Pharmacotherapy of malignant astrocytomas of children and adults: current strategies and future trends.

This article reviews the conceptual progression in the pharmacological therapy of malignant astrocytoma (MA) over the past decade, and its future trends. It is a selective rather than an exhaustive inventory of literature citations. The experience of the Brain Tumour Cooperative Group (BTCG) and earlier phase III trials are summarised to place subsequent phase II and I studies of single and combination agent chemotherapy in perspective. The BTCG experience of the 1970s to 1980s may be summarised to indicate that external beam radiotherapy (EBRT) is therapeutic, although not curative, and not further improved upon by altering fractionation schedules, or the addition of radioenhancers. Whole brain and reduced whole brain EBRT with focal boost were comparable regimens. Nitrosourea-based, adjuvant chemotherapy provided a modest improvement in survival among adult patients, which was comparable with that of other single drugs or multidrug regimes. The multiagent schedules, however, had a correspondingly higher toxicity rate. Intra-arterial administration was associated with significant risk, which conferred no therapeutic advantage. The trend of the past decade has been towards multiagent chemotherapy although its benefit cannot be predicted from the classic prognostic factors. Published experience with investigational trials utilising myeloablative chemotherapy with autologous bone marrow or peripheral blood stem cell haemopoietic support, drug delivery enhancement methods and radiosensitisers is critically reviewed. None of these approaches have achieved wide-spread acceptance in the treatment of adult patients with MA. Greater attention is placed on recent 'chemoradiotherapy' trials, which attempt to integrate and maximise the cytoreductive potential of both modalities. This approach holds promise as an effective means to delay or overcome the evolution of tumour resistance, which is probably one of the dominant determinants of prognosis. However, the efficacy of this approach remains unproven. New chemotherapeutic agents as well as biological response modifiers, protein kinase inhibitors, angiogenesis inhibitors and gene therapy are also discussed; their role in the therapeutic armamentarium has not been defined.

Oligodendroglial ganglioglioma with anaplastic features arising from the thalamus.

Anaplastic gangliogliomas with an oligodendroglial component are exceedingly rare tumors of uncertain growth potential. We report a 17-year-old female with a massive ganglioglioma containing anaplastic oligodendroglioma apparently arising from the thalamus. Two weeks after partial resection, she was started on a regimen including escalated doses of topotecan in combination with a fixed-dosage intensification regimen of cisplatin, cyclophosphamide and vincristine with subsequent hyperfractionated external beam radiotherapy. She currently has stable disease.

December 2000: 6 month old boy with 2 week history of progressive lethargy.

This 6-month-old Caucasian boy presented with a 10-day history of lethargy, obtundation, inability to hold his head up and mild torticollis. MRI and CT scans showed a large solid and cystic mass involving the right temporal, parietal and occipital lobes, pineal, superior pons, mesencephalon and posterior right thalamus. He underwent craniotomy initially for a partial tumor resection with an intraoperative diagnosis of desmoplastic astrocytoma. With immunohistochemistry and special stains the diagnosis of desmoplastic infantile ganglioglioma (DIG) was made. A near total resection was performed a week after initial resection.The patient then was treated with chemotherapy. Two months later an MRI showed tumor growth. Following additional aggressive chemotherapy, an MRI at 5 months post-resection indicated further tumor progression. This case illustrates that some DIGs may behave more aggressively than typical WHO grade I lesions.

The molecular genetics of therapeutic resistance in malignant astrocytomas.

The adverse prognosis associated with malignant astrocytomas (MA) is due in part to the development of resistance by the tumor to chemo- and radiotherapy-induced cytotoxic damage. The mechanisms of resistance are poorly understood but function at the level of the endothelial cell, the blood-brain barrier and the neoplastic cell itself. The classic examples of drug resistance proteins, such as the p-glycoprotein/multidrug resistance protein 1, have been identified within MA biopsy specimens. However, it is questionable to what degree, if at all, these proteins contribute directly to the evolution and prognosis of the MA. Surprisingly, there are specific genes, not traditionally associated with resistance, which appear increasingly relevant to both tumor progression and insensitivity to cytotoxic damage. These genes are involved in cell cycle regulation, and include the retinoblastoma susceptibility gene (Rb), the tumor suppressor gene p53, as well as those encoding the cyclins, their kinases and inhibitors. The interaction between the products of these genes and intratumoral environmental factors appears to involve a dynamic and prognostically adverse selection process. It is from this perspective that the mechanism(s) of hypoxic-ischaemic selection for resistance and its therapeutic repercussions will be analyzed.

