Clinical trials in cancer to guide physicians and that meet regulatory approvals have undergone significant transformations over the last 5 years. Specifically, choice of dose and regimen used to be deduced from extensive in vitro and then in vivo work using a number of animal models and dose-response work. Extensive pharmacology trials would then fine tune dosing for clinical trials and knowledge about likely place in therapy, drug interactions and effects of food on drug effect. Previous recruitment for cancer clinical trials would have come from affected organ and histological subtypes. Now drug development for clinical trials has been significantly abbreviated both in clinical pharmacology and in preparation for the clinical trials. Further, recruitment is now based on genetic subtypes and cell biology. How this has affected the design and then the interpretation of clinical trials, the translation into clinical care, and how has this resulted changes in toxicity and mortality outcomes of new drugs is discussed. Specifically how current clinical trials can predict the the overall risk and benefit of this rapid drug development program is reviewed.