The tumor microenvironment shapes hallmarks of mature B-cell malignancies

Abstract

B-cell tumorigenesis results from a host of known and unknown genetic anomalies, including non-random translocations of genes that normally function as determinants of cell proliferation or cell survival to regions juxtaposed to active immunoglobulin heavy chain enhancer elements, chromosomal aneuploidy, somatic mutations that further affect oncogenic signaling and loss of heterozygosity of tumor-suppressor genes. However, it is critical to recognize that even in the setting of a genetic disease, the B-cell/plasma cell tumor microenvironment (TME) contributes significantly to malignant transformation and pathogenesis. Over a decade ago, we proposed the concept of cell adhesion-mediated drug resistance to delineate a form of TME-mediated drug resistance that protects hematopoietic tumor cells from the initial effect of diverse therapies. In the interim, it has been increasingly appreciated that TME also contributes to tumor initiation and progression through sustained growth/proliferation, self-renewal capacity, immune evasion, migration and invasion as well as resistance to cell death in a host of B-cell malignancies, including mantle cell lymphoma, diffuse large B-cell lymphoma, Waldenstroms macroglobulinemia, chronic lymphocytic leukemia and multiple myeloma. Within this review, we propose that TME and the tumor co-evolve as a consequence of bidirectional signaling networks. As such, TME represents an important target and should be considered integral to tumor progression and drug response.

INTRODUCTION

The advent of functional and structural genomics has greatly accelerated our understanding of oncogenic mechanisms in B-cell tumorigenesis.¹ However, evidence continues to demonstrate that dynamic interactions between the B cells and its tumor microenvironment (TME) profoundly influence the behavior of the other. Over a decade ago, we proposed the concept of cell adhesion-mediated drug resistance to delineate a form of TME-mediated drug resistance that protects B-cell malignancies and other hematopoietic tumor cells from the initial effect of diverse therapies.^2,3 Since then, numerous groups have affirmed these findings and demonstrated that the effects of TME on drug response are multifactorial—involving cytokines, chemokines, growth factors and malignant B-cell interactions with other constituents of TME, including, but not limited to, stromal cells.^4–6 Thus, the term Environmental-Mediated Drug Resistance (EMDR) has been used as a more encompassing term to describe the multiple aspects contributing to the influence of TME on drug response and resistance (in addition to cellular adhesion).⁷ As such, we and the others hypothesized that although the majority tumor cells succumb to therapy, a subset of malignant cells are afforded sanctuary within TME. Within these sanctuaries, malignant cells survive the stresses of therapy resulting in minimal residual disease. Over time, genetic instability inherent in cancer cells combined with the strong selective pressure of therapy (and TME) leads to successive changes that cause the development of more complex, diverse and permanent acquired-resistance phenotypes. These persistent tumor cells eventually cause disease recurrence and are much less likely to respond to subsequent therapy after acquired resistance develops (Figure 1).^5,7 It has been increasingly appreciated that in addition to drug resistance these effectors of TME also contribute to tumor initiation, growth and progression in B-cell malignancies.⁸ As such, this hypothesis can be amended to include not only therapeutic selective pressures but also those required for malignant transformation. Thus TME affords resident clonal B cells a selective advantage contributing to the expansion of a malignant clone. Within this sanctuary, under chronic selective pressures, additional transformative genetic alterations are acquired contributing to lymphomagenesis and myelomagenesis.^7,9,10 Therefore TME represents a critical target for therapeutic intervention and, in our opinion, should also be considered as important to tumor progression and drug response as the tumor itself.

Figure 1.

The development of EMDR, minimal residual disease (MRD), acquired resistance and disease progression. The mechanisms of drug resistance have been defined by genetically acquired changes in the expression or function of specific genes. The conventional explanation is that mutations form spontaneously and randomly before (or during) cancer has undergone chemotherapy, and these rare preexisting mutations may be selected for resistance under therapeutic stress. However, over the past 10 years a large body of evidence has emerged demonstrating that in addition to mechanisms of drug resistance intrinsic to the cancer cell, there exist dynamic, extrinsic mechanisms coordinated by the TME resulting in an EMDR. EMDR is a form of de novo drug resistance that protects tumor cells from the initial effects of diverse therapies by two primary mechanisms: soluble factor-mediated drug resistance and cell adhesion-mediated drug resistance. Specific niches within the lymphoma/myeloma microenvironment provide a sanctuary for subpopulations of tumor cells and provide a survival advantage through stromal cell–tumor cell interaction, allowing them to survive the insult of therapy, resulting in MRD. Over time, the residual tumor (lymphoma) cells are then destined to expand and evolve through (a.) acquisition of additional genetic abnormalities or (b.) selection of preexisting clones during therapy that cause the gradual development of more complex, diverse and permanent acquired-resistance phenotypes. The persistent tumor cells eventually cause disease relapse, much less likely to respond to subsequent therapy after acquired resistance develops and ultimately aggressive progression of lymphoma and multiple myeloma.

