Targeting Intracellular Mutant NPM1 in Acute Myeloid Leukemia with a Novel Peptide-MHC-Specific T-Cell Bispecific Antibody

Background High-frequency NPM1 mutations are essential for initiating malignant transformation at the early stages of leukemogenesis of acute myeloid leukemia (AML). Mutated NPM1 protein is retained in AML cells to maintain its malignant phenotype, with a relatively conserved alternative reading frame, positioning it as an ideal target for AML immunotherapy. However, the current therapeutic antibodies are capable of recognizing surface antigens, but failing to penetrate intracellular domains, thereby limiting the development of therapeutics aimed at NPM1 mutant proteins. To mimic the feature of T-cell receptor (TCR) that recognizes the complex of major histocompatibility class I and peptide on the cell surface derived from the processed intracellular antigen, we used NPM1^mutA_283-291/HLA0201 complex, as neoantigen, to develop a TCR-like fully human IgG1 antibody, for AML immunotherapy.

Methods In this study, the NPM1^mutA_283-291/HLA0201 complex was synthesized and mouse hybridoma technology was used to prepare monoclonal antibody cells. Based on the human IgG1 antibody, we modified the positive clones to construct a variety of chimeric antibodies and humanized antibodies. The physicochemical properties and anti-tumor activity of these antibodies were evaluated in vitro and in vivo.

Result Utilizing the CHO expression system, we successfully produced the recombinant NPM1^mutA_283-291/HLA0201 complex. To generate antibodies specific to this complex, hybridoma technology was employed, and protein-specific monoclonal antibodies were screened using flow cytometry. Notably, among these antibodies, mouse anti-2E2 demonstrated the most significant disparity in binding affinity to NPM1 compared to the control peptide.

To explore the kinetics of antibody binding, surface plasmon resonance analyses were conducted. The Biacore sensorgrams depicting the interaction of mouse 2E2 antibody at different concentrations with NPM1^mutA_283-291/HLA0201 complex are presented, revealing an affinity constant (KD) of 158nM.

The in vitro binding experiment between different concentrations of monoclonal antibody 2E2 and T2-peptide is depicted using flow cytometry. At varying concentrations, the binding strength of monoclonal antibody 2E2 to T2-NPM1 surpassed that to the T2 control peptide, reaching a maximum of 3.75 times. The OCI-AML3 cell line was utilized with the presence of NPM1 A-type mutation and HLA-A2 positivity. Monoclonal antibody 2E2 demonstrated consistently higher binding intensity to OCI-AML3 cells compared to the control antibody at various concentrations, highlighting its effectiveness in targeting NPM1-positive leukemia.

We employed molecular cloning techniques to construct four humanized antibodies and bispecific antibodies based on the 2E2 mouse antibody. Using Jurkat-Lucia NFAT-CD16 Cells, the ability of antibodies hu2E2-Cam16vhh, hu2E2-GAALIE, hu2E2-H11vhh, and hu2E2-L1H1 to mediate antibody-dependent cellular cytotoxicity (ADCC) was evaluated. All antibodies mediated ADCC of OCI-AML3 cells in a dose-dependent manner, with 2E2-GAALIE exhibiting the most potent cytotoxic effect.

To assess the in vivo therapeutic efficacy of the antibody in the NCG humanized mouse model, a humanized mouse tumor model was established by intravenously transplanting OCI-AML3-LUC tumor cells into NCG mice. On the third day after tumor transplantation, a single dose of the antibody (20mg/kg) or chemotherapy drug interventions was initiated, and the tumor burden of the mice was monitored for 12 days. After tumor injection on the 6th, 9th, and 12th days (corresponding to the 3rd, 6th, and 9th days of antibody treatment), the 2E2 antibody treatment group demonstrated a significant inhibition of tumor growth compared to the negative control group.

Through flow cytometry analysis in 23 NPM1-mutated AML patients at initial diagnosis, we assessed the binding of the 2E2 antibody to CD33-positive myeloid cells in PBMCs. Among the 11 initially diagnosed NPM1-mutated AML patients who were HLA-A2 positive, 9 showed significantly enhanced binding affinity of peripheral myeloid cells to the 2E2 antibody.

Conclusion

NPM1^mutA_283-291/HLA0201 complex is the potential agent targeting for AML of NPM1 mutA. The modified antibodies against the above targets have shown good tumor inhibitory effects in vitro and in vivo. It is expected to provide a new direction for clinical treatment of AML.