Abstract Disseminated cancer cells (DCCs) arriving to a new organ may either proliferate and form metastasis, get eliminated by immune cells, or enter a quiescent stage. The factors that determine the DCCs’ fate, including getting quiescent DCCs to re-initiate proliferation and successfully evade the immune system, remain poorly understood, in part due to the lack of robust mouse models. We established a pancreatic ductal adenocarcinoma (PDAC) mouse model of dormant DCCs in the liver. An “immunization protocol” was used to mimic PDAC resection in patients: KPC1199 PDAC cells were injected to establish primary tumors and allow an adaptive immune response to develop against the cancer cells; the tumors were surgically resected; and new KPC1199 cells were intrasplenically injected to establish experimental liver metastases. The overwhelming majority of the KPC1199 cells were eliminated. No metastases spontaneously formed during 20 months of observation, but about 3-8 sporadic, single DCCs were found per liver cross section, and 94% of these DCCs were negative for proliferation markers. Although the model is based on entraining an adaptive immune response, T cell depletion resulted in only 25-30% of mice developing just 1-3 metastases each. This suggests that T cell depletion results in metastasis from the small percentage of DCCs that were proliferating. The DCC-hosting mice therefore represents of model to identify factors that can trigger further DCC proliferation and metastatic recurrence. Patients with PDAC often experience significant stress, leading to excess of endogenous glucocorticoids (GCs). Additionally, synthetic GCs are used during e.g., chemotherapy to manage side effects. To mimic elevated GC levels in patients, we treated DCC-hosting mice with synthetic GCs resulting in an increased percentage of proliferating DCCs (increasing from <6% of the examined cells in control mice to about 36%). Yet, no metastases formed. GC treatment also decreased T cell and increased neutrophil liver infiltration. Neutrophils can form protease-containing neutrophil extracellular traps (NETs), which we previously showed trigger DCC proliferation by proteolytic remodeling of the extracellular matrix protein laminin. We found that neutrophils from GC-treated mice spontaneously formed more NETs, and additionally, NET-generated laminin remodeling was observed in the liver metastasis. Together, the data suggest, that GC treatment drive quiescent DCCs to proliferate via neutrophils but alone this is insufficient to cause metastases. Consistent with this interpretation, T cell depletion combined with GC treatment resulted in multiple metastases in all mice examined. Of note, deleting the GC receptor in the DCCs did not reduce metastases, supporting that GCs act on the host and not the cancer cells. In conclusion, our data suggest that for quiescent DCCs to form metastases, they need signals that trigger proliferation and simultaneously, they must overcome immune surveillance. Elevated GC levels or GC treatment in immune suppressed individuals represents such a scenario. Citation Format: Xiao Han, Xueyan He, Yuan Gao, Sicheng Pan, Mikala Egeblad. Glucocorticoids establish a metastatic-promoting tumor microenvironment in a PDAC mouse model [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Tumor-body Interactions: The Roles of Micro- and Macroenvironment in Cancer; 2024 Nov 17-20; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2024;84(22_Suppl):Abstract nr C004.
Read full abstract