O1009: Reclassifying the Pathogenicity of KCNE1 Variants for Long QT Syndrome

Background: Congenital long QT syndrome (LQTS) is a disorder of impaired cardiomyocyte repolarization characterized by a prolonged QT interval and an increased risk of life-threatening ventricular arrhythmia. The estimated prevalence of LQTS is approximately 1:2000 based on the prevalence of prolonged QT intervals associated with rare variants in the major known causative genes found in a cohort of 44,500 neonates. KCNE1 is estimated to cause 1% to 2% of LQTS, but approximately 85% of disease-associated KCNE1 variants are annotated in the gnomAD v2.1.1 database as variants of uncertain significance (VUS), conflicting, or as having no clinical significance. Only approximately 3.2% of variants are designated as pathogenic and 11.7% are considered likely benign or benign. The Clinical Genome Resource consortium designated KCNE1 as having moderate evidence for causality in autosomal recessive LQTS (Jervell and Lange-Nielsen syndrome type 2), but only limited evidence supporting autosomal dominant LQTS. This study sought to reclassify KCNE1 variants using phenotype segregation analysis using published pedigree data, and newly acquired functional data on 95 variants.

Methods: Variants in the KCNE1 gene were classified in accordance with the standards and guidelines from the American College of Medical Genetics and Genomics (ACMG). Each variant was classified as benign (B), likely benign (LB), variant of unknown significance (VUS), likely pathogenic (LP), or pathogenic (P). KCNE1 variant information was gathered using platforms that compile clinical, functional, and epidemiological information on genetic variation. Literature relevant to each variant was acquired from ClinVar and with Google or PubMed-MEDLINE using variant-specific search strings. Variants were studied using voltage-clamp recordings of heterologously co-expressed KCNE1 and KCNQ1. Differences in current density and voltage-dependent activation from wild-type channels assayed in parallel were used to classify variant function.

Results: Application of ACMG criteria before consideration of functional data resulted in 91 VUS, 2 LP variants, and 2 P classifications of the total 95 variants analyzed. The inclusion of functional data subsequently resulted in a new total of 78 VUS, 14 LP, and 3 P. All newly classified LP variants demonstrated severe LOF in functional assays. Using functional data, we classified 16 severe LOF variants (current density < 25%). When exclusively considering the 23 variants with no counts in gnomAD, 7 of 22 which were initially classified as VUS were reclassified to LP. Only one KCNE1 variant exhibited dominant-negative effects.

Conclusion: We demonstrated that the incorporation of functional data and cosegregation analysis are effective tools for reclassifying novel or rare variants in KCNE1. These reclassifications can help inform KCNE1 carriers and their providers about their risk for LQTS. The majority of KCNE1 variants remain as VUS suggesting that their disease association may require reconsideration.