ABSTRACT
Recent advances in molecular biology, genomics, and targeted agent development have fueled the rapid progress in clinical oncology. In parallel, developments in statistical theory and computation have continued to provide better methods and tools for dealing with complex problems. All these efforts lead to more advanced, yet complicated study design, conduct, and analysis, which depend more and more heavily on computing resources from both the software and hardware points of view Because a wide range of complex calculation methods are involved, it is difficult and complicated for statisticians to develop their own codes from scratch every time when a new design is implemented or an analysis is performed. Moreover, the emerging Bayesian methods and adaptive trial designs (Berry 2006, Biswas et al. 2009, Berry et al. 2010, Lee et al. 2010), which introduced many new concepts and calculation methods, pose new challenges for software development. Developing and debugging codes could take a huge amount of time, and, without thoroughly being tested, the best effort from any individual is also subject to errors. Fortunately, many useful and valuable computer software and resources are now available from both research and commercial entities. Instead of developing their own codes from scratch every time, statisticians and clinical trial researchers will benefit much from using available design and analysis software that has been developed and tested. In this chapter, we will give a broad overview on selected software resources relevant to cancer clinical trials. It is impossible to do a comprehensive review in this knowledge explosion era. The choice of the software is limited by the authors’ knowledge and experience. Undoubtedly, many valuable tools could be omitted and not covered. However, we hope the information provided in this chapter can be used as a starting point for the quest of identifying and developing more and better software for cancer clinical trials.
