S1768
Physics - Dose prediction/calculation, optimisation and applications for particle therapy planning
ESTRO 2026
Conclusion: Static PAT plans exhibited DMcN and LETd values exeeding those observed in IMPT plans, potentially increasing the risk of high-dose-related side effects. LETd-optimized PAT planning reduced LETd values to levels comparable to IMPT, thereby eliminating the potential excess risk associated with elevated RBE at the proton beam end of range. Keywords: Proton arc therapy, Head and Neck cancer, RBE Digital Poster 4133 Study of an image conversion algorithm for offline adaptive proton therapy in hypofractionated prostate cancer treatment Floriana Pansini 1 , Marco Liotta 1 , Stefania Comi 1 , Valerio Ricciardi 2,1 , Giovanni Carlo Mazzola 3 , Daniela Alterio 3 , Barbara Alicia Jereczek-Fossa 3,4 , Federica Cattani 1 1 Unit of Medical Physics, European Institute of Oncology (IEO) IRCCS, Milan, Italy. 2 Unit of Medical Physics, ASST Santi Paolo e Carlo, Milan, Italy. 3 Division of Radiation Oncology, European Institute of Oncology (IEO) IRCCS, Milan, Italy. 4 Department of Oncology and Hemato-Oncology, University of Milan, Milan, Italy Purpose/Objective: Pencil beam treatment remains a complex technique that requires careful management of anatomical changes, setup variations, and range uncertainties. Therefore, verification CTs are essential during prostate treatment, especially in hypofractionated scheme (5fx). To replace routine verification CTs and evaluate the accumulated dose (AD), a CBCT-based algorithm available in RayStation® 2024A has been tested for offline Adaptive Proton Therapy (APT). Material/Methods:
The “corrected CBCT” (cCBCT) algorithm enhances CBCT images using the co-registered planning CT as a reference without altering patient anatomy. Voxel values are converted into a HU-LUT and a correction map is applied to eliminate low-frequency artifacts. The algorithm's accuracy was first evaluated using a Gammex® phantom with inserts of different densities. The mean absolute error (MAE) between reference CT and corresponding cCBCT Hounsfield Units was calculated within predefined VOIs (Tab. 1–2). A box plan (BP) was also computed on both Gammex-CT and corresponding cCBCT to evaluate dosimetric consistency. Then, for six patients, a prostate reference plan (PRP), was recalculated on five cCBCTs (1/fx), after rigid registration. CTV coverage (D95%>95%; D0.03cc<107%) and OARs clinical goals were compared between the PRP and the AD of the recalculated plans. An additional analysis was performed using the gamma passing rate (GPR). Results: Gammex analysis showed that DVH percentage differences were within 1% for most materials, except for cortical bone and air (~7%). These discrepancies were mainly attributed to the limited CBCT image quality in the presence of high-Z materials, which generate artifacts that are difficult to fully correct with the algorithm. GPR between BP on Gammex-CT and
Made with FlippingBook - Share PDF online