Retatrutide ("GLP-3"): triple-agonist mechanism, Phase 3 trial results, and where FDA approval stands

Retatrutide is an investigational synthetic peptide developed by Eli Lilly, internally coded LY3437943 ^[1]. No brand name exists because no approved product exists. In practice it's a once-weekly subcutaneous injection, and its half-life of roughly six days comes from acylation with a fatty diacid — the same pharmacokinetic trick that bought semaglutide and tirzepatide their weekly dosing.

It's the most clinically advanced triple receptor agonist in the obesity pipeline. Lilly runs it under two Phase 3 program families: TRIUMPH for obesity and its complications (including knee osteoarthritis and obstructive sleep apnea) and TRANSCEND for type 2 diabetes ^[4][6]. Together these enroll well over 7,000 participants, and the first Phase 3 readouts arrived in late 2025 and early 2026.

Semaglutide hits one receptor (GLP-1R). Tirzepatide hits two (GLP-1R and GIPR). Retatrutide hits three: GLP-1R, GIPR, and GCGR, the glucagon receptor ^[1]. That third one is where things get interesting, and where the reading has to be careful.

GLP-1R activation does the heavy lifting for the whole class — glucose-dependent insulin secretion, slowed gastric emptying, and hypothalamic satiety signaling that makes people eat less. It's also what drives most of the nausea. GIPR activation augments insulin secretion glucose-dependently and modulates how adipose tissue handles energy; adding GIP to GLP-1 (the tirzepatide design) produces more weight loss than GLP-1 alone, and there's preclinical evidence it may blunt some GLP-1-induced nausea, though that mechanism is still debated.

GCGR activation is the new piece. Glucagon is, in the lay imagination, the hormone that raises blood sugar — seemingly the wrong thing to do in a metabolic drug. But chronic low-grade glucagon-receptor agonism also pushes resting energy expenditure up and strips fat out of the liver, and in this molecular design the hyperglycemic effect is more than offset by co-activating GLP-1R and GIPR. Preclinical work by Coskun and colleagues at Lilly (Cell Metabolism, 2022) described a binding profile with balanced GCGR and GLP-1R activity and more prominent GIPR activity, and documented weight loss persisting 43 days post-dose in obese animals, with energy expenditure rising while intake fell ^[3]. That combination — output up, intake down — is what predicted the size of effect now showing up in humans.

Three molecules, three receptor profiles, three different weight-loss ceilings in their respective programs. The figures below are not from head-to-head trials; they come from separately conducted studies in overlapping but not identical populations, and should be read with that caveat.

The mechanistic distinction predicts a different ceiling. GLP-1 monotherapy drives weight loss mainly by suppressing intake; adding GIP gets somewhat more suppression and possibly better tolerability; adding glucagon agonism adds a push on the expenditure side. That's why retatrutide's Phase 2 weight-loss curves had not plateaued at 48 weeks ^[1], and why the Phase 3 extension data show patients continuing to lose weight into a second year ^[4].

The trial that put retatrutide on the map was published in the New England Journal of Medicine on June 26, 2023, by Jastreboff and colleagues (NCT04881760) ^[1]. It enrolled 338 adults, all with BMI ≥30, or 27–30 with at least one weight-related comorbidity, in a 48-week, randomized, double-blind, placebo-controlled Phase 2 study; people with type 2 diabetes were excluded. A parallel Phase 2 trial in type 2 diabetes landed the same week in the Lancet, from Rosenstock and colleagues ^[2]. Doses ran 1, 4, 8, and 12 mg once weekly, with stepwise titration into the higher-dose arms, against placebo.