The role of transforming growth factor beta in glioma progression.

This review examines the apparently paradoxical conversion of transforming growth factor beta's (TGFbeta) regulatory role as a growth inhibitor among normal glial cells to that of a progression factor among glioblastomas (GM). In vitro, TGFbeta functions as an autocrine growth inhibitor of near-diploid gliomas of any grade. In contrast, hyperdiploid glioblastoma multiforme (HD-GM) cultures proliferate in response to TGFbeta, which is mediated by induction of platelet-derived growth factor B chain (PDGF-BB). The dominant hypothesis of TGFbeta's pathogenetic association with malignant transformation has been predicated upon acquisition of resistance to its growth inhibitory effects. However, the lack of obvious correlation with TGFbeta receptor (TbetaR) expression (or loss) between the HD-GM and the TGFbeta-inhibited GM cultures suggests the existence of intrinsically opposed regulatory mechanisms influenced by TGFbeta. The mechanism of conversion might be explained either by the loss of a putative tumor suppressor gene (TSG) which mediates TGFbeta's inhibition of growth or by enhancement of an active oncogenic pathway among the HD-GM. The frequency of mutations within glioma-associated TSG, such as TP53 and RB, suggests that defects in TGFbeta's inhibitory signaling pathway may have analogous effects in the progression to HD-GM, and TGFbeta's conversion to a mitogen. Alternative sites of inactivation which might explain the loss of TGFbeta's inhibitory effect include inactivating mutation/loss of the TbetaR type II, alterations in post-receptor signal transmission or the cyclin/cyclin dependent kinase system which regulates the phosphorylation of pRB. Loss or inactivation of a glial TSG with a consequent failure of inhibition appears to allow TGFbeta's other constitutive effects, such as induction of c-sis, to become functionally dominant. Mechanistically, TGFbeta's conversion from autocrine inhibitor to mitogen promotes 'clonal dominance' by conferring a Darwinian advantage to the hyperdiploid subpopulations through qualitative and quantitative differences in its modulation of PDGF-A and c-sis, with concomitant paracrine inhibition of competing, near-diploid elements.

Transforming growth factor beta as a potential tumor progression factor among hyperdiploid glioblastoma cultures: evidence for the role of platelet-derived growth factor.

Among early-passage, near-diploid gliomas in vitro, transforming growth factor type beta (TGF beta) has been previously shown to be an autocrine growth inhibitor. In contrast, hyperdiploid (> or = 57 chromosomes/metaphase) glioblastoma multiforme (HD-GM) cultures were autocrinely stimulated by the TGF beta. The mechanism of this 'conversion' from autocrine inhibitor to mitogen is not understood; previous studies have suggested that platelet-derived growth factor (PDGF) might be modulated by TGF beta. The similar expression of TGF beta types 1-3, PDGF-AA; -BB, as well as the PDGF receptor alpha and beta subunits (a/beta PDGFR) between biopsies of the HD-GM and near-diploid, TGF beta-inhibited glioblastomas (GM) by immunohistochemistry did not explain the discrepancy in their regulatory responses. Flow cytometry demonstrated that TGF beta's mitogenic effect was selective for the aneuploid subpopulations of two of three selected HD-GM cultures, while the diploid cells were inhibited. Among the HD-GM, TGF beta 1 induced the RNA of PDGF-A, c-sis and TGF beta 1. The amount of PDGF-AA secreted following TGF beta treatment was sufficient to stimulate the proliferation of a HD-GM culture. Antibodies against PDGF-AA, -BB, -AB, alpha PDGFR and/or beta PDGFR subunits effectively neutralized TGF beta's induction of DNA synthesis among the HD-GM cell lines, indicating that PDGF served as the principal mediator of TGF beta's growth stimulatory effect. By comparison, TGF beta induced only the RNA of PDGF-A and TGF beta 1 among the near-diploid GM, c-sis was not expressed at all. However, the amount of PDGF-A which was secreted in response to TGF beta 1 was insufficient to prevent TGF beta's arrest of the near-diploid cultures in G1 phase. Thus, the emergence of hyperdiploidy was associated with qualitative and quantitative differences in TGF beta's modulation of PDGF-A and c-sis, which provided a mechanism by which the aneuploid glioma cells might achieve 'clonal dominance'. We hypothesize that TGF beta may serve as an autocrine promoter of GM progression by providing a selective advantage to the hyperdiploid subpopulation through the loss of a tumor suppressor gene which mediates TGF beta's inhibitory effect.

