• anthropometry;
  • ovarian cancer;
  • etiology;
  • obesity;
  • menopausal status


  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We examined the associations of measured anthropometric factors, including general and central adiposity and height, with ovarian cancer risk. We also investigated these associations by menopausal status and for specific histological subtypes. Among 226,798 women in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort, there were 611 incident cases of primary, malignant, epithelial ovarian cancer diagnosed during a mean 8.9 years of follow-up. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs), adjusted for potential confounders. Compared to women with body mass index (BMI) < 25 kg/m2, obesity (BMI ≥ 30 kg/m2) was associated with excess ovarian cancer risk for all women combined (HR = 1.33, 95% CI = 1.05–1.68; ptrend = 0.02) and postmenopausal women (HR = 1.59, 95% CI = 1.20–2.10; ptrend = 0.001), but the association was weaker for premenopausal women (HR = 1.16, 95% CI = 0.65–2.06; ptrend = 0.65). Neither height or weight gain, nor BMI-adjusted measures of fat distribution assessed by waist circumference, waist–hip ratio (WHR) or hip circumference were associated with overall risk. WHR was related to increased risk of mucinous tumors (BMI-adjusted HR per 0.05 unit increment = 1.17, 95% CI = 1.00–1.38). For all women combined, no other significant associations with risk were observed for specific histological subtypes. This large, prospective study provides evidence that obesity is an important modifiable risk factor for epithelial ovarian cancer, particularly among postmenopausal women.

Ovarian cancer is the 5th most common incident cancer and cause of cancer deaths in women in western Europe.1 The prognosis upon diagnosis is relatively poor because clinical presentation of the disease often occurs at an advanced stage.2 Thus, it is important to identify potentially modifiable risk factors for the prevention of ovarian cancer.

An association of adiposity with ovarian cancer risk is plausible, given the postulated hormonal etiology of the disease3–5; however, the evidence for an association remains unclear, particularly with regard to menopausal status and specific histological subtypes of the tumor.6 A recent metaanalysis found that obesity [body mass index (BMI) ≥ 30 kg/m2] was associated with a statistically significant 30% increased risk of ovarian cancer when compared with normal weight (BMI < 25 kg/m2), although this was stronger for case–control studies (odds ratio = 1.5) than for cohort studies (odds ratio = 1.1).6 Findings from a pooled analysis of 12 cohort studies indicated that obesity was associated with an increased risk of ovarian cancer in premenopausal women [relative risk (RR) = 1.72] but not postmenopausal women (RR = 1.07).7 It is also unclear whether or not obesity in early adulthood or adult weight change is independently associated with risk.6, 7 It is possible that associations of body measures with ovarian cancer might also differ according to the specific histological subtype, because each subtype has distinct clinical and morphological features and potentially different etiologies.8, 9 However, there have been inconsistent findings for specific histological subtypes in previous epidemiological studies.6 In addition, few studies have evaluated the etiological role of other body measures such as height or measures of central adiposity, and these associations might also differ by menopausal status.7 Abdominal fat is thought to be more strongly associated with metabolic abnormalities than other fat distribution patterns,10, 11 and so could potentially have a different role to general adiposity in influencing cancer risk.12

We conducted an analysis in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort study to examine the associations of measured anthropometric factors, including general and central adiposity and height, with ovarian cancer risk. We also sought to clarify these associations by menopausal status and different histological subtypes.

Material and Methods

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Study cohort

EPIC is a multicenter, prospective cohort study designed primarily to investigate the associations between dietary and lifestyle factors and cancer risk. The design, study population, baseline data collection and follow-up methods have been previously described in detail.13, 14 In brief, standardized questionnaire data on dietary and lifestyle factors were collected from ∼370,000 women and 150,000 men, enrolled between 1992 and 2000 in 23 centers throughout 10 western European countries (Denmark, France, Germany, Greece, Italy, Norway, Spain, Sweden, The Netherlands and United Kingdom).13 Participants were mainly between 35 and 70 years of age at enrolment and were recruited from the general population residing within defined geographic areas (i.e., town or province), with some exceptions: women who were members of a health insurance plan for state school employees (France), women attending breast cancer screening (Utrecht, The Netherlands), blood donors (some centers in Italy and Spain), and predominantly vegetarians (Oxford “health conscious” cohort). Approval for this study was obtained from the ethical review boards of the International Agency for Research on Cancer and from all local recruiting institutions. All participants provided written informed consent.

