Abstracts

Rationale: Large multiplex families with epilepsy have played a critical role in the discovery of genes harbouring rare pathogenic variants. However, the majority of multiplex families with epilepsy remain genetically unsolved. We now know from genome-wide association studies (GWAS) that common variants play an important role in contributing to epilepsy risk. We aimed to determine 1) whether common genetic variation contributes to epilepsy risk in multiplex families, and 2) whether that risk is enriched in familial versus non-familial (sporadic) epilepsy.

Methods: To capture genetic risk in patients with epilepsy, we calculated polygenic risk scores (PRS) using common variants derived from the largest epilepsy GWAS, performed by the International League Against Epilepsy Consortium on Complex Epilepsies in 2018. PRS based on GWAS results for “all epilepsy,” “generalized epilepsy,” and “focal epilepsy” were computed for four groups of European-ancestry subjects: 1) patients with epilepsy and a positive family history of epilepsy (n=1,584; “familial”), 2) sporadic (non-familial) patients (n=914), 3) unaffected relatives of familial patients (n=652) and 4) population controls (n=16,043). Multiplex families with epilepsy were ascertained from the Epi4K Consortium. Additional familial and non-familial patients were recruited from hospitals in Australia and New Zealand and via the Epilepsy Phenome/Genome Project. Controls were randomly sampled residents of Queensland, Australia, collected for population-based study. Patients were clinically heterogeneous with genetic generalized epilepsy (GGE) and focal epilepsy syndromes accounting for the majority. Using a mixed-effects model to account for familial correlations, we compared PRS scores among all four groups with sex and the first ten ancestry principal components as covariates.

Results: Compared to population controls, PRS scores based on the “all epilepsy” model were significantly increased among familial epilepsy cases (OR=1.3, 95% CI=1.23-1.38, p=2 x 10^-18), sporadic cases (OR=1.1, 95% CI=1.05-1.20, p=0.001), and unaffected relatives of familial cases (OR=1.2, 95% CI=1.06-1.28, p=0.001), (Figure 1). Further, PRS scores were significantly higher in familial versus sporadic cases (OR=1.2, 95% CI=1.05-1.26, p=0.003). Results for the “generalized” and “focal” PRS models were consistent with the “all epilepsy” trends, with those two models most strongly associated with their respective phenotype in familial cases. The greatest enrichment was for the “generalized” PRS model applied to familial GGE cases (OR=1.7, 95% CI=1.59-1.90, p=1 x 10^-33) (Figure 2).

Conclusions: The aggregate effects of many common genetic variants throughout the genome, captured by PRS, are significant predictors of epilepsy risk and are particularly enriched in familial epilepsies. This may begin to explain why, despite progress in gene discovery, the majority of familial epilepsies cannot be explained by a single pathogenic variant. Understanding the contribution of common genetic variants to epilepsy risk may eventually carry clinical implications for all individuals with epilepsy, familial and non-familial.

Funding: Please list any funding that was received in support of this abstract.: Australian Commonwealth Government.