IPM Take
The GLP-1 muscle debate risks becoming trapped between two equally unhelpful headlines: that these drugs are “eating muscle”, or that muscle loss does not matter because most weight lost is fat.
The evidence points somewhere more complicated.
Lean mass does fall during substantial GLP-1-induced weight loss. But lean mass is not synonymous with skeletal muscle, some loss is expected whenever body weight falls substantially, and current evidence does not establish that GLP-1 therapies uniquely cause sarcopenia. The more clinically useful question is therefore not whether every patient needs to fear muscle loss. It is which patients begin treatment with too little physiological reserve to lose much of anything except fat.
That distinction could change obesity care. As increasingly potent therapies produce weight reductions that were once unusual outside bariatric surgery, treatment success may need to be measured not just in kilograms, BMI or waist circumference, but also in strength, mobility, nutrition and body composition in patients at greatest risk.
Executive Summary
Muscle and lean-mass loss during GLP-1 therapy was debated at the American Diabetes Association’s 2026 Scientific Sessions and revisited by Medscape in September. Across weight-loss studies, reductions in fat-free or lean mass have been reported alongside much larger reductions in fat mass. In the STEP 1 semaglutide body-composition substudy, body weight fell by 15.0%, total fat mass by 19.3% and lean body mass by 9.7% over 68 weeks. Importantly, lean mass as a proportion of total body weight actually increased because fat mass declined more substantially.
But interpreting those numbers is not straightforward. DXA-derived “lean mass” includes water, organs and other fat-free tissues, not simply skeletal muscle. Researchers have also argued that some apparent fat-free-mass loss reflects the water and protein contained within adipose tissue that disappears as fat stores shrink. In other words, a decline in measured lean mass should not automatically be translated into the same amount of lost functional muscle.
The signal for health systems is therefore becoming more targeted. Older adults, people with sarcopenic obesity and others with limited muscle reserve may warrant closer monitoring, while resistance exercise, appropriate nutrition and adequate protein remain central to preserving function during intentional weight loss. The ADA’s 2026 Standards of Care specifically recommend adequate protein and resistance or weight-bearing exercise for older adults with diabetes, particularly when they are pursuing weight loss.
Why it matters
- HTA bodies: As obesity therapies become more potent, future assessments may increasingly encounter outcomes beyond percentage weight loss. Functional status, body composition and preservation of lean tissue could become more relevant when comparing therapies intended for older or otherwise vulnerable populations.
- Payers: Expanding GLP-1 use into broader and older populations may create a case for treatment pathways that incorporate nutritional and functional support rather than financing pharmacotherapy in isolation.
- Industry / innovation partners: “Quality of weight loss” is becoming a development target of its own. New combinations are already being designed not simply to drive the scale lower, but to preserve lean mass while reducing adiposity.
The success of GLP-1-based obesity medicines has made one outcome exceptionally easy to communicate: percentage body-weight loss.
But as treatment effects become larger, that single number is starting to look incomplete.
At the ADA Scientific Sessions in New Orleans this June, clinicians debated whether concern about GLP-1-associated muscle loss is justified. Medscape’s subsequent review captured the emerging consensus reasonably well: most of the weight lost with these medicines is fat, but a proportion is lean tissue, and the clinical significance may differ sharply depending on who is being treated.
The frequently cited STEP 1 body-composition analysis helps explain both the concern and the confusion. Among participants assessed using DXA, semaglutide treatment reduced body weight by 15.0%, total fat mass by 19.3% and lean body mass by 9.7% over 68 weeks. Because fat declined faster than lean tissue, participants ended treatment with a higher proportion of their body weight classified as lean mass and an improved lean-to-fat ratio.
The word lean, however, is doing a lot of work.
DXA does not directly measure skeletal muscle. Its lean-mass compartment also captures water, organs and other non-fat tissue. Tinsley and Heymsfield have argued that some fat-free-mass reduction during major weight loss is effectively unavoidable because adipose tissue itself contains water and protein. Losing adipose tissue therefore removes not only stored lipid but some material that a body-composition scan categorises as fat-free mass.
That is one reason researchers increasingly caution against using “lean mass loss” and “muscle loss” interchangeably.
Nor is loss of fat-free mass unique to pharmacological weight reduction. Dietary interventions and bariatric surgery also reduce lean tissue as body weight falls. A 2026 review of human evidence found variable reductions in skeletal muscle measures during GLP-1 therapy, generally without corresponding deterioration in strength, while noting that evidence specifically in older adults remains limited.
That limitation matters because average trial participants may not represent the patients in whom relatively modest muscle loss has the greatest consequences. An older adult approaching frailty, a patient with sarcopenic obesity or someone beginning treatment with limited functional reserve is not starting from the same baseline as a younger patient with substantial muscle mass.
The question therefore becomes less dramatic, but more clinically useful: who can afford to lose lean tissue, and who cannot?
There is already evidence that the obesity pipeline is responding.
In a 507-participant Phase 2 trial published in Nature Medicine in March, adding the investigational activin-receptor antibody bimagrumab to semaglutide produced greater fat reduction while substantially reducing the decline in lean mass seen with semaglutide alone. The combination remains investigational, but the trial illustrates where drug development may be heading: away from maximising weight loss at any cost and toward optimising its composition.
Clinical practice may need to make the same transition.
For most patients, the available evidence does not justify treating GLP-1-associated weight loss as synonymous with sarcopenia. But for selected patients, monitoring only the scale could miss the outcome that matters more: whether they remain strong enough to climb stairs, maintain balance, preserve independence and benefit from the metabolic improvements the treatment was intended to deliver.
The GLP-1 era has made major weight loss medically achievable.
The next question is whether obesity care can become equally sophisticated about what patients keep while the weight comes off.
