ESTRO 2023 - Abstract Book

S1075

Digital Posters

ESTRO 2023

Clinical T stage was significantly associated with PFS and the incidence of recurrence after SBRT for lung cancer. There were differences in the failure patterns according to the clinical T stage. Pleural dissemination was a common failure pattern in the T3 group. In addition, 4D-CT might provide significant information for assessing chest wall invasion associated with unfavourable PFS.

Poster (Digital): Upper GI

PO-1342 3-D RT dose prediction in patients with liver and pancreas cancer treated with the Viewray™ system

D. Kim 1 , B. Jang 2 , E. Kim 3 , J.S. Kim 4 , E.K. Chie 5

1 Kyung Hee University Hospital at Gangdong, Radiation Oncology , Seoul, Korea Republic of; 2 Seoul National University Hospital , Radiation Oncology , Seoul, Korea Republic of; 3 Korea Institute of Radiological and Medical Sciences, Radiation Oncology, Seoul, Korea Republic of; 4 Soonchunhyang University Seoul Hospital, Radiation Oncology, Seoul, Korea Republic of; 5 Seoul National University Hospital , Radiation Oncology , Seoul , Korea Republic of Purpose or Objective The radiotherapy (RT) planning process for patients with gastrointestinal tract cancers is regarded as a time-consuming and complicated task due to proximity of multiple organs at risk (OARs). Dose prediction model based on deep learning might be a solution to increase efficiency by reducing the time required to find the optimal plan. Dose distribution model was developed using a three-dimensional (3D) U-Net deep learning. Materials and Methods Altogether RT plan of patients with liver cancer (N=40) and pancreatic cancer (N=25) were acquired from the MRIdian® and TrueBeam® treatment planning system from 2016 to 2018. Magnetic resonance (MR) images for RT, planning dose, and contoured RT structures were retrospectively collected. The 3D U-Net was employed, and transfer-learning was performed on the training set (N=60, 91.3%). The initial learning rate was 0.1, decreasing by 10% every 5 epochs, for a total of 200 epochs. A developed model was evaluated by the testing set (N=5, 7.7%). Prediction values were derived by training the prediction models of Viewray® and TrueBeam® with the above training hyperparameters, and we used 5-fold cross validation. The difference between ground truth and predicted dose was compared by calculating normalized (range, 0-1) minimum, maximum, and mean doses of the planning target volume (PTV) and OARs. Results Root-mean-square error (RMSE) of voxels between ground truth and predictive dose was 3.04 (range 2.29-3.79). The difference of normalized dose of the Viewray® prediction model was smaller than that of the TrueBeam®. Maximum/mean dose difference of PTV were 0.06/0.04 for Viewray® and 0.18/0.11 for TrueBeam®, respectively. Statistical significance was found in maximum dose difference of PTV, duodenum, and liver & mean dose difference of PTV and liver. Overall, the predicted normalized DVH metrics showed better results in prediction models of Viewray®. Conclusion We developed a deep learning model for dose prediction in liver and pancreatic cancer using MR images acquired from Viewray® and TrueBeam® system. Larger-scale training and validation are required to prove the efficacy of the dose prediction model by the 3D U-Net. A.L. Rivero Pérez 1 , F. Pinto Guevara 2 , J. Albendea Roch 3 , M. Gutierrez Ruiz 3 , R. Astudillo Olalla 2 , M. Garcia Lamela 3 , J. Cardenal Carro 3 , A. Garcia Blanco 3 , M. Ferri Molina 3 , J.T. Anchuelo Latorre 3 , R. Fabregat Borras 3 , P.A. Navarrete Solano 3 , J.A. Blanco Rodriguez 2 , M.P. Galdos Barroso 3 , P.J. Prada Gomez 3 1 humv , Oncology Radiotherapy, Santander, Spain; 2 humv, Oncology Radiotherapy, Santander, Spain; 3 humv, Oncology Radiotherapy, Santander, Spain Purpose or Objective To evaluate toxicity and oncological results of intraoperative electron radiotherapy (IOERT) in the treatment of patients with pancreatic cancer in a tertiary hospital. Materials and Methods We evaluated 24 patients with pancreatic cancer treated with IOERT between March 2019 and December 2021 using a mobile linear accelerator in a specially prepared operating room. All cases have been examined in a multidisciplinary oncological committee. Results 11 women and 13 men with a median age of 70 years (43-85) with adenocarcinoma of the pancreas were operated. 18 patients were considered resectable and 6 marginally resectable, of which 5 received neoadjuvant treatment with chemotherapy and one with radio-chemotherapy. The IOERT dose was 15 Gy in all patients. Hyaluronic acid was used between 4-12 ml to homogenize the surface in 12 treatments. 12 patients had no complications. One patient died two days after due to Covid pneumonia and septic shock. 3 patients suffered minor bleeding, 2 pancreatic fistula, 7 abdominal collections and 2 intestinal ischemia. After 2 years follow-up, 50% of patients have relapsed (16,6% locoregional, 20,8% distant and 13% mixed recurrence). Time to occurrence was 8 months after treatment. 37.5% of patients died after a median of 12 months after treatment (56% due to systemic progression). PO-1343 Intraoperative electron radiotherapy in resected pancreatic cancer