Other members of the BG9588 Lupus Nephritis Trial Group were Michael Becker, MD, Chicago, Illinois; Gary Gilkeson, MD, Charleston, South Carolina; Oscar Gluck, MD, Phoenix, Arizona; Katherine Nickerson, MD, New York, New York; Jane Salmon, MD, New York, New York; and Gary Firestein, MD, La Jolla, California.
A short course of BG9588 (anti–CD40 ligand antibody) improves serologic activity and decreases hematuria in patients with proliferative lupus glomerulonephritis
Version of Record online: 28 FEB 2003
Copyright © 2003 by the American College of Rheumatology
Arthritis & Rheumatism
Volume 48, Issue 3, pages 719–727, March 2003
How to Cite
Boumpas, D. T., Furie, R., Manzi, S., Illei, G. G., Wallace, D. J., Balow, J. E., Vaishnaw, A. and on behalf of the BG9588 Lupus Nephritis Trial Group (2003), A short course of BG9588 (anti–CD40 ligand antibody) improves serologic activity and decreases hematuria in patients with proliferative lupus glomerulonephritis. Arthritis & Rheumatism, 48: 719–727. doi: 10.1002/art.10856
- Issue online: 28 FEB 2003
- Version of Record online: 28 FEB 2003
- Manuscript Accepted: 13 DEC 2002
- Manuscript Received: 30 SEP 2002
- Biogen, Inc., Cambridge, Massachusetts
- NIH/NCRR/GCRC. Grant Number: 5M01-RR-00056
CD40–CD40 ligand (CD40L) interactions play a significant role in the production of autoantibodies and tissue injury in lupus nephritis. We performed an open-label, multiple-dose study to evaluate the safety, efficacy, and pharmacokinetics of BG9588, a humanized anti-CD40L antibody, in patients with proliferative lupus nephritis. The primary outcome measure was 50% reduction in proteinuria without worsening of renal function.
Twenty-eight patients with active proliferative lupus nephritis were scheduled to receive 20 mg/kg of BG9588 at biweekly intervals for the first 3 doses and at monthly intervals for 4 additional doses. Safety evaluations were performed on all patients. Eighteen patients receiving at least 3 doses were evaluated for efficacy.
The study was terminated prematurely because of thromboembolic events occurring in patients in this and other BG9588 protocols (2 myocardial infarctions in this study). Of the 18 patients for whom efficacy could be evaluated, 2 had a 50% reduction in proteinuria without worsening of renal function. Mean reductions of 38.9% (P < 0.005), 50.1% (P < 0.005), and 25.3% (P < 0.05) in anti–double-stranded DNA (anti-dsDNA) antibody titers were observed at 1, 2, and 3 months, respectively, after the last treatment. There was a significant increase in serum C3 concentrations at 1 month after the last dose (P < 0.005), and hematuria disappeared in all 5 patients with significant hematuria at baseline. There were no statistically significant reductions in lymphocyte count or serum immunoglobulin, anticardiolipin antibody, or rubella IgG antibody concentrations after therapy.
A short course of BG9588 treatment in patients with proliferative lupus nephritis reduces anti-dsDNA antibodies, increases C3 concentrations, and decreases hematuria, suggesting that the drug has immunomodulatory action. Additional studies will be needed to evaluate its long-term effects.
Proliferative lupus nephritis is a protracted disease with a waxing and waning course (1–3). Although immunosuppressive treatment regimens have led to a dramatic reduction in the frequency of end-stage renal disease (4–18), they fail to induce remission in a significant number of patients (9, 10, 19, 20). Flares are common after remission (8, 21–23), and significant drug toxicities (e.g., gonadal toxicity, avascular bone necrosis, and infections) necessitate the search for alternative agents (9, 10, 24–26).
