Retatrutide Results: What the Trials Reported

By Evan Marsh, EditorUpdated Every figure labeled by source tier and analysis convention

Retatrutide has produced the largest weight reductions ever reported for a pharmacotherapy in an obesity trial. That sentence is doing a lot of work, so this page unpacks it: which number came from which trial, which analysis it used, whether it was peer-reviewed, and which widely quoted figure describes a subgroup rather than a study.

Phase 2: the peer-reviewed dose-response

This is the strongest evidence available, published in the New England Journal of Medicine: 338 adults with obesity, 48 weeks, six dose groups plus placebo. It is the only retatrutide obesity trial to have been through peer review, and it is the one table worth memorizing because it shows the dose-response cleanly.

DoseMean weight change at 24 weeksMean weight change at 48 weeks
Placebo−1.6%−2.1%
1 mg−7.2%−8.7%
4 mg−12.9%−17.1%
8 mg−17.3%−22.8%
12 mg−17.5%−24.2%

Two things stand out. The dose-response is steep between 1 mg and 8 mg and then largely flattens — 8 mg and 12 mg were close at 24 weeks and separated by only 1.4 points at 48. And the weight curve had not plateaued when the trial ended, which earlier incretin trials generally did much sooner.

Phase 3 TRIUMPH: larger, longer, not yet peer-reviewed

TrialPopulationDurationWeight reductionEvidence tier
TRIUMPH-1Obesity without diabetes, n=2,33980 weeks−28.3% efficacy / −25.0% treatment-regimenCompany topline
TRIUMPH-1 extensionSelected subgroup, n=532104 weeks−30.3% efficacy / −29.9% treatment-regimenCompany topline
TRIUMPH-2Type 2 diabetes + obesity, n=1,15280 weeks−20.8% efficacyCompany topline
TRIUMPH-3Severe obesity + cardiovascular disease, n=1,94980 weeks−22.6% efficacyCompany topline

Every Phase 3 figure above comes from Eli Lilly press releases and conference presentations rather than a published paper. That is a weaker tier of evidence and the numbers can move on the way to publication. For TRIUMPH-2 and TRIUMPH-3, Lilly published only the efficacy estimand, so no count-everyone figure exists for those two.

The 30.3% figure needs its caveat attached every time. It is not the whole-trial result at 104 weeks. The extension enrolled only the 532 participants who had a BMI of 35 or above at baseline, completed the 80-week study and tolerated their assigned dose — then escalated them to their maximum tolerated dose for another 24 weeks. It describes a selected, tolerant, higher-BMI subgroup. Placing it beside another drug's whole-population number overstates retatrutide, and that comparison is made constantly.

Which number to use when comparing drugs

Obesity trials report every result twice: once counting all randomized participants whether or not they stayed on the drug, and once assuming everyone adhered. Lilly calls these treatment-regimen and efficacy; Novo Nordisk calls the same two things treatment policy and trial product. The gap between conventions runs 1.6 to 3.6 percentage points — bigger than the difference between some neighbouring drugs. Any cross-drug comparison has to use the same convention on both sides, which is what ourGLP-1 comparison table and thehead-to-head pages do.

Results beyond the scale

The TRIUMPH program measured more than body weight, and these findings are less publicized:

  • Liver fat. A Phase 2a trial in metabolic dysfunction-associated steatotic liver disease reported an 81–82% relative reduction in liver fat at the 8 and 12 mg doses — a direct consequence of the glucagon component.
  • Obstructive sleep apnea. Studied in nested sub-studies inside TRIUMPH-1 and TRIUMPH-2 in participants with an apnea-hypopnea index of 15 or above, where the reported reduction reached 36.1 events per hour from a baseline of 58.6.
  • Body composition. A Phase 2 substudy in type 2 diabetes separated fat mass from lean mass changes — relevant to the lean-mass question covered on the side effects page.
  • Glycemic control. The Phase 2 diabetes trial reported dose-dependent HbA1c reductions.