At 24 weeks the 12 mg arm reached about 17.5% mean body-weight reduction; by week 48 the same arm hit 24.2%, against 2.1% on placebo ^[1]. Essentially every participant on 8 mg and 12 mg cleared the 5% threshold, and a substantial share of the 12 mg group crossed 30% body-weight reduction — territory that used to belong to bariatric surgery. Secondary endpoints moved the way you'd hope: HbA1c, systolic blood pressure, triglycerides, and liver fat all down. The line that mattered most to the field was that weight loss was dose-dependent and had not plateaued at 48 weeks — meaning the Phase 2 numbers were a floor, not a ceiling.

Phase 3 has confirmed that read and then gone past it. TRIUMPH-1 (NCT05929066) randomized 2,339 adults with obesity or overweight and at least one weight-related condition, and without diabetes, in equal parts to retatrutide 4 mg, 9 mg, 12 mg or placebo over 80 weeks ^[5]. Lilly reported topline results on 21 May 2026 ^[4], and the trial was published in the New England Journal of Medicine on 29 September 2026 ^[22].

Two further figures are worth holding onto. At 12 mg, 65.3% of participants finished the trial below a BMI of 30, no longer meeting the threshold for obesity, including 37.5% of those who started at a BMI of 40 or above. And in a pre-specified extension that enrolled 532 participants with a baseline BMI of 35 or above, those continuing to 104 weeks reached 30.3% on the top maintained dose — while the placebo group, crossed over to retatrutide for the extension, reached 19.2% ^[4].

Those are efficacy-estimand numbers: what the drug did in people who stayed on it. The treatment-regimen estimand, which counts everyone as randomized whether they adhered or not, is lower — −17.6%, −23.7% and −25.0% against −3.9% on placebo. Both sets are in the same release, and the gap between them is the honest measure of how much of the headline depends on staying on the drug. The published paper reports the treatment-regimen figures as its primary analysis — the same basis as tirzepatide's 20.9% in SURMOUNT-1 — and adds that the trial met its other primary endpoints: less knee-osteoarthritis pain and fewer sleep-apnoea events per hour than on placebo in the participants with those conditions ^[22].

Two more trials read out on 23 July 2026 ^[15]. TRIUMPH-2, in 1,152 adults with obesity or overweight and type 2 diabetes over 80 weeks, reported −12.7%, −19.1% and −20.8% across the three doses against −4.0% on placebo, with A1c down 1.4 to 1.6 percentage points against 0.2. TRIUMPH-3, in 1,949 adults with severe obesity and established cardiovascular disease over 80 weeks, reported −21.6% at 9 mg and −22.6% at 12 mg against −3.2%. TRIUMPH-2 was published in The Lancet on 29 September 2026 with the treatment-regimen figures: −11.9%, −16.8% and −18.8% against −5.1% on placebo ^[23]. TRIUMPH-3 is so far reported only in the press release.

The diabetes program read out earlier. TRANSCEND-T2D-1 (NCT06354660), announced March 19, 2026, in 537 medication-naive type 2 diabetics over 40 weeks, reported HbA1c reductions of about 1.7–2.0% across active doses alongside meaningful weight loss ^[6][7]; at the American Diabetes Association meeting on 6 June 2026 Lilly put those at up to 2.0% A1c and 16.8% weight loss, with up to 46% of participants reaching a normal A1c ^[16]. As is consistent across the GLP-1 class, the diabetes populations lost less weight than the non-diabetic obesity population.

The second diabetes trial was the first in which retatrutide faced an approved drug rather than placebo. TRANSCEND-T2D-2 (NCT06260722) randomised 1,298 adults with type 2 diabetes on metformin, with or without an SGLT2 inhibitor, 2:2:1 to retatrutide 9 mg, retatrutide 12 mg or Ozempic (semaglutide) 2 mg, the highest approved dose, in an open-label design over 80 weeks ^[19][24]. Lilly reported topline results on 28 October 2026: A1C fell by 2.2 percentage points on both retatrutide doses against 1.8 on Ozempic, from a mean baseline of 8.3%, and body weight fell 19.5% and 20.9% against 9.4%, from a mean baseline of 102.4 kg ^[24]. The 12 mg advantage on both measures was tested with full statistical control; the 9 mg A1C advantage was a nominal result. All figures in the release are efficacy-estimand numbers, and the trial has not yet been published.

