S2173
Physics - Inter-fraction motion management and daily adaptive radiotherapy
ESTRO 2026
and carbon-ion fields.Patient data included 40 CBCTs from 20 patients (10 protons, 10 carbon-ions). From the available CBCTs acquired during treatment, those corresponding to days with a re-evaluation CT (CTRE) were retrospectively selected; CTRE served as ground truth.The study comprised two phases:1.Algorithm performance assessment — sCTs were generated using CTRE as reference to evaluate the accuracy and robustness of the algorithm.2.Simulation of an adaptive workflow — sCTs were generated using the planning CT (CTplan) as reference, simulating a fully synthetic adaptive workflow without additional CTs.Evaluations included morphological assessment and dosimetric validation. The latter comprised plan recalculation on sCTs, comparison of key dosimetric parameters, and local gamma analysis (2%/2mm). Results: In phantom tests, distal range differences between reference CT and sCT were below 1% for most materials and up to 3% for dense bone and lung- equivalent materials. In patients, local gamma evaluation exceeded 95%, confirming good agreement between the clinical plan and recalculated dose on sCTs. Target coverage indexes were consistent, though dose estimation to OARs was slightly less precise, particularly in small-volume optic structures (Table 1).Among the 20 patients, 6 required replanning: 2 due to increased target D1% and 4 due to decreased D95%. Inspection of the sCTs would have led to the same clinical decisions regarding replanning as those based on the CTRE scans (Fig.1).
Conclusion: CBCT-derived sCTs generated with the Virtual CT algorithm accurately reproduce patient anatomy and enable reliable dose calculation for proton and carbon-ion plans. For head and neck cases, sCTs could trigger adaptive replanning, detect significant anatomical changes, reduce unnecessary CTs, and improve workflow efficiency in particle therapy. Further studies on larger cohorts are needed to confirm these findings and consolidate the clinical applicability of sCT-based adaptive workflows. Keywords: Particle therapy, Offline Adaptive, Synthetic CT Evaluation in Healthy Volunteers of 3D-Printed Head and Neck Radiotherapy Immobilization Bertrand Dewit 1 , Ronald Peeters 2 , Laurence Delombaerde 3 , Sandra Nuyts 1,3 , Tom Depuydt 1,3 1 Department of Onclogy, KU Leuven, Leuven, Belgium. 2 Department of Radiology, UZ Leuven, Leuven, Belgium. 3 Department of Radiation Oncology, UZ Leuven, Leuven, Belgium Purpose/Objective: Accurate head and neck immobilization is essential in radiotherapy to ensure precise targeting while sparing healthy tissue. Thermoplastic masks are the current standard, but often cause discomfort and anxiety. 3D- printing offers a digital alternative for personalized immobilization that can be created without direct patient involvement, potentially offering improved accuracy and comfort. This study evaluated a novel in- house developed 3D-printed head immobilization Digital Poster 4087
Made with FlippingBook - Share PDF online