Search

Purpose Microvascular invasion (MVI) is a valuable predictor of survival in hepatocellular carcinoma (HCC) patients. This study developed predictive models using eXtreme Gradient Boosting (XGBoost) and deep learning based on CT images to predict MVI preoperatively. Methods In total, 405 patients were included. A total of 7302 radiomic features and 17 radiological features were extracted by a radiomics feature extraction package and radiologists, respectively. We developed a XGBoost model based on radiomics features, radiological features and clinical variables and a three-dimensional convolutional neural network (3D-CNN) to predict MVI status. Next, we compared the efficacy of the two models. Results Of the 405 patients, 220 (54.3%) were MVI positive, and 185 (45.7%) were MVI negative. The areas under the receiver operating characteristic curves (AUROCs) of the Radiomics-Radiological-Clinical (RRC) Model and 3D-CNN Model in the training set were 0.952 (95% confidence interval (CI) 0.923–0.973) and 0.980 (95% CI 0.959–0.993), respectively (p = 0.14). The AUROCs of the RRC Model and 3D-CNN Model in the validation set were 0.887 (95% CI 0.797–0.947) and 0.906 (95% CI 0.821–0.960), respectively (p = 0.83). Based on the MVI status predicted by the RRC and 3D-CNN Models, the mean recurrence-free survival (RFS) was significantly better in the predicted MVI-negative group than that in the predicted MVI-positive group (RRC Model: 69.95 vs. 24.80 months, p p = 0.027). Conclusion The RRC Model and 3D-CNN models showed considerable efficacy in identifying MVI preoperatively. These machine learning models may facilitate decision-making in HCC treatment but requires further validation.

Background & aims To explore the ability of gadolinium-ethoxybenzyl-diethylenetriamine pentaacetic acid magnetic resonance imaging (EOB-MRI)/ultrasound (US) fusion imaging (FI) to improve the prognosis of radiofrequency ablation (RFA) by ablating the characteristic findings of hepatocellular carcinoma (HCC) in hepatobiliary phase (HBP) imaging. Methods We retrospectively recruited 115 solitary HCC lesions with size of (15.9 ± 4.6) mm. They were all treated by RFA and preoperative EOB-MRI. According to the modalities guiding RFA performance, the lesions were grouped into contrast enhanced US (CEUS)/US guidance group and EOB-MRI/US FI guidance group. For the latter group, the ablation scope was set to cover the HBP findings (peritumoral hypointensity and irregular protruding margin). The presence of HBP findings, the modalities guided RFA, the recurrence rate were observed. Results After an average follow-up of 377 days, local tumor progression (LTP) and intrahepatic distant recurrence (IDR) were 14.8% and 38.4%, respectively. The lesions having HBP findings exhibited a higher recurrence rate (73.7%) than the lesions without HBP findings (42.9%) (p = 0.002) and a low overall recurrence-free curve using the Kaplan-Meier method (p = 0.038). Using EOB-MRI/US FI as guidance, there was no difference in the recurrence rate between the groups with and without HBP findings (p = 0.799). In lesions with HBP findings, RFA guided by EOB-MRI/US FI (53.8%) produced a lower recurrence rate than CEUS/US (84.0%) (p = 0.045). Conclusions The intraprocedurally application of EOB-MRI/US FI to determine ablation scope according to HBP findings is feasible and beneficial for prognosis of RFA.

We appreciate to receive commentary from Dr Guangtong Deng and Dr Liang Xiao to our article, “Nomograms based on inflammatory biomarkers for predicting tumor grade and micro-vascular invasion in stage I/II hepatocellular carcinoma”. First, neutrophil-to-lymphocyte ratio (NLR) and derived NLR (dNLR) are two different parameters. Some studies show that NLR is inconsistent with dNRL in prognostic value through multivariate Cox regression, therefore, it is reasonable that both NLR and dNLR entered into multivariate analysis simultaneously. Second, it is common that articles of predictive nomograms turned continuous variables into categorical variables. The reason is that the categorization of patient clinical variables is beneficial to doctors to make decisions based on the risk level of individual patients in clinical. At last, multicenter validation is quite difficult and we have listed the shortcomings in the limitations of our article. Further validation will need the joint efforts by other institutions.

Background: Increasing evidences reveal that inflammation plays a critical role in tumorigenesis and progression. We aimed to develop the nomograms based on inflammatory biomarkers to predict micro-vascular invasion (MVI) and tumor grade in stage I/II hepatocellular carcinoma (HCC). Methods: A retrospective cohort of 627 patients with stage I/II HCC between January 2007 and December 2014 was included in the study. Logistic regression was performed to identify the independent risk factors of tumor grade and MVI. The significant predictors including neutrophil-to-lymphocyte ratio (NLR), derived neutrophil-to-lymphocyte ratio (dNLR), lymphocyte-to-monocyte ratio (LMR), tumor volume age, and tumor size were subsequently incorporated to build the nomograms. The prediction accuracies of the nomograms were evaluated using the area under the receiver operating characteristic (ROC) curve. Results: The independent risk factors for tumor grade were NLR, dNLR, and tumor volume (P Conclusion: We successfully developed nomograms predicting tumor grade and MVI based on inflammatory biomarkers with high accuracy, leading to a rational therapeutic choice for stage I/II HCC.