Reduced B lymphocyte and immunoglobulin levels after atacicept treatment in patients with systemic lupus erythematosus: Results of a multicenter, phase ib, double-blind, placebo-controlled, dose-escalating trial

Authors


Abstract

Objective

To assess the safety and tolerability of atacicept in patients with systemic lupus erythematosus (SLE) and the biologic effect of atacicept on B lymphocyte and immunoglobulin levels. Atacicept is a TACI-Ig fusion protein that inhibits B cell stimulation by binding to B lymphocyte stimulator and a proliferation-inducing ligand.

Methods

This phase Ib, double-blind, placebo-controlled, dose-escalating trial comprised 6 cohorts of patients treated with atacicept or placebo in a 3:1 ratio of active drug to placebo (n = 8 per group; n = 7 in cohort 5). Cohorts 1–4 received a single subcutaneous dose of placebo or either 0.3 mg/kg, 1 mg/kg, 3 mg/kg, or 9 mg/kg of atacicept. Cohorts 5 and 6 received weekly doses of placebo or either 1 mg/kg or 3 mg/kg of atacicept for 4 weeks. Patients were followed up for 6 weeks (cohorts 1–4) or 9 weeks (cohorts 5 and 6). Patients with mild-to-moderate SLE were enrolled.

Results

Biologic activity of atacicept was demonstrated by dose-dependent reductions in immunoglobulin levels and in mature and total B cell numbers. This effect was most pronounced in the repeated-dose cohorts and was sustained throughout the followup period. There were no changes in the numbers of T cells, natural killer cells, or monocytes. Mild injection-site reactions occurred more frequently among the atacicept group than the placebo group. There were no differences in the frequency or type of adverse events and no severe or serious adverse events in patients treated with atacicept.

Conclusion

Atacicept administered subcutaneously was well tolerated and demonstrated biologic activity consistent with the proposed mechanism of action.

Systemic lupus erythematosus (SLE) is a prototypical autoimmune disease characterized by the production of autoantibodies to a variety of nuclear antigens. B cells are currently thought to play an important role in SLE pathogenesis through both antibody-dependent and antibody-independent mechanisms. Thus, B cells have emerged as rational targets for drug development in SLE (1).

Several B cell–directed strategies have been proposed as possible therapies for SLE. Some of these strategies are designed to eliminate B cells through the use of B cell–directed monoclonal antibodies (mAb) (2–5), while others interfere with B cell stimulation (6–8) or seek to selectively target autoantibody-producing B cells (9–11). Attempts to inhibit B cell stimulation have focused primarily on receptor–ligand interactions that involve molecules called B lymphocyte stimulator (BLyS; trademark of Human Genome Sciences, Rockville, MD) and APRIL. BLyS and APRIL are members of the tumor necrosis factor family of cytokines, which are critical for B cell survival and development. Both molecules bind to TACI and BCMA, while BLyS also binds to BAFF receptor (BAFF-R) and APRIL interacts with proteoglycans.

Mounting evidence in animal models and in humans supports an important role of BLyS and APRIL in the development of autoimmune disease. Transgenic mice that overexpress BLyS display B cell expansion and polyclonal hypergammaglobulinemia (12–14). Some of these mice develop a lupus-like phenotype consisting of anti–double-stranded DNA (anti-dsDNA) antibodies, immunoglobulin deposition in the kidneys, and accelerated development of glomerular disease (15).

Studies in humans also suggest a role of BLyS and APRIL in systemic autoimmune diseases. Patients with SLE have increased serum levels of BLyS that correlate positively with levels of anti-dsDNA antibodies (16–18). Serum levels of APRIL are elevated in patients with SLE as compared with healthy individuals and patients with rheumatoid arthritis (19). BLyS and APRIL have been detected in the synovial fluid of patients with inflammatory arthritis (20).

These compelling observations in mice and humans have led to the development of several BLyS antagonists. One of these agents, atacicept (previously referred to as TACI-Ig), is a recombinant fusion protein comprising the extracellular domain of the TACI receptor joined to a human IgG1 Fc domain. Atacicept blocks B cell stimulation by both BLyS and APRIL. Several lines of investigation provide support for the expectation that atacicept will have potent effects in vivo. First, transgenic mice that express atacicept have few mature B cells and reduced concentrations of immunoglobulin (8). Second, treatment of lupus-prone female (NZB × NZW)F1 (NZB/NZW) mice with atacicept delays the development of proteinuria and increases survival (12). Finally, in a direct comparison of the efficacy of murine atacicept and BAFF-R-Ig (a BLyS-only inhibitor) in lupus-prone female NZB/NZW mice, only atacicept reduced the serum levels of IgM, decreased the frequency of plasma cells in the spleen, and inhibited the IgM response to a T cell–dependent antigen, suggesting a role of APRIL in these processes (21). In light of these encouraging preclinical data, we examined the biologic effects, pharmacokinetics, pharmacodynamics, and safety of atacicept in a phase Ib, double-blind, dose-escalating trial in patients with SLE.

