Identification of a liver cirrhosis signature in plasma for predicting hepatocellular carcinoma risk in a population-based cohort of hepatitis B carriers.

Liver cirrhosis is a critical state in the natural course of hepatocellular carcinoma (HCC). We sought to investigate the potential of in-depth proteomics to reveal plasma protein signatures that reflect common networks/pathways of liver cirrhosis, and to determine whether the cirrhosis-related signature in plasma is linked to the development of HCC among hepatitis B virus (HBV) carriers. We first compared plasma protein profiles using a 174-antibody microarray system between three groups of HBV carriers with different Child's grades of cirrhosis, which revealed a panel of 45 differentially expressed proteins with a high accuracy for discriminating Child's B/C. Ingenuity Pathway Analysis identified two main up-regulated networks connecting the 45 proteins that were most enriched for genes in the pathway of hepatic stellate cell activation. A parsimonious subset of 11 pathway-based proteins was then selected for quantification to correlate with HCC risk among 49 HCC cases and 50 controls in a nested case-control study within a 16-yr follow-up cohort of HBV carriers. A high risk score derived from a principal component analysis, which was used to extract the cluster structure of the 11 proteins, was associated with HCC (odds ratio = 4.83, 95% confidence interval: 1.26-18.56) even after adjustment for viral and clinical variables, implying the involvement of a pattern of coordinated proteins. Stepwise logistic regression on the 11 proteins revealed ICAM-2 as an independent predictor for HCC. These findings may give further insight into the pathobiology of hepatocarcinogenesis, allow testing of the cirrhosis-related plasma protein signature as a potential predictive biomarker for HCC.