Abstract
Background:
Regular physical inactivity may increase ovarian cancer risk, but few studies have investigated whether this association is similar among Black and White women.
Methods:
In a pooled nested case-control within the Ovarian Cancer in Women of African Ancestry consortium, logistic regression models evaluated regular recreational physical inactivity with risk of epithelial ovarian cancer among Black (223 cases; 1,472 controls) and White women (985 cases; 6,212 controls) enrolled in four cohort studies. Models were further stratified by histological type.
Results:
Regular physical inactivity was not associated with risk of overall ovarian cancer among Black (OR=1.16, 95% confidence interval (CI): 0.83–1.61) or White women (OR=1.03, 95% CI: 0.87–1.23). We did not detect associations according to histological type.
Conclusions:
Physical inactivity was not associated with ovarian cancer among Black or White women in a consortium of cohort studies.
Impact:
These results are counter to case control-based studies and emphasize the complexity of investigating physical activity prospectively.
Introduction
Risk and survival of ovarian cancer differ by race and ethnicity.1 The relationship between physical activity and ovarian cancer remains unclear, but modifiable factors have potential for primary prevention and to reduce disparities.2 Due to the differences in prevalence of recreational physical activity by race and ethnicity,3 investigation of inactivity as a potential contributor to ovarian cancer racial disparities is warranted.
To our knowledge, only one study has assessed the influence of race on the association between recreational physical inactivity and ovarian cancer risk. An Ovarian Cancer Association Consortium (OCAC) study reported higher ovarian cancer risk among inactive White women (odds ratio; OR=1.27, 95% confidence interval; CI: 1.17–1.37) and Black women (OR=1.68, 95% CI: 1.01–2.80).4 However, OCAC is comprised of data from case-control studies, which could be subject to recall bias and selection bias.
We evaluate regular recreational physical inactivity with ovarian cancer risk among cohort studies in the Ovarian Cancer in Women of African Ancestry (OCWAA) consortium for Black and White women.5 We hypothesize that regular physical inactivity is associated with higher ovarian cancer risk among Black and White women.
Methods
Study population.
The OCWAA consortium is comprised of eight epidemiologic studies, described in detail elsewhere.5 Briefly, data from questionnaires, medical records, and cancer registries were harmonized across studies. In the present pooled nested case-control study, we used the individual-level data from four cohort studies that collected data on regular recreational physical inactivity: Black Women’s Health Study (BWHS), Southern Community Cohort Study (SCCS), Multiethnic Cohort Study (MEC), and Women’s Health Initiative (WHI). Across the four cohorts, controls were matched to cases on race, age of diagnosis, and last completed questionnaire before diagnosis. Investigators from the corresponding studies obtained written informed consent from participants, and each individual study as well as the OCWAA consortium received approval from institutional review boards. Each study involving human subjects was conducted in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Ovarian cancer diagnosis.
Data on invasive epithelial ovarian cancer diagnosis were obtained from cancer registries or medical records. Morphology and grade were combined into the following histotypes: high-grade serous, low-grade serous, endometrioid, clear cell, mucinous, carcinosarcoma, and other epithelial tumors.6
Assessment of physical inactivity.
All cohorts collected physical activity information across multiple follow-up visits (range:1993–2015) except for SCCS, which collected only at baseline. Women were classified as having regular recreational physical inactivity if they reported no regular recent weekly moderate or vigorous recreational activity in all queried time periods prior to the reference date. The reference date was the date of diagnosis for cases and the date of the interview for controls. All questionnaires prior to the year of diagnosis were used for classification. Physical inactivity was dichotomized, with the participants reporting “yes” to regular recreational activity being the referent category; additional detail provided in Supplemental Table 1.
Covariates.
Harmonized covariates that met confounder criteria included: age, study, education, oral contraceptive use, menopausal status, nulliparity, tubal ligation status, history of postmenopausal hormone use, smoking status, and family history of breast/ovarian cancer.
Statistical analysis.
