How Much Muscle You Lose on Ozempic, Wegovy, and Zepbound — and What to Do About It

When people lose weight on semaglutide (Ozempic, Wegovy) or tirzepatide (Zepbound, Mounjaro), the weight coming off is not all fat. Body composition studies from the STEP and SURMOUNT trials, plus separate DXA-scan analyses done by academic groups, consistently find that roughly 25% to 40% of the weight lost is lean tissue — muscle, connective tissue, and organ mass — rather than fat.

That ratio is close to what people lose during any form of rapid weight loss, including bariatric surgery and very-low-calorie diets. Some muscle loss is expected. The debate is whether the ratio on GLP-1s is worse than what a person would lose eating the same calorie deficit without the drug, and what the loss means for people who plan to stay on the drug for years.

Early answers: the ratio appears similar to diet-alone weight loss, but two things make it worth paying attention. First, GLP-1 users tend to eat less overall, which usually means less protein — and low protein intake accelerates muscle loss. Second, GLP-1s reduce appetite so effectively that many users skip resistance training out of low energy or low hunger, which is the single most protective activity against muscle loss during weight loss.

A new drug class is trying to fix this. Enveda Biosciences reported positive Phase 1 results on August 18, 2026 for ENV-308, an oral pill designed to mimic Lac-Phe (a molecule the body releases during intense exercise). In animal studies, ENV-308 preserved lean muscle during weight loss and prevented weight regain after weight-loss drugs were stopped. The Phase 1 trial in 88 healthy volunteers found the drug safe and well-tolerated, and Enveda is planning Phase 2 trials to test whether it helps people keep weight off after stopping GLP-1s.

ENV-308 is at least two years from potential approval. This piece explains what current GLP-1 users can do right now — resistance training doses, protein targets, monitoring approaches — while the next generation of muscle-preserving drugs works its way through the clinic.

Body composition data on GLP-1s comes from a few main sources: substudies within the large registrational trials (STEP-1 for semaglutide, SURMOUNT-1 for tirzepatide), academic-run DXA-scan (dual-energy X-ray absorptiometry) studies at Yale and Pennington Biomedical Research Center, and post-hoc analyses that reweight trial data by baseline muscle mass.

Semaglutide (Wegovy) body composition. In a STEP-1 substudy of 178 participants who received DXA scans, patients on semaglutide 2.4 mg over 68 weeks lost an average of 15.3 kg total weight. Of that, roughly 10 kg was fat mass and 5 kg was lean mass. That works out to about 33% of the weight lost being lean tissue. The proportion is roughly the same as what a matched calorie-deficit diet without drug produces, though the total weight lost is much greater.

Tirzepatide (Zepbound, Mounjaro) body composition. In a SURMOUNT-1 substudy, tirzepatide 15 mg over 72 weeks produced an average total weight loss of about 24 kg, with fat mass accounting for roughly 17 kg and lean mass roughly 6 kg. The ratio (about 25% lean loss) is slightly better than semaglutide's, which most researchers attribute to the added GIP receptor activity that has direct effects on adipose tissue.

Muscle strength versus muscle mass. Body composition scans measure lean mass; they do not directly measure muscle strength. Several smaller studies have looked at strength and physical function during GLP-1 therapy. Findings are mixed. Some studies show grip strength and 6-minute walk distance drop slightly during rapid weight loss; others show they stay stable or improve as patients carry less weight. The most consistent finding is that older adults (age 65+) lose relatively more muscle function than younger adults during GLP-1 weight loss, even if the absolute lean-mass change looks similar on a DXA scan.

How it compares to bariatric surgery. Gastric bypass and sleeve gastrectomy produce roughly 30% to 35% lean-mass loss as a fraction of total weight loss, similar to semaglutide. Very-low-calorie diets (800 kcal/day meal replacements) produce about 30% lean loss. In other words, the GLP-1 lean-loss ratio is not obviously worse than what anyone loses when they lose weight fast. What's new is the number of people doing it — millions of Americans are now on GLP-1s versus tens of thousands who have bariatric surgery each year.

If some muscle loss is normal during weight loss, why does it matter enough to warrant a whole drug class in development? Three reasons.

