The full results of TRIUMPH-1, the first phase 3 obesity trial of Eli Lilly's retatrutide, were published in the New England Journal of Medicine on 29 September 2026. Counting every adult who was randomised, the 12 mg dose produced an average 25.0% weight loss at 80 weeks, against 3.9% on placebo. The paper also reports less knee pain in people with osteoarthritis and fewer breathing events in people with sleep apnoea.
What happened
TRIUMPH-1 (PubMed 42814954) enrolled 2,339 adults with obesity and without diabetes. They took a once-weekly injection of retatrutide at 4 mg, 9 mg or 12 mg, or placebo, for 80 weeks. Retatrutide acts on three hormone receptors at once: GIP, GLP-1 and glucagon. The trial was funded by Lilly, and its lead author is at Yale School of Medicine.
Inside the main trial sat two smaller ones: 574 people with knee osteoarthritis and 243 with obstructive sleep apnoea.
| Placebo | 4 mg | 9 mg | 12 mg | |
|---|---|---|---|---|
| Weight lost at 80 weeks | 3.9% | 17.6% | 23.7% | 25.0% |
| Fall in knee pain score (osteoarthritis group) | 1.9 | 3.2 | 3.5 | 3.6 |
| Breathing events per hour (sleep apnoea group) | 9.9 fewer | 22.9 fewer | 34.3 fewer | 31.7 fewer |
The knee and sleep figures above use the paper's hybrid estimand. Counting everyone regardless of what happened, the knee pain falls were 3.4, 3.9 and 4.1 points against 2.5 on placebo, and the sleep apnoea falls were 22.8, 34.3 and 32.1 events an hour against 9.6. The 12 mg dose did not beat 9 mg on sleep apnoea.
The most common side effects were gastrointestinal. Lilly's presentation of the same trial at the American Diabetes Association meeting in June put stopping because of side effects at 4% to 11% on retatrutide against 5% on placebo, and named dysesthesia, an unusual skin sensation, as the most common side effect of special interest.
Why 25.0% and not 28.3%?
When Lilly announced TRIUMPH-1 in May, the headline was 28.3% for the 12 mg dose, which we covered in retatrutide explained. That figure used the efficacy estimand: an estimate of what happens if everyone stays on the drug as planned. The NEJM paper leads with the treatment-regimen estimand, which keeps in everyone who stopped, missed doses or changed treatment. That is why it is lower. It is also the fairer number for someone deciding whether to take the drug, because real life includes stopping.
What it does not tell us
- What the weight was made of. The abstract does not report how much of the loss was fat and how much was muscle. On every drug in this class, some of it is muscle.
- What happens after 80 weeks, or after stopping. The trial did not test either.
- Whether it protects the heart. TRIUMPH-1 was not designed to show that.
- How it compares head to head with tirzepatide or semaglutide. There was no active comparator, only placebo.
What it means for you
If you are taking semaglutide or tirzepatide now, nothing changes. Retatrutide is not approved anywhere as of October 2026; Lilly has said it plans to file in the US in the first quarter of 2027. Anything sold online as retatrutide is not the regulated product tested here.
If you have knee pain, the result fits what was seen with semaglutide and knee osteoarthritis: less weight on the joint means less pain. And losing a quarter of body weight is a lot of weight, which makes protein and strength work more important, not less.
What to do with this
A bigger loss applies the same muscle risk to more weight. Whatever you take, eat to a protein target set from your body, not from how little appetite you have.