Nissen 2016: heart safety of naltrexone-bupropion, a trial cut short

What they tested

8,910 overweight or obese adults at higher risk of heart disease were randomly assigned to naltrexone 32 mg plus bupropion 360 mg a day (4,456 people) or placebo (4,454 people) at 266 US centers. Everyone also got an internet-based weight management program. The average age was 61.0, 54.5% were women, 32.1% had a history of cardiovascular disease and 85.2% had diabetes.

The main measure was time to a first major adverse cardiovascular event (MACE): death from cardiovascular causes, a nonfatal stroke or a nonfatal heart attack. The trial was designed to show the drug didn't raise this risk beyond a set limit (a hazard ratio of 1.4), with an early check after about 25% of expected events against a looser limit (2.0) for regulatory approval.

What they found

At the 25% interim analysis, major cardiovascular events had occurred in 35 people on naltrexone-bupropion (0.8%) and 59 on placebo (1.3%), a hazard ratio of 0.59.

After 50% of the planned events, the gap had narrowed: 90 people (2.0%) on the drug and 102 (2.3%) on placebo, a hazard ratio of 0.88 (adjusted 99.7% CI, 0.57-1.34).

Side effects were more common on the drug, including gastrointestinal events (14.2% vs 1.9%) and central nervous system symptoms (5.1% vs 1.2%).

After the sponsor publicly released confidential interim data, the study's academic leaders recommended ending the trial and the sponsor agreed. Because it stopped early, the trial couldn't test the planned limit of 1.4, and the authors concluded that cardiovascular safety remains uncertain.

Limits

• It stopped early, so it never reached the number of events needed for its main safety test.
• Interim results shifted as more events came in: the hazard ratio moved from 0.59 to 0.88 between the two analyses.
• It enrolled people at high cardiovascular risk, most of whom had diabetes (85.2%).

What it means if you're considering naltrexone

This trial is about the weight-loss combination sold as Contrave, not about naltrexone alone for drinking. It left the heart safety of naltrexone-bupropion open: the interim numbers met only the looser limit set for approval, and the trial ended too early to test the stricter one. The site's Contrave guide describes heart safety as unsettled for this reason.