Anaplastic changes of diffuse leptomeningeal glioneuronal tumor with polar spongioblastoma pattern.

Diffuse leptomeningeal glioneuronal tumor (DLGNT) is a rare glioneuronal neoplasm with oligodendroglioma-like cells confined in the subarachnoid spaces. A great majority of DLGNT are histologically low grade. However, some tumors show features of anaplasia with increased mitotic and proliferative activity. Due to the limited number of patients and inadequate clinical follow-up reported to date, the WHO classification does not yet assign a distinct WHO grade to this entity. Polar spongioblastoma pattern, in which bipolar cells are arranged in parallel with palisading nuclei, remains poorly understood about the pathological process of forming this pattern. We experienced a case of 22-year-old man developing DLGNT with extensively distributed anaplastic changes involving polar spongioblastoma pattern and the secondary tumor invasion to brain parenchyma in 4½ years before the autopsy. Clinical and pathological courses of the patient are presented with radiological, histopathological, and genetic examinations. This is the first report demonstrating the immunohistological and genetic evaluation of a DLGNT with polar spongioblastoma pattern.

Immunologic and gene expression profiles of spontaneous canine oligodendrogliomas.

Malignant glioma (MG), the most common primary brain tumor in adults, is extremely aggressive and uniformly fatal. Several treatment strategies have shown significant preclinical promise in murine models of glioma; however, none have produced meaningful clinical responses in human patients. We hypothesize that introduction of an additional preclinical animal model better approximating the complexity of human MG, particularly in interactions with host immune responses, will bridge the existing gap between these two stages of testing. Here, we characterize the immunologic landscape and gene expression profiles of spontaneous canine glioma and evaluate its potential for serving as such a translational model. RNA in situ hybridization, flowcytometry, and RNA sequencing were used to evaluate immune cell presence and gene expression in healthy and glioma-bearing canines. Similar to human MGs, canine gliomas demonstrated increased intratumoral immune cell infiltration (CD4+, CD8+ and CD4+Foxp3+ T cells). The peripheral blood of glioma-bearing dogs also contained a relatively greater proportion of CD4+Foxp3+ regulatory T cells and plasmacytoid dendritic cells. Tumors were strongly positive for PD-L1 expression and glioma-bearing animals also possessed a greater proportion of immune cells expressing the immune checkpoint receptors CTLA-4 and PD-1. Analysis of differentially expressed genes in our canine populations revealed several genetic changes paralleling those known to occur in human disease. Naturally occurring canine glioma has many characteristics closely resembling human disease, particularly with respect to genetic dysregulation and host immune responses to tumors, supporting its use as a translational model in the preclinical testing of prospective anti-glioma therapies proven successful in murine studies.

Characterization of Inflammatory Changes Associated with Canine Oligodendroglioma.

Oligodendroglioma is a common glial tumour in the dog. In human neuropathology, the immune cell microenvironment of gliomas has been investigated; however, the nature of the inflammatory cells within canine gliomas is currently unknown. The aim of this study was to determine the nature of the immune cells and determine an association between the inflammatory cells and tumour grade. Thirty-four (18 of grade II and 16 of grade III) formalin-fixed and paraffin wax-embedded samples of canine oligodendroglioma were evaluated by light microscopy and immunohistochemistry for expression of CD3, PAX5, Iba-1, HLA-DR, Mac387 and CD31. Variations in immune cell recruitment and activation were evident in all cases. Infiltrating CD3(+) T lymphocytes were common in most cases. PAX5(+) B lymphocytes were less common and restricted to perivascular cuffs within or around the tumour. Iba-1(+) cells were common within the tumour and formed a dense infiltrate around the tumour in a subset of cases. HLA-DR(+) cells were common within the tumour and in a subset of cases formed perivascular cuffs. Iba-1(+) cells typically had prominent ramified processes suggestive of activated microglia, while the HLA-DR(+) cells had a more rounded morphology typical of amoeboid microglia. Rare Mac387(+) macrophages were found in the tumour parenchyma, while increased numbers of Mac387(+) monocytes were noted within the vasculature. No association or significance was established between the immune cell infiltrate and the grade of the tumour (all P ≥0.16). This study establishes that there is a robust population of immune cells within canine oligodendrogliomas and indicates that further study is needed to determine the role of these cells in tumour pathogenesis and progression.

Sox2 is required to maintain cancer stem cells in a mouse model of high-grade oligodendroglioma.

