Electroencephalographic insights into variant function and clinical outcomes in SCN2A encephalopathy

Abstract Neither phenotype nor genotype reliably predicts clinical disease severity or neurodevelopmental outcomes in SCN2A developmental and epileptic encephalopathy. In this study we examined the electroencephalographic (EEG) features of children with SCN2A variants to quantify the range of EEG abnormalities and link EEG biomarkers to developmental outcomes. We retrospectively analyzed data from a cohort of 28 children with SCN2A variants and employed a genetics-based consensus framework to infer the functional characterization of each subject’s SCN2A variant. Eleven subjects were predicted to have a gain-of-function variant, and 17 subjects a loss-of-function variant. Overall, variant classifications matched subject phenotypes. 493 EEG recordings from the 28 subjects were analyzed (ages 1 day to 16 years). In addition to the SCN2A recordings, normative data from 1230 children without an epilepsy diagnosis or epileptiform features based on neurologists’ review was analyzed (1704 EEG recordings, ages 0 days to 16 years). We detected interictal epileptiform discharges (IEDs) in the SCN2A recordings using Beacon’s automated IED detection algorithm. We characterized background spectral features by computing relative power in four frequency bands (delta=1-4Hz, theta=4-8Hz, alpha=8-13Hz, beta=13-30Hz) in recordings from both the SCN2A and control cohorts. Additionally, we determined whether each SCN2A recording was associated with a gross motor developmental delay based on reported attainment of gross motor milestones. We then used mixed effects logistic regression models to estimate the effect of EEG biomarkers on developmental delay. We characterized EEG abnormalities in the background spectral features of the SCN2A cohort compared to the controls and identified differences in EEG signatures between the subjects with gain- and loss-of-function variants. Additionally, we showed that background spectral features are correlated with motor developmental outcome when measured relative to age-matched neurotypical children. Furthermore, we showed that interictal epileptiform activity is correlated with delayed motor development in subjects with gain-of-function variants. Taken together, these findings suggest that EEG biomarkers can be used to identify neurological abnormalities that correlate both with SCN2A variants and measures of development. We demonstrate the potential value of EEG as a disease biomarker, and we highlight the potential of such biomarkers to both guide future invasive genetic therapies and to be used as diagnostic tools.