Clinical obesity has been recently defined as a condition of illness that directly results from the effect of excess adiposity on the function of organs and tissues [1]. Weight loss is the primary treatment goal of obesity management because it can improve or resolve many obesity-related complications. People with clinical obesity should receive timely, evidence-based weight-loss treatment, with the aim to induce improvement (or remission, when possible) of clinical manifestations of excessive adiposity (such as type 2 diabetes or T2D) and prevent progression to end-organ damage (for instance, heart failure or renal insufficiency) [2].

Major weight loss resulted in fat mass reduction, an expected benefit, but also in deleterious loss of skeletal muscle mass (SMM) [3]. Maintaining or even increasing SMM during weight reduction is important to overall health, metabolic function, and long-term weight loss maintenance [[4], [5], [6]]. In pooled data from seven prospective cohorts, fat mass and fat-free mass (FFM) showed opposing associations with mortality, excess fat mass being related to increased mortality risk, whereas FFM (mainly SMM) being associated with a lower risk of mortality [7]. Thus, clinicians should consider possible changes in body composition when developing an obesity treatment plan for an individual patient. Especially, taking into account the effects of obesity pharmacotherapy on FFM and SMM appears to be a key objective [5,6].

Sarcopaenia is defined as a generalised skeletal muscle disease, combining both low SMM and reduced muscle strength [8,9]. This complication has been neglected or at least underestimated for a long time in clinical practice, notably among people with overweight or obesity despite the existence of sarcopaenic obesity. Several pharmacological candidates to prevent or treat sarcopaenia are currently under clinical evaluation with some promising early results, but none have been approved for either frailty or sarcopaenia [10]. Sarcopaenic obesity is characterised by a concurrent decline in SMM and function, along with increased adipose tissue [11]. Sarcopaenic obesity is a growing concern, especially among older adults, owing to its association with multiple important health consequences [12,13]. Of note, weight loss therapies could increase the risk of sarcopaenia in adults with obesity [3,14]. If so, the potential health benefits of weight loss therapies might be compromised by the weight loss-associated loss of FFM/SMM, which could increase the risk of sarcopaenia, at least in high risk patients (elderly, frailty with comorbidities) [4]. Furthermore, mounting evidence suggests that the prevalence of sarcopaenia is higher in patients with T2D, especially in older adults, and different mechanisms may be responsible for this association [[15], [16], [17]]. In this context, emerging novel pharmacological agents that were developed for treating T2D commonly associated with overweight/obesity or clinical obesity independently of the presence of T2D may be relevant to sarcopaenia-related health concern [13,18].

When considering the administration of glucose-lowering agents for the management of T2D, it is potentially important to understand not only their weight loss effect but also the degree of SMM loss caused by each drug in clinical practice [[19], [20], [21]]. Drugs currently used to treat T2D may have different mechanisms of action that are relevant to the prevention and treatment of sarcopaenia, for those with T2D but also for those with obesity without diabetes [22]. Glucagon-like peptide-1 receptor agonists (GLP-1RAs) (liraglutide, semaglutide) and the co-agonist GIP (glucose-induced insulinotropic polypeptide)/GLP-1 receptor agonist tirzepatide are now recognised as efficient drugs not only for the management of T2D by combining improved glucose control and weight reduction [23,24], but also for the treatment of clinical obesity in association with reinforcement of lifestyle measurements (healthy well-balanced calorie-restricted diet and regular physical exercise) [25,26]. However, the effects of these agents on muscle mass and function remain controversial [[27], [28], [29]].

The aim of this comprehensive review is to analyse the effects of GLP-1RAs and the dual agonist tirzepatide on the body composition (especially FFM), notably muscle mass and function, by considering and comparing animal findings in rodents and human studies (both randomised controlled trials [RCTs] and observational studies). Whether these incretin-based therapies may be associated with a higher risk of sarcopaenia or in contrast would protect against sarcopaenia within a weight-loss strategy remains indeed a matter of discussion [[27], [28], [29]] (Fig. 1).

A literature search in PubMed, Scopus, Embase and the Cochrane Database of Systematic Reviews has been performed to identify English-language studies published between January 2010 and May 2025. In a first step, the terms used for the research were “GLP-1 receptor agonists” combined with “body composition” OR “fat-free mass” OR “skeletal muscle” OR “sarcopenia”. In a second step, the term “GLP-1 receptor agonists” was divided into the three GLP-1-derived medications that received the indication for the management of obesity, i.e. liraglutide, semaglutide and tirzepatide. The reference lists of previously published reviews regarding this topic were also scrutinised to identify any further reports of potential interest.

