In Vitro Generated Dendritic Cells of Leukemic Origin Predict Response to Allogeneic Stem Cell Transplantation in Patients With AML and MDS.

Allogeneic stem cell transplantation (alloSCT) is the treatment of choice for many patients with acute myeloid leukemia (AML) and myelodysplastic syndrome. The presentation of leukemic or allospecific antigens by malignant blasts is regarded as a crucial trigger for an effective allogeneic immune response. Conversely, insufficient stimulatory capacity by the leukemic blasts is thought to be a relevant escape mechanism from cellular immunotherapy (alloSCT). Our purpose was to test, whether the ability of malignant blasts to differentiate in vitro toward dendritic cells of leukemic origin (DCleu) is associated with clinical outcome. We isolated leukemic blasts from peripheral blood or bone marrow of AML and myelodysplastic syndrome patients before alloSCT (n=47) or at relapse after alloSCT (n=22). A panel of 6 different assays was used to generate DCleu in vitro. Results were correlated with clinical outcome. DCleu could be generated from all 69 samples. Significantly higher mean frequencies of DCleu were found in clinical long-term responders versus nonresponders to SCT (76.8% vs. 58.8%, P=0.006). Vice versa, the chance for response to SCT was significantly higher, if a DCleu+/dendritic cells (DC) ratio of >50% could be reached in vitro (P=0.004). Those patients were characterized by a longer time to relapse (P=0.04) and by a higher probability for leukemia-free survival (P=0.005). In vitro generation of DC and DCleu from leukemic blasts correlated with the clinical outcome. This observation may support a role of leukemic antigen presentation by "leukemia-derived DC" for the stimulation of an allogeneic immune response in AML.

Cytokine Release Patterns in Mixed Lymphocyte Culture (MLC) of T-Cells with Dendritic Cells (DC) Generated from AML Blasts Contribute to Predict anti-Leukaemic T-Cell Reactions and Patients' Response to Immunotherapy.

To enlighten interactions between autologous, allogeneic or T-cells from patients after stem cell transplantation with leukaemia-derived-dendritic-cells containing dendritic cells or blast containing mononuclear cells (n = 21, respectively), we determined cytokine-concentrations (interleukin 2, 4, 6, 10, tumor-necrosis-factor-α, interferon-γ) in supernatants of mixed-lymphocyte-culture and in serum (n = 16) of 20 patients with acute myeloid leukaemia and three patients with myelodysplastic syndromes by cytometric-bead-assay. We correlated our data with lytic capabilities of stimulated T-cells in a fluorolysis-assay and clinical data: Dendritic-cell-/mononuclear-cell-stimulation of T-cells resulted in increased cytokine-levels in culture-medium compared to serum. There were no significant differences between cytokine-patterns of cases with/without lytic T-cell-activity, response to immunotherapy (stem cell transplantation/donor-lymphocyte-infusion) or graft-versus-host-disease. However, some predictive cytokine-cut-off-values for antileukaemic T-cell-activity, patients' response to immunotherapy and graft-versus-host-disease could be defined. Cytokine-profiles alone, without functional assays, are no useful tool to predict antileukaemic T-cell-function, although they can indicate lytic T-cell-activity, patients' response to immunotherapy and graft-versus-host-disease.

Various 'dendritic cell antigens' are already expressed on uncultured blasts in acute myeloid leukemia and myelodysplastic syndromes.

AIM AND METHODS: Leukemia-derived dendritic cells (DC(leu)) potentially present the whole leukemic antigen repertoire. We studied antigen-expression profiles of blasts/dendritic cells (DCs) generated from 137 acute myeloid leukemia (AML)/49 myelodysplastic syndromes (MDS) patients with six different DC-generating media by flow-cytometry combining expression of blast/maturation and DC antigens (DCA:CD1a,b,c, CD25, CD40, CD80, CD83, CD86, CD137-L and CD206). RESULTS: First, DCA are regularly and variably expressed on uncultured blasts/mononuclear cells (MNCs). Individual patients' DCA profiles must be evaluated before DC-culture to find suitable DCA to estimate quality/quantity of DC after culture. Second, after culture in every patient, at least one marker fulfilled these criteria. Third, different DC-generating methods showed varying efficiency to generate DC: not every method was always successful. Fourth, individual FACS-DCA profiles showed a successful DC/DC(leu) generation with at least one of three previously tested methods in every given AML/MDS case. Fifth, pooling results of all selected best methods in every given case, 28/30% DC were generated from AML/MDS samples: >60% viable DC, on average 49/56% mature DC and on average 36% of blasts were convertible to DC(leu) resulting in on average 49% DC(leu) of AML-DC. CONCLUSIONS: Individual DCA-expression profiles should be evaluated before culture to evaluate DC counts/subtypes (mature/viableDC, DC(leu)) in individual patients.

Quality of T-cells after stimulation with leukemia-derived dendritic cells (DC) from patients with acute myeloid leukemia (AML) or myeloid dysplastic syndrome (MDS) is predictive for their leukemia cytotoxic potential.

Myeloid leukemic cells can differentiate into leukemia-derived dendritic cells (DC(leu)), presenting known/unknown leukemic-antigens. Induced anti-leukemic T-cell-responses are variable. To further elicit DC/DC(leu)-induced T-cell-response-patterns we performed (functional)flow-cytometry/fluorolysis-assays before/after mixed lymphocyte cultures (MLC) of matched (allogeneic) donor-T-cells (n=6), T-cells prepared at relapse after stem cell transplantation (n=4) or (autologous) patients'-T-cells (n=7) with blast-containing-mononuclear-cells ('MNC') or DC(leu)-containing DC ('DC'). Compared to 'MNC' 'DC' were better mediators of anti-leukaemic T-cell-activity, although not in every case effective. We could define cut-off proportions of mature DC, DC(leu), proliferating, CD4(+), CD8(+) and non-naive T-cells after 'MNC'- or 'DC'-stimulation, that were predictive for an anti-leukemic-activity of stimulated T-cells as well as a response to immunotherapy. Interestingly especially ratios >1 of CD4:CD8 or CD45RO:CD45RA T-cells were predictive for anti-leukemic function after DC-stimulation. In summary the composition and quality of DC and T-cells after a MLC-stimulating-phase is predictive for a successful ex-vivo and in-vivo anti-leukemic response, especially with respect to proportions of proliferating, CD4(+) and CD45RO(+) T-cells. Successful cytotoxicity and the development of a T-cell-memory after 'DC'-stimulation could be predictive for the clinical course of the disease and may pave the way to develop adoptive immunotherapy, especially for patients at relapse after SCT.

Canine-DCs using different serum-free methods as an approach to provide an animal-model for immunotherapeutic strategies.

Animal-models are the basis of DC-based human immunotherapies. We describe the standardization of a canine-DC-generation protocol using different cytokines and characterize the quality and functional repertoire of the obtained canine-DCs. DCs were generated from healthy dog-PBMCs under serum-free and serum-containing conditions. DC-quality and -quantity was determined by FACS studying the expression-profiles of DC-/costimulatory- and maturation-antigens before/after culture with canine and human monoclonal-antibodies (cmabs/hmabs). Individual DCAgs-(DC-antigens)-expression-profiles were found before and after culture depending on the agents' mode-of-action. With at least one of three serum-free methods (Ca-Ionophore, Picibanil, Cytokines) sufficient DC-amounts were generated. So, canine-DCs can be regularly generated under serum-free conditions and hmabs additionally to cmabs qualify for staining/quantification of canine-cells/DCs. The canine-DCs were functional, shown by T-cell-activation, -proliferation and antigen-specific CTL-responses. In summary, successful, quantitative DC-generation is possible with serum-free methods. DC-based T-cell-vaccination-strategies evaluated for e.g. AML-patients can be tested in the dog and estimated in clinical studies for DC-vaccination-strategies.