GLP-1 muscle loss is usually told as the story of a drug that eats muscle. The trials describe something more ordinary: any large weight loss takes lean tissue with the fat, and on these drugs that was about a quarter of the weight lost in one tirzepatide trial and about a third with semaglutide in pooled data. Lean mass is not the same as muscle, since a DXA scan also counts organs and water. The stakes depend on whether strength follows, and in the one drug trial we found that tested it, training turned a strength loss into a gain.

Key takeaways

1. GLP-1 muscle loss: in trials, about a quarter to a third of the weight lost on semaglutide or tirzepatide was lean mass, and lean mass is not all muscle.

2. More than dieting? Contested: one 2026 meta-analysis found the share similar to lifestyle programs, another found it larger than with diet alone.

3. Training: in the one trial we found, resistance training three times a week cut the lean loss by about 1 kg and turned a strength loss into a gain.

4. Protein and stopping: a European expert group advises people over 65 to eat at least 1.0 to 1.2 g per kg a day, more with illness. About two-thirds of the weight returned within a year of stopping semaglutide in one trial, so any change to the drug belongs with the prescriber.

How much of the weight lost on GLP-1 drugs is muscle?

Yes, lean mass goes with the fat; that much is well established. In the body-composition substudy of SURMOUNT-1, 26 percent of the loss was lean with tirzepatide, against 25 percent on placebo. A 2026 meta-analysis pooled 25.4 percent for tirzepatide and 35.2 percent for semaglutide. Because DXA counts organs and water as lean tissue, only part of that is muscle.

Semaglutide is sold as Wegovy and Ozempic, tirzepatide as Mounjaro and, in the US, as Zepbound.

What the body-composition scans found
Share of weight lost that was lean mass (DXA) in four trials of semaglutide and tirzepatide
TrialTreatmentWho took partLostLean share
SURMOUNT-1 (substudy)Tirzepatide 5 to 15 mg weekly72 weeks160 adults with obesity or overweight, 73 % women, no diabetes21.3 % of body weight26 %25 % with placebo
STEP 1 (substudy)Semaglutide 2.4 mg weekly68 weeksAdults with obesity or overweight, no diabetes10.4 kg fat, 6.9 kg lean mass40 %as reported by a later analysis
SUSTAIN 8 (substudy)Semaglutide 1.0 mg weekly52 weeksAdults with type 2 diabetes, no placebo arm3.4 kg fat, 2.3 kg lean mass40 %our calculation
T-RExTirzepatide up to 15 mg weekly40 weeksTrial of 120 adults aged 50 to 70, half women22.61 kg body weight21 %our calculation; 16 % with resistance training
Lean share = lean mass lost divided by weight lost (SUSTAIN 8: divided by fat plus lean mass lost). Rounded.

The tirzepatide value comes from 160 of the trial's 2,539 adults. In the full trial, weight fell 15.0 to 20.9 percent over 72 weeks, against 3.1 percent on placebo. The drug moved more weight, but its split of fat and lean was close to placebo's.

On semaglutide 2.4 mg, STEP 1 participants lost 14.9 percent of their weight on average (15.3 kg) in 68 weeks. Its lean share, about 40 percent, is known only second-hand, the least certain value in the table. In people with type 2 diabetes, the SUSTAIN 8 substudy tested semaglutide 1.0 mg without a placebo arm. Lean mass fell 2.3 kg, yet the lean share of body weight rose, by 1.2 percentage points.

A 2025 network meta-analysis found about 25 percent in 22 trials, and a 2026 review of 35 trials a median of 28.3 percent across study arms, the middle half between 15.9 and 39.9 percent.

How much muscle will I lose on GLP-1? At those shares, a 20 kg (44 lb) loss means about 5 to 8 kg of lean mass, and only part of that is muscle.

Why does it happen? A review of mechanisms names too little protein and energy, hormonal adaptation, less muscle building and more breakdown, all worse with age, frailty and inactivity.

More than with a diet, or after surgery?