PATIENTS AND METHODS

Study design.

This study (study code 25050) was a multicenter, phase Ib, placebo-controlled, dose-escalating, single- and repeated-dose trial of atacicept in patients with mild-to-moderate SLE. Informed consent was obtained from all patients in accordance with the human subjects Institutional Review Boards of all the participating universities.

Six cohorts comprising 8 patients each were treated with subcutaneously administered atacicept or matching placebo in a 3:1 ratio of those taking active drug and those taking placebo (an additional patient was enrolled in cohort 5 to replace a patient who was withdrawn because of a protocol violation). Cohorts 1–4 received a single subcutaneous dose of placebo or 0.3 mg/kg, 1 mg/kg, 3 mg/kg, or 9 mg/kg of atacicept, respectively. Cohorts 5 and 6 received subcutaneous doses of placebo or 1 mg/kg or 3 mg/kg of atacicept, respectively, once a week for 4 weeks. Patients were followed up for 6 weeks (cohorts 1–4) or 9 weeks (cohorts 5 and 6). Outcome measures included: systemic and local tolerability of atacicept; frequency of adverse events (AEs); pharmacokinetics and pharmacodynamics of atacicept, including effects on lymphocyte subpopulations and immunoglobulin levels; and measures of SLE disease activity.

Selection of study patients.

All patients enrolled in the study were required to have fulfilled at least 4 of the 11 classification criteria for SLE as defined by the American College of Rheumatology (22) and as updated in 1997 (23). Patients also were required to be between the ages of 18 and 70 years, with a body mass index of 18–40 kg/m2. Patients with Safety of Estrogens in Lupus Erythematosus: National Assessment modification of the Systemic Lupus Erythematosus Disease Activity Index (SELENA–SLEDAI) scores of 0–10 at screening were eligible for the trial (24, 25). Patients with a SELENA–SLEDAI score >10 were excluded, based on the rationale that their baseline therapy should be modified to provide better disease control. Patients treated with immunosuppressive medications, such as azathioprine, methotrexate, mycophenolate mofetil, and cyclophosphamide, during the 8 weeks prior to enrollment were excluded. In addition, patients treated with >20 mg/day of prednisone or who were experiencing a change in prednisone dosage during the 4 weeks prior to the start of the trial were excluded. For patients being treated with hydroxychloroquine or nonsteroidal antiinflammatory drugs, stable dosages were required during the 4 weeks prior to the trial. Patients were required to meet the following hematologic criteria for safety: hemoglobin value >8.5 mg/dl, white blood cell count >2.5 × 109/liter, and platelet count >75 × 109/liter. Patients with neurologic symptoms suggestive of central nervous system lupus, congestive heart failure, a history of cancer other than treated basal cell or squamous cell carcinoma of the skin, or the presence of significant liver or kidney disease were excluded from the study. Patients were also excluded if they had been previously treated with biologic agents or had a history of recurrent or active infections, such as human immunodeficiency virus, tuberculosis, hepatitis B virus, or hepatitis C virus.

Assessment of pharmacokinetics.

The pharmacokinetics of the study medication were assessed by measuring serum levels of free atacicept, atacicept–BLyS complex, and composite atacicept (defined as free atacicept plus atacicept–BLyS complex). Serum levels of each of these components were quantified using an enzyme-linked immunosorbent assay (ELISA). Serum was incubated with a biotin-conjugated mAb specific for atacicept (free or total atacicept detection) or BLyS (atacicept–BLyS complex detection) immobilized on a streptavidin-coated microplate. After washing, an atacicept-specific mAb conjugated to horseradish peroxidase (HRP) was added. For detection of total atacicept, a BLyS-specific mAb conjugated to HRP was also added at this stage. In all 3 assays, serum atacicept levels were detected and quantified using standard chemiluminescence methods.

Assessment of pharmacodynamics.

