Retatrutide: about a quarter of body weight lost over 80 weeks in TRIUMPH-1

Design

Phase 3, randomised, double-blind, placebo-controlled, 80 weeks

Population

2,339 adults with obesity without diabetes

Primary outcome

Percent change in body weight (plus knee pain and AHI in subgroups)

Effect

-23.7% (9 mg) and -25.0% (12 mg) vs -3.9% placebo; differences -19.8 and -21.0 points

TRIUMPH-1, published in NEJM on 29 September 2026, randomised 2,339 adults with obesity but without diabetes to once-weekly retatrutide 4 mg, 9 mg or 12 mg, or placebo, for 80 weeks. Retatrutide acts on GIP, GLP-1 and glucagon receptors. Mean weight change was -17.6%, -23.7% and -25.0% across the three doses against -3.9% with placebo. Separate primary endpoints were tested in subgroups with knee osteoarthritis (574 people) and obstructive sleep apnoea (243 people).

In the knee osteoarthritis subgroup, WOMAC pain fell about 1.6 to 1.8 points more than placebo on a 10-point scale. In sleep apnoea, the apnoea-hypopnoea index fell by roughly 22 to 24 more events an hour than placebo at the two higher doses. Gastrointestinal effects were the commonest adverse events.

The weight loss is larger than most readers will be used to from single or dual agonists, but this is one industry-funded trial in people without diabetes. It is not a licensed drug, and the trial does not tell you about cardiovascular outcomes, long-term safety, or what happens after stopping. Treat it as context for patient questions rather than a prescribing decision.

• Retatrutide is investigational; patients asking for it cannot be prescribed it through normal channels yet.
• TRIUMPH-1 excluded people with diabetes, so its weight figures do not transfer directly to type 2 patients.
• Knee pain and sleep apnoea improved alongside weight, which is useful when explaining why weight matters beyond the scale.
• Expect gastrointestinal adverse effects to dominate tolerability, as with other incretin drugs.
• No cardiovascular outcome or long-term safety data exist for this agent yet.

Why it matters

Patients will hear these numbers and ask; the honest answer is that the effect is large but the drug is not available.

Don't overread it

One phase 3 trial in people without diabetes, with no outcome or long-term safety data — not a licence.

The statistics, in plain English

The 9 mg and 12 mg doses beat placebo by about 20 and 21 percentage points of body weight, with P below 0.001, so chance is an unlikely explanation. The knee and sleep apnoea results come from smaller subgroups and used a 'hybrid' estimand that treats some dropouts as failures; the intention-to-treat figures were similar, which is reassuring.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.