Abstract
Purpose :
Primary open-angle glaucoma (POAG) is a highly heritable disease and genome-wide association studies (GWAS) have already identified several loci that predispose to it. Although POAG is more common in African ancestry subjects, little is known about the genetic risk in this group. Previous studies in subjects of African ancestry generally could not replicate known, primarily European-driven genetic associations. The purpose of this study is to evaluate the role of known genetic variants associated with POAG in subjects of African descent.
Methods :
Replication of known POAG GWAS loci was sought in two panels, of African (321 cases and 217 controls) and European descent (306 cases and 297 controls) respectively. Each panel had more than 70% power to replicate associations at nominal level for MAF>0.1 and OR>1.5.
Correlation of observed effect sizes with those of previously published literature as well as genomic risk score associations with the POAG status were used to estimate similarities in genomic architecture of POAG risk.
Results :
There was a nominal replication for the CDKN2B-AS (p=0.038) and the ARHGEF12 locus (p=0.007) in Africans, but despite sufficient power, not at the TMCO1 locus. The published GWAS effect sizes were correlated with those observed in the African (r=0.54, p=0.01) and the European panel (r=0.66, p=0.0008). Effect sizes of POAG GWAS loci observed in our African panel were a function of the determinants of statistical power, such as the originally published effect size (p=0.002) and MAF in Africans (p=0.02). Genetic risk scores built on estimates derived from POAG GWAS were marginally associated with POAG in our African panel (p=0.06) and unsurprisingly stronger in the European panel (p=0.0001), and together with sex and age, explained 7 and 10% of POAG variance in the respective populations.
Conclusions :
Genetic effects observed in European populations were similar to those observed in African cohorts. In the latter, statistically significant replication in some cases may be complicated by differences in LD patterns across populations of different ancestry and by the generally lower MAFs observed for POAG loci. Because known POAG loci tend to have a much lower MAF in subjects of African descent, there is an even larger heritability gap in this ethnic group compared to others, which needs to be addressed by future studies.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.