Correlation between CD105 expression and postoperative recurrence and metastasis of hepatocellular carcinoma

Abstract

Background: Angiogenesis is one of the mechanisms most critical to the postoperative recurrence and metastasis of hepatocellular carcinoma (HCC). Thus, finding the molecular markers associated with angiogenesis may help identify patients at increased risk for recurrence and metastasis of HCC. This study was designed to investigate whether CD105 or CD34 could serve as a valid prognostic marker in patients with HCC by determining if there is a correlation between CD105 or CD34 expression and postoperative recurrence or metastasis.

Methods: Immunohistochemical staining for the CD105, CD34 and vascular endothelial growth factor (VEGF) antibodies was performed in 113 HCC tissue specimens containing paracarcinomatous tissue and in 14 normal liver tissue specimens. The quantitation of microvessels identified by anti-CD105 and anti-CD34 monoclonal antibodies and the semiquantitation of VEGF expression identified by anti-VEGF monoclonal antibody were analyzed in conjunction with the clinicopathological characteristics of the HCC and any available follow-up information about the patients from whom the specimens were obtained.

Results: CD105 was not expressed in the vascular endothelial cells of any normal liver tissue or paracarcinomatous liver tissue but was expressed in the vascular endothelial cells of all HCC tissue. In contrast, CD34 was expressed in the vascular endothelial cells of normal liver tissue, paracarcinomatous tissue, and HCC tissue in the following proportions of specimens: 86.7%, 93.8%, and 100%, respectively. The microvascular densities (MVDs) of HCC determined by using an anti-CD105 mAb (CD105-MVD) and an anti-CD34 mAb (CD34-MVD), were 71.7 +/- 8.3 (SD) and 106.3 +/- 10.4 (SD), respectively. There was a significant correlation between CD105-MVD and CD34-MVD (r = 0.248, P = 0.021). Although CD34-MVD was significantly correlated with VEGF expression (r = 0.243, P = 0.024), CD105-MVD was more closely correlated (r = 0.300, P= 0.005). The correlation between microscopic venous invasion and CD105-MVD, but not CD34-MVD, was also statistically significant (r = 0.254, P = 0.018). Univariate analysis showed that CD105-MVD was significantly correlated with the 2-year overall survival rate (P = 0.014); CD34-MVD was not (P = 0.601). Multivariate analysis confirmed that CD105-MVD was an independent prognostic factor and that CD34-MVD was not.

Conclusion: The anti-CD105 mAb is an ideal instrument to quantify new microvessels in HCC as compared with anti-CD34 mAb. CD105-MVD as compared with CD34-MVD is relevant a significant and independent prognostic indicator for recurrence and metastasis in HCC patients.

Figure 1

A, Positive staining for CD105 in HCC tissue (avidin-biotin-peroxidase complex, original magnification × 200). B, Negative staining for CD105 in paracarcinomatous liver tissue (avidin-biotin-peroxidase complex, original magnification × 200). C, Negative staining for CD105 in normal liver tissue (avidin-biotin-peroxidase complex, original magnification × 200). D, Positive staining for CD34 in HCC tissue (avidin-biotin-peroxidase complex, original magnification × 200). E, Positive staining for CD34 in paracarcinomatous liver tissue (avidin-biotin-peroxidase complex, original magnification × 200). F, Positive staining for CD34 in normal liver tissue (avidin-biotin-peroxidase complex, original magnification × 200).

Figure 2

Correlation between CD34-MVD and CD105-MVD.

Figure 3

A, correlation between VEGF expression and CD34-MVD. B, correlation between VEGF expression and CD105-MVD.

Figure 4

Postoperative overall survival rate of patients with HCC. The comparison shown is according to the MVD determined with an anti-CD105 mAb A or an anti-CD34 mAb B.