How to keep your muscle on a GLP-1.

What actually happens to muscle on these drugs

Any substantial weight loss — dieting, surgery, or medication — costs some lean tissue. That has always been true. What is new is the magnitude of loss these medications produce, and therefore the magnitude of the lean-mass question that comes with it. In the STEP 1 trial of once-weekly semaglutide, the subgroup that underwent body-composition scanning lost a meaningful share of their total weight as lean body mass.¹

Two pieces of nuance matter here, and both cut against the alarmist version of this story. First, "lean body mass" on a DEXA scan is not the same thing as functional skeletal muscle: it includes organ tissue, connective tissue and body water, all of which legitimately decrease as body size falls. Second, when researchers have examined muscle volume in proportion to body size, some of the change looks like an appropriate adaptation to carrying less weight rather than pathological wasting.²

The adaptive share does not account for all of it, though. Reviews of lean mass change with GLP-1-based therapies conclude that the loss is real, that it varies widely between individuals, and — the useful part — that mitigation strategies exist.³ The variation between individuals is the whole point: two people can lose the same 20 kg with very different body-composition outcomes, and the difference is mostly what they did about it.

Why it is worth protecting

Muscle is your largest site of glucose disposal, a major determinant of resting metabolic rate, and the single best-established predictor of physical independence in later life. Losing it during a weight-loss phase has three practical consequences: strength and function decline, the metabolic benefit of the weight loss is partly offset, and — because a lower muscle mass means lower energy expenditure — regain becomes easier if the medication is stopped. Given that weight regain after discontinuation is well documented, the composition of what you lost determines what you are left with afterwards.

For anyone over 40, this is compounded by age-related muscle loss that is already underway. A weight-loss phase handled carelessly at 55 can cost a decade's worth of muscle in under a year.

Lever one: protein, planned rather than hoped for

How much

Two pieces of evidence bracket the useful range. A meta-analysis of 49 randomised trials in 1,863 people found that protein supplementation meaningfully improved gains in strength and fat-free mass during resistance training, but that intakes beyond about 1.6 g/kg/day produced no further benefit — in people who were not in an energy deficit.⁴

In a deficit, the requirement goes up, because protein is being oxidised for energy and the body is in a net catabolic state. A randomised trial put young men on a roughly 40% energy deficit with six days a week of resistance and interval training, assigning either 1.2 or 2.4 g/kg/day of protein. Over four weeks, the higher-protein group gained 1.2 kg of lean mass while the lower-protein group gained 0.1 kg — and the higher-protein group also lost more fat (4.8 kg versus 3.5 kg).⁵

That study used an aggressive deficit in young trained men, so it is a ceiling rather than a prescription. A reasonable practical translation for someone losing weight on a GLP-1 is roughly 1.6 to 2.2 g of protein per kilogram per day — and, importantly, calculated against a target or lean body weight rather than current total weight, since scaling to a high starting weight produces an unrealistic and unnecessary number. For someone with substantial excess weight, using goal weight is the sensible reference; this is precisely the kind of individualisation worth doing with a clinician rather than a calculator.

The problem nobody warns you about

GLP-1 medications work by suppressing appetite and slowing gastric emptying. That is the therapeutic mechanism, and it is directly at odds with eating 140 grams of protein a day. Patients routinely report early satiety, food aversion — often specifically to meat — and simply forgetting to eat. Left alone, total protein intake falls exactly when the requirement per calorie is highest.

This is why the 2025 joint advisory from the American College of Lifestyle Medicine, the American Society for Nutrition, the Obesity Medicine Association and The Obesity Society treats nutritional planning as an integral part of GLP-1 therapy rather than an optional add-on, highlighting adequate protein, sufficient fibre, and attention to micronutrient adequacy given markedly reduced total intake.⁸

What works in practice:

• Protein first at every meal. When you can only finish half a plate, the half you finish should be the protein.
• Spread it across the day — three to four servings of 30–40 g rather than one large evening meal you no longer have the appetite to finish.
• Use liquid protein when solids are unappealing. A shake is far better than a missed target, and tolerance for liquids is usually higher.
• Favour easily tolerated, protein-dense foods — Greek yoghurt, cottage cheese, eggs, fish, poultry, tofu, whey or soy isolate.
• Track it for a couple of weeks at the start and after every dose escalation. Intake almost always drops after a dose increase, and almost nobody notices without measuring.

