|Year : 2020 | Volume
| Issue : 1 | Page : 27-32
Whether Serum Glial Fibrillary Acidic Protein (GFAP) Can Be Used as a Diagnostic Biomarker in Patients With Glioblastoma?
Charandeep Singh Gandhoke1, Ankit S Shah1, Daljit Singh1, Manju Subberwal2, Rohit Kumar Gupta2, Vinod Kumar Gupta2, Ravindra K Saran3
1 Department of Neurosurgery, Maulana Azad Medical College and G. B. Pant Institute of Post Graduate Medical Education and Research (GIPMER), New Delhi, India
2 Department of Biochemistry, Maulana Azad Medical College and G. B. Pant Institute of Post Graduate Medical Education and Research (GIPMER), New Delhi, India
3 Department of Pathology, Maulana Azad Medical College and G. B. Pant Institute of Post Graduate Medical Education and Research (GIPMER), New Delhi, India
|Date of Submission||05-Aug-2019|
|Date of Decision||21-Nov-2019|
|Date of Acceptance||02-Feb-2020|
|Date of Web Publication||30-Apr-2020|
Mch Neurosurgery Charandeep Singh Gandhoke
9B Cycle Merchant Society Rasta Peth (2nd floor), Pune-411011, Maharashtra
Source of Support: None, Conflict of Interest: None
Aim: To study if there is an increase in the serum levels of Glial Fibrillary Acidic Protein (GFAP) in patients with glioblastoma so that it can be used as a diagnostic biomarker for these cases. Design: Prospective observational study. Material and Methods: We prospectively examined 193 patients, out of which 25 were controls (patients with degenerative spinal disease or CV junction anomaly), 79 were tumors other than glioma, 45 were histologically proven gliomas except glioblastoma and 44 were histologically proven glioblastomas. Serum was taken from the patients in the pre-operative period. Serum GFAP levels were determined using a biotin-labeled antibody-based sandwich enzyme immunoassay for the quantitative measurement of GFAP in ng/ml. Statistical Analysis: Data analysis was performed by using SPSS (Statistical Package for Social Sciences) version 20:0. Statistical significance was defined as P value < 0.05. Receiver operating characteristic (ROC) curve analysis was used to find the cut-off value of serum GFAP for glioblastoma patients. Results: In Group A (25 controls) and Group B (79 patients with non-glial brain tumors), no patient had detectable levels of GFAP in their serum. Out of the 45 patients with histologically proven gliomas except glioblastoma (Group C), 12 patients had raised serum GFAP levels. Out of the 44 patients with histologically proven glioblastomas (Group D), 38 patients had raised serum GFAP levels while in the remaining 6, the levels of serum GFAP were not detectable. The sensitivity of serum GFAP levels in diagnosing glioblastoma patients was 86.36% and the specificity was 91.95%. The diagnostic accuracy of the test was 90.67%. Conclusions: Serum GFAP is a sensitive and specific marker for glioblastoma and can be a useful pre-operative serum biochemical investigation in differentiating glioblastoma cases from other brain tumors.
Keywords: Diagnostic biomarker, glioblastoma, glial fibrillary acidic protein (GFAP), serum
|How to cite this article:|
Gandhoke CS, Shah AS, Singh D, Subberwal M, Gupta RK, Gupta VK, Saran RK. Whether Serum Glial Fibrillary Acidic Protein (GFAP) Can Be Used as a Diagnostic Biomarker in Patients With Glioblastoma?. MAMC J Med Sci 2020;6:27-32
|How to cite this URL:|
Gandhoke CS, Shah AS, Singh D, Subberwal M, Gupta RK, Gupta VK, Saran RK. Whether Serum Glial Fibrillary Acidic Protein (GFAP) Can Be Used as a Diagnostic Biomarker in Patients With Glioblastoma?. MAMC J Med Sci [serial online] 2020 [cited 2020 Jul 14];6:27-32. Available from: http://www.mamcjms.in/text.asp?2020/6/1/27/283508
| Introduction|| |
Glioblastoma is the most common primary brain tumor and it is also the most malignant astrocytoma. Glioblastoma accounts for 52% of all primary brain tumors. These tumors were earlier called as “glioblastoma multiforme”, a term coined in 1926 by Percival Bailey and Harvey Cushing, based on the idea that the tumor originates from primitive precursors of glial cells (glioblasts) and the suffix “multiforme” was used to describe the variable appearance of these tumors due to the presence of necrosis, hemorrhage, and cysts.