The pharmacodynamics of the study medication were assessed by measuring serum levels of immunoglobulins (IgG, IgM, IgA), C3 complement, and antinuclear antibodies (ANAs), as determined by flow cytometric analysis of lymphocyte subsets (see below). Levels of immunoglobulins and C3 were measured using standard methods. ANAs were measured using the AtheNA Multi-Lyte ANA test system (Zeus Scientific, Raritan, NJ).

Flow cytometry.

A panel of peripheral blood mononuclear cell types (B cell and T cell subsets, natural killer [NK] cells, and monocytes) was assessed in antibody-stained peripheral blood samples, using 4-color flow cytometry. The analysis included: total T cells (CD45+,CD3+), T helper cells (CD45+,CD3+,CD4+,CD8–), T cytotoxic/suppressor cells (CD45+,CD3+,CD4–,CD8+), total B cells (CD19+), mature B cells (CD19+,IgD+,CD27–), monocytes (CD45+,CD3–,CD14+,CD56–), and NK cells (CD45+,CD3–,CD14–,CD56+). A contract research organization (Esoterix, Groningen, The Netherlands) performed blood sample processing, antibody staining, and acquisition, analysis, and quality control of data. ZymoGenetics (Seattle, WA) performed further analysis and quality control on B cell subsets. For B cell subsets, the analysis gate was enlarged to include small and large lymphocytes, with the latter being similar in size to monocytes.

Measurement of antitetanus antibodies.

In the repeated-dose cohorts only, vaccine immunization status was assessed by measuring titers of antibodies to tetanus toxoid on days 1 and 29 and at the poststudy visit.

Measurement of the immune response to atacicept.

Assays for binding and neutralizing antibodies to atacicept were performed on samples taken at baseline and at the final poststudy visit. Quantification of anti-atacicept antibodies was performed using a bridging ELISA based on streptavidin-precoated plates. For each unknown and control sample, the T:U ratio was calculated (T = treated, representing postdose or spiked sample [control samples only], and U = untreated, representing predose sample). Test samples were considered positive for the presence of antibody to atacicept if the T:U ratio was ≥1.3, corresponding to a minimum sensitivity of 250 ng of control antibody per milliliter (rabbit IgG to atacicept).

Clinical assessments.

Medical history was obtained at study inclusion, and a physical examination was conducted on a weekly basis. Hematologic and serum chemistry profiles were performed on a weekly basis and were evaluated using the Common Toxicity Criteria of the National Cancer Institute (26). Blood samples for pharmacokinetic evaluations were collected on a weekly basis for repeated-dose cohorts and on day 1 at 4 and 8 hours after dosing, days 2, 3, 4, and 8, and on a weekly basis thereafter for the single-dose cohorts. Blood samples for pharmacodynamic evaluations were drawn on a weekly basis in the repeated-dose cohorts and on days 2, 3, and 8 and on a weekly basis thereafter in the single-dose cohorts. Electrocardiogram after day 4 was performed every 2 weeks in the single-dose cohorts and on a weekly basis in patients receiving repeated doses of the study drug.

For practical reasons, there was a difference in collection of local tolerability data between the single-dose and repeated-dose cohorts. Local tolerability of injections was a solicited event and was actively assessed in the single-dose cohorts. Clinically significant local tolerability reactions were to be reported as AEs. In repeated-dose cohorts, local reactions were reported as AEs, but injection sites were not routinely evaluated.

Measures of disease activity.

Although the study was not powered to determine the impact of treatment on disease activity, the following disease activity measurements were obtained to provide preliminary efficacy data. SELENA–SLEDAI scores were determined at baseline and on days 29 and 43 (cohorts 1–4) or on days 22 and 64 (cohorts 5 and 6). Anti-dsDNA antibody and C3 levels were measured at baseline and on days 15, 29, and 43 (cohorts 1–4) or on days 15, 22, 29, 43, and 64 (cohorts 5 and 6).

RESULTS

Clinical characteristics.

Forty-nine patients with mild-to-moderate SLE participated in the study from June 2004 through March 2006 at 5 centers. Characteristics of the study patients are shown in Tables 1 and 2. Among the patients in the single-dose cohorts, 90.6% were female, 71.9% were Caucasian, and their median age was 45.5 years (range 23–64 years). Within the repeated-dose cohorts, 100% were female, 70.6% were Caucasian, and their median age was 49.0 years (range 26–64 years). Clinical manifestations of SLE at baseline included the following: arthritis (90%), photosensitivity (65%), malar rash (51%), oral ulcers (47%), serositis (41%), hematologic (35%), discoid lupus (31%), renal involvement (16%), and neurologic involvement (10%). The median duration of disease was 11.2 years in the single-dose cohorts and 11.0 years in the repeated-dose cohorts. At study entry, 36% of patients were receiving corticosteroids, and 59% of patients were receiving hydroxychloroquine. Stable dosages of these medications were continued for the duration of the study.

Table 1. Baseline demographic and clinical characteristics of patients in the single-dose cohorts*
 Placebo (n = 8)AtaciceptAll (n = 32)
0.3 mg/kg (n = 6)1 mg/kg (n = 6)3 mg/kg (n = 6)9 mg/kg (n = 6)
  • *

    Except where indicated otherwise, values are the median (range). BMI = body mass index; ACR = American College of Rheumatology; SELENA–SLEDAI = Safety of Estrogens in Lupus Erythematosus: National Assessment modification of the Systemic Lupus Erythematosus Disease Activity Index.

Age, years43.5 (26–61)46.5 (23–64)35.0 (24–54)55.5 (44–63)41.0 (30–60)45.5 (23–64)
Female, no. (%)8 (100)6 (100)6 (100)5 (83.3)4 (66.7)29 (90.6)
BMI, kg/m225.0 (18.1–39.6)27.9 (20.2–36.9)23.6 (21.3–27.7)26.8 (23.2–31.1)22.6 (20.8–32.1)25.4 (18.1–39.6)
Race, no. (%)      
 Caucasian6 (75.0)5 (83.3)3 (50.0)4 (66.7)5 (83.3)23 (71.9)
 Black1 (12.5)00001 (3.1)
 Asian002 (33.3)1 (16.7)03 (9.4)
 Other1 (12.5)1 (16.7)1 (16.7)1 (16.7)1 (16.7)5 (15.6)
No. of ACR criteria met,  no. (%) of patients      
 45 (62.5)2 (33.3)2 (33.3)3 (50.0)2 (33.3)14 (43.8)
 52 (25.0)2 (33.3)2 (33.3)02 (33.3)8 (25.0)
 61 (12.5)1 (16.7)001 (16.7)3 (9.4)
 701 (16.7)1 (16.7)1 (16.7)03 (9.4)
 80001 (16.7)1 (16.7)2 (6.3)
 9001 (16.7)001 (3.1)
 100001 (16.7)01 (3.1)
Disease duration, years10.3 (1.2–28.6)9.6 (3.4–34.6)12.4 (0.4–36)12.3 (5.1–25.2)5.0 (0.8–46.5)11.2 (0.4–46.5)
SELENA–SLEDAI score2.0 (0–7)2.0 (0–7)2.0 (0–8)1.0 (0–6)3.0 (2–8)2.0 (0–8)
Table 2. Baseline demographic and clinical characteristics of patients in the repeated-dose cohorts*
 Placebo once weekly for 4 weeks (n = 4)AtaciceptAll (n = 17)
1 mg/kg once weekly for 4 weeks (n = 7)3 mg/kg once weekly for 4 weeks (n = 6)
  • *

    Except were indicated otherwise, values are the median (range). BMI = body mass index; ACR = American College of Rheumatology; SELENA–SLEDAI = Safety of Estrogens in Lupus Erythematosus: National Assessment modification of the Systemic Lupus Erythematosus Disease Activity Index.

Age, years59.0 (42–63)51.0 (26–64)42.5 (36–49)49.0 (26–64)
Female, no. (%)4 (100)7 (100)6 (100)17 (100)
BMI, kg/m232.7 (28.4–36.9)25.3 (21.8–42.6)28.1 (17.7–39.3)29.1 (17.7–42.6)
Race, no. (%)    
 Caucasian4 (100)4 (57.1)4 (66.7)12 (70.6)
 Black01 (14.3)1 (16.7)2 (11.8)
 Other02 (28.6)1 (16.7)3 (17.6)
No. of ACR criteria met,  no. (%) of patients    
 42 (50.0)1 (14.3)2 (33.3)5 (29.4)
 51 (25.0)4 (57.1)2 (33.3)7 (41.2)
 6001 (16.7)1 (5.9)
 71 (25.0)1 (14.3)1 (16.7)3 (17.6)
 801 (14.3)01 (5.9)
Disease duration, years13.3 (6–26)16.8 (2.8–36.8)9.0 (5–27)11.9 (2.8–36.8)
SELENA–SLEDAI score1 (0–6)2 (0–6)4 (0–6)3 (0–6)

Disposition of the study patients.

