Efficacy of Anakinra in Bone: Comparison to Other Biologics

Transcription

1 Advances In Therapy Volume 19 No. 1 January/February 2002 Efficacy of Anakinra in Bone: Comparison to Other Biologics Stephen A. Paget, M.D. Hospital for Special Surgery Department of Rheumatic Disease New York, New York ABSTRACT Three biologic therapies significantly slow radiographic progression in active rheumatoid arthritis. This paper compares the effects of anakinra, a recombinant human interleukin-1 receptor antagonist, with those of etanercept and infliximab, two drugs that target tumor necrosis factor-alpha. A Medline search identified controlled clinical trials that included radiographic progression as an endpoint. Anakinra 30 to 150 mg subcutaneously each day for 24 weeks was significantly more effective than placebo in slowing progression of erosion, joint-space narrowing, and total composite scores, as assessed by the Genant method, and erosive joint count, as assessed by the Larsen method. Erosion scores were slowed even further during a 24-week extension. Etanercept 25 mg subcutaneously twice weekly and infliximab 3 to 10 mg/kg intravenously every 4 or 8 weeks also slowed progressive joint damage, but these agents were studied under different study designs, patient populations, and radiographic assessments than those used in the anakinra study. Despite these differences, however, each biologic therapy appeared to slow progressive joint damage. In some studies, control of clinical symptoms did not correlate with slowing of radiographic progression. Agents that block interleukin-1 or tumor necrosis factor-alpha appear similarly effective in slowing radiographic progression in patients with active rheumatoid arthritis. Treatment strategies for this disease may need to consider clinical symptoms, progressive joint damage, and long-term safety effects separately. Keywords: rheumatoid arthritis; biologics; tumor necrosis factoralpha; interleukin-1 INTRODUCTION Rheumatoid arthritis (RA) is characterized by pain and inflammation, progressive joint destruction, functional and work disability, systemic manifestations, and in severe disease, premature death. 1 Local and systemic bone loss is an integral part of this chronic disease. 2,3 Focal erosions 2002 Health Communications Inc. Transmission and reproduction of this material in whole or part without prior written approval are prohibited Address reprint requests to Stephen A. Paget, M.D. Hospital for Special Surgery Department of Rheumatic Disease 535 East 70th St. New York, NY

2 are found within subchondral bone and at the joint margins where pannus reflection and bone invasion take place. Juxta-articular osteopenia within or adjacent to the inflamed joint likely develops from disuse and the effect of local cytokine activity. Finally, osteoporosis occurs in the axial and appendicular skeleton, even in early disease, increasing the risk of fracture. 2,4,5 Systemic osteoporosis reflects the impact of systemic inflammation, as well as the extent of disability, the menopausal status of women, and use of antiarthritis medications, particularly corticosteroids and other drugs that affect bone metabolism. 3,6 Bone erosions within RA joints progress over time but occur with greatest speed during the first 5 years of disease. 7 Notably, radiographic evidence of substantial erosions has been seen within the first 2 years 8 and may be detected by magnetic resonance imaging before radiographic findings. 9,10 Accumulating joint damage is strongly associated with work and general disability. 11 Although the rate of erosion may slow in some patients by 10 to 15 years, disability, once present, is unlikely to be reversed. Thus, aggressive interventions are needed in the early stages of RA, when slowing or stopping bone erosion may interfere with the natural history of the disease and alter the functional deterioration from joint damage. According to the traditional view of RA, joint inflammation causes clinical signs and symptoms, such as pain and swelling, as well as the processes inherent in joint destruction. However, symptomatic control with disease-modifying antirheumatic drugs (DMARDs), either alone or in combination, does suppress the destructive processes, but not completely. 