ALBERT

Description: Background: The nuclear factor kappa B (NFkB) signaling pathway is constitutively activated and central to the pathogenesis of multiple myeloma (MM). We previously published the antitumor activity of the NFkB inhibitor parthenolide in MM. Dimethylaminoparthenolide (DMAPT) is a water-soluble analog of parthenolide. The improved bioavailability of DMAPT makes its appealing for clinical development. Objectives: To validate the effects of DMAPT both in vitro and in vivo MM models and to develop a rational strategy for a combination of DMAPT with other anti-MM agents. Methods: The in vitro cytotoxicity of DMAPT was tested in human MM cell lines and primary samples both alone and in combination with bortezomib and lenalidomide. The in vivo activity of DMAPT was tested in a model of NOD/Scid mice intravenously transplanted with GFP-tagged NCI-H929 MM cells. Results: Based on MTT assays on MM cell lines and primary MM cells, the IC50 for proliferation inhibition of DMAPT was similar to that of parthenolide (1–3 μM). When used in combination with bortezomib, we observed synergy (combinatorial index 〈 1.0) at higher doses of DMPT and bortezomib (DMAPT 2.5–5 μM and bortezomib 5–12.5 nM), while an antagonistic effect was observed at lower doses. Similarly, when combined with lenalidomide, synergy was observed at DMAPT doses 1.6–6.4 μM and lenalidomide doses 0.5–2 μM, while an antagonistic effect was observed at lower doses. The in vivo cytotoxicity of DMAPT was explored in a model of NOD/Scid mice intravenously transplanted with GFP-tagged NCI-H929 cells. At 3 weeks post-transplantation, all animals had measurable serum human kappa light chain levels, as measured by ELISA assays. Animals were randomized to receive either 50 mg/kg/day DMAPT (n=12), 100 mg/kg/day DMAPT (n=12) or vehicle (n=10) via oral gavage for a duration of 6 weeks or until criteria for euthanasia were met. Weight loss of over 15% and hind leg paralysis were criteria for euthanasia. Toxicities in both treatment groups were minimal. Weight loss correlated with serum human kappa light chain levels and, hence, represented disease progression. At the end of treatment, there were 2 animals alive in the control group, while 5 were alive in each of the treated groups. Using Kaplan-Meier curves and log-rank analysis, the mean overall survival (OS) was 56.3 days in the control cohort versus 61.0 days and 62.4 days in groups treated with 50 and 100 mg/kg/day DMAPT, respectively. There was no statistically significant difference in the mean OS of mice treated with the two doses of DMAPT (p=0.07). Prior to treatment initiation, the serum levels of human kappa light chains were similar in all groups (ANOVA test, p= 0.38). At 2 weeks and 4 weeks post treatment (Week 5 and week 7, respectively; See table), the levels of human kappa light chains were significantly lower in the DMAPT treated groups (ANOVA test p = 0.017 and 0.0061, respectively). Week 3 Week 5 Week 7 Control 10.46 ± 5.79 531.31 ± 238 6272.1 ± 3211 DMAPT 50 mg/kg/day 6.94 ± 4.48 264.03 ± 198 2928.8 ± 2131 DMAPT 100 mg/kg/day 7.99 ± 7.21 334.15 ± 198 3374.9 ± 1835 ANOVA (P value) 0.38 0.017 0.0061 Conclusions: DMAPT demonstrated excellent efficacy in preclinical models of MM and supports a rationale for its clinical development in MM patients with relapsed MM either alone or in combination with bortezomib or lenalidomide.

