Summary
In the 52-week REMODEL trial, semaglutide was associated with changes in kidney vascular measures and tissue signals in people with type 2 diabetes and chronic kidney disease. The three coprimary MRI outcomes did not differ significantly from placebo.
The 52-week REMODEL trial found that semaglutide treatment was associated with changes in kidney vascular measurements and tissue signals in people with type 2 diabetes and chronic kidney disease (CKD). The randomized, placebo-controlled study compared subcutaneous semaglutide, titrated to 1 mg once weekly, with placebo. It enrolled 106 participants at 28 sites in eight countries: 71 received semaglutide and 35 placebo.
REMODEL was designed to investigate how semaglutide may affect the kidney, using functional MRI, biopsies and molecular analysis. Its three coprimary MRI outcomes—kidney oxygenation, global perfusion and T1 mapping, a measure influenced by tissue properties including inflammation and fibrosis—did not differ significantly between treatment groups after 52 weeks.
Kidney imaging showed mixed results
Two secondary MRI measures differed between groups. The renal artery resistive index, an imaging-derived measure related to vascular resistance, was lower with semaglutide than placebo: the estimated treatment ratio was 0.96 (95% confidence interval 0.93–0.99; P = 0.008). The cortical apparent diffusion coefficient (ADC), which measures water movement through tissue, was higher, with an estimated treatment ratio of 1.05 (95% confidence interval 1.01–1.09; P = 0.005). Because fibrosis can restrict water movement, the researchers interpreted the ADC result as consistent with less progression of kidney fibrosis.
Exploratory clinical measures also changed. Urine albumin-to-creatinine ratio was 40% lower with semaglutide than placebo. The estimated between-group difference in 24-hour urine creatinine clearance was 12 ml/min (95% confidence interval 0.7–23.3; P = 0.038). Estimates of change in eGFR had confidence intervals that included no difference between groups.
These findings need to be read in light of the trial’s statistical design. The researchers did not plan a formal testing hierarchy or adjustment for multiple comparisons, so nominal P values for secondary and exploratory outcomes warrant caution. The MRI coprimary results were not statistically significant.
Biopsy analyses point to endothelial changes
Kidney biopsies were obtained from 33 participants. In paired samples, single-nucleus RNA sequencing examined gene activity in kidney cell types, while spatial transcriptomics assessed cells in their tissue locations. The strongest gene-expression responses were in two populations of glomerular endothelial cells, which line the kidney’s small blood vessels. Many genes related to metabolic stress, inflammation and fibrosis had lower expression after semaglutide treatment than after placebo.
In a spatial analysis of 13 paired participants, fewer natural killer and natural killer T cells were found near glomerular endothelial cells after treatment. Together, the imaging and tissue results suggest possible vascular and cellular pathways through which semaglutide may affect the kidney. The study’s authors describe the links among these measurements as associative; it did not establish that the tissue changes caused the clinical findings.
Serious adverse events occurred at similar rates in the two groups. Permanent discontinuation because of adverse events was more frequent with semaglutide: 12.7% (9 of 71 participants), compared with 2.9% (1 of 35) with placebo. Women made up 23.6% of participants, and the biopsy and molecular analyses involved smaller subsets than the overall trial.