A Time Dependent Model of a Passive Scattering Proton Therapy Nozzle Using TOPAS
M H Chequers1*, D Granville1, K Suzuki2, G O Sawakuchi1, (1) Carleton University, Ottawa, ON, (2) The University of Texas MD Anderson Cancer Center, Houston, TXSU-E-T-71 Sunday 3:00:00 PM - 6:00:00 PM Room: Exhibit Hall
Purpose: To model and validate a commercial passive scattering proton therapy nozzle using the time dependent features of the TOPAS Monte Carlo toolkit.
Methods: The geometry and material composition of a commercial passive scattering nozzle were modeled in TOPAS based on specifications found in the literature (Smith, et al. 2009, Med. Phys. 36:4068; Arjomandy et al. 2009, Med. Phys. 36:2269). Energy specific beam shaping devices including range modulator wheels (RMWs) and second scatterers were implemented in TOPAS. Simulations were performed using the time features of TOPAS, which allow geometry update during a single Monte Carlo simulation. The time features allowed us to simulate the rotation of the RMWs to obtain a full spread-out Bragg peak (SOBP) in a single run. Initial parameters of the beam such as initial energy spread, beam spot size and angular spread have large uncertainties and therefore they were considered free parameters of our model. By varying these parameters, simulated dose distributions were matched as closely as possible to measured ones. We compared simulation and measurement results of normalized dose as a function of depth for 4 and 8 cm SOBPs and 140 and 250 MeV beam energies.
Results: In the buildup and SOBP regions, simulations and measurements agreed within 3.6%. Ranges agreed within 1.3 and 0.5 mm for 140 and 250 MeV, respectively.
Conclusions: TOPAS has the potential to accurately model passively scattered therapeutic proton beams. Adjustments of the source parameters and physics list are needed to further improve the agreement between our simulations and measurements. Once validated, our TOPAS model may be used to more accurately perform retrospective studies of the impact of physical parameters, including linear energy transfer, on patient outcome. The model may also be used to investigate the response of detectors to therapeutic proton beams.