The 30-second summary
- Roughly a third to 40% of the weight lost on a GLP-1 can be lean mass, not fat.
- A protein floor of 1.2 to 1.6 grams per kilogram of goal body weight, most women land near 120 grams a day, is the working target to preserve muscle.
- Resistance training twice a week multiplies the effect of protein. Cardio alone does not.
The thing your scale won't tell you
The scale is a liar of omission. It tells you the weight is leaving. It does not tell you what kind.
This article is about one number in isolation. If you want the whole picture of how to eat on this medication, protein included, start with the GLP-1 diet guide and come back here for the detail.
In the STEP 1 trial of semaglutide, the registration study that put Wegovy on the map, participants lost an average of 14.9% of their starting body weight over 68 weeks. In the subset who had DXA body-composition scans, roughly 40% of that weight was lean mass. The tirzepatide SURMOUNT-1 body-composition substudy came out lower, at roughly a quarter of the weight lost, and found the proportion was no worse than in the placebo group, which is a useful corrective to the idea that these drugs are uniquely hard on lean tissue.
One distinction matters before you read another sentence about this. Lean mass is not the same thing as muscle. On a DXA scan, lean mass is everything that is not fat and not bone: skeletal muscle, but also body water, glycogen and organ tissue. Some of the early drop is water and glycogen leaving, which is expected and not a loss worth worrying about. Skeletal muscle is a part of the total, and the trials that measured it directly found smaller losses than the lean-mass figure implies. The proportion of lean loss that is muscle also falls as the deficit gets less aggressive. So 40% is the outer bound of a wider number, not a measurement of how much muscle you are losing. It is still worth acting on, because the part that is muscle is the part you do not get back easily.
This is the uncomfortable math that most weight-loss apps built before the GLP-1 era are not equipped to handle. They congratulate you for the number going down. They do not ask what you lost.
Why muscle loss happens on a GLP-1
The mechanism is not mysterious. It is the mechanism of every aggressive weight-loss protocol, with one amplifier.
When you are in a meaningful caloric deficit, your body meets its energy needs partly from fat and partly from protein. Which ratio depends on three things: how aggressive the deficit is, how much protein you are eating, and whether the muscles are being asked to do anything.
GLP-1s make the deficit aggressive and near-automatic. They reduce appetite so effectively that many people find themselves eating 30 to 50% less than they used to, without any conscious effort to restrict. If the protein proportion of that smaller intake is not deliberately high, the math is unforgiving. A woman who was eating 2,000 calories and 70 grams of protein before the medication, and is now eating 1,300 calories and 50 grams of protein, has cut her protein below any reasonable preservation threshold.
The body does what any well-run biological system does under scarcity: it trims what is not in active use. And for most people, most muscle is not in active use at a given hour of the day.
What the research actually supports
There is no single perfect trial in GLP-1 users that prescribes an exact protein number. What there is, is a convergent body of evidence from adjacent populations.
A systematic review of weight-loss interventions in older adults concluded that higher-protein diets during caloric restriction consistently preserved more lean mass than standard-protein diets, with effects visible in 12- to 24-week timeframes. (Kim et al., Nutrition Reviews, 2016)
A meta-analysis of resistance-training plus protein interventions in caloric deficit found that combining the two, training and elevated protein, produced substantially better lean-mass retention than either alone. (Cermak et al., American Journal of Clinical Nutrition, 2012)
In trials of very-low-calorie diets, protein intakes of at least 1.2 grams per kilogram of body weight per day were associated with significantly less lean-mass loss than intakes below that threshold.
Read across those, and 1.2 to 1.6 grams of protein per kilogram of goal body weight per day is a defensible working range for someone losing weight quickly on a GLP-1, with the upper end for people who are strength training or younger and the lower end for sedentary older adults. That range is our synthesis and it is worth being precise about where it does and does not come from. The International Society of Sports Nutrition position stand recommends 1.4 to 2.0 g/kg of actual body weight for exercising people, and notes that 2.3 to 3.1 g/kg may be needed to hold lean mass in a deficit in resistance-trained subjects. Those are higher numbers than ours, on a different basis, for a more athletic population. We work lower and against goal weight because most people reading this are not resistance-trained athletes, because goal weight avoids inflating the figure with body fat, and because a target nobody can eat on a suppressed appetite is not a useful target.
For a woman whose goal weight is 75 kilograms (165 pounds), that math looks like 90 to 120 grams of protein per day. Most people on a GLP-1 land near the upper end of that range, for two reasons: they are losing weight faster than the populations the guidelines were written for, and they want a margin of safety.
What 120 grams looks like
The number sounds abstract until you translate it.
- 3 ounces of chicken breast: 26 grams
- 1 cup of Greek yogurt (plain, 2%): 20 grams
- 1 scoop of whey protein in water: 24 grams
- 2 large eggs: 12 grams
- ½ cup of cottage cheese: 13 grams
- 4 ounces of salmon: 25 grams
- 1 cup of cooked lentils: 18 grams
Hit any four of those in a day and you are past 80 grams. Add a protein shake and a dinner and you are at 120. On a GLP-1, the challenge is rarely wanting that much food, it is making room for it in a shrunken appetite. The practical answer is protein first on the plate, then fiber, then the rest.
Resistance training multiplies the effect
Protein without training is a flooring, not a ceiling. The studies that show the strongest lean-mass preservation combine protein with two or three sessions per week of resistance work.
The work does not have to be dramatic. Bodyweight squats, push-ups, rows with a resistance band, a 30-minute strength circuit twice a week, the stimulus is the point, not the intensity. Muscle that is used is muscle that the body is reluctant to give up.
Cardio alone does not have the same effect. Walking, while enormously valuable for cardiovascular health and mood, does not create a strong signal for lean-mass retention. A well-designed GLP-1 regimen, for most people, looks like: daily walking, twice-weekly resistance work, and a protein floor held every day regardless of appetite.
If you eat vegetarian or vegan, the same floor applies with a little more planning. See vegetarian or vegan on a GLP-1.
What to do with this
The evidence sets the range; your body sets the number. The calculator works out your protein floor from adjusted body weight, raises it with your dose because higher doses cost more lean mass, and puts protein first in every meal of the three day plan, so it is the thing you eat before you fill up rather than the thing you run out of room for.
For meals that show their protein rather than promise it, the recipe collection calculates every dish from its ingredients: the tarragon chicken with white beans and the Dijon-dill salmon are two places to start.
Sources
- Wilding JPH, Batterham RL, Calanna S, et al. Once-Weekly Semaglutide in Adults with Overweight or Obesity. NEJM 2021; 384:989-1002. NEJM
- Jastreboff AM, Aronne LJ, Ahmad NN, et al. Tirzepatide Once Weekly for the Treatment of Obesity. NEJM 2022; 387:205-216. NEJM
- Kim JE, O'Connor LE, Sands LP, et al. Effects of dietary protein intake on body composition changes after weight loss in older adults: a systematic review and meta-analysis. Nutrition Reviews 2016. PubMed
- Cermak NM, Res PT, de Groot LC, et al. Protein supplementation augments the adaptive response of skeletal muscle to resistance-type exercise training: a meta-analysis. AJCN 2012. PubMed
- Jäger R, Kerksick CM, Campbell BI, et al. ISSN Position Stand: Protein and Exercise. JISSN 2017;14:20. PubMed
Medical disclaimer: Articles in the foodose research library are educational, not medical advice. Do not change your dose, diet, or training program based on anything you read here without talking to your prescriber or a registered dietitian. See our full medical disclaimer.