Abstract:
|
Identifying predictive biomarkers of clinical outcomes is an area of increased interest in precision medicine. There are several biomarker-driven designs that have been implemented in oncology. One common design is the stratified biomarker approach which is the only prospective design that investigates the biomarker-treatment interaction on a clinical endpoint. Some tissue biomarkers are measured with error and biomarker misclassification has profound adverse effects on the coverage of confidence intervals, power of the tests, and required sample sizes. In the talk, methods that will adjust for measurement error for continuous and time to event endpoints will be presented.
|