P-L12 Exploring Pathological Signatures for Predicting Recurrence of Early-stage Hepatocellular Carcinoma Based on Deep Learning

2021 ◽  
Vol 108 (Supplement_9) ◽  
Author(s):  
Yinghong Shi ◽  
Weifeng Qu ◽  
Mengxin Tian ◽  
Jingtao Qiu ◽  
Kun Qian ◽  
...  
Keyword(s):  
Hepatocellular Carcinoma ◽  
Deep Learning ◽  
Validation Cohort ◽  
Early Stage ◽  
External Validation ◽  
Microvascular Invasion ◽  
Risk Scores ◽  
Forecast Performance ◽  
Histological Score ◽  
The Relationship

Abstract Background Early-stage hepatocellular carcinoma (HCC) is the ideal indication for liver resection. High recurrence rate limits the radical possibility. Current clinicopathological determinants are insufficient to reliably evaluate the recurrence risk after surgery. To address this global issue, we aimed to use deep learning to explore novel pathological signatures based on histological slides for predicting early-stage HCC recurrence and to evaluate the relationship between histological features and multi-omics information. Methods 576 pathological images collected from 547 patients with BCLC stage 0-A HCC who underwent hepatectomy from 2006 to 2015 were randomly divided into the training cohort and validation cohort. The external validation cohort was composed of 147 TNM I patients from TCGA database. Weakly supervised convolutional neural networks were used to identify six classes of HCC tissues. Pathological signatures were extracted and two novel risk scores were constructed by LASSO Cox to predict recurrence. The forecast performance of the scores and patients' prognosis were evaluated. The relationship between histological score (HS) and immune infiltrating cells was estimated by clustering analysis. Results The classification accuracy of HCC tissue was 94.17%. The C-indexes of histological score in the training, validation and TCGA cohorts were 0.804, 0.739 and 0.708, respectively. Multivariate analysis showed that microvascular invasion (HR = 1.46, 95% CI: 1.09-1.95) and HS (HR = 4.05, 95% CI: 3.40-4.84) were independent risk factors for recurrence-free survival. Patients in HS high-risk group had elevated  alpha fetoprotein, worse tumor differentiation and higher proportion of microvascular invasion. HS was positively correlated with the expression of CD14 in adjacent normal liver tissue (P = 0.013), and negatively correlated with the expression of CD8 in tumor (P < 0.001). Conclusions This study established and validated two novel risk scores based on the histological slides using deep learning. HS performed well in recurrence prediction for early-stage HCC patients and indication of important clinicopathological features.

scholarly journals Prediction of Microvascular Invasion in Hepatocellular Carcinoma via Deep Learning: A Multi-Center and Prospective Validation Study

Cancers ◽  
2021 ◽  
Vol 13 (10) ◽  
pp. 2368
Author(s):  
Jingwei Wei ◽  
Hanyu Jiang ◽  
Mengsu Zeng ◽  
Meiyun Wang ◽  
Meng Niu ◽  
...  
Keyword(s):  
Hepatocellular Carcinoma ◽  
Deep Learning ◽  
High Risk ◽  
External Validation ◽  
Progression Free Survival ◽  
Risk Groups ◽  
Postoperative Recurrence ◽  
Microvascular Invasion ◽  
Operating Characteristics ◽  
Enhanced Computed Tomography

Microvascular invasion (MVI) is a critical risk factor for postoperative recurrence of hepatocellular carcinoma (HCC). Preknowledge of MVI would assist tailored surgery planning in HCC management. In this multicenter study, we aimed to explore the validity of deep learning (DL) in MVI prediction using two imaging modalities—contrast-enhanced computed tomography (CE-CT) and gadoxetic acid-enhanced magnetic resonance imaging (EOB-MRI). A total of 750 HCCs were enrolled from five Chinese tertiary hospitals. Retrospective CE-CT (n = 306, collected between March, 2013 and July, 2019) and EOB-MRI (n = 329, collected between March, 2012 and March, 2019) data were used to train two DL models, respectively. Prospective external validation (n = 115, collected between July, 2015 and February, 2018) was performed to assess the developed models. Furthermore, DL-based attention maps were utilized to visualize high-risk MVI regions. Our findings revealed that the EOB-MRI-based DL model achieved superior prediction outcome to the CE-CT-based DL model (area under receiver operating characteristics curve (AUC): 0.812 vs. 0.736, p = 0.038; sensitivity: 70.4% vs. 57.4%, p = 0.015; specificity: 80.3% vs. 86.9%, p = 0.052). DL attention maps could visualize peritumoral high-risk areas with genuine histopathologic confirmation. Both DL models could stratify high and low-risk groups regarding progression free survival and overall survival (p < 0.05). Thus, DL can be an efficient tool for MVI prediction, and EOB-MRI was proven to be the modality with advantage for MVI assessment than CE-CT.

