Cytosine arabinoside and mitoxantrone treatment of relapsed or refractory childhood leukemia: initial response and relationship to multidrug resistance gene 1.

The objective of this study was to determine the response rate and toxicity of high-dose cytosine arabinoside (AC) and mitoxantrone (M) in relapsed or refractory childhood acute myeloid leukemia (AML) and acute lymphocytic leukemia (ALL) and to correlate response with the expression of the multidrug resistance gene 1 (mdr1). Twenty-nine patients were treated with AC 1.0 g/m2 infused over 2 h every 12 h for eight doses (days 1-4) and M 12 mg/m2 infused over 1 h (days 3-6). Mdr1 expression was determined by a polymerase chain reaction (pcr) assay. Ten of 15 patients (67%) with AML obtained a complete remission (CR) of 3 to 30+ months duration. Eight of 14 (57%) ALL patients obtained a CR of 1 to 23+ months duration. The major toxicities were hematopoietic and infectious. Seventy-nine per cent of patients developed a documented infection during induction. Mdr1 did not correlate with a lower induction rate. This AC/M regimen is active in childhood AML and ALL.

Induction of growth suppression and modification of gene expression in multi-drug-resistant human glioblastoma multiforme cells by recombinant human fibroblast and immune interferon.

The combination of recombinant human fibroblast (IFN-beta) and immune (IFN-gamma) interferon induces enhanced growth suppression and modifies the antigenic phenotype in parental and multi-drug-resistant (MDR) human glioblastoma multiforme (GBM) cells. The present study was conducted to explore the mechanism underlying this cooperative interaction between interferons in inducing growth suppression in MDR-GBM cells. For this analysis we have utilized 2 MDR-GBM cell lines which display a differential sensitivity to growth suppression when exposed to IFN-beta or IFN-gamma. GBM-18-B3 (MDR) cells are more sensitive to growth inhibition by IFN-gamma than by IFN-beta, whereas GBM-18-A3 (MDR) cells are inhibited to a greater degree by IFN-beta than by IFN-gamma. In both cell types, however, growth is suppressed to a greater degree by the combination of interferons than by equivalent concentrations of either type of interferon used alone. Growth suppression induced by the interferons, alone or in combination, was not associated with comparable changes in the steady-state level of MDRI mRNA. In addition, the anti-proliferative effect of interferon was similar in GBM-18 (MDR) cells grown in the presence or absence of colchicine. GBM-18-A3 and GBM-18-B3 cells differed in their de novo and interferon-inducible expression levels of IFN-beta-responsive genes, isg-15 and isg-54. In contrast, both cell types responded in a similar manner with respect to expression of the IFN-gamma-responsive gene, HLA Class II (HLA-DR beta), and HLA Class I, fibronectin and ICAM-I. No further increase in expression of any of the genes was observed which was unique to the combination of interferons.

Effect of recombinant fibroblast interferon and recombinant immune interferon on growth and the antigenic phenotype of multidrug-resistant human glioblastoma multiforme cells.

To study the effect of drug resistance on the response of stage IV astrocytomas to interferon, a human glioblastoma multiforme cell line, GBM-18, was transfected with an expression-vector plasmid containing a human multidrug resistance (MDR) gene (pHaMDR1/A), and clones surviving in colchicine were isolated. GBM-18 multidrug-resistant subclones displayed cross-resistance to other chemotherapeutic agents, including vincristine, doxorubicin, and dactinomycin. The multidrug-resistant phenotype was reversible when GBM-18 multidrug-resistant cells were cultured in colchicine and the calcium-channel blocker verapamil. The level of the MDR1 gene (also known as PGY1) message was increased in GBM-18 multidrug-resistant cells selected for increased resistance to colchicine, and this effect was not correlated with an amplification of the MDR1 gene. In both parental GBM-18 and GBM-18 multidrug-resistant cells, growth was suppressed to a greater degree when cultures were treated with the combination of fibroblast interferon (IFN-beta) and immune interferon (IFN-gamma). Parental cells and multidrug-resistant subclones varied in their de novo and/or interferon-modulated expression of HLA class I and class II antigens, a high-molecular-weight melanoma-associated antigen, and intercellular adhesion molecule 1 (ICAM-1). Of the antigens tested, ICAM-1 and HLA class I antigens were the most sensitive to enhanced expression induced by IFN-beta and IFN-gamma when used alone or in combination. The results of the present study indicate that multidrug-resistant human glioblastoma multiforme cells retain their increased sensitivity to the antiproliferative activity of the combination of IFN-beta plus IFN-gamma, and differences in antigenic phenotype are apparent in independent multidrug-resistant glioblastoma multiforme clones.

Potentiation of growth suppression and modulation of the antigenic phenotype in human melanoma cells by the combination of recombinant human fibroblast and immune interferons.

