Background: Acute myeloid leukemia (AML) remains one of the highest unmet needs among human cancers. LILRB4 is specifically expressed on M4 myelomonocytic and M5 monocytic AML cells with a lack of expression on normal hematopoietic stem cells and progenitors, making LILRB4 an attractive target for a T-cell redirecting bispecific antibody to treat AML. Using a proprietary bispecific antibody platform, we engineered and optimized a LILRB4/CD3 bispecific antibody demonstrating potent and specific killing of monocytic AML cells in vitro and in vivo. The strong chemistry, manufacturing and control (CMC) attributes and cross-reactivity to non-human primate LILRB4 and CD3 favor this bispecific for rapid advancement to the clinic.

Experimental Procedures: Binding EC50 values for LILRB4/CD3 bispecific antibodies were determined by flow cytometry analysis. The selectivity of the anti-LILRB4 modality to other LILRA/B proteins was assessed by ELISA. The potency of T-cell mediated killing of LILRB4-expressing AML cells was determined by co-culturing human peripheral T cells and monocytic AML THP-1 cells. A human peripheral mononuclear cell (PBMC) assay evaluated potency against LILRB4-positive primary cells (CD14+ monocytes), using CD19+ B cells as a negative control. Tumor bioluminescence kinetics was performed in NSG mice administrated intravenously with human T cells and THP-1-luciferase cells to measure T cell-directed killing of AML cells in vivo. Single intravenous dose of 5 mg/kg was used for PK evaluation in human FcRn transgenic mice.

Results: Two top candidates were selected using biophysical criteria: 1) A monovalent anti-LILRB4 arm together with a monovalent anti-CD3 (1+1); 2) A bivalent anti-LILRB4 arm together with a monovalent anti-CD3 (2+1). The binding EC50 values of the 2+1 or the 1+1 variants to LILRB4 in the THP-1 cells were 0.7 nM and 1.3 nM, respectively. ELISA showed the anti-LILRB4 arm to be highly selective for LILRB4. Both 2+1 and 1+1 variants demonstrated potent killing of the THP-1 cells with EC50 values of 0.5 pM and 4.3 pM, respectively. In the PBMC assay, both variants induced killing of monocytes while sparing B cells, supporting the specificity in inducing T-cell directed killing of the LILRB4-expressing primary cells. The anti-CD3 arm was designed to have a low affinity for CD3 to mitigate the risk of cytokine release while ensuring sufficient T cell activation. Both variants showed potent tumor-growth inhibition at doses as low as 0.2 mg/kg in the NSG mice with the 2+1 molecule showing superiority in vivo. The PK profile of the 2+1 bispecific in human FcRn-transgenic mice was similar to what is expected of human IgG1 with linear clearance.

Conclusions: A novel bispecific LILRB4/CD3 (2+1) has been identified for development to treat AML based on its potency and specificity in killing monocytic AML cells in both in vitro and in vivo preclinical studies.

Citation Format: Tao Huang, Kyu Hong, Krista McCutcheon, Jing-Tyan Ma, Ying Zhu, Mi Deng, Cheng Cheng Zhang, Paul Woodard, Hong Xiang, Xiao Min Schebye, X. Charlene Liao. A novel bispecific LILRB4/CD3 antibody with potent killing of monocytic acute myeloid leukemia cells [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 2 (Clinical Trials and Late-Breaking Research); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(8_Suppl):Abstract nr LB217.