GLP-1 muscle loss is real, and the share is bigger than most people expect. Body scans in the STEP 1 trial found that about 40% of the weight lost on semaglutide was lean mass, not fat (Trial). Across the wider set of trials, the figure runs from about 25% to 39% (Comment). Two things push it down: eating enough protein, and lifting weights two or three times a week.
What Lean Mass Really Means
Lean mass is everything in your body that is not fat and not bone. Muscle is the biggest part of it. The rest is water, your organs, connective tissue, and the sugar stored in muscle, called glycogen. So researchers use lean mass as a stand-in for muscle. It is not a direct measure of it (Comment).
That gap matters most in the first few weeks. When you cut calories hard, glycogen drops, and water leaves with it. So the first lean mass to go is often just fluid.
The best tool for the job is DXA. It is a low-dose X-ray scan that sorts the body into fat, lean tissue and bone. Nearly every number below comes from one.
Bathroom scales that claim to read body fat work differently. They send a tiny current through you and guess from how easily it travels. The method is called bioelectrical impedance. A 2026 review of healthy adults found the average error was small. Single readings, though, were often off by more than six kilograms (Review). Use one to watch a trend if you like. Do not trust it to tell you whether you are losing muscle.
How Big Is GLP-1 Muscle Loss
STEP 1 is the trial that made semaglutide a household name. Inside it, 140 people had body scans at the start and again at 68 weeks.
In the semaglutide group, fat fell by about 10 kilograms and lean mass by about 7 (Trial). That puts lean mass at roughly 40% of the tissue lost. A later review sized up those 7 kilograms another way. They were about 13% of the lean mass the group started with (Review).
Tirzepatide looks better still. In the SURMOUNT-1 scans, 160 people were measured at the start and at 72 weeks. In the drug group, body weight fell about 21%, fat about 34% and lean mass about 11%. So about 75% of the weight lost was fat and 25% was lean mass. The group getting dummy injections split their losses the same way (Study).
Combined results point the same way. Across 19 trials in people with type 2 diabetes, users lost more fat and more lean mass than the comparison groups. Their bodies ended up with a similar share of lean mass (Meta-analysis).
Ordinary dieting costs lean tissue too. Researchers looked at 26 groups who lost more than 10 kilograms by diet or behaviour programs. On a low-calorie diet, lean tissue was about 14% of the weight lost. On a very strict low-calorie diet it was about 23% (Systematic review). The harder you cut, the bigger the share.
A Lancet commentary drew the obvious conclusion. The 25% to 39% figure probably tracks how much weight comes off. It is not a sign that the drug attacks muscle in some special way. The authors add that the idea still needs testing (Comment).
So the percentage is normal, but the number of kilograms is not. Losing a fifth of your body weight moves far more tissue than losing a twentieth. We cover the heart and lifespan side in our piece on GLP-1 drugs and longevity.
Why Strength and Bone Matter
Muscle does real work. It carries you up stairs and catches you when you trip. It also pulls sugar out of your blood after meals. That link with staying well for longer is why we gave strength training and longevity its own article.
Bone gets less attention, and it matters here too. In a Danish trial, a year on liraglutide alone lowered bone density at the hip and spine. The group that exercised instead lost almost as much weight and kept its bone density (Trial).
Weaker evidence points the same way. One study searched health records and matched 33,210 GLP-1 users with the same number of non-users. None of them had diabetes. About 3 in every 100 users broke a bone within a year, against 2.6 in every 100 non-users. Among people aged 78 to 88, it was about 9 in every 100 against 5 (Study). That is a link, not proof. The extra risk sat mostly with the oldest and heaviest.
The evidence does cut both ways. A 2024 review of trials in people with type 2 diabetes found better bone density at the spine and hip (Meta-analysis). That result rests on thin ground. Only 16 of its trials measured the spine, and only 5 measured the hip. Most were run in China, and the Western trials did not show the same gain.
Tests of actual strength have been more reassuring. A French study followed 106 people with obesity on semaglutide for a year. Lean mass fell about 3 kilograms by seven months, then held steady. Grip strength went up by around 4 kilograms. The share with sarcopenic obesity, meaning low muscle and high body fat together, fell from 49% to 33% (Study). Nobody was compared against an untreated group. Part of that last drop is also arithmetic. The measure divides muscle by body weight, and weight fell 13%.
