Dynamic roles of tumor-infiltrating B lymphocytes in cancer immunotherapy

Abstract

The amazing diversity of B cells within the tumor microenvironment is the basis for the diverse development of B cell-based immunotherapies. Here, we focus on elucidating the mechanisms of tumor intervention mediated by four tumor-infiltrating B lymphocytes. Naive B cells present the initial antigen, germinal center B cell subsets enhance antibody affinity, and immunoglobulin subtypes exert multiple immune effects, while regulatory B cells establish immune tolerance. Together they reflect the complexity of the changing dynamics of cancer immunity. Additionally, we have investigated the dynamic effects of tumor-infiltrating B lymphocytes in immunotherapy and their relationship to prognosis, providing new insights into potential treatment strategies for patients.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00262-024-03936-7.

Introduction

The core concept of the cancer immunity cycle, proposed in 2013, is that T cell cytotoxicity leads to the persistence and initiation of emerging or pre-existing canonical T cell responses [1]. This concept has been extended to include the novel view that the tumor microenvironment (TME), and in particular dendritic cells (DCs), play a critical role in regulating and maintaining anti-tumor resistance [2, 3]. However, CD8⁺ tumor-infiltrating lymphocytes (TILs) essentially do not operate in isolation. The research team systematically reviewed articles on the prognostic impact of tumor-infiltrating B cells (TIL-B) and/or plasma cells (PC) on human cancers and found that tumors with CD8⁺ TILs were also frequently infiltrated by CD20⁺ B cells, leading to the conclusion that “B-lineage cells play a critical role in most types of cancer” and suggesting that this finding should be used to design coordinated, multifaceted immunotherapy strategies [4]. B cells have a unique antigen-concentrating function that can enhance the immune response of other immune cells to tumors [5]. In parallel, studies have shown that CD20⁺ TILs colocalize with T cells and express antigen-presenting cell (APC)-specific cell surface markers. In an era dominated by T cell therapies, researchers have broken de novo ground by identifying the close relationship of TIL-B with myeloid cells, T cells, and other inflammatory cells. They have also highlighted the great potential of TIL-B in promoting anti-tumor immunity [6]. This paves the way for further development of TIL-B and its antibody-mediated effector mechanisms to facilitate the emergence of a new generation of cancer immunotherapies.

We will focus on the multiple roles of different B cell subsets in the dynamic regulation of tumor immune responses. Based on previous experimental studies, immunohistochemistry [7, 8] and flow cytometry [9] analyses of tumor and serum samples, evaluation of immunoglobulin molecules by DNA sequencing, and determination of serum autoantibody responses to tumor antigens via ELISA, along with B cell cloning or functional assays, the multiple studies demonstrate the prevalence of an atypical TACI⁺CD27⁺switched memory B cell phenotype (CD19⁺CD20⁺IgD⁻) [10], CD19⁺CD20⁺IgD⁺CD27⁻CD38⁻FCER2⁺TCL1A⁺ unswitched naive B cell phenotype, a minority of CD19⁺IgD⁻CD27⁺CXCR3⁺CD20⁻CD38⁺CD138^−/low plasma cell phenotype [10], CD19⁺CD20⁺IgD⁻CD10⁺CD38⁺BCL-6⁺AID⁺MHC-II⁺ germinal center B cell phenotype [11], and CD1d⁺CD19⁺IL-10⁺IL-35⁺TGF-β⁺ regulatory B cells (Bregs) phenotype [12], etc. TIL-B are enriched in memory B cells and PCs compared to B cells in peripheral blood or healthy tissues [13, 14]. Notably, although tumor-infiltrating germinal center cells (TIL-GCs) have been detected in most studies, they appear to be the most atypical B cell subset within the tumor-infiltrating lymphocyte (TIL) compartment. In the following, the different types of tumor-infiltrating B cells mentioned above will be described one by one, with expansion and discussion.

The intricate journey of germinal center B cells: proliferation, selection, and antibody optimization

We focus on activated B cell follicles (secondary), where germinal centers develop during the T cell-dependent antibody response period [15]. The germinal center (GC) is a transient lymphoid structure that forms when B lymphocytes are activated in response to antigenic stimulation. It plays a critical role in the maturation and diversification of B cells during the immune response. Within this microanatomical structure, B lymphocyte clonal expansion and affinity maturation occur. Somatic antibody mutations also occur in germinal centers, providing a specific environment for their replication and selection [16]. Over time, through rounds of Darwinian selection, a proportion of these “selected B cells” proliferate and differentiate into memory B cells and plasma cells [17]. These cells, which are then released into the circulation, not only contribute to the immune response, but also have a high affinity for serum antibodies.

During B cell maturation, somatic hypermutation (SHM) can fine-tune the low affinity repertoire generated by the variable diversity joining (VDJ) rearrangement. It introduces point mutations that alter the amino acid sequence of the antibody binding site, thereby optimizing antibody affinity [18, 19]. This protective mechanism, similar to the body’s response to irritants, occurs during the primary response, secondary immune response, and reinfection [20]. The “selected B cells” undergo a selection process in conjunction with T follicular helper cells (Tfh cells) that recognize peptide fragments of the same antigen. In this process, B cells with high antigen-binding affinity survive, while those with low affinity are eliminated. Affinity is not only a key factor in this selection, but also serves as the primary driving force behind B cell evolution and the optimization of antibody responses within germinal center responses. During this phase, surviving B cells enter the light zone (LZ) where they can bind to antigens presented on follicular dendritic cells (FDCs). Unlike other antigen-presenting cells, FDCs present intact antigens, allowing B cells to recognize them without prior processing. B cells are then further selected based on the recognition of antigen–antibody complexes by T cells. This process has been termed antigen-driven selection in some studies [21, 22], but is more accurately described as selection mediated indirectly by antigen presentation on MHC II molecules. In addition, research involving predictive modeling and in vivo experiments has suggested that the efficiency of the germinal center response is significantly influenced by intact Toll-like receptor 4 (TLR4)-mediated signaling within FDCs. TLR4 initiates a series of pathogen recognition and inflammatory signaling mechanisms by binding to lipopolysaccharide (LPS), which can regulate SHM mechanisms and immunoglobulin gene rearrangements in B cells, thereby enhancing antibody diversity [23]. B cells in the GC undergo an ordered shuttling between the dark zone (DZ) and the light zone (LZ) regions, forming a rhythmic flow of back-and-forth movement. This process can be referred to as the oscillatory cycle of GC B cells [24] until antibodies of extremely high-affinity antibodies are produced (Fig. 1). It has been reported that this journey takes 2 to 3 weeks. At this point, Tfh cells play a critical selective role by allowing only B cells with high enough affinity to survive and participate in subsequent immune responses, significantly limiting the number and quality of B cells returning to the DZ [22].

Fig. 1.

Generation and optimization of tumor-infiltrating B cells in the germinal center. The germinal center microenvironment provides a unique niche for B cells. Through somatic hypermutation and cellular selection, B cells acquire robust antigen recognition capabilities and eventually differentiate into memory B cells or plasma cells, which form the antibody-mediated immune response. The germinal center is divided into dark and light zones. Rapid proliferation and random mutation in the dark zone, greatly expand the genetic diversity of the B cell repertoire. Once cell division ceases, dark zone B cells bearing to the light zone. In the light zone, B cells carrying receptors of relatively high-affinity capture and process additional antigens and undergo positive selection through interactions with T follicular helper cells. Positive selection of germinal center B cells by T follicular helper cells is mediated primarily by the interaction of CD40L on T follicular helper cells with CD40 on GC B cells. Upon re-entry into the dark zone, these B cells enter a new cycle of rapid clonal expansion. Repeated rounds of “selection-proliferation-mutation” in the germinal center microenvironment further refine the antigen recognition capabilities of the B cell population

On this basis, further research has revealed the existence of a significant positive selection mechanism in the germinal center to optimize antibody responses. During this selection process, B cells with higher affinity can more efficiently process and present antigens to Tfh cells more efficiently, thus showing a strong preference for acquiring survival and proliferation signals. CCL22 is a chemokine that, when expressed at elevated levels, attracts cells that express CCL22 receptors such as CCR4, including Tfh cells. Consequently, by upregulating CCL22, high-affinity GC B cells may be more susceptible to attract Tfh cells. This mechanism ensures that higher affinity B cells are preferentially selected and expanded within the GC, inducing the formation of antibodies with high antigen-binding efficiency, thereby enhancing the immune system’s capacity for antigen recognition and elimination [25]. However, not all B lymphocyte chemokines can induce such a mechanistic insight.

It can be confirmed that the successful implementation of this mechanism depends on the critical role of follicular helper T cells. Tfh cells are a distinct subset of CD4⁺ T cells derived directly from naive T cells that can powerfully orchestrate humoral immune responses and have been shown to be widely localized in human tonsils and lymph nodes, as well as other lymphoid tissues. Under the guidance of Tfh cells and other factors, B cells can undergo somatic hypermutation and class switch recombination in the germinal center, changing antibody isotypes and progressively producing high-affinity antibodies. Current research suggests that Tfh cells have a strong symbiotic interdependence with B cells within the germinal center [26]. GC-resident Tfh cells can precisely regulate B cell proliferation and antibody gene rearrangement within the germinal center by secreting IL-21 and promote the maturation of selective B cells into long-lived plasma cells with memory function, providing reliable protection for immune responses that play a role in mounted and sustained antibody responses or memory B cells in the germinal center, thereby participating in humoral immunity. When Tfh recognize these peptide fragments, they send survival signals to these B cells, and regulate B cell activity mainly by activating the CD40L (also known as CD154)—CD40 signaling pathway and secreting locally produced cytokines such as IL-21, IL-4, IL-6, and others. IL-4 is known to promote the accumulation and maintenance of B cells in germinal centers, particularly in the light zone, during Th2-associated immune responses [22]. Other studies have also reported that knockout of either IL-21 or IL-4 results in a reduced B cell response [27].

In addition, studies using a murine model have observed that the differentiation of Tfh cells during the initial stage of dendritic cell (DC) induction is coordinated and regulated by multiple factors, including IL-6, inducible T cell costimulator (ICOS), IL-2, and TCR signaling strength (Fig. 1). In mouse models lacking ICOS, the absence of ICOS affects germinal center formation [28]. Tfh cell differentiation is a sequential developmental process, in which nascent Tfh cells undergo further maturation and commitment to ultimately give rise to fully functional GC-Tfh cells [29]. In mice, early differentiation of Tfh cells requires the involvement and strong induction of IL-6 and ICOSL. Unlike mice, humans require IL-12 to signal through signal transducer and activator of transcription (STAT) to direct IL-21 secretion [30]. Its mission is to proceed to the site of initial contact and pathogenic engagement or inflammation to enhance the immune response by supporting B cells. At the same time, follicular T helper cells highly express the PD-1 molecule [31]. Given the binding of PD-1 to PD-L1 expressed on follicular naive B cells the expression of CXCR5 and the upregulation of CXCR3 on Tfh cells can be suppressed. This could result in a loss of Tfh cell recruitment capacity, which may significantly reduce the abundance of functionally competent immune cells within germinal centers. However, this mechanism may also affect the infiltration of T cells into tumors and inflamed tissues to some extent. Anti-PD-1 therapy can enhance the activity and function of Tfh cells; paradoxically, research using chimeras formed from PD-L1-deficient and wild-type bone marrow has shown that PD-1 can also induce a sustained increase in the stringency of affinity maturation in GC B cells [31, 32].

We have found that multiple factors influence on the differentiation process of Tfh cells. To date, studies using conditional knockout of the Bcl6 gene in CD4⁺ T cells have shown that the absence of Bcl6 results in impaired differentiation of Tfh cells. A hallmark of Tfh cells in mouse and human models is the high expression of SLAM-associated protein (SAP). The transcription factor Bcl6 coincidentally upregulates the expression of SAP in Tfh cells, suggesting that it may influence and regulate the diversification and function of Tfh cells by regulating SAP expression levels [33]. High expression of SAP mediates signaling that promotes sustained adhesion between Tfh cells and B cells, enhances T cells adhesion to low affinity cells, and amplifies TCR signaling, which is critical for subsequent proliferation and differentiation of B cells in the germinal center [34]. Loss of functional SAP protein expression within Tfh cells would weaken the cognate crosstalk between B and T cells, compromising the stability of GCs [35]. On the other hand, disruption of antigen recognition between resting CD4⁺ T cells and antigen-presenting cells (APCs) under the influence of antigens and microenvironmental signals may also suppress the acquisition of Tfh cell properties by CD4⁺ T cell subsets.

In terminal phases, the formation of intact and mature TLS, induced by certain unknown factors, may have a distinct propensity to induce the differentiation of high-affinity antigen-specific memory B cell (Bmem) and plasma cell (PC) compartments, although this occurrence is not frequent. The formation of effective GCs is only possible with the combination of specific cytokines and functional Tfh cells. There is strong and consistent evidence that treatment with immune checkpoint inhibitors (ICIs) can induce clonal expansion of TIL-B [36], but it remains unknown whether ICIs can induce the development of TLS.

People are curious about the complicated proliferation and selection processes of GC B cells and try to explore their uniqueness compared to other phenotypes. Because GC B cells and other types of B cells encounter antigens in somewhat different ways (the synaptic endocytosis patterns of antigens are different), the traction force generated by their BCRs is also different. For example, researchers have placed naive and GC B cells in environments containing antigen clusters and used traction force and force sensor technologies to measure the physical forces exerted by these cells on the antigen clusters. Studies have shown that GC B cells exhibit more stringent and prolonged antigen-binding capabilities that can be targeted for appropriate affinity discrimination [37].

How are antigens presented to B cells in their native form?

Upon entering the body, tumor cells or exogenous antigens typically reach the local draining lymph nodes via the flow of lymphatic fluid. The anatomy of the draining lymph nodes can be subdivided into several regions, including the cortical, medullary, and peripheral regions. Lymphoid follicles represent pivotal sites for B cell activation [38]. Within the draining lymph nodes, antigen delivery and processing are typically facilitated by the actions of plasma cells and memory B cells. The function of lymph nodes is critical in the tumor immune response, especially during antigen recognition and immune initiation, through mechanisms involving multiple cell types and complex intercellular interactions [39]. It is widely acknowledged that dendritic cells, as the predominant antigen-presenting cells (APCs), assume the responsibility of capturing and processing antigens and subsequently delivering them to T and B cells via the major histocompatibility complex (MHC). In this regard, the present study will focus on the pivotal role of naive B cells in antigen presentation.

After antigens enter the lymph nodes, some are “trapped” in the subcapsular sinus, while others can drain into the spinal cord. B cells are capable of producing antibodies and have all the necessary mechanisms for antigen uptake, processing, and presentation. They can promptly summon activated T cells to better respond to foreign invaders, and are therefore classified as professional antigen-presenting cells (APCs) involved in the regulation of immune responses against pathogens. The lymphoid follicle-subcapsular sinus (SCS) interface in the lymph node represents the initial site of soluble antigen uptake by naive B cells (with a half-life of approximately 6 weeks), followed by extracellular deposition of antigens on follicular dendritic cells (FDCs), where antigens shuttle back and forth. At this stage, antigens binding to complement component 3d (C3d) are readily captured by CR4 on the surface of macrophages or dendritic cells [40]. C3d is a fragment of complement component C3 that can bind to complement receptor 1/2 (CR1/2) on B cells, significantly enhancing the signaling of B cell receptors (BCRs). By binding to antigens and tagging them, C3d facilitates their recognition by B cells, thereby serving as a bridge between the innate immune system (via complement) and the adaptive immune system (B cells). Following uptake of antigens flowing toward the medulla by dendritic cells or macrophages within the medulla, antigens located in the subcapsular sinus macrophages and medullary sinus areas come into contact and are recognized by high-affinity naive B cells expressing abundant complement receptors (CR1/2) (Fig. 1A of ESM). Naive B cells then deliver these antigens to FDCs (for further processing or storage). Visualization tracking revealed that the migration rate of B cells decreased, and their retention time increased at this time [41]. After concentrated particulate antigens are phagocytosed by macrophages in the SCS, these antigens undergo in situ inversion of their surface antigens. Subsequently, the antigens are either indirectly transported by naive B cells or directly taken up by FDCs for intensive processing [42, 43] (Fig. 1B of ESM).

Once the density of “unloaded” naive B cells in the lymphoid follicles decreases accordingly, FDCs assume the role of initial antigen presentation [41]. Particulate antigens are deposited directly on the surface of FDCs, after which the interaction between very late antigen-4 (VLA-4) and vascular cell adhesion molecule-1 (VCAM-1) synergistically enhances the affinity of B cells for antigen contact in conjunction with the BCR. This activation increases the potential of B cells to recognize antigens [44].

In the antigen presentation process mentioned above, this process appears to be further enhanced by the achievements of FDCs. FDCs have long, interlocking dendritic processes that form a dense network, providing an ideal microenvironment for B cells to promote their activation, proliferation, and differentiation. Specific functions of FDCs in the germinal center include: (1) release cytokines and chemokines, such as chemokine C-X-C motif ligand 13 (CXCL13), to maintain the density of germinal center B cells (GC B cells) within the germinal center region; (2) express the CD40 ligand, interact with GC B cells, secrete cytokines, such as BAFF, and promote their survival and proliferation; (3) FDCs are capable of acquiring native, unprocessed antigens in the form of immune repertoires from antigen-presenting cells. FDCs have the propensity to retain these native antigens on the cell surface for over a year, allowing for long-term storage of antigens. During this time, new antigens replace old ones, significantly extending their half-life. As a result, antibody levels remain relatively stable even in the absence of new exogenous immunogenic stimuli [45]; (4) Upon immune activation, remodeling of the FDC network not only increases contact surfaces with B cells, but also facilitates highly efficient antigen display and transfer, which benefits the GC response. By virtue of this dynamic remodeling ability, FDCs play a pivotal role in reshaping the lymph node follicular microenvironment that is conducive to immune responses. Thus, by storing antigens and spontaneously secreting B cell-associated factors, FDCs are important in directing humoral immune responses, regulating immune memory formation and long-term immune defense.

Tumor-Infiltrating B cells: prognostic impact determined by immunoglobulin subtypes

The number and activity of tumor-infiltrating B cells may be coordinated by multiple factors in the tumor microenvironment, including changes in tumor antigen expression on tumor cells, tumor-associated angiogenesis, peripheral conditions, and the abundant amounts of antibodies secreted by B cell subsets, etc. A recent study has shown that high plasma cell infiltration in patient biopsy samples with plasma cells (PCs) in the primary tumor is an important favorable prognostic factor for breast and lung cancer patients [46]. It is well known that various cytokine regulatory networks form a multilayered control system of B cell responses, precisely regulating the expression profiles of different Ig isotypes, such as IgA, IgE, IgM, IgD, IgG1, IgG2a, IgG2b, and IgG3. In this section, we will introduce two of these types: IgG1 and IgA. The prognostic impact of PC depends on the Ig subtype. In certain diseases, the prognostic implications of IgG and IgA are diametrically opposed [47–49]. For example, precancerous livers show an accumulation of IgA⁺ plasma cells. Functional screening has shown that IgA⁺ cells in the liver mediate immune suppression. IgA⁺ PCs are therefore considered an indicator of poor prognosis [50].

IgG1 antibodies can drive a variety of immune responses through antibody-dependent cellular cytotoxicity (ADCC) and phagocytosis, as well as complement activation. Researchers unexpectedly found that the addition of T cell line outgrowths to lipopolysaccharide (LPS)-treated cultures could induce an increased IgG1 response [51]; this soluble substance was later identified as interleukin 4 [52]. Not only is IL-4 a key cytokine in promoting B cell class switching to IgE, but the continued presence of IL-4 is also critical for the durability and potency of the response (Fig. 2A).

Fig. 2.

The phenotypic and functional characteristics of tumor-infiltrating B cells determine their overall impact on tumor progression or suppression. A IgG1 antibodies are often associated with anti-tumor B cell activity. In addition, dendritic cells and macrophages can capture IgG-bound tumor antigens and present these antigen-derived peptides to T cells. B cells are involved in the presentation of tumor-derived antigens to CD4⁺ and CD8⁺ T cells. In addition, studies have shown that B cells in the tumor microenvironment are capable of secreting cytokines that promote cytotoxic T cell responses, such as interferon-γ (IFN-γ) and interleukin-12 (IL-12). While both IgG1 and IgG3 isotypes are capable of mediating antibody-dependent cellular cytotoxicity (ADCC), phagocytosis, complement activation, and antigen cross-presentation through Fc receptor binding, IgG3 exhibits potent ADCC only in the presence of IgG1. B B cells induced the conversion of Tregs cells to IgA subtypes by promoting their production of transforming growth factor-β (TGF-β), forming a B cell-Tregs positive feedback regulatory pathway that promotes tumor immune evasion

However, when the body’s tolerance is reduced and/or immune responses become dysregulated, the pro-inflammatory activity of IL-6 can easily over-activate the immune system, leading to autoimmune diseases. Previous evidence has implicated IL-6 in rheumatoid arthritis, lupus erythematosus, multiple sclerosis, kidney disease, type 2 diabetes [53] and many other diseases. Theoretically, IL-5 can promote the differentiation of B cells into IgA⁺ cells, thereby increasing the secretion of IgA. It has been observed that plasma-like IgA⁺ cells are present in the fibrous tissue of both humans and rodents with live cancer [54], and precancerous livers show aggregation of IgA⁺ plasma cells. Functional screening has shown that IgA⁺ cells in the liver mediate immunosuppression [55] (Fig. 2B).

In addition, the tertiary lymphoid structures (TLS) microenvironment provides a niche that supports B cell maturation and bursts of antibody production, promoting systemic anti-tumor immune responses. PCs that differentiate and mature in the TLS can secrete polyclonal antibodies against tumor cell-specific antigens. Antibodies secreted by TIL PCs could directly mediate apoptosis and lysis of tumor cells. Within the TLS, PCs migrate long distances along the tracks of chemokine (C-X-C motif) ligand 12 (CXCL12)-positive fibroblastic reticular cells (FRCs) in the tumor bed, and disseminate into the depth of the tumor [56]. Furthermore, some researchers claim that the combination of CD8⁺ TILs in TLS with high levels of plasma cell infiltration can produce even more amazing anti-tumor effects [11].

In recent comprehensive analyses of various cancers, researchers have delineated two distinct pathways by which tumor-infiltrating B cells differentiate into antibody-secreting cells: the germinal center (GC) pathway and the extrafollicular (EF) pathway. Tumors dominated by the EF pathway tend to have poorer prognoses and disrupted immune microenvironments, which are associated with the accumulation of atypical memory B cells in immature tertiary lymphoid structures and the influence of glutamine metabolism [57].

The diversity and potential of regulatory B cells in immune regulation

An IL-10-producing B cell subset, termed Bregs, was discovered 20 years ago in murine and human cancers [58] (Fig. 2 of ESM). By modulating the balance between pro- and anti-inflammatory cytokines, Breg cells may help maintain immune homeostasis in the tumor microenvironment and play a critical role in regulating cellular homeostasis [59] in autoimmunity [60], transplantation, and physiological conditions. Breg cells are thought to be derived from mature B cells, as evidenced by their expression of common B cell surface markers such as CD19 and CD20, but different phenotypes of Breg cells express unique markers for specific identification. The phenotype of Breg cells is diverse and complex, and they comprise less than 1% of the peripheral blood mononuclear cells (PBMCs) in the human body. Currently, there is no definitive research demonstrating the existence of specific transcription factors unique to Breg cells [12], and some researchers believe that the study of Breg cells will continue to evolve as the immune environment of disease changes.

Therefore, given the specificity of Breg cell subsets, in order to use Breg cells to address tumor-related issues, we should try to identify a Breg cell-specific transcription factor [61], such as recombinant forkhead box P3 (Foxp3) in regulatory T cells (Treg cells) [62]. However, some researchers have raised objections regarding the diversity of Breg cell subpopulations: Since they were unable to identify unique transcription factors and were largely dependent on their environment, they questioned their lineage specificity and advocated the “inflammatory response” view as more reasonable [63].

Subsequent research has shown that the immunoregulatory mechanisms of Breg cells are more diverse and complex than originally thought and do not rely solely on the secretion of IL-10. It was recognized that the Breg subpopulation could secrete IL-35, which was first described in 2012 [64]. Evidence suggests that LPS-stimulated B cells simultaneously produce IL-35 and IL-10 at a rate of approximately 18.9% [65].

Here, we present three types of regulatory B cell phenotypes. Firstly, B10 cells are a subset of Breg cells capable of producing IL-10 and represent a small fraction of the CD24^hiCD27⁺ subset. When monocytes are co-cultured with activated B10 cells, tumor necrosis factor-alpha (TNF-α) production by monocytes is significantly reduced [66]. During inflammation, researchers have observed a robust derivative event of B10 cells in humans [67]. These cells are present in solid tumor tissues, including invasive breast cancer, hepatocellular carcinoma, gastric cancer, and head and neck squamous cell carcinoma. Furthermore, researchers used a pancreatic cancer model consisting of green fluorescent protein-expressing pancreatic ductal epithelial cells with oncogenic KRas (GFP-KRas^G12D-PDEC) to confirm that CD1d^hiCD5⁺ B cells (human B10 cells) support pancreatic tumor development at both early and late stages by expressing IL-35, thereby promoting tumor progression [68]. Of note, the aryl hydrocarbon receptor (AhR) is a sensor of the response to environmental toxicity, not only that, its activation can inhibit the differentiation of germinal center (GC) B cells and plasma cells to maintain the function of regulatory B cells (Bregs). AhR enhances IL-10 secretion in regulatory B cells (Breg) by silencing pro-inflammatory transcriptional programs, thereby substantially coordinating the immunosuppressive functions of IL-10⁺CD19⁺CD24^hiCD27^hi Breg [69]. The aryl hydrocarbon receptor (AhR) transcription factor on B10 cells regulates regulatory B cell homeostasis through metabolites such as butyrate. Therefore, targeting the AhR to limit the trend of tumor immune suppression warrants further investigation [70].

Secondly, we found that granzyme B⁺ (GZMB⁺) B cells are abundant in the peripheral blood of healthy donors, whereas these cells show a higher propensity for apoptosis in autoimmune patients[71]. IL-21 has been identified as a key cytokine that promotes the differentiation of B cells into GZMB⁺ B cells. From the limited data available, it appears that in addition to secreting IL-10 to exert inhibitory effects, it can also upregulate the PD-1 signaling pathway. IL-21 can induce the production of a substantial reservoir of GZMB in humans, and the proximity of these B cells to IL-21-secreting regulatory T cells enhance this effect and improves tissue immune tolerance [72, 73]. Research has shown that GZMB⁺ B cells are found in the microenvironment of various tumor types, including but not limited to breast cancer, ovarian cancer, cervical cancer, colorectal cancer, and prostate cancer, where they participate in suppressing local immune responses and influence tumor growth rates and metastatic potential [74].

Third, transitional B cells have garnered significant attention due to their high levels of the anti-inflammatory cytokine IL-10 and relatively low levels of the pro-inflammatory cytokine TNF-α. Research suggests that these cells not only inhibit the phenotypic switch of immune cells to pro-inflammatory profiles and restrict the activation of anti-tumor effector cells, such as CD8⁺ T cells, but also interact with immunosuppressive cells like regulatory T cells (Tregs), thereby enhancing immune tolerance in the host. In addition, as previously mentioned, regulatory T follicular helper (TFH) cells play an indispensable regulatory role in the formation of germinal centers and the enhancement of immune responses. Regulatory B cells effectively inhibit Tfh1 cell activity by secreting IL-10 and TGF-β, inducing inhibitory markers on Tfh1 cells, limiting germinal center responses, and enhancing their own function in an immunosuppressive tumor microenvironment. This, in turn, leads to a weakened anti-tumor immune response and tumor evasion. Transitional B cells were previously defined as CD19⁺CD24^hiCD38^hi and are now characterized as CD19⁺CD24^hiCD38^hiCD39⁻. This important advance not only improves the sorting efficiency of transitional B cells, but also opens new avenues for further exploration of their role in tumor immunology [75]. Finally, the study demonstrated that the frequency distribution of transitional B cell subsets is significantly elevated and exhibits high specificity in breast cancer patients.

Furthermore, transplantation tolerance experiments have shown that Breg cells can promote the development of Treg cells by producing transforming growth factor-beta (TGF-β), thereby promoting graft survival. In the transplantation setting, adaptive transfer of highly enriched T cell immunoglobulin domain and mucin domain protein-1 (TIM-1⁺) Breg cells expressing IL-4 and IL-10 can directly establish graft tolerance, and prolong allograft survival [76–78]. In fact, the presence of TIM-1 results in an average 40% increase in IL-10 expression by Tregs [79]. The above findings demonstrate that the interaction between TIM-1 and TGF-β is synergistic, and their combination may achieve a more pronounced immunosuppressive effect.

In organ transplantation tolerance experiments, both regulatory B cells and regulatory T cells have been found together. Breg cells are thought to have a tendency to suppress Treg cells [67], and this mechanism may produce a suppressive inflammatory response in vivo. Complementary experiments performed several years later showed that mice deficient in B cell-specific IL-10 also exhibited features of Treg cell deficiency [80], directly validating the hypothesis of a homologous interaction between the two. It is worth noting that studies have shown that protein kinase inhibitors or agonists can further alleviate the number of regulatory B cells both in vivo and in vitro by affecting the chemotaxis of cell signaling pathways [81]. This mechanism appears to be partially involved in T cell and anti-PD-1 immunotherapy [82]. All of the above examples demonstrate the diversity and comprehensiveness of Breg cell types and highlight their ability to exert broad suppressive effects on immune lineage cells.

Investigating the dynamic change and prognostic value of B cells in clinical immunotherapy

Numerous reports have previously suggested that cancer cells escape T cell-mediated cytotoxicity through tumor immune interactions (tumor immune escape) [83]. We aim to challenge the conventional notion that T cells are the terminal effectors of cancer immunity, and demonstrate surprising efficacy by flexibly exploiting a number of functions of tumor-infiltrating B cells (TIL-B cells) in anti-tumor immunity (Table 1).

Table 1.

Roles and mechanisms of tumor-infiltrating B cell subtypes in tumor microenvironment

Cell types	Subset	Mechanism and function	Tumor type	References
GC B cell	CD19+CD20+BCL6+ GC B cell	Generation of high-affinity antibodies and tumor-specific antibodies through somatic hypermutation (SHM) screening	Diffuse large B cell lymphoma (DLBCL)	Shimin Hu et al. 2013 [110]
Naive B cell	IgD+CD27− naive B cell	Enhance antigen presentation efficiency while differentiating and selecting for memory B cells	Hypogammaglobulinemia	Erika Irie et al. 2010 [111]
Plasma cell	IgG1 B cell/Plasmablast/Plasma Cell (CD20+/CD19+)	Modulation, complement fixation, antibody-dependent cellular cytotoxicity (ADCC), and antibody-mediated phagocytosis kill tumor cells and drive cytotoxic T cell responses	Non-small cell lung cancer (NSCLC)	Zhenzhen Hui et al. 2022 [87]
	IgA B cell/Plasmablast/Plasma Cell (CD20−/CD19+)	Induce CD4+ T cells to differentiate into regulatory T cells (Tregs) for immunosuppression, inhibiting TH1 cells, CD8+ T cells, and natural killer (NK) cells	Non-small cell lung cancer (NSCLC)
Regulatory B cells	B10 cell	Inhibits Th17 cells, CD4+ T cells, macrophages and DCs, expands Tregs	Advanced colorectal cancer	Jiao Yang et al. 2015 [88]
	transitional B cell	Upregulate the expression of PD-L1, affecting T cell activity and promoting the proliferation of regulatory T cells	Severe asthma	Marek Lommatzsch et al. 2023 [112]
	TIM-1 B cell	Expand Tregs and reduce CD4+ T cells	B16F10 melanoma	Qing Ding et al. 2011 [78]; Lloyd Bod et al. 2023 [113]
	GrB+ B cell	Depend on granzyme B (GrB) to degrade T cell receptors (TCR) to suppress T cell proliferation	Ovarian carcinoma	Stefanie Lindner et al. 2013 [74]
	IL-35+ B cell (CD38+CD19+B cells)	Derived IL-10 inhibits the cytotoxicity of natural killer (NK) cells, while the expression of IL-35 affects the gp130/STAT3 pathway, further suppressing the functions of NK cells and CD8+ T cells	Pancreatic ductal adenocarcinoma	Yuliya Pylayeva-Gupta et al. 2016 [68] Heng Zhu et al. 2024 [114]

GC B cell, Germinal center B cell; TIM-1, T cell Immunoglobulin and Mucin-domain containing-1

To begin with, we will focus on the role of B cells in immunotherapy and their dynamic effects. In the clinical treatment of tumors, chemotherapy disrupts the balance between depletion and proliferation of tumor-infiltrating B cells and alters the distribution of B cell subtypes. Several studies have shown that different chemotherapeutic modalities can lead to massive depletion of B cells at the macroscopic level (e.g., GC B cells, memory B cells) [84], while unexpectedly retaining the potential to activate the immune system in some of the surviving B cells.

Rituximab treatment increases the proportion of naive B cells. It promotes their activation and differentiation into memory B cells by clonal selection [85]. In addition, the inhibition of BCR signaling is attenuated. A similar increase in naive B cells has now been observed in hematologic malignancies and esophageal squamous cell carcinoma [86]. According to our description of naive B cells above, an increase in their concentration means that they are able to fulfil their antigen-presenting role more efficiently. Platinum-based chemotherapy, in turn, led to the differentiation and selection of naive B cells into memory B cells, which was able to mitigate the significant decrease in the percentage of memory B cells during chemotherapy and maintain it below baseline.

Although the number of plasma cells decreases during chemotherapy, the suppression of immunoglobulin production is much less than the reduction in cell numbers. A study published in July 2022 characterized 29 different cell populations from responding and non-responding patients with stage IIIA non-small cell lung cancer (NSCLC) treated with neoadjuvant chemotherapy combined with pembrolizumab. The major pathological response (MPR) assessment showed that the number of IgG1 and IgG3 subtype plasma cells tended to increase, while the number of IgA subtype plasma cells decreased. This adjustment in antibody subclasses enhances the body’s anti-tumor immune capabilities [87].

Platinum-treated Breg results in a high frequency of apoptosis. This improves the prognosis of patients [88]. Our hypothesis is to improve the efficacy of chemotherapy by targeting regulatory B cells that express high levels of receptors. Targeting the AhR may be a promising option as it could significantly reduce the immunosuppressive effects of IL-10 on immune cells, thereby increasing the overall efficacy of chemotherapy.

Certain types of chemotherapy have been shown to increase the density and phenotype of tumor-infiltrating B cells (TIL-B) in human cancers and mouse tumor models. However, the dosage and regimen of chemotherapy must be tightly controlled in the clinical setting, making the establishment of metrics for patient heterogeneity critical. In addition, several studies in human cancers have shown a strong positive correlation between TIL-B and clinical response to CTLA-4 and PD-1 targeted immune checkpoint inhibitors. Although no universal clinical phenomenon has been observed, this correlation is unlikely to be coincidental. However, a complicating factor, as mentioned in the discussion of germinal center (GC) B cells, is that PD-1 can induce affinity maturation in GC B cells. This suggests that immune checkpoint inhibitors could potentially lead to adverse effects. Careful consideration of the associated risks and benefits is therefore required. As a result, the complex characteristics of TIL-B make it difficult to implement strategies to broadly enhance their activity, and there may even be a risk of autoimmune responses.

Next, we use tumor-infiltrating B cells or tertiary lymphoid structures as markers to describe cases that have not received anticancer treatment and cases that have received ICB therapy separately. On the one hand, researchers examined tissue sections from surgical cases of hepatocellular carcinoma that had not received preoperative anticancer treatment. They found that nearly one tenth of the lesions were accompanied by significant lymphocyte infiltration, which was localized to tumor areas with angiogenesis and had a non-diffuse distribution [89]. On this basis, researchers compared the proportion of CD20-positive cells at the tumor-non-tumor interface between two groups of patients and assessed the risk of recurrence-free survival (RFS) in a heterogeneous patient cohort. It was observed that CD20-positive B lymphocytes are the predominant cell population at the tumor-non-tumor interface and have the ability to penetrate the tumor capsule [90]. In addition, data analysis revealed that the expression level of CD20 was strongly associated with disease progression. Patients with high CD20 expression had a significantly lower risk of relapse and relatively longer overall survival compared to the low expression group [91]. At the same time, postoperative survival rates vary with age and are more significant in elderly patients aged 58 years and older [92]. This infiltrate is evident at an early stage, indicating the development of CD20-positive B lymphocytes from a mild inflammatory phase [93]. However, this perspective seems to be limited to the tumor-non-tumor interface, and the same conclusion cannot be drawn for areas far from the inflammatory response, and some studies have therefore named them margin-infiltrating CD20⁺ B cells [90]. In 2023, researchers explained the specific localization of tertiary lymphoid structures (TLS) in tumors: TLS are often localized in the peritumoral stroma with the highest immune activity and exhibit high sensitivity, allowing better sampling of mutable antigens or low concentrations of antigens [94].

On the other hand, studies investigating the impact of B cells and tertiary lymphoid structures (TLS) on the efficacy of ICB therapy in various cancers, including melanoma [95], sarcoma, lung adenocarcinoma [96], pancreatic cancer [97], hepatocellular carcinoma, and urothelial cancer [98], have consistently concluded that the presence of TLS is associated with a favorable prognosis [99]. It facilitates the induction of novel anti-tumor T cells responses during ICB treatment. Most bioinformatic analyses suggest that TIL-B cells have a positive or neutral impact on prognosis, with the specific results often dependent on the specific B cell signature used. In cases where TIL-B cells are associated with poor prognosis, the B cell response may be skewed toward a regulatory phenotype. We emphasize that B cells exert anti-tumor and pro-tumor effects in both directions.

Subsequently, it has been proposed to investigate whether early lesions can induce the formation of tertiary lymphoid structures [100]. Research has shown that in 24% of cases, there is an increase in the density of T cells, B cells, and dendritic cells, but the TLS is still in an immature state. The infiltration of these cells is generally not conducive to establishing a microenvironment that promotes tumor growth, and the density of TLS is consistent with the reduction in tumor cell viability. Studies have confirmed that intertumoral TLS in patients is an accurate prognostic marker for early recurrence and is associated with a reduced risk of early tumor recurrence [101]. This finding complements the previous observation of increased intertumoral CD20⁺ B cells [102]. The abundance of TLS in tumors and peritumoral tissues has opposite effects on prognosis, demonstrating dual effects on cancer prognosis and immunity [103]. Their mechanisms and differences will also be rationally analyzed in a large number of clinical samples. In conclusion, activation of TLS in non-tumor liver tissues may play an important role in establishing the inflammatory milieu conducive to tumor development [104].

Conclusion

Herein, we systematically reviewed the studies on the prognostic significance of tumor-infiltrating B cells (TIL-B) and tertiary lymphoid structures (TLS) in human cancers. While most studies suggest a positive or neutral correlation between high levels of TIL-B and GC-TLS and a favorable prognosis, the exact mechanisms by which they influence prognosis in different cancer scenarios remain to be fully elucidated [105]. Furthermore, specific deletion of B cell-related genes in a murine tumor model resulted in a reduction of T cell activation and a significant increase in the number and size of metastatic tumors [106, 107]. This highlights the critical role of the “high CD20, high CD8” phenotype in orchestrating effective cellular immunity against cancer [107–109]. The aim of this study was to propose the hypothesis that differences in the maturity and proportion of immune cells in the tertiary lymphoid structures may be a key factor contributing to inconsistent postoperative recovery and ineffective immunotherapy in patients with solid tumors. Therefore, detailed assessment of individual immune profiles may help to select and optimize personalized treatment strategies. Future studies should consider specific histologic and molecular subtypes of relevant cancers from a clinicopathologic perspective and use multivariate analyses to clarify their epidemiologic and prognostic significance. With a more complete understanding of the mechanisms, it may be possible to induce the formation of mature TLS through the use of specific cytokines or chemokines in combination with immune checkpoint blockade therapies. This approach could lead to broader and more durable tumor suppression by generating synergistic effects.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

ADCC

Antibody-dependent cellular cytotoxicity

APC

Antigen-presenting cell

Bmem

Memory B cell

Breg

Regulatory B cell

CR

Complement receptors

DC

Dendritic cell

DZ

Dark zone

FDC

Follicular dendritic cell

FRC

Fibroblastic reticular cell

GC

Germinal center

GrB/GZMB

Granzyme B

IARC

International Agency for Research on Cancer

ICB

Immune checkpoint blockade

ICIs

Immune checkpoint inhibitors

ICOS

Inducible T cell costimulator

LPS

Lipopolysaccharide

LZ

Light zone

PBMC

Peripheral blood mononuclear cell

RFS

Recurrence-free survival

SAP

SLAM-associated protein

SCS

Subcapsular sinus

SHM

Somatic hypermutation

TCR

T cell receptor

Tfh

T follicular helper

TIL

Tumor-infiltrating lymphocyte

TIL-B

Tumor-infiltrating B cell

TLS

Tertiary lymphoid structures

TME

Tumor microenvironment

WHO

World Health Organization

Author contributions

S Song and H Zhang have access to all the data in this study and take responsibility for the integrity and accuracy of the data analyses. S Song and H Zhang contributed to study concept and design. S Song and C Wang drafted the manuscript, and X Zhou, Y Han, and Y Chen supervised the study and provided modification suggestions. All authors have read and approved the final manuscript.

Funding

This work was supported by the National Natural Science Foundation of China (81401988, 32170915 and 82172931), China Postdoctoral Science Foundation (2019M661907), Jiangsu Postdoctoral Science Foundation (2019K159, 2019Z153), General Project of Jiangsu Provincial Health Committee (H2023136), General Project of Nantong Municipal Health Committee (MS2023013), Jiangsu Provincial Research Hospital (YJXYY202204-YSB28), Beijing Municipal Public Welfare Development and Reform Pilot Project for Medical Research Institutes (JYY2023-14, JYY2023-15), and College student innovation program (202210304128Y and 2023103041055).

Data availability

No datasets were generated or analyzed during the current study.

Declarations

Conflict of interest

The authors declare no competing interests.

Consent for publication

All authors agree on the submission of the manuscript.