Assessing the genetic overlap between myopia and primary open-angle glaucoma

Abstract

Purpose : Previous studies have shown that patients with myopia have an increased risk of primary open-angle glaucoma (POAG). However, the mechanisms behind this association are still unclear. Given the polygenic origin of both diseases, we have explored the genetic overlap between myopia and POAG by using summary statistics from the largest genome-wide association studies (GWAS) of myopia, POAG, and its endophenotypes.

Methods : First, we constructed effect-size polygenic risk scores (PRS) for myopia using the summary statistics from the meta-analysis of GWAS conducted by the Consortium for Refractive Error and Myopia (CREAM) and 23andME (n=160,420). We then tested the association and variance explained of these scores with POAG in 798 cases and 1991 controls from the Australian and New Zealand Registry of Advanced Glaucoma (ANZRAG), and with POAG endophenotypes in the population-based Rotterdam Study (RS) I-III (n=10,792). Second, we used the LD-score regression method to assess the genetic correlation (RhoG) between myopia and POAG endophenotypes. We used summary data from the CREAM and the International Glaucoma Genetics Consortium (IGGC) including only cohorts of European descent. POAG endophenotypes studied include intraocular pressure (IOP), vertical cup-disc ratio (VCDR), disc and rim area.

Results : In total 12 myopia-PRS were derived from different P-value thresholds (from P<5x10^-8 up to P<1). We did not find evidence of an association between myopia-PRS and POAG in the ANZRAG study (P=0.8). In the RS, we found an association of myopia-PRS with rim area (P=2.5x10^-11; r²=0.37%), disc area (P=1.8x10^-8; r²=0.24%) and a nominal association with IOP (P=1.2x10^-2; r²=0.05%). No association was found between myopia-PRS and VCDR. Using the LD-score method, we found a significant genetic correlation between myopia and disc area (RhoG=-0.12, P=1.8x10^-3) while no genetic correlation with VCDR, IOP or POAG was found.

Conclusions : Using two different methods we did not find a significant genetic association between myopia and POAG. However, we observed that myopia-PRS was associated with some of the endophenotypes of POAG (such as rim and disc area), but not with other relevant and expected endophenotypes such as VCDR or IOP. Our findings suggest that the polygenic overlap between myopia in the general population and POAG is not large.

This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.