Study Finds GLP-1 Weight-Loss Drugs May Cause Less Muscle Loss Than Previously Believed

Since semaglutide and tirzepatide became household names, one concern has trailed behind the weight-loss headlines: that these drugs eat muscle along with fat, leaving patients lighter but weaker. A study published in May 2026 in Cell Reports Medicine makes one of the most direct cases yet that this fear has been overstated [1]. But the question of whether GLP-1 receptor agonists are truly safe for long-term musculoskeletal health depends on who is taking them, how muscle is measured, and what counts as "safe enough" — none of which this study fully resolves.

The New Study: What It Found

The paper, led by biologist Henning Langer of Charité Berlin and Keith Baar, a professor of molecular exercise physiology at the University of California Davis, combined preclinical mouse data with a proof-of-concept human trial [1]. In obese mice, GLP-1 medicines predominantly reduced body fat, with only a small decrease in lean body mass. Critically, the researchers found that loss of liver mass exceeded the change in skeletal muscle mass — a finding that reframes earlier body-composition data, which lumped all lean tissue (including organ mass and water) into a single "lean mass" category [1][2].

In the human component, participants with obesity who took GLP-1 receptor agonists showed improved body composition without negative effects on muscle strength. While absolute muscle values declined modestly, relative muscle mass and strength improved, and running performance was better [1].

The authors' central claim: the lean mass loss seen in GLP-1 trials is not a sign of pathological muscle wasting, but an adaptive response — the body appropriately downsizing to match a lighter frame [3].

The Study That Started the Alarm

The muscle-loss narrative traces back largely to the STEP 1 trial, a pivotal study of semaglutide 2.4 mg that helped win FDA approval for Wegovy. An exploratory body-composition analysis of 95 participants in STEP 1 found that lean soft tissue accounted for roughly 40% of total weight lost — about 6.9 kg of lean mass alongside 10.4 kg of fat mass over 68 weeks [4]. That 40% figure became a touchstone for critics of GLP-1 drugs, repeated widely in medical commentary and fitness media.

But that analysis had limitations. Body composition was measured by DEXA (dual-energy X-ray absorptiometry), which cannot distinguish skeletal muscle from organ tissue, water, or connective tissue [4]. The results were explicitly labeled as exploratory and were not corrected for multiple comparisons. And STEP 1 did not assess functional outcomes like grip strength or physical performance [4].

How Does GLP-1 Muscle Loss Compare?

The comparison that matters most — and the one least often made — is against other weight-loss methods. Caloric restriction, the oldest and most common approach, typically produces lean mass loss of about 25% of total weight lost. Bariatric surgery runs around 25–30%. Among GLP-1 drugs, the range varies: tirzepatide trials have reported lean mass loss at about 25% of total weight lost, semaglutide at roughly 39%, and retatrutide at 33% [5][6].

Source: Neeland et al. 2024, Diabetes Obesity & Metabolism; STEP 1 Exploratory Analysis

Data as of May 7, 2026CSV

A 2024 review in Diabetes, Obesity and Metabolism by Neeland and colleagues examined lean body mass changes across GLP-1 therapies and found that reductions in lean mass accounted for approximately 12–40% of total weight loss, with many studies reporting values toward the lower end [5]. The review also noted that 68% of GLP-1 users in one analysis exceeded preset benchmarks for "disproportionate" lean mass loss — but 50% of people who lost weight through lifestyle interventions alone exceeded the same thresholds [7]. In other words, some lean mass loss is a normal feature of losing weight by any means.

A Circulation editorial made this point explicitly, arguing that GLP-1-associated lean mass reductions appear proportional to total weight loss and may represent a physiologic adaptation rather than a harmful side effect [3].

The SEMALEAN Study: Functional Data

The most granular functional data comes from the SEMALEAN study, a prospective trial of 115 patients with obesity taking semaglutide 2.4 mg, published in Diabetes, Obesity and Metabolism [8]. Researchers measured body composition via DEXA and assessed handgrip strength and resting energy expenditure at baseline, 7 months, and 12 months.