Metabolic rate. Muscle burns calories at rest. Lose muscle and your resting metabolic rate drops, which means the number of calories you can eat without regaining weight also drops. Studies of long-term weight-loss maintainers consistently find that people who preserved lean mass during their loss have an easier time keeping weight off. People who lost the same total weight but sacrificed more muscle tend to regain faster and higher.

For GLP-1 users specifically, this matters because most people eventually stop the drug — either because of cost, side effects, plateau, or life change. When they do, whatever metabolic rate they have becomes the new floor. Preserving muscle during the loss phase raises that floor.

Physical function in older adults. For adults over 60, muscle loss during weight loss is a bigger deal than for a 40-year-old. Sarcopenia (age-related muscle loss) is already progressing, and layering rapid muscle loss on top can shift someone from independent daily activities to needing help. The Endocrine Society and several geriatric societies have flagged this concern specifically. They recommend that older adults on GLP-1 therapy receive both nutritional support (adequate protein) and resistance-training guidance, to go alongside the prescription.

Fracture and fall risk. Bone density typically drops alongside muscle mass during rapid weight loss, and some studies have found small but real increases in fracture risk in the first two years of GLP-1 therapy. The mechanism is not fully worked out. The clinical recommendation is straightforward: patients over 60 or with prior fracture history should get baseline and follow-up bone density scans, and should consider weight-bearing exercise as part of the treatment plan.

Weight regain trajectory. Data from the STEP-4 discontinuation trial and follow-up studies of former GLP-1 users show that people who regain weight after stopping tend to regain more fat than lean tissue. In other words, if you lose 20 pounds — 15 pounds fat, 5 pounds muscle — and then regain 15 pounds after stopping, that regain is disproportionately fat. Over multiple cycles of loss and regain, people can end up with lower muscle mass and higher fat mass than when they started, at the same total weight. This pattern is a bigger long-term concern than any single loss episode.

Enveda Biosciences reported positive Phase 1 results for ENV-308 on August 18, 2026. The drug is an oral pill discovered using Enveda's AI-driven natural-product discovery platform, and it is designed to mimic Lac-Phe (short for N-lactoyl-phenylalanine), a compound the body produces during intense exercise.

What Lac-Phe does biologically. Researchers at Stanford and Baylor identified Lac-Phe in 2022 as a signaling molecule that peaks after high-intensity exercise. In animal studies, injected Lac-Phe reduced appetite and produced weight loss, and researchers proposed it as one of several molecules that help explain why intense exercise blunts hunger. The link between Lac-Phe and muscle preservation came later: animals treated with Lac-Phe during calorie restriction lost less lean tissue than animals doing calorie restriction alone.

The Phase 1 trial. Enveda's ENV-308 trial enrolled 88 healthy volunteers and tested single ascending doses plus multiple ascending doses. The primary endpoints were safety and tolerability. Results: ENV-308 was well tolerated with no serious adverse events. Gastrointestinal side effects — the main reason patients stop GLP-1s — were low across all doses tested. The drug also reduced circulating leptin (a hormone that reflects fat mass), suggesting it reaches biologically active levels in humans, though a Phase 1 trial in healthy volunteers cannot demonstrate weight-loss efficacy directly.

What the animal studies showed. In preclinical mouse studies, ENV-308 preserved lean muscle during weight loss and prevented weight regain after weight-loss therapy was stopped. Enveda cites this as the differentiating angle for their Phase 2 program. About 1 in 8 U.S. adults have used a GLP-1 medication, and most stop within a year for cost, side-effect, or plateau reasons. A drug that helps preserve weight loss after stopping a GLP-1 addresses a real problem in the current pipeline.

Timeline. Phase 2 trials are planned to test whether ENV-308 can help people maintain weight after stopping GLP-1s. Phase 2 timelines in obesity typically run 12 to 18 months for enrollment and data readout. Phase 3 would follow, then FDA review. Realistic earliest approval: 2028-2029, and only if Phase 2 shows a substantial signal.

What ENV-308 is not. ENV-308 is a small molecule, not a peptide. It is not a replacement for a GLP-1; the current concept is combination or sequential use. It is not the only drug in development for this indication. Bimagrumab (a myostatin-pathway antibody in Phase 3 with Eli Lilly), amylin agonists like petrelintide (Roche-Zealand), and RNA interference drugs like ARO-INHBE (Arrowhead) are all being tested as muscle-preserving add-ons or alternatives to GLP-1s.

