What GLP-1s Actually Do to Muscle

Man doing barbell back squats in a gym

The state of the science, part one: GLP-1 receptor agonists and body composition

By Rachele Pojednic, PhD, Chief Science Officer, Kíral

The wins for GLP-1s go well beyond the number on the scale.

GLP-1 receptor agonists mimic a gut hormone your body already makes. They slow gastric emptying, blunt appetite signaling, and improve insulin response. The weight loss that follows can be substantial. But weight and fat are not the same thing, and the question that has dominated the field for the past two years is what happens to everything that isn’t fat.

A meta-analysis published in the International Journal of Obesity on June 19, 2026 is the cleanest answer we have so far. Laverde and colleagues pooled seven randomized controlled trials covering 821 adults with obesity treated with liraglutide (3 mg), semaglutide (2.4 mg), or tirzepatide (5 to 15 mg) against placebo. What they found sounds contradictory until you look at the denominators.

Two true things at once

Lean mass as a proportion of total body weight went up by 1.81% (95% CI: 1.10 to 2.52). Absolute lean mass went down by 1.74 kg (95% CI: −3.04 to −0.45). Both findings were statistically significant.

People on these medications lost real lean tissue, and they lost fat faster, so what remained was a leaner body overall.

Body composition improved. Some lean tissue still left with the fat.

The semaglutide subgroup showed the sharpest version of this. Absolute lean mass dropped 5.44 kg (95% CI: −7.07 to −3.81), the largest of any drug pooled in the analysis, while lean mass as a share of total weight rose 2.95%. Bigger total weight loss, bigger absolute lean loss, better proportional composition.

Two caveats deserve as much attention as the headline numbers.

Heterogeneity. The proportional finding was remarkably consistent across trials (I² = 7%). The absolute lean mass findings were not (I² = 98%). An I² near 98% means the trials disagreed with each other almost completely on magnitude. Study duration, drug, dose, scan method, and population all varied. So 1.74 kg is a pooled midpoint across studies that do not really tell the same story, not a number anyone should expect to see on their own scan.

Function was not measured. This is the limitation the authors name directly: none of the seven included trials measured muscle strength or physical function. Not grip strength, not chair stand, not gait speed. “Lean mass” on a DXA or bioimpedance scan includes muscle, organs, connective tissue, and a great deal of water. Losing some of it is not automatically the same as getting weaker, and we do not currently have the trial data to say how much functional capacity, if any, changes. That is a real gap in the literature, not a rhetorical hedge.

The authors’ own conclusion is worth quoting: lean mass loss “should not be considered a limitation for the use of these drugs,” but it is “essential to accompany drug treatment with nutritional and physical exercise interventions to preserve or improve muscle mass.”

What we found outside the trial setting

Randomized trials are tightly controlled environments. Most people taking a GLP-1 are not in one. In 2025 our team published a retrospective analysis in Diabetes, Obesity and Metabolism looking at what body composition does in a real commercial care setting, where adherence, dosing, and behavior are all messier.

We analyzed 94 adults (81 women, 13 men; mean age 46.6; mean BMI 32.5) prescribed once-weekly compounded semaglutide/cyanocobalamin, titrated from 0.25 mg up to a maximum of 2.4 mg. Bioelectrical impedance and bloodwork were collected at baseline and at three months.

At three months, mean weight loss was 4.11 kg, or 4.57% of body weight (p < 0.0001). Nearly half the cohort (44.7%) lost at least 5%.

The composition breakdown:

  1. Total fat mass
    −2.67 kg
  2. Trunk fat mass
    −1.10 kg
  3. Lean body mass
    −1.43 kg
  4. Skeletal muscle mass
    −0.88 kg
  5. Percent body fat
    −1.18%

Lean mass accounted for roughly 34.9% of total weight lost, and skeletal muscle about 21.5%. For comparison, in the STEP 1 body composition substudy, close to 40% of weight loss came from lean body mass. As proportions of total body weight, both lean body mass and skeletal muscle mass increased significantly (p < 0.0001), the same directional finding the 2026 meta-analysis reports.

I want to be careful about what this does and does not show. Three months of treatment produces less total weight loss than a 68-week trial, and less total weight loss means less absolute lean loss. The comparison is not apples to apples. There was no control group, and bioimpedance is not DXA. What the data do support is narrower and still useful: the proportional improvement in body composition that shows up in controlled trials also shows up in ordinary clinical practice, at a dose range and timeline most people actually experience.

Human touch as part of the model

Buried in our methods section is a detail that matters more than any single effect size. Every participant met with a nurse practitioner via telehealth every four weeks. Those visits were not just prescription refills. They covered clinical assessment, side effect management, dose titration, and direct coaching on diet and exercise. Inclusion was based on body composition (measured percent body fat, visceral fat level, metabolic markers) rather than BMI alone. Body composition and a full metabolic panel were measured at baseline and again at three months.

That is a meaningfully different level of contact than a fifteen-minute annual physical, and different from a prescription arriving in the mail with a titration schedule and no one to call.

I am not going to claim our coaching cadence caused the composition results. The study was not designed to test that, and it cannot. But the two independent findings sitting in front of the field right now are these. The strongest evidence-based recommendation for protecting muscle on a GLP-1 is resistance training plus adequate protein. And the most common way these drugs are prescribed involves almost no structured support for either one. Whatever the effect size of solving that turns out to be, the gap itself is not in dispute.

What the evidence currently supports

For anyone on or considering a GLP-1, the research points in three directions.

  1. Load your muscles
    Resistance training at least twice weekly. Free weights, machines, bands, and body weight all produce the stimulus. This is the intervention with the strongest evidence base for attenuating lean mass loss during energy restriction, and it is the one the meta-analysis authors specifically call for.
  2. Eat enough protein
    A suppressed appetite makes chronic underconsumption very easy, and protein is usually the first thing to go. Reduced intake is the mechanism, not the drug itself.
  3. Track function, not just mass
    Stairs, groceries, getting off the floor, how a set feels at a weight you know. Scans tell you about tissue. Function tells you what the tissue can do, and it is the outcome the trials have not yet measured.

Next in this series: what we know and do not know about tirzepatide’s distinct mechanism, and why dual agonism changes the body composition question.

This content is for educational purposes only and is not medical advice, diagnosis, or treatment. Discuss any medication, nutrition, or exercise decisions with your own licensed clinician.

References

1. Laverde LP, Muñoz-Velandia OM, Alfonso D, Gómez Medina AM. Effect of GLP-1 receptor agonists at doses for obesity management on muscle health: systematic review and meta-analysis of randomized controlled trials (RCTs). Int J Obes. 2026. doi:10.1038/s41366-026-02118-y

2. Chun E, Siojo A, Rivera D, Reyna K, Legere H, Joseph R, Pojednic R. Weight loss and body composition after compounded semaglutide treatment in a real world setting. Diabetes Obes Metab. 2025;1-8. doi:10.1111/dom.16162

3. 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. doi:10.1056/NEJMoa2032183

4. McCarthy D, Berg A. Weight loss strategies and the risk of skeletal muscle mass loss. Nutrients. 2021;13(7):2473. doi:10.3390/nu13072473