The same June presentation reported the TRIUMPH-1 basket sub-studies: knee osteoarthritis pain down by up to 4.3 points (73.1%) and moderate-to-severe obstructive sleep apnea down by up to 36.1 events per hour (60.6%) ^[16].

TRIUMPH-4 (NCT05931367), in 445 adults with obesity and knee osteoarthritis, completed in November 2025 ^[8]. Lilly reported topline results on 11 December 2025: 28.7% mean weight loss at 12 mg and 26.4% at 9 mg over 68 weeks, alongside a 4.5-point (75.8%) reduction in WOMAC knee-pain score, with more than one in eight retatrutide-treated participants free of knee pain at trial end ^[13]. These are company-reported topline figures; the full results have not yet been peer-reviewed.

The adverse-event profile in Phase 2 looked like a stronger version of what the field has seen with tirzepatide. GI symptoms dominated — nausea (up to roughly 60% at 12 mg), vomiting, diarrhea, constipation — all dose-dependent, mostly mild to moderate, mostly concentrated during titration ^[1]. Heart rate rose by about 5–7 bpm on average, peaked around week 24, then declined, a pattern broadly consistent with the class. A network meta-analysis found adverse-event frequency higher for retatrutide than tirzepatide (relative risk versus placebo around 4.1 vs 2.8), consistent with the molecule's greater pharmacological reach ^[10].

The signal worth watching emerged at Phase 3 scale and was not prominent in Phase 2: dysesthesia — tingling or altered skin sensation. It has now been reported in five Phase 3 trials, and it is dose-related in all of them.

The first trial against an active comparator shows the same pattern at a smaller scale: in TRANSCEND-T2D-2, dysesthesia was reported in 4.8% on 9 mg and 6.5% on 12 mg against 2.7% on Ozempic ^[24].

Lilly describes the events as generally mild to moderate, mostly resolving during treatment, with most participants continuing on the drug ^{[4][13][15][24]}. It is not unique to retatrutide: a 2026 pharmacovigilance analysis of VigiBase found paraesthesia and dysesthesia signals across the GLP-1 class, including hyperaesthesia with semaglutide and tirzepatide and burning sensation with semaglutide ^[14]. What looks distinctive here is the dose-dependence and the size of the gap over placebo, which fits glucagon-receptor agonism as the driver. Worth understanding before the highest dose, not a reason for alarm.

Discontinuation for adverse events runs the same way. In TRIUMPH-1 it was 4.1%, 6.9% and 11.3% across the three doses against 4.9% on placebo; in TRIUMPH-3, the cardiovascular-disease population, 9.8% and 13.5% at the top two doses against 4.8% ^[4][15]. So at the dose that produces the headline weight-loss figure, roughly one participant in nine left TRIUMPH-1 over side effects, and roughly one in seven left TRIUMPH-3. Against Ozempic in TRANSCEND-T2D-2, 4.8% and 7.9% stopped over adverse events against 2.7% ^[24]. For comparison, tirzepatide's range in SURMOUNT-1 was 4.3% to 7.1% ^[11]. The class warnings still apply — thyroid C-cell tumors (rodent only, human relevance unknown), pancreatitis, gallbladder events, severe constipation.

The Phase 3 file rests on two umbrella programs. TRIUMPH covers obesity and its complications: TRIUMPH-1 is the registrational obesity trial, additional TRIUMPH trials cover obesity with cardiovascular disease and other comorbidities, TRIUMPH-4 covers knee osteoarthritis, and basket sub-studies cover obstructive sleep apnea. Total TRIUMPH enrollment exceeds 5,800 participants, with the design described by Giblin and colleagues in Diabetes, Obesity and Metabolism (2026) ^[9]. TRANSCEND covers type 2 diabetes: the program began in 2024 and has enrolled more than 2,050 participants across three global registrational trials, of which TRANSCEND-T2D-1 (the medication-naive cohort) read out first ^[6] and TRANSCEND-T2D-2 (against Ozempic) second ^[24]. Primary endpoints are percent change in body weight (TRIUMPH) and change in HbA1c (TRANSCEND). With the July 2026 readouts, Lilly stated that it now has the clinical data package to support global submissions ^[15].

