Phase I study of weekly outpatient paclitaxel and concurrent cranial irradiation in adults with astrocytomas.

PURPOSE: Astrocytomas are extremely resistant to currently available treatments. Cranial irradiation is a mainstay of frontline therapy, but tumor recurrence is nearly universal. Paclitaxel has shown antitumor efficacy against astrocytoma cell lines, and is a potent radiosensitizer. For these reasons, we conducted a phase I study of weekly paclitaxel and concurrent cranial irradiation in patients with newly diagnosed astrocytomas. PATIENTS AND METHODS: Patients with astrocytomas were eligible for this study following initial surgery if they had a Karnofsky performance score (KPS) > or = 60%; normal hematologic, liver, and renal function; and could give informed consent. Beginning on day 1 of treatment, patients received paclitaxel by 3-hour infusion once weekly for 6 weeks, concurrent with standard cranial irradiation. Pharmacokinetic studies were performed on 10 patients. RESULTS: Sixty patients were enrolled; 56 were fully assessable. Forty-eight had glioblastomas (GBMs), 10 anaplastic astrocytomas (AAs), and two astrocytomas. Age ranged from 21 to 81 years (median, 55); KPS ranged from 60 to 100 (median, 70). The paclitaxel dose was escalated from 20 mg/m2 to 275 mg/m2. No clinically significant anemia or thrombocytopenia occurred. Only one patient (175 mg/m2) became neutropenic. Sensory neuropathy was dose-limiting. The maximum tolerated dose (MTD) was 250 mg/m2. Paclitaxel pharmacokinetic profiles in study patients were identical to those of previously reported patients with other solid tumors. CONCLUSION: The MTD of paclitaxel administered weekly for 6 weeks by 3-hour infusion is 250 mg/m2. Since patients with brain tumors often have preexisting neurologic deficits, we suggest 225 mg/m2 as the optimum dose for phase II trials in this group of patients.

Detection of tumor-derived DNA in cerebrospinal fluid.

The sensitivity of PCR-based methods for the detection of DNA offers opportunities for tumor diagnosis from the small amounts of tumor-derived DNA released into body fluids. We report the detection of tumor DNA in the cerebrospinal fluid (CSF) of two patients with intracranial neoplasms. One patient had a metastatic breast carcinoma which contained amplified HER-2/neu genes, and amplified HER-2/neu gene sequences were present in her CSF. The other patient had a glioblastoma which contained amplified epidermal growth factor receptor (EGFR) genes, and amplified EGFR gene sequences were present in her CSF. This report demonstrates that CSF sometimes contains tumor-derived DNA and suggests that PCR examination of CSF DNA may be diagnostically useful.

A quantitative electron microscopic study of the intracellular localization of wheat germ agglutinin in retinal neurons.

Previous work has established that, following endocytosis, wheat germ agglutinin, like a number of other plasma membrane bound ligands, is transported to the Golgi apparatus-complex. Previous studies that provided qualitative information about the intracellular distribution of internalized wheat germ agglutinin used techniques that precluded any quantitative conclusions about the relative magnitude of the labeling of endosomes, lysosomes, and the Golgi apparatus-complex. Using quantitative ultrastructural autoradiography, this study compares the time course and relative magnitude of labeling of various intracellular compartments to the labeling in the Golgi area. Fifteen minutes after intraocular injection, wheat germ agglutinin is confined to the inner surface of the retina and the immediate subsurface neuropil with little labeling of the retinal ganglion cell perikarya. Thirty minutes after injection, the plasma membrane (6.97 +/- 1.17), endosomes (10.27 +/- 3.98), smooth vesicles and tubules (1.94 +/- 1.66), and lysosomes (2.42 +/- 1.21) of the retinal ganglion cells are labeled, while the Golgi apparatus-complex is not labeled (0.29 +/- 0.25). (Figures in parentheses are the calculated relative radioactive source density +/- the standard deviation.) The relative labeling density of the plasma membrane and endosomes decreases somewhat during the next 90 minutes (plasma membrane, 4.76 +/- 0.67; endosomes, 7.23 +/- 2.02), while the labeling density of smooth vesicles and tubules and of lysosomes rises (smooth vesicles and tubules, 5.56 +/- 0.94; lysosomes, 7.76 +/- 1.56). The Golgi apparatus-complex, which is unlabeled at 30 minutes, is weakly labeled at 2 hours (1.26 +/- 0.28). The fact that the lysosomes are already labeled at 30 minutes while the Golgi apparatus-complex is unlabeled at that time indicates that the transport of wheat germ agglutinin to the Golgi apparatus-complex is a relatively late phenomenon, and suggests that the bulk of the wheat germ agglutinin destined for lysosomes does not pass through the Golgi apparatus-complex.

Transneuronally transported wheat germ agglutinin labels glia as well as neurons in the rat visual system.

Following intraocular injection in adult rats of 125I-labeled wheat germ agglutinin (I-WGA), the ultrastructural distribution of label in the superior colliculus and lateral geniculate was examined by electron microscope autoradiography. Three days after injection, 5.4% of the label in the lateral geniculate was associated with neuronal perikarya, and 3.6% was associated with glial perikarya. The corresponding figures for the superior colliculus were 5.1% and 0.8%. When the data were expressed as number of grains per micron 2 cytoplasm, there was no statistically significant difference between the grain density over neuronal or glial cytoplasm in either the lateral geniculate or the superior colliculus. A statistical analysis of the distance between the silver grains and the cell membranes showed that in both neurons and glia, the observed labeling was the product of internalized I-WGA and not the result of scatter from the neuropil or from label bound to the surface of the cells. These results indicate that much of the WGA released from axons and axon terminals is not confined to a specific "transsynaptic" pathway, but produces a generalized labeling of nearby cells, much like a microinjection of WGA into the region.

Immunohistochemical analysis of magnocellular elements in rat hypothalamus: distribution and numbers of cells containing neurophysin, oxytocin, and vasopressin.

A cell-by-cell analysis of the magnocellular elements in hypothalami of fifty Long-Evans (normal) and Brattleboro (diabetes insipidus) rats was done using the unlabeled antibody enzyme technique (PAP) with primary antisera directed against oxytocin (OXY), vasopressin (ADH), and the neurophysins. The magnocellular neurons of the hypothalamus were found in the supraoptic (SON), paraventricular (PVN), and anterior commissural (ACN) nuclei, a number of accessory nuclei, and as individual cells in the anterior hypothalamic area. SON was divided by the optic tract into the principal part and retrochiasmatic SON. In retrochiasmatic SON a majority of the cells contained vasopressin. Within the principal part of SON oxytocin-producing cells tended to be found rostrally and dorsally, while the vasopressin cells were more common caudally and ventrally. PVN was divided into three subnuclei, the medial, lateral, and posterior subnuclei, on the basis of cellular morphology and peptide content. The magnocellular cells of the medial and lateral PVN were closely packed together and nearly round, while those of posterior PVN were more separated and fusiform in shape with their long axis running in a medio-lateral direction. Medial PVN consisted primarily of oxytocin-producing cells, while lateral PVN was formed by a core of vasopressin-producing cells with a rim of oxytocin cells. Posterior PVN contained largely oxytocin-producing cells. Both ADH and OXY cells were found in the accessory nuclei. In the Long-Evans rat the SON had, on the average, 1443 OXY and 3236 ADH cells; the PVN had 1174 OXY and 976 ADH cells; and the accessory magnocellular groups in the hypothalamus (including the ACN) had 1286 OXY and 552 ADH cells. The Brattleboro strain animal had similar numbers of cells in these nuclei. (The cells which contain ADH in normal animals were identified in the Brattleboro rat as large, neurophysin-negative cells.) Thus, a large fraction of the magnocellular oxytocin- and vasopressin-producing cells in the rat were located outside of the PVN and SON. One accessory cell group in particular, ACN, had 616 OXY cells, or about 50% as many as PVN. In each nucleus the sum of the numbers of OXY and ADH cells was approximately the number of neurophysin cells.

Autoradiographic localization of thyrotropin-releasing hormone (TRH) receptors in human spinal cord.

Thyrotropin-releasing hormone (TRH) exerts many effects upon spinal cord function in animals, and may also play a role in human spinal cord function. We have used the technique of quantitative autoradiography to anatomically localize specific receptors for TRH within human spinal cord. Highest concentrations of TRH receptors were localized within lamina II, the substantia gelatinosa. A moderate density of TRH receptors was found in lamina IX, the motor neurons of the anterior horn. Low levels of TRH receptors were noted throughout the remainder of the gray matter of the human spinal cord, and no TRH receptors were localized within white matter. This anatomic distribution of TRH receptors within the human spinal cord is consistent with the localization of endogenous TRH and the effects of exogenously applied TRH in animal studies. These results suggest that any effects of TRH on human spinal cord function may be mediated by TRH receptors.

Localization and quantification of beta-adrenergic receptors in human brain.

Little information is currently available on the localization of noradrenergic systems in the human CNS. We used quantitative autoradiography with [125I] iodopindolol to examine beta-adrenergic receptors in postmortem human brain. The concentration of beta-receptors was highest in all subfields of the hippocampus, followed by cerebellum, and then thalamic nuclei, basal ganglia, midbrain, and cerebral cortex. Low levels were found in white matter and hypothalamus. This distribution differed from the distribution of beta-receptors reported in membrane homogenates of human brain and also from the distribution of beta-receptors in rat brain determined by autoradiography. The similarities and differences between the distribution of beta-receptors in the human and rat brains may have implications regarding the role of norepinephrine in the CNS of these two species.

Autoradiographic localization of thyrotropin releasing hormone receptors in human brain.

We used quantitative autoradiography to localize thyrotropin releasing hormone (TRH) receptors in human brain. Highest concentrations of TRH receptors were localized within the cortical, basal, and lateral nuclei of the amygdala and the molecular layer of the hippocampus. Low levels were found in the cortex, diencephalon, and basal ganglia. The radioligand bound with similar affinity and pharmacology to pituitary gland as to brain. These data suggest that authentic TRH receptors in the hippocampus and amygdala may mediate the putative effects of TRH on the human brain.

Similar distribution of monoamine oxidase (MAO) and parkinsonian toxin (MPTP) binding sites in human brain.

1-Methyl-4-phenyl-1,2,5,6-tetrahydropyridine (MPTP) causes parkinsonism in humans and other species. We found [3H] MPTP binding sites that were saturable, specific, and of high affinity. In autoradiographic studies, the highest binding densities of [3H] MPTP occurred in the hypothalamus, interpeduncular nucleus, and ependymal lining of the ventricles. High to moderate binding was seen in the dentate gyrus, caudate, putamen, substantia nigra, and cingulate cortex. The distribution of [3H] MPTP binding correlated with the distribution of [3H] pargyline binding to MAO. Human substantia nigra contains more MPTP binding sites than rat substantia nigra, and this may explain the sensitivity of humans to the neurotoxic effects of MPTP.

Positron emission tomography in a patient with progressive multifocal leukoencephalopathy.

A 56-year-old man with chronic lymphocytic leukemia, progressive multifocal leukoencephalopathy, and a dense left homonymous hemianopia had 18F-fluorodeoxyglucose positron emission tomography. Cortical glucose metabolism was decreased in the right cerebral hemisphere and the left cerebellar hemisphere. To our knowledge, this is the first demonstration of cerebral and cerebellar hypometabolism due solely to white matter disease.

Weekly paclitaxel with and without concurrent radiation therapy: toxicity, pharmacokinetics, and response.

Paclitaxel (Taxol; Bristol-Myers Squibb Company, Princeton, NJ) has shown in vitro and clinical activity against non-small cell lung cancer and astrocytic brain tumors, tumors traditionally thought of as relatively resistant to chemotherapy and radiotherapy. Because of its ability to block dividing cells in the G2/M portion of the cell cycle (the most radiosensitive phase of the cell cycle), paclitaxel is also a potentially potent radiosensitizer. To exploit these and other properties of paclitaxel, we explored a weekly, outpatient administration schedule, with and without concurrent radiation therapy, in patients with non-small cell lung cancer and astrocytic brain tumors. Our experience has shown that weekly outpatient administration is feasible, that remarkably high dose intensities can be achieved with acceptable toxicity, and that the specific dose-limiting toxicity appears to depend on administration schedule, type of concurrent radiotherapy, and certain patient characteristics. Preliminary response data are very encouraging. At the same time, pharmacokinetic studies have suggested possible reasons for our ability to use such exorbitant dose intensities safely, and also have shown that sustained plasma paclitaxel levels above the putative radiosensitizing threshold can be achieved continuously during a 6-week course of radiotherapy. Specific results, dosing recommendations, and plans for future studies are discussed.

Peptidylglycine alpha-amidating monooxygenase (PAM) in Schwann cells and glia as well as neurons.

