Retatrutide: the monograph

Retatrutide is a once-weekly injectable peptide that activates three metabolic hormone receptors at once: GIP, GLP-1, and glucagon¹⁶. In its 48-week phase 2 obesity trial, the 12 mg dose produced a 24.2% mean body-weight reduction¹, a larger mean result than any published phase 3 figure for the approved incretin drugs¹²³²². That single sentence is why a gray market pre-sells this molecule, and it is exactly half the story.

The other half: retatrutide is an unfinished experiment. It has zero Drugs@FDA records²⁶, its 14 phase 3 registry records have zero posted results³⁰³⁹, and FDA states it cannot be used in compounding under federal law and is not a component of any FDA-approved drug²⁵. This page reports the spectacular numbers and their trial stage together, because the stage is part of the number.

Retatrutide began as Eli Lilly compound LY3437943, a single engineered peptide with agonist activity at the glucagon, GIP, and GLP-1 receptors⁶⁵. PubChem records deposited under the name map to compound CID 171390338, molecular formula C221H342N46O68, molecular weight about 4731; the synonym list includes the street nickname Triple G²⁷. The discovery paper published the full arc in 2022: balanced glucagon-receptor and GLP-1-receptor potency with greater GIP-receptor activity in vitro, weight reduction in obese mice, and a first-in-human single-dose study⁶.

From there the program moved fast: a phase 1b trial in type 2 diabetes published in 2022⁵, twin phase 2 trials in obesity and diabetes published in 2023¹², a liver-disease substudy in 2024⁴, and a phase 3 program (TRIUMPH for obesity, TRANSCEND for diabetes and kidney disease) whose first publication appeared in June 2026¹⁰³.

Semaglutide activates one receptor (GLP-1). Tirzepatide activates two (GIP and GLP-1). Retatrutide activates those two plus the glucagon receptor¹⁶. In vitro it shows balanced glucagon- and GLP-1-receptor activity with greater GIP-receptor activity⁶, and cryo-EM structures of the peptide engaging all three receptors were published in 2024¹⁶.

The glucagon component is the interesting and the double-edged part. In obese mice, weight loss was augmented by glucagon-receptor-mediated increases in energy expenditure stacked on top of the calorie-intake reduction driven by the GIP and GLP-1 receptors⁶. A 2026 IUPHAR review is blunt about the evidence level: the energy-expenditure mechanism rests largely on preclinical rodent data, and clinical data for glucagon-receptor agonism are extremely limited²¹. A dedicated human calorie-intake and energy-expenditure trial (85 participants) completed in August 2025; no results are posted yet⁴⁶.

What is measured in humans: participants on 4 mg or more reported less hunger and less disinhibited eating than placebo, and those changes correlated with weight reduction²⁰. A 2023 Lilly report also documents delayed gastric emptying, per its title⁷. The glucagon receptor is also expressed in liver, which is one hypothesis behind the unusually large liver-fat reductions in the MASLD substudy, and it is a candidate explanation for the heart-rate increases covered below⁴¹.

Every published human trial, the key substudies and analyses, and the flagship phase 3 registry rows are in the table below, each row species-labeled and linked to its source. Three results define the published record. Obesity, phase 2: 24.2% mean weight reduction at 48 weeks on 12 mg, with every 12 mg participant losing at least 5% and 83% losing at least 15%¹. Type 2 diabetes, phase 3 (the only phase 3 published so far): HbA1c down 1.94 points and weight down 15.3% at 40 weeks on 12 mg³. Liver disease, phase 2a: liver fat down 82.4% at 24 weeks on 12 mg, with 86% of participants reaching normal liver fat⁴.

Independent syntheses agree on direction and on rank: a meta-analysis of four RCTs found dose-dependent effects maximal at 12 mg¹⁷, and a Bayesian network meta-analysis of 19 trials found retatrutide had the highest odds of 15%-or-greater weight loss among the compared drug classes, alongside the highest adverse-event risk¹⁹. Blood pressure and lipids improved across trials in a 2026 meta-analysis (systolic pressure down 6.79 mmHg; triglycerides down 40.90 mg/dL)¹⁸. Kidney measures moved favorably in post hoc analyses (albuminuria down about a third at 12 mg)⁹, and a fat-versus-lean substudy found the proportion of lean-mass loss similar to other obesity treatments⁸.

