Efficacy and tolerability of zonisamide versus controlled-release carbamazepine for newly diagnosed partial epilepsy: a phase 3, randomised, double-blind, non-inferiority trial.

BACKGROUND Additional options are needed for monotherapy treatment of adults newly diagnosed with partial epilepsy. This trial compares the efficacy and tolerability of once-daily zonisamide with twice-daily controlled-release carbamazepine monotherapy for such patients. METHODS In this phase 3, randomised, double-blind, parallel-group, non-inferiority trial, adults from 120 centres in Asia, Australia, and Europe, aged 18-75 years and newly diagnosed with partial epilepsy, were randomly assigned (in a 1:1 ratio, done with a computer-generated pseudorandom code) to receive zonisamide or carbamazepine. Patients, investigators, and sponsor personnel giving drugs, analysing outcomes, and interpreting data were masked to treatment allocation. After treatment initiation (zonisamide 100 mg/day vs carbamazepine 200 mg/day [given in two doses]) and up-titration (to 300 mg/day vs 600 mg/day), patients entered a 26-78 weeks flexible-dosing period (200-500 mg/day vs 400-1200 mg/day, according to response and tolerance). Once patients were seizure-free for 26 weeks they entered a 26-week maintenance phase. The primary endpoint was the proportion of patients who achieved seizure freedom for 26 weeks or more in the per-protocol population. This trial is registered with ClinicalTrials.gov, number NCT00477295. FINDINGS 583 patients were randomly assigned to treatment groups (282 zonisamide, 301 carbamazepine), of whom 456 were analysed for the primary endpoint (per-protocol population: 223 zonisamide, 233 carbamazepine). 177 of 223 (79·4%) patients in the zonisamide group and 195 of 233 (83·7%) patients in the carbamazepine group were seizure-free for 26 weeks or more (adjusted absolute treatment difference -4·5%, 95% CI -12·2 to 3·1). The incidence of treatment-emergent adverse events was 170 (60%) in the zonisamide group versus 185 (62%) in the carbamazepine group, of which 15 (5%) versus 17 (6%) were serious and 31 (11%) versus 35 (12%) led to withdrawal. INTERPRETATION Zonisamide was non-inferior to controlled-release carbamazepine--according to International League Against Epilepsy guidelines--and could be useful as an initial monotherapy for patients newly diagnosed with partial epilepsy. FUNDING Eisai Ltd.