Generation of leukaemia antigen-specific donor lymphocyte infusions powered by streptamer-based selection

Background: Donor lymphocyte infusions (DLIs) after allogeneic hematopoietic stem cell transplantation have the potential to generate a desirable graft-versus-leukemia (GVL) effect, but bear the risk of eliciting a noxious graft-versus-host disease (GVHD). To improve the GVL effect and to minimize the risk of a GVHD, a positive selection of leukemia (antigen)-specific T cells would be highly desirable. In this study we focused on the leukemia-antigen Wilms Tumor gene 1 (WT1). In patients with myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML), a good correlation of WT1 expression and the number of leukemia blasts could be demonstrated. Several groups described immunogenic T cell epitopes derived from the WT1 protein. Material and methods: Here, we used the technology of streptamers available on a GMP level to detect the frequency of HLA-A2 restricted CD8+ T cells in the naive peripheral blood (PB) from both healthy donors (HDs) and AML patients. Such WT1-specifi c CD8+ T cells were further characterized for the expression of CD27, CD28, CD45RA, CCR7 and CD107a. In the next step, WT1-specific cells were positive selected by MACS columns after labeling with streptamers and thereafter immunopheno-typed. Moreover, mixed lymphocyte peptide cultures (MLPCs) were preformed to enrich WT1 specific T cells derived from the PB of HDs. At last, WT1 specific T cells were evaluated in a standard 51-chromium release assay for their cytotoxicity. Results: 17 of 33 HDs showed naive WT1 specific T cell frequencies of 0.5 to 1.4% of all CD8+ T cells. In four of ten AML patients in complete remission, also 0.6 to 6.0% of WT1-specific T cells could be detected. These cells revealed to be CD8+CD27-CD28-CD45RA+CD107aCCR7- effector T cells in flow cytometry. After positive selection by MACS columns a purity of 20-80% could be achieved for WT1-specific T cells. After a maximum of three rounds of MLPC, only a frequency of 2-5% could be achieved, thus demonstrating the power of the streptamer technology. In cytotoxicity assays, WT1-specific CD8+ T cells were able to lyse 60-100% of HLA-A2+WT1+ AML cells at an effector/target ratio of 20:1. Conclusion: In summary, the streptamer technology allows to select a highly pure fraction of WT1-specific effector T cells with cytotoxic properties. In analogy to DLIs specific for viral antigens, production of leukemia specific DLIs is feasible on a GMP level. Further leukemia antigens are currently evaluated by our group.