There is no retatrutide dose a person can be prescribed, because there is no approved product and no prescribing label²⁶²⁵. What exists are protocol doses given to monitored trial participants. The phase 2 obesity trial randomized participants to 1, 4, 8, or 12 mg once weekly for 48 weeks, with starting doses of 2 or 4 mg and stepwise escalation; the lower 2 mg start partially mitigated gastrointestinal side effects¹. The phase 2 diabetes trial used 0.5 to 12 mg maintenance doses², and the phase 3 TRANSCEND-T2D-1 trial used 4, 9, and 12 mg³. A dedicated phase 3b trial comparing escalation schemes (TRIUMPH-9) is running through 2028, which means the sponsor itself is still working out optimal titration³⁸.

This site publishes no dose calculator and no titration planner for retatrutide, and explains why on the tools page: with no approved dose, no label, and a federal compounding prohibition, a dosing tool would be an instruction manual for an unlawful product²⁵.

The consistent finding across every trial is gastrointestinal: nausea, diarrhoea, vomiting, and constipation, dose-related, mostly mild to moderate, and partially mitigated by a lower starting dose¹². In the phase 2 diabetes trial, GI events affected 35% of retatrutide participants overall (13% to 50% by dose group) versus 13% on placebo, with no severe hypoglycaemia and no deaths². In phase 3 TRANSCEND-T2D-1, GI events again led, subsided over time, and drove discontinuation in 2 to 5% of retatrutide participants versus 0% on placebo; two deaths occurred, both in the 4 mg group and judged unrelated to study drug³.

Two signals deserve their own lines. First, heart rate: dose-dependent increases peaked at 24 weeks and declined thereafter in the phase 2 obesity trial¹; the cardiovascular section below covers what is and is not known. Second, skin sensations: a 2026 pharmacovigilance analysis of the WHO global database concluded its data strengthen evidence for dysesthesias (abnormal skin sensations, including burning) with GLP-1-receptor-agonist therapies, describes them as dose-dependent, and states such events were already observed in clinical trials of semaglutide, tirzepatide, and retatrutide; discontinuation was usually followed by resolution¹².

The honest safety headline is not any single event. It is the denominator: the entire published human safety record covers roughly 1,200 trial participants followed for 12 to 48 weeks under medical monitoring¹²³⁵. There is no label, no long-term file, and no pharmacovigilance system for the gray-market vials people actually inject²⁶²⁵.

Both directions are real and both are cited. Down: a 2026 meta-analysis of the randomized trials found systolic blood pressure fell 6.79 mmHg, diastolic 2.46 mmHg, total cholesterol 21.88 mg/dL, LDL cholesterol 13.10 mg/dL, and triglycerides 40.90 mg/dL versus control¹⁸. Up: dose-dependent heart-rate increases that peaked at 24 weeks and declined thereafter in the 48-week obesity trial¹. Glucagon-receptor agonism is a mechanistic candidate for that increase, but the mechanism literature is largely preclinical²¹.

Whether the net effect helps or harms cardiovascular outcomes is exactly what no one knows yet. Semaglutide has answered this question for itself: a 20% relative reduction in major adverse cardiovascular events in the 17,604-patient SELECT trial²⁴. Retatrutide's answer is scheduled, not known: TRIUMPH-Outcomes (NCT06383390), an event-driven trial in about 10,000 adults with atherosclerotic cardiovascular disease and/or chronic kidney disease, has an estimated primary completion of February 2029³⁹.

The standard answer for an approved drug comes from its label's contraindications. Retatrutide has no label, so that answer does not exist for anyone²⁶. What can be said honestly: the published trials enrolled adults only, under protocol entry criteria: BMI 30 or higher (or 27 to under 30 with a weight-related condition) in the obesity trial¹; HbA1c 7.0 to 10.5% and BMI 25 to 50 in the phase 2 diabetes trial²; HbA1c 7.0 to 9.5% and BMI 23 or higher in the phase 3 diabetes trial³. Everyone outside those windows, including anyone pregnant, under 18, or with significant comorbidities excluded by protocol, has essentially no published exposure data.

