3 years ago

Predictive model for inflammation grades of chronic hepatitis B: Large-scale analysis of clinical parameters and gene expressions

Jingyi Hu, Menghan Zhang, Jiucun Wang, Jiming Zhang, Xinxin Zhang, Xiaonan Zhang, Li Jin, Zhanqing Zhang, Lijun Wu, Jun Zhang, Yitong Zhang, Yida Pan, Yi Wang, Zhenghong Yuan, Hai Li, Yanyun Ma, Yi Li, Weichen Zhou, Jie Liu, Lungen Lu
Background Liver biopsy is the gold standard to assess pathological features (eg inflammation grades) for hepatitis B virus-infected patients although it is invasive and traumatic; meanwhile, several gene profiles of chronic hepatitis B (CHB) have been separately described in relatively small hepatitis B virus (HBV)-infected samples. We aimed to analyse correlations among inflammation grades, gene expressions and clinical parameters (serum alanine amino transaminase, aspartate amino transaminase and HBV-DNA) in large-scale CHB samples and to predict inflammation grades by using clinical parameters and/or gene expressions. Methods We analysed gene expressions with three clinical parameters in 122 CHB samples by an improved regression model. Principal component analysis and machine-learning methods including Random Forest, K-nearest neighbour and support vector machine were used for analysis and further diagnosis models. Six normal samples were conducted to validate the predictive model. Results Significant genes related to clinical parameters were found enriching in the immune system, interferon-stimulated, regulation of cytokine production, anti-apoptosis, and etc. A panel of these genes with clinical parameters can effectively predict binary classifications of inflammation grade (area under the ROC curve [AUC]: 0.88, 95% confidence interval [CI]: 0.77-0.93), validated by normal samples. A panel with only clinical parameters was also valuable (AUC: 0.78, 95% CI: 0.65-0.86), indicating that liquid biopsy method for detecting the pathology of CHB is possible. Conclusions This is the first study to systematically elucidate the relationships among gene expressions, clinical parameters and pathological inflammation grades in CHB, and to build models predicting inflammation grades by gene expressions and/or clinical parameters as well.

Publisher URL: http://onlinelibrary.wiley.com/resolve/doi

DOI: 10.1111/liv.13427

You might also like
Never Miss Important Research

Researcher is an app designed by academics, for academics. Create a personalised feed in two minutes.
Choose from over 15,000 academics journals covering ten research areas then let Researcher deliver you papers tailored to your interests each day.

  • Download from Google Play
  • Download from App Store
  • Download from AppInChina

Researcher displays publicly available abstracts and doesn’t host any full article content. If the content is open access, we will direct clicks from the abstracts to the publisher website and display the PDF copy on our platform. Clicks to view the full text will be directed to the publisher website, where only users with subscriptions or access through their institution are able to view the full article.