ESTRO 2025 - Abstract Book

S3262

Physics - Intra-fraction motion management and real-time adaptive radiotherapy

ESTRO 2025

Amsterdam, Netherlands. 3 Radiation Oncology, University Medical Center Mannheim, Mannheim, Germany. 4 Radiation Oncology, University Medical Center Schleswig-Holstein, Kiel, Germany. 5 Arrhythmia Unit, Hospital Clínico San Carlos, Madrid, Spain. 6 Oncology, University Hospital and Faculty of Medicine, Ostrava, Czech Republic. 7 Radiation Oncology, Maastricht University Medical Centre+, Maastricht, Netherlands. 8 Klinik für Radioonkologie, Klinikum Chemnitz, Chemnitz, Germany. 9 Radiotherapy, Maria Skłodowska-Curie National Research Institute of Oncology, Gliwice, Poland. 10 Electrocardiology and Heart Failure, Medical University of Silesia in Katowice, Katowice, Poland. 11 Oncology and Radiotherapy, Medical University of Gdańsk, Gdańsk, Poland Purpose/Objective: Cardiorespiratory motion (CRM) affects the delivery of stereotactic arrhythmia radioablation (STAR) for ventricular tachycardia (VT) patients. We analysed CRM and its dosimetric impact in an international, multi-institutional STAR patient cohort (STOPSTORM, EU-Horizon-2020-Grant 945119). Material/Methods: Nine centres provided treatment data for 62 STAR-treated patients. Thereof, 47% were treated in free-breathing (FB) with an ITV approach, 16% in FB with target tracking (TR), 16% with abdominal compression (AC), and 21% in breath hold (BH) with mean PTV sizes of 147.6cc, 57.5cc, 65.9cc and 39.0cc, respectively. For FB and AC patients, 4DCT scans with 5-10 respiratory phases and for BH contrast-enhanced CT in diastole and systole were acquired. Sixteen cardiac substructures (CS) were auto-contoured on the planning CTs [1]. To reduce image contrast differences for subsequent registrations, the planning CTs were deformably registered (Elastix [2]) to either the closest 4D-CT breathing-phase (FB/AC) or the diastolic CT (BH), which were assigned as reference CT. The resulting deformation vector field (DVF) was applied to the dose file and contour set. To assess motion, reference CTs were deformably registered to the remaining phases and the resulting DVFs were used to deform all contours. For TR, rigid target motion was compensated. Deformed contours were compared to the original contours using 95% Hausdorff distance (HD95) and Volume Ratio (VR). Planned dose was compared to the dose accumulated over the different phases based on the Conformity Index (CI) for targets, defined as (ptv ∩ V25Gy)^2/(PTV*V25Gy), with Vptv ∩ 25Gy as the overlapping volume between PTV and the 25Gy isovolume (V25Gy), and Dmean and Dmax for CS.

Results: