Short answer: In a 24-week study of 88 adults with type 2 diabetes and obesity, roughly 83 to 92 percent of the weight lost on semaglutide came from fat mass, not lean mass. Lean mass did decrease, but by a small margin, and in the oral semaglutide group the change was not statistically significant.
That is the finding from Rodríguez Jiménez et al., published in Frontiers in Endocrinology in June 2024. Researchers at Hospital Universitario Virgen Macarena in Seville tracked body composition with multi-frequency bioelectrical impedance instead of relying on scale weight alone.
| Subcutaneous (n=55) | Oral (n=33) | |
|---|---|---|
| Total weight lost | 10.0 kg (9.5%) | 8.6 kg (9.4%) |
| Fat mass lost | 8.5 kg | 8.0 kg |
| Lean mass lost | 1.7 kg | 0.7 kg (not significant, p=0.112) |
| Share of loss from fat | ~83% | ~92% |
| Visceral fat area | −30.2 cm² | −42.3 cm² |
Fat mass percentage fell in both groups while fat-free mass percentage rose, by 3.6 points and 5.2 points respectively. That is the direction you want: the body is becoming proportionally leaner, not just lighter.
The authors did not dismiss the concern. Their conclusion states that some muscle decline should be anticipated and addressed through regular physical activity, while noting it is expected to be less pronounced than fat loss.
They also offered an explanation for why the subcutaneous group lost more lean mass. Those participants started with a higher average BMI and therefore a higher baseline lean mass, so a proportionally larger reduction follows from a larger total loss.
A bathroom scale reports one number. It cannot distinguish 1 kg of fat from 1 kg of muscle. Two people can lose the same 8 kg with completely different outcomes underneath, and the display looks identical.
This matters most in two situations. The first is early treatment, when loss is fastest and the ratio is least visible. The second is a plateau, when the number stops moving and the assumption is that nothing is happening. In this study, visceral fat area dropped substantially in both groups, a change no weight-only reading would show.
The study did not test an intervention, so it cannot answer that directly. The authors recommend regular physical activity, specifically noting strength training in their described lifestyle guidance, as the way to mitigate it.
The multivariate analysis found no significant association between administration route and fat mass loss percentage. The apparent difference in lean mass between groups is more plausibly explained by their different starting characteristics than by the route itself.
Only partly. Every participant had type 2 diabetes and obesity with a BMI of 30 or higher and inadequate glycaemic control. If you are taking a GLP-1 for weight management without diabetes, this cohort is not your cohort.
With an InBody 770, a multi-frequency bioelectrical impedance device used in clinical and research settings. The authors note that bioelectrical impedance is not the gold standard, and cite studies comparing it against DXA with broadly consistent results.
There was no control group. It was retrospective. The follow-up ran 24 weeks, which is short relative to how long people stay on these medications. The sample was 88 people. The authors state each of these limitations themselves.
None of that invalidates the finding. It does mean the honest conclusion is narrower than the headline: in this population, over this period, weight loss was predominantly fat loss with modest lean mass reduction.
Group averages describe a group. Your own split depends on your starting composition, your training, your protein intake, and your age, none of which a published mean accounts for.
The InBody Dial H40 uses the same measurement principle as the 770 in this study, multi-frequency bioelectrical impedance with electrodes at the hands and feet, in a device built for home use. It is a wellness device, not a diagnostic one, and it does not replace a clinical assessment. What it provides is a consistent record between appointments so that the trend you bring to your next review covers twelve weeks rather than one morning.
Source: Rodríguez Jiménez B, Rodríguez de Vera Gómez P, Belmonte Lomas S, et al. Transforming body composition with semaglutide in adults with obesity and type 2 diabetes mellitus. Front Endocrinol. 2024;15:1386542. doi:10.3389/fendo.2024.1386542. Open access under CC BY.
This article summarises published research and is general information, not medical advice. It does not recommend any medication. Decisions about GLP-1 treatment belong with your doctor.
