Toward the proteomic identification of biomarkers for the prediction of HBV related hepatocellular carcinoma

Early detection is a key step for effective intervention of hepatocellular carcinoma (HCC), the lack of sensitive and specific biomarkers is a major reason for the high rate of HCC-related mortality. This report described an integrated strategy by combining SELDI-ProteinChip, sophisticated algorithm analysis, acetonitrile (ACN) pre-treatment and two-dimensional electrophoresis (2DE)-peptide mass fingerprinting (PMF) techniques to identify serological markers for the prediction of HBV-related HCC. Proteomic profiling of three groups of serum specimens from HBV-related HCC (50 cases), HBV infection (45 cases), and normal subjects (30 cases) was conducted by using SELDI-ProteinChip system and the resulting different protein peaks were subjected to stepwise statistical analyses. Three most discriminatory peaks at 5890, 11615, and 11724 Da, respectively, were screened out from the statistical algorithm and a predictive model based on the three peaks was constructed and tested using the newly enrolled serum samples. 2DE was applied to separate and compare the serum samples that were pre-treated by ACN precipitation. The protein spots obviously intensified in HCC sera in the 2DE region of 12 kDa were identified by PMF to be serum SAA, which was validated by SELDI-TOF spectra of HCC sera after immunoprecipitation using anti-SAA antibody and by Western blot experiments. Given the fact that SAA is not a specific biomarker, further attempt is being made to identify the other two most discriminatory peaks to realize the possibility of using the predictive model for HCC surveillance and prediction.