Abstract
OBJECTIVE: SCN1A-related seizures first present as febrile seizures (FS). Definitive features emerge later, making early diagnosis challenging. We evaluated the diagnostic yield and clinical characteristics of pathogenic SCN1A variants in FS, and analyzed the effect of variant characteristics on clinical phenotypes required for early identification and intervention in high-risk patients. METHODS: We analyzed data from a multicenter cohort of 2552 children with FS, no history of afebrile seizures, and no clear alternative diagnosis. Patients underwent gene panel testing for molecular screening. Clinical diagnoses were according to patient history and clinical evaluation. Variant classification followed ACMG guidelines, and the consistency of SCN1A missense variant distribution with the clinical diagnoses and established phenotypic severity was assessed. RESULTS: The diagnostic yield of pathogenic SCN1A variants was 5.5% (141/2552; complex febrile seizure [CFS] vs. simple febrile seizure [SFS] group: 6.8% [95/1389] vs. 4.0% [46/1163], p = 0.001). The proportion of variants associated with Dravet syndrome (DS) did not differ significantly between the groups (χ(2) = 1.048; p = 0.306). Missense variants accounted for 76.5% of all pathogenic SCN1A variants, and among previously reported missense variants, those in critical functional domains were frequently associated with severe phenotypes (p = 0.042). However, these domain-specific patterns were not observed when variants were classified into SFS and CFS groups (p = 0.148). Of 106 patients who completed the 1-3-year follow-up, 33 experienced progression to DS. INTERPRETATION: This study demonstrates the limitations of relying solely on clinical classification for FS risk assessment and highlights the clinical value of molecular screening for early risk identification and individualized management.