The stem cell-determining transcription factor Sox2 is required for the maintenance of normal neural stem cells. In this study, we investigated the requirement for Sox2 in neural cancer stem-like cells using a conditional genetic deletion mutant in a mouse model of platelet-derived growth factor-induced malignant oligodendroglioma. Transplanting wild-type oligodendroglioma cells into the brain generated lethal tumors, but mice transplanted with Sox2-deleted cells remained free of tumors. Loss of the tumor-initiating ability of Sox2-deleted cells was reversed by lentiviral-mediated expression of Sox2. In cell culture, Sox2-deleted tumor cells were highly sensitive to differentiation stimuli, displaying impaired proliferation, increased cell death, and aberrant differentiation. Gene expression analysis revealed an early transcriptional response to Sox2 loss. The observed requirement of oligodendroglioma stem cells for Sox2 suggested its relevance as a target for therapy. In support of this possibility, an immunotherapeutic approach based on immunization of mice with SOX2 peptides delayed tumor development and prolonged survival. Taken together, our results showed that Sox2 is essential for tumor initiation by mouse oligodendroglioma cells, and they illustrated a Sox2-directed strategy of immunotherapy to eradicate tumor-initiating cells.

The long pentraxin PTX3 as a correlate of cancer-related inflammation and prognosis of malignancy in gliomas.

Inflammation is a component of glioma microenvironment. PTX3 is a component of the humoral arm of innate immunity and a candidate marker of inflammation. In the present study we assessed the expression of PTX3 in gliomas by immunohistochemistry. PTX3 expression differed across low and high-grade tumors based on histopathological diagnosis and clinical severity, positively correlating with tumor grade and severity. In a multivariate logistic regression model, only the PTX3 score was significantly associated with the presence of a high-grade tumor. Thus, PTX3 may represent a new marker of cancer-related inflammation and glioma malignancy.

High-throughput immunohistochemical profiling of primary brain tumors and non-neoplastic systemic organs with a specific antibody against the mutant isocitrate dehydrogenase 1 R132H protein.

Isocitrate dehydrogenase 1 (IDH1) mutations are common in grade II-III diffuse gliomas and secondary glioblastomas. The aim of this study is to investigate the staining pattern of mIDH1R132H, an antibody specific to mutant IDH1 protein, in primary brain tumors and non-neoplastic systemic organs. Eight of 13 diffuse astrocytomas, 1 of 6 anaplastic astrocytomas, 9 of 11 oligodendrogliomas, 15 of 22 anaplastic oligodendrogliomas, 6 of 7 oligoastrocytomas, and 5 of 8 anaplastic oligoastrocytomas showed both cytoplasmic and nuclear positivity. Two of 25 atypical meningiomas and 2 of 42 pituitary adenomas were positive for mIDH1R132H. The following non-neoplastic systemic organs showed positivity in the cytoplasm alone: the myocardium, peribronchial glands, interlobular ducts of the salivary gland, gastric fundic gland, columnar epithelia of the large bowel, hepatocytes, centroacinar cells, the intercalated ducts of the pancreas, renal proximal and distal tubules, adrenocortex, ovarian granulosa cells, and the choroid plexus epithelia. It was concluded that the immunopositivity detected in non-neoplastic systemic organs was due to cross-reactivity, because immunohistochemistry with an anti-mitochondrial antibody revealed that the mIDH1R132H staining pattern was identical to that of the mitochondria. Therefore, mIDH1R132H positivity should only be considered to be validated when a cell shows both cytoplasmic and nuclear staining.

Systematic immunohistochemical profiling of 378 brain tumors with 37 antibodies using tissue microarray technology.

We performed a systematic immunohistochemical study on 378 brain tumors using 37 antibodies and tissue microarray (TMA) technology. The aim of this study was to find new diagnostic biomarkers using antibodies established in our laboratory. Our TMA consisted of a grid of 1.5-mm cores that were extracted from individual donor blocks. Staining for each antibody was scored using a three-point system. We used hierarchical clustering analysis to interpret these data, which resulted in separation of all the brain tumors into seven groups. Although there were some exceptions, cases with the same histological diagnosis were generally grouped together. We then carried out statistical analyses to find the most useful antibodies for grouping of brain tumors. Ten antibodies [glial fibrillary acidic protein (GFAP), Olig2, vimentin, epithelial membrane antigen (EMA), cytokeratin (AE1/AE3), alpha-internexin, nestin, pinealocytes PP5, aquaporin-4 (AQP4) M13d and AQP4M13e] discriminated between astrocytomas and oligodendroglial tumors. Six antibodies [EMA, AE1/AE3, TUJ1, nestin, neurofilament protein-MH (NF-MH) and perivascular cells GP-1] showed significant differences between high-grade and low-grade gliomas. Our data have revealed new antibodies with potential diagnostic utility (Olig2, PP5, GP-1) and demonstrate that TMA technology is highly useful for evaluating newly established antibodies in brain-tumor research.

