Skip to content
Peptilyze
CompoundsShopsCompareMarketLab testsGuidesMethod

Science · Reference article

GLP-1 Weight Loss and Lean Mass: What the Trials Show

What STEP 1 and SURMOUNT-1 DXA substudies, MRI data and meta-analyses show about lean mass lost on semaglutide and tirzepatide, and where '40%' comes from.

Published · Updated · Rests on: 2 DXA substudies, 1 MRI substudy, 1 phase 2 RCT arm, 3 meta-analyses (2 on GLP-1 drugs, 1 vs lifestyle), 2 diet/surgery reviews, 4 measurement papers, US labels and EU product information

This is a reference article, not a price comparison. Peptilyze compares prices of research peptides; it sells nothing and gives no medical advice. Every claim is numbered and listed under Sources.

People who lose weight on semaglutide or tirzepatide lose lean mass as well as fat. In the two largest DXA substudies lean mass fell by about a tenth, and a quarter to a third of the weight lost was lean. “Lean mass” is not the same as muscle, and almost no trial has shown whether the loss changes strength or function.

Key points

  • STEP 1 (semaglutide, 140 people, 68 weeks) reported lean mass down 9.7%. SURMOUNT-1 (tirzepatide, 124 people on drug, 72 weeks) reported down 10.9%, about 26% of the weight lost [1][2].
  • Pooled estimates of the lean share of weight lost are 25-35%, close to the 26% reported for lifestyle-based weight loss in one 2026 meta-analysis [5][6][7].
  • DXA “lean mass” includes water, organs and connective tissue, not only skeletal muscle [9][10].
  • Strength was rarely measured; grip strength and chair-stand counts did not fall where reported [3][4].
  • “Up to 40%” is the top of a 25-39% range quoted in a 2024 commentary; the trials behind it are not named in the sources we could open [23][24].

How lean mass is measured

DXA (dual-energy X-ray absorptiometry) splits the body into fat, bone and “lean soft tissue”. Accuracy depends on scanner make and model, fluid shifts mainly alter the lean value, and limb lean tissue “is not exclusively skeletal muscle” [10]. The SURMOUNT-1 authors add that a four-compartment model is more accurate than DXA [2].

MRI measures tissue volumes directly and can isolate muscle, but only small substudies used it. Bioimpedance does not measure body components at all: it measures electrical resistance and applies prediction equations. Against MRI, group averages of skeletal muscle mass agreed, but individual differences ranged from -5.3 to +5.3 kg [10].

What the two large DXA substudies found

STEP 1 substudy SURMOUNT-1 substudy
Drug Semaglutide 2.4 mg weekly Tirzepatide 5, 10, 15 mg pooled
Size 140 of 1,961 (95 drug, 45 placebo) 160 of 2,539 (124 drug, 36 placebo)
Duration 68 weeks 72 weeks
Body weight -15.0% vs -3.6% -21.3% vs -5.3%
Fat mass -19.3% -33.9% (-15.9 kg) vs -8.2% (-3.6 kg)
Lean mass -9.7% (kg not reported) -10.9% (-5.6 kg) vs -2.6% (-1.2 kg)
Lean share of loss not reported about 26% (drug), 25% (placebo)
Sponsor Novo Nordisk (no funding statement on the page) Eli Lilly

Sources: [1][2]. STEP 1 is a conference abstract; lean mass fell in absolute terms but rose as a share of body weight by 3.0 percentage points [1]. The SURMOUNT-1 substudy analysed 160 people against a planned 240, applied no multiplicity adjustment and did not evaluate physical activity or nutrient intake [2]. A 2025 correction to it was not read.

The semaglutide arm of the phase 2 BELIEVE trial, funded by Eli Lilly (see the Eli Lilly GLP-1 business overview), is a third DXA data point: in 57 people on semaglutide 2.4 mg, lean mass fell 3.9 kg at week 48 and 7.4% at week 72 [3]. BELIEVE mainly tested a myostatin-pathway drug, covered in our claim check and not repeated here.

Other trials and meta-analyses

A Novo Nordisk-sponsored STEP UP MRI substudy covered 55 people with a valid baseline scan: 49 on semaglutide (43 at 7.2 mg, 6 at 2.4 mg) and 6 on placebo. At 72 weeks, versus placebo, lean body tissue changed by -1.7 L (95% CI -4.0 to 0.7), thigh muscle volume by -1.1 L (-2.3 to 0.1; p = 0.07) and thigh muscle fat by -0.92 percentage points [4][15].