Unconditional logistic regression models adjusted for the aforementioned covariates to estimate the odds ratios (ORs) and corresponding 95% confidence intervals (CIs) to examine the association between physical inactivity and ovarian cancer risk among Black and White participants. Logistic regression models were stratified according to histological type: high-grade serous vs non-high-grade serous tumors. We additionally adjusted for body mass index (BMI), despite it likely being a mediator, as a sensitivity analysis. All analyses were conducted in SAS version 9.4.
Results
The sample includes 223 Black cases and 1,472 Black controls along with 985 White cases and 6,212 White controls. Table 1 shows that Black women had a higher prevalence of physical inactivity than White women.
Table 1.
Descriptive statistics for chronic physical inactivity-ovarian cancer association study in the OCWAA consortium.
| Black
participants N=1,695 |
White participants
N=7,197 |
|||||
|---|---|---|---|---|---|---|
| N (%) or Mean (SD) | Cases N=223 |
Controls N=1,472 |
P | Cases N=985 |
Controls N=6,212 |
P |
| Study site | 0.998 | 0.841 | ||||
| BWHS | 78 (35.0%) | 514 (34.9%) | - | - | ||
| MEC | 60 (26.9%) | 388 (26.4%) | 126 (12.8%) | 754 (12.1%) | ||
| SCCS | 46 (20.6%) | 307 (20.9%) | 31 (3.1%) | 200 (3.2%) | ||
| WHI | 39 (17.5%) | 263 (17.9%) | 828 (84.1%) | 5,258 (84.6%) | ||
| Chronic physical inactivity | 0.205 | 0.976 | ||||
| No | 144 (64.6%) | 1,013 (68.8%) | 784 (79.6%) | 4,947 (79.6%) | ||
| Yes | 79 (35.4%) | 459 (31.2%) | 201 (20.4%) | 1,265 (20.4%) | ||
| Sitting time | 0.769 | 0.769 | ||||
| ≤8 hours | 41 (56.9%) | 383 (58.7%) | 41 (56.9%) | 383 (58.7%) | ||
| >8 hours | 31 (43.1%) | 269 (41.3%) | 31 (43.1%) | 269 (41.3%) | ||
| Age, years | 63.6 (12.7) | 61.5 (12.8) | 0.027 | 72.0 (8.0) | 71.5 (8.3) | 0.081 |
| Education | 0.939 | 0.869 | ||||
| ≤High school graduate/GED | 78 (35.0%) | 489 (33.2%) | 208 (21.1%) | 1,357 (21.8%) | ||
| Some college | 76 (34.1%) | 505 (34.3%) | 339 (34.4%) | 2,152 (34.6%) | ||
| College graduate | 27 (12.1%) | 196 (13.3%) | 141 (14.3%) | 834 (13.4%) | ||
| Graduate/professional school | 42 (18.8%) | 282 (19.2%) | 297 (30.2%) | 1,869 (30.1%) | ||
| Body mass index, kg/m 2 | 0.560 | 0.013 | ||||
| <25 | 37 (16.6%) | 291 (19.8%) | 443 (45.0%) | 2,475 (39.8%) | ||
| 25–29.9 | 72 (32.3%) | 495 (33.6%) | 293 (29.7%) | 2,078 (33.5%) | ||
| 30–34.9 | 63 (28.3%) | 368 (25.0%) | 166 (16.9%) | 1,047 (16.9%) | ||
| 35+ | 51 (22.9%) | 318 (21.6%) | 83 (8.4%) | 612 (9.9%) | ||
| Family history of breast cancer | 0.001 | 0.924 | ||||
| Yes | 38 (17.0%) | 145 (9.9%) | 140 (14.2%) | 890 (14.3%) | ||
| No | 185 (83.0%) | 1,327 (90.1%) | 845 (85.8%) | 5,322 (85.7%) | ||
| Family history of ovarian cancer | 0.982 | 0.019 | ||||
| Yes | 6 (2.7%) | 40 (2.7%) | 37 (3.8%) | 153 (2.5%) | ||
| No | 217 (97.3%) | 1,432 (97.3%) | 948 (96.2%) | 6,059 (97.5%) | ||
| Smoking status | 0.626 | 0.929 | ||||
| Current smoker | 34 (15.2%) | 243 (16.5%) | 59 (6.0%) | 383 (6.2%) | ||
| Never smoker | 120 (53.8%) | 741 (50.3%) | 475 (48.2%) | 3,023 (48.7%) | ||
