Abstract
Background
Obesity disproportionately affects African American (AA) women and has been shown to increase ovarian cancer risk, with some suggestions that the association may differ by race.
Methods
We evaluated body mass index (BMI) and invasive epithelial ovarian cancer (EOC) risk in a pooled study of case–control and nested case–control studies including AA and White women. We evaluated both young adult and recent BMI (within the last 5 years). Associations were estimated using multi-level and multinomial logistic regression models.
Results
The sample included 1078 AA cases, 2582 AA controls, 3240 White cases and 9851 White controls. We observed a higher risk for the non-high-grade serous (NHGS) histotypes for AA women with obesity (ORBMI 30+= 1.62, 95% CI: 1.16, 2.26) and White women with obesity (ORBMI 30+= 1.20, 95% CI: 1.02, 2.42) compared to non-obese. Obesity was associated with higher NHGS risk in White women who never used HT (ORBMI 30+= 1.40, 95% CI: 1.08, 1.82). Higher NHGS ovarian cancer risk was observed for AA women who ever used HT (ORBMI 30+= 2.66, 95% CI: 1.15, 6.13), while in White women, there was an inverse association between recent BMI and risk of EOC and HGS in ever-HT users (EOC ORBMI 30+= 0.81, 95% CI: 0.69, 0.95, HGS ORBMI 30+= 0.73, 95% CI: 0.61, 0.88).
Conclusion
Obesity contributes to NHGS EOC risk in AA and White women, but risk across racial groups studied differs by HT use and histotype.
Subject terms: Cancer epidemiology, Risk factors
Background
Excess body weight, a potentially modifiable risk factor, is responsible for ~11% of all cancers in women [1]. A higher body mass index (BMI) has been associated with higher risks of several cancers [2–5], including ovarian cancer [5–9]. Several meta-analyses and pooled analyses also supported an association. For example, a meta-analysis of 47 studies estimated a 5% increase in ovarian cancer risk per 5 unit increase in BMI [10]. Evidence from the Ovarian Cancer Association Consortium (OCAC) showed that risk varies by histotype, with a higher risk for borderline serous, invasive endometrioid, and invasive mucinous tumours. [11] However, most of this evidence comes from studies conducted primarily in White populations. In the United States, the rates of obesity are high among Black/African American women, and there is suggestive evidence of racial disparities in associations between BMI and cancer [12]. In the first study focused on BMI and ovarian cancer risk specifically in African American (AA) women, a higher risk of ovarian cancer was observed in women who were obese and morbidly obese [13]. However, the association of obesity with ovarian cancer risk by race has not been evaluated in adequately powered analyses. Here, we examined the associations between recent and young adult BMI and invasive epithelial ovarian cancer (EOC) risk by race and histotype in the Ovarian Cancer in Women of African Ancestry (OCWAA) Consortium.
Methods
Study population
Methodology for the OCWAA Consortium has been published [14]. In brief, the OCWAA Consortium has pooled and harmonised data from four case–control studies—African American Cancer Epidemiology Study (AACES) [15], Cook County Case-Control Study (CCCCS) [16], Los Angeles County Ovarian Cancer Study (LACOCS) [17], North Carolina Ovarian Cancer Study (NCOCS) [18]—and three nested case–control studies within prospective cohorts—Black Women’s Health Study (BWHS) [19], Multiethnic Cohort Study (MEC) [20], and the Women’s Health Initiative (WHI) [21]—to examine differences between AA and White women in ovarian cancer incidence and survival, with consideration of histotype. The race was self-reported.
Ovarian cancer diagnoses
Diagnoses were abstracted from medical records or obtained from cancer registry reports; only invasive EOC cases, defined using the International Classification of Diseases for Oncology, Version 3 [22] were included. Accounting for both morphology and grade [23, 24], histotype was classified as high-grade serous, low-grade serous, endometrioid, clear cell, mucinous, carcinosarcoma and other epithelial tumours. Because statistical power was limited for some histotypes, we evaluated the risk for two groups: high-grade serous (HGS) and non-high-grade serous (NHGS) histotypes combined (all other histologies).
Body mass index
Each OCWAA study collected height and weight information before diagnosis and covariate data via self-administered questionnaires and/or interviews. Recent and young adult BMI (kg/m2) were harmonised across studies. Recent BMI within the 5-year period prior to diagnosis (cases) or interview (controls) was calculated from self-reported height and weight (AACES, NCOCS, MEC, BWHS, LACOCS, CCCCS) or a combination of measured of reported height and weight during the same 5-year period, prioritising self-report for comparability with other studies, where available (WHI) to compute BMI (kg/m2). BMI was treated as a continuous variable and categorised using WHO criteria (recent BMI) or race-specific quartiles in controls (young adult BMI) in multivariable models. WHI was excluded from young adult BMI analyses because too few women had data available.
Covariates
We considered the following covariates as potential confounders of the BMI–ovarian cancer association based on established or potential association with BMI and/or EOC risk: age at diagnosis or interview (continuous, years), education (high school or less, some college, college graduate, graduate education), parity (0, 1, 2, ≥3 full-term pregnancies), oral contraceptive use (OC use; never, <60 months, ≥60 months), menopausal status (premenopausal, postmenopausal) and hormone therapy use (HT use; ever, never), tubal ligation (no, yes), cigarette smoking status (never, former, current), first-degree family history of breast or ovarian cancer (yes, no), and study site. We also evaluated differences in HT use by type (oestrogen alone, oestrogen plus progestin).
Statistical analysis
Descriptive statistics were calculated to characterise the distribution of covariates and BMI variables for analysis stratified by both race and case–control status.
Race-specific multi-level logistic regression models were used to test for overall associations of recent and young adult BMI with odds of ovarian cancer, adjusting for the aforementioned covariates. In site-specific models for recent BMI, we estimated odds ratios (ORs) and 95% confidence intervals (CIs) using unconditional logistic regression (AACES, CCCCS, NCOCS, LACOCS) or conditional logistic regression stratifying by pairwise matching (BWHS, MEC, WHI) to estimate the ovarian cancer risk for both women who are overweight (BMI 25–29.9 kg/m2) and women who have obesity (BMI > 30 kg/m2), compared to women who have a normal BMI (BMI 18.5–24.9 kg/m2). Women with a recent BMI < 18.5 kg/m2 (underweight) were excluded because there were too few to estimate meaningful associations (1.1% of AA women and 2.0% of White women). We similarly estimated ovarian cancer risk according to quartiles of young adult BMI. Random-effects models were used first, and if the variance components tests detected no significant heterogeneity by the study [25, 26], random-effects coefficients were removed from the models, leaving only a fixed effect to control for differences between sites; only recent BMI and young adult BMI in Whites used mixed-effect models.
