Science for a Healthier Tomorrow
A meta-analysis of 10 head-to-head studies found tirzepatide produced about 4 percentage points more weight loss than semaglutide — but the evidence is highly heterogeneous, and serious side effects were more common with tirzepatide.
A systematic review pooling 10 head-to-head studies of tirzepatide against semaglutide found that adults with overweight or obesity lost about 4 percentage points more weight on tirzepatide — the same direction as the SURMOUNT-5 trial — but serious adverse events were more common with tirzepatide.
What this means
Pooled across 10 head-to-head studies, adults with overweight or obesity lost on average about 4 percentage points more weight on tirzepatide than on semaglutide — roughly 4.4 kg (about 9.8 lb) more in absolute terms — the same direction as the SURMOUNT-5 trial.
Serious adverse events were more common with tirzepatide (risk ratio 1.83). But the estimate pools mixed designs with very high heterogeneity (I-squared 90%) and no formal certainty grade, so it describes pooled averages between two drugs, not what any one person would experience.
What the review actually compared
The paper is a systematic review and meta-analysis of direct head-to-head studies: comparisons that measured tirzepatide against semaglutide within the same study, rather than inferring one drug’s effect by lining it up against a shared comparator. That distinction is important, because the discovery signal the article surfaced from — a Drug Topics report of a separate, sponsor-commissioned indirect treatment comparison announced at EASD 2026 — used exactly the ‘indirect comparison’ framing this study does not. The review combined 10 studies and 41,381 adults with overweight or obesity, kept to research with at least 24 weeks of follow-up, and searched PubMed, Scopus, CENTRAL and Web of Science from inception to 23 February 2026. It was registered in PROSPERO (CRD420261292090), reported to PRISMA standards, and assessed the randomised trials with the Cochrane RoB 2 tool and the observational studies with ROBINS-I.
Findings:
On the primary outcome — percentage weight change from baseline — tirzepatide was ahead by a mean difference of about 4.3 percentage points (95% CI −5.28 to −3.28; p<0.00001). Among the eight studies that reported absolute weight change, the gap was −4.43 kg (about 9.8 lb; 95% CI −5.56 to −3.30; p<0.00001). Blood-sugar control also tilted toward tirzepatide, with a mean difference in HbA1c of −0.29 percentage points (p=0.0002). The direction matches the pivotal SURMOUNT-5 trial, where 751 adults with obesity but not diabetes lost 20.2% of body weight on tirzepatide over 72 weeks against 13.7% on semaglutide.
The safety trade-off
The weight-loss advantage is not the whole picture. Serious adverse events were more common with tirzepatide: a risk ratio of 1.83 (p=0.007), meaning the pooled rate was about 83% higher with tirzepatide than with semaglutide. Stopping treatment because of side effects, though, did not differ significantly between the drugs (risk ratio 1.28; p=0.54). A higher rate of serious events without a matching rise in discontinuation is a discrepancy the review leaves open, and what counts as a serious adverse event can differ across the studies it pooled. Both safety figures come from only three of the ten pooled studies — the review itself cautions that “these analyses were based on only three studies” — so the serious-adverse-event signal rests on a much narrower base than the 10-study weight-loss pool.
Why the certainty is limited
The review’s headline gap sits under an unusually large warning label. Statistical heterogeneity for the primary outcome was I-squared = 90%, which by convention is very high: it says the studies did not all point to the same underlying effect size. The authors did not run a formal GRADE assessment, so the evidence carries no certainty rating. A funnel plot and Egger’s regression suggested possible publication bias or small-study effects. And the pooled sample is heterogeneous by design — randomised trials and large observational cohorts, different doses, different settings — with the observational cohorts contributing heavily to the total. That last point matters because observational comparisons can carry residual confounding that randomised trials are built to avoid.
What the numbers can’t tell you
This review compares pooled estimates for two groups of people; it does not show what would happen if any one person switched drugs. The supported reading is that across the head-to-head studies to date, adults with overweight or obesity lost on average a few percentage points more on tirzepatide, echoing SURMOUNT-5. It is not evidence that tirzepatide is proven safer or clearly better, and it is not a reason to expect that an individual would lose about 4% more by switching. The clinical meaning of the higher serious-adverse-event rate is unresolved as well: the review does not establish whether it reflects the drug, the populations studied, or differences in how the studies defined and counted events.
Related coverage: why the same GLP-1 drugs work differently for different people (Why do GLP-1 weight-loss drugs work differently for different people?), and what happens when people stop them (What Happens When People Stop Weight-Loss Medications).
Terms explained
- Systematic review and meta-analysis: a study that gathers the existing research on one question according to a preset, reported method, then combines the results statistically into a single pooled estimate.
- Mean difference (MD): the average gap between two groups on a measured outcome, expressed in that outcome’s own units. Here it is the average weight-loss gap between tirzepatide and semaglutide across the pooled studies.
- I-squared heterogeneity: a measure of how much the studies disagree beyond chance. Values near 0% mean the studies agree; the 90% here is very high and signals that they do not all point to one underlying effect.
- Risk ratio (RR): the rate of an outcome in one group divided by the rate in the other. An RR of 1.83 for serious adverse events means that rate was about 83% higher with tirzepatide; an RR of 1.00 would mean no difference.
- Head-to-head (direct) comparison: a study that measures two treatments against each other in the same participants, rather than each against a separate control. Direct comparisons are what this review pooled.
Sources
- Primary source: Pagani Paccola G, Fernandes de Oliveira R, Menão Mochetti M, Pataro Marsola Razera F, Vecchi R, Canale Peres Montanher R. “Comparative Efficacy of Tirzepatide Versus Semaglutide for Weight Loss in Adults With Overweight or Obesity: A Systematic Review and Meta-Analysis of Head-to-Head Studies.” Published 2026-08-30. DOI: 10.1111/cob.70111 · PMID: 42670242 · PMCID: PMC13527552. Peer-reviewed systematic review and meta-analysis of direct head-to-head studies (PROSPERO CRD420261292090; PRISMA-reported). Full text read. The authors declare no conflicts of interest; no funding statement was located.
- Landscape — pivotal head-to-head RCT: “Tirzepatide as Compared with Semaglutide for the Treatment of Obesity (SURMOUNT-5).” N Engl J Med 2025;393(1):26-36. DOI: 10.1056/NEJMoa2416394 · PMID: 40353578 (abstract-level access; phase-3b head-to-head randomised trial, n=751; funded by Eli Lilly).
- Discovery signal (falsely associated with the primary paper in staging — not evidence for it): “Indirect Comparison Links Tirzepatide to Greater Weight Loss Than Semaglutide.” Drug Topics, dated October 3, 2026 (report of an Eli Lilly indirect treatment comparison of tirzepatide 10 mg/15 mg vs semaglutide 7.2 mg from SURMOUNT-1 and STEP UP, presented at EASD). Full text read. It does not cover the primary paper; it is cited only to show a same-week indirect-comparison framing exists in press. Must not be described as ‘press coverage of this review’. drugtopics.com
