New prospects for autoimmune disease therapy: B cells on deathwatch

Authors

  • E. William St.Clair,

    Corresponding author
    1. Duke University Medical Center, Durham, North Carolina
    • Division of Rheumatology and Immunology, Box 3874, Duke University Medical Center, Durham, NC 27710
    Search for more papers by this author
    • Dr. St.Clair has served on the advisory boards of Genentech, Cellective Therapeutics (now owned by MedImmune), and Human Genome Sciences. Dr. Tedder has Tedder has received consulting fees (more than $10,000 per year) from Cellective Therapeutics (now owned by MedImmune), has given expert testimony for Idec Pharmaceuticals, and has patents pending with Cellective Therapeutics (now owned by MedImmune).

  • Thomas F. Tedder

    1. Duke University Medical Center, Durham, North Carolina
    Search for more papers by this author
    • Dr. St.Clair has served on the advisory boards of Genentech, Cellective Therapeutics (now owned by MedImmune), and Human Genome Sciences. Dr. Tedder has Tedder has received consulting fees (more than $10,000 per year) from Cellective Therapeutics (now owned by MedImmune), has given expert testimony for Idec Pharmaceuticals, and has patents pending with Cellective Therapeutics (now owned by MedImmune).


The improvement seen in the signs and symptoms of rheumatoid arthritis (RA) following B cell depletion with rituximab has fueled interest in the role of B cells in autoimmune disease. For several decades, B cells have been the focus of many investigators working to understand a variety of immune-mediated diseases, including RA, Sjögren's syndrome, and systemic lupus erythematosus (SLE). B cells, however, have not generally been thought to play a critical role in the mechanisms of systemic sclerosis (SSc). A study by Matsushita and colleagues, in this issue of Arthritis & Rheumatism (1), calls attention to B cells as potential key players in SSc. They show that patients with this disease have elevated serum levels of soluble BAFF (also known as B lymphocyte stimulator [BLyS; trademark of Human Genome Sciences, Rockville, MD]), a potent B cell survival factor.

The results reported by Matsushita et al raise the possibility that BAFF may promote the survival of pathogenic B cells in SSc and therefore contribute to disease expression. To this point, an additional finding of this study was that SSc B cells express increased amounts of the BAFF receptor (BAFFR) and overproduce IgG and interleukin-6 (IL-6) upon BAFF stimulation (1). Does up-regulated BAFF lead to autoantibody production, and if so, is it linked to disease expression? In the study by Matsushita et al, serum BAFF levels were not significantly correlated with disease-specific autoantibodies (1). Although antinuclear antibodies (ANAs) are a prominent laboratory feature of SSc, they do not appear to have a pathogenic role in this disease, unlike the autoantibodies to native DNA in SLE. Therefore, ANAs and a set of disease-specific autoantibodies (e.g., antibodies to topoisomerase I, centromere antigens, fibrillarin, fibrillin, and RNA polymerases) have heretofore served a diagnostic utility. In contrast, it has been the activated T cell, which is evident at lesional sites as well as in the blood, that is believed to be important in disease expression in SSc, especially the fibrotic components of the disease.

More decisive insights into the role of B cells in the pathogenesis of SSc will come from clinical trials testing B cell–targeted therapies. Importantly, it is now possible to include biologic assays in trials to examine the relationship between clinical response and immune function. Methodologic advances, for example, allow for the design of experiments that can detect and characterize in molecular detail a wide variety of B cell subsets. This approach, coupled with a growing understanding of B cell biology, will undoubtedly pay dividends in the future, and shed light on the importance of B cells in disease pathophysiology.

Rituximab, a CD20-directed monoclonal antibody, represents the initial foray into B cell–targeted therapy. Although anti-CD20 monoclonal antibody therapy is viewed with cautious optimism, it remains unproven whether complete or even partial B cell depletion will become an effective long-term strategy for the treatment of autoimmunity. For example, complete B cell depletion may be required to restrain disease initially, while partial B cell depletion may need to be maintained over the long term to yield sustained clinical benefits; such depletion may compromise the immune system over time. Based on these concerns, it is envisioned that newer therapies can fine-tune B cell function and thereby achieve the same therapeutic aims with a wider margin of safety. Thus, the present development of a monoclonal antibody to BAFF (Lymphostat-B) as a possible treatment for RA and SLE appears to be both timely and scientifically appealing, with the idea that neutralizing BAFF will induce B cells to die normally and keep pathogenic self-reactivity at bay. This approach may also be applicable to SSc, given that BAFF is up-regulated in this disease. Other similar therapies directed at altering B cell survival through CD40 and CD22 cell surface molecules are also in development.

How B cells may contribute to autoimmunity

Knowledge of the functions of B cells in both disease and health is needed to understand the potential benefits and side effects of B cell–targeted therapies. B cells are central players in adaptive immune responses and are cleverly designed to cope with the antigenic heterogeneity of infectious agents. As they develop, B cells traverse a tightly regulated pathway, from their inception as early progenitors to terminally differentiated plasma cells (Figure 1). Progression through this pathway depends on signals controlling both negative and positive selection of B cells, which occur centrally, in the bone marrow, and in the peripheral lymphoid tissues. Self-reactive B cells are mostly eliminated in the bone marrow by a process termed negative selection.

Figure 1.

BAFF, CD40, and CD22 survival factor activities during different B cell development stages compared with expression of cell surface CD40, CD22, and receptors for BAFF. Expression of receptors for BAFF and its activities during B cell development are described in ref.73. GC = germinal center.

The antigens mediating negative selection are usually abundant or membrane-bound self antigens delivering strong signals through the B cell receptor (BCR). Anti-self BCR-induced signals in this context lead to apoptosis. B cells that escape negative selection undergo a second screening for self-reactivity in the periphery, where they may be induced to change their specificity by receptor editing or they may be silenced by deletion or anergy. As part of this complex series of checks and balances, B cells undergo selection for the production of a large and diverse array of high-affinity effector antibodies while at the same time preventing the production of pathologic autoantibodies. Antigen-stimulated B cells may also acquire the capability to survive for long periods as memory cells that can rapidly respond upon subsequent encounters with the same antigen.

As part of host defense, B cells become activated to secrete antibodies that can eliminate foreign antigens. In addition to producing antibodies, B cells are also able to ingest and present antigens (2), express costimulatory molecules (3), secrete chemokines (4, 5) and cytokines (6), and regulate T cell activation (7), formation of lymphoid tissue, and functions of dendritic cells (8). B cell responses are influenced by BCR signaling, cell- surface receptors for extracellular ligands, cytokines, complement components, and growth factors in addition to Toll-like receptors (for review, see ref.9). Appropriate integration of these complex signals is a key factor for B cell negative selection, clonal expansion, isotype switching, affinity maturation, long-lived memory, and the maintenance of tolerance to self antigens. However, at least 2 broad categories of genetic defects promote a loss of B cell tolerance and promote autoreactivity in mice (10). Perturbations that affect B cell thresholds for cellular signaling, activation, or proliferation significantly increase the likelihood of developing autoimmune disease. Likewise, dysregulated apoptotic genes increase B cell life spans, promote survival of self-reactive B cell clones, and lead to the development of autoantibodies and multiple autoimmune syndromes (for review, see ref.11).

Role for B cells in SSc pathogenesis?

While B cell hyperactivity (e.g., hypergammaglobulinemia) and autoantibody production are well-known features of SSc (12, 13), other B cell abnormalities have become apparent that may link directly or indirectly to disease mechanisms. Peripheral blood B cells from patients with SSc overexpress cell-surface CD19, a positive regulator of B cell activation (14). CD19 serves as a general rheostat for B cell signaling thresholds (15) that is counterbalanced by CD22 negative regulation (16). Alterations in components of this signaling pathway can result in autoimmunity (17), and transgenic mice that overexpress CD19 by as little as 20–30% produce ANAs and rheumatoid factor (14). A CD19 gene polymorphism associated with higher CD19 expression in B cells is also significantly more prevalent in patients with SSc than in healthy controls (18).

B cell transcripts were recently discovered in lesional skin of patients with SSc, suggesting that B cells may play a role in the cutaneous manifestations of this disease (19). In addition, examination of peripheral blood from patients reveals expansion of naive B cells and a reduction in the number of memory B cells (20). These memory B cells express higher levels of CD95 (Fas, a cell surface inducer of apoptotic signals), raising the possibility that memory B cells are decreased due to enhanced apoptosis rates. An important role for B cells in SSc is also supported by studies in tight skin mice, a genetic model exhibiting multiple features of human SSc (15). In this model, B cell depletion following treatment with a CD20 monoclonal antibody partially abrogates skin fibrosis if administered before disease onset (Hasegawa M, et al: unpublished observations). Establishing further the role of B cells in the pathogenesis of SSc will require the testing of B cell–targeted therapies in well-designed clinical trials, with mechanistic assays to monitor B cell subsets and their function. While maximal B cell depletion is a potential direction in treating SSc, targeting B cell survival pathways, including those utilizing BAFF, CD40, and CD22, represents an alternative and conceptually attractive approach.

The BAFF/BLyS receptor system

Evidence suggests that BAFF may be a useful therapeutic target to modulate B cell survival in autoimmune disease. BAFF is expressed on neutrophils, monocytes, dendritic cells, lymphoid stromal cells, and malignant B cells as a type II transmembrane protein. The transmembrane form can be cleaved from the membrane by a furin-like convertase, generating a soluble protein fragment (21, 22). BAFF binds to 3 members of the tumor necrosis factor (TNF) family of receptors: BAFFR/BR3, transmembrane activator and calcium modulator and cyclophilin ligand interactor (TACI), and B cell maturation antigen (BCMA) (23). BAFFR is expressed mainly on the surface of B cells, while TACI is expressed on B cells and a subset of T cells, and BCMA is expressed on mature B cells and plasma cells (Figure 1). Two of the BAFF receptors, BCMA and TACI, also bind to the homologous family member, APRIL (a proliferation-inducing ligand), which stimulates B cell proliferation and tumor growth (23). When combined with BCR ligation, BAFF acts as a potent costimulator of B cells (21, 22) and can also rescue self-reactive B cells from death (24). BAFFR ligation up-regulates Bcl-2 expression and NF-κB activation, both of which increase B cell survival (25).

BAFF-deficient mice have a severe defect in mature B2 cell development, reduced levels of serum immunoglobulins, decreased numbers of T2 transitional B cells, impaired maintenance of germinal center reactions, and lowered survival of plasmablasts (26, 27). In contrast, the numbers of bone marrow B cell precursors, B1 cells, and T1 transitional B cells are normal in BAFF-deficient mice. BAFF binds with high selectivity to BAFFR versus BCMA, providing a molecular basis for understanding the different phenotypes of BAFFR- and BCMA-deficient mice (28). BCMA deletion by gene targeting has no effect on B cell development or immunoglobulin production (28), indicating that BAFF primarily mediates its effects on B cell development through BAFFR. TACI is a negative regulator of B cell activation, accounting for the similar phenotypes of TACI-deficient and BAFF-overexpressing mice (29). APRIL-deficient mice show no detectable abnormalities in B cell development (23). Thus, the BAFF receptor system is complex in regulating B cell development, function, and survival.

Mattsushita et al (1) show that SSc patients with diffuse cutaneous involvement have up-regulated BAFF messenger RNA expression in the affected skin and show increased serum levels of BAFF. In this study, higher levels of serum BAFF were associated with severe disease manifestations. Elevated serum levels of BAFF are not unique to SSc patients; they have also been observed in patients with SLE, patients with primary Sjögren's syndrome, and patients with RA (30, 31). BAFF levels are increased in MRLlpr/lpr mice and (NZB × NZW)F1 hybrid mice, both of which are animal models of SLE, and correlate with disease progression (32). Transgenic mice that overexpress BAFF are characterized by features of SLE and Sjögren's syndrome, with elevated levels of antibodies to double-stranded DNA (dsDNA), elevated levels of serum IgM, vasculitis, and glomerulonephritis (31, 33). These findings suggest that elevated BAFF levels could be a common feature of autoimmunity, with complex interactions between BAFF and aberrant genetic factors producing unique clinical phenotypes.

CD40/CD154

CD40, a member of the TNF receptor family, is another positive regulator of B cell survival and is broadly expressed during B cell development (Figure 1) (for review, see ref.34). CD40 was first characterized as an ∼50-kd transmembrane protein on most B lineage cells (Figure 1) but was later also found to be expressed on immature B cells, monocytes, dendritic cells, hematopoietic progenitors, endothelial cells, epithelial cells, and in some B cell malignancies (35, 36). The cellular ligand for CD40 (CD154) is a 33–39-kd type II transmembrane glycoprotein expressed primarily on a subset of activated CD4+ T cells. CD154 is, however, also expressed on activated B cells, natural killer cells, monocytes, eosinophils, basophils, dendritic cells, platelets, endothelial cells, and smooth muscle cells (35).

CD40 ligation promotes B cell proliferation and immunoglobulin production in vitro, and antibody isotype switching, germinal center formation, and induction of B cell memory in vivo (37). When combined with IL-4 in vitro, CD40 antibody binding prevents peripheral B cells from undergoing spontaneous apoptosis and promotes long-term B cell survival and expansion (38). CD40 ligation enhances B cell expression of the early activation markers CD69 and CD154, increases expression of class II major histocompatibility complex (MHC) molecules, CD80 and CD86, adhesion molecules (e.g., very late activation antigen 1), and CD95, and promotes the secretion of IL-6, TNFα, lymphotoxin, and chemokines (39). Thus, CD40 engagement provides a powerful stimulus during T cell–dependent B cell activation and the subsequent steps leading to the formation of memory B cells and antibody-secreting plasma cells.

The critical role that CD40 plays in humoral immune responses is best illustrated by the human immunodeficiency state resulting from CD154 mutations (40). This rare disease, X-linked hyper-IgM syndrome, is characterized by normal or elevated levels of serum IgM but absent IgG, IgA, and IgE, and defects in germinal center and memory B cell formation. Mice deficient in either CD40 or CD154 exhibit phenotypes similar to human hyper-IgM syndrome (41, 42). The potent effects of CD40 on humoral immune responses result in part through its ability to promote B cell survival, because apoptosis is the fate for the vast majority of germinal center B cells. Following MHC-restricted antigen activation, CD154-expressing T cells bind to CD40-expressing B cells, which initiates antigen-activated B cell survival signals and sustains germinal center reactions and the generation of long-lived memory B cells (37). CD40 ligation may also induce apoptosis under some conditions, such as in the context of increased CD95 expression as well as by other poorly understood mechanisms (43–45). Therefore, CD40 signals are essential for normal T cell–dependent B cell survival during humoral immune responses.

CD22

CD22 is a lectin-like member of the immunoglobulin superfamily expressed exclusively by all mature B cells (16). Structurally, CD22 consists of 7 immunoglobulin-like domains, with the 2 amino-terminal domains mediating adhesion to oligosaccharides bearing α2,6-linked sialic acid residues that are present on lymphocytes, neutrophils, monocytes, and erythrocytes, and serum components (46). CD22 expression is limited to the cytoplasm of pro-B and pre-B cells, with cell surface expression beginning around the time of IgD expression (Figure 1). Most IgM+,IgD+ B cells in the circulation express cell-surface CD22. In lymphoid tissues, CD22 is expressed by follicular mantle and marginal zone B cells but is only weakly expressed by germinal center B cells.

CD22 expression influences normal BCR and CD19 signal transduction and BCR-induced cell death, and is critically involved in the survival of follicular B cells (47, 48). These processes are differentially regulated by CD22 binding to its extracellular ligands and/or by the intrinsic activity of the CD22 cytoplasmic domain functioning independently of ligand engagement. The CD22 cytoplasmic domain influences signal transduction and contains potential immunoreceptor tyrosine-based activation motif-like regions (49) and immunoreceptor tyrosine-based inhibition motifs (50, 51). Several intracellular signaling events result from CD22 ligation, including tyrosine phosphorylation of CD22 itself, and increased association of CD22 with Lyn, the p85 subunit of phosphatidylinositol 3-kinase, and Syk (52–54). Rapid CD22 tyrosine phosphorylation also occurs following BCR stimulation, implying functional linkage between these 2 receptors (49, 55, 56). CD22-deficient B cells generate significantly augmented [Ca2+]i responses following BCR crosslinking, consistent with a negative regulatory role for this cell surface molecule (57–59). CD22 also has critically important interactions with CD19 on B cells, forming a regulatory loop in which these 2 molecules modulate each other's functions (60).

B cell survival is the major activity regulated by CD22 ligand binding for which CD22 appears to serve both costimulatory and inhibitory roles depending on the assay system (47, 48, 59). Monoclonal antibodies reactive with the ligand-binding domains of CD22 can promote survival and thereby stimulate B cell proliferation, even in the absence of BCR ligation (52). In costimulation assays, engaging ligand-binding domains of CD22 significantly enhances B cell proliferation in combination with anti-IgM antibodies, IL-2, IL-4, and CD40 monoclonal antibodies. In contrast, antibodies that block CD22 ligand binding are tumoricidal in xenotransplant models and induce apoptosis in tumor cell lines in vitro (61).

The role of CD22 ligand binding in B cell survival is revealed in CD22−/− mice, in which B cells have shorter life spans and increased apoptosis rates (57–59, 62). CD22−/− B cells also undergo apoptosis following BCR crosslinking in vitro, depending on their genetic background (47). Importantly, however, CD40-generated signals can rescue CD22−/− B cells from BCR-induced cell death (47). In fact, CD40 ligation alone also significantly increases the proliferation of CD22−/− B cells relative to that of wild-type B cells. In mice genetically engineered to express mutated CD22 that does not bind ligands, a phenotype remarkably similar to CD22−/− B cells is observed, characterized by shorter life spans and increased apoptosis rates (48). Remarkably, BCR signal transduction is normally regulated in B cells that express these mutated CD22 receptors, except for a delay in BCR-induced proliferation.

Taken together, these data demonstrate that CD22 ligand engagement promotes the survival of the peripheral pool of naive B cells prior to CD40 engagement during germinal center recruitment. In the absence of both CD22 ligand binding and CD40 ligation, mature B cells may not receive sufficient survival signals and undergo apoptosis following BCR engagement. In support of this reasoning, CD22−/− mice show slight increases in serum IgM levels, normal T cell–dependent antigen responses, and decreased T cell–independent responses. Thus, CD22 and CD40 function as reciprocal regulators of B cell survival during different stages of development and BCR-mediated selection (Figure 1).

Clinical trials

Clinical trials using monoclonal antibodies directed against BAFF, CD154, and CD22 are under way or are planned, for many different autoimmune diseases. Thus far, results of these clinical studies are too limited to determine whether modifying B cell survival will become a viable treatment strategy. LymphoStat-B is a fully human antibody that neutralizes soluble BAFF activity in vitro as well as in vivo (63). The results of a phase I study of LymphoStat-B in patients with SLE revealed a significant decrease in circulating B cell numbers and no major safety concerns (63). There has been a preliminary announcement of the results of a phase II, randomized controlled clinical trial of LymphoStat-B in patients with RA. The benefits associated with Lymphostat-B in this trial were relatively modest considering that only 29.4% of patients in the active treatment arms fulfilled the American College of Rheumatology criteria for a 20% improvement (64), compared with 15.9% of the placebo group (65). LymphoStat-B is currently being evaluated in a phase II trial for SLE. The results of this study were recently announced in a press release, which indicated that an analysis of the primary end point failed to show a statistically significant difference between the LymphoStat B–treated and placebo groups.

The clinical efficacy and safety of 2 different humanized CD154 monoclonal antibodies (IDEC-131 and BG9588) have been tested in patients with SLE, idiopathic thrombocytopenic purpura, and Crohn's disease (66–68). In a phase II, randomized, double-blind, placebo-controlled trial involving 85 patients with mild-to-moderately active SLE, multiple infusions of IDEC-131 did not significantly improve disease activity, as measured by the Safety of Estrogens in Lupus Erythematosus: National Assessment version of the SLE Disease Activity Index (69) or a change in the serum levels of anti-dsDNA autoantibodies or complement (66). In another clinical trial, 28 patients with active lupus nephritis were scheduled to receive 3 biweekly doses followed by 3 monthly doses of BG9588 in an open-label trial, but the trial was terminated prematurely due to the occurrence of thromboembolic events in 2 patients (67). Among 18 patients in that trial for whom efficacy could be evaluated, only 2 met the primary end point of a 50% reduction in proteinuria without worsening of renal function. This group of treated patients nevertheless showed a reduction in the mean level of serum anti-dsDNA antibodies and an increase in mean serum complement levels.

B cells from 4 of the patients in the BG9588 study were examined in order to further elucidate the role of CD40–CD154 interactions in SLE (70). Before treatment, the blood of these 4 patients with active SLE contained expanded populations of circulating B cells expressing CD38 at increased levels (e.g., plasmablasts). CD154 treatment reduced the numbers of these circulating CD38bright antibody–secreting cells in parallel with a decline in anti-dsDNA levels. Peripheral blood from children with active SLE had previously been found to contain increased numbers of CD38bright pre–germinal center B cells and plasma cell precursors among blood B cells. Thus, circulating plasmablasts may be a potential biomarker for this disease (71). Despite these promising preliminary results, therapeutic CD154 antibodies currently have an uncertain future because of concerns about their association with thromboembolic complications.

CD22 also represents a potential target for modulating B cell survival. Phase II clinical trials evaluating the effects of a CD22 monoclonal antibody (epratuzumab) in SLE are under way. However, this particular antibody binds CD22 outside of the ligand-binding domains; it may therefore not directly inhibit B cell survival but rather may induce B cell depletion by antibody-dependent or complement-dependent cytotoxicity (72, 73). An antibody that blocks CD22 ligand binding and induces B cell apoptosis (46, 61) is also being developed for clinical trials in autoimmune disease.

Therapeutic opportunities and challenges

There is considerable enthusiasm for strategies that alter B cell survival, and there is a convincing scientific rationale for restoring normal B cell function through this approach. The association of autoimmunity and autoantibody production with specific B cell subsets such as marginal zone B cells suggests that selected B cell depletion may also have advantages over total B cell elimination. Moreover, enhancing B cell turnover may selectively purge the repertoire of pathogenic autoantibody-producing cells while allowing the bone marrow to continue production of nascent B cells that can respond to foreign but not self antigens.

The ability to rescue antigen-specific B cells through other survival pathways also represents a potential advantage for inducing rapid B cell depletion while retaining normal humoral immunity. For example, CD22−/− B cells undergo rapid B cell turnover, but this can be prevented in vivo by CD40-generated survival signals that foster the development of protective humoral immunity. Thus, a great deal remains to be learned regarding the optimal use of BAFF, CD40/CD154, and CD22, but the potential therapeutic benefits justify this effort and focus.

Due to the complexity of the cellular networks governing B cell function, targeting these 3 survival pathways independently for therapeutic purposes may nonetheless be met with initial disappointments and failures. For example, B cell abnormalities in human autoimmune disease are likely to result from alterations in more than one pathway. In addition, some autoimmune diseases may have clinically similar disease manifestations yet result from a variety of different causes or defective pathways in individual patients. In some patients, targeting a single receptor–ligand interaction such as BAFF–BAFFR may be sufficient to restore normal B cell function, while the majority or a subset of patients within each disease category may not respond due to the underlying genetic basis of their disease. Moreover, the timing of the intervention may be critical to its clinical success, because mechanisms triggering the disease may differ from those inducing progression or maintaining established illness. Even in cases in which therapeutic targeting may reestablish normal B cell function, alteration in other immune cell compartments (e.g., T cells, macrophages, inflammation) may be present that can perpetuate the disease phenotype despite normal B cell function or an absence of B cells.

Perhaps the most important consideration in targeting B cell survival is that each of the cell-surface molecules currently known to be involved in this process is a component of a complex regulatory network. Thus, inhibiting the BAFF/BAFFR system, with its dual ligands and 3 receptors and multiple downstream effects on B cell development, may lead to unpredictable positive or negative effects in the context of a complex autoimmune disease such as RA or SLE. The thromboembolic potential of therapeutic CD154 antagonists as currently envisioned may not allow targeting of this molecule without significant risks. CD40-blocking antibodies might similarly inhibit CD154–CD40 interactions but result in varied and possibly undesirable outcomes. CD40 antibodies may also deliver agonistic signals, leading to cellular activation. The expression of CD40 by dendritic cells and nonhematopoietic cells makes the situation even more complex.

CD22 engagement may also stimulate positive and negative signals in B cells that act through overlapping pathways. Moreover, therapeutic antibodies that engage ligand-binding domains versus non–ligand-binding domains of CD22 may have entirely different effects. Thus, while it is convenient to view each of these receptor systems as simple on/off switches, each receptor represents a multipart console of structural and functional domains regulating their overall functions. Each of these receptor systems also interacts with multiple signaling pathways and communicates with other cell surface receptors in complicated ways. In addition, individual B cell subpopulations will have a “history,” providing a context for the interpretation of subsequent signals delivered for therapeutic purposes, adding even further complexity to the predictability of treatment outcomes.

Although the dynamic interrelationships among the pathways controlling B cell survival remain poorly understood, they offer exciting and potentially fruitful targets for therapeutic intervention. Undoubtedly, these complexities will also pose major therapeutic challenges that will require persistence among investigators and the companies currently developing these approaches. The complexity inherent in regulating B cell signaling and survival may soon require therapeutic strategies that target multiple pathways. Although combined targeting approaches may increase the risks of excessive immunosuppression (e.g., infection), these approaches may ultimately provide the most effective and tailored therapies. Altering B cell survival may also require the initial depletion of the peripheral B cell pool, with subsequent therapy regulating the survival of reconstituting B cells for long-term therapy. Regardless, all of these approaches demand a better understanding of the role these molecules play in disease pathogenesis and the molecular effects of these therapies in vivo. In this context, further studies such as those examining the up-regulation of BAFF in SSc (1) take us a step further in that journey seeking to launch a deathwatch for hostile B cells.

Acknowledgements

We thank Drs. Karen Haas and Jonathan Poe for their review of the manuscript.

Ancillary