ESTRO 2024 - Abstract Book

S4358

Physics - Intra-fraction motion management and real-time adaptive radiotherapy

ESTRO 2024

1 Department of Oncology, Copenhagen University Hospital - Herlev and Gentofte, Copenhagen, Denmark. 2 Department of Clinical Medicine, University of Copenhagen, Copenhagen, Denmark

Purpose/Objective:

Cone Beam Computed Tomography (CBCT) based online adaptive radiotherapy (oART) has been clinically implemented for several anatomical sites worldwide. However, for patients with lung cancer, random and abrupt anatomical changes with large density alterations can occur, affecting the accuracy of the density map in the synthetic computed tomography (sCT) generated and thus the dose calculation. The anatomical changes causing density alterations could be the resolution/appearance of atelectasis, tumor shrinkage, tumor position deviation, pleural effusion, pneumonia, etc. This study aimed to quantitatively investigate the uncertainty in dose calculation on sCTs for a CBCT based oART-workflow, and the possibility of using direct dose calculation on CBCT for patients with lung cancer with a broad range of thoracic anatomical changes. Also, qualitative observations of the generated sCT images are reported.

Material/Methods:

A retrospective study of 20 lung cancer patients with different anatomical changes that resulted in re-planning during their course of treatment were included (treated between 15 January 2018 to 15 December 2022). An oART workflow was simulated for each patient in an emulator, running in batch mode. The original planning CT (pCT) was deformed to the re-planning CT (rCT) and the first CBCT post re-planning (representing the anatomy-of-the-day). This generated the sCTrCT and sCTCBCT, respectively. The rCT treatment plan was re-calculated on the sCTrCT, sCTCBCT, and directly on the CBCT (dCBCT) for each patient. Resulting dose distributions were compared with the rCT (considered as the ground truth) using relevant dose volume histogram (DVH) parameters, where the dosimetric agreement was evaluated using Wilcoxon signed rank.

Results:

Among the 20 sCTrCT and 20 sCTCBCT, 12 and 14, respectively, displayed anatomical disagreement with the rCT, attributed anatomical changes that could not be corrected through deformation in the oART workflow. Differences in DVH parameters were found to be within an acceptable range (2.0%) for 13 out of 20 patients for rCT-sCTrCT and for 11 out of 16 for rCT-sCTCBCT (where 4 patients were excluded due to large anatomical disagreement between CBCT and rCT). The rest of the patients resulted in higher (over 2.0%) dose differences for any dosimetry metric, due to incorrect image deformations and erroneously generated sCTs because of anatomical changes. Figure 1 illustrates erroneous deformations due to large density alterations occurring during the course of treatment for three patients. Figure 2 presents the DVH dosimetry metrics’ dose differences for target structures and OARs. DVH metrices from the synthetic CTs (sCTrCT and sCTCBCT) and CBCT were subtracted from the same DVH metrices of the rCT. The Wilcoxon signed rank test found small statistically significant differences for PTV Dmax and body Dmax between rCT sCTrCT, and rCT-sCTCBCT, where DVH differences for rCT-sCTrCT were -0.4 [-7.8; 0.9]% for PTV Dmax and -0.4 [-7.2; 2.0]% for body Dmax. Corresponding results for rCT-sCTCBCT were -1.3 [-9.0; 1.4]% for PTV Dmax and -1.0 [-8.1; 1.4]% for body Dmax. DVH differences for rCT-dCBCT deviated greatly from rCT-sCTrCT, and rCT-sCTCBCT, as seen in Figure 2, where several DVH dosimetry metrics are statistically significant when calculating directly on the CBCT.