PMON243 Metabolic Disorders Genetically Modify Polycystic Ovary Syndrome through its Candidate Hormones

Abstract Introduction As the most common endocrine disorder in reproductive-aged women, polycystic ovary syndrome (PCOS) increases the risk of developing several cardiometabolic disorders like insulin resistance. However, Mendelian randomization (MR) studies provide little support for direct causal associations between PCOS and metabolic conditions with the exception of obesity and type 2 diabetes (T2D). As a result, secondary pathways and biological mediators increasingly emerge as a potential explanation for underlying mechanisms between traits. Objective To identify biomarkers that modify PCOS presentation and its comorbid conditions through MR, which uses significant genetic variants from genome-wide association studies to estimate causal effects between variables. Methods A bidirectional MR analysis was first implemented to identify associations between PCOS and metabolic traits (i.e., body mass index (BMI), T2D, blood pressure, coronary artery disease, and heart failure), with PCOS acting first as the exposure, then as the outcome. Next, MR was applied between significant metabolic phenotypes from the first analysis and thirty-two biomarkers from UK Biobank to identify any overlap with biomarkers that also significantly increased risk for PCOS. Mirroring biomarkers from a hospital population were used as a replication dataset and were further analyzed in a mediation analysis with clinical laboratory measurements. Results Testosterone from females (β = 0.83, p = 4.67e-04) and sex hormone-binding globulin (SHBG, β = -0.01, p = 5.00e-06) were identified as significant exposures for PCOS. These biomarkers also exhibited causal effects on BMI (βFT = 0.10, p = 5.81e-12; βSHBG = 0.001, p = 4.56e-10) and T2D (βSHBG = -0.004, p = 4.99e-26), which were the only two phenotypes that exerted causal risk for PCOS (p < 0.001). Although SHBG had strong bidirectional effects on both traits, female-derived testosterone was the only testosterone exposure (vs sex-combined and male testosterone) that increased BMI (β = 0.04, p = 1.08e-16). This positive association with BMI was exclusively replicated for free testosterone as the outcome. Although no significant associations were observed in the mediation analysis for BMI and free testosterone, total testosterone mediated 29% of the genetic pathway from T2D to PCOS. Conclusions Overall, MR can expand our understanding on the phenotypic irregularities of PCOS through examination of biomarker etiology cross-traits. These results imply a strong interaction between prevalent PCOS characteristics, one that may mitigate the advancement of other chronic comorbidities. Therefore, further research is needed to elucidate PCOS regulation, which may reveal a more complex web of modifiable effects that have yet to be explored. Presentation: Monday, June 13, 2022 12:30 p.m. - 2:30 p.m.