Efficacy and safety of adalimumab in patients with ankylosing spondylitis: Results of a multicenter, randomized, double-blind, placebo-controlled trial

Authors

  • Désirée van der Heijde,

    1. University Hospital Maastricht, Maastricht, The Netherlands
    Search for more papers by this author
    • Dr. van der Heijde has received research support from, or is a consultant for, Abbott, Amgen, Bristol-Myers Squibb, Centocor, Merck, Schering-Plough, and Wyeth. Dr. Schiff has received grant/research support from, is on the Speaker's Bureau of, or is a consultant for, Abbott, Amgen, Aventis, Bristol-Myers Squibb, Centocor, Merck, Novartis, Roche, and Wyeth-Ayerst. Drs. Dijkmans and Dougados have received consulting fees from Abbott. Drs. Wong and Kupper hold stock options as a benefit of their employment with Abbott. Dr. Davis has received research support from, or is a consultant for, Abbott, Amgen, and Centocor.

  • Alan Kivitz,

    1. Altoona Center for Clinical Research, Duncansville, Pennsylvania
    Search for more papers by this author
  • Michael H. Schiff,

    1. Denver Arthritis Clinic Research Unit, Denver, Colorado
    Search for more papers by this author
    • Dr. van der Heijde has received research support from, or is a consultant for, Abbott, Amgen, Bristol-Myers Squibb, Centocor, Merck, Schering-Plough, and Wyeth. Dr. Schiff has received grant/research support from, is on the Speaker's Bureau of, or is a consultant for, Abbott, Amgen, Aventis, Bristol-Myers Squibb, Centocor, Merck, Novartis, Roche, and Wyeth-Ayerst. Drs. Dijkmans and Dougados have received consulting fees from Abbott. Drs. Wong and Kupper hold stock options as a benefit of their employment with Abbott. Dr. Davis has received research support from, or is a consultant for, Abbott, Amgen, and Centocor.

  • Joachim Sieper,

    1. Benjamin Franklin Hospital, Free University Berlin, and German Rheumatism Research Center, Berlin, Germany
    Search for more papers by this author
  • Ben A. C. Dijkmans,

    1. Department of Rheumatology, VU Medical Center, and Jan van Breemen Institute, Amsterdam, The Netherlands
    Search for more papers by this author
    • Dr. van der Heijde has received research support from, or is a consultant for, Abbott, Amgen, Bristol-Myers Squibb, Centocor, Merck, Schering-Plough, and Wyeth. Dr. Schiff has received grant/research support from, is on the Speaker's Bureau of, or is a consultant for, Abbott, Amgen, Aventis, Bristol-Myers Squibb, Centocor, Merck, Novartis, Roche, and Wyeth-Ayerst. Drs. Dijkmans and Dougados have received consulting fees from Abbott. Drs. Wong and Kupper hold stock options as a benefit of their employment with Abbott. Dr. Davis has received research support from, or is a consultant for, Abbott, Amgen, and Centocor.

  • Jürgen Braun,

    1. University Medical Center Ruhrgebiet, Herne, Germany
    Search for more papers by this author
  • Maxime Dougados,

    1. Hôpital Cochin, University of Paris, Paris, France
    Search for more papers by this author
    • Dr. van der Heijde has received research support from, or is a consultant for, Abbott, Amgen, Bristol-Myers Squibb, Centocor, Merck, Schering-Plough, and Wyeth. Dr. Schiff has received grant/research support from, is on the Speaker's Bureau of, or is a consultant for, Abbott, Amgen, Aventis, Bristol-Myers Squibb, Centocor, Merck, Novartis, Roche, and Wyeth-Ayerst. Drs. Dijkmans and Dougados have received consulting fees from Abbott. Drs. Wong and Kupper hold stock options as a benefit of their employment with Abbott. Dr. Davis has received research support from, or is a consultant for, Abbott, Amgen, and Centocor.

  • John D. Reveille,

    1. University of Texas at Houston
    Search for more papers by this author
  • Robert L. Wong,

    1. Abbott Laboratories, Parsippany, New Jersey
    Search for more papers by this author
    • Dr. van der Heijde has received research support from, or is a consultant for, Abbott, Amgen, Bristol-Myers Squibb, Centocor, Merck, Schering-Plough, and Wyeth. Dr. Schiff has received grant/research support from, is on the Speaker's Bureau of, or is a consultant for, Abbott, Amgen, Aventis, Bristol-Myers Squibb, Centocor, Merck, Novartis, Roche, and Wyeth-Ayerst. Drs. Dijkmans and Dougados have received consulting fees from Abbott. Drs. Wong and Kupper hold stock options as a benefit of their employment with Abbott. Dr. Davis has received research support from, or is a consultant for, Abbott, Amgen, and Centocor.

  • Hartmut Kupper,

    1. Abbott GmbH & Co. KG, Ludwigshafen, Germany
    Search for more papers by this author
    • Dr. van der Heijde has received research support from, or is a consultant for, Abbott, Amgen, Bristol-Myers Squibb, Centocor, Merck, Schering-Plough, and Wyeth. Dr. Schiff has received grant/research support from, is on the Speaker's Bureau of, or is a consultant for, Abbott, Amgen, Aventis, Bristol-Myers Squibb, Centocor, Merck, Novartis, Roche, and Wyeth-Ayerst. Drs. Dijkmans and Dougados have received consulting fees from Abbott. Drs. Wong and Kupper hold stock options as a benefit of their employment with Abbott. Dr. Davis has received research support from, or is a consultant for, Abbott, Amgen, and Centocor.

  • John C. Davis Jr.

    Corresponding author
    1. University of California, San Francisco
    • University of California, San Francisco, 533 Parnassus Avenue, Room U386, Box 0633, San Francisco, CA 94143
    Search for more papers by this author
    • Dr. van der Heijde has received research support from, or is a consultant for, Abbott, Amgen, Bristol-Myers Squibb, Centocor, Merck, Schering-Plough, and Wyeth. Dr. Schiff has received grant/research support from, is on the Speaker's Bureau of, or is a consultant for, Abbott, Amgen, Aventis, Bristol-Myers Squibb, Centocor, Merck, Novartis, Roche, and Wyeth-Ayerst. Drs. Dijkmans and Dougados have received consulting fees from Abbott. Drs. Wong and Kupper hold stock options as a benefit of their employment with Abbott. Dr. Davis has received research support from, or is a consultant for, Abbott, Amgen, and Centocor.


Abstract

Objective

To evaluate the safety and efficacy of adalimumab, a fully human recombinant IgG1 monoclonal antibody that specifically targets human tumor necrosis factor, in patients with active ankylosing spondylitis (AS).

Methods

This was a multicenter, randomized (2:1 ratio), double-blind, placebo-controlled study to evaluate a subcutaneous injection of adalimumab, 40 mg every other week, compared with placebo for 24 weeks. The primary efficacy end point was the percentage of patients with a 20% response according to the ASsessment in Ankylosing Spondylitis International Working Group criteria for improvement (ASAS20) at week 12. Secondary outcome measures included the ASAS20 at week 24 and multiple measures of disease activity, spinal mobility, and function, as well as ASAS partial remission.

Results

At week 12, 58.2% of adalimumab-treated patients (121 of 208) achieved an ASAS20 response, compared with 20.6% of placebo-treated patients (22 of 107) (P < 0.001). More patients in the adalimumab group (45.2% [94 of 208]) than in the placebo group (15.9% [17 of 107]) had at least a 50% improvement in the Bath Ankylosing Spondylitis Disease Activity Index at week 12 (P < 0.001). Significant improvements in the ASAS40 response and the response according to the ASAS5/6 criteria at weeks 12 and 24 were also demonstrated (P < 0.001). Partial remission was achieved by more adalimumab-treated patients than placebo-treated patients (22.1% versus 5.6%; P < 0.001). Adalimumab-treated patients reported more adverse events (75.0% versus 59.8% of placebo-treated patients; P < 0.05), but there was no statistically significant difference in the incidence of infections. Most adverse events were mild or moderate in severity.

Conclusion

Adalimumab was well-tolerated during the 24-week study period and was associated with a significant and sustained reduction in the signs and symptoms of active AS.

Ankylosing spondylitis (AS) is a chronic inflammatory disease primarily affecting the axial skeleton, peripheral joints, and entheses. It is prototypical of a group of disorders, the spondylarthritides, which share overlapping features, including enthesitis, sacroiliitis, extraaxial manifestations (e.g., acute anterior uveitis, psoriasis, and inflammatory bowel disease), HLA–B27 positivity, and familial aggregation. AS may result in fusion of the spine, which leads to restriction in spinal mobility and overall function (1). It is likely that the prevalence of AS has been underestimated, and recent estimates indicate a standardized prevalence rate of 0.55% among white patients (2). This is lower than the prevalence of rheumatoid arthritis (RA), and, unlike RA, AS is slightly more common in men than in women and has an earlier age at onset. Because of the early onset and chronicity of the disease, the lifetime costs and socioeconomic impairment in individual AS patients are high (3–5). There is a negative impact on the health-related quality of life in patients with AS and an unmet need for long-term, effective therapy in many patients (6–8).

Tumor necrosis factor (TNF) appears to be key in the inflammatory response observed in AS. In fact, increased concentrations of TNF messenger RNA in biopsy tissues from the sacroiliac joints of AS patients have been reported (9). Until recently, relatively few treatment options for AS existed. However, these provided only symptomatic relief, did not influence spinal mobility or concentrations of acute-phase reactants (e.g., C-reactive protein [CRP]), and have not been shown to affect axial symptoms or the long-term course of disease (1, 10). Current evidence does not support the use of conventional antirheumatic drugs for the treatment of AS (11).

Several placebo-controlled trials have demonstrated the significant and sustained efficacy of TNF blockade in the treatment of AS, using either a monoclonal antibody (infliximab) or a soluble receptor construct (etanercept). Both agents have been shown to result in improvement in symptoms, function, mobility, and concentrations of acute-phase reactants (11–20). Importantly, improvements in health-related quality of life also have been demonstrated (6, 7).

Adalimumab is the first fully human monoclonal antibody with a very high affinity for human TNF. Controlled clinical trials in patients with RA and psoriatic arthritis (PsA) have shown that adalimumab is effective and safe when self-administered subcutaneously every other week (21–25). Long-term studies have demonstrated the sustained efficacy of adalimumab in the treatment of RA, with a safety profile comparable with that of other anti-TNF agents (23, 26–29).

The present study was conducted by the Adalimumab Trial Evaluating Long-term Efficacy and Safety for Ankylosing Spondylitis (ATLAS) Study Group to assess the safety and efficacy of adalimumab treatment in patients with active AS.

PATIENTS AND METHODS

Role of the funding source.

An Advisory Committee was developed for this study, which consisted of many of the authors from academic institutions and Abbott Laboratories. The Advisory Committee and members of the Abbott Laboratories clinical trial team designed the study, with input from several of the study investigators. The study was conducted at 43 centers in the US and Europe. Clinical data were collected and analyzed by Abbott Laboratories. Data analysis was reviewed by the Advisory Committee. All authors reviewed and contributed to the manuscript during its development, agreed to submit the manuscript, and approved the content of the submitted manuscript.

Patient population.

The patients were at least 18 years of age and were classified as having definite AS based on the modified New York criteria (30). All had active disease, which was defined as fulfillment of at least 2 of the following 3 criteria: a Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) score ≥4, a total back pain score ≥4 by visual analog scale (VAS; 0–10 cm), or a duration of morning stiffness ≥1 hour. The percentage of patients with radiographic evidence of total spinal ankylosis of the cervical and lumbar lateral spine was limited a priori to 10% of the study population. Patients with stable and well-controlled psoriasis, uveitis, inflammatory bowel disease (i.e., ulcerative colitis, Crohn's disease), and reactive arthritis were allowed to participate. Inadequate response or intolerance to 1 or more nonsteroidal antiinflammatory drugs (NSAIDs) was defined by the investigators. Patients in whom 1 or more disease-modifying antirheumatic drugs (DMARDs) had failed were also allowed to participate. Patients were allowed to continue any of the following medications if the dosage had remained stable for at least 4 weeks before the baseline visit: sulfasalazine (≤3 gm/day), methotrexate (≤25 mg/week), hydroxychloroquine (≤400 mg/day), prednisone or prednisone equivalent (≤10 mg/day), and NSAIDs.

