Anakinra, a recombinant human interleukin-1 receptor antagonist (r-metHuIL-1ra), in patients with rheumatoid arthritis: A large, international, multicenter, placebo-controlled trial




To evaluate the safety of anakinra (a recombinant human interleukin-1 receptor antagonist) in a large population of patients with rheumatoid arthritis (RA), typical of those seen in clinical practice.


A total of 1,414 patients were randomly assigned to treatment with 100 mg of anakinra or placebo, administered daily by subcutaneous injection. Background medications included disease-modifying antirheumatic drugs, corticosteroids, and nonsteroidal antiinflammatory drugs, alone or in combination. The primary end point was safety, which was evaluated by adverse events (including infections), discontinuation from study due to adverse events, and death.


Safety was evaluated in 1,399 patients (1,116 in the anakinra group and 283 in the placebo group; 15 patients were randomized but did not receive any study drug) during the initial 6-month, double-blind, placebo-controlled phase of this long-term safety study. Baseline demographics, disease characteristics, and concomitant medications were similar between the 2 groups. The study group included patients with numerous comorbid conditions and a wide range of RA disease activity. Serious adverse events occurred at a similar rate in the anakinra group and the placebo group (7.7% and 7.8%, respectively). Serious infectious episodes were observed more frequently in the anakinra group (2.1% versus 0.4% in the placebo group). The rate of withdrawal due to adverse events was 13.4% in the anakinra group and 9.2% in the placebo group.


Results from this large, placebo-controlled safety study demonstrate that anakinra is safe and well tolerated in a diverse population of patients with RA, including those with comorbid conditions and those using multiple combinations of concomitant therapies. Although the frequency of serious infection was slightly higher in the anakinra group, no infection was attributed to opportunistic microorganisms or resulted in death.

Rheumatoid arthritis (RA) is a systemic autoimmune disorder characterized by pain and inflammation of the joints (1). Although the severity of disease may fluctuate over time, there is often progressive destruction of the articular and periarticular joint structures leading to deformity and disability (2). In an effort to prevent or slow progressive joint destruction and improve long-term outcomes, the trend over the past 10 years has been to use disease-modifying antirheumatic drugs (DMARDs) earlier and more aggressively (3–6). However, many of the commonly used DMARDs are often associated with significant toxicity or diminishing efficacy after a few years of use; as a consequence, many patients are unable to obtain maximally effective disease management over the long term (7–9). Thus, there is a need for new and novel treatments for patients with RA that are safe and effective and that can be used, alone or in combination with existing treatments, for long-term therapy.

The etiology of RA remains poorly understood, but it is thought to be driven by antigen-activated T cells and autoreactive B cells, both of which produce proinflammatory mediators. These proinflammatory mediators, in turn, cause chronic inflammation and joint destruction (10). Studies in both experimental animal models and patients suggest that the cytokine interleukin-1 (IL-1) plays an important role in promoting tissue inflammation and remodeling (11–14). Of particular interest, IL-1 may contribute to joint damage in RA by stimulating the release of matrix metalloproteinases, by inhibiting mechanisms of cartilage repair, and by augmenting bone resorption through activation of osteoclasts (15–18). These observations suggest that inhibition of IL-1 may be a very important interventional strategy for decreasing inflammation and bone resorption in patients with RA.

IL-1 receptor antagonist (IL-1Ra) is an endogenous, competitive antagonist of the IL-1 receptor, which modulates the biologic actions of IL-1 by prevention of signal transduction (19). Administration of IL-1Ra, either systemically or locally by gene transfer, reduces inflammation and joint damage in animal models of RA (20–23). A recombinant form of human IL-1Ra, anakinra (recombinant methionyl human IL-1 receptor antagonist [r-metHuIL-1ra]), is a competitive antagonist of IL-1 that blocks the actions of IL-1 without any detectable agonist activity. In previously completed controlled clinical trials, anakinra provided significant symptomatic improvement when used as monotherapy or in combination with methotrexate (24, 25). In addition, anakinra monotherapy slowed the radiographically observed progression of RA (24, 26).

