Retatrutide is a single peptide that activates three receptors at once: the glucose-dependent insulinotropic polypeptide receptor, the GLP-1 receptor, and the glucagon receptor.[1] That third target is what separates it from the two molecules this market already sells, and the reason its phase 2 numbers landed above the ones in the tirzepatide and semaglutide comparison. Those numbers are real and they are published. They were also produced by a drug that no pharmacy in the United States may lawfully dispense, which is the part most pages about this molecule leave out.
The phase 2 obesity trial, dose by dose
The obesity trial enrolled 338 adults with a body-mass index of 30 or higher, or 27 to under 30 with at least one weight-related condition, and ran for 48 weeks against placebo. At 24 weeks, the primary endpoint, the least-squares mean change in body weight was −7.2% at 1 mg, −12.9% in the combined 4-mg group, −17.3% in the combined 8-mg group and −17.5% at 12 mg, against −1.6% with placebo.[1]
At 48 weeks the same arms reached −8.7%, −17.1%, −22.8% and −24.2%, against −2.1% with placebo. A reduction of 15% or more had occurred in 60% of the 4-mg group, 75% of the 8-mg group and 83% of the 12-mg group, against 2% of placebo.[1] The common adverse events were gastrointestinal and dose-related, and starting at 2 mg rather than 4 mg partially blunted them. Heart-rate increases were dose-dependent, peaked at 24 weeks and then declined.[1]
In diabetes, the active comparator lost less than placebo
The parallel phase 2 trial randomized 281 adults with type 2 diabetes to placebo, dulaglutide 1.5 mg, or one of six retatrutide regimens. At 24 weeks the least-squares mean change in glycated hemoglobin was −2.02% in the 12-mg group against −0.01% with placebo and −1.41% with dulaglutide.[2]
The weight arm of that trial carries a figure worth pausing on. At 36 weeks, body weight fell 16.94% in the 12-mg group and 3.00% in the placebo group — but only 2.02% on dulaglutide, an approved GLP-1 receptor agonist taken weekly.[2] The licensed comparator did slightly worse than placebo on weight in a diabetes population. That is a reminder that a drug approved for glycemic control is not thereby a weight drug, which is the distinction set out in the article on the two indications.
More weight, less fat: the top dose split the two endpoints
A prespecified substudy of that diabetes trial measured body composition by dual-energy X-ray absorptiometry in 189 participants, of whom 103 completed treatment with usable scans at both baseline and week 36. Total fat mass fell 4.9% at 0.5 mg, 15.2% in the pooled 4-mg groups, 26.1% in the pooled 8-mg groups and 23.2% at 12 mg, against 4.5% with placebo and 2.6% with dulaglutide.[3]
Against placebo, the least-squares mean difference was −21.6 (95% CI, −27.1 to −16.1; p < 0.0001) for the pooled 8-mg groups and −18.7 (95% CI, −25.1 to −12.3; p < 0.0001) at 12 mg. The 0.5-mg arm did not separate from placebo at all: −0.4 (95% CI, −4.0 to 3.2; p = 0.83).[3] So within one trial, the highest dose produced marginally more total weight loss — 16.94% at 12 mg against 16.81% and 16.34% in the two 8-mg arms at 36 weeks[2] — and measurably less fat loss than the dose below it. The authors' own conclusion was that the ratio of lean mass lost to weight lost resembled other obesity treatments, and the pooled figures for that ratio across the class are in the muscle article.