For this analysis, the initial analytical cohort consisted of 335,581 women who, at baseline, did not have prevalent cancer at any site and had not undergone bilateral ovariectomy. We further a priori excluded women with missing dietary or lifestyle questionnaire data and those who were in the top or bottom 1% of the distribution of the ratio of energy intake to estimated energy requirement15 to reduce the impact on the analysis of implausible extreme values. In addition, we excluded women without baseline measurements of height, weight, waist and hip values taken by trained observers at study centers; this included all women from the Norwegian cohort, Umeå (Sweden) and about 70% of French participants. Women from the Oxford health conscious cohort had self-reported measurements but were included in the analysis because we were able to correct the measurements for reporting error using age- and sex-specific linear regression model prediction equations. The prediction equations were derived from 965 men and 1464 women aged 50–64 years old from the health conscious cohort, with both measured and self-reported anthropometric measures determined within 3 weeks; details have been previously published.16–18 The final analytical cohort for this study therefore consisted of 226,798 women from 9 countries. The analyses of weight change were limited to a subcohort of 118,493 women from centers that collected data on recalled weight in early adulthood: Denmark, Germany (Potsdam), Greece, Italy (Varese), Sweden (Malmö) and the United Kingdom.

Anthropometric measurements and other baseline exposure variables

Details on the standardized procedures for taking anthropometric measurements in the EPIC study centers have been previously described in detail.16 Briefly, weight was measured to the nearest 0.1 kg and height was measured to the nearest 0.1, 0.5 or 1.0 cm depending on the study center, in subjects wearing no shoes. Waist circumference was measured either at the narrowest torso circumference (most centers) or at the midpoint between the lower ribs and iliac crest. Hip circumference was measured at the widest circumference or over the buttocks. Weight, waist and hip measurements were corrected to account for protocol differences between centers in clothing worn by participants during body measurements.16 The waist and hip circumferences of each participant were used to construct waist–hip ratio (WHR; cm/cm) as an additional measure of fat distribution. Adult weight change was estimated as the difference between measured weight at enrolment and recalled weight at age 20 (25 years in 1 center).

Diet over the previous 12 months was assessed using country-specific, validated dietary assessment instruments.13, 19 Data on lifestyle, health and sociodemographic characteristics were collected via standardized questionnaires, which included menstrual and reproductive history, use of oral contraceptives (OCs) and postmenopausal hormone replacement therapy (HRT), medical history, lifetime history of tobacco smoking and alcohol consumption, physical activity, brief occupational history and level of education. Menopausal status at enrolment was defined as follows: women were “premenopausal” if they reported having had regular menses over the past 12 months; “postmenopausal” if they reported not having had any menses over the past 12 months; and “perimenopausal/unknown” if they reported irregular menses over the past 12 months (1–9 cycles) or if they indicated having had menses over the past 12 months but were no longer menstruating at the time of recruitment. Women with incomplete or missing questionnaire data, or who reported having had a hysterectomy or who were currently using exogenous hormones were classified as premenopausal if they were less than 46 years of age, perimenopausal/unknown if they were between 46 and 55 years of age, and postmenopausal if they were older than 55 years. For this analysis, presentation of results by menopausal status excluded women classified as perimenopausal or unknown (n = 35,8567 including 85 cases).

Follow-up for cancer incidence and vital status

Incident cases of ovarian cancer were identified through population-based cancer registries, except in France, Germany and Greece, where a combination of methods, including health insurance records, cancer and pathology registries, and active follow-up through study subjects and their next-of-kin was used. Data on vital status in most EPIC study centers were collected from mortality registries at the regional or national level, in combination with data collected by active follow-up (Greece). Vital status was known for 98.4% of all EPIC participants as of March 2007. Women were followed from the date of enrolment until ovarian cancer diagnosis, death, emigration or end of the follow-up period. The closure date for this follow-up period was the date of the last complete follow-up for both cancer incidence and vital status, which varied between 2004 and 2006 for most centers.

After excluding women diagnosed with nonepithelial, metastatic or in situ ovarian tumors (69 cases), there were 611 incident cases of primary, malignant, epithelial ovarian tumors diagnosed during this follow-up period and meeting the eligibility criteria for this analysis. Of these cases, 275 were serous tumors, 57 mucinous, 59 endometrioid, 26 clear cell, 4 undifferentiated and 176 were not otherwise specified or could not be classified; histology information was missing for 14 cases.20

Statistical analyses

We evaluated the association between anthropometric characteristics and risk of ovarian cancer using Cox proportional hazards models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). Age was used as the underlying time variable, with entry and exit time defined as the subject's age at recruitment and age at ovarian cancer diagnosis or censoring (death, lost to follow-up and end of follow-up), respectively. Models were stratified by study center and age at recruitment, and adjusted for potential confounders including parity (nulliparous, parous), age at menarche (<11, 12–14, ≥15, unknown), menopausal status (premenopausal, perimenopausal, postmenopausal), smoking status (current, former, never), education (none/primary school, secondary/technical/university, unknown), unilateral ovariectomy (yes, no, unknown), ever use of OCs (yes, no, unknown), and use of HRT at baseline (yes, no, unknown). In addition, we examined the effect of mutual adjustment of body measures to determine whether central or general obesity were independent determinants of risk. We also considered adjustment for energy intake and physical activity but did not retain them in the final models, because adjustment for these factors did not appreciably change the risk estimates. Trend tests were estimated on integer scores applied to the categories or quartiles and entered as a continuous term in the regression models.

Body measures were categorized into quartiles with cutoff points based on the overall distribution for all centers combined. BMI was also categorized into internationally standardized categories of <25 kg/m2 (normal weight), 25 to <30 kg/m2 (overweight), ≥30 kg/m2 (obese).21 We also analyzed anthropometric measures as continuous variables to estimate the HR per unit difference.

We investigated the associations between anthropometric factors and specific histological subtypes. We also examined whether the associations between anthropometric factors and ovarian cancer risk differed according to menopausal status, parity, ever use of OCs, baseline HRT use, country and follow-up time by adding subgroup interaction terms to the continuous multivariate models. All analyses were performed using SAS version 9.1.3 (SAS Institute, Cary, NC), and all statistical tests were two sided.


  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

During an average 8.9 (SD ± 1.8) years of follow-up (2,016,263 person years), there were 611 cases of primary invasive epithelial ovarian cancer diagnosed in this cohort of 226,798 women with a mean age at baseline of 52 years (Table 1). The prevalence of overweight was 31.4% and ranged from 19.5% in the United Kingdom health conscious cohort to 42.1% in Spain. The prevalence of obesity was 15.0% and ranged from 5.5% in France to 35.8% in Greece. Table 2 shows the mean values for the anthropometric measures and the distribution of covariates for the women who developed ovarian cancer during follow-up (cases) and those who did not develop ovarian cancer (noncases). Except for height, the mean anthropometric measures appeared higher in cases compared with noncases.

Table 1. EPIC cohort characteristics for the analyses of anthropometry and epithelial ovarian cancer
inline image
Table 2. Anthropometric measures and covariates at baseline for women in the EPIC cohort by ovarian case status after a mean of 8.9 years of follow-up
inline image

Higher BMI was associated with increased risk of ovarian cancer for all women combined and for postmenopausal women (Table 3). For all women combined, obesity (BMI ≥ 30 kg/m2) was associated with a 33% higher multivariate risk (95% CI = 1.05–1.68; ptrend = 0.02) compared with women of normal weight, and for postmenopausal women, the HR was 1.59 (95% CI = 1.20–2.10; ptrend = 0.001). The association was weaker for premenopausal women (HR = 1.16, 95% CI = 0.65–2.06; ptrend = 0.65). All tests of interaction by menopausal status were not statistically significant (p > 0.05).

Table 3. Multivariate-adjusted hazard ratio estimates of epithelial ovarian cancer according to anthropometric characteristics, for all women and by menopausal status
inline image

The risk estimates for weight were similar to those for BMI quartiles. After adjustment for height, the BMI and weight estimates were slightly attenuated for premenopausal women only (data not shown). Height was not associated with risk of ovarian cancer. Weight gain during adulthood after adjusting for weight at age 20 was not associated with ovarian cancer risk overall. For premenopausal women, the association with weight gain was statistically significant for those who had gained 10–15 kg compared with women who stayed within ±5 kg of their weight (HR = 2.41, 95% CI = 1.19–4.87) but the trend was not statistically significant (ptrend = 0.15).

For measures of fat distribution, including waist and hip circumferences and WHR, a statistically significant association for all women combined was noted only for hip circumference, for which the risk for the highest versus lowest quartile was 1.33 (95% CI = 1.04–1.70; ptrend = 0.02) (Table 3). There was a weak positive association between both waist and hip circumferences and ovarian cancer risk for postmenopausal women. After additional adjustment for BMI, these associations were attenuated, particularly for postmenopausal women, and were no longer statistically significant (Table 4).

Table 4. Multivariate-adjusted hazard ratio estimates of epithelial ovarian cancer according to measures of central adiposity after adjustment for BMI and other covariates, for all women and by menopausal status
inline image

Table 5 shows the associations between anthropometric factors as continuous variables and ovarian cancer risk for all women for different histological subtypes. There was a positive association between WHR and risk of mucinous tumors (HR per 0.05 unit increment = 1.19, 95% CI = 1.02–1.38), which was slightly attenuated after adjustment for BMI (HR = 1.17, 95% CI = 1.00–1.38). No other statistically significant associations were observed for all women combined, including the examination of serous invasive (n = 255) and serous borderline tumors (n = 20) separately (data not shown). When examining these associations by menopausal status (data not shown) unadjusted for BMI, we observed positive associations between height and risk of serous invasive tumors (p = 0.05) and between WHR and risk of serous borderline tumors (p = 0.003) in premenopausal women. For postmenopausal women, there was a positive association between hip circumference and risk of serous invasive tumors (p = 0.03).