We undertook this large, randomized, placebo-controlled trial to characterize further the safety profile of anakinra, a new biologic agent with a novel mechanism of action, in a patient population with a diverse range of RA disease activity and a variety of comorbid conditions. The trial was designed to assess overall safety as measured by treatment-emergent adverse events that may have a low incidence of occurrence and for which a larger number of patients is required to permit detection. The study consists of an initial 6-month randomized, double-blind, placebo-controlled phase followed by an open-label phase in which patients continue in the study and are treated with anakinra for up to 30 additional months. The results of the initial 6-month placebo-controlled treatment phase of this trial are presented here.



Patients ages 18 years and older with a diagnosis of RA based on the American College of Rheumatology (formerly, the American Rheumatism Association) 1987 diagnostic criteria (27) were enrolled at 169 centers in the US, Canada, Europe, and Australia. The study protocol was approved by the institutional review boards or institutional ethics committees of the participating centers, and all patients gave written informed consent before any study-specific procedures were performed. Patients were required to have had RA for at least 3 months, with evidence of active disease defined by a minimum of 3 swollen joints and 3 tender or painful joints, or 45 minutes of morning stiffness.

To achieve conditions similar to those expected in conventional, real-world rheumatology practice, patients were allowed to use nonsteroidal antiinflammatory drugs (NSAIDs), corticosteroids, and DMARDs (with the exception of tumor necrosis factor [TNF] inhibitors), either alone or in any combination. Before randomization, patients already using NSAIDs and/or oral corticosteroids were required to have been receiving stable doses for at least 1 month, and those using DMARDs (other than TNF inhibitors) were required to have been receiving stable doses for at least 2 months. Changes in these medications were allowed during the course of the study if such changes were clinically indicated by the treating physician.

Although the study was designed to exclude as few patients as possible, patients were excluded if they had uncontrolled medical conditions (e.g., diabetes with a hemoglobin A1c value >8%), a malignancy other than basal cell carcinoma of the skin or in situ carcinoma of the cervix within the previous 5 years, Felty's syndrome, leukopenia (white blood cell count <2.0 × 109/liter), neutropenia (neutrophil count <1.0 × 109/liter), thrombocytopenia (platelet count <100 × 109/liter), or an abnormal liver function test result (aspartate aminotransferase or alanine aminotransferase level ≥1.5-fold the upper limit of normal), or if they were known to be positive for hepatitis B, hepatitis C, or human immunodeficiency virus. Also excluded were pregnant or breastfeeding women.

Study drug.

Anakinra (Amgen, Thousand Oaks, CA) was supplied in vials as a clear, colorless, sterile solution containing 100 mg r-metHuIL-1ra. Placebo, consisting of the vehicle used in the preparation of anakinra, was provided in similar vials. Both anakinra and placebo were self-administered daily by subcutaneous injection (1.0 ml). Patients, health care providers, and the sponsor were blinded to the identity of the drug being administered.

Study design.

This was a prospective, multicenter, double-blind, randomized, placebo-controlled study conducted at 169 sites in 9 countries (the US, Australia, Belgium, Canada, Germany, Ireland, Netherlands, Sweden, and the UK) in patients with RA who were representative of those seen by physicians in the typical community practice setting.

The sample was planned to include 1,000 patients, although 1,399 patients were eventually enrolled and received study drug. With a sample size of 1,000 and a 4-to-1 allocation ratio (800 anakinra patients to 200 placebo patients), the probability of observing at least 1 serious adverse event or infectious episode at a 1% incidence rate was 100% for the anakinra group and 87% for the placebo group.

The study was designed to include a 6-month, placebo-controlled, double-blind phase followed by a 30-month, open-label phase. All patients who completed 6 months of treatment during the blinded phase of the study, including those in the placebo group, were eligible to receive open-label anakinra, 100 mg daily, for the remaining 30 months of the study. Treatment compliance was assessed by drug accountability and patient diaries.

Safety end points.

The objective of this study was to assess the safety of anakinra, administered over a period of 36 months, in patients with RA of varying severity and with various comorbid conditions. This report presents results from the initial 6-month, placebo-controlled phase. Safety was assessed by summarizing reports of adverse events, serious adverse events, infections, discontinuation from study due to adverse events, and death. All adverse events were grouped according to the body system affected and by preferred term within the body system according to a modified World Health Organization adverse reaction term dictionary (28).

Anti–IL-1ra immunoassay and bioassay.

