Retatrutide

The story

1. 2019-03 First-in-human single-ascending-dose study begins in healthy participants (n=45)
2. 2019-12 Phase 1b multiple-ascending-dose trial in type 2 diabetes begins (n=72)
3. 2022-09 Discovery-to-proof-of-concept paper published: LY3437943 characterized as a GCGR/GIPR/GLP-1R triple agonist, with energy expenditure attributed to the glucagon arm in obese mice
4. 2022-11 Phase 1b results published in The Lancet
5. 2023-06 Phase 2 obesity trial published in NEJM (T1; −24.2% body weight at 48 weeks on the highest dose arm) and phase 2 type 2 diabetes trial in The Lancet (T1) — the compound's public arrival
6. 2023-07 Eli Lilly's TRIUMPH phase 3 program begins enrolling (TRIUMPH-1, -2, -4)
7. 2024-06 Randomized phase 2a hepatic-fat substudy published in Nature Medicine (T1): liver fat down 81–82% at 24 weeks on the two highest dose arms
8. 2024-11 TRIUMPH-5, the first randomized head-to-head against tirzepatide, begins enrolling (n=800), registered under the name Eli Lilly and Company — the posting is the sole source for this study
9. 2025-10 TRANSCEND-CKD design and baseline paper published (T1): a phase 2b mechanistic study of retatrutide in adults with overweight or obesity and chronic kidney disease (eGFR 25–75 mL/min/1.73 m²), with and without type 2 diabetes; 146 randomized 1:1 against placebo, primary objective the change in measured GFR by iohexol clearance at 24 weeks, with MRI kidney haemodynamic and volumetric measures alongside. No administered quantity from this paper is imported (R-B).
10. 2025-10 TRIUMPH design paper published, describing the basket structure across obesity, OSA and knee OA
11. 2026-05-21 TRIUMPH-1 topline announced: weight loss reported as increasing across the three maintenance-dose arms against placebo at 80 weeks, with a 104-week extension — per sponsor announcement (T3), unpublished. No efficacy figure from this announcement is reported in this dossier. Registered enrolment n=2,335
12. 2026-07-23 TRIUMPH-2 and TRIUMPH-3 topline announced; Lilly states intent to file with FDA in Q1 2027 — per sponsor announcement (T3), unpublished. No efficacy figure from this announcement is reported in this dossier
13. 2026-07 Peer-reviewed commentary takes up the phase 3 urinary-tract-infection signal, framing the open question as one of timing (T1, independent of the sponsor) — the first non-sponsor home for that signal
14. 2026-08-06 First peer-reviewed analysis of products sold as retatrutide (T1, independent): three Australian samples all contained the expected peptide, and none contained the labelled amount of it
15. 2026-08-12 Lilly files six lawsuits against sellers of unapproved retatrutide
16. 2026-08-17 Post hoc analysis of cardiovascular risk biomarkers in both phase 2 trials published (T1 by venue, sponsor-authored): reductions in atherogenic lipoproteins in both trials, and in hs-CRP and interleukin-6 in the obesity trial only
17. 2026-08-18 Four patient-safety organizations issue coordinated warnings; a non-peer-reviewed preprint on real-world users is posted the same day (no figure from it is reported in this dossier)