12 Treatment with nonsteroidal anti-inflammatory drugs (NSAIDs) reduces pain and swelling but does not slow bone erosion or cartilage destruction. These and other findings support a new understanding of RA: joint inflammation causes pain and swelling, but synovial hyperplasia is responsible, in large part, for joint destruction. Accordingly, it may be possible to target treatment at processes directly involved in joint destruction. The exact etiology of RA remains to be defined, but key pathogenetic events include infiltration of macrophages and lymphocytes and activation of synovial fibroblasts, chondrocytes, and osteoclasts, which release inflammatory and destructive mediators. 12 The most important mediators are interleukin-1 (IL-1) and tumor necrosis factor-alpha (TNF-α). 13,14 Both of these cytokines promote infiltration of inflammatory cells into the joint by upregulating expression of endothelial cell adhesion molecules and stimulating production of chemokines. IL-1 and TNFα also activate joint mononuclear cells, stimulate synovial proliferation, and increase production of proinflammatory mediators and proteolytic enzymes. Both promote osteoclast differentiation by upregulating the receptor activator NF-kappa B ligand (RANKL) and down-regulating osteoprotegerin; IL-1 also augments osteoclastic activity. 15 As bone resorbers, osteoclasts produce the erosion seen at the pannus bone interface in RA patients. 16 The effects of IL-1, TNF-α, and other proinflammatory cytokines are normally balanced by anti-inflammatory cytokines and regulatory factors. 17 IL-1, for example, is usually held in balance by the IL-1 receptor antagonist (IL-1Ra) and soluble IL-1 receptors; however, synovial cells from RA patients do not produce sufficient IL-1Ra to completely counteract the effects of IL Observations in experimental models support the importance of IL-1 in arthritis. Erosive arthritis is not found in IL-1 deficient mice, whereas destructive arthritis occurs spontaneously in IL-1Ra deficient 28 S. A. Paget Anakinra in Bone

3 mice. 19,20 Given the central role of IL-1 in RA, exogenous administration of IL-1Ra to correct the relative deficiency may be an option for treating RA and preventing its long-term sequelae. Anakinra is a recombinant human Il-1Ra that is identical to the naturally occurring nonglycosylated form except for the addition of an N-terminal methionine. 21 This article reviews the effect of anakinra on bone erosion in RA and compares it with two biologics that reduce TNF-α activity: etanercept, a recombinant fusion protein derived from the human soluble TNF-α receptor, 22,23 and infliximab, a chimeric anti-tnf-α monoclonal antibody. 24 METHODS OF ASSESSING BONE EROSION On radiographic analysis, bone erosion, joint-space narrowing, ankylosis, juxtaarticular osteopenia, and soft-tissue swelling may be evident in RA joints. The presence of bone erosion and joint-space narrowing, in particular, has been used to monitor the efficacy of antiarthritis drugs in clinical trials. X-ray findings in RA are scored by one of two methods: a global method proposed by Larsen 25 or an analysis of erosions and joint-space narrowing as proposed by Sharp. 26 The Larsen method rapidly assesses all abnormalities in each joint and reports a single global score. The more time-consuming Sharp method reports findings for erosions and joint-space narrowing separately and as a total composite score. The Sharp approach may be advantageous in that these two changes may occur independently, and treatments may have different effects on these separate aspects of joint damage. Both methods have been modified to increase their sensitivity and reproducibility. 