Abstract 4021: The class I HDAC inhibitor, mocetinostat, induces expression of PD-L1 and tumor antigen presentation machinery and modifies tumor immune cellular subsets providing a rationale for immune checkpoint inhibitor combinations

Immunotherapy has led to major treatment breakthroughs for a number of cancers including non-small cell lung cancer (NSCLC). Although initial responses to immune checkpoint inhibitors are promising, a significant percentage of patients do not respond or rapidly acquire resistance. Although the mechanisms underlying intrinsic and acquired resistance remain largely unexplained; the expression of programmed cell death-ligand 1 (PD-L1), lack of tumor cell capacity to effectively present neoantigens, and presence of immunosuppressive cellular subsets have been implicated as potential mechanisms. Histone deacetylase (HDAC) inhibitors have emerged as a class of agents that may combat checkpoint inhibitor resistance by reversing immune evasion and eliciting an anti-tumor activity through a multi-faceted immuno-stimulatory mechanism of action. Mocetinostat is a spectrum-selective Class I/IV HDAC inhibitor specifically targeting HDAC-1, -2, -3 and -11. The present studies were designed to explore mocetinostat's effect as an immune-enhancer and ultimately, to evaluate its potential to be used in combination with immune checkpoint inhibitors (e.g., PD-1/PD-L1 antagonists). Specifically, we assessed mocetinostat's effect on the expression of various immunomodulatory factors by tumor cells as well as its effect on immune cell sub-populations in the tumor microenvironment in vivo. Mocetinostat elicited a concentration-dependent increase in PD-L1 mRNA expression which translated into increased PD-L1 surface protein expression in a panel of NSCLC cell lines. In addition, mocetinostat elicited a concentration-dependent increase in expression of MHC-class I related polypeptide-related sequence A (MIC-A) and MIC-B, and cluster of differentiation 86 (CD86). Furthermore, mocetinostat induced expression of several human leukocyte antigen (HLA) gene complex family members including HLA-A, -B, -DRA, and -DPA among others. To determine the effect of mocetinostat on systemic and tumor immune cell subpopulations we treated CT26 tumor-bearing mice. Mocetinostat increased splenic CD4-positive T effector cells and tumor mature cytolytic CD8-postive T cells and at the same time decreased tumor FoxP3-positive T regulatory cells and CD11b/Gr1-positive myeloid-derived suppressor cells (MDSC). These data provide evidence that mocetinostat modulates key immune regulators both in tumor cells as well as in relevant immune cell types in the tumor microenvironment and provides strong rationale for combination with immune checkpoint inhibitors.

Citation Format: David Briere, Niranjan Sudhakar, Lars Engstrom, Jill Hallin, Ruth Tang, Harrah Chiang, Maryland Rosenfeld-Franklin, Peter Olson, James Christensen. The class I HDAC inhibitor, mocetinostat, induces expression of PD-L1 and tumor antigen presentation machinery and modifies tumor immune cellular subsets providing a rationale for immune checkpoint inhibitor combinations. [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 4021.

Abstract 3242: In depth myeloid cell characterization in the murine syngeneic CT26 colon carcinoma model by 10 color flow cytometry

The efficacy of immune-modulating anti-cancer therapeutic antibodies that have been FDA-approved in recent years, such as anti-CTLA-4 and anti-PD-1, has driven growing interest in methods that provide a mechanistic understanding of drug function. Development of new mono- and combination therapies with immune-modulatory effects requires more powerful immunophenotyping techniques capable of in depth cell characterization. To this end, using the CT26 murine syngeneic colorectal cancer model we have developed a 10 color flow cytometry antibody panel that focuses on the identification of tumor-infiltrating immune cell subsets derived from myeloid lineage precursors utilizing the high-throughput-capable 4-laser, 14-color Attune NxT Flow Cytometer with autosampler. The panel includes a combination of antibodies against CD45, CD3, CD19, CD49b, CD335, CD11b, CD11c, Ly-6G, Ly-6C, F4/80, and CD115. By excluding cells of lymphoid lineage, we show that this panel facilitates analysis of myeloid derived cells including natural killer (NK) cells, macrophages, neutrophils, dendritic cells (DCs), and monocytic or granulocytic myeloid-derived suppressor cell (mMDSCs and gMDSCs) subsets in tumor and peripheral blood. In addition, this combination of antibodies allows for a more complete analysis of MDSCs which can differentially express several disease-relevant myeloid specific markers including Ly-6G, Ly-6C, F4/80, CD11c, and CD115. In the tumor, the Ly-6G-high population demonstrates differential expression of Ly-6C, 21% Ly-6C-high (granulocytes) and 77% Ly-6C-low (mMDSCs). The majority of the granulocytic population was identified as gMDSCs and the remainder as neutrophils based on CD115 expression. Macrophages constitute 27% of Ly-6G-low cells. In blood, 98% of the Ly-6G-high population was also Ly-6C-high, and this population is predominantly neutrophils. No macrophages (Ly6G-low and F4/80+) were identified in the peripheral blood. These data confirm the expected distribution of myeloid lineages in the tissue types investigated. Finally, we show that the accuracy of analysis is enhanced by the use of fluorescence minus-one (FMO) controls to identify those markers that generate dim signals, as well as a viability dye used to exclude dead cells from analysis. Identification of additional potentially responsive immune compartments will facilitate identification and development of potential combination therapies otherwise overlooked by looking primarily at T-cells. This panel allows for a significant expansion of our ability to provide a complex description of the myeloid subset.