What it does

• body weight
• metabolic

What the evidence shows

The finding is a collision. Retatrutide has some of the strongest evidence in this library — two large randomized, double-blind, placebo-controlled phase 2 trials from a major sponsor, published in the New England Journal of Medicine and The Lancet, plus a randomized hepatic-fat substudy in Nature Medicine — and it is approved by no regulatory authority anywhere, so the route by which essentially every non-trial user actually obtains it is a gray market whose product has now been measured once, in a peer-reviewed Australian analysis of three vials sold as retatrutide that found the right molecule in all three and roughly half to nearly double the labelled amount of it, and no peer-reviewed study of outcomes in those non-trial users exists: the only dataset that has looked is a single non-peer-reviewed preprint from a commercial analytics company, and no figure from it is reported anywhere in this dossier. What that strong evidence showed is real and large: on the highest dose arm the published phase 2 obesity trial recorded the biggest mean weight reduction then reported for any drug in that setting, against roughly two percent on placebo. The limitations are independence and stage — every efficacy figure in this file originates with the molecule's originator, and a Lilly registry posting plus a Lilly press release plus a Lilly-authored design paper is one entity and not three sources; and the phase 3 programme's results exist mainly as sponsor announcements, because no TRIUMPH trial has been published in a peer-reviewed journal, as a PubMed and Crossref search on 2026-08-29 confirmed, so every phase 3 efficacy and safety figure is withheld here. No head-to-head against tirzepatide or semaglutide has completed, which makes retatrutide's advantage over approved agents an expectation rather than a finding, and the glucagon-receptor component that makes the molecule novel is also what leaves its long-term safety unestablished: the clinical evidence for glucagon-receptor agonism is thin by the field's own account, and no published human study has isolated its contribution.

Overall evidence strength: Moderate certainty

• Jastreboff AM, Kaplan LM, Frías JP, et al. (2023) — phase 2, obesity (2023) Randomized controlled trial

  Over 48 weeks average body weight fell about 24% on the highest dose against about 2% on placebo, and every participant on the two highest dose arms lost at least 5% — the largest reduction reported for any drug in a trial of this kind at the time. It does not show retatrutide beats semaglutide or tirzepatide, because neither was in the trial, and says nothing about what happens after 48 weeks or after stopping.

  Source PMID 37366315

• Rosenstock J, Frías J, Jastreboff AM, et al. (2023) — phase 2, type 2 diabetes (2023) Randomized controlled trial

  HbA1c fell by about 2 percentage points on the top doses versus essentially nothing on placebo, and the mid-dose slow-escalation and highest-dose arms beat an approved GLP-1 drug, dulaglutide 1.5 mg, inside the same trial. It does not show superiority over the strongest approved comparators: dulaglutide 1.5 mg is a modest dose of a modest agent.

  Source PMID 37385280

• Sanyal AJ, Kaplan LM, Frias JP, Brouwers B, et al. (2024) — phase 2a MASLD substudy (2024) Randomized controlled trial

  Liver fat fell by roughly 81% and 82% on the two highest dose arms at 24 weeks, against a 0.3% increase on placebo, and 79% and 86% of those participants reached a normal liver fat fraction (<5%). This is an imaging endpoint, not a histological one — it shows fat leaving the liver, not that fibrosis or liver-related outcomes improved.

  Source PMID 38858523

• TRIUMPH-1 (2026) — phase 3, obesity (2026) Randomized controlled trial

  The sponsor announced that mean weight loss at 80 weeks increased across the three maintenance-dose arms and exceeded placebo on all three, and that the highest-dose group in the BMI ≥35 extension lost more still by 104 weeks with no plateau evident. No efficacy figure is reported here — not the arm-by-arm weight changes, not the placebo change, not the 104-week extension value, and not the difference between the efficacy and treatment-regimen estimands, which the release itself reported as materially different for the same arm. The release also reported a higher rate of stopping for adverse events on the top dose than on placebo; the discontinuation figures are not reported here either. Why: the sole source is a sponsor announcement (T3), and T3 may carry neither an efficacy nor a safety figure. What would restore these figures: peer-reviewed publication of the TRIUMPH-1 result. A PubMed and Crossref search on 2026-08-29 found none. Every figure was reported by the sponsor and is not independently re-sourceable.

