Obesity is a well-established risk factor for endometrial cancer in western countries. A meta-analysis of 25 prospective studies, including 23 from western countries, found a 50% increased risk of endometrial cancer per 5 BMI units (kg/m2) (WCRF/AICR Systematic Literature Review Continuous Update Project, 2012). However, evidence on the association between BMI and endometrial cancer risk among Asian populations remains limited, with only three prospective cohort studies (Kuriyama et al., 2005; Song et al., 2008; Liu et al., 2016) and five case–control studies (Shu et al., 1992; Okamura et al., 2006; Xu et al., 2006; Zhang et al., 2010; Hosono et al., 2011) having evaluated the association. Three of the five case–control studies used BMI at diagnosis (Okamura et al., 2006; Zhang et al., 2010; Hosono et al., 2011), which might have led to underestimation of the influence of obesity on endometrial cancer risk because of weight loss induced by the cancer itself (Theologides, 1977).
In addition, no previous prospective study in an Asian population has assessed the increase in the risk of endometrial cancer by its subtypes, which are categorized into type 1 and type 2 according to histological features (Bokhman, 1983; Sherman, 2000). Although type 1 cancer expresses estrogen receptors and arises from hyperplastic epithelium, which is induced by chronic estrogen stimulation, type 2 endometrial cancer does not show estrogen receptors (Murali et al., 2014). As expected from the estrogen-dependent mechanisms of the development of type 1 cancer (Murali et al., 2014), several studies from western countries have reported that BMI is associated with a greater risk of type 1 than type 2 endometrial cancer (Bjørge et al., 2007; McCullough et al., 2008; Amankwah et al., 2013; Setiawan et al., 2013; Trabert et al., 2015). Furthermore, the lower prevalence of exogenous female hormone use in Asia (Collaborative Group on Epidemiological Studies on Endometrial Cancer, 2015) might make easier to detect an association between BMI and endometrial cancer, including its subtypes, because hormone use might mask an association (Crosbie et al., 2010).
Height is also considered a potential risk factor for the development of endometrial cancer, but the evidence that adult attained height increases the risk of endometrial cancer is limited, but suggestive (WCRF/AICR Systematic Literature Review Continuous Update Project, 2012). Some studies have suggested associations (Green et al., 2011; Aune et al., 2015), whereas others have not (Bjørge et al., 2007; Sung et al., 2009). In Asian countries, only one prospective cohort study (Sung et al., 2009) has reported an association. Further evidence is required.
The aim of our study was to clarify the impact of BMI and height on the risk of endometrial cancer, overall and by histological subtype, among Japanese women using data from a prospective large-scale population-based study.
Participants and methods
The Japan Public Health Center-based Prospective Study (JPHC study) is an ongoing cohort study that was designed to investigate the association between nutritional, lifestyle risk factors and cancer, cardiovascular disease, and lifestyle-related disease. Details of the study have been described elsewhere (Tsugane and Sawada, 2014). In brief, the JPHC study is composed of two cohorts, with cohort I initiated in 1990 in patients aged 40–59 years and cohort II initiated in 1993, and including women aged 40–69 years. A total of 140 420 individuals (71 698 women and 68 722 men) were identified and recruited using data from the population registries of the municipalities of 11 public health center (PHC) areas at the beginning of each baseline survey. One PHC area was excluded because of a lack of cancer incidence data (4178 women). An additional 149 women were excluded because of non-Japanese nationality (n=20), late report of emigration occurring before the start of follow-up (n=122), or incorrect birth date or duplicate registration (n=7).
The study commenced with a baseline self-administered questionnaire survey on various health habits, including anthropometric factors, menstrual and reproductive history, personal medical histories, smoking history, and other lifestyle factors in 1990 for cohort I and in 1993–1994 for cohort II. Exogenous hormone users were determined by answers to the question ‘Have you ever taken female hormone drugs?’ in cohort I and by answers to the question ‘Have you ever taken hormone therapy for dysmenorrhea, contraception or for menopausal problems?’ in cohort II. Therefore, we cannot distinguish between oral contraceptive and hormone replacement therapy users. Among eligible participants, 55 837 (82.9%) women returned the questionnaire. In cohort I, weight change from 20 years of age was obtained as categorical data from answers to the question ‘have you had a weight gain/loss of more than 5 kg or within 5 kg’. In cohort II, the actual weight data of 20-year-olds was collected.