Studies in both animal models (27–30) and humans (31–38) have demonstrated the essential role of the costimulatory molecule CD40 and its ligand CD40L (gp39, CD154, 5c8 antigen, T-BAM) in the production of pathogenic autoantibodies and tissue injury in lupus nephritis. CD40L is found mainly on T cells and platelets and binds to CD40 that is present on endothelial cells, professional antigen-presenting cells, macrophages, B cells, and renal parenchymal cells, such as mesangial cells, epithelial cells, and distal tubular epithelial cells (27–43). Glomerular and tubular CD40 expression is markedly up-regulated in proliferative, but not in membranous, lupus nephritis (33). Several groups have reported hyperexpression of CD40L by T cells (31, 32) and elevated soluble CD40L concentrations (33) in human lupus. Anti-CD40L therapy ameliorates renal disease in animal models of systemic lupus erythematosus (SLE) and improves survival even when used in animals with established disease (27–30). This therapy is also effective in retarding renal disease in animals with membranous and heavy metal–induced glomerulonephritis (30).
BG9588 is a humanized anti-human CD40L antibody that blocks antigen-specific IgG responses in nonhuman primates (baboons and rhesus monkeys) immunized with a variety of T-dependent antigens. In this report, we describe the results of a phase II, multicenter, open-label study evaluating the toxicity and efficacy of BG9588 in patients with proliferative lupus glomerulonephritis.
PATIENTS AND METHODS
This phase II, multicenter, open-label study of BG9588 was sponsored by Biogen, Inc. (Cambridge, MA). The trial was conducted in 28 patients at 10 sites from July 1999 to April 2000. Informed consent was obtained from all patients in accordance with the Institutional Review Boards of each of the participating institutions.
Twenty-eight patients between 18 and 75 years of age and with a diagnosis of SLE according to the 1982 criteria of the American College of Rheumatology (44) were enrolled. Inclusion criteria included a renal biopsy showing proliferative lupus nephritis within 5 years prior to the first dose of study drug, proteinuria of >1.0 gm/day at 2 separate screening visits, and any 1 of the following 4 criteria at each of the 2 screening visits: 1) anti–double-stranded DNA (anti-dsDNA) antibody >2 times the upper limit of normal; 2) C3 <80 mg/dl; 3) hematuria (>5 red blood cells/high-power field); and 4) granular or red blood cell casts detected on urinalysis.
Exclusion criteria included antiphospholipid syndrome (presence of anticardiolipin antibodies in combination with either prior arterial or venous thrombosis or history of 3 or more spontaneous abortions; patients with a history of thromboembolic events were allowed to enroll if they were receiving anticoagulation therapy), severe renal disease (rapidly progressive glomerulonephritis or fibrinoid necrosis and/or cellular crescents affecting >25% of glomeruli in any renal biopsy performed within 3 months prior to the date the first dose of study drug was to be given), serum creatinine >2.0 mg/dl, active psychiatric disease, chronic liver dysfunction, chronic or serious acute infections, a CD4+ cell count of >150/μl, any vaccinations given within the 4 weeks prior to the date the first dose of study medication was to be given, treatment with prednisone >0.5 mg/kg/day or with intravenous or oral cyclophosphamide, or treatment with other immunosuppressive agents (including methylprednisolone) or angiotensin-converting enzyme (ACE) inhibitors initiated within the 4 weeks prior to the date the first dose of study medication was to be given. Patients were screened twice (2 screenings performed within 1 month and separated by at least 5 days) prior to administration of the first dose of study drug to assure fulfillment of entry criteria. A total of 44 patients were screened for the study, 16 of whom were excluded based on the above criteria.
Criteria for withdrawal from the study included pregnancy, substantial worsening of renal function (>33% or 0.3-mg/dl increase from baseline serum creatinine level [whichever was greater], and >25% decrease from baseline creatinine clearance sustained for >1 week) or proteinuria (>50% increase in proteinuria if baseline level was >3.5 gm/day or >1.5-gm increase in proteinuria if baseline level was <3.5 gm/day), or severe extrarenal lupus.
Patients receiving ACE inhibitors, azathioprine, or mycophenolate mofetil for >4 weeks prior to the first dose of BG9588 were allowed to continue their therapies during the study provided that the treatments were maintained at a constant level until BG9588 dosing had been completed. For patients who were taking prednisone at study entry, a tapering schedule, to a dosage of 10 mg/day beginning 1 month after the first dose of BG9588, was required. Each dosage reduction occurred at an interval of 4 weeks and consisted of decrements of 10 mg for prednisone dosages >20 mg/day and 5 mg for prednisone dosages ≤20 mg/day.
Because of reports of thromboembolic events occurring in subjects enrolled in this study and other studies involving BG9588, the protocol was amended in October 1999 to include prophylactic anticoagulation therapy consisting of subcutaneous daily low molecular weight heparin and aspirin (81 mg/day orally). In November 1999, study enrollment and dosing were halted because of several thromboembolic events in BG9588 protocols. Because no further BG9588 dosing was to occur, disallowed medications, including immunosuppressive therapy and ACE inhibitors, were subsequently resumed.
Preparation and administration of BG9588.
BG9588 consists of the complementarity-determining regions of the murine monoclonal antibody 5c8 (anti-human CD40L antibody) with human variable-region framework residues and IgG1 constant region. BG9588 (Biogen, Inc.) is secreted from murine NSO myeloma cells, which overexpress the protein. Purified BG9588 specifically binds to human, baboon, and rhesus CD40L. Lyophilized BG9588 was administered intravenously for 30 minutes after reconstitution with preservative-free sterile water. It was given every 14 days for 3 doses (days 1, 15, and 29) and then every 28 days for 4 doses (days 57, 85, 113, and 141), for a total of 7 doses.
Safety evaluations, including monitoring of adverse events and vital signs and administration of routine laboratory tests, were performed during each followup visit. Patients were monitored for a 12-month period after initiation of therapy, to assess safety. Pharmacokinetics, laboratory results, and scores on the Safety of Estrogens in Lupus Erythematosus: National Assessment–Systemic Lupus Erythematosus Disease Activity Index (45, 46) and Short Form 36 (47) were recorded at regular intervals. The presence of antibodies to rubella was determined prior to dosing and at 3 months after the last dose. Adverse events were recorded throughout the study. Laboratory and clinical efficacy evaluations, complete blood cell counts, determination of prothrombin times, activated partial thromboplastin times, and erythrocyte sedimentation rates, serum chemistry studies, testing for antinuclear antibodies, anti-dsDNA antibodies, and complement components C3 and C4, lymphocyte subset analysis, and 24-hour urine protein and creatinine determination were performed at a central laboratory. Anti-dsDNA antibodies were tested by Farr assay; C3 and C4 concentrations and lymphocyte subsets were determined by standard methods. To assure consistency and reproducibility at each site, a formal protocol for urine collection and evaluation was established and followed. Urinalysis was performed at the local laboratories of each participating center; the second morning specimen (10–13 cc) was collected.
Outcomes and data analysis.
The primary outcome was a ≥50% reduction of proteinuria from baseline at 2 consecutive visits between days 57 and 141 of treatment, without worsening of renal function. Secondary outcomes included the determination of safety and pharmacokinetics of BG9588 as well as the determination of the effects of BG9588 on serum C3 concentrations, hematuria, cellular casts, and anti-dsDNA antibody.
All P values shown are 2-sided. Values are expressed as the mean ± SD. Changes in the absolute values and percentage change from baseline at different time points were compared by repeated-measures analysis of variance. Adjustments for multiple comparisons were made using the Bonferroni/Dunn method. All statistical analyses were performed with the Statview V.5 statistical software package (SAS Institute, Cary, NC).
Baseline characteristics of the 28 patients who entered the trial are shown in Table 1. A significant number of patients exhibited features that have been associated with poor renal outcomes (1), such as male sex (6 patients), African American race (10 patients), nephrotic-range proteinuria (15 patients), mixed membranous and proliferative disease (8 patients), and abnormal serum creatinine level (8 patients). The mean duration of nephritis was 2.5 years, and patients had previously received a variety of immunosuppressive drugs, including cyclophosphamide in 19. All but 1 patient had anti-dsDNA antibodies, and 21 of the 28 had low C3 concentrations. Only 2 patients had high-titer IgG anticardiolipin antibodies, but they had not previously experienced thrombotic complications. Concomitant therapy included azathioprine (3 patients), methotrexate (1 patient), and glucocorticoids (25 patients).
|Age, mean ± SD years||34.7 ± 8.8|
|Duration of SLE, mean ± SD years||8.3 ± 6.4|
|Duration of nephritis, mean ± SD years||2.5 ± 5.2|
|Nephrotic-range proteinuria (>3 gm/24 hours)||15|
|Abnormal creatinine (>1.1 mg/dl)||8|
|Low C3 (<88 mg/dl)||21|
|Low C4 (<16 mg/dl)||14|
|Anti-dsDNA antibodies (>7.2 IU/ml)||27|
|Mixed membranous and proliferative||8|
|Previous immunosuppressive therapy|
|Concomitant immunosuppressive therapy|
Patient exposure and adverse events.
