AbstractBackgroundTreatment on risk adapted intensive pediatric protocols has improved outcome for teenagers and young adults (TYA) with T‐cell acute lymphoblastic leukemia (T‐ALL). Understanding the biology of disease in this age group and the genetic basis of relapse is a key goal as patients with relapsed/refractory disease have poor outcomes with conventional chemotherapy and novel molecular targets are required. This study examines the question of whether TYA T‐ALL has a specific biological‐molecular profile distinct from pediatric or adult T‐ALL.MethodsGenomic characterization was undertaken of a retrospective discovery cohort of 80 patients aged 15–26 years with primary or relapsed T‐ALL, using a combination of Genome‐Wide Human SNP Array 6.0, targeted gene mutation and promoter methylation analyses. Findings were confirmed by MLPA, real‐time quantitative PCR, and FISH. Whole Exome Sequencing was performed in 4 patients with matched presentation and relapse to model clonal evolution. A prevalence analysis was performed on a final data set of 1,792 individual cases to identify genetic lesions with age specific frequency patterns, including 972 pediatric (1–14 years), 439 TYA (15–24 years) and 381 adult (≥25 years) cases. These cases were extracted from 19 publications with comparable genomic data identified through a PubMed search.ResultsGenomic characterization of this large cohort of TYA T‐ALL patients identified recurrent isochromosome 7q i(7q) in our discovery cohort (n = 3). Prevalence analysis did not identify any age specific genetic abnormalities. Genomic analysis of 6 pairs of matched presentation – relapsed T‐ALL established that all relapses were clonally related to the initial leukemia. Whole exome sequencing analysis revealed recurrent, targetable, mutations disrupting NOTCH, PI3K/AKT/mTOR, FLT3, NRAS as well as drug metabolism pathways.ConclusionsAll genetic aberrations in TYA T‐ALL occurred with an incidence similar or intermediate to that reported in the pediatric and adult literature, demonstrating that overall TYA T‐ALL exhibits a transitional genomic profile. Analysis of matched presentation – relapse supported the hypothesis that relapse is driven by the Darwinian evolution of sub‐clones associated with drug resistance (NT5C2 and TP53 mutations) and re‐iterative mutation of known key T‐ALL drivers, including NOTCH1.

Loss-of-function mutations in KMT2D are a striking feature of the germinal centre (GC) lymphomas, resulting in decreased H3K4 methylation and altered gene expression. We hypothesised that inhibition of the KDM5 family, which demethylates H3K4me3/me2, would re-establish H3K4 methylation and restore the expression of genes repressed upon loss of KMT2D . KDM5-inhibition increased H3K4me3 levels and caused an anti-proliferative response in vitro , which was markedly greater in both endogenous and CRISPR-edited KMT2D mutant DLBCL cell lines, whilst tumour growth was inhibited in KMT2D mutant xenografts in vivo . KDM5-inhibition reactivated both KMT2D-dependent and -independent genes, resulting in diminished B-cell receptor signalling and altered expression of BCL2 family members, including BCL2 itself, allowing it to synergise with agents targeting these pathways. KDM5-inhibition may offer an effective therapeutic strategy for ameliorating KMT2D loss-of-function mutations in GC-lymphomas. Statement of significance We detail a novel way of reverting the effects of loss-of-function mutations in the histone methyltransferase KMT2D by inhibiting the KDM5 demethylase family, increasing levels of H3K4me3 and restoring expression of KMT2D regulated genes.

Background. Treatment failure after allogeneic haematopoietic stem-cell transplantation (AHST) using reduced-intensity conditioning (RIC) results from too much alloreactivity and harmful acute Graft-versus-Host Disease (aGVHD). Studies have identified many reconstituting immune cell subsets associated with development of clinical alloreactivity but the functionally dominant parameters at different time-points remain unknown. We therefore used mass cytometry (MS) to simultaneously assess multiple alloreactive and immunoregulatory cell populations to identify dominant immune reconstitution signatures associated with subsequent development of aGvHD after AHST. Methods. Phenotypic markers identifying more than 30 immune cell subsets known to influence alloreactivity were combined in a single MS panel. Peripheral blood from 58 patients with haematological cancers was analysed after T-replete HLA-matched RIC-AHST using uniform conditioning. Normalization of individual test samples spiked with CD45-barcoded healthy control cells was used to reduce batch effects. Complementary high-dimensional analytic tools were used to generate cellular profiles across the whole cohort and identify differences between patients grouped by subsequent development of aGVHD. Results. Unsupervised clustering analysis identified 40 phenotypically distinct T, B and NK cell clusters post-transplant. Significant batch effects were effectively reduced with a novel R-based algorithm normalising data to control cells. Cluster diversity analysis early post-transplant demonstrated lower cluster diversity in patients who subsequently developed aGvHD consistent with perturbation of phenotypic clusters in these patients. Two specific clusters were significantly different in abundance at D+30 in patients who went on to develop aGvHD and those who remained aGVHD-free. A cluster with a CD56brightCD16negCD27+/- regulatory NK cell (NKreg) phenotype was reduced in patients going on to develop aGvHD using both Phenograph and FlowSOM algorithms (p=0.001). These findings were validated by forward analysis using the CITRUS algorithm, revealing a similar differentiating cell population. CD56bright NKreg reconstitution was independent of CMV reactivation and did not impede reconstitution of WT1 and PR1 tumor-associated antigen-specific T cells. The reduction in NKreg in patients who subsequently developed aGvHD was accompanied by a significant increase in alloreactive CCR5+CD45RA-CCR7- CD4 effector memory T cells (Tem). We next used correlation analysis of cluster abundance across the whole cohort to identify all clusters contributing to the immune 'regulome' (those inversely correlated with alloreactive CD4 EM and/or CD8 EM T cell clusters). Notably, at D+30 the regulome consisted of 4 phenotypically distinct CD56bright NKreg clusters and a CD4+CD8+ double positive (DP) cluster, but not FOXP3+ CD4 regulatory T cells (Treg), Figure 1A Both the identity of differentiating clusters between patients subsequently developing aGVHD and those who remained aGvHD-free, and the dominant constituents of the regulome changed over time. By D+60 a CD56bright NKreg cluster (with a distinct phenotype to the differentiating cluster identified at D+30) and a DP T cell cluster were significantly reduced in patients subsequently developing aGvHD. The D+60 regulome consisted of multiple distinct CD56bright clusters, a DP T cell cluster and CD4 Treg, Figure 1B. Importantly by D+90 the immune regulome consisted of a reduced number of CD56bright NKreg clusters and increasing dominance of CD4Treg, Figure 1C. Conclusion. We show proof-of-concept that a novel acquisition and analysis pipeline can be applied to MS data to identify multiple immunoregulatory cells after AHST that contribute to the control of reconstituting alloreactive T cells. This approach identified a loss of NK cell-mediated control of alloreactive CD4 Tem cells as the dominant immune process preceding the development of aGvHD early post-transplant. Importantly, we show that specific immunoregulatory subsets are dominant at different time-points, with increasing influence of DP T cells and CD4 Treg at later time points. Our data provide mechanistic insight into the dynamic pattern of control of alloreactivity over time and show that strategies to expand or potentiate immunoregulatory cells to prevent aGvHD should be time-dependent. Disclosures Gribben: Acerta/Astra Zeneca: Consultancy, Honoraria, Research Funding; Janssen: Consultancy, Honoraria, Research Funding; Celgene: Consultancy, Honoraria, Research Funding; Abbvie: Consultancy, Honoraria, Research Funding.

In the original version of this article the authors noted an omission in the author affiliations where the university details: Queen Mary University of London was not included in the original affiliation for the majority of the authors. The correct affiliations are as follows 1. Centre for Haemato-Oncology, Barts Cancer Institute, Queen Mary University of London, London, UK 3. Centre for Molecular Oncology, Barts Cancer Institute, Queen Mary University of London, London, UK 6. Evolution and Cancer Laboratory, Barts Cancer Institute, Queen Mary University of London, London, UK

Saudi Arabian Ministry of Higher Education through a doctoral scholarship awarded to A.F.A.S. and a Bloodwise Programme grant (14032) awarded to J.F., T.V., and I.D.

Proteomic and genomic integration identifies kinase and differentiation determinants of kinase inhibitor sensitivity in leukemia cells

Nat. Genet.; 10.1038/ng.3473; corrected online 12 January 2016 In the version of this article initially published online, several funding sources were omitted from the Acknowledgments section. The error has been corrected for the print, PDF and HTML versions of this article.

Donor T-cell immune responses can eradicate lymphomas after allogeneic hematopoietic stem cell transplantation (AHSCT), but can also damage healthy tissues resulting in harmful graft-versus-host disease (GVHD). Next-generation sequencing has recently identified many new genetic lesions in follicular lymphoma (FL). One such gene, tumor necrosis factor receptor superfamily 14 (TNFRSF14), abnormal in 40% of FL patients, encodes the herpes virus entry mediator (HVEM) which limits T-cell activation via ligation of the B- and T-lymphocyte attenuator. As lymphoma B cells can act as antigen-presenting cells, we hypothesized that TNFRSF14 aberrations that reduce HVEM expression could alter the capacity of FL B cells to stimulate allogeneic T-cell responses and impact the outcome of AHSCT. In an in vitro model of alloreactivity, human lymphoma B cells with TNFRSF14 aberrations had reduced HVEM expression and greater alloantigen-presenting capacity than wild-type lymphoma B cells. The increased immune-stimulatory capacity of lymphoma B cells with TNFRSF14 aberrations had clinical relevance, associating with higher incidence of acute GVHD in patients undergoing AHSCT. FL patients with TNFRSF14 aberrations may benefit from more aggressive immunosuppression to reduce harmful GVHD after transplantation. Importantly, this study is the first to demonstrate the impact of an acquired genetic lesion on the capacity of tumor cells to stimulate allogeneic T-cell immune responses which may have wider consequences for adoptive immunotherapy strategies.

Background: While myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML) are considered sporadic hematopoietic stem cell clonal disorders, there are rare occurrences of familial cases (<5%) where two or more individuals within the same family are affected. These high-risk examples are characterised by wide variations in the age of onset, disease latency and outcome between and within families, making their investigation, follow-up and treatment all the more challenging.To date, germline mutations in 11 disease genes have been described, with mutations in the myeloid transcription factor GATA2 representing one of the best-characterised genetic loci predisposing to inherited hematological malignancies. We have noted that within GATA2 families, particularly those segregating a germline p.Thr354Met mutation, there is striking evidence of reduced penetrance. In our example, two first-degree cousins (III.1 and III.3) developed high-risk MDS with monosomy 7 with a third cousin (III.7) presenting with significant leukopenia (monocytopenia [0.1x109/L] and neutropenia [0.8x109/L]). This contrasts with the parental generation (II.1, II.3 and II.5) who all remain hematologically normal and symptom free into their mid-late 60s (Figure 1). We therefore set out to understand these differences in clinical presentation between mutation carriers.

Abstract Introduction Protein kinases play a key role in how cells respond and adapt to intra and extracellular stimuli. By the addition of phosphate groups to serine, threonine or tyrosine residues, these enzymes modify the activity and properties of the targeted proteins which in turn modulate biological processes like proliferation, differentiation and cell death. Kinase signalling pathways are deregulated in most cancer types including haematological malignancies. Indeed, the kinases FLt-3, c-Kit and JAK2 as well as the up-stream kinase signalling regulators KRAS and NRAS are among the most frequently mutated genes in acute myeloid leukaemia (AML). Consequently, protein kinases have attracted the attention of the pharmaceutical and biotechnology companies and inhibitors have been found for one fifth of human kinases. In the case of AML, midostaurin, a multi-kinase inhibitor that targets, among others, the tyrosine kinase Flt-3, has granted a breakthrough therapy designation by the FDA and several other kinase inhibitors are in clinical trials or under preclinical investigation. Molecular profiling of patient samples will play a pivotal role for the development and implementation of personalized therapies including those based on kinase inhibitors. We used a molecular profile generated by a phosphoproteomics approach to rationalize why some primary AML cells respond to treatment with different kinase inhibitors while others are resistant to the same treatments. Methods Label free phosphoproteomics based on trypsin digestion and TiO2 phosphoenrichment was used to quantify > 5,000 phosphorylation sites in mononuclear cells extracted from the peripheral blood of 36 AML patients. KSEA technology was applied to infer kinase activity from the phosphoproteomics data and DAVID software was used to determine gene ontology enrichments based on the genes that code for the proteins where the phosphorylation sites were detected. Guava EasyCyte Flow Cytometry was used to determine cell viability after the treatment of the same patient samples with different kinase inhibitors. Mass cytometry was used to measure the expression at the plasma membrane of 17 surface markers in 30 of the previously analysed AML primary samples. Results The FAB classification subdivide AML cases depending on cytomorphological features. We compared the phosphoproteomes of M1 and M4 classes that are associated with early and late states of differentiation. Based on the 150 phosphopeptides more significantly regulated between FAB-M1 and FAB-M4 groups, hierarchical clustering analysis was used to stratify AML patient samples into two subsets named M1-Like and M4-Like. Phosphoproteome reanalysis showed that the M4-Like set upregulated 1255 phosphopeptides and downregulated 446 when compared with the M1-Like set. The upregulated group comprised regulatory phosphorylation sites in several kinases including PAK1 and PCK delta. Kinase activity analysis using KSEA (Kinase Substrate Enrichment Analysis) also showed an increased activity of PAK, PKCδ and other kinases like P38 alpha in the M4-Like group. Interestingly, the PAK inhibitor PF03758309 reduced more efficiently the viability in M4-Like group than in the M1-Like group (average reduction after a 72h treatment with 1µM of 55.2% for M4-Like compared to 33.8% for M1-Like, p-value = 0.0078). This difference was not observed for other inhibitors such as those targeting CK2 or p38. CyTOF analysis showed that the M4-Like group upregulated the surface expression of several differentiation markers. Discussion Predicting the effectiveness of a drug for a particular patient is a major goal of personalized medicine. In the case of kinase inhibitors, responses may be influenced by several factors including the activity of the targeted kinase as well as the activity of other kinases that act in parallel pro-survival pathways. In this work, we have found that differentiation leads to a particular activation pattern of the signalling networks, a phenomenon that determines the response to signalling inhibitors. Conclusion We found phosphoproteomics signatures in primary AML that are associated with distinct haematopoietic differentiation stages. These signatures are in turn associated with how AML cells respond to kinase inhibitors. Disclosures Fitzgibbon: Epizyme: Research Funding; Gilead: Honoraria; Janssen: Honoraria; Celgene: Honoraria.

Mantle cell lymphoma (MCL) is a clinically heterogeneous, but often aggressive lymphoma characterized by the IGH:CCND1 translocation and cyclin D1 (CCND1) over-expression. Chromosomal instability, due to disrupted DNA damage response, in conjunction with abnormal activation of cell survival mechanisms underlies the aggressive clinical course in MCL (Jares et al, 2012). In recent years, improved understanding of lymphoma biology has led to the development of a number of small molecule inhibitors. However, the relative rarity of MCL (incidence 0·55 per 100 000) (Smedby & Hjalgrim, 2011) poses a challenge in effectively evaluating these drugs in patients. In vitro studies have been limited by the difficulty of culturing primary MCL cells. Murine models of MCL cell lines are relatively easy to establish in SCID or NOD/SCID/IL2Rγ null (NSG) mice (Wang et al, 2007, 2008a; Weston et al, 2010) but have their limitations. Until a couple of years ago, the only primary mouse model of human MCL described in the literature was established by injection of primary MCL cells into subcutaneous human bone grafts implanted in SCID mice (SCID-Hu model) (Wang et al, 2008b). Recently, however, disseminated models of human primary MCL have been established in NSG mice (Iyengar et al, 2012; Klanova et al, 2014). We report our experience here in further detail, focussing on the characteristics of MCL engraftment in this model. We used 8- to 12-week-old NSG mice that were sub-lethally irradiated (3·75 Gy) 24 h prior to transplantation. Before undertaking xenograft studies with primary cells, we used the MCL cell line JEKO-1 to assess kinetics, disease burden and distribution of MCL cells in NSG mice. JEKO-1 cells were transduced with firefly luciferase and injected intravenously into irradiated mice at two doses – 0·5 × 106 and 2 × 106 cells. Bioluminescent imaging was performed at weekly intervals following injection of D-luciferin. All mice became ill with marked weight loss and had to be sacrificed by day 29. Bioluminescence was observed in the bone marrow and spleen in all mice. Mice injected with the higher cell dose had more rapid disease progression, developed hind leg weakness and had bioluminescence in the central nervous system (CNS) on imaging, indicating involvement (Fig 1). Following this, seven cryopreserved primary MCL samples were identified from the Barts Cancer Institute tissue bank. An additional fresh primary sample derived from a splenectomy was included in the cohort. Ethical approval was obtained from East London and the City Local Research Ethics Committee. Written informed consent was obtained from patients according to the Declaration of Helsinki. All samples had a classical MCL phenotype with CD5/CD20 positivity and were confirmed to have the IGH:CCND1 translocation by fluorescence in situ hybridization (FISH). Irradiated NSG mice were injected intravenously with a dose of 107 unselected MCL cells each. Flow cytometry was performed for mouse CD45 and human CD45, CD3, CD5 and CD20 on peripheral blood samples taken from mice at 3, 6 and 12 weeks. Mice were sacrificed at 20 weeks, or earlier if they met Home Office guidelines, and tissue was harvested for immunohistochemistry (IHC). Cells were flushed from mouse femur for flow cytometry. At 20 weeks, MCL cells were found in the bone marrow and spleen of mice injected with 2 out of the 7 cryopreserved primary samples. Both samples that engrafted had blastoid morphology and one was obtained from a patient with relapsed disease. FISH for IGH:CCND1 on cell suspensions prepared from spleen of NSG mice further confirmed engraftment. None of the mice that engrafted appeared to have bowel involvement as assessed by IHC. Lymphadenopathy was not found at sacrifice. Scattered human CD20-positive cells were seen in the liver but this was not a consistent feature. Mice remained relatively well until sacrifice (Fig 2A–F). As a next step, secondary transplantation of MCL cells isolated from NSG spleen (107 cells per mouse) was undertaken. Once again, engraftment was seen in mouse spleen and bone marrow on sacrifice at 20 weeks (Fig 2G). In addition to the two cryopreserved samples, evidence of engraftment was also seen in the spleen of NSG mice injected with the fresh primary sample (non-blastoid). Interestingly, there appeared to be co-existence of MCL cells and T-cells in the spleen of mice injected with fresh MCL cells, with tumour cells concentrated around blood vessels. However, these mice had to be sacrificed at 7 weeks due to illness and T-cell infiltration was found in the liver and bone marrow, without evidence of MCL. We found a similar proliferation of T cells but without evidence of MCL in one of the seven cryopreserved samples that had high T-cell content (>10%), indicating T-cell depletion may be important in this scenario. Therefore, similar to the recent report by Klanova et al (2014), we demonstrate human primary MCL engraftment in NSG mice. In contrast to their study where mice were injected with a variable cell dose (1–8 × 107 cells), we injected all mice with a fixed dose of 107 cells. This may explain the lower rate of engraftment in our study. Both cryopreserved samples that engrafted in our study had blastoid morphology, suggesting that a higher cell dose may be required for engraftment of non-blastoid MCL in this model. In our experiments, mice with primary MCL engraftment were not visibly ill at 20 weeks and disease burden was heaviest in the spleen. In contrast, disease progression was rapid in the JEKO-1 xenograft, with CNS involvement and hind leg weakness developing by 4 weeks. These findings mirror those of Klanova et al (2014), and are important considerations when designing pre-clinical experiments involving these models. The longer overall survival of NSG mouse models of primary human MCL could be an advantage for pre-clinical testing of newer agents, which often require longer periods of administration for efficacy. Finally, our study demonstrates, similar to the findings of Klanova et al (2014), that secondary transplantation can be successfully carried out in this model, highlighting the self-renewal and tumour-initiating capacity of primary MCL cells. In summary, this NSG model of human primary MCL is a promising in vivo model for both pre-clinical drug testing and further understanding MCL biology. Our research provides further insight into the advantages and limitations of this model, which will be crucial for its effective use in pre-clinical research. This work was funded by Cancer Research UK (D.B.) and a Roger Counter Foundation award from the British Society of Haematology (S.I.). We are indebted to patients who gave samples. We thank the Animal Care and Flow Cytometry core facilities staff at the London Research Institute for valuable technical help. S.I.: design and performance of experiments, data analysis and interpretation, manuscript writing; L.A.M.: design and performance of experiments, data analysis and interpretation; A.C. and S.I.: help in sample processing, research and data analysis; R.A., D.T. and J.G.: provision of vital patient samples, materials and data; A.R. and D.L. performance of FISH analysis; D.B.: design of experiments, data analysis, manuscript writing. The authors declare no potential conflicts of interest.

Understanding the dynamics of evolution of Follicular Lymphoma (FL) clones during disease progression is important for monitoring and targeting this tumor effectively. Genetic profiling of serial FL biopsies and examples of FL transmission following bone marrow transplant suggest that this disease may evolve by divergent evolution from a common ancestor cell. However where this ancestor cell resides and how it evolves is still unclear. The analysis of the pattern of somatic hypermutation of the immunoglobulin gene (Ig) is traditionally used for tracking the physiological clonal evolution of B cells within the germinal center and allows to discriminate those cells that have just entered the germinal center and display features of ancestor cells from those B cells that keep re-circulating across different lymphoid organs. Here we investigated the pattern of somatic hypermutation of the heavy chain of the immunoglobulin gene (IgH-VH) in 4 flow-sorted B cells subpopulations belonging to different stages of differentiation, from sequential lymph node biopsies of cases displaying diverse patterns of evolution, using the GS-FLX Titanium sequencing platform. We observed an unexpectedly high level of clonality, with hundreds of distinct tumor subclones in the different subpopulations from the same sample, the majority detected at a frequency <10(-2). By using a lineage trees analysis we observed in all our FL and t-FL cases that the oligoclonal FL population was trapped in a narrow intermediate stage of maturation that maintains the capacity to undergo SHM, but was unable to further differentiate. The presence of such a complex architecture highlights challenges currently encountered in finding a cure for this disease.

In-depth molecular investigation of familial leukemia has been limited by the rarity of recognized cases. This study examines the genetic events initiating leukemia and details the clinical progression of disease across multiple families harboring germ-line CEBPA mutations. Clinical data were collected from 10 CEBPA-mutated families, representing 24 members with acute myeloid leukemia (AML). Whole-exome (WES) and deep sequencing were performed to genetically profile tumors and define patterns of clonal evolution. Germline CEBPA mutations clustered within the N-terminal and were highly penetrant, with AML presenting at a median age of 24.5 years (range, 1.75-46 years). In all diagnostic tumors tested (n = 18), double CEBPA mutations (CEBPAdm) were detected, with acquired (somatic) mutations preferentially targeting the C-terminal. Somatic CEBPA mutations were unstable throughout the disease course, with different mutations identified at recurrence. Deep sequencing of diagnostic and relapse paired samples confirmed that relapse-associated CEBPA mutations were absent at diagnosis, suggesting recurrence was triggered by novel, independent clones. Integrated WES and deep sequencing subsequently revealed an entirely new complement of mutations at relapse, verifying the presentation of a de novo leukemic episode. The cumulative incidence of relapse in familial AML was 56% at 10 years (n = 11), and 3 patients experienced ≥3 disease episodes over a period of 17 to 20 years. Durable responses to secondary therapies were observed, with prolonged median survival after relapse (8 years) and long-term overall survival (10-year overall survival, 67%). Our data reveal that familial CEBPA-mutated AML exhibits a unique model of disease progression, associated with favorable long-term outcomes.

