ESTRO 2020 Abstract Book

S307 ESTRO 2020

The probability distributions of the SDs per voxel of the three vector components (RL: right-left, CC: cranial- caudal, PA: posterior-anterior) and vector length (VL) of all pCT-rCT DVFs per patient are shown in Figure 1. Similarly, the voxel displacements in DVF 2 are shown in Figure 2.

method is used to assess the dosimetric effect of systematic anatomical changes seen on a representative CBCT. As a result, the decision whether adaptive RT (ART) is required can be made on dosimetric rather than geometric grounds. In this work we investigate how access to the delivered dose influences how often and in which Data was gathered for 190 HNC patients curatively treated between January 2018 and July 2019 (patients receiving ART due to a study protocol were excluded). Patients were treated with dual arc VMAT beams and a CTV-PTV margin of 3mm. For each patient we determined whether a dose recalculation was performed and if an ART plan was created. We analyzed the resulting groups and identified per type of anatomical change whether or not its occurrence led to the application of ART. Results Dose recalculation was performed for 50 patients. ART was applied for 16 patients. In 6 cases this was the direct result of a dose recalculation, in the other 10 ART was applied with a different cause. • ART after recalculate (6); In 4 cases the dose deviation was caused by a difference in shoulder position, leading to hot spots in the target in 2 cases (Figure 1A), and to insufficient target coverage in the other 2. In the final 2 cases the brainstem was adjacent to the target, and a small contour decrease resulted in a violation of the max dose constraint on the brainstem. • Dose recalculations that did not lead to ART (44); In 70% of cases, there was an anatomical shift (residual deformations up to 1cm / 10°), typically resulting in the target moving out of the PTV (Figure 1B). In 30%, dose was recomputed because of a contour change (up to 2.5cm). • ART without recalculate (10); Causes for ART: modification of positioning device (5), tumor shift (5) . In 3 out of the latter 5 cases, a retrospective dose recalculate showed that ART was not required. cases ART is applied. Material and Methods

Conclusion The validation methods presented here provide a straightforward way to assess the geometric reproducibility of any DIR or AAM algorithm. The results indicate that the DIR and AAM registration errors are small with respect to the anatomical differences the AAM aims to correct, thereby allowing for further clinical implementation of the AAM. References: 1. S. Van Kranen et al, Radiother. Oncol. 109 (2013), 463-468 2. M. Abdoli et al, Med. Phys. 44, 7 (2017) 3570- 3578 3. Z. H. Saleh et al, Phys. Med. Biol. 59, 3 (2014) 733-746 PD-0553 Efficient adaptive RT using CBCT-based dose recalculation for head and neck cancer patients G. Wortel 1 , S. Van Beek 1 , C. Carbaat 1 , S. Van Kranen 1 , R. Van der Bel 1 , K. Kiers 1 , O. Hamming-Vrieze 1 , E. Damen 1 , P. Remeijer 1 , T. Janssen 1 1 Netherlands Cancer Institute, Department of Radiation Oncology, Amsterdam, The Netherlands Purpose or Objective In 2017 our institute introduced dose recalculation on CBCT for head and neck cancer (HNC) patients. This