Ghosh, Asish K.

Abstract B-cell chronic lymphocytic leukemia (CLL) is an incurable disease and represents a significant health problem in the western world. We and others have reported that primary CLL B-cells spontaneously produce increased levels of proangiogenic basic fibroblast growth factor (bFGF) in vitro and that most CLL plasma contains elevated levels of bFGF. However, the precise role of bFGF in CLL pathobiology is not clearly understood. In this study we investigated the functional implication of the FGF/FGF receptor (FGFR) signaling axis in CLL B-cell biology. We have detected expression of FGFR1 and FGFR3 with comparatively higher levels of the latter receptor tyrosine kinase (RTK), but no or notably low levels of FGFR2/FGFR4, by flow cytometry and Western blot analyses in primary CLL B-cells. This observation was further supported by detection of FGFR1/FGFR3 transcripts in CLL B-cells by semi-quantitative reverse transcriptase polymerase chain reaction. Although both FGFR1 and FGFR3 in CLL B-cells remain as constitutively phosphorylated, we found significantly higher levels of phosphorylation on FGFR3 and thus this latter receptor is likely the predominant RTK of the FGFR family in these leukemic B-cells. Of note, in vitro stimulation of FGFRs with recombinant bFGF was unable to increase total phosphorylation on FGFRs from their constitutive basal levels in CLL B-cells. Further analysis using a bFGF neutralizing antibody suggested that FGFR phosphorylation in CLL B-cells is likely independent of bFGF ligation. We then interrogated the mechanism of how FGFRs were being phosphorylated and/or maintained at the observed constitutive levels of phosphorylation in CLL B-cells. Our previous studies established that Axl is a critical RTK in CLL B-cells since it acts as a docking site for multiple cellular kinases/lipase, an observation supported by earlier literatures in human malignancies. Given this, Axl is likely capable of cross talk with other RTKs including FGFRs to regulate FGFR-signaling in CLL B-cells. Therefore, in an effort to determine whether Axl is functionally associated with FGFR, we examined if these two RTKs exist in the same molecular complex in CLL B-cells. Indeed, immunoprecipitation assays demonstrated that Axl formed a complex with FGFR3 in CLL B-cells, suggesting that Axl is likely functionally linked to the FGFR signaling. In this regard we found that Axl inhibition, using a high-affinity Axl inhibitor (TP-0903; Tolero Pharmaceuticals), resulted in significant reduction of total FGFR phosphorylation in CLL B-cells. Additionally, siRNA-mediated partial depletion of Axl in CLL B-cells reduced total FGFR phosphorylation. In contrast, inhibition of FGFR phosphorylation using a high-affinity FGFR inhibitor could not alter phosphorylation levels on Axl RTK in CLL B-cells. Together, these findings suggest that Axl has a dominant role in the regulation of FGFR signaling in CLL B-cells. To find out if inhibition of FGFR can induce apoptosis in CLL B-cells we used a specific inhibitor for FGFR (TKI-258; Novartis) to treat CLL B-cells. Here we found a substantial level of apoptosis induction in the leukemic B-cells with a mean LD50 dose of ~2.5 μM. Interestingly, Axl inhibition by TP-0903 induced a robust level of apoptosis in CLL B-cells in the nanomolar dose range with a mean LD50 dose of 0.14 mM. Thus Axl inhibition exerts a very robust cytotoxic effect on CLL B-cell survival likely targeting both Axl and FGFR signaling pathways via Axl inhibition. In conclusion, we have detected expression of constitutively active FGFR1 and 3 in primary CLL B-cells and that inhibition of FGFR signaling induces considerable levels of CLL B-cell apoptosis albeit lower than that observed on Axl RTK inhibition. Interestingly, our findings here suggest that Axl forms an active RTK complex with FGFR and that Axl inhibition modifies FGFR phosphorylation levels. Thus it is likely that Axl RTK can regulate FGFR signaling in the CLL B-cells. In total these observations suggest that the finding of robust induction of apoptosis in CLL B-cells is as a result of targeting two signaling pathways with Axl inhibition: Axl and FGFR. These studies further support investigation of Axl inhibition as a way to develop a more effective and efficient therapeutic intervention for CLL patients. Disclosures Warner: Tolero Pharmaceuticals: Employment, Equity Ownership, Patents & Royalties. Kay:Genetech: Research Funding; Pharmacyclics: Research Funding; Hospira: Research Funding; Celgene: Membership on an entity's Board of Directors or advisory committees; Gilead: Membership on an entity's Board of Directors or advisory committees.

Abstract Background: A challenge for novel therapeutic strategies will be the fine tuning of intracellular reactive oxygen species (ROS) signaling in order to effectively deprive cells from ROS-induced tumor promoting events and subsequent tipping the balance to ROS-induced apoptotic signaling. ROS plays a critical role in regulation of the pro-survival receptor tyrosine kinase (RTK) signaling pathways in human cancers. Studies have identified mitochondrial metabolism as the key source for abundant ROS in chronic lymphocytic leukemia (CLL). Unlike in other malignant cells, increased oxidative phosphorylation but not increased aerobic glycolysis has been found in CLL B-cells. While we have detected constitutively phosphorylated/active RTKs like Axl in CLL B-cells, the mechanism of activation remains largely undefined. Here we report that (i) elevated ROS activates the Axl signaling pathway in CLL B-cells and, (ii) mechanism of increased ROS accumulation in the leukemic B-cells. Methods: CLL B- and normal B-cells were purified from peripheral blood of previously untreated CLL patients and normal healthy individuals, respectively using a RossetteSep B-Cell enrichment kit. CLL B-cells were exposed to H2O2 for 5 min and cell lysates were analyzed for activation of RTKs including Axl by immunoprecipitation/Western blots. We also examined expression status of SIRT3, acetylated-superoxide dismutase (SOD)2 and catalase in CLL B-cells by Western blots. Catalase mRNA levels in CLL B-/normal B-cells were determined by qRT-PCR. In some experiments, genomic DNA was isolated from CLL B-/normal B-cells for catalase promoter methylation studies. Finally, accumulation of O2‾ and H2O2 in CLL B- and normal B-cells were measured by flow cytometry after treating the cells withdihydroethidium (DHE) and dichlorohydrofluorescein diacetate (DCFDA), respectively. Results: We found that enforced induction of ROS significantly increases tyrosine phosphorylation levels on multiple cellular proteins in CLL B-cells (Fig. 1). Further analysis finds that increased ROS activates Axl and its downstream targets AKT/Erk1/2 MAPK and the fibroblast growth factor receptor (FGFR) which we recently defined as a downstream target of Axl; while ROS accumulation did not show any significant effect on other RTKs, cMET or IGFR1, in CLL B-cells. Interestingly, the histone deacetylase SIRT3 which activates mitochondrial SOD2 via deacetylation, we found, was overexpressed in CLL B-cells as compared to normal B-cells indicating more efficient conversion of O2‾ into H2O2 in the leukemic B-cells (Fig. 2). However expression of catalase, which converts H2O2 into O2 and H2O, was reduced significantly in CLL B-cells as compared to normal B-cells both at mRNA and protein levels. To delineate the mechanism of reduced expression of catalase in CLL B-cells we studied the human catalase promoter region for potential methylation sites using the method of HpaII restriction enzyme digestion where HpaII is unable to cut DNA (CCGG site) when the internal cytosine is methylated. Initial findings from this approach suggest that while the catalase gene promoter in CLL B-cells may contain methylated cytosines, the catalase gene in normal B-cells may not confer methylation in the promoter region. Next, genomic DNA from CLL B-cells (n=10) and normal B-cells (n=6) was subjected to bisulfite treatment, followed by PCR amplification and sequencing. This approach revealed no significant promoter methylation of the catalase gene in normal B-cells, while moderate to heavily methylated promoter of the catalase gene was detected in CLL B-cells from majority of CLL patients (9 of 10). Further studies are in progress to examine catalase promoter methylation status in a larger cohort of CLL patients and its association with H2O2 accumulation, RTK phosphorylation and risk-factors. Indeed, flow cytometric analysis finds lower levels of O2‾ but accumulation of H2O2 in CLL B-cells as compared to normal B-cells. Taken together, these findings suggest that although SOD2 remains highly active, increased accumulation of H2O2 may occur in CLL B-cells due to epigenetic silencing of the catalase gene. Conclusion: These observations may explain, at least in part, the expression of constitutively active Axl signaling pathway in CLL B-cells that plays a role in CLL B cell survival and may account for the prolonged survival and/or apoptotic resistance of the leukemic B-cells. Disclosures No relevant conflicts of interest to declare.

Abstract Introduction: The use of B-cell receptor (BCR) signal inhibitors-based therapies (e.g., Ibrutinib) for B-chronic lymphocytic leukemia (CLL) was initiated just a few years ago but has rapidly escalated due to their clinical efficacy and relative ease of use. However newer therapeutic approaches are needed due to multiple issues including the continued need to improve complete responses and reduce toxicity profiles. To that end our group has discovered a novel membrane target in the ubiquitous presence of Axl receptor tyrosine kinase (Axl RTK) on CLL B-cells and has reported that the Axl RTK inhibitor TP-0903 is able to induce apoptosis of CLL B-cells at nanomolar doses (Sinha, Clin Cancer Res, 2015). Given this we assessed if TP-0903 would be effective in the induction of apoptosis of leukemic B-cells from CLL patients who are currently on Ibrutinib therapy or whom have relapsed while on Ibrutinib treatment. Methods: Relapsed/refractory CLL patients (n=22) who were placed on Ibrutinib for progressive disease provided blood samples at a median of 3.2 months after Ibrutinib therapy initiation for these studies. We also obtained sequential samples on 8 patients from initial start of ibrutinib therapy and then over a 6 month follow-up period. CLL B-cells from these blood samples were subject to Ficoll separation, purified by using a Rosette Sep B-cell enrichment kit and then studied by flow cytometry to determine Axl RTK expression levels by flow cytometric analysis. Purified CLL B-cells (CD19+/CD5+) were cultured with TP-0903 in vitroat increasing doses (0.01µM - 0.50µM) for 24 hours and the LD50 dose was determined. In addition, 3 CLL patients who had been on Ibrutinib therapy and had a documented relapse were studied in similar fashion using TP-0903. LD50-sensitivity was measured. "LD50-sensitivity" was defined as an LD50 ≤0.50µM and "insensitive" was defined as an LD50 dose >0.50µM. CLL prognostic factors (e.g., FISH, IGHV mutation status, Rai stage, CD38, and CD49d) were evaluated at the time of ibrutinib treatment. Differences in factors between sensitive and insensitive cases were computed using the Kruskal-Wallis test for continuous variables and Chi-square test for categorical variables. Results: Twenty-two CLL patients (5 female, 17 male) were included in the analysis. Fourteen (64%) patients were found to be TP-0903 LD50-sensitive. Axl expression on CLL B-cells for this cohort was heterogeneous with a median of CD19+/CD5+ cells positive for Axl at 69.9% (range of 2.7-91.3%). The sensitive subjects tended to be younger with a median age at Ibrutinib treatment initiation of 62 vs 75.5 years (p=0.004). There were no significant differences in gender, FISH, IGHV mutation status, CD38, CD49d, or Rai stage between the sensitive and insensitive LD50 groups. There were no significant differences in relation to median Axl expression on CLL B-cells (sensitive: 72.6%, range: 2.7-91.3%; insensitive: 41.5%, range: 16.5-83.1%; p=0.35). The median number of treatments prior to initiation of ibrutinib did not differ between sensitivity groups (sensitive: 2.53, range: 8-10; insensitive: 43.5, range 12-20; p=0.2833). Association for ZAP70+ CLL B-cells tended to have more apoptosis induction by TP-0903 (sensitive: 84.6% ZAP70+; insensitive: 42.9% ZAP70+; p=0.052). In 8 CLL patients that were studied sequentially while on Ibrutinib continued to express Axl or increased their Axl expression (n=2) over a 3-6 month follow-up period. Three CLL patients who had relapsed on Ibrutinib were sensitive to TP-0903 with LD50 values of ≤0.50µM. Summary: Here we find that CLL B-cells from over 60% of relapsed CLL patients on Ibrutinib therapy were highly sensitive to the high-affinity Axl inhibitor TP-0903 with induction of apoptosis at nanomolar doses (≤0.50µM). The sensitivity of CLL B-cells to TP-0903 appears to be independent of Axl expression levels and of the known CLL prognostic factors but more evident for younger patients and for ZAP70+ expression status. Given this level of activity for apoptosis induction of CLL B-cells by TP-0903 encourages the further testing of this drug in clinical trials for CLL patients. Disclosures Parikh: Pharmacyclics: Honoraria, Research Funding. Shanafelt:Pharmacyclics: Research Funding; Janssen: Research Funding; Genentech: Research Funding; GlaxoSmithKline: Research Funding; Celgene: Research Funding; Cephalon: Research Funding; Hospira: Research Funding. Warner:Tolero Pharmaceuticals: Employment, Equity Ownership, Patents & Royalties. Bearss:Tolero Pharmaceuticals: Employment, Equity Ownership, Patents & Royalties. Kay:Pharmacyclics: Research Funding; Tolero Pharmaceuticals: Research Funding; Acerta: Research Funding; Gilead: Membership on an entity's Board of Directors or advisory committees; Morpho-Sys: Membership on an entity's Board of Directors or advisory committees; Infinity Pharm: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees.

