ESTRO 2025 - Abstract Book

S3638

Physics - Quality assurance and auditing

ESTRO 2025

concerns about treatment precision. The conclusions are applicable only for our particular machines and software configuration.

Keywords: synthetic CT, MR simulation, spatial accuracy

References: 1. Fusella M, Alvarez Andres E, Villegas F, Milan L, Janssen TM, Dal Bello R, Garibaldi C, Placidi L, Cusumano D. Results of 2023 survey on the use of synthetic computed tomography for magnetic resonance Imaging-only radiotherapy: Current status and future steps. Phys Imaging Radiat Oncol. 2024 Sep 26;32:100652. doi: 10.1016/j.phro.2024.100652. PMID: 39381612; PMCID: PMC11460247. 2. Bahloul MA, Jabeen S, Benoumhani S, Alsaleh HA, Belkhatir Z, Al-Wabil A. Advancements in synthetic CT generation from MRI: A review of techniques, and trends in radiation therapy planning. J Appl Clin Med Phys. 2024 Nov;25(11):e14499. doi: 10.1002/acm2.14499. Epub 2024 Sep 26. PMID: 39325781; PMCID: PMC11539972.

2337

Digital Poster Use of the EPID to verify multi-isocenter craniospinal irradiation treatments Paul J Doolan 1 , Melka Benjamin 2 , Mary Peratikou 2 , Constantinos Zamboglou 2,3 , Konstantinos Ferentinos 2 , Efstratios Karagiannis 2,4 1 Medical Physics, German Oncology Center, Limassol, Cyprus. 2 Radiation Oncology, German Oncology Center, Limassol, Cyprus. 3 Radiation Oncology, University of Freiberg, Freiberg, Germany. 4 Radiation Oncology, Cantonal Hospital of Graubünden, Chur, Switzerland Purpose/Objective: To investigation the possibility of using the EPID to verify the delivery of multi-isocentric craniospinal irradiation (CSI). Material/Methods: An adult medulloblastoma patient was treated (in the first phase) with 35.2 Gy in 22# to the brain and whole spine. For the treatment length and sufficient overlap (76 + 5 cm), four isocentres were required. A treatment plan, robust to a 1.5 mm shift in the cranial-caudal direction, was made in RayStation (Raysearch Labs, Stockholm, Sweden), with low gradient sloping junctions between isocentres. The IGRT approach further aimed to avoid possible hot and cold spots. At each isocentre, CBCT images were acquired, matches were performed, shifts were recorded, but no shifts were applied. After all CBCTs had been acquired, a single shift was applied in each direction, so the patient was only moved once for the whole treatment. This shift was either: (i) equal to the brain shifts; or (ii) an equally weighted average of the four CBCT shifts. With IGRT approach (i), the brain was matched but all other isocentres were not matched. In IGRT approach (ii) (equal weight), no isocentres were perfectly matched. For all fields, EPID in vivo dosimetry was performed, using gamma analysis (4%/3 mm, 10% threshold, global normalisation) assessed in PerFRACTION (Sun Nuclear, Melbourne, USA). The correlation between the root mean square error (RMSE) of the match (X, Y, Z) and the EPID passing rate was assessed using Spearman’s rank coefficient. Results: When matching only on the brain, the EPID passing rate for the brain isocentre was 90.8%, with the other isocentres showing passing rates of 95.3%, 94.8% and 71.2%. When matching on an average, equally weighted across the four shifts, the brain EPID passing rate dropped to 85.0%, but the other isocentres were equal or better: 95.3%, 95.1% and 74.6%. Examples in Figure 1.