ENV-308 is one of several drugs testing whether the muscle-loss ratio on GLP-1s can be improved. Understanding the pipeline helps set expectations for what will be commercially available and when.

Bimagrumab (Eli Lilly). An anti-activin type II receptor antibody in Phase 3 obesity trials as a combination with tirzepatide (Zepbound). Phase 2 data reported in 2024 showed that bimagrumab plus semaglutide produced substantially higher fat loss and lower lean loss than semaglutide alone. If Phase 3 confirms the pattern, bimagrumab could be the first specifically-approved muscle-preserving GLP-1 add-on. Timeline: 2027-2028.

Amylin agonists (petrelintide, cagrilintide, AZD6234, MET-233i). Amylin is a peptide hormone that acts alongside insulin on food intake and gastric emptying. Amylin analog drugs produce weight loss similar in size to semaglutide but appear to preserve more lean mass in early studies, likely because the appetite reduction is milder and slower. Novo Nordisk's cagrilintide is in the CagriSema combination already under FDA review. Roche/Zealand's petrelintide is entering Phase 3 in late 2026. AstraZeneca's AZD6234 has a Phase 2 readout at EASD 2026 (September 28 to October 2 in Rome).

UCN2 analog HM17321 (Hanmi/Genentech). Genentech licensed a urocortin-2 analog peptide from Hanmi Pharm in August 2026 for $190 million upfront plus up to $2.3 billion in milestones. UCN2 activates a receptor called CRFR2 that appears to promote weight loss while preserving muscle in preclinical studies. HM17321 is in Phase 1. Timeline: earliest 2029-2030 if the mechanism holds up in humans.

RNA interference drugs (Arrowhead ARO-INHBE, ARO-ALK7). Arrowhead's ARO-INHBE targets Activin E, a hormone involved in fat storage. In Phase 1 combination with tirzepatide, the drug enhanced fat loss and preserved lean mass better than tirzepatide alone. Dosing is quarterly rather than weekly. Phase 2 is ongoing.

The realistic near-term picture. Between now and roughly 2028, no muscle-preserving drug will be approved specifically for that use. The commercial reality for current GLP-1 users is that the drug you're taking is the drug that's available. The behavioral tools described below matter, because pharmacology is not going to save the muscle question in the next 18 months.

Two behaviors preserve muscle during weight loss, and both have strong evidence bases. Neither requires a new drug.

Resistance training, at a dose that matters. Systematic reviews of weight-loss studies consistently find that adding resistance training reduces lean-mass loss by roughly 30% to 50% versus diet alone or diet plus aerobic exercise alone. The dose that shows up in the studies is real: 2 to 3 sessions per week, working most major muscle groups (legs, back, chest, shoulders, arms), 8 to 12 exercises per session, 2 to 3 sets of 8 to 12 repetitions at a weight that is genuinely challenging by the last few reps. This is not walking; it is not yoga; it is not "active recovery" days. Barbells, dumbbells, machines, or bodyweight progressions all work if the load is heavy enough.

For GLP-1 users specifically, energy for training can be a challenge, especially in the first three months when appetite suppression is strongest. Practical approach: schedule training on days when nausea and fullness are lower (typically 3-4 days after weekly injection for once-weekly drugs), eat a small pre-workout meal or snack even if not hungry, and expect the first few sessions to feel harder than they should. Consistency matters more than intensity in the first month; intensity can build after the body adapts.

Protein target: 1.2 to 1.6 grams per kilogram of body weight per day. Under normal (non-weight-loss) conditions, adults need about 0.8 g/kg/day of protein. During weight loss, that number rises. Multiple studies find that people who hit 1.2 to 1.6 g/kg/day preserve substantially more lean mass than people who eat lower protein amounts, even at the same total calorie intake.

For a 200-pound (90 kg) person losing weight on a GLP-1, the protein target is roughly 108 to 145 grams per day. That is achievable but not automatic — most Americans eat 60 to 90 grams per day. Practical sources: 4-6 oz of chicken breast or fish per meal, Greek yogurt (15-20 g per cup), cottage cheese (25 g per cup), eggs (6 g each), tofu (10 g per half cup), whey or plant protein shakes (20-25 g per scoop). For GLP-1 users whose appetite is small, whey protein shakes are practical because they get 20-25 grams of high-quality protein into a low-volume drink.

