S1761
Physics - Dose prediction/calculation, optimisation and applications for particle therapy planning
ESTRO 2026
system (RaySearch Laboratories, Sweden); for the phantom, 22 two-field plans with varying angles were created. All the plans were similar under the assumption of RBE=1.1. Each plan was evaluated in terms of physical dose (PD), LETd and DD (Figure 1). The DD/PD(%) and LETd were analyzed in all clinical structures, as well as in concentric shells surrounding the CTV. The DD and PD for D5%(near-maximum) and D50%(median) were evaluated. Results: An example of axial distribution of PD, LETd, DD for one proton field is shown in Figure 1. The location of the brainstem, one of the most sensitive organs of interest in this example, is indicated for guidance, to illustrate the complementing value of the evaluation of LETd and DD/PD in addition to the physical dose. Analysis of patients plans and concentric shell structures (Figure 2) showed that under the applied robustness parameters, the LETd increased even beyond 10 mm from the CTV, where PD was low, indicating limited clinical relevance of this rice. The highest DD/PD5 for the different plans were observed in the shell located 4.5-7.5 mm from the CTV, where physical dose remains substantial. This suggests that neglecting the dirty dose in this region may increase the risk for biological side effects.
Conclusion: Evaluation of DD provides information on potential biological effects that is complementary to LETd and should be considered when selecting proton therapy plans to better manage toxicity risks. In regions with overlapping fields or near the distal end of the proton range, LETd may not directly predict the biological effect. Furthermore, as LETd is an averaged metric, it might need to be assessed for each field individually, unlike additive parameters such as DD. These findings highlight the complexity of beam arrangement in proton therapy. Keywords: LET, Dirty Dose, RBE Digital Poster 3930 Fast and memory efficient Monte-Carlo based plan optimization in proton therapy : A pairwise beamlet approach Romain Schyns, Ana Maria Barragan Montero, John Lee MIRO, UCLouvain, Brussels, Belgium Purpose/Objective: Beamlet-based plan optimization, relying on Monte Carlo (MC) dose calculation engines [1] to precompute spot beamlets, currently represents the gold standard in proton therapy.However, this gain in accuracy with MC simulations comes at a prohibitive computational cost, particularly for advanced and time-constrained delivery modalities such as proton arc therapy [2] or online adaptive treatments, respectively. A
Made with FlippingBook - Share PDF online