Retatrutide is the compound with the largest weight-loss numbers in the field and no approval in any country. Both halves of that sentence matter, and vendor copy tends to carry only the first. On those vendor pages it usually goes by a nickname — reta, or reta peptide — while the trial record calls it LY3437943. Three names, one molecule, and only one of them appears on a regulator's paperwork.
What it is
An investigational peptide, Eli Lilly's LY3437943, that activates three receptors: GIP, GLP-1 and glucagon. The first of those is the gut hormone in the GIP entry. Semaglutide occupies one of those. Tirzepatide occupies two. The receptor for glucagon, the pancreas's blood-sugar-raising hormone, is the addition, and it is not a small one — glucagon raises energy expenditure as well as blood glucose, so the design works on both sides of energy balance rather than on appetite alone.
What the phase 2 trial found
The trial that made its reputation is NCT04881760, published in the New England Journal of Medicine in 2023: 338 adults with obesity, or with overweight plus a weight-related condition, randomised to subcutaneous retatrutide once weekly or placebo for 48 weeks.
| Weekly dose in the trial | Weight change at 48 weeks |
|---|---|
| 1 mg | −8.7% |
| 4 mg | −17.1% |
| 8 mg | −22.8% |
| 12 mg | −24.2% |
| Placebo | −2.1% |
In the 12 mg group, every participant lost at least 5% of body weight, 93% lost at least 10%, and 83% lost at least 15%. On placebo the same figures were 27%, 9% and 2%.
These are doses a published trial protocol used, recorded here as what that protocol did. There is no approved label for retatrutide, so there is no recommended dose to report at all.
The finding inside the finding
The trial did not only test dose; it tested how the dose was reached. The 4 mg and 8 mg arms were each run twice, once starting from 2 mg and once from 4 mg. Side effects were gastrointestinal and dose-related, and the paper reports they were "partially mitigated with a lower starting dose (2 mg vs. 4 mg)."
That is a result about escalation rather than about the drug, and it is the part of a protocol most often stripped out when a figure is repeated. A number quoted without the ramp that reached it describes something the trial did not measure.
The other signal the paper reports plainly: heart rate rose in a dose-dependent way, peaking at 24 weeks and declining afterwards.
Why weekly at all? Lilly's phase 1b trial in type 2 diabetes, published in The Lancet in 2022, reported that retatrutide's pharmacokinetics were dose-proportional and its half-life approximately 6 days (PMID 36354040) — between tirzepatide's labelled 5 days and semaglutide's one week. Like every other number about this compound, it comes from the developer.
One sponsor owns the entire evidence base
Re-counted on 2026-09-22, ClinicalTrials.gov holds 33 registrations naming retatrutide. Eli Lilly is the lead sponsor of 32 — 14 phase 1, 4 phase 2 and all 14 phase 3 registrations among them. The 33rd, NCT07467447, was filed by Hudson Biotech, and the correction note under Sources explains why it does not count as an independent trial. Only two of the 33 have posted results, both Lilly phase 2 studies (NCT04881760 and NCT04867785).
That is normal for a compound in development and it is still worth stating, because it means every number in circulation about retatrutide comes from its developer. There is no independent replication of any of it, and there cannot be until the programme reports and others can run their own trials.
The phase 3 programme does contain the comparisons that matter: retatrutide against tirzepatide (800 participants) and against semaglutide (1,250 participants). Neither has reported. Until they do, comparing retatrutide's 24.2% to a figure from a different trial in a different population is arithmetic rather than evidence.
Why the vendor market is a special case here
For an approved drug, a research-chemical vial can at least be held against a label: a known molecule, a known strength, a known purity standard. For retatrutide none of that exists, because no approved product exists. A related confusion is worth heading off here, because it sends buyers to sellers with nothing to sell: this compound's three-receptor mechanism is what people are usually reaching for when they search for a hormone called GLP-3, and there is no such hormone in a human being. There is no reference standard, no approved concentration, and no regulator that has reviewed anything about the compound outside a trial file.
One further detail, checkable in a browser: a ClinicalTrials.gov registration naming retatrutide has been filed carrying the internal Lilly protocol identifier belonging to the phase 2 study cited on this page. Peptifact's MOTS-c dosage page publishes the method for verifying that. It is a useful reminder that a registry entry is a filing, not evidence that a study took place.
Where it stands
A compound with the strongest trial numbers in its class, an unfinished phase 3 programme, a single sponsor behind every credible study, a known dose-dependent heart-rate signal, and no approval anywhere. All five of those are true at the same time, and an accurate description of retatrutide contains all of them.
Sources
- Jastreboff AM, Kaplan LM, Frías JP, et al. Triple-Hormone-Receptor Agonist Retatrutide for Obesity — A Phase 2 Trial. N Engl J Med 2023;389(6):514–526. PMID 37366315. ClinicalTrials.gov NCT04881760. Funded by Eli Lilly.
- ClinicalTrials.gov API v2, searched 2026-09-08 for interventional studies naming retatrutide: 33 registrations. Head-to-head trials NCT06662383 (versus tirzepatide) and NCT06260722 (versus semaglutide) were active and unreported on that date.
- Correction, 2026-09-09. This entry previously stated that Eli Lilly was the lead sponsor of all 33 registrations. Re-counted through the same API on 2026-09-09, Lilly sponsors 32; the remaining one, NCT07467447, was filed by Hudson Biotech — a sponsor whose eight filings do not withstand checking, two of them carrying Lilly's own internal protocol identifiers. The corrected count and the reason it matters are set out on mazdutide, where the contrast with a compound that does have independent sponsors is the point.
- PubMed, searched 2026-09-08: 175 records naming retatrutide.
- Urva S, Coskun T, Loh MT, et al. LY3437943, a novel triple GIP, GLP-1, and glucagon receptor agonist in people with type 2 diabetes: a phase 1b, multicentre, double-blind, placebo-controlled, randomised, multiple-ascending dose trial. Lancet 2022;400(10366):1869–1881. PMID 36354040. Half-life approximately 6 days.
- Second correction, 2026-09-22. The 2026-09-09 correction above was made in this Sources list but not in the body, the facts or the standing panel, which went on saying "all 33" for two weeks. All three now carry the re-counted figure (32 of 33, API v2, 2026-09-22).