We raised an antiserum against the synthetic peptide FKETTRSFSNECLGTTR corresponding to the amino terminus of the enzyme peptidylglycine alpha-amidating monooxygenase (PAM). Control experiments were performed to determine the specificity of the antiserum and its suitability for the immunohistochemical identification of PAM-containing cells. An immunoaffinity column made with the antibody coupled to Sepharose permitted the isolation of the active enzyme. Peptide-agarose immunoadsorbant removed the antibodies responsible for the characteristic staining patterns in immunohistochemical experiments. As expected from the widespread distribution of amidated peptides in the nervous system, PAM immunoreactivity was detected in perikarya in a variety of locations, including the pituitary, the hypothalamic periventricular and supraoptic nuclei, neocortex, and sensory ganglia. Punctate immunostained fibers, especially around neuronal perikarya, were observed in regions known to receive amidated peptidergic afferents. In addition, PAM immunoreactivity was observed in some neurons not known to produce amidated peptides (e.g., pyramidal cells of the hippocampus). This result suggests that these neurons also produce an amidated peptide. PAM immunoreactivity was also detected in several unexpected cell types, including ependyma, choroid plexus, oligodendroglia, and Schwann cells. The presence of enzymatically active PAM in Schwann cells was confirmed by measurements of amidating activity in ligated and control sciatic nerve. These results suggest that these non-neuronal cells may produce amidated peptides.

PCR-detection of tumor-derived p53 DNA in cerebrospinal fluid.

The sensitivity of polymerase chain reaction (PCR)-based methods for the detection of DNA offers opportunities for tumor diagnosis from the small amounts of tumor-derived DNA released into body fluids. Tumor-derived DNA can be distinguished from DNA derived from non-neoplastic cells by the presence of tumor specific genomic alterations, such as mutations in the p53 gene. This case report describes the use of allele-specific PCR (A-PCR) to detect a C-->T transition in p53 codon 273 in DNA extracted from the cerebrospinal fluid (CSF) of a patient whose glioblastoma contained the same mutation. The results of this study were confirmed by a second independent A-PCR reaction that detected the corresponding G-->A transition on the opposite strand. The specificity of the A-PCR protocol was demonstrated by negative controls, including pooled human placental DNA and the patient's non-tumor DNA, and by the use of A-PCR primers to detect all four possible bases at the site of the mutation. The methodology used in this study is suitable for use as a diagnostic clinical test. Because about half of all human tumors contain p53 mutations, PCR examination of CSF for the presence of mutant p53 sequences may be useful in the diagnosis of recurrent or metastatic tumors. Patients with known carcinoma of the breast or lung might be particularly benefited by this test.

Analysis of the allele-specific PCR method for the detection of neoplastic disease.

PCR assays for the presence of mutant K-ras or p53 sequences are potentially useful as sensitive tests for tumor diagnosis. The technical challenge is to design assays sensitive enough to detect a few molecules of mutant DNA yet sufficiently specific that a false positive signal is not produced by a 10(5)- or 10(6)-fold excess of normal DNA. We determined the detection limit of allele-specific PCR (ASA) as a function of the particular mismatch involved using all 12 possible mismatches in two different DNA sequence contexts (K-ras codon 12 and p53 codon 273). Depending on the identity of the mismatch, mismatched template was amplified 10(2)-10(4)-fold less than perfectly matched template. In other words, a mutant allele could be detected by ASA if it represented > 1-0.01% of the total DNA from that locus. Peptide nucleic acid (PNA) clamping was used to improve the K-ras ASA assay. Selective amplification of mutant sequences was achieved using a PNA complementary to the normal sequence to inhibit the amplification of wild-type DNA. PNA clamping followed by ASA resulted in significant improvement in sensitivity and specificity, permitting the detection of tumor DNA diluted with a 300,000-fold excess of normal human DNA.

Telomere associations in desmoplastic infantile ganglioglioma.

A four and a half year old male was diagnosed with desmoplastic infantile ganglioglioma. To our knowledge, the cytogenetics of this tumor have never been reported. In our analysis of 40 cells, no consistent clonal abnormalities were observed; however, the majority of cells (25 of 40) showed structural rearrangements (telomere associations) resulting in dicentrics and other derivative chromosomes. Breakpoints most often observed included 17q25 (6 of 40), 19p13.3 (4 of 40), 17p13 (3 of 40), 14q32 (3 of 40), 11q25 (3 of 40), 9p24 (2 of 40), 5q35 (2 of 40), and 22q13 (2 of 40).

Constitutional de novo t(1;22)(p22;q11.2) and ependymoma.

