TOPLINE

A real-world study of 4554 patients with type 2 diabetes (T2D) and stage 3 or 4 chronic kidney disease (CKD) found that those who were started on a GLP-1-based drug, including semaglutide and tirzepatide, had a 29% lower relative risk of progressing to advanced kidney disease than those who were started on finerenone. The absolute risk difference over 2.5 years was small and not statistically significant.

METHODOLOGY

• GLP-1-based drugs and finerenone are both guideline-recommended add-on treatments to renin-angiotensin-aldosterone system inhibitors and SGLT2 inhibitors in patients with T2D and CKD. However, no head-to-head trial has compared them, and doctors often choose between them based on cost, how many medicines a patient already takes, and patients’ preferences.
• Researchers conducted a retrospective cohort analysis of data obtained from a global health database to compare kidney outcomes with GLP-1-based drugs vs finerenone in patients with T2D and stage 3 or 4 CKD who began therapy between July 2021 and November 2025, following up on them for up to 2.5 years.
• After 1:1 matching, the analysis included 4554 adults (43.3% women; mean age, 70.7 years; mean estimated glomerular filtration rate [eGFR], 39.4 mL/min/1.73 m²): 2277 who were started on GLP-1-based drugs such as semaglutide, dulaglutide, liraglutide, or tirzepatide and 2277 matched patients who were started on finerenone. All had at least one clinic visit before starting treatment.
• The study excluded patients with very advanced kidney disease at baseline, an eGFR above 60 mL/min/1.73 m², an A1c level above 9%, and recent prior use of either of the drugs.
• The primary endpoint was a composite kidney outcome of end-stage kidney disease (ESKD), stage 5 CKD, and dialysis initiation.

TAKEAWAY

• The primary endpoint occurred in 110 patients (4.8%) who were started on a GLP-1-based drug and 136 patients (6.0%) who were started on finerenone, corresponding to a 29% lower relative hazard with GLP-1-based drugs (hazard ratio [HR], 0.71; P = .007). However, the absolute risk difference at the end of follow-up (up to 2.5 years) was -1.14 percentage points, which was not statistically significant (P = .088).
• GLP-1-based drugs were also linked to a lower relative hazard for the composite of ESKD and stage 5 CKD (HR, 0.66) and for the composite of kidney events or death (HR, 0.80) compared with finerenone. However, dialysis initiation, all-cause death, and acute kidney injury did not differ significantly between the groups.
• Kidney function declined more slowly with GLP-1-based drugs than with finerenone over the 2.5 years of follow-up (-0.43 vs -1.47 mL/min/1.73 m² per year; P = .004), but the gap disappeared once the first six months were excluded (P = .899).
• The association between GLP-1-based drugs and the primary endpoint appeared stronger in patients with a BMI of 30 or higher (HR, 0.55) than in those with a BMI below 30 (HR, 0.84; not statistically significant).

IN PRACTICE

"Given the mechanistic heterogeneity of CKD in T2D, distinct patient profiles may benefit differently from GLP-1/GIP [glucose-dependent insulinotropic polypeptide] RAs [receptor agonists] and nonsteroidal MRAs [mineralocorticoid receptor antagonists]. Our findings, which did not find a consistently significant effect favoring one treatment over the other, might stem from between-patient heterogeneity in treatment effect, supporting this concept," the authors wrote.

SOURCE

The study was led by Dvora R. Sehtman-Shachar, Hebrew University of Jerusalem, Israel. It was published online in Diabetes, Obesity and Metabolism.

LIMITATIONS

Information on the albuminuria status, a key marker of kidney damage, was missing for most patients and could not be used in matching. Finerenone users were followed up for less time, partly because it was approved more recently. The cohort also represented a selected group of patients with T2D and CKD, who may be eligible for either of the drugs.

DISCLOSURES

The study received no specific funding. Several authors reported serving on advisory boards or receiving research grants, travel support, or lecture fees from pharmaceutical companies. One author was an employee of Regeneron Pharmaceuticals Inc., but the company had no role in the study. 

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.