The study protocol was approved by the Institutional Review Board of the National Cancer Center, Tokyo, Japan.
Using the self-reported height and weight collected in the baseline survey, BMI was calculated as the weight divided by the square of the height (kg/m2). On evaluation of the validity of the self-reported BMI and height in a subsample of the cohort, Spearman’s correlation coefficients were 0.90 (Tsugane et al., 2002) and 0.91, respectively.
Identification of cancer cases
The end of follow-up for cancer incidence was 31 December 2012. Eligible women contributed person-years from the date of completion of the baseline questionnaire until the diagnosis of endometrial cancer, death from any cause, or end of follow-up, whichever came first. Changes in residential status and survival were obtained annually from the residential registry of each PHC. Information on the cause of death was confirmed by death certificates provided by the Ministry of Health, Labor and Welfare. Incidence of endometrial cancer was identified by voluntary reports from major local hospitals in the study areas, and data linkage with population-based cancer registries, with permission. Death certificates were used as supplementary information to identify cancer incidence. In our cancer registry system, information was obtained from death certificates only in 3.0% of cases during the study period. Cases were coded using International Classification of Diseases for Oncology, 3rd ed. (ICD-O-3) (World Health Organization, 2000) (C54.0-C54.9). Type 1 cases were defined as low-grade endometroid adenocarcinoma, adenocarcinoma not otherwise specified, mucinous, and adenocarcinoma with squamous differentiation (ICD-O-3 morphology codes of 8140, 8380, 8482, 8560, and 8570; n=119), and type 2 cases as serous, squamous, and clear cell pathology (ICD-O-3 morphology codes of 8070, 8260, 8310, 8441, and 8460; n=12) or grade 3 or worse endometrioid and adenocarcinoma not otherwise specified (total; n=9) on the basis of the classification used in previous studies (Bjørge et al., 2007; McCullough et al., 2008; Setiawan et al., 2013).
For the present analysis, we excluded women with a history of any cancer (n=1498) or missing information on height or weight information (n=593) from all analyses. Height was categorized into five groups (<148, 148–151, 152–155, 156–159, ≥160 cm). BMI was categorized into seven groups (<19.0, 19.0–20.9, 21.0–22.9, 23.0–24.9, 25.0–26.9, 27.0–29.9, ≥30.0 kg/m2) as used in our previous reports (Tsugane et al., 2002). In addition, we also carried out an analysis using the WHO BMI classification for Asians (<18.5, 18.5–22.9, 23.0–24.9, 25.0–27.4, 27.5–29, ≥30.0 kg/m2) (WHO Expert Consultation, 2004). Weight change from 20 years of age was categorized into three groups according to the answer in cohort I (less than −5, −4.9 to +4.9, more than +5.0 kg). In cohort II, it was calculated as gain in weight from 20 years of age to baseline, and categorized into three. The Cox proportional hazards model was used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) in each group compared with the reference BMI (23.0–24.9kg/m2) or height (<148 cm) group. SAS statistical software, version 9.3 (SAS Institute Inc., Cary, North Carolina, USA), was used for all analyses. Cox models were stratified according to PHC and adjusted for the following potentially cofounding variables: age (continuous), smoking (never or ever), alcohol consumption (none or more than once per week), coffee consumption (≤2 cups/week, 3–4 cups/week, or ≥1 cup/day), history of diabetes mellitus (none or yes), sports (none or at least once per week), use of exogenous female hormones (ever or never), age at menarche (<14, 14–14.9, 15–15.9, and ≥16), age at menopause (<48, 48–50, and ≥51), number of deliveries (nulliparous or ≥1), and menopausal status at baseline survey (premenopausal or postmenopausal). These potential cofounding factors were referenced from previous studies (WCRF/AICR Systematic Literature Review Continuous Update Project, 2012, Shimazu et al., 2008).
We observed 53 651 women for an average of 18.6 years. During the follow-up period, 180 newly diagnosed endometrial cancers were identified. Among the study participants, 3710 (6.9%) moved away and 175 (0.3%) were lost to follow-up during the study period. Histology and differentiation data were confirmed in 97.3 and 90.0% of cases, respectively. Women with type 1 endometrial cancer were slightly younger than those with type 2 endometrial cancer (mean 61.5 vs. 64.5 years old, respectively) and had a higher mean BMI (24.4 vs. 23.1 kg/m2).
