Abstracts

Rationale: In integrated safety analyses and in the literature, estimates of long-term retention on antiepileptic drugs (AED) consider all patients exiting a study to have discontinued. However, patients might discontinue studies for reasons that are not study drug-related. A less conservative approach is to use Kaplan-Meier methods to censor all patients who discontinue for reasons other than lack (or loss) of efficacy or adverse events (AE). Conversely, most published analyses of long-term AED retention include only patients who entered open-label extension (OLE) studies and as such do not account for discontinuation in the preceding double-blind (DB) studies. A more conservative approach is to define retention as the time from first dose to discontinuation, thus including DB data for patients randomized to active drug. Brivaracetam (BRV) is a high-affinity synaptic vesicle protein 2A ligand. Disposition data for DB studies of BRV have been presented previously and showed low rates of discontinuation. This study assesses the distribution of reasons for and time to discontinuation of BRV in DB and OLE studies.Methods: This study included patients from six DB Phase 2 and Phase 3 studies and four ongoing OLE studies of BRV as adjunctive treatment in adults with partial-onset seizures. Patients were included if their modal daily dose was between 50 mg and 200 mg. The start of exposure was defined as the day of first BRV dose; namely Day 1 of the DB studies for patients randomized to BRV and Day 1 of the OLE studies for patients initially randomized to placebo. Kaplan-Meier methods were used to estimate time to discontinuation due to AE or lack of efficacy. Patients still ongoing at the time of analysis, or discontinuing due to reasons other than the aforementioned, were censored. Retention rate at Week 12 and Week 52 was estimated.Results: The first exposure to BRV in the included DB studies was in July 2005. At the time of analysis (January 2014 data cut-off), the OLE studies were still ongoing. From a total of 2051 patients exposed to BRV and eligible for this analysis, 42.6% remained in the OLE, 12.2% had discontinued due to AE, 23.1% due to lack of efficacy, and 22.1% due to other reasons. After 12 weeks of exposure, the Kaplan-Meier estimated retention rate was 95.9%. The rate of discontinuation due to AE, after censoring all other discontinuation reasons including lack of efficacy, was 3.8%; due to lack of efficacy, after censoring AE and all other discontinuation reasons, it was 0.3%. After 52 weeks of exposure, the Kaplan-Meier estimated retention rate was 79.8%. The rate of discontinuation due to AE was 8.7%; due to lack of efficacy, it was 12.6%.Conclusions: Long-term retention rates on BRV were found to be high with about 20% of patients having discontinued due to lack of efficacy or AE by Week 52. The method presented here differs in several respects from what is generally reported in the literature and as such the results are difficult to compare. UCB supported.