  Source

• TRIUMPH-2 and TRIUMPH-3 (2026) — phase 3 (2026) Randomized controlled trial

  The sponsor announced that in people with type 2 diabetes both weight and HbA1c fell on retatrutide, and that in severe obesity with established cardiovascular disease weight fell. No efficacy figure is reported here — not the TRIUMPH-2 weight or HbA1c toplines, not the TRIUMPH-3 weight topline. The release also reported hazard ratios on a five-component and a three-component MACE endpoint that pointed in opposite directions, on trials powered for neither; those hazard ratios are not reported here either. Why: the sole source is a sponsor announcement (T3), which may carry neither an efficacy nor a safety figure. What would restore these figures: peer-reviewed publication of the TRIUMPH-2 and TRIUMPH-3 results; a PubMed and Crossref search on 2026-08-29 found none. Everything here was reported by the sponsor and is not independently re-sourceable. A dedicated 10,000-participant cardiovascular and kidney outcomes trial, TRIUMPH-Outcomes, is registered under the name Eli Lilly and Company (NCT06383390 — registry record (T2): existence and status only; self-asserted, unverified); its readout date is a sponsor estimate, not asserted here. Its existence is no longer registry-only: a peer-reviewed design and baseline paper in Nephrology Dialysis Transplantation describes TRANSCEND-CKD — a phase 2b mechanistic study of retatrutide in adults with overweight or obesity and chronic kidney disease, 146 randomized 1:1 against placebo, primary objective the change in measured GFR by iohexol clearance at 24 weeks — and states that it is designed "to inform clinical findings in the ongoing cardio-kidney outcome trial TRIUMPH-Outcomes (NCT06383390)" (Heerspink et al., 2026 — T1 by venue, part sponsor-authored). TRANSCEND-CKD itself is posted as COMPLETED with no results posted and has published no result (NCT05936151 — registry record (T2): existence and status only; self-asserted, unverified; verified via the ClinicalTrials.gov API 2026-08-29). No administered quantity from the design paper is reproduced here (R-B).

  Source

• Murugadoss K, Venkatakrishnan AJ, Soundararajan V (2026) — real-world unapproved use (2026) Observational

  The preprint reports that people who obtained retatrutide outside a trial lost substantially less weight than a trial-participant cohort and no more than users of approved tirzepatide, and that they showed a heart-rate rise and excesses of new-onset cardiovascular and neuropsychiatric symptoms relative to users of approved agents. No figure from this preprint is printed anywhere in this dossier, here or in,,,, or the front matter. It is a single non-peer-reviewed source from a commercial analytics company; the Endpoints News and Partnership for Safe Medicines items report this same preprint rather than corroborating it, so the apparent three citations are one source. The neuropsychiatric signal in particular has no corresponding signal in any trial and must not be treated as established. Even taken at face value the study cannot say why — wrong product, wrong strength, wrong schedule, or wrong patients — and the authors list residual confounding, disclosure-dependent exposure capture, and a trial cohort that necessarily includes undisclosed placebo recipients. What is needed: peer-reviewed publication, or an independent replication in a different dataset.

  Source

Who it's for, who should skip

Investigational — not FDA-approved. The FDA is explicit that it cannot be used in compounding; the only lawful routes are a clinical trial or Lilly's expanded-access program (narrow, physician-applied). Tracked pending approval, projected 2027–28.

Running it

No dosing protocol is published for this compound — there is no lawful supervised channel for the use this guide covers. What follows is monitoring, expectations, and safety framing only.