Glioblastoma can arise “de novo” (primary) or evolve from low-grade astrocytomas (secondary). Primary glioblastoma is more aggressive and tends to occur in older patients. These tumors are characterized by Epidermal Growth Factor Receptor (EGFR) amplification and/or overexpression, Phosphatase and TENsin homolog (PTEN) mutations, Mouse Double Minute 2 homolog (MDM2) overexpression and/or loss of heterozygosity of chromosome 10p. Secondary glioblastoma is less aggressive and occurs in younger patients. These tumors are characterized by Isocitrate dehydrogenase 1 (IDH 1) mutations, p53 mutations, amplification of Platelet-Derived Growth Factor A (PDGF-A) and allelic loss of chromosomes 19q and 10q. Over 80% of secondary glioblastomas carry a mutation in IDH 1, whereas this mutation is rare in primary glioblastomas (5-10%). This helps to differentiate between primary and secondary glioblastomas.,
Survival from glioblastoma rarely exceeds 1 year. Post-operative concurrent chemo-radiation followed by oral Temozolomide promises better results. Patients with both IDH 1 mutation and O6 methylguanine DNA methyltransferase (MGMT) methylation have the longest survival; patients with either one of the above have intermediate survival and patients with neither of the above have the worst prognosis.
Glial fibrillary acidic protein (GFAP) is an intermediate filament protein that is highly specific for cells of astroglial lineage. It was named, first isolated and characterized by Lawrence F. Eng in 1969. Bongcam-Rudloff et al. mapped the GFAP gene to chromosome 17q21.
Advances in immunology, genetics, and immunohistochemistry have led to the discovery of more specific tumor markers, thus allowing for improved application of these markers in the diagnosis, treatment, and surveillance of patients with neoplasms. Despite these advances, no reliable marker has yet been identified for use in the diagnosis and surveillance of patients with high-grade gliomas.
In histopathology, confirmation of the diagnosis of glioblastoma utilizes GFAP staining on tumor tissue. We hypothesized that GFAP may be released from astroglial tumors, especially high-grade gliomas, into the serum and could therefore be used as a diagnostic biomarker for glioblastoma. Based upon this hypothesis, we studied GFAP levels in the serum of glioblastoma patients and various other groups.
| Material and Methods|| |
After obtaining ethical clearance from the Institutional Review Board (IRB), we conducted a prospective observational pilot study. The study included all patients who were admitted with a diagnosis of supratentorial or infratentorial space-occupying lesion and who were operated between August 2016 and March 2017. Control group included patients with degenerative spinal disease or CV junction anomaly. Patients who were operated outside and then referred to us, patients who had received neo-adjuvant chemotherapy and/or radiotherapy, patients with multiple intracranial space-occupying lesions and patients with recurrent glial malignancy were excluded from the study.
Sample collection and GFAP measurement
Three ml blood sample of patients enrolled in the study was taken prior to surgery. Serum samples were then centrifuged immediately in the laboratory and supernatants were stored at −700 C till the serum GFAP levels were assessed. Serum GFAP levels were determined using a biotin-labeled monoclonal anti-human GFAP antibody-based sandwich enzyme immunoassay for the quantitative measurement of human GFAP. GFAP-ELISA test conditions and quantification were as described by the manufacturer (Bio Vendor, Czech Republic). Samples, quality controls and calibrators were diluted 1:3 prior to analysis. Duplicates of 100 microlitre (μl) samples and calibrators were then pipetted into the ELISA wells and incubated for 2 hours at room temperature, followed by 1 hour incubation with a biotin-labeled anti-GFAP-antibody solution and 1 hour of incubation with a Streptavidin-HRP conjugate. Between each step, 3 cycles of washing (350 μl washing solution per well) were performed. Finally, the plate was incubated for 20 minutes at room temperature under avoidance of direct sunlight with the substrate solution of the kit. The color development was stopped by adding the stop solution of the kit. The absorbance was measured by reading the ELISA plate at 450 nm. The total assay time was about 5 hours. Results were expressed as ng/ml.
Formation of four study groups
Patients were divided into four groups after the histopathological report was available.
Group A: Patients with degenerative spinal disease or CV junction anomaly.
Group B: Patients with supratentorial or infratentorial space-occupying lesion (brain tumor) other than glioma.
Group C: Patients with histologically proven gliomas other than glioblastoma.
Group D: Patients with histologically proven glioblastomas.
GFAP analysis in tumor tissue
An association of serum GFAP levels and GFAP expression in tumor cells, on immunohistochemistry (IHC), was analyzed in patients with glioblastoma. GFAP expression was assessed in five separate microscopic fields of view under a magnification of 40x and was ranked as < 25%, 25 to 50%, 51 to 75% and > 75% GFAP positive tumor cells.
Data analysis was performed by using SPSS (Statistical Package for Social Sciences) version 20:0. Qualitative data variables were expressed by using frequency and percentage (%). Quantitative data variables were expressed by using mean and standard deviation (SD). Statistical significance was defined as P value < 0.05. Receiver operating characteristic (ROC) curve analysis was used to find the cut off value of serum GFAP for glioblastoma patients.
| Results|| |
Out of the 193 patients that were enrolled in the study, 25 patients belonged to group A (control group), 79 patients belonged to group B (Tumors except glioma), 45 patients were histologically proven gliomas except glioblastoma (Group C) and 44 patients were histologically proven glioblastoma cases (Group D) [Figure 1]. Tumors except glioma (Group B—79 cases) included meningiomas (26), schwannomas (12), craniopharyngiomas (9), pituitary adenomas (7), epidermoids (5), arachnoid cyst (4), metastasis (4), pineal masses (3), tuberculomas (3), colloid cyst (2), medulloblastomas (2) and one case each of inflammatory pathology and Ewing’s sarcoma. Gliomas except glioblastoma (Group C—45 cases) included cases of pilocytic astrocytoma (6), diffuse astrocytoma (13), anaplastic astrocytoma (7), oligodendroglioma (ODG—5), anaplastic oligodendroglioma (5), ependymoma (4), dysembryoplastic neuroepithelial tumor (DNET—2), ganglioglioma (1), papillary glio-neuronal tumor (1) and oligoastrocytoma (1). The average age at presentation for the gliomas except glioblastoma cases (Group C) was 30.22 years and there was almost equal male: female distribution. The average age at presentation for the glioblastoma cases (Group D) was 46.59 years and 77% were males. In Group A (25 patients of degenerative spinal disease or CV junction anomaly) and Group B (79 patients with non-glial brain tumors), no patient had detectable levels of GFAP in their serum. Out of the 45 patients with histologically proven gliomas except glioblastoma (Group C), 12 patients had raised serum GFAP levels (26.66%). These included cases of anaplastic astrocytoma (4), anaplastic ODG (3), ependymoma (2) and one case each of ODG, papillary glio-neuronal tumor and oligoastrocytoma [Figure 2]. Out of the 44 patients with histologically proven glioblastoma (Group D), 38 patients had raised serum GFAP levels (86.36%) while in the remaining 6, the level of serum GFAP was not detectable [Figure 3]. The average of GFAP levels in gliomas except glioblastoma cases (Group C) was 0.24ng/ml (median—0). The average of GFAP levels in glioblastoma cases (Group D) was 1.04ng/ml (median—0.145ng/ml) [Figure 4]. We observed that the serum GFAP levels were 4 times higher in glioblastoma patients than in other gliomas, which was statistically significant (P < 0.001). In our study, the sensitivity of serum GFAP levels in diagnosing glioblastoma patients was 86.36%, the specificity was 91.95%, the positive predictive value (PPV) was 76% and the negative predictive value was 95.80%. The diagnostic accuracy of the test was 90.67%. On ROC (Receiver Operating Characteristic) curve analysis, a serum GFAP level of more than zero was 86.4% sensitive and 91.9 % specific for diagnosing glioblastoma. Glioblastoma tumor tissue samples showed a strong variability in GFAP expression on IHC ranging from < 25% in some patients to almost 100% in others. The association of serum GFAP levels and GFAP expression on IHC was studied by using ANOVA (Analysis of Variance) test. It revealed that P value was more than 0.05, therefore, there was no significant association between percentage of GFAP positive tumor cells on IHC and serum GFAP levels in our study.
