Search

Supplemental Figure 2: STING agonists stimulate cytokines in tumor, plasma and peripheral tissues following IT dosing.

Supplementary Methods and Figures

Targeting the programmed death 1/programmed death ligand 1 (PD-1/PD-L1) pathway with immunotherapy has revolutionized the treatment of many cancers. Somatic tumor mutational burden (TMB) and T-cell–inflamed gene expression profile (GEP) are clinically validated pan-tumor genomic biomarkers that can predict responsiveness to anti-PD-1/PD-L1 monotherapy in many tumor types. We analyzed the association between these biomarkers and the efficacy of PD-1 inhibitor in 11 commonly used preclinical syngeneic tumor mouse models using murinized rat anti-mouse PD-1 DX400 antibody muDX400, a surrogate for pembrolizumab. Response to muDX400 treatment was broadly classified into three categories: highly responsive, partially responsive, and intrinsically resistant to therapy. Molecular and cellular profiling validated differences in immune cell infiltration and activation in the tumor microenvironment of muDX400-responsive tumors. Baseline and on-treatment genomic analysis showed an association between TMB, murine T-cell–inflamed gene expression profile (murine-GEP), and response to muDX400 treatment. We extended our analysis to investigate a canonical set of cancer and immune biology-related gene signatures, including signatures of angiogenesis, myeloid-derived suppressor cells, and stromal/epithelial-to-mesenchymal transition/TGFβ biology previously shown to be inversely associated with the clinical efficacy of immune checkpoint blockade. Finally, we evaluated the association between murine-GEP and preclinical efficacy with standard-of-care chemotherapy or antiangiogenic agents that previously demonstrated promising clinical activity, in combination with muDX400. Our profiling studies begin to elucidate the underlying biological mechanisms of response and resistance to PD-1/PD-L1 blockade represented by these models, thereby providing insight into which models are most appropriate for the evaluation of orthogonal combination strategies.

Supplementary Data from Reverse Translating Molecular Determinants of Anti–Programmed Death 1 Immunotherapy Response in Mouse Syngeneic Tumor Models

Supplemental Table 1: Log2Ratio and p values for all 310 genes shown in Figure 6

Targeting the programmed death 1/programmed death ligand 1 (PD-1/PD-L1) pathway with immunotherapy has revolutionized the treatment of many cancers. Somatic tumor mutational burden (TMB) and T-cell–inflamed gene expression profile (GEP) are clinically validated pan-tumor genomic biomarkers that can predict responsiveness to anti-PD-1/PD-L1 monotherapy in many tumor types. We analyzed the association between these biomarkers and the efficacy of PD-1 inhibitor in 11 commonly used preclinical syngeneic tumor mouse models using murinized rat anti-mouse PD-1 DX400 antibody muDX400, a surrogate for pembrolizumab. Response to muDX400 treatment was broadly classified into three categories: highly responsive, partially responsive, and intrinsically resistant to therapy. Molecular and cellular profiling validated differences in immune cell infiltration and activation in the tumor microenvironment of muDX400-responsive tumors. Baseline and on-treatment genomic analysis showed an association between TMB, murine T-cell–inflamed gene expression profile (murine-GEP), and response to muDX400 treatment. We extended our analysis to investigate a canonical set of cancer and immune biology-related gene signatures, including signatures of angiogenesis, myeloid-derived suppressor cells, and stromal/epithelial-to-mesenchymal transition/TGFβ biology previously shown to be inversely associated with the clinical efficacy of immune checkpoint blockade. Finally, we evaluated the association between murine-GEP and preclinical efficacy with standard-of-care chemotherapy or antiangiogenic agents that previously demonstrated promising clinical activity, in combination with muDX400. Our profiling studies begin to elucidate the underlying biological mechanisms of response and resistance to PD-1/PD-L1 blockade represented by these models, thereby providing insight into which models are most appropriate for the evaluation of orthogonal combination strategies.

