Image-matching digital macro-slide-a novel pathological examination method for microvascular invasion detection in hepatocellular carcinoma

A new scoring system for predicting the outcome of hepatocellular carcinoma patients without microvascular invasion-a large-scale multicentre study

BACKGROUND

The prognosis of HCC patients without MVI (so called M0) is highly heterogeneous and the need for adjuvant therapy is still controversial.

METHODS

Patients with HCC with M0 who underwent liver resection (LR) or liver transplantation (LT) as an initial therapy were included. The Eastern Hepatobiliary Surgery Hospital (EHBH)-M0 score was developed from a retrospective cohort to form the training cohort. The classification which was developed using multivariate cox regression analysis was externally validated.

RESULTS

The score was developed using the following factors: α-fetoprotein level, tumour diameter, liver cirrhosis, total bilirubin, albumin and aspartate aminotransferase. The score differentiated two groups of M0 patients (≤3, >3 points) with distinct long-term prognoses outcomes (median overall survival (OS), 98.0 vs. 46.0 months; p < 0.001). The predictive accuracy of the score was greater than the other commonly used staging systems for HCC. And for M0 patients with a higher score underwent LR. Adjuvant transcatheter arterial chemoembolization (TACE) was effective to prolong OS.

CONCLUSIONS

The EHBH M0 scoring system was more accurate in predicting the prognosis of HCC patients with M0 after LR or LT. Adjuvant therapy is recommended for HCC patients who have a higher score.

Deep learning-based accurate diagnosis and quantitative evaluation of microvascular invasion in hepatocellular carcinoma on whole-slide histopathology images

BACKGROUND

Microvascular invasion (MVI) is an independent prognostic factor that is associated with early recurrence and poor survival after resection of hepatocellular carcinoma (HCC). However, the traditional pathology approach is relatively subjective, time-consuming, and heterogeneous in the diagnosis of MVI. The aim of this study was to develop a deep-learning model that could significantly improve the efficiency and accuracy of MVI diagnosis.

MATERIALS AND METHODS

We collected H&E-stained slides from 753 patients with HCC at the First Affiliated Hospital of Zhejiang University. An external validation set with 358 patients was selected from The Cancer Genome Atlas database. The deep-learning model was trained by simulating the method used by pathologists to diagnose MVI. Model performance was evaluated with accuracy, precision, recall, F1 score, and the area under the receiver operating characteristic curve.

RESULTS

We successfully developed a MVI artificial intelligence diagnostic model (MVI-AIDM) which achieved an accuracy of 94.25% in the independent external validation set. The MVI positive detection rate of MVI-AIDM was significantly higher than the results of pathologists. Visualization results demonstrated the recognition of micro MVIs that were difficult to differentiate by the traditional pathology. Additionally, the model provided automatic quantification of the number of cancer cells and spatial information regarding MVI.

CONCLUSIONS

We developed a deep learning diagnostic model, which performed well and improved the efficiency and accuracy of MVI diagnosis. The model provided spatial information of MVI that was essential to accurately predict HCC recurrence after surgery.

Precise localization of microvascular invasion in hepatocellular carcinoma based on three-dimensional histology-MR image fusion: an ex vivo experimental study

Background

Microvascular invasion (MVI) is an independent risk factor for postoperative recurrence of hepatocellular carcinoma (HCC). However, MVI cannot be detected by conventional imaging. To localize MVI precisely on magnetic resonance (MR) images, we evaluated the feasibility and accuracy of 3-dimensional (3D) histology-MR image fusion of the liver.

Methods

Animal models of VX2 liver tumors were established in 10 New Zealand white rabbits under ultrasonographic guidance. The whole liver lobe containing the VX2 tumor was extracted and divided into 4 specimens, for a total of 40 specimens. MR images were obtained with a T2-weighted sequence for each specimen, and then histological images were obtained by intermittent, serial pathological sections. 3D histology-MR image fusion was performed via landmark registration in 3D Slicer software. We calculated the success rate and registration errors of image fusion, and then we located the MVI on MR images. Regarding influencing factors, we evaluated the uniformity of tissue thickness after sampling and the uniformity of tissue shrinkage after dehydration.

Results

The VX2 liver tumor model was successfully established in the 10 rabbits. The incidence of MVI was 80% (8/10). 3D histology-MR image fusion was successfully performed in the 39 specimens, and the success rate was 97.5% (39/40). The average registration error was 0.44±0.15 mm. MVI was detected in 20 of the 39 successfully registered specimens, resulting in a total of 166 MVI lesions. The specific location of all MVI lesions was accurately identified on MR images using 3D histology-MR image fusion. All MVI lesions showed as slightly hyperintense on the high-resolution MR T2-weighted images. The results of the influencing factor assessment showed that the tissue thickness was uniform after sampling (P=0.38), but the rates of the tissue shrinkage was inconsistent after dehydration (P<0.001).

Conclusions

3D histology-MR image fusion of the isolated liver tumor model is feasible and accurate and allows for the successful identification of the specific location of MVI on MR images.

