Two ways of countingEli Lilly headlines the efficacy estimand: what happened in people who kept taking the drug. Journal papers lead with the treatment-regimen estimand: everyone who was randomized, including people who stopped. The second number is always lower. Both are shown below and labeled.
TRIUMPH-1: obesity without diabetes
2,339 adults, 80 weeks, published in the New England Journal of Medicine in September 2026.[1][2]
| Dose | Journal (all randomized) | Lilly (stayed on drug) |
|---|---|---|
| 4 mg | −17.6% | −19.0% |
| 9 mg | −23.7% | −25.9% |
| 12 mg | −25.0% | −28.3% (−70.3 lb) |
| Placebo | −3.9% | −2.2% |
In a 104-week extension in people with a BMI of 35 or more, the 12 mg group reached −30.3% (Lilly's efficacy figure).[2] Sleep apnea substudies in TRIUMPH-1 also showed large falls in breathing interruptions during sleep.[11]
TRIUMPH-4: obesity and knee osteoarthritis
445 adults, 68 weeks. This was the first phase 3 result, announced in December 2025.[4][3]
| Dose | Journal (all randomized) | Lilly (stayed on drug) |
|---|---|---|
| 9 mg | −20.0% | −26.4% |
| 12 mg | −23.7% | −28.7% (−71.2 lb) |
| Placebo | −4.6% | −2.1% |
Knee pain scores (WOMAC) fell by up to 4.5 points, or 75.8%, against 2.4 points on placebo.[4]
TRIUMPH-2 and TRIUMPH-3
| Trial | Population | 4 mg | 9 mg | 12 mg | Placebo |
|---|---|---|---|---|---|
| TRIUMPH-2 (journal) | Obesity + type 2 diabetes, 1,152 people | −11.9% | −16.8% | −18.8% | −5.1% |
| TRIUMPH-2 (Lilly) | Same trial, stayed on drug | −12.7% | −19.1% | −20.8% | −4.0% |
| TRIUMPH-3 (Lilly) | BMI ≥35 + heart disease, 1,949 people | – | −21.6% | −22.6% | −3.2% |
Weight loss in people with type 2 diabetes (TRIUMPH-2) was lower than in TRIUMPH-1, which enrolled people without diabetes.[5][6] TRIUMPH-3 has been announced but not yet published in a journal. Its heart-outcome estimates were imprecise because there were fewer events than expected.[6]
Type 2 diabetes trials (TRANSCEND-T2D)
| Dose | HbA1c change | Weight change |
|---|---|---|
| 4 mg | −1.69% | −11.5% |
| 9 mg | −1.86% | −13.9% |
| 12 mg | −1.94% | −15.3% |
| Placebo | −0.81% | −2.6% |
Average HbA1c at the start was 7.9%.[7] In TRANSCEND-T2D-2, a head-to-head trial against semaglutide 2 mg, retatrutide 12 mg lowered weight by 20.9% against 9.4% at 80 weeks (Lilly figures).[9] See retatrutide vs semaglutide.
Retatrutide before and after: the average result in pounds
This site does not show before-and-after photos, because individual pictures cannot be checked and often come from unapproved products. The trials give a verified "before and after" for the average participant instead. Lilly reported these average losses in pounds (efficacy estimand):
| Trial | Dose | Average loss | Time on treatment |
|---|---|---|---|
| TRIUMPH-1[2] | 4 mg | 47.2 lb (19.0%) | 80 weeks |
| TRIUMPH-1[2] | 9 mg | 64.4 lb (25.9%) | 80 weeks |
| TRIUMPH-1[2] | 12 mg | 70.3 lb (28.3%) | 80 weeks |
| TRIUMPH-1 extension, BMI ≥35[8] | 12 mg | 85.0 lb (30.3%) | 104 weeks |
| TRIUMPH-4[4] | 12 mg | 71.2 lb (28.7%) | 68 weeks |
| TRIUMPH-3[6] | 12 mg | 55.8 lb (22.6%) | 80 weeks |
| TRIUMPH-2 (type 2 diabetes)[6] | 12 mg | 49.6 lb (20.8%) | 80 weeks |
| TRANSCEND-T2D-1 (type 2 diabetes)[8] | 12 mg | 36.6 lb (16.8%), still falling | 40 weeks |
Working backwards from the TRIUMPH-1 figures, the 12 mg group started at roughly 248 lb on average (70.3 ÷ 0.283; our arithmetic, not a published number). Averages hide a wide range: in phase 2, 26% of people on 12 mg lost 30% or more of their weight, while some lost far less.[10]
Retatrutide weight loss per week
Weight loss on retatrutide was fastest in the first months and slowed later. Averaged over the whole 80 weeks of TRIUMPH-1, the 12 mg group lost about 0.9 lb a week (70.3 lb ÷ 80 weeks).[2] The phase 2 trial shows how uneven the pace was:[10]
| Period | Weight change in the period | Average per week |
|---|---|---|
| Weeks 0–24 | −17.5% | About 0.73% of body weight |
| Weeks 24–48 | A further −6.7% (to −24.2%) | About 0.28% of body weight |
The weekly figures are our arithmetic from the published 24- and 48-week averages. For a person starting at 250 lb, 0.73% a week is about 1.8 lb, and 0.28% is about 0.7 lb. Individual results vary widely.
