Efficacy and safety of infliximab in patients with ankylosing spondylitis: Results of a randomized, placebo-controlled trial (ASSERT)

Authors

  • Désirée van der Heijde,

    Corresponding author
    1. University Hospital Maastricht, Maastricht, The Netherlands
    • University Hospital Maastricht, PO Box 5800, Maastricht 6202 AZ, The Netherlands
    Search for more papers by this author
    • Drs. van der Heijde, Dijkmans, Geusens, Sieper, and Braun have received research support from Centocor, Inc. and have received consulting fees or honoraria totaling less than $10,000 from Centocor, Inc. or Schering-Plough Research Institute. In 2002, Dr. van der Heijde gave expert testimony at the FDA hearing on inhibition of structural damage and received a fee from Centocor, Inc. Drs. DeWoody and Williamson own stock in Johnson and Johnson, of which Centocor, Inc. is a subsidiary.

  • Ben Dijkmans,

    1. VU University Medical Center, Amsterdam, The Netherlands
    Search for more papers by this author
    • Drs. van der Heijde, Dijkmans, Geusens, Sieper, and Braun have received research support from Centocor, Inc. and have received consulting fees or honoraria totaling less than $10,000 from Centocor, Inc. or Schering-Plough Research Institute. In 2002, Dr. van der Heijde gave expert testimony at the FDA hearing on inhibition of structural damage and received a fee from Centocor, Inc. Drs. DeWoody and Williamson own stock in Johnson and Johnson, of which Centocor, Inc. is a subsidiary.

  • Piet Geusens,

    1. Limburg Universitaire Centrum, Maastricht, The Netherlands
    Search for more papers by this author
    • Drs. van der Heijde, Dijkmans, Geusens, Sieper, and Braun have received research support from Centocor, Inc. and have received consulting fees or honoraria totaling less than $10,000 from Centocor, Inc. or Schering-Plough Research Institute. In 2002, Dr. van der Heijde gave expert testimony at the FDA hearing on inhibition of structural damage and received a fee from Centocor, Inc. Drs. DeWoody and Williamson own stock in Johnson and Johnson, of which Centocor, Inc. is a subsidiary.

  • Joachim Sieper,

    1. Universitaetsklinikum Benjamin Franklin–Free University of Berlin, Berlin, Germany
    Search for more papers by this author
    • Drs. van der Heijde, Dijkmans, Geusens, Sieper, and Braun have received research support from Centocor, Inc. and have received consulting fees or honoraria totaling less than $10,000 from Centocor, Inc. or Schering-Plough Research Institute. In 2002, Dr. van der Heijde gave expert testimony at the FDA hearing on inhibition of structural damage and received a fee from Centocor, Inc. Drs. DeWoody and Williamson own stock in Johnson and Johnson, of which Centocor, Inc. is a subsidiary.

  • Kimberly DeWoody,

    1. Centocor, Inc., Malvern, Pennsylvania
    Search for more papers by this author
    • Drs. van der Heijde, Dijkmans, Geusens, Sieper, and Braun have received research support from Centocor, Inc. and have received consulting fees or honoraria totaling less than $10,000 from Centocor, Inc. or Schering-Plough Research Institute. In 2002, Dr. van der Heijde gave expert testimony at the FDA hearing on inhibition of structural damage and received a fee from Centocor, Inc. Drs. DeWoody and Williamson own stock in Johnson and Johnson, of which Centocor, Inc. is a subsidiary.

  • Paul Williamson,

    1. Centocor, Inc., Malvern, Pennsylvania
    Search for more papers by this author
    • Drs. van der Heijde, Dijkmans, Geusens, Sieper, and Braun have received research support from Centocor, Inc. and have received consulting fees or honoraria totaling less than $10,000 from Centocor, Inc. or Schering-Plough Research Institute. In 2002, Dr. van der Heijde gave expert testimony at the FDA hearing on inhibition of structural damage and received a fee from Centocor, Inc. Drs. DeWoody and Williamson own stock in Johnson and Johnson, of which Centocor, Inc. is a subsidiary.

  • Jürgen Braun

    1. Rheumazentrum Ruhrgebiet, Herne, Germany
    Search for more papers by this author
    • Drs. van der Heijde, Dijkmans, Geusens, Sieper, and Braun have received research support from Centocor, Inc. and have received consulting fees or honoraria totaling less than $10,000 from Centocor, Inc. or Schering-Plough Research Institute. In 2002, Dr. van der Heijde gave expert testimony at the FDA hearing on inhibition of structural damage and received a fee from Centocor, Inc. Drs. DeWoody and Williamson own stock in Johnson and Johnson, of which Centocor, Inc. is a subsidiary.


Abstract

Objective

The signs and symptoms of ankylosing spondylitis (AS) respond inadequately to nonsteroidal antiinflammatory drugs, corticosteroids, and disease-modifying antirheumatic drugs in quite a number of patients. Tumor necrosis factor inhibitors have demonstrated success in reducing AS disease activity in a limited number of clinical trials. The objective of this multicenter, randomized, placebo-controlled study was to evaluate the efficacy and safety of infliximab in patients with AS.

