SGLT-2 Inhibitors Versus GLP-1 Receptor Agonists Effects on Kidney and Clinical Outcomes in Veterans with Type 2 Diabetes.

Background: The primary aim compared kidney endpoints between patients with type 2 diabetes (T2D) 36 months after initiation on a sodium-glucose cotransporter-2 inhibitor (SGLT2i) or a GLP-1 receptor agonist (GLP-1RA). Secondary aims compared estimated glomerular filtration rate (eGFR), hemoglobin A1c (HbA1c), weight, and urine albumin-to-creatinine ratio (UACR) changes.

Methods: We conducted a retrospective cohort study of propensity score matched veterans with T2D, baseline eGFR>20mL/min/1.73m2, and initiated on a SGLT2i vs GLP-1RA between 4/1/2009-9/1/2020. Cox proportional hazard models were constructed to evaluate effectiveness between both groups on composite endpoint (decline of >=40% in eGFR from baseline, ESRD event, and all-cause mortality) and its components adjusting for baseline characteristics. Spline models were constructed to evaluate eGFR change and linear mixed effects models were constructed to evaluate changes in HbA1c, weight, and UACR. We used an intent-to-treat (ITT) approach as our main analysis followed by a per-protocol (PP) approach excluding veterans who discontinued or switched therapy during the study period.

Results: A total of 29,146 propensity score matched veterans were included in SGLT2i and GLP-1RA groups (14,573 per group). In the ITT and PP analyses, veterans initiated on SGLT2i had a 35% (HR=0.65; 95% CI: 0.62, 0.68) and 34% (HR=0.66; 95% CI: 0.62, 0.69) reduction in the hazard of experiencing the composite endpoint compared to veterans initiated on GLP-1RA adjusting for baseline characteristics, respectively. Between 6-36 months, we found an improved chronic eGFR slope with SGLT2i compared to GLP-1RA in both ITT and PP analyses; +1.19 mL/min/1.73 m2 (95% CI: 0.93, 1.45) and +1.29 mL/min/1.73m2 (95% CI: 1.01, 1.57), respectively. The annual difference in chronic eGFR slope in both ITT and PP analyses were +0.97 mL/min/1.73m2/year (95% CI: 0.82, 1.11) and +1.08 mL/min/1.73m2/year (95% CI: 0.92, 1.25). Improved HbA1c, weight loss and UACR were reported for both groups.

Conclusion: In this real-world study, veterans with T2D initiated on SGLT2i were associated with reduced hazard of experiencing mortality, worsening eGFR, or developing ESRD and improved glycemic, metabolic, and renal endpoints compared to GLP-1RA use.