Retatrutide benefits beyond weight loss
Retatrutide also improved several weight-related conditions in trials:
| Outcome | Result | Trial |
|---|---|---|
| Blood sugar (HbA1c) | −1.94 points on 12 mg vs −0.81 on placebo from a 7.9% baseline; up to 90% reached HbA1c below 7.0%[7][8] | TRANSCEND-T2D-1, 40 weeks |
| Knee osteoarthritis pain | WOMAC pain down up to 4.5 points (75.8%) vs 2.4 on placebo; more than 1 in 8 people on retatrutide pain-free at the end[4][13] | TRIUMPH-4, 68 weeks |
| Knee osteoarthritis pain | WOMAC pain down up to 4.3 points (73.1%) from 6.0; more than 1 in 6 pain-free at the end[13] | TRIUMPH-1 knee substudy, 80 weeks |
| Obstructive sleep apnea | Breathing pauses down 31.7–34.3 per hour on 9–12 mg vs 9.9 on placebo[11] | TRIUMPH-1-OSA, 80 weeks |
The knee results were presented at the American College of Rheumatology's ACR Convergence 2026 in November 2026. Lilly notes that the percentage reductions in knee pain come from a post-hoc analysis, meaning they were not pre-specified.[13] These benefits come with the side effects described on the side effects page, and none has been reviewed by the FDA yet.
Retatrutide and sleep apnea: what Lilly's trial showed
Retatrutide reduced sleep apnea severity by more than half in its phase 3 sleep apnea study. June 2026 headlines that an "Eli Lilly trial shows retatrutide treats obesity and sleep apnea" refer to results Lilly presented at the American Diabetes Association meeting: in people with moderate-to-severe obstructive sleep apnea, the apnea-hypopnea index fell by up to 36.1 events per hour (60.6%) from a starting average of 58.6 (efficacy estimand, 80 weeks).[8]
The full results of two sleep apnea trials were published in the New England Journal of Medicine in October 2026. Both lasted 80 weeks and compared 4, 9 and 12 mg with placebo:[11]
| Trial | People | 4 mg | 9 mg | 12 mg | Placebo |
|---|---|---|---|---|---|
| TRIUMPH-1-OSA (no diabetes) | 243 | −22.9 | −34.3 | −31.7 | −9.9 |
| TRIUMPH-2-OSA (type 2 diabetes) | 448 | −19.2 | −24.7 | −25.4 | −10.3 |
The effect was smaller in people with type 2 diabetes, as with weight loss. Retatrutide is not approved for sleep apnea or any other condition.
Phase 2 results
For how the drug works, see what retatrutide is. The 48-week phase 2 obesity trial reported −24.2% on 12 mg against −2.1% on placebo. 26% of people on 12 mg lost 30% or more of their body weight.[10] Full arm-by-arm numbers are in the dosage chart.
Which trials are still running, and when they finish, is tracked on the retatrutide clinical trials page.
How to read these numbers
- These are averages. Individual results in every group ranged widely.
- Trials differ in length (40 to 104 weeks), population and counting method, so compare within a trial, not across trials.
- Results describe people in controlled trials with regular medical follow-up.
Related
- Side effectsWhat came with these results
- Retatrutide vs tirzepatideComparing the trial programs
- FDA approval statusWhen it could be available
Frequently asked questions
How much weight do people lose on retatrutide?
Is retatrutide stronger than tirzepatide?
The trial numbers are higher, but the drugs have not yet been compared directly in a published trial. The head-to-head TRIUMPH-5 trial has not reported results. See retatrutide vs tirzepatide.
Are there real retatrutide before and after results?
Does retatrutide help knee pain?
In people with knee osteoarthritis and obesity, retatrutide lowered WOMAC knee pain by up to 4.5 points (75.8%) in TRIUMPH-4, against 2.4 points on placebo.[4] It is not approved for this use.
Does retatrutide help sleep apnea?
In trials, yes. In TRIUMPH-1-OSA, breathing interruptions during sleep fell by up to 34.3 events per hour on retatrutide against 9.9 on placebo; in people with type 2 diabetes (TRIUMPH-2-OSA) the fall was up to 25.4 against 10.3.[11] It is not approved for this use.
What is the efficacy estimand?
A way of analyzing trial results that estimates the effect if everyone had kept taking the drug as planned. It gives a higher number than the treatment-regimen estimand, which counts everyone who was randomized.
Sources
- Jastreboff AM et al. Retatrutide for obesity (TRIUMPH-1)
- Lilly: retatrutide delivered powerful weight loss in pivotal phase 3 obesity trial (TRIUMPH-1)
- Schnitzer TJ et al. Retatrutide in obesity and knee osteoarthritis (TRIUMPH-4)
- Lilly: retatrutide delivered weight loss of up to 71.2 lbs in TRIUMPH-4
- Bellido V et al. Retatrutide in obesity with type 2 diabetes (TRIUMPH-2)
- Lilly: retatrutide successful in two additional phase 3 obesity trials (TRIUMPH-2, TRIUMPH-3)
- Bajaj HS et al. Retatrutide in type 2 diabetes (TRANSCEND-T2D-1)
- Lilly: retatrutide results presented at ADA 2026
- Lilly: retatrutide vs semaglutide 2 mg in TRANSCEND-T2D-2
- Jastreboff AM et al. Triple-hormone-receptor agonist retatrutide for obesity: a phase 2 trial
- Grunstein RR et al. Retatrutide for obstructive sleep apnea (TRIUMPH-1-OSA, TRIUMPH-2-OSA)
- Rosenstock J et al. Retatrutide in people with type 2 diabetes: a phase 2 trial
- Lilly to showcase new retatrutide, Taltz (ixekizumab) and Zepbound (tirzepatide) data at ACR Convergence 2026