BIOANALYSISPAPER
A peer reviewed review argues the question is how many kilograms of lean (non fat, non bone) mass you lost against your starting baseline — not the percentages dominating public discussion.
For a patient on Ozempic, Wegovy, Mounjaro, or Zepbound, the clinical question has rarely been "how much did you lose?" It has been "how much of what you lost was muscle?" A new peer-reviewed review of body-composition data argues the question is the wrong unit of analysis, and that the public conversation built on top of it has been mismeasuring the problem for years.
The review, indexed in PubMed (PMID 42601996) and available in full through PubMed Central (PMC13473246), examines how body composition changes during therapy with glucagon-like peptide-1 receptor agonists, the drug class most readers know by those four brand names. These drugs were originally developed to regulate insulin and blood sugar in type 2 diabetes. They became public phenomena because patients on them lost substantial weight. The review's argument is that we have been doing the math on that weight loss in the least useful way.
To follow the reframing, two terms need plain definitions. "Body composition" is what your body is made of: fat mass and lean mass, with lean mass further split into muscle, organs, water, and bone. "Lean body mass," in the technical sense the review uses, is everything that is not fat and not bone, so it tracks muscle closely without being a perfect proxy for it. When a GLP-1 drug takes weight off a patient, the question is how that weight divides between fat and lean, and how the lean side of the ledger changes a person's function.
Public discussion has been dominated by the relative share. Headlines and social posts have leaned on the share of total weight lost that came from lean mass, often in the 25 to 40 percent range cited from older trials of caloric restriction. The review's point is that the share is the number that misleads. A patient who loses 30 kilograms will shed more lean mass in absolute terms than a patient who loses 10 kilograms, even if the share of lean mass in each loss is identical. The clinically meaningful number is kilograms of lean mass against the patient's starting baseline, set against a clinical outcome the patient can feel: grip strength, gait speed, mobility, metabolic markers, or the ability to recover from illness.
Randomized controlled trials cited in the review report that tirzepatide and semaglutide, the two leading molecules in the class, can produce a reduction in lean body mass of roughly 2.3 to 5.6 kilograms during a treatment course. That range is the absolute number the review argues the conversation should center on, because it is the number that maps onto functional outcomes. The relative share, by contrast, is a function of how much fat the patient also lost, and a patient losing a large share of lean mass can still be better off in absolute clinical terms if the absolute loss is modest and the fat loss is large.
The reframing matters because the public has been carrying a generic muscle-loss panic that the underlying trials do not actually support as written. The trials do not show catastrophic muscle loss on these drugs. They show measurable lean-mass reduction, on the order of a few kilograms, in patients who are simultaneously losing tens of kilograms of fat. The review's authors call the field's current framing "discrepant" and argue for balancing body-composition changes against metabolic-health outcomes rather than treating any lean-mass loss as a failure of the drug.
For a reader, the practical change is small but durable. When a future headline says "GLP-1 patients lose 30 percent of their weight from muscle," the question to ask is not whether the percentage is true. It is how many kilograms of lean mass that percentage represents, against what starting baseline, in a patient who started at what body composition, and what functional outcome the trial measured at the end. The percentage is a ratio between two unknowns. The kilograms are the number a clinician acts on.
The review does not specify every outcome threshold, and any patient decision belongs with a clinician who has the full chart. What it does is give readers a better question to bring into the room, and a more honest unit of account for the next round of coverage.