Joint longitudinal and time-to-event cure models for the assessment of being cured
mercredi 16 mai 2018, 11h00 - 12h00
In clinical studies with a time-to-event endpoint, we may have, in some situations, to take into account the presence of two sub-populations: the patients who will not experience the event of interest and are said to be ”cured”, and those who will develop this event, and are said to be ”susceptible”. However, a difficulty is that the cure status is unobserved in (right-)censored patients. Cure models have been developed to address this situation.
While most of the work on cure models focus on the time-to-event for the uncured patients (latency) or on the baseline probability to cure or not (incidence), we focus in this research on the conditional probability of being cured given that the patient is event-free until a certain time and taking into account the information collected for a longitudinal biomarker until that time point. We therefore consider joint models for the longitudinal and survival data given a cure fraction. These models are composed of a mixed model to fit the trajectory of longitudinal measurement, a mixture cure model and a shared latent structure linking these sub-models.
In this talk, we will discuss several possibilities for this shared latent structure, present the results of a simulation study comparing the various approaches and illustrate those results with real data.