[Immunohistochemistry in oligodendrogliomas]. / Imunoistoquímica em oligodendrogliomas.

Oligodendrogliomas (OL) are neuroepithelial tumors characterized by the presence of uniformly round nuclei with a clear cytoplasm around it. These features can also be seen in central neurocytomas, DNTs and clear cell ependymomas. Immunohistochemistry with glial and neuronal markers may be helpful in differential diagnosis. The aim of this study was to determine the glial and neuronal differentiation in 42 specimens of otherwise typical OL using immunohistochemical techniques. Ten cases showed anaplastic characteristics. Thirty-three samples (78.5%) were positive to GFAP with few cells stained in ten cases and many positive cells in six. Twelve cases (28.5%) were focally positive to NSE and/or synaptophysin showing neuronal differentiation. Thirty-four cases (80.9%) expressed S-100. In conclusion, glial proteins may be present focally in OL due to presence of mature reactive astrocytes or transitional forms between astrocytes and oligodendrocytes. Focal areas of neuronal differentiation can also be found in typical OL. The widespread staining with neuronal marker suggests central neurocytoma, but this diagnosis should not be done with small amount of tissue.

Microwave-assisted antigen retrieval and incubation with cox-2 antibody of archival paraffin-embedded human oligodendroglioma and astrocytomas.

Immunohistochemistry is an important tool that is often used for the diagnosis of pathologies; however, the length of time required to process the tissue is relatively long. Furthermore, the quality and sensitivity of immunohistochemical staining is affected by formalin fixation which results in variable loss of antigenicity, known as masking effect. Here we assess the effect of microwave irradiation on the incubation time required to obtain high quality immunohistochemical staining for cox-2 using archival formalin-fixed, paraffin-embedded human oligodendrogliomas and astrocytomas. The results show that intermittent microwave irradiation during the incubation with the primary antibody reduced the time requirement to 5 min while the staining quality was indistinguishable from 1 or 24 h long incubations. Thus, the use of this procedure results in a significant saving of time which is important for a timely diagnosis of pathological conditions that await treatment.

Imunoistoquímica em oligodendrogliomas / Immunohistochemstry in oligodendrogliomas

Os oligodendrogliomas (OL) são tumores gliais caracterizados histologicamente pela presença de núcleo redondo e homogêneo com halo claro perinuclear. A diferenciação microscópica desses tumores com neurocitoma central, DNT e algumas vezes com ependimoma de células claras pode ser difícil. O estudo imunoistoquímico com marcadores glial e neuronal tem sido utilizado e pode auxiliar no diagnóstico diferencial. O objetivo do presente estudo foi determinar a diferenciação neuronal e glial por meio de técnica imunoistoquímica utilizando anticorpos de rotina em tumores com características microscópicas de OL. Foram estudados 42 pacientes com idade entre 4 e 60 anos. Dez apresentavam sinais de maior malignidade (anaplásico). Trinta e três casos (78,5%) mostraram positividade para GFAP, sendo em 10 focal e 6 casos com expressão intensa. Doze casos (28,5%) apresentaram positividade para NSE e/ou sinaptofisina, demonstrando alguma diferenciação neuronal, principalmente focal. Trinta e quatro casos (80,9%) foram positivos para S-100 e três casos (7,1%) foram positivos focalmente para NeuN. Concluimos que áreas focais de diferenciação neuronal e/ou glial podem estar presente em OL típicos e, portanto, é necessário cautela no diagnóstico diferencial em amostras pequenas de tumor. A positividade difusa para marcadores neuronais deve sugerir o diagnóstico de neurocitoma central.

Immunohistochemical detection of dUTPase in intracranial tumors.

