Reviewed & approved by Dr. Nova Reyes · Last checked 2026-10-24. ← Back to Evidence Library

Retatrutide: what the evidence says

Legend: HUMAN data · ANIMAL / LAB data · UNKNOWN · UNVERIFIED: not yet confirmed against a primary source

---

What is it?

Retatrutide is an experimental triple receptor agonist, meaning it switches on the receptors for three gut and pancreas hormones (GIP, GLP-1, and glucagon) that affect insulin, appetite, and how the body uses energy. In clinical trials it is injected under the skin once a week. ^R1

Does it work?

All results below are from trials funded by Eli Lilly, compared with placebo.

Human Weight (phase 2): in 338 adults with obesity or overweight, weight fell up to 24.2% on retatrutide vs 2.1% on placebo at 48 weeks. ^R1

Human Weight (phase 3, TRIUMPH-1, published): in 2,339 adults with obesity without diabetes, average weight change at 80 weeks was −17.6% to −25.0% across study arms vs −3.9% on placebo, though more people had skin-sensation changes and, in some arms, more stopped because of side effects (company-reported; see "Is it safe?"). (Lilly's press release quoted up to 28.3%, which counts only people who stayed on treatment. We use the published figure that counts everyone.) It also reduced knee osteoarthritis pain and sleep-apnea events more than placebo in those subgroups. ^R1

Human Weight with type 2 diabetes (phase 3, TRIUMPH-2, published): in 1,152 adults, weight changed −11.9% to −18.8% vs −5.1% on placebo at 80 weeks. ^R1

Human Type 2 diabetes: in phase 2 (281 adults), HbA1c (average blood sugar) fell up to 2.02 points vs 0.01 on placebo at 24 weeks. ^R1 In phase 3 TRANSCEND-T2D-1 (537 adults), it fell 1.69 to 1.94 points vs 0.81 on placebo at 40 weeks. ^R1

Human Heart and severe obesity (phase 3, TRIUMPH-3, press release only, not yet peer-reviewed): Lilly reports weight loss up to 22.6% at 80 weeks, counting only people who stayed on treatment. Major heart events were fewer than expected in both groups. The results don't show whether heart risk goes up or down. ^R2

Is it safe?

What clinical trials reported, compared with placebo: ^{R1, R2} - Caution Stomach side effects were the most common, and they increased with dose. In TRIUMPH-2, diarrhea happened in 27–34% on retatrutide vs 13% on placebo, and nausea in 14–28% vs 8%. ^R1 - Caution Heart rate: phase 2 found dose-dependent heart-rate increases that peaked at 24 weeks and then declined. Unknown The size of the increase (beats per minute) and any heart-rhythm problems are not in the abstracts, and we could not access the full texts. ^R1 - Caution Dysesthesia (abnormal skin sensations, such as tingling or sensitivity): in TRIUMPH-2, 4–7% vs 1% on placebo. Lilly reports 5–13% vs 1% in TRIUMPH-1 (press release). ^{R1, R2} - Caution Low blood pressure: in TRIUMPH-2, 1–6% vs under 1% on placebo. ^R1 - Caution Stopping early: in TRIUMPH-2, more people on the two higher study arms than on placebo stopped because of side effects or death (8–12% vs 5%). For TRIUMPH-1, Lilly reports 4.1–11.3% vs 4.9% (press release). In TRIUMPH-3, Lilly reports 9.8–13.5% vs 4.8% on placebo. In TRANSCEND-T2D-1 it was 2–5% vs 0%. ^{R1, R2} - Unknown Long-term safety and rare harms are unknown. Trials lasted up to about 2 years.

Gray-market "research use only" retatrutide: extra risks

• FDA warnings on "research" vials and compounding. FDA has warned companies selling unapproved retatrutide labeled "for research purposes" or "not for human consumption," as well as telehealth companies, ingredient distributors, and outsourcing facilities, noting that these products are of unknown quality and cannot be compounded under federal law. ^R10
• Reported harms. A published case report describes a woman in her mid-20s hospitalized with persistent nausea and vomiting after injections of an unverified product reported to contain retatrutide (adverse events: problems reported after use, which may or may not have been caused by the drug). ^R12
• Measuring mistakes. FDA has reported hospitalizations after people drew up 10× too much from multi-dose GLP-1 vials using U-100 insulin syringes. The same unit-versus-mg problem applies to any reconstituted vial; see the measuring-mistakes section on our semaglutide page. ^R9

What we don't know

• Unknown What is actually in a given gray-market vial (identity, strength, sterility)
• Unknown Long-term safety, and whether heart-rate increases matter over years
• Unknown Full phase 3 safety data for TRIUMPH-3 (press release only)

Is it legal?

Not FDA-approved for any use, and it cannot be used in compounding under federal law. ^R10 Lilly plans to apply for US approval in Q1 2027. ^R2 Retatrutide is not named on WADA's 2026 Prohibited List. The S0 class bans at all times "any pharmacological substance which is not addressed by any of the subsequent sections of the List and with no current approval by any governmental regulatory health authority for human therapeutic use (e.g. drugs under pre-clinical or clinical development...)," which appears to cover it while it is unapproved. Caution [UNVERIFIED: no WADA statement naming retatrutide] As of Oct 24, 2026. ^R8

What's in the vial?

FDA warns that fake (counterfeit) GLP-1 products could contain the wrong ingredients, too little, too much or no active ingredient, or other harmful ingredients. ^R10 We found no published lab test of bought retatrutide (PubMed search, Oct 24, 2026). In a small study of semaglutide (not retatrutide) vials bought from illegal online pharmacies, all 3 vials held far more drug than labeled, had low purity, and contained bacterial toxins (endotoxin). ^R11 Under the Open Peptide Standard, a GLP-1 vial overfilled by more than 20% is a critical fail (draft threshold). See results at /tests/. A passing test covers one batch, tested on one date. It does not mean the compound itself is safe.

What do users report?

Self-report, not evidence.

Unknown TBD. We'll only list this once we can cite a public source.

Study details

^R1 Human Published human trials

All funded by Eli Lilly. Study arms are listed only to report what each trial tested.

Study	People	Study arms (weekly, subcutaneous)	Follow-up	Reported	Limitations	Citation
Phase 2 obesity (Jastreboff 2023)	338	Placebo; 1, 4, 8, 12 mg arms (some with different starting steps)	48 wk	Weight −8.7% / −17.1% / −22.8% / −24.2% vs −2.1%; GI events dose-related; heart rate rose dose-dependently, peaked wk 24. Heart-rate size (bpm), arrhythmias, discontinuation: Unknown not in abstract (full text not accessible)	Phase 2, small arms	PMID 37366315
Phase 2 T2D (Rosenstock 2023)	281	Placebo; dulaglutide 1.5 mg; retatrutide 0.5, 4, 8, 12 mg arms	36 wk	HbA1c at 24 wk up to −2.02 vs −0.01 (placebo), −1.41 (dulaglutide); weight at 36 wk up to −16.94% vs −3.00%; GI events 35% vs 13% placebo; 79% completed study treatment overall. Heart rate, arrhythmias, AE discontinuation: Unknown not in abstract	Phase 2, US only	PMID 37385280
TRIUMPH-1 (Jastreboff 2026), obesity without diabetes	2,339 (574 knee OA, 243 OSA subgroups)	Placebo; 4, 9, 12 mg arms	80 wk	Weight −17.6% / −23.7% / −25.0% vs −3.9%; WOMAC pain and AHI improved vs placebo; GI events most common. Discontinuation, heart rate, arrhythmias: Unknown not in abstract (full text not accessible); company figures in R2	Weight figures use the treatment-regimen (ITT) estimand	PMID 42814954
TRIUMPH-2 (Bellido 2026), obesity + T2D	1,152	Placebo; 4, 9, 12 mg arms	80 wk	Weight −11.9% / −16.8% / −18.8% vs −5.1%; diarrhea 27–34% vs 13%; nausea 14–28% vs 8%; dysesthesia 4–7% vs 1%; hypotension 1–6% vs <1%; stopped for AEs or death 4% / 12% / 8% vs 5%; 84% completed study drug overall. Heart rate, arrhythmias: Unknown not in abstract	—	PMID 42810372
TRIUMPH-3 (company-reported, not peer-reviewed), severe obesity + cardiovascular disease	Unknown Not in release	Placebo; 9, 12 mg arms	80 wk	Weight loss up to 22.6% (efficacy estimand); stopped for AEs 9.8% / 13.5% vs 4.8%; dysesthesia 6.4% / 6.4% vs 1.3%	Press release only R2	Lilly press release, Jul 23, 2026
TRANSCEND-T2D-1 (Bajaj 2026), T2D on diet/exercise	537	Placebo; 4, 9, 12 mg arms	40 wk	HbA1c −1.69 / −1.86 / −1.94 vs −0.81; weight −11.5% / −13.9% / −15.3% vs −2.6%; stopped for AEs 2–5% vs 0%; 91% completed treatment overall. Heart rate, arrhythmias: Unknown not in abstract	—	PMID 42250575

Note: Lilly's TRIUMPH-1 press release (May 21, 2026) headlines −19.0% / −25.9% / −28.3% vs −2.2% using the efficacy estimand (people who stayed on treatment without starting other prohibited weight treatments). The same release gives the treatment-regimen estimand (everyone, regardless of adherence), −17.6% / −23.7% / −25.0% vs −3.9%, which matches the published NEJM paper. This page uses the treatment-regimen figures (press release).

