- Top of page
- Materials and methods
Meningiomas are slow-growing neoplasms that recur locally. Their morphologic grading does not always correlate with patient outcome. We evaluated the status of several immunohistochemical markers with histopathologic parameters in various grades of meningioma.Eighty-eight meningioma specimens were examined immunohistochemically to determine the status of Ki-67, cyclin D1, epidermal growth factor receptor (EGFR), cyclooxygenase-2 (COX-2), vascular endothelial growth factor (VEGF), and bcl-2. Several clinical and pathological parameters were investigated.Forty-nine Grade I, 33 Grade II, and 6 Grade III meningiomas were observed. VEGF and Ki-67 expression was correlated with higher tumor grade. The association between grade and other immunohistochemical markers expression was not significant. A correlation was observed between COX-2 expression and invasiveness to the brain or adjacent soft tissue. Tumor recurrence was correlated with brain or adjacent soft tissue invasion. We also observed a relationship between VEGF level and COX-2 expression, and they were both correlated with necrosis.Immunohistochemical evaluation of VEGF, COX-2, and Ki-67 expression can provide information regarding the behavior of meningiomas, particularly for cases in which histological grading is not straightforward.
Meningiomas are common among intracranial tumors and are estimated to account for 15–20% of all primary brain tumors , with an annual incidence of up to 13 per 100 000 persons . Ninety percent of meningiomas are classified as histologically benign and are typically curable after complete removal . A minority of these tumors present with clinical and pathologic features, suggesting aggressive clinical behavior, and are histologically classified as atypical (World Health Organization grade II) and malignant (WHO grade III) [2, 3]. However, even histologically benign meningiomas often recur, thus, their morphologic grading does not always correlate with patient outcome . These limitations of the current grading system for predicting tumor behavior have prompted a search for other approaches to identify more aggressive meningiomas.
Brain invasion is defined as irregular, tongue-like protrusions of tumor cells that infiltrate the parenchyma, without intervening of the meningeal tissue layer. Brain invasion is a powerful predictor of shorter recurrence-free survival; according to the WHO classification, otherwise benign meningiomas with brain invasion should be considered as WHO grade II . Invasion into adjacent tissues such as the skull, scalp, and even paranasal sinus does not alter the grade, but can make resection more difficult and shows a high recurrence rate and poor prognosis .
The prognostic significance of diverse proliferative indices in meningiomas has been assessed in various studies, and evaluation of the proliferative potential of tumors may be used as an ancillary technique in predicting the clinical course for these patients [6-9]. Cyclooxygenase-2 (COX-2), an enzyme that converts arachidonic acid into prostaglandins, plays a critical role in the development of various tumors including those of the colon, pancreatic, prostate, lung, and head and neck. This enzyme can stimulate gene transcription, tumoral growth, angiogenesis, and metastasis; inhibit apoptosis; and cause resistance to chemotherapy [10-12]. COX-2 overexpression is also frequently observed in brain tumors. Numerous studies have demonstrated the presence of COX-2 in gliomas, indicating the therapeutic effectiveness of COX inhibitors in gliomas [13, 14]. However, there are relatively few reports regarding COX-2 expression in meningiomas [10, 15, 16].
Increased vascular permeability and angiogenesis are essential for tumorigenesis. Vascular endothelial growth factor (VEGF), also known as vascular permeability factor, is a prime regulator of vascular permeability and angiogenesis. The VEGF system has been shown to be upregulated in various types of tumors, and there is evidence for interactions between this system and COX-2-derived prostaglandins .
A wide variety of normal and neoplastic tissues express epidermal growth factor receptor (EGFR). Its overexpression has been detected in a number of human tumors including breast, lung, head and neck, and colorectal carcinomas. EGFR protein expression has been suggested to be of prognostic importance . Immunoreactivity of EGFR protein has been also observed in central nervous system tumors such as meningiomas and gliomas, suggesting EGFR involvement in the proliferation and/or differentiation of meningothelial cells [18, 19].
In addition, other biological parameters such as apoptosis, cell cycle-related factors including bcl-2 and cyclin D1, and hormonal receptor status have been investigated in several studies. However, their predictive values have not led to the development of a comprehensive molecular prognostic model [20-22].