Results showed mean weight loss of 13% at 12 months. Lean mass decreased initially at 7 months but stabilized through month 12. Handgrip strength actually improved by 4.5 kg at 12 months. And the prevalence of sarcopenic obesity — the dangerous combination of low muscle mass and high body fat — fell from 49% at baseline to 33% at 12 months [8].

These findings suggest that, at least in a structured clinical setting, semaglutide can improve the ratio of muscle to fat and increase measured strength, even as absolute lean mass declines. But the SEMALEAN study tracked patients for only 12 months, leaving open the question of what happens over years of use.

The Measurement Problem

Much of the disagreement in this field comes down to measurement. DEXA scans — the most commonly used tool in GLP-1 trials — measure "lean soft tissue," a category that includes skeletal muscle but also organ mass, water, and glycogen stores [4]. When someone loses weight, they lose intramuscular and hepatic glycogen (which binds water), reducing measured lean mass even if contractile muscle fibers are intact.

The Langer and Baar study found that in mice, liver mass loss was the dominant contributor to the measured lean mass decline, not skeletal muscle [1]. If this holds true in humans, much of the reported 40% lean mass loss from trials like STEP 1 reflects organ and fluid shifts, not actual muscle wasting.

Bioelectrical impedance analysis (BIA), used in some studies, is even less precise than DEXA and can be confounded by hydration status [9]. MRI and CT scans can isolate skeletal muscle from other tissues but are expensive and rarely used in large trials. This measurement gap means the field has been arguing about a number — percent of lean mass lost — that may not reflect the clinical outcome patients and doctors actually care about: whether the person can walk, lift, and function.

Who Is at Risk: The Older Adult Question

The reassuring findings from SEMALEAN and the Langer-Baar study carry a significant caveat: their participants were generally middle-aged adults in structured clinical settings. The populations for whom muscle loss is most dangerous — older adults over 65, people with pre-existing sarcopenia, and patients not engaged in exercise programs — remain underrepresented in the evidence base [10].

Natural aging reduces skeletal muscle mass by 12–16% in adults over 65, and up to half of adults over 80 experience sarcopenia [10]. An editorial in Annals of Internal Medicine warned that GLP-1 receptor agonists could exacerbate age-related muscle decline in seniors, potentially pushing vulnerable patients into frailty [10]. The editorial called for baseline assessment of muscle mass and physical function before prescribing GLP-1 drugs to older patients — a recommendation that is not yet standard in prescribing guidelines.

Roughly 9 million Americans are currently prescribed GLP-1 drugs [11], and Medicare will begin covering Wegovy and Zepbound starting July 1, 2026, under the Treat and Reduce Obesity Act [12]. That coverage expansion will bring a surge of older patients onto these medications — many of whom may not have access to the resistance training programs and high-protein diets that studies suggest can mitigate lean mass loss.

The Obesity Context

The debate over GLP-1 muscle loss unfolds against a background of rising obesity rates worldwide. The United States leads major economies with an adult obesity prevalence of 42%, followed by South Africa at 30.8% and Australia at 30.2% [13].

Source: WHO Global Health Observatory

Data as of Dec 31, 2022CSV

For patients with BMIs above 30 or 35, the health risks of obesity — type 2 diabetes, cardiovascular disease, joint damage, sleep apnea — are well established. The relevant clinical question is not whether GLP-1 drugs cause any muscle loss, but whether the net health benefit of significant fat reduction outweighs the cost of modest lean mass decline. For most obese patients, obesity medicine specialists argue, the answer is clearly yes [3][7].

Funding and Conflicts of Interest

The full conflict-of-interest disclosures for the Langer-Baar study were not accessible outside the journal's paywall at the time of this reporting. The study was published in Cell Reports Medicine, a peer-reviewed Cell Press journal [1].

On the other side of the debate, the STEP 1 trial that generated the 40% lean-mass-loss figure was funded by Novo Nordisk, the maker of Wegovy and Ozempic [4]. The SEMALEAN study did not disclose pharmaceutical industry funding [8]. The 2024 Diabetes, Obesity and Metabolism review that synthesized lean mass data across trials included authors who disclosed consulting relationships with both Novo Nordisk and Eli Lilly [5].

This creates a situation where research on both sides of the muscle-loss question has potential financial entanglements with the companies that stand to gain or lose from the narrative.

The Bias Question: Was the Alarm Always Overblown?