Two 2026 meta-analyses disagree. One, across 20 trials with 15,782 people, found lifestyle programs lost 26.2 percent as lean, not clearly different from the drugs. The other pooled 21 trials with at least 10 percent weight loss (1,334 people), only five of them with incretin drugs and four with surgery. It found 33.3 percent with incretin drugs against 22.3 percent with diet alone, 34.2 percent after surgery and 7.7 percent with diet plus exercise.

Lean mass as a share of the weight lost
Two 2026 meta-analyses of randomized trials: share of weight lost that was lean mass, by treatment (pooled estimates)
Bars: pooled share of the weight lost that was lean mass. In color: drugs; pale: no drug. Band: middle half of drug arms in 35 trials (median 28.3 %), 15.9 to 39.9 %.
20 trials, 15,782 people (drugs vs lifestyle: difference not statistically significant)
15.9 to 39.9 %Semaglutide35.2 %Tirzepatide25.4 %Lifestyle programs26.2 %Lifestyle plus resistance training17.5 %
15.9 to 39.9 %Semaglutide35.2 %Tirzepatide25.4 %Lifestyle programs26.2 %Lifestyle plus resistance training17.5 %
21 trials with at least 10 % weight loss, 1,334 people (5 with incretin drugs, 4 with surgery)
15.9 to 39.9 %Incretin drugs33.3 %Diet alone22.3 %Diet plus exercise7.7 %Bariatric surgery34.2 %
15.9 to 39.9 %Incretin drugs33.3 %Diet alone22.3 %Diet plus exercise7.7 %Bariatric surgery34.2 %
Pooled shares from different trials, not head-to-head comparisons. Lean mass by DXA includes water and organs, not only muscle.
Two 2026 meta-analyses of randomized trials: share of weight lost that was lean mass, by treatment (pooled estimates)
TreatmentLean mass, % of weight lostMeta-analysis
Semaglutide35.220 trials, 15,782 people (drugs vs lifestyle: difference not statistically significant)
Tirzepatide25.420 trials, 15,782 people (drugs vs lifestyle: difference not statistically significant)
Lifestyle programs26.220 trials, 15,782 people (drugs vs lifestyle: difference not statistically significant)
Lifestyle plus resistance training17.520 trials, 15,782 people (drugs vs lifestyle: difference not statistically significant)
Incretin drugs33.321 trials with at least 10 % weight loss, 1,334 people (5 with incretin drugs, 4 with surgery)
Diet alone22.321 trials with at least 10 % weight loss, 1,334 people (5 with incretin drugs, 4 with surgery)
Diet plus exercise7.721 trials with at least 10 % weight loss, 1,334 people (5 with incretin drugs, 4 with surgery)
Bariatric surgery34.221 trials with at least 10 % weight loss, 1,334 people (5 with incretin drugs, 4 with surgery)

Each pooled different trials, not head-to-head comparisons, so neither settles it. On one point both agree: the share was smallest where exercise was added, 17.5 percent when lifestyle programs included resistance training.

The older benchmark for dieting is a rule of one quarter fat-free mass, whose precision its own review questioned. A 2026 review counted study arms, not people: 65 percent of drug arms exceeded the expected share, as did nearly half of the non-drug arms, which lost far less weight. That benchmark is an expected value, not a limit for harm.

Why does the number look so large? Kilograms grow with the total. As an example, 25 percent of a 5 kg diet loss is 1.25 kg of lean mass and 3.75 kg of fat. The same share of 20 kg is 5 kg lean and 15 kg fat; at 40 percent, 8 kg.

Worked example
Same share, more kilograms: lean mass lost when 25 % of a 5 kg and of a 20 kg weight loss is lean (example)
Solid: lean mass lost; pale: fat lost. Every bar to scale from zero.
5 kg loss (diet-sized)
25 % lean: 1.25 kg lean mass
5 kg (11 lb)
5 kg (11 lb)
20 kg loss (drug-trial-sized)
25 % lean: 5 kg lean mass
20 kg (44 lb)
20 kg (44 lb)
The same 20 kg at a 40 % share
40 % lean: 8 kg lean mass
20 kg (44 lb)
20 kg (44 lb)
Illustration with round numbers; shares ranged widely across trials.
Same share, more kilograms: lean mass lost when 25 % of a 5 kg and of a 20 kg weight loss is lean (example)
Weight lostLean shareLean mass lost
5 kg25 %1.25 kg
20 kg25 %5 kg
20 kg40 %8 kg
Sources: huuman calculation from Look et al. 2025 and Eisa et al. 2026; spread: Batsis et al. 2026

In seven placebo-controlled trials, the lean fraction of body weight increased by 1.81 percentage points against placebo, even as lean mass dropped by 1.74 kg.

