Abstracts

Rationale: The advent of next-generation sequencing has led to an exponential increase in the identification of epilepsy-related genes.  Multi-gene epilepsy panels are now widely available, with test yields reported of 14-57%, depending upon the age at seizure onset, patient phenotype and the test methodology.  Whole exome sequencing (WES) has also become increasingly utilized, especially for patients with refractory epilepsy and complex phenotypes, with test yields reported of up to 39%.  However, WES findings reported to date include gene variants detectable by panel testing, and there are no published data regarding the yield of WES among patients with previous non-diagnostic epilepsy panel results.  Here we report on outcomes of WES among a cohort of pediatric patients with refractory epilepsy that had previous non-diagnostic epilepsy gene panel results. Methods: We reviewed the medical records of all patients with refractory epilepsy seen at Boston Children’s Hospital who had WES between January of 2013 and June of 2017.  We excluded patients who did not have a primary epilepsy phenotype, patients whose epilepsy panel results were diagnostic, and patients for whom an epilepsy panel had not been performed prior to WES. We categorized the WES results as either positive, negative, candidate gene or VUS. Results: We identified 56 pediatric patients with refractory epilepsy who had non-diagnostic epilepsy panels and subsequent WES.  Forty-nine of the tests were sent as trios. A total of 11 patients (20%) had positive WES results, 13 (23%) had candidate gene findings, 13 (23%) had VUS findings (all of which we determined to be non-diagnostic) and 19 (33%) had negative results.  Among the 7 patients sent without both parental samples, 1 (14%) received a positive result. Diagnostic results included autosomal recessive phenotypes associated with ALG11, ASNS, CAD, PROSC, SEPSECS and TBCK, and autosomal dominant phenotypes due to de novo pathogenic variants in CHD2, GABRG2, GNAO1 and SCN8A. Candidate genes included AQP1, AGO1, DOCK1, CASR, HSD17B4, NRXN2, PACS2, PPP3CA, SLC12A4, SLITRK4, THOC2 and ZW10. Conclusions: In our cohort of patients with refractory epilepsy, all of whom had prior epilepsy panels that were non-diagnostic, the positive yield of WES was lower than what has been previously reported.  This is likely due to the fact that previous studies included patients who were panel naïve, such that positive WES results included gene findings that would have been identified on panels if they had been performed. Indeed, 3 of the 11 patients in our cohort were found to have pathogenic variants in either SCN8A or CHD2. These were not detected through previous epilepsy panel testing, as neither SCN8A nor CHD2 had been identified as epilepsy genes when the panel testing was performed.  Despite the lower WES yield, the positive rate of 20% is still considerable, and when taken in conjunction with the 23% candidate gene rate suggests that WES testing is indicated for pediatric patients with refractory epilepsy who have had prior non-diagnostic epilepsy panel testing. Funding: N/A