Table 1 shows the characteristics of the study participants according to BMI and height categories. Compared with women with a BMI of 23.0–24.9 kg/m2, those in higher BMI categories were more likely to be older, nulliparous, and premenopausal, and have a history of diabetes, and less likely to drink alcohol or coffee. Participants with higher height tended to be younger, premenopausal, nulliparous, and smokers, and tended to drink more alcohol or coffee and to play sports regularly.
Table 2 presents the associations of BMI and height with total endometrial cancer by multivariate models with or without a history of diabetes mellitus. We found a significant positive risk increase in the category of BMI of at least 27.0 kg/m2. Compared with the reference BMI (23.0–24.9 kg/m2) group, the multivariate-adjusted HR in the highest group (≥30 kg/m2) was 2.37 (95% CI: 1.20–4.66, Supplementary Table 2, Supplemental Digital Content 1, http://links.lww.com/EJCP/A204). Similar results were obtained using the WHO BMI classification (Supplementary Table 1, Supplemental digital content 1, http://links.lww.com/EJCP/A204). To compare these results with those from a previous meta-analysis (Crosbie et al., 2010), we excluded participants with a BMI less than the reference category (BMI<23.0 kg/m2). The HR per 5-U increase in BMI was 1.63 (95% CI: 1.18–2.24). In contrast, no statistically significant association was observed between height and endometrial cancer risk (HR for per 5 cm increase in height: 1.11, 95% CI: 0.95–1.28).
Table 3 shows the association of each histological subtype of endometrial cancer and BMI. With each 5-U increase in BMI, the estimated HR of type 1 endometrial cancer was increased (HR=1.54, 95% CI: 1.21–1.98). In contrast, no risk elevation was observed with type 2 endometrial cancer (HR=0.90, 95% CI: 0.44–1.80). HRs (95% CIs) per 5 cm increase in height were 1.26 (0.88–1.79) for type 2 endometrial cancer and 1.01 (0.86–1.19) for type 1. Among premenopausal women (21 630 participants with 94 cases), women with a BMI of at least 27.0 kg/m2 did not have a statistically significant increase in risk compared with the reference group (HR=1.55, 95% CI: 0.81–2.98). In contrast, among postmenopausal women (31 029 participants with 83 cases), those with a BMI of at least 27.0 kg/m2 had an increased risk compared with the reference group (HR=2.75, 95% CI: 1.41–5.37). None of these results changed after exclusion of cases occurring within the first 3 years of baseline (n=16).
We also carried out a subgroup analysis according to the use of exogenous female hormones. A trend similar to that in the overall analysis was observed in never users of hormones (Supplementary Table 3, Supplemental digital content 1, http://links.lww.com/EJCP/A204). Among ever users of hormones, because only 19 events occurred, we could not carry out an analysis by the category of BMI and height (HR=0.80, 95% CI: 0.38–1.68, per 5 kg/m2 increase in BMI; HR=0.85, 95% CI: 0.52–1.39, per 5 cm increase in height, respectively). We also evaluated the risk of endometrial cancer and weight change from the age of 20 years. No risk elevation or reduction was observed in the group with a weight gain or loss of more than 5 kg compared with the group with a weight change of less than 5 kg (Supplementary Table 4, Supplemental digital content 1, http://links.lww.com/EJCP/A204). Among cohort II participants, in whom actual weight data were collected at the age of 20 years, after exclusion of participants who experienced weight loss, HR was 1.12 (95% CI: 0.78–1.61) per 5 kg weight gain.
In this large prospective cohort study in Japan, we found a positive association between BMI and the risk of endometrial cancer. In the analysis of endometrial cancer subtypes, BMI was associated with an increased risk of type 1. but not type 2 endometrial cancer. No association was observed between height and the incidence of endometrial cancer.
