Primary open angle glaucoma polygenic risk score can identify individuals at low risk of disease onset in the Ocular Hypertension Treatment Study

Abstract

Purpose : Assess utility of a primary open angle glaucoma (POAG) polygenic risk score (PRS) to identify ocular hypertensive individuals at low risk of disease onset.

Methods : We included 1010 participants from the Ocular Hypertension Treatment Study (OHTS) with genotype data. Participants' eyes were categorized into disease onset risk tertiles using baseline OHTS model of age, intraocular pressure (IOP), vertical cup-to-disc ratio (VCDR), central corneal thickness (CCT), and pattern standard deviation (PSD). POAG PRS was computed using summary statistics from a large cross-ancestry POAG genome wide association study meta-analysis. PRS was dichotomized at a threshold that maximized Youden’s index. Disease onset and survival probability were analyzed using mixed-effect logistic models and stratified Kaplan-Meier curves, considering the correlation between an individual’s eyes.

Results : Among 2018 eyes (1010 participants), the optimal PRS threshold was -0.05 (48th percentile). In the logistic model adjusted for age, VCDR, IOP, CCT, PSD, and randomization status (RS), POAG PRS under threshold was associated with 2.15-fold increased likelihood of remaining disease-free at 20 years (95% CI 1.65-2.81, p<0.001). Among those randmized to early treatment, individuals below the PRS threshold showed conversion rates of 3.5% at 10 years and 13.2% at 15 years. Those above the threshold had rates of 7.6% at 10 years and 30.1% at 15 years. For the observation group, below-threshold individuals had rates of 9.3% at 10 years and 29.4% at 15 years, while those above had rates of 18.8% at 10 years and 45.0% at 15 years. When the study population was grouped into OHTS baseline risk tertiles, the largest differences in survival probability based on PRS threshold was observed in eyes in the high risk tertile (p<0.0001) with randomization to early treatment mitigating effect of high genetic risk (p=0.002) but providing limited benefit to those with low genetic risk (p=0.86) (Fig1, Table 1).

Conclusions : A PRS threshold can identify those at lower risk of POAG onset, with substantial differences in conversion rates over 20 years. Amongst those considered at highest risk based on the OHTS baseline model, early treatment may partially offset the effect of high genetic risk but may provide limited benefit to those with low genetic risk.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.

 

 

20 year POAG conversion by OHTS risk tertile