Multi-State Models for the Time to Event Post-Transplantation Cancer Data: A Competing Risks Approach

In clinical research studies complex end points are most common. Especially in studies on diseased patients undergo therapy, a relapse can occur, or patients can die after relapse or without former relapse (death). Sometimes, endpoints can be reasonably combined in a composite endpoint, as relapse and death combined into disease-free survival (DFS). In this case, standard survival techniques, as Kaplan-Meier estimation of the DFS probability, can be applied. Often, interest focuses on endpoints for which competing risks are present; a competing risks model plays a special case of a multistate model. A more complex multistate model is required when the effects of events occurring in the course of the study on further disease progress. Another endpoint of interest is time to experience multiple episodes; the multistate model used for analysis must be adapted for this event structure. The aim of this report is to explain use and interpretation of both non proportional hazard Cox (non PH) and proportional hazard Cox (PH)type multistate models for the suitability of assessing different covariate effects in situations of multiple endpoints and also different baseline hazard assumptions in a comparison manner for the European Group for Blood and Marrow Transplantation (EBMT) competing risks data.