We further explored associations of recent BMI with histotype-specific ovarian cancer using the same multi-level approach and covariates with polytomous outcomes instead of dichotomous: HGS cancer and NHGS cancers (all other invasive ovarian cancers), compared to controls. P values for trend were calculated by including the median in each category as a continuous variable in the models. All modelling and forest plots were created in SAS 9.4 [26–28].
We evaluated possible effect modification by HT use (any HT use or oestrogen alone HT use, classified as ever/never) for recent BMI and ovarian cancer associations in women aged 50 or older that was guided by prior literature [8]. We also evaluated differences in HT use by type (oestrogen alone, oestrogen plus progestin).
To examine whether undiagnosed advanced ovarian cancer leading to weight loss influenced our results, we conducted stratified analyses by stage (localised/regional and distant) using the same covariates. We used a SEER summary stage variable where localised stage is equivalent to FIGO Stage I, the regional stage is equivalent to FIGO Stage II and the distant stage is equivalent to FIGO Stage III/IV.
Results
This OCWAA sample included 1078 AA cases and 2582 AA controls, and 3240 White cases and 9851 White controls. Among AA women, compared to controls, cases had fewer years of education (40% have high school or less vs 33%), were more likely to be nulliparous (19% vs 16%), were less likely to use OCs for 5 years or more (25% vs 28%), were more likely to never use HT (76% vs 72%), and more likely to have a first-degree family history of breast or ovarian cancer (26% vs 14%; Table 1).
Table 1.
Descriptive statistics for BMI–ovarian cancer association study in the OCWAA consortium.
| N (%) or mean (SD) | African American (N = 3660) | White (N = 13,091) | ||||
|---|---|---|---|---|---|---|
| Cases (N = 1078) | Controls (N = 2582) | P value | Cases (N = 3240) | Controls (N = 9851) | P value | |
| Age, year | 58.9 (11.6) | 59.5 (13.5) | 0.130 | 62.9 (11.9) | 67.0 (11.9) | <0.0001 |
| Site, % | ||||||
| AACES | 570 (52.9) | 739 (28.6) | <0.0001 | 0 (0.0) | 0 (0.0) | <0.0001 |
| CCCCS | 43 (4.0) | 77 (3.0) | 222 (6.9) | 404 (4.1) | ||
| LACOCS | 124 (11.5) | 141 (5.5) | 1158 (35.7) | 1764 (17.9) | ||
| NCOCS | 115 (10.7) | 187 (7.2) | 785 (24.2) | 834 (8.5) | ||
| BWHS | 89 (8.3) | 598 (23.2) | 0 (0.0) | 0 (0.0) | ||
| MEC | 90 (8.4) | 528 (20.5) | 142 (4.4) | 852 (8.7) | ||
| WHI | 47 (4.4) | 312 (12.1) | 933 (28.8) | 5997 (60.9) | ||
| Education, % | ||||||
| HS or less | 438 (40.6) | 859 (33.3) | <0.0001 | 687 (21.2) | 1962 (19.9) | <0.0001 |
| Some college | 281 (26.1) | 780 (30.2) | 937 (28.9) | 3065 (31.1) | ||
| College graduate | 205 (19.0) | 470 (18.2) | 730 (22.5) | 1718 (17.4) | ||
| Grad/prof school | 154 (14.3) | 456 (17.7) | 878 (27.1) | 3065 (31.1) | ||
| Missing | 0 (0.0) | 17 (0.7) | 8 (0.3) | 41 (0.4) | ||
| Parity, % | ||||||
| No pregnancies | 205 (19.0) | 420 (16.3) | 0.028 | 747 (23.1) | 1756 (17.8) | <0.0001 |
| 1–2 pregnancies | 417 (38.7) | 1106 (42.8) | 1344 (41.5) | 3649 (37.0) | ||
| 3+ pregnancies | 455 (42.2) | 1042 (40.4) | 1142 (35.3) | 4387 (44.5) | ||
| Missing | 1 (0.1) | 14 (0.5) | 7 (0.2) | 59 (0.6) | ||
| OC use/duration, % | ||||||
| Never | 421 (39.1) | 1053 (40.8) | 0.010 | 1572 (48.5) | 5308 (53.9) | <0.0001 |
| <5 years | 374 (34.7) | 769 (29.8) | 950 (29.3) | 2125 (21.6) | ||
| ≥5 years | 265 (24.6) | 716 (27.7) | 682 (21.1) | 2371 (24.1) | ||
| Missing | 18 (1.7) | 44 (1.7) | 36 (1.1) | 47 (0.5) | ||
| Tubal ligation, % | ||||||
| No | 751 (69.7) | 1746 (67.6) | 0.551 | 2760 (85.2) | 8098 (82.2) | 0.0001 |
| Yes | 316 (29.3) | 772 (29.9) | 477 (14.7) | 1733 (17.6) | ||
| Missing | 11 (1.0) | 64 (2.5) | 3 (0.1) | 20 (0.2) | ||
| Age at menarche, % | ||||||
| <11 years | 107 (9.9) | 247 (9.6) | 0.940 | 210 (6.5) | 610 (6.2) | 0.231 |
| 11–12 years | 437 (40.5) | 1077 (41.7) | 1341 (41.4) | 4008 (40.7) | ||
| 13–14 years | 398 (36.9) | 923 (35.8) | 1367 (42.2) | 4237 (43.0) | ||
| 15–16 years | 108 (10.0) | 265 (10.3) | 271 (8.4) | 881 (8.9) | ||
| 17+ years | 20 (1.9) | 52 (2.0) | 45 (1.4) | 97 (1.0) | ||
| Missing | 8 (0.7) | 18 (0.7) | 6 (0.2) | 18 (0.2) | ||
| Menopausal status, % | ||||||
| Premenopausal | 264 (24.5) | 686 (26.6) | 0.200 | 552 (17.0) | 1142 (11.6) | <0.0001 |
| Postmenopausal | 812 (75.3) | 1892 (73.3) | 2687 (82.9) | 8700 (88.3) | ||
