Abstracts

Rationale: LGS is a severe refractory childhood-onset epilepsy characterized by a triad of multiple seizure types (including drop seizures), intellectual disability, and certain electroencephalographic (EEG) abnormalities. High baseline seizure frequency was previously identified as a factor associated with poor seizure control among patients with epilepsy; identification of clinical factors associated with improved response may help inform clinicians on optimal treatment. This post hoc analysis identified predictive clinical factors associated with reduced drop seizure frequency in patients with LGS aged ≥ 2 years with uncontrolled seizures (Study 338; NCT02834793).

Methods: Study 338 enrolled patients with clinical and EEG confirmation of LGS diagnosis who were receiving 1–4 concomitant anti-seizure medications (ASMs), and had an average of ≥ 2 drop seizures/week during baseline. The study consisted of a randomized, double-blind, placebo-controlled Core Study (6-week Titration, 12-week Maintenance), and open-label Extension A (52 weeks) and Extension B (for patients in Japan or countries without an Extended Access Program). Primary endpoint: median percent change from baseline in drop seizure frequency/28 days during the Titration and Maintenance Periods. Secondary endpoints: seizure-freedom rates and safety outcomes. This analysis assessed variables influencing drop seizure frequency using rank analysis of covariance during the Double-blind Period. The base model included treatment, rank-transformed baseline seizure frequency, region, and age group. Baseline seizure frequency, number of ASMs, age at diagnosis, time since diagnosis, and perampanel plasma concentration were assessed as continuous variables, and etiology and seizure type were assessed as categorical variables in univariate/multivariate models.

Results: The Full Analysis Set included 70 patients with baseline and relevant post-baseline seizure data that were included in this analysis. In univariate analyses, there was borderline evidence that increasing plasma concentration resulted in decreasing median percent change in drop seizures (regression coefficient [RC] -0.0103; 95% CI -0.02137, 0.0007; P=0.0671). When treatment was excluded as a variable (plasma concentration and treatment are highly correlated), there was significant evidence that increasing plasma  concentration resulted in decreasing percent change in drop seizures in both univariate (RC -0.0105; 95% CI -0.01909, -0.00192; P=0.0173) and multivariate (RC -0.01076; 95% CI -0.02122, -0.00029; P=0.0441) analyses. No other variables influenced drop seizure frequency (Figures 1 and 2).

Conclusions: The best predictive factor for greater reduction in drop seizure frequency was higher perampanel plasma concentration; no other variables were significant. This is in line with previous analyses, where greater response was linked with higher perampanel plasma concentration. Small patient numbers may limit interpretation of the data.

Funding: Eisai Co., Ltd., Eisai Inc., and Eisai Ltd.