Integrating KCNH2 Variant-Specific Features and Heterozygote Phenotypes to Estimate Long QT Penetrance

点击翻译按钮获取中文摘要

PROJECT SUMMARY/ABSTRACT Sequencing an individual’s genome now costs less than many routine medical procedures. A resulting vision is that everyone will have their genome sequenced early in life to enable individualized medical advice about disease prevention and drug selection. A major concern with this vision, however, is proper interpretation of the overwhelming volume of discovered novel and rare variants. In other contexts, a new diagnostic test can be benchmarked and validated in studies that compare large populations with and without a disorder to determine the predictive value of a positive result. In the genetic sequencing context, however, a positive test for most variants cannot be applied to enough heterozygous individuals for a definitive association with disease. Pathogenic variants in KCNH2 (a.k.a. hERG, a cardiac potassium channel gene critical for cardiomyocyte repolarization) can cause sudden cardiac death in the young and can predispose carriers to drug-induced arrhythmias. Genetic variants in KCNH2 are responsible for ~ 6% of autopsy-negative sudden unexplained death in the young and ~ 1% of sudden infant death syndrome cases. Additionally, among heterozygous carriers of variants in KCNH2, women are at greater overall risk of a severe event (including sudden cardiac death) which is increased in the postpartum period. Since most disease-associated variants in KCNH2 are rare, full exome sequencing has the potential to identify those individuals at greater risk early in life before any phenotype manifests. However, even among heterozygous carriers of KCNH2 variants definitively associated with disease, not all develop the same phenotype. To address the challenge of variant interpretation, the American College of Medical Genetics and Genomics suggests criteria to incorporate variant population, functional, computational, and segregation data using several described heuristics. Our foundational hypothesis is that clinically meaningful knowledge is lost in the compression of these variant-specific data to a dichotomous classification. To investigate this hypothesis, we will generate in vitro data for all missense variants in KCNH2 prospectively, build a prediction model of disease penetrance, and validate resulting predictions against the incidence of arrhythmias and cardiac events for variants observed in the Electronic Medical Records and Genomics Network (eMERGE), a Leducq Transatlantic Network, and the UK Biobank.