Lilly has disclosed positive topline results from five Phase 3 retatrutide trials. Weight loss ranged from about 11.5% to about 28.7% depending on the trial, dose and duration, against 2.1% to 4.0% on placebo. Four of the five readouts are sponsor-reported topline figures that have not yet been through peer review. One, TRANSCEND-T2D-1, was published in The Lancet in June 2026.
Before the numbers
These are averages from controlled trials in selected populations, most with a mean baseline BMI around 40. They describe what happened to groups of participants under study conditions. They are not a prediction about any individual, and nothing on this page is medical advice or a recommendation to seek this drug.
Why “estimand” appears next to every number
A weight-loss percentage on its own is close to meaningless, and the reason is boring but decisive. Trials report results under different estimands — formal definitions of what question the number answers.
The efficacy estimand answers: what happens to people who take the drug as directed for the full period. It excludes the effect of stopping early. The treatment-regimen estimand answers: what happens across everyone randomised, including those who discontinued. The second is always the smaller number, and it is usually closer to what happens outside a trial.
Headlines quote the first. Both are given below where Lilly disclosed both. In TRIUMPH-1 the gap is 28.3% against 25.0% — not enormous, but the kind of difference that quietly inflates every downstream comparison.
Adults with severe obesity and established cardiovascular disease, with or without type 2 diabetes
Duration
80 weeks
Baseline
Mean body weight 111.4 kg (245.6 lb); mean BMI 40.4
Body weight change — efficacy estimand
Dose
Percent change
Kilograms
Pounds
9 mg
-21.6%
-23.9
-52.7
12 mg
-22.6%
-25.3
-55.8
Placebo
-3.2%
-3.5
-7.7
Cardiovascular events: Lilly reported that major adverse cardiovascular events occurred less frequently than anticipated in both the retatrutide and placebo arms. For the five-component composite, 44 events occurred on retatrutide (pooled) against 52 on placebo, a hazard ratio of 0.82 with a 95% confidence interval of 0.55 to 1.22. For the three-component composite, 27 against 23, a hazard ratio of 1.12 with a 95% confidence interval of 0.64 to 1.96. Both intervals cross 1.0, so this trial does not establish a cardiovascular benefit.
Discontinuation due to adverse events: 9.8% (9 mg) and 13.5% (12 mg), compared with 4.8% on placebo.
TRIUMPH-3 was powered for weight loss, not for cardiovascular outcomes. The dedicated outcomes trial, TRIUMPH-Outcomes, is not expected to complete before 2029.
Adults with obesity or overweight and at least one weight-related comorbidity, without diabetes
Duration
80 weeks
Baseline
Mean body weight 112.7 kg (248.5 lb); mean BMI 40.0
Body weight change — efficacy estimand
Dose
Percent change
Kilograms
Pounds
4 mg
-19.0%
-21.4
-47.2
9 mg
-25.9%
-29.2
-64.4
12 mg
-28.3%
-31.9
-70.3
Placebo
-2.2%
-2.5
-5.5
Treatment-regimen estimand (everyone randomised, including those who stopped): -17.6% on 4 mg, -23.7% on 9 mg, -25.0% on 12 mg, -3.9% on Placebo.
Discontinuation due to adverse events: 4.1% (4 mg), 6.9% (9 mg) and 11.3% (12 mg), compared with 4.9% on placebo.
A pre-specified 104-week extension in 532 participants with a BMI of 35 or above reported up to -30.3% (-38.5 kg; -85.0 lb) on the 12 mg arm escalated to maximum tolerated dose.
Lilly stated that analyses of the two basket sub-studies, in knee osteoarthritis pain and in obstructive sleep apnea, were not included in this release.
Adults with obesity or overweight and knee osteoarthritis, without diabetes
Duration
68 weeks
Baseline
Mean body weight 112.7 kg (248.5 lb); mean BMI 40.4; mean WOMAC pain subscale 6.0 points
Body weight change — efficacy estimand
Dose
Percent change
Kilograms
Pounds
9 mg
-26.4%
-29.1
-64.2
12 mg
-28.7%
-32.3
-71.2
Placebo
-2.1%
-2.1
-4.6
Pain: WOMAC pain subscale fell by 4.5 points on 9 mg and 4.4 points on 12 mg, against 2.4 points on placebo, from a baseline of 6.0 points. The percentage forms of these changes were post-hoc rather than pre-specified.
Discontinuation due to adverse events: 12.2% (9 mg) and 18.2% (12 mg), compared with 4.0% on placebo.
Dysesthesia — altered or uncomfortable skin sensation — was reported by 8.8% on 9 mg and 20.9% on 12 mg, against 0.7% on placebo.
Four of these five readouts are topline results disclosed by Lilly and have not yet been fully evaluated in a peer-reviewed publication. Only TRANSCEND-T2D-1 has been published in a peer-reviewed journal. Press releases from Lilly have been unusually detailed and have included unflattering figures such as discontinuation and dysesthesia rates, which counts in their favour. They are still documents written and released by the company that makes the drug.
Tolerability is a real signal, not a footnote
Discontinuation because of adverse events reached 18.2% on the 12 mg arm in TRIUMPH-4 against 4.0% on placebo. Dysesthesia — altered or uncomfortable skin sensation — was reported by 20.9% on that arm against 0.7% on placebo. These rates rise with dose, and the highest weight-loss numbers come from the highest doses. Reporting the efficacy without the discontinuation rate describes a group of people who are not the same as the group that started.
No cardiovascular benefit has been shown
This is the claim most likely to be overstated. TRIUMPH-3 enrolled people with established cardiovascular disease and reported event counts, but the confidence intervals crossed 1.0 in both composites. The trial was powered for weight, not for outcomes. TRIUMPH-Outcomes is the trial designed to answer the question, and it is not expected to complete before 2029.
Trial populations are selected
Mean baseline BMI in the obesity trials was around 40, which is class 3 obesity. Participants were screened, monitored and supported in ways that do not resemble ordinary care. Results in a general population would be expected to differ.
Detailed results are still to come
Lilly has said detailed results from the TRIUMPH trials will be presented at medical meetings and published in peer-reviewed journals. Until then, what exists publicly is a summary chosen by the sponsor. When the full papers appear, this page will be updated and the change logged.
Sources, with the exact release each figure came from.
Questions
How much weight did people lose on retatrutide in Phase 3?
It depends on the trial, the dose, the duration and which estimand is used. In TRIUMPH-1, adults with obesity and no diabetes lost an average of 28.3% of body weight on the 12 mg dose at 80 weeks under the efficacy estimand, against 2.2% on placebo. In trials of people with type 2 diabetes the figures were lower. These are sponsor-reported topline numbers, not peer-reviewed results.
Have the Phase 3 retatrutide results been peer reviewed?
Only one of them. TRANSCEND-T2D-1 was published in The Lancet in June 2026. The four TRIUMPH readouts are topline results disclosed by Lilly and have not yet been fully evaluated in a peer-reviewed publication.
Does retatrutide reduce heart attacks and strokes?
That has not been established. In TRIUMPH-3 the confidence intervals around the cardiovascular event comparisons crossed 1.0, meaning no benefit was demonstrated. The trial designed to answer this question, TRIUMPH-Outcomes, is not expected to complete until 2029.