ALBERT

Description: Allogeneic hematopoetic stem cell transplantation (HSCT) is the treatment of choice for a variety of hematologic malignancies. Graft-versus-Host disease (GvHD) is a key contributor to treatment related morbidity and mortality and consequently limits the efficacy of allogeneic HSCT. Interleukin 10 (IL-10) is a well-known cytokine with immunoregulatory and anti-inflammatory properties, also important in context of GvHD. B cells have been described as potent IL-10 producers in various situations. Here we show how host as well as donor derived B cells contribute to GvHD amelioration through IL-10 production. We address the role of IL-10 in GvHD in an acute murine MHC mismatch model: Mice on a C57BL/6 background received bone marrow and CD90+ T cells from mice on a BALB/c background or vice versa. Transplantation experiments with IL-10 deficient donor or host cells clearly show the importance of donor derived IL-10 in general. To further dissect the cells contributing to IL-10 production in this situation we employed an IL-10 knock-in reporter mouse in which expression of eGFP is under control of the Il-10 locus. Lethal irradiation as used in the conditioning regiment before transplantation revealed B cells as major contributors of host derived IL-10. In addition, transfer of cells from reporter mice into preconditioned recipients showed also donor B cells as contributors to IL-10 production. A phenotypical characterization of the eGFP+ B cells exhibited a CD1d+TIM-1+CD5int phenotype, in line with immunoregulatory B cells. To finally confirm the relevance of B cells derived IL-10 in GvHD, we employed B6.B-IL-10-/- mice that have a B cell specific IL-10 knock-out as donors or recipients. Here we found a reduced survival associated with the incapability of the B cells to produce IL-10 in both cases. Taken together, our results provide new insights in the mechanisms and the variety of cells contributing to the course of GvHD. An improved understanding of these aspects might help to pave the way for new treatment options to overcome current limitations of allogenic HSCT. Disclosures: No relevant conflicts of interest to declare.

Description: Ibrutinib (IBR) is a small molecule inhibitor of Bruton's Tyrosine Kinase, a key component of the BCR pathway. In phase II studies of single-agent IBR in CLL and MCL, the overall response rate was 89% and 68% respectively. However, only a minority of responses were complete and few are durable, suggesting that drug combinations will be required to achieve broader and more durable responses. In a previous report from our laboratory, Venetoclax (VEN) (GDC-0199; ABT-199), an inhibitor of the anti-apoptotic protein BCL2, showed synergistic cytotoxicity with IBR in MCL cell lines (Axelrod M et al, Leukemia 2014). A clinical trial testing this combination for MCL patients is now open (NCT02419560). Here, we sought to investigate the cytotoxic effects of the IBR and VEN combination in CLL and MCL patient samples and the mechanisms of resistance to drug treatment that might be expected to occur in the tissue microenvironment of these malignancies in patients. Peripheral blood mononuclear cells (PBMCs) from CLL/MCL patients containing circulating neoplastic B cells were incubated with IBR (0.1 μM), VEN (25 nM), and the combination of IBR and VEN for different time intervals. Cytotoxicity following drug treatment was determined by FACS analysis of cleaved PARP in CD5/CD19 double positive cells (Portell CA et al ASH 2014) and loss of CD5/CD19 double positive cells (i.e., neoplastic B cells). Cytotoxicity induced by IBR was variable, but was invariably enhanced by the addition of VEN. In a cohort of CLL and MCL patient samples that were treated (n=14), eleven patients showed synergistic cytotoxicity following treatment with IBR and VEN as evaluated by the Bliss model of independence. Recognizing that CLL and MCL cells populate replication centers (pseudofollicles) of the tissue microenvironment such as lymph node and bone marrow, we attempted to emulate a pseudofollicle in vitro by co-culturing or pre-incubating PBMC with agonistic components of that microenvironment. We screened several cytokines, other agonistic proteins, and supporting cells, including stromal cells (HS-5 cell line), follicular dendritic cells (HK cell line), soluble CD40L, CxCL13, IL-10, IL-2, CpG oligodeoxynucleotides (CpG ODN), BAFF, or IgM. A 12 h pre-incubation with soluble CD40L, IL-10, or CpG ODN generated significant levels of resistance to IBR and VEN. Moreover, the combination of soluble CD40L, IL-10, and CpG ODN resulted in nearly complete resistance to the IBR-VEN combination in cells from three CLL patient samples. The resistance was not overcome by increasing concentrations of IBR up to 10 μM, suggesting that resistance is not due to upregulation of BTK. To gain insight into the mechanism(s) of resistance and ways to overcome it, CLL patient PBMC pre-treated with soluble CD40L, IL-10, and CpG ODN were treated in three-way combination with IBR, VEN and inhibitors of potential resistance pathways known to be downstream from the tested ligands. These included small molecule inhibitors of IKKα and IKKβ, JAK1, 2, and 3, MAP Kinase, or PI3Kδ as well as the anti-apoptotic proteins MCL1, BCL-xL, or Survivin. Cytotoxicity was evaluated with an alamarBlue® assay. The NF-kB, JAK-STAT, PI3Kδ, and anti-apoptotic protein inhibitors all exhibited synergistic interactions with the IBR-VEN combination, as well as substantial cytotoxicity as single agents at higher doses. Inhibitors of the MAP Kinase pathway were ineffective. Taken together our data are consistent with the hypothesis that resistance to IBR, VEN and the combination in patients with CLL or MCL could arise by NF-kB, JAK-STAT, or PI3K signaling from CD40, IL10R and Toll-like receptors. In conclusion, IBR and VEN induced synergistic cytotoxicity in a majority of CLL and MCL patient samples. An in vitro model of the pseudofollicle microenvironment provided protection against cytotoxicity of IBR and VEN, suggesting that the tissue microenvironment provides a niche that supports drug resistance. This resistance was overcome by inhibition of NF-kB, JAK-STAT, or PI3Kδ, or the anti-apoptotic proteins MCL1, BCL-xL, or Survivin. Disclosures Off Label Use: Ibrutinib and Venitoclax drug combination. This drug combination was used in this study to evaluate cytotoxicity in CLL and MCL ex vivo patient samples. . Portell:AbbVie: Research Funding. Williams:Genentech: Other: Research funding to my institution; Celgene: Consultancy, Other: Research funding to my institution; Takeda: Consultancy, Other: Research Funding to my institution. Petricoin:Theranostics Health: Consultancy, Equity Ownership, Patents & Royalties.

Description: Lymphomas are assumed to originate at different stages of lymphocyte development through chromosomal aberrations. Thus, different lymphomas resemble lymphocytes at distinct differentiation stages and show characteristic morphologic, genetic, and transcriptional features. Here, we have performed a microarray-based DNA methylation profiling of 83 mature aggressive B-cell non-Hodgkin lymphomas (maB-NHLs) characterized for their morphologic, genetic, and transcriptional features, including molecular Burkitt lymphomas and diffuse large B-cell lymphomas. Hierarchic clustering indicated that methylation patterns in maB-NHLs were not strictly associated with morphologic, genetic, or transcriptional features. By supervised analyses, we identified 56 genes de novo methylated in all lymphoma subtypes studied and 22 methylated in a lymphoma subtype–specific manner. Remarkably, the group of genes de novo methylated in all lymphoma subtypes was significantly enriched for polycomb targets in embryonic stem cells. De novo methylated genes in all maB-NHLs studied were expressed at low levels in lymphomas and normal hematopoietic tissues but not in nonhematopoietic tissues. These findings, especially the enrichment for polycomb targets in stem cells, indicate that maB-NHLs with different morphologic, genetic, and transcriptional background share a similar stem cell–like epigenetic pattern. This suggests that maB-NHLs originate from cells with stem cell features or that stemness was acquired during lymphomagenesis by epigenetic remodeling.

Description: Responses to single-agent ibrutinib (IBR) or venetoclax (VEN) in Chronic Lymphocytic Leukemia (CLL) or Mantle Cell Lymphoma (MCL) are often incomplete suggesting drug combinations are needed to overcome de novo and acquired resistance. We previously reported synergistic cytotoxicity for the IBR+VEN combination in CLL/MCL cells ex vivo and have initiated an IBR+VEN clinical trial in MCL (NCT02419560). However, ex vivo analysis of patient samples showed de novo resistance even to the IBR+VEN combination in some samples, consistent with recent clinical experience (Tam et al. N Engl J Med, 2018). We noted that sensitivity to IBR+VEN was lower in CLL cells showing an "activation" phenotype (CD5+/19+/69+) that can be induced by extracellular factors in vivo, suggesting that the microenvironment could induce drug tolerance in the cancer cells.This was supported by our finding that ex vivo exposure of CLL/MCL samples to a mixture of microenvironmental agonists (CpG-ODN, sCD40L, and IL10; "agonist mix") induced significant loss of sensitivity to IBR+VEN (Jayappa et al. Blood Adv., 2017). Here we show that various microenvironmental agonists, including of innate immunity, are able to reduce sensitivity to IBR+VEN as well as other pro-apoptotic drugs, generating a multi-drug tolerant phenotype. To explore drug tolerance, we performed flow cytometric analysis of apoptosis induced by IBR+VEN in CD5+/19+cancer cells in CLL/MCL PBMCs cultured with agonists of TLRs, NOD1/2, CD40, and IL10R and various stromal cells. Tolerance to IBR+VEN as determined by resistance to apoptosis was noted in most samples cultured with TLR9 agonist CpG-ODN, sCD40L, or Jurkat cells expressing CD40L. IL10 and HUVEC induced modest levels of drug tolerance in a few CLL/MCL samples, and TLR1/2, TLR7, and NOD1/2 agonists were effective only in MCL samples. Prior exposure to CpG-ODN enhanced the ability of sCD40L to induce proliferation and drug tolerance and vice versa in CLL/MCL, predicting mutually reinforcing interactions in vivo. We noted that CLL cells exposed to CpG-ODN displayed higher levels of CD40 and downstream signaling proteins (TRAF2, RelB, and p100/52). Conversely, CD40L induced a modest increase in NF-kB transcription factors, providing a possible mechanism for mutual reinforcement. Using flow cytometric analysis of apoptosis, we noted tolerance to several pro-apoptotic agents (bendamustine, fludarabine, and inhibitors of Mcl-1, Bcl-xL, Bcl-2, and Bcl-2/Bcl-xL) including IBR+VEN in CD5+/19+cells from CLL samples treated with agonist mix, showing development of multi-drug tolerance in these cells. CpG-ODN or agonist mix induced NFkB-dependent over-expression of pro-survival proteins (Mcl-1 and Bcl-xL), and increased ratio of these to cognate pro-apoptotic proteins. This increased expression of pro-survival proteins underlies the multi-drug tolerant phenotype. Consistently, multi-drug tolerance was rescued with inhibitors of NFkB (BMS345541 or Bortezomib) or drug combinations simultaneously inhibiting multiple anti-apoptotic proteins (inhibition of Mcl-1 with Bcl-2, Bcl-xL, or Bcl-2 and Bcl-xL). CLL cells with activation phenotype (CD5+/19+/69+) in patient PBMCs cultured without the agonist mix also showed tolerance to several apoptotic protein inhibitors ex vivo, and this was also effectively rescued with NF-kB inhibition or combination of apoptotic protein inhibitors ex vivo. These results suggest that the cancer cells with activation phenotype exist in vivo anddisplay multi-drug tolerant phenotype consistent with the drug tolerance induced by agonist mix ex vivo. In conclusion, several microenvironmental factors, particularly agonists of TLR9 and CD40, induce tolerance to IBR+VEN and cell proliferation in CLL/MCL, and response to these agonists is enhanced by combinatorial exposure. These agonists generate tolerance to several apoptosis-inducing agents beyond IBR+VEN. This microenvironment-induced multi-drug tolerance is mediated by NFkB dependent over-expression of multiple pro-survival proteins. Inhibitors of NFkB, or drug combinations targeting multiple pro-survival proteins, overcame multi-drug tolerance in agonist mix-treated samples and in CD69+CLL cells in patient PBMCs that represent multi-drug tolerant cells in vivo. Thus, more complete and durable responses might be achieved in MCL/CLL with therapies that target multi-drug tolerant persister cells. Disclosures Williams: Celgene: Consultancy, Research Funding; Kite: Consultancy; Juno: Consultancy; Seattle Genetics: Consultancy; Gilead: Consultancy, Research Funding; Novartis: Research Funding; TG Therapeutics: Consultancy; Sandoz: Consultancy; Astra-Zeneca: Consultancy; Abbvie: Consultancy; Takeda: Research Funding; Pharmacyclics: Research Funding; Janssen: Consultancy, Honoraria, Research Funding; Verastem: Consultancy. Portell:AbbVie: Research Funding; Infinity: Research Funding; Genentech/Roche: Consultancy, Research Funding; Acerta: Research Funding; BeiGene: Research Funding; Kite: Research Funding; Amgen: Consultancy; TG therapeutics: Research Funding.

Description: Ibrutinib (IBR), an inhibitor of Bruton's Tyrosine Kinase (BTK), has been FDA approved for Chronic Lymphocytic Leukemia (CLL) and Mantle Cell Lymphoma (MCL). However, IBR responses are incomplete in most cases, and of short duration for MCL and higher-risk CLL subtypes. We hypothesize that intrinsic resistance, incomplete responses, and rapid recurrence can be due to adaptive signaling that should be co-targeted with BTK inhibition to achieve deeper and more durable responses. We previously showed that venetoclax (VEN), an inhibitor of Bcl-2, generated synergistic cytotoxicity with IBR in MCL lines as well as circulating leukemic B cells in 13/19 CLL and 4/5 MCL patient samples treated ex vivo (Jayappa KD et al. ASH 2015; Portell CA et al. ASH 2014). However, the sensitivity to VEN or IBR+VEN was highly variable among patient samples, and did not correlate with the diagnostic genetic lesions in the CLL/MCL cells or clinical characteristics of the patients. Here, we show that resistance to IBR+VEN can be induced by interaction with soluble factors found in the in vivo "macroenvironment" of the circulating cancer cells. To gain insight into changes in signaling pathways that might underlie mechanisms of drug synergy and resistance, we analyzed drug resistant MCL lines treated with IBR, VEN, and the combination by Reverse Phase Protein Arrays. We observed downregulation of PI3K-Akt, MAPK, JAK-STAT, and NOTCH signaling after IBR and IBR+VEN treatment, cleavage of caspases and PARP after VEN-treatment, and greatly enhanced cleavage of caspases and PARP after IBR+VEN treatment. A notable exception was significantly increased phosphorylation on p65 S536 of the NF-kB pathway at longer times after IBR+VEN treatment, indicating a possible role of NF-kB signalling in generating resistance to IBR and VEN in cells that survived treatment. To determine whether extrinsic factors in the cancer cell environment might be able to induce a drug-resistant phenotype, we co-cultured or pre-incubated PBMC with a diverse panel of stromal cells, cytokines, and other agonists. We found that the combination of soluble CD40L, IL-10, and CpG DNA (agonist mix) generated nearly complete resistance to IBR+VEN in CLL and MCL patient samples, with CpG DNA being the most effective single agent. Every sample treated with agonist mix displayed increased resistance to IBR+VEN drug combination, suggesting that responsiveness transcends genetic heterogeneity and reflects the developmental lineage of the cancer cells. We investigated whether the extrinsic agonists induce drug resistance by activating the NF-κB pathway, by analyzing nuclear localization of NF-kB transcription factors RelA and RelB using ImageStream and cell fractionation. We observed robust activation of alternative-NF-kB signaling in primary CLL and MCL cells treated with agonist mix, with little effect on canonical NF-κB. PKC, MAPK and PI3K-Akt signaling also showed evidence of activation by agonist mix. Agonist mix treatment also caused significant overexpression of anti-apoptotic proteins Mcl-1, Bcl-xL, and survivin, but not Bcl-2. Inhibitors of NF-κB blocked RelB translocation and overexpression of Mcl-1, Bcl-xL and survivin. Inhibitors of NF-kB or of upregulated anti-apoptotic proteins overcame drug resistance induced by agonist mix. Inhibitors of the other activated pathways (MAPK, PI3K-Akt, PKC) did not block agonist-induced drug resistance at pharmacologically relevant concentrations. To determine whether extra-nodal agonists in the blood of patients could generate resistance to IBR and VEN, we analyzed drug cytotoxicity in CLL patient PBMCs cultured with autologous plasma. We found that autologous plasma was able to induce increased resistance to IBR+VEN in 3/7 CLL samples and this resistance was blocked by treatment with an NF-kB inhibitor. In conclusion, soluble agonists in the patient macroenvironment of circulating CLL/MCL cells can generate resistance to IBR+VEN by activating alternative-NF-kB signaling and over-expression of multiple anti-apoptotic proteins. Inhibitors of NF-kB or of the upregulated anti-apoptotic proteins overcame IBR+VEN resistance generated by these extrinsic factors. We suggest that survival signals generated by extra-nodal agonists in the patient macroenvironment generate drug resistance in CLL and MCL, and that improved responses could occur with interventions that block these survival signals. Disclosures Portell: AbbVie: Research Funding; Roche/Genentech: Research Funding; Infinity: Research Funding; Acerta: Research Funding. Wulfkuhle:Theranostics Health, LLC: Equity Ownership. Petricoin:Perthera, Inc.: Consultancy, Equity Ownership; Ceres Nanosciences, Inc.: Consultancy, Equity Ownership, Patents & Royalties; Avant Diagnostics, Inc.: Equity Ownership. Williams:University of Virginia: Employment; Janssen and Pharmacyclics: Research Funding.

