Next-generation incretin drugs delivered the largest weight reductions without diabetes
A network meta-analysis published in BMJ Medicine compared the weight-loss performance of GLP-1 based drugs in adults with overweight or obesity who did not have diabetes, and the spread between agents was substantial. Across 58 randomized controlled trials involving 24,214 participants, the strongest effect emerged with retatrutide, followed by tirzepatide and CagriSema, while conventional GLP-1 receptor agonists produced smaller reductions in body weight.
Retatrutide produced the largest modeled reduction versus placebo, at -22.10% (95% confidence interval -25.60% to -18.60%). Tirzepatide followed at -19.28% (-20.39% to -18.16%), and CagriSema, a combination of cagrilintide and semaglutide, reached -17.32% (-19.32% to -15.32%). Those differences are not cosmetic. In obesity medicine, a 15% to 20% reduction in body weight can move patients across clinically meaningful thresholds for blood pressure, glycemic risk, mobility, and sleep-disordered breathing, and these estimates place the newer incretin-based agents in a different efficacy tier from older GLP-1 therapies.
Trial network mapped a clear efficacy gradient across incretin pathways
The analysis drew on Embase, PubMed, and Web of Science, covering studies published from 1 January 2000 through 6 March 2026. Eligible trials enrolled adults with overweight or obesity, excluded diabetes, and compared GLP-1 receptor agonists or related co-agonists against placebo or active comparators over at least 12 weeks. That design matters because it isolates the obesity indication and avoids the confounding effect of glucose-lowering treatment, which can alter weight trajectories and adverse-event profiles.
A network meta-analysis is especially useful here because direct head-to-head comparisons are sparse. By linking placebo-controlled and comparator-controlled trials into one analytic framework, the investigators were able to rank agents across a broader therapeutic class than any single study could do alone. The trade-off is familiar to clinicians: the estimates are powerful, but they still depend on the assumptions of indirect comparison, and some intervals overlapped even when the treatment rankings were distinct.
Conventional GLP-1 receptor agonists did work, but the magnitude was more modest than with the newer incretin strategies. Similar patterns appeared for waist circumference and lipid outcomes, suggesting that the body-weight signal was not isolated to the scale alone. Visceral adiposity falls when appetite suppression, delayed gastric emptying, and energy intake reduction persist over time, and those downstream changes often track with improvements in atherogenic lipid measures as fat mass declines.
The ranking analysis favored the next generation of incretin-based treatments, yet the confidence intervals for several comparisons overlapped. In clinical terms, that means the hierarchy is useful for framing expectations, but it does not erase uncertainty around the exact size of the difference between adjacent agents.
Tolerability separated the strongest weight-loss drugs from the most usable ones
Efficacy was only one side of the equation. Low-certainty evidence suggested higher rates of treatment discontinuation with danuglipron and retatrutide, while mazdutide appeared better tolerated. In obesity pharmacotherapy, discontinuation often determines real-world success more than nominal efficacy does, because a drug that drives large weight loss but cannot be maintained rarely delivers its full benefit at the bedside.
One sentence captures the central trade-off: the most potent agents were not automatically the easiest to keep on board.
The analysis did not present a simple winner across all outcomes. Retatrutide led for weight loss, but its higher discontinuation signal means clinicians must weigh efficacy against persistence on therapy, gastrointestinal burden, and individual patient preference. Mazdutide’s better tolerability may make it more practical in some settings even if its weight-loss magnitude is not the highest in the network.
This is where mechanism matters. GLP-1 receptor agonism lowers appetite and slows gastric emptying, but next-generation incretin strategies extend beyond that single pathway. Multi-receptor activity may amplify central satiety signaling and peripheral metabolic effects, helping explain why retatrutide and tirzepatide outperformed conventional GLP-1 agents in the pooled comparisons. The biological logic is consistent with the clinical signal: broader incretin engagement appears to deliver larger weight loss, but with a potential cost in tolerability for some compounds.
What the comparative ranking can and cannot tell clinicians
The findings support a stepped view of obesity pharmacotherapy. If the goal is maximal average weight reduction in adults without diabetes, retatrutide, tirzepatide, and CagriSema sit at the top of the modeled hierarchy. If the goal is long-term use, the tolerability profile becomes just as important as the percentage on the scale. Patients do not benefit from an agent they stop taking after the first few months.
Because most of the evidence came from indirect comparisons, the certainty around some estimates remains limited. Head-to-head trials are still scarce, and the analysis necessarily inherits the usual constraints of network meta-analysis, including heterogeneity in trial populations, dose escalation schedules, treatment duration, and outcome reporting. The comparison is strongest as a map of relative performance, not as a substitute for direct randomized evidence between the leading agents.
The population studied also matters. These were adults with overweight or obesity without diabetes, so the results should not be automatically extended to patients with type 2 diabetes, to older frail patients, or to groups with different baseline cardiometabolic risk. Weight-loss pharmacology is not one-size-fits-all, and the benefit-risk balance can shift quickly when diabetes, cardiovascular disease, or medication adherence issues enter the picture.
A second punchy finding deserves emphasis: the best weight-loss drugs were not the best tolerated drugs.
In practical terms, the study strengthens the case for considering newer incretin-based therapies when substantial weight loss is the target, while also reinforcing the need to monitor discontinuation risk and adverse effects. The evidence base now points to a clear efficacy gradient, but the next clinical question is whether the most potent agents can sustain enough use in routine care to match their trial performance. Residual uncertainty remains around comparative tolerability, long-term maintenance, and the extent to which indirect rankings will hold once more direct head-to-head data arrive.