Abstract
Purpose :
Macular ganglion cell complex (GCC) thickness is an important biomarker for early glaucoma. Understanding its determinants may teach us about the biological processes underlying glaucoma. To date, genome-wide association studies (GWAS) of GCC have been small and not identified any significant associations. We carried out the largest GWAS of GCC to date in 41,504 participants of the UK Biobank.
Methods :
The UK Biobank is a very large British multisite cohort study. Participants underwent spectral-domain optical coherence tomography (SD-OCT) imaging of both eyes using the Topcon 3D OCT- 1000 Mark II and genotyping using the Affymetrix Axiom UK Biobank chip. Version 1.6.1.1 of the Topcon Advanced Boundary Segmentation (TABS) algorithm was used to delineate the retinal surfaces. Following exclusion of images based on poor image quality or segmentation, 41,504 European, unrelated participants were used to perform a linear model GWAS on GCC thickness. We adjusted for age, sex, height, weight, refractive error, batch effect, and the first 20 principal components of the genotype data.
Results :
There were a total of 89 independent genomic regions associated with GCC at genome-wide significance. The most significant locus was in TSPAN10, (17:79614932_TTAAC_T, P=1.42x10-57) a known myopia locus, followed by SLC6A20 (rs17279437, P=3.48x10-37), previously associated with hyperglycinuria and metabolite levels. There was strong association with several known oculocutaneous albinism loci including TYR (rs1042602, P=5.58x10-24) and OCA2 (rs1800407P=2.37x10-20). The known glaucoma-related SIX6 locus was also genome-wide significant (rs33912345, P=1.93x10-12). However, the majority of known glaucoma loci were not associated with GCC at genome-wide significance (P>5x10-8).
Conclusions :
We have identified 89 novel loci that are associated with GCC in the largest GWAS of inner retinal anatomy to date. Despite adjusting for refractive error, the strong association of GCC with a known myopia locus suggests common processes underlying general eye anatomical size. The relative lack of glaucoma-related loci among GCC-significant loci suggests the majority of GCC variation between individuals in populations is not due to glaucomatous processes and perhaps more reflective of baseline anatomy. Future work aims to examine all GCC-significant loci for association with glaucoma in independent studies.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.