Skip to main content

Archived Comments for: Correlation between CD105 expression and postoperative recurrence and metastasis of hepatocellular carcinoma

Back to article

  1. CD105 (endoglin) expression: an unfavorable prognostic marker in hepatocellular carcinoma?

    Koert de Jong, Departments of HPB Surgery and Pathology, University Medical Center Groningen, The Netherlands

    22 June 2007

    Koert P. de Jong and Annette S.H. Gouw

    Working in the field of angiogenesis in hepatocellular carcinoma we read with interest the paper by Yang et al (BMC Cancer 2006, 6:110). They concluded that CD105 is superior to CD34 as a marker for new microvessels in HCC and that CD105 is a better prognostic indicator for recurrence and metastasis of HCC. These conclusions were based on three major findings. After partial liver resection for HCC there was a significantly lower patient survival if their tumors contained a higher microvessel density (MVD), quantified by CD105 expression. However, the authors did not find a significant influence on patient survival when tumor MVD was quantified by CD34. Secondly, CD105-MVD correlated with venous invasion whereas CD34-MVD did not. And thirdly, in contrast with CD34, CD105 vascular expression was not present in normal and paracarcinomatous liver tissue.

    Several papers evaluated MVD in resected HCC using CD34 as the endothelial cell marker. Most of these papers (1-3) describe a poor outcome with high MVD expression, one (4) a favorable outcome, and one paper (5) did not find a difference in survival. One of these papers failed to identify CD105 expression as a prognostic marker (5).

    We would greatly appreciate if Yang and colleagues could address the following items. First, the survival curves of the low versus high CD105 MVD expression strongly suggests that in the high MVD patient group the early postoperative mortality is higher than in the low MVD group. The early drop in the line of the survival curve can very likely not be explained by deaths due to tumor recurrences, because it is difficult to conceive that death due to tumor recurrence occurs so early after partial liver resection. So, it is suggested that the significant difference in survival is explained by a coincidental higher early death rate in the high MVD group as compared to that in the low MVD group. How is the survival when the early postoperative (not tumor recurrence death related) mortality is not taken into account. Or, alternatively, is the early postoperative death rate comparable in the high and low CD105 MVD groups?

    The staining pattern of CD105 is pivotal for the correlation studies of this paper. The results concerning the expression of CD105 in non carcinomatous tissue are discordant with several other studies. CD105 vascular expression has been documented both in normal livers and in chronic viral hepatitis in which there was an increased expression (6,7). Yang et al. did not mention the underlying liver disease of their patient groups. If these were HCC’s arising in chronic viral hepatitis, the total absence of CD105expression could represent a false negative result, which could very well be due to differences in techniques/antibodies. The latter is much less a problem with CD34 because practically all studies applied the same clone (QBEnd /10). From a pathogenetic point of view it is also hard to conceive that the paracancerous tissue is totally devoid of CD105 expression because liver tissue adjacent to tumor is rarely normal. More frequently it contains substantial amount of inflammatory cells evoking phenotypic changes on endothelial cells. Moreover, CD105 expression has also been described in activated hepatic stellate cells and portal myofibroblasts which should be present in paracancerous tissue. Subsequently, it might well be that the differences between CD34 and CD105 expression described in this study are not as prominent; which might weaken the statistical correlation, e.g. with venous invasion. The more so since the other tested parameters did not show differences between CD105 and CD34.

    Another important issue is the essential oncogenetic difference between HCC and other solid tumors in which many other CD105 studies were performed and thus cited. One should bear in mind that the vast majority of HCC develop in inflamed and scarred liver tissue in contrast with other solid tumors. Hepatitic and cirrhotic livers are known to have undergone angiogenetic processes without tumor development, thus an increased CD105 expression in these livers apart from tumor angiogenetic activity would not be an unexpected finding.

    We would appreciate the elucidation of these items by Yang and colleagues because it will contribute to a better understanding of angiogenesis in hepatocellular carcinoma.

    Reference List

    1. Tanigawa N, Lu C, Mitsui T, Miura S. Quantitation of sinusoid-like vessels in hepatocellular carcinoma: its clinical and prognostic significance. Hepatology 1997;26:1216-1223.

    2. Sun HC, Tang ZY, Li XM, Zhou YN, Sun BR, Ma ZC. Microvessel density of hepatocellular carcinoma: its relationship with prognosis. J Cancer Res Clin Oncol 1999;125:419-426.

    3. Poon RT, Ng IO, Lau C, Yu WC, Yang ZF, Fan ST, Wong J. Tumor microvessel density as a predictor of recurrence after resection of hepatocellular carcinoma: a prospective study. J Clin Oncol 2002;20:1775-1785.

    4. Nanashima A, Yano H, Yamaguchi H, Tanaka K, Shibasaki S, Sumida Y, Sawai T, Shindou H, Nakagoe T. Immunohistochemical analysis of tumor biological factors in hepatocellular carcinoma: relationship to clinicopathological factors and prognosis after hepatic resection. J Gastroenterol 2004;39:148-154.

    5. Ho JW, Poon RT, Sun CK, Xue WC, Fan ST. Clinicopathological and prognostic implications of endoglin (CD105) expression in hepatocellular carcinoma and its adjacent non-tumorous liver. World J Gastroenterol 2005;11:176-181.

    6. Clemente M, Nunez O, Lorente R, Rincon D, Matilla A, Salcedo M, Catalina MV, Ripoll C, Iacono OL, Banares R, Clemente G, Garcia-Monzon C. Increased intrahepatic and circulating levels of endoglin, a TGF-beta1 co-receptor, in patients with chronic hepatitis C virus infection: relationship to histological and serum markers of hepatic fibrosis. J Viral Hepat 2006;13:625-632.

    7. Theuerkauf I, Zhou H, Fischer HP. Immunohistochemical patterns of human liver sinusoids under different conditions of pathologic perfusion. Virchows Arch 2001;438:498-504.

    Competing interests

    No competing interest