Three meta-analyses:

  • Karakasis et al. (Metabolism, 2024): 22 RCTs, 2,258 participants. Lean mass fell 0.86 kg more than with comparators, about 25% of the weight lost, while percent change in lean mass was not significant. Semaglutide 2.4 mg and tirzepatide 15 mg were among the least favourable for lean mass [5].
  • Beavers et al. (Obesity, 2024): 9 placebo-controlled trials with DXA, 659 participants in the 7 that reported lean mass (liraglutide in six; semaglutide and tirzepatide in one each). Lean mass was 1.9 kg lower than placebo (95% credible interval -3.5 to -0.2) against 6.9 kg more weight lost, about 30% [6][16].
  • Eisa and Barood (Diabetes Obes Metab, 2026): 20 RCTs using DXA or MRI. Lean share of weight lost was 35.2% for semaglutide, 25.4% for tirzepatide and 26.8% for liraglutide, against 26.2% for lifestyle interventions (p = 0.42), with moderate heterogeneity (I² = 68%) [7]. The abstract reports 15,782 participants, which looks like whole-trial populations; we could not check.

A 2026 review found only two or three comparative studies per outcome and low to very low certainty. It concluded that lean mass reductions should not be read as synonymous with sarcopenia, but that the evidence cannot exclude clinically relevant muscle loss in susceptible people [8].

Diet-only and surgical weight loss for comparison

The older benchmark is a “quarter rule”: about one-fourth of weight lost from dieting is fat-free mass. Heymsfield et al. (Obesity Reviews, 2014) traced that rule and wrote that lean tissue losses with dieting “typically tend to be small” [11]. Chaston et al. (2007) pooled 26 diet cohorts and 29 bariatric cohorts with losses above 10 kg: the fat-free share rose with stricter calorie restriction, fell with exercise in three RCTs, and was higher after gastric bypass than after gastric banding [12].

After surgery, a meta-analysis of 59 studies found lean body mass down 8.13 kg at 12 months (fat-free mass 8.23 kg; skeletal muscle 3.18 kg, from three studies), with about 55% of the lean loss in the first three months [13].

Is lean mass the same as muscle?

No. An international expert panel (Prado et al., 2025) says “lean mass” equals fat-free mass, which includes bone mineral content, that skeletal muscle is a different, tissue-level component, and that “lean body mass” should not be used [9]. Water, organs and connective tissue sit inside the number, and hydration shifts DXA lean values with little effect on fat [10]. DXA alone cannot separate muscle from the rest.

Did strength or function change?

  • BELIEVE (phase 2): grip strength changed by +2.2 kg on semaglutide 2.4 mg and +1.7 kg on placebo, which the authors describe as similar; SF-36 physical functioning did not differ significantly from placebo [3].
  • STEP UP MRI substudy: sit-to-stand repetitions in 30 seconds were about 13 at baseline and 15 at week 72 in the pooled groups; no separate comparison was reported [4].
  • SURMOUNT-1 substudy: strength and performance were not measured [2].
  • Questionnaires: EU product information for both drugs reports small improvements in self-reported physical functioning (SF-36) [21][22].

Resistance exercise and protein trials with GLP-1 drugs

We found no completed randomized trial of resistance exercise or added protein with semaglutide or tirzepatide that reports lean mass or strength. What exists:

  • Liraglutide with exercise (S-LiTE; NEJM, 2021): 195 adults who had lost 13.1 kg on a low-calorie diet, then 1 year of exercise, liraglutide, both or placebo. The combination lowered body-fat percentage by 3.9 points, about twice the exercise or drug alone. The abstract reports no lean mass, and the exercise was not specifically resistance training [14]. Funding: Novo Nordisk Foundation and others.
  • LEAN-PREP (NCT06885736): a registered 232-person, 6-month trial in Kuwait of resistance exercise, protein, both or neither with semaglutide or tirzepatide; primary outcome MRI quadriceps area. Recruiting, no results [17][18].
  • ClinicalTrials.gov also lists NCT07777575 (150 planned) and NCT07700030 (32 planned), without posted results on 2026-11-08 [18].

For drug-free comparison, lifestyle plus resistance training lost 17.5% of weight as lean mass in the 2026 meta-analysis [7].

