ABSTRACT

This chapter focuses on the areas where biomarkers are developed for enhancing the effectiveness and safety of patient care by allowing physicians to tailor treatment for individual patients-an approach known as personalized medicine, or precision medicine, according to the US FDA's initiative on personalized medicine. Expressed in notation, the null hypothesis is for untargeted designs, where and p is the prevalence rate of marker-positives. Targeted designs perhaps aim on optimizing ethics rather than efficiency in terms of sample size; see discussion in Appendix 3.3. With untargeted designs, as seen above, treatment groups may be balanced with respect to the biomarker-defined cohorts, despite randomization. The marker-independent arm is just like that in untargeted designs; hence, the pooled treatment means and are estimable directly in this arm. In a targeted design, only marker-positive patients, who are predicted to be responsive, are randomized to the treatment groups.