Abstracts

Rationale: The effect of first generation antiepileptic drugs (AEDs) on cognition and behavior in patients with epilepsy has been widely studied, however, less information is available about children and adolescents with epilepsy treated with newer AEDs. We report the neurocognitive effects of levetiracetam (LEV, Keppra^®) as adjunctive treatment for partial-onset seizures (POS) in children 4-16 years old.Methods: This is a randomized, double-blind, multicenter, placebo-controlled safety study. Eligible patients having ≥1 POS during 4 weeks prior to screening, treated with 1-2 AEDs, with IQ ≥65 were randomized (2:1) to either LEV or placebo (PBO) for 12-week treatment. LEV dose was titrated up to 60 mg/kg/day followed by an 8-week maintenance period. Neuropsychological measures were administered at baseline and Week 12. The Leiter International Performance Scale-Revised (Leiter-R) Attention and Memory Battery Memory Screen Composite score change from baseline was utilized as the primary endpoint for non-inferiority hypothesis tests with non-inferiority margin of -9.0. Per protocol (PP) population was used for primary analyses. The Wide Range Assessment of Memory and Learning-2 (WRAML-2) indices were utilized as secondary endpoints. Adverse events, ECG, and laboratory tests were assessed during the treatment period.Results: 99 patients were randomized with 73 (46 on LEV, 27 on PBO) in PP population. Mean dose for LEV patients during maintenance was 55.5 mg/kg/day. The Leiter-R Memory Screen Composite Change from Baseline Least Square Means (LSM) and standard error (SE) was 5.36(1.78) for LEV; 5.17(2.33) for PBO, with LEV-PBO LSM and 2-sided 90% confidence interval (CI) of 0.19 (-4.69, 5.08). Since the 90% CI lower bound was greater than the non-inferiority margin of -9.0, non-inferiority of LEV vs. PBO was demonstrated. No statistically significant differences between LEV and PBO at 5% level were observed in WRAML-2 Indices. The change from baseline LSM (SE) and 95% CI of the difference between LEV and PBO for each index were: General Memory, LEV 3.43(1.54), PBO 6.86(2.13), 95% CI (-8.75, 1.89); Visual Memory, LEV 7.31(1.95), PBO 10.76(2.70), 95% CI (-10.19, 3.29); Verbal Memory, LEV 2.62(1.74), PBO 1.29(2.42), 95% CI (-4.65, 7.32); Attention/Concentration, LEV -0.18(1.34), PBO 1.96(1.86), 95% CI (-6.73, 2.44). Of 98 Intent-to-Treat patients, 89.1% of LEV and 85.3% of PBO patients reported ≥1 Treatment Emergent Adverse Events (TEAEs). The most frequently reported CNS related TEAEs were headache (LEV 26.6% vs. PBO 14.7%), somnolence (LEV 14.1% vs. PBO 8.8%) and aggression (LEV 12.5% vs. PBO 8.8%).Conclusions: The neurocognitive effects, mainly memory or attention as measured by well standardized neuropsychological instruments, were not different between LEV and placebo as adjunctive therapy in treatment of POS in patients of 4-16 years. UCB Funded.