Table 5. Multivariate-adjusted1 hazard ratio estimates of epithelial ovarian cancer according to anthropometric characteristics for all women2, by histological subtype
inline image

For all the body measures shown, there was minimal change to the risk estimates between the age- and center-adjusted models and multivariate models or after excluding the 34 borderline tumors (data not shown). There was no statistically significant interaction between body measures and overall follow-up time as continuous variables. However, when we stratified the HR estimates by <2 years (132 cases) and ≥2 years (479 cases) follow-up, the associations between adiposity and ovarian cancer risk appeared weaker among women diagnosed with ovarian cancer during the first 2 years of follow-up; for BMI, the HR per 2 kg/m2 increment was 0.97 (95% CI = 0.90–1.06) for women diagnosed during the first 2 years of follow-up and 1.06 (95% CI = 1.02–1.11; pheterogeneity = 0.06) for women diagnosed 2 or more years after enrolment. A similar proportion of premenopausal and postmenopausal cases were diagnosed during the first 2 years of follow-up (24 and 22%, respectively), and the pattern of results was consistent for both groups of women.

There was no evidence of effect modification by use of HRT at baseline, ever use of OCs, parity or country of recruitment. For postmenopausal women who were current users of HRT, the HR per 2 kg/m2 increment in BMI was 1.08 (95% CI = 0.98–1.19) and for nonusers of HRT was 1.05 (95% CI = 1.00–1.11). For women who had ever-used OCs, the HR per 2 kg/m2 increment in BMI was 1.06 (95% CI =1.00–1.12) and for never users of OCs was 1.03 (95% CI = 0.98–1.08).


  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

In the largest, prospective cohort study to date, to comprehensively examine associations of anthropometry with ovarian cancer risk, we found modest, positive associations between the degree of adiposity and risk. The associations were generally stronger for measures of general adiposity (BMI, weight) than for measures of fat distribution (waist and hip circumferences, WHR) and appeared stronger for postmenopausal women than for premenopausal women.

A major strength of our study is the standardized body measurements that were taken by study personnel for all participants. In contrast, most previous studies have used self-reported anthropometric measures.6 We did, however, rely on recalled weight at age 20 to estimate weight gain during adulthood. Although recall of past adult weight has been shown to be reliable, its accuracy may be poorer for certain subgroups, e.g., those currently overweight, which could have introduced some differential error for this body measure.22, 23 Other strengths of this study include the large, heterogeneous cohort that maximizes differences in the exposure measurements and the prospective design. Prospective study designs are important for minimizing recall bias and reverse-causation bias. However, we still found some evidence of reverse-causation bias in this study, because the risk estimates were lower for women diagnosed early in follow-up compared with those diagnosed after 2 or more years. This finding may be due to preclinical, tumor-related weight loss that had already occurred before the baseline assessment24 and could have slightly attenuated the risk estimates for the associations with adiposity. Alternatively, the difference by follow-up might be due to anthropometric measurements taken at a time that is less etiologically relevant. Another possibility is that stronger risks observed later in follow-up might be considered as strengthening the causality of the association due to a potentially longer exposure period.

A limitation of the study is the lack of updated information during follow-up on anthropometric characteristics and other factors that might influence these associations, such as menopausal status and use of exogenous hormones. Thus, changes that might have occurred during the mean 9 years of follow-up were unable to be accounted for in the analyses and may have attenuated the risk estimates. Premenopausal women composed 37% of the analytical cohort at baseline, and some of these women would have been perimenopausal or postmenopausal at the end of the follow-up period for this analysis. Strong residual confounding due to incompletely measured or other unmeasured risk factors (such as family history) is unlikely because of the minimal changes to risk estimates after adjustment for numerous important potential confounders in our and other cohort studies.7

Compared with women of normal weight, obese women had a 33% increased risk of ovarian cancer and overweight women had a 14% nonstatistically significant increased risk. Our results for BMI are very similar to the overall pooled effect estimates of 1.30 for obesity and 1.16 for overweight from a recent metaanalysis.6 Their pooled estimates were weaker for cohort studies than for case–control studies, but sensitivity analyses showed that studies in which BMI was measured 25 years or more prior to diagnosis had lower risk estimates; after omitting these studies, the pooled RR for cohort studies changed from 1.1 to 1.3.6 Differences in results for different study designs might also be due to recall bias in case–control studies or to whether self-reported or measured anthropometric values were used. A different metaanalysis25 of 13 prospective observational studies reported a RR of 1.03 (95% CI = 0.99–1.08) associated with a 5 kg/m2 increment in BMI.

In our study, the associations of overweight and obesity with ovarian cancer risk appeared stronger for postmenopausal women than for premenopausal women, although differences by menopausal status may be due to chance, because the tests for interaction were not statistically significant. The finding of stronger associations for postmenopausal women contrasts with most other studies (summarized in previous reviews6, 7), which have usually reported positive associations for premenopausal women and weaker or null associations for postmenopausal women. Schouten et al.7 pooled data from 12 cohort studies and found that obesity was associated with an increased risk of ovarian cancer in premenopausal women (RR = 1.72) but not postmenopausal women (RR = 1.07) when compared to women with a BMI less than 23 kg/m2; however, their test for trend across categories of BMI for premenopausal women lacked statistical significance (ptrend = 0.13) and the test for interaction by menopausal status was borderline significant.

