Functional PRAME Specific T Cells Can Be Cultured From CD8+ Cells From Healthy Donors but Not From Patients at First CR: Implications for Immunotherapeutic Strategies in AML.

Abstract

Abstract 4139 BackgroundTherapeutic vaccination with dendritic cells (DCs) is regarded as an option for the treatment of minimal residual disease in acute myeloid leukemia (AML). In this respect, defined epitope sequences derived from leukemia-associated antigens (LAA) are an attractive source of antigen. WT1 and PRAME are both LAA that are over-expressed in various hematological malignancies and that share a potential for the induction of cellular immune responses. Expression rates in AML of each of these LAA are high, ranging from 60 to 90%. To evaluate the feasibility of using these LAA in DC based vaccine formulations in AML, we investigated the relative cytotoxic T lymphocyte (CTL) priming efficiencies in response to WT and PRAME derived HLA-A2 binding epitopes in healthy donors (HD) and leucapheresis (LF) material or peripheral blood (PB) samples of AML patients in first complete remission (CR). MethodsCD8+ cells were isolated from peripheral blood of HLA-A2+ HD and from G-CSF stimulated HLA-A2+ patients and stimulated with peptide-loaded DCs. DCs were derived from the human HLA-A2 positive AML cell line MUTZ-3 (MUTZ3-DCs) (Santegoets, et al. Cancer Immunol Immunother. 2006). The peptides used were WT1 (126-134) and PRAME (100-108). Irradiated CD8-cells from the same donor were also added as helper cells. CD8+ cells were plated at one million per well in 8 to 30 wells per donor or patient. CD8+ cells were weekly restimulated with peptide-loaded MUTZ3-DCs. HLA-tetramer positivity (Tm+) for the WT1 and PRAME epitopes was weekly screened by flow cytometry. Wells were scored positive if more than 0.02% Tm+, viable CTLs were present for at least 2 weeks. Functionality of Tm+ cells was studied by analysis of IFN-gamma production after stimulation with the JY cell line loaded with relevant peptides or with an irrelevant control peptide. PRAME Tm+ CTLs were isolated by flow cytometric sorting and a CTL clone was derived by limiting dilution. Recognition of HLA-A2 and PRAME positive tumor cell lines and patient samples by the CTL clone was tested in a direct cytotoxicity assay and ELISPOT assay. ResultsOn the whole, frequencies of the induced peptide-specific CTLs were low as shown in the Table below. Priming efficiencies were highest for PRAME. Furthermore, both PRAME and WT1 specific T cells derived from HD produced IFN-gamma in an epitope specific fashion, confirming functionality of the primed T cells. Most notably, no PRAME specific CTLs could be cultured from the patients' blood, whereas a PRAME specific CTL clone, cultured from sorted Tm+ CTLs from HD, recognized and lysed HLA-A2 PRAME positive tumor cell lines and patient-derived AML blastsHealthy Donor or Leucapheresis ProductTm+ rates after co-culture (per no. of tested HD/patients)Total no. of Tm+ co-culturesMedian % Tm+ cells per positive culture (range)Number of restimulations at maximum of Tm +PRAMEHD3/421/500.10% (0.02-1.07)1-5WT1HD2/54/1040.07% (0.03-0.09)1-4PRAMELF0/40/44*0.00%-WT1LF1/41/440.05%4*LF vs HD, p<0.001 by Chi-Square test ConclusionsLAA-specific CD8+ T cell frequencies were generally low. Most consistently successful priming results were obtained for PRAME in HD and functional CTL were derived thereof indicating that this peptide may be an attractive candidate for vaccination strategies. However, priming of patient samples was unsuccessful, implying possible anergy induction or clonal deletion. As a consequence, active immunotherapy with PRAME-loaded DC may require combined therapy with matched allogeneic donor lymphocyte Infusion (DLI). Disclosures:No relevant conflicts of interest to declare.