Abstract 15631: A Precision Medicine Approach to Predicting Pathogenicity Among Population Variants in Hypertrophic Cardiomyopathy-Associated Genes in the UK Biobank

Abstract

Introduction: Genomic and precision medicine have opened the door to predictive genetic testing to identify those at risk of cardiac disease before its development. This is important for hypertrophic cardiomyopathy (HCM), the most prevalent heritable heart muscle disease. A major hurdle to accurate predictive testing is the markedly higher prevalence of HCM-associated variants than of disease. New approaches are needed to discern which disease-causative, “actionable,” variants. Objective: To develop and validate a precision medicine approach for predicting pathogenicity of population-identified HCM-associated variants. Methods: We used phenotypic and genotypic data from the UK Biobank (UKBB). We filtered for missense variants in “definitive” (ClinGen) HCM genes. Our novel precision medicine tool, DiscoVari , provides signal-to-noise (S:N) analysis comparing disease to population variant frequency to identify pathogenic genetic ’hotspots’. We used reverse phenotyping to correlate evidence of HCM and sequelae (sudden death, heart failure, ventricular arrhythmias, atrial fibrillation, chest pain, pacemaker/ICD) to S:N hotspots. Variants were re-evaluated using American College of Medical Genetics and Genomics (ACMG) criteria; those in S:N hotspots met PM1 criteria. Results: Variants of uncertain significance (VUSs) in HCM genes had a prevalence of 21.8% in the UKBB, with PM1 determining variant call in 25.7% [24.9-26.6] of HCM-gene variants. Using SN enriched for HCM (0.35% with HCM without PM1 and 0.83% with PM1, p < 0.01). InterVar did not enrich for HCM (0.71% with HCM with PM1 and 0.51% without PM1). DiscoVari reduced the likely pathogenic/pathogenic (LPP) variant burden in individuals without HCM (11.8% compared to InterVar (25.9%, p < 0.0001). DiscoVari enriched for LPP variants in those with HCM (36.1%, p < 0.0001) compared to healthy individuals, while InterVar did not. Conclusion: VUSs are common in HCM genes in the population and PM1 is crucial for accurate variant classification. Incorporating DiscoVari S:N into ACMG guidelines improves prediction of variants associated with HCM, or subclinical cardiomyopathy.