Although serum miR-122 was also elevated in HBV patients with HCC comparing with those without HCC, the difference was at the border line. Serum miR-122 yielded an AUC of 0.869 for discriminating HCC from healthy subjects and only 0.630 for discriminating HBV patients with HCC from those without HCC. At a cut-off value of 0.474, the sensitivity was 81.6% and the specificity was 83.3% in discriminating HCC from healthy subjects, and 77.6% sensitivity and 57.8% specificity in discriminating HBV patients with HCC from those without HCC at the cut-off value of 0.651. More importantly, it was found that the levels of miR-122 were significantly reduced in the post-operative serum samples when compared to the pre-operative samples, reaching levels comparable with healthy subjects, indicating that the elevation of serum miR-122 is likely derived from HCC. MiR-122 not only is evolutionary conserved across species and but also was identified as the most abundant liver specific miRNA constituting 70% of total hepatic miRNAs while cloning small RNAs from different tissues in mice. MiR-122 facilitates replication and translation of hepatitis C viral RNA and positively regulates cholesterol and triglyceride level. Significantly, the down-regulation of miR-122 was detected in more than 70% of HCC. It was shown that the level of miR122 expression increases in the mouse liver throughout development, to reach the maximum just Ibrutinib before birth. Thus, the loss of expression of miR-122 of HCC cells may represent either a differentiation reversion or a block to a less differentiated status of liver cells. In our study, it appears contrary and unexpected that the levels of miR-122 are elevated in serum of HCC patients. Our results showed that the elevated serum miR-122 is presented not only in HBV patients with HCC but also in HBV patients without HCC, suggesting that the elevated miR-122 in the serum of patients may also reflect liver injury. Hepatocytes contain abundant miR-122 and damage of hepatocytes caused by inflammation due to virus infection or cancer would be expected to release significant amount of this miRNA into the circulation. Because serum miRNAs have been shown to be very stable, miRNAs leaked from damaged hepatocytes would accumulate in blood to a high level. This might explain why miR-122 is downregulated in HCC tissues but elevated in serum of HBV patients without or with HCC. Interestingly, our data indicated that expression levels of miR-122 in serum were significantly higher in HCC patients than disease controls or healthy controls, while Xu et al. showed that expression levels of serum miR-122 were significantly higher in HBV patients than HCC or healthy controls. The reason may be that we and Xu et al. use the different normalization control. MiR-223 is one of the miRNAs that has been given much attention in the literature. In addition to this, a recent study observed that miR-223 was commonly repressed in HCC, suggesting a potential role of this miRNA in liver disease. In our study, levels of miR-223 were significantly elevated in HCC patients than in healthy controls, while no significant difference was observed for this miRNA between HBV subjects with and without HCC.