Abstracts

Rationale: Systematic phenotype definition in the epilepsies can help direct the search for susceptibility genes. Prior studies have identified distinct genetic effects on localization related (LRE) vs. generalized epilepsy (GE), myoclonic vs. absence seizures, and the generalized tonic-clonic seizure types within the idiopathic generalized epilepsies (IGEs). Here, we examine the genetic effects on specific seizure semiology within LREs, and on pharmacosensitivity (PS) to antiepileptic drugs (AEDs). Methods: We examined seizure semiology and pharmacosensitivity (PS) in sibling pairs from EPGP, a multicenter collaborative consortium that collects in-depth phenotype and genotype data from a large number of patients with epilepsy to investigate the genetic influences on common and rare forms of epilepsy and pharmacosensitivity. We completed seizure symptom data on 102 sibling pairs with IGE and/or LRE. 51 of these pairs are concordant for IGE, 31 are concordant for LRE, and 20 are discordant (one sibling has IGE, the other LRE). Within the 31 pairs concordant for LRE, we examined concordance of focal seizure symptoms in the following categories: aphasia, autonomic, psychic, sensory, and motor. We also examined the concordance of PS within 20 sibling pairs in which both siblings were definitively classified according to PS. The analytic approach fit a logistic regression model in which presence of a symptom category in the proband was the predictor of presence of the same category in the sibling. Under the null hypothesis of no familial aggregation of symptom types, there should be no association between proband and sibling. The extent to which the estimated log-odds differs from zero was used to test for association. Results: Probands and siblings were significantly associated for aphasic (1-sided p-value=0.003), autonomic (0.04), psychic (0.012), and sensory (0.004) symptom categories. The sibpair association for motor symptoms was suggestive (p=.074). The results of analysis of pharmacosensitivity data were suggestive of sibling concordance, though numbers were too small for statistical significance. Among siblings of probands who were PS (N=12), all were also PS, whereas among siblings of probands who were not PS (N=8), 6/8 were PS (p=0.15). Conclusions: Analysis of focal seizure symptom concordance in EPGP sibling pairs provides evidence for distinct genetic effects on aphasic, autonomic, sensory, and possibly motor symptoms. This evidence can help clarify the role of genes in determining the specific manifestations of the epilepsies, and may allow a priori identification of disease subtypes that are more likely to share susceptibility genes within the focal epilepsies. Analysis of PS data in sibling pairs suggests that there may be a genetic effect on pharmacosensitivity. As enrollment and data collection proceeds in EPGP, analyses using larger numbers of subjects will better address this question, and will allow stratification of PS by epilepsy type.