SGLT2 inhibitors provide the greatest risk reduction for kidney outcomes in type 2 diabetesSGLT2 Inhibitors Lead in Kidney Protection for Diabetes and CKD

BackgroundDespite optimal renin-angiotensin system (RAS) blockade, patients with diabetic kidney disease (DKD) face substantial residual cardiorenal risk. The 2024 FLOW trial established semaglutide as the first GLP-1 receptor agonist with a dedicated primary renal outcome benefit. We conducted a Bayesian network meta-analysis comparing SGLT2 inhibitors, semaglutide, and finerenone—the three therapies with contemporary phase 3 cardiorenal evidence—to define their comparative efficacy and safety hierarchy in type 2 diabetes with chronic kidney disease.MethodsWe systematically searched PubMed, Embase, and the Cochrane Library through 31 July 2025 for phase 3 randomized trials reporting kidney, cardiovascular, or safety outcomes. We treated FIDELIO-DKD and FIGARO-DKD as separate evidence nodes to preserve between-trial heterogeneity. For DAPA-CKD and EMPA-KIDNEY, only the pre-specified type 2 diabetes subgroup data were used. Efficacy was synthesized using hazard ratios (HRs) with 95% credible intervals (CrIs); safety was modeled using a binomial likelihood for treatment discontinuation. Treatment hierarchy was quantified by the surface under the cumulative ranking curve (SUCRA). Four pre-specified sensitivity analyses evaluated robustness (±SCORED trial; fixed-vs. random-effects models).ResultsSix trial-level datasets (CREDENCE, DAPA-CKD T2D subgroup, EMPA-KIDNEY T2D subgroup, FIDELIO-DKD, FIGARO-DKD, FLOW; N = 20,521) were included in the main analysis. All trials were at low risk of bias. Kidney outcomes: SGLT2 inhibitors demonstrated the greatest risk reduction (HR 0.70, 95% CrI 0.64–0.76; SUCRA 94.5%), followed by semaglutide (HR 0.76, 95% CrI 0.66–0.88; SUCRA 68.0%) and finerenone (HR 0.84, 95% CrI 0.77–0.92; SUCRA 37.5%). The indirect comparison between SGLT2 inhibitors and finerenone reached statistical significance (HR 0.83, 95% CrI 0.73–0.95; posterior probability of SGLT2i superiority >99.9% [fixed effects]; 96% [random effects]). In contrast to prior NMAs that pooled finerenone trials via the FIDELITY estimate (HR 0.77), our approach of separating FIDELIO-DKD and FIGARO-DKD yielded a pooled finerenone HR of 0.84, allowing this indirect detection. This finding is sensitive to the analytic decision to separate the two finerenone trials and should be interpreted as hypothesis-generating. Semaglutide was numerically but not statistically superior to finerenone (HR 0.90, 95% CrI 0.76–1.07; posterior probability = 88%).Cardiovascular outcomesAll three active agents significantly reduced cardiovascular risk versus placebo, with no statistically significant differences between active agents. Robustness: All key findings were stable across four sensitivity analyses; the posterior probability of SGLT2 inhibitor superiority to finerenone for kidney protection remained ≥96% in all specifications.ConclusionSGLT2 inhibitors provide the greatest kidney protection in this network; when the two finerenone trials are analyzed separately, an indirect statistically significant SGLT2i advantage emerges (HR 0.83, 95% CrI 0.73–0.95 [fixed effects]). This finding is analytic-choice-dependent and hypothesis-generating. Semaglutide provides substantial renal protection that is numerically superior to finerenone. These findings support a four-pillar framework—RAS blockade and SGLT2 inhibition as the foundational combination, with semaglutide and finerenone as complementary phenotype-matched additions—and provide a probabilistic evidence base for individualized prescribing.