilliamsA. GouldingBody mass index, waist girth, and waist-to-hip ratio as indexes of total and regional adiposity in women: Evaluation using receiver operating characteristic curvesAm. J. Clin. Nutr.19986744499440374
  48. J.O. HillS. SidneyC.E. LewisK. TolanA.L. ScherzingerE.R. StammRacial differences in amounts of visceral adipose tissue in young adults: The CARDIA (Coronary Artery Risk Development in Young Adults) studyAm. J. Clin. Nutr.19996938138710075320
  49. I. JanssenS.B. HeymsfieldD.B. AllisonD.P. KotlerR. RossBody mass index and waist circumference independently contribute to the prediction of nonabdominal, abdominal subcutaneous, and visceral fatAm. J. Clin. Nutr.20027568368811916754
  50. J. Neamat-AllahD. WaldA. HüsingB. TeucherA. WendtS. DelormeJ. DinkelM. ViglM.M. BergmannS. FellerValidation of anthropometric indices of adiposity against whole-body magnetic resonance imaging—A study within the German European Prospective Investigation into Cancer and Nutrition (EPIC) cohortsPLoS ONE20149e9158610.1371/journal.pone.009158624626110
  51. J.P. DesprésI. LemieuxJ. BergeronP. PibarotP. MathieuE. LaroseJ. Rodés-CabauO.F. BertrandP. PoirierAbdominal obesity and the metabolic syndrome: Contribution to global cardiometabolic riskArterioscler. Thromb. Vasc. Biol.2008281039104910.1161/ATVBAHA.107.15922818356555
  52. H. EyreR. KahnR.M. RobertsonN.G. ClarkC. DoyleY. HongT. GanslerT. GlynnR.A. SmithK. TaubertPreventing cancer, cardiovascular disease, and diabetes: A common agenda for the American Cancer Society, the American Diabetes Association, and the American Heart AssociationCirculation20041093244325510.1161/01.CIR.0000133321.00456.0015198946
The underlying source XML for this text is taken from The license for the article is Creative Commons Attribution. The main subject has been identified as obesity.