Voltage-gated sodium channel epilepsies in a tertiary care center: Phenotypic spectrum with correlation to predicted functional effects.

Background: Variants in sodium channel genes (SCN) are strongly associated with epilepsy phenotypes. Our aim in this study to evaluate the genotype and phenotype correlation of patients with SCN variants in our tertiary care center.

Methods: In this retrospective study, patients with SCN variants and epilepsy who were followed up at our clinic between 2018 and 2022 were evaluated. Our study discussed the demographics of the patients, the seizure types, the age of seizure onset, the SCN variants, the domains and the functions of the variants, the magnetic resonance imaging findings, the motor, cognitive, and psychiatric comorbidities, and the response to anti-seizure medication. Genetic testing was conducted using a next-generation sequencing gene panel (epilepsy panel) or a whole-exome sequencing. For evaluating variant function, we used a prediction tool (https://funnc.shinyapps.io/shinyappweb/ site). To assess protein domains, we used the PER viewer (http://per.broadinstitute.org/).

Results: Twenty-three patients with SCN variants and epilepsy have been identified. Sixteen patients had variants in the SCN1A, six patients had variants in the SCN2A, and one patient had a variant in the SCN3A. Two novel SCN1A variants and two novel SCN2A variants were identified. The analysis revealed 14/23 missense, 6/23 nonsense, 2/23 frameshift, and 1/23 splice site variants in the SCN. There are seven variants predicted to be gain-of-function and 13 predicted to be loss-of-function. Among 23 patients; 11 had Dravet Syndrome, 6 had early infantile developmental and epileptic encephalopathy, three had genetic epilepsy with febrile seizures plus spectrum disorder, one had self-limited familial neonatal-infantile epilepsy, one had self-limited infantile epilepsy and one had infantile childhood development epileptic encephalopathy.

Conclusion: Our cohort consists of mainly SCN1 variants, most of them were predicted to be loss of function. Dravet syndrome was the most common phenotype. The prediction tool used in our study demonstrated overall compatibility with clinical findings. Due to the diverse clinical manifestations of variant functions, it may assist in guiding medication selection and predicting outcomes. We believe that such a tool will help the clinician in both prognosis prediction and solving therapeutic challenges in this group where refractory seizures are common.