scholarly journals Radiomics Models for Predicting Microvascular Invasion in Hepatocellular Carcinoma: A Systematic Review and Radiomics Quality Score Assessment

Cancers ◽  
2021 ◽  
Vol 13 (22) ◽  
pp. 5864
Author(s):  
Qiang Wang ◽  
Changfeng Li ◽  
Jiaxing Zhang ◽  
Xiaojun Hu ◽  
Yingfang Fan ◽  
...  
Keyword(s):  
Hepatocellular Carcinoma ◽  
Methodological Quality ◽  
Prediction Models ◽  
Validation Cohort ◽  
External Validation ◽  
Quality Score ◽  
Risk Of Bias ◽  
Microvascular Invasion ◽  
Patient Treatment ◽  
Preoperative Prediction

Preoperative prediction of microvascular invasion (MVI) is of importance in hepatocellular carcinoma (HCC) patient treatment management. Plenty of radiomics models for MVI prediction have been proposed. This study aimed to elucidate the role of radiomics models in the prediction of MVI and to evaluate their methodological quality. The methodological quality was assessed by the Radiomics Quality Score (RQS), and the risk of bias was evaluated by the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2). Twenty-two studies using CT, MRI, or PET/CT for MVI prediction were included. All were retrospective studies, and only two had an external validation cohort. The AUC values of the prediction models ranged from 0.69 to 0.94 in the test cohort. Substantial methodological heterogeneity existed, and the methodological quality was low, with an average RQS score of 10 (28% of the total). Most studies demonstrated a low or unclear risk of bias in the domains of QUADAS-2. In conclusion, a radiomics model could be an accurate and effective tool for MVI prediction in HCC patients, although the methodological quality has so far been insufficient. Future prospective studies with an external validation cohort in accordance with a standardized radiomics workflow are expected to supply a reliable model that translates into clinical utilization.

scholarly journals CYSTM1: A Novel Biomarker for Hepatocellular Carcinoma Prognosis

2021 ◽  
Author(s):  
Jun Du ◽  
Jinguo Wang
Keyword(s):  
Hepatocellular Carcinoma ◽  
Poor Prognosis ◽  
Early Stage ◽  
Human Tumor ◽  
The Cancer Genome Atlas ◽  
Data Set ◽  
Kaplan Meier ◽  
Cancer Genome Atlas ◽  
Cox Proportional Hazard Regression ◽  
The Relationship

Abstract Background: The expression and molecular mechanism of cysteine rich transmembrane module containing 1 (CYSTM1) in human tumor cells remains unclear. The aim of this study was to determine whether CYSTM1 could be used as a potential prognostic biomarker for hepatocellular carcinoma (HCC).Methods: We first demonstrated the relationship between CYSTM1 expression and HCC in various public databases. Secondly, Kaplan–Meier analysis and Cox proportional hazard regression model were performed to evaluate the relationship between the expression of CYSTM1 and the survival of HCC patients which data was downloaded in the cancer genome atlas (TCGA) database. Finally, we used the expression data of CYSTM1 in TCGA database to predict CYSTM1-related signaling pathways through bioinformatics analysis.Results: The expression level of CYSTM1 in HCC tissues was significantly correlated with T stage (p = 0.039). In addition, Kaplan–Meier analysis showed that the expression of CYSTM1 was significantly associated with poor prognosis in patients with early-stage HCC (p = 0.003). Multivariate analysis indicated that CYSTM1 is a potential predictor of poor prognosis in HCC patients (p = 0.036). The results of biosynthesis analysis demonstrated that the data set of CYSTM1 high expression was mainly enriched in neurodegeneration and oxidative phosphorylation pathways.Conclusion: CYSTM1 is an effective biomarker for the prognosis of patients with early-stage HCC and may play a key role in the occurrence and progression of HCC.

