Do GLP-1 drugs destroy muscle?

One of the biggest concerns surrounding GLP-1 medications like semaglutide and tirzepatide is the fear of “muscle loss.” But according to researcher Stuart Phillips, one of the first questions we should ask is: What exactly are we measuring? In a recent discussion on the topic, Phillips highlighted that lean mass is not the same thing as skeletal muscle mass.

Most studies looking at body composition changes with GLP-1 medications use DXA or BIA scans. These scans can report reductions in “lean mass,” but lean mass includes much more than skeletal muscle. It also includes:

- Water

- Glycogen (stored carbohydrate)

- Organs

- Connective tissue

- Blood volume

- Other non-fat tissues

⁠

So when someone says, “GLP-1s cause muscle loss,” Phillips argues that we need to be careful about what was actually measured. Two recent reviews help explain why the conversation has become confusing. One meta-analysis concluded that GLP-1 receptor agonists produce weight loss primarily through reductions in fat mass, with comparatively modest reductions in lean body mass.

Another systematic review found that the proportion of weight lost from fat-free mass sometimes exceeded the commonly quoted benchmark that around 25% of weight loss comes from lean tissue. At first glance, those findings sound contradictory — but they are actually asking different questions.

One asks:

“How much lean mass decreased overall?”⁠

The other asks:

“What proportion of total weight loss came from lean-related tissue?”

⁠

Those are not the same thing. Phillips also points out that the commonly referenced “quarter fat-free mass rule” is a heuristic, not a biological law. The proportion of lean tissue lost during weight loss depends on many factors, including:

- Baseline body fat levels

- Age

- Sex

- Protein intake

- Resistance training

- Physical activity levels

- Rate of weight loss

- Duration of dieting

- Overall metabolic health

⁠

So do GLP-1 medications cause more muscle loss than expected? Possibly in some settings... but the evidence is not definitive. What we do know is that GLP-1 and GLP-1/GIP medications clearly reduce:

- Total body weight

- Fat mass

- Visceral fat

- Ectopic fat

⁠

However, much of what is currently reported as “lean mass loss” has not been conclusively proven to represent clinically meaningful skeletal muscle loss. A recent paper by Langer and colleagues is particularly interesting because it attempted to move beyond DXA-only measurements. In a small proof-of-concept human trial, semaglutide reduced the cross-sectional area of the vastus lateralis muscle over 12 weeks, but researchers did not observe reductions in knee-extension strength or grip strength.

⁠That distinction matters.

Changes in imaging measurements do not automatically mean someone becomes weaker or less functional. And this just shows one of the biggest gaps in the current research: Most GLP-1 studies were designed to study weight loss... not muscle physiology or long-term physical function.

Very few studies:

- Directly measure skeletal muscle tissue

- Include older adults at high risk of frailty

- Assess long-term strength outcomes

- Measure real-world physical function

⁠

The responsible takeaway is not: “GLP-1 drugs destroy muscle.”

But it is also not: “There’s nothing to consider.”

As Phillips explains, substantial weight loss usually includes some loss of lean tissue. With incretin-based medications, some of the reported lean mass loss may represent skeletal muscle, but much may also reflect reductions in water, glycogen, and other non-fat tissues.

Right now, we simply need better research to determine how clinically meaningful these changes actually are. In the meantime, the practical advice remains clear:

- Resistance train regularly

- Consume adequate protein

- Monitor strength and physical function

- Focus on preserving muscle while reducing fat

⁠

Because the goal should never be just weight loss. The goal is to lose fat while preserving the tissues that keep us strong, mobile, independent, and functioning well long-term.

⁠

Credit to Stuart Phillips for much of the scientific discussion and interpretation that informed this article.