Patients who had previously received anti-TNF therapy, cyclosporine, azathioprine, or DMARDs (other than the medications and dosages listed above) at any time and patients who had received intraarticular injection(s) with corticosteroids within 4 weeks prior to baseline were excluded. All patients were evaluated for latent tuberculosis (TB) infections using a purified protein derivative skin test and a chest radiograph. Patients with latent TB were allowed to participate in the study if a documented history of treatment was available or if treatment for latent TB was initiated before the first dose of study medication. Patients with clinically active TB were excluded from the study. Additional major exclusion criteria included a history of any recent infections requiring antibiotic treatment; hepatitis or human immunodeficiency virus; a significant history of cardiac, renal, neurologic, psychiatric, endocrinologic, metabolic, or hepatic disease; and a history of demyelinating disease or multiple sclerosis. Patients with a history of cancer or lymphoproliferative disease other than a successfully treated nonmetastatic squamous cell or basal cell carcinoma and/or localized carcinoma in situ of the cervix were not allowed to participate.

This study was conducted at 43 centers in the US and Europe (Belgium, France, Germany, Italy, The Netherlands, Spain, Sweden, and the UK). See Appendix A for additional members of the ATLAS Study Group and their locations. All centers received approval from independent ethics committees, and the study was conducted in accordance with the ethics principles of the Declaration of Helsinki. Investigators assured that the study complied with prevailing local laws and customs. Signed informed consent was obtained from each patient before any study-related procedures were performed or medications were withheld.

Study medications and dosage.

The dosage regimen for this study was chosen based on the demonstrated safety and efficacy of 40 mg of adalimumab delivered subcutaneously every other week in clinical trials of RA and PsA. The study drug was provided as a subcutaneous injection solution in 1-ml prefilled syringes containing either adalimumab 40 mg or matching placebo (Abbott Laboratories, Abbott Park, IL).

Study design.

This was a multicenter, randomized, double-blind, placebo-controlled study. Following a 2-week screening period, enrolled patients were randomly assigned in a 2:1 ratio to receive adalimumab 40 mg every other week or placebo, respectively, for a 24-week period. Treatment efficacy and safety were assessed at scheduled visits at weeks 2, 4, 8, 12, 16, 20, and 24. Patients who did not achieve a 20% response according to the ASsessment in Ankylosing Spondylitis International Working Group criteria for improvement (ASAS20) at weeks 12, 16, or 20 were eligible for “early-escape” open-label treatment with adalimumab 40 mg every other week. Adverse events and other safety assessments were completed throughout the study. Following the initial 24-week period, patients were eligible to receive open-label treatment with adalimumab at the same dosage for up to an additional 80 weeks.

Efficacy evaluations.

Primary efficacy end point.

The clinical response to adalimumab was assessed using the ASAS20 response criteria (31). The primary efficacy end point was the percentage of ASAS20 responders at week 12. An ASAS20 responder was defined as a patient experiencing improvement of at least 20% and absolute improvement of at least 1 unit as compared with baseline in at least 3 of the following 4 domains, with no deterioration (defined as a worsening of at least 20% or an absolute increase of at least 1 unit) in the remaining domain: patient's global assessment of disease activity during the previous week (assessed with a 0–10-cm VAS); patient's assessment of pain during the previous week, represented by the total back pain score (assessed with a 0–10-cm VAS); function, represented by the Bath Ankylosing Spondylitis Functional Index (BASFI) score (assessed with a 0–10-cm VAS) (32); and inflammation, represented by the mean of the severity and duration of morning stiffness (i.e., questions 5 and 6 of the BASDAI [33]) (assessed with a 0–10-cm VAS). Patients who did not achieve an ASAS20 response at weeks 12, 16, or 20 were considered for open-label adalimumab treatment (early-escape open-label group).

Secondary efficacy end points.

The ASAS International Working Group has suggested that, in addition to the ASAS20 response, patients receiving a drug with potential disease-modifying ability should attain a response according to the ASAS5/6 criteria or a 40% improvement response (ASAS40) (34). The ASAS5/6 response criteria require at least 20% improvement in 5 of 6 domains: spinal mobility (according to the Bath Ankylosing Spondylitis Metrology Index [BASMI]; other instruments may be used) and acute-phase reactants (the CRP concentration) in addition to the 4 domains included in the ASAS20 response criteria. The BASMI is a composite index with a range of 0–10, based on a 0–2-point scale for each of 5 clinical measurements: tragus-to-wall distance (in cm), anterior lumbar flexion (modified Schober's test; in cm), lumbar side flexion (in cm), intermalleolar distance (in cm), and cervical rotation (in degrees) (35). The ASAS40 response represents improvement of at least 40% and absolute improvement of at least 2 units (on a scale of 0–10) compared with baseline in at least 3 of the 4 domains of the ASAS20 criteria, with no deterioration in the remaining domain. Partial remission is defined as a value of <2 on a 0–10 scale in each of the 4 domains of the ASAS20.

Additional secondary efficacy end points were performed to further assess metrology, enthesitis, peripheral arthritis, and disease activity. The BASDAI measures the severity of fatigue, spinal and peripheral joint pain, localized tenderness, and morning stiffness (both qualitative and quantitative) as assessed on a 10-cm VAS. A 50% improvement in the BASDAI score is considered clinically meaningful. Chest expansion, defined as the difference in chest circumference (in cm) between maximal inspiration and maximal expiration, was also assessed (36). Enthesitis was assessed by the Maastricht Ankylosing Spondylitis Enthesitis Score (MASES; range 0–13) (37). Additional musculoskeletal assessments included the swollen joint count (of 44 joints) and the tender joint count (of 46 joints). The Bath Ankylosing Spondylitis Patient Global Score (BAS-G) indicates the impact of AS on the patient's well-being over the previous week and the previous 6 months, using a 10-cm VAS (38). The physician's global assessment of disease activity is a physician assessment of the patient's current disease activity, using a 0–10-cm VAS. Nocturnal pain during the previous week was assessed on a 0–10-cm VAS. Self-reported quality-of-life assessments were also completed and will be reported separately.

Safety evaluations.

Adverse events and vital signs were assessed at every visit. Laboratory measurements were assessed at weeks 4, 12, 16, and 24.

Statistical analysis.

Safety and efficacy analyses were performed on the intention-to-treat population, which was defined as all randomized patients who received at least 1 dose of the study medication. At least 90% power to detect a difference between a placebo response rate of 28% and an adalimumab response rate of 59% according to the ASAS20 at week 12, using a chi-square test with an alpha level of 0.05 (Type I error), was assumed to begin trial enrollment.

All response rates for the adalimumab and placebo treatment groups were compared using Pearson's chi-square test. Fisher's exact test was used if at least 25% of cells had expected counts of <5. Patients who withdrew from the study before week 12 or who had missing values were considered nonresponders for purposes of the analysis of categorical variables at week 12. Patients who withdrew from the study before week 24, had missing values at week 24, or entered early-escape therapy before week 24 were classified as nonresponders for purposes of the analysis of discrete variables at week 24. To assess the impact of the missing data, a sensitivity analysis of ASAS20 at week 12 was performed without including nonresponder imputations.

Analysis of covariance was used to compare the mean change in each treatment group for the continuous secondary efficacy variables. The analysis of covariance model included a factor for treatment and was adjusted for baseline. A last observation carried forward analysis was used for continuous end points. For laboratory values, mean changes at week 12 were compared between treatment groups using one-way analysis of variance. For adverse events, Fisher's exact test was used for comparisons between treatment groups. All statistical tests were 2-sided, and comparisons were performed at an alpha level of 0.05, unless stated otherwise.

RESULTS

Data for this study were collected between January 27, 2004 and December 23, 2004. Of the 315 patients enrolled, 208 were randomly assigned to receive adalimumab 40 mg every other week and 107 to receive placebo (Figure 1). Of these patients, 6 (2.9%) in the adalimumab group and 5 (4.7%) in the placebo group had complete ankylosis according to their local rheumatologists. Most patients were white (95.6%), male (74.9%), and positive for the HLA–B27 allele (78.7%) (Table 1). Patients had a mean disease duration of 10.6 years; 30.2% of the patients had a history of uveitis (past or concurrent). Baseline demographic and clinical characteristics were similar across both treatment groups, except for history of psoriasis (Table 1). However, this difference was not considered clinically significant.

Figure 1.

Flow chart showing the disposition of the ankylosing spondylitis study patients from initial screening through week 24. Patients may have withdrawn from the study for more than 1 reason. eow = every other week.

Table 1. Baseline demographic and clinical characteristics of the ankylosing spondylitis patients, by treatment group*
VariablePlacebo (n = 107)Adalimumab (n = 208)
  • *

    Patients were randomized at a ratio of 2:1 to receive treatment with 40 mg of adalimumab or placebo every other week for 24 weeks. IBD = inflammatory bowel disease; VAS = visual analog scale; BASDAI = Bath Ankylosing Spondylitis Disease Activity Index; BASFI = Bath Ankylosing Spondylitis Functional Index; CRP = C-reactive protein; BASMI = Bath Ankylosing Spondylitis Metrology Index (sum of the scores of the 5 individual items that are scored on a 0–2-point scale); MASES = Maastricht Ankylosing Spondylitis Enthesitis Score; BAS-G = Bath Ankylosing Spondylitis Patient Global Score; DMARDs = disease-modifying antirheumatic drugs; NSAIDs = nonsteroidal antiinflammatory drugs.

Male, no. (%)79 (73.8)157 (75.5)
White, no. (%)99 (92.5)202 (97.1)
Age, mean ± SD years43.4 ± 11.3241.7 ± 11.69
Disease duration, mean ± SD years10.0 ± 8.3411.3 ± 9.99
History of uveitis, no. (%)27 (25.2)68 (32.7)
History of psoriasis, no. (%)17 (15.9)16 (7.7)
History of IBD, no. (%)  
 Crohn's disease1 (0.9)6 (2.9)
 Ulcerative colitis1 (0.9)9 (4.3)
History of peripheral arthritis, no. (%)11 (10.3)12 (5.8)
HLA–B27 positive, no. (%)85 (79.4)163 (78.4)
Patient's global assessment of disease activity, mean ± SD cm (0–10-cm VAS)6.5 ± 2.06.3 ± 2.2
Total back pain, mean ± SD cm (0–10-cm VAS)6.7 ± 2.26.4 ± 2.1
Inflammation (mean of questions 5 and 6 of the BASDAI), mean ± SD cm (0–10-cm VAS)6.7 ± 1.96.7 ± 2.0
BASFI, mean ± SD cm (0–10-cm VAS)5.6 ± 2.25.2 ± 2.2
BASDAI, mean ± SD cm (0–10-cm VAS)6.3 ± 1.76.3 ± 1.7
CRP, mean ± SD mg/dl2.2 ± 2.91.8 ± 2.2
BASMI, mean ± SD (range 0–10)4.2 ± 2.13.8 ± 2.2
Chest expansion, mean ± SD cm3.0 ± 1.93.4 ± 1.8
MASES (range 0–13)  
 Mean ± SD score6.7 ± 7.56.4 ± 6.8
 No. (%) with scores >081 (75.7)152 (73.1)
BAS-G, mean ± SD cm (0–10-cm VAS)6.9 ± 1.96.8 ± 1.9
Nocturnal pain, mean ± SD cm (0–10-cm VAS)6.5 ± 2.46.1 ± 2.4
Physician's global assessment of disease activity, mean ± SD cm (0–10-cm VAS)5.8 ± 1.95.6 ± 1.9
Swollen joint count (range 0–44 joints)  
 Mean ± SD (median)1.4 ± 2.8 (0)1.5 ± 3.3 (0)
 No. (%) with ≥1 swollen joint44 (41.1)75 (36.1)
Tender joint count, mean ± SD (median) (range 0–46 joints)5.6 ± 6.8 (3)5.1 ± 7.4 (2)
Concomitant DMARDs, no. (%)22 (20.6)40 (19.2)
 Sulfasalazine15 (14.0)26 (12.5)
 Methotrexate8 (7.5)20 (9.6)
 Leflunomide1 (0.9)0
Concomitant oral corticosteroids, no. (%)6 (5.6)25 (12.0)
Concomitant NSAIDs, no. (%)84 (78.5)166 (79.8)

Most of the adalimumab-treated patients (98.1%) and the placebo-treated patients (96.3%) completed the 12-week double-blind portion of the study (Figure 1). At week 12, 55 patients randomized to placebo (51.4%) were deemed eligible to enter the early-escape period, compared with 54 patients randomized to adalimumab (26.0%). By week 24, 94.0% of patients (296 of 315) remained in the study; however, only 29 of the 107 patients originally randomized to receive placebo (27.1%) completed week 24 while continuing to take the initial study drug; the remaining 74 placebo-treated patients (69.2%) entered the early-escape protocol and received open-label adalimumab. In contrast, only 81 of the 208 patients originally randomized to adalimumab (38.9%) entered the early-escape protocol.

Efficacy.

Of the patients receiving adalimumab, 58.2% (121 of 208) achieved an ASAS20 response at week 12, compared with 20.6% (22 of 107) of the patients receiving placebo, a difference that was highly statistically significant (37.6% [95% confidence interval (95% CI) 27.4–47.8]; P < 0.001). A further sensitivity analysis indicated that the nonresponder imputation did not alter the outcome of the primary end point analysis. A clinical response to adalimumab occurred as early as 2 weeks after treatment, and this magnitude of response at week 8 was sustained through week 24 (Figure 2). Overall, 155 patients (49.2%) received early-escape open-label adalimumab treatment beginning at week 12. Of the 74 patients who had previously received placebo, 44 (59.5%) rapidly became ASAS20 responders at week 16 during open-label adalimumab treatment. By comparison, of the 81 patients who had previously received adalimumab, 32 (39.5%) became ASAS20 responders at week 16 during open-label adalimumab treatment. The response rates for both groups increased over the remaining weeks.

Figure 2.

Percentage of patients who achieved a 20% response over time according to the ASsessment in Ankylosing Spondylitis International Working Group criteria for improvement. ∗ = P < 0.001 versus placebo, based on an intention-to-treat analysis using nonresponder imputation.

Adalimumab treatment was associated with statistically significant improvements in all 4 subcomponents of the ASAS20 response at weeks 12 and 24 (Table 2). For the small percentage of patients with total spinal ankylosis, 50% (3 of 6) of the adalimumab-treated patients achieved an ASAS20 response at week 12, compared with 0% (0 of 5) of the placebo-treated patients (difference of 50% [95% CI 10.0–90.0]). At week 24, 66.7% (4 of 6) of the adalimumab group had an ASAS20 response, compared with 0% (0 of 5) of the placebo group (difference of 66.7% [95% CI 28.9–100.0]).

Table 2. Summary of adjusted mean changes or mean percentage changes in clinical signs and symptoms from baseline to week 12 and from baseline to week 24, by treatment group*
AssessmentBaseline to week 12Baseline to week 24
Placebo (n = 107)Adalimumab (n = 208)PPlacebo (n = 107)Adalimumab (n = 208)P
  • *

    Except where indicated otherwise, values are the adjusted mean ± SEM change. VAS = visual analog scale; BASFI = Bath Ankylosing Spondylitis Functional Index; BASDAI = Bath Ankylosing Spondylitis Disease Activity Index; BASMI = Bath Ankylosing Spondylitis Metrology Index (sum of the scores of the 5 individual items that are scored on a 0–2-point scale); MASES = Maastricht Ankylosing Spondylitis Enthesitis Score; BAS-G = Bath Ankylosing Spondylitis Patient Global Score.

  • P values were determined by analysis of covariance, comparing adalimumab with placebo treatment. If a patient received early-escape open-label adalimumab prior to week 24, the last observation prior to early-escape open-label adalimumab was carried forward when week 24 data were analyzed.

  • Values are the mean ± SEM percentage change.

Patient's global assessment of disease activity (0–10-cm VAS)6.5 ± 6.3−39.1 ± 4.6<0.0018.7 ± 6.9−37.8 ± 5.0<0.001
Total back pain (0–10-cm VAS)−9.5 ± 4.3−40.5 ± 3.1<0.001−10.0 ± 4.3−42.4 ± 3.1<0.001
BASFI (0–10-cm VAS)−8.0 ± 4.0−35.8 ± 2.8<0.001−8.5 ± 4.2−37.7 ± 3.0<0.001
Inflammation (mean of questions 5 and 6 of the BASDAI) (0–10-cm VAS)−15.2 ± 4.9−41.6 ± 3.5<0.001−12.5 ± 5.0−42.9 ± 3.6<0.001
BASDAI (0–10-cm VAS)−0.8 ± 0.2−2.6 ± 0.2<0.001−0.8 ± 0.2−2.6 ± 0.2<0.001
C-reactive protein, mg/dl−0.1 ± 0.1−1.3 ± 0.1<0.001−0.1 ± 0.1−1.3 ± 0.1<0.001
BASMI (range 0–10)0.1 ± 0.1−0.5 ± 0.1<0.0010.0 ± 0.1−0.6 ± 0.1<0.001
 Tragus-to-wall0.1 ± 0.3−0.3 ± 0.20.299−0.1 ± 0.3−0.4 ± 0.30.475
 Anterior lumbar flexion−0.1 ± 0.30.3 ± 0.20.229−0.1 ± 0.30.3 ± 0.20.239
 Lumbar side flexion−0.2 ± 0.41.7 ± 0.3<0.0010.0 ± 0.42.1 ± 0.3<0.001
 Intermalleolar distance0.1 ± 2.17.9 ± 1.50.0021.0 ± 2.27.4 ± 1.50.017
 Cervical rotation0.2 ± 1.43.2 ± 1.00.077−1.0 ± 1.53.6 ± 1.10.011
Chest expansion0.3 ± 0.70.9 ± 0.50.5040.4 ± 0.10.3 ± 0.10.580
MASES (range 0–13)−1.3 ± 0.5−2.7 ± 0.40.018−1.6 ± 0.5−3.2 ± 0.30.005
BAS-G (0–10-cm VAS)−0.6 ± 0.2−2.3 ± 0.2<0.001−0.8 ± 0.2−2.7 ± 0.2<0.001
Nocturnal pain (0–10-cm VAS)−0.8 ± 0.3−2.6 ± 0.2<0.001−0.9 ± 0.3−2.7 ± 0.2<0.001
Physician's global assessment of disease activity (0–10-cm VAS)−1.0 ± 0.2−2.5 ± 0.2<0.001−1.0 ± 0.2−2.5 ± 0.2<0.001
Swollen joint count (range 0–44 joints)−0.5 ± 0.3−0.4 ± 0.20.810−0.4 ± 0.3−0.4 ± 0.20.870
Tender joint count (range 0–46 joints)−0.3 ± 0.6−0.8 ± 0.40.481−0.3 ± 0.6−0.9 ± 0.40.400

Mean BASDAI scores were also significantly improved and sustained in the adalimumab-treated patients (see Table 2). At week 12, 45.2% (94 of 208) of adalimumab-treated patients achieved at least a 50% improvement on the BASDAI, compared with only 15.9% (17 of 107) of placebo-treated patients (difference of 29.3% [95% CI 19.6–39.0]; P < 0.001). This level of improvement was sustained through week 24, with 42.3% (88 of 208) of the adalimumab group achieving a BASDAI 50% response at week 24, compared with 15.0% (16 of 107) of the placebo group (difference of 27.4% [95% CI 17.8–36.9]; P < 0.001).

Adalimumab-treated patients achieved significantly higher ASAS5/6, ASAS40, and partial remission responses at weeks 12 and 24 compared with placebo-treated patients (Figure 3). At week 24, 44.7% of the adalimumab group had at least a 20% improvement in 5 of the 6 ASAS assessment domains, compared with only 12.1% of the placebo group (difference of 32.6% [95% CI 23.4–41.7]; P < 0.001). Similarly, at week 24, 39.4% of the adalimumab group were ASAS40 responders, compared with only 13.1% of the placebo group (difference of 26.3% [95% CI 17.1–35.6]; P < 0.001). In addition, 22.1% of the adalimumab group were in partial remission, compared with only 5.6% of the placebo group (difference of 16.5% [95% CI 9.4–23.6]; P < 0.001).

Figure 3.

Percentages of patients who achieved a 20% response according to the ASsessment in Ankylosing Spondylitis International Working Group 5/6 criteria (ASAS5/6) (i.e., 20% improvement in 5 of 6 domains, consisting of the 4 domains of the ASAS20 criteria plus spinal mobility and acute-phase reactants), a 40% improvement response (ASAS40) (i.e., at least 40% improvement and 20 units of absolute change in 3 of the 4 domains of the ASAS20 criteria, with no worsening in the fourth domain), and partial remission (i.e., a value of <2 on a 0–10 scale in each of the 4 ASAS20 domains) at weeks 12 and 24 of study. ∗ = P < 0.001 versus placebo, based on an intention-to-treat analysis using nonresponder imputation. Numbers at the top of the bars are the percentages represented by the bars.

Compared with placebo treatment, adalimumab treatment was associated with statistically significant improvements in all other signs and symptoms of AS, including CRP concentrations, physician's global assessment of disease activity, BAS-G scores, and nocturnal pain (Table 2). Adalimumab treatment was also associated with statistically significantly improved metrology (BASMI) and enthesitis (MASES) scores at weeks 12 and 24 (Table 2). At week 12, adalimumab-treated patients had a greater reduction in BASMI scores (−0.5) compared with placebo-treated patients (0.1; P < 0.001), and a sustained response was demonstrated at week 24 (−0.58 for adalimumab versus −0.04 for placebo; P < 0.001). The lumbar side flexion, cervical rotation (week 24 only), and intermalleolar distance components of the BASMI demonstrated statistically significant improvement in adalimumab-treated patients compared with placebo-treated patients, but the mean changes in the tragus-to-wall and anterior lumbar flexion components of the BASMI were not statistically different between adalimumab- and placebo-treated patients at weeks 12 or 24. At week 12, adalimumab-treated patients had a statistically significantly greater improvement in the MASES (−2.7) compared with placebo-treated patients (−1.3; P = 0.018), and a sustained response was demonstrated at week 24 (−3.2 for adalimumab versus −1.6 for placebo; P = 0.005). The maintenance of statistically significant improvements in virtually all measures of disease activity through week 24 demonstrates the durability of the response to adalimumab in this 24-week study.

Despite a lower percentage of patients in this study who were HLA–B27 positive as compared with previous clinical studies of patients with AS (15, 17), all patients met the definition of active AS based on the modified New York criteria (30). Of the HLA–B27–positive patients receiving adalimumab at week 12, 62.0% (101 of 163) achieved an ASAS20 response, compared with 23.5% (20 of 85) of those receiving placebo (difference of 38.5% [95% CI 26.7–50.1]; P < 0.001). Of the HLA–B27–negative patients, 43.9% (18 of 41) of those receiving adalimumab achieved an ASAS20 response at week 12, compared with 10.0% (2 of 20) of those receiving placebo (difference of 33.9% [95% CI 13.8–54.0]; P < 0.001). This level of response was maintained through week 24.

Another analysis of this subgroup also indicated no statistically significant treatment difference in the ASAS20 response rate between HLA–B27–positive and HLA–B27–negative patients at week 12 (P = 0.740) or at week 24 (P = 0.212). Additional efficacy analyses of other subgroups indicated no differences in the treatment effect among the various subgroups analyzed.

Safety.

For the 24-week placebo-controlled period of this study, the overall incidence of adverse events reported by the adalimumab-treated patients (75.0%) was statistically significantly higher than that reported by the placebo-treated patients (59.8%), as was the incidence of injection site reactions (10.1% versus 2.8%) (P < 0.05 for both comparisons) (Table 3). There were no other statistically significant differences in the incidence of any other adverse events between treatment groups. The incidence of infectious adverse events was higher in the adalimumab treatment group (31.7%) compared with the placebo group (21.5%), but the difference was not statistically significant, and no serious infections were reported in the adalimumab-treated group. Five individual nonserious infectious adverse events (herpes simplex, influenza, nasopharyngitis, pharyngitis, and upper respiratory tract infection) occurred at a ≥2% higher incidence in adalimumab-treated patients than in placebo-treated patients. There were no cases of tuberculosis/granulomatous infections, demyelination, drug-induced lupus, congestive heart failure, and malignancies, and no deaths occurred during the 24-week period of the study.

