Retatrutide has no prescribing label, so there is no official list of its side effects. It is an experimental once-weekly injection from Eli Lilly that acts on three hormone receptors — GIP, GLP-1 and glucagon — and it is not approved anywhere. Every figure on this page is a trial figure, and every one is set beside the placebo group from the same trial, because a side-effect rate without its placebo column cannot be read.
This record was read on 27 October 2026 from the phase 2 results posted on ClinicalTrials.gov, the TRIUMPH-1 paper in the New England Journal of Medicine, the TRIUMPH-2 paper in The Lancet, the TRANSCEND-T2D-1 paper, and Lilly's releases of 21 May and 23 July 2026. What retatrutide is, its regulatory position and its weight-loss results are on our retatrutide page; how it compares with Zepbound is on retatrutide vs tirzepatide. This page is only about safety.
Where the record stands
Three phase 3 trials now have peer-reviewed papers. TRIUMPH-1 (2,339 adults with obesity and no diabetes, 80 weeks) was published in NEJM on 29 September 2026 (Jastreboff et al.). TRIUMPH-2 (1,152 adults with obesity and type 2 diabetes, 80 weeks) was published in The Lancet on 10 October 2026 (Bellido et al.). TRANSCEND-T2D-1 (537 adults with type 2 diabetes, 40 weeks) appeared in June 2026 (Bajaj et al.). TRIUMPH-3 (people with severe obesity and cardiovascular disease) is still reported only in Lilly's 23 July 2026 release.
The NEJM abstract for TRIUMPH-1 says only that "the most common adverse events were gastrointestinal"; the by-dose table is in Lilly's 21 May 2026 release, which is what we quote below. On 27 October 2026 ClinicalTrials.gov showed TRIUMPH-1 (NCT05929066) and TRIUMPH-2 (NCT05929079) as completed with no results posted to the registry yet. The phase 2 trial (NCT04881760) is the one study whose full adverse-event table is public, arm by arm.
The stomach and bowel effects, by dose
TRIUMPH-1 — adults with obesity, no diabetes, 80 weeks (Lilly release, 21 May 2026):
| Event | Placebo | 4 mg | 9 mg | 12 mg |
|---|---|---|---|---|
| Nausea | 14.8% | 28.6% | 38.4% | 42.4% |
| Diarrhea | 13.5% | 25.2% | 34.1% | 32.0% |
| Constipation | 10.9% | 23.8% | 25.9% | 26.1% |
| Vomiting | 4.8% | 10.6% | 22.8% | 25.3% |
| Stopped for adverse events | 4.9% | 4.1% | 6.9% | 11.3% |
TRIUMPH-2 — adults with obesity and type 2 diabetes, 80 weeks (Lancet paper; diarrhea and nausea from the abstract, the rest from Lilly's 23 July release):
| Event | Placebo | 4 mg | 9 mg | 12 mg |
|---|---|---|---|---|
| Diarrhea | 13% | 27% | 34% | 34% |
| Nausea | 8% | 14% | 21% | 28% |
| Constipation | 9.4% | 14.0% | 16.2% | 16.8% |
| Vomiting | 4.2% | 5.5% | 10.2% | 15.7% |
| Decreased appetite | 4.5% | 5.8% | 12.3% | 17.1% |
| Stopped for adverse events or death | 5% | 4% | 12% | 8% |
Two patterns stand out. Vomiting rises most steeply with dose — about five times placebo at 12 mg in TRIUMPH-1, against roughly three times for nausea. And people with type 2 diabetes reported less nausea at every dose (28% against 42.4% at 12 mg). The two TRIUMPH trials were not designed to compare populations, and the placebo groups differed too (14.8% against 8%), so it is an observation, not a finding.
In TRIUMPH-3 (severe obesity with cardiovascular disease), Lilly reported diarrhea in 30.1% and 24.4% at 9 and 12 mg against 8.7% on placebo, nausea in 21.7% and 22.4% against 5.8%, and stopping for adverse events in 9.8% and 13.5% against 4.8%. In TRANSCEND-T2D-1, a 40-week trial in people with recently diagnosed type 2 diabetes, 2–5% stopped for adverse events against 0% on placebo, and no severe hypoglycaemia was reported.
What the starting dose did in phase 2
The phase 2 trial (338 adults with obesity, 48 weeks, Jastreboff et al., NEJM 2023) tested two starting doses for the same target dose, which no phase 3 trial did. The paper says gastrointestinal events were "partially mitigated with a lower starting dose (2 mg vs. 4 mg)". The results posted on ClinicalTrials.gov let the size of that be counted. Percentages are our arithmetic from the posted counts:
| Arm (target dose, starting dose) | People | Nausea | Vomiting | Decreased appetite |
|---|---|---|---|---|
| Placebo | 70 | 8 (11.4%) | 1 (1.4%) | 6 (8.6%) |
| 1 mg | 69 | 10 (14.5%) | 2 (2.9%) | 9 (13.0%) |
| 4 mg, started at 2 mg | 33 | 6 (18.2%) | 4 (12.1%) | 6 (18.2%) |
| 4 mg, started at 4 mg | 33 | 12 (36.4%) | 4 (12.1%) | 8 (24.2%) |
| 8 mg, started at 2 mg | 35 | 6 (17.1%) | 2 (5.7%) | 4 (11.4%) |
| 8 mg, started at 4 mg | 35 | 21 (60.0%) | 9 (25.7%) | 11 (31.4%) |
| 12 mg, started at 2 mg | 62 | 28 (45.2%) | 12 (19.4%) | 18 (29.0%) |
The same 8 mg target produced nausea in 60% of people when the first dose was 4 mg and in 17% when it was 2 mg. The arms are small — 33 to 35 people — so the exact gap is uncertain, but the direction is consistent at both target doses. It is also why the phase 3 trials escalated from a low starting dose. The registry lists only events reported by at least 5% of an arm as "other" adverse events, so rarer events are not in this table.
Skin sensitivity (dysesthesia)
The side effect most specific to retatrutide in these releases is dysesthesia: skin that feels sore, burning, tingling or painful to light touch. The approved GLP-1 labels group the same sensations under that term, and our page on skin sensitivity on GLP-1 drugs sets out what the Wegovy and Zepbound labels report.
| Trial | Placebo | 4 mg | 9 mg | 12 mg |
|---|---|---|---|---|
| TRIUMPH-1 (obesity) | 0.9% | 5.1% | 12.3% | 12.5% |
| TRIUMPH-2 (type 2 diabetes, Lancet) | 1% | 4% | 6% | 7% |
| TRIUMPH-3 (cardiovascular disease) | 1.3% | — | 6.4% | 6.4% |
Lilly describes these events as "generally mild to moderate", says "the majority resolved during treatment" and that most participants continued on the drug. The phase 2 registry entries point the same way on a smaller scale: "sensitive skin" in 2 of 62 people on 12 mg and none on placebo, "hyperaesthesia" in 2 of 62 against 1 of 70. For context, the Wegovy label reports dysesthesia in 22% of people on its new 7.2 mg dose against 0.3% on placebo, so the signal is not unique to retatrutide; it appears to track how strongly and how long the receptors are driven.
Heart rate and blood pressure
Every drug in this class raises resting heart rate a little. The phase 2 retatrutide paper reports "dose-dependent increases in heart rate" that "peaked at 24 weeks and declined thereafter". A 2026 meta-analysis of randomised trials in people without diabetes (Zhang et al., Eur J Med Res) estimated the average rise against placebo at 3.46 beats per minute for retatrutide (95% CI 1.74 to 5.18), against 3.35 for semaglutide, 2.37 for liraglutide and 2.05 for tirzepatide. On that pooled evidence retatrutide sits inside the range of the approved drugs, not above it. The TRIUMPH papers' own heart-rate tables were not in their abstracts when we read them.
TRIUMPH-2 reported hypotension (low blood pressure) in 1%, 5% and 6% of people on 4, 9 and 12 mg against fewer than 1% on placebo. Most of those participants were taking other diabetes medicines, and weight loss itself lowers blood pressure, so the trial cannot say how much is the drug and how much the weight.
Serious events
The phase 2 registry results count serious adverse events per arm: 3 of 70 on placebo and between 0 and 3 per retatrutide arm. Among them were one case of acute pancreatitis (12 mg arm) and one of acute cholecystitis (8 mg arm, 4 mg start). Pancreatitis and gallbladder disease are both on the labels of the approved GLP-1 drugs; our pages on pancreatitis and gallbladder problems set out the rates those labels give.
Deaths: TRIUMPH-2 recorded two on 4 mg, three on 9 mg, one on 12 mg and one on placebo, which the investigators judged unrelated to the study drug. TRANSCEND-T2D-1 recorded two, both on 4 mg, also judged unrelated. One death occurred in the phase 2 trial (4 mg arm, 4 mg start). Those counts come from trials of hundreds of people followed for under two years and cannot rule a rare harm in or out.
What the record cannot tell you yet
- Long-term safety. No participant has been followed beyond about two years. The cardiovascular-outcomes trial has not reported.
- Rare harms. There is no marketed product, so there is no FDA adverse-event reporting data, which is where rare problems with the approved drugs first surfaced.
- What happens after stopping. No retatrutide trial has published an off-drug period. What the approved drugs show is on our page about stopping a GLP-1 drug.
- Products sold online. Vials labelled "retatrutide" and sold as research chemicals are not Lilly's medicine. Their contents, purity and dose are unverified, and none of the figures on this page applies to them. FDA has sent warning letters to sellers.
How it compares
Placed beside the approved drugs, the shape is familiar and the numbers are higher at the top dose. TRIUMPH-1's 11.3% stopping rate at 12 mg is about twice the 6.2% recorded at tirzepatide's top dose in SURMOUNT-1 — a comparison across different trials and years, set out with its caveats on retatrutide vs tirzepatide. The class-wide rates from the approved labels are on our GLP-1 side-effects overview, and the other glucagon-acting candidate is compared on survodutide vs retatrutide. Trial dose labels appear on this page only as the names of the trial arms; what the trials used and when is reported on Peptifact's retatrutide dosage page.
Anyone weighing a trial or an unapproved product should take the questions on this page to a doctor, who can set them against their own health record.
Sources
Read 27 October 2026: ClinicalTrials.gov records NCT04881760 (results posted), NCT05929066 and NCT05929079 (no results posted); Jastreboff et al., NEJM 2023 (PMID 37366315); Jastreboff et al., NEJM 2026, TRIUMPH-1 (PMID 42814954); Bellido et al., Lancet 2026, TRIUMPH-2 (PMID 42810372); Bajaj et al., Lancet 2026, TRANSCEND-T2D-1 (PMID 42250575); Zhang et al., Eur J Med Res 2026 (PMID 41582189); Eli Lilly releases of 21 May 2026 and 23 July 2026. Corrections go to the contact page.