The nuclear isoform of deoxyuridine 5'-triphosphate nucleotidohydrolase (dUTPase, OMIM *601266, EC 3.6.1.23) is immunohistochemically detectable in all proliferating tissues and may thus be a useful adjunct for the grading of tumors analogous to Ki-67 labeling. A hundred and twenty-seven human intracranial tumors, including 56 astrocytomas, 12 oligodendrogliomas, 8 oligoastrocytomas, 34 meningiomas, 7 ependymomas, and 10 metastatic carcinomas, were stained using the monoclonal rat anti-human dUTPase antibody (clone 3E6) with formalin-fixed and paraffin-embedded tissue. The labeling indices were compared with those obtained with the proliferation marker Ki-67 on parallel tissue sections. All tumors contained dUTPase-positive nuclei, whereas the percentage of positive tumor cells generally increased with grade of malignancy. Meningiomas of higher grades, i.e., World Health Organization (WHO) grades II and III, contained additional cells with cytoplasmic reactivity. There were usually fewer dUTPase- than Ki-67-positive nuclei detectable. Unlike Ki-67, dUTPase was not detectable in mitotic figures. Labeling indices for dUTPase, but not for Ki-67, showed significant differences between all 3 WHO grades of diffuse astrocytomas. In summary, dUTPase staining provides a useful measure of cell proliferation distinct from that offered by Ki-67 labeling. It proved particularly useful for the evaluation of diffuse astrocytomas.

T lymphocyte subsets and immunoglobulins in intracranial tumor patients before and after treatment, and based on histological type of tumors.

It has been reported that nervous system and peripheral immune system communicate with each other and the peripheral immune status is depressed in some intracranial tumor (ICT) patients pre operatively. Little is known about the immune status of intracranial tumor patients during the post operative survival period. We thus investigated total T cells (CD 11+), helper/inducer (CD4+) T cells, suppressor/cytotoxic (CD8+) T cells, B cells (CD19+) and serum immunoglobulins in peripheral blood in certain ICT patients before and after treatment, and based on the histological type of the tumors. Post treatment analysis were conducted 30 days after surgical removal of tumor tissue in benign brain tumor patients and 30 days after chemo therapy (CT)/radiotherapy (RT) following surgical removal of tumor tissue in malignant brain tumor patients. Decreased CD11+, CD4+ and increased CD8+ T cell counts were observed in both benign and malignant tumor cases before treatment compared with control subjects. After treatment, CD4+ T cell count increased and CD8+ T cell count decreased than their pre treatment levels. Serum IgA and IgG levels were decreased in both benign and malignant brain tumor patients before treatment than in control subjects. Serum IgM level has been increased in both benign and malignant tumor patients before and after treatment than in control subjects. Anaplastic malignant astrocytoma, medulloblastoma and glioblastoma multiforme patients showed higher IgM level than astrocytoma, meningioma and ependymoma patients. In conclusions, the depressed host cellular immunity in benign and malignant tumor patients before treatment may be due to the changes in CD4+ and CD8+ counts in addition to tumour specific immunosuppressive factors. Treatment procedures such as surgery, CT and RT may play certain role in the post operative depressed immunosuppression in malignant tumor patients. Humoral immune mechanism (CD19+) in the ICT patients was less markedly affected.

Oligodendroglioma: pathology and molecular biology.

BACKGROUND: Recently, important new information has become available concerning the histologic recognition and molecular biology of oligodendrogliomas. This information, in turn, impacts the way neurosurgeons diagnose and treat patients with these tumors. The purpose of this paper is to review the pathology and basic science of oligodendroglioma, highlighting these developments. METHODS: Information for this review was obtained by a Medline search using the term "oligodendroglioma," and limiting the results to articles dealing with pathology. Chapters from standard textbooks were also used, and bibliographies were checked for additional key articles contributing to the understanding of the pathobiology of this disease. RESULTS: On histologic examination, oligodendrogliomas must be differentiated from tumors including the fibrillary astrocytoma, clear cell ependymoma, central neurocytoma, and dysembryoplastic neuroepithelial tumor (DNT). There is no specific immunocytochemical marker allowing for the recognition of human oligodendroglial tumor cells. A current simplified grading scheme separates these tumors into low grade (WHO grade II) and anaplastic (WHO grade III) oligodendrogliomas. New molecular and genetic markers may aid in grading oligodendrogliomas and identifying patients with a better prognosis or response to chemotherapy. Markers studied include Ki-67, PCNA, EGFr, VEGF, platelet-derived growth factor, p16, p18, p53, bcl-2, COX-1, and chromosomal deletions. The combination of allelic losses on chromosomes 1p and 19q has been statistically associated with a longer recurrence-free survival after chemotherapy. CONCLUSIONS: A patient with an oligodendroglioma may at times still present a diagnostic challenge for the neuropathologist. Yet making an accurate diagnosis is essential, since the clinical course and optimal therapeutic approach differs from that of other gliomas. In the near future, molecular characterization of oligodendrogliomas is expected to play an even greater clinical role.

Microglia in brain tumors.

Microglia have long been ignored by neurooncologists. This has changed with the realization that microglial cells not only occur within and around brain tumors but also contribute significantly to the actual tumor mass, notably in astrocytic gliomas. In addition, it has been speculated that microglia could play a role in the defense against neoplasms of the nervous system. However, the biological success of these tumors, i.e., their highly malignant behavior, indicates that natural microglial defense mechanisms do not function properly in astrocytomas. In fact, there is evidence that microglial behavior is controlled by tumor cells, supporting their growth and infiltration. This unexpected "Achilles heel" of microglial immune defense illustrates the risk of generalizing on the basis of a single aspect of microglial biology. Microglia are highly plastic cells, capable of exerting cytotoxic functions under conditions of CNS infections, but not necessarily during glioma progression. Thus, the suggestion that microglial activation through stimulation by cytokines (e.g., interferon-gamma) will benefit patients with brain tumors could prove fatally wrong. Therapeutic recruitment of microglia to treat such diffusely infiltrative brain tumors as astrocytic gliomas must be considered premature.

Prognostic factors in children and adolescents with low-grade oligodendrogliomas.

Few reports exist describing the progression-free and overall survival of children with low-grade (WHO grade II) oligodendrogliomas treated uniformly with aggressive surgery but without adjuvant chemotherapy or radiation therapy. Furthermore, significant prognostic features, including the MIB-1 labeling index (LI), have not been reported for children with oligodendrogliomas. The medical records of 20 consecutive patients with low-grade oligodendrogliomas were reviewed. All patients had been treated with aggressive surgical resection. Adjuvant chemotherapy and radiation therapy were reserved for radiographic or clinical progression. These patients have been followed for a median of 5.5 years (range 0.5-11.5 years) after diagnosis. To date, there have been no patient deaths. Six of the 20 patients experienced tumor progression at a median of 2.2 years (range 0.4-4.8 years) following the initial surgery. The MIB-1 LI was infrequently greater than 5. Of several prognostic factors for subsequent tumor progression that were examined, only tumors located within the parietal lobes were associated with a worse progression-free survival. Other risk factors, including presenting symptoms, age at diagnosis, MIB-1 LI and the extent of tumor resection, were not associated with an increased frequency of tumor progression.

T cell adoptive immunotherapy of newly diagnosed gliomas.

Patients with newly diagnosed gliomas were treated with adoptive transfer of ex vivo activated T lymphocytes, derived from lymph nodes (LNs) draining autologous tumor vaccines, to determine the long-term toxicity of this treatment. Twelve consecutive patients were enrolled: 2 with grade II astrocytoma, 4 with anaplastic gliomas, and 6 with glioblastoma multiforme. Patients were injected intradermally with short-term cultured autologous irradiated tumor cells, admixed with granulocyte macrophage colony-stimulating factor, to stimulate draining LNs. The LN cells were activated with staphylococcal enterotoxin A for 48 h and then cultured in medium containing interleukin 2 for an additional 6-8 days and subsequently transferred i.v. to the patients. The number of cells obtained from the LNs ranged from 9 x 10(7) to 1.1 x 10(9), and the median cell proliferation was 41-fold. The dose of T cells infused ranged from 0.6 to 5.5 x 10(10) with a median of 1.1 x 10(10), the majority of which were CD 4+ (mean, 71%). The entire treatment was performed as outpatient therapy and was associated with a toxicity of grade 2 or less, consisting mainly of fever, nausea, and myalgias during the first 24 h. There were no indications of late adverse events from this treatment even among three patients with follow-up greater than 2 years post T cell transfer. Moreover, four patients demonstrated partial regression of residual tumor. This Phase I clinical trial of adoptive immunotherapy for patients with newly diagnosed malignant gliomas demonstrates feasibility, lack of long-term toxicity, and several objective clinical responses.

CD44 expression in primary and recurrent oligodendrogliomas and in adjacent gliotic brain tissue.

CD44 expression was evaluated in 114 primary and recurrent oligodendrogliomas (46 primary oligodendrogliomas grade II and 15 recurrences grade II; 17 primary anaplastic oligodendrogliomas and nine recurrent oligodendrogliomas grade III; 14 glioblastomas with oligodendroglial growth pattern and 13 tumour recurrences grade IV). CD44 expression was found to correlate with tumour grading (P<0.001) and with survival (Kaplan-Meier Log Rank P<0.01, median survival 28 months in oligodendrogliomas with CD44 expression vs. 108 months in CD44-negative tumours). However, multivariate Cox regression analysis of grading and CD44 expression revealed that CD44 expression had no prognostic relevance independent of histological grading. Characterization of CD44 positive cells by double labelling with GFAP revealed that in addition to oligodendroglioma cells, reactive astrocytes within the tumour, at the invasive margin and along the pathways of oligodendroglioma invasion in the subpial matrix, and in the vicinity of vessels, frequently expressed CD44. It is suggested that in analogy to carcinoma invasion where a tumour-induced production of hyaluronan was found in fibroblasts at the invasive margin of the tumour, in the brain reactive astrocytes may produce hyaluronan which would facilitate the adhesion of new CD44-positive astrocytic processes but which would also promote tumour invasion.

Expression of granulocyte colony-stimulating factor in recurrent glial tumors is inversely correlated with tumor progression.

We have previously reported that in vivo-expression of granulocyte colony-stimulating factor (G-CSF) is a characteristic feature of reactive and neoplastic astrocytes. The aim of the present study was to analyze the expression of G-CSF protein in primary and recurrent astroglial, oligodendroglial and mixed glial-differentiated tumors. G-CSF expression was present in all GFAP-positive tumors excepting glioblastomas. G-CSF expression was significantly reduced in recurrent tumors more dedifferentiated than their primary counterparts. G-CSF expression was absent in all GFAP-negative tumors such as oligodendrogliomas. Our results demonstrate that G-CSF is a highly sensitive differentiation marker of neoplastic astrocytes, which is lost during tumor progression.

Spinal oligodendroglioma with diffuse arachnoidal dissemination in a Japanese Black heifer.

A gelatinous focus with cystic spaces, was found in the posterior funiculus of the 2nd to 3rd lumbar levels of the spinal cord of a Japanese Black heifer, 2 years old, with clinical signs of severe dysstasia. Histopathological examination revealed that the spinal lesion consisted of multifocal and diffuse proliferation of round cells with abundant vacuolar cytoplasm and hyperchromatic nuclei. In the lesions there was a number of cystic spaces containing aggregates of small round cells. The neoplastic foci showed a honeycomb structure divided by thin blood vessels, representing typical lesions of oligodendroglioma. Diffuse and multifocal proliferation of these round cells were also recognized in the subarachnoidal space in the sacral spinal cord. Immunohistochemically, the proliferating round cells were negative for glial fibrillary acidic protein. Based on these morphological features, the case was diagnosed as lumbar spinal oligodendroglioma with diffuse arachnoidal dissemination.

Expression and distribution of vascular endothelial growth factor protein in human brain tumors.

Marked neovascularization is a hallmark of many neoplasms in the nervous system. Recent reports indicate that the endothelial mitogen vascular endothelial growth factor (VEGF) may play a critical role in the regulation of vascular endothelial proliferation in malignant gliomas. Using novel monoclonal antibodies to the VEGF polypeptide we have determined the expression and cellular distribution of VEGF protein in a representative series of 171 human central nervous system (CNS) tumors by immunohistochemistry and immunoblotting. In agreement with previous in situ hybridization data, 19 out of 20 glioblastomas (95%) showed immunoreactivity for VEGF, whereas both the percentage of immunoreactive tumors and the extent of immunoreactivity for VEGF were significantly lower in astrocytomas. Of the pilocytic astrocytomas (WHO grade I) 44% were immunoreactive for VEGF, but we observed several cases with pronounced vascular proliferates in the absence of VEGF. In ependymomas, meningiomas, hemangioblastomas, and primitive neuroectodermal tumors, there was no correlation between VEGF expression, vascular endothelial proliferation and the grade of malignancy. Oligodendrogliomas and the oligodendroglial component of mixed gliomas lacked immunoreactive VEGF, indicating that endothelial growth factors other than VEGF may regulate tumor angiogenesis in these neoplasms. Western blot analysis showed a predominant VEGF protein species of 23 kDa and confirmed the immunohistochemical data in all cases. Our findings demonstrate that VEGF is expressed in a wide spectrum of brain tumors in which it may induce neovascularization. However, other angiogenic factors also appear to contribute to the vascularization of CNS neoplasms.