|Figure 1 Pie chart showing the percentage of distribution of cases in the four groups.|
Click here to view
|Figure 2 Bar diagram showing serum GFAP positive cases in “Glioma except Glioblastoma” group.|
Click here to view
|Figure 3 Bar diagram showing the serum GFAP positive and negative cases in Glioblastoma and non-Glioblastoma groups.|
Click here to view
|Figure 4 Box plots of serum GFAP levels in “Glioma except Glioblastoma” and Glioblastoma groups. The average of serum GFAP levels in gliomas except glioblastoma cases was 0.24ng/ml (median—0). The average of serum GFAP levels in glioblastoma cases was 1.04ng/ml (median—0.145ng/ml).|
Click here to view
| Discussion|| |
Unlike malignancy of other parts of the body where tumor markers in the blood can be a useful diagnostic tool, brain (glial tumor) does not have any established marker. There are some reports in which various CSF tumor markers have been suggested, however, none has been accepted widely. Largely, it is due to the pain of obtaining CSF from lumbar puncture, more so if repeated lumbar punctures are required and also due to poor sensitivity and specificity. Finding a reliable tumor marker in the blood can minimize the trauma and the burden of repeated radiological investigations like CT/MRI on follow up.
GFAP is a protein that is highly specific for cells of astroglial lineage and is found to be expressed in tumor cells on IHC. Pathologists, on regular basis, subject tumor tissue specimens to GFAP staining for the diagnosis of glial malignancy. This pilot study was conducted to determine, if there is any evidence of raised GFAP levels in the serum of patients suffering from glioblastoma so that it can be used as a promising new diagnostic biomarker.
In our study, many patients with glioblastoma had detectable levels of GFAP in their serum. The serum GFAP levels of patients with glioblastoma were significantly higher than patients with “gliomas other than glioblastoma” group (on an average, four times higher), whereas none of the non-glial tumors or controls showed raised serum GFAP levels.
Serum GFAP levels have also been reported to be elevated after traumatic brain injury and in patients with hemorrhagic stroke. Pelinka et al. in their paper, “Glial fibrillary acidic protein in serum after traumatic brain injury (TBI) and multiple trauma” concluded that GFAP is released in the circulation after TBI and it is related to brain injury severity and outcome. Foerch et al. in their paper, “Serum glial fibrillary acidic protein as a biomarker for intracerebral hemorrhage (ICH) in patients with acute stroke” concluded that serum GFAP can reliably detect ICH in the acute phase of stroke.