Regulatory status

  • United States (FDA labels on DailyMed): Wegovy (semaglutide; label version 2026-06-30) and Zepbound (tirzepatide; version 2026-09-02) are prescription medicines for chronic weight management in adults with obesity or overweight with a comorbidity; Wegovy also has cardiovascular and liver indications [19][20]. Both labels state in section 12.2 that the drug “lowers body weight with greater fat mass loss than lean mass loss.” We found no warning about muscle loss or sarcopenia in either label text [19][20].
  • European Union (EMA product information, read 2026-11-08): Wegovy (first authorised 2022-01-06) cites the STEP 1 DEXA (N = 140) and STEP UP MRI (N = 55) substudies; Mounjaro (first authorised 2022-09-15) cites SURMOUNT-1 DEXA. Both say most weight loss was fat tissue; we found no muscle-loss warning in either [21][22].
  • UK (MHRA) and Germany (BfArM): not checked.
  • “Research use only” does not apply to these drugs; the results concern the products used in the trials.

The “up to 40%” figure

The figure traces to a Lancet Diabetes & Endocrinology commentary by Prado, Phillips, Gonzalez and Heymsfield (November 2024). A release on it says muscle loss, measured as fat-free mass, “can account for 25 to 39 percent of total weight lost” over 36 to 72 weeks, with no trials named. The release says the rate is “significantly higher” than with non-drug calorie restriction or ageing, giving no numbers [23][24]. The circulating “40%” looks like 39 rounded up. We could not open the commentary itself, so the trials behind the range are unverified.

How it holds: the range fits the pooled data (25.4% tirzepatide, 35.2% semaglutide), but 40% is above every pooled estimate we found [5][6][7]. The “higher than diet” claim is not borne out by the one pooled comparison we found (26.2% for lifestyle, no significant difference), though trial populations and measurements differ [7].

Risks and unknowns

  • Unknown: whether the lean mass lost is mainly muscle, what happens to it after stopping, and whether it affects falls, fractures or mortality. The Beavers authors call its clinical significance unclear [16].
  • The MRI substudy had six placebo participants [4].
  • The substudies were sponsor-funded and covered a small fraction of their parent trials; SURMOUNT-1 applied no multiplicity adjustment [1][2][4].
  • Documented adverse effects in the US labels include severe gastrointestinal reactions, gallbladder disease, pancreatitis and hypersensitivity; the most common are nausea, diarrhoea and vomiting [19][20]. These are separate from the body-composition question.

Frequently asked questions

Do semaglutide and tirzepatide cause muscle loss? Lean mass falls on DXA, about 10% of the starting value in the two large substudies. Whether this is muscle, and whether it matters clinically, is not established [1][2][8].

Is 40% of GLP-1 weight loss muscle? No source we opened supports 40% as typical. Pooled estimates are 25-35% lean mass, and lean mass is not muscle [5][6][7][9].

Is the lean share larger than with dieting? One 2026 meta-analysis found no significant difference (26.2% for lifestyle, 25.4-35.2% for incretins), but populations differ [7].

What do the drug labels say? US labels and EU product information say fat mass loss exceeds lean mass loss. We found no muscle-loss warning in any of them [19][20][21][22].