Administration of interferon as a single therapeutic regimen in cancer patients with various neoplasias has had only limited efficacy in ameliorating the negative clinical course of their disease. In the present study, we have evaluated the effect of recombinant human fibroblast (IFN beta) and immune (IFN gamma) interferon, alone and in combination, on growth, differentiation and the expression of class I and II histocompatibility locus antigens (HLA) and melanoma-associated antigens on the human melanoma cell line H0-1. The effect of combinations of interferons on the antigenic profile of human melanoma cells displaying different organ colonization and spontaneous metastatic potential in athymic nude mice was also determined. H0-1 cells were more sensitive to the antiproliferative activity of IFN beta than to IFN gamma and the combination of interferons resulted in a potentiation of growth suppression. The antiproliferative effect of both interferons was greater in later-passage than in earlier-passage H0-1 cells, possibly reflecting alterations in the evolving tumor cell population as a result of long-term in vitro propagation and/or the selective outgrowth of cells with an increased growth rate. The enhanced growth suppression observed in H0-1 cells treated with the combination of IFN beta plus IFN gamma was not associated with a significant increase in the level of melanin, a marker of melanoma differentiation, above that observed with either interferon used alone. IFN beta and IFN gamma differentially modulated the expression of class I and II HLA and melanoma-associated antigens in H0-1 cells and a series of melanoma cells with different organ colonization and metastatic potential, including MeWo, MeM 50-10, MeM 50-17, 3S5 and 70W. No consistent potentiation or antagonism in the expression of any specific antigen was observed in any of the melanoma cell lines exposed to the combination of interferons. The present study demonstrates that the combination of IFN beta plus IFN gamma can potentiate growth suppression in H0-1 human melanoma cells and that this effect is not associated with an increase in differentiation or a potentiation in antigenic modulation. In addition, no direct correlation between the expression of any specific antigen or its modulation by IFN beta or IFN gamma, alone or in combination, and organ colonization and metastatic potential in nude mice was observed in the different melanoma cell lines.

Expression and modulation by cytokines of the intercellular adhesion molecule-1 (ICAM-1) in human central nervous system tumor cell cultures.

The intercellular adhesion molecule (ICAM-1) has been shown to be important in interactions involving cells of the immune system and to be upregulated in a number of cell culture systems by cytokines, including immune interferon (IFN-gamma) and tumor necrosis factor-alpha (TNF-alpha). In the present study, we have determined by fluorescence-activated cell sorter (FACS) analysis and the anti-ICAM-1 monoclonal antibody (MAb) CL203.4 the base-line expression of ICAM-1 and its modulation by recombinant IFN-beta, IFN-gamma and TNF-alpha in early passage (less than 15) human central nervous system (CNS) tumor-derived cell cultures. These cultures were established from various malignancies, including glioblastoma multiforme (GBM), astrocytoma (AST), ganglioglioma, medulloblastoma, meningioma and a pineal tumor. ICAM-1 expression was highest in the GBM- and AST-derived cell cultures and was lowest in the ganglioglioma and normal pineal cell cultures. Variable ICAM-1 expression was found, however, in tumors of the same histological group. In several cell cultures the variable expression observed by FACS was substantiated by the intensity of the molecular species immunoprecipitated by the anti-ICAM-1 MAb CL203.4 from these cells. All the cell cultures displayed variable but consistent increases in ICAM-1 expression following treatment with IFN-gamma or TNF-alpha. In general, the degree of increase in ICAM-1 expression was greatest in cultures exposed to TNF-alpha. Upregulation of ICAM-1 expression in an established glioblastoma multiforme cell line was of greater magnitude and more rapid following TNF-alpha treatment (within 2 to 3 hr) than exposure to IFN-gamma (by 24 hr). In several cultures, IFN-beta also increased ICAM-1 expression and enhanced the increase induced by TNF-alpha. The results of the present study indicate that variable expression of ICAM-1 is a common property of early passage cultures of CNS tumors and recombinant interferons and TNF-alpha can differentially upregulate ICAM-1 expression in these CNS tumor cell cultures.

Large scale isolation of human erythrocyte membranes by high volume molecular filtration.

A molecular filtration procedure for preparing large quantities of human erythrocyte ghost membranes is presented. Hemolysate ghost membranes are rapidly cycled in the retentate channel of the filtration apparatus, while hemoglobin is removed as it passes through Pellicon filters into the filtrate channel. Several-liter quantities of washed packed erythrocytes can be processed in a few hours with this system, and the filtration procedure does not appear to alter intact erythrocyte or ghost membranes. Intact erythrocytes in isotonic solution can be circulated through the retentate channel for 16 h with only 3% hemolysis and with preservation of their original morphology in scanning electron microscopy. Ghost membranes isolated by the procedure are virtually identical in morphology, polypeptide composition and acetylcholinesterase content to membranes isolated by conventional centrifugation techniques.