LEAN-PREP protocol published: the first randomized trial of lifting weights on semaglutide or tirzepatide

Researchers have published the design for what they describe as the first randomized controlled trial testing resistance exercise and protein supplementation in people already taking a GLP-1 medication for weight loss. The protocol, called LEAN-PREP, appeared in BMJ Open and describes a six-month study enrolling 232 adults on semaglutide or tirzepatide ^[1].

Participants will be randomly assigned to one of four groups: usual care (control), a resistance exercise program, protein supplementation, or both exercise and protein together ^[1]. The main outcome is not weight on a scale. Instead, investigators will measure the cross-sectional area of the quadriceps muscle using MRI, a direct imaging measure of thigh muscle size ^[1].

That design choice matters because the debate about GLP-1 medications and muscle has been hard to settle with the tools most trials use. Body composition estimates from DXA scans or bioelectrical impedance lump together muscle, organs, connective tissue and water as "lean mass." MRI of a specific muscle group is a more targeted measurement, and a randomized comparison of four conditions is designed to separate the effect of lifting weights from the effect of eating more protein, and to see whether combining them adds anything.

What the protocol does not settle

A protocol is a publication of a study plan, not results. It tells other researchers and regulators in advance what will be measured and how, which limits the temptation to cherry-pick findings later. It does not tell anyone whether resistance training preserves muscle on these drugs. Until LEAN-PREP reports, guidance about exercise and protein for people on semaglutide or tirzepatide is extrapolated from studies done in other populations, such as older adults losing weight through diet alone or people after bariatric surgery ^[1].

Several practical details are not available in the material reviewed for this article, because the full protocol text could not be retrieved. That includes the countries and sites involved, the funding source, the specific exercise prescription (how many sessions per week, which lifts, how the load progresses), the protein dose and type, the enrollment start and expected completion dates, how long participants must have been on medication before joining, and what secondary outcomes such as strength, physical function or fat mass will be tracked ^[1]. Those specifics will shape how much the results can be applied to everyday practice, so readers may want to check the published protocol directly.

It is also not stated in the available summary whether the trial will analyze semaglutide and tirzepatide users separately, or pool them. The two drugs work differently: semaglutide (sold as Ozempic, Wegovy and Rybelsus) targets the GLP-1 receptor, while tirzepatide (Mounjaro and Zepbound) targets both GLP-1 and GIP receptors. Whether that difference affects muscle loss is an open question that a 232-person trial split four ways may not be large enough to answer.

Why it matters for patients

Muscle loss during rapid weight loss is one of the most common concerns raised by people taking these medications, and one of the least well answered. Some of lean mass that comes off during weight loss is normal and expected with any method of losing weight. What is unclear is whether the amount lost on a GLP-1 drug is greater than expected for the amount of fat lost, whether it affects strength and day-to-day function, and whether anything reliably prevents it.

Right now, when a clinician suggests strength training or a higher protein intake to someone on semaglutide or tirzepatide, that advice is based on reasoning from other settings rather than on a trial in this exact population ^[1]. That is not the same as the advice being wrong. It does mean the size of the benefit, if any, is unmeasured for these drugs.

A trial with an MRI primary outcome and four arms is built to produce a clearer answer than observational data or secondary analyses of drug trials can. It is also a reminder that the field is still early: the first randomized test of a very ordinary intervention, lifting weights, is only now beginning in this drug class.

What happens next

The protocol was published in BMJ Open in 2026 ^[1]. The trial is designed to run six months per participant ^[1]. Results timing depends on when enrollment finishes, which is not stated in the available information. Anyone following this question should watch for the primary results publication, and for whether the reported outcomes match what the protocol laid out.