HLA Class I Ligandome Analysis In Acute Myeloid Leukemia – Novel T-Cell Epitopes For Peptide-Based Immunotherapy

Data regarding the graft-versus-leukemia (GVL) effect after allogeneic stem cell transplantation (SCT) and donor lymphocyte infusion strongly suggest that T lymphocytes play a major role in the rejection of leukemic cells. Immunotherapy directed against leukemia- associated antigens might elicit specific immune responses that may serve to eliminate minimal residual disease after chemotherapy, or enhance the GVL effect after SCT. To achieve this goal there is need to identify appropriate leukemia associated HLA ligands, which are able to induce specific T cell responses. We here aimed to characterize the HLA class I ligandome in AML patients to provide novel tumor associated antigens (TAA) for peptide-based immunotherapy employing our recently implemented approach of direct isolation and identification of naturally presented HLA ligands by affinity chromatography and mass spectrometry (LC-MS/MS) in AML (Stickel et.al., abstract in Blood 2012). Absolute HLA surface expression on AML cells and autologous monocytes and granulocytes was quantified by flow cytometry. HLA class I ligands were isolated from AML cells as well as bone marrow and peripheral blood mononuclear cell (BMNCs/PBMCs) of healthy donors. LC-MS/MS peptide analysis provided qualitative and semi-quantitative information regarding the composition of the respective ligandomes. Comparative analysis of malignant and benign samples served to identify ligandome-derived TAA (LiTAA) and to select peptide vaccine candidates. The most abundantly detected peptide candidates were checked for immunogenicity by ELISpot and confirmed by intracellular interferon-g staining of CD8⁺ T-cells. Meanwhile 15 AML patients (8 FLT3-ITD mutant) and 35 healthy donors were analyzed. We observed overexpression of HLA class I and II on AML cells as compared to autologous monocytes and granulocytes, with the level of significance reached for HLA class II (p=0,04). A total of more than 12,000 AML derived HLA ligands representing >6,000 different source proteins were identified; of which 2,220 were exclusively represented in AML, but not in healthy PBMC/BMNC. Data mining for broadly represented LiTAA pinpointed 98 TAA as most promising targets. HLA ligands derived from these TAA were presented exclusively on more than 33% of all analyzed AML samples, amongst them already described TAA (e.g. JUP, FAF1) as well as several new leukemia-associated proteins (e.g. MTCH2, METTL7A). Subset analysis of the FLT3-ITD positive AML cohort revealed 21 LiTAA presented exclusively on more than 50% of FLT3-ITD positive AML cases. Additional screening for HLA ligands derived from described leukemia associated antigens revealed overrepresentation for e.g. FLT3, NUSAP, RHAMM and RGS5. Specific CD8⁺ T cell responses were detected against two A*03 epitope pools (pool 1: APLP2, DKGZ, FAF1, MTCH2; pool 2: KLF2, METTL7A, VCIP1, WIPI1) in AML patients. Notably, the chosen A*03 epitope pools did not elicit specific responses of CTL from healthy donors. Taken together, our HLA class I ligandome analysis in AML for the first time identified naturally presented HLA ligands from patients including a vast array of new leukemia associated antigens representing promising targets for a multipeptide-based immunotherapy approach in AML.