Retatrutide: strong numbers, measured on stand-ins

The short version

Retatrutide, also called LY3437943, is an investigational injectable peptide that switches on three gut-hormone receptors at once. Two of them, GLP-1 and GIP, damp appetite and improve the release of insulin when blood sugar is high. The third, the glucagon receptor, adds energy expenditure. The combination is why it belongs to the same broad family as semaglutide (Ozempic, Wegovy) and tirzepatide (Mounjaro, Zepbound), and why it is expected to go further than either.

The human evidence here is real, and it is the strongest on this desk after PT-141. In a 48-week Phase 2 trial in 338 adults with obesity, once-weekly retatrutide at the highest amount tested produced a mean body-weight change of -24.2 percent against -2.1 percent on placebo ^[11]. In 281 adults with type 2 diabetes over 36 weeks, the same amount lowered HbA1c by 2.02 percentage points at week 24 and body weight by 16.94 percent at week 36 ^[12]. A substudy in 98 people with fatty liver disease saw liver fat fall by 82.4 percent at 24 weeks, with 86 percent reaching a normal level ^[10].

So where is the gap? In what those numbers are. Body weight, HbA1c and liver fat are surrogate endpoints: things that are easy to measure and are believed to track something that matters more. Not one published retatrutide trial has yet reported whether the people taking it had fewer heart attacks, fewer strokes, less kidney decline, or longer lives. Those trials are running. None has read out. And nothing published says what happens to the weight after the injections stop.

What it is

Retatrutide is a 39-amino-acid synthetic peptide built on a backbone derived from GIP, one of the gut hormones released after a meal. A twenty-carbon fatty diacid arm is attached to it, which makes the molecule stick to albumin in the blood and slows its clearance enough for once-weekly injection. Its molecular formula as the free acid is C221H342N46O68.

It is an investigational drug in Phase 3 trials, developed by Eli Lilly. It has not been approved by the FDA or any other regulator. It is not available on prescription and it is not a consumer product. Everything on this page comes from published clinical trials rather than from approved labelling, which is a meaningful distinction: an approved label has been argued over by a regulator, and a trial report has not.

That also sets up the compound's most practical problem. Because it is not approved, there is no legitimate route to it outside a trial, and a grey market has grown up selling material labelled for research. Vials from that channel cannot be confirmed to contain authentic retatrutide at the stated concentration, and injectable material of unverified sterility carries risks that have nothing to do with the molecule itself ^[8][11].

How it works

The simplest way to picture retatrutide is as one key cut to fit three locks. Most drugs in this family fit one or two.

The GLP-1 and GIP arms do what that class is known for: they slow the stomach's emptying, reduce appetite, and increase insulin release in a way that depends on blood sugar already being elevated, so the effect fades when it is not needed. The glucagon arm is the unusual addition. Glucagon normally tells the liver to release stored glucose, which sounds like the opposite of what a diabetes drug should do. But alongside simultaneous GLP-1 and GIP activity, mild glucagon-receptor stimulation instead raises energy expenditure and mobilises fat, without much net rise in blood sugar. The result works on both halves of the energy equation at once: less taken in, more spent.

Structural work published in 2024 shows how uneven that three-way engagement is. Cryo-electron microscopy resolved retatrutide bound to all three receptors, at resolutions of 2.68, 3.26 and 2.84 angstroms, and measured its potency against the body's own hormones: about 8.9-fold more potent than native GIP at the GIP receptor, but 0.3-fold and 0.4-fold at the glucagon and GLP-1 receptors respectively. A loop on the receptor called ECL1 forms a rigid helix at GLP-1R and GCGR but stays flexible at GIPR ^[9].

That asymmetry is itself a gap. It means the three arms are not contributing equally, and no published trial has taken the molecule apart to establish which arm delivers which share of the weight loss, the heart-rate change, or the nausea. The pharmacology is described in detail; the attribution is not.

What the research shows, and what each study leaves open

Phase 2 obesity trial, 2023. 338 adults with obesity, 48 weeks, once-weekly injection. Mean body-weight change of -24.2 percent at the 12 mg amount against -2.1 percent on placebo. Gastrointestinal adverse events were dose-related and mostly mild to moderate, and a dose-dependent increase in heart rate was observed, peaking around week 24 ^[11]. Left open: the trial stopped at week 48. The heart-rate curve peaked halfway through it and what that measure does over two or five years is not in the published literature. Nothing here is an outcome; it is weight and tolerability.

Phase 2 type 2 diabetes trial, 2023. 281 adults over 36 weeks. HbA1c fell by 2.02 percentage points at week 24 against 0.01 on placebo, and body weight by 16.94 percent at week 36 against 3.00 percent. Mild-to-moderate gastrointestinal events occurred in about 35 percent of participants, with no severe low-blood-sugar episodes and no deaths ^[12]. Left open: HbA1c is a stand-in for the complications of diabetes, not a count of them. Thirty-six weeks is also too short to see the complications it stands in for.