Other markers that have been detected in the cerebrospinal fluid (CSF) of glioma patients include Deoxythymidine-kinase, neuron-specific enolase (NSE) and vascular endothelial growth factor (VEGF). A few potential markers that have been detected in the serum of brain tumor patients include NSE, recoverin, Low-Molecular Weight Caldesmon, and Cathepsin D. Gronowitz et al. in their paper, “Deoxythymidine-kinase (dTK) in cerebrospinal fluid: a new potential ‘marker’ for brain tumors” found that no dTK activity could be detected in the CSF of healthy individuals or in patients with hydrocephalus or cranio-cerebral trauma. dTK levels ranging from detectable to high were found in the CSF of patients with malignant primary brain tumours or secondary brain tumors. Taomoto et al. in their paper, “The value of neuron specific enolase (NSE) in patients with brain tumors” found that serum and CSF levels of NSE were higher in malignant gliomas and primitive neuroectodermal tumors. Sampath et al. in their paper, “Cerebrospinal fluid (vascular endothelial growth factor—VEGF) and serologic (recoverin) tumor markers for malignant glioma” concluded that VEGF is detectable in the CSF and may be a potential marker for differentiating astrocytic from nonastrocytic tumors and recoverin is detectable in serum and may be a useful glioma tumor marker, especially for recurrent active disease. Zheng et al. in their paper, “Low-Molecular Weight Caldesmon (l-CaD) as a Potential Serum Marker for Glioma” found that patients with biopsy-proven gliomas have significantly increased serum levels of l-CaD as compared with the control group. Fukuda et al. in their paper, “Cathepsin D is a Potential Serum Marker for Poor Prognosis in Glioma Patients” concluded that the serum cathepsin D level correlated with the histologic grade of gliomas, suggesting that the serum cathepsin D level could be a potential indicator for disease aggressiveness in human gliomas.
GFAP is highly specific for cells of astroglial lineage and is widely used as a reliable marker in the immunohistochemical (IHC) diagnosis and differentiation of brain tumours. On IHC, glioblastoma patients show a strong variability in GFAP expression and distribution. One may argue as to why there are elevated levels of serum GFAP in patients with glioblastoma when both low grade as well as high grade gliomas show GFAP staining on IHC. GFAP has a high molecular weight of 52 kilodaltons (kDa) which limits its passage through the blood brain barrier (BBB) under normal physiological conditions. The pathophysiological mechanisms that may explain the increased GFAP levels in the serum of glioblastoma patients include GFAP leakage from glial cells (tumor necrosis) and disruption of the BBB. In fact, both are among the diagnostic criteria defining glioblastoma. These pathophysiological mechanisms may also explain the slightly increased GFAP levels, seen in our study, in 12 cases of gliomas except glioblastoma group. Jung et al. in their paper, “Serum GFAP is a diagnostic marker for glioblastoma multiforme” stated that a serum GFAP level of more than 0.05ng/ml was 76% sensitive and 100% specific for the diagnosis of glioblastoma in patients with a single supratentorial mass lesion. Furthermore, serum GFAP concentration was closely linked with glioblastoma tumor volume, tumor necrosis volume and the estimated amount of necrotic GFAP positive cells.
A decrease in GFAP expression of tumor cells is associated with growth and enhanced malignant behavior of glial tumors. In vitro, a decreased expression of GFAP in some glioblastoma cell lines was also associated with more aggressive and invasive potentials.,, While some factors like transforming growth factor (TGF)-alpha lead to decreased GFAP expression,, other factors like glial growth factor (GGF) and glial maturation factor (GMF) upregulated GFAP expression in vitro.,
One of the limitations of our study was that the association of serum GFAP levels with tumor size was not evaluated.
| Conclusion|| |
Serum GFAP is a sensitive and specific marker for Glioblastoma. It can become a useful pre-operative serum biochemical investigation in differentiating Glioblastoma cases from other brain tumors. This study opens up the possibility of using serum GFAP as a diagnostic biomarker for glioblastoma patients.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Ohgaki H, Kleihues P. The definition of primary and secondary glioblastoma. Clin. Cancer Res 2013;19:764-72.
Molenaar RJ, Radivoyevitch T, Maciejewski JP, Van Noorden CJ, Bleeker FE. The driver and passenger effects of isocitrate dehydrogenase 1 and 2 mutations in oncogenesis and survival prolongation. Biochimica et Biophysica Acta 2014;1846:326-41.
Molenaar RJ, Verbaan D, Lamba S, Zanon C, Jeuken JW, Boots-Sprenger SH et al.
The combination of IDH1 mutations and MGMT methylation status predicts survival in glioblastoma better than either IDH1 or MGMT alone. Neuro Oncol 2014;16:1263-73.
