Early detection and prognosis evaluation for hepatocellular carcinoma by circulating tumour DNA methylation: A multicentre cohort study.

BACKGROUND: Early diagnosis of hepatocellular carcinoma (HCC) can significantly improve patient survival. We aimed to develop a blood-based assay to aid in the diagnosis, detection and prognostic evaluation of HCC. METHODS: A three-phase multicentre study was conducted to screen, optimise and validate HCC-specific differentially methylated regions (DMRs) using next-generation sequencing and quantitative methylation-specific PCR (qMSP). RESULTS: Genome-wide methylation profiling was conducted to identify DMRs distinguishing HCC tumours from peritumoural tissues and healthy plasmas. The twenty most effective DMRs were verified and incorporated into a multilocus qMSP assay (HepaAiQ). The HepaAiQ model was trained to separate 293 HCC patients (Barcelona Clinic Liver Cancer (BCLC) stage 0/A, 224) from 266 controls including chronic hepatitis B (CHB) or liver cirrhosis (LC) (CHB/LC, 96), benign hepatic lesions (BHL, 23), and healthy controls (HC, 147). The model achieved an area under the curve (AUC) of 0.944 with a sensitivity of 86.0% in HCC and a specificity of 92.1% in controls. Blind validation of the HepaAiQ model in a cohort of 523 participants resulted in an AUC of 0.940 with a sensitivity of 84.4% in 205 HCC cases (BCLC stage 0/A, 167) and a specificity of 90.3% in 318 controls (CHB/LC, 100; BHL, 102; HC, 116). When evaluated in an independent test set, the HepaAiQ model exhibited a sensitivity of 70.8% in 65 HCC patients at BCLC stage 0/A and a specificity of 89.5% in 124 patients with CHB/LC. Moreover, HepaAiQ model was assessed in paired pre- and postoperative plasma samples from 103 HCC patients and correlated with 2-year patient outcomes. Patients with high postoperative HepaAiQ score showed a higher recurrence risk (Hazard ratio, 3.33, p < .001). CONCLUSIONS: HepaAiQ, a noninvasive qMSP assay, was developed to accurately measure HCC-specific DMRs and shows great potential for the diagnosis, detection and prognosis of HCC, benefiting at-risk populations.

Genomic and transcriptomic characteristics of 12 novel primary cell lines derived from three patients with cholangiocarcinoma.

Cholangiocarcinoma (CCA) is an aggressive bile duct malignancy with poor prognosis. To improve our understanding of the biological characteristics of CCA and develop effective therapies, appropriate preclinical models are required. Here, we established and characterized 12 novel patient-derived primary cancer cell (PDPC) models using multi-region sampling. At the genomic level of PDPCs, we observed not only commonly mutated genes, such as TP53, JAK3, and KMT2C, consistent with the reports in CCA, but also specific mutation patterns in each cell line. In addition, specific expression patterns with distinct biological functions and pathways involved were also observed in the PDPCs at the transcriptomic level. Furthermore, the drug-sensitivity results revealed that the PDPCs exhibited different responses to the six commonly used compounds. Our findings indicate that the established PDPCs can serve as novel in vitro reliable models to provide a crucial molecular basis for improving the understanding of tumorigenesis and its treatment.

Deformable Generator Networks: Unsupervised Disentanglement of Appearance and Geometry.

We present a deformable generator model to disentangle the appearance and geometric information for both image and video data in a purely unsupervised manner. The appearance generator network models the information related to appearance, including color, illumination, identity or category, while the geometric generator performs geometric warping, such as rotation and stretching, through generating deformation field which is used to warp the generated appearance to obtain the final image or video sequences. Two generators take independent latent vectors as input to disentangle the appearance and geometric information from image or video sequences. For video data, a nonlinear transition model is introduced to both the appearance and geometric generators to capture the dynamics over time. The proposed scheme is general and can be easily integrated into different generative models. An extensive set of qualitative and quantitative experiments shows that the appearance and geometric information can be well disentangled, and the learned geometric generator can be conveniently transferred to other image datasets that share similar structure regularity to facilitate knowledge transfer tasks.

