Abstracts

Rationale: The standard evaluation for epilepsy includes clinical history, physical exam, blood tests, EEGs, and imaging. Despite this workup, the etiology of epilepsy may not be known in up to 60-70% of patients. Recently, genetic testing has become more readily available and less expensive, making it practical for routine clinical testing especially when standard workup is inconclusive. While there are over 500 established epilepsy-associated genes (Weber, et al. Expert Rev Mol Diag 2017), genetic testing often leaves clinicians with the dilemma of variants of uncertain significance (VUS). In order to better categorize VUS, more data are necessary to understand their clinical impact. Because most epilepsy genetic studies have been performed on pediatric populations, the purpose of our study was to identify possible genetic causes of epilepsy in a cohort of adult patients.  Methods: Genetic testing was performed on patients at the Stanford Comprehensive Epilepsy Center as part of their routine evaluation at the discretion of the ordering clinician. Genetic etiology was explored using an epilepsy gene panel (ranging from 87-182 genes) or via whole exome sequencing. See Table 2 for a list of genes. Diagnostic yield was calculated based on the frequency of identified likely pathogenic/pathogenic variants.  Results: Fourteen patients (10 female and 4 male) were included in the study. See Table 1 for demographics. Age range was 18-52y (mean 28.6y). Genetic testing identified some variant in 12/14 (86%) of patients. Out of those, 6/12 (50%) had more than one variant reported. 3/14 (21%) had at least one known pathogenic variant and 11/14 (79%) had at least one VUS. There was no other clearly identified cause for epilepsy in 8/12 (67%) of patients with any genetic mutation. All VUS mutations were heterozygous with 10/20 (50%) autosomal dominant and 50% autosomal recessive. There was a positive family history of seizures in 7/11 (64%) patients specifically with VUS. Furthermore, there was no other identifiable epilepsy risk factor in 7/11 (64%) patients with VUS. Notably, patients 7 and 8 are identical twins with the same VUS, but different clinical phenotypes.  Conclusions: In our cohort, 21% of epilepsy patients had a pathogenic variant. This falls into a range from previous genetic studies in the pediatric epilepsy population (Wang, et al. Sci Rep 2017; Oates, et al. NPJ Genom Med 2018; Butler, et al. Pediatr Neurol 2017). Because most of these studies were skewed by neonatal and infantile epilepsy patients (which may carry higher rates of genetic epilepsies), our pathogenic variant rate is much higher than expected. Additionally, 79% of our epilepsy patients had a VUS. Of these patients, 64% had a positive family history of seizures and 64% had no other identifiable risk factor for epilepsy. While it is possible that the VUS mutations are not significant, the substantial family histories and the low rates of other epilepsy risk factors in this group suggest that these VUS may contribute in some way to their epilepsy. Interestingly, there were two identical twins with the same VUS, but different epilepsy risk factors. This pair exemplifies the heterogeneity of genotype-phenotype expression and supports the need for larger studies in adult epilepsy patients and control patients to better understand how a genetic variant (whether pathogenic or VUS) may contribute to disease. This would allow patients to understand their disease and plan for their families. Furthermore, a better understanding of genetics may help therapy decisions. If we can identify mutations involving specific ion channels/proteins, we may be able to tailor treatment as we do for patients with SCN2A mutations.  Funding: No funding