Precise Prediction Of Pathological Staging And HER2 Gene Expression By CT Radiomics In Advanced Gastric Cancer

Prediction of the Depth of Tumor Invasion in Gastric Cancer:Potential Role of CT RadiomicsObjective:Accurate evaluation of the T staging---the depth of tumor invasion may facilitate the selection of the optimal treatment strategy,individualized therapy and improvement of the patient prognosis.T staging the one of the important criteria for selecting appropriate patients for neoadjuvant chemotherapy(NAC).The aim of this study was to investigate the value of computed tomography(CT)radiomics for the differentiation between T2 and T3/4 stage lesions in gastric cancer.Materials and methods:A total of 244 consecutive patients with pathologically proven gastric cancer who received radical surgery at Peking Union Medical College Hospital were retrospectively included and split into a training cohort(171 patients)and a test cohort(73 patients).Preoperative arterial phase(AP)and portal phase(PP)contrast enhanced CT images were retrieved for tumor segmentation and feature extraction by using a dedicated postprocessing software.The random forest(RF)method was used to build the classifier models.The performances of the radiomics models in both the training and test cohorts were evaluated with receiver operating characteristics(ROC)curvesResults:AP-based radiomics model exhibited area under curve(AUC)of 0.899(95%CI:0.812 to 0.955)and accuracy of 84.1%for differentiation between stage T2 and T3/4 cases in the training cohort,the AUC was 0.825(95%CI:0.718 to 0.904)and accuracy was 75.3%in the test cohort.PP-based radiomics model showed AUCs of 0.843(95%CI:0,746 to 0.914)and 0.818(95%CI:0.711 to 0.899)in the training and test cohort for differentiation between stage T2 and T3/4 lesions,respectively.The accuracy of the PP-based radiomics model were 78.0%and 61.6%,respectively?Conclusion:Radiomics models based on the standard-of-care CT images has a potential role in differentiation between T2 and T3/4 stage tumors in gastric cancer.CT Radiomics Nomogram for the Preoperative Prediction of Lymph Node Metastasis in Gastric CancerPurpose:Lymph node(LN)metastasis is not only one of prognostic factors of gastric cancer patients.It also matters much when selecting the appropriate candidates to receive neoadjuvant chemotherapy(NAC).The aim of this retrospective analysis was to investigate the role of computed tomography(CT)radiomics for the preoperative prediction of LN metastasis in gastric cancer.Materials and methods:This retrospective study included 247 consecutive patients(training cohort:197 patients;test cohort:50 patients)with pathologically proven gastric cancer.A dedicated radiomics prototype software was used to segment lesions on preoperative arterial phase(AP)CT images and extract features.A radiomics model was constructed to predict the LN metastasis by using a random forest(RF)algorithm.Multivariate logistic analysis was used to evaluate all the clinical variables and radiomics scores to select the independent predictors of LN metastasis.Finally,a nomogram was built incorporating the radiomics scores and selected clinical predictors.Receiver operating characteristic(ROC)curves were used to validate the capability of the radiomics model and nomogram on both the training and test cohorts.Results:The radiomics model showed a favorable discriminatory ability for predicting LN metastasis.The area under the curve(AUC)in the training cohort could be up to 0.844(95%CI:0.759 to 0.909),which was confirmed in the test cohort with an AUC of 0.837(95%CI:0.705 to 0.926).The accuracy of the radiomics model for prediction of the LN metastasis was 80%in the training cohort,and 84%in the test cohort.Multivariate logistics analysis showed that only CT-reported LN status as well as the radiomics scores are independent predictors.The nomogram incorporating the radiomics scores and the CT-reported LN status showed excellent discrimination in the training and test cohorts with AUCs of 0.886(95%CI:0.808 to 0.941)and 0.881(95%CI:0.759 to 0.956),respectively.The accuracy of the radiomics nomogram was 83%in the training cohort and 84%in the test cohort.Conclusions:The radiomics nomogram based on the standard-of-care CT images holds promise for use as a noninvasive tool in the individual prediction of LN metastasis in gastric cancer.CT Radiomics for the Distinction of Human Epidermal Growth Factor Receptor 2-Negative Gastric CancerObjective:Evaluation of the human epidermal growth factor 2(HER2)status in gastric cancer is important for timely target chemotherapy to improve prognosis.The purpose of this study was to develop and validate the computed tomography(CT)based radiomics model for the prediction of HER2 status in patients with gastric cancer.Materials and methods:One hundred and thirty-two consecutive patients with advanced gastric cancer undergoing radical gastrectomy were retrospectively reviewed.All patients received preoperative contrast CT examination,and immunohistochemistry(IHC)results of their HER2 status were available.All the subjects were randomly divided into a training cohort(n=90)and a validation cohort(n=42).Arterial phase(AP)and portal phase(PP)contrast CT images were retrieved for tumor segmentation and feature extraction.One of the popular machine learning method—random forest(RF)algorithm was used to build the radiomics model.The performances of the radiomics classifiers were evaluated with receiver operating characteristics(ROC)curvesResults:Among the 132 patients,a total of 99 patients were HER2 negative(ICH=0 or 1+),and the remaining 33 patients were border line or positive(ICH=2+or 3+).The AP radiomics model could distinguish HER2-negative cases with an AUC(area under curve)of 0.756(95%CI:0.656-0.840)in the training cohort,which was confirmed in the test cohort with AUC of 0.830(95%CI:0.678-0.930).The accuracy in the training and test cohort was 78.2%and 77.5%,respectively?The PP radiomics model showed AUCs of 0.715(95%CI:0.612-0.804)and 0.718(95%CI:0.554-0.849)in the training and test cohort for distinction of negative HER2 cases,respectively.The accuracy of the PP-based radiomics model for prediction of the HER2-negative cases was 78.3%in the training cohort and 77.5%in the test cohort.Conclusions:Radiomics models based on standard-of-care CT images hold promise for distinguishing HER2-negative gastric cancer.