Mature B-cell malignancies have been proposed to originate from B cells at different stages of B-cell development, primarily derived from antigen-experienced germinal center or postgerminal center B cells.^9,11,12 Furthermore, the DNA repair/remodeling machinery that facilitates the great diversity of the antibody repitoires has also been shown to drive aberrant chromosomal translocations and other molecular anomalies.^1,11 In turn, a sequence of genetic alterations within a malignant clone facilitates an escape from the constraints governing proliferation and death. The initial requirements are at least partially fulfilled by the non-random translocation of a gene(s) that normally function (s) as determinants of cell proliferation or cell survival to regions juxtaposed to active immunoglobulin heavy chain/immunoglobulin light chain enhancer elements that occur secondary to errors of Ig switch recombination.^6,11,13 These, with additional genetic and epigenetic anomalies, lead to constitutive activation of oncogenes, deletion and inactivation of tumor-suppressor genes providing critical tumor hallmarks leading to malignant transformation.^6,11,13–15

However, even in the context of molecular-derived transformative lesions, malignant B cells do not act in isolation. These cells subsist in a rich microenvironment composed of resident fibroblast-like cells, endothelial cells, immune cells and extracellular matrix. TME has the ability to support lymphoma development and progression by fostering key hallmarks of malignancy: resistance to cell death, cell homing and invasion, sustained proliferation, immunosuppression, self-renewal and stemness as well as angiogenesis (Figure 2). It is also important to recognize that malignant B cells are not only simply receiving signals but also have an impact on the biology of stroma and other cellular (and non-cellular) elements of TME, fashioning a bidirectional relationship between the tumor and TME. Within hematological malignancies, as in models of solid tumorigenesis, TME has been shown to evolve with malignancy and demonstrate sustained gene expression changes that support critical aspects of tumor growth.^16–18 Thus the bidirectional relationship between malignant B cells and their TME is an integral part of tumor initiation, progression and responsiveness to anticancer therapies.^4,5,7

Figure 2.

TME shapes the hallmarks of B-cell malignancies. Malignant B cells (lymphoma and myeloma) recruit and activate TME stromal cells, including immune cells, angiogenic vascular cells and fibroblastic cells, and, in turn, TME stromal cells profoundly influence the behavior of B-cell malignancies. The collaborative interaction between neoplastic B cells and their supporting stromal cells enable and sustain the hallmarks of cancer, including resistance to cell death (anti-apoptosis and drug resistance), sustaining cell proliferation, angiogenesis, immune suppression, stemness and self-renewal and cell homing and invasion, thus promoting tumor progression.

TME CONFERS THERAPY RESISTANCE, MINIMAL RESIDUAL DISEASE AND SURVIVAL

The role of TME in drug response and survival has been relatively well characterized. Even with significant improvements in therapeutics and supportive care available for the management of non-Hodgkin's lymphoma and multiple myeloma (MM), these malignancies represent largely incurable diseases. Mortality is frequently the consequence of refractory disease, a phenomenon that had long been attributed to the acquisition of heritable mechanisms of multidrug resistance (MDR). However, it is clear that treatment failure was the consequence of more than a stochastic drug exposure phenomenon and that other microenvironmental factors also contributed to MDR.¹⁹ The unique tropism of mature B-cell tumors to lymph nodes, the bone marrow (BM) or other secondary lymphoid organs suggested that the niches affords a survival advantage. As such, local microenvironmental determinants would afford resident cells a therapy resistance resulting in small foci of residual disease that subsequently develop complex genetic or epigenetic means of acquired resistance in response to the selective pressure of therapy. TME consists of physical extracellular matrix components and supportive fibroblast-like stromal cells that are critical for the dynamic interplay between these chemokine/cytokine networks, cellular adhesion molecules and the cellular components of TME.

These extrinsic effectors of TME translate to intrinsic biochemical signaling cascades modulating drug resistance.^4,9,12 Over 10 years ago, we were among the first to observe that malignant B-cell survival and drug response was subject to regulation by TME.^2,20 Instead of stable acquired mechanisms of MDR, we and the others demonstrated that there also exist dynamic, de novo mechanisms of drug resistance coordinated by both the soluble and physical effectors of TME.^4,7,21 In 1999, Damiano et al.² demonstrated that interactions between the extracellular matrix component fibronectin and myeloma cells via β1 integrin-containing heterodimers conferred cell adhesion-mediated drug resistance, a form of de novo drug resistance that protects tumor cells from the effects of diverse therapies. Interleukin (IL)-6, a putative myeloma survival cytokine, has also been shown to confer a survival advantage to myeloma cells via JAK/STAT3 (Janus kinase/signal transducer and activator of transcription) induced regulation of the Bcl-2 family member Bcl-XL.²⁰ Subsequently, adhesion of malignant hematopoietic cells to extracellular matrix and stromal cells has been demonstrated to confer a survival advantage in the face of therapeutics and host antitumor cell death pathways via a multitude of processes.^4,10,22,23 Lwin et al.²⁴ demonstrated that direct contact with the human BM stromal cells (BMSCs) afforded resistance to spontaneous and drug-induced apoptosis in various lymphomas. In MM, stroma-induced activation of nuclear factor (NF)-κB, PIM1, Notch and MYC and other critical survival and drug response pathways have been observed and contribute to EMDR.⁵ These data demonstrated that interactions of malignant B cells with TME facilitates a drug-resistant phenotype.

In the context of these tumor-stroma induced events, it is apparent that EMDR arises from adaptive, reciprocal signaling circuits between tumor cells and the stromal elements of TME via positive feedback/feedforward loops.⁵ These signaling dialogues are the result of alterations in intracellular signaling within the tumor cells and stromal cells induced by their interactions within TME.^4,5,25 TME-interacting malignant B cells undergo wholescale changes in the transcriptome with alteration expression of cellular processess controlling interactions with TME as well as the expression of cytokines and growth factors that control critical pathways for survival and drug resistance.^10,17,26 In B-cell lymphomas, this has been indirectly demonstrated with distinct TME-induced gene expression profiles when comparing lymph node-resident tumor cells with circulating tumor cells.¹⁷ In chronic lymphocytic leukemia, gene expression profiles from peripheral blood versus lymph node-resident tumor cells demonstrated a significant increase in the expression of B-cell receptor (BCR) pathway genes in lymph node microenvironment relative to circulating cells,²⁷ indicating that TME acts by modulating sustained BCR signaling. Further evidence supporting BCR signaling pathway in TME-induced survival signals comes from the in vitro effects of BCR kinase inhibitors SYK inhibition with R406,²⁸ phosphatidylinositol 3′-kinase δ (PI3Kδ) inhibitor Idelalisib/GS1101^29,30 and Ibrutinib/PCI32765-mediated inhibition of BTK in non-Hodgkin's lymphoma and MM.^31–33 Collectively, these data indicate that BCR and likely other apical proteins/kinases are critical to orchestrating bidirectional collaborative communication between tumor cells and BMSCs (or other cellular elements of TME) that facilitate EMDR and ultimately result in residual disease in the setting of therapeutic stress and disease recurrence.

TME SUSTAINS MALIGNANT B-CELL GROWTH AND PROLIFERATION

As discussed, chemokines and cytokines function to recruit tumor cells to TME where positive cytokine- and cell adhesion-mediated feedback loops are established between the malignant B cells and stromal cells.⁴ These intercellular positive feedback loops not only support the survival and drug resistance but also facilitate the growth and proliferation of malignant B cells. Paracrine mitogenic signals supplied by stromal cells in lymph node and BM are implicated in different lymphoma types ranging from the initiation of aberrant proliferation to the cell-autonomous tumor progression. Upon adhesion to stromal cells, malignant cells as well as stromal cells produce and secrete a host of soluble factors associated with malignant B-cell growth.^4,24,34 Further, TME, specifically stromal cells, have also been shown to regulate sustained c-Myc activation facilitating lymphoma cell growth and proliferation.³⁵

B-cell activating factor (BAFF) as well as its receptors, particularly BAFF-R, are critical factors for the growth and survival of both normal and malignant B cells.³⁶ Our group also demonstrated that B-cell lymphoma adhesion-induced BAFF expression by BMSCs promoted lymphoma cells growth in B-cell lymphomas via activation of NF-κB signaling events.^37,38 Overall, B-cell and stromal cell interactions generate TME niches for lymphomas with high BAFF availability to promote lymphoma growth. In addition to BAFF, IL-6 remains the prototypical growth and survival factor in myeloma as well as B-cell lymphomas.³⁹ IL-6 binds its receptors on lymphoma cells facilitating the phosphorylation of gp130 and initiation of Ras/Raf–mitogen-activated protein kinase, kinase–extracellular signal-related kinase (ERK)1/2, PI3K and JAK/STAT3 signaling.⁴⁰ Each of these signal transduction pathways has been implicated in IL-6-mediated proliferation. Autocrine and/or paracrine IL-6 is involved in mantle cell lymphoma growth.⁴¹ Importantly, paracrine IL-6 in lymphomagenesis is context-specific. In transformed but not immature B cells, B-cell lymphomas use paracrine IL-6 signaling as a growth-promoting signal, implicating complex regulation of lymphoma development by the microenvironment and transformative genetic events.⁴⁰ It is critical to note that other cytokines and/or adhesion molecules also contribute to the cross-talk between stroma and lymphoma/myeloma cells: IL-7, vascular endothelial growth fractor (VEGF), C-X-C motif chemokine receptor 4 (CXCR4), RANTES (regulated and normal T cell expressed and secreted), macrophage-inflammatory protein-1, JAGGED-1, insulin-like growth factor-1 and IL-21 to list a few.^42–46 TME can also support malignant B-cell proliferation via additional ligand-receptor siganling conduits. Hedgehog ligands produced by stromal cells have been demonstrated to promote proliferation of B-cell lymphomas and myeloma in the context of TME both in vitro and in vivo.^47,48 Hedgehog ligands produced by follicular dendritic cells promote growth of germinal center B cells.⁴⁷ In MM, BMSCs promoted CD138-isolated patient cell expansion in a hedgehog-dependent fashion.⁴⁹ Together, these data demonstrate that dynamic interections between tumor cells and TME can promote malignant B-cell proliferation. It may also be expected that within TME these effectors do not act in isolation; as such, the combination of secreted factors and cell adhesion mechanisms act in concert to dictate malignant B-cell proliferation.²⁵

