ESTRO 2024 - Abstract Book

S4602

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

ESTR0 2024

3. Breedveld, Sebastiaan, Pascal RM Storchi, and Ben JM Heijmen. "The equivalence of multi-criteria methods for radiotherapy plan optimization." Physics in Medicine & Biology 54.23 (2009): 7199.

4. J. A. Langendijk et al., “National protocol for model-based selection for proton therapy in head and neck cancer,” International journal of particle therapy, vol. 8, no. 1, pp. 354–365, 2021.

667

Digital Poster

LET d -based proton-therapy optimization strategy for brainstem sparing in neuro-oncological patients

Silvia Molinelli 1 , Alessandro Vai 1 , Giuseppe Magro 1 , Giulia Riva 2 , Stefania Russo 1 , Alessia Bazani 1 , Lucia Ciccone 2 , Alberto Iannalfi 2 , Ester Orlandi 2 , Mario Ciocca 1

1 CNAO, medical physics, pavia, Italy. 2 CNAO, clinical department, pavia, Italy

Purpose/Objective:

The current plan optimization strategy in proton therapy (PT) relies solely on dose-based objectives, with a constant relative biological effectiveness (RBE) of 1.1. Protons linear energy transfer (LET) and RBE increase at the end of their range, with potential deposition of higher than expected RBE-weighted doses (D RBE ) in healthy tissues, distally located with respect to the target volume (1). As the high-LET radiation burden in radiosensitive organs, such as the brainstem, is suggested to be potentially related to increased treatment toxicities (2), multiple dose-averaged LET (LET d ) or variable RBE optimization approaches have been proposed to address this issue (3). In this regard, LET d -based objectives have been recently integrated in the plan optimization module of RayStation (v2023B). This work aims to quantify potential improvements in brainstem sparing by the addition of LET d -based objectives to proton plan optimization, while maintaining robust clinical goals with respect to purely dose-based planning strategies, in intra cranial patients. We selected ten neuro-oncological patients, with the clinical target volume abducting the brainstem, previously treated with dose-based intensity modulated (IMPT) optimized plans, delivered with a gantry-less pencil beam scanning technique. Prescription doses ranged from 54 to 59.4 Gy(RBE) in 1.8 to 2 Gy(RBE) per fraction. Standard dose-based robust (3 mm setup, 3% range uncertainties) optimization used a brainstem near-to-maximum dose constraint of D1% ≤ 54 Gy(RBE). Plans were re-optimized including a LET d -based objective in the cost function: the maximum brainstem LET d of particles depositing a D RBE higher than a fixed dose threshold was minimized in each plan. The optimal patient-specific combination of minimum LET d and dose threshold must guarantee constancy of robust clinical goals, with respect to the clinically delivered plan, and acceptable balance of beam contributions. Brainstem sparing improvement between dose-based and LET d -based optimizations, in terms of dose-volume (the percentage of volume receiving a dose ranging from 30 to 55 Gy(RBE) (V D ) and D 1% of brainstem) and LET d -volume (the median (LET d50% ) and near-to-maximum (LET d1% )) points, were statistically tested with the Wilcoxon signed rank (α = 0.05), over different RBE model scenarios (linear LET-weighted RBE model (LWD) with a k=0.055 µm/keV) (4). Material/Methods: