ESTRO 2020 Abstract book

S872 ESTRO 2020

A programmable phantom with a sin 6 motion patterns (20- mm diameter “target”; period = 4 s; motion magnitude of 0.5, 1, 2 and 3 cm) were used for this simulation study. Air cavities of sizes 0, 0.5 and 1.0 cm were introduced near the target using three custom made inserts. For all inserts, 4DCT scans were acquired and VMAT treatment plans were generated using the ITV approach. The PTV was created by an isotropic 5 mm margin of the ITV. The clinical plans were optimized and dose distributions were calculated on the AIP. 4D dose calculations were utilized to evaluate the effect of an air cavity near the target cases by a poor pulmonary lung function. The calculations were performed on the static phantom scan with series of isocentre shifts corresponding to the target position of each 4DCT phase. The 4D dose calculations were accumulated and compared to the 3D dose calculations on the TPS. Dose metric of minimum dose, D 99% , maximum dose, D 1% , and mean dose, D mean were compared to obtain the dose discrepancy in the target of each air cavity insert. The same procedures were repeated to assess the dose deficiency due to set up errors of 0, 0.2, 0.3, 0.5, 0.6 and 1.0 cm in the D 99% . Results Figure 1 shows the dose discrepancy in the investigated dosimetric parameters between 3D and 4D dose calculations were found to be within -0.15% to 2.81 %, - 0.02% to 2.56% and -0.05% to 5.62% for the target of 0, 0.5 and 1.0 cm air cavity, respectively. Figure 2 illustrates the dose deficiency in the investigated D 99% between 4D dose calculations and the prescription dose was found to be less than 5% up to 5 mm setup error for targets of 0 and 0.5 cm air cavities, but exceeded 5% (clinically significant) with only 3 mm setup errors for target of 1.0 cm air cavity. Conclusion The evaluated dose discrepancy due to air cavity and deficiency due to setup errors during lung VMAT SBRT treatments resulted in non-negligible dose variations. Therefore, performing the 3D dose calculation based on the AIP, rather than the 4D dose calculations may not be adequate, particularly for a target near a large air cavity. Furthermore, target near an air cavity needs special care prior to treatment to ensure high accuracy in the target localization as they are more sensitive to setup errors. Future work includes patients study could offer a better understanding of the dosimetric effect of an air cavity near a moving tumor. PO-1523 Predictive factors of over Grade 2 radiation pneumonitis for advanced non-small cell lung cancer. K. Muraki 1 , E. Ogo 1 , H. Suefuji 1 , H. Eto 1 , C. Tsuji 1 , C. Hattori 1 , Y. Miyata 1 , T. Abe 1 1 Kurume University, Radiology, Kurume, Japan Purpose or Objective The standard of treatment for patients with locally advanced non-small cell lung cancer (LA-NSCLC) is the platinum based chemoradiotherapy (CRT) and followed by anti PD-L1 antibody durvulmab. However, we should withhold durvalumab for patients with Grade ≧ 2 radiation pneumonitis (RP). We aimed to identify dosimetric and radiobiological factors for predicting Grade ≧ 2 radiation pneumonitis after CRT for LA-NSCLC. Material and Methods We retrospectively analyzed data from 73 patients (RP Grade1 ： Grade2 = 43 : 30) with LA-NSCLC treated with CRT in our hospital between 2013 to 2018. The NTCP was calculated with the biological evaluation tool RayBiology in RayStation (Ver. 6.2). Univariate and multivariate analyses were performed to test the association between radiation pneumonitis and dosimetric and radiobiological factors; the mean lung dose (MLD), the normal lung Poster: Physics track: Radiobiological and predictive modelling, and radiomics

the CTV volumes for each balloon filling was checked to be <2%. A balloon volume up to 150cc was well-tolerated by all patients. A balloon volume of 200 cc produced some discomfort in 2 of the patients. On average dose tolerances were met for all organs at risk (fig2). However, two patient plans exceeded UW dose tolerances for the lowest balloon fillings. RW tolerance dose was exceeded in three patient plans for the highest balloon filling. Individual plan optimization would improve these results, but also increase the subjectivity of the study. No significant effect was observed (<3%) for BW, RW, and UW with respect to balloon filling (fig 2). Regarding RW, the highest priority was needed for the D1cc leading the optimization to the limit. For UGD and NVBs the Dmax appears to decrease with balloon volume. However, the increasing plan difficulty for large balloon volumes mostly manifests itself on the PTV coverage, which decreased 8% from 0cc to 250cc (fig.2a).

Conclusion An endorectal balloon filling of 150cc is well tolerated. Increased rectal balloon filling reduces PTV coverage if organ at risk doses are to be maintained. For our prostate hypofractionation we have chosen 150cc as the best compromise, accepting 5% decrease in dose coverage and prioritizing anatomy fixation and OAR sparing. PO-1522 3D vs 4D dose calculations for moving target near air cavity for lung VMAT SBRT planning Y. AlmutairI 1 , L. Leon Vintro 2 , B. McClean 3 1 University College Dublin & Saint Lukes Radiation Oncology Network, Physics, Dublin, Ireland ; 2 University College Dublin, School of Physics, Dublin, Ireland ; 3 Saint Lukes Radiation Oncology Network, Department of Physics, Dublin, Ireland Purpose or Objective An air cavity near a moving target is a non-trivial source of uncertainty in lung SBRT. A common method is to calculate a lung SBRT treatment plan using the average intensity projection (AIP) of the 4DCT scans. In this study, the extent of air cavity caused by emphysema in a poor pulmonary lung function, in addition to the dose deficiency due to set up errors was investigated. This effect is often ignored when the ITV concept is utilized. Material and Methods