Intentional weight loss in people with obesity primarily decreases body fat, but also reduces FFM. SMM accounts for about 50 % of FFM, whereas the remainder amount is composed of the FFMs of other fluids, organs, and body tissues. As recently emphasised [30], bringing body composition science into the modern era mandates the use of the chemically correct term FFM with the elimination of the duplicative term “lean body mass” (LBM) that today has value primarily in a historical context (thus, FFM was used instead of LBM throughout this article). A number of technologies exist to measure body composition, each with advantages and disadvantages. The most relevant ones are hydrostatic weighing (densitometry) and air displacement plethysmography, bioelectrical impedance analysis (BIA), dual-energy X-ray absorptiometry (DEXA), computed tomography (CT) and magnetic resonance imaging (MRI) [14,31,32].

A widely cited rule guiding expected loss of lean tissue states that approximately one-fourth of weight loss will be FFM. However, this rule has been controverted, depending on how FFM has been measured [33]. In addition, this fraction may be influenced by initial body composition and total weight loss [33]. Furthermore, the rate of weight loss [34], the degree and type (protein content) of caloric restriction, and the impact of concomitant physical exercise also influence the proportion of weight lost as FFM (SMM) after dietary interventions [35]. There is consensus that FFM and SMM can be preserved, albeit to varying degrees, by including both endurance and resistance forms of exercise (but especially resistance) in every weight management intervention. Higher intakes of protein can also protect loss of these body compartments, acting either separately or synergistically with exercise [3,[36], [37], [38]].

Bariatric surgery is the weight-loss approach that induces the most rapid and important weight reduction associated with changes in body composition. In a recent study, at 12 months post-surgery, Roux-en-Y gastric bypass and sleeve gastrectomy resulted in comparable weight loss and changes in body composition assessed by DEXA (reduction in FFM by almost 18 %), yet improvements in relative muscle strength (handgrip, sit-to-stand test) and physical function (6-minute walking test) were noticed [39]. However, another meta-analysis of 24 articles (666 participants), showed that bariatric surgery reduces absolute lower-limb isometric strength, directly related to body weight loss, and also diminishes absolute handgrip strength [40].

Of note, changes in FFM should not be directly conflated with changes in SMM (see below, discussion). Additionally, the potential for obligatory loss of FFM from adipose tissue should be considered when interpreting FFM changes with large magnitudes of weight loss, such as those frequently obtained with GLP-1RAs (especially semaglutide) and co-agonist tirzepatide to be compared with those achieved with bariatric surgery [14,18,36].

In a systematic review and network meta-analysis of 22 RCTs (2258 participants), GLP-1RAs significantly reduced (mean difference [95 % confidence interval]) total body weight (−3.55 [−4.81;−2.29] kg), fat mass (−2.95 [−4.11;−1.79] kg), and lean body mass (−0.86 [−1.30;−0.42] kg), with FFM loss comprising approximately 25 % of the total weight loss [41]. However, the relative lean mass, defined as percentage change from baseline, was unaffected. Of note, potent GLP-1 RAs, such as tirzepatide and semaglutide, demonstrated greater overall weight loss but were associated with a significant reduction in FFM, in contrast to liraglutide [41].

In another meta-analysis of 19 RCTs, greater reductions in FFM (weighted mean difference [95 % confidence interval]) were also observed in GLP-1RA users compared with non-users: −1.02 [−1.46;−0.57 kg]. Again, the changes in FFM percentage were comparable between GLP-1 receptor-based agonist users and non-users [42].

In a meta-analysis of six RCTs in people with overweight and obesity, tirzepatide was associated with a significant reduction in total fat mass and visceral adipose tissue, whereas the effect on FFM was considered as uncertain because the findings remained inconclusive [43]. In a substudy of SURMOUNT 1 among adults with obesity or overweight who underwent DEXA, the changes in body weight, fat mass and FFM from baseline to week 72 was −21.3 %, −33.9 % and −10.9 % with tirzepatide, respectively (compared with −5.3 %, −8.2 % and −2.6 % with placebo, respectively) [44].

