ESTRO 2020 Abstract book

S777 ESTRO 2020

Gamma index pass rate (local DD=3%, DTA=1.25 mm, threshold 10%) of expected sinogram was 96.9% and 98.2% for a chest (50Gy/25fr) and H&N (66Gy/33fr) plans respectively. Conclusion The results obtained with this preliminar study were promising. Anyway, it needs further checks against measurements obtained with independent instruments. Furthermore, efforts should be addressed to investigate the sensibility against induced MLC errors and validate data for a significative amount of cases. References: PO‐1372 Feasibility of treatment planning on contrast enhanced CTs for HN cancer M. Gober 1 , A. Riegler 1 , N. Zagler 1 , D. Cires 1 , F. Beck 1 , M. Kirschner 1 , E. Steiner 1 , M. Metz 1 1 Landesklinikum Wiener Neustadt, Institute of Radiatiotherapy and Radiooncology, Wiener Neustadt, Austria Purpose or Objective For head and neck (H&N) treatment planning, contrast- enhanced CTs (CECT) are helpful for telling the difference between lymph nodes/vessels and blood vessels. For this purpose, a native CT scan and a subsequent contrast agent (CA) enriched CT scan are registered rigidly. Uncertainties in the registration, additional imaging dose for the patient and additional steps in the workflow for radio-oncologists are disadvantages of this method. This study is looking into the feasibility of treatment planning on the CECT itself, as intravenous CA not only increases the contrast but also the Hounsfield units (HU) of the tissue with CA uptake. The aim of this work is to evaluate the influence of CA on dose calculations in H&N treatment planning with focus on VMAT. Material and Methods A treatment-planning study was performed which involved the generation of treatment plans on unenhanced CTs and a subsequent transfer of those plans onto the registered CECT. The study included 43 VMAT plans of 19 H&N patients who were treated from August 2018 to January 2019. The dose calculation was performed in Varian Eclipse using the algorithms AAA 11031 and AcurosXB 11031 with dose to medium. The derived dose distributions were compared and analysed by evaluating the near-minimum D98(%), the near-maximum D2(%) and the median D50(%) absorbed dose values for the PTV. Furthermore, the dose volume histograms (DVHs) were compared visually The analysis of the dose distributions of the native CT and the dose distribution of the CECT indicated discrepancies in patient positions (deglutition, respiration, etc.) between the two CT data sets. In order to avoid the uncertainty introduced by the setup error, the tissue with CA uptake was contoured and altered by assigning new CT values of 40-50 HU. By using the CA corrected CTs, an evaluation of the effect of CA on dose calculation without the uncertainty arising from setup errors could be performed. Results The differences in dose distributions between the native CT and the CECT were 0.13% ± 0.19% for D50(%), -0.36% ± 1.19% for D98(%) and -0.04% ± 0.19% for AAA. For AcurosXB dose differences of 0.15% ± 0.40% for D50(%), -0.20% ± 1.16% for D98(%) and 0.06% ± 0.20% were found. The comparison of the dose distributions between the CECT and the CA corrected CT showed dose differences of -0.03% ± 0.08% for D50(%), -0.03% ± 0.08% for D98(%) and - 0.04% ± 0.07% for AAA and -0.06% ± 0.09% for D50(%), - 0.05% ± 0.12% for D98(%) and -0.03% ± 0.08% for AcurosXB. [1] AAPM TG-218 (2018) [2] AAPM TG-219 (2018) [3] AAPM TG-120 (2011)

Conclusion The setup error between the native CT and the CECT showed to have a larger influence on the dose calculation than the CA itself. The dose difference in treatment planning caused by the presence of the CA is small and no clinical effect is expected. The use of the CECT for VMAT treatment planning is admissible for H&N cancer. PO‐1373 Evaluation of X‐ray beam focal spot position for Varian TrueBeam linear accelerators O. Juda 1 , W. Skrzyński 1 , A. Walewska 1 1 Maria Sklodowska-Curie Memorial Cancer Center and Instytute of Oncology in Warsaw, Medical Physics, Warsaw, Poland Purpose or Objective The position of X-ray beam focal spot can affect the position of the radiation isocentre, symmetry of the beam, and the width of penumbra, which is especially important for small fields used in stereotactic radiotherapy. The purpose of the work was to determine position of the X- ray beam focal spot relative to the rotation axis of the collimator for two TrueBeam Varian accelerators Material and Methods Based on the Chojnowski’s paper (J Appl Clin Med Phys. 2017;18(5):175-183), a method for determination of the location of the focal spot X-ray beam relative to the collimator rotation axis was developed and implemented. Portal images generated with high-energy X-ray beams with a flattening filter were registered regularly over a period of four months for two Varian TrueBeam linear accelerators. Each time a set of images was obtained, with separate fields shaped by the jaws of the main collimator and by the multi-leaf collimator (MLC). The images were analyzed using own plugin for ImageJ software to find positions of beam centers. If the beam focal spot was not on the collimator's rotation axis, the location of the center of the field shaped by the jaws of the main collimator differed from the location of the center of the field shaped by the MLC. Based on this difference, the distance between the focal spot and the axis of collimator rotation was calculated. Results For both accelerators, the distance between the beam focal spot and the collimator rotation axis measured in the cross-plane direction did not exceed 0.2 mm, and in more than 90% of cases, it was within 0.1 mm. The distances measured in the in-plane direction were larger and for one accelerator reached even 0.76 mm, whereas for the other one the range of results was smaller and did not exceed 0.4 mm. Conclusion Differences between two linear accelerators of the same type were observed. Accelerator with a smaller variation of distance between the beam focal spot and the collimator rotation axis should be probably chosen as a primary unit for stereotactic radiotherapy