ALBERT

Description: Myeloma (MM) cells grow and expand almost exclusively in the bone marrow while creating a cellular microenvironment suitable for MM cell growth and survival (MM niche). In pursuing the molecular mechanisms whereby MM cells gain drug resistance in the “MM niche”, we have found that the serine/threonine kinase Pim-2 is constitutively over-expressed in MM cells, and further up-regulated by co-cultures with bone marrow stromal cells (BMSCs) as well as osteoclasts (Leukemia, 2011), and that Pim-2 is an important therapeutic target in MM for the progression of MM tumor and bone disease (Leukemia, 2014). The ABC transporter BCRP is preferentially expressed in drug resistant MM cells as well as in MM progenitors or stem cells. BCRP has been demonstrated to be phosphorylated by Pim kinases to trigger its dimerization and function; Pim inhibition may suppress the BCRP function to sensitize BCRP-expressing MM cells to chemotherapeutic agents. In the present study we therefore explored whether Pim inhibition is able to target and impair BCRP-expressing drug-resistant MM cells and MM progenitors. We analyzed an ABC transporter activity in BCRP-expressing RPMI8226 and KMS11 cells by intracellular accumulation and retention of BCRP substrates with auto-fluorescence emission, mitoxantrone and doxorubicin, in flow cytometry. Treatment with Pim inhibitors, SMI-16a or SMI-4a, increased the incorporation of these drugs into the MM cells and enhanced their subsequent intracellular retention after 6-hour incubation without these drugs, although BCRP expression on their surface was only marginally affected by the Pim inhibition. Interestingly, acidic conditions up-regulated Pim-2 expression while reducing the accumulation and retention of these drugs in BCRP-expressing RPMI8226 and KMS11 cells. However, the Pim inhibitors efficaciously restored the drug accumulation and retention reduced by extracellular acidification, and enhanced the cytotoxic activity of the BCRP substrate doxorubicin against RPMI8226 cells rather preferentially in acidic conditions. Furthermore, the Pim inhibition minimized the sizes of “side populations”, highly drug-resistant fractions with enhanced BCRP activity, and the ability of colony formation in RPMI8226 and KMS11 cells, which was more marked in acidic conditions. We previously demonstrated the in vivo effects of the Pim inhibitors in human INA-6 cell-bearing SCID-rab MM models and syngeneic mouse MM models with an intra-tibial inoculation of 5TGM1 MM cells (Leukemia, 2014). To further examine the acid-tropism of anti-tumorigenic activity of Pim inhibition, we pretreated murine 5TGM1 MM cells in vitro with or without SMI16a at pH6.8 for 24 hours, and transplanted to the tibiae in mice the same numbers of viable MM cells remaining in each treatment group. Treatment with SMI16a at pH6.8 almost completely abrogated in vivo tumorigenic capacity of 5TGM1 cells, while MM cells without the treatment rapidly grew and expanded in and outside of the tibiae, suggesting targeting clonogenic MM cells by Pim inhibition preferentially in acidic conditions. Taken together, Pim-2 may become an important therapeutic target of drug-resistant BCRP-expressing MM cells and their progenitors which appear to gain more drug resistance in acidic bone lesions. Combinatory treatment with Pim inhibitors warrants further study to overcome drug resistance in MM cells, including their tumorigenic cancer stem cells. Disclosures No relevant conflicts of interest to declare.

Description: Abstract 1823 Poster Board I-849 The proteasome inhibitor bortezomib, a novel anti-myeloma (MM) agent, has recently drawn considerable attention to its anabolic actions on bone formation in patients with MM. Bortezomib was reported to enhance the activity of Runx2/cbfa1, an essential transcription factor for osteoblast (OB) induction, in mesenchymal stem cells to induce OB differentiation. However, because over-expression of Runx2 unexpectedly suppresses terminal OB differentiation or mineralization, there may be critical factors involved in OB differentiation in concert with Runx2 to achieve terminal OB differentiation in the treatment with bortezomib. Proteasome inhibition accumulates a variety of proteins and induces ER stress or unfolded protein response. Among proteins induced by ER stress, activating transcription factor-4 (ATF-4) plays a critical role in OB differentiation. ATF-4 is expressed in osteoprogenitors and preOBs following Runx2, and acts in concert with Runx2 to facilitate terminal maturation of OBs. However, it is unknown whether a change in ATF-4 protein levels plays any role in OB differentiation induced by proteasome inhibition. In the present study, we therefore explored the role of ATF-4 in OB differentiation by proteasome inhibition in Runx2-expressing immature OB lineage cells. Bortezomib dose-dependently increased ATF-4 protein levels in primary bone marrow stromal cells and ST-2 stromal and MC3T3-E1 preosteoblastic cell lines at concentrations higher than 10 nM as early as 3 hours. Because serum bortezomib levels reach around 100 nM (Cmax) with T1/2 of 3 hours after iv injection at therapeutic doses, bortezomib treatment in MM patients is expected to enhance ATF-4 protein levels in OB lineage cells. Interestingly, bortezomib treatment did not change mRNA levels of ATF-4 as well as βTrCP1, E3 ligase for ATF-4. Because translation of ATF-4 mRNA is triggered by ER stress response, it is plausible that the ATF-4 accumulation by bortezomib is mediated by the suppression of proteasomal degradation with subsequent induction of ER stress response. MM cell-derived factors and TGF-β released from bone by enhanced bone resorption suppress OB differentiation in MM bone lesions. Treatment with bortezomib was able to accumulate ATF-4 in the presence of MM conditioned media (CM) or TGF-β to the levels similar to those without MM CM nor TGF-β. Furthermore, bortezomib enhanced promotor activity of osteocalcin, a marker of mature OBs, as well as BMP-2-induced mineralized nodule formation in MC3T3-E1 cells, and these effects of bortezomib were suppressed by ATF-4 siRNA. These results demonstrate that bortezomib treatment accumulate ATF-4, and suggest that the effect of bortezomib on OB differentiation is mediated via an accumulation of ATF-4 protein in OB lineage cells. We have previously demonstrated that OB differentiation is suppressed in MM bone lesions, and that differentiated OBs suppress MM cell growth and survival. Thus, resumption of bone formation by bortezomib may further suppress MM cell growth in concert with its direct anti-MM actions. Disclosures No relevant conflicts of interest to declare.

Description: TNF-related apoptosis-inducing ligand/Apo2 ligand (TRAIL/Apo2L) selectively induces apoptosis in various cancer cells including myeloma (MM) cells. Although TRAIL-mediated immunotherapy is as attractive maneuver against MM because of its tumor-specific cytotoxic activity, the susceptibility of MM cells to TRAIL has been demonstrated to vary among MM cells, which limits clinical applications of TRAIL-mediated immunotherapy. TRAIL is known to bind to 2 different proapoptotic receptors, death receptor 4 (DR4) and DR5. The membrane-bound TRAIL receptors have been reported to be subjected to proteolytic cleavage in melanoma cells in a manner inhibitable by tissue inhibitor of metalloproteinases-3 (TIMP-3), an endogenous inhibitor for TNF-alpha converting enzyme (TACE) (Ahonen M, Oncogene. 2003). Although TACE is regarded as a sheddase for TNF-like ligands/receptors, the role for TACE in TRAIL-mediated cytotoxicity in MM cells is largely unknown. In the present study, we therefore investigated the role for TACE in TRAIL and its receptor editing as well as the effect of TACE inhibition on TRAIL-triggered cytotoxicity in MM cells. MM cells constitutively expressed DR4 on their surface as determined in flow cytometry, while the expression of DR5 as well as TRAIL was negligible. TACE mRNA was constitutively overexpressed in all MM cell lines including RPMI8226, MM.1S, KMS12 and U266, whereas its endogenous inhibitor, TIMP-3, was not detected, suggesting enhanced TACE activity along with down-regulation of TIMP-3 in MM cells. TAPI-0, a TACE inhibitor, up-regulated the cell-surface expression of DR4 about 2-fold in RPMI8226 and U266 cells, while DR5 and TRAIL expression remained undetectable. Immunoblot analyses for DR4 revealed that conditioned media from RPMI8226 and U266 cells contained DR4 protein with a molecular size smaller than that observed in their cell lysates. Interestingly, TAPI-0 treatment down-regulated DR4 levels in culture supernatants of these MM cells, while increasing DR4 levels in their cell lysates as well as on their surface, suggesting ectodomain shedding of DR4 by TACE in MM cells. Importantly, TAPI-0 treatment potentiated TRAIL-triggered cytotoxicity against MM cells including RPMI8226 and MM.1S, which was almost completely antagonized by addition of osteoprotegerin, a soluble inhibitor for TRAIL, indicating a TRAIL-dependent mechanism. Taken together, the present observations demonstrate that cell-surface TRAIL receptor expression is post-transcriptionally down-regulated by endogenous TACE activity enhanced in MM cells, which may protect MM cells from TRAIL-mediated apoptosis by surrounding immune cells expressing TRAIL, and that up-regulation of TRAIL receptor expression by inhibiting TACE activity is able to restore the susceptibility of MM cells to TRAIL. Therefore, these preclinical data provide that TACE inhibitors in combination with TRAIL-based immunotherapy may be a novel therapeutic approach against MM which can not be cured by the present chemotherapeutic modalities.

Description: Multiple myeloma (MM) still remains incurable even with the implementation of novel therapeutic modalities, leading to the idea to develop various forms of immunotherapies. In this regard, γδ T cells bearing Vγ9Vδ2 TCR expanded from peripheral blood mononuclear cells (PBMCs) have attracted attention as potent effectors available in a novel immunotherapy against MM. Human Vγ9Vδ2 γδ T cells can be expanded ex vivo by aminobisphosphonates in combination with IL-2, and effectively target and impair MM cells. However, MM cells appear to protect themselves from external insults by immune cells in a unique bone marrow microenvironment created by the accumulation of mesenchymal stem cells/bone marrow stromal cells (BMSCs) with defective osteoblastic differentiation and acid-producing osteoclasts. To improve the therapeutic efficacy of γδ T cells, therefore, we need to develop a maneuver to effectively enhance the expansion and activity of γδ T cells while disrupting the MM cell-bone marrow interaction. Lenalidomide (Len), a novel immunomodulatory anti-MM agent, shows pivotal anti-MM activity by targeting immune cells as well as the interaction of MM cells and their surrounding cells in the bone marrow. The present study was undertaken to explore the efficacy of Len in combination with zoledronic acid (Zol) or a precursor of isopentenyl pyrophosphate (IPP) (E)-4 hydroxy-3-methyl-but-2-enyl pyrophosphate (HMB-PP), a microbial antigen for Vγ9Vδ2 TCR, on the induction and expansion of Th1-like γδ T cells with enhanced cytotoxic activity against MM cells in the skewed bone marrow microenvironment in MM. When combined with Zol (1μM), clinically relevant doses of Len (around 1 μM) substantially expanded γδ T cells from PBMCs to the levels similar to IL-2 (100 U/ml). Len was able to expand γδ T cells more robustly in combination with HMB-PP (1 μM) than Zol from PBMCs from the majority of normal donors. However, Len alone did not show any significant effects on γδ T cell expansion and activation, suggesting a costimulatory role of Len on Zol or HMB-PP-primed γδ T cells. The surface expression of LFA-1, and the cytotoxicity-associated molecules NKG2D, DNAX accessory molecule-1 (DNAM-1; CD226) and TRAIL were up-regulated in the expanded γδ T cells. Although functional diversity has been demonstrated in γδ T cells expanded by various stimuli, Len in combination with either Zol or HMB-PP enhanced intracellular IFN-γ along with the surface NKG2D but not Foxp3 in γδ T cells at higher levels than IL-2, suggesting robust induction of Th1-like γδ T cells by Len. Importantly, γδ T cells expanded with the combinatory treatments with Len and Zol or HMB-PP exerted potent cytotoxic activity against MM cells but not normal cells surrounding MM cells in bone marrow samples from patients with MM. Such treatments with Len was able to maintain the cytotoxic activity of the γδ T cells against MM cells in acidic conditions with lactic acid, and restored their anti-MM activity blunted in the presence of BMSCs. Interestingly, the expanded γδ T cells markedly suppressed the colony formation in semi-solid methylcellulose assays of RPMI8226 and KMS-11 cells [81±1 (mean ± SD) vs. 0±0 and 40±1 vs. 16±4 colonies/dish, respectively, p

Description: Abstract 454 Multiple myeloma (MM) develops and expands in the bone marrow, and causes devastating bone destruction by enhancing osteoclastic bone resorption in their close vicinity. In MM bone lesions, thus induced osteoclasts (OCs) in turn enhance MM cell growth and survival, thereby forming a vicious cycle between the progression of bone destruction and MM tumor expansion. Such cellular interactions create an acidic milieu not only through acids produced by OCs but also through a large amount of lactate by proliferating tumor cells (Warburg effect). Reveromycin A (RM-A), a small microbial metabolite, preferentially induces cellular apoptosis in an acidic milieu, and draws considerable attention as a novel anti-resorptive agent. In the present study, we explored whether an acidic condition induced by MM-OC interaction affects MM expansion and whether RM-A targets not only OCs but also such an acidic microenvironment to regress tumor expansion in MM. INA6 and RPMI8226 MM cells potently enhanced osteoclastogenesis and osteoclastic pit formation when cocultured with rabbit bone marrow cells on bone slices. Notably, large multinucleated OCs were almost completely disappeared and pit formation on bone slices was abolished upon the treatment with RM-A at concentrations as low as 100nM. The cocultures with rabbit bone marrow cells stimulated INA6 MM cell growth; RM-A at 1microM was however able to substantially decrease the MM cell viability in the cocultures after 12 hours, although RM-A at this concentration did not affect MM cell growth when MM cells were cultured alone at pH7.4. The suppression of INA6 MM cell viability by RM-A was obviously more potent than that under bisphosphonate treatment in which mature OCs and pits on bone slices similarly decreased in number, suggesting that the anti-MM effects of RM-A is not merely due to depletion of mature OCs. Blockade of acid release by the proton pump inhibitor concanamycin A abolished such RM-A effects. Because an acidic microenvironment increases cell permeability of RM-A to cause apoptosis, it is plausible that a highly acidic milieu created by OC-MM interaction allows RM-A to act on nearby MM cells as well as OCs. In order to clarify a role of tumor acidity in RM-A-triggered cell death, we examined the effects of RM-A on MM cell growth upon acidification with lactic acid. When lactic acid was added to media to adjust their pH to be 7.0 and 6.75, the growth of INA6 and RPMI8226 MM cells was enhanced up to 150 and 120%, respectively, after 24 hours compared to that at pH7.4. However, RM-A at 1microM induced cell death in these MM cells at pH7.0 (60-70% reduction of alive MM cells compared to those at pH7.4) and at pH6.75 (〉90%), suggesting cytocidal effects of RM-A on lactate-producing MM cells densely proliferated in an acidic milieu. Because metoformin, anti-diabetic agent, up-regulates lactate production through stimulation of glycolysis, we next examined the effects of RM-A on MM cells in combination with metoformin. Metoformin dose-dependently enhanced lactate production by MM cells to decrease pH in their culture media over time; RM-A at 1microM showed potent cytotoxic effects on MM cells upon 24-hour preceded treatment with metoformin at 5 mM even when MM cells were started to be cultured at pH7.4, suggesting induction of anti-MM activity of RM-A with metoformin. Finally, in vivo RM-A effects were studied using INA6 MM cell-bearing SCID-rab mice. We injected RM-A sc at 4mg/kg twice daily for 18 days to the mice after confirming MM cell growth at 4 weeks after the MM cell inoculation. The RM-A treatment substantially decreased osteolytic lesions in X-ray and microCT images and MM tumor area in bone sections along with a reduction of INA6 cell-derived human soluble IL-6 receptor levels in mouse sera as a marker of MM tumor burden. These results collectively suggest that acidic microenvironment produced by MM-OC interaction enhances MM tumor progression but can trigger cytotoxic effects of RM-A on MM cells besides acid-producing OCs. RM-A may become a candidate for a novel therapeutic agent against MM with extensive bone resorption. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 1839 The implementation of hematopoietic stem cell transplantation and new agents such as bortezomib, thalidomide, and lenalidomide (Len) has dramatically improved survival in patients with multiple myeloma (MM). However, most MM patients eventually relapse after achieving of complete response. The existence of cancer stem cells is suggested to cause the relapse and considered as an important therapeutic target. We have demonstrated that HM1.24 (CD317) is selectively over-expressed on neoplastic plasma cells and that a defucosylated humanized monoclonal antibody (mAb) specific to HM1.24 (YB-AHM) is able to induce potent antibody-dependent cellular cytotoxicity (ADCC) against human MM cells in the presence of human effector cells. On the other hand, Len, an immunomodulatory drug, has been shown to activate NK cells to enhance their ADCC activity. Recently, we have reported that “side population (SP)” cells expelling a Hoechst 33342 dye represent a fraction with cancer stem cell-like property in MM cells. Moreover, we have found that MM cancer stem-like cells over-express pluripotency-associated transcription factors such as Sox2 and Nanog, and that these factors are useful for evaluating the potential of MM cancer stem-like cells. Here, we evaluated the efficacy of the combinatory therapy of YB-AHM and Len on MM cancer stem-like cells. We first examined the expression levels of the target molecule, HM1.24 on SP fraction of MM cells. Although SP cells expressed higher levels of ABC transporter ABCG2 compared with main population (MP) cells, HM1.24 was highly expressed in both SP and MP fractions in MM cells. We next examined the inhibitory effect of YB-AHM and Len on clonogenic activity of MM cell lines. RPMI 8226, U266, and OPM-2 cells were pre-incubated with YB-AHM (0.1 μg/ml) and Len (3 μM)-stimulated peripheral blood mononuclear cells (PBMCs) at an effector/target (E/T) ratio of 10 for 4 hours, and then were cultured into H4034 methylcellulose medium. Colony formation of MM cells was examined after 14 days. Treatment with YB-AHM and Len-stimulated PBMCs significantly suppressed the colony formation of MM cell lines compared with the no-treatment group (4±5 vs 62±2 colonies/well in RPMI 8226, p

Description: Although a therapeutic outcome with extended survival has been improved in patients with multiple myeloma (MM) owing to implementation new anti-MM agents, MM eventually relapses and still remains incurable. MM-initiating cells or MM progenitors are generally accepted to contribute to disease relapse through their drug-resistant nature. In order to overcome this drug resistance, various forms of immunotherapies have been studied. One such approach is the application of γδT cells. Vγ9Vδ2γδT (Vδ2 γδT) cells are important effectors in the first-line defense and tumor surveillance, and can be activated and expanded from peripheral blood mononuclear cells (PBMCs) upon treatment with zoledronic acid (Zol) and IL-2. New generation IMiDs, namely lenalidomide (LEN) and pomalidomide (POM), not only exert direct anti-MM actions, but also activate immune cells surrounding MM cells. In the present study, we aimed to clarify the effects of new generation IMiDs, LEN and POM, on activation and expansion of Vδ2 γδT cells in combination with Zol and the cytotoxic effects of ex-vivo expanded Vδ2 γδT cells against MM progenitors. After PBMCs isolated from normal donors were incubated for one week with Zol at 1 microM and LEN at 1 microM in combination, Vδ2 γδT cells were robustly expanded to be a majority of CD3-positive T cells as observed with Zol at 1 microM and IL-2 at 100 U/ml in combination. The expanded Vδ2 γδT cells expressed intracellular IFN-γ but not Foxp3 along with increased surface expression of NKG2D and DNAM-1, molecules known to be associated with NK or cytotoxic T cell-mediated cytotoxic activity, indicating induction of Th1-like Vδ2 γδT cells. In the same way as LEN, POM at 0.01, 0.1 and 1 microM was able to expand Vδ2 γδT cells from PBMCs isolated from all 10 normal donors. The expansion efficiency was slightly better with POM over LEN in some donors. The expansion efficiency with Zol plus these IMiDs appeared to be weaker in PBMCs from a majority of patients with MM than those from normal donors, as previously reported with Zol plus IL-2. Intriguingly, these IMiDs further expanded upon treatment with Zol plus IL-2 Vδ2 γδT cells from PBMCs from some donors. However, these IMiDs alone only marginally expanded Vδ2 γδT cells in the absence of Zol, suggesting costimulation of antigen-primed Vδ2 γδT cells by these IMiDs. Thus, LEN or POM appears to be able to expand Th1-like Vδ2 γδT cells in the presence of Zolwithout exogenous addition of IL-2. Such expanded Th1-like Vδ2 γδT cells exerted potent cytotoxic activity against MM cells. Interestingly, the expanded Vδ2 γδT cells also markedly minimized the sizes of side populations in RPMI8226 and KMS-11 cells, and suppressed their clonogenic capacity as determined by in vitro colony formation and tumorigenic capacity in SCID mice, suggesting targeting a drug-resistant clonogenic MM cells. Panobinostat, a pan-HDAC inhibitor, upregulated surface expression of NKG2D ligands, ULBPs and MICA/B, on MM cells, while impairing MM cells. Pretreatment with panobinostat may further potentiate the cytotoxic activity of Vδ2 γδT cells against MM cells. Combination with novel anti-MM agents warrants further study in terms of further enhancement of anti-MM effects with the ex-vivo expanded Th1-like Vδ2 γδT cells. The precise underlying mechanisms for ex-vivo expansion of Th1-like Vδ2 γδT cells should be clarified to make the best use of these new generation IMiDs with this unique and interesting phenomenon. Disclosures No relevant conflicts of interest to declare.

Description: Multiple myeloma (MM) is a heterogeneous clonal plasma cell proliferative disorder with CRAB features. Although survival of MM patients has been greatly prolonged by recent implementation of various combinatory treatments with novel anti-MM agents, MM still remains incurable. MM cells preferentially grow and expand in the bone marrow to elicit the alteration of gene expression and thereby drug resistance. To improve the therapeutic efficacy, we urgently need to develop novel treatment strategies targeting the BM microenvironment-mediated drug resistance. The serine/threonine kinase Pim-2 is constitutively over-expressed and acts as a pro-survival mediator in MM cells. We have reported that cocultures with bone marrow stromal cells (BMSCs) or osteoclasts (OCs) further up-regulate Pim-2 expression in MM cells to confer drug resistance (Leukemia 2011, 2015). Therefore, Pim-2 appears to be an important therapeutic target to impair the BM microenvironment-mediated drug resistance in MM. Histone deacetylases (HDACs) are generally accepted to be therapeutic targets for MM treatment. However, clinical application of currently available pan-HDAC inhibitors is limited with their adverse effects induced by a non-selective HDAC inhibition. To develop safe and effective HDAC inhibitor-based treatment, the therapeutic roles of HDAC isoform-specific inhibition should be elucidated. In this regard, we have recently reported therapeutic impacts on MM cells of inhibition of class-I HDACs, especially HDAC1 and HDAC3. HDAC3-selective inhibitor BG45 induces anti-MM activity in combination with DNA methyltransferase (DNMT) inhibitor azacytidine (Leukemia 2017). In the present study, we aimed to clarify the underlying mechanisms for impairment of MM cell growth and survival by HDAC1 inhibition. We first referenced the expression of class-I HDACs using a publicly available GSE6691 data set. Among class-I HDACs, HDAC1 and HDAC3 were highly expressed in MM cells. We then knockdowned HDAC1 gene using lentiviral shRNA system in MM cell lines. The HDAC1 gene silencing induced MM cell death with caspase-3 activation, indicating the critical role of HDAC1 in MM cell growth and survival. To determine target molecules of HDAC1, we carried out RNA-sequencing with and without the HDAC1 gene silencing in RPMI 8226 cells. Among genes whose expression significantly changed by the HDAC1 knockdown (adjusted P values 〈 0.05, log fold change 〉 0.5), we focused on IRF4 together with PIM2, because MM cell has been demonstrated to addict to aberrant IRF4-c-Myc regulatory network (Nature 2008). Downregulation of IRF4 and Pim-2 by the HDAC1 knockdown was further confirmed by quantitative PCR (Q-PCR) and immunoblotting in RPMI 8226 and MM.1S cells. Treatment with the class I HDAC-selective inhibitor MS-275 (entinostat) also induced MM cell death along with reduction of IRF4 and Pim-2 expression. Since previous study has shown that IRF4 binds to PIM2 promoter in MM cells (Nature 2008), we examined whether IRF4 regulates PIM2 expression. We found that IRF4 binds to the PIM2 promoter region by analyzing ChIP-Seq data in KMS-12 cells (GSE22901). We further confirmed the binding of IRF4 on PIM2 promoter by ChIP-Q-PCR. Indeed, the IRF4 knockdown downregulated Pim-2 expression in RPMI 8226 cells. These results suggest that HDAC1 inhibition downregulates IRF4 expression, thereby transcriptionally reducing PIM2 expression in MM cells. Pim-2 expression can also be augmented by multiple signaling pathways, including HIF-1a, JAK-STAT and NF-kB-mediated ones in MM cells through the interaction with BM microenvironment. Interestingly, the Pim inhibitor SMI-16a and MS-275 cooperatively induced apoptotic cell death in MM cell lines and CD138-positive primary MM cells even in the presence of BMSCs. Taken together, our results demonstrate the critical role of the HDAC1-IRF4-Pim-2 axis in MM cell growth and survival, and provoke the novel treatment strategy targeting the HDAC1-IRF4-Pim-2 axis in MM cells. Disclosures Anderson: Takeda Millennium: Consultancy; Gilead: Membership on an entity's Board of Directors or advisory committees; Oncopep: Equity Ownership; C4 Therapeutics: Equity Ownership; Celgene: Consultancy; Bristol Myers Squibb: Consultancy.

