Radiomics Analysis Using Contrast-Enhanced CT for Preoperative Prediction of Occult Peritoneal Metastasis in Advanced Gastric Cancer

Abstract

Background: To evaluate the predictive value of CT radiomics features derived from primary lesions in discriminating occult peritoneal metastasis (PM) in advanced gastric cancer (AGC). Methods: Preoperative CT images of 233 patients with AGC were retrospectively analyzed. The region of interest was manually drawn along the margin of the lesion on the largest slice of venous CT images, and a total of 539 quantified features were extracted automatically. The intraclass correlation coefficient (ICC) and the absolute correlation coefficient (ACC) were calculated for selecting influential features. The predictive value of each influential feature was assessed using the Mann-Whitney U test and receiver operating characteristic (ROC) analysis. A multivariate logistic regression model was constructed based on the training cohort (124 without PM, 34 with PM), and the testing cohort (64 without PM, 11 with PM) validated the reliability of the model. Moreover, another model based on the preoperative clinicopathological features was also developed. A comparison between the diagnostic performances of the two models was performed using ROC analysis and the Akaike information criterion (AIC) value. Results: Our results showed that 6 radiomics features (ID_Energy, LoG(0.5)_Energy, Compactness2, Max Diameter, Orientation and Surface Area Density) differed significantly between AGCs with and without PM (all P < 0.05) and performed well in distinguishing AGCs with PM from those without PM in the primary cohort (AUC = 0.618-0.658, all P <0.05). The radiomics model including Compactness2, Max Diameter and Orientation showed a higher AUC value than each single radiomics feature in the primary cohort (0.741 vs 0.618-0.658) and similar diagnosis performance in the validation cohort. The radiomics model showed better diagnostic efficacy and model fit accuracy than the clinicopathological model (AUC, 0.724 vs 0.649; AIC, 157.2 vs 172.1). Conclusion: Preoperative radiomics analysis using venous CT images provided valuable information for predicting occult PM in AGC. Funding Statement: This work was supported by National Natural Science Foundation of China (ID: 81501441, 81601463, 81871410), Foundation of National Health and Family Planning Commission of China (W201306), Social Development Foundation of Jiangsu Province (BE2015605), the Natural Science Foundation of Jiangsu Province (ID: BK20131281, BK20150109), Jiangsu Province Health and Family Planning Commission Youth Scientific Research Project (ID: Q201508), Six Talent Peaks Project of Jiangsu Province (ID: 2015-WSN-079), and Jiangsu Provincial Medical Youth Talent (ID: QNRC2016040) Declaration of Interests: The authors declare that they have no competing interests. Ethics Approval Statement: This retrospective study was approved by the ethics committee of the Institutional Review Board of Nanjing Drum Tower Hospital, and the requirement for informed consent was waived.