Thirty-two patients were enrolled in the single-dose cohorts and 17 patients were enrolled in the repeated-dose cohorts. All of the patients except for 2 completed the trial. One patient in the 1-mg/kg atacicept repeated-dose group (cohort 5) withdrew and was replaced after day 8, when a protocol violation was discovered (history of melanoma). One patient in the repeated-dose placebo group in cohort 6 was lost to followup after day 8.

Atacicept pharmacokinetics.

Evidence of nonlinear pharmacokinetics, consistent with saturable binding pharmacokinetics of ligand–receptor interactions, was demonstrated (Figure 1). Free and composite atacicept concentration–time profiles displayed multiphasic pharmacokinetics with fairly rapid absorption, with a time to maximum concentration of ∼24 hours after the first dose, and an initial distribution phase lasting 7–14 days. Low accumulation of free atacicept was observed in the repeated-dose cohort; the accumulation of composite atacicept was marginally higher and the atacicept–BLyS complex was found to accumulate throughout the dosing period.

Figure 1.

Pharmacokinetics of free atacicept when administered as A, a single dose or B, repeated doses in patients with systemic lupus erythematosus. Pharmacokinetic parameters were measured before each dose and no sooner than 1 week after the last dose (see Patients and Methods for details). Values are the median.

Effect of atacicept on B lymphocytes.

Treatment with atacicept was associated with an initial, transient increase in mature and total B cells, followed by a sustained, dose-related reduction (Figure 2). In the single-dose 3 mg/kg and 9 mg/kg groups (data not shown) and in the repeated-dose groups (Figure 2), a reduction in mature B cells of ∼35% from baseline was seen on day 29. In the single-dose groups, this reduction was sustained through day 43; in the repeated-dose groups, a reduction of ∼60% was seen on day 43 and was sustained at 45–60% through to the last assessment on day 64. The patterns observed for total B cells were similar to those for mature B cells. In the 3-mg/kg single-dose group (data not shown), a reduction in total B cells of ∼30% from baseline was seen on day 29, which was sustained through day 43; in the repeated-dose groups, reductions of ∼40–50% were seen on day 43 and were sustained at 35–60% through to the last assessment on day 64 (Figure 2). There were no significant changes in the number of total T cells, T helper cells, T suppressor/cytotoxic cells, monocytes, or NK cells.

Figure 2.

Effect of atacicept versus placebo on B lymphocyte numbers in patients with systemic lupus erythematosus. A, Mature B cells (CD19+,IgD+,CD27–), gated on small and large lymphocytes. B, Total B cells (CD19+), gated on small and large lymphocytes. Values are the median percentage change in absolute numbers from baseline.

Effect of atacicept on immunoglobulin levels.

Dose-dependent reductions in immunoglobulin levels were observed in the atacicept-treated patients (Figure 3). This effect was most notable in the repeated-dose groups. IgM levels showed the greatest declines with treatment, reaching nearly 50% on day 43 in the 3-mg/kg repeated-dose group. IgA levels decreased by ∼33% in the 3-mg/kg repeated-dose group on day 29, and IgG levels decreased by ∼16% in the 3-mg/kg repeated-dose group on day 36. Nadirs occurred between days 15 and 29 in the single-dose cohorts and between days 29 and 43 in the repeated-dose cohorts. Thereafter, values began to return to baseline. The last observed values were ∼5–30% below baseline in the single-dose cohorts (with the exception of the 0.3-mg/kg group, in which IgM values were above baseline) and 8–45% below baseline in the repeated-dose cohorts.

Figure 3.

Levels of immunoglobulin present after treatment with atacicept in patients with systemic lupus erythematosus. A, IgM levels in patients receiving single or repeated doses. 1+ 8h = 8 hours after the dose. B, IgG and C, IgA levels in patients receiving repeated doses. Values are the median percentage change in absolute levels from baseline.

IgM levels dropped below the lower limit of normal during treatment (range 0.31–0.39 mg/ml) in a total of 6 patients (1 in each cohort). However, for all of these patients, levels were already low at baseline, varying from 0.42 to 0.81 mg/ml (normal range 0.4–2.3). IgM values had returned to normal by the end of the study for 3 of the patients who received single doses.

