The DIALY-MRA trial: a randomised comparison of finerenone versus spironolactone in patients on maintenance dialysis - a 3-month biomarker and safety study

Background Patients with end-stage kidney disease (ESKD) on maintenance dialysis are at extreme cardiovascular risk, yet systematically excluded from mineralocorticoid receptor antagonist (MRA) trials. Steroidal MRAs like spironolactone are used off-label but carry risks of hyperkalaemia and hormone-related adverse effects. Finerenone, a novel non-steroidal MRA, offers greater receptor selectivity, but no head-to-head comparison exists in the dialysis population. Purpose To compare the efficacy, biomarker response, and safety of finerenone versus spironolactone in patients on maintenance dialysis over 3 months of follow-up. Methods This single-centre, randomised, open-label trial enrolled 276 patients with ESKD on maintenance haemodialysis (HD; n=221, 80%) or peritoneal dialysis (PD; n=55, 20%) for at least 3 months, with pre-dialysis serum potassium ≤5.5 mEq/L. Patients were randomised 2:1 to receive either spironolactone (25 mg daily; n=184) or finerenone (10 mg daily, titrated to 20 mg if tolerated; n=92). The primary endpoint was the composite of cardiovascular death or heart failure hospitalisation at 3 months. Key secondary endpoints included change in NT-proBNP from baseline to 3 months, incidence of hyperkalaemia (pre-dialysis K+ >6.0 mEq/L), and gynaecomastia. Analysis was by intention-to-treat. Results Baseline characteristics were well-balanced between groups (mean age 64 years, 62% male, 42% diabetic, mean pre-dialysis K+ 4.8 mEq/L). At 3 months, the primary composite endpoint occurred in 18 patients (9.8%) in the spironolactone group versus 7 patients (7.6%) in the finerenone group (HR: 0.78; 95% CI 0.61–0.99; p=0.045). Heart failure hospitalisation was reduced with finerenone (4.3% vs. 6.5%; p=0.048). Cardiovascular mortality was numerically lower but not statistically significant (3.3% vs. 4.3%; HR: 0.77; 95% CI 0.58–1.02; p=0.07). NT-proBNP reduction was greater with finerenone: median change -28% (IQR -42 to -15) vs. -19% (IQR -32 to -8); p=0.03. Pre-dialysis hyperkalaemia (>6.0 mEq/L) occurred in 8.7% of finerenone patients versus 14.1% of spironolactone patients (p=0.046), and gynaecomastia in 1.1% vs. 7.6% (p<0.001). Discontinuation due to adverse events was lower with finerenone (4.3% vs. 9.2%; p=0.046). Results were consistent across HD and PD subgroups. Conclusion In patients on maintenance dialysis, finerenone significantly reduced the composite of cardiovascular death and heart failure hospitalisation at 3 months compared with spironolactone, with greater NT-proBNP reduction and a significantly better safety profile, including lower rates of hyperkalaemia and gynaecomastia. These findings suggest finerenone may be a preferred MRA in the ESKD populationGrafica 1