Abstract 16950: The Genetic Basis of Atrial Fibrillation in a Large-Scale Multi-Ancestry Sample

Abstract

Introduction: Atrial fibrillation (AF) is a heritable arrhythmia and associated with substantial morbidity. Hypothesis/Aims: We sought to expand our understanding of the genetic basis of AF by examining a large and ancestrally diverse sample. Methods: We performed a meta-analysis of ancestry-specific GWAS results from 600K unrelated individuals (85K with AF) in the Million Veteran Program (MVP) with GWAS results from nearly 2.25 million individuals (267K with AF) in the Atrial Fibrillation Consortium (AFGen). We estimated genetic correlations across ancestry groups and traits in MVP. To evaluate GWAS loci, we examined possible gene functions, prioritized relevant genes and pathways, and explored potential drug targets. We derived a multi-ancestry polygenic risk score (PRS) and independently evaluated associations with AF in the All of Us cohort (nearly 100K individuals; 3K with AF). Results: We identified 376 loci associated with AF, including 63 that have not been reported. We observed closer genetic correlation of AF between individuals of European and Hispanic (r2=1.0) than with African ancestry (r2=0.6). AF is highly correlated genetically with multiple cardiovascular traits, especially coronary artery disease, heart failure, and hypertension. GWAS loci implicate 15 genetic regulators of the cardiac action potential, including targets of commonly used AF drugs and 2 potential therapeutic targets whose predicted cardiac expression is positively associated with AF risk (KCNJ5 and KCND3). A PRS derived from the meta-analysis outperforms all previously reported PRSs in European, African and Hispanic ancestry participants and is strongly associated with multiple cardiovascular outcomes. Conclusions: In the largest genetic study of AF to date, we identify disease-relevant pathways and improve polygenic prediction for AF. We demonstrate the importance of ancestry-specific analyses in assessing AF risk and suggest that further inclusion of participants from more diverse ancestries is likely to accelerate genetic discoveries.