About 25-45% of weight lost on GLP-1 drugs is lean tissue, though 2026 analyses contest whether that is disproportionate to ordinary dieting. The strongest mitigation data is BELIEVE, a 507-person phase 2 trial where bimagrumab plus semaglutide cut fat mass 45.7% while losing only 2.9% lean mass at week 72. A five-arm trial in hypogonadal men found testosterone plus resistance training produced net lean-mass gain. Larger trials with MRI muscle endpoints read out in 2027.
Summary drafted from this entry’s sources; editorial review pending.
01Key questions
Do GLP-1 drugs cause muscle loss?
DXA substudies put lean tissue at 25-45% of total weight lost, but a 2026 Cell Reports Medicine analysis argues the loss is proportional to what equivalent diet-induced weight loss produces. Concern concentrates in older and low-muscle-mass patients.111
What preserves muscle on GLP-1 drugs?
The strongest data is BELIEVE, a 507-person phase 2 trial where bimagrumab plus semaglutide cut fat mass 45.7% while losing only 2.9% lean mass at week 72. A five-arm trial in hypogonadal men found testosterone plus resistance training produced net lean-mass gain. Resistance training and adequate protein are the accessible options while larger trials run.11
Is bimagrumab approved?
No. Bimagrumab is not approved for obesity or muscle preservation, and no GLP-1 drug carries an FDA muscle-preservation claim.
What would change this rating?
LEAN-PREP (n=232, resistance exercise and protein arms) and RESIST (n=150, the first large trial with MRI muscle volume as the primary endpoint, completing May 2027) will supply causal data on resistance training during GLP-1 therapy.4
02Problem Statement
25–45% of weight lost on GLP-1 receptor agonists (semaglutide, tirzepatide, retatrutide) is lean tissue, not fat mass, per DXA substudies. This creates secondary market need for muscle preservation protocols in longevity and weight-management clinics.
03Human Clinical Trial Evidence (Primary Sources)
Key Trials Showing Lean Mass Loss
- SURPASS-3 MRI post-hoc (Tirzepatide): MRI substudy (n=246): muscle volume decrease proportional to weight loss; maintained fat-free muscle volume, improved muscle quality (less fat infiltration). web:0
Meta-analyses (2025–2026) confirm heterogeneity: lean mass loss 15–60% of total weight lost depending on measurement (DXA vs MRI) and population. See 2026 counter-evidence section below — the "muscle wasting" framing is increasingly contested.
Animal/in-vitro: Limited; mouse models show predominant fat loss with minor LBM decrease (Langer et al. 2026, Cell Rep Med).
042026 Practice-Changing Evidence
1. Bimagrumab + Semaglutide — Full Publication (BELIEVE) — PMID 41772149
Heymsfield SB et al. "Bimagrumab plus semaglutide alone or in combination for the treatment of obesity: a randomized phase 2 trial." Nature Medicine. 32(3):... (Mar 2026). PMID 41772149. DOI 10.1038/s41591-026-04204-0. PMC13004672. Trial NCT05616013.
Double-blind, placebo-controlled, full factorial (2×2 + monotherapy arms, 9 groups total), 26 sites (US/Australia/New Zealand), n=507 randomized, BMI ≥30 or ≥27 + obesity complication (no diabetes). IV bimagrumab 10 or 30 mg/kg every 12 weeks; SC semaglutide 1.0 or 2.4 mg weekly. 48-week primary period → 72-week OLE → 104-week withdrawal follow-up (still ongoing). Funded by Versanis Bio (acquired by Eli Lilly 2023).
