Microvessel density assessment in benign and malignant endometrial changes

Tumor angiogenesis is believed to be a prognostic indicator associated with tumor growth and metastasis. Microvessel density (MVD) assessment with common endothelial markers such as CD34 has been found to influence prognosis among endometrial carcinoma patients. The CD105/endoglin antibody has been reported to preferentially bind to proliferated endothelial cells in tissues participating in angiogenesis. The aim of this study was to evaluate the quantification of angiogenesis by assessing MVD in endometrial lesions when comparing the performance of anti-CD34 and anti-CD105 in women with benign and malignant endometrial changes. The study included 58 women (37 postmenopausal) with normal, hyperplastic and malignant endometrium in which preoperative transvaginal sonography was performed. Histological results of the removed endometrium were correlated with MVD assessed in "hot areas" where high densities of microvessels were detected within tumoral tissue. Endometrial cancer was confirmed in 37 women (3 premenopausal). Benign hyperplasia (14 cases), secretory or proliferative endometrium (5 cases) or endometrial atrophy (2 cases) was found in the remaining women. Malignant changes were mostly noted as FIGO stage I and II (28 cases) and had a low (1 or 2) histological grade (29 cases). Median MVD's assessed with CD105 and CD34 were 10.4 and 32.3, respectively. Median MVD assessed with CD34 was almost twice higher in women with endometrial cancer than in women with benign endometrium (CD34 MVD = 41.8 vs. 27.6, p=0.004). In cases of CD105 MVD significant differences between women with benign and malignant endometrial changes were also found (CD105 MVD = 11.8, vs. 6.4; p=0.00007). The menopausal status, but not the clinical stage or histological grading was significantly correlated with both CD34 MVD (p=0.02) and CD105 MVD (p=0.0003). A significant correlation was also found between CD34 and CD105 measured MVD (p=0.000001). In conclusion, transition from endometrial hyperplasia to endometrial cancer appears to be accompanied by microvessel density changes. MVD assessed with both CD34 and CD105 antibodies could be used as a potential prognostic factor in women with endometrial cancer. Our study showed that endoglin, by staining the proliferating microvessels could be more specific and sensitive marker for tumor neoangiogenesis than the more commonly used marker, CD34.

Microvessel density and CpG island methylation of the THBS2 gene in malignant ovarian tumors

We aimed to investigate the role of thrombospondin-2 (THBS2) related angiogenic activity in malignant ovarian tumors and to determine if aberrant methylation associated inactivation is involved in down-regulating THBS2 expression in ovarian cancer. The methylation status of the THBS2 promoter region and microvessel density (MVD) was studied in 70 malignant ovarian tumors and in 15 control ovarian samples. A methylation specific PCR (MSP) method was used to distinguish methylated from unmethylated DNA in the promoter regions of the THBS2 gene. MVD was assessed with anti-CD34 antibodies and the results were compared between tumors with average (AVD) and high (HVD) microvessel density. Alterations in the expression of trombospondin-2 were more often seen in early (FIGO stage I and II ) than in late stage tumors (66% vs. 30%, p=0.01). Age, menopausal status, the histological type and tumor grade did not correlate with thrombospondin-2 expression, however, silencing of THBS2 gene was more often seen in higher rather than in lower grade (50% vs. 28%) cancers and in nonserous rather than in serous (43% vs. 32%) tumors. In 81% of THBS2 mRNA-negative tumors, ahypermethylated promoter region of THBS2 was found (p=0.00003). An unmethylated product of the MSP reaction was more often detected in high grade tumors (93% vs. 76%, p=0.04). The incidence of THBS2 hypermethylation was not related to the tumor histological type, but unmethylated THBS2 was more often found in serous rather than in nonserous tumor (96% vs. 74%, p=0.01). The median MVD in malignant the tumor samples was 21,7 (range: 7.6-55.2). In the group with HVD, 54% were THBS2 mRNAnegative, conversely, in the group with AVD tumors only 26% of the cases had undetectable THSB2 mRNA. A significant correlation between microvessel density and the expression of trombospondin-2 (p=0.009) was found. In the samples with HVD, 51% had hypermethylated THBS2, however methylation pattern had no significant influence on microvessel density. In conclusion, hypermethylation might be responsible for altered expression of thrombospondin-2 in ovarian cancer. The THSB2 methylation pattern had no significant influence on microvessel density.