Assessment Of Neoadjuvant Chemotherapy Efficacy In Progressive Gastric Cancer By CT And The Study Of Imaging Histology

Objective: To evaluate the role of volume in the efficacy of neoadjuvant chemotherapy for gastric cancer by three-dimensional image analysis of multilayer spiral CT(MSCT),to compare the diagnostic efficacy of volume with the subjective model of common CT indices.Methods: Sixty-one patients with gastric cancer diagnosed by gastroscopy and pathological examination from January 2019 to March 2022 were divided into chemotherapy effective group(n=39)and ineffective group(n=22)according to postoperative pathological regression grading criteria(TRG),underwent pre-and preoperative MSCT examination before neoadjuvant chemotherapy,and were analyzed by binary logistic regression analysis combined with CT commonly used observational indexes(including tumor maximum thickness,gastric wall motility,enhancement pattern,lymph node metastasis,distant metastasis,peritoneal thickening or peritoneal nodules)to construct a predictive model of visual signs.The value of preoperative scan phase texture was evaluated with the imaging software Mazda,and the value of arterial phase and venous phase in predicting the efficacy of neoadjuvant chemotherapy was compared,and the more valuable phase was used to measure the tumor volume and thickness with the imaging histology software ITK-snap.In order to understand the relationship between observed indexes and postoperative pathological TRG,the relationship between the change of each index and postoperative pathological TRG was analyzed by using Spearman correlation test,and the ROC curve was drawn to analyze the diagnostic efficacy of volume and CT visual model.Results: The textural efficacy was better in the intravenous phase with a minimum false positive rate of 8.2%,which was more valuable in predicting the efficacy.In the evaluation of chemotherapy efficacy,the rate of volume change of cancer foci could evaluate the efficacy after chemotherapy to a certain extent(p <0.05).Further analysis revealed that the volume change rate was more negatively correlated with TRG grading(r=-0.9196);for the univariate analysis of subjective tumor CT signs,the indicators associated with the effectiveness of chemotherapy were tumor thickness change rate,gastric wall peristalsis,lesion intensification mode,and peritoneal thickening and nodules(p<0.05).For univariate analysis of subjective tumor CT signs,significant findings(tumor thickness change rate,gastric wall peristalsis,lesion intensification mode,and peritoneal thickening and nodules)were included in binary logistic regression to screen for independent factors: gastric wall peristalsis(OR=0.294,95% CI=0.093 to 0.928),thickness change rate(OR=0.282,95% CI=0.083 to 0.957),and the new independent variables derived from logistic regression analysis were included in the ROC curve,and the tumor volume change rate AUC(0.998),sensitivity(100%),and specificity(95.5%)were higher than the CT visual sign model AUC(0.822),sensitivity(75.0%),and specificity(85.7%).Conclusion:MSCT combined with the imaging histology software Ma Zda and ITK-snap to measure tumor volume provides an objective basis for predicting the efficacy of neoadjuvant chemotherapy,and its diagnostic efficacy is better than that of the CT visual sign model.