ROLE OF PD-1/PD-L1 AXIS IN REGULATING FUNCTIONS OF INNATE IMMUNE CELLS IN THE TREATMENT OF CANCER

Background. Programmed cell death-1 (PD-1), a member of the CD28 receptor family, is reported to be expressed by activated T lymphocytes and upon the interaction with its ligands (PD-L1 and -2) determines T cells exhaustion. PD-1/PD-Ls axis is crucial in controlling T cell activity in order to maintain peripheral tolerance and to prevent autoimmunity. For this reason, tumor cells, up-regulating PD-L1, exploit this pathway to escape immune elimination. Although deeply investigated in T lymphocytes, PD-1 can be also expressed by innate immune cells (NK cells, monocytes, macrophages, dendritic cells) where its function seems to be similar to that described for T cells. In recent years, the discovery of immune checkpoint inhibitors (ICIs), antibodies directed against PD-1 or PD-L1, revolutionized the paradigm of tumors treatment. Clinical studies demonstrated the efficacy of PD-1 blockade in different types of cancer, even in those considered non-responsive to adaptive immunotherapy, suggesting that also innate immune system can be involved in the final anti-tumor activity of PD-1 blockade. Despite the enthusiasm of the beginning, only a fraction of patients really benefits from this treatment, while all the others remain unresponsive. Therefore, efforts are required to improve the efficacy of ICIs treatment. Rationale. Since innate immune cells may influence the efficacy of PD-1 blockade therapy, the therapeutic stimulation of the innate immune system might improve PD-1 antibody efficacy. Among innate cells immunostimulants, Toll-like receptors (TLRs) agonists are the best known for their ability to strongly activate innate immune cells and for their anti-tumor activity. Aim. The present thesis is aimed to evaluate the role of innate immune system in the context of PD-1 blockade therapy and to explore novel combination therapies. Results. Immunodeficient mice xenografted with human ovarian cancer cells IGROV-1 were treated with CpG-ODN, TLR9 agonist, and anti-PD-1 antibody alone or in combination. Surprisingly, we observed a drastic reduction in CpG-ODN antitumor activity when combined with anti-PD-1 blocking antibody. Subsequent experiments revealed that anti-PD-1 antibody alone produces an acceleration of tumor growth paralleled by an increase of ArgI+ myeloid cells in the tumor microenvironment. These results suggested that myeloid cells can play a role in determining the deleterious effect of anti-PD-1 antibody on CpG-ODN activity. This idea was also supported by data showing that in vivo depletion of this innate immune cell population abrogated the negative effect of anti-PD-1 antibody. Similar experiments were performed in another mouse xenograft model. Animals were injected with NCI-H460 human non-small cell lung cancer (NSCLC) cell line and treated as above. Again, we found that PD-1 blockade determined an acceleration of tumor growth together with an increase in intratumoral macrophages and ArgI+ cells. Since blocking PD-1 in macrophages is reported to promote their anti-tumoricidal effector functions, it is possible to speculate that the detrimental effect of anti-PD-1 antibody may be not due to such blockade but instead to an interaction of the Fc portion of the antibody with macrophages/myeloid cells. Indeed, when H460 xenograft model was treated with anti-PD-1 F(ab)2, lacking the Fc domain, no increase in tumors growth was detected. Several publications reported that there is a fraction of patients experiencing hyperprogression, a particular aggressive and unpredictable spread of the disease following anti-PD-1 administration. This condition may resemble our preclinical models. By immunohistochemistry, we analyzed tumor biopsies of NSCLC patients treated with ICI. We found a peculiar cluster of CD163+ CD33+ PD-L1+ macrophages with epithelioid morphology in all patients that experienced hyperprogression, highlighting the role of myeloid cells in inducing the detrimental effect of anti-PD-1 antibody. Conclusion. Our findings, describing myeloid cells as crucial player in causing tumor acceleration after ICI therapy, may allow the identification of patients not to be treated with this type of immunotherapy.