LJP 394 for the prevention of renal flare in patients with systemic lupus erythematosus: Results from a randomized, double-blind, placebo-controlled study

Authors

  • Donato Alarcón-Segovia,

    1. Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán, Mexico City, Mexico
    Search for more papers by this author
  • James A. Tumlin,

    1. Emory University School of Medicine, Atlanta, Georgia
    Search for more papers by this author
  • Richard A. Furie,

    1. North Shore University Hospital, Manhasset, New York
    Search for more papers by this author
  • James D. McKay,

    1. Tulsa, Oklahoma
    Search for more papers by this author
  • Mario H. Cardiel,

    1. Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán, Mexico City, Mexico
    Search for more papers by this author
  • Vibeke Strand,

    1. Stanford University, Palo Alto, California
    Search for more papers by this author
    • Dr. Strand serves as a consultant to LJP.

  • Robert G. Bagin,

    1. Santarus, Inc., San Diego, California
    Search for more papers by this author
  • Matthew D. Linnik,

    Corresponding author
    1. La Jolla Pharmaceutical Company, San Diego, California
    • Chief Scientific Officer and Executive Vice President, Research, La Jolla Pharmaceutical Company, 6455 Nancy Ridge Drive, San Diego, CA 92121
    Search for more papers by this author
  • Bonnie Hepburn,

    1. Santarus, Inc., San Diego, California
    Search for more papers by this author
    • Dr. Hepburn was an employee of LJP during the conduct of this study and serves as a consultant to LJP.

  • LJP 394 Investigator Consortium

    Search for more papers by this author
    • Members of the LJP 394 Investigator Consortium are as follows: Jacob Aelion, MD, Gerald B. Appel, MD, Cynthia Aranow, MD, Stanley Ballou, MD, Michael Becker, MD, Nancy J. Becker, MD, H. Michael Belmont, MD, Jill P. Buyon, MD, Anthony Bohan, MD, William G. Brelsford, MD, Nancy L. Carteron, MD, FACR, Mary E. Cronin, MD, Raphael J. DeHoratius, MD, Luis R. Espinoza, MD, Mark C. Genovese, MD, Gary Gilkeson, MD, Antonio Gil-Aguado, MD, Oscar Gluck, MD, Jose L. Granda, MD, PhD, Maria Hill, MD, Paul Howard, MD, FACR, Miguel Ingelmo, MD, Kenneth C. Kalunian, MD, Gary M. Kammer, MD, Keith Kanik, MD, Stanley Kaplan, MD, Howard M. Kenney, MD, Neil A. Kurtzman, MD, Robert G. Lahita, MD, Michael R. Liebling, MD, Jill Suzanne Lindberg, MD, Susan Manzi, MD, MPH, Joan T. Merrill, MD, PhD, Larry W. Moreland, MD, C. Michael Neuwelt, MD, Michelle A. Petri, MD, MPH, Rosalind Ramsey-Goldman, MD, Steven G. Rosenblatt, MD, Naomi Rothfield, MD, John T. Schousboe, MD, K. Lea Sewell, MD, Yvonne Sherrer, MD, FACR, Douglas B. Smith, MD, Bruce S. Spinowitz, MD, FACP, Roland M. Staub, MD, Steven Stern, MD, Jon T. Stevenson, MD, Ronald van Vollenhoven, MD, John Varga, MD, Miguel Vilardell-Tarres, MD, and Daniel J. Wallace, MD.

    • La Jolla Pharmaceutical (LJP) provided the study medication.


Abstract

Objective

To determine whether LJP 394 delays or prevents renal flare in patients with systemic lupus erythematosus (SLE) and a history of renal disease.

Methods

In a 76-week, double-blind, placebo-controlled study, 230 SLE patients were randomized to receive 16 weekly doses of 100 mg of LJP 394 or placebo, followed by alternating 8-week drug holidays and 12 weekly doses of 50 mg of LJP 394 or placebo. An assay measuring the affinity of the serum IgG fraction for the DNA epitope of LJP 394 identified a high-affinity population of patients (189 of 213 patients; 89% taking LJP 394 and 90% taking placebo). Analyses were performed on both the intent-to-treat population and the high-affinity population.

Results

Anti–double-stranded DNA antibodies decreased and C3 levels tended to increase during treatment with LJP 394. In the intent-to-treat population, the time to renal flare was not significantly different between treatment groups, but patients taking LJP 394 had a longer time to institution of high-dose corticosteroids and/or cyclophosphamide (HDCC) and required 41% fewer treatments with HDCC. In the high-affinity population, the LJP 394 group experienced a longer time to renal flare, 67% fewer renal flares, longer time to institution of HDCC, and 62% fewer HDCC treatments compared with the placebo group. In patients with serum creatinine levels ≥1.5 mg/dl at study entry, those taking LJP 394 had 50% fewer renal flares; no renal flares were observed in the high-affinity group taking LJP 394. Serious adverse events were observed in 25 of the 114 LJP 394–treated patients (21.9%) and 34 of the 116 placebo-treated patients (29.3%).

Conclusion

Treatment with LJP 394 in patients with high-affinity antibodies to its DNA epitope prolonged the time to renal flare, decreased the number of renal flares, and required fewer HDCC treatments compared with placebo. The study drug appeared to be well tolerated.

