This week the headlines said GLP-1s eat your muscle. Now there’s a study that says: mostly they don’t, if you do two boring things. Both can be true, and the gap between them is the most useful thing you’ll read about these drugs this week.
This article summarises new research and is not medical advice. It does not give dosing or treatment guidance. Talk to your prescriber or dietitian about decisions specific to you.
The change came from one paper. In Cell Reports Medicine this June, researchers pulled together four preclinical experiments and a small proof-of-concept clinical arm, and asked a sharper question than “do you lose muscle.” They asked what kind of weight comes off. The answer: overwhelmingly fat. Absolute muscle mass dipped a little, the way it does in almost any real weight loss. But measured against a now-lighter body, relative muscle mass and strength held steady or nudged up. The drug strips fat and hands you back a higher proportion of the muscle you walked in with. The wasting headline doesn’t survive that.
So why did this week’s news feel like a warning siren? Because the 16 June work — Stanford’s mouse study on slowed muscle repair, and the ENDO tracker data showing people quietly moving less — described what happens when nothing intervenes. Drug plus drift. This week’s data describes the other branch: drug plus protein plus lifting. Same medication, two completely different outcomes for your legs, and the only variable is what you do in the weeks the fat is melting.
The number worth memorising came out of ADA 2026’s “quality of weight loss” sessions, where the muscle question got its own slot for the first time. The figure clinicians kept circling: about 1.2 to 1.6 grams of protein per kilogram of body weight per day, plus resistance training two or so times a week. Put a body on that. An 80 kg person is looking at roughly 96 to 128 grams of protein a day. A 65 kg person, about 78 to 104. That’s a lot more than most people eat once a GLP-1 has quietly halved their appetite — which is the whole problem.
Because protein is the most filling macronutrient, and the drug’s entire job is to make you feel full sooner, protein is usually the first casualty. You sit down hungry, eat a third of the plate, feel done, and the third you ate skewed toward the easy carbs. Repeat that for three months while the scale hands you a win every Monday, and you’ve quietly under-eaten the exact nutrient your muscle needed to survive the cut. Nobody decides to do this. It’s just what an appetite-suppressant does to a normal day if you’re not watching the one number that matters.
Resistance training is the other half, and the data is blunt about it: skip the lifting and the protein alone won’t hold the line. Lifting is the signal that tells your body to keep the muscle rather than recycle it. Two sessions a week clears most of the benefit in the studies. You don’t need a barbell or a programme with a name. You need to make muscle worth keeping, repeatedly, while the weight comes off.
None of this is a reason to fear your medication, or to stop it. The drug is doing the hard part. The research is narrower and more practical than the scare headlines, and its point is plain: the quality of your weight loss is something you steer. The steering wheel is two numbers you can see, protein on the plate and weight on the bar, not the single number on the scale that can’t tell muscle from fat.
That’s the reason protein tracking sits next to your dose in Traqr rather than buried in a food diary. The scale already told you the good news. Whether that good news cost you strength or kept it is a different question, and it’s the one you can still do something about.
Sources
- Cell Reports Medicine — GLP-1 receptor agonist weight loss is predominantly fat-mass; relative lean mass and strength preserved (4 preclinical models + proof-of-concept clinical arm), June 2026.
- American Diabetes Association, 86th Scientific Sessions (ADA 2026, 5–8 June, Chicago) — “quality of weight loss” sessions: protein target ~1.2–1.6 g/kg/day + resistance training for lean-mass preservation.
- Companion reporting: Stanford Medicine (muscle repair in mice, PNAS, 2 June 2026) and ENDO 2026 activity-decline tracker data (Maharjan et al.) — see our 16 June piece, “Losing muscle, moving less.”