In the body composition substudy of SURMOUNT-1, 160 participants had a DXA scan at the start and again at week 72. Of the weight they lost on tirzepatide, roughly 75% was fat mass and 25% was lean mass (PMID 39996356). Lean mass fell 10.9% while fat mass fell 33.9%, against a 21.3% drop in total body weight.
The number that gets left out of most coverage is the placebo arm in that same substudy. It split the same way, about 75% fat and 25% lean. Losing lean tissue alongside fat is what weight loss does, and the drug did not change the ratio. What it changed was how much total weight came off.
- SURMOUNT-1 DXA substudy, tirzepatide, 72 weeks: fat mass -33.9%, lean mass -10.9%, total weight -21.3%.
- About a quarter of the weight lost was lean mass on tirzepatide and about a quarter on placebo, so the proportion is a feature of weight loss rather than of the drug.
- DXA lean mass is not the same thing as muscle. It includes water, glycogen and organ tissue, all of which fall as body size falls.
- In a real world 12 month cohort on semaglutide, skeletal muscle mass fell from 29.9 kg to 28.7 kg while absolute handgrip strength was preserved.
- The only strategy with randomised evidence behind it is adding exercise, which roughly doubled the fall in body fat percentage compared with the drug alone in a one year trial.
What the SURMOUNT-1 body composition substudy actually measured
SURMOUNT-1 randomly assigned 2539 adults with obesity or overweight and no diabetes to tirzepatide 5 mg, 10 mg, 15 mg or placebo for 72 weeks (PMID 35658024). A substudy of 160 of those participants, 124 on pooled tirzepatide doses and 36 on placebo, had dual energy X-ray absorptiometry at baseline and at week 72. They were 73% female, with a mean weight of 102.5 kg and a mean BMI of 38.0.
DXA separates body mass into fat, lean soft tissue and bone, which is why a substudy like this is worth more than a bathroom scale reading. Here is what it recorded over the 72 weeks.
| Measure | Tirzepatide (pooled doses) | Placebo |
|---|---|---|
| Total body weight | -21.3% | -5.3% |
| Fat mass | -33.9% | -8.2% |
| Lean mass | -10.9% | -2.6% |
| Share of lost weight that was fat | About 75% | About 75% |
| Share of lost weight that was lean | About 25% | About 25% |
All three comparisons against placebo reached p below 0.001. The proportions held across the substudy subgroups the authors looked at, which were sex, age band and tertile of total weight lost. Someone who lost more than 25.9 kg lost more lean mass in kilograms than someone who lost under 15.3 kg, but not a different share of it.
Lean mass is not the same thing as muscle
This is where most of the alarm around the topic comes from. DXA lean mass, sometimes reported as fat free mass, is everything that is not fat or bone mineral. That includes skeletal muscle, but also intracellular and extracellular water, glycogen and the mass of organs such as the liver and kidneys.
Several of those shrink as body size falls, for reasons that have nothing to do with muscle wasting. Glycogen carries water with it, so a drop in intake lowers both. A smaller body needs less blood volume. A liver that was enlarged by fat gets smaller. A 10.9% fall in DXA lean mass therefore sets an upper bound on how much muscle was lost, not a measurement of it.
The Zepbound label puts the same point in one sentence in its pharmacodynamics section: tirzepatide lowers body weight with greater fat mass loss than lean mass loss. That is the direction the substudy confirmed, and the 75 to 25 split is what greater means in practice.
What happened to strength in a real world cohort
A retrospective study at the obesity unit of Hospital Universitari Vall d Hebron in Barcelona followed 44 adults with obesity and no diabetes on semaglutide alongside a lifestyle programme, and 39 on the lifestyle programme alone, for 12 months (PMID 42640861). Body composition was measured by bioelectrical impedance rather than DXA, which is a real limitation, and the design was observational.
The semaglutide group lost 11.0 kg more than the lifestyle group. Skeletal muscle mass fell from 29.9 kg to 28.7 kg and fat free mass from 53.8 kg to 52.7 kg, both statistically significant. So the absolute lean tissue loss was there, at about 1.2 kg of skeletal muscle over a year.
Function told a different story from mass. Absolute handgrip strength was preserved in both groups, and handgrip strength relative to body weight improved in the semaglutide group. A reference based skeletal muscle index, standardised for age and BMI, also improved more in the semaglutide group than the lifestyle group. Less muscle on a much smaller body was still enough muscle to do the same work.
How large the loss is next to other things that shrink muscle
A 2024 narrative review in Diabetes Care gathered the body composition findings across liraglutide, semaglutide, tirzepatide and retatrutide and put the typical lean mass loss at around 10%, or roughly 6 kg (PMID 38687506). Its comparison is the memorable part: that is similar to what a decade or more of ageing does.
Read that carefully in both directions. It is a real amount of tissue, and it is also the price of losing 15% to 20% of body weight by any means, including surgery and dieting. The review is a narrative synthesis rather than a pooled analysis, so treat the 10% as a central tendency across trials rather than a measured figure from one.
