Enhancing glaucoma prediction across ancestries: integrating functional annotation into multi-trait polygenic risk scores.

Introduction: Primary open-angle glaucoma (POAG) remains a leading cause of irreversible blindness worldwide, with early detection crucial for preventing vision loss. Current polygenic risk scores (PRS) for POAG, however, demonstrate limited predictive power. Methods: Here, we present a multi-trait polygenic probability risk score (PPRS) approach that integrates PRSs of multiple glaucoma-related traits, including POAG, intraocular pressure (IOP), vertical cup-to-disc ratio (VCDR), and retinal nerve fiber layer thickness, while leveraging functional genomic annotations and extensive genomic coverage (>7 million variants) to enhance POAG prediction across diverse ancestries. We evaluated the PPRS in the UK Biobank (n = 324,713, European ancestry) and the Mexican American Glaucoma Genetic Study (MAGGS, n = 4,549, Latino ancestry). Results: The PPRS improved prediction compared with conventional approaches, achieving area under the receiver operating characteristic curve (AUC) values of 0.814 in Europeans and 0.802 in Latinos, versus 0.721 and 0.753 for baseline models with age and sex alone. Genetic contributions varied by ancestry: IOP PRS showed the strongest association in Europeans (OR = 1.63, P = 5.37 × 10 Discussion: These findings show that integrating multiple disease-relevant PRSs and functional annotations significantly improves genetic prediction of POAG across diverse populations, supporting applications in targeted screening and early intervention.