Tirzepatide vs. semaglutide: what the head-to-head trial found

Why a head-to-head trial is unusual, and why this one exists

Almost every compound comparison on this site rests on separate trials run at separate times against separate placebo groups. A controlled trial pitting two active compounds directly against each other is expensive, and most manufacturers have no reason to fund one against a competitor.

SURMOUNT-5 is the exception. It is a randomized, open-label, phase 3b trial that put tirzepatide and semaglutide head to head in the same population, at the same time, under the same protocol. That design is rarer than a reader might expect, and it beats asking a reader to compare two separate trials against two separate placebo groups years apart.

It enrolled 751 adults with obesity, or overweight plus a weight-related condition, none of whom had diabetes. Participants were randomized roughly one to one to tirzepatide or semaglutide, each titrated to the maximum tolerated dose within its own approved range, and followed for 72 weeks.

Who funded it, and why that belongs in the first paragraph

SURMOUNT-5 was funded by Eli Lilly, which manufactures tirzepatide, not by a neutral third party and not by Novo Nordisk, which manufactures semaglutide. The trial's own registration and its published funding disclosure both state this.

That fact does not, on its own, invalidate the result. The trial was still randomized, still used the maximum tolerated dose of the comparator rather than a deliberately weak one, and was still published in a peer-reviewed journal after external review. Those are real safeguards against the crudest kind of biased comparison.

It does mean a reader should hold the result the way any single-sponsor trial should be held. It is real evidence, generated under real methodological safeguards. It was also generated by the party with the clearest financial interest in one particular outcome. Both of those things are true, and stating one without the other would be the failure.

What the trial measured, in its own numbers

At 72 weeks, participants assigned to tirzepatide had an average body weight reduction of 20.2%. Participants assigned to semaglutide had an average reduction of 13.7%. The difference between the two arms was statistically significant.

Waist circumference moved the same direction: an average reduction of 18.4 centimeters with tirzepatide against 13.0 centimeters with semaglutide. Looking at how many people reached a given threshold rather than the average alone, 31.6% of the tirzepatide group achieved at least a 25% body weight reduction, against 16.1% of the semaglutide group.

Those are the trial's own reported figures, not a rounded or simplified restatement. What they describe is the specific comparison this specific trial ran: two drugs, each at its own maximum tolerated approved dose, over 72 weeks, in adults with obesity or overweight without diabetes.

What "maximum tolerated dose" means for reading this result

The trial did not compare a fixed dose of one drug against a fixed dose of the other. It titrated each participant to the highest dose they could tolerate within that drug's own approved range, which is the standard the drugs are actually prescribed under.

That design choice is defensible: it compares how the two drugs perform as they are meant to be used, rather than at some arbitrary matched dose neither manufacturer recommends. It also means the result is a comparison of two treatment strategies, not a comparison of two molecules at equivalent exposure.

A reader should not read this result as 'tirzepatide is one-third more potent than semaglutide molecule-for-molecule.' It measured something narrower: what happened when each drug was pushed to the top of its own approved dosing range, in this population, over this time frame. That is a narrower and more useful claim than the popular shorthand version of it.

What the trial did not measure

SURMOUNT-5 enrolled adults without diabetes. Its result does not describe what happens in adults with type 2 diabetes, a population studied separately for each drug and not compared head to head here.

It was open-label, meaning participants and investigators knew which drug each person was taking. That is a real limitation for an outcome like body weight, where knowledge of one's own treatment can plausibly influence behavior, even though the weight measurement itself is objective.

It reported weight and waist circumference at 72 weeks. On its own, it is not a comparison of long-term safety or of outcomes beyond a year and a half. It also says nothing about a compounded version of either molecule, which is a different product entirely and is covered on its own page on this site.

How this fits with each drug's own evidence grade

Both tirzepatide and semaglutide are graded strong evidence on this site, for the same reason each cleared it independently: an approved indication with randomized controlled trial data behind it. SURMOUNT-5 does not change either grade, because the grade describes whether an approved use has controlled trial evidence at all, not which of two approved drugs performs better against the other.

A superiority result in one trial is not the same question as an evidence grade, and this site keeps the two separate on purpose. A drug can lose a head-to-head comparison and still sit at the same evidence tier as the drug that won it. Both were tested the same rigorous way for the question the tier actually asks.

Sources

Each document below is named as it names itself, with the date printed on that document rather than the day it was read.

1. Aronne LJ, et al. Tirzepatide as Compared with Semaglutide for the Treatment of Obesity (SURMOUNT-5) — New England Journal of Medicine, July 2025
2. A Study of Tirzepatide (LY3298176) Compared With Semaglutide in Participants With Obesity or Overweight (SURMOUNT-5), NCT05822830 — ClinicalTrials.gov, National Library of Medicine, July 2025