Apoptosis In CLL Cells Research Articles

Abstract 3522: Vincristine activates c-Jun N-terminal kinase in chronic lymphocytic leukemia in vivo.

Abstract Chronic lymphocytic leukemia (CLL) is a cancer of B cell lymphocytes whose survival depends on the anti-apoptotic activity of Bcl-2 family proteins. Due to the low growth fraction of CLL cells, drugs that kill during mitosis, such as the microtubule-targeting vinca alkaloids, are not expected to be effective. However, we have found that vincristine and vinblastine can act independently of mitosis and induce acute apoptosis in freshly isolated CLL cells. The mechanism of this effect still depends on destabilization of microtubules as similar events occur with combretastatin A4 which destabilizes microtubules through binding to a different site on tubulin. Vinca-mediated apoptosis in CLL cells can be enhanced by addition of the cyclin-dependent kinase (CDK) inhibitor dinaciclib, killing cells acutely within 6 h. Dinaciclib functions by inhibiting CDK9, suppressing transcription and preventing expression of Mcl-1. The combination of dinaciclib and vinca alkaloids depends on vinca-induced phosphorylation of c-Jun N-terminal kinase (JNK). Therefore, we investigated whether JNK phosphorylation occurs in peripheral blood CLL cells in patients receiving vincristine. Patients received 2 mg vincristine intravenously by infusion over 5 minutes. Venous blood samples were obtained at time zero (pre-treatment) and post drug administration for up to 6 h. Patients then continued to receive chemotherapy treatment as prescribed by their primary oncologist. Pre-treatment CLL cells were studied ex vivo for JNK phosphorylation on exposure to vincristine. These studies showed that CLL cells phosphorylated JNK in response to clinically relevant concentrations of vincristine (10-50 nM). In the CLL cells from all 4 patients studied thus far, vincristine led to rapid (≤ 1 h) JNK phosphorylation which was sustained for 4-6 h post vincristine administration. These proof-of-principle study results provide justification for the development of a Phase I study to define the effect of the combination of vinca alkaloids and dinaciclib in patients with CLL. Citation Format: Darcy JP Bates, Alexey V. Danilov, Bernard B. Beaulieu, Lionel D. Lewis, Alan Eastman. Vincristine activates c-Jun N-terminal kinase in chronic lymphocytic leukemia in vivo. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 3522. doi:10.1158/1538-7445.AM2013-3522

Abstract 3338: Novel caspase-3 activator, L14R8, induces apoptosis in chronic lymphocytic leukemia cells.

Abstract Chronic lymphocytic leukemia (CLL) is a B-cell malignancy and leukemia cells accumulate in bone marrow, lymph node, and peripheral blood due to decreased apoptosis. High levels of Bcl-2 family anti-apoptotic and inhibitors of apoptosis proteins (IAPs) in leukemic lymphocytes are primarily responsible for defective apoptosis in CLL. Both intrinsic caspase-9-mediated and extrinsic caspase-8-dependent apoptosis pathways activate downstream procaspase-3 to active casapce-3. In normal (n=4) as well as CLL lymphocytes (n=9), there is consistent expression of procaspase-3. However, active caspase-3 was present in normal lymphocytes but not in CLL cells suggesting that CLL cells have defective endogenous apoptosis. Hence, we hypothesized that compounds that can directly activate the functional ability of procaspase-3, should overcome the checkpoints of cell death and trigger apoptosis in CLL lymphocytes. Procaspase activating compounds (PAC; Vanquish Oncology) convert executioner procaspases to their active forms by chelating labile zinc on these enzymes. We have tested L14R8, a potent analogue of PAC, in primary CLL cells. The IC50 of L14R8-induced apoptosis in CLL was 7.7μM (n=12, 24hrs). L14R8 (10 μM) induced apoptosis as early as 6 hrs along with mitochondrial outer membrane permeabilization (MOMP). The drug significantly induced apoptosis in CLL patient samples (n=4) cultured in suspension cultures (60%, p=0.023,) as well as CLL lymphocytes under microenvironmental protection of marrow stromal cells (40%, p=0.011) albeit at a lower rate. Mechanistically, L14R8-treatment reduced uncleaved procaspases-3 and -9 expression and increased cleaved active caspases-3 and -9 expression (n=3). Consistent with the mechanism of action of L14R18 on terminal caspases, WT as well as cells lacking BAX and BAK (mouse embryo fibroblasts) were equally sensitive to this agent. Consistent with L14R8 mechanism, addition of exogenous Zinc ion completely reversed L14R8-mediated apoptosis of CLL lymphocytes (p=0.0063, n=4), but did not affect staurosporin-induced apoptosis in these lymphocytes. Moreover, L14R8-induced 86% (range 58-90%, n=4) cell death in CLL lymphocytes, which was reduced by pan caspase-inhibitors, Z-VAD to 58% (range 32-79%) and Q-VD-OPh to 35% (range 20-55%) cell death, which indicated the mechanistic recovery of L14R8-mediated apoptosis. Smac (the inhibitor of IAPs) protein expression was increased in CLL cells after treatment with L14R8. With respect to the therapeutic index, at 24 hrs, L14R8 induced 79% apoptosis in CLL cells (range 46%-97%, n=14) and only 18% (16-28%, n=3) in lymphocytes from healthy donors. We conclude that L14R8 induced apoptosis through activation of procaspase-3 to active caspase-3 in CLL cells. L14R8 and other PAC provide therapeutic index and may be developed into targeted therapeutics for patients with CLL. *Funded by Vanquish Oncology Citation Format: Viralkumar M. Patel, Kumudha Balakrishnan, William Wierda, Varsha Gandhi. Novel caspase-3 activator, L14R8, induces apoptosis in chronic lymphocytic leukemia cells. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 3338. doi:10.1158/1538-7445.AM2013-3338

Abstract SY24-02: Remarkable early clinical activity with the Bcl-2 selective inhibitor ABT-199: Proving the concept.

Abstract SY24-02: Remarkable early clinical activity with the Bcl-2 selective inhibitor ABT-199: Proving the concept.

IGF1R Inhibition Induces Apoptosis in Chronic Lymphocytic Leukemia

AbstractAbstract 3864Chronic lymphocytic leukemia (CLL) is characterized by the accumulation of monoclonal B lymphocytes. For decades, nucleoside analogs, alkylating agents, and immunotherapeutics have remained the mainstay in treating this disease. Despite major advances in this field, CLL remains incurable with standard therapy. In recent years, preclinical and early clinical data on the use of kinase inhibitors have sparked new hope in the treatment of CLL. The multikinase inhibitor sorafenib, targeting RAF, platelet-derived growth factor receptor (PDGFR), KIT, FMS-like tyrosine kinase 3 (FLT3), and vascular endothelial growth factor receptor (VEGFR), has been approved for the treatment of renal cell carcinoma and hepatocellular carcinoma. Recent studies suggested that CLL cells might be also susceptible to this compound, however the precise mode of action in CLL cells remains elusive. In this study, we identified the Insulin-like growth factor receptor-I (IGF1R) pathway as novel target of sorafenib inducing cell death in CLL cells.Treatment with 10 μM of sorafenib significantly increased apoptosis in primary CLL cells as determined by AnnexinV/PI staining via flow cytometry. Commensurate with its RAF inhibiting properties, the apoptotic effect of sorafenib was accompanied with ERK pathway inhibition. Moreover, sorafenib treatment decreased phosphorylation of SRC and AKT, molecules implicated with IGF1R and insulin receptor (IR) signaling. Interestingly, the latter were strongly expressed in primary CLL cells compared with healthy B cells. Similar to sorafenib, 24 hour treatment of CLL cells with the three structurally distinct IGF1R inhibitors Picropodophyllin, AG1024, and Linsitinib significantly increased apoptosis compared with vehicle control resulting in decreased phophorylation of MEK, ERK, SRC, and AKT. Sorafenib and the IGF1R inhibitor AG1024 also downregulated the expression of IGF1R on CLL cells but not on healthy B cells.To test whether sorafenib modulates IGF-1 binding and thereby influences the IGF1R activation, we biotinylated recombinant IGF-1 and tested its binding to the IGF1R via flow cytometry. We observed a reduced IGF-1 binding after sorafenib treatment. IGF-1 binding after treatment with different IGF1R inhibitors was performed as an internal control. In order to further establish the functional relevance of IGF1R expression in CLL, we performed IGF1R specific and non-silencing siRNA experiments in primary CLL cells. In line with our previous results, IGF1R knockdown resulted in a significant decrease of cell viability and in downregulation of RAF-1 expression, and MEK, ERK, SRC, and AKT phosphorylation. The stromal microenvironment protects CLL cells from spontaneous and drug-induced apoptosis. Sorafenib, AG1024, and Picropodophyllin counteracted the protective effect of microenvironmental factors simulated by the presence of the murine stromal cell line M210B4, the chemokine CXCL12, and the integrin CD49d.Finally, we used the Eμ-Tcl1 transgenic mouse model to validate these results in vivo. Male and female mice (n=8) were treated with 25 mg/kg of the IGF1R inhibitor Linsitinib per oral gavage for 7 days and the amount of CD5/CD19 positive cells was determined flow cytometrically at different time points. We observed a reduction of CD5/CD19-positive cells by 26,1% and 23,2% after 4 and 8 days of treatment, respectively.Our results provide a novel mechanism of action of the multikinase inhibitor sofarenib in CLL cells by blocking IGF1R mediated signaling. IGF1R inhibition by itself induced apoptosis in CLL cells in vitro and in vivo, thus identifying IGF1R as promising target for therapeutic approaches and proposing IGF1R inhibitors for clinical assessment in the therapy of CLL. Disclosures:No relevant conflicts of interest to declare.

Differential Expression and Functions of NOTCH Family Members and Involvement of NR4A1 in the Regulation of Apoptosis in CLL

AbstractAbstract 3919Chronic lymphocytic leukemia (CLL) cells express constitutively activated NOTCH2 in a protein kinase C (PKC) dependent manner linking NOTCH2 to the activated state of the leukemic cells. The transcriptional activity of NOTCH2 is associated with the expression of CD23 and enhanced CLL cell viability. However, the regulation and possible functions of the individual NOTCH family members (NOTCH1–4) in CLL cells remain to be clarified.We took advantage of targeting nuclear NOTCH2 using the recently identified NOTCH2 transactivation inhibitor gliotoxin (WO 2006/135949). We also analysed the regulation and possible function of NOTCH1–4 in PKC stimulated CLL cells using a PMA model (Hubmann et al., BJH 2010) and a microenvironment model where CLL lymphocytes were co-cultured with primary bone marrow stromal cells (BMSC) (Shehata et al., BLOOD 2010).Electrophoretic mobility shift assays (EMSA) demonstrated that gliotoxin inhibited DNA-bound NOTCH2 complexes in PMA stimulated CLL cells in parallel to increasing the rate of apoptosis (mean±SD: 67±31% in gliotoxin treated cells versus 13±14% in the untreated controls, n=21). This was associated with downregulation of CD23A mRNA expression and CD23 surface expression (mean±SD: 42±32% versus 83±17%, n=21) as assessed by RT-PCR and FACS analysis. Exceptionally, one CLL case with a recently described NOTCH1 gain of function mutation appeared to be less sensitive to gliotoxin and had a persistent high expression of CD23.We next tested whether NOTCH2 inhibition by gliotoxin is a selective process or indirectly mediated by effects on proteasome regulated apoptosis. Proteasome assays showed that gliotoxin had a minimal or no effect on the chymotrypsin like activity of the proteasome in CLL cells. In addition, the activity of the proteasome regulated transcription factor NFκB and the expression of its target genes like BCL2 and MCL1 were also not influenced by gliotoxin. These data point to the selectivity of targeting NOTCH2 signaling by gliotoxin rather than indirectly through the regulation of proteasome activity.Short term (4 hours) exposure of CLL cells revealed that NOTCH1 was equally transcribed in unstimulated and in PMA activated CLL cells. NOTCH2 was upregulated in PMA activated CLL lymphocytes whereas NOTCH4 was only weakly detectable in unstimulated CLL cells. Gliotoxin treatment resulted in the downregulation of NOTCH1, NOTCH2 and NOTCH4 mRNA expression.Interestingly, the inhibition of NOTCH2 activity by gliotoxin was associated with the concomitant induction of NOTCH3 signaling especially in the presence of PMA. This was indicated by the induced mRNA expression of NOTCH3 and its preferred target gene HEY1. Moreover, the induced transcription of HEY1 correlated with the upregulation of NR4A1, a key regulator of apoptosis in activated lymphocytes. These data may thus point to a pro-apoptotic role for NOTCH3/HEY1/NR4A1 signaling in CLL cells. The data also suggest that gliotoxin induced apoptosis is associated with differential regulation of the anti-apoptotic and pro-apoptotic arms of NOTCH signalling in CLL cells.RT-PCR revealed that NOTCH1 and NOTCH2 are the main NOTCH family members which are expressed in CLL cells under co-culture conditions with BMSC and in freshly isolated CLL cells. Exposure to gliotoxin in co-culture selectively induced apoptosis in CLL cells and led to downregulation of NOTCH1 and NOTCH2 together with upregulation of NOTCH3 mRNA expression.In summary, the data suggest that nuclear NOTCH2 activity might protect activated CLL cells from apoptosis by modulating the expression of NR4A1. The induced expression of NOTCH3 and its target gene HEY1 by gliotoxin reveals the complex role of different NOTCH family members in the regulation of apoptosis in CLL cells. Therefore, the individual NOTCH receptors may have opposite effects on CLL cell viability which should be considered in therapeutic approaches aimed to target NOTCH signaling in CLL. Disclosures:No relevant conflicts of interest to declare.

Monoclonal Antibody Against ROR1 in Chronic Lymphocytic Leukemia Cells Induced Apoptosis Via PI3-Kinase/AKT/CREB Pathway

AbstractAbstract 1769 Background:Phosphoinositide 3-kinase (PI3K)/AKT cascade regulates cell survival, proliferation and differentiation in a variety of cells. In CLL cells PI3K pathway is constitutively activated leading to AKT activation and phosphorylation of cAMP response element-binding protein (CREB). CREB is a transcription factor overexpressed and constitutively phosphorylated in a variety of cancers and seems to have a role in tumor pathobiology. There is a great need to develop novel strategies for targeted therapy in CLL. Monoclonal antibodies (mAbs) specifically targeting leukemic cells might be a rewarding approach. ROR1 is a type I transmembrane receptor tyrosine kinase belonging to one of the twenty families of receptor tyrosine kinases (RTKs). ROR1 is overexpressed on CLL cells but not in white blood cells of healthy donors. ROR1 is constitutively phosphorylated in CLL and siRNA transfection induced apoptosis. We have developed a unique anti-ROR1 mAb directed against CRD (cysteine-rich domain) of the extracellular region of ROR1 capable of inducing direct apoptosis of primary CLL cells. Our anti-CRD mAb induced dephosphorylation of the ROR1 molecule. Aims:To study the apoptotic effect of an anti-ROR1 CRD mAb and effects on downstream signaling pathways involved in CLL, specially the PI3-kinase/AKT/CREB pathway using primary CLL cells. Methods:Using a peptide-based mouse mAb generation method we produced several mAbs against the three extracellular domains of ROR1. In the current study we used one of the best anti-ROR1 antibodies, an anti-CRD mAb raised against the CRD region of ROR1 (Daneshmanesh et al., Leukemia. 2012 Jun;26(6):1348-55). Flow cytometry was used for surface staining of ROR1. Primary CLL cells were incubated with the anti-ROR1 CRD mAb and apoptosis was detected by the MTT assay and Annexin V/propidium iodide (flow cytometry) methods in a 24 h assay. Antibody untreated and treated cell lysates were prepared and subjected to Western blot analysis for identification of signaling molecules involved in apoptosis induced by the anti-ROR1 CRD mAb. We analysed total and phosphorylated levels of the following signaling proteins: AKT, p-AKT, PI3K, p-PI3K, CREB, p-CREB, ERK, p-ERK, PKC and p-PKC. Phosphoproteins were measured before incubation with the mAb and after 20 min-2 h. Results:ROR1 surface expression was detected on 80–85% of the CLL cells. The frequency of apoptotic cells induced by the anti-CRD mAb was in the range of 45–50% which is in accordance with our previous reports (see above). Time kinetics experiments using anti-ROR1 CRD mAb incubated with primary CLL cells revealed dephosphorylation of ROR1 downstream signaling molecules. We analysed the following molecules known to be involved in CLL: PKC, PI3-kinase and ERK1/2. After co-culturing CLL cells with the anti-ROR1 CRD mAb, Western blot analysis showed decreased level of phosphorylated AKT in treated compared to untreated samples. No changes in the phosphorylation levels of ERK1/2 and PKC proteins were seen. Furthermore, we analysed the PI3-kinase protein which is upstream of AKT, and noticed that in CLL cells treated with the anti-ROR1 CRD mAb, the phosphorylation intensity of PI3-kinase p85 isoform has decreased but not p55 isoforrn. Moreover, we also studied the CREB phosphorylation in treated and untreated CLL samples and detected dephosphorylation of CREB in treated as compared to untreated samples. Conclusion:Incubation of CLL cells with an anti-ROR1 CRD mAb induced apoptosis of primary CLL cells. Apoptosis was preceded by dephosphorylation within 2 h of PI3-kinase, AKT and CREB proteins indicating deactivation of these signaling proteins by the anti-ROR1 mab. In untreated CLL cells no effect on phosphorylation of these proteins was noted. Furthermore our ROR1 mAb did not dephosphorylate PKC or ERK. Our data may suggest that activation of CREB molecule might occur via the PI3K/AKT pathway and may be a survival signal in CLL cells associated with the aberrant expression of ROR1. The constitutive phosphorylation of PKC and ERK1/2 seen in CLL might not be related to the overexpression of ROR1. Further studies are warranted for a better understanding of signaling pathways associated with ROR1 and the downstream signaling effects of ROR1 targeting drugs. Disclosures:No relevant conflicts of interest to declare.

