Elsevier

Journal of Hepatology

Volume 70, Issue 6, June 2019, Pages 1133-1144
Journal of Hepatology

Research Article
Radiomic analysis of contrast-enhanced CT predicts microvascular invasion and outcome in hepatocellular carcinoma

https://doi.org/10.1016/j.jhep.2019.02.023Get rights and content

Highlights

  • We identified 8 MVI preoperative risk factors in HCC, including radiomic features.

  • Radiomic features do not provide significant added value to radiologist scores.

  • A model integrating clinic-radiologic and radiomic features demonstrates good performance for predicting MVI.

Background & Aims

Microvascular invasion (MVI) impairs surgical outcomes in patients with hepatocellular carcinoma (HCC). As there is no single highly reliable factor to preoperatively predict MVI, we developed a computational approach integrating large-scale clinical and imaging modalities, especially radiomic features from contrast-enhanced CT, to predict MVI and clinical outcomes in patients with HCC.

Methods

In total, 495 surgically resected patients were retrospectively included. MVI-related radiomic scores (R-scores) were built from 7,260 radiomic features in 6 target volumes. Six R-scores, 15 clinical factors, and 12 radiographic scores were integrated into a predictive model, the radiographic-radiomic (RR) model, with multivariate logistic regression.

Results

Radiomics related to tumor size and intratumoral heterogeneity were the top-ranked MVI predicting features. The related R-scores showed significant differences according to MVI status (p <0.001). Regression analysis identified 8 MVI risk factors, including 5 radiographic features and an R-score. The R-score (odds ratio [OR] 2.34) was less important than tumor capsule (OR 5.12), tumor margin (OR 4.20), and peritumoral enhancement (OR 3.03). The RR model using these predictors achieved an area under the curve (AUC) of 0.909 in training/validation and 0.889 in the test set. Progression-free survival (PFS) and overall survival (OS) were significantly different between the RR-predicted MVI-absent and MVI-present groups (median PFS: 49.5 vs. 12.9 months; median OS: 76.3 vs. 47.3 months). RR-computed MVI probability, histologic MVI, tumor size, and Edmondson-Steiner grade were independently associated with disease-specific recurrence and mortality.

Conclusions

The computational approach, integrating large-scale clinico-radiologic and radiomic features, demonstrates good performance for predicting MVI and clinical outcomes. However, radiomics with current CT imaging analysis protocols do not provide statistically significant added value to radiographic scores.

Lay summary

The most effective treatment for hepatocellular carcinoma (HCC) is surgical removal of the tumor but often recurrence occurs, partly due to the presence of microvascular invasion (MVI). Lacking a single highly reliable factor able to preoperatively predict MVI, we developed a computational approach to predict MVI and the long-term clinical outcome of patients with HCC. In particular, the added value of radiomics, a newly emerging form of radiography, was comprehensively investigated. This computational method can enhance the communication with the patient about the likely success of the treatment and guide clinical management, with the aim of finding drugs that reduce the risk of recurrence.

Introduction

Hepatocellular carcinoma (HCC) is one of the most common primary hepatic malignant tumors and its incidence is increasing worldwide.1 It is the second leading cause of cancer-specific mortality in the Asia-Pacific regions, and especially in China.2 Surgical resection and liver transplantation (LT) are potentially curative for patients with HCC,3 but recurrence after surgical treatment is common. Some studies maintained that approximately 70% of patients would suffer from recurrence within 5 years after surgical resection, and 35% after LT.[4], [5], [6], [7], [8]

Microvascular invasion (MVI) is one of the most important prognostic factors for HCC after surgical treatment.[9], [10], [11] Contrary to macrovascular invasion, which can be detected with diagnostic imaging, MVI is a histologic finding that can only be postoperatively diagnosed with a surgical specimen.12 Preoperative prediction of MVI is still challenging. A variety of imaging findings have been described, with variable diagnostic utility. Previous studies found that imaging features such as tumor size, multinodular tumor morphology, tumor margins, and peritumoral enhancement were associated with MVI.[13], [14], [15] In addition, Renzulli et al. showed that a 2-trait predictor of venous invasion can be a useful preoperative predictor of MVI.14 Banerjee et al. showed that a radio-genomic venous invasion (RVI) predictor, based on the association between imaging features and gene expression, achieves high accuracy in predicting MVI in HCC.8 However, these criteria for a preoperative imaging diagnosis of MVI in HCC have not yet been widely recognized.