There is a credible argument that the fitness and medical establishment amplified GLP-1 muscle-loss concerns beyond what the evidence supported. Yoni Freedhoff, an obesity medicine physician, has called the muscle-loss panic a "manufactured controversy" [2]. The argument runs as follows: weight loss by any method produces lean mass loss, bariatric surgery has caused similar or greater lean mass loss for decades without generating equivalent alarm, and the outsized attention to GLP-1 muscle effects reflects a broader cultural discomfort with pharmaceutical approaches to weight loss.

The academic research trend supports the idea that this topic attracted disproportionate attention. Over 25,000 papers related to GLP-1 and muscle loss have been published, with the volume peaking at 4,476 papers in 2025 alone [14].

Source: OpenAlex

Data as of Jan 1, 2026CSV

Critics of this view counter that the scale of GLP-1 prescribing is unprecedented — millions of patients, many of them not medically monitored — making caution appropriate even if the per-patient risk is modest.

Beyond the Scale: Functional Outcomes That Matter

Lean mass expressed in kilograms tells only part of the story. The outcomes that determine whether muscle loss is clinically meaningful include grip strength (a predictor of mortality in older adults), fall risk, bone mineral density, and resting metabolic rate.

On bone density, a study of semaglutide found a 2.6% loss at the total hip and 2.1% at the lumbar spine at 52 weeks [15]. This is driven partly by reduced mechanical loading from lighter body weight and partly by inadequate calcium and protein intake. On grip strength, the SEMALEAN data showed improvement rather than decline at 12 months [8]. On metabolic rate, the Harvard Science Review noted that rapid skeletal muscle loss can reduce basal metabolic rate, increasing the risk of fat regain if the medication is discontinued [16].

No published GLP-1 trial has tracked fall incidence as a primary outcome. The LEAN-PREP trial (NCT06885736), currently enrolling, will be the first randomized controlled trial to test resistance training as a specific countermeasure to GLP-1-related lean mass loss [11].

Market and Policy Implications

The GLP-1 drug market has grown from $18.2 billion in 2021 to an estimated $58 billion in 2026, with projections reaching $185 billion by 2033 [17]. Eli Lilly now holds approximately 57% of the U.S. GLP-1 market, overtaking Novo Nordisk [18]. Both companies have launched oral GLP-1 formulations in 2026, further expanding the potential patient base [18].

Source: Towards Healthcare; Grand View Research

Data as of May 7, 2026CSV

If the muscle-loss narrative fades, the primary beneficiaries are the drug manufacturers and the patients who might have been deterred from trying these medications. Insurers, who have resisted covering GLP-1 drugs partly by citing incomplete safety profiles, would face increased pressure to expand coverage. Bariatric surgeons, whose procedures produce comparable lean mass loss but attract far less scrutiny, would see a competitive disadvantage deepen.

Medicare's decision to cover GLP-1 drugs starting July 2026 was driven primarily by cardiovascular benefit data, not body-composition outcomes [12]. But the muscle-loss question could influence how aggressively these drugs are prescribed to older Medicare beneficiaries, and whether coverage will require concurrent physical therapy or exercise programs.

What Would It Take to Close the Debate?

Endocrinologists and geriatricians interviewed in recent coverage have outlined what a definitive answer would require: randomized controlled trials of at least 24 months, using MRI or CT to isolate skeletal muscle (not DEXA lean mass as a proxy), in populations that include adults over 65, patients with pre-existing sarcopenia, and participants not enrolled in exercise programs [10][11]. The studies would need to track functional outcomes — grip strength, gait speed, fall incidence, bone density — not just mass on a scan.

No such trial has been completed. The Langer-Baar study advances the argument that muscle loss is less alarming than feared, but it does not meet this evidentiary bar. Its human component was small, its mouse data — while informative about mechanisms — cannot be directly extrapolated to clinical practice, and it did not follow high-risk populations.

The current evidence supports a qualified reassurance: for middle-aged adults with obesity who exercise and maintain adequate protein intake, GLP-1 drugs do not appear to cause disproportionate or functionally meaningful muscle loss. For older adults, sedentary patients, or those on long-term therapy without structured support, the data remains insufficient to make that claim with confidence.