Retatrutide and muscle loss: what the trial data shows so far

Retatrutide (LY3437943) is a once-weekly injectable that activates three receptors at once: GIP, GLP-1, and glucagon. Semaglutide (Ozempic, Wegovy) activates one of those. Tirzepatide (Mounjaro, Zepbound) activates two. The third receptor, glucagon, is what makes retatrutide different, and it is part of why the weight-loss numbers coming out of its trials are larger than anything reported for an approved GLP-1 medication.

That size is exactly why the muscle question matters more here, not less. A bigger loss, spread across the same rough ratio of fat to lean tissue, means more absolute muscle at stake. But the composition data that would answer the question directly, the kind of DXA substudy that exists for semaglutide and tirzepatide, is only partially published for retatrutide. This article separates what is actually known from what is still assumption.

How much weight does retatrutide produce?

The Phase 2 obesity trial, published in the New England Journal of Medicine in 2023, randomized 338 adults with obesity to placebo or one of several retatrutide doses for 48 weeks. At the highest dose, mean weight loss reached about 24.2 percent from baseline, compared with about 2.1 percent on placebo. The 8 mg dose group averaged about 22.8 percent, and the 4 mg group about 17.1 percent.¹ For comparison, the highest tirzepatide dose in SURMOUNT-1 reached about 20.9 percent, and the highest semaglutide dose in STEP 1 reached about 14.9 percent, both also at close to a year of treatment.⁴⁵

More recent, larger numbers come from Eli Lilly's Phase 3 program. On May 21, 2026, the company reported topline results from TRIUMPH-1, an 80-week trial of 2,339 adults with obesity or overweight and at least one weight-related condition, without diabetes. The 12 mg dose averaged about 28.3 percent weight loss at 80 weeks, the 9 mg dose about 25.9 percent, and the 4 mg dose about 19.0 percent, against roughly 2.2 percent on placebo. In a 104-week extension limited to participants with a starting BMI of 35 or higher, the 12 mg group averaged about 30.3 percent loss.²

Two things are worth flagging about that Phase 3 number. First, it is topline, company-reported data from a press release, not yet a peer-reviewed publication with full methods and prespecified statistical detail. Second, retatrutide itself is still investigational: as of mid-2026 it carries no FDA approval, and full regulatory review has not begun for any indication.

What the body-composition data actually shows

Weight-loss percentages do not tell you what was lost. For that you need a body-composition substudy, typically using dual-energy X-ray absorptiometry (DXA) to separate fat from lean tissue. For retatrutide, exactly one such substudy has been published, and it was not conducted in the obesity trial population.

The substudy, led by Coskun and colleagues and published in The Lancet Diabetes & Endocrinology in 2025, enrolled 189 people with type 2 diabetes as part of a separate Phase 2 trial, with 103 completing baseline and 36-week DXA scans. Retatrutide produced dose-dependent fat-mass reductions of about 4.9 percent at 0.5 mg, 15.2 percent at 4 mg, 26.1 percent at 8 mg, and 23.2 percent at 12 mg, all clearly ahead of placebo (4.5 percent) and the comparator dulaglutide (2.6 percent). On lean mass specifically, the authors reported that the proportion of lean mass lost relative to total weight lost was similar to other obesity treatments, but the published abstract does not give a single precise lean-mass percentage the way the SURMOUNT-1 tirzepatide substudy reported "about 26 percent" of tirzepatide's weight loss as lean tissue.³⁴

That is a meaningfully different population from the one asking this question. People with type 2 diabetes differ from the general obesity population in insulin sensitivity, baseline muscle mass, and typically age, all of which can affect how much lean tissue is lost during rapid weight loss. It is a real data point, and a reassuring one on its face, but it is not a substitute for a composition substudy inside the obesity-only trials, including TRIUMPH-1, which as of this writing has reported no DXA or lean-mass figures at all.²

Why the size of the loss raises the stakes

If retatrutide holds a similar fat-to-lean ratio to tirzepatide or semaglutide, roughly a quarter of total weight lost being lean tissue, then a 28 to 30 percent total weight loss simply produces more absolute muscle loss than a 15 to 21 percent loss does, even at an identical ratio. Nobody has published the ratio for retatrutide in the population that matters most, so this is presented as a plausible extrapolation, not a demonstrated finding. It is also possible retatrutide's added glucagon-receptor activity changes energy expenditure or appetite suppression in ways that shift the ratio in either direction; glucagon signaling affects metabolic rate independently of weight loss itself, and no published trial has isolated that effect on lean mass.

For a detailed look at how the three major GLP-1-class drugs compare on the composition data that does exist, see our Ozempic vs Mounjaro vs Zepbound comparison. And for every muscle-related figure across all these drugs collected in one place, with sources, see GLP-1 muscle loss by the numbers.

What still protects your muscle, regardless of the drug

Nothing about retatrutide changes the two levers with the strongest evidence behind them, because those levers were established independently of any specific medication. In a meta-analysis of older adults losing weight through caloric restriction, resistance training offset about 93.5 percent of the diet-induced lean-mass loss that dieting alone produced.⁶ A separate dose-response meta-analysis found that protein intake up to about 1.6 grams per kilogram of body weight per day continued to improve resistance-training outcomes, after which more protein added little.⁷ Both findings come from broader weight-loss and resistance-training populations, not from GLP-1 trials specifically, but they are the best evidence available and they do not depend on which appetite-suppressing drug produced the deficit.

Hitting a protein target on an appetite that a triple agonist has suppressed hard is its own problem, distinct from the training question. Our Protein Playbook walks through a practical system for hitting a lean-mass-protective protein target on a severely reduced appetite, including which foods and timing patterns make it easiest.

What the evidence does not say

Several honest limits apply to everything above. The Phase 3 TRIUMPH-1 figures are topline, company-announced results, not a peer-reviewed publication; the numbers could shift, and important detail (confidence intervals, full adverse-event accounting, subgroup breakdowns) is not yet public. No DXA or other body-composition data has been reported from TRIUMPH-1 or any other retatrutide obesity trial as of this writing, so every claim about the fat-to-lean ratio of retatrutide weight loss in a non-diabetic population is an extrapolation from a different drug or a different population, not a direct finding. The one composition substudy that does exist was conducted in people with type 2 diabetes over 36 weeks, a shorter window and a different metabolic context than the 80-to-104-week obesity trials generating the largest headline numbers. And retatrutide is not FDA-approved; nothing here is guidance for obtaining or using an unapproved drug, and this article provides no dosing or titration information.

The bottom line

Retatrutide is producing weight loss beyond what any approved GLP-1-class drug has shown in a randomized trial, and that alone is reason to expect the muscle question will matter, possibly more than it does for semaglutide or tirzepatide. But "expect" is doing real work in that sentence. The only body-composition data published so far comes from a diabetes trial, not the obesity program driving the headlines, and it stops short of a precise lean-mass number. Until a DXA substudy from the obesity trials is published, the honest answer is that retatrutide's muscle cost is a reasonable open question, not a settled one, and the same two protective habits, resistance training and adequate protein, remain the best answer regardless of which GLP-1-class drug someone ends up using.

Frequently asked