Lenalidomide Reduces Survival of Chronic Lymphocytic Leukemia Cells in Primary Co-Cultures by Altering the Myeloid Microenvironment

Abstract 3894Chronic lymphocytic leukemia (CLL) cells are highly dependent on their microenvironment. External stimuli provided by bone marrow stromal cells or non-malignant leukocytes are required for their survival and proliferation. Interestingly, peripheral blood-derived monocytes differentiate in the presence of CLL cells to so-called Nurse-like cells (NLCs), which are round or fibroblast-shaped adherent cells that were shown to promote survival of CLL cells in vitro and to exist in lymph nodes of CLL patients. In search of new therapeutic options for patients with CLL, the immunomodulatory drug lenalidomide turned out to have significant clinical activity in CLL. Lenalidomide does not induce apoptosis in CLL cells directly, but is rather believed to act via the microenvironment. Several studies described that it alters cytokine levels and the activation status of the cells. Further, a CLL-specific T-cell defect was shown to be repaired by lenalidomide, which might represent a major activity of this drug in CLL. However, its mechanism of action seems to be complex and is not well understood.As monocytes as well as NLCs are very effective in maintaining survival of CLL cells, we aimed to investigate whether lenalidomide interferes with these supportive cell-cell interactions. To do this, we established primary co-cultures of monocytes and CLL cells in the presence or absence of lenalidomide and observed a significantly decreased viability of CLL cells after 14 days of treatment, suggesting an impact of this drug on the survival support of NLCs. Therefore, we analyzed the immunophenotype of NLCs by flow cytometry, as well as the secretion of cytokines in the co-cultures by ELISA and antibody-coupled bead arrays. Among the effects induced by lenalidomide, we observed reduced cell surface expression of the MHC II protein HLA-DR on NLCs as well as lower levels of the chemokine CCL2, but higher levels of IL-10 in the culture supernatant, indicating an altered inflammatory milieu in the co-cultures. The enhanced IL-10 levels resulted in an increase in STAT1 phosphorylation in CLL cells as measured by Western blot analysis. As a consequence, enhanced expression of the adhesion molecule ICAM-1 (CD54) and an altered expression of cytoskeletal genes (e.g. RHOC and CORO1B) were observed in CLL cells after lenalidomide treatment. Chemotaxis assays using transwell culture dishes and SDF1-α as chemoattractant revealed an impaired migratory potential of lenalidomide-treated CLL cells, which was not due to reduced expression of the SDF1-α receptor CXCR4. In summary, our data show that lenalidomide reduces the survival support of NLCs for CLL cells in vitro, suggesting that this drug effects the myeloid microenvironment in CLL in vivo. Furthermore, lenalidomide impairs the migratory potential of CLL cells which may affect circulation and homing of CLL cells in vivo. Disclosures:No relevant conflicts of interest to declare.

BMS-936564 (MDX1338): A Fully Human Anti-CXCR4 Antibody Induces Apoptosis in an in Vitro Model of Stromal – Leukemia Cell Interaction for Chronic Lymphocytic Leukemia.

AbstractAbstract 2887Chronic lymphocytic leukemia (CLL) remains incurable despite advances in the biology and treatment of this disease. Current data support the notion that resistance to therapy is promoted by a “protective” tumor microenvironment in which non-leukemia cells produce factors that enhance the resistance of CLL cells to spontaneous or drug-induced apoptosis. One such factor is the chemokine CXCL12, which interacts with its receptor CXCR4 on CLL cells to promote cancer cell survival.To examine the therapeutic potential of blocking CXCL12-CXCR4 interactions, we studied the effect of BMS-936564, a fully human IgG4 anti-CXCR4 antibody, using an in vitro co-culture model of human bone marrow derived stomal-NKter cells – leukemia cell interaction. Such stromal-NKter cells secrete CXCL12 and enhance the resistance of CLL cells to apoptosis in vitro.We observed that primary CLL cells co-cultured with stromal-NKter cells had significantly greater viability than CLL cells cultured alone (20–60% above baseline at 48 hours). Moreover, CLL cells co-cultured with stromal cells had enhanced resistance to drug-induced apoptosis. We found that BMS-936564 antibody at concentrations of 2–200nM could enhance the rate of apoptosis of CLL cells cultured alone or in the presence of stromal cells. CLL cells that expressed unmutated IgVH genes or ZAP-70 appeared equally susceptible to treatment with BMS-936564 as did CLL cells that lack these adverse prognostic markers, as did CLL cells that harbored deletions in 17p13.2 and that were resistant to chemotherapeutic agents, such a fludarabine monophosphate. BMS-936564 antibody inhibited CXCL12 mediated F-Actin polymerization in CLL cells at lower concentrations (20–200nM) compared to AMD-3100 (Mozobil), a small molecule CXCR4 inhibitor (50–150μM). In addition, AMD-3100 did not induce apoptosis in CLL cells (10–300μM).In summary, we observed that the anti-CXCR4 antibody BMS-936564 inhibited CXCL12 mediated activation of the CXCR4 receptor in CLL cells and induced apoptosis in leukemia cells. The pro-apoptotic activity of BMS-936564 was observed in cells cultured alone or together with stromal cells suggesting that this antibody had direct cytotoxic effect on leukemia cells and that it can overcome the protective tumor microenvironment. More over, the activity of BMS-936564 was independent of the presence of poor prognostic factors such as del(17p) suggesting that its mechanism of action is P53 independent. These findings show evidence that the CXCR4-CXCL12 pathway is a valid therapeutic target in CLL and provide additional biological rationale for ongoing clinical trials in CLL and other hematological malignancies using BMS-936564. Disclosures:Kuhne:Bristol-Myers Squibb: Employment. Sabbatini:Bristol-Myers Squibb: Employment. Cohen:Bristol-Myers Squibb: Employment. Shelat:Bristol-Myers Squibb: Employment. Cardarelli:Bristol-Myers Squibb: Employment. Kipps:Abbott: Consultancy, Research Funding.

Rictor Overexpression and mTORC2 Signaling in Chronic Lymphocytic Leukemia

AbstractAbstract 3884Signaling via the B-cell receptor (BCR) stimulates growth and survival of CLL leukemic cells and inhibits apoptosis by phosphorylating immunoreceptor tyrosine based activation motifs. This signaling subsequently activates PI3 Kinase/AKT, mTOR, ERK and other pathways. Activation of Akt in turn requires phosphorylation by mTOR kinase, which assembles in two complexes mTORC1 and mTORC2 and it is the mTORC2 complex that phosphorylates and activates Akt. This phosphorylation of Akt in CLL specimen's upregulates anti-apoptotic proteins such as Mcl-1, Bcl-xl and XIAP. We have identified that Rictor, a component of mTORC2 complex is over-expressed in CLL specimens as compared to normal peripheral mononuclear B cells. This over-expression was noted by real time PCR that showed 1.5 to 4 fold upregulation (n=12). Western blot analysis also showed Rictor overexpression in all the twelve CLL specimens tested. Rictor overexpression was also seen in Mantle cell lymphoma cell lines and to study its role in BCR signaling, stable Mantle cell lymphoma lines with SiRNA mediated Rictor knockdown were established. Rictor knockdown resulted in a significant decrease in Akt activation as phosphorylation (phospho S473) of Akt both in unstimulated cells and when the cells were stimulated with BCR crosslinking was decreased.To determine the effect of Rictor and mTORC2 inhibition on CLL specimens, we tested the activity of three compounds isolated via yeast two hybrid drug screen designed to identify molecules that inhibit Rictor/mTOR interaction. When tested on CLL specimens in the presence of BCR crosslinking, these mTORC2 inhibitor compounds inhibited the downstream phosphorylation of Akt S473. Functionally the inhibitors also induced apoptosis in CLL cells with 40–60% of CLL cells undergoing apoptosis (1.0mM, cells treated for 48 hours). In comparison, rapamycin (mTORC1 inhibitor) and ppp242 (mTORC1, 2 inhibitor) were comparatively less active in CLL specimens as they were less effective in the induction of apoptosis. Western blot analysis of mTORC2 inhibitor treated cells also showed PARP cleavage and an increase in the pro-apoptotic protein BAD. Our data indicates that Rictor overexpression in CLL specimens is required for Akt phosphorylation activation and downstream BCR signaling. Inhibition of this pathway by mTORC2 inhibitors in CLL will be an effective therapeutic strategy. Disclosures:No relevant conflicts of interest to declare.

Phase I Study of Bortezomib and Romidepsin in Patients with Chronic Lymphocytic Leukemia/Small Lymphocytic Lymphoma, Indolent B-Cell Lymphoma, or Peripheral T-Cell Lymphoma

Abstract 1794Preclinical studies have demonstrated synergistic interactions between proteasome and histone deacetylase (HDAC) inhibitors in diverse hematologic malignancies, including those of B-cell origin. We have previously shown that the combination of the proteasome inhibitor bortezomib and the HDAC inhibitor romidepsin, administered at extremely low concentrations (∼3–5 nM), results in a striking increase in apoptosis in primary CLL cells, including cells from patients with CLL refractory to standard treatments (Dai Y et al. Clin Cancer Res. 2008). These findings prompted initiation of a phase I trial, using a 3+3 design, with the primary objective of determining the maximum tolerated dose (MTD) for the combination of bortezomib and romidepsin in patients with relapsed or refractory chronic lymphocyte leukemia/small lymphocytic lymphoma (CLL/SLL), indolent B-cell lymphoma, or peripheral T-cell lymphoma (PTCL).To date, 11 patients have been enrolled and treated. All of the patients treated have been diagnosed with CLL/SLL (n=11). The male:female ratio was n = 10 (91%):1 (9%); the median age was 56 (range 46–66) years; ECOG performance scores ranged from 0 to 1; and the median number of prior therapies was 4 (range 2–6). The schedule of administration was intravenous bolus bortezomib followed by a 4-hour intravenous infusion of romidepsin on days 1, 8, and 15; on a 28-day cycle. Dose level 1 = bortezomib 1.3 mg/m2, romidepsin 8 mg/m2 (n=3); level 2A = bortezomib 1.3 mg/m2, romidepsin 10 mg/m2 (n=5); and level 2B = bortezomib 1.6 mg/m2, romidepsin 8 mg/m2(n=3). The study is currently enrolling to dose levels 2A and 2B.Adverse events were determined using CTCAE version 4. Dose limiting toxicities (DLTs) were determined per protocol and occur in cycle 1 only. Two DLTs have been observed: 1 at dose level 2A (grade 3 fatigue) and 1 at dose level 2B (grade 3 chills, associated with cytokine release syndrome). Both dose levels will be expanded to 6 DLT-evaluable patients. Non-DLT grade 3 or greater adverse events have included anemia (grade 3, 9%), fatigue (grade 3, 9%), leukopenia (grade 3, 9%), nausea (grade 3, 9%), neutropenia (grade 4, 18%), soft tissue infection (grade 3, 9%), and vomiting (grade 3, 9%). Serious adverse events that were deemed at least possibly related to treatment were grade 3 soft tissue infection with grade 4 neutropenia (1 patient); grade 2 fever/nausea/vomiting which precipitated hospitalization (1 patient); and grade 2 cytokine release syndrome with associated grade 3 shaking chills, and grade 2 nausea/vomiting/diarrhea/oral dysethesia/sweats/photophobia/macular rash (1 patient).All 11 patients treated were evaluable for response. One partial response has been observed to date; the patient (positive for ZAP70; 4 lines of prior chemotherapy) proceeded, after 4 cycles of romidepsin and bortezomib, to allogeneic stem cell transplant. Five patients had a best response of stable disease and 5 had progressive disease. Correlative studies examining pre- and post-treatment expression of NF-κB (nuclear RelA and p52 as a marker of p100 processing), the NF-κB-dependent proteins XIAP and Bcl-xL, and Bim are currently underway.Collectively, these findings indicate that the safety profile is consistent with those reported for bortezomib and romidepsin, with reversible grade 2 to 4 adverse events, in heavily pretreated patients with relapsed/refractory CLL/SLL. The MTD has not yet been reached. Pending identification of the MTD, phase II evaluation of this therapeutic strategy, if warranted, should determine its activity more definitively in this population. Disclosures:Shea:Millennium Pharmaceuticals: Research Funding.

Effect of auranofin on oxidative and endoplasmic reticulum stress as well as anti-CLL activity with proteasome inhibitor.

e13568 Background: The gold-containing compound auranofin (AF) (Ridaura), is a treatment of rheumatoid arthritis. Recently, AF was shown to inhibit thioredoxin reductase (TRR), increase reactive oxygen species (ROS) and induce apoptosis in cancer cells. Methods: We determined the ability of AF (250 to 1000 nM) to induce oxidative, proteotoxic and lethal endoplasmic reticulum (ER) stress in CD19+ primary CLL cells, including those with deletion of chromosome 17p. Results: Treatment with AF induced 20 to 40% increase in ROS levels, decreased TRR activity (mean of 45%), but not its protein expression, in CLL cells. AF-mediated oxidative stress induced NRF2 activity with increase in hemeoxygenase-1 (HO-1) and glutamate cysteine ligase modifier (GCLM) levels. AF induced ER stress, associated with increase in GRP78 and the pro-apoptotic transcription factor CHOP protein levels. AF also induced the pro-apoptotic BH3-only domain protein BIM. Exposure to AF increased the intracellular levels of misfolded polyubiquitylated proteins. This disrupted the binding of heat shock protein (hsp) 90 with histone deacetylase 6 (HDAC6), resulting in hyperacetylation and inhibition of the chaperone function of hsp90. This led to proteasomal degradation of CLL-relevant hsp90 client proteins, including ZAP70, c-Raf and AKT. Nuclear STAT3 levels also declined. Exposure to AF induced significantly more apoptosis (range 40 to 60%) in primary CLL cells, as compared to CD19+ normal B cells and CD34+ human cord blood and bone marrow progenitor cells (< 15% apoptosis) (p < 0.01). AF treatment also induced apoptosis in CD19+ cells from patients with poor prognosis CLL with deletion of 17p or with deletion of 13q. Co-treatment with AF and the proteasome inhibitor carfilzomib or the GCLM antagonist buthionine sulfoximine (BSO) synergistically induced apoptosis in primary CLL cells. A FDA IND-supported phase II clinical trial of AF (6 to 9 mg PO/day) with correlative biomarker analysis (as above) is underway and three patients are enrolled. Conclusions: These findings support the full evaluation of clinical activity and predictive biomarkers of response to the re-purposed AF in patients with CLL.