Radiomics is a newly emerging form of imaging analysis using a series of data-mining algorithms or statistical analysis tools on high-throughput imaging features to obtain predictive or prognostic information. By building appropriate models with refined features, it achieved successful assessment and prediction abilities in various challenging clinical tasks.[16], [17], [18], [19] A landmark study in colorectal cancer revealed clear associations between CT radiomics and lymph node metastases, and a combination of radiomic and clinico-radiologic factors could achieve significant clinical benefits.20 However, despite its potential, the use of radiomics as a clinical biomarker still necessitates amelioration and standardization.21 Greater integration between radiomics and other sources of data is required for clinicians to fully and confidently accept its role in patient management. To the best of our knowledge, only one study to date has assessed the prognostic aspect of radiomics for MVI in a group of 304 patients with HCC.22 However, stronger evidence is needed in support of the implications for tumor progression and the reliability of the methodology. Additionally, MVI concerns tumor edges, while in previous studies radiomic features were only extracted inside the tumor, where by definition there is no microvascular involvement. A more effective evaluation should focus on the radiomic features at the tumor periphery.

The purpose of this study was to investigate whether a computational approach integrating large-scale clinical and imaging modalities, especially radiomic features extracted from contrast-enhanced CT (CECT), could be useful to predict MVI and the long-term clinical outcomes in patients with HCC. Additionally, the added value provided by radiomics to evidence-based clinico-radiologic factors was investigated.

Section snippets

Patients

This retrospective study involved standard care performed at a single medical institution. Ethics committee approval was granted by the local institutional ethics review board, and the requirement for written informed consent was waived. All procedures involving human participants were performed in accordance with the 1975 Helsinki declaration and its later amendments.

We queried our institution’s medical records to derive all surgically confirmed cases of HCC between January 2009 and August

Basic clinico-radiologic characteristics

Out of 495 patients included, histologic MVI was diagnosed in explanted tissue of 149 patients (30.1%). Patients with MVI were younger, and had higher ALT, AST, PLT, INR and AFP levels than those without MVI. The 2 groups were similar in their distribution of sex, hepatic virus infection, cirrhosis, Child-Pugh stage, TB, CB, ALB, PT and Scr test. Risk coefficient estimated by univariate analysis is summarized in Table 1. HCCs with and without MVI demonstrated significantly different imaging

Discussion

The aim of this study was to investigate the prognostic aspects of computational-assisted models derived from large-scale clinical and imaging data, especially radiomic features from CECT, for preoperative prediction of histologic MVI status and clinical outcomes in a cohort of 495 patients with HCC. We concluded that CT tumor radiomic features, converted into quantitative R-scores, can be independent predictors of MVI, but less relevant than radiologist scores. A risk model integrating

Financial support

This work was supported by China Postdoctoral Fund (grant number 2015M580453, Z.Y.) and a Key Social Development Program for the Ministry of Science and Technology of Jiangsu Province (grant number BE2017756, Z.Y.).

Conflict of interest

The authors declare no conflicts of interest that pertain to this work.

Authors' contributions

Conception and design: Yu-Dong Zhang, Xi-Sheng Liu. Development of methodology: Xun Xu, Hai-Long Zhang, Qiu-Ping Liu, Shu-Wen Sun, Fei-Peng Zhu, Jing Zhang, Xu Yan, Guang Yang, Yu-Dong Zhang, Xi-Sheng Liu. Acquisition of data (acquired and managed patients, provided facilities, etc.): Xun Xu, Hai-Long Zhang, Qiu-Ping Liu, Shu-Wen Sun, Fei-Peng Zhu, Jing Zhang, Xu Yan, Guang Yang, Yu-Dong Zhang, Xi-Sheng Liu. Analysis and interpretation of data (e.g., statistical analysis, biostatistics,

Acknowledgement

We thank the Department of Pathology of the First Affiliated Hospital of Nanjing Medical University for assistance with histopathology.

References (31)

  • F. Kanwal et al.

    Potentially curative treatment in patients with hepatocellular cancer–results from the liver cancer research network

    Aliment Pharmacol Ther

    (2012)
  • A.E. Marshall et al.

    Tumor recurrence following liver transplantation for hepatocellular carcinoma: role of tumor proliferation status

    Liver Transplant

    (2010)
  • M.A. Zimmerman et al.

    Recurrence of hepatocellular carcinoma following liver transplantation: a review of preoperative and postoperative prognostic indicators

    Arch Surg

    (2008)
  • S. Banerjee et al.

    A computed tomography radiogenomic biomarker predicts microvascular invasion and clinical outcomes in hepatocellular carcinoma

    Hepatology

    (2015)
  • F. D'Amico et al.

    Predicting recurrence after liver transplantation in patients with hepatocellular carcinoma exceeding the up-to-seven criteria

    Liver Transplant

    (2009)
  • Cited by (427)

    View all citing articles on Scopus
    4

    Xun Xu and Hai-Long Zhang are co-first authors.

    5

    Equally contributed to the study.

    View full text