In terms of IgG levels, a decrease below the lower limit of normal occurred only in 1 patient; this patient had received a single 0.3-mg dose. IgG levels dropped from 7.55 mg/ml at baseline (normal range 7–16) to 6.31 mg/ml on day 8. IgG levels in this patient had returned to normal by day 15. It is important to note that 7 patients had IgG levels below the lower limit of normal at baseline and that these levels decreased during treatment. However, no patient had IgG levels below 4 mg/ml at any point during the study.

Adverse events.

There were no statistical differences in the frequency or type of AEs, including infections, between the placebo and atacicept groups (Table 3). At least 1 AE was reported by 78% of patients in the single-dose cohorts and by 77% of patients in the repeated-dose cohorts. All reported events except 1 were graded as mild or moderate in severity: 107 events were graded as mild, 33 were graded as moderate, and 1 was graded as severe (paresthesia in a patient in the repeated-dose placebo group). There were no treatment withdrawals or withdrawals from the trial because of AEs.

Table 3. Types of adverse events occurring in ≥3 patients
Adverse eventNo. (%) of atacicept-treated patients (n = 37)No. (%) of placebo-treated patients (n = 12)
Rhinitis5 (13.5)0
Fatigue5 (13.5)1 (8.3)
Nausea5 (13.5)1 (8.3)
Headache4 (10.8)0
Upper respiratory tract infection3 (8.1)3 (25.0)
Peripheral edema3 (8.1)0
Arthritis3 (8.1)2 (16.7)
Arthralgia3 (8.1)0
Prolonged prothrombin time3 (8.1)0
Dizziness2 (5.4)1 (8.3)
Depression2 (5.4)1 (8.3)
Sinusitis2 (5.4)1 (8.3)
Vaginal mycosis2 (5.4)1 (8.3)
Pain in extremity2 (5.4)1 (8.3)

Overall, 42% of patients receiving placebo and 35% of patients receiving atacicept experienced an infection-related event. The most frequently reported infectious AEs were upper respiratory tract infections (6 patients, 3 receiving placebo and 3 receiving atacicept), rhinitis (none receiving placebo and 5 receiving atacicept), and sinusitis (1 receiving placebo and 2 receiving atacicept). No infection-related event was reported as severe or serious, and all patients recovered without complications. Two patients experienced serious AEs; both were in the placebo groups. One patient was hospitalized because of ventricular bigeminy and pyrexia, and the other patient was hospitalized because of acute right-sided paresthesia. There were no severe or serious AEs in patients treated with atacicept.

Local tolerability.

For the single-dose cohorts, the injection sites were assessed systematically; symptoms were noted, but none was reported as an AE. In the repeated-dose cohorts, local tolerability was assessed through AE reporting. All of the injection-site AEs reported, therefore, occurred in patients in the repeated-dose cohorts. Overall, mild injection-site reactions occurred more frequently in patients receiving atacicept than in patients receiving placebo. Mild injection-site redness was observed in 2 of the 8 patients in the single-dose placebo group. Mild-to-moderate injection-site redness, bruising, and swelling were noted in all of the atacicept single-dose groups, without notable differences between doses except for redness, which was less frequent in the 0.3-mg/kg group (affecting only 1 patient). Injection-site redness was observed in 50% of all of the patients who were receiving atacicept. No itching was reported, and no severe injection-site reactions were observed. Three local injection-site reactions were reported as AEs: a mild injection-site hemorrhage in a patient in the placebo group and moderate injection-site rashes in 2 patients in the repeated-dose atacicept group.

SLE exacerbations.

A total of 4 exacerbations of SLE were reported during the trial. One patient in the placebo group experienced increased arthralgias and was treated with corticosteroids and adrenocorticotropic hormone. One patient in the atacicept 0.3-mg/kg group experienced increased arthritis and rash 11 days after dosing; this was treated with acetaminophen, corticosteroids, and azathioprine. One patient in the atacicept 1-mg/kg group experienced a new mouth ulcer 4 weeks after dosing; no treatment was required. One patient in the atacicept 3-mg/kg repeated-dose group experienced pleuritic chest pain 3 days after the first dose, which required treatment with ibuprofen.

Findings of laboratory and other assessments.

There was no evidence of hematologic, hepatic, or renal toxicity in any patient treated with atacicept. There were no differences in blood pressure, heart rate, temperature, or electrocardiographic findings between the placebo and atacicept groups.

Anti-atacicept antibody response.

No patient developed detectable binding antibodies to atacicept.

Antitetanus protective antibodies.