Pro-Apoptotic Activity of Honokiol Analogues in B-Cell Lymphoid Malignancies

Abstract 1663 Backround:Honokiol, a small-molecule polyphenol isolated from Magnolia officinalis, was recently found to have antitumor properties. The agent showed cytotoxicity against CLL cells, but so far there are no data on its activity in other lymphoid tumors. More recently, new honokiol analogues (HAs) were synthetized, however, their cytotoxicity has not been investigated in leukemia/lymphoma yet. The aim: This is the first study assessing pro-apoptotic activity of new HAs on tumor cells from B-cell lymphoid malignancies. Material and Methods: Seven HAs: BR MMK + phloroG-3,3’ dibromo4,4’ bis (dimethlyamino) benzylidenephloroglucinol (HA 1), dibromoimipramine blue (HA 2), 3 bromo-4 dimethylamino diuhydroxyphenoxane (HA 3), bromo Gentian Violet-tribromogentian violet (HA 4), Gentian Violet (HA 5), BR dimethylaminobenzaldehyde + phloroG 3 bromo-4 dimethlyaminobenzylidenephloroglucinol (HA 6) and hexafluoro-diallylhexaflurobisphenol (HA 7) were studied. We assed ex vivo CLL cells (21 pts) and, in in vitro settings, pre-B-cell acute lymphoblastic leukemia (NALM-6), Burkitt lymphoma (Raji), diffuse large B-cell lymphoma; DLBCL (Toledo) and multiple myeloma (MM)-derived (RPMI 8226) cell lines. Lymphocytes obtained from 10 healthy volunteers were also treated. HAs were tested in concentrations 1–10 mM (24 and 48 hr incubation), then the minimal doses that triggered significant apoptosis and 48 hr time point were chosen for further experiments. Overall cytotoxicity of HAs was estimated using propydium iodide (PI) flow cytometry assay. The half maximal inhibitory concentration (IC50) of the study-drugs was estimated. Drug-induced apoptosis was assessed by the Annexin-V assay. Compensated apoptotic index (CAI) was calculated as a difference in percent of Annexin-V-positive cells between the drug-treated sample and the parallel untreated culture. Activation of caspases-8, -9 and -3 as well as expression of several apoptosis–regulating proteins, including Bcl-2 family (Bax, Bak, Bcl-2, Mcl-1), and inhibitor of apoptosis protein (IAP) family (cIAP1, cIAP2, XIAP, Smac/DIABLO) and IAP antagonists (survivin, HTRA2/Omi) were also investigated. Results: HA 1 triggered significant apoptosis in CLL cells starting from the dose 5mM, with minimal significant CAI (msCAI) 12.5%; vs. control - p=0.043 (IC50 10mM). In Raji cells msCAI was 17% at the dose 0.5mM; p=0.025 (IC50 2.5mM). In Toledo msCAI was 28% (at 0.1mM); p=0.007 (IC50 0.5mM) and RPMI 8226 (msCAI 58% at 0.5mM; p<0.001, IC50 0.5mM) cells. HA 2 induced both msCAI and IC50 at 2.5mM (msCAI 51.9%; p<0.001) In Raji cells msCAI was 22.1% (at 1mM); p=0.025, with IC50 2.5mM. The highest anti-tumor effect of HA 2 was found for Toledo (msCAI 52% at 0.5mM; p<0.001, IC50 0.5mM) and RPMI 8226 (msCAI 45.7% at 0.1mM; p<0.001, IC50 0.5mM) cell lines. HA 4 in CLL cells msCAI induced 15,8% (0.1mM); p=0.0270, IC50 10mM). In Raji cells msCAI was 27.4% (at 2.5mM); p=0.015 (IC50 5mM). Toledo and RPMI 8226 showed msCAI and IC50 at the same dose 2.5mM; msCAIs were 65.2% and 67.5%, respectively; p<0.0001. HA 5 in CLL cells triggered msCAI 23% (p=0.012) at 2.5mM (IC50 0.5mM). In Raji model msCAI was 78.5 % at 0.1mM; p<0.001 (IC50 0.1mM). In Toledo msCAI was 65% (at 0.5mM); p<0.001 (IC50 0.5mM). Similarly, in RPMI 8226 cells msCAI was 59.3% (at 0.5mM); p<0.001 (IC50 0.5mM). Interestingly, HAs did not trigger apoptosis of healthy lymphocytes at doses inducing msCAI. Their mechanism of action of HAs 1,2 4 and 5 was caspase-dependent apoptosis, triggered through both TNF receptor and mitochondrial pathways. In CLL cells their significantly downregulated Bcl-2, Mcl-1, cIAP-1, XIAP and Smac/DIABLO proteins and upregulated survivin. In Raji cells, significant Bcl-2 downregulation, with upregulation of Bax and Bak proteins was found. In Toledo cells decreased expression Bcl-2 and XIAP and overexpression of Bax and survivin were observed. In RPMI 8226 cells, HAs significantly downregulated Bcl-2 and XIAP and upregulated Bax, Bak and survivin proteins. In contrast, NALM-60 cells were resistant to all HAs. Moreover, HA 3, 6 and 7 did not significantly affect the examined malignant cells. Conclusions: These data indicate that HAs some are potent tumor-selective inducers of apoptosis in ex vivo CLL cells and in in vitro model of Burkitt lymphoma, DLBCL and MM. Those HAs should be examined for further clinical application either as a single agents or in combination with other anti-cancer drugs. Disclosures:No relevant conflicts of interest to declare.

Highly Specific Inhibitor for Syk Induces Chronic Lymphocytic Leukemia Cell Apoptosis,

Abstract 3875The antibodies expressed by CLL cells are highly selected, implicating a role for antigen in the pathogenesis of this disease. The B-cell receptor (BCR) for antigen may become engaged by microbial or self-antigens to induce activation of downstream adapter/signaling molecules. This, either alone or in concert with activation of other receptors, can induce CLL-cell expression of genes encoding proteins involved in cell survival or proliferation, e.g. c-MYC, potentially enhancing the tendency for disease progression. Our prior studies found that ZAP-70 in CLL cells enhanced BCR signaling, which might account for the relatively aggressive clinical behavior of patients who have CLL cells that express this protein. Although transduction of ZAP-70-negative CLL cells with vectors encoding ZAP-70 enhanced BCR signaling, we found this effect did not require an active ZAP-70 kinase domain. Instead ZAP-70 apparently acts as an adaptor protein that facilitates recruitment/activation of Syk, which in turn serves as the principal kinase responsible for initiating BCR signaling. As such, inhibitors of Syk could be useful therapeutically, as indicated by a clinical study demonstrating that one Syk inhibitor, fostamatinib disodium (R406/R788), has activity in the treatment of patients with this disease. However, at concentrations of fostamatinib required for clinical activity this drug also can inhibit several kinases other than Syk, including Lck, JAK, and Flt3, making it difficult to conclude that the activity of this drug solely was due to its capacity to inhibit Syk. Coffey and colleagues developed a series of highly potent and selective small molecule Syk inhibitors. One of these inhibitors, PRT060318, is a highly specific inhibitor of Syk that is more selective than fostamatinib. We examined whether this specific Syk inhibitor could affect BCR signaling and CLL-cell survival. We evaluated the effect of PRT060318 on BCR signaling by calcium influx induced through surface IgM ligation. We found that PRT060318 inhibited IgM induced BCR signaling with an IC50∼200nM. CLL cells that express ZAP-70 (n=13) or that lacked expression of this kinase (n=10) were treated with 5 μM PRT060318 and examined for cell viability 24 hours later. The average viability of CLL cells that expressed ZAP-70 was significantly lower (70.4 ± 16.1) than that of CLL cells that lacked expression of ZAP-70 (93.6 ± 9.0) (P<0.01), revealing that PRT060318 had greater activity in inducing apoptosis of CLL cells that had high-level expression of ZAP-70. Moreover, Syk inhibition abrogated the capacity of surface IgM ligation to enhance CLL cell survival, the viability of CLL cells treated with PRT060318 and anti-μ (38.8 ± 5.4) was significantly lower than CLL cells with anti-μ stimulation after 48 hours (64.0 ± 6.0) (P<0.01, n=3). PRT060318 also inhibited the capacity of marrow stromal cells to support the survival of CLL cells in vitro (35.3 ± 5.2 vs. 61.8 ± 4.7) (P<0.01, n=4). These results indicate that selective Syk inhibition by PRT060318 can induce apoptosis in CLL cells, has greater activity against CLL cells that express ZAP-70 and might be most active in treating patients with aggressive disease characteristics. Disclosures:Coffey:Portola Pharmaceuticals Inc.: Employment. Carson:Wintherix: Equity Ownership. Kipps:Igenica: Equity Ownership, Membership on an entity's Board of Directors or advisory committees; Celgene: Consultancy, Research Funding; Abbot Industries: Research Funding; Pharmacyclics: Membership on an entity's Board of Directors or advisory committees; Genentech: Research Funding; GSK: Research Funding; Gilead Sciences: Research Funding; Amgen: Research Funding.

