Censored quantile regression survival models with a cure proportion

A new quantile regression model for survival data is proposed that permits a positive proportion of subjects to become unsusceptible to recurrence of disease following treatment or based on other observable characteristics. In contrast to prior proposals for quantile regression estimation of censored survival models, we propose a new data augmentation approach to estimation. Our approach has computational advantages over earlier approaches proposed by Wu and Yin (2013, 2017). We compare our method with the two estimation strategies proposed by Wu and Yin and demonstrate its advantageous empirical performance in simulations. The methods are also illustrated with data from a Lung Cancer survival study. (C) 2020 Elsevier B.V. All rights reserved.

Automated summary

A new quantile regression model for survival data is proposed, allowing some subjects to become unsusceptible to disease recurrence after treatment. The new approach of data augmentation estimation has computational advantages over prior methods, demonstrating advantageous empirical performance in simulations. The proposed method is compared against existing strategies and illustrated with data from a Lung Cancer survival study.