Fatty liver substudy, 2024. 98 participants with obesity or overweight and metabolic dysfunction-associated steatotic liver disease, at least 10 percent liver fat measured by MRI. Relative liver-fat change at 24 weeks ran -42.9, -57.0, -81.4 and -82.4 percent across the four ascending amounts against +0.3 percent on placebo; 86 percent of participants at the highest amount reached a normal liver-fat level under 5 percent, and the reduction was sustained to 48 weeks at -86.0 percent ^[10]. Left open: liver fat on a scan is not liver scarring on a biopsy. No published retatrutide trial reports fibrosis stage, and fibrosis is the part of liver disease that predicts what happens to the person.

Structural pharmacology, 2024. The cryo-EM work described above, resolving the triple engagement and the uneven potencies ^[9]. Left open: the arm-by-arm attribution question, and everything about long-term receptor biology.

2025 narrative review. A synthesis of the triple-agonist pharmacology and the Phase 1 and 2 data, characterising the weight change of roughly 24 percent at 12 mg over 48 weeks as a step-change against earlier incretin therapies, and describing the gastrointestinal and heart-rate safety profile and the ongoing Phase 3 TRIUMPH programme ^[8]. Left open: a review can only summarise what exists. The word ongoing is doing the heavy lifting in that sentence, and it is the single most important word on this page.

Reported effects, cautions and safety

What follows in this first part is anecdotal, not clinical evidence. It comes from research-use communities discussing grey-market material with no clinical oversight and no verified product identity, and no amounts are attached to any of it.

The effect described most consistently is the near-total silencing of intrusive thoughts about food, which community members call food noise going quiet, and which is reported as disinterest in eating rather than active fullness. Rapid and pronounced weight reduction is frequently reported. A distinct sensation of running warm, sweating more easily, or a low-grade heat unlike exertion is commonly reported and is widely attributed within those communities to the glucagon arm. A minority describe a mood lift and an easier relationship with eating.

On the unwanted side, nausea in the hours after injection is among the most common reports, described as peaking roughly four to eight hours afterwards and easing over the first weeks. Sulfur-smelling burps and constipation are commonly reported and attributed to slowed gut motility. Fatigue in the early weeks is commonly reported and often linked by the people reporting it to eating far less. An elevated resting heart rate, noticed on wearables, is commonly reported. Sleep disturbance is occasionally reported, as is minor injection-site itching. People who track body composition closely raise a recurring concern that rapid loss feels soft, meaning they suspect muscle is going with the fat.

The documented cautions are these.

• Unapproved status makes supply the primary risk. Retatrutide is investigational and has not been approved by any regulator. Material obtained outside a trial has unverified identity, concentration and sterility, and injecting unverified material carries infection risk independent of the compound ^[8][11].
• Gastrointestinal effects were dose-related and were the main reason participants stopped. They arise from GLP-1-mediated slowing of the stomach and altered gut motility, and in the Phase 2 obesity trial they were dose-related and drove discontinuation at the highest amounts ^[11]. Trials escalate the amount gradually to manage this; unsupervised use has no such structure.
• Heart rate rises in a dose-dependent way. The increase was observed in Phase 2 and peaked around week 24 ^[11]. The glucagon arm drives cardiac rate through cAMP signalling. A dedicated cardiovascular outcomes trial is running and has not reported, so what this means for arrhythmia burden or cardiac remodelling is genuinely unknown ^[8].
• Combining it with insulin or a sulfonylurea raises the risk of low blood sugar. Its GLP-1 and GIP activity augments insulin release; on top of injected insulin or a drug that forces insulin release, blood sugar can fall below safe levels. Trial participants on background insulin had that insulin reduced during the study under supervision ^[12].
• Body composition is an open question rather than a settled one. Rapid loss of this magnitude reduces lean mass alongside fat across this class, and the dedicated body-composition analysis is not among the sources listed on this site, so it is flagged here rather than summarised.
• Long-term safety, durability and outcomes are all unreported. The Phase 3 programme and the dedicated cardiovascular and kidney outcome trials are ongoing ^[8]. Substantial weight regain after stopping is documented for related agents in the class, which raises an unanswered question about what open-ended unsupervised use amounts to.

Where it fits in the gap register

Retatrutide is the opposite failure mode to BPC-157. Its human evidence is not thin; it is substantial, controlled, placebo-compared and published in major journals. What is missing is at the far end of the study design rather than the near end.

Every headline figure on this page is a surrogate. Weight, HbA1c and liver fat are measured because they are quick, cheap and reliable, and because decades of epidemiology suggest they track things that matter. But suggestion is not demonstration, and the history of metabolic medicine contains drugs that moved a surrogate in the right direction and an outcome in the wrong one. That is the specific reason regulators require outcome trials, and it is the specific reason no honest page about retatrutide can end at -24.2 percent ^[11].

The second missing piece is time. The longest published exposure here is 48 weeks ^[10][11]. The heart-rate signal peaked at week 24, inside that window ^[11], which means the shape of the curve after the trial ended is unknown rather than reassuring. And nothing published follows participants past the end of treatment, so the question people most want answered, whether the change lasts, has no published retatrutide answer at all.

Read alongside NAD+, the contrast is instructive: NAD+ trials measure a stand-in because the real endpoint is decades away, while retatrutide trials measure a stand-in because the real endpoint is expensive and slow. Same word, different problem. The comparison page sets the four out side by side.