S1790
Physics - Dose prediction/calculation, optimisation and applications for particle therapy planning
ESTRO 2026
degrees of freedom in optimization and treatment delivery. Material/Methods: The SPArc-4pi plan optimization algorithm was developed based on dynamic programming (DP), in which it searches for an efficient treatment delivery arc trajectory iteratively while ensuring superior plan quality. Five skull-based chordoma patients’ CT datasets were retrospectively selected. Treatments were planned using sequential boosts based on individual clinical scenarios with a common objective of treating residual disease (Gross Target Volume; GTV) to 78 Gy in all patients. SPArc-4pi and 5 field clinical IMPT plans were generated for evaluation. Robust optimization parameters, such as a ±2 mm setup and a ±3 % range uncertainty, were applied to the GTV. DVHs metrics were used to assess the target coverage and OARs dose sparing. A published DynamicARC® system delivery sequence model was used to simulate the SPArc-4pi treatment delivery2. Results: The SPArc-4pi demonstrated significant improvement in target coverage of all 5 cases. Specifically, GTV V100% improved from 59.96±18.92% to 73.28±16.96% (p<0.01) on average. In addition, superior sparing of OARs was achieved using SPArc-4pi in all five cases, with the most clinically relevant being the brainstem (mean dose 30.61±6.91Gy for IMPT vs 20.48±3.51Gy for SPArc-4pi, p<0.02). The SPArc-4pi treatment delivery time was 359.27(286.39-456.06) seconds, which was shorter than that of the IMPT plans, 553.74(459.32-727.65) seconds, representing a 35.12±8.85% reduction in treatment time(p<0.01).
Conclusion: We present pyRadPlan as a holistic open-source treatment planning software toolkit, featuring a modular design that enables intuitive incorporation of AI-based tools. The exemplary head-and-neck treatment case highlights the feasibility and efficacy of AI methods in the radiotherapy planning workflow. References: Wieser, H.-P., Cisternas, E., Wahl, N., Ulrich, S., Stadler, A., Mescher, H., Müller, L.-R., Klinge, T., Gabrys, H., Burigo, L., Mairani, A., Ecker, S., Ackermann, B., Ellerbrock, M., Parodi, K., Jäkel, O. and Bangert, M. (2017). Development of the open-source dose calculation and optimization toolkit matRad. Med. Phys., 44: 2556- 2568.https://doi.org/10.1002/mp.12251Wahl, N., Becher, T., Bucher, L., Leininger, F., Ortkamp, T., & Stanic, G. (2025). pyRadPlan (v0.2.8). Zenodo.https://doi.org/10.5281/zenodo.16757343 Keywords: open-source software, artificial intelligence Digital Poster 5037 Overcoming dosimetric limitations in the management of complex clival chordomas using 4pi spot-scanning proton arc therapy Xiaoda Cong 1 , Peilin Liu 1 , shupeng chen 2 , Prakash chinnaiyan 1 , peter chen 1 , xuanfeng ding 1 1 radiation oncology, corewell health, royal oak, USA. 2 radiation oncology, emory, Atlanta, USA Purpose/Objective: The management of skull-based chordomas represents a therapeutic challenge due to their intrinsic radiation resistance and location, typically abutting critical organs at risk (OARs), including the optic nerves, chiasm, and brainstem. Even with state- of-the-art techniques, as Intensity Modulated Proton Therapy (IMPT), it is still difficult to achieve optimal target coverage with therapeutic doses for radiation1. To improve tumor coverage, which is expected to translate to improved local control of skull-based chordomas and quality of life for this patient population, we introduce a 4pi spot-scanning proton arc therapy technique (SPArc-4pi) to fully utilize
Made with FlippingBook - Share PDF online