Multiple Myeloma Cross-Presents Neutrophil Elastase and Proteinase 3 and Becomes Susceptible to PR1-Targeting Immunotherapy

Abstract 4973

PR1 (VLQELNVTV) is a human leukocyte antigen (HLA)-A2 restricted nonameric peptide that has been effectively targeted in myeloid leukemia using PR1-peptide vaccine and PR1-specific cytotoxic T cells (CTL). A phase I/II clinical trial has been initiated with PR1 peptide vaccine, which demonstrated complete remission and immunologic responses in patients with acute (AML) and chronic (CML) myeloid leukemia, as well as myelodysplastic syndrome. We recently developed an anti-PR1/HLA-A2 antibody (8F4) and have demonstrated its activity against myeloid leukemia blasts in vitro and in a human leukemia xenograft mouse model.

PR1 is derived from the serine proteases proteinase-3 (P3) and neutrophil elastase (NE), which are normally found within neutrophil azurophil granules and are released into the inflammatory milieu. We have shown that through cross-presentation, a process whereby exogenous antigen is taken up, processed and expressed on cell surface in association with human leukocyte antigen class I molecules, PR1 can be expressed on the surface of cells that do not endogenously express NE and P3. Specifically, we showed that P3 and NE are taken up and cross-presented by antigen presenting cells (APC), including B cells, as well as breast cancer and melanoma cells. Furthermore, cross-presentation by these non-myeloid cells led to their killing by PR1-CTL and 8F4 antibody.

Since multiple myeloma is derived from B cells, which are deficient in NE and P3 but have intrinsic APC functions including cross-presentation, we hypothesized that P3 and/or NE may be taken up and cross-presented by multiple myeloma cells, hence rendering them targets for PR1-CTL and 8F4 antibody. Using RT-PCR and western blotting, we first show that NE and P3 are absent in myeloma cell lines including ARK, ARP-1, OMP-2, LP-1, U-266, IM9, and RPMI 8226. Intracellular staining for P3 and NE after culturing myeloma cell lines with P3 (10 ug/mL) or NE (10 ug/mL) confirmed uptake of P3 and NE by multiple myeloma cell lines; uptake is time dependent with variable plateauing between the cell lines over a 30-hour time course. Using 8F4, the novel PR1-HLA-A2 monoclonal antibody, we demonstrate that P3 and NE are cross-presented by the HLA-A2 positive U-266 myeloma cell line. Peak P3 and NE cross-presentation occurs at 24 hours and 6 hours, respectively. Next, we studied whether PR1 cross-presentation causes cells to be susceptible to PR1-specific killing by PR1-CTL and 8F4 monoclonal antibody. We show that cross-presentation increases susceptibility of U-266 myeloma cell line to killing by PR1-CTL. Specifically, following incubation with P3 (10 ug/mL) and NE (10 ug/mL), we show 30% and 25% killing, respectively, of U-266 cells by PR1-CTL versus unpulsed U-266 cells (10% killing). Furthermore, we show dose-dependent killing of the U-266 myeloma cell line by 8F4 antibody in a complement dependent cytotoxicity assay, with maximum 30% and 20% killing of U-266 cells after incubation with P3 (10 ug/mL) or NE (10 ug/mL), respectively, versus no killing in the unpulsed group. Finally, using PR1/HLA-A2 dextramer staining, we show PR1-CTL in peripheral blood from patients with multiple myeloma following stem cell transplantation (Median, 0. 06%; range, 0. 03%-0. 1%). Our data therefore show that if the uptake of P3 or NE, present in the inflammatory milieu of multiple myeloma bone marrow and extramedullary tumors, leads to PR1 cross-presentation, then PR1-based immunotherapy may be useful to treat patients with multiple myeloma. These results also support a new paradigm linking inflammation and innate immunity to adaptive immune responses to cancer.