Table 3. Summary of adverse events through week 24, by treatment group*
 Placebo (n = 107)Adalimumab (n = 208)
  • *

    Values are the number (%) of patients. Patients may be counted in more than 1 adverse event category.

  • P < 0.05, by Fisher's exact test.

Patients with any adverse event64 (59.8)156 (75.0)
Patients with any serious adverse event3 (2.8)6 (2.9)
Patients with any injection site reaction3 (2.8)21 (10.1)
Patients with any infection23 (21.5)66 (31.7)
Patients with any serious infection1 (0.9)0
Adverse events occurring in ≥5% of patients in either treatment group  
 Nasopharyngitis8 (7.5)26 (12.5)
 Headache9 (8.4)20 (9.6)

Only 2 treatment-emergent adverse events were reported by at least 5% of patients in the adalimumab group: nasopharyngitis and headache. Only nasopharyngitis occurred at a notably higher rate than the rate in the placebo group. Through week 24, only 6 of the 208 adalimumab-treated patients (2.9%) reported a serious adverse event, including 1 case each of head injury, chest pain, hypersensitivity to adalimumab, patella fracture, noncardiac chest pain, and increase in hepatic enzyme concentrations requiring a liver biopsy and hospitalization. Of these 6 patients, only the patient who developed hypersensitivity to adalimumab withdrew from the study. Three of the 107 placebo-treated patients (2.8%) reported a serious adverse event, including 1 case each of pancreatitis, appendicitis, and drug hypersensitivity. Of these 3 patients, only the patient who reported drug hypersensitivity (allergic reaction to rifampicin) withdrew from the study. Three patients randomized to adalimumab withdrew from the study because of nonserious adverse events, including hypergammaglobulinemia, upper respiratory tract infection, and urticaria, and 1 patient randomized to placebo withdrew from the study because of a nonserious adverse event (rash). None of the injection site reactions was considered severe.

At week 12, mean increases in albumin, total cholesterol, triglyceride, alanine aminotransferase (ALT), aspartate aminotransferase (AST), and total bilirubin concentrations and a mean decrease in the alkaline phosphatase concentration were statistically significantly different between the 2 groups (P < 0.05). Seven patients (6 taking adalimumab and 1 taking placebo) had a postbaseline ALT concentration ≥3 times the upper limit of normal. The ALT concentration returned to normal during continued adalimumab treatment in 4 of the 6 patients. Two adalimumab-treated patients and 1 placebo-treated patient had postbaseline AST concentrations ≥3 times the upper limit of normal. One other patient (ALT value 112 units/liter and AST value 70 units/liter) was hospitalized for a liver biopsy; however, the biopsy procedure itself resulted in no liver tissue for histologic evaluation. This patient had a history of moderate alcohol consumption (2–4 drinks/day), and concomitant medications included indomethacin.

Mean increases in the hemoglobin value and the lymphocyte count and mean decreases in the white blood cell, neutrophil, and platelet counts also were statistically significantly different between the 2 groups (P < 0.05). Mean changes in clinical chemistry and hematology values from baseline to week 24 were similar in magnitude and direction to the changes observed through week 12, and none of the changes at week 12 or 24 were considered clinically significant. Mean changes in vital signs at weeks 12 and 24 were small and were not statistically or clinically significantly different between treatment groups.

The percentage of patients with a prior diagnosis of either Crohn's disease or ulcerative colitis was extremely low (Table 1). None of the 7 patients with a prior diagnosis of Crohn's disease experienced flares of underlying disease. Of the 10 patients with a previous diagnosis of ulcerative colitis, 2 (adalimumab group) experienced flares of underlying disease during the 24-week period. No new cases of Crohn's disease or ulcerative colitis were diagnosed during the 24-week treatment period.

DISCUSSION

The results of this study indicate that adalimumab substantially improves the signs and symptoms of active AS for up to 24 weeks, with 58.2% of patients achieving an ASAS20 response at week 12. By week 2, after just 1 dose of adalimumab, >40% of the patients were ASAS20 responders. In contrast to previous studies of anti-TNF therapies in patients with AS (15, 17), this study design allowed early-escape open-label adalimumab treatment for patients who did not achieve a response at week 12, 16, or 20. By week 24, 94.0% of patients (296 of 315) remained in the study. The ASAS20 response in placebo-treated patients who received early-escape open-label adalimumab treatment (56.1% at week 24, after 12 weeks of treatment) was similar to the response in patients initially randomized to receive adalimumab.

Adalimumab significantly improved virtually all measures of disease activity, including patient's and physician's global assessments of disease activity, total and nocturnal back pain, inflammation, BASFI scores, BASDAI scores, and CRP concentrations. Lumbar side flexion is a particularly sensitive indicator of spinal mobility (39), and the magnitude of improvement observed at week 12 (34.2%) suggests that adalimumab treatment results in a clinically meaningful improvement in spinal mobility in a short period of time. Adalimumab-treated patients also experienced a significant reduction in enthesis pain, as indicated by both the MASES (37) and the enthesis question on the BASDAI.

A substantial percentage of patients fulfilled the ASAS40 and ASAS5/6 response criteria advocated by the ASAS International Working Group as more stringent criteria sets for demonstrating major clinical responses in clinical trials in AS (34, 39). At week 12, nearly 50% of patients met the ASAS5/6 response criteria and 40% were ASAS40 responders. A sustained, clinical response was demonstrated by maintenance of the response on the ASAS5/6 criteria and the ASAS40 response through week 24. The ASAS5/6 criteria response is a particularly useful assessment of disease activity in AS because it reflects changes in spinal mobility (BASMI in this study) and acute-phase reactants (CRP) that are not captured in the ASAS20 response. In this group of patients with active AS unresponsive to NSAIDs, >20% experienced partial remission at week 12, which was sustained through week 24.