27,28 Although the total Sharp scores do not add much to the individual scores for erosion and jointspace narrowing, they do allow comparison with global scoring methods. Three controlled clinical trials have evaluated the effects of anakinra and the TNF-α modulators etanercept and infliximab on radiographic progression in RA These studies differed in design, populations, and use of the Larson method and varying modifications of the Sharp method. In each study, investigators experienced in the respective scoring system but unaware of the treatment evaluated the x-rays. The anakinra study 29 used the Genant modification of the Sharp method 27 to evaluate radiographs of the hands before and after 24 weeks of treatment, and after 48 weeks for those continuing in an extension study; joints in the feet were not assessed. The Genant method noted erosions in 28 joints and narrowing in 26 joint spaces, for a maximum total composite score of 202 (Table 1). Individual scores were normalized to a total possible composite score of 200. The Larsen method scored 30 joints globally on a scale of 0 to 6 (total score = 180). In addition, erosive joint counts were derived from the Larsen score by determining the number of joints with definite bone erosion (a score of 2 or higher). Modifications of the Sharp method were used to score radiographs of the hands and feet in the etanercept and infliximab trials. In the etanercept study, 30 x-rays were taken at baseline and after 6 and 12 months and were evaluated by means of van der Heijde and Rau modifications. The infliximab study 31 used the van der Heijde modification to rate x-rays taken at baseline and after 30 and 54 weeks of treatment. Radiographic progression was also assessed in several recent studies of the new DMARD leflunomide. 32 Amodified Sharp method rated erosions in 34 hand joints Advances In Therapy Volume 19 No. 1, January/February

4 and 12 foot joints on a scale of 0 to 5 and joint-space narrowing in 36 hand joints and 12 foot joints on a scale of 0 to 4. In this evaluation, the maximum total score was 422. Any attempt to compare the treatments used in these studies must recognize that differences in design (monotherapy vs combination therapy), populations (early-stage vs later-stage RA), treatment duration, and radiographic scoring methods may affect the results. Table 1. Comparison of Radiographic Scoring Methods Used in Clinical Trials of Biologics Anakinra Etanercept Infliximab Study 29 Study 30 Study 31 Radiographs Hands Only Hands Only Hands and Feet Hands and Feet Genant Larsen Method Van der Heijde/ Van der Heijde Method Modification Modification Rau Modification Modification Erosions No. of joints 28 ND (hands); 12 (feet) Scoring range (hands); 0 10 (feet) Maximum score Joint-space narrowing No. of joints 26 ND Scoring range Maximum score Total composite score No. of joints Maximum 200 * ND = not determined. * Normalized to a total of 200. EFFECT OF ANAKINRA ON BONE EROSION A randomized, placebo-controlled study conducted at 41 centers in 11 European countries evaluated the effect of anakinra on bone erosion. 21 Patients meeting American College of Rheumatology (ACR) criteria for RA were eligible for this study if they had active disease with symptoms for 6 months to 8 years. Active disease was characterized by the presence of 10 or more swollen joints and three of the following criteria: 10 or more tender or painful joints, severe or very severe disease on physician or patient assessment, or C-reactive protein level of 1.5 mg/dl or higher. Previous DMARD therapy was discontinued at least 6 weeks before enrollment. A total of 472 patients were assigned to 24 weeks of anakinra 30, 75, or 150 mg or 30 S. A. Paget Anakinra in Bone

5 placebo once daily by subcutaneous injection. During a 24-week extension, anakinra patients continued on their given dose, and placebo patients were randomly assigned to one of the three anakinra doses. At baseline, demographic and clinical characteristics were comparable in the four treatment groups (Table 2). 21 Previous DMARD use ranged from 66% (anakinra 150-mg group) to 81% but did not predict response to treatment. At baseline, absence of erosions ranged from 13% (Genant method) to 25% (Larsen method). 