Week 48 (treatment-regimen estimand, LSM ± SE):
| Arm | Body weight (kg) | Fat mass (kg) | Lean mass (kg) | Fat-loss index |
|---|---|---|---|---|
| Placebo | −3.3 ± 1.4 | −1.9 ± 1.2 | −0.5 ± 0.7 | — |
| Bimag 10 mg/kg | −6.0 ± 1.4 | −5.7 ± 1.2 | +0.5 ± 0.7 | — |
| Bimag 30 mg/kg | −9.3 ± 1.3 | −8.6 ± 1.2 | +0.4 ± 0.7 | 100% (all weight lost = fat) |
| Sema 1.0 mg | −9.8 ± 1.3 | −6.7 ± 1.0 | −2.6 ± 0.6 | — |
| Sema 2.4 mg | −14.2 ± 1.2 | −9.6 ± 1.0 | −3.9 ± 0.6 | 71.1% |
| Bimag 10 + Sema 2.4 | −13.9 ± 1.3 | −12.2 ± 1.1 | −0.6 ± 0.6 | — |
Swipe table →
Week 72 (efficacy estimand, LSM): - Bimag 30: −10.8% body weight, −28.5% fat mass, +2.5% lean mass (P<0.001 vs placebo) - Sema 2.4: −15.7% body weight, −27.8% fat mass, −7.4% lean mass - High-dose combo: −22.1% body weight, −45.7% fat mass, only −2.9% lean mass (P<0.001 combo vs sema 2.4 for lean preservation)
Other notable endpoints: - 94.0% of high-dose-combo participants achieved ≥30% fat-mass reduction at week 72 (vs 36.4% sema 2.4, 50.0% bimag 30) - Mean percent body fat: 43.7% → 32.5% with high-dose combo; 41.9% → 37.9% with sema 2.4 - Estimated VAT: −58.2% (combo) vs −35.8% (sema) at week 72 - hsCRP down up to 83%, adiponectin substantially increased - Prediabetic subgroup: 100% reversion to normoglycemia in some two-drug combination groups - Appendicular lean mass (skeletal-muscle proxy): better preserved with combo vs sema alone (Table 2 / Extended Data Table 1) - Total body BMD small increases across active arms (consistent with reduced mechanical loading + anabolic effect of preserved lean)
Safety (bimagrumab signature AEs): muscle spasms, diarrhea, acne (ActRII blockade → known class effect). Semaglutide AEs as expected (nausea, diarrhea, constipation, fatigue). Discontinuation for AE: bimagrumab 14–21% vs sema 3.6–8.8% vs combo 5.3–12.5% vs placebo 3.6%. No deaths. Importantly: combo did NOT stack bimagrumab's discontinuation rate — the combination was better tolerated than bimagrumab monotherapy.
Critical interpretation: - High-dose combo achieved fat-mass reduction (45.7%) in the range reported for bariatric surgery at similar timepoints - Combination produced "nearly additive" fat loss because mechanisms are independent: sema targets CNS appetite, bimag targets ActRII→ALK7 adipose lipid mobilization + ALK4 muscle anabolism - PMID 41772149 confirms the 2024 ADA late-breaking preview numbers but provides week-72 OLE data showing continued divergence of fat vs lean curves through the open-label extension - Limitation: DXA measures lean mass, not muscle directly (includes connective tissue, water); intramuscular fat quantified only in earlier bimagrumab-T2D study. Phase 3 with MRI muscle-volume primary endpoint is the next data milestone.
2. CagriSema Body Composition — REDEFINE-1 DXA Substudy (ECO 2026, May 2026)
Presented by Ravussin (Pennington) at European Congress on Obesity 2026; DXA subgroup, week 68: - CagriSema (cagrilintide 2.4 mg + semaglutide 2.4 mg): −23.9% body weight — vs semaglutide 2.4 mg −16.6%, cagrilintide −15.0%, placebo −2.8%. - Composition of loss: 66.9% fat mass / 33.1% lean soft tissue (CagriSema); semaglutide 69.7% fat; cagrilintide 62.9% fat. - In CagriSema responders losing ≥30% body weight, lean-soft-tissue share of loss increased over time (51.3% → 63.2% between measurement points) — i.e., loss became progressively more fat-predominant; muscle strength was preserved. - Read-through: amylin co-agonism does not worsen lean-mass fraction vs GLP-1 alone at class level.
3. REDEFINE-4 Head-to-Head + Regulatory Milestone (Feb 2026)
- CagriSema NDA submitted to FDA December 2025 (weight management, based on REDEFINE-1/2); decision expected late 2026. First amylin-combination approval would expand the muscle-preservation adjunct market (cagrilintide's lean fraction profile is slightly worse than semaglutide's in REDEFINE-1 DXA).
