Perfluoroalkyl & Polyfluoroalkyl Substances (PFAS), Endometriosis, & Ovarian Reserve

Introduction: Perfluoroalkyl and polyfluoroalkyl substances (PFAS), also known as the “forever chemicals,” are a chemical family consisting of thousands of congeners and a major public health concern due to their persistence and ubiquity in the environment. In women, PFAS have been associated with menstrual cycle disruption, reduced fecundability, menopause, fibroids, polycystic ovary syndrome (PCOS), and endometriosis. Endometriosis is a steroid-dependent, chronic gynecologic condition. Symptoms of endometriosis include infertility and chronic pain and women with endometriosis have been shown to have lower levels of anti-mullerian hormone (AMH). AMH is secreted in the granulosa cells in the ovaries and is a reliable proxy for functional ovarian reserve. High levels of AMH are associated with PCOS and certain ovarian cancers, while low levels are associated with infertility and a shortened time to menopause. Limited studies have evaluated the effect of PFAS on serum AMH levels and have mostly yielded mixed findings. The objective of this dissertation was to identify and evaluate how PFAS impact endometriosis presentation and AMH and examine endometriosis diagnosis’ impact on AMH among a sample of women from the Endometriosis, Natural History, and Diagnosis (ENDO) Study. Methods: The ENDO Study enrolled women 18-44 between 2007 and 2009 and consisted of an operative cohort and a population cohort in Utah and San Franscico. The operative cohort included women who were undergoing laparotomy/laparoscopy for any indication while the population cohort included women matched on age and residency to the operative sample. AIM 1: PFAS and Endometriosis: Evaluate the association between individual and mixtures of PFAS and endometriosis typology, symptomology, and symptom severity. Methods: Participants with surgically confirmed endometriosis and baseline serum PFAS were included in this analysis (n=200). Logistic and linear regression models were conducted to evaluate the association between individual PFAS and endometriosis typology, symptomology, and symptom severity. Bayesian Kernel Machine Regression (BKMR) models were conducted to evaluate mixtures of PFAS on endometriosis diagnosis, surgical staging, typology, and symptomology. All models were adjusted for age, body mass index (BMI), and parity. AIM 2: PFAS and Ovarian Reserve: Evaluate the effect of individual and mixtures of PFAS on serum AMH levels. Methods: Participants from the Utah study site with PFAS and AMH serum levels were included in this analysis (N=416). Multivariable linear regression models adjusting for age (squared and continuous), BMI, serum cotinine (ng/mL), parity, and current exogenous hormonal contraceptive use were conducted to evaluate the association between individual PFAS and AMH concentrations. BKMR was used to evaluate mixture effects. AIM 3: Endometriosis and Ovarian Reserve: Evaluate the effect of endometriosis diagnosis, stage, and typology on serum AMH levels. Methods: Baseline demographic variables, surgical reports, and serum AMH levels among patients diagnosed with endometriosis and those not diagnosed with endometriosis were used to determine the association between endometriosis overall, as well as surgical staging and type, and functional ovarian reserve (N=348). Results: AIM 1: PFAS and Endometriosis: Individual and mixtures of PFAS were associated with endometriosis typology, symptomology, and symptom severity. PFOS was inversely associated with deep endometriosis (Odds Ratio (OR): O.6). PFHxS was inversely associated with dyschezia and dyschezia severity (OR: 0.5). Mixture analyses yielded joint effects between the PFAS and deep endometriosis, incident endometriosis, surgical staging, dyschezia, and dyspareunia. AIM 2: PFAS and Ovarian Reserve: PFAS were positively associated with serum AMH concentration. We observed a statistically significant positive association between PFOS, PFOA, PFNA, and PFDA on serum AMH concentration (PFOS: β 0.2; PFOA: β 0.02; PFNA Tertile 3 vs. 1 β 0.4; PFDA Tertile 3 vs. 1: β 0.3). We did not observe a significant mixture effect. AIM 3: Endometriosis and Ovarian Reserve: Pre-operative endometriosis was associated with 19.8% lower AMH levels. Both ovarian endometrioma and moderate surgical severity (stage III-IV) were statistically significantly associated with lower AMH levels compared to those without endometriosis (Ovarian: -54.3%; Stage III-IV: -40.1%). When compared to a post-operative diagnosis of normal pelvis, endometriosis diagnosis, type (deep, ovarian, and superficial), and moderate surgical severity were associated with statistically significantly lower AMH levels (Endometriosis Diagnosis: -26.8%; Stage III-IV: -47.8 Ovarian: -60.8%; Deep: -34.3%; Superficial: -24.8%). Conclusion: Overall, these findings contribute to a broader understanding of factors that are associated with ovarian function and endometriosis presentation. We observed associations between PFAS exposure, ovarian reserve, and endometriosis presentation, suggesting potential pathways through which PFAS may contribute to gynecologic disease pathology and progression. Additionally, our findings indicate that endometriosis lesions themselves, independent of surgical intervention, may adversely impact ovarian reserve. Future research is needed to further examine the role of endocrine-disrupting chemicals, including PFAS, in the development and progression of adverse gynecologic disease. Moreover, when evaluating endometriosis and its association with downstream health outcomes, it is critical to move beyond incident diagnosis and consider disease typology and symptomology.