Abstract While cellular communication and interactive signaling typically relies on multiple mechanisms including secretion of various growth factors, cytokines, bioactive lipids, and adhesion molecules, we have recently focused on cell-to-cell communication that involves circular membrane vesicles designated as microvesicles (MVs). These are normal constituents of blood plasma released by leukocytes, endothelium, platelets and erythrocytes and are generated by cell activation and growth. In normal serum, the order of MV abundance is platelet-derived (80%), endothelial (10%) and leukocyte-derived (10%) but MV can also be released by malignant cells. The size varies from 0.1–1.0 μm and, MVs express antigens characteristic of the cell of origin, carry other membrane and cytoplasmic constituents and exhibit negatively charged phospholipids (phosphatidylserine) detectable by Annexin V binding. To begin to characterize MV and their potential for communication in CLL, we isolated and characterized MVs from the plasma of chronic lymphocytic leukemia B-cell (CLL B) patients (Rai 0-IV). After isolation from platelet-free plasma, MVs were characterized by flow cytometry for forward and side scatter using 1μm fluorescent beads and stained with Annexin V. Our results suggest that MVs are heterogeneous in size with the majority of MVs within 1μm. Transmission electron microscopy confirmed the size heterogeneity of MVs and exhibited the morphology of the membrane-bound vesicles after phosphotungstic acid staining onto parlodium-coated 300 mesh copper grids. Flow cytometric analysis further demonstrated the presence of CD61 (platelet/megakaryocyte marker), CD19, CD5, CD52, and CD20 on the surface of MV, albeit in differential amounts depending on the stage of the disease and the total leukocyte count. For example, in Rai stage IV (n= 5), we found lesser numbers of MV carrying platelet-derived marker CD61 and more CD19 (41–79%), and CD5 compared to normal MV; 1–5% MV from healthy controls expressed CD19 while Rai stage 4 patients MV expressed CD19 from 41–79%. These results suggest that a significant number of MVs are generated from the leukemic B-lymphocytes in more advanced Rai stage. However, we also found that patients with a higher blood lymphocyte count but lower Rai-stage (Rai I/II) contain moderate levels of CD19-bearing MVs (13–43%). Importantly, presence of MVs carrying CD52/CD20 in the blood of CLL patients may undermine the therapeutic effectiveness of alemtuzumab and rituximab. To this end, we observed that MVs could be incorporated into the CLL B cells as evident from the expression of the platelet antigen CD61 on CLL B cell surfaces after co-culture of CD61-bearing MVs with CLL B cells. Global proteomic analysis of MVs isolated from an advanced stage (Rai IV) CLL B patient detected the presence of about 700 proteins that normally reside in the nucleus, cytoplasm or membrane. In summary, we have isolated and characterized plasma MV from CLL B patients and found them to be more likely generated from CLL B cells in relation to disease stage or total lymphocyte count. These MVs contain a multitude of proteins and are able to integrate into bystander CLL B cells. We also found that these MVs carry a large set of proteins, which could potentially not only modulate cell-cell communication but also interrupt monoclonal antibody directed therapy. Further studies are underway to dissect their roles in CLL B malignancy.

Abstract Introduction: B-cell chronic lymphocytic leukemia (CLL) is an incurable disease despite that still requires options for treatment despite recent advances with BTK therapy. We previously found that Axl receptor tyrosine kinase (RTK), a member of the TAM (Tyro3, Axl, MER) family of RTKs, plays a critical role in CLL B-cell survival (Blood 2011, 117:1928). Here, we explored the levels of apoptosis induction of CLL B-cells using a novel high-affinity Axl inhibitor as a single agent or in combination with BTK inhibitors. Methods: Expression/activation status of the Tyro3, Axl and MER in CLL B-cells was evaluated by Immunoprecipitation/Western blots. Cells were treated with a high-affinity orally bioavailable Axl inhibitor TP-0903 (Tolero Pharmaceuticals) or FGFR inhibitor TKI-258 (Novartis) with or without the presence of CLL bone marrow stromal cells (BMSCs). Impact of Axl or FGFR signal inhibition on CLL B-cell survival was also evaluated by flow cytometry after staining with annexin V/propidium iodide. Finally, cells were exposed to TP-0903 in combination with BTK inhibitors to determine any synergistic/additive effects of the combination in induction of apoptosis. Combination index (CI) of two agents was calculated using the method of Chou and Talalay, 1984. Results: Although we defined pro-survival role of Axl in CLL B-cells, contribution of other members of the TAM family of RTKs including Tyro3 and MER to prolonged survival of the leukemic B-cells remains largely undefined. In this study we found that CLL B-cells overexpress Tyro3, but not MER, and that Tyro3 is constitutively phosphorylated. Interestingly, immunoprecipitation of Tyro3 from CLL B-cell lysates shows retention of highly active Axl in the immune complex suggesting that these two RTKs either heterodimerize or form a molecular complex in CLL B-cells. Our earlier work found that Axl regulates multiple cell survival signaling pathways including PI3K/AKT and Src and that, Axl inhibition using a first generation Axl inhibitor induced apoptosis in CLL B-cells (Blood 2011, 117:1928). Thus, we hypothesized that targeting Axl using a very high affinity Axl inhibitor would induce even higher levels of apoptosis in CLL B-cells compared to the first generation Axl inhibitor. To address this, we utilized a high-affinity orally bioavailable Axl inhibitor TP-0903 to generate preliminary information on its efficiency to specifically target Axl and the level of apoptosis induction in CLL B-cells. Indeed TP-0903 as a single agent induced a robust apoptotic cell death in CLL B-cells within 24 hours of exposure at nanomolar doses, which are easily achievable in vivo (information provided by Tolero Pharmaceuticals). Interestingly, while TP-0903 effectively reduces phosphorylation on Axl in CLL B-cells, it was unable to target P-Tyro3 suggesting that Axl is likely the predominant RTK of the TAM family in CLL B-cells. A marked reduction of the anti-apoptotic proteins Mcl-1, Bcl-2, XIAP and upregulation of the pro-apoptotic protein BIM in CLL B-cells via activation of its upstream regulator FOXO3a were noted as a result of inhibition of P-Axl and its downstream signaling pathways, P-AKT/P-Src. Importantly, TP-0903 was able to overcome CLL BMSC mediated protection of the leukemic B-cells from apoptosis induction. Given the observed robust induction of apoptosis with TP-0903 and the known impact of BTK inhibitors on CLL patient clinical outcome, we then exposed CLL B-cells to TP-0903 in combination with two BTK inhibitors that included Ibrutinib and a reversible BTK inhibitor TP-4216 (a kind gift from Tolero Pharmaceuticals). Based on the CI values, we found that the in vitro combination of TP-0903 and Ibrutinib showed additive effects in CLL B-cells from 6 of 11 CLL patients, CLL B-cells from the remaining 5 CLL patients showed antagonistic effects (Fig. 1A). However, we observed synergistic effects of TP-0903 in combination with TP-4216 in the induction of apoptosis in CLL B-cells from 2 of 10 patients and additive effects in CLL B-cells from 6 patients (Fig. 1B). Cells from the remaining 2 patients exhibited a slightly antagonisticeffect. Conclusion: In total, these results suggest that TP-0903 is quite effective in inducing robust CLL B-cell apoptosis as a single agent even at a nanomolar concentration and its combination with the reversible BTK inhibitor TP-4216 could be an effective and unique strategy for the treatment of CLL patients. Figure 1 Figure 1. Disclosures Warner: Tolero Pharmaceuticals: Employment, Equity Ownership, Patents & Royalties. Bearss:Tolero Pharmaceuticals: Employment. Shanafelt:Genentech: Research Funding; GlaxoSmithKline: Research Funding; Celegene: Research Funding; Cephalon: Research Funding; Pharmacyclics/Jannsen: Research Funding; Hospira: Research Funding; Polyphenon E Int'l: Research Funding. Kay:Genetech: Research Funding; Pharmacyclics: Research Funding; Hospira: Research Funding; Celgene: Membership on an entity's Board of Directors or advisory committees; Gilead: Membership on an entity's Board of Directors or advisory committees.

Abstract Introduction: We demonstrated that elevated levels ofcirculating microvesicles (MV) in CLL activate CLL bone marrow stromal cells (BMSCs) with resultant enhanced vascular endothelial growth factor (VEGF) production, a pro-survival factor for CLL B-cells (Ghosh 2009 Blood 113:5568-74). We also detected a phenotypic shift from predominantly platelet/megakaryocyte-derived MV (CD61+) in early stages towards a more leukemic B-cell derived MV (CD19+) in advanced Rai stages (Ghosh 2010 Blood 115:1755-64). These observations suggest that the MV parameters of their levels and surface phenotypes may predict or be associated with the clinical outcome and/or therapeutic response in a given CLL patient. In this study we therefore designed an approach to being to assess these latter possibilities in regard to MV in CLL. Methods: In vitro stimulation – Purified CLL B-cells were stimulated in vitro using CpG oligonucleotide (n=10) or anti-IgM antibody (n=10) for 72 hours. MV were isolated from the used media, levels measured and then phenotyped by flow cytometry (BD Canto-I) using standard TRUCount beads (BD Biosciences). In vivo (plasma) work – MV were isolated from normal plasma (n=5) or plasma of 17 CLL patients entered onto a previously reported Pentostatin, Cyclophosphamide, Ofatumumab (PCO) clinical trial at their entry to the trial and at 6-months after therapy and then sequentially over a period of 18 months (Shanafelt 2013 Cancer 119:3788-96). We determined the phenotypes and plasma levels of MV as described above. Results Phenotypes of in vitro generated MV: We found that, for the most part, CLL B-cells shed MV upon in vitro stimulation however; some CLL B-cells of unmutated IGHV status, but not mutated status, spontaneously generated MV in culture (Fig. 1). Most interestingly, we found that the surface phenotype of MV generated from CLL B-cells carrying predominantly the CD52 surface marker, not the CD19 or CD5 marker. Interestingly, none of the in vitro generated MV was CD20+, another known marker of CLL B-cells. We found similar phenotypes of the in vitro generated MV by stimulating two different signaling pathways, Toll-like receptor and BCR, in CLL B-cells. Together these results suggest that CLL B-cells generate predominantly CD52+ MV, but not CD19 or CD5, upon in vitro stimulation. MV phenotypes in pre- and post-therapy CLL plasma: Here we focused on the kinetics of MV levels and phenotype in relation to disease outcome for PCO treated CLL. Thus, we found total MV levels were elevated in all patients compared to normal plasma at entry to the PCO trial and then dropped substantially after therapy when compared to the trial entry. Also most CLL patients (n=11) continued to maintain low plasma MV levels until the end of the follow up period (18 months). Interestingly, in a few cases (n=6) total plasma MV levels decreased initially after therapy but then started to rise. These patients are now being followed for disease status in order to determine if rising MV levels are related to a disease relapse. Of interest on flow phenotyping the plasma MV obtained at entry to the PCO trial we detected predominantly CD52+ MV but not CD19+ MV, similar to the findings of the in vitro MV generated from CLL B-cells. Normal plasma (n=5) does not contain significant levels CD52+ MV or CD19+ MV. Together these results suggest that the malignant B-cells shed preferentially CD52+ MV, but not CD19+ MV. We are currently analyzing the MV levels and phenotypes pre- and post-therapy response levels in relation to the patients’ response to therapy in a larger cohort of treated CLL patients (n=60). Conclusion: Current understanding about the phenotypes of MV is based on the concept that MV carry the unique surface marker of the originator cells, e.g., CD19 on B-cell derived MV. However, findings from our in vitro stimulation studies demonstrate that the CD52 but not CD19 is preferentially expressed on CLL B-cell derived MV. We are currently studying the mechanism for such a preferential surface expression on these leukemic derived MV. To our knowledge, this is the first report describing that CLL B-cells generate preferentially CD52+ MV. Our studies are now committed to a more extensive analysis of MV parameters in association with CLL disease progression, depth of clinical response and if the elevation of plasma MV heralds a relapse in responding patients. Figure 1 Figure 1. Disclosures Shanafelt: Genentech: Research Funding; GlaxoSmithKline: Research Funding; Celegene: Research Funding; Cephalon: Research Funding; Pharmacyclics/Jannsen: Research Funding; Hospira: Research Funding; Polyphenon E Int'l: Research Funding. Kay:Genetech: Research Funding; Pharmacyclics: Research Funding; Hospira: Research Funding; Celgene: Membership on an entity's Board of Directors or advisory committees; Gilead: Membership on an entity's Board of Directors or advisory committees.

Abstract Abstract 3927 PURPOSE: CLL B-cells interact with their microenvironment in a variety of ways that influence leukemia cell survival. Contact mediated interactions are the most widely studied. More recently, bidirectional cytokine mediated interactions have been found to modulate spontaneous and drug induced CLL B-cell death. Specifically, upregulation of CCL2, CCL3, and CCL4 are found when CLL B-cells are co-cultured with various accessory cells. We set out to more comprehensively profile the effect of CLL-stromal cell co-culture on immune cytokine regulation and to evaluate the effect of novel therapeutic agents in clinical testing for patients with CLL on these interactions. METHODS: Freshly isolated CLL B-cells (n=24) were cultured in AIM V media for 24 hours alone or in direct contact with HS-5 stromal cells. HS-5 cells were cultured alone in AIM V to determine basal level of immune cytokine secretion. Immune cytokine levels in the media under these 3 conditions (CLL alone; HS5 alone; co-culture) were assessed using multiplex cytokine arrays (Invitrogen). The protective effects of the 6 most highly upregulated cytokines on spontaneous apoptosis of CLL B-cells after 7 day culture were then examined by culturing leukemic cells with and without recombinant cytokines. The effect of sub-lethal doses of the PI3K inhibitor CAL-101, the AKT inhibitor MK2206 and the Bruton's Tyrosine Kinase inhibitor PCI-32765 were examined. To compare the effect of these drugs to a more conventional chemotherapeutic agent, we evaluated fludarabine (F-Ara-A) under the same conditions. To explore the ability of these novel agents to overcome stromal rescue, CLL B-cells pre-exposed to bendamustine (BEN) were cultured in media alone, or in direct contact with stroma with and without sub-lethal doses of CAL-101, MK2206, or PCI-32756. Leukemic cell viability was assessed by flow cytometry using annexin/PI staining. RESULTS: Six cytokines increased ≥4 fold (e.g. >400%, all p<0.0001) on coculture relative to the expected additive value of individual cultures. Consistent with previous reports, (Schulz 2011; Burger 2009) CCL3 and CCL4 were found to have the greatest upregulation (both >20 fold the additive value). The other highly upregulated cytokines detected have not been previously reported to be influenced by coculture: IL12 (15 fold additive value), IL1β (11 fold additive value), TNF-α (11 fold additive value), and IL-1RA (4 fold additive value). Although transwell experiments (n=6) suggested upregulation of these cytokines was largely contact dependent, substantial upregulation of several cytokines still occurred without direct contact. Each of the 6 most upregulated cytokines reduced spontaneous apoptosis when CLL B-cells were cultured for 7 days in media plus recombinant cytokine. When CLL B-cells and HS-5 cells were co-cultured with CAL-101, MK-2206, or PCI-32765, the effects on cytokine secreted upregulation were cytokine and drug specific (e.g. one drug affected individual cytokines differently; different drugs affected different cytokines). For example, PCI-3276 and CAL-101 had similar affects on the upregulation of CCL3 and CCL4 on co-culture but opposite effects on TNF-α and IL-1β. In contrast, F-Ara-A had minimal effect on upregulation of any of these 6 cytokines. Finally, we evaluated the ability of these novel agents to overcome stromal rescue of CLL B-cells treated with conventional chemotherapeutic agents. At sublethal doses CAL-101, MK2206, PCI-32765 each significantly overcame stromal rescue of BEN exposed CLL B-cells. (Fig.1) CONCLUSIONS: Direct physical interaction of CLL B-cells with bone marrow stromal cells resulted in profound upregulation of multiple immune cytokines. The 6 most upregulated cytokines identified in these experiments all reduced spontaneous apoptosis of CLL B-cells on extended culture. Sublethal doses of the PI3K inhibitor CAL-101, the AKT inhibitor MK2206, and the Bruton's Tyrosine Kinase inhibitor PCI-32765 had distinct effects on stromal cell mediated cytokine upregulation. Importantly, each of these 3 drugs also overcame stromal mediated rescue of BEN exposed CLL B-cells. These in vitro findings support the ongoing clinical trials of BEN in combination with CAL-101 (ASH 2011), PCI-32765 (ASCO 2012) or MK2206 (Ding et. al., unpublished observation). Disclosures: Kay: Celgene: Research Funding. Shanafelt:Celgene: Research Funding; Hospira: Research Funding; Cephalon: Research Funding; Genentech: Research Funding; Glaxo-Smith-Kline: Research Funding; Polyphenon E International: Research Funding.