What community reports can and cannot tell you

Self-reported results circulate widely — progress threads, before-and-after photos, spreadsheets of weekly weights. We do not republish specific figures from them, and the reasoning is worth stating plainly rather than hiding behind a disclaimer.

An anonymous report has no verified starting weight, no verified dose, and — the part that matters most here — no verification of what was in the vial. Research-grade retatrutide is unregulated material whose contents no regulator has evaluated, so a self-reported outcome is an outcome of an unknown compound at an unknown dose. Add the absence of dietary control, the absence of a comparison group, and severe selection bias — people who lose nothing, or who quit from side effects, rarely post an update — and the result is a number that cannot be attributed to anything.

What those reports are useful for: understanding what people are attempting, which side effects drive discontinuation, and how titration is being handled outside a trial. That is real information about behaviour. It is not evidence about the compound, and presenting it as though it were is how a health page becomes actively harmful.

Frequently asked questions

How much weight do people lose on retatrutide?

In the peer-reviewed Phase 2 trial, the 12 mg dose produced a mean 24.2% reduction at 48 weeks. In Phase 3 TRIUMPH-1, the company reported 28.3% at 80 weeks assuming full adherence, or 25.0% counting every randomized participant. Individual results within those trials varied widely around the mean.

What is the 30.3% retatrutide result?

It comes from a 104-week extension of TRIUMPH-1 that included only 532 participants who had a BMI of 35 or above, completed the first 80 weeks and tolerated their dose — then escalated them to their maximum tolerated dose. It describes a selected, tolerant subgroup on the highest dose each could take, not the whole trial.

How fast does retatrutide work?

In Phase 2 the dose-response was already clear at 24 weeks — 17.5% at the 12 mg dose — and continued descending to 24.2% by week 48. Notably the curve had not flattened when the trial ended, which is unusual in this drug class.

Are online retatrutide before-and-after results reliable?

No, and we do not republish them. Self-reported results carry no verified starting weight, no verified dose, no confirmation of what was actually in the vial, no dietary control, and heavy selection bias — people who see nothing rarely post. They are worth reading as a description of what people are attempting, not as evidence of what the compound does.

Do the results last after stopping?

No published discontinuation data exists for retatrutide. For semaglutide, the STEP-1 trial extension documented substantial weight regain after withdrawal, and there is no reason to expect a different pattern here.

Sources

  • Jastreboff AM et al. Triple–Hormone-Receptor Agonist Retatrutide for Obesity — A Phase 2 Trial. New England Journal of Medicine, 2023;389(6):514–526. PubMed
  • Rosenstock J et al. Retatrutide for people with type 2 diabetes: a phase 2 trial. The Lancet, 2023;402(10401):529–544. PubMed
  • Sanyal AJ et al. Triple hormone receptor agonist retatrutide for metabolic dysfunction-associated steatotic liver disease: a randomized phase 2a trial. Nature Medicine, 2024;30(7):2037–2048. PubMed
  • Coskun T et al. Effects of retatrutide on body composition in people with type 2 diabetes. The Lancet Diabetes & Endocrinology, 2025;13(8):674–684. PubMed
  • Eli Lilly. TRIUMPH-1 topline results, 21 May 2026. Company announcement. Not peer-reviewed.
  • Eli Lilly. TRIUMPH-2 and TRIUMPH-3 topline results, 23 July 2026. Company announcement. Not peer-reviewed.
  • Giblin K et al. Rationale and design of the TRIUMPH registrational clinical trials. Diabetes, Obesity and Metabolism, 2026;28(1):83–93. PubMed
  • Wilding JPH et al. Weight regain after withdrawal of semaglutide: the STEP 1 trial extension. Diabetes, Obesity and Metabolism, 2022;24(8):1553–1564. PubMed

Related guides

Research use only. Retatrutide is investigational and not approved for human use. This page summarizes published trial data and is not medical advice.