| Missing | 2 (0.2) | 4 (0.2) | 1 (0.03) | 9 (0.10) | ||
| HT use ever, % | ||||||
| No | 821 (76.2) | 1852 (71.7) | 0.018 | 1512 (46.7) | 4251 (43.2) | 0.0006 |
| Yes | 247 (22.9) | 684 (26.5) | 1721 (53.1) | 5569 (56.5) | ||
| Missing | 10 (0.9) | 46 (1.8) | 7 (0.2) | 31 (0.3) | ||
| HT duration, % | ||||||
| None | 821 (76.2) | 1852 (71.7) | 0.031 | 1512 (46.7) | 4251 (43.2) | 0.0002 |
| <5 years | 140 (13.0) | 409 (15.8) | 561 (17.3) | 1950 (19.8) | ||
| 5+ years | 102 (9.5) | 271 (10.5) | 1115 (34.4) | 3569 (36.2) | ||
| Missing | 15 (1.4) | 50 (1.9) | 52 (1.6) | 81 (0.8) | ||
| Smoking status, % | ||||||
| Never | 574 (53.3) | 1352 (52.4) | 0.003 | 1627 (50.2) | 4891 (49.7) | 0.010 |
| Former | 365 (33.9) | 775 (30.0) | 1293 (39.9) | 4076 (41.4) | ||
| Current | 138 (12.8) | 434 (16.8) | 297 (9.2) | 800 (8.1) | ||
| Missing | 1 (0.10) | 21 (0.8) | 23 (0.7) | 84 (0.9) | ||
| Pack-years of smoking, % | ||||||
| 0 | 574 (53.3) | 1354 (52.4) | 0.783 | 1627 (50.22) | 4891 (49.65) | 0.583 |
| <10 years | 243 (22.5) | 546 (21.2) | 587 (18.12) | 1846 (18.74) | ||
| 10+ years | 256 (23.8) | 619 (24.0) | 957 (29.54) | 2829 (28.72) | ||
| Missing | 5 (0.46) | 63 (2.44) | 69 (2.13) | 285 (2.89) | ||
| Family history of breast or ovarian cancer, % | ||||||
| No | 720 (66.8) | 2081 (80.6) | <0.0001 | 2484 (76.7) | 7748 (78.7) | <0.0001 |
| Yes | 278 (25.8) | 366 (14.2) | 643 (19.9) | 1560 (15.8) | ||
| Missing | 80 (7.4) | 135 (5.2) | 113 (3.5) | 543 (5.5) | ||
| Recent BMI, kg/m2 | 32.0 (7.8) | 30.6 (7.2) | <0.0001 | 26.6 (5.9) | 27.0 (5.7) | 0.003 |
| Recent BMI categories | ||||||
| 18.5–24.99 kg/m2 | 183 (17.0) | 544 (21.1) | 0.0001 | 1580 (48.8) | 4260 (43.2) | <0.0001 |
| 25–29.99 kg/m2 | 314 (29.1) | 842 (32.6) | 912 (28.2) | 3217 (32.7) | ||
| 30+ kg/m2 | 581 (53.9) | 1196 (46.3) | 748 (23.1) | 2374 (24.1) | ||
| Young adult BMI,* kg/m2 | 22.3 (4.6) | 21.9 (4.5) | 0.027 | 21.1 (3.2) | 21.1 (3.2) | 0.662 |
| Young adult BMI, race-specific quartiles,* kg/m2 | ||||||
| Q1 | 228 (22.1) | 542 (23.9) | 0.014 | 505 (21.9) | 899 (23.3) | 0.053 |
| Q2 | 219 (21.2) | 537 (23.7) | 648 (28.1) | 957 (24.8) | ||
| Q3 | 249 (24.2) | 551 (24.3) | 558 (24.2) | 966 (25.1) | ||
| Q4 | 304 (29.5) | 547 (24.1) | 573 (24.8) | 974 (25.3) | ||
| Missing | 31 (3.0) | 93 (4.1) | 23 (1.0) | 58 (1.5) | ||
*WHI was excluded from young adult BMI analyses due to low numbers. The reduced subset excluding WHI includes 1031 African American cases, 2270 African American controls, 2307 White cases, and 3854 White controls.
AA quartiles young adult BMI: Q1 (13.15, 19.06), Q2 (19.06, 20.98), Q3 (20.98, 23.30), Q4 (23.30, 59.52).
White quartiles for young adult BMI: Q1 (12.27, 19.05), Q2 (19.05, 20.55), Q3 (20.55, 22.17), Q4 (22.17, 54.87).
Among White women, compared to controls, cases were younger (63 vs 67 years), were more likely to be nulliparous (23% vs 18%), were less likely to use OCs for 5 or more years (21% vs 24%), were less likely to have had a tubal ligation (15% vs 18%), were more likely to be premenopausal (17% vs 12%), and were less likely to use HT (52% vs 57%). White cases were also more likely to have a first-degree breast or ovarian cancer family history compared to White controls (20% vs 16%).
In a comparison of recent BMI, 54% of AA cases were obese (BMI > 30 kg/m2) compared to 23% of White cases. Similarly, the prevalence of obesity was higher in AA controls (46%) compared to White controls (24%). On average, AA cases had a 0.4 kg/m2 higher young adult BMI, and a 1.5 kg/m2 higher recent BMI compared to AA controls. In White women, cases did not differ with respect to young adult BMI, but on average, cases had a lower recent BMI (0.4 kg/m2) compared to controls.
Table 2 shows the pooled ORs for invasive ovarian cancer by race for recent and young adult BMI. In AA women, there was a higher risk for women with a BMI ≥ 30 kg/m2 compared to normal BMI (OR = 1.23, 95% CI: 0.98,1.55). This finding is driven by the NHGS histotype, where the OR for AA women with obesity was 1.62 (95% CI: 1.16, 2.26) compared to AA women with a normal BMI, while the association for HGS was 1.04 (95% CI: 0.80, 1.36), P-heterogeneity = 0.022. For young adult BMI, there was a suggestion of increased risk with higher BMI in AA women (OR = 1.36, 95% CI: 0.97, 1.91) and the NHGS EOC, but no significant associations of BMI and the HGS histotype.
