Hierarchical Analysis of Composite Time-to-Event End Points in Heart Failure Clinical Trials Using Time in Clinical...

Area of Science:

• Cardiology
• Biostatistics
• Clinical Trials

Background:

• Hierarchical composite endpoint analysis is crucial for evaluating heart failure treatments.
• Existing methods like win ratio, win odds, and proportion in favor of treatment address treatment effects.
• Motivations include weighting severe outcomes and combining diverse endpoint types (e.g., death, hospitalizations, continuous measures).

Purpose of the Study:

• To review established hierarchical composite endpoint analysis methods for clinical endpoints in heart failure.
• To introduce and discuss recent methods that incorporate time-dependent clinical states.
• To provide recommendations based on US Food and Drug Administration guidances.

Main Methods:

• Focus on methods applicable when all components are clinical endpoints (death, hospitalizations).
• Utilize the HF-ACTION trial (Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training) as a case study.
• Describe and analyze time-dependent methods such as pairwise win time and restricted mean time in favor of treatment.

Main Results:

• The article reviews and compares various hierarchical composite endpoint analysis methods.
• It highlights the advantages of time-dependent methods for capturing nuanced treatment effects in clinical trials.
• The HF-ACTION trial data serves to illustrate the application of these statistical approaches.

Conclusions:

• Established and novel hierarchical composite endpoint analysis methods offer valuable tools for heart failure research.
• Time-dependent analyses provide a more comprehensive understanding of treatment efficacy.
• Adherence to regulatory guidances is essential for robust clinical trial analysis.