ESTRO 2022 - Abstract Book

S337

Abstract book

ESTRO 2022

Conclusion Replacing manual with automated CTV and OAR delineation could lead to missing the required re-planning and inadequate estimation of individual target and OAR doses. Deep learning CT segmentation and registration could be promising to improve the quality of dose accumulation in a time effective manner.

PD-0398 Accumulated dose comparison of stereotactic MRgRT and proton therapy for central lung tumors

M. Rabe 1 , M.A. Palacios 2 , J. van Sörnsen de Koste 2 , C. Eze 1 , M. Hillbrand 3 , C. Belka 1,4 , G. Landry 1 , S. Senan 2 , C. Kurz 1

1 University Hospital, LMU Munich, Department of Radiation Oncology, Munich, Germany; 2 Amsterdam University Medical Centers, location VUmc, Department of Radiation Oncology, Amsterdam, The Netherlands; 3 Kantonsspital Graubünden, Institut für Radio-Onkologie, Chur, Switzerland; 4 German Cancer Consortium (DKTK), Partner Site Munich, Munich, Germany Purpose or Objective Stereotactic MR-guided adaptive photon radiotherapy (MRgRT) and hypofractionated intensity-modulated proton therapy (IMPT) are techniques to reduce the potential toxicity of radiotherapy for central lung tumors. This study compared MRgRT with state-of-the-art robustly optimized IMPT for central lung tumors. We used MRgRT fraction images to calculate accumulated doses and assess the dosimetric effects of interfractional anatomical changes. Materials and Methods Datasets of ten central lung tumor patients, treated with MRgRT to a dose of 8 × 7.5 Gy prescribed to the PTV (GTV + 5 mm) at a ViewRay MRIdian 0.35 T MR-Linac, were included in this study. The setup 3D-MRIs acquired at the beginning of each treatment fraction were deformably registered to the baseline 3D-MRI (pMRI) acquired at the treatment planning stage. The delivered daily adapted fraction doses were deformed to the pMRI to accumulate the dose. For IMPT, treatment plans with 8 × 7.5 Gy(RBE) were created on the baseline virtual CT (planning CT deformably registered to the pMRI) used for initial MRgRT treatment planning. Assuming gated proton beam delivery, an ITV with muscle tissue density override was created by isotropically expanding the GTV by 3 mm to account for residual motion within the gating window. The IMPT plans were robustly optimized with 3% range and isotropic 6 mm position uncertainty to account for interfractional changes and residual uncertainties as per published protocols. The IMPT plans were recalculated on all eight fraction virtual CTs using a routine approach, i.e., without online plan adaptation, which is not yet standard practice in proton therapy. The resulting dose distributions were mapped to the pMRIs to accumulate the dose. A DVH parameter analysis was conducted for the GTV, the ipsilateral lung, and organs at risk (OARs) located within 2 cm of the PTV that was clinically used for treatment planning. Only OARs with at least five patient observations fulfilling this proximity criterion were included in the analysis. Results Exemplary MRgRT and IMPT accumulated dose distributions are depicted in Figure 1. The mean DVH parameters and their differences averaged over all patients are summarized in Table 1. The accumulated GTV D 98% was above the prescribed dose