Table 2.
Associations from logistic regression models for recent and young adult BMI and invasive ovarian cancer risk and multinomial models for high-grade serous and non-high-grade serous epithelial ovarian cancer in the OCWAA consortium, by race.
| All invasivea | High-grade serousa | Non-high-grade serousa | P-heterogeneity (between histotypes) | ||||
|---|---|---|---|---|---|---|---|
| N cases/N controls | OR (95% CI) | N cases/N controls | OR (95% CI) | N cases/N controls | OR (95% CI) | ||
| African American | |||||||
| Recent BMI, kg/m2 | 969/2289 | 610/2289 | 359/2289 | ||||
| 18.5–<25 | 166/477 | 1.00 | 109/477 | 1.00 | 57/477 | 1.00 | N/A |
| 25–<30 | 283/747 | 1.13 (0.88, 1.43) | 189/747 | 1.08 (0.82, 1.43) | 94/747 | 1.19 (0.83, 1.71) | 0.649 |
| 30+ | 520/1065 | 1.23 (0.98, 1.55) | 312/1065 | 1.04 (0.80, 1.36) | 208/1065 | 1.62 (1.16, 2.26) | 0.022 |
| P for trend | – | 0.072 | 0.932 | 0.004 | – | ||
| Young adult BMI | 910/1991 | 574/1991 | 336/1991 | ||||
| Q1 | 216/497 | 1.00 | 144/497 | 1.00 | 72/497 | 1.00 | N/A |
| Q2 | 190/493 | 0.98 (0.77, 1.26) | 127/493 | 1.00 (0.75, 1.33) | 63/493 | 0.96 (0.66, 1.40) | 0.875 |
| Q3 | 226/496 | 1.10 (0.87, 1.41) | 134/496 | 0.97 (0.73, 1.29) | 92/496 | 1.35 (0.95, 1.91) | 0.107 |
| Q4 | 278/505 | 1.18 (0.93, 1.49) | 169/505 | 1.08 (0.82, 1.42) | 109/505 | 1.36 (0.97, 1.91) | 0.235 |
| P for trend | 0.118 | 0.661 | 0.052 | – | |||
| White | |||||||
| Recent BMIb, kg/m2 | 3050/9077 | 1896/9077 | 1154/9077 | ||||
| 18.5–<25 | 1494/3984 | 1.00 | 962/3984 | 1.00 | 532/3984 | 1.00 | N/A |
| 25–<30 | 855/2935 | 0.87 (0.74, 1.03) | 520/2935 | 0.82 (0.73, 0.93) | 335/2935 | 1.04 (0.89, 1.21) | 0.011 |
| 30+ | 701/2158 | 0.98 (0.74, 1.29) | 414/2158 | 0.91 (0.80, 1.04) | 287/2158 | 1.20 (1.02, 1.42) | 0.004 |
| P for trend | 0.497 | 0.159 | 0.031 | – | |||
| Young adult BMIb | 2256/3772 | 1338/3772 | 918/3772 | ||||
| Q1 | 521/941 | 1.00 | 308/941 | 1.00 | 213/941 | 1.00 | N/A |
| Q2 | 634/945 | 1.36 (0.86, 2.16) | 395/945 | 1.38 (1.15, 1.65) | 239/945 | 1.22 (0.99, 1.51) | 0.333 |
| Q3 | 541/943 | 1.20 (0.90, 1.60) | 320/943 | 1.16 (0.96, 1.41) | 221/943 | 1.20 (0.97, 1.49) | 0.795 |
| Q4 | 560/943 | 1.16 (0.72, 1.86) | 315/943 | 1.20 (0.99, 1.45) | 245/943 | 1.24 (1.00, 1.54) | 0.799 |
| P for trend | 0.955 | 0.405 | 0.066 | – | |||
aModels adjusted for study, age, education, parity, oral contraceptive use, menopausal status, tubal ligation status, postmenopausal hormone use (ever/never), smoking status and family history of breast/ovarian cancer.
bRandom effects by the site are included in the model.
In White women, there was no evidence for an association between obesity and invasive EOC risk overall. However, in histotype-stratified analyses, we observed a higher risk for the NHGS histotype for women with a BMI ≥ 30 kg/m2 before diagnosis (OR = 1.20, 95% CI: 1.02, 1.42), but no association for HGS, P-heterogeneity = 0.004. Similarly for young adult BMI, we observed a higher risk for White women in the highest BMI quartile for the NHGS histotype (ORQ4 = 1.24, 95% CI: 1.00, 1.54), compared to the lowest and some suggestion of increased HGS EOC risk but no clear trend of increasing risk with higher BMI. The results of elevated risk of NHGS EOC for women with obesity remained essentially unchanged when women with previous cancers were excluded (OR: 1.57; 95% CI: 1.11, 2.22 for AA women and OR: 1.19; 95% CI: 0.99–1.42 for White women; data not shown).
Associations also differed by HT use (Table 3). In never HT users, we observed a higher risk of all invasive EOC and the NHGS histotype in White women with obesity (ORBMI 30+= 1.40, 95% CI: 1.08, 1.82). In AA women who never used HT, we observed an association of similar magnitude that did not reach statistical significance (ORBMI 30+= 1.36, 95% CI: 0.85, 2.17) for; the NHGS histotype. In AA women who used HT, we observed a higher risk of NHGS histotype in women with obesity (ORBMI 30+= 2.66, 95% CI: 1.15, 6.13) compared to normal BMI. In White women who used HT, we observed inverse associations between women who are overweight and women with obesity for risk of invasive EOC and the HGS histotype (ORBMI 30+= 0.73, 95% CI: 0.61, 0.88). When we analysed oestrogen alone HT use, the results were consistent.
Table 3.