Description: Although monocytic zinc finger protein (MOZ/MYST3) maintains normal hematopoietic stem cells, its fusion to the coactivator CREB-binding protein (CBP/CREBBP) induces acute myeloid leukemia (AML). Since leukemic stem cells in AML often exhibit excessive signal-dependent activity of the transcription factor NFκB, we hypothesized that cooperation between NFκB and MOZ-CBP represents an alternative mechanism for enhancing NFκB transcriptional activity. In reporter assays, MOZ and CBP separately induce transcription from the NFκB-dependent interleukin-8 promoter; however, these two proteins together markedly activate its expression. Although MOZ has less potent transcriptional activity than MOZ-CBP, both proteins cooperate with steroid receptor coactivator-1 (SRC1/NCOA1) to activate transcription. Since cooperation between MOZ-CBP and SRC1 is strongly reminiscent of the interaction between MOZ-TIF2/SRC2 and CBP required to induce AML (Deguchi et al, 2003), these findings suggest that MYST family fusion proteins may assemble a conserved leukemogenic transcriptional complex composed of a MYST protein (MOZ, MORF), molecular integrator (CBP, p300), and nuclear receptor coactivator (SRC1, TIF2/SRC2). MOZ also induces multiple NFκB-dependent viral promoters. Importantly, MOZ associates in a protein complex with the DNA-binding p65/RELA subunit of NFκB in vivo and interacts directly with p65 in vitro. Whereas deletion of the leukemia-associated protein (LAP/PHD) or MYST domains decreases the transcriptional activity of MOZ, deletion of either the acidic domain or C-terminal domain completely abrogates its activity. Since the C-terminal domain is absent from MOZ-CBP, these results indicate that the potent transcriptional activity of MOZ-CBP represents a gain-of-function property derived from the retained portion of CBP. Collectively, these studies not only demonstrate that MOZ and MOZ-CBP cooperate with NFκB to enhance expression of NFκB-dependent promoters but also suggest that aberrant interaction between MOZ-CBP and NFκB may play an important role in the pathogenesis of certain acute myeloid leukemias.

Description: Abstract 1966 Poster Board I-989 The Effect of Targeting the Antiapoptotic Protein c-FLIP in Multiple Myeloma Background: Multiple myeloma (MM) is an incurable blood cancer. Treatments which target key proteins that play a role in drug resistance should improve treatment outcomes. The cellular FLICE-inhibitory protein (c-FLIP) is an antiapoptotic protein which confers resistance to death receptor-mediated apoptosis. Further, c-FLIP overexpression has been identified in various cancers, and in MM specifically, gene expression profiling has demonstrated that c-FLIP is overexpressed in a patient's MM cells compared to the normal plasma cells of an identical twin. However, the precise clinical significance of c-FLIP overexpression and its potential as a therapeutic target in MM have not been established. Objectives: To determine the potential of c-FLIP as a therapeutic target by exploring the in vitro and in vivo effects of c-FLIP knockdown in MM cells, using a doxycycline-inducible lentiviral vector containing c-FLIP shRNA. Results: Doxycycline treatment in vitro induced c-FLIP shRNA transcription and reduced c-FLIP levels more than 80% in H929 cells. Correspondingly, we observed a greater than 95 % reduction in the growth of H929 cells, compared to the cell line containing an empty vector control. In addition, immunoblots of MM cell lysates showed an accumulation of the cleaved products of caspases 8 and 3, suggesting that c-FLIP knockdown sufficiently induced caspase activation and apoptosis in MM cells. To explore the cytotoxicity of c-FLIP knockdown in MM cells in vivo, we used two different mouse models. In the first model, MM cells containing the doxycycline inducible, c-FLIP shRNA containing, lentiviral vector, were transplanted subcutaneously into NOD/Scid mice. Two weeks post transplantation, when accurate tumor measurements could be made, the mice were divided into two groups, of approximately equal tumor volumes, and one group was given doxycycline treated water. A significant reduction in tumor size was observed in doxycycline treated mice compared to the controls (Figure 1). In a second mouse model, NOD/Scid mice were transplanted with MM cells by tail vein injection and divided into two groups. The first group (“early dox”; n=10) was given doxycycline treated water 4 hours post transplantation, while the second group (n=20) was given regular water until they developed tumors which could be determined by measuring human IgA kappa levels in the mouse serum. Within this group, mice were randomized base on the IgA levels to receive doxycycline treatment (“late dox”; n=6) or regular water (n=6). ELISA measurements of human IgA kappa light chain from treated mice are shown in Figure 2. The “early dox” group had barely detectable IgA kappa levels while the “late dox” group had significantly lower levels than the mice that never received any doxycycline. Summary: An inducible c-FLIP shRNA system was created in order to observe the effects of c-FLIP knockdown on the viability of MM cells both in vitro and in vivo. A nearly complete c-FLIP knockdown could be generated in vitro which had a dramatic effect on the viability of the MM cells, and this translated to a significant reduction in tumor volumes when these cells were later transplanted into mice, both subcutaneously and intravenously, and then treated with doxycycline. This work provides evidence that targeting c-FLIP in MM, either directly or via gene regulation, holds significant promise. Disclosures: Abonour: Celgene: Honoraria; Millennium: Honoraria.