The development of anti–IL-1ra antibodies was assessed in serum samples collected at regular intervals. Three methods were used. The first method was an enzyme-linked immunosorbent assay (ELISA), which served as a screening assay. Reactive samples were further processed in a BIAcore 3000 biosensor immunoassay (BIAcore, Stevenage, UK) to verify whether ELISA reactivity was attributable to anti–IL-1ra antibodies. Samples that were BIAcore-reactive were further processed in a cell-based bioassay. The cell-based bioassay was designed to determine whether serum antibodies (if present) were capable of neutralizing a biologic effect of IL-1ra in vitro.

Statistical analysis.

Statistical analyses were conducted using intent-to-treat analysis of all patients who were randomized and received at least 1 dose of study drug. Baseline demographic and clinical characteristics were summarized by mean (± SD) values for continuous measures and by numbers (%) for categoric measures. For adverse events, the number and the proportion of patients experiencing at least 1 adverse event were summarized by preferred term and treatment group. Comparisons between treatments in the time to an event, such as withdrawal from the study, were made using the log rank test. Comparisons in proportions of patients, such as proportions of patients with adverse events, were made using Fisher's exact test.


A total of 1,414 patients were randomly assigned to treatment in a 4-to-1 (anakinra-to-placebo) allocation ratio, and 1,399 patients received at least 1 dose of study drug or placebo. Of these 1,399 patients, 79.0% (1,105 of 1,399) completed 6 months of treatment, with 78.4% (875 of 1,116) receiving anakinra and 81.3% (230 of 283) receiving placebo.

The 2 treatment groups were well balanced at baseline with respect to demographics, concomitant DMARD or corticosteroid therapy (Table 1), and baseline disease status (Table 2). The demographics of the patients randomized in this study were similar to those reported in previous trials, with a majority of the patients being white women with a mean age of 55 years. Patients who were enrolled had a broad spectrum of RA disease activity (Table 2), a variety of comorbid conditions (Table 3), and varied usage of concomitant DMARDs and corticosteroids (Table 4). Approximately 30% of patients received methotrexate (MTX) alone, 22% were treated with MTX in addition to other DMARDs, and 21% received no DMARD therapy (Figure 1). Almost half (47%) of patients were treated with a combination of corticosteroids and DMARDs, while 10.5% received neither DMARDs nor corticosteroids.

Table 1. Baseline patient demographics*
CharacteristicPlacebo (n = 283)Anakinra (n = 1,116)
  • *

    Except where indicated otherwise, values are the number (%). Tumor necrosis factor inhibitors were excluded from the list of allowable disease-modifying antirheumatic drugs (DMARDs). NSAID = nonsteroidal antiinflammatory drug.

  • Mean corticosteroid dose (prednisone equivalent): for placebo 6.62 mg/day (n = 159), for anakinra 6.60 mg/day (n = 572). Some steroids could not be converted into a prednisone equivalent dose due to lack of conversion factors for these steroids.

  • Mean methotrexate dose: for placebo 14.7 mg/week (n = 168), for anakinra 14.6 mg/week (n = 579).

 Men72 (25.4)282 (25.3)
 Women211 (74.6)834 (74.7)
 White255 (90.1)980 (87.8)
 Hispanic15 (5.3)68 (6.1)
 Black10 (3.5)49 (4.4)
 Other3 (1.1)19 (1.7)
Age, mean ± SD years55.7 ± 13.454.6 ± 12.3
Weight, mean ± SD kg75.7 ± 18.177.1 ± 19.7
Concomitant drug use  
 NSAID244 (86.2)973 (87.2)
 Corticosteroid172 (60.8)636 (57.0)
 Methotrexate168 (59.4)579 (51.9)
 Other DMARDs (excluding   methotrexate)135 (47.7)532 (47.7)
  2 DMARDs29 (10.2)120 (10.8)
  3 DMARDs7 (2.5)20 (1.8)
  >3 DMARDs0 (0.0)2 (0.2)
Table 2. Disease characteristics at baseline
CharacteristicPlacebo (n = 283)Anakinra (n = 1,116)
  • *

    C-reactive protein values below the detection limit (0.1) are set to 0.09 for summarization.

Disease duration, years  
 Mean ± SD10.7 ± 9.510.2 ± 9.6
Tender/painful joint count  
 Mean ± SD22.6 ± 14.522.6 ± 14.7
Swollen joint count  
 Mean ± SD18.3 ± 11.718.8 ± 11.9
C-reactive protein level, mg/dl*  
 Mean ± SD2.7 ± 3.32.7 ± 3.3
Table 3. Medical history of patients*
Medical historyPlacebo (n = 283)Anakinra (n = 1,116)
  • *

    Values are the percent. COPD = chronic obstructive pulmonary disease.

Coronary artery disease5.657.53
Congestive heart failure3.183.23
Table 4. DMARDs used by >2% of patients at baseline*
DMARDPlacebo (n = 283)Anakinra (n = 1,116)
  • *

    Values are the number (%). Tumor necrosis factor inhibitors were excluded from the list of allowable disease-modifying antirheumatic drugs (DMARDs).

Methotrexate168 (59.4)579 (51.9)
Hydroxychloroquine63 (22.3)239 (21.4)
Sulfasalazine40 (14.1)153 (13.7)
Leflunomide28 (9.9)110 (9.9)
Azathioprine6 (2.1)47 (4.2)
Figure 1.

Use of disease-modifying antirheumatic drugs (DMARDs) at baseline. MTX = methotrexate.

Patients in both groups were exposed to the study drug for a median of 0.471 years (Table 5). Among those completing the first 6 months of blinded treatment, compliance was high, with >40% of patients missing no injections and >90% of patients receiving ≥90% of their intended doses.

Table 5. Summary of treatment compliance*
CharacteristicPlacebo (n = 283)Anakinra (n = 1,116)
  • *

    Unless indicated otherwise, values are the number (%). Compliance with use of study drug is calculated as a percentage of the total doses taken compared with the total intended doses during the 6 months of treatment.

Duration of exposure to study  medication, years  
 Mean ± SD0.426 ± 0.1270.410 ± 0.150
No. of missed doses as a percentage  of intended doses, mean ± SD3.5 ± 8.64.3 ± 10.6
Patients who completed 6 months  of treatment230 (81.3)875 (78.4)
Patients in compliance with use of  study drug  
 100% compliant110 (47.8)383 (43.8)
 90–99% compliant102 (44.3)431 (49.2)
 80–89% compliant11 (4.8)48 (5.5)
 70–79% compliant3 (1.3)6 (0.7)
 <70% compliance4 (1.7)7 (0.8)

Adverse events were reported by 1,027 (92.0%) and 261 (92.2%) patients in the anakinra and placebo groups, respectively (Table 6). Injection-site reactions (ISRs) were the most commonly reported adverse event, occurring in 72.6% of patients in the anakinra group and 32.9% of those in the placebo group. Most ISRs were of mild or moderate severity and were transient in nature. In any individual patient, the first ISR typically occurred during the first month of treatment; if no such reaction was seen by this time, it was unlikely to occur. The most frequently reported ISRs in the anakinra group were erythema, pruritus, and rash of mild-to-moderate severity; all of these ISRs occurred more frequently with anakinra than with placebo. In contrast, the most common ISRs in the placebo group were pain and ecchymosis, which occurred at a similar rate in both treatment groups. Other than ISRs, the most common adverse event was worsening of symptoms of RA, occurring in 223 anakinra patients (20% of those reporting adverse events) and 78 placebo patients (27.6% of those reporting adverse events). These adverse event rates are consistent with those observed in previous clinical studies of anakinra in patients with RA (24, 25). In the anakinra group, the rate of ISRs was higher among female patients than among male patients (80% versus 50%). No other noteworthy differences in the rate of adverse events according to age, sex, or race were observed.

Table 6. Summary of adverse events*
Adverse eventPlacebo (n = 283)Anakinra (n = 1,116)P
  • *

    Values are the percent. P values were calculated using Fisher's exact test.

Overall adverse events92.292.01.000
Serious adverse events7.87.71.000
Severe adverse events13.115.50.351
Withdrawal due to adverse event9.213.40.057
Infectious episode43.541.20.500
Serious infection0.42.10.068

Serious adverse events occurred at a similar frequency in the 2 treatment groups, with 86 patients in the anakinra group (7.7%) and 22 patients in the placebo group (7.8%) experiencing serious events. Worsening of RA was the only serious adverse event occurring in ≥1% of patients in either treatment group, with rates of 1.8% (placebo) and 0.9% (anakinra). The incidence of infectious episodes was similar in the 2 groups, occurring in 460 patients in the anakinra group (41.2%) and 123 patients in the placebo group (43.5%). The most common infectious episodes involved the respiratory system, most frequently upper respiratory system infection and sinusitis, both of which occurred at a similar rate in the anakinra and placebo groups.

However, serious infections occurred more frequently in the anakinra group (23 patients, 2.1%) than in the placebo group (1 patient, 0.4%). Such infections consisted primarily of pneumonia (10 patients) and cellulitis (3 patients). Of the 10 patients with pneumonia, 5 had a history of underlying pulmonary disease, and 3 had a history of pneumonia at baseline. Of the 3 patients with serious cellulitis, all had risk factors for infection, including 2 patients with diabetes and 1 patient in whom a toe ulcer was present at baseline. The available microbiologic data on cases of serious infection were limited but suggested the presence of commonly seen organisms such as Streptococcus pneumoniae in those with pneumonia and Staphylococcusaureus in those with bone or joint infections. Unusual or opportunistic infections such as tuberculosis, histoplasmosis, listeriosis, and aspergillosis, which have been observed in patients using other biologic RA therapies, were not observed in this study. None of the serious infectious events were fatal, and all had resolved by the time of the last evaluation. In the anakinra group, most patients experiencing serious infections (17 of 23 [74%]) resumed therapy with anakinra once the infection resolved and had no further problems with serious infections.

An analysis of potential risk factors for infection was conducted to evaluate any association between serious infections and patient age, extent of RA disease, duration of RA, concomitant medications (corticosteroids, MTX), diabetes mellitus, chronic heart failure, history of pneumonia, history of asthma, and underlying pulmonary disease. Although the results were not statistically significant, the risk for serious infection appeared to be higher for patients with asthma who were receiving anakinra than for patients with asthma who were receiving placebo. No other factors appeared to be associated with increased risk of infection while using anakinra.

Malignancies were diagnosed in 9 patients during the study (Table 7). Four malignancies were diagnosed in the anakinra treatment group (0.4% of all anakinra patients). In one of these patients, who had a history of melanoma, diffuse metastatic melanoma (brain and lungs) was diagnosed. In the other 3 anakinra patients, the diagnoses were moderately to poorly differentiated adenocarcinoma of the cecum, uterine carcinoma, and basal cell carcinoma, respectively. Of the 5 malignancies diagnosed in the placebo group (1.8% of all placebo patients), 2 were basal cell and squamous cell carcinomas of the skin. The others were Hodgkin's lymphoma, bladder carcinoma, and squamous cell carcinoma of the lung.

Table 7. Malignancies diagnosed in study patients
TreatmentMalignancyStudy day of onsetRelated to treatment*
  • *

    As determined by the investigator (blinded).

PlaceboBasal cell carcinoma (nose)31No
PlaceboSquamous cell carcinoma of the  skin112No
PlaceboHodgkin's lymphoma179No
PlaceboBladder carcinoma151Yes
PlaceboSquamous cell carcinoma of the  lung157No
AnakinraMalignant melanoma105No
AnakinraAdenocarcinoma of the cecum113No
AnakinraUterine carcinoma138No
AnakinraBasal cell carcinoma143No

A total of 241 anakinra patients (21.6%) and 53 placebo patients (18.7%) withdrew prematurely from the placebo-controlled phase of the study. The most common reason for study withdrawal was an adverse event (13.4% in the anakinra group and 9.2% in the placebo group). Other reasons for early withdrawal included ineligibility, noncompliance, consent withdrawn, administrative decision, and lost to followup. The time to study withdrawal did not differ significantly between treatments (P = 0.155). Among anakinra-treated patients, the use of DMARDs (excluding MTX) and MTX was higher in patients who completed the study than in those who did not complete the study (51% versus 35% and 54% versus 44%, respectively).

Five patients died during the placebo-controlled phase of this study, 4 in the anakinra group (0.4%) and 1 in the placebo group (0.4%). None of these deaths was considered unusual for the RA patient population studied. In the anakinra group, deaths were attributed to deterioration of interstitial lung disease (most likely related to chronic MTX therapy), suicide (in a patient with a history of depression), metastatic melanoma (in a patient with a history of melanoma), and upper gastrointestinal hemorrhage (in a patient with a history of gastrointestinal ulcerative disease). The patient in the placebo group died of myocardial infarction. Serum chemistry, hematologic analysis, and urinalysis indicated no evidence of hepatotoxicity, renal toxicity, agranulocytosis, or other drug-related changes.