4. 2026 Counter-Evidence — "Disproportionate Muscle Loss" Challenged
The debate has tightened; balance both directions: - Langer HT et al. Cell Rep Med. 2026;7:102665 — "Weight loss with GLP-1 medicines does not result in a disproportionate loss of muscle mass or function in obese mice and humans": lean loss proportional to what diet-induced weight loss of same magnitude produces. - SEMALEAN (Alissou M et al. Diabetes Obes Metab. 2026;28:112–121) — semaglutide fat/lean mass + muscle function outcomes. - Sawicka-Gutaj N et al. Int J Obes. 2026 (systematic review + network meta-analysis; doi 10.1038/s41366-026-02088-1) — class-level body-composition comparison. - Corso H et al. Front Endocrinol. 2026;17:1834580 (July 2026) — real-world tirzepatide body composition. - medRxiv 2026.04.11.26350687 — phenotyping study: baseline phenotype determines position on fat-loss ↔ mixed-loss spectrum; most reproducible signal = greater relative LBM decline with tirzepatide vs semaglutide.
Synthesis: lean loss is real but proportional-ish; the clinical concern concentrates in sarcopenic-risk, elderly, low-baseline-lean cohorts — exactly where DXA/MRI monitoring and resistance training matter most.
05Mitigation Strategies and Combination Evidence
Ongoing RCT — LEAN-PREP (primary protocol source): Alawadhi AA et al. BMJ Open. 2026;16(4):e116911. doi 10.1136/bmjopen-2025-116911. Trial reg: NCT06885736 (recruiting).
Design: Single-centre (Dasman Diabetes Institute), 6-month, 1:1:1:1 RCT, n=232 obese adults (BMI 27–45) starting semaglutide or tirzepatide. - Arms: Control (usual care) vs Resistance exercise (home-based 3×/wk, progressive) vs Protein (1.6 g/kg/day) vs Combined. - Primary outcome: MRI quadriceps CSA (vastus lateralis). Secondary: DEXA body comp, strength/function, liver fat/stiffness (MRI PDFF/MRE), QoL, sleep, biomarkers. - Protein target based on Morton meta-regression (1.6 g/kg threshold).
Pharmacologic adjunct hierarchy (Aug 2026): 1. Bimagrumab + semaglutide — only combination with full published Ph2 RCT evidence (Heymsfield, Nat Med 2026). Not approved; IV administration. 2. Myostatin/activin-pathway inhibitors (Regeneron, BioAge, Veru) — Ph1/2; mechanism validated by bimagrumab data. 3. Resistance exercise + protein 1.6 g/kg/day (the protein level used in trials), the most accessible option while LEAN-PREP results are pending; zero cost, functional-outcome track record. 4. Monitoring: baseline + serial DXA or MRI in at-risk cohorts (older, low lean mass, rapid losers >1.5%/wk).
5. Multimodal Anabolic Stack — TRT + Resistance Training + High-Protein (2026 RCT)
Reference: Abstract 470, J Sex Med 2026; doi 10.1093/jsxmed/qdag118.413 — "Preservation of Lean Mass under GLP-1/GIP-Based Pharmacotherapy: Effects of Testosterone and Resistance Training — A Five-Arm, Randomized, Open-Label Parallel-Group Study in Obese Men."
- 26-week, multicenter, randomized, open-label (assessors blinded), n=266 obese men, BMI ≥30, age 18–70, total testosterone <3.5 ng/mL with hypogonadism-compatible symptoms, eligible for GLP-1/GIP therapy.
- Primary endpoint: change in fat-free mass by bioimpedance. Secondary: body weight, waist circumference, maximal strength (leg press/row/bench), IIEF.
FFM share of weight loss (26 weeks, by arm):
| Arm | FFM share of weight loss | Strength change (leg press) | IIEF (sexual function) |
|---|---|---|---|
| GLP-1/GIP alone | 38% | up to −23% | −2.1 (decline) |
| + TRT | 26% | (modest gain) | +3.8 to +4.6 |
| + TRT + training | 16% | (improving) | +3.8 to +4.6 |
| — | 0% — net FFM gain +4.2% | +43.4% leg press, +31.4% row, +27.1% bench | +4.6 |
| + high-protein alone | 32% | partial | — |
Swipe table →
Tradeoff observation: Greatest absolute weight loss in GLP-1/GIP-only arm (−12.1 kg) but also highest muscle loss. Waist circumference reduction was LEAST in the GLP-1/GIP-only arm (−6.3 cm) and MOST in the combined arm (−8.9 cm) — i.e., the anabolic stack changes body recomposition more than absolute weight loss.
