ARTHRITIS & RHEUMATISM Vol. 46, No. 3, March 2002, pp 755–765 DOI 10.1002/art.511 © 2002, American College of Rheumatology Randomized Double-Blind Comparison of Chimeric Monoclonal Antibody to Tumor Necrosis Factor ␣ (Infliximab) Versus Placebo in Active Spondylarthropathy Filip Van den Bosch, Elli Kruithof, Dominique Baeten, Annemie Herssens, Filip de Keyser, Herman Mielants, and Eric M. Veys Objective. To confirm in a placebo-controlled trial the safety and efficacy profile of infliximab in shortterm treatment of patients with active spondylarthropathy (SpA). Methods. Forty patients with active SpA were randomly assigned to receive an intravenous loading dose (weeks 0, 2, and 6) of 5 mg/kg infliximab or placebo. Evaluations for efficacy and safety were performed at weeks 1, 2, 6, 8, and 12. The primary end points of this study were the improvements in patient and physician global assessments of disease activity on a 100-mm visual analog scale. Results. Both primary end points improved significantly in the infliximab group compared with the baseline value, with no improvement in the placebo group. As early as week 2 and sustained up to week 12, there was a highly statistically significant difference between the values for these 2 end points in the infliximab versus the placebo group. In most of the other assessments of disease activity (laboratory measures, assessments of specific peripheral and/or axial disease), significant improvements were observed in the infliximab group compared with the baseline value and compared with placebo. Minor adverse events not causing discontinuation were equally observed in both treattocor. ment groups. There was one severe drug-related adverse event, in which a patient developed disseminated tuberculosis. Conclusion. Tumor necrosis factor ␣ blockade with infliximab in patients with active SpA was well tolerated and resulted in significant clinical and laboratory improvements in this short-term, placebocontrolled study. However, the occurrence of tuberculosis in one patient necessitates strict inclusion criteria and long-term followup. The spondylarthropathies (SpA) are a group of chronic autoimmune disorders of the joint (1). Entities belonging to this concept are ankylosing spondylitis (AS), reactive arthritis (ReA), psoriatic arthritis (PsA), undifferentiated SpA (uSpA) (2), arthritis associated with inflammatory bowel disease, and arthritis associated with acute anterior uveitis. This frequent rheumatic condition has a global prevalence of 0.5–1%, although there are important racial and geographic differences and a clear predominance among men. Juveniles as well as adults can be affected by one of the SpA. The different types of SpA share a number of common clinical, radiographic, and genetic features that are clearly distinct from those of other autoimmune arthritides, such as rheumatoid arthritis (RA). The hallmark symptoms include sacroiliitis, spondylitis, pauciarticular synovitis, and enthesitis. In an important number of cases, subclinical gut inflammation with pathologic findings resembling those of Crohn’s disease (CD) can be found, appearing in 25–75% of patients depending upon the type of SpA (3). Some of these patients (those with uSpA with histologic evidence of chronic gut inflammation) may eventually develop overt CD by the 5-year followup visit (4,5). In repeat ileo- Supported in part by grants from Schering-Plough and Cen- Filip Van den Bosch, MD, Elli Kruithof, MD, Dominique Baeten, MD, PhD, Annemie Herssens, Filip de Keyser, MD, PhD, Herman Mielants, MD, PhD, Eric M. Veys, MD, PhD: Ghent University Hospital, Ghent, Belgium. Address correspondence and reprint requests to Filip Van den Bosch, MD, Department of Rheumatology, Ghent University Hospital, De Pintelaan 185, B-9000 Ghent, Belgium. E-mail: firstname.lastname@example.org. Submitted for publication July 25, 2001; accepted in revised form October 15, 2001. 755 756 colonoscopy studies, clinical remission of articular symptoms was consistently associated with normalization of gut histology, while persistence of locomotor inflammation was usually associated with persistence of gut inflammation (6). Therapeutic options for patients with SpA are limited. Nearly all patients take nonsteroidal antiinflammatory drugs (NSAIDs) for control of pain and stiffness. However, their use is sometimes limited by major side effects and drug interactions. The only diseasemodifying agent that has been demonstrated to be useful for SpA is sulfasalazine (SSZ), which has a proven beneficial effect on gastrointestinal symptoms in inflammatory bowel disease (7,8). Although SSZ has also been found to have a favorable effect on articular symptoms in SpA patients, this effect appears to be modest for peripheral arthritis and enthesitis and nearly imperceptible for spondylitis (9,10). In patients with PsA, methotrexate (MTX) has become the most widely used therapy (11). Although cyclosporine has a proven effect on skin disease in psoriasis (12), the only data for its use in PsA come from a 1-year prospective trial which compared cyclosporine with MTX and showed that both treatments were effective (13); however, no doubleblind, placebo-controlled trials were performed with this agent in PsA. Recently, the use of biologic therapies that block tumor necrosis factor ␣ (TNF␣) has opened new perspectives for the treatment of patients with SpA. Infliximab (Remicade; Centocor, Malvern, PA) is a chimeric anti-TNF␣ monoclonal IgG1 antibody that neutralizes the soluble cytokine and blocks the membrane-bound cytokine (14). Investigators in an open-label trial with infliximab in patients with active CD despite steroid treatment reported clinical remission with endoscopic healing of mucosal ulcers after a single intravenous dose (15). On the basis of two randomized, double-blind, placebo-controlled trials (16,17), infliximab was approved by the health authorities in the US (the Food and Drug Administration [FDA]) and Europe (the European Medicine Evaluation Agency [EMEA]) as a treatment for therapy-resistant moderate-to-severe CD and CD with fistulas. However, studies with infliximab in inflammatory bowel disease have not evaluated the effect on associated rheumatologic manifestations, such as spondylitis, synovitis, or enthesitis. Recently, we observed a fast and significant