Prediction Of Microvascular Invasion And Prognosis Of Intrahepatic Mass-forming Cholangiocarcinoma Based On MRI Features

Objective: To explore the predictive value of magnetic resonance imaging features on microvascular invasion and prognosis of intrahepatic mass-forming cholangiocarcinoma.Methods: A retrospective collection of 70 patients with pathologically confirmed IMCC who were surgically resected in the The First Hospital of China Medical University from December 2014 to March 2022 was retrospectively collected,and the patients were divided into MVI-positive group(37 cases)and MVI-negative group(33 cases)according to the pathological results.Qualitative and quantitative features were observed on Gd-DTPA enhanced MRI: qualitative indicators included tumor size,number of lesions,tumor morphology,tumor location,tumor margin,intrahepatic bile duct dilation,hepatic capsular retraction,cystic change/necrosis,lymphadenectasis,ascites,abnormal perfusion around the tumor,dynamic enhancement,arterial phase enhancement,delayed phase intensification;Quantitative indexes included maximum edge thickness in the arterial phase and the arterial ring enhancement ratio.Univariate and multivariate logistic regression was used to analyze the correlation between each image features and MVI.The Kaplan-Meier(K-M)method was used to plot the curve,and the log-rank test was used for the between-group comparison.The receiver operating characteristic(ROC)curve analysis was carried out based on the prediction probability of the regression equation to evaluate the prediction efficiency.The optimal diagnostic cut-off value was determined according to the nearest approximation index,and the sensitivity and specificity of the area under the curve(AUC)were obtained.The Kaplan-Meier(K-M)method was used to plot the curve,and the log-rank test was used for the between-group comparison.P<0.05 is statistically significant.Result In terms of clinicopathological features,the negative rate of HBs Ag or HCVAb(p=0.017),CA19-9 value(p=0.001)and surgical resection range(p=0.010)were higher in the MVI-positive group than those in the MVI-negative group.In terms of imaging features,the results showed the number of tumors(p=0.030),tumor size(p=0.036),the arterial phase enhancement(p=0.024),and intrahepatic bile duct dilation(p <0.001),tumor margin(p=0.008),and hepatic capsular retraction(p=0.034).There were significant differences between in the MVI-positive group and the MVI-negative group.Multivariate logistic regression showed that hepatic capsular retraction(OR=4.298,95% CI: 1.097-16.835,p=0.036)and intrahepatic bile duct dilation(OR=7.412,95% CI: 2.02-27.198,p=0.003)were independent risk factors for MVI.The sensitivity of the joint model for the diagnosis of MVI was 67.8% and the specificity was 78.8%,the calibration curve showed that the predictive model had good predictive performance,and the decision curve analysis showed that the predictive model had good clinical benefits.The log-rank results showed that the OS of MVI-positive patients was significantly lower than that of MVI-negative patients(p=0.001).Patients with one or two imaging features had a significant reduction in OS compared with those without shadow features(p =0.018).Conclusion: Preoperative magnetic resonance imaging features can help predict the potential risk and prognosis of IMCC microvascular invasion,and can provide more imaging reference indicators for the formulation of individualized treatment plans for IMCC patients.