Role of Tumor Acidosis and Hypoxia in Immune Escape by IFN-γ-induced Transcriptional Induction of PD-L1 via the elF4F Axis on Cancer Cells

Abstract

Immunotherapy is a promising approach in cancer treatment. In recent years, several antibodies targeting immune checkpoint proteins have been approved and are now applied in clinical routine. Immune escape mechanisms as well as response biomarkers and resistance mechanisms of an immune checkpoint inhibitor therapy are of great interest and are subject to many research studies. The TME and its role during immunotherapies attracted special attention in recent years. Different factors in the TME including immune cell-derived IFN- and hypoxia induce expression of the checkpoint protein PD-L1 and thereby mediate T cell exhaustion or apoptosis. In the first chapter of this thesis, the role of an acidic TME and IFN-on cancer cell PD-L1 expression and subsequent immune escape was studied. It is proposed that PD-L1 induction by an acidic tumor pHe combined with IFN- represents an immune escape mechanism which can be targeted by an anti-PD-L1 mAb therapy. This conjoint acidosis- and IFN--mediated PD-L1 induction was exclusively found in immunotherapy-responsive murine cancer cell lines. Therefore, it might serve as a biomarker to distinguish immunotherapy responders from non-responders. Further it could be shown that acidosis- and IFN--induced PD-L1 is mediated by the expression and activation of the transcription factor STAT1. The translation of the Stat1 mRNA was experimentally proven to be elF4F-dependent, whereas the translation initiation complex elF4F can be inhibited by silvestrol representing an additional therapeutic option. Most importantly, the in vitro findings on conjoint acidosis- and IFN--mediated cancer cell PD-L1 expression could be confirmed in vivo. The tumor pHe neutralization with sodium bicarbonate was proven in vivo by acidoCEST-MRI measurements in MC38wt tumors and IFN--dependency of the immune escape was confirmed by IFN--/- mice. In summary, conjoint IFN-- and acidosis-inducible cancer cell PD-L1 expression represents a novel immune escape mechanism and might display a biomarker for immunotherapy response. In the second chapter, hypoxia which, besides acidosis and IFN- induces PD-L1 expression, was imaged non-invasively in vivo. Imaging tumor hypoxia by in vivo OI with reporter cancer cell lines did not prove to be suitable serving as a non-invasive surrogate marker for tumor PD-L1 expression. Short-lived tumor hypoxia neither correlated with tumor PD-L1 expression, nor with CD3+ T cell infiltration. In addition to tumor-resident immune cells, a systemic immune response including the lymphoid system is pivotal for anti-tumor immunity. Lymph nodes are secondary lymphoid organs and play an essential role in the anti-tumor immune response during a checkpoint inhibitor therapy. This underlines the importance of non-invasive in vivo imaging of TDLNs prior to lymphadenectomy. Therefore, in the third chapter of this thesis, different contrast agents of varying molecular weight were assessed for their suitability to detect TDLNs. Further, the non-invasive imaging technologies OI and PET/MRI were compared. In contrast to OI with Patent Blue V, which was only suitable for ex vivo identification of the accessory and proper axillary TDLNs, functional in vivo together with ex vivo [18F]FDG-PET/MRI revealed the accessory axillary LN as the main TDLN of a s.c. MC38wt tumor on the right shoulder of mice. Due to the link between immune checkpoint inhibitor therapies and senescence, the final chapter of this thesis addressed the identification of cancer senescence metabolic biomarkers. Several anti-tumor therapies, as well as immune cell-derived cytokines which are released upon immunotherapy induce cancer cell senescence. Thus, potential biomarkers for senescence including GPC and PCr were successfully identified by 1H MRS and must be further studied and evaluated in continuative pre-clinical in vivo studies. In summary, within this thesis a so far unknown mechanism of immune escape by cancer cells was uncovered enabling the identification of anti-PD-1 or anti PD-L1 mAb immune checkpoint inhibitor responsive patients. Further non-invasive in vivo imaging modalities to identify the TDLNs were evaluated. Finally, metabolic studies on cancer senescence revealed several potential biomarkers of a cancer cell growth arrest.