In a review of studies reporting the effects of GLP-1RAs and sodium-glucose cotransporter 2 (SGLT2) inhibitors on body composition, no clear differences existed between the two pharmacological classes when considering the reduction in FFM. In over half of the studies identified, the proportion of FFM reduction ranged between 20 % and 50 % of total weight lost, which is consistent with diet-induced weight loss and bariatric surgery [45]. An even higher heterogeneity in the reported effects of GLP-1-based therapies on FFM changes was noticed in another report: in some clinical trials, reductions in FFM ranged between 40 % and 60 % as a proportion of total weight lost, while other studies showed FFM reductions of only about 15 % or less of total weight lost [46]. There are several potential reasons to explain the heterogeneity in the reported effects of GLP-1RAs on FFM changes in clinical trials. These may include the specific effects of different molecules (for instance liraglutide versus semaglutide versus tirzepatide), the heterogeneity in dosing leading to different weight loss kinetics (for instance 1.0 versus 2.4 mg for semaglutide), varying duration of follow-up (< one year versus > two years), different population characteristics (e.g., with versus without diabetes, variable proportion of men versus women, different age groups), interference of concomitant lifestyle interventions (diet and exercise) prescribed with the pharmacological therapy, and finally possible methodological heterogeneity and bias in FFM and SMM assessments, as further analysed below (see answer to the first question in the discussion).

The impact of GLP-1RA-based treatments on FFM has been carefully analysed in 28 clinical trials that used DEXA for measuring FFM [47]. The percentage of FFM loss using GLP-1RA-based agents ranged between 20 % and 40 %. However, this review was limited to small substudies. Furthermore, DEXA does not measure SMM directly (FFM should be considered only a crude surrogate for SMM, because FFM may contain a variable amount of muscle, approximately 55 %). Assessing quantity and quality of skeletal muscle using more accurate advanced imaging techniques (such as MRI) combined with functional testing should help fill the gaps in our current understanding about the impact of GLP-1-based therapies on FFM and SMM [47]. It has been proposed that tailored resistance exercise training should be recommended as an adjunct to incretin therapy to optimise changes in body composition by preserving SMM while achieving fat loss, and thus avoiding sarcopaenia [48].

Besides SMM, other muscle properties appear to be critical, especially muscle strength, function, and quality. While SMM (or at least its proxy FFM) has been widely assessed in human studies with GLP-1RAs, it was less the case for the evaluation of muscle strength and function. The study of skeletal muscle structure and quality requires biopsies, not available in human studies with GLP-1-derived therapies yet. Thus, specific observations about skeletal muscle are mostly derived from animal models, especially in rodents (mice and rats).

Initial animal studies that investigated the effects of GLP-1RAs on the skeletal muscle were primarily interested in the improvement of insulin sensitivity [49]. It has been shown that exenatide ameliorates intramyocellular lipid deposition in ob/ob mice and in diet-induced obese mice, an effect associated with the activation of AMP-activated protein kinase (AMPK) signalling pathway and improvement in insulin sensitivity, independent of weight loss [50]. Numerous more recent rodent studies have shown a positive impact on skeletal muscle quality and function [51]. These effects were demonstrated with different GLP-1RAs: exenatide [[52], [53], [54]], liraglutide [[55], [56], [57], [58], [59], [60]], semaglutide [55,[61], [62], [63]], dulaglutide [54,64]. No such studies were carried out with the dual GIP/GLP-1 agonist tirzepatide yet.

