The Study Of Gd-EOB-DPTA Enhanced MRI Contrast Agent To Predict The Sfficacy Of Targeted Combined Immunotherapy In Unresectable Hepatocellular Carcinoma

Background:Primary hepatocellular carcinoma(HCC)is the fourth most common malignant tumor and the second most lethal cause of cancer in China,including 75% ～ 85% hepatocellular carcinoma.The onset of HCC is occult and the early symptoms are not obvious.More than 70% of patients with HCC cannot receive radical treatment such as surgical excision when first diagnosed.In recent years,molecular targeted drug therapy,immunotherapy and other therapeutic methods have provided a new treatment option for advanced liver cancer,but the overall efficacy is limited.Predicting the efficacy of targeted and immunotherapy for HCC has become a research hotspot.Objective:To explore the relationship between the degree of nodule enhancement in MRI hepatobiliary stage images enhanced by Gadoxetic Acid Disodium(Gd-EOB-DTPA)and the effect of targeted combined immunotherapy on HCC.Correlation imaging omics was used to construct a high efficiency prediction model to predict the efficacy of targeted combined immunotherapy.Method:Data of 53 patients diagnosed with HCC in Henan Provincial People’s Hospital from January 2020 to January 2022 were retrospectively collected.The signal intensity of 107 intrahepatic nodular lesions was measured by Radi Ant DICOM Viewer.The Relative Enhancement Ratio(RER)was calculated and the 20%Time to nodular progression(TTNP)per nodular progression between high and low enhancement HCC nodules was compared.According to the immune Response Evaluation Criteria in Solid Tumors(i RECIST),The Progression-Free Survival(PFS)and Objective response rate(ORR)of patients with high enhancement and low enhancement of HCC nodules in hepatobiliary stage were compared.ORR,Disease Control Rate(DCR),etc.According to i RECIST,images of 15 patients with disease remission and 38 patients without disease remission were used to extract 1130 image omics features from MRI images using 3D Slicer software.After feature screening through variance homogeneity test,independent sample t test,Least absolute shrinkage and selection operator(LASSO)regression analysis,7 features were selected for model construction.The Receiver operating characteristic(ROC)curve was used to verify the model,and the efficacy of the imaging omics characteristics in predicting the treatment effect was evaluated.Results:The high strengthen HCC nodules in patients with median PFS was 3.4(95% confidence interval[Confidence intervals,CI],2.5-4.3)months(n = 21),ORR was 14.3%(3/21),DCR was 47.5%(10/21).The median PFS,ORR,and DCR were 6.8(95%CI: 6.2-7.3)months(n = 32),37.5%(12/32),and 75.0%(24/32)for patients with low enhancement of HCC nodules.The median TTNP of advanced high-enhancement HCC nodules(n = 23)was 2.3(95%CI: 2.0--2.6)months and that of low-enhancement HCC nodules(n = 62)was 6.1(95%CI: 5.0--7.2)months.After subsequent feature screening,seven imaging omics features were used to construct a predictive model of treatment effect.In the test group,the score of random forest classifier was 0.789,the area under ROC curve was 0.821,the accuracy was 0.79,the sensitivity was 0.92,and the specificity was 0.67.Conclusion:Gd-EOB-DTPA enhanced MRI hepatic HCC nodules in hepatobiliary stage were significantly correlated with nodule TTNP,PFS and DCR,and the imaging omics model based on Gd-EOB-DTPA enhanced MRI nodules in hepatobiliary stage had high predictive efficacy for the effect of targeted combined immunotherapy.