Search

Recently there has been interest in developing cell and gene therapies with adult stem cells from human bone marrow referred to as mesenchymal stem cells or marrow stromal cells (hMSCs). We incubated early-passage hMSCs in serum-free medium without cytokines or other supplements for 2 to 4 weeks. Surprisingly, a subpopulation of the cells survived serum deprivation and then began to proliferate in serum-containing medium. The cells selected by serum deprivation had longer telomeres than control cells. Also, the patterns of gene expression revealed by reverse transcriptase–polymerase chain reaction (RT-PCR) assays and microarray data indicated that the cells selected by serum deprivation were a subpopulation of very early progenitor cells with enhanced expression of octomer-binding transcription factor 4 (OCT-4) and several other genes characteristically expressed in embryonic cells.

Abstract 234 Background: NF-κB is a family of transcription factors known to play an essential role in the development & survival of lymphocytes. In recent years, it has been clear that aberrant NF-κB activation is a hallmark of various lymphoid malignancies and appears to be associated with chemotherapy resistance and adverse prognosis. The canonical NF-κB pathway is frequently engaged in lymphoid malignancies wherein activated IKK phosphorylates IκB proteins inducing IκB polyubiquination and subsequent proteasomal proteolytic degradation; this allows for release and nuclear translocation of NF-κB dimers to activate target gene transcription. Gene Expression Profiling (GEP) has identified two distinct sub groups of Diffuse Large B cell Lymphoma (DLBCL). While the Activated B cell (ABC) type shows constitutive activation of NF-κB, the role for NF-κB activation in Germinal Center B Cell (GCB) DLBCL is currently unclear. Since NF-κB inhibition has been identified as a therapeutic possibility in DLBCLs, it is important to define the role of this pathway and its modulators. In this study, we sought to investigate key regulators of the NF-κB pathway that might mediate a therapeutic response to IKKβ inhibition of NF-κB in DLBCL including the GCB subtype. Through GEP and Exome Sequencing, we demonstrate that ANTXR1 is a key mediator of response to IKKβ inhibition in DLBCL. Methods/Results: We obtained a novel selective inhibitor of IKKβ, TLX-2001 that has been found to be safe in animal models. IC50 were obtained on 61 cell lines representing various lymphomas including DLBCL (N=25) using cell viability MTT assays. The drug showed efficacy in both ABC and GCB DLBCL cell lines at physiologically achievable concentrations. These results were unsurprising in ABC DLBCLs which are known to depend on NF-κB activation, but the lethality of this selective drug in GCB DLBCLs was unexpected. To better understand the role of individual genes in the response in GCB DLBCLs, gene expression profiling was performed on 61 cell lines using Human Gene 1.0 ST Array. We found that ANTXR1 expression significantly correlated with NF-κB resistance (p = 0.035). Additionally, we sequenced the exomes of DLBCL tumors (N=95) and matched normal tissue (N=34). 95 cases of DLBCLs consisted of 73 cases of primary human DLBCLs and 22 DLBCL cell lines. Whole exome sequencing was performed using the Agilent solution-based system of exon capture to sequence all protein coding exons in the CCDS database. We identified 465 recurrently somatically mutated genes in these DLBCL cases, and found that mutation status of ANTXR1 was associated with high sensitivity to IKKβ inhibition (p =0.015) of NF-κB. Cell lines with non-synonymous mutations in ANTXR1 had over 3-fold lower IC50 (mean= 2.39 μM) compared to cell lines with no mutation in ANTXR1 (mean IC50 = 8.72μM). Discussion/Conclusion: ANTXR1 is the docking receptor for bacillus anthracis toxin, and anti-tumor responses have been observed in mice injected with recombinant engineered anthrax toxin. It is also known as TEM8 and maps to chromosome 2p13.1. Increased levels of TEM8 (tumor endothelial marker 8) have been noted in various malignancies including melanoma. Our data suggest that pharmacogenetic approaches that combine gene expression profiling and whole exome sequencing are useful tools for identifying novel genes that modulate therapeutic responses in lymphoma. Disclosures: No relevant conflicts of interest to declare.

Abstract 1848 Background: The hedgehog signaling pathway has been shown to be important in the development of malignancies, however, the role of this pathway in lymphomas is uncertain. GDC-0449 is a novel small molecule inhibitor that has been shown to block hedgehog activity by inhibiting SMO. The purpose of this study was to determine a potential role for hedgehog inhibition in the treatment strategies for lymphomas. We further sought to understand the determinants of sensitivity and potential resistance to therapeutic inhibition of this pathway. Methods and Results: We tested the small molecule inhibitor GDC-0449 in cell lines that span the spectrum of lymphoma including Burkitt lymphoma (N=4), ABC DLBCL (N=3), GCB DLBCL (N=7), mantle cell lymphoma (N=2), multiple myeloma (N=2), lymphoid leukemia (N=2), and T cell leukemia (N=2). Using the MTT assay, we identified the IC50 of drug response in these cell lines. We performed gene expression profiling in these cell lines and applied the Cox proportional hazards model using IC50 as the survival variable to identify genes that were associated with either sensitivity or resistance to the drug. We found that the drug was lethal in these cell lines at physiologically achievable concentrations with IC50s ranging from 5 to 22μM. We applied hierarchical clustering to the genes associated with sensitivity and resistance and found that the method distinguished two groups of samples that differed in their average IC50s by over 2-fold (8μM vs 17μM). There was a suggestion of association with lymphoma cell type with Burkitt lymphoma lines being the dominant cell type in the resistant group and DLBCL being the dominant cell type in the sensitive group. Resistance to hedgehog inhibition was found to be highly associated with the phosphatidylinositol 3′-kinase (PI3K) signaling pathway, suggesting that the hedgehog pathway and PI3K pathways act through independent, complementary mechanisms to promote tumor proliferation. In order to test the hypothesis that PI3K signaling is independent of hedgehog signaling we tested the lethality of PI3K inhibition using a dual PI3K/mTOR inhibitor (BEZ235) in the same cell lines. We found an inverse relationship between the hedgehog inhibitor and PI3K inhibitor, i.e. the cell lines that were relatively sensitive to hedgehog inhibition (mean IC50 of 8μm) were resistant to PI3K inhibition(mean IC50 of 0.69μm) and the cell lines that were relatively resistant to hedgehog inhibition (17μm) were relatively sensitive to PI3K inhibition (0.32μm). Conclusion: Hedgehog pathway inhibition is a promising therapeutic approach in a carefully defined subset of patients with lymphoma. Hedgehog and PI3K inhibitors appear to act by distinct mechanisms and may be complementary in the treatment of lymphomas. Disclosures: No relevant conflicts of interest to declare.

Abstract 1849 Background: The NF-kappaB pathway is an essential and tightly regulated signaling cascade that mediates development, activation and survival of normal lymphocytes. In recent years, there has been convincing evidence implicating this pathway in several lymphoid malignancies including the activated B cell-like (ABC) subtype of Diffuse Large B cell Lymphoma(DLBCL), Primary Mediastinal B Cell Lymphoma(PMBL), marginal zone lymphoma, and Hodgkins lymphoma. These tumors frequently engage the classic NF-kappaB pathway in which the IKK β subunit phosphorylates IκBα triggering its ubiquitination and subsequent proteasomal degradation. Therefore, therapies targeting this pathway through selected IKKB inhibition may provide a novel approach for targeting NF-kappaB activity in lymphoid malignancies. We therefore decided to explore the efficacy of a novel IKKB inhibitor CMPDA in a broad group of lymphomas. Methods: We chose 37 cell lines that broadly represented a diverse group of lymphomas including Burkitt lymphoma, mantle cell lymphoma, Hodgkin lymphoma, multiple myeloma, prolymphocytic leukemia, and the activated B cell-like (ABC) and germinal center (GCB) subtypes of DLBCL. Cell viability assays using MTT were performed to identify the IC50 in these cell lines. We performed gene expression profiling on all cell lines using a GeneChip Human Gene 1.0 ST Array which comprises 33297 probes that target all known genes. Utilizing hierarchial clustering algorithms in conjunction with cox regression analysis, the cell lines were classified into 2 groups based on gene expression: the predicted sensitive(Group A: 20 cell lines) versus predicted less sensitive cell lines (Group B: 17 cell lines). Results: CMPDA was broadly effective in lymphomas demonstrating lethality in most lymphoma types at physiologically achievable concentrations with a mean IC50 of 9.15 micromoles (Range: 1.47 to 58.24 micromoles). The Mean IC50 for cell lines in Group A which were predicted as being sensitive by gene expression was 5 micromoles, whereas the mean IC50 for the cell lines in Group B which were predicted as being relatively resistant to NF-KB inhibition was 3-fold higher (∼15micromoles). Group A included a number of cell lines with known dependence on NF-kappa B activity including several ABC DLBCL cell lines. It is notable that while GCB DLBCL have generally been thought to have little dependence of NF -KB activity, we nevertheless found that there were 4 GCB DLCBL cell lines that were also sensitive to NFKB inhibition at IC50s comparable to a number of ABC DLBCLs. Group B comprised of a disproportionate number of Burkitt lymphoma cell lines (4/17), which is consistent with the known biology of this tumor as it typically expresses NF-KB at lower levels. Genes associated with response to NF-KB inhibition showed robust concordance with the known biology of the NF-kappaB pathway, with gene expression signatures related to CD40-stimulation (an upstream activator of NF-kappa B) and NF-kappa B activity showing the strongest associations with response to CMPDA (P=0.004 and P=0.01 respectively). Conclusions: Thus our data demonstrate that selective IKKβ inhibition of the NF-kappa B pathway using CMPDA is a promising therapeutic strategy in lymphoma. Gene expression profiling provides a useful tool for potentially identifying tumors that are most likely to respond to this therapy. Disclosures: No relevant conflicts of interest to declare.