TME has been shown to promote malignant B-cell proliferation via the regulation of the oncogene c-Myc. Adhesion of mantle cell lymphoma and other non-Hodgkin's lymphoma cells to lymph node and BM stromal cells has been shown to promote lymphoma clonogenic growth in vitro and in vivo.³⁵ This stroma-enhanced clonogenicity and tumor formation has also been observed in MM.⁵⁰ This clonogenic advantage is, at least in part, modulated by intracellular bidirectional circuitry induced by TME. Stroma-induced c-Myc activation was shown to contribute to lymphoma growth and progression via a stroma-triggered c-Myc–miR-548 feedforward loop leading to sustained c-Myc activation.³⁵ Critically, disruption of this circuit suppressed clonogenicity and lymphoma growth ex vivo and in vivo. We also observed induction of enhancer of zeste homolog 2 (EZH2) by stroma in lymphoma cells consistent with a MYC–miRNA–EZH2 positive feedback loop in aggressively transformed lymphoma.^51,52 Similarly, targeting c-Myc produces a potent antiproliferative effect in in vivo models of MM.⁵³

These data indicate that the mechanisms by which stroma influence tumor proliferation are complex, potentially representing convergence of stroma-induced and genetically determined cellular functions. Consistent with this, stromal interactions have been shown to enhance the transcriptional output of these signaling cascades, even when they are constitutively activated by mutational events.⁵ For instance, recurrent lesions for genetically driven alterations in B-cell malignancies such as mutational activation of RAS, NF-κB, PI3K-AKT and STAT3 pathways as well as MYC amplifications can be further activated by stromal cells.^5,7 Malignant B cells are therefore poised for an increased capacity for proliferation via the production of ligands from tumor and/or stromal cells resulting in autocrine and paracrine networks. During the multistep transformation and malignant progression, both malignant B cells and stromal cells adapt to maintain the constitutive activation of signaling pathways.^35,54,55

TME IS CENTRAL TO THE ADHESION, MIGRATION AND INVASION OF MALIGNANT B CELLS

Localization to the appropriate niche is critical for B-cell development and selection. Migration and adhesion of malignant and normal B cells to the BM and secondary lymphoid tissues are, in part, modulated by the homeostatic chemokine C-X-C motif chemokine ligand 12 (CXCL12)/stromal cell-derived factor 1 produced by stromal cells of a given niche and its cognate receptor CXCR4.^56,57 Additional chemokines are also critical to this process, including CXCR13 and IL-8.¹¹ As an example, we will focus on CXCL12 and CXCR4. CXCL12 produced by TME modulates homing of B cells to TME by inducing both the expression of integrin subunits and by inducing inside-out activation of integrins.⁴ Of note, the converse has also been observed; integrins also modulate both increased CXCR4 expression and function.⁴ To this end, the relationship between CXCR4 and the integrin heterodimer VLA4 (very late antigen 4) is complex, constituting bidirectional outside-in and inside-out signaling system modulating malignant B-cell motility.

As with drug response and proliferation, malignant B-cell motility arises from adaptive, reciprocal signaling dialogues between tumor cells and TME cells via positive feedback/feedforward loops.^4,35,56 Accumulating evidence indicates that BCR is integral in this bidirectional signaling between chemokine receptors and integrins contributing to migration and cell adhesion in B-cell malignancies, such as chronic lymphocytic leukemia, mantle cell lymphoma and MM.^58,59 BCR governs the interplay between intrinsic and extrinsic CXCR4/CXCL12-mediated migration and VLA4-mediated adhesion.^56,60 Specific targeting of SYK with the inhibitor R406 attenuated CXCL12/stromal cell-derived factor 1-mediated malignant B-cell homing signals.⁶¹ BCR has also been shown to induce VLA4 high-affinity binding via sequential activation of the components of BCR effectors.⁵⁸ Direct clinical translation of the critical role of BCR in malignant B-cell motility has been demonstrated in early-stage clinical trials with SYK inhibitor FosD,⁶² BTK inhibitor Ibrutinib⁶³ and PI3Kδ inhibitor GS-1101/Idelalisib⁶⁴ with clinical responses characterized by an early redistribution of tissue-resident chronic lymphocytic leukemia cells into the blood, resulting in rapid resolution of lymphadenopathy and organomegaly, along with a transient surge in lymphocytosis during the first weeks of therapy consistent with the attenuated B-cell migration to and adhesion within TME.⁶⁵ Together, these data underscore the clinical opportunities defined by targeting TME to disrupt tumor–stromal interactions releasing cells from the niche and sensitizing cells to therapy.

TME PROMOTES ANGIOGENESIS

A number of studies have implicated increased microvessel density with disease state and worse outcomes, suggesting that increased niche angiogenesis is an important requirement for disease progression.^66–69 Compelling evidence stems from the examination of BM microvessel density along the myeloma continuum showing a greater density in patients with active disease relative to patients with monoclonal gammopathy of undetermined significance.⁶⁶ Du et al.⁶⁶ demonstrated that microvessel density increased from normal BM samples to monoclonal gammopathy of undetermined significance to active disease. Tumor vascularization has also been demonstrated to be higher in lymphoma tissue than in reactive lymph nodes.⁶⁷ These results indicate a critical role of the vascular component of TME in the progression of B-cell malignancies.

As we have discussed, malignant B cells and MM cells not only benefit from the normal effectors within TME but also contribute to the environment. Malignant B cells, endothelial cells (and other stromal cells) and non-cellular components of TME contribute to neovascularization within lymph nodes and the BM.^70,71 The most prominent effector of angiogenesis is the VEGF. In myeloma, VEGF has been shown to induce growth, survival as well as migration of cells in an autocrine manner via VEGFR-1.⁷² Moreover, paracrine activation of BM endothelial cells (ECs) VEGFR-2 modulates angiogenesis.⁷⁰ Studies have also identified BM ECs that produce growth and invasive factors for myeloma cells, including VEGF, fibroblast growth factor-2, matrix metalloproteinase-2, IL-8, stromal cell-derived factor 1-α and monocyte chemotactic protein-1 as well as matrix metalloproteinase-9.^70,73 In turn, VEGF produced by myeloma cells can similarly prolong survival of ECs.⁷² Myeloma cells and BMSCs can also promote EC function by attenuating the secretion of antiangiogenic factors.⁷⁰ Similar scenarios are also seen in TME of other B-cell malignancies. In mantle cell lymphoma, both lymphoma cell-derived autocrine signaling and paracrine signaling to stromal cells of blood and lymphatic vessels have been shown to mediate lymphoma cell survival and lymph node EC activities, respectively.^71,74 Together, these results demonstrate that in B-cell lymphomas and MM neoangiogenesis is regulated by an reciprocal network of interactions based on the production of growth factors deriving from malignant cells and from accessory cells of TME.

TME FACILITATES IMMUNOSUPPRESSIVE NICHE

Similar to many other cancers, B-cell malignant progression is accompanied by profound immune suppression that interferes with an effective antitumor response and tumor elimination.^14,75–77 A growing body of evidence indicates that interaction among lymphoma cells, stromal cells and neighboring immune cells regulate antitumor immunity. Using gene expression profiles on whole lymphoma tissues, Lenz et al.⁷⁶ identified ‘stromal’ signatures predictive of good and bad survival rates, respectively, in diffuse large B-cell lymphomas that were differentiated by an immune component to the stromal signature. Gene expression profiles of follicular lymphomas further supports that the immune microenvironment has an important role in the outcome as genes expressed by infiltrating T cells and macrophages are among the most important predictors of survival.¹⁸ Moreover, upon transformation of follicular lymphoma to aggressive diffuse large B-cell lymphoma, nearly two-thirds of the most discriminative genes were related cellular immune response and inflammatory processes.^78,79 These data indicate that interactions between malignant B cells, immune cells and stromal cells engender an ‘immune-suppressive’ TME via the regulation of immune cell differentiation, recruitment of regulatory T cells, suppressive tumor-associated monocytes/marcophages and myeloid-derived suppressor cells.

Regulatory T cells are a specific subset of T cells that are essential for maintaining immune tolerance that have been observed in increased numbers in TME of B-cell malignancies.^80,81 Regulatory T cells inhibit T-cell responses against both foreign antigens and self-antigens such as tumor antigens by attenuating effector immune responses mediated by CD4 and CD8 T cells through the production of immunosuppressive cytokines.⁸² These data indicate that regulatory T cells can contribute to immune escape of the neoplastic clone as well as influencing B-cell differentiation and maturation process. Macrophages are part of the immune infiltrate found in a variety of lymphomas and myeloma. Increased numbers and/or density of intratumoral macrophages is associated with both progression and prognosis B-cell malignancies.⁸³ Lymphoma cells and stromal cells produce chemokines and growth factors that produce tumor-associated monocytes/marcophages. These tumor-associated monocytes/marcophages induce immune suppression through several mechanisms involving regulation of antitumoural immunity through interactions with the malignant clone as well as other TME cellular components.^84,85 Myeloid-derived suppressor cells represent another immunosuppressive cell population related to macrophages. These cells have the capacity to suppress both the cytotoxic activities of natural killer and natural killer T cells as well as the adaptive immune response mediated by CD4⁺ and CD8⁺ T cells.⁸⁶ Recent studies demonstrated that immunosuppressive myeloid-derived suppressor cells accumulate in the BM of patients with MM as well as in the BM of MM-bearing immunocompetent mice promoting tumor progression.^87,88

