(P 233) DETERMINING THE CONTRIBUTIONS OF PD-L1 AND STAT1 TO THE UPS ANTI-TUMOR MICROENVIRONMENT

Objective: Sarcoma is a heterogeneous group of cancers with over 70 subtypes and survival rates have remained relatively stagnant in recent decades. Sarcoma treatment strategies rely on surgical resection and immunotherapy interventions applied to other malignancies are largely underutilized. We previously found that a subset of Undifferentiated Pleiomorphic Sarcoma (UPS) cases harboured tumor-infiltrating lymphocytes (TILs), expressed high levels of PD-1 ligand (PD-L1) RNA and had good clinical outcomes. Through RNA-sequencing, a list of genes differentially expressed between UPS cases with high and low PD-L1 expression was determined. Genes in the Th1 pathway were significantly overexpressed within this list. We hypothesize that tumor-related genes from the Th1 pathway may be related to an anti-tumor microenvironment. Our aims are to identify the potential roles of Th1 pathway genes, PD-L1 and STAT1, in tumor-immune interactions. This will be accomplished by first creating and characterizing patient-derived cell cultures followed by examining the functional differences in cytokine output and TIL-sensitivity with the use of siRNA pathway modulation.

Methods: Sarcoma specimens from patients undergoing open biopsy or surgical resection have been collected for the downstream isolation of both tumor cell lines and the respective TIL cultures. Various dissociation and tumor culturing protocols were optimized for each case, and methods were tested on multiple subtypes. Cultured UPS cells were validated as tumor cells using Whole Exome Sequencing (WES) to confirm that the cells reflected the patient samples. The gene expression of PD-L1 and STAT1 in bulk tumor and cultured cells was investigated using qPCR. A multiplexed Luminex assay was used to identify what cytokines are secreted from UPS cell lines. The expression of STAT1 and PD-L1 was disrupted using siRNA methodologies within the cell lines before interferon gamma (IFN- γ) treatment. The siRNA-mediated changes in gene expression profiles with or without IFN- γ will be analyzed.

Results: Due to the heterogeneity of sarcoma the optimal culturing conditions were variable, therefore techniques were optimized on a case-specific basis. Of 27 attempted cases, 21 showed growth beyond passage 5. WES of bulk tumor, patient blood and cell line DNA confirmed that the cultured cells reflected the patient sample as variants unique to the patient tumor were identified in the cell lines (Fig. 1). Cell lines resulting from bulk tumors with relatively lower PD-L1 mRNA levels were found to secrete high concentrations of IL-6 and CCL2, where as a cell line from a PD-L1 high tumor secreted more IL-10 and soluble PD-L1 (Fig. 2).

Conclusion: A subset of patient tumors can form successful cell cultures and reflect molecular characteristics of the original patient sample. Bulk UPS tumors co-express PD-L1 and STAT1 in both RNA-sequencing and qPCR experiments. Further, UPS cell lines have unique cytokine secretion profiles. Future experiments will explore if PD-L1 and STAT1 in UPS cells functionally contribute to cell-cell interactions and cytokine profiles, which could suggest tumor-driven microenvironment conditions. Specifically, differences in cell-cell interactions between wildtype cells and siRNA-modulated cells will be determined via immune cell co-culturing assays that use HLA-A matched TIL populations. Simultaneously, conditioned media from both wildtype and siRNA-modulated cell lines will be assayed using the Luminex platform. By creating a model, characterizing it, and assessing the functional consequences of STAT1 and PD-L1 within cell autonomous and non-autonomous settings, we aim to illustrate how tumor-directed interactions may shape the UPS microenvironment.