Form

injectable

What to expect

• First weeks — Gastrointestinal adverse events are the dominant early experience and are dose-related. The phase 2 obesity trial was built to examine this: it ran duplicate arms at two dose levels differing only in their starting dose, and reported a difference in gastrointestinal event rates between those arms. Reported here as a feature of how the trial was designed and what it observed; no inference is drawn about how any exposure should be begun (Confidence: Trial-documented (NEJM 2023))
• Weeks to ~24 — Weight falls steeply and dose-dependently; at 24 weeks in phase 2 the mean changes ran from −7.2% on the lowest dose arm through −12.9% and −17.3% on the pooled middle arms to −17.5% on the highest, vs −1.6% placebo. Heart rate rises over the same window, peaking at week 24 (Confidence: Trial-documented (NEJM 2023))
• ~24 weeks onward — Heart rate declines from its week-24 peak while weight continues to fall — the two curves separate (Confidence: Trial-documented (NEJM 2023))
• 48 weeks — Phase 2: −24.2% on the highest dose arm vs −2.1% placebo; 93% of that arm had lost ≥10% and 83% ≥15%. Liver fat in the MASLD substudy was already down 81–82% at 24 weeks (Confidence: Trial-documented (NEJM 2023; Nat Med 2024))
• 80 weeks — Phase 3: weight loss reported as increasing across the low, mid and high maintenance-dose arms and greater than placebo on all three, with no plateau evident, and with a visible gap between the efficacy and treatment-regimen estimands for the same arm. No figure is reported here. What would restore them: peer-reviewed publication of TRIUMPH-1 — none exists as of 2026-08-29 (Confidence: Sponsor announcement (T3) only — figures withheld; per sponsor announcement (Lilly, 2026-05-21))
• 104 weeks — The BMI ≥35 extension is the longest exposure in the public record and was reported as showing further weight loss beyond 80 weeks in the highest dose arm. No figure is reported here. What would restore it: peer-reviewed publication of TRIUMPH-1 (Confidence: Sponsor announcement (T3) only — figure withheld; per sponsor announcement (Lilly, 2026-05-21))
• After stopping — No data specific to retatrutide. No published trial has followed participants after cessation. A maintenance-of-weight-reduction study is registered under the name Eli Lilly and Company — TRIUMPH-6 (NCT06859268, n=643, registry record (T2): existence and status only; self-asserted, unverified, and the sole source for that study); its completion estimate is the sponsor's own, not stated here. Weight regain on cessation is documented for the incretin class generally (IUPHAR review) (Confidence: No data (compound-specific))
• Beyond 2 years — No exposure data of any kind exists beyond 104 weeks (Confidence: No data)

