(P 105) STING THERAPY INDUCES IMMUNOGENIC ANTI-TUMOR RESPONSES THAT ARE INDEPENDENT OF TUMOR STING EXPRESSION IN THE KP MODEL OF SOFT TISSUE SARCOMA

Objective: Undifferentiated pleomorphic sarcoma (UPS), is one of the most common, aggressive, and metastatic soft tissue sarcomas (STS) in adults. Systemic therapies against UPS are lacking and STS tumors are insensitive to most immunotherapies, including immune checkpoint blockade therapy. Generally, the tumor microenvironment (TME) of many STS, including UPS, are characteristically low in tumor infiltrating lymphocytes (TILs) and contain an abundance of tumor associated macrophages and myeloid derived suppressor cells. In previous studies, intratumoural (i.t.) activation of the STimulator of INterferon Genes (STING) pathway, using the murine specific agonist DMXAA, recruited TILs into the TME in the murine KP model of UPS and induced survival in 50-75% of treated mice. To assess the clinical translation feasibility of this immunotherapeutic strategy, investigating agonists capable of activating both human and murine STING must be evaluated. As a mechanism of evading the immune system’s ability to identify and destroy cancer cells, many cancers have evolved to downregulate STING expression or have defective STING pathway signaling abilities. Although STING immunotherapy can induce immunogenic tumor clearance, it is unclear which cell types are critical contributors to this outcome, and if STING activation within tumor cells is conducive to this anti-tumor phenotype. To determine whether tumor STING signaling contributes to STING-mediated tumor clearance, STING therapy must be assessed in UPS tumor cell lines that do and do not engage in competent STING pathway signaling. Therefore, the objectives of this study are 1) to evaluate the anti-tumour potential of and TME modulation of the translational STING agonist, E7766 and 2) to characterize the importance of STING expression in host and UPS cells in the context of E7766 therapy.

Methods: Immune competent C57Bl/6 mice were orthotopically engrafted with a syngeneic murine KP UPS cell line in the right hindlimb muscle. Once palpable, mice were treated with 4mg/kg of E7766 or 18mg/kg of DMXAA i.t. Following treatment, mice were divided into STING monotherapy or STING + anti-PD1 immune checkpoint blockade therapy groups which were followed for survival. Another cohort of DMXAA or E7766 treated mice were sacrificed at various time points to 1) evaluate the systemic effects of STING therapy using a Luminex 45-plex assay or 2) assess the impacts of STING therapy on the TME using flow cytometry and transcriptomics. Tumour volume measurements and bioluminescence were measured over time during survival studies. Any surviving mice were re-challenged with UPS in the contralateral limb. To assess STING expression in our UPS STING wildtype cells and to confirm that STING was knocked out of our UPS cell line, we completed western blots probing for STING and known proteins that are phosphorylated downstream of STING. To understand the role of STING expression in host cells, mice lacking the Sting1 gene (Goldenticket mice) were engrafted with UPS cells, treated with 4mg/kg of E7766 or 18mg/kg of DMXAA i.t, and followed for survival. Cells were treated with equimolar amounts of E7766 or DMXAA and were compared to an HBSS control. To determine the importance of UPS STING expression, immune competent C57Bl/6 mice were engrafted with UPS cells that have been genetically modified to lack Sting1 expression and treated with 4mg/kg of E7766 or 18mg/kg of DMXAA i.t and followed for survival.

Results: Of mice treated with E7766 monotherapy, 25% cleared their primary UPS tumors (Fig 1B). Mice who cleared their initial UPS tumors following E7766 treatment were re-challenged with an additional inoculation of UPS cells and displayed 100% clearance, indicating that adaptive immune protection was conferred (Fig 1 C). Interestingly, E7766 induced the greatest systemic production of interferon beta and CXCL9 at 6hrs following therapy relative to control and DMXAA treated groups (Fig 1D). Flow cytometry studies of tumors 1-week following STING therapy displayed that E7766 induces similar lymphocyte infiltration signatures to DMXAA (Fig 1E&F). Moreover, E7766 appeared to induce a greater transcriptomic enrichment of markers associated with negative regulation of immune responses, including PD-1/PD-L1 (Fig 1G). Excitingly, in the E7766 + anti-PD1 combination therapy group, survival increased as 42% of treated mice eradicated their UPS tumors (Fig 1H). Western blots of protein lysates from UPS STING WT and UPS STING k/o cell lines showed that in response to E7766 and DMXAA treatment, the WT cell line expresses STING and both STING and IRF3 are phosphorylated in response to STING therapy (Fig 2A). Conversely, these observations are abrogated in the UPS STING k/o blots (Fig 2A). In vivo experiments evaluating the contributions of STING expression in host and tumor cells, we found that host STING expression was required to observe the survivor phenotype in both DMXAA and E7766 treated mice (Fig 2B&C). Moreover, mice bearing STING k/o UPS tumors demonstrated an improved survival phenotype in response to E7766 therapy, as 57% of treated mice eradicated their tumors (Fig 2D&E). Single cell mRNA sequencing experiments will be completed to determine which cell types in the UPS TME are key contributors to STING signalling following therapy.

Conclusion: We have shown that E7766 mediated STING activation is a promising immunotherapeutic strategy for UPS in our murine model. E7766 therapy can induce tumor clearance and adaptive immune protection against UPS re-challenge. Excitingly, E7766 STING therapy can also sensitize our model to anti-PD1 ICB therapy. We have shown that therapeutic responses to STING agonist therapies are dependent on host STING expression and not tumor STING expression. In fact, UPS tumors deficient in STING may show improved therapeutic responses as our data currently suggests that UPS tumors lacking STING expression respond better to E7766 therapy. Ultimately, this study demonstrates the potential opportunity for clinical translation of STING as an immunotherapy for STS which could significantly improve outcomes for this patient demographic.