In our cohort, weight gain during adulthood was not significantly associated with overall risk after accounting for weight at age 20, which is consistent with the pooled cohort analysis7 but not with another cohort study that found increased risk associated with prospectively measured weight change over a 7-year period.26 For premenopausal women, we observed an increased risk for women who had gained 10–15 kg but not more than 15 kg, although there were few women in the latter group. In the metaanalysis,6 obesity in early adulthood was associated with a statistically significant 22% increased risk, although this estimate was unadjusted for BMI in later adulthood. There are conflicting findings as to whether recent27 or early adulthood28, 29 body weight may be more etiologically relevant to ovarian tumorigenesis.

Few studies have examined how different measures of fat distribution are related to ovarian cancer.29–31 In the multivariate models unadjusted for BMI, hip circumference was positively associated with overall risk, but there was no association with waist circumference or WHR, except for a weak positive association of waist circumference with ovarian cancer in postmenopausal women. These associations were no longer present after adjustment for BMI, suggesting that these measures of fat distribution are not independently related to ovarian cancer risk. Similarly, another cohort study found no association with waist circumference30; however, 2 studies have reported significant but nonlinear positive associations of WHR with risk independent of BMI.30, 31 Slight differences in the measurement protocols for waist and hip circumferences between EPIC centers could have attenuated our results for fat distribution measures, although stratification of the Cox models by study center should have minimized this effect.32 Interindividual variation is also more likely to be a much more important source of variation than any potential systematic error caused by different measurement protocols.16

The biological mechanisms mediating an association between adiposity and ovarian cancer risk are thought to involve elevations in levels of androgens and oestrogens, decreases in progesterone levels, and modulation of the insulin-like growth factor (IGF) axis, resulting in proliferative and antiapoptotic effects on ovarian epithelial tissue.3–5 In addition to the well-documented associations of adiposity with circulating sex steroid hormone levels in premenopausal and postmenopausal women,3, 33–37 excess weight also increases insulin levels, which can upregulate ovarian androgen synthesis and could also increase IGF-I bioactivity, resulting in direct effects on ovarian epithelial cells or indirect effects on the synthesis and bioavailability of sex-steroid hormones.3, 36, 38, 39 However, the epidemiological evidence to date for a role of endogenous hormones and the IGF axis in ovarian cancer development is limited,3, 5 and the biological mechanisms may differ considerably with menopausal status.3, 35, 38–40

In our study, height was not associated with overall risk. There was some evidence of a positive association among premenopausal women, although this did not reach statistical significance. A recent cohort study by Lacey et al.41 found no association of height with risk among predominantly postmenopausal women (RR = 0.90). However, most previous studies have shown that height is associated with a modest increased risk of ovarian cancer.42 In the pooled cohort analysis,7 height was positively associated with risk for both pre- and postmenopausal women, but the association was stronger for premenopausal women. The categorical estimates were strongest and statistically significant for those at least 170 cm tall,7 but in our study, the highest quartile cutoff point was 166.2 cm; using similar cutoff points, we found that the overall RR for women ≥170cm was 1.19 (95% CI = 0.90–1.58) and the trend remained nonsignificant.

It is hypothesized that different histological subtypes of ovarian cancer have different etiologies8, 43; however, we found no clear associations among all women, except for a minimally increased risk of mucinous tumors with increasing WHR. Our ability to examine associations for specific histological subtypes was limited because of few cases for some of the subtypes. Adiposity did not seem to be related to serous tumors, which account for ∼50% of malignant epithelial ovarian tumors.9 Overall, previous findings by histological subtype have been inconsistent across individual studies.6 A pooled analysis of 10 case–control studies8 from the United States found a nonsignificant elevated risk of endometrioid and mucinous tumors associated with elevated BMI, which is consistent with our results, and an inverse association for serous tumors. Conversely, the pooled cohort analysis7 found no association between BMI and risk of endometrioid, mucinous or serous tumors. Other studies have reported a positive association of obesity with serous borderline tumors.44, 45 Few studies have examined clear cell tumors separately because they represent only about 5% of malignant ovarian tumors9; some studies found no association with BMI,6, 8 whereas others did report a significantly increased risk.44, 46, 47 Some studies have reported a positive association of height with risk of serous tumors,7, 29, 45 which is consistent with the findings in premenopausal women in this study, and of height with endometrioid tumors.7

Although there is some evidence that use of exogenous hormones may modify associations of adiposity with risk of hormone-related cancers including ovarian,48 breast,49 endometrial50 and colon,12, 51 in this study, we did not find evidence of stronger associations for nonusers of HRT or OCs.