Bromodomain and extraterminal domain (BET) proteins help direct the differentiation of helper T cell subsets, but their role in activated T cell function has not been described in detail. In this study, we investigate various consequences of epigenetic perturbation in human T lymphocytes using MK-8628, a potent and highly selective inhibitor of BET proteins. MK-8628 reduces the expression of canonical transcripts directing the proliferation, activation, and effector function of T lymphocytes. Treatment with MK-8628 abolishes the expression of key cyclins required for cell cycle progression and induces G1 cell cycle arrest in TCR-activated lymphocytes. This antiproliferative phenotype partially results from T lymphocyte apoptosis, which is exacerbated by MK-8628. In naive and memory T cell subsets, MK-8628 antagonizes T cell activation and suppresses polyfunctional cytokine production. Collectively, our results describe potent immunosuppressive effects of BET inhibition on human T cell biology. These results have important implications for immune modulatory targeting of BET proteins in the settings of T cell–driven autoimmune inflammation.

The innate immune agonist STING (STimulator of INterferon Genes) binds its natural ligand 2′3′-cGAMP (cyclic guanosine-adenosine monophosphate) and initiates type I IFN production. This promotes systemic antigen-specific CD8+ T-cell priming that eventually provides potent antitumor activity. To exploit this mechanism, we synthesized a novel STING agonist, MSA-1, that activates both mouse and human STING with higher in vitro potency than cGAMP. Following intratumoral administration of MSA-1 to a panel of syngeneic mouse tumors on immune-competent mice, cytokine upregulation and its exposure were detected in plasma, other tissues, injected tumors, and noninjected tumors. This was accompanied by effective antitumor activity. Mechanistic studies in immune-deficient mice suggested that antitumor activity of intratumorally dosed STING agonists is in part due to necrosis and/or innate immune responses such as TNF-α activity, but development of a robust adaptive antitumor immunity is necessary for complete tumor elimination. Combination with PD-1 blockade in anti–PD-1–resistant murine models showed that MSA-1 may synergize with checkpoint inhibitors but can also provide superior tumor control as a single agent. We show for the first time that potent cyclic dinucleotides can promote a rapid and stronger induction of the same genes eventually regulated by PD-1 blockade. This may have contributed to the relatively early tumor control observed with MSA-1. Taken together, these data strongly support the development of STING agonists as therapy for patients with aggressive tumors that are partially responsive or nonresponsive to single-agent anti–PD-1 treatment by enhancing the anti–PD-1 immune profile.

PD-L1 expression has been evaluated as a predictive biomarker for immunotherapy in numerous tumor types. However, very limited data are available in pediatric brain tumors. The aim of this study was to characterize PD-1 and PD-L1 expressions of four pediatric malignant brain tumors and gene expression profile. This study included 89 pediatric patients receiving standard treatment at Seoul National University Children’s Hospital and Seoul National University Bundang Hospital between 1990 and 2014: atypical teratoid/rhabdoid tumor (AT/RT) 20; ependymoma (EPN) 20; high grade glioma (HGG) 21; and medulloblastoma (MBL) 28. We performed immunohistochemistry assays for PD-1 and PD-L1. To characterize the gene expression, a custom immune-response focused gene panel was used. PD-1 expression was positive in 7 (35%) AT/RT, 7 (35%) EPN, 4 (19%) HGG, and 3 (11%) MBL patients. PD-L1 expression was positive in 8 (40%) AT/RT, 4 (20%) EPN, and 4 (19%) HGG; negative in all MBL patients. There was no statistically significant difference in the overall survival of PD-L1 positive patients. The gene expression analysis demonstrated differences in two clustering functional categories: cell–cell signaling and antigen presentation pathway. AT/RT, EPN, and HGG showed a relatively higher expression rate of PD-L1 (19–40%). This suggests these tumor types might be good candidates for PD-1 checkpoint blockade. We determined that gene expression may potentially serve as a molecular tool in predicting which patients will respond to immunotherapy. Further investigation is required to better understand the predictive and prognostic role of PD-L1 in pediatric brain tumors.

LAG3 is an important regulator of T cell homeostasis and studies in mouse tumor models have demonstrated that simultaneously antagonizing LAG3 and PD1 can augment tumor-specific T cell responses and induce tumor rejection. The combined use of LAG3 antagonist antibodies with established anti-PD1 therapies is currently being evaluated in human clinical trials. A functional assay for human LAG3 was developed by co-culture of a Jurkat T-cell lymphoma line overexpressing LAG3 with a Raji B-cell lymphoma line in the presence of staphylococcal enterotoxins. Reversal of LAG3 repression was measured as an increase in IL-2 production or NFAT activation in response to treatment with MK-4280, an anti-human LAG3 antagonist antibody. Changes in cytokines, chemokines, and other mRNA transcripts were in agreement with published in vitro and in vivo models for LAG3 biology which highlights the physiological relevance of the Jurkat functional assay. Additional engineering of PD1 and PDL1 components into the LAG3 assay resulted in a bi-functional assay that is capable of inducing a 10-fold response to individual antibodies blocking either PD1 or LAG3. Importantly, when MK-4280 and pembrolizumab were combined to block both pathways, a synergistic 50-fold increase in response was observed.