Citation Format: Matt Thayer, David Draper, Daniel Saims, Maryland Rosenfeld-Franklin, Scott C. Wise. In depth myeloid cell characterization in the murine syngeneic CT26 colon carcinoma model by 10 color flow cytometry. [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 3242.

Inducible expression of TGFβ, snail and Zeb1 recapitulates EMT in vitro and in vivo in a NSCLC model

The progression of cancer from non-metastatic to metastatic is the critical transition in the course of the disease. The epithelial to mesenchymal transition (EMT) is a mechanism by which tumor cells acquire characteristics that improve metastatic efficiency. Targeting EMT processes in patients is therefore a potential strategy to block the transition to metastatic cancer and improve patient outcome. To develop models of EMT applicable to in vitro and in vivo settings, we engineered NCI-H358 non-small cell lung carcinoma cells to inducibly express three well-established drivers of EMT: activated transforming growth factor β (aTGFβ), Snail or Zeb1. We characterized the morphological, molecular and phenotypic changes induced by each of the drivers and compared the different end-states of EMT between the models. Both in vitro and in vivo, induction of the transgenes Snail and Zeb1 resulted in downregulation of epithelial markers and upregulation of mesenchymal markers, and reduced the ability of the cells to proliferate. Induced autocrine expression of aTGFβ caused marker and phenotypic changes consistent with EMT, a modest effect on growth rate, and a shift to a more invasive phenotype. In vivo, this manifested as tumor cell infiltration of the surrounding mouse stromal tissue. Overall, Snail and Zeb1 were sufficient to induce EMT in the cells, but aTGFβ induced a more complex EMT, in which changes in extracellular matrix remodeling components were pronounced.

Abstract 3370: The role of epithelial to mesenchymal transition (EMT) in resistance to Erlotinib in EGFR mutant NSCLC cell line models

The EGFR kinase inhibitor erlotinib is approved as a maintenance therapy in 1st line NSCLC as well as for treatment of 2nd/3rd line NSCLC and in combination with Gemcitabine for pancreatic cancer. It has been observed that the most pronounced responses to EGFR tyrosine kinase inhibitors (TKI's) were observed in patients whose tumors expressed a mutated form of the EGFR kinase. These mutations mapped to the kinase domain of the receptor and functionally have been shown to render tumors and cells lines onco-addicted to EGFR signaling. Although patients expressing a mutated EGFR show a dramatic initial response to EGFR TKI's, ∼50% of these patients will progress while on therapy after 1-2 years. The mechanisms that underlie this acquired resistance to EGFR TKI therapy have been intensively studied and include but are not limited to the presence of a second mutation, T790M, or increased HGF-MET signaling.

