Semaglutide's weight edge comes with a gut-trouble bill

Semaglutide, a long-acting GLP-1 receptor agonist, is now approved for chronic weight management and has become one of the most talked-about drugs in metabolic medicine. Much of the earliest evidence came from people with type 2 diabetes, where weight loss was a welcome side effect rather than the goal. The open question is how well it performs — and how well it is tolerated — in adults who carry excess weight but do not have diabetes. Obesity itself is described as a chronic, multifactorial condition marked by excess adiposity and persistent low-grade inflammation, contributing to cardiometabolic illness and death, so the size of any drug effect matters for more than the scales.

The authors ran a systematic review and meta-analysis following PRISMA 2020 guidance. They searched PubMed/MEDLINE, Embase, Scopus and CINAHL from inception to March 2025 for randomised controlled trials of subcutaneous semaglutide in adults with overweight or obesity but without type 2 diabetes. The primary outcome was percentage change in body weight from baseline. Secondary outcomes covered overall gastrointestinal adverse events, nausea, vomiting, diarrhoea, constipation, discontinuation because of adverse events, and serious adverse events. Pooled estimates were calculated with a random-effects model. Four eligible randomised controlled trials involving 3613 participants were included.

Semaglutide produced significantly greater weight loss than placebo, with a mean difference of -11.85% (95% CI -12.81 to -10.90). Gastrointestinal adverse events were significantly more frequent on semaglutide, with nausea, vomiting, diarrhoea and constipation the most commonly reported. Discontinuation of treatment because of adverse events was also significantly higher, with a risk ratio of 2.62 (95% CI 1.70 to 4.03). Serious adverse events, including acute pancreatitis and cholelithiasis, were uncommon in the pooled data. So the efficacy signal was large and consistent, while the tolerability signal was the main counterweight.

Earlier evaluations of semaglutide have often mixed diabetic and non-diabetic populations, which muddies the picture of how much weight loss to expect in people whose only issue is excess adiposity. This analysis narrows the question to non-diabetic adults and confirms a substantial average effect against placebo. It does not settle what happens after the trials end: the authors explicitly call for further long-term randomised trials to evaluate the durability of weight loss and the long-term safety profile. With only four trials pooled, questions about who tolerates the drug, who stops it, and what happens on withdrawal remain open.

This is a reason to treat the large average weight losses reported for semaglutide in non-diabetic adults as reasonably well established across trials, rather than as marketing. It is equally a reason to take the gastrointestinal side effects seriously, since they were significantly more common and were linked to more than double the rate of stopping treatment. Serious events were rare in these trials, which is reassuring as far as it goes, but the follow-up periods were not designed to answer long-term safety. Anyone weighing up a GLP-1 medication is looking at a genuine trade-off between effect size and tolerability, and that conversation belongs with a clinician.