Summary: In an exploratory imaging substudy of a 32-week randomised trial in 80 adults with type 2 diabetes and high cardiovascular risk, semaglutide and empagliflozin each reduced cortical apparent diffusion coefficient (ADC) on diffusion-weighted MRI versus placebo (semaglutide -0.20×10-3 mm2/s, p<0.001; empagliflozin -0.15×10-3 mm2/s, p=0.01), whereas the combination arm showed no significant change. Total kidney volume fell modestly (3-5%) across all active arms. These microstructural changes were not associated with GFR, albuminuria or inflammation.
PICO Summary
| Element | Detail |
|---|---|
| Population | 80 adults with type 2 diabetes and high cardiovascular risk; post hoc MRI substudy of a 32-week randomised, four-arm trial; single-centre, Denmark. |
| Intervention | Semaglutide (with tablet placebo) for 32 weeks (n=20); or combination therapy, semaglutide for 16 weeks then empagliflozin added for a further 16 weeks (n=20). Outcomes assessed by diffusion-weighted MRI (cortical, medullary and cortico-medullary ADC) and total kidney volume (TKV). |
| Comparison | Tablet placebo (n=20) and empagliflozin monotherapy for 32 weeks (n=20). |
| Outcome | Cortical ADC vs placebo: semaglutide -0.20×10-3 mm2/s (95% CI -0.30, -0.10; p<0.001); empagliflozin -0.15×10-3 mm2/s (95% CI -0.26, -0.04; p=0.01); combination not significant (-0.05×10-3 mm2/s; 95% CI -0.15, 0.05; p=0.29). Medullary ADC unchanged in all arms. Cortico-medullary difference fell only with semaglutide (-0.13×10-3 mm2/s; 95% CI -0.22, -0.04; p=0.01). TKV vs placebo: -3% semaglutide (p=0.04), -3% empagliflozin (p=0.02), -5% combination (p<0.001). Cortical ADC changes were not associated with GFR, albuminuria, TKV or inflammation; TKV changes were associated with GFR, albuminuria and HbA1c. |
Renal diffusion MRI with semaglutide and empagliflozin
RCT substudy · type 2 diabetes · 32 weeks
Semaglutide and empagliflozin each lowered cortical ADC versus placebo, but the change did not track with GFR, albuminuria or inflammation. A hypothesis-generating imaging signal, not yet a clinical tool.
Expert Commentary
This is a mechanistic imaging substudy, not an outcome trial, and it should be read as hypothesis-generating rather than confirmatory. The finding that semaglutide and empagliflozin each lowered cortical ADC is internally consistent and statistically convincing, yet its biological meaning is uncertain. ADC was proposed as a surrogate for kidney microstructure and fibrosis, but the observed reductions were not associated with GFR, albuminuria or inflammatory markers, and the authors concede the signal may reflect processes unrelated to fibrosis. The modest 3-5% fall in total kidney volume, which did track with GFR, albuminuria and HbA1c, is more plausibly explained by reduced glomerular hyperfiltration, a familiar early effect of both drug classes. The most weighed limitation is power: with only 20 patients per arm, the non-significant combination result cannot be read as evidence of no effect, and the sequential add-on design for that arm complicates interpretation. Can I use this with my patients? Not yet as a clinical tool; diffusion MRI remains a research instrument here, and nothing in these data changes prescribing for any patient type. The renoprotective case for these agents already rests on hard endpoints elsewhere. Adequately powered studies linking ADC change to measured histology and long-term function are needed before this imaging readout earns a clinical role.
References
Vernstrom L, Gullaksen S, Sorensen SS, Ringgaard S, Laustsen C, Birn H, et al. Effects of semaglutide, empagliflozin and their combination on renal diffusion-weighted MRI and total kidney volume in patients with type 2 diabetes: a post hoc analysis from a 32 week randomised trial. Diabetologia. 2024;67(10):2175-2187. doi:10.1007/s00125-024-06228-y