The number of doses of BG9588 received by the patients ranged from 1 to 5. Eighteen patients received 3 or more doses, 6 received 2 doses, and 4 received 1 dose. The total cumulative dose received in the 27 evaluable patients (the amount administered to 1 patient was not recorded) ranged from 18.5 to 99.6 mg/kg, and the mean cumulative dose was 55.8 mg/kg. Twenty-seven patients reported 1 or more adverse events. The most frequently reported adverse events were headache (32%), fatigue (25%), chest pain (21%), and pharyngeal pain (18%). Severe or moderately severe adverse events occurred in 17 patients (61%).
Table 2 shows the serious adverse events that occurred during the study period. There were 2 non–Q-wave myocardial infarctions during the study. In the first patient (a 29-year-old man), the myocardial infarction developed within 9 days after the first infusion; in the other (a 40-year-old woman), the infarct occurred in the second month of the study, after the fourth infusion. Both patients recovered fully. Renal function deteriorated in 1 patient. Evidence of worsening renal function was already present in this patient during the screening period, and it continued to deteriorate after a single infusion of study drug.
|Event*||Day after first infusion||Severity as judged by investigator||Comment|
|Fever and chills||29||Mild|
|Fetal death||145||Moderate||Patient receiving known teratogenic treatment (coumadin prophylaxis) during pregnancy|
|Worsening renal function||2||Severe||Worsened during screening period; progressed to end-stage renal disease after a single BG9588 infusion|
Effects on lymphocytes and immunoglobulins.
CD19+ cells had increased significantly on day 28 after the last dose (mean 95 cells/μl at baseline and 120 cells/μl posttreatment; P < 0.05), but returned to baseline concentrations within 1 month later (Table 3). There were no statistically significant changes in neutrophil or total lymphocyte counts (including T cell subsets such as CD4+ and CD8+ cells), hematocrit values, platelet counts, or serum anticardiolipin antibodies after therapy. Serum immunoglobulin (IgA, IgG, and IgM) concentrations exhibited transient reductions from baseline to 28 days after the last dose (IgA mean 324 to 284 mg/dl, IgG 1,047 to 911 mg/dl, IgM 214 to 194 mg/dl; these reductions were neither clinically nor statistically significant and resolved by day 84. Twenty-seven patients were immune to rubella prior to infusions and remained immune at the end of therapy.
|Variable (normal range)||Study entry (n = 28)||Days after last infusion|
|28 (n = 26)||56 (n = 25)†||84 (n = 21)|
|IgG (672–1,440 mg/dl)||1,047 ± 497||911 ± 401||ND||1,062 ± 466|
|IgM (57–285 mg/dl)||214 ± 709||194 ± 621||ND||245 ± 833|
|IgA (59–396 mg/dl)||324 ± 128||284 ± 130||ND||311 ± 145|
|All (0.91–4.28 × 103/μl)||1,195 ± 584||1,148 ± 362||1,025 ± 371||1,953 ± 400|
|CD4 (404–1,612/μl)||396 ± 218||376 ± 115||356 ± 74||340 ± 67|
|CD8 (220–1,129/μl)||462 ± 247||423 ± 38||396 ± 129||366 ± 111|
|CD19 (80–616/μl)||95 ± 85||120 ± 60||99 ± 45||88 ± 44|
A total of 18 patients who received at least 3 infusions were evaluated for efficacy (Table 4). Seven of these patients had nephrotic-range proteinuria at baseline. The mean 24-hour urinary protein level decreased after treatment, but changes from baseline were not statistically significant (Table 4). Two patients had 50% reductions in proteinuria; reductions in anti-dsDNA antibodies were also noted in these patients. When the definition of response was modified to 40% or 30% reduction in proteinuria (rather than 50%) without worsening of renal function, a total of 3 patients (17%) and 4 patients (22%), respectively, qualified as responders. One patient who received 5 infusions showed reduction of proteinuria by 50% on the day of the fourth infusion (day 57). This was associated with a decrease in anti-dsDNA titer and increase in C3 concentration. Another patient exhibited a nonsustained reduction of proteinuria also associated with a decrease in anti-dsDNA titers after 3 infusions. Hematuria disappeared in all 5 patients who had significant hematuria at baseline. Reductions in anti-dsDNA titers and increases in C3 concentrations were observed in 4 of these 5 patients.