Description: Abstract 3025 Multiple myeloma (MM) remains incurable by the present therapeutic modalities, leading to the idea to develop various forms of immunotherapy. γδT cells are important effectors in the first-line defense against infections and tumors, and play a critical role in host defense and tumor surveillance. Aminobisphosphonates can effectively expand human γδT cells in combination with IL-2. Because expanded γδT cells have been demonstrated to exert potent anti-MM effects in vitro, γδT cell-based immunotherapy has attracted considerable attention as a novel treatment against MM. However, the clinical efficacy of γδT cell-based immunotherapy has been limited in patients with MM. We previously demonstrated that bone marrow stromal cells (BMSCs) blunt γδT cell activity and thus protect MM cells. To improve the therapeutic efficacy of γδT cells in vivo, therefore, we need to develop a maneuver to restore the anti-MM activity of γδT cells in the bone marrow microenvironment in which MM cells reside. Lenalidomide (LEN), a novel immunomodulatory anti-MM agent, shows pivotal anti-MM activity by targeting immune cells as well as MM cells and their surrounding cells in the bone marrow microenvironment. The present study was undertaken to further clarify the role of the bone marrow microenvironment in MM in the cytotoxic activity of γδT cells against MM cells and to determine whether LEN is able to modulate γδT cell activity and restore its anti-MM effects in the bone marrow. When γδT cells expanded with zoledronic acid (Zol) and IL-2 were cocultured with peripheral blood mononuclear cells (PBMC)-derived OCs or MM cell lines, γδT cells almost completely destroyed both types of cells, suggesting the susceptibility of OCs as well as MM cells to γδT cells. However, the cytotoxic activity of γδT cells against MM cells was markedly attenuated in the presence of BMSCs but not OCs. γδT cells expanded with Zol and IL-2 up-regulated their surface expression of LFA-1, CD26, and the molecules associated with NK or cytotoxic T cell activation including NKG2D and DNAX accessory molecule-1 (DNAM-1; CD226). Among these molecules, stromal cells markedly down-regulated DNAM-1 expression along with a decrease in interferon-γ production by γδT cells. DNAM-1 is regarded as an essential factor in NK cell-mediated cytotoxicity against various cancers. MM cells constitutively expressed DNAM-1 ligands, poliovirus receptor (CD155) and nectin-2; treatment with a blocking antibody against DNAM-1 substantially reduced the cytotoxic activity of γδT cells against MM cells, suggesting a critical role of DNAM-1 in the anti-MM effects of γδT cells. Interestingly, when γδT cells were cocultured with BMSCs in the presence of clinically relevant doses of LEN (1 microM) and Zol (0.1-1 microM), the suppression of interferon-γ production and DNAM-1 expression of γδT cells by BMSCs were substantially reduced. Moreover, such γδT cells exerted potent anti-MM effects, suggesting antagonism of the suppressive effects of BMSCs on γδT cell activity and its anti-MM effects. LEN (0.1-10 microM) and Zol (0.1-5 microM) in combination were able to substantially expand γδT cells from PBMCs in the absence of IL-2. These expanded γδT cells expressed the activation markers LFA-1, CD26, NKG2D and DNAM-1 at levels similar to those in γδT cells expanded by Zol and IL-2. However, LEN alone did not show any significant effects on γδT cell expansion and activation, suggesting a costimulatory role of LEN on Zol-primed γδT cells. These results collectively suggest that MM cells evade attack by γδT cells in the bone marrow through attenuation of γδT cell activity by BMSCs, and that LEN and Zol in combination may restore γδT cell activity and its anti-MM effects to improve the efficacy of γδT cell-based immunotherapy against MM. Disclosures: No relevant conflicts of interest to declare.