Malignant gliomas: MR imaging spectrum of radiation therapy- and chemotherapy-induced necrosis of the brain after treatment.

PURPOSE: To describe both the common and less frequently encountered magnetic resonance (MR) imaging features of radiation therapy- and chemotherapy-induced brain injury, with particular emphasis on radiation necrosis. MATERIALS AND METHODS: A cohort of 148 adult patients underwent surgical resection of malignant brain (glial) tumors and were subsequently entered into a research protocol that consisted of accelerated radiation therapy with carboplatin followed by chemotherapy with procarbazine, lomustine, and vincristine. Patients typically underwent sequential MR imaging at 6-8-week intervals during the 1st year and at 3-6-month intervals during subsequent years. In all patients, histopathologic confirmation of lesion composition was performed by board-certified neuropathologists. RESULTS: The patients exhibited different types of MR imaging-detected abnormalities of the brain: pure radiation necrosis in 20 patients, a mixture of predominantly radiation necrosis with limited recurrent and/or residual tumor (less than 20% of resected tissue) in 16 patients, radiation necrosis of the cranial nerves and/or their pathways in two patients, radiation-induced enhancement of the white matter in 52 patients, and radiation-induced enhancement of the cortex in nine patients. CONCLUSION: The frequent diagnostic dilemma of recurrent neoplasm versus radiation necrosis is addressed in this study through a description of the varying spatial and temporal patterns of radiation necrosis at MR imaging.

Toward a molecular classification of the gliomas: histopathology, molecular genetics, and gene expression profiling.

As many as 100,000 new cases of brain tumor are diagnosed each year in the United States. About half of these are primary gliomas and the remaining half are metastatic tumors and non-glial primary tumors. Currently, gliomas are classified based on phenotypic characteristics. Recent progress in the elucidation of genetic alterations found in gliomas have raised the exciting possibility of using genetic and molecular analyses to resolve some of the problematic issues currently associated with the histological approach to glioma classification. Recently, immunohistochemical studies using novel proliferation markers have significantly advanced the assessment of tumor growth potential and the grading criteria of some tumor subtypes. Preliminary studies using cDNA array technologies suggest that the profiling of gene expression patterns may provide a novel and meaningful approach to glioma classification and subclassification. Furthermore, cDNA array technologies may also be used to identify candidate genes involved in glioma tumor development, invasion, and progression. This review summarizes current glioma classification schemes that are based on histopathological characteristics and discusses the potential for using cDNA array technology in the molecular classification of gliomas.

Combined activation of Ras and Akt in neural progenitors induces glioblastoma formation in mice.

Gliomas are the most common primary malignant brain tumours and are classified into four clinical grades, with the most aggressive tumours being grade 4 astrocytomas (also known as glioblastoma multiforme; GBM). Frequent genetic alterations in GBMs (refs 2-5) result in stimulation of common signal transduction pathways involving Ras, Akt and other proteins. It is not known which of these pathways, if any, are sufficient to induce GBM formation. Here we transfer, in a tissue-specific manner, genes encoding activated forms of Ras and Akt to astrocytes and neural progenitors in mice. We found that although neither activated Ras nor Akt alone is sufficient to induce GBM formation, the combination of activated Ras and Akt induces high-grade gliomas with the histological features of human GBMs. These tumours appear to arise after gene transfer to neural progenitors, but not after transfer to differentiated astrocytes. Increased activity of RAS is found in many human GBMs (ref. 11), and we show here that Akt activity is increased in most of these tumours, implying that combined activation of these two pathways accurately models the biology of this disease.

A phase II trial of high-dose bromodeoxyuridine with accelerated fractionation radiotherapy followed by procarbazine, lomustine, and vincristine for glioblastoma multiforme.

PURPOSE: To conduct a Phase II study to evaluate the long-term efficacy and safety of high-dose 5'-bromodeoxyuridine (BrdU) and accelerated radiotherapy followed by procarbazine, lomustine (CCNU), and vincristine (PCV) chemotherapy in patients with glioblastoma multiforme. METHODS AND MATERIALS: Between 1994 and 1996, 88 patients were enrolled to receive 1.9 Gy of radiation three times a day for two 5-day cycles separated by 2 weeks; each 5-day cycle was preceded by a continuous 96-hour infusion of BrdU at a dose of 2.1 g/m2/day. After radiotherapy, patients received PCV chemotherapy. RESULTS: Median survival for all 88 patients was 50 weeks. Seventy (79.5 %) received one or more courses of PCV; their median survival was 57 weeks. Covariates predictive of improved survival were gross total versus subtotal resection or biopsy (p = 0.0048) and radiation dose > or = 56 Gy (p = 0.019). While receiving BrdU, 47 patients (53%) suffered grade 3 or 4 thrombocytopenia or leukopenia; 22 patients (25%) suffered grade 3 or 4 dermatologic toxicity. CONCLUSION: Survival was not extended in patients with glioblastoma or gliosarcoma who received BrdU at the dose and administration schedule used in this study. The BrdU dose used in this study resulted in substantial myelosuppressive and dermatologic toxicity.

Role of surgery in the treatment of brain metastases in patients with breast cancer.

BACKGROUND: Patients whose brain metastases from breast cancer are treated nonsurgically have a median length of survival ranging from 2.5 to 7.5 months, and a median time to recurrence ranging from 2 to 5 months. Patients treated with radiotherapy have a median length of survival ranging from 3 to 4 months. Those treated with chemotherapy have a median length of survival ranging from 5.5 to 7.5 months. METHODS: We conducted a retrospective analysis on 63 patients treated over a 10-year period. Only patients who underwent surgery for nonrecurrent brain metastases were studied. Sixty-one patients (97%) underwent surgery within 2 weeks of diagnosis of the brain metastases. RESULTS: The median length of survival was 16 months (95% confidence interval [CI] 11 to 22 months), and the 5-year survival rate was 17% (CI 9% to 29%). Brain metastases recurred in 27 patients at a median interval of 15 months (CI 12 to 24 months). Eleven patients had local recurrence, 10 had distal recurrence, and seven developed leptomeningeal disease. Significant prognosticators of length of survival were age (p = 0.011), menopause status (p = 0.10), postoperative radiotherapy (p = 0.054), preoperative neurologic status (p = 0.011), and preoperative systemic disease status (p = 0.0003). Systemic disease status had a significant effect on the length of survival but not on the time to recurrence.

Diagnostic potential of laser-induced autofluorescence emission in brain tissue.

Laser-induced autofluorescence measurement of the brain was performed to assess its spectroscopic properties and to distinguish brain tumors from the normal tissues. The excitation-induced emission spectra were plotted on a 2-dimensional map, the excitation-emission matrix, to determine the excitation wavelengths most sensitive for the spectroscopic identification of brain tumors. The excitation-emission matrices of various types of human brain tumors and normal brain samples lead to the selection of three fluorescence peaks at 470, 520, and 630 nm, corresponding excitation light at 360, 440, and 490 nm, respectively for comparing the autofluorescence signatures of brain tissue. The fluorophores most likely related to each of these peaks are NAD(P)H, various flavins, and porphyrins, respectively. In vivo studies of rat gliomas showed that "NAD(P)H", "flavin", and "porphyrin" fluorescence were lower in gliomas than in normal brain. This finding suggests that there are certain relationship between brain tissue autofluorescence intensity and metabolic activity. In vitro human normal brain tissue fluorescence signals were lower in gray matter than in white matter and "NAD(P)H" fluorescence were lower in all measured human brain tumors than in normal brain. "Flavin" and "porphyrin" fluorescence in the neoplastic tissues was lower or higher than normal tissue depending on their nature. In conclusion, the fluorescence spectroscopic diagnostic system might be able to distinguish brain tumors from the normal brain tissue. The results of this study need to be verified and the investigation extended to human brain tumors in the operating room.

Cranial irradiation after surgical excision of brain metastases in melanoma patients.

BACKGROUND: Brain metastases account for 20-54% of reported deaths from melanoma. Duration and quality of survival depend on the extent of metastatic disease and response to treatment. Treatment goals are palliation of symptoms and prolongation of life. No studies have directly compared surgery alone and surgery with adjunctive cranial irradiation in patients with solitary brain metastases. METHODS: We evaluated postoperative adjunctive cranial irradiation in 34 patients with solitary brain metastases. RESULTS: Overall survival was significantly improved in the 22 patients who received adjunctive cranial irradiation versus that in the 12 patients who had surgery alone. Twenty-eight patients subsequently relapsed. Nine of 10 patients with surgery alone had brain recurrence as a component of failure. Six of 10 patients not receiving irradiation had brain recurrences as a component of relapse at multiple sites whereas only 1 of 18 patients receiving irradiation relapsed with the brain. CONCLUSIONS: Adjunctive cranial irradiation is justified for melanoma patients who undergo surgical therapy for solitary brain metastases. Survival in patients presenting with solitary brain metastases was improved by a reduction of relapse in the brain as a component of failure by combined surgery and irradiation.

