HDAC6 and DNMT inhibition affect immunogenicity of ovarian cancer cells: A rationale for combining epigenetic and immune therapy in ovarian cancer

Background: Therapies that activate the immune system to fight cancer have shown robust responses in most solid tumors. However, in ovarian cancer, most patients do not respond to these therapies alone. Inhibitors of epigenetic modifying enzymes increase immune signaling from cancer cells. Epigenetic modifiers DNA methyltransferase inhibitors (DNMTi) and selective histone deacetylase inhibitors (HDACi), such as HDAC6i, modulate immune-related pathways involved in anti-tumor immune responses. HDAC6i downregulates immunosuppressive ligands PD-L1 and PD-L2 by dephosphorylating pSTAT3 and upregulates tumor associated antigens (TAA) and antigen presentation machinery. Similarly, DNMTi activate anti-viral signaling via expression of Endogenous Retroviruses (ERVs) to trigger the type I interferon response. Our aim is to test if the combination of epigenetic modulators Nexturastat A (Next A), a selective HDAC6i, and 5-azacytidine (AZA), a DNMTi, can be safely used to increase an immune response in ovarian cancer. We hypothesize that these drugs will enhance tumor immunity alone and when combined with immune checkpoint blockades targeting PD-1. Results: HDAC enzymes are differentially expressed in A2780, HEY, Kuramochi, SKOV3, and TykNu human ovarian cancer cell lines. HDAC6 was expressed at lower levels in HEY and TykNu but at higher levels in SKOV3 and A2780. As previously reported, we believe this is due to the overexpression of the chromatin modifier ARID1A in SKOV3 and A2780. Upregulation of HDAC6 also correlated with a higher IC50 for Next A treatment in those cell lines. Immunoblots showed that PD-L1 protein, a marker of poor prognosis in ovarian cancer, decreased after treatment with Next A and even more in combination with AZA. qPCR demonstrates that combination therapy induce an additive type 1 interferon response and alters the expression of markers that modulate the tumor-immune reaction. DNMT1, the known target of AZA, was decreased after treatment with AZA and NextA, independently and in combination, a finding that has not been previously reported. Conclusion: As shown previously, HDAC6 enzyme levels are higher in cell lines with ARID1A mutations. DNMT1 decreased after treatment with AZA, and surprisingly also after treatment with Next A. PD-L1 decreased after treatment with Next A and even more so when adding AZA. Thus, combining these epigenetic modifiers could lead to an additive effect on immune signaling through stimulation of antiviral signaling (DNMTi), which upregulates the immunosuppressive ligand PD-L1, which is then reduced by HDAC6 inhibition. We are testing this combination with anti-PD-1 in an immunocompetent mouse model of ovarian cancer