Lack of Prognostic Significance of C-erbB-2 Expression in Low- and High- grade Astrocytomas

Astrocytic tumors (astrocytomas) are the most common primary glial tumors of the central nervous system (CNS), and are classified according to a grading system developed by the World Health Organization (WHO), from low grade astrocytomas (diffuse astrocytoma, pilocytic astrocytoma) to high grade astrocytomas (anaplastic astrocytoma, glioblastoma) (Louis et al., 2007). Glioblastoma (GBM) is the most common primary malignant tumor of the CNS in adults, representing 50% of all gliomas and 20% of all solid intracranial lesions that undergo surgery. With a mean incidence of 3 per year per 100.000 persons (Mineo et al., 2002; Rainov et al., 2007), it ranks fourth among the causes of death due to cancer in the middle-aged population (Hiesiger et al., 1993). Despite advances in the therapeutic management of GBM, prognosis remains poor and the median overall survival rate is <1 year. Several factors, such as age, performance status, radiotherapy, chemotherapy, quality of surgery, location and nature (de novo or secondary glioma)


Introduction
Astrocytic tumors (astrocytomas) are the most common primary glial tumors of the central nervous system (CNS), and are classified according to a grading system developed by the World Health Organization (WHO), from low grade astrocytomas (diffuse astrocytoma, pilocytic astrocytoma) to high grade astrocytomas (anaplastic astrocytoma, glioblastoma) (Louis et al., 2007).
Glioblastoma (GBM) is the most common primary malignant tumor of the CNS in adults, representing 50% of all gliomas and 20% of all solid intracranial lesions that undergo surgery.With a mean incidence of 3 per year per 100.000persons (Mineo et al., 2002;Rainov et al., 2007), it ranks fourth among the causes of death due to cancer in the middle-aged population (Hiesiger et al., 1993).Despite advances in the therapeutic management of GBM, prognosis remains poor and the median overall
Epidermal growth factor receptors (EGFR), transmembrane receptors located on the surface of epithelial cells, are crucial in the activation of subcellular signal transduction pathways controlling epithelial cell growth differentiation (Haapasala et al., 1996).Overexpression of EGFR has been shown to be a negative prognostic marker for survival in multiple cancer types, including breast, bladder, esophagus, cervix and ovarian carcinomas (Neal et al., 1985;Sainsbury et al., 1987;Ozawa et al., 1989;Pfeiffer et al., 1989;Bauknecht et al., 1989).EGFR targeted therapies have been introduced in carcinomas of the lung and colon (Wan den Eynde et al., 2011).In astrocytic tumors, however, the prognostic and diagnostic role of EGFR has shown divergent results (Pigott et al., 1993;Koka et al., 2003;Potti et al., 2004;Haynik et al., 2007).
C-erb-B2 (also known as Her-2/neu) is a member of ErbB family, located on chromosome region 17q12, encoding for a 185 kD transmembrane glycoprotein with intracellular tyrosine kinase activity.The c-erb-B2 receptor belongs to the family of EGFRs.
In this study, we investigated the frequency and prognostic significance of c-erb-B2 as well as the effects of tumor characteristics on survival in a small group of patients with low-and high-grade astrocytic tumors.

Materials and Methods
After the study was approved by the Institutional Review Board of Baskent University, the records of 72 patients with low-and high-grade astrocytic tumors diagnosed, followed and treated at the Medical Oncology Department of Adana Hospital of Baskent University, between 2006-2011, were investigated.Gender, age, type of operation, treatment modality, and tumor characteristics as location, grade, histological subtype, Ki-67 levels and C-erbB-2 status were evaluated retrospectively.The vital status of each patient (alive, dead and if dead the death date) was noted or inquired by phone, when required.
Tumor specimens were obtained from the pathology laboratory archives and prepared for immunohistochemistry. Four micron thick sections cut from the selected paraffin blocks were immunohistochemically stained by the same pathologist with polyclonal rabbit anti-human c-erbB-2 oncoprotein (Ref: A0485, Lot: 00076694, DAKO, Carpinteria, Ca, USA) in Dako Autostainer Link 48.Brown-red coloration in tumor cytoplasmic membrane by light microscopy was considered c-erbB-2 positive.If there was no staining, it was considered as 0. Pale and partial membranous staining in less than 10% of tumor cells was recorded as weak staining (1+), pale and complete staining in more than 10% of tumor cells was recorded as moderate staining (2+) and strong and complete staining in more than 10% of tumor cells was recorded as strong (3+).
Data analysis was performed using Statistical Package for Social Sciences (SPSS) for Windows version 11.5 (SPSS Inc., Chicago, IL, USA) software.Continuous variables were expressed as mean±standard deviation or median (minimum-maximum), while categorical variables were expressed as percentages (%) and numbers.
We evaluated the effects of factors such as gender, type of operation, tumor location, histological subtype, C-erbB-2 status, and treatment modality on overall survival by using log rank test with Kaplan Meier survival analysis.Univariate and multivariate analyses were performed using the Cox proportional hazards regression model.A p value<0.05 was considered to be statistically significant.
The neoplasm was located in the left cerebral hemisphere in 28 patients (12 in the frontal lobe, 6 in the parietal lobe, 6 in the temporal lobe, 3 in the occipital lobe and 4 in the cerebellar region).Of those located in the right hemisphere (n=44), 18 were in the temporal lobe, 13 were in the frontal lobe, 9 were in the parietal lobe and 1 was in the occipital lobe.Gross total tumor resection was achieved in 12 (16.7%)patients, subtotal resection was achieved in 42 (58.3%)patients, and decompression was achieved in 10 (13.9%) patients.Only biopsies were taken in 8 (11.1%) patients.
While 19 (26.4%) patients had received no treatment, 21 (29.2%)patients had received radiotherapy, 7 (9.7%)patients had received chemoradiotherapy, 17 (23.6%)patients had received chemotherapy (temozolomide) followed by chemoradiotherapy, and 8 (11.1%) patients had received chemotherapy followed by radiotherapy.Forty-eight patients had died by the end of follow-up period.Demographic and clinical characteristics of the patients are shown in Table 1.
Location of the tumor was significantly associated with survival (p=0.049).The median overall survival for frontal lobe tumors and temporal lobe tumors were 30.2 and 31.9 months, while the median overall survival in occipital lobe tumors and parietal lobe tumors were 15.7 and 15.6 months, respectively.Median overall survival was 17.4 months for GBM patients, 28.1 months for anaplastic astrocytomas, and 55.1 months for diffuse astrocytomas (p<0.001).
For the histological subtype and grade multivariate analysis could not be performed as there was a multiple relation problem.The histological subtype of cases with Grade I tumor was pilocytic astrocytoma, and all the cases were alive, while the histological subtype of cases with Grade IV tumor was GBM, and all the cases were dead.