Systemic lupus erythematosus (SLE) is a multisystem disease in which tissue damage is mediated by autoantibodies and immune complex deposition. High-avidity IgG autoantibodies against double-stranded DNA (anti-dsDNA) have a high degree of specificity for SLE and are thought to contribute to certain disease manifestations, including nephritis. Anti-dsDNA antibodies can be eluted from diseased glomeruli (1), and high serum titers of anti-dsDNA antibodies correlate with the presence of active glomerulonephritis (1–5). Monoclonal anti-dsDNA antibodies infused into normal mice or human IgG anti-dsDNA antibodies infused into SCID mice were shown to deposit in the glomeruli and induce proteinuria (6, 7). Factors that enhance the renal pathogenicity of anti-dsDNA antibodies include the ability to fix complement (8–10), a cationic charge (11), and a high avidity (1, 12).

Nephritis is a primary cause of morbidity and mortality in SLE (13–16). Management of the renal manifestations of SLE often requires immunosuppression with high-dose corticosteroids, azathioprine, methotrexate, mycophenolate mofetil, and/or cyclophosphamide. However, the utility of these agents is limited by significant drug-induced toxicity (17, 18).

LJP 394 (abetimus sodium) was designed to arrest the renal disease of SLE and prevent renal flares by selectively reducing antibodies to dsDNA and their parent B cells via antigen-specific tolerance. Accordingly, this nonimmunogenic immunomodulatory agent has been demonstrated to specifically reduce anti-dsDNA antibody titers in mice as well as in humans with SLE (19–21). It consists of four 20-mer dsDNA epitopes conjugated to a pharmacologically inert triethylene glycol platform. LJP 394 is capable of crosslinking anti-dsDNA antibodies in solution or on the surface of B cells (22). Previous studies have shown that the crosslinking of membrane immunoglobulin on the surface of naive B cells in the absence of T cell help can tolerize B cells via anergy or apoptosis (23–25).

Administration of LJP 394 is thought to reduce circulating anti-dsDNA antibodies by at least 2 mechanisms. First, LJP 394 acutely depletes circulating anti-dsDNA antibodies, presumably by forming small, soluble complexes that do not appear to result in significant activation of the complement system. Second, as demonstrated in rodents, LJP 394 appears to induce B cell tolerance by crosslinking anti-dsDNA surface immunoglobulin receptors on the B cell and triggering the signal transduction pathways that lead to B cell anergy or apoptosis (19). The relative contribution of these 2 mechanisms to the reduction in anti-dsDNA antibodies in SLE patients has not been fully established, and other potential mechanisms are possible.

An assay was developed that measures the strength of binding between a patient's serum IgG fraction and the LJP 394 DNA epitope to determine if the efficacy of LJP 394 correlates with its binding to anti-DNA antibodies (26, 27). A direct relationship between affinity and efficacy would be expected if the affinity of the patient's antibody for LJP 394 influenced the patient's response to a fixed dose of drug. For example, patients with high-affinity antibody binding should exhibit greater responses than patients with low-affinity antibody binding. The affinity assay was used to identify study patients with high-affinity antibody binding to the LJP 394 DNA epitope, using serum samples collected prior to initial treatment with study drug.

We present the results of a randomized, double-blind, placebo-controlled trial designed to determine whether LJP 394 prevents or delays subsequent renal flares in patients with SLE and a history of renal disease. We also examined anti-dsDNA antibody and serum complement (C3) levels, treatment with high-dose corticosteroids and/or cyclophosphamide (HDCC), and major SLE flares. Analyses were performed using the intent-to-treat population, the group with high-affinity antibodies to the LJP 394 DNA epitope at study entry, and the group with serum creatinine levels ≥1.5 mg/dl at study entry.

PATIENTS AND METHODS

Patient population.

Patients were eligible for entry into the study if they 1) fulfilled the American College of Rheumatology criteria for the classification of SLE (28); 2) had an episode of SLE renal disease within 4 years before study entry, as documented in clinical or hospital records, and were treated with cyclophosphamide or ≥30 mg/day of prednisone, or if they had significant, confirmed laboratory findings or renal biopsy findings (class III, IV, or V) of renal disease; 3) had an anti-dsDNA titer ≥15 IU/ml by Farr assay at study entry; and 4) provided voluntary informed consent.

Patients were excluded if they 1) had evidence of a renal flare within 3 months of screening, as documented by significant changes in laboratory tests for hematuria, proteinuria, or serum creatinine levels or the presence of new red blood cell (RBC) casts in the urine; 2) were receiving >20 mg/day of prednisone (or equivalent), >200 mg/day of azathioprine, >25 mg/week of methotrexate, and/or any dosage of cyclophosphamide within 3 months of screening; or 3) had a serum creatinine level of >2.5 mg/dl.

Results of any previous renal biopsies were collected, although they were not required for study entry. The protocol was approved by appropriate institutional review boards and ethics review committees.

Study design.

The study was a multicenter, randomized, double-blind, placebo-controlled trial comparing intravenously administered LJP 394 versus placebo in patients with a history of SLE and renal disease. Prospective patients were observed for 4–6 weeks prior to randomization to ensure that inclusion and exclusion criteria were met. In the initial protocol, patients received LJP 394 (La Jolla Pharmaceutical, San Diego, CA) at a dosage of 100 mg/week or placebo in a volume of 2 ml/week for 52 weeks with a 6-month followup period.

The protocol was later amended to a 76-week treatment period that included a 16-week induction period and a 60-week maintenance period. Patients received 16 weekly doses of 100 mg of LJP 394 or placebo during the induction period, followed by alternating 8-week drug holidays and 12 weekly treatments with 50 mg LJP 394 or placebo for a total of 76 weeks. The change in dosing schedule was based in part on a phase II study in which 4 patients taking 50 mg/week for 4 months showed a sustained reduction in antibodies for 2 additional months (21). All patients enrolled in the original protocol were transferred into the amended protocol prior to completing the first 12-week dosing cycle of the maintenance period.

Renal flares.

A protocol-defined renal flare required that it be attributed to SLE by the treating physician and medical monitor. In addition, 1 or more of the following 3 criteria were required: 1) a reproducible increase in 24-hour urine protein levels to (a) >1,000 mg if the baseline value was <200 mg, (b) >2,000 mg if the baseline value was 200–1,000 mg, or (c) more than twice the value at baseline if the baseline value was >1,000 mg; 2) a reproducible increase in serum creatinine of >20% or at least 0.3 mg/dl, whichever was greater, accompanied by proteinuria (>1,000 mg/24 hours), hematuria (≥4 RBCs/high-power field), and/or RBC casts; or 3) new, reproducible hematuria (≥11–20 RBCs/high-power field) or a reproducible increase in hematuria by 2 grades compared with baseline, associated with >25% dysmorphic RBCs, glomerular in origin, exclusive of menses, accompanied by either an 800-mg increase in 24-hour urinary protein levels or new RBC casts.

Assessment of laboratory values.

All laboratory values were determined at a central laboratory. Baseline anti-dsDNA antibody levels were calculated as the mean of the last 2 determinations prior to administration of the study drug. Baseline values for all other laboratory measures were determined immediately before administration of the study drug. The upper limit of normal for the anti-dsDNA antibody assay at the central laboratory was 5 IU/ml.

Concomitant medications.

Therapeutic intervention with HDCC was performed at the investigator's discretion. Patients were encouraged to remain in the trial following renal flares and HDCC therapy. HDCC was defined as treatment with cyclophosphamide; an increase in the dosage of oral, intravenous, or intramuscular prednisone (or equivalent) of ≥15 mg/day over baseline to a dosage >20 mg/day for more than 2 days; or any dose of prednisone (or equivalent) exceeding 200 mg in a single day. Topical, intraarticular, intralesional, or intraocular corticosteroid administration was excluded from this analysis. In many patients, treatment with both high-dose corticosteroids and cyclophosphamide was instituted. HDCC analyses were based on the first exposure of either or both therapies in a given patient.

Assessment of the signs and symptoms of SLE.

The signs and symptoms of SLE were evaluated by the SLE Disease Activity Index (SLEDAI), the Systemic Lupus International Cooperating Clinics/American College of Rheumatology Damage Index, and Medical Outcomes Short Form 36 health survey (29). These instruments were completed at baseline and at regular intervals during the trial.

Affinity determination.

An assay was utilized to measure the affinity of a patient's serum IgG antibody fraction for the DNA epitope of LJP 394. Briefly, the total IgG fraction was isolated from serum using protein G beads. Binding affinity to the LJP 394 DNA epitope was measured using surface plasmon resonance (BIAcore Model 2000 system; BIAcore, Uppsala, Sweden). Data are reported as the apparent equilibrium dissociation constant (Kd′), which defines the concentration of IgG required to reach half-maximum binding at equilibrium and is expressed in milligrams of IgG per milliliter. The Kd′ reflects the titer-weighted average affinity of the patient's IgG fraction for the LJP 394 DNA epitope. This implies that the measured Kd′ is influenced by both the titer and the affinity of the combined antibody populations. Previous studies have demonstrated that the titration curves for the polyclonal antibody population appear as if there is 1 monoclonal population with that titer-weighted average Kd′ (26). The experimental and analytical bases for this assay have been published elsewhere (26, 27, 30).

All affinity assays were performed by individuals blinded to clinical outcomes, and the segregation value was defined prior to analyzing the clinical outcomes in these populations. Plasma samples from 186 patients (91 taking LJP 394; 95 taking placebo) were available from both the screening (pretreatment) period and the period following induction (4 months of weekly treatment). Selection of a specific segregation value to identify patients with high-affinity antibodies to LJP 394 was based on a comparison of the affinity before initial exposure to study drug with the affinity after 4 months of treatment with 100 mg/week of LJP 394 or placebo. The patients with high-affinity antibodies to LJP 394 were those with a Kd′ ≤0.8 mg/ml.

Patient populations used for statistical analysis.

Statistical analyses were conducted on the following groups of patients: 1) the intent-to-treat population, which was defined as all randomized patients who received at least 1 dose of study drug; 2) patients with high- and low-affinity antibodies to the LJP 394 DNA epitope prior to treatment with study drug; and 3) patients with a serum creatinine value ≥1.5 mg/dl at study entry.

Statistical analysis.

The sample size estimate assumed an 18-month flare rate of 35% in the placebo-treated group and 17.5% in the LJP 394–treated group. The trial was designed to detect a statistically significant difference in time to renal flare between treatment arms at a power of 80% and a 2-sided significance level of 0.05. The comparison of continuous variables was performed using analysis of variance. Fisher's exact test was used for the comparison of incidence rates and all other categorical comparisons. P values were not adjusted for multiple comparisons.

All time-to-event comparisons were performed using the Kaplan-Meier product-limit method and the log rank test. When comparing time-to-event variables for the entire study period, all patients were included in the analysis until the time of the event (renal flare or HDCC treatment) or their last exposure to the study drug. Only the first renal flare and the first treatment with HDCC were used in the time-to-event and incidence analyses.

RESULTS

Characteristics of the patient population.

Two hundred thirty patients were randomized to receive study drug (114 to receive LJP 394; 116 to receive placebo); 214 patients were randomized in North America and 16 patients were randomized in Europe. Pretreatment serum samples for the affinity analysis were available for 213 of the 230 patients, including 41 of the 42 patients who had a renal flare during the trial. Pretreatment samples were not available from the 16 patients treated at the European sites. These 16 European patients received a median of 3 doses (range 1–4 doses, mean 2.7 doses), and no renal flares were reported.

Demographic and baseline disease characteristics were similar between groups at the time of randomization. Figure 1 summarizes the disposition of the patients according to the 2001 Consolidated Standards of Reporting Trials (CONSORT) standards (31); the demographic data are shown in Table 1. The trial was discontinued after an interim analysis suggested that the trial was unlikely to reach statistical significance for differences in time to renal flare between the treatment groups in the intent-to-treat population.

Figure 1.

Flow chart defining the screening, enrollment, intervention allocation, followup, and analysis data sets. anti-dsDNA = anti–double-stranded DNA. = Administrative terminations were due to premature termination of the study.

Table 1. Baseline demographics, by treatment group*
 LJP 394PlaceboTotal (n = 230)
ITT (n = 114)DNA binding affinityITT (n = 116)DNA binding affinity
High (n = 92)Low (n = 13)High (n = 97)Low (n = 11)
  • *

    Study populations were subgrouped by intent-to-treat (ITT) and by the presence of high- or low-affinity binding to the LJP 394 DNA epitope at study entry.

  • Mean of values at visits 2 and 3.

  • Values at visit 3.

  • §

    P < 0.05 by chi-square analysis.

  • If more than one class was indicated, the higher class is shown.

  • #

    The mean Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) score was determined for the subgroup of patients with positive findings on the renal component of the SLEDAI at baseline.

Age, mean ± SD years37 ± 1138 ± 1132 ± 1035 ± 1035 ± 1033 ± 836 ± 10
Sex       
 Female10183111079010208
 Male139297122
Ethnic group       
 White4633556444102
 Black413473127472
 Asian76155012
 Hispanic181702320341
 Other2201103
Weight, mean ± SD kg71 ± 1871 ± 1880 ± 2074 ± 2074 ± 2079 ± 2372 ± 19
Disease duration, mean ± SD years9 ± 79 ± 77 ± 48 ± 68 ± 66 ± 38 ± 6
Anti-DNA antibodies at entry       
 15–49 IU/ml6756552464119
 50–200 IU/ml352575140686
 >200 IU/ml121111311125
Serum creatinine at entry       
 Higher than upper limit of normal (n = 230)68531061517129
 ≥1.5 mg/ml (n = 230)17116§11110§27
Creatinine clearance <70 ml/minute at entry (n = 221)332663634269
24-hour urinary protein ≥500 mg/day at entry (n = 214)483612§49454§97
Renal biopsy classification before baseline       
 Class I0001101
 Class II43065010
 Class III171521914536
 Class IV382853835076
 Class V181432016338
 Class VI2100002
 Class missing/unknown353133226367
SLEDAI score at baseline, mean       
 All patients6687797
 Patients with abnormal renal component#11111012111111

Clinical outcomes in the intent-to-treat population.

In the intent-to-treat population, the time to renal flare (P = 0.506) (Figure 2A) and the number of renal flares (P = 0.610) were not significantly different between treatment groups (19 LJP 394 patients, 23 placebo patients). However, LJP 394–treated patients had a longer time to institution of HDCC (P = 0.033) and required 41% fewer HDCC treatments (23 LJP 394 patients, 39 placebo patients; P = 0.026) (Figure 3A). The median time to institution of HDCC was 52.6 months in the LJP 394–treated patients and 26.9 months in placebo patients. The mean and median dosages of prednisone or prednisone equivalent in patients who received HDCC were 151 ± 42 mg/day and 50 mg/day, respectively.

Figure 2.

Analysis of time to first renal flare in the LJP 394–treated and the placebo–treated groups of the A, intent-to-treat population and B, high-affinity population. The terminal decrease in the proportion of patients who did not experience a renal flare was a result of the premature termination of the trial. Insets, Number of patients in each group at 2-month intervals over the course of the 18-month trial. PBO = placebo.

Figure 3.

Analysis of time to first administration of high-dose corticosteroids and/or cyclophosphamide (HDCC) in the LJP 394–treated and the placebo-treated groups in the A, intent-to-treat population and B, high-affinity population. Insets, Number of patients in each group at 2-month intervals over the course of the 18-month trial. PBO = placebo.

The introduction of HDCC was predominantly indicated for the treatment of SLE, accounting for 21 of 23 HDCC treatment events in the LJP 394 group and 32 of 39 in the placebo group. The indications for the 9 non–SLE-related HDCC exposures were as follows: inflammatory myopathy, cellulitis, adrenal crisis prophylaxis, polyarteritis nodosum after insect bites, viral syndrome, herpes zoster, fever, flu-like symptoms, and revision surgery for a left hip prosthesis.

During the initial 16-week period when patients received weekly doses of 100 mg of LJP 394, 4 LJP 394–treated patients experienced a renal flare. During this same induction period, 9 renal flares occurred in the placebo-treated group (Table 2).

