Low-grade Diffuse Astrocytomas Research Articles

Genetic evidence of the neoplastic nature of gemistocytes in astrocytomas.

Gemistocytic astrocytoma is characterized by a predominance of large astrocytes with plump processes and massive accumulation of glial fibrillary acidic protein (gemistocytes). This histological variant of low-grade diffuse astrocytoma (WHO grade II) is prone to more rapid progression to anaplastic astrocytoma and glioblastoma than the ordinary fibrillary astrocytoma. The biological basis of this unfavorable prognosis is unclear, since gemistocytes themselves have low proliferative activity, even if present in anaplastic astrocytomas or glioblastomas. This has raised the question of whether gemistocytes are neoplastic cells or dysplastic reactive astrocytes. In this study, gemistocytes and non-gemistocytic neoplastic cells were separated by laser-assisted microdissection from six gemistocytic astrocytomas carrying TP53 mutations. In all cases, identical TP53 mutations were identified in both cell types, indicating that gemistocytes are indeed neoplastic cells. Their lack of proliferative activity may indicate terminal differentiation.

Promoter methylation of the DNA repair gene MGMT in astrocytomas is frequently associated with G:C --> A:T mutations of the TP53 tumor suppressor gene.

O(6)-Methylguanine-DNA methyltransferase (MGMT) is a repair protein that specifically removes promutagenic alkyl groups from the O(6) position of guanine in DNA. Repair of O(6)-alkylguanine adducts by tumour cells has been implicated in drug resistance since it reduces the cytotoxicity of alkylating chemotherapeutic agents. We assessed promoter methylation of the MGMT gene in astrocytic brain tumours by methylation-specific PCR. MGMT promoter methylation was detected in 26 of 54 (48%) low-grade diffuse astrocytomas (WHO grade II) and in 12 of 16 (75%) of secondary glioblastomas (WHO grade IV) that had progressed from low-grade astrocytomas. The frequency of MGMT methylation was significantly lower in primary (de novo) glioblastomas (13 of 36, 36%, P = 0.0155). The majority of low-grade astrocytomas with MGMT methylation (24/26, 92%) contained a TP53 mutation, whereas only 11 out of 28 (39%) cases without MGMT methylation carried a TP53 mutation (P < 0.0001). Furthermore, G:C --> A:T transition mutations at CpG sites were significantly more frequent in low-grade astrocytomas with MGMT methylation (15/26, 58%) than in those without (3/28, 11%, P = 0.0004). These results suggest that loss of MGMT expression as a result of promoter methylation, which frequently occurs at an early stage in the pathway leading to secondary glioblastomas, appears to be associated with increased frequency of TP53 mutations, in particular G:C --> A:T transitions.

Adult brain low-grade astrocytomas: survival after surgery and radiotherapy.

In order to identify prognostic factors of survival, twelve elements of disease and treatment have been evaluated for a population of 49 patients with diffuse low-grade astrocytoma treated with surgical resection and radiotherapy. The survival values were inversely correlated with age and major residual portion. On the other hand, KPS, lobar site, grade II Daumas-Duport lesions, protoplasmatic variant, early epilepsy, hyperfractionated radiotherapy and extent of exeresis were prognostic factors correlated with survival. Tumor extent and radiation total dose were not correlated in a meaningful way. Only KPS was statistically significant when compared to all the prognostic factors. We believe that patient selection according to age, lesion site and histological features are not sufficient to generate a homogeneous tumoral population. The most appropriate therapy for treating low-grade astrocytomas is still an open subject. However, recent studies have shown that the prognostic value of a group of factors is useful to plan controlled studies that compare differentiated treatment protocols.

P14ARF deletion and methylation in genetic pathways to glioblastomas.

The CDKN2A locus on chromosome 9p21 contains the p14ARF and p16INK4a genes, and is frequently deleted in human neoplasms, including brain tumors. In this study, we screened 34 primary (de novo) glioblastomas and 16 secondary glioblastomas that had progressed from low-grade diffuse astrocytomas for alterations of the p14ARF and p16INK4a genes, including homozygous deletion by differential PCR, promoter hypermethylation by methylation-specific PCR, and protein expression by immunohistochemistry. A total of 29 glioblastomas (58%) had a p14ARF homozygous deletion or methylation, and 17 (34%) showed p16INK4a homozygous deletion or methylation. Thirteen glioblastomas showed both p14ARF and p16INK4a homozygous deletion, while nine showed only a p14ARF deletion. Immunohistochemistry revealed loss of p14ARF expression in the majority of glioblastomas (38/50, 76%), and this correlated with the gene status, i.e. homozygous deletion or promoter hypermethylation. There was no significant difference in the overall frequency of p14ARF and p16INK4a alterations between primary and secondary glioblastomas. The analysis of multiple biopsies from the same patients revealed hypermethylation of p14ARF (5/15 cases) and p16INK4a (1/15 cases) already at the stage of low-grade diffuse astrocytoma but consistent absence of homozygous deletions. These results suggest that aberrant p14ARF expression due to homozygous deletion or promoter hypermethylation is associated with the evolution of both primary and secondary glioblastomas, and that p14ARF promoter methylation is an early event in subset of astrocytomas that undergo malignant progression to secondary glioblastoma.

Promoter Hypermethylation of the RB1 Gene in Glioblastomas

Loss of expression of the retinoblastoma gene (RB1) has been shown to occur in up to 25% of glioblastomas (WHO Grade IV). To elucidate the underlying mechanism, we assessed RB1 promoter hypermethylation using methylation-specific polymerase chain reaction and RB1 expression by immunohistochemistry in 35 primary (de novo) glioblastomas and in 21 secondary glioblastomas that had progressed from low-grade diffuse astrocytoma (WHO Grade II) or anaplastic astrocytoma (WHO Grade III). Promoter hypermethylation was significantly more frequent in secondary (9 of 21, 43%) than in primary glioblastomas (5 of 35, 14%; p = 0.0258). There was a clear correlation between loss of RB1 expression and promoter hypermethylation. In the majority of glioblastomas with loss of RB1 expression, there was promoter hypermethylation (11 of 13, 85%), whereas 93% of tumors with RB1 expression had a normal RB1 gene status (p < 0.0001). In three glioblastomas, areas with and without RB1 expression were microdissected; promoter hypermethylation was detected only in areas lacking RB1 expression. In patients with multiple biopsies, methylation of the RB1 promoter was not detectable in the less malignant precursor lesions, ie, low-grade diffuse and anaplastic astrocytoma. These results indicate that promoter hypermethylation is a late event during astrocytoma progression and is the major mechanism underlying loss of RB1 expression in glioblastomas.

Intramedullary low-grade astrocytomas: long-term outcome following radical surgery.

The management of low-grade intramedullary astrocytomas is controversial. Unlike ependymomas, which have a distinct cleavage plane, astrocytomas are diffuse infiltrative tumors. The intramedullary tumor database at NYU Medical Center was searched to identify all patients with intramedullary astrocytoma from 1988 to 1994. Seventeen patients presented with a low-grade diffuse astrocytoma. The medical, surgical and office records were reviewed retrospectively and statistically analyzed. All patients underwent a radical resection of the intramedullary tumor; of these, 12 patients had gross-total removal and 5 had subtotal removal. Nine patients received adjuvant radiotherapy. The median follow-up period was 7.4 years. Fourteen patients are alive and have stable disease. Eleven patients (79%) are functionally independent at last follow-up. The remaining patients are at functional Grade III. The 5-year overall survival was 82% for this group. These results indicate that aggressive surgery is associated with a prolonged survival for patients with intramedullary astrocytomas. Radiation therapy should only be reserved for patients who have radiographic progression or inoperable disease.

Gliomas of the optic nerve: histological, immunohistochemical (MIB-1 and p53), and MRI analysis.

Gliomas of the optic nerve, although typically of pilocytic (WHO grade I) histology, can present within the spectrum of astrocytic neoplasia including glioblastoma (WHO grade IV). In certain cases, histologic features alone make the distinction between pilocytic and diffuse astrocytomas difficult. We reviewed 22 cases of optic nerve gliomas, 19 of which were pilocytic astrocytomas (PA), and 3 of which were diffuse, non-pilocytic astrocytomas. The cases were evaluated for their clinical course, radiographic appearance, histologic grade, and proliferation indices as detected by MIB-1 (Ki-67) and p53 antibodies. Of the 19 PA, 14 showed no tumor growth by magnetic resonance imaging, and had Ki-67 and p53 labeling indices (LI) of < 1%. The other 5 PA exhibited aggressive behavior manifest by marked diffuse infiltrative tumor growth causing death in 2 patients, 1 of whom was diagnosed with neurofibromatosis type 1 (immunoperoxidase and radiographs not available), and marked local growth with an average time to growth of 39.3 months, a Ki-67 LI of 2-3%, and a p53 LI of < 1% in three others. Three of the five aggressive PA histologically demonstrated a finely reticulated pattern, a pattern that appears as an exaggeration or expansion of the normal neuroglia of the optic nerve, and may simulate a diffuse low-grade astrocytoma. Two demonstrated the coarsely reticulated pattern, with the biphasic and microcystic pattern typical of PA. Three diffuse astrocytomas (2 anaplastic astrocytomas and 1 glioblastoma) originated clinically and radiographically from the optic nerve, and revealed a Ki-67 LI of 2-12%, a p53 LI of 2-8%, and an average time to growth of 8 months. We conclude that the majority of PA of the optic nerve are non-aggressive, stabilize radiographically, and have Ki-67 and p53 LI < 1%. However, a subpopulation of PA has a propensity for aggressive behavior, and are identified by a Ki-67 LI of 2-3% and a p53 LI of < 1%. Diffuse astrocytomas have both Ki-67 and p53 LI > 2%. Thus, in cases of aggressive optic nerve tumors in which the histologic review of biopsy material cannot confidently confirm the diagnosis of pilocytic or diffuse fibrillary glioma, a p53 LI of > 1% appears to favor the diagnosis of diffuse astrocytoma.

Absence of p53 gene mutations in a tumor panel representative of pilocytic astrocytoma diversity using a p53 functional assay

Although p53-gene mutations occur with significant frequency in diffuse low-grade and high-grade astrocytomas, and are postulated to play an important role in tumorigenesis in these cases, the role of the p53 gene in pilocytic astrocytomas remains unclear. Published data using DNA-based assays for p53-gene analysis in these tumors have shown contradictory results in mutation frequency (0–14%). It is not known whether these heterogeneous results stem from the biological diversity of this tumor group or from technical problems. To re-evaluate p53-gene status in pilocytic tumors, we analyzed 18 tumors chosen to represent the clinical and biological heterogeneity of this tumor type with respect to anatomical location, patient age, gender, ethnic origin (Caucasian or Japanese) and the concomitant occurrence of neurofibromatosis type 1 (NF1). All primary tumors were histologically diagnosed as pilocytic astrocytoma (WHO grade I), except for one anaplastic pilocytic astrocytoma (WHO grade III) which developed in an NF1 patient and recurred as glioblastoma multiforme (WHO grade IV). p53 mutations were detected using an assay in yeast which tests the transcriptional activity of p53 proteins synthesized from tumor mRNA-derived p53-cDNA templates. None of 18 tumors, including 3 NF1-related tumors, showed p53-gene mutations between and including exons 4 and 11. We conclude that p53-gene mutations are extremely rare findings in pilocytic astrocytomas, and are absent even in those exceptional cases in which malignant progression of such tumors has occurred. Int. J. Cancer 76:797–800, 1998.© 1998 Wiley-Liss, Inc.

Absence of p53 gene mutations in a tumor panel representative of pilocytic astrocytoma diversity using a p53 functional assay.

Although p53-gene mutations occur with significant frequency in diffuse low-grade and high-grade astrocytomas, and are postulated to play an important role in tumorigenesis in these cases, the role of the p53 gene in pilocytic astrocytomas remains unclear. Published data using DNA-based assays for p53-gene analysis in these tumors have shown contradictory results in mutation frequency (0-14%). It is not known whether these heterogeneous results stem from the biological diversity of this tumor group or from technical problems. To re-evaluate p53-gene status in pilocytic tumors, we analyzed 18 tumors chosen to represent the clinical and biological heterogeneity of this tumor type with respect to anatomical location, patient age, gender, ethnic origin (Caucasian or Japanese) and the concomitant occurrence of neurofibromatosis type 1 (NF1). All primary tumors were histologically diagnosed as pilocytic astrocytoma (WHO grade I), except for one anaplastic pilocytic astrocytoma (WHO grade III) which developed in an NF1 patient and recurred as glioblastoma multiforme (WHO grade IV). p53 mutations were detected using an assay in yeast which tests the transcriptional activity of p53 proteins synthesized from tumor mRNA-derived p53-cDNA templates. None of 18 tumors, including 3 NF1-related tumors, showed p53-gene mutations between and including exons 4 and 11. We conclude that p53-gene mutations are extremely rare findings in pilocytic astrocytomas, and are absent even in those exceptional cases in which malignant progression of such tumors has occurred.

Immunohistochemical analysis of cyclin A in astrocytic tumours.

Cyclins are important regulators of the cell cycle; there is increasing evidence that some cyclins are positively involved in carcinogenesis. Amplification and translocation of the cyclin genes and overexpression of their mRNAs and proteins have been observed in a variety of tumours. We studied cyclin A protein in astrocytic tumours by immunohistochemical analysis. Immunohistochemistry with microwave antigen retrieval was carried out on formalin fixed, paraffin embedded material from 15 glioblastomas (WHO grade IV), 10 anaplastic astrocytomas (WHO grade III), seven diffuse low grade astrocytomas (WHO grade II) and nine pilocytic astrocytomas (WHO grade I) using antibodies against cyclin A and a proliferation marker MIB1. Staining for these antibodies was seen mainly in the tumour cell nuclei; 66% of all cases showing staining for cyclin A and 95% of all cases staining for MIB1. Mean labelling indices (LI) for cyclin a were higher in glioblastoma (mean LI-6.7) and anaplastic astrocytoma (mean LI-5.9) than low grade diffuse astrocytoma (mean LI-1.7) and pilocytic astrocytoma (mean LI-0.12), although there was no clear cut off point between the various tumour types. A good correlation was seen between labelling indices of cyclin A and MIB1 (Pearson correlation coefficient r = 0.59, P < 0.0001). Cyclin A is variably expressed in astrocytic tumours, either reflecting increased tumour proliferation (cyclin A being an integral component of the cell cycle), an alteration of its gene, protein upregulation or regulation of apoptosis. The genetic basis of expression of cyclin A in astrocytic tumours remains to be determined.

