ABSTRACT
Importance Epilepsy is defined as a group of neurological disorders characterized by epileptic seizures, brief episodes of symptoms that are caused by abnormal or excessive neuronal activity in the brain. Epilepsy affects around 3 percent of individuals. In the past 10 years, many groups have been working to better understand the complex genetic mechanisms underlying epilepsy. Together, they studied many different genetic mechanisms, but there is still a substantial missing heritability component in epilepsy genetics.
Objective Here, we used polygenic risk scores (PRS) to quantify the cumulative effects of a number of variants, which may individually have a very small effect on susceptibility.
Design We calculated PRS in 522 French-Canadian epilepsy patients divided into seven subtypes and French-Canadian controls.
Setting All study participants (cases and controls) were selected based on their French-Canadian ancestry.
Participants The epilepsy cohort was composed of families of at least three affected individuals with Idiopathic Generalized Epilepsy (IGE) or Non-acquired Focal Epilepsy (NAFE) previously collected and diagnosed by neurologists following the International League Against Epilepsy (ILAE) criteria.
Exposures All samples were processed on a common genotyping array.
Main outcomes and Results We show that the area under the curve (AUC) is almost always slightly greater than 0.5, especially in patients with IGE and subtypes. We also looked at the association of the PRS with the different phenotypes using a linear mixed effects model estimated by generalized estimating equation (GEE) with the pairwise identity-by-descent (IBD) matrix as a random effect. P-values of GEE were consistent with AUC calculations.
Conclusions and Relevance Globally, we support the notion that PRS and SNP-based heritability provide reliable measures to rightfully estimate the contribution of genetic factors to the pathophysiological mechanism of epilepsies, but further studies are needed on PRS before they can be used clinically.