Abstracts

Rationale: Sudden Unexpected Death in Epilepsy (SUDEP) is rare, affecting approximately 1 per 1000 individuals with epilepsy per year. The underlying mechanism(s) of SUDEP remain elusive and robustly identified risk factors are limited to the presence and frequency of generalized tonic-clonic seizures. A genetic contribution to SUDEP could include genetic cardiac conditions mimicking epilepsy, the co-existence of epilepsy and inherited arrhythmia condition genes and genetic variants that confer both a risk of epilepsy and a risk of sudden cardiac death. To explore if those who die of SUDEP may be predisposed to sudden death by underlying genetic variants, we investigate the yield of cardiac- and epilepsy- related gene panels in a series of SUDEP cases. Methods: Cardiac arrhythmia and epilepsy gene panels were performed in 39 cases of SUDEP, identified via autopsy by the Ontario Forensic Pathology Service. All cases were sudden and unexpected deaths in individuals with epilepsy in which autopsy did not demonstrate an anatomic or toxicologic cause of death. Testing was performed by GeneDx and included the comprehensive arrhythmia panel including deletion/duplication analysis and comprehensive epilepsy panel including deletion/duplication analysis. At time of sequencing, the epilepsy and arrhythmia panels included 87 and 46 genes respectively. The proportion of cases with at least one variant was calculated. Variants were summarized by in-silico evidence for pathogenicity from 4 algorithms (SIFT, PolyPhen-2, PROVEAN, Mutation Taster), allele frequencies in the general population (GnomAD), and whether they had been previously reported in the literature or by clinical laboratories.Assuming an epilepsy prevalence of 5/1000 and a SUDEP incidence of 1/1000 per year, the method described by Whiffin et al (2017, Genetics in Medicine) determined a maximum credible population allele frequency of 0.00004. Results: Median age at time of death was 33 years (range: 2, 60). Fifty-nine percent (n=23) were male.Of the 39 cases, 28 (72%) had at least one variant identified on either the epilepsy or cardiac panel. Age and sex were similar between groups with and without variants. Among those with at least one identified variant, most individuals (36%) had 1 variant (n = 10) or two variants (29%; n = 8).62 unique variants were detected in 45 genes: 20 on the arrhythmia panel and 25 on the epilepsy panel. All were reported to be variants of uncertain significance by GeneDx except one likely pathogenic EPM2A variant associated with Lafora Disease and one pathogenic SCN1A variant. Missense mutations comprised 57 of the observed variants.Analysis of the cardiac arrhythmia panel missense variants indicated 13 of 27 (48%) had population allele frequencies <0.00004, of which 7 were considered damaging by multiple in-silico algorithms. Nine of the 27 (33%) variants had previously been reported in the literature, including variants in the RYR2, SCN10A and KCNQ1 genes.Ten of 30 epilepsy panel variants had population allele frequencies < 0.00004, of which 8 were considered damaging by multiple in-silico algorithms. Seven of the 30 (23%) variants were previously reported in the literature, and 2 had not been previously identified.Other variant types included duplication of exons (n=1), deletion of exon (n=1), intron variants (n=2), and variants in untranslated regions (n=1). All variants were observed in single cases, except one SCN10A variant,  documented in 3 cases. Conclusions: Targeted epilepsy and cardiac arrhythmia gene panels in SUDEP cases detected variants in the majority of cases; however, the role these genetic variants played in death is unclear. Further work to determine the significance of the identified variants is ongoing. Funding: Ontario Brain Institute