Associations for recent body mass index and ovarian cancer risk, by hormone therapy use and race, among women aged 50 + in the OCWAA Consortium.
| Recent body mass index (BMI) | All invasivea | High-grade serousa | Non-high-grade serousa | P-heterogeneity (between histotypes) | |||
|---|---|---|---|---|---|---|---|
| N cases/N controls | OR (95% CI) | N cases/N controls | OR (95% CI) | N cases/N controls | OR (95% CI) | ||
| Never users of hormone therapy | |||||||
| African American | 545/1132 | 364/1132 | 181/1132 | ||||
| 18.5–<25 | 93/216 | 1.00 | 64/216 | 1.00 | 29/216 | 1.00 | N/A |
| 25–<30 | 149/356 | 1.02 (0.72, 1.42) | 103/356 | 0.99 (0.68, 1.46) | 46/356 | 1.05 (0.63, 1.76) | 0.846 |
| 30+ | 303/560 | 1.10 (0.80, 1.50) | 197/560 | 0.98 (0.68, 1.39) | 106/560 | 1.36 (0.85, 2.17) | 0.212 |
| P for trend | 0.490 | 0.761 | 0.103 | – | |||
| White | 1041/3123 | 607/3123 | 434/3123 | ||||
| 18.5–<25 | 433/1273 | 1.00 | 260/1273 | 1.00 | 173/1273 | 1.00 | N/A |
| 25–<30 | 308/982 | 1.10 (0.91, 1.32) | 177/982 | 1.03 (0.83, 1.29) | 131/982 | 1.20 (0.93, 1.55) | 0.331 |
| 30+ | 300/868 | 1.26 (1.04, 1.52) | 170/868 | 1.17 (0.93, 1.47) | 130/868 | 1.40 (1.08, 1.82) | 0.256 |
| P for trend | 0.018 | 0.151 | 0.014 | – | |||
| Ever users of any hormone therapy | |||||||
| African American | 199/576 | 119/576 | 80/576 | ||||
| 18.5–<25 | 29/117 | 1.00 | 21/117 | 1.00 | 8/117 | 1.00 | N/A |
| 25–<30 | 72/213 | 1.38 (0.80, 2.38) | 43/213 | 1.10 (0.58, 2.08) | 29/213 | 2.06 (0.87, 4.86) | 0.210 |
| 30+ | 98/246 | 1.49 (0.88, 2.53) | 55/246 | 1.08 (0.58, 2.00) | 43/246 | 2.66 (1.15, 6.13) | 0.062 |
| P for trend | 0.198 | 0.858 | 0.065 | – | |||
| White | 1541/5021 | 1075/5021 | 466/5021 | ||||
| 18.5–<25 | 795/2154 | 1.00 | 576/2154 | 1.00 | 219/2154 | 1.00 | N/A |
| 25–<30 | 448/1726 | 0.79 (0.68, 0.91) | 301/1726 | 0.72 (0.62, 0.85) | 147/1726 | 0.96 (0.77, 1.21) | 0.030 |
| 30+ | 298/1141 | 0.81 (0.69, 0.95) | 198/1141 | 0.73 (0.61, 0.88) | 100/1141 | 1.02 (0.79, 1.33) | 0.024 |
| P for trend | 0.005 | <0.001 | 0.961 | – | |||
aModels adjusted for study, age, education, parity, oral contraceptive use, tubal ligation status, smoking status and family history of breast/ovarian cancer.
In models stratified by menopausal status, there are no significant trends observed among premenopausal women. In premenopausal Black women, there was a non-significant trend of increasing risk with obesity for NHGS (ORBMI 30+: 1.58, 95% CI: 0.87, 2.84) P-trend = 0.063). In postmenopausal Black women, there is a significant increased risk of NHGS ovarian cancer with obesity (ORBMI 30+: 1.64, 95% CI: 1.08, 2.47) P-trend = 0.038). However, in postmenopausal White women, there is a significantly decreased risk of all invasive and HGS ovarian cancer with a BMI 25-30 kg/m2 (ORBMI 25-30: 0.81, 95% CI: 0.71, 0.92) P-trend = 0.077). In postmenopausal women, the P-for heterogeneity for obesity risk across histotypes was significant, 0.043 for African Americans and 0.005 for whites (data not shown). In stratified analyses by stage (Table 4), for overall invasive EOC, a positive association with risk for recent BMI was observed for both AA and White women with localised/regional disease, but stronger in magnitude for AA women (AA ORBMI 30+= 1.51, 95% CI: 1.04, 2.18, White ORBMI 30+= 1.26, 95% CI: 1.05, 1.52). In histotype-stratified analyses, higher risks were observed for recent BMI and the NHGS histotype only among women with localised/regional disease; ORs were slightly higher for AA compared to Whites (AA ORBMI 30+= 1.74, 95% CI: 1.08, 2.80, White ORBMI 30+= 1.51, 95% CI: 1.21, 1.88). There was no evidence that recent BMI was associated with ovarian cancer risk in women diagnosed with distant disease.
Table 4.
Associations from logistic regression models for recent BMI and ovarian cancer risk, overall and by histotype, by localised/regional and distant disease and race, in the OCWAA Consortium.
| All invasivea | High-grade serousa | Non-high-grade serousa | P-heterogeneity (between histotypes) | ||||
|---|---|---|---|---|---|---|---|
| N cases/N controls | OR (95% CI) | N cases/N controls | OR (95% CI) | N cases/N controls | OR (95% CI) | ||
| Localised/regional | |||||||
| African Americanb | 265/2289 | 94/2289 | 171/2289 | ||||
| 18.5–<25 | 42/477 | 1.00 | 16/477 | 1.00 | 26/477 | 1.00 | N/A |
| 25–<30 | 76/747 | 1.25 (0.90, 1.73) | 31/747 | 1.21 (0.64, 2.30) | 45/747 | 1.39 (0.83, 2.35) | 0.732 |
| 30+ | 147/1065 | 1.51 (1.04, 2.18) | 47/1065 | 0.93 (0.51, 1.72) | 100/1065 | 1.74 (1.08, 2.80) | 0.103 |
| P for trend | 0.156 | 0.454 | 0.029 | – | |||
| White | 861/9077 | 288/9077 | 573/9077 | ||||
| 18.5–<25 | 408/3984 | 1.00 | 150/3984 | 1.00 | 258/3984 | 1.00 | N/A |
| 25–<30 | 234/2935 | 0.97 (0.81, 1.15) | 74/2935 | 0.75 (0.56, 1.00) | 160/2935 | 1.10 (0.89, 1.37) | 0.030 |
| 30+ | 219/2158 | 1.26 (1.05, 1.52) | 64/2158 | 0.90 (0.67, 1.23) | 155/2158 | 1.51 (1.21, 1.88) | 0.006 |
| P for trend | 0.019 | 0.420 | <0.001 | – | |||
| Distant | |||||||
| African American | 649/2289 | 486/2289 | 163/2289 | ||||
| 18.5–<25 | 122/477 | 1.00 | 92/477 | 1.00 | 30/477 | 1.00 | N/A |
| 25–<30 | 190/747 | 1.00 (0.76, 1.31) | 146/747 | 0.99 (0.73, 1.34) | 44/747 | 1.00 (0.61, 1.63) | 0.975 |
| 30+ | 337/1065 | 1.05 (0.81, 1.36) | 248/1065 | 0.97 (0.73, 1.29) | 89/1065 | 1.30 (0.83, 2.03) | 0.252 |
| P for trend | 0.631 | 0.741 | 0.220 | – | |||
| Whiteb | 2158/9077 | 1593/9077 | 565/9077 | ||||
| 18.5–<25 | 1078/3984 | 1.00 | 809/3984 | 1.00 | 269/3984 | 1.00 | N/A |
| 25–<30 | 611/2935 | 0.85 (0.72, 1.01) | 441/2935 | 0.83 (0.73, 0.95) | 170/2935 | 0.97 (0.79, 1.19) | 0.179 |
| 30+ | 469/2158 | 0.92 (0.74, 1.14) | 343/2158 | 0.90 (0.78, 1.04) | 126/2158 | 0.98 (0.78, 1.23) | 0.506 |
| P for trend | 0.214 | 0.132 | 0.894 | – | |||
aModels adjusted for study, age, education, parity, oral contraceptive use, menopausal status, tubal ligation status, postmenopausal hormone use (ever/never), smoking status and family history of breast/ovarian cancer.
