Estimating the Time to Benefit for Therapies in Heart Failure with Reduced Ejection Fraction: A Case Study of Sacubitril-Valsartan Using Reconstructed Data from a Randomized Controlled Trial

Background: Foundational therapies in heart failure improve clinical outcomes in heart failure with a reduced ejection fraction (HFrEF). Underuse of these life-prolonging heart failure therapies, such as sacubitril-valsartan, is common in older adults and has been associated with worse clinical outcomes. Characterizing the early benefits seen with these therapies might help increase their uptake in older adults.

Objective: We applied several methods to estimate the time to benefit of an HFrEF therapy, using sacubitril-valsartan as a case study.

Methods: PARADIGM-HF was a randomized controlled study on sacubitril-valsartan versus enalapril in stable, ambulatory HFrEF patients (n = 8399). The primary endpoint, a composite of death from cardiovascular causes or a first hospitalization for heart failure, was significantly reduced (sacubitril-valsartan (21.8%) versus enalapril (26.5%), hazard ratio (HR) 0.80 (95% confidence interval [CI] 0.73-0.87). We extracted and tabulated the Kaplan-Meier (KM) curves of the primary endpoint. An individual patient dataset was then reconstructed. The following methods were applied to explore the time to benefit of sacubitril-valsartan versus enalapril: visual estimation of the point of divergence of the KM curves, statistical process control (SPC), unadjusted landmark analyses using Cox proportional hazards analysis with 30-day increments until significance was persistently achieved, and comparing the survival probabilities of the extracted life tables.

Results: Six raters visually estimated the time to benefit at a median of 60 days (interquartile range 38-10 days). Using SPC we found an early benefit from 28 days on, using the longest predefined control period of 28 days. An absolute risk reduction of 1 and 2% was found after 59 and 250 days, respectively. The reconstructed dataset provided a similar HR of 0.8004 (95% CI 0.7331-0.8739). Landmark analyses persistently showed statistical significance from 390 days and later. Survival probabilities differed from 35 days onward.

Conclusion: Using multiple approaches, the earliest benefit of sacubitril-valsartan compared to enalapril in stable HFrEF was found at about 1 month after initiation.