Large Effect Size in Composite Kidney Outcomes than in Majors Cardiovascular Events of SGLT2 inhibitors compared with GLP‐1 RAs: A Pooled Analysis of Type 2 Diabetes Trials

International audienceAim: Sodium glucose cotransporter 2 (SGLT2) inhibitors and glucagon-like peptide 1 receptor agonists (GLP-1 RAs) reduce cardiorenal events in at-risk patients with type 2 diabetes (T2D). However, uncertainty remains regarding difference in treatment effect sizes between major adverse cardiovascular event (MACE-3) and composite kidney outcome (CKO). We aimed to compare treatment effect sizes between CKO and MACE-3 outcomes, and to investigate relationship between treatment effects on CKO and MACE-3 in T2D patients.Materials and methods: We performed a MEDLINE database search up to December 31, 2021 to identify all placebo-controlled phase 3 trials which investigate the efficacy of glucose lowering interventions, and selected those reporting results for CKO and MACE-3. Hazard ratios (HRs) with 95% CI for both outcomes were extracted for each trial, and we evaluated differences in treatment effect sizes by using a ratio of HRs (rHR): the HR for CKO to that for MACE-3. A random-effect meta-analysis was used to obtain the overall rHR across trials and according subgroup. We investigated the relationship between treatment effects on CKO and MACE-3 by the coefficient of determination (R2 ) using a weighted meta-regression. The study protocol was registered on PROSPERO (CRD42022299690).Results: A total of 12 studies fulfilled the prespecified criteria, comprising 104,987 patients with T2D were included. On average, treatment effect sizes were 17% greater with CKO than with MACE-3 (rHR, 0.83; 95% CI, 0.74 to 0.92; I2 = 50%; P = 0.03; τ2 = 0.0161), especially for trials of iSGLT2 than for GLP-1 RAs. For secondary outcomes, treatment effect size was 22%, 21%, 16% and 9% greater with CKO than with myocardial infarction, strokes, death from cardiovascular cause, and hospitalization for heart disease, respectively. MACE-3 and CKO were moderately correlated (ρ=0.40; p-value = 0.21), with only 11% (95% CI, 1% to 54%) of the variability in the MACE-3 effect could be explained by the variability in the CKO effect.Conclusion: In T2D patients, treatment effect sizes were greater for kidney than for macrovascular (MACE-3) outcomes, with important differences according to the drugs considered. CKO and MACE-3 are independent. Caution must be taken when interpreting CKO in the absence of MACE-3 data