ESTRO 2021 Abstract Book

S515

ESTRO 2021

moderate to severe after effects of the available treatment techniques and prevent them by choosing the most appropriate treatment modality.

OC-0647 Improved robustness in oesophageal cancer treatment by diaphragm based position verification S. Visser 1 , L.A. den Otter 1 , C. O.Ribeiro 1 , E.W. Korevaar 1 , S. Both 1 , J.A. Langendijk 1 , C.T. Muijs 1 , N.M. Sijtsema 1 , A. Knopf 1 1 UMCG, Radiotherapy, Groningen, The Netherlands Purpose or Objective Compared to conventional radiotherapy techniques, improved organs-at-risk sparing can be achieved with advanced radiotherapy techniques, such as volumetric modulated arc therapy (VMAT) and intensity modulated proton therapy (IMPT). However, VMAT and IMPT are sensitive to motion and anatomical variations. In the treatment of distal oesophageal cancer, the target dose coverage may be affected by target displacements and density changes in the beam path due to diaphragm position variations. CBCT images used for position verification (PV) in distal oesophageal cancer show the diaphragm in addition to the target volume and the bony anatomy. Therefore, we evaluated whether a correction for diaphragm position changes (PV Diaphragm ) observed during PV would improve daily target dose coverage, compared to PV based on bony anatomy (PV Bones ) and PV based on the target volume (PV Target ). Materials and Methods Weekly verification 4DCTs of ten oesophageal cancer patients were rigidly registered to the planning 4DCT simulating PV Bones . Inter-fractional cranial-caudal target displacements and their relation with diaphragm variations were investigated, using deformation vector fields resulting from deformable image registration. This information was used to determine the most optimal registration method based on both bony anatomy and diaphragm shifts, simulating PV Diaphragm . A third registration was performed based on the target volume, simulating PV Target . Subsequently, VMAT and IMPT plans were robustly evaluated on each verification 4DCT using the three PV methods. Inter-fractional target dose coverage (dose that 98% of the target volume receives [D 98% ]) was evaluated on the resulting voxel-wise minimum dose distributions. Results The cranial-caudal mean target displacement was congruent with nearly half of the diaphragm displacement (Fig. 1; y=0.439x), which was used for the diaphragm correction in PV Diaphragm . Target dose coverage using PV Bones was adequate (D 98% ≥94%) for patients with diaphragm displacements up till 12 mm (Fig. 2). For larger diaphragm displacements, the target coverage was better maintained by PV Target and PV Diaphragm , as target displacements were better accounted for (Fig. 2). An additional benefit was observed for PV Diaphragm , as the densities in the beam path were more similar to the planning situation (D 98% : IMPT 95% ± 5%, VMAT 97% ± 6%). A diaphragm displacement of more than 12 mm was observed in 16% of all verification 4DCTs. Conclusion Position verification in VMAT and IMPT for distal oesophageal cancer based on bony anatomy can be improved employing a diaphragm position correction for diaphragm displacements >12 mm. Improved target dose coverage can be achieved compared to image registration based on bony anatomy alone or based on the target. To detect the cases where target dose coverage could be compromised due to diaphragm variations, we recommend monitoring of the diaphragm position prior to treatment through online CBCT imaging.