Abstract
BACKGROUND: In 2011, the US Food and Drug Administration requested that GlaxoSmithKline perform retrospective adjudication of cardiovascular (CV) events reported in the bupropion drug-development trials for smoking cessation. HYPOTHESIS: Retrospective adjudication of clinical trial data will not increase the identification of adverse events. METHODS: We performed a comprehensive retrospective analysis of adverse events in 19 previously completed controlled US clinical trials of bupropion marketed for the treatment of smoking cessation, yielding 9479 subjects (5290 bupropion, 2927 placebo, 1018 active control [ACT], and 244 treated concurrently with bupropion and ACT). All adverse events were sent to the Duke Clinical Research Institute for adjudication by Clinical Events Classification (CEC) physician reviewers. The primary endpoint was a composite of major adverse CV events: CV death, nonfatal myocardial infarction (MI), and nonfatal stroke. RESULTS: Overall, 416 nonfatal CV events in 366 subjects, and 22 deaths, were identified and processed for adjudication. Of these, 7 nonfatal MIs (4 bupropion, 3 placebo, 0 ACT), 5 nonfatal strokes (1 bupropion, 3 placebo, 1 ACT), and 9 CV deaths (4 bupropion, 4 placebo, 1 ACT) were confirmed by the CEC Committee. The primary endpoint occurred in 3/4297 (0.07%) subjects in the bupropion group and in 4/2927 (0.14%) subjects in the placebo group (log-rank P value: 0.613). CONCLUSIONS: CV events in bupropion clinical trials for smoking cessation were uncommon, with no observed increase among subjects assigned to bupropion vs placebo. However, this effort was limited by a paucity of quality data.