ESTRO 2021 Abstract Book

S62

ESTRO 2021

Conclusion This study shows that the relatively simple ATS workflow can robustly perform high quality adaptations, maintaining the quality of the reference plan throughout the treatment fractions for rectal cancer MR guided radiotherapy. OC-0088 Margin calculation for multiple metastases in single-isocenter lung SBRT J. van Timmeren 1 , M. Hoogeman 2 , S. Ehrbar 1 , M. Mayinger 1 , N. Andratschke 1 , M. Guckenberger 1 , S. Tanadini- Lang 1 1 University Hospital Zurich and University of Zurich, Department of Radiation Oncology, Zurich, Switzerland; 2 Erasmus MC Cancer Institute, University Medical Center Rotterdam, Department of Radiotherapy, Rotterdam, The Netherlands Purpose or Objective A single-isocenter SBRT approach for multiple lung metastases has the potential to reduce treatment time and lower cumulative patient dose. However, the magnitude of inter-lesion motion is currently unknown and not incorporated in margin calculations. The aim of this study was to quantify relative inter-lesion motion and calculate ITV to PTV margins. Materials and Methods A cohort of 42 NSCLC patients with >1 pulmonary metastases treated with SBRT formed the basis for the calculation of random and systematic errors caused by relative inter-lesion motion. Inter-lesion motion was assessed by noting the required shifts in AP, SI and LR direction for a perfect lesion-based registration of the planning CT and CBCT, for 83 lesion-pairs. Wilcoxon signed-rank test was used to assess differences between average shifts in AP, LR, SI and 3D for several situations of inter-lesion distance, location (e.g. in the lower lobe, both in the same lung or lobe) and attachment to the pleura. Population-based safety margins for 5- fraction treatments were calculated for subgroups depending on significant differences between the inter- lesion shifts using the van Herk formula: M=M r +M s =β√(σ tot 2 +σ p 2 )-βσ p +αΣ tot ≈β heterogeneous σ tot + αΣ tot . An �� -value of 2.5 was used, the SDs of delineation- and intra-fractional motion errors were retrieved from literature and the SD of random linac errors was set to 0.6 mm according to Winston-Lutz tests. The value of β heterogeneous was assessed by simulating a spherical lung target and applying a 65%-isodose prescription. Dose profiles were subjected to Gaussian smoothing with σ-values ranging from 1 to 10 mm and distances between original and blurred profiles were calculated at the 100% isodose line to assess β heterogeneous . Results The relative inter-lesion motion was 1.2±1.1 mm (mean±SD) as 3D-vector. Significant larger inter-lesion motion was observed when at least one lesion was not attached to the pleura (3D,p=0.015 and SI,p=0.0026), or both were attached (3D,p=0.043), inter-lesion distance was large (AP,p=0.016 and LR,p=0.0016) or both lesions were in the same lobe (LR,p=0.0011). Therefore, margins were calculated for the entire population as well as for subgroups (Table1).