ESTRO 2020 Abstract book

S809 ESTRO 2020

1.5%) for lung, breast, and esophageal cancer. For PRT it was significantly reduced to 1.6% (range 0.7-2.1%), 4.5% (range 0.0-15.5), and 0.8% (range 0.0-1.6%), respectively (p≤0.01). The fatal risk for secondary malignancies also significantly decreased to 1.1% (range 0.5-1.5%), 0.9% (range 0.0-3.0%), and 0.7% (range 0.0-1.5%) for lung, breast, and esophageal tumors in PRT, compared to 3.1% (range 1.3-4.1%) , 1.9% (range 0.4-5.4%), and 1.0% (range 0.5-1.4%) with IMRT (p≤0.01). Using the Schneider model, a significant risk reduction of 54.4% (range 32.2-84.0%), 56.4% (range 16.0-99.4%), and 24.4% (range 0.0-99.0%) was seen for secondary lung, breast, and esophageal tumors when treating with PRT instead of photon IMRT (p≤0.01). Conclusion Based on two prediction models, PRT for malignant mediastinal lymphoma significantly reduced the risk for radiation-induced secondary malignancies compared to photon irradiation. Young patients with malignant lymphoma of the mediastinum, at high risk for secondary malignancies after treatment, should be treated with PRT. PO-1430 A Monte Carlo study of infinitely long phantoms used for CT dosimetry A. Abuhaimed 1 , C.J. Martin 2 1 King Abdulaziz City for Science and Technology, The National Center for Applied Physics, Riyadh, Saudi Arabia ; 2 University of Glasgow, Department of Clinical Physics and Bioengineering, Glasgow, United Kingdom Purpose or Objective Assessment of doses resulting from CT scans over infinitely long phantoms is required for some purposes. For example, they are used to estimate CT dose index (CTDI ∞ ), from which the efficiency of CTDI 100 can be assessed. They can also be employed to measure the cumulative dose from a given CT scan as described by the American Association of Physicists in Medicine (AAPM) in the task group 111 report. The dose profile integral to assess total dose absorbed in a patient undergoing a given CT scan can also be estimated with the long phantoms. However, in practice, use of an infinitely long phantom is difficult. The aim of this study was to investigate a practical phantom length that includes all primary and scattered photons. This length would allow the detection of almost all photons that contribute to the scan dose, such that the difference from results with the infinite one would be negligible. Material and Methods The assumption has been made that the longest beam width for a cone beam that can be utilized in the clinic is 300 mm. Monte Carlo (MC) simulations were conducted to track trajectories of photons resulting from this beam. A previously validated MC model of a kV on-board imaging system integrated with a Truebeam linear accelerator was utilized to run head and body scans using two tube potentials of 80 and 140 kV that would cover the range used for CT scans. Photons trajectories were tracked in head and body phantoms, 16 and 32 cm in diameters, respectively. In order to approach the infinite length, a length of 900 mm that equates to three times the beam width, was used for the phantoms. Results Photon trajectories resulting from head and body scans are shown in figure 1. For all scans, almost all photons were within the length used. Thus, use of 900 mm indicates a possible option to approach the scan dose that would be assessed in an infinitely long phantom. In practice, the 900 mm length may be formed by attaching several phantoms made of PMMA similar to that of CTDI phantoms. This length may also be considered when MC simulations are performed to improve efficiency of the simulations and minimize the time required for long phantoms.

Conclusion Use of an infinity long phantom for some purposes is difficult in practice. This study shows that head and body phantoms with a length of 900 mm would include almost all photons resulting from CT scans with the longest possible beam width. It can be concluded that this length can be considered to mimic the infinite length and provide an equilibrium condition, beyond which contribution of photons into the scan dose becomes negligible. The suggested length is for wide beams, but a shorter length, for example 450 mm, may also be used for scans acquired with narrow beams. PO-1431 Patient-specific dose calculations from a proton gantry mounted CBCT system: implementation in TOPAS T. Henry 1 , A. Dasu 1 1 Skandionkliniken, Skandionkliniken, Uppsala, Sweden Purpose or Objective Studies on anthropomorphic phantoms have shown that proton CBCT units can deliver up to 2 Gy in scenarios where the scans are performed before each treatment fraction (Ardenfors et al. 2018). In light of these findings and the increased number of patients receiving proton therapy, the determination of patient-specific doses would be of great interest, especially for paediatric patients. The aim of this project was to evaluate the possibility to easily and automatically perform calculations of the total doses delivered by a proton CBCT unit to specific organs for each patient undergoing CBCT examinations during the treatment course. Material and Methods The gantry-mounted CBCT system from the Skandion Clinic in Uppsala, Sweden, was re-modelled in the TOPAS Monte Carlo code. Simulated depth-dose curves and lateral profiles obtained from three clinical protocols (head, pelvis and thorax), with different tube spectra outputs, were compared to measurements. Thereafter, the capacity of TOPAS to efficiently load and correctly position patient CT data, as well as to record doses from CBCT examinations to different organs, was evaluated. For this, DICOM images and RT structures set were exported from a TPS and imported in TOPAS using the embedded tools. To ensure that the beam isocenter in the treatment plan was correctly aligned to the world’s origin in the Monte Carlo code, DicomOrigin, UserOrigin and Isocenter positions were also exported from the TPS. Organ specific doses were calculated using the geometries