Producing a Beam Model of the Varian ProBeam Proton Therapy System using TOPAS Monte Carlo Toolkit.

Abstract

A Geant4-based TOPAS Monte Carlo toolkit was utilized to model a Varian ProBeam proton therapy system, with the aim of providing an independent computational platform for validating advanced dosimetric methods. The model was tested for accuracy of dose and linear energy transfer (LET) prediction relative to the commissioning data, which included integral depth dose (IDD) in water and spot profiles in air measured at varying depths (for energies of 70 to 240MeV in increments of 10MeV, and 242MeV), and absolute dose calibration. Emittance was defined based on depth-dependent spot profiles and Courant-Snyder's particle transport theory, which provided spot size and angular divergence along the inline and crossline plane. Energy spectra were defined as Gaussian distributions that best matched the range and maximum dose of the IDD. The validity of the model was assessed based on measurements of range, dose to peak difference, mean point to point difference, spot sizes at different depths, and spread-out Bragg peak (SOBP) IDD and was compared to the current treatment planning software (TPS). Simulated and commissioned spot sizes agreed within 2.5%. The single spot IDD range, maximum dose, and mean point to point difference of each commissioned energy agreed with the simulated profiles generally within 0.07mm, 0.4%, and 0.6%, respectively. A simulated SOBP plan agreed with the measured dose within 2% for the plateau region. The protons/MU and absolute dose agreed with the current TPS to within 1.6% and exhibited the greatest discrepancy at higher energies. The TOPAS model agreed well with the commissioning data and included inline and crossline asymmetry of the beam profiles. The discrepancy between the measured and TOPAS-simulated SOBP plan may be due to beam modeling simplifications of the current TPS and the nuclear halo effect. The model can compute LET, and motivates future studies in understanding equivalent dose prediction in treatment planning, and investigating scintillation quenching.