Abstract
This case-control study sought to confirm the exploratory finding of an association between polycystic ovary syndrome and risk of pancreatic cancer.
Given that only 11% of patients with pancreatic cancer survive 5 years after diagnosis,1 clinical surveillance is recommended for individuals at high risk.2 Although pancreatic cancer has relatively few established risk factors, knowledge of these can help identify other subgroups of individuals who may benefit from increased education and clinical vigilance. Polycystic ovary syndrome (PCOS) is a common endocrine disorder3 associated with several carcinogenic processes and cancers.4 A 2019 Swedish register study reported that women with PCOS had 3.4-fold increased risk of pancreatic cancer,4 but it included only 5 women with pancreatic cancer who had PCOS. We sought to confirm this exploratory finding.
Methods
In this case-control study, we used data from the Memorial Sloan Kettering Cancer Center Pancreatic Tumor Registry.5 Participants were female, English speaking, and 21 years of age or older. Cases (n = 446) had pathologically or cytologically confirmed pancreatic adenocarcinoma, and controls (n = 209) were visitors accompanying patients to other Memorial Sloan Kettering Cancer Center clinics or spouses of patients with pancreatic cancer with no personal history of cancer. We followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Women were asked, “Have you ever had polycystic ovaries?” If they answered “yes,” then they reported their age at diagnosis and any treatment. We considered pancreatic cancer diagnosed less than 1 year after PCOS (n = 1) as occurring in the unexposed period.4 We collected covariate information, including factors potentially associated with PCOS or pancreatic cancer. This study was approved by the institutional review board of the Memorial Sloan Kettering Cancer Center. Participants provided written informed consent.
We used multivariable logistic regression to estimate odds ratios (ORs) and 95% CIs for the association between PCOS and risk of pancreatic cancer. We used minimally adjusted (age and race and ethnicity) and fully adjusted (age, race and ethnicity, body mass index [BMI], and estrogen use) models. Adjustment for additional covariates (Table 1) did not appreciably alter estimates. We separately assessed type 2 diabetes given its potential mediating role.6 We performed stratified analyses and tested for multiplicative interactions by menopausal status, type 2 diabetes, and obesity. We conducted sensitivity analyses, excluding those with pancreatic cancer diagnosed less than 5 years after PCOS (n = 6), pancreatitis (n = 55), and familial cases of pancreatic cancer (n = 86). Statistical tests were 2-tailed with α = .05. Statistical analyses were performed using SAS, version 9.4 (SAS Institute) and RStudio, version 4.2.1 (RStudio, PBC). Analyses were conducted in multiple intervals: from June to August 2019 and then updated and completed in June 2022.
Table 1. Descriptive Characteristics of Study Population.
| Characteristic | No. (%) | |
|---|---|---|
| Cases (n = 446) | Controls (n = 209) | |
| Age at diagnosis or enrollment, mean (SD), y | 63.8 (11.3) | 57.7 (11.2) |
| Race and ethnicity | ||
| White | 395 (88) | 183 (88) |
| Black | 25 (6) | 11 (5) |
| Asian | 14 (3) | 8 (4) |
| Othera | 12 (3) | 7 (3) |
| Education | ||
| High school or less | 141 (32) | 47 (23) |
| Some college or college | 178 (40) | 97 (46) |
| Graduate | 127 (28) | 65 (31) |
| Smoking status | ||
| >100 Cigarettes | 215 (48) | 106 (51) |
| ≤100 Cigarettes | 231 (52) | 103 (49) |
| BMI | ||
| Underweight (<18.5) | 33 (7) | 8 (4) |
| Normal (18.5 to <25) | 216 (48) | 92 (44) |
| Overweight (25 to <30) | 96 (22) | 70 (34) |
| Obese (≥30) | 47 (11) | 24 (11) |
| Missing | 54 (12) | 15 (7) |
| Diabetes | ||
| Yes | 116 (26) | 11 (5) |
| No | 330 (74) | 198 (95) |
| History of hysterectomy | ||
| Yes | 98 (22) | 49 (23) |
| No | 348 (78) | 160 (77) |
| History of oophorectomy | ||
| Yes | 83 (19) | 41 (20) |
| No | 363 (81) | 168 (80) |
| Menopause | ||
| Yes | 394 (88) | 163 (78) |
| No | 52 (12) | 46 (22) |
| Postmenopausal estrogen use | ||
| Yes | 120 (27) | 77 (37) |
| No | 321 (72) | 130 (62) |
| Missing | 5 (1) | 2 (1) |
Abbreviation: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared).