Distinct patterns of DNA methylation characterize the epigenetic landscape of promyelocytic leukemia/retinoic acid receptor-α (PML-RARα)-associated acute promyelocytic leukemia (APL). We previously reported that the microRNAs (miRNAs) clustered on chromosome 14q32 are overexpressed only in APL. Here, using high-throughput bisulfite sequencing, we identified an APL-associated hypermethylation at the upstream differentially methylated region (DMR), which also included the site motifs for the enhancer blocking protein CCCTC-binding factor (CTCF). Comparing the profiles of diagnostic/remission paired patient samples, we show that hypermethylation was acquired in APL in a monoallelic manner. The cytosine guanine dinucleotide status of the DMR correlated with expression of the miRNAs following a characteristic position-dependent pattern. Moreover, a signature of hypermethylation was also detected in leukemic cells from an established transgenic PML-RARA APL mouse model at the orthologous region on chromosome 12, including the CTCF binding site located upstream from the mouse miRNA cluster. These results, together with the demonstration that the region does not show DNA methylation changes during myeloid differentiation, provide evidence that 14q32 hypermethylation is implicated in the pathogenesis of APL. We propose a model in which loss of imprinting at the 14q32 domain leads to overexpression of the miRNAs in APL.

Chronic lymphocytic leukemia (CLL) is a disease of an accumulation of mature B cells that are highly dependent on the microenvironment for maintenance and expansion. However, little is known regarding the mechanisms whereby CLL cells create their favorable microenvironment for survival. High-mobility group protein B-1 (HMGB1) is a highly conserved nuclear protein that can be actively secreted by innate immune cells and passively released by injured or dying cells. We found significantly increased HMGB1 levels in the plasma of CLL patients compared with healthy controls, and HMGB1 concentration is associated with absolute lymphocyte count. We therefore sought to determine potential roles of HMGB1 in modulating the CLL microenvironment. CLL cells passively released HMGB1, and the timing and concentrations of HMGB1 in the medium were associated with differentiation of nurse-like cells (NLCs). Higher CD68 expression in CLL lymph nodes, one of the markers for NLCs, was associated with shorter overall survival of CLL patients. HMGB1-mediated NLC differentiation involved internalization of both receptor for advanced glycation end products (RAGE) and Toll-like receptor-9 (TLR9). Differentiation of NLCs can be prevented by blocking the HMGB1-RAGE-TLR9 pathway. In conclusion, this study demonstrates for the first time that CLL cells might modulate their microenvironment by releasing HMGB1.

Background: Human follicular lymphoma (FL) and diffuse large B-cell lymphoma (DLBCL) are the most common forms of indolent and aggressive NHL, respectively. The t(14;18) translocation characterizes approximately 85% of FL and 20% of DLBCL and results in constitutive overexpression of the anti-apoptotic protein BCL-2. It was previously reported that BCL-2 plays dual roles in preventing apoptosis and autophagy. Autophagy is a physical and pathological process whereby cells sequester portions of cytoplasm including organelles to form autophagosomes where they are degraded and recycled. Growing evidence demonstrates that autophagy plays important roles in tumorigenesis, tumor progression, and resistance to chemotherapy.

Abstract Introduction Genetic aberrations of Tumor Necrosis Factor Receptor Superfamily 14 (TNFRSF14, also known as HVEM) have been shown to occur at high frequencies in patients (pts) with follicular lymphoma (FL). HVEM is a ligand for B and T lymphocyte attenuator (BTLA) which negatively regulates T cell responses and BTLA stimulation reduces acute graft-versus-host disease (aGvHD) in murine allogeneic hematopoietic cell transplantation (AHCT) models. As activated FL B cells are potent alloantigen presenting cells, we hypothesized that TNFRSF14 aberrations in FL B cells would reduce expression of HVEM and potentiate capacity of FL B cells to stimulate allogeneic T cell responses. We therefore sought to determine the functional effect of TNFRSF14 aberrations on FL B cell-stimulated donor T cell alloresponses in vitro. We also examined the impact of TNFRSF14 aberrations on the outcome of FL pts after HLA-matched reduced intensity conditioning (RIC) AHCT. Results FL B cells from lymph nodes were FACS-sorted (>90% purity and > 95% light chain restriction), activated and used as stimulators in mixed lymphocyte reactions with purified allogeneic responder CD3+ T cells. HVEM expression on FL B cells from pts with biallelic TNFRSF14 aberrations (Mut/Del cases) was undetectable whereas 40% of FL B cells from TNFRSF14 WT cases expressed HVEM (Fig 1 A). In contrast, FL B cells from Mut/Del and WT cases expressed similar levels of MHC class I/II, CD80, CD86 and CD58 before and after activation. Allostimulation with Mut/Del FL B cells resulted in significantly greater expression of activation markers on responder CD4+ T cells, increased secretion of pro-inflammatory cytokines (IFN-γ, TNF-α, and IL-2) measured by ELISA and increased frequencies of cytokine-secreting CD4+ and CD8+ T cells enumerated by intracellular cytokine staining. Responder T cell proliferation by thymidine incorporation was significantly greater after stimulation with Mut/Del FL B cells compared to WT FL B cells. CFSE labeling studies demonstrated that this effect resulted from increased proliferation of CD4+ and CD8+ responder T cells after both primary (Fig 1B) and secondary allostimulation. To determine if the increased alloresponses we observed using FL B cells from TNFRSF14 Mut/Del cases was due to reduced HVEM-BTLA signaling, we performed allogeneic co-cultures in the presence of antagonist or agonist BTLA antibodies (ab). Antagonist anti-BTLA ab increased proliferation of responder T cells after stimulation with WT FL B cells confirming that BTLA limits alloresponses in our in vitro model. Importantly, agonist BTLA ab reduced alloresponses stimulated by Mut/Del FL B cells. We next sought to determine if the increased alloresponses we detected in vitro in FL pts with TNFRSF14 aberrations resulted in an increase in clinical alloreactivity after AHCT. DNA from lymph nodes from FL pts undergoing T-cell replete RIC AHSCT was screened for TNFRSF14 mutations and deletions by Sanger sequencing and multiplex ligation-probe amplification respectively. Cumulative incidences (CI) of aGvHD and GvHD-related death were calculated with FL progression as a competing risk. TNFRSF14 aberrations were identified in 10/21 pts prior to RIC AHCT (4 Mut/Del, 1 Del/Del, 1 Mut/WT, 4 Del/WT). Most (18/21) pts had evidence of ongoing FL pre-transplant. Disease and donor characteristics were similar in pts with and without aberrations. There was no significant difference in CI of aGvHD in pts with or without TNFRSF14 aberrations. However there was a significantly higher CI of fatal aGvHD in patients with TNFRSF14 aberrations (45%) compared to those without aberrations (0%, p<0.01). Interestingly, relapse was less frequent in patients with TNFRSF14 aberrations consistent with increased graft-versus-tumor effects, although this did not reach statistical significance. Conclusion This study is the first to describe the impact of TNFRSF14 aberrations on the allostimulatory capacity of FL B cells. TNFRSF14 aberrations were associated with enhanced T-cell alloresponses in vitro and increased death from aGvHD. Importantly, our results suggest FL patients with TNFRSF14 aberrations may benefit from more aggressive immunosuppression to prevent fatal aGvHD after AHCT. The increased antigen-presenting capacity of FL B cells with TNFRSF14 aberrations could also influence autologous anti-tumor responses and impact outcome after other treatment modalities. Figure 1 Figure 1. Disclosures Gribben: Celgene: Research Funding; Pharmacyclics: Honoraria; Roche: Honoraria.

Overexpression of the anti-apoptotic protein BCL-2 is characteristic of human follicular lymphoma (FL) and some cases of diffuse large B cell lymphoma (DLBCL). We aimed to determine autophagy status in primary FL and DLBCL samples and the BCL-2(+)/BCL-2(-) lymphoma cell lines using both autophagy PCR array and tissue microarray (TMA). A greater number of autophagy machinery genes were up-regulated in the BCL-2(+) Su-DHL4 cell line compared with BCL-2-Su-DHL8 cells, at both the basal level and in response to autophagic stress. The autophagy-related gene expression profiles were determined in purified and unpurified malignant human lymph node biopsies. Seven autophagy machinery genes were up-regulated in purified FL B-cells compared with reactive B-cells. Only 2 autophagy machinery genes were up-regulated in DLBCL B-cells. In unpurified tissue biopsies, 20 of 46 genes in FL and 2 of 5 genes in DLBCL with increased expression were autophagy machinery genes. Expression of autophagy substrates p62 and LC3 were determined by TMAs. FL samples showed significantly decreased levels of both p62 and LC3 compared with reactive and DLBCL, indicative of an increased autophagy activity in FL. In summary, these results demonstrate that FL showed increased basal autophagy activity, regardless of overexpression of BCL-2 in this disease.