Predicting overall survival of patients with hepatocellular carcinoma using a three-category method based on DNA methylation and machine learning

Rui-Zhao Dong; Xuan Yang; Xin-Yu Zhang; Ping-Ting Gao; Ai-Wu Ke; Hui-Chuan Sun; Jian Zhou; Jia Fan; Jia-Bin Cai; Guo-Ming Shi

doi:10.1111/jcmm.14231

Predicting overall survival of patients with hepatocellular carcinoma using a three-category method based on DNA methylation and machine learning

J Cell Mol Med. 2019 May;23(5):3369-3374. doi: 10.1111/jcmm.14231. Epub 2019 Feb 19.

Authors

Rui-Zhao Dong¹, Xuan Yang^{1

2}, Xin-Yu Zhang^{1

2}, Ping-Ting Gao^{1

2}, Ai-Wu Ke^{1

2}, Hui-Chuan Sun^{1

2}, Jian Zhou^{1

2

3}, Jia Fan^{1

2

3}, Jia-Bin Cai^{1

2}, Guo-Ming Shi^{1

2}

Affiliations

¹ Department of Liver Surgery and Transplantation, Liver Cancer Institute, Zhongshan Hospital, Fudan University, Shanghai, China.
² Key Laboratory of Carcinogenesis and Cancer Invasion of Ministry of Education, Shanghai, China.
³ Key Laboratory of Medical Epigenetics and Metabolism, Institutes of Biomedical Sciences, Fudan University, Shanghai, China.

Abstract

Hepatocellular carcinoma (HCC) is closely associated with abnormal DNA methylation. In this study, we analyzed 450K methylation chip data from 377 HCC samples and 50 adjacent normal samples in the TCGA database. We screened 47,099 differentially methylated sites using Cox regression as well as SVM-RFE and FW-SVM algorithms, and constructed a model using three risk categories to predict the overall survival based on 134 methylation sites. The model showed a 10-fold cross-validation score of 0.95 and satisfactory predictive power, and correctly classified 26 of 33 samples in testing set obtained by stratified sampling from high, intermediate and low risk groups.

Keywords: DNA methylation; hepatocellular carcinoma; machine learning.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Carcinoma, Hepatocellular / genetics*
Carcinoma, Hepatocellular / surgery
DNA Methylation / genetics*
Humans
Liver Neoplasms / genetics*
Liver Neoplasms / surgery
Machine Learning*
Models, Biological
Reproducibility of Results
Survival Analysis