On Statistical Theory for Clinical Trials with Predictive Biomarker under Unselected Design

Project: National Science and Technology CouncilNational Science and Technology Council Academic Grants

Project Details

Abstract

Translational medicine is the “bench-to-beside” research which applies a basic laboratory discovery to prevention, diagnosis, and individualized treatment of a certain disease. After completion of the Human Genome Project (HGP), the disease biomarker can be identified. As a result, treatment modality for predictive biomarker can be developed. However, the accuracy of diagnostic devices for identification of such predictive biomarker is usually not perfect. Therefore, the treatment effects of the targeted therapy estimated from predictive biomarker clinical trials could be biased. We will propose the method to incorporate the inaccuracy of the diagnostic variance for statistical inference of the predictive biomarker clinical trials under unselected design, with respect to the censored data. Hence, this research proposal will be devoted to develop the statistical methodology for evaluation of following areas: (1) Statistical inference for two types of designs under exponential distribution model. (2) Statistical inference for two types of designs under exponential distribution model.

Project IDs

Project ID:PA10507-1180
External Project ID:MOST105-2118-M182-001
StatusFinished
Effective start/end date01/08/1631/07/17

Keywords

  • Predictive biomarker
  • Unselected design
  • EM algorithm

Fingerprint

Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.