Expression and localization of 92 kDa type IV collagenase/gelatinase B (MMP-9) in human gliomas.

Matrix metalloproteinases play an important regulatory role in tissue morphogenesis, cell differentiation and motility, and tumor cell invasiveness. We have recently demonstrated elevated activity of the 92 kDa type IV collagenase (MMP-9) in human glioblastoma and in the present study examine the relative amounts of MMP-9 protein and mRNA in human gliomas and as well as the distribution of MMP-9 in human glioma tumors in vivo. Using an enzyme-linked immunosorbent assay for the quantitative determination of MMP-9 protein, we found that levels were significantly higher in malignant astrocytomas, especially in glioblastoma multiforme, than in normal brain tissues and low-grade gliomas. In addition, the amount of MMP-9 mRNA, as determined by northern blot analysis was higher in anaplastic astrocytomas and glioblastoma multiforme than in normal brain tissue and low-grade gliomas. Immunocytochemical staining for MMP-9 showed strong cytoplasmic immunoreactivity in the tumor cells and the proliferating endothelial cells of glioblastoma multiforme and anaplastic astrocytomas. The staining intensity was lowe in low-grade astrocytomas, and was undetectable or very low in normal brain astrocytes. The results indicate that expression of MMP-9 is dramatically upregulated in highly malignant gliomas and correlates with the highly malignant progression of human gliomas in vivo, and support a role for the MMP-9 in facilitating the invasiveness seen in malignant gliomas in vivo.

Expression and localization of 72 kDa type IV collagenase (MMP-2) in human malignant gliomas in vivo.

The 72 kDa type IV collagenase (gelatinase), a matrix metalloproteinase (MMP-2), has been proposed to potentiate the invasion and metastasis of malignant tumors. To determine the potential role of the MMP-2 in human gliomas and normal brain tissue, we examined the relative amounts of protein, mRNA, and distribution. Using gelatin zymography, densitometry, and an enzyme-linked immunosorbent assay for the quantitative determination of the MMP-2, we found that the enzyme's activity was significantly elevated in malignant astrocytomas, especially in glioblastoma multiforme, compared to low-grade glioma and normal brain tissues. As determined by Northern blot analysis, the amount of MMP-2 mRNA transcript was higher in anaplastic astrocytomas and glioblastoma multiforme tumors than in normal brain tissues or low-grade gliomas, a finding that was consistent with the amounts of MMP-2 protein detected in these tissues. Immunohistochemical studies demonstrated that MMP-2 was localized in tumor cells and vasculature cells of malignant astrocytomas. Staining intensity was clearly lower in low-grade astrocytomas, and immunoreactivity was very low or undetectable in normal brain astrocytes. The results suggest that expression of the MMP-2 is dramatically upregulated in malignant gliomas, correlating with the malignant progression of human gliomas in vivo.

Alternative RNA splicing of the hyaluronic acid receptor CD44 in the normal human brain and in brain tumors.

The cell-surface receptor for hyaluronic acid, CD44, is expressed by both normal and malignant cells. Numerous CD44 isoforms have recently been identified that are derived by alternative ribonucleic acid splicing. The expression of some CD44 isoforms has been shown to be involved in tumor progression and metastatic spread in a rat carcinoma model and in human carcinomas. In the present study, CD44 isoform expression was evaluated by reverse transcriptase-polymerase chain reaction (PCR) analysis in frozen sections derived from three samples of normal brain tissue and from 40 brain tumors, including samples of glioblastoma multiforme, anaplastic astrocytoma, low-grade astrocytoma, cerebral primitive neuroectodermal tumor, medulloblastoma, metastatic colon carcinoma, and metastatic melanoma. Normal brain tissue adjacent to the tumors was also examined in 14 of 18 glioblastomas. In all normal brain and tumor samples, with the exception of metastases from colon carcinoma, PCR analysis demonstrated one prominent product that corresponded to the CD44H hematopoietic form of CD44. Metastases from colon carcinoma demonstrated two prominent PCR amplification products corresponding to CD44H and CD44R1. These results suggest that CD44H is the predominant isoform of this protein in normal human brain tissue and in human neuroectodermal tumors of varying degrees of malignancy. The ability of CD44H to mediate tumor cell motility and invasiveness (in contrast to CD44R1) suggests that the CD44 alternative splicing pattern of neuroectoderm-derived tumors may enhance their local biological aggressiveness and intracerebral spread. The lack of expression of larger molecular weight CD44 variants by primary brain tumors may also partially explain why these tumors rarely metastasize to distant sites.

