Abstracts

Rationale: Previous analyses of perampanel Phase III data from patients with POS identified a significant relationship between increases in perampanel plasma levels and reductions in seizure frequency.1 Concomitant CYP3A4 enzyme-inducing antiepileptic drugs (EIAEDs; carbamazepine, oxcarbazepine, or phenytoin) reduced perampanel exposure 2-3 fold but had no impact on the concentration-effect relationship.1 Perampanel has also been investigated in a Phase III, randomized, double‑blind, placebo‑controlled study (study 332) of patients with idiopathic generalized epilepsy (IGE) and uncontrolled PGTC seizures. In Study 332, adjunctive treatment with perampanel improved seizure control.2 Here we present equivalent pharmacokinetic/pharmacodynamic (PK/PD) analyses for Phase III data from patients with POS and PGTC seizures.Methods: Patients aged ≥12 years were enrolled in four double-blind, randomized, Phase III studies. Patients with uncontrolled POS received placebo or perampanel 8 or 12 mg/day (study 304; study 305), or placebo or perampanel 2, 4, or 8 mg/day (study 306). Patients with IGE and uncontrolled PGTC seizures received placebo or perampanel at a target dose of 8 mg/day (study 332). Dose adjustments were permitted at the discretion of the investigators. For the pooled POS and PGTC population, PK/PD models were used to describe relationships between perampanel exposure and percent reduction in seizure frequency per 28 days from baseline to the double-blind treatment period.Results: Perampanel exposure was significantly correlated with reductions in seizure frequency according to a log-linear relationship. Relative reductions in seizure frequency were greater in the PGTC population (n=149), and were characterized by a steeper concentration-effect curve compared with patients in the POS population (n=1109; Figure 1). This suggested that patients with PGTC seizures showed a higher relative response at lower concentrations of perampanel compared with patients in the POS population. In addition, the placebo effect in patients with POS was less than that in patients with PGTC seizures. The presence/absence of CYP3A4 EIAEDs had no impact on the slope of the relationship for either seizure type. There was corresponding attenuation of model-predicted reductions in seizure frequency by perampanel dose in the presence of EIAEDs (Figure 2).Conclusions: In both POS and PGTC seizures, there was no pharmacodynamic interaction between perampanel efficacy and any of the concomitant AEDs; the lower response observed in the presence of concomitant EIAEDs was exclusively due to lower exposure to perampanel. As depicted by the steeper reduction in seizure frequency, patients with PGTC seizures showed a higher relative response at lower concentrations of perampanel compared with POS. Funding: This study was funded by Eisai Inc. Reference 1. Gidal BE et al. Epilepsia 2013; 54: 1490-1497. 2. French J et al. Epilepsy Currents 2015;15(s1):367 abstract 2.389.