A Nomogram Based on Contrast-Enhanced Ultrasound to Predict the Microvascular Invasion in Hepatocellular Carcinoma

OBJECTIVE

The aim of this study was to establish and validate a contrast-enhanced ultrasound (CEUS) nomogram for pre-operative microvascular invasion (MVI) prediction in hepatocellular carcinoma (HCC), and compare it with the nomogram based on gadopentetate dimeglumine-enhanced magnetic resonance imaging (Gd-MRI).

METHODS

A total of 251 patients with a single HCC were enrolled in this prospective study, including 176 patients in the training cohort and 75 patients in the validation cohort. Contrast-enhanced ultrasound (CEUS) with Sonazoid and Gd-MRI was performed pre-operatively. Post-operative histopathology was the gold standard for MVI. Univariate and multivariate logistic regression was performed to determine independent risk factors for MVI. Nomograms based on CEUS and Gd-MRI were established, and their discrimination, calibration and decision curve analysis were evaluated and compared.

RESULTS

Multivariate logistic regression revealed that arterial circular enhancement, non-enhancing area and thick ring-like enhancement in the post-vascular phase were independent risk factors for MVI. The areas under the receiver operating characteristic curve of the nomogram were 0.841 (0.779-0.892) and 0.914 (0.827-0.966) in the training and validation cohorts, with no significant difference compared with the Gd-MRI nomogram (p = 0.294, 0.321). The C-indexes were 0.821 and 0.870 in the training and validation cohorts. Decision curve analysis revealed that the CEUS nomogram had better clinical applicability than the Gd-MRI nomogram when the threshold probability was between 0.35 and 0.95.

CONCLUSION

The CEUS-based nomogram was available for predicting MVI in HCC, and its predictive performance was not inferior to that of Gd-MRI.

Predictive machine learning model for microvascular invasion identification in hepatocellular carcinoma based on the LI-RADS system

Purposes

This study aimed to establish a predictive model of microvascular invasion (MVI) in hepatocellular carcinoma (HCC) by contrast-enhanced computed tomography (CT), which relied on a combination of machine learning approach and imaging features covering Liver Imaging and Reporting and Data System (LI-RADS) features.

Methods

The retrospective study included 279 patients with surgery who underwent preoperative enhanced CT. They were randomly allocated to training set, validation set, and test set (167 patients vs. 56 patients vs. 56 patients, respectively). Significant imaging findings for predicting MVI were identified through the Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression method. Predictive models were performed by machine learning algorithm, support vector machine (SVM), in the training set and validation set, and evaluated in the test set. Further, a combined model adding clinical findings to the radiologic model was developed. Based on the LI-RADS category, subgroup analyses were conducted.

Results

We included 116 patients with MVI which were diagnosed through pathological confirmation. Six imaging features were selected about MVI prediction: four LI-RADS features (corona enhancement, enhancing capsule, non-rim aterial phase hyperehancement, tumor size) and two non-LI-RADS features (internal arteries, non-smooth tumor margin). The radiological feature with the best accuracy was corona enhancement followed by internal arteries and tumor size. The accuracies of the radiological model and combined model were 0.725–0.714 and 0.802–0.732 in the training set, validation set, and test set, respectively. In the LR-4/5 subgroup, a sensitivity of 100% and an NPV of 100% were obtained by the high-sensitivity threshold. A specificity of 100% and a PPV of 100% were acquired through the high specificity threshold in the LR-M subgroup.

Conclusion

A combination of LI-RADS features and non-LI-RADS features and serum alpha-fetoprotein value could be applied as a preoperative biomarker for predicting MVI by the machine learning approach. Furthermore, its good performance in the subgroup by LI-RADS category may help optimize the management of HCC patients.

A deep learning model with incorporation of microvascular invasion area as a factor in predicting prognosis of hepatocellular carcinoma after R0 hepatectomy

INTRODUCTION

Microvascular invasion (MVI) is a known risk factor for prognosis after R0 liver resection for hepatocellular carcinoma (HCC). The aim of this study was to develop a deep learning prognostic prediction model by incorporating a new factor of MVI area to the other independent risk factors.

METHODS

Consecutive patients with HCC who underwent R0 liver resection from January to December 2016 at the Eastern Hepatobiliary Surgery Hospital were included in this retrospective study. For patients with MVI detected on resected specimens, they were divided into two groups according to the size of the maximal MVI area: the small-MVI group and the large-MVI group.

RESULTS

Of 193 patients who had MVI in the 337 HCC patients, 130 patients formed the training cohort and 63 patients formed the validation cohort. The large-MVI group of patients had worse overall survival (OS) when compared with the small-MVI group (p = 0.009). A deep learning model was developed based on the following independent risk factors found in this study: MVI stage, maximal MVI area, presence/absence of cirrhosis, and maximal tumor diameter. The areas under the receiver operating characteristic of the deep learning model for the 1-, 3-, and 5-year predictions of OS were 80.65, 74.04, and 79.44, respectively, which outperformed the traditional COX proportional hazards model.

CONCLUSION

The deep learning model, by incorporating the maximal MVI area as an additional prognostic factor to the other previously known independent risk factors, predicted more accurately postoperative long-term OS for HCC patients with MVI after R0 liver resection.