Blockade of the checkpoint inhibitor programmed death 1 (PD1) has demonstrated remarkable success in the clinic for the treatment of cancer; however, a majority of tumors are resistant to anti-PD1 monotherapy. Numerous ongoing clinical combination therapy studies will likely reveal additional therapeutics that complement anti-PD1 blockade. Recent studies found that immunogenic cell death (ICD) improves T cell responses against different tumors, thus indicating that ICD may further augment antitumor immunity elicited by anti-PD1. Here, we observed antitumor activity following combinatorial therapy with anti-PD1 Ab and the cyclin-dependent kinase inhibitor dinaciclib in immunocompetent mouse tumor models. Dinaciclib induced a type I IFN gene signature within the tumor, leading us to hypothesize that dinaciclib potentiates the effects of anti-PD1 by eliciting ICD. Indeed, tumor cells treated with dinaciclib showed the hallmarks of ICD including surface calreticulin expression and release of high mobility group box 1 (HMGB1) and ATP. Mice treated with both anti-PD1 and dinaciclib showed increased T cell infiltration and DC activation within the tumor, indicating that this combination improves the overall quality of the immune response generated. These findings identify a potential mechanism for the observed benefit of combining dinaciclib and anti-PD1, in which dinaciclib induces ICD, thereby converting the tumor cell into an endogenous vaccine and boosting the effects of anti-PD1.

Meeting abstracts Pembrolizumab (MK-3475), a humanized monoclonal IgG4 antibody against programmed death receptor 1 (PD-1), is currently being studied in clinical trials across more than 30 types of cancers. Immunotherapy with anti–PD-1 monoclonal antibodies such as pembrolizumab shows robust

Improving clinical outcomes using immunotherapeutics in cancer treatments will depend on a variety of factors including the composition and phenotype of effector T cell in tumors, the inhibitory factors from myeloid, tumor and stromal compartments of the tumor and selection of efficacious combination of therapeutic immunomodulatory agents that either neutralize immune checkpoint inhibitors or activate T cell costimulatory pathways. To address the issue of selecting optimal combinations of immune modulating agents we developed an in-vitro culture system for allo-stimulating and expanding tumor infiltrating lymphocytes (TILS) from tumor digests with an MHC Class II alloantigen expressing cell line. We selected allo-specific CD4 and CD8 clones from expanded TILS with an exhausted phenotype confirmed by cell surface expression of PD-1, LAG3 and TIGIT. T cell clones were incubated with the alloantigen expressing cell lines transduced with different combinations of immodulatory ligands and compared the effect of single agent versus combinations of checkpoint inhibitory monoclonal antibodies on T cell proliferation and cytokine production. Taqman and flow cytometry analysis was performed on several clones following allo-stimulation. A decrease in TIL clonality with repeated allo-stimulation was confirmed by TcR beta chain sequencing. In-vitro IFN gamma and Granzyme B production was enhanced with anti-PD-1 antibodies in combination with either anti-LAG3 or anti-TIGIT antibodies. This in-vitro system has the potential of screening combinations of immodulatory antibodies in a variety of TILS from different tumor types and can improve our understanding of new checkpoint inhibitor combinations. Citation Format: Robert L. Shields, Sabine Le Saux, Sophie Li, Manjiri Sathe, Terri McClanahan, Rene De Waal Malefyt. Tumor derived T cell clones for evaluation of check point inhibitor therapeutics [abstract]. In: Proceedings of the AACR Special Conference on Tumor Immunology and Immunotherapy; 2017 Oct 1-4; Boston, MA. Philadelphia (PA): AACR; Cancer Immunol Res 2018;6(9 Suppl):Abstract nr A33.