Description: Introduction Mantle cell lymphoma (MCL) which relapses or becomes refractory to frontline chemotherapy can be a clinical challenge. There have been several targeted agents approved in relapsed MCL including bortezomib, lenalidomide and ibrutinib (IBR) with the best single-agent responses seen with IBR. IBR is an oral, Bruton tyrosine kinase (BTK) inhibitor which has an overall response rate (ORR) of 67% with a median duration of response of 17.5 months in relapsed MCL (Wang, Blood 2015). While these responses are impressive in this population, only 1/3 of patients will have a complete response and only 1/3 of responding patients will have a 24 month PFS. Thus, improvements are needed. Venetoclax (VEN) is an oral selective BCL2 inhibitor which is currently FDA approved in relapsed 17p-deleted chronic lymphocytic leukemia. We and others have shown synergistic cytotoxicity with VEN and IBR (Axelrod Leukemia 2014) which prompted us to explore the combination in a Phase I/Ib clinical trial (clinicaltrials.gov ID: NCT02419560). This study was supported by a grant from AbbVie Inc. Methods Given overlapping toxicities with VEN and IBR, namely neutropenia and GI toxicities, potential for drug-drug interactions given both are metabolized by CYP3A, and a wide range of therapeutic dosing for the two drugs, a dose finding study is appropriate. A continual-reassessment model was designed to test six dosing strategies (table 1). Subjects start treatment with single agent VEN at 100mg PO daily and increase to the allocated dose per table 1. After 1 week of VEN, subjects start the allocated dose of IBR. Subjects are monitored closely for tumor lysis syndrome (TLS) and hospitalized for TLS monitoring when starting IBR. Subjects are treated with the combination for 6 months and are encouraged to continue IBR after that time. The study enrolls in 2 stages. In the first stage, subjects are enrolled one at a time to sequential arms. Up to 2 subjects are allowed on each arm in a zone before enrollment in the zone is paused. Subsequent zones are enrolled once at least one subject in every arm of the zone does not have a DLT during the DLT window. The second stage begins when a subject has a DLT or all arms have enrolled at least 1 subject. In the second stage, subsequent subjects are allocated to an arm based on DLT's and ORR at 2 months occurring in prior patients on the study, thus the study aims to find the optimal dosing combination of IBR and VEN for both toxicity and response. Enrollment will continue until 10 subjects are allocated to an arm or 28 total subjects are enrolled. Eligible patients must have documented relapsed MCL after at least 1 line of chemotherapy. Subjects must not have bulky disease, no evidence of TLS, and must not have been previously treated with IBR. Results Enrollment began 10/2015 and at the time of submission, we have treated 8 subjects and have finished stage I of the study. Subjects were enrolled on arms A to E. Mean age is 63 years (range 49-81). 7 of the 8 subjects are male. 5 subjects were refractory to their prior treatment and 3 subjects have progressed after an autologous bone marrow transplant. Seven of the 8 subjects are evaluable for adverse events. 5 subjects have completed the 6-week DLT window. There have been 15 related adverse events reported with 14 of these being grade 1 or 2. No TLS has been reported. One DLT at arm E was identified (grade 4 neutropenia) which prompted us to move to stage II of the study. Three subjects (arm A, B, and C) are evaluable for response with all achieving at least a partial response. One subject on arm C, had a complete response at 4 months of the combination. Conclusion: Early results suggest tolerability for the combination of IBR and VEN in Relapsed MCL. There have been no signs of TLS, though subjects with high risk for TLS are excluded. One DLT (neutropenia) has been reported in Arm E and thus modeling will start to find the optimal combination using both toxicity and response. Responses have been seen across the various treatment arms. Continual re-assessment modeling is an adequate study design for combination studies with targeted agents to identify optimal dosing, accounting for both toxicity and response. Disclosures Portell: Infinity: Research Funding; Roche/Genentech: Research Funding; Acerta: Research Funding; AbbVie: Research Funding. Chen:Seattle Genetics: Consultancy, Honoraria, Research Funding, Speakers Bureau; Millenium: Consultancy, Research Funding, Speakers Bureau; Genentech: Consultancy, Speakers Bureau; Merck: Consultancy, Research Funding. Cohen:Millennium/Takeda: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Infinity: Consultancy, Membership on an entity's Board of Directors or advisory committees; Seattle Genetics: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Pharmacyclics: Consultancy, Membership on an entity's Board of Directors or advisory committees; Bristol-Myers Squibb: Research Funding; Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees. Kahl:Seattle Genetics: Consultancy; Celgene: Consultancy; Infinity: Consultancy; Gilead: Consultancy; Juno: Consultancy; Pharmacyclics: Consultancy. Williams:Jansen: Research Funding; Pharmacyclics: Research Funding.