ALBERT

Description: Abstract 2757 MCL is typically characterized by an aggressive clinical course and inevitable development of refractory disease despite early intervention that often includes: immunotherapy (e.g., rituximab), multi-agent induction chemotherapy and consolidation with high dose chemotherapy and autologous stem cell transplant in first remission. Residual disease at the time of stem cell collection is an important cause for treatment failure. There is a need to evaluate more potent anti-CD20 mAbs capable to kill lymphoma cells with low CD20 surface levels. In Burkitt's lymphoma (BL) and diffuse large B-cell lymphoma (DLBCL) pre-clinical models we previously demonstrated that OFA was more potent than rituximab (RIT) in vitro and in vivo. In order to characterize the activity of OFA against MCL, we evaluated the activity of OFA against cytarabine (Ara-C)-sensitive (eg. Mino, Jeko-1, Rec-1, HBL-2, Granta and Z-138); –resistant MCL cell lines (eg. MinoAraCR, Jeko-1AraCR, Rec-1AraCR, HBL-2AraCR and GrantaAraCR); and primary tumor cells derived from MCL patients (n=2). Antibody-dependent cellular cytotoxicity (ADCC) and complement dependent cytotoxicity (CDC) were measured by standard 51Cr release assays in MCL exposed to OFA, RIT or isotype control. OFA vs. RIT direct anti-proliferative effects were measured in by alamar blue reduction assay. Apoptosis following in vitro exposure to OFA or RIT was detected by caspase 3/PARP cleavage. Patient tumor cells were isolated from biopsy specimens by negative selection using magnetic beads and incubated with OFA or RIT +/− human serum as a complement source. Cell viability was determined at 48 hours by CellTiterGlo assay. Surface CD20 and the complement inhibitory proteins (CIPs) (CD55 and CD59) density in MCL cell lines was determined by flow cytometry (Image stream) and compared to BL or DLBCL cell lines. For in vivo experiments 6–8 week-old SID mice were inoculated subcutaneously with 5×106 matrigel suspended Z-138 cells. Upon tumor engraftment, mice were assigned to RIT (10mg/kg), OFA (10mg/kg) or control groups. Tumor growth curves were calculated for each group. Mice were sacrificed if tumor size reached 〉2cm in any dimension. After 6 months, survival was analyzed by Kaplan-Meier analysis and compared by log-rank test. OFA induced significantly higher levels of CDC associated cell lysis compared to RIT in almost all MCL cell lines tested (10/11) (Mino: 53.2% vs 0.2%; MinoAraCR: 72.6% vs. 0.6%; Jeko-1: 33.4% vs. 9.8%; Jeko-1AraCR: 38.3% vs. 2.8%; REC-1: 17% vs 3%; Rec-1AraCR: 7.8% vs. 0.2%; HBL-2: 27.1% vs. 19.2%; HBL-2AraCR: 86.6% vs. 72.2%; GrantaAraCR: 17% vs 0.9%; Z-138: 56.4% vs. 0.65%; all p-values

Description: Background Burkitt Lymphoma (BL) is an aggressive non-Hodgkin B cell neoplasm primarily affecting children. While overall cure rates are high, prognosis for the 20% of patients with relapsed or refractory is dismal with a 16% survival rate at four years, indicating the need for novel therapeutic approaches. While epigenetic modulators have shown potential therapeutic benefit in other hematologic malignancies, their use in BL has been limited. Here, we evaluate whether histone deacetylase (HDAC) inhibitors could enhance cell death in chemotherapy resistant BL cells. Methods The BL cell lines Ramos, Raji, or previously generated/characterized immunochemotherapy resistant Raji BL cell lines (Raji 2R and Raji 4RH) (Barth et al., Br J Haematol, 2012) were treated for 48 hours with 4-hydroperoxycyclophosphamide (4-HC, the active metabolite of cyclophosphamide), cytarabine, prednisolone, HDAC inhibitors (suberoyanilide hydroxamic acid (SAHA, vorinostat), and romidepsin), or vehicle control. Cell viability was measured on a Via Cell, by Alamar Blue staining (Invitrogen), or by measuring Caspase-3 activation by flow cytometry. Levels of pro- (survivin, XIAP, MCL-1, livin, and BCL-XL) and anti- (Bax, Bak) apoptotic proteins were evaluated using standard western blot techniques. To evaluate the wiring of signaling networks in the basal state or in response to drug treatment, samples were fixed, permeabilized, and simultaneously stained for Caspase-3, phospho GSK-3 (pGSK3, the inactive form of this protein), active β-catenin, and survivin. Cells were processed on a BD FACSVerse and analyzed by gating on Caspase-3 negative (chemotherapy resistant) cells and monitoring the activity of the pGSK3, active β-catenin, and survivin using FlowJO (V9.6) software. Results Relative to the BL cell line Ramos, Raji BL cells were resistant to all doses of conventional chemotherapy (cyclophosphamide, ara-c, and prednisolone) tested. Because hyper-activation of the WNT/b-catenin network due to epigenetic modulation has been implicated in chemotherapy resistance in other B-lineage pediatric malignancies (Hogan et al., Blood 2011), we treated cells with chemotherapy or vehicle and used phosphoflow cytometry to evaluate the expression of pGSK-3, active β-catenin, and its downstream target survivin. Relative to vehicle control, we found that exposure to cytotoxic chemotherapy resulted in rewiring of the cellular networks with increased levels of pGSK3, β-catenin, and survivin. Western blot analyses of the immunochemotherapy resistant Raji 2R and Raji 4RH cells revealed similar upregulation of anti-apoptotic proteins including survivin, downregulation of pro-apoptotic proteins Bak and Bax, and increased expression of pGSK3 compared to Raji cells. To test whether alterations in these signaling axes might enhance the survival of chemorefractory BL cells, we treated Raji cells with the HDAC inhibitor SAHA and found induction of apoptosis at nanomolar doses. Sensitivity to SAHA correlated with down-regulation of active β-catenin and survivin in a dose-dependent manner. To extend these results, we also tested whether immunochemotherapy resistant Raji 2R and Raji 4RH cells were sensitive to HDAC inhibition and found similar robust induction of cell death upon treatment with a panel of HDAC inhibitors. Conclusions Active β-catenin and survivin are upregulated in chemotherapy resistant BL cells. HDAC inhibition results in their rapid down regulation and enhanced apoptotic cell death. Further evaluation of inhibitors of the WNT/ β-catenin pathway and HDAC inhibitors is warranted in chemotherapy resistant BL and could have a role in treating refractory or relapsed BL patients. Disclosures: No relevant conflicts of interest to declare.

Description: Introduction: Novel therapies are urgently needed in pediatric Burkitt lymphoma (pBL) where the survival for relapsed disease is less than 20%. Heat shock protein 90 (Hsp90) is molecular chaperone that protects proteins from proteolytic degradation including oncogenic signaling complexes. The clinical development of broad-spectrum Hsp90 inhibitors has previously been limited by suboptimal target inhibition and off-target toxicities. PU-H71 is a next-generation Hsp90 inhibitor that preferentially targets tumor enriched (te-Hsp90), the functionally distinct pool of Hsp90 present in tumor cells. PU-H71 is not toxic to normal B-cells and has demonstrated pre-clinical efficacy in diffuse large B-cell lymphoma, but has not been studied in Burkitt lymphoma. In the current study, we evaluated te-Hsp90 as a potential therapeutic target in pediatric Burkitt lymphoma. Methods and Results: To evaluate overall Hsp90 protein expression in primary pBL tumors we performed immunohistochemistry on a tissue microarray. Fifty-three of 59 cases (90%) demonstrated high levels of Hsp90 expression defined as 〉90% tumor cell positivity (Fig. 1A). To evaluate the sensitivity of pBL to te-Hsp90 inhibition we performed in-vitro viability assays with the ATP-based CellTiter-Glo¨ in a panel of pBL cell lines (Ramos, DG-75, Raji, Namalwa, Daudi, Jiyoye, CA-46, Raji 2R, Raji 4RH) and in-vivo studies with a Ramos xenograft model of pBL. pBL cells were sensitive to inhibition by PU-H71 with IC50s in the low nanomolar range (151-337nM, Fig 1B). The Raji 2R and Raji 4RH pBL cell lines with acquired resistance to chemotherapy (Czuczman et al, Clin Cancer Res 2008) were also sensitive to PU-H71 (IC50 175-181nM). In contrast, normal peripheral blood lymphocytes were resistant (IC50 〉37,000nM). In pBL xenograft studies, PU-H71 decreased tumor volume (p

Description: Treatment with multi-agent chemotherapy regimens has significantly improved survival in pediatric Burkitt Lymphoma (BL) leading to long-term survival in over 80% of cases. The incorporation of rituximab in the treatment of pediatric B-cell non-Hodgkin lymphoma (B-NHL) has been slower than in adults, but may improve clinical outcomes in high risk pediatric BL patients when combined with chemotherapy regimens. On the other hand, BL patients with primary refractory or relapsed disease have a dismal prognosis, stressing the need to identify the mechanism(s) resulting in chemotherapy/rituximab resistance and to develop novel therapeutic approaches. To this end, we exposed a BL cell line (Raji) to escalating doses of rituximab with or without human serum and generated/characterized several BL rituximab-resistant (Raji 7R and Raji 8RH) (RRCL) or rituximab-chemotherapy resistant (Raji 2R and Raji 4RH) (RCRCL) cell lines. Subsequently, we screened for aberrant activation of signal transduction pathways between RSCL (Raji), RRCL (Raji 7R and Raji 8RH), or RCRCL (Raji 2R and Raji 4RH) in an attempt to define what pathways were associated with resistance to both rituximab and chemotherapy agents. This was accomplished by the analysis of phosphorylation patterns on key-regulatory members of pre-defined signal transduction pathways using Western blotting, phospho-flow cytometry studies and phosphoproteomics. Effects on cell viability or cell cycle distribution of RRCL or RCRCL following pharmacological inhibition of key-regulatory pathways identified was then performed using the alamar blue reduction assay or flow cytometry respectively. While total Akt (Protein kinase B) expression was similar between all the types of BL cell lines tested, we found that RCRCL (Raji 2R and Raji 4RH) had an increase in basal phosphorylation levels of Akt at the Ser473 and Thr308 phosphorylation sites when compared to RSCL (Raji cells) or RRCL (Raji 7R and Raji 8RH). These findings were confirmed by phospho-flow cytometry studies. Phosphoproteomic analysis comparing Raji (RSCL) against Raji 4RH (RCRCL) cells identified an increase of at least 2 fold in the phosphorylation of 315 proteins in RCRCL including several direct targets of AKT such as GSK3B, WEE1, FOXO1 and PRAS40. Altered phosphorylation of multiple Akt/mTOR downstream proteins (BAD, 4EBP1, GSK3B and ERK) was detected by western blot in RCRCLs (Raji 2R and Raji 4RH) compared to RSCL and RRCLs. In vitro exposure of RCRCL (Raji 2R and Raji 4RH) to escalating doses of MK-2206, a selective Akt inhibitor, or idelalisib, a selective PI3 kinase delta inhibitor, resulted in a dose- and time-dependent decrease in cell viability of RCRCL and to a lesser degree RRCL and RSCL (RCRCL vs. RRCL/RSCL, p

Description: Abstract 2761 Clinical outcome of patients with relapsed/refractory diffuse large B-cell lymphoma (DLBCL) and mantle cell lymphoma (MCL) remains poor with currently available therapies. Recently, the ubiquitin-proteasome system (UPS) appears to play an important role in the development of resistance in MCL and some subtypes of DLBCL. Targeting UPS represents a rational approach in an attempt to eradicate drug-resistant lymphoma clones. MLN4924 is a novel, potent and selective inhibitor of the NEDD8-activating enzyme (NAE) that is necessary for the modification of cullin-RING ubiquitin ligases. We evaluated the anti-tumor activity of MLN4924 against a panel of rituximab-sensitive (RSCL) or rituximab/chemotherapy–resistant (RCRCL) DLBCL and Burkitt lymphoma cell lines, cytarabine-sensitive or -resistant (AraCR) MCL cell lines, and primary tumor cells freshly isolated from lymphoma patients (n=13). Lymphoma cells were exposed to escalating doses of MLN4924 alone or in combination with selected chemotherapy agents for up to 72 hrs. Changes in the cell viability or ATP content were determined by alamar Blue reduction or CellTiterGlo assays, respectively. Induction of apoptosis and changes in the levels of NFkB and UPS regulatory proteins were analyzed by Western blotting. Cell cycle alterations were determined by propidium iodide staining and NFkB activity was quantified by flow cytometry using the Imagestream technology. MLN4924 demonstrated time- and dose-dependent anti-lymphoma activity in all cell lines tested. The IC50 in RSCLs were Raji=400nM, RL=1uM and U2932=〉3uM. All RCRCLs were less responsive to MLN4924 as a single agent with IC50 concentrations 4–10× those of their respective sensitive parental cell lines. The MCL cell lines Mino, MinoAraCR, Z-138, HBL-2 and HBL-2AraCR were most sensitive to MLN4924 anti-tumor effects (IC50=250nM) with no significant difference between cytarabine-sensitive and -resistant cell lines; while the MCL cell lines Rec-1, Rec-1AraCR, Jeko-1 and Jeko-1AraCR were less sensitive (IC50=500–1000nM). A variable degree of anti-tumor activity was also observed in primary lymphoma cells. In addition to single-agent activity, MLN4924 plus selected anti-lymphoma chemotherapy agents (bortezomib, bendamustine and cytarabine) demonstrated synergy in cytarabine-sensitive and (to a lesser degree) cytarabine-resistant MCL cell lines. Combinations with additional chemotherapeutic agents (doxorubicin and vincristine) resulted in additive effects. Exposure of MCL cells to MLN4924 resulted in G1 cell cycle arrest. In vitro exposure of the more sensitive MCL cell lines Mino and MinoAraCR to MLN4924 resulted in an increase in p-IkBα and down-regulation of both total and nuclear NFkB. The less sensitive cell lines Rec-1 and Rec-1AraCR demonstrated little to no change in NFkB activation following exposure to MLN4924. Additional studies are ongoing to further define the molecular mechanisms of the anti-tumor activity observed following NAE inhibition by MLN4924 in these pre-clinical models and to further evaluate the activity of MLN4924 in in vivo SCID mouse models of B-cell lymphoma. Our data suggests that MLN4924, a novel NAE inhibitor, is active against B-cell lymphomas, particularly MCL, and is a promising agent warranting further investigation in relapsed/refractory aggressive B-cell lymphomas. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 1649 Deacetylases (DACs) are enzymes that remove the acetyl groups from target protein lysines leading to regulation of gene transcription and other cellular processes. Entinostat is a novel and potent class I DAC inhibitor undergoing pre-clinical and clinical testing. We previously demonstrated that entinostat was active against a panel of rituximab-sensitive (RSCL) and –resistant cell lines (RRCL); and enhanced the anti-tumor activity of chemotherapy agents. In the current work, we studied the mechanisms-of-action of entinostat and its effect on rituximab activity and proteasome inhibition. Our results indicate that entinostat potentiates both rituximab and proteasome inhibitor activity in NHL cells through a dual caspase dependent and independent mechanism-of-action. Studies were conducted in RSCL, RRCL and in primary tumor cells isolated from patients with B-cell non-Hodgkin's lymphoma (n = 43). For studies with a proteasome inhibitor, cells were exposed to escalating doses of entinostat +/− bortezomib of up to 72 hrs. Studies were repeated with or without a pan-caspase inhibitor (Q-VD-OPh, 5uM). Changes in mitochondrial potential and ATP synthesis were determined by alamar blue reduction and cell titer glo luminescent assays, respectively. Flow cytometric analysis was used to determine changes in cell cycle. Protein lysates from entinostat +/− BTZ exposed cells were evaluated for changes in members of Bcl-2 and cell cycle family proteins. Microarray analysis was performed to discern differences in gene expression between RRCL and RSCL after incubation with entinostat. Separately, we studied the effects of entinostat in vitro and in vivo on rituximab-associated complement-mediated cytotoxicity (CMC) or antibody-dependent cellular cytotoxicity (ADCC). NHL cells were exposed to entinostat or DMSO 0.01% for 48hrs and then labeled with 51Cr. Labeled target cells were then exposed to rituximab or isotype control (10mg/ml) and human serum (25%) (CMC) or PBMCs (Effector:Target ratio, 40:1) (ADCC). For in vivo studies, 6–8 week old SCID mice were inoculated with 1′106 Raji cells via tail vein injection and after a period of 72 hours animals were divided into six cohorts: control, entinostat alone at 5mg/kg/dose, entinostat alone at 20mg/kg/dose alone, rituximab single agent at 10mg/kg/dose, rituximab and entinostat at 5mg/kg/dose and rituximab + entinostat at 20mg/kg/dose. Entinostat was administered by gastric lavage daily on days +3–8 and +11–16, and rituximab was administered via tail vein injection on days +3, +8, +11, and +16. Difference in survival between treatment groups was performed by Kaplan-Meier analysis. Entinostat exhibited dose-dependent activity as a single agent against RSCL, RRCL and patient-derived primary lymphoma cells. In vitro exposure of lymphoma cells to entinostat resulted in an increase in G1 and a decrease in S phase. Synergistic activity was observed by combining entinostat with BTZ but not rituximab in vitro. Microarray data demonstrated that p21 was up-regulated after exposure to entinostat. Additionally, we found an upregulation of ICAM2, down-regulation of E2F2 but no changes in CD20 mRNA. Findings were confirmed at the protein level. Caspase inhibition diminished entinostat anti-tumor activity in RSCL but not in RRCL suggesting that entinostat has a dual mechanism-of-action and can induce cell death by caspase-dependent and independent pathways (i.e. cell cycle arrest). Furthermore, In vivo treatment of SCID mice with high dose entinostat in combination with rituximab led to prolongation of the mean survival (87 days, 72–103 95% C.I. P=0.05) compared to rituximab (67 days, 52–82 95% C.I.) entinostat 20 mg/kg (19 days, 19–20 95% C.I.) alone. Our data suggests that entinostat is active against RSCL and RRCL and potentiates the anti-tumor activity of BTZ or rituximab. In addition, entinostat appears to have a dual mechanism of action that supports its clinical development in rituximab-sensitive or resistant lymphomas. Ongoing studies are aimed to further define the molecular events responsible for entinostat mechanisms-of-action and optimize its spectrum of anti-tumor activity. (Research, in part, supported by a NIH grant R01 CA136907-01A1 awarded to Roswell Park Cancer Institute). Disclosures: No relevant conflicts of interest to declare.