Includes American Indian or Alaska Native, Native Hawaiian or Other Pacific Islander, and other.
Results
We found that PCOS was positively associated with risk of pancreatic cancer after adjusting for age, race and ethnicity, BMI, and estrogen use (OR, 1.88; 95% CI, 1.02-3.46) (Table 2). The association was slightly attenuated after adjusting for potential mediator type 2 diabetes (OR, 1.78; 95% CI, 0.95-3.34). We did not observe statistically significant interactions. Excluding those with pancreatic cancer diagnosed less than 5 years after PCOS, pancreatitis, and a history of familial pancreatic cancer did not materially alter estimates.
Table 2. Association Between Polycystic Ovary Syndrome and Pancreatic Cancer Risk.
| Polycystic ovary syndrome | No. (%) | Odds ratio (95% CI) | |||
|---|---|---|---|---|---|
| Cases (n = 446) | Controls (n = 209) | Minimally adjusteda | Fully adjustedb | Diabetes adjustedc | |
| Yes | 49 (11) | 18 (9) | 1.50 (0.83-2.69) | 1.88 (1.02-3.46) | 1.78 (0.95-3.34) |
| No | 397 (89) | 191 (91) | 1 [Reference] | 1 [Reference] | 1 [Reference] |
For age and race and ethnicity.
For age, race and ethnicity, body mass index (BMI), and estrogen. Multiple imputation used to fill in missing values for BMI and estrogen.
For age, race and ethnicity, BMI, estrogen, and diabetes. Multiple imputation used to fill in missing values for BMI and estrogen.
Discussion
A diagnosis of PCOS was associated with a 1.9-fold higher risk of pancreatic cancer, confirming the exploratory findings of Yin et al.4 This association was independent of BMI, largely not driven by type 2 diabetes, and robust to additional adjustment for several covariates and sensitivity analyses. These data suggest some individuals may have unknown metabolic derangements that may underly the development of both conditions.
To our knowledge, this study is the second to examine the association between PCOS and risk of pancreatic cancer and the first with an adequate sample size. We designed our study to minimize the inherent limitations of the case-control design. Results were similar after restricting analysis to cases with a time from PCOS to pancreatic cancer diagnosis of 5 years or more, which suggests that reverse causation and detection bias did not materially influence the findings. Although differential recall in cases and controls is a concern, self-reported PCOS or age at PCOS diagnosis is likely more accurate compared with self-reported lifestyle factors. Hospital-based controls may have introduced selection bias; however, most controls were visitors accompanying patients and were not selected based on health conditions that may be inadvertently associated with pancreatic cancer.
In this case-control study, PCOS was strongly associated with increased risk of pancreatic cancer. Combined with previous findings, diagnosis of PCOS may warrant increased education and clinical vigilance for pancreatic cancer. Prospective studies are needed to examine underlying biologic mechanisms and confirm our findings.
References
- 1.Surveillance Epidemiology, and End Results Program. Cancer stat facts: pancreatic cancer. National Cancer Institute. Accessed June 27, 2022. https://seer.cancer.gov/statfacts/html/pancreas.html
- 2.Goggins M, Overbeek KA, Brand R, et al. ; International Cancer of the Pancreas Screening (CAPS) consortium . Management of patients with increased risk for familial pancreatic cancer: updated recommendations from the International Cancer of the Pancreas Screening (CAPS) Consortium. Gut. 2020;69(1):7-17. doi: 10.1136/gutjnl-2019-319352 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Sirmans SM, Pate KA. Epidemiology, diagnosis, and management of polycystic ovary syndrome. Clin Epidemiol. 2013;6:1-13. doi: 10.2147/CLEP.S37559 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Yin W, Falconer H, Yin L, Xu L, Ye W. Association between polycystic ovary syndrome and cancer risk. JAMA Oncol. 2019;5(1):106-107. doi: 10.1001/jamaoncol.2018.5188 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Olson SH, Xu Y, Herzog K, et al. Weight loss, diabetes, fatigue, and depression preceding pancreatic cancer. Pancreas. 2016;45(7):986-991. doi: 10.1097/MPA.0000000000000590 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rubin KH, Glintborg D, Nybo M, Abrahamsen B, Andersen M. Development and risk factors of type 2 diabetes in a nationwide population of women with polycystic ovary syndrome. J Clin Endocrinol Metab. 2017;102(10):3848-3857. doi: 10.1210/jc.2017-01354 [DOI] [PubMed] [Google Scholar]