Several mechanisms have been proposed to explain a protective effect of GLP-1RAs on SMM and muscle function in rodents and thus contributing to mitigate sarcopaenia. Let us cite a few experimental results of potential interest, successively with exendin-4, liraglutide, semaglutide, and dulaglutide. In skeletal muscle cells of obese nondiabetic fa/fa Zucker rats, exendin-4 strongly restructured the extracellular matrix and reinforced muscle contractile capabilities, while optimizing the cellular metabolism through the AMPK signalling pathway [52]. GLP-1RAs (exendin-4) ameliorated muscle wasting by suppressing myostatin and muscle atrophic factors and enhancing myogenic factors through GLP-1R-mediated signalling pathways [54]. Using a streptozotocin-induced diabetic mouse model, exendin-4 counteracted diabetes-induced muscle weight loss, weaker grip, and changes in muscle fibre cross-sectional area distribution, thus globally improving muscular dysfunction [53]. While high glucose promotes muscle cell ageing and sarcopaenia, liraglutide attenuated these processes by modulating the YAP/TAZ signalling, a critical ageing pathway [60]. In another experimental study, liraglutide-treated KK-A y mice showed, despite final body weight loss, significant increase in SMM counteracting diabetic muscular atrophy, which coincided with a significant decrease in the expression levels of ubiquitin protease E3 MuRF1 and MAFbx [57]. Liraglutide exerted protection and restored myofibrillar architecture in diverse models of muscle atrophy in Sprague Dawley rats. Anti-atrophy actions of liraglutide involved suppression of atrogene expression and enhancement in expression of myogenic factors [59]. In in vitro studies in healthy C2C12 mouse skeletal muscle cells, liraglutide at an appropriate dose enhanced myotube differentiation and muscle contractile activity upon electric pulse stimulation; these positive changes included the promotion of muscle hypertrophy [56]. Liraglutide led to greater citrate synthase activity and cytochrome c oxidase subunit 5 B protein expression in the soleus of spontaneously diabetic torii fatty rats, independently of blood glucose. These findings suggest that the GLP-1RA may contribute to preservation of mitochondrial content on soleus muscle in a model of T2D [58]. A study that investigated the effects of liraglutide and semaglutide on obesity-induced muscle atrophy concluded that these GLP-1RAs protect skeletal muscle against obesity-induced muscle atrophy via the SIRT1 pathway [55]. Semaglutide significantly reduced the accumulation of intramuscular fat in the gastrocnemius, promoted muscle protein synthesis, increased the relative proportion of skeletal muscle (increased type I/type II muscle fibre ratio, total muscle fibre area, muscle fibre density, sarcomere length, mitochondrial number, and mitochondrial area), and improved muscle function of obese mice, possibly by altering the metabolism of muscle lipids and organic acids [61]. Semaglutide effectively inhibited psoas muscle atrophy and suppressed declines in grip strength in a diet-fed diabetic KK-A(y) mouse model. This effect of semaglutide on skeletal muscle atrophy was mediated by multiple functional pathways, notably associated with decreased proinflammatory cytokines and reduced accumulation of reactive oxygen species [62]. Treatment with a long-acting GLP-1RA, dulaglutide, recovered muscle mass and function in DBA/2J-mdx mice, a Duchenne muscular dystrophy model [54]. In db/db mice, high glucose inhibited the differentiation of C2C12 cells and decreased the mRNA and protein levels of myokines whereas dulaglutide could reverse the differentiation state induced in C2C12 cells by high glucose [64]. These findings suggest that a treatment with the GLP-1RA protects db/db mice against skeletal muscle injury by inhibiting inflammation and regulating the differentiation of myoblasts [64]. Another study showed that dulaglutide improved muscle mass and strength in aged mice by attenuating the expression of inflammatory cytokines. These observations suggest that the GLP-1RA may exert beneficial effects in the treatment of muscle wasting due to ageing [65]. Finally, dulaglutide inhibited the decrease of muscle fibre size and the expression of atrophic factors in a model of disuse muscle atrophy in C57BL/6 mice. These beneficial effects against disuse-induced muscle atrophy were mainly attributed to the inhibition of inflammation and apoptosis by induction of heat shock protein 72 expression via the regulation of AMPK signalling [66].

In contrast to all positive abovementioned studies, an investigation using a cell model revealed that GLP-1 dose-dependently suppressed the expression of the myogenic marker, impeding myocyte fusion and the formation of polarised myotubes during differentiation [67].

Of potential interest, even if still preliminary, blockade of activin type II receptors (ActRII) signalling using bimagrumab, a monoclonal antibody against ActRII, improved body composition (superior fat mass loss while simultaneously preserving lean mass) and metabolic parameters (associated with improved exercise performance) during calorie deficit driven by semaglutide in diet-induced obese mice [68].

Finally, according to a systematic review of studies that investigated the effects of GLP-1RAs on mitochondrial function within skeletal muscle, the emerging evidence derived from animal and in vitro models suggested that GLP-1RAs increase mitochondrial area and number while improving mitochondrial morphology (i.e., reduced swelling), but effects upon other aspects of mitochondrial health remain inconclusive [69]. It has been concluded that a protective effect on skeletal muscle by GLP-1-based therapies may result from a reduction in inflammation and apoptosis [[70], [71], [72]].