[Biostatistical issues]

In this contribution we discuss some fundamental aspects of biostatistics in the context of research in clinical oncology. Starting with a general presentation of basic principles of confirmatory statistics we deal with specific questions which typically arise in practical situations. This covers the importance of prespecification of statistical analyses before looking at the data as well as the judgment of multiple p-values and the interpretation of subgroup analyses. Subsequently we propose essential principles and techniques for the design and analysis of controlled clinical trials, which are relevant to achieve valid results with regard to cause-effect relationships from clinical observations. We specifically point out that an exact definition of the whole research design together with the methods of analysis and evaluation as well as planning of interim analyses and possible termination strategies have to be defined clearly in advance. General hints with regard to reporting standards conclude this chapter. In the final chapter we discuss general ideas behind meta-studies and demonstrate by means of a small example some basic analytical techniques for the summarization of single study results.