Title: Individual dynamic predictions using landmarking and joint modelling: Validation of estimators and robustness assessment
Authors: Cecile Proust-Lima - INSERM (France) [presenting]
Loic Ferrer - INSERM (France)
Abstract: After the diagnosis of a disease, one objective is to predict cumulative probabilities of events such as clinical relapse from the individual information collected up to a prediction time, usually including biomarker repeated measurements. Even before a diagnosis, cumulative probability of disease can be computed from the individual screening history. Several estimators have been proposed to calculate individual dynamic predictions, mainly from joint modelling and landmarking approaches. These approaches differ by the information used, the model assumptions and the computational complexity. To provide key elements for the development and use of individual dynamic predictions in clinical follow-up, it is essential to properly validate these estimators, quantify their variability and compare them. Motivated by the prediction of two competing causes of prostate cancer progression from the history of prostate-specific antigen, we conducted an in-depth simulation study to validate and compare the dynamic predictions derived from joint models and landmark models. After formally defining the quantity to estimate, we introduce its estimators and propose techniques to assess their uncertainty. We also compare the individual dynamic predictions derived from both approaches in terms of predictive accuracy, efficiency and robustness to model assumptions. We conclude with some recommendations.