TRIUMPH-2: Retatrutide in Type 2 Diabetes and Obesity, With the Adverse Events

Two estimands, two sets of numbers. The Lancet abstract leads with the treatment-regimen estimand. Lilly’s press release leads with the efficacy estimand, which it defines as efficacy “had all randomized participants remained on study intervention.” The second set of figures is larger. Both are below, labeled, and neither is converted into the other.

The trial

TRIUMPH-2 (ClinicalTrials.gov NCT05929079) is an 80-week, randomized, double-blind, placebo-controlled phase 3 trial under a basket design, sponsored by Eli Lilly. According to the company release it randomized 1,152 participants 1:1:1:1 to retatrutide 4 mg, 9 mg or 12 mg or placebo, with a stepwise escalation every four weeks toward the target dose, and included a separate obstructive sleep apnea basket whose results Lilly says will be presented later.

Weight and blood sugar

• Treatment-regimen estimand (the Lancet abstract). Mean percent change in body weight at week 80 was −11.9% (4 mg), −16.8% (9 mg) and −18.8% (12 mg) versus −5.1% with placebo. Differences from placebo were −6.9% (95% CI −8.7 to −5.1), −11.8% (−13.7 to −9.8) and −13.8% (−15.8 to −11.8), P<0.0001 for all.
• Efficacy estimand (Lilly’s release). Mean change was −12.7%, −19.1% and −20.8% versus −4.0% with placebo, from an average baseline of 106.4 kg. A1C fell by an average of 1.4, 1.6 and 1.5 percentage points versus 0.2 with placebo, from a baseline of 7.7%. The company also reports that 28.4%, 40.0% and 39.3% reached an A1C below 5.7% versus 4.4%; the release marks that endpoint as not controlled for type 1 error.

Adverse events and discontinuations

The abstract reports that gastrointestinal events were the most frequent and more common with retatrutide than placebo. Diarrhoea occurred in 27% (4 mg), 34% (9 mg) and 34% (12 mg) versus 13% with placebo, and nausea in 14%, 21% and 28% versus 8%. Hypotension (1%, 5%, 6% versus under 1%) and dysesthesia (4%, 6%, 7% versus 1%) were also more frequent with retatrutide.

Permanent discontinuation due to adverse events or death was 4% (4 mg), 12% (9 mg) and 8% (12 mg) versus 5% with placebo. The abstract reports seven deaths across the trial arms (two, three and one in the 4, 9 and 12 mg groups and one with placebo), all judged unrelated to the study intervention by the investigators. Lilly’s release adds that other events were “generally mild to moderate” and lists decreased appetite, constipation and vomiting rates; it also lists urinary tract infection, at 3.8% to 8.0% across retatrutide arms versus 6.6% with placebo. The authors conclude that safety was “generally consistent with other molecules with GLP-1 receptor agonism.”

What the release adds, and what it is

The company release reports secondary results the abstract does not, including reductions in triglycerides, blood pressure, waist circumference and hsCRP at the highest dose. Those come from the sponsor’s own summary rather than from the paper’s abstract and have not been checked against the full text. Lilly states it plans to submit an application in Q1 2027.

For the companion obesity trial without diabetes, which uses different endpoints and populations, see the report on TRIUMPH-1 in the NEJM.

What this is, and isn’t

This is reporting on a clinical trial from its published abstract and the sponsor’s press release. It is not a review of the full paper. The trial administered escalating once-weekly doses to selected adults under medical supervision, and its results describe that setting only. Retatrutide is investigational and is not approved for human use anywhere. Nothing here is medical advice, a comparison with other drugs, or a dosing or usage recommendation.

Sources

• Retatrutide in adults with obesity and type 2 diabetes (TRIUMPH-2): a double-blind, parallel-group, randomised, placebo-controlled, phase 3 trial. The Lancet, online Sept. 29, 2026 (PubMed record and abstract) — primary: treatment-regimen estimand, adverse events, discontinuations and deaths. DOI 10.1016/S0140-6736(26)01861-1.
• Lilly press release, Sept. 29, 2026 — company release: efficacy-estimand figures, A1C, other adverse-event rates, the Q1 2027 plan.
• TRIUMPH-2, ClinicalTrials.gov NCT05929079 — primary: the trial registration.