Abstract
Purpose :
Glaucoma is one of the most heritable human diseases and hence mapping the specific genes is expected to both provide insights into disease biology as well as enhance our ability to detect who will be at highest risk. While genome-wide association studies (GWAS) have identified 127 open angle glaucoma risk loci to date, these only account for a small fraction of disease heritability. Traditionally glaucoma genes were mapped using a case-control design but we recently showed (Craig et al, Nature Genetics, 2020) that power to identify genes can be dramatically increased (effective sample size increased 2.5 fold) by augmenting cases and controls with glaucoma risk factor data. Our aim here is to leverage the correlation between glaucoma, intraocular pressure (IOP) and vertical cup to disc ratio (VCDR) to identify novel glaucoma risk loci.
Methods :
We conducted a multi-trait analysis of glaucoma and its risk factors, using new and existing GWASs we have collected from around the world (34,179 glaucoma cases, 349,321 controls, supplemented by 111,724 individuals with VCDR and 153,604 individuals with IOP measures, drawn from studies in the International Glaucoma Genetics Consortium, UK Biobank and the Canadian Longitudinal Study of Aging). A key novelty over previous work is that VCDR measures were derived using a machine learning approach instead of using human graders, dramatically improving accuracy and power. The input traits were combined in a multi-trait GWAS framework using the software MTAG; an important feature of the approach is that it generates results which are specific to glaucoma risk.
Results :
Our analysis of European ancestry samples identified 263 genome-wide significant loci, with most showing very good concordance in other ancestries. Combining across ancestries further improved power and identified 312 independent loci associated with glaucoma. Novel loci implicated by our analysis include associations at the genes TCF7 (implicating overlapping pathways with cataract) and SYN3 (implicating overlapping pathways with macular degeneration). We are currently working on replicating the novel loci in a large case-control cohort.
Conclusions :
This work dramatically expands our knowledge of glaucoma genetics by identifying a large number of new risk loci, providing new insights into glaucoma aetiology and helping enable improved prediction of who is at highest disease risk.
This is a 2021 ARVO Annual Meeting abstract.