Does the muscle loss matter? Quality and strength

Likely less than the share suggests in midlife; for older adults, the trials we found do not answer it. MRI shows muscle volume falling about as much as the weight lost predicts, and fat inside the muscle falling more than predicted. Grip strength held in short and mid-term trials. But none of 35 trials in one 2026 review reported an objective test of physical function, and strength fell in the training trial's group that did not train.

An MRI analysis of SURPASS-3 scanned 246 adults with type 2 diabetes, mean age 56, 190 of them on tirzepatide. In that group thigh muscle volume fell 0.64 liters over 52 weeks, about what population data predict for that weight loss. Fat stored inside the muscle (muscle fat infiltration) fell 0.36 percentage points, more than predicted. The analysis was post hoc and sponsor-funded.

In STEP UP, 55 adults with obesity, only 6 on placebo, fat inside the thigh muscle fell 0.92 percentage points more with semaglutide than with placebo, mostly at a high dose of 7.2 mg. Muscle volume fell 1.1 liters against placebo, a difference that was not statistically reliable.

Reviews read this as mostly an adaptation to a lighter body: an interpretation, not a measurement. Over 12 weeks, 10 people with type 2 diabetes on semaglutide showed no measurable strength change, too few and too short for a verdict. A 2026 review found grip strength held in short trials, while longer observational studies linked prolonged semaglutide use in older adults with type 2 diabetes to lower grip strength, an association rather than proof.

After bariatric surgery, 64 mostly observational studies found physical performance generally improved while muscle quantity fell with the weight.

Who should be most careful

The advice changes with age, frailty and function, not with the lean share alone. Specialists say age and disease severity can matter in choosing who gets these drugs, given the risk of sarcopenia, and set no age cutoff. A four-society US advisory asks for a baseline check of strength and function, with body composition, when treatment starts.

Sarcopenic obesity means excess fat with low muscle mass or function, and it is diagnosed by testing muscle function first. The self-tests in our sarcopenia guide follow the same order.

Data on older users are thin. In the tirzepatide substudy, fat made up 76 percent of the loss in those 65 and older, close to the 74 and 75 percent of younger bands, but these groups were small and formed after the fact. That is no proof of safety, and the grip-strength signal above comes from older adults.

What kept the muscle: resistance training

The first randomized trial of resistance training with tirzepatide that we found enrolled 120 adults aged 50 to 70, all on the drug. Half were assigned supervised training three times a week for 40 weeks, half kept their usual activity. Those who trained lost 3.57 kg of lean tissue against 4.67 kg without training, 1.09 kg less, at a similar weight loss of about 22 kg. Their strength rose instead of falling.

What adding resistance training changed
Tirzepatide with or without resistance training 3 times a week: change in lean tissue and strength over 40 weeks (120 adults aged 50 to 70, mean changes)
Bars: mean change over 40 weeks in kilograms, per group. Left of the line: a loss; right: a gain.
Lean soft tissue
Tirzepatide alone−4.67 kgTirzepatide plus resistance training−3.57 kg
Tirzepatide alone−4.67 kgTirzepatide plus resistance training−3.57 kg
Strength: bench press
Tirzepatide alone−4.09 kgTirzepatide plus resistance training+4.68 kg
Tirzepatide alone−4.09 kgTirzepatide plus resistance training+4.68 kg
Strength: lat pulldown
Tirzepatide alone−4.42 kgTirzepatide plus resistance training+8.58 kg
Tirzepatide alone−4.42 kgTirzepatide plus resistance training+8.58 kg
Strength: knee extension
Tirzepatide alone−4.42 kgTirzepatide plus resistance training+5.88 kg
Tirzepatide alone−4.42 kgTirzepatide plus resistance training+5.88 kg
In diet trials without the drug
Older adults dieting, 6 months: lean mass −2 % with resistance training, −3 % combined, −5 % aerobic only.
6 diet trials in older adults: resistance training prevented 93.5 % of the lean mass loss.
Weight loss was similar in both groups (about 22 kg). Blinding of the strength testers could not be ensured.
Tirzepatide with or without resistance training 3 times a week: change in lean tissue and strength over 40 weeks (120 adults aged 50 to 70, mean changes)
MeasureTirzepatide alone, kgTirzepatide plus resistance training, kg
Lean soft tissue-4.67-3.57
Strength: bench press-4.094.68
Strength: lat pulldown-4.428.58
Strength: knee extension-4.425.88

