Trial review

Tirzepatide, Lean Mass and Muscle Loss

What is measured, and what is not

Body-composition work through 2025 and 2026 has consistently found that a substantial fraction of weight lost on these drugs is lean mass rather than fat — figures around 20% to 40% appear across reviews, with analyses of older adults clustering at the higher end.

Some lean-mass loss accompanies weight loss by any method; that is normal physiology. The question specific to high-efficacy GLP-1s is whether the rate and magnitude shift the ratio unfavourably.

The trial that showed the ratio is modifiable

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%, presented at ADA 2026.

That is the first strong evidence that the composition of loss is pharmacologically adjustable rather than fixed. It reframes the problem from an unavoidable cost of treatment into something that might be engineered away.

The gap nobody has closed

Function. Lean mass on a DEXA scan is a surrogate for strength, mobility and independence. Researchers at ADA 2026 flagged repeatedly that large trials measuring muscle function are largely absent, and that clinical measurement of function is not standardised.

So the position is: composition has been improved in a phase 2 trial, and the outcome anyone actually cares about — whether a person can rise from a chair unaided in ten years — has not been measured at scale.

A field that has repeatedly learned surrogates can mislead should hold that distinction visibly rather than assume it away.

Four tiers of evidence, separated

1. Established: weight loss costs lean mass; the share on high-efficacy GLP-1s is substantial.
2. Demonstrated in a trial: myostatin inhibition reduces that share.
3. Physiologically reasonable, untested here: that protein intake and resistance training preserve lean mass during GLP-1-driven loss specifically.
4. Not established at all: that preserving lean mass preserves function.

What this changes about treatment

It argues for monitoring rather than for avoidance. Untreated obesity carries its own well-documented risks, and this is not a reason to decline treatment — it is a reason to treat it as a clinical process with measurement rather than as a purchase.

Practically: ask what is measured before starting and what is re-measured; ask what would make the prescriber slow escalation. The tirzepatide label already directs increases based on tolerability and response rather than on a calendar, and names three maintenance dosages rather than one destination.

Specific protein targets and training prescriptions belong with a clinician or dietitian who can see your bloods, not with a website publishing numbers for a general audience.

Questions readers actually ask