Validation of XiO Electron Monte Carlo-based calculations by measurements in a homogeneous phantom and by EGSnrc calculations in a heterogeneous phantom

The purpose of the present study is to perform a clinical validation of a new commercial Monte Carlo (MC) based treatment planning system (TPS) for electron beams, i.e. the XiO 4.60 electron MC (XiO eMC). Firstly, MC models for electron beams (4, 8, 12 and 18MeV) have been simulated using BEAMnrc user code and validated by measurements in a homogeneous water phantom. Secondly, these BEAMnrc models have been set as the reference tool to evaluate the ability of XiO eMC to reproduce dose perturbations in the heterogeneous phantom. In the homogeneous phantom calculations, differences between MC computations (BEAMnrc, XiO eMC) and measurements are less than 2% in the homogeneous dose regions and less than 1mm shifting in the high dose gradient regions. As for the heterogeneous phantom, the accuracy of XiO eMC has been benchmarked with predicted BEAMnrc models. In the lung tissue, the overall agreement between the two schemes lies under 2.5% for the most tested dose distributions at 8, 12 and 18MeV and is better than the 4MeV one. In the non-lung tissue, a good agreement has been found between BEAMnrc simulation and XiO eMC computation for 8, 12 and 18MeV. Results are worse in the case of 4MeV calculations (discrepancies≈4%). XiO eMC can predict dose perturbation induced by high-density heterogeneities for 8, 12 and 18MeV. However, significant deviations found in the case of 4MeV demonstrate that caution is necessary in using XiO eMC at lower electron energies. © 2012 Associazione Italiana di Fisica Medica.