bRandom effects by site are included in the model.
Discussion
Our study adds to the epidemiologic evidence that obesity increases risk of invasive EOC and suggests possible effect modification by race, histotype, HT use, and stage at diagnosis. Recent BMI was associated with higher NHGS risk in both AA and White women. Further, our finding of a positive association of young adult BMI and NHGS EOC in White and AA women provides additional information about obesity during early adult life as an exposure, with a higher risk for women who were obese both during young adulthood as well as during the recent period before diagnosis. Conversely, obesity may be protective for HGS EOC in Whites who used HT, or vice versa. Interestingly, we found contrasting associations according to HT use and race, where we observed higher NHGS risk among White women with obesity who were never users of HT, and higher NHGS risk for AA women with obesity who used HT.
Evidence from meta-analyses and prospective cohort studies support a role for obesity and invasive ovarian cancer [7, 29] and in particular, for the NHGS histotype [6, 10, 11, 30–32]. These findings are consistent with our own, where we found that higher BMI in adulthood was associated with increased risk of NHGS EOC in AA and White women. Our findings of stronger association of a BMI ≥ 30 kg/m2, which defines obesity, for AA women than White women could be attributed to a higher proportion of women with morbid obesity (i.e., BMI > 40 kg/m2) among the AA group. However, we were unable to compare associations with higher categories of BMI by race due to limited power.
Previous studies have reported effect modification by HT use, where higher BMI was found to be associated with higher risk of ovarian cancer in women who never used HT and risk was attenuated in women who ever used HT [8, 10]. Our findings are in partial agreement, where White women who are overweight and obese who used HT had lower HGS EOC risk. In AA women, our findings do not suggest an attenuated association of BMI and overall, HGS or NHGS risk in HT users. In fact, AA women with obesity who used HT had higher NHGS risk. The relatively small AA studies published to date suggest that HT is associated with higher EOC risk or suggestive of higher risk (all of which are included in the OCWAA consortium) [15, 18, 33]; therefore our finding of higher risk with obesity and ever-HT use may represent the joint effects of both risk factors for the NHGS histotype and a threshold of oestrogen exposure that may increase risk. In 12 case–control studies in OCAC, there was a suggestion of higher risk for HT use (OR = 1.07, 95% CI: 0.83, 1.38) [34]. Higher risk in the presence of both HT and obesity is biologically plausible given that HT and obesity in menopause are both associated with higher circulating oestrogens, and oestrogen and oestrogen metabolites are associated with nonserous histotypes [35].
Our sensitivity analyses by stage at diagnosis, where the associations were limited to local/regional stage, support a role for recent BMI that is not influenced by weight loss that could have occurred due to the disease process.
Social determinants of health play an important role in individual behaviours related to healthy eating and patterns of physical activity [36], therefore driving obesity prevalence. When working in cancer health disparities, it is important to recognise that African Americans have been disproportionately exposed to a variety of social and structural factors limiting access to healthy foods and opportunities for safe physical activity that contributes to higher risks for obesity-related cancers. While diet and physical activity are typically labelled as modifiable, they remain a significant issue in cancer prevention efforts tailored to address cancer health disparities.
Mendelian randomisation analysis using data for 87 BMI-associated genetic variants from 39 OCAC studies builds additional biologic support our obesity-NHGS findings; each genetically predicted 5 unit increase in BMI conferred a significantly increased risk of NHGS ovarian cancer (OR: 1.29, 95% CI: 1.03–1.61), but there was no association with HGS ovarian cancer risk (OR: 1.06, 95% CI: 0.88–1.27) [37]. These BMI-associated alleles were identified from studies that included predominantly women of European ancestry, and thus, the generalisability of these findings to AA women is unclear.
The sample size and comprehensive harmonisation of variables in the OCWAA consortium enabled us to stratify by histotype and control for a variety of important confounders. We were able to present results in the largest available sample of AA women for the study of ovarian cancer risk factors to date, overcoming the limitation of existing large meta-analyses where other racial and ethnic groups besides Non-Hispanic Whites are often combined as a single group and not able to draw population-specific inferences. However, our study was limited in the analysis of specific NHGS histotypes individually due to the low sample size. The limitations of BMI as a proxy for adiposity are well-known. Additional studies where specific compartments of adipose tissue are measured can provide additional information about these associations, given the known racial differences in body composition [38].
In summary, our study provides new insight into associations of BMI and risk of EOC by race, revealing complex associations by HT use, histotype, and stage at diagnosis. Future prospective studies with more detailed measures of body composition are needed to explore heterogeneity in associations according to histotype and HT use in diverse populations.