One trial in the program answers a question this page cannot. TRIUMPH-5 (NCT06662383) is a Phase 3 comparison of retatrutide against tirzepatide in 800 adults with obesity; recruitment has closed, and its primary completion date falls in November 2026 ^[18]. Every claim anywhere about the size of the gap between the two drugs — including the one in the table above — rests on cross-trial arithmetic until that reads out. A second head-to-head, TRANSCEND-T2D-2 (NCT06260722), set it against semaglutide (Ozempic 2 mg) in type 2 diabetes and read out on 28 October 2026 in retatrutide's favour on both A1C and weight ^[24]; it says nothing directly about tirzepatide. The cardiovascular and kidney outcomes trial (NCT06383390), with 10,000 participants, does not reach primary completion until February 2029 ^[20].

One smaller kidney question has now been answered, and the answer is a flat line. A Phase 2b double-blind trial (NCT05936151) randomised 146 adults who had overweight or obesity together with chronic kidney disease, with or without type 2 diabetes, to retatrutide or placebo for about 24 weeks, and measured glomerular filtration rate directly by iohexol clearance rather than estimating it from creatinine. Results were posted on 30 September 2026 ^[25]. Measured GFR fell slightly in both arms — a least-squares mean change of −1.62 mL/min/1.73 m² on retatrutide against −0.19 on placebo, a difference of −1.44 with a 95% confidence interval running from −9.17 to 6.30, p=0.714. In plain terms: over six months, in people whose kidneys were already impaired, the drug neither protected filtration nor harmed it to any degree this trial could detect, and the confidence interval is wide enough to be consistent with a modest effect in either direction. Whether retatrutide does for kidneys what GLP-1 drugs have been shown to do is a question for NCT06383390 in 2029, not for this trial.

This is the section most often misread on social media, so to be exact: retatrutide is not FDA approved. It's not approved by the EMA, the MHRA, or any other major regulator. It's investigational only. There are exactly two legal routes for a U.S. patient to receive it, a clinical trial and a pre-approval expanded access program, and both run through a physician rather than a pharmacy ^[4][21].

As of November 2026 there is no application before the FDA. Drugs@FDA, the register of approved drug applications, returns nothing for retatrutide, which is exactly what an unapproved molecule should return.

What changed in July is that the timeline stopped being analyst speculation. Alongside the TRIUMPH-2 and TRIUMPH-3 results, Lilly stated that it plans to submit a Biologics License Application for retatrutide to the FDA in the first quarter of 2027 ^[15]. A plan to file is not a filing, and a filing is not an approval: after submission comes a review that typically runs several months to a year, which puts the earliest realistic approval in late 2027, with 2028 at least as likely. There is no PDUFA date, because there is no application yet.

On compounding: retatrutide isn't on the FDA drug-shortage list, and it can't be, because it's not an approved drug. The 503A and 503B pathways don't authorize compounding of unapproved investigational molecules, so any "retatrutide" sold by compounding pharmacies or peptide vendors is operating outside any FDA-recognized authority. In August 2026 Lilly published a statement setting out the FDA's position in the agency's own terms: unapproved drugs sold for human use are illegal and may pose significant risks, retatrutide cannot lawfully be compounded, and products marketed as "research-use only" are of unknown quality and may be harmful ^[17].

The honest answer is: two routes, neither of which is a purchase and both of which run through a clinician.

Clinical trial enrollment is the first, and recruitment for the largest studies is now substantially closed, though extensions and sub-studies continue ^[5][8].

