Related Experiment Video
Updated: Mar 2, 2026
Proton Therapy Delivery and Its Clinical Application in Select Solid Tumor Malignancies
Published on: February 6, 2019
SU-E-T-500: Pencil-Beam versus Monte Carlo Based Dose Calculation for Proton Therapy Patients with Complex
J Schuemann1,2,3, J Shin1,2,3, J Perl1,2,3
1Massachusetts General Hospital, Boston, MA.
Purpose:
To investigate the necessity of the verification of dose distributions using Monte Carlo (MC) simulations for proton therapy of head and neck patients and other complex patient geometries.
Methods:
TOPAS, a TOol for PArticle Simulations that makes MC simulations easy-to-use for research and clinical use and is layered on top of Geant4, has been used to simulate the treatments of head and neck patients at the Massachusetts General Hospital (MGH). The resulting dose distributions have been compared to pencil beam calculations based on the XiO treatment planning system. Dose difference distributions were used to highlight areas where the two algorithms did not agree. Dose volume histograms are utilized to investigate the overall agreement of the planned doses in target structures.
Results:
21 head and neck patients, both nasopharynx and spinal cord, were investigated. The field complexity ranges from a single field up to 13 fields. For all patients, the dose in the clinical target volume agrees well. Nevertheless, differences in critical structures around the targets have been observed mostly due to range differences between the two algorithms.
Conclusions:
Pencil beam algorithms provide an accurate description of dose in the target volume. However, we conclude that the differences between MC simulations and pencil beam algorithms in regions outside the target for complex geometries, such as present in head and neck patients, support the necessity of routine use of MC simulations for treatment verifications before treatment. TOPAS is aiming to make such routine simulations available to all researchers and clinics. An automated interface utilizing TOPAS to enable such simulations has been developed at MGH and should become routinely used in the near future for patients with complex geometries. NIH/NCI R01 CA140735.
