Conventional pharmacokinetic trials in oncology measure the total amount of drug and its metabolites in the plasma of patients. These total drug concentrations are then correlated with clinical parameters such as toxicity and/or drug efficacy. However, total drug concentrations in plasma are often not directly related to tumor or tissue concentrations and the kinetics in (tumor) tissue may be very different from the kinetics of the drug in plasma. As only the unbound fraction of drugs can passively diffuse or can be actively transported across the cell membrane of endothelial cells and penetrate the tissues and tumor cells, binding of drugs to plasma components may, in part, be responsible for the variability in clinical outcomes. Also, variation in binding to cellular components may be important. As an overall consequence, in in vivo models high inter-individual variabilities of drug concentrations in the tumor have been observed, which affected drug activity. Many current cancer drug-therapies aim for a clinical benefit at the edge of acceptable toxicity, without paying attention to (variability in) plasma and tumor or tissue pharmacokinetics. This review discusses the potential applicability of microdialysis, a minimally invasive method used to determine the unbound fraction of drugs in blood or in the interstitial fluid of tissues, in clinical oncology. Although faced with some methodological challenges, the technique offers many advantages over others, and enables direct tumor assessment for pharmacokinetics and even pharmacodynamics.