Eng LF, Ghirnikar RS, Lee YL. Glial fibrillary acidic protein: GFAP-thirty-one years (1969-2000). Neurochem Res 2000;25:1439-51.
Bongcam-Rudloff E, Nister M, Betsholtz C, Wang JL, Stenman G, Huebner K, Croce CM, Westermark B. Human glial fibrillary acidic protein: complementary DNA cloning, chromosome localization and messenger RNA expression in human glioma cell lines of various phenotypes. Cancer Res 1991;51:1553-60.
Pelinka LE, Kroepfl A, Schmidhammer R, Krenn M, Buchinger W, Redl H et al.
Glial fibrillary acidic protein in serum after traumatic brain injury and multiple trauma. J Trauma 2004;57:1006-12.
Foerch C, Curdt I, Yan B, Dvorak F, Hermans M, Berkefeld J et al.
Serum glial fibrillary acidic protein as a biomarker for intracerebral haemorrhage in patients with acute stroke. J Neurol Neurosurg Psychiatry 2006;77:181-4.
Gronowitz JS, Kallander CF, Hagberg H, Persson L. Deoxythymidine-kinase in cerebrospinal fluid: a new potential “marker” for brain tumours. Acta Neurochir (Wien) 1984;73:1-12.
Taomoto K, Kokunai T, Okuda T, Tamaki N, Matsumoto S, Hashimoto N. The value of neuron specific enolase (NSE) in patients with brain tumors. No To Shinkei 1987;39:169-73.
Sampath P, Weaver CE, Sungarian A, Cortez S, Alderson L, Stopa EG. Cerebrospinal fluid (vascular endothelial growth factor) and serologic (recoverin) tumor markers for malignant glioma. Cancer Control 2004;11:174-80.
Zheng PP, Hop WC, Sillevis Smitt PA, van den Bent MJ, Avezaat CJ, Luider TM et al.
Low-molecular weight caldesmon as a potential serum marker for glioma. Clin Cancer Res 2005;11:4388-92.
Fukuda ME, Iwadate Y, Machida T, Hiwasa T, Nimura Y, Nagai Y et al.
Cathepsin D is a potential serum marker for poor prognosis in glioma patients. Cancer Res 2005;65:5190-4.
Bonnin JM, Rubinstein LJ. Immunohistochemistry of central nervous system tumors. Its contributions to neurosurgical diagnosis. J Neurosurg 1984;60:1121-33.
Yen SH, Dahl D, Schachner M, Shelanski ML. Biochemistry of the filaments of brain. Proc Natl Acad Sci USA 1976;73:529-33.
Jung CS, Foerch C, Schanzer A, Heck A, Plate KH, Seifert V et al.
Serum GFAP is a diagnostic marker for glioblastoma multiforme. Brain 2007;130:3336-41.
Rutka JT, Murakami M, Dirks PB, Hubbard SL, Becker LE, Fukuyama K et al.
Role of glial filaments in cells and tumors of glial origin: a review. J Neurosurg 1997;87:420-30.
Murphy KG, Hatton JD, Hoi Sang U. Role of glial fibrillary acidic protein expression in the biology of human glioblastoma U-373MG cells. J Neurosurg 1998;89:997-1006.
Zhou R, Skalli O. TGF-alpha differentially regulates GFAP, vimentin, and nestin gene expression in U-373 MG glioblastoma cells: correlation with cell shape and motility. Exp Cell Res 2000;254:269-78.
Lee K, Jeon K, Kim JM, Kim VN, Choi DH, Kim SU et al.
Downregulation of GFAP, TSP-1, and p53 in human glioblastoma cell line, U373MG, by IE1 protein from human cytomegalovirus. Glia 2005;51:1-12.
Brockes JP, Lemke GE, Balzer DR Jr. Purification and preliminary characterization of a glial growth factor from the bovine pituitary. J Biol Chem 1980;255:8374-7.
Lim R, Mitsunobu K. Brain cells in culture: morphological transformation by a protein. Science 1974;185:63-6.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]