On this page

SURMOUNT-5 compared tirzepatide with semaglutide for 72 weeks in adults without diabetes who had obesity, or overweight with a weight-related complication. The primary modified treatment-regimen analysis estimated mean weight changes of −20.2% and −13.7%, respectively: a difference of −6.5 percentage points. The trial randomized 751 people; 750 received trial treatment and entered the analysis. These are group estimates under the studied regimens, not predictions for an individual.

Trial design

The study enrolled 751 adults with obesity, or overweight with at least one weight-related complication, who did not have diabetes. Participants were randomly assigned to maximum tolerated doses of once-weekly tirzepatide or semaglutide for 72 weeks. The trial was open label: participants and investigators knew which treatment was assigned.

The primary endpoint was percentage change in body weight from baseline to week 72. Waist circumference and categorical weight-loss thresholds were among the secondary outcomes.

SURMOUNT-5 primary result at 72 weeks

Least-squares mean percentage change in body weight at week 72; primary modified treatment-regimen estimand in adults without diabetes.

Tirzepatide−20.2%
Semaglutide−13.7%

Bars are scaled to the larger observed mean, not to 100% body weight. This randomised trial supports a group-level comparison within its studied population and regimens; it does not determine individual treatment choice.

Source: Aronne et al., New England Journal of Medicine 2025;393:26–36. DOI 10.1056/NEJMoa2416394.

Headline result

Week-72 primary estimates: 750 treated participants; modified treatment-regimen estimand
Assigned treatment Analysis participants Mean weight change 95% confidence interval
Tirzepatide 374 −20.2% −21.4% to −19.1%
Semaglutide 376 −13.7% −14.9% to −12.6%

Estimated difference, tirzepatide minus semaglutide: −6.5 percentage points (95% CI, −8.1 to −4.9). The negative sign indicates a greater mean weight reduction with tirzepatide. The interval describes uncertainty in the group comparison; it is not a range of expected individual weight losses.

Why the estimand is called modified: early discontinuation did not by itself remove a participant from the estimated treatment effect. The paper specified exceptions for switching to the other trial medicine outside the study and special handling of weight-reduction surgery. Read those rules before describing the result as a general treatment-policy estimate or an estimate assuming uninterrupted treatment.

6.5 percentage points and about 47% relative difference describe different calculations. Using the rounded means, 20.2 − 13.7 = 6.5 percentage points; (20.2 ÷ 13.7 − 1) × 100 ≈ 47% greater mean weight-loss magnitude. That is not a 47-percentage-point weight reduction or a 47% reduction in a clinical-event risk. See our guide to percentages, percentage points and relative risk.

Source check for this clarification: 26 October 2026. Estimates, denominators and estimand rules were checked against the published trial report, methods and Table 2 and posted SURMOUNT-5 registry results. The relative comparison above is arithmetic using rounded published means.

These are group averages, not predictions for an individual. They do not mean every participant reached the average or that the difference automatically outweighs tolerability, contraindications, access, cost or clinical circumstances.

Safety interpretation

Gastrointestinal events were the most commonly reported adverse events in both groups and were generally described as mild to moderate. A head-to-head efficacy result should not be compressed into “better overall”: safety, discontinuation, long-term maintenance and outcomes beyond weight require their own analysis.

What the trial can establish

  • It directly compares the two assigned treatment strategies in a randomised population without diabetes.
  • It supports a difference in average weight reduction at 72 weeks under the study conditions.
  • It provides comparative waist-circumference and adverse-event observations.

What it cannot establish

  • Which medicine is appropriate for a particular person.
  • Comparative outcomes in every population, including people with diabetes or under-represented clinical groups.
  • What happens over many years or after treatment is stopped.
  • That unapproved or online products bearing either ingredient name are equivalent to regulated medicines.

After treatment stops: use the GLP-1 withdrawal-trial evidence reference to compare continued-treatment, randomized-withdrawal and off-treatment follow-up designs. SURMOUNT-5 does not establish a head-to-head ranking after withdrawal.

Funding and evidence context

The study was funded by Eli Lilly, the manufacturer of tirzepatide. Industry funding does not invalidate a trial, but it is a material context field. Readers should examine the protocol, statistical analysis, outcome hierarchy, missing-data handling and full adverse-event tables rather than relying only on the abstract.

Primary records

Related: How to read a clinical-trial registry record and what trial phases can and cannot establish.

Scientific information only; not medical advice or a treatment recommendation.

Citation record

How to cite this article

Tirzepatide vs Semaglutide: What the SURMOUNT-5 Head-to-Head Trial Actually Shows. Elite Peps Australia. Published 14 September 2026; updated 26 October 2026. https://elitepepsaustralia.com/tirzepatide-vs-semaglutide-surmount-5-head-to-head-trial/

Source checks: Consult the primary studies, trial records and regulator documents linked in this article for current evidence and status. Publication and update dates do not by themselves indicate an evidence review.