| Former smoker | 69 (30.9%) | 488 (33.2%) | 451 (45.8%) | 2,806 (45.2%) | ||
| Nulliparous | 0.027 | 0.049 | ||||
| Yes | 49 (22.0%) | 236 (16.0%) | 191 (19.4%) | 1,046 (16.8%) | ||
| No | 174 (78.0%) | 1,236 (84.0%) | 794 (80.6%) | 5,166 (83.2%) | ||
| Postmenopausal hormone use duration | 0.616 | <0.001 | ||||
| Never | 143 (64.4%) | 978 (66.6%) | 320 (32.5%) | 2,194 (35.3%) | ||
| <5 years | 43 (19.4%) | 289 (19.7%) | 161 (16.3%) | 1,296 (20.9%) | ||
| 5+ years | 36 (16.2%) | 198 (13.5%) | 504 (51.2%) | 2,712 (43.7%) | ||
| Unknown | 0 (0.0%) | 4 (0.3%) | 0 (0.0%) | 5 (0.1%) | ||
| Oral contraceptive use | 0.150 | 0.988 | ||||
| Yes | 103 (46.2%) | 756 (51.4%) | 380 (38.6%) | 2,395 (38.6%) | ||
| No | 120 (53.8%) | 716 (48.6%) | 605 (61.4%) | 3,817 (61.4%) | ||
| Tubal ligation | 0.193 | 0.004 | ||||
| Yes | 167 (74.9%) | 1,040 (70.7%) | 850 (86.3%) | 5,134 (82.6%) | ||
| No | 56 (25.1%) | 432 (29.3%) | 135 (13.7%) | 1,078 (17.4%) | ||
| Age at menarche | 0.341 | 0.543 | ||||
| <11 years | 30 (13.5%) | 135 (9.2%) | 51 (5.2%) | 393 (6.3%) | ||
| 11–12 years | 93 (41.7%) | 609 (41.4%) | 413 (41.9%) | 2,494 (40.1%) | ||
| 13–14 years | 75 (33.6%) | 549 (37.3%) | 432 (43.9%) | 2,720 (43.8%) | ||
| 15–16 years | 21 (9.4%) | 149 (10.1%) | 80 (8.1%) | 552 (8.9%) | ||
| 17+ years | 4 (1.8%) | 30 (2.0%) | 9 (0.9%) | 53 (0.9%) | ||
| Hysterectomy | 0.007 | 0.774 | ||||
| Yes | 136 (61.5%) | 1,035 (70.5%) | 711 (72.4%) | 4,517 (72.8%) | ||
| No | 85 (38.5%) | 433 (29.5%) | 271 (27.6%) | 1,684 (27.2%) | ||
| Diabetes | 0.042 | 0.759 | ||||
| Yes | 191 (85.7%) | 1,176 (79.9%) | 941 (95.6%) | 5,925 (95.4%) | ||
| No | 32 (14.3%) | 296 (20.1%) | 43 (4.4%) | 285 (4.6%) | ||
| Menopausal status | 0.127 | 0.595 | ||||
| Premenopausal | 43 (19.3%) | 352 (23.9%) | 12 (1.2%) | 89 (1.4%) | ||
| Postmenopausal | 180 (80.7%) | 1120 (76.1%) | 973 (98.8%) | 6,123 (98.6%) | ||
| Stage | <0.001 | <0.001 | ||||
| Localized | 29 (14.6%) | 0 (0.0%) | 95 (9.8%) | 0 (0.0%) | ||
| Regional | 23 (11.6%) | 0 (0.0%) | 128 (13.2%) | 0 (0.0%) | ||
| Distant | 146 (73.7%) | 0 (0.0%) | 746 (77.0%) | 0 (0.0%) | ||
| Structural missing | 0 (0.0%) | 1,472 (100.0%) | 0 (0.0%) | 6,212 (100.0%) | ||
| Histological type | <0.001 | <0.001 | ||||
| High-grade serous | 124 (55.6%) | 0 (0.0%) | 661 (67.1%) | 0 (0.0%) | ||
| Low-grade serous | 1 (0.4%) | 0 (0.0%) | 12 (1.2%) | 0 (0.0%) | ||
| Endometrioid | 17 (7.6%) | 0 (0.0%) | 35 (3.6%) | 0 (0.0%) | ||
| Clear cell | 3 (1.3%) | 0 (0.0%) | 43 (4.4%) | 0 (0.0%) | ||
| Mucinous | 12 (5.4%) | 0 (0.0%) | 13 (1.3%) | 0 (0.0%) | ||
| Carcinosarcoma | 6 (2.7%) | 0 (0.0%) | 31 (3.1%) | 0 (0.0%) | ||
| Other specified epithelial | 60 (26.9%) | 0 (0.0%) | 190 (19.3%) | 0 (0.0%) | ||
| Structural missing | 0 (0.0%) | 1,472 (100.0%) | 0 (0.0%) | 6,212 (100.0%) | ||
We did not detect associations between regular physical inactivity and risk of overall ovarian cancer among Black or White women (Table 2). For Black women, the OR was 1.16 (95% CI: 0.83–1.61). For White women, the OR was 1.03 (95% CI: 0.87–1.23). Results did not differ when assessed according to histological type or with adjustment for BMI.
