Glaucoma Polygenic Risk Scores Demonstrate Heterogeneous Performance across 2 Large Multiethnic Cohorts

Area of Science:

• Ophthalmology
• Genetics
• Biostatistics

Background:

• Glaucoma is a leading cause of irreversible blindness globally.
• Polygenic risk scores (PRS) are emerging tools for stratifying patient risk.
• The consistency and reliability of glaucoma PRS across diverse populations require thorough investigation.

Purpose of the Study:

• To evaluate the population-level performance and individual-level risk assignment of published glaucoma polygenic risk scores (PRS).
• To assess the consistency of glaucoma PRS across two large, multiethnic cohorts: the All of Us (AoU) Research Program and Mass General Brigham (MGB) biobank.

Main Methods:

• A cross-sectional study design was employed.
• Eleven published glaucoma PRS were evaluated using logistic regression models.
• Performance was assessed by examining associations with open-angle glaucoma (OAG) and agreement in risk classification using Pearson correlation and Jaccard index.

Main Results:

• All evaluated PRS were significantly associated with increased odds of OAG at the population level.
• Performance varied by ancestry, with strongest associations in European ancestry and attenuated effects in African and admixed American ancestries.
• Individual risk classification showed low concordance across PRSs, with median Pearson r = 0.42 and Jaccard index = 0.16, indicating instability.

Conclusions:

• Published glaucoma PRS demonstrate consistent population-level prediction but lack individual-level reliability for standalone clinical application.
• Glaucoma PRS should be used as adjuncts to established risk factors, not as primary diagnostic tools.
• Future research should focus on developing methods to assess PRS reliability and standardize analytical approaches for improved clinical utility.