Previously, the role of epithelial to mesenchymal transition (EMT) in resistance to EGFR kinase inhibitors has been described in the context of wild type EGFR. EMT. We were therefore interested in understanding whether EMT could play a role in resistance to EGFR TKIs in the context of an EGFR mutation. Here we show that a panel of NSCLC cell lines, expressing mutant EGFR, can undergo an EMT in response to TGFβ treatment. The cell lines take on a scattered and spindle-like morphology and also down regulate the expression of E-cadherin and up regulate expression of vimentin, classic protein markers of an EMT. Importantly we show that after undergoing EMT, the EGFR mutant line HCC827 has reduced sensitivity to erlotnib treatment which is regained after reversal of the EMT. To further explore whether EMT could play a role in acquired resistance to erlotinib, we generated in vitro cell line models that were resistant to EGFR inhibition through continued culturing in the presence of erlotinib over a 6 month period. Resistant clones generated from parental HCC4006 cells acquired a more scattered and spindle-like morphology consistent with an EMT. These clones had down-regulated E-cadherin and ErbB3 expression and upregulated vimentin, fibronectin and Zeb1 expression and also showed a gene expression pattern consistent with having undergone an EMT. In addition, the resistant H4006 clones were more migratory and invasive than their parental counterpart. Finally we show that the resistant clones are enriched for stem cell markers and have enhanced signaling through the Src family kinases and the JAK-STAT pathway suggesting a mechanistic rationale for their reduced sensitivity to EGFR inhibitors.

Taken together. these data indicate that NSCLC cell lines that express a mutant version of EGFR are able to undergo an EMT which can influence the efficacy of EGFR TKIs, suggesting that this may be an additional mechanism underlying the acquired resistance of NSCLC patients to EGFR therapy in the clinic.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 3370. doi:10.1158/1538-7445.AM2011-3370

Abstract 1631: Tumorigenicity of IGF-1R and IR: Rationale for co-targeting IGF-1R and IR in cancer

The Type 1 insulin-like growth factor receptor (IGF-1R) is a well established mediator of tumor cell proliferation and survival. IGF-1R is required for cellular transformation by a number of oncogenes including Ras, and expression of IGF-1R can promote tumor formation in vivo. Blockade of IGF-1R signaling by either genetic or pharmacological methods results in inhibition of tumor cell proliferation. This understanding has spurred the evaluation of a number of IGF-1R inhibitors in the clinic, including both neutralizing antibodies and small molecule kinase inhibitors.

IGF-1R is structurally and functionally related to the insulin receptor (IR). Although IR is appreciated for its classical role in glucose metabolism, IR can also regulate cellular proliferation. Furthermore, there is evidence for compensatory crosstalk between IGF-1R and IR. In embryonic development IR signaling can compensate for loss of IGF-1R to maintain normal embryonic weight. A growing body of data indicates that tumor cells may also exploit IR signaling for proliferation and survival. Tumor cells frequently co-express both IGF-1R and IR, and increased expression of both IGF ligands and insulin are associated with increased risk of cancer. Elevated expression of the IR(A) fetal variant, which is potently activated by both insulin and IGF-2, is observed in select human tumors. Although the tumorigenicity for IGF-1R is well described, the potential for IR as a tumor maintenance gene in vivo has thus far not been established. Herein, we sought to address the tumorigenic potential for IGF-1R compared with IR(A). We used a mouse mammary tumor model driven by an inducible human HER2 oncogene under doxycyclin-directed expression, where repression of HER2 expression upon doxycyclin withdrawal was followed by introduction of genes encoding either IGF-1R or IR(A) in combination with IGF2. We find that either IGF-1R or IR(A), in combination with the ligand IGF2, can complement tumor growth. The growth of both IGF-1R and IR(A) direct complementation (DC) tumor models could be inhibited by the dual IGF-1R/IR small molecule inhibitor OSI-906. Cell lines derived from the IGF-1R and IR(A) DC tumors show that expression of either receptor individually can maintain signaling through the IRS-AKT signaling pathway. Collectively, these observations show that either IGF-1R or IR(A) have tumorigenic potential, and indicate that dual targeting of IGF-1R and IR may be required for optimal activity in tumors where both receptors are present and activated.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 1631. doi:10.1158/1538-7445.AM2011-1631

Abstract LB-388: Combination treatment of a dual IGF-1R/IR kinase inhibitor, OSI-906, with EGFR inhibitor erlotinib in models of non-small cell lung cancer with an EGFR activating mutation

Erlotinib is a tyrosine kinase inhibitor of EGFR approved for use in non-small cell lung cancer (NSCLC) and pancreatic cancer in combination with Gemcitabine. Within the context of NSCLC it has been observed that patients harboring an activating mutation within the kinase domain of EGFR demonstrate a high response to treatment. However, their disease often progresses within 1 year. OSI-906 targets IGF-1R and IR receptor tyrosine kinases and has shown combination efficacy with erlotinib in preclinical models of NSCLC with wild type EGFR. Here we report preclinical data suggesting that the combination of erlotinib and OSI-906 is more efficacious in NSCLC models expressing an EGFR mutation than either single agent. A panel of NSCLC cells lines with confirmed mutations in the EGFR kinase domain was screened for erlotinib and OSI-906 single agent drug sensitivities. All mutant EGFR models exhibited sensitivity to EGFR inhibition by erlotinib (IC50 7–33 nM) while none showed sensitivity to OSI-906 (IC50 7gt;10 uM). In vitro, the combination of erlotinib and OSI-906 resulted in synergistic inhibition of cell proliferation and induction of apoptosis in PC3 (JCP-1) cells. Mechanistically we found the combination had enhanced inhibition of signaling through the PI3K/AKT pathway compared to treatment with either single agent. We also observed that while OSI-906 showed no induction of cell death, erlotinib treatment caused almost complete cell death. However after 9 days of erlotinib treatment small colonies of viable cells still remained. Removal of drug allowed the cells to grow and expand. When cells were treated with both erlotinib and OSI-906, complete cell death occurred without any evidence of cell regrowth following removal of both drugs. In order to evaluate this combination effect in vivo the EGFR mutant human tumor xenograft models NCI-H1650 and PC-14 were employed. Both lines are highly sensitive to erlotinib single agent treatment but show no anti-tumor activity with OSI-906 single agent treatment. The combination of 100 mg/kg erlotinib and 10 mg/kg OSI-906 demonstrated enhanced tumor growth delay when compared to either single agent dosed at MTD. However, as a first step towards understanding preclinical mechanism of this interaction we investigated the combination of 25 mg/kg of erlotinib with 30 mg/kg of OSI-906 and identified potential in vivo synergy. These studies with 25 mg/kg of erlotinib and 30 mg/kg of OSI-906 showed initial tumor regressions, enhanced tumor inhibition and substantially prolonged tumor growth delay when compared to either single agent. Drug-drug interaction PK studies suggest that OSI-906 is not acting to enhance erlotinib exposure. This data provides preclinical proof-of-concept for the use of OSI-906 in combination with erlotinib to obtain greater anti-tumor activity in the mutant EGFR setting.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr LB-388. doi:10.1158/1538-7445.AM2011-LB-388

Compensatory insulin receptor (IR) activation on inhibition of insulin-like growth factor-1 receptor (IGF-1R): rationale for cotargeting IGF-1R and IR in cancer

Insulin-like growth factor-1 receptor (IGF-1R) is a receptor tyrosine kinase (RTK) and critical activator of the phosphatidylinositol 3-kinase-AKT pathway. IGF-1R is required for oncogenic transformation and tumorigenesis. These observations have spurred anticancer drug discovery and development efforts for both biological and small-molecule IGF-1R inhibitors. The ability for one RTK to compensate for another to maintain tumor cell viability is emerging as a common resistance mechanism to antitumor agents targeting individual RTKs. As IGF-1R is structurally and functionally related to the insulin receptor (IR), we asked whether IR is tumorigenic and whether IR-AKT signaling contributes to resistance to IGF-1R inhibition. Both IGF-1R and IR(A) are tumorigenic in a mouse mammary tumor model. In human tumor cells coexpressing IGF-1R and IR, bidirectional cross talk was observed following either knockdown of IR expression or treatment with a selective anti-IGF-1R antibody, MAB391. MAB391 treatment resulted in a compensatory increase in phospho-IR, which was associated with resistance to inhibition of IRS1 and AKT. In contrast, treatment with OSI-906, a small-molecule dual inhibitor of IGF-1R/IR, resulted in enhanced reduction in phospho-IRS1/phospho-AKT relative to MAB391. Insulin or IGF-2 activated the IR-AKT pathway and decreased sensitivity to MAB391 but not to OSI-906. In tumor cells with an autocrine IGF-2 loop, both OSI-906 and an anti-IGF-2 antibody reduced phospho-IR/phospho-AKT, whereas MAB391 was ineffective. Finally, OSI-906 showed superior efficacy compared with MAB391 in human tumor xenograft models in which both IGF-1R and IR were phosphorylated. Collectively, these data indicate that cotargeting IGF-1R and IR may provide superior antitumor efficacy compared with targeting IGF-1R alone.

Feedback mechanisms promote cooperativity for small molecule inhibitors of epidermal and insulin-like growth factor receptors

Epidermal growth factor receptor (EGFR) and insulin-like growth factor-I receptor (IGF-IR) can cooperate to regulate tumor growth and survival, and synergistic growth inhibition has been reported for combined blockade of EGFR and IGF-IR. However, in preclinical models, only a subset of tumors exhibit high sensitivity to this combination, highlighting the potential need for patient selection to optimize clinical efficacy. Herein, we have characterized the molecular basis for cooperative growth inhibition upon dual EGFR and IGF-IR blockade and provide biomarkers that seem to differentiate response. We find for epithelial, but not for mesenchymal-like, tumor cells that Akt is controlled cooperatively by EGFR and IGF-IR. This correlates with synergistic apoptosis and growth inhibition in vitro and growth regression in vivo upon combined blockade of both receptors. We identified two molecular aspects contributing to synergy: (a) inhibition of EGFR or IGF-IR individually promotes activation of the reciprocal receptor; (b) inhibition of EGFR-directed mitogen-activated protein kinase (MAPK) shifts regulation of Akt from EGFR toward IGF-IR. Targeting the MAPK pathway through downstream MAPK/extracellular signal-regulated kinase kinase (MEK) antagonism similarly promoted IGF-driven pAkt and synergism with IGF-IR inhibition. Mechanistically, we find that inhibition of the MAPK pathway circumvents a negative feedback loop imposed on the IGF-IR- insulin receptor substrate 1 (IRS-1) signaling complex, a molecular scenario that parallels the negative feedback loop between mTOR-p70S6K and IRS-1 that mediates rapamycin-directed IGF-IR signaling. Collectively, these data show that resistance to inhibition of MEK, mTOR, and EGFR is associated with enhanced IGF-IR-directed Akt signaling, where all affect feedback loops converging at the level of IRS-1.

A novel, potent, and selective insulin-like growth factor-I receptor kinase inhibitor blocks insulin-like growth factor-I receptor signaling in vitro and inhibits insulin-like growth factor-I receptor dependent tumor growth in vivo

Insulin-like growth factor-I receptor (IGF-IR) and its ligands, IGF-I and IGF-II, are up-regulated in a variety of human cancers. In tumors, such as colorectal, non-small cell lung, ovarian, and pediatric cancers, which may drive their own growth and survival through autocrine IGF-II expression, the role of IGF-IR is especially critical. Here, we present a novel small-molecule IGF-IR kinase inhibitor, cis-3-[3-(4-methyl-piperazin-l-yl)-cyclobutyl]-1-(2-phenyl-quinolin-7-yl)-imidazo[1,5-a]pyrazin-8-ylamine (PQIP), which displayed a cellular IC(50) of 19 nmol/L for inhibition of ligand-dependent autophosphorylation of human IGF-IR with 14-fold cellular selectivity relative to the human insulin receptor. PQIP showed minimal activity against a panel of 32 other protein kinases. It also abolished the ligand-induced activation of downstream phosphorylated AKT and phosphorylated extracellular signal-regulated kinase 1/2 in both IGF-IR transfectant cells and a GEO human colorectal cancer cell line. Analysis of GEO cells revealed a significant level of both phosphorylated IGF-IR and IGF-II expression. Furthermore, inactivation of IGF-II in conditioned GEO culture medium by a neutralizing antibody diminished IGF-IR activation, indicating the presence of a functional IGF-II/IGF-IR autocrine loop in GEO cells. Once daily oral dosing of PQIP induced robust antitumor efficacy in GEO xenografts. The antitumor efficacy correlated with the degree and duration of inhibition of tumor IGF-IR phosphorylation in vivo by this compound. Moreover, when mice were treated for 3 days with a dose of PQIP that maximally inhibited tumor growth, only minor changes in blood glucose were observed. Thus, PQIP represents a potent and selective IGF-IR kinase inhibitor that is especially efficacious in an IGF-II-driven human tumor model.

Interleukin 31, a cytokine produced by activated T cells, induces dermatitis in mice

T cell-derived cytokines are important in the development of an effective immune response, but when dysregulated they can promote disease. Here we identify a four-helix bundle cytokine we have called interleukin 31 (IL-31), which is preferentially produced by T helper type 2 cells. IL-31 signals through a receptor composed of IL-31 receptor A and oncostatin M receptor. Expression of IL-31 receptor A and oncostatin M receptor mRNA was induced in activated monocytes, whereas epithelial cells expressed both mRNAs constitutively. Transgenic mice overexpressing IL-31 developed severe pruritus, alopecia and skin lesions. Furthermore, IL-31 receptor expression was increased in diseased tissues derived from an animal model of airway hypersensitivity. These data indicate that IL-31 may be involved in promoting the dermatitis and epithelial responses that characterize allergic and non-allergic diseases.