Abstracts

Rationale: Pharmacoresistance and comorbid neurological disorders (NDs) are two key features of epilepsy, contributing to the negative impact of epilepsy on the quality of life, yet our understating of the genetics behind both features is limited. Pharmacoresistance may be of genetic etiology, but the exact degree of genetic contribution is unknown, with no robust evidence for association with any single gene. Whilst genetic overlap between NDs and epilepsy has been quantified epidemiologically, the role of comorbid NDs to genetic analyses and treatment solutions in epilepsy remains elusive. Methods: To identify risk factors for pharmacoresistance, a genome-wide association study (GWAS) was performed in 2354 individuals with pharmacoresistant and 1580 with pharmacoresponsive epilepsy of European ancestry. Pharmacoresistance was defined as: seizures recurring at a frequency of =4/year over the year preceding the latest data entry, despite adequate trials of =2 tolerated and appropriately chosen and used anti-epileptic drug schedules, whether as monotherapies or in combination. Pharmacoresponsiveness was defined as freedom from seizures for =12 months up to the latest recorded visit. The nucleotide polymorphism (SNP) based heritability of pharmacoresistance was estimated in the same sample using LDSC. A linear mixed model GWAS (as implemented in BOLT-LMM) was performed using SHAPEIT/IMPUTE2 imputed Illumina OmniExpress SNP-chip data. To evaluate the predictive power of genetic risk factors for comorbid NDs, we used published ND-GWAS summary statistics in PLINK and R, following best-practices for allelic scoring. Results: The GWAS did not reveal any genome-wide significant associations with pharmacoresistance in epilepsy. In the same sample, the estimated observed-scale single nucleotide polymorphism (SNP) heritability h²_SNP was 0.22 (standard error=0.13). Among four comorbid NDs (Alzheimer’s disease, schizophrenia, major depressive disorder, attention deficit hyperactivity disorder) and two traits (educational attainment and neuroticism), genetic risk factors for Alzheimer’s disease significantly predicted pharmacoresistance in epilepsy patients with P=1.3x10^-4, explaining 0.05% of the trait variance. Interestingly, the best prediction was achieved, when considering only strong associations with Alzheimer’s disease with P<10^-6. Conclusions: This study provides evidence for a common genetic contribution to pharmacoresistance in epilepsy as a phenotype and implicates comorbid NDs in the biology of pharmacoresistance. Specifically, common variation associated with Alzheimer’s disease might help predict each epilepsy patient’s response to pharmacotherapy, potentially adding to the body of evidence increasingly implicating inflammatory processes in epilepsy. Funding: This project has received funding from the European Union’s 7th Framework Programme under grant agreement No 279062, EpiPGX.