Comparative Effectiveness of SGLT2 Inhibitor and GLP-1 Receptor Agonist on Kidney and Cardiovascular Outcomes by Kidney Failure Risk

KEY POINTS: In veterans with type 2 diabetes and low kidney failure risk, sodium-glucose cotransporter 2 inhibitors (SGLT2is) were more kidney protective while glucagon-like peptide-1 receptor agonists (GLP-1 RAs) were more cardioprotective. For cardiovascular-kidney-metabolic outcomes, GLP-1 RAs were more protective at moderate kidney failure risk and SGLT2is were more protective at high kidney failure risk. The kidney failure risk equation might be a clinically useful tool to guide therapy in type 2 diabetes. BACKGROUND: Head-to-head comparisons of non-exendin glucagon-like peptide-1 receptor agonists (GLP-1 RAs) and sodium-glucose cotransporter 2 inhibitors (SGLT2is) on kidney failure or cardiovascular-kidney-metabolic (CKM) composite end points are lacking. Whether kidney failure risk modifies the comparative effectiveness of GLP-1 RA versus SGLT2i is clinically relevant. METHODS: We defined a national veterans cohort with type 2 diabetes who initiated an SGLT2i, non-exendin GLP-1 RA, or insulin glargine between January 1, 2018, and December 31, 2021. After applying inverse probability of treatment weighting to balance baseline characteristics, outcomes were compared across new-user groups through March 31, 2023. Outcomes included kidney failure (stage 5 CKD or long-term KRT), major adverse cardiovascular events (MACEs: heart failure, myocardial infarction, or stroke), CKM composite (kidney failure or MACE), all-cause death, and composites of outcomes with death. We tested for effect modification by the kidney failure risk equation (KFRE) score on comparative pairwise drug effectiveness. RESULTS: Out of 160,428 veterans, 53%, 14%, and 34% were new users of SGLT2i, GLP-1 RA, and insulin glargine, respectively. Relative to GLP-1 RA new-users, SGLT2i new-users had similar mortality risk, a trend toward lower kidney failure risk (hazard ratio [HR], 0.89; 95% confidence interval [CI], 0.74 to 1.06), but higher risk of MACE (HR, 1.14; 95% CI, 1.09 to 1.20) and CKM composite (HR, 1.13; 95% CI, 1.08 to 1.19). The effect of SGLT2i versus GLP-1 RA differed significantly between moderate-risk (2%-6%) and high-risk (≥6%) KFRE subgroups for all outcomes except all-cause death. In those with moderate-risk KFRE, GLP-1 RA seemed more protective for kidney failure, MACE, and CKM composite, while SGLT2i appeared more protective in those with high-risk KFRE. CONCLUSIONS: Compared with GLP-1 RA, SGLT2i had comparable risks of mortality, perhaps a lower risk of kidney failure, but modestly higher risk of cardiovascular events in the entire cohort. However, GLP-1 RA were more beneficial in those with moderate kidney failure risk and SGLT2i more beneficial in those with high kidney failure risk.