Kelsey Sivick Gauthier, Sarah M. McWhirter, Laura Hix Glickman, Chudi Ndubaku, Charles Y. Cho, John A. Tallarico, Bender Steven Lee, Anne Marie Pferdekamper, Weiwen Deng, Anthony L. Desbien, Lianxing Zheng, David B. Kanne, Leticia Corrales, Thomas W. Dubensky, Ken Metchette, Thomas F. Gajewski, Yan Feng, Justin Leong, Shailaja Kasibhatla, Jeffery M. McKenna, Jie Li, George E. Katibah, and Leonard Sung
Recent reports have provided the mechanistic insight of how innate immune activation promotes priming of anti-tumor immunity and inform the development of clinical approaches to facilitate this process. Spontaneous T cell infiltration of melanoma lesions in humans is correlated with a type I interferon (IFN) transcriptional profile in the tumor microenvironment (TME) and infiltration of lymphocytes, indicative of ongoing innate immune recognition within the tumor. Substantial evidence indicates that tumor infiltrating lymphocytes (TILs), including activated CD8+ T cells, is predictive of a positive clinical outcome in response to several immunotherapy strategies. Similarly, in mice bearing melanoma, there is a correlation between expression of IFN-β by tumor-resident dendritic cells (DCs), and spontaneous priming of tumor-specific immunity. Induction of IFN-β expression and co-regulated IFN-responsive genes and pro-inflammatory chemokines is dependent upon activation of the STING (Stimulator of Interferon Genes) pathway, mediated through sensing of tumor dsDNA in TME-resident CD8α+ DCs by cyclic GMP-AMP (cGAMP) synthase (cGAS), which in turn synthesizes cGAMP. The cyclic dinucleotide (CDN) cGAMP produced by cGAS is the natural STING agonist ligand. Thus, the cGAS-STING signaling axis has emerged as a central node for sensing damage in the host. We hypothesized that direct activation of the STING pathway in the TME by intratumoral (IT) injection of specific CDNs would be an effective therapeutic strategy to promote broad tumor-initiated T cell priming against an individual's tumor antigen repertoire. There are five variant human STING alleles that exist at varying frequencies. While the natural STING ligand cGAMP activates signaling in all variants, other natural CDNs, including those produced by bacteria, have structural differences and are unable to activate particular STING variants, such as the REF (R232H) allele, informing the development of compounds that activate all human STING alleles. We sought to develop synthetic CDN compounds with increased activity in human cells as well as the ability to engage all known polymorphic human STING molecules. Using human 293T cell lines engineered to express the various STING proteins, we screened a large panel of CDN derivatives that varied in purine nucleotide base, structure of the phosphate bridge linkage, and substitution of the non-bridging oxygen atoms at the phosphate bridge with sulfur atoms. ADU-S100 is composed of two adenosine monophosphate (AMP) analogues cyclized via a 2’-5’ (non-canonical) and a 3’-5’ (canonical) phosphodiester bond, and was selected for clinical translation based on properties of enhanced cellular uptake, human STING activation, stability and anti-tumor efficacy, as compared to bacterial and mammalian derived CDNs. Induced cytokine expression from a panel of donor human peripheral blood mononuclear cells (PBMCs) expressing a variety of STING haplotypes, including donors with a homozygous haplotype for the refractory human REF allele, indicates that ADU-S100 activates STING across a diverse human population. Direct engagement of STING through IT administration of ADU-S100 results in effective anti-tumor therapy and long-term survival in various mouse syngeneic tumor models. IT injection of ADU-S100 also generates substantial systemic immune responses capable of rejecting distant metastases and provided long-lived immunologic memory. A bell-shaped ADU-S100 dose response curve (which varied based on tumor model) delineated regression of injected tumor, induction of tumor-specific CD8+ T cell immunity, and regression of distal non-injected tumors (abscopal effect), and/or protection against autologous tumor challenge. At low dose levels, regression of the treated tumor was suboptimal. At optimal doses, regression of the treated and distal untreated tumors, or protection against tumor re-challenge was observed, and correlated with induction of a robust tumor Ag-specific CD8+ T cell response. At higher dose levels there was a loss of protection against tumor re-challenge which correlated with increases in acute systemic cytokines and reduction in CD8+ T cell responses even though growth of the treated tumor was inhibited. These results suggest that the mechanism of ADU-S100-induced tumor regression is due to both an acute pro-inflammatory cytokine response and also tumor-specific CD8+ T cell immunity. The local anti-tumor effect without systemic immunity is consistent with well-established data in which excessive innate immune stimulation and induction of pro-inflammatory cytokines such as TNF-α are known to inhibit both priming of CD8+ T cell immunity and establishment of a stable and self-renewing memory CD8+ T cell population. In addition, anti-tumor efficacy was enhanced by combination with immune checkpoint inhibitors, for example α-PD1, informing future clinical development. By virtue of the ability to elicit innate and T cell-mediated anti-tumor immunity in the TME, these results demonstrate that CDNs have high translational potential for the treatment of patients with advanced/metastatic solid tumors. A Phase 1 clinical study to evaluate the safety and tolerability and possible anti-tumor effects in subjects with cutaneously accessible non UV-induced and UV-induced malignancies or lymphomas given repeated IT doses of ADU-S100 is planned. Citation Format: Laura Hix Glickman, Leticia Corrales, David B. Kanne, Shailaja Kasibhatla, Jie Li, Anne Marie Culazzo Pferdekamper, Kelsey Sivick Gauthier, George E. Katibah, Justin J. Leong, Leonard Sung, Ken Metchette, Weiwen Deng, Anthony L. Desbien, Chudi Ndubaku, Lianxing Zheng, Charles Cho, Yan Feng, Jeffery M. McKenna, John A. Tallarico, Steven L. Bender, Sarah M. McWhirter, Thomas F. Gajewski, Thomas W. Dubensky. Direct activation of STING in the tumor microenvironment leads to potent and systemic tumor regression and immunity. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr SY39-02.