Imposing implicit feasibility constraints on deformable image registration using a statistical generative model.

Purpose: Deformable registration problems are conventionally posed in a regularized optimization framework, where balance between fidelity and prescribed regularization usually needs to be tuned for each case. Even so, using a single weight to control regularization strength may be insufficient to reflect spatially variant tissue properties and limit registration performance. In this study, we proposed to incorporate a spatially variant deformation prior into image registration framework using a statistical generative model. Approach: A generator network is trained in an unsupervised setting to maximize the likelihood of observing the moving and fixed image pairs, using an alternating back-propagation approach. The trained model imposes constraints on deformation and serves as an effective low-dimensional deformation parametrization. During registration, optimization is performed over this learned parametrization, eliminating the need for explicit regularization and tuning. The proposed method was tested against SimpleElastix, DIRNet, and Voxelmorph. Results: Experiments with synthetic images and simulated CTs showed that our method yielded registration errors significantly lower than SimpleElastix and DIRNet. Experiments with cardiac magnetic resonance images showed that the method encouraged physical and physiological feasibility of deformation. Evaluation with left ventricle contours showed that our method achieved a dice of ( 0.93 ± 0.03 ) with significant improvement over all SimpleElastix options, DIRNet, and VoxelMorph. Mean average surface distance was on millimeter level, comparable to the best SimpleElastix setting. The average 3D registration time was 12.78 s, faster than 24.70 s in SimpleElastix. Conclusions: The learned implicit parametrization could be an efficacious alternative to regularized B-spline model, more flexible in admitting spatial heterogeneity.

Replicating Neuroscience Observations on ML/MF and AM Face Patches by Deep Generative Model.

A recent Cell paper (Chang & Tsao, 2017) reports an interesting discovery. For the face stimuli generated by a pretrained active appearance model (AAM), the responses of neurons in the areas of the primate brain that are responsible for face recognition exhibit a strong linear relationship with the shape variables and appearance variables of the AAM that generates the face stimuli. In this letter, we show that this behavior can be replicated by a deep generative model, the generator network, that assumes that the observed signals are generated by latent random variables via a top-down convolutional neural network. Specifically, we learn the generator network from the face images generated by a pretrained AAM model using a variational autoencoder, and we show that the inferred latent variables of the learned generator network have a strong linear relationship with the shape and appearance variables of the AAM model that generates the face images. Unlike the AAM model, which has an explicit shape model where the shape variables generate the control points or landmarks, the generator network has no such shape model and shape variables. Yet it can learn the shape knowledge in the sense that some of the latent variables of the learned generator network capture the shape variations in the face images generated by AAM.

Long non-coding RNA DILC promotes the progression of gallbladder carcinoma.

Increasing evidence has demonstrated that long non-coding RNAs (lncRNAs) contribute to tumorigenesis, progression and recurrence of various malignancies including Gallbladder carcinoma (GBC). Lnc-DILC is reported to be the tumor suppressor gene to play an important role in liver cancer stem cells (CSCs). However, the role of lnc-DILC in GBC remains to be elucidated. Herein, we show that lnc-DILC is upregulated in gallbladder CSCs and GBC patients' tissues. Knockdown of lnc-DILC attenuates the self-renewal, tumorigenicity, proliferation and metastasis of gallbladder CSCs. Mechanistically, lnc-DILC promotes gallbladder CSCs expansion via Wnt/ß-catenin pathway. Special Wnt/ß-catenin inhibitor FH535 diminishes the discrepancy of self-renewal, growth and metastasis between lnc-DILC interference GBC cells and their control cells. In conclusion, lnc-DILC drives gallbladder CSCs self-renewal, tumorigenicity, proliferation and metastasis by activating Wnt/ß-catenin signaling, and may therefore prove to be a potential therapeutic target for GBC patients.

Hydroxylase Activity of ASPH Promotes Hepatocellular Carcinoma Metastasis Through Epithelial-to-Mesenchymal Transition Pathway.

