Our clinical and literature analyses suggest that in patients with HCC with preserved liver function, the presence of large, solitary tumors, multinodular tumors, macrovascular invasion, or PHT are not contraindications for HR.

To clarify the significance of circulating tumor cells (CTC) undergoing epithelial-mesenchymal transition (EMT) in patients with hepatocellular carcinoma (HCC), we used an advanced CanPatrol CTC-enrichment technique and hybridization to enrich and classify CTC from blood samples. One hundred and one of 112 (90.18%) patients with HCC were CTC positive, even with early-stage disease. CTCs were also detected in 2 of 12 patients with hepatitis B virus (HBV), both of whom had small HCC tumors detected within 5 months. CTC count ≥16 and mesenchymal-CTC (M-CTC) percentage ≥2% prior to resection were significantly associated with early recurrence, multi-intrahepatic recurrence, and lung metastasis. Postoperative CTC monitoring in 10 patients found that most had an increased CTC count and M-CTC percentage before clinically detectable recurrence nodules appeared. Analysis of HCC with high CTC count and high M-CTC percentage identified 67 differentially expressed cancer-related genes involved in cancer-related biological pathways (e.g., cell adhesion and migration, tumor angiogenesis, and apoptosis). One of the identified genes, BCAT1, was significantly upregulated, and knockdown in Hepg2, Hep3B, and Huh7 cells reduced cell proliferation, migration, and invasion while promoting apoptosis. A concomitant increase in epithelial marker expression (EpCAM and E-cadherin) and reduced mesenchymal marker expression (vimentin and Twist) suggest that BCAT1 may trigger the EMT process. Overall, CTCs were highly correlated with HCC characteristics, representing a novel marker for early diagnosis and a prognostic factor for early recurrence. BCAT1 overexpression may induce CTC release by triggering EMT and may be an important biomarker of HCC metastasis. In liver cancer, CTC examination may represent an important "liquid biopsy" tool to detect both early disease and recurrent or metastatic disease, providing cues for early intervention or adjuvant therapy. .

Background and AimsTreatment of patients with Barcelona Clinic Liver Cancer Stage B hepatocellular carcinoma (BCLC-B HCC) is controversial. This study compared the long-term survival of patients with BCLC-B HCC who received liver resection (LR) or transarterial chemoembolization (TACE).MethodsA total of 257 and 135 BCLC-B HCC patients undergoing LR and TACE, respectively, were retrospectively evaluated. Kaplan–Meier method was used for long-term survival analysis. Independent prognostic predictors were determined by the Cox proportional hazards model.ResultsThe hospital mortality rate was similar between groups (3.1% vs. 3.7%; P = 0.76). However, the LR group showed a significantly higher postoperative complication rate than the TACE group (28 vs. 18.5%; P = 0.04). At the same time, the LR group showed significantly higher overall survival rates (1 year, 84 vs. 69%; 3 years, 59 vs. 29%; 5 years, 37 vs. 14%; P<0.001). Moreover, similar results were observed in the propensity score model. Three independent prognostic factors were associated with worse overall survival: serum AFP level (≥400 ng/ml), serum ALT level, and TACE.ConclusionsLR appears to be as safe as TACE for patients with BCLC-B HCC, and it provides better long-term overall survival. However, prospective studies are needed to disclose if LR may be regarded as the preferred treatment for these patients as long as liver function is preserved.

BackgroundWe attempted to train and validate a model of deep learning for the preoperative prediction of the response of patients with intermediate-stage hepatocellular carcinoma (HCC) undergoing transarterial chemoembolization (TACE).MethodAll computed tomography (CT) images were acquired for 562 patients from the Nan Fang Hospital (NFH), 89 patients from Zhu Hai Hospital Affiliated with Jinan University (ZHHAJU), and 138 patients from the Sun Yat-sen University Cancer Center (SYUCC). We built a predictive model from the outputs using the transfer learning techniques of a residual convolutional neural network (ResNet50). The prediction accuracy for each patch was revaluated in two independent validation cohorts.ResultsIn the training set (NFH), the deep learning model had an accuracy of 84.3% and areas under curves (AUCs) of 0.97, 0.96, 0.95, and 0.96 for complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD), respectively. In the other two validation sets (ZHHAJU and SYUCC), the deep learning model had accuracies of 85.1% and 82.8% for CR, PR, SD, and PD. The ResNet50 model also had high AUCs for predicting the objective response of TACE therapy in patches and patients of three cohorts. Decision curve analysis (DCA) showed that the ResNet50 model had a high net benefit in the two validation cohorts.ConclusionThe deep learning model presented a good performance for predicting the response of TACE therapy and could help clinicians in better screening patients with HCC who can benefit from the interventional treatment.Key Points• Therapy response of TACE can be predicted by a deep learning model based on CT images. • The probability value from a trained or validation deep learning model showed significant correlation with different therapy responses. • Further improvement is necessary before clinical utilization. Electronic supplementary materialThe online version of this article (10.1007/s00330-019-06318-1) contains supplementary material, which is available to authorized users.