Search for a command to run...
Abstract Mutant KRAS (mKRAS) is a dominant oncogenic driver in pancreatic ductal adenocarcinoma (PDAC), making it an attractive neoantigen target for T cell receptor (TCR) T cell therapy. While early clinical trials of mKRAS-targeted TCR-T cell therapy have been initiated, efficacy in PDAC is potentially hindered by low mKRAS epitope abundance, MHC-I downregulation, and resistance mediated by its dense tumor stroma. Hence, robust preclinical models of PDAC are needed to dissect these therapeutic barriers and guide strategies to improve mKRAS-directed TCR-T cell therapies. Here, we present results of a humanized co-culture and mouse model system incorporating human antigen-presentation and recognition components to better reflect clinically relevant immune interactions in PDAC. Murine KPC cell lines derived from the pancreatic KPC model (LSL-KrasG12D/+;LSL-Trp53R172H/+;Pdx1-Cre) and human PDAC cell lines harboring mKRAS were transduced to express a chimeric human-mouse fusion HLA-A*11.01 protein containing human α1 and α2 domains, while the α3 to cytoplasmic domains are of murine origin. Engineered mouse T cells expressing human mKRAS-specific TCRs were co-cultured with HLA-A*11.01 expressing tumor cell lines presenting natural or overexpressed levels of mKRAS epitopes. Antigen recognition by engineered T cells was measured by IFN-γ production. For in vivo studies, chimeric HLA-A*11:01 expressing murine KPC or human PDAC cells were implanted into Rag1-/- mice, followed by transfer of engineered mKRAS-specific mouse TCR-transgenic T cells. Tumor growth and resistance mechanisms were evaluated. Using this humanized TCR-HLA system, both overexpressed and natural amounts of mKRAS in PDAC cell lines led to a recognition and activation of co-cultured T cells in vitro. In the mouse model, KPC tumors overexpressing mKRAS epitope exhibited significantly reduced growth following transfer of mKRAS-specific T cells. Tumor cells isolated from relapsed tumors predominantly lacked mKRAS epitope expression, suggesting selection of antigen-loss variants as mechanism of treatment failure. In summary, our humanized TCR-HLA PDAC models demonstrate recognition of mKRAS epitope in vitro and tumor growth control by engineered mKRAS-specific T cells in vivo. They provide clinically relevant platforms for evaluating strategies to surmount barriers that constrain mKRAS-targeted TCR-T cell therapy in PDAC. Citation Format: Georg Hilfenhaus, Johnathan Arnon, Christian Friese, Lucia Poncette, Arunraj Dhamodaran, Sebastian Stintzing, Thomas Kammertoens, Thomas Blankenstein. TCR-T cell therapy targeting mutant KRAS in pancreatic cancer using a humanized co-culture and mouse model system [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: RAS Oncogenesis and Therapeutics; 2026 Mar 5-8; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(5_Suppl_1):Abstract nr A035.
Published in: Cancer Research
Volume 86, Issue 5_Supplement_1, pp. A035-A035