Vascular endothelial growth factor expression and vascular density as prognostic markers of survival in patients with low-grade astrocytoma.

It has long been recognized that some patients with low-grade astrocytoma may survive for many years, whereas in others the disease follows a more malignant course resulting in a short survival time, usually due to malignant transformation into higher-grade tumors. The aim of this study was to investigate angiogenesis in the initial biopsy specimen of tumor tissue as a biological marker to identify patients with low-grade astrocytoma who are at high risk of malignant tumor transformation or death. Tumor tissue was studied in 74 consecutively treated adult patients in whom a diagnosis of diffuse supratentorial hemispheric histologically proven fibrillary low-grade astrocytoma was made and who underwent surgery between January 1972 and January 1994. Studies were conducted using monoclonal antibodies to the antigens of the proliferation-associated Ki-67 (MIB-1), factor VIII, vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), and epidermal growth factor (EGF). The overall 5-year survival rate for the entire patient population was 65%, with a median survival time of 7.5 years. The total mean follow-up period was 6.1 years. All tumors showed a low proliferative potential at the time of the initial operation, as demonstrated by an MIB-1 labeling index of less than 1.5%. Patients with more than seven microvessels in tumor tissue (29 cases) had a shorter survival time (mean 3.8 years) than those with seven or fewer microvessels (mean survival 11.2 years). This difference in survival times was significant by univariate (p = 0.001) and stepwise multivariate analyses (p < 0.001). Tumors with a larger number of microvessels also had a greater chance of undergoing malignant transformation (p = 0.001). Similarly, significant staining for VEGF was correlated with shorter survival times when using univariate (p = 0.003) and multivariate (p = 0.008) analyses and with a greater chance of malignant transformation (p = 0.002). Patients with tumors staining positive for VEGF (39 individuals) had a median survival time of 5.3 years, and those with tumors negative for VEGF (35 patients) had a median survival time of 11.2 years. No association was observed between bFGF, EGF, and survival or malignant transformation. The stepwise multivariate analysis included histological and clinical variables simultaneously. The authors have shown that microvessel density and VEGF levels are independent prognostic markers of survival in fibrillary low-grade astrocytoma. This finding leads them to propose that fibrillary diffuse low-grade astrocytoma is not a single pathological entity but is composed of a spectrum of tumors with differing propensities to undergo malignant transformation that is at least partly based on their inherent angiogenic potential.

Genetics of pediatric central nervous system tumors.

Pediatric central nervous system (CNS) tumors comprise a wide variety of histologic subtypes ranging from the benign juvenile pilocytic astrocytoma to the highly aggressive atypical teratoid/rhabdoid tumor. Although some brain tumors are seen in association with inherited genetic disorders which predispose to malignancies, most are sporadic. Current knowledge regarding the cytogenetic and molecular genetic events which have been implicated in the development or progression of common brain tumors in children in the subject of this review. Combined cytogenetic and molecular genetic approaches, including fluorescence in situ hybridization, have been used to identify genomic alterations in different histologic types of pediatric brain tumors. The most frequent abnormality in primitive neuroectodermal tumor/medulloblastoma is an i(17q), present in approximately 50% of cases. This finding implicates the presence of a tumor suppressor gene on 17p, which is important in tumor development. A number of genes on 17p have been eliminated as candidates for this locus, including TP53. A tumor suppressor gene in chromosome band 22q11.2 has been hypothesized to play a role in atypical teratoid/rhabdoid tumors, and positional cloning strategies are in progress to identify a rhabdoid tumor gene. Chromosome 22 deletions are also seen in meningiomas and a small percentage of ependymomas, but it is not yet known whether the same gene is responsible for more than one malignancy. With regard to childhood astrocytomas, tumor-associated genetic changes have not yet been identified for the common juvenile pilocytic or low grade diffuse astrocytoma. In contrast, malignant anaplastic astrocytomas and glioblastoma multiforme have abnormalities similar to those seen in adults, including loss of alleles on 17p13 and TP53 mutations, trisomy 7, EGFR rearrangements, and loss of chromosomes 10 and 22. The presence of tumor-associated genetic abnormalities has clinical utility in a differential diagnostic setting, and has lead to the identification of genes which contribute to tumorigenesis.