Autoreactive B cells are one of the key immune cells that have been implicated in the pathogenesis of systemic lupus erythematosus (SLE). In addition to the production of harmful auto-antibodies (auto-Abs), B cells prime autoreactive T cells as antigen-presenting cells and secrete a wide range of pro-inflammatory cytokines that have both autocrine and paracrine effects. Agents that modulate B cells may therefore be of potential therapeutic value. Current strategies include targeting B-cell surface antigens, cytokines that promote B-cell growth and functions, and B- and T-cell interactions. In this article, we review the role of B cells in SLE in animal and human studies, and we examine previous reports that support B-cell modulation as a promising strategy for the treatment of this condition. In addition, we present an update on the clinical trials that have evaluated the therapeutic efficacy and safety of agents that antagonize CD20, CD22 and B-lymphocyte stimulator (BLyS) in human SLE. While the results of many of these studies remain inconclusive, belimumab, a human monoclonal antibody against BLyS, has shown promise and has recently been approved by the US Food and Drug Administration as an indicated therapy for patients with mild to moderate SLE. Undoubtedly, advances in B-cell immunology will continue to lead us to a better understanding of SLE pathogenesis and the development of novel specific therapies that target B cells.
Systemic lupus erythematosus (SLE) is a prototypic autoimmune disorder that predominately affects young women who present with heterogeneous clinical symptoms in an array of tissues and organs. Because of the diversity of the symptoms, SLE diagnosis is challenging and is made principally based on patients' clinical presentations upon meeting four or more of the 11 SLE Classification Criteria that have been designated by the American College of Rheumatology with supporting laboratory tests.1 SLE, therefore, likely represents a spectrum of diseases with similar pathological manifestations but is caused by the interplay of multiple etiological factors. Immunologically, SLE is generalized as a disease that is triggered by a combination of environmental and genetic factors that result in a loss of tolerance towards self-antigens.2 Among the diverse clinical symptoms, the prominent increase in anti-nuclear antigen auto-antibodies (auto-Abs) is a hallmark in nearly all SLE cases and is the key pathological factor leading to the formation and deposition of the immune complexes that cause tissue inflammation and damage in the kidneys, skin, joints and central nervous system.3 The common consensus is that there is a general breakdown in B-cell (and T-cell) tolerance in SLE patients. Subsequently, autoreactive B cells are often considered one of the central effector cell types that are responsible for the perpetuation of inflammatory responses, and B cells are considered the main therapeutic target of SLE treatment.4,5
Roles of B cells in SLE
Auto-Abs can appear years before the clinical onset of SLE, and there is a progressive accumulation of auto-Abs before disease onset which suggests that they are a key pathological factor.6 In SLE, a well-recognized role of B cells is to produce a panoply of auto-Abs against self-antigens such as DNA in both the double-stranded (ds) and single-stranded (ss) conformations, RNA-containing/binding nuclear antigens including Smith antigen, ribonucleoprotein, the Ro/SAA and La antigens, and non-nuclear components including ribosomal P antigen and phospholipids. The prevalence of different types of auto-Abs varies and may be associated with different clinical presentations in SLE patients. For example, anti-ribosomal P Abs are present in 5%–10% of SLE patients, and their presence is correlated with the risk of developing neuropsychiatric symptoms.7 Anti-dsDNA Abs are one of the most commonly observed auto-Abs in SLE patients by far; they are present in 30%–70% of patients, and their serum levels correlate with the development and progression of glomerulonephritis.8,9,10 The pathogenic attributes of anti-DNA Abs in the development of lupus extend beyond their ability to bind DNA and also require the contributions of other cellular components. It has been shown that anti-DNA Abs have the potential to cross-react with various cellular proteins,11 including actinin-α12 and heparan sulphate, a major glycosaminoglycan in glomerular basement membranes.13 Injecting anti-DNA IgG-producing hybridomas derived from SLE patients into SCID mice caused only mild proteinuria, which correlated with the immunoreactivity of the IgG in the glomeruli and in the cellular nuclei in the kidneys and other organs.14 However, no pathological change was observed in the kidney, which suggested that anti-DNA Abs alone was not sufficient to induce nephritic damage and that T- and B-cell contributions were critical. Indeed, more profound changes was observed when similar experiments were performed in normal mice. Normal BABL/c mice that received anti-DNA mAbs derived from MRLlpr/lpr and NZB×SWR F1 lupus strains exhibited intranuclear Ig deposition in multiple organs, and the Ig deposition in the glomeruli was also associated with pathological and functional changes such as increasing cellularity, signs of collagen synthesis and the induction of proteinuria.15
In SLE pathogenesis, the role of B cells is more than just as a source of auto-Abs. In rheumatoid arthritis models, antigen-specific B cells have been shown to help prime autoreactive T cells as antigen-presenting cells (APCs).16 In SLE, a series of earlier studies by Shlomchik and co-workers have provided evidence of this direction. Targeted JH deletion in MRLlpr/lpr lupus mice caused B-cell deficiencies and the concomitant absence of nephritis and vasculitis, which indicated that B cells were critical for disease initiation and development.17 The numbers of activated and memory T cells were also drastically reduced in these mice.18 Chan et al.19 subsequently created an MRLlpr/lpr mutant mouse line with B cells that only expressed surface Ig but not secretory Abs. These B-cell-intact but Ab-deficient mice spontaneously developed nephritis with associated cellular infiltration. In vivo spontaneous T-cell activation was also evident, confirming the Ab-independent role of B cells in either serving as APCs for antigen-specific autoreactive T cells, or by contributing directly to local inflammation. In support of this hypothesis, B-cell aggregates have been observed in lupus nephritic kidneys, and most of these intrarenal B cells display a mature, non-Ab-producing, and antigen-presenting phenotype.20 Evidence has also shown that auto-antigen-primed B cells are capable of activating autoreactive T cells in vivo.21
In addition to functioning as APCs, B cells are also a rich source of various cytokines that exhibit both autocrine and paracrine activities.22 The cytokine responses of B cells depend highly on the nature of the stimuli. Upon appropriate stimulation of both the B-cell receptor (BCR) and CD40, human B cells secrete predominately pro-inflammatory cytokines, such as lymphotoxin-α, tumor-necrosis factor alpha (TNF-α) and interleukin-6 (IL-6), which may act in an autocrine fashion as growth and differentiation factors and also function to amplify other immune responses. In contrast, bystander stimulation of CD40 without BCR engagement induces B cells to produce regulatory IL-10, which has a suppressive effect on other immune cells.23 The cytokine network dysregulation observed in SLE is highly intricate, and many cytokines are produced by multiple immune and non-immune components under inflammatory conditions. It is, therefore, rather difficult to delineate the unique contributions of B-cell-derived cytokines. The pathological relevance of various critical cytokines that are related to SLE has been evaluated and reviewed elsewhere.24 Here, we will instead discuss the therapeutic applicability of a few B-cell-related cytokines and growth factors.
Strategies in B-cell-targeted therapy
Corticosteroids and broad spectrum immunosuppressants are the mainstream therapy for standard treatment of SLE. The therapeutic outcomes of these conventional treatments are often variable and are usually associated with severe side effects. In view of their central role as the main effectors in SLE pathogenesis, various new biologics that target B cells have been developed and evaluated in both preclinical and clinical settings. Summarized below and in Figure 1, we will discuss the rationales of the various strategies that have been adopted and their supporting evidence from in vitro studies and animal models.
Targeting B-cell surface antigens to achieve cell depletion/inhibition
CD20 belongs to the tetraspan family of integral membrane proteins. Its function is largely unknown but a recent report suggests that this B-cell differentiation antigen may play a central role in T-cell-independent antibody responses.25 CD20 is widely expressed across different stages of B-cell development, only except in early pre-B cells and terminally differentiated plasma cells. Anti-CD20 B-cell depletion was initially explored for the treatment of B-cell malignancies. Rituximab, a chimeric antibody comprising the mouse Fab and the human IgG1κFc portion of an anti-CD20 monoclonal antibody, was first marketed and approved by the FDA to treat B-cell lymphomas and showed significant clinical benefits.26 Subsequent trials in SLE patients, however, yielded variable outcomes.27 Rituximab achieves B-cell depletion through a number of different mechanisms, including direct induction of programmed cell death, antibody-dependent cellular cytotoxicity and complement-dependent cytotoxicity.26 The relative contributions of these pathways in vivo may vary with different clinical situations, but in the context of SLE, the defects associated with apoptotic machinery, complement insufficiency, Fcγ receptor expression variations and genetic polymorphisms28 would have a strong effect on treatment efficacy. Indeed, the FcγRIIIa genotype has been reported as an independent predictor of B-cell depletion efficacy following rituximab treatment in SLE patients.29 Furthermore, human anti-chimeric antibody responses may also account for the poor B-cell depletion that has been observed in some patients receiving rituximab.30 These factors may, at least in part, account for its lack of significant clinical benefits in several randomized controlled trials (details will be discussed in a later section). Although the rituximab response in SLE patients falls short of the results observed in cancer settings, newer generations of anti-CD20 biologics, including the fully human MAb ofatumumab and other engineered versions with improved antibody-dependent cellular cytotoxicity killing, are currently under clinical development.26,31
Anti-CD22 and CD19
Based on a similar rationale to target B-cell-specific surface antigens that are expressed broadly across different B-cell subsets, therapeutics against CD22 and CD19 are also being explored.32 CD22, a member of the sialoadhesin subclass of the Ig superfamily, is a lectin receptor that binds α2,6-linked sialic acid residues. The immunoreceptor tyrosine-based inhibition motifs in the CD22 signaling domain confer its negative regulatory effect on BCR, CD19/21 and CD45 signaling, and consequently, it plays a role in BCR-induced cell death and B-cell survival in the periphery.33 In contrast, CD19 is a transmembrane protein that is physically associated with BCR and functions to potentiate BCR signaling.34 Similar to rituximab, the humanized anti-CD22 MAb epratuzumab was also first tested as a B-cell lymphoma therapy. Its in vivo effects, however, likely extend beyond cell depletion. Epratuzumab induces a moderate B-cell depletion mainly in the CD27− naive and transitional B-cell compartments in SLE patients,35 and this finding is consistent with a higher expression of CD22 in this cell population.36 In vitro analyses only indicate a mildly toxic effect. However, additional regulatory effects in down-modulating the exaggerated activation and proliferation of lupus B cells but not normal B cells have been reported.35 Epratuzumab binding also induces a change in the expression pattern of the adhesion molecules CD62L and the β7 and β1 integrins with an associated enhancement in the migration of CD27− B cells towards CXCL12.36 While the impact of these regulatory effects on the in vivo efficacy of epratuzumab remains unclear, a newer version of the anti-CD22 MAb with improved cytotoxicity has been developed. BL22, an anti-CD22 MAb fused with Pseudomonas aeruginosa exotoxin A, is currently in phase I trial for the treatment of B-cell malignancies.37 A similar version of anti-CD19 with a maytansinoid conjugate is also under consideration.38 The rationale is that by tagging a B-cell specific MAb with a toxin, irrespective of the target molecule function, cellular cytotoxicity would be expected upon internalization.
With our increasing knowledge in defining the markers and functions for different B-cell subsets, one obvious caveat of targeting a pan-B antigen is the lack of specificity against harmful autoreactive B cells. Invariably, B cells with regulatory functions may also be affected. Much has been learned in the past decade on the characteristics, functions and origins of regulatory T cells, and it is now apparent that certain B-cell subsets also exhibit regulatory functions.39 Recent evidence also indicates that regulatory B cells may play a prominent role in suppressing SLE development. In NZB/W lupus mice, CD19 deficiency promoted early-onset nephritis and reduced survival, which were associated with a reduction of CD1dhiCD5+B220+ IL-10-producing regulatory B10 cells when compared with wild-type mice.40 Adoptive transfer of B10 cells from healthy mice improved animal survival in CD19−/− hosts, demonstrating a protective function of regulatory B cells in SLE development. In humans, a similar subset of IL-10-producing B cells that are defined as CD19+CD24hiCD38hi has been shown to inhibit T helper cell differentiation in vitro, and this cell subset is functionally impaired in SLE patients.41 However, the in vivo role of these newly identified IL-10-producing B cells in humans remains unclear.
Targeting cytokines to limit B-cell growth and functions
Anti-BAFF (B-cell activating factor of the TNF family)/APRIL (A proliferation-inducing ligand)
Of the many B-cell growth factors, the BAFF (also known as B-lymphocyte stimulator (BLyS), THANK or TALL-1) and its homologue APRIL have gained much attention recently as therapeutic targets for rheumatic diseases. BAFF was initially discovered as a cytokine that induced B-cell proliferation and immunoglobulin secretion, but later studies demonstrated its critical role in peripheral B-cell survival and development.42,43,44 BAFF has three known receptors, namely the BAFF receptor (BAFF-R or BR3), the transmembrane activator-1 and CMCL interactor (TACI), and the B-cell maturation antigen (BCMA). APRIL, which shares similar functions with BAFF, binds to TACI and BCMA but not BR3, and it appears to play a distinct role in T-cell-independent type II antigen responses.45 The expression of these receptors also varies in different B-cell subsets. Plasma cells express BCMA and TACI, and thus they are sensitive to both BAFF and APRIL stimulation; however, BAFF has a unique role in immature and transitional B cells, which express BR3 but not the other two receptors.
In view of its prominent effect on B-cell growth and differentiation, it is not surprising to discover the pathogenic attributes of the BAFF/APRIL system in SLE development. BAFF is overexpressed in patients with SLE and other autoimmune conditions,46,47,48,49 and BAFF concentration has been shown to correlate with disease activity and anti-dsDNA antibody titers.46,48,49,50 Interestingly, in addition to dendritic cells and macrophages, the aberrant BAFF/APRIL expression in patients can be observed in naive, memory and plasma cells, which can produce functional BAFF/APRIL upon activation in vitro. The upregulation of BAFF/APRIL mRNA levels that is observed in lupus CD19+ B cells correlates with anti-dsDNA Abs levels and the SLEDAI.51
The relevance of the BAFF/APRIL system in SLE pathogenesis is further revealed in animal models. Various transgenic mouse lines that overexpress BAFF have high levels of circulating anti-dsDNA auto-Abs and immune complexes, and develop lupus-like symptoms even in the absence of T-cell help.52,53,54 BAFF-driven lupus disease in these mice depends on MyD88 signaling, which enhances TLR-mediated auto-Ab production in marginal zone and B1 B cells.54 In addition, both NZB/W F1 and MRLlpr/lpr mice exhibit an increase in serum BAFF levels during disease onset and progression.53 In the absence of BAFF, NZM2328 lupus-prone mice are protected from overt disease and present less severe proteinuria and reduced mortality; however, high serum Ig, immune complex deposition and proliferative glomerulonephritis are still apparent.55 Further inactivation of APRIL in these BAFF knockout mice does not improve the renal immunopathology,56 which suggests that there may be no clinical advantage in blocking both BAFF and APRIL. However, when using adenovirus receptor-Ig fusion proteins as pathway blockers, there are qualitative differences between blocking BAFF alone or BAFF and APRIL together. BAFF-R-Ig, which blocks only BAFF, and TACI-Ig, which blocks both BAFF and APRIL, both prolong the survival of NZB/W F1 mice and induce similar in vivo effects by depleting B cells and preventing progressive T-cell activation. However, there is an insignificant effect on the anti-dsDNA response. Significantly, TACI-Ig (but not BAFF-R-Ig) can further suppress serum IgM antibodies, reduce plasma cell frequency in the spleen and inhibit IgM responses to a T-cell-dependent antigen.57 More detailed investigations are thus needed to fully characterize the differences between the outcomes when using selective and non-selective BAFF blockers because they may have a profound influence on the choice of therapeutics when treating B-cell-related diseases.
IL-6 is a pleiotropic cytokine with a broad spectrum of biological functions and is produced by a multitude of cell types.58 Of relevance to SLE pathogenesis, IL-6 is also known as the B-cell stimulatory factor 2 for its essential role in the differentiation of B cells into Ig-producing plasma cells.59 Exogenous IL-6 can enhance spontaneous IgG production by lupus B cells in vitro.60 In lupus patients, elevated levels of serum IL-6 have been observed,60,61,62 and in some cases, IL-6 levels correlate directly with disease activity60 and inversely with treatment response.62 Serum IL-6 elevation is also consistent with the observation of increased frequencies of IL-6-producing leukocytes in the peripheral blood of patients.61,63 Indeed, IL-6 expression has been observed in both monocytes and lymphocytes in patients.60,64,65,66 Autoreactive T-cell clones that are derived from SLE patients support polyclonal B-cell activation, which can be inhibited by anti-IL-6.64 In addition to the constitutive expression of IL-6 receptor, SLE autoreactive B cells also spontaneously produce high levels of IL-6, which acts in an autocrine manner to cause further cell activation and auto-Ab production.65,66 Importantly, IL-6 neutralizing antibodies have been shown to inhibit anti-ssDNA IgG antibodies and total IgG production by SLE B cells in vitro, which highlights the potential therapeutic value of these agents.60,65
Evidence from murine lupus models further supports the pathogenic role of this critical cytokine. Splenic B cells from NZB/W F1 (but not BALB/c) mice respond to exogenous IL-6 in vitro by producing IgM, IgG and anti-DNA antibodies.67 Interactions with autoreactive T cells also promote B cell auto-Ab production partly through IL-6. In coculturing T and B cells from aged NZB/W F1 mice, the addition of neutralizing anti-IL-6 abrogates anti-DNA IgG production in B cells.68 Additionally, the capacity of NZB/W F1 mice to produce anti-DNA IgG correlates with the ability of B cells to respond to IL-6 and with the disease stage of the animals.68 The success of neutralizing IL-6 antibodies to suppress auto-Ab production in vitro has set the scene for exploring anti-IL-6 therapy in vivo. An early study by Finck et al.69 reported that the weekly administration of a rat anti-mouse IL-6 Ab alone to pre-symptomatic 3-month-old NZB/W F1 mice did not improve the disease course due to the anti-idiotypic Ab response mounted by the host. However, a tolerizing dose of anti-CD4 concurrently with the first injection of anti-IL-6 resulted in a significant reduction in anti-dsDNA auto-Ab production, a lower incidence of severe proteinuria and improved survival in mice at 9 months of age. A similar study by Liang et al.70 later showed that anti-IL-6 alone was sufficient to ameliorate disease in NZB/W F1 mice with improved kidney pathology and less auto-Ab production in B cells. In addition, both T and B cells from anti-IL-6-treated mice showed reduced proliferation responses towards antigen receptor stimulation ex vivo, which suggests that the T-cell compartment was also affected by the treatment. This observation signifies a wider scope of IL-6 involvement as more than a B-cell stimulatory factor in SLE development. IL-6 also regulates other immune components such as the differentiation and function of various T-cell subsets.
Recent studies suggest that an imbalance between IL-17-producing T (Th17) cells and regulatory T (Treg) cells is an important pathologic factor in several autoimmune diseases. Interestingly, IL-6 plays opposite roles in the development of these two types of T cells. In the presence of TGF-β, IL-6 favors the differentiation of naive T cells into Th17 cells, and at the same time, it has an inhibitory effect on TGF-β-induced Treg differentiation.71,72 In SLE, this perturbation of the Th17/Treg balance may also be important. In B6.Sle1.Sle2.Sle3 triple congenic lupus mice, the overproduction of IL-6 by intrinsically dysregulated dendritic cells (DCs) has been shown to not only block Treg cell development, but also to inhibit Treg cell suppression activities.73
Targeting B- and T-cell interactions
T-cell help is a critical parameter that enables the development of a full spectrum of humoral responses. Different T-cell subsets produce specific cytokines that can skew the nature of humoral responses. For example, IL-4 secreted by Th2 cells is a critical B-cell growth factor that also promotes class switching in B cells to produce IgE, while interferon (IFN)-γ produced by Th1 cells favors the production of IgG2a. In addition to soluble factors, direct cell–cell contact to establish ‘cross-talk’ between T and B cells is also indispensable for generating optimal adaptive immune responses.
This pathway is perhaps one of the most studied areas in T- and B-cell interactions. B cells constitutively express CD40, while CD40L (also called CD154 or gp39) is found primarily, though not exclusively, on activated T cells. These receptors belong to the TNF receptor and TNF super-families, and they reciprocally regulate the cellular functions of their binding partners.74 CD40L is a costimulatory molecule for T cells, and its signaling is crucial for the development of cellular immunity, including cytotoxic T-cell responses to viruses, tumors and alloantigens. In contrast, CD40 signaling is essential for the generation of a full spectrum of thymus-dependent humoral responses. Its critical roles in germinal center formation and development, antibody isotype class switching and affinity maturation, and the development of plasma and memory cells have been firmly established.75
In SLE, a high basal expression of CD40L has been found on both T and B cells in active patients. In inactive patients, its expression on lymphocytes, although comparable to healthy subjects, is readily induced to much higher levels than normal controls upon activation, confirming that lupus T and B cells are intrinsically dysregulated.76 Additionally, lupus CD4+ T cells have been shown to have prolonged CD40L upregulation upon activation, which in turn induces higher CD80 expression on B cells when the cells are cocultured together.77 The expression of CD40L on lupus B cells is also unique because CD40L is seldom expressed by normal B cells even after stimulation. CD40L also exists in a soluble form, sCD40L. A higher serum sCD40L level, which correlates with anti-dsDNA auto-Ab levels and the SLEDAI, has been observed in patients, and these sCD40L molecules are functionally active to induce B-cell activation.78,79 The in vivo pathological contribution of heightened CD40L expression, either cell-bound or soluble, is not completely clear. However, anti-CD40L mAb significantly prevents lymphocytes from SLE patients with active disease from producing a variety of auto-Abs in vitro, which suggests the potential therapeutic value of blocking the CD40–CD40L pathway.76
Interestingly, parallel or similar observations of ectopic basal CD40L expression in T and B lymphocytes from different murine lupus models have also been reported, and the expression levels induced upon activation were significantly higher than the normal counterparts.80,81 Forced over-expression of CD40L on B cells induces spontaneous lupus-like symptoms in some but not all of the transgenic mice in a non-lupus-prone background, which suggests that mere ectopic CD40L expression by B cells is not sufficient to cause disease and that some other pathological factor(s) is/are required for induction.82 Nevertheless, the in vivo efficacy of anti-CD40L Ab treatment in ameliorating disease has been demonstrated in NZ-related lupus strains. Anti-CD40L Ab treatment of presymptomatic lupus mice has been shown to delay disease onset, reduce the incidence and severity of nephritis, and prolong animal survival.80,83,84 In one study, alleviation of clinical symptoms and improved kidney pathology were even observed when treatment was given to older mice that had established disease. In that case, a more aggressive treatment regimen was needed.84 These data have prompted the clinical testing of anti-CD40L mAbs in human SLE.
Other strategies for disrupting or compromising the T-cell help for B cells are blocking or interfering with T-cell costimulatory pathways. The classical CD80/CD86-CD28 costimulatory pathway is a prime target for immunotherapy in various autoimmune diseases. Cytotoxic T lymphocyte antigen 4 (CTLA4), a negative regulator that is induced on activated T cells to limit excessive T-cell activation, has a higher binding affinity for CD80/CD86 than CD28 and is thus exploited as therapeutics to suppress T-cell activity through competitive blockade of CD28. CTL4-Ig has been widely tested as an immunotherapeutic in a number of disease systems and models.85,86 In SLE, CTLA4-Ig has been a highly promising therapeutic agent with encouraging preclinical data from various models.87 Its efficacy in patients remains to be validated because the results from a few phase II/III clinical trials are not yet available.
B-cell-targeted therapies of SLE: clinical trials update
Numerous B-cell-related biologics are currently at different stages of clinical development.5 In this section, we will only focus on those therapies that are at the advanced stage of testing in the clinic. A number of randomized controlled trials (RCT) have been performed on SLE patients using anti-CD20 antibodies (e.g., rituximab, ocrelizumab), anti-CD22 antibodies (e.g., epratuzumab) and anti-BAFF antibodies (e.g., belimumab). Off-label use and uncontrolled studies of rituximab have also been extensively reported in patients with SLE who do not respond to conventional treatment or who have life-threatening manifestations. The following information summarizes the RCTs for various B-cell-targeted agents as well as uncontrolled studies involving rituximab.
Rituximab is a chimeric monoclonal antibody against the B-cell surface antigen CD20. Two RCTs evaluating the use of rituximab in patients with SLE have been published. In the Exploratory Phase II/III SLE Evaluation of Rituximab (EXPLORER) trial, 257 patients with moderate to severe extra-renal SLE were randomized in a 2∶1 ratio to receive rituximab or placebo in addition to their baseline immunosuppressive agents and a 10-week course of moderate-to-high-dose oral prednisone.88 No statistically significant differences were observed between the two groups in terms of a reduction in disease activity using British Isles Lupus Assessment Group (BILAG) scores. The second RCT evaluated the efficacy and safety of rituximab in patients with ISN/RPS class III and IV lupus nephritis (LUNAR trial).89 In the LUNAR trial, 144 patients were randomized in a 1∶1 ratio to receive either rituximab or placebo on a background of treatment with mycophenolate mofetil and corticosteroids. The results showed a statistically significant improvement in serum complement C3, C4 and anti-dsDNA levels, but there were no differences in the renal response rates at week 52.
In contrast to the results from these two RCTs, numerous observational studies and case series have reported favorable responses to rituximab in patients with active or refractory SLE. In these observational studies, patients with active or refractory SLE that was unresponsive to conventional immunosuppressive agents were given rituximab. The overall response rates range from 55% to 87%, and significant responses have been documented in lupus nephritis90,91,92,93 and neuropsychiatric lupus.94 The most common adverse events were infections, with cutaneous infections and pneumonia being more frequent.94 Although these studies were uncontrolled, they provide ample evidence that rituximab is a potentially useful agent in the treatment of patients with severe SLE that is refractory or intolerant to conventional immunosuppressive agents. In addition, they can even be considered as the first-line therapy in life-threatening scenarios, such as pulmonary hemorrhage,95,96 renal failure and central nervous system involvement.97
Ocrelizumab is a humanized monoclonal antibody against the B-cell surface antigen CD20. In the Study to Evaluate Ocrelizumab in Patients With Nephritis Due to Systemic Lupus Erythematosus (BELONG), 381 patients with active class III or IV lupus nephritis were randomized to receive either placebo or ocrelizumab in addition to standard of care (SOC) mycophenolate mofetil or cyclophosphamide.98 The study was suspended in March 2010 following the recommendations of an independent monitoring board that based their studies on patients with rheumatoid arthritis and SLE. The board review detected an infection-related safety signal, including severe and opportunistic infections, some of which were fatal. The week-48 renal responses for patients enrolled for 32 weeks or more prior to stopping the drug showed a 12% better response in patients receiving ocrelizumab (63%) compared with those receiving placebo (51%, P=0.075). An interim safety analysis also revealed an imbalance in serious infections, particularly in patients receiving ocrelizumab with concurrent mycophenolate mofetil and also in patients from Asia across all arms.98
Epratuzumab is a human monoclonal antibody against the B-cell surface antigen CD22. It has been evaluated in two phase III RCTs that involved SLE patients with moderate to severe SLE.99,100 These trials were prematurely terminated due to interruptions in medication supply, and the analyses were combined to increase the available data. Of the 55 SLE patients treated with epratuzumab, a clinically meaningful steroid-sparing effect was noted, and there was also a greater reduction in total BILAG scores when compared with placebo.101 Treatment with epratuzumab was found to significantly improve patients' quality of life in terms of Short Form 36 Health Survey scores, as well as physician's and patient's global assessment scores.102 Another phase III study (NCT00383513) involving patients recruited in the previous two trials has recently been completed and aims to provide long-term efficacy and safety data concerning epratuzumab.103
Belimumab is a human monoclonal antibody against BLyS, a key survival factor for B lymphocytes. In a phase II dose-ranging RCT of belimumab in 449 patients with active SLE, no significant differences were detected between the treatment and placebo groups for the primary end points.104 However, a subsequent subgroup analysis of the serologically active patients (71.5%), which was defined as being positive for anti-nuclear Abs or anti-dsDNA antibodies, detected a significantly better response in the belimumab-treated group in terms of reduction in SELENA-SLEDAI scores, physician's global assessment and Short Form 36 physical component scores. A 6-year follow-up of these patients showed a sustained improvement in disease activity, a reduced frequency of flares and a decrease in corticosteroid use.105
Two phase III RCTs have studied the use of belimumab in patients with active SLE and positive serology. In these two studies, namely the BLISS-52106 and BLISS-76107 trials, patients were randomized to receiving 1 or 10 mg/kg belimumab or placebo in addition to SOC therapy. The combined results from these two trials (total n=1684) demonstrated that significantly more belimumab-treated patients had improvements in the musculoskeletal (1 mg/kg), mucocutaneous (10 mg/kg) and immunological (1 and 10 mg/kg) domains by week-52, which were the three most frequently affected domains at baseline.108 Improvements in other organ systems, including renal, hematological and central nervous systems were also observed. In addition, significantly fewer patients treated with belimumab experienced worsening in the BILAG hematological domain (1 mg/kg) or in the SELENA-SLEDAI immunological (10 mg/kg), hematological (10 mg/kg) and renal (1 mg/kg) domains.108 Multivariate analysis of the pooled data identified high baseline disease activity, anti-dsDNA positivity, low C3 or C4 levels and pre-existing prednisolone use as the predictors of a better response to belimumab plus SOC therapy compared to SOC treatment alone.109
Treatment of SLE, particularly those cases of refractory disease, still poses a major challenge for rheumatologists. To date, randomized placebo-controlled trials using B-cell-depleting mAbs, either anti-CD20 or anti-CD22, have not yielded satisfactory results. Much optimism, however, has been offered by the latest trials with belimumab. The actual clinical efficacies of other B-cell-related therapies are still waiting to be revealed by more advanced trial studies. Compared to other rheumatic diseases, the interplay of various known (and unknown) etiological factors in SLE is far more complicated, which is supported by data from the latest genetic association studies. Genome-wide association analyses of SLE patients have disclosed a broad collection of genetic risk loci that are related to immune complex and antigen clearance (ITGAM, TREX1, FcγR genes, complement genes), T-cell activation (PTPN22, TNFSF4, HLA class II), B-cell signaling (BANK1, BLK, PRDM1), as well as TLR–IFN pathways (IRF5, STAT4, IRAK1).28 Perhaps targeting B cells alone may not be sufficient, and combination therapy would offer better promise. In fact, the focus of immunotherapy has shifted more toward the upstream regulators of the immune response, i.e., DCs. Various studies over the past decade have brought attention to the key pathogenic attributes of plasmacytoid (p-)DCs, which uniquely can secrete a large amount of type-I interferons upon nucleic acid and immune complex stimulation through toll-like receptors 7 and 9.110 pDC-related biologics for SLE are also being actively pursued. Sifalimumab, a fully human anti-IFN-α mAb is in phase I trial, while DV1179, a TLR7 and TLR9 antagonist, is under preclinical development for lupus.111 As far as B-cell-targeted therapies are concerned, a deeper understanding of the development and characteristics of various B-cell subsets is indispensable. The in vivo functions of the newly identified CD19+CD24hiCD38hi IL-10-producing B cells in humans and their role in SLE are still unknown.41 Recently, a previously unidentified distinct population of CD27− B cells with a memory phenotype has been observed in lupus patients. This homogeneous subset of CD27−IgD−CD95+ memory B cells has an activated phenotype, and its frequency in patients correlates with disease activity and serologic abnormalities.112 Again, their function and response to B-cell-targeted therapy is not clear. Advances in this area will most certainly help us to re-design some of the current B-cell therapeutics in the future.