SCN1A variants from bench to bedside-improved clinical prediction from functional characterization.

Abstract

Variants in the SCN1A gene are associated with a wide range of disorders including genetic epilepsy with febrile seizures plus (GEFS+), familial hemiplegic migraine (FHM), and the severe childhood epilepsy Dravet syndrome (DS). Predicting disease outcomes based on variant type remains challenging. Despite thousands of SCN1A variants being reported, only a minority has been functionally assessed. We review the functional SCN1A work performed to date, critically appraise electrophysiological measurements, compare this to in silico predictions, and relate our findings to the clinical phenotype. Our results show, regardless of the underlying phenotype, that conventional in silico software correctly predicted benign from pathogenic variants in nearly 90%, however was unable to differentiate within the disease spectrum (DS vs. GEFS+ vs. FHM). In contrast, patch-clamp data from mammalian expression systems revealed functional differences among missense variants allowing discrimination between disease severities. Those presenting with milder phenotypes retained a degree of channel function measured as residual whole-cell current, whereas those without any whole-cell current were often associated with DS (p = .024). These findings demonstrate that electrophysiological data from mammalian expression systems can serve as useful disease biomarker when evaluating SCN1A variants, particularly in view of new and emerging treatment options in DS.