Application of the Multi-Model Monte-Carlo Treatment Planning System Combined with PHITS to Proton Radiotherapy

In proton radiotherapy, a proton generates secondary neutrons and photons in a patient’s body. This leads to irradiation of the area in proximity to the target region with incidental doses caused by secondary radiation in addition to the therapeutic dose. However in current treatment planning of proton radiotherapy, conventional dose estimation algorithms like the pencil beam method can not estimate the influence of secondary radiation. Thus, to improve accuracy of radiotherapy's dose estimation, a new multi-modal Monte-Carlo treatment planning system is being developed. The new system employs the Monte-Carlo transport code PHITS as the dose calculation engine of the system. The final goal of this study is to investigate the application of the Monte-Carlo treatment planning system at the proton radiotherapy facility of the University of Tsukuba. To achieve this goal, first, the influence of the secondary radiations and the incidental doses generated by the proton irradiation was estimated by the Monte-Carlo simulation utilizing PHITS. Secondly, we aim at performing Monte-Carlo treatment planning for proton radiotherapy using JCDS-FX combined with PHITS. The estimation results for the secondary radiation caused by proton irradiation demonstrated that the maximum value of absorbed dose due to recoil protons was approximately 0.64 mGy/Gy at a depth of 6.5 cm on the central axis of the phantom. The absorbed dose accounted for about 72% of the total incidental equivalent dose. A lot of external neutrons generated at upstream devices irradiate the surface of the phantom. However, at deeper located regions, influences of the external neutrons were small as neutrons generated inside the body accounted for a large amount of the total neutron count in this region. The results demonstrated that it is impossible to eliminate the incidental dose completely, even if shielding of the external neutrons and application of the active scanning method are achieved.