Supplementary Material for: Effect of finerenone treatment discontinuation on kidney and cardiovascular outcomes: a FIDELITY analysis

Introduction
Finerenone reduced the risk of heart and kidney events in patients with chronic kidney disease (CKD) and type 2 diabetes (T2D) in FIDELITY, a prespecified pooled analysis combining data from the phase III FIDELIO-DKD (NCT02540993) and FIGARO-DKD (NCT02545049) trials. This FIDELITY analysis aimed to identify and assess key predictors of finerenone discontinuation and evaluate the impact of discontinuation on its efficacy in people with CKD and T2D.

Methods
Adults with CKD (urine albumin-to-creatinine ratio 30–≤5000 mg/g, estimated glomerular filtration rate [eGFR] ≥25 mL/min/1.73 m2) and T2D on optimized renin–angiotensin system inhibition were randomized 1:1 to finerenone or placebo. Baseline characteristics were identified and assessed as predictors of treatment discontinuation using a multivariate Cox proportional hazards model. Stratified Cox models with treatment discontinuation as a time-varying covariate were used to assess the effect of treatment discontinuation on composite kidney and cardiovascular (CV) outcomes (kidney: kidney failure, sustained ≥57% eGFR decrease from baseline over ≥4 weeks, or kidney-related death; CV: CV death, non-fatal myocardial infarction, non-fatal stroke, or hospitalization for heart failure).

Results
Among 12,990 participants included in the analysis, 22.8% and 21.6% prematurely discontinued treatment in the finerenone and placebo arms, respectively. Advanced age, White/Black race, lower eGFR, higher urine albumin-to-creatinine ratio, and higher serum potassium at baseline were identified as predictors of finerenone discontinuation. Crude event rates per 100 patient-years for the composite kidney and CV outcomes were lower with finerenone versus placebo under treatment (kidney: 1.09 vs 1.71; CV: 2.98 vs 3.78) as well as after discontinuation (kidney: 11.95 vs 13.67; CV: 14.07 vs 14.73). The effect of finerenone on composite kidney and CV outcomes appeared to be reduced after discontinuation (hazard ratio [HR]=0.82; 95% confidence interval [CI] 0.66–1.02; HR=0.93; 95% CI 0.79–1.09, respectively) versus the time on-treatment (HR=0.65; 95% CI 0.54–0.78; pinteraction=0.0959; HR=0.79; 95% CI 0.70–0.88; pinteraction=0.0960, respectively).

Conclusion
In FIDELITY, treatment discontinuation rates were similar in the finerenone and placebo arms. Finerenone demonstrated numerically higher kidney and CV benefits during treatment versus after discontinuation.