In our study, the HR of endometrial cancer incidence per 5-U increase in BMI was 1.63 (95% CI: 1.18–2.24). This is consistent with the findings of a meta-analysis that showed a slightly higher point estimate of pooled HR – 1.79 – from two studies in Asia than in western countries, which ranged from 1.57 to 1.59 (Crosbie et al., 2010). In contrast, Aune et al. (2015) reported no significant difference in subgroup analysis by geographic location (HR=1.48 in Europe, HR=1.47 in Asia). The multiethnic cohort study also reported a more pronounced association between BMI and endometrial cancer risk in Japanese Americans than in other non-Asian populations (Park et al., 2010). One potential reason for this difference in association might be the lower prevalence of exogenous female hormone use in Asia (Shimazu et al., 2008; Collaborative Group on Epidemiological Studies on Endometrial Cancer, 2015). Several epidemiological studies have reported a weak or a null association of BMI with endometrial cancer among hormone replacement therapy users, and positive associations among never users (Beral et al., 2005; Crosbie et al., 2010; Aune et al., 2015; Renehan et al., 2015). Thus, the greater relative risk with increasing BMI in our population is reasonable, given its lower prevalence of exogenous female hormone use. Another possible reason is that differences in fat distribution might enhance the association between BMI and endometrial cancer risk. Visceral adipose tissue is metabolically active (Lim et al., 2012), and secretes proinflammatory substances, adipokine, or insulin growth factor (Dossus et al., 2010), which could promote carcinogenesis. Considering that the accumulation of visceral adipose tissue is higher in East Asian than in White women at the same level of BMI (Nazare et al., 2012), the effect of increasing BMI on endometrial cancer risk might be enhanced. In our study, weight change from age 20 years was not associated with endometrial cancer risk. Among participants of cohort II, however, a slight but not statistically significant risk elevation per 5 kg weight gain was observed. This result was consistent with two previous meta-analyses (Aune et al., 2015; Keum et al., 2015), relative risk of 1.16–1.39 per 5 kg increase in weight.
To our knowledge, this is the first prospective study to assess the association between BMI and endometrial cancer by subtype in Asia. The proportion of type 2 endometrial cancer in our study was 12.2% (22/180), which is comparable to previous reports, and among other clinical features, the mean age of our type 2 group was well compatible with type 2 endometrial cancer examined elsewhere (Shu et al., 1992; Okamura et al., 2006; Xu et al., 2006; Zhang et al., 2010; Hosono et al., 2011). In contrast to recent studies that suggested a positive association between BMI and type 2 endometrial cancer (Bjørge et al., 2007; McCullough et al., 2008; Amankwah et al., 2013; Setiawan et al., 2013; Trabert et al., 2015), our results did not show an association between BMI and type 2 endometrial cancer risk. This discrepancy might be because of differences in the prevalence of severe obesity: the prevalence (BMI≥30.0 kg/m2) was only 3% in our study, versus more than 10% in previous studies (Bjørge et al., 2007; McCullough et al., 2008; Amankwah et al., 2013; Setiawan et al., 2013; Trabert et al., 2015). Most of these other studies showed a weaker association with type 2 than type 1 endometrial cancer, and the risk elevation was detectable in the group with BMI of at least 30 kg/m2 (Bjørge et al., 2007; McCullough et al., 2008; Setiawan et al., 2013; Trabert et al., 2015). Body fatness might induce carcinogenesis through three pathways: (i) estrogen-driven, (ii) insulin-driven or insulin growth factor-driven, and (iii) adipokine-driven systemic inflammation (Beral et al., 2005). No evidence has as yet shown that excess estrogen exposure is a risk factor for the development of type 2 endometrial cancer (Sherman, 2000), and the development of type 2 endometrial cancer is considered to be estrogen independent. In the development of type 1 cancer, these three pathways might all play a role, but in that of type 2 endometrial cancer, the latter two might be dominant. Thus, the development of type 2 endometrial cancer might require higher BMI levels.
Our study had several strengths. The study participants were selected from a general population, and response rates were high and loss to follow-up was low. Data were collected using our own research-based cancer registry, in addition to the registries maintained by the local governments. The completeness of information on endometrial cancer incidence with our cancer registry system is adequate for the present study. In addition, the study’s prospective design means that recall bias was avoided because information on anthropometric measures was collected before the diagnosis of endometrial cancer.
