Dr. Tu is a Senior Biostatistician in the Canadian Cancer Trials Group with appointments in both the Departments of Community Health & Epidemiology, and Mathematics & Statistics. He came to Queen's in 1996 from the Therapeutic Products Directorate of Health Canada. His primary responsibilities are the design, management and analysis of the cancer clinical trials. He is also interested in the methodological research on the development and evaluation of statistical procedures for the design and analysis of cancer clinical trials and maintains an interest in the general mathematical theory of statistics.
Title: A Bayesian approach for identification and assessment of biomarkers predictive of cancer treatment
Time and location:14:00-15:00, May 30, 2014, Room 510 Graduate school building
Abstract: Identification and assessment of biomarkers which can be used to predict the response of a patient to a specific treatment is a very active research area of health sciences. The significance of the biomarker is usually assessed by a test of interaction between the biomarker and treatment. Many biomarkers are measured on a continuous scale but in clinical practise, it is preferred patients can be classified into two categories such as biomarker positive and negative based on a cutpoint. In this talk, I will present a Bayesian approach my colleagues and I developed recently which can provide statistical inference simultaneously on the cutpoint and biomarker-treatment interaction.