Body-composition change on GLP-1 therapy: what newer evidence actually shows

What the verified numbers show

Conte, Hall, and Klein reanalyzed the body-composition data from the major GLP-1-class trials and reported the fat-mass to fat-free-mass split for each agent:

• Semaglutide: approximately 61 percent fat mass, 39 percent fat-free mass.
• Tirzepatide: approximately 75 percent fat mass, 25 percent fat-free mass.
• Retatrutide: approximately 67 percent fat mass, 33 percent fat-free mass.

They also asked the framing question directly in the title of the paper: is weight loss-induced muscle mass loss clinically relevant? Their answer is that the relevance depends on what the patient is doing — protein intake, resistance-training stimulus, age, and starting body composition all move the result more than the agent does.

The SEMALEAN study added functional outcomes that the DEXA-only data could not provide. Across 115 patients on semaglutide 2.4 mg followed for 12 months:

• Body weight fell a mean of 12.5 percent. Ninety-three percent of patients lost at least 5 percent, 59 percent lost at least 10 percent, and 26 percent lost at least 15 percent.
• Fat mass fell substantially in both absolute terms and as a percentage of body composition.
• Lean mass as a percentage of body composition rose from baseline.
• Handgrip strength improved at both the 7-month and 12-month checkpoints.
• Sarcopenic obesity prevalence dropped from 49 percent at baseline to 33 percent at 12 months. Twenty-two percent of patients transitioned from sarcopenic obesity to no sarcopenic obesity; 5 percent moved the other way.

Why the 40 percent practitioner estimate has been retired

The 40 percent figure that circulated in practitioner discussions before 2024 was a clinical-experience estimate, not a published trial statistic. The verified figures from Conte, Hall, and Klein (39 percent fat-free mass on semaglutide; 25 percent on tirzepatide; 33 percent on retatrutide) are now the load-bearing numbers for any GLP-1-related lean-mass claim on EDRD content. We do not cite 40 percent as established fact and we do not extend it to agents where Conte and colleagues did not report a value.

The DEXA versus MRI methodological note

DEXA measures lean soft tissue. Lean soft tissue includes contractile skeletal muscle, but it also includes connective tissue, skin, water, and a portion of the non-fat components of adipose tissue. Crucially, DEXA cannot separate intramuscular fat from contractile muscle. When intramuscular fat is reduced — which is generally favorable for metabolic health — DEXA reports a portion of that change as "lean mass loss."

MRI using proton-density fat fraction can separate intramuscular fat from contractile muscle directly. Heymsfield and colleagues (Advances in Nutrition 2024) describe the methodological gap. The clinical implication: older DEXA-based estimates of "muscle loss" on GLP-1 therapy likely include a portion of intramuscular fat reduction, which is favorable rather than concerning. The newer MRI-based and functional-outcome data (SEMALEAN handgrip strength gain on semaglutide) are more reliable indicators of whether contractile muscle is actually changing.

What this means for the practical intervention

The actionable target is unchanged regardless of which body-composition measurement is used:

• Protein intake of 1.2 to 1.6 g/kg of ideal body weight per day during active weight loss, distributed across meals with at least 30 grams per meal (Layman 2015; Mechanick 2025; Almandoz 2024).
• Progressive resistance training at least three times per week. EASO 2015 and AACE 2016 guideline language is the most direct: resistance training is required to preserve or increase lean body mass during weight loss in middle-aged adults.
• Functional monitoring matters more than body-composition scan numbers. Handgrip dynamometer (validated as a longevity correlate), 30-second sit-to-stand, and the SARC-F questionnaire are self-administered, accessible, and track what the body actually does. DEXA every 6 to 12 months is fine for in-person trend tracking but is not a research-grade differentiator of contractile muscle from intramuscular fat.

The coming-off and weight-cycling caveat

The largest documented post-discontinuation risk to muscle is repeated on-and-off cycling, not the act of coming off itself. Rossi and colleagues (Obesity 2019) found that severe weight cyclers — six or more cycles — carry roughly 5.0 times the risk of low skeletal muscle mass and 5.5 times the risk of reduced muscle strength compared with non-cyclers. Sarcopenia prevalence climbed from 9 percent in non-cyclers to 49 percent in severe cyclers. A planned taper with continued protein and training protects the result better than yo-yo restarts.

Where this evidence informs EDRD content

This entry is referenced by the following cluster pages:

• /glp-1-nutrition/muscle-loss — anchors the headline reframe.
• /glp-1-nutrition/protein-requirements — supports the 1.2 to 1.6 g/kg range with the Mechanick and Almandoz framework.
• /glp-1-nutrition/exercise-and-muscle-preservation — supports the resistance-training requirement with EASO 2015 and the Mechanick 2025 mitigation strategy framework.
• /glp-1-nutrition/wegovy — replaces the previous 40 percent practitioner estimate with the Conte, Hall, and Klein 61/39 split for semaglutide.
• /glp-1-nutrition/coming-off-glp-1 — supports the Rossi 2019 weight-cycling caveat.

Provenance

The reframe in this entry was driven by the DUO26 Toronto conference (April 23 to 25, 2026), specifically Dr. Akshay Jain's "Muscle Matters with GLP-1 Therapy" plenary session on April 24. The supporting deck, transcript, and operational audit document are stored at:

• customer-development/conference-signals/duo26-toronto/synthesis/01-jain-muscle-matters.md
• customer-development/conference-signals/duo26-toronto/wbs-3.3.1-claim-audit.md