Abstract B087: Impact of RAS inhibition on T cell maturation in the PDAC microenvironment

Abstract

Pancreatic ductal adenocarcinoma (PDAC) is the third leading cause of cancer deaths in the United States with a dismal 5-year survival rate of 13%. More than 90% of pancreatic tumors are driven by oncogenic mutations in Kirsten rat sarcoma viral oncogene homolog (KRAS). Mutation-specific and pan-RAS inhibitors have shown great promise in ongoing clinical trials, but primary and acquired resistance is common. It is therefore critical to understand how these inhibitors impact the tumor immune microenvironment to inform how these drugs can potentially be combined with immunotherapies. To understand the direct effect of RAS inhibition on T cells, an orthotopic PDAC mouse tumor model with inducible and reversible pancreas epithelial cell-specific oncogenic KRAS expression was generated. Mice were either implanted with inducible KRAS (iKRAS) tumor cells or basement membrane extract control and administered doxycycline to induce KRAS G12D expression. A 25-color spectral flow cytometry panel was generated to understand the impact of pharmacologic and genetic RAS inhibition on infiltrating T cells. With this we were able to characterize differences in the composition of tumor-infiltrating T cells and in the activation, memory, and exhaustion status of CD8+ and CD4+ T lymphocytes across these test conditions. In addition, we ran the same panel on liver tissue to study potential changes in a pre-metastatic niche. As expected, both genetic inactivation of KRAS (RAS-OFF) and pharmacologic inhibition of RAS (RASi) reduced the tumor burden compared to the untreated control (RAS-ON). We did not observe differences in the total proportion of T cells (%CD3+ of CD45+) within the tumor. However, expansion of effector T cells within the CD8+ T cell compartment was only observed in the RAS-OFF condition within the pancreatic tumor microenvironment. A higher frequency of terminally exhausted CD8+ T cells was also noted in the RAS-OFF mice compared to the RASi and RAS-ON conditions, further highlighting discrepancies in the differentiation of these cells upon RAS inhibition. In the liver, the phenotype of CD8+ T cells treated with RAS inhibitor was similar to that of RAS-OFF. Additionally, while no differences were observed in the frequency of regulatory T cells (%Foxp3+ of CD4+) in tumors, a lower proportion of regulatory T cells was noted in the liver, indicating potential tissue specific differences. Together, these findings suggest that pharmacologic and genetic KRAS inhibition differentially remodel the intratumoral T cell compartment, whereas both induce similar T cell changes within the liver premetastatic niche. Future studies will investigate the mechanisms underlying these phenomena to inform new therapeutic combinations for PDAC.