Lever two: resistance training, with real weekly volume

Protein is the raw material. Resistance training is the signal that tells the body to keep the muscle it has. Without the signal, extra protein largely goes to waste.

The evidence for combining exercise with weight loss is strong. In a randomised trial of obese older adults, the group assigned to both weight loss and exercise achieved better physical function than either intervention alone, and exercise attenuated the loss of lean mass that accompanied dieting.⁷ Reviews of preserving muscle during weight loss reach the same conclusion: the combination of adequate protein and resistance exercise is the intervention with the best supporting evidence.⁹

On dose, a meta-regression of 34 treatment groups found a graded dose-response relationship between weekly set volume and muscle growth — each additional weekly set per muscle group was associated with further gains.⁶ More is better, within recoverable limits.

A defensible starting prescription for someone on a GLP-1:

• Two to three full-body sessions per week, on non-consecutive days. Frequency beats duration; three 40-minute sessions outperform one heroic weekend effort.
• Around 10 or more hard sets per muscle group per week, built up gradually rather than started at. If you are new to lifting, half that is a fine beginning and still produces results.
• Compound movements as the backbone — a squat or leg press, a hip hinge, a horizontal push, a vertical or horizontal pull, and a carry or core movement. These cover the most muscle for the least time.
• Sets taken close to genuine effort — within a few repetitions of failure. Anything in roughly the 6–20 repetition range builds muscle if the effort is real.
• Progression that is written down. Adding weight or repetitions over time is the actual stimulus. A log is what separates training from exercising.

Walking, cycling and other cardiovascular work are genuinely valuable — for cardiorespiratory fitness, glucose control and appetite regulation — but they do not substitute for loading the muscle. Steps are not a muscle-preservation strategy.

Lever three: pace

The faster you lose, the greater the share that comes from lean tissue. This is one of the oldest findings in the weight-loss literature and it applies to pharmacological loss as much as dietary. Two practical implications:

• Do not escalate the dose faster than you need to. Dose titration is meant to manage side effects, but it is also a lever on rate of loss. If you are losing steadily and tolerating the dose, there is often no reason to keep climbing.
• Do not stack a severe self-imposed calorie restriction on top of the medication's appetite suppression. The drug is already creating the deficit. Adding an aggressive diet on top of it drives the rate of loss into the range where lean tissue goes disproportionately, and makes hitting protein targets nearly impossible.

The rest of the basics are worth stating plainly because they get neglected: adequate sleep, since short sleep during energy restriction shifts loss toward lean tissue; resistance training maintained through the whole weight-loss phase rather than started at the end; and enough total energy to support training. Someone eating 900 calories a day will not recover from hard sessions no matter how the protein is distributed.

How to know whether it is working

The scale cannot tell you what you lost, which is why relying on it is the core mistake. Better signals:

• Strength in the gym. The most accessible proxy there is. If your working weights are holding or climbing during weight loss, you are almost certainly preserving muscle. If they are falling steadily, something needs to change.
• Body composition measurement. A DEXA scan at baseline and every six months quantifies fat versus lean change directly. Bioimpedance is less accurate but useful for trends if you keep conditions consistent.
• Grip strength. Cheap, quick, and a well-validated marker of overall muscle function.
• Waist circumference alongside weight. If waist is falling faster than weight, the composition of your loss is favourable.

The honest summary

GLP-1 medications are a genuine advance, and none of this argues against using one when it is indicated. But a prescription with no plan for muscle is doing half the job. The three things that decide how you come out the other side — protein intake, resistance training volume, and rate of loss — are all within your control, and all of them need to be in place from the first injection rather than added once the scale has moved. If you want the broader risk picture, our article on the risks and questions worth asking before starting a GLP-1 covers the rest.