Design and Analysis of Clinical Trials, and Analysis of Correlated Data

[摘要] Today, human disorders are thought to be heterogeneous withregard to their molecular and genomic properties. Recent advancesin biotechnology have resulted in a shift toward molecularly targetedagent. These new therapeutics works on specific genetic pathwayswhich are only dysregulated in some smaller subset of patients with thedisease. New and general statistical methods are needed so that geneticsor other biomarkers can be used to assist the analysis of clinical data forpersonalized medicine. The challenge raised by personalized medicineis how to find the predictive biomarkers that can identify patientswho are likely or unlikely to benefit from a specific treatment withinlarge number of biomarkers to be considered and how to incorporatebiomarkers information in randomized clinical trial while consideringissues of both efficiency and medical ethics. Several statistical methodsfor evaluating and comparing biomarkers for patient treatmentselection have been developed. Currently available methods are farfrom sufficient to deal with the complications of patient treatmentselection in clinical trial practice. A good design of clinical trials can useless number of total patients, send more patient to the better treatmentarms, and expose less patients to highly toxic treatments. We will workin these areas to help advance clinical trial design.