Each year, the Developmental Research Program supports pilot projects that take maximum advantage of new research opportunities in lymphoma. Through the DRP, the MSK SPORE in Lymphoma provides multidisciplinary resources to advance novel scientific ideas and support innovative pilot projects in translational lymphoma research. Projects are intended to rapidly advance a new idea or concept that has the potential to substantially impact our understanding of lymphoma and advance treatment of the disease.
“Role of BCL10 somatic mutations in lymphomagenesis and response to BCR-targeted therapies”
Lorena Fontan (Renewal)
Weill Cornell Medicine
Abstract: Diffuse large B-cell lymphoma (DLBCL), the most common lymphoid malignancy, is a molecularly heterogenous disease. Two recent genomic profiling studies independently identified a previously unnoticed subtype of DLBCL reminiscent of Marginal Zone Lymphoma (MZL), namely C1 or BN2 lymphoma. This novel subtype, characterized by frequent BCL6 translocations and activation of NOTCH and NF-kB pathways, displays frequent BCL10 mutations (30%), which are rare in other DLBCL subtypes (<2%) but relatively common in MZL (8%). BCL10 forms a high order complex (CBM) with CARD11 and MALT1, also lymphoma oncogenes, leading to NFkB activation. BCL10 is critical for MZ B-cell development and its overexpression mediates hyperproliferation and, eventually, MZL. However, the effect of BCL10 somatic mutations has not been studied. BCL10 mutations can be classified in: CARD domain missense and C-terminus truncating mutations. We hypothesize that CARD and C-terminal mutations induce gain-of-function through distinct molecular mechanisms.
Based in our preliminary results, we predict that: i) BCL10 gain-of-function mutations will accelerate BCL10 polymerization, rewire complex structure and composition, and induce constitutive activation of NF-kB and MALT1 protease activity through distinct molecular mechanisms; and ii) BCL10 gain-of-function mutations will confer resistance to classical BCR pathway kinase inhibitors, thus requiring targeting of downstream proteins such as MALT1 or alternative pathways. Our goals for this proposal are to elucidate the molecular mechanism by which specific BCL10 somatic mutation classes alter the high order molecular structure of the CBM complex and, to leverage this information to design novel therapeutic approaches for C1/BN2 lymphomas.
“Dual targeting of PRMT5 and MSI-2 as a novel therapy in lymphoma”
Memorial Sloan Kettering Cancer Center
Abstract: Non-Hodgkin lymphoma is one the most common cancers in adults that requires new treatment strategies. Recently B-cell lymphomas have been found to overexpress and require protein arginine methyltrasferase-5 (PRMT5) for their survival. Thus, small molecule inhibitors to PRTM5 have been developed as a new therapeutic strategy in lymphoma and other cancers. However, resistance to these inhibitors have been observed. In order to understand which factors drive resistance, the Younes laboratory performed a genome wide CRISPR screen to discover new regulators. The top hit as a driver of resistance is the MUSASHI2 (MSI2) RNA binding protein that has been implicated in a variety of hematological diseases, including AML, as a poor prognostic marker and regulator of leukemia stem cell function. The role for MSI2 in lymphoma is not known. Our preliminary data suggests a new role for MSI2 in lymphoma and as a key driver of resistance to PRMT5 therapy. This proposal will 1) characterize the relationship between MSI2 and PRMT5 and its regulation in lymphoma; 2) develop a mechanism-based combination strategy to overcome resistance mechanism. Overall, we will use new technologies to study MSI2’s targets and study the therapeutic consequences of targeting PRMT5 and MSI2 in lymphoma cell lines and in patient samples.
“Exploring the role of the H3K4 methyltransferase SETD1B in drug resistance and Lymphomagenesis”
Memorial Sloan Kettering Cancer Center
Abstract: Epigenetic regulators are among the most frequently altered genes in lymphoma. KMT2D/MLL2 is the most frequently mutated gene in lymphoma (frequencies between 50%-85%). KMT2D is one component of the COMPASS (Complex of Proteins Associated with Set1) involved in histone H3 lysine 4 (H3K4) methylation2. Other components of the complex (e.g. KMT2A, KMT2B, KMT2C, SETD1A and SETD1B) share the highly conserved catalytic SET domain and are less mutated (frequencies 1-15%). Across three genetic screens for resistance to inhibitors of BCL2, MCL1 and BTK, we repeatedly identified SETD1B/KMT2G as a highly significant screen hit. We have confirmed this resistance function. SETD1B is mutated in 15% of Diffuse Lager B Cell Lymphoma (DLBCL) and represents one of the 150 oncogenic drivers in DLBCL3. We found it is also mutated in 7% of follicular lymphoma (FL). Our preliminary results suggest loss of Setd1b leads to disease acceleration in vivo in our vavP-Bcl2 FL mouse model. We now propose to examine the role of SETD1B in lymphoma development and drug resistance. In order to define the key genes and pathways affected by SETD1B mutations, we will use gene expression, ChIP-seq and ATAC-seq studies on genetically engineered murine FLs, primary human lymphoma specimens, and isogenic pairs of cell lines with/without SETD1B. In additional, we will decipher potential avenues to bypass resistance or target SETD1B mutant disease by synthetical lethal screen with our druggable libraries. Our study will yield functional insights into the pathogenic role of a novel epigenetic lesion in lymphoma and explore potential therapeutic vulnerabilities.
“Targeting the NAD Salvage pathway in DLBCL”
Abstract: Novel therapies are needed for approximately 40% of Diffuse-Large-B-Cell Lymphomas (DLBCL) patients who do not respond to the standard immunochemotherapy. Classic drug-development starting from lead-compounds requires extensive time and resources and shows an abysmal 3.8% approval rate for cancer treatments. Repositioning is the process of redeveloping an existing drug for a different disease. Repositioning can uncover unexpected sensitivities in DLBCL and offer ready-to-use clinical strategies.
As a proof of concept, we have recently provided pre-clinical evidence that dasatinib, a Src/Abl inhibitor approved B-ALL and CML, can be repositioned for DLBCL therapy. Dasatinib is highly effective against a sizable group (50%) of DLBCL including both the GCB-and ABC-subtypes, widens the range covered by the Bruton Kinase (BTK) inhibitor Ibrutinib and can overcome Ibrutinib resistance by acting independently of BTK. Dasatinib resistance is characterized by activation of the PI3K pathway, identified by loss of PTEN expression as a specific biomarker. Accordingly, combination with mTORC2 inhibitors reverts dasatinib resistance in vitro and in vivo, suggesting dasatinib/mTORC2 inhibition as a promising therapeutic strategy. These studies are currently being translated to clinical trials for DLBCL. (Scuoppo et al., PNAS 2019).
The focus of this proposal is to develop the results of a subsequent effort, involving the screening of the whole FDA-approved set and the Columbia University Irving Medical Center (CUIMC) Experimental Oncology library, an additional set of promising compounds in advanced clinical testing. We identified FK866, a specific inhibitor of Nicotinamide Phosphoribosyl Transferase (NAMPT), a key enzyme in the NAD Salvage pathway, as the top candidate for a specific DLBCL subtype in a panel of 34 DLBCL lines. FK866 (Daporinad, Klinge Pharma) was discovered as the most potent hit in a screen for drugs able to induce apoptosis without concurrent DNA damage in myeloid leukemia cells and hepatocytes. FK866 is a specific inhibitor of Nicotinamide Phosphoribosyl Transferase (NAMPT), a key enzyme in the NAD Salvage pathway, that is the major route of NAD regeneration in mammalian cells. Of note, FK866 safety has already been positively evaluated in phase I/II clinical trials not involving lymphoma. A second molecularly distinct drug targeting NAMPT recapitulated FK866 activity, indicating FK866 acts as a specific NAMPT inhibitor. Based on these preliminary results, the goal of this proposal is to thoroughly develop the pre-clinical rationale for FK866 and other clinically developed NAMPT drugs against DLBCL. We will identify biomarkers of sensitivity and resistance and validate targets and compounds that can overcome resistance or synergize with NAMPT targeting in a large panel of genetically characterized DLBCL lines and Patient Derived Xenotransplant (PDX) models. These studies will provide the basis to establish clinical trials for DLBCL.
“Role of BCL10 somatic mutations in lymphomagenesis and response to BCR-targeted therapies”
Weill Cornell Medicine
Abstract: Diffuse large B-cell lymphoma (DLBCL) is the most common lymphoid malignancy. Among DLBCLs, the ABC DLBCL subtype is the most aggressive and manifests constitutive activation of NF-κB pathway, caused by somatic mutations of proteins involved in different signaling pathways. ABC DLBCL have been subclassified in C1 and C5 lymphomas characterized by different mutations and pathways. C5 features frequent B-cell receptor (BCR) and Toll-like receptor (TLR) mutations, while C1 constitutes a new subtype with frequent structural variants of BCL6 (amplification, gain or translocation) and NOTCH2 activation. C1 DLBCL has been postulated as transformed Marginal Zone Lymphoma.
Of central importance in ABC-DLBCLs, the CARMA1/BCL10/MALT1 (CBM) complex integrates signals from the BCR and TLR pathways and induces canonical and/or non-canonical NF-κB activation. This complex is overactive in C5 lymphomas because of mutations in CD79B, MYD88, CARD11 or MALT1 gain. Interestingly, C1 lymphomas are devoid of those mutations but feature frequent BCL10 mutations, which significance is currently unknown. BCL10 mutations affect two different regions in the protein mediating binding to CARD11 or homodimerization vs binding to MALT1. We hypothesize that BCL10 mutations induce gain of function and drive lymphomagenesis by activating CBM complex and NF-κB activity or perhaps other pathways. We hypothesize that: i) BCL10 gain-of-function mutations cause enhanced CBM complex activity by disrupting BCL10 autoinhibitory structure through distinct mechanisms based on specific biochemical effects of these different mutations; and ii) BCL10 gain-of-function mutations confer resistance to classical BCR pathway kinase inhibitors such as Ibrutinib, thus requiring targeting of downstream proteins such as MALT1.
“Mechanistic Role of VAV1 Alterations in T-cell Differentiation and Peripheral T-cell Lymphoma”
Columbia University – Institute of Cancer Genetics
Abstract: Peripheral T-cell lymphomas (PTCLs) represent a heterogeneous and poorly understood pathological group of non-Hodgkin lymphomas associated with poor prognosis. Despite major progress in recent years in the identification of genomic drivers in PTCLs, understanding their mechanisms of transformation and identifying therapeutic targets remain a high priority in the field. Recently, we have identified novel genetic alterations in the VAV1 oncogene using a combination of RNAseq analysis and targeted sequencing of candidate genes. Our data identified recurrent alterations in VAV1 in angioimmunoblastic T-cell lymphomas (AITL) and PTCL not otherwise specified (PTCL-NOS) samples. Most genomic alterations are gene fusions and small intragenic deletions affecting the C-terminal domain of the protein. Our central hypothesis is that the VAV1 alterations lead to increase activation of signaling pathways downstream of VAV1 and act as oncogenic drivers of PTCL. Here we will identify the mechanisms and pathways by which these genomic alterations contribute to T-cell transformation and analyze the oncogenic effects of the Vav1-myo1f fusion oncogene in the pathogenesis of PTCL using lymphoma mouse models.
“Functional bypass of cell cycle entry checkpoints by MYC mutations in aggressive B-cell
Mount Sinai - Icahn School of Medicine
Abstract: In a large fraction of Burkitt lymphomas, the MYC gene is targeted by recurrent missense mutations clustered around mutational hotspots. These MYC mutants have enhanced oncogenic potencies linked to lower apoptotic responses when overexpressed. However, the mechanistic basis and true biological meaning for the recurrent selection of these mutations in MYC-driven aggressive lymphomas is unclear. Based on recent observations, we hypothesize that these mutations bypass a PI3K-dependent cell cycle entry checkpoint triggered by incongruent mitogenic signals. This checkpoint is strictly controlled through T cell-dependent
affinity selection events in normal germinal center B cells. We propose to use a combination of genetic and pharmacological strategies in newly developed mouse models to specifically address the relevance of this circuitry during normal B cell responses and B cell lymphomagenesis.
Specifically, we will: (1) Investigate the functional dependence of MYC missense mutants on PI3K and antigen receptor signals, by manipulating the strength of PI3K signaling through pharmacological and genetic strategies and evaluate the dependency of MYC mutants on PI3K signaling during clonal B cell expansion, both ex vivo and in vivo. And (2) Determine the contribution of affinity selection events to MYC-dependent B cell lymphomagenesis, by editing in vivo B cell receptor affinities in the context of adaptive immune responses in mice carrying wild type or mutant MYC transgenes. These studies address an important aspect of B cell lymphoma pathogenesis, and can provide a mechanistic explanation to the observed synergism between MYC and PI3K during lymphomagenesis, with potential therapeutic implications.
“Targeting oncoprotein expression in follicular lymphoma and mantle cell lymphoma patients with indolent disease: a biomarker-driven proof-of-concept clinical trial”
Weill Cornell Medicine
Abstract: Follicular Lymphoma (FL) usually presents with disseminated disease and follows an indolent, although incurable, clinical course. During the indolent phase, rather than increased proliferation, accumulation of FL cells is a consequence of the almost universal expression of BCL2 that make these cells resist apoptosis. Mantle cell lymphoma (MCL) is another type of lymphoma that can have an indolent course. It is characterized by (11,14) translocation which leads to cyclin D1 (CCND1) overexpression.
We and others identified eIF4E as a critical factor for the nuclear export and preferential translation of lymphoma oncogenic mRNAs. Accordingly, eIF4E inhibition using the antiviral ribavirin results in decreased production of BCL2, CCND1, BCL6, LYN and AKT among others proteins. The advantage of oncogenic mRNA translation as a therapeutic target is that iFL cells require constant production of these proteins to survive even if they are not proliferating. We therefore postulate that ribavirin could constitute a therapeutic strategy for lymphoma patients with indolent disease.
Our specific aim is to conduct a proof-of-mechanism clinical trial (Phase 0) to determine the tolerability and biological effect of ribavirin in patients with iFL and iMCL. The main endpoint will be a change in a pharmacodynamics biomarker (i.e. amount of DNA with BCL2 or CCND1 translocation in cell-free circulating DNA). We will enroll iFL and iMCL patients with BCL2 or CCND1 translocations and expression of eIF4E in the nucleus of lymphoma cells. Patients will be followed by serial assessment of BCL2/CCND1 translocation in cf-DNA. We expect that ribavirin treatment will lead to a decrease of BCL2/CCND1 presence in cf-DNA that will be followed by clinical benefit.
“Genetic determinants of primary chemoresistance in diffuse large B cell lymphoma”
Columbia University Institute for Cancer Genetics
Abstract: Diffuse large B-cell lymphoma (DLBCL) represents the most common B cell lymphoma in adulthood, accounting for 30-40% of all diagnoses. Despite significant advances in the management and treatment of this disease, a significant proportion of patients are not cured; prognosis is especially dismal in patients progressing during first-line therapy or early after treatment (primary-refractory DLBCL), with 4-yr overall survival of only 13%. The molecular basis and the specific mechanisms that are responsible for this poor therapeutic response are currently unknown. The general goal of this project is to provide pilot data toward the identification of genetic determinants of primary chemo-resistance in DLBCL, with the following specific aims:
1) Identify genetic alterations associated with primary-refractory DLBCL, by integrating state-of-the-art genomic profiling approaches and computational methods in clinically-annotated, pre-treatment tumor specimens from rigorously defined chemoresistant (CR) and chemo-sensitive (CS) DLBCL patients; 2) Provide initial functional characterization of the top scoring mutant alleles, including the DTX1 candidate gene identified in a preliminary discovery screen. The results of this study are expected to provide insights into the genetic landscape of CR-DLBCL, with relevant implications for the diagnosis and clinical management of this poor prognostic category of patients.
“Potentiating Antibody-Dependent Killing of Lymphomas by Use of a CAR T Cell that Blocks CD47”
Memorial Sloan Kettering Cancer Center
Abstract: CAR T cells have emerged as an effective therapy for B-cell hematopoietic cancers. However, greater potency and mechanisms to defeat resistance are still needed. This includes antigen loss variant escape and relapse. To mitigate this problem, we propose an innovative strategy that overcomes tumor resistance by modifying CAR T cells to constitutively express and secrete a drug that activates macrophages and vastly stimulates their ability to kill via antibody mediated phagocytosis (ADCP), which is the major mechanism of antibody mediated killing in vivo. Selective synthesis and local secretion of the activating drug at the lymphoma by the CAR T cell should reduce systemic toxicity of the combination. The mechanism of action of these advanced CAR T cells is that the drug, CV1, blocks the phagocytosis and ADCP-inhibiting protein, CD47, on the lymphomas. CD47 transduces an anti-phagocytic signal via binding to its cognate ligand, SIRPα, expressed on macrophages, neutrophils and NK cells. CV1 is a soluble, truncated SIRPα protein variant that potently blocks binding of tumor cell CD47 to phagocyte SIRPα, and improves macrophage-mediated ADCP of cancer cells in vitro and in mice. Here we plan to construct a CAR T cell that secretes CV1 to increase the phagocytic function of macrophages locally and more specifically to engage targets of infused antibodies such as CD20. Orexigenic drugs stimulate the appetite; thus CV1 is orexigenic to macrophages. Accordingly, we term this new CAR T cell technology as, “OrexiCARs.” Local CV1 secretion should improve potency through synergy with antibody therapy, and reduce systemic toxicity compared to CD47 antibody therapies.
“Characterization of clonal evolution using cell free DNA from lymphoma patients”
Abstract: Cell free DNA is known to be elevated in malignant conditions and has been applied to solid tumors to monitor tumor dynamics. Analysis of cell free tumor DNA is a young field with early successes of tracking minimal residual disease by high throughput sequencing of the immunoglobulin gene. However this approach does little to reveal clonal evolution and heterogeneity throughout a disease course. Analysis of genomic dynamics through cell free tumor DNA sequencing will subsequently aid biomarker development by non-invasive molecular stratification, monitoring treatment responses and identifying genomic mechanisms of resistance. This proposal is a proof of principle study using cell free DNA in patients with lymphoma to study tumor dynamics.
“Characterizing and modulating the immune evasion landscape in Primary CNS Lymphoma disease models and patients”
Abstract: Diffuse large-B-cell lymphoma represents the most common type of malignant lymphoma and is heterogeneous in its clinical presentation. Involvement of the central nervous system (CNS) occurs as primary CNS lymphoma (PCNSL), with exclusive disease manifestation in the CNS, or as secondary CNS dissemination (SCNSL). Patients with disease that does not respond to or recurs after initial therapy have a poor prognosis. As PCNSL tissue samples are mainly collected through biopsies, there is a paucity of tumor material with a lack of detailed pathophysiologic investigations. Therefore, the mechanisms of immune evasion in PCNSL are largely unknown. We have established a large PCNSL tissue bank with more than 170 tumor samples. We will characterize immune checkpoint markers in our tumor bank with a novel multiplexing immunofluorescence technology platform (MultiOmyx) allowing to quantify the expression of >60 proteins at single cell resolution in a single tissue section. All PCNSL tumor samples have been sequenced with the MSK-HemePACT panel and immune checkpoint marker protein expression can be associated with underlying genomic alterations. Unique PCNSL xenografts established by our group will be used to modulate immune evasion in vivo. The Bruton Tyrosine Kinase (BTK) inhibitor ibrutinib was been used in the MSKCC IRB#14-184 trial and shown to be effective in CNS lymphoma. Furthermore it has been shown to trigger tumor growth inhibition independent of its BTK function through immune modulation. We will build on these findings and propose a combination trial with ibrutinib and an immune checkpoint inhibitor to further improve clinical outcome.
“Mechanisms of response and resistance to checkpoint inhibitor-based treatments in lymphomas”
Abstract: Relapsed lymphomas are typically incurable in the absence of a stem cell transplant, underscoring the need for durable, well-tolerated treatments. Recently, treatments aimed at amplifying the anti-tumor immune response through blockade of the inhibitory immunoreceptor PD-1 have shown activity in multiple lymphoma subtypes; however, complete remissions remain infrequent. As a result, a series of combination strategies have recently been developed at MSKCC with the goal of deepening responses to immunotherapy across multiple lymphoma subtypes. These include facilitation of antigen release (by combining anti-PD-1 therapy with radiation therapy) as well as increasing antigen and co-stimulatory machinery while depleting inhibitory regulatory T-cells (by combining with histone deacetylase inhibition).
I propose to investigate the mechanisms by which combination immunotherapeutic strategies promote anti-tumor responses in relapsed and refractory lymphomas through a series of correlative analyses performed on primary patient samples obtained as part of 3 clinical trials opening in the coming year at MSKCC. Specifically, through a combination of immunohistochemistry, T-cell immunophenotyping, and high throughput T cell receptor sequencing, I will determine biomarkers of response and resistance to these therapeutic strategies. These analyses are critical to gaining a better understanding of how anti- PD-1 therapy functions in lymphoma and will lead to both identification of clinically applicable biomarkers as well as development of next-generation, biologically tailored treatment strategies.
“Dissecting SHMT2 function as a metabolic driver of lymphomagenesis”
Collaborator: Sara Parsa
Abstract: We find that the serine hydroxymethyltransferase-2 gene (SHMT2) is amplified in approx. 30% of human B cell lymphomas including follicular and diffuse large B cell lymphomas. Moreover, our preliminary data show that SHMT2 can act as a driver of lymphoma development in a mouse lymphoma model in vivo. To our knowledge SHMT2 is therefore a first metabolic enzyme that can act as an oncogene in vivo. We propose a study to dissect the metabolic effects of SHMT2 activation in lymphoma, its contribution to lymphoma progression, and we will explore therapeutic implications of SHMT2 activity. Specifically, we have developed a murine model of follicular lymphoma that captures the early stages of the disease and is ideally suited to study genetic drivers of lymphoma development and progression and characterize their biochemical and gene expression effects as well as test new therapies in genetically engineered tumors. We will further investigate the mechanistic effect of SHMT2 activation on lymphomagenesis using a combination of metabolomics and in vivo CRISPR-Cas9 based genetic screening approaches. A better understanding of the role of SHMT2 in Myc amplified lymphoma cells will enable identification of new therapies directed towards aberrant metabolism in these cancers. Metabolic aspects of lymphoma biology and their immediate goal is to strengthen the existing preliminary data towards a full research project.