On the basis of the limitations of the current grading system for predicting tumor behavior, we assessed whether these markers can be used to predict and distinguish differences in prognosis among meningioma cases, even cases of the same-grade meningioma.
The goal of this study was to establish a relationship between the expression of Ki-67, cyclin D1, EGFR, COX-2, VEGF, and bcl-2 and histological grading with other histopathologic parameters, including invasion into adjacent tissues, to determine the value of these molecular markers for predicting disease prognosis.
- Top of page
- Materials and methods
In our series of 88 meningiomas treated at a single institution and diagnosed according to the 2007 WHO classification of nervous system tumors, we confirmed that tumor grade was strongly associated with all histologic para-meters for meningioma grading criteria. Meanwhile, invasion into surrounding tissues such as bone and soft tissue makes tumor resection more difficult, but does not change the grade . No correlation was observed between tumor grade and brain or adjacent soft tissue invasion in our study. This lack of correlation may be explained, in part, by the inclusion of adjacent soft tissue invasion as well as brain invasion in our statistical analysis. However, invasion into adjacent soft tissues, like brain invasion, results in a high recurrence rate and poor prognosis . Therefore, we recommend that meningiomas with invasion into adjacent soft tissues should be monitored extremely closely with radiographic follow-up. We also observed a significantly higher rate of recurrence in brain- or soft tissue-invasive meningiomas. This result indicates that despite recurrence of brain-invasive meningiomas being well-recognized , adjacent soft tissue invasion may also be associated with tumor recurrence. In addition, our data confirm that most meningiomas (90.9%) recur with a similar histology, both at first operation and recurrence. This indicates that meningioma recurrence is not associated with an increased potential for tumor growth. We attempted to define the role of ancillary markers; however, there was no significant correlation between immunohistochemical markers and tumor recurrence.
Prognostic prediction based on tumor location has led to conflicting results. Among intracranial and intraspinal meningiomas, no obvious regional differences regarding proliferative activity have been observed . However, genetic studies by Gezen et al. have demonstrated a loss of 1 homologue of chromosome 22 localized on the long arm in patients with spinal meningioma; a tumor suppressor gene may be present at this site . These results suggest that the prognosis for intraspinal meningiomas is less favorable than that for intracranial meningiomas. Presently, most intraspinal meningioma cases can be categorized as WHO grade I (7/8, 87.5%), but the occurrence of solid-sheet formation and necrosis was statistically higher in intraspinal meningiomas, although we did not perform chromosomal studies. Additional studies are needed to confirm our data.
Immunohistochemistry of the Ki-67 antigen has been widely used in numerous studies of meningiomas as a powerful prognostic marker and supplement to histopathology for meningioma grading. A high Ki-67 index is associated with aggressiveness and poor prognosis for meningiomas [4, 26]. Although we observed no significant correlation between Ki-67 labeling index and tumor recurrence, a strong correlation was observed between Ki-67 labeling index and tumor grade as well as various histopathologic parameters. In addition, tumor invasiveness was significantly influenced by Ki-67 labeling index according to multivariate analysis. These results are consistent with those of previous reports [4, 7, 8, 27, 28]. Thus, Ki-67 labeling index may be helpful, particularly in borderline cases of meningiomas in which mitoses are difficult to identify or other histopathologic criteria for atypical meningioma are not clear.
