Abstracts

Rationale: Data from three randomized controlled trials (2093-301, -302, and -304) demonstrated that eslicarbazepine acetate (ESL) was effective as adjunctive therapy for partial-onset seizures (Biton et al. Epilepsy Curr 2014;14 [Suppl. 1]:209-10). Data from these studies were used to examine the relationship between eslicarbazepine exposure and selected efficacy endpoints. Methods: Each trial comprised an 8-wk baseline period and a 14-wk double-blind period (2-wks titration, 12-wks maintenance). ESL doses ranged from 400mg to 1200mg QD. Eslicarbazepine exposures in individual patients were calculated using a population pharmacokinetic model. Data for the 14-wk double-blind period were analyzed. Predictive models were developed through data analysis, base structural modeling and covariate evaluation, and then validated for concordance between simulated and observed data. Results: Data from 1150 patients (80% Caucasian, 53% male) were included in the analyses; median age was 37 years. At baseline, 48.3% were receiving concomitant carbamazepine; standardized seizure frequency (SSF) was highly variable. Predictive models were developed to describe the relationship between eslicarbazepine exposure and: SSF; probability of response (PR), response being ≤50% reduction in SSF from baseline; and weekly seizure frequency over time. The model predicted a decrease in SSF with increasing ESL dose (ESL 400mg, 5.4; 800mg, 4.6; 1200mg, 4.3; placebo, 6.5 seizures/4wks). The reduction in SSF with ESL (E_max;; Table 1) was predicted to be smaller for patients who were taking carbamazepine at baseline, and in patients from Western Europe (WE). A lower PR was predicted for patients from WE than for non-WE patients. For the WE group, predicted PRs were: ESL 400mg, 0.18; 800mg, 0.22; 1200mg, 0.26; placebo, 0.12. Outside WE, predicted PRs were: ESL 400mg, 0.30; 800mg, 0.35; 1200mg, 0.40; placebo, 0.21. The predicted maximum reduction in weekly seizure frequency from baseline with ESL was 56%. The model indicated that the effect was related to both time (i.e., a placebo effect; 39%) and eslicarbazepine average steady-state concentration (C_av-ss; 61%). The estimated eslicarbazepine EC₅₀ (half maximal effective concentration) was 9450ng/mL; this is similar to the median C_av-ss with ESL 800mg QD. Conclusions: The predicted relationship between exposure and SSF was shallow over the studied concentration range; only slight improvements in seizure control are expected at higher concentrations of eslicarbazepine, and therefore routine monitoring of plasma concentrations is not required. Together with a related analysis of key safety endpoints (Penovich et al. AES 2014, submitted), this analysis supports the ESL dosing recommendations; the optimal risk-benefit profile is associated with a starting dose of 400mg and titration to 800mg or 1200mg QD.