Semaglutide beats insulin on weight and lows

Type 2 diabetes affects over 537 million adults worldwide, and when tablets stop controlling blood glucose, treatment usually escalates to an injection. The choice at that point is consequential: basal insulin has decades of use behind it, while GLP-1 receptor agonists offer glucose control alongside weight change. Until now, no systematic review had directly compared once-weekly semaglutide against both basal insulin and the other injectable GLP-1 agonists in people who had never used insulin. That gap leaves clinicians and patients guessing about the relative trade-offs between glucose control, weight, hypoglycaemia risk and tolerability.

The reviewers searched five databases from inception to April 2026 for phase 2b-4 randomised controlled trials lasting at least 12 weeks that compared once-weekly semaglutide with injectable therapy in insulin-naive adults with type 2 diabetes inadequately controlled on oral antidiabetic drugs. Comparators were insulin glargine, exenatide ER, dulaglutide and liraglutide. Two reviewers handled selection, extraction and risk-of-bias assessment using Cochrane RoB 2, and certainty of evidence was graded with GRADE. Random-effects pairwise meta-analyses pooled the results. Four SUSTAIN trials met criteria, totalling 3,680 participants, with 2,705 analysable for the primary comparison.

Semaglutide 1.0 mg reduced HbA1c compared with all comparators, by a mean difference of -0.64% (95% CI -0.80 to -0.47), rated low certainty. Body weight fell by a mean of 4.38 kg more (95% CI -5.76 to -3.01), and against the other GLP-1 agonists specifically the weight advantage was highly consistent at -3.72 kg (I² = 0%, moderate certainty). Systolic blood pressure was 2.32 mmHg lower, and reaching HbA1c below 7.0% was 1.60 times more likely. Hypoglycaemia risk was roughly half that of insulin glargine (RR 0.53). Against that, gastrointestinal adverse events and treatment discontinuation were both higher with semaglutide than with insulin.

The findings line up with the general expectation that GLP-1 agonists deliver weight reduction that insulin does not, and they put numbers on semaglutide's edge over the other drugs in its own class. The authors themselves flag serious limits on confidence: every included trial was open-label, heterogeneity was present in several comparisons, and all four trials were funded exclusively by the manufacturer of semaglutide. Only the weight advantage over other GLP-1 agonists, the blood-pressure change and the reduced hypoglycaemia risk reached moderate certainty. The authors call explicitly for independent trials.

If you are reading about GLP-1 drugs in the context of weight, this review is a reminder that the evidence base is real but narrower than the headlines imply: four industry-funded, open-label trials in people with type 2 diabetes, not in healthy adults seeking weight loss. The size of the weight difference against other GLP-1 agonists appears reasonably solid; the glucose benefits are less certain. The tolerability trade-off is also part of the picture, with more gut side effects and more people stopping treatment than on insulin. Any decision about these medicines belongs with a clinician who knows your history.