Intravascular delivery (1.5 x 10(9) particles and higher) of recombinant adenovirus (rAd) induces myeloid cell mediated, self-limiting hemodynamic responses in normal mice. However, we observed anaphylactoid-type reactions and exacerbated hemodynamic events following rAd injection in mice bearing malignant 4T1 mammary carcinoma. Because 4T1 tumors induce significant CD11b(+)Gr-1(+) myeloid cell expansion and activation, we set to determine whether this causes rAd-induced exaggerated responses. When treated with a single intravenous dose (1 x 10(10) particles) of rAd, mice implanted with 4T1 carcinoma succumbed due to the anaphylactoid-type reactions. In contrast, normal mice and mice implanted with a related mammary carcinoma (66cl4) that does not induce CD11b(+)Gr-1(+) cell expansion, showed minimal responses. Depletion of phagocytic CD11b(+)Gr-1(+) cells prior to rAd delivery protected 4T1 tumor-bearing animals, whereas passive transfer of CD11b(+)Gr-1(+) cells from 4T1 tumor-bearing animals was sufficient to convey susceptibility to anaphylactoid-type reactions in normal animals. We further show that there is upregulation of nitric oxide and leukotriene signaling pathways in the 4T1 tumor-induced CD11b(+)Gr-1(+) myeloid cells and that pretreating mice with inhibitors of nitric oxide synthetase and leukotrienes can attenuate the anaphylactoid-type reactions. These data show that malignant tumor growth can alter CD11b(+)Gr-1(+) myeloid cells, rendering hosts susceptible to anaphylactoid-type reactions upon intravascular treatment with rAd.

STAT5A and STAT5B proteins belong to the family of signal transducers and activators of transcription. They are encoded by two separate genes with 91% identity in their amino acid sequences. Despite their high degree of conservation, STAT5A and STAT5B exert non-redundant functions, resulting at least in part from differences in target gene activation. To better characterize the differential contribution of STAT5A and STAT5B in gene regulation, we performed single or double knockdown of STAT5A and STAT5B using small interfering RNA. Subsequent gene expression profiling and RT-qPCR analyses of IL-3-stimulated Ba/F3-beta cells led to the identification of putative novel STAT5 target genes. Chromatin immunoprecipitation assays analyzing the corresponding gene loci identified unusual STAT5 binding sites compared to conventional STAT5 responsive elements. Some of the STAT5 targets identified are upregulated in several human cancers, suggesting that they might represent potential oncogenes in STAT5-associated malignancies.

MK-4280 is a humanized IgG4 monoclonal antibody (mAb) that binds to the immune checkpoint receptor Lymphocyte Activation Gene-3 (LAG-3) to block the interaction with its ligand, Major Histocompatibility Complex (MHC) Class II. LAG-3 is frequently co-expressed with other immune checkpoint receptors, most notably programmed cell death protein-1 (PD-1), on T cells with an exhausted phenotype. LAG-3 and PD-1 cooperate to regulate peripheral immune tolerance in healthy individuals, and, conversely, play critical roles in several diseases, including autoimmunity, graft rejection, viral infections, and cancer. Co-blockade of LAG-3 and PD-1 in immunocompetent mouse tumor models have demonstrated augmented anti-tumor activity over single agents. However, the molecular mechanisms behind these combination effects have not been fully investigated. Here, preclinical proof-of-biology studies are presented for co-targeting LAG-3 and PD-1 in cancer. c28G10-mG1-[D265A] (abbreviated 28G10-mG1) is a rat:mouse chimera that mimics MK-4280 by its ability to directly block the mouse LAG-3:MHC Class II interaction without initiating Fc-mediated effector functions. As a single agent, 28G10-mG1 demonstrated modest anti-tumor activity across several syngeneic mouse tumor models, despite evidence of systemic drug exposure and target engagement (as assessed by sLAG-3-mAb complex accumulation). The combination of 28G10-mG1 and the anti-mouse PD-1 blocking antibody mDX400 resulted in greater tumor growth inhibition and increased numbers of complete responses (CR) over mDX400 alone in the MBT-2 tumor model. Furthermore, animals that had achieved CR to combination therapy were subsequently protected from MBT-2 re-challenge, suggesting the establishment of immune memory. RT-qPCR analyses revealed up-regulation of immune-related genes, primarily at Day 4 in the blood and Day 8 in the tumor with mDX400, but not 28G10-mG1, treatment. However, when combined with mDX400, 28G10-mG1 further altered the expression of many immune-related genes that were perturbed by mDX400 single agent therapy. Genes unique to combination treatment were also observed. Significantly, immune pathway signatures associated with clinical efficacy to Keytruda were upregulated with combination therapy. Tumor transcriptome and network analysis by RNAseq revealed enrichment in several immune- and cytokine-related pathways with combination therapy compared to mDX400 single agent therapy. Taken together, these preclinical oncology studies support the concept of co-targeting LAG-3 to increase the therapeutic efficacy of PD-1 blockade. Clinical investigation of MK-4280 in combination with anti-PD-1 therapy (pembrolizumab/Keytruda) is ongoing. Citation Format: Brian B. Haines, Sarah Javaid, Long Cui, Heather Hirsch, Saso Cemerski, Terri McClanahan, Manjiri Sathe, Shuli Zhang, Michael Rosenzweig, Brian Long, Rene de Waal Malefyt. Blockade of LAG-3 amplifies immune activation signatures and augments curative antitumor responses to anti-PD-1 therapy in immune competent mouse models of cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 4714. doi:10.1158/1538-7445.AM2017-4714

Monocytes recruited from the blood are key contributors to the nature of an immune response. While monocyte recruitment in a subset of immunopathologies has been well studied and largely attributed to the chemokine monocyte chemoattractant protein (MCP)-1, mechanisms mediating such recruitment to other sites of inflammation remain elusive. Here, we showed that localized inflammation resulted in an increased binding of monocytes to perifollicular high endothelial venules (HEVs) of lymph nodes draining a local inflammatory site. Quantitative PCR analyses revealed the upregulation of many chemokines in the inflamed lymph node, including MCP-1 and MIG. HEVs did not express detectable levels of MCP-1; however, a subset of HEVs in inflamed lymph nodes in wild-type (but not tumor necrosis factor [TNF] null mice) expressed MIG and this subset of HEVs preferentially supported monocyte binding. Expression of CXCR3, the receptor for MIG, was detected on a small subset of peripheral blood monocytes and on a significant percentage of recruited monocytes. Most importantly, in both ex vivo and in vivo assays, neutralizing anti-MIG antibodies blocked monocyte binding to inflamed lymph node HEVs. Together, these results suggest that the lymph node microenvironment can dictate the nature of molecules expressed on HEV subsets in a TNF-dependent fashion and that inflammation-induced MIG expression by HEVs can mediate monocyte recruitment.

Chronic inflammation has been associated with increased risk for developing gastrointestinal cancer. Interleukin-23 (IL-23) receptor signaling has been correlated with inflammatory bowel disease pathogenesis, as well as promotion of tumor growth. However, little is known about the relative potential for IL-23-directed causality in gut tumorigenesis. We report that IL-23 transgene expression was sufficient to induce rapid (3-4 weeks) de novo development of intestinal adenomas with 100% incidence. Initiation of tumorigenesis was independent of exogenous carcinogens, Helicobacter colonization, or pre-existing tumor-suppressor gene mutations. Tumorigenesis was mediated by Thy1(+)IL-23R(+) innate lymphoid cells (ILC3), in part, through IL-17 responses as tumor development was inhibited in RAG(-/-) × IL-17(-/-) double knockout mice. Remarkably, IL-23 initiation of tumorigenesis by resident ILCs consistently occurred before recruitment of conspicuous inflammatory infiltrates. Our results reveal an explicit role for IL-23-mediated initiation of gut tumorigenesis and implicate a key role for IL-23R(+) ILC3 in the absence of overt cellular infiltrate recruitment.

The spondyloarthropathies are a group of rheumatic diseases that are associated with inflammation at anatomically distal sites, particularly the tendon-bone attachments (entheses) and the aortic root. Serum concentrations of interleukin-23 (IL-23) are elevated and polymorphisms in the IL-23 receptor are associated with ankyosing spondylitis, however, it remains unclear whether IL-23 acts locally at the enthesis or distally on circulating cell populations. We show here that IL-23 is essential in enthesitis and acts on previously unidentified IL-23 receptor (IL-23R)(+), RAR-related orphan receptor γt (ROR-γt)(+)CD3(+)CD4(-)CD8(-), stem cell antigen 1 (Sca1)(+) entheseal resident T cells. These cells allow entheses to respond to IL-23 in vitro-in the absence of further cellular recruitment--and to elaborate inflammatory mediators including IL-6, IL-17, IL-22 and chemokine (C-X-C motif) ligand 1 (CXCL1). Notably, the in vivo expression of IL-23 is sufficient to phenocopy the human disease, with the specific and characteristic development of enthesitis and entheseal new bone formation in the initial complete absence of synovitis. As in the human condition, inflammation also develops in vivo at the aortic root and valve, which are structurally similar to entheses. The presence of these entheseal resident cells and their production of IL-22, which activates signal transducer and activator of transcription 3 (STAT3)-dependent osteoblast-mediated bone remodeling, explains why dysregulation of IL-23 results in inflammation at this precise anatomical site.

Paired immunoglobulin-like receptors beta, PILRβ, and alpha, PILRα, are related to the Siglec family of receptors and are expressed primarily on cells of the myeloid lineage. PILRβ is a DAP12 binding partner expressed on both human and mouse myeloid cells. The potential ligand, CD99, is found on many cell types, such as epithelial cells where it plays a role in migration of immune cells to sites of inflammation. Pilrb deficient mice were challenged with the parasite Toxoplasma gondii in two different models of infection induced inflammation; one involving the establishment of chronic encephalitis and a second mimicking inflammatory bowel disease in order to understand the potential role of this receptor in persistent inflammatory responses. It was found that in the absence of activating signals from PILRβ, antigen-presenting cells (APCs) produced increased amounts of IL-27, p28 and promoted IL-10 production in effector T cells. The sustained production of IL-27 led ultimately to enhanced survival after challenge due to dampened immune pathology in the gut. Similar protection was also observed in the CNS during chronic T. gondii infection after i.p. challenge again providing evidence that PILRβ is important for regulating aberrant inflammatory responses.

Paired immunoglobulin-like type 2 receptors (PILRs) inhibitory PILRα and activating PILRβ are predominantly expressed on myeloid cells. Their functions in host defense and inflammation are largely unknown, and in this study, we evaluated their roles in an acuteStaphylococcus aureuspneumonia model. Compared to their respective controls,Pilrb−/−mice or mice in which PILRα was activated with an agonistic antibody showed improved clearance of pulmonary staphylococci and improved survival. These mice had reduced serum or bronchoalveolar lavage fluid levels of interleukin-1β (IL-1β), tumor necrosis factor alpha (TNF-α), and IL-6 and elevated levels of gamma interferon (IFN-γ), IL-12, and IL-10. In contrast, mice in which PILRβ was activated had increased lung bacterial burdens and higher mortality coupled with an intense proinflammatory response with highly elevated levels of IL-1β, TNF-α, and IL-6. Treatment groups with reduced bacterial burdens had higher levels of Keratinocyte-derived chemokine (KC), macrophage inflammatory protein 2 (MIP-2), and MIP-1α in bronchoalveolar lavage fluid and an increased influx of neutrophils and macrophages to the lungs. Consistent with ourin vivofindings, bone marrow-derived macrophages fromPilrb−/−mice released significantly less IL-1β and TNF-α and more IFN-γ and IL-12 than did the wild-type macrophages when directly stimulated with heat-killedS. aureus. To our knowledge, this is the first evidence thatS. aureusdirectly interacts with PILRβ. It provides a mechanism by which manipulating the balance in favor of an inhibitory PILR signal, by activation of PILRα or deletion of PILRβ, helps to control acuteS. aureus-mediated pneumonia and attenuate the inflammatory response. These results highlight the importance of PILRs in innate immunity and the control of inflammation.

IL-27 is secreted by APCs in response to inflammatory stimuli and exerts a proinflammatory Th1-enhancing activity but also has significant anti-inflammatory functions. We examined the molecular mechanism by which IL-27 regulates TGFβ plus IL-6- or IL-23-dependent Th17 development in the mouse and human systems. IL-27 inhibited the production of IL-17A and IL-17F in naive T cells by suppressing, in a STAT1-dependent manner, the expression of the Th17-specific transcription factor RORγt. The in vivo significance of the role of IL-27 was addressed in delayed-type hypersensitivity response and experimental autoimmune encephalomyelitis (EAE). By generating mice deficient for the p28 subunit of IL-27, we showed that IL-27 regulated the severity of delayed-type hypersensitivity response and EAE through its effects on Th17 cells. Furthermore, up-regulation of IL-10 in the CNS, which usually occurs late after EAE onset and plays a role in the resolution of the disease, was notably absent in IL-27p28−/− mice. These results show that IL-27 acts as a negative regulator of the developing IL-17A response in vivo, suggesting a potential therapeutic role for IL-27 in autoimmune diseases.

Although there are a variety of immune cells localized in the tumor microenvironment, they are generally present in a quiescent state. The goal of cancer immunotherapy is to activate immune effector cells in the tumor microenvironment to recognize tumor cells and elicit an immune response that results in elimination of the tumor. Checkpoint inhibitors and activators that target T-cell receptors provide attractive targets for cancer immunotherapies by either enhancing T-cell activation pathways or inhibiting T-cell checkpoint/anergy pathways. In 2014, the anti-programmed death receptor 1 (PD-1) antibody pembrolizumab (MK-3475) was approved for use in patients with unresectable or metastatic melanoma and disease progression following ipilimumab and, if BRAFV600 mutation positive, a BRAF inhibitor. Pembrolizumab is currently being explored in multiple clinical trials covering ∼30 different cancer types. A more complete understanding of the mechanism of action and biology associated with both response and resistance to pembrolizumab is critical to better inform future clinical development and to provide insight into the development of additional immuno-oncology therapies. Preclinical mouse syngeneic tumor models have been used extensively to support the clinical development of drugs such as pembrolizumab and to help identify novel targets that have the potential to synergize with anti-PD-1 therapies. We employed a panel of mouse syngeneic tumor models that have been used to assess the efficacy of muDX400, a fully murinized surrogate antibody of pembrolizumab. Tumors from these models were extensively characterized at the molecular and cellular levels at baseline and after muDX400 treatment. Data obtained from multiple experimental sources, including RNA expression, DNA mutation and copy number, fluorescence-activated cell sorting, and immunohistochemistry, have been integrated to inform pembrolizumab mechanisms of action and resistance, pharmacodynamic biomarkers, responder identification, and indication selection. These findings are also being compared to data being obtained from ongoing clinical trials of pembrolizumab to help better understand whether these preclinical models can be translated to the clinic. These preclinical syngeneic tumor models also provide the opportunity to formulate and test specific hypotheses to more completely understand the biology behind the clinical successes currently observed with novel cancer immunotherapies, including anti-PD-1 antibodies such as pembrolizumab. The hypotheses under evaluation include mutational burden, immune cell activation and migration, interferon signaling, antigen presentation (MHC class I and II), and expression of novel targets that will help lead to the development of the next generation of cancer immunotherapies. Citation Format: Heather A. Hirsch, Elaine M. Pinheiro, Mingmei Cai, Yanhong Ma, Manjiri Sathe, Mark Ayers, Terrill K. McClanahan. Molecular characterization of mouse syngeneic tumor models in response to treatment with anti-PD-1 immunotherapy. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1328. doi:10.1158/1538-7445.AM2015-1328

To investigate the potential molecular mediators of tissue-specific recruitment, we explored the influence of different cytokine challenges on gene expression regulation in five primary endothelial cells (ECs), representing two different phenotypes: iliac artery and aortic (macrovascular); lung, colon and dermal (microvascular). We challenged ECs with cytokines that elicit different patterns of inflammatory and immune responses in immune cells: tumor necrosis factor (TNF-alpha), interferon-gamma (IFN-gamma) or interleukin-4 (IL-4), and used microarrays containing approximately 40,000 unique cDNAs, to assess changes in differential gene expression relative to untreated cells. Five hundred and sixty three sequences changed by at least 2.5 fold in one or more of the 15 possible EC /cytokine combinations. The list included highly regulated adhesion molecules, chemokines, cytokines, metalloproteases, and IFN-gamma-induced genes. Overall, IFN-gamma caused the largest number of gene expression changes and its profile was least correlated with IL-4. In addition to clusters that were predominantly EC/cytokine specific, we also observed several clusters that were regulated by more than one cytokine across several ECs. Furthermore, we identified genes that were reciprocally expressed in response to different cytokines that could serve as markers of inflammatory and immune expression. These results confirm the importance of microenvironment in primary ECs that could have important applications in developing targeted therapies for vascular diseases.

Uncontrolled mucosal immunity in the gastrointestinal tract of humans results in chronic inflammatory bowel disease (IBD), such as Crohn disease and ulcerative colitis. In early clinical trials as well as in animal models, IL-12 has been implicated as a major mediator of these diseases based on the ability of anti-p40 mAb treatment to reverse intestinal inflammation. The cytokine IL-23 shares the same p40 subunit with IL-12, and the anti-p40 mAbs used in human and mouse IBD studies neutralized the activities of both IL-12 and IL-23. IL-10-deficient mice spontaneously develop enterocolitis. To determine how IL-23 contributes to intestinal inflammation, we studied the disease susceptibility in the absence of either IL-23 or IL-12 in this model, as well as the ability of recombinant IL-23 to exacerbate IBD induced by T cell transfer. Our study shows that in these models, IL-23 is essential for manifestation of chronic intestinal inflammation, whereas IL-12 is not. A critical target of IL-23 is a unique subset of tissue-homing memory T cells, which are specifically activated by IL-23 to produce the proinflammatory mediators IL-17 and IL-6. This pathway may be responsible for chronic intestinal inflammation as well as other chronic autoimmune inflammatory diseases.

Spondyloarthropathy is characterised by inflammation, bone erosion, and new bone formation at the entheseal insertions of tendons and ligaments to bone. Lack of understanding of the underlying mechanisms that drive entheseal disease has seriously inhibited design of therapeutics. Although anti-tumour necrosis factor (TNF) therapy reduces signs and symptoms of inflammation, residual inflammation continues and bone growth is not inhibited. This suggests that TNF is not the optimum target to modify entheseal disease. We previously demonstrated that interleukin (IL) 23 is pivotal in autoimmune inflammation. Recently, genetic variants in the IL-23 receptor (IL23R) have been associated with development of spondyloarthropathy. Moreover, HLA-B27 (present in 90% of patients with ankylosing spondylitis) as well as associated bowel inflammation have been shown to induce IL-23 expression. However, the site and mechanism of action of IL23 are unknown and the reason why such dysregulation of IL23 is associated with inflammation specifically at the enthesis has been inexplicable. We now demonstrate that the entheses and aortic root contain a novel population of resident IL23R+ T cells, which allow the tissue to rapidly respond to IL23. Intravital microscopy reveals that these cells display an extremely restricted entheseal localisation and are absent from synovium and tendon proper. The cells express the molecule PLZF, which allows them to respond to cytokines extremely rapidly, and indeed entheses respond within hours to IL23 in vitro. Moreover, IL-23 expression in vivo in mice is sufficient by itself to rapidly induce the hallmark features of spondyloarthropathy with development of exceedingly specific enthesitis, aortitis, and typical bone erosion and new bone formation. The highly restricted distribution of IL23R+ cells provides the fundamental basis for the anatomical localisation of inflammation observed in spondyloarthropathy, as well as allowing a unified understanding of the genetic associations. These findings suggest that neutralisation of IL23 may be a truly disease modifying therapeutic approach. Funding Merck.

Cancer can induce expansion of CD11b+Gr-1+ myeloid cells, attributed to immune suppression. We noted tumor-activation of these cells further correlated with predisposition for anaphylactoid-type reactions in mice bearing 4T1 mammary carcinomas, which induces strong CD11b+Gr-1+ expansion. To test this hypothesis, 4T1-tumor bearing (4T1+) mice were challenged with recombinant adenoviral vector (rAd), previously shown to activate myeloid lineage cells. When challenged with a single intravenous dose of rAd, 4T1+ mice succumbed to acute anaphylactoid-type reactions. In contrast, minimal responses were seen in rAd treated mice implanted with the related cell line, 66cl4, which do not induce CD11b+Gr-1+ cells. Liposome depletion prior to rAd challenge protected 4T1+ mice, whereas, passive transfer of CD11b+Gr-1+ cells from 4T1+ mice was sufficient to convey susceptibility for anaphylactoid reactions in normal mice. Metabolic pathways were upregulated for nitric oxide and leukotrienes in the 4T1-induced CD11b+Gr-1+ cells. Pre-treating 4T1+ mice with inhibitors of NO-synthetase and leukotrienes inhibited rAd-induced anaphylaxis. Thus, malignant tumor growth can alter CD11b+Gr-1+ myeloid cells, rendering hosts susceptible to anaphylactoid-type reactions. Importantly, there was no requirement for pre-existing Ag-specific IgE or IgG. This reflects clinical reports in which cancer patients undergo treatment-induced anaphylactoid responses in absence of pre-exposure and may provide insight for modeling alternative pathways induced by direct activation of inflammation- or tumor-modified myeloid cells. * Funded by the Schering-Plough Research Institute