Abstracts

Rationale: SCN1A gene mutations are associated with various pediatric epilepsy syndromes. MRI scans typically show normal findings or nonspecific atrophy in these children and are not useful biomarkers.  In the present study, we evaluated the evolution of brain glucose metabolism in children with epilepsy and SCN1A gene mutations who had serial F-18-Fluorodeoxyglucose positron emission tomography (FDG PET) scans. Methods: Four children (3 females) with medically refractory epilepsy (including 1 with infantile spasms) and SCN1A mutation underwent FDG PET scans at around 1 year of age (0.9 ± 0.2 years) and then a few years later (at 5.7 ± 2.0 years; range: 4-8 years). MRI was normal in all four patients. Interictal FDG PET scans were evaluated both visually and semi-quantitatively, by region of interest (ROI) analysis in lobar cortical and subcortical brain regions. FDG uptake in these regions was normalized to basal ganglia values. Basal ganglia was chosen for normalization, as its metabolism appears to be least affected by the epileptic process. Subsequently, normalized glucose metabolism in various brain regions, and its change over time, was compared to those measured in age-matched pseudo-controls (i.e., children with epilepsy and normal development, normal MRI and normal FDG PET on visual assessment; Jeong et al., Human Brain Mapping, 2017;38:3098-3112). Developmental level was correlated with metabolic changes. Results: At the baseline studies, the brain glucose metabolic pattern and various regional values in the SCN1A group were similar to those of the pseudo-controls (Table). At follow-up, decreased normalized FDG uptake values were found in bilateral frontal, parietal and temporal cortex, falling approximately 20% below the mean values of age-matched pseudo-controls (Table). There was only a minimal decrease in occipital cortex, no change in thalamus, but increased normalized values in cerebellum. The two children with maximum/minimum decreases in cortical glucose metabolism had the most/least delay in developmental level, respectively. Conclusions: Children with epilepsy and SCN1A mutation appear to have a normal pattern of cerebral glucose metabolism at around 1 year of age. However, this normal pattern evolves to show various degrees of cortical glucose hypometabolism by age 4 years and later, with maximal decreases in bilateral frontal, parietal and temporal cortex. Together with increased cerebellar and preserved subcortical values, this metabolic pattern may be characteristic of epilepsy associated with SCN1A mutation, and may serve as a useful biomarker to monitor disease progression, neurocognitive outcome and response to treatments being developed for SCN1A gene mutations. Funding: None