Methods

Patients were randomly assigned to receive infusions of placebo or 5 mg/kg infliximab at weeks 0, 2, 6, 12, and 18. Efficacy was assessed using the ASsessment in Ankylosing Spondylitis (ASAS) International Working Group criteria, the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), night pain, patient's global assessment, the Bath Ankylosing Spondylitis Functional Index (BASFI), the Bath Ankylosing Spondylitis Metrology Index (BASMI), chest expansion, the Mander enthesis index, the total swollen joint index, the C-reactive protein level, and the Short Form 36 (SF-36) health survey questionnaire. The primary end point in this study was the proportion of patients with a 20% improvement response according to the ASAS International Working Group criteria (ASAS20 responders) at week 24.

Results

Of the 357 patients screened, 201 were assigned to receive 5 mg/kg infliximab and 78 were assigned to receive placebo. After 24 weeks, 61.2% of patients in the infliximab group were ASAS20 responders compared with 19.2% of patients in the placebo group (P < 0.001). Clinical benefit was observed in patients receiving infliximab as early as week 2 and was maintained over the 24-week study period. Patients receiving infliximab also showed significant improvements in the BASDAI, BASFI, BASMI, chest expansion, and physical component summary score of the SF-36. Adverse events were reported by 82.2% of patients receiving infliximab and by 72.0% of patients receiving placebo; however, most adverse events in both treatment groups were mild or moderate in severity.

Conclusion

Infliximab was well tolerated and effective in a large cohort of patients with AS during a 24-week study period.

Ankylosing spondylitis (AS) is a chronic, progressive, immune-mediated inflammatory disorder that affects 0.1–1.1% of the Caucasian adult population worldwide (1). Inflammation of sacroiliac joints, the spine, and entheses leads to new bone formation, syndesmophytes, and ankylosis of joints (2–4). Patients with AS may also have associated peripheral arthritis, enthesitis, osteoporosis, or extraarticular involvement such as uveitis and inflammatory bowel disease (2). Compared with the general population, patients with AS have increased rates of work disability and unemployment (5) and a higher mortality rate (6, 7). Although the exact etiology of AS is unknown, ∼90% of patients with the disease have the HLA–B27 allele (8).

Until recently, treatment options for patients with AS have been limited to nonsteroidal antiinflammatory drugs (NSAIDs), disease-modifying antirheumatic drugs (DMARDs), and physical therapy. NSAIDs are often used continuously and are not efficacious for some patients (2, 9). They relieve pain sufficiently to allow increased spinal movement and an improved ability to exercise, but they appear to have little effect on the underlying inflammatory process (10). DMARDs that have shown efficacy in rheumatoid arthritis, such as methotrexate, do not have efficacy in patients with spinal manifestations of AS (10, 11). There is evidence that sulfasalazine may provide benefit for the peripheral articular manifestations of AS, but it has not been shown to be effective in treating axial disease (12–14).

Therapeutic agents that target the proinflammatory cytokine tumor necrosis factor α (TNFα) may be the most viable alternative to existing treatment options. Abundant TNFα messenger RNA and protein have been found in the sacroiliac joints of patients with AS (15, 16). Etanercept, a dimeric fusion protein of the TNF receptor and the Fc portion of IgG1, has been successfully used to treat patients with AS (17, 18). Infliximab, a monoclonal antibody that targets TNFα, has also been shown to be effective in treating patients with AS (19–26) and other spondylarthropathies (27–32). In a randomized, placebo-controlled, multicenter study (22), patients who received an induction regimen of 5 mg/kg infliximab showed significant improvement in disease activity after 12 weeks compared with patients who received placebo. The response was sustained over a 1-year, open-label observation period (24). Immunologic studies conducted in a subset of patients from this trial demonstrated that infliximab down-regulated interferon-γ and TNFα secreted by T cells (33).

Although the approved standard of care for maintenance infliximab treatment of patients with rheumatoid arthritis is infusions every 8 weeks, there is evidence that patients with AS would need to receive infusions every 6 weeks. In an open-label trial, 48 subjects with AS were treated with an induction regimen of 5 mg/kg infliximab (26). The mean time to relapse (defined as a loss of ≥50% in the global pain improvement observed at week 8) was 7 weeks (34) (median 14 weeks [26]) after the induction regimen. A second difference from the use of infliximab for rheumatoid arthritis is that methotrexate therapy has not been administered concomitantly in patients with AS.

Although the results of prior studies were promising, the effect of infliximab induction and maintenance therapy needed to be evaluated in a larger patient population. The primary objective of the current study, conducted by members of the Ankylosing Spondylitis Study for the Evaluation of Recombinant Infliximab Therapy (ASSERT) Study Group, was to assess the reduction in the signs and symptoms of AS at week 24 after an induction regimen of 5 mg/kg infliximab followed by maintenance infusions every 6 weeks without concomitant methotrexate therapy.

PATIENTS AND METHODS

Patients.

Adult patients classified as having AS (according to the modified New York criteria [35]) for at least 3 months prior to screening, with a Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) score of ≥4 (range 0–10), and with a spinal pain assessment score of ≥4 on a visual analog scale (VAS; range 0–10 cm) were eligible for the study. Patients were also required to have a normal chest radiograph within 3 months prior to randomization and either a negative purified protein derivative (PPD) skin test result for latent tuberculosis (in the US and Canada) or adequate screening with documented negative results for latent tuberculosis using local guidelines for high-risk or immunocompromised patients (in Europe). Patients were excluded from the study if they had any of the following diagnoses or medical history: total ankylosis of the spine (defined by syndesmophytes present on the lateral views of spinal radiographs at all intervertebral levels from T6 through S1), any other inflammatory rheumatic disease, fibromyalgia, a serious infection within 2 months prior to randomization, tuberculosis (active or latent) or recent contact with a person with active tuberculosis, an opportunistic infection within 6 months of screening, hepatitis, human immunodeficiency virus, a transplanted organ, malignancy, multiple sclerosis, or congestive heart failure.

Patients were allowed to receive concurrent stable doses of NSAIDs, acetaminophen (paracetamol), or tramadol during the study. Patients were not permitted to receive sulfasalazine or methotrexate within 2 weeks prior to screening, systemic corticosteroids within 1 month prior to screening, anti-TNF therapy other than infliximab within 3 months prior to screening, infliximab at any time prior to screening, DMARDs other than sulfasalazine or methotrexate within 6 months prior to screening, or cytotoxic drugs within 12 months prior to screening.

The study was conducted in 33 centers throughout the US, Canada, and Europe. The study protocol was reviewed and approved by the respective institutional review board or independent ethics committee at each site. All patients provided written informed consent.

Study agents.

Infliximab is a recombinant IgG1-κ human-murine chimeric monoclonal antibody that specifically binds to both the soluble and membrane-bound forms of TNFα. Both infliximab and placebo were supplied as sterile, white, lyophilized powders in single-use 20-ml vials.

Study protocol.

In this 24-week, double-blind, placebo-controlled study, patients were randomly assigned in a 3:8 ratio to receive infusions of placebo or 5 mg/kg infliximab at weeks 0, 2, 6, 12, and 18. Patients were allocated to treatment groups using an adaptive treatment allocation stratified by investigational site and C-reactive protein (CRP) level (within or above 3 times the upper limit of normal).

Efficacy evaluations.

The primary end point was the proportion of patients with a 20% improvement response according to the criteria of the ASsessment in Ankylosing Spondylitis (ASAS) International Working Group (ASAS20 responders) (36) at week 24. An ASAS20 responder was defined as a patient who showed at least 20% improvement from baseline and had an absolute improvement from baseline of at least 1 unit (on a scale of 0–10) in at least 3 of the following 4 assessment domains: patient's global assessment, spinal pain, function according to the Bath Ankylosing Spondylitis Functional Index (BASFI), and morning stiffness (the average of the last 2 questions of the BASDAI). In addition, ASAS20 responders must not have had deterioration from baseline (defined as a worsening of ≥20% and an absolute worsening of at least 1 unit [on a scale of 0–10]) in the potential remaining assessment domain.

Secondary end points included the ASAS40 response (40% improvement from baseline and an absolute improvement of at least 2 units [on a scale of 0–10] in at least 3 of the 4 assessment domains defined in the ASAS20 response criteria, with no deterioration from baseline in the potential remaining assessment domain), ASAS partial remission (an absolute score of <2 in each of the above 4 ASAS assessment domains) (36), and 20% improvement in at least 5 of the following 6 ASAS assessment domains (an ASAS 5 of 6 response): spinal pain, patient's global assessment, function according to the BASFI, morning stiffness, CRP level, and the Bath Ankylosing Spondylitis Metrology Index (BASMI) score (37). Other secondary end points were disease activity, physical function, range-of-motion assessments, other musculoskeletal assessments, and quality of life.

Disease activity was evaluated using the BASDAI, the night pain VAS, the patient's global assessment, and the CRP level. The BASDAI (score ranging from 0 to 10) is a combined assessment of fatigue, spinal pain, joint pain, enthesitis, and morning stiffness. In addition to the change from baseline in the BASDAI score, the proportion of patients who had at least 50% improvement in the BASDAI score was also assessed for each treatment group (38). Physical function was evaluated using the BASFI (score ranging from 0 to 10), which includes 8 questions relating to the patient's function and 2 questions relating to a patient's ability to cope with everyday life (39). Range of motion was assessed using the BASMI and chest expansion. The BASMI is an aggregate score (ranging from 0 to 10) of patient mobility assessments, including tragus-to-wall, lumbar flexion (Schober test), cervical rotation, lumbar side flexion, and intermalleolar distance (40). Chest expansion is the difference between the circumference of the chest in maximal inspiration and that in maximal expiration.

