A radiobiological database produced by the BIANCA model to predict the biological effectiveness of hadrontherapy beams

The BIANCA biophysical model was used to simulate cell survival curves by protons, He-and C-ions over a wide LET range and for several doses. Each simulated curve was then fitted by a linear-quadratic exponential function of the form S(D)=exp(-αD-βD 2). This allowed to produce an almost continuous set of α and β values as a function of LET for each ion type. The same procedure was repeated for chromosome aberration dose-response curves, using the following fitting function: A(D)=αAD+βAD 2. In the context of hadrontherapy, the tables of α and β (as well as αA and βA) were read by the FLUKA radiation transport code, which provides the necessary information about particle type, LET and absorbed dose, thus allowing fast computing of biological outputs in every position of a therapeutic dose profile. Some examples of the variation of the two considered biological quantities, i.e. probability of cell survival and chromosome aberrations, along Spread Out Bragg Peaks in water are reported as preliminary results. 1 Introduction The main rationale for the use of charged particles in cancer therapy relies in their physical properties: when traversing a material, they deposit most of their energy at the end of their path, in the so-called Bragg peak. By using several beams with different initial energies, it is possible to shape a " Spread Out Bragg Peak " (SOBP), thus delivering most of the physical dose in the tumour region, while minimizing the dose in the surrounding healthy tissues [1]. Nevertheless, also the ion biological effectiveness plays an important role in this context: the higher Relative Biological Effectiveness (RBE, defined as the ratio between a photon dose and the ion dose necessary to induce the same effect) is the main advantage in the use of ions heavier than protons (e.g. carbon ions) [2]. However, it is well- known that, in general, RBE is not constant along a flat SOBP, but it tends to increase with increasing depth. Therefore, it is fundamental to use a tool (e.g. a biophysical model) able to quantify the RBE, in principle in each position and for each physical and biological configuration. Currently, in clinical practice the variability of RBE is ignored in the case of protons, for which a constant RBE of 1.1 is usually assumed. Nevertheless, it would be important to quantify the consequences of the well-known increase of RBE in the last few millimetres of proton SOBPs, as well as the " shift " of the biological peak beyond the physical dose fall-off [3]; this may be critical, especially for mono- directional irradiations, when an organ at risk is present beyond the tumour. On the other hand, the variability of RBE is usually taken into account for carbon ions; two biophysical models are currently coupled with the Treatment Planning Systems (TPS) used in hadrontherapy centres worldwide. In particular, the Local Effect Model (LEM) is used in Europe (Germany, Italy and Austria) [4], whereas the Microdosimetric Kinetic Model (MKM) is applied in Japan [5]. In clinical practice, the physical dose distribution for the treatments is provided by the TPSs, which are typically fast-performing analytical algorithms; these tools needs to adopt some