Response signs

Getting it right

• Monitoring: Heart rate is the compound-specific marker — the one thing the trials measured that moves in the wrong direction and is attributable to the novel receptor. It rose dose-dependently, peaked at 24 weeks, then declined (NEJM 2023). Whether the same happens outside a trial is unmeasured in any peer-reviewed source. A pre-exposure baseline makes any later reading interpretable. mitigate side effect
• Monitoring: Glycemia, especially alongside other glucose-lowering therapy. Glucagon and incretin agonism pull in opposite directions on hepatic glucose output; the class-level hypoglycemia signal is a threefold relative risk (DOI 10.1097/CRD.0000000000001209). mitigate side effect
• Monitoring: Body composition, not just scale weight — the DXA substudy is the reason the lean-mass question has an answer at all (Lancet Diab Endo 2025). mitigate side effect
• Monitoring: Atherogenic lipoproteins and inflammatory markers. A peer-reviewed post hoc of both phase 2 trials found consistent reductions in non-HDL cholesterol, apolipoprotein B and triglyceride-rich and LDL particle measures, and reductions in hs-CRP and interleukin-6 in the obesity trial but not the diabetes trial (Ruotolo et al., 2026 — T1 by venue, sponsor-authored). These are the markers a clinician would already be tracking; the point is that the trial record now says what they did. mitigate side effect
• Monitoring: Hepatic markers. Liver fat improved substantially in trials (Nat Med 2024); reported serious harms outside trials include liver failure (Partnership for Safe Medicines). Different exposures — do not conflate them in either direction. mitigate side effect
• Monitoring: Trial-design observation relocated. What the phase 2 trial's duplicate-arm design showed about gastrointestinal tolerability is a description of trial conduct, not a monitoring action, and now sits in EXPECTATIONS. It is not repeated here: under the standing restriction, a restricted-compound carries no directional administration guidance, quantity-free or otherwise. mitigate side effect
• General safety: Dose accuracy is the measured risk — more than identity. The only peer-reviewed analysis of product sold as retatrutide found the correct molecule in all three samples tested, and metal and inorganic impurities below the permitted daily exposure thresholds ICH Q3D sets for injectables — but content at 51.3%, 165.0% and 190.0% of the labelled amount, so that a buyer following the label would have taken roughly half, or nearly double, what they intended (Piatkowski et al., Drug Alcohol Rev 2026 — T1, independent of the sponsor). Three samples is a small base, one country, one testing service, and sterility and endotoxin were not assessed at all, so this narrows the uncertainty without closing it. mitigate side effect
• General safety: Litigation allegations are not assays. Lilly's August 2026 lawsuits allege the products at issue contain potential contaminants (CBS News); that is a pleading, and the one published assay did not find contaminants at concentrations of toxicological concern. A seller-supplied certificate of analysis is not independent verification either — the tested products carried labels, and the labels were wrong. mitigate side effect
• General safety: What the product does in people is still unmeasured. No peer-reviewed study has ever measured outcomes in non-trial users — that risk is unquantified, not small. mitigate side effect
• General safety: A "research use only" label describes the seller's legal posture, not the product's quality; such products are being marketed for human consumption (Axios, 2026-08-24). mitigate side effect
• General safety: No combination containing retatrutide has been studied; concurrent use with another incretin agent stacks overlapping GI and heart-rate effects with no data behind it. mitigate side effect
• General safety: Athletes face strict liability on a status question the sources themselves disagree about. mitigate side effect
• The physician conversation: Have the approved options — semaglutide, tirzepatide — actually been tried to an adequate dose and duration? (. Those have labels, published outcome data and a lawful channel; retatrutide has none of the three.) mitigate side effect
• The physician conversation: Do I meet the expanded-access criteria (severe obesity, at least two serious obesity-related complications, refractory to best approved therapy, unable to enrol in a trial)? Would you initiate that request? (NCT07629401 — registered under the name Eli Lilly and Company; registry record (T2): existence and status only; self-asserted, unverified) mitigate side effect
• The physician conversation: Is there a TRIUMPH site enrolling for my situation? (Fourteen phase 3 records were registered as of 2026-08-29, of which one was posted as RECRUITING on that date; recruitment status is the sponsor's own registry submission and should be checked directly.) mitigate side effect
• The physician conversation: What is my baseline resting heart rate, and what change would concern you? mitigate side effect
• The physician conversation: What other glucose-lowering medication am I on, and does that change the hypoglycemia picture? mitigate side effect
• The physician conversation: How will lean mass be tracked, not just scale weight? mitigate side effect
• The physician conversation: If I am already using a non-pharmacy product, what should be checked now? (Clinicians frequently learn about peptide use after the fact rather than before.) mitigate side effect
• The physician conversation: Do I have any sport or military testing obligation? (.) mitigate side effect

Measuring it

Retest HbA1c — per trial / expanded-access protocol.

• HbA1c baseline
• HbA1c quarterly
• Fasting glucose baseline
• Fasting glucose on cycle
• LDL-C baseline
• LDL-C annual
• Triglycerides baseline
• Triglycerides annual

How biomarkers respond

• Blood pressure Improves · 36–48 weeks — Improved — Trial-documented, reported via review — the phase 2 papers report this as a secondary endpoint; the summary figure here still comes from a 2025 review rather than the primary papers. (The lipid half of this row was replaced above by the peer-reviewed post hoc; blood pressure was not covered by it and the defect stands.) (Curr Cardiovasc Risk Rep 2025) Trial-documented

Stacking & alternatives

• Alternative to GLP-1 Receptor Agonist — The same glucagon thesis with two receptors instead of three — the closest mechanistic relatives outside Lilly (IUPHAR review)

Access & cost

Available only through a clinical trial or an expanded-access program. Not a consumer purchase.