Over-expression of aspartyl (asparagynal)-ß-hydroxylase (ASPH) contributes to hepatocellular carcinoma (HCC) invasiveness, but the role of ASPH hydroxylase activity in this process remains to be defined. As such, the current study investigated the role of ASPH hydroxylase activity in downstream signalling of HCC tumorgenesis and, specifically, metastasis development. Over-expression of wild-type ASPH, but not a hydroxylase mutant, promoted HCC cell migration in vitro, as well as intrahepatic and distant metastases in vivo. The enhanced migration and epithelial to mesenchymal transition (EMT) activation was notably absent in response to hydroxylase activity blockade. Vimentin, a regulator of EMT, interacted with ASPH and likely mediated the effect of ASPH hydroxylase activity with cell migration. The enhanced hydroxylase activity in tumor tissues predicted worse prognoses of HCC patients. Collectively, the hydroxylase activity of ASPH affected HCC metastasis through interacting with vimentin and regulating EMT. As such, ASPH might be a promising therapeutic target of HCC.

Long non-coding RNA DILC suppresses cell proliferation and metastasis in colorectal cancer.

Colorectal cancer (CRC) is one of the most common malignant tumors and one of the leading causes of cancer-related death in both men and women. The prognosis of CRC remains poor due to the advanced stage and cancer metastasis at the time of diagnosis. However, the exact mechanism of tumorigenesis in CRC remains unclear. Long non-coding RNAs (lncRNAs), which refer to transcripts longer than 200 nucleotides that are not translated into protein, are known to play important roles in multiple human cancers. Lnc-DILC is reported to be an important tumor suppressor gene and its inactivation is closely associated with liver cancer stem cells. However, the role of lnc-DILC in CRC remains to be elucidated. In the present study, we observed that lnc-DILC overexpression inhibited the growth and metastasis of CRC cells. Consistently, lnc-DILC knockdown facilitated the proliferation and metastasis of CRC cells. Mechanically, lnc-DILC suppressed CRC cell progression via IL-6/STAT3 signaling inactivation. More importantly, the specific STAT3 inhibitor S3I-201 and IL-6R inhibitor tocilizumab abolished the discrepancy of growth and metastasis capacity between lnc-DILC-interference CRC cells and control cells, which further confirmed that IL-6/STAT3 signaling was required in lnc-DILC-disrupted CRC cell growth and metastasis. Taken together, our results suggest that lnc-DILC is a novel CRC suppressor and may prove to be an inhibitor of CRC progression by inactivating IL-6/STAT3 signaling.

Effectiveness of repeat hepatic resection for patients with recurrent intrahepatic cholangiocarcinoma: Factors associated with long-term outcomes.

BACKGROUND: Tumor recurrence after liver resection for intrahepatic cholangiocarcinoma is common. The effective treatment for recurrent intrahepatic cholangiocarcinoma remains to be established. This study evaluated the short- and long-term prognoses of patients after repeat hepatic resection for recurrent intrahepatic cholangiocarcinoma. METHODS: Data for 72 patients who underwent R0 repeat hepatic resection for recurrent intrahepatic cholangiocarcinoma at the Eastern Hepatobiliary Surgery Hospital between 2005 and 2013 were analyzed. Tumor re-recurrence, recurrence-to-death survival, and overall survival were calculated and compared using the Kaplan-Meier method and the log-rank test. Independent risk factors were identified by Cox regression analysis. RESULTS: Operative morbidity and mortality rates were 18.1% and 1.4%, respectively. The 1-, 2-, and 3-year re-recurrence rates were 53.2%, 80.2%, and 92.6%, respectively, and the corresponding recurrence-to-death survival was 82.9%, 53.0%, and 35.3%, respectively. The 1-, 3-, and 5-year overall survival was 97.2%, 67.0%, and 41.9%, respectively. Patients with a time to recurrence of >1 year from the initial hepatectomy achieved higher 1-, 2-, and 3-year recurrence-to-death survival than patients with a time to recurrence of ≤1 year (92.5%, 61.7%. and 46.6% vs 70.4%, 42.2%, and 23.0%, P = .022). Multivariate analysis identified that recurrent tumor >3 cm (hazard ratio: 2.346; 95% confidence interval: 1.288-4.274), multiple recurrent nodules (2.304; 1.049-5.059), cirrhosis (3.165; 1.543-6.491), and a time to recurrence of ≤1 year (1.872; 1.055-3.324) were independent risk factors of recurrence-to-death survival. CONCLUSION: Repeat hepatic resection for recurrent intrahepatic cholangiocarcinoma was safe and produced long-term survival outcomes in selected patients based on prognostic stratification with the presence of the independent risk factors of recurrence-to-death survival.