|Variable (normal range)||Study entry||Days after last infusion|
|Proteinuria (<0.15 gm/24 hours)||3.26 ± 2||2.81 ± 1.4||2.86 ± 1.3||2.49 ± 1.8|
|Creatinine (0.4–101 mg/dl)||0.88 ± 0.3||0.86 ± 0.1||0.87 ± 0.14||0.86 ± 0.2|
|Creatinine clearance (75–125 ml/minute/1.7 m2)||114 ± 54||106 ± 48||116 ± 30||116 ± 31|
|Anti-dsDNA antibody titer (<3.61 IU/ml)||121 ± 240||63 ± 96||58 ± 73||86 ± 144|
|C3 (88–201 mg/dl)||90 ± 50||106 ± 63||101 ± 43||96 ± 38|
|C4 (16–47 mg/dl)||25 ± 20||25 ± 16||30 ± 32||27 ± 25|
|Anticardiolipin antibody titers|
|GPL (<23 units)||16 ± 7||11 ± 5||14 ± 2||17 ± 8|
|MPL (<11.0 units)||3 ± 3||4 ± 4||4 ± 4||5 ± 8|
All patients except 1 had elevated anti-dsDNA antibody titers at baseline. Among the 18 patients evaluated for efficacy, there was a significant reduction in anti-dsDNA titers and a significant increase in mean serum C3 concentrations following treatment (Figures 1 and 2). Anti-dsDNA titers decreased from a mean of 121.7 IU/ml at baseline to 68.1 IU/ml on day 29 of treatment. This reduction was sustained through day 56 posttreatment (mean 63.8 IU/ml). Antibody titers returned to baseline levels by day 168 after the last dose (Figure 1A). Because anti-dsDNA antibody titers varied over a wide range among individual patients, we also analyzed the percent change in anti-dsDNA antibody levels in each patient. This analysis confirmed a significant treatment-related decline in anti-dsDNA antibody titers (Figure 1B), with mean reductions of 25.1% (P < 0.005) on day 29 and 23.9% (P < 0.005) and 18.9% (P < 0.05), respectively, 1 month and 2 months after the last treatment. Mean C3 concentrations increased during the treatment phase, with a maximal increase of 18% (P < 0.005) 28 days after the last treatment, and declined after BG9588 was discontinued (Figure 2).
The goal of this open-label pilot study was to obtain safety information about BG9588 in patients with lupus nephritis and assess its effects on renal function. Reduction of proteinuria, provided that renal function was stable, was the primary clinical end point of the study; reduction in proteinuria is almost invariably observed in patients with proliferative lupus nephritis that responds to therapy, and it often portends a favorable long-term prognosis (48–51). Because CD40–CD40L interactions are essential for autoantibody production in animals and humans (27–43), we used anti-dsDNA antibody titers and serum complement concentrations as markers of biologic activity. Of the 18 patients in whom efficacy could be evaluated, 2 had a 50% reduction in proteinuria without worsening of renal function. Substantial reductions in anti-dsDNA antibody titers and improvements in serum C3 concentrations were observed. Hematuria disappeared in all 5 patients who had significant hematuria at baseline; the small number of patients with hematuria at baseline probably reflects difficulties in assessing urine sediment in hospital or commercial laboratories (48, 49). The study was terminated prematurely after 2 patients in this study had myocardial infarctions and following serious thromboembolic events in other concurrent BG9588 protocols.
This study demonstrates that, similar to findings in models using lupus-prone animals, a short course of anti-CD40L antibodies reduces anti-dsDNA production and improves C3 concentrations in patients with SLE. The concurrent improvement in serologic activity and disappearance of hematuria in 5 patients (with improvement of proteinuria in 2 patients) also suggest that treatment had favorable effects in some patients. Despite the absence of a control group, the observed temporal relationship between the biologic response and drug treatment supports the notion of a true effect.