The intervention with randomised evidence behind it
One trial has tested adding exercise to a GLP-1 head to head. After an eight week low calorie diet in which 195 adults with obesity lost a mean of 13.1 kg, participants were randomly assigned for one year to supervised moderate to vigorous exercise plus placebo, liraglutide 3.0 mg daily plus usual activity, both combined, or neither (PMID 33951361).
| One year strategy | Weight change vs placebo | Change in body fat percentage |
|---|---|---|
| Exercise plus placebo | -4.1 kg | -1.7 points |
| Liraglutide 3.0 mg plus usual activity | -6.8 kg | -1.9 points |
| Exercise plus liraglutide 3.0 mg | -9.5 kg | -3.9 points |
The combination cut body fat percentage by about twice as much as either part alone, and it was the only arm that improved HbA1c, insulin sensitivity and cardiorespiratory fitness. Two caveats matter: the drug was liraglutide 3.0 mg daily, not a weekly semaglutide or tirzepatide dose, and the exercise programme was supervised rather than self directed.
The Diabetes Care review above argues specifically for resistance training on top of that, noting that supervised resistance programmes running longer than 10 weeks have produced gains of around 3 kg of lean mass and about 25% in strength. Those figures come from exercise trials in people not taking these drugs, so they describe what resistance training can do rather than what it has been shown to do during incretin treatment.
Where intake fits
Every trial quoted here ran its drug on top of a prescribed deficit of roughly 500 kcal a day, not on top of eating as little as appetite allows. Because these drugs cut how much food you want, the practical risk is landing far below that deficit without meaning to, which is the condition under which lean tissue loss gets worse. We go through the arithmetic in how many calories to eat on a GLP-1, and you can get a starting estimate for your own numbers from the GLP-1 weight loss calculator.
If you want the trial by trial weight loss figures behind these body composition numbers, the tirzepatide ones are collected in how much weight you can lose on Zepbound.
FAQ
Does a GLP-1 cause muscle loss?
It causes lean mass loss, which is not quite the same claim. In the SURMOUNT-1 DXA substudy, lean mass fell 10.9% over 72 weeks on tirzepatide while fat mass fell 33.9%. DXA lean mass includes water, glycogen and organ tissue as well as skeletal muscle, so that figure is an upper bound on muscle loss rather than a measurement of it.
How much of the weight you lose on a GLP-1 is muscle?
In the SURMOUNT-1 substudy about 75% of the weight lost was fat mass and 25% was lean mass, on tirzepatide and on placebo alike. The share stayed close to that across sex, age band and how much total weight participants lost.
Is lean mass loss on these drugs worse than losing weight by dieting?
The substudy that can answer this found the same 75 to 25 split in its placebo arm, where participants lost 5.3% of body weight with lifestyle support alone. The drug arm lost far more total weight, so more lean mass in kilograms, but not a different proportion of it.
Does losing lean mass make you weaker?
Not necessarily over a year. In a 12 month real world cohort on semaglutide, skeletal muscle mass fell by about 1.2 kg while absolute handgrip strength was preserved and strength relative to body weight improved. That is one observational study using impedance rather than DXA, so it is suggestive rather than conclusive.
Does resistance training prevent it?
No trial has tested resistance training against no resistance training during semaglutide or tirzepatide treatment. The closest evidence is a one year randomised trial in which supervised aerobic exercise plus liraglutide 3.0 mg cut body fat percentage by 3.9 points, about twice what either exercise or the drug achieved alone.
Do you regain lean mass if you stop?
The withdrawal trials measured body weight rather than body composition, so there is no clean answer from them. What they do show is that most of the lost weight comes back, which is covered in what happens when you stop a GLP-1.
Sources
- Look M, Dunn JP, Kushner RF, et al. Body composition changes during weight reduction with tirzepatide in the SURMOUNT-1 study of adults with obesity or overweight. *Diabetes Obes Metab*. 2025;27(5):2720-2729. PMID: 39996356
- Jastreboff AM, Aronne LJ, Ahmad NN, et al. Tirzepatide Once Weekly for the Treatment of Obesity (SURMOUNT-1). *N Engl J Med*. 2022;387(3):205-216. PMID: 35658024
- Filippi-Arriaga F, Comas M, Prats A, et al. Longitudinal Changes in Body Composition, Adaptive Thermogenesis, and Muscle Strength in Patients with Obesity Treated with Semaglutide. *Obes Facts*. 2026. PMID: 42640861
- Locatelli JC, Costa JG, Haynes A, et al. Incretin-Based Weight Loss Pharmacotherapy: Can Resistance Exercise Optimize Changes in Body Composition? *Diabetes Care*. 2024;47(10):1718-1730. PMID: 38687506
- Lundgren JR, Janus C, Jensen SBK, et al. Healthy Weight Loss Maintenance with Exercise, Liraglutide, or Both Combined. *N Engl J Med*. 2021;384(18):1719-1730. PMID: 33951361
- Wilding JPH, Batterham RL, Calanna S, et al. Once-Weekly Semaglutide in Adults with Overweight or Obesity (STEP 1). *N Engl J Med*. 2021;384(11):989-1002. PMID: 33567185
- ZEPBOUND (tirzepatide) prescribing information, sections 6.1 Adverse Reactions, 12.2 Pharmacodynamics and 14.1 Clinical Studies. Eli Lilly and Company, via DailyMed.
This article is for information only and is not medical advice. Dosing, suitability and any change to treatment are decisions for your prescriber.