After years of trying to change my weight, I understand how easily a big percentage can begin to feel like a promise. I want to slow that moment down enough to ask: who was studied, what happened and what was treatment like?
This entry was updated on 1 October 2026 after the peer-reviewed TRIUMPH-2 paper appeared in The Lancet on 29 September. The July company announcement remains linked below. TRIUMPH-3 is still described here as company-reported evidence.
What does the fuller study add?
TRIUMPH-2 followed 1,152 adults with type 2 diabetes and obesity or overweight for 80 weeks. Participants were assigned by chance, and neither they nor the study team knew which treatment they were receiving. Women made up 48% of the group. Lilly funded the study. [2]
Why have I seen both 18.8% and 20.8%?
Both figures relate to the highest-dose group, but they answer different questions. The paper’s main approach includes everyone assigned to a group and estimates missing results. The company’s other analysis estimates what would happen if people stayed on treatment without prohibited weight-management treatment. [2] [3]
| Study group | Main paper analysis: includes everyone assigned | Company analysis: estimates continued treatment |
|---|---|---|
| 4 mg | 11.9% | 12.7% |
| 9 mg | 16.8% | 19.1% |
| 12 mg | 18.8% | 20.8% |
| Placebo | 5.1% | 4.0% |
These doses identify trial groups, not personal dosing instructions. Neither column predicts an individual result, and each needs its own placebo comparison. For a hypothetical starting weight of 100 kg, 18.8% would mean 18.8 kg. That is just an illustration of the calculation.
What else changed?
In Lilly’s continued-treatment analysis, 34.9% of the 12 mg group lost at least a quarter of their starting weight, and 59.5% no longer met the BMI definition of obesity. HbA1c, a measure of blood sugar over time, fell by up to 1.6 percentage points across retatrutide groups. That does not establish diabetes remission without medication. [3]
The company also reported changes in the 12 mg group’s blood fats: triglycerides fell 39.5% and non-HDL cholesterol fell 19.6%. The top blood-pressure number fell 10.8 mmHg, waist measurement fell 16.9 cm, and hsCRP, a marker of inflammation, fell 58.3%. Those are changes from the starting measurements, not the extra benefit over placebo. They do not prove fewer heart attacks or strokes. [3]
What was the difficult part?
At 12 mg, Lilly reported diarrhoea in 33.6%, nausea in 28.0%, constipation in 16.8% and vomiting in 15.7%. The corresponding placebo figures were 13.2%, 8.0%, 9.4% and 4.2%. Adverse events led 7.7% to stop, compared with 4.9% on placebo. [3]
The published abstract also reports more low blood pressure and altered sensations with retatrutide. [2] I want those experiences beside the weight figures, because being able to live with treatment matters too.
This is strong weight-loss evidence for the people and time period studied. It does not establish lifetime safety or benefits for menopause symptoms, ADHD, migraine or addiction. Including women is not the same as studying menopause specifically.
And what about TRIUMPH-3?
The July announcement reported 22.6% weight reduction versus 3.2% with placebo in TRIUMPH-3’s highest-dose group, using the continued-treatment analysis. Those participants had severe obesity and established cardiovascular disease, with or without type 2 diabetes. It was a different group, and this entry still treats that result as company-reported. [1]
A publication is not an approval
Retatrutide remains investigational. Lilly’s planned US submission in Q1 2027 is a plan to apply, not an approval already granted. [3] And, as my own story makes clear, clinical-trial evidence cannot verify something bought online.
Read the sources and follow the update
- Lilly: July TRIUMPH-2 and TRIUMPH-3 announcement ↗23 July 2026 · Original manufacturer report, retained for the article history.
- Bellido and colleagues: TRIUMPH-2 in The Lancet ↗29 September 2026 · Peer-reviewed Phase 3 trial · Published abstract reviewed · DOI 10.1016/S0140-6736(26)01861-1.
- Lilly: detailed TRIUMPH-2 results ↗29 September 2026 · Manufacturer account; headline numbers use the efficacy estimand.