Title: On the validity of time-dependent AUC estimation in the presence of cure fraction
Authors: Anouar El Ghouch - The University catholique de Louvain (Belgium) [presenting]
Abderrahim Oulhaj - United Arab Emirates University (United Arab Emirates)
Kassu Mehari Beyene - UCLouvain (Belgium)
Abstract: During the last decades, several approaches have been proposed to estimate the time-dependent area under the ROC curve (AUC) of risk tools derived from survival data. The validity of these estimators relies on some regularity assumptions among which a survival function being proper. In practice, this assumption is not always satisfied because a fraction of the population may not be susceptible to experience the event of interest even for long follow-up. Studying the sensitivity of the proposed estimators to the violation of this assumption is of substantial interest. We investigate the performance of the Li's estimator, a recently proposed estimator of the time-dependent AUC, when the population exhibits a cure fraction. Motivated from the current practice of deriving risk tools in cardiovascular disease, we also assess the loss, in terms of predictive performance, when deriving risk tools from survival models that do not acknowledge the presence of cure. The simulation results show that the Li's estimator is still valid even under the presence of cure. They also show that risk tools derived from survival models that ignore the presence of cure have smaller AUC compared to those derived from survival models that acknowledge the presence of cure.