Because there is no approved product, there is also no validated guidance for drug interactions, kidney or liver dose adjustment, or use in any special population. Those chapters get written during and after regulatory review, which has not happened²⁶²⁵.

No one outside Lilly and FDA can date an approval, and this site will not guess. What the public record shows, with registry dates: the TRIUMPH registrational program's four core trials (over 5,800 participants, basket design spanning obesity, obstructive sleep apnea, and knee osteoarthritis) are described in a 2026 design paper¹⁰. TRIUMPH-4 reached primary completion in November 2025³³; TRIUMPH-1 in April 2026³⁰; TRIUMPH-3 in April 2026³²; TRIUMPH-2 in June 2026³¹. All four are marked completed, and none has posted results to the registry as of August 14, 2026^30313233.

The first phase 3 publication anywhere in the program is TRANSCEND-T2D-1, in type 2 diabetes, in The Lancet in June 2026³. Two more signals worth knowing: ClinicalTrials.gov lists a pre-approval expanded access record for retatrutide in obesity and obesity-related complications, status Available⁴⁴, a step The BMJ covered in August 2026¹⁵; and later-readout trials run to 2027 through 2030 (head-to-head versus tirzepatide, November 2026³⁴; maintenance of weight reduction, 2028³⁵; cardiovascular outcomes, 2029³⁹; liver-disease master protocol, 2030⁴⁷). The full registry table with every date is below and on the tools page.

Unknown, and that is the point. FDA warns that illegally marketed GLP-1-class drugs may be counterfeit and could contain the wrong ingredients, harmful ingredients, or too little, too much, or no active ingredient at all²⁵. Sellers label vials 'for research purposes' or 'not for human consumption' while shipping them to consumers with dosing instructions, a practice FDA has issued warning letters over²⁵. The composition question has reached the scientific literature: a 2026 peer-reviewed letter reports formal testing of the composition and labelling accuracy of products sold as retatrutide in Australia¹³, and The BMJ ran a fact check on a reported death after use of the unapproved injection¹⁴.

Two structural facts separate retatrutide's gray market from semaglutide's or tirzepatide's. First, there is no lawful compounded version at all: FDA states retatrutide cannot be used in compounding under federal law²⁵. Second, there is no reference product to compare a vial against: no approved manufacturing standard, no label, no lot-release oversight²⁶. Every real trial result on this page came from pharmaceutical-grade material given under medical monitoring inside a protocol; none of it validates anything sold from a website¹.

The published human pharmacokinetics come from the phase 1b multiple-ascending-dose trial: exposure was dose proportional, and the half-life was approximately 6 days, which is the basis for once-weekly subcutaneous dosing across the entire program⁵. After a single dose in the first-in-human study, a body-weight reduction persisted up to day 43⁶. No steady-state timing, absolute bioavailability, or exposure-by-population figures have been published in the abstracts of record; those chapters normally arrive with a regulatory label, which does not exist²⁶.

The list is long because the program is unfinished. No phase 3 obesity efficacy or safety results have been published or posted; the only phase 3 publication is 40-week diabetes monotherapy data³, and all 14 phase 3 registry records show hasResults false³⁰³⁹. The longest published randomized exposure is 48 weeks¹, so durability, plateau behavior, and year-two effects are unmeasured. What happens on stopping is unstudied for retatrutide; the maintenance-of-weight-reduction trial reads out around 2028³⁵. Whether the heart-rate increase¹ matters for events, and whether the blood-pressure and lipid improvements¹⁸ translate into fewer heart attacks and strokes, waits on TRIUMPH-Outcomes around February 2029³⁹. The energy-expenditure mechanism attributed to the glucagon receptor is demonstrated in rodents, not yet in published human data⁶²¹; the human study is completed but unposted⁴⁶. There is no approval anywhere in the US record, no prescribing label, no validated dose, and no lawful compounding pathway²⁶²⁵. And nothing on this page describes the vials sold online, whose contents are unverified by any regulator²⁵.

Rendered from the machine-readable citation manifest; every numbered reference resolves to a peer-reviewed journal record (PMC or Europe PMC primary), an FDA page or openFDA API record, ClinicalTrials.gov, or PubChem.