Influence of tumor microenvironment and cancer cells interplay on cutaneous squamous cell carcinoma progression and immunotherapy response

Abstract

[eng] Squamous cell carcinoma of the skin (cSCC) is the second most common type of nonmelanoma skin cancer. Treatment with immunotherapy based on immune checkpoint inhibitors (ICI) was recently approved by the FDA and EMA for this type of tumour. However, a significant percentage of patients do not benefit from this treatment, showing primary or acquired resistance. Dr. Muñoz's group studies the molecular and cellular mechanisms that cause the progression of cSCCs from a well-differentiated profile (tumor cells with epithelial characteristics) to a poorly differentiated profile (tumor cells with mesenchymal characteristics). Previous data from the group in a mouse model of cSCC progression showed that the presence of plastic cells (hybrid tumor cells with intermediate features between epithelial and mesenchymal) caused the progression of these tumors to a poorly differentiated profile. In addition, these poorly differentiated and mesenchymal tumors are strongly infiltrated by immunosuppressive cells (M2 macrophages, MDSCs and regulatory T cells) which caused a state of exhaustion (dysfunctional state) in cytotoxic T cells, causing resistance to ICI treatments in these tumor types. In the present thesis we propose to study the intercommunication between tumor cells and immunological cells present in the tumor microenvironment (TME), which are responsible for promoting the progression of cSCC tumors and the recruitment of immunosuppressive cells. To study the cytokines released by M2 macrophages and MDSCs that could cause the acquisition of the mesenchymal profile by tumor cells, primary cultures of BMDM (bone marrow-derived macrophages) and MDSCs were established from the spleen of mice carrying WD-SCCs and PD-SCCs. We observed that the factors derived from the M-MDSCs subpopulation isolated from mice carrying poorly differentiated tumors had a greater capacity to induce the loss of epithelial characteristics of tumor cells. Subsequently, we identified the cytokines differentially secreted by this subpopulation and obtained a list of five candidate cytokines (BAFF, IL-28, CCL20, CCL21 and IL-7) that could be responsible for the acquisition of mesenchymal characteristics, as has been seen in other tumor types. Future experiments should clarify whether the signaling of these cytokines promotes the acquisition of a mesenchymal profile and are responsible for the progression of cSCCs. We also studied the factors secreted by mesenchymal tumor cells from poorly differentiated tumors that could be responsible for recruiting immunosuppressive cells and establishing a hostile TME for the antitumor function of cytotoxic T cells. First, we identified the cytokines secreted by these mesenchymal tumor cells: CXCL1, CSF3 and CCL2. Subsequently, we blocked its signaling in vivo and characterized the TME profile of the treated tumors. We observed that the separate blockade of the 3 cytokines increased the presence of GranzymaB+ cells, which are cytotoxic T cells and active NK. However, this was not enough to slow tumor growth or increase tumor necrosis. Thus, we hypothesize that blocking the signaling of these cytokines alone is not sufficient to obtain strong antitumor effects, but their combination with ICI could give synergistic antitumor effects. Next, we study alternative strategies to promote the antitumor response of ICI therapies in advanced cSCCs. Thus, we validated previous data from the group that demonstrated that mesenchymal tumor cells do not express the MHC-I protein, which participates in the presentation of antigens to cytotoxic T cells and is essential for these cells to identify and eliminate tumor cells. We observed that mesenchymal tumor cells had reduced transcriptional expression of Mhc-I compared to epithelial tumor cells. In addition, these mesenchymal tumor cells responded correctly to IFN-γ and exhibited increased Mhc-I expression. However, this transcriptional increase was not reflected in an increase in the MHC-I protein in the membrane. We were able to verify that the MHC-I protein was accumulating in perinuclear vesicles and was not being translocated to the cell membrane. In addition, this process was independent of the autophagic pathway unlike what has been described in other tumor types. Thus, we concluded that there was a post-translational dysregulation of MHC-I that prevented this molecule from translocating to the cell surface of mesenchymal tumor cells, making them invisible to the action of cytotoxic T cells. Finally, in an initial study in a small cohort of patients with sensitive cSCC and squamous cell carcinoma of the head and neck (HNSCC) refractory to ICI treatment, we obtained indications that ICI-resistant patients may have a higher content of hybrid and mesenchymal tumor cells and that these tumors would have a higher infiltrate of exhausted CD8 cells. However, these studies were carried out with a very limited number of patient samples. The observations obtained were valid in a larger cohort of cSCC patients, while a larger number of HNSCC patients would need to be analyzed to validate the results obtained.