A novel scoring system predicts adjuvant chemolipiodolization benefit for hepatocellular carcinoma patients after hepatectomy.

Our aim in this study was to develop a prognostic scoring system with which to identify patients most likely to benefit from adjuvant chemolipiodolization (ACL) after liver resection for hepatocellular carcinoma (HCC). Data from 1150 HCC patients who underwent liver resection between 2002 and 2008 at the Eastern Hepatobiliary Surgery Hospital were used to develop the scoring system. Patients were stratified into prognostic subgroups using the new scoring system, and the outcomes of patients who received ACL and those who did not were compared in each subgroup. Using data from 379 patients operated on between 2008 and 2010 for validation, the scoring system had a concordance index (C-index) of 0.75 for predicting post-resectional overall survival (OS). It optimally stratified patients into three prognostic subgroups with scores of 0-5, 6-9 and ≥ 10, having better, medium and worse survival outcomes, respectively. A difference in OS between ACL and non-ACL patients was only detected in the subgroup with scores ≥ 10 (1-, 3-, and 5-year OS rates: 63.9%, 22.6%, and 9.0% vs. 33.8%, 5.6%, and 2.8%, p = 0.001). Our proposed scoring system provides an effective tool for selecting the patients most likely to benefit from ACL.

Class Energy Image analysis for video sensor-based gait recognition: a review.

Gait is a unique perceptible biometric feature at larger distances, and the gait representation approach plays a key role in a video sensor-based gait recognition system. Class Energy Image is one of the most important gait representation methods based on appearance, which has received lots of attentions. In this paper, we reviewed the expressions and meanings of various Class Energy Image approaches, and analyzed the information in the Class Energy Images. Furthermore, the effectiveness and robustness of these approaches were compared on the benchmark gait databases. We outlined the research challenges and provided promising future directions for the field. To the best of our knowledge, this is the first review that focuses on Class Energy Image. It can provide a useful reference in the literature of video sensor-based gait representation approach.

Reexpression of Let-7g microRNA inhibits the proliferation and migration via K-Ras/HMGA2/snail axis in hepatocellular carcinoma.

Let-7 family microRNAs have been reported to be downregulated in human hepatocellular carcinoma in comparison with normal hepatic tissues. Among them, let-7g was identified as the lowest expression using real-time RT-PCR. However, the mechanism by which let-7g works in hepatocellular carcinoma remains unknown. Here, in our present study, we have had let-7g reexpressed in vitro in hepatocellular carcinoma cell lines MHCC97-H and HCCLM3 via transfection. The proliferation after reexpression of let-7g was assayed using MTT method; the migration and invasion after restoration were detected by wound-healing and Transwell assay, respectively. We found using Western-blotting that let-7g can regulate epithelial-mesenchymal transition (EMT) by downregulating K-Ras and HMGA2A after reexpresssion. Xenografted nude mice were used to observe whether or not reexpression of let-7g could have potential therapeutic ability. In vivo, to observe the association with let-7g expression and overall prognosis, 40 paired cases of hepatocellular carcinoma were analyzed using in situ hybridization (ISH). It was found that reexpression of let-7g can inhibit the proliferation, migration, and invasion significantly, and that low expression of let-7g was significantly associated with poorer overall survival. Taken together, let-7g could be used as a promising therapeutic agent in vivo in the treatment of hepatocellular carcinoma at the earlier stage.