Table 2. Effect of dosing regimen on number of renal flares*
 Induction (16 weeks)Maintenance (60 weeks)
LJP 394PlaceboLJP 394Placebo
  • *

    The induction period consists of the first 16 weeks of the trial, when patients received weekly doses of 100 mg of LJP 394 or placebo. The maintenance period consists of the 60-week period after induction, when patients were exposed to alternating 8-week drug holidays and 12 weekly treatments with 50 mg of LJP 394 or placebo. Over the entire 76-week study period, a total of 42 renal flares occurred.

Intent-to-treat491514
High affinity18613

Of the 42 renal flares that occurred during the trial, 25 met the serum creatinine end point, 21 met the proteinuria end point, and 5 met the hematuria end point. Ten patients met 2 end point criteria, and 2 patients met all 3 end point criteria simultaneously.

Of the 42 patients with renal flares, 35 (83%) were treated with HDCC and 19 (45%) were treated with cyclophosphamide. Twenty patients (48%) were hospitalized during the trial. Nineteen of 21 patients who were treated with cyclophosphamide (91%) had a renal flare due to the SLE. Of the remaining 2 cyclophosphamide-treated patients, one received treatment for “SLE syndrome” and the other for lupus nephritis.

Anti-dsDNA antibody levels decreased to −27 ± 4% (mean ± SEM) of baseline after 16 weekly doses of 100 mg of LJP 394, compared with an increase of 11 ± 6% in the placebo group (Figure 4). In the intent-to-treat population, anti-dsDNA antibody levels were reduced by >30% in 51 of the 86 LJP 394–treated patients (59%) versus 22 of 96 of the placebo-treated patients (23%) (P < 0.0001 by Fisher's exact test). Antibody levels increased during each 8-week drug-free period and decreased during each 50-mg/week LJP 394 dosing cycle over the 18-month trial. Conversely, C3 levels tended to increase as the anti-dsDNA antibody levels decreased in the LJP 394 group, although this was not statistically significant. During weeks 44–48, the large increase in anti-dsDNA antibody level in the LJP 394 group was primarily attributed to 1 patient with low-affinity binding whose antibody level rose several thousand percent (from 20 IU/ml at study entry to 601 IU/ml at week 47).

Figure 4.

Anti–double-stranded DNA (anti-dsDNA) antibody levels (▪) and C3 levels (○) in the LJP 394–treated and the placebo-treated groups. Bars at the top indicate the active dosing periods. A, Intent-to-treat group receiving LJP 394. B, Intent-to-treat group receiving placebo. C, High-affinity group receiving LJP 394. D, High-affinity group receiving placebo. The high-affinity group represented 89% of the intent-to-treat population. Values are the mean and SEM.

Clinical outcomes in the high- and low-affinity populations.

Pretreatment serum samples were available for all but 1 of the patients treated in North America. The high-affinity population included 189 of 213 patients from the intent-to-treat population (89%) and was equally distributed between treatment groups: 92 of 105 patients taking LJP 349 (88%) and 97 of 108 patients taking placebo (90%). All affinity assays were conducted by individuals who were blinded to the clinical outcomes after the trial was completed.

Patients with high-affinity antibody binding exhibited a mean ± SEM change in affinity, from 0.30 ± 0.02 mg of IgG/ml at study entry to 0.81 ± 0.03 mg of IgG/ml after 4 months of weekly treatment with 100 mg of study drug (Figure 5). Reductions in affinity were dependent on the initial affinity values. Patients with the highest-affinity antibody binding exhibited the greatest reductions from baseline. In contrast, patients with low-affinity antibodies to LJP 394 did not exhibit a significant change in affinity at the end of the induction phase.

Figure 5.

Affinity analysis in the high- and low-affinity populations of the LJP 394–treated and the placebo-treated groups. Serum samples were evaluated at study entry (baseline; shaded bars) and after 4 months of treatment with LJP 394 (100 mg/week; open bars) or placebo. The initial affinities (Kd′) in the high- and low-affinity groups of both treatment cohorts were similar. There was a significant reduction in mean affinity in the high-affinity LJP 394 group after 4 months of treatment. Values are the mean and SEM.

In the high-affinity population, patients treated with LJP 394 experienced a significantly longer time to renal flare (P = 0.007) and had fewer renal flares (P = 0.008) compared with patients treated with placebo (Figure 2B). There were 3 times as many renal flares in placebo-treated patients (21 of 97 [22%]) as in LJP 394–treated patients (7 of 92 [8%]). During the induction period, patients treated with LJP 394 had fewer renal flares (n = 1) than did patients treated with placebo (n = 8) (P = 0.035) (Table 2). The median time to renal flare in the high-affinity population was 157.8 months for the LJP 394 group and 51.3 months for the placebo group.

In the high-affinity population, the LJP 394–treated patients had an increased time to institution of HDCC (P = 0.002) and required fewer treatments with HDCC (13 of 92 [14%]) compared with the placebo-treated patients (34 of 97 [35%]) (P = 0.001) (Figure 3B). The median time to institution of HDCC in the high-affinity population was 82.1 months for the LJP 394 group and 25.1 months for the placebo group.

In the low-affinity population, the LJP 394 group had a greater number of patients with serum creatinine levels ≥1.5 mg/dl at study entry (6 of 13 in the LJP 394 group; 0 of 11 in the placebo group). There were also more patients in the LJP 394 group who had proteinuria >500 mg/day at study entry (12 of 13 patients, compared with 4 of 11 patients in the placebo group). Renal flares were observed in 11 of 13 patients in the LJP 394 group (85%) compared with 2 of 11 patients in the placebo group (18%) (P = 0.003). Treatment with HDCC was instituted in 9 of 13 patients in the LJP 394 group (69%) and 5 of 11 patients in the placebo group (46%) (P = 0.408).

