Naltrexone–bupropion (Mysimba) in management of obesity: A systematic review and meta-analysis of unpublished clinical study reports
Onakpoya IJ., Lee JJ., Mahtani KR., Aronson JK., Heneghan CJ.
© 2020 The British Pharmacological Society Aims: To compare the benefits and harms of naltrexone–bupropion using evidence from clinical study reports. Methods: We searched Food and Drug Administration and European Medicines Agency websites, PubMed, and Clinicaltrials.gov (May 2016) to identify pivotal trials; we then sent a freedom of information request to the European Medicines Agency (July 2016). We included pivotal, phase III placebo-controlled trials. We assessed the risks of bias using the Cochrane criteria, and the quality of the evidence using GRADE. We used a random-effects model for meta-analyses. Results: Over a 27-month period (July 2016 to August 2018), we received 31 batches of clinical study report documents containing over 65 000 pages of data from 4 pivotal trials (n = 4536). Significantly more participants who took naltrexone–bupropion achieved ≥5% reduction in body weight: risk ratio (RR) = 2.1 (95% confidence interval 1.35–3.28), P =.001, GRADE = low, number needed to treat (NNT) to benefit = 5 (3–17); this represents a 2.53 kg (1.85–3.21) reduction in baseline body weight compared with placebo. Naltrexone–bupropion had significantly beneficial effects on other cardiovascular risk factors; however, the true effect sizes for these are uncertain because of incomplete outcome data. Naltrexone–bupropion significantly increased the risk of adverse events: RR = 1.11 (1.05–1.18, P =.0004, GRADE = low, NNT to harm = 12 7–27); serious adverse events: RR = 1.70 (1.38–2.1, P <.00001, GRADE = moderate, NNT to harm = 21 13–38); and discontinuation because of adverse events: RR = 1.92 (1.65–2.24, P <.00001, GRADE = moderate, NNT to discontinue treatment = 9 8–13). Conclusions: Naltrexone–bupropion significantly reduces body weight by a small amount but significantly increases the risk of adverse events. A rigorous process of postmarketing surveillance is required.