Belimumab in systemic lupus erythematosus: an update for clinicians

Abstract

Systemic lupus erythematosus (SLE) is a chronic inflammatory disorder that is driven by autoantibodies that target multiple organ systems. B-lymphocyte stimulator (BLyS) and its receptors on B-cell subsets play an important role in autoimmune B-cell development and SLE pathogenesis. Targeted therapy with belimumab, the monoclonal antibody against BLyS, has shown clinical benefit in two large-scale, multicenter phase III trials leading to US Food and Drug Administration approval for patients with serologically positive SLE who have active disease despite standard therapy. This review will discuss the challenges in lupus drug development and clinical trials, the basics of B-cell pathogenesis in SLE, the recent lupus clinical trials of B-cell targeted treatments, and other potential targeted therapies under investigation for patients with lupus.

Introduction

Systemic lupus erythematosus (SLE) is a chronic inflammatory disorder that is driven by autoantibodies, which target multiple organ systems including joints, skin, and kidneys. The relapsing-remitting pattern of disease, along with the clinical heterogeneity, makes SLE not only one of the challenging autoimmune disorders to diagnose, but also to treat and assess drug efficacy on a large-scale basis. Another complexity of SLE is the marked health disparities that exist within the lupus population, related to modifiable and nonmodifiable risk factors, such as socioeconomic factors, race and ethnic heterogeneity [Gillis et al. 2007]. The American College of Rheumatology (ACR) criteria for SLE classification proposed in 1982 and revised in 1997 give a framework to identify patients for clinical studies (Table 1) [Hochberg, 1997]. Given the clinical heterogeneity of lupus, ACR criteria were developed to ‘classify’ patients in a research setting; the purpose is not necessarily to exclude or confirm the lupus diagnosis in a clinical setting.

Table 1.

The 1997 revised American College of Rheumatology classification criteria for the diagnosis of systemic lupus erythematosus [Hochberg, 1997].

Malar rash	Fixed erythema, flat or raised, over the malar eminences
Discoid rash	Erythematous circular raised patches with adherent keratotic scaling and follicular plugging; atrophic scarring may occur
Photosensitivity	Exposure to ultraviolet light causes rash
Oral ulcers	Includes oral and nasopharyngeal ulcers, observed by physician
Arthritis	Nonerosive arthritis of two or more peripheral joints, with tenderness, swelling, or effusion
Serositis	Pleuritis or pericarditis documented by electrocardiogram or rub or evidence of effusion
Renal disorder	Proteinuria >0.5 g/day or 3+, or cellular casts
Neurologic disorder	Seizures or psychosis without other causes
Hematologic disorder	Hemolytic anemia or leukopenia (<4000/liter) or lymphopenia (<1500/liter) or thrombocytopenia (<100,000/liter) in the absence of offending drugs
Immunologic disorder	Anti-dsDNA, anti-Sm, and/or false-positive serologic test for syphilis known to be positive for at least 6 months and confirmed by Treponema pallidum immobilization or fluorescent treponemal antibody absorption test
Antinuclear antibodies (ANAs)	An abnormal titer of ANAs by immunofluorescence or an equivalent assay at any point in time in the absence of drugs known to induce ANAs

Any combination of four or more of 11 criteria, well documented at any time during a patient’s history, makes it likely that the patient has systemic lupus erythematosus (specificity and sensitivity are 95% and 75%, respectively).

The current treatments for SLE aim to restore the balance in a dysregulated immune system. The mainstay of therapy is glucocorticoids, which since their introduction in the 1950s have altered the management of most rheumatic diseases, and have led to gradual improvements in disease management and quality of life for patients with lupus. However, there are major toxicities from long-term glucocorticoid use, involving its effects on infection risks, bone health, and glucose homeostasis. Thus, antimalarials, nonsteroidal anti-inflammatory drugs, azathioprine, methotrexate, cyclosporine, mycophenolate mofetil, and cyclophosphamide have been used for steroid-sparing and immunuosuppressive effects. Antimalarials, particularly hydroxychloroquine, are widely used for milder manifestations of disease (skin and joint inflammation); these agents have long-term protective effects against lupus flares [Tsakonas et al. 1998]. Cyclophosphamide has been shown in prospective, controlled trials to enhance renal survival in patients with proliferative lupus nephritis [Gourley et al. 1996]. In addition, cyclophosphamide is regarded as the first-line treatment for severe manifestations of lupus, such as lupus vasculitis [Klippel, 1998]. Notably, most of these medications are used on an empirical basis and not approved by the US Food and Drug Administration (FDA), with the exception of glucocorticoids, hydroxychloroquine, and aspirin.

On 9 March 2011, FDA approved belimumab (Benlysta, Human Genome Sciences and GlaxoSmithKline) to treat patients with autoantibody [antinuclear antibody (ANA) and/or anti-double-stranded DNA (dsDNA)]-positive SLE who have active disease despite standard therapy. The recommended dose is 10 mg/kg at 2-week intervals for the first three doses and at 4-week intervals thereafter. Belimumab is a fully human recombinant immunoglobulin G (IgG) 1λ monoclonal antibody to soluble B-lymphocyte stimulator (BLyS) [Baker et al. 2003]. FDA approval comes after the positive results showing clinical efficacy and safety in two large-scale phase III randomized controlled trials, involving 1684 patients with lupus, one conducted in Asia, South America, and Eastern Europe, and the other in the USA, Canada, and Europe [Navarra et al. 2011; Furie et al. 2010]. The approval and introduction of belimumab to the limited list of lupus medication armamentarium heralds the first new lupus drug that has been approved in more than 50 years.

The purpose of this article is to briefly review the following: the challenges in lupus drug development and clinical trials; the basics of B-cell pathogenesis in SLE and the importance of BLyS as a key factor in B-cell survival and selection; the recent lupus clinical trials of B-cell targeted treatments (rituximab and belimumab); and other potential targeted therapies under investigation for patients with lupus.

Challenges in lupus drug development and clinical trials

Despite clear advances in our understanding of the pathophysiology of lupus, several challenges exist that impede drug development for patients with lupus.

First, lupus pathogenesis is complex due to nonlinear immune pathways. Although the disease is characterized by pathogenic autoantibodies that target specific tissues, many additional cell types (e.g. B cells, T cells), cytokines [e.g. type I interferon (IFN-I)-α], and proteins are involved in the inflammatory response. Lupus can arise from various molecular junctures ranging from defective proteins that regulate T cells to the dysfunctional clearance of immune cells, suggesting part of the pathology lies in loss of the immune tolerance and the persistence of autoreactive B- and T-cell populations [Ravirajan et al. 1996].

Second, the diverse clinical spectrum of lupus that can initially present with vague manifestations (e.g. fever, fatigue, and arthralgia) can pose as a diagnostic challenge for many clinicians who may not recognize the systemic nature of the inflammation. In addition, both the temporal sequence of organ involvement and the severity of its course are often unpredictable. Beyond the clinical complexity of diagnosis and course of illness, the challenge often becomes offering treatment modalities that strike the fine balance between immunosuppression and immune dysregulation.

Third, the most effective instrument to measure SLE disease activity as well as the response to therapy is still open to debate. There are several validated measures including SLAM (Systemic Lupus Activity Measure), SLEDAI (Systemic Lupus Erythematous Disease Activity Index), LAI (Lupus Activity Index), ECLAM (European Consensus Lupus Activity Measurement), and BILAG (British Isles Lupus Activity Group) [Gladman, 1994]. The SLEDAI and the BILAG, each with updated versions (SLEDAI-2000, SELENA-SLEDAI, and BILAG-2004), are two indices that are most widely used in randomized clinical trials [Merrill et al. 2004]. The SLEDAI consists of a list of ‘descriptors’ based on different organ manifestations that are weighted by importance of an organ system, according to its ‘presence’ or ‘absence’ in the past 10 days of evaluation [Petri et al. 1999]. The SLEDAI does not take into account the severity of involvement within the organ system and does not include the patient’s assessment of disease activity, which may be an important part in the evaluation of overall lupus disease activity [Petri, 2007]. The BILAG combines the strengths of SLEDAI, of weighting for the perceived severity of a manifestation, along with a sliding scale scoring for the organ system involved [Stoll et al. 1996]. The BILAG consists of 86 ‘descriptors’ based on eight organ manifestations that are scored according to its ‘presence (new, worse, same, or improving)’ or ‘absence’ in the past 28 days of evaluation. Within each organ system involvement, there are gradations of severity [A: ‘active’ (severe disease); B: ‘beware’ (less active disease); C: ‘contentment’ (mild stable disease); D: ‘discount’ (inactive, but previously active); E: ‘no evidence’ (inactive and never affected)] that are scored to summarize the activity within each domain such that, for example, a patient with severe thrombocytopenia will have a higher organ system score than a patient with mildly active mucosal oral ulcerations [Petri, 2007; Hay et al. 1993]. The complexity and length of the BILAG index clearly make it a complicated but stringent tool to longitudinally follow each organ involvement.

In summary, the lack of effective lupus drug development does not in any way represent a paucity of effort in the scientific and medical community for the search of effective therapies; rather, it reflects the complexity of lupus pathogenesis and the clinical heterogeneity, which translates to the difficulty in developing targeted therapies.

Basics of B-cell pathogenesis in systemic lupus erythematosus

B cells play a central role in the pathogenesis of SLE, mainly by producing autoantibodies, but also by producing cytokines and by presenting antigens to T cells. In addition to the adaptive immunity (the antigen-specific, more precise immune response carried out by T and B cells), the innate immunity (the nonspecific, initial defense line of natural killer cells, phagocytic cells such as neutrophils, macrophages, and dendritic cells) also plays an important role in SLE pathogenesis; there is a dynamic interplay between the two immune responses. For instance, immune complexes of autoantibodies with endogenous RNA and DNA can be taken up by plasmacytoid dendritic cells, which activate toll-like receptor (TLR)-7 and TLR9, respectively, in their endosomal compartment to generate IFN-I [Crow and Kirou, 2004]. IFN-I can then further activate the adaptive immunity by enhancing the antigen-presenting function of monocytes and dendritic cells, activating B cells for class switching and expression of TLR7, and stimulating T-helper (Th) cells and their production of Th1 cytokines [Kim et al. 2009].

Another important link between adaptive and innate immunity with a special role in SLE pathogenesis is BLyS, also known as B-cell activating factor (BAFF). BLyS is a cytokine member of the tumor necrosis factor (TNF) family, which plays a crucial role in B-cell selection, maturation, and survival. Many immune cells, such as monocytes/macrophages, dendritic cells, T cells, and activated neutrophils, have been implicated in the production of BLyS, which is a 285-amino-acid cytokine [Scapini et al. 2003; Nardelli et al. 2001]. Interleukin-2, TNF-α, IFN-I, and IFN-γ can stimulate BLyS production. BLyS, along with a related cytokine [a proliferation-inducing ligand (APRIL)], binds to three different receptors that are situated on B cells, known as BAFF-receptor 3 (BR3), transmembrane activator and calcium modulator and cyclophylin ligand interactor (TACI), and B-cell maturation antigen (BCMA). This binding stimulates B-cell maturation and survival from immature B cells to plasma cells with BR3 expression more dominant in early stages and TACI/BCMA expression in late stages (Figure 1) [Litinskiy et al. 2002]. BLyS can bind to all three receptors, whereas APRIL interacts only with TACI and BCMA.

Figure 1.

Phases of B-cell development and activation (adapted from [Cancro et al. 2009]).***, High-affinity binding; *, low-affinity binding, relatively; , increased dependence; , decreased dependence, relatively.

(A/B) Pre-B cells originate from mesenchymal stem cells in bone marrow. Immature B cells leave bone marrow and enter into circulation as transitional B cells. B cells complete maturation in the spleen. In the periphery, signals from antigen activation and co-stimulation from T cells, help transition mature, preimmune B cells to form germinal centers, ultimately leading to memory and plasma cell development.

(C) Three B-lymphocyte stimulator (BLyS) family receptors (BR3, BAFF-receptor 3; TACI, transmembrane activator and calcium modulator and cyclophylin ligand interactor; and BCMA, B-cell maturation antigen) vary in their expression patterns and levels across stages of B-cell development.

(D) BLyS binding affinity is dominant in the immature, pre-B cell phase in the bone marrow, and continues to increase onto the transitional cell phase. BR3 expression is highest in the follicular and marginal zone subsets of B cells in the spleen. After antigen activation, BLyS/APRIL receptor expression on mature B-cells transitions from BR3 expression to TACI and/or BCMA expression in the memory/plasma cell population. APRIL-TACI and APRIL-BCMA interactions have been shown to influence plasma cell survival and Ig class-switching.

Mouse models have shown that BLyS/BR3 binding provides the crucial survival signal in the early stages of B-cell development by antagonizing apoptosis, and by potentiating further B-cell differentiation into mature, preimmune B-cell populations [Bossen et al. 2008; Schiemann et al. 2001]. The early intracellular signaling pathway triggered by BLyS/BR3 binding governs the homeostasis of the preimmune B-cell population, from its generation in the bone marrow through the transitional development to mature follicular B cells or marginal zone B cells. Normally, only about one-third of transitional B cells survive through various check points for self tolerance, due to negative selection of those with autoreactivity [Stohl et al. 2011]. High BLyS levels allow autoimmune B-cell survival that would otherwise be eliminated at the checkpoint. Mouse models have illustrated that overexpression of BLyS leads to B-cell hyperplasia; an increased population of self-reactive B cells may lead to production of autoantibodies, similar to the pathogenesis of SLE and Sjogren’s syndrome [Mackay et al. 1999]. In addition, cross-sectional studies have shown elevated BLyS levels in patients with SLE and other autoimmune diseases, as well as association of BLyS levels with SLE disease activity and dsDNA levels [Stohl et al. 2003].

In summary, SLE pathogenesis involves both the activation and dysregulation of innate and adaptive arms of the immune system. Recent research has implicated that the IFN-I signaling pathway plays a central role; its induction from the TLR pathway activates players of the adaptive immune response (T and B cells), which leads to generation of autoantibodies and immune complexes that target self-antigens or tissues. B-cell related pathways [e.g. B-cell survival factors (BLyS and APRIL)] that are critical for proper B-cell selection, maturation, and survival, and improper regulation of BLyS levels have been implicated as the key potential mechanisms for autoimmune dysregulation in patients with lupus.

The recent lupus clinical trials of B-cell targeted treatments

Rituximab

Rituximab (RTX) is an anti-CD20 monoclonal antibody composed of a chimera of the murine variable regions against CD20 and the human IgG Fc constant region. RTX is FDA approved for non-Hodgkin’s lymphoma, chronic lymphocytic leukemia, moderate to severe rheumatoid arthritis not responsive to TNF antagonists, and most recently, for two forms of antineutrophil cytoplasmic antibody associated vasculitis, granulomatosis with polyangiitis (formerly known as Wegener’s granulomatosis), and microscopic polyangiitis. Its off-label use in the autoimmunity world has been validated in the treatment of hematologic conditions, such as idiopathic thrombocytopenic purpura or autoimmune hemolytic anemia [Stone et al. 2010; Cohen et al. 2006].

There have been case reports of off-label use of RTX in patients with severe refractory SLE with or without nephritis; various dosing regimens (375 mg/m² weekly for 4 weeks, or 1000 mg 2 weeks apart) showed efficacy with an acceptable safety and tolerability profile [Murray and Perry, 2010; Terrier et al. 2010; Hughes, 2009]. These promising studies were then followed by two large-scale phase III, randomized, double-blinded, placebo-controlled studies, the Efficacy and Safety of Rituximab in Patients with Severe SLE (EXPLORER trial) and the Efficacy and Safety of Rituximab in Class III or IV Lupus Nephritis (LUNAR trial). Both of these trials failed to show significant difference between RTX and placebo when added to the standard treatment in patients with active lupus [Merrill et al. 2010; Furie et al. 2009].

The EXPLORER trial consisted of 257 patients with moderate to severe lupus disease activity (excluding renal disease), as defined by one BILAG A (severe) score or two BILAG B (moderate) scores. Patients were randomized to receive placebo versus RTX (1000 mg on days 1, 15, 168, 182) in addition to the baseline immunosuppressive therapy (azathioprine, mycophenolate mofetil, or methotrexate) and a 10-week course of standardized corticosteroid taper [Merrill et al. 2010]. The follow-up duration was 52 weeks. The BILAG index was used to assess SLE activity and response. The primary endpoints were defined as major clinical response (MCR; achieving BILAG C scores or better in all organs at week 24 without having a severe flare, and maintaining this response without a moderate or severe flare to week 52); partial clinical response (PCR; protocol-defined improvements in BILAG scores but not meeting MCR requirements); or no clinical response at week 52. There was no difference between the placebo and RTX groups (MCR was 12.4% versus 15.9%, and PCR was 12.5% versus 17.2%, respectively) [Merrill et al. 2010]. Furthermore, no difference was found in preventing or delaying the time to first moderate to severe disease flare and improvement in quality of life. However, a subgroup analysis of African American and Hispanic patients showed the beneficial effect of RTX on the primary endpoint measures (33.8% of African Americans and Hispanics had an MCR or PCR versus 15.7% of non-African Americans and non-Hispanics, p = 0.04) [Merrill et al. 2010]. Of interest, a recent post hoc analysis of the EXPLORER trial showed that RTX reduced the risk of subsequent first BILAG A flares compared with placebo (hazard ratio 0.61; p = 0.052) [Merrill et al. 2011].

The second large phase III trial (LUNAR) randomized 144 patients with class III or IV lupus nephritis to placebo versus RTX, in the background of steroid and mycophenolate mofetil use [Furie et al. 2009]. The trial was designed to show a beneficial effect of RTX for the induction of a renal response. Complete renal response was defined as normalization of serum creatinine, inactive urinary sediment, and urine protein/creatinine ratio ≤ 0.5. Partial renal response was defined as serum creatinine ≤ 15% above baseline value, no significant worsening of urinary sediment, and 50% improvement of urine protein/creatinine ratio. The results failed to detect any significant difference in renal response between the placebo and RTX groups.

The EXPLORER and LUNAR studies were large-scale trials that failed to show benefit of RTX in patients with moderate to severe nonrenal SLE and with class III/IV lupus nephritis, respectively. One potential explanation for the failure is the use of high-dose corticosteroids along with immunosuppressive therapies, which would make any incremental benefit from the experimental drug difficult to detect unless there is a large effect. Other potential explanations for the failure are related to the study design: inclusion of both serologically active and inactive patients with SLE, and not using the SLE responder index (SRI) as was done in the belimumab trial (discussed below). In addition, the duration of the study period (52 weeks) may not have been long enough to observe the renal response, in which clinical outcome measures in lupus nephritis may become apparent over several years. Furthermore, the effect of RTX in combination with cyclophosphamide in the treatment of severe, refractory SLE, which has appeared beneficial in small open-label studies, has yet to be determined in controlled trials [Gunnarsson et al. 2007 and Jonsdottir et al. 2007]. These negative studies also stressed the methodological problems of any research involving patients with SLE given the heterogeneity of disease manifestations, and how to use the appropriate outcome measures to detect overall clinically significant changes in nonrenal lupus. Despite the negative trial results for rituximab, this agent is still used in patients with lupus refractory to conventional therapies [Murray and Perry, 2010; Lu et al. 2009; Reynolds et al. 2009].

Belimumab

The mounting evidence from animal studies has identified the BLyS family of ligands and receptors as playing a central or facilitating role in B-cell pathogenesis in SLE. BLyS-neutralizing monoclonal antibody (belimumab) was developed with the goal of inhibiting the binding of soluble circulating BLyS to its target receptors (BR3, TACI, BCMA) on B cells. The ultimate goal was to block the crucial survival signal in the early stages of B-cell development and to decrease the survival of autoreactive B cells.

There is overexpression of BLyS in patients with SLE; 50% and 61% of patients compared with controls were found to have persistently or intermittently elevated serum BLyS and BLyS mRNA phenotypes, respectively [Stohl et al. 2003]. Interestingly, there were wide variations of serum BLyS levels in many of the patients with SLE at baseline compared with controls, and patients were distributed among the persistently normal, persistently elevated, and intermittently elevated groups. Treatment of patients with SLE with high-dose steroids in this study resulted in a marked decrease in serum BLyS levels. Furthermore, another prospective study showed that serum BLyS levels are related to lupus disease activity, as evaluated by SELENA-SLEDAI scores [Petri et al. 2008]; a greater increase in the SELENA-SLEDAI score from a previous visit was associated with higher BLyS levels, thereby implying an interesting relationship between serum BLyS levels and SLE disease activity.

The phase I, double-blind, randomized controlled trial of belimumab showed a relative safety profile in patients with SLE (n = 70); there was no difference in the prevalence of adverse events between the belimumab-treated and placebo-treated groups during the 12-week study period [Furie et al. 2008]. Significantly greater percent reductions in peripheral B-cell levels were achieved after one or two doses in the belimumab-treated group compared with the placebo-treated group. Overall, there was no difference in disease activity, as measured by SELENA-SLEDAI, between the two groups. Although, the short duration of the trial and the relatively small number of patients possibly precluded demonstration of a clinical benefit, the purpose of this phase I trial was not to test the efficacy of belimumab.

Subsequently, a phase II double-blinded randomized controlled trial examined the safety, efficacy, and biological activity of a much larger cohort of patients with SLE (n = 449), who were assigned to either belimumab (1, 4, or 10 mg/kg) or placebo in a 52-week study [Wallace et al. 2009]. The inclusion criteria were adult patients fulfilling ACR criteria for SLE, who had active disease as defined by a SELENA-SLEDAI score of at least 4, as well as a history of autoantibodies (ANAs, dsDNA, anti-Smith, antiribonucleoprotein, anti-Ro, anti-La, or anticardiolipin antibodies), but autoantibodies did not have to be present at screening. Of note, 28% of the patients were seronegative at the screening. The primary endpoints were percentage change in disease activity, as defined by SELENA-SLEDAI score, at week 24; and time to the first SLE flare over 52 weeks, as defined by SELENA-SLEDAI Flare Index. All patients with SLE received the standard of care therapy, which included a stable regimen of steroids, antimalarials, or other immunosuppressants for 60 days prior to the first belimumab infusion. Of note, after the screening visit, steroid and immunosuppressive regimens were allowed to be modified as clinically needed. There was no statistically significant difference between the two groups in attaining either of the primary endpoints. The percentage reduction in SELENA-SLEDAI score from baseline visit was 19.5% in the combined belimumab groups versus 17.2% in the placebo group at 24 weeks, and 27.2% versus 20.6% at week 52, respectively. Furthermore, there was no significant difference in time to first SLE flare over the 52-week course between the combined belimumab and placebo groups (67 versus 83 days, respectively).

Despite the failure of this phase II study to meet the coprimary endpoints, an extensive post hoc analysis of 321 patients (71.5% of the entire cohort) with serologically active (ANA ≥ 1:80 and/or anti-dsDNA ≥ 30 IU/ml) lupus demonstrated that patients receiving belimumab had significant improvements at 52 weeks in various secondary endpoints: SELENA-SLEDAI score (−28.8% in the combined belimumab group versus −14.2% in the placebo group; p = 0.04); physician’s global assessment (PGA) score (−32.7% versus −10.7%, respectively; p = 0.001); and short form 36 (SF-36) physical component score (+3.0 versus +1.2 points, respectively; p = 0.04). In addition, the serologically active group included more African Americans (27% versus 16%; p = 0.01), and fulfilled a greater number of ACR SLE criteria, with more major organ involvement (renal 34% versus 19%; hematologic 59% versus 33%), less cutaneous involvement, higher mean SELENA-SLEDAI scores (9.8 versus 8.9), and greater prednisone use (72.6% versus 57.8%; p = 0.002). Furthermore, the immunological profile of this group showed lower C3 and C4 levels (p < 0.0001), higher Ig levels (IgG, IgA, and IgE, p ≤ 0.001), lower baseline CD19+/CD20+ B-cell counts (p ≤ 0.01), and interestingly, higher levels of BLyS compared with patients with seronegative SLE (51% versus 24%; p < 0.0001).

An important development from this study was the SRI, a more sensitive composite tool to measure SLE disease activity for future clinical trials [Furie et al. 2009]. The SRI is composed of the SELENA-SLEDAI score to define overall improvement in disease activity; the BILAG domain score to assess for any change in specific organ involvement; and the PGA to assess that any change in disease activity took into account the patient’s overall condition. The SRI response is defined by at least a four-point reduction in SELENA-SLEDAI score; no new BILAG A or no more than one new BILAG B domain score; and no deterioration from baseline in the PGA by at least 0.3 points. The fulfillment of all three criteria labeled a study patient as a ‘responder’ at a specific assessment point. The post hoc analysis of patients with serologically active SLE (n = 321) included in the above phase II trial was retrospectively evaluated using the SRI; the response rate was 46% in the belimumab-treated group at week 52 compared with 29% in the placebo group (p = 0.006). Of note, the analysis of all patients (n = 449) revealed that the SRI responses were independent of baseline autoantibody subtype; the SRI response rate was significantly higher at week 52 in the belimumab group (45.9%) versus the placebo group (35.4%) (p = 0.045) [Furie et al. 2008]. This composite index was used as the primary endpoint for the two large phase III trials of belimumab.

On 20 July 2009, Human Genome Sciences and GlaxoSmithKline announced positive results from their phase III, double-blind, placebo-controlled lupus clinical trial [Navarra et al. 2011]. The 52-week study, also known as BLISS-52 (a Study of Belimumab in Subjects With Systemic Lupus Erythematosus), enrolled 865 patients in 13 countries outside North America (Latin America, Asia Pacific, and eastern Europe) and randomized patients with active lupus to intravenous belimumab 10 mg/kg, 1 mg/kg, or placebo in addition to their standard lupus medications. Patients had to fulfill the ACR lupus classification criteria, have active lupus (defined as SELENA-SLEDAI score ≥6), and remain on a stable treatment regimen for at least 30 days (excluding other biological therapies or cyclophosphamide) in order to be eligible for the trial. Patients with severe lupus kidney disease and lupus-related neurologic problems were excluded. All patients were serologically positive, in regards to positive ANA (titer ≥ 1:80) or anti-dsDNA antibody (≥30 IU/ml). Selected baseline characteristics of the study cohort are outlined in Table 2. The study medication was given 2 weeks apart for the first three doses, and then every 4 weeks. The study was successful in meeting its primary outcome, as defined by the SRI; the SRI response rates were 57.6% (p = 0.0006), 51.4% (p = 0.013), and 43.6% for belimumab 10 mg/kg, 1 mg/kg, and placebo, respectively (Table 3, Figure 2). In addition to improvement in various clinical measurements of disease activity, patients were also able to reduce steroid dosages (Table 4). Specifically, a significantly greater proportion of patients treated with belimumab 10 mg/kg (28%) were able to attain at least a 50% reduction in prednisone dose by visit 52 compared with placebo (18%) (p = 0.01). Notably, the drug was tolerated well and the safety profile, including the infection rate, was comparable to the placebo group.

Table 2.

Selected baseline demographic and clinical characteristics of the BLISS-52 study [Navarra et al. 2011].

Baseline characteristics	Placebo (n = 287)	Belimumab1 mg/kg (n = 288)	Belimumab10 mg/kg (n = 290)
Disease duration, years [SD]	5.9 (6.2)	5.0 (4.6)	5.0 (5.1)
SELENA-SLEDAI score, mean [SD]	9.7 (3.6)	9.6 (3.8)	10.0 (3.9)
BILAG 1 A or 2B score	166 (58%)	166 (58%)	172 (59%)
Prednisone > 7.5 mg/day at baseline	192 (67%)	204 (71%)	204 (70%)
Immunosuppressive drugs	122 (43%)	120 (42%)	123 (42%)
Antimalarial drug	201 (70%)	195 (68%)	185 (64%)
ANAs (≥1:80)	264 (92%)	272 (94%)	276 (95%)
Anti-dsDNA (≥30 IU/ml)	205 (71%)	221 (77%)	218 (75%)
Low C3 levels	132 (46%)	148 (51%)	147 (51%)
Low C4 levels	160 (56%)	173 (60%)	180 (62%)
Proteinuria (≥2 g/24 h)	21 (7%)	26 (9%)	19 (7%)

Immunosuppressive drugs are mycophenolate mofetil, azathioprine and methotrexate.

ANA, antinuclear antibody; BILAG, British Isles Lupus Activity Group; BLISS, A Study of Belimumab in Subjects With Systemic Lupus Erythematosus; C3, C3 complement; C4, C4 complement; SELENA-SLEDAI, Safety of Estrogens in Lupus Erythematosus National Assessment-Systemic Lupus Erythematosus Disease Activity Index; SD, Standard Deviation.

Table 3.

BLISS-52 and BLISS-76 results: primary endpoint – Systemic lupus erythematosus Responder Index (SRI) at week 52 [Navarra et al. 2011; Furie et al. 2010].

BLISS-52	Placebo (n = 287)	1 mg/kg (n = 288)	10 mg/kg (n = 290)
SRI at week 52, n (%)	125 (43.6)	148 (51.4) (p = 0.013)	167 (57.6) (p = 0.0006)
SELENA-SLEDAI ≥4-point reduction, n (%)	132 (46)	153 (53) (p = 0.019)	169 (58) (p = 0.0024)
BILAG: no worsening (no new 1 A/2B flares), n (%)	210 (73)	226 (78) (p = 0.1064)	236 (81) (p = 0.0181)
PGA: no worsening (<0.3 point increase), n (%)	199 (69)	227 (79) (p = 0.0078)	231 (80) (p = 0.0048)
BLISS-76	Placebo (n = 275)	1 mg/kg (n = 271)	10 mg/kg (n = 273)
SRI at week 52, n (%)	93 (33.5)	110 (40.6) (p = 0.104)	118 (43.2) (p = 0.021)
SELENA-SLEDAI ≥4-point reduction, n (%)	98 (36)	116 (43) (p = 0.087)	128 (47) (p = 0.006)
BILAG: no worsening (no new 1 A/2B flares), n (%)	179 (76)	203 (75) (p = 0.011)	189 (69) (p = 0.319)
PGA: no worsening (<0.3 point increase), n (%)	173 (63)	197 (73) (p = 0.012)	189 (69) (p = 0.126)

BILAG: British Isles Lupus Activity Group; BLISS, A Study of Belimumab in Subjects With Systemic Lupus Erythematosus; PGA, Physician’s Global Assessment; SELENA-SLEDAI, Safety of Estrogens in Lupus Erythematosus National Assessment-Systemic Lupus Erythematosus Disease Activity Index.

Figure 2.

Systemic Lupus Erythematosus Responder Index (SRI) for BLISS-52, BLISS 76, and BLISS 52/76 combined results at week 52 [Navarra et al. 2011; van Vollenhoven et al. 2011; Furie et al. 2010].

*Statistically significant difference from placebo.

Table 4.

BLISS-52 and BLISS-76 results: selected secondary endpoints [Navarra et al. 2011; Furie et al. 2010].

BLISS-52	Placebo (n = 287)	1 mg/kg (n = 288)	10 mg/kg (n = 290)
Patients reduced prednisone by ≥25% to ≤7.5 mg/day during weeks 40–52 (%)	12	21 (p = 0.025)	19 (p = 0.053)
Patients increased prednisone to >7.5 mg/day at week 52 from ≤7.5 mg/day (%)	36	30 (p = 0.558)	20 (p = 0.019)
Patients with flare (SFI) during week 52 (%)	80	70 (p < 0.01)*	71 (p < 0.03)*
Patients with severe flare (SFI) during week 52 (%)	23	18 (p = 0.134)	14 (p = 0.006)
BLISS-76	Placebo (n = 275)	1 mg/kg (n = 271)	10 mg/kg (n = 273)
Patients reduced prednisone by ≥25% to ≤7.5 mg/day during weeks 40–52 (%)	13	19 (p = 0.203)	17 (p = 0.532)
Patients increased prednisone to >7.5 mg/day at week 52 from ≤7.5 mg/day (%)	32	23 (p = 0.109)	29 (p = 0.609)
Patients with flare (SFI) during week 52 (%)	83	79 (p = 0.232)	79 (p = 0.480)
Patients with severe flare (SFI) during week 52 (%)	24	16 (p = 0.023)	18 (p = 0.087)

^*p-Value calculated by authors using chi-square test based on available data.

BLISS, a Study of Belimumab in Subjects With Systemic Lupus Erythematosus. SFI, SELENA-SLEDAI (Safety of Estrogens in Lupus Erythematosus National Assessment-Systemic Lupus Erythematosus Disease Activity Index) Flare Index.

The results of BLISS-76 were announced on 2 November 2009 [Furie et al. 2010; van Vollenhoven et al. 2010]. This trial of 819 patients took place in North America and Europe, and was identical in design to the BLISS-52 trial, except for the longer study period of 76 weeks. The primary outcome, defined by SRI at week 52, showed response rates of 43.2% (p = 0.021), 40.6% (p = 0.104) and 33.5% for the belimumab 10 mg/kg, 1 mg/kg, and placebo group, respectively (Table 3, Figure 2). Only the higher dose of belimumab (10 mg/kg) was significantly more effective than placebo. Secondary outcomes, including ability to reduce steroid dosage, also favored the active drug but were not as statistically significant as they were in the BLISS-52 trial (Table 4). Notably, a significant proportion of patients treated with the lower belimumab dose (1 mg/kg) experienced fewer severe flares compared with placebo. The drug was tolerated well and its safety profile, including the infection rate, was comparable to the placebo arm.

The pooled data analysis of 1684 patients from both the BLISS-52 and BLISS-76 trials confirmed the clinical response, showing that the SRI response rate remained higher among patients receiving belimumab 10 mg/kg (50.6%) (p < 0.0001) and 1 mg/kg (46.2 %) (p = 0.006) versus placebo (38.8%) (Figure 2) [van Vollenhoven et al. 2011]. Furthermore, among the proportion of patients from both trials (58%) who were on corticosteroids at a dose greater than 7.5 mg/day at baseline, a statistically greater percentage of the belimumab-treated group were able to reduce the corticosteroids by at least 25% to up to 7.5 mg/day during weeks 40–52 compared with placebo [placebo, 12%, belimumab 1 mg/kg, 20% (p = 0.0097), and belimumab 10 mg/kg, 18% (p = 0.045)] [Cervera et al. 2011]. Notably, the rates of severe flares in the second half of the trial were almost half as likely in the actively treated group. Clinical responses were accompanied by an increase in depressed complement levels in a dose-dependent manner, and seroconversion from positive to negative titers of several autoantibodies [anti-dsDNA: placebo 6.8% versus belimumab 10 mg/kg 16.0% (p < 0.001), anti-Sm 18.3% versus 32.1%, respectively (p < 0.05), anti-ribosomal-P 21.6% versus 51.7%, respectively (p < 0.001), and anticardiolipin-IgG 40.0% versus 55.8%, respectively (p < 0.05)] [Hiepe et al. 2011]. Among the SELENA-SLEDAI domains, the majority of patients had mucocutaneous (82%), immunologic (80%), and musculoskeletal (65%) system involvement, in which the belimumab-treated group had improvement in all of these organ systems at week 52 [Cruz et al. 2011].

Given that belimumab is now available for patients with lupus who have active disease despite standard medications, physicians should be aware of the ‘warning/precautions’ included in the package insert [Human Genome Sciences, Inc., 2011]. Although postmarketing experience will better define the safety profile of belimumab, pooled data from the phase II/III trials showed that belimumab achieved a favorable safety profile. Serious/severe infection rates were comparable across all groups, and there was no difference between overall adverse event rates and serious adverse event rates resulting in discontinuation compared with placebo [Wallace et al. 2011].

Other targeted biological therapies for systemic lupus erythematosus under development

There are other promising biological interventions that have been developed based on our improved understanding of lupus pathogenesis. These potential future targets, along with anti-B-cell agents, include costimulatory molecules and cytokines such as the IFN-I signaling pathway.

Atacicept is a soluble, recombinant fusion protein of the humanized Fc portion of IgG and the TACI receptor (one of the BLyS binding receptors) [Ramanujam et al. 2004]. In contrast to belimumab, atacicept binds and neutralizes both BLyS and APRIL, and thereby inhibits their action on B cells. The effects of atacicept on murine models of SLE have shown delayed onset of SLE disease and reduction in B-cell populations, however its broader impact on B-cell lineage in humans remains to be seen. The phase I clinical trial of atacicept in patients with moderate SLE has shown no significant difference in adverse events between the atacicept- and placebo-treated groups [Dall’Era et al. 2007]. Dose-dependent reductions of mature and total B cells and Ig levels were demonstrated in the atacicept group. Despite limited numbers (n = 12), patients with SLE in the treatment group showed a trend toward clinical improvement as assessed by SELENA-SLEDAI scores. The phase II/III clinical trial is currently active (NCT00624338), assessing the efficacy of atacicept in reducing the number of flares in patients with SLE.

Epratuzumab is an anti-CD22 fully humanized, monoclonal antibody which modulates B-cell function leading to B-cell apoptosis [Dorner et al. 2006]. Two phase II randomized controlled trials were both terminated early due to manufacturing shortages. However, the limited analysis of their results suggested that patients with moderate to severe SLE receiving epratuzumab showed improvement in their BILAG scores, a clinically significant reduction of their steroid doses, and an acceptable safety profile. Currently, two phase III multicenter (approximately 300 participating sites in 30 countries), double-blinded, placebo-controlled studies (EMBODY 1 and EMBODY 2) are currently active and enrolling patients with moderate to severe SLE to further investigate the efficacy, safety and tolerability, and immunogenicity of epratuzumab (NCT01261793/NCT01262365).

Abatacept, a cytotoxic T-lymphocyte antigen 4 (CTLA-4) Ig that works as a T-cell costimulation blocker, has already been shown to be efficacious in rheumatoid arthritis and more recently psoriatic arthritis has also been studied in patients with nonrenal SLE. However, a phase II trial (n = 175) of patients with nonrenal SLE (skin, pleurisy, arthritis involvement) showed no significant difference in primary outcome of new SLE flare (defined as new BILAG A or BILAG B score) between the abatacept (79.7%) and placebo (82.5%) groups [Merrill et al. 2010]. Currently, there is an ongoing phase II multicenter study to evaluate the addition of abatacept to standard cyclophosphamide therapy compared with cyclophosphamide alone for the treatment of lupus nephritis (NCT00774852).

Several anti-IFN-I trials have been launched, including studies of the monoclonal antibodies, sifalimumab and rontalizumab. Of note, these agents cause potent inhibition of IFN-I-inducible genes in the blood and skin of treated patients [Yao et al. 2009]. They are currently in phase II studies of development and results are awaited with great interest.

Conclusions

BLyS is one of the important players in B-cell development and pathogenesis in SLE; its overexpression in murine models of SLE and the persistently or intermittently elevated serum levels in patients with SLE suggest the apparent link between BLyS and lupus pathogenesis. The extent to which disease activity is driven by the pathogenic B-cell response to apoptosis versus survival, and the extent to which BLyS serves as a reliable biomarker for SLE disease activity is under investigation. Belimumab offers a new therapeutic option for patients with SLE who have active disease despite existing therapies. More data are needed before this medication can be recommended for the therapy of more severe lupus disease, especially those with nephritis and neurologic disease. Equally prominent is that these trials have opened the way for better designed lupus clinical trials and it is expected that other novel therapeutic agents will follow suit soon.