ESTRO 2021 Abstract Book

S316

ESTRO 2021

Purpose or Objective Proton therapy (PT) may offer superior conformity but variations in relative biological effectiveness (RBE) with linear energy transfer (LET), especially at the distal Bragg peak region, remain a large source of uncertainty. With the increasing use of intensity-modulated PT, there is growing concern regarding unintended RBE hot- spots in healthy structures due to inhomogeneous LET distributions, which may potentially induce radiation- induced side effects. Although knowledge on biological parameters is limited, LET is a computable physical quantity correlated with RBE. In this study, LET and RBE distributions were retrospectively investigated for a large number of neurological organs-at-risk (OARs) in a cohort of PT patients. Materials and Methods 23 patients with tumors at different locations inside the brain, who received PT between 2019 and 2020 were selected for this study. The dose-averaged LET (LET d ), the product between LET d and absorbed dose (D∙LET d ) and the RBE (i.e. RBE Un for Unkelbach and REB Mc for McNamara models) were computed for 26 OARs delineated according to the European Particle Therapy Network (EPTN) neuro contouring atlas[1]. Calculations were done using the research version of Raystation 9A with the Monte Carlo dose engine commissioned for a Mevion S250i Hyperscan PT system. Results The heterogeneous distribution of tumor positions and arrangement of treatment beams promote large planning dose variation per healthy structure over the patient population, and to a smaller extent LET D variation (Figure 1a,b). Variation was seen in both RBE models, in comparison to the clinically used constant value of 1.1 (straight line, Figure 1c), with McNamara’s model results being invariably higher for all OARs (depicted in pink). Distribution maps (Figure 1d) and individual patient parameters (Figure 1e) for the chiasm (yellow; Figure 1d), exhibit a decrease in LET d with increasing dose throughout the entire population, also expressed in the D∙LET d decrease for patients where the chiasm dose was close to the constraint. This behavior was consistent for other critical organs of small volume, e.g. optical nerve, pituitary, cochlea. The clinical plan optimization strategy allows the use of multiple distal layers beyond the CTV, likely resulting in higher LET d regions being placed beyond critical structures in cases they are abutting the CTV.

Conclusion This analysis promotes the use of biophysical surrogate parameters such as D∙LET d

to contribute to the

treatment planning process. Further validation of the LET d calculation is still necessary as well as evaluating follow-up functional imaging of this cohort to identify if potential radiation-induced lesions could be correlated with their LET d distributions. [1] Eekers DB, In 't Ven L, Roelofs E, Postma A, Alapetite C, Burnet NG, et al. The EPTN consensus-based atlas for CT- and MR-based contouring in neuro-oncology. Radiother Oncol. 2018;128:37-43.