Journal of Medical Oncology and Therapeutics

Journal Banner

Harnessing B Cells for Cancer Immunotherapy

Yanqiu Li, Shala Yuan and Jiusheng Deng*

Department of Hematology and Medical Oncology, Winship Cancer Institute, Emory University, Atlanta, USA.

*Corresponding Author:
Jiusheng Deng
Department of Hematology and Medical Oncology
Winship Cancer Institute, Emory University
1365B Clifton Road, Atlanta, GA, USA
Tel: 4047781763
E-mail: [email protected]

Accepted Date: July 25, 2016

Visit for more related articles at Journal of Medical Oncology and Therapeutics


B cells are a heterogeneous population in immune defense system with multidirectional functions. In cancer patients, B cell infiltrates are associated with a significant increase of overall survival. We have recently developed a GM-CSF and IL-4 derived fusion cytokine named GIFT4, which has the capability to prime naïve B cells into anti-tumor immune effector cells. Herein, we overview current research findings on B cell anti-tumor functions and B cell-based approaches for cancer immunotherapy. We predict that GIFT4- augmented B cells as a potent cellular therapeutic could provide a new approach for cancer immunotherapy.


B cells, Tumor microenvironment, GIFT4 fusokine, Cancer immunotherapy.


B Cell Functions

B cells are one of the two adaptive arms partnering with T cells in immune defense system against infections by virus, bacteria, fungi and parasites [1]. Originated from hematopoietic stem cells in bone marrow, B cells contain multiple subsets including antibody-secreting cells, antigen-presenting cells (APC), innate B effector cells and regulatory B cells [2]. As a heterogeneous population, B cells possess multidirectional immune functions. For instance, B cells can produce antigen-specific antibodies in response to infectious pathogens or sterile self-antigens [3]. B cells can also present pathogen-derived antigens to T cells during infections [4]. B effector cells can further produce a variety of immune-stimulatory cytokines such as IL-1, IL-6, IL-12 [5], Granulocyte-macrophage colonystimulating factor (GM-CSF) [6], augment immune response against infections or promote inflammation in autoimmune diseases [6,7]. In contrast, regulatory B cells secrete immune-suppressive cytokines including IL-10 and transforming growth factor-β (TGF-β) to attenuate proinflammatory immune response [8]. Emerging evidences show that B cells also have anti-tumor function [9,10]. In preclinical animal model, B cells are required for the successful combined antibody-immunotherapy against murine mesotheliomas [11]. In patients with malignancies, B cells are also found to correlate with a significant increase of overall survival, and higher number of B cell infiltrates lead to better prognosis [12]. However, B cells in particular regulatory B cells can also act as immunesuppressive cells and facilitate tumor immune escape [13,14]. The dual functional faces of B cells on tumors are likely due to the different B cell subpopulations, which have distinguished phenotypes and secretomes that either inhibit tumor growth or facilitate malignancy [9].

B Cells in Tumor Microenvironment

B cells as well as T cells, natural killer cells, monocytes and other immune cells can infiltrate into tumor microenvironment, distributing from the tumor margin to the tumor core. In patients with tongue squamous cell carcinoma, infiltrated B cells are commonly found in the carcinoma stroma with tumor-suppressive effect [12]. In pancreatic ductal adenocarcinoma, human B cells reside in tertiary lymphoid tissue with two distinct infiltrating patterns: scattered or organized [15,16]. High density of organized infiltrating B cells predicts longer survival for patients; highlighting B cells are essential effector cells in the tumor microenvironment of human pancreatic ductal adenocarcinoma [16]. In bladder cancer, human CD20+ B cells preferentially migrate into the lamina propria area, and have positive correlation with T cell infiltration [17]. Further analyses demonstrated that B cell infiltrates have no link with Foxp3 positive regulatory T cells in tumor microenvironment [17]. Moreover, tertiary lymphoid structures with aggregating B cells are associated with lung cancer prognosis [18]. In patients with gastric cancer, B cells abundantly infiltrate and aggregate in the gastric cancer stromal microenvironment, accompanied with infiltrated T-bet+ T cells to form a tertiary lymphoid structure surrounding the tumor [19]. Tumor-associated B cells in gastric cancer microenvironment are proliferating and express Ki67. Importantly, infiltrated B cell number is positively linked to relapse-free survival, and B-cell gene expression is significantly connected with improved outcome [19]. It is found that tumor-infiltrating B cells have beneficial effects on prognosis in patients with tongue squamous cell carcinoma [12], pancreatic adenocarcinoma [16], gastric cancer [19], cutaneous melanoma [20,21], breast cancer [22], ovarian cancer [23] and colorectal cancer [24]. However, the mechanisms by which B cells accumulate in the tumor microenvironment and result in better prognosis are not fully understood. One explanation is that tumor-infiltrating B cells express antigen-presentation molecules and function as professional APC to orchestrate T cell-mediated anticancer immunity [25,26]. Another reason could be that infiltrated B cells have potent capacity to produce antitumor antigen-specific antibodies, since CD138+ and immunoglobulin kappa C-positive plasma cells have positive impact on anti-tumor immunity and are related to favorable prognosis in cancer patients [24,27]. MUC1 (The polymorphic epithelial mucin) is one of the most specific tumor-associated antigens in human cancers [24,28]. Anti-MUC1 IgG antibodies but not IgM in patients are significantly related to better prognosis [22]. Consistently, high density of plasma cells was found surrounding the tertiary lymphoid structures and correlated to T cell cytotoxicity [29]. Infiltrating B cells can also undergo somatic mutation, clonal expansion, intraclonal variation and isotype switching, eliciting humoral immunity against tumors [30,31]. Collectively, the preclinical and clinical investigations strongly support the notion that B cell infiltrates in the tumor microenvironment not only serve as a valuable predictive biomarker, but also play a profound protective role in anti-tumor immunity [32-35].

Cancer Immunotherapy

During the last decade, great progress has been made on cancer immunotherapy including dendritic cell-based cell immunotherapy [36], chimeric antigen receptor (CAR)-T cell [37,38] and immune checkpoint blockade including CTLA-4 (Cytotoxic T-lymphocyte-associated protein 4) or PD-1 (Programmed cell death protein 1)/ PD-L1 (Programmed death-ligand 1) inhibitors [39-41]. Dendritic cells as the most professional APC possess the capability to orchestrate innate and adaptive cellular and humoral immune responses against cancer cells. Cancerantigen loaded or bioengineered dendritic cells that expressing tumor antigens have been utilized as cancer vaccines for cancer patients [42]. However dendritic cells as a tumor vaccine in clinical trials are not as effective as in preclinical animal tumor models, with the limitation of high-cost, small number and short life (2-3 days after maturation) of dendritic cells generated from peripheral blood monocytes. CAR-T cells have been successfully used to treat B-cell malignancies by targeting CD19, CD20, CD22, CD30, CD33, CD123, CD133, CD138, ROR1, κ light chain and B-cell maturation antigen [43]. The killing of normal B cells besides malignant B cells by CAR-T cells and its serious treatment-related toxicities remains a challenge [44]. Current clinical trials reveal that CAR-T therapy have very limited efficacy on nonhematological solid tumors. Expression of regulatory molecules such as CTLA-4 and PD-1 on cytotoxic T cells has been shown to suppress the anti-tumor functions of T cells. Thus immune checkpoint blockade using antagonistic antibodies against the negative regulators can overcome cancer immune resistance and demonstrates promising therapeutic efficacy [45-47]. However, clinical trials showed that only partial cancer patients respond to immune checkpoint blockade [48,49]. B cells have multiple functions as antibody-producing cells; antigen-presenting cells, immune effector cells, and are required for adaptive T cell immune responses against tumors [50]. B cells also have an advantage to be easily expanded ex vivo in comparison with dendritic cells. Moreover, activated B cells can effectively present tumor lysate, antigen peptide or antigen cDNA and induce antigen-specific T cell immunoreaction against tumors [51]. Thus, B cells represent a promising approach for cancer immunotherapy, complementing the use of dendritic cells.

B Cell Based Approaches for Cancer Immunotherapy

B cells have been widely explored as a cellular adjuvant for cancer immunotherapy due to its immune-stimulatory activities. As antigen-presenting cells, B cells express CD40 and ligation with CD40 ligand on B cells robustly enhances the expression of co-stimulatory molecules CD80 and CD86 [52]. Consequently, CD40-activated B cells have potent capability to promote naïve and memory T activation and expansion and induce cytotoxic T cells immunity [53]. When pulsed with a melanoma antigen, CD40-activated B cells efficiently propel the generation of melanoma-specific T cells in vitro [54]. CD40-activated B cells also express adhesion molecules and chemokine receptors facilitating the cells to migrate into the secondary lymphoid organs, attract and interact with antigen-specific T cells [52-55]. CD40-activated B cells also function similarly to plasma cells and produce IgG [52]. In vivo, CD40-activated B cells have protective effect on various tumor models [56,57], with little toxicity to the mice [56]. Alternatively, CD40-ligated B cells loaded with tumorspecific RNA as a cancer vaccine induce tumor-specific cytotoxic T cell immune response, inhibit the growth of non-Hodgkin’s lymphoma and improve overall survival in preclinical animal model [57]. It is interesting that leukemia B cells activated by CD40 ligation are also functionally similar to antigen-presenting cells and induce both IFN-γ+ CD4 and cytotoxic CD8 T cell proliferating and expansion [58]. Those data together inform that CD40-activated B cells have the potential to serve as a potent cellular agent for cancer immunotherapy.

Tumor-infiltrated B cells provide another approach for B cell cancer immunotherapy. B cells infiltrated into tumor stroma function as both antigen-presenting cells and tumor antigen-specific antibody-producing cells, and play essential roles in anti-tumor immunity [34,59]. A Epstein-Barr virus immortalization in vitro assay demonstrates that primary colorectal carcinoma harbor infiltrated B cells that are consistent of CD23+CD80+ activated antigen-presenting cells and IgG-secreting cells. Those infiltrated B cells not only produce functional carcinoma-specific antibodies [59], but are also associated with cytolytic T cell response and superior prognosis in cancers [21,22,24,29]. Adoptive transfer of tumor-derived B cells further promotes anti-tumor T cell immunity and leads to tumor regression in preclinical breast cancer and pulmonary metastatic tumor animal models [60,61]. The anti-tumor property of tumor-primed B cells suggests that ex vivo expanded tumor-primed B cells could be utilized as potent T helper cells for cancer immunotherapy. B cells loaded with tumor-derived autophagosomes have the ability to present tumor-specific antigens selectively captured by autophagosomes and induce robust anti-tumor T cell response as well as antibody-mediated humoral response [62]. Administration of tumor-antigen loaded B cells as a vaccine further prevents the growth of tumors in mice [62], indicating that B cells activated by tumorderived autophagosomes represent a new strategy for cancer immunotherapy.

Recently, we have developed an immune-stimulatory fusion cytokine (Fusokine) named GIFT4 (Figure 1), which is a granulocyte macrophage colony-stimulating factor (GM-CSF) and common γ-chain Interleukins 4 (IL- 4) fusion transgene [63]. In comparison with its parental cytokines, GIFT4 fusokine gains new function distinct from its parental cytokines GM-CSF and IL-4. GIFT4 has potent capability to activate and program naive B cells into immune effector cells. Programming of naïve B cells by GIFT4 fusokine involves both GM-CSF and IL-4 domains through a synergistic recruitment of GM-CSF receptor and IL-4 receptor clustered on B cell surface, which further triggers the formation of downstream signaling complex of JAK1 (The Janus kinase 1), 2, 3 and STAT1 (The signal transducer and activator of transcription 1), 3, 5 and 6 [63]. Inhibition of JAK signaling by its specific inhibitors completely interrupted GIFT4-induced STAT1, STAT3, STAT5 and STAT6 signaling in the treated B cells and consequent B cell expansion. In contrast, combined use of parental cytokines GM-CSF and IL-4 is unable to cluster the two receptors on B cell surface and induce B cell proliferation.


Figure 1: Structure of GIFT4 protein. (A) GIFT4 protein structure that contains GM-CSF and IL-4 domains. (B) Amino acids of human GIFT4

Interesting, GIFT4 protein has no effect on monocytes, although GM-CSF and IL-4 together have the capability to promote monocytes differentiation into dendritic cells. GIFT4-augmented B cells (GIFT4-B cells) express co-stimulatory molecules CD40, CD80 and CD86, and produce unique immune-stimulatory cytokines, chemokines and adhesion molecules including IL-1α, IL-6, IL-12, GM-CSF, CCL3, CCL4 and CD54, but little IL-10 and IFN-γ [63], apart from CD40-activated B cells [52] or innate response activator B cells [6]. With those immune properties, GIFT4-B cells function as APC-like effector cells, and consequently promote the expansion of CD314+, granzyme B-, granulysin- and IFN- -producing cytotoxic T cells that selectively kill human melanoma cells both in vitro and in vivo [63]. Moreover, GIFT4 fusokine induces B cell-dependent anti-tumor immunity in murine melanoma models [63], involving both APC-like B effector cells and GM-CSFproducing innate response activator B-cells [6,63]. In our investigation of GIFT4 as a potential vaccine adjuvant, we also discovered that GIFT4-coated virus-like particles enhance anti-HIV antigen-specific antibody production in vivo [64], suggesting additional effect of GIFT4 on the antibody-secreting cells. Indeed, we have found that administration of GIFT4 protein induces robust anti-melanoma specific-antibody production in murine melanoma model (Unpublished data). We have further extended our investigation to human chronic lymphocytic leukemic (CLL) B cells, and examined the immune activity of GIFT4-stimulated CLL B cells (GIFT4-CLL cells). Unlike CD40-activated CLL cells [58], TLR9 ligand-treated CLL cells [65] or normal GIFT4-B cells [63], GIFT4-CLL cells produce immune-stimulatory cytokines including IL-1β, IL-2, IL-6, IL-8, ICAM-1 and prime autologous T cells to proliferate, express tumorkilling molecules IFN-γ, CD314, perforin and granzyme B, and lyse autologous primary leukemic cells [66]. Taken together, GIFT4 induces broad anti-tumor B cell immune responses either through GIFT4-programmed B effector cells that further prime tumor-killing cytotoxic T cell response, or through the augmentation of tumor-specific antibody production. Those results provide a strong basis for the potential utilization of GIFT4 fusokine and GIFT4- augmented B cells as well as GIFT4-converted CLL cells for cancer immunotherapy in human.


B cells play pivotal roles in immune defense system, which bridge the innate and the adaptive immunities against cancers. Augmented B cells including GIFT4-B cells and expanded tumor-infiltrated B cells have potent immunestimulatory activities and anti-tumor function by either priming cytotoxic T cell response or producing anti-tumor specific antibodies. We predict that GIFT4 and GIFT4- augmented B cells as potential immune therapeutics could provide a new approach for cancer immunotherapy.


This work was supported by the Winship Robbins Scholar Award, the Winship Melanoma Research Fund and the Developmental Fund of the Winship Cancer Center Support Grant (5P30CA138292-06) (to JD).


The authors declare that they have no competing interest.