29 Radiographic scores were similar across the four groups at baseline and on comparisons between completers and noncompleters of the first 24 weeks of treatment. Table 2. Demographic and Clinical Characteristics in Clinical Trials of Biologics Anakinra vs Etanercept vs Infliximab vs Parameter Placebo 29 Methotrexate 30 Placebo 31* Treatment arms 4 (3 active, 3 (all active) 4 (3 active, 1 placebo) 1 placebo) Patients, no Mean age range, y Female, % Mean duration of RA, y Rheumatoid factor, % Corticosteroid use, % NSAID use, % Swollen joints, no Tender joints, no C-reactive protein, mg/dl * Methotrexate included in both treatment arms. All three doses of anakinra significantly reduced progression of the Genant total score relative to placebo after 24 weeks (Fig 1). 29 In the placebo group, the total composite score increased by 3.5 from baseline, comparable to the predicted change (3.6) calculated from the baseline score and RA duration. In each anakinra group, however, the actual change in total score was less than predicted and significantly lower than that with placebo for all doses combined (P.0004) (1.87, 1.86, and 1.81 for the 30-, 75-, and 150-mg doses); the predicted changes from baseline were 3.4, 3.5, and 3.2. Compared with placebo, anakinra, across all three doses, significantly reduced the progression of erosions (1.18 vs 1.91; P =.0097) and joint-space narrowing (0.68 vs 1.62; P =.0003), based on the Genant evaluation. Significant reductions were noted with the 30- and 150-mg doses in erosions (P.03 vs placebo) and with each dose in joint-space narrowing (P<.01). Advances In Therapy Volume 19 No. 1, January/February

6 Fig 1. Effect of anakinra on radiographic progression, as assessed by the Genant and Larsen methods. Shown are the least-square mean changes from baseline to week 24 for total composite, erosion, and joint-space narrowing scores (Genant method) and erosive joint count (Larsen). From Jiang et al. 29 LS Mean Change From Baseline * * * * * Placebo Anakinra 30 mg Anakinra 75 mg Anakinra 150 mg * * P.004. P< Total Genant Score Erosion Joint-Space Narrowing Larsen Erosive Joint Count Results were similar in an analysis by the Larsen method. 29 When all three doses were combined, anakinra significantly reduced the progression in erosive joint count relative to placebo (1.44 vs 2.62; P =.0005); only the 75-mg dose achieved statistical significance (P =.003 vs placebo). The change from baseline was 6.22 with placebo and 3.76 with all anakinra doses (P =.0294). In the extension period (weeks 24 to 48), changes in total Genant (1.32) and Larsen (3.13) scores for patients switched from placebo to anakinra were comparable to the change from baseline with anakinra during the first 24 weeks of the study (1.85 and 3.76, respectively). 29 The switch to anakinra was accompanied by significant slowing of erosions (P =.0001) and decrease in total score (P =.0005) according to the Genant method; joint-space narrowing slowed to a lesser extent (P =.0195). In the anakinra groups, radiographic progression appeared delayed during the extension, particularly with the 75- and 150-mg doses. Notably, the benefit of anakinra was more evident in joint-space narrowing during the first 24 weeks, whereas the effect on erosions was greater during the extension (Fig 2). Overall, 106 of 248 patients (43%) receiving anakinra showed no radiographic progression after the first 24 weeks of treatment, according to the Genant total score, including 10 patients (4%) whose scores decreased. 29 In comparison, 26 of 78 patients (33%) did not have radiographic progression with placebo; no scores in this group decreased. Erosions did not progress in 130 patients (53%) in the anakinra groups, 32 S. A. Paget Anakinra in Bone

7 compared with 33 (42%) in the placebo group; respective figures for maintenance of joint-space narrowing were 147 patients (59%) and 34 patients (44%). A higher percentage of anakinra-treated patients had decreased scores; a higher percentage of placebo patients had increased scores. Fig 2. Effect of anakinra on radiographic progression from weeks 0 to 24 and weeks 24 to 48, as assessed by the Genant method. Data for the three groups of patients treated with anakinra are combined. From Jiang et al Weeks 0 24 Weeks Mean Change in Radiographic Score P = P = Total Genant Score Erosions Joint-Space Narrowing The findings were highly correlated at baseline and at weeks 24 and 48 (all P<.001) with both methods. Correlation coefficients were 0.84 at baseline and 0.83 at weeks 24 and 48 between the Genant total score and Larsen scores and 0.83 at all three assessments between the Genant erosion score and the Larsen erosive joint count. 29 EFFECT OF OTHER BIOLOGICS ON BONE EROSION Etanercept The effect of etanercept on radiographic progression was evaluated in a randomized, controlled study of 632 patients with active early-stage RA. 30 Eligible patients had RA for less than 3 years; either seropositivity for rheumatoid factor or at least three bone erosions on radiographs of the hands, wrist, and feet; at least 10 swollen and 12 tender joints; and an erythrocyte sedimentation rate (ESR) of 28 mm/h or higher, C-reactive protein of 2 mg/dl or higher, or morning stiffness for at least Advances In Therapy Volume 19 No. 1, January/February

8 45 minutes. DMARDs were withdrawn at least 4 weeks before enrollment. Etanercept 10 or 25 mg twice weekly by subcutaneous injection or methotrexate once weekly orally was continued for 12 months. The three treatment groups were demographically and clinically comparable at baseline (see Table 2). 30 Mean duration of disease was 11 months in the etanercept 10-mg group and 12 months in the other two groups. This study differed from the previously described anakinra study in three ways: (1) these patients had earlierstage RA (mean duration, 1 year vs 4 years in the anakinra study); (2) a higher percentage of patients in the etanercept study were seropositive for rheumatoid factor; and (3) the etanercept study included an active control (methotrexate); the anakinra study used a placebo control. Changes from baseline in total Sharp score over 1 year tended to be lower with etanercept 25 mg than with the other treatments (P =.11) (Fig 3) 30 and were lower than the 12-point change predicted from the total score at baseline and the disease duration. The short duration of disease may introduce an artifact into the predicted change in Sharp score owing to the relatively small number used in the denominator. The mean increase from baseline in erosion score was significantly lower with etanercept 25 mg than with methotrexate at the month 6 (0.3 vs 0.68; P =.001) and month 12 (0.47 vs 1.03; P =.002) evaluations. Etanercept 10 mg and methotrexate had comparable effects on bone erosion. The joint-space narrowing score increased by approximately 0.4 in each treatment group over the 1-year study. Fig 3. Effect of etanercept and methotrexate on radiographic progression at 6 and 12 months. Shown are mean changes from baseline in total scores, as assessed by the van der Heijde modification of the Sharp scale. From Bathon et al Etanercept 25 mg Etanercept 10 mg Methotrexate Mean Change in Sharp Score Months 12 Months 34 S. A. Paget Anakinra in Bone

9 Most patients in this study 72% taking etanercept 25 mg and 60% taking methotrexate showed no increase in erosion score (P =.007). The absence of radiographic progression was correlated with clinical evidence of reduced disease activity, particularly for decreases in C-reactive protein levels. In this study, etanercept was an effective treatment of early RA. Infliximab The effect of infliximab on radiographic progression was evaluated in a randomized, controlled trial conducted at 34 centers in the United States and Europe. 31 Patients with active RA despite at least 12.5 mg of methotrexate were eligible. Active disease was defined by the presence of at least six swollen and six tender joints, and two or more of the following: morning stiffness for 45 minutes or longer, ESR 28 mm/h or higher; or C-reactive protein 2.0 mg/dl or more. A total of 428 patients continued to receive methotrexate and then were randomly assigned to one of five treatment arms: infliximab 3 mg/kg every 4 or 8 weeks, infliximab 10 mg/kg every 4 or 8 weeks, or placebo. Infliximab was administered intravenously. Although this study included a placebo arm, infliximab monotherapy was not evaluated. Instead, the effect of adding infliximab to existing methotrexate therapy was compared with continuing methotrexate that had failed to adequately control disease activity. Patients in the five treatment arms were well matched for demographic and clinical characteristics (see Table 2). The weekly dose