Xiaolong Gu1, Xianbo Yu2, Gaofeng Shi3, Yang Li1, Li Yang1. 1. Department of Radiology, The Fourth Hospital of Hebei Medical University, Shijiazhuang, Hebei, 050011, People's Republic of China. 2. CT Collaboration, Siemens Healthineers Ltd., Beijing, People's Republic of China. 3. Department of Radiology, The Fourth Hospital of Hebei Medical University, Shijiazhuang, Hebei, 050011, People's Republic of China. uni_uni@163.com.
Abstract
BACKGROUND: This study aimed to construct a computed tomography (CT) radiomics model to predict programmed cell death-ligand 1 (PD-L1) expression in gastric adenocarcinoma patients using radiomics features. METHODS: A total of 169 patients with gastric adenocarcinoma were studied retrospectively and randomly divided into training and testing datasets. The clinical data of the patients were recorded. Radiomics features were extracted to construct a radiomics model. The random forest-based Boruta algorithm was used to screen the features of the training dataset. A receiver operating characteristic (ROC) curve was used to evaluate the predictive performance of the model. RESULTS: Four radiomics features were selected to construct a radiomics model. The radiomics signature showed good efficacy in predicting PD-L1 expression, with an area under the receiver operating characteristic curve (AUC) of 0.786 (p < 0.001), a sensitivity of 0.681, and a specificity of 0.826. The radiomics model achieved the greatest areas under the curve (AUCs) in the training dataset (AUC = 0.786) and testing dataset (AUC = 0.774). The calibration curves of the radiomics model showed great calibration performances outcomes in the training dataset and testing dataset. The net clinical benefit for the radiomics model was high. CONCLUSION: CT radiomics has important value in predicting the expression of PD-L1 in patients with gastric adenocarcinoma.
BACKGROUND: This study aimed to construct a computed tomography (CT) radiomics model to predict programmed cell death-ligand 1 (PD-L1) expression in gastric adenocarcinoma patients using radiomics features. METHODS: A total of 169 patients with gastric adenocarcinoma were studied retrospectively and randomly divided into training and testing datasets. The clinical data of the patients were recorded. Radiomics features were extracted to construct a radiomics model. The random forest-based Boruta algorithm was used to screen the features of the training dataset. A receiver operating characteristic (ROC) curve was used to evaluate the predictive performance of the model. RESULTS: Four radiomics features were selected to construct a radiomics model. The radiomics signature showed good efficacy in predicting PD-L1 expression, with an area under the receiver operating characteristic curve (AUC) of 0.786 (p < 0.001), a sensitivity of 0.681, and a specificity of 0.826. The radiomics model achieved the greatest areas under the curve (AUCs) in the training dataset (AUC = 0.786) and testing dataset (AUC = 0.774). The calibration curves of the radiomics model showed great calibration performances outcomes in the training dataset and testing dataset. The net clinical benefit for the radiomics model was high. CONCLUSION: CT radiomics has important value in predicting the expression of PD-L1 in patients with gastric adenocarcinoma.
Authors: Ronan J Kelly; Jeeyun Lee; Yung-Jue Bang; Khaldoun Almhanna; Mariela Blum-Murphy; Daniel V T Catenacci; Hyun Cheol Chung; Zev A Wainberg; Michael K Gibson; Keun-Wook Lee; Johanna C Bendell; Crystal S Denlinger; Cheng Ean Chee; Takeshi Omori; Rom Leidner; Heinz-Josef Lenz; Yee Chao; Marlon C Rebelatto; Philip Z Brohawn; Peng He; Jennifer McDevitt; Siddharth Sheth; Judson M Englert; Geoffrey Y Ku Journal: Clin Cancer Res Date: 2019-11-01 Impact factor: 12.531
Authors: Katrin Aslan; Verena Turco; Jens Blobner; Jana K Sonner; Anna Rita Liuzzi; Nicolás Gonzalo Núñez; Donatella De Feo; Philipp Kickingereder; Manuel Fischer; Ed Green; Ahmed Sadik; Mirco Friedrich; Khwab Sanghvi; Michael Kilian; Frederik Cichon; Lara Wolf; Kristine Jähne; Anna von Landenberg; Lukas Bunse; Felix Sahm; Daniel Schrimpf; Jochen Meyer; Allen Alexander; Gianluca Brugnara; Ralph Röth; Kira Pfleiderer; Beate Niesler; Andreas von Deimling; Christiane Opitz; Michael O Breckwoldt; Sabine Heiland; Martin Bendszus; Wolfgang Wick; Burkhard Becher; Michael Platten Journal: Nat Commun Date: 2020-02-18 Impact factor: 14.919