Discussion
The present study performed in a small sample of low-and high-grade astrocytomas, failed to identify a statistically significant difference in terms of survival between patients with C-erbB-2 positive and C-erbB-2 negative tumors, although the median overall survival in these two groups of patients was 30 and 16.9 months.Astrocytic tumors, particularly anaplastic astrocytomas and GBM, face a poor prognosis despite major efforts to improve radiation therapy, chemotherapy, and surgery.While median survival for patients with anaplastic astrocytoma was reported to be between 3 and 5 years, median survival for patients with GBM was 12 to18 months, with the best available treatment.Although several prognostic factors such as age, performance status, quality of surgery, chemotherapy, radiotherapy and location of the tumor, have been proposed, there is a need for reliable prognostic indicators in patients with brain tumors.
In this study, 23 (64%) of 36 patients with C-erbB-2 overexpression were patients with GBM.In our study, C-erbB-2 positivity (2+ and 3+ staining) was observed in 11% of patients with low and high-grade astrocytomas.
It was not surprising to find out that all tumors with Her-2 overexpression were GBMs, 19.5% of patients with GBM showed C-erbB-2 overexpression.C-erbB-2 overexpression was not identified in any of the other astrocytomas.
Similarly, conflicting findings exist regarding the association between prognosis and C-erbB-2 expression.Mineo et al. (2002) showed that survival time was significantly longer when C-erbB-2 expression was low (p=0.04).Koka et al. (2003) in their study, revealed that C-erbB-2 overexpression significantly increased the odds of early mortality, and stated that C-erbB-2 overexpression may be a poor prognostic marker in patients with GBM.In another study, Potti et al. (2004) showed that C-erbB-2 overexpression predicts increased mortality.Published studies on C-erbB-2 overexpression and prognostic significance in astrocytic tumors are shown in Table 3.
In our study, there was no significant difference between the two groups with C-erbB-2 negative (30 months) and C-erbB-2 positive tumors (16.9 months) in terms of median overall survival (p=0.244).Although the difference between the two groups is clinically significant, the small sample size of our study limited our ability to detect a statistically significant difference.
Previous studies have demonstrated a variety of possible clinical and molecular prognostic indicators, but results have not been consistent.Gehan et al. (1997) showed that while young age, radiotherapy, chemotherapy and encapsulated tumor were favorable prognostic factors, parietal location was an unfavorable prognostic factor.In another study, Potti et al. (2004) demonstrated that tumor location (occipital, parietal), tumor histology (GBM), and presenting symptoms (nausea/vomiting) were related to poor prognosis in patients with primary malignant brain tumors.Similar to the literature, our findings suggest that tumor location (occipital and parietal) and tumor histology (GBM) are related to poor prognosis.
This retrospective study has limitations typical of immunohistochemical approaches, including limited technical reproducibility and subjective interpretation.The limited sample size is another important limitation.
Thus far, studies investigating the prognostic value of C-erbB-2 overexpression in human astrocytic tumors have produced mixed findings.In our study, although there was no difference in survival, C-erbB-2 overexpression was observed only in the GBM subtype.Large studies are necessary to fully clarify the prognostic role of C-erbB-2 in human astrocytic tumors.

Figure 1 .
Figure 1.Cumulative Survival Curves for Patients with c-erbB-2 Negative and Positive Tumors