Cancer is the leading cause of death in the world. Development of minimally invasive biomarkers for early detection of cancer is urgently needed to reduce high morbidity and mortality associated with malignancy. MicroRNAs (miRNAs) are small regulatory RNAs that modulate the activity of specific mRNA targets and play important roles in a wide range of physiologic and pathologic processes. Recently, miRNAs were found to be dysregulated in a variety of diseases including cancer. Emerging evidence suggests that miRNAs are involved in tumor initiation and progression. Together, the different expression profiles of miRNAs in cancer, and the stability of circulating miRNAs, make them new potentially clinical biomarkers for cancer diagnosis, classification, therapeutic decisions, and prognosis.

Background: Endoglin (CD105) has been considered a prognostic marker for hepatocellular carcinoma (HCC), and widely used as an appropriate targeting for antiangenesis therapy in some cancers. Our aim was to evaluate the distribution and expression of CD105 in the liver of patients with HCC, and to discuss whether CD105 may be used as an appropriate targeting for antiangenesis therapy in HCC. Methods: Three parts of liver tissues from each of 64 patients with HCC were collected: tumor tissues (TT), adjacent non-tumor (AT) liver tissues within 2 cm, and tumor free tissues (TF) 5 cm far from the tumor edge. Liver samples from 8 patients without liver diseases served as healthy controls (HC). The distribution and expression of CD105 in tissues were evaluated by immunohistochemistry, Western blotting analysis, and real-time PCR. HIF-1alpha and VEGF₁₆₅ protein levels in tissues were analyzed by Immunohistochemistry and Western blotting analysis or ELISA. Results: CD105 was positively stained mostly in a subset of microvessels 'endothelial sprouts' in TT of all patients while CD105 showed diffuse positive staining, predominantly on hepatic sinus endothelial cells in the surrounding of draining veins in TF and AT. The mean score of MVD-CD105 (mean ± SD/0.74 mm²) was 19.00 ± 9.08 in HC, 153.12 ± 53.26 in TF, 191.12 ± 59.17 in AT, and 85.43 ± 44.71 in TT, respectively. Using a paired t test, the expression of CD105 in AT and TF was higher than in TT at protein (MVD, p = 0.012 and p = 0.007, respectively) and mRNA levels (p < 0.001 and p = 0.009, respectively). Moreover, distribution and expression of CD105 protein were consistent with those of HIF-1alpha and VEGF₁₆₅ protein in liver of patients with HCC. The level of CD105 mRNA correlated with VEGF₁₆₅ level in TF (r = 0.790, p = 0.002), AT (r = 0.723, p < 0.001), and TT (r = 0.473, p = 0.048), respectively. Conclusion: It is demonstrated that CD105 was not only present in neovessels in tumor tissues, but also more abundant in hepatic sinus endothelium in non-tumor tissues with cirrhosis. Therefore, CD105 may not be an appropriate targeting for antiangenesis therapy in HCC, especially with cirrhosis.

Our previous investigations showed that mobilized endothelial progenitor cells (EPCs) are enriched in non-tumor tissues (NT) surrounding hepatocellular carcinoma (HCC), compared to in tumor tissues (TT). This particular recruitment of EPCs is worth investigating further. The mobilization, recruitment, homing, and incorporation of EPCs into tumors require the participation of multiple factors, including angiogenic factors, adherent molecules, endothelial cells, hypoxic environment, etc. Therefore, we hypothesized that NT might be a hypoxic and highly angiogenic area, into which many more EPCs are recruited and homed. In the last three years, we evaluated the hypoxic condition, angiogenic factors and angiogenic index using frozen tissues or tissue microarrays from 105 patients who had undergone hepatectomy for HCC, and here we review our results and the studies of others. All results showed the expression of Hypoxia-inducible factor-1α was higher in NT than in TT. The expression of VEGFA, bFGF, TGF-β, MCP-1, MMP-9, TIMP-2, and endostatin in NT was significantly higher than in normal liver and TT. Meanwhile, the expression of CD105—the surface marker of activated endothelial cells—was also higher in NT than in TT at the protein and mRNA levels. These investigations showed that NT is a hypoxic and highly angiogenic area, which may be the ‘niche’ of EPCs. The particular background in HCC may be related to liver cirrhosis. Therefore, non-tumor tissues surrounding HCC may be the ‘niche’ of endothelial progenitor cells.

Background: We investigated the distribution and clinical significance of mobilized endothelial progenitor cells (EPCs) in hepatocellular carcinoma (HCC). We found that many more EPCs were recruited to nonmalignant liver tissue (especially into adjacent non-tumor tissues (AT)) than to tumor vessels. These results suggest that the mechanism underlying the recruitment of EPCs into microvessels in AT merits further investigation Methods: Angiogenic factors were detected in three tissue microarrays comprising normal liver, paired tumor tissue (TT) and AT from 105 patients (who had undergone hepatectomy for HCC) using immunohistochemistry. Also, the number of EPCs (positive for Sca-1, Flk-1 and c-Kit) in the blood and liver of cirrhotic mice were determined by flow cytometry and immunohistochemistry. The distribution of these labeled EPCs in tumor and non-tumor tissues was then studied. Results: The results from the tissue microarrays showed that the expression levels of VEGF-A, bFGF, TGF-β, MCP-1, TSP-1, MMP-9, TIMP-2, and endostatin were significantly higher in AT than in either normal liver or TT (p < 0.05), but no significant difference was found in the expression levels of COX-2 and NOS-2 between AT and TT. The expression of VEGF-A, bFGF, TGF-β, MCP-1, TSP-1, MMP-9, TIMP-2, endostatin, COX-2, and NOS-2 in normal liver tissue was weaker than that in AT or TT. In cirrhotic mice, the number of circulating endothelial progenitor cells gradually increased, before decreasing again. In this mouse model, increased numbers of EPCs were recruited and homed specifically to the cirrhotic liver. Conclusions: Both liver cirrhosis and HCC led to increased expression of pro-angiogenic factors, which resulted in the recruitment of EPCs into AT. Also, EPCs were mobilized, recruited and homed to cirrhotic liver. The unique pathology of HCC coupled with liver cirrhosis may, therefore, be associated with the distribution and function of EPCs.