Novel TCR-like Bi-Specific T Cell Engaging Antibody Targeting the PR1/HLA-A2 Myeloid Leukemia Antigen

PR1 is a HLA-A2 restricted immune-dominant peptide derived from proteinase 3 and neutrophil elastase - leukemia associated self-antigens aberrantly over-expressed in myeloid leukemia. The persistent PR1/HLA-A2 specific cytotoxic T cell responses are associated with prolonged remission from myeloid leukemia, suggesting that PR1/HLA-A2 would be a potent target for T cell based immunotherapy. Previously, we have successfully developed a high affinity T cell receptor (TCR)-like antibody specific for the PR1/HLA-A2 complex (h8F4), and demonstrated that h8F4 mediated specific lyses of myeloid leukemia cells invitro via antibody dependent cell mediated cytotoxicity (ADCC), and significantly reduced leukemic blasts invivo in murine xenograft model of treatment refractory acute myeloid leukemia. We hypothesize that the therapeutic efficacy of h8F4 would be enhanced by combining T cell mediated cytotoxicity through a novel antibody engineering technology, called, Bi-Specific T cell Engager (BiTE) specific for T cells and tumor associated antigens. The BiTE molecules redirect effector function of polyclonal T cells against the tumor leading to lyses of tumor cells, and early clinical data show promising therapeutic efficacy of this approach. In the current study, we constructed Bi-Specific T cell Engaging antibody: h8F4-BiTE, a single chain Fragment of variable domain (scFv) of h8F4 conjugated with scFv of anti-CD3 antibody – OKT3, and produced h8F4-BiTE proteins using eukaryotic expression system. The h8F4-BiTE proteins retained the similar binding specificity to PR1/HLA-A2 as compared to the parent antibody (h8F4), demonstrated by flowcytometry based binding assays with a series of alanine substituted PR1 peptide/HLA-A2 complexes on T2 cells. The 8F4-BiTE bound to native and endogenous PR1/HLA-A2 complexes presented by HLA-A2 positive or transduced myeloid cell lines such as K562, THP-1, and HL-60. In addition, h8F4-BiTE showed comparable binding affinity to CD3 moiety of T cells. Lastly, h8F4-BiTE proteins activated T cells in a PR1/HLA-A2 specific and dose dependent manner, shown by increased expression of activation markers such as CD69 when co-cultured with PR1-pulsed T2 target cells in various concentrations of h8F4-BiTE proteins. Importantly, the PR1/HLA-A2 specific activation of T cells by h8F4-BiTE proteins was equally efficient among all subsets of T cells regardless of their phenotypes, e.g., CD4 vs CD8, and naïve vs memory, suggesting that h8F4-BiTE would be a highly potent inducer of cytotoxic T cells against PR1/HLA-A2 positive myeloid leukemia. Moreover, we generated multiple h8F4-BiTE variants with specific mutations in frameworks and/or complement determining regions (CDRs), and compared binding affinity and function to wildtype h8F4-BiTE. The h8F4-BiTE variants indeed improved specificity to PR1/HLA-A2 complex by complete elimination of residual non-specific binding to HLA-A2, and increased the binding affinity to CD3 by 10-folds resulting in dramatic 50 fold increase in potency to activate T cells. In conclusion, we have developed a novel immunetherapeutics: h8F4-BiTE targeting PR1/HLA-A2 myeloid leukemia antigen. We demonstrated high binding affinity and specificity of h8F4-BiTE to both PR1/HLA-A2 on myeloid leukemia cells and CD3 on T cells and in vitro activation of T cells in the presence of PR1/HLA-A2 expressing target cells in a dose dependent manner. Finally, we improved antigen specificity and functional activity of h8F4-BiTE via selective site-directed mutagenesis. Further preclinical evaluation of therapeutic efficacy is thus warranted using animal in vivo animal treatment model with NOD/SCID/IL2Rγ ^−/− mice xenografted with human myeloid leukemia cells. This unique TCR-like therapeutic agent adds an additional mechanism of action to PR1/HLA-A2 specific antibody (h8F4) and promises to enhance T cell function against myeloid leukemia.