Abstracts

Rationale: The epilepsies are a heterogeneous group of disorders with strong evidence for heritability in many forms. Genome-wide association studies in the epilepsies have had limited success in identifying risk loci probably due to relatively small sample sizes and lack of power. Here we report three meta-analyses conducted for all epilepsy and the two largest clinical subgroups -genetic generalized epilepsy (GGE) and focal epilepsy. Methods: We examined 12 case/control cohorts comprising 8,696 cases and 26,157 controls. Subjects were phenotyped into categories of GGE, focal epilepsy and unclassified epilepsy. A fixed-effects meta-analysis was performed after standardized imputation to the 1000 Genomes reference (August 2011) and association using a linear mixed model. Logistic regression of genome-wide significant signals was conducted as a technical validation and to estimate odds ratios. Results: Meta-analysis of the "all epilepsy" cohort identified loci at 2q24.3 implicating SCN1A, and at 4p15.1 harboring PCDH7 as the lead candidate. For the GGE cohort, a single signal at 2p16.1 was observed, possibly implicating VRK2 or FANCL. No SNP achieved genome-wide significance for focal epilepsy. Conclusions: Taken together, this first meta-genome-wide association study in epilepsy has uncovered new loci for the common forms of epilepsy. The finding that individual loci can have pleiotropic or specific epilepsy phenotype effects suggests that future genetic analysis may benefit from both "lumping" and "splitting" clinical subgroups of epilepsy.