Analyzing semi-competing risks data with missing cause of informative terminal event

Area of Science:

• Biostatistics
• Survival Analysis
• Cancer Research

Background:

• Cancer studies often involve multiple event times, such as recurrence and death, leading to semi-competing risks data.
• Existing methods for semi-competing risks data typically require the cause of the terminal event to be known.
• Missing cause-of-failure data for the terminal event presents a challenge in accurately analyzing disease progression.

Purpose of the Study:

• To propose a statistical method for handling missing cause-of-failure data in semi-competing risks analysis.
• To develop a robust approach for estimating survival functions in the presence of incomplete cause information for terminal events.
• To address the challenge of differentiating informative from non-informative terminal events when cause data is missing.

Main Methods:

• Utilized the expectation-maximization algorithm for nonparametric estimation of the terminal event time survival function with missing cause data.
• Developed a semiparametric copula model for estimating semi-competing risks data when the cause of the terminal event is missing.
• Conducted simulation studies to evaluate the performance and accuracy of the proposed methodology.

Main Results:

• The proposed method effectively handles missing cause-of-failure data in semi-competing risks scenarios.
• Simulation studies demonstrated the reliability and accuracy of the estimation techniques.
• The methodology was successfully applied to real-world data from an early-stage breast cancer study.

Conclusions:

• The developed statistical approach provides a valuable tool for analyzing semi-competing risks data with missing cause information.
• This method enhances the understanding of cancer progression and survival by accounting for incomplete data.
• The findings have significant implications for cancer research, particularly in studies with complex event structures.