Effects of semaglutide and empagliflozin on markers of endothelial function in persons with type 2 diabetes: A post hoc analysis of a randomized clinical trial.
Gullaksen S, Vernstrøm L, et al. • Journal of diabetes and its complications • 2026
Semaglutide improved endothelial function compared to baseline but not significantly versus placebo, and CAM responses were heterogeneous, suggesting distinct roles in endothelial dysfunction and atherosclerosis.
Key Findings
Results
Semaglutide increased the reactive hyperaemic index (RHI) compared to baseline but not compared to placebo.
RHI increased from baseline in the semaglutide group by 0.11 (95%CI [0.008;0.21], p = 0.03)
The increase in RHI compared to placebo was 0.11 (95%CI [-0.04;0.24], p = 0.16), which was not statistically significant
Authors attributed the lack of significance versus placebo to limited statistical power
Study involved 120 participants with type 2 diabetes aged ≥50 randomized over 32 weeks
Results
Empagliflozin had no significant effect on RHI.
There was no effect on RHI in the empagliflozin group compared to placebo
This was assessed in a 32-week randomized trial with four arms: semaglutide, empagliflozin, combination, and placebo
Participants were adults with type 2 diabetes aged ≥50
Results
E-Selectin decreased significantly in the semaglutide and combination groups compared to placebo.
In the semaglutide group, E-Selectin decreased by -9 units (95%CI [-14.1;-5.1], p < 0.01) compared to placebo
In the combination group (semaglutide + empagliflozin), E-Selectin decreased by -9 units (95%CI [-14.3;-5.2], p < 0.01) compared to placebo
E-Selectin was not reported to be significantly affected in the empagliflozin-alone group
Results
VCAM-1 increased significantly in the semaglutide and combination groups compared to placebo.
VCAM-1 increased by 12.3 units (95%CI [2.8;20.8], p = 0.01) in the semaglutide group compared to placebo
VCAM-1 increased by 16.2 units (95%CI [7.2;24.3], p < 0.01) in the combination group compared to placebo
This increase in VCAM-1 was considered a heterogeneous and potentially unfavorable finding relative to the decrease in E-Selectin
Results
P-Selectin and ICAM-1 were not significantly affected in any treatment group compared to placebo.
P-Selectin showed no significant change in any group compared to placebo (p ≥ 0.11)
ICAM-1 showed no significant change in any group compared to placebo (p ≥ 0.09)
These findings were consistent across all four treatment arms
Discussion
The cell adhesion molecule (CAM) responses across treatment groups were heterogeneous, suggesting distinct roles for individual CAMs in endothelial dysfunction and atherosclerosis.
E-Selectin decreased while VCAM-1 increased in the same treatment groups (semaglutide and combination), indicating divergent CAM responses
P-Selectin and ICAM-1 were unaffected, further demonstrating non-uniform CAM responses
Authors concluded that heterogeneous CAM responses suggest distinct roles in endothelial dysfunction and atherosclerosis
Methods
The trial was a 32-week randomized study of 120 participants with type 2 diabetes assigned to four treatment groups.
Participants were randomized to semaglutide, empagliflozin, the combination, or placebo
Inclusion criteria required age ≥50 and a diagnosis of type 2 diabetes
This report is a post-hoc analysis of the primary randomized trial, which evaluated separate and combined effects of semaglutide and empagliflozin on cardio-renal organ damage
The trial was registered at ClinicalTrialsRegister.eu: EudraCT 2019-000781-38
What This Means
This research examined whether two commonly used diabetes medications—semaglutide (a GLP-1 receptor agonist) and empagliflozin (an SGLT2 inhibitor)—improve the health of blood vessel linings (endothelial function) in people with type 2 diabetes. The study measured blood vessel flexibility using a test called the reactive hyperaemic index (RHI) and also measured proteins in the blood called cell adhesion molecules (CAMs), which are markers of blood vessel inflammation. The analysis included 120 adults with type 2 diabetes who were treated for 32 weeks.
The study found that semaglutide improved blood vessel flexibility compared to where participants started, but this improvement was not statistically significant when compared to the placebo group, likely because the study did not have enough participants to detect a smaller difference. For the CAM markers, semaglutide (both alone and in combination with empagliflozin) significantly reduced E-Selectin, a marker linked to early blood vessel inflammation, which could be considered beneficial. However, the same groups also showed increases in VCAM-1, another CAM, which complicates the interpretation. Empagliflozin alone did not show significant effects on any of these markers. Two other CAMs (P-Selectin and ICAM-1) were unchanged in all groups.
This research suggests that the cardiovascular benefits of semaglutide and empagliflozin seen in large clinical trials may not be primarily explained by broad improvements in endothelial function, or at least not in a straightforward way. The mixed results across different CAMs indicate that these markers reflect different aspects of blood vessel health, and that semaglutide may selectively influence certain inflammatory pathways. Larger studies would be needed to determine whether the trend toward improved blood vessel flexibility with semaglutide is a true treatment effect.
Gullaksen S, Vernstrøm L, Sørensen S, Funck K, Poulsen P, Laugesen E. (2026). Effects of semaglutide and empagliflozin on markers of endothelial function in persons with type 2 diabetes: A post hoc analysis of a randomized clinical trial.. Journal of diabetes and its complications. https://doi.org/10.1016/j.jdiacomp.2026.109397