Stroke genetics informs drug discovery and risk prediction across ancestries

Abstract

Previous genome-wide association studies (GWAS) of stroke, the second leading cause of death, have been conducted in populations of predominantly European ancestry.1,2 We undertook cross-ancestry GWAS meta-analyses of stroke and its subtypes in 110,182 stroke patients (33% non-European) and 1,503,898 control individuals of five ancestries from population- and clinic-based studies, nearly doubling the number of cases in previous stroke GWAS. We identified association signals at 89 independent loci, of which 61 were novel. Effect sizes were overall highly correlated across ancestries. Cross-ancestry fine-mapping, in silico mutagenesis analysis using a novel machine-learning approach,3 transcriptome and proteome-wide association analyses revealed putative causal genes (e.g. SH3PXD2A and FURIN) and variants (e.g. at GRK5 and NOS3). Using a novel three-pronged approach,4 we provided genetic evidence for putative drug effects, highlighting F11, KLKB1, PROC, GP1BA, and VCAM1 as possible targets, with drugs already under investigation for stroke for F11 and PROC. A polygenic score integrating cross-ancestry and ancestry-specific stroke GWAS with vascular risk factor GWAS (iPGS) showed strong prediction of ischemic stroke risk in European and, for the first time, East-Asian populations.5,6 The iPGS performed better than stroke PGS alone and better than previous best iPGS, in Europeans and East-Asians. Transferability of European-specific iPGS to East-Asians was limited. Stroke genetic risk scores were predictive of ischemic stroke independent of clinical risk factors in 52,600 clinical trial participants with cardiometabolic disease and performed considerably better than previous scores, both in Europeans and East-Asians. Altogether our results provide critical insight to inform biology, reveal potential drug targets for intervention, and provide genetic risk prediction tools across ancestries for targeted prevention.

Genal: A Python Toolkit for Genetic Risk Scoring and Mendelian Randomization

“…This analysis leverages public GWAS summary statistics 9,10 . For the MR analyses, we achieve results that match those obtained with TwoSampleMR, the most commonly used tool for such analyses.…”

“…In the supplementary material, we provide a tutorial along with an example where we construct a PRS for systolic blood pressure and examine its genetic influence on the risk of all-cause stroke using MR. This analysis leverages public GWAS summary statistics 9,10 . For the MR analyses, we achieve results that match those obtained with TwoSampleMR, the most commonly used tool for such analyses.…”

Genome-Wide Studies in Ischaemic Stroke: Are Genetics Only Useful for Finding Genes?

“…However, the largest step forward was taken by the GIGASTROKE project, which involved the largest multiethnic meta-analysis GWAS in stroke (110,182 stroke patients and 1,503,898 controls) [ 15 ], and in which the International Stroke Genetics Consortium (ISGC) collaborated [ 6 ]. The pre-printed results represent the most comprehensive description of stroke-risk genetic variants in Europeans, East Asians, South Asians, AfricanAmericans, and Hispanics to date.…”

“…Finally, the GIGASTROKE study also explored the MTAG approach by adding the genetic information for related traits such as coronary artery disease, atrial fibrillation, and white matter hyperintensity volume. These analyses led to the identification of 24 novel loci associated with stroke and its subtypes [ 15 ].…”

“…However, the beneficial effect in patients is lacking, as there are no phase III clinical trials assessing the viability of HDAC inhibitors in managing stroke [ 180 ]. Interestingly, an intronic variant in HDAC9 has been associated with the risk of LAS in several studies [ 2 , 15 , 181 ].…”