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 abnormalities. The burden of disease remains high among patients with LGS due to persistent seizures throughout their lifetime and poor prognosis. Study 338 (NCT02834793) was a Phase III study of adjunctive perampanel in patients aged ≥ 2 years with uncontrolled seizures with LGS. Here, we report changes in burden of disease from baseline during Study 338 Core Study using the Impact of Childhood Neurologic Disability Scale (ICNDS).
Methods: Study 338 enrolled patients with clinical and electroencephalogram confirmation of LGS diagnosis with disease onset before 11 years of age, who were receiving 1–4 concomitant anti-seizure medications at stable doses for ≥ 30 days before screening, and had an average of ≥ 2 drop seizures/week during baseline. The Core Study was a randomized, double-blind, placebo-controlled phase (4–8-week screening/baseline, 6-week Titration [2 to ≤ 8 mg/day based on response and tolerability], 12-week Maintenance). The changes from baseline in individual ICNDS realms and the overall score were exploratory endpoints. ICNDS comprises four realms (behavior, cognition,
neurologic/physical limitations, and epilepsy); each realm is scored from 0 (not at all) to 3 (a lot) to evaluate the quality of life (QoL) in pediatric patients based on the parent’s/caregiver’s assessment. The maximum overall score is 132 (33 in each realm); higher ICNDS scores indicate greater impact on behavior/cognition/neurologic disability and epilepsy severity.
Results: Forty-eight pediatric patients were randomized in the Core Study to receive perampanel (n=26) or placebo (n=22). At baseline, ICNDS scores in the overall and individual realms (particularly epilepsy) were high in both groups, indicating great neurologic disability and poor QoL in patients with LGS (Table 1). At Week 18, the mean (standard deviation [SD]) changes in score vs baseline in the behavior realm were -0.6 (4.7) with perampanel and -0.5 (7.3) for placebo; in the cognition realm, the mean (SD) changes vs baseline were ‑1.9 (5.7) with perampanel and -1.5 (8.8) for placebo; in the epilepsy realm, the mean changes vs baseline were similar between perampanel and placebo; in the neurologic or physical limitations realm, a numerically greater mean (SD) reduction vs baseline was observed with perampanel (‑3.8 [7.6]) than placebo (0.0 [6.4]). Following 18-week treatment, the mean (SD) changes in the overall ICNDS scores were -3.1 (16.9) with perampanel and -3.3 (22.9) with placebo. The wide variability of changes in ICNDS scores among patients should be noted.
Conclusions: These results did not provide evidence that adjunctive perampanel negatively impacts burden of disease in pediatric patients with LGS, assessed by ICNDS. However, interpretation of these data is limited due to small sample size.
Funding: Eisai Co., Ltd., Eisai Inc., and Eisai Ltd.