SAT-190 - High Prevalence of Adolescents "at Risk for PCOS" and Links to Genetic Susceptibility to PCOS

Background: Diagnosis of polycystic ovary syndrome (PCOS) in adolescence is challenging because menstrual irregularity and hyperandrogenic findings (e.g., acne) are common in adolescents. Recent international guidelines highlighted an “at risk for PCOS” category, defined by the presence of either menstrual irregularity or hyperandrogenism, to guide clinical surveillance. However, whether the “at risk” category identifies adolescents with elevated PCOS risk is unknown, limiting our ability to develop evidence-based surveillance and risk-stratification strategies.

Objective: To estimate the prevalence of PCOS and “at risk for PCOS” in adolescence and to evaluate associations with genetic risk for PCOS.

Design, Settings, and Participants: We performed a population-based analysis of female participants in the Avon Longitudinal Study of Parents and Children (ALSPAC) in the United Kingdom.

Exposure: Individual polygenic score (PGS) for PCOS was calculated based on the largest genome-wide association study in European women to date.

Main Outcomes and Measures: The primary outcomes were guideline-defined PCOS (presence of both irregular menses and hyperandrogenism) and “at risk for PCOS” (presence of just one feature). Hyperandrogenism was defined by elevated testosterone and/or free androgen indices (≥ 2 SD the cohort mean) and/or a modified Ferriman-Gallwey score ≥4. Menstrual irregularity was defined using age- and menarche-specific guideline criteria for cycle length. Exclusion criteria included a diagnosis of premature ovarian sufficiency, contraception use or pregnancy, or incomplete reproductive data.

Results: Of the 1,533 adolescent females with sufficient data to classify menstrual regularity and hyperandrogenism, 49 (3.2%) met criteria for PCOS and 417 (27.2%) were “at risk for PCOS” (hyperandrogenism only: 276 [18.0%]; irregular menses only: 141 [9.2%]). Among the 1,371 of these individuals with genotyping data, a higher PCOS PGS was associated with increased odds of the at-risk hyperandrogenism phenotype (odds ratio [OR] per SD of PGS 1.22; 95% CI, 1.07-1.39; P=4x10^-3) but not the at-risk irregular menses phenotype (OR 0.98; 95% CI, 0.83-1.15; P=0.8). Although the association with PCOS was positive, it was not statistically significant (OR 1.28; 95% CI, 0.88-1.87, P=0.2), likely reflecting limited statistical power from low prevalence.

Conclusion: In this population-based cohort, more than one-fourth of adolescents met criteria for being “at risk for PCOS”. Genetic risk for PCOS was associated with hyperandrogenism but not isolated irregular menses, suggesting that androgen excess is a more specific early manifestation of inherited PCOS liability. These findings support current guideline recommendations for clinical monitoring of adolescents at risk for PCOS and highlight the need for longitudinal studies to define progression to PCOS.