Intensification of cytotoxic chemotherapy enhances the outcome of several malignancies but is limited by haematotoxicity. While neutropenia and anaemia can be treated with supportive growth factor applications, thrombocytopenia remains a dose-limiting side effect due to the lack of clinically approved pharmaceutical growth factors. Hence, it is necessary to assess the degree of thrombocytopenia of newly designed intensified regimens in the planning phase of a clinical trial. We present a simple ordinary differential equations model of thrombopoiesis under chemotherapy which maps the dynamics of stem cells, CFU-Mk, megakaryocytes and platelets in spleen and circulation. Major regulatory cytokine of thrombopoiesis is thrombopoietin (TPO) whose production and consumption is explicitly modelled. TPO acts by increasing the number of mitoses of CFU-Mk and increasing the mass and maturation of megakaryocytes. Chemotherapy is modelled by a drug-dose and cell-stage specific acute cell loss. Most of the cell kinetic parameters of the model were taken from literature. Parameters regarding TPO regulation and chemotherapy toxicity were estimated by fitting the predictions of the model to time series data of platelets received from large clinical data sets of patients under seven different chemotherapies. We obtained a good agreement between model and data for all scenarios. Parameter estimates were biologically plausible throughout. For validation, the model also explains data of TPO and platelet dynamics after thrombopheresis taken from literature. We used the model to make clinically relevant predictions. Regarding thrombocytopenia we estimated that the CHOP regimen for the treatment of high-grade non-Hodgkin's lymphoma can be time-intensified to a cycle duration of 12 days while the time-intensified CHOEP regimen would result in severe cumulative toxicity. We conclude that our proposed model proved validity for both, different chemotherapeutic regimens and thrombopheresis as well. It is useful to assess the thrombocytopenic risk in the planning phase of a clinical trial.
Corresponding Author Institution
Institut für Medizinische Informatik, Statistik und Epidemiologie