of methotrexate at baseline averaged 16 mg in four groups and 17 mg in the group receiving infliximab 10 mg/kg every 4 weeks. Notably, patients in this study had long-standing disease, with a mean duration of 9 to 12 years. Thus, this study differed from the anakinra study in several ways: (1) these patients had later-stage RA (mean disease duration, 10 vs 4 years in the anakinra study); (2) they tended to have fewer swollen and tender joints; and (3) this study evaluated the effect of adding infliximab to existing methotrexate therapy, whereas anakinra was evaluated in the absence of any DMARD therapy. The radiographic analysis involved 349 patients; a sensitivity analysis accounted for any missing data. Mean changes from baseline in the van der Heijde modification of the total Sharp score were significantly lower with infliximab plus methotrexate than with placebo and methotrexate alone (P<.001) (Fig 4). 31 The mean changes in total score were 1.6 and 1.3 in the groups receiving the lower dose of infliximab every 4 and 8 weeks, 0.7 and 0.2 in the higher-dose infliximab + methotrexate groups every 4 and 8 weeks, and 7.0 in the placebo + methotrexate group. Similar effects were observed when erosions and joint-space narrowing were evaluated separately. Thus, treatment with infliximab 10 mg/kg appeared to have stopped radiographic progression during the 54-week study. Notably, the benefit of infliximab on the total radiographic score was similarly evident in patients achieving an ACR20 response as in those with no clinical response to treatment. Advances In Therapy Volume 19 No. 1, January/February

10 Fig 4. Comparison of infliximab (+ methotrexate) and placebo (+ methotrexate) on radiographic progression after 54 weeks of treatment, as assessed by the van der Heijde modification of the Sharp scale. From Lipsky et al. 31 Mean Change From Baseline in Radiographic Score Total score Erosions score Joint-space narrowing 2 Placebo (+ methotrexate) 3 mg/kg q8wk 3 mg/kg q4wk 10 mg/kg q8wk 10 mg/kg q4wk *All P<.001 vs placebo. Infliximab (+ Methotrexate) * Leflunomide Leflunomide, the newest DMARD, also slowed radiographic progression of RA in three randomized, controlled studies. 32 In the first study, 480 patients with a mean disease duration of 6.5 to 7 years were treated with leflunomide 20 mg daily, methotrexate 7.5 to 15 mg weekly, or placebo for 1 year. In the second study, 356 patients with RA for a mean of 5.7 to 7.6 years received leflunomide 20 mg daily, sulfasalazine 0.5 to 2.0 g daily, or placebo for 6 months. In the third study, 985 patients with a mean disease duration of 3.7 to 3.8 years were treated with either leflunomide 20 mg daily or methotrexate 7.5 to 15 mg weekly for 1 year. In each study, approximately 40% of patients had disease for 2 years or less. In the two placebo-controlled studies, treatment with leflunomide, methotrexate, or sulfasalazine produced significantly less radiographic progression than did placebo. 32 Changes from baseline in Sharp score were 0.53 with leflunomide, 0.89 with methotrexate, and 2.16 with placebo (P =.0007 for leflunomide vs placebo, P =.02 for methotrexate vs placebo) in the first study; 1.23 with leflunomide, 2.32 with sulfasalazine, and 5.88 with placebo (P =.0004 for leflunomide vs placebo, P =.05 for sulfasalazine vs placebo) in the second study (changes with active treatment were lower than predicted from baseline Sharp score and disease duration); and 2.48 with leflunomide and 1.62 with methotrexate (P =.29) in the third study (both less than predicted 36 S. A. Paget Anakinra in Bone

