Semiparametric model for semi-competing risks data with application to breast cancer study

Area of Science:

• Biostatistics
• Cancer Epidemiology
• Translational Oncology

Background:

• Understanding cancer progression and death is crucial for clinical and research advancements.
• Informative censoring, where death affects cancer progression data, presents analytical challenges.
• Existing methods struggle with the dependent relationship between cancer relapse and mortality.

Purpose of the Study:

• To develop a robust statistical method for analyzing cancer patient data with informative censoring.
• To simultaneously estimate cancer relapse survival and the association between relapse and death.
• To address the limitations of traditional survival analysis in complex cancer data.

Main Methods:

• Utilized a copula modeling approach for dependent survival data.
• Employed an exploratory diagnostic strategy for copula model selection.
• Developed an inference procedure for simultaneous estimation of marginal survival and association parameters.
• Validated the method through simulation studies and application to breast cancer data.

Main Results:

• The proposed estimators demonstrate consistency and weak convergence.
• The method effectively handles informative censoring in cancer progression data.
• Simulations confirmed the finite sample performance of the statistical approach.

Conclusions:

• The developed copula-based method provides a reliable tool for analyzing cancer survival data with informative censoring.
• This approach enhances the understanding of cancer disease processes in clinical and epidemiological research.
• Accurate estimation of survival functions and association parameters is vital for advancing cancer treatment strategies.