In conclusion, this large European prospective cohort study provides evidence that obesity, particularly excess relative weight (BMI), is an important modifiable risk factor for epithelial ovarian cancer, especially among postmenopausal women. Despite the large size of our study, more studies are needed to clarify the associations of different anthropometric measures for specific histological subtypes.


  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We thank Mr. Wolfgang Bernigau (Potsdam, Germany) for statistical support. AEC is supported by a NHMRC Public Health Fellowship #520018. CMF was supported by a Health Senior Scholar Award from the Alberta Heritage Foundation for Medical Research.


  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • 1
    Ferlay J, Bray F, Pisani P, Parkin DM. GLOBOCAN 2002: cancer incidence, mortality and prevalence worldwide IARC cancerbase No. 5, version 2.0. Available: Lyon: IARC Press, 2004.
  • 2
    Cannistra SA. Cancer of the ovary. N Engl J Med 2004; 351: 251929.
  • 3
    Lukanova A, Kaaks R. Endogenous hormones and ovarian cancer: epidemiology and current hypotheses. Cancer Epidemiol Biomarkers Prev 2005; 14: 98107.
  • 4
    Risch HA. Hormonal etiology of epithelial ovarian cancer, with a hypothesis concerning the role of androgens and progesterone. J Natl Cancer Inst 1998; 90: 177486.
  • 5
    Olsen CM, Green AC, Nagle CM, Jordan SJ, Whiteman DC, Bain CJ, Webb PM. Epithelial ovarian cancer: testing the ‘androgens hypothesis.’ Endocr Relat Cancer 2008; 15: 10618.
  • 6
    Olsen CM, Green AC, Whiteman DC, Sadeghi S, Kolahdooz F, Webb PM. Obesity and the risk of epithelial ovarian cancer: a systematic review and meta-analysis. Eur J Cancer 2007; 43: 690709.
  • 7
    Schouten LJ, Rivera C, Hunter DJ, Spiegelman D, Adami HO, Arslan A, Beeson WL, van den Brandt PA, Buring JE, Folsom AR, Fraser GE, Freudenheim JL, et al. Height, body mass index, and ovarian cancer: a pooled analysis of 12 cohort studies. Cancer Epidemiol Biomarkers Prev 2008; 17: 90212.
  • 8
    Kurian AW, Balise RR, McGuire V, Whittemore AS. Histologic types of epithelial ovarian cancer: have they different risk factors? Gynecol Oncol 2005; 96: 52030.
  • 9
    Chen VW, Ruiz B, Killeen JL, Cote TR, Wu XC, Correa CN. Pathology and classification of ovarian tumors. Cancer 2003; 97: 263142.
  • 10
    Krotkiewski M, Bjorntorp P, Sjostrom L, Smith U. Impact of obesity on metabolism in men and women. Importance of regional adipose tissue distribution. J Clin Invest 1983; 72: 115062.
  • 11
    Garaulet M, Perez-Llamas F, Baraza JC, Garcia-Prieto MD, Fardy PS, Tebar FJ, Zamora S. Body fat distribution in pre-and post-menopausal women: metabolic and anthropometric variables. J Nutr Health Aging 2002; 6: 1236.
  • 12
    Pischon T, Lahmann PH, Boeing H, Friedenreich C, Norat T, Tjonneland A, Halkjaer J, Overvad K, Clavel-Chapelon F, Boutron-Ruault MC, Guernec G, Bergmann MM, et al. Body size and risk of colon and rectal cancer in the European Prospective Investigation Into Cancer and Nutrition (EPIC). J Natl Cancer Inst 2006; 98: 92031.
  • 13
    Riboli E, Hunt KJ, Slimani N, Ferrari P, Norat T, Fahey M, Charrondiere UR, Hemon B, Casagrande C, Vignat J, Overvad K, Tjonneland A, et al. European Prospective Investigation into Cancer and Nutrition (EPIC): study populations and data collection. Public Health Nutr 2002; 5: 111324.
  • 14
    Riboli E, Kaaks R. The EPIC Project: rationale and study design. European Prospective Investigation into Cancer and Nutrition. Int J Epidemiol 1997; 26 ( Suppl 1): S614.
  • 15
    Ferrari P, Slimani N, Ciampi A, Trichopoulou A, Naska A, Lauria C, Veglia F, Bueno-de-Mesquita HB, Ocke MC, Brustad M, Braaten T, Jose Tormo M, et al. Evaluation of under- and overreporting of energy intake in the 24-hour diet recalls in the European Prospective Investigation into Cancer and Nutrition (EPIC). Public Health Nutr 2002; 5: 132945.
  • 16
    Haftenberger M, Lahmann PH, Panico S, Gonzalez CA, Seidell JC, Boeing H, Giurdanella MC, Krogh V, Bueno- de-Mesquita HB, Peeters PH, Skeie G, Hjartaker A, et al. Overweight, obesity and fat distribution in 50- to 64-year-old participants in the European Prospective Investigation into Cancer and Nutrition (EPIC). Public Health Nutr 2002; 5: 114762.
  • 17
    Spencer EA, Appleby PN, Davey GK, Key TJ. Validity of self-reported height and weight in 4808 EPIC-Oxford participants. Public Health Nutr 2002; 5: 5615.
  • 18
    Spencer EA, Roddam AW, Key TJ. Accuracy of self-reported waist and hip measurements in 4492 EPIC-Oxford participants. Public Health Nutr 2004; 7: 7237.
  • 19
    Margetts BM, Pietinen P. European prospective investigation into cancer and nutrition: validity studies on dietary assessment methods. Int J Epidemiol 1997; 26 ( Suppl 1): S15.
  • 20
    Pathology and genetics. Tumours of the breast and female genital organs. In: TavassoliFA, DevileeD, eds. World Health Organization classification of tumours. Lyon: IARC Press, 2003.
  • 21
    International Agency for Research on Cancer (IARC)/World Health Organization (WHO). Weight control and physical activity (IARC Handbook for Cancer Prevention, vol. 6). Lyon, France: IARC Press, 2002.
  • 22
    Perry GS, Byers TE, Mokdad AH, Serdula MK, Williamson DF. The validity of self-reports of past body weights by U.S. adults. Epidemiology 1995; 6: 616.
  • 23
    Casey VA, Dwyer JT, Berkey CS, Coleman KA, Gardner J, Valadian I. Long-term memory of body weight and past weight satisfaction: a longitudinal follow-up study. Am J Clin Nutr 1991; 53: 14938.
  • 24
    Stevens J, Juhaeri Cai J. Changes in body mass index prior to baseline among participants who are ill or who die during the early years of follow-up. Am J Epidemiol 2001; 153: 94653.
  • 25
    Renehan AG, Tyson M, Egger M, Heller RF, Zwahlen M. Body-mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studies. Lancet 2008; 371: 56978.
  • 26
    Rapp K, Klenk J, Ulmer H, Concin H, Diem G, Oberaigner W, Schroeder J. Weight change and cancer risk in a cohort of more than 65,000 adults in Austria. Ann Oncol 2008; 19: 6418.
  • 27
    Rossing MA, Tang MT, Flagg EW, Weiss LK, Wicklund KG, Weiss NS. Body size and risk of epithelial ovarian cancer (United States). Cancer Causes Control 2006; 17: 71320.
  • 28
    Peterson NB, Trentham-Dietz A, Newcomb PA, Chen Z, Gebretsadik T, Hampton JM, Stampfer MJ, Willett WC, Egan KM. Relation of anthropometric measurements to ovarian cancer risk in a population-based case-control study (United States). Cancer Causes Control 2006; 17: 45967.
  • 29
    Anderson JP, Ross JA, Folsom AR. Anthropometric variables, physical activity, and incidence of ovarian cancer: The Iowa Women's Health Study. Cancer 2004; 100: 151521.
  • 30
    Folsom AR, Kushi LH, Anderson KE, Mink PJ, Olson JE, Hong CP, Sellers TA, Lazovich D, Prineas RJ. Associations of general and abdominal obesity with multiple health outcomes in older women: the Iowa Women's Health Study. Arch Intern Med 2000; 160: 211728.
  • 31
    Dal Maso L, Franceschi S, Negri E, Conti E, Montella M, Vaccarella S, Canzonieri V, Parazzini F, La Vecchia C. Body size indices at different ages and epithelial ovarian cancer risk. Eur J Cancer 2002; 38: 176974.
  • 32
    Mason C, Katzmarzyk PT. Variability in waist circumference measurements according to anatomic measurement site. Obesity 2009; 17: 178995.
  • 33
    Tworoger SS, Eliassen AH, Missmer SA, Baer H, Rich-Edwards J, Michels KB, Barbieri RL, Dowsett M, Hankinson SE. Birthweight and body size throughout life in relation to sex hormones and prolactin concentrations in premenopausal women. Cancer Epidemiol Biomarkers Prev 2006; 15: 2494501.
  • 34
    Kaaks R, Berrino F, Key T, Rinaldi S, Dossus L, Biessy C, Secreto G, Amiano P, Bingham S, Boeing H, Bueno de Mesquita HB, Chang-Claude J, et al. Serum sex steroids in premenopausal women and breast cancer risk within the European Prospective Investigation into Cancer and Nutrition (EPIC). J Natl Cancer Inst 2005; 97: 75565.
  • 35
