Abstracts

Rationale: A previous double-blind, parallel-group, non-inferiority trial has shown levetiracetam (LEV) and controlled-release carbamazepine (CBZ-CR) to produce equivalent seizure freedom rates in patients with newly diagnosed partial-onset seizures (Brodie et al., 2007;68:402-408). The purpose of the present open-label study was to assess the efficacy and tolerability of LEV, extended-release sodium valproate (VPA-ER) and CBZ-CR as monotherapy in patients with newly diagnosed partial or generalized epilepsies. Methods: In this open-label, parallel-group, 52-week trial, patients (≥16 years old) with ≥2 unprovoked seizures in the past 2 years and ≥1 in the last 6 months were stratified to physicians' best recommended treatment (VPA-ER or CBZ-CR). By randomization, patients were allocated to best recommended treatment (VPA-ER or CBZ-CR) or LEV. The initial target doses for each treatment were 1000 mg/day for VPA-ER, 600 mg/day for CBZ-CR and 1000 mg/day for LEV. The primary endpoint was the time to withdrawal. Results: 1688 patients (ITT population) were randomized to receive either LEV (n=841) or standard AEDs (n=847). The mean age of the population was 41 years and 44% were female. Patients experienced a median of 3 seizures in the last 2 years (partial-onset, 62%; primary generalized, 32%; mixed seizure types, 6%) and had a median epilepsy duration of 0.9 years. Completion rates (52-week treatment) were similar between groups (LEV, 76.0%; standard AEDs, 74.0%). Time to withdrawal was not significantly different between treatment groups overall and by strata: hazard ratios (95% CI) i) LEV/standard AEDs, 0.90 (0.74-1.08), p=0.258 (Fig. 1); ii) LEV (n=349)/VPA-ER (n=347), 1.02 (0.74-1.41), p=0.882; iii) LEV (n=492)/CBZ-CR (n=500), 0.84 (0.66-1.07), p=0.161. The hazard ratios (95% CI) for time to first seizure after treatment initiation were i) LEV/standard AEDs, 1.20 (1.03-1.39), p=0.022; ii) LEV/VPA-ER, 1.19 (0.93-1.54), p=0.167; iii) LEV/CBZ-CR, 1.20 (0.99-1.46), p=0.061. The discontinuation rate due to lack of efficacy was 4.2% for LEV-treated patients and 3.0% for those patients receiving standard AEDs (VPA-ER, 3.5%; CBZ-CR, 2.6%). Discontinuation rates due to adverse events were 8.4% and 13.0% for patients in the LEV and standard AEDs groups (VPA-ER, 4.6%; CBZ-CR, 18.8%). A similar proportion of patients in all treatment groups experienced ≥1 drug-related adverse event (LEV, 45.6%; VPA-ER, 45.9%; CBZ-CR, 52.3%). The most frequently reported treatment-emergent adverse events were headache (LEV, 19.3%; VPA-ER, 17.0%; CBZ-CR, 22.4%), fatigue (LEV, 14.4%; VPA-ER, 11.4%; CBZ-CR, 19.0%), and weight increase (LEV, 5.6%; VPA-ER, 19.0%; CBZ-CR, 6.6%). Conclusions: In this large open-label, randomized study, LEV demonstrated similar effectiveness to standard AED treatments as monotherapy in patients with newly diagnosed partial and generalized epilepsies, further supporting its broad spectrum of efficacy. Study sponsored by UCB.