鉴定NSCLC中KRAS抑制剂耐药的免疫抑制机制

Mutations in the KRAS oncogene by substitution of glycine at position 12 (G12C, G12D, G12V) which locks the protein in its active conformation, are commonly associated with pancreatic, colorectal, and NSCLC. Recent breakthroughs in drug design, have resulted in generation of covalent inhibitors such as Sotorasib (AMG510) and Adagrasib (MRTX849), which target the G12C mutation and have received FDA approval following successful clinical trials. Research efforts are now focused on discovering inhibitors for other KRAS mutations, including the noncovalent KRAS-G12D inhibitor MRTX1133, offering new hope for KRAS-driven cancers. While these inhibitors have shown early success, resistance to treatment frequently arises, and the mechanisms behind this resistance remain unclear. To address this critical challenge, we are leveraging syngeneic mouse models and KRAS mutant allele-specific cell lines to iidentify the molecular pathways responsible for resistance to allele-specific KRAS inhibitors (KRASi). By generating a panel of acquired resistant cell lines from both murine and human NSCLC models, we have aim to identify the molecular drivers of resistance. Proteomic profiling using RPPA analysis elucidated YAP1/TEAD and PDK1 (3-phosphoinositide-dependent kinase-1) pathway to be consistently upregulated in cells resistant to MRTX849 (G12Ci) and MRTX1133 (G12Di). Notably, resistant cells regained sensitivity to KRAS inhibitors when treated in combination with a TEAD inhibitor or a PDK1 inhibitor in vitro and in vivo . Using genetic knockout and gain-of-function models, we further established that YAP1 and PDK1 were necessary and sufficient to impart resistance to KRASi. To elucidate the tumor immune microenvironment of KRASi resistant tumors, we performed flow cytometry-based immune profiling of the primary tumors to assess immune cell subsets. Overall, in the KRASi resistant tumors there were robust increases in CD4+ T cells, but significant decreases in the proliferating CD8+ TILs and Effector/Memory CD8+ T-cells while having a significant elevation in the number of exhausted CD8+ T-cells, compared to the sensitive tumors. These results indicate that the acquisition of resistance to these direct KRAS inhibitors grossly alters the immune microenvironment. We utilized the genetic knock-in and knock-out murine models for PDK1 and YAP1 and performed gene expression screen and identified differentially regulated chemokines and transcription factors. There was a significant correlation between YAP1 expression and multiple chemoattractants of immunosuppressive populations, such as CXCL1, CXCL3, and CXCL5 which could function in recruitment of immunosuppressive cells into the tumor microenvironment, such as neutrophils and MDSCs, which play significant roles in dampening the antitumor T cell response. Our preliminary data indicates a reversal of the immunosuppressive tumor microenvironment of KRASi-resistant tumors that have either lost YAP1 or PDK1 expression or in which these pathways have been pharmacologically inhibited. These results indicate that signaling pathways like YAP/TEAD and PDK1 may have a functional role in modulating the tumor immune responses in KRASi resistance, and therefore understanding the underlying mechanisms may be critical to developing novel approaches for immunotherapy combinations to target KRASi-resistant NSCLC.