Table 2.
Associations of chronic physical inactivity 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.
| Histological type | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
|
|
|||||||||||
| Invasive1 | High-grade serous1 | Non-high grade serous1 | |||||||||
|
| |||||||||||
| Cases | Controls | Adjusted Model1,2 | BMI Model3 | Cases | Controls | Adjusted Model1 | Cases | Controls | Adjusted Model1 |
P
heterogeneity |
|
| Black Participants (BWHS, MEC, SCCS, WHI) | |||||||||||
|
| |||||||||||
| Chronic physical inactivity | |||||||||||
|
| |||||||||||
| No | 144 | 1,013 | 1.00 | 1.00 | 84 | 1,013 | 1.00 | 60 | 1,013 | 1.00 | 0.91 |
| Yes | 79 | 459 | 1.16 (0.83–1.61) | 1.13 (0.81–1.58) | 40 | 459 | 1.24 (0.78–1.99) | 39 | 459 | 1.27 (0.76–2.13) | |
|
| |||||||||||
| White Participants (MEC, SCCS, WHI) | |||||||||||
|
| |||||||||||
| Chronic physical inactivity | |||||||||||
|
| |||||||||||
| No | 784 | 4,947 | 1.00 | 1.00 | 535 | 4,947 | 1.00 | 249 | 4,947 | 1.00 | 0.94 |
| Yes | 201 | 1,265 | 1.03 (0.87–1.23) | 1.04 (0.87–1.24) | 126 | 1,265 | 1.01 (0.82–1.26) | 75 | 1,265 | 1.02 (0.77–1.37) | |
Model adjusted for study, age, education, nulliparity, oral contraceptive use, menopausal status, tubal ligation status, postmenopausal hormone use (ever/never), smoking status, and family history of breast/ovarian cancer.
Random effects by site are included in the main model only.
Model additionally adjusted for BMI.
Discussion
In this cohort-only consortium study, we did not find regular physical inactivity was associated with risk of invasive epithelial ovarian cancer for either Black or White women. Results are counter to those from a prior OCAC study, which reported a positive association between physical inactivity and ovarian cancer risk, with a larger effect size among Black women compared to White women.4 Similar to our results, a meta-analysis found associations attenuate and become null when restricted to cohort studies.7
The differences in results between the current study and the OCAC study presents a significant question about the validity of prior studies on physical inactivity and ovarian cancer risk.4 While OCAC has a larger sample size, physical activity is a notoriously difficult exposure to measure, and misclassification of physical inactivity may help explain study differences.8 Exposure misclassification among cohort studies would likely be non-differential and bias toward the null, whereas misclassification among case-control studies could be differential due to recall bias, and thus potentially lead to inflated effects.
