ESTRO 38 Abstract book

S38 ESTRO 38

ArcCheck (Sun Nuclear, Melbourne, FL) plus an ionization chamber (IC) and correlated with the independent dose calculation passing rates. An analysis of average changes in monitor units (MU) from one adapted plan to another was also performed. Results The average difference between the measured and TPS- predicted values for the end-to-end tests was 1.64% (maximum of 2.8%). The average gamma (2%/2mm) passing rate for the independently calculated dose vs the TPS-predicted dose was 98.5%. An example of the in- house software results for an adapted plan is shown in Figure 1. Table 1 shows the average IC, ArcCheck (3%/3mm) and independent QA (2%/2mm) rates for each patient. The average change in monitor units from fraction to fraction was 7.5%, with maximum observed of 20%.

program. To suppress stray radiation, the detector’s crystal is fully surrounded by a cylindrical steel cylinder. Collimators of 10 mm, 5 mm, 3 mm and 0.9 mm diameter are available. The detector’s efficiency was calibrated by means of radioactivity standards. The intensity of the discrete X-ray lines is analysed by a software package that follows the methodology of the “GUPIX” X-ray spectra-analysing software package. The source holder is mounted to a computer-interfaced rotary stage, thus providing angular control while maintaining the proper orientation of the collimator axis to the source. In this way, spectra for a specific angle, as well as integral spectra, can be determined. Results In the figure below the step of automated peak detection is exemplify with a spectrum of an I-125 Bebig S17 seed. The upper panel shows the whole spectrum. The lower panel is an enlargement of the region marked red. The black frames represent the values for the detection limits calculated as the sum of three times the peak’s uncertainty plus the uncertainty of the fitted background at the peak’s position. As the peaks are of the Au-L-Lines at 9.7, 11.4 and 13.4 keV are below the frame they are not detected by the algorithm. This result is realistic, as the seed doesn’t feature any golden components. By contrast, the Ti-K-lines at 4.5 and 4.9 keV, arising from the capsule, are clearly detected, as well as the Mo-K- lines at 17.4 and 19.7 keV, arising from the marker. Conclusion An automated spectrometry set-up for LDR brachytherapy seeds will provide spectrometry as an integral part of the calibration service in the future. OC-0080 Comprehensive commissioning of MR-Linac online adaptive radiotherapy QA O. Green 1 , A. Price 1 , B. Cai 1 , J. Cammin 1 , V. Rodriguez 1 , J. Park 1 , S. Mutic 1 , D. Yang 1 1 Washington University School of Medicine, Radiation Oncology- Physics Division, Saint Louis, USA Purpose or Objective To establish the quality assurance process for online adaptive radiotherapy (ART) performed with a hybrid MR- linac system. Material and Methods The commissioning process consisted of three parts: independent validation of the vendor-provided Monte Carlo-based secondary dose calculation module; development of an in-house software package for plan integrity and deliverability checks; patient-specific phantom-based QA for each adapted fraction for the first five patients (a total of 42 fractions). The independent validation of the secondary dose calculation was performed via end-to-end tests with an MR-imageable phantom with inserts for ionization chambers (CIRS, Norfolk, VA). Five adaptive scenarios were simulated that included changes in target volume, changes in organ-at- risk positions, and changes in electron density (this was done by adding or subtracting water to the phantom). For each scenario, IC measurements were compared to predicted treatment planning system (TPS) dose values and correlated with independent dose calculation passing rates. The in-house software package was developed in MatLab (MathWorks, Natick, MA) to check contour integrity (gaps, islands, assigned density overrides, volumes, and whether it was used in optimization), flag undeliverable segments, and estimate plan delivery times. Patient-specific QA was performed after each adapted fraction for the first five patients using the MR- Proffered Papers: PH 1: Adaptive radiotherapy: tools and technologies

Figure 1. An example of the in-house software check for contour integrity flagging the Kidney_R structure as one for which the dose in the treatment plan is high enough to suggest being included in optimization objectives. The discontinuity of the Skin contour is also flagged as a potential issue.

Table 1 Phantom-based and independent dose calculation results for patient-specific adapted fractions using ionization chamber (IC) and ArcCheck (AC). Conclusion The complexity of online adaptation necessitates not only thorough commissioning but the establishment of on-going comprehensive quality assurance for each fraction that includes not only a phantom-less QA but also a method to ensure that all other components of the plan are accounted for and checked. In this work we have shown an example of such a comprehensive commissioning method for a hybrid MR-linac. OC-0081 Plan-library supported automated replanning for online-adaptive IMPT of cervical cancer T. Jagt 1 , S. Breedveld 1 , R. Van Haveren 1 , R. Nout 2 , E. Astreinidou 2 , M. Staring 3 , B. Heijmen 1 , M. Hoogeman 1 1 Erasmus MC University Medical Center Rotterdam, Department of Radiation Oncology, Rotterdam, The Netherlands; 2 Leiden University Medical Center, Department of Radiation Oncology, Leiden, The Netherlands ; 3 Leiden University Medical Center, Division of Image Processing- Department of Radiology, Leiden, The Netherlands Purpose or Objective Intensity-modulated proton therapy (IMPT) is very sensitive to small daily density variations along the pencil beam paths and variations in target and OAR shapes. This makes IMPT for sites with large inter-fraction target