Abstract
Background:
Meningiomas, benign central nervous system tumors, express progesterone and estrogen receptors. Their proliferation has been associated with hormonal and demographic factors, including female sex, obesity, and race. Prior studies on oral contraceptive pill (OCP) use and meningioma risk have been limited in their analysis of the modifying effects of race, obesity, and polycystic ovary syndrome (PCOS).
Objectives:
To assess the association between combined OCP use and meningioma development, and to assess how race, obesity, and PCOS influence this relationship.
Design:
Retrospective cohort study using aggregated electronic health record data in Epic Cosmos.
Methods:
Women aged 13–50 from 2005 to 2023 with and without combined OCP use were identified. Patients with history of radiation, neurofibromatosis 2, progestin-only contraceptive use, and hormone replacement therapy were excluded. The cohort was then stratified by PCOS status, obesity, and race.
Results:
Combined OCP users saw a 40% lower risk of meningioma compared to non-users. PCOS was associated with a bidirectional effect on meningioma risk, modified by obesity. Among patients with obesity, those with PCOS had a 30% lower risk of developing meningioma compared to those without PCOS. Among non-obese patients, those with PCOS had a 108% greater risk of developing meningioma compared to those without PCOS. After adjusting for both PCOS and obesity, women with a history of combined OCP usage had 42% reduced odds of developing meningioma. When stratified by race, combined OCP use was associated with 47% decreased risk in White patients, 34% lower risk in Black patients, and 28% lower risk in Asian patients. Controlling for race overall, combined OCP use remained significantly protective, with 43% reduced odds of meningioma development.
Conclusion:
Findings suggest a potential protective association between combined OCP use and meningioma that remained significant after controlling for obesity, PCOS status, and race. Additionally, this study found that meningioma risk in patients with PCOS differed based on obesity status.
Introduction
Meningiomas, benign tumors of the meninges, are the most common primary brain tumor, comprising 38.3% of all primary brain and central nervous system tumors in the United States from 2013 to 2017. 1 These tumors have been found to commonly express progesterone and estrogen receptors. 2 Risk of meningioma has been associated with hormonal and demographic factors, including female sex, obesity, and Black race.1,3,4 Additionally, tumor proliferation has been observed in patients using estrogen-only hormone replacement therapy (HRT), as well as during states of elevated progesterone, including pregnancy and the luteal phase of the menstrual cycle, suggesting both estrogen and progesterone influences on meningioma proliferation.2,5
Oral contraceptive pills (OCPs) are used by an estimated 14% of women aged 15–49. 6 Combined OCPs, which contain estrogen and progestin, function by way of altering sex hormone levels throughout the menstrual cycle to suppress ovulation and thicken cervical mucus, preventing pregnancy.
Epidemiological studies have found an association between progestogen use and meningioma risk. Namely, use of the French National Health Data System (SNDS) has enabled studies on odds of meningioma associated with various progestogen formulations. Roland et al. 7 found increased odds for meningioma associated with prolonged use of medrogestone, medroxyprogesterone acetate, and promegestone use, in addition to previously established increased odds of meningioma associated with cyproterone acetate, nomegestrol acetate, and chlormadinone acetate use. However, another SNDS study found no statistically significant increased odds of meningioma associated with use of combined levonorgestrel-estrogen contraception. 8
Notably, these epidemiological studies of OCP use in association with meningioma development have chiefly been conducted in European populations, in which oral contraceptive formulations and prescribing patterns, population racial composition, and obesity rates vary from the United States. Black individuals comprise approximately 14.4% of the U.S. population, whereas only 3.2% of the French population used in the SNDS identifies as Black.9,10 Similarly, obesity is more prevalent in the United States, affecting roughly 41.9% of adults in 2021, compared to approximately 17%–20% in France in 2020.11,12 These differences in population structure and health profiles may limit the generalizability of European findings to a U.S. context.
Polycystic ovary syndrome (PCOS) is often associated with obesity in women of reproductive age. PCOS is a common endocrine disorder characterized by hyperandrogenism, hypothalamic–pituitary–ovarian axis dysfunction, and cystic ovaries secondary to anovulatory cycles. The pathophysiology of PCOS is multifactorial and complex but is commonly associated with insulin resistance, which occurs independent of obesity, but is also exacerbated by it. Reflexive compensatory hyperinsulinemia stimulates excessive androgen production from ovarian theca cells and reduces hepatic production of sex hormone binding globulin, increasing free circulating androgens. These hormonal imbalances disrupt GnRH pulsatility, which leads to an increased lutenizing hormone/follicle-stimulating hormone ratio and inhibited follicular development. The resulting anovulation contributes to the estrogen-predominant state of PCOS.13–15 Though obesity is an established risk factor for meningioma, its association with meningioma risk in the context of PCOS diagnosis is not well documented. 3
This study aims to investigate the relationship between altered estrogen and progesterone states in premenopausal women, including combined OCP use and PCOS, and subsequent diagnosis of meningioma later in life, particularly within the context of a large and demographically diverse population.