Description: ENMD-2076 is a novel, orally-active molecule that has been shown to have significant activity against Aurora A kinase as well as multiple receptor tyrosine kinases (RTK). We investigated the single agent activity of ENMD-2076 against MM cells in vitro and in vivo, and in combination with lenalidomide. ENMD-2076 free base showed significant cytotoxicity against MM cells with a mean LC50 of 3.84±0.86 μM at 48 hours in vitro. Cytotoxicity was associated with cleavage of caspase 3, 8, 9 and PARP, and loss of mitochondrial membrane potential as early as 6 hours. ENMD-2076 free base inhibited c-kit, FGFR-1, 3 and VEGFR1 and subsequently inhibition of downstream targets phosphorylated (p)-BAD, p-Foxo1a and p-GSK-3β was observed at 6 hours. NOD/SCID mice implanted with H929 human plasmacytoma xenografts and treated for 30 days with 50, 100, 200mg/kg/d ENMD-2076 showed a dose-dependent inhibition of tumor growth (Figure 1), with minimal toxicity as assessed by the stable weight of treated animals. Immunohistochemical staining of tumors from sacrificed animals showed significant reduction in Ki67 at all dose levels of treatment compared to control tumors. An increase in cleaved caspase-3 was observed on Western blot from the lysates of H929 tumors obtained from treated animals. ENMD-2076 free base also showed synergistic cytotoxic activity when combined with lenalidomide against H929, MM1.R and MM1.S cells as assessed by MTT assay and Annexin-V/PI staining. Using the Chou-Talalay method, the combination indices (CI) were 〈 1 for all three cell lines across a range of concentrations of ENMD-2076 free base (0.25–1.0 μM) plus lenalidomide (2.5–10 μM) indicating synergistic activity (CI=0.362 H929; CI=0.315 MM1.R; CI=0.415 MM1.S). Our results provide rationale for the investigation of ENMD-2076 alone and in combination with lenalidomide in patients with multiple myeloma. Figure Figure

Description: Parthenolide is a sesquiterpene lactone, a new class of antitumor agents which target nuclear factor kappa B (NFkB). Although the activation of NFkB has been implicated in the growth and drug resistance of multiple myeloma (MM) cells, the efficacy of parthenolide in MM is unknown. Here we show that parthenolide inhibits the NFkB DNA binding in MM cells in a time- and dose-dependent fashion. Parthenolide inhibits the proliferation of MM cell lines, as well as tumor cells from chemotherapy-resistant MM patients; however, at the effective doses it is not toxic to peripheral blood mononuclear cells or bone marrow stromal cells. Neither exogenous interleukin-6 (IL-6) nor insulin-like growth factor 1 (IGF-1) overcome the parthenolide-induced cytotoxicity. In addition, parthenolide inhibits the proliferation of MM cells adherent to bone marrow stromal cells. Parthenolide has a minimal effect on cell cycle progression but strongly induces apoptosis. Parthenolide activates caspases 8 and 3 more strongly than caspase 9, resulting in the cleavage of poly (ADP)-ribose polymerase (PARP) and XIAP. Z-VAD-FMK partially blunts the PARP cleavage but does not protect MM cells against apoptosis. Parthenolide induces p21 accumulation, while the levels of Bax, IAP2, cyclin D, Bcl-X and p53 or the phosphorylation status of JNK, AKT and ERK are unaffected. In addition, parthenolide augments dexamethasone-induced cytotoxicity. Our study therefore provides the rationale for the future clinical development of parthenolide, or its combination with dexamethasone, in MM therapy.