Strengths of this study include the relatively large sample size, particularly for Black ovarian cancer cases, and the restriction to cohort studies, lending a strong study design. Limitations include potential misclassification of regular physical inactivity. Physical inactivity was dichotomized due to harmonization challenges, thus preventing us from evaluating light, moderate, and vigorous activity, and non-linear effects. Early-life physical activity may be influential for cancer risk,9 however information on physical activity during earlier life periods was not routinely collected. Further, due to small sample of premenopausal women, we were unable to evaluate by menopausal status.
Physical activity is protective for many chronic conditions and may lead to metabolic improvements that reduces ovarian cancer risk.2 Further research is needed, particularly among cohorts with racially and ethnically diverse samples.
Supplementary Material
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.
Funding information:
This study is supported by the National Institutes of Health (R01-CA207260 to J.M. Schildkraut and L. Rosenberg and K01-CA212056 to T.N. Bethea). AACES was funded by NCI (R01-CA142081 to J.M. Schildkraut); BWHS is funded by NIH (R01-CA058420, UM1-CA164974, and U01-CA164974 to L. Rosenberg); CCCCS was funded by NIH/NCI (R01-CA61093 to K.A. Rosenblatt); LACOCS was funded by NCI (P01-CA17054 to M.C. Pike, R01-CA058598 to M.R. Goodman and A.H. Wu, and Cancer Center Core Grant P30-CA014089 to B.E. Henderson and A.H. Wu) and by the California Cancer Research Program (2II0200 to A.H. Wu); SCCS is supported by U01 CA202979 (to W. Zheng); and NCOCS was funded by NCI (R01-CA076016 to J.M. 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 G.L. Anderson) and NIH/NHLBI-CSB-WH-2016-01-CM.
Footnotes
Conflicts: The authors declare no potential conflicts of interest.
Data availability.
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1.Park HK, Ruterbusch JJ, Cote ML. Recent Trends in Ovarian Cancer Incidence and Relative Survival in the United States by Race/Ethnicity and Histologic Subtypes. Cancer Epidemiology, Biomarkers & Prevention. 2017;26(10):1511–1518. doi: 10.1158/1055-9965.Epi-17-0290 [DOI] [Google Scholar]
- 2.Gapstur SM, Drope JM, Jacobs EJ, Teras LR, McCullough ML, Douglas CE, et al. A blueprint for the primary prevention of cancer: Targeting established, modifiable risk factors. CA Cancer J Clin Nov 2018;68(6):446–470. doi: 10.3322/caac.21496 [DOI] [PubMed] [Google Scholar]
- 3.Elgaddal N, Kramarow EA, Reuben C. Physical Activity Among Adults Aged 18 and Over: United States, 2020. NCHS Data Brief. Aug 2022;(443):1–8. [Google Scholar]
- 4.Cannioto R, LaMonte MJ, Risch HA, Hong CC, Sucheston-Campbell LE, Eng KH, et al. Chronic Recreational Physical Inactivity and Epithelial Ovarian Cancer Risk: Evidence from the Ovarian Cancer Association Consortium. Cancer Epidemiol Biomarkers Prev Jul 2016;25(7):1114–24. doi: 10.1158/1055-9965.Epi-15-1330 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.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. Sep 2019;30(9):967–978. doi: 10.1007/s10552-019-01199-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Young R WHO classification of tumours of female reproductive organs. IARC; 2014. [Google Scholar]
- 7.Zhong S, Chen L, Lv M, Ma T, Zhang X, Zhao J. Nonoccupational physical activity and risk of ovarian cancer: a meta-analysis. Tumour Biol Nov 2014;35(11):11065–73. doi: 10.1007/s13277-014-2385-z [DOI] [PubMed] [Google Scholar]
- 8.Ferrari P, Friedenreich C, Matthews CE. The role of measurement error in estimating levels of physical activity. Am J Epidemiol Oct 1 2007;166(7):832–40. doi: 10.1093/aje/kwm148 [DOI] [PubMed] [Google Scholar]
- 9.Huang T, Eliassen AH, Hankinson SE, Okereke OI, Kubzansky LD, Wang M, et al. A prospective study of leisure-time physical activity and risk of incident epithelial ovarian cancer: Impact by menopausal status. Int J Cancer. Feb 15 2016;138(4):843–52. doi: 10.1002/ijc.29834 [DOI] [PMC free article] [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.