VEGF has been implicated as a critical factor in the formation of peritumoral brain edema associated with meningiomas [15, 29]. Sakuma et al. revealed that the expression of VEGF-A is related to the development of peritumoral brain edema with meningiomas and histological grade; thus, VEGF-A expression may influence the prognosis of patients with meningiomas . Furthermore, VEGF is thought to play an essential role in the positive regulation of tumor angiogenesis by promoting the migration, proliferation, and tube formation of endothelial cells; thus, up-regulation in meningiomas confirms its role as a pro-angiogenic factor . Presently, VEGF was shown to be correlated with grading of meningiomas and necrosis. Our findings indicate that overexpression of VEGF is correlated with poor prognosis of meningiomas by the mechanisms mentioned above. Thus, in the clinical setting, VEGF can be used as a prognostic marker for meningiomas. However, the prognostic value of VEGF expression regarding recurrences of these tumors appears questionable [24, 32]. VEGF secretion from microscopic remaining tumor lesions after surgery may induce neovascularization, which promotes meningioma recurrence, suggesting that high VEGF expression is significantly related to recurrence and poor prognosis of meningioma . However, in another study, despite more prominent vascularity in high-grade tumors, no difference in VEGF expression has been observed between tumor grades of meningiomas . We did not observe a significant association between VEGF expression status and tumor recurrence. However, few studies examining VEGF as a prognostic factor in meningiomas have been conducted. As information regarding VEGF status in meningiomas increases, correlation of this marker with diagnostic potential and clinical outcomes will become possible.
COX-2 plays an important role in neoplasm development in various manners. For brain neoplasms, increased levels of COX-2 have been noted in human gliomas, demonstrating an association between COX-2 protein expression, clinical aggressiveness and poor survival . Moreover, the utility of COX-2 inhibitors on proliferation and the invasion ability of cultured glioma cell lines have been reported . In meningiomas, Lin et al. revealed that the association between tumor grade and COX-2 expression is highly significant based on the 1993 WHO classification. More aggressive tumors were associated with increasingly high levels of COX-2 . In our study, COX-2 expression was significantly correlated with necrosis and brain or adjacent soft tissue invasion in meningioma patients. Moreover, we showed that COX-2 expression had a major impact on adjacent tissue invasion according to multivariate analysis. A hypothesis regarding increased COX-2 expression in more aggressive meningioma states that COX-2 levels may be an indicator of ischemia. COX-2 expression appears to be elevated near areas of necrosis, which may reflect increased angiogenesis . According to Ragel et al., overexpression of COX-2 contributed to tumor growth in mouse xenograft models with meningioma cell lines by increasing angiogenesis and cellular proliferation, as well as by decreasing apoptosis . In addition, the study revealed that celecoxib, a selective COX-2 inhibitor, decreases meningioma growth with evidence of decreased microvascular density, increased apoptosis, and decreased COX-2 and VEGF expression [36, 37]. Thus, the association between COX-2 and meningioma provides a new therapeutic strategy for treatment using COX-2 inhibitors, either as an adjunct or in combination with radiation therapy. Additional systemic studies are necessary before COX-2 inhibitors can be used in clinical practice.
We observed a correlation between COX-2 and VEGF expression. A previous study reported a trend of moderate to high COX-2 levels with VEGF expression, although no statistical association between COX-2 expression and VEGF was observed . VEGF interacts with COX-derived prostaglandins in angiogenesis . Thus, COX-2 as well as VEGF may be ancillary molecular markers used to indicate aggressive meningioma behavior.
The bcl-2 and cyclin D1 proteins have been implicated in prolonged cell survival by inhibiting apoptosis and influencing the cell cycle [6, 38]. Several studies demonstrated that bcl-2 and cyclin D1 are useful as proliferative markers in defining subgroups of meningiomas with more aggressive biological behavior [6, 20, 38]. However, another study did not observe an alteration of the cyclin D1 gene in non-astrocytic brain tumors, including meningiomas . We were unable to demonstrate other correlations of bcl-2 and cyclin D1 with histopathologic parameters and other molecular markers, except for cellularity in various grades of meningiomas.
Several studies have demonstrated a significant association between the staining intensity/percentage of EGFR and tumor grade in meningiomas [19, 40]. However, Andersson et al. identified no relationship between EGFR expression level and clinical outcome using immunohistochemical and quantitative real-time PCR analyses . We observed no correlation between EGFR and tumor grade.
In summary, our data revealed a statistically significant correlation between expression of Ki-67 and VEGF proteins and meningioma grade; COX-2 expression was correlated with several histopathologic parameters, including necrosis and brain or adjacent tissue invasion and with VEGF expression. These molecular factors may be helpful for determining meningioma grade in histologically controversial cases and identifying tumors with benign histological features, but unfavorable clinical outcomes.