Identification and validation of a potent multi-mRNA signature for the prediction of early relapse in hepatocellular carcinoma

2019 ◽  
Vol 40 (7) ◽  
pp. 840-852 ◽  
Author(s):  
Jie Cai ◽  
Ying Tong ◽  
Lifeng Huang ◽  
Lei Xia ◽  
Han Guo ◽  
...  
Keyword(s):  
Gene Expression ◽  
Hepatocellular Carcinoma ◽  
Messenger Rna ◽  
Validation Cohort ◽  
Expression Profiles ◽  
External Validation ◽  
Early Recurrence ◽  
Discovery Cohort ◽  
Early Relapse ◽  
Staging Systems

Abstract Early recurrence of hepatocellular carcinoma (HCC) is implicated in poor patient survival and is the major obstacle to improving prognosis. The current staging systems are insufficient for accurate prediction of early recurrence, suggesting that additional indicators for early recurrence are needed. Here, by analyzing the gene expression profiles of 12 Gene Expression Omnibus data sets (n = 1533), we identified 257 differentially expressed genes between HCC and non-tumor tissues. Least absolute shrinkage and selection operator regression model was used to identify a 24-messenger RNA (mRNA)-based signature in discovery cohort GSE14520. With specific risk score formula, patients were divided into high- and low-risk groups. Recurrence-free survival within 2 years (early-RFS) was significantly different between these two groups in discovery cohort [hazard ratio (HR): 7.954, 95% confidence interval (CI): 4.596–13.767, P < 0.001], internal validation cohort (HR: 8.693, 95% CI: 4.029–18.754, P < 0.001) and external validation cohort (HR: 5.982, 95% CI: 3.414–10.480, P < 0.001). Multivariable and subgroup analyses revealed that the 24-mRNA-based classifier was an independent prognostic factor for predicting early relapse of patients with HCC. We further developed a nomogram integrating the 24-mRNA-based signature and clinicopathological risk factors to predict the early-RFS. The 24-mRNA-signature-integrated nomogram showed good discrimination (concordance index: 0.883, 95% CI: 0.836–0.929) and calibration. Decision curve analysis demonstrated that the 24-mRNA-signature-integrated nomogram was clinically useful. In conclusion, our 24-mRNA signature is a powerful tool for early-relapse prediction and will facilitate individual management of HCC patients.

Predictive model for microvascular invasion of hepatocellular carcinoma among candidates for either hepatectomy or liver transplantation.

2017 ◽  
Vol 35 (15_suppl) ◽  
pp. 4079-4079
Author(s):  
Hidetoshi Nitta ◽  
Marc Antoine Allard ◽  
Mylene Sebagh ◽  
Gabriella Pittau ◽  
Oriana Ciacio ◽  
...  
Keyword(s):  
Hepatocellular Carcinoma ◽  
Liver Transplantation ◽  
Predictive Model ◽  
Validation Cohort ◽  
Microvascular Invasion ◽  
Roc Curve Analysis ◽  
Training Cohort ◽  
Contrast Enhanced Ct ◽  
Contrast Enhanced ◽  
Enhanced Ct

4079 Background: Microvascular invasion (MVI) is the strongest prognostic factor following surgery of hepatocellular carcinoma (HCC). However, it is usually not available on the preoperative setting. A predictive model of MVI in patients scheduled for hepatic resection (HR) or liver transplantation (LT) would thus help guiding treatment strategy. The aim of this study was to develop a predictive model for MVI of HCC before either HR or LT. Methods: HCC patients who consecutively performed HR or LT from January 1994 to June 2016 at a single institution were subdivided into a training and validation cohort. Risk factors for MVI in the training cohort were used to develop a predictive model for MVI, to be validated in the validation cohort. The outcomes of the HR and LT patients with high or low MVI probability based on the model, were compared using propensity score matching (PSM). Cut-off values for continuous factors were determined based on ROC curve analysis. Results: A total of 910 patients (425 HR, 485 LT) were included in the training (n = 637) and validation (n = 273) cohorts. In the training cohort, multivariate analysis demonstrated that alpha-fetoprotein ≥100ng/ml ( p < 0.0001), largest tumor size ≥40mm ( p = 0.0002), non-boundary HCC type on contrast-enhanced CT ( p = 0.001), neutrophils-to-lymphocytes ratio ≥3.2 ( p = 0.002), aspartate aminotransferase ≥62U/l ( p = 0.02) were independently associated with MVI. Combinations of these 5 factors varied the MVI probability from 15.5% to 91.1%. This predictive model achieved a good c-index of 0.76 in the validation cohort. In PSM (109 HR, 109 LT), there was no difference in survival between HR and LT patients among the high MVI probability (≥50%) patients, (5y-OS; 46.3% vs 42.2%, p = 0.77, 5y-RFS; 54.0% vs 28.8%, p = 0.21). Among the low probability ( < 50%), survival was significantly decreased following HR compared with LT (5y-OS; 54.1% vs 78.8%, p = 0.007, 5y-RFS; 17.3% vs 86.1%, p< 0.0001). Conclusions: This model developed from preoperative data allows reliable prediction of MVI, and may thus help with preoperative decisions about the suitability of HR or LT in patients with HCC.

Sign in / Sign up

Export Citation Format

Share Document