Retatrutide: what phase 2 actually showed, and what it didn't

When the phase 2 results for retatrutide published in 2023, the top dose had produced a 24.2% mean reduction in body weight at 48 weeks — the largest figure any obesity trial had reported at the time. Communities organized around the molecule almost immediately, and much of what circulates in them now runs well ahead of what one trial can support. This article separates the layers: what phase 2 established, what it merely observed, what only phase 3 can answer, and what the absence of any approval means for what is actually in the vials people are buying.

What retatrutide is

Retatrutide (Eli Lilly's LY3437943) is an investigational once-weekly peptide that activates three receptors at once: GLP-1, GIP, and glucagon. It is the next step in a deliberate design progression — semaglutide is a single GLP-1 agonist, tirzepatide adds GIP, and retatrutide adds glucagon on top of both. Its roughly 6-day half-life supports the weekly injection schedule used in its trials.

The regulatory status comes first because everything downstream depends on it: retatrutide is not approved anywhere, for anything. It is an investigational drug in phase 3 trials. Every product sold as "retatrutide" today is an unapproved research chemical, full stop.

Why the glucagon arm is mechanistically interesting

GLP-1 agonism suppresses appetite and slows gastric emptying; GIP agonism adds insulinotropic and adipose-tissue signaling. Both work the intake side of the energy budget — you lose weight because you eat less.

Glucagon-receptor agonism is the novel arm: it works the output side, raising energy expenditure and driving hepatic fat oxidation. That makes retatrutide one of the few clinical-stage obesity drugs with an explicit mechanism for burning more rather than only eating less, and it is the leading explanation for the striking liver-fat results covered below.

The same arm carries the design tension. Glucagon raises hepatic glucose output, which alone would push blood sugar the wrong way — the two incretin arms are what hold glycemia in check. A triple agonist is therefore a balancing act by construction, which is exactly why its cardiovascular and metabolic profile needs full characterization before anyone can claim to understand it.

What the phase 2 trial showed

The obesity trial is Jastreboff et al., published in the New England Journal of Medicine in 2023.

Three findings deserve precision. First, the response was dose-dependent across the 1, 4, 8, and 12 mg arms — this was a real dose-response curve, not a single lucky arm. Second, the paper reported that at 48 weeks every participant in the 8 mg and 12 mg groups had lost at least 5% of body weight. Third, in the higher-dose groups the weight curves had not plateaued when the trial ended — weight was still falling at week 48. That last point is frequently quoted as evidence the "real" number is higher. It is suggestive, and it is also extrapolation beyond the data, which is precisely the habit this article is trying to discourage.

A separate phase 2 trial in type 2 diabetes (Rosenstock et al., The Lancet, 2023) showed substantial reductions in both HbA1c and body weight, consistent with the obesity findings.

Now the caveats, which are structural rather than nitpicks. This was 338 people for one year. A trial that size cannot surface rare adverse events — you need thousands of patient-years for that. The population was screened and excluded diabetes in the obesity trial; it does not resemble the unscreened gray-market user base. Effect sizes routinely shrink between phase 2 and phase 3 as populations broaden. And 24.2% is a mean: individual responses spread widely around it, in both directions.

The phase 2 observations worth knowing

Heart rate. Retatrutide produced a dose-dependent increase in resting heart rate that peaked around week 24 and partially attenuated by week 48. Whether that pattern matters clinically over years is unknown — it is one of the central questions the phase 3 program, including its cardiovascular outcomes work, exists to answer. It is also the observation community discussion most consistently skips.

GI profile. The most common adverse events were gastrointestinal — nausea, vomiting, diarrhea, constipation — mostly mild to moderate and concentrated during dose escalation. Notably, the trial itself treated escalation speed as a variable worth studying: the top-dose group was split between lower and higher starting doses, and gentler starts produced fewer early GI events. The titration schedule was not an afterthought; it was part of the experiment.

The liver-fat sub-study

Participants with elevated liver fat at baseline were followed in an imaging sub-study, published by Sanyal et al. in Nature Medicine in 2024. At the two highest doses, mean relative reductions in liver fat exceeded 80% by week 48, and most of those participants reached normal liver-fat content (below 5%). That is a remarkable result, and it is mechanistically coherent with the glucagon arm's hepatic fat oxidation.