The second is newer. Lilly has opened a pre-approval expanded access record for retatrutide on ClinicalTrials.gov, NCT07629401, listed as available ^[21]. Expanded access is a genuine FDA pathway for patients outside a trial, arranged by a treating physician with the manufacturer's agreement and FDA authorization. We do not summarize its eligibility criteria here — they are on the record itself, they change, and the person who applies is the physician, not the patient.

Everything else sits outside both. The websites selling "retatrutide for research use only," the peptide vendors offering vials at various weights, and the compounding pharmacies advertising the molecule are not operating within any FDA-authorized framework. The buyer-beware concern isn't only legal; it's pharmacological — no verified potency, no sterility certification, no identity assurance a buyer in that market can rely on. In August 2026 Lilly went public against that market, calling on online platforms, payment companies and regulators to shut it down and describing sellers who label products "research-use only" while intending them for human use ^[17].

The peptide subreddits have been running well ahead of both the regulatory and clinical record. The pattern, roughly: the community gets the headline right and the extrapolations wrong.

What's right: the Phase 2 and Phase 3 weight-loss numbers are real, the triple-agonist mechanism is real, and the dose-response curve is steeper than the field assumed possible. What is right with a caveat: retatrutide produces more percent body-weight reduction than tirzepatide across separately conducted trials, but the direct comparison, TRIUMPH-5, does not read out until later this year ^[18], so nobody yet knows the size of the gap.

What's overstated or unsupported: that gray-market vials are pharmacologically equivalent to the trial drug (they aren't necessarily); that forum self-titration protocols will replicate trial outcomes (they won't necessarily); and that the safety picture is clean (the heart-rate rise is modest and self-limiting, but the dysesthesia signal is real and emerged only at Phase 3 scale). Anecdotal reports of muscle loss circulate without imaging data to support or refute them — the published trials have not reported body-composition endpoints in a form that settles the question.

The largest gap between hype and evidence concerns durability: the longest published follow-up runs to about two years, the class as a whole shows large regain after stopping, and whether retatrutide's glucagon component changes that is an empirical question without an answer yet.

The Phase 2 obesity trial took adults 18–75 with BMI ≥30, or 27–30 with a weight-related comorbidity; type 2 diabetes ruled people out, and the mean baseline BMI sat around 37 ^[1]. TRIUMPH-1 went heavier, with the same comorbidity-or-BMI gating but a higher mean BMI — shifting the study toward severe obesity, where comorbid load stacks and absolute pounds-lost numbers look larger ^[5]. The diabetes program enrolled a different cohort: TRANSCEND-T2D-1 specifically took medication-naive type 2 diabetics with a mean diabetes duration around 2.5 years ^[6]. Not enrolled, and therefore unknown: adolescents, pregnant or lactating women, people with low eGFR, and people with a history of pancreatitis or of medullary thyroid carcinoma / MEN-2. Extrapolation beyond the studied population requires caution.

Long-term safety tops the list — two years is the longest published exposure, and the rodent thyroid C-cell finding remains a class warning whose human relevance has never been cleanly resolved. The dysesthesia signal needs a mechanism, a natural history, and dose-response data. Body composition (lean-mass change) needs DEXA/MRI substudies at Phase 3 doses. Bone density after large, rapid weight loss is an open question. The hepatic effects of chronic glucagon-receptor agonism beyond fat reduction aren't fully characterized. Cardiovascular outcomes await the dedicated 10,000-participant outcomes trial, which does not reach primary completion until February 2029 ^[20] — until that reads out, retatrutide is a weight-loss drug with cardiometabolic biomarker improvements, not yet a proven cardiovascular drug. And there's no published retatrutide discontinuation data, which is the single biggest practical question for a chronic obesity medication.

