National Cancer Center/National Clinical Research Center for Cancer Cancer Hospital, CAMS & PUMC Beijing, Beijing
Y. Liu1, Y. Wang2, Q. Pang3, X. Wang1, L. Xue4, H. Zhang2, X. Hu5, Z. Ma1, H. Deng6, Z. Y. Yang7, Y. Men8, F. Ye9, K. Men10, J. Qin11, N. Bi1, J. Zhang12, Q. Wang2, and Z. Hui8; 1Department of Radiation Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China, 2Department of Radiation Oncology, Sichuan Cancer Hospital and Institution, Sichuan Cancer Center, School of Medicine, University of Electronic Science and Technology of China, Radiation Oncology Key Laboratory of Sichuan Province, Chengdu, China, 3Department of Radiation oncology, Tianjin Medical University Cancer Institute and Hospital, National Clinical Research Center for Cancer, Key Laboratory of Cancer Prevention Therapy, Tianjin, Tianjin, China, 4Department of Pathology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China, 5Department of Radiation oncology, Tianjin Medical University Cancer Institute and Hospital, National Clinical Research Center for Cancer, Key Laboratory of Cancer Prevention Therapy, Tianjin, China, 6Department of Diagnostic Radiology, Sichuan Cancer Hospital & Institution, Sichuan Cancer Center, School of Medicine, University of Electronic Science and Technology of China, Chengdu, China, 7Department of Pathology, National Cancer Center/Cancer Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, China, 8Department of VIP Medical Services, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China, 9Department of Diagnostic Radiology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China, 10National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China, 11Department of Thoracic Surgery, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China, 12Laboratory for Biomechanics and Mechanobiology of Ministry of Education, Beijing Advanced Innovation Center for Biomedical Engineering, School of Engineering Medicine, School of Biological Science and Medical Engineering, Beihang University, Beijing, China
Purpose/Objective(s): This study aimed to develop and validate a novel deep learning radiomics model using CT, T2 and DWI images for predicting pathological complete response (pCR) in patients with esophageal squamous cell carcinoma (ESCC) undergoing neoadjuvant chemoradiotherapy (nCRT). Materials/
Methods: Patients with ESCC undergoing nCRT were enrolled retrospectively from Institution 1 and 2, and prospectively from Institution 3, and then divided into training and testing cohorts based on their institution of origin. Both traditional and deep learning radiomics features were extracted from pre-treatment CT, T2, and DWI scans. Feature selection was conducted using the Mann-Whitney U test and LASSO with cross-validation. Multiple radiomics models were developed, both single modality and integrated, using machine learning algorithms. The models performance was assessed using Receiver Operating Characteristic curve analysis, with the area under the curve (AUC) as a primary metric, alongside sensitivity and specificity from cut-off analysis. Results: The study involved 151 patients. The training cohort consisted of 89 patients from Institution 1 (median age 62, 73 males) and the testing cohort included 52 patients from Institution 2 (median age 62, 41 males) and 10 from Institution 3 (median age 69, 9 males). The integrated model, combining traditional and deep learning radiomics features from CT, T2, and DWI scans, demonstrated best performance with an AUC of 0.868 (95%CI 0.766-0.959), sensitivity of 88% (95%CI 73.9-100), and specificity of 78.4% (95%CI 63.6-90.2) in the testing cohort. This model outperformed single-modality models and the clinical model. Conclusion: A multimodality deep learning radiomics model, utilizing CT, T2, and DWI images, was developed and validated for accurately predicting pCR of ESCC following nCRT.