The BH3-Only Protein Puma Plays An Essential Role in p53-Mediated Apoptosis of Chronic Lymphocytic Leukemia Cells

Abstract 4592 ObjectiveChronic lymphocytic leukemia (CLL) is a heterogeneous malignant hematologic disease; the response to chemotherapy partly determines the prognosis of patients. Fludarabine, which is frequently used in the treatment of CLL, induces CLL cell apoptosis by activating p53 and then down stream BH3-only family members, consequent to the induction of DNA damage. To identify which is the key factor in this apoptosis pathway, we incubated CLL cells with an active metabolite of fludarabine, 9-β-D-arabinosyl-2-fluoroadenine (F-ara-A), in vitro and then analyzed the role of Puma, Noxa and Bim in p53-mediated apoptosis of CLL cells. MethodsCLL cells from 15 CLL patients were incubated in medium with 3.5 mM fludarabine. Isometric cells were incubated with medium only, as blank control. CLL cells were harvested after 48-hour incubation. The apoptosis of cells was measured by staining with annexin V-FITC/propidium iodide (PI). The mRNA levels of Puma, Noxa, Bim and Bcl-2 genes were quantified using real-time quantitative reverse transcription-polymerase chain reaction (qRT-PCR) with SYBR Green by ABI LightCycler. Fluorescence in situ hybridization (FISH) analysis was used to detect ATM and p53 gene deletions. CLL cells were evaluated for immunoglobulin heavy chain variable region (IgVH) gene and p53 gene mutational status by PCR and sequencing. The function of p53 gene was detected by flow cytometry. Results1. The percentage of apoptosis at the beginning of incubation was 7.41% (4.11%-10.71%), and the percentages of spontaneous and fludarabine-induced apoptosis after 48 h incubation were 31.40% (18.14%-44.66%) and 59.6% (43.72%-75.48%), respectively. When CLL cells were cultured in vitro, the apoptosis was raised obviously (P=0.0002). Mean fludarabine-induced apoptosis was higher than the mean spontaneous apoptosis of CLL cells (P=0.0019).2. After 48 h incubation without fludarabine, the expression level of Puma was significantly higher in CLL cells (P=0.0317), but the expression levels Noxa (P=0.5404), Bim (P=0.3067) and Bcl-2 (P=0.1747) were not changed. Treatment of CLL cells with fludarabine further enhanced Puma expression in all samples compared to cells in medium only (P=0.0359), but the expression levels of Noxa (P=0.8063), Bim (P=0.5703) and Bcl-2 (P=0.5322) were not changed.3. In this analysis, 4 patients carried p53 abnormality and 3 patients carried heterozygous ATM gene deletion. The level of p21 in cytoplasm was not raised in CLL cells from patients with p53 abnormality after incubated with fludarabine. Elevation of p21 was detected in CLL cells carried heterozygous ATM gene deletion, after fludarabine induction.4. It was appeared that apoptosis level (P=0.1704) and Puma expression level (P=0.5334) of CLL cells cultured in medium only were indepented of p53 status, but apoptosis level (P=0.0109) and Puma expression level (P=0.0430) were up-regulated higher by fludarabine in CLL cells with functional p53 than in ones with dysfunctional p53. Fludarabine-induced apoptosis level (P=0.228) and Puma expression level (P=0.164) in CLL cells with heterozygous ATM gene deletion were similar to p53 wild type ones.5. Fludarabine-induced apoptosis and the expression level of Puma seemed to be higher in CLL cells with mutated IgVH than those with ummutated IgVH, but the difference was not significant. ConclusionFunctional p53 was necessary in fludarabine-inducd apoptosis. CLL cells with p53 gene deletion and/or p53 mutation showed dysfunction of p53. CLL cells with heterozygous ATM gene deletion carried functional p53. Fludarabine induced CLL cells apoptosis via activating p53 and then up-regulating its downstream effector Puma. Only Puma was the essential factor for p53-mediated apoptosis, and Noxa and Bim seemed to play a more restricted role. Disclosures:No relevant conflicts of interest to declare.

The p90RSK Inhibitor BI-D1870 Induces Apoptosis in CLL Cells

Abstract 2492Chronic Lymphocytic leukaemia (CLL) is the most common leukemia in the western world but is currently incurable using conventional therapies. Therefore novel agents which can kill CLL cells either alone or in combination with conventional agents are being continually sought. Activation of the MAPK pathway has been implicated as one of the protective signaling pathway in CLL and is characterised by expression of phosphorylated ERK. In CLL cells ERK can be activated by phosphorylation after BCR activation with αIgM. However in a proportion of CLL isolates ERK is constitutively active. MEK inhibitors (MEKi) are currently being used in clinical trials to treat relapsed or refractory leukemia including CLL (www.clinicaltrials.gov, NCT00920140). Therefore we hypothesised that inhibiting the ribosomal S6 kinase p90RSK which is immediately downstream of the MAPK pathway, may be of therapeutic value for the treatment of CLL. Since p90RSK regulates the majority of the anti-apoptotic signaling of the ERK pathway, we investigated the effect of inhibiting p90RSK with its specific inhibitor BI-D1870 (BI-D) on apoptosis in CLL.Immunoblotting was carried out to investigate phosphorylation of components of the ERK/p90RSK pathway using phosphorylation state-specific and pan-reactive antibodies. BI-D toxicity was evaluated by Annexin V/PI staining, MTT assay and immunoblotting for cleavage of the caspase 3 substrate poly(ADP ribose) polymerase (PARP) from its 116KDa to its 85KDa form.BI-D treatment of CLL cells decreased S6 phosphorylation, and this down regulation correlated with apoptosis. BI-D killed CLL cells rapidly, with a mean IC50 of 12μM following 18h incubation. Treatment of CLL cells with BI-D induced cytochrome C release indicating inhibition of p90RSK induced apoptosis by the intrinsic death pathway. However its toxicity was not due to inhibition of protein translation since inhibition of mTOR, a parallel pathway which also regulates S6 phosphorylation, was minimally toxic to CLL cells at concentrations as high as 500nM. However apoptosis did coincide with a reduction of the antiapoptotic protein Mcl-1. We further demonstrated that BI-D augmented apoptosis induction by cytotoxic drugs, as shown by PARP cleavage and by annexin V/PI binding assays. Analysis of cytotoxicity data using CalcuSyn software provided evidence for synergistic killing of the majority of CLL isolates by combinations of BI-D with either chlorambucil, fludarabine or nutlin 3.Constitutive or antigen induced activation of p90RSK via the MAPK pathway in CLL cells may contribute to apoptosis resistance and its inhibition may be of value in devising novel therapeutic strategies, either alone or in conjunction with conventional cytotoxic drugs. p90RSK inhibition led to apoptosis independently of protein translation inhibition and may be attributable in part to downregulation of Mcl-1. Given the recent interest in using MEK inhibitors (MEKi) for the treatment of CLL, these data suggests MEKi may induce apoptosis in CLL in part by inhibiting pathways downstream of p90RSK. In addition BI-D is a highly specific inhibitor of p90RSK and therefore may negate the off target effects shown with some MEKi. This is the first report on the consequences of inhibition of p90RSK in CLL cells and suggests that inhibiting p90RSK may be of considerable interest for the treatment of CLL. Disclosures:No relevant conflicts of interest to declare.

