Rationale: Epilepsy is one of the most prevalent and troubling features of Tuberous Sclerosis Complex (TSC), with early onset, multiple seizure types and intractability often reported. Less well-described is the day-to-day and patient-to-patient variability in seizure frequency and seizure type. Along with age of seizure onset, these measures can be used to define seizure burden in this population, at both the individual and group levels. Methods: TSC patients ages 0-12 months were enrolled and followed until 36 months in the TSC Autism Centers of Excellence Research Network (TACERN). Caregivers kept daily seizure diaries, and detailed developmental assessments were performed at 36 months. For inclusion in the current analysis, diary compliance had to be >70% and include a minimum of 365 days. Onset for each seizure type was captured, along with daily seizure counts by type. Hierarchical clustering analysis was performed to identify sub-groups based on six metrics of seizure burden. Developmental outcomes at 36 months were compared between subgroups. Results: Epilepsy was seen in 124 of 156 (79%) enrolled participants. Range of seizure onset was 0 to 29.5 months; nearly all (93%) had onset prior to age 12 months (median 4.7 months). Average diary compliance was 85.5%, and 123 participants met diary inclusion criteria. Overall, more than 78,000 patient diary days and >75,000 separate seizures were analyzed. Median reporting duration per participant was approximately 2.5 years (range 498-1261 days). Eighty-eight participants recorded at least one seizure. Focal seizures and infantile spasms were most common. Total seizures ranged from one to 9,128 per patient. Seizures days ranged from one to 803 per patient, representing between 0.1% and 90.0% of each individual’s total reported days. During each patient’s worst seven-day stretch, number of seizures ranged from one to 547 (median 14.5). During each patient’s worst 30-day stretch, number of seizure days ranged from one to 30 (median 18.5). Hierarchical clustering based on metrics of seizure burden (age of seizure onset, total seizures, ratio of seizure to non-seizure days, seizures per seizure day, and worst seven- and 30-day stretches) revealed two distinct patient groups with broadly favorable and unfavorable epilepsy phenotypes (Figure 1). Furthermore, within each group were identifiable subpopulations with differing patterns of seizure burden. Groups with higher seizure burden had worse developmental outcomes at 36 months. Conclusions: Despite high prevalence of epilepsy overall, not all young children with TSC have the same epilepsy phenotype. At least two subpopulations are discernible based on different metrics of seizure burden, with worse developmental outcomes seen in patients with more severe epilepsy profiles. Epilepsy in TSC should be treated aggressively and treated early, with potentially different approaches targeting specific subgroups. Funding: Please list any funding that was received in support of this abstract.: National Institutes of Health U01-NS082320, P20-NS080199, U54-NS092080. Click here to view image/table
