Lifetime benefits of finerenone on cardiorenal morbidity and mortality in chronic kidney disease and type 2 diabetes: a pooled analysis of FIDELIO-DKD and FIGARO-DKD trials

Résumé fourni par la source

Chronic kidney disease (CKD) impacts nearly 800 million persons worldwide and is a leading risk factor for premature cardiovascular disease (e.g. incident heart failure), kidney failure, and mortality.1,2 On the basis of findings in placebo-controlled randomized trials,3 international guidelines support long-term use of finerenone—a non-steroidal mineralocorticoid receptor antagonist—to reduce cardiorenal events in persons with CKD, type 2 diabetes, and albuminuria.4 Improving understanding of expected lifetime benefits on this broad range of cardiorenal risk may enhance personalized decisions around implementation of finerenone in clinical care. In this participant-level analysis of FIDELITY (Finerenone in Chronic Kidney Disease and Type 2 Diabetes: Combined FIDELIO-DKD and FIGARO-DKD Trial Programme Analysis), we estimated gains in survival free of cardiorenal morbidity and mortality with long-term use of finerenone in persons with CKD and type 2 diabetes. The design and primary findings of the FIDELITY programme have been previously reported.3 In brief, FIDELITY was a prespecified pooled analysis of the complementary, randomized, double-blind, and placebo-controlled FIDELIO-DKD (Finerenone in Reducing Kidney Failure and Disease Progression in Diabetic Kidney Disease) and FIGARO-DKD (Finerenone in Reducing Cardiovascular Mortality and Morbidity in Diabetic Kidney Disease) trials, which evaluated the efficacy of finerenone in persons with CKD, type 2 diabetes, and albuminuria.3 Study protocols were approved by local institutional review boards or ethics committees at each site, and all participants provided informed consent. In this analysis, we estimated the effects of finerenone on long-term survival free from cardiorenal morbidity and mortality using validated, age-based (actuarial), non-parametric restricted mean survival time methods.5 We evaluated a cardiorenal composite outcome comprised of all-cause death and the individual non-fatal components of the prespecified cardiovascular (heart failure hospitalization, myocardial infarction, or stroke) and kidney [sustained ≥57% decrease in estimated glomerular filtration rate (eGFR) or kidney failure] composite endpoints.3 First, the effect (and 95% CI) of finerenone vs placebo on the composite cardiorenal outcome was estimated through Cox proportional hazards regression, stratified for trial and the protocol-specified stratification factors.3 Second, we estimated residual survival free of the composite outcome at all enrolled ages, separately in the finerenone and placebo arms. Third, we constructed lifetime event-free survival curves by treatment arm using Kaplan–Meier methods; projected event-free survival was estimated as the difference in the area under the survival curves, up to 90 years. Fourth, we repeated this process to estimate treatment effects on residual event-free survival at all ages between 50 and 80 years. A locally weighted scatterplot smoothing procedure was applied to model event-free survival gains (and 95% CI) with finerenone,6 across the spectrum of age. Analyses included 12 990 participants validly randomized at sites without critical Good Clinical Practice violations and followed intention-to-treat principles.3 Trial regimen discontinuations (28% for any reason in both treatment arms) and censoring reasons are previously reported.3 P values < .05 were considered statistically significant; statistical analyses were conducted using Stata, version 19.5 (StataCorp, LLC). Over a median follow-up of 3.0 (2.3, 3.8) years, the incidence rate of the composite cardiorenal outcome was lower with finerenone (71.8 events per 1000 person-years) vs placebo (85.4 events per 1000 person-years; hazard ratio, 0.84; 95% CI, 0.78–0.90). In age-based analyses, mean survival free from cardiorenal morbidity or mortality was 10.2 years (95% CI, 9.7–10.7) with placebo and 11.5 years (95% CI, 11.0–12.0) with finerenone, representing an additional 1.3 years (95% CI, 0.6–2.0 years; P < .001) of event-free survival from age 65 years (Figure 1A). Similar findings from age 65 years were observed with baseline eGFR < 60 (1.0 years; 95% CI, 0.2–1.9; P = .017) and ≥60 (1.9 years; 95% CI, 0.6–3.2; P = .004) mL/min/1.73 m2 and baseline urine albumin-to-creatinine ratio < 300 (1.2 years; 95% CI, 0.0–2.3; P = .042) and ≥300 (1.4 years; 95% CI, 0.5–2.2; P = .001) mg/g. From age 65 years, similar increments in survival free of composite kidney events (1.0 years; 95% CI, 0.2–1.9; P = .014), and composite cardiovascular events (1.3 years; 95% CI, 0.5–2.1; P = .002), were also observed. Cardiorenal event-free survival gains were observed across a broad age range, from 1.2 years (95% CI, 0.2–2.3) at age 55 years to 0.8 years (95% CI, 0.1–1.4) at age 75 years (Figure 1B). Estimated lifetime benefits of finerenone on survival free of cardiorenal events in FIDELITY participants with type 2 diabetes and chronic kidney disease. (A) Survival free from cardiorenal morbidity and mortality after age 65 years. (B) lifetime gains in cardiorenal event-free survival, by baseline age. Panel (A) shows the probability of survival free of cardiorenal morbidity and mortality in FIDELITY after age 65 years, by treatment arm. The origin, start, and end times for the survival analysis were age 65 years, age at study entry, and age at censoring or outcome of interest, respectively. The 95% CI was calculated using the variance of the between-group difference, estimated as the sum of the treatment arm-specific variances. The P value for the between-group difference in mean event-free survival was calculated from a Z score under the assumption of normality. Panel (B) shows the estimated survival gains (in years) free of cardiorenal morbidity and mortality with finerenone vs placebo in FIDELITY according to baseline age, with the solid line reflecting the smoothed estimate (tricube weighting function with bandwidth of 0.8) and the dashed lines indicating 95% CI. FIDELITY, Finerenone in Chronic Kidney Disease and Type 2 Diabetes: Combined FIDELIO-DKD and FIGARO-DKD Trial Programme Analysis Based on this FIDELITY analysis, treatment with finerenone is projected to substantially extend survival free from cardiorenal morbidity and mortality in persons with CKD and type 2 diabetes. Meaningful gains in residual lifespan were estimated across a broad age range, consistent with the prior observation of similar relative treatment benefits of finerenone across the age spectrum in FIDELITY.7 However, absolute gains in event-free survival were expectedly higher among younger adults given longer remaining lifespan. These non-parametric, time-based estimates may provide a complementary and more digestible perspective of expected treatment effect when considered alongside the findings of pivotal clinical trials. Indeed, outcomes trials that inform guideline recommendations and regulatory approvals are typically conducted over relatively short timescales and traditionally report findings in the form of hazard ratios, which may be challenging to translate at the bedside. Hence, these estimates of long-term absolute benefit may enhance individualized decisions around implementation of finerenone in eligible persons with CKD. Such estimates may be especially informative for persons with CKD, who may be asymptomatic or have few health status impairments at the time of treatment initiation. These findings may additionally support health policy initiatives to ensure long-term access to disease-modifying therapies. This analysis has some limitations. Although intention-to-treat approaches considering in-trial discontinuation of study medication may approximate treatment effectiveness in routine care, longitudinal adherence may be different in non-trial settings due to multiple factors, including access, tolerability, and cost. Alternatively, intention-to-treat methods may underestimate treatment benefits with optimal adherence. As such, these findings may not be f