11 from baseline scores and disease duration). In each study, an effect of active treatment was evident on scores for both erosions and joint-space narrowing. The effect of treatment on radiographic progression did not correlate with clinical response. In the two leflunomide-methotrexate studies, ACR20 responders had smaller changes in total Sharp score than nonresponders; however, in the leflunomide-sulfasalazine study, ACR20 responders showed greater radiographic progression than nonresponders. As in the studies with anakinra and the TNF-α modulators, only small percentages of patients experienced new erosions during treatment: 28% to 34% of the placebo group, 21% to 32% of those receiving leflunomide, 23% to 30% of those receiving methotrexate, and 27% of patients receiving sulfasalazine. SUMMARY Several controlled clinical trials have shown that biologic therapy directed against either IL-1 or TNF-α slows radiographic progression of RA Despite differences in study design, patient populations, and radiographic scoring methods, anakinra, etanercept, and infliximab had comparable effects, as determined by total radiographic scores as well as separate scores for erosions and joint-space narrowing. Clearly, in the absence of direct clinical comparisons, standard methods for assessing radiographic progression are needed to provide a more nuanced interpretation of trial results. In general, all three biologics were well tolerated, although the risk of infection may be higher with therapies directed against TNF-α. The incidence of upper respiratory tract infections (34% vs 22%), sinusitis (17% vs 6%), and pharyngitis (11% vs 6%) was higher with infliximab (plus methotrexate) than with placebo (plus methotrexate). 31 Similarly, upper respiratory tract infections (27% 35%), rhinitis (15% 17%), and sinusitis (10% 13%) occurred at comparable incidences with etanercept 10 and 25 mg and with methotrexate. 30 In comparison, 15% to 17% of patients treated with anakinra had infections, compared with 12% of placebo patients. 21 Only three anakinra-treated patients and one placebo patient withdrew because of infection. The types of infection in the anakinra study were not reported. Of note, recent warnings were added to the labeling for TNF-α blocking agents. Though rare, warnings for disseminated tuberculosis and sepsis were added to etanercept and infliximab labeling. 32 These biologics can be distinguished by their mechanism of action and route and frequency of administration. Anakinra blocks IL-1, whereas etanercept and infliximab target TNF-α. Both anakinra and etanercept are administered by subcutaneous injection; infliximab is given intravenously. Frequency of administration is daily with anakinra, twice weekly with etanercept, and every 4 to 8 weeks with infliximab. Finally, control of clinical symptoms does not ensure a halt to radiographic progression. Evidence from the etanercept study and one leflunomide study shows that patients with ACR20 responses had similar or greater radiographic progression than nonresponders. 30,33 These observations suggest that treatment may need to address clinical symptoms and radiographic progression separately, recognizing that clinical symptoms are due to inflammation, whereas radiographic progression results from synovial hyperplasia. Advances In Therapy Volume 19 No. 1, January/February

12 REFERENCES 1. Pincus T. Long-term outcomes in rheumatoid arthritis. Br J Rheumatol. 1995;34(suppl 2): Gough AK, Lilley J, Eyre S, Holder RL, Emery P. Generalised bone loss in patients with early rheumatoid arthritis. Lancet. 1994;344: Celiker R, Gokce-Kutsal Y, Cindas A, et al. Osteoporosis in rheumatoid arthritis: effects of disease activity. Clin Rheumatol. 1995;14: Hooyman JR, Melton LJ, Nelson AM, O Fallon WM, Riggs BL. Fractures after rheumatoid arthritis. A population-based study. Arthritis Rheum. 1984;27: Spector TD, Hall GM, McCloskey EV, Kanis JA. Risk of vertebral fracture in women with rheumatoid arthritis. BMJ. 1993;306: Kirwan JR. The effect of glucocorticoids on joint destruction in rheumatoid arthritis. N Engl J Med. 1995;333: Pincus T, Callahan LF, Fuchs HA, Larsen A, Kaye J. Quantitative analysis of hand radiographs in rheumatoid arthritis: time course of radiographic changes, relation to joint examination measures, and comparison of different scoring methods. J Rheumatol. 1995;22: Fuchs HA, Kaye JJ, Callahan LF, Nance EP, Pincus T. Evidence of significant radiographic damage in rheumatoid arthritis within the first 2 years of disease. J Rheumatol. 1989;16: McGonagle D, Conaghan PG, O Connor P, et al. The relationship between synovitis and bone changes in early untreated rheumatoid arthritis: a controlled magnetic resonance imaging study. Arthritis Rheum. 