Nonparametric estimation of time-dependent ROC curves conditional on a continuous covariate

Area of Science:

• Biostatistics
• Medical Informatics
• Epidemiology

Background:

• Receiver-operating characteristic (ROC) curves are standard for diagnostic biomarker performance evaluation.
• Traditional ROC analysis is insufficient for time-to-event data due to dynamic disease status and censoring.
• Time-dependent extensions are needed to accurately assess biomarkers in survival studies.

Purpose of the Study:

• To develop novel nonparametric estimators for cumulative/dynamic time-dependent ROC curves.
• To account for the modifying effects of covariates on biomarker discrimination.
• To provide robust methods for evaluating biomarkers in the presence of time-dependent outcomes and censoring.

Main Methods:

• Development of new nonparametric estimators for time-dependent ROC curves.
• Incorporation of time-dependent covariates and covariate-dependent censoring.
• Validation through simulation studies and application to real patient data.

Main Results:

• The proposed estimators effectively handle time-dependent disease status and censored data.
• New methods accurately assess biomarker discriminatory power in dynamic settings.
• The estimators demonstrate flexibility in accommodating complex censoring patterns.

Conclusions:

• The novel estimators offer a significant advancement in evaluating diagnostic biomarkers for time-to-event outcomes.
• These methods enhance the reliability of biomarker assessment in survival analysis.
• The approach is applicable to various clinical settings with time-dependent disease progression and censoring.