Characterization of CBX-250, a First-in-class TCR-mimetic-based T-cell Engager Targeting a Cathepsin G Peptide-Hla Complex for the Treatment of Myeloid Leukemia

Abstract While T-cell engaging therapies have demonstrated clear clinical benefit in the treatment of B-cell malignancies, their application to myeloid malignancies remains a challenge due to the limited repertoire of suitable tumor specific surface antigens. Targeting HLA-restricted peptides (pHLA) derived from intracellular cancer antigens provides an opportunity to explore the entire cancer proteome. CG1 (FLLPTGAEA) has been validated as an HLA-A*02:01 restricted leader peptide from Cathepsin G (CTSG) and is abundantly presented by leukemic versus normal myeloid cells. TCR-mimetic (TCRm) antibodies are ideally suited to target pHLA with high potency and specificity and can be engineered and manufactured using standard antibody technologies. Here we report the preclinical characterization of CBX-250, a novel CG1/HLA-A2 TCRm-CD3 bispecific T-Cell Engager (TCE). CBX-250 induced potent T-cell activation and T-cell mediated killing of leukemia cell lines with varying levels of target antigen density with sub-nM EC50 in vitro. In vivo, a closely related precursor of CBX-250 demonstrated potent tumor control in various AML and CML CDX models at doses as low as 0.01mpk. Although CTSG is a serine protease stored in neutrophil azurophilic granules, the CG1 peptide is preferentially presented by leukemia cells, as validated by mass spectrometry. Moreover, when co-culturing PBMCs with HLA-A2 neutrophils, we did not observe any CBX-250-mediated T-cell activation or IFNγ production. In silico predictions identified HLA-A2-restricted human peptides that share sequence or structural similarity to CG1. We determined the cross-reactivity risk of CBX-250 to be low, based on T-cell activation and T-cell mediated cytotoxicity assays against cells pulsed with supra-physiological concentrations of these peptides. Moreover, we observed no activity against target-negative cancer cells, including a CTSG-KO cell line. To further assess the specificity of CBX-250, we screened a panel of normal cells and found no meaningful T-cell activation or IFNγ production at CBX-250 concentrations exceeding its EC90. Finally, the CBX-250 precursor molecule demonstrated favorable safety in a double transgenic mouse model expressing human HLA-A2 and CG1. Following sequence optimization, CBX-250 demonstrated robust serum and pH stress stability and a favorable melting temperature, while retaining excellent potency and specificity. A full suite of analytical and biophysical assessments supports CBX-250’s favorable developability profile. In summary, these data provide strong preclinical evidence of the potency, specificity, safety, and developability of CBX-250, a novel, first-in-class, off-the-shelf, TCRm-based TCE for the treatment of myeloid malignancies. Citation Format: Geraldine Paulus, Benjamin Lee, Preethi Sankaran, Tanzila Rahman, Jessica Jimenez, Delainey O'Connor, Simon Yue, Yu Huang, Melissa Bikowitz, Sarah Jaffe, Shawn O'Malley, Bhupal Ban, Galina Gabriely, Tao Wang, Amanda Mak, Michael Princiotta, Chunhua Shi, Helen He, Ningping Feng, Jun Yan, Timothy Heffernan, Gheath Alatrash, Jeffrey Molldrem, Dmitri Wiederschain. Characterization of CBX-250, a first-in-class TCR-mimetic-based T-cell engager targeting a cathepsin G peptide-HLA complex for the treatment of myeloid leukemia [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1236.