Human sera have shown antitumor effects mediated by tumor-specific immunoglobulin M (IgM) antibodies. Most people who have cytotoxic serum are in good health and show no evidence of exposure to tumor antigens. We characterized the serum of a healthy female adult that was highly lytic to a neuroblastoma cell line via IgM-activated complement (>60% of malignant cells were killed during the 60-min assay). Complement-dependent lysis was not mediated by other classes of serum antibodies (data not shown) which is consistent with the findings of Ollert et al. To identify the target antigen on neuroblastoma cells, we fractionated neuroblastoma cell lysates by ion-exchange chromatography. In the fraction that showed maximal IgM binding, the dominant protein was identified as the 47-kDa translational elongation factor 1alpha (eEF1alpha). We used the donor's B-cells to create hybridomas producing the antibody (B12.6.22) that bound to neuroblastoma cells and mediated cytotoxicity. This antibody recognized eEF1alpha in a specific manner. Sequence analysis of the heavy chain of B12.6.22 showed usage of VH3-23 and JH6 gene segments, with no somatic mutation. The structural similarity of B12.6.22 to antibodies of the innate immune system supports the assumption that natural antibodies are a potential source of therapeutic antibodies.

We have investigated the purging efficacy of positive selection of autologous mobilized CD34 + peripheral stem cells in 22 children with high-risk neuroblastoma. CD34 + cell selection was performed using the method of magnetic-activated cell sorting (MACS). The median purity of the CD34 + cells post selection was 97.6% (range 81.7–99.7). For detection of contaminating neuroblastoma cells before and after CD34 + selection, the chimeric anti-disialoganglioside GD2 antibody delta ch 14.18 was used. Prior to positive selection, various numbers of contaminating neuroblastoma cells were found in 17 patients. After positive CD34 + cell selection, low numbers of neuroblastoma cells were only detectable in four patients. In 18 patients, high-dose chemotherapy was performed and the isolated CD34 + cells were reinfused. In all patients, a rapid neutrophil recovery was seen with a median time to reach 0.5 × 10 9 /l neutrophils of 12 days (range 8–24 days). Nine of the 18 patients are free of progression with a median follow-up of 55 months (range 45–70 months). Two patients are alive with relapse, six patients died due to progression or relapse and one patient died due to secondary AML 10 months after transplant while in remission from neuroblastoma. In summary, we show that, through a highly effective positive selection method, a high purging efficacy can be obtained without compromising the hematopoietic reconstitution capacity of the graft.

We read with interest the article by Kunzmann et al, which states cytoreductive effects of aminobisphosphonates mediated by gamma delta T cells (γδ cells).[1][1] All tested aminobisphosphonates induced significant expansion of γδ cells. This proliferative response was interleukin-2 (IL-2)

We performed HLA-mismatched stem cell transplantation with megadoses of purified positively selected mobilized peripheral blood CD34 + progenitor cells (PBPC) from related adult donors in 39 children lacking an otherwise suitable donor. The patients received a mean number of 20.7 ± 9.8 × 10 6 /kg purified CD34 + and a mean number of 15.5 ± 20.4 × 10 3 /kg CD3 + T lymphocytes. The first seven patients received short term (<4 weeks) GVHD prophylaxis with cyclosporin A, whereas in all the following 32 patients no GVHD prophylaxis was used. In 38 evaluable patients, five patients experienced primary acute GVHD grade I and one patient grade II. In 32 patients, no signs of primary GVHD were seen and GVHD only occurred after T cell add backs. T cell reconstitution was more rapid if the number of transplanted CD34 + cells exceeded 20 × 10 6 /kg. Of the 39 patients, 15 are alive and well, 13 died due to relapse and 10 transplant-related deaths occurred. We conclude that the HLA barrier can be overcome by transplantation of megadoses of highly purified mismatched CD34 + stem cells. GVHD can be prevented without pharmacological immunosuppression by the efficient T cell depletion associated with the CD34 + positive selection procedure. This approach offers a promising therapeutic option for every child without an otherwise suitable donor. Bone Marrow Transplantation (2001) 27, 777–783.

Human gammadelta T lymphocytes play an important role in nonadaptive reactions to infection and early tumor defense. This is the first report that freshly isolated, native gammadelta T cells of some healthy donors can kill human neuroblastoma cells to varying degrees. Their killing ability was increased and maintained during expansion and cultivation with interleukin-2 (IL-2; 400 IU/mL) for as long as 30 days (100% specific lysis at an effector-to-target cell (E:T) ratio of 20:1). gammadelta T lymphocytes without this spontaneous killing ability gained a specific cytolytic activity of 81% +/- 10.4% SD after stimulation with IL-2 for 24 hours. gammadelta cells were isolated from peripheral blood by positive enrichment (using a magnetic cell sorting system; purity, 95.2% +/- 3.2% SD, n = 21). High natural cytotoxic activity against human neuroblastoma cell lines (>50% specific lysis at an E:T ratio of 20:1) was exhibited by one of 11 donors, whereas two of 11 showed medium cytotoxicity (30% to 50% specific lysis). Eight of 11 donors showed very slight or no lytic activity against human neuroblastoma cells (<30% specific lysis). gammadelta T cells were also cytotoxic against Daudi (32.7% specific lysis at an E:T ratio of 20:1), Raji (10.3%), Colo 205 (23.1%), A 204 (54%), K 562 (100%), and SK-N-MC (100%) cells. Isolated gammadelta T cells were grown in Iscove modified Dulbecco medium with IL-2 (400 IU/mL). Increased cell proliferation (38.5% to 182%) was induced with phytohemagglutinin, IL-15, Clodronat, OKT3, or various combinations of these. Results of cold target inhibition assays suggest a natural killer-like activity of the gammadelta T-cell killing mechanism. Peptidase or papain render neuroblastoma cells unsusceptible to gammadelta T-cell killing, suggesting the involvement of antigen peptide(s) in the process of neuroblastoma cell killing. Treatment with acid phosphatase reduced specific lysis by 66.5% +/- 34.1% SD, which suggests a binding to phosphorylated neuroblastoma cell-surface structures in the killing mechanism of gammadelta T cells. Heat shock did not affect the extent of neuroblastoma killing by gammadelta cells. Recognition of neuroblastoma cells by gammadelta cytotoxic T lymphocytes is negatively regulated by major histocompatibility complex I receptors. Evidence for natural and inducible cell cytotoxicity of gammadelta T cells against human neuroblastoma cells, easy propagation, purification, and missing alloreactivity in mixed lymphocytes cultures indicates a role for this subpopulation of T lymphocytes in adoptive immunotherapy.

Recently, we have shown that patients with acute leukaemias and myelodysplastic syndromes (MDS), who showed increasing mixed chimaerism (MC) upon serial PCR analysis after transplant, have a significantly increased risk of relapse. To determine whether the increasing MC in these patients is caused by the reappearance of normal recipient: haematopoiesis or by the reoccurrence of malignant cells, we purified different leucocyte subpopulations and analysed these subfractions with regard to their donor-recipient ratio by a PCR-based method for the analysis of minisatellite DNA regions. In 14 patients [eight acute lymphoblastic leukaemia (ALL), three acute myelogenous leukaemia (AML) and three MDS] subfractions were analysed when increasing MC was first noted upon serial analysis of the peripheral blood. In seven of these 14 patients (four ALL, two AML and one MDS), subfractions were characterized at the time of frank haematological relapse, In all 14 patients investigated with increasing MC, recipient cells were detected in different mononuclear cell subpopulations, In patients characterized during frank relapse, two distinct distribution patterns were found. Patients who relapsed before day +300 (one ALL, two AML and one MDS) showed recipient-derived (normal) cells in addition to blast populations in different mononuclear subsets as well as granulocytes. In patients with acute leukaemias who relapsed after day +300 (two ALL and one AML), only leukaemic cells were found that were of recipient origin, whereas all other haematopoietic cell lines were donor derived.These data show that persistent MC in the early posttransplant period is caused predominantly by normal recipient haematopoietic cells. This finding further supports the hypothesis that a state of mixed haematopoietic chimaerism may reduce the clinical graft-versus-leukaemia (GVL) effect of alloreactive donor-derived effector cells in patients with acute leukaemias and MDS, and thus facilitate the proliferation of residual malignant cells that may have survived the preparative regimen.

We developed a cellular approach to the identification of circulating melanoma cells in peripheral blood using immunomagnetic cell sorting. One hundred seventy-eight blood samples from 129 melanoma patients and 30 samples from healthy persons and nonmelanoma patients were examined. After density gradient centrifugation the interphase was incubated with the mAb 9.2.27. Positive cells were labeled with magnetic microbeads and enriched by immunomagnetic cell sorting. Cells were stained using an alkaline phosphatase-antialkaline phosphatase assay and examined by light microscopy. In spiking experiments, melanoma cells seeded at a concentration of one melanoma cell per ml whole blood could be detected reliably with the assay. Circulating melanoma cells were not found in 30 controls examined, nor were 9.2.27-positive cells found in 41 patients with primary malignant melanoma. In patients with regional lymph node metastases and in patients with disseminated disease, circulating 9.2.27-positive cells could be detected in 3 out of 22 patients (13.6%) and 10 out of 66 patients (15.2%) examined. We present a sensitive and specific immunocytological approach to detect circulating melanoma cells in peripheral blood. The method is not suitable for early detection of metastases but is a valuable tool for further investigating biological characteristics of circulating melanoma cells.

We present our experience with three clinical scale isolation methods for positive selection of CD34+ progenitors from peripheral blood for autologous and allogeneic transplantation in children. A combination of the CellPro device and the Magnetic Activated Cell Sorting system (MACS), as well as two different combinations of MACS systems were used (VarioMACS-SuperMACS and SuperMACS-SuperMACS). With the CellPro-MACS combination (16 separations), a median purity of 96.2% and a median recovery of 42% CD34+ cells could be achieved, whereas the two step MACS systems (55 and 29 separations) showed a median purity of 97.6% and 98.0% and a median recovery of 96.5% and 97%, respectively. Depletion of T cells was profound (4–5 log). A total of 34 patients in the autologous and 18 patients in the allogeneic setting have been transplanted with highly enriched CD34+ cells, obtained by these methods. Only one patient failed to engraft, all other patients showed a rapid and sustained hematological engraftment with the longest follow-up of 3 years. In summary, especially the two step MACS systems have proven to be appropriate tools for enrichment of CD34+ cells, yielding both high purity and good recovery, and can thus be used for tumor cell purging in the autologous setting and for effective T cell depletion in the allogeneic setting.

Killer-cell inhibitory receptors inhibit cytolysis when natural killer cells encounter cells bearing HLA class I molecules.1,2 This phenomenon explains the preferential killing by natural killer cells of tumor cells with reduced expression of HLA class I molecules.3 This finding is the basis of the “missing self” model, in which the HLA class I allele lost by the tumor cells is not available to inhibit the killer cell's lytic machinery. Different types of killer-cell inhibitory receptors have distinct specificity for HLA-A, B, C, and E antigens.1,4 In addition, some natural killer cells also possess receptors that induce lysis of . . .

A new method for the isolation of CD56+ lymphocytes from peripheral mononuclear blood cells is described. Magnetic microbeads conjugated to goat antimouse antiserum in combination with a murine monoclonal anti-CD56 antibody were coated to the CD56+ target cells. CD56+ cells were then isolated with the use of a magnetic cell sorter. The purity of the CD56+ cells was 98.4 +/- 1% (n = 12) with a recovery of the CD56+ lymphocytes of 57.2 +/- 9% (range 48-77%). The natural killer, activity of the CD56+ lymphocytes as well as the interleukin-2-induced proliferative response were not affected by the isolation procedure or the presence of the magnetic microbeads on the CD56+ cells. The described method might be a useful tool for the further characterization of CD56+ cells and their subsets and can easily be upgraded for clinical use in adoptive immunotherapy with CD56+ lymphocytes.

CD34-positive cells were isolated from cord blood ( n = 8), bone marrow ( n = 4) and leukapheresed material ( n = 7), using an immunomagnetic isolation technique, MACS (Miltenyi Biotec, Bergisch Gladbach, Germany). In flow cytometric analysis, cell populations after enrichment revealed a fraction of 96.1% (cord blood), 96.2% (bone marrow) and 98.6% (leukapheresis material) CD34-positive cells. Cells were further stained with antibodies specific for CD44 isoforms: CD44s (SFF-2), CD44v5 (VFF-8) and CD44v6 (VFF-18). CD44-positive cells were detected by directly (FITC, fluorescein isothiocyanate) or indirectly (streptavidin-PE, phycoerythrin)-conjugated fluorochromes in flow cytometric analysis. Analysis was restricted to CD34-positive cells. A high expression of CD44s was noted in all kinds of material under investigation with mean values in the range of 98.6–100%. There was little expression of CD44v6 (mean values in the range of 1.5–3.6%) and very slight expression of CD44v5 (mean values in the range of 0.6–1.4%). The finding that CD34-positive hematopoietic stem cells express CD44v5 and CD44v6 to a very small extent offers the possibility of using antibodies specific to CD44v5 and CD44v6 in immunopurging in the course of autologous stem cell transplantation.

Numerous studies have suggested that interferon-gamma (IFN-gamma) exhibits an inhibitory effect on conceptus development during pregnancy, and recent investigations have shown that decidual CD56(+ +), CD16(-) large granular lymphocytes (LGL) contain mRNA for IFN-gamma. We have investigated the influence of exogenous interleukin-12 (IL-12) and interleukin-2 (IL-2) on IFN-gamma secretion by cultivated LGL and macrophages isolated from first trimester human decidua. The effect of decidual macrophages on IFN-gamma secretion by LGL was also assessed using co-incubation experiments. Neither IL-12 nor IL-2 stimulated the secretion of IFN-gamma by decidual macrophages. IL-12 alone, but not IL-2 alone, stimulated the release of IFN-gamma by LGL. However, IL-2 acted synergistically with IL-12 to enhance the release of IFN-gamma by LGL. Unstimulated and IL-12- and IL-2-stimulated LGL incubated with macrophages exhibited a marked increase in secretion of IFN-gamma compared to those in monoculture. This effect was also seen when the LGL and macrophages were separated by a semi-permeable membrane. The results suggest that interactions between decidual LGL and macrophages, possibly mediated by soluble factors, could play a role in regulating IFN-gamma secretion at the materno-fetal interface and thus contribute to the control of invasion by the trophoblast.