Twelve subjects (3 assigned to placebo and 9 assigned to atacicept) had detectable levels of antitetanus antibodies at baseline. Levels posttreatment were available in 11 of these patients. None of these patients experienced a decrease in antibody levels into the nonprotective range. Antibody titers of <0.1 IU/ml are considered nonprotective, and levels >0.49 IU/ml are considered protective; patients with titers falling between these values are said to be in the intermediate range.

Five patients receiving atacicept and 2 receiving placebo who had protective levels of antibodies at baseline continued to have levels within the protective range. One patient receiving atacicept who had a protective level of antibodies at baseline showed intermediate-range levels on day 29, which then returned to the protective range on day 64. Two patients receiving atacicept who had baseline levels of antibodies in the intermediate range continued to have levels within the intermediate range. One patient receiving atacicept who had baseline levels of antibodies in the intermediate range showed protective levels on days 29 and 64. One patient receiving atacicept who had nonprotective levels of antibodies at baseline and on day 29 had levels in the protective range on day 64.

Effect of atacicept on SELENA–SLEDAI scores.

The majority of patients had mild SLE disease activity at baseline; the median SELENA–SLEDAI score was 2.0 in the single-dose cohorts and 3.0 in the repeated-dose cohorts. However, 12 patients (3 receiving placebo, 6 receiving single-dose atacicept, and 3 receiving repeated-dose atacicept) had baseline SELENA–SLEDAI scores of ≥6, denoting moderate disease activity. To evaluate a possible trend in the effect of atacicept on disease activity, we determined how many of those 12 patients experienced a reduction of >3 points in the SELENA–SLEDAI score by the end of the trial. Our choice of 3 points was based on a study of 230 patients with SLE, which determined that meaningful improvement in a patient's clinical disease activity correlated with a reduction in the SELENA–SLEDAI score of >3 (27). Using this criterion, we observed that 1 of the 3 patients from the placebo group, 2 of the 6 patients from the single-dose cohorts, and 2 of the 3 patients from the repeated-dose cohorts experienced clinically meaningful improvements in disease activity.

We also determined how many of the 12 patients with baseline SELENA–SLEDAI scores of ≥6 achieved a SELENA–SLEDAI score of 0 at the end of the study. Interestingly, 2 of the 3 patients in the repeated-dose cohorts achieved a SELENA–SLEDAI score of 0, compared with 2 of the 6 patients in the single-dose cohorts and none of the 3 patients in the placebo cohorts. While these data are encouraging, they are insufficient to draw conclusions regarding efficacy, which will be assessed in a subsequent phase II/III trial.

Effect of atacicept on C3 complement levels.

Only 8 patients in this study had low C3 levels at baseline (1 receiving placebo, 5 receiving single-dose atacicept, and 1 each receiving 1 mg/kg and 3 mg/kg repeated-dose atacicept). Hypocomplementemia persisted in all 6 of the patients receiving placebo and single-dose atacicept, whereas C3 levels normalized in both of the patients receiving repeated-dose atacicept.

Effect of atacicept on anti-dsDNA antibodies.

Since this phase Ib study was conducted in patients with relatively mild disease, only 5 patients had anti-dsDNA antibodies at baseline (2 receiving placebo and 1 each receiving single-dose 0.3 mg/kg, 1 mg/kg, and 3 mg/kg atacicept). Each of these patients also had anti-dsDNA antibodies at the end of the study. None of the patients in the repeated-dose cohorts had detectable anti-dsDNA antibodies at baseline.

DISCUSSION

The primary objectives of this study were 1) to assess the short-term safety and tolerability of single and repeated subcutaneous doses of atacicept in patients with mild-to-moderate SLE and 2) to determine the biologic effects of atacicept on B lymphocyte numbers and immunoglobulin levels. Treatment with atacicept was well tolerated, although mild injection-site reactions were observed more commonly in the atacicept groups than in the placebo group. The impact of atacicept on B cell and immunoglobulin levels documents a potent biologic effect that supports the hope that this approach may be effective in autoantibody-mediated autoimmune diseases such as SLE.

Interest in the role of B cells in the pathogenesis of SLE has led to the development of several potential therapeutic agents that specifically target B cells by distinct mechanisms. Rituximab (anti-CD20) and epratuzumab (anti-CD22) deplete B cells by mAb-mediated mechanisms. In contrast, atacicept and belimumab (anti-BLyS) deprive B cells of signals necessary for growth and development, thereby causing partial depletion of B cells and reducing immunoglobulin production.

