ESTRO 2025 - Abstract Book

S2804

Physics - Dose prediction, optimisation and applications of photon and electron planning

ESTRO 2025

2078

Proffered Paper Pre-simulation of geometric uncertainties instead of margins for cranial radiotherapy allows an integrated boost while reducing irradiated volumes Dan Welsh 1 , Joseph Wood 1 , Eliana Vasquez Osorio 2 1 Christie Medical Physics and Engineering, The Christie NHS Foundation Trust, Manchester, United Kingdom. 2 Radiotherapy Related Research, The University of Manchester, Manchester, United Kingdom Purpose/Objective: This work investigates Monte Carlo simulations of geometric uncertainties prior to plan creation (“PreSim”) as an alternative to margin expansions for handling geometric uncertainty in cranial radiotherapy. This method allows probabilistic information to be incorporated into conventional plan optimisation; boosting a probabilistically derived target within the CTV was also investigated. Material/Methods: Thirteen glioblastoma patients where the PTV overlapped at least one critical OAR PRV were selected. The clinical margin-based plans used a 4mm PTV margin and were prescribed 60Gy/30#. All plans and simulations were produced in Raystation 11B. Spatial probability maps for CTVs and OARs were produced via simulation of 500 treatment courses (random and systematic errors sampled for each fraction/course). Iso-probability thresholds were used to generate PreSim alternative psPTV and psPRV volumes (90% confidence level for comparison to margin PTV) and each case replanned (PreSim_Standard plans). A boost target was derived from the PreSim maps, (GTV within the CTV 95% probability region, avoiding the 0.1% OAR probability volume). PreSim_Boost plans with a simultaneous integrated boost of 72Gy to this volume (60Gy/30# to CTV remainder) were produced. The performance of all plans (Margin, PreSim_Standard and PreSim_Boost) under uncertainty was assessed using the previously described probabilistic evaluator [1], and the 90 th centile of achieved clinical goals compared. The TCP (progression-free survival at 1 year) of the conventional and boost plans was estimated using published empirical data [2], [3]. Results: All plans achieved equivalent CTV coverage (D99% and D95%) and met OAR constraints to 90% confidence. The psPTVs were significantly smaller than margin PTVs (median volume ratio 0.88, range 0.85-0.89, p<0.001); mean brain dose and irradiated volumes (V15Gy, V30Gy, V57Gy) were reduced in PreSim_Standard compared with Margin plans. Figure 1 shows the relative change in OAR and normal tissue dose statistics. The PreSim Boost plans resulted in further significant reductions in irradiated volumes and mean brain doses. Figure 2 shows the calculated TCP for standard and boost plans.