Bimagrumab is an experimental antibody being developed to make weight loss come more from fat and less from muscle. It blocks the activin type II receptors — a signalling route that normally puts a brake on muscle growth — and in its main obesity trial it was given alongside semaglutide, the molecule in Wegovy. It is not approved anywhere, and no phase 3 trial is registered.
This page is the trial record. Everything below was read from ClinicalTrials.gov's posted results and PubMed on 15 October 2026. Why muscle loss on GLP-1 drugs became a question at all, and the other muscle-sparing agents in testing, are on our muscle loss page; this page does not repeat them.
Where the molecule came from
Bimagrumab was developed by Novartis as BYM338. Its registry history runs well before obesity: Novartis tested it in sporadic inclusion body myositis, sarcopenia in older adults, recovery after hip-fracture surgery, cancer-related weight loss and COPD with muscle wasting. ClinicalTrials.gov lists 17 bimagrumab studies in all, and none of those earlier programmes produced an approval.
The obesity signal came from a small Novartis trial in 75 adults with type 2 diabetes and obesity (NCT03005288), published in 2021. Over 48 weeks of monthly infusions, fat mass fell 20.5% on bimagrumab against 0.5% on placebo, lean mass rose 3.6% against a 0.8% fall, and body weight fell 6.5% against 0.8%. That trial analysed only people who completed the full regimen, so its figures flatter the drug compared with an intention-to-treat analysis.
The programme then passed to Versanis Bio, which designed the semaglutide trial below, and then to Eli Lilly: the 2026 publication's disclosures list several authors as former Versanis staff now employed by Lilly, and Lilly is the registry sponsor.
BELIEVE: nine arms, one table
The BELIEVE trial (NCT05616013) randomised 507 adults in the United States, Australia and New Zealand with a BMI of 30 or more, or 27 or more with a weight-related condition other than diabetes. People were assigned to placebo, bimagrumab alone at 10 or 30 mg/kg, semaglutide alone at 1.0 or 2.4 mg, or one of four combinations. Bimagrumab was infused at baseline and at weeks 4, 16, 28 and 40; semaglutide was injected weekly. The 48-week results were posted to the registry in July 2025 and published in Nature Medicine in March 2026.
| Arm (48 weeks) | Weight change | Fat mass change (DXA) | Lean mass change (DXA) | Lost ≥15% of weight |
|---|---|---|---|---|
| Placebo | -3.31 kg | -2.41 kg | -0.33 kg | 5.7% |
| Bimagrumab 10 mg/kg | -5.99 kg | -6.90 kg | +1.48 kg | 16.1% |
| Bimagrumab 30 mg/kg | -9.25 kg | -11.86 kg | +1.19 kg | 32.4% |
| Semaglutide 1.0 mg | -9.76 kg | -7.49 kg | -2.86 kg | 35.6% |
| Semaglutide 2.4 mg | -14.24 kg | -11.06 kg | -4.43 kg | 47.1% |
| Bimagrumab 10 + semaglutide 1.0 | -12.73 kg | -14.17 kg | -1.28 kg | 41.5% |
| Bimagrumab 10 + semaglutide 2.4 | -14.31 kg | -15.93 kg | -1.12 kg | 55.0% |
| Bimagrumab 30 + semaglutide 1.0 | -13.86 kg | -15.33 kg | -0.65 kg | 60.5% |
| Bimagrumab 30 + semaglutide 2.4 | -17.79 kg | -19.30 kg | -1.43 kg | 79.1% |
Source: ClinicalTrials.gov posted results, least-squares means. Arms held 31 to 51 people with a week-48 weight measurement; the DXA measures come from slightly smaller subsets.
Two things in the table are easy to miss. Bimagrumab on its own produced weight loss with lean mass going up, so the fat it removed was more than the scale showed. And the fat column does not add up to the weight column in several rows — the high-dose combination lost 19.30 kg of fat and 1.43 kg of lean mass, more than its 17.79 kg weight change. The scale and the DXA scan are separate measurements, modelled separately on different subsets of each arm, so the columns are best read against their own placebo rather than summed.
The figure the trial was built to produce
The question bimagrumab exists to answer is how much of the weight lost is lean tissue. The registry does not state it as a single percentage, so glp1ledger divided each arm's lean-mass change by its weight change:
| Arm | Lean mass as a share of weight lost (approximate) |
|---|---|
| Semaglutide 1.0 mg alone | about 29% |
| Semaglutide 2.4 mg alone | about 31% |
| Bimagrumab 10 + semaglutide 2.4 | about 8% |
| Bimagrumab 30 + semaglutide 2.4 | about 8% |
| Bimagrumab 30 + semaglutide 1.0 | about 5% |
Because the two means come from slightly different groups of people, these ratios are indicative, not trial estimates. The registry's own pre-specified version points the same way: the share of people who lost at least 10% of their fat while losing less than 5% of their lean mass was 19.1% on semaglutide 2.4 mg alone and 67.4% on the high-dose combination.
