Abstract:
|
Assumption of proportional hazard is a common practice in clinical trials involving time-to-event endpoints, which includes indications like Cardio-vascular and Oncology. In general, the assumption holds valid for the study duration. However, with recent development in immuno-oncology field, where the anti-cancer therapy takes longer to show its effect on efficacy, it is getting relevant to consider non-proportional hazard between the active arm involving immune-oncology treatment and the control arm involving some other treatment. Different methods, namely weighted log-rank test, weighted Kaplan-Maier method etc. have been investigated to analyze such data. However, no comparative study is available using modulated Markov models and the modulated renewal process. In this project, we intend to explore the proposed models in simulated and real-life data. Efforts will be made to compare the results and evaluate the performance of the models against the standard methodologies. This project will help us to understand the behavior of the investigated methods and therefore, will provide recommendations for an adequate statistical analysis strategy and meaningful interpretation.
|