S1731
Physics - Dose prediction/calculation, optimisation and applications for particle therapy planning
ESTRO 2026
Habraken, S. J., & Hoogeman, M. S. (2023). PTV-based VMAT vs. robust IMPT for head-and-neck cancer: A probabilistic uncertainty analysis of clinical plan evaluation with the Dutch model-based selection. Radiotherapy and Oncology, 186, 109729.[2] Andrzej Niemierko. Reporting and analyzing dose distributions: A concept of equivalent uniform dose. Medical Physics, 24(1):103–110, 1997[3] Unkelbach, J., Botas, P., Giantsoudi, D., Gorissen, B. L., & Paganetti, H. (2016). Reoptimization of intensity modulated proton therapy plans based on linear energy transfer. International Journal of Radiation Oncology* Biology* Physics, 96(5), 1097-1106. Keywords: LET optimization, IMPT, lung cancer Sparse probabilistic evaluation to enable fast probabilistic evaluation guided planning: a feasibility study in head and neck patients. Jenneke I de Jong 1,2 , Steven J.M. Habraken 3,2 , Sebastiaan Breedveld 1 , Albin Fredriksson 4 , Johan Sundström 4 , Erik Engwall 4 , Mischa S Hoogeman 1,2 1 Radiotherapy, Erasmus MC Cancer Institute, Rotterdam, Netherlands. 2 Medical Physics and Informatics, HollandPTC, Delft, Netherlands. 3 Radiation Oncology, Leiden University Medical Center, Leiden, Netherlands. 4 R&D department, RaySearch Laboratories, Stockholm, Sweden Purpose/Objective: Current robustness optimization and evaluation methods in proton therapy lead to inter-patient variation in probabilistic CTV robustness and suboptimal trade-offs between target coverage and OAR sparing [1]. Probabilistic evaluation guided planning can mitigate these limitations but remains computationally demanding. This study proposes sparse probabilistic evaluation (SPE), a Digital Poster Highlight 2388 computationally efficient approach integrated into a clinical TPS. We investigate the tradeoff between computational efficiency and accuracy achieved by this method. Material/Methods: Clinical plans of 20 IMPT HNC patients treated at our center in 2024 were included, divided into a calibration (5) and a validation (15) group. SPE relied on predefined setup error grids with Monte-Carlo computed dose distributions. Two settings were calibrated: (1) the number of errors (nsetup error=7, 33 or 123) and (2) the maximum included error Emax (3 σ or 4 σ ), with σ = √ ( σ random error ² + σ systematic error ² ), in the grid. The setup error grids were combined with seven range errors (-4.5% to 4.5% in 1.5%-point increments) for the final dose calculations (ntotal=49, 231, 861). Each setup error grid was
Conclusion: LET painting in combination with probabilistic evaluation guided CTV-based planning leads to an increased CTV-RBE and TCP, at the cost of increased dose inhomogeneity. More fine-tuned LET- optimization approaches may further enhance the trade-off between inhomogeneity and CTV-LETd. References: [1] Rojo-Santiago, J., Korevaar, E., Perkó, Z., Both, S.,
Made with FlippingBook - Share PDF online