Abstract
Purpose :
Genome-wide association study (GWAS) meta-analyses have identified hundreds of genes associated with primary open-angle glaucoma (POAG) and related traits, though most individuals enrolled in these studies are of European Caucasian descent. Here, we perform a GWAS meta-analysis for POAG in the ethnically diverse All of Us dataset.
Methods :
POAG cases were defined as individuals age >40 with an ICD 9/10 diagnosis code for POAG in the electronic medical record. Controls were also >40 years old and did not have any ICD 9/10 codes for POAG or any other forms of glaucoma, glaucoma suspect, or ocular hypertension. Sample and variant quality control (QC) were performed prior to data analysis and genetically inferred ancestry was derived for all individuals using principal component (PC) analysis and the 1000 Genomes Project as the reference panel. POAG GWAS were first performed individually for individuals of European (1386 cases/91461 controls), African (855 cases/37859 controls) and Hispanic (302 cases/24419 controls) descent. The regression analyses adjusted for age, sex and top 6 genotype PCs as covariates. The three GWAS all had a lambda inflation value <1.05. A meta-analysis was then performed using a fixed-effects inverse-variance weighted approach in METAL.
Results :
A total of 16 loci reached genome-wide significance (p<5x10-8) and 37 were subthreshold (p<1x10-5) in the cross-ancestry GWAS meta-analysis, all of which have previously been associated with POAG and/or intraocular pressure (IOP) in GWAS meta-analyses performed to date. For the genome-wide significant SNPs, the European SNP effects (beta) correlated with effects in African and Hispanic ancestries (Pearson correlation coefficient r=0.58 and r=0.47, respectively). Intriguingly, IOP-independent associations for POAG, such as CDKN2B-AS1, did not reach significance levels in this GWAS meta-analysis.
Conclusions :
These data demonstrate that known GWAS loci for POAG are reproducible in an ethnically diverse dataset. The lack of association of IOP-independent genes in this GWAS suggests that use of ICD 9/10 codes for disease ascertainment may bias towards selection of individuals who have high-tension glaucoma.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.