Critical interpretation: First RCT to demonstrate that pharmacologic + behavioral + hormonal synergy can produce net FFM gain during GLP-1 therapy in hypogonadal men. Major caveat: open-label design, single-gender (men only), industry-symposium framing (J Sex Med abstract session), and protein target (2.2 g/kg FFM/day) is ~38% above LEAN-PREP's 1.6 g/kg/day. Generalization to eugonadal men and to women not established.
6. RESIST — Phase IV Resistance Training + Tirzepatide (NCT07777575, started 2026-08-17)
- Sponsor: Technical University of Munich, with EGYM SE (smart-gym equipment partner).
- Design: Phase IV, randomized, double-blind (assessor), parallel-group, n=150 adults with obesity.
- Primary outcome: Change in contractile thigh muscle volume by MRI (mL) — first large RCT using direct MRI muscle volume as the primary endpoint (more rigorous than DXA lean mass).
- Secondary: muscle mass by bioimpedance, training adherence (automatic EGYM device logs), strength, body composition.
- Completion: 2027-05-31 (primary).
- Significance: Will provide the highest-quality causal evidence for the "resistance training alone preserves muscle on GLP-1/GIP" claim that LEAN-PREP (lifestyle + protein) is testing. Watch for late-2027/2028 readouts.
06Regulatory & Claims Audit
- No FDA approval for muscle-preservation claims with GLP-1s; bimagrumab not approved for obesity. Clinics selling "GLP-1 muscle protection" stacks are ahead of evidence — only exercise/protein has guideline backing.
- CagriSema NDA (Dec 2025, decision late 2026) is the class regulatory event to watch; orforglipron/oral GLP-1s will re-run the same lean-mass debate.
What studies and practitioners report
Amounts as reported in the cited sources: trial doses with the population studied, and practitioner-reported ranges marked as such. Not medical advice.
- Trial 2.4 mg/week · Semaglutide (STEP 1) adults with obesity or overweight, n=1961 PMID 33567185
- Trial 5–15 mg/week · Tirzepatide (SURMOUNT-1) adults with obesity or overweight, n=2539 PMID 35658024
07Connections
08Sources
- Heymsfield SB et al. (2026). Bimagrumab plus semaglutide alone or in combination for the treatment of obesity: a randomized phase 2 trial. Nature Medicine . PMID 41772149 . DOI 10.1038/s41591-026-04204-0 . PMC13004672 . NCT05616013 . The single most important paper in this file.PubMed 41772149 · DOI 10.1038/s41591-026-04204-0 · PMC13004672 · NCT05616013
- DOI 10.1038/s41366-026-02088-1DOI 10.1038/s41366-026-02088-1
- DOI 10.1136/bmjopen-2025-116911DOI 10.1136/bmjopen-2025-116911
- Alawadhi AA et al. (2026). LEAN mass Preservation with Resistance Exercise and Protein during semaglutide and tirzepatide therapy (LEAN-PREP study): protocol for a randomised controlled trial. BMJ Open 16(4):e116911. PMID 42020128 . NCT06885736 .NCT06885736 · PubMed 42020128
- Abstract 470 (2026). Preservation of Lean Mass under GLP-1/GIP-Based Pharmacotherapy: Effects of Testosterone and Resistance Training — A Five-Arm, Randomized, Open-Label Parallel-Group Study in Obese Men. J Sex Med . doi 10.1093/jsxmed/qdag118.413 .DOI 10.1093/jsxmed/qdag118.413
- NCT07777575 (RESIST, started 2026-08-17). Resistance Training to Prevent Muscle Loss During Treatment With Tirzepatide — Phase IV, Technical University of Munich + EGYM SE. Primary completion 2027-05-31.NCT07777575
09Entry history
Latest content update.
Heretic informs and never prescribes. Where amounts appear, they describe what published studies used. Talk to a qualified clinician before acting on anything here.