Other musculoskeletal assessments included the total swollen joint index and the enthesis index. The swollen joint index, an assessment tool typically used in studies of rheumatoid arthritis, is an evaluation of swelling in 44 joints, including the elbows, wrists, metacarpophalangeal joints of the fingers and thumbs, interphalangeal joints of the thumbs, proximal interphalangeal joints of the fingers, shoulders, acromioclavicular joints, sternoclavicular joints, knees, ankles (mortices), and metatarsophalangeal joints (41). The Mander enthesis index is an assessment (score ranging from 0 to 90) of the degree of tenderness at a variety of anatomic sites and regions (42). The Short Form 36 (SF-36) health survey questionnaire was used to assess quality of life (43).

Safety evaluations.

Safety assessments included adverse events, infections, infusion reactions, premature discontinuation, and laboratory tests. Serum samples were assessed at baseline and at week 24 for the presence of antinuclear antibodies and antibodies to infliximab. Samples that tested positive for antinuclear antibodies were also tested for antibodies to double-stranded DNA (dsDNA).

Statistical analysis.

Assuming that 60% of infliximab-treated patients and 25% of placebo-treated patients achieved the primary study end point, the study required only 40 subjects in each treatment group to achieve 90% power. The targeted sample size for the study was 75 patients in the placebo group and 200 patients in the infliximab group in order to provide a sufficient number of patients for safety assessments. With this sample size, the study had >99% power to detect a treatment effect using a 2-sided chi-square test at the α = 0.05 level given the abovementioned assumptions of response rates.

All requirements for inclusion of patients in analysis populations were prespecified. All patients were analyzed for efficacy according to their randomly assigned treatment group, regardless of the actual treatment they received. All patients who received at least 1 study infusion were analyzed for safety according to the actual treatment they received. The proportions of ASAS20 responders (primary end point) and the proportions of patients who achieved the ASAS partial remission were analyzed using the Cochran-Mantel-Haenszel chi-square test stratified by the CRP value observed at screening (values within versus values above 3 times the upper limit of normal).

Patients in whom non–study-related anti-TNF therapy was initiated or who prematurely discontinued study treatment due to lack of efficacy were considered to be nonresponders. Data for patients who initiated prohibited antiinflammatory medications (e.g., systemic corticosteroids, cytotoxic drugs, or DMARDs) or who discontinued study treatment for reasons other than lack of efficacy (e.g., adverse events) were assessed for treatment response based on their actual observed values at week 24. The last observation carried forward approach was used in the ASAS20 and ASAS partial remission analyses to impute the missing value for patients who did not return for an evaluation or for whom there were insufficient data for us to assess their treatment response at week 24. Secondary efficacy end points were not stratified by CRP level, and no adjustment was made for initiation of anti-TNF therapies, discontinuation of treatment, or missing values. Continuous measures were analyzed using an analysis of variance on the van der Waerden normal scores. All statistical tests were 2-sided and were performed at α = 0.05 with no statistical adjustments for multiple comparisons. All analyses and summaries were conducted using SAS software, version 8.2 (SAS Institute, Cary, NC).

RESULTS

Data for this study were collected between November 2002 and September 2003. Overall, 279 patients were randomly assigned to treatment with placebo (78 patients) or infliximab (201 patients). The study population, typical of patients with active AS, consisted mostly of men (80.6%), had a median age of 40.0 years, and had a median AS disease duration of 8.8 years (Table 1). Eighty-seven percent of patients were positive for the HLA–B27 allele. Median scores for baseline measurements of pain and disease activity showed that most patients had significant AS disease activity (Table 1). Patients also had significant physical impairment at baseline with an overall median SF-36 physical component summary score (28.9) well below that of the general population of the US and Europe (range 49.7–52.7) (44). However, the median SF-36 mental component summary score (47.3) was similar to that of the general population of the US and Europe (range 47.6–54.0) (44).

Table 1. Summary of baseline demographics and clinical characteristics*
 Placebo (n = 78)5 mg/kg infliximab (n = 201)
  • *

    Except where indicated otherwise, values are the median (interquartile range). IBD = inflammatory bowel disease; BASDAI = Bath Ankylosing Spondylitis Disease Activity Index; BASFI = Bath Ankylosing Spondylitis Functional Index; BASMI = Bath Ankylosing Spondylitis Metrology Index; VAS = visual analog scale; CRP = C-reactive protein; ULN = upper limit of normal; SF-36 = Short Form 36 health survey questionnaire.

  • Only 200 patients were assessed at baseline.

  • Normal range 0–0.5 mg/dl.

Men, no. (%)68 (87.2)157 (78.1)
Caucasian, no. (%)76 (97.4)197 (98.0)
Age, years41.0 (34.0, 47.0)40.0 (32.0, 47.0)
Disease duration, years13.2 (3.7, 17.9)7.7 (3.3, 14.9)
History of uveitis, no. (%)25 (32.1)72 (35.8)
History of psoriasis, no. (%)5 (6.4)16 (8.0)
History of IBD, no. (%)6 (7.7)13 (6.5)
HLA–B27 positive, no. (%)69 (88.5)173 (86.5)
BASDAI score, 0–106.5 (5.2, 7.1)6.6 (5.3, 7.6)
BASFI score, 0–106.0 (4.1, 7.2)5.7 (4.5, 7.1)
BASMI score, 0–104.0 (2.0, 6.0)4.0 (2.0, 5.0)
Chest expansion, cm3.0 (2.0, 4.0)3.0 (2.0, 4.0)
Mander enthesis index, 0–908.0 (2.0, 16.0)8.0 (3.0, 15.0)
Swollen joint index, 0–440.0 (0.0, 1.0)0.0 (0.0, 1.0)
Night pain, 0–10 VAS6.7 (4.7, 7.9)6.6 (4.8, 8.1)
Patient's global assessment, 0–10 VAS6.7 (5.8, 7.7)6.9 (5.7, 8.0)
CRP level, mg/dl1.7 (0.7, 3.3)1.5 (0.7, 3.2)
 Screening CRP level ≤3 times ULN, no. (%)38 (48.7)95 (47.3)
 Screening CRP level >3 times ULN, no. (%)40 (51.3)106 (52.7)
SF-36 summary scores  
 Physical component30.1 (24.9, 36.2)28.8 (23.8, 33.7)
 Mental component45.0 (33.7, 55.5)47.6 (37.6, 54.9)

The treatment groups were well matched with regard to demographics and baseline clinical characteristics; however, there was a higher percentage of men in the placebo group (87.2%) compared with that in the infliximab group (78.1%). Although the median baseline disease duration in the placebo group was higher than that in the infliximab group (13.2 years versus 7.7 years), the means were similar between treatment groups (10.1 years and 11.9 years in the infliximab and placebo groups, respectively). The median baseline CRP level was 1.5 mg/dl in the infliximab group and 1.7 mg/dl in the placebo group.

The majority of patients in both treatment groups completed the 24-week study period (Figure 1). Two patients in the infliximab group discontinued study infusions because of adverse events, 1 patient withdrew due to lack of efficacy, and 1 patient was withdrawn because of a protocol violation (positive tuberculin PPD test result with induration of >5 mm). In the placebo group, 1 patient discontinued because of adverse events, and 1 patient withdrew consent.

Figure 1.

Patient disposition. Patients may have discontinued study treatment but continued participation in the study. * = Of the 2 patients in the placebo group who discontinued study treatment, 1 patient discontinued participation in the study. ** = Of the 4 patients in the infliximab group who discontinued study treatment, 2 patients discontinued participation in the study.

One patient was assigned to the placebo group but received an infusion of infliximab. This patient was included in the placebo group for the efficacy analysis and in the infliximab group for the safety analysis. Two other patients were assigned to the placebo group but withdrew consent before receiving any study infusions. These patients were included in the placebo group for the efficacy analysis but were not included in the safety analysis.

Efficacy.

After 24 weeks, 123 of 201 patients (61.2%) in the infliximab group were ASAS20 responders compared with 15 of 78 patients (19.2%) in the placebo group (P < 0.001) (Figure 2A). A difference between the treatment groups was observed as early as week 2 and was maintained over the 24-week observation period (P < 0.001) (Figure 2A). Treatment response was observed in both CRP level strata. For patients with baseline CRP levels ≤3 times the upper limit of normal, 46.3% of 95 infliximab-treated patients achieved an ASAS20 response compared with 21.1% of 38 placebo-treated patients (P = 0.007). For patients with baseline CRP levels >3 times the upper limit of normal, 74.5% of 106 infliximab-treated patients achieved an ASAS20 response compared with 17.5% of 40 placebo-treated patients (P < 0.001).

Figure 2.

Proportion of patients who achieved A, 20% and B, 40% improvement responses over time according to the criteria of the ASsessment in Ankylosing Spondylitis (ASAS) Working Group (see Patients and Methods for explanations of ASAS Working Group criteria). ∗ = P < 0.001 versus placebo group.

Significantly more patients in the infliximab group were also ASAS40 responders over the 24-week observation period compared with those in the placebo group (P < 0.001) (Figure 2B). At week 24, 93 patients in the infliximab group (47.0%) and 9 patients in the placebo group (12.0%) were ASAS40 responders (P < 0.001). In addition, at week 24, 97 patients in the infliximab group (49.0%) achieved 20% improvement in 5 of the 6 ASAS assessment domains compared with 6 patients in the placebo group (8.0%) (P < 0.001) (Figure 3). Also, at week 24, 45 patients in the infliximab group (22.4%) achieved ASAS partial remission compared with 1 patient in the placebo group (1.3%) (P < 0.001) (Figure 3).

Figure 3.

Proportion of patients at week 24 with 20% improvement in 5 of the 6 ASsessment in Ankylosing Spondylitis (ASAS) Working Group response domains and proportion of patients at week 24 with ASAS partial remission (see Patients and Methods for explanations of ASAS Working Group criteria). ∗ = P < 0.001 versus placebo group.

Measures of disease activity (BASDAI score, night pain, patient's global assessment, and CRP level) and physical function (BASFI score) were significantly improved in the infliximab group compared with the placebo group (Table 2). Significantly more patients in the infliximab group had at least a 50% improvement in BASDAI score from baseline to week 24 compared with those in the placebo group (101 patients [51.0%] versus 8 patients [10.7%], respectively). Ninety-four patients (47.5%) in the infliximab group improved by ≥2 in their BASFI score compared with 10 patients (13.3%) in the placebo group (P < 0.001).

Table 2. Summary of median change or median percent change from baseline to week 24 in major clinical outcomes*
 Placebo5 mg/kg infliximabP
  • *

    Values are the median (interquartile range) change from baseline except for chest expansion, patient's global assessment, and CRP level, which are the median (interquartile range) percent change from baseline. See Table 1 for definitions.

BASDAI score, 0–10−0.4 (−1.4, 0.7)−2.9 (−4.9, −0.9)<0.001
BASFI score, 0–100.0 (−1.0, 1.0)−1.7 (−3.6, −0.6)<0.001
BASMI score, 0–100.0 (−1.0, 0.0)−1.0 (−1.0, 0.0)0.019
 Tragus-to-wall0.0 (−1.0, 1.0)−0.5 (−2.0, 0.7)0.013
 Lumbar flexion0.4 (−0.2, 1.0)0.5 (−0.3, 1.0)0.749
 Cervical rotation0.0 (−7.0, 5.0)5.0 (−2.0, 15.0)<0.001
 Lumbar side flexion0.3 (−1.0, 2.7)2.0 (−0.3, 4.5)0.011
 Intermalleolar distance0.0 (−9.0, 7.0)3.5 (−5.0, 12.0)0.045
Chest expansion, cm0.0 (−25.0, 42.9)16.7 (−16.7, 66.7)0.037
Mander enthesis index, 0–90−3.0 (−8.0, 0.0)−3.0 (−9.0, 0.0)0.800
Swollen joint index, 0–440.0 (0.0, 1.0)0.0 (−1.0, 0.0)0.019
Night pain, 0–10 VAS−0.3 (−1.7, 0.9)−2.9 (−5.6, −0.8)<0.001
Patient's global assessment, 0–10 VAS6.1 (−16.4, 30.0)49.2 (11.4, 77.3)<0.001
CRP level, mg/dl0.0 (−31.7, 25.0)−68.7 (−85.2, −16.7)<0.001
SF-36 summary scores   
 Physical component0.8 (−1.9, 6.0)10.2 (3.9, 17.1)<0.001
 Mental component2.0 (−2.6, 7.5)2.7 (−2.9, 8.8)0.547

Patients treated with infliximab in this trial demonstrated statistically significant improvement in measures of range of motion compared with those treated with placebo. The median improvement from baseline in BASMI scores was 1.0 in the infliximab group and 0.0 in the placebo group (P = 0.019) (Table 2). Patients in the infliximab group had significantly greater improvement in all components of the BASMI except for lumbar flexion (Table 2). Improvement in chest expansion was also significantly greater for patients in the infliximab group compared with that for patients in the placebo group (median improvement 16.7% versus 0.0%, respectively; P = 0.037). Although no significant change was observed in the Mander enthesis index scores (P = 0.800) (Table 2), there was a significant improvement from baseline at week 24 in the enthesitis component of the BASDAI for patients in the infliximab group (median improvement of 2.9 compared with a median worsening of 0.2 in the placebo group; P < 0.001), as well as a significant improvement in their median swollen joint index score (P = 0.019) (Table 2).

At week 24, patients in the infliximab group showed a significant improvement from baseline in the physical component summary score of the SF-36 compared with patients in the placebo group (median change from baseline 10.2 versus 0.8, respectively; P < 0.001). No significant treatment effect was observed for the mental component summary score of the SF-36 (2.7 versus 2.0, respectively; P = 0.547); however, the patients participating in this trial had approximately normal mental component summary scores at baseline.

Safety.

A higher proportion of patients receiving infliximab (82.2%) experienced ≥1 adverse events compared with patients receiving placebo (72.0%) (Table 3). Most adverse events in both treatment groups were mild or moderate in severity. Forty-three percent of patients receiving infliximab reported ≥1 infections compared with 36.0% of patients receiving placebo. Eleven percent of patients receiving infliximab reported an infusion reaction compared with 9.3% of patients receiving placebo, and the proportion of infusions associated with infusion reactions was low and identical between the treatment groups (2.7%).

Table 3. Incidence of adverse events through week 24*
 Placebo (n = 75)5 mg/kg infliximab (n = 202)
  • *

    Values are the number (%) of patients. ALT = alanine aminotransferase; AST = aspartate aminotransferase.

Patients with any adverse event54 (72.0)166 (82.2)
Patients with any serious adverse event2 (2.7)7 (3.5)
Patients with any infusion reaction7 (9.3)22 (10.9)
Patients with any infection27 (36.0)86 (42.6)
Patients with any serious infection0 (0.0)2 (1.0)
Adverse events occurring in ≥5% of  patients in either treatment group  
 Upper respiratory tract infection11 (14.7)28 (13.9)
 Pharyngitis2 (2.7)21 (10.4)
 ALT level increased3 (4.0)19 (9.4)
 Headache6 (8.0)18 (8.9)
 Rhinitis2 (2.7)15 (7.4)
 Diarrhea4 (5.3)11 (5.4)
 Pain4 (5.3)11 (5.4)
 AST level increased2 (2.7)11 (5.4)
 Fatigue3 (4.0)10 (5.0)
 Pruritus5 (6.7)8 (4.0)
 Nausea8 (10.7)7 (3.5)
 Arthritis4 (5.3)6 (3.0)
 Rash4 (5.3)5 (2.5)

Seven patients (3.5%) who received infliximab reported serious adverse events, including dizziness (1 patient); cholecystitis (1 patient); arthritis (1 patient); leukocytosis and pneumonia (1 patient); inguinal hernia (1 patient); hemiparesis (1 patient); and abdominal pain, back pain, fever, and ganglioneuroma (1 patient). None of the serious adverse events reported by patients in the infliximab group led to discontinuation of study infusions. However, 2 patients in the infliximab group discontinued study infusions due to nonserious adverse events, including chills (1 patient) and otitis (1 patient). Two patients in the placebo group (2.7%) reported serious adverse events, including pain and arthralgia (1 patient) and myelitis (1 patient). The myelitis event led to the only adverse event–related withdrawal from the study in the placebo group. No deaths, malignancies, or tuberculosis events were reported in this trial.

Adverse events that were reported in at least 5% of patients in either group occurred at comparable or lower rates in the infliximab group compared with the placebo group except for pharyngitis, rhinitis, and increased liver enzymes (Table 3). Nausea was reported by 3.5% of infliximab-treated patients and by 10.7% of patients receiving placebo. Increases in levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were reported in 19 (9.4%) and 11 (5.4%) of the infliximab-treated patients, respectively, and in 3 (4.0%) and 2 (2.7%) of the patients receiving placebo, respectively. Twelve infliximab-treated patients (5.9%) had postbaseline ALT or AST values that were ≥2 times higher than the baseline value and ≥3 times higher than the upper limit of the normal range. These criteria were not met by patients in the placebo group. None of the infliximab-treated patients with increased ALT or AST values had their treatment discontinued or had their enzyme increases reported as serious adverse events. None of the patients in this study developed symptoms of clinical hepatitis. As of week 24, 4 of the 12 patients had liver enzyme levels that had returned to normal or were decreasing.

The majority of patients in the infliximab group (93%) had an inconclusive result for antibodies to infliximab at week 24 due to detectable serum concentrations of infliximab, a phenomenon that has been previously described (see refs. 45 and46). However, 6 patients in the infliximab group (3%) had antibodies to infliximab at week 24 with titers of 1:640 or less.

Of the 172 patients in the infliximab group and the 65 patients in the placebo group who were negative for antinuclear antibodies at baseline, 70 infliximab-treated patients (40.7%) were found to be positive with a titer of 1:40 or more at week 24 compared with 6 patients (9.2%) in the placebo group. Twenty-four of 189 patients (12.7%) in the infliximab group who were negative for antibodies to dsDNA at baseline were found to be positive at week 24 compared with 0 of 70 patients (0%), respectively, in the placebo group. Using a titer of 1:320 as the measure of positivity, 16 of 186 patients (8.6%) who were negative for antinuclear antibodies at baseline were found to be positive at week 24 compared with 0 of 69 patients (0%), respectively, in the placebo group.

DISCUSSION

The results of this study indicate that patients receiving infliximab showed substantial improvement in multiple measures of disease activity, physical function, and quality of life. In addition to meeting the primary end point of the study (an ASAS20 response), nearly half of the patients in the infliximab group met the more rigorous ASAS40 response criteria as well as the ASAS 5 of 6 response criteria, which have recently been shown to be better-performing outcome measures for trials of anti-TNFα therapy in patients with AS (37). In addition, the proportion of patients in the infliximab group showing at least 50% improvement in BASDAI score at week 24 (51%) was similar to that observed by Braun et al (22) at week 12 (53%). Finally, patients receiving infliximab showed significant improvement in quality of life, as assessed by the physical component summary score of the SF-36. Although no significant improvement was observed in the mental component summary score of the SF-36, none was expected because patients participating in this trial had nearly normal mental component summary scores at baseline.