improvement of articular as well as axial inflammation in 4 patients with SpA associated with CD treated with infliximab (18). The observations in these patients suggested that refractory joint manifestations in CD might VAN DEN BOSCH ET AL be a potential indication for infliximab treatment, and warranted further investigation of the therapeutic potential of TNF␣ blockade in patients with other subtypes of SpA. Observations in polyarticular PsA (19,20) also suggested a beneficial role of TNF␣ blockade in alleviating articular symptoms. Few data exist concerning the expression of TNF␣ in joints of patients with SpA; however, in sacroiliac joint biopsy specimens from patients with AS, an abundant TNF␣ message could be found by in situ hybridization (21). Finally, in contrast to SpA, the indication for TNF␣ blockade with infliximab in RA, another form of chronic autoimmune arthritis, has been well documented (22,23), and the drug has been approved by the FDA and the EMEA for use in therapy-resistant RA. Based on these data, a pilot study was set up to evaluate the efficacy of TNF␣ blockade with infliximab in patients with different subtypes of active SpA (24). This was an open-label study in 21 patients who received a loading-dose regimen of 3 infusions of infliximab (5 mg/kg) at weeks 0, 2, and 6. All the measured variables (global disease activity, peripheral arthritis assessments, axial disease assessments, and skin disease) improved significantly, with most parameters already reaching statistical significance at day 3. The beneficial effect was maintained up to day 84 (6 weeks after the third infusion). Simultaneously, successful treatment with infliximab in 11 patients with AS was reported by investigators in an open-label German study (25). We now report the results of a double-blind, placebo-controlled trial of infliximab in patients with different subtypes of active SpA. The study was designed to confirm the safety and efficacy profile of infliximab in short-term treatment of SpA. PATIENTS AND METHODS Study design. The study was a 12-week, monocenter, randomized, double-blind, placebo-controlled trial. The study protocol, the information for the patients, and the consent form were approved by the Ethics Committee of the Ghent University Hospital. All patients enrolled in the trial signed an informed consent form at an initial screening visit. If eligible, they were entered into the study within 4 weeks. No studyrelated procedures were performed before written informed consent was obtained. The primary end points of the study were the improvements in the patient and physician assessments of global disease activity measured on a 100-mm visual analog scale (VAS) at week 12. These end points were chosen because there are no validated outcome measurements for the diseases of the SpA concept as a whole. Wherever applicable, specific subgroup-focused measurements (e.g., in AS) were determined, as described below under clinical evaluation. All INFLIXIMAB VERSUS PLACEBO IN ACTIVE SpA clinical assessments throughout the study were performed by one investigator (FvdB). In patients with active knee arthritis, needle arthroscopy was performed as described previously (26); synovial biopsies were performed at baseline and at weeks 1 and 12. Patients. Patients had to fulfill the European Spondylarthropathy Study Group criteria for SpA (27). Patients were at least 18 years old. For patients of childbearing potential, both a negative pregnancy test result and adequate contraception during the study period and for 6 months after the last infusion were criteria for entry. At the time of enrollment, all patients had active SpA. Since there are no validated disease activity criteria for the SpA disease concept as a whole, this was defined as the presence of at least 1 swollen joint or 1 current episode of active tendinitis or dactylitis and/or inflammatory spinal pain (typical “night pain”). Disease-modifying drugs, such as SSZ, MTX, and cyclosporine, were not allowed during the study and were discontinued at least 4 weeks prior to baseline. Exclusion criteria included serious infections (e.g., hepatitis, pneumonia, pyelonephritis) in the previous 3 months, opportunistic infections within 2 months of screening, documented human immunodeficiency virus infection, proven urogenital or gastrointestinal ReA, overt CD, known malignancy, and current signs of severe, progressive, or uncontrolled concomitant disease in the opinion of the investigator. Chest radiography was performed at screening, and patients with lesions suggestive of previous tuberculosis were excluded from the study. Patients who received an investigational drug within the previous 3 months were excluded, and previous administration of any therapeutic agent targeted at reducing levels of TNF␣ was also a criterion for exclusion. Concurrent medications. Patients were allowed to continue NSAIDs and/or corticosteroids (ⱕ10 mg/day prednisolone), provided that the dosage and schedule regimen were stable for at least 4 weeks prior to baseline and throughout the study. No intraarticular injections of steroids were allowed during the trial. Randomization. A randomization list was generated by an individual (unconnected with the study) who ensured that equal numbers of patients with AS, PsA, and uSpA were entered into each of the two treatment groups. Randomization codes were provided to the hospital pharmacists, who ensured appropriate study drug preparation. The randomization code was concealed from study participants, the clinical observer, and the study coordinator. Study drug administrations. Infliximab (5 mg/kg in 250 ml 0.9% NaCl) or placebo was prepared by the hospital pharmacy. The dose of 5 mg/kg was chosen based on the pilot observations in patients with CD and associated SpA (18); as a consequence, this dose was also used in the open-label pilot trial (24). Since the present trial was designed to confirm these results in a controlled way, the same dosage regimen was used. Aseptic procedures were used during the preparation and administration of the study drug. The characteristics of the placebo and infliximab infusion bags were identical, and the investigators and patients did not know which infusion was being administered. The total dose was administered over a period of at least 2 hours. A loading-dose regimen identical to that in the open-label study (24) was used: 3 infusions were administered at weeks 0, 2, and 6. 