Supplementary information
Acknowledgements
The authors thank the WHI investigators and staff for their dedication, and the study participants for making the study possible. A full listing of WHI investigators can be found at https://www.whi.org/doc/WHI-Investigator-Long-List.pdf. Pathology data were obtained from the following state cancer registries (AZ, CA, CO, CT, DE, DC, FL, GA, IL, IN, KY, LA, MD, MA, MI, NJ, NY, NC, OK, PA, SC, TN, TX, VA), and results reported do not necessarily represent their views. The IRBs of participating institutions and cancer registries have approved these studies, as required. Opinions expressed by the authors are their own, and this material should not be interpreted as representing the official viewpoint of the U.S. Department of Health and Human Services, the National Institutes of Health, or the National Cancer Institute.
Author contributions
Conceptualisation: EVB, HMO, JMS and LR. Data curation: TNB, HRH, CEJ, PGM, WR, EM, HMO, VWS, AHW, LR, JMS and EVB. Formal analysis: CEJ. Funding acquisition: JMS and LR. Methodology: HMO, EVB, CEJ, BQ, JMS and LR. Writing—original draft: HMO, EVB and JMS. Writing—review and editing: all authors.
Funding
This study is supported by the National Institutes of Health (R01-CA207260 to Schildkraut and Rosenberg and K01-CA212056 to Bethea). AACES was funded by NCI (R01-CA142081 to Schildkraut); BWHS is funded by NIH (R01-CA058420,UM1-CA164974, and U01-CA164974 to Rosenberg); CCCCS was funded by NIH/NCI (R01-CA61093 to Rosenblatt); LACOCS was funded by NCI (R01-CA17054 to Pike, R01-CA58598 to Goodman and Wu, and Cancer Center Core Grant P30-CA014089 to Henderson and Wu) and by the California Cancer Research Program (2II0200 to A Wu); and NCOCS was funded by NCI (R01-CA076016 to Schildkraut). The WHI program is funded by the National Heart, Lung, and Blood Institute through contracts HHSN268201600018C, HHSN268201600001C, HHSN268201600002C, HHSN268201600003C and HHSN268201600004C. Additional grants to support WHI inclusion in OCWAA include UM1-CA173642-05 (to Anderson) and NIH/NHLBI-CSB-WH-2016-01-CM.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Competing interests
PGM has received compensation for work related to litigation in regard to talc and ovarian cancer. The remaining authors declare no competing interests.
Ethics approval and consent to participate
Each study obtained informed consent from its participants; the individual studies and the OCWAA Consortium were approved by the relevant Institutional Review Boards.
Consent to publish
Not applicable.
Footnotes
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
The online version contains supplementary material available at 10.1038/s41416-022-01981-6.
References
- 1.Islami F, Goding Sauer A, Miller KD, Siegel RL, Fedewa SA, Jacobs EJ, et al. Proportion and number of cancer cases and deaths attributable to potentially modifiable risk factors in the United States. CA Cancer J Clin. 2018;68:31–54. doi: 10.3322/caac.21440. [DOI] [PubMed] [Google Scholar]
- 2.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;371:569–78. doi: 10.1016/S0140-6736(08)60269-X. [DOI] [PubMed] [Google Scholar]
- 3.Basen-Engquist K, Chang M. Obesity and cancer risk: recent review and evidence. Curr Oncol Rep. 2011;13:71–6. doi: 10.1007/s11912-010-0139-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.World Cancer Research Fund/American Institute for Cancer Research. Diet, nutrition, physical activity and cancer: a global perspective. 2018; www.dietandcancerreport.org.
- 5.Lauby-Secretan B, Scoccianti C, Loomis D, Grosse Y, Bianchini F, Straif K, et al. Body fatness and cancer-viewpoint of the IARC Working Group. N Engl J Med. 2016;375:794–8. doi: 10.1056/NEJMsr1606602. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 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;43:690–709. doi: 10.1016/j.ejca.2006.11.010. [DOI] [PubMed] [Google Scholar]
- 7.Aune D, Navarro Rosenblatt DA, Chan DS, Abar L, Vingeliene S, Vieira AR, et al. Anthropometric factors and ovarian cancer risk: a systematic review and nonlinear dose-response meta-analysis of prospective studies. Int J Cancer. 2015;136:1888–98. doi: 10.1002/ijc.29207. [DOI] [PubMed] [Google Scholar]
- 8.World Cancer Research Fund/American Institute for Cancer Research. Continuous Update Project Expert Report, 2018. Diet, nutrition, physical activity and ovarian cancer. 2018; www.dietandcancerreport.org.
- 9.National Academies of Sciences, Engineering, and Medicine 2016. Ovarian Cancers: Evolving Paradigms in Research and Care. Washington, DC: The National Academies Press. 10.17226/21841. [PubMed]
- 10.Collaborative Group on Epidemiological Studies of Ovarian Cancer. Ovarian cancer and body size: individual participant meta-analysis including 25,157 women with ovarian cancer from 47 epidemiological studies. PLoS Med. 2012;9:e1001200. doi: 10.1371/journal.pmed.1001200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Olsen CM, Nagle CM, Whiteman DC, Ness R, Pearce CL, Pike MC, et al. Obesity and risk of ovarian cancer subtypes: evidence from the Ovarian Cancer Association Consortium. Endocr Relat Cancer. 2013;20:251–62. doi: 10.1530/ERC-12-0395. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Bandera EV, Alfano CM, Qin B, Kang D-W, Friel CP, Dieli-Conwright CM. Harnessing nutrition and physical activity for breast cancer prevention and control to reduce racial/ethnic cancer health disparities. Am Soc Clin Oncol Educ Book. 2021;41:1–17. doi: 10.1200/EDBK_321315. [DOI] [PubMed] [Google Scholar]