BH3 Profiling Demonstrates That Restoration of Apoptotic Priming Contributes to Increased Sensitivity to PI3K Inhibition in Stroma-Exposed Chronic Lymphocytic Leukemia Cells

Abstract 974 We and others have previously found that CLL cells exposed to stroma are resistant to a wide variety of treatments. While this phenomenon is thought largely to be due to inhibition of apoptosis, this has not been directly tested. BH3 profiling was developed in our laboratory as a functional assay to assess the degree to which malignant cells are primed to undergo apoptosis via the mitochondrial pathway. BH3 profiling demonstrates that decreased priming for apoptosis is indeed associated with treatment resistance in CLL cells co-cultured with mouse fibroblast L cells. Here, we confirm these preliminary results, showing that CLL cells co-cultured with either primary human stroma or a human stromal cell line also exhibit decreased priming for apoptosis. We demonstrate that the delta-isoform specific PI3K inhibitor CAL-101 (GS1101) effectively kills stroma-exposed CLL cells and sensitizes them to treatment with other agents by releasing the cells from stroma and/or increasing their level of mitochondrial apoptotic priming. CLL cells were co-cultured in the presence or absence of drugs with either the human stromal cell line StromaNKTert, with nurse-like cells (NLC) grown from primary CLL patient samples, or with no stroma for 24 hours, as previously described (Burger et al., Blood, 2000). A calcein-based adhesion assay was used to assess CLL cell adhesion to stroma in the presence or absence of CAL-101 (GS1101). CLL cell apoptosis in response to various drug treatments was assessed by Annexin/PI analysis. BH3 profiling was performed by exposing gently permeabilized malignant CD19+ B cells from CLL patients to a panel of BH3-domain peptides, and mitochondrial apoptotic priming was quantified by JC-1-based FACS to assess mitchondrial outer membrane permeabilization, as previously described (Ryan et al., PNAS 2010). Stroma-exposed CLL cells were highly resistant to the purine analog fludarabine and the BH3-mimetic ABT-263. CAL-101 (GS1101) was able to partially overcome this resistance, and the combination of CAL-101 (GS1101) and fludarabine led to a marked increase in sensitivity of stroma-exposed CLL cells. To evaluate the underlying mechanism of increased CLL cell sensitivity in the presence of CAL-101 (GS1101), we labelled CLL cells with calcein, and co-cultured them with the StromaNKTert cell line. In short term culture (1 hour), CLL cells were highly adherent to stroma, and were significantly less adherent in the presence of CAL-101 (GS1101). After longer co-culture (24 hours), calcein-labelled CLL cells were released from stroma within one hour of the addition of CAL-101 (GS1101). To evaluate whether the release of CLL cells from stroma affected their apoptotic priming, we performed BH3 profiling on CLL cells co-cultured in the presence or absence of stroma. CLL cells cultured without stroma were significantly more primed for apoptosis than CLL cells co-cultured with either the StromaNKTert cell line or primary human NLC. The addition of CAL-101 (GS1101), with or without fludarabine, to CLL cells in culture with stroma, led to increased apoptotic priming compared to CLL cells in culture that were untreated or treated with fludarabine alone. Similar alterations in apoptotic priming were observed when CLL cells were co-cultured with CAL-101 (GS1101) with or without ABT-263. Overall, these results suggest that the sensitivity of stroma-exposed CLL cells to CAL-101 (GS1101) may be related to the increased mitochondrial apoptotic priming that is facilitated by release of these cells from the stroma. These data may partially explain the clinical observation that, unlike traditional chemotherapy agents, CAL-101 (GS1101) has been efficacious in CLL despite causing an increase in the white blood cell count. Disclosures: Lannutti: Gilead Sciences: Employment. Letai:Eutropics Pharmaceuticals: Consultancy, Membership on an entity's Board of Directors or advisory committees.

CRM1/XPO1 Represents a Promising Therapeutic Target for Treatment of Chronic Lymphocytic Leukemia

Abstract 232Exportin 1 (CRM1, XPO1) is a nuclear exporter that promotes the transit of tumor suppressor proteins (TSPs) including p53, I-κB, and FOXO3A out of the nucleus, thereby preventing their activity and contributing to disrupted apoptosis and enhanced proliferation. Recently, whole-genome sequencing in patients with CLL allowed the identification of recurrent mutations in a highly conserved region of CRM1 that can potentially affects its gene function, suggesting a direct role for CRM1 in the pathogenesis of CLL (Puente XS, et al: Nature 75:101, 2011). However the role of CRM1 and the consequences of its mutation in the development of CLL have yet to be explored. CRM1 has been shown to be up-regulated in hematologic and various solid tumors, making it a highly attractive molecular target impacting multiple pro apoptotic pathways. KPT-SINEs are new, potent and irreversible small molecule selective inhibitors of nuclear export developed by Karyopharm that specifically and irreversibly bind to CRM1 and block the function of this protein. CLL is characterized by disrupted apoptosis caused both by co-dependent stromal elements and aberrant activation of several survival-promoting signaling/transcriptional pathways including PI3K/Akt, NF-kB, and p53. Because of the distinct subtypes of CLL and multiple signaling pathways dysregulated, a therapeutic agent targeting a single biological pathway is unlikely to be effective. Thus, pursuit of CRM1 inhibition as a novel strategy aimed to restore multiple death pathways is crucial and has broad implications for many types of patients. Our preliminary work demonstrated CRM1 is over-expressed in CLL cells compared to normal B cells at a protein (3 fold, p<0.005) and mRNA level (2.6 fold p=0.014). Inhibition of CRM1 by KPT-185 induced apoptosis in primary patient CLL cells in a dose and time dependent manner (EC50<500nM) while limited cytotoxicity against normal PBMC and isolated B, NK and T cells was observed (EC50 values >20 μM). Additionally, KPT-185 treatment of NK cells had no effect on their function as measured by ability of NK cells to mediate antibody dependent (ADCC) as wekk as direct cytotoxicity. The effect of KPT-185 on T function is currently under evaluation. Nuclear accumulation of FOXO3, p53 and IkB was also observed in primary CLL cells in a time dependent manner as shown by western blot and confocal microscopy. The evaluation of activated target genes is currently ongoing. Given the importance of microenvironmental stimuli on survival of CLL cells and response to therapy, we evaluated the ability of KPT-185 to induce cytotoxicity of CLL cells in the presence or absence of soluble factors such as CPG, CD40L, BAFF, TNF-α, IL-6, or IL-4, which are known to reduce the spontaneous apoptosis associated with CLL cells. KPT-185 treatment abrogated the protection induced by each of these factors suggesting that KPT-SINEs can disrupt signaling from the microenvironment that lead to in vivo CLL cell survival and potentially drug resistance. Interestingly the cytotoxic effect elicited by KPT-185 was enhanced in CPG activated cells (p=0.02). We also tested the ability of KPT-185 to kill CLL cells under coculture conditions with Hs5 stromal cell line. Coculture of CLL cells alone for 48 hours on the Hs5 stromal cell line resulted in a marked reduction of spontaneous apoptosis suggesting a strong protective effect elicited (P<0.001) by the stromal cells. Interestingly the cytotoxic effect mediated by KPT-185 was enhanced under coculture conditions (p=0.013). KPT-185 was also proven to be effective on murine TCL1+ cells (EC50<500nM) in vitro. The in vivo efficacy of this compound and other structurally related analogs is currently being assessed in an ongoing study in theTCL1 mouse model of CLL. In conclusion CRM1 represents a novel target that has not been adequately explored in CLL. KPT-SINEs are a class of promising therapeutic agents with proven selective in vitro activity in CLL cells providing the rationale for developing small molecule, drug-like CRM1 inhibitors for the treatment of this disease. Disclosures:Sandanayaka:Karyopharm Therapeutics: Employment. Shechter:Karyopharm Therapeutics: Employment. McCauley:Karyopharm Therapeutics: Employment. Shacham:Karyopharm: Equity Ownership. Kauffman:Karyopharm: Equity Ownership.

