A growth factor that is mitogenic for vascular endothelial cells, with an ED50 of~'1 ng/ml, has been purified 170,000-fold to apparent homogeneity from tissue culture medium conditioned by a rat glioma-derived cell line. The pure protein is a 46-kDa dimer composed of two subunits of equivalent mass as established by comparison of migration in SDS/polyacrylamide gels with and without prior reduction.This glioma-derived growth factor is a glycoprotein and is not mitogenic for BALB/c 3T3 fibroblasts, properties that further distinguish it from other well-characterized vascular endothelial cell mitogens. In contrast to acidic and basic fibroblast growth factors and to platelet-derived endothelial cell growth factor, which have no secretory leader sequences and might only be released by leakage from damaged cells, the glycoprotein nature of this mitogen implies that it is processed through the glycosylating secretory pathway. This secretable growth factor could, therefore, be readily available in the extracellular space under normal physiological conditions in vivo to promote vascular endothelial cell proliferation associated with bloodvessel growth and maintenance. genes (1-7). All of these FGFs, with the exception of the int-2 and FGF-6 gene products, which have not been isolated, have been shown to be potent vascular endothelial cell mitogens in vitro. Furthermore, exogenously applied aFGF (8) and bFGF (9) induce angiogenesis in vivo. An angiogenic platelet-derived endothelial cell growth factor (PD-ECGF), a monomeric 45-kDa protein, has also been purified and structurally characterized. This protein, like aFGF and bFGF, lacks a secretory leader sequence and so is presumably released by lysis of platelets (10,11 We have identified and purified a vascular endothelial cell mitogen that is not only produced but also apparently secreted by cultured cells derived from a chemically induced rat glioma (13). This glioma-derived vascular endothelial cell growth factor (GD-VEGF) is structurally unrelated to the previously well-characterized mitogens for large-vessel endothelium. Pure GD-VEGF approaches the specific mitogenic activity of FGFs for cultured human umbilical vein endothelial (HUVE) cells. MATERIALS AND METHODSMitogenic Assays. HUVE cells were isolated and maintained as described (14). The cells were plated in gelatincoated 48-well tissue culture dishes (Costar) at 6000 cells per well in 400 ,ul of medium 199 (GIBCO) containing 15 mM Hepes buffer and supplemented with 20% heat-inactivated fetal bovine serum (Hazelton Research Products, Reston, VA) and antibiotics (penicillin G, 100 units/ml; streptomycin sulfate, 100lg/ml; amphotericin B, 0.25 ,ug/ml; GIBCO). Samples to be assayed were added at the time of cell plating and the tissue culture dishes were incubated at 37°C under 5% CO2. After a 12-hr incubation, 0.64 ,Ci of [methyl-3H]-thymidine (20 Ci/mmol; New England Nuclear; 1 Ci = 37 GBq) was added per well and the dishes were incubated for an additional 60 hr. The cells were then washed with Hanks' ...

Despite expectations that 2-chlorodeoxyadenosine (2-CdA) would prove active primarily in lymphoproliferative diseases, early reports suggested unexpected high activity of this drug in heavily pretreated children with acute myeloblastic leukemia (AML) at a maximally tolerated dose of 8.9 mg/m2/day for 5 days. In view of these findings, we conducted an escalating dose trial of 2-CdA in adult patients with relapsed or resistant AML. Thirty-six patients who had received extensive prior therapy were treated at 9 dose levels of 2-CdA at daily doses ranging from 5 to 21 mg/m2 for 5 days. 2-CdA eliminated leukemic blasts from the peripheral blood in 32 of 36 cases; however, bone marrow hypoplasia was seen only at daily dose levels > or = 15 mg/m2. We observed a total of 3 complete remissions: 1 at the 15 mg/m2/d dose level and 2 at the 21 mg/m2/d dose level; these responses persisted for 3, 2, and 3 months, respectively. Although prolonged myelosuppression would have been dose-limiting at 21 mg/m2/d for 5 days, the most important adverse effect was the development of a sensorimotor peripheral neuropathy. This reaction, whose onset was substantially delayed after completion of drug treatment, was observed in 2 of 5 patients at the 19 mg/m2/d level and in 4 of 4 evaluable patients at the 21 mg/m2/d level. Pathologically, this process was characterized by axonal degeneration and secondary demyelination. Other side effects included reactivation of a posttransplant Epstein-Barr virus-related lymphoma in 1 patient and tumor lysis syndrome. We conclude that the maximally tolerable dose of 2-CdA in adult patients (17 mg/m2/d for 5 days) in approximately twofold in excess of that previously reported in children and that the limiting toxic effect is a degenerative neuropathic disorder. We confirm that this drug has definite activity in AML, but the magnitude of this effect needs to be determined in larger numbers of patients who have received less extensive therapy. This agent deserves further evaluation in patients with both AML and acute lymphoblastic leukemia at these higher doses and perhaps as part of a preparative regimen for patients undergoing bone marrow transplantation.

Both direct viral cytopathic effects and host immune responses appear to be important in the pathogenesis of hepatitis C virus (HCV) infection. Liver transplantation provides a means to explore the role of the immune system in the development of HCV-related liver damage through comparing the natural history of HCV in patients with different degrees of donor-recipient human leukocyte antigen (HLA) matching. We evaluated 36 patients with recurrent hepatitis C viremia following liver transplantation to determine whether hepatocellular injury or progression to bridging fibrosis occur more rapidly when donor and recipients share HLA alleles. HLA typing for the HLA-A and HLA-B loci was performed by serological techniques and PCR-based oligotyping was used to type alleles of the DRB1, DRB3, DQA1, and DQB1 loci. A median of eight liver biopsies, obtained during a median follow-up of 36 months, were reviewed per patient. Donor-recipient sharing of alleles of HLA-DQB1 or DRB1 was associated with more rapid development of recurrent hepatitis by univariate analysis (chi2=5.7, P=0.02 and chi2=5.54, P=0.02 respectively). However, only sharing of HLA-DRB1 alleles was identified as an independent predictor of reduced time to recurrent histologic injury by multivariate analysis (chi2 =5.74, P=0.02). Furthermore, sharing of HLA-DRB3 and histologic evidence of rejection were associated with more rapid progression to bridging fibrosis both by univariate methods (chi2=4.12, P=0.04 and chi2=4.66, P=0.03 respectively), and by multivariate analysis (chi2=13.01, P=0.001). These findings suggest that HLA class II-restricted immune responses may contribute to the pathogenesis of HCV-related liver injury in liver transplant recipients.

We developed logistic regression models that combine information from the automated CBC and manual 100-cell differential counts to predict bacterial infection. The logistic models were fitted from a case group of 116 patients with proven bacterial infection and a control group of 930 presumably uninfected outpatients. A 4-variable, 15-parameter model, which includes automated absolute neutrophil, manual band, and manual immature granulocyte counts, performed best with a receiver operating characteristic (ROC) curve area of 89%. A more practical 2-variable model including automated absolute neutrophil and manual band counts performed almost as well with an ROC curve area of 86%. The automated neutrophil count-only model is less informative with an ROC curve area of 78%. The combined information from automated and manual differential cell counts more accurately predicts bacterial infection than automated counting alone. Despite these modest improvements, the high cost of manual differential cell counts dictates careful patient selection. The supplemental information gained from manual differential counts is most useful for patients with low to normal neutrophil counts (8,000/microL [8.0 x 10(9)/L] or less). Further studies are indicated to determine the characteristic patient populations deriving maximal benefit from this information.