757 Clinical evaluation. Patients were seen for clinical evaluation at baseline and at weeks 1, 2, 6, 8, and 12. At these visits, the following variables were evaluated: patient global assessment of disease activity (100-mm VAS), patient assessment of pain (100-mm VAS), physician global assessment of disease activity (100-mm VAS), erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP) level. The clinical observer was blinded to the results of laboratory testing for acute-phase measures (ESR and CRP level). Patients with peripheral arthritis evaluated their own duration of morning stiffness (in minutes) and degree of peripheral joint pain (100-mm VAS); tender joint count (68 joints) and swollen joint count (66 joints) were assessed by the physician. In patients fulfilling modified New York criteria for AS (28), axial examination was performed at baseline and at weeks 2, 6, and 12. This consisted of patient and physician assessments. Tests scored by the patient included sleep disturbance due to night pain (0 ⫽ not bothered, no pain at all; 1 ⫽ bothered a little, pain is present part of the time, but mild in character; 2 ⫽ bothered a lot, steady or intermittent pain, which usually interferes with sleep; and 3 ⫽ bothered terribly, the night pain is constant, causes marked interference with sleep and the patient is quite miserable), duration of morning stiffness of the spine (in minutes), spinal pain (100-mm VAS), Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) (29), Bath Ankylosing Spondylitis Functional Index (BASFI) (30), and Dougados Functional Index (DFI) (31). Tests performed by the physician included clinical assessment of spinal pain (0 ⫽ no pain on firm palpation, percussion, or motion; 1 ⫽ slight pain on firm palpation, percussion, or motion; 2 ⫽ moderate pain on moderate palpation, percussion, or motion; 3 ⫽ moderate-to-severe pain on slight palpation, percussion, or slight motion; and 4 ⫽ extreme pain with inability to withstand even slight palpation or percussion), Bath Ankylosing Spondylitis Metrology Index (32), chest expansion, fingersto-floor test, occiput-to-wall test, modified Schober test, cervical rotation, lateral spinal flexion, intermalleolar distance, and articular index according to Dougados (AID) (31). The extent of skin disease in psoriasis patients was measured using the Psoriasis Area and Severity Index (PASI) (33) at baseline and at weeks 6 and 12. Safety evaluation. During the infusion and for 1 hour afterward, the patient was monitored for adverse experiences and vital parameters (blood pressure, pulse, temperature). If the infusion was well tolerated, the patient was discharged. At each visit, patients were asked about side effects, and routine laboratory tests were performed, which consisted of a full blood count with white blood cell (WBC) differentiation and a biochemical screening, including electrolytes, urea, creatinine, total protein, and liver function tests. Urinalysis was performed at screening and at the final visit. Positivity for antinuclear antibodies (ANAs) and antibodies to double-stranded DNA (anti-dsDNA antibodies) was determined at baseline and at week 12. ANAs were scored by indirect immunofluorescence on HEp-2 cells, using a semiquantitative scale from 0 to 5⫹. Anti-dsDNA antibodies were identified by Crithidia luciliae indirect immunofluorescence test. Statistical analysis. The analysis was performed on the basis of intent-to-treat. Variables were summarized by presenting the median and range. Statistical power was calculated based on the data for the primary end points obtained in the 758 VAN DEN BOSCH ET AL Table 1. Demographic and descriptive characteristics of the study population* Patients† Treatment group Placebo Age, years No. men/no. women Disease duration, years Patient global assessment, 0–100-mm VAS Physician global assessment, 0–100-mm VAS Swollen joint count, 0–66 Axial night pain score, 0–3 ESR, mm/first hour CRP level, mg/dl HLA–B27, no. positive/no. negative Previous DMARDs used NSAIDs, no. using/no. not using Corticosteroids, no. using/no. not using Infliximab Age, years No. men/no. women Disease duration, years Patient global assessment, 0–100-mm VAS Physician global assessment, 0–100-mm VAS Swollen joint count, 0–66 Axial night pain score, 0–3 ESR, mm/first hour CRP level, mg/dl HLA–B27, no. positive/no. negative Previous DMARDs used NSAIDs, no. using/no. not using Corticosteroids, no. using/no. not using AS ⫹ PA Total AS 47.5 (29–66) 14/6 8 (2–34) 53.5 (24–92) 66.5 (51–94) 6 (1–17) 2 (0–3) 20 (2–107) 1.63 (0–8.37) 11/9 1 (0–3) 18/2 0/20 46.5 (29–58) 6/0 17 (6–26) 53.5 (27–69) 77 (56–78) – 1.5 (1–3) 22.5 (12–53) 2.57 (1.75–7.13) 5/1 1 (0–2) 6/0 0/6 46 (31–66) 3/1 19 (2–34) 66.5 (45–90) 65.5 (54–94) 3 (1–11) 2 (1–2) 27 (2–107) 3.45 (0–8.37) 3/1 0.5 (0–2) 3/1 0/4 46 (26–65) 13/7 6.5 (1–20) 67 (17–96) 67.5 (59–88) 9 (1–19) 1 (0–3) 17 (1–69) 1.92 (0–6.96) 11/9 1 (0–4) 19/1 3/17 44 (40–46) 2/1 10 (5–20) 84 (37–96) 85 (72–87) – 3 (2–3) 11 (5–40) 3.33 (0.8–4.29) 3/0 0 (0–1) 3/0 0/3 44.5 (35–60) 5/1 12 (2–19) 64 (17–83) 72 (59–88) 6.5 (1–19) 1.5 (1–2) 30 (11–69) 4.49 (1.05–6.96) 5/1 1.5 (1–2) 6/0 1/5 PsA uSpA 47 (39–63) 4/5 5 (3–26) 47 (24–92) 65 (51–77) 8 (1–17) 2 (0–2) 8 (4–43) 0.8 (0.27–1.51) 3/6 2 (0–3) 8/1 0/9 54 1/0 5 85 83 2 2 23 2.47 0/1 2 1/0 0/1 48 (30–65) 5/4 5 (1–18) 66 (42–84) 65 (60–85) 9 (3–14) 1 (1–2) 11 (1–26) 0.96 (0–2.41) 2/7 2 (1–4) 8/1 2/7 41.5 (26–57) 1/1 2 (2–2) 80.5 (80–81) 68 (62–74) 1.5 (1–2) 1 (0–2) 21.5 (16–27) 2.18 (2.16–2.19) 1/1 1 (1–1) 2/0 0/2 * Except where otherwise indicated, values are the median (range). AS ⫽ ankylosing spondylitis; AS ⫹ PA ⫽ AS with peripheral arthritis; PsA ⫽ psoriatic arthritis; uSpA ⫽ undifferentiated spondylarthropathy; VAS ⫽ visual analog scale; ESR ⫽ erythrocyte sedimentation rate; CRP ⫽ C-reactive protein; DMARDs ⫽ disease-modifying antirheumatic drugs; NSAIDs ⫽ nonsteroidal antiinflammatory drugs. † The placebo group had a total of 20 patients (6 with AS, 4 with AS ⫹ PA, 9 with PsA, and 1 with uSpA). The infliximab group had a total of 20 patients (3 with AS, 6 with AS ⫹ PA, 9 with PsA, and 2 with uSpA). Table 2. Global disease assessments* Patient global assessment, 0–100-mm VAS Placebo