Abstract Abstract 2368 Poster Board II-345 Background: B-cell chronic lymphocytic leukemia (CLL) is characterized by the progressive accumulation of CD5+ B lymphocytes in the peripheral blood, lymphoid organs and bone marrow. Despite aggressive therapy, CLL is still incurable partly because of intrinsic defect in apoptosis induction. A novel therapeutic agent, Bosutinib, was initially developed as an inhibitor of Src and Abl kinases and is currently in phase II clinical trials for the treatment of several human malignancies. Recently, Bosutinib has been shown to inhibit phosphorylation of a novel receptor tyrosine kinase, Axl which has been reported to be overexpressed in several types of human cancers including colon, prostatic, thyroid, breast, gastric, renal and lung. Previously, Dasatinib, another Src/Abl kinase inhibitor, showed cytotoxic effects on CLL B cells by decreasing levels of activated, phosphorylated forms of Akt, Erk1/2 and p38 and reducing expression of anti-apoptotic proteins Mcl-1 and Bcl-xL in CLL B cells. Here, we wished to examine receptor or non-receptor tyrosine kinases active in primary CLL B-cells and determine their status after exposure to Bosutinib as well as the latter drug's effect on CLL B-cell viability. Methods: We used freshly isolated CLL B cells after obtaining written consent from patients. Bosutinib was used at various doses (2.5, 5.0, 10.0 and 20.0 μM) to treat CLL B cells in vitro for 24/48 hrs. Induction of apoptosis was assessed by annexin/propidium staining. To examine the impact of stroma on Bosutinib induced CLL B-cell death, primary CLL-bone marrow stromal cells (BMSC) were cocultured with CLL B cells at a cell density ratio of 1:20 and treated with various doses of Bosutinib for 24 hrs. Expression status of various kinases and downstream targets were analyzed in CLL B cell lysates with or without Bosutinib-treatment by Western blot using specific antibodies. Results: Treatment of CLL B-cells with Bosutinib induces a massive apoptotic cell death in a dose- and time-dependent manner (IC50 for 24 h; ∼10 μM and IC50 for 48 h: 5-10 μM) which involves PARP cleavage as demonstrated by Western blot analysis. Moreover, Bosutinib-treatment reduced expression of several key anti-apoptotic proteins, Mcl-1, XIAP and Bcl-2 reported to be overexpressed in CLL B cells. Interestingly, we detected that the majority of CLL B-cells express constitutively active Axl. Importantly, Bosutinib treatment inhibited phosphorylation of Axl in CLL B cells resulting in inhibition of AKT-activation, one of its downstream signaling pathways. Previous studies have suggested a possible physical association between Axl and Src kinase. We observed that expression of constitutively active Axl was associated with the presence of highly phosphorylated Src kinase when compared with that in CLL B-cells with low or unphosphorylated Axl. These observations suggest that phosphorylation of Axl may be an upstream event for Src activation in CLL. We found inhibition of constitutively active Axl resulted in subsequent inhibition of Src kinase activation in CLL B-cells following Bosutinib-treatment. We also detected inhibition of ZAP70/Syk-phosphorylation in CLL B cells upon Bosutinib-treatment. Finally, we found Bosutinib was able to overcome stomal protection of CLL B cells at a dose of 10 μM in an in vitro coculture system suggesting its potential as a therapeutic agent against CLL. Conclusion: Together, these observations suggest that Bosutinib induces apoptosis in CLL B-cells, even in the presence of stromal cells, in association with the down regulation of multiple kinases including the novel receptor-tyrosine kinase, Axl, and reduces expression of the anti-apoptotic proteins critical to CLL B-cell survival. In total, these findings for the first time indicate that Bosutinib has the potential to be a very potent therapeutic agent for CLL patients. Disclosures: Kay: Biogenc-Idec, Celgene, Genentech, genmab: Membership on an entity's Board of Directors or advisory committees; Genentech, Celgene, Hospira, Polyphenon Pharma, Sanofi-Aventis: Research Funding.

Abstract Abstract 804 Background: We have demonstrated that crosstalk between CLL B-cells and marrow-derived mesenchymal stem cells (MSC) can modulate the activation of both. Soluble factors in the CLL-condition media (CM) regulate the migration and proliferation of MSC. We previously found that platelet-derived growth factor (PDGF) receptors were selectively activated in MSC by CLL-CM and are necessary for the downstream Akt activation in MSC. PDGF is capable of promoting vascular endothelial growth factor (VEGF) production in MSC; however, several questions remain unanswered. Can CLL B-cells produce PDGF; is PDGF the specific mediator in CLL-CM responsible for PDGFR activation in MSC; what is the mechanism of PDGF mediated VEGF production in MSC; and is there any clinical significance of PDGF mediated VEGF production by MSC for CLL patients? Methods: MSC were generated from CLL patients and controls as previously described. To determine the specific ligand responsible for PDGFR activation mediated by CLL-CM blocking antibody to PDGF or VEGF or both was used to neutralize the binding activity of PDGF or VEGF in the CLL-CM. Subsequently, the pattern of membrane receptor activation was evaluated using a commercially available receptor tyrosine kinase (RTK) array assay. ELISA was performed to measure PDGF and VEGF levels from plasma and CM of CLL patients and normals. The protein expression of PDGF isoforms in CLL and normal B-cells was assessed using Western blot. To assess the mechanism of PDGF mediated VEGF production by MSC, we used either a PI3K inhibitor, LY294002 (20 mM), or a p38 MAPK inhibitor, SB202190 (100 nM), to block these pathways in MSC. Results: The known ligands for PDGF receptors, PDGF and VEGF, were detected in the CLL-CM (PDGF, 113.7 ± 23.6 pg/ml, VEGF, 186.94 ± 54.57 ng/ml; n = 12). We also found that PDGF is present in CLL cells by Western blot analysis and expression of PDGF in CLL cells was up regulated by hypoxia. By using PDGF- or VEGF-blocking antibody or both to neutralize the binding activity of PDGF or VEGF in CLL-CM, we found only PDGF-blocking antibody was able to neutralize the majority of PDGFRa activation, indicating that PDGF in the CM is the predominant ligand for PDGFR activation in MSC. CLL MSC was exposed to either pooled CLL (n = 5) or pooled normal plasma (n = 5). Interestingly, PDGFR was more potently activated by CLL plasma than the normal plasma, suggesting PDGF levels were elevated in the CLL plasma. Indeed, PDGF in the normal plasma appeared to be much lower compared to the CLL plasma [normal (n = 11): 554.43 ± 88.55 pg/ml; CLL (n=43): 3115.36 ± 428.79 pg/ml. p < 0.0001]. In addition, the plasma PDGF level was found to be weakly correlated with VEGF level in 43 CLL patients assessed using Spearman correlation analysis (r = 0.3, p = 0.001). However, when CLL patients were grouped by clinical prognostic factors, the plasma PDGF level was found to be strongly correlated with plasma VEGF level in CLL patients with high-risk features including more advanced Rai stage 2-4 (r = 0.87, p < 0.0001), ZAP-70 positive (r = 0.8, p =0.04) and CLL patients who had required treatment (r = 0.8, p = 0.0008). To examine the mechanism for PDGF augmentation of VEGF by MSC, we found that PI3K pathway interruption using LY294002 (n = 6, p = 0.04), but not p38 MAPK inhibition (n = 6, p = 0.5), was able to abrogate the PDGF mediated VEGF production in MSC. Importantly, we also found that the secreted VEGF level in the CM of unstimulated CLL MSC appeared to be significantly higher compared to the level from culture media of unstimulated normal MSC [CLL (n = 14) vs. normal (n = 5), 298.1± 61.2 vs. 109.9 ± 52.3, p = 0.03]. Conclusions: These results indicate that PDGF secreted by CLL B-cells is necessary for PDGFR activation in MSC and can lead to increased VEGF production by MSC via a PI3K-Akt dependent mechanism. These findings have clinical implications as plasma PDGF levels are strongly correlated with plasma VEGF levels in high risk CLL. The finding that CLL MSC are more prone to secrete VEGF than normal MSC suggests the stromal microenvironment may function to facilitate CLL disease progression. Finally, as VEGF is known to enhance CLL B-cell survival and drug resistance, interrogation of the PDGF-VEGF regulation in the microenvironment of CLL patients should yield important information that can be used for therapeutic strategies in CLL. Disclosures: Kay: Biogenc-Idec, Celgene, Genentech, genmab: Membership on an entity's Board of Directors or advisory committees; Genentech, Celgene, Hospira, Polyphenon Pharma, Sanofi-Aventis: Research Funding.