ESTRO 38 Abstract book

S324 ESTRO 38

each segmented structure were compared between cases and controls using two-way ANOVA tests with significance level 0.05.

Conclusion The brainstem morbidities manifested in the ten cases could not be explained by the investigated dose metrics, including the specific brainstem substructures. The variations in radio-sensitivity may therefore be due to other patient-specific factors not controlled for in this study, or may be better explained by other metrics including variable RBE doses. OC-0613 Spatial dose patterns of radiation pneumonitis in lung cancer patients treated by photons or protons G. Palma 1 , S. Monti 2 , T. Xu 3 , E. Scifoni 4 , P. Yang 3 , S. Hahn 3 , M. Durante 5 , R. Mohan 6 , Z. Liao 3 , L. Cella 1 1 National Research Council, Institute of Biostructures and Bioimaging, Napoli, Italy ; 2 Irccs SDN, Image Processing Department, Napoli, Italy ; 3 MD Anderson Cancer Center, Department of Radiation Oncology, Houston, USA ; 4 Istituto Nazionale di Fisica Nucleare, Trento Institute for Fundamental Physics and Applications, Trento, Italy ; 5 GSI Helmholtz Centre for Heavy Ion Research, Department of Biophysics, Darmstadt, Germany ; 6 MD Anderson Cancer Center, Department of Radiation Physics, Houston, USA Purpose or Objective Radiation pneumonitis (RP) is a common side effect of thoracic Radiation Therapy (RT), and its incidence has been reported similarly in both Intensity-Modulated RT (IMRT) and Passive Scattering Proton Therapy (PSPT) [Liao et al. , JCO 2018]. Aim of this study is to investigate the spatial pattern of pulmonary and cardiac radiosensitivity (RS) to RP in patients enrolled in a prospective randomized trial of IMRT versus PSPT for locally advanced Non-Small-Cell Lung Cancer (NSCLC). Material and Methods We analyzed data of 178 prospectively treated at a single institution with PSPT (64 patients) or IMRT (114 patients) for NSCLC. All patients were treated to a prescribed dose of 60 to 74 Gy in conventional daily fractionation with concurrent chemo-radiotherapy. Median patient age was 66 years (range: 33–85 years). Fifty-five patients (31%) in the trial developed RP of any grade scored according to CTCAE v. 3.0. Each planning CT and dose map was spatially normalized to a common anatomical reference using a B-spline inter- patient registration algorithm after masking the gross tumor volume. A Voxel-Based Analysis (VBA) was performed according to a non-parametric permutation test accounting for multiple comparison, based on a cluster analysis method [Monti et al. , SciRep 2018]. The underlying general linear model of RP was designed to include dose maps and potential non-dosimetric variables significantly correlated with RP. A 3D significance map was derived and the clusters of voxels with dose differences that significantly correlated with RP at a p -level of 0.05 ( S 0.05 ) were generated accordingly. The VBA was performed on the Entire Cohort (EC) and, separately, on the sub-cohorts of IMRT and PSPT. In order to quantify the match between the significance p -maps obtained on a sub-cohort ( p IMRT or p PSPT ) and the p -map obtained on EC ( p EC ), the metrics DIV was computed as the Dice Index between the sublevel sets of p IMRT or p PSPT and p EC with same relative volume V [Monti et al. , SciRep 2018]. Results The VBA on EC highlighted extended areas of significant dose differences between patients with and without RP in the lower part of the right lung and in the heart (Figure).

Results Average D0.1cc, D10% and D50% of each brainstem substructure were not statistically significant different between cases and controls (Figure 2). Significant difference was also not found between the dose/volume parameters for the full brainstem. For both cases and controls, the medulla oblongata and anterior pons had the largest variance in all dose/volume parameters, whereas the variance of D50% was also high for the posterior pons.