Chronic Lymphocytic Leukemia Lymph Nodes Research Articles

Metabolic reprogramming in the CLL TME; potential for new therapeutic targets

Chronic lymphocytic leukemia (CLL) cells circulate between peripheral (PB) blood and lymph node (LN) compartments, and strictly depend on microenvironmental factors for proliferation, survival and drug resistance. All cancer cells display metabolic reprogramming and CLL is no exception – though the inert status of the PB CLL cells has hampered detailed insight into these processes. We summarize previous work on reactive oxygen species (ROS), oxidative stress, and hypoxia, as well as the important roles of Myc, and PI3K/Akt/mTor pathways. In vitro co-culture systems and gene expression analyses have provided a partial picture of CLL LN metabolism. New broad omics techniques allow to obtain molecular and also single-cell level understanding of CLL plasticity and metabolic reprogramming. We summarize recent developments and describe the new concept of glutamine addiction for CLL, which may hold therapeutic promise.

Investigation into Intraclonal Heterogeneity of CXCR4 DimCD5 Bright Chronic Lymphocytic Leukemia Cells Identifies Distinct Activation Signatures

Chronic Lymphocytic Leukemia (CLL) is a heterogeneous disease associated with diverse clinical and molecular profiles, suggesting long-term clonal evolution driven by genetic diversification from a small, often undetected, subset of leukemic cells endowed with superior growth capacities. More importantly, this diversification process enhances the opportunity of CLL subclones to survive selective pressure, such as BTK or PLCG2 mutations that can abolish the effect of BTK inhibitors. While this evolutionary process is inevitable in leukemic cells, therapeutic intervention targeting the cells that have just replicated their DNA, and hence could have accumulated new genomic mutation, would be an effective strategy to significantly delay or abort the emergence of deleterious mutations. In CLL, a series of studies involving consumption of 2H 2O determined that CXCR4 DimCD5 Bright cells (proliferative fraction, PF) of CLL clones contain most of the recently divided cells, identified by the presence of 2H-DNA. This conclusion was supported by finding elevated numbers of Ki67 + and MCM6 + cells in the same fraction, when compared to other CLL cells. Furthermore, PF is superior in: lymphoid tissue homing in xenograft models; activating T cells; the capacity to alter tumor microenvironment (TME) by Th2 skewing; and expression of AICDA which can increase the likelihood of off-target genomic mutations. To permit a focused analysis of PF that would allow us to intercept this evolutionary loop, we characterized the single cell transcriptomes of PF from 4 U-CLL and 4 M-CLL patients immediately prior to ibrutinib treatment. Consistent with previous reports, the majority of PF (herein APC-PF) cells express certain antigen presentation molecules with a gene signature suggestive of T cell regulation and cell-cell interactions. Additionally, we discovered a small subpopulation with an upregulation of MIR155HG, HSP90AB, CCL3, CCL4, MYC and a concomitant MYC activation signature (herein MYC-PF). Since miR-155 expression can be induced by CD40L or B-cell-activating factors, this PF subset may reflect prior BCR ligation and/or T cell co-stimulation. However, the strongest BCR activation signature was found in another PF subpopulation (herein BCR-PF) with an enrichment of germinal center-related genes (e.g., BACH2, PAX5, FOXP1) and CLL-associated molecules ( BCL2, ROR1, MSI2, PLCG2). Furthermore, the BCR-PF cells are least likely to be resting (in G1/G0) compared to others. Collectively, these findings suggest effective BCR engagement, either from the TME or cell-autonomously. Corroborating these findings with our previous studies documenting the CXCR4/CD5 fractions and the expression of protein proliferation markers, our transcriptomic analyses suggest that the BCR-PF may be the most immediate emigrant from the proliferation centers (PC) within CLL lymph nodes (LN). To probe this possibility, spatial transcriptomic analysis was performed on FFPE LN tissues from surgical biopsies from 3 U-CLL and 1 M-CLL patients. Single cell deconvolution was used to determine the relative abundance of cells within the anatomically defined PCs that resemble the three PF subpopulations transcriptomically. Our results suggest that PCs are more enriched in cells resembling BCR-PF, followed by cells bearing the APC-PF signature. To extend on these observations, we have documented that the unique surface markers expressed in these PF subfractions positively correlate with our transcriptomic results. Taken together, these results not only highlighted the molecular heterogeneity of CLL, but also lead to the identification of two PF subpopulations that may provide novel actionable therapeutic targets to significantly halt CLL clonal evolution.

In Vitro 3D Spheroid Culture System Displays Sustained T Cell-dependent CLL Proliferation and Survival.

Chronic lymphocytic leukemia (CLL) cells are highly dependent on microenvironmental cells and signals. The lymph node (LN) is the critical site of in vivo CLL proliferation and development of resistance to both chemotherapy and targeted agents. We present a new model that incorporates key aspects of the CLL LN, which enables investigation of CLL cells in the context of a protective niche. We describe a three-dimensional (3D) in vitro culture system using ultra-low attachment plates to create spheroids of CLL cells derived from peripheral blood. Starting from CLL:T cell ratios as observed in LN samples, CLL activation was induced by either direct stimulation and/or indirectly via T cells. Compared with two-dimensional cultures, 3D cultures promoted CLL proliferation in a T cell-dependent manner, and enabled expansion for up to 7 weeks, including the formation of follicle-like structures after several weeks of culture. This model enables high-throughput drug screening, of which we describe response to Btk inhibition, venetoclax resistance, and T cell-mediated cytotoxicity as examples. In summary, we present the first LN-mimicking in vitro 3D culture for primary CLL, which enables readouts such as real-time drug screens, kinetic growth assays, and spatial localization. This is the first in vitro CLL system that allows testing of response and resistance to venetoclax and Bruton's tyrosine kinase inhibitors in the context of the tumor microenvironment, thereby opening up new possibilities for clinically useful applications.

Single-Cell Omics Analyses Identify the TIM-3 Ligand Galectin-9 As Novel Immunotherapy Target for Chronic Lymphocytic Leukemia

Failure of response to immunotherapy including checkpoint inhibitors or CAR-T-cell therapy in chronic lymphocytic leukemia (CLL) has been linked to dysfunctional effector T cells. Using single-cell analyses, we performed an in-depth characterization of the T-cell compartment in blood and tissue samples of CLL patients and mouse models to gain insights into the spectrum of phenotypes and transcriptional programs of T cells and the underlying mechanisms of their development. By mass cytometry (CyTOF) using 35 antibodies, we characterized T cells in blood (n=8), bone marrow (n=3), and lymph nodes (n=21) of CLL patients, as well as reactive lymph nodes of non-tumor patients (n=13). Integrative analyses of all data sets allowed us to identify and quantify 15 clusters of CD4+ and 14 clusters of CD8+ T cells. First, our data showed that T cells in blood and bone marrow are similar, but clearly distinct from lymph node derived cells. Second, we observed an accumulation of several regulatory T-cell subsets, as well as T cells harboring an exhausted and dysfunctional phenotype in the lymph node samples, and a positively correlated abundance of these T-cell subsets. Third, an increased frequency of regulatory and exhausted T cells was detected in CLL in comparison to non-malignant lymph nodes. We further performed single-cell RNA-sequencing generating transcriptome and T-cell receptor (TCR) data of T cells from lymph nodes of CLL patients (n=5) and spleen samples of the Eµ-TCL1 mouse model (n=3). These data confirmed the presence of several exhausted T-cell clusters in CLL lymph nodes and spleens of Eµ-TCL1 mice, and allowed for the identification and transcriptional characterization of terminally exhausted T cells and their precursor state (Figure A). By integrating single-cell TCR data, a clonal expansion mainly of the precursor exhausted T cells was observed, suggesting their reactivity for CLL cells. Finally, we used the single-cell transcriptome data for interactome analyses and identified both known and novel ligand-receptor-interactions between CLL and different T-cell clusters with a presumable function in survival and growth support of CLL cells, but also in suppression of T cells. Among the latter, we focused on galectin-9 which is expressed by CLL cells in patients and mice, and known as ligand for the immunoregulatory receptor TIM-3. We treated mice that had developed CLL after injection of Eµ-TCL1 leukemic cells with galectin-9 blocking antibodies and showed that this treatment slowed down disease development (Figure B). Altogether, our study provides a detailed characterization of the T-cell compartment in CLL that helps to understand T-cell exhaustion and suggests the TIM-3 ligand galectin-9 as novel target for immunotherapy in CLL. MZ, EM, JP and MS are co-senior authors of this study. Figure: A) UMAP of single-cell transcriptome data of T cells from lymph node samples of CLL patients (n=5) displaying a precursor exhaustion score described by Guo et al. in Nat.Med. in 2018. B) Mice that have developed CLL-like disease after adoptive transfer (AT) of Eµ-TCL1 tumor cells were treated with blocking anti-galectin-9 (α-GAL9) or isotype antibodies. CLL cell counts were monitored in blood over time. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

P607: A NOVEL LYMPH NODE-MIMICKING 3D CULTURE SYSTEM DISPLAYS LONG-TERM T CELL-DEPENDENT CLL PROLIFERATION AND SURVIVAL

Background: Primary chronic lymphocytic leukemia (CLL) cells, despite originating from a proliferative disease, rapidly undergo apoptosis in vitro in the absence of microenvironmental survival signals1. No current system permits long-term expansion of CLL cells in vitro due to difficulties of mimicking a physiological microenvironment. The lymph node (LN) is the critical site of in vivo CLL proliferation and is also thought to play a key role in the development of resistance to targeted agents2,3. Developing an in vitro culture system for CLL with pathophysiological relevance will greatly benefit studies of CLL proliferation, microenvironment, clonal outgrowth and relevant drug screening. Aims: To design an in vitro CLL culture system that incorporates key aspects of the CLL LN; contribution of non-CLL cells and cytokines, and induction of drug resistance and prolonged proliferation of CLL cells. Methods: Primary CLL cells were cultured in ultra-low attachment (ULA) plates in parallel to standard 2D cultures. PBMCs were cultured with or without T cells (in a specific CLL:T cell ratio to mimic the LN composition), or PBMCs were co-cultured with primary lymph node fibroblasts. CLL cells were either stimulated directly with a B cell cocktail (BCC) consisting of IL-2, IL-15, IL-21 and CpG and/or indirectly via a T cell stimulation of anti-CD3/CD28. Results: Compared to 2D cultures, 3D cultures showed several important CLL LN features. First, significantly increased CLL proliferation independent of IGHV mutation status, which was abrogated in the absence of T cells, or by JAK inhibitors. Notably, treatment with BTK inhibitors significantly inhibited CLL proliferation and resulted in disintegration of the 3D spheroid architecture. Second, co-culture with LN-derived stromal cells further increased CLL proliferation, reaching a maximum of 8 generations. Third, 3D cultures could be expanded approximately 3-4-fold over a course of 6 weeks using the 3D model. Fourth, when PBMCs were stimulated with BCC, spheroids developed proliferation center-like structures after 4 weeks of culture where T cells localized together with enrichment of Ki-67+ CLL cells. Finally, either B or T cell stimulation resulted in an induction of venetoclax resistance, showing that T cells are able to confer drug resistance to CLL cells in this model. Summary/Conclusion: We present a 3D culture system that underlines the role of T cells in sustained CLL proliferation, permitting investigation of CLL cells in the context of a protective niche consisting of multiple cell types, thereby opening up new avenues for clinically useful applications.

Nurselike cells sequester B cells in disorganized lymph nodes in chronic lymphocytic leukemia via alternative production of CCL21

AbstractTumor microenvironment exerts a critical role in sustaining homing, retention, and survival of chronic lymphocytic leukemia (CLL) cells in secondary lymphoid organs. Such conditions foster immune surveillance escape and resistance to therapies. The physiological microenvironment is rendered tumor permissive by an interplay of chemokines, chemokine receptors, and adhesion molecules as well as by direct interactions between malignant lymphocytes and stromal cells, T cells, and specialized macrophages referred to as nurselike cells (NLCs). To characterize this complex interplay, we investigated the altered architecture on CLL lymph nodes biopsies and observed a dramatic loss of tissue subcompartments and stromal cell networks as compared with nonmalignant lymph nodes. A supplemental high density of CD68+ cells expressing the homeostatic chemokine CCL21 was randomly distributed. Using an imaging flow cytometry approach, CCL21 mRNA and the corresponding protein were observed in single CD68+ NLCs differentiated in vitro from CLL peripheral blood mononuclear cells. The chemokine was sequestered at the NLC membrane, helping capture of CCR7-high-expressing CLL B cells. Inhibiting the CCL21/CCR7 interaction by blocking antibodies or using therapeutic ibrutinib altered the adhesion of leukemic cells. Our results indicate NLCs as providers of an alternative source of CCL21, taking over the physiological task of follicular reticular cells, whose network is deeply altered in CLL lymph nodes. By retaining malignant B cells, CCL21 provides a protective environment for their niching and survival, thus allowing tumor evasion and resistance to treatment. These findings argue for a specific targeting or reeducation of NLCs as a new immunotherapy strategy for this disease.

The scaffold protein NEDD9 is necessary for leukemia-cell migration and disease progression in a mouse model of chronic lymphocytic leukemia

The scaffold protein NEDD9 is frequently upregulated and hyperphosphorylated in cancers, and is associated with poor clinical outcome. NEDD9 promotes B-cell adhesion, migration and chemotaxis, pivotal processes for malignant development. We show that global or B-cell-specific deletion of Nedd9 in chronic lymphocytic leukemia (CLL) mouse models delayed CLL development, markedly reduced disease burden and resulted in significant survival benefit. NEDD9 was required for efficient CLL cell homing, chemotaxis, migration and adhesion. In CLL patients, peripheral NEDD9 expression was associated with adhesion and migration signatures as well as leukocyte count. Additionally, CLL lymph nodes frequently expressed high NEDD9 levels, with a subset of patients showing NEDD9 expression enriched in the CLL proliferation centers. Blocking activity of prominent NEDD9 effectors, including AURKA and HDAC6, effectively reduced CLL cell migration and chemotaxis. Collectively, our study provides evidence for a functional role of NEDD9 in CLL pathogenesis that involves intrinsic defects in adhesion, migration and homing.

Characterization of metabolic alterations of chronic lymphocytic leukemia in the lymph node microenvironment

AbstractAltered metabolism is a hallmark of both cell division and cancer. Chronic lymphocytic leukemia (CLL) cells circulate between peripheral blood (PB) and lymph nodes (LNs), where they receive proliferative and prosurvival signals from surrounding cells. However, insight into the metabolism of LN CLL and how this may relate to therapeutic response is lacking. To obtain insight into CLL LN metabolism, we applied a 2-tiered strategy. First, we sampled PB from 8 patients at baseline and after 3-month ibrutinib (IBR) treatment, which forces egress of CLL cells from LNs. Second, we applied in vitro B-cell receptor (BCR) or CD40 stimulation to mimic the LN microenvironment and performed metabolomic and transcriptomic analyses. The combined analyses indicated prominent changes in purine, glucose, and glutamate metabolism occurring in the LNs. CD40 signaling mostly regulated amino acid metabolism, tricarboxylic acid cycle (TCA), and energy production. BCR signaling preferably engaged glucose and glycerol metabolism and several biosynthesis routes. Pathway analyses demonstrated opposite effects of in vitro stimulation vs IBR treatment. In agreement, the metabolic regulator MYC and its target genes were induced after BCR/CD40 stimulation and suppressed by IBR. Next, 13C fluxomics performed on CD40/BCR-stimulated cells confirmed a strong contribution of glutamine as fuel for the TCA cycle, whereas glucose was mainly converted into lactate and ribose-5-phosphate. Finally, inhibition of glutamine import with V9302 attenuated CD40/BCR-induced resistance to venetoclax. Together, these data provide insight into crucial metabolic changes driven by the CLL LN microenvironment. The prominent use of amino acids as fuel for the TCA cycle suggests new therapeutic vulnerabilities.

Activation and expansion of T-follicular helper cells in chronic lymphocytic leukemia nurselike cell co-cultures.

Interactions between chronic lymphocytic leukemia (CLL) cells and T-cell subsets in the lymph node microenvironment are thought to play a central role in disease biology. To study these interactions in a model of the CLL lymph node microenvironment, we characterized T-cell subsets in CLL nurselike cell (NLC) co-cultures. We focused on T-follicular helper (Tfh) cells, which are characterized by CXCR5 expression and localization to B-cell follicles. In co-cultures from 28 different CLL patients, we detected an expansion of Tfh cells based on PD-1, BCL6, and ICOS expression, with increased IL-21 and downmodulated CD40L surface expression. Regulatory T cells (Treg), which promote immune tolerance, also expanded in NLC co-cultures. T-cell receptor (TR) gene repertoire analyses confirmed the clonal expansion of CD4+ T cells, with an enrichment of TR clonotypes commonly expanded also in primary CLL samples. Multicolor confocal microscopy revealed that Tfh, but not Treg co-localize with proliferating CLL cells in CLL lymph node sections. Collectively, these data provide new insight into the cellular and molecular cross-talk between CLL and T-cell subsets, resulting in clonal expansion of T-helper cells and interaction of Tfh cells with proliferating CLL cells which may open new avenues for therapeutic targeting.

The Tigit/CD226/CD155 Immunomodulatory Axis Is Deregulated in CLL and Contributes to B-Cell Anergy

The Tigit/CD226/CD155 Immunomodulatory Axis Is Deregulated in CLL and Contributes to B-Cell Anergy

Targeting Metabolic Alterations in CLL Microenvironment; Inhibition of Glutamine Import Attenuates Venetoclax Resistance

Introduction: For chronic lymphocytic leukemia (CLL), especially in the lymph node (LN) setting where cells receive proliferative and pro-survival signals, in-depth studies of altered metabolism and its relationship with therapeutic responses are still lacking. Venetoclax, a BCL-2 inhibitor currently in wide clinical use for CLL, has shown high efficiency yet emerging resistance is a growing clinical problem. In cell line models, induced resistance to Venetoclax was accompanied by profound metabolic changes 1. This is in accordance with our earlier findings on metabolic and apoptotic changes that CLL cells undergo within the LN environment 2. In the current study, we performed RNA sequencing and applied fluxomics with 13C 6-glucose and 13C 5-glutamine to investigate in detail the metabolic routes in LN CLL. This led to studies to manipulate glutamine metabolism in a venetoclax resistance model.Methods: Peripheral blood (PB) samples from CLL patients were in vitro stimulated for 24 hrs by CD40 or B cell receptor (BCR), which are two potential key signals in LN. RNAseq analysis was compared with microarray data of paired PB/LN patient samples 3. For fluxomics, CLL cells were cultured for 2 hrs in medium containing either 5 mM 13C 6-glucose or 1 mM 13C 5-glutamine. Incorporation of 13C in metabolic intermediates was analyzed by LC-MS. For glutamine blockade, CLL cells were stimulated in presence of specific inhibitors of glutamine/glutamate metabolism or amino acid transporters. Cells were then treated with venetoclax, and viability was measured.Results: Gene expression profiles demonstrated that CLL cells obtained from LN tissue as well as after in vitro CD40 or BCR stimulation showed increased expression of gene sets involved in glycolysis, oxidative phosphorylation / citric acid cycle (OXPHOS/TCA) and amino acid metabolism as well as Myc activation. This confirmed that in vitro stimulation can be used to model the CLL LN setting. For unstimulated PB CLL cells, fluxomics data demonstrated low uptake of either glucose or glutamine, with 13C labelling close to zero for most metabolites. In contrast, both CD40 or BCR stimulation increased the uptake and utilization of glucose and glutamine. 13C labelling from glucose was detected in all glycolytic intermediates analyzed in both CD40- and BCR-stimulated CLL cells. Glucose was catalyzed to lactate and also partly converted to acetyl-CoA, which entered the TCA cycle. Additionally, labelling from glucose was also increased in several metabolites of the pentose phosphate pathway (PPP) suggesting it entered nucleotide synthetic routes. Compared to glucose, the contribution of glutamine was much higher in the TCA cycle in both BCR and CD40-stimulated cells. All intermediates of the TCA cycle were highly enriched with 13C from glutamine (Figure 1A). Combined, these data revealed that glutamine is the key metabolite to fuel the TCA cycle in LN CLL cells, and prompted us to study effects of glutamine blockade in conditions of Venetoclax resistance. It was found that venetoclax resistance induced by CD40 or BCR stimulation was clearly attenuated by glutamine uptake inhibition. CLL cells became re-sensitized to Venetoclax in both CD40- or BCR-stimulated samples, with an approximate 100-fold shift in IC50 (Figure 1B).Conclusions: Our study highlights the role of glutamine, in addition to glucose, in the metabolic reprogramming that CLL cells undergo in the LN (Figure 1C). These processes show potential for therapeutic targeting. Inhibition of glutamine import could contribute to dampen tumor microenvironment-induced Venetoclax resistance.

Transcriptome Analysis of the Lymph Node Microenvironment in Chronic Lymphocytic Leukemia

Gene expression profiling with microarrays and in vivo deuterated water labeling in patients with chronic lymphocytic leukemia (CLL) have shown that tumor activation and proliferation predominantly occur in the lymph node (LN). It is thought that CLL cells actively shape the surrounding microenvironment to support tumor survival and growth.To further dissect the biology of CLL and its interactions with the tumor microenvironment, we performed total RNA sequencing of purified tumor cells from paired peripheral blood (PB) and LN in 29 patients. Samples separated by anatomic compartment in principal component analysis. A total of 883 protein-coding genes were differentially expressed (DE) between PB and LN (fold-change >2, Q <0.1). Among these DE genes, 82% were upregulated and 18% were downregulated in LN compared to PB. Gene Set Enrichment Analysis using experimentally-derived lymphoid gene signatures (https://lymphochip.nih.gov/signaturedb) revealed activation of B-cell receptor (BCR), NF-κB, JAK-STAT, cytokine and NOTCH signaling as well as increased proliferation and metabolism in LN (normalized enrichment score >1.6, Q <0.05). In contrast, PB tumor cells were characterized by a quiescent transcriptional program.Long non-coding RNAs (lncRNAs) are a novel class of transcripts with emerging significance in cancer. We detected 134 DE lncRNAs between PB and LN. To gain insight into the function of lncRNAs, DE lncRNAs and protein-coding genes were subjected to weighted gene co-expression network analysis grouping highly co-expressed transcripts into modules. One of 5 identified modules contained protein-coding genes representing BCR activation and cell proliferation. Among 36 lncRNAs in this module, CCAT1 was overexpressed in LN-resident CLL cells. Overexpression of CCAT1 has also been reported in acute myeloid leukemia and some solid tumors, in which CCAT1 regulates MYC transcription. While the function of most lncRNAs remains unknown, our data suggest a network of lncRNAs may regulate survival pathways and proliferation of CLL cells in the microenvironment.Next, we profiled the transcriptome of bulk LN samples from 33 patients with CLL and 4 normal donors to characterize the cellular composition of CLL-LNs. We applied CIBERSORT (Newman, Nat Methods 2015) to infer the relative proportion of leukocyte subsets. First, B cells were more abundant in CLL than normal LNs (median 66% versus 34% of cellular elements, Q < 0.0001). Among non-B cells, CLL-LNs had more γδ-T cells (median 12% versus 1%, Q = 0.04), but fewer CD4+ T cells (median 34% versus 57%, Q = 0.008), NK cells (median 0% versus 4%, Q = 0.008), and mast cells (median 0% versus 1%, Q = 0.01) than normal LNs. No difference was observed in the frequency of CD8+ T-cells or monocyte-macrophage lineage cells between CLL and normal LNs. Because M2-polarized tumor-associated macrophages have been implicated in tumor development, we also assessed the ratio of M1 to M2 macrophages. However, macrophages from CLL-LNs did not exhibit increased M2 polarization as compared to normal LNs.In summary, we confirmed BCR and NF-κB activation reported in prior microarray studies and additionally identified the upregulation of JAK-STAT and NOTCH signaling in LN-resident tumor cells. In addition, network analysis suggested a possible role for lncRNAs in the response of CLL cells to microenvironmental stimuli. Significant alterations in the cellular composition of CLL-LNs included the relative expansion of B cells and γδ-T cells. Our data highlight the complexity of tumor-microenvironment interactions in CLL and reveal novel aspects that merit further investigation.This research was supported by the Intramural Research Program of the NHLBI. DisclosuresWiestner:Acerta Pharma: Research Funding; Pharmacyclics: Research Funding.

Tumor Microenvironment (TME) Regulation of Programmed Death-1 (PD1) Receptor and Its Ligands PDL1 and PDL2 in Chronic Lymphocytic Leukemia (CLL)

The PD1 pathway is involved in the inactivation of immune effectors and is potentially reactive against neoplastic cells in the TME. T cells from CLL patients exhibit defective immunity, decreased proliferative capacity, and impaired effector function leading to T cell exhaustion. These functional defects coincide with higher CD244, CD160 and PD1 expression on T cells. In the Eμ-TCL1 transgenic CLL model, PDL1 checkpoint blockade restores immune dysfunction and impairs leukemia growth. PD1/PDL1 ligation also affects BCR signaling, and because PD1 is also expressed on CLL cells, PD1 interference might directly influence tumor growth and proliferation directly.We studied the 1) expression the PD1 receptor and its ligands and their correlation with TFTT; 2) role of the TME in controlling PD1 axis and miRNA expression; 3) role of the BCRi ibrutinib (IB) in the expression of the receptor and ligands.Early-stage CLL patients were prospectively enrolled at diagnosis (n=211, O-CLL1: clinicaltrial.govID:NCT00917540). Gene expression (GE) profiling was performed using the GeneChipVR Gene1.0 STArray (Affymetrix) and validated by qPCR (Thermofisher) and flow-cytometry (FC, BD Biosciences) in an independent subset of patients. CLL lymph node samples underwent in situ immunolocalization analyses. Autologous T-cells (AAT) were obtained by in vitro exposure of patient T-cells with anti CD3/CD28 Dynabeads (Thermofisher) and IL2 in co-culture with CLL cells. Cultures were monitored until substantial clumping was observed and tested for PD1 axis expression. HS5 stromal cells were used in co-culture experiments with B-CLL (ratio 1:100). MiR-424 expression was evaluated by qPCR using validated TaqMan® MicroRNA Assays (Thermofisher). In selected experiments IB was added to cell culture. PD1 GEP of circulating B-CLL cells was significantly higher than of ligands. No differences in GEP of PD1 or its ligands were observed stratifying cases based on CD38, ZAP-70, or FISH markers. Significantly higher PDL1 GE was detected in IGHV-M cases. GEP results were validated by qPCR on 27 independent CLL samples. The impact of GE of the PD1 axis on clinical outcomes in CLL cases (median follow-up 39 months, range 6−82 months) indicated TTFT was significantly shorter in cases with higher levels of PDL2, while no significant differences were detected based on higher or lower PD1 or PDL1 levels. Multivariate analysis indicated higher PDL2 gene expression retained an independent prognostic power (HR=1.9, 95%CI 1.1-3.4, P=.022) in predicting TTFT together with IGHV-UM status, B-lymphocytosis≥5000/mm3 (P=.020), and CD38 expression (P=.021). In situ immunolocalization of CLL tissue infiltrates showed variable expression of PD1 that characterized small lymphoid elements, while PDL1 and PDL2 (PDL2>PDL1) characterized larger medium-sized elements within proliferation centers. Co-localization studies revealed PD1 co-expression of CD20+CLL cells and in scattered CD3+ cells. Both PDL1 and PDL2 were variably expressed in CD20+CLL cells; few T-cells and macrophages showed expression of both ligands within CLL infiltrates. FC analysis of CLL cells after AAT co-culture showed a higher percentage expression of circulating PD1 axis members (n=27, PDL1>>PDL2>PD1) compared to baseline indicating that the changes in PD1 axis expression may be due to events triggered by activation of the TME. In contrast, T cell populations had higher expression of CD3+, CD4+ and CD8+ cells bearing both PD1 and PDL1, while no substantial increment for either CD3+, CD4+ or CD8+/PDL2+ cells following AAT co-culture. GE analysis also showed a similar increase in PDLs (n=15, PDL2>>PDL1>PD1) expression with respect to non-activated B-CLL cells. Interestingly, a 48h co-culture of B-CLL with HS5 stromal cells (n=6) showed a similar increase in PD1 axis GE pattern as AAT cultures. Expression of miR-424 also decreased significantly (<80%) after AAT co-cultures (n=6), indicating its potential involvement in PDL1 regulation. IB reduced PD1/PDL1 gene and protein expression on CLL B-cells; the IB inhibitory effect was also observed only on CD8+ cells.We indicate that PDL2 expression characterizes a subset of high-risk early stage CLL patients. PD1 axis expression is characteristic of the CLL TME whereby modulatory co-stimulatory signals may be counteracted by IB. Further studies will examine the role of miR-424 regulation with this cellular context. DisclosuresNo relevant conflicts of interest to declare.

Microenvironment Remodeling and Subsequent Clinical Implications in Diffuse Large B-Cell Histologic Variant of Richter Syndrome.

IntroductionRichter Syndrome (RS) is defined as the development of an aggressive lymphoma in the context of Chronic Lymphocytic Leukemia (CLL), with a Diffuse Large B-Cell Lymphoma (DLBCL) histology in 95% cases. RS genomic landscape shares only a few features with de novo DLBCLs and is marked by a wide spectrum of cytogenetic abnormalities. Little is known about RS microenvironment. Therapeutic options and efficacy are limited, leading to a 12 months median overall survival. The new targeted treatments usually effective in CLL fail to obtain long-term remissions in RS.MethodsWe reviewed available PubMed literature about RS genomics, PD-1/PD-L1 (Programmed Death 1/Programmed Death Ligand 1) pathway triggering and subsequent new therapeutic options.ResultsData from about 207 patients from four landmark papers were compiled to build an overview of RS genomic lesions and point mutations. A number of these abnormalities may be involved in tumor microenvironment reshaping. T lymphocyte exhaustion through PD-L1 overexpression by tumor cells and subsequent PD-1/PD-L1 pathway triggering is frequently reported in solid cancers. This immune checkpoint inhibitor is also described in B lymphoid malignancies, particularly CLL: PD-1 expression is reported in a subset of prolymphocytes from the CLL lymph node proliferation centers. However, there is only few data about PD-1/PD-L1 pathway in RS. In RS, PD-1 expression is a hallmark of recently described « Regulatory B-cells », which interact with tumor microenvironment by producing inhibiting cytokines such as TGF-β and IL-10, impairing T lymphocytes anti-tumoral function. Based upon the discovery of high PD-1 expression on tumoral B lymphocyte from RS, immune checkpoint blockade therapies such as anti-PD-1 antibodies have been tested on small RS cohorts and provided heterogeneous but encouraging results.ConclusionRS genetic landscape and immune evasion mechanisms are being progressively unraveled. New protocols using targeted treatments such as checkpoint inhibitors as single agents or in combination with immunochemotherapy are currently being evaluated.

Eomes and IL-10 Regulate Anti-Tumor Activity of T Cells in Chronic Lymphocytic Leukemia

Introduction: Genome-wide association studies showed that a single-nucleotide polymorphism (SNP) affecting the transcription factor Eomesodermin (Eomes) is associated with a significantly increased risk to develop chronic lymphocytic leukemia (CLL). Eomes and its paralogue T-bet are known master regulators of CD8+ effector T cells and CD4+ T helper cells and critical for T cell-mediated immune responses against pathogens and cancer. While Eomes has been shown to be essential for effector function of CD8 T cells, its role in CD4 T cells is less well understood. Recent work suggested that Eomes drives the development of IL-10 and IFNγ co-producing, FoxP3-negative, regulatory T cells, named Tr1 cells, a population that was found enriched in inflamed tissues in patients with chronic inflammatory disorders. In CLL, the T cell compartment is altered with an enrichment of effector and memory T cells that were shown to control leukemia development in a mouse model, but also to acquire a dysfunctional or exhausted state. Methods: We investigated Eomes-expressing CD4 and CD8 T cells in blood and lymph node (LN) samples of patients with CLL, as well as in the Eµ-TCL1 mouse model of CLL by flow cytometry, CyTOF, mRNA sequencing, and ex vivo functional assays. The role and function of these cells was explored in bone marrow-chimeric mice harbouring an Eomes-deficient hematopoietic microenvironment that were used for adoptive transfer of TCL1 leukemia, as well as by co-transfer experiments of Eomes- or IL-10R-deficent CD4 or CD8 T cells with TCL1 leukemia cells in Rag2-/- mice lacking B and T cells. Results: We detected an accumulation of Eomes-expressing CD4 and CD8 T cells in CLL patients, which was more severe in LN compared to blood samples, and significantly stronger in CLL LN compared to reactive LN samples as non-cancer control (Fig. 1A). This was in line with an observed expansion of Eomes-positive T cells in the spleen of leukemic Eµ-TCL1 mice and upon adoptive transfer of TCL1 leukemia. Eomes expression in CD8 T cells correlated with the expression of CD69, IFNγ and PD-1, suggesting a link between Eomes and CD8 T cell activation and function in CLL. The importance of Eomes in CD8 T cell-mediated control of CLL was demonstrated in mice that were transplanted with TCL1 leukemia, where Eomes-deficient CD8 T cells failed to control leukemia development. As we detected significantly less Eomes-deficient CD8 T cells in these mice compared to respective wildtype controls, and a lower percentage of them was positive for the proliferation marker Ki-67, we conclude that Eomes drives the differentiation and expansion of CD8 T cells in mice with CLL-like disease. We further explored Eomes-expressing CD4 T cells in CLL by transcriptome analysis, flow cytometry and ex vivo functional assays, and observed increased expression of IFNγ and IL-10, as well as inhibitory receptors, like PD-1, BLIMP-1 and LAG3, features that are described for Tr1 cells. Transfer of Eomes-deficient or wildtype CD4 T cells in Rag2-/- mice that were injected with TCL1 leukemia cells, lead to a comparable expansion of CD4 T cells independent of Eomes. But even though wildtype CD4 T cells were able to control leukemia development in this setting, Eomes-deficient CD4 T cells failed to do so (Fig. 1B). As Eomes is a known driver of IL-10 expression, we tested whether IL-10R signalling in CD4 T cells is involved in the anti-tumor activity by performing respective CD4 T cell transfer experiments comparing this time wildtype with IL-10R-deficent CD4 T cells. Interestingly, lack of IL-10R in CD4 T cells lead to a reduction in anti-tumor control (Fig. 1C) and therefore suggests that IL-10 is involved in Eomes-driven regulation of CD4 T cell-mediated immune control. Conclusions: In summary, we conclude that Eomes is required for CD8 T cell-mediated control of CLL, and Eomes+ PD-1+ IL-10-producing CD4 T cells contribute to adaptive immunity in CLL. The increased risk of developing CLL in individuals harbouring a SNP in the Eomes gene might be therefore explained by a negative impact of this alteration on CD4 and/or CD8 T cell-mediated immune control of CLL. Disclosures Stilgenbauer: Novartis: Consultancy, Honoraria, Research Funding, Speakers Bureau; Amgen: Consultancy, Honoraria, Research Funding, Speakers Bureau; Celgene: Consultancy, Honoraria, Research Funding, Speakers Bureau; Gilead: Consultancy, Honoraria, Research Funding, Speakers Bureau; GSK: Consultancy, Honoraria, Research Funding, Speakers Bureau; Hoffmann La-Roche: Consultancy, Honoraria, Research Funding, Speakers Bureau; Pharmacyclics: Other: Travel support; AbbVie: Consultancy, Honoraria, Research Funding, Speakers Bureau; Janssen: Consultancy, Honoraria, Research Funding, Speakers Bureau; AstraZeneca: Consultancy, Honoraria, Research Funding, Speakers Bureau.

Ibrutinib and Venetoclax Target Distinct Subpopulation of CLL Cells: Rationale for Drug Combination and Implication of Minimal Residual Disease Eradication

Ibrutinib (Ibr) is a specific inhibitor of Bruton tyrosine kinase, a component of the B-cell receptor (BCR) signaling. Venetoclax (Veneto) inhibits the anti-apoptotic protein BCL2. Both drugs are highly effective as monotherapy against chronic lymphocytic leukemia (CLL) (Byrd NEJM, Roberts, NEJM, 2015 and Roberts, Blood, 2019) and clinical trials using the combination therapy are ongoing. An interesting clinical observation is that the tumor cell sensitivity to these two drugs differ between different anatomic compartments. While lymph node-resident CLL cells are more sensitive to ibr, circulating CLL cells in the peripheral blood are more sensitive to the action of veneto. When these two drugs are used in combination to treat relapsed/refractory (Hillman ASH 2018 and Kater EHA 2019) or previously untreated CLL patients (Jain NEJM 2019), rate of complete response significantly increases compared to single drug alone. Moreover, negativity of bone marrow minimal residual disease continues to improve over time. However, the reason behind these observed compartmental responses is largely unknown, and we investigated the differential response using an ex vivo model that promotes CLL proliferation (CLL proliferation model). A better understanding of how these two drugs synergize would eventually help develop other rational combination strategies. We established the ex vivo model using fibroblasts derived from the bone marrow of a CLL patient. In this model, CLL cells closely interact with the bone marrow fibroblasts (BMF) (Fig.1) and divide and proliferate for several generations, as measured by CFSE labeling. The culture, in some cases, may continue for 10 weeks before cell death ensues. With this ability to promote long-term cell proliferation, the model recapitulates one of the most salient features observed histologically in the "proliferation centers" of the CLL lymph nodes. With the proliferation model, we first tested the effects of ibrutinib on cell division/proliferation as well as cell viability in samples taken from 30 CLL patients consisting of 22 ibr-naïve (thus sensitive) and 8 ibr-relapsed (thus resistant) patients. We demonstrate that there was lack of a significant inhibition of ibr on cell viability. In comparison, Ibr markedly inhibited cell proliferation in cells from ibr-naïve/sensitive patients (p&amp;lt;0.0001), while such inhibition was much less prominent in cells from ibr-relapsed/resistant patients (p=0.053). Overall, our data demonstrate that only the dividing subpopulation of CLL responds to ibr, and this response is highly correlated with patients' clinical response to the drug while cell viability responses are not. With the CLL proliferation model, we also evaluated how veneto works. Veneto, as expected, markedly decreased cell viability at clinically achievable concentrations. CFSE profiles of the remaining live cells, however, revealed an interesting pattern of distribution, showing that veneto induces cell death but it preferentially kills the resting CLL population (Fig. 2). In other words, the resting subpopulation of CLL, instead of the dividing one, preferentially responds to veneto as compared to ibr. Combination of the two drugs, as expected, worked most effectively, significantly reducing the total number of live cells, both resting and dividing subpopulations, in all cases. Analyses of CFSE profiles became less meaningful in many of these cases due to the small number of live cells remaining after the combination treatment. In conclusion, our study, comparing the actions of ibr and veneto in a clinically relevant CLL proliferation model, demonstrates for the first time that the drugs target distinct subpopulations of CLL cells with different proliferative capacity. The data also suggest that single drug therapy may leave a subpopulation behind that would potentially rise to cause a future relapse. Our study provides a strong laboratory rationale explaining why combining these two drugs has the potential to eradicate CLL disease. The findings are also in line with the clinical observations regarding the compartmental response; ibrutinib works primarily on the lymph nodes where CLL cells divide in the "proliferation centers" and veneto acts preferentially on the circulating CLL cells that are mostly non-dividing and resting. Disclosures Wang: Incyte Reaserch Institute: Employment. Ma:Abbvie: Research Funding; Xeme: Research Funding; Novartis: Research Funding; Bioverativ: Consultancy; Janssen: Consultancy, Speakers Bureau; Beigene: Research Funding; Kite: Consultancy; Gilead: Research Funding; Pharmacyclics: Consultancy, Research Funding, Speakers Bureau; Acerta: Research Funding; Astra Zeneca: Consultancy, Research Funding, Speakers Bureau; Genentech: Consultancy; Incyte: Research Funding; Juno: Research Funding. Larson:Celgene: Consultancy; Agios: Consultancy; Novartis: Honoraria, Other: Contracts for clinical trials.

PS1125 LINKING MICROENVIRONMENTAL SIGNALS TO METABOLIC SWITCHES AND IBRUTINIB RESPONSE IN CHRONIC LYMPHOCYTIC LEUKEMIA

Background:Chronic lymphocytic leukemia (CLL) is a clinically heterogeneous B cell malignancy, with accumulation of clonal CD5+ B cells in blood, lymph nodes (LN) and bone marrow. CLL has become a paradigm for a cancer that depends on signals from the microenvironment. This is evidenced by the clinical success of the Btk inhibitor ibrutinib, which drives CLL cells out of the LN to eventually die by neglect. In peripheral blood, CLL cells are quiescent, non‐cycling, with low metabolic activity. In contrast, in LN CLL cells receive various signals from surrounding cells, crucial for proliferation, and protection from chemotherapeutic drugs. These acquired advantages are possibly linked with altered metabolism, but it is unclear if and how CLL cells change their metabolism in the LN environment.Aims:We aimed to investigate the altered metabolism of CLL cells in the tumor microenvironment, and what signals determine these changes.Methods:Glucose uptake, glycolysis, mitochondrial mass, oxidative phosphorylation and lipid uptake were measured by flow cytometer or Seahorse XF96 analyzer, and 90 metabolites were analyzed by LC mass‐spectrometry.Results:We first established that the metabolic activity profile of circulating recent LN emigrants (CXCR4high/CD5dim PB lymphocytes) resembled that of actual LN residents (obtained by a core biopsy), which showed both higher mitochondrial mass and glucose uptake than paired peripheral blood CLL cells. In vitro stimulation of isolated blood‐derived CLL samples demonstrated that CD40 but not BCR signaling recapitulate these metabolic changes. Next, 10 patient samples were stimulated with/without CD40L for 48 hours and analyzed by mass‐spec for metabolic intermediates. To draw a further parallel with the situation in LN, metabolomics was performed on 13 patients sampled before and after 3 months of ibrutinib treatment. Combined, the data shows that various metabolic pathways are activated in the tumor microenvironment, particularly TCA cycle, pyruvate metabolism, glycolysis, and fatty acid metabolism. Apart from these overall changes, the highest ranking shifts in metabolites point to involvement of amino acids to fuel the TCA cycle. Moreover, aberrant patterns in metabolic changes after ibrutinib were noted in some patients, possibly linked with clinical response.Summary/Conclusion:CD40 signaling leads to increased oxidative phosphorylation and glycolysis in the CLL microenvironment. Opposite changes were observed in ibrutinib treated samples, together providing indirect insight into the metabolism in the CLL LN.

