Endocan expression and localization in human glioblastomas.

Glioblastomas (GBMs) are highly malignant tumors characterized by microvascular proliferation and the pseudopalisading pattern of necrosis. Investigations have, therefore, focused on vascular and endothelial cell biology in GBM. Endocan, also called endothelial cell-specific molecule-1, is a proteoglycan that is secreted by endothelial cells and upregulated by proangiogenic factors. We found that endocan is not only expressed in vitro by endothelial cells but also in the T98G and U118MG human GBM cell lines. In U118MG cells, tumor necrosis factor and fibroblast growth factor 2 upregulated endocan production, whereas exposure to hypoxia or cobalt chloride, an inducer of hypoxia inducible factor 1, increased endocan release without affecting cell viability. Endocan expression in 82 brain tumors was studied by immunohistochemistry. Endocan immunoreactivity was detected in hyperplastic endothelial cells in high-grade gliomas, mostly at the tumor margins; endothelial cells were mostly endocan negative in low-grade gliomas, and it was never detected in the cerebral cortex distant from the tumors. Tumor cells in high-grade but not low-grade gliomas also expressed endocan, and it was detected in palisading cells surrounding areas of necrosis in GBM. Endothelial cell endocan immunoreactivity also correlated with shorter survival in glioma patients. Taken together, these results suggest that endocan is associated with abnormal vasculature in high-grade gliomas.

Low HER2-expressing glioblastomas are more often secondary to anaplastic transformation of low-grade glioma.

BACKGROUND: Anti-Human Epithelial Receptor Type 2 (HER2) antibodies have the ability to induce in vitro apoptosis of glioblastoma (GBM) cells. This study was designed to evaluate the variability of HER2 expression in GBM and its role as a possible prognosis factor. METHODS: Data of 57 patients with GBM and 16 patients with grade III gliomas were retrospectively analyzed. The expression of HER2 was determined by immunohistochemistry and intensity was noted from 0+ to 3+. We compared the HER2 expression in de novo GBM and in GBM resulting from anaplastic transformation of low-grade glioma ("secondary GBM"). Statistical analysis was performed using univariate analysis and the Kaplan-Meier method. FINDINGS: All GBM expressing highly HER2 (2+ and 3+) were de novo GBM. All secondary GBM expressed HER2 with low intensity (0+ and 1+). Survival time was significantly longer when HER2 expression was low (Log Rank test P = 0.04). The patterns of HER2 expression were similar between grade III gliomas and secondary GBM. CONCLUSIONS: To our best knowledge, our study showed for the first time a significant association between HER2 expression and the type of GBM, with subsequent influence on survival rate. GBM with low-HER2 expression are more likely to be secondary GBM, carrying a better prognosis than de novo GBM.