TME PROMOTES ‘STEMNESS’ LINKED TO SUSTAINED SELF-RENEWAL CAPACITY

Emerging evidence now has shown that TME has a critical role in supporting what are termed cancer stem cells (CSC). CSCs, also known as ‘tumor-initiating cells,’ are the cancer cell population that carries functional properties of stem cells: self-renewal, the capability to differentiate into multiple lineages in the tumor, and the potential to proliferate extensively to expand the malignant cell population.⁸⁹ Although the exact nature of the CSC in B-cell malignancies remains controversial, subpopulations of cells have been identfied with self-renewal capacity.^50,90,91 Hematological CSCs may arise from self-renewing stem cells by the acquisition of mutations or from more differentiated cells that gain CSC-like properties.^17,92,93 Similar to normal stem cells that are maintained within specialized niches, CSCs also rely on TME. These niches are important for the control of CSC survival and proliferation.⁹² Thus the microenvironment likely has an important role in promoting and conferring stemness, CSC self-renewal and primary tumor progression.⁹⁴ Unlike most stem cells that are restricted to the least differentiated cell types, both germinal center and memory B cells reside in germinal centers, interact with stromal cells, differentiate into mature cells and possess self-renewal potential that allows rapid lymphocyte division and antigen-specific immunity. The renewal and differentiation capacity of the germinal center lymphocytes upon antigen stimulation has suggested that secondary lymphoid organs are ideal niches to confer lymphocyte stem cell-like features (stemness) through stroma-mediated genetic and epigenetic means. Indeed, most B-cell non-Hodgkin's lymphomas are truly derived from either committed lymphoid progenitor/precursor cells or from more mature B-lymphocytes in germinal centers.¹¹ Given the close contact of lymphoma cells and stromal cells in the lymph node and BM microenvironment, malignant reprogramming may be achieved through stroma-mediated regulation of genes controlling a stemness pathway.³⁵ Our recent studies on TME have revealed that stromal cells can modulate stemness by promoting B-cell differentiation and enhancing lymphoma cell clonogenicity.^35,51 Further, recent experiments have shown that stromal cells activate c-Myc, EZH2, Oct-4, Nanong and Sox2, key elements in acquisition and maintenance of stem cell properties.^95,96 Collectively, these results suggest that lymphoma cells interacting with stroma display characteristics of CSC. It is tempting to speculate that lymphoma microenvironment or lymphoma/stroma interactions are the driving force for the initiation and maintenance of the replicative immortality of CSC in malignant B-cell disorders.

Another critical aspect of replicative immortality is the control of telomere length. Evidence of the necessity of telomerase activity in B-cell malignancies comes from the observation of increased telomerase activity with progression from monoclonal gammopathy of undetermined significance to active MM.⁹⁷ Increased telomerase activity observed in myeloma is regulated by TME effectors IL-6 and insulin-like growth factor-1 in myeloma cell lines and myeloma patient specimens.⁹⁸ Furthermore, both IL-6 and insulin-like growth factor-1 overcame dexamethasone-induced attenuation of telomerase activity. Together, these results indicated that TME has profound effects on stemness and self-renewal in mature B-cell neoplasms.

CONCLUSIONS AND PERSPECTIVES

Malignant B-cell tumorigenesis is a complex collection of distinct genetic diseases characterized by common hallmarks. Here we describe a conceptual framework by which TME contributes to these hallmarks in B-cell malignancies. In the context of malignant B cells, TME shapes a common set of malignant properties (hallmarks), including sustained proliferation, self-renewal capacity and stemness, homing and invasion, immune evasion and angiogenesis as well as resistance to cell death.⁹⁹ For lymphomas and myeloma to evolve and progress, malignant B cells manipulate their TME to fulfill these hallmarks through reciprocal interactions with surrounding cellular and non-cellular elements. As shown in Figure 3, the malignant progression is a complex process of interconnected levels of signaling events. First, the extracellular circuits of malignant B cells with tumor-associated monocytes/marcophages, immune cells, stromal cells and EC (Figure 3a). The interactions between the malignant B cells and TME cells are dynamic, bidirectional paracrine (or autocrine) leading to recruitment and activation (or suppression) of partner cellular populations. Both neoplastic cells and the cellular elements of TME evolve progressively through heterotypic signaling between tumor and stroma cells that modulate the release of cytokines, chemokines and growth factors and recruitment of cellular TME effectors facilitating malignant progression. Second, the intercellular circuits between lymphoma/myeloma cells and TME that establish positive cytokine and cell-adhesion-mediated feedforward loops that promote homing of tumor cells as well as cell-adhesion and soluble factor-induced signaling events (Figure 3b). BCR can function as a central modulator of the interplay between the inside-out and outside-in CXCR4/CXCL12-mediated migration and VLA4-mediated adhesion leading to enhanced receptor signaling and cell adhesion.⁴ Third, intracellular circuits—the intercellular interaction-triggered intracellular signaling circuits that drive cancer hallmarks (Figure 3c). Stroma through combination of secreted factors and cell adhesion mechanisms further enhance numerous inter-related signaling pathways, including BCR–PI3K–AKT, NF-κB, RAS–RAF–MAPK, JAK–STAT–HIF1α, c-MYC, ERK and Notch pathways. Moreover, in order to sustain hallmark features, tumor cells’ positive feedforward signaling loops such as BCR-NF-κB, IL-6-STAT3 and miRNA-Myc circuitry lead to increased transcriptional and posttranscriptional events.^35,54,55 The collaboration and concerted action between these three levels of reciprocal networks drive the hallmarks of B-cell malignancies.

Figure 3.

Reciprocal signaling networks modulated by TME drive malignant B-cell tumor progression and response to therapy. As malignant B cells accumulate in the niche, positive cytokine and cell adhesion-mediated feedforward loops/circuits are established between the tumor and TME that facilitate the progression of the malignant clone. (a) Extracellular circuits: Tumor (lymphoma/myeloma) cells recruit and orchestrate the activity of accessory cells, including vascular endothelial cells, fibroblast-like stromal cells, immune and inflammatory cells, via autocrine and paracrine signaling. These extracellular signals trigger amplification loops fostering increased production of cytokines/chemokines and recruitment of additional cellular effectors. This, in turn, supports a survival and proliferative advantage in the face of endogenous and therapeutic stresses. (b) Intercellular circuits: Interactions between malignant B cells and TME facilitate critical alterations in signaling in both tumor cells and stromal cells. Central to these events are ‘inside-in’ and ‘outside-in’ signaling loops involving chemokines, cytokines, integrins and BCR signaling activation. Stromal cells induce sustained BCR signaling leading to chemokine and integrin activation in malignant B cells. In turn, tumor cells directly and indirectly elicit alterations in stromal cells via cell–cell receptor–ligand contact and soluble factors. Together these networks result in enhanced receptor signaling and cellular adhesion. (c) Intracellular circuits: Tumor/TME signaling stimulates intracellular feedforward signaling loops that sustain signals emitted by ligand-activated receptors and integrins. Examples of signaling loops include BCR-NF-κB, IL-6-STAT3 and miRNA-Myc circuitry leading to sustained transcriptional and posttranscriptional events. Collectively, these inter-related and intra-related signaling cascades dictate the nature of TME vis-à-vis the hallmarks of B-cell malignancies. As the tumor progresses, the tumor and TME co-evolves into an activated state through continuous bidirectional communication, thus creating dynamic signaling circuitries that promote tumor/TME interactions.

Together, these results demonstrated that it is critical that we identify the appropriate factors within the complex network of the tumor cell microenvironment to target in the context of therapy. By elucidating the role of the TME and the crucial steps of TME–lymphoma interaction in the pathogenesis of B-cell malignancy, recent studies have provided the framework for identifying and validating novel therapies that target both tumor cells and cells in their surrounding microenvironment. Targeting the TME can be implemented through several strategies to discrupt the circuits outlined above: (1) attacking cellular component of TME niche and inhibiting abnormally activated pathways within the cells of niche; (2) affecting homing and adhesion through interference with chemokines and adhesion molecules to target the communication between malignant cells and TME; and (3) targeting TME-induced intracellular signaling pathways to block the prosurvival and proliferative signals in malignant B cells as well as stromal cells.