Infliximab P, vs. placebo Patient pain assessment, 0–100-mm VAS Placebo Infliximab P, vs. placebo Physician global assessment, 0–100-mm VAS Placebo Infliximab P, vs. placebo Night pain score, 0–3 Placebo Infliximab P, vs. placebo ESR, mm/hour Placebo Infliximab P, vs. placebo CRP level, mg/dl Placebo Infliximab P, vs. placebo Baseline (n ⫽ 40) Week 1 (n ⫽ 40) Week 8 (n ⫽ 39) Week 12 (n ⫽ 38) 53.5 (24–92) 67 (17–96) NS 64.5 (7–97) 26 (10–76)† NS 52.5 (14–91) 30 (0–75)† 0.038 59.5 (0–94) 18 (0–66)† ⱕ0.001 58 (2–81) 11 (1–68)† ⱕ0.001 69 (0–100) 18 (2–78)† ⱕ0.001 50 (16–90) 50.5 (14–95) NS 60 (6–94) 31 (11–69)† 0.012 57 (18–93) 29 (0–71)‡ 0.016 49.5 (1–91) 19 (0–68)† 0.002 58 (3–89) 15 (0–80)† ⱕ0.001 58.5 (2–97) 18.5 (0–79)† ⱕ0.001 66.5 (51–94) 67.5 (59–88) NS 61.5 (15–97) 37 (10–74)† 0.009 70 (22–80)§ 35 (11–65)† ⱕ0.001 65.5 (7–93)§ 18 (9–52)† ⱕ0.001 66.5 (6–86) 14 (7–65)† ⱕ0.001 72 (4–89) 16.5 (7–57)† ⱕ0.001 2 (0–3) 1 (0–3) NS 2 (0–3) 1 (0–2) NS 2 (0–3) 1 (0–2)‡ 0.006 2 (0–3) 1 (0–2)‡ 0.016 2 (0–3) 1 (0–2)‡ 0.01 20 (2–107) 17 (1–69) NS 18.5 (1–113) 6 (1–40)† 0.003 15 (1–68) 3 (1–25)† ⱕ0.001 14 (2–114) 3 (1–20)† ⱕ0.001 20 (1–117) 3 (1–23)† ⱕ0.001 1.9 (0–9.68) 0 (0–2.31)† ⱕ0.001 1.6 (0–7.04)§ 0.2 (0–2.63)† 0.003 1.3 (0–10.22) 0 (0–3.86)‡ ⱕ0.001 1.6 (0–8.37) 1.9 (0–6.96) NS 1.5 (0–9.74) 0.3 (0–1.1)† ⱕ0.001 Week 2 (n ⫽ 39) Week 6 (n ⫽ 39) 2 (0–3) 1 (0–2)‡ 0.008 17.5 (2–101) 3 (1–19)† ⱕ0.001 1.6 (0–7.42) 0 (0–7.93)‡ ⱕ0.001 * Values are the median (range). Significance of comparisons with placebo is determined by the Mann-Whitney test. NS ⫽ not significant (see Table 1 for other definitions). † P ⱕ 0.001 versus baseline, by Wilcoxon signed rank test. ‡ P ⱕ 0.01 versus baseline, by Wilcoxon signed rank test. § P ⱕ 0.05 versus baseline, by Wilcoxon signed rank test. INFLIXIMAB VERSUS PLACEBO IN ACTIVE SpA 759 Figure 1. Patient and physician global assessments of disease activity and C-reactive protein (CRP) levels in 40 patients with spondylarthropathy (20 treated with infliximab and 20 treated with placebo). A, Patient global assessment on 100-mm visual analog scale (VAS). B, Physician global assessment on 100-mm VAS. C, CRP levels in mg/dl. Values at each point are medians. † ⫽ P ⱕ 0.05, ⴱ ⫽ P ⱕ 0.01, and ⴱⴱ ⫽ P ⱕ 0.001 versus placebo, by Mann-Whitney test. ns ⫽ not significant. open-label study (24), assuming a 25% placebo response and a 0.05 Type I and Type II error. Statistical significance of the change from baseline (intragroup analysis) was measured using the Wilcoxon signed rank test. Comparison of the placebo and infliximab groups was performed using the Mann-Whitney test. P values of ⱕ0.05 were considered significant. RESULTS Forty patients were enrolled in the study and subsequently randomly assigned to treatment. Table 1 shows the demographic and descriptive characteristics of the patients in the placebo and infliximab groups. Patients were classified as having AS with only axial disease, AS with peripheral arthritis (AS ⫹ PA), PsA, and uSpA. The infliximab and placebo groups were well matched at entry, with no significant differences in age, sex ratio, disease duration, number of failed diseasemodifying antirheumatic drugs (DMARDs), the two primary end points of the study (patient and physician assessment of global disease activity), ESR, and CRP level. Nearly all patients were taking NSAIDs at study entry; 3 patients in the infliximab group were taking low-dose prednisolone (2 with PsA, 1 with AS ⫹ PA). HLA–B27 status was comparable in the two groups. General assessments were performed in all patients. Peripheral disease was evaluated in 30 patients who presented with at least 1 swollen joint (13 in the placebo group, 17 in the infliximab group). Axial evaluation was performed in 21 patients fulfilling the modified New York criteria for AS (12 in the placebo group, 9 in the infliximab group), including 2 patients with PsA who also fulfilled these criteria (both in the placebo group). Global assessments. Global assessments were performed in all patients at baseline and at weeks 1, 2, 6, 8, and 12. The results are shown in Table 2. A significant improvement in the infliximab group (but not in the placebo group) was observed for both primary end point parameters (patient and physician global assessment of disease activity), as well as for the patient assessment of pain and the intensity of night pain. Among the laboratory variables, a significant decrease was noted for ESR and CRP level in the infliximab group. This improvement was statistically significant (compared with the 760 VAN DEN BOSCH ET AL Table 3. Peripheral disease assessments* Morning stiffness, minutes Placebo Infliximab P, vs. placebo Peripheral joint pain, 0–100-mm VAS Placebo Infliximab P, vs. placebo Tender joint count, 0–68 Placebo Infliximab P, vs. placebo Swollen joint count, 0–66 Placebo Infliximab P, vs. placebo Baseline (n ⫽ 30) Week 1 (n ⫽ 30) Week 2 (n ⫽ 29) Week 6 (n ⫽ 29) 30 (0–300) 60 (0–300) NS 25 (0–300) 30 (0–200)† NS 60 (0–300) 30 (0–180)‡ NS 60 (0–300) 15 (0–180)† NS 60 (7–95) 28.5 (3–67)† NS 59 (10–83) 27 (0–57)‡ 0.005 41 (6–87) 31 (2–60)† NS 65.5 (11–91) 21 (2–56)† 0.001 6 (0–31) 10 (3–26) NS 3 (0–41) 6 (0–26)‡ NS 4 (0–36) 4.5 (0–15)‡ NS 5 (0–38) 3 (0–12)‡ NS 5 (0–40) 1 (0–11)‡ 0.032 6 (1–17) 9 (1–19) NS 3 (0–16)§ 3 (0–13)† NS 3 (0–13)§ 2 (0–14)‡ NS 63 (5–91) 52.5 (20–68) NS 2 (0–18) 1 (0–8)‡ NS Week 8 (n ⫽ 29) 60 (0–280) 30 (0–120)‡ NS 2 (0–21) 1.5 (0–8)‡ NS Week 12 (n ⫽ 28) 60 (0–240) 17.5 (0–90)‡ 0.038 70 (0–91) 19 (1–51)‡ 0.002 8 (0–42) 2 (0–8)‡ 0.015 3 (0–32) 1 (0–9)‡ NS * Values are the median (range). Significance of comparisons with placebo is determined by the Mann-Whitney test. See Tables 1 and 2 for definitions. † P ⱕ 0.01 versus baseline, by Wilcoxon signed rank test. ‡ P ⱕ 0.001 versus baseline, by Wilcoxon signed rank test. § P ⱕ 0.05 versus baseline, by Wilcoxon signed rank test. baseline value) for all of these variables from week 1 onward, with the exception of night pain, which became significant at week 2. Moreover, for all variables, there was a statistically significant difference between the infliximab and placebo groups which was present from week 2 onward. The evolution of the median value for patient and physician global assessment of disease activity as well as that for CRP level is illustrated in Figure 1. Peripheral assessments. The