Retatrutide is the most potent obesity pharmacotherapy that has entered late-stage trials. The weight-loss numbers are unprecedented for a non-surgical intervention, the mechanism is distinct from what came before, and the Phase 3 readouts have so far confirmed and modestly exceeded what Phase 2 predicted. It's also unapproved, unobtainable outside a trial or expanded access, and carries a safety signal at the top dose that didn't surface at Phase 2 scale, along with a discontinuation rate there of roughly one in nine.

What changed this summer is the shape of the wait. Lilly has now put a date on the filing rather than on the approval — a BLA in the first quarter of 2027 ^[15] — and the approval that follows it remains a projection, not a commitment. The gray market is not a safe substitute for the trial drug, and the FDA has said as much in terms that leave no room ^[17]. The right posture for a careful reader in November 2026 is informed patience: understand what the trial record shows, understand what it doesn't yet show, and note that the head-to-head against the drug you can actually get reads out before the application is even filed. The drug is coming. It isn't here yet.

The practical comparison is not with semaglutide but with tirzepatide, which is approved, prescribable today, and the most effective approved obesity drug on head-to-head evidence. Tirzepatide vs retatrutide sets the trial numbers side by side, explains why they do not subtract, and gives the date the real head-to-head reads out.

Counted from PubMed and ClinicalTrials.gov on October 9, 2026. A study counts only if participants were given retatrutide. How we count, and the same figures for every peptide we cover.

Most recent published trials

• Retatrutide-Associated Improvements in Cardiovascular Risk Biomarkers in Adults With Obesity With or Without Type 2 Diabetes. Diabetes Obes Metab, 2026.
• Efficacy and safety of retatrutide, a GIP, GLP-1, and glucagon receptor agonist, in people with type 2 diabetes and inadequate glycaemic control with diet and exercise (TRANSCEND-T2D-1): a double-blind, randomised, phase 3 trial. Lancet, 2026.
• Appetite, eating attitudes, and eating behaviours during treatment with retatrutide in adults with type 2 diabetes: Results of a phase 2 study. Diabetes Obes Metab, 2025.
• Decreases in circulating ANGPTL3/8 concentrations following retatrutide treatment parallel reductions in serum lipids. Diabetes Obes Metab, 2025.
• Effects of retatrutide on body composition in people with type 2 diabetes: a substudy of a phase 2, double-blind, parallel-group, placebo-controlled, randomised trial. Lancet Diabetes Endocrinol, 2025.

Most recent registrations

• A Study of Retatrutide (LY3437943) in Participants Without Type 2 Diabetes Who Have Obesity or Overweight NCT07357415, active, not recruiting, started 2026.
• A Study of Retatrutide (LY3437943) in Participants With Obesity or Overweight NCT07232719, active, not recruiting, started 2025.
• A Master Protocol of Multiple Agents in Adults With Metabolic Dysfunction-Associated Steatotic Liver Disease (SYNERGY-Outcomes) NCT07165028, recruiting, started 2025.
• A Study to Investigate the Response of Participants With Type 2 Diabetes Mellitus on Once-Weekly Retatrutide to Hypoglycemia NCT06982846, completed, started 2025.
• A Study to Evaluate the Effect of Retatrutide on Insulin Secretion and Insulin Sensitivity in Adult Participants With Type 2 Diabetes Mellitus NCT06982859, active, not recruiting, started 2025.

A registration is a plan filed before a study starts, not a result.

Editorial note: Informational only — not medical advice. Retatrutide is an investigational drug: it is not approved by the FDA for any use, it is not legally available by prescription or from a compounding pharmacy, and anything sold under the name outside a clinical trial or an authorized expanded access program is unverified. The doses described come from clinical trials and are not a prescribing recommendation. TRIUMPH-1 and TRIUMPH-2 are published; the other Phase 3 figures on this page come from company topline announcements that have not yet been peer-reviewed, and are labelled as such. Decisions about weight management or diabetes treatment should be made with a licensed healthcare provider familiar with your medical history. See our methodology. Last reviewed October 2026.