The aim of this study is to identify candidate biomarkers for the detection of hepatocellular carcinoma (HCC) through pre-cancerous gene expression analysis in an HBx transgenic mouse model. The gene expression profiles of liver and tumor tissues from 6-, 12- and 18-month-old HBx transgenic and littermate control mice were monitored using DNA oligo microarrays. Data analysis revealed changes in 684 genes in the tumor tissues from HBx transgenic mice. Based on their pre-cancerous expression profiles, two separate gene groups, corresponding to HCC specific and non-specific groups respectively, were identified. Gene ontology analysis identified 47 genes encoding secretory or transmembrane proteins among 155 upregulated genes in the HCC-specific group. Among these, four genes encoding TFF3, IGF2, LPL and SPP1 were found to be promising biomarker candidates for the detection of HCC as compared with AFP. This work describes a new way to identify novel biomarker candidates for HCC diagnosis.