Serum peptide profiles during progression of chronic hepatitis B virus infection to liver failure

Chronic hepatitis B virus (HBV)-infected patients with liver failure have a poor prognosis, and no satisfactory biomarkers are available for diagnosis before the end-stage. We explored serum peptide profiling for diagnosis and prediction of progression to liver failure in HBV-infected patients. Serum samples (164) from healthy subjects (n = 20), or subjects with chronic hepatitis B without cirrhosis and liver failure [chronic hepatitis B subjects without cirrhosis and liver failure (CHB); n = 33], with compensated liver cirrhosis (compensated liver cirrhosis (LC); n = 35), with acute-on-chronic liver failure [acute-on-chronic liver failure (ACLF); n = 38] or with chronic liver failure [chronic liver failure (CLF), n = 38] were applied to ClinProt magnetic beads, and bound peptides/proteins were analyzed by matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry. Our classification diagnostic models of liver disease were generated based on the Genetic Algorithm (GA) and Quick Classifier Algorithm (QC). Differentially expressed peptides were found among all test groups, with patterns of difference that readily distinguished between healthy and various HBV-associated liver disease samples. The model generated seven characteristic peptide peaks at 4053 m/z, 3506 m/z, 4963 m/z, 9289 m/z, 2628 m/z, 3193 m/z and 6432 m/z, giving overall predictive capability of 54.27%. Two-way comparisons of LC, ACLF or CLF vs CHB had predictive capabilities of 79.8%, 91.41% and 97.99%, respectively. Comparisons of ACLF or CLF vs LC were predictive at 87.72% and 82.18%, respectively and ACLF vs CLF was predictive at 75.05%. These classification diagnostic models generated by different peptide peaks were further validated in blinded tests with 67-100% accuracy. Serum peptide patterns vary during progression of chronic HBV infection to liver failure and may be used to distinguish different stages of the disease.