    Rinaldi S, Dossus L, Lukanova A, Peeters PH, Allen NE, Key T, Bingham S, Khaw KT, Trichopoulos D, Trichopoulou A, Oikonomou E, Pera G, et al. Endogenous androgens and risk of epithelial ovarian cancer: results from the European Prospective Investigation into Cancer and Nutrition (EPIC). Cancer Epidemiol Biomarkers Prev 2007; 16: 239.
  • 36
    Calle EE, Kaaks R. Overweight, obesity and cancer: epidemiological evidence and proposed mechanisms. Nat Rev Cancer 2004; 4: 57991.
  • 37
    Rinaldi S, Key TJ, Peeters PH, Lahmann PH, Lukanova A, Dossus L, Biessy C, Vineis P, Sacerdote C, Berrino F, Panico S, Tumino R, et al. Anthropometric measures, endogenous sex steroids and breast cancer risk in postmenopausal women: a study within the EPIC cohort. Int J Cancer 2006; 118: 28329.
  • 38
    Lukanova A, Lundin E, Toniolo P, Micheli A, Akhmedkhanov A, Rinaldi S, Muti P, Lenner P, Biessy C, Krogh V, Zeleniuch-Jacquotte A, Berrino F, et al. Circulating levels of insulin-like growth factor-I and risk of ovarian cancer. Int J Cancer 2002; 101: 54954.
  • 39
    Peeters PH, Lukanova A, Allen N, Berrino F, Key T, Dossus L, Rinaldi S, van Gils CH, Bueno-de-Mesquita HB, Boeing H, Schulz M, Chang-Claude J, et al. Serum IGF-I, its major binding protein (IGFBP-3) and epithelial ovarian cancer risk: the European Prospective Investigation into Cancer and Nutrition (EPIC). Endocr Relat Cancer 2007; 14: 8190.
  • 40
    Lukanova A, Lundin E, Akhmedkhanov A, Micheli A, Rinaldi S, Zeleniuch-Jacquotte A, Lenner P, Muti P, Biessy C, Krogh V, Berrino F, Hallmans G, et al. Circulating levels of sex steroid hormones and risk of ovarian cancer. Int J Cancer 2003; 104: 63642.
  • 41
    Lacey JV Jr, Leitzmann M, Brinton LA, Lubin JH, Sherman ME, Schatzkin A, Schairer C. Weight, height, and body mass index and risk for ovarian cancer in a cohort study. Ann Epidemiol 2006; 16: 86976.
  • 42
    World Cancer Research Fund and American Institute for Cancer Research, Food, Nutrition, Physical Activity, and the Prevention of Cancer: a Global Perspective. Washington, DC: AICR, 2007.
  • 43
    Risch HA, Marrett LD, Jain M, Howe GR. Differences in risk factors for epithelial ovarian cancer by histologic type. Results of a case-control study. Am J Epidemiol 1996; 144: 36372.
  • 44
    Olsen CM, Nagle CM, Whiteman DC, Purdie DM, Green AC, Webb PM. Body size and risk of epithelial ovarian and related cancers: a population-based case-control study. Int J Cancer 2008; 123: 4506.
  • 45
    Kuper H, Cramer DW, Titus-Ernstoff L. Risk of ovarian cancer in the United States in relation to anthropometric measures: does the association depend on menopausal status? Cancer Causes Control 2002; 13: 45563.
  • 46
    Nagle CM, Olsen CM, Webb PM, Jordan SJ, Whiteman DC, Green AC. Endometrioid and clear cell ovarian cancers—A comparative analysis of risk factors. Eur J Cancer 2008; 44: 247784.
  • 47
    Riman T, Dickman PW, Nilsson S, Nordlinder H, Magnusson CM, Persson IR. Some life-style factors and the risk of invasive epithelial ovarian cancer in Swedish women. Eur J Epidemiol 2004; 19: 10119.
  • 48
    Leitzmann MF, Koebnick C, Danforth KN, Brinton LA, Moore SC, Hollenbeck AR, Schatzkin A, Lacey JV Jr. Body mass index and risk of ovarian cancer. Cancer 2009; 115: 81222.
  • 49
    Lahmann PH, Hoffmann K, Allen N, Van Gils CH, Khaw KT, Tehard B, Berrino F, Tjonneland A, Bigaard J, Olsen A, Overvad K, Clavel-Chapelon F, et al. Body size and breast cancer risk: findings from the European Prospective Investigation into Cancer and Nutrition (EPIC). Int J Cancer 2004; 111: 76271.
  • 50
    Friedenreich C, Cust A, Lahmann PH, Steindorf K, Boutron-Ruault MC, Clavel-Chapelon F, Mesrine S, Linseisen J, Rohrmann S, Boeing H, Pischon T, Tjonneland A, et al. Anthropometric factors and risk of endometrial cancer: the European prospective investigation into cancer and nutrition. Cancer Causes Control 2007; 18: 399413.
  • 51
    Wolf LA, Terry PD, Potter JD, Bostick RM. Do factors related to endogenous and exogenous estrogens modify the relationship between obesity and risk of colorectal adenomas in women? Cancer Epidemiol Biomarkers Prev 2007; 16: 67683.