Mendelian randomization in ovarian research: applications in PCOS, ovarian aging, and assisted reproductive technology

Ovarian disorders such as polycystic ovary syndrome (PCOS) and primary ovarian insufficiency (POI) have complex etiologies that are difficult to disentangle using observational studies. Mendelian randomization (MR) offers a powerful approach to identify causal drivers of these conditions, with direct implications for assisted reproductive technology (ART), but its application is blocked by the ART phenotype gap, namely the absence of large-scale GWAS for cycle-specific outcomes. This review synthesizes a conceptual framework to navigate this gap. In the near term, researchers can use existing genetic data through two complementary strategies, upstream exposure analysis and proxy outcome analysis, both requiring rigorous validation and sensitivity testing. In the long term, the field must build deeply phenotyped ART genomic consortia with standardized data and ethical governance. This approach enables causal hypothesis generation using existing data, while recognizing that definitive evidence will require deeply phenotyped consortia in the future.