Biological Models

Radiotherapy of loco-regional tumors aims to irradiate the tumor with sufficient dose to achieve local control while minimizing the dose to adjacent organs at risk to avoid complication in normal tissues. As tumor control probability (TCP) as well as normal tissue complication probability (NTCP) increase with increasing dose, there is a dose range (“therapeutic window”) where the probability for tumor control without complications receives its maximum (fig 1.) Biological models in radiotherapy aim to predict these TCP- and NTCP-values already at the stage of treatment planning to optimize the treatment for the individual patient.

To describe the growth and radiation response of tumors and to better understand the underlying biological mechanisms, a single-cell based simulation model was developed. In this model, the individual cells of the tumor are arranged on a cubic grid and the action of each cell (e.g. proliferation, cell death etc.) at any point in time is individually determined using Monte Carlo methods.
As the clinical radiation response strongly depends on tumor oxygenation, the model additionally includes capillary cells, which may also proliferate (i.e. show angiogenesis). The density and distribution of the capillary cells determines the distribution of the oxygen concentration within the tumor. Furthermore, hypoxic cells in the tumor model are assumed to excrete tumor angiogesis factors (TAF), which stimulates angiogenesis in in hypoxic regions of the tumor. The dynamical equilibrium between tumor cell proliferation and angiogenesis finally determines whether the tumor will contain hypoxic or well-oxygenated regions only.
This simulation model can be used to systematically investigate the influence of specific tumor- or treatment parameters (e.g. cell sensitivities, proliferation rates, fractionation schemes, total doses etc.) on tumor control probability. Furthermore, the model can be used to predict the development of tumor oxygenation under radiotherapy, which could in principle be used to adapt and to optimize the individual irradiation protocol.
In the ext step, the model needs to be validated and some biological input-parameters (e.g. the amount clonogenic cells in the tumor, proliferation rates etc.) remain to be determined. As a proof-of-principle, the model parameters shall be adapted for an experimental tumor model and the actual radiation response shall be compared with the model predictions.
To apply the simulation model to clinical tumors in patients, the concept of simulating of tumor growth and radiation response has to be replaced by simulation the radiation response only. The initial status of the tumor prior to radiotherapy (e.g. size, shape, activity, oxygenation etc.) can then be characterized using multimodal biological imaging methods. The development of the tumor under irradiation may then be predicted by translation the macroscopic imaging information into the required microscopic information and by applying the simulation model.