FC083: Finerenone and Canagliflozin in the Treatment of Chronic Kidney Disease and Type 2 Diabetes: Matching-Adjusted Indirect Treatment Comparison of Fidelio-DKD and Credence. (3rd May 2022)

• Record Type: Journal Article
• Authors: Cherney, David; Folkerts, Kerstin; Mernagh, Paul; Nikodem, Mateusz; Pawlitschko, Joerg; Rossing, Peter
• Is Part Of: Nephrology dialysis transplantation. Volume 37(2022)Supplement 3
• Publication Date: 2022-05-03
• Subjects: Nephrology -- Periodicals; Hemodialysis -- Periodicals; Kidneys -- Transplantation -- Periodicals; Hemodialysis; Kidneys -- Transplantation; Nephrology; Periodicals; 616.61
• DOI: 10.1093/ndt/gfac115.003
• Languages: English
• ISSNs: 0931-0509

Abstract

Abstract: BACKGROUND AND AIMS: Finerenone (FIN: an oral, nonsteroidal mineralocorticoid receptor antagonist) and canagliflozin (CAN: a sodium-glucose cotransporter 2 inhibitor) demonstrated cardiorenal efficacy in patients with chronic kidney disease and type 2 diabetes on top of renin–angiotensin system blockade in phase III placebo-controlled studies (FIDELIO-DKD [NCT02540993] for FIN; CREDENCE [NCT02065791] for CAN) [3, 2 ]. In the absence of head-to-head studies, we assessed FIN and CAN in this patient population while appropriately accounting for meaningful differences between the trials. METHOD: We performed an anchored matching-adjusted indirect comparison (MAIC), with placebo (PBO) as a common comparator, to generate measures of efficacy and safety while accounting for differences between the two study populations [3 ]. Individual patient-level data from FIDELIO-DKD and published data from CREDENCE were used [1, [2 ]. Weights were calculated and assigned to each patient in FIDELIO-DKD so the weighted population of FIDELIO-DKD matched that of CREDENCE for selected baseline characteristics, e.g. mean estimated glomerular filtration rate (eGFR; 44.3 and 56.2 mL/min/1.73 m 2 in FIDELIO-DKD and CREDENCE, respectively). Weights were obtained from a logistic regression model of the odds of enrolment in CREDENCE and FIDELIO-DKD for baseline characteristics believed to be effect modifiers. Hazard ratios (HRs) with 95% confidence intervals (CIs) comparing FIN and PBO for time-to-event endpoints were estimated based on the weighted population of FIDELIO-DKD. HRs with 95% CIs comparing FIN and CAN were then calculated from results of the previous step and published data from CREDENCE [2 ]. This analysis evaluated the cardiorenal composite endpoint from CREDENCE (kidney failure [dialysis, transplantation or sustained eGFR < 15 mL/min/1.73 m 2 ), a doubling of serum creatinine level or death from kidney or cardiovascular disease) [2 ] and hyperkalaemia. A sensitivity analysis that matched patients based on their history of heart failure was performed. RESULTS: Calculation of the weights for the FIDELIO-DKD population ( N = 5674) resulted in an effective sample size of 1288 for the pseudo-population formed by the weighting to compare with the CREDENCE population ( N = 4401). For the cardiorenal composite endpoint, the HR (95% CI) for FIN and PBO based on reweighted FIDELIO-DKD data was 0.72 (0.59–0.90) and the MAIC-based HR (95% CI) for FIN and CAN was 1.03 (0.79–1.35) (P = 0.802). For hyperkalaemia, the MAIC-based HR (95% CI) for FIN and CAN was 2.25 (1.67–3.03) (P < 0.001). Similar efficacy and safety results were demonstrated in the sensitivity analysis. CONCLUSION: The MAIC for FIDELIO-DKD and CREDENCE enabled more robust assessment of FIN and CAN when a similar patient population was considered. There was no evidence of a significant difference between FIN and CAN in the cardiorenal composite endpoint as assessed in CREDENCE. These results are consistent with a recent analysis using a different payer-accepted method that also accounted for differences between the FIDELIO-DKD and CREDENCE inclusion criteria and endpoints [3, 4 ]. GRAPHICAL ABSTRACT: