Our Science – Staudt Website
Louis M. Staudt, M.D., Ph.D.
Our laboratory studies the molecular pathogenesis of human lymphoid malignancies and has three primary goals: to establish a new molecular diagnosis of human lymphoid malignancies using gene expression profiling, to elucidate the oncogenic pathways that result in malignant transformation of normal B lymphocytes, and to identify molecular targets for development of novel therapeutics for these cancers.
To provide a molecular basis for the diagnosis of human lymphoid malignancies, we are exploiting DNA microarray technology to profile gene expression in these cancers on a genomic scale. The laboratory created a novel DNA microarray, the 'Lymphochip', which is enriched in genes that are expressed in and/or function in lymphocytes (1). We have used Lymphochip and Affymetrix microarrays to profile gene expression in diffuse large B cell lymphoma (DLBCL) (2-4), chronic lymphocytic leukemia (CLL) (5, 6), mantle cell lymphoma (7), follicular lymphoma (8), multiple myeloma (9), and in a wide variety of normal lymphoid subsets (2, 10-13).
One central goal of these studies is to relate gene expression to clinical outcome, thereby establishing a quantitative, reproducible and informative molecular diagnosis of the lymphoid malignancies (14). Our studies have revealed previously unknown types of diffuse large B cell lymphoma that are indistinguishable by current diagnostic methods, but which have strikingly distinct gene expression profiles, originate from different stages of B cell differentiation, utilize distinct oncogenic mechanisms, and differ in their ability to be cured by current chemotherapy (2-4). For several lymphoid malignancies, we have identified molecular profiles that predict the length of survival or the ability to be cured by chemotherapy, thereby providing clinically useful prognostic indicators. Our laboratory has mounted a major effort to create a diagnostic microarray that could provide these molecular diagnoses and prognoses to patients with lymphoid malignancies.
Our laboratory uses functional genomics, chemical genetics and molecular biological techniques to identify new molecular targets for therapy of lymphoid malignancies. Some of the genes that are associated with clinical prognosis have provided new molecular targets. For example, our laboratory discovered that the subgroup of DLBCL with the worst prognosis relies on constitutive activity of the NF-kB signaling pathway for survival; molecular or pharmacological inhibition of this pathway kills this type of lymphoma.
A major new initiative aims to identify molecular targets in lymphoid malignancies using large scale RNA interference. The laboratory has created a library of over 10,000 retroviruses that can inducibly express small hairpin RNAs (shRNAs) targeting more than 3,000 human genes. When expressed in a cell, each shRNA can be processed into a small interfering RNA that can decrease the mRNA expression of a single human gene. We are using this library to identify genes that are important for the proliferation and survival of lymphoma cells.
MOLECULAR DIAGNOSIS OF LYMPHOID MALIGNANCIES
Molecularly and clinically distinct diseases within diffuse large B cell lymphoma (DLBCL)
DLBCL has long been enigmatic in that 40 percent of patients can be cured by combination chemotherapy whereas the remainder succumb to this disease. By gene expression profiling, the laboratory discovered that DLBCL is actually comprised of at least three different diseases that are indistinguishable by current diagnostic methods (2-4, 15). As detailed below, these DLBCL subgroups can be considered distinct diseases in that they originate from B cells at different stages of differentiation, utilize distinct oncogenic mechanisms, and differ significantly in their survival rates following chemotherapy.
DLBCL subgroups originate from distinct stages of B cell development
One subgroup of DLBCL, termed germinal center B cell-like (GCB) DLBCL, expresses genes that are hallmarks of normal germinal center B cells. By contrast, another DLBCL subgroup, termed activated B cell-like (ABC) DLBCL, lacks expression of germinal center B cell-restricted genes and instead expresses genes that are induced during mitogenic stimulation of blood B cells (2). These two subgroups of DLBCL differ in the expression of thousands of genes, and in this respect they are as different as acute myelogenous leukemia is from acute lymphoblastic leukemia.
Clues to the normal cellular counterparts of these DLBCL subgroups have been provided by our laboratory's analysis of regulatory factors that control the differentiation of germinal center B cells to plasma cells. We and others showed that BCL-6 is a transcriptional repressor that is required for mature B cells to differentiate into germinal center B cells during an immune response (16, 17). Normal germinal center B cells express BCL-6 at high levels but BCL-6 expression is silenced during plasmacytic differentiation. DLBCLs belonging to the GCB subgroup express BCL-6 at high levels but those belonging to the ABC subgroup do not. BCL-6 is deregulated by chromosomal translocations in roughly 20% of DLBCLs, but the high expression of BCL-6 in GCB DLBCLs is not accounted for by these translocations. Rather, BCL-6 is expressed in GCB DLBCLs along with a host of other germinal center B cell restricted-genes because these DLBCLs are derived from normal germinal center B cells and retain much of their biology. In keeping with this notion, GCB DLBCLs have ongoing somatic hypermutation of their immunoglobulin genes, a characteristic feature of normal germinal center B cells (18).
The cell of origin of ABC DLBCL has not been fully elucidated, but may be a plasmablastic B cell that is poised to exit the germinal center. Support for this notion comes from our laboratory's analysis of two regulatory factors that are required for plasmacytic differentiation, Blimp-1 (19) and XBP1 (20). By gene expression profiling, our laboratory demonstrated that BCL-6 blocks the expression of Blimp-1, and when BCL-6 activity was inhibited in a lymphoma cell line, Blimp-1 was induced and plasmacytic differentiation was initiated (21). We went on to show that Blimp-1 is a transcriptional repressor that extinguishes the expression of virtually all germinal center B cell genes, including BCL-6 (22). Blimp-1 and BCL-6 thus form a double negative autoregulatory loop that controls plasmacytic differentiation. Blimp-1 enables the expression of XBP1, which our laboratory showed is a master regulator of the secretory phenotype of plasma cells (23). XBP1 induces the expression of a large set of genes encoding components of the endoplasmic reticulum and golgi, leading to a dramatic expansion of the secretory apparatus (23). In addition, XBP1 increases the overall rate of protein synthesis by 50%, which contributes to the high rate of immunoglobulin secretion by plasma cells (23).
In comparison to GCB DLBCLs, ABC DLBCLs are characterized by high expression of XBP1 and its target genes, as well as Blimp-1 (4). This phenotype is similar to that of a rare subpopulation of plasmablasts in the germinal center, which are thought to be in the process of migrating to the bone marrow where they will differentiate fully into plasma cells (24, 25). ABC DLBCLs do not express a variety of other genes that characterize normal plasma cells and multiple myeloma, suggesting that they are derived from a cell that is intermediate between a germinal center B cell and a plasma cell. In support of this notion, ABC DLBCLs have somatically mutated immunoglobulin genes, and therefore are derived from a B cell that has likely traversed the germinal center (18). However, in contrast to GCB DLBCLs, ABC DLBCLs have a fixed complement of immunoglobulin gene mutations, suggesting that the somatic hypermutation machinery has been inactivated as occurs normally during plasmacytic differentiation.
Primary mediastinal B cell lymphoma: A distinct subgroup of DLBCL related to Hodgkin lymphoma
Recently, we and others developed a molecular diagnosis of a third subgroup of DLBCL, termed primary mediastinal B cell lymphoma (PMBL) (15, 26). PMBL cannot be reliably distinguished from other types of DLBCL by current clinical criteria. PMBL was readily distinguished from GCB and ABC DLBCL by the expression of hundreds of genes, and we were able to develop a molecular diagnosis of PMBL. Clinically, PMBL patients were younger (median age 33) than patients with GCB or ABC DLBCL (median age over 60). PMBL lymphomas frequently extended from the mediastinum into other thoracic structures, but did not involve the extrathoracic sites that are typical of other DLBCLs.
A striking relationship was found between PMBL and Hodgkin lymphoma by gene expression profiling. Over one third of the genes that were characteristically expressed in PMBL were also expressed in Hodgkin lymphoma cell lines. We confirmed the expression of several of these genes in primary Hodgkin Reed Sternberg cells from nodular sclerosing Hodgkin lymphoma. In light of this strong molecular resemblance, it is particularly notable that PMBL and nodular sclerosing Hodgkin lymphoma share numerous clinical and pathological features, including prevalence in young patients, especially women, prominent sclerosis histologically, and primary presentation in the mediastinum, often with thymic remnants evident upon pathological examination.
It is likely, therefore, that PMBL and some forms of Hodgkin lymphoma arise from a rare subpopulations of thymic B cells (27, 28).
Distinct Oncogenic Mechanisms in DLBCL subgroups
Analysis of recurrent oncogenic changes in DLBCL support the view that the DLBCL subgroups represent distinct disease entities. Two recurrent oncogenic abnormalities in DLBCL, BCL-2 translocation and c-rel amplification, are present in 45% and 16% of GCB DLBCLs, respectively, but have never been detected in ABC DLBCLs (3, 29). Comparative genomic hybridization has revealed that these two DLBCLs have strikingly different frequencies of certain genomic abnormalities. For example, gains of chromosome arm 3q are detected in 24% of ABC DLBCLs, but are never detected in GCB DLBCLs. As mentioned above, the 9p24 chromosomal region is amplified in 43% of PMBLs, but this abnormality is not observed in GCB DLBCL.
The DLBCL subgroups also utilize distinct signaling pathways to promote their survival and/or proliferation. Our laboratory demonstrated that the anti-apoptotic NF-kB pathway is constitutively active in ABC DLBCLs but not GCB DLBCLs (30). More recently, we and others demonstrated constitutive activity of the NF-kB pathway in PMBL (15, 26). Importantly, inhibition of the NF-kB pathway in ABC DLBCL or PMBL cell lines induces apoptosis whereas GCB DLBCL cell lines are not affected (30, 31). These findings demonstrate that the molecular circuits of the DLBCL subgroups are dramatically different, which may translate into differential susceptibility to particular therapies.
DLBCL subgroups have distinct cure rates with multiagent chemotherapy
Our initial study of 42 patients with DLBCL demonstrated that patients with GCB DLBCL had a strikingly superior rate of cure following chemotherapy compared to patients with ABC DLBCL (1). Based on these results, the NCI expanded the scope of our laboratory's gene expression profiling analysis by sponsoring the Lymphoma/Leukemia Molecular Profiling Project (LLMPP; http://llmpp.nih.gov). This international collaborative project involves eight groups that have maintained biopsy samples and clinical data from hundreds of lymphoma patients. In the initial study by the LLMPP consortium, we profiled gene expression in 240 biopsy samples of DLBCL (3). This study confirmed that patients with GCB DLBCL and ABC DLBCL had distinct 5-year survival rates of 60% and 35%, respectively.
To provide a robust method for the molecular diagnosis of cancer subgroups clinically, we created a Bayesian statistical algorithm that determines the probability that a patient's tumor belongs to a particular cancer subgroup (4). Based in this method, the GCB, ABC, and PMBL subgroups account for 40%, 34%, and 8% of DLBCLs, respectively. The 5-year survival rates for the GCB, ABC, and PMBL subgroups defined in this fashion were 59%, 31%, and 64%, respectively.
GENE EXPRESSION-BASED SURVIVAL PREDICTION
A molecular predictor of survival following chemotherapy for diffuse large B cell lymphoma
The subdivision of DLBCL into distinct gene expression subgroups accounts, in part, for the differences in survival of these patients following multiagent chemotherapy. However, further scrutiny of the gene expression profiling data revealed many additional genes whose expression patterns were associated with the length of survival. These predictive genes reflected different biological features of the tumors that influenced the ability of chemotherapy to cure these patients. The relationship of these predictive genes to particular biological processes was revealed by assigning them to 'gene expression signatures' (13). A gene expression signature is operationally defined as a group of coordinately regulated genes that are characteristically expressed in a particular cell type or stage of differentiation, or are expressed during a particular cellular response to internal or external stimuli.
All together, 55% of the survival-associated genes could be classified into one of these four signatures, suggesting that these signatures capture the majority of biological variation that influences survival in DLBCL. A composite molecular predictor of survival was created from these four gene expression signatures and was used to divide the DLBCL patients into four equal groups with widely differing 5-year survival rates of 73%, 71%, 36% and 15%.
The majority of the genes that predicted outcome following chemotherapy for DLBCL could be classified into one of four gene expression signatures. Expression of genes in the germinal center B-cell signature predicted favorable outcome, a finding that mirrors the DLBCL subgroup distinction described above. Two other gene expression signatures that predicted favorable outcome may reflect differences in the way the malignant cell interacts with the non-malignant immune cells in the lymph node. One of these signatures includes components of the MHC class II antigen presentation pathway. Variation in the MHC class II gene expression signature is associated with differences in expression of MHC class II molecules on the surface of the malignant cells, which could influence the anti-tumor T cell immune response. The 'lymph node' gene expression signature reflects variation in the quantity and characteristics of the non-malignant cells in the involved lymph nodes of DLBCL patients. Many of the lymph node signature genes are expressed in macrophages and NK cells, suggesting that the favorable prognosis associated with this gene expression signature could reflect an innate immune response to the tumor that contributes to curative responses with chemotherapy. Many of the genes whose expression was associated with poor outcome in DLBCL belong to the proliferation gene expression signature. This signature includes genes that are expressed at high levels in dividing cells and at lower levels in quiescent cells (13). One of the most predictive genes in the proliferation signature was c-myc, which encodes a growth-promoting oncogene that has already been implicated in the pathogenesis of DLBCL and other non-Hodgkin lymphomas.
A molecular predictor of survival following diagnosis of mantle cell lymphoma
Mantle cell lymphoma accounts for roughly 8% of all non-Hodgkin lymphomas, but contributes disproportionately to the number of deaths from lymphoma since there is no curative chemotherapy. The hallmark of mantle cell lymphoma is the t(11;14) translocation, which deregulates the expression of cyclin D1, a key regulator of transition from G1 to S phase in the cell cycle. By gene expression profiling, mantle cell lymphoma could be readily distinguished from DLBCLs and from small lymphocytic lymphoma (7). The characteristic gene expression signature of mantle cell lymphoma allowed us to identify a new subgroup of mantle cell lymphoma that is cyclin D1-negative. This subgroup, which accounts for ~3% of cases, was indistinguishable from cyclin D1-positive mantle cell lymphomas by histology and by gene expression but lacked the t(11;14) translocation.
In mantle cell lymphoma, survival times range from less than 1 year to more than 10 years following diagnosis. DNA microarrays were used to correlate gene expression in 91 cyclin D1-positive mantle cell lymphoma biopsies with the length of survival following diagnosis (7). Genes belonging to the proliferation signature were more highly expressed in tumors of patients with short survival. A molecular predictor of survival based on the expression of these proliferation signature genes could stratify patients into 4 quartile groups with median survival times of 0.8 yr, 2.3 yr, 3.3 yr, and 6.7 yr. This finding suggests that the proliferation signature provides a quantitative measure of the tumor proliferation rate in mantle cell lymphoma that accounts for much of the variability in survival following diagnosis.
The proliferation gene expression signature integrates multiple molecular features of the mantle cell lymphomas that influence their proliferation rate, including variation in cyclin D1 expression and deletion of the INK4A/ARF locus (7). Based on these observations, we proposed a model in which oncogenic changes in mantle cell lymphoma quantitatively alter the rate of progression from G1 to S phase and are reflected in the proliferation gene expression signature.
A molecular predictor of survival following diagnosis of follicular lymphoma
The clinical course of follicular lymphoma is highly variable: the median survival is approximately 10 years, but some patients live more than 15 years following diagnosis whereas others succumb to this disease in less than five years. In some cases, the malignancy transforms into DLBCL, which is rapidly fatal. Patients with follicular lymphoma are managed by watchful waiting or are treated with chemotherapy and/or various forms of immunotherapy. However, no definitive evidence has been presented that any of these approaches provide a survival advantage, and therefore there is no consensus as to the best treatment for these patients.
Using DNA microarray analysis, our laboratory revealed that the length of survival in follicular lymphoma can be predicted by the gene expression profile of the tumor at the time of diagnosis (8). We created a multivariate model of the length of survival in follicular lymphoma that was comprised of two gene expression signatures, termed immune response-1 and immune response-2. Expression of the immune response-1 signature was associated with long survival whereas expression of the immune response-2 signature was associated with short survival. Patients were assigned a survival predictor score based on this statistical model. These scores were used to divide the patients into four quartiles that had strikingly disparate median survival times of 3.9, 10.8, 11.1 and 13.6 years, respectively.
We demonstrated that the majority of the genes in the immune response-1 and immune response-2 signatures were expressed preferentially in the CD19-negative non-malignant cells of the tumor. These finding demonstrate that the character of the tumor-infiltrating immune cells in follicular lymphoma is the predominant feature that predicts the length of survival following diagnosis. The immune response-1 signature reflects a complex mixture of T cells and other immune cells that is associated with long survival in follicular lymphoma. The immune response-2 signature is associated with short survival in follicular lymphoma and appears to reflect an immune infiltrate that is relatively low in T cell content and relatively enriched in cells of the myeloid lineage.
Follicular lymphoma is one of the cancers in which spontaneous regression has been reported, although this occurs rarely. Remissions in the absence of therapy have also been noted in melanoma and renal cell carcinoma, a phenomenon that has been ascribed to an anti-tumor immune response in some patients. In this regard, the favorable prognosis associated with expression of the immune response-1 signature suggests that this signature may reflect a type of immune response that is capable of limiting the progression of follicular lymphoma. Thus, the immune response-1 signature could represent an adaptive immune response to the lymphoma. By contrast, the genes that constitute the immune response-2 signature do not encode T cell markers but rather encode markers of cells in the innate immune system. In follicular lymphomas with high expression of the immune response-2 signature, the infiltrating immune cells may be responding to 'danger' signals derived from the malignant cells.
While it is possible that the immune response-1 signature reflects an adaptive immune response to the malignant cells that is associated with longer survival, other models are conceivable. The infiltrating immune cells may provide trophic signals that either increase proliferation or prolong survival of the malignant clone. In this scenario, the malignant clone may be 'addicted' to the trophic factors from the immune cells, and this dependence may limit the ability of tumor cells to spread beyond the lymph node to anatomical locations lacking these immune cells.
The prognostic power of the gene expression-based survival predictor should prove helpful in patient management. Watchful waiting is a reasonable clinical approach for those patients in the top three quartiles since they have a rather indolent form of this lymphoma. On the other hand, those patients assigned to the fourth quartile have an aggressive lymphoma and should be considered for clinical trials involving novel therapies. The gene expression-based survival predictor should be particularly helpful in designing clinical trials in follicular lymphoma. Since the median survival of these patients is roughly 10 years, it has not been possible to conduct clinical trials in follicular lymphoma in which overall survival is the primary endpoint. If a clinical trial is designed to enroll patients in the least favorable quartile of the gene expression-based survival predictor, overall survival could be an achievable endpoint since the median survival of these patients is 3.9 years.
ZAP70 expression as the best predictor of clinically distinct types of chronic lymphocytic leukemia
Two clinically distinct subgroups of chronic lymphocytic leukemia (CLL) have been described that are distinguished by the presence or absence of immunoglobulin variable gene (IgVH) mutations in the leukemic cells (32, 33). Patients whose leukemic cells have unmutated IgVH genes have a median survival of ~8 years whereas patients whose leukemic cells have mutated IgVH genes have a median survival of greater than 20 years (34, 35).
By gene expression profiling, our laboratory identified ZAP-70 as the gene whose expression best discriminated these two types of CLL (6). Among 107 cases of chronic lymphocytic leukemia, the expression of ZAP70 predicted the dichotomy between the two CLL subgroups with 93% accuracy (5). The prognostic significance of ZAP70 expression has been reproduced in several independent patient cohorts (36-38). Since IgVH sequencing is impractical in a clinical diagnostics laboratory, mRNA or protein-based assays for ZAP70 expression should be of clinical utility..
MOLECULAR TARGETS IN LYMPHOID MALIGNANCIES
Targets in diffuse large B cell lymphoma
The subdivision of DLBCL into gene expression subgroups has also identified new molecular targets in these lymphomas. Our laboratory discovered that a critical molecular difference between the GCB and ABC DLBCL subgroups is the activation of the NF-kB pathway (30). ABC DLBCLs, but not GCB DLBCLs, express a number of known target genes of the NF-kB transcription factors, including IRF-4 and cyclin D2. More recently, we and others showed that PMBL tumors also express NF-kB target genes (15, 26).
ABC DBLCLs express NF-kB target genes because they have constitutive activity of the IkB kinase, which phosphorylates IkB, leading to its proteosomal degradation (30). Since IkB inhibits the activity of the NF-kB transcription factors, its degradation in ABC DLBCLs accounts for their expression of NF-kB target genes. GCB DLBCLs, by contrast, have little if any activity of the NF-kB pathway, in keeping with the fact that normal germinal center B cells do not express NF-kB target genes appreciably (13). Inhibition of the NF-kB pathway in ABC DLBCL cells is lethal whereas the survival of GCB DLBCLs does not depend upon NF-kB. These results validate the NF-kB pathway as a molecular target in ABC DLBCL.
To develop more specific and selective inhibitors of the NF-kB pathway for the treatment of lymphoma patients, we have evaluated a small molecule IkB kinase inhibitor of the beta-carbolene class, termed MLX105 (31). Treatment of ABC DLBCL or PMBL cell lines with MLX105 induces apoptosis, but this treatment has no effect on GCB DLBCL cell lines. The toxicity of MLX105 for ABC DLBCL and PMBL cells can be reversed by expression of the p65 subunit of NF-kB, demonstrating that the toxicity of this compound is due to NF-kB inhibition and not to off-target effects. Thus, both ABC DLBCL and PMBL have constitutive IkB kinase activity, which they require for survival. The as yet unknown mechanisms leading to IkB kinase activity in these lymphomas is currently under investigation. These results support the further development of selective IkB kinase inhibitors for the treatment of patients with these lymphoma types.
Targets in multiple myeloma
Translocations of the c-maf gene occur in roughly 5-10% of myelomas. However, we discovered that c-maf is overexpressed in ~50% of myeloma cell lines and patient samples, including many that lack a c-maf translocation (9). These findings demonstrate that c-maf overexpression is one of the most frequent oncogenic events in multiple myeloma. c-maf is a transcription factor belonging to the B-ZIP family. By gene expression profiling, we found that c-maf activates expression of three important genes: integrin ÃƒÅ¸7, C-C chemokine receptor 1, and cyclin D2, each of which may contribute to its action as an oncogene.
Inhibition of c-maf function in myeloma cell lines had two important biological consequences: decreased proliferation and decreased ability to interact with bone marrow stromal cells (9). It is likely that c-maf regulates proliferation, at least in part, by transactivating cyclin D2. Importantly, dominant inhibition of c-maf function blocked the ability of c-maf-expressing myelomas to form tumors in immunodeficient mice, presumably due to decreased proliferation. In addition, the increased integrin ÃƒÅ¸7 on the surface of c-maf-expressing myelomas caused them to adhere more avidly to bone marrow stroma. As a consequence, the bone marrow stroma produced larger quantities of VEGF, a cytokine that increases angiogenesis and promotes proliferation of myeloma cells. These findings suggest that therapeutic agents aimed at c-maf or its downstream targets could be beneficial for patients whose myelomas express c-maf.
Achilles heel RNA interference screens for new therapeutic targets
We have embarked on a major new initiative to use large-scale RNA interference screens to identify genes that are required for the abnormal proliferation and survival of cancer cells (39). To perform such Ã¢â‚¬Å“Achilles heelÃ¢â‚¬Â screens, we have created a retroviral library to express small hairpin RNAs (shRNAs), each of which can knock down the expression of an individual gene within a cell. To date, the library consists of 7,500 retroviruses that express shRNAs targeting 2,500 human genes. To enhance our ability to detect shRNA-induced proliferation or viability phenotypes in cancer cells, we have created a retroviral vector that can express shRNAs in an inducible fashion, based on the tetracycline repressor system. In addition, each shRNA construct is tagged with a unique 60 base pair 'bar code' to allow us to monitor the abundance of each shRNA vector in a cell population.
To perform an Achilles heel screen, the retroviral library is introduced into a cancer cell line and the cell population is divided in two, with one half receiving doxycycline to induce shRNA expression and the other half used as a control cell population. Any shRNA that knocks down the expression of a gene that is critical for proliferation or survival of the cancer cells will be selectively eliminated from the doxycycline-induced culture. At various times points, genomic DNA is harvested from the two populations and PCR is used to amplify the bar code sequences present in the genomic DNA. Amplified DNAs from the doxycycline-induced and control cultures are fluorescently labeled with different dyes and co-hybridized to a DNA microarray consisting of the bar code oligonucleotides. The microarray is scanned to reveal the relative abundance of each bar code in the two populations and hence the relative depletion or enrichment of cells expressing a given shRNA.
We have successfully used this approach to identify genes that are required for the proliferation or survival of different subgroups of diffuse large B cell lymphoma as well as multiple myeloma. In diffuse large B cell lymphoma, the Achilles heel screen revealed that the CARD11/MALT1/BCL10 pathway is responsible for the constitutive NF-kB signaling in the ABC DLBCL subgroup (39).
1.Alizadeh, A, et al: Cold Spring Harbor Symp Quant Biol 1999; 64:71-78.
2.Alizadeh, AA, et al: Nature 2000; 403:503-511.
3.Rosenwald, A, et al: N Engl J Med 2002; 346:1937-47.
4.Wright, G, et al: Proc Natl Acad Sci U S A 2003; 100:9991-6.
5.Wiestner, A, et al: Blood 2003; 101:4944-51.
6.Rosenwald, A, et al: J Exp Med 2001; 194:1639-1648.
7.Rosenwald, A, et al: Cancer Cell 2003; 3:185-97.
8.Dave, SS, et al: N Engl J Med 2003;
9.Hurt, EM, et al: Cancer Cell 2004; 5:191-9.
10.Ma, C, et al: Philos Trans R Soc Lond B Biol Sci 2001; 356:83-9.
11.Kovanen, PE, et al: J Biol Chem 2003; 278:5205-13.
12.Feske, S, et al: Nat Immunol 2001; 2:316-24.
13.Shaffer, AL, et al: Immunity 2001; 15:375-85.
14.Staudt, LM N Engl J Med 2003; 348:1777-85.
15.Rosenwald, A, et al: J Exp Med 2003; 198:851-62.
16.Dent, AL, et al: Science 1997; 276:589-92.
17.Ye, BH, et al: Nat Genet 1997; 16:161-70.
18.Lossos, IS, et al: Proc Natl Acad Sci U S A 2000; 97:10209-13.
19.Shapiro-Shelef, M, et al: Immunity 2003; 19:607-20.
20.Reimold, AM, et al: Nature 2001; 412:300-7.
21.Shaffer, AL, et al: Immunity 2000; 13:199-212.
22.Shaffer, AL, et al: Immunity 2002; 17:51-62.
23.Shaffer, AL, et al: Immunity 2004; 21:81-93.
24.Angelin-Duclos, C, et al: J Immunol 2000; 165:5462-71.
25.Falini, B, et al: Blood 2000; 95:2084-92.
26.Savage, KJ, et al: Blood 2003; 102:3871-9.
27.Copie-Bergman, C, et al: Mod Pathol 2002; 15:1172-80.
28.Addis, BJ, et al: Histopathology 1986; 10:379-90.
29.Huang, JZ, et al: Blood 2002; 99:2285-90.
30.Davis, RE, et al: J Exp Med 2001; 194:1861-1874.
31.Lam, LT, et al: Clin Cancer Res 2005 11:1-13.
32.Damle, RN, et al: Blood 1999; 94:1840-7.
33.Hamblin, TJ, et al: Blood 1999; 94:1848-54.
34.Hamblin, TJ, et al: Blood 2002; 99:1023-9.
35.Stilgenbauer, S, et al: Leukemia 2002; 16:993-1007.
36.Crespo, M, et al: N Engl J Med 2003; 348:1764-75.
37.Rassenti, LZ, et al: N Engl J Med 2004; 351:893-901.
38.Orchard, JA, et al: Lancet 2004; 363:105-11.
39.Ngo, VN et al: Nature 2006 441:106-110.
This page was last updated on 6/7/2013.