Short answer: 25–40% of the weight lost on a GLP-1 is lean mass by DXA (a ceiling, not a muscle measurement), more than half of users stop within a year, about two-thirds of the lost weight returns after stopping — and resistance training plus 1.2–1.6 g/kg of protein are the two numbers that change that trajectory.
This page is the curated reference behind that answer: the numbers that actually matter about muscle and weight on GLP-1 medications, each one linked to its primary source, with the caveat that belongs next to it. No number here is estimated, rounded up for effect, or quoted from another blog. We update it when the evidence changes, and the date below the title is real.
Citing these numbers? Link the study — that is what the links are for. If this page helped you find them, a link here helps the next person do the same.
How much of the weight lost is lean mass
- 25–40% of total weight lost on GLP-1 therapy is lean mass, across the DXA data reviewed in Neeland et al., Diabetes, Obesity and Metabolism (2024) — the reference review on this question.
- The caveat that changes the headline: on a DXA scan, “lean mass” counts organ tissue, water and glycogen together with muscle. The 25–40% figure is a ceiling on muscle loss, not a measurement of it — our evidence review walks through why reading it as “muscle” overstates what is happening.
What the lost weight was actually made of
- 70% fat / 30% lean body mass — the human proof-of-concept arm of Langer et al., Cell Reports Medicine (2026): ten adults on semaglutide for twelve weeks, measured without DXA.
- In the same paper’s animal arm, a meaningful share of “lean” loss was liver tissue, not skeletal muscle — and a fatty liver shrinking is a good outcome. The human muscle-vs-organ split has not been measured; that number does not exist yet, which is why it is not on this page.
What happens after stopping
- Two-thirds of the lost weight returns within one year of stopping semaglutide — participants who lost 17.3% regained 11.6 percentage points, ending 5.6% below where they started (STEP 1 trial extension).
- +14% of body weight regained in 52 weeks after switching from tirzepatide to placebo, versus a further −6.7% for those who stayed on (SURMOUNT-4).
- The composition caveat: weight lost was partly lean mass; weight regained without resistance training is predominantly fat. “Back to the same weight” can mean a worse body composition than baseline — the arithmetic is in our after-GLP-1 guide.
How many people actually stop
- More than half stop within the first year — 20.4% within three months, 52.4% by twelve, in a Cleveland Clinic analysis of nearly 8,000 real-world patients (reported at Medscape).
- ~32% persistence at one year in commercially insured adults without diabetes, in payer claims data (Journal of Managed Care & Specialty Pharmacy, 2024).
- Why it matters: the year after the pen is the rule, not the exception — which is what the regain numbers above describe.
What holds muscle, by the numbers
- +1.1 kg (~2.4 lb) of lean mass in ~20.5 weeks — the average gain for aging adults doing full-body resistance training, across the trials in Peterson et al.’s meta-analysis; more training volume produced more. Our starter plan is built on the underlying prescription.
- 1.2–1.6 g of protein per kg of body weight per day is the lean-mass-sparing corridor in weight-loss contexts (American Journal of Clinical Nutrition), consistent with the PROT-AGE recommendations of 1.0–1.2 g/kg for older adults and 1.2–1.5 g/kg for those training — turned into meals in our protein guide.
- 0 — the number of controlled trials showing that creatine alone reduces lean-mass loss during GLP-1 use (the evidence, reviewed honestly). Creatine supports the strength work; it does not replace it. Our evidence review covers what it does and does not do.
The numbers being generated right now
Two registered trials sit exactly on this page’s questions and have not reported yet: NCT07480109 (higher protein intake in midlife women on a GLP-1) and NCT07457437 (lean-mass preservation strategies during GLP-1 therapy). When they report, this page changes — that is what the update date is for.
How to read all of this together
Weight lost on a GLP-1 is roughly one-quarter to one-third lean tissue by DXA — a ceiling, not a muscle measurement. Most people stop the drug within a year, and most of the weight returns, as fat, unless resistance training and adequate protein give the body a reason to rebuild muscle instead. The numbers for doing that are on this page too. That is the whole picture, sourced.