ESTRO 2024 - Abstract Book

S2257

Clinical - Upper GI

ESTRO 2024

Department of Radiation Oncology (Maastro), GROW School for Oncology and Reproduction, Maastricht University Medical Centre+, Maastricht, Netherlands

Purpose/Objective:

About 50% of esophageal cancer patients treated with intensity modulated proton therapy (IMPT) in our institution require plan adaptation based on two criteria, i.e., target underdosage or increase in mean heart dose (MHD), evaluated on weekly repeated CTs. Due to the proximity of the target volume to the diaphragm and the sensitivity of the protons to path length differences, changes in diaphragm position and amplitude are significant predictors of plan adaptation [1-3]. The current treatment plan strategy consists of three-posterior coplanar beams (Figure 1A) with one posterior beam, and additional posterior oblique beams on left and right which partially go through the diaphragm. In a previous study, we contoured the lungs on the two extreme phases of the planning 4DCT and the difference of both contours resulted in the diaphragm amplitude. We concluded that contouring and overriding this structure to water material in the planning CT, in general, did not produce more robust plans. In this study, we evaluated if a three-posterior non-coplanar beam arrangement, with two beams that do not cross the diaphragm and one posterior oblique beam, could be more robust against changes in diaphragm position (Figure 1B). A total of fourteen patients who received a minimum of two repeat CTs (reCTs) (range 2 to 6 reCTs per patient) and who had to be adapted during the course of treatment according to the current clinical criteria were included in this study. For all patients, CTVs were defined on all phases of a 4DCT, and subsequently combined to an ITV. Plans were robustly optimized using a 3 mm isotropic margin around the ITV as well as 5 mm residual setup and 3% range uncertainty. Seven patients received a plan adaptation because of MHD increase (more than 1.5Gy on the reCT), 6 patients because of insufficient target coverage after robust evaluation on the reCT (ITV D98% lower than 94% of the prescription dose in voxel-wise minimum) and 1 patient due to both criteria. Two plans per patient were made: the clinical (3Bclin) and the new beam arrangement (3Bnew). A two-tailed paired sample t-test was performed to compare the differences in organ at risk (OAR). To assess which beam arrangement was more robust over time we recomputed the dose and performed a robust plan evaluation on the reCTs accounting for 2 mm residual setup uncertainty and 3% range uncertainty. Material/Methods: