Illustration note: The featured exercise image and consultation scene are editorial illustrations, not study participants or proof that a workout prevents medication-related muscle loss.
A semaglutide muscle loss study needs a label: who was studied, what was measured and what the comparison was. A change in estimated lean tissue, the size of one muscle and performance on a strength test answer different questions.
Selected studies have found reductions in lean mass or measured muscle size during semaglutide treatment. Their findings do not establish that every patient loses the same tissue, becomes weaker or can prevent those changes with one workout.
The concern raised in the original CBS19 report is worth examining through those distinctions. This guide compares human observations, an animal experiment and a planned resistance-exercise trial, then shows how to check what a headline actually supports.
A study passport for the muscle-loss question
Start with the population and study design before reading a percentage. The following is a selected evidence map, not a systematic review of every semaglutide study.
| Study | Population and comparison | Measurement | What the design can support |
|---|---|---|---|
| STEP 1 exploratory body-composition analysis | 140 adults without diabetes in a semaglutide-versus-placebo trial. | DXA estimates of body composition over 68 weeks. | Body-composition changes in this subset; no randomized resistance-training comparison. |
| SEMALEAN | Adults with obesity receiving semaglutide; no untreated control group. | DXA lean mass and handgrip at baseline, seven months and 12 months. | Changes observed during treatment, with limited ability to separate medication effects from other changes. |
| SLIM LIVER muscle analysis | People with HIV and metabolic dysfunction-associated steatotic liver disease; single treatment arm. | Psoas MRI volume, chair-rise time and walking speed over 24 weeks. | A specific muscle-size finding alongside function tests in a specific clinical population. |
| Karasawa and colleagues’ mouse experiment | Mice, not patients. | Lean mass, skeletal muscles and force-generating capacity. | Preclinical evidence about measurements and mechanisms, requiring human validation. |
| LEAN-PREP | Planned randomized comparison of four exercise/protein-support groups during semaglutide or tirzepatide therapy. | Thigh MRI plus body composition, strength and function. | A published protocol and registered question, not a reported treatment benefit. |
A drug name alone cannot make these studies interchangeable. They differ in health conditions, treatment context, follow-up and the question their design can answer.
The human MRI study measured one muscle, alongside function
The SLIM LIVER analysis had muscle measures for 46 people with HIV and MASLD over 24 weeks. Mean psoas volume fell 9.3% (95% confidence interval: 5.2% to 13.4% reduction), not whole-body muscle loss.
Nonsignificant average chair-rise and walking-speed changes do not establish equivalence or exclude meaningful change. Without a randomized untreated comparator, direct drug causation and generalization remain limited.
SEMALEAN shows why the measurement and time point matter
SEMALEAN enrolled 115 adults with obesity; 106 completed follow-up, with semaglutide targeted at 2.4 mg. Lean mass fell initially, then stabilized, while handgrip improved.
There was no control group, and early discontinuers before seven months were excluded. The retained cohort cannot establish muscle protection: DXA lean tissue is not all skeletal muscle, grip is not every muscle, and later stabilization is not absence of early loss.
The mouse result is a reason to ask better human questions
A 2025 Cell Metabolism letter examined semaglutide, skeletal muscle and force-generating capacity in mice. Its findings cannot be converted into a percentage forecast for people taking Ozempic or Wegovy.
The University of Utah’s explanation describes changes that differed between muscles, including weaker force generation in some muscles without a matching decrease in their size. It explicitly cautions against drawing a one-to-one parallel between mice and people.
That distinction prevents two opposite mistakes: assuming smaller lean mass always means the same loss of strength, or assuming a relatively stable muscle size guarantees unchanged strength. Human studies need to measure the outcomes they want to claim.
Animal experiments can help develop those questions. They do not establish a patient’s diagnosis, a safe personal training load or an exercise program that eliminates risk.
Check the denominator and the endpoint before sharing a number
The STEP 1 exploratory DXA report found lower absolute lean mass during semaglutide treatment while lean tissue became a larger share of body weight. Its body-composition subset did not test a particular resistance-training program.
Our separate muscle-gain and measurement guide covers those DXA findings in more detail. Here, the useful next step is checking whether the headline keeps its original denominator and endpoint.
| Claim you encounter | Detail to locate in the source | Why it changes the meaning |
|---|---|---|
| “A percentage of weight lost was lean mass.” | Is the denominator total weight lost, starting lean mass or starting body weight? | These fractions describe different quantities and cannot be substituted for one another. |
| “Muscle was preserved.” | Was the outcome DXA lean mass, a particular muscle on MRI, grip strength or a functional task? | One favorable measure does not establish all the others. |
| “Exercise prevented the loss.” | Was exercise actually randomized, and what did its comparison group receive? | A cohort observed during medication treatment cannot isolate the effect of a training program. |
| “Function did not change.” | Was this a nonsignificant result, and how precise was the estimate? | Failure to detect a difference does not prove that all meaningful changes were ruled out. |
| “A trial proves the plan works.” | Is the document a protocol, registry entry or completed results paper? | Publishing a plan establishes methods, not the outcome. |
Where resistance-training evidence fits
Resistance exercise remains relevant to the conversation, but the evidence needs its original population attached. A 2017 randomized trial led by Dennis Villareal studied weight management with different exercise types in older adults with obesity, without GLP-1 medication.
