Association of polygenic scores for glaucoma with measures of retinal ganglion cell integrity in young and older adults.

PURPOSE Polygenic scores (PGS) for glaucoma is predictive of the disease in older adults. This study tested the hypothesis that multitrait PGS for primary open-angle glaucoma (POAG) and its associated traits are associated with glaucoma endophenotypes from a young age, but with larger effects in older adults. DESIGN Cross-sectional and cohort analyses PARTICIPANTS: Young (<30 years; n=1,400) and older (45+ years; n∼ 3,500) community-based adults. METHODS Participants underwent ocular tonometry, optical coherence tomography imaging, and genotyping. Their PGS for POAG, IOP, and vertical cup-to-disc ratio (VCDR) were generated. A subset of young participants (n∼614) had follow-up measurements 8 years later. Cross-sectional associations in both cohorts and the 8-year change in the young cohort were analysed against each PGS. MAIN OUTCOME MEASURES Intraocular pressure (IOP), peripapillary retinal nerve fibre layer (pRNFL) thickness, and Bruch's membrane opening minimum rim width (BMO-MRW). RESULTS IOP-PGS explained 4 and 8% of the variance in IOP (p≤0.001) in the young and older cohorts. Weak associations between pRNFL thickness and all 3 PGS were observed in the older group (all p<0.017), but none were significant in the young participants. All 3 PGS were significantly associated with BMO-MRW, explaining 0.3-14.5% and 0.1-12.8% of the phenotypic variance in the older and younger cohorts, respectively. None of the PGS were associated with longitudinal IOP or pRNFL change in the young cohort. CONCLUSIONS Associations between PGS and optic disc measures were present from young adulthood, but the effect sizes were greater in older adults. This, coupled with the lack of associations in the 8-year change in the young adults, suggests that glaucoma-related genetic effects on the optic nerve are not apparent until older age.