Body composition monitoring (optional but useful). DXA scans provide the most accurate measurement of muscle versus fat loss during weight loss. Scans cost $50 to $150 out of pocket at many hospitals and standalone imaging centers. A baseline scan before starting a GLP-1, then a follow-up scan at 6 to 12 months, tells you what fraction of the weight loss is coming from fat versus lean tissue. If lean loss looks high (over 35% of total loss), that is a signal to increase resistance training or protein intake or both.

Cheaper alternatives include bioelectrical impedance scales (accuracy is modest but the trend is useful), circumference measurements (waist, hips, upper arm, thigh at reference points), and strength benchmarks in the gym (grip strength, leg press or squat max, plank duration). Any consistent measurement helps; the specific tool matters less than doing the same measurement repeatedly.

A word on creatine. Creatine monohydrate (3-5 g per day) has strong evidence for preserving strength during weight loss. It is one of the most studied supplements in exercise science and one of the few with consistent muscle-preservation data. It is cheap ($15-25 for a 6-month supply), safe in healthy adults, and worth considering for GLP-1 users doing resistance training. Talk to a physician if you have kidney disease or take medications that affect kidney function.

The clinical care model for GLP-1 prescribing is still catching up to the muscle question. In 2024 and early 2025, the standard prescribing conversation was: side effects, cost, dosing schedule, expected weight loss. Muscle preservation was rarely part of the conversation.

By late 2026, several practice guidelines have started to include it. The Obesity Medicine Association's 2026 clinical practice statements and the Endocrine Society's updated obesity guidelines both recommend that GLP-1 prescriptions be paired with structured lifestyle guidance covering protein intake and resistance training, especially for patients over 60. Neither guideline requires body composition scans, though both suggest they are useful for patients at higher risk of sarcopenia.

What to ask your prescriber. A few concrete questions that address the muscle question directly.

What protein target should I aim for during weight loss? A good answer will name a range in grams per kilogram of body weight per day (typical range 1.2-1.6 g/kg/day). A vague answer like "eat a healthy diet" suggests the prescriber has not thought about it specifically.

Is resistance training safe for me, and can I get a referral to a physical therapist or exercise physiologist? Some patients (with joint issues, cardiovascular concerns, or history of falls) benefit from professional guidance to build a safe program. Others can start on their own.

Should I get a baseline DXA scan? For patients over 60 or with prior fracture history, this is worth considering. For younger patients without risk factors, it is optional but informative.

At what point should we consider adding a muscle-preserving drug if one becomes available? Bimagrumab and possibly others may be on the market within a few years. Knowing that the prescriber is tracking the pipeline is a signal of good ongoing care.

Telehealth prescribing. Many current GLP-1 prescriptions come from telehealth vendors without a longitudinal relationship. If your prescription is from a telehealth service, consider adding a primary care or endocrinology relationship for the lifestyle side of the treatment. The drug is only half of what matters for long-term outcomes.

About 25% to 35% of the weight lost on semaglutide (Ozempic, Wegovy) and 25% to 30% on tirzepatide (Zepbound, Mounjaro) is lean tissue rather than fat, based on DXA scan substudies within the STEP-1 and SURMOUNT-1 trials. That ratio is close to what any form of rapid weight loss produces, including bariatric surgery and low-calorie diets.

The muscle question matters because muscle preservation affects long-term metabolic rate, physical function in older adults, fracture risk, and the trajectory of weight regain after discontinuation.

Enveda's ENV-308 pill reported positive Phase 1 safety data on August 18, 2026 and is designed to mimic Lac-Phe, an exercise-produced molecule that preserves lean tissue during weight loss in animal models. Phase 2 trials are planned. Realistic earliest approval: 2028-2029. Other candidates — bimagrumab (Lilly), petrelintide (Roche-Zealand), HM17321 (Hanmi-Genentech), ARO-INHBE (Arrowhead) — are in later or comparable stages, but none is commercially available yet.

The behavioral tools that work now: resistance training 2-3 times per week at a load that is genuinely challenging, protein intake of 1.2-1.6 g/kg/day, and creatine monohydrate at 3-5 g/day for people doing resistance work. Body composition monitoring with DXA scans or bioelectrical impedance can track whether these tools are working for a specific person.

The drug pipeline will eventually offer pharmacological help. The behaviors will still matter after that.