Multiplexed Assays of Variant Effect and Automated Patch-clamping ImproveKCNH2-LQTS Variant Classification and Cardiac Event Risk Stratification

• DOI: 10.1101/2024.02.01.24301443
• Publication Date: 2024-02-05T18:35:11Z
• Authors: Matthew J. O’Neill; Chai-Ann Ng; Takanori Aizawa; Luca Sala; Sahej Bains; Annika Winbo; Rizwan Ullah; Qianyi Shen; Chek-Ying Tan; Krystian Kozek; Loren R. Vanags; Devyn W. Mitchell; Alex Shen; Yuko Wada; Asami Kashiwa; Lia Crotti; Federica Dagradi; Giulia Musu; Carla Spazzolini; Raquel Neves; J. Martijn Bos; John R. Giudicessi; Xavier Bledsoe; Eric R. Gamazon; Megan Lancaster; Andrew M. Glazer; Bjorn C. Knollmann; Dan M. Roden; Jochen Weile; Frederick Roth; Joe-Elie Salem; Nikki Earle; Rachael Stiles; Taylor Agee; Christopher N. Jo...; Minoru Horie; Jonathan Skinner; Michael J. Ackerman; Peter J. Schwartz; Seiko Ohno; Jamie I. Vandenberg; Brett M. Kroncke
• Keywords: Risk Stratification; Stratification (seeds)

Abstract Background Long QT syndrome (LQTS) is a lethal arrhythmia syndrome, frequently caused by rare loss-of-function variants in the potassium channel encoded KCNH2 . Variant classification difficult, often owing to lack of functional data. Moreover, variant-based risk stratification also complicated heterogenous clinical data and incomplete penetrance. Here, we sought test whether variant-specific information, primarily from high-throughput assays, could improve both cardiac event large, harmonized cohort missense variant heterozygotes. Methods We quantified cell-surface trafficking 18,796 using Multiplexed Assay Effect (MAVE). recorded current density for 533 automated patch clamping (APC). calibrated strength evidence MAVE according ClinGen guidelines. deeply phenotyped 1,458 patients with variants, including QTc, history, mortality. correlated Bayesian LQTS penetrance estimates phenotypes assessed hazard ratios events. Results scores APC peak tail currents were highly (Spearman Rank-order ρ = 0.69). The found provide up pathogenic very strong severe variants. In cohort, assays significantly predictive events when independently modeled patient sex adjusted interval (QTc); however, became non-significant peak-tail available. area under ROC 20-year outcomes based on patient-specific QTc (AUC 0.80 [0.76-0.83]) was improved prospectively available conditioned 0.86 [0.83-0.89]) or attainable 0.84 [0.81-0.88]). Conclusion High throughput meaningfully contribute at scale while measurements heterozygous Clinical Perspective What new? A two-order magnitude increase set -LQTS provided two complementary Proactively are presented all possible estimation framework data, addition features corrected sex, modeling implications? Readily thousands may facilitate new discovered individuals suspected Scores readily searchable variantbrowser.org community inquiry Both patch-clamp quantitative estimates, prediction large heterozygotes