Preserving muscle during weight loss with GLP-1 RAs

GLP-1 receptor agonists (RAs) such as semaglutide (e.g. Ozempic), tirzepatide (e.g. Zepbound), and liraglutide lower body weight by reducing energy-intake, but a portion of the loss typically comes from lean tissue (i.e. muscle) as well as adipose (fat). That tradeoff matters: skeletal muscle underpins glycemic control, mobility, and long-term weight-loss maintenance, and so losing muscle alongside fat can blunt the very metabolic and functional gains patients seek. The question is how to make weight loss more "fat-selective" -- maximizing adipose loss while preserving, or even improving, muscle.

This is especially a concern because the amount of weight loss can be substantial. Rough ballpark estimates from obesity trials show that semaglutide can lead to about 15% mean weight loss at ~68 weeks (STEP-1), while tirzepatide 10–15 mg can yield roughly 21–23% weight loss at ~72 weeks (SURMOUNT-1).

One practical answer is to combine a GLP-1 RA with structured exercise. In a randomized, placebo-controlled trial published in the New England Journal of Medicine (NEJM), adults with obesity who completed an 8-week low-calorie diet were assigned for 52 weeks to exercise, liraglutide (an older generation GLP-1 RA), both, or placebo. The idea of starting the trial with the diet before the treatments was to see how well the treatments could keep the weight/fat off. 

Each participant in an exercise group was assigned an instructor to supervise their activity programs. In addition, they were invited to join group exercise sessions. The combination arm achieved superior "healthy weight loss maintenance," including greater reductions in body-fat percentage and improved cardiorespiratory fitness, outperforming either component alone.

Figure 1 shows that after the 8-week diet, the subjects lost a substantial amount of body fat. The placebo control group regained some of that fat over the 52 week trial period. Those in the exercise alone or liraglutide alone groups, continued to lose body fat. However, the combination exercise and liraglutide group clearly did the best, losing the most weight (not shown) and the most body fat (Figure 1).

In an off-treatment extension, participants who had been on the combination maintained more of their lost weight and body-fat reduction one year after stopping therapy compared with those previously on liraglutide alone, underscoring that exercise layered onto GLP-1 RA not only enhances the quality of weight loss during treatment but also makes the results more durable after treatment ends.

A second, mechanistically complementary approach to preserving muscle during weight loss is to pair GLP-1 RAs with inhibitors of the myostatin–activin axis. Myostatin (GDF-8) and activin A are members of the TGF-β superfamily, which signals through activin type-II receptors (ActRIIA/B), recruiting ALK4/5/7 kinases to activate Smad2/3 transcriptional programs that suppress muscle protein synthesis and promote proteolysis. Blocking this pathway "releases the brake," allowing muscle hypertrophy (growth) in adult tissue. Reviews of the pathway and translational work converge on the same principle: attenuating ActRII–Smad2/3 signaling increases muscle mass and can prevent or reverse wasting.

Recent data (without GLP-1 RA) show that this biology can be leveraged to re-partition body composition. In a 48-week, phase-2 randomized clinical trial in adults with type 2 diabetes and overweight/obesity, the anti-ActRII monoclonal antibody bimagrumab reduced total body fat mass by about 20.5% while increasing lean mass by 3.6%, with parallel improvements in waist circumference and glycemia versus placebo. That profile -- less fat, more muscle -- demonstrates that ActRII blockade can tilt weight change toward healthier composition, addressing the concern of drug-induced weight loss that sheds lean tissue.

Early reports now suggest that combining ActRII-pathway inhibitors with a GLP-1 RA may make weight loss even more fat-selective than GLP-1 therapy alone. However, clinicians should await full peer-review before adopting such combinations outside trials. What is already actionable is the exercise pairing: a supervised, progressive program as described above. It should be acknowledged that the exercise component was substantial, blending aerobic intervals with resistance/circuit work, performed at least 3-4 days per week and paired with adequate protein under the supervision of an instructor.

In summary, the field is moving from counting pounds to auditing tissues and function. Future obesity trials should elevate body composition (less fat/more muscle) and functional endpoints (strength and cardiorespiratory fitness) to co-primary status, standardize exercise within pharmacotherapy protocols, and power randomized studies of GLP-1 RA plus ActRII-axis inhibition that report lean mass, strength, and visceral-fat outcomes alongside safety. Until those data arrive, pairing GLP-1 RAs with structured exercise is the most evidence-based way to keep what matters -- muscle and function -- while losing what harms -- excess adipose.

Figure 1. Changes in body-fat percentage during 52-week trial period. Subjects underwent low-calorie diet for 8 weeks (-8 to 0 weeks), and then were randomly divided into 4 groups (week 0): placebo control, exercise alone, liraglutide alone, and the combination of liraglutide and exercise. After 52 weeks body fat percentage (y-axis) was measured and compared to week 0. Bars indicate standard error. Figure reproduced from Lundgren et al. NEJM, 2021.