Abstract 434 Background Histone deacetylase inhibitors (HDACis) have demonstrated significant clinical activity in hematologic malignancies; however, single agent response rates have ranged between 20–50% with the duration of response often measured in months, suggesting that drug resistance is a major mode of failure. The pathways through which these agents work and the means by which tumors develop resistance to them are poorly understood. Combination therapy targeting multiple oncogenic pathways holds the promise to improve upon both the depth and durability of these responses. We investigated the mechanisms of inherent and acquired resistance to HDACis in a broad range of lymphomas. By detailing the molecular pathways implicated in resistance to HDACi, we sought to identify novel combinations of compounds that could overcome potential mechanisms that confer resistance. Methods and Results We tested two separate HDACis, LBH589 and SAHA in 51 cell lines representing a wide range of lymphomas including Burkitt lymphoma, diffuse large B cell lymphoma (DLBCL), mantle cell lymphoma, and Hodgkin lymphoma. Gene expression array data was generated for all these cell lines. We then identified genes that were significantly associated with resistance to both LBH589 and SAHA (p To determine whether these genes that we found to be associated with resistance reflected potential mechanisms of acquired resistance to HDACi therapy, we separately engineered resistance to LBH589 and SAHA in three DLBCL cell lines (LY3, BJAB, Farage) through incremental dose escalation over a period of up to 6 months. Each of these three cell lines demonstrated sustained growth at drug concentrations that were at or above their original IC50. Each of these cell lines were then exposed to the other HDACi and tested for cross resistance. In each case, the cell lines demonstrated complete cross-resistance to the other drug. We then profiled the gene expression of these cell lines that had acquired resistance. Similar to our previous results, these cell lines demonstrated increased signaling through the JAK/STAT pathway, suggesting that mechanisms of inherent and acquired resistance are similar. We therefore reasoned that combining HDAC and JAK inhibition may overcome both inherent and acquired resistance. To investigate this hypothesis, we tested LBH589 and INCB018424, a JAK1/2 inhibitor, alone and in combination in the LY3, TMD-8, U2932, and BJAB cell lines. While INCB018424 demonstrated no single agent cytotoxicity, it yielded a high degree of synergy when combined with LBH589 with the combination index computed by the Chou-Talalay method ranging from .19 to .9. Conclusion HDACis show single agent activity in the treatment of a number of hematologic malignancies, however most patients develop resistance to these drugs after relatively short-lived remissions. Thus, the greatest promise of these drugs may lie in combination with other agents that target molecular pathways that underlie resistance to these drugs. Using gene expression profiling of a broad range of tumor types and sensitivity to HDACis we were able to identify activation of the JAK/STAT pathway as a common feature of inherent and acquired resistance to HDACis. We combined the JAK1/2 inhibitor INCB018424 with LBH589 and demonstrated a high degree of synergy. As the number of small molecule inhibitors with clinical activity increases, the need to identify rational preclinical combinations becomes greater. Pairing gene expression profiling and resistant cell lines is a promising approach to the selection of combinations likely to maximize clinical benefit while limiting toxicity. Disclosures: No relevant conflicts of interest to declare.