Optimising the design of phase II oncology trials: The importance of randomisation

摘要: Oncology trial end-points continue to receive considerable attention, as illustrated by the development and revisions RECIST criteria. In this article, we focus reader away from issue of for phase II trials towards what believe be an even more important issue, fundamental need randomisation in oncology trials, ideally with blinding dose-ranging. We present arguments support proposition that will enable greater clarity interpretation results, well allowing precise estimates effect size sample requirements definitive III trials. Randomisation also reduce potential bias resulting inter-trial variability, which inflates both type I errors if historical controls are utilised. context a randomised blinded trial, exact choice end-point is less critical, although favour such change tumour or progression status at fixed early time point (i.e. 8-12 weeks after randomisation). Although based on criteria can should utilised do not revision result improvement drug decisions absence clinical stage development.