Incretin Drugs and the Sarcopenia Question

A review published in the Diabetes & Metabolism Journal in September 2026 argues that the obesity field has been watching the wrong number on incretin-based weight-loss drugs. Kim Jae Yong and Kim Nan Hee looked across GLP-1 receptor agonists, dual GIP/GLP-1 agonists such as tirzepatide, and the emerging multi-agonist drugs still in trials, and made a class-wide case: large weight loss on these drugs does come with some lean-tissue loss, and that loss deserves active management instead of being waved off as a side effect of success^[1].

Their paper is not about one molecule. It reads across semaglutide, tirzepatide, and the newer triple agonists in development, and its central claim is careful, not alarmist: lean-mass loss on a body scan does not automatically mean a patient has lost contractile muscle, and it does not automatically mean sarcopenia. But the authors argue that nobody has ruled either one out for every patient, and current practice mostly tracks the scale, not the muscle^[1].

How much of the weight lost is muscle

The clearest human numbers come from a body-composition analysis of SURMOUNT-1, the trial that established tirzepatide's headline weight-loss figures. At 72 weeks, tirzepatide produced a mean 21.3% reduction in total body weight against 5.3% on placebo. Of that, fat mass fell 33.9% and lean mass fell 10.9%^[2]. Roughly three-quarters of the weight lost was fat and one-quarter was lean tissue, a split that also showed up in the placebo group losing weight through diet alone. Fat comes off faster than muscle, but muscle still comes off.

That ratio is the review's starting point, not its conclusion. Lean mass on a DXA or bioimpedance scan includes organs, bone, and body water, not only skeletal muscle, so a 10.9% lean-mass drop overstates how much muscle actually left the body^[1]. The review's concern is narrower and more specific: patients who start treatment with limited muscle reserve, especially older adults, are the group where a proportionally modest lean-mass loss could still matter functionally, and that group is under-represented in the trials that established these figures.

What the review recommends

Rather than one fix, the paper proposes a package: routine body-composition assessment alongside the usual weight and metabolic labs, functional checks such as grip strength or gait speed instead of the scale alone, resistance training, and adequate protein intake, all under clinical supervision^[1]. It also points to a newer, still-early category: drugs designed to protect muscle while an incretin therapy removes fat.

The clearest example is a phase 2 trial published in 2026 that paired semaglutide with bimagrumab, an antibody that blocks the activin and myostatin signalling pathway that normally caps muscle growth. At 72 weeks, semaglutide alone reduced lean mass by 7.4%; bimagrumab alone increased it by 2.5%; and the combination held the lean-mass loss to 2.9% while still producing a 22.1% total weight reduction, with 92% of that loss coming from fat^[3]. It is one trial in one population, and nothing from it is an approved treatment, but it is the shape of evidence the review says the field needs more of.

What's established and what's still open

Some of this is reasonably settled. Fat mass falls faster than lean mass across the incretin class, so the proportion of the body made up of lean tissue tends to rise even while the absolute amount falls^[2]. What is not settled is the size of that absolute drop across the wider class: a 2024 review of lean-mass change across GLP-1-based therapies found estimates ranging from roughly 15% of total weight lost at one end to 40% or 60% at the other, depending on the population, the molecule, and existing disease^[4].

Also open: whether the SURMOUNT-1-style ratio holds up over years of treatment, not just months, whether it looks the same in frailer or older patients who start with less muscle to spare, and whether monitoring protocols like the one the review proposes actually change outcomes when tested in practice. None of the trials behind this review were designed to answer that last question.

What this is not

This is not evidence that tirzepatide, semaglutide, or the multi-agonist drugs behind them, including the investigational retatrutide, are unsafe, and the review does not argue for avoiding them. It is also not evidence that resistance training or higher protein intake has been proven, in a trial of people actually taking these drugs, to prevent lean-mass loss. Most of the exercise evidence comes from general dieting trials, not incretin-drug trials, and the bimagrumab combination data comes from a single phase 2 study that has not been replicated^[3]. Nothing here is a personal exercise or protein prescription. The review's recommendations are for clinical, supervised care, not a routine to start on your own alongside a prescription.

What this means for readers

For someone currently taking tirzepatide or semaglutide, the practical read on this review is not alarm. It is a case for asking about monitoring. A patient with a large weight-loss target and limited starting muscle, an older adult especially, is a reasonable candidate for the body-composition and functional checks the review describes.

That conversation belongs with a prescriber who knows the individual case, not with an article. Check the tirzepatide regulation and safety page and the semaglutide regulation and safety page for the current label information in your country before assuming the trial averages above describe your situation specifically.