Nationwide Functional Profiling Enables Clinical Reclassification of KCNH2 Variants in the French LQT2 Population

Background: The increasing use of genetic testing has identified thousands of KCNH2 variants associated with long QT syndrome type 2 (LQT2), yet most remain variants of uncertain significance (VUS). Translating functional abnormalities into clinically actionable evidence remains a major challenge. We established a nationwide functional genomics strategy to systematically characterize KCNH2 variants identified in the French LQT2 population and improve their clinical interpretation. Methods: We characterized 303 KCNH2 variants, including 201 VUS, using standardized high-throughput automated patch-clamp electrophysiology and complementary membrane trafficking assays. Nine electrophysiological parameters were integrated using MeTAL to predict variant-specific repolarization phenotypes under homozygous and heterozygous conditions. Modeled QTc values were compared with patient electrocardiographic data and variant-specific penetrance estimates. Functional evidence was calibrated against established benign and pathogenic variants using ClinGen-recommended categorical OddsPath and continuous likelihood-ratio approaches. Results: Multidimensional functional profiling revealed abnormalities not captured by current amplitude alone, identifying additional variants with clinically relevant repolarization defects. Among variants assessed under heterozygous conditions, MeTAL predictions showed 82.9% concordance with patient QTc phenotypes for ACMG class 5 variants, with apparent discordances partly consistent with incomplete penetrance. Continuous likelihood-ratio calibration demonstrated strong discrimination between established benign and pathogenic variants (AUC=0.936) and provided graded functional evidence for 76.1% of VUS. Integration of calibrated functional evidence into the ACMG/AMP framework enabled reclassification of 64 of 201 VUS (31.8%), including 7 as pathogenic, 46 as likely pathogenic, and 11 as likely benign. Conclusions: This nationwide study establishes a scalable framework integrating high-throughput electrophysiology, computational modeling, and clinical phenotyping to translate complex KCNH2 channel dysfunction into clinically interpretable evidence. The resulting reclassification of nearly one-third of VUS, together with the open-access HAMMER database, provides a resource to improve molecular diagnosis and risk assessment in LQT2.