Pharmacokinetic optimisation of cancer chemotherapy. Effect on outcomes.

Cancer chemotherapy doses are empirical in that the majority are administered at a fixed dose (mg/m2 or mg/kg). One reason for this is the intrinsic sensitivity of the tumour or host cells to one particular chemotherapy agent is unknown. Therefore, the likelihood of response or toxicity is unpredictable a priori. This contrasts with antimicrobial chemotherapy where sensitivity (minimum inhibitory concentration) can be determined for a specific bacterium. The pharmacokinetics of cancer chemotherapy agents is also highly variable between patients. In addition, the small therapeutic index of these drugs, combined with the lack of good surrogate markers of toxicity or response, adds to the empiricism of the administration of cancer chemotherapy. In the past few years, numerous studies have established good relationships between systemic exposure to cancer chemotherapy and both response and toxicity. These relationships have been used to individualise chemotherapy dose administration a priori and a posteriori. Some examples of drugs which are individualised based on their pharmacokinetics are methotrexate, busulfan and carboplatin. Other examples of antineoplastic agents which may eventually be individualised based on their pharmacokinetics are mercaptopurine, fluorouracil, etoposide and teniposide, topotecan and suramin. New strategies are being investigated to improve the therapeutic index of cancer chemotherapy agents such as biomodulation, pharmacogenetics, circadian administration and the modification of drug scheduling. Pharmacokinetic studies have also played a major role in these areas. Thus, despite the empiricism associate with cancer chemotherapy administration, some progress has been made and shown to have an impact on outcome. However, more studies are needed to improve cancer chemotherapy administration.