Multi-trait Polygenic Probability Risk Score Enhances Glaucoma Prediction Across Ancestries

Area of Science:

• Genetics and Genomics
• Ophthalmology
• Biostatistics

Background:

• Primary open-angle glaucoma (POAG) is a leading cause of irreversible blindness globally.
• Current risk assessment tools for POAG lack sufficient predictive power for early detection.
• Effective strategies for early POAG detection are crucial to prevent vision loss.

Purpose of the Study:

• To develop and validate a multi-trait polygenic probability risk score (PPRS) for enhanced POAG prediction.
• To integrate multiple glaucoma-related traits and functional genomic annotations for improved predictive accuracy.
• To assess the performance of the PPRS across diverse ancestral populations.

Main Methods:

• Constructed polygenic risk scores (PRSs) for POAG, intraocular pressure (IOP), vertical cup-to-disc ratio (VCDR), and retinal nerve fiber layer (RNFL) thickness using >7 million variants and 96 functional annotations.
• Employed the SBayesRC method for PRS construction.
• Validated the multi-trait PPRS in European (UK Biobank) and Latino (MAGGS) cohorts.

Main Results:

• The multi-trait PPRS achieved superior predictive accuracy with AUC values of 0.814 (Europeans) and 0.801 (Latinos), outperforming single-trait models (AUC ≤0.79).
• Identified ancestry-specific genetic contributions, with IOP strongest in Europeans and VCDR in Latinos.
• Demonstrated significant risk stratification, with the highest PPRS decile showing substantially increased POAG risk (80.2-fold in Europeans, 51.1-fold in Latinos) compared to the lowest decile.

Conclusions:

• Integrating multiple disease-relevant traits and functional annotations significantly enhances polygenic prediction of POAG.
• The developed PPRS offers improved risk stratification and case capture across diverse populations.
• This approach has significant implications for targeted screening, early intervention, and reducing the global burden of POAG.