Published: October 15, 2026 · Last updated: October 15, 2026
You are losing weight on a GLP-1 medicine, but a headline says you may be losing muscle. Then your scan says your lean-mass percentage has improved. Can both be true? Yes. The useful question is not whether one number went up or down. It is what was measured, what changed compared with a control group, and whether everyday strength or function changed too.
Quick answer
GLP-1 muscle loss concerns deserve attention, but lean-mass loss is not the same as proven muscle failure. Semaglutide and tirzepatide body-composition studies report reductions in lean mass alongside larger reductions in fat. A scan alone does not establish loss of strength. Human studies that actually tested function show why both measurements matter. Plan nutrition, activity, and follow-up with your clinician rather than stopping treatment because of a headline. Tirzepatide DXA substudy; SEMALEAN; exercise and liraglutide trial analysis.
Four things to keep separate
- Lean mass versus skeletal muscle. Do not automatically relabel a body-composition result as an equal amount of contractile muscle lost.
- Kilograms versus percentages. A smaller absolute amount can make up a larger share of a much smaller body.
- Mass versus function. Grip strength, stair climbing, and a scan answer different questions.
- Evidence versus advice. Sensible exercise and nutrition support does not mean every proposed protein dose or program has been proven with every medicine.
What does “lean mass” actually tell you?
Lean-mass measurements include more than skeletal muscle. In particular, DXA lean soft tissue includes organs and water as well as muscle; bone mineral is measured separately. It is not a direct test of how strongly a muscle contracts. Terminology also varies between reports, so read the study’s definition rather than assuming every use of “lean,” “fat-free,” and “muscle” means exactly the same thing. 2026 methodological discussion of DXA and muscle-health assessment.
Sarcopenia is not simply a disappointing number on a scale. The EWGSOP2 clinical consensus emphasizes low strength, uses muscle quantity or quality to confirm the diagnosis, and uses poor physical performance to identify severe disease. That is a clinical framework, not a home diagnostic test or a GLP-1-specific trial. Original consensus, DOI 10.1093/ageing/afy169.
Here is a purely mathematical example, not a trial result: someone weighing 100 kg with 50 kg of lean tissue has a 50% lean proportion. At 80 kg with 46 kg of lean tissue, the proportion is 57.5%. The percentage improved even though the absolute amount fell by 4 kg. Neither percentage alone tells us whether that person’s strength improved.
What the human studies measured
These are selected comparisons, not a ranking of medicines. Different populations, measurements, follow-up periods, and study designs prevent a fair head-to-head conclusion from these rows. On a phone, scroll inside the table.
| Study and design | Finding | Important limit |
|---|---|---|
| STEP 1 exploratory DXA analysis, 2021 140 adults without diabetes; 95 semaglutide / 45 placebo; 68 weeks. | Semaglutide: weight −15.0%, fat mass −19.3%, lean mass −9.7%; lean share of body weight rose 3.0 percentage points. Placebo weight −3.6%. | Conference-supplement abstract, not a full muscle-function report. Do not call a 9.7% lean-mass change a 9.7% strength loss. |
| SURMOUNT-1 DXA substudy, 2025 160 with baseline/end scans: 124 pooled tirzepatide / 36 placebo; 72 weeks. | Weight −21.3% versus −5.3%; lean mass −10.9% versus −2.6%. Approximately 25% of weight lost was lean mass in both groups. | Selected scan substudy, not all 2,539 parent-trial participants. Lean-mass proportion is not the proportion of starting muscle lost. |
| SEMALEAN, 2026 Prospective semaglutide cohort; 115 enrolled, 106 completed; 12 months. | Lean mass fell about 3 kg by month 7, then stabilized. Handgrip strength increased 4.5 kg at month 12. | Within-group changes, not placebo-adjusted effects. Does not prove semaglutide itself strengthens muscle. |
| Jensen et al., 2026 Secondary randomized-trial analysis; 193 adults without diabetes; 52 weeks after a diet run-in. | Exercise plus liraglutide versus liraglutide alone: stair test 1.2 seconds faster (95% CI 0.6–1.9). Absolute strength was preserved across groups. | Fitness outcome, not proof that every exercise program preserves muscle mass with semaglutide or tirzepatide. |
Why “25% muscle loss” can be misleading
The tirzepatide report’s approximate 25% figure refers to the share of lost weight that was lean mass, not losing one-quarter of all starting muscle. Those denominators are different. The study was funded by Eli Lilly, and several authors were employees/shareholders; those disclosures matter when assessing replication. Its published correction changes Figure 4’s fat/lean change axis from percent to kilograms. Do not mix that figure’s units with the percentage results above. Original report and disclosures; correction, DOI 10.1111/dom.70050.