The effect of LJP 394 and placebo on anti-dsDNA antibody levels in the high-affinity population was similar to the effect observed in the intent-to-treat population (Figures 4C and D). Antibody levels decreased in the LJP 394 population during each dosing cycle and increased during each 8-week drug-free period over the 18-month trial. Mean anti-dsDNA antibody levels were unchanged in the placebo group over the course of the trial.

Patients with serum creatinine levels ≥1.5 mg/dl at baseline.

Twenty-seven patients (17 in the LJP 394 group; 10 in the placebo group) had serum creatinine levels between 1.5 and 2.5 mg/dl (upper limit for inclusion) at baseline. Renal flares occurred in 3 of the 17 patients receiving LJP 394 (18%) and 6 of the 10 patients receiving placebo (60%), and there was a significantly longer time to renal flare in the LJP 394 group than in the placebo group (P = 0.003). HDCC was administered to 5 of the 17 patients receiving LJP 394 (29%) and 7 of the 10 patients receiving placebo (70%), and there was a significantly longer time to institution of HDCC in the LJP 394 group than in the placebo group (P = 0.043). Of the 17 patients receiving LJP 394, 11 (65%) had high-affinity antibodies to LJP 394 prior to treatment, compared with all 10 of the patients receiving placebo (100%).

Among patients with high-affinity antibodies to LJP 394, renal flares were observed in none of the 11 patients of the LJP 394 treatment group, compared with 3 of the 5 patients (60%) in the low-affinity LJP 394 treatment group and 6 of the 10 patients (60%) in the placebo treatment group (P = 0.004). A longer time to institution of HDCC was also observed in patients with high-affinity antibodies to LJP 394 in the LJP 394 treatment group (2 of 11 patients [18%]) compared with the placebo treatment group (7 of 10 patients [70%]) (P = 0.009).

Serum creatinine levels increased from a mean ± SEM of 1.9 ± 0.4 mg/dl at baseline to 5.0 ± 3.0 mg/dl at the final visit for the 9 patients who experienced a renal flare. In contrast, mean serum creatinine levels remained unchanged at 1.8 mg/dl between baseline and the final visit in patients who had a renal flare during the trial.

Placebo-treated patients with serum creatinine levels ≥1.5 mg/dl at baseline appeared to have a higher rate of renal flares compared with the placebo-treated patients in the intent-to-treat population. In patients with serum creatinine levels ≥1.5 mg/dl at baseline, 6 of 10 patients in the placebo group (60%) developed renal flares, compared with 23 of 116 patients in the placebo group of the intent-to-treat population (20%).

Changes in disease activity.

The SLEDAI score was determined at regular intervals during the trial. No significant differences in the SLEDAI scores were found between treatment groups in comparisons of baseline values with values at the end of the induction phase or the final visit.

Safety.

Seventy-seven patients (40 in the LJP 394 group; 37 in the placebo group) completed 18 months of protocol treatment; 133 patients completed more than 12 months, and 182 patients completed at least 4 months. Fifty patients (25 in the LJP 394 group; 25 in the placebo group) discontinued the protocol treatment prior to the premature termination of the trial.

LJP 394 appeared to be well tolerated, with no significant differences in the pattern of treatment-emergent adverse events between the LJP 394 and placebo groups (Tables 3 and 4). One hundred eighty-one serious adverse events (87 in the LJP 394 group; 94 in the placebo group) were reported in 25 of 114 patients receiving LJP 394 (22%) and 34 of 116 patients receiving placebo (29%). Three serious adverse events (1 LJP 394–treated patient; 2 placebo-treated patients) were considered by investigators to be possibly related to administration of the study drug. The LJP 394–treated patient, a 37-year-old woman with history of hypertension, Raynaud's disease, pericarditis, and pleuritis, had a left popliteal vein thromboembolism. One of the 2 placebo-treated patients had nephritis and the other had a urinary tract infection. There were 104 hospitalizations (39 for the LJP 394 group; 65 for the placebo group) in 56 patients (23 patients in the LJP 394 group; 33 in the placebo group). An independent review of the medical records indicated that the hospitalizations were primarily due to manifestations of SLE or complications of treatment.