Blinding of the strength testers could not be ensured, a limit the authors name; 91 percent of sessions were attended.

A six-month diet trial randomized 160 older adults with obesity, including a control group. Every exercise group lost 9 percent of its weight, but lean mass fell 2 percent with resistance training and 3 percent with combined training, against 5 percent with aerobic training alone. In six pooled diet trials in older adults with obesity, resistance training prevented 93.5 percent of the lean loss, 0.819 kg.

Can lost muscle be rebuilt on the drug? The trials we found did not test that. Strength rose with training while participants kept taking it, though lean tissue still fell, only less. A four-society advisory recommends resistance training and an appropriate diet to preserve muscle and bone during treatment, and there is a full plan for strength training after 40.

An investigational antibody, apitegromab, cut lean loss by 1.9 kg against placebo in a 24-week phase 2 trial of 102 adults on tirzepatide; it is not approved.

Protein: how much, and when appetite is low

The four-society advisory asks for an appropriate diet, but its summary names no protein number for people on these drugs. The closest values come from European geriatric experts (PROT-AGE): for people over 65, at least 1.0 to 1.2 g per kg of body weight a day, at least 1.2 with exercise and 1.2 to 1.5 with acute or chronic illness. Severe kidney disease without dialysis (eGFR under 30) may call for less.

Grams of protein a day, by body weight
Daily protein at the European geriatric values for people over 65, worked out for 60 to 90 kg of body weight
1.0 to 1.2 g/kg: over 65 1.5 g/kg: with acute or chronic illness, top of range
Body weight1.0 to 1.2 g/kgg a day1.5 g/kgg a day
60 kg132 lb60 to 72 g90 g
70 kg154 lb70 to 84 g105 g
80 kg176 lb80 to 96 g120 g
90 kg198 lb90 to 108 g135 g
Set for people over 65, not specifically for GLP-1 treatment; with kidney disease, ask first. With obesity, a dietitian may use a lower reference weight.
Sources: PROT-AGE Study Group (Bauer et al. 2013); huuman calculation

The table covers 60 to 90 kg (132 to 198 lb), and the protein calculator does any weight. With obesity, a dietitian may calculate from a lower reference weight.

Whether more protein keeps more muscle is contested. In adults over 50 losing weight, 20 trials found higher-protein diets kept more lean mass (at least 25 percent of energy or 1.0 g per kg). Across adults without kidney disease, 22 trials of high protein showed no significant change in fat-free mass; body fat fell and kidney markers showed no consistent harm, in mostly short trials.

The training trial set no diet: participants were asked to keep their usual protein intake. If nausea keeps you from eating enough, protein included, tell the prescriber. Protein needs by age explains the values.

After stopping: what comes back

Much of the weight, within a year. One year after stopping semaglutide 2.4 mg and the lifestyle program, people in the STEP 1 extension had regained about two-thirds of their loss: 11.6 of 17.3 percentage points, a net loss of 5.6 percent. In SURMOUNT-4, adults switched from tirzepatide to placebo after 36 weeks regained 14.0 percent of their weight in a year, while those who continued lost 5.5 percent more.

The two withdrawal trials did not report what the regained weight is made of. In 78 women after menopause who regained weight after a diet, less lean tissue came back than had gone: 0.26 kg of lean tissue went per kg of fat lost, but only 0.12 kg returned per kg of fat regained. That is one diet cohort, not a drug trial.