Treatment with Auranofin Induces Oxidative and Lethal Endoplasmic Reticulum (ER) Stress Exerting Single Agent Activity Against Primary CLL Cells

Abstract 929Auranofin (AF) (Ridaura®) is an oral, FDA-approved, lipophilic, gold-containing compound used for the treatment of rheumatoid arthritis, based upon its anti-inflammatory and immunosuppressive effects. In the present studies, we determined for the first time the lethal activity of AF (250 to 1000 nM) and its underlying mechanism(s) in CD19+ primary CLL cells, including those with del(17p13.1). We demonstrate that treatment with AF induced 20 to 40% increase in reactive oxygen species (ROS) levels (assessed by flow cytometry), decrease in thioredoxin reductase (TRR) activity (mean reduction of 45%) without alteration in the expression of TRR, as determined in the AF treated cell lysates of CLL cells. AF treatment also increased the levels of thioredoxin binding protein 2. These findings were associated with induction of nuclear factor erythroid 2-related factor 2 (NRF2) activity marked by substantial increase in hemeoxygenase-1 (HO-1) levels. Thus, AF treatment perturbed the redox status in CLL cells. Consistent with increased levels of intracellular ROS, AF treatment also induced the markers of ER stress response including the induction of GRP78 and a sustained increase in the levels of the pro-apoptotic transcription factor CHOP (CAAT/enhancer binding protein homologous protein). This was associated with induction of the levels of the pro-death BH3 only domain protein BIM, which is a known mediator of lethal ER stress. Exposure to 1000 nM AF induced significantly more apoptosis (range 40 to 60%) in primary CLL cells, as compared to CD19+ normal B cells and CD34+ human cord blood and bone marrow progenitor cells (< 15% apoptosis) (p < 0.01). AF treatment induced apoptosis to a similar extent in primary CLL cells with del(17p13.1) as compared to other genetic abnormalities. Based on the observations that AF treatment induced oxidative and ER stress, we determined whether exposure to AF also increased the intracellular levels of misfolded and polyubiquitylated proteins, which would disrupt the cytosolic complex consisting of the heat shock protein (hsp) 90, heat shock factor (hsf) 1, histone deacetylase 6 (HDAC6) and p97/VCP, resulting in hyperacetylation and inhibition of the chaperone function of hsp90. Indeed, treatment with AF disrupted the association between hsp90 and HDAC6, resulting in hyperacetylation of hsp90 and depletion of the levels of HDAC6 in CLL cells. AF mediated hyperacetylation of hsp90 and α-tubulin was associated with proteasomal degradation of several of the CLL-relevant hsp90 client proteins, including ZAP70, c-Raf and AKT. Co-treatment of CLL cells with AF and the proteasome inhibitor carfilzomib (20 nM) significantly restored the levels of ZAP70, c-Raf, AKT and HDAC6. Finally, relatively lower levels of AF (250 nM) combined with the pan-HDAC inhibitor panobinostat (20 nM) or carfilzomib (5 nM) induced more cell death of primary CLL cells than each agent alone. These findings indicate that treatment of primary CLL cells with AF results in oxidative and ER stress. AF treatment also promotes the depletion of pro-survival CLL-relevant hsp90 client proteins. Collectively, our data creates a strong rationale for determining the in vivo activity of AF in patients with CLL including those with del(17p13.1). Disclosures:Reyes:Millennium, Sanofi Aventis: Consultancy.

Obatoclax in Combination with Fludarabine and Rituximab (FR) Is Well-Tolerated and Shows Promising Clinical Activity in Relapsed CLL/SLL

Abstract 2865Obatoclax is a small molecule mimetic of the BH3 domain of Bcl-2 family proteins. Obatoclax is broadly specific, with activity against Bcl-2, Bcl-X, Bcl-w and Mcl-1. CLL cells overexpress Bcl-2, Bcl-XL and Mcl-1 in particular, and obatoclax can induce apoptosis of CLL cells in vitro. A phase 1 study of single-agent obatoclax in heavily pretreated largely refractory CLL patients demonstrated that the dose-limiting toxicity was neurologic, including euphoria and ataxia, leading to a maximum tolerated dose of 28 mg/m2 given over 3 hours every 3 weeks. One PR was observed along with biologic activity demonstrated by reductions in lymphocyte counts and improvements in cytopenias. We therefore undertook this phase 1 study of the combination of obatoclax with FR in CLL patients relapsed after at least one prior therapy and in need of therapy again. Obatoclax was given as a three hour infusion on days 1 and 3 at three dose levels, 10, 14, and 20 mg/m2 per dose. Fludarabine was given at 25 mg/m2 days 1–5, and rituximab 375 mg/m2 day 1 following an option to split the dose in cycle 1. Thirteen patients were enrolled, seven men and six women, with median age 58. 5 (38%) had stage 3–4 disease. FISH showed one patient with del17p, one with complex karyotype, and five with del11q. Six of nine patients evaluable had high risk unmutated IGHV, and nine of ten patients evaluable were positive for ZAP-70. The median number of prior therapies was two, with 9 patients having had prior fludarabine-based combination chemotherapy, 10 patients having had prior rituximab, and 8 patients having had prior alkylator-based combination chemotherapy. The study therapy was well-tolerated, with a median of five cycles administered. One dose-limiting toxicity (DLT) was observed at the 20 mg/m2 obatoclax dose; this DLT was a greater than two week treatment delay for persistent grade 2–3 neutropenia in a patient who had had a similar event previously with FR alone. This DLT led to expansion of the third and highest cohort, which enrolled seven patients with no further DLTs observed. Other grade 3–4 toxicities have been limited and include neutropenia (n=5), thrombocytopenia (n=2), fever without neutropenia (n=2), increased ALT/AST (n=1), and dizziness (n=1). Neurologic side effects were easily managed and resembled alcohol intoxication, including grade 1–2 euphoria (n=6), ataxia (n=5), dizziness (n=6), anxiety (n=4), speech impairment (n=4) and confusion (n=3). The ORR by NCI-WG criteria was 85% (11/13; 90% CI 59–97%) with 15% CR (2/13; 90% CI 3–41%) and 38% nPRs (5/13; 90% CI 17–65%). With the addition of CT scan measurement of lymphadenopathy the ORR declined to 54% (7/13; 90% CI 29–78%) with no CRs. With a median follow-up of 26 months from the start of the study, three patients are in ongoing remission, six have relapsed with three receiving further therapy to date, two patients have gone on to stem cell transplantation, and two patients have died of disease. The median time to progression is 20 months (95% CI 9, 35 mos). We were able to demonstrate increased apoptosis compared to baseline in peripheral blood CLL lymphocytes during cycle 1 therapy in 9 of 13 patients, using Annexin V propidium iodide staining. We conclude that the FR-obatoclax regimen is well-tolerated and highly active in a relapsed CLL population. An extension of this study to increase the frequency of obatoclax dosing to days 1–3, and to change the chemotherapy backbone to FCR, is planned pending the availability of obatoclax. Disclosures:Brown:Calistoga: Consultancy, Research Funding; Celgene: Honoraria, Research Funding; Genzyme: Research Funding; GSK: Research Funding; Pharmacyclics: Consultancy.

2.47 p53 Homolog p63 Promotes Apoptosis and Chemosensitivity in Chronic Lymphocytic Leukemia B-Cells

in cell culture medium were measured by enzyme-linked immunosorbent assay (ELISA) and activation of STAT3 and NFB was indicated by the ratio of p-STAT3/STAT3 and p-NFB/NFB. Patients with high levels of p-STAT3/STAT3 and/or p-NFB/ NFB had higher levels of IL-6 production. The level of IL-6 production significantly correlated with STAT3 activation (R 0.856, p 0.001) and NFB activation (R 0.815, p 0.001). Inhibition of constitutive STAT3 and NFB activation by the STAT3 phosphorylation inhibitor, diphenylporphyrin, and the NFB inhibitor, caffeic acid phenethyl ester (CAPE), blocked IL-6 production by 40% on average (p 0.004, p 0.003), active STAT3 or NFB significant increase IL-6 secretion. Analyzing the level of autocrine IL-6 production with in vitro spontaneous apoptosis in CLL cells from 38 patients, we found that patients with higher levels of IL-6 production (more than median) had significantly less apoptosis compared with those below the median (23% vs. 42.2% in average, p 0.0001). Correlation between the levels of autocrine IL-6 production and cell viability in 38 patients (CLL cells were incubated for 48 hours in vitro) was significant (R 0.72, p 0.0001). High IL-6 production was also associated with a shorter average time to first treatment (0.5 vs. 3 years, p 0.0023). This study demonstrates that autocrine IL-6 production correlates with STAT3 and NFB activation and apoptosis resistance in CLL. Furthermore, higher IL-6 production is associated with biological markers of poor prognosis and earlier treatment.