All of the cellular components of the TME are potential therapeutic targets, because dysfunctional cytokine and chemokine network that promote mature B-cell malignancy survival and progression are established in concert with these effector cells. Controlling angiogenesis is clearly a strategy for the treatment of B-cell malignancies.⁶⁹ This can be achieved either by directly targeting VEGF with monoclonal antibody, bevacizumab or targeting the upstream or downstream molecules of VEGF signaling. It is possible that diffuse large B-cell lymphoma patients with an increased tumor blood-vessel density/stromal-2 signature⁷⁶ and other aggressive B-cell malignancies would benefit from targeting inhibitors of angiogenesis. Disruption of migratory and adhesion signals represents the strategy of attenuating malignant B-cell homing to a niche and/or sensitizing tumor cells to chemotherapy or targeted therapies. Small-molecule pharmacological inhibitors of CXCR4 have been shown to be efficacious in preclinical models of chronic lymphocytic leukemia, B-cell lymphomas and MM,^100,101 presumably through interference of TME homing and recruitment of tumor cells out of their protective microenvironmental niches. Cell-penetrating lipidated peptides targeting CXCR4 intracellular domains also significantly increase the antitumoral activity of rituximab in vitro and in vivo.¹⁰² Another approach is to implement neutralizing VLA4 antibody and small-molecule inhibitors of VLA4/vascular cell adhesion molecule-1 to disrupt the TME interactions and cause impressive mobilization out of TME protective niche. Anti-VLA4 antibody natalizumab has been evaluated as an alternative stromal adhesion-disruptive drug.¹⁰³ Targeting TME using d-amino acid peptides to prevent cell adhesion has proven to induce myeloma cell death and increase sensitivity to known active agents, including melphalan and bortezomib.¹⁰⁴ Additionally, lenalidomide, an immunomodulatory drug clinically active in several mature B-cell malignancies and under evaluation in follicular lymphomas, could disrupt B cell–stroma interaction through a decrease in both CXCL12 production by stromal cells and RhoH expression in malignant B cells.^105,106 Further, the end result of TME–tumor interaction is activation of various inter-related signaling cascades in malignant B cells and subsequently dictate growth, response to therapy and drive TME–tumor interaction and hallmarks. Therefore, these signaling pathways serve as potential therapeutic targets. Of these signaling pathways, the BCR and its downstream effectors are emerging as central modulators of B-cell homing, survival and drug resistance within the context of B-cell microenvironment.^{6,28,107–109} As such, the BCR may serve as a central hub at the crossroads between extrinsic and intrinsic events. More specifically, in addition to activation of PI3K–AKT and NF-κB pathways, BCR orchestrates the interplay between outside-in and inside-out by CXCR4, integrins and other key effectors of the TME, thereby having a critical role in malignant B-cell homing, survival and EMDR.⁴ Therefore targeting the BCR pathway molecules will attenuate growth and survival signals emanating from both B-cell intrinsic abnormalities and from the TME, serving as a novel ‘double-hit’ strategy: targeting both BCR-regulated survival signaling and BCR-regulated lymphoma–TME interactions releasing lymphoma cells from their microenvironment, resulting in sensitization and enhanced cytotoxic killing. This hypothesis has been substantiated by recent clinical trials of BTK inhibitors in B-cell lymphoma patients with encouraging results.^110–112 In support of this, an example is the recent success of using the PI3Kδ-specific inhibitor CAL-101/GS1011/Idalelisib in the treatment of B-cell lymphoid malignancies.^113–117 These promising effects are likely due to the concomitant action of the drug in both tumor and TME. For example, in malignant B cells, PI3Kδ blockade impairs pro-survival signaling and induces tumor cell death. On the other hand, inhibition of PI3Kδ in monocyte-derived cells of the lymph node stroma blocks the secretion of pro-survival chemokines sensed by tumor cells and strengthens the pro-apoptotic effect of the drug.¹¹³ These demonstrate the principles behind drug design in the context of the microenvironment. The BCR inhibitors discussed above target both extracellular and intracellular determinants of the BM niche. The hope is that we can identify the appropriate factors within the complex network of the tumor cell microenvironment to target. To this end, we may divine therapies to overcome the coordinated effort between lymphoma/MM cells and the microenvironment. In so doing, we may be also able interrupt the sequence of events (de novo and acquired) facilitating minimal residual disease culminating in therapy resistance.

However, it is not likely that singly targeted therapies will demonstrate lasting control of disease with the significant heterogeneity within the BM and/or lymph node niche. As such, the development of rationally designed combination therapies will be critical to further clinical success. To this end, targeting of multiple pathways either simultaneously or in sequence may be the measure by which to overcome the sanctuary of TME milieu. Stroma-induced BCR/BTK/PI3K and c-Myc pathways cooperatively dictate interaction (adhesion) between lymphoma and stroma and promote lymphoma drug resistance and proliferation. Disruption of both pathways establishes a novel ‘triple-hit’ or ‘multi-hit’ strategy: targeting the pathway (BCR/PI3K) related to survival, targeting the pathway (c-Myc) related to cell proliferation and lymphoma angiogenesis, and targeting lymphoma–stroma adhesion to release lymphoma cells from their microenvironment to sensitize and enhance cytotoxic killing. Additionally, tumor cells respond, adapt and co-evolve before, during and after treatment and results in consequence of therapy resistance.^118,119 It will be important to consider the merit of combinatorially targeting downstream (BCR) signaling pathways to hold advantage of direct modulation of the cell survival and proliferation machinery with targeting upstream or parallel pathways to circumvent compensatory survival pathway. In contrast, tumor-associated stromal cells presumably may also ‘adapt and evolve’ and, therefore, constitute an alternative therapeutic target. A recent in vitro and in vivo study demonstrated that upregulated stromal PKC-βII and NF-κB are prerequisites to support the survival of malignant B cells in a variety of B-cell malignancies and constitute a potential target.¹²⁰ Going forward, the challenge will be to identify the optimal combinations of drugs to appropriately modulate both tumor cells and TME in the most selective manner possible. Furthermore, the continued success of BCR and other inhibitor therapy in lymphoma/MM will require the rational design of combination of targeted agents and a deep understanding of the nature between the malignant B cells and their TME.