ESTRO 38 Abstract book

S479 ESTRO 38

calibrated in dose to water in a water phantom. Monte Carlo modelling is required to convert the measurements to either dose to water (D w,w ) or dose to medium ( D m,m ). The EGSnrc code was used to correct the raw measurements with a microdiamond in trabecular bone as shown in Figure 1. Before correction the measurements were higher than planned doses for both D w,w and D m,m with averages of -2.1% and -4.1% respectively. After correction the averages are -0.8% and +0.2% respectively.

Material and Methods The PTW RW3 and CIRS4D thorax phantom were used to verify the dose distribution of 10 lung cancer patients with GTV volumes below 1cm 3 . EBT-XD film measurements were performed in the RW3 slab phantom at 5 cm depth. Treatment plan verification was performed in the CIRS4D phantom in a film rod containing a 0.1 cm 3 water- equivalent target. All measurements were performed on a 6MV Varian Truebeam STx linac for single fraction doses of 15 or 18Gy. Treatment plans were calculated in Eclipse 15.5 with the AcurosXB 15.5.11 dose algorithm. Films were analyzed using an EPSON 11000 XL film scanner and FilmQA Pro software (Ashland, USA), following a dedicated protocol (Lewis et al., 2012; Lewis et al. 2016). Results The average GTV volume for the 10 patients was 0.6±0.5 cm 3 . Regarding the homogeneous RW3 verifications, the average difference between maximum dose on film and maximum dose on the corresponding dose plane in the TPS was 2.6±1.6%. Recalculating the patient plans in the CIRS4D phantom in lung with an artificial lung tumor matching the patient tumor size yields differences between maximum dose on film and maximum dose on the corresponding dose plane in Eclipse up to 16.5% with an average overdose of 10.8±4.7%. In general, larger differences were found between measurement and predicted dose by the AcurosXB algorithm as tumor size Dose verification of small lung tumors should be performed in a heterogeneous phantom incorporating a water-equivalent target that is of a comparable size to the original tumor to be treated. Current verification methods in homogeneous phantoms do not seem suitable for these challenging stereotactic lung treatments. The AcurosXB dose calculation consistently underestimates dose in (very) small lung tumors. - Fogliata et al., Physica Medica, 44, 167-162, 2017 - Lewis et al., Med. Phys. 39(10), 6339, 2012 - Lewis et al., Med.Phys., 42(2), 643-649, 2016 PO-0904 Benchmarking of a module for Monte Carlo simulation of proton transport in the PENELOPE code N. Verbeek 1,2,3 , S. Smyczek 1,3,4 , J. Wulff 1,3 , C. Bäumer 1,3,5 , B. Timmermann 1,3,5,6 , L. Brualla 1,2,3 1 University Hospital Essen, West German Proton Therapy Centre Essen WPE, Essen, Germany ; 2 University of Duisburg-Essen, Faculty of Medicine, Essen, Germany ; 3 University Hospital Essen, West German Cancer Center WTZ, Essen, Germany ; 4 Heinrich Heine University decreased. Conclusion

Figure 1 Summary of dose differences ((planned- measured)/measured) from spine point dose in trabecular bone. For measurements in solid water, an algorithm dependent correction for dose to tissue is planned for TPS reporting dose to medium. Conclusion For measurements in bone, the ACDS applies Monte Carlo corrections to report D w,w and D m,m with a detector calibrated with dose to water. The ACDS does not calculate and report on Dose to water in medium, as determined by applying stopping power ratios to dose to medium calculations. PO-0903 AcurosXB dose verification of ultra-small lung lesions with EBT-XD film in a heterogeneous phantom M. Öllers 1 , A. Swinnen 1 , F. Verhaegen 1 1 Maastro Clinic, Medical Physics, Maastricht, The Netherlands Purpose or Objective Modern type ‘c’ dose calculation algorithms can predict dose for stereotactic treated lung tumors larger than 4cm 3 with an uncertainty up to 5% (Fogliata et al., 2017). However, the trend to treat tumors with volumes below 1cm 3 with hypofractionation poses a serious challenge even for type ‘c’ algorithms. Common recommendations for QA and dose verification in SBRT treatments are to use phantoms that are homogeneous: either water-equivalent (e.g. PTW RW3/Octavius, SunNuclear Delta4) or lung- equivalent materials (e.g. CIRS thorax). The CIRS4D thorax phantom provides a lung-equivalent film rod containing a water-equivalent spherical target (figure 1) that provides a more realistic verification set-up. Our aim was to evaluate the accuracy of the AcurosXB dose algorithm with EBT-XD film in a homogeneous and heterogeneous phantom for targets below 1cm 3