There is a body of evidence suggesting the presence of a tumor suppressor gene on chromosome 22 which plays a role in the pathogenesis of ependymomas. We report a patient with a de novo constitutional t(1;22)(p22;q11.2) who developed a malignant ependymoma at age 5. The patient is otherwise phenotypically normal. By fluorescence in situ hybridization (FISH) analysis, the chromosome 22 breakpoint has been localized to the region between the DiGeorge locus and BCR. Since NF2 and EWS are both distal to BCR, the are presumable not involved in this rearrangement. This patient may offer a unique opportunity to identify the chromosome 22 ependymoma tumor suppressor gene by cloning the translocation breakpoint.

Lectin affinity and PAGE analysis of soluble axonally transported glycoconjugates in the rat visual system.

Recent reports of the transsynaptic transfer of wheat germ agglutinin (WGA) in several systems are consistent with the hypothesis that WGA binds to endogenous glycoproteins(s) which undergo transsynaptic transfer. Such molecules might be involved in trophic interactions between neurons. Since these hypothetical glycoproteins must be soluble proteins (it is unlikely that integral membrane proteins are exchanged between neurons) this preliminary study of the soluble, axonally transported glycoconjugates of the visual system was undertaken. Soluble, tritium-labeled macromolecules which accumulated in the rat lateral geniculate nucleus (LGN) and visual cortex after intraocular injections of tritiated fucose were studied by lectin affinity chromatography and polyacrylamide gel electrophoresis (PAGE). After correction for hematogenously or cerebrospinal fluid transported label, 9.6% of the radioactivity in the LGN and 17.9% of the radioactivity in the visual cortex was found in the soluble fraction of a phosphate-buffered saline homogenate. In the LGN, 31.3% of the axonally transported soluble label was bound by concanavalin-A (Con-A) agarose. The corresponding figure in the visual cortex was 25.7%. Most (greater than 90%) of the label which did not bind Con-A was soluble in 10% trichloroacetic acid (TCA). Eighty-eight percent of the label which bound Con-A was precipitated by 10% TCA. On 7.5% polyacrylamide SDS reducing gels, the Con-A bound material from either the LGN or the visual cortex migrated as a single peak near the origin of the gel (apparent molecular weight greater than 300,000 Da).(ABSTRACT TRUNCATED AT 250 WORDS)

Expression of neuropeptide processing enzymes and neurosecretory proteins in ependyma and choroid plexus epithelium.

Recent studies suggest that brain ependyma and choroid plexus produce neuropeptide processing enzymes. To facilitate the understanding of these cells and their ability to produce biologically active peptides, we developed cultures of defined cell type. Ependymal cells were characterized by morphological criteria, and choroid plexus epithelial cell lines were characterized by the presence of the mRNA for IGF-II and transthyretin, a thyroxine binding protein produced in liver and choroid plexus. The ependymal cells and the choroid plexus epithelial cell lines were then examined for the presence of mRNAs for various neuropeptide processing enzymes. Northern blot analysis revealed high levels of furin, carboxypeptidase E, and peptidyl glycine alpha-amidating monooxygenase mRNAs, with levels in ependymal cells comparable to those in brain or pituitary. Carboxypeptidase E activity was detected in medium from cultured ependymal cells; this activity was identified as carboxypeptidase E based on the acidic pH optimum and sensitivity to various inhibitors. The mRNAs for other neuropeptide processing enzymes, such as prohormone convertases 1 and 2, were not detected on Northern blots of RNA from ependyma or choroid plexus epithelium. Since ependyma and choroid plexus epithelium express a subset of processing enzymes, we suggest that these cells have the capacity to produce biologically active peptides. Initial screening by reverse transcriptase-polymerase chain reaction assays has demonstrated the presence of mRNA for the neurosecretory proteins chromogranin B and secretogranin II in both ependyma and choroid plexus epithelium.

A quantitative comparison of the efficiency of orthograde axonal transport and transsynaptic transport of iodinated (125I) wheat germ agglutinin (I-WGA) and horseradish peroxidase labeled I-WGA (I-WGA-HRP) in the rat visual system.

A quantitative study of the orthograde axonal transport of iodinated wheat germ agglutinin (I-WGA) and I-WGA conjugated to horseradish peroxidase (I-WGA-HRP) demonstrated labeling of both the lateral geniculate nucleus and the visual cortex following intraocular injections. I-WGA was 13 times more efficient than I-WGA-HRP as a marker of the primary projection from the retina to the geniculate and 3 times more efficient than I-WGA-HRP as a marker of the second order projection to the cortex.

Pigmented villonodular synovitis of a lumbar facet joint.

We describe the CT appearance of suspected pigmented villonodular synovitis involving a lumbar facet in a 51-year-old woman, and discuss how the histologic and radiologic appearances may differ from those of synovial cysts.