Abstracts

Rationale: Few placebo-controlled trials have been performed that assess efficacy of new AEDs for primary generalized epilepsy. In part, this is because some seizure types, such as generalized tonic-clonic seizures are infrequent, leading to long trials with slow enrollment. A new methodology, time to nth seizure, might reduce the study duration, by eliminating the need for a baseline, and reducing treatment period for non-responders. Lamotrigine (LTG) as adjunctive therapy in primary generalized tonic-clonic (PGTC) seizures has been proven effective in a standard baseline/escalation/maintenance trial assessing reduction from baseline seizure frequency. A post-hoc analysis was conducted to assess whether time to nth seizure would have been a successful trial design.Methods: This analysis utilized data from an international, multicenter, randomized, double-blind, placebo-controlled trial with LTG. LTG-naive patients (>2 years, weighing >13 kg) with PGTC seizures, EEG at screen showing generalized spike and wave (GSW) activity or no GSW, and receiving a stable regimen of 1 or 2 AED(s) were enrolled in an 8-week baseline phase. Patients having >3 PGTC seizures were randomized (1:1) to receive either LTG or placebo (PBO). The treatment period consisted of Escalation (12 weeks for 2 to 12 years and 7 weeks for >13 years) and Maintenance (12 weeks) phases with dosing based on background AEDs. Patients were categorized as taking an AED regimen with enzyme inducers (induced), enzyme inhibitors (inhibited) or no effect (neutral). The primary endpoint was percent change from baseline in average monthly PGTC seizures during the entire treatment period. Tns from the start of escalation for n=3, 6, 9 and 12 seizures (T3, T6, T9 and T12) were estimated using Kaplan-Meier analysis with the treatments compared by log-rank test.Results: Of the 117 patients randomized, 58 received LTG and 59 received PBO. There was nearly an even distribution of patients on 1 or 2 concomitant AEDs. The median frequency of PTGC seizures per month during baseline was 2.43 in the LTG group and 2.85 in the PBO group. Tn analysis showed superiority of LTG over PBO for T3, T6, T9 and T12 (p= 0.007, 0.003, 0.003 and 0.022, respectively). This compares to the median percent decrease from baseline in seizure frequency for PGTC seizures for the entire treatment period of 66.5% for LTG and 34.2% for PBO (p=0.006). The Kaplan-Meier plot for T3 is presented in the figure. Of the 117 patients, 74 (63%) had a 3rd PGTC seizure, 60 (51%) had a 6th PGTC seizure, 42 (36%) had a 9th PGTC seizure and 34 (29%) had a 12th PGTC seizure.Conclusions: Time to nth seizure analysis in PGTC seizures demonstrated superiority of LTG over PBO. As PGTC seizure frequency was low, nearly two thirds of patients had fewer than 9 seizures during the trial, and therefore low N’s were optimal. Time to nth seizure appears to be a viable study design for trials of low-frequency events.