Results obtained ex vivo further demonstrate the biologic effects of BG9588. Huang et al studied peripheral blood lymphocytes from 5 patients treated for various periods during this study (52). With treatment there was a reduction in the frequency of IgG- and IgM-secreting B cells, and the frequency of B cells secreting IgG and IgM anti-dsDNA antibodies decreased by as much as 99% from baseline values. In some patients, these cells were not detectable until 168 days after the last treatment. In yet another cohort of patients treated with BG9588 during this study, Grammer et al (53) and Gur et al (54) reported normalization of aberrations in the peripheral B cell compartment of patients with active lupus, such as increased numbers of activated CD5bright, CD38bright, and CD19+ B cells (these cells spontaneously proliferate and secrete Ig in vitro). B cells expressing CD154 (CD40L) or CD69 and germinal center B cells (founders, centroblasts, and centrocytes) also disappeared from the periphery after treatment (55). In animal models of lupus nephritis, anti-CD154 treatment was shown to interfere with the autostimulatory loop that sustains the autoimmune response, by decreasing the enhanced apoptosis of splenocytes and limiting the increased proliferation and migration of splenic dendritic cells (56).
In SLE patients, CD40L may have roles beyond those involved in antibody production (36–38, 57). Immunohistologic analysis demonstrates that glomerular and tubular CD40 expression is markedly up-regulated in proliferative lupus nephritis (33). Inflammatory cytokines that are up-regulated in lupus nephritis, most notably interferon-γ (58, 59), enhance CD40 expression on dendritic cells, monocytes, epithelial cells, and endothelial cells in vitro. Ligation of CD40L to CD40 1) up-regulates endothelial cell surface adhesion molecules, including E-selectin, intercellular adhesion molecule 1 (ICAM-1), and vascular cell adhesion molecule 1; 2) up-regulates ICAM-1 on B cells, dendritic cells, and fibroblasts; and 3) induces the secretion of proinflammatory cytokines and chemokines from a variety of cells, including renal tubular epithelial cells (60). Anti-CD40L monoclonal antibody therapy markedly reduces the incidence and severity of nephritis in lupus-prone mice, even when administered after the disease is established (28). Collectively, these findings suggest that CD40–CD40L interactions are involved not only in the initiation and maintenance of the pathogenic immune response, but also in the effector phase of established nephritis.
Anti-CD40L (CD154) administration has been shown to prevent acute renal allograft rejection in nonhuman (rhesus monkey) primates without depleting T or B cells (61). In that study, long-term survivors lost their mixed lymphocyte reactivity in a donor-specific manner, but still formed donor-specific antibody and generated T cells that infiltrated the grafted organ without any obvious effect on graft function. Similar to studies in nonhuman primates, the present study did not demonstrate a global depletion of CD4+, CD8+, or CD19+ cells. However, the possibility exists that depletion of lymphocyte subsets (especially by clearance of activated T cells that bear CD40L) occurred but was not detectable with the methodology we used. In the same study in nonhuman primates (61), glucocorticoids and calcineurin inhibitors that limit the expression of CD40L on T cells antagonized the antirejection effects of this antibody. In our study, patients receiving calcineurin inhibitors, but not those receiving glucocorticoids, were excluded. It is possible that the concomitant use of glucocorticoids may have reduced the efficacy of this antibody.
CD40L is essential for the growth, differentiation, resistance to apoptosis, and effector functions of B cells (60). CD40L stimulation after syngeneic bone marrow transplantation in mice accelerates both immune and hematopoietic recovery (62). It is surprising that the number of B cells increased after treatment. The clinical significance of this increase and its mechanism are uncertain. Lymphocytopenia is a feature of active lupus, and this increase may reflect improvement in overall lupus activity. The increase in B lymphocyte number without a concomitant increase in T lymphocytes may reflect the superior capacity of the former to regenerate.
Mutations in the CD40L gene are associated with a rare immunodeficiency state, the X-linked hyper-IgM syndrome. Patients with this syndrome have very low concentrations of IgG, IgA, and IgE but have elevated titers of IgM. Mice lacking either CD40 or CD40L produce essentially no IgG, IgA, or IgE responses to T cell–dependent antigens (63, 64). It is encouraging that a short course of therapy did not result in a clinically significant decrease in serum immunoglobulin concentrations or in a change in immune status against rubella. However, the effect of anti-CD40L on protective immunity and overall risk for infections requires further investigation.