No lawful channel to price. As of 2026-08-28 retatrutide is approved by no regulatory authority and has no lawful commercial supply in the United States, so there is no brand, pharmacy or telehealth price to report. The only lawful routes — trial enrolment and physician-initiated expanded access — do not have a consumer price (NCT07629401 — registered under the name Eli Lilly and Company; registry record (T2): existence and status only; self-asserted, unverified). A substantial illicit market nonetheless exists and has been documented by mainstream reporting, spanning online vendors, med spas, some compounding pharmacies and, in at least one documented instance, over-the-counter retail, with prices set entirely outside any regulated channel (Axios, 2026-08-24; CBS News, 2026-08-12).

Questions

Does retatrutide actually work?

In randomized, double-blind, placebo-controlled trials, yes, by a large margin: 24.2% mean weight loss at 48 weeks on the highest dose arm against about 2% on placebo, in the published phase 2 trial (T1, NEJM 2023). Phase 3 was announced as showing more weight loss again over 80 and 104 weeks, but that figure is not reported here: it rests on a sponsor announcement (T3), which may not carry an efficacy figure, and no TRIUMPH publication exists as of 2026-08-29 (Lilly, 2026-05-21 — per sponsor announcement). Whether it works for someone buying it online is a separate question, and it has no answer: no peer-reviewed study has measured outcomes in non-trial users.

Is retatrutide better than tirzepatide or semaglutide?

Unknown — nobody has finished the trial that would tell you. Cross-trial numbers favour retatrutide and an indirect network meta-analysis of 102 trials ranked it first for weight loss (DOI 10.3389/fphar.2026.1846714), but the head-to-head against tirzepatide (TRIUMPH-5 — registered under the name Eli Lilly and Company; registry record (T2): existence and status only; self-asserted, unverified) has not reported. See.

Is it approved? Can I get it legally?

Approved by no regulator anywhere; Lilly has stated it plans to file with FDA in Q1 2027, per sponsor announcement (T3) (Lilly, 2026-07-23) — a company statement of intent, not a regulatory action. The only lawful routes are enrolment in an ongoing trial, or a physician-initiated single-patient expanded-access request for severe obesity refractory to approved therapy — a route registered under the name Eli Lilly and Company, the posting being its sole source (NCT07629401 — registry record (T2): existence and status only; self-asserted, unverified). Both are investigational routes; neither is a supervised access channel.

Is "research use only" retatrutide the same thing?

The molecule usually is; the amount in the vial is not. The only peer-reviewed analysis of such product — three samples sold as retatrutide in Australia — found the correct peptide in all three and between about half and nearly double the labelled content, with metal impurities below the thresholds set for injectables (Piatkowski et al., Drug Alcohol Rev 2026 — T1). Lilly filed six lawsuits in August 2026 alleging such products contain potential contaminants (CBS News, 2026-08-12) — an allegation in litigation, not an assay. And no peer-reviewed study has ever established that gray-market product performs like the trial material — or measured what it does in people at all.

Does it cause muscle loss? Does it raise heart rate?

Some lean mass is lost, as with any large weight loss, but the DXA substudy found the proportion comparable to other obesity treatments despite the larger total loss — from only 103 paired scans (DOI 10.1016/S2213-8587(25)00092-0). Heart rate rose dose-dependently in phase 2, peaking at 24 weeks then declining; a commentary put the peak at up to 6.7 bpm (NEJM 2023; DOI 10.1080/13543784.2023.2283020).

Is there a pill or nasal version?

No registered trial has used any route other than weekly subcutaneous injection. Oral and drop products are sold; nothing in the trial record supports them.

What happens when you stop?

No retatrutide-specific data exists — no published trial has followed participants after cessation, and the maintenance study TRIUMPH-6, registered under the name Eli Lilly and Company, has not reported (NCT06859268 — registry record (T2): existence and status only; self-asserted, unverified). Weight regain after stopping is documented across the incretin class (IUPHAR review).

Is it banned in sport?

See — the sources conflict, and the conflict is not resolved here. ---