The concern is clearest in older people. Researchers followed 220 older adults with type 2 diabetes for two years. Those on semaglutide ended with a weaker grip and a slower walk than similar people not taking it. Higher doses came with more muscle loss (Study). The researchers only watched people rather than assigning treatment, so this cannot prove cause.
So the risk is not spread evenly. A 45-year-old with muscle to spare is in a very different position from an 80-year-old who is already frail.
How Much Protein You Need
Protein is the first fix, and these drugs make it harder to get right. They work partly by turning your appetite down. When you are barely hungry, protein is easy to miss.
A 2025 analysis pooled 29 studies of dieting plus training. More protein went with better lean-mass results all the way up. The researchers found no ceiling, even at 3.2 grams per kilogram of body weight a day. The trend crossed from losing lean mass to holding on to it at around 1.9 grams per kilogram (Systematic review).
Treat that number with care. The authors call their own analysis a first look, because the studies disagreed with each other so much. The people in them were lean and used to training. They were not people with obesity taking one of these drugs.
The trial now testing protein alongside these drugs picked a smaller target: 1.6 grams per kilogram a day (Protocol). For a 90-kilogram person, that works out at about 145 grams. It is a sensible number to aim for.
Protein does far more when training comes with it. In a four-week trial, 40 overweight young men trained six days a week. None had trained before, and all ate about 40% less than they needed. The group eating 2.4 grams per kilogram gained 1.2 kilograms of lean mass. The group eating 1.2 grams simply held theirs steady (Trial).
Without training, protein alone buys much less. Across 24 trials in dieters, higher-protein diets saved about 0.4 kilograms more lean tissue than standard ones (Meta-analysis). The authors call that modest. It is worth having, but it will not carry the job on its own.
Food first is the simplest rule. Eggs, dairy, fish, poultry, meat, beans, lentils and tofu all do the job. Spread it across your meals. That is easier on a small appetite than one big serving at night.
Lift Two or Three Times Weekly
The strongest direct evidence is a single Danish trial. Adults with obesity first lost about 13 kilograms on a strict eight-week diet. A coin flip then sorted 195 of them into four groups for a year: exercise plus a dummy injection, liraglutide alone, both, or neither (Trial).
The group doing both kept off the most weight. Its lead over the drug alone was small enough that chance could explain it. That group also cut body-fat percentage about twice as much as either treatment on its own. Both exercising groups gained lean mass. The group on liraglutide alone did not gain any.
The bone results are the clearest part. On the drug alone, hip and spine density fell. In the group doing both, who lost more weight than anyone, it held (Trial). A follow-up analysis looked at what people could actually do. Next to the drug alone, the group doing both climbed stairs about 9% faster. Their fitness improved as well (Trial).
The exercise was mostly indoor cycling and circuit training, not a barbell program. And liraglutide is an older, weaker drug than semaglutide or tirzepatide. Both facts limit what the trial can tell you.
For lifting on its own, the evidence comes from dieting studies with no drug involved. Researchers pooled 25 trials in people with overweight or obesity. Adding weight training to a diet protected lean mass and increased fat loss (Meta-analysis). Both groups lost about the same total weight. So the scale reads the same, but more of what came off was fat.
Six trials followed older adults with obesity through a calorie cut. Lifting three times a week saved almost all the lean mass that dieting alone cost, about 0.8 kilograms (Meta-analysis). An older review of 52 studies in adults over 50 counted it another way. Among the diet-only groups, 81% lost 15% or more of their weight as lean tissue. Among the diet-plus-exercise groups, 39% did (Systematic review).
The dose is ordinary. Most trials that protected lean mass used three supervised sessions a week, 30 to 60 minutes each (Meta-analysis). Two sessions a week is a real program, and three is better. Cover the big movements: a squat, a hinge, a push, a pull, and something for your middle. Then add weight or extra reps as it gets easier. That habit of adding a little is called progressive overload.
No finished trial has yet tested lifting during semaglutide or tirzepatide. LEAN-PREP is recruiting now, and a coin flip will sort 232 adults into weight training, higher protein, both, or neither (Protocol). It is not due to finish until 2029 (Registry). A second trial pairs supervised exercise with semaglutide, and it should not report before 2028 (Registry).
So this is widely agreed advice, and it carries very little risk. The firm proof is still missing.