Sources

  1. Wilding JPH et al., Journal of the Endocrine Society, “Impact of semaglutide on body composition in adults with overweight or obesity: exploratory analysis of the STEP 1 study” (conference abstract, 2021). https://pmc.ncbi.nlm.nih.gov/articles/PMC8089287/
  2. Look M et al., Diabetes, Obesity and Metabolism, “Body composition changes during weight reduction with tirzepatide in the SURMOUNT-1 study” (2025; correction 2025-08-20 not read). https://pmc.ncbi.nlm.nih.gov/articles/PMC11965027/
  3. Heymsfield SB et al., Nature Medicine, “Bimagrumab plus semaglutide alone or in combination for the treatment of obesity: a randomized phase 2 trial” (2026-03-02), semaglutide-alone arms only. https://pmc.ncbi.nlm.nih.gov/articles/PMC13004672/
  4. Hjelmesaeth J et al., Diabetes, Obesity and Metabolism, “Effect of Semaglutide on Thigh Skeletal Muscle Volume, Fat Infiltration and Physical Function in People With Obesity: Secondary Analysis of STEP UP” (2026-09-20; abstract read via Europe PMC). https://doi.org/10.1111/dom.71323
  5. Karakasis P et al., Metabolism, “Effect of glucagon-like peptide-1 receptor agonists and co-agonists on body composition: systematic review and network meta-analysis” (2024-12-22; abstract via Europe PMC). https://doi.org/10.1016/j.metabol.2024.156113
  6. Beavers KM et al., Obesity, “GLP1Ra-based therapies and DXA-acquired musculoskeletal health outcomes: a focused meta-analysis of placebo-controlled trials” (2024-12-22; abstract via Europe PMC). https://doi.org/10.1002/oby.24172
  7. Eisa N, Barood O, Diabetes, Obesity and Metabolism, “Lean Mass Changes With Incretin Therapy Versus Lifestyle Intervention: A Systematic Review and Meta-Analysis of Randomised Controlled Trials” (2026-03-24; abstract via Europe PMC). https://doi.org/10.1111/dom.70666
  8. Lopes Menezes ML et al., Diabetes, Obesity and Metabolism, “Body Composition Remodelling During GLP-1-Based Therapy: A Systematic Review and Meta-Analysis” (2026-08-17; abstract via Europe PMC). https://doi.org/10.1111/dom.71220
  9. Prado CM et al., American Journal of Clinical Nutrition, “Methodological standards for body composition: levels, models, and terminology” (2025-07-17; abstract via Europe PMC). https://doi.org/10.1016/j.ajcnut.2025.05.022
  10. Prado CM et al., American Journal of Clinical Nutrition, “Methodological standards for body composition assessment: bioimpedance, DXA, CT and ultrasound” (2026-03-19; first 100,000 characters read). https://pmc.ncbi.nlm.nih.gov/articles/PMC13197919/
  11. Heymsfield SB et al., Obesity Reviews, “Weight loss composition is one-fourth fat-free mass: a critical review and critique of this widely cited rule” (2014-01-22; abstract via Europe PMC). https://doi.org/10.1111/obr.12143
  12. Chaston TB, Dixon JB, O’Brien PE, International Journal of Obesity, “Changes in fat-free mass during significant weight loss: a systematic review” (2007; abstract via Europe PMC). https://doi.org/10.1038/sj.ijo.0803483
  13. Nuijten MAH et al., Obesity Reviews, “The magnitude and progress of lean body mass, fat-free mass, and skeletal muscle mass loss following bariatric surgery” (2021-10-19 online; abstract via Europe PMC). https://doi.org/10.1111/obr.13370
  14. Lundgren JR et al., New England Journal of Medicine, “Healthy Weight Loss Maintenance with Exercise, Liraglutide, or Both Combined” (2021-05; abstract via Europe PMC). https://doi.org/10.1056/nejmoa2028198
  15. ClinicalTrials.gov, NCT05646706, STEP UP (Novo Nordisk A/S), via API. https://clinicaltrials.gov/study/NCT05646706
  16. Healio, “About 31% of weight lost by adults during GLP-1 therapy comes from lean mass” (2025-03-13), for the Beavers conflict statement. https://www.healio.com/news/endocrinology/20250313/about-31-of-weight-lost-by-adults-during-glp1-therapy-comes-from-lean-mass
  17. Alawadhi AA et al., BMJ Open, “LEAN mass Preservation with Resistance Exercise and Protein during semaglutide and tirzepatide therapy (LEAN-PREP study): a protocol” (2026-04-22; abstract via Europe PMC). https://doi.org/10.1136/bmjopen-2026-116911
  18. ClinicalTrials.gov, registry search for semaglutide or tirzepatide with resistance training or resistance exercise (API, run 2026-11-08; records NCT06885736, NCT07777575, NCT07700030). https://clinicaltrials.gov/search?intr=semaglutide%20OR%20tirzepatide&term=resistance%20training
  19. DailyMed, ZEPBOUND label (version 40, published 2026-09-02). https://dailymed.nlm.nih.gov/dailymed/services/v2/spls/487cd7e7-434c-4925-99fa-aa80b1cc776b.xml
  20. DailyMed, WEGOVY label (version 19, published 2026-06-30). https://dailymed.nlm.nih.gov/dailymed/services/v2/spls/ee06186f-2aa3-4990-a760-757579d8f77b.xml
  21. EMA, Wegovy product information (SmPC; first authorised 2022-01-06; downloaded 2026-11-08). https://www.ema.europa.eu/en/documents/product-information/wegovy-epar-product-information_en.pdf
  22. EMA, Mounjaro product information (SmPC; first authorised 2022-09-15; downloaded 2026-11-08). https://www.ema.europa.eu/en/documents/product-information/mounjaro-epar-product-information_en.pdf
  23. EurekAlert, release on Prado CM, Phillips SM, Gonzalez MC, Heymsfield SB, “Muscle matters: the effects of medically induced weight loss on skeletal muscle”, Lancet Diabetes Endocrinol (release 2024-10-25; commentary text not opened). https://www.eurekalert.org/news-releases/1062733
  24. Pennington Biomedical Research Center, “Commentary: Skeletal Muscle Health Amid Growing Use of Weight Loss Medications” (2024). https://www.pbrc.edu/news/media/2024/skeletalmuscle.aspx

This article is information, not medical advice.

More in Science