Abstract B023: Identification of TME-dependent resistance and emergence of persister cells to KRAS inhibition

Abstract Non-small cell lung cancer (NSCLC) remains a major cause of cancer-related deaths worldwide and is often driven by mutations in the KRAS oncogene. Interestingly, smokers most often bear KRASG12C mutations, whereas KRASG12D mutations are more common in non-smokers. KRASG12C inhibitors were FDA-approved in 2021 and have revolutionized personalized therapeutic approaches for lung cancer. However, resistance is widespread, comprising genetic alterations in RAS, gene amplification, and compensatory pathway activation, along with changes in the tumor microenvironment (TME). For KRASG12D inhibitors now emerging for clinical testing, therapeutic resistance is also predicted. To understand both tumor microenvironmental changes induced by oncogenic KRAS and TME-dependent resistance mechanisms, we developed a new orthotopic model of KRASG12D-driven lung cancer, derived from our published genetically engineered mouse model of KRASG12D NSCLC (Lasse-Opsahl and Barravecchia et al. , JCI Insight 2025). Single-cell RNA sequencing data from lung tumor tissue of mice in which KRAS was inhibited, either genetically or pharmacologically with the pan-RAS inhibitor RMC-7977, demonstrated major changes in several cell clusters. Not surprisingly, both genetic and therapeutic inhibition of oncogenic KRAS reduced proliferating tumor cells. In fact, 95% of proliferating cells were lost in lungs where KRASG12D expression was turned ‘OFF’ and 90% were lost in RMC-7977-treated mice. Gene set enrichment analysis (GSEA) of the proliferating tumor cell cluster confirmed strong oncogenic KRAS signaling activity in these cells and thus, dependency on oncogenic KRAS. Despite loss of most proliferating tumor cells, further sub-clustering revealed that a small subset of tumor cells persisted. These ‘persister’ cells displayed a low GSEA ‘hallmark oncogenic KRAS signaling’ signature in ‘ON’ tumors, and thereby likely allowed evasion of KRAS-targeted elimination. However, upon KRAS inhibition, persister cells were activated and exhibited dramatic upregulation of the GSEA hallmark KRAS signaling pathway signature (39% in OFF and 16% in RMC-7977-treated tumors, respectively). Furthermore, persister cells expressed multiple RAS/MAPK signaling genes, including Raf and Cdc42, likely contributing to evasion and resistance to KRAS inhibition. Genetic and pharmacological inhibition of oncogenic KRAS eliminated not only tumor cells, but also IFN-responsive CD8+ cytotoxic T-cells. Loss of these CTLs may have been caused by reduced antigen presentation and inflammatory cues associated with the dramatic loss of tumor cells observed upon KRAS inhibition. In contrast, KRAS inhibition drove a marked expansion of CTLA4+ CD4+ T cells, indicating adaptive resistance in the TME to KRAS inhibition. In conclusion, our findings reveal major changes in both tumor and T-cell composition following KRASG12D inhibition and potential TME-dependent mechanisms of resistance to KRAS inhibitors. Citation Format: Niloofar Khairkhah, Ali Namvar, Emily L. Lasse-Opsahl, Mostafa M H. Ibrahim, Carlos E. Espinoza, Megan Faunce, Lily Rober, Marina Pasca di Magliano, Stefanie Galban. Identification of TME-dependent resistance and emergence of persister cells to KRAS inhibition 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 (5Suppl₁): Abstract nr B023.