What Happens If You Stop
STEP 1 has a sequel that deserves more attention. It followed 327 people for another year after treatment ended. The 228 who had been on semaglutide had lost about 17% of their body weight. A year later, they had put roughly two thirds of it back on (Trial).
Blood pressure climbed back to where it started. Blood sugar and cholesterol drifted back as well, though both stayed better than before the drug. The researchers treat these numbers as a first look, not a firm result.
A 2026 analysis of six trials mapped the pattern. About 60% of the lost weight returns within a year. The model expects it to level off near 75% (Systematic review). What nobody has measured is what kind of weight comes back. No study has checked fat and muscle after people stop one of these drugs.
The concern comes from older weight-loss research. Women past menopause lost weight, regained some, and got back far more fat than lean tissue. For every kilogram of fat they lost, about a quarter of a kilogram of lean went with it. For every kilogram of fat they regained, only about a tenth of a kilogram of lean came back (Study).
That is a leap, not GLP-1 data, and it comes from one small group of women. The Lancet commentary makes the same point about these drugs. It warns that the cycle may be sharper here (Comment).
If the same pattern holds, the risk is easy to see. Stop without protecting your muscle, regain the weight, and the scale can read what it did before. You would be carrying less muscle at that weight. That is why the muscle you keep now matters.
None of this is a reason to start or stop anything. These are prescription medicines, taken under medical supervision. That decision belongs to you and your prescriber.
Frequently Asked Questions
Is GLP-1 muscle loss permanent?
Nobody has measured it yet. No study has tracked fat and muscle after people stop one of these drugs (Systematic review). Muscle can be rebuilt with training and enough protein, which is why the advice here matters. In the wider weight-loss research, though, regained weight tends to come back mostly as fat (Study).
Is eating more protein enough on its own?
Probably not. Across 24 trials, higher-protein diets on their own saved about 0.4 kilograms of extra lean tissue. The authors call that modest (Meta-analysis). Weight training appears to do more. It held on to lean mass and increased fat loss for the same total weight lost (Meta-analysis). Protein works best alongside training.
Does cardio count for GLP-1 muscle loss?
It helps. The Danish trial used mostly cycling and circuit work, and those groups still gained lean mass and kept their bone density (Trial). But the studies built to protect lean mass while dieting almost all used weight training three times a week (Meta-analysis). Do both if you can, and keep the lifting.
Are body-composition scales worth buying?
Not for this job. Home devices often miss lean mass by more than six kilograms in either direction (Review). One tracked a group’s fat loss well over 12 months, yet it still read muscle 1 to 2 kilograms low every time (Study). Track your waist, your strength in the gym and how your clothes fit instead.
Key Takeaways
- Roughly a quarter to 40% is lean mass. STEP 1 put it near 40% for semaglutide (Trial). The tirzepatide scans put it at about 25%, the same split as the dummy injections (Study).
- Lean mass is not all muscle. It includes water, stored sugar, organs and connective tissue. Ordinary dieting also costs 14% to 23% of the weight lost as lean tissue (Systematic review).
- Aim at 1.6 grams of protein per kilogram a day. That is the target chosen by the trial now testing protein in people on semaglutide and tirzepatide (Protocol).
- Lift two or three times a week. In older adults, three sessions a week during a calorie cut saved almost all the lean mass that dieting alone cost (Meta-analysis).
- Exercise protected bone where the drug alone did not. Hip and spine density fell on liraglutide by itself, and held in the group doing both (Trial).
- Regain may come back mostly as fat. Two thirds of the weight returned within a year of stopping semaglutide (Trial), and older research shows regain favouring fat over lean tissue (Study).
Lose Fat, Keep Your Strength
The advice here is ordinary, and that is the good news. Eat enough protein. Lift something heavy two or three times a week. Add a little weight when it starts to feel easy.
None of that depends on taking a GLP-1 drug. It is the same advice for anyone losing weight, and it carries almost no downside. Do it and you arrive at a lighter body with your strength intact.
The trials built to answer this are still recruiting. So treat all of it as the best reading we have today, not a closed case. Protein and lifting have the strongest support behind them, and neither one asks you to wait.
This article is for educational purposes and is not medical advice. Talk to a qualified clinician before changing your health regimen, and never start or stop a prescribed medicine because of something you read online.
Bio-Hacking.Blog 
Leave a comment