|
Abstract:
|
Statistical tests for biomarker identification and classification methods for patients grouping are two important topics in adaptive designs of clinical trials. In this article, we evaluate three test methods for biomarker identification: a model-based identification method, the popular t-test, and the nonparametric Wilconxon Rank Sum test. For selecting the best classification methods in Stage 2 of an adaptive design, we examine classification methods including the recent developed machine learning approaches such as Random Forest, the Lasso and Elastic-Net Regularized Generalized Linear Models (Glmnet) , Support Vector Machine (SVM), the Gradient Boosting Machine (GBM), and the Extreme Gradient Boosting (XGBoost). Statistical simulations are carried out in our study to assess the performance of biomarker identification methods and the classification methods. The best identification method and the classification method will be selected based on the True Positive Rate (TPR) and the False Positive Rate (FPR). The optimal test method for gene identification and classification method for patients grouping will be applied to the Adaptive Signature Design (ASD) for the purpose of evalua
|