Table 3. Frequency of adverse events reported in more than 5% of either treatment group
Body system, conditionLJP 394 (n = 114)Placebo (n = 116)
Frequency%Frequency%
Body as a whole    
 Allergic reaction1815.81512.9
 Asthenia3228.12925.0
 Chills108.854.3
 Edema of the face87.076.0
 Fever1614.02420.7
 Flu syndrome108.81210.3
 Headache4438.64437.9
 Infection5750.05446.6
 Accidental injury2824.62824.1
 Pain2320.22017.2
 Abdominal pain2017.52219.0
 Back pain1714.986.9
 Chest pain1714.91512.9
Cardiovascular system    
 Cardiovascular disorder65.343.5
 Hypertension1412.31311.2
 Migraine21.876.0
 Peripheral vascular disease43.576.0
 Vasculitis65.343.5
 Vasodilation65.343.5
Digestive system    
 Anorexia65.343.5
 Constipation87.054.3
 Diarrhea1714.93328.4
 Dyspepsia119.61412.1
 Gastroenteritis76.197.8
 Oral monilia65.3
 Nausea3127.23227.6
 Mouth ulcers1614.01311.2
 Vomiting1311.41613.8
Hematologic and lymphatic systems    
 Anemia65.386.9
 Ecchymosis1815.886.9
 Lymphadenopathy65.332.6
Metabolic/nutritional systems    
 Hypokalemia65.343.5
 Peripheral edema2320.22420.7
Musculoskeletal system    
 Arthralgia4035.13631.0
 Arthritis1412.31815.5
 Arthrosis87.076.0
 Myalgia3026.32925.0
 Tendon disease21.876.0
 Joint disease65.386.9
Nervous system    
 Depression76.176.0
 Dizziness2017.52622.4
 Hypertonia87.0119.5
 Insomnia1311.41412.1
 Paresthesia1714.92219.0
Respiratory system    
 Bronchitis119.61311.2
 Increased cough2925.41613.8
 Dyspnea1513.2108.6
 Pharyngitis2017.53025.9
 Pleural disease97.91916.4
 Pneumonia21.876.0
 Rhinitis2320.22319.8
 Sinusitis97.91512.9
Skin and appendages    
 Acne76.143.5
 Alopecia1614.01512.9
 Herpes zoster87.076.0
 Pruritus87.086.9
 Rash1815.81613.8
 Lupus rash1412.32118.1
 Skin ulcers87.021.7
 Urticaria54.465.2
Special senses    
 Conjunctivitis1210.565.2
 Eye disease21.897.8
 Ear pain43.597.8
Urogenital system    
 Dysuria65.365.2
 Metrorrhagia65.397.8
 Monilia vagina97.91311.2
 Nephritis2219.32420.7
 Urinary tract infection1614.01714.7
 Vaginitis65.321.7
Table 4. Treatment-emergent serious adverse events among all study patients (n = 230), by treatment group
 No. (%) of patients taking LJP 394No. (%) of patients taking placebo
Patients with serious adverse events25 (21.9)34 (29.3)
Body system affected  
 Body as a whole13 (11.4)16 (13.8)
 Cardiovascular system10 (8.8)7 (6.0)
 Digestive system4 (3.5)7 (6.0)
 Endocrine system0 (0.0)0 (0.0)
 Hematologic and lymphatic systems0 (0.0)2 (1.7)
 Metabolic/nutritional systems1 (0.9)4 (3.4)
 Musculoskeletal system3 (2.6)5 (4.3)
 Nervous system1 (0.9)7 (6.0)
 Respiratory system2 (1.8)10 (8.6)
 Skin and appendages0 (0.0)2 (1.7)
 Special senses0 (0.0)0 (0.0)
 Urogenital system10 (8.8)8 (6.9)

Antiphospholipid antibodies are present in up to 40% of SLE patients and predispose these patients to thromboembolic events (32). Ten thromboembolic events, including deep vein thrombosis, myocardial infarction, mesenteric infarction, and cerebellar infarction, were reported (8 in the LJP 394 group; 2 in the placebo group). The difference was not statistically significant between treatment groups. Thromboembolic events occurred 6–32 days after the last exposure to study drug, well after circulating levels of the study drug would be detectable (plasma half-life for LJP 394 ∼1 hour). Antiphospholipid antibodies were not determined in these patients, either at randomization or at the time of the thromboembolic event.

Two deaths occurred during the trial: 1 in the LJP 394 group and 1 in the placebo group. The death in the LJP 394 treatment group was attributed to intraabdominal sepsis following mesenteric infarction, which was presumed to be secondary to pancreatitis with a subsequent large bowel resection. Death was not attributed to administration of the study drug. The death in the placebo patient was attributed to suicide.

DISCUSSION

Treatment with LJP 394 has been shown to reduce anti-dsDNA antibody levels in SLE patients in 6 clinical trials (20, 21). The present trial was the largest and longest study thus far, with 230 patients enrolled and 114 treated with LJP 394 for a mean of 11.4 months. The study demonstrated that LJP 394 treatment over 18 months reproducibly decreased anti-dsDNA levels when patients were given weekly treatments and that antibody levels increased during drug holidays. In the LJP 394 treatment group, serum C3 levels exhibited an inverse relationship to anti-dsDNA antibody levels. While the absolute changes in C3 levels were numerically small, this apparent relationship supports the hypothesis that reductions in anti-dsDNA antibody levels correlate with increases in serum complement levels, presumably due to a reduction in complement activation and consumption.

The reduction in anti-dsDNA antibodies was accompanied by an apparent improvement in clinical outcomes in 89% of patients with high-affinity antibodies to LJP 394. There was a 67% reduction in renal flares in patients with high-affinity antibodies who were treated with LJP 394, compared with placebo treatment, although the number of renal flares in the intent-to-treat population was not statistically significantly different. The treatment effect in the high-affinity population appeared to be more pronounced when patients received 100 mg/week during the first 4 months, during which time there was an 88% reduction in renal flares. Treatment with LJP 394 resulted in a 41% reduction in the number of HDCC treatments in the intent-to-treat population and a 62% reduction in patients with high-affinity antibodies compared with placebo. The clinical benefit of delaying renal flares is supported by the reduction in the requirement for HDCC treatment and the reduction in number of SLE-related hospitalizations.

Patients with serum creatinine levels ≥1.5 mg/dl at study entry were a small, but potentially important, group. Placebo-treated patients in this group appeared to experience a higher percentage of renal flares (6 of 11 patients [55%]) as compared with placebo-treated patients with serum creatinine levels <1.5 mg/dl at entry (17 of 106 patients [16%]). Treatment with LJP 394 reduced the number of renal flares, especially in the patients with high-affinity binding, in whom renal flares did not occur. Because of the highly specific mechanism of action of LJP 394, these data support the hypothesis that anti-dsDNA antibodies increase renal exacerbations in patients in whom renal function is already significantly compromised.