peripheral disease assessments were performed at baseline and at weeks 1, 2, 6, 8, and 12 in 30 patients who had at least 1 swollen joint at baseline. The results are shown in Table 3. In the infliximab group, an improvement compared with baseline was observed for all variables. With the exception of the swollen joint count, these improvements were statistically significant compared with the placebo group. Individual patient values for tender and swollen joint count at baseline and week 12 are shown in Figure 2. Axial assessments. The questionnaires were filled in by the patient at baseline and at weeks 1, 2, 6, 8, and 12, and physician assessments and metrology of the spine were performed at baseline and at weeks 2, 6, and 12. The results are shown in Tables 4 and 5, respectively. Morning stiffness of the spine as well as functional (BASFI, DFI) and disease activity (BASDAI) scores improved significantly in the infliximab group compared with baseline (Table 4). For some variables (morning stiffness, BASDAI and BASFI scores), this improve- Figure 2. Peripheral assessments of disease activity in 30 patients with spondylarthropathy, by treatment group. Individual patient data are shown at baseline (week 0) and at week 12. A, Tender joint count. B, Swollen joint count. Shown are P values for week 12 versus baseline, by Wilcoxon signed rank test. INFLIXIMAB VERSUS PLACEBO IN ACTIVE SpA 761 Table 4. Axial disease assessments by 21 patients* Baseline Morning stiffness, minutes Placebo Infliximab P, vs. placebo Spinal pain, 0–100-mm VAS Placebo Infliximab P, vs. placebo BASDAI score, 0–100 Placebo Infliximab P, vs. placebo BASFI score, 0–100 Placebo Infliximab P, vs. placebo DFI score, 0–40 Placebo Infliximab P, vs. placebo 75 (5–210) 60 (0–230) NS 63.5 (40–95) 30 (0–87) NS Week 1 Week 2 Week 6 75 (5–210) 25 (0–120)† NS 75 (2–220) 15 (0–120)† 0.041 60 (5–220) 15 (0–100)† 0.049 97 (33–91) 13 (0–69)† 0.002 65.5 (17–90) 29 (0–75) 0.009 Week 8 56 (3–72)† 15 (0–64) 0.018 90 (0–220) 15 (0–60)† 0.018 Week 12 95 (10–600) 15 (0–60)† 0.006 71 (1–88) 11.5 (0–60) 0.002 53.5 (3–96) 16 (0–43) 0.002 52.7 (27.5–85.4) 58.9 (16.3–93.1) NS 53.2 (14.6–100) 26.2 (8.7–72.6)† 0.025 56.6 (13.3–95) 24.2 (3.8–67.7)‡ 0.023 50.8 (2–90.5) 16.8 (3–59.7)‡ NS 62.1 (2.1–88.5) 8.5 (0.6–47.9)† 0.002 50.1 (3.2–94.1) 26.6 (1.2–30.7)‡ 0.002 58.9 (14.8–94.6) 46.5 (31.2–98.4) NS 57.6 (8.5–92.8) 35.8 (13–91)† NS 59.5 (4.1–92.2) 31.7 (6–90.5)† NS 59.1 (1.4–92.5) 27.9 (0.4–87.3)‡ NS 68.4 (0.7–91) 23.2 (0.3–81.1)† NS 71.9 (0.4–97.4) 27.4 (1.5–77.6)‡ 0.041 15.3 (4–28.5) 15.5 (9.5–31.5) NS 14 (4–28) 10 (4–29.5)† NS 15.3 (3–28) 10.5 (1.5–29)† NS 15.8 (1.5–24.5) 8 (1–32)† NS 16.5 (0–25.5) 7.3 (0.5–25.5)† NS 17.3 (0–26) 8 (1–27)‡ NS * Values are the median (range). Significance of comparisons with placebo is determined by the Mann-Whitney test. BASDAI ⫽ Bath Ankylosing Spondylitis Disease Activity Index; BASFI ⫽ Bath Ankylosing Spondylitis Functional Index; DFI ⫽ Dougados Functional Index (see Tables 1 and 2 for other definitions). † P ⱕ 0.05 versus baseline, by Wilcoxon signed rank test. ‡ P ⱕ 0.01 versus baseline, by Wilcoxon signed rank test. ment was statistically significant compared with placebo (Table 4). The physician-assessed spinal pain score and the AID score improved significantly in the infliximab group compared with baseline and placebo (Table 5), whereas in the metrologic assessments of the spine, no significant differences (infliximab versus placebo group) Table 5. Axial disease assessments of 21 patients by the physician* Spinal pain score, 0–4 Placebo Infliximab P, vs. placebo BASMI score, 0–10 Placebo Infliximab P, vs. placebo Occiput to wall, cm Placebo Infliximab P, vs. placebo Chest expansion, cm Placebo Infliximab P, vs. placebo Fingers to floor, cm Placebo Infliximab P, vs. placebo AID score, 0–30 Placebo Infliximab P, vs. placebo Baseline Week 2 Week 6 Week 12 2 (0–3) 3 (0–4) NS 2 (0–4) 1 (0–3)† NS 2.5 (0–3) 1 (0–3)† NS 3 (0–4) 1 (0–2)† 0.006 4 (2–7) 5 (2–10) NS 4 (1–8) 4 (1–10)† NS 4 (1–8) 4 (1–10)† NS 4 (1–8) 4 (1–10) NS 7.3 (3–16) 6 (3–26) NS 7 (2.5–15) 3 (2–24)‡ NS 7.5 (2–14.5) 2 (2–26)† NS 7.8 (2–15) 2 (1.5–25)‡ NS 3.3 (1.2–5.5) 3.5 (0.8–6.5) NS 3.3 (2.2–5.5) 4 (1.5–6) NS 3.5 (2–6.5) 5 (1.5–7)† NS 4 (2.2–6) 4.5 (2–7) NS 23 (0–45) 25 (0–63) NS 21 (2–44) 30 (0–63) NS 21 (0–41) 23 (0–60) NS 4.5 (0–11) 5 (0–12) NS 4 (0–12) 3 (0–10) NS 5.5 (0–14) 0 (0–5)† 0.003 21.5 (0–42) 30 (0–59) NS 7.5 (0–19)† 1 (0–7)† 0.015 * Values are the median (range). Significance of comparisons with placebo is determined by the Mann-Whitney test. BASMI ⫽ Bath Ankylosing Spondylitis Metrology Index; AID ⫽ articular index according to Dougados (see Table 2 for other definitions). † P ⱕ 0.05 versus baseline, by Wilcoxon signed rank test. ‡ P ⱕ 0.01 versus baseline, by Wilcoxon signed rank test. 762 VAN DEN BOSCH ET AL Figure 3. Axial assessments of disease activity in 21 patients with spondylarthropathy, by treatment group. Individual patient data are shown at baseline (week 0) and at week 12. A, Bath Ankylosing Spondylitis Disease Activity Index scores. B, Bath Ankylosing Spondylitis Functional Index scores. Shown are P values for week 12 versus baseline, by Wilcoxon signed rank test. could be detected in this short-term trial. Individual patient values for the BASDAI and BASFI are shown in Figure 3. Psoriasis skin disease. The effect on skin disease was evaluated using the PASI. No conclusions could be drawn, since the placebo and infliximab groups were not matched for this variable and there was already a significant difference between the groups at baseline. The median score in the placebo group was 8.9 (range 0.06–22.4), while that in the infliximab group was 0.235 (range 0–2.98). Safety analysis. Two serious adverse events (SAEs) were observed in the infliximab-treated group, and both patients were withdrawn from the study. The first SAE was considered to be drug related. The patient was a 65-year-old man with PsA. All screening examinations (including chest radiograph) yielded results within the normal range; there was no history of previous tuberculosis in this patient. There were clear improvements in the assessments of disease activity by the patient and physician on a 100-mm VAS from baseline values of 42 and 78, respectively; at week 8, these values had decreased to 10 and 