Belimumab is a mAb to BLyS. As such, it inhibits B cell stimulation by BLyS, but not by APRIL. Belimumab was recently studied in a phase II, randomized, double-blind, placebo-controlled trial in 449 patients with active SLE (7). Although this trial did not meet its primary end points in the overall study population, subsequent subset analysis suggested possible benefit. It is expected that this issue will be clarified in 2 upcoming phase III trials. A potentially important distinguishing feature between atacicept and belimumab is the fact that atacicept binds to both BLyS and APRIL, while belimumab is a mAb directed solely against BLyS. In addition, the receptors for BLyS and APRIL are expressed differentially on B cells according to their developmental stage. TACI is strongly expressed by transitional type 2 B cells, marginal-zone B cells, and activated B cells, whereas BCMA is preferentially expressed by plasma cells, plasmablasts, and tonsillar germinal center B cells (28). Thus, the various proposed B cell–directed therapies may well differ in their potency and in their safety profiles. Atacicept may be more potent than belimumab because of its ability to block both BLyS and APRIL; however, it remains to be determined whether this difference will result in a better risk/benefit ratio. Similarly, atacicept may be safer than rituximab or epratuzumab because it might spare patients severe and prolonged B cell depletion. How this difference will translate in terms of relative risk and benefit, however, also remains to be determined in controlled clinical trials.

It is reassuring that there were no differences in the frequency of infections between the atacicept and placebo groups. The most commonly reported infectious AE was upper respiratory tract infection (3 patients receiving atacicept and 3 patients receiving placebo). Despite the overall reductions in immunoglobulin levels observed in this study, it is encouraging that all patients maintained protective levels of antitetanus antibodies. Additionally, immunogenicity against atacicept was not detected, even though the majority of patients were not receiving concomitant antimetabolic or cytotoxic therapies.

This small study was not powered to reach definitive conclusions regarding the effect of atacicept on measures of disease activity. In addition, for ethical reasons, only patients with mild-to-moderate SLE were enrolled in this phase Ib study. Nevertheless, reassuring trends were observed in complement levels and SELENA–SLEDAI scores. These very preliminary indications will be the subject of systematic investigation in a phase II/III trial.

AUTHOR CONTRIBUTIONS

Dr. Dall'Era had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study design. Dall'Era, Chakravarty, Genovese, Vincent, Pena-Rossi, Wofsy.

Acquisition of data. Dall'Era, Chakravarty, Wallace, Genovese, Kavanaugh, Kalunian, Dhar, Vincent, Pena-Rossi, Wofsy.

Analysis and interpretation of data. Dall'Era, Genovese, Kavanaugh, Kalunian, Vincent, Pena-Rossi, Wofsy.

Manuscript preparation. Dall'Era, Chakravarty, Wallace, Genovese, Weisman, Kavanaugh, Kalunian, Vincent, Pena-Rossi, Wofsy.

Statistical analysis. Vincent, Pena-Rossi.

ROLE OF THE STUDY SPONSOR

A committee composed of the study investigators and members of the Merck Serono and ZymoGenetics Study Group was responsible for the design of the study protocol. Merck Serono was responsible for collecting data from the study sites and for preparing and maintaining the study database. The study investigators had full access to the study data and were responsible for data analysis and manuscript preparation. The manuscript was reviewed by Merck Serono and ZymoGenetics prior to submission.

Acknowledgements

We thank Steve Lund, NP, and Anne Marie Duhme, RN (Clinical Trial Center, University of California, San Francisco), for their contributions to the study design and protocol development. We are indebted to the Rosalind Russell Medical Research Center for Arthritis for their support of the Clinical Trials Center at the University of California, San Francisco.

APPENDIX A:

THE MERCK SERONO AND ZYMOGENETICS ATACICEPT STUDY GROUP

Members of the Merck Serono and ZymoGenetics Atacicept Study Group are as follows: Nasreen Alam, PhD, Annette Dubois, MA, Nils Kinnman, MD, PhD, and Marie Picard, MS (Merck Serono International SA, Geneva, Switzerland); Alessandro Bortolotti, PhD (Industria Farmaceutica Serono SpA, Rome, Italy); Laura O'Grady, BS (EMD Serono, Rockland, MA); and Julieann Hill, MS, Ivan Nestorov, PhD, and Elisabeth Salmon, MS (ZymoGenetics Inc., Seattle, WA).

Ancillary