PF358 FUNCTIONAL HIERARCHY AMONG PRO-SURVIVAL BCL-2 MEMBERS IN DETERMINING RESISTANCE TO VENETOCLAX IN CLL

Background: The Bcl-2 inhibitor Venetoclax rapidly induces apoptosis in chronic lymphocytic leukemia (CLL) cells in peripheral blood (PB) whereas lymph node (LN) responses are less complete. This may be linked to pro-survival signals in the LN, where micro-environmental signals strongly induce Bcl-XL, Mcl-1 and Bfl-1 which are not targeted by Venetoclax. In some patients resistance occurs over time which sometimes can be explained by acquired mutations, but in most cases the mechanism(s) remain unknown. Aims: We hypothesize that changes in expression or shifts in functional interactions among Bcl-2 family members are key to clinical response and refractory disease. Methods: Intracellular FACS staining for Bcl-2, Mcl-1 and Bcl-XL in circulating recent LN emigrants (CD19+/CXCR4high/CD5dim) versus returning (CD19+/CXCR4llow/CD5high) lymphocytes in patient samples treated with Venetoclax. To investigate the functional impact of expression of Bcl-2 family members, profiling with additional BH3 mimetics against Bcl-XL (A1331852) and Mcl-1 (S63845) was applied in a CLL LN model. Primary CLL cells were co-cultured with fibroblasts expressing CD40L in order to mimic T cell help in the LN. Results: Under Venetoclax treatment, levels of Bcl-2 itself and other family members may increase in some patients, early in clinical response. During short-term in vitro incubation with Venetoclax or other BH3 mimetics, we established that Mcl-1 and Bfl-1 can increase but not Bcl-2 or Bcl-XL. The effect on Mcl-1 and Bfl-1 can be explained by a stabilization of these normally short-lived proteins. In vitro CD40 activation renders CLL cells fully resistant to single Venetoclax treatment. Using this model, we find that dual or triple BH3 mimetic combinations targeting Bcl-2, Bcl-XL and Mcl-1 restored killing. Specifically, whereas the IC50 of dual Bcl-2+Mcl-1 targeting is >1uM, combined inhibition of Bcl-2+Bcl-XL has an IC50 around 0.3uM. Bcl-2 and Bcl-XL thus seem to be the most dominant players. However, even when Bcl-2, Bcl-XL and Mcl-1 were inhibited using triple BH3 mimetic treatment, Venetoclax resistance was not fully reversed to unstimulated control (IC50 1 versus 10 nM), highlighting a potential role for Bfl-1. To further dissect the functional roles, Bim co-immunoprecipitation was performed. In PB CLL cells, Bim only binds to Bcl-2 and upon Venetoclax treatment it is released and targets Bax/Bak to kills cells. After CD40 stimulation, Bim shifts predominantly to Bcl-XL and less to Mcl-1. Only when applying triple BH3 mimetic combinations targeting Bcl-2, Bcl-XL and Mcl-1, Bim starts to bind Bfl-1. These observations suggest a hierarchical model in which Bcl-2 is most important and where upon Venetoclax treatment, Bcl-XL and Mcl-1 come into play. When all of these are targeted by triple BH3 mimetics, Bfl-1 is engaged to intercept apoptosis. Summary/Conclusion: We applied BH3 mimetics in primary CLL and showed that combined targeting of Bcl-2 members can abrogate resistance to Venetoclax. Moreover, we dissected the roles of the different Bcl-2 family members and constructed a hierarchical model of Bcl-2 family members, providing new insights into Venetoclax resistance which may have clinical utility.

A B Cell Regulome Links Notch to Downstream Oncogenic Pathways in Small B Cell Lymphomas

Gain-of-function Notch mutations are recurrent in mature small B cell lymphomas such as mantle cell lymphoma (MCL) and chronic lymphocytic leukemia (CLL), but the Notch target genes that contribute to B cell oncogenesis are largely unknown. We performed integrative analysis of Notch-regulated transcripts, genomic binding of Notch transcription complexes, and genome conformation data to identify direct Notch target genes in MCL cell lines. This Bcell Notch regulome is largely controlled through Notch-bound distal enhancers and includes genes involved in B cell receptor and cytokine signaling and the oncogene MYC, which sustains proliferation of Notch-dependent MCL cell lines via a Notch-regulated lineage-restricted enhancer complex. Expression of direct Notch target genes is associated with Notch activity in an MCL xenograft model and in CLL lymph node biopsies. Our findings provide key insights into the role of Notch in MCL and other Bcell malignancies and have important implications for therapeutic targeting of Notch-dependent oncogenic pathways.

CD40 signaling instructs chronic lymphocytic leukemia cells to attract monocytes via the CCR2 axis.

Chronic lymphocytic leukemia (CLL) cells are provided with essential survival and proliferative signals in the lymph node microenvironment. Here, CLL cells engage in various interactions with bystander cells such as T cells and macrophages. Phenotypically distinct types of tumor infiltrating macrophages can either be tumor supportive (M2) or play a role in tumor immune surveillance (M1). Although recent in vitro findings suggest a protective role for macrophages in CLL, the actual balance between these macrophage subsets in CLL lymphoid tissue is still unclear. Furthermore, the mechanism of recruitment of monocytes towards the CLL lymph node is currently unknown. Both questions are addressed in this paper. Immunofluorescence staining of lymph node samples showed macrophage skewing towards an M2 tumor-promoting phenotype. This polarization likely results from CLL-secreted soluble factors, as both patient serum and CLL-conditioned medium recapitulated the skewing effect. Considering that CLL cell cytokine secretion is affected by adjacent T cells, we next studied CLL-mediated monocyte recruitment in the presence or absence of T-cell signals. While unstimulated CLL cells were inactive, T cell-stimulated CLL cells actively recruited monocytes. This correlated with secretion of various chemokines such as C-C-motif-ligand-2,3,4,5,7,24, C-X-C-motif-ligand-5,10, and Interleukin-10. We also identified CD40L as the responsible T-cell factor that mediated recruitment, and showed that recruitment critically depended on the C-C-motif-chemokine-receptor-2 axis. These studies show that the shaping of a tumor supportive microenvironment depends on cytokinome alterations (including C-C-motif-ligand-2) that occur after interactions between CLL, T cells and monocytes. Therefore, targeted inhibition of CD40L or C-C-motif-chemokine-receptor-2 may be relevant therapeutic options.