13, respectively. At week 9 (3 weeks after the third infusion), the patient developed a systemic illness with general malaise and high-peaking fever; measures of inflammation, such as ESR and CRP level, were markedly elevated. Further examinations revealed nodular lesions in the spleen and liver, as well as mediastinal lymphadenopathies. Biopsy samples of a mediastinal lymph node, obtained through mediastinoscopy, revealed specific histologic features of tuberculosis, with the presence of acid-fast bacilli. Cultures of gastric fluid obtained in the initial phase of the illness were positive for Mycobacterium tuberculosis. The patient was treated with isoniazid, rifampin, pyrazinamide, and ethambutol, and is slowly recovering. The second SAE was considered to be procedure related rather than drug related. The patient was a 26-year-old man with uSpA who presented with a therapy-resistant arthritis of the right knee. One week after the start of the blinded study drug, the patient experienced a marked improvement in signs and symptoms (at week 1, patient global assessment dropped from 80 to 34 and physician global assessment decreased from 62 to 25 [both on a 100-mm VAS], and CRP level decreased from 2.16 mg/dl to 0.2 mg/dl). Synovial biopsy samples were obtained by needle arthroscopy to further document the clinical response seen in this patient. One day after this procedure, the patient developed an acute exacerbation of arthritis of the right knee, with marked systemic (CRP level 13.16 mg/dl) and local (synovial fluid WBC count 20,800/l; 95% neutrophils) inflammation. Culture and polymerase chain reaction of the synovial fluid yielded negative results. For safety reasons, this arthritis was considered a septic arthritis, and the patient was treated with intravenous antibiotics, recovering fully within 48 hours. All other minor AEs were recorded and are shown in Table 6. There were no significant differences between the incidence of these AEs in either of the two treatment arms. No peri-infusional AEs were observed; vital parameters remained within normal range during the infusions and for 1 hour afterward. Laboratory tests were performed as described above, and no significant changes were seen. INFLIXIMAB VERSUS PLACEBO IN ACTIVE SpA 763 Table 6. Minor adverse events recorded from baseline to week 12 Treatment group Infections Common cold Cystitis Tooth abscess Sore throat Cardiovascular Palpitations Gastrointestinal Nausea Abdominal pain Skin Itching (no rash) Injection-site reaction Neurologic Headache Dizziness Paresthesia Eye Uveitis Conjunctivitis Other Fatigue Hyperventilation Epistaxis Infliximab Placebo 5 0 0 1 3 1 1 1 0 1 1 1 0 0 2 0 2 1 2 2 1 0 0 1 1 1 0 0 3 0 1 2 1 0 Detection of ANAs. At baseline, 4 patients in the placebo group and 3 patients in the infliximab group were ANA positive. In the placebo group, 3 of the 4 patients remained positive at week 12 (with the same ANA fluorescence score); 1 patient had a negative test result at week 12, and no specific reactivities to dsDNA or extractable nuclear antigens could be determined. In the infliximab group, 1 patient remained ANA positive with the same ANA score, 1 patient became ANA negative, and 1 patient had a rise in his ANA score from 2⫹ to 4⫹. Moreover, 4 patients in the infliximab group who were ANA negative at the beginning of the study developed a positive ANA score (ranging from 3⫹ to 4⫹) at week 12; 2 patients in this group also developed anti-dsDNA antibodies. However, no lupus-like symptoms were observed. Nevertheless, all patients are still being followed up. DISCUSSION We report here the results of the first doubleblind, placebo-controlled trial of infliximab in patients with different subtypes of active SpA, including AS (with or without peripheral arthritis), PsA, and uSpA. The aim of this study was to confirm the fast and significant improvement of global, peripheral, and axial disease manifestations seen in a previously reported open pilot study (24). All patients are still in followup to enable the reporting of longer term efficacy and safety. The design of the present trial included treatment of all patients in one center with wide experience in the use of biologic therapies for autoimmune arthritis, comparison of active treatment (infliximab) with placebo in a population not receiving other DMARDs, blinding of the clinical observer to the treatment group and the results of the laboratory tests, and the use of a single observer for all patients throughout the study. The two treatment groups were well matched at entry, both for demographic features and for baseline clinical and laboratory measures of disease activity, attesting to the effectiveness of the randomization procedure. Large and highly significant differences were seen in the response rates between the infliximab and placebo groups. When looking at the global assessments of disease activity in these 40 patients, it is important to note that there was virtually no placebo response in this population. Only 3 individual patients (1 with AS, 2 with PsA) in the placebo group displayed an improvement, while most of them experienced stable or worsening disease. The sample size was calculated based on the data from the pilot trial concerning the two primary end points. While this number of patients was appropriate to confirm the efficacy seen in the open-label trial, it was not designed to yield statistically significant results in the subgroup analysis (e.g., tender and swollen joint counts in patients with peripheral arthritis, BASDAI and BASFI scores in patients with AS). Therefore, the number of patients in each subgroup is relatively small and does not allow the drawing of definite conclusions about the response in these subgroups. However, of the 18 patients who completed the study in the infliximab group, 17 experienced a major improvement, suggesting that all patients responded to the treatment equally. The medians for patient and physician global assessments of disease activity decreased from 67 and 67.5 to 18 and 16.5, respectively, representing improvements of 73% and 76%. Analysis of laboratory indices of inflammation (ESR, CRP level) showed the same impressive improvement in the infliximab-treated group compared with placebo. Other disease activity assessments for both peripheral and axial disease also