ESTRO 2023 - Abstract Book

S1825

Digital Posters

ESTRO 2023

We observed CWP in 28% of our lung SBRT series. We did not find any statistical difference in ribs’ Vx and Dx parameters between the non-CWP and the CWP groups. Near-maximum ribs doses showed the highest correlation with CWP appearance. Larger datasets might help predict radiation-induced CWP better.

PO-2051 Set-up errors evaluation of CSI using longitudinally extended CBCT & its comparison with normal CBCT

P. Agarwal 1 , U. Mukherjee 2 , K. Kaushik 2 , T. Shrivastava 2 , J. Verma 2 , D. Nandi 2 , N. Sharma 2 , S. Pradhan 2

1 Homi Bhabha Cancer Hospital (Tata Memorial Center), Department of Radiation Oncology, Varanasi, India; 2 Homi Bhabha Cancer Hospital, Department of Radiation Oncology, Varanasi, India Purpose or Objective Craniospinal irradiation (CSI) treatment is a complex radiotherapy technique due to its junctions, proximities of OARs and extensive PTV length. For the accuracy of treatment, it is necessary to carry out daily IGRT. However, in conventional linear accelerator, the IGRT length is a constraints for extensive PTV length, which provides the uncertainty in the evaluation of setup error. Now, it is feasible in Truebeam (Varian Medical System, ver 2.7) to carry out the image guidance with longitudinally extended CBCT & match with patient DRR. The purpose of this is to evaluate the setup error of CSI using longitudinally extended CBCT & its comparison with normal length CBCT. Materials and Methods A total 12 CSI patients treated in 2021-2022 were reviewed for this study, retrospectively. The CT simulation was carried out in the supine position with NNR, AIO baseplate and 4 clamp orfit. The CTV, PTV & OARs were segmented and CTV to PTV margin was applied 5mm in the brain region and 8mm in the spinal region isotropically according to the institutional protocol. The mean (± SD) PTV volume for brain and spine was 1680cc (± 305cc) & 642cc, respectively. The patients were planned on Eclipse TPS (ver 15.5) with 3D-CRT treatment modality. The total dose prescription was 36Gy in 20 fractions. As per the institutional protocol for pre-treatment setup verification on Truebeam, the daily kV-CBCT was acquired till full length of PTV from the head (Fig.1). The matching was carried out on the axial images guided by sagittal, coronal images. Bony anatomy based matching was carried out with longitudinally extended kV-CBCT (E-CBCT) to document the anterior- posterior(AP), cranio-caudal(CC), & medio-lateral(ML) shifts. Similarly, the same patients shifts (no extra images was carried out) were carried out for normal length kV-CBCT (N-CBCT) through offline review. The CTV to PTV margin was calculated using systematic & random error by Ven-Herk margin recipe for both E-CBCT & N-CBCT. Results A total 240 kV-CBCT images for E-CBCT and 240 for N-CBCT were analyzed. The systematic error for E-CBCT were 1.26mm, 1.45mm and 1.04mm in AP, CC and ML plane, respectively. The same for N-CBCT were 1.08mm, 1.01mm and 0.97mm in AP, CC and ML plane, respectively. The random error for E-CBCT in AP, CC and ML plane were 0.54mm, 0.87mm and 0.63mm, respectively. The same for N-CBCT were 0.56mm, 0.62mm and 0.38mm, respectively. The CTV to PTV margin was calculated for E-CBCT 3.4mm, 4.3mm an 3.1mm in AP, CC and ML plane, respectively. The same for N-CBCT were 3.1mm, 3.0mm and 2.7mm in AP, CC and ML plane, respectively (Fig 2). The couch shifts and CTV to PTV margins were larger in all three directions for E-CBCT compare to N-CBCT. However, all the margins were within the actual applied margins in both the cases. Conclusion Our study suggests that E-CBCT margins is the superior choice to avoid the uncertainty in treatment of CSI in any direction by matching the entire PTV at a