Journal

Muscle Loss on GLP-1s: What the 2026 Body-Composition Data Shows

A substantial fraction of weight lost on these drugs is lean mass. New trial data suggests it may be modifiable.

What the concern actually is

Weight loss by any method costs some lean mass; that is normal physiology. The question specific to high-efficacy GLP-1s is whether the rate and magnitude of loss shifts that ratio unfavourably.

Body-composition work published through 2025 and 2026 puts the lean-mass share of total loss in a range around 20% to 40% depending on the analysis and the population. Reviews of older adults cite figures at the higher end of that range.

The trial that changed the conversation

A phase 2 study combining semaglutide with bimagrumab, a myostatin inhibitor, reduced the lean-mass fraction of total weight loss from roughly 21% to about 7%. That is a large effect on the composition of loss, and it is the first strong signal that the ratio is pharmacologically modifiable rather than fixed.

The gap researchers keep flagging is function. Lean mass measured on a scan is not the same as strength, mobility or independence, and large trials measuring muscle function are largely absent. There is also no settled clinical method for measuring it routinely. Preserving mass on a DEXA scan is a surrogate; whether it translates to a person climbing stairs more easily has not been demonstrated at scale.

Who this matters most for

Adults over 65 carry the highest concern, because they begin with less lean mass, lose it faster, and have more to lose functionally when it goes. Reviews in 2026 have raised malnutrition and bone density alongside muscle in this group.

What clinicians treating older patients increasingly recommend — and this belongs in a conversation with your own clinician rather than as a rule from a website — includes establishing baseline body composition before starting, attention to protein intake, resistance training alongside drug therapy rather than instead of it, and slower dose escalation to moderate the rate of loss.

Those are clinical decisions with individual answers. The specifics depend on your history, your kidney function, your other medications and what you can actually sustain.

What the evidence does not establish

• Whether lean-mass loss on GLP-1s produces worse functional outcomes than equivalent weight loss by other means. The comparison has not been made rigorously.
• Whether bimagrumab combination therapy improves function, as opposed to composition. Phase 2 measured the latter.
• What happens to lean mass after stopping. The regain literature measures weight, not composition.
• Anything about compounded preparations, microdoses or oral formulations, none of which appear in body-composition trials.

The practical framing

This is not a reason not to treat obesity, which carries its own well-documented risks. It is a reason to treat it as a clinical process with monitoring rather than as a purchase.

Ask your prescriber what baseline measurements they take, whether they monitor composition, and what their approach is to escalation rate. A programme that titrates on a fixed calendar without assessment is not doing that work — and the label itself sets a minimum interval based on tolerability and response rather than a schedule.

Questions readers actually ask