ABSTRACT
The extensive application of advanced molecular technology has generated a vast amount of data on human biology in a very short time. For example, the Human Genome Project, completed in 2003, has successfully sequenced approximately 3 billion base pairs and identied a majority of the genes in the human genome [1]. Shortly thereafter, the advancement of the next generation sequencing methods has also made genomic sequencing more
CONTENTS
6.1 Introduction ................................................................................................ 149 6.2 Motivations and Concepts ........................................................................ 150 6.3 Adaptive Staggered Dose Design ............................................................ 152
6.3.1 General Design ............................................................................... 152 6.3.2 Specic Design ............................................................................... 153
6.4 Other Design Specications ..................................................................... 156 6.4.1 Family-Wise Type I Error .............................................................. 156 6.4.2 Calculation of Stopping Boundaries ........................................... 157 6.4.3 Statistical Power ............................................................................. 158
6.5 Operating Characteristics ......................................................................... 160 6.5.1 Numerical Studies ......................................................................... 160 6.5.2 Simulation Studies ......................................................................... 162 6.5.3 Efcacy Stopping Boundaries ...................................................... 163 6.5.4 Cohort Sizes and Planned Trial Sample Size ............................. 164 6.5.5 Expected Stages and Expected Trial Sample Sizes ................... 167 6.5.6 Probabilities of Dose Selection ..................................................... 167 6.5.7 Comparison of Statistical Power .................................................. 168
6.6 Discussion and Summary ........................................................................ 170 Appendix A: Proposition of Strong Control of Type I Error ......................... 176 Acknowledgments .............................................................................................. 176 References ............................................................................................................. 176
efcient but at much lower cost [2]. The availability of genetic and genomic information has allowed teams of scientists to further pinpoint the genetic predispositions to common but complex diseases such as cancer, cardiovascular diseases, and diabetes. With the cost of sequencing a personal genome further dwindling in the near future, it is predicted that we may be able to formally incorporate our genomic, epigenetic, and even proteomic information in a real-time fashion into our personal health care. Specically, this has signaled the coming of the eld of personalized medicine [3].
