Cancer is one of the common diseases. Quantitative biomarkers extracted from standard-of-care computed tomography (CT) scan can create a robust clinical decision tool for the diagnosis of hepatocellular carcinoma (HCC). According to the current clinical methods, the situation usually accounts for high expenditure of time and resources. To improve the current clinical diagnosis and therapeutic procedure, this paper proposes a deep learning-based approach, called Successive Encoder-Decoder (SED), to assist in the automatic interpretation of liver lesion/tumor segmentation through CT images. The SED framework consists of two different encoder-decoder networks connected in series. The first network aims to remove unwanted voxels and organs and to extract liver locations from CT images. The second network uses the results of the first network to further segment the lesions. For practical purpose, the predicted lesions on individual CTs were extracted and reconstructed on 3D images. The experiments conducted on 4300 CT images and LiTS dataset demonstrate that the liver segmentation and the tumor prediction achieved 0.92 and 0.75 in Dice score, respectively, by as-proposed SED method.

Accumulating evidence suggests the involvement of tumor-derived exosomes in the development and recurrence of hepatocellular carcinoma (HCC). We previously identified miR-4669 as a highly expressed microRNA in circulating exosomes obtained from patients with post-transplant HCC recurrence. This study aimed to explore how overexpression of miR-4669 affects HCC development and recurrence. The impact of miR-4669 overexpression in Hep3B cells on tumor cell behavior and the tumor microenvironment was evaluated in vitro. In addition, the clinical value of exosomal miR-4669 for the prediction of treatment response to HCC downstaging therapies and following post-transplant HCC recurrence was explored. Overexpression of miR-4669 enhanced migration ability and led to acquired sorafenib resistance with an elevation of sirtuin 1 and long noncoding RNA associated with microvascular invasion. Active release of tumor-derived exosomes and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) contributed to generating an immunosuppressive tumor microenvironment through the induction of M2 macrophage polarization. The retrospective analysis demonstrated the clinical value of exosomal miR-4669 for predicting treatment response to HCC downstaging therapies and for risk assessment of post-transplant HCC recurrence. In summary, the present data demonstrate the impact of exosomal miR-4669 on HCC recurrence through the enhancement of tumor aggressiveness and generation of an immunosuppressive tumor microenvironment.

Background: The purpose of this study was to assess the safety and efficacy of Yttrium-90 radioembolization using in unresectable hepatocellular carcinoma. Methods: From 2017 to 2021, 32 patients with unresectable hepatocellular carcinoma, with mean tumor diameter about 7cm (21 males, 11 females; median age, 57.5 years of age), treated with Yttrium-90 radioembolization using resin microspheres were reviewed at pre-Yttrium-90 and post-Yttrium-90 follow-up. Tumor response was assessed according to the modified Response Evaluation Criteria in Solid Tumors. Outcomes including overall survival and progression-free survival were reported. Results: Median follow-up was 18 months. At follow-up examinations at 3-, 6-, and 12-months follow-up, the overall survival rates were 94%, 87% and 59%, and the progression-free survival rates were 78%, 64% and 60%, respectively. Complete response, partial response, stable disease, and progressive disease were noted in 7 (21.9%), 14 (43.7%), 4 (12.5%), and 7 (21.9%) patients, respectively. The disease control rate was 78.1%, the objective response rate was 65.6%, and the successful downstage rate was 34.4% (11 of 32). Nine of thirty-two patients underwent resection or transplantation after Yttrium-90 radioembolization with 2-year overall survival being 100%. No serious adverse events occurred after Yttrium-90 treatment. Worse overall survival was related to the larger tumor, higher stage, Eastern Cooperative Oncology Group performance status, and Child-Pugh score. And worse progression-free survival was related to the higher tumor burden, and pre-Yttrium-90 serum α-fetoprotein level > 100. Conclusion: Yttrium-90 Radioembolization can control hepatocellular carcinoma well even in advanced diseases. Patients successfully downstaging/bridging to resection or transplantation have excellent overall survival.

PURPOSE We aimed to compare the safety and effectiveness of 100-300 mu m versus 300-500 mu m drug-eluting bead transarterial chemoembolization (DEB-TACE) and to investigate the impact of tumor and feeding artery size on treatment outcome of different particle sizes in the treatment of hepatocellular carcinoma (HCC). METHODS This retrospective cohort study enrolled 234 consecutive patients who underwent TACE using 100-300 mu m DEB (Group A, n=75) and 300-500 mu m DEB (Group B, n=159) in a tertiary center between August 2012 and March 2017. Initial treatment response and adverse events were assessed using modified Response Evaluation Criteria in Solid Tumors (mRECIST) and National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE) version 5.0, respectively. RESULTS A total of 704 HCCs in 234 patients were evaluated. The average index tumor size was 3.8 cm. Multivariate analysis showed that tumor size, lobe inv olvement, particle size, and tumor location were significant predictive factors of complete response. The overall rate of complete response in groups A and B were 56.0% and 33.3% (P = 0.001), respectively. Group A had higher complete response rate than group B in the subgroup of BCLC B with tumor = 0.9 mm (55.2% vs. 30.9%; P = 0.014). There were fewer major complications in group A compared with group B (0% vs. 6.9%, P = 0.018). CONCLUSION TACE with 100-300 mu m DEB is associated with better initial treatment response and fewer major complications compared with 300-500 mu m. Our study also highlights the impact of tumor characteristics on treatment outcome of different DEB size, which might help to select the optimal sphere size for TACE in the treatment of HCC.

The research was based on the image recognition technology of artificial intelligence, which is expected to assist physicians in making correct decisions through deep learning. The liver dataset used in this study was derived from the open source website (LiTS) and the data provided by the Kaohsiung Chang Gung Memorial Hospital. CT images were used for organ recognition and lesion segmentation; the proposed Residual-Dense-Attention (RDA) U-Net can achieve high accuracy without the use of contrast. In this study, U-Net neural network was used to combine ResBlock in ResNet with Dense Block in DenseNet in the coder part, allowing the training to maintain the parameters while reducing the overall recognition computation time. The decoder was equipped with Attention Gates to suppress the irrelevant areas of the image while focusing on the significant features. The RDA model was used to identify and segment liver organs and lesions from CT images of the abdominal cavity, and excellent segmentation was achieved for the liver located on the left side, right side, near the heart, and near the lower abdomen with other organs. Better recognition was also achieved for large, small, and single and multiple lesions. The study was able to reduce the overall computation time by about 28% compared to other convolutions, and the accuracy of liver and lesion segmentation reached 96% and 94.8%, with IoU values of 89.5% and 87%, and AVGDIST of 0.28 and 0.80, respectively.