Several limitations of our study also warrant mention. First, we have a small number of type 2 cases (n=21); further prospective studies are needed to confirm the association of BMI with type 2 endometrial cancer in Asian populations. Second, we calculated BMI on the basis of self-reported height and weight. However, a previous study from our group showed that self-reported and health check-up values are highly correlated (Tsugane et al., 2002). Third, we could not exclude the possibility of residual or uncontrolled cofounding, including details of hormone therapy, although we adjusted for generally considered lifestyle and dietary factors.
We found that the risk of endometrial cancer was elevated in women with a BMI of 27.0 kg/m2 or more. On analysis by histological subtype, BMI was associated with the risk of type 1 endometrial cancer risk. but not with that of type 2 endometrial cancer among a population with a relatively low BMI compared with western populations.
Members of the Japan Public Health Center-based Prospective Study (JPHC Study; Principal Investigator, S. Tsugane) Group (as of April 2017) are listed at: http://epi.ncc.go.jp/en/jphc/781/7951.html.
The authors are indebted to the Aomori, Iwate, Ibaraki, Niigata, Osaka, Kochi, Nagasaki, and Okinawa Cancer Registries for providing their incidence data.
This study was supported by the National Cancer Center Research and Development Fund [23-A-31 (toku) and 26-A-2] (since 2011) and a grant-in-aid for Cancer Research from the Ministry of Health, Labour and Welfare of Japan (from 1989 to 2010).
Conflicts of interest
There are no conflicts of interest.
Amankwah EK, Friedenreich CM, Magliocco AM, Brant R, Courneya KS, Speidel T, et al (2013). Anthropometric measures and the risk of endometrial cancer, overall and by tumor microsatellite status and histological subtype
. Am J Epidemiol 177:1378–1387.
Aune D, Navarro Rosenblatt DA, Chan DS, Vingeliene S, Abar L, Vieira AR, et al (2015). Anthropometric factors and endometrial cancer risk: a systematic review and dose-response meta-analysis of prospective studies
. Ann Oncol 26:1635–1648.
Beral V, Bull D, Reeves G, Collaborators MWS (2005). Endometrial cancer and hormone-replacement therapy in the Million Women Study. Lancet 365:1543–1551.
Bjørge T, Engeland A, Tretli S, Weiderpass E (2007). Body size in relation to cancer of the uterine corpus in 1 million Norwegian women. Int J Cancer 120:378–383.
Bokhman JV (1983). Two pathogenetic types of endometrial carcinoma. Gynecol Oncol 15:10–17.
Collaborative Group on Epidemiological Studies on Endometrial Cancer (2015). Endometrial cancer and oral contraceptives: an individual participant meta-analysis of 27 276 women with endometrial cancer from 36 epidemiological studies. Lancet Oncol 16:1061–1070.
Crosbie EJ, Zwahlen M, Kitchener HC, Egger M, Renehan AG (2010). Body mass index, hormone replacement therapy, and endometrial cancer risk: a meta-analysis. Cancer Epidemiol Biomarkers Prev 19:3119–3130.
Dossus L, Rinaldi S, Becker S, Lukanova A, Tjonneland A, Olsen A, et al (2010). Obesity, inflammatory markers, and endometrial cancer risk: a prospective case-control study. Endocr Relat Cancer 17:1007–1019.
Green J, Cairns BJ, Casabonne D, Wright FL, Reeves G, Beral V, et al (2011). Height
and cancer incidence in the Million Women Study: prospective cohort, and meta-analysis of prospective studies
and total cancer risk. Lancet Oncol 12:785–794.
Hosono S, Matsuo K, Hirose K, Ito H, Suzuki T, Kawase T, et al (2011). Weight gain during adulthood and body weight at age 20 are associated with the risk of endometrial cancer in Japanese women. J Epidemiol 21:466–473.
Keum N, Greenwood DC, Lee DH, Kim R, Aune D, Ju W, et al (2015). Adult weight gain and adiposity-related cancers: a dose–response meta-analysis of prospective observational studies. J Natl Cancer Inst 107.
Kuriyama S, Tsubono Y, Hozawa A, Shimazu T, Suzuki Y, Koizumi Y, et al (2005). Obesity and risk of cancer in Japan. Int J Cancer 113:148–157.
Lim U, Turner SD, Franke AA, Cooney RV, Wilkens LR, Ernst T, et al (2012). Predicting total, abdominal, visceral and hepatic adiposity with circulating biomarkers in Caucasian and Japanese American women. PLoS ONE 7:e43502.