The honest boundary: this measured steatosis — fat content — by imaging. Resolving fat is not the same as resolving fibrosis, the scarring that actually drives liver outcomes, and no biopsy-based fibrosis results have been reported for retatrutide. MASH ambitions rest on trials that have not read out.

What phase 3 is testing — and what the approval reality is

The phase 3 program, TRIUMPH, spans obesity, type 2 diabetes, and obesity-related conditions — including a trial in knee osteoarthritis — alongside cardiovascular outcomes work. Phase 3 is where the open questions get answered at scale: whether the effect size holds in broader populations, whether rare adverse events emerge, and what the heart-rate signal means over time.

Until then, the approval ledger is simple: no regulator anywhere has approved retatrutide. There is no label, no agreed dosing, no pharmacovigilance system watching for problems. The entirety of controlled human knowledge in obesity is what you read above — one year, 338 people, plus the diabetes trial. Everything else is anecdote.

What "not approved" means for sourcing

This is where trial results and community reality diverge hardest. An unapproved investigational drug has no lawful pharmacy channel and no compounding pathway — there is no regulated route by which retatrutide reaches an individual outside a clinical trial. What exists instead is the research-chemical gray market.

Stated flatly, without moralizing: gray-market vials carry no GMP guarantee. Identity, quantity, sterility, and endotoxin status are unverified by default, and a vendor's certificate of analysis is marketing until independently confirmed. Independent testing labs can verify a given vial — and that verification covers that vial only.

This matters for how you read community knowledge. Forum dosing lore assumes the number on the label is real. A protocol built on vials of unverified content isn't a replication of the trial at home; it is a different, uncontrolled experiment with borrowed precision.

Honest unknowns

• Long-term safety. Controlled exposure tops out at 48 weeks. The durability of the heart-rate change, and anything that only shows up over years, is unknown.
• Rare adverse events. 338 participants cannot surface a 1-in-1,000 problem. Only phase 3 scale and post-market surveillance can.
• Durability and discontinuation. No retatrutide withdrawal data exists. The class precedent is sobering: in the STEP 1 trial extension (Wilding et al., Diabetes, Obesity and Metabolism, 2022), participants regained roughly two-thirds of lost weight in the year after stopping semaglutide. Whether retatrutide behaves the same is unstudied.
• Cardiovascular outcomes. Semaglutide earned its cardiovascular evidence through the SELECT trial (Lincoff et al., NEJM, 2023). Retatrutide has no outcome data yet, and the heart-rate observation makes this the load-bearing unknown rather than a formality.
• Composition of the loss. How the weight reduction splits between fat and lean mass is a question the phase 2 publication was not designed to settle.

None of this is a prediction that retatrutide fails. The phase 2 data is genuinely exceptional, and the mechanism is coherent — it may well earn approval on the strength of TRIUMPH. But "will probably be a great drug" and "is a characterized drug today" are different claims, and the communities organized around this molecule routinely swap one for the other. Today's evidence is one year, 338 people, striking numbers, real open questions, and zero approvals. Calibration means holding all five at once.

Sources

• Jastreboff AM, et al. Triple-hormone-receptor agonist retatrutide for obesity — a phase 2 trial. New England Journal of Medicine, 2023.
• Rosenstock J, et al. — phase 2 trial of retatrutide in type 2 diabetes. The Lancet, 2023.
• Sanyal AJ, et al. — phase 2 sub-study of retatrutide in metabolic dysfunction-associated steatotic liver disease. Nature Medicine, 2024.
• Wilding JPH, et al. Weight regain and cardiometabolic effects after withdrawal of semaglutide: the STEP 1 trial extension. Diabetes, Obesity and Metabolism, 2022.
• Lincoff AM, et al. Semaglutide and cardiovascular outcomes in obesity without diabetes (SELECT). New England Journal of Medicine, 2023.
• ClinicalTrials.gov — TRIUMPH phase 3 program registrations for retatrutide (Eli Lilly), 2023 onward.