The Combination of PARP Inhibitors and DNMT Inhibitors Modulates Immune Activity and Suggests a Role for Immune Therapy in AML

Acute myeloid leukemia (AML) is a genetically heterogeneous disease where malignant clonal proliferation of immature myeloid cells in the bone marrow leads to disruption of normal hematopoiesis and bone marrow failure. Most treatment improvements have consisted of limited chemotherapeutic interventions, and morbidity and mortality remain unacceptably high. While patients unfit for intensive chemotherapy have the options of epigenetic treatments with DNA methyltransferase inhibitors (DNMTis), and the preclinical research underpinning these treatments is robust and promising, clinical response rates have been poor. Therefore, we have derived a unique treatment strategy combining poly-ADP ribose polymerase inhibitors (PARPis) with DNMTis, and we have previously demonstrated efficacy of this combination in AML in both in vitroand in vivomodels. This combination increases the amount and retention time of PARP1 at laser-induced DNA damage sites, increases DNA DSBs and tumor cytotoxicity, blunts cell self-renewal, and leads to strong anti-tumor responses. These results form the basis for our ongoing Phase I/II clinical trial of this combination therapy in patients with relapsed/refractory AML.

To identify molecular pathways underpinning this combination, genome-wide expression microarray analysis was performed on RNA extracted from multiple BRCA-intact sporadic AML cell lines treated with single agents or drug combination. Interestingly, combination drug treatment, more than either single treatment alone, up regulated genes in immune response pathways, including interferons, natural killer cell-mediated cytotoxicity and Toll-like receptor signaling pathway genes. In both AML cell lines (N=10) and primary samples (N=8), we used qPCR to validate these expression increases, with particular focus on interferon type 1-inducible genes, including IFI6, IFI27, IFI30, IRF7, ISG15, and OASL. These interferon-inducible genes were similarly increased in selected AML clinical trial samples post combination treatment. Interferon stimulation studies showed that treatment of multiple cell lines with IFN-β similarly up regulates our genes of interest, and suggests that the immune system plays a pivotal role in mediating the effects of our combination treatment.

To functionally validate this immune involvement in vivo, we use an immune competent mouse model that spontaneously develops AML. The bone marrow cells collected from leukemic FLT3/ITD;NHD13 mice are transplanted into sublethally irradiated syngeneic recipients where they undergo clonal leukemic expansion and expression of specific surface markers (CD45.1 vs CD45.2) allow for the tracking of both host and transplanted blood cells. After validating that our combination treatment increases survival and reduces tumor burden in this in vivo model, we isolated mononuclear cells from spleens and livers and found that mice in the combination therapy group had increased T cell populations with increased PD1, CD28 and 4-1BB expression. Tal single agent and combination therapy treated mice also had decreased myeloid-derived suppressor cells and increased antigen presenting cells, specifically dendritic cells, monocytes and macrophages. Finally, upon ex vivo stimulation, T cells from mice receiving combination therapy produced significantly more IFN-γ and GM-CSF than mice from other treatment groups.

These results indicate that our combination therapy results in immune activation, and suggest the potential to target the immune system in order to induce anti-tumor effects, enhance efficacy of our combination treatment and improve clinical outcomes in AML.