Concepts elucidated from preclinical pharmacology studies have made a substantial impact on the clinical use of anticancer drugs. However, the majority of animal pharmacology results have not been available until after drugs have entered clinical trials. Since clinical pharmacokinetic measurements are already part of many phase I trials, human data could be directly compared with mouse data if mouse pharmacology studies were completed before clinical trials were initiated. Once the starting dose in a phase I clinical trial has been evaluated, subsequent doses are escalated until the maximum tolerated dose is reached. The rate of escalation is empirically defined by a modified Fibonacci series. This universal escalation scheme is applied to all drugs, with no modifications based upon pharmacology or other factors. If the starting dose is far removed from the maximum tolerated dose, a large number of dose escalations are required. Consequently, most patients receive subtherapeutic doses, and the amount of resources allocated to each drug increases. We are exploring potential strategies for controlling the rate of dose escalation based upon pharmacokinetic determinations in mouse and man. Retrospective analyses indicate that 20%-50% savings in the total number of dose escalations are possible.