Exercise may help hold the result. In one trial, adults first lost 13.1 kg by diet, then had a year of exercise, liraglutide (an older GLP-1 drug), both or placebo. A year after all treatment ended, the combination group weighed 5.1 kg less than the liraglutide-only group; 109 of 195 came to that follow-up. Any change to the drug belongs with the prescriber.

More health topics to explore

References

  1. Look M et al. — Body composition changes during weight reduction with tirzepatide in the... (2025)
  2. Eisa N et al. — Lean Mass Changes With Incretin Therapy Versus Lifestyle Intervention: A... (2026)
  3. Neeland IJ et al. — Changes in lean body mass with glucagon-like peptide-1-based therapies and... (2024)
  4. Green DJ et al. — Resistance exercise to attenuate lean mass loss during tirzepatide therapy... (2026)
  5. Jastreboff AM et al. — Tirzepatide Once Weekly for the Treatment of Obesity (2022)
  6. Wilding JPH et al. — Once-Weekly Semaglutide in Adults with Overweight or Obesity (2021)
  7. McCrimmon RJ et al. — Effects of once-weekly semaglutide vs once-daily canagliflozin on body... (2020)
  8. Karakasis P et al. — Effect of glucagon-like peptide-1 receptor agonists and co-agonists on body... (2025)
  9. Batsis JA et al. — Effect of Incretin-Based and Nonpharmacologic Weight Loss on Body... (2026)
  10. Rossi G et al. — Muscle loss and GLP-1R agonists use (2026)
  11. Busk-Cirera L et al. — Effects of Incretin-Based Therapies, Diet and Exercise Interventions, and... (2026)
  12. Heymsfield SB et al. — Weight loss composition is one-fourth fat-free mass: a critical review and... (2014)
  13. Laverde LP et al. — Effect of GLP-1 receptor agonists at doses for obesity management on muscle... (2026)
  14. Sattar N et al. — Tirzepatide and muscle composition changes in people with type 2 diabetes... (2025)
  15. Hjelmesæth J et al. — Effect of Semaglutide on Thigh Skeletal Muscle Volume, Fat Infiltration and... (2026)
  16. Langer HT et al. — Weight loss with GLP-1 medicines does not result in a disproportionate loss... (2026)
  17. Prokopidis K et al. — Glucagon-like peptide-1 receptor agonists and muscle strength changes in... (2026)
  18. Eksteen G et al. — Beyond Sarcopenia: A Systematic Review and Meta-Analysis on Muscle Outcomes... (2026)
  19. Mozaffarian D et al. — Nutritional priorities to support GLP-1 therapy for obesity: a joint... (2025)
  20. Donini LM et al. — Definition and diagnostic criteria for sarcopenic obesity: ESPEN and EASO... (2022)
  21. Villareal DT et al. — Aerobic or Resistance Exercise, or Both, in Dieting Obese Older Adults (2017)
  22. Sardeli AV et al. — Resistance Training Prevents Muscle Loss Induced by Caloric Restriction in... (2018)
  23. Pratley RE et al. — Apitegromab for lean mass preservation during tirzepatide-induced weight... (2026)
  24. Bauer J et al. — Evidence-based recommendations for optimal dietary protein intake in older... (2013)
  25. Kim JE et al. — Effects of dietary protein intake on body composition changes after weight... (2016)
  26. Teixeira Silva PR et al. — Effects of High-Protein Diets on Renal Function and Body Composition in... (2026)
  27. Wilding JPH et al. — Weight regain and cardiometabolic effects after withdrawal of semaglutide:... (2022)
  28. Aronne LJ et al. — Continued Treatment With Tirzepatide for Maintenance of Weight Reduction in... (2024)
  29. Beavers KM et al. — Is lost lean mass from intentional weight loss recovered during weight... (2011)
  30. Jensen SBK et al. — Healthy weight loss maintenance with exercise, GLP-1 receptor agonist, or... (2024)

About this article · Written by the huuman Team. Our content is based on peer-reviewed research and clinical guidelines. We follow editorial standards grounded in scientific evidence.

This article is for educational purposes only and does not constitute medical advice. Health and training decisions should be discussed with qualified professionals.

November 14, 2026
November 19, 2026