Phase 3 has reported once, and not in obesity
The first phase 3 result to publish was TRANSCEND-T2D-1, a 40-week monotherapy trial in adults with type 2 diabetes inadequately controlled by diet and exercise. It randomized 537 participants across sites in the United States, Mexico and India. Glycated hemoglobin fell 1.94% at 12 mg against 0.81% with placebo, an estimated treatment difference of −1.12 percentage points (95% CI, −1.39 to −0.85). Body weight fell 15.3% at 12 mg against 2.6% with placebo. Discontinuations for adverse events ran 2% to 5% on retatrutide and 0% on placebo, and two deaths occurred, both in the 4-mg group and both judged unrelated to the study drug.[4]
The obesity program is still running. TRIUMPH comprises four phase 3 randomized trials in more than 5,800 participants, using a basket design that nests obstructive sleep apnea and knee osteoarthritis protocols inside two weight-management trials, with a third trial in people with cardiovascular disease and a fourth in osteoarthritis alone.[5] Until those report, the obesity evidence for this molecule is the 48-week phase 2 trial above. A 2025 meta-analysis that searched through May 2024 found three randomized trials covering 878 patients in total, with a pooled mean difference in body weight of −14.33% and no significant difference in adverse events against comparators (relative risk 1.11; p = 0.24).[6]
It is not approved, and it cannot be compounded
No product containing retatrutide appears in Drugs@FDA. There is no approved label, no approved dose ladder, and no approved indication. The FDA states the position directly: retatrutide and cagrilintide cannot be used in compounding under federal law, are not components of FDA-approved drugs, and have not been found safe and effective for any condition.[9]
That last clause is not a technicality about paperwork. It means the legal route by which compounded semaglutide and tirzepatide reach buyers — a prescription filled by a pharmacy preparing a drug for an individual patient — is closed for this substance. Compounded drugs are in any case not FDA-approved and are not reviewed by the agency for safety, efficacy or quality before they are dispensed, a distinction set out in the compounded-versus-brand article. Retatrutide sits a further step outside that: it may not be compounded at all. The agency has said it warned telehealth companies for marketing it, active-ingredient distributors for selling it to compounders, and outsourcing facilities for repackaging it.[9]
What has been found inside vials bought this way
Substances sold under the label “for research purposes only” or “not for human consumption” reach consumers without a prescriber, without a pharmacist, and frequently without the supplies or the instructions needed to reconstitute a lyophilized powder and inject it.[7] Nothing in that channel is monitored, and no interaction check happens.
The contents have been measured. A 2024 study collected 1,080 search results, identified 317 links to online pharmacies of which 134 led to 59 distinct illegal sites, then made test purchases from six of them. Three prefilled pens were paid for and never delivered. Of the three vials that did arrive, visual inspection found noncompliance on 59% to 63% of the criteria checked. The measured peptide content exceeded the labeled amount by 28.56% to 38.69%, while the measured purity was 7.7% to 14.37% against the 99% claimed on the labels. Endotoxin was present in every sample, at 2.1645 to 8.9511 EU/mg.[8]
Read those two numbers together. More peptide than the label says, and roughly a tenth of the purity the label claims, means the vial contains both a larger dose than intended and a large mass of something that was never identified. Those samples were sold as semaglutide, a molecule with an approved reference product to compare against. Retatrutide has no approved reference product at all, so there is nothing to compare a vial of it to. European pharmacovigilance data on suspected counterfeit semaglutide, covering 234 individual case safety reports through the end of 2025, found 89.3% of the suspected reactions were serious, with hypoglycemia and drug ineffectiveness both reported disproportionately.[10]
What this page does and does not settle
What the trials show is that a triple agonist produced larger 48-week weight reduction in phase 2 than anything currently approved produces in phase 3, at the cost of a dose-related gastrointestinal burden and a transient rise in heart rate. What they do not show is anything about durability past 48 weeks, cardiovascular outcomes, or how the molecule behaves in the populations phase 2 excluded. The horizon problem across this whole class is covered in the long-term article.
And none of the published evidence is evidence about a vial bought online. The trials used a manufactured, assayed, cold-chain-controlled investigational product administered under a protocol. A seller quoting −24.2% is quoting that product, not the one being shipped. A service that offers to prescribe it at all has misdescribed what it is allowed to do, which belongs on the list in the telehealth red flags article.