The thrombotic effects observed in this and other trials raise important issues regarding future studies in humans. Among patients with SLE, patients with idiopathic thrombocytopenic purpura, and transplant recipients, there have been several cases of thrombotic events, including atherothrombotic central nervous system events (n = 3), myocardial infarction (n = 2), pulmonary embolism (n = 2; 1 fatal) and deep vein thrombosis (n = 1). All events reported in lupus patients (n = 3) were observed in patients without demonstrable antiphospholipid antibodies.
Platelets express CD40L within seconds after activation in vitro and during thrombus formation in vivo (65). Kawai et al treated monkeys with a different anti-CD40L (ATTC 5C8.33) and observed an unusually high incidence of thromboembolic complications (66). When fresh thrombi were analyzed, CD40L was expressed in areas where platelets had not yet formed an amorphous mass and on platelets directly adhering to the vessel endothelium (66). In the same study, administration of heparin in conjunction with the antibody reduced the frequency of thromboembolic complications. The relative contribution of these in vitro findings to the observed thrombotic events remains to be defined. In a study using another humanized monoclonal IgG antibody against CD40L, IDEC-131, produced by IDEC Pharmaceuticals (San Diego, CA), treatment of SLE patients proceeded without apparent thromboembolic events (34), although a thromboembolic event in a Crohn's disease patient was recently reported, leading to a halt in all IDEC-131 studies (67).
Recent data suggest that CD40L stabilizes arterial thrombi by a β3 integrin–dependent mechanism (68); it is conceivable that inhibition of these interactions by anti-CD40L may render platelet plugs unstable and thus ready to embolize. Alternatively, interaction of the antibody with the platelet Fc receptors may promote platelet aggregation and thrombosis.
In summary, this pilot study of anti-CD40L in patients with proliferative nephritis has provided preliminary information regarding the role of CD40L in the production of autoantibodies and immunologic activity in these patients. Although the data obtained regarding clinical outcomes are encouraging, concerns about the potential prothrombotic effects of anti-CD40L agents should be addressed prior to further evaluation in humans.
The authors thank Drs. John H. Klippel and Howard A. Austin, III for support and useful discussions regarding the design of the protocol; Dr. Peter Lipsky for support and insightful comments; Emily Stark for editorial assistance; and the referring physicians, the patients, and nurses at each site. We also thank the following protocol coordinators for their invaluable assistance with the study: C. H. Yarboro and M. Wilson (NIH), Laurie Payne, RN (North Shore–Long Island Jewish Health System), Christine Kochi and Megan O'Leary (Medical Operations, Biogen), and Jimmy Scaramucci, MPH (Biostatistics, Biogen).
- 23Renal flares are common in patients with severe proliferative lupus nephritis treated with pulse immunosuppressive therapy: long-term followup of a cohort of 145 patients participating in randomized controlled studies. Arthritis Rheum 2002; 46: 995–1002., , , , , , et al.
- 46Reliability of SELENA SLEDAI and flare as clinical trial outcome measures [abstract]. Lupus 1998; 7 Suppl: 110., , .
- 47SF-36 health survey: manual and interpretation guide. Boston: The Health Institute, New England Medical Center; 1993., , , .
- 53Normalization of peripheral B cells following treatment of active SLE patients with humanized anti-CD154 Mab (5C8, BG9588) [abstract]. Arthritis Rheum 2001; 44 Suppl 9: S282., , , , , .
- 54Analysis of CD5 expression on peripheral B cells following treatment of active SLE patients with humanized anti-CD154 mAb (5C8, BG9588) [abstract]. Arthritis Rheum 2001; 44 Suppl 9: S282., , , , , .
- 56Apoptosis and altered dendritic cell homeostasis in lupus nephritis are limited by anti-CD154 treatment. J Immunol 2001; 157: 1740–7., , .
- 67IDEC Pharmaceuticals Web site. Accessed Jan. 20, 2003. URL: http://www.idecpharm.com/site/science/idec131.htm.