- 13.Bandera EV, Qin B, Moorman PG, et al. Obesity, weight gain, and ovarian cancer risk in African American women. Int J Cancer. 2016;139:593–600. doi: 10.1002/ijc.30115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Schildkraut JM, Peres LC, Bethea TN, Camacho F, Chyn D, Cloyd EK, et al. Ovarian Cancer in Women of African Ancestry (OCWAA) consortium: a resource of harmonized data from eight epidemiologic studies of African American and white women. Cancer Causes Control. 2019;30:967–78. doi: 10.1007/s10552-019-01199-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Schildkraut JM, Alberg AJ, Bandera EV, Barnholtz-Sloan J, Bondy M, Cote ML, et al. A multi-center population-based case-control study of ovarian cancer in African-American women: the African American Cancer Epidemiology Study (AACES) BMC Cancer. 2014;14:688. doi: 10.1186/1471-2407-14-688. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Peterson CE, Rauscher GH, Johnson TP, Kirschner CV, Freels S, Barrett RE, et al. The effect of neighborhood disadvantage on the racial disparity in ovarian cancer-specific survival in a large hospital-based study in cook county, Illinois. Front Public Health. 2015;3:8. doi: 10.3389/fpubh.2015.00008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Wu AH, Pearce CL, Tseng CC, Templeman C, Pike MC. Markers of inflammation and risk of ovarian cancer in Los Angeles county. Int J Cancer. 2009;124:1409–15. doi: 10.1002/ijc.24091. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Moorman PG, Palmieri RT, Akushevich L, Berchuck A, Schildkraut JM. Ovarian cancer risk factors in African-American and white women. Am J Epidemiol. 2009;170:598–606. doi: 10.1093/aje/kwp176. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Rosenberg L, Adams-Campbell L, Palmer JR. The Black Women’s Health Study: a follow-up study for causes and preventions of illness. J Am Med Women’s Assoc. 1995;50:56–8. [PubMed] [Google Scholar]
- 20.Kolonel LN, Henderson BE, Hankin JH, Nomura AM, Wilkens LR, Pike MC, et al. A multiethnic cohort in Hawaii and Los Angeles: baseline characteristics. Am J Epidemiol. 2000;151:346–57. doi: 10.1093/oxfordjournals.aje.a010213. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Hays J, Hunt JR, Hubbell FA, Anderson GL, Limacher M, Allen C, et al. The Women’s Health Initiative recruitment methods and results. Ann Epidemiol. 2003;13:S18–77. doi: 10.1016/S1047-2797(03)00042-5. [DOI] [PubMed] [Google Scholar]
- 22.World Health Organization. International classification of diseases for oncology (ICD-O). 3rd ed., 1st revision. World Health Organization; 2013. https://apps.who.int/iris/handle/10665/96612.
- 23.Peres LC, Cushing-Haugen KL, Köbel M, Harris HR, Berchuck A, Rossing MA, et al. Invasive epithelial ovarian cancer survival by histotype and disease stage. J Natl Cancer Inst. 2019;111:60–68. doi: 10.1093/jnci/djy071. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Clarke BA, Gilks B. Ovarian carcinoma: recent developments of tumour histological subtype. Can J Pathol. 2011;3:33–42. [Google Scholar]
- 25.Veroniki AA, Jackson D, Viechtbauer W, Bender R, Bowden J, Knapp G, et al. Methods to estimate the between-study variance and its uncertainty in meta-analysis. Res Synth Methods. 2016;7:55–79. doi: 10.1002/jrsm.1164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Cole SR, Chu H, Greenland S. Maximum likelihood, profile likelihood, and penalized likelihood: a primer. Am J Epidemiol. 2014;179:252–60. doi: 10.1093/aje/kwt245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Chang BH, Hoaglin DC. Meta-analysis of odds ratios: current good practices. Med Care. 2017;55:328–35. doi: 10.1097/MLR.0000000000000696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Bangdiwala SI, Bhargava A, O’Connor DP, Robinson TN, Michie S, Murray DM, et al. Statistical methodologies to pool across multiple intervention studies. Transl Behav Med. 2016;6:228–35. doi: 10.1007/s13142-016-0386-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Poorolajal J, Jenabi E, Masoumi SZ. Body mass index effects on risk of ovarian cancer: a meta-analysis. Asian Pac J Cancer Prev. 2014;15:7665–71. doi: 10.7314/APJCP.2014.15.18.7665. [DOI] [PubMed] [Google Scholar]
- 30.Fortner RT, Ose J, Merritt MA, Schock H, Tjønneland A, Hanson L, et al. Reproductive and hormone-related risk factors for epithelial ovarian cancer by histologic pathways, invasiveness and histologic subtypes: results from the EPIC cohort. Int J Cancer. 2015;137:1196–208. doi: 10.1002/ijc.29471. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Gates MA, Rosner BA, Hecht JL, Tworoger SS. Risk factors for epithelial ovarian cancer by histologic subtype. Am J Epidemiol. 2010;171:45–53. doi: 10.1093/aje/kwp314. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Yang HP, Trabert B, Murphy MA, Sherman ME, Sampson JN, Brinton LA, et al. Ovarian cancer risk factors by histologic subtypes in the NIH-AARP Diet and Health Study. Int J Cancer. 2012;131:938–48. doi: 10.1002/ijc.26469. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Bethea TN, Palmer JR, Adams-Campbell LL, Rosenberg L. A prospective study of reproductive factors and exogenous hormone use in relation to ovarian cancer risk among Black women. Cancer Causes Control. 2017;28:385–91. doi: 10.1007/s10552-016-0840-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Peres LC, Risch H, Terry KL, Webb PM, Goodman MT, Wu AH, et al. Racial/ethnic differences in the epidemiology of ovarian cancer: a pooled analysis of 12 case-control studies. Int J Epidemiol. 2018;47:1011. doi: 10.1093/ije/dyy054. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Trabert B, Brinton LA, Anderson GL, Pfeiffer RM, Falk RT, Strickler HD, et al. Circulating estrogens and postmenopausal ovarian cancer risk in the women’s health initiative observational study. Cancer Epidemiol Biomark Prev. 2016;25:648–56. doi: 10.1158/1055-9965.EPI-15-1272-T. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.CancerDisparitiesProgressReport.org [Internet]. Philadelphia: American Association for Cancer Research; ©2022 [cited August 2022]. http://www.CancerDisparitiesProgressReport.org/.
- 37.Dixon SC, Nagle CM, Thrift AP, Pharoah PD, Pearce CL, Zheng W, et al. Adult body mass index and risk of ovarian cancer by subtype: a Mendelian randomization study. Int J Epidemiol. 2016;45:884–95. doi: 10.1093/ije/dyw158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Bandera EV, Fay SH, Giovannucci E, Leitzmann MF, Marklew R, McTiernan A, et al. The use and interpretation of anthropometric measures in cancer epidemiology: a perspective from the world cancer research fund international continuous update project. Int J Cancer. 2016;139:2391–7. doi: 10.1002/ijc.30248. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.