showed significant improvements in the infliximab group, but not in the placebo group. Skin disease was not evaluated, since the two treatment groups already differed significantly in the extent and severity of psoriasis at study 764 entry, making it impossible to draw conclusions about the effect of the drug. The infliximab infusions were well tolerated; no allergic or anaphylactic reactions were observed. Two patients in the infliximab group were withdrawn from the trial; one was diagnosed as having disseminated tuberculosis, while the other had an acute exacerbation of arthritis following needle arthroscopy. As of January 2001, 33 incident cases of tuberculosis (30 in postmarketing experience) have been described in patients treated with infliximab (data on file with Centocor). Eleven of these 33 patients had normal pretreatment chest radiographs, which was also the case in our patient. Skin tests were not systematically performed and were only available in a minority of patients. With regard to this potentially fatal complication, a careful pretreatment screening with chest radiograph and skin test seems appropriate. Minor AEs not leading to discontinuation of the trial were observed equally in the two treatment groups. We observed the occurrence in the infliximab group (but not in the placebo group) of ANAs in 4 patients (20%) who were ANA negative at the start of the study; 2 patients (10%) also developed anti-dsDNA antibodies. However, no lupus-like symptoms were observed. In RA patients treated with infliximab, a similar occurrence of ANAs and anti-dsDNA antibodies was observed (34). The incidence of ANAs in the infliximab-treated patients rose from 29% pretreatment to 53% posttreatment. No RA patients were anti-dsDNA antibody positive prior to treatment with infliximab; 14% became anti-dsDNA antibody positive posttreatment. Only 1 of the 156 patients who were treated with infliximab developed a self-limiting clinical lupus syndrome. Fewer data exist concerning the development of anti-dsDNA antibodies in etanercept-treated RA patients. In one study, 10 of 214 etanercept-treated patients were newly positive for anti-dsDNA antibodies (35). In a combination study with MTX, 4 of 59 patients newly developed anti-dsDNA antibodies after 24 weeks of treatment with 25 mg etanercept twice weekly (36). In these studies, no symptoms suggestive of drug-induced lupus occurred, with the exception of 1 patient who had a systemic lupus erythematosus/RA overlap syndrome that had been diagnosed many years before study enrollment (35). These data confirm the findings of the open-label trial of infliximab (24) in a similar patient group. Together with the results from previous open-label studies with infliximab (18,19,25) and a placebo-controlled trial of etanercept in PsA (20), they indicate that TNF␣ plays a prominent role in the pathogenesis and disease man- VAN DEN BOSCH ET AL ifestations of SpA, and that blockade of this cytokine is highly effective in reducing signs and symptoms in this disease group as a whole. However, the mechanism by which infliximab exerts its effect in SpA remains unknown. We previously demonstrated that the clinical effect of TNF␣ blockade, seen in the open-label study, was paralleled by a restoration of the impaired Th1 cytokine profile (37), pointing toward a systemic effect of the drug. Moreover, a local biologic effect at the site of the joint was also observed (38). As part of the protocol of the present study, we performed needle arthroscopy with synovial biopsy sampling in patients with active knee arthritis. The immunohistologic features observed before and after infliximab treatment (versus placebo) are still being analyzed; these results will be communicated separately. We report the results of a successful randomized, placebo-controlled, short-term trial of infliximab in different subtypes of active SpA. The data indicate for the first time that there may be an effective therapeutic option in severe SpA. However, the occurrence of tuberculosis, which was observed in one patient treated with infliximab, necessitates strict inclusion criteria with careful screening by chest radiograph and skin test. Moreover, longer term experience and followup are needed to determine the optimal maintenance regimen (dose and interval) and to detect possible long-term side effects. ACKNOWLEDGMENTS We especially thank Dr. Freddy Cornillie of Centocor. We thank Dr. Ilse Hoffman for her help in this study, and the following rheumatologists for referring patients: Dr. Eric Dhondt, Dr. Rik Joos, Dr. Hilde Luyten, Dr. Stefaan Poriau, Dr. Benedicte Vanneuville, Dr. Ann Verbruggen, Prof. Gust Verbruggen, and Dr. Hans Zmierczak. REFERENCES 1. Wright V. Seronegative polyarthritis: a unified concept. Arthritis Rheum 1978;21:619–33. 2. Zeidler H, Mau W, Khan MA. Undifferentiated spondyloarthropathies. Rheum Dis Clin North Am 1992;18:187–202. 3. Mielants H, Veys EM, Cuvelier C, de Vos M. Ileocolonoscopic findings in seronegative spondylarthropathies. Br J Rheumatol 1988;27 Suppl 2:95–105. 4. Mielants H, Veys EM, de Vos M, Cuvelier C, Goemaere S, de Clercq L, et al. The evolution of the spondyloarthropathies in relation to gut histology. I. Clinical aspects. J Rheumatol 1995;22: 2266–72. 5. De Vos M, Mielants H, Cuvelier C, Elewaut D, Veys E. Long-term evolution of gut inflammation in patients with spondyloarthropathy. Gastroenterology 1996;110:1696–703. 