Resistance training preserved more measured lean mass than aerobic training alone during that intervention. This supports a discussion about training during weight loss; it does not quantify protection during semaglutide treatment.
In this six-month trial, 160 adults were randomized and 141 completed; approximate mean lean-mass reductions were 5% with aerobic training, 2% with resistance training and 3% with both. All training groups received weight management, combined training also improved function, and controls had neither program; lean mass is not identical to muscle.
Our exercise-on-Ozempic guide explains how general activity guidance and clinical recommendations differ. A generic sets-and-repetitions target should not be presented as a medication-specific guarantee.
LEAN-PREP separates the exercise question from the medication question
The 2026 LEAN-PREP protocol plans a six-month randomized study of 232 adults at the Dasman Diabetes Institute. Its four groups are control, resistance exercise, protein support, and both exercise and protein support during semaglutide or tirzepatide therapy.
The primary outcome is MRI-measured quadriceps cross-sectional area, a measure of muscle size at the thigh. DXA body composition, strength and physical-function measures are additional outcomes.
When results appear, check the enrolled population, actual participation and comparison groups beside the assigned intervention. A protocol explains what will be tested; it does not show that the support worked.
On 27 October 2026, the official registry displayed “Recruiting” and “No Results Posted,” with its last update posted on 21 April 2026. The record listed estimated primary completion in August 2029 and study completion in December 2029.
Six months describes each participant’s planned follow-up, not a promise that the entire study ends six months after the protocol appears. Estimated completion dates can also change.
The registered intervention includes initial supervised exercise sessions and recurring support. Copying its exercise frequency or protein amount into an unsupervised personal plan would leave out assessment, selection and follow-up.
Use the evidence to make the care conversation more specific
When a new study appears, copy five details beside the headline: population, comparison, measurement, follow-up and whether results were actually reported. That short record makes it harder to confuse a mouse experiment, an observation and a randomized treatment test.
For your own care, describe changes you have noticed in everyday strength, activity or ability to eat adequately to your prescribing team. Ask which outcomes matter in your situation and how they will be reviewed.

A published average cannot decide whether you should change medication, diet or exercise. The useful takeaway is a more accurate interpretation of the evidence and a clearer question for the professional who knows your health history.
Get the next evidence check.
Five practical fitness stories worth your time, every Friday.
One email a week. No product pitches. Unsubscribe in one click.
References
Alissou, M., Demangeat, T., Folope, V., et al. (2026; online 2025). Impact of semaglutide on fat mass, lean mass and muscle function in patients with obesity: The SEMALEAN study. Diabetes, Obesity and Metabolism, 28(1), 112 to 121. Full text.
Ditzenberger, G. L., Lake, J. E., Kitch, D. W., et al. (2025; online 2024). Effects of semaglutide on muscle structure and function in the SLIM LIVER study. Clinical Infectious Diseases, 80(2), 389 to 396. Primary abstract.
Karasawa, T., Choi, R. H., Meza, C. A., et al. (2025). Unexpected effects of semaglutide on skeletal muscle mass and force-generating capacity in mice. Cell Metabolism, 37(8), 1619 to 1620. Primary abstract.
University of Utah Health. (2025, August 5). New study raises questions about how Ozempic affects muscle size and strength. Institutional explanation of the mouse experiment cited above.
Wilding, J. P. H., Batterham, R. L., Calanna, S., et al. (2021). Impact of semaglutide on body composition in adults with overweight or obesity: Exploratory analysis of the STEP 1 study. Journal of the Endocrine Society, 5(Supplement 1), A16 to A17. Conference abstract.
Villareal, D. T., Aguirre, L., Gurney, A. B., et al. (2017). Aerobic or resistance exercise, or both, in dieting obese older adults. New England Journal of Medicine, 376(20), 1943 to 1955. Primary abstract.
Alawadhi, A. A., Alroudhan, D., Alsaeed, D. J., et al. (2026). LEAN mass Preservation with Resistance Exercise and Protein during semaglutide and tirzepatide therapy (LEAN-PREP study): a protocol for a randomised controlled trial. BMJ Open, 16(4), e116911. Primary abstract.
ClinicalTrials.gov. (2026). NCT06885736, LEAN-PREP. Study record, accessed 27 October 2026; last update posted 21 April 2026.
Watch the original report
The original CBS19 segment supplied the news context for this article. Its reporting is separate from the primary studies above and does not establish a proven muscle-preservation program.
Watch the original report on YouTube if the embedded player does not load.
Medical disclaimer: This article provides general information about weight-management research, not personal medical, medication, nutrition or exercise advice. Discuss treatment decisions and changes in strength, function or food intake with an appropriately qualified healthcare professional who knows your health history.