Adalimumab treatment was also associated with substantial improvements in patients with radiographic evidence of total spinal ankylosis. Patients with total spinal ankylosis have been excluded from participation in other clinical trials of anti-TNF therapy because the response of this patient population to TNF blockade was unknown. The present study provides preliminary evidence suggesting that patients with total spinal ankylosis may benefit from anti-TNF treatment, although further study in a larger sample of patients is required to confirm this finding.

The efficacy of adalimumab for the treatment of AS is similar to that reported for both infliximab and etanercept (12, 13, 15–17). The clinical benefit of TNF blockers in AS is apparent early in the course of treatment and is sustained during long-term treatment (18–20). Anti-TNF therapy is associated with persistent regression of spinal inflammation, as measured by magnetic resonance imaging (40–42), and withdrawal of anti-TNF therapy leads to relapse of disease activity in patients with AS (43). Depending on the anti-TNF agent administered, improvements in other spondylitis-associated symptoms, including psoriasis, inflammatory bowel disease, and uveitis, have been reported (25, 44–47).

Adalimumab was generally well-tolerated by the AS patients in this study, and the adverse event profile was similar to that previously reported in patients with RA or PsA (21–25). Compared with placebo-treated patients, a higher percentage of adalimumab-treated patients experienced infectious adverse events; however, the incidence and type of infections were similar to those observed in adalimumab-treated patients with RA and PsA and are considered to be medically manageable. Most adverse events were mild or moderate, and there were no unexpected adverse events or infections. The small increases in mean concentrations of cholesterol and triglycerides observed in this and previous studies of patients with RA and PsA are probably related to correction of the dyslipoproteinemia associated with the inflammatory state. The small changes in liver enzyme, total bilirubin, and hematology values also have been observed in previous adalimumab studies and are not considered clinically important.

Most injection site reactions were mild. The incidence of injection site reactions (10% in adalimumab-treated patients and 3% in placebo-treated patients) was lower than that reported by patients in controlled studies of adalimumab treatment in RA (20% versus 14%) (48) or etanercept treatment in AS (30% versus 9%) (15). The overall safety profile of adalimumab is favorable and is comparable with that of other anti-TNF therapies for the treatment of AS (15, 17).

In conclusion, adalimumab significantly reduced the signs and symptoms of active AS and established a sustained clinical response in patients who had an inadequate response or intolerance to NSAID therapy. The overall safety profile of adalimumab is favorable and is similar to that observed in adalimumab-treated patients with RA and PsA. Further studies are needed to determine the long-term safety of adalimumab and the durability of response in patients with AS.

Acknowledgements

The authors thank Eric D. Bauer, BS, Shyanne M. Douma, BS, Winnie Lau, BS, and Hans H. Kissel, PhD, as well as all of the study site coordinators for assistance with the conduct of the study. The authors also thank Dana L. Randall, MS, PharmD, for assistance with manuscript preparation.

APPENDIX A

MEMBERS OF THE ATLAS STUDY GROUP

In addition to the authors, other members of the ATLAS Study Group are as follows: Charles Birbara, MD (Clinical Pharmacology Study Group, Worcester, MA), Michael Borofsky, MD (Arthritis and Osteoporosis Center, Clinical Research Center of Reading, West Reading, PA), Johan Bratt, MD (Karolinska University Hospital, Karolinska, Sweden), Maxime Breban, MD (Hôpital Ambroise Paré, Boulogne-Billancourt, France), Fabrizio Cantini, MD (Stabilimento Ospedaliero Misericordia e Dolce, Prato, Italy), Eduardo Collantes, MD (Hospital Universitario Reina Sofia, Cordova, Spain), John J. Cush, MD (Presbyterian Hospital of Dallas, Dallas, TX), Gino DiVittorio, MD (Coastal Clinical Research, Mobile, AL), John Flynn, MD (Johns Hopkins University, Baltimore, MD), Dale Halter, MD (Houston Institute for Clinical Research, Houston, TX), Christopher Jackson, MD (University of Utah School of Medicine, Salt Lake City), Brian Keroack, MD (Portland, ME), Joel Kremer, MD (Center for Rheumatology, Albany, NY), Richard Lautzenheiser, MD (Rheumatology Associates, Indianapolis, IN), Xavier Le Loët, MD (Hôpital de Bois-Guillaume, Rouen, France), Alan K. Matsumoto, MD (Center for Rheumatology and Bone Research, Wheaton, MD), Phillip J. Mease, MD (Seattle Rheumatology Associates, Seattle, WA), Boel Mörck, MD (Sahlgrenska Universitetssjukhus, Gothenburg, Sweden), Larry Moreland, MD (University of Alabama at Birmingham), James Morgan, MD (Dartmouth-Hitchcock Medical Center, Lebanon, NH), Juan Mulero, MD (Hospital Universitario Puerta de Hierro, Madrid, Spain), Ignazio Olivieri, MD (Ospedale S. Carlo di Potenza, Potenza, Italy), William Palmer, MD (Westroads Medical Group, Omaha, NE), Kaushik Patra, PhD (Abbott Laboratories, Parsippany, NJ), Eric Ruderman, MD (Feinberg School of Medicine of Northwestern University, Chicago, IL), Peggy Rupp, MD (Boise Arthritis Clinic, Boise, ID), Carlo Salvarani, MD (Arcispedale S. Maria Nuova, Reggio Emilia, Italy), Michael Sayers, MD (Arthritis Associates, Colorado Springs, CO), David Scott, MD (Norfolk & Norwich University Hospital NHS Trust, Norfolk, UK), Roger D. Sturrock, MD (Royal Infirmary, Glasgow, UK), Juan Torre, MD (Hospital Monte Naranco, Oviedo, Spain), Eric Veys, MD (UZ Ghent, Ghent, Belgium), Rene Westhovens, MD (UZ Gasthursberg, Leuven, Belgium), Larry Willis, MD (Lynn Health Science Institute, Oklahoma City, OK), Brian P. Wordsworth, MD (Nuffield Orthopaedic Centre, Oxford, UK), and Henning Zeidler, MD (Medizinische Hochschule Hannover, Hannover, Germany).

Ancillary