Formation of anti-anakinra antibodies in patient sera was evaluated at regular intervals over the course of the study. Nine patients (0.8%) seroconverted and developed neutralizing antibodies, as defined by a BIAcore immunoassay and cell-based bioassay. Overall, the serum antibody concentration was transient. Neutralizing antibodies could be detected at either 3 months (n = 6) or 6 months (n = 3); no patient had evidence of neutralization at more than 1 time point. The safety profile of anakinra in these 9 patients was similar to that in the rest of the study population.


This is the first large, placebo-controlled, prospective study of the safety of a biologic response modifier in a clinical setting intended to resemble typical rheumatology practice. The unrestricted severity of patients' disease, use of background medications (excluding TNF inhibitors), and the wide range of comorbid conditions permitted are in notable contrast to the restrictions placed on these variables in typical efficacy studies used to gain approval of new therapies. Previous studies of anakinra have either excluded all concomitant DMARD use (24) or limited concomitant drug use to MTX (25). Studies of other recently introduced biologic response modifiers have had similar restrictions (29–33). It is thus notable that during the initial 6 months of this study, the safety profile of anakinra as administered in the present setting did not differ meaningfully from the safety profile observed in previous studies with more restrictively defined patient samples.

This study was performed to assess the safety profile of anakinra compared with placebo in a large and diverse population of patients with RA. The results showed a very similar safety profile in both treatment groups. In addition, the safety observations from this study were similar to observations in other anakinra studies conducted in more restrictive patient populations (24, 25). During the initial 6 months of the study, no treatment-related difference was apparent in the rates of adverse events (92.0% in the anakinra group versus 92.2% in the placebo group), serious adverse events (7.7% versus 7.8%), infections (41.2% versus 43.5%), discontinuation from the study due to adverse events (13.4% versus 9.2%), malignancy (0.4% versus 1.8%), or death (0.4% in both groups). Patients in both groups remained in the study for the same amount of time and were equally compliant with the medication regimen. Most discontinuations due to adverse events were attributable to ISRs. At the time of this publication (>2 years after study initiation), >50% of the patients initially enrolled remain on open-label treatment with anakinra. These results will be presented in a future report.

Although it was not statistically significant, there was a difference in the rate of serious infectious episodes, which occurred more frequently in the anakinra group (2.1%) than in the placebo group (0.4%). However, a rate of 2.1% is consistent with previous reports in similar RA populations (34). The other notable difference between treatments was the occurrence of ISRs. There were significantly more ISRs in the anakinra group (72.6%) than in the placebo group (32.9%), an observation that is consistent with previous studies of anakinra (24, 25) and other injected biologic agents (29–33) in the treatment of patients with RA. As in those studies, the ISRs observed in the present study were typically mild to moderate. The results of this study demonstrate that anakinra does not appear to increase the risk of death or malignancy in the broad patient group studied, and that it is well tolerated by most patients.

As opposed to previous reports of reactivation of tuberculosis in association with use of other biologic response modifiers approved for use in RA (35), no opportunistic infections (including tuberculosis, histoplasmosis, listeriosis, and aspergillosis) were reported during the first 6 months of this large, prospective study. In addition, anakinra does not appear to cause hepatic toxicity, serious hematologic events, or serious cardiovascular events, as has been reported with use of other DMARDs.

Nine patients had evidence of antibodies that neutralized IL-1ra, according to a COS-1 cell bioassay that used IL-8 production as the assay readout. Although these antibodies have the potential to neutralize IL-1ra in vivo, there was no clinical evidence of adverse effects. Moreover, this response was transient. Although 3 patients had evidence of neutralizing antibodies at month 6 (the last time point described in this report), they have been tested subsequently and were determined to be neutralizing antibody negative (data not shown). It is therefore unlikely that the development of neutralizing anti–IL-1ra antibodies represents a major factor of clinical risk for anakinra.

In conclusion, during the initial 6 months of this placebo-controlled study, anakinra was demonstrated to be safe and well tolerated in a diverse population of patients with RA, including those with comorbid conditions and those using multiple combinations of concomitant therapies. Further evaluation in this population is ongoing, to confirm the safety of anakinra beyond 6 months of treatment.


One hundred sixty-nine investigators participated in this study. The authors would like to thank all members of the 990757 Study Group. Moraye Bear, Robert Baltera, and Barb Thomson provided assistance with the conduct of the study. Antibody testing was performed by the Department of Clinical Immunology, Amgen Inc.