S2152
Physics - Inter-fraction motion management and daily adaptive radiotherapy
ESTRO 2026
Digital Poster 3246 Automated Workflow for Online Adaptive Upright Proton Therapy Chenni Xu 1 , Alexander Pryanichnikov 2,3 , Shimshon Winograd 1 , Yair Hillman 4 , Aviad Berger 4 , Philip Blumenfeld 4 , Jon Feldman 4 , Aron Popovtzer 4 1 Research and development, P-Cure Ltd./Inc, Shilat, Israel. 2 Institute of Biomedical Engineering, Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany. 3 Division of Biomedical Physics in Radiation Oncology, German Cancer Research Center (DKFZ), Heidelberg, Germany. 4 Sharett Institute of Oncology, Hadassah Medical Center, Hebrew University of Jerusalem, Jerusalem, Israel Purpose/Objective: Online adaptive proton therapy enables plan re- optimization on the treatment day to account for daily anatomical variations such as bladder or rectal filling, bowel motion, and tumor displacement. It requires daily high-quality imaging to enable plan re- optimization based on day-to-day anatomical changes. This capability is available in upright image-guided proton therapy systems equipped with an integrated vertical CT in the treatment room. Because the adaptive process is time-critical, typically limited to 15– 30 minutes, this work aimed to develop a fully scripted workflow to automate key steps, minimizing manual input and reducing adaptive cycle time. Material/Methods: This workflow was developed for the P-Cure upright proton therapy solution, which provides image-guided proton therapy with a diagnostic-quality vertical CT scanner (Philips Brilliance Big Bore) integrated in the treatment room [1–3]. Following the semi-automated approach [4], full automation was implemented in RayStation v2025SP1 using Python scripting interface.After daily CT acquisition, the script performs deformable registration between the planning and daily CTs, automatically propagates targets and organs-at-risk (OARs), and generates a valid external contour within the field of view (FOV). The process includes clearing existing ROI geometries, creating frame-of-reference and rigid registrations, and aligning CTs by translation and rotation. The script then defines FOV and external ROIs using RayStation’s limited FOV algorithm, crops the external contour to the valid image stack, removes holes, and assigns air override for dose pass-through beyond the FOV. A hybrid deformable registration is then created and set as the default for dose deformation. Based on this registration, ROIs, excluding external-related ones, are automatically mapped to the daily CT.Finally, two adaptive plans are generated: (1) a recomputed plan using the original beamset and (2) an adapted plan re- optimized for current anatomy (Figure 1). Radiation
was performed daily; for prostate and rectal cases, during the first three fractions and weekly thereafter.Detected deviations prompted a review of stored treatment-day images to determine their origin. Common findings included patient misidentification, weight loss, air cavities, or incorrect placement of immobilisation devices (e.g., shoulder displacement, compression arc misalignment, or incorrect support position). Results: Analysis showed that 49% of detected deviations were related to shoulder mispositioning (Figure 1), revealing limitations in the existing immobilisation systems. Anatomical changes due to patient weight loss accounted for 31% of anomalies, particularly in head and neck treatments. Tumour volume reduction was also observed during lung and head and neck radiotherapy. Implementation of this workflow enabled early identification and correction of systematic set-up deviations. Following these findings, new shoulder immobilisers were introduced, significantly reducing positioning-related errors. The workflow also facilitated clinician review and replanning in cases of relevant anatomical or tumour volume change. Conclusion: Flat-panel in vivo dosimetry proved to be a powerful tool for offline adaptive radiotherapy and daily treatment verification. Beyond detecting deviations, it provided actionable clinical feedback that resulted in improved immobilisation, workflow optimisation, and patient safety. Incorporating this approach into routine clinical practice enhances overall treatment quality assurance in EBRT. References: 1. Olaciregui-Ruiz I, Mijnheer B, Rozendaal R, et al. In vivo dosimetry in external beam photon radiotherapy: Requirements and future directions for research, development and clinical practice. Phys Imaging Radiat Oncol. 2020;15:108-116. doi:10.1016/j.phro.2020.08.0032. Mijnheer B, Beddar S, Izewska J, Reft C. In vivo dosimetry in external beam radiotherapy. Med Phys. 2013;40(7):070903. doi:10.1118/1.48112163. D’Agostino E, Koivisto J, Persoon L, et al. Assessing the impact of adaptations to the clinical workflow in radiotherapy using transit in vivo dosimetry. Phys Imaging Radiat Oncol. 2023;27:100474. doi:10.1016/j.phro.2023.100474 Keywords: In vivo dosimetry, adaptive radiotherapy
Made with FlippingBook - Share PDF online