Welcome international healthcare professionals

This site is no longer supported and will not be updated with new content. You are welcome to browse and download all content already included in the site. Please note you will have to register your email address to access the site.

You are here

The microenvironment in follicular lymphoma

Best Practice & Research Clinical Haematology, 2, 24, pages 135 - 146

It has become increasingly clear that proliferation and survival in FL is not only driven by genetic changes, but also and possibly even predominantly by the close interaction with the immune microenvironment and stromal cells. Based on in vitro studies and experimental models and supported by immunohistochemical studies in biopsy specimens of FL patients, classes of CD4+ T-cell populations including follicular helper T cells and regulatory T cells are now identified as major players to regulate the delicate balance of effector populations into a supportive microenvironment. These insights may thoroughly change the therapeutic approaches in FL and translate into programs that combine direct cytotoxic and indirect immunomodulatory aspects.

Keywords: Follicular lymphoma, Microenvironment, T-helper cell, TFH cell, Follicular dendritic cell.

Tumor microenvironment may play a role in the biology of follicular lymphoma; evidence from clinico-pathological data

Although the presence of non-tumorous cells have long been appreciated in human cancers, their important role in the biology and clinical behavior of malignancies has been underestimated in favor of the dominant impact of the genetic make-up of tumor cells. The same holds true for follicular lymphoma. Since the routine introduction of immunohistochemistry in the 1990s, pathologists have realized from daily diagnostic work that non-malignant T cells account for a large part of the tumor volume with infiltration of up to 50% of the cell mass in FL. Also the presence of follicular dendritic cells, macrophages, microvasculature and mast cells was noted. Clinico-pathological studies to understand the variations in clinical behavior and to identify the prognostic factors have most been focused on genetic aspects, however. Various genetic alterations in addition to the basic t(14; 18) translocation have been shown to occur at relatively high frequencies and some of these correlate to survival as reviewed in Chapter 1 of this issue.

Evidence from gene-expression profiling

With the generation of the first gene-expression studies, it was expected that these would add further insight in the oncogenic signaling pathways that may be involved in the molecular deregulation of tumor cells. The very first published gene-expression study by Husson and coworkers was performed on enriched tumor cells and indeed showed significant molecular differences between FL tumor cells and normal germinal center cells with upregulation of cell cycle regulating genes (CDK10, p120, p21, p16INK4A), apoptosis regulating genes (BCL2, ERCC3, BIK, CASP4) and various transcription factors (PAX5, ID2, JUN) in line with these expectations [1] . A large proportion of the 588 differentially expressed genes were related to cell surface receptors and proteins involved in cell–cell communication and especially immune cell communication. Such factors especially included cytokines and chemokines and their receptors, involved in interactions between B- and T-cells, macrophages and stroma components (TNF, IL-4Rα, IL-2Rγ, TNFRSF5, CSF1). In subsequent studies on mRNA isolated from full biopsies that included both the non-malignant cells and the tumor cells, the tumor cell derived signatures proved to be very hard to appreciate. In both published studies, the signature was overruled by dominant information derived from non-tumor cells that showed correlation to prognosis in terms of overall survival in the study by the LLMPP group and to risk for transformation by the Dutch group respectively [2], [3], and [4]. The classes of genes are largely overlapping and both show involvement of various T-cell classes and activation and of accessory cells, including macrophages. Based on these studies, a preliminary hypothesis may be brought forward that implicates various T-cell classes as prognostically favorable, others as unfavorable and dense infiltrates or activation of macrophages as prognostically unfavorable. Due to the involvement of genes in more than one single cell class and complicated by the complexity of the signaling and activation networks, gene-expression profiling may not be the method of choice to identify the relevant cell populations in more detail, however.

Immunohistochemical identification of relevant cell populations

In contrast to gene-expression profiling, immunohistochemical analysis on tissue sections provides information on specific cell populations as well as their spatial distribution in the tissue. Therefore, various groups have tried to apply the insights from gene-expression studies by analyzing T-cell populations, including helper T cells (CD4), cytotoxic T cells (CD8, granzymeB), T-helper subsets such as regulatory T cells (FOXP3) and follicular T-helper cells (PD-1), macrophages (CD68) and follicular dendritic cells (CD21, CD23, CD35) as well as mast cells [3], [5], [6], [7], [8], [9], [10], [11], [12], and [13]. Based on various published series, it may be suggested that high numbers of macrophages may confer an unfavorable clinical course, whereas a high content of CD8+ and CD4+ lymphocytes may indicate prolonged survival. The results of various studies are very contradictory, however, with specific cell populations significantly correlating with poor prognosis in some series, but with good prognosis or without any significant impact in others [14] . Several explanations may be given related to study end points, patient selection, treatment and technical scoring variations.

In most studies, overall survival (OS) is chosen as an end point, while in other studies alternative end points such as progression-free survival (PFS), time to relapse (TTR) or time to transformation (TTT) are chosen. OS and TTT may be directly correlated, but may both be influenced in an unpredictable way by the various second, third and further lines of treatment in contrast to PFS and TTR. Differences in patient selection may be a further confounding matter precluding full comparisons. The proportion of high-risk patients as reflected by the FL IPI distribution is very different between series with young, low-risk patients in one series as compared to a relative enrichment of elderly and higher risk patients in others ( Table 1 ).

Table 1 Clinical characteristics of 11 publications on the impact of various microenvironment aspects in follicular lymphoma.

Publication Type of analysis Number of patients End point Type of series Treatment (FL) IPI range (%) low-intermediate-high  
Lee End-of-spectrum 59 OS Single center Various, not specified Unknown Untreated, unselected series, avarage risk
Glas End-of-spectrum 66 TTT Single center Various, mostly CVP, no Rituximab Significant between outcome groups Untreated, unselected series, avarage risk
Carreras Continuous 97 OS Single center Various, 58% CHOP, 14% fludarabine, 23% Rituximab 37/27/35 Untreated, unselected series, avarage risk
Farina Continuous 99 OS, PFS Single center BP-VACOP+RT 59/40/1 Untreated, young, good risk, aggressive treatment
Alvaro Continuous 211 OS, PFS Multicenter Various, 44% CHOP, 15% CVP 39/38/23 Untreated, unselected series, avarage risk
Wahlin End-of-spectrum 70 OS Population-based cohort Various Significant between outcome groups Untreated, unselected series, avarage risk
Taskinen Continuous 96 OS Single center clinical trial R-CHOP versus CHOP 61/31/7 Both untreated and relapse, young, good risk
de Jong Continuous 45 PFS EORTC/BNLI 20921 CVP versus fludarabine, no Rituximab 26/44/30 Untreated, relatively low risk
Sweetenham Continuous 180 OS SWOG S8809, S9800, S9911 Promace-MOPP, CHOP, CHOP + tositumomab 59/33/7 Untreated, relatively lower risk, aggressive treatment
Klapper Continuous 158 OS, TTF GLSG 1996 MCF versus CHOP ± ASCT or αIFN 14/40/46 Untreated, older age groups, high risk
Canioni Continuous 194 EFS GELA-GOELAMS FL-2000 CHVP-I versus R-CHVP-I 54/43 Untreated, high tumor burden, relatively high risk

TTT: time to transformation, OS: overall survival, PFS: progression-free survival, ESF: event-free survival, TTF: time to failure.

Moreover, treatments both within and between studies are also very various with mostly CHOP, CVP, chlorambucil or fludarabin containing schemes, but also very aggressive treatments and a variable inclusion of immunotherapy (Rituximab). More systematic studies in uniformly treated patients have shed some more light on the influence of treatment on the impact of tumor microenvironment as a prognostic factor in FL. In a series of patients treated within an EORTC/BNLI clinical trial comparing fludarabine to cyclophosphamide, vincristine and prednisone (CVP) chemotherapy, some markers showed a homogeneous prognostic impact, while others had a different and sometimes opposite effect in the treatment arms [14] . An interfollicular infiltrate of FoxP3 positive T cells was a good prognostic sign irrespective of the treatment arm. Data were suggestive, however for a dense infiltrate of FoxP3 positive T cells, a dense and interfollicular infiltrate of CD68-positive macrophages and complete follicular dendritic meshworks to be associated with a favorable time to progression in CVP-treated patients, while being a poor prognostic sign in fludarabine-treated patients.

The dominant influence of treatment is further underpinned by studies in patients treated with chemo-immunotherapy (R-CHOP), but the results from various studies are equally contradictory. Based on 194 patients from the FL-2000 trial, the French GELA-GOELAMS group showed that a low macrophage content was a favorable feature in CHOP-treated patients as measured by EFS, while this effect was lost in R-CHOP-treated patients [15] . The Finnish group studied 96 patients from a single institution treated with R-CHOP compared to 45 historical treated differently (chemotherapy and/or radiotherapy). In this series, a similar impact of low macrophage content and good prognosis was found for patients treated without Rituximab [10] . The effect was reversed, however in the R-CHOP cohort and low macrophage content was then a poor prognostic sign. In a series by SWOG, no impact of TAM was demonstrated irrespective of addition of Rituximab to chemotherapy, however [16] . Similarly, the impact of FOXP3 positive regulatory T cells may be modulated by the addition of Rituximab.

It is attractive to speculate about the biological background of the differential impact of non-tumorous cell populations on the clinical behavior of FL and the response to chemotherapy. Based on the modes of action of different chemo(immuno)therapeutical agents, effective therapeutical strategies in FL act both directly on the tumor B cells and on the composition of the tumor microenvironment and disturb the immune interactions differently. Indeed, Cleary and coworkers have aimed to find a biological basis for the relation of tumor microenvironment and clinical behavior by studying the relation between neovasculature and macrophages. Early results showed an association between increased angiogenic activity, TAM and poor prognosis [17] .

A final confounding issue in the interpretation of immunohistochemistry-based studies is the reproducibility of scoring cell densities and patterns of infiltration. Although pre-ported studies are all well-performed, the Lunenburg Lymphoma Biomarker Consortium (LLBC) previously showed that the reproducibility of scoring of immunohistochemical stainings is rather disappointing and actually quite insufficient for several frequently reported markers. These validation studies were performed for markers in DLBCL and showed a major influence on reproducibility caused by technical variations, but after correction of this parameter, very significant variation caused by reproducibility of scoring remained [18] and [19]. This is quite comparable to reproducibility studies in solid malignancies as for Her2 expression in breast cancer [20] . These studies were all devoted to protein markers that are expressed on tumor cells and supposedly relatively easy to score. Preliminary validation studies of FL show a rather similar disappointing level of reproducibility for densities and distribution patterns of cell populations. More objective scoring methods such as flow cytometric (FACS) analysis and computer assisted scoring by image analysis could be a solution to this problem. Although also this approach should be formally validated, better reproducibility may be assumed.

Two studies using computerized image analysis and with concordant results should be discussed. Both concentrate on regulating T-cell populations: FOXP3 regulatory T cells (Treg) and PD-1 positive follicular helper T cells (TFH). Treatments in both series were heterogeneous in respect to chemotherapy schedules and Rituximab. Carreras and coworkers initially showed the correlation of dense Treg infiltration, especially perifollicular and intrafollicular as a favorable prognostic feature and subsequently also a high level of infiltration with TFH cells was identified as a favorable feature [21] . In a series especially enriched for poor prognosis cases (end-of-spectrum model) Wahlin et al. identified follicular CD4+ T cells as a poor prognostic sign, whereas follicular TFH, and Treg and interfollicular CD8+ T cells and CD68+ macrophages were associated with good outcome [22] .

In view of the disturbing heterogeneity of the reported results from various, all renowned, groups and even in uniformly treated patients, it may be too early for biological and certainly for practical clinical interpretations based on immunohistochemistry studies however. Further consequences for biological insights in FL as a disease will be discussed below.

The role of tumor microenvironment: experimental evidence

Functional assessment of FL tumors includes the characterization of cell populations and subsets along with the exploration of cell signaling events and their kinetics. In order to accede to functional approaches tumor tissues have to be freshly collected and handled through a standardized procedure starting with tissue dilacerations and cell re-suspension followed by several other steps depending on the final goal. The first technical approach opens directly on freshly or DMSO-conserved purified mononuclear cell suspensions is an immunophenotype analysis by multicolor flow cytometry, which one authorize, yet a multiparametric assay in a routine use owing to analyze up to 8 or 10 colors for a single cell in a single tube. The team headed by R. Levy in Stanford was the first to apply this technique in FL cases for the monitoring of phospho-protein responses to environmental cues in FL at the single cell level in order to identify profiles of aberrant B-cell receptor (BCR) signaling in individual malignant or normal B cells within lymphoma biopsies [23] . Following BCR cross-linking in the presence of H2O2 as phophatase inhibitor, levels of protein phophorylation accumulate rapidly and significantly higher in the 4 different tested FL B cells, including individual differences in term of signal kinetic between FL cases, compared to B cells issued from peripheral blood of healthy individuals and more interestingly to non-malignant tumor-infiltrating host B cells. Indeed, stimulated FL cells displayed greater levels of BCR-induced phosporylation of Syk (Y352), Btk, and MAPK suggesting that enhanced signaling through BCR may contribute to the malignant phenotype of FL. Thus, FL cells represent models suitable for evaluating pharmacologic targeting of BCR signaling. In a following paper this group developed a strategy involving 288 signaling measurements per patient including both lymphoma B cells and tumor-infiltrating T cells providing a large picture of signal transduction pathways known to play a role in cancer and capturing responses of key cancer-associated effectors for the stimuli used in the panel [24] . The study was carried on 56 FL tumor samples collected before any treatment from patients who received uniform initial CVP chemotherapy. They found a negative prognostic value independent of the FL IPI scoring for FL cases that presented a subpopulation of tumor B cells higher than 40% among all lymphoma B cells characterized by their lack of activatable BCR signaling associated to a CD20 down expression. Interestingly this emerging subclone, which shares identical heavy and light chain Ig isotype that the other lymphoma B cells, has a tumor growth advantage and accumulates extensive abrogation of the BCR signaling over time for a same patient after therapy and progression.

Despite the over-expression of anti-apoptotic protein Bcl-2 in almost all FL, this latter cannot explain by itself the selective advantage given to tumor cells raising the question of the role of immune microenvironment and stromal compartment as well, in the biology and pathogenesis of the tumor. Analysis of intraclonal sequence diversity and their follicular growth pattern showed that FL B cells migrate and “metastasize” distinct follicles where they found tumor follicles meaning that the follicular microenvironment is required for their clonal expansion [25] . The microenvironment dependency is also supported by the fact that FL B cells are hard to grow in vitro in absence of stromal cells and CD40 cytokine, both of them providing survival signals [26], [27], and [28]. The FL tumor is characterized by the maintenance of the follicular structure indicating that FL B cells remain dependent on cellular and molecular events that contribute to the normal germinal center (GC) reaction. In this latter, FDCs which are restricted to GCs allow B cell migration, selection and differentiation through a complex set of factors including BCR-mediated signal, chemokines, cytokines and adhesions molecules [29] . Meantime, the GC supports cross-talk between B cells and GC-dedicated CD4+ T cells that control their survival, proliferation and differentiation. The cell surface expression of programmed cell death 1 (PD-1) associated to co-stimulatory molecule ICOS and chemokine receptor CXCR5 allow identifying the CD4+ follicular helper T cells (TFH cell) which secrete IL-4 and IL-21 along with high levels of the chemokine CXCL13 [30] . Co-stimulating molecules positively regulate B-cell differentiation and for instance CD40L, transiently expressed by activated CD4+ T cells, stimulates B-cell proliferation and facilitates cytokine-induced class switching [31] . In FL, a high percentage of CD40L-expressing CD4+ T cells were found admixed with tumor cells [32] indicating that this factor remain crucial in this disease sustaining B-cell proliferation and protecting cell from apoptosis through the activation of the NF-kB pathway, which in turn up-regulates c-FLIP and Bcl-xL [24], [28], [33], and [34]. This effect has to be taken into account in the therapeutic context because of the CD40 signal provided by TFH that could abolish a beneficial anti-tumor effect mediated by TRAIL or eventually FAS [34] and [35]. Finally, the fact that a full PD-1 signaling is supported by the TFH in the tumor tissue this may explain the maintenance of functional features of normal GC in FL including high production of IL-4 and IL-21, both cytokines sustaining normal B-cell differentiation [30], [36], and [37] ( Fig. 1 ).


Fig. 1 Schematic illustration of interactions between B cells and their microenvironment in the context of the normal germinal center reaction and the follicular lymphoma niche. A. To make high-affinity, class-switched antibody, B cells must receive cognate help from TFH cells during the GC reaction leading to maturation of activated B cells along with production of memory B cells and plasma cells. In absence of T-cell help during B-cell priming by DCs followed by FDCs, B cells are driving to apoptosis. A range of cytokines including CD40L, IL-21 and IL-4 produced by TFH cells can direct antibody class switching. Moreover, TFH cells produce high levels of chemokine CXCL13 along with FDCs allowing the B-cell migration within an appropriate GC area where rescued B cells undergo final maturation. In the opposite, B cells produce ICOS-L which engages ICOS-driving production of cytokines in TFH. The next critical cell in the development of the GC reaction is the FDC that produces under specific and coordinated signaling from immune accessory cells and B cells themselves, a wide range of factors which support recruitment and survival of B cells. FDCs concentrate also antigen as immune complexes on their surface bridging BCR on B cells leading to a specific B-cell signaling involved in the cell activation and maturation. Several other hematopoietic cells are present during the GC reaction holding specific functions such as antigen presentation for DCs and macrophages, innate immune response for macrophages, NK cells and γδ T cells and adaptive immune response for CD8+ and Treg cells. B. Early at the de novo FL emergence, specific changes take place in the microenvironment induce either directly by the BCL2-translocated B cells (nuclear green-red bar code) or indirectly by emerging cell subsets including Treg which attenuate CD8+ T-cell function. TFH are highly represented in the FL tumor and they upregulate their IL-4 production sustaining B-cell survival. FDCs modify most likely their released factors in response to cross-talk modifications along with FL B cells but also through other cell subsets which show significant perturbation like macrophages. BCL2-translocated FL B cells present specific modifications including the BCR membrane complex and its secondary signaling. C. Progressed-FL disease shows large modifications of the tumors landscape. B cells present genetic instability (several bar codes) driving several cell function modifications including a constitutive BCR signal (red star). Meantime cells belonging to the normal GC reaction are vanishing (TFH, FDC, CD8+ T cells and others), FRC-like (pink stromal cells) along with TAMs appear in response to stress signals building a microenvironment specific of tumor aggressiveness including angiogenesis promotion. DC, dendritic cell; FDC, follicular dendritic cell; TFH, T follicular helper cells; GC, germinal centre reaction; ICOS, inducible T cell co-stimulator; Treg, regulatory T cells; NK, natural killer cells; MΦ, macrophage; TAM, tumor-associated macrophage; FL, follicular lymphoma.

The results for more experimental and basic research correlate to obvious observations in clinic and immunohistochemistry based studies. In FL, T cell number, functionality and distribution may impact patient prognosis. In order to explore more precisely the cross-talk between FL B-cells and their microenvironment, Pangault et al. settled an original approach on fresh tumor tissues and cell sorting procedures allowing comparison between the B-cell compartment and non-B compartment in FL versus reactive lymph nodes (Leukemia, in press). Gene-expression profiling completed with functional experiments shed new lights on FL biology and especially CD4+ T cells and lymphoma B cells cross-talk. Indeed, the TFH cell compartment in FL was found enriched compared to normal lymph node and at a level equivalent of chronically inflamed tonsils (Leukemia, in press). In parallel, FL microenvironment presents a strong IL-4 signature at the mRNA level confirming previous results obtained at the protein levels [38] associated with a functional signaling characterized by a significant increase of p-STAT6-positive B cells compared to benign follicular hyperplasia. Taken altogether these findings demonstrate in FL the presence of an IL-4-dependent TFH B-cell axis that could be involved in lymphomagenesis by inducing a STAT6 signaling, which one sustains tumor growth as described earlier in GC-like diffuse large B-cell lymphomas and Hodgkin’s diseases [39] and [40]. Using highly purified CD4-positive and non-B-cell subsets, IL-4 cytokine was shown to be predominantly present in FL-derived TFH compared to TFH from non-malignant human tonsil or hyperplasic lymph nodes suggesting a specific activation of CD4+ T cells in FL (Leukemia in press). Taken in account the fact that TFH promotes the emergence of B cell–T cell conjugates, in FL the increase of the CD4+ compartment could shape the FL B-cell biology with maintenance of a GC-like B-cell phenotype including the expression of activation-induced cytidine deaminase and ongoing somatic hypermutation [41], [42], [43], and [44]. Furthermore, formation of B cell–TFH cell conjugates associated with the resulting cytokine secretion regulates cell fate decisions in normal GC and could drive a chronic antigenic stimulation through a process called idiotypic-driven T–B collaboration [45] . Interestingly, it has been possible to demonstrate in mice models that chronically active idiotypic-specific T cells can help or even promote the emergence of indolent-type B-cell lymphomas [46] . Collectively these results contribute to reinforce the idea that in FL it is into the microenvironment that lies the major lymphomagenesis contributors, which sustain cell growth and survival creating a specific FL tumor niche. Of course, FL B cells modify the microenvironment and contribute to the emergence of the FL tumor niche by impacting for instance the CD4+ T-cell differentiation [47] and inducing therefore T-cell defect after short-term tumor cell contact affecting the recruitment of critical signaling in the immunological synapse [48] .

The other main activated CD4+ regulating T-cell population in FL was first identified in a study by Yang and coworkers as a mean of 17% [range: 6–43%] of T cells highly positive at the cell surface for CD25 and a vast majority of these positive for both Foxp3 and CTLA-4 corresponding therefore to CD4+ Treg cells [49] . These cells were found in a much higher proportion compared to non-malignant reactive lymph nodes and, functionally in B-cell lymphomas including FLs they were able to inhibit the production of IFN-γ, and in a less extent IL-4, by tumor-infiltrating CD4+ CD25− T cells. Interestingly, tumor-associated Treg interact also with autologous tumor-infiltrating CD8+ cytotoxic T cells by attenuating their proliferation, granule production & release, and their capacity to lyse lymphoma cells. These findings were further explored by the authors of the paper who finally described a strong inverse relationship between the frequency of intratumoral Treg cells and the percentage of CD8 T cells in biopsy specimens suggesting that an increased number of Treg damps the immune anti-tumor response within the tumor microenvironment [50] . In this scenario the same team proposed that lymphoma B cells – and especially FL B cells which were dominantly investigated compared to other types of B lymphomas – seem to play a major role by secreting the CCL22 chemokine, which induces migration of CCR4+ Treg cells toward the tumor niche [49] , and also by producing IL-2 cytokine establishing therefore an imbalance between Treg and TH17 leading to an upregulation of Treg cells and inhibition of TH17 cells. Altogether these results indicate that tumor B cells favor the establishment of a profoundly inhibitory tumor microenvironment [51] .

Therefore T-cell modulation by specific drugs may improve therapeutic response in FL to chemotherapy or vaccine [48] and [52]. The Swedish group of Kimby and coworkers conducted a flow cytometry study involving 139 cases based on the evaluation in the patient’s initial FL tumor of T-cell subsets and showed that higher levels of CD8+ T cells predict a better outcome [11] . This data militate to the restoration of an efficient anti-tumor CD8+ cytotoxic T cell immune adaptive response in FL through a vaccine therapy for instance. Using a TLR-9 agonist strategy aimed to boost the CD8 T-cell response, Ron Levy’s group was able to show lymphoma regression in a mouse model but also in a phase I/II study involving 15 patients [53] and [54]. Besides CD8+ T cells, the γδ T lymphocytes exhibit also a cytotoxic function but belong to the innate immune response. Both T cells, which seem to participate to the immune response-1 described in FL patients with good prognosis and long survival, reside in the perifollicular zone in both, FL and non-malignant hyperplastic lymph nodes, but their number is significantly reduced in FL compared to non-tumoral cases [4], [55], and [56].

The stromal compartment of the lymph node was recently more extensively studied using for this purpose specific purification and cell culture conditions in order to isolate viable tumor-associated stromal cells. In normal lymph nodes three different subsets exist and participate in the anatomical compartmentalization [57] . In FL, the first paper with extensive data about stromal cells demonstrated that lymphoma B cells modify the tumor-associated stromal compartment through the differentiation of resident mesenchymal stem-cells to fully competent fibroblastic reticular cells able to recruit B tumor cells and to sustain their survival [58] . This effect is related partially to the production of TNFα and lymphotoxin-α1β2 either by FL B cells themselves or FL microenvironment [59] . Macrophages may be the source of TNFα and especially when they are exposed to IL-4 and IL-10, both inducing an M2 phenotype characterized by their ability to suppress T cell adaptive immunity and promote angiogenesis [60] and [61]. Macrophages are recruited into tumor as monocytes from the bloodstream by the release from the tumor microenvironment, in particular stromal cells, of chemokines like CCL2, VEGF, and IL-8 [62] . In FL, M2 macrophages have been shown to be associated with vascular sprouts [17] demonstrating the close connection between activated monocytes/macrophages and angiogenesis [62] . Furthermore, both monocytes and macrophages can induce CD40-independent Ig class switching of normal B cells and support survival and proliferation of GC-derived B-cell lymphomas [63] and [64]. Together, stromal and myeloid cells mediated many of the changes in the microenvironment during tumor progression and can directly compromise normal mature B-cell differentiation and finally build up with the T-cell compartment a specific FL niche in the secondary lymphoid organs but also in the bone marrow [57] and [65]. A large body of work as still to be done to better understand how the multiple cell populations present in the lymph nodes contribute together to host the BCL2-translocated tumor initiating cells which could secondary give rise to FL [66] . Specific cross-talk between cell populations of diverse origin may take place – for instance neutrophils, monocytes and macrophages that mediate inflammatory responses – involving specific signaling pathways and specific extracellular protein processes that can balance further the immune control and promote cancer emergence [67], [68], and [69].

Novel approaches for treatment based on insights in the role of tumor microenvironment

A tentative model for the role of the tumor microenvironment in FL would encompass a role for the effector T-cell compartment of CD4 helper and CD8 suppressor cells providing potential growth support and cytotoxicity respectively ( Fig. 2 A). Regulator populations such as FoxP3 positive Treg and PD-1 positive FTH cells may exert their effects through regulation of these effector populations. The cytokine, chemokines and other intermediates such as IL-4, IL-10, TNFα, lymphotoxin-α1β2 and TLR receptors mediate these regulatory functions and may be partly produced by macrophages and by the tumor microvasculature that itself can also serve as an effector compartment. Interference with the net growth supportive balance of the tumor microenvironment forms an attractive new target for treatment approaches ( Fig. 2 B). It may also explain the effectiveness of conventional chemotherapeutical agents. For both, fludarabin and cyclophophamide known effective chemotherapeuticals in FL, a direct effect on regulatory T-cell populations, especially FoxP3 positive Treg cells has been shown. For fludarabine reduced frequencies of FOXP3 positive T cells after treatment but also reduced inhibitory functions have been demonstrated [70] . Next to and likely related to an effect on Treg by cyclophosphamide, restoration of the proliferative capacity of effector T cells has been demonstrated [71] and [72]. It should be noted, however, that these effects are measured either in vitro or in the peripheral blood and not at the tissue level. Monitoring of T-cell populations in follow-up endoscopial biopsies during fludarabin treatment in gastric MALT-lymphoma patients, we could show that quite reverse alterations may take place at the tissue level and that the effects may be less predictable than theoretically expected (de Boer et al. submitted).


Fig. 2 Functional results of the interactions of regulating and effector T-cell populations as target for treatment in follicular lymphoma. A. Regulating T-cell populations such as TFH and Treg exert their influence primarily via effector T-cell populations and together with FDC and other accessory cell populations such as histiocytes result in growth supportive and cytotoxic signals to the FL tumor cells. B. The interactions between microenvironment and FL tumor cells can serve as a target for treatment with the aim to boost the cytotoxic influence and to interfere with growth support. Various classical cytotoxic drugs are known to have a dominant effect on specific T-cell populations, while newer compounds are specifically designed for this purpose. Low-dose radiotherapy has been shown to operate specifically on activation of accessory cell populations. The results of interventions largely depend on the relative balance between the various microenvironmental factors and their net growth supportive or cytotoxic effect and are difficult to predict at this stage.

Agents that are more specifically designed to modulate T-cell functions have also been studied in lymphoma. The TLR-9 agonist 1018ISS is designed to interfere with both the innate and adaptive immune responses and especially by enhancing cytotoxicity. In FL patients, this agent together with Rituximab has shown to be effective with a 48% response rate [73] . In the peripheral blood, increase of CD3+ T cells, elevated levels of IFN-γ and interferon-inducible genes could be demonstrated and increased infiltration of CD8+ T cells at the injection site. No data of alterations at the tumor site could be studied and therefore the same caveats hold true as discussed for conventional chemotherapeutical agents. The approaches by the group of Levy in conjunction with vaccination point in the same direction [53] . Histone-deacetylases (HDACs) form another class of immune modulating agents and may specifically target FOXP3 positive Treg and thereby influence the activity of effector cells, dendritic cells and other components of the inflammatory cascade [74] . A net anti-inflammatory response has been demonstrated for HDAC inhibitors and thereby these agents have proven effectiveness in inflammatory and autoimmune diseases such as colitis, arthritis and graft versus host disease, No data on clinical effects in FL are known and in view of the balances between growth support and suppression in this disease, the results are difficult to predict. Theoretically, CXCR4 antagonists as plerixafor/AMD3100 are similarly attractive agents for immune microenvironment interference especially through blockage of T helper growth supportive effects. Although this agent is used for stem-cell mobilization in lymphoma patients, also here, no data are available on its effect on the lymphoma itself or specifically on lymphoma microenvironment. Various other potential immunomodulatory agents, including thalidomide and lenalidomide have been proposed as instrumental in a favorable response in FL [48], [75], and [76]. In many cases, however, the precise signaling pathways that are targeted are largely unknown and often multiple immunomodulatory and other pathways and both directed toward tumor cells and non-tumor cells are involved precluding any mechanistic conclusions at this stage.

In conclusion, it is clear that the tumor microenvironment in FL is such a strong determining factor in the clinical behavior of the disease that this will be the prime target of experimental treatments in the near future. Adequate in vitro and especially mouse models are sorely needed to study the precise balances between the various components of the immune microenvironment to better predict the effect of specific interferences. In the end, the net clinical effect of tumor response and survival will be the ultimate read-out of these theoretical considerations.

Conflict of interest

No conflict of interests to declare.


  • [1] H. Husson, E.G. Carideo, D. Neuberg, et al. Gene expression profiling of follicular lymphoma and normal germinal center B cells using cDNA arrays. Blood. 2002;99:282-289 Crossref.
  • [2] A.M. Glas, M.J. Kersten, L.J. Delahaye, et al. Gene expression profiling in follicular lymphoma to assess clinical aggressiveness and to guide the choice of treatment. Blood. 2005;105:301-307 Crossref.
  • [3] A.M. Glas, L. Knoops, L. Delahaye, et al. Gene-expression and immunohistochemical study of specific T-cell subsets and accessory cell types in the transformation and prognosis of follicular lymphoma. J Clin Oncol. 2007;25:390-398 Crossref.
  • [4] S.S. Dave, G. Wright, B. Tan, et al. Prediction of survival in follicular lymphoma based on molecular features of tumor-infiltrating immune cells. N Engl J Med. 2004;351:2159-2169 Crossref.
  • [5] J. Carreras, A. Lopez-Guillermo, B.C. Fox, et al. High numbers of tumor-infiltrating FOXP3-positive regulatory T cells are associated with improved overall survival in follicular lymphoma. Blood. 2006;108:2957-2964 Crossref.
  • [6] P. Farinha, H. Masoudi, B.F. Skinnider, et al. Analysis of multiple biomarkers shows that lymphoma-associated macrophage (LAM) content is an independent predictor of survival in follicular lymphoma (FL). Blood. 2005;106:2169-2174 Crossref.
  • [7] T. Kelley, R. Beck, A. Absi, et al. Biologic predictors in follicular lymphoma: importance of markers of immune response. Leuk Lymphoma. 2007;48:2403-2411 Crossref.
  • [8] W. Klapper, E. Hoster, L. Rolver, et al. Tumor sclerosis but not cell proliferation or malignancy grade is a prognostic marker in advanced-stage follicular lymphoma: the German Low Grade Lymphoma Study Group. J Clin Oncol. 2007;25:3330-3336 Crossref.
  • [9] A.M. Lee, A.J. Clear, M. Calaminici, et al. Number of CD4+ cells and location of forkhead box protein P3-positive cells in diagnostic follicular lymphoma tissue microarrays correlates with outcome. J Clin Oncol. 2006;24:5052-5059 Crossref.
  • [10] M. Taskinen, M.L. Karjalainen-Lindsberg, H. Nyman, et al. A high tumor-associated macrophage content predicts favorable outcome in follicular lymphoma patients treated with rituximab and cyclophosphamide-doxorubicin-vincristine-prednisone. Clin Cancer Res. 2007;13:5784-5789 Crossref.
  • [11] B.E. Wahlin, B. Sander, B. Christensson, et al. CD8+ T-cell content in diagnostic lymph nodes measured by flow cytometry is a predictor of survival in follicular lymphoma. Clin Cancer Res. 2007;13:388-397 Crossref.
  • [12] T. Alvaro, M. Lejeune, M.T. Salvado, et al. Immunohistochemical patterns of reactive microenvironment are associated with clinicobiologic behavior in follicular lymphoma patients. J Clin Oncol. 2006;24:5350-5357 Crossref.
  • [13] M. Taskinen, M.L. Karjalainen-Lindsberg, S. Leppa. Prognostic influence of tumor-infiltrating mast cells in patients with follicular lymphoma treated with rituximab and CHOP. Blood. 2008;111:4664-4667 Crossref.
  • *[14] D. de Jong, A. Koster, A. Hagenbeek, et al. Impact of the tumor microenvironment on prognosis in follicular lymphoma is dependent on specific treatment protocols. Haematologica. 2009;94:70-77 Crossref.
  • [15] D. Canioni, G. Salles, N. Mounier, et al. High numbers of tumor-associated macrophages have an adverse prognostic value that can be circumvented by rituximab in patients with follicular lymphoma enrolled onto the GELA-GOELAMS FL-2000 trial. J Clin Oncol. 2008;26:440-446 Crossref.
  • [16] J.W. Sweetenham, B. Goldman, M.L. LeBlanc, et al. Prognostic value of regulatory T cells, lymphoma-associated macrophages, and MUM-1 expression in follicular lymphoma treated before and after the introduction of monoclonal antibody therapy: a Southwest Oncology Group Study. Ann Oncol. 2010;21:1196-1202 Crossref.
  • *[17] A.J. Clear, A.M. Lee, M. Calaminici, et al. Increased angiogenic sprouting in poor prognosis FL is associated with elevated numbers of CD163+ macrophages within the immediate sprouting microenvironment. Blood. 2010;115:5053-5056 Crossref.
  • *[18] de Jong D, Rosenwald A, Chhanabhai M, et al. Immunohistochemical prognostic markers in diffuse large B-cell lymphoma: validation of tissue microarray as a prerequisite for broad clinical applications–a study from the Lunenburg Lymphoma Biomarker Consortium. J Clin Oncol 2007;25:805–812.
  • [19] D. de Jong, W. Xie, A. Rosenwald, et al. Immunohistochemical prognostic markers in diffuse large B-cell lymphoma: validation of tissue microarray as a prerequisite for broad clinical applications (a study from the Lunenburg Lymphoma Biomarker Consortium). J Clin Pathol. 2009;62:128-138 Crossref.
  • [20] E.A. Perez, V.J. Suman, N.E. Davidson, et al. HER2 testing by local, central, and reference laboratories in specimens from the North Central Cancer Treatment Group N9831 intergroup adjuvant trial. J Clin Oncol. 2006;24:3032-3038 Crossref.
  • *[21] J. Carreras, A. Lopez-Guillermo, G. Roncador, et al. High numbers of tumor-infiltrating programmed cell death 1-positive regulatory lymphocytes are associated with improved overall survival in follicular lymphoma. J Clin Oncol. 2009;27:1470-1476 Crossref.
  • *[22] B.E. Wahlin, M. Aggarwal, S. Montes-Moreno, et al. A unifying microenvironment model in follicular lymphoma: outcome is predicted by programmed death-1–positive, regulatory, cytotoxic, and helper T cells and macrophages. Clin Cancer Res. 2010;16:637-650 Crossref.
  • [23] J.M. Irish, D.K. Czerwinski, G.P. Nolan, et al. Altered B-cell receptor signaling kinetics distinguish human follicular lymphoma B cells from tumor-infiltrating nonmalignant B cells. Blood. 2006;108:3135-3142 Crossref.
  • *[24] J.M. Irish, J.H. Myklebust, A.A. Alizadeh, et al. B-cell signaling networks reveal a negative prognostic human lymphoma cell subset that emerges during tumor progression. Proc Natl Acad Sci U S A. 2010;107:12747-12754 Crossref.
  • [25] S. Oeschger, A. Brauninger, R. Kuppers, et al. Tumor cell dissemination in follicular lymphoma. Blood. 2002;99:2192-2198 Crossref.
  • [26] C. Buske, A. Twiling, G. Gogowski, et al. In vitro activation of low-grade non-Hodgkin's lymphoma by murine fibroblasts, IL-4, anti-CD40 antibodies and the soluble CD40 ligand. Leukemia. 1997;11:1862-1867
  • [27] P. Ghia, V.A. Boussiotis, J.L. Schultze, et al. Unbalanced expression of bcl-2 family proteins in follicular lymphoma: contribution of CD40 signaling in promoting survival. Blood. 1998;91:244-251
  • [28] P.W. Johnson, S.M. Watt, D.R. Betts, et al. Isolated follicular lymphoma cells are resistant to apoptosis and can be grown in vitro in the CD40/stromal cell system. Blood. 1993;82:1848-1857
  • [29] J.G. Tew, J. Wu, M. Fakher, et al. Follicular dendritic cells: beyond the necessity of T-cell help. Trends Immunol. 2001;22:361-367 Crossref.
  • [30] C. King. New insights into the differentiation and function of T follicular helper cells. Nat Rev Immunol. 2009;9:757-766 Crossref.
  • [31] C. van Kooten, J. Banchereau. Functions of CD40 on B cells, dendritic cells and other cells. Curr Opin Immunol. 1997;9:330-337
  • [32] A. Carbone, A. Gloghini, H.J. Gruss, et al. CD40 ligand is constitutively expressed in a subset of T cell lymphomas and on the microenvironmental reactive T cells of follicular lymphomas and Hodgkin's disease. Am J Pathol. 1995;147:912-922
  • [33] D. Schmitter, M. Koss, E. Niederer, et al. T-cell derived cytokines co-stimulate proliferation of CD40-activated germinal centre as well as follicular lymphoma cells. Hematol Oncol. 1997;15:197-207 Crossref.
  • [34] M. Travert, P. Ame-Thomas, C. Pangault, et al. CD40 ligand protects from TRAIL-induced apoptosis in follicular lymphomas through NF-kappaB activation and up-regulation of c-FLIP and Bcl-xL. J Immunol. 2008;181:1001-1011
  • [35] A.M. Cleary, S.M. Fortune, M.J. Yellin, et al. Opposing roles of CD95 (Fas/APO-1) and CD40 in the death and rescue of human low density tonsillar B cells. J Immunol. 1995;155:3329-3337
  • [36] K.L. Good-Jacobson, C.G. Szumilas, L. Chen, et al. PD-1 regulates germinal center B cell survival and the formation and affinity of long-lived plasma cells. Nat Immunol. 2010;11:535-542 Crossref.
  • [37] M.A. Linterman, L. Beaton, D. Yu, et al. IL-21 acts directly on B cells to regulate Bcl-6 expression and germinal center responses. J Exp Med. 2010;207:353-363 Crossref.
  • [38] C. Pangault, P. Ame-Thomas, P. Ruminy, et al. Follicular lymphoma cell niche: identification of a preeminent IL-4-dependent T(FH)-B cell axis. Leukemia. 2011;24:2080-2089
  • [39] K.R. Calvo, B. Dabir, A. Kovach, et al. IL-4 protein expression and basal activation of Erk in vivo in follicular lymphoma. Blood. 2008;112:3818-3826 Crossref.
  • [40] X. Lu, H. Nechushtan, F. Ding, et al. Distinct IL-4-induced gene expression, proliferation, and intracellular signaling in germinal center B-cell-like and activated B-cell-like diffuse large-cell lymphomas. Blood. 2005;105:2924-2932 Crossref.
  • [41] D. Baus, F. Nonnenmacher, S. Jankowski, et al. STAT6 and STAT1 are essential antagonistic regulators of cell survival in classical Hodgkin lymphoma cell line. Leukemia. 2009;23:1885-1893 Crossref.
  • [42] R.L. Reinhardt, H.E. Liang, R.M. Locksley. Cytokine-secreting follicular T cells shape the antibody repertoire. Nat Immunol. 2009;10:385-393 Crossref.
  • [43] M.S. Hardianti, E. Tatsumi, M. Syampurnawati, et al. Activation-induced cytidine deaminase expression in follicular lymphoma: association between AID expression and ongoing mutation in FL. Leukemia. 2004;18:826-831 Crossref.
  • [44] Okada T, Miller MJ, Parker I, et al. 2005. Antigen-engaged B cells undergo chemotaxis toward the T zone and form motile conjugates with helper T cells. PLoS Biol 3:e150.
  • [45] C. Zhou, A. Saxon, K. Zhang. Human activation-induced cytidine deaminase is induced by IL-4 and negatively regulated by CD45: implication of CD45 as a Janus kinase phosphatase in antibody diversification. J Immunol. 2003;170:1887-1893
  • [46] B. Bogen, S. Weiss. Processing and presentation of idiotypes to MHC-restricted T cells. Int Rev Immunol. 1993;10:337-355 Crossref.
  • [47] M.M. Zangani, M. Froyland, G.Y. Qiu, et al. Lymphomas can develop from B cells chronically helped by idiotype-specific T cells. J Exp Med. 2007;204:1181-1191 Crossref.
  • *[48] W.Z. Ai, J.Z. Hou, R. Zeiser, et al. Follicular lymphoma B cells induce the conversion of conventional CD4+ T cells to T-regulatory cells. Int J Cancer. 2009;124:239-244 Crossref.
  • [49] A.G. Ramsay, A.J. Clear, G. Kelly, et al. Follicular lymphoma cells induce T-cell immunologic synapse dysfunction that can be repaired with lenalidomide: implications for the tumor microenvironment and immunotherapy. Blood. 2009;114:4713-4720 Crossref.
  • [50] Z.Z. Yang, A.J. Novak, M.J. Stenson, et al. Intratumoral CD4+CD25+ regulatory T-cell-mediated suppression of infiltrating CD4+ T cells in B-cell non-Hodgkin lymphoma. Blood. 2006;107:3639-3646 Crossref.
  • *[51] Z.Z. Yang, A.J. Novak, S.C. Ziesmer, et al. Attenuation of CD8(+) T-cell function by CD4(+)CD25(+) regulatory T cells in B-cell non-Hodgkin's lymphoma. Cancer Res. 2006;66:10145-10152 Crossref.
  • *[52] Z.Z. Yang, A.J. Novak, S.C. Ziesmer, et al. Malignant B cells skew the balance of regulatory T cells and TH17 cells in B-cell non-Hodgkin's lymphoma. Cancer Res. 2009;69:5522-5530 Crossref.
  • [53] R. Houot, R. Levy. T-cell modulation combined with intratumoral CpG cures lymphoma in a mouse model without the need for chemotherapy. Blood. 2009;113:3546-3552 Crossref.
  • [54] J.D. Brody, W.Z. Ai, D.K. Czerwinski, et al. In Situ Vaccination With a TLR9 Agonist Induces Systemic Lymphoma Regression: A Phase I/II Study. J Clin Oncol. 2010;
  • [55] B. Varghese, A. Widman, J. Do, et al. Generation of CD8+ T cell-mediated immunity against idiotype-negative lymphoma escapees. Blood. 2009;114:4477-4485 Crossref.
  • [56] M.S. Braza, A. Caraux, T. Rousset, et al. gammadelta T lymphocytes count is normal and expandable in peripheral blood of patients with follicular lymphoma, whereas it is decreased in tumor lymph nodes compared with inflammatory lymph nodes. J Immunol. 2010;184:134-140 Crossref.
  • [57] L.M. Staudt, S. Dave. The biology of human lymphoid malignancies revealed by gene expression profiling. Adv Immunol. 2005;87:163-208 Crossref.
  • [58] S.N. Mueller, R.N. Germain. Stromal cell contributions to the homeostasis and functionality of the immune system. Nat Rev Immunol. 2009;9:618-629
  • [59] P. Ame-Thomas, H. Maby-El Hajjami, C. Monvoisin, et al. Human mesenchymal stem cells isolated from bone marrow and lymphoid organs support tumor B-cell growth: role of stromal cells in follicular lymphoma pathogenesis. Blood. 2007;109:693-702 Crossref.
  • [60] H. Maby-El Hajjami, P. Ame-Thomas, C. Pangault, et al. Functional alteration of the lymphoma stromal cell niche by the cytokine context: role of indoleamine-2,3 dioxygenase. Cancer Res. 2009;69:3228-3237 Crossref.
  • [61] A. Mantovani, S. Sozzani, M. Locati, et al. Macrophage polarization: tumor-associated macrophages as a paradigm for polarized M2 mononuclear phagocytes. Trends Immunol. 2002;23:549-555 Crossref.
  • [62] L. Yang, L.M. DeBusk, K. Fukuda, et al. Expansion of myeloid immune suppressor Gr+CD11b+ cells in tumor-bearing host directly promotes tumor angiogenesis. Cancer Cell. 2004;6:409-421 Crossref.
  • [63] A. Mantovani, P. Allavena, A. Sica, et al. Cancer-related inflammation. Nature. 2008;454:436-444 Crossref.
  • [64] M.B. Litinskiy, B. Nardelli, D.M. Hilbert, et al. DCs induce CD40-independent immunoglobulin class switching through BLyS and APRIL. Nat Immunol. 2002;3:822-829 Crossref.
  • [65] C.G. Mueller, C. Boix, W.H. Kwan, et al. Critical role of monocytes to support normal B cell and diffuse large B cell lymphoma survival and proliferation. J Leukoc Biol. 2007;82:567-575 Crossref.
  • [66] M. Rafei, J. Hsieh, S. Fortier, et al. Mesenchymal stromal cell-derived CCL2 suppresses plasma cell immunoglobulin production via STAT3 inactivation and PAX5 induction. Blood. 2008;112:4991-4998 Crossref.
  • [67] S. Roulland, J.M. Navarro, P. Grenot, et al. Follicular lymphoma-like B cells in healthy individuals: a novel intermediate step in early lymphomagenesis. J Exp Med. 2006;03:2425-2431
  • [68] D.I. Gabrilovich, S. Nagaraj. Myeloid-derived suppressor cells as regulators of the immune system. Nat Rev Immunol. 2009;9:162-174 Crossref.
  • [69] O. Soehnlein, L. Lindbom. Phagocyte partnership during the onset and resolution of inflammation. Nat Rev Immunol. 2010;10:427-439 Crossref.
  • [70] K. Kessenbrock, V. Plaks, Z. Werb. Matrix metalloproteinases: regulators of the tumor microenvironment. Cell. 2010;141:52-67 Crossref.
  • [71] M. Beyer, M. Kochanek, K. Darabi, et al. Reduced frequencies and suppressive function of CD4+CD25hi regulatory T cells in patients with chronic lymphocytic leukemia after therapy with fludarabine. Blood. 2005;106:2018-2025 Crossref.
  • [72] F. Ghiringhelli, C. Menard, P.E. Puig, et al. Metronomic cyclophosphamide regimen selectively depletes CD4+CD25+ regulatory T cells and restores T and NK effector functions in end stage cancer patients. Cancer Immunol Immunother. 2007;56:641-648 Crossref.
  • *[73] M.E. Lutsiak, R.T. Semnani, R. De Pascalis, et al. Inhibition of CD4(+)25+ T regulatory cell function implicated in enhanced immune response by low-dose cyclophosphamide. Blood. 2005;105:2862-2868 Crossref.
  • [74] J.W. Friedberg, J.L. Kelly, D. Neuberg, et al. Phase II study of a TLR-9 agonist (1018 ISS) with rituximab in patients with relapsed or refractory follicular lymphoma. Br J Haematol. 2009;146:282-291 Crossref.
  • [75] L. Wang, E.F. de Zoeten, M.I. Greene, et al. Immunomodulatory effects of deacetylase inhibitors: therapeutic targeting of FOXP3+ regulatory T cells. Nat Rev Drug Discov. 2009;8:969-981
  • [76] J.A. Burger, P. Ghia, A. Rosenwald, et al. The microenvironment in mature B-cell malignancies: a target for new treatment strategies. Blood. 2009;114:3367-3375 Crossref.


a The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands

b Unité INSERM U917, Pôle Cellules & Tissus, CHU Pontchaillou, 35043 Rennes, France

Corresponding author. Tel.: +31 20 5122752; Fax: +31 20 5122759.

1 Tel.: +33 2 99284272; Fax: +33 2 99284152.