Retatrutide

What it is

Retatrutide, development name LY3437943, is a synthetic peptide injected under the skin. It belongs to the family of incretin drugs, which work with gut hormones. Semaglutide acts on one switch, GLP-1. Tirzepatide acts on two, GLP-1 and GIP. Retatrutide adds the glucagon receptor as a third. Hence the nickname “Triple G”, and in the scene also “GLP-3”, which is not quite scientifically accurate.

The class began as diabetes research. Only later was it fully understood how strongly these drugs act on weight loss.

How it is supposed to work

GLP-1 and GIP reduce appetite and smooth out blood sugar. Glucagon is actually known as a hormone that raises blood sugar, but it can do something else as well: it boosts energy expenditure and mobilizes fat, especially in the liver. Eating less and burning more, in one molecule.

GLP-1 acts not only in the gut but also in brain regions that control hunger and reward. Many users report that constant thinking about food becomes quieter, the so-called “food noise”. That is a user report, not a measured endpoint.

An honest limitation belongs here: how much the third lever contributes on its own has not yet been tested separately by anyone. That glucagon makes the difference compared with tirzepatide is a plausible story, not a measurement.

Which number applies to whom

Three numbers from TRIUMPH-1 are circulating, and all are correct, just for different groups. The 28.3 percent applies to the highest tested dose level after 80 weeks, calculated for everyone who continued treatment as planned. Under the second prespecified rule, which counts everyone randomized regardless of whether they stayed on, it is 25.0 percent versus 3.9 percent on placebo. This second number is closer to everyday reality.

The widely cited 30.3 percent comes from an extension to 104 weeks. It enrolled 532 participants with a body mass index of 35 or more at baseline who had completed the 80 weeks and tolerated their dose. The 30.3 percent applies only to the part of them who received the highest level from the start. There was no longer a placebo group in the extension; the former placebo group switched to the drug and was at 19.2 percent after 104 weeks.

Liver, blood sugar and the other trials

Almost more impressive than the weight is the liver. In a substudy of the phase 2 trial, liver fat in people with fatty liver fell by 82.4 percent at the highest level after 24 weeks, while it rose by 0.3 percent on placebo. 86 percent reached a normal liver fat content, and no one in the placebo group did.

For type 2 diabetes, a peer-reviewed phase 3 trial is now available: TRANSCEND-T2D-1 in the Lancet 2026 with 537 participants. After 40 weeks, HbA1c fell by 1.94 percentage points at the highest level, versus 0.81 on placebo. Weight fell by 15.3 percent, versus 2.6 on placebo. According to the manufacturer, TRIUMPH-4 in knee osteoarthritis with 445 participants reached 28.7 percent versus 2.1 percent after 68 weeks, and TRIUMPH-3 in 1,949 people with cardiovascular disease 22.6 percent versus 3.2 percent.

Gray market and user reports

Retatrutide can be ordered anyway, and there are now numbers on this. A non-peer-reviewed preprint searched a US network of 29 million patient records. For 531 people, it was possible to trace where they got retatrutide; 71.2 percent of them obtained it outside a trial. They lost 7.2 percent of their weight after six to twelve months, while trial participants in the same analysis lost 15.5 percent. That is roughly what comparable tirzepatide users achieved, 7.7 percent.

A second preprint analyzed 148,640 Reddit posts from 27 forums, including 13,589 users reporting their own use. The most frequently mentioned effect was not nausea but increased appetite, followed by fatigue and more energy, with nausea only in fourth place. The authors themselves stress that neither frequency nor cause can be inferred from this, especially since no one knows what was injected.

What the studies show

Phase 2, NEJM 2023 — the peer-reviewed core

Jastreboff and colleagues randomized 338 adults with obesity, or overweight plus a comorbidity, to several dose levels or placebo, 48 weeks, double-blind. The primary endpoint was the change in weight after 24 weeks: 17.5 percent less at the highest level versus 1.6 percent on placebo. Endpoint met. After 48 weeks, it was 24.2 versus 2.1 percent. Side effects mainly affected the gastrointestinal tract and were mostly mild to moderate.

