The main finding
In one sentence: In a small planned substudy of the SUSTAIN 8 trial, body scans showed that both semaglutide 1.0 mg and canagliflozin reduced fat mass and lean mass over 52 weeks, and the differences between the two medicines were not statistically significant.
Group averages, not an individual prediction. A trial result describes what happened to a defined group over a defined time.
On this page
Why this study matters
When people lose weight, they lose a mix of fat and lean tissue. Lean tissue includes muscle, and many readers worry that weight-loss medicines cause too much of it to go. This substudy used a scanning method called DXA to measure what kind of tissue was lost with semaglutide (Wegovy, Ozempic) at the 1.0 mg diabetes dose, compared with a different diabetes medicine, canagliflozin. Canadian sites took part.
Who was in the study
The main SUSTAIN 8 trial enrolled adults aged 18 and over with type 2 diabetes. Everyone was taking a stable daily dose of metformin (at least 1500 mg, or the highest dose they could tolerate). HbA1c, a blood test showing average blood sugar over about three months, was between 7.0% and 10.5%. Kidney function, measured as eGFR, was 60 or higher.
The substudy included 178 of these participants who had a DXA scan at screening: 88 were assigned to semaglutide and 90 to canagliflozin. Of these, 114 people (53 on semaglutide and 61 on canagliflozin) had a scan at the end of treatment that could be included in the main analysis. The substudy ran for 52 weeks. The CSV record for this study notes that Canadian authors or sites were involved.
What the researchers did
This was a randomized, double-blind, double-dummy comparison. Randomized means chance decided the treatment. Double-blind means neither participants nor investigators knew who was on which medicine. Double-dummy means everyone took both a weekly injection and a daily pill, one of which was a placebo (an inactive copy), so the treatments could not be told apart.
One group received semaglutide 1.0 mg once a week by injection under the skin plus a daily placebo pill. The other received canagliflozin 300 mg once a day by mouth plus a weekly placebo injection.
DXA stands for dual-energy X-ray absorptiometry. It is a whole-body scan that estimates fat mass, lean mass and visceral fat (fat around the organs). The main measure was the change in total fat mass in kilograms from the start to week 52.
What they found
At the start, average total fat mass was 33.2 kg. After 52 weeks it had fallen by 3.4 kg with semaglutide and by 2.6 kg with canagliflozin. The estimated difference was -0.79 kg, with a 95% confidence interval of -2.10 to 0.51 (R32-C34). A confidence interval is the range in which the true difference is likely to sit. Because this range crosses zero, the difference between the two medicines was not statistically significant.
| Measure | Starting average | Semaglutide 1.0 mg | Canagliflozin 300 mg | Difference (95% CI) |
|---|---|---|---|---|
| Total fat mass | 33.2 kg | -3.4 kg | -2.6 kg | -0.79 kg (-2.10 to 0.51) |
| Total lean mass | 51.3 kg | -2.3 kg | -1.5 kg | -0.78 kg (-1.61 to 0.04) |
| Lean mass as a share of body mass | 59.4% | +1.2 percentage points | +1.1 percentage points | 0.14 (-0.89 to 1.17) |
Both medicines reduced lean mass, but because fat fell by more, the proportion of the body that was lean tissue rose slightly in both groups (R32-C33). Changes in visceral fat mass, and overall changes in the fat-to-lean ratio, were similar between the groups.
Side effects and people who stopped
The published summary we reviewed does not report side effects or the number of people who stopped treatment in this substudy. Of the 178 people randomized, 114 had an end-of-treatment scan included in the main analysis, which means many did not have complete scan data.
What this study does not tell you
- This was a small substudy, and only 114 of the 178 people had an end-of-treatment scan included in the main analysis. The researchers used a statistical method called multiple imputation to fill in missing values under the assumption that people had stayed on treatment without rescue medicine.
- There was no placebo group. The authors say the specific effect of either medicine on body composition, compared with no treatment, is speculative.
- The trial used the 1.0 mg diabetes dose of semaglutide, not the 2.4 mg weight-management dose.
- Lean mass on a DXA scan is not the same as muscle strength or physical function.
- Everyone had type 2 diabetes and was on metformin, so the results may not apply to people without diabetes.
- The study was funded by the company that makes semaglutide.
What it means in Canada
This substudy adds a small piece of information about what kind of tissue is lost with semaglutide, at a dose lower than the one used for weight management. It does not answer whether semaglutide protects or reduces muscle in everyday use. For how the medicine is used in Canada, see semaglutide in Canada. For questions about what provincial plans pay for, see public drug coverage by province.
Common questions
How much fat mass did people lose with semaglutide in the SUSTAIN 8 substudy?
Total fat mass fell by 3.4 kg with semaglutide 1.0 mg and by 2.6 kg with canagliflozin over 52 weeks. The difference of -0.79 kg was not statistically significant.
Did semaglutide cause loss of lean mass in SUSTAIN 8?
Lean mass fell by 2.3 kg with semaglutide and 1.5 kg with canagliflozin, but the proportion of body mass that was lean rose in both groups, by 1.2 and 1.1 percentage points.
Was there a placebo group in the SUSTAIN 8 body-composition substudy?
No. The substudy compared two active medicines, so it cannot say what either one does compared with no treatment.
Related reading
Other studies in this library that bear on the same question.
STEP 2: semaglutide for weight loss in type 2 diabetes
SURPASS-2: tirzepatide versus semaglutide 1 mg in diabetes
Retatrutide body-composition substudy in type 2 diabetes
STEP 1: semaglutide and weight loss in adults without diabetes
Source
Original paper: Rory J McCrimmon, SUSTAIN 8 DXA substudy, Diabetologia, 2020. PubMed: https://pubmed.ncbi.nlm.nih.gov/31897524/. DOI: https://doi.org/10.1007/s00125-019-05065-8. Funding: The trial was supported by Novo Nordisk A/S, the company that makes semaglutide. How we summarized it: full text and extracted data.
This page explains a published study. It is not medical advice and does not describe whether a medicine is right for you. Talk to your doctor, nurse practitioner or pharmacist about your own situation. Reviewed and signed by two pharmacists registered in British Columbia, 2026-09-18.
Evidence records behind this page
Each record is a row in the clinical evidence file, taken from the published paper named above. The caution column is the limit the researcher recorded for that number.
This table scrolls sideways.
| Record | What the paper reports | Type | Population | Caution | Source |
|---|---|---|---|---|---|
| R32-C33 | In the SUSTAIN 8 DXA substudy, lean mass fell by 2.3 kg with semaglutide 1.0 mg and 1.5 kg with canagliflozin, while the proportion of body mass that was lean increased in both groups. | direct evidence | Adults with T2D on stable metformin; HbA1c 7.0-10.5%; eGFR ≥60 | Small substudy with missing DXA outcomes; no placebo; lean mass is not equivalent to skeletal-muscle function; 1.0 mg diabetes dose. Separate post hoc correlations and total-weight calculations excluded. | PubMed 31897524 |
| R32-C34 | The body-composition differences between semaglutide and canagliflozin were not statistically significant in this small substudy, which had no placebo group. | direct evidence | Adults with T2D on stable metformin; HbA1c 7.0-10.5%; eGFR ≥60 | Small substudy with missing DXA outcomes; no placebo; lean mass is not equivalent to skeletal-muscle function; 1.0 mg diabetes dose. Separate post hoc correlations and total-weight calculations excluded. | PubMed 31897524 |