Bimagrumab: Weight Loss That Is Entirely Fat Mass

Every other drug in this pipeline works on appetite. Bimagrumab does not. It is an antibody against the activin type II receptor that changes what the body does with muscle and fat, and in the BELIEVE trial that difference produced a result no incretin has matched: every kilogram lost on bimagrumab alone was fat mass, and three-quarters of the loss was still gone six months after treatment stopped.

What Bimagrumab Is — and Is Not

Bimagrumab is a monoclonal antibody against ActRIIA and ActRIIB, the activin type II receptors. Those receptors sit downstream of a family of signalling molecules — myostatin (GDF8), activins, GDF11 and related ligands — that regulate muscle protein balance and, in adipose tissue, lipolysis and fat distribution.

Blocking the receptor does two things at once. In muscle, it shifts the balance toward protein synthesis and away from degradation, and reduces intramuscular fat. In adipose tissue, it increases lipolysis and reduces visceral fat. Neither effect depends on appetite or food intake, which is why bimagrumab's mechanism is genuinely separate from every incretin in this pipeline.

BELIEVE measured that separation directly. At week 24, median daily calorie change was −182 kcal with high-dose bimagrumab, versus −482.5 kcal with semaglutide 2.4 mg and −487 kcal with the combination. Bimagrumab lost weight without materially reducing how much people ate.

BELIEVE: Design and Population

• 507 adults with obesity or overweight plus at least one complication, without diabetes (HbA1c below 6.5%).
• Mean baseline: age 47.5 years, weight 107.5 kg, BMI 37.3, fat mass 45.8 kg, lean mass 58.3 kg.
• Arms: bimagrumab 10 or 30 mg/kg, semaglutide 1.0 or 2.4 mg, the combinations, and placebo.
• Administration: a 30-minute IV infusion at a clinical site, with loading doses at weeks 1 and 4, then maintenance at weeks 16, 28, 40, 52 and 64.
• Every arm received the same lifestyle framework, including a protein target of at least 1.2 g/kg/day and at least 150 minutes of weekly activity.
• 48-week primary period, a 24-week extension, then a 32-week off-treatment withdrawal period — a design almost no obesity trial runs.

That last design choice is the most interesting thing about BELIEVE. Most obesity trials stop at the point of maximum weight loss. This one deliberately took the drug away to see what the body did next.

The Body-Composition Result

The headline everyone quotes is the 22.1% combination figure. The more important number is what happened to lean mass.

Body composition measured by DXA. Week-72 analyses were post hoc and not adjusted for multiple comparisons. A DXA lean-mass increase is not proof of increased strength or function; BELIEVE did not demonstrate a matching functional benefit.

Two caveats belong next to that table. First, "lean mass" on a DXA scan includes water, glycogen and connective tissue, so a rise is not automatically a rise in contractile muscle. Second, the combination still lost lean mass — less than semaglutide, but not zero. Bimagrumab attenuated the loss; it did not abolish it.

The Withdrawal Data Is the Real Story

Weight regain after stopping an incretin is the category's least comfortable fact. In BELIEVE's withdrawal phase, the arms behaved very differently:

• Bimagrumab 30 mg/kg retained 76% of its week-72 weight loss at week 104.
• Semaglutide 2.4 mg retained 51%.
• The combination retained 54%.

The composition of the regain differed too. For bimagrumab monotherapy, 100% of both the loss and the regain was fat mass — lean mass returned to baseline after withdrawal, having risen during treatment. For semaglutide, roughly 75% of both loss and regain was fat.

This is a mechanistic distinction with commercial consequences. If a drug changes body composition rather than appetite, the off-treatment trajectory may look different from the incretin pattern of rapid regain. BELIEVE is a 507-person Phase 2 and cannot settle that question, but it is the first obesity trial in this class to generate the data at all.

Cardiometabolic Signals

Safety and Tolerability

Three findings matter for anyone tracking this asset:

• Discontinuation. Adverse-event discontinuation was 14.0–21.4% with bimagrumab alone, 5.3–12.5% with the combination, and higher in the monotherapy arms than in the semaglutide arms. The trial authors suggested the IV loading schedule and peak exposure may have contributed.
• LDL cholesterol. High-dose bimagrumab raised LDL by 17.6% at week 72, while semaglutide lowered it by 8.9% and the combination was roughly neutral (+0.1%). LDL fell below baseline after bimagrumab was withdrawn, and the rise was smaller in statin users.
• Laboratory changes. Alkaline phosphatase and creatine kinase rose, with transient ALT/AST increases and small magnesium decreases. Lipase rose transiently with bimagrumab.

Serious adverse events occurred in 8.8–12.5% of bimagrumab participants and 3.6% of placebo participants. No deaths were reported. Three pancreatitis events occurred in the 48-week period, one each in the placebo, bimagrumab 10 mg/kg and semaglutide 1.0 mg arms.

Where the Program Stands

No Phase 3 obesity trial for bimagrumab appears in ClinicalTrials.gov as of this review. One Phase 2 diabetes study was withdrawn for stated strategic business reasons before a single participant enrolled; that withdrawal produced no efficacy or safety result and did not stop the non-diabetes program.

Why It Matters Beyond Bimagrumab

The muscle question has moved to the centre of obesity medicine. If incretins cause a quarter of weight loss to come from lean tissue, then the next competitive axis is not how much weight comes off but what the weight is made of. Bimagrumab is the clearest test of whether that can be engineered, and it is why our tirzepatide body-composition analysis is worth reading alongside this one.

For the broader mechanism context, see dual-incretin mechanisms and the peptide fundamentals guide.

Related Reading

Explore Our Research Tools

• Calculate volume and concentration parameters with the Peptide Calculator.
• Structure research intervals inside the Protocol Builder.
• Log baseline metrics with the Protocol Tracker Tool.

Sources

• Heymsfield SB, et al. Bimagrumab and semaglutide in adults with overweight or obesity (BELIEVE). Nature Medicine. 2026;32(3):869–882.
• Heymsfield SB, et al. Bimagrumab in adults with type 2 diabetes. JAMA Network Open. 2021;4(1):e2033457.
• Eli Lilly and Company. Cardiometabolic changes during 6-month withdrawal from bimagrumab and/or semaglutide: BELIEVE study extension. EASD 2026.
• ClinicalTrials.gov records NCT05616013, NCT06643728, NCT05933499, NCT06890611 and NCT06901349.
• FDA. Compounding laws and policies — biological products are not eligible for 503A or 503B exemptions.