Sarcoma metastatic to the brain: results of surgical treatment.

We report on 21 patients surgically treated for intraparenchymal brain metastasis from sarcoma, including six osteosarcomas, four leiomyosarcomas, three malignant fibrous histiocytomas, two alveolar soft-part sarcomas, two Ewing's bone sarcomas, one extraskeletal osteosarcoma, one extraskeletal Ewing's sarcoma, and two unclassified sarcomas. Median survival after craniotomy was 11.8 months. Patients with a preoperative Karnofsky performance score of > 70 survived for 15.7 versus 6.6 months for those with a Karnofsky performance score < or = 70. Patients. undergoing complete resection survived 14.0 versus 6.2 months for patients undergoing incomplete resection. Patients with evidence of lung metastases at the time of surgery survived 11.8 months, which was similar to the 10.5-month survival for patients with disease limited to the brain. The two patients with alveolar soft-part sarcoma are alive at 16 and 25 months after surgery. We conclude that surgery is effective in treating selected patients with sarcoma metastatic to the brain and that patients with metastasis from alveolar soft-part sarcoma may have a relatively good prognosis if they are surgically treated. The complete removal of all brain metastases and a Karnofsky performance score > 70 are associated with a favorable prognosis, whereas the presence of concurrent lung metastases is not a contraindication to surgery.

Proteolysis and invasiveness of brain tumors: role of urokinase-type plasminogen activator receptor.

The cellular receptor for urokinase-type plasminogen activator (uPAR) in glioblastoma cell lines has been identified and found to be similar to the uPAR expressed by other tumor cell lines. Increased levels of uPAR have been found in primary malignant brain tumor tissues, especially highly malignant glioblastoma, and, to a lesser degree, in malignant astrocytomas, suggesting that this receptor might be involved in efficient activation of pro-uPA and confinement of uPA activity on the cell surface of invading brain tumors. The cell surface uPARs in gliomas could constitute an optimum environment for the generation and activity of plasmin, which is known to play a crucial role in the dissolution of the extracellular matrix during tumor cell invasion. In situ hybridization studies have shown that uPAR mRNA is expressed abundantly in tumor cells and is consistently present at the invasive edges of malignant gliomas. These results imply that uPAR is involved in plasmin-catalyzed proteolysis during glioma invasion and that interference with the uPA:uPAR interactions could constitute a novel approach for developing therapeutic strategies to counteract invasion of brain tumors.

Thromboembolic complications associated with brain tumors.

Thromboembolic complications are the second most common cause of death in hospitalized cancer patients; they are caused by alterations of hemostasis and include hypercoagulable states, acute and chronic disseminated intravascular coagulation, and primary fibrinolysis. The fibrinolytic system is comprised of several serine protease enzymes and their inhibitors and is associated in various biological systems with physiological and pathological events such as tissue development, remodeling, invasiveness, and migratory potentials of both normal and malignant cells. It also plays a key role in the dissolution of fibrin strands. Defective fibrinolysis, which is often associated with the pathogenesis of venous thrombosis and other thromboembolic complications, occurs when the balance is disrupted, resulting in either inhibition or enhancement of fibrinolysis. The association between thromboembolic complications and neoplastic disease has been well-established since Trousseau in 1865 first reported a high incidence of venous thrombosis in a series of patients with gastric carcinoma. In this article, we discuss the factors that have been shown to be associated with thromboembolic complications in patients who harbor brain tumors, namely, hemostatic alterations caused by the tumors themselves or through interactions with neural tissue around the tumors, pre-operative hemostatic alterations in certain patients, and defective fibrinolysis associated with specific tumor types and/or tumor locations.