What the table cannot say is whether the preserved lean mass is functioning muscle. DXA lean mass counts water, organs and connective tissue along with muscle, and the trial's quality-of-life measures (SF-36 physical functioning, IWQOL-Lite) were secondary outcomes. Strength and physical performance — the outcomes that would matter most to someone worried about frailty — were not the endpoints.
Side effects, from the registry counts
The Nature Medicine abstract names muscle spasms, diarrhoea and acne as the common bimagrumab events, and nausea, diarrhoea, constipation and fatigue for semaglutide. The registry's adverse-event table gives the counts, and glp1ledger converted them to percentages:
| Event | Bimagrumab 30 mg/kg alone (57) | Semaglutide 2.4 mg alone (56) | Combination 30 + 2.4 (55) |
|---|---|---|---|
| Muscle spasms | 72% | 11% | 64% |
| Acne | 49% | 11% | 53% |
| Diarrhoea | 51% | 38% | 49% |
| Nausea | 14% | 46% | 49% |
| Vomiting | 5% | 18% | 15% |
One caveat governs the whole table. The adverse-event window runs to week 72, and in the extension every group — the placebo group included — received bimagrumab infusions at weeks 52 and 64. That is why the "placebo" group records muscle spasms in 22 of 55 people and acne in 20 of 55. There is no clean placebo column for bimagrumab's side effects in this table, and comparing against the semaglutide-alone arm is the closer reference. Serious adverse events ranged from 4 to 9 people per arm.
What is running now
ClinicalTrials.gov lists four Lilly-era bimagrumab studies in obesity beyond BELIEVE:
| Registry number | Design | Size | Status (15 October 2026) |
|---|---|---|---|
| NCT06643728 | Bimagrumab and tirzepatide, alone or combined, phase 2 | 252 | Active, not recruiting; primary completion January 2026; no results posted |
| NCT06890611 | Bimagrumab alone and with tirzepatide, healthy volunteers, phase 1 | 125 | Completed September 2025 |
| NCT06901349 | Bimagrumab and tirzepatide in obesity with type 2 diabetes, phase 2 | 0 | Withdrawn |
| NCT05933499 | Tirzepatide and bimagrumab, body composition and bone, phase 2 (Massachusetts General Hospital) | 63 | Recruiting; completion December 2028 |
No phase 3 obesity trial is registered. The two "phase 2/3" and "phase 3" bimagrumab records on the registry belong to Novartis's 2010s myositis programme. That Lilly is pairing the antibody with tirzepatide — its own weight-loss drug — is the clearest signal of where the programme is heading. Tirzepatide's own trial record is on our Zepbound weight-loss results page.
What is not established
- Long-term safety. The longest data run to 72 weeks in a few hundred people per dose.
- Function. Whether the lean mass preserved on the combination translates into strength, mobility or fewer falls has not been tested as a primary outcome.
- The route. BELIEVE used intravenous infusions dosed by body weight. What a marketed product's dose, route and schedule would be is not published.
- Whether it works with tirzepatide. The phase 2 combination trial has not posted results.
Peptide Lexicon's bimagrumab entry gives the definition of the molecule; the wider pipeline is on how GLP-1 drugs work. Questions about muscle, weight-loss medicines and individual risk belong with a clinician who knows the person's history.
Sources
- ClinicalTrials.gov, NCT05616013 (BELIEVE), protocol and posted results (results first posted 18 July 2025; record updated 8 July 2026): weight, DXA fat and lean mass, response rates, participant flow and adverse events, clinicaltrials.gov/study/NCT05616013, read 2026-10-15.
- Heymsfield SB et al. "Bimagrumab plus semaglutide alone or in combination for the treatment of obesity: a randomized phase 2 trial." Nat Med 2026;32(3):869-882, PubMed 41772149, doi 10.1038/s41591-026-04204-0. Disclosures list current and former Lilly and Versanis staff.
- Heymsfield SB et al. "Effect of bimagrumab vs placebo on body fat mass among adults with type 2 diabetes and obesity: a phase 2 randomized clinical trial." JAMA Netw Open 2021;4(1):e2033457, PubMed 33439265 (NCT03005288).
- ClinicalTrials.gov, intervention search "bimagrumab": 17 records, including NCT06643728, NCT06890611, NCT06901349 and NCT05933499, read 2026-10-15.
- PubMed search "bimagrumab": 93 records on 15 October 2026.