Matched pancreatic cancer liver metastatic model system reveals cancer cell-dependent organotropism and site-specific tumor microenvironment reflective of human disease

Pancreatic ductal adenocarcinoma (PDAC) is a deadly, highly metastatic disease, driven by an interplay between cancer cells and the metastatic site-specific microenvironment. However, pre-clinical models that robustly capture these interactions within the context of matched primary and metastatic tumors are limited. Here, we present a novel transplant model system for matched pancreas and liver tumors to study PDAC metastatic progression. Using this model, we identified murine PDAC cell lines with distinct liver tropism potentials and defined a transcriptional program associated with enhanced liver metastasis. This signature was enriched in malignant cells from human PDAC liver metastases across multiple independent datasets and was predictive of survival. Integrative ligand-receptor interaction analyses, multiplex protein profiling, and spatial immune cell profiling revealed that PDAC liver metastases develop within a distinct immunosuppressive microenvironment characterized by predicted enhanced inhibitory signaling and altered immune organization. Notably, CD4 and CD8 T cells were more proximal to cancer cells in liver metastases compared to primary pancreatic tumors, suggesting site-specific tumor-immune interactions. Finally, we demonstrate the utility of these model systems for interrogating cell line-dependent and T cell-regulated mechanisms of metastatic progression. Collectively, this work establishes a tractable platform for studying matched primary and metastatic PDAC and identifies tumor-immune signaling networks associated with immunosuppressive liver metastatic progression.