Strategies in the treatment of diffuse pontine gliomas: the therapeutic role of hyperfractionated radiotherapy and chemotherapy.

This article will review the current treatment of pediatric patients with diffuse pontine gliomas (DPG) and discuss three potential avenues of therapeutic research including (i) radiotherapy (RT) in combination with radiation sensitizers, (ii) dose-intensive, induction chemotherapy with hematopoietic support followed in sequence with RT applied as a "consolidation' therapy, and (iii) the interleafed application of phase-specific chemotherapeutic agents and hyperfractionated external beam radiotherapy (HFEBRT) referred to as "chemoradiotherapy'.

Optimal cutoff levels of F-18 fluorodeoxyglucose uptake in the differentiation of low-grade from high-grade brain tumors with PET.

PURPOSE: To determine the optimal cutoff level of fluorine-18-labeled fluorodeoxyglucose (FDG) uptake in the differentiation of low-grade from high-grade cerebral tumors at position emission tomography (PET). MATERIALS AND METHODS: The authors retrospectively reviewed images from PET, magnetic resonance imaging, and computed tomography performed in 58 consecutive patients with histologically proved brain tumors. There were 32 high-grade tumors (20 gliomas) and 26 low-grade tumors (18 gliomas). RESULTS: The best cutoff level of FDG uptake ratios in the differentiation of high-grade from low-grade tumors was 1.5 for tumor-to-white matter (T/WM) ratios and 0.6 for tumor-to-cortex (T/C) ratios. These levels were the same when only gliomas were analyzed and when all tumors were analyzed. When a T/WM ratio of more than 1.5 was considered indicative of a high-grade tumor, the sensitivity and specificity were 94% and 77%, respectively. The results were similar for the T/C ratio. CONCLUSION: Cutoff levels of 1.5 for the T/WM FDG uptake ratio and 0.6 for the T/C ratio are useful in the differentiation of low-grade from high-grade gliomas with PET.

Gliomatosis cerebri presenting as intractable epilepsy during early childhood.

We review 160 cases of gliomatosis cerebri from the literature and report an additional three infants and young children who presented with intractable epilepsy, corticospinal tract deficits, and developmental delay in whom a pathologic diagnosis was made. The progressive nature of the encephalopathy in our cases was documented by serial clinical examination, electroencephalograms, magnetic resonance imaging, and positron emission tomographic scans. The natural history of gliomatosis cerebri was determined by a retrospective review of the literature of 160 cases in 85 reports. The most common neurologic symptoms and signs included corticospinal tract deficits (58%), dementia/mental retardation (44%), headache (39%), seizures (38%), cranioneuropathies (37%), increased intracranial pressure (34%), and spinocerebellar deficits (33%). The most commonly involved central nervous system structures were the centrum semiovale and cerebrum (76%), mesencephalon (52%), pons (52%), thalamus (43%), basal ganglia (34%), and the cerebellum (29%). Fifty-two percent of patients were dead within 12 months of onset. Different grades of glial neoplasm may also coexist within gliomatosis cerebri such as astrocytoma with anaplastic astrocytoma, atypical or anaplastic oligodendroglioma, and glioblastoma multiforme. Hypotheses regarding the pathogenesis of gliomatosis cerebri include blastomatous dysgenesis, diffuse infiltration, multicentric origin, in situ proliferation, and "field transformation." The biologic determinants of whether a transformed glial cell behaves as a relatively localized tumor mass or truly loses anchorage dependence to become migratory as well as proliferative are not understood.