ESTRO 2020 Abstract book

S1095 ESTRO 2020

5x7Gy 0%(=0/30), 3x10Gy 0%(=0/30) and 1x20Gy 43%(=13/30). The resulting 5x7Gy and 3x10Gy treatment plans were comparable (Figure 1). The 1x20Gy treatment plans resulted into a significantly lower HI, but significantly higher R 100% , R 50% and D 2cm (Wilcoxon matched- pairs signed rank test, p<0.05). Plan quality for these plans can potentially be improved by introducing more beam angles.

larger the extent of rotational shift, the larger the differences in loss of PTV coverage between 2 approaches. For OARs doses, there were no apparent and consistent results that can be concluded. Conclusion Single-isocenter plans can have comparable dosimetric results as multiple-isocenter plans. However, after application of isocenter shifts, coverage for PTVs could be compromised using single-isocenter. The risk increased with intra-fractional rotational errors.These results would provide guidance in understanding benefits and risks of 2 approaches. PO‐1868 In silico hypofractionation feasibility study for lymph node oligometastases SBRT on the MRI‐linac D. Winkel 1 , W.S.C. Eppinga 1 , M.P.W. Intven 1 , A.M. Werensteijn-Honingh 1 , J. Hes 1 , L.M.W. Snoeren 1 , G.H. Bol 1 , B.W. Raaymakers 1 , I.M. Jürgenliemk-Schulz 1 , P.S. Kroon 1 1 UMC Utrecht, Department of Radiotherapy, Utrecht, The Netherlands Purpose or Objective At our department, MR-guided SBRT using the 1.5T MR- linac system (MRL) (Unity, Elekta AB, Stockholm, Sweden) has been initiated for patients with lymph node oligometastases (LNO). Superior soft tissue contrast and the possibility for online plan adaptation on the MRL allows for hypofractionated treatment. The purpose of this study was to investigate the feasibility and compare plan quality of different hypofractionation schemes. Material and Methods Data was used from 12 patients with single LNO (10 pelvic, 2 para-aortic), which were all treated on the MRL with a prescribed dose of 5x7Gy to 95% of the PTV. Possibility for further hypofractionation was investigated based on the pre-treatment plan data. If all planning constraints (Table 1) were met for the 5x7Gy plan a 3x10Gy plan was created and if this plan met all constraints a 1x20Gy plan was generated (all BED 60Gy with α/β=10). All plans were made using IMRT with 7 or 9 beams, respectively. In case all three differently fractionated pre-treatment plans met all clinical dose criteria, the plans were adapted to the daily patient anatomy [1]. These were then used for quality and dose criteria based evaluation. The quality of the different plans was evaluated using four parameters: heterogeneity index HI (=PD* the actual prescription dose defined as D 95% of the PTV as a % of D max ), conformity index R 100% (=V PD* /V PTV ), R 50% (=V 0.5xPD* /V PTV ) and D 2cm (=D max at 2cm from PTV as % of PD*). These values were compared with benchmarks, of which R 50% and D 2cm depend on the size of the PTV [2].

Conclusion Hypofractionation of LNO on the MRL is feasible, but the most suitable fractionation scheme should be investigated by pre-treatment imaging and planning. The optimal fractionation scheme differs per patient and is dependable on the presence of OARs near the target. For single fraction treatment, in which OAR constraints are more limiting, pre-treatment planning alone is not an optimal indicator for the online daily situation. [1] D. Winkel et al. Adaptive radiotherapy: The Elekta Unity MR-linac concept. ctRO 2019;18:54-59. [2] S. Chmura et al. A Phase 1 Study of Stereotactic Body Radiotherapy (SBRT) for the Treatment of Multiple Metastases. 2015: ClinicalTrials.gov NCT02206334. PO‐1869 Dosimetry impact of manual vs. automated segmentation of lung contours in RayStation (RS) F. Vallejo Castañeda 1 , M. Hinse 1 1 Centre intégré de cancérologie de Laval, Département de radio-oncologie, Laval, Canada Purpose or Objective To investigate the dosimetric impact of manual vs. automated segmentation of lung contours created by ATLAS base segmentation (ABS) and model base segmentation (MBS) considering specific dose constraints and contouring creation time. Material and Methods A total number of 45 lung patients was used for the study. 20 of them were used to create an ATLAS while the remaining 25 were used for validation. The original RS MBS was used where no patient data was required. Each validation patient had both lungs contoured manually in Pinnacle and further exported to RS. In RS two other sets of lungs contours were created using ABS and MBS. Lungs-

Results It was possible to create pre-treatment plans that met all dose constraints in 10/12, 9/12 and 6/12 patient cases for the 5x7Gy, 3x10Gy and 1x20Gy scheme, respectively. Violations were caused by OARs overlapping or adjacent to the PTV. Dose criteria were violated in the online plans of