GWAS of Retinal Vessel Tortuosity Identifies 173 Novel Loci Revealing Genes and Pathways Associated with Vascular Pathomechanics and Diseases

Abstract

AbstractBackgroundFundus images of the eye allow for a non-invasive assessment of the retinal vasculature, whose features provide important information on cardiovascular health. In particular, blood vessel tortuosity (or “bendiness”) is a morphological feature associated with hypertension and other common diseases.MethodsWe analysed a total of 175 821 fundus images across 63 662 participants from three cohorts of genotyped subjects, namely the UK Biobank (n=62 751), SKIPOGH (n=397), and OphtalmoLaus (n=514). We set up a fully automated image processing pipeline, using the ARIA software for positional annotation of the vasculature and the DL tool Little W-Net to determine vessel type, enabling the massively parallel characterisation of such images in terms of seven distinct vessel tortuosity measures for arteries, veins, and combined vessels. Using these measures as traits, we performed the largest Genome-Wide Association Study (GWAS) of vessel tortuosity conducted to date. We calculated gene set enrichment using a novel high-precision statistical method PascalX.ResultsHigher tortuosity was significantly associated with the incidence of angina, myocardial infarction, stroke, deep vein thrombosis, and hypertension. Heritability was highest (25%) when using the tortuosity measure known as Distance Factor (curve over chord length). Using this measure, we identified 175 significantly associated loci, 173 of which were novel, 4 of which replicated in an independent meta-cohort. Vessel-type specific GWAS revealed 114 and 63 loci for arteries and veins, respectively. Aggregating SNP-wise effects revealed key genes, such as COL4A2, ACTN4, LGALS4, LGALS7, LGALS7B, TNS1, MAP4K1, EIF3K, CAPN12, ECH1, and SYNPO2. Tortuosity genes were overexpressed in arteries and heart muscle, forming pathways modulating the structural properties of the vasculature, and involved in tissue remodelling and angiogenesis. We demonstrated that tortuosity loci served pleiotropic functions as cardiometabolic disease variants and risk factors. In line with this, Mendelian Randomisation revealed causal effects of LDL on tortuosity.ConclusionsWe showed that several alleles associated with retinal vessel tortuosity point to a common genetic architecture of this trait with cardiovascular diseases and metabolic syndrome. Our results shed new light on the genetics of vascular diseases and their pathomechanisms and highlight how GWAS and heritability can be used to improve phenotype extraction from high-dimensional data, such as images.