Kidney outcomes in T2D better with SGLT2i vs GLP1-RA

In adults with type 2 diabetes (T2D), treatment with a sodium-glucose cotransporter-2 inhibitor (SGLT2i) confers better protection against chronic and acute kidney events compared with a glucagon-like peptide-1 receptor agonist (GLP-1RA), according to a comparative effectiveness study with a target trial emulation design.

In a large population-based cohort of individuals with T2D, those who initiated an SGLT2i had a lower 5-year risk of chronic kidney disease (CKD) compared with those who initiated a GLP-1RA (6.7 percent vs 8.2 percent; risk ratio [RR], 0.81, 95 percent confidence interval [CI], 0.76–0.87). [JAMA Intern Med 2026;doi:10.1001/jamainternmed.2025.7409]

Similarly, the 5-year mean cumulative count (MCC) of acute kidney injury (AKI) was lower among SGLT2i initiators vs GLP-1RA initiators (25.2 vs 28.7 per 100 individuals; MCC ratio, 0.88, 95 percent CI, 0.83–0.93).

First study author Dr Simon Jensen from Aarhus University Hospital in Aarhus, Denmark, and colleagues noted that the reduced risk of CKD with SGLT2i was driven by a lower risk of sustained reduction in eGFR (RR, 0.75, 95 percent CI, 0.69–0.82) and kidney failure (RR, 0.77, 95 percent CI, 0.54–1.11) relative to GLP-1RA. The risk of severe albuminuria did not significantly differ between the two treatment groups (RR, 0.98, 95 percent CI, 0.89–1.09).

“In contrast, the secondary outcomes of albuminuria and mortality were slightly reduced in GLP-1RA initiators,” Jensen and colleagues said.

The 5-year risk of incident or progressive albuminuria was 16.1 percent with SGLT2i vs 15 percent with GLP-1RA (RR, 1.07, 95 percent CI, 1.02–1.12), while the 5-year mortality was 9.7 percent vs 9.3 percent, respectively (RR, 1.05, 95 percent CI, 0.99–1.11).

Subgroup analyses yielded consistent results, with the most pronounced CKD and AKI reductions with SGLT2i seen among individuals without pre-existing kidney disease (urine albumin-creatinine ratio <30 vs 30–300 mg/g or eGFR ≥60 vs <60 mL/min/1.73 m²).

“To our knowledge, no head-to-head randomized clinical trials have directly compared the effectiveness of SGLT2is vs GLP-1RAs in reducing CKD and AKI,” according to Jensen and colleagues.

Evidence from network meta-analyses of randomized trials comparing SGLT2i and GLP-1RA treatment are mixed. [Prim Care Diabetes 2021;15:208-211; BMJ 2021;372:m4573; Diabetologia 2021;64:2676-2686; Cardiovasc Diabetol 2022;21:42; Diabetes 2025;74:672-681]

Most trials included in these meta-analyses measured eGFR and only a few assessed albuminuria, which could have affected the results “as GLP-1RAs may exert their kidney-protective effects mainly by preventing albuminuria,” Jensen and colleagues pointed out. “This aligns with our findings, as development of incident or progressive albuminuria was less common among individuals treated with a GLP-1RA compared with an SGLT2i.”

Overall, the present study provides evidence of improved kidney outcomes with SGLT2i vs GLP-1RA in T2D, especially among individuals with a low a priori risk of kidney disease, they said.

Nevertheless, Jensen and colleagues emphasized that kidney outcomes are only one of several considerations when selecting treatment for individuals with T2D. “In addition to potentially divergent effects on albuminuria, GLP-1RAs appear more effective for stroke and peripheral artery disease, while SGLT2is seem superior in preventing heart failure.”

Future studies should explore the potential of combined treatment, as it may offer additive benefits, they added.

The present study included 36,279 SGLT2i initiators (median age 63 years, 64 percent male, 13 percent had obesity) and 18,782 GLP-1RA initiators (median age 61 years, 58 percent male, 23 percent had obesity). Diabetes duration, eGFR, and urine albumin-creatinine ratios were comparable between the two groups.

CKD (ie, 40-percent reduction in eGFR, severe albuminuria, or kidney failure) and AKI were the two coprimary outcomes. Secondary outcomes included the individual components of CKD, albuminuria, and death.