Title: Joint Models and Tests for Time to Tumor Recurrence and Disease Stage in Oncology Clinical Trials
Advisor: Professor Robert Keener
Committee Members: Associate Professor Ed Ionides, Professor George Michailidis, Professor Alexander Tsodikov (Biostatistics)
Abstract: In this presentation, a clinical trial with bladder cancer patients who went through surgery and were followed up for tumor recurrence will be discussed. The surgery was conducted on patients with an early cancer stage. There was a control group using standard procedures, and an experimental group with a drug designed to enhance observation of suspected cancer lesions. One of the primary objectives of the study was to evaluate and compare the time to tumor recurrence (or progression) of patients in the control and experimental groups. But at the time of tumor recurrence, the disease stage was also evaluated. The majority of these stages were less advanced, while some patients progressed to more aggressive stages. The stage of the disease at recurrence significantly impacts future treatment and quality of life. Therefore, analyzing and comparing the time to tumor recurrence and the stage at recurrence jointly makes more sense than an analysis based primarily on the time to recurrence. This trial served as the motivation for the current research. Parametric and semi-parametric methods to model the joint distribution of recurrence stage and time to recurrence will be discussed. The relevant literature and solution approaches will be briefly reviewed. Using these models, the proposed research will develop methods to estimate and test treatment efficacy, and study the performance of the tests and estimators derived.