Examination of renal flares during the induction and maintenance phases of the trial suggested that clinical response was correlated with the dosing regimen. LJP 394 was more efficacious when patients were treated with weekly doses of 100 mg. In the intent-to-treat population, there were 9 renal flares in the placebo group and 4 in the LJP 394 group during the induction period, compared with 14 in the placebo group and 15 in the LJP 394 group during the maintenance phase. In the high-affinity population, there were 8 renal flares in the placebo group and 1 in the LJP 394 group during the induction period, compared with 13 in the placebo group and 6 in the LJP 394 group during the maintenance phase.

The trial was initially designed to expose patients to LJP 394 at a dosage of 100 mg/week for 52 consecutive weeks. The protocol was amended after a change in sponsorship to an induction/maintenance design, in which the maintenance phase consisted of repeating cycles of 8-week drug holidays and 12 weekly doses of 50 mg LJP 394 or placebo. The anti-dsDNA data from the current trial suggest that the 50-mg dose was less effective than the 100-mg dose and that the drug holidays were sufficiently long to allow anti-dsDNA antibodies to return to or exceed their pretreatment levels.

The affinity assay provided a way to identify patients who were most likely to exhibit the desired pharmacodynamic response to LJP 394. On average, a 235% reduction in titer-weighted average affinity for LJP 394 was observed in the high-affinity population after 4 months of LJP 394 treatment, compared with little to no change in the low-affinity population. Virtually all LJP 394–treated patients in the high-affinity population (99%) exhibited a reduction in affinity during this period. One potential mechanism underlying the effect of the 100-mg dosage on posttreatment antibody affinity in the high-affinity population is that LJP 394 preferentially tolerizes B cells with high-affinity surface immunoglobulin for the LJP 394 DNA epitope. However, this mechanism has been demonstrated only in animals and remains to be proven in humans. In addition, other mechanisms, including direct depletion of circulating antibodies, are likely contributors to the drug-induced change in affinity.

The affinity segregation value, Kd′ ≤0.8 mg of IgG/ml, was based on pharmacodynamic data. A subsequent sensitivity analysis was conducted on a range of affinity values to evaluate the robustness of the selected value. The LJP 394 treatment group experienced a longer time to renal flare (P < 0.05) and fewer renal flares (P < 0.05) for Kd′ values from 0.4 to 0.95 mg of IgG/ml. Based on this analysis, the segregation value appears to be robust.

There were several issues to consider in evaluating the low-affinity population. The number of patients was small (13 in the LJP 394 group; 11 in the placebo group), and an imbalance in renal function existed at study entry. The low-affinity LJP 394 treatment group had a greater number of patients whose serum creatinine level was ≥1.5 mg/ml at entry (6 of 13; compared with 0 of 11 in the placebo group) and whose proteinuria was >500 mg/day at entry (12 of 13; compared with 4 of 11 in the placebo group). These imbalances may have contributed to the increased rate of renal flare in the low-affinity LJP 394 population. Even though patients with significantly impaired renal function are more likely to progress to renal insufficiency and/or renal failure, it is reassuring that the flare rate in patients with serum creatinine levels ≥1.5 mg/ml at entry were similar between the low-affinity LJP 394 treatment group (3 of 5 patients) and the low-affinity placebo treatment group (6 of 10 patients). The apparently dichotomous outcome between the high-affinity and low-affinity populations with regard to clinical efficacy should be considered in interpreting the results of this trial.

The study also showed that the 100-mg/week dosage of LJP 394 reduced anti-dsDNA antibody levels in the low-affinity group, while the 50-mg/week dosage was inadequate to control anti-dsDNA antibodies (data not shown). This finding suggests that patients with low-affinity antibodies require higher doses of LJP 394 to control anti-dsDNA antibody levels. Another possibility is that suboptimal dosing, perhaps in combination with the on–off dosing regimen, may have also contributed to the renal flares in this group. Patients with low-affinity antibodies may respond if treated weekly with doses higher than 50 mg. However, further research is necessary.

In summary, LJP 394 is a novel immunomodulatory agent that selectively decreases anti-dsDNA antibody levels when administered weekly at doses of 100 mg. The pharmacodynamic effect of LJP 394 correlated with the clinical benefit in patients with SLE and previous renal manifestations. In the intent-to-treat population, the time to institution of HDCC was significantly prolonged and the number of treatments was significantly lower compared with placebo. In patients with high-affinity antibodies to LJP 394, representing 89% of the intent-to-treat population, the times to renal flare and to institution of HDCC were significantly longer compared with placebo, and the numbers of renal flares and treatments with HDCC were significantly lower compared with placebo. Treatment benefit was also evident in patients with serum creatinine levels ≥1.5 mg/dl at study entry. Administration of LJP 394 appeared to be well tolerated, with no evidence of immunogenicity or infusion reactions. A multinational, prospective, randomized controlled trial to assess time to renal flare in SLE patients with a history of renal disease and high-affinity antibodies at baseline is ongoing.

Acknowledgements

The authors would like to acknowledge the generous assistance of Patricia McNeeley and Yun Hardiman and the thoughtful comments and analyses provided by Drs. K. Frank Austen, Jay Hu, Sheela Talwalker, W. Leigh Thompson, Shaw-Ling Wang, Andrew Wiseman, and Irina Yushmanova.

Ancillary