The treatment groups were generally well matched with regard to baseline characteristics. Although patients in the placebo group had longer median disease duration at baseline (13.2 years) than those in the infliximab group (7.7 years), the mean disease durations were similar (11.9 years and 10.1 years, respectively). In general, patients in this trial had shorter disease durations than those in the previous randomized controlled trial of infliximab (∼15 years) (22), even though the ages of the respective patient populations were similar (age ∼40 years). Shorter disease duration has recently been shown to be a strong predictor of clinical response to anti-TNFα therapy in patients with AS, presumably because irreversible structural damage occurs as the disease progresses (47). These results suggest that earlier treatment with anti-TNFα therapy may provide a better overall prognosis for patients with AS, and the substantial response exhibited by infliximab-treated patients with relatively short disease duration who participated in the present study provides additional support for early treatment. However, additional studies must be conducted in patients who are recently diagnosed as having AS in order to determine the appropriate time to initiate anti-TNFα therapy.

Infliximab-treated patients showed significant improvement in range of motion in this study and in the previous randomized controlled study (22). In addition to the total BASMI score and chest expansion, patients in the infliximab group showed significant improvements in all components of the BASMI except for lumbar flexion. However, improvements were seen in lumbar side flexion, which may be more responsive to change. Lumbar side flexion has been included as one of the ASAS 5 of 6 response criteria because it had the highest standardized response mean among all 5 BASMI items (37).

Patients receiving infliximab showed significant improvement in assessments of disease activity, physical function, range of motion, and quality of life. However, the results with regard to enthesitis were not consistent. According to the enthesis question on the BASDAI, significant improvement in enthesis pain was observed for patients in the infliximab group. However, no significant difference between the treatment groups was observed in the Mander enthesis index scores. The Mander enthesis index is not widely used in clinical trials or routine clinical practice, and the feasibility of assessment and accuracy of this index have recently been called into question (48). It is possible that the baseline enthesis index scores for this study population were low, thus making it difficult for an instrument with inherent variability to detect a change. Braun et al (22) reported a significant improvement in the number of enthesitic regions of infliximab-treated patients compared with those receiving placebo. Furthermore, a significant improvement was observed in the number of infliximab-treated patients with psoriatic arthritis who had enthesopathies in the Infliximab Multinational Psoriatic Arthritis Controlled Trial (49). Clearly, the effect of infliximab on enthesitis in patients with AS requires further study using a validated assessment tool that is less cumbersome and easier to use at the bedside.

Infliximab was generally well tolerated by patients with AS who participated in this study. Twelve patients receiving infliximab had elevations in transaminase levels; however, none of these patients showed symptoms of hepatitis or discontinued treatment due to the elevations. Although elevations in transaminase levels have been reported in patients with rheumatoid arthritis after receiving infliximab (50), the clinical implications of these elevations require further exploration in patients with AS. No patients in this study discontinued infliximab due to serious adverse events, and only 2 patients discontinued due to nonserious adverse events (chills and otitis). The proportion of infliximab-treated patients who were found to be newly positive for antinuclear antibodies (40.7%) was lower than the rate reported in the product labeling (51%) (50). The incidence of infusion reactions was equivalent between the infliximab and placebo groups, even though patients were not receiving concomitant methotrexate. The overall adverse event profile was comparable with that reported previously for patients with rheumatoid arthritis (45, 51, 52), and no unexpected adverse events were observed.

In conclusion, patients receiving infliximab as monotherapy exhibited significant improvement in the signs and symptoms of AS, physical function, and quality of life. Additional studies must be conducted to determine the impact of infliximab treatment on early disease, the sustainability of the treatment effect, and the safety of infliximab therapy in patients with AS over the long term.

Acknowledgements

The authors wish to acknowledge Scott Newcomer, MS, for his assistance in the preparation of the manuscript. We thank the following investigators who, in addition to the authors of this article, participated in the ASSERT trial: M. Breban (Paris, France); G. Burmester (Berlin, Germany); M. Clark (Chicago, IL); A. Deadhar (Portland, OR); M. Dougados (Paris, France); W. Edwards (Charleston, SC); P. Emery (Leeds, UK); J. Gaston (Cambridge, UK); R. Inman (Toronto, Ontario, Canada); H. Kellner (Munich, Germany); M. Leirisalo-Repo (Helsinki, Finland); W. Maksymowych (Edmonton, Alberta, Canada); L. Moreland (Birmingham, AL); P. Peloso (Iowa City, IA); J. Reveille (Houston, TX); C. Ritchlin (Rochester, NY); M. Schneider (Dusseldorf, Germany); D. Smith (Indianapolis, IN); S. Steinfeld (Brussels, Belgium); E. Veys (Ghent, Belgium); K. de Vlam (Leuven, Belgium); H. Zeidler (Hanover, Germany).

Ancillary