1999;42: Klarlund M, Ostergaard M, Jensen KE, Madsen JL, Skjodt H, Lorenzen I. Magnetic resonance imaging, radiography, and scintigraphy of the finger joints: one year follow up of patients with early arthritis. Ann Rheum Dis. 2000;59: Scott DL, Pugner K, Kaarela K, et al. The links between joint damage and disability in rheumatoid arthritis. Rheumatology. 2000;39: Harris ED. Rheumatoid arthritis: pathophysiology and implications for therapy. N Engl J Med. 1990;322: Arend WP, Dayer J-M. Inhibition of the production and effects of interleukin-1 and tumor necrosis factor α in rheumatoid arthritis. Arthritis Rheum. 1995;38: Dinarello CA, Wolff SM. The role of interleukin-1 in disease. N Engl J Med. 1993;328: Roux S, Orcel P. Bone loss: factors that regulate osteoclast differentiation: an update. Arthritis Res. 2000;2: Goldring SR, Gravallese EM. Pathogenesis of bone erosions in rheumatoid arthritis. Curr Opin Rheumatol. 2000;12: Burger D, Dayer JM. Inhibitory cytokines and cytokine inhibitors. Neurology. 1995;45(suppl 6): S39-S Firestein GS, Boyle DL, Yu C, et al. Synovial interleukin-1 receptor antagonist and interleukin-1 balance in rheumatoid arthritis. Arthritis Rheum. 1994;37: van den Berg W, Joosten LAB, Kollias G, van de Loo F. Role of tumour necrosis factor α in experimental arthritis: separate activity of interleukin 1β in chronicity and cartilage destruction. Ann Rheum Dis. 1999;58(suppl I):I40-I Horai R, Saijo S, Tanioka H, et al. Development of chronic inflammatory arthropathy resembling rheumatoid arthritis in interleukin 1 receptor antagonist-deficient mice. J Exp Med. 2000;191: Bresnihan B, Alvaro-Gracia JM, Cobby M, et al. Treatment of rheumatoid arthritis with recombinant human interleukin-1 receptor antagonist. Arthritis Rheum. 1998; S. A. Paget Anakinra in Bone

13 22. Moreland LW, Schiff MH, Baumgartner SW, et al. Etanercept therapy in rheumatoid arthritis. A randomized, controlled trial. Ann Intern Med. 1999;130: Weinblatt ME, Kremer JM, Bankhurst AD, 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: Maini R, St Clair EW, Breedveld F, et al. Infliximab (chimeric anti-tumour necrosis factor alpha monoclonal antibody) versus placebo in rheumatoid arthritis patients receiving concomitant methotrexate: a randomised phase III trial. Lancet. 1999;354: Larsen A, Dale K, Eek M. Radiographic evaluation of rheumatoid arthritis and related conditions by standard reference films. Acta Radiol Diagn. 1977;8: Sharp JT, Lidsky MD, Collins LS, Moreland J. Methods of scoring the progression of radiologic changes in rheumatoid arthritis: correlation of radiologic, clinical and laboratory abnormalities. Arthritis Rheum. 1971;14: Genant HK, Jiang Y, Peterfy C, Lu Y, Redei J, Countryman PJ. Assessment of rheumatoid arthritis using a modified scoring method on digitized and original radiographs. Arthritis Rheum. 1998; 41: van der Heijde DM, van Leeuwen MA, van Riel PL, et al. Biannual radiographic assessment of hands and feet in the three-year prospective followup of patients with early rheumatoid arthritis. Arthritis Rheum. 1992;35: Jiang Y, Genant HK, Watt I, et al. A multicenter, double-blind, dose-ranging, randomized, placebo-controlled study of recombinant human interleukin-1 receptor antagonist in patients with rheumatoid arthritis: radiologic progression and correlation of Genant and Larsen scores. Arthritis Rheum. 2000;43: Bathon JM, Martin RW, Fleischmann RM, et al. A comparison of etanercept and methotrexate in patients with early rheumatoid arthritis. N Engl J Med. 2000;343: Lipsky PE, van der Heijde DM, St Clair EW, et al. Infliximab and methotrexate in the treatment of rheumatoid arthritis. N Engl J Med. 2000;343: Safety update on TNF-α antagonist: infliximab and etanercept. Available at: dockets/ac/01/briefing/3779b2_01_cber_safety_revision2.htm. Accessed September 4, Sharp JT, Strand V, Leung H, Hurley F, Loew-Friedrich, for Leflunomide Rheumatoid Arthritis Investigators Group. Treatment with leflunomide slows radiographic progression of rheumatoid arthritis: results from three randomized controlled trials of leflunomide in patients with active rheumatoid arthritis. Arthritis Rheum. 2000;43: Advances In Therapy Volume 19 No. 1, January/February