Liu Y, Warren Andersen S, Wen W, Gao YT, Lan Q, Rothman N, et al (2016). Prospective cohort study of general and central obesity, weight change trajectory and risk of major cancers among Chinese women. Int J Cancer 139:1461–1470.
McCullough ML, Patel AV, Patel R, Rodriguez C, Feigelson HS, Bandera EV, et al (2008). Body mass and endometrial cancer risk by hormone replacement therapy and cancer subtype. Cancer Epidemiol Biomarkers Prev 17:73–79.
Murali R, Soslow RA, Weigelt B (2014). Classification of endometrial carcinoma: more than two types. Lancet Oncol 15:e268–e278.
Nazare JA, Smith JD, Borel AL, Haffner SM, Balkau B, Ross R, et al (2012). Ethnic influences on the relations between abdominal subcutaneous and visceral adiposity, liver fat, and cardiometabolic risk profile: the International Study of Prediction of Intra-Abdominal Adiposity and Its Relationship With Cardiometabolic Risk/Intra-Abdominal Adiposity. Am J Clin Nutr 96:714–726.
Okamura C, Tsubono Y, Ito K, Niikura H, Takano T, Nagase S, et al (2006). Lactation and risk of endometrial cancer in Japan: a case–control study. Tohoku J Exp Med 208:109–115.
Park SL, Goodman MT, Zhang ZF, Kolonel LN, Henderson BE, Setiawan VW (2010). Body size, adult BMI
gain and endometrial cancer risk: the multiethnic cohort. Int J Cancer 126:490–499.
Renehan AG, Zwahlen M, Egger M (2015). Adiposity and cancer risk: new mechanistic insights from epidemiology. Nat Rev Cancer 15:484–498.
Setiawan VW, Yang HP, Pike MC, McCann SE, Yu H, Xiang YB, et al (2013). Type I and II endometrial cancers: have they different risk factors? J Clin Oncol 31:2607–2618.
Sherman ME (2000). Theories of endometrial carcinogenesis: a multidisciplinary approach. Mod Pathol 13:295–308.
Shimazu T, Inoue M, Sasazuki S, Iwasaki M, Kurahashi N, Yamaji T, et al (2008). Coffee consumption and risk of endometrial cancer: a prospective study in Japan. Int J Cancer 123:2406–2410.
Shu XO, Brinton LA, Zheng W, Swanson CA, Hatch MC, Gao YT, et al (1992). Relation of obesity and body fat distribution to endometrial cancer in Shanghai, China. Cancer Res 52:3865–3870.
Song YM, Sung J, Ha M (2008). Obesity and risk of cancer in postmenopausal Korean women. J Clin Oncol 26:3395–3402.
Sung J, Song YM, Lawlor DA, Smith GD, Ebrahim S (2009). Height
and site-specific cancer risk: a cohort study of a korean adult population. Am J Epidemiol 170:53–64.
Theologides A (1977). Weight loss in cancer patients. CA Cancer J Clin 27:205–208.
Trabert B, Wentzensen N, Felix AS, Yang HP, Sherman ME, Brinton LA (2015). Metabolic syndrome and risk of endometrial cancer in the United States: a study in the SEER-medicare linked database. Cancer Epidemiol Biomarkers Prev 24:261–267.
Tsugane S, Sawada N (2014). The JPHC study: design and some findings on the typical Japanese diet. Jpn J Clin Oncol 44:777–782.
Tsugane S, Sasaki S, Tsubono Y (2002). Under- and overweight impact on mortality among middle-aged Japanese men and women: a 10-y follow-up of JPHC study cohort I. Int J Obes Relat Metab Disord 26:529–537.
WHO Expert Consultation (2004). Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet 363:157–163.
World Health Organization (2000). International classification of diseases for oncology, 3rd ed. Geneva, Switzerland: WHO.
Xu WH, Xiang YB, Zheng W, Zhang X, Ruan ZX, Cheng JR, et al (2006). Weight history and risk of endometrial cancer among Chinese women. Int J Epidemiol 35:159–166.
Zhang Y, Liu Z, Yu X, Zhang X, Lü S, Chen X, et al (2010). The association between metabolic abnormality and endometrial cancer: a large case–control study in China. Gynecol Oncol 117:41–46.