Models in cytostatic chemotherapy.

Models play a fundamental role in all sciences. Scientific theories can be understood as "intellectual models" of the relations and interrelations between the elements of the science. Such relational models consider the scientific topics as "systems," and may therefore be called "system models." The basic elements of chemotherapy are the processes of kinetics (absorption, distribution, metabolizing, and elimination of drugs in organisms), interaction of drug molecules or their metabolites with specific receptors, and the effect-formation. The basic system models for these processes with special reference to cancer chemotherapy are discussed. For the evaluation of these system models appropriate experimental models are necessary. For cancer chemotherapy the experiment with living tumor-bearing animals is the basis as well for evaluation of scientific theories (models) as well as for inference in clinical situations. Such inferences are "inductive" inferences, i.e., inferences made under uncertain conditions. To assess probabilities for the degree of uncertainty the theorem of Bayes is a valuable tool. Experimental models, as models for clinical situations, are used for screening, pharmacologic characterization, and preclinical pharmacology of drugs. In each of these phases of preclinical drug development the experimental models must be appropriate for the relevant clinical situation. So, for the screening of anticancer drugs appropriate experimental tumor systems must be used to reveal the clinical relevant toxic and therapeutic potency of the drug. In our laboratory, test systems with relatively chemosensible experimental tumors on rats and mice have proved as efficient for primary screening. In this system a total cure could be achieved with cyclophosphamide as standard. For the pharmacologic characterization, test systems must be used which are able to reveal the clinically relevant safety margins, time-effect relations, and organotoxic effects of the drug. For the preclinical pharmacology, the experimental models should give some information about the relevant pharmacokinetics, pharmacodynamics, and possible (toxic or therapeutic) interactions with other drugs. The most important aspects of these models for cancer chemotherapy are discussed. Special attention is given to the clinical aspects.