Methods
A cohort study using the Epic Cosmos database included female patients aged 13–50 years between January 1, 2005 and December 31, 2023 with at least two encounters within a 2-year period. Combined OCP exposure was defined by having a prescription for a combined estrogen–progestin oral contraceptive during this timeframe, along with at least one encounter for a medication refill, regardless of the medication type. Patients using progestin-only oral contraceptives, depot injections, and long-acting reversible contraception formulations were not included in the exposure group.
All eligible patients meeting the inclusion criteria during the study period were included in the analysis. Patients diagnosed with neurofibromatosis type 2 (International Classification of Diseases, 10th Revision (ICD-10): Q85.2) and patients with a history of irradiation (ICD-10: Z92.3) prior to meningioma development were excluded. Additionally, patients with a history of HRT were excluded, given the established association between HRT and meningioma development. 16 HRT exposure was identified using standardized medication aggregators within the Epic Cosmos network, including: “hormone therapy for dyspareunia (painful intercourse) associated with menopause,” “hormone therapy for menopause-associated vaginal atrophy,” “estrogens,” “progesterone, micronized,” “medroxyprogesterone acetate,” and “medroxyprogesterone acetate, micronized.”
Patients were stratified by those who did and did not have a prior history of combined OCP use, a diagnosis of PCOS, a body mass index (BMI) greater than 30.0, and by race. For patients with both meningioma diagnosis and OCP use, the search strategy necessitated that combined OCP use preceded the diagnosis of meningioma. Age at diagnosis of meningioma, age during use of OCP, and duration of OCP use were not available.
To assess internal validity of the query structure, similar parameters were reapplied in a search for the incidence of hepatocellular adenomas (ICD-10: D13.4) and malignant ovarian neoplasms (ICD-10: C56*) in patients with and without a history of OCP use, as these cancers both have well-documented correlations with OCP use.17,18
Cohorts were analyzed by comparing meningioma incidence rates and relative risks (RRs) between groups. Statistical significance was assessed using chi-square goodness-of-fit tests, and significance was defined as p < 0.001. Given the aggregate-only format of the Cosmos dataset, analyses requiring patient-level data, such as multivariable regression, could not be performed. Instead, Cochran–Mantel–Haenszel tests were used to perform stratified analyses, generating odds ratios (ORs) based on mutually exclusive categorical predictor variables that serve as a substitute for multivariate regression to account for variable effects in Cosmos aggregated data.
The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology guidelines. 19
Carle Institutional Review Board and the Carle Human Subject Protection Program waived the need for further International Review Board approval, ethics approval, and patient consent for the collection, analysis and publication of the retrospectively obtained and anonymized Epic Cosmos data for this non-interventional study. Consent to participate was not applicable.
Results
Of the 47,262,625 female patients included in this study, 4,663,249 (9.87%) had a documented history of combined OCP use, of whom 4023 (0.086%) were subsequently diagnosed with meningioma. In comparison, among the 42,599,376 patients without a documented history of OCP use, 61,109 (0.143%) developed a meningioma. This corresponds to a RR of 0.60 (95% CI: 0.58–0.62, p < 0.001) in patients with a documented history of combined OCP use compared to those without, indicating a statistically significantly lower risk of meningioma in women with prior combined OCP use.
Similarly, 1,054,917 (2.23%) of patients in the cohort were diagnosed with PCOS. The overall incidence of meningiomas in patients with PCOS was 0.158%, slightly higher than the overall meningioma incidence of 0.139% (RR = 1.14, 95% CI: 1.09–1.20, p < 0.001).
PCOS was associated with opposing effects on meningioma development. Among obese women, a diagnosis of PCOS was associated with reduced risk of meningioma (RR = 0.70, 95% CI: 0.66–0.74, p < 0.001). Conversely, among non-obese women, PCOS was associated with increased risk of meningioma (RR = 2.08, 95% CI: 1.91–2.27, p < 0.001).
