ESTRO 2025 - Abstract Book

S2776

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

ESTRO 2025

1640

Digital Poster Fully automated megavoltage photon treatment planning for skin cancer with comparison to electron therapy James L Robar Medical Physics, Nova Scotia Health, Halifax, Canada Purpose/Objective: Electron therapy has been a mainstay of treatment for skin cancer, however newer O-ring treatment platforms do not offer electron beams. Megavoltage photon treatment is possible [1] for these superficial targets but requires specific strategy in defining beam/arc angles, tuning structures and optimization objectives. In this work we present and evaluate a novel technology for automated treatment planning for skin cancer with methodology dedicated to this clinical application. Material/Methods: A software application (app, Figure 1) was developed for operation within the Eclipse treatment planning system (Varian Medical). Design goals included i) fully automated planning, i.e., in a single button press; and ii) planning completion within two minutes. The app automates creation of planning target volumes, generation of tuning structures to optimize dose, bolus design, optimization of beam and arc angles based on prediction of normal tissue dose, IMRT or VMAT optimization, and forward dose calculation. Plan quality was assessed using a sample of common skin sites including scalp vertex, forehead, nose, cheek, ear, and upper leg. Target volumes ranged from 2.3 to 20.3 cc in volume. Plans were completed automatically using the application for Halcyon 4 platform, as well as manually for electron beams on the TrueBeam platform (Varian Medical). Plan quality was assessed in terms of PTV dose homogeneity and dose received by normal tissues. For the automated photon plans the time for plan completion was recorded.

Results: Target dose homogeneity was improved compared to electrons, with the PTV maximum dose lower by 21% ±16% and 22% ± 16% for IMRT and VMAT, respectively. Similarly, the normal tissue maximum dose was lower by 14% ± 13% and 15% ± 13% for IMRT and VMAT. The volume of normal tissue receiving 90% dose (V90%D) was lower by 3cc ± 2cc and 4cc ± 3cc and V50%D was lower by 10cc ± 12cc and 9 cc ± 15cc for IMRT and VMAT, respectively. All normal tissue volumes were lower above approximately the 20%D level, below which IMRT and VMAT plans were