^R2 Company-reported, not yet peer-reviewed

• Lilly / PR Newswire, May 21, 2026: TRIUMPH-1 discontinuation for adverse events 4.1% / 6.9% / 11.3% (low / middle / high arm; arms in the trial table) vs 4.9% placebo; dysesthesia 5.1% / 12.3% / 12.5% vs 0.9%. No heart-rate figures in the release.
• Lilly / PR Newswire, Jul 23, 2026: TRIUMPH-3 (severe obesity + cardiovascular disease; two dose arms vs placebo, arms in the trial table) weight loss up to 22.6% at 80 weeks; discontinuation for adverse events 9.8% / 13.5% vs 4.8% placebo; dysesthesia 6.4% / 6.4% vs 1.3%; weight figures are efficacy estimand; MACE-5 44 vs 52 events (HR 0.82, 95% CI 0.55–1.22); MACE-3 27 vs 23 events; fewer events than anticipated overall. TRIUMPH-2 discontinuation for AEs 3.8% / 11.6% / 7.7% vs 4.9%. "Plans to submit a Biologics License Application (BLA) for retatrutide to FDA in Q1 2027." No heart-rate figures in the release.

^R8–R12 Regulatory and safety sources

• ^R8 WADA: 2026 Prohibited List, S0 Non-approved substances, prohibited at all times. Retatrutide not named individually (checked against the official 2026 PDF text).
• ^R9 FDA alert on dosing errors with compounded semaglutide. Semaglutide-specific; used only for the unit-versus-mg context.
• ^R10 FDA's concerns with unapproved GLP-1 drugs used for weight loss: "Retatrutide and cagrilintide cannot be used in compounding under federal law"; "not found safe and effective for any condition"; warnings to telehealth companies, API distributors and outsourcing facilities; "research purposes" / "not for human consumption" warnings naming retatrutide; counterfeit wording.
• ^R11 Online-purchase testing (semaglutide, not retatrutide): Ashraf AR, et al. J Med Internet Res. 2024;26:e65440. PMID 39509151. Three vials: content 28.56%–38.69% above label; purity 7.7%–14.37% vs 99% claimed; endotoxin in all samples. Very small sample.
• Context only, not evidence: Rubin R. "Unapproved GLP-1 Drugs Retatrutide and Zenagamtide Sold Online Are No Bargain." JAMA (Medical News), Oct 9, 2026. doi:10.1001/jama.2026.20514. PMID 42853576. A news article on the gray market; only its summary was reviewed.
• ^R12 Case report: Buzzi RM, Huber LC. BMJ Case Rep. 2026;19(10):e276106. PMID 42843917. Single case; product unverified.

Full reference list (all verified on PubMed, Oct 24, 2026)

• Jastreboff AM, et al. N Engl J Med. 2023;389(6):514-526. doi:10.1056/NEJMoa2301972. PMID 37366315
• Rosenstock J, et al. Lancet. 2023;402(10401):529-544. doi:10.1016/S0140-6736(23)01053-X. PMID 37385280
• Jastreboff AM, et al. N Engl J Med. 2026 (Sep 29). doi:10.1056/NEJMoa2604169. PMID 42814954
• Bellido V, et al. Lancet. 2026 (Oct 10). doi:10.1016/S0140-6736(26)01861-1. PMID 42810372
• Bajaj HS, et al. Lancet. 2026;407(10546):2402-2413. doi:10.1016/S0140-6736(26)00967-0. PMID 42250575
• Ashraf AR, et al. J Med Internet Res. 2024;26:e65440. doi:10.2196/65440. PMID 39509151
• Rubin R. JAMA. 2026 (Oct 9; news). doi:10.1001/jama.2026.20514. PMID 42853576 (context only)
• Buzzi RM, Huber LC. BMJ Case Rep. 2026;19(10):e276106. doi:10.1136/bcr-2026-276106. PMID 42843917

---

Disclaimer

This page is for education and harm reduction, not medical advice. Seek care right away for severe stomach pain, repeated vomiting, a racing heartbeat, or fainting, and report problems to FDA MedWatch (https://www.fda.gov/safety/medwatch).