To further examine the effects of combined OCP use within PCOS and obesity strata, odds of meningioma were assessed. Controlling for both obesity and PCOS, women with a history of combined OCP use had significantly lower odds of meningioma (OR = 0.58, 95% CI: 0.56–0.59, p < 0.001) compared to women without combined OCP use. When controlling for obesity alone, the protective association remained similar (OR = 0.57, 95% CI: 0.56–0.59, p < 0.001). Among all patients who were obese, combined OCP use was associated with a RR of 0.55 (95% CI: 0.52–0.57, p < 0.001) for meningioma development.
Lastly, patients were stratified by race to evaluate the effect of OCP use on meningioma risk across racial groups. Table 1 presents the available demographic data.
Demographic composition of study population. Considerations for this data include: multiple races can be selected for a single patient in the chart, patients without a documented race are categorized as “Not listed,” and obesity status is defined as a recorded BMI ⩾ 30.0. Patient age was not a reliable measure that could be reported from the Cosmos query.
OCP: oral contraceptive pill (OCP); PCOS: polycystic ovary syndrome.
Patient counts do not equal study counts as multiple races can be selected for a patient in Epic demographics entry.
After stratifying for race, meningioma RRs were calculated for patients with a positive history of combined OCP use compared to patients who had no history of OCP use (see Table 2).
Risk of meningioma development in patients with combined OCP use compared to patients without OCP use, stratified by race.
Statistical testing showed insignificance in the RR calculated for patients with a race of not listed assigned to their demographic data.
For patients whose demographic data included a race of White, Black or African American, Asian, and for those whose race was other, a history of combined OCP use was associated with a statistically significant reduction in RR for meningioma as compared to their counterparts without combined OCP use. When controlling for race, the overall OR for meningioma in patients with combined OCP use was 0.57 (95% CI: 0.55–0.58, p < 0.001).
Validation studies found hepatocellular adenoma risk was significantly increased among female patients with past OCP use (RR: 2.97, 95% CI: 2.87–3.08, p < 0.001), while ovarian cancer risk was significantly decreased among female patients with past OCP use (RR: 0.59, 95% CI: 0.57–0.61, p < 0.001). The directionalities of these trends are consistent with current literature that finds OCP use is correlated with an increased hepatocellular adenoma risk and a decreased ovarian cancer risk.17,18
Discussion
This observational cohort study is the first to use the Epic Cosmos database, which contains over 274 million patient records across the United States and Lebanon, to assess the association between premenopausal combined OCP use and subsequent meningioma development. The findings of this study support the hypothesis that premenopausal combined OCP use is associated with a decreased risk of meningioma.
Although a diagnosis of PCOS was associated with an overall increased RR for meningioma as compared to patients without a PCOS diagnosis, this relationship varied depending upon obesity status. PCOS was associated with a decreased risk in obese women and an increased risk in non-obese women, suggesting possible effect modification by obesity.
One possible explanation could be attributed to the differing hormonal profiles of PCOS in patients with obesity compared to those without obesity. Deng et al. 20 observed that both non-obese and obese women with PCOS had significantly elevated levels of testosterone and androstenedione compared to women without PCOS. Additionally, obese women with PCOS were found to have lower levels of estrone compared to obese women without PCOS. However, non-obese, normal BMI women with PCOS (“lean PCOS”) were associated with significantly elevated levels of 17-hydroxyprogesterone, progesterone, and estrone as compared to non-obese patients without PCOS.
There is a well-documented association between obesity and meningioma risk, which is thought to arise from elevated levels of estrone caused by peripheral aromatization of androgens by excess adipose tissue.3,21 This study showed an association between a diagnosis of PCOS in patients with obesity and a relatively lower risk of meningioma. This is consistent with the aforementioned reduced levels of estrone in patients with obesity and PCOS relative to patients with obesity but without PCOS. Conversely, the observed harmful effect of PCOS in women who were not obese may be consistent with the finding of elevated progesterone by Deng et al., 20 reflective of the well-established correlation between progestogen exposure and meningioma risk. 7
Despite potentially differing mechanisms by which PCOS may contribute to meningioma risk in non-obese versus obese women, the protective association of combined OCP use was maintained across both strata. Combined OCP use was significantly associated with a reduced risk of meningioma while controlling for obesity and PCOS, as well as obesity alone.
This study also evaluated the association between combined OCP use and meningioma risk stratified by patient race as documented in Epic. A significantly reduced risk of meningioma with combined OCP use was observed for those with a documented race of White, Black or African American, Asian, and for those whose race was other. However, this reduced risk was not statistically significant in the not listed race stratum.
The observed protective effect of combined OCP use was approximately 20% more pronounced in the White stratum compared to the Black stratum. Prior research has shown that Black race is associated with a greater incidence of meningioma compared to non-Black populations. 4 Additionally, Black patients with meningioma have been associated with an increased prevalence of somatic hedgehog mutations driving meningioma development. 22 These mutations could possibly contribute to the slightly attenuated protective effect of combined OCP use observed in Black patients relative to White patients.
Overall, findings of this study suggest a potential protective association between combined OCP use and meningioma that remained significant after controlling for obesity, PCOS status, and race. Additionally, this study found that meningioma risk in PCOS patients differed based upon obesity status. These findings underscore the importance of considering both hormonal and metabolic contexts in reproductive conditions.
Clinically, these results suggest an opportunity for providers to consider combined oral contraceptives in patients with an elevated risk of meningioma, including patients with obesity or certain subtypes of PCOS. PCOS is a heterogenous condition with hormonal levels that vary by phase of the menstrual cycle and subtype.7,8 More granular subtype diagnoses, hormonal profiles, genetic factors (e.g., CYP17A1 polymorphisms), and environmental factors such as diet may yield insights into specific meningioma risk.23–28
Future research could investigate the association between the duration of combined OCP use and various molecular formulations of combined OCPs on meningioma risk, variables that were limited in this study.
Study strengths and limitations
Strengths of this study include its utilization of the Epic Cosmos database, with access to over 274 million de-identified patient charts across the United States and Lebanon, allowing for highly specific studies. Correspondingly, validation of the search strategy by hepatocellular adenoma and malignant ovarian neoplasm searches showed trends in risk consistent with current literature.6,7
Limitations to this study include reliance on ICD-10 coding accuracy and potentially incomplete diagnostic and prescription data in Epic Cosmos. Medication adherence, duration of use, prescription refill type, and age at diagnosis were inaccessible, which may obscure specific risks. Data completeness may be lacking in older records. Due to the aggregate nature of Epic Cosmos data, a multivariate analysis could not be executed on this dataset. Additionally, the patient race field in Epic can be completed with no entry, a single race, or multiple races, meaning patient counts once stratified by race do not equal discrete patient counts.
A formal sample size calculation was not performed because this retrospective database study included all eligible patients meeting inclusion criteria rather than a predetermined enrollment target. Thus, no a priori power analysis was calculated for this study.
This study was conducted using data from the Epic Cosmos database, which reflects clinical practices and demographic patterns mainly within U.S. healthcare systems. Therefore, these findings may not be fully generalizable to populations in other countries or where differing formulations of combined oral contraceptives are used.
Conclusion
This study found an association between a history of combined oral contraceptive use and a reduction in meningioma RR. This reduced risk remained statistically significant after controlling for obesity, PCOS status, and race. This study also found that a diagnosis of PCOS was associated with a greater risk of meningioma in patients who were not obese, as compared to a lower risk of meningioma in patients who were obese.
Supplemental Material
sj-pdf-1-whe-10.1177_17455057261421724 – Supplemental material for Incidence of meningioma in women with a history of combined oral contraceptive pill use and polycystic ovary syndrome
Supplemental material, sj-pdf-1-whe-10.1177_17455057261421724 for Incidence of meningioma in women with a history of combined oral contraceptive pill use and polycystic ovary syndrome by Emily Edwards, Kathryn Tsai, Suguna Pappu, Heidi Gaddey, Tyler B. Nofzinger, Emily Vachon and Bryan McConomy in Women's Health
Footnotes
Acknowledgements
We thank Alexander Smith for his valuable feedback in reviewing the study design. We also appreciate the administrative support provided by Carle Foundation Hospital, especially Dr. Emily Wee, Dr. Linda Owens, and Debby Vannoy. Data used in this study came from Epic Cosmos, a dataset created in collaboration with a community of Epic health systems representing more than 274 million patient records from over 1583 hospitals and 36,100 clinics from all 50 states and Lebanon.
Ethical considerations
Carle Institutional Review Board and the Carle Human Subject Protection Program waived the need for further IRB approval, ethics approval, and patient consent for the collection, analysis and publication of the retrospectively obtained and anonymized Epic Cosmos data for this non-interventional study.
Consent to participate
Not applicable.
Consent for publication
Not applicable.
Author contributions
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding for the publication of this project was made possible through the Stephens Family Clinical Research Institute and Carle Health Center for Philanthropy.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data availability statement
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
