Design and Analysis of Prognostic Biomarker Validation Studies.
*Yingye Zheng, Fred Hutchinson Cancer Research Center 


Zheng will also introduce R software for analysis and interactive tools for power calculation.

Novel markers promise to dramatically change clinical decision making in early diagnosis of cancer, prognosis and in selecting patient-specific treatments. But translating the promise into reality is not easy. Most marker based tests are imperfect, and incorrect test results can have enormous consequences in both financial and human terms. Prior to incorporating a biomarker into standard clinical care, rigorous evaluation is required. In this talk Dr. Zheng will discuss current statistical methods to evaluate the clinical performance of novel biomarkers with prospective cohort studies with time-to-event outcome, and discuss study designs related to such validation. Statistical package for calculating time-dependent accuracy of a biomarker and conducting power analysis of a prospective validation study will be introduced. The goal is to provide biomarker researchers robust and rigorous tools for conducting biomarker validation studies.