The Association of Polygenic Risk Score With Glaucoma and Its Related Traits: Results From Two German Population-Based Studies.

PURPOSE

Genome-wide association studies have revealed many loci linked to glaucoma. The aim of this study was to explore the associations between a glaucoma polygenic risk score (GPRS) with glaucoma prevalence, intraocular pressure (IOP), and peripapillary retinal nerve fibre layer thickness (pRNFL) on a population-based level.

METHODS

Participants of the Gutenberg Health Study (aged 40-80 years) and the AugUR Study (aged 70-95 years), two German population-based studies, were included. Of participants with genetic assessment 12,147 had glaucoma grading, 8938 data on IOP, and 7844 data on pRNFL. Spectral-domain optical coherence tomography was used to measure pRNFL thickness and noncontact tonometry for IOP measurements. Glaucoma was graded according to the classification of the International Society for Geographic and Epidemiological Ophthalmology. GPRS was calculated using 2453 locus lead variants. Linear regression was conducted to evaluate the association of GPRS with IOP and pRNFL thickness and Poisson regression analysis for the association of GPRS with glaucoma prevalence. Age groups were divided in quartiles: 40-51, ≥52-62, ≥63-72, and ≥73-95 years.

RESULTS

Increased GPRS was significantly associated with increased glaucoma prevalence (Prevalence ratio: 1.84, 95% confidence interval [CI], 1.60-2.11; P = 2 × 10-16) and higher IOP (B-estimate = 0.54 mm Hg per SD of GRPS; 95% CI, 0.48-0.59; P = 2 × 10-16). Higher GPRS was associated with thinner global pRNFL (B-estimate = -0.61 µm per SD of GPRS on global pRNFL; 95% CI, -0.86 to -0.36; P = 1.2 × 10-6) with increasing age. There was an association in temporal-inferior, temporal-superior, nasal-inferior, nasal-superior, and nasal pRNFL sector.

CONCLUSIONS

Higher GPRS is associated with glaucoma prevalence, higher IOP and thinner pRNFL in two population-based studies in Germany. These associations were particularly seen in older participants.