Abstracts

Rapid protection against seizures is an important aspect of epilepsy care when initiating treatment with an antiepileptic drug. The objective of this analysis is to determine whether rapid achievement of steady state of levetiracetam (LEV) plasma levels is translated into an immediate measurable efficacy. To determine this, we measured time to onset of antiepileptic action after LEV initiation.
Investigations were conducted on a pooled database of patient diary cards for the three randomized, double-blind, placebo-controlled, multicenter studies with adjunctive LEV in patients with refractory partial onset seizures. These trials included a minimum of 8 week baseline period; a 4-6 week titration followed by a 12-14 week evaluation period. Drug starting dose was either 1000mg/D or 333mg/D. The treatment effect was assessed by computing the proportion of patients free of any seizure type on a daily basis. Time to onset of action was evaluated by comparing proportion of seizure-free patients on a given day before and after drug administration. McNemar test was used for within group comparison while Logistic Regression (LR) with last baseline day value (D-1) as covariate was used for between group comparison. Odds-Ratios (OR) and 95% confidence intervals are presented. Out of 904 patients, 883 were evaluated in the pooled analysis. Ninety-six initiated LEV at 333 mg/day, 480 at 1000 mg/day and 307 were on placebo.
The proportion of seizure-free patients was stable during baseline with values ranging between 67 and 71% for D-3 to D-1. The effect of LEV was observed as of D+1 with an increased proportion of seizure free patients over baseline of 15, 17 and 17% for the D+1,D+2 and D+3 for the 1000 mg LEV group (p value[lt]0.001). For the 333mg LEV group, the increase was respectively 7, 9, and 9% for D+1, D+2 and D+3 (differences NS). No major changes were observed in the placebo group. Regarding differences between LEV 1000 mg and placebo, OR(LR) were 2.3 (1.58-3.21), 2.5 (1.69-3.54) and 2.7 (1.84-3.86) respectively for D+1 to D+3 (all differences statistically significant). For the 333 mg, differences did not reach statistical significance.
Evidence of a rapid onset of action was demonstrated through a significant increase in proportion of seizure-free patients as of D+1. Results based on ORs indicate that for D+1 to D+3, seizure freedom is at least twice as likely to occur in the LEV 1000 mg than in the placebo group.
[Supported by: UCB Pharma SA]