Room: Exhibit Hall | Forum 7
Purpose: The â€œMCsquareâ€? simulation package is an open source Monte Carlo dose engine, which has been successfully implemented in proton centers employing cyclotron accelerators. However, it has not been implemented for synchrotron based proton centers. Due to its fast speed and open source nature, MCsquare is an ideal dose engine for being implemented into robust optimization algorithms for intensity-modulated proton therapy (IMPT), which will give more accurate dose calculation in heterogeneous disease sites such as lung and head and neck cancer. Hence, MCsquare was commissioned for our proton system with a synchrotron accelerator.
Methods: The commissioning process includes three steps. Firstly, a double Gaussian beam model was generated based on our measurements of the in-air pencil beam dose lateral profiles and the integrated depth dose curves of 97 clinical energies using a synchrotron accelerator. Secondly, the beam model parameters were fine-tuned according to about two hundreds Spread-Out-Bragg-Peak (SOBP) dose measurements using ionization chambers in water. Finally, the doses to twelve patients with different disease sites was calculated by MCsquare and compared to a well-benchmarked closed source GPU-accelerated MC (gMC) dose engine.
Results: The relative differences between MCsquare calculations and SOBP measurements are all within 2.5% (within 1.5% for ~85% cases). The 3D gamma analysis (3%/3mm) results between MCsquare and gMC calculations for twelve patients are all above 99%. The typical time to simulate 20 million particles using a Dual Intel Xeon E5-2680 2.50GHz workstation is within 2 minutes, which is hundreds of times faster than the traditional Geant4 simulation.
Conclusion: MCsquare was successfully commissioned for a synchrotron-based proton system. The dose distributions calculated by MCsquare agreed well with SOBP measurements in water and gMC calculations in patient geometries. MCsquare is fast enough to be integrated into our robust optimization algorithms and to achieve MC-based robust optimization in IMPT.
Funding Support, Disclosures, and Conflict of Interest: This research was supported by Arizona Biomedical Research Commission Investigator Award, the National Cancer Institute (NCI) Career Developmental Award, the Fraternal Order of Eagles Cancer Research Fund Career Development Award, the Lawrence W. and Marilyn W. Matteson Fund for Cancer Research, Mayo ASU Seed Grant, and the Kemper Marley Foundation.