ESTRO 2025 - Abstract Book

S3469

Physics - Optimisation, algorithms and applications for ion beam treatment planning

ESTRO 2025

References: 1. Tsujii H, Kamada T. A review of update clinical results. Jpn J Clin Oncol . 2012

2. Manganaro et al. A Monte Carlo approach, Med Phys . 2017 3. Kase Y et al. Microdosimetric measurements. Radiat Res . 2006 4. Inaniwa T et al. Treatment planning. Phys Med Biol . 2010 5. Kase Y et al. Biophysical calculation of cell survival probabilities. Phys Med Biol . 2008 6. Chatterjee A et al. Microdosimetric Structure. Radiat Environ Biophys . 1976 7. Kiefer J et al. A Model of Ion Track Structure . Vol 31.; 1986. 8. Hawkins RB. A microdosimetric-kinetic theory of the dependence of the RBE. Med Phys . 1998 9. Furusawa Y et al. Inactivation of Aerobic and Hypoxic Cells. Radiat Res . 2000 10. Sato T et al. Cell survival fraction estimation. Radiat Res . 2012

1743

Digital Poster Reducing ΔNTCP by improving proton radiotherapy planning in advanced stage non-small cell lung cancer (NSCLC) patients Petra Klinker, Robin Wijsman, Fred Ubbels, Pieter Deseyne, Olga Chouvalova, Stefanie A. de Boer, Johannes A. Langendijk, Erik Korevaar Radiation Oncology, University Medical Centre Groningen, University of Groningen, Groningen, Netherlands Purpose/Objective: Since October 2019, over 300 patients have been treated with intensity-modulated proton therapy (IMPT) for loco regionally advanced stage non-small cell lung cancer. Patient selection was based on ΔNTCP differences between photon (VMAT) and proton treatment plans. We performed a retrospective planning study to evaluate whether we could further optimize our planning approach in terms of dose to organs at risk (OARs), robustness and ΔNTCP. Material/Methods: Ten NSCLC patients were included, receiving chemo-radiation of 60 Gy in 25 fractions. Patients qualified for proton therapy if one of the following criteria were met; ΔNTCP for acute grade 2 esophageal toxicity (AET) or grade 2 radiation pneumonitis (RP) ≥10% or multiple grade 2 for AET and RP ≥ 15% or 2-year mortality (2yM) ≥2%. Preparation was based on a 4DCT to establish target motion (maximum 15 mm allowed) and internal target volume (ITV) construction. Treatment plans were robustly optimized with three beams for 6 mm setup and 3% range uncertainty. For robustness against density changes, the ITV was overridden to ‘muscle’ during optimisation (removed for final dose calculation).10% modulation per beam was allowed by a minimal dose of 18 Gy per beam to the ITV. We compared the standard optimisation (sOpt) workflow with an optimisation strategy without minimum dose per beam restrictions and increased focus on dose conformity (reOpt). An extra CT was used during optimisation; one with override, one without. Differences in dose to OARs between the sOpt and reOpt plans were evaluated using the Wilcoxon signed rank test. Results: Differences in mean dose to the heart, lungs and esophagus, and differences in ΔNTCP between the sOpt plans and reOpt plans are listed in table 1. For all OARs, the mean dose was significantly lower in the reOpt plans compared to the sOpt plans. This resulted in an increase in ΔNTCP of 1.4%, 0.5%, 5.9% for 2yM, RP and AET, respectively. All patients already qualified on 2yM in both planning strategies. The number of patients that qualified on RP was zero and one in the sOpt and reOpt plans, respectively. For AET these numbers were three and six. All plans were robustly evaluated and met the goal of V94%>98% ITV coverage for the voxelwise minimal dose distribution (figure 1).