Pharmacokinetic variability in children with cancer is substantial and confounds drawing conclusions regarding optimal therapy based only on dose-response relationships. Careful pharmacokinetic studies performed during drug development in conjunction with an assessment of patient characteristics, such as age, renal and hepatic function, and concomitant therapy, is essential for defining those factors that may alter drug disposition. By integrating pharmacokinetic studies with measures of efficacy and toxicity, a pharmacodynamic framework can be established for guiding therapy to minimize differences in systemic exposure among subpopulations of patients (eg, impaired renal function and neonates). In selected instances when pharmacokinetic variability cannot be predicted by patient covariates, the potential for individualizing dosages based on patient-specific pharmacokinetic parameters is now a clinically feasible option. The need for and benefits of incorporating such strategies into routine therapy represents an exciting area for further clinical research.