Abstract 3795: Cross-HLA targeting of synthetic neoepitopes with T cell engagers to eradicate sotorasib-resistant, KRAS-G12C mutant cancer

Abstract Despite the advances provided by covalent inhibitors for the treatment of KRASG12C-driven cancers, the emergence of drug resistance is a challenge. We have exploited the fact that the covalent inhibitor-modified KRASG12C protein is degraded by the proteasome, and drug-modified peptides (hapten-peptides, p*) are presented by MHC-I molecules (p*MHCs), offering tumor-specific, synthetic neoantigens. The marked dissimilarity of a haptenated-neoantigen to self-peptides or traditional neoantigens provides unique advantages, including the ability to engineer companion high-affinity antibodies that can specifically recognize p*MHCs presented on multiple HLA alleles. When formatted as anti-CD3, T cell engager bispecific antibodies (TCEs), these molecules can eradicate drug-resistant tumor cells. Here, we present the development of AETX-R114, a TCE targeting sotorasib-p*MHCs presented on 3 different HLA-A3 supertype alleles, A*03:01, A*11:01, and A*68:01, which is not competed by free drug and demonstrates in vitro and in vivo tumor cell-specific potency. Although we previously provided indirect evidence of p*MHCs (PMID 36250888), here we use immunopeptidomics to directly identify sotorasib, divarasib, and adagrasib-modified p*MHCs presented on the cell surface of inhibitor-resistant cells. Characterization of sotorasib-p*MHCs revealed low copy-number presentation on the cell surface (<1 copy-per-cell, NCI-H2122, A*03) and poor inherent stability (42 min, A*03), particularly relative to other well characterized antigens (PRAME, 17 h, A*02, ) suggesting poor intrinsic immunogenicity. This insight prompted the development of AETX-R114, a 3rdgeneration antibody obtained through affinity maturation of our previously reported R023 clone. R114 achieves pM affinities at 37°C (A*03 KD = 7 pM, A*11, 130 pM, A*68, 335 pM) and exhibits slower dissociation from the target. The high affinity of AETX-R114 dramatically extends the half-life of p*MHCs (>24 h, A*03), and mass spectrometry provides the first direct evidence of increased p*MHC surface density via TCE-mediated stabilization (e.g., 15-fold, NCI-H2122), overcoming the inherent poor stability and low antigen density of KRAS p*MHCs. T cell-mediated cytotoxicity was evaluated in drug-resistant, KRASG12C cancer cell lines. AETX-R114 demonstrated potent killing in multiple adherent & spheroid models expressing A*03, A*11, or A*68. In vivo, AETX-R114 versus isotype control showed significant tumor growth inhibition (84%, day 23) in combination with sotorasib in an NCI-H2122 xenograft mouse model. The unique properties of drug-modified HLA peptides provide new opportunities in anti-MHC targeting immunotherapies. Taken together, our approach unites targeted, mutation-specific therapy with immunotherapy to enable cancer specific, immune-mediated cell killing. Citation Format: Lauren Stopfer, Lorenzo Maso, Sarah Mosure, Weifeng Liu, Sergio Rodriguez, Diamond Mensah, Matthew Southard, Samantha Sze, Takamitsu Hattori, Epsa Rajak, Akiko Koide, Angelina Pizzo, Anusha Keerthi, Sean Toenjes, Brian Vash, Paul Da Silva Jardine, Benjamin G. Neel, Shohei Koide, Christoph Rader, Rajesh Chopra. Cross-HLA targeting of synthetic neoepitopes with T cell engagers to eradicate sotorasib-resistant, KRAS-G12C mutant cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3795.