6. Mielants H, Veys EM, Cuvelier C, de Vos M, Goemaere S, de Clercq L, et al. The evolution of spondyloarthropathies in relation INFLIXIMAB VERSUS PLACEBO IN ACTIVE SpA 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. to gut histology. III. Relation between gut and joint. J Rheumatol 1995;22:2273–8. Dick AP, Grayson MJ, Carpenter RC, Petrie A. Controlled trial of sulfasalazine in the treatment of ulcerative colitis. Gut 1964;5: 437–42. Van Hees PA, van Lier HJ, van Elteren PH, Driessen M, van Hogezand RA, Ten Velde GP, et al. Effect of sulfasalazine in patients with active Crohn’s disease: a controlled double-blind study. Gut 1981;22:404–9. Dougados M, van der Linden S, Leirisalo-Repo M, Huitfeldt B, Juhlin R, Veys E, et al. Sulfasalazine in the treatment of spondylarthropathy: a randomized, multicenter, double-blind, placebocontrolled study. Arthritis Rheum 1995;38:618–27. Clegg DO, Reda DJ, Abdellatif M. Comparison of sulfasalazine and placebo for the treatment of axial and peripheral articular manifestations of the seronegative spondylarthropathies: a Department of Veterans Affairs cooperative study. Arthritis Rheum 1999;42:2325–9. Cuellar ML, Espinoza LR. Methotrexate use in psoriasis and psoriatic arthritis. Rheum Dis Clin North Am 1997;23:797–809. Ellis CN, Fradin MS, Messana JM, Brown MD, Siegel MT, Hartley AH, et al. Cyclosporine for plaque-type psoriasis: results of a multidose, double-blind trial. N Engl J Med 1991;324:277–84. Spadaro A, Riccieri V, Sili-Scavalli A, Sensi F, Taccari E, Zoppini A. Comparison of cyclosporin A and methotrexate in the treatment of psoriatic arthritis: a one-year prospective study. Clin Exp Rheumatol 1995;13:589–93. Knight DM, Trinh H, Le J, Siegel S, Shealy D, McDonough M, et al. Construction and initial characterization of a mouse-human chimeric anti-TNF antibody. Mol Immunol 1993;30:1443–53. Van Dullemen HM, van Deventer SJ, Hommes DW, Bijl HA, Jansen J, Tytgat GN, et al. Treatment of Crohn’s disease with anti-tumor necrosis factor chimeric monoclonal antibody (cA2). Gastroenterology 1995;109:129–35. Targan SR, Hanauer SB, van Deventer SJ, Mayer L, Present DH, Braakman T, et al, for the Crohn’s Disease cA2 Study Group. A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor alpha for Crohn’s disease. N Engl J Med 1997;337: 1029–35. Present DH, Rutgeerts P, Targan S, Hanauer SB, Mayer L, van Hogezand RA, et al. Infliximab for the treatment of fistulas in patients with Crohn’s disease. N Engl J Med 1999;340:1398–405. Van den Bosch F, Kruithof E, de Vos M, de Keyser F, Mielants H. Crohn’s disease associated with spondyloarthropathy: effect of TNF-␣ blockade with infliximab on the articular symptoms. Lancet 2000;356:1821–2. Antoni C, Dechant C, Lorenz H, Olgivie A, Kalden-Nemeth D, Kalden J, et al. Successful treatment of severe psoriatic arthritis with infliximab [abstract]. Arthritis Rheum 1999;42 Suppl 9:S371. Mease PJ, Goffe BS, Metz J, VanderStoep A, Finck B, Burge DJ. Etanercept in the treatment of psoriatic arthritis and psoriasis: a randomised trial. Lancet 2000;356:385–90. Braun J, Bollow M, Neure L, Seipelt E, Seyrekbasan F, Herbst H, et al. Use of immunohistologic and in situ hybridization techniques in the examination of sacroiliac joint biopsy specimens from patients with ankylosing spondylitis. Arthritis Rheum 1995;38: 499–505. Elliott MJ, Maini RN, Feldmann M, Kalden JR, Antoni C, Smolen J, et al. Randomised double-blind comparison of chimeric monoclonal antibody to tumour necrosis factor ␣ (cA2) versus placebo in rheumatoid arthritis. Lancet 1994;344:1105–10. Maini RN, St Clair EW, Breedveld F, Furst D, Kalden J, Weisman M, et al, for the ATTRACT Study Group. Infliximab (chimeric anti-tumour necrosis factor ␣ monoclonal antibody) versus pla- 765 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. cebo in rheumatoid arthritis patients receiving concomitant methotrexate: a randomised phase III trial. Lancet 1999;354:1932–9. Van den Bosch F, Kruithof E, Baeten D, de Keyser F, Mielants H, Veys EM. Effects of a loading dose regimen of 3 infusions of chimeric monoclonal antibody to tumour necrosis factor ␣ (infliximab) in spondyloarthropathy: an open pilot study. Ann Rheum Dis 2000;59:428–33. Brandt J, Haibel H, Cornely D, Golder W, Gonzalez J, Reddig J, et al. Successful treatment of active ankylosing spondylitis with the anti–tumor necrosis factor ␣ monoclonal antibody infliximab. Arthritis Rheum 2000;43:1346–52. Baeten D, van den Bosch F, Elewaut D, Stuer A, Veys EM, de Keyser F. Needle arthroscopy of the knee with synovial biopsy sampling: technical experience in 150 patients. Clin Rheumatol 1999;18:434–41. Dougados M, van der Linden S, Juhlin R, Huitfeldt B, Amor B, Calin A, et al, for the European Spondylarthropathy Study Group. The European Spondylarthropathy Study Group preliminary criteria for the classification of spondylarthropathy. Arthritis Rheum 1991;34:1218–27. Van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis: a proposal for modification of the New York criteria. Arthritis Rheum 1984;27:361–8. Garrett S, Jenkinson T, Kennedy LG, Whitelock H, Gaisford P, Calin A. A new approach to defining disease status in ankylosing spondylitis: the Bath ankylosing spondylitis disease activity index. J Rheumatol 1994;21:2286–91. Calin A, Garrett S, Whitelock H, Kennedy LG, O’Hea J, Mallorie P, et al. A new approach to defining functional ability in ankylosing spondylitis: the development of the Bath ankylosing spondylitis functional index. J Rheumatol 1994;21:2281–5. Dougados M, Gueguen A, Nakache JP, Nguyen M, Mery C, Amor B. Evaluation of a functional index and an articular index in ankylosing spondylitis. J Rheumatol 1988;15:302–7. Jenkinson TR, Mallorie PA, Whitelock HC, Kennedy LG, Garrett SL, Calin A. Defining spinal mobility in ankylosing spondylitis (AS): the Bath AS metrology index. J Rheumatol 1994;21:1694–8. Fredriksson T, Pettersson U. Severe psoriasis: oral therapy with a new retinoid. Dermatologica 1978;157:238–44. Charles PJ, Smeenk RJT, de Jong J, Feldmann M, Maini RN. Assessment of antibodies to double-stranded DNA induced in rheumatoid arthritis patients following treatment with infliximab, a monoclonal antibody to tumor necrosis factor ␣: findings in open-label and randomized placebo-controlled trials. Arthritis Rheum 2000;43:2383–90. Moreland LW, Schiff MH, Baumgartner SW, Tindall EA, Fleischmann RM, Bulpitt KJ, et al. Etanercept therapy in rheumatoid arthritis: a randomized, controlled trial. Ann Intern Med 1999;130: 478–86. Weinblatt ME, Kremer JM, Bankhurst AD, Bulpitt KJ, Fleischmann RM, Fox RI, et al. A trial of etanercept, a recombinant tumor necrosis factor receptor:Fc fusion protein, in patients with rheumatoid arthritis receiving methotrexate. N Engl J Med 1999; 340:253–9. Baeten D, van Damme N, van den Bosch F, Kruithof E, de Vos M, Mielants H, et al. Impaired Th1 cytokine production in spondyloarthropathy is restored by anti-TNF␣. Ann Rheum Dis 2001;60: 750–5. Baeten D, Kruithof E, van den Bosch F, Demetter P, van Damme N, Cuvelier C, et al. Immunomodulatory effects of anti–tumor necrosis factor ␣ therapy on synovium in spondylarthropathy: histologic findings in eight patients from an open-label pilot study. Arthritis Rheum 2001;44:186–95.