TRIUMPH-1 — phase 3, so far only as a company announcement

2,339 participants without diabetes, randomized and double-blind to three dose levels and placebo, 80 weeks. According to Eli Lilly's announcement of May 21, 2026, they lost 19.0, 25.9 and 28.3 percent, versus 2.2 percent on placebo. In the stricter analysis of everyone randomized, it was 17.6, 23.7 and 25.0 percent versus 3.9 percent. The trial is strong; what is weak is only what can be seen of it: no confidence intervals, no weight curve.

Liver substudy, Nature Medicine 2024

From the phase 2 trial, 98 participants with at least 10 percent liver fat were followed by magnetic resonance imaging. The primary endpoint was the relative change in liver fat after 24 weeks. Depending on the dose level, it ranged from 42.9 to 82.4 percent less, versus 0.3 percent more on placebo. Endpoint met.

TRANSCEND-T2D-1, Lancet 2026

The first peer-reviewed phase 3 trial: 537 people with type 2 diabetes, 40 weeks, three dose levels against placebo. HbA1c fell by 1.69 to 1.94 percentage points versus 0.81 on placebo, primary endpoint met. Weight fell by 11.5 to 15.3 percent, versus 2.6 on placebo. No severe hypoglycemia occurred.

Where the data stop

The most important limit is mundane: to this day, there is no peer-reviewed full publication on the TRIUMPH trials. Only the description of how the program is planned has been peer-reviewed. All phase 3 weight numbers come from the manufacturer's announcements and from a conference presentation. None of these trials has published values after stopping the drug.

Hard endpoints such as heart attack, stroke or mortality are not the goal of these trials. In TRIUMPH-3 in people with heart disease, they were counted anyway: 27 major events versus 23 on placebo by the narrower measure, 44 versus 52 by the broader one. That points in two directions, and both numbers are too small to support a conclusion. There is no direct comparison with tirzepatide; the claim that retatrutide works more strongly is based on comparing different trials.

Status, approval and legal situation

Retatrutide is not approved anywhere in the world. Eli Lilly intends to submit the application for approval to the US FDA in the first quarter of 2027. In Germany, it is not a marketable medicine. The FDA has warned about unapproved products sold directly to consumers, often labeled “for research purposes only”. Such goods are not tested. In sport, retatrutide, as a substance not approved anywhere, falls under class S0 of the WADA Prohibited List and is prohibited at all times.

Safety

The side effect profile resembles the GLP-1 family but is more pronounced. In TRIUMPH-1, according to the manufacturer, nausea affected 28.6, 38.4 and 42.4 percent by dose level versus 14.8 percent on placebo; 4.1, 6.9 and 11.3 percent stopped because of side effects versus 4.9 percent. In TRIUMPH-4, discontinuation at the highest level was 18.2 percent. New is dysesthesia, that is, tingling or abnormal sensations of the skin: 12.5 percent at the highest level versus 0.9 percent in TRIUMPH-1, and 20.9 versus 0.7 percent in TRIUMPH-4. The glucagon arm raises heart rate; in the preprint on real-world use, it rose by 2.5 beats per minute in gray-market users. Long-term data are lacking. A case report describes impending ketoacidosis in a person with type 1 diabetes after retatrutide bought online.

Sources

• Jastreboff et al., N Engl J Med 2023 (phase 2)
• Sanyal et al., Nat Med 2024 (liver substudy)
• Bajaj et al., Lancet 2026 (TRANSCEND-T2D-1)
• Giblin et al., Diabetes Obes Metab 2026 (design of the TRIUMPH trials)
• Eli Lilly, company announcement on TRIUMPH-1, 2026-05-21
• Eli Lilly, company announcement on TRIUMPH-2 and TRIUMPH-3, 2026-07-23
• Eli Lilly, company announcement on TRIUMPH-4, 2025-12-11
• Murugadoss et al., Preprints 2026 (gray-market use, not peer-reviewed)
• Sehgal et al., medRxiv 2026 (Reddit analysis, not peer-reviewed)
• Branine, Cureus 2026 (case report, retatrutide bought online)