Background:The increasing number of studies dealing with linear energy transfer (LET)-based evaluation and optimization in the field of carbon ion radiotherapy (CIRT) indicates the rising demand for LET implementation in commercial treatment planning systems (TPS). Benchmarking studies could play a key role in detecting (and thus preventing) computation errors prior implementing such functionalities in a TPS. Purpose: This in silico study was conducted to benchmark the following two LET-related functionalities in a commercial TPS against Monte Carlo simulations: (1) dose averaged LET (LET d ) scoring and (2) physical dose filtration based on LET for future LET-based treatment plan evaluation and optimization studies. Methods: The LET d scoring and LET-based dose filtering (in which the deposited dose can be separated into the dose below and above the user specified LET threshold) functionalities for carbon ions in the research version RayStation (RS) 9A-IonPG TPS (RaySearch Laboratories, Sweden) were benchmarked against GATE/Geant4 simulations. Pristine Bragg peaks (BPs) and cuboid targets, positioned at different depths in a homogeneous water phantom and a setup with heterogeneity were used for this study. Results: For all setups (homogeneous and heterogeneous), the mean absolute (and relative) LET d difference was less than 1 keV/μm (3.5%) in the plateau and target and less than 2 keV/μm (8.3%) in the fragmentation tail. The maximum local differences were 4 and 6 keV/μm, respectively. The mean absolute (and relative) physical dose differences for both low-and high-LET doses were less than 1 cGy (1.5%) in the plateau, target and tail with a maximum absolute difference of 2 cGy. Conclusions: No computation error was found in the tested functionalities except for LET d in lateral direction outside the target, showing the limitation of the implemented monochrome model in the tested TPS version.
Purpose: The Local Effect Model version one (LEM I) is applied clinically acrossEurope to quantify the relative biological effectiveness (RBE) of carbon ion beams. It requires the full particle fluence spectrum differential in energy in each voxel as input parameter. Treatment planning systems (TPSs) use beamlinespecific look-up tables generated with Monte Carlo (MC) codes. In this study, the changes in RBE weighted dose were quantified using different levels of details in the simulation or different MC codes. Methods: The particle fluence differential in energy was simulated with FLUKA and Geant4 at 500 depths in water in 1-mm steps for 58 initial carbon ion energies (between 120.0 and 402.8 MeV/u). A dedicated beam model was applied, including the full description of the Nozzle using GATE-RTionV1.0 (Geant4.10.03p03). In addition, two tables generated with FLUKA were compared. The starting points of the FLUKA simulations were phase space (PhS) files from, firstly, the Geant4 nozzle simulations, and secondly, a clinical beam model where an analytic approach was used to mimic the beamline. Treatment plans (TPs) were generated with RayStation 8B (RaySearch Laboratories AB, Sweden) for cubic targets in water and 10 clinical patient cases using the clinical beam model. Subsequently, the RBE weighted dose was re-computed using the two other fluence tables (FLUKA PhS or Geant4). Results: The fluence spectra of the primary and secondary particles simulated with Geant4 and FLUKA generally agreed well for the primary particles. Differences were mainly observed for the secondary particles. Interchanging the two energy spectra (FLUKA vs. GEANT4) to calculate the RBE weighted dose distributions resulted in average deviations of less than 1% in the entrance up to the end of the target region, with a maximum local deviation at the distal edge of the target. In the fragment tail, larger discrepancies of up to 5% on average were found for deep-seated targets. The patient and water phantom cases demonstrated similar results. Conclusion: RBE weighted doses agreed well within all tested setups, confirming the clinical beam model provided by the TPS vendor. Furthermore, the results showed that the open source and generally available MC code Geant4This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.