Abstract IA22: Myeloid derived suppressor cells as mediators of local immune suppression in head and neck cancer

Abstract

Listen

Translate article

Abstract Development of local immune suppression is a major barrier to effective antitumor immunotherapy. Within head and neck squamous cell carcinomas, infiltration of both immature myeloid cells (myeloid derived suppressor cells, or MDSCs) and regulatory T-cells (Tregs) contributes to local immunosuppression. Analysis of The Cancer Genome Atlas head and neck cancer subset reveals that patients with T-cell inflamed tumors also demonstrate robust MDSC and Treg gene expression profiles, indicating that both cell subsets may be limiting effective antitumor immunity in the setting of an otherwise activated immune response. The carcinogen-induced, syngeneic mouse oral cancer (MOC) model demonstrates robust MDSC infiltration and provides a model to study how modulation of MDSCs can alter anti-tumor immunity at baseline and in response to checkpoint inhibitor immunotherapy. Time course analysis of immune cell infiltration into MOC tumors demonstrates an inverse relationship between granulocytic MDSC (gMDSC) and CD8 tumor-infiltrating T-cells (TIL). Antibody-mediated elimination of gMDSC from the MOC tumor microenvironment rescues antigen-specific T-cell responses lost with tumor progression. While CTLA4 mAb treatment alone controls growth of a subset of MOC tumors, combining gMDSC depletion with CTLA4 checkpoint blockade induces consistent CD8 TIL-dependent rejection of established MOC tumors, resulting in immunologic memory and rejection of subsequent MOC cell challenge. Within MOC tumors, MDSCs express the largest individual pool of PD-L1. Combination PD-L1 and CTLA4 checkpoint blockade adds no benefit over CTLA4 blockade alone, suggesting that the effects of gMDSC depletion are not simply due to the elimination of PD-L1 from the tumor microenvironment. CTLA4 mAb treatment appears to have multiple mechanisms including partial depletion of tumor-infiltrating Tregs and direct activation of CD8 TIL. Given these results, we have studied multiple translational approaches to inhibit the function of MDSCs within the MOC tumor microenvironment with the goal of sensitizing tumors to checkpoint inhibition. Given that immune cells exhibit dependence on the γ/δ subunits for signaling through PI3K, we hypothesized that a PI3Kγ/δ-selective small molecule inhibitor (IPI-145) would functionally inhibit MDSCs. We demonstrated that low-dose IPI-145 reduced arginase-dependent MDSC suppression of TIL function and enhanced responses to PD-L1 blockade in MOC tumor-bearing mice. However, higher doses of IPI-145 reversed this enhancement of tumor control due to direct suppression of TIL function. As an alternative approach, we have evaluated the effect of blockade of sema4D, a semaphoring family protein involved in cell migration and potentially involved in the polarization of myeloid cells to an immunosuppressive phenotype. We demonstrated that sema4D mAb monotherapy reduced gMDSC tumor infiltration and T-cell suppressive capacity through multiple mechanisms. Similar to complete gMDSC depletion, sema4D plus CTLA4 mAb treatment induced immune-mediated rejection of established MOC tumors. Blockade of immunosuppressive cell recruitment into tumor through interruption of chemokine signaling has been demonstrated in other solid tumor types. Historical and current evidence demonstrates that gMDSCs recruitment into head and neck tumors appears to be dependent on the CXCL1(tumor)/CXCR2(expressed on gMDSC) signaling axis. Experiments combining small molecule or mAb-based CXCR2 inhibitors with PD-L1 or CTLA4 checkpoint inhibition are underway in the MOC model. Strong lines of evidence suggest that modulation of MDSCs within the tumor microenvironment can partially reverse local immunosuppression and improve responses to checkpoint inhibitor therapies. These and similar findings have widespread implications as local immunosuppression likely limits effective responses to immune-activating anticancer therapies across many tumor types. Citation Format: Clint T. Allen. Myeloid derived suppressor cells as mediators of local immune suppression in head and neck cancer [abstract]. In: Proceedings of the AACR-AHNS Head and Neck Cancer Conference: Optimizing Survival and Quality of Life through Basic, Clinical, and Translational Research; April 23-25, 2017; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2017;23(23_Suppl):Abstract nr IA22.