Also ask whether a number is a change from baseline or a difference between groups. A treatment group can lose lean mass while improving its lean-to-fat ratio. Neither observation cancels the other. The responsible interpretation keeps both, rather than selecting the more reassuring or alarming statistic.
Does lower lean mass necessarily mean lower strength?
No. SEMALEAN is useful precisely because it measured both. However, its participants had a mean BMI of 46.3, and the cohort had no randomized untreated comparison. The abstract does not supply a confidence interval for the reported grip change. We should not translate that observation into a guaranteed response for older, frail, or less severely obese adults. SEMALEAN original abstract.
The randomized liraglutide analysis asks a different question: what did adding structured exercise contribute? It reported better stair performance with the combination than medication alone; the 95% confidence interval above shows uncertainty around the mean difference. Exercise was moderate-to-vigorous, not a test of protein supplements alone. Novo Nordisk supplied medicine/placebo pens, and the paper lists foundation and institutional support. These results support measuring function, not assuming a scan captures it. Original analysis and funding statement.
Does resistance training guarantee lean-mass preservation?
“Guarantee” goes beyond the evidence. A 2025 dietary-weight-loss pilot randomized 48 adults to diet alone or diet plus home resistance exercise for 12 weeks; 39 completed. It found no significant between-group body-composition benefit, but strength and chair-stand performance improved with training. Body composition was assessed by bioimpedance and muscle thickness by ultrasound, not DXA. Small size and unmeasured exercise adherence limit interpretation. This was not a GLP-1 trial. Binmahfoz et al.; PMID 40760444.
A null scan result does not make the functional benefit irrelevant. Conversely, stronger grip is not proof of larger muscles. When discussing a program, ask what it is intended to improve and how that will be checked. A plan can be worthwhile without promising a perfect scan.
What remains unproven about protein and exercise together?
The 2026 LEAN-PREP publication describes a six-month, four-group trial designed for 232 adults starting semaglutide or tirzepatide: usual care, resistance exercise, protein support, or both. Its primary outcome is MRI-measured quadriceps area. This publication is a protocol, not a result. Its planned 1.6 g/kg/day protein target should not be presented as a proven universal prescription. We did not verify current recruitment or results-posting status here. Protocol; DOI 10.1136/bmjopen-2026-116911.
A separate 2026 expert consensus explicitly acknowledges that many supportive-care recommendations rely on indirect evidence and clinical experience. That does not make them useless. It means we should distinguish practical care guidance from a direct randomized demonstration in the exact population. Nutritional and lifestyle supportive-care consensus.
A practical conversation plan for your next visit
The following is a discussion checklist, not an individual exercise, protein, or medication prescription. Bring your treatment timeline and any body-composition report so the clinician can interpret the context rather than just one alarming percentage.
- Define the concern. Is it lower measured lean mass, new weakness, difficulty eating, or reduced ability to do usual tasks? Write down when it started and whether it is progressing.
- Check the measurement. Ask what the report measures, its units, whether the same method was used before, and whether a change is large enough to interpret confidently.
- Agree on a functional measure. Ask whether a standardized grip, chair-rise, or walking assessment would be useful. Do not improvise strenuous home tests if balance or mobility is a concern.
- Discuss what you can actually eat. Ask for a food-first nutrition assessment and an individualized protein plan, especially if low appetite or gastrointestinal symptoms make eating difficult. A supplement label is not a personalized care plan.
- Choose achievable activity and a review point. Ask what resistance activity is appropriate, how to progress it, and when to reassess intake, function, symptoms, and weight together.
CDC’s general adult guidance includes 150 minutes of moderate aerobic activity and at least two days of muscle-strengthening activity each week. That is population guidance, not a guarantee of muscle preservation during medication treatment. Start from your abilities; people with chronic conditions should discuss appropriate types and amounts with their clinician. CDC activity guidance.
Walking and resistance work answer different needs. Our post-meal walking evidence review focuses on glucose and timing, not proof of muscle retention. Likewise, probiotic GLP-1 claims do not establish a way to preserve muscle. Avoid treating a benefit in one domain as an answer to another.
Frequently asked questions
Should I stop a GLP-1 medicine because lean mass fell?
A scan alone cannot make that decision. Discuss its meaning, treatment benefits, symptoms, and function with the prescriber. This review does not provide a dose change or stopping rule.
Can lean percentage improve while I lose lean tissue?
Yes. Percentage and absolute amount use different denominators. The arithmetic example above shows how this happens; it is not evidence that every such change is harmless.
Is tirzepatide better than semaglutide for muscle?
The studies compared here do not establish that. Comparing percentages from separate trials is not a randomized head-to-head muscle-health comparison.
Does protein powder solve the problem?
This review does not establish a powder, brand, or universal dose that prevents GLP-1 muscle loss. Ask what your actual intake and nutritional needs are before adding products.
Does a lower lean-mass number mean sarcopenia?
Not by itself. Clinical assessment considers strength, muscle quantity or quality, and performance; interpretation depends on the framework and the person. EWGSOP2 consensus.
Bottom line
Do not minimize a meaningful functional decline, and do not turn every lean-mass change into a diagnosis. The evidence supports asking two questions together: “What tissue measurement changed?” and “What can this person do?” A useful follow-up plan should answer both.
Methods, access limits, and update history
This targeted narrative comparison was prepared October 15, 2026, after checking the site’s 27 published titles for overlap. Focused web/PubMed discovery used “semaglutide lean mass muscle strength resistance exercise trial 2025 2026,” “tirzepatide body composition SURMOUNT-1,” and “LEAN-PREP.” We selected human body-composition and function comparisons, one dietary exercise pilot, a protocol, and methodological/official guidance. This is not a systematic review, exhaustive evidence inventory, independent duplicate extraction, or meta-analysis.
Access: SEMALEAN and SURMOUNT-1 numerical findings were checked in original PubMed abstracts and available figure captions/disclosures. The STEP 1 source is a published conference-supplement abstract. Springer original pages were read for the exercise analysis’s abstract/funding and the pilot’s methods/results/limitations. The SURMOUNT correction text was available through indexed primary publisher/PMC records; direct full-page retrieval failed. LEAN-PREP and supportive-care consensus were used at abstract level; the latter was accessible through its indexed PubMed record. The DXA methodological discussion was assessed through its indexed publisher summary. Failed PMC browser checks were not counted as full-text access. Several endpoint-specific denominators, confidence intervals, supplements, and current registry results remain unverified. We did not pool effects or claim a complete correction-history audit of every source.
Transparency: Prepared with AI assistance. The author works in healthcare and biotechnology. No affiliate purchase links or specific supplement recommendations are included. No independent scientific or medical review is claimed.
Update history: October 15, 2026 — Initial publication; measurement denominators, mass-versus-function findings, a null body-composition result, and protocol-versus-results boundaries documented.
Sources and further reading
- STEP 1 exploratory body-composition abstract, 2021. DOI 10.1210/jendso/bvab048.030.
- Look et al., SURMOUNT-1 DXA substudy, 2025. DOI 10.1111/dom.16275. Figure-axis correction.
- Alissou et al., SEMALEAN, 2026. DOI 10.1111/dom.70141.
- Jensen et al., physical fitness after exercise/liraglutide, 2026. PMID 41579235.
- Binmahfoz et al., home-based resistance pilot, 2025. PMID 40760444.
- Alawadhi et al., LEAN-PREP protocol, 2026. DOI 10.1136/bmjopen-2026-116911.
- 2026 nutritional/lifestyle supportive-care expert consensus; CDC adult activity guidance.
- EWGSOP2 definition/diagnosis consensus; 2026 DXA/functional-assessment discussion.
Written and edited by: Denny Jaeyong Choi (JD), INSEAD MBA
Medical information notice: General education only. This article does not diagnose muscle disease or prescribe medication changes, protein targets, or exercise. Discuss persistent weakness, difficulty eating, or declining daily function with a qualified clinician.