MEtastases MOdeling and Reseach in EXperimental PharmacoKinetics – MEMOREX-PK

Efficient cancer therapy requires a throughout and upgraded understanding of cancer biochemistry, pharmacology, kinetics and medicine. By using mathematical models, numerous authors attempted since several years to manage all this available information in order to optimize cancer therapy (Swan, 1990). Particularly in metastatic cancer, the decision for the forthcoming treatment depends on the disease progression, but evidence for the existence of occult micro-metastases at the time of diagnosis is overwhelming. In these cases, chemotherapy is used as a systemic therapy because it is the only treatment which may have an impact on the problem of disseminated tumors (Early, 1992). To improve efficiency, the traditional chemotherapy involving cytotoxic agents is today enhanced to associate cytostatic, anti-angionesis, or target therapy agents. In all these schemes, both drug amounts (intensification) and schedule (densification) of drug administrations should be determined to ensure a desired rate of tumor cell kill without unacceptable toxicity. Dose-dense schedules may have an advantage over conventional schedules of drug administration (Norton, 2001). Several years ago, we developed a mathematical model able to calculate densified administration protocols (Iliadis, 2000). Recently, we also proposed an adaptive approach for controlling dose intensification for the phase I trials in oncology (Meille, 2008). The methodology combining dose densification and intensification was applied to the metastatic breast cancer in phase I clinical trial (You, 2007). However, it is important to point out that the approach does not yet take into account the disease progression. Research conducted for more than 30 years at the department of Pharmacokinetics lead to precise recommendations for a valuable use of drugs in humans. Mathematical model is the basic tool for describing the gathered observations, predicting the drug fate in the organism following various administration schemes and individualizing the drug treatment (Iliadis 1992). In these developments, the disease progression is entirely ignored. The current proposal supplies the lacking component to establish the analysis of complete 'disease-therapy' system in a more realistic context. A first dynamical model described by partial differential equation (a conservation law supplied with appropriated boundary conditions) has already been designed (Iwata, et al., 2000) (Barbolosi, 2008) (Devys, et al., 2008) and has now to be validated and enriched to take into account the therapies and their administration protocols. Before validation in the humans, an animal validation is required. This is the purpose of the interdisciplinary collaboration between applied mathematicians (Applied Mathematics team of LATP University of Provence) and life and health sciences researchers (Oncopharmacology team of the department of Pharmacokinetics, faculty of Pharmacy Marseille), that we propose. The contribution of the mathematicians consists in bringing out the parameters of the models that are fundamental in the prediction of the disease progression. These parameters must be uniquely defined from a set of realistic experimental data. The underlying problem of identifiability of parameters is a crucial question in the mathematical-biology field. Approximation of these structural parameters assorted with robust numerical algorithms to simulate these predictive models constitutes the in silico models. The expertise of life and health researchers will help through the animal validation to adapt and enrich these original in silico models of prediction of micro-metastatic process underlying the growth of a primary tumor, and to develop a chemotherapeutic strategy and to improve, ultimately, the treatment of cancer patients.